Management of

Insect Pests:
Nuclear and Related
Molecular and
Genetic Techniques

PROCEEDINGS
OF A SYMPOSIUM
VIENNA
19-23 OCTOBER 1992
JOINTLY ORGANIZED BY
IAEA AND FAO
 PROCEEDINGS SERIES

MANAGEMENT OF INSECT PESTS

NUGLEAR AND
RELATED MOLECULAR
AND GENETIC TECHNIQUES

PROCEEDINGS OF AN INTERNATIONAL SYMPOSIUM

ON MANAGEMENT OF INSECT PESTS:
NUCLEAR AND RELATED MOLECULAR
AND GENETIC TECHNIQUES
JOINTLY ORGANIZED BY THE
INTERNATIONAL ATOMIC ENERGY AGENCY
AND THE
FOOD AND AGRICULTURE ORGANIZATION
OF THE UNITED NATIONS
AND HELD IN VIENNA, 19-23 OCTOBER 1992

INTERNATIONAL ATOMIC ENERGY AGENCY

VIENNA, 1993
Permission to reproduce or translate the information contained in this
publication may be obtained by writing to the International Atomic Energy
Agency, Wagramerstrasse 5, P.O. Box 100, A-1400 Vienna, Austria.

© IAEA, 1993

VIC Library Cataloguing in Publication Data

International Symposium on Management o f Insect Pests : Nuclear and Related

Molecular and Genetic Techniques (1992 : Vienna, Austria)
Management o f insect pests : nuclear and related molecular and genetic
techniques : proceedings o f an International Symposium on Management of
Insect Pests : Nuclear and Related Molecular and Genetic Techniques jointly
organized by the International Atomic Energy Agency and the Food and
Agriculture Organization of the United Nations and held in Vienna,
19-23 October 1992. — Vienna : The Agency, 1993.
p. ; 24 cm. — (Proceedings series, ISSN 0074-1884)
STI/PUB/909
ISBN 9 2 -0 -0 0 0 2 9 3 -5
Includes bibliographical references.

1. Insect pests—Biological control. 2. Insect radiosterilization.

I. International Atomic Energy Agency. II. Food and Agriculture Organi­
zation. III. Title. IV . Series: Proceedings series (International Atomic
Energy Agency).

VICL 93-00066

Printed by the IAEA in Austria

September Í993
STI/PUB/909
 FOREWORD

The balance of the relationship between people and insects is quite un­
acceptable in many regions of the world. The incidence of malaria, which is
transmitted by mosquitoes, is far greater today than thirty years ago, when the
A n o p h e l e s vectors were being suppressed strongly with DDT and other organo-
chlorines in a co-ordinated global campaign. At that time, many hoped that the
mosquitoes that transmit malaria would be reduced to such low numbers that this
deadly disease would be eradicated from large sections of the globe. Subsequently,
the programmes against these vectors lost their effectiveness and there has been a
widespread resurgence of the disease. Similarly in Africa, ground has been lost
against tsetse flies and trypanosomiasis in livestock and man. Tropical fruit flies des­
troy fruit and vegetables and they are a serious barrier for exports from developing
countries to markets in several industrialized countries. Some major fruit fly pests
have spread to other continents and threaten to spread even more widely as interna­
tional travel continues to increase. Intermittently, locusts and other acridids devastate
crops in Africa, the Middle East and Asia. The dreaded cotton boll, weevil expanded
its range into the cotton growing areas of Brazil, where it has resulted in widespread
economic losses. The populations of more than six hundred species of insects have
developed resistance to insecticides, still the major weapon used to combat them.
The severity of certain insect pests, such as the whitefly, has increased as older insec­
ticides have been replaced with synthetic pyrethroids, which decimate some of the
natural enemies that have hitherto effectively kept some of these pests under a
modicum of control.
Yet there are substantial grounds for optimism, based on some major and last­
ing advances against certain dangerous pests, and on very promising developments
in science and technology. Thus, the New World screwworm has been eradicated
by means of the sterile insect technique (SIT) from all of Mexico and the United
States of America, and this campaign is rapidly advancing toward Panama. This
deadly parasite was also eradicated from the Libyan Arab Jamahiriya, where it posed
a tremendous threat to the livestock, wildlife and people of Africa and the Mediterra­
nean region. The SIT was also used to eradicate both the melon fly and the Oriental
fruit fly from Japan. The greatest single achievement in classical biological control
occurred recently in Africa, where the ravages of the cassava mealybug were brought
under lasting control throughout the cassava growing zone extending over 34 sub-
Saharan countries. This was accomplished by mass rearing its natural enemy, an
encyrtid wasp from South America, and distributing it over this vast region. In
Indonesia, the brown plant hopper and other insects on rice were controlled by
fostering the resurgence of their natural enemies, brought about by strongly limiting
the excessive use of insecticides.
 Progress has been made during the past decade in overcoming many of the
impediments of biologically based methods of pest management. The reliability and
economy of mass rearing many insects have been improved significantly. Significant
advances have been made in formulating naturally occurring compounds, such as
pheromones and biological control agents. It has become possible to develop robust
genetic sexing strains of insects, so that only sexually sterile males can be released.
This will definitely increase the effectiveness and economy of SIT.
A topic of particular importance, dealt with at length in these Proceedings,
concerns the advances made in the field of molecular technology and biotechnology.
DNA probes and other molecular techniques are being used to identify cryptic
species and individual insects with genes for resistance to insecticides. Bacteria and
several crops have been genetically engineered to express the Ô endotoxin of B a c i l l u s
t h u r i n g i e n s i s for effective insect control on a commercial basis. The excessively
narrow host range o f some nuclear polyhedrosis viruses and other biological control
agents is being overcome through genetic engineering. However, many of the major
benefits from such approaches will not be realized until practical technology for the
genetic transformation of economically important arthropods has been developed.
The presentations in this Symposium focused on advances and trends in insect
control and eradication, genetic engineering and molecular biology, insect genetics,
operational SIT programmes, Fj sterility and behaviour, biocontrol, tsetse fly R&D
and quarantine. The Symposium was attended by 83 participants from 36 countries
and 3 international organizations. Sixty papers and four posters were presented and
are included in these Proceedings.
 The Symposium participants expressed their desire to preserve in memory the life and
work of the late André Van Der Vloedt. Dr. Van Der Vloedt pioneered in the development and
use of the sterile insect technique in the combat o f tsetse fly vectors o f trypanosomiasis. He
became ill while on a mission in the United Republic o f Tanzania and Zambia and died in
Vienna, after a brief illness, on 31 December 1991. Dr. Van Der Vloedt was an outstanding
leader, teacher and researcher. His infectious enthusiasm and boundless good will enabled
him to reach across the chasms o f race and culture and to stimulate scientific advancement
on several continents.
 E D IT O R IA L N O T E

The Proceedings have been edited by the editorial staff o f the IAEA to the extent
considered necessary for the reader’s assistance. The views expressed remain, however, the
responsibility o f the named authors or participants. In addition, the views are not necessarily
those o f the governments o f the nominating Member States or o f the nominating organizations.
Although great care has been taken to maintain the accuracy of information contained
in this publication, neither the IAEA nor its Member States assume any responsibility for
consequences which may arise from its use.
The use o f particular designations o f countries or territories does not imply any judge­
ment by the publisher, the IAEA, as to the legal status of such countries or territories, of their
authorities and institutions or o f the delimitation o f their boundaries.
The mention o f names o f specific companies or products (whether or not indicated as
registered) does not imply any intention to infringe proprietary rights, nor should it be
construed as an endorsement or recommendation on the part o f the IAEA.
The authors are responsible for having obtained the necessary permission for the IAEA
to reproduce, translate or use material from sources already protected by copyrights.
Material prepared by authors who are in contractual relation with governments is
copyrighted by the IAEA, as publisher, only to the extent permitted by the appropriate national
regulations.
 CONTENTS

OPENING (Session 1)

Advances and trends in managing insect pests (IAEA-SM-327/1) ................. 3

R . L . R i d g w a y , M :N . I n s c o e , K .W . T h o r p e
Biotechnological prospects for managing insect pests (IAEA-SM-327/2) ...... 17
J . O a k e s h o t t , P. C h r i s t i a n , P . W. A t k in s o n
Genetic engineering of insects and applications in basic and
applied entomology (IAEA-SM-327/5) ............ ........................................ 33
J .M . C ram p ton
Eradication of the melon fly from Okinawa, Japan, by means of the
sterile insect technique (IÁEA-SM-327/4) .................................................. 49
M . Y a m a g is h i , H . K a k i n o h a n a , H . K u b a , T. K o h a m a , Y. N a k a m o t o ,
Y. S o k e i , K . K i n j o

GENETIC ENGINEERING AND MOLECULAR BIOLOGY (Session 2)

Advances in the preservation of insect germplasm (IAEA-SM-327/27) ........ 63

R .A . L e o p o l d
Technology for transforming the germ line in economically
important insects: Current status-and prospects for advancements
(IAEA-SM-327/6) ...................................................................................... 79
A .M . H a n d l e r
Prospects for non-drosophilid germ line transformation
(IAEA-SM-327/7) ....................................................................................... 93
P .W . A t k i n s o n , C . J . C o a t e s , A .Ç . P i n k e r t o n ,
H.A. Mende, A.J. Howells> D.A. O ’Brochta
Development and use of DNA probes in mapping insect genomes
(IAEA-SM-327/8) ...................................................................................... 101
A . S. R o b i n s o n
Are immediate early baculovirus gene promoters useful tools
in insect control? (IAEA-SM-327/10) ........................ .............................. 115
R . H u y b r e c h t s , M . V a n B r u s s e l , V. V u ls t e k e , L . S c h o o f s
J . R o b b e n , G . V o lc k a e r t
Development of an alternative baculovirus expression system
using an immediate early gene of the baculovirus
of B o m b y x m o r i (IAEA-SM-327/11) ...................................................... . 121
V. V u ls t e k e , M . V a n B r u s s e l , I. J a n s s e n , R . H u y b r e c h t s
Molecular cloning of restriction endonuclease fragments of DNA
isolated from the nuclear polyhedrosis virus of H e l i o t h i s a r m í g e r a
(IAEA-SM-327/12) ..................................................... .................. ............ 125
T. A t t a t h o m , B . I s a n o n t , S . A t t a t h o m
Contrôle génétique des insectes: Caractérisation d’éléments génétiques
mobiles chez les moustiques (IAEA-SM-327/13) ...................................... 133
C . M o u c h e s , J . C . S a lv a d o , N . B e n s a a d i, S. K a r a m a , P . B la n c h a r d
Use of the biolistic technique for gene transfer to mosquito embryos
(IAEA-SM-327/14) ........................................ ............................................ 145
E . M i a l h e , A . L a u g h i n g h o u s e , L .H . M i l l e r
Polymerase chain reaction amplification of RAPD markers to distinguish
populations of the Mediterranean fruit fly, C e r a t i t i s c a p i t a t a
(IAEA-SM-327/15) ...................... ....................................................... ....... 149
D . H a y m e r , D . M c l n n is
Ribosomal DNA of A p h id i u s (Hymenoptera: Braconidae) Nees:
Structure and intraspecific variations (IAEA-SM-327/16) .......................... 153
G . G a r g iu lo , P . V a r r ic c h io , F . P e n n a c c h io ,
F . G r a z ia n i, E . T r e m b la y , C . M a lv a
Mitochondrial DNA variability in geographical populations
of the Brazilian screwworm fly (IAEÀ-SM-327/17) ................................. 161
A .M .L . d e A z e r e d o - E s p i n
Genetic fingerprinting applied to tsetse fly species (IAEA-SM-327/18) ........ 167
A . B l a n c h e t o t , R .H . G o o d i n g
Isolation and preliminary restriction site map of mitochondrial DNA
from C e r a t i t i s c a p i t a t a in Brazil (IAEA-SM-327/19) .................... ........... 175
S .R . M a t i o l i , P . D o s S a n t o s , J . S . M o r g a n t e

GENETICS (Session 3)

Radiation induced chromosome aberrations for the genetic analysis

and manipulation of the Mediterranean fruit fly, C e r a t i t i s c a p i t a t a
(IAEA-SM-327/20) .......................................................................................187
G. F ran z, P. K errem an s
Inheritance of refractoriness to trypanosome infection in tsetse
(IAEA-SM-327/21) .................................................... ..................... ............195
I . M a u d l i n , S .C . W e lb u r n
Active quality control in mass reared melon flies: Quantitative genetic
aspects (IAEA-SM-327/22) ...................................... .................................. 201
T. M i y a t a k e , M . Y a m a g is h i
Search for sex specific genes in the medfly, C e r a t i t i s c a p i t a t a :
Preliminary data on S x l (IAEA-SM-327/23) ..... ........................................ 215
M . F u r i a , D . A r t i a c o , G . S a c c o n e , P . V ito ,
I. P e l u s o , E . G i o r d a n o , L . G . P o l i t o
Transfections of DNA into cultured insect cells and embryos:
Progress and prospects (IAEA-SM-327/24) ................................................ 225
R .A . L e o p o l d
Use of Y linked translocations in locating mutant loci (B l , d p ) on polytene
salivary gland chromosomes of A n o p h e l e s s t e p h e n s i Liston
(IAEA-SM-327/25) ........................................................................... ........ .235
P h a m V a n L a p , A . S. R o b i n s o n
Analysis of structural rearrangements of Lepidoptera chromosomes
using the centrifugation spreading technique (IAEA-SM-327/26) .............. 243
F . M a r e e , W. T rau t
Suitable markers for population genetics
C e r a titis c a p it a t a :
and genome organization analysis (LAEA-SM-327/66) ............................... 251
G . G a s p e r i , A .R . M a l a c r i d a , C .R . G u g l i e l m i n o ,
L . B a r u f f i , C . T o r t i, R . M i l a n i , G . D a m i a n i , C . B a n d i
Comparison of B a c i l l u s t h u r i n g i e n s i s and B a c i l l u s c e r e u s
(IAEA-SM-327/46) ........................................ ...... ................................... 257
A .- B . K o ls t e , C . C a r ls o n

Poster presentation

Molecular genetics of insects: Oogenesis and maternal control

. of early development (IAEA-SM-327/23P) ................................................. 265
C. M a l v a , F . G r a z i a n i

OPERATIONAL PROGRAMMES (Session 4)

Genetic control of cotton insects: The pink bollworm as a working

programme (IAEA-SM-327/28) ........................ ....... ............ .................... 269
R .T . S t a t e n , R .W . R o s a n d e r , D . F . K e a v e n y
Implementation of the sterile insect release programme to eradicate
the codling moth, C y d i a p o m o n e l l a (L.) (Lepidoptera: Olethreutidae),
in British Columbia, Canada (LAEA-SM-327/29) ...................................... 285
V .A . D y c k , S .H . G r a h a m , K .A . B l o e h m
Advances in sheep blowfly genetic control in Australia
(IAEA-SM-327/30) ........................... ........ ............ .................................... 299
G . G . F o s t e r , G . L . W e l le r , W .J . J a m e s ,
K .M . P a s c h a l i d i s , L . J . M c K e n z i e
Erradicación del gusano barrenador del Nuevo Mundo (IAEA-SM-327/31) .. 313
J .M . I r a s t o r z a , C . B a ja t t a , J . O r t e g a , S .J . M a r tin e z
Eradication of the New World screwworm from
the Libyan Arab Jamahiriya (IAEA-SM-327/32) ........................ .............. 319
D .A . L i n d q u i s t , M . A b u s o w a , W . K l a s s e n
Effective control of the Mediterranean fruit fly by genetic sexing male only
sterile insect technique releases during 1989-1990
(IAEA-SM-327/70) .................................................................... ;........... ...331
Y. N it z a n , Y. R o s s l e r , A . P . E c o n o m o p o u l o s
Tsetse sterile insect technique programmes in Africa: Review and
analysis of future prospects (IAEA-SM-327/71) ....................................... 345
E .D . O ffo r i

Poster presentation

Evaluation of Oriental fruit fly control using the sterile insect technique
at Doi Ang Khang (IAEA-SM-327/16P) ........................................... . 359
M . S u ta n ta w o n g , P . K a o c h o n g ,
W. L im o p a s m a n e e , M . L u a n g a p ic h a tk u l

Fj STERILITY AND INSECT BEHAVIOUR (Session 5)

Integration of inherited sterility and other pest management strategies

for H e l i c o v e r p a z e a : Status and potential (IAEA-SM-327/35) .................. 363
J . E . C a rp en ter
Evaluation of the Ft sterility technique for population suppression
of the pink bollworm, Pectinophora gossypiella (Saunders)
(IAEA-SM-327/36) ................................. ................ ................... ....... . 371
Z .A . Q u r e s h i , T. H u s s a i n , N . A h m e d
Evaluation of the potential control o f the European corn borer
( O s t r i n i a n u b i l a l i s Hb.) in the field by radiation induced Fj sterility .
(IAEA-SM-327/37) .......... !........................................................................ 379
1.1. R o $ c a , A l. B a r b u l e s c u

The sterile insect technique and transmission of .inherited sterility

to control the diamondback moth, P l u t e l l a x y l o s t e l l a (L.), and the
cabbage webworm, C r o c i d o l o m i a b i n o t a l i s Zell. (IAEA-SM-327/38) ...... 395
S . S u t r is n o , M . H o e d a y a
Esterilidad en la F[ de D i a t r a e a s a c c h a r a l i s (Fab.), Lepidoptera:
Crambidae: I. Efectos de las dosis subesterilizantes sobre la
reproducción y la competividad (IAEA-SM-327/39) ....................... ......... 405
J . C . G a r c ía , J .R . G o n z á le z
Esterilidad en la Fj de D i a t r a e a s a c c h a r a l i s (Fab.), Lepidoptera:
Crambidae: П. Dinámica de apareamiento y efectos sobre
su descendencia (IAEA-SM-327/56) ........................................................... 419
J .R . G o n z á le z , J .C . G a r c ía
Partial sterilizing radiation dose effect on the Fj progeny
of S p o d o p t e r a l i t u r a (Fabr.): Growth, bioenergetics and reproductive
competence (IAEA-SM-327/40) ................................................................. 427
R .K . S e th , S. S. S e h g a l
Mating behaviour of the melon fly: Sexual selection
and sperm competition (IAEA-SM-327/41) ............................................... 441
Y. I t ô , M . Y a m a g is h i , H . K u b a
First field assessment of the dispersal and survival of mass reared sterile
Mediterranean fruit fly maies of an embryonal, temperature sensitive
genetic sexing strain (IAEA-SM-327/42) ................................................... 453
J . H e n d r i c h s , V. W o m o a y p o m , B . I . K a t s o y a n n o s , K . G a g g l
Recent progress in the development of attractants for monitoring
the Mediterranean fruit fly and several A n a s t r e p h a species
(IAEA-SM-327/43) ........................................................................ ........... 463
R .R . H e a t h , N .D . E p s k y
Effect of the sex ratio on egg collection and hatch of the Mediterranean
fruit fly (Diptera: Tephritidae) (IAEA-SM-327/33) .................................. 473
A . P . E c o n o m o p o u lo s

Poster presentation

Recovery of fertility in irradiated E p h e s t i a c a u t e l l a (Walker)

(Lepidoptera: Phycitidae) (IAEA-SM-327/IP) ........................................... 481
H. М акее

BIOCONTROL (Session 6)

Protein hydrolysate components attractive to tephritids

(IAEA-SM-327/44) ...................................................................................... 487
R . T e r a n i s h i , S . K in t , R .A . F l a t h , D .M . L i g h t ,
S .B . O p p , K .M . R e y n o l d s , W . H a m e r s k y
Ecohormones for the management of fruit fly pests: Understanding
plant-fruit fly-predator interrelationships (IAEA-SM-327/45) .................. 495
K e n g H o n g T an
Control of the Chinese citrus fly, D a c u s c i t r i (Chen), using the sterile
insect technique (IAEA-SM-327/47) ........................................................... 505
H u a s o n g W an g , H e q in Z h a n g
B a c illu s t h u r in g ie n s is endotoxins active against C h i l o p a r t e l l u s
and (IAEA-SM-327/48) ................................ 513
G lo s s in a m o r s ita n s m o r s ita n s
E . O . O s ir , G .N . M a g o m a , M .W . V u n d la , E . K e n y a
Natural factors as potential insecticides (IAEA-SM-327/49) ..... ........... ...... 523
H . B u ed s, C. P y c k e, A. D e L o o f
Quantitative analysis of the fate of a pesticide after its application
to insects (IAEA-SM-327/50) ............................. ................. ............. . 529
R . T ykva, B . B en n etto v á
Use of juvenoids for suppression of insect reproduction
(IAEA-SM-327/51) ...................................... ............................................ 537
F. S eh n al

RESEARCH AND DEVELOPMENT ON THE TSETSE FLY (Session 7)

Age dependent sampling biases in tsetse flies ( G l o s s i n a ): Problems

associated with estimating mortality from sample age distributions
(IAEA-SM-327/52) ..................................................... .............................. 549
J .W . H a r g r o v e
Detection and identification of tissue specific lectins of the tsetse fly,
G l o s s i n a t a c h i n o i d e s : Midgut lectin activity with lipopolysaccharide
binding specificity (IAEA-SM-327/53) ...................................................... 557
P . V o lf, L . G r u b h o f f e r , M . M u í k a
Influence combinée de la basse température et de Г irradiation durant
la dernière phase pupale sur la capacité vectorielle des mâles de
G l o s s i n a p a l p a l i s g a m b i e n s i s (IAEA-SM-327/54) ...................................... 567
A . D jite y e , B . B a u e r , A . V an d e r V lo e d t,
U. F e ld m a n n , M .J .B . V r ey se n
Rearing tsetse flies for use in sterile insect technique vector control
programmes (IAEA-SM-327/55) .......................................................... . 579
U. F e l d m a n n
Hybridization of G l o s s i n a s w y n n e r t o n i with subspecies of G l o s s i n a m o r s i t a n s
(Diptera: Glossinidae): Implications for use of hybrid sterility
and satyrs for genetic control of tsetse (IÀEA-SM-327/57) ............... 603
R .H . G o o d i n g
The various vector control activities of the Centre de recherches sur
les trypansosomes animales (CRTA) (lAEA-SM-327/77) ......................... 619
B . B a u e r , S. A m s l e r

QUARANTINE (Session 8)

Irradiation as a quarantine treatment of agricultural commodities

against arthropod pests (IAEA-SM-327/58) ..... ...................... ........... .6 2 7
. N . W. H e a t h e r
A biochemical marker for detection of irradiated pupae of the ,
Oriental fruit fly (D a c u s d o r s a l i s ) (IAEA-SM-327/60) ................. ...___ ; 641
M .T . Y u l o - N a z a r e a , E . C . M a n o t o
Poster presentation

Irradiation as a quarantine treatment for European red mite

eggs on apples (IAEA-SM-327/19P) ......................................................... 649
S. Ig n a to w ic z

Chairmen of Sessions and Secretariatof the Symposium ............................... 653

List of Participants ............................................... ........................................... 655

Author Index .................................................................................................... 667

Index of Papers and Posters by Number 669

OPENING
(Session 1)

Chairman
W. KLASSEN
FAO/IAEA
 IAEA-SM-327/1

ADVANCES AND TRENDS

IN MANAGING INSECT PESTS

f Rib. RIDGWAY, M.N. INSCOE, '

K.W. THORPE
Beltsville Ágriciiltural Research Center,
Agricultural Research Service,
United Statés Department of Agriculture,
' Beltsviile, Maryland,
United States of America

Abstract

ADVANCES AND TRENDS IN MANAGING INSECT PESTS. •

Providing thè needed food, 1clothing and shelter for the world’s peoplés, while maintain­
ing a long term sustainable nàtural resource base and a healthful environment, will be a major
challenge. One very important áspect of this challenge is the protection of plants and animals
from insect pests. Although many different insect control methods are currently in use, there
has been a heavy reliance on conventional insecticides for more than forty years. Because of
the development of insecticide resistance, the public concern about environmental and health
risks, and the high cost of development, approvals for specific insecticide uses are now being
removed at a faster rate than they are being replaced. Additipnal alternative insect manage­
ment technologies will be required, and a strategy for their use should be selected based on
the characteristics of the. technology to bè used and the pest involved. However, thé basic
methods aváilable for insect control in the future are not likely to change drastically. Included
will be natural enemies, semiochemicals and other natural products, synthetic insect toxicants,
host plant resistance, autocidal methods and cultural controls. Since a number of alternative
pest controls, such as importation of natural enemies, autocidal methods and cultural controls,
usually do not have asignificant market value, the research and. development responsibilities
necessary'to. bring such components into practice will rest primarily with the'public sector.
Also, desirable area-wide management, and eradication programmes often will entail major
involvement of public institutions. Most insect problems can best be dealt with by. a pest
management programme that includes integration of two or more suppression methods with
the aid of a strong population dynamics knowledge base, decision support systems, economic
analyses and environmental assessments. Although trends cannot be predicted with certainty,
the market for conventional insecticides probably will decline and the market for biologically
based products probably'will increase. Thus, àn increased number of biological,suppression
methods will probably be availablé in the future. However, advances in production, storage
and delivery to reduce costs and increase efficacy are needed in order for these methods to
producé the desired-result. Therefore, increased efforts by the public sector, closely
co-ordinated with the'private sector; will likely be necessary to meet future néeds for economi­
cally feasible and socially acceptable pest control products. Additional emphasis should also
be placed on the development of non-product oriented, biologically based control methods.-

3
6 RIDGWAY et al.

4. SUPPRESSION METHODS AND.STRATEGIES

; . Suppression methods can be placed into three broad categories: biological, cul­
tural and chemical [11]. Howeyer,. many methods.cannot be exclusively classed in
a single:category. For instance, Br is a biological organism, but much of its insect
suppression activity is associated with a protein toxin that is a naturally, occurring
chemical. Therefore, an in depth.understanding of the characteristics and properties
of'various suppression methods,.particularly those that are biologically based, is
usually.necessary in order to develop and successfully utilize a suppression method.
The various suppression methods ,to be considered include microbial agents, nema­
todes,, arthropods (augmentation), pheromones, botanicals, insect growth regulators,
synthetic toxicants, host plant resistance, arthropods (importation of exotics), auto-
çidal (genetic) methods and various cultural controls.

FIG. 1. Insect management strategies.

 IAEA-SM-327/1 7

Suppression methods should be utilized within a deliberate strategy. The

articulation of the various strategies has evolved during recent years; these now
include prevention, quarantine and containment, temporary alleviation, management
of localized populations and total population management, including area wide popu­
lation management, reproduction management and eradication (Fig. 1) [12-14]. The
successful implementation of a particular suppression method will be facilitated by
the selection of the most appropriate strategy after careful consideration of the
characteristics of the method to be employed and of the population dynamics of the
insect to be managed.

5. PRODUCT ORIENTED TECHNOLOGIES

A wide range of biologically based product oriented technologies are in limited

use and include products based on B t [15], insect viruses [16], fungi [17], parasitic
nematodes [18] and mass reared arthropods [19]. Other technologies of natural
origin, but more closely associated with defined chemistry, include microbially
produced toxins (other than B t endotoxin) [20], pheromones [21] and botanical insec­
ticides [22]. Other products include synthetic insect growth regulators [23] and
synthetic organic toxicants [10], with the latter comprising the vast majority; of
products currently being sold.

6. NON-PRODUCT-ORIENTED TECHNOLOGIES

A number of insect suppression technologies that have made a major contribu­

tion are not particularly product related. For instance, biological controls with
imported natural enemies may provide considerable public good .[24], particularly
through the control of exotic pests on perennial crops, but these controls do not
usually have significant market value. Therefore, the research and development
responsibilities necessary to bring these components into practice rest primarily with
the public sector. Also, area wide management and eradication programmes [25] will
continue to play a significant role in insect control and often will involve institutions
that are not market oriented. For example, autocidal methods, such as the sterile
insect technique, have had major impacts, but application of these methods usually
requires the involvement of governmental or grower organizations.
Host plant resistance, although not usually considered to involve an insect
control product, often comprises a commercial product if the seed or other plant
material is sold. On the other hand, some seeds and other plant materials can be
propagated by individual growers or by grower organizations. However; regardless
of the source of the resistant plant material, there are many opportunities for reduc­
ing pests through the use of resistant plants [26].
8 RIDGWAY et al.

Finally, cultural controls [27] such as tillage, crop rotation, irrigation and fer­
tilization management, manure management, trap cropping and companion cropping
provide a wide range of options for reducing insect populations, but they do not have
a direct market value. '

7. INTEGRATED PEST MANAGEMENT

7.1. Integrated pest management defined

During the debate in the 1970s and 1980s, when IPM was frequently touted
as the preferred approach to pest control, many different interpretations were placed
on IPM. These ranged from ‘eliminate all pesticides’ to ‘make some changes, but
don’t reduce pesticide market size’ , and often did not include total population
management. However, now, in the early 1990s, two predominant interpretations of
IPM have emerged: (1) improved pesticide use management, which emphasizes
application only when needed, with a wide range of precautions to reduce exposure
of both workers and the environment, and (2) reduction in pest losses and/or pesti­
cide use through expanded use of alternative pest control methods.
Thus, a common goal, which can encompass various insect management strate­
gies designed to suppress insect populations, appears to be evolving and pest
management, as earlier defined by the Council on Environmental Quality [28], can
still provide the framework for moving toward that common goal:

“ Integrated pest management is the selection, integration, and implementation

of pest control methods based on predicted economic, ecological, and socio­
logical consequences. ’ ’

Within this context, regardless of the strategy and methods to be employed, the basic
components ôf an IPM programme are similar. They include: (a) suppression
methods; (b) decision support technologies (sampling, treatment thresholds, decision
guides and population models); and (c) integration into production systems, includ­
ing environmental assessments and economic analyses. In addition, a thorough
knowledge of population dynamics (host/pest interactions, pest density/damage/yield
relationships and management of pests and their natural enemies) is often needed or
useful in designing and implementing IPM programmes. Perhaps the single most
important element common to all strategies is monitoring or sampling. The develop­
ment of pheromones and other attractants for use in traps is a major advancement
in that area [29]. Two examples of insect pest management are presented here to
illustrate local management and reproduction management/eradication programmes.
 IAEA-SM-327/1 9

7.2. Local management

Advanced local management programmes often include the use of computer­

ized decision support, expert or knowledge based systems [30]. For instance, an IPM
programme for apples has been developed that includes a computerized Expert Advi­
sory System for Managing Apple Cropping Systems (EASY-MACS) [31]. This sys­
tem provides a programme for the management of aphids, leafminers, leafrollers,
mites and fruit flies. The system provides for inputs of phenological data, pest
density information and weather data and for outputs on additional sampling recom­
mendations if needed, damage forecasting, treatment recommendations and record
keeping. A companion guide for sampling has also been developed which includes
sampling data forms that enable the grower to make decisions without access to the
computerized expert system [32].

7.3. Reproduction management/eradication

A number of very important programmes involving autocidal methods for

eradication of both dipteran and lepidopteran pests are discussed elsewhere in these
Proceedings, so a programme for reproduction management and eradication of a
coleopteran pest, the boll weevil, A n t h o n o m u s g r a n d i s , the Southeastern Boll Weevil
Eradication Program in the USA, has been selected here for illustration. The compo­
nents of the programme include: (a) pheromone trap sampling for treatment deci­
sions; (b) fall diapause control with insecticides; (c) selected presquaring (flower
bud) control with insecticides; (d) fall diapause control as needed; and (e) pheromone
trap monitoring to detect reinfestation. The programme, with these components,
eliminated boll weevil reproduction in North Carolina and essentially eliminated it
in eastern South Carolina between 1978 and 1986 [33]. In 1987, the programme was
expanded throughout Florida, most of Georgia and a major portion of Alabama.
Over 400 000 ha (1 million acres) were involved throughout the southeastern USA
in 1991 [34].

8. TRENDS

Worldwide numerical data available for analysing insect management trends

are limited primarily to information on sales of insecticides and other pest control
methods that have a market vàlue. However, these data do provide some insight into
overall trends. For instance, the rapid growth in pesticide sales that occurred in the
1960s (11%) slowed in the 1970s (7%) and 1980s (3%) [7]. Also, a recent market
study on insect control products indicates that the conventional insecticide market in
the USA and the world will likely decline over the next ten years and that the market
for biologically based products will increase significantly (Table II) [35, 36].
10 RIDGWAY et al.

TABLE П. ESTIMATED AND PROJECTED SALES OF INSECT CONTROL

PRODUCTS AT THE USER LEVEL (FROM REFS [35, 36])

US market (millions of constant US dollars)

Product group 1991 1996 2001

Microbial agentsa,b 79 118 232

Nematodes0 3 ' 15 35
Árthropodsc,d 8 12 18
Behaviour modifying chemicals15 30 45 111
Botanicalsb 45 51 55

Subtotal 165 241 451

Speciality chemicalsb,e 22 ' 30 57
Conventional insecticides1* 2 119 2 021 1 600

Total 2 306 2 292 2 108

World market (millions of constant US dollars)

Product group 1991 1996 2001

Microbial agentsab , 157- 219 381

Nematodesc 4 30 70
Arthropodsc,d 35. 47 60
Behaviour modifying chemicalsb 60 80 158
Botanicalsb 70 81 90

Subtotal 326 • 457. 759

Speciality chemicalsb,c 100 116 181
Conventional insecticidesb 9 358 9 597 9 212

Total 9 784 10 170 10 152

a Bacteria, viruses and fungi, and toxins produced by these organisms.

b Estimates from Ref. [35].
c Estimates from Ref. [36].
d Includes primarily predators and parasitoids; some pollinators for use in glasshouses are
also included. (
e Includes primarily insect growth regulators.
 IAEA-SM-327/1 11

Although the use of biological insect control products is likely to increase, their cost
and the complexity of their use will likely limit the extent to which they will replace
conventional insecticides. In addition, the extent to which new insecticidal chemis­
tries become available will significantly influence the rate of increase in the use of
biologically based insect suppression methods. Several hew organic synthetic insecti­
cides and miticides from several chemical classes, such as pyrroles, nitroguanidines,
phenylpyrazoles, dibenzoylhÿdrazines, substituted triazones and quinazolines, are in
various stages of development.
The differences in the total values for insect control product sales in Tables I
and П should be noted. Even though both values are at the end user level, the values
in Table I are calculated on the basis that all materials are sold at the farm level and
the values in Table II include some sales at the consumer and professional pest con­
trol levels. Also, estimates in Table II on projected sales for arthropods and nema­
todes are from two different sources that used different criteria for the projections.
Therefore, the estimates for arthropods are more conservative than the estimates for
nematodes.
In addition to changing trends in the use of insect control products, there.are
numerous examples of the application of alternative insect control methods that are
not product oriented which have resulted in reductions in insecticide use [37], even
though quantitative worldwide data on the extent of use of these methods are not
available.

9. OPPORTUNITIES FOR BIOLOGICALLY BASED METHODS

Conventional insecticides have received wide acceptance because they provide

a means of rapidly reducing pest populations in highly manipulated agroecosystems
in which natural regulation of pest populations does not maintain populations at
acceptable levels. If biologically based pest controls are to be substituted for conven­
tional insecticides, they need to be'modified and/or managed in siich a manner as
to produce a similar result either by rapidly reducing a pest population or by prevent­
ing it from reaching damaging levels.
Therefore, the perception that biological pest controls are readily available,
easy to use and effective, and that their use in lieu of conventional insecticides should
therefore be simple and rather straightforward, is not consistent with actual circum­
stances. The characteristics generally associated with biologically based products,
which often require different approaches to development and delivery compared with
conventional insecticides, include: (a) narrow spectrum of activity; (b) short shelf
life (c) slow action; (d) lower efficacy; (e) relatively small market size and fewer
opportunities for proprietary protection. Although it is important to be cognizant of
these characteristics, there are opportunities for overcoming some of these limita­
tions. For instance, in vitro virus production [38] and improved artificial diets [39]
12 RIDGWAY et al.

offer considerable potential for reducing the costs of virus and insect production.
Further; chemical enhancers are showing considerable promise for increasing the
effectiveness of insect viruses [40]. Also, nutrient based phagostimulants can be used
to improve the efficacy of microbial agents [41], while some semiochemical baits
hold promise for controlling insects with very small quantities of insecticide [42].
Also, the development of total population management programmes would favour
the use of biologically based suppression methods [25].
Transgenic plants may also be a significant component in agricultural product
sales in the future, but their role in insect control continues to be uncertain because
of the time required to incorporate new traits and the concern about their potential
to increase the rate of resistance development. However, the great increase in our
ability to manipulate genes and, therefore, to transfer various traits from one
organism to another, likely will greatly increase our ability to utilize various genetic
traits. Therefore, the biologically based methods available in the future probably
will include increased numbers of genetically modified organisms or products of
such organisms [14, 43]. See related papers by Oakeshott1 and others in these
Proceedings. ,

10. VALUE COMPONENTS

Economic analyses and environmental assessments should be an integral part

of programme development and these analyses and assessments, coupled with social
and political considerations, should have a major influence on insect control deci­
sions, since almost all pest control decisions affect people other than those who make
the decisions. The value judgements made will vary greatly from country to country,
so no attempt will be made here to review the total process. However, an example
can be used to illustrate the value of obtaining applicable quantitative data on the
impacts of pest control actions. An economic study of the US Southeastern Boll
Weevil Eradication Program, discussed in Section 7.3, which included an analysis
of the numbers of insecticide applications, cotton acreage and cotton yields, indicated
that the programme in North Carolina, Virginia and South Carolina increased annual
returns by about US $ 187/ha [33]. Also, the 50-70% reduction in insecticides
associated with this programme represents a substantial and continuing environmen­
tal benefit. These results were particularly useful in guiding programme expansion.

1 Paper IA E A -S M -327/ 2..

 IAEA-SM t327/1 13

11. CONCLUSIONS

If current trends continue, a substantial increase in biologically based methods

of insect control will be needed to replace conventional insecticides that are being
lost. However, because of the characteristics of most alternative technologies,
increased efforts by the public sector will likely be necessary to meet future needs
for economically feasible and socially acceptable pest control products. Additional
emphasis should also be placed on the development of non-product oriented bio­
logically based pest control methods.

REFERENCES

[1] GORE, A ., Earth in the Balance: Ecology and'the Human Spirit, Houghton Mifflin,
Boston, MA (1992).
[2] PIMENTEL, D ., “ Introduction” , CRC Handbook of Pest Management in Agriculture
(PIMENTEL, D ., Ed.), Vol. 1, CRC Press, Boca Raton, FL (1991) 3-11.
[3] CRAMER, H .H ., Plant protection and world crop production, Pflanzensch. Nachr. 20
(1967) 1-524.
[4] METCALF, R.L., “ Introduction” , Entomology Serving Society: Emerging Techno­
logies and Challenges (VINSON, S.B., METCALF, R.L., Eds), Entomological Soci­
ety of America, Lanham, MD (1991) 1-6.
[5] CARSON, R., Silent Spring , Houghton Mifflin, Boston, MA (1962).
[6] PERKINS, J.H., Insects, Experts, and the Insecticide Crisis: The Quest for New Pest
Management Strategies, Plenum Press, New York (1982).
[7] COUNTY NATWEST WOODMÀC, Business Consultancy Unit, Edinburgh, United
Kingdom, personal communication, Oct. 1992.
[8] GEORGHIOU, G.P., LAGUNES-TEJEDA, A ., The Occurrence of Resistance to
Pesticides in Arthropods, FAO, Rome (1991).
[9] Enhancing the Safety of America’s Food Supply (Proc. Round Table, Washington, DC,
1991) (RIDGWAY, R.L., INSCOE, M .N ., ROBENS, J.F., Eds), Riley Memorial
Foundation, Lanham, MD (1992).
[10] WILLIAMS, C .S., “ Urban pesticides — 2000: The outlook for conventional pesti­
cides” , Annual Meeting of the Chemical Specialties Manufacturers Association, Fort
Lauderdale, FL (Dec. 1991) (oral presentation).
[11] GARCIA, R., et al., Comments on a redefinition of biological control, BioScience 38
(1988) 692-694.
[12] RABB, R.L., “ Principles and concepts of pest management” , Implementing Practical
Pest Management Strategies (Proc. National Extension Insect-Pest Management Work­
shop, West Lafayette, 1972), Purdue University, West Lafayette, IN (1972) 6-29.
[13] KLASSEN, W ., “ Concepts of pest management” , Introduction to Crop Protection
(ENNIS, W .B ., Jr., Ed.), Crop Science Society of America, Madison, WI (1979)
403-419.
14 RIDGWAY et al.

[14] RIDGWAY, R.L., et al., “ Implementation of emerging technologies for management

of insect pests” , Entomology Serving Society: Emerging Technologies and Challenges
(VINSON, S.B., METCALF, R.L., Eds), Entomological Society of America, Lan-
ham, MD (1991) 392-424.
[15] CARLTON, B.C., “ Genetic improvements of Bacillus thuringiensis as a bioinsecti­
cide” , Biotechnology, Biological Pesticides, and Novel Plant Pest Resistance for Insect
Pest Management (Proc. Symp. Ithaca, 1988) (ROBERTS, D .W ., GRANADOS, R.R.,
Eds), Boyce Thompson Institute for Plant Research, Ithaca, NY (1988) 38-43.
[16] FUXA, J.R., “ New directions for insect control with baculoviruses” , New Directions
in Biological Control: Alternatives for Suppressing Agricultural Pests and Diseases
(Proc. UCLA Symp. on Molecular and Cellular Biology, Frisco, CO, 1989)
(BAKER, R.R., DUNN, P.E., Eds), Alan R. Liss, New York (1990) 97-113.
[17] McCOY, C .W ., “ Entomogenous fungi as microbial pesticides” , ibid., pp. 139-159.
[18] Entomopathogenic Nematodes in Biological Control (GAUGLER, R., KAYA, H.K.,
Eds), CRC Press, Boca Raton, FL (1990).
[19] VAN LENTEREN, J.C., Implementation of biological control, Am. J. Altern.
Agrie. 13 (1988) 102-109.
[20] LASOTA, J.A., DYBAS, R .A ., Avermectins, a novel class of compounds: Implica­
tions for use in arthropod pest control, Annu. Rev. Entomol. 36 (1991) 91-117.
[21] Behavior-Modifying Chemicals for Insect Management: Applications of Pheromones
and Other Attractants (RIDGWAY, R.L., et al., Eds), Marcel Dekker, New York
0990). ,
[22] Insecticides of Plant Origin (ARNASON, J.T., et al., Eds), ACS Symposium Series
No. 387, American Chemical Society, Washington, DC (1989).
[23] MIYAMOTO, J., et al., “ Newer insect growth regulators for pest control” , Proc.
203rd ACS National Meeting San Francisco, СA , 1992, American Chemical Society,
Washington, DC (1992) AGRO 158 (abstract only).
[24] LAING, J.E., HAMAI, J., “ Biological control of insect pests and weeds by imported
parasites, predators, and pathogens” , Thepry and Practice of Biological Control
(HUFFAKER, C.B., MESSENGER, P.S., Eds), Academic Press, New York (1976)
685-743.
[25] KNIPLING, E.F., The Basic Principles of Insect Population Suppression and Manage­
ment , USDA Agriculture Handbook No. 512, United States Government Printing
Office, Washington, DC (1979).
[26] Breeding Plants Resistant to Insects (MAXWELL, F.G., JENNINGS, P.R., Eds),
Wiley, New York (1980).
[27] COAKER, Т.Н ., “ Cultural methods: The crop” , Integrated Pest Management
(BURN, A.J., et al., Eds), Academic Press, London (1987) 69-88.
[28] BOTTRELL, D.R., Integrated Pest Management, Council on Environmental Quality,
Washington, DC (1979). .
[29] INSCOE, M .N ., et al., “ Commercial availability of insect pheromones and other
attractants” , Behavior-Modifying Chemicals for Insect Management: Applications of
Pheromones and Other Attractants (RIDGWAY, R.L., et al., Eds), Marcel Dekker,
New York (1990) 631-715,
 IAEA-SM-327/1 15

[30] CROFT, B .A ., GUTIERREZ, A .P ., “ Systems analysis role in modeling and decision­

making” , Entomology Serving Society: Emerging Technologies and Challenges
(VINSON, S.B., METCALF, R.L., Eds), Entomological Society of America,
Lanham, MD (1991) 298-319.
[31] HUBER, B., et al., Development of a knowledge-based system supporting IPM deci­
sion making in apples, Comput. Electron. Agrie. 4 (1990) 315-331.
[32] AGNELLO, A ., et al., Simplified Integrated Management Program: A Guide for Apple
Sampling Procedures in New York, New York State Agricultural Experiment Station,
Geneva, NY (1991).
[33] RIDGWAY, R.L., et al., “ Role of the boll weevil pheromone in pest management” ,
Behavior-Modifying Chemicals for Insect Management: Applications of Pheromones
and Other Attractants (RIDGWAY, R.L., et al., Eds), Marcel Dekker, New York
(1990) 437-471.
[34] Insect Pheromones and Other Behaviour-Modifying Chemicals: Applications and Regu­
lation (RIDGWAY, R.L., et al., Eds), Monograph No. 51, British Crop Protection
Council, Farnham, Surrey, UK (1992).
[35] SRI INTERNATIONAL, Agribusiness Consulting Group, Health and Performance
Chemicals Center, Menlo Park, CA, personal communication, Aug. 1992.
[36] ANONYMOUS, USA estimates on beneficial arthropod market based on anonymous
sources and general knowledge; world estimates on beneficial arthropod market based
on reports in the Wall Street Journal and projections by an anonymous source; estimates
on nematode market based on an anonymous source but substantially modified, 1992.
[37] Food, Crop Pests, and the Environment: The Need and Potential for Biologically Inten­
sive Integrated Pest Management (ZALOM, F.G ., FRY, W .E ., Eds), APS Press,
St. Paul, MN (1992).
[38] LYNN, D .E ., et al., Comparative replication of Lymantria dispar nuclear polyhedrosis
virus strains in three continuous-culture cell lines, Appl. Environ. Microbiol. 55 (1989)
1049-1051.
[39] LEPPLA, N .C ., ASHLEY, T.R ., Quality control in insect mass production: A review
and model, Bull. Entomol. Soc. Am. 35 4 (1989) 33-44.
[40] SHAPIRO, М ., Stilbene fluorescent brighteners as radiation protectants and activity
enhancers for insect pathogens, US Patent Application No. 609,848, filed 1/7/1990,
US Patent Office, Washington, DC (1990).
[41] FARRAR, R .R ., Jr., RIDGWAY, R .L ., Comparative studies of the effects of nutrient-
based phagostimulants on six lepidopterous insect pests (in preparation).
[42] METCALF, R.L., METCALF, E.R., “ Plant kairomones in insect ecology and
control” , Contemporary Topics in Entomology 1, Chapman and Hall, New York
(1992).
[43] KLINE AND COMPANY, INC., The US Pesticide Market, 1989: Abstract, Fairfield,
NJ (1990).
 IAEA-SM-327/2

BIOTECHNOLOGICAL PROSPECTS
FOR MANAGING INSECT PESTS

J. OAKESHOTT, P. CHRISTIAN, P.W. ATKINSON

Division of Entomology,
Commonwealth Scientific and Industrial
Research Organization,
Canberra ACT,
Australia

Abstract
BIOTECHNOLOGICAL PROSPECTS FOR MANAGING INSECT PESTS.
Mounting problems with resistance and residues threaten the long term utility of many
chemical insecticides and drive the search for biotechnological alternatives. The potential
impact that molecular and other biotechnologies may have on three pest control strategies,
involving biological insecticides, insect resistant hosts and genetically engineered insects, is
discussed. Regarding biological insecticides, it is argued that recent improvements in produc­
tion technologies and relatively low development costs should allow exploitation of a number
of non-engineered microbials, generally for situations involving relatively high damage
thresholds where speed of action is not the major priority. The much greater development
costs associated with engineered microbials may restrict their use in the short term to situations
involving large markets in the developed world. Nevertheless, there are now excellent
prospects for genetically engineering insect viruses to achieve kill times approaching those of
many chemicals. Likewise, genetic engineering now permits the transfer of various insectici­
dal delta endotoxin genes from Bacillus thuringiensis (Bt) to other bacteria with different eco­
logical niches. Concerning insect resistant hosts, the prospects of developing vaccines against
several haematophagus animal ectoparasites such as ticks are now good, but genetic engineer­
ing offers the possibility of heritable protection against a wider range of pests in both animals
and plants. The basic engineering technology is now available for many production animals
and non-cereal crops, but major difficulties remain in developing an engineering technology
for cereals; achieving adequate control of transgene expression in both plants and animals; the
shortage of orally active alternatives to Bt genes to engineer into plants; and the lack of any
efficacious insecticidal genes to engineer into animals. With regard to genetically engineered
insects, two of three basic components needed to generalize the existing capability for
Drosophila transformation to other insects are now available. These are marker genes to iden­
tify transformants and promoters to drive the expression of foreign genes in these insects. The
third component is a transposable element system to act as a vector for transferring foreign
DNA into the recipient’ s genome. This component has proved more problematic, but recent
developments indicate that the hobo or mariner elements could function as such vectors for
non-drosophilid transformation. The resulting ability to analyse specific genes within pest
species will greatly enhance the ability to use these genes, or mutant alleles of them, as the
basis for future control strategies.

17
18 OAKESHOTT et al.

1. INTRODUCTION

Several problems are now emerging with chemical insecticides which

jeopardize their continued use and efficacy. Perhaps the most important of these has
been the evolution of target pest resistance; over eight hundred cases are now
reported and no major class of chemicals has proved immune to the problem. A
second issue arises from their broad spectrum activity; while of value in many
multipest situations, there are also many untoward effects on beneficial species. With
such chemicals, implementation of insect pest management (IPM) programmes often
becomes impractical and problems with secondary pests can also arise. Moreover,
the persistence of chemical residues both in commodities and in the environment can
present real or perceived downstream problems for human health. On top of all these
problems, new chemical insecticides are becoming more expensive to develop and
the probability of finding new classes may be diminishing.
As the problems with chemical insecticides have mounted, so has the pressure
to develop biotechnological alternatives. This pressure intensified with the advent of
recombinant DNA technologies in the early 1970s, but two decades later very few
biotechnological alternatives have yet become available, although several may do so
within the next decade. It is therefore timely to assess the prospects for these alterna­
tives. In this paper, we argue that the alternatives under development do have the
potential to deliver effective control, but only if their development and management
incorporate the lessons learned from the misuse of chemical insecticides.
In the following sections, we deal with three major biotechnological
approaches that should bring tangible benefits over the next 20 years. The first of
these involves the development of biological insecticides, which could be used like
conventional chemicals and may or may not be genetically engineered. The second
involves the development of hosts carrying in built protection against insect attack,
which could be achieved in non-heritable form by vaccination or in heritable form
by the engineering of either plants or animals with protective genes. The third
approach involves genetic manipulation of the insects themselves, which could lead
to biotechnological versions of genetic control programmes, but could have other
applications as well.

2. BIOLOGICAL INSECTICIDES

2.1. Non-engineered biological insecticides

Most insect species are susceptible to infection by a range of viruses, bacteria,

protozoans, fungi and nematodes. However, few of these agents will meet all the
criteria for use as a microbial insecticide. Some of the salient criteria concern
pathogenicity and speed of action, host range and safety issues, ability of the
 IAEA-SM-327/2 19

microorganism to be máss produced and stability under realistic storage and applica­
tion conditions...............
The majority of microbials have only limited pathogenicity and/or speed of
action, presumably because they have evolved together with their host as a balanced
host-parasite system. In some cases, these problems may be remedied by genetic
engineering technologies and we discuss these in Section 2.2. In other pest control
situations, factors such as slow speed of action are less of a problem because the host
has a relatively high damage threshold and the losses incurred between application
and control can be tolerated. In the latter situations at least, there are significant
prospects of finding a pathogen that is sufficiently active in its wild type form to
provide effective control. ,
The next set of criteria that determine the suitability of a pathogen as a biologi­
cal insecticide concerns host range and safety issues. Before any field trial of a candi-,
date pathogen can be contemplated it is necessary to have some information on these
two issues. Most entomopathogens have fairly restricted host ranges and for some
groups, such as baculoviruses, there are now extensive data to show that this range
is limited at the broadest to the ordinal level [1].; At least for many non-engineered
forms of such entomopathogens, their vertebrate safety can generally be demon­
strated with only,-the.minimum of acute toxicity testing. However, this will, not be
true, for other groups of entomopathogens. For instance, the type virus of the
Nodaviridae, Nodamura virus, is capable of replicating after inoculation into the
brains of suckling mice [2]; obviously, any attempt to use a nodavirus as an insecti­
cide would require more extensive safety testing before release. Such safety.testing,
can form a significant part of total development costs.
The next criterion that a new insecticide must meet concerns the means of its
mass production, which could entail the use of either living insects or artificial cul­
ture systems (in vivo or in vitro systems, respectively). In vivo systems have
predominated, historically, but they are only possible when a suitable host can be
easily reared and even then problems with quality control can occur [1]. One
example of a successful in vivo method concerns the baculovirus of the soybean
looper,■ A n t i c a r s i a g e m m a t a l i s , , which was produced under field conditions by a
government agency in Brazil. A different in vivo production system was used,by
Sandoz in the United States of America to produce a viral insecticide marketed under
the trade name of ELCAR™ for use against H e l i o t h i s and H e l i c o v e r p a species.
Instead of infecting insects that had naturally infested field crops like the Brazilian
system, a dedicated large scale ‘caterpillar factory’ was constructed to provide larvae
for infection with the virus.
Nevertheless, there remain few cases where in vivo production systems have
proved adequate for large scale and/or commercial production. Indeed, the success­
ful commercial development of a biological insecticide has often been predicated on
the development of a reliable, cost effective in vitro production system. The lack of
such a system has been a major reason why virtually no protozoan insecticides have
20 OAKESHOTT et al.

been commercialized. Conversely, recent improvements in in vitro production sys­

tems have been important to the commercialization of the bacterium B a c i l l u s t h u r in ­
g i e n s i s ( B t ) , various nematodes directed against horticultural pests and fungi such as
M e t a r h i z i u m [ 3]. As yet, no viral insecticides have been produced commercially
in vitro, although recent interest in the large scale production of pharmaceutical pep­
tides using baculovirus expression systems in insect cell culture has led to qualitative
improvements in the large scale production of baculoviruses and now makes the cost
effective production of baculovirus insecticides in vitro a feasible proposition [4].
Another crucial criterion which a candidate biological insecticide must meet is
its stability under realistic conditions of storage and use. Formulation technology for
biologicals has generally lagged behind that for chemical insecticides and remains
limiting for some fungi and nematodes with critical ecological requirements. On the
other hand, the intense recent interest in B t and engineered baculoviruses has led to
improvements in their formulation, which now enable their use under broadly com­
parable application technologies to those used for conventional chemicals.
The future for some non-engineered biological insecticides therefore looks
extremely promising. Many have economic levels of pathogenicity and kill times for
particular situations, safety testing for some can be considerably cheaper than for
broad spectrum chemicals, and production technologies are rapidly improving for
some bacteria, fungi and viruses, as are the formulations for bacteria and viruses.
In some instances, like B t , the markets will be large enough that they will be devel­
oped and produced by large multinational companies. However, since development
costs can be relatively small compared with chemicals and some engineered biologi­
cals, some will also be suitable for smaller ‘niche’ markets where development and
production could be undertaken by government agencies and/or small companies.
As more biological insecticides become available it will become critical to
learn the lessons from the misuse of chemicals. Problems of B t resistance have
already emerged in the field for the Indian mealmoth, almondmoth [5] and diamond-
back moth [6] and laboratory studies indicate a high probability that it will quickly
emerge in other major pests such as the Colorado potato beetle and H e l i o t h i s [7].
Unless the use of microbial B t is integrated into a soundly based resistance manage­
ment strategy, the exciting prospects for crop plants engineered with B t toxin genes
may be stymied before they even become commercially available.

2.2. Engineered biological insecticides

For those instances in which wild type microbials are incapable of delivering
acceptable levels of control (e.g. in high value crops with low damage thresholds
such as cotton), genetic engineering may be required to accelerate their pathogenic
effects. The technology has two parts: the ability to engineer the agent and a gene
encoding an insecticidal protein with which to engineer it.
 IAEA-SM-327/2 21

At present, the most advanced genetic engineering technologies for entomo-

pathogens have been developed for insect viruses and, in particular, the baculo­
viruses. Over five hundred isolates of baculoviruses have been recovered from a
wide range of insects, albeit mostly lepidopterans. These viruses have large, double
stranded DNA genomes and some are readily grown in cell culture, enabling the
development of an engineering technology analogous to those previously developed
for several mammalian viruses [8]. This technology was originally based on the
promoter that drives the expression of polyhedrin, a structural protein produced late
in the viral replicative cycle in very large amounts. The high level of expression from
this promoter makes it attractive for producing insecticidal proteins, but the late tim­
ing of its expression limits the improvement in kill time over wild type virus. While
40% improvements have been achieved with one toxin [9], there is also intense
interest in developing alternative promoters that are active earlier in the viral replica­
tive cycle.
There are also other groups of insect viruses that should be amenable to genetic
engineering [10]. For instance, the entomopoxviruses and iridoviruses are large
double stranded DNA viruses and for some of them there are good cell culture
systems. Neither of these virus groups are particularly pathogenic in their wild type
forms, so engineering will be important for their exploitation. Moreover, these
viruses infect many orthopteran and dipteran species for which there are no
baculoviruses known. It is now also possible to engineer several genera of bacteria
and much effort is currently being invested in the transfer of toxin genes from B t
to other bacteria with different environmental stabilities (e.g. pseudomonads) or eco­
logical niches (e.g. cyanobacteria) [11]. The prospects for engineering other non-
viral pathogens are not as clear; fungi, nematodes and protozoans have much larger
and more complex genomes and their life histories and modes of action would make
the manipulation of their kill times problematic. As discussed above, many have
other problems associated with their production and formulation which may limit
their potential anyway.
The second part of the technology for developing an engineered biological
insecticide is the gene to be engineered into the pathogen to increase its control
potential. Broadly speaking, two classes of genes are considered to have potential
for insertion into viruses. The first of these are neurotoxins and the second are genes
whose products disrupt hormonal functions. Two neourotóxin genes haVe been
engineered into the baculovirus from A u t o g r a p h a c a l i f o m i c a , one encoding a toxin
from the straw itch mite and the other encoding a venom protein from a scorpion
[9, 12]; improvements in kill time of 40% and 25%, respectively, were obtained.
Two hormone related genes, encoding the juvenile hormone esterase [13] and
diuretic hormone [14], have also shown some promise in baculoviruses, but further
work is required on these systems before their efficacy compares with the toxins
above. ■
22 OAKESHOTT et al.

None of the neurotoxin or hormone related genes mentioned above are suitable
for engineering into bacteria which, unlike viruses, cannot deliver gene products
beyond the gut barrier. For bacteria, the only toxins currently available are from the
B t delta endotoxin group. These act directly on the midgut cells of a number of
insects, causing cessation of feeding, lysis of the midgut cells and eventual death.
Delta endotoxin genes specific for many lepidopterans, some coleopterans and dipte-
rans are now available for engineering into various other bacteria [15].
Obviously, an engineered biological insecticide must meet similar downstream
criteria.regarding the safety, production and formulation as a wild type biological
insecticide. In many respects; the technologies for production and formulation are
the same as those used for the wild type forms. However, the safety issues are much
more complex for a genetically modified organism, (GMO). Many countries now
have a legislative framework to govern the release of GMOs and the requirements
relating to safety are more stringent than for non-engineered biologicals. For
instance, host range studies need to be more comprehensive to ensure that the
engineering.has not significantly altered the host range. Environmental fate studies
must also assess not only the way in which the organism disperses through the
environment, but also whether the introduced gene can be transferred to other organ­
isms. For some agents, it is likely that additional technology will require develop­
ment to restrict their persistence and recombination potential. These clear needs for
extra safety data and possible requirements for additional technology will add
substantially to the development costs of such engineered biological insecticides.
Consequently, they are less likely to be produced for small markets or by small
companies than non-engineered alternatives, and are less likely to be developed for
pest problems in developing nations in the short term.
The prospects of. pests evolving resistance to engineered microbials have so far
received much less attention than they have for the engineered hosts discussed below.
This is a problem that needs urgent study, since several engineered viruses are well
down the route to commercialization. Basic issues such as whether resistance would
arise to viral functions or to the insecticidal proteins they express have important
long term consequences but have so far not been explored in any depth.

3. BIOTECHNOLOGICAL APPROACHES TO HOST RESISTANCE

3.1. Vaccines

Vaccination technology is a major component of disease control strategies in

both humans and production animals. There is now, great interest in adapting this
technology to provide at least short term protection against some ectoparasitic pests
such as ticks and miasis flies. This approach will generally be more attractive for
ectoparasites that cause production losses themselves than for insects which are
 IAEA-SM-327/2 23

vectors for microbial pathogens. For the latter, it will generally be more appropriate
to vaccinate directly against the microbe rather than its vector.
Successful adaptation of vaccine technology to control ectoparasitic insects
clearly requires isolation of antigens which are capable of stimulating a vigorous
immunological response to some aspect of the insect’ s physiology that is accessible
via the digestive tract. For some haematophagus parasites such as ticks which have
highly porous guts, ingestion of the antibody may give access to many internal
organs; for other species such as miasis flies, gut porosity will generally be limited
and the only accessible organ will be the digestive tract itself. Against this back­
ground, it is not surprising that the first ectoparasitic vaccines developed have been
against ticks [16], although some progress has now also been made using essential
gut antigens as the basis for vaccines against blowflies [17].
Modern biotechnologies could play several roles in the development of
vaccines against ectoparasites. One lies in the identification of protective antigens
and the cloning of cognate genes. A second lies in the expression of these genes in
large scale in vitro expression systems to produce economic quantities of the anti­
gens. Both of these functions have enabled the development of the tick vaccines dis­
cussed above. Eventually, a third role for molecular biology in this area could also
arise as a result of emerging antibody library technologies [18]. These could allow
genes for the host’ s antibodies to be be cloned and then engineered into the same,
or another, host to confer long term and heritable protection. Particularly noteworthy
here is the prospect of engineering such genes into other hosts; coupled with the
recent demonstration that engineered plants can express functional antibodies [19],
it now becomes possible to contemplate the transfer of protective antibody genes
from laboratory animals into crop plants.

3.2. Engineered host resistance

Classical genetic procedures have been deployed with variable success for
decades in efforts to build protection against insect pest attack into commercial varie­
ties of plants and animals. The aim has been to produce varieties with sufficient in
built protection that no further intervention is required to produce economic levels
of control. In theory, the further attraction of genetic engineering solutions in this
area is that protective genes which do not occur naturally within the species can be
introduced into a commercial variety in a single generation, without the need for
extensive wide crossing and back crossing programmes. As was the case with the
engineered microbials above, progress towards this aim requires both the technology
to engineer foreign genes into the organism and the availability of cloned insecticidal
genes.
A basic engineering technology has now been developed for several production
animals and broad acre crops. For production animals, the technology relies on
24 OAKESHOTT et al.

injection of the foreign gene located on a vector into the developing embryo and sub­
sequent transplantation of the embryo [20]. For plants, the technology has generally
used the Ti plasmid from the bacterium A g r o b a c t e r i u m t u m e f a c i e n s as the transfer
vector [21], while the host tissues used more protoplasts. A g r o b a c t e r i u m t u m e f a c i e n s
has a limited host range which does not include many important crops such as
cereals, but recent advances in ‘ballistic’ technology, in which DNA is ‘ shot’ into
the plant cell, have led to the transformation of some cereals not tractable to
engineering with the Ti element [22]. On the other hand, problems still remain in
regenerating many species or varieties from transformed protoplasts.
Even when the basic engineering technology is available for commercial varie­
ties, there are additional problems in adapting them for the expression of currently
available insecticidal insert genes. One of these problems concerns the level and
specificity of expression of the transgene product,'which depend in large part on
promoter technology. For plants, some high level promoters are now available
which, together with enhancements in the post-transcriptional aspects of expression
such as RNA processing and codon usage, achieve concentrations of transgene
products that exceed 0.1% of the total cellular protein [23]. Unfortunately, these
high level promoters are all systemic in their expression profiles and there are con­
cerns about the public acceptability of commodities expressing high amounts of
insecticidal proteins. However, currently available tissue specific promoters gener­
ally do not express at sufficient levels to be useful in this context. Some parallel
problems have been met in the search for appropriate animal promoters, although
some wound response and sweat gland promoters that express at relatively high
levels are now under development [24].
Another problem in applying engineering technologies to the insertion of insec­
ticidal genes relates to the specificity of the integration event. For no commercial
crop or production animal is it yet possible to target the insertion of the foreign gene
to a particular location in the host genome. This unpredictability means that a certain
proportion of insertion events will either inhibit adequate expression of the inserted
foreign gene or disrupt the functioning of an important host gene. This latter problem
has already proved a significant impediment to the introduction of many desirable
traits into both plants and animals.
While not dismissing some of the above problems, the major limitation in
attempts to engineer pest protection into plants and animals is the shortage of cloned
insecticidal genes. These genes must encode insecticidal proteins that are orally
active and, as we discussed above, for most insects this means that they must disrupt
the function of the digestive tract. Currently, the only group of genes that satisfies
these criteria are those encoding B t delta endotoxins. B t toxin genes have now been
expressed in many different plant species, providing effective protection from certain
lepidopteran and coleopteran pests in the glasshouse, and in some cases in field trials
as well. On the other hand, no currently identified B t toxins are suitable for engineer­
ing into animals, for which many of the insect pests are members of the Diptera.
 IAEA-SM-327/2 25

Some currently available B t endotoxins display activity against particular dipteran

pests such as mosquitoes and blackflies but, even then, their toxicity is too low for
use in engineering pest reisistant animals.
Clearly, alternatives to B t toxin genes are urgently needed, both in the event
of resistance arising to fir and to broaden the control prospects of the strategy against
pests not susceptible to B t . Some plant protease inhibitor genes show promise for use
in plants [25], as do certain chitinases for animals [24], and we discussed the possi­
bilities of antibody genes earlier. Another exciting prospect suggested by
Sivàsubramanian et al. [26] is to modify the mode of action of B t delta endotoxins
themselves. These toxins bind to a gut membrane receptor in a specific manner, the
binding moeity of the toxin being located at its carboxy terminal. Sivasubramanian
et al. [26] suggest that changing the binding moeity could change the specificity of
the toxin. They constructed a chimeric B t toxin composed of the amino terminal
portion (the toxin moiety) of a B t toxin and a carboxy terminus derived from the g p 6 7
gene of a baculovirus. They reported that the chimeric protein was then active in a
normal host of the baculovirus, even though the toxin moiety in itself was not. This
approach could be useful, both for extending the effective host range of B t toxins
to more pests and (because resistance generally seems to involve mutant receptors)
for overcoming resistance problems with current targets.
Given that alternatives to B t toxins for engineering into plants may not be avail­
able within the next ten years, good management of B t in the interim is critical to
minimize resistance problems. Two crucial ‘design’ issues impinge upon this
problem, relating to the level and specificity of transgene expression. First, if suffi­
cient expression levels are not achieved, then the situation is ripe for rapid selection
of tolerance traits in the pest insect, in much the same way as continual use of
sublethal doses of chemical insecticides selects effectively for tolerance. Indeed,
such a situation has already been observed in insects selected on sublethal doses of
the B t delta endotoxin (e.g. Ref. [27]). Second, strong arguments have been made
that the expression of the toxin should be restricted to those tissues of the plant
requiring protection; other tissues would then be available as refuges for the pest,
which theory suggests should impede the selection for resistance. Quite apart from
these ‘design’ issues, many management issues are also critical to the likelihood of
resistance evolving. There is fierce debate but little agreement over the benefits of
rotations versus mixtures of protected and unprotected plants in this context [28, 29].
The development of genetically engineered host resistance is obviously an
exciting prospect, but there áre several technical problems that need to be overcome
before this approach will realize its potential. Most of the problems with the
engineering technologies should be soluble in the short term, but there are virtually
no insecticidal genes suitable for engineering into production animals and there is a
great need for alternatives to B t toxins for engineering into plants. Until such alterna­
tives are in commercial use, there is a great danger that misuse of B t toxins, either
delivered in transgenic plants or as microbial insecticides, could lead to resistance
problems and the loss of B t as a control option.
26 OAKESHOTT et al.

4. GENETIC ENGINEERING OF INSECTS

The availability of general insect genetic engineering technologies could have

three potential benefits in pest management. First would be the engineering of pests
to create genetic control systems such as have been developed with chromosomal
translocations for the Australian sheep blowfly, L u c i l i a c a p r i n a [30]. The second
would be the engineering of beneficiáis to enhance their efficacy as natural enemies.
An example would be the engineering of an insecticide resistance gene into a para­
site/predator. Finally, a genetic engineering technology for insects would have value
throughout applied entomology as a tool for understanding the function of strategic
genes in insects of economic significance. For instance, characterization of genes
involved in insecticide resistance would not only assist the management of current
insecticides by providing tools for monitoring resistance, but could also lead to the
design of new insecticides or synergists that break the resistance cycle. As another
example, characterization of the genes involved in sexual differentiation could per­
mit the apparent sex ratio to be manipulated, thereby providing an ability to rear only
phenotypic males for use in the sterile insect technique (SIT).
So far, the only insects for which a reliable and repeatable engineering technol­
ogy is available are certain D r o s o p h i l a species. This technology has generally used
the transposable element P as a vector. Briefly, the procedures involve the cloning
of the gene of interest into a modified copy, of the P element, injection of the gene-P
element chimera into a preblastoderm embryo and then use of the transposition abil­
ity of the P element as a vector to insert the foreign gene into the host genome.
However, the P element proves to have a highly restrictive functional range and
efforts to use the element as a transfer vector to introduce marker genes into non-
drosophilid insects have so far met with little success. Some transgenic mosquitoes
have been recovered, but only at very low frequencies and subsequent analyses have
shown that integration occurred independently of P element transposition (e.g.
Ref. [31]).
Three other more mechanical approaches not based upon transposable ele­
ments have involved attempts to incorporate foreign DNA into sperm [32] and use
of electroporation or ballistic technology to bombard embryos with foreign DNA in
much the same.way as has successfully been carried out with some plants [33]. To
our knowledge, no insect species has been stably transformed by sperm absorption
or electroporation as yet, although transient expression of foreign genes has recently
been achieved in D r o s o p h i l a embryos using electroporators [34]. However, ballistic
technology has recently yielded putative germ line transformants in both D r o s o p h i l a
and a mosquito, A n o p h e l e s g a m b i a e [35]. Transformation frequencies were very low
and the technique would only be sufficient by itself for species where large numbers
of synchronized eggs are relatively easily obtained. Nevertheless, the technique
could still have value as a component of a transformation technology for other spe­
cies; it could be a very useful alternative to microinjection as a means of introducing
 IAEA-SM-327/2 27

the foreign DNA into the host nucleus, providing that an efficient biological vector
system was then able to integrate the DNA into the host chromosome.
There have in fact been several recent advances in the biological aspects of the
technology as well. These three critical aspects are: a suitable marker gene that
allows transformants to be identified by a reliable and unambiguous phenotype; a
promoter(s) to drive the expression of the marker and other foreign genes once
integrated into the host genome; and the vector system to direct the incorporation
of the foreign DNA into the host genome. Most marker systems in D r o s o p h i l a have
been based around eye pigmentation and antibiotic resistance; however, recent
studies have shown that non-homologous markers such as the E . c o l i ¡3-g l u c u r o n i ­
d a s e gene (GUS) may have generic application [36]. Likewise, the promoter system
of first choice was based on the D r o s o p h i l a heat shock protein 70 promoter, but this
system is inadequate, at least in L . c u p r i n a embryos, where the D r o s o p h i l a actinC
promoter gives better expression [37].
The vector system has remained the most elusive biological component of a
non-drosophilid transformation technology. Nevertheless, the recent development of
an in vitro transposable element excision assay has greatly accelerated work on this
component. These assays, which detect and measure the excision of transposable ele­
ments in preblastoderm embryos, are both easy to perform and rapid — results are
obtained within several days of microinjection. In these assays, a reporter plasmid
containing a transposable element into which a genetic marker has been inserted is
co-injected into preblastoderm embryos with a plasmid containing the functional
transposase gene (the transposase enzyme is encoded by the transposable element and
mediates the transposition event). The following day, the reporter plasmids are res­
cued from developed embryos and transformed into a strain of E . c o l i which permits
the loss of the genetic marker to be assayed by a simple colour change. Candidate
plasmids are then sequenced to determine whether the loss of the marker is associated
with the excision of the transposable element. Results gained from these assays
enable any particular transposable element’ s suitability as a transformation vector for
a given species to be determined.
Excision assays have been used to demonstrate that the P element is capable
of mobility in members of the Drosophilidae [38], but is effectively non-functional,
or at least highly inefficient, in the various non-drosophilids examined. Major
modifications to this transposable element would be needed if it is to be used to
achieve transformation in these species.
Conversely, excision assays based on the h o b o transposable element of
D r o s o p h i l a demonstrate that this element can be cross-mobilized in L . c u p r i n a and
M . d o m e s t i c a [ 3 9 ] . Furthermore, the ability of h o b o to be cross-mobilized in non-
drosophilids in the absence of h o b o transposase indicates that these species may
possess a similar transposable element system [39]. Another significant feature of
h o b o elements with respect to non-drosophilid transformation is that h o b o trans­
posase shows a high degree of amino acid similarity to both the A c element of maize
28 OAKESHOTT et al.

and the Т а т З element from snapdragons, suggesting a common phylogenetic origin

[40]. These two plant transposable elements display the widest host range of all
eukaryotic transposable elements so far examined. Our current understanding of
h o b o elements thus gives cause for optimism about their utility in the genetic trans­
formation of L . c u p r i n a and M . d o m e s t i c a , and perhaps other non-drosophilids as
well.
There may also be cause for optimism about the use of m a r i n e r , another small
transposable element originally isolated from D . m a u r i t i a n a [41]. Like P and h o b o ,
it transposes through a ‘DNA only mechanism’ but, unlike P and h o b o , is capable
of transposition in somatic as well as germ line tissue. M a r i n e r is widely distributed
in the D . m e l a n o g a s t e r species group and has also been found in a distantly related
genus, Z a p r i o n u s . M a r i n e r can be used to achieve transformation of D . m e l a n o g a s t e r
using the same strategy as for P and h o b o [42] and, given its wide distribution, may
also be useful in achieving non-drosophilid transformation.
Once a transposable element based transformation system is available, two
other recent developments can then be exploited to enhance both the efficiency and
specificity of this technology. The first of these addresses the efficiency issue by
overcoming the inability of injected plasmid DNA to replicate efficiently within
insect nuclei. If the plasmids containing transposable elements could be made to
replicate autonomously, there should be. an increase in transformation frequency,
simply because more plasmids survive through to the full development of germ line
tissue. This prediction has been fulfilled in other organisms such as yeast, where use
is made of autonomously replicating sequences (ARSs) which enable plasmids to
replicate within yeast cells. Attempts to isolate analogous sequences from D r o s o p h i l a
have not been successful as yet; however, small palinodromic sequences have been
isolated from the genome of a baculovirus from A . c a l i f o m i c a that, once cloned into
plasmids, permit the autonomous replication of these plasmids in S p o d o p t e r a
f r u g i p e r d a cells [43]. This offers, some hope that ARS like sequences from insects
or their viruses may soon be available to increase the efficiency of insect
transformation.
The second development which would improve insect transformation technol­
ogy once established involves gene targeting technology. A major limitation of cur­
rent transposable element based transformation technology in D r o s o p h i l a has been
that the transposable element, together with the foreign DNA contained in it, inserts
randomly into the recipient genome. This makes it practically impossible to replace
an endogenous gene with a custom mutagenized homologue; it means, for example,
that the foreign gene will only affect the phenotype of the transformants if its expres­
sion is dominant over the existing endogenous gene. Recently, however, both the
P element and a system from yeast called FLP recombinase have been manipulated
so as to permit targeted gene replacement in D . m e l a n o g a s t e r [44, 45]. The condi­
tions under which gene replacements occur are still highly , specific in that they
require, either the presence of a pre-existing P element or FRT sequence, respec­
 IAEA-SM-327/2 29

tively, in the genome adjacent to the target sequence. Nevertheless, the FLP recom-
binase has been demonstrated to facilitate site directed plasmid to plasmid
recombination in mosquitoes [46]. Clearly, neither the h o b o equivalent of the P sys­
tem or the FLP system are currently sufficient to provide a gene replacement technol­
ogy for non-drosophilids. However, with further development of these systems,
either could become the basis for highly specific gene replacement procedures in pest
insects once the basic transformation technology has been established.
Finally, we consider some of the downstream research priorities that will arise
once an efficient transformation technology has been developed for pest insects.
Perhaps the most often discussed application is in the design of genetic control strate­
gies. We would concur that genetic control strategies would be greatly aided by
transformation technology. However, significant difficulties in the extension of this
technology into the field would still remain. A system or gene must be identified
which will lead to a significant and rapid decrease in the relevant insect population
and this gene must first be efficiently spread throughout the population. In practice,
this means that a conditional field lethal gene is needed. One such system which
shows some potential in this respect is the process of somatic sexual differentiation.
Investigations into this process in D r o s o p h i l a have revealed that the apparent sex
ratio of this species could be manipulated if so desired. Thus, a population of pheno­
typic males could be obtained and, if applied to insects such as the Mediterranean
fruit fly which are currently the subject of SIT, would ensure the mass release of
only phenotypically male flies.
While not devaluing such applications of transformation technology in genetic
control, we believe that more powerful benefits may accrue in other areas. These
benefits follow from the fact that the technology enables the molecular analysis of
gene function in vivo. To give just one example, the study of P l a s m o d i u m - m o s q u i t o
interactions will benefit greatly from the development of mosquito transformation.
Genes expressed within the mosquito gut and salivary glands which are involved in
P l a s m o d i u m transmission could be identified, isolated and analysed. Thus, the
molecular basis for the P l a s m o d i u m refractory phenotype could be elucidated which,
in turn, could provide new approaches to malaria control. For example, specific
biorational insecticides could be designed which inhibit the function of those
molecules and processes in the mosquito required for the passage of P l a s m o d i u m ,
thereby preventing the parasite from infecting livestock and man.

REFERENCES

[1] CUNNINGHAM, J.C. , Baculoviruses: Their status compared to Bacillus thuringiensis

as microbial insecticides, Outlook Agrie. 17 (1988) 10-17.
[2] HENDRY, D .A ., “ Nodaviridae of invertebrates” , Viruses of Invertebrates
(KURSTAK, E., Ed.), Dekker, New York (1991) 227-276.
30 OAKESHOTT et al.

3] FRIEDMAN, M.J., “ Commercial production and development” , Entomopathogenic

Nematodes in Biological Control (GAUGLER, R., KAYA, H .K., Eds), CRC Press,
Boca Raton, FL (1990) 153-172.
4] VLAK, J.M., “ Genetic engineering of baculoviruses. for insect control” , Mole­
cular Approaches to Environmental and Applied Entomology (OAKESHOTT, J.G.,
WHITTEN, M.J., Eds.), Springer-Verlag, New York (1992) 90-127.
5] McGAUGHEY, W .H ., Insect resistance to the biological insecticide Bacillus thurin-
giensis, Science 229 (1985) 193-195.
6 ] TABASHNIK, B.E., et al., Field development of resistance to Bacillus thuringiensis
in diamondback moth (Lepidoptera: Plutellidae), J. Econ. Entomol. 83 (1990)
1671-1676.
7] GOULD, F., et al., Broad-spectrum resistance to Bacillus thuringiensis toxins in
Heliothis virescens, Proc. Natl. Acad. Sci. USA 89 (in press).
8 ] SMITH, G.E., SUMMERS, M .D ., Method for producing a recombinant baculovirus
expression vector, United States Patent No. 4,745,051, US Patent Office, Washington,
DC.
9] TOMALSKI, M .D ., MILLER, L .K ., Insect paralysis by baculovirus-mediated expres­
sion of a mite neurotoxin gene, Nature (London) 352 (1991) 82-85.
10] CHRISTIAN, P.D., et al., “ Insect viruses: New strategies for insect control” ,
Molecular Approaches to Environmental and Applied Entomology (OAKESHOTT,
J.G., WHITTEN, M.J., Eds), Springer-Verlag, New York (1992) 128-163.
11] THANABALU, T ., et al., Expression of the mosquitocidal toxins of Bacillus sphaeri-
cus and Bacillus thuringiensis subspecies israelensis by recombinant Caulobacter
crescentus, a vehicle for biological control of aquatic insect larvae, Appl. Environ.
Microbiol. 58 (1992) 905-910.
12] STEWART, L .M .D ., et al., Construction of an improved baculovirus insecticide con­
taining an insect-specific toxin gene, Nature (London) 352 (1991) 85-88.
13] HAMMOCK, B .D ., et al., Expression and effects of the juvenile hormone esterase in.
a baculovirus vector, Nature (London) 244 (1990) 458-461.
14] MAEDA, S., Increased insecticidal effect by a recombinant baculovirus carrying a syn­
thetic diuretic hormone gene, Biochem. Biophys. Res. Commun. 165 (1989)
1177-1183.
15] FEITELSON, J.S., et al., Bacillus thuringiensis: Insects and beyond, Biotechnology
10 (1992) 271-275.
16] RAND, K ., et al., Cloning and expression of a protective antigen from cattle tick
Boophilus microplus, Proc. Natl. Acad. Sci. USA 8 6 (1989) 9657-9661.
17] DAVIDSON, S., New technologies bring vaccines closer. Rural Research 153 (Sum­
mer 1991/92) 19-24.
18] WINTER, G ., MILSTEIN, C ., Man-made antibodies, Nature (London) 349 (1991)
293-299.
19] HIATT, A ., et al., Production of antibodies in transgenic plants, Nature (London) 342
(1989) 76-78.
20] JAENISCH, R., Transgenic animals, Science 240 (1988) 1468-1473.
21] KLEE, H.J., ROGERS, S.G., Plant gene vectors and genetic transformation: Plant
transformation systems based on the use of Agrobacterium tumefaciens, Cell Cuit.
Somatic Genet. 6 (1989) 1-23.
 IAEA-SM-327/2 31

[22] KLEIN, T .M ., et al., “ Applications of the particle gun in plant biology” , Progress in
Plant Cellular and Molecular Biology (NUKAMP, VAN DER PLA, L .H .W .,
VAN AARTRIKJ, A ., Eds), Kluwer, Dordrecht (1990) 56-66.
[23] PERLAK, F.J., et al., Modifications of the coding sequence enhance plant expression
of insect control protein genes, Proc. Natl. Acad. Sci. USA 8 8 (1991) 3324-3328.
[24] WARD, K., “ Transgenic farm animals and enhanced productivity” , Animal Genetics
of Pigs, Oncomice and Men, Athene Trust, London (in press).
[25] BINNINGTON, K .C ., BAULE, V.J., “ Naturally occurring insecticidal molecules as
candidates for genetic engineering” , Molecular Approaches to Environmental and
Applied’ Entomology (OAKESHOTT, J.G., WHITTEN, M.J., Eds), Springer-Verlag,
New York (1992) 38-89.
[26] SIVASUBRAMANIAN, N ., et al., “ Use of baculovirus proteins for developing novel
protein insecticides” , Invertebrate Pathology and Microbial Control (Proc. 5th Int.
Colloq. Adelaide, 1990), Society for Invertebrate Pathology, Adelaide (1990).
[27] SIMS, S.R., STONE, T.B ., Genetic basis of tobacco budworm resistance to an
engineered Pseudomonas fluorescens expressing the delta endotoxin of Bacillus thurin-
giensis kurstaki, J. Invert. Pathol. 57 (1991) 206-210.
[28] GOULD, F., Evolutionary biology and genetically engineered crops, BioScience 38
(1988) 26-33.
[29] VAN RIE, J., Insect control with transgenic plants: Resistance proof? Trends Biotech.
9 (1991) 177-179.
[30] WHITTEN, M.J., HOY, M .A ., “ Genetic improvement and other genetic considera­
tions for improving the efficacy and success rate of biological control” , Biological Con­
trol (FISHER, T .W ., Ed.), University of California Press, Los Angeles (in press).
[31] MILLER, L.H ., et al., Stable integration and expression of a bacterial gene in the
mosquito, Anopheles gambiae, Science 237 (1987) 779-781.
[32] ATKINSON, P .W ., et al., Association of exogenous DNA with cattle and insect sper­
matozoa in vitro, Mol. Reprod. Develop. 29 (1991) 1-5.
[33] KLEIN, T. M ., et al., Transformation of microbes, plants and animals.by particle bom­
bardment, Biotechnology 10 (1992) 286-291.
[34] KAMDAR, P., et al., Transient expression of DNA in Drosophila via electroporation,
Nucleic Acids Res. 20 (1992) 3526.
[35] MIALHE, E., LAUGHINGHOUSE, A ., MILLER, L.H ., Paper IAEA-SM-327/14,
these Proceedings.
[36] O ’BROCHTA, D .A ., ATKINSON, P .W ., unpublished data.
[37] ATKINSON, P .W ., O’BROCHTA, D .A ., In vivo expression of two highly conserved
Drosophila genes in the Australian sheep blowfly, Lucilia cuprina, Insect Biochem.
Mol. Biol. 22 (1992) 423-431.
[38] O ’BROCHTA, HANDLER, A .M ., P element excision in Drosophila
D .A .,
melanogaster and related drosophilids, Mol. Gen. Genet. 225 (1991) 387-394.
[39] ATKINSON, P .W ., et al., Paper LA.EA-SM-327/7, these Proceedings.
[40] CALVI, B.R., et al., Evidence for a common evolutionary origin of inverted repeat
transposons in Drosophila and plants: hobo, Activator, and ТатЗ, Cell 6 6 (1991)
465-471.
32 OAKESHOTT et al.

[41] H A Y M E R ,D .S., MARSH, J.L ., Germ line and somatic instability of a white mutation
in Drosophila mauritiana due to a transposable element, Dev. Genet. 6 (1986)
281-291.
[42] GARZA, D ., et al., Introduction of the transposable element mariner into the germ line
of Drosophila melanogaster. Genetics 128 (1991) 303-310.
[43] PEARSON, M ., et al., The Autographa califomica baculovirus genome: Evidence for
multiple replication origins, Science 257 (1992) 1382-1384.
[44] GOLIC, K.G., LINDQUIST, S., The FLP recombinase of yeast catalyzes site-specific
recombination in the Drosophila genome, Cell 59 (1989) 499-509.
[45] GLOOR, G .B ., et al., Targeted gene replacement in Drosophila via P element-induced
gap repair, Science 253 (1991) 1110-1117.
[46] MORRIS, A .C ., et al., FLP-mediated recombination in the vector mosquito, Aedes
aegypti, Nucleic Acids Res. 19 (1991) 5895-5900.
 IAEA-SM-327/5

GENETIC ENGINEERING OF INSECTS

AND APPLICATIONS IN BASIC
AND APPLIED ENTOMOLOGY

J.M. CRAMPTON
Wolfson Unit of Molecular Genetics,
Liverpool School of Tropical Medicine,
Liverpool,
United Kingdom

Abstract
GENETIC ENGINEERING OF INSECTS AND APPLICATIONS IN BASIC AND
APPLIED ENTOMOLOGY.
Insects are responsible for transmitting a wide variety of organisms to man and
agricultural animals and are major agricultural pests. Control of such pest and vector popula­
tions has until now relied on the elimination of breeding sites and the widespread application
of chemical insecticides. The appearance of insecticide resistance, coupled with rapidly
escalating costs for developing new insecticidal compounds, and the increasing awareness of
the detrimental effect insecticides have on the environment, has stimulated interest in alterna­
tive methods for controlling insect populations or their ability to transmit disease causing
organisms to man and animals. What is required is the development and evaluation of a new
generation of methodologies which will have a profound and long lasting effect on insect pest
populations or their efficiency as vectors of disease. Biotechnology, genetic engineering and
transgenic technology can play a central role in developing such methodologies. Transgenic
technology in relation to insects may have potential application for population suppression.
Perhaps, more interestingly, it may also provide a means for altering the vectorial capacity
of insect populations. The various aspects of applying transgenic technology to insects are
discussed in the paper, highlighting the requirements for being able to genetically manipulate
insect genomes. The potential of the technology is then illustrated by outlining recent research
by the author aimed at creating an ‘incompetent’ mosquito, i.e. one which blocks the trans­
mission of malaria. There is clearly some way to go before any release of transgenic insects
can be considered. The power of the technology is, however, so enormous that it must be
explored not only for its potential for manipulating wild populations, but also as an analytical
tool to help us understand the biology of insects and how they interact with man and his
environment.

1. INTRODUCTION

Insects are responsible for transmitting a wide variety of organisms to man.

This is particularly true in tropical countries where the diseases which these
organisms cause, including malaria, filariasis, leishmaniasis, river blindness and

33
34 CRAMPTON

TABLE I. THE IMPORTANCE OF MOSQUITO VECTORS OF. DISEASE

Persons Principal
Disease Causative organism
infected annually mosquito vector

Malaria Protozoa

Plasmodium > 5 0 0 million Anopheles

Filariasis Filarial worms

Wuchereria > 2 5 0 million

Culex
Brugia > 5 million

Yellow fever Arbovirus -л

Dengue Arbovirus

DHF Arbovirus 1 million Aedes

Japanese encephalitis Neurotropic virus

La Crosse Neurotropic virus

various viral diseases, represent the most important health care problems in the
world today. Mosquitoes are particularly important as vectors of disease and Table I
lists just some of the different organisms which they transmit to man and the esti­
mated number of individuals affected by each disease annually.
The most important of the diseases transmitted by insects is malaria which,
despite enormous efforts over many years, is again an increasingly important health
problem. Recent estimates indicate that malaria is endemic in 102 countries with
2000 million people at risk from the disease, representing over half the world’s
population. There are perhaps 200 million malaria infections and 1-2 million deaths
annually and clearly this one disease has an enormous impact on the health and
economies of many tropical countries [1]. The incidence of malaria is increasing, due
largely to the development of insecticide resistance by the mosquito vectors and by
the appearance of drug resistance in the malaria parasite. These factors are exacer­
bated by climatic changes and migration of increasing numbers of people from non­
endemic areas to regions where malaria is prevalent. Until now, control of malaria
on a global scale has relied on the application of chemical insecticides to limit
Anopheline vector populations. The appearance of insecticide resistance, coupled
with rapidly escalating costs for developing new insecticidal compounds, and the
increasing awareness of the detrimental effect insecticides have on the environment,
has stimulated interest in alternative methods for malaria control. What is required
is the development and evaluation of a new generation of methodologies which will
 IAEA-SM-327/S 35

have a profound and long lasting effect on malaria transmission. Biotechnology and
molecular biology can play a central role in the search for, and production of, such
new tools. The development of potential anti-malarial vaccines and larvicidal
compounds produced by biotechnology are obvious examples of the power of the
approach. Perhaps, surprisingly, this technology has not been applied to the
Anopheline vectors of malaria or other insect vectors of disease until very recently.
Advances in the molecular analysis of vector-parasite relationships and vector
molecular biology now make such an approach very attractive, especially as past
experience has shown that vector control is an effective way of disrupting disease
transmission.

2. GENETIC MANIPULATION OF INSECT VECTORS OF DISEASE

Control of insect populations by genetic means is not new. A number of vector

control programmes have utilized the mass release of sterile males, the generation
of cytogenetically induced sterility through translocation and the use of mating
behaviour to transfer lethal or sterilizing agents between members of the populations.
The strategy of sterile male release has proved to be particularly effective for the
eradication of the screwworm fly from the southern part of the United States of
America [2]. However, such strategies of autocidal population control have proved
prohibitively expensive, since they involve the repeated mass release of treated
males [3].
The advent of recombinant DNA and transgenic techniques now provides the
means for the controlled genetic manipulation of insect vector genomes by the direct
introduction of DNA into the germ line of these insects. Two particular advantages
of using transgenic technology over classical genetics for future manipulation are
evident and should be emphasized. One is the potential to exploit genes and gene con­
structs across species barriers and the other is the ability to introduce particular,
defined sequences without the génome disruption of a conventional cross.
Transgenic technology in relation to insect vectors may have potential applica­
tion for population suppression. Perhaps, more interestingly, it may also provide a
means for altering the vectorial capacity of insect populations to transmit disease.
For example, it may be possible to produce strains which are refractory for the
pathogen, strains of the vector which have reduced vector competence or reproduc­
tion potential, or to increasé the susceptibility of the vector to existing control strate­
gies. These áre clearly applications for the futuré. More immediately, this type of
technology, and molecular biology in general, may be used as an exquisite analytical
tool to begin to dissect the complex relationships between the insect vector and the
disease causing organism which it transmits.
In any consideration of transgenic technology and how it may be applied to
medically important insects, the following factors need to be considered:
36 CRAMPTON

(a) The practical requirements for creating transgenic insects,

(b) How best to apply.the technology,
(c) What gene systems from insects or other organisms need to be defined in order
to undertake the desired manipulations,
(d) Once transgenic insects incorporating the desired characteristics have been
created, what further research needs to be carried out prior to their release into
natural populations?

Each of these factors is considered in turn.

3. REQUIREMENTS FOR GENETIC MANIPULATION

Initially, we have concentrated our studies on the mosquito A e d e s a e g y p t i , the

major urban vector of arboviral diseases such as yellow fever, dengue and dengue
haemorrhagic fever [4]. These acute diseases affect millions of people and result in
substantial mortality . A e d e s a e g y p t i is therefore an important disease carrying vector
and, coupled with its suitability for laboratory research, has great potential for
studies in molecular genetics [5].
In order to be able to carry out the genetic manipulation of mosquito vectors,
a number of areas of basic research need to be developed. The areas to be considered
include the analysis of genome complexity and organization, the methods for
introducing DNA into mosquito embryos and cells, and the search for a mosquito
DNA transformation system. Each of these areas is considered in turn in order to
indicate what has been achieved in relation to what still needs to be done.

3.1. Genome organization and complexity

If the genome of mosquito vectors is to be genetically manipulated in a con­

trolled and directed fashion, it is important to understand the size of the genome
which is to be manipulated. Genome organization, i.e. the nature and dispersion
pattern of repetitive sequences and how they are organized in relation to the coding
sequences, is also important. This is because it will have a profound influence on
the types of manipulation which can be envisaged and the approaches to be adopted
to identify and clone sequences of interest.
Until recently, very little was known about the size and organization of
mosquito genomes. Black and Rai [6] have analysed the genomic DNA from
four species of mosquito, A n . q u a d r i m a c u l a t u s , C . p i p i e n s , A e . a l b o p i c t u s and
A e . t r i s e r i a t u s . More recently, Cockburn and Mitchell [7] have shown that the
genomes of Anopheline mosquitoes are generally relatively small and exhibit what
is called a long period interspersion (LPI) pattern of repetitive sequences (the
euchromatic DNA consists of long, single copy sequences interrupted by a few
 IAEA-SM-327/5 37

moderately repeated sequences). This LPI type of genome organization in

Anopheline mosquitoes is in marked contrast to that found in the A e . a e g y p t i
genome, which is both large and much more complex in its organization [8]. The
fact that A n o p h e l e s mosquitoes have genomes which are relatively small [9] and an
LPI organization means that, in molecular terms, they are simpler to work with and
that many of the approaches used for analysis of the D r o s o p h i l a genome (which also
has an LPI organization pattern) are directly applicable to A n o p h e l e s .

3.2. Methods for introducing DNA into mosquito cells and embryos

The generation of transgenic mosquitoes requires a means for introducing

DNA into the germ line of the insect as well as an efficient DNA vector system which
will allow stable integration of the introduced DNA. Microinjection systems have
now been developed for a number of mosquitoes including A n . g a m b i a e [10],
A e . t r i s e r i a t u s [11] and A e . a e g y p t i [12]. The systems which have been developed
are derived from that employed for D r o s o p h i l a transformation modified to take
account of differences in the physiology and development times between D r o s o p h i l a
and the mosquito species studied.
The DNA which was introduced into the mosquito embryos in these experi­
ments was the pUChsneo vector/helper system based on the-P transposable element
from D . m e l a n o g a s t e r [13]. An antibiotic resistance gene ( n e o ) forms part of this
DNA vector system. The presence of this gene, if it becomes integrated into the
mosquito genome, confers resistance to the synthetic antibiotic G418. Progeny of
those individuals which survive microinjection are exposed to G418, so that only
those mosquitoes which have integrated the injected DNA may be selected. Experi­
ments with this vector/helper system have served to demonstrate that DNA may be
injected into mosquito embryos with survival rates comparable to those in work with
D . m e l a n o g a s t e r [14]. Integrations of the introduced P element DNA have been
observed in both A n o p h e l e s and A e d e s mosquitoes and the integration events appear,
in some cases, to be heritable and clearly involve the germ line of the mosquitoes
involved. Although these events did not result from normal P element transposition,
some functional role of the P sequence cannot be excluded.
As indicated above, embryonic transformation has shown some limited
success. However, the transformation of cultured cells has, thus far, provided
greater information regarding mosquito gene expression and its control. This is
due, in part, to the simpler methods of introducing DNA into cells in culture and
the more efficient screening of transformed lines. Several transfection methods are
successful for mosquito cell cultures [15, 16]. These include the use of polybrene,
a divalent cation, together with a glycerol shock treatment [17]> and Lipofectin® ,
a patented lipid compound, which forms complexes with DNA and fuses with cell
membranes [18].
38 CRAMPTON

. Development of this transfection technology has allowed recent advances in

mosquito molecular biology. Transient expression assays, with a DNA vector com­
prising the D r o s o p h i l a heat shock protein promoter (h s p 7 0 ) fused to the chlor­
amphenicol acetyl transferase ( C A T ) gene, have elucidated various aspects of the
heat shock response of mosquitoes. Similar studies have shown that other hetero­
logous promoters, including the constitutive rat actin promoter, and the inducible
D r o s o p h i l a metallothionein promoter [19, 20] are recognized by the mosquito.
Further work will define control sequences from cloned mosquito genes, such as
those of the tubulin and heat shock genes.
More recently, we have identified a vector construct that stably transforms
mosquito cells to antibiotic resistance. Southern blot analysis and in situ hybridiza­
tion of transformed lines have revealed that the vector becomes integrated, both
singly and in tandem arrays. In addition, the construct remains in an extremely stable
chromosomal position during prolonged culture in the absence of selection. In an
extension of this work, we have demonstrated that separate non-selected DNA can
be stably transferred and expressed in mosquito cells along with the selectable
marker gene of the vector DNA. Thus, it is possible to introduce virtually any gene
into a mosquito cell. In the near future, such gene expression studies will be used
to overproduce useful proteins in culture [20].
It is clear from this work, however, and the experiments involving microinjec­
tion of the P element into embryos that the P element system in its present form is
not suitable for routine use in the mosquito. Thus, while the means are currently
available for introducing DNA into both mosquito embryos and cultured cells, the
major stumbling block is currently the lack of an appropriate DNA vector system
for manipulating the mosquito genome.

3.3. Search for a mosquito DNA transformation vector system

As indicated above, the DNA vector systems currently available are not ideal
for creating transgenic mosquitoes. Attempts are being made to modify the P system
for more general use [21] but alternative elements, such as h o b o from D r o s o p h i l a
and the A c element'from maize, are also being investigated for their usefulness in
the mosquito system. In addition, there is increasing interest in identifying mosquito
elements with the properties of mobile or transposable genetic elements (TGEs).
Such an element could form the core of a mosquito transformation vector. Develop­
ment of this system first requires the identification, isolation and characterization of
such endogenous mosquito TGEs.
A number of approaches have been taken to identify such mobile elements in
mosquitoes. One of these is to analyse specific gene systems, such as the ribosomal
DNA of mosquitoes, in order to detect variants of these genes arising from the inser­
tion o f á mobile element. No such insertions have, as yet, been detected in
A e . a e g y p t i rDNA [22] but insertion events have been detected in the rDNA of
 IAEA-SM-327/5 39

A n. g a m b ia e and these elements are being fully defined [23]. The elements appear
to resemble a particular class of TGEs, non-viral retroposons. It is unlikely, how­
ever, that these elements will prove ideal as transformation vectors because of the
ill defined nature of their mode of transposition.
We have recently adopted an alternative strategy to directly identifying a
specific class of TGEs, known as retrotransposons, in the mosquito DNA. The
approach relies on utilizing the characteristic biochemical and structural properties
of these elements to identify them. This has led to the successful isolation of several
retrotransposon like elements from t h c A e . a e g y p t i genome [24, 25]. More recently,
we have used the polymerase chain reaction (PCR) to develop a particularly rapid
methodology for identifying endogenous retrotransposon like elements in mosquito
DNA [8, 26]. Once such elements have been isolated, fully characterized and their
ability to transpose autonomously established, they may be engineered to form the
core of a transformation vector system.
Any DNA vector system which is to be of practical value will have to incor­
porate a number of features in addition to its ability to transpose. Most important
of these are a selectable marker system and specific promoter or enhancer sequences.
As indicated above, current DNA vectors incorporate an antibiotic resistance gene
which allows for selection of transformed individuals by exposure to G418.
However, this system is less than satisfactory because different mosquitoes exhibit
a, spectrum of sensitivity to G418 and, more importantly, because it depends on high
levels of expression of the resistance gene in the transformed mosquitoes. The
method of choice would be the use of a phenotypic marker, such as eye colour, so
that transformed individuals may be identified by direct visual inspection. Such a
method requires mutant strains of mosquitoes and the corresponding cloned gene
coding for an easily scored phenotype which can be incorporated into a transforma­
tion vector. Although eye colour mutants of both A e . a e g y p t i and A n . g a m b i a e are
available, cloning of the genes responsible for these phenotypes has not yet been
completed.
Finally, a considerable body of work using the pUChsneo transformation
vector, both in cultured mosquito cells and injected embryos, has indicated that the
D r o s o p h i l a heat shock promoter sequence is not entirely satisfactory for driving the
expression of genes in mosquitoes because of its low efficiency and the high tempera­
tures required to induce expression [20]. In addition, at some stage it will be
desirable to express defined genes in mosquitoes in a tissue or stage specific fashion:
For this to be envisaged, stage and'tissue specific promoters that function in the
mosquito have to be defined. None are, as yet, available but attempts to characterize
the DNA sequences responsible for regulating the expression of genes encoding
abundant proteins in mosquitoes are well under way [27].
All of these different aspects of basic research need to be pursued in parallel
and eventually amalgamated to provide a mosquito DNA vector system of real
practical value.
40 CRAMPTON

4. POTENTIAL APPLICATION OF TRANSGENIC TECHNOLOGY

IN INSECT SYSTEMS

Once the systems necessary to create transgenic insects have been developed,
how can this technology be applied? Two aspects are discussed in order to illustrate
the potential of the technology. The first deals with the use of the technique for
analytical purposes and the second with applying transgenic technology to medically
significant mosquito populations.

4.1. Transgenic technology as an analytical tool

The introduction and insertion of a mobile genetic element at or near a particu­

lar locus can cause that allele to mutate, producing a structural or developmental
effect. In D r o s o p h i l a , in particular, TGEs have been used as mutagens in order to
clone genes or gene clusters of interest via transposon tagging [28]. In essence, the
TGE is introduced into the germ line of the insect by microinjection of the embryo,
and the progeny scored for mutants in the phenotype of interest. Subsequently,
cloned TGE probes are used for in situ hybridization to chromosomes of mutant and
wild type individuals. This identifies a TGE ‘newly’ integrated at or near the genetic
locus of interest. DNA clones are then retrieved from a genomic library prepared
from the mutant stock using the TGE DNA as a probe. DNA sequences adjacent to
the TGE in such clones represent the gene for the locus of interest. This approach
is an extremely powerful application of the technology, since it allows the cloning
of genes purely on the basis of their function.

4.2. Transgenic technology in insect populations of medical significance

As discussed above, insects are important vectors of disease to both man and
agricultural animals. In this way, vector populations may have a profound impact on
the health and economy of a region which in many cases, particularly in the tropics,
are the most fertile and potentially productive areas. Transgenic technology may
eventually have a role to play in controlling vector borne disease by providing the
means to suppress vector populations by rendering them vulnerable to subsequent
control measures, such as insecticide susceptibility, temperature sensitivity or ability
to survive diapause. A second possibility and, perhaps, a more exciting approach,
would be to alter the ability of the insect to transmit the disease. Clearly, such possi­
bilities are for the future, but in the shorter term it is quite feasible to consider genetic
manipulation of insect populations of direct commercial value, such as the honey bee
or silk moth. Here, transgenic technology may be employed to confer a. number of
beneficial characteristics to these insects to create novel and highly productive
strains. For example, the insect may be manipulated to increase the yield of the
product by increasing the growth rate or by enhancing the resistance of the insect
to infection, temperature shock or other detrimental factors.
 IAEA-SM-327/5 41

4 .2 .1 . P o t e n t ia l t a r g e t g e n e s f o r m a n ip u la tio n

. Having discussed the types of manipulation of insect genomes which may be

beneficial in economic or health terms, it is worth considering the types of gene
system which may be potential targets for manipulation to achieve these aims. In this
respect, there are a number of obvious targets for manipulation, including the genes
involved in the insect immune system, developmental control genes and insecticide
resistance genes. Genes influencing all of these factors have now been characterized
at the molecular level for a number of different insects and it is now feasible to con­
sider manipulating them in the germ line of these insects. In addition, a number of
genes are of particular interest because they are directly implicated in the ability of
insects to transmit disease causing organisms. Examples include the filarial suscepti­
bility ( f m) and Plasmodium susceptibility ( p i s ) loci of the mosquito, A e . a e g y p t i .
T h e /"1locus is genetically well defined and there is good data on its linkage rela­
tionships. Refractoriness to infection is due to a partially sex linked, dominant gene
[29]. There is marked variation in the susceptibility of A e . a e g y p t i to different filarial
worms, although all of the alleles concerned map at about the same place on the sex
determining chromosome. Also of particular interest is a strain of A n . g a m b i a e
which has been selected for refractoriness to the malaria parasite and characterized
genetically [30]. Attempts are currently under way to clone these genes, but it is
difficult to undertake such a cloning exercise in the absence of any knowledge of the
gene product. Clearly, the use of transgenic technology through transposon tagging
will assist in the characterization of refractory genes and their products.

4 .2 .2 . C r ea tin g a n ‘i n c o m p e t e n t ’ m o s q u i t o

An important genotypic characteristic not met by the majority of genes

encoding refractoriness is that any such gene introduced into the insect would have
to be capable of altering the phenotype through the expression of a single gene copy.
Unfortunately, at present there is no gene or gene product defined at the molecular
level which is known to directly affect phenotype in relation to pathogen develop­
ment in, or transmission by, any insect. However, in the mosquito system a number
of molecules are known to affect the transmission of malaria by Anophelines. Fore­
most among these are the so called transmission blocking vaccines, which can
achieve a total transmission blockade [31]. These vaccines attack antigens present
on the gametes and ookinetes of the malaria parasite, and antibodies which recognize
these antigens are able to block the development of the parasite in the mosquito
midgut. A very exciting possibility therefore is to introduce the genes coding for such
antibodies into the mosquito genome, thus directly conferring the transmission
blocking phenotype to the insect. In this case, a transgenic mosquito would be
created incorporating an antibody gene expressed in the insect midgut in response
to a blood meal and which therefore blocks transmission of malaria.
42 CRAMPTON

This type of approach is an attractive one for a number of reasons. It eliminates

the need for detailed molecular analysis of refractory mechanisms in mosquitoes and
it would be a ‘dominant’ gene system (i.e. one gene copy only would be needed in
each cell of the mosquito). The antigen target.on the stage of the malaria parasite
present in the mosquito may not be under the same selection pressure as other
malarial antigens because it is not normally exposed to the human immune system.
One implication of this may be that the transmission blocking antigens are less likely
to exhibit extensive variation, so the parasite may be less able to avoid this type of
transmission control mechanism. To date, evidence suggests that the ookinete anti­
gens are, indeed, highly conserved. Another advantage of the approach is that the
antibody molecule would only be expressed by the mosquito midgut following a
blood meal, so that such a genetic manipulation may not reduce the overall fitness
of the individual mosquito to any great extent. Finally, use of transgenic insects
incorporating an antibody gene could be applied to any vector transmitted pathogen
(parasite or viral) where a target antigen can be identified as being inhibited by the
expressed molecule.
We are currently involved in creating a transgenic mosquito incorporating a
transmission blocking antibody gene under the control of a blood meal induced, gut
specific promoter sequence. This project is being undertaken with funding from the
European Economic Community by R. Sinden (Imperial College, United Kingdom),
A. Crisanti (University of Rome, Italy) and. R. Galler (Fiocruz, Brasil); Fig. 1
illustrates the steps involved in generating this type of transgenic mosquito. Figure 2
illustrates how such an ‘incompetent’ transgenic mosquito would disrupt the trans­
mission of malaria. If successful, transgenic mosquitoes expressing antimalarial anti­
bodies may represent a potential strategy for controlling malaria and establish a
precedent for a wide range of new antidisease strategies in many vector borne
diseases.

4 .2 .3 . T r a n s g e n i c m o s q u i t o e s in n a t u r a l p o p u l a t i o n s

Once transgenic insects with the necessary characteristics have been created,
there remains the question, what next? Clearly, if the manipulated insects are them­
selves to be cultivated for production, it may .be possible to directly apply novel
strains created by transgenic means. However, where this is not the case, it is neces­
sary to consider the problems likely to be faced in applying the technology in
experimental and natural populations. It may well be that such a situation would dis­
rupt the normal adaptive process and therefore be opposed by natural selection. If
this is so, then some form of drive mechanism may be needed to force the desired
gene through the population. This is not an alien concept to those who have worked
on the genetic control of insect populations. However, testing of such mechanisms
has been limited since, in reality, they have awaited the advent of recombinant DNA
technology to provide the necessary raw material.
 IAEA-SM-327/5 43

FIG. 1. Creation of a transmission blocking or ‘incompetent’ mosquito.

44 CRAMPTON

IN THE GUT OF AN
‘INCOMPETENT’ MOSQUITO

FIG. 2. Life-cycle of the malaria parasite and how transmission may be disrupted by
generating a mosquito which expresses transmission blocking antibodies in the gut when the
mosquito takes a blood meal infected with the malaria parasite.

Two types of drive mechanism have been suggested. One is meiotic drive,
where a given chromosome is transmitted to more than the expected 50% of off­
spring. Any desirable genes linked to the driven chromosome would eventually
approach fixation, even with the release of relatively few individuals. There is
experimental evidence to support the use of meiotic drive in A e . a e g y p t i . This
mechanism, driven by the M ° locus, has been used to force the marker gene r e (red
eye) into a laboratory cage population [32]. Interestingly, meiotic drive also occurs
during hybrid dysgenesis and it might, therefore, also be possible to exploit this
phenomenon by using either the P element itself, or a mobile element with properties
similar to P , as an efficient mechanism to drive a specific gene construct through
 IAEA-SM-327/5 45

an insect population. The second type of drive mechanism is the exploitation of

genetic traits that reduce heterozygote fitness [33]. For example, the gene to be
driven could be introduced into a translocation chromosome such that viable and fer­
tile homozygotes were formed, whereas heterozygotes would display reduced fertil­
ity or viability. In this way, translocations, pericentric inversions, inter-racial hybrid
sterility, cytoplasmic incompatibility and compound chromosomes all have potential
since, in each case, hybrids have reduced fitness. Such mechanisms require larger
release numbers, since there is no exponential increase in the frequency of the driven
chromosome as with meiotic drive. However, fixation of desirable genes would
occur more quickly than with meiotic drive because of the reduced fitness of the
heterozygous combinations. Certainly in the case of vector populations, the most
useful end result of such programmes would be the progressive replacement rather
than the eradication of disease transmitting populations since an emptied ecological
niche might be colonized rapidly by the migration of wild type insects.

5. TRANSGENIC INSECTS: THE FUTURE

Eventually, embryo transformation will provide the raw material to test the
proposed drive mechanisms in laboratory and natural populations. The questions
posed by considering the release of transgenic insects emphasize the need to assess
the biological consequences of such a release. It is, however, difficult to gauge the
possible hazards of such a release in the absence of experimental evidence and these
ethical and safety considerations need to be faced at an early stage. In order to under­
take an informed appraisal where the possible net benefits may be balanced against
the potential hazards, considerable effort will have to be devoted to utilizing caged
populations and the controlled release of molecularly tagged individuals together
with mathematical modelling of these populations. There is clearly some way to go
before any release of transgenic insects can be considered. The power of the technol­
ogy is, however, so enormous that it must be explored and there is every indication
that over the next few years the potential of transgenic technology in insects will be
fully exploited.

ACKNOWLEDGEMENTS

The author wishes to thank P. Eggleston, G. Lycett, A. Warren, A. Morris,

M. Hughes and T. Knapp, all of whom have been involved in the work described
in this paper. He would also like to thank the Wellcome Trust, the Medical Research
Council, the Wolfson Foundation, Liverpool University and the Commission of the
European Communities (Life Sciences and Technology for Developing Countries)
for providing financial support. The author is a Wellcome Trust Senior Research
Fellow in Basic Biomedical Sciences.
46 CRAMPTON

REFERENCES

[1] STURCHLER, D., How much malaria is there worldwide? Parasitol. Today 5 (1989)
39.
[2] KRAFSUR, E.S., WHITTEN, C.J., NOVY, J.E., Screwworm eradication in North
and Central America, Parasitol. Today 3 (1987) 131-137.
[3] MCDONALD, P.T., HAUSERMANN, N ., LORIMER, N .. Sterility introduced by
release of genetically altered males to a domestic population of Aedes aegypti at the
Kenya coast, Am. J. Trop. Med. Hyg. 26 (1977) 553-561.
[4] RUDNICK, A ., Aedes aegypti and haemorrhagic fever, Bull. WHO 36 (1967)
528-532.
[5] KNIPLING, E.F., et al., Genetic control of insects of public health importance, Bull.
WHO 38 (1968) 421-438.
[6 ] BLACK, W .C ., RAI, К .S., Genome evolution in mosquitoes: Intraspecific and inter­
specific variation in repetitive DNA amounts and organization, Genet. Res. 51 (1988)
185-195.
[7] COCKBURN, A .F ., MITCHELL, S.F., Repetitive DNA interspersion patterns in
Diptera, Arch. Insect. Biochem. Physiol. 10 (1989) 105-113.
[8 ] WARREN, A .M ., CRAMPTON, J.M., The Aedes aegypti genome: Complexity and
organization, Genet. Res. 58 (Í991) 225-232.
[9] BESANSKY, N.J., POWELL, J.R., Reassociation kinetics of Anopheles gambiae
(Diptera:Culicidae) DNA, J. Med. Entomol. 29 (1992) 125-128.
[10] MILLER, L.H.', et al., Stable integration and expression of a bacterial gene in the
mosquito Anopheles gambiae, Science 237 (1987) 779-781.
[11] McGRANE, V ., CARLSON, J.O., MILLER, B.R., BEATTY, B.J., Microinjection
of DNA into Aedes triseriatus ova and detection of integration, Am. J. Trop. Med.
Hyg. 39 (1988) 502-510.
[12] MORRIS, A .C ., EGGLESTON, P., CRAMPTON, J.M., Genetic transformation of
the mosquito Aedes aegypti by micro-injection of DNA, Med. Vet. Entomol. 3 (1989)
1-7.
[13] STELLER, H., PIRROTTA, V ., Transposable P vector that confers selectable G418
resistance to Drosophila larvae, EMBO J. 4 (1985) 167-171.
[14] SPRADLING, A .C ., RUBIN, G .M ., Transposition of cloned P elements into
Drosophila germ line chromosomes, Science 218 (1982) 341-347.
[15] LYCETT, G ., EGGLESTON, P., CRAMPTON, J.M., DNA transfection of an Aedes
aegypti mosquito cell line, Heredity 63 (1989) 277.
[16] LYCETT, G.J., DNA transfection of mosquito cells in culture, Insect Mol. Genet.
Newsl. 4 (1990) 1-3.
[17] DURBIN, J.E., FALLON, A .M ., Transient expression of the chloramphenicol acetyl
transferase gene in cultured mosquito cells, Gene 36 (1985) 173-178.
[18] FELGNER, P .L., et al., Lipofection: A highly efficient, lipid mediated DNA transfec­
tion procedure, Proc. Natl. Acad. Sci. USA 84 (1987) 7413-7417.
[19] KOVACH, M.J., CARLSON, J.O., BEATY, B.J., A Drosophila metallothionein
promoter is inducible in mosquito cells, Insect Mol. Biol. 1 (1992) 37-43.
 IAEA-SM-327/5 47

[20] LYCETT, G.J., CRAMPTON, J.M., Transient and stable expression of a heat
inducible reporter gene in an Aedes aegypti cell line, Insect Biochem. Mol. Biol, (sub­
mitted for publication).
[21] O ’BROCHTA, D .A ., Genetic transformation and its potential in insect pest control,
Bull. Entomol. Res. 80 (1990) 241-244.
[22] GALE, K.R., CRAMPTON, J.M., The ribosomal genes of the mosquito, Aedes
aegypti, Eur. J. Biochem. 185 (1989) 311-317.
[23] PASKEWITZ, S.М ., COLLINS, F.H., Site-specific ribosomal DNA insertion ele­
ments in Anopheles gambiae and A. arabiensis: Nucleotide sequence of gene-element
boundaries, Nucleic Acids Res. 17 (1989) 8125-8133.
[24] CRAMPTON, J.M., MORRIS, A .C ., LYCETT, G.J., WARREN, A .M .,
EGGLESTON, P., Transgenic mosquitoes: A future vector control strategy? Parasitol.
Today 6 (1990) 31-36.
[25] CRAMPTON, J.M., MORRIS, A .C ., LYCETT, G.J., WARREN, A .M .,
EGGLESTON, P., “ Molecular characterisation and genome manipulation of the
mosquito, Aedes aegypti” , Molecular Insect Science (HAGEDORN, H.H.,
HILDEBRAND, J.G., KID WELL, M .G ., LAW , J.H., Eds), Plenum Press, New
York (1990) 1-11.
[26] WARREN, A ., CRAMPTON, J.M., Identification of retrotransposon-like elements in
the genomes of the mosquitoes, Aedes aegypti and Anopheles gambiae, Insect Biochem.
Mol. Biol, (submitted for publication).
[27] JAMES, A .A ., BLACKMER, K., RACIOPPI, J.V., A salivary gland-specific,
maltase-like gene of the vector mosquito, Aedes aegypti, Gene75 (1989) 73-83.
[28] BINGHAM, P.M ., LEVIS, R., RUBIN, G .M ., Cloning of DNA sequences from the
white locus o fû . melanogaster by a novel and general method, Cell 25 (1981) 693-704.
[29] MACDONALD, W .W ., RAMACHANDRAN, C.P., The influence of the g e n e /"
(filarial susceptibility, Brugia malayi) on the susceptibility of Aedes aegypti to seven
strains of Brugia, Wuchereria and Dirofilaria, Ann. Trop. Med. Parasitol. 59 (1965)
64-73.
[30] COLLINS, A .F ., et al., Genetic selection of a Plasmodium refractory strain of the
malaria vector Anopheles gambiae, Science 236 (1986) 607-610.
[31] WINGER, L., et al., Ookinete antigens of Plasmodium berghei: The appearance of a
21 kd transmission blocking determinant on the developing ookinete, Parasite
Immunol. 10 (1987) 193-207.
[32] WOOD, R.J., COOK, L .M ., HAMILTON, A ., WHITELAW, A ., Transporting the
marker gene re (red eye) into a laboratory cage population of Aedes aegypti (Diptera:
Culicidae), using meiotic drive at the M D locus, J. Med. Entomol. 14 (1977)
461-464.
[33] CURTIS, C .F., GRAVES, P.M ., Methods for replacement of malaria vector popula­
tions, J. Trop. Med. Hyg. 91 (1988) 43-48.
 IAEA-SM-327/4

ERADICATION OF THE MELON FLY

FROM OKINAWA, JAPAN, BY MEANS
OF THE STERILE INSECT TECHNIQUE

M. YAMAGISHI, H. KAKINOHANA, H. KUBA,

T. KOHAMA, Y. NAKAMOTO, Y.'SOKEI, K. KINJO
Okinawa Fruit Fly Eradication Project Office,
Naha, Okinawa,
Japan

Abstract
ERADICATION OF THE MELON FLY FROM OKINAWA, JAPAN, BY MEANS OF
THE STERILE INSECT TECHNIQUE.
A national project to eradicate an introduced pest, the melon fly, from Okinawa was
carried out from 1972 to 1992 using the sterile insect technique. After the release of about
50 000 million sterile flies, involving about ¥. 10 000 million as investment, the melon fly
was completely eradicated. The most important conditions for success were the maintenance
of a quality mass reared strain and availability of precise information on the temporal and spa­
tial distribution of wild and released flies for improvement of the release plan.

1. HISTORY

The project to eradicate the melon fly (B a c t r o c e r a c u c u r b i t a e ) from Okinawa

Prefecture was begun in 1972. There are many islands between mainland Japan and
Taiwan (China) and these are called the Southwestern Islands. Okinawa Prefecture
is located at the southwestern part of the island group, while the northern part
belongs to Kagoshima Prefecture. Okinawa Prefecture consists o f three groups of
islands, the Okinawa Islands, the Miyako Islands and the Yaeyama Islands (Fig. 1).
All are situated in the subtropical rain forest zone. Although most parts are now culti­
vated, rain forests still remain in the mountainous areas of Iriomote and Ishigaki and
the northern part of Okinawa Hontô (the main island of Okinawa). The population
density of the melon fly in these rain forest areas was low because host plants are
not abundant. The Miyako Islands and the southern part of Okinawa Hontô are flat
and the original vegetation has been destroyed and turned into crop fields and bush.
The melon fly density in these areas was high.
Figure 1 also shows the spread of distribution of the melon fly to these islands.
The melon fly was first discovered on Ishigaki Island in 1919, and it then invaded
the Miyako Islands in 1929. It was found on Kume Island in 1970 and spread to
Okinawa Hontô in 1972. After that, the distribution range widened rapidly to the
northern islands belonging to Kagoshima Prefecture [1].

49
 Kyushu, Japan

Kagoshima Prefecture

0, 100 200km
'O

Taiwan (China)
1929

1919 Miyako
Islands

Yaeyama
Islands

FIG. 1. Distribution and eradication o f the melon fly in the Southwestern Islands, Japan. Unboxed and boxed num­
bers show years o f distribution and eradication, respectively. The broken line shows the border of -the-Prefecture.
 IAEA-SM-327/4 51

In Okinawa, a pilot eradication project was initiated on Kume Island in 1972,

and the species was successfully eradicated in 1978 from this island [1, 2]. A large
scale project to eradicate the melon fly from all of Okinawa was then carried out in
the following sequence: Miyako Islands (from 1984 to 1987), Okinawa Islands (from
1986 to 1990) and Yaeyama Islands (from 1989 to now).

2. MASS REARING, STERILIZATION, RELEASE AND MONITORING

For this large project, the Government established large mass rearing and
irradiation facilities which could produce a maximum of more than 200 million flies
(males and females) per week [3]. Although the composition of larval and adult diets
was basically the same as that developed in Hawaii, it must be noted that we used
(a) pumpkin juice for the oviposition stimulant; (b) a round surface egging apparatus
(in place of a flat surface of gauze, as in many medfly mass rearing facilities), and
(c) natural day length in the adult room [1,3,4]. These rearing conditions are differ­
ent from many mass rearing systems of fruit flies used in other countries. We decided
to use them for the maintenance of natural genes in our mass rearing stock. Although
the male mating competitiveness of long term mass reared fly stock is weaker than
that of wild stock (see Ref. [5]), the quality of mass reared/sterilized males was kept
at a reasonable level until the last stage of the project.
The pupae were irradiated by gamma rays at a dose of 70 Gy three days before
adult eclosion. The pupae were mixed with fluorescent dye and sent by aircraft to
the target islands. Although pupal release was accomplished in the Kume
project [1, 2], we later used a chilled adult release system. Chilled adults were
distributed over the islands from helicopters.
For monitoring, traps baited with cuelure arid insecticide (naled, etc.) were set
in the target areas. The total number of trap sites was about 550. The traps were sur­
veyed once every two weeks and captured specimens were checked for fluorescent
dye marking under an ultraviolet light; thus sterile and wild flies were counted
separately. Another method used to evaluate the efficacy of SIT was to examine the
infestation rate of host plants. Thousands of host fruits, mainly fruits of a wild cucur­
bit, B r y o n o p s i s l a c i n i o s a , were collected once a month.

3. ESTIMATION OF POPULATION DENSITY

Before beginning the mass releases of sterile flies, we estimated the population
density of the wild fly in various vegetations using a mark-recapture method. A
modified Jackson positive method (the method of Itô and Hamada, Refs [6, 7]) and
the Jackson negative method [8] were used to calculate the density and survival rate
of wild males. We found that the density of the melon fly in mountainous areas was
 Additional release

No. of flies per 1000 trap-days

1983 1984 1985 1986 1987

Year

FIG. 2. Monthly change in the abundance of sterile (S) and wild (W) melon flies caught by monitor traps on the Miyako Islands.
 IAEA-SM-327/4 53

usually less than 10/ha [9], while the density in crop fields and bushy areas some­
times reached 600/ha [7, 10]. As the maximum density in non-mountainous areas
was higher than the capability of the mass production facilities, we suppressed the
fly population in high density areas using the male annihilation technique before
releasing sterile flies. Cotton strings soaked in cuelure and insecticide were
distributed from a helicopter at a dose of 40 strings per hectare per month. We
carried out this population suppression technique to decrease the density of wild flies
to one tenth to one twentieth of the natural density.

4. ERADICATION FROM THE MIYAKO ISLANDS

Figure 2 shows the monthly change in the number of sterile and wild melon
fly males caught by monitor traps in the Miyako Islands (total area: 227 km2).
Sterile fly release on the Miyako Islands began in August 1984; at first 30 million
flies per week were released evenly in the islands. During the initial seven months,
the number of sterile males caught by traps was low (about 100 per 1000 trap-days)
because of low temperature and low fly quality. Early in 1985, the fly quality was
improved and the number of sterile flies increased to about 10 000 per
1000 trap-days.
Figure 3 shows the local distribution of the abundance of sterile and wild melon
flies in August 1985. There were high density areas in some parts of Shimoji Village,
which is a well known vegetable producing area. We therefore decided to release
additional flies into these high density areas. Figure 4 shows the additional release
area in the Miyako Islands. At first, an additional six million flies were released per
week in the Shimoji area from October 1985. The second and third additional
releases were carried out in high density areas. The maximum number of flies
released in the Miyako Islands was 48 million per week. Figure 5 shows the distribu­
tion and abundance of flies caught in June 1986. At this time, wild flies were caught
at only a few points and the number was very small. Since February 1987, no wild
flies have been caught in the Miyako Islands.
Figure 6 shows the change in infestation rate of host fruits. Until April 1986,
the infestation rate lay between 0.1 and 10%, but it rapidly decreased from
May 1986, and became zero in November 1986. In November 1987, the Govern­
ment announced that the melon fly had been eradicated from the Miyako Islands after
the release of 6340 million sterile flies.

5. ERADICATION FROM THE OKINAWA ISLANDS

The project to eradicate the melon fly from the Okinawa Islands was begun in
November 1986, while the Miyako project was still under way. The release of sterile
 t-Л
(a). (b)

YAMAGISHI et al.
Ueno

• ° оо о о ОО
О 1-3 4-9 10-20 21-50 51-100 101-200 201-

FIG. 3. Distribution and abundance o f (aj sterile ( о ) and (b) wild (• ) melon flies caught
by monitor traps on the Miyako Islands in August 1985. The size of the circles indicates the
number o f flies trapped.
 ^\^^kerna

IAEA-SM-327/4.
>?>: First additional release (six million)
from October 1985

Second additional release (six million)

s if e from February 1986

[| Third additional release (eight million)

Ш ill from May 1986

FIG. 4. The additional release areas on the Miyako Islands.

 (a) (b)

YAMAGISHI et al.
• ° о ОО ООО
О 1-3 4-9 10-20 21-50 51-100 101-200 201-

FIG. 5. Distribution and the abundance o f (a) sterile ( о ) and (b) wild ( • j melon flies caught by monitor traps in June 1986.
 140000

30000

No. of surveyed fruits

Infestation rate (%)

20000

10000

Year

Ln
FIG. 6. Number of surveyed host fruits (histograms) and rate of melon fly infestation ( • ) on the Miyako Islands.
58 YAMAGISHI et al.

1986 1987 1988 1989 1990

Year

FIG. 7. Monthly change in the abundance of sterile (Q ) and wild (• ) melon flies caught
by monitor traps in the Okinawa Islands.

flies was begun from the southern part of Okinawa Hontô; the target areas were then
expanded to the northern part and adjacent islets. As in the Miyako Islands, wild flies
were caught until 1989 in the main crop producing areas and concentrated releases
were made into these areas. The maximum number of sterile flies released in the
Okinawa Islands was about 170 million per week.
Figure 7 shows the results of monitor trap surveys on the Okinawa Islands.
Since December 1989, wild flies have not been detected. The Government
announced in November 1990 that the eradication of the melon fly from the
Okinawa Islands had been achieved. The total number of sterile flies released was
30 940 million.

6. ERADICATION FROM THE YAEYAMA ISLANDS

The project to eradicate the melon fly from the last target area, the Yaeyama
Islands, was begun in October 1989. The number of sterile flies released was
90 million per week. From September 1991, no wild flies were caught with monitor
traps (Fig. 8). We consider that melon flies from the Yaeyama Islands were eradi­
cated. About 12 000 million sterile flies are planned to be released in this area until
March 1993.
 IAEA-SM-327/4 59

1988 1989 1990 1991 1992

Year

FIG. 8. Monthly change in the abundance of sterile (©) and wild (• ) melon flies caught
by monitor traps in the Yaeyama Islands.

7. CONCLUSIONS

It can be concluded that the melon fly has been completely eradicated from
Okinawa. The total number of flies released (from the Kume to Yaeyama projects)
was about 50 000 million, and the total investment was about ¥. 10 000, including
¥. 3600 million for the construction of facilities.
As the melon fly was also eradicated from the Amami Islands by the
Kagoshima Prefectura] Government by 1989 [11], Japan now has no melon flies.
This is the first complete success of the sterile insect technique (SIT) eradication
project after 1963 (eradication of the melon fly from Rota [12]).
Our experience, based on the success of the eradication programme using SIT,
shows that the quality of mass reared, sterilized melon flies to be released is of fun­
damental importance and estimation of the wild population density by the mark-
recapture technique and appropriate monitoring systems are necessary; the latter can
detect the temporal and spatial distributions of wild and released insects. Detailed
field studies on the spatial distribution pattern of the target insect are essential for
improvement of the sterile insect release plan. Over the twenty years of our SIT
project, we carried out research on the fundamental biology of the target insect and
published more than one hundred and eighty original papers. Such basic research
might have provided the conditions for our success.
60 YAMAGISHI et al.

ACKNOW LEDGEM ENTS

The authors thank Y. Itô for reading the first draft of the manuscript. They also
thank the past and present staff of the Fruit Fly Laboratory, Okinawa Agricultural
Experiment Station.

REFERENCES

[1] ITÔ, Y., K O Y A M A , J., Eradication of the melon fly: Role of population ecology in
the successful implementation of the sterile insect release method, Prot. Ecol. 4 (1982)
1-28.
[2] I W A H A S H I , O., Eradication of the melon fly, Dacus cucurbitae, from K u m e Is.,
Okinawa, with the sterile insect release method, Res. Popul. Ecol. 19 (1977) 87-97.
[3] K A K I N O H A N À , H., “A plan to construct the new mass production facility for the
melon fly, Dacus cucurbitae Coquillett, in Okinawa, Japan”, Sterile Insect Technique
and Radiation in Insect Control (Proc. Symp. Neuherberg, 1981), IAEA, Vienna
(1982) 477-482.
[4] N A K A M O R I , H., K A K I N O H A N A , H., Mass-production of the melon fly, Dacus
cucurbitae Coquillet, in Okinawa, Japan, Rev. Plant Prot. Res. 13 (1980) 37-53.
[5] ITÔ, Y., Y A M A G I S H I , М., K U B A , H., IAEA-SM-327/41, these Proceedings.
[6] H A M A D A , R., Density estimation by the modified Jackson method, Appl. Entomol.
Zool. 11 (1976) 194-201.
[7] ITÔ, Y., M U R A I , М., T E R U Y A , T., H A M A D A , R., S U G I M O T O , A., A n estima­
tion of population density of Dacus cucurbitae with mark-recapture methods, Res.
Popul. Ecol. 15 (1974) 213-222.
[8] J A C K S O N , C.H.N., The analysis of an animal population, J. Anim. Ecol. 8 (1939)
234-246.
[9] K O Y A M A , J., et al., A n estimation of the adult population density of the melon fly,
Dacus cucurbitae Coquillett (Diptera:Tephritidae) in Okinawa Island, Japan, Appl.
Entomol. Zool. 17 (1982) 550-558.
[10] I C H I N O H E , F., T A N A K A , K., M I U R A , M., ITÔ, Y., A n estimation of the winter
population density of Dacus cucurbitae Coq. in southern Okinawa and the effect of
g a m m a radiation on the released male flies, Appl. Entomol. Zool. 13 (1978) 316-318.
[11] Éradication of the Melon Fly from A m a m i Islands, special publication, Kagoshima
Prefecture (1991) (in Japanese).
[12] STEINER, L.F., H A R M S , E.J., M I T C H E L L , W.C., F U J I M O T O , M.S.,
C H R I S T E N S E N , L.D., Melon fly eradication by overflooding with sterile flies,
J. Econ. Entomol. 58 (1965) 519-522.
GENETIC ENGINEERING AND MOLECULAR BIOLOGY
(Session 2)

Chairmen

J. OA K E SH O TT
Australia
A .S . ROBINSON

United Kingdom
 IAEA-SM-327/27

ADVANCES IN THE PRESERVATION

OF INSECT GERMPLASM

R.A. LEOPOLD
Biosciences Research Laboratory,
Agricultural Research Service,
United States Department of Agriculture,
Fargo, North Dakota,
United States of America

Abstract

A D V A N C E S IN T H E P R E S E R V A T I O N O F I N S E C T G E R M P L A S M :
The current means of preserving insects that are freezing intolerant or have no
dormancy capabilities for use in the laboratory or in management programmes is by continu­
ous culture. Not only can continuous culture be a costly venture, but it can effect genetic drift
and is subject to accidental loss of colonies, genetic strains and transformants. Further, the
ability to be able to stockpile insects for later use in sterile insect technique and biocontrol
programmes would be of tremendous benefit. Since preservation of mammalian embryos by
low temperature technology has become a c o m m o n procedure, researchers, insectary
managers and those involved in control programmes have been looking to cryobiologists for
assistance in solving the insect germplasm storage problem. The paper examines the concepts
of the conventional methodology that is used for cryopreservation of cells and mammalian
embryos. Also pointed out are several inherent barriers posed by embryos of insects such as
muscoid flies which are incompatible with the use of the conventional techniques. Of the
obstacles thus far identified, chilling intolerance and egg membrane impermeability have been
given the most attention by researchers attempting to develop low temperature storage
methods. Limited but promising success has been obtained using chemical dissolution of
membrane waxes, infusion of embryos with multimolar ciyoprotectants and avoidance of
chilling injury by ultrarapid cooling and warming. The feasibility of incorporating techniques
which facilitate natural insect cold hardiness into a cryopreservation protocol and alternatives
to preservation of embryos are discussed.

1. INTRODUCTION

The demand for a long term storage methodology for insects has grown sub­
stantially in recent years with its successful use in sterile insect technique (SIT) and
biocontrol programmes and with the advancements made in genetic engineering and
molecular techniques. The need to increase the shelf-life of those insects having high
genetic control or research utility relates directly to reducing costs in the main­
tenance of large numbers of genetic strains, combating genetic drift and catastrophic
loss under mass rearing conditions while having the capability to stockpile insects

63
64 LEOPOLD

prior to release in a management system. To meet these demands, research was

initiated about a decade ago by the United States Department of Agriculture in an
attempt to transfer to insects the cold storage technology developed for mammalian
embryos that is routinely used for preservation of domesticated and some wild zoo
animals. The efforts made by Heacox et al. [1] to use these standard cryopreservation
methods for preservation of M u s c a d o m e s t i c a embryos were largely unsuccessful,
but their research identified several important barriers, unique to insects, that
precluded the use of the then current ‘ state of the art’ technology for preserving
animal embryos. Since that initial study, others have joined the effort to provide
researchers and those involved in insect management with long term storage methods
that employ cryopreservation techniques. Thus, the objective of this paper is to
review the most recent findings that relate to the preservation of insect germplasm
and hopefully to give some insight to where efforts might be directed to achieve the
goal of providing storage methods for certain insects. Because many insects have
inherent strategies to survive subzero temperatures, this author feels that it is impera­
tive that discussion be included of how these mechanisms could (and possibly should)
be incorporated into certain preservation techniques. Further, since previous reports
[2, 3] have summarized the early work on preserving insects in cold storage prior
to the successful use of chemical cryoprotectants in the preservation of mousë
embryos [4], mention of those studies is limited to developing the background to or
an understanding of the most recent research efforts.

2. CRYOPRESERVATION: BASIC PRINCIPLES

The conventional method for cryopreservation of mammalian embryos, mul-

ticellular systems and cells requires that sufficient dehydration of the intracellular
domain be achieved prior to cooling in order to avoid lethal ice formation. This is
accomplished with molar concentrations of protective solutes that also protect cells
from a number of potentially lethal chemical and physical changes that can occur
upon freezing (Fig. 1).

2.1. Cooling rate

The cooling rate is a critical factor in conventional cryopreservation. Under

slow cooling conditions, most cells and embryos tend to supercool, even in the
presence of extracellular ice. Slow cooling allows cells to dehydrate osmotically
through the difference in the chemical potential of the supercooled water within the
cells and that of the water and ice in the surrounding medium [5]. The bulk water
content of cells, as opposed to the osmotically inactive water, can be reduced to a
low level via slow cooling and cryoprotectant loading and ice will either not form
upon reaching the normal nucléation temperature of the cells or will be present in
 IAEA-SM-327/27 65

CHEMICAL AND PHYSICAL

CHANGES

Solute concentration
pH shift
Enzyme reaction rates '
Structural distortion

Structural damage ■
Ice surface effects
Proton mobility
Dielectric constant
Enzyme reaction rates

Structural damage
(recrystallization)
pH shift
Enzyme reaction rates

FIG. 1. Effects of three arbitrarily selected cooling rates on the movement and freezing of
intracellular water in the presence of extracellular ice. Also shown are the chemical andphysi­
cal changes that accompany cell dehydration and intracellular freezing.

quantities too small to become damaging. When cells are cooled at faster rates, the
water efflux is limited and freezing occurs intracellularly (Fig. 1). For each system
to be frozen there is an optimum cooling rate and cooling too slowly can be as harm­
ful as cooling too rapidly. Cell shrinkage below a critical volume and severe defor­
mation within the unfrozen fraction of water remaining in the ice channels have been
suggested as the source of damage caused by a suboptimal rate of cooling [6, 7].

2.2. W arming rate

The warming process (fast versus slow) after storage at liquid nitrogen (LN2)
temperature appears to be directly related to the rate that the cells or embryos were
cooled and at what subzero temperature the transfer to liquid N2 was made [8].
Cells sensitive to slow warming are thought to have been cooled too rapidly,
allowing the formation of small intracellular ice crystals, which grow by migratory
recrystallization and create damage during the thaw [9]. Thus, fast warming
(>300°C/min) is needed to prevent damaging recrystallization. Cells or embryos
requiring slow warming (<25°C/min) have usually been cooled at a slow rate
(<l°C/m in) and may also have been transferred to LN2 at temperatures below
-60°C . In this case considerable dehydration has occurred, leading Leibo et al. [10]
to propose that rapid warming causes injury to cells and embryos through the osmotic
66 LEOPOLD

stress created by rapid rehydration as the extracellular ice melts. However, others
believe that neither recrystallization, melting of intracellular ice nor osmotic stress
are major factors in causing injury upon warming but that it is some as yet unidenti­
fied component [8, 11]. A schematic representation of a conventional method of
cryopreservation using, cryoprotectant loading and slow cooling is depicted in Fig. 2.

3. INHERENT CRYOPRESERVATION BARRIERS

In general, cryopreservation procedures require that the’ membranes of cells

and embryos be permeable to water and the solutes used for cryoprotection. Insects,
except for the viviparous species and those ovipositing in protected humid environ­
ments, lay eggs which are endowed with elaborate membrane systems and protective
chorions that impede water loss and entry of harmful chemicals.

3.1. Permeabilization

. The egg chorions of Drosophila melanogaster, M. domestica and other mus-

coid flies are easily removed by immersing in dilute commercial bleach or by
mechanical means. However, much effort has been spent rendering the underlying
vitelline membrane permeable. A hydrophobic waxy layer is associated with the
vitelline membranes surrounding D. melanogaster and M. domestica embryos and
embryos of many other insects [12]. Limbourg and Zalokar [13] devised a method
for permeabilizing vitelline membranes using octane. Heacox et al. [1] modified this
method slightly and used it for permeabilizing the vitelline membranes of M. domes-
ticà embryos before treating with dimethylsulfoxide. The inconsistent results and
reduced hatchability of D. melanogaster embryos have led researchers to make
several changes in the original procedure. The apparent essential steps of the most
recent methods employ either an initial rinse in isopropanol before immersion in
hexane or treating the embryos with a mixture of 1-butanol and an alkane after the
isopropanol rinse [14-16]. It has been reported that both methods yield >90% per­
meabilization and from 70-90% hatching.

3.2. The yolk system

Intuitively, the amount and/or type of yolk present in insect eggs presents a
potential barrier to cryopreservation. Since thè insect egg is a self-sustaining
developmental system, all the materials needed for embryogenesis must be contained
within the yolk nutrients. For example, the lipid content of insect eggs typically
ranges from about 1.5 to 18.5% [17]. On the basis of measurements of respiratory
metabolism and a decrease in the lipid content during embryogenesis, stored lipids
are the main source of energy for the developing embryos of most1of the insects
 IAEA-SM-327/27 67

CONVENTIONAL METHOD
(1 )
PERMEABILIZATION
NaOCI
Alkane/A Icohol

(2)
(7)
CRYOPROTECTANT RECOVERY
LOADING
+ 5 » — -5«C ' - jjg
H 20
DEHYDRATION
(3) SLOW COOL
-5 ° -*-< -3 0 ° C

(4)
LIQUID N j FAST THAW A HYDRATION
STORAGE
-Î9 6 °-* -+ 3 0 °C <6>
(5)

VITRIFICATION METHOD
(1)
PERMEABILIZATION
. NaOCI — «►
Alkane/Alcohol

(2)
VITRIFICATION (7)
FLUID LOADING RECOVERY
0°C

(3) ULTRA RAPID COOL ; I VITRIFICATION

+0* -205°C 1

(4)
LIQUID N j HYDRATION
STORAGE (6)

FIG. 2. Comparisons of two methods of gaining LN2 storage for an insect embryo. The
essential differences between the two methods are in the amount of cell dehydration obtained
during the loading of cryoprôtectants and cooling, the concentration of cryoprotectants, and
the rates of cooling and warming. ■The asterisk at step 3 of the conventional method denotes
a process which precludes use on chilling intolerant embryos (see text for details).
68 LEOPOLD

studied [18]. Early work by Lea and Hawke [19] has shown that the lipovitellin
component produced by egg laying animals is sensitive to cold temperatures and dis­
associates upon freezing. Further, Lovelock [20] indicated that the protein-lipid
complexes within cells are particularly susceptible to chilling and freezing because
they are held together by only weak associations and not through covalent bonding.
Mazur et al. [21] observed during the testing of their process for cryopreservation
of D . m e l a n o g a s t e r embryos that 47 % developed from the time of treatment (predor­
sal closure) up to hatching, but hatching mostly did not occur. This may be.an indica­
tion that the energy needed for larval emergence from the egg has been lost because
of cold instituted damage to the yolk system.

3.3. Chilling sensitivity

The lack of success thus far in applying conventional cryopreservation proce­

dures for M. d o m e s t i c a and D . m e l a n o g a s t e r has been attributed to their prefreeze
sensitivity to chilling. The theoretical low temperature at which freezing intolerant
insect species such as these two flies can no longer survive is the temperature of crys­
tallization (Tc). This is the subzero temperature at which spontaneous nucléation of
body fluids occurs. Depending on the stage of. development assayed, many insects
can supercool to —25° С or lower before spontaneous freezing of their body fluids
occurs. However, the Tc is not a good indicator of the lowest lethal temperature.
Several recent studies have shown that a substantial number of insects succumb at
temperatures of 10-15°C above the Tc [22-25]. These studies show that insects,
like other organisms ranging from bacteria to higher plants, mammalian sperm and
embryos, are sensitive to the stress caused by exposure to subzero non-freezing
temperatures.
In addition to the possible membrane damage caused by chilling, we have
suggested that sensitivity during the early embryogenesis of insects is probably inher­
ent to the process of rapid cellulation [24]. During blastoderm formation, syn­
chronous mitotic divisions of cleavage energids in D . m e l a n o g a s t e r occur about
every 10 min [26]. Further, Callaini and Marchini [27] have shown that during this
period of cellularization of the blastoderm, chilling for 1 h at 0°C causes the forma­
tion of abnormal centrosomes and treatment of earlier embryos results in the arrest
of cleavage divisions at metaphase.

3 .3 .1 . D i r e c t a n d i n d i r e c t c h i l l i n g in ju r y

Direct chilling injury or ‘cold shock’ is injury caused by a rapid reduction in

temperature, while indirect chilling injury involves damage incurred only after long
exposure to low temperature, and is independent of the rate of cooling [28]. Cold
shock injury in some insects has been linked to membrane failure [29, 30]. The
damage caused by cold shock is thought to be due to the induction of phase transi-
 IAEA-SM-327/27 69

tions in membrane lipids, which results in leakage and loss of the membrane function
[31]. An alternative view is that chilling causes damaging thermoelastic stress
through an unequal condensation of membranes relative to the cell contents [32].
Indirect chilling injury has been suggested to be associated with the formation of
irreversible metabolic imbalances involving cellular energetics [33].

3 .3 .2 . D e v e l o p m e n t a l s t a g e s e n s i t iv it y •

Different developmental stages possess varying capacities for tolerance to both

cold shock and long term chilling. Further, there are also significant differences in
the capacity for cold tolerance within a particular developmental stage of an insect.
As mentioned earlier, these phenomena are important considerations when devising
a cryopreservation protocol, since the conditions for cryoprotectant loading and
dehydration under a conventional procedure require that the cells/embryos be chill­
ing tolerant.
A survey of the within and between stage tolerance to long term chilling
throughout the pre-adult development of the house fly showed that at 5 and 0 °C, the
two-day-old pupae were the most resistant to the effects of long term chilling, while
at -5 ° C the late stage embryos were the most resistant (Fig. 3). Interestingly, three-
day post-emergent adults were as sensitive as the larvae and pupae to —5°C (data
not shown).

300 -

L
~1 I— I— I----- 1— Г 1— I— I— Г 1— I— I— I— Г
2 4 6 8 10 12 1 2 3 4 0 1 2 3 4
Embryos Larvae Pupae
• (h) (d) (d)

FIG. 3. Within and between stage survival to hatching (LT50) o f house fly embryos, larvae
and pupae subjected to long term chilling at three temperatures (modified from Strong-
Gunderson and Leopold [24]).
70 LEOPOLD

The within stage sensitivity of house fly embryos subjected to a cold shock at
a temperature of —20°C is shown in Figs 4(a) and 4(b). House fly embryos super­
cool to a range of -2 6 to -3 4 °C before freezing [24]. As indicated in Fig.. 4(a),
there is a rate limiting effect of cooling on survival up to and including 9 h post-
oviposition. The slower cooling rate of - 10°C/min was clearly better than the faster
rate at all ages, except for 12 h post-oviposition, where they do not differ. Further,
the younger ages are mostly more sensitive to rapid, brief chilling than are the.older
embryos.
When the hold time is extended from 3 to 30 min at —20°C, survival is drasti­
cally reduced (Fig. 4(b)). Only the 9-h-old embryos show any significant survival.
Survival at the two faster rates of cooling, -1 0 and -65°C/min, does not differ,
while cooling at a much slower rate, -0.5°C/min, increases survival almost four
fold.

3.4. Preservation of post-embryonic stages

The foregoing discussion has assumed that the insect embryo would be the
most desirable stage to use in developing a long term storage protocol. This assump­
tion may not be applicable to a particular insect or situation. For example, preserva­
tion of embryos of a larvapositing insect such as the tsetse fly would not be
attainable, since the embryo cannot be maintained outside its mother. Further, it may
be more desirable to stockpile an insect to be used in a control programme in the
stage closest to the one that is to be released. This would allow rapid implementation
of the control programme.
Cryopreservation of the post-embryonic stages generally presents even more
difficult problems than that of embryos. The post-embryonic stages are larger, more
compartmentalized and often possess almost impenetrable cuticular surfaces, which
would obstruct the permeabilizátion and even the distribution of cryoprotectants.
Moreover, the older stages may be more chilling intolerant than their embryonic
counterparts (see Section 3.3).

4. CIRCUMVENTING THE BÁRRIERS

4.1. Cryoprotectants

Avoiding or dealing with the natural barriers that interfere with the use of low
temperature to effect a long term storage condition usually requires that chemical
cryoprotectants be employed in addition to the reduction of freezable water.
Cryoprotectants are variously classified by molecular weight, permeating ability, ice
promoting or deterring and as antifreeze agents. Permeating cryoprotectants of low
molecular weight can protect cells from freezing damage simply on a colligative
 IAEA-SM-327/27 71

Cooled to - 2 0 ° G

Cooled to —20°C

FIG. 4. (a) Survival to hatching o f house fly embryos subjected to à short term cold shock
and cooled to —20°С at two different ratés, (b) Similar to (a) except that chilling was extended
to 30 min and an additional cooling rate o f 0.5°C/min was tested for the 3, 6 and 9 h (lines
not connected for clarity) .. The warming rates for all the groups were the same as each'respec­
tive cooling rate.
72 LEOPOLD

basis and have the capacity to keep extra- and intracellular solutes from reaching
toxic levels by remaining in the aqueous phase during the freezing process. Thus,
cell dehydration and shrinkage is reduced, electrolyte balance is maintained and the
amount of extracellular water remaining as a liquid increases [6, 7, 34]. Others have
suggested that alternative or additional actions of permeating cryoprotectants may be
to stabilize the unfreezable water [34], to reduce cellular damage by changing the
ice crystal configuration [35] and to resist the dénaturation of the essential macro­
molecules during subzero dehydration [33].
Non-permeating cryoprotectants such as trehalose appear to protect cell mem­
branes during and upon recovery from chilling and/or freezing [36]. Combining tre­
halose with glycerol was found to increase the viability of mouse embryos upon
recovery from a freeze-thaw process [37]. The protective action of trehalose has
been attributed to take the place of the H20 molecules that normally hydrate mem­
brane phospholipids [38]. Further, several insects have been found to accumulate
large amounts of trehalose (4 0 -7 0 pig/mg body weight) during the overwintering
period, leading to the speculation that trehalose gives protection from chilling/freez­
ing injury [39, 40].

4.2. Vitrification

If M. domestica and D. melanogaster were not chilling intolerant, the calcu­

lated optimum cooling rate for permeabilized embryos would be < l°C/m in [1, 41].
In theory, this rate would effect the efflux of water, the influx of a cryoprotectant
and allow a conventional slow freezing strategy to be used. One means of dealing
with high chilling sensitivity is to cool the embryos at extremely rapid rates. Mazur
et al. [42] have calculated that a cooling rate of > 2 0 000°C/m in is needed to reduce
the median lethal temperature of D. melanogaster embryos from - 2 5 to - 6 5 ° C .
Further, to avoid lethal ice formation at such high cooling rates, glass forming
solutes must be infused into the embryos which, upon cooling, would produce a vitri­
fied state. Simply stated, vitrification is a process whereby a liquid through rapid
cooling becomes solidified, not by the formation of ice, but by an extraordinary
increase in viscosity. The way in which conventional and vitrification procedures
differ is that highly concentrated solutions ( > 5 0 wt%) are used to dehydrate the
embryos and to concentrate the intracellular solutes. When a concentrated
cryoprotectant is present both inside and outside the embryo and cooled at ultrarapid
rates, an amorphous glass is formed in the surrounding medium and the embryo that
precludes the formation of ice. Also, very rapid warming of the vitrified specimen
must also be employed to eliminate the possibility of devitrification, which results
in the transient formation of damaging ice crystals. (Figure 2 gives a graphic com­
parison of the conventional and vitrification methods.)
Steponkus et al. [43] were the first to report successful use of a vitrification
protocol to gain survival of D. melanogaster embryos after storage at —196°C .
 IAEA-SM-327/27 73

Their two stèp method for infusing 12-h-old embryos with 2 . 1M and then 8.5M ethy­
lene glycol (EG ), followed by cooling at 54 000°C/m in in liquid propane, resulted
in an 18% hatch, 3% of which developed into adults. Mazur et al. [21] reported
slightly better results by using 14-15-h-old embryos, altering the permeabilization
solution and cooling at rates approaching 100 000°C/m in in LN2 slush. However,
recent refinements made by Steponkus and Caldwell [44] have raised the hatching
to near 50% and survival to adulthood to about 10%. Their refinements include
optimizing the permeabilization procedure [16], choosing 13.5-14.5-h-old embryos
for treatment, increasing the duration of time for the exposure to EG, using LN2
slush as a coolant and increasing the osmotic potential of the vitrification solution.
Mazur et al. [21, 42] have made extensive studies on the cooling and warming
rates required to circumvent the chilling sensitivity of Drosophila embryos during
vitrification and they contend that very high warming rates are more important to
survival than the cooling rates. It was suggested that the requirement for rapid warm­
ing was an indication that devitrification was occurring and was caused by the uneven
distribution of EG in the compartmentalized embryo.

4.3. Enhancing cold tolerance for cryopreservation

Development of a cryopreservation procedure can be aided considerably by

judiciously choosing the most cold tolerant stage. Many insects respond to certain
environmental cues by producing natural cryoprotectants in the form of polyols,
sugars and antifreeze or ice nucleating proteins, which enable them to become more
cold tolerant (see reviews of Storey and Storey [33] and Duman et al. [45]). Even
insects that ,do not exhibit overwintering strategies or experience dormancy may be
induced to increase their cold tolerance. Czajka and Lee [46] found that
D. melanogaster larvae, pupae and adults had increased tolerance to subzero chilling
if they were first exposed to 0 or 5°C for 1 h. It has been suggested that this rapid
cold hardening phenomenon enables insects to deal with a short term drop in the
ambient temperature [47].
When house fly embryos are subjected to a cold hardening treatment at 0°C
before cooling to - 2 0 ° С at a rapid rate, survival was observed to increase with the
6-, 9- and 12-h-old embryos but not with the 1- and 3-h-old embryos (Table I). Slow
cooling and the hold at 0°C appear to have the same cold hardening effect, since
cooling of the 9-h-oId embryos at a rate of —0.5°C /m in from 2 0 to 0°C before fast
cooling to - 2 0 ° C produce almost identical survival rates. Mazur et al. [42], con­
cluded that for D. melanogaster embryos, cold hardening cannot alleviate the need
for very high cooling rates ( > 1 0 0 000°C/m in) to out run the accelerating effects of
chilling injury at low temperatures.
Burton et al. [47] found a connection between the heat and cold shock
phenomena in that a moderate heat shock could protect D. melanogaster larvae from
cold shock injury. Exposure of house fly embryos to a heat shock before cooling to
74 LEOPOLD

TA BLE I. PRETREATMENT OF HOUSE F L Y EMBRYOS WITH COLD

HARDENING OR HEAT SHOCK BEFORE COOLING TO - 2 0 ° C

Mean per cent survival to hatching

Embryonic age (h)
No cold hardening
Cold hardened3 Heat shocked^
or heat shock

1 0 0 0

3 0 0 0

6 0 .2 3.7 0

9 9.7 17.7 (17.6)c 2 2 .1

12 0.3 25.4 0.3 .

a 1 h at 0°C before cooling to -2 0 ° C at 65°C/min, holding for 30 min and warming at the
same rate.
b 30 min at 38°C before cooling, holding and warming as above.
c Cooled at 0.5°C/min from 20° to 0°C before cooling, holding and warming as above.

- 2 0 ° С imparted increased chilling tolerance only to the 9 h embryos (Table I). The
mechanism of how heat shock or rapid cold hardening increases tolerance to chilling
is not understood. More information on the etiology of chilling injury in insect
embryos and how cold and heat stress tend to reduce its severity is needed before
determ ining whether it can be used in a cryopreservation protocol.
Besides cold or heat stress, other factors such as changes in the photoperiod,
humidity, diet quality and quantity and oxygen tension, or a combination of these
factors, have been reported to enhance cold tolerance [4 8 -5 1 ]. Even oral injections
of glycerol given to the larvae of Añagasta kuehniella resulted in an increase in short
term low temperature survival [52]. In most cases, gaining enhancement of the cold
tolerance of an insect requires that its low temperature biology and physiology be
thoroughly studied, and understood.

5. ALTERNATIVES TO EMBRYO CRYOPRESERVATION

The alternatives to embryo cryopreservation include storage of the post-

embryonic stages, gonads or primordial germ cells (pole cells) for transplantation,
cleavage energids, sperm and unfertilized eggs. Except for the preservation of post-
embryonic stages, the other possibilities are labour intensive and involve complex
manipulative techniques such as microsurgery and injection. Techniques for the
cryopreservation of gonadal imaginai discs and germ cells have been developed, so
for some insects these methods offer viable options (see Ref. [3]). Also, it should
 IAEA-SM-327/27 .75

be pointed out that the preservation of gonads and germ cells allows only one-half
the genome to be retained. However, recent success with the in vitro fertilization of
sawflies [54], if expandable to other insects, would alleviate this problem. The
current obstacle lies in the lack of knowledge on egg activation occurring at
fertilization.

6. SUMMARY

The cryopreservation of insect germplasm is still largely in the developmental

stages, and for those insects for which long term low temperature storage techniques
have been developed, the methods are cumbersome, require considerable expertise
and result in a relatively low yield of viable adults. This is not unexpected, in a new
area of endeavour where the research effort is being conducted by relatively few
scientists on limited budgets. However, considering the progress made over the last
five to six years, it can be anticipated that the technology for storage of D r o s o p h i l a
spp. and related flies will be forthcoming. The task of developing less complicated,
universally utilizable methods will be made easier once effective and efficient means
of dealing with chilling injury and cryoprotectant permeation of complex systems are
found. There is as yet much to be learned from the study of insect cryobiology which
no doubt will have application to the development of cryopreservation methods.
Given the amount of diversity among insects, the only limitations to solving
problems in the preservation of insect germplasm lie in the imagination of the scien­
tists and the provision of adequate funding.

REFERENCES

[1] HEACOX, A .E ., LEOPOLD; R: A ., BRAMMER, J.D., Survival of house fly embryos

cooled in the presence of dimethylsulfoxidè, Cryo-Lett. 6 (1985) 305-312.
[2] LEOPOLD, R .A ., ROJAS, R.R., “ Invertebrate germplasm cryopreservation: Poten­
tial, problems and prospects” , Biotic Diversity and Germplasm Preservation, Global
Imperatives (Proc. Annu. Symp. Beltsville, 1988) (KNUTSON, L., STONER, A .K .,
Eds), Kluwer, Boston, MA (1989) 355-377-.
[3] LEOPOLD, R .A ., “ Cryopreservation of insect germplasm: Cells, tissues and organ­
isms” , Insects at Low Temperature (LEE, R.E., DENLINGER, D .L ., Eds), Chapman
and Hall, New York (1991) 379-407. '
[4] WHITTINGHAM, D .G ., LEIBO, S.P., MAZUR, P., Survival of mouse embryos
frozen to - 1 9 6 ° and 269°C, Science 178 (1972) 411-414.
[5] MAZUR, P., Kinetics of water loss from cells at subzero temperatures and the likeli­
hood of intracellular freezing, J. Gen. Physiol. 47 (1963) 347-369.
[6 ] MAZUR, P., Freezing of living cells: Mechanisms and implications, Am. J. Physiol.
247 (1984) 125-142.
76 LEOPOLD

[7] M E R Y M A N , H .T . , F re e zin g injury and its prevention in liv in g cells, A n n u . Rev.

B io p h y s. Bioe ng. 3 (197 4) 34 1 -3 6 3 .
[8] R A L L , W . F . , R E I D , D . S . , P O L G E , G ., A n a ly s is o f slo w w a rm in g injury o f m ouse
em b ryos b y cryom icroscopical and physiochem ical methods, C ry o b io lo g y 21 (1984)
1 0 6-121 .
[9] M A Z U R , P., “ P h y sic a l and chem ical b asis o f injury in single-celled organ ism s sub­
jected to freezing and th a w in g” , C ry o b io lo g y ( M E R Y M A N , H .T . , Ed.), A cad e m ic
Press, N e w Y o r k (1966) 2 1 4 -3 1 6 .
[10] L E I B O , S .P ., M A Z U R , P ., J A C K O W S K I , S .C ., Factors affecting su rviv a l o f m ouse
em b ryos du rin g freezing and thawing, E x p . C e ll Res. 8 9 (1974) 7 9 -8 8 .
[11] M A C K E N S I E , A .P ., “ D eath o f frozen yeast in the course o f fre ezin g” , T h e Froze n
C e ll (Proc. C ib a Foundation Sym p. 1970) ( W O L S T E Ñ H O L M E , G .E ., O ’C O N N O R ,
М . , E d s), C h u rch ill, L o n d o n (1970) 8 9 -9 6 .
[12] M A R G A R I T I S , L .H ., “ Structure and p h y sio lo g y o f the egg sh e ll” , C om p re he nsive
Insect' P h y sio lo g y , Biochem istry and Ph arm acology (K E R K U T , G .A .;
G I L B E R T , L .I. , E d s), V o l. 1, Pergam on Press, L o n d o n (1985) 15 4-230 .
[13] L I M B O U R G , B ., Z A L O K A R , М ., Perm eabilization o f Drosophila eggs, D e v . Biol.
3 5 (1973) 3 8 2 -3 8 7 .
[14] L Y N C H , D . V . , et al., A two-step method for perm eabilization o f Drosophila eggs,
C ry o b io lo g y 2 6 (1989) 4 4 5 -4 5 2 .
[15] M A Z U R , P., C O L E , K . W . , M A H O W A L D , A .P ., C ritical factors affecting the per­
m eabilization o f Drosophila em bryos b y alkanes, C ry o b io lo g y 2 9 (1 99 2) 3 9 -6 8 .
[16] C I C E R O , М ., C A L D W E L L , S., S T E P O N K U S , P .L ., Perm eabilization o f Drosophila
em bryos, C ry o b io lo g y (in press).
[17] G I L B E R T , L .I. , L ip id m etabolism and function in insects, A d v . Insect P h ysio l. 4
(1967) 7 0 -2 1 1 .
[18] S A N D E R , K ., G U T Z E I T , H .O ., J A C K L E , H ., “ Insect em bryogenesis: P h ysio lo gy,
genetical and m olecular aspects’ ’, C om p re h e n sive Insect P h y s io lo g y , Biochem istry and
Ph arm ac o lo gy ( K E R K U T , G .A ., G I L B E R T , L .I., E d s), V o l. 1, Perga m on Press,
L o n d o n (1985) 3 1 9 -3 8 5 .
[19] L E A , C .H ., H A W K E , J .C ., T h e influence o f water o n stability o f lipovitellin and the
effects o f freezing and d ryin g, J. Biochem . 5 2 (1952) 10 5-114 .
[20] L O V E L O C K , J.E., T h e dénaturation o f lipid protein com plexes as cause o f dam age by
. freezing, Proc. R. Soc. L o n d ., Ser. В B io l. Sci. 147 (1957) 4 2 7 -4 3 3 .
[21] M A Z U R , P., C O L E , K . W . , S C H R E U D E R S , P .D ., M A H O W A L D , A .P ., C o n trib u ­
tions o f coo lin g and w a rm in g rate and developm ental stage to the s u rviv a l o f Drosophila
• em b ryos cooled to — 2 0 5 ° C , C ry o b io lo g y (in press).
[22] L E E , R . E ., D E N L I N G E R , D . L . , C o ld tolerance in diapausing and nondiapausing
stages o f the flesh fly, Sarcophaga crassipalpis, Ph ysiol. Entom ol. 10 (1985) 3 0 9 -3 1 5 .
[23] K N I G H T , J.D ., B A L E , J.S., F R A N K S , F ., M A T H I A S , S .F ,, B A U S T , J.G ., Insect
cold hardiness: Su p ercoolin g points and pre-freeze mortality, C ryo-Lett. 7 (1986)
19 4 -2 0 3 . - ,
[24] S T R O N G - G U N D E R S O N , J. М . , L E O P O L D , R . A . , C ry o b io lo g y o f Musca domestica:
Su p ercoolin g capacity and low-tem perature tolerance, A n n . Entom ol. Soc. A m . 18
(1989) 7 5 6 -7 6 2 .
 IAEA-SM-327/27 77

[25] M E Y E R S , S .P ., L Y N C H , D . V . , K N I P P L E , D . C . , L E I B O , S.P ., S T E P O N K U S ,
P .L . , L o w temperature sensitivity o f Drosophila melanogaster em bryos, C ry o b io lo g y
2 5 (1988) 5 4 4 -5 4 5 .
[26] S O N N E N B L I C K , B .P ., “ T h e early em b ryo lo gy o f Drosophila melanogaster” , The
B io lo g y o f Drosophila ( D E M E R E C , M . , E d .), W ile y, N e w Y o r k (1 95 0) 6 2 -1 6 7 .
[27] C A L L A I N I , G ., M A R C H I N I , D ., A b n o rm a l centrosom es in cold-treated Drosophila
em bryos, E x p . C e ll Res. 18 4 (1989) 3 6 7 -3 7 4 .
[28] L E V I T T , J., “ R e sp onses o f plants to h ig h temperature stress” , C h illin g , F ree zing and
H ig h Tem perature Stress, V o l. 1, A ca d e m ic Press, N e w Y o r k (1 9 8 0 ) 2 3 -6 4 .
[29] B A L E , J.S., Insect cold hardiness — A n e cophysiological perspective, J. Insect P h y s ­
iol. 13 (1987) 121-129 .
[30] L E E , R . E . , C H E N , C .P ., D E N L I N G E R , D . L . , A rapid cold-hardening process in
insects, Science 2 3 8 (198 7) 14 15 -14 17 .
[31] Q U I N N , P J . , A lipid-phase separation m odel o f low-temperature dam age to b iological
m em branes, C ry o b io lo g y 2 2 (1985) 1 2 8-146 .
[32] M c G R A T H , J.J., “ C o ld shock: A thermoelastic stress in chilled biologica l m em ­
branes” , N e tw o rk T h erm odynam ics, H eat and M a s s T ran sfe r in B iotech no logy
( D I L L E R , K . R . , Ed .), A S M E B e d V o l. 5, H T D V o l. 9 0 (1987) 5 7 -6 6 .
[33] S T O R E Y , K . B . , S T O R E Y , J . M . , Freeze tolerance in anim als, Ph ysiol. R e v. 6 8 (1988)
. 2 7 -8 4 .
[34] F R A N K S , F ., B io p h y sic s and Biochem istry at L o w Tem peratures, C a m b rid ge U n iv e r­
sity Press, C a m b rid ge (1985).
[35] B A U S T , J.G ., M e c h a n ism s o f cryoprotection in freezing tolerant anim al systems,
C ry o b io lo g y 10 (1973) 1 9 7-203 .
[36] R U D O L P H , A . S . , C R O W E , J .H ., M e m b ra n e stabilization d u rin g freezing: T h e role
o f two natural cryoprotectants, trehalose and proline, C ry o b io lo g y 22 (1985) 3 6 7 -3 7 7 .
[37] H O N A D E L , T . E ., K I L L I A N , G .J., C ryop reserva tion o f m urine em bryos with treha­
lose and glycerol, C ry o b io lo g y 2 5 (1 98 8) 3 3 1 -3 3 7 .
[38] A N C H O R D O Q U Y , T.J., R U D O L P H , A . S . , C A R P E N T E R , J.F., C R O W E , J.H .,
M o d e s o f interaction o f cryoprotectants with m em brane p ho sph olipid s du rin g freezing,
C ry o b io lo g y 2 4 (1987) 3 2 4 -3 3 1 .
[39] A S A H I N A , E ., T A N N O , K ., A large am ount o f trehalose in a frost resistant insect,
Nature (L o n d o n ) 2 0 4 (1964) 1222.
[40] R O J A S , R . R . , C H A R L E T , L . D . , L E O P O L D , R . A . , Biochem istry and P h y sio lo g y o f
overw intering in the mature larva o f the red sunflow er weevil, Simicronyx fulvus
Leconte (Coleóptera: Curculionidae), J. Insect Ph ysiol. 3 7 (199 1) 4 8 9 -4 9 6 .
[41] M E Y E R S , S .P ., P IT T , R . E ., L Y N C H , D . V . , S T E P O N K U S , P .L ., Characterization
o f intracellular ice form ation in Drosophila melanogaster em bryos, C ry o b io lo g y 26
(1989) 4 7 2 -4 8 4 .
[42] M A Z U R , P., S C H N E I D E R , U ., M A H O W A L D , A .P ., Characteristics and kinetics o f
subzero ch illin g injury in Drosophila em bryos, C ry o b io lo g y 2 9 (1992) 3 9 -6 8 .
[43] S T E P O N K U S , P .L . , et al., C ryopreserva tion o f Drosophila melanogaster em bryos,
N ature (L o n d o n ) 3 4 5 (1990) 17 0-172 .
[44] S T E P O N K U S , P .L ., C A L D W E L L , S., C ryop reserva tion o f Drosophila melanogaster
em b ryos b y vitrification, C ry o b io lo g y (in press).
78 LEOPOLD

[45] D U M A N , J.G ., X U , L ., N E V E N , L .G :, T U R S M A N , D ., W U , D . W . , “ H e m o lym p h

proteins involved in insect subzero-temperature tolerance: Ice nucleators and antifreeze
proteins” , Insects at L o w Tem perature ( L E E , R . E ., D E N L I N G E R , D . L . , Ed s),
C h a p m a n and H a ll, N e w Y o r k (199 1) 9 4 -1 3 0 .
[46] C Z A J K A , M . C . , L E E , R ; E . , A rapid cold-hardening response protecting against cold
sh o c k injury in Drosophila melanogaster, J. E x p . B io l. 14 8 (1 99 0) 2 4 5 -2 5 4 .
[47] B U R T O N , V ., M I T C H E L L , H .K ., Y O U N G , P., P E T E R S E N , N . S ., Heat shock p ro ­
tection against cold stress o f Drosophila melanogaster, M o l. C ell. B io l. 8 (1988)
35 5 0 -3 5 5 2 .
[48] B A U S T , J.G ., Enviro nm e ntal triggers to cold-hardening, C o m p . B iochem . P h ysio l. 7 3
(1982) 5 6 3 -5 7 0 .
[49] H O R W A T H , K . L . , D U M A N , I.G ., Induction o f antifreeze protein production by
ju ve nile horm one in larvae o f the beetle Dendroides canadensis, J. C o m p . Ph ysiol. 151
• (1983) 2 3 3 -2 4 0 .
[50] C A N N O N , R.J., D ie t and acclim ation effects on the cold tolerance and su rviv a l o f an
Antarctic springtail ( Cryptopygus antarticus), B r. Antarctic Su rv. B u ll. 71 (1986)
7 3 -8 0 .
[51] O H L S S O N , L . , V E R H O E F , H . A . , Effects o f diet com position o n cold adaptation in
temperate C ollenbola, C o m p . Biochem . P h ysio l. 91 (1988) 4 7 5 -4 7 9 .
[52] S O M M E , L . , T h e effect o f glycerol injection on mortality and pupation in larvae o f
Ephestia kuehniella after exposures to lo w temperatures, Entom ol. E x p . A p p l. 11
(1 96 8) 14 3-148 .
[53] S A W A , М . , Fertilization o f hetero-specific insect eg gs b y sperm injection, Jpn J.
Genet. 6 6 (1991) 2 9 7 -3 0 3 .
 IAEA-SM-327/6

TECHNOLOGY FOR TRANSFORMING

THE GERM LINE IN ECONOMICALLY
IMPORTANT INSECTS
Current status andprospects
for advancements
A .M . HANDLER
Insect Attractants, Behavior, and Basic
Biology Research Laboratory,
Agricultural Research Service,
United States Department o f Agriculture,
Gainesville, Florida,
United States o f America

Abstract
TECHNOLO GY FOR T R A N S F O R M IN G THE GERM L IN E IN E C O N O M IC A L L Y
IM P O R T A N T IN S E C T S : CURRENT STATU S AND PRO SPECTS FO R
ADVAN CEM ENTS.
T h e developm ent o f a ge rm line transform ation m ethodology fo r insects o f agricultural
and m edical im portance is critical to the im plem entation o f n o ve l'a n d h ig h ly efficient m eans
o f insect m anagem ent u sin g m olecular b iolog ica l techniques. C urrently, an efficient m eans o f
gene transfer, u sin g the P transposable element based vector system, is p ossib le in on ly one
insect, Drosophila melanogaster, and this m ethodology has not been applicable to other
insects. P ro gre ss has been made in m o d ify in g the P element so that it m igh t function in non-
d ro sop h ilid s, although a system fully useful fo r a w ide range o f insects is still prospective.
O ther transposable element system s fro m Drosophila ha v in g the potential fo r gene vector
developm ent have been identified, specifically the hobo and mariner transposons. F o r both
these elements, h o m o lo g o u s genes or D N A sequences have been found in distantly related
insects, suggesting that their m ob ility properties m igh t be m aintained in these and other
insects, o r that they m ay be used as probes to isolate related transposons. Continued studies
along these lines ho ld the greatest prom ise, at present, fo r gene vector development. A n
immediate use o f gene transform ation in econom ically important insects m a y be for genetic
sexin g and m ale sterilization in the sterile insect technique. T h is is due to the understanding
o f sex specific gene expression and sex determination activity in D. melanogaster, as w ell as
the genetic material available fro m this and other organism s. M o d e l system s fo r h ig h ly effi­
cient genetic se xin g and sterilization based on recom binant D N A m anipulation and gene
transform ation are discussed.

79
80 HANDLER

1. INTRODUCTION

The manipulation o f genes and their subsequent transfer into insect host
genomes offers a vast potential for improving our ability to control insects o f agricul­
tural and medical importance. Gene transfer technology might directly impact insect
management programmes by facilitating: (1) genetic sexing and sterilization for
sterile male release programmes; (2) insecticide resistance in beneficial insects;
(3) cold hardiness or freeze tolerance to store insects for mass release or for main­
taining germplasm; and (4) introduction o f genetic markers into populations for
monitoring gene flow, insect migration and dispersal patterns, among others.
Perhaps more important than some o f these immediate applications, the basic
knowledge gained by gene transfer experiments will certainly enhance our under­
standing o f insect biology relevant to insect management and control, such as
resistance mechanisms, sex determination, hybrid sterility, and hormone action and
metabolism. In particular, this information will be essential to the development o f
biological control agents, enhancement o f natural enemies and developing resistance
mechanisms against insect pests in plants and animals.
Despite the benefits to be derived from gene transfer methods in insects, at
present the routine introduction o f exogenous DNA into a host insect’ s genome is
limited to the genus Drosophila using the P transposable element based gene vector.
Non-vector mediated gene transfer has been reported sporadically for many years,
although in most studies it was concluded that somatic inheritance was occurring,
and in no case was chromosomal integration verified (see Ref. [1]). Quite recently,
gene transformation was re p o rte d for the p re d a t o ry mite, Metaseiulus occidentalis,
at relatively high frequencies but, while stable transformation was concluded due to
the presence o f injected DNA after several generations, chromosomal integration
was not determined [2]. Other recent efforts to achieve P vector mediated gene
transfer in non-drosophilids, including mosquitoes, tephritids and locusts, have all
been unsuccessful. However, in the mosquito studies non-P mediated DNA
transposition was observed [3-5], suggesting that non-vector mediated transforma­
tion may be possible in this insect, albeit at low frequencies. A caveat to the reported
failure o f P mediàted transformation is that all o f these tests relied, at least initially,
on the use o f neomycin analog resistance as a selection method. For tephritids, at
least, this appears to have been unreliable, since drug resistant lines were selected
in which they did not contain the vector, the vector was not chromosomally
integrated, or resistance was lost in succeeding generations [6-8].
While the P vector function in non-drosophilids remains uncertain on the basis
o f transformation experiments, a lack o f P mobility in these insects is supported by
our tests using transient embryonic P mobility assays (see Ref. [9]). Although
progress has been made in defining the specific restrictions on P mobility and
ameliorating some o f them using these assays, development o f a highly efficient P
vector for non-drosophilids may be a long term effort. Given this prospect, it is
 IAEA-SM-327/6 81

reasonable that research be continued on the discovery and analysis o f other vector
systems, in addition to methods to efficiently introduce DNA into germ cells and
select for transformants. Potential candidates for vector development include
identified transposable elements, viruses and retroviruses and symbiotic organisms.
The relatively high number o f transposons within middle repetitive DNA has also
encouraged the search for new transposon systems for development into species
specific vector systems. O f these possibilities, the best prospects for rapid develop­
ment o f an efficient, reliable and stable gene vector system are known transposable
element systems which are in the inverted terminal repeat class, such as the
P element. O f particular interest are the hobo and mariner elements, which were also
discovered in Drosophila and which have subsequently been developed into gene
vectors for that insect. Unlike the P element, these transposons have an apparent
evolutionary relationship to other transposons, or other genes, in distantly related
insect and non-insect organisms. This raises the possibility that the hobo and mariner
elements may retain mobility properties in non-drosophilid insects, or that they may
be used as probes to isolate related transposons.

2. P ELEMENT VECTOR

The first vector helper system for stable germ line transformation in
Drosophila melanogaster was developed by Rubin and Spradling [10] using the
transposable element P system (see Ref. [11] for a review o f P). Within a few years
o f this development, several findings suggested that the P vector might similarly be
effective in non-drosophilids. First, Brennan et al. [12] found evidence for
P transposition in Drosophila species which are distantly related to D. melanogaster
and in which P elements had not been detected. Second, the basis for a major
restriction in P element mobility in D. melanogaster was discovered and
ameliorated. Laski et al. [13] found that P mobility was restricted to the germ line
due to a lack o f intron 3 processing in somatic cells, resulting in a truncated
non-functional transposase polypeptide. Creation o f a helper (pUChsxA2-3) having
intron 3 DNA deleted from the transposase gene resulted in somatic production o f
a full length functional product which supports P mobility. Third, a dominant drug
resistance selection system for transformants, using a vector (pUChsneo) with the
bacterial neomycin phosphotransferase gene, was developed, allowing relatively
easy selection o f transgenic insects [14]. This finding was especially important for
non-drosophilid insects in which the usual selectable markers are not available
(requiring a selectable mutant allele and cloned DNA for the wild type allele).
Using the pUChsneo vector and the pUChsirA2-3 helper, several laboratories
attempted to create non-drosophilid transformants. Most notably, three separate
82 HANDLER

groups working with different mosquito species isolated neomycin resistant

transformants, although upon molecular analysis none o f them appeared to be
P mediated [3-5]. Several laboratories working on the Mediterranean fruit fly,
Ceratitis capitata [6,7], and our laboratory working on another tephritid, the
Caribbean fruit fly, Anastrepha suspensa [8], obtained similar results which were
unusual. Putative neomycin resistant G1 lines were selected, and in some cases
vector DNA was observed in these insects [15]. However, chromosomal integrations
could not be verified, and in all cases neomycin resistance was lost in subsequent
generations. This suggests that neomycin resistance’ is not a reliable selection
method, at least for tephritids. Although the basis for this has not been unequivocally
determined, it may have relevance to the consideration o f other selection systems.
Anecdotal evidence already exists for inconsistencies in selection by neomycin
resistance in Drosophila which are related to the amount and methods o f neomycin
administration, the presence o f yeast, rearing conditions, etc. (see Ref. [16]). The
transient nature o f neomycin resistance in tephritids would suggest that
extrachromosomal expression and inheritance o f the vector may be occurring,
perhaps borne by bacterial endosymbionts. If this is the case, then a similar situation
may occur with other chemical selections, or any selection which is not defined by
visible expression in insect cells (assuming that extrachromosomal expression would
usually be mosaic).

Realizing that the inability to isolate transformants could be due to restrictions

other than a non-functional vector, and considering vector functionality to be the
primary necessity for efficient gene transfer, we developed methods to test vector
functional directly, independent o f chromosomal integration or dominant selection.
This was achieved by modification o f an assay for P mobility in Drosophila cell lines
[17]. The assay consisted o f an indicator plasmid, pISP, having P element sequences
within a reporter gene, and a helper plasmid, pUChs7rA2-3, encoding transposase,
being injected into insect embryos [18]. The functionality o f the transposase gene in
the host embryo was determined by P element excision from the indicator plasmid.
After incubation and harvesting o f the plasmids from the embryos, and subsequent
bacterial transformation, P element excision could be assayed as a function o f
restoration o f the reporter gene expression. In our initial assays, the reporter gene
was j8-galactosidase expression, which is monitored by blue bacterial colonies grown
on the X-gal substrate. This assay only reported P excisions which restored the
/З-galactosidase reading frame, which represents approximately one third o f all exci­
sions. A subsequent new indicator plasmid, containing a gene which confers
sensitivity to streptomycin inserted within the P element sequence, was able to report
all excisions as a function o f streptomycin resistance [19]. Expression o f
/З-galactosidase further distinguished precise from imprecise excisions.

Studies with these in vivo transient expression excision assays in

D. melanogaster indicated an approximate total excision frequency o f
 IAEA-SM-327/6 83

5.5 x 10"3/pISP, with a precise excision frequency o f about 1.7 x 10"3/pISP [19].
In other drosophilids, the excision frequency decreased as a general function o f
relatedness to D. melanogaster, with both total and precise excisions occurring at
about a 10% level in distantly related drosophilids such as Chymomyza procnemis
and D. virilis [9, 18, 19]. In other dipterans, excision was not detected. These
results indicated that P mobility is restricted in non-drosophilid insects, and that the
P gene vector would most likely be non-functional as well in these insects.

Since a restriction in P transposition in somatic tissue is due to defective intron

3 splicing, we tested the possibility that splicing o f the other introns might be
defective in non-drosophilids [9]. Using a transposase gene lacking introns 2 and
3 (pUChs7rA 1-2-3), and a transposase cDNA (pKhsircDNA) lacking all 3 introns,
P mobility was tested in D. melanogaster and A. suspensa. Both intron deleted genes
supported P excision in Drosophila. In Anastrepha, pUChs7rAl-2-3 was not
functional; however, a relatively low level o f P mobility was supported by the cDNA
representing the first observation o f a significant level o f P mobility, and thus
transposase function, in a non-drosophilid. While this result suggested that transcript
processing is a restriction on the transposase function in Anastrepha, the twenty fold
lower level o f excision relative to Drosophila indicated that other limiting factors
also exist. The most straightforward possibilities are that either requisite positive
acting factors are lacking or inhibiting factors exist in non-drosophilids. To test the
first possibility, nuclear extracts from D. melanogaster embryos were co-injected
with the plasmids (pISP and pKhsircDNA) in the mobility assay [9]. Similar nuclear
extract preparations were found to contain a 66 kilodalton polypeptide which binds
specifically to the P element termini [20]. The addition o f the nuclear extract further
increased excision by two fold in Anastrepha, although, interestingly, excision was
decreased in Drosophila by more than one third. This decreased the difference in P
excision between the species from twenty fold using only the transposase cDNA to
an eight fold difference with the addition o f nuclear extract. Importantly, the nega­
tive influence o f the extract in Drosophila raises the possibility that the action o f
specific positive acting factors was not fully revealed in Anastrepha owing to a
general inhibitory effect o f the extract. This should be clarified by testing more
highly purified nuclear extract fractions, as well as the 66 kilodalton polypeptide, to
identify specific factors. Although this objective , is straightforward and has a
reasonable probability for success, the protein purification and the number o f
mobility assays required make this a formidable task nevertheless.

P, or P like, elements are widely distributed in Drosophila [21, 22] and their
presence in dipterans outside o f the Drosophila genus has been reported [23]. While
our results suggest that P would not be an effective vector in these insects, these
P like elements may have vector capability in the insects in which they are found.
However, if they retain the functional characteristics o f P, their mobility properties
also might be restricted to a few closely related species.
84 HANDLER

3. Hobo

The hobo transposable element was discovered in D. melanogaster by its

association with a number o f mutant alleles o f cloned genes, which made its isolation
and initial molecular analysis straightforward (see Ref. [24] for a review o f hobo).
The eventual identification o f a complete hobo element revealed basic structural
similarities with the P element in that it is a 3.0 kb element with short terminal
inverted repeat sequences. A 2 kb open reading frame exists that is capable o f
encoding a 73 kilodalton polypeptide. Beyond these general structural similarities
with P, there is no apparent DNA or amino acid sequence homology with P, nor do
the elements act to mobilize one another. Like P, hobo has been developed into a
vector helper transformation system in D. melanogaster yielding similar trans­
formation frequencies [25].
The molecular analysis o f a functional hobo element revealed domains o f
homology in the amino acid sequence between hobo and the plant transposons
Activator, from maize, and ТатЗ, from Antirrhinum [26]. This suggested that these
transposons may have a common evolutionary history and, if so, related transposons
might exist in a variety o f insect species, and perhaps their mobility properties are
less phylogenetically restricted compared to P. Preliminary investigations give
support to both these possibilities. Atkinson et al. [27] used hobo mobility assays,
similar to those described for P, in D. melanogaster and Musca domestica. Interest­
ingly, hobo mobility was detected in Musca without a helper plasmid, but not in a
D. melanogaster strain known to be devoid o f hobo elements. This suggests that the
Musca s tra in tested m a y h a v e hobo lik e e le m e n ts c a p a b le o f m o b il iz in g hobo. I n a n
effort to identify hobo or hobo like elements, we screened a wide variety o f insects
using polymerase chain reaction (PCR) gene amplification techniques. First,
degenerate PCR primers were made to DNA sequences within the hobo transcrip­
tional unit, which encode four to five amino acid sequences identical to that found
in Activator. These primers amplified the intervening DNA sequence in both hobo
and Activator. PCR amplified DNA sequences were also generated from genomic
DNA ifrom various Drosophila species, Anastrepha suspensa and striata, Heliothis
zea and virescens, Musca domestica and Stomoxys calcitrans. O f the insects tested,
it is notable that PCR fragments were not generated from Ceratitis capitata or
Anopheles quadrimaculatus. Given our primary interest in tephritid insects, the
single 450 bp fragment generated in A. suspensa was subcloned, sequenced and
compared with the corresponding sequence in hobo. After sequence alignment, a
55% homology was determined between the sequences, which was the highest level
o f homology among all the transposable element sequences in the GenBank.
Translation o f the aligned sequences yielded an amino acid sequence homology o f
approximately 30%.
Transposable elements often exist as defective (non-mobile) deletion
derivatives o f complete elements. Thus, it is possible that transposon related
 IAEA-SM-327/6 85

sequences exist but are missed owing to deletion o f one or both o f the primer sites.
For both P and hobo, most o f the deletion derivatives are internal, leaving the
inverted terminal repeat sequences intact or with minor sequence variation. Since
some homology exists between the terminal sequences o f hobo and Activator (as well
as other transposable elements) [28], a PCR screen was also conducted using the
terminal inverted repeat sequence as a primer. Unfortunately, the PCR reàction is
usually limited to efficient amplification o f sequences smaller than 2 kb; thus we
would not expect to generate complete sequences in the range o f 3 kb. However,
smaller internal deletion derivatives can be detected. As expected, numerous
fragments were generated in D. melanogaster and closely related species, as well as
drosophilids in which hobo had not yet been detected. Analysis o f a wide range o f
dipterans, lepidopterans, hymenopterans and coleopterans also revealed several
fragments in almost all the species. Notably, one o f the only species which did not
reveal hobo like fragments was A. quadrimaculatus, consistent with the internal
primer result. However, several polymorphic fragments were detected in several
strains o f C. capitata.
The relatedness o f these amplified sequences to hobo was indicated by DNA
hybridization using Ûie hobo Xho I fragment (a 2.6 kb internal sequence not
including the termini) as a probe. A very strong signal was obtained from
D. melanogaster strains and Drosophila species, such as simulons, known to have
elements. Previous evidence from Daniels et al. [29] indicated a very limited
distribution o f hobo among drosophilids based on low stringency hybridization. Thus
it was interesting to detect signals from D. willistoni and D. saltans, which were
probably missed in the Daniels et al. study owing to sequence divergence or low copy
number. Given the 55% homology between hobo and the hobo like internal fragment
in A. suspensa, we were not optimistic that an unambiguous signal would occur in
non-drosophilids. Among non-drosophilids, a signed was detected only from
Anticarsia gemmatalis, the velvetbean caterpillar, suggesting a relatively high level
o f homology. These fragments are currently being sequenced to better define this
possibility.
The apparent mobility o f hobo in non-drosophilids, along with the discovery
o f DNA elements having homology with hobo, suggest that hobo based gene vectors
may be functional in these insects, or that new transposable elements may be isolated
on the basis o f their relatedness to hobo. The isolation o f a functional hobo like
element, if necessary, will be a challenge, and may be pursued by testing for
transcriptional activity or transient mobilization o f defective elements. Despite these
difficulties, these lines o f investigation hold great promise for development o f gene
vectors for non-drosophilid insects.
86 HANDLER

4. Mariner

Quite recently, another transposable element from Drosophila, mariner, has

been developed into a gene vector. Mariner was discovered in D. mauritiana,
associated with the H'/iziepeach mutation [30]. Like P and hobo, mariner is a terminal
inverted repeat transposon, but it is different in some significant ways (see Ref. [31]
for a review o f mariner). First, it is smaller than either P or hobo, being about 1.3 kb
in length with an uninterrupted open reading frame o f about 1 kb, and having
inverted terminal repeats o f 28 bp. Second, in contrast to both P and hobo, mariner
exhibits a significant level o f somatic mobility. This is evidenced in the white**30*1
strain where, somatic excision o f the element from the wild type gene results in a
mosaic phenotype o f patches o f wild type red eye pigmentation in a mutant peach
colour background.
Like hobo, but unlike P thus far, mariner like elements have been discovered
recently in non-dipteran insects. Somewhat serendipitously, Lidholm et al. [32]
discovered a mariner like element (MLE) inserted within an intron o f an antibacterial
protein gene, cecropin, in Hyalophora cecropia. The MLE is 1255 bp in length with
38 bp terminal inverted repeats. The internal sequence has 48% homology with
mariner, with the terminal repeats having 68% homology. Interestingly, it was
estimated that the Cecropia species has approximately 1000 copies o f the MLE,
while mariner has, about 20-30 copies in D. mauritiana. Including conservative
replacements, the MLE reading frame has a 56% similarity with mariner. Stretches
o f an identical amino acid sequence for six and seven residues indicates that gene
amplification techniques can be used to reveal additional MLEs in a variety, o f
insects, as we have done with hobo. Clearly, this should be a high priority.
MLE transcripts have not been detected in Cecropia, and the reading frames
sequenced apparently have mutations capable o f limiting full length or functional
translation products [32]. However, if.the inverted repeats represent true terminal
sequences, then it might be relatively simple to attempt the mobilization o f the MLEs
from either chromosomes or plasmid constructs, by transient expression o f mariner.
Similarly, the mobilization o f mariner itself should be tested by excision or
transposition assays. If neither o f these approaches is successful, then an MLE vector
would require the isolation o f an autonomous functional element, or an element
encoding functional transposase. As with hobo like elements, identification o f an
autonomous element may not be trivial. However, the somatic expression o f mariner
presents the possibility that the. initial identification o f transcriptionally active
elements may be considerably easier than with the.germ line restricted P and hobo.
Since transposase from other systems is normally produced at very low levels, more
extensive and sensitive tests are required to rule out MLE transcripts in the Cecropia
strain from which it was discovered. In addition, screening by PCR and reverse
transcriptase-PCR may reveal transcripts in other Cecropia strains, and in other
insects in which MLEs are discovered. Similar to hobo, the development o f a gene
 IAEA-SM-327/6 87

vector from mariner or mariner like elements may be difficult, but it also has one
o f the best potentials for success.

5. APPLICATION OF GENE TRANSFER TO GENETIC

SEXING TECHNIQUES

O f the many potential uses o f gene transfer technology for the management o f
economically and medically important insects, an important application with
potential for immediate use is for genetic sexing and male sterilization utilized in the
sterile male release technique (see Ref. [33]). The knowledge and techniques
available from recombinant DNA, and transformation studies in Drosophila, suggest
a variety o f means to genetically sex and sterilize male insects. The rapid
implementation o f these techniques may be easier relative to other applications
requiring the release o f transgenic insects into the environment, since release would
be limited to sterile males.
The basis for genetic sexing using molecular techniques would, most simply,
follow methods already being developed with classical genetic techniques. These
include the sex specific expression o f a selectable gene product. A useful system for
fruit flies is alcohol dehydrogenase (adh) gene activity [34], since it can be used with
both positive and negative selections [35]. A lack o f activity, as in the null mutant
adh, is selected against with ethanol, while normal activity from the wild type gene
is selected against with secondary alcohols such as pentenol. Thus, female specific
expression o f adh + , when driven by a promoter such as the female specific yolk
protein (YP) gene promoter [36], could be selected against by pentenol. However,
while female specific YP promoters exist in almost all insects, they are only active,
in adults or late pupae, thereby limiting male selection after rearing costs have been
incurred. A more highly efficient selection system which acts early in development
and without toxic chemicals is based on temperature sensitive lethal (tsl) gene
activity [37]. Organisms having tsl mutations die at elevated temperature owing to
gene products which are necessary for survival becoming non-functional. The male
specific expression o f a wild type allele o f a tsl gene, in an otherwise tsl mutant
background, would allow males to survive, while females would die.at an elevated
temperature. Most simply, this system would require a male specific promoter which
acts in most tissues or those critical to survival. Unfortunately, the identified male
specific promoters in D. melanogaster are all limited to tissues necessary for
reproduction, but which are not critical to survival [38]. Nevertheless, such male
specific promoters could find use in causing male sterility by linking them to toxin
genes. An alternative to the recessive tsl genes are mutations in genes such as a
dominant temperature sensitive lethal (DTS) gene, which has recently been
cloned [39]. This mutation acts as a neomorph or antimorph to cause lethality even
88 HANDLER

in the presence o f wild type alleles. If expressed female specifically, the mutant gene
product would kill only females at elevated temperatures.
Although the recombinant DNA molecules for the systems described are
available or can be created, the most effective genetic sexing and sterilization strains,
similarly based upon genetic information from D. melanogaster, are much more
speculative. For example, several sex determination genes act female specifically
throughout development owing to female specific intron splicing [40]. In females,
the spliced intron removes a translational stop signal, allowing production o f a
functional product. In males, use o f an upstream 3' splice site allows the stop codon
to remain in the transcript, resulting in a truncated non-functional product. If the
female specific splice site were inserted in front o f (5') to a gene which encodes a
selectable or toxic product, in conjunction with a conditional promoter, expression
would be limited to females, allowing their selection early in development.
Interestingly, the most efficient model system for genetic sexing and male
sterilization, also from D. melanogaster, does not directly require recombinant DNA
or gene transfer techniques. This model is based upon the transformer-2 temperature
sensitive (rra-2ts) allele [41]. The tra-2 gene is required in chromosomal females
for female differentiation. In the absence o f tra-2 (as in a tra-2 homozygous mutant),
chromosomal females develop as fully differentiated males, but they have
rudimentary testes and are sterile. Although mutant X Y males are not somatically
affected, they are also sterile. Therefore, for the temperature sensitive allele,
X X females are fertile (although weakly) at the permissive 16°C temperature, but
develop as morphological sterile males at 29°C. Mutant X Y males are morpho­
logically normal at both temperatures, but sterile at 29°C. Thus, an idealized
extrapolation o f this homozygous mutant strain would yield breeding parentals at
16°C, while genetically identical offspring reared at 29°C would all develop as
sterile males. In this situation, there would be no treatment necessary other than
elevated temperature for sexing and sterilization. Indeed, genetic sexing would not
be required, since all the zygotes would be usable for release, thereby reducing by
half the number o f parental insects needed for breeding. Although, as noted,
molecular genetic manipulations would not necessarily be required to develop an
analogous system, molecular analysis using gene transformation would clearly be
essential to discover sex determination genes in non-drosophilid insects and to create
conditional mutations by site directed mutagenesis. This system well illustrates that
while gene transfer techniques may have some immediate uses for insect manage­
ment, their greatest potential will only be realized as a result o f their use in more
basic genetic and biological studies.

ACKN O W LED GEM EN T

Appreciation is extended to the United States Department o f Agriculture-

National Research Initiative for support. N
 IAEA-SM-327/6 89

REFERENCES

[1] H A N D L E R , A . M . , O ’B R O C H T A , D . A . , Prospects for gene transform ation in insects,

A n n u . R e v. Entom ol. 3 6 (1991) 1 5 9-183 .
[2] P R E S N A I L , J .K ., H O Y , M . A . , Stable genetic transform ation o f a beneficial arthro­
pod, Metaseiulus occidentalis (A ca ri: Phytoseiidae), b y a m icroinjection technique,
Proc. Natl. Acad. Sci. U S A 8 9 (1992) 7 7 3 2 -7 7 3 6 .
[3] M I L L E R , L . H . , et al., Stable integration and expression o f a bacterial gene in the
m osquito Anopheles gambiae, Science 2 3 7 (1 98 7) 7 7 9 -7 8 1 .
[4] M c G R A N E , V . , C A R L S O N , J.O ., M I L L E R , B . R . , B E A T Y , B.J., M icroin je ction o f
D N A into Aedes triseriatus o va and detection o f integration, A m . J. Trop. M e d . H y g .
3 9 (1987) 5 0 2 -5 1 0 .
[5] M O R R I S , A . C . , E G G L E S T O N , P., C R A M P T O N , J .M ., Genetic transform ation o f
the m osquito Aedes aegypti by m icro-injection o f D N A , M e d . Vet. Entom ol. 3 (1989)
1 -7
[6] M c I N N I S , D .O ., et al., “ T h e M editerranean fruit fly: Pro gre ss in developing a genetic
sexing strain u sin g genetic engineering m e th od ology” , M o d e rn Insect C ontrol: N u cle ar
Techniques and Biote ch n o logy (Proc. Sym p . Vie nn a, 1987), I A E A , V ie n n a (1988)
2 5 1 -2 5 6 .
[7] R O B I N S O N , A .S . , S A V A K I S , C ., L O U I S , C ., “ Status o f m olecular genetic studies
in the m edfly, Ceratitis capitata, in relation to genetic se x in g ” , ibid., pp. 2 4 1 -2 5 0 .
[8] H A N D L E R , A . M . , G O M E Z , S .P ., unpublished results, 1992.
[9] H A N D L E R , A . M . , G O M E Z , S.P ., O ’B R O C H T A , D . A . , A functional analysis o f the
P-elem ent gene-transfer vector in insects, A rc h . Insect Biochem . Ph ysiol, (in press).
[10] R U B I N , G .M . , S P R A D L I N G , A . C . , Genetic transform ation o f Drosophila with trans­
posable element vectors, Science 21 8 (1982) 3 4 8 -3 5 3 .
[11] E N G E L S , W . R . , “ P e lem e nts'in Drosophila melanogaster” , M o b ile D N A ( B E R G ,
D .E . , H O W E , M . M . , E d s), A m e ric a n Society fo r M ic ro b io lo g y , W ashington , DC
(1989) 437-484.
[12] B R E N N A N , M . D . , R O W A N , R .G ., D I C K I N S O N , W .J., Introduction o f a functional
P element into the germ line o f Drosophila hawaiiensis, C e ll 3 8 (1984) 14 7-151 .
[13] L A S K I , F . A . , R IO , D . C . , R U B I N , G . M . , T issu e specificity o f Drosophila P element
transposition is regulated at the level o f m R N A splicing, C e ll 4 4 (1986) 7 —19.
[14] S T E L L E R , H ., P I R O T T A , V . , A transposable P vector that confers selectable G 4 1 8
resistance to Drosophila larvae, E M B O J. 4 (1985) 167-171.
[15] M c I N N I S , D .O ., H A Y M E R , D . S ., T A M , S . Y . T . , T H A N A P H U M , S., Ceratitis
capitata (Diptera: Tephritidae): Transient expression o f a heterologous gene for
resistance to the antibiotic geneticin, A n n . Entom ol. Soc. A m . 8 3 (1990) 9 8 2 -9 8 6 .
[16] ASH BU RN ER, М . , Drosophila, A Labo ra tory H an dbook, C o ld S p rin g H a rb o r
Lab o ra tory Press, C o ld S p rin g H arbor, N Y (1989).
[17] R IO , D . C . , L A S K I , F . A . , R U B I N , G . M . , Identification and im m unochem ical analysis
o f b io lo g ica lly active Drosophila P element transposase, C e ll 4 4 (1986) 2 1 -3 2 .
[18] O ’B R O C H T A , D . A . , H A N D L E R , A . M . , M o b ility o f P elements in dro soph ilids and
nondrosophilids, Proc. Natl. A cad. Sci. U S A 8 5 (1988) 6 0 5 2 -6 0 5 6 .
90 HANDLER

[19] O ’B R O C H T A , D . A . , G O M E Z , S .P ., H A N D L E R , A . M . , P element excision in

Drosophila melanogaster and related drosophilid s, M o l. Gen. Genet. 2 2 5 (1991)
3 8 7 -3 9 4 .
[20] R IO , D . C . , R U B I N , G . M . , Identification and purification o f a Drosophila protein that
b ind s to the term inal 3 1 -base p a ir inverted repeats o f the P transposable element, Proc.
Natl. A cad . Sci. U S A 85 (198 8) 8 9 2 9 -8 9 3 3 .
[21] A N X O L E B E H E R E , D ., N O U A U D , D ., P E R I Q U E T , G ., Séquences ho m ologue s à
l ’élément P chez des espèces de Drosophila du groupe obscura et chez Scaptomyza pal­
lida (D roso philida e), Genet. Sel. E v o l. 17 (198 5) 5 7 9 -5 8 4 .
[22] L A N S M A N , R . A . , S T A C E Y , S .N ., G R I G L I A T T I , T . A . , B R O C K , H .W . , Sequences
ho m o lo g o u s to the P m obile element o f Drosophila melanogaster are w id ely distributed
in the subgenus Sophophora, Nature (L o n d o n ) 31 8 (1985) 5 6 1 -5 6 3 .
[23] A N X O L E B E H E R E , D ., P E R I Q U E T , G ., P -h o m o lo g o u s sequences in D ípte ra are not
restricted to the D ro so p h ilid a e fam ily, Genet. Iber. 3 9 (1987) 2 1 1 - 2 2 2 ..
[24] BLACKM AN, R .K ., G ELBART, W .M ., “ The transposable element hobo o f
Drosophila melanogaster", M o b ile D N A ( B E R G , E .E . , H O W E , M . M . , E d s), A m e r i­
can Society for M ic ro b io lo g y , W a shington , D C (1989) 5 2 3 -5 2 5 .
[25] B L A C K M A N , R . K . , G E L B A R T , W . M . , Identification o f a fully-functional hobo
transposable element and its use for germ -line transform ation o f Drosophila, E M B O
J. 8 (1989) 2 1 1 -2 1 7 .
[26] C A L V I , В ., Evid e n ce fo r a com m on e volutionary o rig in o f inverted repeat transposons
in Drosophila and plants: Hobo, Activator, and ТатЗ, C e ll 6 6 (1991) 4 6 5 -4 7 1 '
[27] A T K IN S O N , P .W ., CO ATES, C.J., P IN K E R T O N , A .C ., M ENDE, H .A .,
H O W E L L S , A .J ., O ’B R O C H T A , D . A . , Paper IA E A - S M - 3 2 7 / 7 , these Proceedings.
[28] S T R E C K , R . D . , M a c G A F F E Y , J .E ., B E C K E N D O R F , S . K . , T h e structure o f hobo
transposable elements and their insertion sites, E M B O J. 5 (198 6) 3 6 1 5 -3 6 2 3 .
[29] D A N I E L S , S .B ., C H O V N I C K , A ., B O U S S Y , I. A . , D istribu tion o f hobo transposable
elements in the genus Drosophila, M o l. B io l. E v o l. 7 (1990) 5 8 9 -6 0 6 .
[30] H A Y M E R , D .S ., M A R S H , J .L ., G e rm line and som atic instability o f a white mutation
in Drosophila mauritiana due to a transposable element, D e v . Genet. 6 (1986)
2 8 1 -2 9 1 .
[31] H A R T L , D . L . , ‘ ‘T ran spo sable element mariner in Drosophila species” , M o b ile D N A ,
( B E R G , D . E . , H O W E , M . M . , Ed s), A m e ric a n Society for M ic ro b io lo g y , W ashington ,
D C (1989) 5 3 1 -5 3 6 .
[32] L I D H O L M , D . - A . , G U D M U N D S S O N , G .H ., B O M A N , H .G ., A h ig h ly repetitive,
mariner-like element in the genom e o f Hyalophora cecropia, J. B io l. C hem . 2 6 6 (1991)
11 51 8-1 15 21 .
[33] HANDLER A .M ., “ M o le c u la r genetic m echanism s for sex-specific selection” ,
. A d va n c e s in Insect R e a rin g fo r Research and Pest M anagem ent ( A N D E R S O N , T .E .,
L E P P L A , N . C . , E d s), W e stvie w Press, Boulder, C O (1992) 11-32.
[34] R O B I N S O N , A .S . , V A N H E E M E R T , C ., G enetic sexing in Drosophila melanogaster
u sin g the alcohol dehydrogenase locus and a Y -lin k e d translocation, Theor. A p p l.
Genet. 5 9 (1980) 2 3 -2 4 .
[35] S O F E R , W ., H A T K O F F , М ., C he m ica l s e le c t i o n ^ alcohol dehydrogenase negative
mutants in Drosophila, Genetics 72 (1972) 545^549.
 IAEA-SM-327/6 91

[36] S H I R R A S , A ., B O W N E S , М ., Separate D N A sequences are required fo r norm al

female and ecdysone-induced m ale e xpression o f Drosophila melanogaster y o lk protein
1, M o l. Gen. Genet. 210 (198 7) 1 5 3-155 .
[37] B U S C H - P E T E R S E N , E ., “ D evelopm ent o f induced sex-separation m echanism s in
Ceratitis capitata (W ied.): E M S tolerance and suppression o f fem ale recom bination” ,
F ru it F lie s (Proc. 2n d Int. Sym p . C o lym b a ri, Crete, 1986) ( E C O N O M O P O U L O S ,
A .P ., Ed .), Else vie r, A m ste rd am (1987) 2 0 9 -2 1 6 .
[38] W O L F N E R , M . F . , Se x-specific gene expression in som atic tissues o f Drosophila
melanogaster, T ren d s Genet. 4 (1 98 8) 3 3 3 -3 3 7 .
[39] S A V I L L E , K .J ., B E L O T E , J .M ., A n essential gene, w ith a dom inant temperature-
sensitive lethal allele, en codin g a Drosophila proteasom e subunit, Science (submitted
for publication).
[40] B A K E R , B .S ., S e x in flies: T h e splice o f life, Nature (L o n d o n ) 340 (198 9) 5 2 1 -5 2 4 .
[41] B E L O T E , J . M . , B A K E R , B . S . , S e x determ ination in Drosophila melanogaster: A n a ly ­
sis o f transformer-2, a sex-determ ining locus, Proc. Natl. A cad . Sc i. U S A 79 (1982)
1568.
 IAEA-SM-327/7

PROSPECTS FOR NON-DROSOPHILID

GERM LINE TRANSFORMATION

P .W . ATK IN SO N , C.J. CO A TE S,
A .C . PINKERTON, H A . M EN DE
D ivision o f Entom ology,
Commonwealth Scientific and Industrial
Research Organization

A .J. HO W ELLS
Division o f Biochemistry and M olecular B iology,
Australian National University

Canberra A C T ,
Australia

D .A . O ’B R O C H TA
Center for Agricultural Biotechnology
and Department o f Entom ology,
University o f Maryland,
College Park, Maryland,
United States o f Am erica

Abstract
P R O S P E C T S F O R N O N - D R O S O P H IL ID G E R M L IN E T R A N S F O R M A T IO N .
T h e hobo transposable element o f Drosophila melanogaster has been used to achieve
stable ge rm line transform ation o f this insect. It is exam ined whether this element, o r related
transposable elements, can be used as vectors fo r achieving the ge rm line transform ation o f
tw o non-d rosop hilid s, the A u stralia n sheep blow fly, Lucilia cuprina, and the housefly, Musca
domestica. T h ro u g h the use o f e xcision a ssays designed to m easure the m obility o f hobo in
preblastoderm em bryos, it is dem onstrated that this transposable element is capable o f m o v e ­
ment in these species and that this m obility is independent o f the supplied hobo transposase,
suggesting the presence o f host encoded factors w h ich are capable o f interacting w ith hobo
D N A . T h e presence o f hobo like elements in the genom es o f each o f these insects is supported
b y the clonin g o f hobo like sequences from both these species. A n a ly s is o f these sequences
is discussed, as is the possible role o f hobo and hobo like elements in achieving non-
d ro sop h ilid ge rm line transformation.

93
94 ATKINSON et al.

1. IN TROD U CTION

The hobo transposable element o f Drosophila melanogaster is approximately

3 kb in length and is a member o f a class o f transposable elements that contains short,
inverted repeat sequences at their termini. In some respects, hobo elements resemble
the P transposable elements o f D . melanogaster. They are o f similar size, they trans­
pose through a ‘ D N A on ly’ mechanism, their presence or absence can result in a
phenotype characterized as hybrid dysgenesis, both have been proposed to have
invaded D. melanogaster wild populations within the past 50 years, and they create
duplications o f flanking sequences upon insertion into the genome [1, 2]. Further­
m ore, as for P, hobo elements have been used to genetically transform
D . melanogaster [3]. H ow ever, there are a . number o f , significant differences
between these two Drosophila transposable elements. There is no sequence similarity
between hobo and P and the internal organization o f each element is also quite differ­
ent [4, 5]. The P element transposase is encoded by four exons and differential splic­
ing o f the last intron is the basis o f the tissue specific expression o f P transposase.
W hile expression o f the putative hobo transposase is not yet fully understood, the
presence o f one long open reading frame encoding hobo transposase would seem to
preclude differential splicing playing a role in hobo transposase expression.
The ability o f hobo elements to act as vectors o f Drosophila transformation
suggests that they may be also capable o f achieving non-drosophilid insect transfor­
mation. T o explore this further, we developed an excision assay which allows the
detection o f hobo element excision in insect embryos [6]. Using these assays, we
demonstrated that hobo can be mobilized in the embryos o f D. melanogaster, the
Australian sheep blow fly, Lucilia cuprina, and the housefly, Musca domestica .
M oreover, in both Lucilia and Musca, excision appears to be independent o f hobo
transposase, suggesting that endogenous genes are involved in mobility. In
support o f this, we have cloned sequences from both o f these non-drosophilids which
display strong similarity to the hobo element o f D . melanogàster.

2. M A TE R IA L S A N D M ETH ODS

2 .1 . Strains

The strain o f L. cuprina used for both microinjections and cloning was the
standard wild type (SW T) strain maintained at the Division o f Entom ology,
Commonwealth Scientific and Industrial Research Organization (C SIRO ), Canberra.
Three different populations o f M . domestica were used. One had been maintained
at the United States Department o f Agriculture (U SD A ) in Beltsville, Maryland,
another at the U SD A in Gainesville, Florida, and the third at the CSIRO Division
o f Entom ology, Canberra.
 IAEA-SM-327/7 95

2.2. Plasmid constructions and hobo excision assays

The hobo excision assay, together with the construction o f the plasmids used
and the techniques used for the sequencing o f excision events, have been described
previously [6].

2.3. Cloning of hobo like sequences from L. cuprina

A size fractionated genom ic library cloned into the EM BL3 derived lambda
vector GEM 11 (Promega) was transferred to Hybond N + membranes (Amersham)
and screened with radiolabelled D N A containing the hobo 108 element [4] under
standard conditions o f low stringency (5X SSC, 5 4 °C , 0% formamide) [7]. Posi­
tively hybridizing plaques were rescreened with the same D N A and subsequent
hybridizing plaques were purified for further analysis. One o f these was found,
through conventional techniques o f cloning and D N A sequencing [8], to contain
sequences similar to the hobo element. D N A sequence analysis was perform ed using
the G C G database program o f the University o f W isconsin, M adison, U SA.

3. RESULTS

3.1. H obo excision assays in L . cuprina and M . domestica

The hobo excision assay is similar in concept to the P element excision assay
described previously [9]. An indicator plasmid containing a defective hobo element
was co-injected into pre-blastoderm embryos together with a helper plasmid which
encodes hobo transposase. In some experiments, the indicator plasmid is injected
alone. When injected alone into embryos o f a strain o f D . melanogaster that lacks
autonomous elements, no excision events are recovered [6]. Co-injection o f the hobo
transposase encoding plasmid results in the recovery o f excision events at a fre­
quency o f at least 0.04% [6]. Sequence analysis o f these excision events revealed
that, in the majority o f cases, all o f the hobo element had been rem oved, together
with a small amount o f flanking D N A sequence. Significantly, additional D N A was
present at the empty excision site. This additional D N A was not from the hobo ele­
ment but was an inverted duplication o f flanking genom ic D N A . In some excision
events, this additional D N A was duplicated several times. Injection o f a plasmid
lacking hobo sequences did not result in the recovery o f excision events, indicating
that excision was dependent on the presence o f hobo sequences in the indicator
plasmid.
The pattern o f hobo excision observed in D. melanogaster was found to be
distinctly different to the pattern o f P element excision events recovered from this
species under identical experimental conditions [10]. Specifically, follow ing P exci-
 о
о
TA BLE I. hobo EXCISION IN L. cuprina (SWT) AND M. domestica

—hobo
W ith hobo excision indicator plasm id
sequences

L. cuprina ( S W T ) M. domestica E. coli M. domestica

+ hsplO hobo transposase — + — + — —

helper plasm id

N o . o f independent 9 5 9 7 5 6

ATKINSON
experiments

N o. o f p lasm ids screened 75 713 44 157 42 142 4 7 437 1 x 105 55 722

N o . o f tetRlacZ~ colonies 40 25 95 14 7 16

et a l.
recovered (0 .0 5 3 % ) (0 .0 5 7 % ) (0 . 2 2 5 % ) (0 . 0 2 9 % ) (0 .0 0 7 % ) (0 .0 2 9 % )

N o. with deletions 19 11 55 11 0 2
(0 .0 2 5 % ) (0 .0 2 5 % ) (0 . 1 3 0 % ) (0 .0 2 3 % ) (0 .0 0 3 % )

N o . with deletions in v o lv in g 4 2 20 11 0
hobo sequences3 (0 .0 0 5 % ) (0 .0 0 4 % ) (0 . 0 4 7 % ) (0 . 0 2 3 % )

a W e have not as yet exam ined the break points o f the rem aining 3 4 and 35 deletions recovered from L. cuprina and M. domestica, respectively.
A g a ro se gel electrophoresis revealed that none o f these deletions с о -m igrated w ith a k n o w n p recise e x c isio n event recovered fro m
D. melanogaster in w h ich hobo excision involved both termini, rather that they contained deletions w h ich w ere either larger o r sm aller in size.
W e therefore suggest that it is unlikely that these deletions have resulted from excision s in w h ich both hobo term ini have been involved . W e
are currently e xa m in ing these plasm ids in order to determine the structure o f the e xcision events.
 IAEA-SM-327/7 97

sion no additional D N A is found at the excision site, not all o f the P element is
excised and there is no duplication o f flanking genom ic D N A . The pattern o f hobo
excision in D. melanogaster resembled the excision o f A с elements and ТатЗ ele­
ments from maize and snapdragon, respectively, suggesting that these transposable
elements are members o f a single family o f eukaryotic transposable elements.
Sequence similarities between these transposable elements have previously been
identified [5]. Given the wide phylogenetic range o f the Ac element in particular, we
were interested to determine whether hobo could be.excised in non-drosophilids.
Demonstration o f hobo excision in non-drosophilids would indicate that hobo, or
hobo like, elements could be useful as transformation vectors in these insects. Pre­
blastoderm embryos o f L. cuprina and M. domestica were injected with indicator
plasmid in the presence or absence o f hobo transposase encoding plasmid. Excision
was observed in both species, however, at a lower frequency than that.observed in
D. melanogaster (Table I). Significantly, injection o f indicator plasmid alone
resulted in the recovery o f excision events from both species. This suggests that
L. cuprina and M. domestica contain endogenous factors which are capable o f
recognizing the hobo element and removing it from the indicator plasmid.
Sequence analysis o f the excision events recovered from L. cuprina and
M. domestica revealed that they did not resemble the excision events recovered from
D. melanogaster. Rather, the excision events appeared to be imprecise. In many
cases, large parts o f hobo remained at the empty excision site and, in addition, dele­
tions often extended several hundred base pairs into the flanking plasmid sequence.
In those excision events so far characterized from M. domestica and L. cuprina, in
only four and two cases, respectively, did at least one o f the break points occur
within hobo element terminal sequences. Excision assays conducted in M. domestica
using an indicator plasmid lacking hobo sequences resulted in the recovery o f lacZ
p la s m id s at a f r e q u e n c y fo rt y -t h re e f o ld lo w e r th a n that o b s e r v e d in a s s a y s u s in g the
hobo containing indicator plasmid, indicating that, even though imprecise, excision
was dependent on the presence o f the hobo sequences in the indicator plasmid.

3 .2 . C lon in g o f hobo like sequences fro m L. cuprina

A library constructed from size fractionated L. cuprina genom ic D N A was

screened with hobo element D N A under low stringency conditions [7]. One clone
that consistently hybridized to this probe was selected for further analysis. Partial
sequence analysis revealed this clone to contain sequences which were 50% similar
overall to the hobo element HFL1. M oreover, sequence similarity increased in three
regions previously identified as being conserved both at the nucleic acid and the
amino acid levels between hobo, Ac and ТатЗ. W hile this clone is still to be entirely
sequenced, the presence o f these conserved regions, together with a putative terminal
repeat sequence which is 83% similar to the hobo terminal sequences, suggests that
it is closely related to the hobo transposable element. In addition, translation o f
98 ATKINSON et al.

the putative hobo like transposase from this clone shows 39% identity and 59%
similarity over the entire region that has been sequenced.
Further evidence for the presence o f hobo like sequences in the genome o f
L. cuprina has been gained through the use o f the polymerase chain reaction (PCR).
Degenerate oligonucleotide primers based on the three regions o f similarity between
hobo, A c and ТатЗ were synthesized and used to amplify the intervening sequences.
Using primers from regions 1 and 2, an expected fragment o f 450 bp was amplified
from D . melanogaster D N A and, in addition, fragments o f similar size were ampli­
fied from L. cuprina and M . domestica. A ll the fragments hybridized strongly to a
radiolabelled hobo D N A probe. The sequence o f the 264 bp M . domestica clone
revealed it to have 66% similarity to hobo element HFL1. M oreover, translation o f
the putative open reading frame o f this clone showed an 81% similarity to hobo
transposase [11].

4. DISCUSSION

Through the use o f hobo excision assays we had previously demonstrated that
the hobo element can be m obilized in the soma o f D. melanogaster embryos and that
this excision was dependent on the presence o f hobo sequences in the indicator
plasmid. M oreover, the structure o f empty excision sites follow ing excision o f the
hobo element resembled the structure o f empty sites follow ing the excision o f the A c
element o f maize and the ТатЗ element o f snapdragon in that all o f the hobo element
w a s r e m o v e d a n d a d d it io n a l D N A , w h ic h w a s a n in v e rte d d u p lic a t io n o f f la n k in g
genom ic D N A , was left at the excision site. This supported the hypothesis proposed
by Calvi et al. [5] that these three transposable elements were members o f the one
family o f transposable elements. They identified three regions o f strong nucleic acid
and amino acid similarity between A c and hobo, one o f which was also com m on to
A c, ТатЗ and hobo.
Given the broad phylogenetic range o f species in which A c can be mobilized,
w e were interested in determining whether hobo could also be mobilized in a number
o f distantly related species to D . melanogaster. Previous studies based on the low
stringency hybridization o f hobo element D N A to genom ic D N A prepared from
134 dipteran species suggested that hobo had a very narrow host distribution within
dipterans, confined to the melanogaster and montium subgroups o f the melanogaster
species group [12].
W e chose two independent strategies to examine for the presence o f hobo, or
hobo like, elements in two non-drosophilids, L. cuprina and M . domestica. The
first, based on a genetic approach, assayed for the mobilization o f the hobo element
HFL1 in pre-blastoderm em bryos. Excision events were recovered from both
species. How ever, sequence analysis o f these excision products revealed significant
differences between their structure and the structure o f excision events recovered
 IAEA-SM-327/7 99

from D. melanogaster. Excision in both L. cuprina and M. domestica was impre­

cise, only part o f hobo was excised, and deletions extended into flanking genomic
D N A . That these excisions were dependent on hobo sequences in the indicator
plasmid was shown by the significant reduction in the recovery o f lacZ' plasmids
when excision assays were carried out with an indicator plasmid lacking hobo
sequences. Significantly, the presence or absence o f helper hobo encoded trans­
posase had little effect on either the frequency or the structure o f these excision
events.
The recovery o f hobo excision events from L. cuprina and M. domestica in
experiments in which hobo encoded transposase was absent suggests that both o f
these non-drosophilids possess factors functionally related to hobo transposase which
are capable o f recognizing and excising, albeit imperfectly, hobo elements. It is
clear, however, based on the structure o f the empty excision sites recovered from
L. cuprina and M. domestica, that these factors differ from hobo transposase in their
interactions with hobo element D N A . Whether this difference in the structure o f
these sites reflects differences in the excision process itself or in the, subsequent
repair o f these sites remains to be examined.
Further support for the existence o f hobo like elements in the genomes o f these
non-drosophilids was obtained by the cloning o f L. cuprina and M. domestica
sequences which have significant similarity to thé hobo element. Partial nucleic acid
and amino acid sequence analysis o f the L. cuprina clone in particular has shown that
the three regions identified by Calvi et al. [5] ás being conserved between hobo, Ac
and ТатЗ are also present in this clone. The question as to whether or not each o f
these clones represents a transposable element and, if so, whether they are functional
and autonomous, should becom e clear once sequence analysis is complete. Excision
and transposition assays using these elements in L. cuprina, M. domestica and
D. melanogaster should provide information on the mobility o f these elements in
each o f these species.
The presence o f hobo like sequences in L. cuprina and M. domestica, mem­
bers o f two families within the Diptera, suggests that hobo like sequences, and
perhaps hobo like transposable elements, may be more widely distributed throughout
the Insecta than previously thought. A s for the P element, the origin' o f hobo ele­
ments and the basis for their dispersal amongst drosophilids remain largely
unknown, although there is strong evidence for both elements being recent invaders
o f D. melanogaster [2]. The contribution, if any, o f horizontal transfer in the distri­
bution o f hobo is unknown. H ow ever, it has recently been suggested that the parasitic
mite, Proctolaelaps regalis, may be responsible for the physical movement o f P ele­
ments between some species o f the family Drosophilidae [13]. Given the diversity
o f organisms possessing the hobo/АсПатЗ family o f transposable elements, it seems
possible that horizontal transfer may have also once played a role in the distribution
o f these elements in extant species. Distribution o f this family o f elements within the
Insecta is currently the focus o f much activity within our laboratories. Isolation and
100 ATKINSON et al.

characterization o f hobo like elements in selected non-drosophilids might be

expected to be o f some use in the extension o f the techniques o f modern molecular
biology to comm ercially and medically important insects.

REFERENCES

[1] B L A C K M A N , R .K ., G E L B A R T , W .M ., “ The transposable element hobo of

Drosophila melanogaster", M o b ile D N A ( B E R G , D ., H O W E , M . M . / E d s ) , A m e r i­
can Society for M ic ro b io lo g y , W a shington , D C (1989) 5 2 3 -5 3 0 .
[2] P A S C U A L , L . , P E R I Q U E T , G . , D istribu tion o f hobo transposable elements in natural
populations o f Drosophila melanogaster, M o l. B io l. E v o l. 8 (1991) 2 8 2 -2 9 6 .
[3] B L A C K M A N , R .K ., K O E H L E R , M .M .D ., G R I M A I L À , R ., G E L B A R T , W . М .,
Identification o f a fully functional hobo transposable element and its use for germ -line
transform ation o f Drosophila, E M B O J. 8 (1989) 2 1 1 -2 1 7 .
[4] S T R E C K , R . D . , M a c G A F F E Y , J.E., B E C K E N D O R F , S . K . , T h e structure o f hobo
transposable elements and their insertion sites, E M B O J. 5 (1986) 3 6 1 5 -3 6 2 3 .
[5] C A L V I , B . R . , H O N G , T.J., F I N D L E Y , S . D . , G E L B A R T , W . M . , E v id e n c e fo r a
com m on evolutionary o rig in o f inverted repeat transposons in Drosophila and plants:
hobo, Activator and ТатЗ, C e ll 66 (1991) 4 6 5 -4 7 1 .
[6] A T K I N S O N , P .W ., O ’B R O C H T A , D . A . , The hobo transposable element of
Drosophila can be cross-m obilized, in houseflies and excised like the Ac element o f
maize, Proc. Natl. A cad . Sci. U S A (submitted fo r publication).
[7] P E R K I N S , H . D . , H O W E L L S , A . J ., D iv is io n o f Biochem istry and M o le c u la r B io lo g y ,
A u stralia n N ational U n ive rsity, C an b erra A C T (unpublished data).
[8] S A M B R O O K , J., F R I T S C H , E ÎF . , M A N I A T I S , T ., M o le c u la r C lo n in g : A L a b o ra ­
tory M a n u a l, 2n d edn, C o ld Sp rin g H a rb o r Laboratory, C o ld S p rin g H arbor, N Y
(1989).
[9] O ’B R O C H T A , D . A . , H A N D L E R , A . M . , M o b ility o f P elements in d ro soph ilid s and
nondrosophilids, Proc. Natl. Acad. Sci. U S A 8 5 (1988) 6 0 5 2 -6 0 5 6 .
[10] O ’B R O C H T A , D . A . , G O M E Z , S.P ., H A N D L E R , A .M ., P element excision in
Drosophila melanogaster and related drosophilids, M o l. Gen. Genet. 2 2 5 (1991)
.3 8 7 -3 9 4 . ,
[11] O ’B R O C H T A , D . A . , W ARREN, W . D . , C enter for A gric u ltu ral Biotechnology,
U n iv e rsity o f M a ry la n d , C o lle g e Park, M D (unpublished data).
[12] D A N I E L S , S .B ., C H O V N I C K , A ., B O U S S Y , I.A ., D istribution o f hobo transposable
elements in the genus Drosophila, M o l. B io l. E v o l. 7 (1990) 5 8 9 -6 0 6 .
[13] H O U C K , M .A ., C L A R K , J.B., P E T E R S O N , K .R ., K ID W E L L , M .G ., Possible
horizontal transfer o f Drosophila genes by the mite Proctolaelaps regalis, Science 253
(1991) 11 25 -11 29 .
 IAEA-SM-327/8

DEVELOPMENT AND USE OF DNA PROBES

IN MAPPING INSECT GENOMES

A .S . ROBINSON
Insect M olecular Genetics Group,
Institute o f M olecular B iology and Biotechnology,
Heraklion, Crete,
Greece

Abstract
D E V E L O P M E N T A N D U S E O F D N A P R O B E S IN M A P P IN G IN S E C T G E N O M E S .
C la ssica l genetic m a p p in g ha s relied on thé use o f m orphological, biochem ical and
D N A variation (restriction fragm ent length po lym o rp h ism s) to establish lin kage relationships
between loci in order to construct genetic m aps. T h is genetic a n alysis has been com plem ented
b y a cytological com ponent. Recent advances in m olecular b iological techniques w ill re volu­
tionize the m anner and the speed w ith w h ich genetic m aps can be constructed. The ‘m otor’
o f this revolution is the polym erase chain reaction. T h e paper concentrates on tw o techniques
w h ich are no w available, nam ely, m icrosatellites and random am plified polym orp h ic D N A
( R A P D ) , b y p ro v id in g an explanation o f the prin ciple s involved together w ith exam ples o f
their use. A com parative assessm ent is m ade o f the tw o approaches and som e em phasis given
to their use as tools fo r m apping specific genes.

1. H ISTO R IC AL PERSPECTIVE

Recent advances in molecular biology have dramatically increased the acces­

sibility o f insect genomes to genetic analysis. Given certain assumptions, it is possi­
ble to generate a detailed genetic map o f any insect species within an extremely short
time.
Analysis and mapping o f insect genomes has in the past relied on the avail­
ability o f spontaneous or induced mutations at specific loci, which were then used
to establish linkage groups and recombination distances between the loci. Using this
approach, linkage maps can be assembled and when combined with cytological tech­
niques, linkage group chrom osom e correlations can be established. This , is a
laborious procedure as it relies on the availability o f visible mutations, by. definition
rare events, and the collaboration o f many laboratories working on a single species.
This approach is typified in the example o f Drosophila melanogaster, where an
extremely detailed map has been assembled over 80 years by a large number o f
geneticists and cytogeneticists. The use o f protein variation, i.e. isozym es, improved
dramatically the analysis o f insect genomes as variation at this level is much more

101
102 ROBINSON

widespread and is readily accessible for analysis. H ow ever, there are real limits to
the number o f loci that can be studied. Nevertheless, using a combination o f
biochemical and m orphological markers, detailed genetic maps have been con ­
structed for several important pest or beneficial species, e .g . Lucilia cuprina [1],
Aedes aegypti [2], Musca domestica [3], Bombyx morí [4] and Ceratitis capitata [5].
For many years it has been recognized that variation at the D N A level can be
accessed using cloned D N A sequences and looking for restriction fragment length
polymorphisms (RFLPs), either within the cloned regions or in the flanking
sequences. Variations are monitored as changes in the length o f the defined D N A
fragments produced by digestion o f the D N A with restriction endonucleases. This
technology has been extensively used to produce detailed genetic maps in many spe­
cies [6]. Although this variation is ubiquitous, exhibits co-dom inance and is stably
inherited, serious practical limitations have been acknowledged: it is laborious, the
quality o f the D N A has to be high and the amount o f D N A required is such that only
a few RFLPs can be scored for any individual. This paper will not deal further with
RFLPs.

1.1. Importance of cytology

A s indicated above, cytology has been a key element in the development o f

genetic maps; this is especially so in many dipteran species where polytene chrom o­
somes are available. In polyteny, multiple rounds o f replication o f each chrom osom e
occur within a single nucleus, and the chrom osom e copies pair precisely, leading to
very long and thick chrom osom es characterized by a banding pattern that identifies
a particular chrom osom e and its various sections. In D . melanogaster, the com bina­
tion o f a large number o f m orphological markers and extremely fine cytological map­
ping has enabled a wonderfully precise genetic map to be assembled, encompassing
close to four thousand loci. The usefulness o f polytene chrom osom es for genome
mapping cannot be overemphasized as any cloned sequence can be mapped precisely
to its cytological location(s) using in situ hybridization.

1.2. Current approaches

The key to being able to map any particular genome is the availability o f varia­
tion at the D N A level and the accessibility o f that variation to analysis. The sequence
is the key, with variation occurring as specific changes in the nucleotide order or as
changes in the number o f times a particular sequence is repeated. Both these types
o f variation are well known, but up until now they have not been routinely accessible
to molecular analysis. This situation has now changed with the introduction o f the
polymerase chain reaction (PC R) [7].
The PCR utilizes in vitro enzymatic amplification o f a specific segment o f
D N A . Three nucleic acid segments are used, the double stranded D N A template and
 IAEA-SM-327/8 103

— Original DNA
New DNA
Primers

FIG. 1. Schematic o f the polymerase chain reaction.

two single stranded oligonucleotide primers that are complementary to the two
strands, at opposite ends o f the D N A segment. By adding a thermostable D N A poly­
merase, dNTPs and the appropriate buffer, the D N A segment between the primers
can be exponentially amplified (Fig. 1). This is done by multiple cycles o f denatur­
ing the double stranded D N A , annealing the primers and synthesizing a new double
strand o f D N A between the primers. This amplification product can be visualized
on slab gels follow ing ethidium bromide staining. The pow er o f this technique is
illustrated by the fact that if 30 o f these cycles were repeated, then a 27 million fold
amplification can be obtained from a specific segment. This means o f course that
very small amounts o f template D N A are needed and conversely many assays could
be carried out from a single individual. M ost o f the approaches outlined below can
be carried out on insect species where no genetic information is available.
104 ROBINSON

2. ANALYTICAL TOOLS USING PCR

2.1. Microsatellites

Microsatellites are dinucleotide or other simple repeat sequences, e.g. (C A )n,

and they have proved to be highly polym orphic and widely distributed in diverse
genomes, including the mouse and human. For example, there are 105 sequences
containing (C A )n repeats in the human genom e, mainly in the range o f 9 -3 0 repeat
units, and they seem to be homogeneously distributed, approximately one every
5 0 -1 5 0 kb [8]. When assayed using PCR, they define sequence tagged sites [9].
They are differentiated from minisatellites only by the size o f the repeat unit which,
in minisatellites, can be much larger. The principle (Fig. 2) is to screen a small insert
genom ic library with an end labelled (GA)n oligonucleotide. Once clones containing
these repeats are identified, then the unique sequences flanking them can be
sequenced and specific oligonucleotide primers designed which will then amplify the
repeat sequence. Following amplification, the products can be visualized on non­
denaturing acrylamide sequencing gels by ethidium bromide staining or autoradio­
graphy (if amplification is performed in the presence o f radioactive dNTP) and
differences as small as one repeat can be resolved. T o generate a high density map
using microsatellites, hundreds o f markers are necessary. Using conventional
libraries, this presents some technical problems but techniques are now available
to construct libraries which contain up to 50% o f clones with long microsatellite
repeats [10].
In the mouse, this technique has been extensively used to generate a high reso­
lution genetic map [11, 12]. For example, o f 50 microsatellites tested, 78% were
polym orphic between inbred strains and 92% were polym orphic between a wild
mouse strain and the inbred lines. The authors showed that the larger the number
o f repeats, the m ore alleles per locus were available for analysis. In the mouse, there
are about 105 copies in the genome and, if approximately 70% are polym orphic,
then a marker can be generated every 20 kb. Recently, the approach has been
extended to a mosquito, Anopheles gambiae [13], and here, out o f 34 (G T)n markers
tested, 27 revealed polymorphism. The presence o f polytene chrom osom es in this
species enabled in situ hybridization to be carried out to identify the exact location
o f specific markers. In general, clones with large numbers o f repeats hybridized to
many sites on the chrom osom es. When clones with less than 12 repeats were used,
a single hybridization site could be identified [13].

2.2. Arbitrarily primed DNA (AP-PCR) and random amplified

polymorphic DNA (RAPD)

This is the amplification o f genomic D N A with single primers o f arbitrary

nucleotide sequence. These primers will detect polymorphisms in the absence o f any
 IAEA-SM-327/8 105

Microsatellites

Screen library
with
(CA) probe

Sequence unique
TCTAGTTCAG (CA)n TCGATGACT
flanking DNA

Synthesize primers
to unique sequence

Use set of primers to amplify

genomic DNA
and analyse on agarose gel

! = ------------------- (CA)n ---------------------------

2 = -------------------- (CA)n ---------------------------------
3 = --------------- (CA)n+ i o ------------------------
4 = --------------------(C A )n- 1 0 -----------------------------

FIG. 2. Isolation and use o f microsatellites to reveal polymorphisms.

sequence information and they can be used as genetic markers to construct genetic
maps. The polymorphisms are visualized on ethidium stained agarose gels as the
presence or absence o f bands. The difference between the two approaches, A P-PCR
and R A P D , is quantitative, with R A P D being generated by 10 mer oligonucleotides,
while A P-PC R products are primed by much larger primers, up to 34 mers. In
general, there are more amplified bands using A P-PC R . Primers for R A P D analysis
 Mutation Small insertion within ©
a\
in annealing site the priming sites

Dominant versus recessive Co-dominant

(1) (2) (3) (1) (2) (3)

X X * X X X

> < X

R O B IN S O N
(1) (2) (3) (1)

FIG. 3. RAPD polymorphisms illustrating both dominant and co-dominant phenotypes from a single priming site. An asterisk represents a mutation
in an annealing site and an arrow the RAPD primers.
 IAEA-SM-327/8 107

are 9 -1 0 nucleotides in length, are between 5 0 -8 0 % G C and contain no palindromic

sequences. Using mammalian, plant and prokaryotic D N A , it was shown that these
primers can amplify polym orphic D N A which can be used for genetic mapping [14,
15]. The follow ing are some characteristics o f R A P D and A P -P C R products: they
are generally dominant, they can contain repetitive sequences, they can be used for
population studies [16] and those with repeat sequences cannot be used for correla­
tion o f cytological and genetic maps.
Figure 3 shows how R A P D can be visualized if á single site on the chrom o­
some is being amplified. I f there is a mutation in the annealing site, then both
hom ozygote 1 and the heterozygote will produce a single band, hom ozygote 3 will
produce no band and hence the R A P D is dominant. I f a small insertion is present
between the primer sites, then a co-dominant phenotype can be visualized. I f the
primer anneals to multiple sites, a much m ore com plex pattern o f amplified products
can be produced. In Fig. 4, four different individuals are depicted which have the
same mutation in different priming sites. In this case, the cryptic variation is not rev­
ealed, with all four, individuals producing an identical band on the agarose gel. In
Fig. 5 , with insertions and deletions in multiple sites, a much m ore com plex pattern
can be produced. For genetic mapping, the occurrence o f multiple sites is not a
problem . H ow ever, it is impossible to be specific about the relationships o f the
bands, i.e. are they alleles o r loci, until individual families are examined.
In the An. gambiae com plex, R A P D has been used to develop genetic markers
for differentiation o f the sibling species. So far, about fifty markers have been identi­
fied. How ever, in situ hybridization to polytene chrom osom es revealed that many
o f them hybridized to multiple sites [17]. In the Mediterranean fruit fly (m edfly),
initial results using 20 primers have clearly shown that differences in banding pattern
occur between populations [18] and a m ore detailed genetic analysis has enabled
R A P D to b e a s s ig n e d to the f iv e a u t o s o m e s [19]. H o w e v e r , the r e p r o d u c ib ilit y o f
the results has to be improved.
In the mouse, A P -P C R primers have been used very successfully to identify
strains and for genetic mapping [20, 21] and in Arabidopsis thaliana, R APD was
used- to contruct a detailed genetic map in a very short time [22].

3. M A PPIN G S T R A TE G Y

B efore embarking on a genome mapping exercise in an insect where there is

little formal genetics, it is important to consider the strains to be used and the strategy
for their use. A s already discussed, variation is the key to being able to map loci,
so the question is, what is the best way to maximize such variation (differences)
between the strains?
In plants, near isogenic lines and recombinant inbred lines have been exten­
sively used in conventional RFLP analysis to map genes o f econom ic importance and
108 ROBINSON

Mutations in annealing site

-> <— X * *

(1)

> < - * *

> '■ < -

(2)

> < - X "

(3)

* * *
(4)

> < * * *

(1) (2) (3) (4)

FIG. 4. RAPD amplification at multiple loci where cryptic variation is not revealed. An
asterisk represents a mutation in an annealing site and an arrow the RAPD primers.
 IAEA-SM-327/8 109

Insert
Small insertions and deletions
we

(1)

>
<~X~- + < T *
we
X -X X
(2)

we

X- -X--X-|^X
(3)

X X
Insert

Deletion we
> < >T< -1 ><
(4)

X -- ^ — ><— X
Deletion

(1) . (2) (3) (4)

FIG. 5. RAPD amplification at multiple loci. Inserts and deletions lead to different sized
bands which can be from different loci or the same locus. An asterisk represents a mutation
in an annealing site and an arrow the RAPD primers.
110 ROBINSON

to assemble genetic maps. These lines are now being very efficiently exploited using
R A P D . For example, in one month three R A P D markers were identified as being
closely linked to a Pseudomonas resistance gene in the tomato, whereas using the
conventional approach o f RFLP mapping, a period o f two years was needed [23].
In A. thaliana, 225 lo ci were identified and mapped in 8 person-months [22]. In
most insect species, however, such lines are not available and in many cases cannot
be constructed or, even if they could, would be extremely time consuming to
produce; therefore other strategies are necessary. T o demonstrate an alternative
approach, the m edfly, Ceratitis capitata, can be used as an example. In this species,
a collection o f mutations, both biochemical and m orphological, is available and mul­
tiple marker strains can be synthesized. Initially, one can construct a strain which
is hom ozygous for at least one marker per aùtosome and different wild type strains
can be crossed with this strain to produce an F! generation. Following backcrossing
o f individual F]S to the multiple marker strain, a segregating . F 2 generation is
produced which can then be uséd to assign molecular markers, either R A P D or
microsatellites, to their autosomes. A s these techniques are PCR driven, very small
amounts o f D N A are necessary and up to 40 different molecular markers can be
assayed from a single pair mating in a day.
The type o f strategy adopted will depend on the species being investigated and
the availability o f markers, inbred lines, etc.; however, an investment o f time in the
construction o f the relevant strains will increase significantly the speed with which
a genetic map can be assembled.
A note o f caution is necessary, however, when using R A P D . A s the D N A
being amplified is random and o f unknown sequence, it is essential that molecular
analysis be performed on a defined cross, preferably with markers. This is to ensure
that the polymorphism observed on the gel can be shown to demonstrate Mendelian
inheritance and that it is not simply a PCR artefact.

3.1. Identification of molecular markers linked to specific genes

In som e insect vectors o f disease, genes have been identified which interfere
with the ability o f the insect to transmit the disease, the so-called refractory genes
[24]. A study o f such genes will lead to a better understanding o f the disease and
perhaps to new methods o f control.
In general, the cloning o f such genes requires some knowledge o f the transcript
or polypeptide and in most insect vectors that knowledge is unavailable. If molecular
markers can be identified which are tightly linked, they can be used to initiate a
chromosomal ‘ walk’ to the gene o f interest. T o use this approach it is necessary to
saturate the genomic area o f interest with molecular markers and assess linkage rela­
tionships. A new technique called ‘bulked segregant analysis’ will be an extremely
valuable asset to this approach [25]. Bulked segregant analysis involves the screening
o f tw o pools o f D N A in a segregating population which has been generated from a
 IAEA-SM-327/8 111

single cross. Screening is done on D N A from the parental lines and from the
segregating F 2 populations. Using this approach with R A P D , it is possible to
differentiate between polymorphisms resulting from variation between the parental
lines and polymorphisms linked to the genom ic area o f interest.

3.2. Correlation of a molecular map with a cytological map

Polytene chrom osom es offer a unique possibility o f combining a recombina 1

tional map with a cytological map. Once cloned or amplified sequences have been
mapped recombinationally, they can be assigned to their chromosomal location using
in situ hybridization to polytene chrom osom es. Not only does this enable genetic dis­
tances to be correlated with.cytological distances, it also enables the recombinational
map o f a linkage group to be correctly orientated on the polytene chrom osom e; this
approach has been successfully demonstrated in the medfly [26].

4. CONCLUSIONS

Before concluding with a presentation o f the relative merits o f microsatellites

versus R A P D for genom e mapping, it is important to stress the significance o f both
these methods for the genome analysis o f insect pests:

— For neither o f the approaches is radioactivity necessary;

— Many assays can be performed per individual;
— The quantity and quality o f the D N A need not be high;
— Southern blotting is not used, so polymorphisms can be detected in fragments
containing highly repeated sequences;
— Both approaches can be carried out with low technology input;
— N o prior genetic or molecular information is required;
— An almost unlimited number o f markers is available.

It is clear that molecular markers o f the type outlined in this paper can provide
a rapid entry point into the genomes o f insect pests; there are, however, important
differences between R A P D and microsatellitès. First, technical differences: for
R A P D no libraries are necessary, only D N A and primers, which are comm ercially
available, whereas for microsatellites a library must first be constructed, screened
for microsatellites, follow ed by sequencing o f the unique flanking region and the
synthesis o f specific primers. The reproducibility o f results using R A P D can be a
problem due to PCR artefacts occurring. This might lead to difficulties in comparing
results between different laboratories, whereas for microsatellites, a collection o f
specific primers will be a very powerful tool for genome analysis with high repeata­
bility. Second, genetic differences: the most important is the fact that R APD is
usually a dominant marker, whereas microsatellites are co-dominant. H ow ever, for
112 ROBINSON

mapping this is not a serious problem as backcross populations map dominant and
co-dominant markers with equal efficiency [22]. This difference does, however,
becom e important when these two systems are applied to population genetic
problems [16]. With microsatellites, loci with high numbers o f repeats frequently
have multiple alleles/loci and in general the larger the number o f repeats the more
alleles/loci; R APD does not in general exhibit multi-allelism. An important factor
if in situ hybridization is to be included in the genome analysis is the occurrence o f
repeat sequences in the probe used for the hybridization. Data from plants have indi­
cated that up to 50% o f R A P D loci hybridize to single copy D N A [14]; this has now
been supported by data from the mosquito [17]. For microsatellites, the situation is
clearly different and without further modification only a small proportion o f the
microsatellite clones will hybridize to a single polytene site, i.e. those containing low
numbers o f repeats.
It is at the moment unwise to be dogmatic as to which o f the two approaches
is best; it all depends on the question being asked and the species under considera­
tion. How ever, independent o f the choice o f the method, the ready availability o f
these molecular tools will open up pest insect genomes to sophisticated analysis.

ACKNOW LEDGEM ENTS

The author wishes to thank F .C . Kafatos for his comments on the manuscript.
The discussions with colleagues in the Insect M olecular Genetics Group have been
much appreciated. N. Kelaidi is thanked for secretarial help.

REFERENCES

[1] F O S T E R , G .G ., W H I T T E N , M .J ., K O N O V A L O V , C ., A R N O L D , S . T . A . , M A F F I ,
G . , A u toso m al genetic m aps o f the A u stralia n sheep b low fly, Lucilia cupriana dorsalis,
and possible correlations w ith the linkage m aps o f Musca domestica and Drosophila
melanogaster, Genet. R e s. 3 7 (1981) 5 5 -6 9 .
[2] 0 ‘B R I E N , S.J., Ed ., Genetic M a p s 1990, C o ld S p rin g H a rb o r Laboratory, C o ld Sp rin g
H arb o r, N Y (1990).
[3] M A L A C R I D A , A ., G A S P E R I , G ., B I S C A L D I , G .F ., M I L A N I , R „ Signifícate funzi-
onale di grup p i di concatenazione in Musca domestica e altri ditteri, S. It. E . Atti
5 (1985) 3 0 7 -3 1 0 .
[4] S T R U N N IK O V , V .A ., S e x control in silkw o rm s, Nature (L o n d o n ) 2 5 5 (1975)
11 1-113 .
[5] M A L A C R I D A , A ., G A S P E R I , G ., M I L A N I , R ., “ G enom e organization o f Ceratitis
capitata: L in k a g e grou p s and evidence fo r sex-ratio distorters” , F ru it F lie s (Proc. 2nd
Int. Sym p . C o lym b a ri, Crete, 1986) ( E C O N O M O P O U L O S , A .P ., Ed .), Elsevier,
A m sterd am (1987) 16 9-174 .
 IAEA-SM-327/8 113

[6] B E C K M A N N , J.S., S O L L E R , М . , Restriction fragm ent length p o lym o rp h ism s and

genetic im provem ent o f agricultural species, E up h ytica 3 5 (1986) 1 1 1-124 .
[7] S A K I , R . K . , et al., Prim er-directed enzym atic am plification o f D N A w ith a therm o­
stable D N A polym erase, Science 2 3 9 (1988) 4 8 7 -4 9 1 .
[8] H A M A D A , H ., P E T R I N O , M . G . , K A K U N A G A , T ., A novel repeat element with a
Z -D N A fo rm in g potential, is w id ely found in evolutionarily diverse eukaryotic
genom es, Proc. N a tl A cad . Sci. U S A 7 9 (1982) 6 4 6 5 -6 4 6 9 .
[9] O L S O N , М . , H O O D , L ., C A N T O R , C ., B O T S T E I N , D ., A com m on language for
p hy sical m apping o f the hum an genom e, Science 245 (1 98 9) 1 4 3 4 -1 4 3 5 .
[10] S T R A N D E R , E . A . , J O N G , P . M . , R I N E , J., D U Y K , G ., C onstruction o f sm all-insert
ge nom ic D N A libraries h ig h ly enriched fo r m icrosatellite repeat sequences, Proc. Natl.
A cad . Sci. U S A 8 9 (1992) 3 4 1 9 -3 4 2 3 .
[11] C O R N A L L , R .J., A I T M A N , T.J., H E A R N E , C M . , T O D D , J .A ., T h e generation o f
a lib rary o f P C R -a n a ly z e d m icrosatellite variants fo r genetic m apping on the m ouse
genom e, G e n o m ics 10 (1 99 1) 8 7 4 -8 8 1 .
[12] L O V E , J .M ., K N I G H T , A . M . , M c A L E E R , M . A . , T O D D , J .A ., T o w a rd s construc­
tion o f a h ig h resolution m ap o f the m ouse genom e u sin g P C R analyzed microsatellites,
N u cle ic A c id s Res. 18 (1990) 4 1 2 3 -4 1 3 0 .
[13] Z H E N G , B ., K A F A T O S , F . C ., Departm ent o f C e ll and D evelopm ental B io lo g y , H a r ­
vard U niv e rsity , C am bridge, M A , personal com m unication.
[14] W I L L I A M S , J .G .R ., K U B E L IK , A .R ., L I V A K , J., R A F A L S K I , J.A .,
T I N G E Y , S . V ., D N A p o lym o rp h ism s am plified by arbitrary prim ers are useful as
genetic m arkers, N u cle ic A c id s Res. 18 (1990) 6 5 3 1 -6 5 3 5 .
[15] W E L S H , S., M c C L E L L A N D , м . , Fin ge rp rin tin g genom es u sin g P C R with arbitrary
prim ers, N u cle ic A c id s Res. 18 (1990) 7 2 1 3 -7 2 1 8 .
[16] K A M B H A M P A T I , S., B L A C K , W . C . , IV , R A I , K . S . , R A P D - P C R fo r identification
and differentiation o f m osquito species and populations: Techniques and statistical ana­
lysis, J. M e d . Entom ol. (in press).
[17] L O U I S , C ., Insect M o le c u la r G enetics G ro u p , Institute o f M o le c u la r B io lo g y and
Biotech no logy, H e ra k lio n , Crete, Greece, personal com m unication.
[18] R O B I N S O N , A . S . , Insect M o le c u la r G enetics G rou p, Institute o f M o le c u la r B io lo g y
and B iotech nology, H era klion, Crete, Greece, unpublished results.
[19] M A L A C R I D A , A ., G A S P E R I , G ., D ipartim ento d i B io lo g ía A nim a le, U niversità di
Pavia, Pavia, Italy, personal com m unication.
[20] W E L S H , J., P E T E R S O N , C ., M c C L E L L A N D , М ., P o ly m o rp h ism s generated by
arbitrarily prim ed P C R in the m ouse: A p p lica tion to strain identification and genetic
m apping, N u cle ic A c id s Res. 18 (1991) 3 0 3 -3 0 6 .
[21] W E L S H , J., M c L E L L A N D , М ., G en om ic fingerprinting u sin g arbitrarily prim ed P C R
and a m atrix o f pairw ise com binations o f prim ers, N u cle ic A c id s Res. 19 (1991)
5 2 7 5 -5 2 7 9 .
[22] R E I T E R , R .S ., et al., G lo b a l and local genom e m apping in Arabidopsis thaliana by
u sin g recom binant inbred lines and random am plified p olym orp h ic D N A s , Proc. Natl.
A cad . Sci. U S A 8 9 (199 2) 1 4 7 7 -14 81 .
114 ROBINSON

[23] M A R T I N , G .B ., W I L L I A M S , J .G .K ., T A N K S L E Y , S . D ., R a p id identification o f
m arkers linked to a Pseudomonas resistance gene in tomato by u sin g random prim ers
and near-isogenic lines, Proc. Natl. A cad. Sci. U S A 8 8 (1991) 2 3 3 6 -2 3 4 0 .
[24] C O L L I N S , F . H . , et al., Genetic selection o f a Plasmodium-rtinciory strain o f the
: m alaria .vector, Anopheles gambiae, Science 2 3 4 (1986) 6 0 7 -6 1 0 .
[25] M I C H E L M O R E , R . W . , P A R A N , I . , K E S S E L I , R . V . , Identification o f m arkers linked
to disease resistance genes b y bulked segregant analysis: A rapid method to detect m ar­
kers in specific ge nom ic regions by u sin g segregating populations, Proc. Natl. Acad.
S c i. U S A 8 8 (1991) 9 8 2 8 -9 8 3 2 .'
[26]. S C O T T , М ., K R I T I C O U , D ., R O B I N S O N , A .S . , Isolation o f c D N A s encoding
6-phosphogluconate dehydrogenase and glucose-6-phosphate dehydrogenase fro m the
Mediterranean fruit fly, Ceratitis capitata: C orrelatin g genetic and physical m aps o f
chrom osom e 5, Insect M o l. Biol, (submitted for publication).
 IAEA-SM-327/10

ARE IMMEDIATE EARLY

BACULOVIRUS GENE PROMOTERS
USEFUL TOOLS IN INSECT CONTROL?

R . H U YB RE C H TS, M . V A N BRUSSEL,
V . V U LSTÉ K E , L . SCHOOFS
Z oological Institute,
Katholieke Universiteit Leuven,
Leuven

J. ROBBEN, G. V O L C K A E R T
Gene Technology ,
Katholieke Universiteit Leuven,
Heverlee

Belgium

Abstract
A R E I M M E D I A T E E A R L Y B A C U L O V IR U S G E N E P R O M O T E R S U S E F U L T O O L S IN
IN S E C T C O N T R O L ?
Research o n the genetic m anipulation o f baculoviruses fo r insect pest control has raised
tw o problem s. T h e first concerns the choice o f v iral prom oter used to direct the expression
o f an introduced insecticidal gene. T h e second points tow ards the k in d o f insecticidal gene
w h ich should be preferentially used in order to select m a xim u m effect on the target insect but
the m in im u m impact o n the environm ent. A sum m ary is g iv e n o f the m ajor argum ents in
favour o f the use o f im m ediate early b a culovirus gene prom oters w h ich direct the expression
o f insect derived peptide genes. It is thought that a correct com bination w ill affect the insect
at the first site o f infection, the insect gut.

1. IN TR O D U C TIO N

Nuclear polyhedrosis viruses (N PVs) bear no morphological resemblance to

any known plant or vertebrate viruses. They are fairly host specific! They do not
affect beneficial non-target insects and have successfully passed extensive safety tests
on mammals, birds and fish. Thus, baculoviruses can be considered safe and ecolog i­
cally acceptable for the regulation o f insect pest populations [1]. H ow ever, field
experiments have shown that baculovirus epizootics frequently terminate infestations
o f insect pests, but that collapse o f the target populations usually occurs after severe
crop damage has been inflicted [2]. This phenomenon may limit the use o f
baculoviruses to the biocontrol o f the forest ecosystem , rather than to an agricultural

115
116 HUYBRECHTS et al.

environment in which either a m ore straightforward ‘ knock d ow n ’ effect or at least

a direct block in food intake is envisaged.
At the laboratory level, this inconvenience can be overcom e by the intro­
duction o f an extra gene, bearing insecticidal potentials, into the baculovirus
genom e [3]. Several models o f genetically improved recombinant baculoviruses have
been proposed. The use o f either an extra polyhedrin promoter or the authentic
P10 promoter, or a combination o f both, was favoured in order to ensure production
o f the occluded virus necessary for oral infestation [4]. Depending on the insecticidal
activity spectrum o f the expressed recombinant protein, a more or less pronounced
reduction in survival time follow ing a single application has been reported.
H ow ever, both the polyhedrin promoter and the P10 promoter are ascribed to the
very late phase o f viral gene regulation and even appear not to becom e activated in
insect gut cells, the first site o f attack by the virus. A minimal latency time before
observing any reduction in crop damage is unavoidable for these types o f control
vector. Already in 1988, Crawford and M iller [5] suggested the potential o f early
baculovirus genes for the expression o f foreign genes in the early stages o f infection.
In analogy to the situation in Autographa califomica, we cloned a sequentially
related immediate early gene o f the Bombyx mori NPV [ 6], studied the promoter
characteristics o f the corresponding upstream sequence by using transient transfec­
tion assays, and are now trying to creaté an insect control véctor based on the expres­
sion o f insect derived m yo-active peptide genes under immediate early gene
promoter control.

2. THE M O D E L SYSTEM : BmNPVIEG

The 3.8 kb C L a l fragment, spanning the 9 4 -9 6 % map unit (mu) region o f the
Bombyx mori N PV physical map [7], includes an open reading frame o f 1572 bp as
well as a 631 bp upstream regulatory region showing a high degree o f sequence
hom ology with A cM N PVIE-1 [ 6]. Using a polymerase chain reaction, the existence
o f an extra upstream regulatory region, responsible for an extended spliced expres­
sion product, corresponding to A cM N P V IE -0 was also evidenced. Apart from dem­
onstrating that the BmNPVIEG gene product has transactivating capabilities towards
viral delayed early genes [6], it became clear that the conditions needed for promoter
activation o f the cloned leader sequence were fulfilled in permissive as well as non-
permissive insect cells [ 8]. On the other hand, transfection o f a fusion construct in
which the bacterial beta-galactosidase gene was inserted in frame downstream o f the
BmNPVIEG promoter allowed us to conclude, that neither V ero cells (monkey
kidney cells) nor. developing zebra fish em bryo cells were suitable hosts because
beta-galactosidase activity was never observed. Although further testing o f the
transcriptional activation o f the BmNPVIEG-ZacZ construct by vertebrate cell type
transcriptional machinery will be needed, present results point to the assumption that
 IAEA-SM-327/10 117

the Bm NPVIEG promoter could be safely used in the field. In vitro, this BmNPVIEG
promoter, being isolated from any other cis or transacting viral sequence, remains
operational on a continuous, prolonged base [9]. On the basis o f the pronounced
sequence and the functional similarity to A cM N P V IE -1, transcription o f a gene
inserted downstream o f the BmNPVIEG promoter might also be expected to occur
throughout the entire infection cycle [10]. W e also tried a gene knock out experiment
in which the endogenous BmNPVIEG coding sequence was replaced, through
hom ologous recombination, by a fused lacZ-IEG sequence. T o our surprise, apart
from a wild type virus, the resulting inoculum also contained a virus that produced
blue plaques after reinfection and addition o f an X -gal substrate. Unfortunately, with
respect to the wild type virus, the recombinant virus titre was decreased some
1000 times, making it quite impossible to purify the plaque. A t this point, it remains
open whether the original BmNPVIEG was indeed knocked out or whether the blue
plaques were rather a result o f a random integration into a site, effecting optimal
viral replication. In any case, in view o f the wide range o f viral transactivating
activity o f this immediate early gene, its functional presence in the viral genome will
most probably be obligatory. In our m odel, we are considering duplicating the
BmNPVIEG promoter and inserting the second promoter in tandem downstream ó f
the P I0 promoter.

3. CHOICE OF IN SE CTIC ID A L GENE

A s pointed out in Section 1, the insecticidal activity spectrum o f the gene used
for the realization o f a recombinant baculovirus will be a major determinant for
success. It might appear obvious to use well known toxin genes from which the
resulting toxin has already shown its potential for insect control. A good example
is Bacillus thuringiensis toxins. H ow ever, one should take into consideration that
these proteins exert their effect follow ing proteolytic activation and subsequent
binding to a receptor on the luminal side o f the gut. A first report on experiments
using an occluded recombinant virus carrying the Bikurstaki H D -73 delta-endotoxin
was not encouraging [11]. Neither were those experiments in which a baculovirus
expressing juvenile hormone esterase, a naturally occurring enzyme responsible for
juvenile hormone (JH) inactivation, was used [12]. So far, only the baculovirus
carrying a gene encoding the insect specific toxin, A alT , isolated from the venom
o f the scorpion Androctonus autralis, reduced the killing time o f the test insect by
40% [13]. Although these examples might still be somewhat disappointing, one
should not forget that the exploration o f this field has just .begun and that a major
breakthrough has still to be made.
In this context, we decided to explore further the potentials o f typical insect
neuropeptides that participate in the normal physiology o f the insect. Recently, our
research group [14], as well as several others, realized a major breakthrough by
118 HUYBRECHTS et al.

isolating and sequencing whole families o f bioactive neuropeptides. How ever, so far
only a very, limited success in cloning the corresponding genes has been reported.
In view o f our approach, and until the original genes becom e available, we can only
rely on the construction o f synthetic genes. In a first attempt, the genes encoding two
Locusta migratoria myotropins and one myo-inhibiting peptide were expressed in a
prokaryotic system and tested for bioactivity in a midgut contraction assay. Final
conclusions cannot yet be drawn, but we consider it appropriate to test these genes
for correct expression and processing in our eucaryotic, BmNPVIEG based stable
expression system [9]. This, in combination with practical in vitro hindgut contrac­
tion assay, will enable us to screen for the correct genes that might be worth while
testing in vivo by a recombinant baculovirus in which the peptide gene will becom e
expressed by a doubled immediate early gene promoter.

4. CONCLUSIONS

It is generally accepted that recombinant NPVs expressing insect selective

toxins, hormones or enzymes could enhance their insecticidal properties. How ever,
since a straightforward approach is not yet available, we started to develop an alter­
native model system: the immediate early gene promoter directed expression o f a
synthetic peptide gene where the expression product interacts with the muscle layer
surrounding the intestinal tract. From the explanation given it is clear that we are
still quite far from the ready-to-use realization o f an effective recombinant virus.
H ow ever, we are convinced that experimental incorporation o f an easy in vitro
screening system will help in choosing the right insecticidal gene. On the other hand,
we are quite aware o f the many difficulties we still have to circumvent.

REFERENCES

[1] B a cu lo viru se s fo r Insect Pest Control: Safety C onsiderations ( S U M M E R S , М .,

E N G L E R , R .„ F A L C O N , L . A . , V A I L , P., E d s), A m e ric a n Society for M ic ro b io lo g y ,
W ashington , D C (1975).
[2] C U N N I N G H A M , J.C ., “ Fie ld trials w ith baculoviruses: C o n tro l o f forest, insect
pests” , M ic ro b ia l and V ir a l Pesticides ( K U R S T À K , E., Ed .), D ekke r, N e w Y o r k
(1986) 33 5 -3 8 6 .
[3] K E E L Y , L . L . , H A Y E S , Т .К ., Speculations on biotechnology ápplicátions for inseót
neuroendocrine research, Insect Biochem . 17 (1987) 6 3 9 -6 5 1 .
[4] M I L L E R , L . K . , Insect baculoviruses: Pow e rful gene e xp re ssio n vectors, B io E s s a y s 11
(1989) 9 1 -9 5 .
[5] C R A W F O R D , A . M . , M I L L E R , L . K . , Characterization o f an early gene accelerating
expression o f late genes o f the b aculovirus Autographa califomica nuclear p olyh ed rosis
virus, J. V iro l. 62 (1988) 2 7 7 3 -2 7 8 1 .
 IAEA-SM-327/10 119

[6] H U Y B R E C H T S , R ., G U A R IN O , L ., VAN B R U S S E L , М ., V U L S T E K E , V .,
Nucleotide sequence o f a transactivating Bombyx mori nuclear p olyh e d rosis virus,
B io p h y s. A c ta 1 1 29 (199 2) 3 2 8 -3 3 0 .
[7] M A E D A , S., M A J I M A , K . , M o le c u la r clo n in g and p hysical m apping o f the genom e
o f Bombyx mori nuclear p olyh e d rosis v iru s, J. Gen. V iro l. 71 (1990) 18 51 -18 55 .
[8] H U Y B R E C H T S , R ., V U L S T E K E , V . , C lo n in g o f an early gene prom oter o f the
Bombyx mori b a culovirus and its use fo r transient and permanent expression o f foreign
genes in eucaryotic cells, M L F R A 3 , M e d e d . L an d b H o o ge sch . O p zock Stnts Gent 56
(1 99 1) 1 4 1 1 -1 4 1 4 .
[9] V U L S T E K E , V ., V A N B R U S S E L , М ., J A N S S E N , I., H U Y B R E C H T S , R ., Paper
IA E A - S M - 3 2 7 / 1 1, these Proceedings.
[10] K O V A C S , G .R ., G U A R I N O , L . A . , S U M M E R S , M . D . , N o v e l regulatory properties
o f the IE 1 and IE O transactivators encoded b y the ba culovirus Autographa califomica
m ulticapsid nuclear p olyh e d rosis v irus, J. V ir o l. 6 5 (1991) 5 2 8 1 -5 2 8 8 .
[11] M E R R Y W E A T H E R , A .T . , et al., C onstruction o f genetically engineered b aculovirus
insecticides containing the Bacillus thuringiensis subsp. hurstaki H D - 7 3 delta-
endotoxin, J. G en. V ir o l. 71 (1990) 1 5 3 5 -1 5 4 4 .
[12] H A M M O C K , B .D ., B O N N IN G , B .C ., P O S S E E , R .D ., H A N Z L IK , T .N .,
M A E D A , S., E x p re ssio n and effects o f the ju ve nile horm one esterase in a b aculovirus
vector, Nature (L o n d o n ) 3 4 4 (199 0) 4 5 8 -4 6 1 .
[13] M c C U T C H E N , B . F . , et al., Developm ent o f a recom binant b a culovirus expressing an
insect-selective neurotoxin: Potential fo r pest control, B iotech no logy 9 (199 1) 84 8 -8 5 2 .
[14] S C H O O F S , L ., H O L M A N , G .M ., H A Y E S , Т .К ., N A C H M A N , R.J.,
D E L O O F , A ., Isolation, identification and synthesis o f locustam yoinhibiting peptide
( L O M - M I P ) , a n o ve l b io lo g ica lly active neuropeptide fro m Locusta migratoria, R egul.
Peptides 3 6 (1 99 1) 11 1 -1 1 9 .
 IAEA-SM-327/11

DEVELOPMENT OF AN ALTERNATIVE
BACULOVIRUS EXPRESSION SYSTEM
USING AN IMMEDIATE EARLY GENE
OF THE BACULOVIRUS OF Bombyx morí

V . V U LSTE K E , M . V A N BRUSSEL,
I. JANSSEN, R. HUYBRECHTS
Z oological Institute,
Katholieke Universiteit Leuven,
Leuven,
Belgium

Abstract
DEVELOPMENT OF AN ALTERNATIVE BACULOVIRUS EXPRESSION SYSTEM
USING AN IMMEDIATE EARLY GENE OF THE BACULOVIRUS OF Bombyx mon.
The classic baculovirus expression system makes use o f recombinant viruses to
infect cells. This infection can, by the lytic activity o f the virus, only result in transient
expression o f the recombinant protein. In this study, an alternative baculovirus expression
system is demonstrated which is based on the property o f viral immediate early genes, to be
transcribed by uninfected cells in the absence o f other viral gene products. An immediate early
gene o f the baculovirus o f Bombyx mori (BmNPVIEG) has been isolated and sequenced.
By using an expression vector in which the lacZ reporter gene is placed under BmNPVIEG
promoter control, expression o f the lacZ reported gene in insect cells in a transient way was
found. When the same expression vector was co-transfected with a construct containing the
hygromycin В resistance gene, transformed Drosophila S2 cells were able to express the
reporter géne continuously.

1. IN TROD U CTION

Baculoviruses are insect pathogenic viruses with a circular, double stranded

D N A genome. In recent years, a subgroup o f these baculoviruses, the nuclear
polyhedrosis viruses (N PV s), and particularly the Autog rapha califomica nuclear
polyhedrosis viruses (A cN P V ), has received considerable attention because o f its
potential use as expression vector. Baculovirus genes, based on their temporal
expression during infection, can be divided into four classes: immediate early,
delayed early, late and very late genes.
The classic baculovirus expression system makes use o f the hyperexpressed
late viral polyhedrin gene. Insect cells are infected with recombinant viruses in which
the desired coding.sequence is placed under the control o f the polyhedrin promoter.
Because o f the strength o f this promoter, there is hypertranscription o f the foreign
gene insert at a late stage in the infection cycle.

121
122 VULSTEKE et al.

A disadvantage o f this classic baculovirus expression system is the lytic

activity o f the recombinant virus. Recombinant proteins can only be transiently
expressed and at the late phase o f infection, when this hyperexpression occurs,
the ability o f the host cells to process newly synthesized proteins might be
com prom ised [1].
In order to avoid the use o f these lytic viruses, we decided to follow the
approach used by Jarvis et al. [2]. Instead o f using the polyhedrin promoter, they
constructed an expression vector based on the promoter o f a viral immediate early
gene isolated from the Autographa califomica nuclear polyhedrosis virus
(A cM N PV IE-1 gene).
Immediate early genes are defined as those genes that can be transcribed
without earlier expressed viral gene products in transient assays.
By using an immediate early gene o f the baculovirus o f Bombyx mori
(Bm NPVIEG ), we seek to develop an alternative baculovirus expression system.

2. RESULTS A N D DISCUSSION

2.1. Isolation and sequencing of BmNPVIEG

A 3823 bp long C lal fragment, cross-hybridizing to the A cM N PVIE-1

sequence, has been cloned, subcloned and entirely sequenced by Huybrechts and c o ­
workers [3]. This fragment contains an open reading frame o f 1752 bp, encoding
a protein with a molecular weight o f 67 kilodalton, flanked by a 631 bp leader
sequence and 1440 downstream bases. Comparison o f this gene sequence with that
o f the AcM N PVIE -1 gene shows 95 .3 % hom ology.
The ‘ C A G T ’ m otif, the transcription initiation sequence which is perfectly
conserved for a number o f early baculovirus genes [4], even as putative C A A T and
T A T A like sequences, was found in the leader sequence.
Experiments already performed with the AcM N PVIE -1 gene revealed that the
gene product is involved in the transactivation o f a number o f delayed early genes
[5 -7 ] and at least one late viral gene [8].
A co-transfection o f Sf9 cells with a plasmid containing the BmNPVIEG
sequence and a construct in which the chloramphenicol acetyl transferase (C A T )
gene was placed under control o f the delayed A cM N P V 39K gene promoter dem on­
strated the cross-transactivating properties o f BmNPVIEG. The activity o f the
delayed early gene promoter was dependent on the presence o f the BmNPVIEG
product.
 IAEA-SM-327/11 - 123

2.2. Use of the BmNPVIEG promoter for foreign gene expression

in uninfected insect cells

Because immediate early genes are defined as those genes that, can be tran­
scribed by uninfected cells in, the absence o f other viral gene products, we wanted
to prove that the 631 bp leader sequence contains the regulatory sequences necessary
for transcription o f the transactivating gene in transient assays.
For this, we used the comm ercial pCH 110 promoter screening vector
(Pharmacia) in which a functional E. coli lacZ gene is cloned. W e inserted the
631 bp leader sequence in such a way that expression o f the lacZ gene was dependent
on the promoter activity o f the Bm NPVIEG leader sequence. For all the transfection
experiments, we used the transfection agent D O T A P (Boehringer Mannheim
Gm bH, Germany) and 5 -1 0 /ig circular supercoiled plasmid D N A .
After transfection o f Bombyx mori Bm-5 cells, we found cells expressing
the lacZ reporter gene. Even when a Bm NPV semi-permissive cell line such as
Spodopiem frugiperda Sf9 cells, or a Bm NPV non-permissive cell line such as
Drosophila S2 cells was transformed, expression o f the reporter gene was clearly
found. In this way, we concluded that the BmNPVIEG leader sequence contains
promoter activity by which foreign genes can be expressed in uninfected insect cells.
So far, the lacZ reporter gene is only transiently expressed under BmNPVIEG
promoter control. A further objective o f our study was to produce permanent trans­
formants. For this, we had to extend our transforming vector with an antibiotic
resistance gene. pUChshyg (kindly provided by J. Carlson, C olorado State Univer­
sity, Fort Collins, C olorado, United States o f Am erica) contains the hygrom ycin В
resistance gene under control o f the Drosophila hsplO prom oter; which can be
activated in the available Drosophila S2 cells.
In the coding sequence o f Bm NPVIEG, cloned into pU C 19, the lacZ gene is
inserted in a frame 54 bp downstream o f the BmNPVIEG transcriptional start site.
Transformation o f S2 cells with this construct (pBm N PVIEG /acZ) is also successful.
This prompted us to co-transfèct S2 cells with pBm N PVIËG lacZ and
pUChshyg. The heterogeneous mixtures o f the transformed cells are cultured in
selective medium containing 300 /¿g o f hygrom ycin В and those cells that even
contain pBm N PVIEG /acZ can be m icroscopically detected by adding X -G al as sub­
strate for /3-galactosidase.
W e follow ed the transformed cells for over one hundred passages and still
found /3-galactosidase expressing cells. Even when they are cultured, for over
30 passages, in non-selective medium they retain their transformed state.
Total cellular D N A , extracted from the transformed cells, was digested with
X bal and X h ol. X bal has one restriction site in pBm N PVIEG /acZ, and pUChshyg
is cleaved once by X h ol. Southern blots were probed for plasmid specific sequences.
The hybridization pattern o f digested total cellular D N A showed a fragment with the
same length as that o f digested plasmid D N A . It seems that the transforming
124 VULSTEKE et al.

plasmids are maintained as stable extrachromosomal D N A . On Southern blots o f

non-cleaved genomic D N A , hybridized with the same probe, this extrachromosomal
D N A appears as a high molecular weight molecule.
For the moment, it is still unclear in which conformation these multicopy
plasmids are maintained in the transformed cells. Phenomena such as concatameriza-
tion or chromosomal linkage remain possibilities.

REFERENCES

[1] J A R V I S , D . L . , S U M M E R S , M . D . , G lyco syla tio n and secretion o f hum an tissue plas­

m in ogen activator in recom binant baculovirus-infected insect cells, M o l. Cell. B io l. 9
(1989) 2 1 4 -2 2 3 .
[2] J A R V IS , D .L ., F L E M I N G , J .G .W ., K O V A C S , G .R ., S U M M E R S , M .D .,
G U A R I N O , L . A . , U s e o f early ba culovirus prom oters fo r continuous exp ression and
efficient p rocessin g o f fore ign gene products in stably transform ed lepidopteran cells,
B iotech no logy 8 (1990) 9 5 0 -9 5 5 .
[3] H U Y B R E C H T S , R ., G U A R IN O , L .A ., V A N B R U S S E L , М ., V U L S T E K E , V .,
Nucleotide sequence o f a transactivating Bombyx mori nuclear p olyh e d rosis v iru s
immediate early gene, B ioch im . B io p h ys. A c ta 1129 (1992) 3 2 8 -3 3 0 .
[4] B L I S S A R D , G .W ., R O H R M A N N , G .F ., Location, seqiience, transcriptional
m apping, and temporal expression o f the g p 6 4 envelope glycoprotein gene o f the
Orgyia pseudotsugata m ulticapsid nuclear polyh edrosis v irus, V ir o lo g y 170 (1989)
5 3 7 -5 5 5 .
[5] G U A R I N O , L . A . , S U M M E R S , M . D . , Functional m apping o f a transactivating gene
required fo r expression o f a b aculovirus delayed-early gene, J. V iro l. 5 7 (1986)
5 6 3 -5 7 1 .
[6] G U A R I N O , L . A . , S U M M E R S , M . D . , N u cle otide sequence and tem poral expression
o f a b aculovirus regulatory gene, J. V iro l. 61 (198 7) 2 0 9 1 -2 0 9 9 .
[7] N I S S E N , M . S . , F R I E S E N , P .D ., M o le c u la r a n alysis o f the transcriptional regulatory
region o f an early ba culovirus gene, J. V iro l. 63 (1989) 4 9 3 -5 0 3 .
[8] G U A R IN O , L .A ., S U M M E R S , M .D ., Functional m apping of A cM N P V genes
required fo r late gene expression, J. V iro l. 6 2 (1988) 4 6 3 -4 7 1 .
 IAEA-SM-327/12

MOLECULAR CLONING
OF RESTRICTION ENDONUCLEASE
FRAGMENTS OF DNA ISOLATED
FROM THE NUCLEAR POLYHEDROSIS VIRUS
OF Heliothis armígera

T. A T T A T H O M , B. ISAN O N T
Department o f Entom ology

S. A T T A T H O M
Plant Genetic Engineering Unit

Kasetsart University,
Kamphaengsaen, Nakorn Pathom,
Thailand

Abstract
M O LECU LAR C L O N IN G O F R E S T R IC T IO N EN DO N U CLEASE FRAGM ENTS OF
DNA IS O L A T E D FROM THE NU CLEAR P O L Y H E D R O S IS V IR U S OF Heliothis
armígera.
Restriction fragm ents o f Heliothis armígera nuclear polyh edrosis v iru s ( H a N P V ) D N A
double digested w ith E c o R I and B a m H I were cloned in E. coli D H 5 a F ' u sin g the Bluescript
plasm id as a vector. T h e inserted D N A s w ere estimated to be 1 .5 -2 .0 k b in size. T h e D N A
probe w as constructed fro m a 1.5 k b D N A fragm ent labelled w ith d igoxigen in , i.e. no n­
radioactive D N A labelling. T h e probe hybridize d w ell w ith D N A isolated from positive
colonies, but not w ith the Bluescrip t plasm id. D o t blot hybridization data indicated that this
d igo xig e n in labelled D N A probe can detect viral D N A in the infected larva two days after,
inoculation. T h is probe can also be used to detect viral D N A isolated from a single N P V
infected H. armígera larva.

1. IN TR O D U C TIO N

The nuclear polyhedrosis virus (N PV ) is considered to be a potential bio­

insecticide for the control o f several lepidopterous insects. A m ong these pests, the
cotton bollw orm , Heliothis armígera, is the most destructive to crop plants because
o f its resistance to a wide range o f chemical insecticides.
The N PV belongs to subgroup A o f the Baculoviridae family o f insect viruses.
This virus has an envelope, rod shaped nucleocapsid, and a large, covalently closed,
circular, double stranded D N A [1 ,2 ]. The molecular weight o f the viral genome was
estimated at 5 0 -1 0 0 million dalton [3]. Physical mapping o f the restriction
endonuclease fragments o f some viruses in this group has been constructed [4 -7 ].

125
126 ATTATHOM et al.

In Thailand, an isolate o f H. armígera NPV (HaNPV) was isolated from

infected cotton bollworm s from the field and is now being developed as a bioinsecti­
cide [ 8]. Preliminary characterization o f HaNPV was done by determining the
specific restriction endonuclease cleavage pattern, arid its molecular, weight o f D N A
was estimated to be 65 X 10 6 dàlton [ 8] . M olecular characterization o f HaNPV is
needed to better understand the genome organization and its functions for further
improvement o f HaNPV as a bioinsecticide.
In this study, we report on the cloning o f the restriction endonuclease
fragments o f the HaNPV genom e and the construction o f a D N A probe for NPV
diagnosis in order to study the distribution and transmission o f this virus, in nature.
This is part o f the effort to understand the molecular biology and ecology o f the Thai
N PV isolate.

2. M A TE R IA L S A N D M ETHODS

2.1. Virus purification and viral DNA isolation

The nuclear polyhedrosis virus o f the cotton bollw orm , H. armígera, was iso­
lated from diseased larvae collected from .cotton fields in Utong District, Supanburi
Province, Thailand. The virus was propagated in host larvae that were mass reared
on an artificial diet. Virus purification and viral D N A isolation follow ed the method
described elsewhere [ 8]. Briefly, virus particles were extracted by dissolving the
polyhedra in 0 .1 M Na2C 0 3, pH 11.2, for 20 min. After removing undissolved poly-
hedra and debris by low speed centrifugation, the virus particles were pelleted by
centrifugation at 36 000 rev./m in fo r 2 h. Viral D N A was isolated from the purified
virus by phenol:chloroform extraction, precipitated by ethanol and resuspended in
TE buffer for further analysis.

2.2. Construction of recombinant plasmids

Bluescript (Stratagene) and H aN PV -D N A were each double digested to com ­

pletion with E coRI and BamHI. Enzyme treated Bluescript was.separated by 0 .7 %
agarose gel electrophoresis [9] and. eluted from the gel by using a Geneclean kit
(BIO, 101, La Jolla, C A , United States o f Am erica). The restriction fragments o f
viral D N A were collected by phenol-chloroform :isoam yl alcohol (24:1) extraction
and ethanol precipitation. The restriction fragments o f H aN PV -D N A and Bluescript
were then ligated at 14°C for 16 h. The ligation reaction mixture consisted o f 10 ¡xL
o f 100 ng digested N P V -D N A , 5 pL o f 100 ng digested Bluescript, 1 pL o f T4 D N A
ligase (Biolabs, Inc., U SA ), 2 /xL o f 10X.ligation buffer and 2 p L o f lOOmM A TP.
 IAEA-SM-327/12 127

2.3. Cloning of the NPV-DNA fragments

Recombinant plasmids were used to transform E. coli DH 5aF' by mixing

10 /*L o f ligated vector and 100 piL o f competent cell suspension. The mixture was
maintained on ice for 30 min, incubated at 4 2 °C for 50 s and immediately chilled
for 5 min. After supplementation with 900 ¡iL o f growth medium, the cells were
incubated for 1 h at 37°C and then plated on selective medium. Positive colonies
were selected and the recombinant plasmids were purified as outlined by Maniatis
et al. [10]. The DNA inserts were analysed by EcoRI and BamHI double digestion
o f the purified plasmids, followed by agarose gel electrophoresis.

2.4. Probe preparation

A fragment o f EcoRI and BamHI double digested HaNPV-DNA was used to

construct the probe. An eluted DNA fragment was labelled with
digoxigenin-ll-dUTP, i.e. non-radioactive DNA labelling (Boehringer Mannheim
GmbH, Germany). The reaction mixture o f 20 /¿L was incubated for 16 h at 37°С
and stopped with 2 /zL o f 0.2M EDTA, pH8.0. The probe was precipitated with
2.5 fiL o f 4M LiCl and 75 /¿L o f prechilled ethanol and kept overnight at -2 0 ° C .
After centrifugation, the digoxigenin labelled probe was vacuum dried and
resuspended in 50 /¿L o f ТЕ buffer, pH8.G.

2.5. Sample preparation

To prepare the DNA samples for hybridization tests, each heálthy and NPV
infected H. armígera larva was homogenized in 1 mL o f ddH20 and 1 mL o f 2X
lysis buffer (0.02M tris, pH7.5, Ó.3M NaCl and 4% SDS), centrifuged at
14 000 rev./min for 5 min and the supernatant collected. The DNA was precipitated
from the supernatant by cold absolute ethanol, vacuum dried and dissolved in
ddH20 for hybridization.

2.6. Dot blot hybridization

The DNA samples were dotted on a nitrocellulose membrane. After pre-

hybridizàtion at 68°C for 1 h in a sealed plastic bag, the samples were hybridized
overnight at 68° С in a shaker bath with a hybridized solution containing 5 ^L/mL
o f denatured DNA probe. Detection o f the target DNA with a digoxigenin labelled
probe was performed according to the standard experimental procedure (Boehringer
Mannheim GmbH).
128 ATTATHOM et al.

3. RESULTS

3.1. Cloning of the NPV-DNA fragments

O f the over tw o thousand colonies screened, 23 colonies were shown to be

positive to the ampicillin resistant test. The plasmids isolated from colonies 3 , 5 , 9 ,
1 2 , 15, 18 and 20 revealed distinct bands under agarose gel electrophoresis, with the
sizes ranging from 3.0 to 3.5 kb (Fig. 1 ). These bands migrated more slowly than
the 2 .9 kb Bluescript plasmid, indicating that they contained inserts.

3.2. Analysis of the NPV-DNA inserts

Double digestion o f the isolated plasmids with E coRI and BamHI restriction
endonucleases showed inserted D N A fragments o f similar size (Fig. 2). Colonies 3,
9 and 18 contained inserted D N A with the sizes o f 1 .3 -1 .5 kb. The digestion product
o fc o lo n ie s 5 , 12 and 20 showed D N A sizes o f 2 . 0 k b . The D N A fragment o f 1.3 kb
from colony 3 was used for probe construction.

Colonies
9 12 15 18 20

kb

2 3 .1 —-------- ►

2 .0 — -------- ►

FIG. 1. Agarose gel electrophoresis o f recombinant Bluescript plasmids contained HaNPV

DNA EcoRI and BamHI double digested fragments. The plasmids were isolated from seven
selected colonies o f ampicillin resistant E. coli DH 5a The Hindlllfragments o f lambda DNA
were used as the standards fo r size determination.
 IAEA-SM-327/12 129

Colonies
3 5 9 12 15 18 20

FIG. 2. Agarose gel electrophoresis o f recombinant Bluescript plasmids after double

digestion with EcoRI and BamHI restriction endonucleases. The plasmids were isolated from
seven selected colonies o f ampicillin resistant E . coli DH 5aF'.

Plasmids from colonies

Bluescript 3 5 9 10 15 18 20
I

« • •* • ® *

FIG. 3. Dot blot hybridization o f plasmids isolated from ampicillin resistant colonies to a
digoxigenin labelled NPV-cDNA probe. The DNA loaded per dot was 5 ng (row 1) and 10 ng
(row 2). The non-insert Bluescript showed negative reaction to the probe.

3.3. DNA probe and dot blot hybridization

The digoxigenin labelled D N A probe constructed from the 1.3 kb inserted

D N A from colony 3 reacted positively with all plasmids isolated from colonies 3,
5, 9, 15, 18 and 20. N o reaction was observed when the probe was hybridized with
non-insert Bluescript (Fig. 3). It is suggested that these D N A inserts may derive from
the same region o f the HaNPV genome.
130 ATTATHOM et al.

When the probe was dot blot hybridized with D N A isolated from healthy and
NPV infected H. armígera, positive reactions were observed only with diseased sam­
ples (Fig. 4). The viral D N A isolated from one, two and three infected larvae gave
the same result when hybridized with the probe. The intensity o f colour development
showed the same level o f hybridization, independent o f the amount o f D N A loaded
per dot. This result indicated that D N A isolated from á single N PV infected larva
was sufficient for dot blot hybridization with the 1.3 kb fragment N P V -D N A probe.

Healthy Diseased

FIG. 4. Dot blot hybridization o f DNA isolated from healthy and NPV infected (diseased)
H eliothis arm ígera larvae to a digoxigenin labelled NPV-cDNA probe. The NPV-DNA was
isolated from one, two or three larvae, as indicated in rows 1, 2 and 3, respectively.

Days after inoculation

Healthy 1 2 3 4 5 6

FIG. 5. Dot blot hybridization o f DNA isolated from healthy and NPV infected H eliothis
arm ígera larvae to a digoxigenin labelled NPV-DNA probe.
 IAEA-SM-327/12 131

This study also demonstrated that the concentration o f N P V -D N A increased in

infected larvae after inoculation (Fig. 5). W hen N P V -D N A was isolated from
H. armígera larvae 1, 2, 3, 4, 5 and 6 d after inoculation and hybridized with the
probe, colour development was observed with the increasing intensity. The results
indicated that N P V -D N A was first detected on the second day after inoculation. The
viral D N A concentration increased up to the sixth day after inoculation. By that time,
most o f the infected larvae had died because o f the viral infection.

4. DISCUSSION

The restriction endonuclease pattern has proved to be useful for the analysis
o f D N A because the restriction enzymes cleave D N A at specific nucleotide
sequences. Its usefulness for identifying strains and genom ic variants o f NPV has
been well recognized [11, 12]. Our previous w ork has shown the difference in the
restriction endonuclease pattern o f H aN PV -D N A after being digested with EcoRI,
H indlll, PstI, SacII, Hpal and Seal [ 8]. How ever, application o f this technique to
differentiate the strains o f NPV found in Thailand is very limited. This is due to the
fact that there is no rapid and reliable method for detecting viruses in the field so
that more strains can be collected and studied. Molecular cloning o f restriction
endonuclease fragments o f N P V -D N A will prove to be o f great value in the search
for and development o f m ore effective strains o f N PV in Thailand. W e selected the
E coRI and BamHI fragments because o f their constant 1 .3 -2 .0 kb products, which
are suitable for cloning. Although this w ork is at the initial stage o f constructing
genome mapping and full length sequence analysis, the probe derived from a cloned
D N A fragment has a high potential for application. Using this probe and the digox­
igenin labelling, technique, N PV can be detected even in a single infected larva. It
is hoped that m ore naturally occurring strains o f N P V . will be detected by this
method, thereby accomplishing the selection and development o f an effective strain
o f NPV for the control o f H. armígera.

ACKNOW LEDGEM ENTS

The authors wish to thank the Japan International Co-operation A gency,

T okyo, and the National Center o f Genetic Engineering and Biotechnology, Ministry
o f Science, Technology and Environment, Thailand, for their support.

REFERENCES

[1] S U M M E R S , M . D . , “ B a cu lo v iru se s (B ac u lo virid a e )” , T h e A tla s o f Plant and A n im a l

V ir u s e s ( M A R A M O R O S C H , K ., E d .), A cad e m ic Press, N e w Y o r k (1977) 3 -2 7 .
132 ATTATHOM et al.

[2] T I N S L E Y , T . W ., H A R A P , K . A . , ‘‘V iru se s o f invertebrates” , C om pre he nsive V ir o l­

og y ( F R A E N K E L - C O N R A T , H ., W A G N E R , R . R . , E d s), V o l. 12, Ple nu m Press,
N e w Y o r k (1977) 1-144.
[3] B I L I M O R I A , S . L ., “ T a x o n o m y and identification o f b a culoviruses” , T h e B io lo g y o f
Ba cu loviru se s, V o l. 1, B io lo g ic a l Properties and M o le c u la r B io lo g y ( G R A N A D O S ,
R . R . , F E D E R I C I , B . A . , E d s), C R C Press, B o c a Raton, F L (1986) 3 8 -5 4 .
[4] S U M M E R S , M . D . , S M I T H , G .E ., Restriction m aps o f five Autographa califomica
M N P V variants, Trichoplusia ni M N P V , and Galleria mellonella M N P V D N A s with
endonucleases Sm al, K p n l, B a m H I, Sa cI, X h o l and E c o R I, J. V iro l. 3 0 (1979)
8 2 8 -8 3 8 .
[5] V L A K , J .M ., M a p p in g o f B a m H I and S m a l D N A restriction sites on the genom e o f
the nuclear polyh edrosis v iru s o f the alfalfa looper, Autographa califomica, J. In v e r­
t e d . Pathol. 3 6 (1979) 4 0 9 -4 1 4 .
[6] L O H , L . C . , H A M M , J.J., H U A N G , E .S ., Spodoptera frugipeda nuclear p olyh edrosis
v iru s genom e: Ph ysical m aps fo r restriction endonucleases B a m H I and H in d III, J. V ir o l
3 8 3 (1981) 2 2 -3 1 .
[7] K N E E L , J .D ., S U M M E R S , M . D . , A p hysical m ap for the Heliothis zea S N P V
genom e, J. Gen. V iro l. 6 5 (1984) 4 4 5 -4 5 0 .
[8] A T T A T H O M , T., C H A E Y C H O M S R I , S., C H A I C H U C H O T , S., A T T A T H O M , S.,
C H I E M S O M B A T , P., C haracterization o f the nuclear polyh edrosis v iru s o f the cotton
b ollw orm , Heliothis armígera, Kasetsart J. (Nat. Sci. Suppl.) 2 2 (1988) 14-23.
[9] P E R B A L , B ., A Practical G u id e to M o le c u la r C lo n in g , 2nd edn, W ile y , N e w Y o r k
(1988).
[10] M A N I A T I S , T., F R I T S C H , E .F ., S A M B R O O K S , J., M o le c u la r C lo n in g : A L a b o ra ­
tory M a n u a l, C o ld S p rin g H a rb o r Lab oratory, C o ld S p rin g H arbor, N Y (1982).
[11] C O C H R A N , M . A . , C A R S T E N S , E .B ., E A T O N , B .T ., F A U L K N E R , P., M o le c u la r
clonin g and physical m apping o f restriction endonuclease fragm ents o f Autographa
califomica nuclear polyh ed rosis v iru s D N A , J. V iro l. 41 (1982) 9 4 0 -9 4 6 .
[12] BROW N, S .E ., M A R U N IA K , J.E., KN U DSO N , D .L ., B a c u lo v iru s (M N P V )
genom ic variants: Characterization o f Spodoptera exempta M N P V D N A s and com pari­
son w ith other Autographa califomica M N P V D N A s , J. Gen. V iro l. 6 6 (1985)
2 4 3 1 -2 4 4 1 .
 IAEA-SM-327/13

CONTROLE GENETIQUE DES INSECTES:

CARACTERISATION D’ELEMENTS
GENETIQUES MOBILES CHEZ LES MOUSTIQUES

C . M O U CH ES, J.C. S A L V A D O , N. BEN SA AD I,

S. K A R A M A , P. B LA N C H A R D
Laboratoire d ’ écologie moléculaire,
Université de Pau et des Pays de l ’ Adour,
Pau, France

A bstract-R ésu m é

G E N E T IC C O N T R O L O F IN S E C T S : C H A R A C T E R IZ A T IO N O F M O B IL E G E N E T IC
E L E M E N T S F R O M M O S Q U IT O G E N O M E S .
T h e Juan elements constitute a fam ily o f L I N E ’S retroposons w h ich are dispersed in the
genom e o f m any strains, if not all, o f the three m osquito species A. aegypti, C. pipiens and
C. tarsalis. A specific Juan sub fam ily is am plified and dispersed in the genom e o f each o f
these species. T h e y have been designated respectively as Juan A in A. aegypti, Juan С in
C. pipiens and Juan C t in C. tarsalis. T h e distribution o f Juan retroposons am ong m osquito
species does not reflect the ph y loge n y o f these species. Furtherm ore, the Juan retroposons
form hom ogeneous subfam ilies: full-length copies w h ich are reiterated in strains collected
from regions covering different continents are nearly identical. T h ese data are interpreted to
indicate that the Juan retroposons have spread recently in the m osquito genom es harbouring
them, perhaps upon horizontal transfer from other species. Juan A elements have been found
in two isolates o f A. albopictus and one isolate o f A. polynesiensis, whereas num erous other
isolates o f these two species are devoided o f any Juan-like sequence. T h e unfrequent presence
o f Juan A elements in som e strains o f A. albopictus and A. polynesiensis can be the result o f
an horizontal invasion, but m ore p robably from cross-hybridizations w h ich have been reported
som etim es to occur between the form er species and A. aegypti. N o p ro gen y is obtained w hen
A. albopictus m ales containing Juan A retroposons are crossed w ith females la cking these
elements, whereas the reciprocal cross is fertile. Su c h results suggest that the Juan retroposons
m ay be responsible for-incom patibilities between, strains.

CO NTRO LE G E N E T IQ U E DES IN S E C T E S : C A R A C T E R IS A T IO N D ’E L E M E N T S
G E N E T IQ U E S M O B IL E S C H E Z L E S M O U S T IQ U E S .
L e s éléments Juan constituent une fam ille de rétroposons de type L I N E , répétés et d is­
persés dans le génom e des m oustiques Aedes aegypti, Culex pipiens et Culex tarsalis. U n e
sous-fam ille Juan particulière, désignée respectivement Juan A , Juan C ou Juan C t, est
am plifiée et dispersée dans le génom e de chacune des souches de ces trois espèces analysées
à ce jour. L a distribution des éléments Juan parm i les différentes espèces de m oustiques ne
reflète pas la p hylogénie de celles-ci. D e plus, les copies com plètes des rétroposons Juan
réitérées dans des souches de différents continents form ent des sous-fam illes hom ogènes. C e s
observations indiquent que ces éléments ont envahi récemment les génom es q ui les hébergent,
peut-être à l ’issue d ’un transfert horizontal, de façon infectieuse. D e s éléments Juan A ont été

133
134 MOUCHES et al.

trouvés dans seulement deux isolats d’A. albopictus et.un isolât d’A. polynesiensis, alors que
de nom breux autres isolats de ces espèces sont dépou rvus de tels,éléments. Cette contam ina­
tion de rares isolats d ’A albopictus et d’A. polynesiensis par des éléments Juan A peut être
le résultat d ’une in va sio n horizontale, m ais p lus probablement de croisem ents fertiles peu
fréquents entre souches de ces deux espèces et A. aegypti. A u c u n e descendance n ’est obtenue
quand des m âles d ’une souche à'A. albopictus hébergeant les rétroposons Juan A sont croisés
avec des femelles dépourvues de ces éléments, a lors que le croisem ent réciproque est fertile.
C e s résultats prélim inaires suggèrent que les rétroposons Juan pourraient être responsables
d ’incom patibilités ou de dysgénésies entre souches de m oustiques.

1. IN TROD U CTION

La stimulation, par certains facteurs génétiques ou environnementaux, de

l ’ activité d ’ un élément génétique mobile peut augmenter la fréquence des mutations
et causer des variations alléliques. Ces composants instables du génome eucaryote
jouent donc très certainement un rôle majeur dans la plasticité de l ’ information géné­
tique et les remarquables facultés d ’ adaptation des insectes des populations
naturelles. Dès lors, la caractérisation de tels éléments est importante pour
comprendre leur rôle dans le polymorphisme et les processus d ’ adaptation des
espèces des populations naturelles qui les hébergent et leur origine évolutive.
Surtout, des éléments transposables peuvent permettre le développement de
méthodes plus efficaces de lutte génétique chez les insectes d ’ importance agrono­
mique ou médicale. La lutte génétique utilise plusieurs stratégies: introduction dans
la population d ’ insectes visés de mâles stériles (lâchers d ’ insectes stériles ou Sterile
Insect Technique), d ’ une espèce proche, ou contenant des gènes défavorables pour
l ’ espèce (allèles de stérilités, d ’ incompatibilités et de dysgénésies ou modifiant le
rapport de sexe). Parce qu’ elle a une spécificité absolue, ne sélectionne théorique­
ment pas d ’ individus résistants et permet en principe d ’ éradiquer une population sans
pollution, la lutte génétique devrait occuper une place privilégiée dans les pro­
grammes de lutte intégrée. L ’ utilisation des transposons peut permettre de produire,
à des coûts raisonnables, des insectes mâles stériles ou porteurs d ’ informations
létales compétitifs dans les milieux naturels. D ’ abord, des transposons peuvent
constituer des vecteurs de gènes [ 1 ] pour m odifier par transgénèse les chromosomes
des insectes afin de disposer de méthodes simples de sélection des mâles (sexage par
un gène de résistance à l ’ alcool sur le chrom osom e Y ) ou de transférer des gènes
de stérilisation, modifiant le rapport de sexe ou létaux pour la descendance (gènes
de thermosensibilité, de dérive méiotique, de dysgénésies, ...) ou les parasites
véhiculés (gène de diphénoloxydase nocive pour les plasmodiums du paludisme). En
outre, beaucoup d ’entre eux appartiennent à la catégorie de séquences d ’ A D N dites
‘ égoïstes’ , capables d ’ envahir le génome grâce à leur propre machinerie enzy­
matique, responsable de leur amplification et leur dispersion. Il serait donc
 IAEA-SM-327/13 135

particulièrement intéressant de leur associer des informations que l ’ on veut répandre

dans les populations naturelles. Enfin, certains transposons peuvent être des agents
d ’ incompatibilités ou de dysgénésies hybrides entre souches d ’ une espèce [1, 2].
Dans la mesure où des espèces particulièrement nuisibles pourraient être contrôlées,
voire éradiquées par des procédés faisant appel aux techniques moléculaires, le
bénéfice pour l ’environnement serait considérable, malgré les problèmes éthiques
soulevés.
Enfin, certains éléments mobiles peuvent constituer des outils remarquables
pour identifier par mutagénèse par transposons les gènes impliques chez les insectes
dans les phénomènes com plexes d ’ adaptation, d ’ apprentissage, comportementaux ou
de vection d ’ agents pathogènes. Ainsi, la microinjection du transposon P dans la
lignée germinale de Drosophila melanogaster a permis de muter et d ’ étiqueter, puis
d ’ isoler, des gènes difficiles à étudier par les approches biologiques traditionnelles.

2. R ESU LTATS ET DISCUSSION

2.1. Stratégies d’identification d’éléments génétiques mobiles de moustiques

Plusieurs approches sont possibles pour identifier des éléments génétiques

mobiles. La plus simple consiste à sonder le génome à l ’ aide d ’ éléments déjà
identifiés chez d ’ autres espèces, en espérant des hybridations grâce aux analogies
de séquences qui se présentent assez fréquemment. Mais cette recherche n’ est
envisageable en principe qu’ entre groupes phylogénétiquement voisins et ne permet
pas forcément d ’ accéder aux éléments les plus intéressants. L ’analyse de mutants
phénotypiques est probablement la stratégie la plus rigoureuse pour détecter des
séquences m obiles; en effet, plus de la moitié des mutations dans certains gènes de
D . melanogaster sont dues à l ’ insertion dans ces gènes, ou dans leurs régions régula­
trices, de transposons. Des phénomènes de dysgénésies hybrides, qui ont été à
l ’ origine de la découverte des transposons chez les invertébrés, peuvent aussi révéler
leur présence; ils se manifestent par des stérilités et mutations lors de croisements
entre femelles et mâles d ’ une même espèce, en particulier d ’ origines géographiques
différentes [1, 2]. Les éléments mobiles constituent également des ensembles d ’ A D N
moyennement ou très répétés et on peut les rechercher dans ce type de séquence.
Enfin, des m olécules d ’ acides nucléiques, A D N ou A R N , extrachromosomiques et
dés particules pseudovirales susceptibles de représenter des intermédiaires de trans­
position spécifiques de diverses séquences mobiles peuvent s’ accumuler à certains
stades du développement de l ’ insecte, en particulier ceux correspondant à des étapes
précoces, ou dans des cellules en culture. Nous avons mis en route ces diverses
stratégies pour identifier et isoler des éléments génétiques mobiles spécifiques des
moustiques Aedes et Culex. Nous avons en particulier analysé des mutations dans
136 MOUCHES et al.

lesquelles des transposons pourraient être impliqués, mutations de résistances aux

insecticides ou d ’ autres toxiques ou pour la coloration de l ’ œil.
Chez une population de moustiques C. pipiens quinquefasciatus de Californie,
la résistance aux insecticides organophosphorés est due à l ’amplification du gène de
l ’ estérase de détoxication B1 [3, 4]. Cette amplification résulte de remaniements
chromosomiques auxquels sont associés des éléments génétiques mobiles: des
séquences appartenant à au moins trois familles différentes d ’éléments génétiques
mobiles sont coamplifiées, dans l ’ amplicon, avec le gène de l ’estérase B1 [5, 6,
M O U CH ES, résultats non publiés].
La première famille de transposons ayant été ainsi identifiée est celle des élé­
ments désignés Juan C [5, 6]. Une copie tronquée de cette famille est coamplifiée
avec le gène de l ’ estérase B1 et de nombreuses autres séquences de ce type sont dis­
persées dans tout le génome du moustique. Le séquençage de ces copies tronquées
de la famille Juan C a révélé qu’il s’agissait de rétroposons de type LINE («long
interspersed repetitive elements») [6], c ’ est à dire de transposons dont l ’ amplification
et la dispersion dans le génome résultent d ’ un processus de transcription réverse d ’ un
intermédiaire A R N [7]. En outre, l ’ hybridation moléculaire nous a indiqué que des
séquences homologues à Juan C existent chez d ’ autres populations ou espèces de
moustiques: C. pipiens, C. tarsalis, Aedes aegypd, etc.
Les deux autres familles d ’ éléments génétiques mobiles qui ont été identifiées
dans l ’ amplicon du gène d ’ estérase B1 sont la famille des transposons CEI d ’ une part
et celle des éléments Pau d ’ autre part, ces derniers contenant en outre une séquence
de type minisatellite. Les copies de ces transposons coamplifiées dans l ’ amplicon de
l ’ estérase B1 sont des copies semble-t-il incomplètes et qui ne sont probablement pas
responsables du processus d ’ amplification lui-même; des représentants complets de
ces séquences mobiles sont en cours d ’ isolement.

2.2. Caractérisation de la famille d’éléments génétiques mobiles Juan

chez les moustiques A. aegypti et C. pipiens

Des copies complètes. (4487 pb) du rétroposon Juan C des moustiques

C. pipiens ont été clonées et sëquencées (accès à Genbank: M 91082) [8]. Des
fragments de ce rétroposon ont été utilisés com m e sondes pour purifier, par hybrida­
tion hétérologue, des copies complètes d ’une sous-famille de rétroposons amplifiés
et dispersés dans le génome des moustiques Aedes et désignés Juan A (4727 pb)
(accès à Genbank: M 95171) [9]. Le degré d ’ amplification dans le génome des
éléments Juan complets est compris entre 200 et 500 copies. Ces valeurs sont
voisines de celles de certaines séquences LINE de mammifères, mais supérieures à
celles observées pour les rétroposons complets de Drosophila melanogaster qui sont
présents seulement à moins de 50 copies par génome.
Les rétroposons Juan A et Juan C contiennent deux phases de lecture ouvertes
(Open reading frame ou ORF) parfaites, codant potentiellement la première une
 0,5 kpb
Juan A (4727 p b*) i______________ i

CYS

ш т т Ш

E X

IA E A -SM -3 2 7 / 1 3
Juan С (4487 pb)
С YS

OBF1

X Е

* pb: paire de base; 1 kb d ’A D N double b rin a une m a sse m oléculaire de 6,6 X 105.

FIG. 1. Structure et organisation génétique des rétroposons Juan A d ’A . aegypti et Juan C d e C . pipiens. L es dom aines hautem ent con servés sont
indiqués p a r CYS (domaine riche en cystéine typique des p rotéin es d e liaison aux acides n ucléiques) dans la p h a se d e lectu re ou verte O R F 1
et R T (domaines typiques des transcriptases réverses) dans la ph ase de lectu re ou verte O R F 2. A n indique une séq u en ce rich e en adénosine à
l ’extrém ité 3 ’ d es éléments. L es sites de coupure par d es enzym es d e restriction sont indiqués p a r: B, Bam H I; E, E coR I; X , X bal.

ы
-J
138 MOUCHES et al.

protéine de liaison aux acides nucléiques (nucleic acid binding ou «gag» protein), la
seconde une transcriptase réverse (Fig. 1). Ils présentent entre eux de fortes homolo­
gies au niveau de leurs séquences nucléotidiques et de celles des polypeptides poten­
tiellement codés (37% d’homologies pour la protéine «gag», 67% d’homologies au
niveau dé la transcriptase reverse).
Les deux familles ont la même organisation génétique que les éléments Jockey,
F, G et I de D. m e l a n o g a s t e r [10, 11, 12]. En outre, les séquences nucléotidiques
des rétroposons Juan A et Juan C présentent des régions d’homologies avec les
éléments Jockey, F et G, mais non pas avec les rétroposons I de D . m e l a n o g a s t e r
et D . t e i s s i e r i , ou Tl à ' A n o p h e l e s g a m b i a e [13]. Ces résultats indiquent que les
rétroposons Juan et certains rétroposons de drosophiles dérivent d’un même élément
ancestral.

2.3. Ecologie des rétroposons Juan chez les moustiques Culex et Aedes

Des copies des rétroposons Juan C et Juan A, ou d’éléments voisins, ont été
recherchées dans des souches ou clés individus issus des populations naturelles, en
analysant les ADN génomiques extraits de ces insectes par hybridation avec les
éléments clonés, détermination ou polymorphisme de longueur des fragments de
restriction hybridés et par amplification génique in vitro par la polymérase (poly­
merase chain reaction ou PCR), en utilisant comme amorces des oligonucléotides
spécifiques de ces éléments (Tableau I). '
Approximativement le mêmë nombre de copies du rétroposon Juan C sont
amplifiées dans le génome les souches de C . p i p i e n s de différents continents. Une
sous-famille d’éléments très proche de celle des rétroposons Juan C, désignée
Juan Ct, est amplifiée chez C . t a r s a l i s . Ces éléments Juan Ct présentent un poly­
morphisme de restriction différent de celui des rétroposons Juan C, montrant que
chacune des deux sous-familles a été amplifiée à partir d’un variant d’un ancêtre
commun.
Des rétroposons Juan A sont amplifiés dans toutes les souches de moustiques
A . a e g y p t i . Des éléments identiques ont également été trouvés dans deux isolats de
l’espèce A . a l b o p i c t u s et un isolât d ’A . p o l y n e s i e n s i s , mais dé nombreux autres iso­
lats de ces deux espèces sont dépourvus de tout élément de type Juan, quelle que soit
la stringence des conditions d’hybridations utilisées.

2.4. Histoire évolutive des rétroposons Juan A et Juan C

Toutes les souches de moustiques appartenant aux espèces C. p i p i e n s ,

et A . a e g y p t i que nous avons analysées jusqu’à présent hébergent, sous
C. ta r s a lis
une forme répétée et dispersée, une sous-famille de rétroposons Juan, respectivement
Juan C, Juan Ct et Juan A. Cependant, la distribution des rétroposons Juan ne reflète
pas la phylogénie des espèces de moustiques. En effet, ces éléments semblent être
 IAEA-SM-327/13 139

TABLEAU I. DISTRIBUTION DES RETROPOSONS JUAN

CHEZ DES MOUSTIQUES D’ESPECES ET D’ORIGINES GEOGRAPHIQUES
DIFFERENTES

Espèce Isolât Origine Rétroposon

C. pipiens Say Tem-R Californie Juan C

C. pipiens Say S-Lab Californie Juan C
C. pipiens Say S54 Montpellier Juan C
C. pipiens Say MSE Montpellier Juan C
C. pipiens Say ' Bleuet Montpellier ' Juan C
C. pipiens Say Lahore Pakistan Juan C
C. pipiens Say - Selax Californie Juan C
C. pipiens Say Bouaké Côte d’ivoire . Juan C

C. tarsalis NR Californie Juan Ct

C. tarsalis Sus Californie Juan Ct '

A. aegypti L. Bobo-Dioulasso Afrique Juan A

A. aegypti L. Meya Pacifique Juan A
A. aegypti L. MD Trinidad Juan A

A. albopictus Skuse Oahu 1971 Hawaii — „■

A. albopictus Skuse . Honolulu 1986-01 Hawaii - (a)

A. albopictus Skuse Honolulu 1986-97 Hawaii . Juan A
A. albopictus Skuse Poona 1988-01 Inde -(b )
A. albopictus Skuse Poona 1988-109 Inde Juan A

A. polynesiensis Marks Pap 1987-01 Tahiti — (c)

A. polynesiensis Marks Pap 1987-132 Tahiti Juan A

(a), (b), (c): cent dix-sept autres isolats d 'A:albopictus prélevés en 1986 a Hawaii, cent vingt-
deux autres collectés en 1988 à Poona et cent quatre-vingt-sept autres, isolats d 'A. poly­
nesiensis prélevés en 1987 à Tahiti ne contenaient aucun rétroposon Juan.

absents chez la plupart des isolats d’autres espèces d ' A e d e s , pourtant plus proches
d ’A . a e g y p t i que des C u le x . Comme il est peu vraisemblable qu’un élément amplifié
et dispersé à plusieurs centaines de copies dans un génome puisse avoir été éliminé
totalement sans laisser de traces, les rétroposons Juan ont donc envahi le génome des
espèces qui les hébergent après la divergence évolutive des différentes espèces de
moustiques. • ■
En outre, les copies complètes d’une sous-famille Juan amplifiées dans une
espèce sont remarquablement homogènes, non seulement .au sein d’une souche
140 MOUCHES et al.

donnée, mais aussi entre souches d’origines géographiques différentes. Ce résultat

est surprenant pour une famille d’éléments dont l’amplification et la dispersion
procèdent via des étapes de transcription en ARN, puis de réverse-transcription en
ADN. En effet, et contrairement aux ADN polymérases, les ARN polymérases sont
dépourvues d’ activités éditrices correctrices des erreurs survenant lors de la poly­
mérisation de leurs produits. Une replication via de tels processus devrait donc
conduire à des copies très hétérogènes des éléments Juan.
La distribution particulière des rétroposons Juan et l’homogénéité de chacune
des sous-familles amplifiée au sein d’une espèce donnée impliquent que: ( 1) l’ampli­
fication d’une sous-famille Juan s’est faite chez une population unique d’une espèce
donnée à partir d’une seule copie du rétroposon ou seulement de quelques
exemplaires tous identiques de cet élément; (2 ) cette amplification a conféré un avan­
tage sélectif à l’hôte, puisque les individus dans lesquels les rétroposons Juan ont été
amplifiés ont été capables d’envahir une grande partie de l’aire géographique des
espèces en cause; (3) l’amplification de ces éléments s’est produite très récemment
dans les populations de moustiques. Ces données nous conduisent à émettre
l’hypothèse que les rétroposons Juan ont envahi très récemment le génome des
souches ou espèces de moustiques qui les hébergent, à l’issue d’un transfert hori­
zontal entre espèces, c ’est-à-dire que des copies du rétroposon Juan ont une capacité
«infectieuse» à se répandre.
Des copies complètes du rétroposon Juan A semblent avoir envahi très récem­
ment des populations prélevées récemment dans la nature des espèces A . a l b o p i c t u s
et A . p o l y n e s i e n s i s , alors que la plupart des populations de ces espèces, et en
particulier les plus anciennes, en sont dépourvues (Tableau I). Cette invasion peut
résulter d’un processus de transfert horizontal comme évoqué ci-dessus; plus
probablement, il est le résultat de croisements entre ces deux espèces et A . a e g y p t i .
En effet, si des croisements entre A . a e g y p t i , A . a l b o p i c t u s et A . p o l y n e s i e n s i s sont
généralement considérés comme infertiles, plusieurs auteurs ont montré que des
croisements entre individus de certaines souches de ces trois espèces pouvaient être
interfertiles [14,15, 16, 17].

2.5. Etude de la transposition des éléments Juan et interactions avec

le génome de l’hôte: ces rétroposons sont-ils responsables
d’incompatibilités entre populations de moustiques?

La transposition des éléments mobiles est habituellement réprimée chez les

hôtes qui les hébergent. Les transposons peuvent être mobilisés à l’issu de différents
stress subis par l’hôte, par exemples des chocs thermiques ou des croisements entre
souches «éloignées» [18, 19]. Dans ce dernier cas peut survenir le phénomène de
dysgénésie hybride qui, avec trois transposons différents de drosophiles, P, I et
Hobo, survient en particulier lorsqu’un de ces éléments est introduit, par croisement
sexuel ou par injection, dans de jeunes embryons issus de femelles elles-mêmes
 IAEA-SM-327/13 141

dépourvues de la séquence considérée. La ressemblance de l’organisation génétique

des rétroposons Juan avec celle des LINE de D . m e l a n o g a s t e r suggère que, comme
le facteur I, ils pourraient être responsables de phénomènes de dysgénésie hybride
entre souches géographiques différentes de moustiques.
Nous avons croisé des individus des souches d’Æ a l b o p i c t u s contenant le
rétroposon Juan A avec l’une des nombreuses souches qui sont dépourvues de ce
transposon. Les croisements sont féconds seulement de façon unidirectionnelle: les
mâles sans rétroposons Juan A sont compatibles avec les femelles hébergeant
l’élément, alors que le croisement réciproque est infertile. Ce résultat est du même
ordre que celui obtenu dans les cas de dysgénésie hybride dus à des transposons chez
les drosophiles. Il suggère que les rétroposons Juan sont facteurs d’incompatibilités
entre souches d’une même espèce de moustiques. Cependant, de telles incompati­
bilités unidirectionnelles entre souches peuvent. aussi résulter de la présence de
bactéries intraovariennes de type rickettsies ou W o l b a c h i a . Des expériences sont en
cours pour introduire directement, par transgénèse dans des œufs embryonnés, des
copies complètes du rétroposon Juan A dans les.souches d ’A . a l b o p i c t u s qui en sont
dépourvues.
Les souches de C. p i p i e n s , C . t a r s a l i s et A . a e g y p t i que nous avons analysées
jusqu’à présent ont été prélevées au cours de ces vingt-cinq dernières années dans
des populations naturelles géographiquement différentes. La recherche des rétro­
posons Juan dans des souches géographiquement isolées et des collections anciennes
a été entreprise dans le but d’identifier des individus dépourvus de ces éléments, ce
qui nous permettrait d’évaluer l’impact des rétroposons Juan sur le génome de ces
trois espèces.

3. CONCLUSIONS

Depuis leur découverte, les éléments génétiques mobiles ont été étudiés chez
un nombre limité d’espèces. Dès lors, leur histoire évolutive et les stratégies qui leur
permettent d’évoluer apparemment de concert avec le génome de leurs hôtes respec­
tifs sont peu connues. La caractérisation de nouveaux éléments à partir de
moustiques permettra de mieux connaître l’évolution de ces séquences, de
comprendre les mécanismes de leur interaction avec le génome hôte et d’apprécier
leur importance dans les mutations adaptatives. Il sera en particulier intéressant de
vérifier l’éventualité que ces éléments puissent être «infectieux», c’est-à-dire
transmis de façon horizontale, entre espèces différentes. Chez les mammifères et la
drosophile, certains éléments mobiles semblent avoir été fréquemment perdus et
gagnés par les diverses espèces, et il a été suggéré (et illustré chez la drosophile) que
certains d’entre eux ont envahi une espèce donnée à l’issue d’un transfert horizontal
à partir d’une autre espèce.
142 MOUCHES et al.

L’étude plus approfondie des éléments Juan nous permettra également de

déterminer s’ils peuvent être utilisés comme transposons pour muter des fonctions
importantes des insectes, servir de vecteurs de gènes ou comme agents
d’incompatibilités entre souches de'moustiques.

REFERENCES

[1] RUBIN, G .М ., SPRADLING, A .C ., Genetic transformation of Drosophila with trans­

posable element vectors, Science 218 (1982) 348-353.
[2] LOUIS, C ., YANNOPÓULOS, G ., The transposable elements involved in hybrid dys­
genesis in Drosophila melanogaster, Oxford Surv. Eukar. Genes 5 (1988) 205-250.
[3] MOUCHES, С ., PASTEUR, N ., BERGE, J.B., HYRIEN, О ., RAYMOND, М .,
ROBERT DE SAINT VINCENT, B., DE SYLVESTRI, М ., GEORGHIOU, G.P.,
Amplification of an esterase gene is responsible for insecticide resistance in a California
Culex mosquito, Science 233 (1986) 778-780.
[4] MOUCHES, С ., MAGNIN, М ., BERGE, J.B., DE SYLVESTRI, М .,
BEYSSAT, V ., PASTEUR, N ., GEORGHIOU, G.P., Overproduction of detoxifying
esterases in organophosphate resistant Culex mosquitoes and their presence in other
insects, Proc. Natl. Acad. Sci. USA 84 (1987) 2113-2116.
[5] MOUCHES, С ., PAUPLIN, Y ., AGARWAL, М ., LEMIEUX, L., HERZOG, М .,
ABADON, М ., BEYSSAT-ARNAOUTY, V ., HYRIEN, O ., ROBERT DE SAINT
VINCENT, B., GEORGHIOU, G.P., PASTEUR, N ., Characterization of amplifica­
tion core and esterase B1 gene responsible for insecticide resistance in Culex, Proc.
Natl. Acad. Sci. USA 87 (1990) 2574-2578.
[6 ] MOUCHES, С ., AGARWAL, М ., CAMPBELL, К., LEMIEUX, L., ABADON, М .,
Sequence of a truncated LINE-like rétroposon dispersed in the genome of Culex
mosquitoes, Gene 106 (1991) 279-280.
[7] DI NOCERA, P.P., SAKAKI, Y ., LINEs: a superfamily of retrotransposable ubiqui­
tous DNA elements, Trends Genet. 6 (1990) 29-30.
[8 ] AGARWAL, М ., BENSAADI, N ., SALVADO, J.C., CAMPBELL, K.,
MOUCHES, С ., Insect Physiol. Molecul. Biol., soumis (1992).
[9] MOUCHES, С ., BENSAADI, N., SALVADO, J.C., Characterization of a LINE
rétroposon dispersed in the genome of three non-sibling Aedes mosquito species, Gene,
in press (1992).
[10] PRIIMÀGI, A .F ., MIZROKHI, L. V ., ILYIN, Y .V ., The Drosophila mobile element
jockey belongs to LINEs and contains coding sequences homologous to some retroviral
proteins, Gene 70 (1988) 253-262.
[11] BUCHETON, A ., PARO, R., SANG, H .M ., PELISSON, A ., FINNEGAN, D.J.,
The molecular basis of I-R hybrid dysgenesis in Drosophila melanogaster. identifica­
tion, cloning; and properties of the I factor, Cell 38 (1984) 153-163.
[12] FAWCETT, D.H ., LISTER, C .K ., KELLET; E., FINNEGAN, D.J., Transposable
elements controlling I-R hybrid dysgenesis in D. melanogaster are similar to mam­
malian LINEs, Cell 47 (1986) 1007-1015.
 IAEA-SM-327/13 143

[13] BESANSKY, N.J., A retrotransposable element from the mosquito Anopheles

gambiae, Mol. Cell. Biol. 10 (1990) 863-871.
[14] GUBLER, D.J., Induced sterility in Aedes (Stegomyia) polynesiensis marks by cross­
insemination with Aedes (Stegomyia) albopictus Skuse., J. Med. Entomol. 7(1) (1970)
65-70.
[15] LEAHY, M .G ., CRAIG, G.B., Barriers to hybridization between Aedes aegypti and
Aedes albopictus (Diptera: Culicidae), Evolution 21 (1967) 41-58.
[16] ROZEBOOM, L.E., GILFORD, B .N ., The genetic relationships of Aedes pseudo-
scutellaris Theobald and A. polynesiensis Marks (Diptera: Culicidae), Amer. J. Hyg.
60 (1954) 117-134.
[17] WOODHILL, A .R ., Experimental crossing of Aedes (Stegomyia) pseudoscutellaris
Theobald and Aedes (Stegomyia) polynesiensis Marks (Diptera: Culicidae), Proc. Linn.
Soc. NSW 79 (1954) 19-20.
[18] BUCHETON, A ., I transposable elements and I-R hybrid dysgenesis in Drosophila,
Trends Genet. 6 (1990) 16-21.
[19] RIO, D .C ., Regulation of Drosophila P element transposition, Trends Genet. 7 (1991)
282-287.
 IAEA-SM-327/14

USE OF THE BIOLISTIC TECHNIQUE

FOR GENE TRANSFER TO MOSQUITO EMBRYOS

E. MIALHE* +> A. LAUGHINGHOUSE*, L.H. MILLER*

* Laboratory of Malaria Research,
National Institute of Allergy and Infectious Diseases,
National Institutes of Health,
Bethesda, Maryland,
United States of America

+ Unité de recherches en pathologie, immunologie

et génétique des invertébrés marins,
Institut français de recherches poiir l’exploitation
de la mer,
La Tremblade,
France

Abstract
USE OF THE BIOLISTIC TECHNIQUE FOR GENE TRANSFER TO MOSQUITO
EMBRYOS.
Gene transformation for the control of mosquito transmitted diseases is currently
limited by the lack of an efficient method for the transfection and integration of exogenous
DNA. The biolistic technique, which uses high velocity DNA coated microprojectiles, has
been developed to introduce DNA into embryos of Anopheles gambiae. The biolistic
parameters have been characterized and optimized on the basis of transient expression of the
luciferase reporter gene placed under the control of the heat shock protein 70 promoter of
Drosophila. High luciferase activities were observed for biolistic DNA introduction per­
formed in the early embryonic stages. Because of the very large number of embryos which
can easily be transfected biolistically, and because DNA is probably delivered directly into
the nuclei, the biolistic method could be useful for obtaining insect genetic transformation,
despite the low frequency of exogenous DNA integration. Moreover, this technique has
already been proved suitable for embryo transfection in other invertebrates, such as
crustaceans.

The introduction of genes into mosquitoes that would make them refractory to
pathogens is a possible strategy for controlling vector borne diseases [1-5].
However, techniques are not available to efficiently introduce genes into mosquitoes,
as DNA injected into embryos rarely integrates into the host genome. This barrier
was overcome in D r o s o p h i l a by identifying the P transposable element that effi­
ciently introduce genes into the germ line. Because of the lack of identified high

145
146 MIALHE et al.

efficiency integration elements in mosqúitoés, alternative approaches need to be

developed. To compensate for the low frequency of integration, we have explored
use of the biolistic technique [6 ] to introduce DNA into a large number of A n o p h e l e s
g a m b i a e embryos. This technique, which uses high velocity DNA coated micro­
projectiles to deliver DNA directly into the nuclei, has been successfully applied to
introduce DNA into a large number of fish eggs [7] and dechorionated D r o s o p h i l a
embryos [8 ].
Our experiments were performed with the Biolistic PDS-1000/He particle
delivery system (Bio-Rad). The experimental parameters were optimized for three
different embryonic stages:

(1) 0 to 2 0 -m in -o ld e m b r y o s : At this stage, the chorion is soft, which may facilitate

penetration by the microprojectile. Moreover, because pronuclei fusion is in
progress, delivery of DNA coated particles at this time may lead to early
integration of exogenous DNA. Consequently, some fully transgenic
mosquitoes could be obtained as soon as the G0 generation.
(2) 6 0 t o 8 0 - m i n - o l d e m b r y o s : At this stage, the chorion has not fully hardened.
Numerous syncytial nuclei are scattered inside the ooplasm and, consequently,
there is a higher probability of microprojectiles penetrating the nuclei.
Moreover, some nuclei may have migrated to the germinal pole. Thus, bom­
bardment of such embryos may lead to partially transformed G0 embryos,
fully transgenics being eventually obtained only at the G[ generation.
(3) 1 5 0 t o 1 8 0 - m i n - o l d e m b r y o s : At this stage, hardening of the chorion and vitel­
line membrane has been completed. The nuclei are numerous and distributed
homogeneously along the embryo surface. The pole cells begin to form.
The biolistic parameters were optimized on the basis of the transient expression
of a luciferase gene placed under the control of the D r o s o p h i l a heat shock protein
( h s p ) promoter. This reporter gene was selected because of the ease of sample treat­
ment of luminometry analyses, which are quantitative and sensitive for samples con­
taining as few as a thousand embryos. For all the preliminary comparative
experiments carried out, the transient expression of luciferase was analysed in
24-h-old embryos.
For each biolistic operation, the embryos (3000-5000) were placed immedi­
ately in a monolayer inside a circle ( 2 cm diameter) corresponding to the micro­
projectile beam size. After bombardment, the embryos were maintained in an
antibiotic solution to prevent the development of microorganisms transformed by the
DNA construct, which would have a luciferase activity.
Relatively low levels of luciferase transient expression were observed with 0
to 20-min-old embryos, possibly due to the small number of nuclei. Higher levels
of luciferase activity were observed in the 60 to 80 and 150 to 180-min-old embryos.
The biolistic parameters were then progressively optimized, coating of particles with
DNA and their subsequent loading on to the macroprojectile being the most variable
 IAEA-SM-327/14 147

operations. The best ratio of DNA/particles was about 10 ¿ig/l mg. The higher
ratios resulted in the formation of particle aggregates. Using 900 psi rupture discs,
the embryo survival rates were similar to those of the controls. The transient expres­
sion of luciferase was also analysed in 48-h-old embryos and in 1-d-old larvae. The
highest luciferase activities were found in 24-h-old embryos. Both circular and linear
DNA induced transient expression efficiently. The biolistic method led to transient
expression in about one hundred experiments.
Experiments are now in progress, using the biolistic conditions developed from
transient expression, to introduce DNA with a selectable marker into large numbers
of embryos in order to select for stable integration. The biolistic technique has
already been proved to be suitable for embryo transfection in other invertebrates,
such as crustaceans.

REFERENCES

[1] MILLER, L.H ., et al., Stable integration and expression of a bacterial gene in the
mosquito Anopheles gambiae, Science 237 (1987) 779-781.
[2] EGGLESTON, P., The control of insect-borne disease through recombinant DNA
technology, Heredity 6 6 (1991) 161-172.
[3] HANDLER, A .M ., O ’BROCHTA, D ., Prospects for gene transformation in insects,
Annu. Rev. Entomol. 36 (1991) 159-183.
[4] McGRANE, V ., CARLSON, J.O., MILLER, B.R., BEATY, B.J., Microinjection of
DNA into Aedes triseriatus ova and detection of integration, Am. J. Trop. Med. Hyg.
39 (1988) 502-510.
[5] MORRIS, A .C ., EGGLESTON, P., CRAMPTON, J.M., Genetic transformation of
the mosquito Aedes aegypti by microinjection of DNA, Med. Vet. Entomol. 3 (1989)
1-7.
[6 ] SANFORD, J.C., SMITH, F.D ., RUSSEL, J.A., Optimizing the biolistic process for
different biological applications, Methods Enzymol. (in press).
[7] ZELENIN, A .V ., et al., The delivery of foreign genes into fertilized eggs using high-
velocity microprojectiles, FEBS Lett. 287 (1991) 118-120.
[8 ] BALDARELLI, R .M ., LENGYEL, J.A., Transient expression of DNA after ballistic
introduction into Drosophila embryos, Nucleic Acids Res. 18 (1990) 5903-5904.
 IAEA-SM-327/15

POLYMERASE CHAIN REACTION AMPLIFICATION

OF RAPD MARKERS TO DISTINGUISH
POPULATIONS OF THE MEDITERRANEAN
FRUIT FLY, Ceratitis capitata

D. HAYMER
Department of Genetics and Molecular Biology,
University of Hawaii

D. McINNIS
Agricultural Research Service,
United States Department of Agriculture

Honolulu, Hawaii,
United States of America

Abstract
POLYMERASE CHAIN REACTION AMPLIFICATION OF RAPD MARKERS TO
DISTINGUISH POPULATIONS OF THE MEDITERRANEAN FRUIT FLY,
Ceratitis capitata.
Polymerase chain reaction amplification of RAPD markers was used to document the
existence of molecular markers that can distinguish different populations of the Mediterranean
fruit fly, Ceratitis capitata. The populations used include strains which differ primarily in
terms of geographical origin, but, in addition, strains that have been in long tèrm laboratory
culture can also be distinguished from wild caught strains. The molecular markers identified
can be used for a variety of purposes, including documentation of new forms of genetic varia­
tion, identification of strain specific molecular markers and establishment of linkage relation­
ships between anonymous DNA. sequences and desirable genes or traits of interest.

1. INTRODUCTION

Many of the techniques recently developed in molecular genetics and biology

hold great promise for answering questions of importance in research on insect
pests. We have initiated several molecular genetic studies of the Mediterranean fruit
fly (medfly), C e r a t i t i s c a p i t a t a , as well as other Tephritid pest species found here
in Hawaii [1-3]. In this paper, we describe our work showing that in the médfly,
so called randomly amplified polymorphic DNA (RAPD) markers can be amplified
using the polymerase chain reaction (PCR), and that these markers can be used to
assess genetic variation within arid between populations of this pest species. The
populations surveyed include geographical strains originating from different

149
150 HAYMER and McINNIS

islands within the Hawaiian archipelago, various countries in Latin America, in

Israel and limited amounts of material captured in recent infestations in California,
United States of America.
Beyond simply identifying new forms of genetic variation, the markers identi­
fied here have many other potential uses. For example, once suitable markers have
been described that identify populations of different geographical origin, it may be
possible to determine the source of origin of pest species moving into a new habitat.
In addition, these markers can be used to establish linkage relationships between
anonymous DNA sequences and genes or traits of interest in pest species. Once such
linkage relationships have been established, whole new approaches to augmenting
genetically based control programmes may become practical by making it possible
to clone specific genes or by more easily monitoring the introgression of desirable
traits into strains through classical genetic crossing schemes.

2. MATERIALS AND METHODS

The PCR procedure we hâve used here has the great advantage that it can be
carried out on limited amounts of material, such as dessicated or ethanol preserved
material, even that which has been rendered unusable for most other types of
analysis. In our procedure, DNA is first extracted from material such as a single fly
or from body parts oif a fly according to established protocols [1]. Typically, from
a single medfly we dilute an aliquot of the. extracted DNA to approximately
10 ng//xL. This small fraction of the extracted DNA serves as a template for the
amplification of specific regions of the genome in the PCR.procedure. The regions
of the genome that are amplified can serve as molecular markers which are specific
to an individual, strain, population, or species, whatever the case may be.
The particular regions of the genome that will be amplified are determined
by ‘primers’ added to the PCR reaction. These primers are short, synthetic strands
of DNA that serve as the initiation point in the amplification of a specific region
of the genome. In the RAPD approach, primers are chosen at random, and they
consist of strings of DNA 10 nucleotides in length [4]. After PCR amplification,
the DNA fragments are visualized using ethidium bromide staining. The number and
sizes of these fragments can differ between individuals or strains. The specific
pattern of the DNA fragments observed are representative of the genetic make-up
of the individual. Identically sized bands observed between different individuals are
considered to represent genetic relatedness or similarity. To quantify these relation­
ships, we calculated band sharing percentages for different medfly populations.
This number represents the proportion of bands generated by the RAPD PCR
procedure that are common, or shared, between individuals within or between
populations.
 IAEA-SM-327/15 151

3. RESULTS AND DISCUSSION

To date, fairly extensive analyses have been made of medfly populations

using three different primers. These three primers are those considered to be most
informative at a variety of. levels of genetic differentiation. These primers were
identified from tests carried out on approximately 14 different 10-mers (primers
10 nucleotides in length) synthesized for the PCR reactions. One of these primers
(referred to as primer 67) is extremely sensitive to differences in the genetic make-up
o f individuals. We used this primer to distinguish between laboratory reared and
wild caught strains of the medfly derived from the same geographical locality. The
other two primers (14 and 74) identified appear to correspond to more conserved
regions of the genome. However, these primers are extremely useful in terms of
detecting the differences between strains of different geographical origin.
Using primers such as these, we can identify RAPD markers which discri­
minate between medflies originating from different geographical localities.
However, in many cases where individuals from different populations are compared,
some RAPD marker bands produced by the PCR method appear to be common
between populations. We used this commonality to devise a measurement of
similarity based on the number of RAPD marker bands which are shared within or
between samples. We infer that this ‘band sharing’ reflects similarities in the genetic
make-up of these populations.
Table I presents band sharing calculations made for samples from several
different geographical localities, including a small sample of flies from a recent
infestation in California. The numbers shown represent band sharing values between
the different populations compared here. As is shown, the flies from California share

TABLE I. PROPORTION OF RAPD MARKER BANDS COMMON TO

DIFFERENT GEOGRAPHICAL SAMPLES

California Hawaii Guatemala C osta Rica lira/il Argentina Israel

California — 0.27 0.18 0.18 0.09 0 0

Hawaii — — 0.18 0.18 0.09 0 0

Guatemala — — — 0.18 0.09 0 0

Costa Rica — — — — 0.09 0 0.09

Brazil — — — — — 0.09 0

Argentina - — — 0.09

Israel — — —
152 HAYMER and McINNIS

a greater proportion of bands in common with samples from Hawaii and Central
America as opposed to other geographical localities. Similar results have been
obtainéd with other primers as well. Assuming that these bands represent DNA
polymorphisms which are common to different populations, these genetic markers
can be indicative of the origin and genetic relatedness of populations.
We are attempting to obtain additional material from California to examine this
further. In addition we need to examine other, presumably unrelated, geographical
populations to better understand these results. We are encouraged, however, that we
can produce these DNA polymorphisms to be used as population specific molecular
markers using the PCR technology.

REFERENCES

[1] HAYMER, D., McINNIS, D ., ARCANGELI, L., Genetic variation between strains
of the Mediterranean fruit fly, Ceratitis capitata, detected by DNA fingerprinting,
Genome 35 (1992) 528-533.
[2] ANLEITNER, J., HAYMER, D., Y specific and Y enriched sequences from the
genome of the Mediterranean fruit fly, Ceratitis capitata, Chromosoma 101 (1992)
271-278.
[3] HE, М ., HAYMER, D., Isolation and characterization of actin homologous sequences
from the Dacus dorsalis, Ann. Entomol. Soc. Am. 84 (1991) 601-607.
[4] WILLIAMS, J., KUBELIK, A ., LIVAK, K ., RAFALSKI, J., TINGEY, S., DNA
polymorphisms amplified by arbitrary primers are useful as genetic markers, Nucleic
Acids Res. 18 (1990) 6531-6535.
 IAEA-SM-327/16

RIBOSOMAL DNA OF Aphidius

(HYMENOPTERA: BRACONIDAE) NEES
Structure and intraspecific variations

G. GARGIULO*, P. VARRICCHIO**, F. PENNACCHIO***,

F. GRAZIANI+, E. TREMBLAY**, C. MALVA +

* Dipartmento di Genetica,
Biología Generate e Molecolare,
Facoltà di Scienze,
Università di Napoli,
Naples
** Dipartimento di Entomologia e Zoologia Agraria,
Università di Napoli,
Portici
*** Istituto di Entomologia Agraria e Forestale,
Università della Basilicata,
Potenza

+ Istituto Internazionale di Genetica

e Biofísica,
Consiglio Nazionale delle Ricerche,
Naples

Italy

Abstract
RIBOSOMAL DNA OF Aphidius (HYMENOPTERA: BRACONIDAE) NEES: STRUC­
TURE AND INTRASPECIFIC VARIATIONS.
An investigation is being carried out on the ribosomal DNA (rDNA) molecular organi­
zation of closely related Aphidiinae species belonging to the genus Aphidius Nees, of relevant
interest in biological control, with the aim of evaluating the variability within and between
species. After construction of a restriction map of the most represented A. ervi rDNA cistrons,
the molecular organization of the rDNA repeating units of single individuals and populations
was studied in Southern blot analyses. Heterogeneity within the A. ervi rDNA cistrons and
between populations of different geographical origin was identified. This approach allowed
the conclusion to be reached that differences in the rDNA cistrons can be diagnostic of species
and populations, therefore providing a useful tool in biological control programmes.

153
154 GARGIULO et al.

1. INTRODUCTION

Among the insects, most of the detailed studies on rDNA using the well devel­
oped recombinant DNA technologies for DNA structural analysis have been per­
formed on the dipteran fly D r o s o p h i l a m e l a n o g a s t e r [1]. More recently, the study
of rDNA has been extended to several insect species of medical and agricultural
importance in order to start the molecular characterization of the genome and to
investigate the molecular systematics of an ever-growing number of species: Díp­
tera, Orthoptera, Lepidoptera and Hymenoptera [2-4].
The study of the rDNA molecular structure of insects takes advantage of the
well described characteristics of these genes in D . m e l a n o g a s t e r , mainly their high
copy number and the presence in each repeated unit of a highly conserved coding
region and of less conserved sequences: non-transcribed spacers (NTS) and
introns [5-8]. Restriction analyses of the spacer structure of cloned rDNA gene units
have indicated that the spacers are internally repetitious and that the observed
heterogeneity in length derives in part from a different number of internal repeated
sequences [9-12]. Another particular feature of D . m e l a n o g a s t e r rDNA is that the
28S coding region sequences are interrupted by non-coding DNA sequences (rDNA
introns) in about 40% of the repeated units [13]. Almost all the dipteran species so
far investigated contain introns in some fraction of their 28S rRNA coding
regions [2 ].
We are investigating the rDNA molecular organization of closely related
Aphidiinae species, taking special interest in A p h i d i u s e r v i Haliday. We have
already demonstrated the real potentialities of this approach, both in the discrimina­
tion of species and in the determination of the phyletic relationships between taxa.
The results reported in this paper show that the rDNA structure and organization of
A . e r v i is complex and heterogeneity exists both within the rDNA cluster and
between individuals and populations of different geographical origin.

2. MATERIALS AND METHODS

A p h id iu s e r v i Haliday cultures were partially laboratory reared on their natural

hosts: A c y r t h o s i p h o n p i s u m (Harris), starting from field collections of mummies in
different areas of Campania. A p h i d i u s e r v i was also continuously reared on A . p i s u m
on potted broad bean plants. Both the aphids and the parasitoids were kept in two
separate climatic cabinets at 22 ± 1°C, 70 ± 10% relative humidity and under long
day photoperiodic conditions (18 h light: 6 h dark).
Only adults were treated according to the methodologies described below for
DNA extraction and subsequent processing. All the specimens utilized were
individually anaesthetized with C 0 2 for species identification.
 IAEA-SM-327/16 155

(1) D N A ex tr a c tio n :
The DNA from single individual and from pooled samples
was extracted according to Endow and Glover [14].
(2) C l o n e s : A complete, uninterrupted genetic ribosomal unit (cDm Y22) inserted
into pMB9 was obtained from P.K. Wellauer (Institut Suisse de Recherches
Expérimentales sur le Cancer, Lausanne, Switzerland) and its different frag­
ments were used as probes in Southern blot experiments. Cloned DNA was
prepared according to Dawid et al. [6 ]. Nick translation was performed
according to Rigby et al. [15].
(3) R e s tr ic tio n a n a ly s is , e le c tr o p h o r e s is , t r a n s fe r a n d h y b r id iz a tio n o f DNA\
Digestions with restriction enzymes were carried out under the conditions
recommended by the manufacturers (Amersham, Biolabs, Boehringer
Mannheim GmbH). Restriction maps were determined by analysing single and
double digestions on 0.6-1.5% agarose gels in Tris-borate EDTA buffer [16].
The DNA was transferred on to nitrocellulose filters, hybridized as outlined
by Southern [17] and washed four times in 2 X SSC, 0.1% SDS for 15 min
at 56°C.

3. RESULTS AND DISCUSSION

3.1. The A . erv i rDNA restriction map

To complete the A . e r v i rDNA restriction map reported on in our previous

paper [18], total DNA was extracted from pooled samples of adult specimens and
digested with several restriction enzymes in single and double digestions. Southern
blot experiments were performed using as probes the labelled D . m e l a n o g a s t e r com­
plete genetic ribosomal unit (cDm Y22) or its different parts: 18S, 28Sa and 28S/S.
An example of the numerous experiments performed is given in Fig. 1, where we
chose conditions under which only the main bands, corresponding to the most
represented rDNA cistrons, were observed. The restriction map of A . e r v i rDNA is
shown in Fig. 2. The restriction enzymes Bgin, PstI and Kpnl have only one site
in the complete repeating unit, giving the first indication of the length of most of the
rDNA cistrons, which is approximately 12 kb.
Many of the sites conserved between the different species of insects were
present also in A p h i d i u s rDNA. The highly conserved RI site in 18S, the Bgin and
Hindin sites in 28S and PstI site in the internal transcribed spacer (ITS) between 18S
and 28S were also présent in this gene, which enabled us to determine the position
of thèse regions in the A . e r v i repeating unit. On the basis of the length of the main
bands obtained with enzymes with only one site in the repeating units and the fact
that the transcribed region was almost identical in length in all the species so far
examined (about 7 kb), the spacer length was found to be 5.5-6 kb. This calculation
156 GARGIULO et al.

1 2 3 4 5 6 7

- 21

5.2
5.0
4.2
3.4

kb

2.0
1.9

1.57

1.3

- 0.98

- 0.84

FIG. 1. Southern blot analysis o f genomic DNA from A. ervi digested with (1) Hindlll/SacI;
(2) Hindlll/PvuII; (3) Hindlll/Pstl; (4) Hindlll/Kpnl; (5) Hindlll/Dral; (6) Hindlll/accI; and
(7) Hindlll and hybridized with labelled D. melanogaster 28a.

does not take into account the possible occurrence of other sources of variability,
such as the insertion sequences in the 28S region.

3.2. Differences in length of the rDNA cistrons within

and between A. e r v i individuals

We have demonstrated in a previous paper [18] that differences exist between

four A p h i d i u s species (Æ e r v i and A . m i c r o l o p h i i belonging to the e r v i species group;
A . u r t i c a e and A . s m it h i belonging to the u r t i c a e species group) that could be a means
of distinguishing some species. We have now focused our attention on A . e r v i to
investigate whether the differences were detectable within the cluster of rDNA of
each single individual and between populations of different geographical origin. In
this set of experiments, we used conditions and restriction enzymes that allow
maximum detection of heterogeneity in an attempt to also identify less represented
repeating units.
 IAEA-SM-327/16 157

RI Pst B im R1 Bgl Bam HIII R1 Bam H ill R1

JLU
Kpn

ITS 28S a NTS 18S

1 kb

FIG. 2. Restriction map o f A. ervi rDNA. As stressed in the text, the most represented rDNA
cistrons are about 12 kb long. Without taking into account the possible occurrence o f other
sources o f variability, their spacer length would be 5.5-6 kb.

1 2 3 4 5

FIG. 3. Southern blot o f genomic DNA extracted from single laboratory reared A. ervi
individuals digested with the restriction enzyme BamHI and hybridized with the labelled 28S02
fragment o f D. melanogaster rDNA.

A s can be seen in the map (Fig. 2), the BamHI enzyme sites in the 18S and
28S/32 coding regions identify a fragment which, from comparison with D r o s o p h i l a ,
should contain the NTS as well as the site where 28S insertions occur in many
insects. We therefore used this enzyme in Southern blot experiments (Fig. 3) of
genomic DNAs extracted from single laboratory reared A . e r v i individuals hybri­
dized with the labelled 28S/32 fragment of D . m e l a n o g a s t e r rDNA as probe.
158 GARGIULO et al.

1 2 3 4 5 6 7

1 5 --
13 _
12- -

9 .5 -

F/G. 4. Southern blot o f genomic DNA from A. ervi single individuals collected from
different areas (lanes 1-2, Nola; lane 3, Samo; lanes 4-6, Ischitella; and lane 7,
Castelvoltumo). The enzyme and probe are the same as those used in Fig. 3.

Numerous bands appear in each lane, indicating that a considerable amount of varia­
bility exists within the rDNA repeating units. A main band of about 7.5 kb appears
in all the individuals, indicating that this is the most represented class in single
individuals, but this band seems to differ slightly between individuals (either 7.5 or
8 kb). The other most represented bands (11 and 13 kb) are present in most of the

individuals, sometimes split as doublets, even if in different relative amounts. Faint

bands with lengths up to 2 0 kb also appear, indicating that very long repeating units
exist in A . e r v i .
It can be concluded that there is heterogeneity in this species within the rDNA
cistrons. In addition, differences at quantitative and qualitative levels also exist
between single individuals reared in the laboratory starting from heterogeneous wild
populations.
 IAEA-SM-327/16 159

3.3. Heterogeneity of the rDNA structure between different

populations of A. ervi

To investigate the rDNA population structure further, we analysed single

individuals collected from wild populations of different geographical origin.
Figure 4 shows the restriction patterns of samples collected in four different areas
of Campania, corresponding to Lanes 1-2, 3, 4-6 and 7. Differences are visible
between samples of the same and of different origin. Also in these experiments, the
main band changes between individuals (7.5-8 kb), but strong bands are also present
at a higher and/or a smaller molecular weight. There are bands present only in single
individuals; however, the variability within a single population is as high as between
populations. Therefore, more individuals have to be analysed to define the
population rDNA restriction profiles.
In conclusion, we have identified differences in the rDNA structure both within
and between A . e r v i individuals and populations. However, to characterize better,the
origin of the identified differences and to understand their evolutionary importance,
it is necessary to isolate the rDNA genes of A . e r v i and to proceed through their
sequence analysis, an effort which has already started in our laboratory.

ACKNOWLEDGEMENT

This research was supported by the National Research Council of Italy, Special
Project RAISA, Sub-Project No. 2, Paper No. N524.

REFERENCES

[1] LONG, E.O., DAWID, I.B., Restriction analysis of spacer in ribosomal DNA of
Drosophila melanogaster, Nucleic Acids Res. 7 (1979) 205-215.
[2] BECKINGHAM, K ., “ Insect rDNA” , The Cell Nucleus, Vol. X (BUSCH, H.,
ROTHBLUM, L., Eds), Academic Press, London (1982) 205-269.
[3] SCHAFER, М ., KUNZ, W ., rDNA in Locusta migratoria is very variable: Two
introns and extensive restriction site polymorphisms in the spacer, Nucleic Acids Res.
13 (1985) 1251-1265.
[4] CROSS, N.C.P., DOVER, G .A ., Tsetse fly rDNA: An analysis of structure and
sequence, Nucleic Acids Res. 15 (1987) 15-30.
[5] RITOSSA, F .M ., ATWOOD, K .C ., SPIEGELMAN, S., A molecular explanation of
the bobbed mutant in Drosophila as partial deficiencies of ribosomal DNA, Genetics
54 (1966) 819-834.
[6 ] DAWID, I.B., WELLAUER, P.K., LONG, E.O., Ribosomal DNA in Drosophila
melanogaster. Isolation and characterization of cloned fragments, J. Molec. Biol. 126
(1978) 749-768.
160 GARGIULO et al.

[7] WELLAUER, P.К ., DAWID, I.B., The structural organization of ribosomal DNA in
D. melanogaster, Cell 10 (1977) 193-212.
[8 ] COEN, E.S., DOVER, D .A ., Unequal exchange and the coevolution of the X and Y
rDNA arrays in D. melanogaster, Cell 33 (1983) 849-855.
[9] LONG, E.O., DAWID, I.B., Repeated genes in eukaryotes, Annu. Rev. Biochem. 49
(1980) 727-764.
[10] DOVER, G ., Molecular drive: A cohesive mode of species evolution, Nature (London)
299 (1982) 111-117.
[11] BONCINELLI, E., et al., Inheritance of the rDNA spacer in D. melanogaster, Mol.
Gen. Genet. 189 (1983) 370-374.
[12] SALZANO, G ., MALVA, С ., Non-random loss of uninterrupted ribosomal DNA
repeating units upon induction of a bobbed mutation, J. Molec. Biol. 177 (1984)
189-200.
[13] GLOVER, D .M ., HOGNESS, D .M ., A novel arrangement of the 18S and 28S
sequences in a repeating unit of D. melanogaster rDNA, Cell 10 (1977) 167-176.
[14] ENDOW, E.S., GLOVER, D .M ., Differential replication of ribosomal gene repeat in
polytene nuclei of D. melanogaster, Cell 17 (1979) 597-605.
[15] RIGBY, P.W.J., DIECKMANN, М ., RHODES, C ., BERG, P., Labelling deoxy­
ribonucleic acid to high specific activity in vitro by nick translation with DNA poly­
merase I, J. Molec. Biol. 113 (1977) 237-251.
[16] MANIATIS, T., FRITSCH, E.F., SAMBROOK, J., Molecular Cloning - A
Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY (1982).
[17] SOUTHERN, E., Detection of specific sequences among DNA fragments separated by
gel electrophoresis, J. Molec. Biol. 98 (1975) 503-519.
[18] GARGIULO, G ., MALVA, С ., PENNACCHIO, F., TREMBLAY, E., Structure of
Aphidius Nees rDNA: A molecular tool in biosystematic research, Boll. Lab. Entomol.
Agrie. F. Silvestri 45 (1988) 203-219.
 IAEA-SM-327/17

MITOCHONDMAL DNA VARIABILITY

IN GEOGRAPHICAL POPULATIONS
OF THE BRAZILIAN SCREWWORM FLY

A.M.L. de AZEREDO-ESPIN
Centro de Biología Molecular e Engenharia Genética,
Universidade Estadual de Campinas,
Campinas, Sào Paulo,
Brazil

Abstract
MITOCHONDRIAL DNA VARIABILITY IN GEOGRAPHICAL POPULATIONS OF THE
BRAZILIAN SCREWWORM FLY.
Restriction endonucleases analysis of mitochondrial DNA (mtDNA) was used to
examine the genetic variability of the screwworm in twelve populations collected in seven
locations of southern and northern Brazil. In two distinct areas of Sâo Paulo State, seasonal
sampling was performed to detect any possible variation in mtDNA. A group of five enzymes,
Clal, НаеШ, Hindin, MspI and PvuII, was sufficient to differentiate some of the 12 screw­
worm sample populations tested. The results of this study support theoretical predictions that
mtDNA analysis is a highly sensitive method for examining the population structure of Coch-
liomyia hominivorax mtDNA. Variability in C. hominivorax mtDNA should prove useful in
the efforts to trace the origin and dispersion of the species in Brazil.

1. INTRODUCTION

The screwworm fly, C o c h l i o m y i a h o m i n i v o r a x , is an important pest which

parasitizes livestock and other warm blooded animals of the New World. Adult
females lay eggs on open wounds or in active myiases; this results in injuries contain­
ing hundreds to thousands of larvae at different stages of development. If the myiases
are not treated in time, aggravation through reinfections may become fatal to the
host. Many cases are reported in the literature of human beings infected by screw­
worm larvae.
The main host for this pest in Brazil is cattle, but other domestic and wild
animals may be parasitized.
A basic knowledge of the genetic variability of C . h o m i n i v o r a x is necessary
to understand the population structure and evolution of this speciès in order to
develop efficient methods of pest control in Brazil.
The molecular approach to the study of genetic variability has been shown to
be very effective and to generate fast and reliable data in cases where other types
of analysis may be time consuming or impractical. In the case of insect populations,

161
162 de AZEREDO-ESPIN

one of the preferred methods is mitochondrial DNA analysis, which has been shown
to represent a powerful tool for obtaining information on the gene flow, population
structure, biogeographical hybridization zones and phylogenetic relationships.
Mitochondrial DNA variation has also become an important tool for detailed studies
of the intraspecific population structure because of its considerable polymorphism
and strong sensitivity to founder events or population subdivision. The main reasons
for this are the maternal mode of inheritance and the lack of recombination.
In this study, mtDNA variability was analysed in twelve screwworm samples
collected from seven localities of southern and northern Brazil. In two different areas
of Sâo Paulo State, seasonal sampling was performed to detect any possible variation
in mtDNA. The specific objectives of this work are to answer the following ques­
tions: (1) Can mtDNA restriction site analyses provide a sensitive method for
examining genetic variability in these natural populations; and (2) Can this analysis
reveal convenient genetic markers to trace the maternal lineages in these natural
populations?

2. MATERIALS AND METHODS

Live screwworm samples (larvae and adults) were obtained from wounds in
infested cattle at the' following localities: Adamantina (Ad-1, Ad-2, Ad-3),
Caraguatatuba (Ca-1, Ca-2, Ca-3, Ca-4), Botucatu (Bot-1), Paulinia (Pa-1), Amparo
(Amp-1), Alfenas (Alf-1) and Morro do Chapeu (Ba-1). MtDNA was purified from
individual larvae, pupae or adults, as described in Ref. [1]. The screwworm larvae
were reared in the laboratory in a medium of fresh cattle groundmeat supplemented
with blood and water (2:1). The larvae were maintained at approximately 30°C.
Mature larvae were allowed to pupate in sawdust. Adults were maintained in screen
cages at 25°C, with dried milk, sugar and ferment.
MtDNA was purified from individual larvae, pupae or adults as described in
Ref. [1]. The following 15 restriction enzymes were used: AccI, BamHI, Clal,
EcoRI, ECoRV, Haelll, Hindlll, Hpal, MspI, PvuII, PstI, SppI, SttI, Xbal and
Xhol. The purified mtDNA was nick translated to yield a 32P labelled probe.
Hybridization was carried out under standard conditions [1]. Visualization of the
mtDNA fragments was conducted using autoradiography and the fragment sizes
were estimated from autoradiographs using a digitizer and Bioscan software.

3. RESULTS AND DISCUSSION

The initial screening survey revealed that five of the fifteen enzymes used were
suitable for detecting mtDNA variation among the sampled populations. Three
different restriction patterns were obtained for Clal and PvuII, four for Hindlll and
 IAEA-SM-327/17 163

(a) (b)
1 2 3 4 5 6 7 8 9 .1 2 3 4 5 6 7 8

A С D ■ *
С D A B E

FIG. 1. Autoradiographs o f a Southern blot showing C. hominivorax mtDNA from distinct

populations, (a) Digested with Hindlll: lane I = \DNA/HindIII plus фХ 174 DNA/Haelll
standards. Fragment sizes in kilobase pairs on the left; lane 2 = Caraguatatuba (Sâo Paulo),
summer 1991; lane 3 = Botucatu (Süo Paulo); lanes 4 and 5 = Caraguatatuba (Sâo Paulo),
winter 1991; lane 6 = Alfenas (Minéis Gerais); lane 7 = Caraguatatuba (Sâo Paulo), summer
1991; lane 8 = Paulinia (Sâo Paulo), (b) Digested with MspI; lane 1 — XDNA/Hindlll plus
фХ 174 DNA/Haelll standards; lanes 2 and 3 =' Caraguatatuba (Sao Paulo), winter 1991;
lane 4 = Paulinià (Sâo Paulo); lane 5 = Botucatu (Sâo Paulo); lane 6 = Caraguatatuba
(Sâo Paulo), summer 1991; lane 7 = Caraguatatuba (Sâo Paulo)', winter 1991;
lane 8 '= Caraguatatuba (Sao Paulo), summer 1991.

five for НаеШ and MspI. The different restriction patterns for each enzyme were
designated by capital letters in the order that they happened to be discovered. As seen
in Fig. 1, the enzyme Hindlll showed three patterns, designated A (lanes 2, 3, 5,
6 , 8 and 9), В (lane 4) and С (lanes 4, 5 and 6 ); MspI showed A (lanes 4, 5 and
7), В (lane 7), С (lane 2), D (lane 3) and E (lane 8 ).
The estimated mtDNA length was similar to that reported by Roehrdanz and
Johnson and Roehrdanz [2, 3] for screwworm populations from North and Central
America and has approximately 16.3 kilobase pairs.
On the basis of the fragment patterns of four enzymes, Haelll, Hindlll, MspI
and PvuII, 14 different haplotypes were defined for the sampled populations. The
interpopulation analysis has shown high variability in mtDNA. Three geographical
populations exhibited different single haplotypes. Two haplotypes were found in
three other localities. In one specific locality, Caraguatatuba (Sâo Paulo State), nine
164 de AZEREDO-ESPIN

different haplotypes were found. The occurrence of shared haplotypes was in some
cases independent of the geographical distance. In the populations that showed more
than one haplotype, the frequencies varied and in most cases one type was prevalent.
The results indicated that different geographical populations have different haplotype
compositions and therefore different maternal lineages for mtDNA.
High variability in mtDNA was also found by Roehrdanz and Johnson and
Roehrdanz [2, 3] in 30 screwworm lines from North and Central America. Sixteen
haplotypes found individually in only two screwworm lines in Mexico showed ample
distribution, while the remaining haplotypes were foiind individually in only one
location. A comparison of these results with our data showed that no common haplo­
types were found on both the areas analysed. Taylor et al. [4] described a single
haplotype in a population found in Tripoli, Libyan Arab Jamahiriya, but there was
no similarity to any of the haplotypes found in Brazil. This sample was collected
from a recent introduction of the screwworm, probably from South America. A
further screening from other regions of Brazil and other South American countries
could perhaps indicate its possible origin.
The seasonal analysis also indicated genetic heterogeneity. In one of the areas
(Caraguatatuba), two predominant haplotypes were found; these were the pre­
dominant types found in all four samples taken in this locality. However, seven other
types were found in the samples collected during the summer (Ca-2) and in autumn
(Ca-4), but they were not detected in other samples at this locality. This variability,
plus the fact that large numbers of cattle are being taken in and out of the area, could
be the reasons for the presence or absence of flies of these haplotypes. Only further
temporal analysis of the mtDNA in this locality will provide more conclusive data
on the screwworm population in this area: In another locality (Adamantina, Sâo
Paulo State), no significant variation was observed during the year;
These preliminary data on the high variability observed on the mtDNA of
C. h o m i n i v o r a x samples collected in seven localities have indicated the existence of
different maternal lineages in Brazil. Despite the high dispersal capacity of the
screwworm flies and the movement of livestock in some regions, screening of some
of the samples has suggested local differentiation in mtDNA’and unique haplotypes.
Only further analysis of mtDNA, involving more samples from different areas
of Brazil, is needed, especially concerning the spreading and introduction of this pest
in other areas of the world, with special concern for Africa.
It has become increasingly clear that genetic variability among and within
populations of insect pests affects the success of biological control agents and insecti­
cide resistance management. The high interpopulation mtDNA variation found in 12
screwworm populations from 7 locations suggests that further studies of mtDNA will
be valuable to address the evolutionary, genetic and pest management questions of
this insect in South America. . ■
 IAEA-SM-327/17 165

REFERENCES

[1] AZEREDO-ESPIN, A .M .L ., SCHRODER, R .F.W ., HUETTEL, M .D .,

SHEPPARD, W .S ., Mitochondrial DNA variation in geographic population of
Colorado potato beetle, Leptnotarsa decemlineata (Coleoptera: Chrysomelidae),
Experientia 47 (1991) 483-485.
[2] ROEHRDANZ, R.L., JOHNSON, D .A ., Mitochondrial DNA variation among
geographical populations of the screwworm fly, Cochliomyia hominivorax, J. Med.
Entomol. 25 2 (1988) 136-141.
[3] ROEHRDANZ, R.L., Intraspecific genetic variability in mitochondrial DNA of the
screwworm fly (Cochliomyia hominivorax), Biochem. Genet. 27 9/10 (1989) 551-569.
[4] TAYLOR, D.B., HAMMACK, L., ROEHRDANZ, R.L., Reproductive compatibility
and mitochondrial DNA restriction site analysis of New World screwworm,
C. hominivorax, from North Africa and Central America, Med. Vet. Entomol. 5 (1991)
145-151.
 IAEA-SM-327/18

GENETIC FINGERPRINTING
APPLIED TO TSETSE FLY SPECIES

A. BLANCHETOT
Department of Biochemistry,
University of Saskatchewan,
Saskatoon, Saskatchewan

R.H. GOODING
■ Department of Entomology,
University of Alberta,
Edmonton, Alberta

Canada

Abstract
GENETIC FINGERPRINTING APPLIED TO TSETSE FLY SPECIES.
The bacteriophage M13 was used as a probe to detect DNA fingerprinting (DNAfp)
profiles in adults from laboratory colonies of the three subgenera of tsetse flies (Austenina,
Nemorhina and Glossina). In all three subgenera, the probe revealed profiles of multiple
components similar to those found in other organisms. The general complexity of the profiles
varied between subgenera and between species and subspecies. A common overall DNAfp
pattern was observed within a subspecies but variations occurred at the intrapopulation level.
Evidence is presented that DNAfp provides a means for population biology studies, such as
comparisons between field collected flies and those from established laboratory colonies.
Pedigree analysis was performed in the context of further development of a genetic linkage
map using DNAfp markers and studies related to the molecular basis of hybrid sterility in
tsetse flies. A pedigree established by mating a male and a female from different lines
(‘ RUCA’ x ‘Cent’) of G. m. centralis showed, in addition to a Mendelian inheritance of
DNAfp fragments, an amplification of the intensity of certain bands in the offspring. It is
suggested that DNAfp offers a tool for analysing the molecular genetic aspects of mating
flies from different geographical areas.

1. INTRODUCTION

The information accumulated in the fields of biology and genetics has now
made tsetse flies, the vectors of African trypanosomiases, interesting and amenable
models for studies at the molecular level [1]. There are 31 species and subspecies
within the genus G l o s s i n a (Diptera: Glossinidae) and they are arranged in three
subgenera or species groups [2-5]: subgenus A u s t e n in a Townsend (= f u s c a group,
15 taxa); subgenus N e m o r h i n a Robineau-Desvoidy (= p a l p a l i s group, 9 taxa); and

167
168 BLANCHETOT and GOODING

subgenus Glossina s. str. ( = morsitans group, 7 taxa). Thé divisions are based
mainly on structural characters, but are supported also by ecological informa­
tion [ 1 ].
DNA probes would be useful in research on tsetse flies, particularly in the
study of population genetics, expansion of the existing linkage map and in the search
for new methods of genetic control. Tandemly repeated sequences are well known
for their ability to detect, by DNA hybridization, high levels of variation between
individuals and to generate profiles commonly termed DNA fingerprinting
(DNAfp). DNAfp has found a wide range of applications in population biology
studies, linkage analysis studies, genetic relationships in vertebrate species and for
individual identification in forensic science [6 ]. The objective of this paper is to
demonstrate some potentials of DNAfp for developing the molecular biology of
tsetse flies.
The experimental procedures to perform DNAfp from individual insects have
been described previously [7]. In the present studies, the bacteriophage M13, known
to reveal DNAfp profiles in many organisms, including insects, was used as a
probe [7, 8 ].

2. RESULTS

2.1. DNAfp variability in tsetse colonies

Adult flies from laboratory colonies of tsetse flies of the three subgenera
(Austenina, Nemorhina and Glossina) were analysed by DNAfp (Fig. 1). The M13
sequence, detecting multiple hybridizing components in all three subgenera, acted
as a multilocus probe. Comparison of DNAfp profiles between subgenera and
between species and subspecies showed a great deal of variation in terms of the
number and position of the bands detected by the probe. Except for G. brevipalpis,
the DNAfp profiles of each subspecies were characterized by a conserved overall
pattern. In addition, the DNAfp profiles revealed intrapopulation variation within
a subspecies. Except for the ‘ RUCA’ and ‘Cent’ colonies of G. m. centralis, each
species of tsetse was represented, in the present study, by a laboratory colony that
originated from a single geographical area. Therefore, it is riot possible for us to
generalize on the geographical or ecological variations in DNAfp within a sub­
species. However, the DNAfp profiles of flies from the ‘RUCA’ and ‘Cent’
colonies have a conserved DNAfp pattern. Although these colonies descended from
puparia collected in Zambia and the United Republic of Tanzania, they originate
from within the same main fly belt. The similarity of the DNAfp in the two colonies
suggests that along this fly belt, which is several hundred kilometres in length, a
continuous gene flow has occurred between the Zambian and Tanzanian populations,
 IAEA-SM-327/18 169

(a )
kb 1 2 3 4 k b R R C C R R C C kb 1 2 3 4 kb 1 2 3 4
iiflflflft .y.* -ЯДО
- Ï
7 1 10
-W A%V 7 — 40M» *10
'I- ШШШ!!!
Дом . -Юф s <Mr

5 — ' 7 iiliiili 7
-Ж »* » * “ f T 41

4 1-
Ш Ч Й Ж râffir |Щ ^
I *
m - aw s
^
I(il

w ™
■#£: í88S-

# | i #
* * -* j 4 ”’V ^

Ш 2

G. pallidipes
т щ ь
G. m. centralis
2

G. m. morsitans
m *

G. m. submorsitans

(b) (C)
kb 1 2 3 4 kb 1 2 3 4 kb 1 2 3 4 kb 1 2 3 4

% "Ш
-тШШж 7 - i Щ 10
A#
_ ■•■■•V.V.VAW.V.W.^WMfr.
у т « < 'у М Ъ !Ш Ь > Я И » >

1 1 1 Ж
• w 5 7
- | | f ,

У л * *

- ^ ^ * щ ! i **т Ш
i |« l
S SX

^^Ш т ш 2 L * • * *
-nTW-
JÍ 2

Щ щ З Ш Ш У Й Н
9 нМ |
я я я ш
Щ 8* ^ В
Я ш Ш ж
lm
G.p. palpalis G.p. gambiensis G. brevipalpis G. longipennis

FIG. 1. DNA fingerprinting o f adult tsetse flies, (a) Subgenus Glossina: G. pallidipes,
G. m. centralis (lines: R = ‘RUCA’, С = ‘Cent’), G. m. morsitans (line: 190),
G. m. submorsitans (line: ‘Brist’); (b) subgenus Nemorhina: G. p. palpalis, G. p. gambiensis;
(c) subgenus Austenina: G. brevipalpis, G. longipennis.
170 BLANCHETOT and GOODING

TABLE I. GENETIC VARIABILITY IN ADULT TSETSE FLIES FROM

DIFFERENT SUBGENERA

Fragment Mean Band share Probability

Subgenus Species size (kb) No. of bands (N) x ± SD (P)
(No. of individuals)

Austenina G. brevipalpis 1 0 - 1 2 1 . 2 ± 0.3 (20) 0.31 ± 0.01 1 .6 x 1 0 '"

G. longipennis 1 0 - 1 1 0 .2 ± 1.7 (14) 0.77 + 0.17 6.9 x 10‘ i2

Nemorhina G. p. palpalis 1 0 - 1 13 + 2 . 2 (8 ) 0.85 + 0.6 0 . 1 2

G. p. gambiensis 7 -1 6 ± 3.4 (6 ) 0.3 ± 0.23 7.2 x 10~4

Glossina G. pallidipes 7 -1 13.0 ± 2.7 (11) 0.75 ± 0.08 2.3 x 10' 2

G. m. centralis
Line: ‘RUCA’ 1 0 - 1 2 2 . 2 ± 0.9 (4) 0.76 ± 0.07 2.2 x lO' 3

Line: ‘Cent’ 1 0 - 1 2 0 . 0 ± 2 (1 2 ) 0.84 ± 0.1 3 x 10 ‘ 2

G. m. morsitans
line: 190 1 0 - 1 10 ± 0.7 (12) 0.60 ± 0.03 6 x 10 ~ 3

G. m. submorsitans
line: ‘Brist’ 1 0 - 1 1 2 .0 + 0.8 (13) 0.93 ± 0.06 0.28

x = ((Nab/Na) + (Nab/Nb))/2, where Na and Nb represent the number of bands in the

individuals a and b and Nab is the number shared by both [9].
P = xN: Probability of two individuals having identical DNAfp.

making them indistinguishable by DNAfp. Alternatively, it is also possible that the

results suggest that colonization has selected for markedly similar DNAfp patterns.
The average number of bands detected by the M13 probe and band sharing
estimates, x, between individuals were used to quantify the amount of intra­
population variation within the different colonies (Table I) [9]. G l o s s i n a b r e v i p a l p i s
and G . p . g a m b i e n s i s display a large amount of intrapopulation variation,
comparable to that found in other species [ 1 0 ] and indicating that individual specific
DNAfp can be obtained in tsetse flies. Allozyme frequency data in G . b r e v i p a l p i s
colonies indicated that the mean level of heterozygosity per locus is among the
highest recorded in any tsetse population [11]. Except for G. b r e v i p a l p i s and
G . p . g a m b i e n s i s , the amount of intrapopulation variation in the laboratory colonies
is notably reduced (Table I). There is a dramatic reduction in the general complexity
of the profile and in the appearance of bands that are fixed in the population.
Several factors may have contributed to this: (1) the relatively small size of the
initial breeding population in some colonies (known to be the case in the
G. m . s u b m o r s i t a n s colony); (2) during laboratory maintenance each colony may
 IAEA-SM-327/ 18 171

1 2 O fD 3 4 5 6 7

1 2 3 41 < L 4 5 H 3 0

kb f2 ô ó ó á á

FIG. 2. DNA fingerprinting in pedigree analysis o f tsetse flies. Analysis o f some o f the
individuals resulting from an extensive G. m. centralis pedigree which originated by mating
a ‘RUCA’ virgin female with a ‘Cent’ male. Males are represented by squares and females
by circles and the offspring are numbered consecutively. The star indicates the position of
the band showing an amplification o f intensity in offspring females Fn and F]7.

have gone through genetic bottlenecks (almost certainly true for all of the small
colonies); and (3) all the colonies (except G . b r e v i p a l p i s and G . l o n g i p e n n i s ) were
subjected to inbreeding and selection to increase the frequency of specific alleles.
Nonetheless, the amount of variation in DNAfp provides a means for population
genetic comparisons between field collected and colonized tsetse flies. In this
context, one can propose that DNAfp be used as a routine technique to monitor
genetic variability and genetic changes in tsetse colonies.

2.2. Genetic inheritance in pedigree analysis

A pedigree analysis was performed in anticipation of the establishment of a

genetic linkage map using DNAfp and of studies of the molecular basis of hybrid
sterility. Figure 2 represents the M l3 hybridization of part of a large pedigree that
resulted from mating a virgin female from the ‘ RUCA’ line ( G . m . c e n t r a l i s from
Zambia) with a male from the ‘Cent’ line (G. m . c e n t r a l i s , originating from the
172 BLANCHETOT and GOODING

United Republic of Tanzania). For the reasons explained above, the DNAfp profiles
showed little variation between individuals and a large majority of bands are
common to both parents. In this two generation pedigree, neither the parents nor the
different offspring have an identical profile and therefore each DNAfp is unique to
an individual. The pedigree demonstrates that there is a germ line stability of
inherited fragments, that the transmission of polymorphic fragments of maternal and
paternal origin can be followed in the progeny, and that each inherited fragment has
a counterpart in at least one parent. In this particular pedigree, there was no specific
linkage of any fragment to a sex chromosome, nor did any mutated bands appear.
Some fragments showed a dramatic increase in intensity in the segregation profile
of offspring, revealing that they are inherited as a homozygous allele. Two offspring,
females Fu and F 17, showed an amplification in intensity of a fragment that cannot
be explained through normal Mendelian inheritance. Similar observations were made
during the analysis of other pedigrees involving the same intercolony cross.
Although it is rather speculative at this stage of the study, this observation may
suggest the mobilization of repeated sequences such as transposable elements.

ACKNOWLEDGEMENTS

The authors would like to thank V. Catinot for technical assistance. This work
was supported by a Research Development Grant from the Medical Research
Council of Canada (to A. Blanchetot) and by a grant (No. A-3900) from the Natural
Sciences and Engineering Research Council of Canada (to R.H. Gooding).

REFERENCES

[1] GOODING, R.H., Tsetse genetics: A review, Quaestiones Entomologicae 20 (1984)

89-128.
[2] NEWSTEAD, J.R., et al., Guide to the Study of Tsetse Flies, Mem. Liverpool School
of Tropical Medicine (New Series 1) (1924).
[3] GLASGOW, J.P., “ The genus Glossina: Introduction” , The African Trypano­
somiases (MULLIGAN, H .W ., Ed.), George Allen and Unwin, London (1970)
225-242.
[4] POTTS, W .H ., “ Glossinidae (tsetse-flies)” , Insects and Other Arthropods of Medical
Importance (SMITH, K .G .V ., Ed.), British Museum (Natural History), London (1973)
209-249.
[5] JORDAN, A .M ., “ Systematics” , Tsetse: The Future for Biological Methods in
Integrated Control (LAIRD, M ., Ed.), International Development Research Centre,
Ottawa, ON (1977) 13-22.
[6 ] JEFFREYS, A.J., et al., Hypervariable .minisatellite regions in human DNA, Nature
(London) 314 (1985) 67-73.
 IAEA-SM-327/18 173

[7] BLANCHETOT, A ., Genetic relatedness in honey bees as established by DNA finger­

printing, J. Hered. 82 (1991) 391-396.
[8] BLANCHETOT, A ., DNA fingerprinting analysis in the solitary bee Megachile
rotundata, variability and nest mate genetic relationships, Genome (in press).
[9] WETTON, J.H., CARTER, R.E., PARKIN, D .T., WALTERS, D ., Demographic
study of a wild house sparrow population by DNA fingerprinting, Nature (London)
327 (1987) 147-149.
[10] GEORGES, M ., et al., DNA fingerprinting in domestic animals using four different
minisatellite probes, Cytogenet. Cell Genet. 47 (1988) 127-131.
[11] GOODING, R.H., etal., Genetic variation in Glossina brevipalpis, G. longipennis
and G. pallidipes, and the phenetic relationships of Glossina species, Med. Vet.
Entomol. 5 (1991) 165-173.
 IAEA-SM-327/19

ISOLATION AND PRELIMINARY

RESTRICTION SITE MAP
OF MITOCHONDRIAL DNA
FROM Ceratitis capitata IN BRAZIL

S.R. MATIOLI, P. DOS SANTOS, J.S. MORGANTE

Departamento de Biología,
Instituto de Biociências,
Universidade de Sâo Paulo,
Sao Paulo,
Brazil

Abstract
ISOLATION AND PRELIMINARY RESTRICTION SITE MAP OF MITOCHONDRIAL
DNA FROM Ceratitis capitata IN BRAZIL.
Ceratitis capitata, as other tephritid fruit fly species, is a fresh fruit parasite during its
larval stages. A pest introduced to Brazil, it has now spread over large areas of the country.
Unfortunately, this species has shown very few genetic markers at the protein level that can
be used to track population scattering. Therefore, a search was started for suitable genetic
markers, the initial goal being to find genetic variation in the mitochondrial genome.
Mitochondria were isolated from pupae reared in the laboratory for several years by the usual
cell fractionation. The mitochondrial pellet was lysate; its DNA was further purified by the
CsCl equilibrium gradient in an ultracentrifuge. The purified mtDNA was cut with restriction
endonucleases and analysed by agarose gel electrophoresis, followed by direct visualization
with ethydium bromide or by Southern blot hybridization with biotinylated probes, revealed
by the histochemical reactions with streptavidin alkaline phosphatase conjugates. Eleven sites
were detected: one with BamHI, four with EcoRI, four with Hindlll, two with EcoRV and
none with Pstl. Small fragments were detected with silver stain after polyacrylamide gel
electrophoresis. The sites were mapped with multiple restriction enzyme digestions. The total
size of the mitochondrial genome was estimated to be about 16 kb.

1. INTRODUCTION

The Mediterranean fruit fly (medfly), C e r a t i t i s c a p i t a t a , was first recorded in

Brazil at the beginning of the century [1]. Until the last decade, it had not been
detected in the northern and northeastern regions of Brazil, the Reconcavo Baiano
being its northernmost limit [2]. However, some populations have recently been
recorded in Maranhâo State (in the 1990s) in previously unrecorded hosts. These
populations could have resulted from early introductions, with subsequent spreading,
or they could have originated from more recent introduction events. The study of
genetic markers can be useful in deciding between these alternative hypotheses.

175
176 MATIOLI et al.

It is known from studies made at the phenotypic level that Brazilian populations
of C . c a p i t a t a have sufficient genetic plasticity to adapt to new environments [3].
However, previous studies performed with the use of isozyme markers have shown
little variation in the sampled populations [4].
Over the past decade, analysis of genetic variability at the mitochondrial
genome has proved to be a powerful tool for tracking dispersion and for studying
the population structure of animal species [5-6]. Maternal inheritance and the high
degree of variation were believed to be the cause of this success [7]. To characterize
the mitochondrial genome of Brazilian populations, we started a programme that
began with the determination of the restriction sites of a long term laboratory
population.

2. MATERIALS AND METHODS

2.1. Isolation of mtDNA

All the flies were obtained from a long term, laboratory reared colony of
first collected in Jundiai in 1975 [3]. MtDNA from pupae at least 3 d
C. c a p ita ta
old were isolated according to Ref. [8 ] (typically 10-20 g). We used TES buffer
(Tris-HCl 30mM, pH7.6, EDTA lOmM, Saccharose 0.25M) instead of the original
MIM buffer for mitochondria isolation in order to improve the yield. We also used
a single ultracentrifugation step in a caesium chloride-ethidium bromide equilibrium,
gradient.

2.2. Restriction digestion of the mtDNA samples

The two isolated mtDNA samples were either digested with the restriction
enzymes or used to prepare biotinylated probes. In order to make precise measure­
ments of the electrophoretic migration rates, all the digestions were carried out in
a single buffer (‘One-phor-all’ Pharmacia). Typically, two units of each enzyme
were used to digest 50-200 ng of DNA for each slot.

2.3. Agarose gel electrophoresis

Gels of 0.4-0. 8 % agarose were prepared in TBE 0.5X buffer according to

Ref. [9]. Low voltages were employed (20-30 V) to ensure better resolution. The
gels were stained with ethidium bromide after running and then photographed. Frag­
ments of lambda DNA obtained from restriction nuclease digestion were used as
standards.
 IAEA-SM-327/19 177

2.4. Polyacrylamide gel electrophoresis

To analyse small DNA fragments, vertical slab gels were also cast in TBE
0.5X buffer. A thickness of 0.7 mm was used for the 5% polyacrylamide gels. After
running, the gels were fixed and silver stained.

2.5. Detection of DNA fragments with biotinylated probes

Biotinylated probes were prepared according to Ref. [10] by means of a nick

translation reaction using 14 biotin dATP as labelled precursor. After electrophoretic
running, the gels (with samples of 5 ng of mtDNA or less if pure, or more than
100 ng if total) were blotted against nitrocellulose membranes. The membranes were
hybridized with probes in formamide 45% at 42 °C overnight (further details can be
found in Ref. [10]). Detection after membrane washing was performed by a reaction
with streptavidin-alkaline phosphatase conjugates and BCIP-NBT histochemical
staining.

3. RESULTS

3.1. Estimation of restriction fragment lengths

After careful measurement of the migration distances of the bands in photo­

graphs of the gels (Fig. 1), we performed a polynomial regression analysis between
migration distances and the logarithm of the size from the known fragments of
lambda phage DNA. Table I shows length estimates of fragments obtained after
digestion with single and double restriction enzymes. The overall size of the mtDNA
from C . c a p i t a t a was calculated to be 16 209 bp, with 15 930 to 16 488 bp as the
95% confidence interval.
Some bands that appeared to be the result of a single fragment in the agarose
gels could be resolved into two bands with polyacrylamide gel electrophoresis
(Fig. 2 ).

3.2. Map construction

A tentative map is shown in Fig. 3. The site distances used to construct the
map were those obtained from the length estimates of the presumptive fragments
themselves (Table II). The map shown is therefore one of minimum distances. Dou­
ble digestions with Hindin and EcoRI once showed a slightly increased distance
between the two smaller bands than the distance observed in Hindlll digestion. We
therefore concluded that there was an EcoRI site located in the smaller Hindin
fragment.
178 MATIOLI et al.

FIG. 1. Ethidium bromide stained 0.8% agarose gel showing DNAfragments after restriction
enzyme digestion. Lane 1, lambda phage, and lanes 2-11, C. capitata mtDNA. The enzymes
used were: 1, EcoRI and Hindlll; 2, Hindlll; 3, Hindlll, BamHI and EcoRV; 4, EcoRI;
5, EcoRI and BamHI; 6, EcoRI and Hindlll; 7, EcoRI and Hindlll; 8, EcoRI and EcoRV;
9, EcoRV; 10, EcoRV and Hindlll; and 11, EcoRV and BamHI.

3.3. Detection of the biotinylated probes

Figure 4 shows a nitrocellulose membrane after biotin detection. In this mem­

brane, pure digested mtDNA fragments were analysed. In the PstI digestion, a very
high molecular weight DNA can be observed.

4. DISCUSSION

The mitochondrial DNA from C . c a p i t a t a has a length that is a model value

in animals [11]. The incompatibility that existed among the overall size estimates
from different digestions can be explained by poor estimates in the longer fragments,
 IAEA-SM-327/19 179

TABLE I. LENGTH ESTIMATES OF SOME FRAGMENTS OF DIGESTED

C . c a p i t a t a mtDNA- THE VALUES ARE THE MEANS OF TWO DIFFERENT
MEASUREMENTS. FOR DOUBLE DIGESTIONS, ONLY THE DATA FROM
NEW GENERATED FRAGMENTS ARE PRESENTED

No. of Length
Digestion 95% confidence interval
fragment (bp)

BamHI 1 16 543 16 019 17 085

EcoRI 1 7 444 7 230 7 664

2 6 172 6 006 6 343

3 1 636 1 603 1 670

4 1 078 1 050 1 107

Hindlll 1 8 262 8 0.18 8 514

2 6 849 6 659 7 048

3 473 448 501

4 467 445 490

EcoRV 1 11 563 11225 11 912

2 4 858 , 4 753 4.965

EcoRI + BamHI . - 2 4 290 . 4 192 4 393

3 3 029 2 969 3 092

Hindlll + BamHI 2 4 318 4 230 44 087

3 2 1 0 2 2 059 2 146

EcoRV + BamHI 1 9 436 9 144 9 737

3 1 998 1 960 2 039

or by the undetected loss of very small fragments, or by a combination of both fac­

tors. None of the data obtained are compatible with the hypothesis of polymorphic
sites in the strain studied.
The observation of very high molecular weight mtDNA in undigested samples
(as the PstI digest in Fig. 3) can be explained by the existence of chained quaternary
mtDNA structures caused by topoisomerases [12].
180 MATIOLI et al.

FIG. 2. Silver stained 5% polyacrylamide gel. The samples are the same as in Fig. 1, with
the addition o f lane 13, which is lambda phage DNA cut with Hindlll. The open arrowheads
indicate the positions of double bands that were detected as single bands in the agarose gels.
The filled arrowheads indicate the 564 bp lambda DNA fragment.

14 774, BamHI

FIG. 3. Ceratitis capitata restriction site map o f the enzymes studied. The distances between
the sites are minima (in bp). The starting point was chosen to be the only BamHI site observed.-
 IAEA-SM-327/19 181

TABLE П. MAP DISTANCES BETWEEN SUCCESSIVE RESTRICTION SITES

(CLOCKWISE)

Distance
Sites 95% confidence interval
(bp)

BamHI and Hindlll 2 1 0 2 2059 2146

Hindlll and EcoRI 1577 ' 1540 1615

EcoRI and EcoRI 1078 1050 1107

EcoRI and EcoRV 3720 3630 3811

EcoRV and Hindlll 1113 1088 1139

Hindlll and EcoRI 1 0 2 a Not determined

EcoRI and Hindlll 473 448 501

Hindin and Hindlll 457 435 480

HindHI and EcoRI 1145 1 1 2 0 1170

EcoRI and EcoRV 1 1 0 1 1069 1133

EcoRV and BamHI 1998 1960 2039

a Estimated by difference between digestion, as shown in lanes 2 and 6 of Fig. 2.

ЯЩМ|
1 2

FIG. 4. Detection o f C. capitata mtDNA by hybridization after transfer to a nitrocellulose

membrane. The digestions were: 1, EcoRI; 2, Hindlll; 3, PstI; 4, EcoRI and BamHI; 5, EcoRI
and Hindlll;-6, EcoRI, Hindlll and BamHI; 7, BamHI; 8, EcoRV; 9, EcoRV and BamHI; and
10, EcoRV and Hindlll. The open arrowhead indicates the position o f thé quaternary forms
o f mtDNA. The filled arrowhead indicates whole mtDNA molecules.
182 MATIOLI et al.

The procedures used to construct the map are suitable for scaling up and use
in population studies. The use of non-radioactive probes provides a safer research
environment. The map obtained here can also be used for comparisons and the
construction of phylogenetic trees if different patterns are to be obtained in the
future.

ACKNOWLEDGEMENTS

The authors wish to thank F.G. da Nobrega, A.M. Azeredo-Espin, F.J. Lara
and C.F.M. Menck for their help and suggestions, and J. Gomes da Silva for earlier
revisions of the manuscript. This work was supported by grants from the Conselho
Nacional de Desenvolvimento Científico e Tecnológico (CNPq).

REFERENCES

[1] VON IHERING, H „ Laranjas bichadas, Rev. Agrie. 6 (1901) 179.

[2] NASCIMENTO, A .S ., ZUCCHI, R .A ., Dinámica populacional das moscas-das-frutas
do género Anastrepha (Dip.:Tephritidae) no Recôncavo Baiano. I. Levantamento das
espécies, Pesq. Agropec. Bras. 16 (1981) 763-667.
[3] SOUZA, H .M .L ., MATIOLI, S.R ., SOUZA, W .N ., The process of laboratory adapta­
tion of Ceratitis capitata'. Analysis of life-history traits, Entomol. Exp. Appl. 49 (1988)
195-201.
[4] MORGANTE, J.S., SOUZA, H .M .L., DE CONTI, E., CYTRYNOWICZ, М .,
Allozymic variability in an introduced fruit fly pest Ceratitis capitata (Diptera,
Tephritidae), Rev. Bras. Genet. 4 (1981) 183-191.
15] MORITZ, C ., DOWLING, T.E ., BROWN, W .M ., Evolution of animal mitochondrial
DNA: Relevance for population biology and systematics, Annu. Rev. Ecol. Syst. 18
(1987) 269-292.
[6 ] HARRISON, R.G., Animal mitochondrial DNA as a genetic marker in population and
evolutionary biology, TREE 4 (1989) 6-11.
[7] AVISE, J.C., et al., Intraspecific phylogeography: The mitochondrial DNA bridge
between population genetics and systematics, Annu. Rev. Ecol. Syst. 18 (1987)
489-522.
[8 ] AZEREDO-ESPIN, A .M .L ., SCHRODER, R .F.W ., HUETTEL, M .D .,
SHEPPARD, W .S., Mitochondrial DNA variation in geographic populations of
Colorado potato beetle, Leptinotarsa decemlineata (Coleoptera:Chrysomelidae),
, Experientia 47 (1991) 483-485.
[9] SAMBROCK, J., FRITSCH,.E.F., MANIATIS, T ., Molecular Cloning, A Laboratory
Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY (1989).
[10] BluGENE (TM): Non-Radioactive Nucleic Acid Detection System; Manual of Proce­
dures, Bethesda Research Laboratories, Life Technologies, Inc., Bethesda, MD.
 IAEA-SM-327/19 183

[11] GRAY, М ., Origin and evolution of mitochondrial DNA, Annu. Rev. Cell Biol. 5
(1989) 25-50.
[12] NOBREGA, F.G ., Departamento de Biología, Instituto de Biociências, Universidade
de Sâo Paulo, Sâo Paulo, personal communication, 1992.
GENETICS
(Session 3)

Chairman

Y. ITÔ
Japan
 IAEA-SM-327/20

RADIATION INDUCED CHROMOSOME ABERRATIONS

FOR THE GENETIC ANALYSIS AND MANIPULATION
OF THE MEDITERRANEAN FRUIT FLY,
Ceratitis capitata

G. FRANZ, P. KERREMANS
FAO/IAEA Entomology Unit,
Agency’s Laboratory Seibersdorf, ‘
International Atomic Energy Agency,
Vienna

Abstract
RADIATION INDUCED CHROMOSOME ABERRATIONS FOR THE GENETIC
ANALYSIS AND MANIPULATION OF THE MEDITERRANEAN FRUIT FLY, Ceratitis
capitata.
Gamma radiation was used to induce chromosome aberrations for the analysis and
genetic manipulation of the Mediterranean fruit fly (medfly). Several new translocations
linking different autosomes to the male determining Y chromosome were detected. In three
separate experiments, a total of 34 translocation strains were recovered. Sixteen of these
strains were subjected to more detailed analysis. The results concerning their overall viability
are presented. In addition, an attempt has been made to determine the cytological location of
the white pupa mutation. For this purpose, deletions were induced. Of the over 8000 pupae
screened, four strains with visible chromosome aberrations were found.

1. INTRODUCTION , ,

The Mediterranean fruit fly (medfly), C e r a t i t i s c a p i t a t a , is a serious agricul­

tural pest in tropical and subtropical areas. To eliminate the damage caused by this
fly, the sterile.insect technique (SIT) has been applied as an environmentally safe
alternative to insecticides (for a review, see Ref. [1]).
However, the cost effectiveness and application range of this technique could
be improved if methods for male-only releases were developed. The released sterile
males are the primary active agent in SIT and, consequently, the cost of mass rearing
could be reduced if procedures to eliminate the females as early as possible were
available. Furthermore, the number of flies required to achieve control can be
reduced, since the problem of preferential mating among the released flies no longer
exists. [2]. Additional savings in programme costs result from the reduction in
monitoring, which is greatly simplified if the only females trapped are those from
the wild population [3].

187
188 FRANZ and KERREMANS

The application range of the SIT is extended in two respects if only males are
released. First, those countries that did not allow use of this technique because of
the damage caused by the sterile females attempting to deposit their eggs into the fruit
can now also apply SIT. Second, SIT can also be used to control rather than to
eradicate the pest.
To be able to release only males, genetic manipulation of the medfly is neces­
sary. In all existing genetic sexing systems, the wild type allele of a selectable marker
is linked to the male sex through translocations between the respective autpsome and
the Y chromosome. This type of chromosome aberration is usually induced by radia­
tion. The selection of the appropriate sexing gene determines primarily the cost
effectiveness of the sex separation (i.e. it should be as early and as accurate as possi­
ble and should not require expensive equipment), while choosing which translocation
to use in the strain determines the genetic stability of the sexing system during mass
rearing. It has been shown that recombination in the males occurs [4, 5] and that
genetic exchange in the chromosomal region between the sexing gene and the trans­
location break point leads to the breakdown of the sexing system [6 ].

2. RESULTS AND DISCUSSION

2.1. Induction of Y autosome translocations

In the first experiment (T7-1) (see Table I) aimed at the induction of Y auto­
some translocations, pupae of the wild type strain Egypt II (EglI) were irradiated
with 50 Gy in a ^Co source 1 d before adult emergence. Single males, 61 in total,
were mated with females homozygous for the mutation white pupa ( w p ) [7].
F[ males were also crossed with females from the w p strain. The F2 progeny of
these families, originating from a single irradiated male, were scored for pseudo­
linkage between w p and sex, i.e. for the appearance of only mutant females and only
wild type males. Seven families were found. In parallel, the remaining families were
analysed for translocations linking either chromosome 3 or chromosome 4 to the
Y chromosome, i.e. the same crossing scheme as that described for w p was used,
but in this case the mutations dark pupa ( d p ) [8 ] and apricot eye ( a p ) [8 ] were
used [9, 10].
Further genetic and cytological analyses of these 13 new translocation strains
showed that seven (11.5% of 61 families) displayed pseudolinkage with chromo­
some 5 and six (9.8% of 61 families) with chromosome 3 or 4, respectively. This
means that under these experimental conditions approximately 1 0 % of all the scored
families carry at least one Y autosome traiislocation for any of the five autosomes,
assuming that the autosomes not tested here (chromosomes 2 and 6 ) exhibit the same
characteristics.
 IAEA-SM-327/20 189

TABLE I. INDUCTION OF Y AUTOSOME TRANSLOCATIONS

No. of single pair families

Experiment Dose Screened With With more than Producing

(Gy) ' pseudolinkage 1 autosome adjacent- 1

(marker used) involved adults

T7-1 50 61 7 {wp) 5 3 "

(52) 3 (ар) 1 0

(52) 3 (dp) 1 1

T7-2 50 150 8 (wp) 6 4

T7-3 40 300 1 2 (we wp) 5 5 -8

Note: Numbers in parentheses denote remaining families after screening for pseudolinkage
with wp.

In the second experiment (T7-2), 150 irradiated EglI males were scored for
pseudolinkage between w p and sex. In this case, eight (5.3% of 150 families) new
translocations were detected. In both experiments (T7-1 and T7-2), a high frequency
of complex translocations was observed, i.e. of 21 translocations, 13 involved more
than one autosome. In the most extreme case, three autosomes were indirectly linked
via autosome-autosome translocations to the Y chromosome. The presence and the
complexity of the translocation has a severe impact on the viability of these strains
(Table П) [10-12]. Strains with simple Y autosome translocations produce approxi­
mately half the number of males per 1 0 0 0 fertilized eggs as the standard non­
translocation control (Fig. 1). If two autosomes are involved, this value is again
reduced by 50% and in the case of translocations linking three autosomes to the
Y chromosome, male production is reduced to approximately 13% of the control.
Consequently, the most suitable strains for mass rearing carry translocations
where only one autosome is involved. Experiments T7-1 and T7-2 resulted in four
strains where only chromosome 5 and the Y chromosome were linked. To increase
this number, a third experiment (T7-3) was initiated. This time, a dose of 40 Gy was
used and 300 single pair families were analysed (Table I). Of 13 families with
pseudolinkage, eight contained simple T(Y;5) translocations.
To construct genetic sexing strains, the position of the Y autosome break point
has to be known. Genetic and cytological experiments were performed to determine
the break points in the existing strains [ 1 0 , 1 1 ] and to analyse their stability using
either w p or .a temperature sensitive lethal (ts l) [13] as the selectable marker [11].
190 FRANZ and KERREMANS

TABLE II. VIABILITY OF Y AUTOSOME TRANSLOCATIONS INVOLVING

ONE, TWO OR THREE AUTOSOMES [10-12]

Strains No. of eggs % hatch % pupae % males No. of males/1000 Refs

T(Y;5)30C 5 380 74.7 .83.5 46.7 291 [1 1 ]

T(Y;5)1-61 9 210 64.2 82.3 44.5 235 [1 2 ]
T(Y;5)2-22 2 700 69.8 80.9 41.4 234 [1 2 ]
T(Y;5)3-129 1 800 69.7 67.2 40.0 187
T(Y;5)3-245 1 800 84.6 69.1 36.6 214
T(Y;5)I-59 5 600 67.8 79.5 37.0 .. 2 0 0 [1 2 ]
T(Y;5)2-40 500 56.6 66.4 41.5 156 [1 2 ]
T(Y;3)l-5 300 62.0 75.3 34.3 160 [1 0 ]
T(Y;3)l-30 300 85.7 66.9 30.8 177 [1 0 ]
T(Y;4)l-9 300 74.7 82.6 36.2 223 [1 0 ]
T(Y;4)l-33 300 61.7 87.0 36.6 197 [1 0 ]

Total T(Y;A) 28 640 6 8 . 2 79.7 42.0 227

T(Y;2;5)1-15 500 39.4 56.3 42.3 94 [1 2 ]

T(Y;3;5)l-56 4 900 48.6 60.4 39.0 114 [1 1 ]
T(Y;2;5)2-82 5 1 0 0 46.2 65.6 ‘ 34.6 105 [1 1 ]
T(Y;3;4)l-43 300 47.7 83.2 43.7 173 [1 0 ]

Total T(Y;A;A) 1 0 800 . 47.0 63.3 37.2,- 111

T(Y;2;4;5)2-54 500 36.2 39.8 26.4 38 [1 2 ]

T(Y;3;4;6)l-50 300 31.0 49.5 43.5 67 [1 0 ]

Total T(Y;A;A;A) 800 34.3 43.1 33,1 .4 9

2.2. Induction of deletions

The second essential prerequisite for the construction of stable genetic sexing
strains is to determine the chromosomal location of the selectable marker. In several
experiments we have attempted to induce deletions of the w p locus (Table III) [14].
Because no balancer chromosomes exist in the medfly, the deletions were induced
in a Y autosome translocation (T(Y;5)30C, Refs [14, 15]), i.e. the deletions
remained linked to the Y chromosome and, therefore, to the male sex. The irradiated
males ( J ( Y ' , 5 ) w p + / w p + ) were crossed in large cages with w p females and the
Fl generation was screened for the occurrence of white pupae. Males emerging
 IAEA-SM-327/20 191

500

400
to
О)
О)
Ф
■о
<D
N

fe 300 -
о
о
о
ф
Q.
73
Ф
О
3 200 -
тз
о
(Л
ф
<с

100 -

0 1 2 3

No. of autosomes involved in T(Y;A)

FIG. I: Production of males in strains with no, one, two or three autosomes linked to the
Y chromosome. Male survival is calculated per 1000 fertilized eggs.

from these pupae were used to establish single pair families and their offspring were
analysed cytologically. In the first three experiments (T6-1, T6-2 and T6-3) (see
Table III), five w p males were recovered. However, analysis of the trichogen poly­
tene chromosomes revealed a chromosome aberration in only one of them, and this
was a transposition and not a deletion. In the fourth experiment, the dose was
reduced to 30 Gy. This resulted in the recovery of four w p males, three of which
showed a deletion and were used in combination, with the transposition to locate the
w p locus on chromosome 5 [12].
192 FRANZ and KERREMANS

TABLE Ш. INDUCTION OF DELETIONS IN T(Y;5)30C [12]

No. of No. of No. of No. of

Dose Strain
Experiment brown white wp wp Aberration
(Gy) [Ref.]
pupae pupae females males

T6-1 50 Not counted 3 0 2 T6-1-9 No

T6-1-12 [12] Transposition

T6-2 50 1540 7 0 3 T6-2-1 No

T6-2-5 No
T6-2-6 No

T6-3 50 1338 2 1 0

T6-4 50 639 0

40 1052 0

30 3483 6 1 4 T6-4-1 [12] No

T6-4-2 [12] Deletion?
T6-4-3 [12] Deletion
T6-4-4 [12] Deletion

T o ta l >8052 18 2 9 4

3. CONCLUSIONS

Many of the genetic and cytogenetic tools required to analyse and manipulate
the medfly genome are now available [16]. Approximately ninety mutations have
been identified and the polytene chromosomes have been mapped [17-20]. In addi­
tion, many Y autosome translocations have been induced. These can be used either
to locate mutations or to construct genetic sexing strains. Mapping through radiation
induced deletion has also been established in the medfly. The first case was the
mapping of the alcohol dehydrogenáse ( A d h ) loci on chromosome 2 [21-23]. We
used this technique to locate the w p gene on the right arm of chromosome 5 [15] and,
recently, we isolated three potential deletions for the mutation white eye ( w e ) [24]
(data not shown).

REFERENCES

[1] KLASSEN, W ., LINDQUIST, D .A ., BUYCKX, E.J., “ Overview of the Joint

FAO/IAEA Division’s involvement in fruit fly sterile insect technique programmes” ,
Proc. XIX Int. Congress on Entomology, Beijing, 1992 (in press).
 IAEA-SM-327/20 193

[2] R O BINSO N, A .S ., C IR IO , U ., HOOPER, G .H .S ., C A P P A R E L L A , М ., F ield cage

studies w ith a genetic sexing strain in the Mediterranean fru it fly , Ceratitis capitata,
Entomol. Exp. A pp l. 41 (1986) 231-235.
[3] Economic Evaluation o f M ed fly Control in the Maghreb Using the Sterile Insect Tech­
nique, Report by Expert Group, Joint F A O /IA E A D ivision o f Nuclear Techniques in
Food and A griculture, D ivision o f Technical Co-operation Programmes, IA E A ,
Vienna, Sep. 1992 (internal report).
[4] RÔSSLER, Y ., Genetic recombination in males o f the Mediterranean fru it fly , and its
relation to automated sexing methods, Ann. Entomol. Soc. Am . 75 (1982) 28-31.
[5] ROSSLER, Y ., Recombination in males and females o f the Mediterranean fru it fly
(Diptera: Tephritidae) w ith and without chromosome aberrations, Ann. Entomol. Soc.
Am . 75 (1982) 619-622.
[6] BUSC H-PETERSEN, E ., K A F U , A ., Stability o f two mass-reared genetic sexing
strain o f Ceratitis capitata (Diptera: Tephritidae) based on pupal color dimorphism,
Environ. Entomol. 18 (1989) 315-322.
[7] ROSSLER, Y ., The genetics o f the Mediterranean fru it fly : A “ white pupae” mutant,
Ann. Entomol. Soc. A m . 72 (1979) 583-585.
[8] ROSSLER, Y ., K O L T IN , Y ., The genetics o f the Mediterranean fru it fly Ceratitis
capitata-. Three morphological mutations, Ann. Entomol. Soc. A m . 69 (1976)
604-608.
[9] K E R R E M A N S , P., G E N C H E V A , E ., F R A N Z , G ., The sterile insect technique:
Im proving genetic sex-separation methods fo r the m edfly, Ceratitis capitata, Med. Fac.
Landbouww. R ijksuniv. Gent 56 (1991) 1053-1064.
[10] K E R R E M A N S , P., G E N C H E V A , E ., F R A N Z , G ., Genetic cytogenetic analysis o f
Y-autosome translocations in the Mediterranean fru it fly , Ceratitis capitata, Genome
35 (1992) 264-272.
[11] F R A N Z , G ., K E R R E M A N S , P., Improved stability o f genetic sex-separation strains
fo r the Mediterranean fru it fly , Ceratitis capitata (in preparation).
[12] K E R R E M A N S , P., F R A N Z , G ., Cytogenetic analysis o f chromosome 5 from the
Mediterranean fru it fly , Ceratitis capitata (in preparation).
[13] F R A N Z , G ., BUSCH-PETERSEN, E ., Analysis o f a temperature sensitive lethal
mutation used as selectable marker in a genetic sexing strain o f the Mediterranean fru it
fly , Ceratitis capitata (in preparation).
[14] K E R R E M A N S , P., B O U R TZIS, K ., Z A C H A R O P O U L O U , A ., Cytogenetic analysis
o f three genetic sexing strains o f Ceratitis capitata, Theor. A ppl. Genet. 80 (1990)
177-182.
[15] BUSC H-PETERSEN, E ., SO UTHERN , D .I., Induced suppression o f genetic recom­
bination in females o f the Mediterranean fru it fly , Ceratitis capitata (W ied.), by trans­
location heterozygosity, Genetica 72 (1987) 161-169.
[16] F R A N Z , G ., K E R R E M A N S , P., Requirements and strategies fo r the development o f
genetic sex separation systems w ith special reference to the Mediterranean fru it fly ,
Ceratitis capitata, Proc. X IX Int. Congress on Entomology, Beijing 1992 (in press).
[17] BEDO , D .G ., Polytene and m itotic chromosome analysis in Ceratitis capitata
(Diptera: Tephritidae), Genome 28 (1986) 180-188.
[18] BEDO , D .G ., Polytene chromosome mapping in Ceratitis capitata (Diptera:
Tephritidae), Genome 29 (1987) 589-611.
194 FRANZ and KERREMANS

[19] ZACHAROPOULOU, A ., Cytogenetic analysis of mitotic and salivary gland chromo­

somes in the Medfly, Ceratitis capitata, Genome 29 (1987) 67-71.
[20] ZACHAROPOULOU, A ., Polytene chromosome maps in the Medfly, Ceratitis
capitata, Genome 33 (1990) 184-197.
[21] RIVA, M .E ., ROBINSON, A .S ., Induction of alcohol dehydrogenase null mutations
in the Mediterranean fruit fly, Ceratitis capitata, Biochem. Genet. 24 (1986) 765-774.
[22] ROBINSON, A .S ., RIVA, M .E ., ZAPATER, М ., Genetic sexing in the Mediterra­
nean fruit fly, Ceratitis capitata, using the Adh locus, Theor. Appl. Genet. 72 (1986)
455-457/
[23] ZACHAROPOULOU, A ., RIVA, M .E ., MALACRIDA, A ., GASPERI, G.,
Cytogenetic characterization of a genetic sexirig strain in Ceratitis capitata, Genome
34 (1991) 606-611.
[24] RÔSSLER, Y ., ROSENTHAL, H., Genetics of the Mediterranean fruit fly (Diptera:
Tephritidae): Morphological mutants on chromosome 5, Ann. Entomol. Soc. Am. 85
(1992) 525-531.
 IAEA-SM-327/21

INHERITANCE OF REFRACTORINESS
TO TRYPANOSOME INFECTION IN TSETSE

I. MAUDLIN, S.C. WELBURN

Tsetse Research Laboratory,
Department of Veterinary Medicine,
University of Bristol,
Bristol,
United Kingdom

Abstract .
INHERITANCE OF REFRACTORINESS TO TRYPANOSOME INFECTION IN TSETSE.
Differences in susceptibility to midgut infection between teneral flies from susceptible
and outbred stocks disappear in non-teneral flies, showing that maternally inherited suscep­
tibility to midgut infection is a condition expressed only in teneral flies. The increased suscep­
tibility of colonized Glossina morsitans morsitans to trypanosome infection compared with
wild flies can be related to the spread of flies carrying rickettsia like organisms through coloni­
zation. The relative refractoriness of non-teneral flies suggests that flies fed prior to release
in a sterile insect technique programme would not play a significant part in the spread of
Trypanozoon or Trypanosoma congolense infections.

1. INTRODUCTION

The enhanced susceptibility to infection of teneral flies has long been recog­
nized. Duke [1] found that repeatedly infecting G l o s s i n a f u s c i p e s with T r y p a n o s o m a
b r u c e i sensu lato (si) did not significantly increase the infection rates over flies
infected only at the first feed. Wijers [2] and Harley [3] later confirmed the superior
susceptibility of teneral flies to T r y p a n o z o o n infections. It was later shown [4, 5] that
teneral flies are also more susceptible to T. c o n g o l e n s e infections than fed flies.
The present work examines the relationship between the heritability of refrac­
toriness to trypanosome infection in tsetse and the teneral state of the fly.

2. MATERIALS AND METHODS .

2.1. Teneral and non-teneral infection rates

G l o s s i n a m o r s i t a n s m o r s i t a n s from two stocks were used: (1) outbred, and

(2 ) susceptible an iso-female line selected for susceptibility to trypanosome
infection [6 ]. Flies were infected [7] with bloodstream form T. c o n g o l e n s e

195
196 MAUDLIN and WELBURN

(stock 1/148) [8 ] either when teneral or at the second to seventh subsequent feed.
All the flies were starved for three days before infection.

2.2. Comparison of laboratory and wild flies

puparia were collected in Zimbabwe from the

G lo s s in a m o r s ita n s m o r s ita n s
same population from which the original Langford colony was derived in 1967.
Haemolymph was collected from flies emerging from wild collected puparia
and from a control group taken from the Langford G . m . m o r s i t a n s colony. Haemo-
cytes were stained and examined by light microscopy for the presence or absence
of rickettsia like organisms (RLO) infections.
The emergent female flies from wild collected puparia and a control group of
females from the Langford G . m . m o r s i t a n s colony were also infected with
T. b . r h o d e s i e n s e (EATRO 2340) [9].

2.3. Dissection

Mouthparts (labrum and hypopharynx) and midguts were dissected and exam­
ined for trypanosomes 21 d post-infection (Г. c o n g o l e n s e infections), and salivary
glands and midguts were dissected and examined 28 d post-infection (Г. b . r h o d e ­
s i e n s e infections).

3. RESULTS

3.1. Teneral and non-teneral infection rates

Table I shows the infection rates in two lines of G . m . m o r s i t a n s infected with

at increasing ages. The non-teneral flies of both lines had significantly
T. c o n g o l e n s e
lower infection rates than the teneral flies. While all the midgut infections matured
when the flies were infected as teñerais, the non-teneral transmission index
(% hypopharynx/midgut) was significantly reduced at all stages compared with the
teñerais.
The midgut infection rates in the susceptible line of teneral G . m . m o r s i t a n s
were significantly greater than in the outbred teneral flies (P < 0.001). A compari­
son with non-teneral flies showed no significant differences in the midgut infection
rates between the lines of flies.

3.2. Comparison of laboratory and wild flies

Significantly less (21.8%, n = 202) haemolymph samples from wild G . m .

were found to be infected with RLO than Langford colony flies (76.0%;
m o r s ita n s
 IAEA-SM-327/21 197

TABLE I. FLY AGE AND T. c o n g o l e n s e INFECTION RATES IN SUSCEP­

TIBLE AND OUTBRED LINES OF G . m . m o r s i t a n s . THE TENERAL FLIES
WERE INFECTED ON THE DAY AFTER EMERGENCE FROM THE
PUPARIUM (GROUP 0). THE NON-TENERAL FLIES WERE INFECTED AT
SUBSEQUENT BLOODMEALS (GROUPS 1-6). THE INTERVAL BETWEEN
BLOODMEALS WAS THREE DAYS
(N : n u m b e r o f f l i e s d i s s e c t e d ; M G : % m i d g u t ; H y p : % h y p o p h a r y n x in f e c t i o n s )

Susceptible Outbred
No. of feeds
N MG Hyp N MG Hyp

Teneral 44 95.5 95.5 77 57.1 57.1

1 74 16.2 8 .1 103 20.4 16.5

2 80 16.3 .2 .5 117 5.1 3.4

3 81 3.7 1 .2 111 3.6 1.8

4 76 1 1 .8 5.3 195 6.7 3.1

5 83 , 8.4 3.6 96 9.4 6.3

6 37- 2.7 0 . 0 115 13.0 4 :3

n = 105; P < 0.001). Wild flies also showed significantly lower midgut infection
rates (22%, n = 46) than colonized flies (82%, n = 32) (P < 0.001); the salivary
gland infection rates were 0.0% and 6.5%, respectively.

4. DISCUSSION

The results presented here show that tsetse flies, once fed, remain highly
refractory to infection throughout their lives. The differences in susceptibility to
midgut infection observed between outbred G . m . m o r s i t a n s and a line from the same
population selected for susceptibility to infection disappeared in the non-teneral flies.
This suggests that enhanced susceptibility to T. c o n g o l e n s e and T r y p a n o z o o n midgut
infection in G . m . m o r s i t a n s , which has been shown to be maternally inherited [10],
is expressed only in teñerais and is abolished after the second feed.
We have previously shown that susceptibility to trypanosome infection is
related to RLO infection in teneral G . m . m o r s i t a n s [11]. The increased susceptibility
to midgut infection of laboratory reared G . m . m o r s i t a n s compared with wild flies
shown here may be attributed to the accumulation of RLO in the absence of selection
to which the field population would normally be exposed.
198 MAUDLIN and WELBURN

Refractoriness to trypanosome infection has been shown to be due to trypano-

some killing by tsetse midgut lectin [12]. Teneral flies would be expected to produce
less midgut lectin or to produce inhibitors which would disable the lectin prior to
contact with the parasites. Two such inhibitors, D glucosamine and N-acetyl-D-
glucosamine, when added to the tsetse bloodmeal, have been shown to permit fed
flies to become superinfected; this could be generated by the enzyme digestion of
chitin [13]. Tsetse RLO produce chitinases in vitro [14] and their action in the pupar­
ium could increase the susceptibility of the teneral fly. This effect would, however,
be abolished by the large amounts of lectin produced in response to feeding. The fact
that older flies can become superinfected by the addition of lectin inhibitory sugars
[13] suggests that tsetse midgut lectin normally prevents parasitism in non-teneral
flies arid that there is no physical barrier (for example, the peritrophic membrane)
to infection in older flies.
The significance of non-teneral flies in the transmission of trypanosomiases
remains controversial. Harley [3] found that 3.0% of the non-teneral G . f f u s c i p e s
acquired mature infections of T. b . r h o d e s i e n s e compared with 14.3% of the teneral
flies, challenging the opinion of Wijers [2] that flies could only be infected with
T r y p a n o z o o n at the teneral feed. Wijers [2] was, however, working with T. b . g a m -
b i e n s e , which produces few mature infections, even in teneral flies [15]. Recent
related experiments suggesting that non-teneral flies may be of epidemiological
significance [16-19] have often been conducted with colonized G . m . m o r s i t a n s
which, as we have shown, have lost much of their innate refractoriness to midgut
infection.
The present work shows that a laboratory colony of tsetse susceptible to
trypanosome infection when teneral becomes highly refractory to midgut infection
when infected as non-teneral. If flies are to be released in a sterile insect technique
programme, then it is important to ensure that such flies are fed prior to release in
order to minimize disease transmission. However, logistically it would be advanta­
geous to breed stocks of tsetse which are as intrinsically refractory to trypanosome
infection as teñerais.

ACKNOWLEDGEMENTS

The financial support of the The Wellcome Trust; the Overseas Development
Administration of the United Kingdom Government (through the Livestock Protec­
tion Programme of the Natural Resources Institute), the UNDP/World Bank/WHO
Special Programme for Research and Training in Tropical Diseases and the Euro­
pean Economic Community (Science, Twinning and Operations programme) is
gratefully acknowledged.
 IAEA-SM-327/21 199

REFERENCES

[1] DUKE, H .L ., Studies on the factors that may influence the transmission of the poly­
morphic trypanosomes by tsetse. Di. On the infectivity to Glossina of the trypanosome
in the blood of mammal, Ann. Trop. Med. Parasitol. 29 (1935) 131-143.
[2] WDERS, D.J.B., Factors that may influence the infection rate of Glossinapalpalis with
Trypanosoma gambiense. I. The age of the fly at the time of the infective feed, Ann.
Trop. Med. Parasitol. 52 (1958) 385-390.
[3] HARLEY, J.M.B., The influence of age of the fly at the time of the infecting feed on
infection of Glossina fuscipes with Trypanosoma rhodesiense, Ann. Trop. Med.
Parasitol. 65 (1971) 191-196.
[4] DISTELMANS, W ., et al., The susceptibility of Glossina palpalis palpalis at different
ages to infection with Trypanosoma congolense, Ann. Soc. Belg. Med. Trop. 62 (1982)
41-47.
[5] MWANGELWA, M .I., et al., Some barriers to Trypanosoma congolense development
in Glossina morsitans morsitans, Insect Sci. Appl. 8 (1987) 33-37.
[6 ] MAUDLIN, I., et al., Extrachromosomal inheritance of susceptibility to trypanosome
infection in tsetse flies. П. Susceptibility of selected lines of Glossina morsitans morsi­
tans to different stocks and species of trypanosome, Ann. Trop. Med. Parasitol.
80 (1986) 97-105.
[7] WELBURN, S.C., MAUDLIN, I., A simple in vitro method for infecting tsetse with
trypanosomes, Ann. Trop. Med. Parasitol. 81 (1987) 453-455.
[8 ] YOUNG, C.J., GODFREY, D .G ., Enzyme polymorphism and the distribution of
Trypanosoma congolense isolates, Ann. Trop. Med. Parasitol. 77 (1983) 467-481.
[9] CORNELISSEN, A .W .C .A ., et al., Characteristics of trypanosome variant antigen
genes active in the tsetse fly, Nucleic Acids Res. 13 (1985) 4661-4676.
[10] MAUDLIN, I., DUKES, P., Extrachromosomal inheritance of susceptibility to
trypanosome infection in tsetse flies. I. Selection of susceptible and refractoiy lines of
Glossina morsitans morsitans, Ann. Trop. Med. Parasitol. 79 (1985) 317-324.
[11] WELBURN, S.C., MAUDLIN, I., Rickettsia-like organisms, puparial temperature
and susceptibility to trypanosome infection in Glossina morsitans, Parasitology 102
(1991) 201-206.
[12] MAUDLIN, I., WELBURN, S.C., Lectin mediated establishment of midgut infections
of Trypanosoma congolense and Trypanosoma brucei in Glossina morsitans, Trop.
Med. Parasitol. 38 (1987) 167-170.
[13] WELBURN, S.C., MAUDLIN, I., The nature of the teneral state in Glossina and its
role in the acquisition of trypanosome infections in tsetse, Ann. Trop. Med. Parasitol.
(in press).
[14] WELBURN, S.C., The Rickettsia-Like Organisms of Glossina spp., PhD Thesis,
University of Bristol, Bristol (1991).
[15] DUKES, P., et al., A new method of isolating Trypanosoma brucei gambiense from
sleeping sickness patients, Trans. R. Soc. Trop. Med. Hyg. 83 (1989) 636-639.
[16] GINGRICH, J.B., et al., African sleeping sickness: New evidence that mature tsetse
flies (Glossina morsitans) can become potent vectors, Trans. R. Soc. Trop. Med. Hyg.
76 (1982) 479-481.
200 MAUDLIN and WELBURN

[17] G I N G R I C H , J.B., et al., Trypanosoma brucei rhodesiense (Trypanomastidae): Factors

influencing infection rates of a recent human isolate in the tsetse Glossina morsitans
(Diptera: Glossinidae), J. Med. Entomol. 19 (1982) 268-274.
[18] G O O D I N G , R.H., Infection of post-teneral tsetse flies (Glossina morsitans morsitans
and Glossina morsitans centralis) with Trypanosoma brucei brucei, Can. J. Zool. 66
(1988) 1289-1292.
[19] GIBSON, W., FERRIS, V., Sequential infection of tsetse flies with Trypanosoma
congolense and Trypanosoma brucei, Acta Trop. 50 (1992) 345-352.
 IAEA-SM-327/22

ACTIVE QUALITY CONTROL

IN MASS REARED MELON FLIES
Quantitative genetic aspects

T. MIYATAKE
Fruit Flies Laboratory,
Okinawa Préfectoral Agricultural Experiment Station

M. YAMAGISHI
Okinawa Fruit Fly Eradication Project Office

Naha, Okinawa,
Japan

Abstract

A C T I V E Q U A L I T Y C O N T R O L IN M A S S R E A R E D M E L O N FLIES: Q U A N T I T A T I V E
G E N E T I C A SPECTS.
It is important to maintain the quality of mass reared flies throughout many generations
in establishing the sterile insect technique. In practical mass rearing of the melon fly, Bac-
trocera cucurbitae, many traits have already been differentiated between mass reared and wild
flies. First, the differing traits of wild and mass reared melon flies are reviewed and the factors
which have caused these differences are considered. In the mass rearing procedure, some
artificial selection pressures have been attributed to many traits of the flies. Next, an attempt
is made to predict the change in larval period in mass rearing. As a result, a quantitative
genetic model successfully predicts the change in traits. Finally, consideration is given to some
correlated responses to artificial selection in mass rearing. M a n y such responses to artificial
selection in the mass rearing process are explored. The survival rate that correlated to
reproduction was successfully controlled by artificial selection for reproduction in mass
rearing. O n the basis of these results, active quality control for mass reared melon flies in the
future is discussed.

1. INTRODUCTION

The complete eradication o f the melon fly, Bactrocera cucurbitae Coquillett,

from Japan, using the sterile insect technique (SIT) will be announced in 1993 [1].
After eradication, we must make efforts to prevent reinvasion by the melon fly from
foreign countries. Should reinvasion occur, we will also make an effort to eradicate
the invading population from the location where reinfestation originated, immedi­
ately after detection. Thus, we have to continue mass rearing o f the melon fly for
SIT even after eradication. There is no other valid method for eradication except
SIT [2]. In this context, it is important to maintain mass reared flies over many

201
202 M IYАТАКЕ and YAMAGISHI

generations continuously. However, a method to maintain the high quality o f mass

reared flies that are suitable for use with SIT has not been established. Our goal is
to establish a method to control and improve the quality o f mass reared flies. In this
paper, we review the differences in traits between wild and mass reared melon flies
and estimate the change in these traits caused by mass rearing. Two questions are
addressed. First, are there any correlated responses among these traits? Second, can
we control the changes in these traits by intentional selection in mass rearing?

2. DIFFERENCES IN TRAITS BETWEEN WILD

AND MASS REARED MELON FLIES

The differences in some characteristics between wild and mass reared melon
flies have been extensively studied in Okinawa, Japan. We summarize the results o f
these reports in Table I [3-17]. In the table, the traits are divided into five categories:
reproductive activity (RA), duration and timing o f the life history (LH), dispersal
ability (DA), activity rhythm (AR) and variation within individuals (VI). Forty-six
comparisons were conducted independently for 18 traits in both wild and mass reared
melon flies. For ‘ R A ’ , mass reared flies were always more prolific than wild flies
in all ten comparisons. For ‘ LH ’ , mass reared flies always had shorter life history
stages than did wild flies in all 20 cases. The dispersal ability o f mass reared flies
was not superior to wild flies. Diurnal rhythm, except for the initiation time o f
mating in the day, was not affected by mass rearing in two studies, while the initia­
tion time o f mating in the day for mass reared flies w as earlier than that for wild
flies in three studies. Variations within individuals in the pre-oviposition period,
reproduction, oviposition, longevity, pre-mating period and the number o f matings
in mass reared flies were smaller than those in wild flies (Table I) [6 ].
In addition to these traits, sensitivity to space during copulation [18] and the
mating behaviour o f wild flies [19] also differed from mass reared flies, while the
mating site o f wild flies in a field cage (i.e. on leaves or fruits or stems) did not differ
from that o f mass reared flies [16].
What factors have caused these differences between wild and mass reared
flies? In a laboratory reared insect strain, in-breeding depression by random drift is
the important problem when effective population size, Ne, is small, while artificial
selection pressures are important when it is large [20]. Artificial selection pressures
are significant in SIT, which requires mass rearing o f the insect.
In our mass rearing process, three types o f artificial selection are in operation.
The first process for mass rearing was selection for flies that displayed a high fitness
value, considering the abnormally high density in an adult cage. These mass reared
flies mated about 1 h earlier than wild flies ( ‘ A R ’) [9, 15, 16] and had somewhat
decreased flight ability compared with wild flies (‘ D A ’) [12]. Also, the mating
behaviour o f mass reared flies differed from that o f wild flies [18, 19]. This could
 IAEA-SM-327/22 203

TABLE I. COMPARISON OF TRAITS OF WILD AND MASS REARED

MELON FLIES IN Д Н Е MASS REARING FACILITIES OF OKINAWA [3-17]

Resultsb

Traits3 W > M W = M W < M References

Type 'RA ’
Fecundity 0 0 5 [3-6]
% of females that laid eggs 0 0 1 [7]
Hatchability of eggs. 0 0 1 [7]
Frequency of oviposition0 0 0 1 [6]
Number of matings0 0 0 2 [6, 7]

Type ‘LH’
Development timed 1 0 0 : [8]
Pre-oviposition period 6 0 0 [3-6]
Age of peak fecundity 1 0 0 [5]
Ovarian development time 1 0 0 [9]
Post-ovipositional life span 1 0 0 [6]
Longevity 4 0 0 [5-7, 10]
Pre-mating period 5 0 0 [5-7, 9, 11]
Remating interval 1 0 0 [7]

Type ‘DA’
Flight ability 1 0 0 [12]
Dispersal ability in thé field 3 1 0 [10, 13, 14]

Type 'AR ’
Initiation of mating in a day 3 0 0 [9, 15, 16]
Diurnal rhythm 0 2 0 [9, 17]

Type ‘VI’
Variation within individuals0 4 2 0 [6]

a See text for classification of types.

b W and M indicate wild and mass reared populations, respectively. Wild flies were main­
tained on pumpkin as a larval diet, and were used in experiments of less than three genera­
tions after introduction from the field.
c Per lifetime.
d Wild flies used were six generations after introduction from the field.
e Pre-oviposition period, fecundity, frequency of oviposition, longevity, pre-mating period
and number of matings observed in a lifetime.
204 MIYATAKE and YAMAGISHI

have been caused by the first type o f selection. In the second selection process, flies
that adapted well to the artificial diet and the artificial egging devices were selected
for the next generation. The difference in reproduction ( ‘ R A ’ ) between wild and
mass reared flies may be a result o f adaptation to the artificial egging device and the
artificial diet. In the third process, flies that developed earlier in each stage were
selected for the next generation. The SIT requires a high level o f efficiency in the
production o f mass reared flies. The mass reared flies actually had a shorter develop­
ment time [8 ] and shorter pre-mating and pre-oviposition periods, but they also had
shorter longevity compared with wild flies (‘ LH’ ) [6 ].

3. PREDICTION OF CHANGE IN THE LARVAL PERIOD

The genetic bases o f the traits mentioned above (see Table I) must be clarified
in order to obtain high quality mass reared flies. If the selection intensities in the
mass rearing process and the heritability for a trait are obtained, we can predict the
degree o f change in the trait using a quantitative genetic method [21]. W e predicted
the amount o f genetic change in the larval period as depicted in our study [2 2 ].
If a selection is by truncation, the response to selection per generation is expected
to be R = ih 2 ffp [21]. About 3% o f the larvae that developed most slowly were
selected out artificially for each generation in the mass rearing procedure for the
melon fly in Okinawa, i.e. 3% was the selection intensity (i) [23]. The heritability
(h2) o f the larval period for the melon fly was estimated to be 0.2704 [24]. The
phenotypic standard deviation, ap, in the wild strain ranged from 1.20 to 1.48 d
[25]. The value calculated by the equation closely corresponded with the observed
value at the 40th generation. The results showed that the larval period o f the melon
fly was getting shorter as generations passed through our mass rearing system. We
analysed all the records o f the mass rearing o f the melon fly and found similar
changes in larval periods to our prediction in the actual mass rearing process
(Table II).

4. SELECTION EXPERIMENT FOR DEVELOPMENT TIME

The quality o f mass reared flies might be greatly influenced by characters cor­
related to the shortened larval period resulting from mass rearing. W e tried a selec­
tion experiment for development time (from egg to adult) and then examined some
correlated responses to the selection for development time.

4.1. Selection response

Eggs were collected from the mass reared strain o f the 41st generation and
about 1600 eggs (0.1 mL in volume) were placed on 110 g o f the larval medium.
 IAEA-SM-327/22 205

TABLE II. CHANGE IN THE LARVAL

PERIOD IN MASS REARED MELON FLIES
AFTER INTRODUCTION FROM THE
FIELD

Larval period (d)

Generation
(mean (SD))

2 7.47 (0.18)
4 7.20 (0.31)
6 7.01 (0.10)
8 6.80(0.15)
10 6.09 (0.15)
12 6.39 (0.23)
14 6.33 (0.10)
16 6.40 (0.11)
18 6.28 (0.10
20 6.27 (0.14)

The first 50 males and 50 females that emerged were selected to propagate the short
development time lines (S line) and the last 50 males and females were selected to
propagate the long development time lines (L line). Two replicates were conducted.
One set o f short and long lines (designated S-l and L -l) was tested and maintained
together, as was the second replicate (S-2 and L-2). The two replicates were initiated
at the same time.
There was a clear response to selection for long development time, but no
response for the short one (Fig. 1). Realized heritabilities were calculated for the first
ten generations o f selection as the regression o f the population mean on the cumula­
tive selection differential. The significance o f each regression was determined by
analysis o f variance. The realized heritabilities o f two replicates were significantly
different from zero for the L lines and divergence (Table Ш). The response to selec­
tion for development time was mainly caused by the change in the larval period
rather than the pupal period [ 8 ].

4.2. Correlated responses

Correlated responses to selection for development time were measured

(Table IV) [22, 26, 27]. Head width and wing length were measured in generation 13
206 MIYATAKE and YAMAGISHI

Generation

FIG. 1. Response to selection for short and long development times ( © : first replicate;
о; second replicate).

for the S line and in generation 11 for the L line. We found significant correlated
responses for head width and wing length to selection for development time. The size
o f the L lines was larger than the size o f the S lines (P < 0.01, A N O V A 1). In
Drosophila melanogaster, the line selected for larger flies always took longer to
develop [28]. Total fecundity, peak fecundity, longevity, number o f matings and the
pre-mating periods were measured in generation 6 for the S line and in 5 for the
L line. Total fecundity, peak fecundity and longevity did not correlate with develop­
ment time, in spite o f the size differences in the flies o f both lines (not significant,
two way AN O VA; see Ref. [29]). The number o f matings and the pre-mating period
did correlate with the development time, but the interaction between the selection
regime and replicates was also significant (P < 0.01, two way ANOVA). The
mating time in the day was measured in generation 19 for the S line and in 15 for
the L line. The mating time in the day o f the S lines was earlier than in the L lines
(P < 0.05, Kruskal-Wallis test).

1 A N O V A : Analysis of variance.
 IAEA-SM-327/22 207

TABLE Ш. REALIZED HERITABILITIES CALCULATED AS THE REGRES­

SION OF THE POPULATION MEAN ON THE CUMULATIVE SELECTION
DIFFERENTIAL FOR THE FIRST TEN GENERATIONS

Short line Long line Divergence

Replicate 1
Male -0.0761 0.3057*** 0.1716***
Female -0.0610 0.3257*** 0.1822***

Replicate 2
Male -0.0424 0.3781*** 0.2072***
Female -0.0041 0.3719*** 0.2126***

*** P < 0.001 (ANOVA).

5. CORRELATED RESPONSES BETWEEN TRAITS

The correlated responses o f the melon fly to artificial two way selection have
been studied in four cases (Table IV). In two selection experiments conducted for
early and late reproduction, there were correlated»responses to the traits in early and
late fecundity, sexual maturation and mating time in the day. There were no cor­
related responses to the traits in total fecundity, flight ability, flight velocity and the
beginning o f wing vibration by males in the day. Different results for longevity were
obtained from these two experiments [26, 27].
. In two selection experiments conducted for pre-adult developmental time,
there were correlated responses to the traits o f size and mating time in the day. There
were no correlated responses to the trait in fecundity, number o f lifetime matings,
the pre-mating period, or longevity [22, 27]. The evidence that mass reared flies
mated about 1 h earlier than wild flies [8 , 14, 15] should be related to this correla­
tion. The difference in the change in mating time may be a mixed result o f the abnor­
mally high adult density in the egg collection cage and the correlated response to
artificial selections for early reproduction or short developmental periods in mass
rearing. These three selection experiments, except for development time, have no
replications. Thus Nakamori [26] and Suenaga [27] cannot rule out the possibility
that the results were affected by mutation or genetic drift. More systematic selection
experiments are needed to elucidate the nature o f the correlated responses between
traits in the melon fly.
208 MIYATAKE and YAMAGISHI

TABLE IV. CORRELATED RESPONSES TO TWO W A Y SELECTED TRAITS

IN THE MELON FLY

, . _ , . . . Correlated
Selected traits Reference Assayed traits
response

Reproduction, [26] 1. Total fecundity No

early versus late 2. Male longevity No
3. Female longevity No
4. Initiation time in
sexual maturation Early < late
5. Flight ability No
6 . Flight velocity No
7. Beginning of wing
vibration by male
in the day No

Reproduction, [27] 1. Early fecundity Early > late

early versus late 2. Late fecundity Early < late
3. Longevity Early < late
4. Mating time in the day Early < late

Larval period, [27] 1. Mating time in the day

short versus long Fast < slow2

Development time, [22] 1. Fecundity No

short versus long 2. Number of matings No
3. Pre-mating period No
4. Male longevity No
5. Female longevity No
6. Head width Fast < slow
7. Wing length Fast < slow
8 . Mating time in the day Fast < slow

a A symbol of < shows that there was only a little difference between both strains.

6 . APPLICATION OF CORRELATED RESPONSES

TO QUALITY CONTROL

The longevity o f the mass reared melon flies has gradually decreased [30].
However, in Drosophila melanogaster, the decreased longevity, which has a genetic
correlation with the onset o f reproduction, can be recovered by selection for late
 IAEA-SM-327/22 209

TABLE V. DIFFERENCE IN SURVIVAL RATE OF THE STRAIN SELECTED

FOR LATE REPRODUCTION AND THE NON-SELECTED STRAIN OF MASS
PRODUCED MELON FLIES

Strains
(mean ± standard deviation)

Selected for Mann-Whitney

Generation Non-selected , , .
late reproduction U test

34 0.30 ±0.11 0.43 ± 0.09 P < 0.05

(5)a (6)

35 0.42 ± 0.04 0.51 ± 0.06 P < 0.05

(5) (6)
. 36 0.40 ± 0.08 0;48 ± 0.09 P = 0.07.
(5) (ID

37 0.32 ± 0.06 0.49 ±0.11 P < 0.05

(5) (6)

38 , 0.29 ± 0.01 0.38 ± 0.07 P < 0.05

(5) (6)

39 0.25 ± 0.08 0.45 ± 0.04 P < 0.01

(5) (6)

40 0.27 ± 0.09 0.40 ± 0.07 P < 0.05

(3) (6)

a Number of populations assayed for the survival rate.

fecundity (Refs [31, 32], but see Ref. [33]). To introduce such a selection technique
to our mass rearing o f the melon fly, we divided the mass reared flies into two strains
after 34 generations. In the first strain, flies were derived from eggs collected from
the second to the sixth week after adult emergence (non-selected strain). In the
second strain, flies were derived from eggs collected from older adults only five to
six weeks after emergence (selected for late reproduction strain) [30]. The survival
rate at the tenth week in the selected strain was significantly longer than the non-
selected one, except at the 36th generation (Table V). Thus, we can easily control
one o f the life history traits by changing the timing o f egg collection.
210 MIYATAKE and YAMAGISHI

7. ACTIVE QUALITY CONTROL

What is the important trait o f mass reared insects for successful SIT? It is the
ability o f the mass reared males to mate with wild female flies in the field. In order
to accomplish this, we must select for males with increased flight ability, providing
the capability o f dispersing and attracting wild female flies. If wild females will mate
more frequently with our mass reared mâles, rather than with wild males, the quality
control o f the mass reared flies based on artificial selection will become ‘ active’ . The
actual selection process is, however, complicated.
We should search for the genetic correlation between the traits controlling
mating behaviour and the traits that can be easily selectable in mass rearing. An
example o f these traits is life history. The correlations between longevity and flight
ability found in Drosophila melanogaster [34] may be a valuable candidate for the
SIT process.
We were able to control change in a trait by using the response o f a correlated
trait. As o f now, there is only one successful case in our mass rearing o f the melon
fly. To use the direct and correlated responses to artificial selection in mass rearing
for active quality control, we must accumulate information on many genetic
parameters, i.e. selection intensity, additive variance, heritability, genetic correla­
tion, genotype environment correlation, etc. There are two main tools to obtain these
parameters: parent/offspring regression analysis and family analysis [21]. W e can
predict the change caused by selection in any trait using these parameters.
W e hope this report will be the first step in active quality control based on
quantitative genetic aspects and that this technique can be easily incorporated into
future SIT projects.

ACKNO W LED GEM ENTS

The authors thank K. Kawasaki for a critical review o f the manuscript and
E. Kasuya and B.E. Spix for their comments on the manuscript.

REFERENCES

[1] Y A M A G I S H I , M., et al., Paper IAEA-SM-327/4, these Proceedings.

[2] M A T S U I , М., N A K A M O R I , H., K O H A M A , T., N A G A M I N E , Y., The effect of
male annihilation on a population of wild melon flies, Dacus cucurbitae Coquillett
(Diptera: Tephritidae) in northern Okinawa, Jpn J. Appl. Entomol. Zool. 34 (1990)
315-317 (in Japanese with an English sumihary).
[3] N A K A M O R I , H., K A K I N O H A N A , H., S O O E M O R I , H., Method of mass rearing
melon flies, Dacus cucurbitae Coquillett (Diptera: Tephritidae). 3. Mass collection of
eggs, J. Okinawa Agrie. 14 (1976) 1-5 (in Japanese with an English summary).
 IAEA-SM-327/22 211

[4] S U G I M O T O , A., Egg collection method in mass rearing of the melon fly, Dacus
cucurbitae Coquillett (Diptera: Tephritidae), Jpn J. Appl. Entomol. Zool. 22 (1978)
60-67 (in Japanese with an English summary).
[5] S O E M O R I , H.,N A K A M O R I , H.,Production of successive generations of a new strain
of the melon fly, Dacus cucurbitae Coquillett (Diptera: Tephritidae) and reproductive
characteristics in mass rearing, Jpn J. Appl. Entomol. Zool. 25 (1981) 229-235 (in
Japanese with an English summary).
[6] N A K A M O R I , H., Variation of reproductive characters in wild and mass-reared melon
flies, Dacus cucurbitae Coquillett (Diptera: Tephritidae), Jpn J. Appl. Entomol. Zool.
31 (1987) 309-314 (in Japanese with an English summary).
[7] K U B A , H., S O E M O R I , H., Characteristics of copulation duration, hatchability of eggs
and remating intervals in the melon fly, Dacus cucurbitae Coquillett (Diptera: Tephriti­
dae), Jpn J. Appl. Entomol. Zool. 32 (1988) 321-324 (in Japanese with an English
summary).
[8] M I Y АТАКЕ, T., “Response to artificial selection for development time in the melon
fly”, Biology and Control of Fruit Flies (Proc. Int. Symp. Okinawa, Japan 1991)
( K A W A S A K I , K., I W A H A S H I , O., K A N E S H I R O , K.; Eds), University of the
Ryukus, Okinawa, Japan (1991)-317-321.
[9] SUZUKI, Y., K O Y A M A , J., Temporal aspects of mating behaviour of the melon fly,
Dacus cucurbitae Coquillett (Diptera: Tephritidae): A comparison between laboratory
and wild strains, Appl. Entomol. Zool. 15 (1980) 215-224.
[10] N A K A M O R I , H., S O E M O R I , H., Comparison of dispersal ability and longevity for
■ ■wild and mass-reared melon flies, Dacus cucurbitae Coquillett (Diptera: Tephritidae),
under field conditions, Appl. Entomol. Zool. 16 (1981) 321-327.
[11] S O E M O R I , H., Difference in mating occurrence between mass-reared and wild strains
of the melon fly, Dacus cucurbitae Coquillett, Bull. Okinawa Agrie. Exp. Stn. 5 (1980)
69-71 (in Japanese with an English summary).
[12] N A K A M O R I , H., SIMIZU, K., Comparison of flight ability between wild and mass-
reared melon fly, Dacus cucurbitae Coquillett (Diptera: Tephritidae), using a flight
mill, Appl. Entomol. Zool. 18 (1983) 371-381.
[13] K A K I N O H A N A , H., S O E M O R I , H., N A K A M O R I , H., “Dispersal ability of mass-
reared melon fly adults under field conditions”, Report of the Eradication Program of
the Melon Fly in Okinawa Prefecture (1977) 23-24 (in Japanese).
[14] S O E M O R I , H., K U B A , H., Comparison of dispersal ability among one wild and two
mass-reared strains of the melon fly, Dacus cucurbitae Coquillett (Diptera: Tephriti­
dae), under field conditions, Bull. Okinawa Agrie. Exp. Stn. 8 (1983) 37-41 (in
Japanese with an English summary).
[15] K U B A , H., K O Y A M A , J., Mating behaviour of the melon fly, Dacus cucurbitae
Coquillett (Diptera: Tephritidae): Comparative studies of one wild and two laboratory
strains, Appl. Entomol. Zool. 17 (1982) 559-568.
[16] K O Y A M A , J., N A K A M O R I , H., K U B A , H., Mating behaviour of wild and mass-
reared strains of the melon fly, Dacus cucurbitae Coquillett (Diptera: Tephritidae), in
a field cage, Appl. Entomol. Zool. 21 (1986) 203-209.
212 M IYАТАКЕ and YAMAGISHI

[17] K A K I N O H A N A , H., “Qualitative change in the mass-reared melon fly, Dacus cucur­
bitae Coq.”,Fruit Fly Problems (Proc. Symp. Kyoto and Naha, Japan, 1980), National
Institute of Agricultural Science, Yatabe, Ibaraki (1980) 27-36.
[18] S O E M O R I , H., T S U K A G U C H I , S., N A K A M O R I , H., Comparison of mating ability
and mating competitiveness between mass-reared and wild strains of the melon fly,
Dacus cucurbitae Coquillett (Diptera: Tephritidae), Jpn J. Appl. Entomol. Zool. 24
(1980) 246-250 (in Japanese with an English summary).
[19] HIBINO, Y., I W A H A S H I , O., Mating receptivity of wild type females for wild type
males and mass-reared males in the melon fly, Dacus cucurbitae Coquillett (Diptera:
Tephritidae), Appl. Entomol. Zool. 24 (1989) 152-154.
[20] C R O W , J.F., K I M U R A , М., A n Introduction to Population Genetics Theory, Harper
and Row, N e w York (1970).
[21] F A L C O N E R , D.S., A n Introduction to Quantitative Genetics, 3rd edn, Wiley,
N e w York (1989).
[22] M I Y АТАКЕ, T., Okinawa Prefectural Agricultural Experiment Station, Okinawa
Prefectoral Fruit Fly Eradication Project Office, Naha, Okinawa (in preparation).
[23] N A K A M O R I , H., K A K I N O H A N A , H., Mass production of the melon fly, Dacus
cucurbitae Coquillett, in Okinawa, Japan, Rev. Plant. Prot. Res. 13 (1980) 37-53.
[24] S U E N A G A , H., Kagoshima Prefectural Agricultural Station, Kagoshima, Japan,
personal communication.
[25] M I Y АТАКЕ, T., Okinawa Prefectural Agricultural Experiment Station, Naha,
Okinawa, Japan, unpublished data.
[26] N A K A M O R I , H., Behavioural and ecological studies on sexual competitiveness in the
melon fly, Dacus cucurbitae Coquillett, mass production, Bull. Okinawa Pref. Agrie.
Exp. Stn. 2 (1988) 1-64 (in Japanese with an English summary).
[27] S U E N A G A , H., “Correlated responses of mating behaviour of Dacus cucurbitae to
selections for early and late reproduction or short and long larval period”, Biology and
Control of Fruit Flies (Proc. Int. Symp. Okinawa, Japan, 1991) ( K A W A S A K I , K.,
I W A H A S H I , O., K A N E S H I R O , K., Eds), University of the Ryukus, Okinawa Prefec­
ture, Okinawa, Japan (1991) 327-330.
[28] H I L L E S H E I M , E., S T E A R N S , S.C., The responses of Drosophila melanogaster to
artificial selection on body weight and its phenotypic plasticity in two larval food
environments, Evolution 45 (1991) 1909-1923.
[29] H I L L E S H E I M , E., S T E A R N S , S.C., Correlated responses in life history traits to
artificial selection for body weight in Drosophila melanogaster, Evolution 46 (1992)
745-752.
[30] K A K I N O H A N A , H., Y A M A G I S H I , М., “The mass production of the melon fly —
techniques and problems”, Biology and Control of Fruit Flies (Proc. Int. Symp.
Okinawa, Japan, 1991) ( K A W A S A K I , K., I W A H A S H I , O., K A N E S H I R O , K., Eds),
University of the Ryukus, Okinawa Prefecture, Okinawa, Japan (1991) 1-10.
[31] L U C K I N BI L L , L.S., A R K I N G , R., C L A R E , M.J., C I R O C C O , W.C., B U C K , S.A.,
Selection for delayed senescence in Drosphila melanogaster, Evolution 38 (1984)
996-1003.
[32] L U C KI N B I L L, L.S., C L A R E , M.J., Selection for life span in Drosophila melano­
gaster, Heredity 55 (1985) 9-18.
 IAEA-SM-327/22 213

[33] P A R T R I D G E , L., F O W L E R , K., Direct and correlated responses to selection on age

at reproduction in Drosophila melanogaster, Evolution 46 (1992) 76-91.
[34] G R A V E S , J.L., L UC K I N B I L L , L.S., N I C H O L S , A., Flight duration and wing beat
.frequency in long and short-lived Drosophila melanogaster, J. Insect Physiol. 34 (1988)
1021-1026.
 IAEA-SM-327/23

SEARCH FOR SEX SPECIFIC GENES

IN THE MEDFLY, Ceratitis capitata
Preliminary data on Sxl

M. FURIA, D. ARTIACO, G. SACCONE,

P. VITO, I. PELUSO, E. GIÔRDANO, L.C. POLITO
Dipartimento di Genetica Biologia Generate e Molecolare,
Facoltà di Scienze,
Université di Napoli,
Naples, #
Italy

Abstract

S E A R C H F O R S E X SPECIFIC G E N E S IN T H E M E D F L Y , Ceratitis capitata: P R E L I M I ­

N A R Y D A T A O N Sxl.
The genes involved in sex determination seem to be among the most valuable for
developing efficient biological control strategies based on the sterile insect technique (SIT) in
Ceratitis capitata and in other insects of economic importance. In addition, these genes are
of primary interest in studies of the basic biology and genetics of systems such as C. capitata
that are relatively little characterized. To examine both of these topics, Drosophila was chosen
as the model system and the, strategy, was to utilize probes derived .from Drosophila
melanogaster genes to isolate by cross-hybridization the homologous Ceratitis genes. The first
intent was to develop rapidly a good basis of the molecular genetics of the sex determining
genes, as a preliminary step in achieving the proposed goals. To start, sex lethal (Sxl) was
chosen, the key control gene involved in sex differentiation and dosage compensation of the
X linked genes. In a second set of experiments, a female specific gène, recently identified in
Drosophila, was used to probe C. capitata. Preliminary data are reported here on thé identifi­
cation of a C. capitata gene homologous to the Drosophila Sxl. Using as a probe a D N A sub­
fragment derived from the Drosophila femále Sxl c D N A clone, an adult C. capitata female
c D N A library was screened and a clone isolated that, on the basis of nucleotide and'amino
acid sequence analysis, was found to have originated from a C. capitata gene homologous to
the Drosophila Sxl. Some observations are also presented on the molecular characterization
in C. capitata of the function corresponding to the Drosophila female specific gene.

1. INTRODUCTION

The Mediterranean fruit fly (medfly), Ceratitis capitata, is an extremely

important pest o f fruit and vegetable crops. Control o f this pest has been historically
based on the use o f insecticidal sprays and, more recently, on the sterile insect
technique (SIT).

215
216 FUMA et al.

To overcome some limitations connected with SIT. application, namely to

obtain better biological efficiency and reduce costs, it is necessary to release only
sterilized males. To attain these results, the development o f suitable genetic sexing
strains is required, and there are two potential approaches to achieve elimination at
the early developmental stages o f females from a mass reared medfly culture: the
first involves classical genetics [ 1 ] and the second involves the application o f recom­
binant DNA technology [2].
Our group is participating in an internàtional project focused on the develop­
ment o f a germ line transformation system for the medfly and is presently engaged
in the search for sex specific genes in C. capitata [3].
The isolation o f such genes is, in our project, a preliminary but essential step
to building up hybrid DNA constructs that, once the transformation is operating in
this organism, could allow female specific lethality [2]. On the other hand, these
genes are also intrinsically o f primary interest for studies o f the basic biology and
genetics o f systems which have been characterized relatively little, such as
C. capitata. The impressive recent advances in the molecular genetics o f another
fruit fly, Drosophila melanogaster, where the genetic and molecular bases o f sex
determination have been largely elucidated, make this insect a perfect model system
from which to transfer the know-how to insects o f economic importance such as
C. capitáta, which diverged from D. melanogaster about 125 million years ago. For
all these reasons, and to isolate C. capitata sex specific genes, we chose the strategy
o f cross-hybridization with probes derived from Drosophila genes. The first gene
used was sex lethal (Sxl), the key control gene in sex differentiation (reviewed in
Ref. [4]) and dosage compensation in D. melanogaster (reviewed in Ref. [5]).
The process o f sex determination in Drosophila involves a cascade o f regula­
tory genes that connects the primary sex determining signal (the X :A ratio) to the
terminal differentiation genes whose products are responsible for the sexual
dimorphic characteristics o f the adult. The special feature o f this regulatory cascade
is that most o f the regulatory interactions are based on sex specific splicing o f the
primary transcripts common to both sexes. The most recent molecular findings are
concerned with four genes in this regulatory pathway: sex lethal (Sxl), transformer
(ira), transformer-2 (tra-2) and doublesex (dsx). Sxl is the gene that responds
directly to the primary sex determining signal and is active only in females. It has
been shown that the Sxl transcripts are different in the two sexes. Without entering
into details, the male specific transcript carries an exon that is absent in the female
specific one, and this extra exon contains an in-frame stop codon which leads to the
production o f a truncated, ‘ function-less’ polypeptide. Moreover, the Sxl female
protein has RNA binding properties. This protein has recently been shown to bind
specifically to its own transcript, thus autocatalysing the female specific pattern o f
splicing and maintaining the gene in the active female mode [6 , 7]. In addition to
its own regulation, Sxl controls the activity o f the somatic sex determining genes
entirely through its control on the ira gene. As with Sxl, the ira gene is active only
 IAEA-SM-327/23 217

in females and its activity is similarly regulated at the level o f RNA splicing. Analo­
gous mechanisms based on sex specific alternative splicing ensure the production and
function o f the other two relevant genes in the sex cascade, tra-2 [8 ] and dsx [9].
The genetic programme specifying the sexual fate is then achieved through the
activation, mediated by the dsx gene products, o f different sets o f structural genes
in males and females.
In this paper, we mainly report the isolation, by cross-hybridization with
Drosophila probes, o f a C. capitata gene homologous to Sxl.

2. MATERIALS AND METHODS

All methodologies and buffers not specifically described were carried out
according to Sambrook et al. [10].

2.1. Labelling of DNA probes

For Northern and Southern analyses and for screening o f libraries, the DNA
probes consisted o f DNA fragments isolated from 5% acrylamide gel and labelled
to a specific activity o f 5 X 108 (counts/min)//¿g with a Multiprime labelling kit
(Amersham).

2.2. cDNA library

The total C. capitata RNA was extracted from female adults according to the
guanidinium/CsCl method. The poly (A )+ RNA was purified with the Stratagene
poly (A )+ Quick™ mRNA purification kit.
A cDNA library from C. capitata female adults poly (A )+ RNA was pre­
pared in Uni-Zap™ X R using the Stratagene Zap-cDNA synthesis kit. The titre o f
the non-amplified library was 2.8 x 106 plaque forming units (PFU)/mL. The titre
o f the amplified library was 1 X 10 10 PFU/mL.

2.3. Screening of cDNA libraries and Southern analyses

In screening an embryonic genomic library (kindly provided by the colleagues

o f the Insect Group o f the University o f Crete) o f the cDNA library and in Southern
analyses, we utilized low stringency conditions: 55°C in 5X SSPE, 5X Denhardt’ s
solution, 0.5% SDS and salmon sperm DNA, with washes at 55°C in 2X SSC,
0.1% SDS.
218 FURIA et al.

2.4. Northern analyses

The RNA was fractionated on 1.2% agarose formaldehyde gel, transferred to

a nylon membrane (Hybond N, Amersham), hybridized and washed following the
protocols suggested by Amersham. Low stringency hybridizations were conducted
at 37°C in 50% formamide, 5X SSPE, 5X Denhardt’ s solution, 0.5% SDS and
salmon sperm DNA, with washes at 42°C in 2X SSC, 0.1% SDS.

2.5. DNA sequencing and analysis

The nucleotide sequences were determined by the dideoxy method. Sequencing

data were analysed using the FASTA analysis program [11].

3. RESULTS

Two C. capitata libraries, an embryonic genomic library and a cDNA library

from female adults, constructed in the phage vectors EMBL4 and Uni-Zap™ XR,
respectively, were screened at low stringency for homologues o f the Sxl gene o f
D. melanogaster using as a probe a subfragment derived from the Drosophila female
'CDNA (fragment A in Fig. 1, panel (b)) [6 ]. Several recombinant clones were iso-
■lated from the genomic library, but were all ‘ false positive’ to the final control o f
the sequence analysis, and all o f them were shown to be relatively within reach o f
TA or CA stretches. Similar results were obtained in another screening carried out
utilizing the entire cDNA as a probe [12]. Actually, the choice to utilize the subfrag­
ment was made on the basis o f the observation that the Sxl cDNA itself contains some
(TA )n or (CA )n repeats and because o f the notion that working at low stringency
could be responsible for the false positive signals. The subfragment, however, is
relatively free o f such sequences. In the cDNA screening, approximately 1 X 106
phage plaques were plated and four positive independent clones were isolated. The
four clones were restricted with the enzymes EcoRI and Xhol to extract the inserts
and subjected to Southern hybridization with the same,probe. Three clones contain
an insert o f 1.2 kb and the fourth o f only 0.6 kb. At the moment, we have character­
ized one o f these clones, SxlCf\ at 1.2 kb. A restriction map was constructed and
sequence analysis was conducted; the nucleotide and amino acid sequences were
compared with the corresponding sequences o f the male and female cDNA
Drosophila clones utilizing the FASTA program for scoring similarities and
constructing alignments (data not shown) [ 1 1 ].
In Fig. 1, we show a schematic representation o f the Ceratitis clone and the
. female and male SxUDrosophila cDNAs.
The major difference between the two Drosophila cDNA clones is the presence
o f an extra 190 bp exon (exon 3) in the male cDNA. The structure and the nucleotide
 IAEA-SM-327/23 219

1 kb (a )

XP H P H X P XX H H
II I _ l_ _J____ u I I
AUG UGA
2t 3t 4 56 7 8
MALE 5'

RBD1 R B D 2

FEMALE 4 167!L
4 *
AUG UGA

0.1 kb (b)

AUG
HAAAA

EXON 3

AAAA

5 6 7 8 EXONS
FRAGMENTA NDE1

□ 45-55 % Ш 55-65 % 0 65-75 % 75-85 % 85-95 %

F/G. 7. (a) Genomic map о/Sxl in D. melanogaster, with a schematic diagram illustrating
the approximate positions o f the exons present in adult.male and female specific transcripts
[6]. The open readingframes are indicated by arrows. RBD1 and KBD2 indicate the two RNA
binding domains present in the Sxl Drosophila protein; they extendfrom exon 5 to exon 8; the
brackets limit the first and the second domain, respectively. The restriction sites are: X, Xho,
I; H, Hindlll; P, Pstl. (b) A comparison between the D. melanogaster Sxl cDNA (middle)
and the homologous cDNA in C. capitata (top). The boxes on the Drosophila cDNA illustrate
the different percentages o f homology. Fragment A (bottom) was used to screen the adult
female cDNA library.
220 FURIA et al.

TABLE I. BASE UTILIZATION IN RNA

BINDING DOMAIN 2

Drosophila Ceratitis

207 bp

48 A ' 72 A
58 С 47 C
59 G 44 G
42 T 44 T

Base at position IIIa

8.5% A 36.1% A
33.1% С 24.3% C
36% G 15.7% G
21.4% T 23% T

G + С contents3

С + G 69.1% C + G 40%

a The percentages are approximated to the first

decimal value.

and amino acid sequences o f the Ceratitis cDNA clone appear to be highly conserved
with respect to the Drosophila femalë cDNA.
The overall nucleotide sequence homology is about 58 %, while the amino acid
homology is about 62% ; these are, in absolute, very significant values. The sequence
corresponding to the male exon 3 is not present in the capitata cDNA that, as in the
Drosophila female, contains a single long open reading frame that starts at an
AUG codon in a region corresponding to the Drosophila exon 2. Sequences homolo­
gous to exons 4, 5, 6 , 7 and 8 are present in the same order and are more or less
the same length. The codon usage in this open reading frame agrees with the
C. capitata codon bias [13].
Table I shows the base composition in one o f the two RNA binding domains,
RBD2, present in the D. melanogaster. female Sxl cDNA and in the corresponding
region o f the C. capitata clone (Fig. 1, panel (a)). .
In the latter clone, the coding region is preceded by a 5' untranslated leader
o f 108 bp, instead o f the 482 bp present in both the female and male Drosophila
clones. In the Drosophila cDNA clone, the termination codon UGA is in exon 8 and
is followed by 217 bp, ending with 3 A [6 ]. In our clone, after the stop codon there
are only 6 bp followed by 25 A.
 IAEA-SM-327/23 221

4. DISCUSSION AND CONCLUSIONS

The evidence presented here indicates that we have isolated a cD NA derived

from the C. capitata gene homologous to the Drosophila Sxl gene. This clone was
selected from an adult female cDNA library and appears to be related to the
Drosophila female specific transcript, since any sequence corresponding to the
Drosophila male specific exon 3 is absent.
As reported in Section 3, the overall nucleotide and amino acid sequence
homologies are significantly high (58 and 62% , respectively), orders o f magnitude
that prove an identity o f function in the two organisms. Or, more precisely, indicate
a very conserved function in the two species, even though they have been separated,
as we said, by more or less 125 million years. Indeed, a recent analysis o f the evolu­
tionary divergence o f the transformer gene in Drosophila gives lower values o f
nucleotide and amino acid homologies between the sequences in D. melanogaster
and D. hydei, even though they diverged only 60 million years ago [14].
We do not yet have any idea o f the genomic organization o f the Sxl gene in
Ceratitis, but the fact that the dimensions o f the inserts o f three out o f four clones
we have isolated are 1 .2 kb seems to us to indicate, with a good probability, that
these are the dimensions o f the transcript in C. capitata.
If this hypothesis is true, then the main differences between the female
Drosophila and Ceratitis transcripts are in the 5' and 3' untranslated regions.
As regards the 5 ' end, and again on the basis o f the insert dimension, we think
that our clone is complete and that in Capitata there is a shorter untranslated leader.
With respect to the 3' end, the clone ends with a relatively long poly A tail and there
is a clear poly adénylation consensus signal.
A very important point concerns the topological analysis o f the homology
patterns. The reported values for the nucleotide and amino acid sequences are aver­
age values but, as shown in Fig. 1, panel (b), there is basically a symmetric distribu­
tion in the value o f nucleotide contents.
As regards the nucleotide sequences, from the 5' to 3' ends o f the clone the
values range from 45-55 %, to a maximum o f 85-95 %, more or less in the middle
o f the clone, and end at a range between 55 and 65% . The same pattern, with higher
values, is obtained for the amino acid sequence values, i.e. in the region o f the two
RNA binding domains there are long stretches with more than 95% homology.
It is interesting to note that the codon bias in the two genomes is very different.
As given in Table I, even in a region so highly conserved in terms o f amino acid
sequences as RBD2, the usage o f bases in position П1 strongly differs, while the
G + С content is 69.1% in D. melanogaster and only 40% in C. capitata [13].
The cDNA sequences and a more detailed analysis, together with a comparison
o f the genomic organization o f the Sxl in the two insects, will be published elsewhere
(manuscript in preparation), but we would, like to stress here the very high degree
o f homology between the Drosophila and Ceratitis Sxl gene and its topological
distribution.
222 FURIA et.al.

In the products o f the Drosophila Sxl gene, two RNA binding domains, RBD1
and RBD2, are present. These domains are crucial in the process o f sex specific
splicing o f Sxl itself [6 ] and tra [ 8 ]. In conclusion, the fact that this part o f the gene
and the gene product are so highly homologous in Drosophila and Ceratitis is confir­
mation o f the functional relevance in evolutionary terms. In addition, all these obser­
vations strongly suggest that, even if the chromosomal basis o f sex determination
differs between these two insects [3, 15], the molecular regulatory pathways should
be extensively conserved.
The cloning o f the C. capitata Sxl gene will therefore provide the immediate
possibility o f studying the molecular bases o f sex determination in this organism and
checking if here also the sex regulatory cascade is based on the differential splicing
o f primary transcripts common to both sexes.
Once transformation in C. capitata is achieved, the introduction o f the
homologous Sxl. minigene in a construct that allows its inducible expression as
antisense RNA should permit selective elimination o f females from the mass reared
culture at an early stage o f development [2 ].
The antisense RNA will in fact interact with the Sxl transcripts both in female
and male insects, inducing loss o f function, [2 ], but because only the females
need it to maintain the sex pathway o f the Sxl gene, they will be the only ones
affected [4, 6 , 7].
At the same time, for the reasons presented in Section 1, we are also interested
in genes involved in a more terminal sex differentiation event, i.e. oocyte develop­
ment. Nucleotide and amino acid sequences o f genes involved in the ovarian specific
functions are also more likely to be conserved, and, in fact, D. melanogaster and
C. capitata chorion [16] and vitellogenin [13] gene sequences show good homology.
Recently, our group reported in D. melanogaster the identification and the
initial molecular characterization o f a function expressed at high levels during
embryogenesis and oogenesis [17], as visualized by in situ hybridization to whole
mount preparation o f ovaries and embryos [18]. In Northern blots o f these stages a
single transcript o f 2.6 kb was revealed. From the patterns o f expression we
concluded that this gene is involved in establishing the egg morphology and in germ
line formation. Again, using a subfragment derived from a D. melanogaster cDNA
as probe, we identified and partially characterized, at the molecular level, a
corresponding function in Ceratitis.
On a Southern blot o f genomic Ceratitis DNA, restricted with .EcoRI and
Hindin, two bands o f 3.5 and 2.7 kb are visible in the EcoRI track and four (3.4,
2,9, 2.1 and 1.8 kb) in the Hindlll track.
At the RNA level, the Drosophila probe identifies at least two transcripts o f
about 1.3 and 2.2 kb, respectively, present at different stages o f Ceratitis
development.
As a direct approach towards the isolation o f C. capitata sex specific genes,
we have prepared Ceratitis cDNA libraries from adult males and females, and
 IAEA-SM-327/23 223

ovaries and testes, and we will proceed to identify, by subtractive analysis, clones
expressed in a sex specific manner.
From the data we have shown, we think it is possible to conclude that the cross­
hybridization approach, Drosophila versus Ceratitis, to isolate genes o f the' latter
organism is still in certain contexts (i.e. highly conserved genes utilizing cDNA
libraries) a valid, if not general, strategy.

ACKNOWLEDGEMENTS

This research was supported by the National Research Council o f Italy Special
Project RAISA, Subproject N.2 Paper N536. D. Artiaco has received a RAISA
Training Fellowship.

REFERENCES

[1] R O B I N S O N , A.S., “Genes for genetic sexing in the Mediterranean fruit fly, Ceratitis
capitata (Wied.), and the mosquito, Anopheles stephensi” , Genetic Sexing of the
Mediterranean Fruit Fly (Proc. Final Research Co-ordination Meeting Colymbari,
Crete, 1988), IAEA, Vienna (1990) 189-198.
[2] R O B I N S O N , A.S., SAVAKIS,.C., LOUIS, C., “Status of molecular genetic studies
in the medfly Ceratitis capitata in relation to genetic sexing”, M o d e m Insect Control:
Nuclear Techniques and Biotechnology (Proc. Symp. Vienna, 1987), IAEA, Vienna
(1988) 241-250.
[3] LIFSCHITZ, E., C L A D E R A , J., “Cytogenesis and sex determination in Ceratitis
capitata” , Fruit Flies and their Biology: Natural Enemies and Control
(ROB IN S O N , A.S., H O O P E R , G.H.S., Eds), Elsevier, Amsterdam (1989).
[4] S T E I N M A N N - Z W I C K E Y , M . , A M R E I N , H . , N O T H I N G E R , R., Genetic control of
sex determination in Drosophila, Adv. Genet. 27 (1990) 189-237.
[5] L U C C H E S I , J.C., M A N N I N G , J.E., Gene dosage compensation in Drosophila
melanogaster, Adv. Genet. 24 (1987) 371-429.
[6] BELL, L.R., M A I N E , E.M., S C H E D L , P., CLINE, T.W., Sex-lethal, a Drosophila
sex determination switch gene, exhibits sex-specific R N A splicing and sequence
similarity to R N A binding proteins, Cell 55 (1988) 1037-1048.
[7] BELL, L.R., H O R A B I N , J.I., S C H E D L , P., CLINE, T.W., Positive autoregulation
of Sex-lethal by alternative splicing maintains the female determined state in
Drosophila, Cell 65 (1991) 229-239.
[8] S O S N O W S K I , B.A., B E L O T A , J.M., M c K E O W N , М., Sex-specific alternative
splicing of R N A from the transformer gene results from sequence-dependent splice site
blockage, Cell 58 (1989) 449-459.
[9] BURTIS, K.C., B A K E R , B.S., Drosophila double-sex gene controls somatic sexual
differentiation by producing alternatively spliced m R N A encoding related sex-specific
polypeptides, Cell 56 (1989) 997-1010.
224 FURIA et al.

[10] S A M B R O O K , J., FRITSCH, E.F., M A N I A T I S , T., Molecular Cloning: A Labora­

tory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N Y (1989).
[11] P E A R S O N , W.R., L I P M A N , D.J., Improved tools for biological sequence compari­
son, Proc. Natl Acad. Sci. U S A 85 (1988) 2444-2448.
[12] S I D EN - K I A SM O S , I., et al., Opa-like repeats in the genome of Medfly, Ceratitis
capitata (submitted for publication).
[13] RINA, М., SAVAKIS, C., A cluster of vitellogenin in the Mediterranean fruit fly
Ceratitis capitata: Sequence and structural conservation in dipteran yolk proteins and
their genes, Genetics 127 (1991) 769-780.
[14] O’ NEIL, M.T., B E L O T E , J.M., Interspecific comparison of the transformer gene of
Drosophila reveals an unusually high degree of evolutionary divergence, Genetics 131
(1992) 113-128.
[15] Z A P A T E R , М., R O B I N S O N , A.S., Sex chromosome aneuploidy in a male-linked
translocation in Ceratitis capitata, Can. J. Genet. Cytol. 28 (1986) 161-167.
[16] K O N S O L A K I , М., etal., The chorion genes of the Medfly, Ceratitis capitata.
I: Structural and regulatory conservation of the s36 gene relative to two Drosophila
species, Nucleic Acids Res. 18 (1990) 1731-1737.
[17] POLITO, L.C., FURIA, М., Sex determination and dosage compensation in
Drosophila melanogaster, Collana U.Z.I., selected symposium and monographs (in
press).
[18] DIGILIO, A.F., etal., Molecular characterization of a new gene of Drosophila
melanogaster with maternal effect, Atti Assoc. Genet. Ital. X X X V I I I (in press).
 IAEA-SM-327/24

TRANSFECTIONS OF DNA INTO

CULTURED INSECT CELLS AND EMBRYOS
Progress and prospects*

R .A . LEOPOLD
Biosciences Research Laboratory,
Agricultural Research Service,
United States Department o f Agriculture,
Fargo, North Dakota,
United States o f America

Abstract

T R A N S F E C T I O N S O F D N A INTO C U L T U R E D INSECT C ELLS A N D E M B R YO S :

P R O G R E S S A N D PROSPECTS.
The ability to manipulate insect genomes to develop or enhance methods of control for
pest and beneficial species has been a goal of geneticists working in the areas of medical and
agricultural entomology. In vitro gene manipulation techniques combined with methods that
enable integration of exogenous D N A into a host genome have become routine procedures for
many organisms. Except for the drosophilid species, these advancements have not been
realized for other insects. A lack of functional gene vectors and the means to introduce foreign
D N A into non-drosophilid species have limited most studies to the analysis of transient expres­
sion of manipulated genes in cells and embryos of relatively few insects. The paper reviews
the current technology used to deliver D N A to cultured cells and embryos of insects, examines
the mechanisms effecting transfection and compares the relative transfection frequency of the
various methods. A discussion of possible modifications and combinations of transfection
methodology that could aid in the integration of D N A into a host embryo’ s genome is
presented.

1. INTRODUCTION

The introduction o f exogenous DNA into cultured tells o f insects has been
accomplished by a variety o f methods [ 1 ], while introduction into embryos has been
done mostly by microinjection [2]. The search for alternative DNA delivery methods
has become a goal o f many researchers, since the transfection frequency elicited by
a particular technique may vary widely among cell lines [1 ,3 ] and the microinjection
o f embryos tends to be labour intensive and inefficient [2]. While the prospects o f

* Mention of a trademark or a proprietary product does not constitute endorsement,

a guarantee or warranty of the product by the United States Department of Agriculture, and
does not imply its approval to the exclusion of other products that may be suitable.

225
226 LEOPOLD

producing transgenic insects o f economically important species for use in genetic

control programmes are encouraging, the lack o f efficient DNA vector, systems such
as the P element transposon o f Drosophila [4, 5] prevents their development.
Insect cell culture provides an easy and economical system for testing the effec­
tiveness o f transfection methods and evaluating the activity o f various promoters
combined with cloned genetic material o f interest. Since cultured cells are not
exposed to potential regulators such as hormones, electrolytes and the positional
effects when cells are organized into tissues and organs, gene expression can be more
easily analysed. Further, cell cultures provide a sufficient number o f cells when
vigorous selection schemes are used in conjunction with transformation systems
where integration occurs only rarely (e.g. gene replacement).
After gaining success in cultured cells, the DNA constructs can then be tested
in embryos using DNA delivery systems developed for whole organisms. While
microinjection o f individual insect embryos is the commonly used procedure,
development o f methods allowing concurrent delivery o f a DNA construct into large
numbers o f embryos is needed. Such methods would permit the use o f gene transfer
systems that are typically employed only in cell cultures.
Thus, the intent o f this paper is to examine the current methods being used for
the transfection o f insect cells and, using this information as background, to suggest
approaches that could.be used to increase the frequency o f transfection in insect
embryos and to decrease the amount o f labour or handling required to achieve
transfection.

2. CHEMICALLY MEDIATED TRANSFECTION

OF INSECT CELL LINES

The most common technique used to transfer foreign DNA to cultured insect
cells from the culture medium has been the calcium phosphate co-precipitation
method. Calcium phosphate is thought to aid uptake by the cells through precipitating
the DNA on the cell membrane, inducing endocytosis and protecting the internalized
DNA from degradation by endogenous nucleases [6 ]. Calcium phosphate co ­
precipitation has largely been used to transfer purified DNA to a variety o f
Drosophila cell lines [7-12], but it has also been used successfully in cultured
mosquito and lepidopteran cells'[3, 13]. Further, the temperature at which the DNA
and calcium phosphate are co-precipitated can also be an important factor in gaining
higher transfection frequencies. Preparation o f the DNA-calcium phosphate
co-precipitate at 50°C instead o f at ambient temperature gave a tenfold higher trans­
fection frequency in mosquito cells [3].
Polycations such as DEAE-dextran, polybrene, poly ornithine and poly lysine
have been used to transfect dipteran, coleopteran and lepidopteran cell lines
[1, 14-18]. The negatively charged DNA molecule forms a complex with the posi-
 IAEA-SM-327/24 227

TABLE I. COMPARISON OF THE TRANSFECTION EFFICIENCY OF

THREE CHEMICAL MEDIATORS WITH ELECTROPORATION IN THE BOLL
WEEVIL CELL LINE 3 BRL-AG-3C, WHEN USING THE hsp 70-CA T VECTOR

Mediator
Transfection D N A concentration % conversion of
concentration
mediator Oig/mL) .chloramphenicolb
(itg/mL)

Polybrene 12.5 16 <1

Polylysine 12.5 8 11.0
Lipofectin 12.5 4 75.1
Electroporation0 — 30 2.6

a Cell concentration = 1 x 107 cells/mL.

b Measured as per cent conversion of available chloramphenicol to the acetylated products.
c Cells given three pulses at 0.6 m s each, 175 V, field strength 0.9 kV/cm (data from Stiles
et al. [17]).

tively charged polycation molecule and together they bind irreversibly to the cell
membrane and are apparently internalized by phagocytosis. Occasionally, DMSO is
used to facilitate entry o f the polycation-DNA complex into the cells, but it was
found to be o f no advantage when used in transfecting Aedes albopictus or Anthono-
mus grandis cells [10, 17].
Feigner et al. [19] developed a transfection procedure using the synthetic
cationic lipid, N-[l-(2,3-dioleyoxy)propyl]-N,N,N-trimethylammonium chloride.
Transfection is accomplished through a fusion o f cells with liposomes containing the
DNA. To this author’ s knowledge, this chemical mediator, marketed under the label
Lipofectin™, has not been used extensively in the transfection o f insect cells. Stiles
et al. [17] recently compared the transfection efficiency o f the cationic liposome
method with that o f polylysine, polybrene and electroporation and found it to be far
superior in gaining transient expression o f the chloramphenicol acetyltransferase
(CAT) gene in a boll weevil cell line (Table I).

3. NON-CHEMICAL METHODS OF CELL CULTURE TRANSFECTION

The formation o f somatic cell hybrids by controlled fusion o f cells derived

from two different parental lines can serve as a method o f introducing exogenous
DNA into a cell culture. Somatic cell fusion has been performed by treating with
polyethylene glycol, inactivated Sendai virus, lectins and electrical stimuli [20, 21].
228 LEOPOLD

While cell fusion has been accomplished in Drosophila and mosquito cell lines
[21-23], the lack o f selective media needed for transformant recovery by com­
plementation o f a mutation in each parental line and problems deriving cell lines
from mutant insect strains have limited the use o f this method.
Another method that uses electric current to effect transfection o f cell cultures
is electroporation. This is a rapid procedure in which a suspension o f cells and DNA
is placed between two electrodes and given a high voltage pulse. Transient holes or
pores in the cell membrane are producèd through which the DNA is able to enter
the cell. After a short period o f time, the holes reseal and the cells resume normal
functions. Mann and King [24] used electroporation to aid entry o f a baculovirus
vector into a Spodoptera frugiperda cell line and Stiles et al. [17] obtained a low
transfection frequency in A. grandis cells when using the heat shock protein 70
(hsp70)-CAT plasmid. One o f the advantages o f this method is that internalization
o f the exogenous DNA is not dependent on the phagocytic capacity o f cells. Other
advantages include the ability to obtain integration o f low gene copy numbers, as
opposed to the large tandem arrays seen in the chemical methods [25], and to use
either plasmid or genomic DNA [26].
One o f the most effective methods o f transferring genes to lepidopteran cell
lines is by using insect viruses. The most commonly used o f the insect viruses is the
nuclear polyhedrosis baculovirus, Autographa californica (AcNPV). For a detailed
description o f how baculoviruses infect cells, enter nuclei and replicate, see the
review o f Faulkner and Carstens [27]. Baculovirus expression vectors (BEVs) are
recombinant insect viruses that have foreign genes inserted into the viral genome
under the regulation o f a strong promoter for the viral gene that produces the poly­
hedrin protein. While BEVs have been used to transfect lepidopteran cell lines with
a variety o f foreign genes [28-30], they function poorly in most other insect cell
lines. However, some success was gained in the expression o f the CAT gene linked
with a retroviral promoter that was transferred by a BEV to Drosophila and
Ae. aegypti cells [31].
For the most part, transfection by a BEV results in transient gene expression,
since the host cells are eventually killed by the viral infection. Jarvis et al. [32] were
able to overcome this obstacle by placing foreign DNA under an early viral promoter
(IE1) and eliminating that part o f the viral genome which produces the cytopathic
effects occurring late in the infective cycle. With this method, they were able to
obtain stable transformation o f a S. frugiperda cell line. Some o f the advantages o f
using BEVs are that very high levels o f gene expression can be obtained, large
amounts o f DNA can be inserted into the host genome and the virus is non-infectious
for vertebrates [33, 34].
Except for the recent reports by Kjér and Fallon [3], Stiles et al. [17] and
Hartig et al. [35], little attention has been given to comparing the effectiveness o f
the various DNA delivery systems that are available for transfection o f the insect cell
lines. Cell lines can differ widely in their susceptibility to the often toxic effects o f
 IAEA-SM-327/24 229

a chemical or physical facilitator for transfection. It is often essential to balance the

toxic effects o f a method with the optimal transfection frequency. Some methods
employing chemical mediators or electroporation typically produce > 5 0 % cell death
at the chemical concentration or field strength where the transfection frequency is
the highest. Other factors contained in a particular transfection protocol which affect
the fate o f the transfected DNA include the type o f plasmid vector used, linearized
versus circular DNA, the cell line, the selection regime, the DNA concentration and
the suitability o f the regulatory sequence (see the. review o f Walker [1]). Thus,
except for the dipteran and coleopteran cell lines previously mentioned, there have
been few studies where effective transfection protocols have been optimized for the
cell lines derived from other medically and agriculturally important insects.

4. TRANSFECTION OF INSECT EMBRYOS BY MICROINJECTION

The transfer o f DN A to Drosophila embryos by microinjection has become a

routine procedure where the preliminary steps involve the chorion removal o f
recently oviposited eggs in a dilute commercial bleaching agent or by mechanical
means, desiccation o f about 1-5% o f the egg volume and covering with a halocarbon
oil [36]. The injection site is usually at the posterior end o f the egg where the pole
cells are located and injection is done before cellularization o f the blastoderm.
Injected embryos are then returned to a humid environment until hatching. This tech­
nique has been modified variously to accommodate the insect embryos o f other spe­
cies. A growing number o f insect embryos, including two tephritid species [37, 38],
the silkworm moth [39-41], the migratory locust [42], the sheep blowfly [43],
mosquitos [44, 45] and the house fly [46], have all been transfected successfully with
various DNA constructs using microinjection procedures. Apparently, removal o f
the egg chorion was not possible in several species, but some embryos survived
injection through the intact chorion regardless.
It has been estimated that, under the best o f conditions, about 80% o f the
injected Drosophila embryos hatch, 40-50% o f these survive to adulthood and then
10-90% o f the surviving adults are sterile because o f their genetic background or
have injection induced abnormalities [47]. Further, the survival to adulthood from
injected embryos o f three mosquito species ranged from 6 to 15% [44, 48, 49] and
for the silkworm moth embryos it was 3.2% [40].
It should be mentioned that the physical manipulations required to facilitate
transfection by microinjection may influence expression o f a heterologous gene when
it is linked to the Drosophila hsp70 promoter. Eberlein and Mitchell [50] suggested
that the stress o f dechorionation, desiccation, injection and incubation under halo-
carbon oil was responsible for inducing expression o f hsp70-CAT activity in
Drosophila embryos, which was above that expected to be expressed constitutively
in the absence o f a heat shock. A microinjection procedure involving injection o f an
230 LEOPOLD

egg within the gravid female mite, Metaseiulus occidentalis, has provided a means
o f gaining stable transformation for this species [51]. A transformation frequency o f
0.5% was obtained without benefit of a transposable element system and survival o f
the injected females was described as an order o f magnitude higher than that gained
with injection o f mosquito embryos. It is not known whether the stable transforma­
tion achieved by this method is related to the reproductive biology o f this phytoseiid
mite, or to the introduction o f exogenous DNA prior to fertilization o f the egg.
Regardless, the potential o f this method for transforming other arthropods certainly
deserves further study. It may be the only method whereby insects having a unique
reproductive cycle like that o f the tsetse fly can be transformed.

5. OTHER TYPES OF DNA DELIVERY METHODS

The search for methods that can deliver DNA simultaneously to large numbers
o f embryos rather than treating them individually by microinjection has been
encouraged by recent studies. Baldarelli and Lengyel [52] used a bombardment tech­
nique whereby particles 1.2 ц т in diameter were coated with a DNA construct. The
particles were blasted shotgun like into samples containing 10 0 0 0 -2 0 000 dechorio-
nated Drosophila embryos using a commercially available biolistic bioparticle deliv­
ery system. The survival to adulthood was not reported in this study nor was the
transfection frequency, but it was described as ‘high’ .
Another method which has the capability o f delivering exogenous DNA simul­
taneously to samples containing hundreds o f embryos involves the use o f electropo­
ration. Kamdar et al. [53] were able to obtain a 95% frequency o f transient
expression o f the aldehyde oxidase gene transfected into ma-l mutant Drosophila
embryos. These results were obtained without dechorionation and about 75% o f the
treated insects survived to third instar larvae. Dechorionation reduced survival to
1 0 % at the third instar, but frequency o f expression was 1 0 0 %.

6 . ENHANCEMENT OF DNA DELIVERY TO GERM CELLS

Since it is essential to integrate DNA into the germ line when stable transfor­
mation o f an organism is desired, manipulations which have the potential to selec­
tively increase the likelihood o f uptake by germ cells would benefit any scheme
devised for gene integration. For example, injection o f gene constructs conjugated
to yolk proteins into the haemocoel of female insects during egg development has
been suggested as a means o f gaining selective uptake by oocytes [2]. Another
strategy could involve transfection o f sperm cells, either by direct injection o f the
testes or by transfecting sperm before artificial insemination (e.g. the honey bee) or
before in vitro fertilization [54]. Lipofection o f murine sperm was found to be an
 IAEA-SM-327/24 231

efficient method for transfering DNA but, unlike an earlier report [55], no transgenic
mice were obtained upon testing by in vitro fertilization [56].
Pole cell transplants between mutant lines o f Drosophila have been routinely
used to produce germ line mosaics for developmental and genetic studies. While it
may not be beneficial to transfect these primordial germ cells outside the embryo,
combining the exogenous DNA with a chemical mediator such as a cationic liposome
or polybrene before injection into the pole cell region o f the egg may increase the
efficiency o f transfection. Extending the life o f an introduced DNA construct
through several cycles o f cell division or in the case o f the preblastoderm embryo,
cleavage divisions, could also enhance the chances for integration by a vector or by
random integration. This, too, could be accomplished by combining the construct
with a chemical mediator before transfecting by injection, electroporation or
bombardment o f the embryos.

7. SUMMARY

There is a definite need for development and refinement o f DNA delivery

systems for cells, embryos and perhaps gonads to realize the potential that genetic
engineering by gene transformation presents for enhancing insect management pro­
grammes. Progress in this area is clearly indicated by the encouraging results gained
with optimizing cell culture transfection procedures and the recent successes
achieved in the transfection o f embryos obtained by electroporation and particle
bombardment. However, with the development o f new gene vectors to accommodate
different species o f economically important and biologically diverse insects, con­
tinued attention to optimization and modification o f DNA delivery systems will no
doubt be required. Further, the development o f new gene vectors, which depends
in part on gaining reproducible transfection efficiencies and, consequently, creating
reliable and reproducible DNA delivery systems, increases the probability that trans­
genic insects important to the fields o f medicine and agriculture will be obtained.

REFERENCES

[1] W A L K E R , V.K., “Gene transfer in insects”, Advances in Cell Culture,

( M A R A M O R O S C H , K., S A T O , G.H., Eds), Vol. 7, Academic Press, San Diego
(1989) 87-124.
[2] H A N D L E R , A.M., O ’ B R O C H T A , D.A., Prospects for gene transformation in insects,
Annu. Rev. Entomol. 36 (1991) 159-183.
[3] KJER, K.M., F A L L O N , A.M., Efficient transfection of mosquito cells is influenced
by the temperature at which DNA-calcium coprecipitates are prepared, Arch. Insect
Biochem. Physiol. 16 (1991) 189-200.
232 LEOPOLD

[4] RUBIN, G . M . , S P R A D L I N G , A.C., Genetic transformation of Drosophila with trans­

posable element vectors, Science 218 (1982) 348-353.
[5] S P R A D L I N G , A.C., RUBIN, G.M., Transposition of cloned P elements into
Drosophila germline chromosomes, Science 218 (1982) 341-347.
[6] L O Y T E R , A., S C A N G O S , G., R U D D L E , F., Mechanisms of D N A uptake by m a m ­
malian cells: Fate of exogenously added D N A monitored by the use of fluorescent dyes,
Proc. Natl. Acad. Sci. U S A 79 (1982) 422-426.
[7] S A N G , J.H., Drosophila cells and cell lines, Adv. Cell Cult. 1 (1981) 125-182.
[8] C H E R B U S , L., C H E R B U S , P., The effects of ecdysteroid hormones on Drosophila
melanogaster cells, Adv. Cell Cult. 1 (1981) 91—124.
[9] D I N O C E R A , P.P., D A W I D , I.B., Transient expression of genes introduced into
cultured cells of Drosophila, Proc. Natl. Acad. Sci. U S A 80 (1983) 7095-7098.
[10] SINCLAIR, J.H., S A N G , J.H., B U R K E , J.F., I S H - H O R O W I C Z , D., Extra chromo­
somal replication of copia-based vectors in cultured Drosophila cells, Nature (London)
306 (1983) 198-200.
[11] B U R K E , J.F., SINCLAIR, J.H., S A N G , J.H., I S H - H O R O W I C Z , D., A n assay for
transient gene expression in transfected Drosophila cells, using [3H]guanine incorpo­
ration, E M B O J. 3 (1984) 2549-2554.
[12] B E N Y A J A T I , C., D R A Y , J.F., Cloned Drosophila alcohol dehydrogenase genes are
correctly expressed after transfection into Drosophila cells in culture, Proc. Natl Acad.
Sci. U S A 81 (1984) 170Ы 705.
[13] B E R G E R , E.M., M A R I N O , G., T O R R E Y , D., Expression of Drosophila hsp
7 0 - C A T hybrid gene in Aedes cells induced by heat shock, Somat. Cell Mol. Genet. 11
(1985) 371-377.
[14] D U R B I N , J.E., F A L L O N , A.M., Transient expression of the chloramphenicol
acetyltransferase gene in cultured mosquito cells, Gene 36 (1985) 173-178.
[15] F A L L O N , A.M., Factors affecting polybrene-mediated transfection of cultured Aedes
albopictus (mosquito) cells, Exp. Cell Res. 166 (1986) 535-542.
[16] M O R G A N E L L I , C.M., B E R G E R , E.M., Transient expression of homologous genes
in Drosophila cells, Science 224 (1984) 1004-1006.
[17] STILES, B., H E I L M A N N , L.J., S P A RKS, R.B., S A N T O S O , A., L E O P O L D , R.A.,
Transfection of cultured cells of the cotton boll weevil with a heat shock promoter-
chloramphenicol construct, J. Insect Mol. Biol. 1 (in press).
[18] H E L G E N , J.C., F A L L O N , A.M., Polybrene mediated transfection of cultured
lepidopteran cells: Induction of a Drosophila heat shock promoter, In Vitro Cell. Dev.
Biol. 26 (1990) 731-736.
[19] F E L G N E R , P.L., et al., Lipofection: A highly efficient, lipid-mediated D N A -
transfection procedure, Proc. Natl Acad. Sci. U S A 84 (1987) 7413-7417.
[20] M I T S U H A S H I , J., “Fusion of insect cells”, Biotechnology in Invertebrate Pathology
and Cell Culture, ( M A R A M O R O S C H , K., Ed.), Academic Press, N e w York (1987)
387-400.
[21] C H I P L O N K A R , J.M., Electrofusion of mosquito cells, In Vitro Cell. Dev. Biol.
(1992) 230-232.
[22] N A K A J I M A , S., M I Y A K E , T., Cell fusion between temperature sensitive mutants of
a Drosophila melanogaster cell line, Somat. Cell Mol. Genet. 4 (1978) 131-141.
 IAEA-SM-327/24 233

[23] W Y S S , С., The loss of ecdysone sensitivity of a Drosophila cell line after hybridization
with embryonic cells? Exp. Cell Res. 125 (1980) 121-126.
[24] M A N N , S.G.,KING, L. A . ,Efficient transfection of insect cells with baculovirus D N A
using electroporation, J. Gen. Virol. 70 (1989) 3501-3505.
[25] T O N E G U Z Z O , F., K E A T I N G , A., G L Y N N , S., M c D O N A L D , K., Electric field-
mediated gene transfer: Characterization of D N A transfer and patterns of integration
in lymphoid cells, Nucleic Acids Res. (1988) 5515-5532.
[26] J A S T R E B O F F , M.M., ITO, E., B E R T I N O , J.R., N A R A Y A N A N , R., Use of electro­
poration for high-molecular-weight DNA-mediated gene transfer, Exp. Cell Res. 171
(1987) 513-517.
[27] F A U L K N E R , P., C A R S T E N S , E.B., A n overview of the structure and replication of
baculoviruses, Curr. Top. Microbiol. Immunol. 131 (1986) 1-19.
[28] SMITH, G.E., S U M M E R S , M.D., F R A S E R , M.J., Production of human beta inter­
feron in insect cells with a baculovirus expression vector, Mol. Cell. Biol. 3 (1983)
2156-2165.
[29] P E N N O C K , G.D., S H O E M A K E R , C., M I L L E R , L.K., Strong and regulated expres­
sion of Escherichia coli |3-galactosidase in insect cells with a baculovirus vector, Mol.
Cell. Biol. 4 (1984) 399-406.
[30] E L D R I D G E , R., O ’REILLY, D.R, H A M M O C K , B.D., M I L L E R , L.K., Insecticidal
properties of genetically engineered baculoviruses expressing an insect hormone ester­
ase gene, Appl. Environ. Microbiol. 58 (1992) 1583-1591.
[31] C A R B O N E L L , L.F., K L O W D E N , M.J., M I L L E R , L.K., Baculovirus-mediated
expression of bacterial genes in dipteran and mammalian cells, J. Virol. 56 (1985)
153-160.
[32] JARVIS, D.L., F L E M I N G , J.G., K O V A C S , G.R., S U M M E R S , M.D.,
G U A R I N O , L.A., Use of early baculovirus promoters for continuous expression and
efficient processing of foreign gene products in stably transformed lepidopteran cells,
Bio/Technology 8 (1990) 950-955.
[33] L U C K O W , V.A., S U M M E R S , M.D., Trends in the development of baculovirus
vectors, Bio/Technology 6 (1988) 47-55.
[34] M I L L E R , L.K., Baculoviruses as gene expression vectors, Annu. Rev. Microbiol. 42
(1988) 177-199.
[35] H A R T I G , P.C., C A R D O N , M.C., K A W A N I S H I , C.Y., Generation of recombinant
baculovirus via liposome-mediated transfection, Biotechniques 11 (1991) 310-313.
[36] A S H B U R N E R , М., Drosophila — A Laboratory Handbook, Cold Spring Harbor
Laboratory, Cold Spring Harbor, N Y (1989).
[37] O’ B R O C H T A , D.A., H A N D L E R , A.M., Mobility of P elements in drosophilids and
non-drosophilids, Proc. Natl Acad. Sci. U S A 85 (1988) 779-781.
[38] McINNIS, D.O., H A Y M E R , D.S., T A M , S.Y.T., T H A N A P H U M , S., Ceratitis
capitata (Diptera: Tephritidae): Transient expression of a heterologous gene for
resistance to the antibiotic geneticin, Ann. Entomol. Soc. A m . 83 (1990) 982-986.
[39] T A M U R A , T., K A N D A , T., T A K I Y A , S., O K A N O , K., M A E K A W A , H., Transient
expression of chimeric C A T genes injected into early embryos of the domesticated silk­
worm, Bombyx mori, Jpn J. Genet. 65 (1990) 401-410.
234 LEOPOLD

[40] NIKOLAEV, A.I., CHKONIYA, T.I., KAFIANI-ERISTAVI, K.A., Extrachromo­

somal localization and hereditary transfer of recombinant plasmid microinjected into
silkworm eggs, Mol. Biol. (English transi.) (US) (1991) 896-904.
[41] KANDA, T., TAMURA, T., Microinjection method for DNA embryos of the silk­
worm, Bombyx morí, using air pressure, Bull. Natl Inst. Serie. Entomol. Soc. 2 (1991)
47-63 (in Japanese, with English summary).
[42] MATHI, S., WALKER, V.K., WYATT, G.R., Expression from two Drosophila
promoters in embryos of the migratory locust, Insect Biochem. (1991) 623-629.
[43] ATKINSON, P.W., O’BROCHTA, D.A., In vivo expression of two highly conserved
Drosophila genes in the Australian sheep blowfly, Lucilia cuprina, Insect Biochem.
(1992) 423-431.
[44] MILLER, L.K., et al., Stable integration of a bacterial gene in-the mosquito, Anopheles
gambiae, Science 237 (1987) 779-781.
[45] CRAMPTON, J., MORRIS, A., LYCETT, G., WARREN, A., EGGLESTON, P.,
Transgenic mosquitoes: A future vector control strategy? Parasitai. Today 2 (1990)
31-36.
[46] LEOPOLD, R.A., unpublished data.
[47] PIRROTTA, V., “ Vectors for P-mediated transformation in Drosophila” , Vectors —
A Survey of Molecular Cloning Vectors and their Uses, (RODRIGUEZ, R.L.,
DENHARDT, D.T., Eds), Butterworths, Boston, MA (1988) 437-456.
[48] McGRANE, V., CARLSON, J.O., MILLER, B.R., BEATY, B.J,, Microinjection of
DNA into Aedes triseriatus ova and detection of integration, Am. J. Trop. Med. Hyg.
39 (1988) 502-510.
[49] MORRIS, A.C., EGGLESTON, P., CRAMPTON, J.M., Genetic transformation of
the mosquito, Aedes aegypti, by microinjection of DNA, Med. Vet Entomol. 3 (1989)
1-7.
[50] EBERLEIN, S., MITCHELL, H.K., Protein synthesis patterns following stage-
specific heat shock in early Drosophila embryos, Mol. Gen. Geiiet. 210 (1987)
407-412.
[51] PRESNAIL, J.K., HOY, M.A., Stable genetic transformation of a beneficial arthro­
pod, Metaseiulus occidentalis (Acari: Phytoseiidae), by a microinjection technique,
Proc. Natl Acad. Sci. USA 89 (1992) 7732-7736.
[52] BALDARELLI, R.M., LENGYEL, J.A., Transient expression of DNA after ballistic
introduction Into Drosophila embrÿos, Nucleic Acids Res. 18 (1992) 5903-5904.
[53] KAMDAR, P., VON ALLMEN, G., FINNERTY, V., Transient expression of DNA
in Drosophila via electroporation, Nucleic Acids Res. 20 (1992) 3526.
[54] SAWA, М., Fertilization of hetero-specific insect eggs by sperm injection, Jpn J.
Genet. 66 (1991) 297-303.
[55] BACHILLER, D., SCHELLANDER, K., PELI, J., RUTHER, U., Liposome-
mediated DNA uptake by sperm cells, Mol. Reprod. Dev. (1991) 194-200.
[56] LAVITRANO, M.L., et al., Sperm cells as vectors for introducing foreign DNA into
eggs: Genetic transformation of mice, Cell (1989) 717-723.
 IAEA-SM-327/25

USE OF Y LINKED TRANSLOCATIONS

IN LOCATING MUTANT LOCI (Bl, dp)
ON POLYTENE SALIVARY GLAND
CHROMOSOMES OF Anopheles stephensi LISTON

PHAM VA N LAP
Department o f Genetics,
Faculty o f Biology,
University o f Hanoi,
Hanoi, Viet Nam

A.S. ROBINSON*
Insect Genetics Unit,
Research Institute ITAL,
Wageningen, Netherlands

Abstract

USE OF Y LINKED TRANSLOCATIONS IN LOCATING MUTANT LOCI (Bl, dp)

ON POLYTENE SALIVARY GLAND CHROMOSOMES OF Anopheles stephensi
LISTON.
Using two Y linked translocations, in which the break points were tightly linked to the
morphological mutants (Bl, dp), the location of mutant loci on polytene salivary gland
chromosomes of Anopheles stephensi was determined. In searching for discontinuities in the
polytene chromosomes of male larvae from the T(Y-3)20 translocation involving a black larva
mutant, a single break point was found in region 36D/37 of 3R. Analysing the polytene
chromosomes of male larvae from the T(Y-3)12 translocation involving the diamond palpus,
the translocation break point was determined at position 34A of 3R. Because the Y/autosome
breakpoint in T(Y-3)20 was tightly linked to the black larva mutant (Bl), and the break point
in T(Y-3)12 was tightly linked to the diamond palpus mutant dp, it was concluded that gene
Bl is located very close to the map reference 36D/37 of 3R and that the gene dp is located
at position 34A of 3R. The mitotic chromosomes of these Y linked translocations are
described.

1. INTRODUCTION

Anopheles stephensi Liston is an important malaria vector in the Indian sub­

continent and has been the subject o f genetic and cytogenetic studies aimed at the
development o f genetic control strategies [1-5]. These strategies are mainly involved

* Present address: Institute of Molecular Biology and Biotechnology, P.O. Box 1527,
GR-71110 Heraklion, Crete, Greece.

235
236 PHAM VAN LAP and ROBINSON

in the release o f sterile mosquitoes carrying chromosomal rearrangements into field

populations with the aim o f population suppression or replacement. In the isolation
o f chromosomal rearrangements or other genetic studies, it is very necessary to
employ mutant markers in order to make the crosses manageable, simple and
efficient. Further, Y/autosome translocations and deficiencies have been used to
determine the location o f some mutant loci in different mosquito species
[6 , 7]. In An. stephensi, a genetic sexing system has been developed that is based
on the translocation o f dieldrin resistance to the male determining chromosome [5].
Analysing the salivary gland chromosomes o f this genetic sexing line, Robinson and
Pham van Lap [ 8 ] were able to determine that the location o f the dieldrin locus in
3L was very close to map reference 43/44. Sakai et al. [1], in studying different
translocations, suggested that the diamond palpus (dp) and black larva (Bt) mutants
are associated with the shorter o f the mitotic autosomes and with the 3R element in
polytene configurations. During the isolation o f the genetic sexing system in
An. stephensi, Robinson et al. [9] isolated two male linked translocations in which
the translocation break points were tightly linked to BI and dp. These Y linked
translocation strains provided us with an ideal opportunity for localizing the above-
mentioned genes on the polytene chromosomes o f An. stephensi. The results o f our
study represent another step towards a better understanding o f the genetics o f
An. stephensi and may be useful in future investigations relating to the development
o f genetic control strategies.

2. MATERIALS AND METHODS

2.1. Strains

The following Y linked translocation strains isolated by Robinson et al. [9]

were used:

— T(Y-3)20: A translocation between the male determining chromosome and

chromosome 3. The B l+ allele was completely linked to the translocation
break point. BI is co-dominant and BI/+ has an intermediate colour between
+ / + and Bl/Bl.
— T(Y-3)12: A Y linked translocation in which dp, an autosomal recessive on
chromosome 3 [10], was tightly linked to the translocation break point.

2.2. Rearing

All the mosquitoes were held at 28° С and 80% relative humidity (RH). The
larvae were fed on fish food (Tetramicromin) and the adult blood meals were
provided by anaesthetized mice. Since thé translocation strains were very stable, they
were maintained generation by generation without any selection.
 IAEA-SM-327/25 237

2.3. Cytology

The mitotic and meiotic chromosomes were prepared from the testes o f fourth
instar larvae, as described by French et al. [11], to confirm the presence o f the trans­
locations. The salivary glands o f early fourth instar larvae were dissected in a droplet
o f 5% diluted Carnoy’ s fixative and then transferred to a small drop o f concentrated
Carnoy (3 ethylalcohol:l acetic acid) on a siliconized coverslip. This tissue was
stained with 2% lacto-aceto-orcein. The squashed preparations were sealed using
photo glue and stored at 2°C . The chromosomes were investigated under a Carl
Zeiss microscope. Before dissection o f the salivary glands, the late third instar larvae
or early fourth instar larvae were kept overnight in an insectary set at 20°C. Accord­
ing to Seawright et al. [7], low temperature treatment enhances the preparation o f
readable polytene chromosomes. The salivary gland chromosome map o f Sharma
et al. [ 1 2 ] was used to identify the break points.

3. RESULTS AND DISCUSSION

3.1. Standard mitotic and salivary gland chromosome complements

The Lahor wild type line (■+/+) described by Robinson et al. [9] was used for
making the preparations o f the standard mitotic and polytene salivary gland chromo­
somes. The mitotic chromosome complement o f An. stephensi consists o f one pair
o f short subtelocentric sex chromosomes and two pairs o f longer autosomes, one
metacentric and the other submetacentric. The longer and shorter autosomes are
designated as chromosomes 2 and 3, respectively. The sex chromosomes are
morphologically distinguishable in males. The shorter sex chromosome has been
designated as the Y and the longer one the X chromosome. The short arms o f
the Y and X chromosomes are equal in length, but the long arm o f X is longer than
that o f Y.
Analysis o f the polytene salivary gland chromosomes o f the wild type line
showed a complement composed o f five arms, the shortest bèing X , which was often
separated from the chromocentre, and four autosomal arms. The Y chromosome is
heterochromatic and does not polytenize [1]. The longest arm in the whole comple­
ment is designated as 2R, which is characterized by a big puff in zone 7. The shortest
autosomal arm is designated 2L, which can easily be recognized by its length and
by other characteristics. The centromere part o f 2L (region 20) always formed a
loop, as described by Mittal and Dev [13]. The 3L and 3R arms are nearly equal
in length. They can be recognized by several characteristic landmarks. The free end
o f 3R is flared, with two dark bands and some dot line bands in region 29A. The
free end o f 3L is characterized by several light bands. In general, the banding pattern
in the polytene salivary gland chromosomes o f our An. stephensi specimen was
238 PHAM VAN LAP and ROBINSON

». f\
■y
X t У
‘.y j
Ü \
£ > ^ - л 'гг' r

U»
Яг

t e s » 3

F/G. /. Mitotic chromosome complement of the T(Y-3)20 translocation heterozygote. The

arrows denote the breaks.

FIG. 2. Discontinuity in region 36D/37 o f 3R.

 IAEA-SM-327/25 239

similar to that described in the map o f Sharma et al. [12] and to that shown in the
photomap o f Mittal and Dev [13]. Therefore, the polytene salivary gland
chromosome map o f Sharma et al. [12] was used to designate the chromosome breaks
in our translocation strains. By analysing the mitotic and polytene chromosomes
in different translocations o f An. stephensi, in which visible alterations occurred,
Sakai et al. [1] showed that the 2R -2L arms in the polytene chromosome complement
are associated with the longer o f the mitotic autosomes and that the 3R -3L polytene
elements are associated with the shorter o f the mitotic autosomes.

3.2. Cytological characterization of T(Y-3)20

Analysis o f the mitotic chromosomes o f the male larvae o f translocation

T(Y-3)20 revealed an unusual karyotype in all cases. There was unequal exchange
between 3 and Y , with chromosome 3 losing a very long segment and receiving a
short one from the long arm o f Y. Therefore, the translocated chromosome 3 was
nearly one arm shorter than its normal homologue and the Y chromosome became
longer than normal. In Fig. 1, the synapsis was clearly observed between the
long arm o f Y and the normal 3. Therefore, the break points can be seen in the
middle o f the Y chromosome and in the region very near to the centromere o f
chromosome 3 in the mitotic chromosomes.
Analysis o f the polytene salivary gland chromosomes o f the male larvae o f the
translocation strain showed a complement composed o f five arms as normal. But by
searching for discontinuities in the autosomal arms, the position o f the Y/autosome
break was determined. In Fig. 2, the 3R and 3L arms can be seen, the break point
being localized in region 36D/37 o f 3R. The Y chromosome is heterochromatic and
was not observed in the polytene chromosome complement, but Fig. 2 shows that
it has a strong affinity to the chromocentre. As the black larva locus must be tightly
linked to the translocation breakpoint [9], it was concluded that the gene coding for
larva colour (Bt) is located in 3R, very close to map reference 36D/37. Sakai et al.
[1], using different translocations, showed that the gene BI is located in 3R. Robinson
et al. [9] suggested that the BI locus is located near to the centromere. Our results
now confirm that it is located on the polytene chromosome.

3.3. Cytological characterization of T(Y-3) 12

Analysis o f the mitotic chromosomes o f the male larvae o f this strain showed
an abnormal karyotype (Fig. 3). Mitotic configurations showed an unequal exchange
between chromosome 3 and the long arm o f the Y chromosome. Chromosome 3 lost
a long segment and received a short one from the Y chromosome, so becoming about
half an arm shorter than its normal homologue. Analysing the polytene salivary gland
chromosomes o f male larvae showed that a single break point was found in region
34A o f 3R (Fig. 4). As the ф locus must be tightly linked to the break point [9],
240 PHAM VAN LAP and ROBINSON

мймишая

T m fjT ? '
ШШшёшЛШ

F/G. 5. Mitotic chromosome complement of the T(Y-3)12 translocation heterozygote.

~:%jF * .

» . Í *Ч • ** ** \
7* к •* ** *’ ** * *
J ' Ч V

V !» ^ . ' • W J
**« - <*'• ....................................

F/G. 4. Translocation break ofT(Y-3)12 in region 34A of 3R. The arrow denotes the break.
 IAEA-SM-327/25 241

it was concluded that the dp locus was located in 3R, veiy near to region 34A. By
analysing several translocations in which the dp locus is involved, Sakai et al. [1]
showed that the diamond palpus mutant is associated with the shorter o f the mitotic
autosomes and with the 3R element in polytene configurations. The break points in
the 3R aberrations studied in their experiment were distributed from zones 30-35 and
all showed moderately strong linkage with the dp locus. Our T(Y-3)12 translocation
showed strong linkage to dp, therefore it should be located very near to the above
mentioned region.

ACKNOWLEDGEMENTS

Pham Van Lap received a fellowship from the Netherlands University

Foundation for International Co-operation (NUFFIC) (VH3 project) for carrying out
this research in the Insect Genetics Unit, Research Institute ITAL, Wageningen,
Netherlands.

REFERENCES

[1] SAKAI, R.K., MAHMOOD, F., AKTAR, K., DUBASH, J., BAKER, R.H., Induced
chromosomal aberrations and linkage group-chromosome correlation in Anopheles
stephensi, J. Hered. 74 (1983) 232-238.
[2] SAKAI, R.K., MAHMOOD, F., Homozygous chromosome aberrations in Anopheles
stephensi, J. Hered. 76 (1985) 230-236.
[3] VAN HEEMERT, C., TOAN, T., ROBINSON A.S., FELDMAN, A., Induction and
isolation of translocations in Anopheles stephensi, Mosq. News 43 (1983) 480-484.
[4] R O B I N S O N , A.S., Sex-ratio manipulation in relation to insect pest control, Annu.
Rev. Genet. 17 (1983) 191-214.
[5] ROBINSON, A.S., Genetic sexing in Anopheles stephensi using dieldrin resistance,
J. Am. Mosq. Control Assoc. 2 (1986) 93-95.
[6] RABBANI, M.G., SEAWRIGHT, J.A., Use of Y-autosome translocations in assigning
the stripe locus to chromosome 3 in the mosquito Anopheles albimanus, Ann. Entomol.
Soc. Am. 69 (1976) 2.
[7] SEAWRIGHT, J.A., BENEDICT, M.Q., NARANG, S., Use of deficiencies for map­
ping four mutant loci on salivary gland chromosomes of Anopheles albimanus, Mosq.
News 44 4 (1984).
[8] ROBINSON, A.S., PHAM VAN LAP, Cytological, linkage and insecticide studies on
a genetic sexing line in Anopheles stephensi Liston, Heredity 58 (1987) 95-101.
[9] ROBINSON, A.S., MALCOLM, C., MALI, P., SCHELLING, G., Breakpoint distri­
bution in male-linked translocations in Anopheles stephensi Liston, J. Hered. 77 6
(1986).
[10] SAKAI, R.S., BAKER, R.H., DUBASH, C.J., RAANA, K., The genetics of diamond
palpus in Anopheles stephensi, Mosq. News 41 (1981) 125-128.
242 PHAM VAN LAP and ROBINSON

[11] F R E N C H , W.L., B A K E R , R.H., K I T Z M I L L E R , J.B., Preparation of mosquito

chromosomes, Mosq. News 22 (1962) 377-388.
[12] S H A R M A , . G.P., P A R S H A D , R., N A R A N G , S.L., K I T Z M I L L E R , J.B., The
salivary gland chromosomes of Anopheles stephensi,J. Med. Entomol. 6 (1969) 68-71.
[13] M I T T A L , O.P., D E V , V., A photomap of the salivary gland chromosomes of
Anopheles stephensi Liston (Culicide: Diptera), Cytobios 19 (1978) 151-157.
 IAEA-SM-327/26

ANALYSIS OF STRUCTURAL REARRANGEMENTS

OF LEPIDOPTERA CHROMOSOMES USING
THE CENTRIFUGATION SPREADING TECHNIQUE

F. MAREC
Institute o f Entomology,
Czechoslovak Academy o f Sciences,
Ceské Budéjovice,
Czechoslovakia

W . TRAUT
Institut für Biologie der Medizinischen
Universitât zu Liibeck,
Liibeck,
Germany

Abstract

ANALYSIS OF S T R U C T U R A L R E A R R A N G E M E N T S OF LEPIDOPTERA C H R O M O ­
S O M E S USING T H E CENTRIFUGATION SPREADING TECHNIQUE.
A centrifugation spreading technique (CST) was used to analyse 12 sex chromosome
mutant female lines of the Mediterranean flour moth, Ephestia kuehniella. The method permit­
ted visualization of the long synaptonemal complexes of microspread pachytene nuclei in the
electron microscope. In four lines, T(W;Z)l-4, a stable translocation of a Z chromosome
segment on the W chromosome was confirmed. In the other lines, designated A S F (abnormal
segregating females), unexpected phenotypes segregated in atypical ratios. A quadrivalent
with both sex chromosomes and an autosome bivàlent was observed in females of the ÁSF-3
line, indicating that part of the Z chromosome was translocated on to one of the autosomes.
In seven A S F lines, an additional fragment of the Z chromosome was found to be responsible
for the segregation of unexpected phenotypes. It was shown that C S T is useful for the study
of chromosome aberrations in species with extremely small metaphase chromosomes.

1. INTRODUCTION

Concerning the use o f genetic methods for the suppression o f lepidopteran

species, there is a great need for additional data on Lepidoptera cytogenetics. For
instance, very little is known about the cytogenetic basis for the high resistance o f
Lepidoptera to ionizing radiation; little is also known about the chromosomal aberra­
tions leading to the inherited sterility phenomenon. However, the chromosomes o f
Lepidoptera are difficult to investigate using standard cytogenetic methods because
they are too small, too numerous and uniform in shape during metaphase. These

243
244 MAREC and TRAUT

technical problems can partially be overcome when a modified centrifugation

spreading technique (CST) [1] is used. Thé method, also called the microspreading
technique [2], permits visualization o f the long synaptonemal complexes (SCs) o f
spread pachytene nuclei in the electron microscope (EM).
We used CST to analyse chromosome aberrations in females o f 12 sex chromo­
some mutant lines o f the Mediterranean flour moth, Ephestia kuehniella Zeller.
These structural rearrangements o f the sex chromosomes were induced by gamma
rays in wild type females and isolated using the sex linked recessive marker dz [3].
Ephestia kuehniella offers the advantage that a great deal o f cytogenetic work
has already been done on it. The moth possesses a W Z /Z Z sex chromosome
mechanism, which is typical for most lepidopteran species, and females are the
heterogametic sex. Although the diploid chromosome set o f this moth is large
(2n = 60), the W Z bivalent can frequently be recognized in spread pachytene
oocytes from remnants o f the compact heterochromatin associated with the
W chromosome [1, 2, 4].

2. MATERIALS AND METHODS

2.1. Insects

In the present study, we used wild.type strain С (kept in Ceské Budéjovice)

[5] and 12 sex chromosome.mutant lines [3] o f E. kuehniella. The sex linked reces­
sive mutation dz (dark central area o f the forewings) [6 ] served as a genetic marker.
In each mutant line, all structurally normal Z chromosomes were marked with the
dz allele. All the mutant females were heterozygous for the dz locus and, therefore,
were phenotypically o f the wild type. To maintain these lines, the sex chromosome
mutant (but wild type) females were mated to dz males in each generation. In four
lines, T (W ;Z )l-4 , females were proved by segregation data to possess part o f the
Z chromosome, including the dz + allele translocated on to the W chromosome. In
the other lines, designated ASF (abnormal segregating females), unexpected pheno­
types segregated in atypical ratios; the genetic mechanism for this is unknown
(Table I).
The insects were reared on flaked oats at 20-21 °C. For EM preparations, the
ovaries were dissected from both the feeding and wandering larvae o f fifth instar or
from prepupae; the testes o f third and fourth instar larvae were used.

2.2. EM microspreading of pachytene nuclei

The CST was performed according to the following protocol [1]. Gonads,
dissected in a physiological solution, were transferred to a hypotonic solution
(83 mM KCI and 17 mM NaCl) and disrupted using fine tungsten needles. After
 IAEA-SM-327/26 245

TABLE I. CHARACTERISTICS OF SEX CHROMOSOME MUTANT LINES

OF E. kuehniella ACCORDING TO PREVIOUSLY OBTAINED SEGREGATION
DATA [3]. IN EACH MUTANT LINE, AS WELL AS IN THE CONTROL
CROSSES, THE PROGENY ORIGINATED FROM CROSSES BETWEEN
PHENOTYPIC A LLY WILD TYPE FEMALES ( + ) AND dz MALES

Segregated progeny (%)

Line code Females Males

+ dz + dz

T(W;Z)1 46 — — 54
T(W;Z)2 48 — — 52
T(W;Z)3 45 — — 55
T(W;Z)4 50 — — 50
ASF-1 40 12 — 48
ASF-2 39 17 — 44
ASF-3 32 34 — 34
ASF-4 37 9 11 43
ASF-5 33 16 14 37
ASF-6 33 ■ 14 15 38
ASF-7 33 16 14 37
ASF-8 30 9 11 50
Control — 52 48 ' —

10-20 min, the swollen material was transferred to a spreading solution (0.02% Joy
detergent, Procter and Gamble, United States o f America, adjusted to pH8.6-8.9
with a 0.01M borate buffer solution o f pH9.22, Merck, Germany) and allowed to
disperse for 5-10 min. Then the material was centrifuged in a Teflon microcentri­
fugation chamber on to a hydrophilized EM grid through a 0.1M sucrose cushion
containing 1 % formaldehyde (pH 8.2-8.5). The centrifugation was performed at 4°C
in a minifuge 2 (Heraeus, Germany) at 2200 rev./min for 10 min. Then the
specimen was fixed for 2 min with a fresh formaldehyde-sucrose solution, rinsed for
30 s in 0.4% Kodak Foto Flo detergent (pH 8.0-8.2), air dried and stained for 30 s
in 1 % ethanolic phosphotungstic acid. Micrographs were taken in a Philips EM 400
operated at 80 kV.
246 MAREC and TRAUT

In each line, at least 30 well spread nuclei o f pachytene oocytes were examined
for the configuration o f sex chromosomes and their structural changes. A digitizer
tablet and two computer programs, MESSCHRO and EVALCHRO (written by
W . Traut), were used to measure chromosomes on the EM micrographs.

3. RESULTS

3.1. Wild type females

In the oocyte pachytene nuclei, the complete set o f 30 bivalents was observed.
Synaptonemal complexes formed by 29 autosomal homologues consisted o f two
completely paired lateral elements o f equal length with chromatin loops radiating
from them. In some preparations, a central element was visible in the free space
between the two lateral elements. In most o f the pachytene complements, the W Z
bivalent could be identified as the only SC with both lateral elements o f unequal
length and from the remnants o f compact heterochromatin tangles decorating the
W chromosome axis [1]. Measurement o f 47 W Z bivalents revealed that the
W chromosome axis was, on average, 20% shorter than that o f the Z chromosome.
Only 60% o f the W Z bivalents were completely paired, often showing characteristic
twisting o f the longer Z lateral element along the shorter W lateral element (see
W T in Fig. 1). In the remaining cases, the W Z bivalents displayed partial pairing,
and in a few complements the sex chromosomes were not paired at all. This may
indicate delayed pairing o f these non-homologous chromosomes in the early
pachytene.

3.2. Wild type males

The male pachytene complements were composed o f 30 well paired bivalents.

The sex chromosomes (ZZ ) could not be recognized because o f homologous pairing
and the absence o f any morphological marker. In contrast to the female SCs, almost
every SC in the male complements had one recombination nodule, usually seen in
the free space between the lateral elements. This finding is in accordance with the
fact that male meiosis is chiasmatic, in contrast to the achiasmatic meiosis o f
Ephestia females [7]. Recombination nodules were also found in three dimensional
reconstructions o f Bombyx mori male SCs [ 8 ].

3.3. T(W ;Z) females

In the females o f four T (W ;Z ) lines, part o f the Z chromosome, genetically

marked by dz + , is linked to the W chromosome [3]. Therefore, we used the symbol
W + for this neo-W chromosome. The W +Z bivalents in all four lines resembled,
 IAEA-SM-327/26 247

FIG: 1. Schematic drawing of the most frequent pachytene configurations of sex chromo­
somes in WT, T(W;Z) and ASF females.

with small differences, the W Z bivalent o f wild type females. No heterochromatin

tangles were observed at one end o f the W + lateral element documenting the
presence o f the translocated segment o f the Z chromosome (Fig. 1). This segment
was relatively long, especially in T(W ;Z)1 and T(W ;Z)3 females, in which its length
represented 30-45% o f the whole W + axis. In. accordance with previously obtained
data [9], the shortest segment without heterochromatin tangles was, observed in
T(W ;Z)4 females (about 25% o f the W + axis). In all the lines, the W + lateral
element was shorter than the Z lateral element in spite o f the translocation presence.
This indicated that the original W chromosome had,been shortened by gamma rays.
In contrast to the wild type W Z bivalent, most o f the sex chromosome SCs in
T (W ;Z ) females displayed complete synapsis, probably due to homologous pairing
between the translocated segment and the corresponding part o f the normal
248 MAREC and TRAUT

Z chromosome. Concerning W +Z pairing, exceptions were only observed in

T(W ;Z)1 females. In 15% o f the nuclei o f these females, the W + chromosome was
autosynapsed, while the Z chromosome remained unpaired.

3.4. ASF females

In most ASF-1 and ASF-2 females, the pachytene complements displayed a

fragment o f the Z chromosome ( Z +) in addition to the normal Z chromosome. The
length o f the Z + axis was about 40% that o f the Z axis. The W chromosome was
most probably deleted because its axis (w “) was evidently shorter than that o f the
wild type chromosome W. Three main types o f configuration o f sex chromosomes
were observed (Fig. 1). In about 60% o f the nuclei, a W “Z Z + trivalent was
formed. The remaining nuclei showed either a W “Z bivalent and an autosynapsed
Z + fragment, or a Z Z + bivalent with very unequal lateral elements and an
autosynapsed (or free) W " chromosome.
A quadrivalent with the W Z bivalent and an autosome bivalent (A A +) was
frequently found in ASF-3 females (Fig. 1). This indicated that a part o f the
Z chromosome marked with the dz + allele had been translocated on to one o f the
autosomes. Besides the quadrivalent, other configurations o f W Z and A A + were
observed. For example, both the bivalents were separated, and A A + could be
recognized as an SC with unequal lateral elements. The unpaired end o f this SC
might represent the trarislocation. However, investigations o f a number o f ASF-3
female larvae did not reveal any structural changes in their SC complements. These
larvae were regarded as dz females, which are segregated in the progeny o f the
ASF-3 line with a high frequency.
More than 70% o f the microspread pachytene nuclei in ASF-4 females showed
a well synapsed W Z Z + trivalent (Fig. 1). The Z + fragment was significantly
longer than fragments occurring in the other ASF lines. It was typical that the
Z chromosome axis was considerably extended and appeared to be thinner in the
middle part. Thus, the Z lateral element was adjusted to synapse with both the W
and Z + lateral elements.
A shorter Z + fragment, representing about 30-40% of the length o f the
Z chromosome axis, was regularly observed in the pachytene complements o f
ASF-5, ASF- 6 , ASF-7 and ASF - 8 lines. Similar to the ASF-1 and ASF-2 lines, three
types o f sex chromosome pairing were recorded. W Z Z + trivalents were less fre­
quent than those o f the ASF-l and ASF-2 lines; they were found in half o f the nuclei.
Many complements displayed a W Z bivalent and a Z + fragment, which was
usually autosynapsed. In the remaining complements, a Z Z + bivalent with very
unequal lateral elements and a free W univalent were observed.
 IAEA-SM-327/26 249

4. DISCUSSION

With the CST used it was possible not only to describe the structural rearrange­
ments in the 12 sex chromosome mutants o f E. kuehniella, but also to correlate the
previously obtained cross-breeding data [3] with the pairing behaviour o f sex
chromosomes.
In the T (W ;Z ) lines, a stable translocation o f a Z chromosome segment on the
W chromosome was confirmed. Owing to the homology between this segment and
the corresponding part o f the Z chromosome, the translocation seemed to increase
the pairing affinity o f the W and Z chromosomes, which are largely — if not
completely — non-homologous in females o f the wild type strain [2 ].
In the ASF-3 line, another type o f translocation was found to be responsible
for the segregation o f wild type and dz females at a ratio o f 1:1 (see Table I). The
dz+ allele was translocated on to an autosome and, therefore, exhibited autosomal
inheritance. However, this finding did not explain the total absence o f wild type
males in the progeny. One may only speculate that these males were lethal because
o f trisomy for some important genes o f the Z chromosome.
The atypical phenotypic ratios in the other ASF lines were due to the presence
and pairing behaviour o f those Z chromosome fragments possessing the dz + allele.
In W Z Z + trivalents, which appeared to be the predominating configuration o f sex
chromosomes in the females o f these lines, the Z + fragments were synapsed with
their structurally normal homologues. In such cases, they should segregate together
with the W chromosome, simulating a W linkage. On the other hand, free or
autosynapsed fragments could segregate randomly, resulting in the occurrence o f
exceptional phenotypes o f both sexes in the progeny. However, only exceptional
dz females segregated in the ASF-1 and ASF-2 lines but, apart from the deleted
W chromosome, we did not find any substantial difference between these and the
remaining five ASF lines in the configuration o f sex chromosomes. Thus, the
absence o f exceptional wild type males in the ASF-1 and ASF-2 lines remains to be
clarified.
In conclusion, the present paper shows that CST is a suitable method for
chromosome analysis in species with extremely small metàphase chromosomes.
Therefore, this method may be useful in investigating the effects o f ionizing radiation
on the chromosomes o f any lepidopteran species that are candidates for genetic pest
control.

ACKNOWLEDGEMENT

F. Maree was supported by the Alexander von Humboldt Foundation, Bonn,

Germany.
250 MAREC and TRAUT

REFERENCES

[1] W E I T H , A., T R A U T , W., Synaptonemal complexes with associated chromatin in a

moth, Ephestia kuehniella Z.: The fine structure of the W chromosomal hetero­
chromatin, Chromosoma 78 (1980) 275-291.
[2] W E I T H , A., T R A U T , W., Synaptic adjustment, non-homologous pairing, and non­
pairing of homologous segments in sex chromosome mutants of Ephestia kuehniella
(Insecta, Lepidoptera), Chromosoma 94 (1986) 125-131.
[3] M A R E C , F., MIRCHI, R., Genetic control-of the pest Lepidoptera: Gamma-ray
induction of translocations between sex chromosomes of Ephestia kuehniella Zeller
(Lepidoptera: Pyralidae), J. Stored Prod. Res. 26 (1990) 109-116.
[4] T R A U T , W., W E I T H , A., T R A U T , G., Structural mutants of the W chromosome in
Ephestia (Insecta, Lepidoptera), Genetica 70 (1986) 69-79.
[5] M A R E C , .F., Genetic control of pest Lepidoptera: Induction of sex-linked recessive
lethal mutations in Ephestia kuehniella (Pyralidae), Acta. Entomol. Bohemoslov. 87
(1990) 445-458.
[6] K Ü H N , A., Über eine geschlechtsgekoppelte Mutation des Zeichnungsmusters dz bei
Ephestia ktihniella Z., Biol. Zbl. 59 (1939) 347-357.
[7] T R A U T , W-, A study of recombination, formation of chiasmata and synaptonemal
complexes in female and male mèiosis of Ephestia kuehniella (Lepidoptera), Genetica
47 (1977) 135-142.
[8] H O L M , B.H., R A S M U S S E N , S.W., Chromosome pairing, recombination nodules
and chiasma formation in diploid Bombyx males, Carlsberg Res. C o m mun. 45 (1980)
483-548.
[9] M A R E C , F., Genetic control of pest Lepidoptera: Construction of a balanced lethal
strain in Ephestia kuehniella, Entomol. Exp. Appl. 61 (1991) 271-283.
 IAEA-SM-327/66

Ceratitis capitata: SUITABLE MARKERS

FOR POPULATION GENETICS
AND GENOME ORGANIZATION ANALYSIS

G. GASPERI*, A .R . M ALACRIDA*, C.R. GUGLIELMINO**,

L. BARUFFI*, C. TORTI*, R. MILANI*,
G. DAM IANI***, C. BAÑDI***

* Dipartimento di Biología Animale,

** Dipartimento di Genetica e Microbiología,

Université di Pavía,
Pavía

*** IDVGA,
Consiglio Nazionale delle Ricerche,
Milan

Italy

Abstract

Ceratitis capitata: S U I T A B L E M A R K E R S F O R P O P U L A T I O N G E N E T I C S A N D
G E N O M E O R G A N I Z A T I O N A N A L YS I S .
Biochemical and molecular markers (RAPD) càn be used for population genetics and
genome analysis of the Mediterranean fruit fly (medfly), Ceratitis capitata. At the population
level, biochemical markers allowed recognition of: (a) the presence of geographical genetic
heterogeneity, and (b) provided information on the relative contribution of different types of
evolutionary forces during the process of colonization. The polymorphisms in genomic finger­
prints, generated by the R A P D approach, provide a tool to improve the significance of the
estimates of genetic relatedness between medfly populations, as well as being able to distin­
guish between slightly divergent flies. For genome organization analysis, biochemical markers
provided information on the relationship of genome structure to genome function. The R A P D
approach may provide a new tool to explore and completely m a p the medfly genome.

!.. INTRODUCTION

The Mediterranean fruit fly (medfly), Ceratitis capitata Wied., is one o f the
major fruit crop pests in the world. It is a polyphagous and multivoltine tropical spe­
cies which in the last hundred years has spread from its supposed origin in Africa
to a number o f countries, including the Mediterranean basin, parts o f South and Cen­
tral America and Australia [1]. .

251
252 GASPERI et al.

The use o f biochemical markers [2] has broadened the available knowledge on
the genetic features o f this species at two main levels:

(a) Population genetics and taxonomic relationships [3-5],

(b) Genome organization [6 , 7].

In addition, the study o f the genetic control o f enzyme systems, such as

ADH [8 , 9], and o f proteins typical o f critical developmental stages (LSP) [10] has
permitted study o f the evolution o f protein patterns and facilitated the approach to
gene duplication and gene expression problems during the developmental cycle o f
the medfly.
The extension o f the analyses o f genetic variability in the medfly at the DNA
level will provide opportunities for delving deeper into knowledge on the genetic
features o f the medfly, as well as for integrating new ideas and new tools for a many-
faceted approach. Among the recent molecular advances, random amplified poly­
morphic DNA (RAPD) analysis [11, 12] seems to be an extraordinarily powerful
approach to generating very quickly a large number o f markers that are useful in
population genetics analysis and in genome mapping o f the medfly.

2. POPULATION GENETICS

The use o f biochemical markers permits examination, at different levels o f

analysis, o f the genetic consequences o f the geographical spread o f this species. At
a first level, geographical genetic heterogeneity has been recognized in this species
range. At the second level, the study o f enzyme variability provided information on
the relative contribution o f different types o f evolutionary forces during the, process
o f colonization (selection, genetic drift and gene flow).
Intraspecific genetic differentiation is believed to be associated with the coloni­
zation history o f the medfly. From the point o f view o f the colonization pattern,
world populations o f C. capitata can be divided into three main categories:
(a) ancestral populations from sub-Saharan Africa; (b) ancient populations from the
Mediterranean basin and (c) new populations from South and Central America.
Genetic parameter estimates, such as the average number o f alleles per locus, the
proportion o f polymorphic loci and the average number o f heterozygous individuals,
seem to be correlated with the population’ s evolutionary history. In fact, they reveal
a trend o f decreasing genetic variability from the putative source area o f the species
towards the periphery o f its geographical range. These estimates, together with the
results o f multivariate and cluster analyses o f allele frequencies, confirmed the above
mentioned subdivision o f world medfly populations [3, 4].
Data on the spatial distribution o f allelic variation show that the pattern is
highly heterogeneous for different loci. Parameters o f variability or distances involv­
ing gene frequencies enlightened the dynamic aspects o f evolutionary processes,
 IAEA-SM-327/66 253

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

FIG. 1. Polymorphic amplification patterns obtainedfrom genomic DNA of Ceratitis capitata

flies from Kenya (lanes 1-7), from the Metapa laboratory strain (lanes 8-13) and from the
M84 line (lanes 14-19) . Lane 20 shows the molecular size markers. NP4 primer was usedfor
the amplification reaction.

such, as selection, drift and gene flow. In fact (i) comparison o f Fsx distribution
among loci led to the identification o f groups o f loci probably affected by selection;
(ii) genetic distance analysis permitted inference o f the amount o f drift between
geographical populations (this factor seems to have played a major role in the disper­
sion processes o f the peripheral populations); and (iii) gene flow estimates, in terms
o f the number o f immigrants per generation, are significant between the original and
the derived Mediterranean populations and support the hypothesis o f recent coloniza­
tion [13]. The sub-Saharan Kenyan population has all the properties o f a native popu­
lation and could belong to the centre o f diffusion o f the species. It has probably
maintained a very large size, as deduced by the high number o f rare alleles present
only in this population [14].
A parallel study o f protein and DNA variation by means o f allozyme and
RAPD markers is in progress on samples o f wild medfly populations and on labora­
tory strains. Although such markers may be influenced by different types and
degrees o f selection pressure, they would be affected similarly by such factors as
population size and gene flow. An unexpectedly high degree o f individual variability
in the genomic DNA is evident when the wild flies from Kenya are screened for
RAPD polymorphisms. The DNA o f the majority o f the Kenyan flies exhibits a
254 GASPERI et al.

unique pattern, differing in several fragments from that o f any other. The observed
profiles are typical DNA fingerprints and can be generated using different single
primers o f arbitrary sequence. A lesser degree o f genomic variability is detected in
the colonized laboratory strains which, however, maintain a certain degree o f DNA
polymorphism. An example is illustrated in Fig. 1, in which the polymorphic RAPD
profiles from African flies are compared with the ones derived from the Metapa
laboratory strain and from the M84 line. These data, as expected, support the previ­
ous results obtained with allozyme markers on the role of, drift in the colonization
process and on the high degree o f genetic variation present in the medfly African
ancestral populations. However, RAPD analysis, compared with the allozyme
approach, detects a larger amount o f hidden variation both in wild and laboratory,
medfly samples.
The polymorphisms in genomic fingerprints, generated by the RAPD
approach, provide a tool to improve the significance o f the estimates o f genetic relat­
edness between medfly populations, as well as to distinguish between slightly diver­
gent flies. A great deal o f information datà will be available for the evolutionary
population genetics o f the medfly. From the point o f view o f application, the avail­
ability o f markers which can unambiguously discriminate between strains as well as
samples o f C. capitata is o f utmost importance for the biological approach to medfly
control, such as the sterile insect technique (SIT).

3. LINKAGE MAPS AND COMPARATIVE GENETICS

The available genetic maps o f C. capitata, which rely heavily on isozyme

markers [6 , 7], provide useful information on the relationship between the genome
structure and genome function in this insect. To date, a total o f 30 biochemical
markers and 12 morphological and functional genes have been linked, linearly
arranged and mapped on five out o f six linkage groups o f C. capitata. These groups
include structural gene coding for enzyme functions, lethal and sex ratio distorter
genes and developmental proteins. Map distances on the five linkage groups show
that the marked loci are distributed over wide map intervals. Therefore, sufficient
loci are now available for extending the physical chromosome maps to also include
chromosome characters. Because many o f the biochemical loci considered have been
mapped in other Tephritidae flies [15], in Drosophila melanogaster [16] and in
Musca domestica [17], it is possible to compare the linkage arrangements o f
horthologous genes in these different species. This type o f analysis provides the basis
for establishing chromosomal homologies, and for examining the overall organiza­
tion o f the genome in these dipteran species.
For high resolution linkage mapping in C. capitata, the availability o f molecu­
lar markers is essential. High density genetic maps, comprised o f molecular
markers, may lead to the identification o f several previously unidentified loci o f bio­
 IAEA-SM-327/66 255

logical importance in C. capitata that can be exploited also to improve autocidal con­
trol o f this insect by genetic manipulation. From thè methodological point o f view,
the RAPD approach is easier and faster than other established molecular methods
used for genetic mapping. Using only one primer, in one multilocus cross, it is
possible to study the linkage relationships among several markers and to recognize
their linkage groups. Other advantages in utilizing RAPD polymorphisms are that
map planning can be approached without having to first identify RFLP probes and,
subsequently, RAPD generated DNA polymorphic fragments can be isolated directly
from gels and reamplified for use o f probes in restriction mapping strategies and for
mapping on polytene chromosomes by in situ hybridization.

4. CONCLUSIONS

Significant work remains to be done to examine the geographical distribution

o f DNA variation, particularly in light o f the contrasting distribution o f protein elec­
trophoretic variation observed at particular biochemical loci across medfly popula­
tions. As concerns the study o f medfly genome organization, the approach may
provide a new tool to explore and to completely map the genome o f this insect.

ACKNOWLEDGEMENTS

This research was supported by the National Research Council o f Italy, Special
Project RAISA, Sub-project 2, Paper N. 560. Grants from the IAEA also
contributed to this work.

REFERENCES

[1] F L E T C H E R , B.S., “Life history strategies of tephritid fruit flies”, Fruit Flies: Their
Biology, Natural Enemies and Control (ROBINSON, A.S,, H O O P E R , G., Eds),
Vol. 38, Elsevier, Amsterdam (1989) 195.
[2] MILANI, R., GASPERI, G., M A L A C R I D A , A., “Biochemical genetics (of Ceratitis
capitata)” , ibid., p. 35.
[3] GASPERI, G . ,G U G L I E L M I N O , C.R., M A L A C R I D A , A.R., MILANI, R., Genetic
variability and gene flow in geographic populations of the medfly Ceratitis capitata,
Heredity 67 (1991) 347.
[4] M A L A C R I D A , A.R., G U G L I E L M I N O , C.R., GASPERI, G., BARUFF I , L.,
MINALI, R., Spatial and temporal differentiation in colonizing populations of Ceratitis
capitata, Heredity 69 (1992) 101.
[5] M A L A C R I D A , A.R., etal., Genetical approach to systematics and phylogeny of
Trypetinae (Diptera: Tephritidae), Boll. Zool. 58 (1991) 355.
256 GASPERI et al.

[6] MALACRIDA, A.R., GASPERI, G., BARUFFI, L., MILANI, R., “ The contribu­
tion of formai genetic studies to the species characterization of Ceratitis capitata
(Wied.)” , Genetic Sexing of the Mediterranean Fruit Fly (Proc. Final Research Co­
ordination Mtg Colymbari, Crete, 1988), IAEA, Vienna (1990) 85.
[7] MALACRIDA, A.R., GASPERI, G., BARUFFI, L., BISCALDI, G.F.,
MILANI, R., “ Updating of the genetics of Ceratitis capitata (Wied.)” , Modem Insect
Control: Nuclear Techniques and Biotechnology (Proc. Symp. Vienna, 1987), IAEA,
Vienna (1988) 221.
[8] MALACRIDA, A., etal., Evidence for a genetic duplication involving alcohol
dehydrogenase genes in Ceratitis capitata, Biochem. Gen. 30 (1992) 35.
[9] GASPERI, G., BARUFFI, L., MALACRIDA, A.R., ROBINSON, A.S., A biochem­
ical genetic study of alcohol dehydrogenase isozymes of the medfly, Ceratitis capitata
Wied., Biochem. Genet. 30 (1992) 289.
[10] MALACRIDA, A.R., GASPERI, G., BISCALDI, G.F., MILANI, R., “ The gene
family of larval serum proteins (LSP) of Ceratitis capitata” , Fruit Flies of Economic
Importance ’87 (CAVALLORO, R., Ed.), A.A. Balkema, Rotterdam (1989) 283.
[11] WELSH, J., McCLELLAND, м ., Fingerprinting genomes using PCR with arbitrary
primer, Nucleic Acids Res. 18 (1990) 7213.
[12] WILLIAMS, J.G.K., KUBELIK, A.R., LIVAK, K.J., RAFALSKY, J.A., TIN-
GEY, S.V., DAN polymorphisms amplified by arbitrary primers are useful as genetic
markers, Nucleic Acids Res. 18 (1990) 6531.
[13] HAGEN, K.S., WILLIAM, W.W., TASSAN, R., Mediterranean fruit fly: The worst
may be yet to come, Calif. Agrie. 35 (1981) 5.
[14] NEI, М., MARUYAMA, T., CHAKRABORTY, R., The bottíeneck effect and
genetic variability in populations, Evolution 29 (1975) 1.
[15] BERLOCHER, S!H., SMITH, D.C., Segregation and mapping of allozymes of the
apple maggot, J. Hered. 74 (1983) 337.
[16] COLLIER, G.E., “Drosophila melanogaster, genetic map of biochemical loci” ,
Genetic Map 1987 (O’BRIEN, S.J., Ed.), Cold Spring Harbor Laboratory, Cold
Spring Harbor, NY (1987) 374.
[17] MALACRIDA, A., GASPERI, G., BISCALDI, G.F., MILANI, R., “ Signifícate?
fiinzionale di blocchi di geni concatenad in Musca domestica ed altri ditteri”
(MORONI, A., ANELLI, A., RAVERA, O., Eds), S. It. E. Atti 5 (1985) 307.
 IAEA-SM-327/46

COMPARISON OF Bacillus thuringiensis

AND Bacillus cereus

A.-B. K O LST0, C. CARLSON

Biotechnology Centre o f Oslo,
University o f Oslo,
Oslo, Norway

Abstract

COMPARISON OF Bacillus thuringiensis AND Bacillus cereus.

Bacillus cereus and Bacillus thuringiensis are closely related, spore forming soil
bacteria. B. thuringiensis produces insecticidal crystal proteins during sporulation and these
toxins are the most important biopesticide in thé world today. Genomes of the B. thuringiensis
and B. cereus strains were analysed by pulsed field gel electrophoresis after treatment of the
DNA with the restriction enzyme Notl. The Notl fingerprint patterns varied both within the
B. thuringiensis and the B. cereus strains. The size of the fragments varied between 15 and
1350 kb. When physical maps of the B. thuringiensis and B. cereus strains were compared,
B. thuringiensis appeared to be as closely related to B. cereus as the B. cereus strains were
to each other. Nine out of 12 B. thuringiensis strains and,18 out of 25 B. cereus strains
produced enterotoxins. The close relationship between B. thuringiensis and B. cereus should
be taken into consideration when B. thuringiensis is used as a biopesticide.

1. INTRODUCTION

Bacillus cereus and Bacillus thuringiensis are spore forming, common soil bac­
teria. They are very closely related, and in some taxonomic studies B. thuringiensis
is regarded as a variant o f B. cereus. This implies that the differences between the
strains may not be o f taxonomic importance — yet the differences may be o f vast
practical and medical importance.
B. thuringiensis is well known for its insecticidal activity, producing toxins
active against lepidopteran, dipteran or coleopteran larvae [1, 2]. The toxin genes
are usually localized on large plasmids, and the protoxins are found inside the
bacteria during sporulation as crystalline inclusion bodies. Four classes o f cry genes
are characterized and one bacterial strain may produce several toxins. Today,
B. thuringiensis is the most widely used biopesticide in the world, and several
companies produce toxin crystals/spores with different insect specificities. Hundreds
o f tonnes o f spores/toxins are used in various parts o f the world every year.
Like other members o f the Bacillaceae family, B. cereus secretes proteins very
efficiently. Some strains produce and secrete enterotoxins, and may cause food
poisoning in humans [3]. B. cereus is the most ,common contaminant in dairies [4].

257
258, KOLST0 and CARLSON

В. cereus and B. thuringiensis are very closely related, yet one is regarded as
a contaminant while the other is useful and o f great commercial interest. We wanted
to compare strains o f the two bacteria by analysing their DNA fingerprint patterns,
looking for characteristic patterns.

2. MATERIALS AND METHODS

2.1. Bacteria strains

The bacteria strains were obtained from the American Type Culture Collection
(ATCC), from the Bacillus Genetic Stock Center (BGSC), Ohio, or through
H. Nissèn o f Ás, Norway, from the Public Health Laboratory (PHLS), United
Kingdom. One strain (В. cereus 45) was isolated and provided by the National
Institute for Public Health, Oslo, Norway.

2.2. Preparation of the DNA

The bacteria were grown on liquid cultures, harvested by centrifugation and

embedded in low melting agarose plugs o f 10 0 fiL each, about 10 6 bacteria per plug
[5, 6 ]. The plugs were then treated with detergents and proteolytic enzymes that dis­
solved and degraded all the cellular components o f the bacteria, except the DNA.
The purified chromosomal (and large plasmid) DNA thus obtained was protected
within the agarose plug.

2.3. Analysis of the DNA by pulsed field gel electrophoresis

A DNA plug was incubated with a restriction enzyme that cut infrequently into
the genome, like Notl, which recognizes the sequence GCGGCCGC. The DNA
fragments were then separated by pulsed field gel electrophoresis, where the direc­
tion o f the electric field changes, thus allowing for separation o f DNA fragments o f
more than 1500 kb [7]. The fragment patterns from individual strains after staining
with ethidium bromide were analysed at pulse times ranging from 2 to 90 s; the
electrophoresis was routinely run for 18-24 h in a Beckman Gene Line apparatus.

2.4. Hybridization

. After staining the gel, the DNA was denatured and blotted on to supported
nitrocellulose filters [8 ]. The filters were baked at 80°C for 2 h and hybridized to
32P labelled DNA probes as described previously [6 , 9]. The probes used were
either genes or gene fragments from B. cereus (phospholipase С genes, beta-
lactamase genes), B. thuringiensis (the cryI gene), or B. subtilis (AbrB, the pyruvate
 IAEA-SM-327/46 259

dehydrogenase gene), or they were random fragments from a library o f B. cereus

ATCC 10987 prepared in our laboratory (BC probes). Hybridization was performed
in 3 x SSC/0.1% SD S/10% dextran sulphate (1 X SSC is 0.3M NaCl plus 0.03M
sodium citrate). The filters were washed for 2 x 30 min in each o f the solutions:
(1) 3 X SSC/0.1% SDS; (2) 1 X SSC/0.1% SDS; and (3) 0.3 x SSC/0.1% SDS.

2.5. Analysis of enterotoxin

Bacteria were grown overnight at 30°C , centrifuged (10 000g, 20 min, 4°C ),
sterile filtered and the enterotoxin was measured in the filtrate using a kit for
B. cereus enterotoxin, BCET-RPLA, obtained from Oxoid, UK.

2.6. Haemolytic activity

Bacteria were spread on agar plates containing 1 % human erythrocytes. Zones

o f haemolysis around the colonies were observed after incubation overnight at 37 °C.

(a) (b)

1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7

FIG. 1. Pulsed field gel electrophoresis of DNA from Bt strains after digestion with Notl.
The electrophoresis was run (a) at 60 s pulses at 150 mA for 22 h, and (b) at 30 s pulses at
150 mA for 18 h. (a) (1) Size markers, yeast chromosomes (Bio-Labs); (2) B. t. kurstaki;
(3) B. t. kenya; (4) B. t. canadensis; (5) B. t. entomocidus; (6) B. t. dendrolimus;
(7) B. t. subtoxicus; (8) B. t. morrisoni; and (9) B.t. darmstadiensis. (b) (1) Yeast chromo­
somes; (2) B. t. kurstaki; (3) B. t. kenya; (4) B. t. canadensis; (6) B. t. morrisoni; and
(7) B. t. darmstadiensis.
260 KOLST0 and CARLSON

(a) (b)
1 2 3 4 5 6 7 1 2 3 4 5 6 7

FIG. 2. Pulsed field gel electrophoresis of DNA from B. cereus strains after digestion with
Notl. The electrophoresis was run (a) atóOs pubes at 150 mAfor 20 h, and fb) at 30 s pulses
at 150 mA for 18 h. (a) (1) Yeast chromosomes; (2) B. cereus ATCC 11778; (3) B. cereus
ATCC 1457; (4) B. cereus ATCC 10876; (5) B. cereus ATCC 10987; and (6) B. cereus F
4810/72; (7) F837/76. (b) (1) Size marker, lambda concatemers; (2) B. cereus 6EI;
(3) B. cereus 6E2; (4) B. cereus 45; (5) B. cereus ATCC 21281; (6) B. cereus ATCC 27877;
and (7) B. cereus ATCC 38018.

3. RESULTS

3.1. Notl fingerprint patterns

Both the B. cereus and B. thuringiensis strains contained Notl fragments

ranging in size from 15 to 1350 kb (Figs 1 and 2). The Notl fingerprint patterns o f
eight Bt strains (Fig. 1) were fairly similar in some strains (В. t. entomocidus,
B. t. dendrolimus and B. t. subtoxicus), while the fragment patterns were
very different in other strains (B. t. kurstaki, B. t. kenya, B. t. canadensis,
B. t. morrisoni and B. t. darmstadiensis). A similar variation among the B. cereus
strains was observed (Fig. 2(a), (b)).

3.2. Physical maps

Probes were assigned to individual Notl fragments by hybridization (Fig. 3).

The sizes of. the neighbouring fragments were obtained when partial Notl digests
were analysed. We used more than 50 probes and aligned all the Notl fragments o f
 п ш 1У V

А В А В

F kb

1000-

770-
Í '

1200 -
kb
kb
500-
юоо- И
*»"!

IAEA-SM-327/46
t .
790- 270-

i -
410-

50- i PDH
200 -

Вс 209 Вс 201

Вс 72
рАК!

FIG. 3. (A) Pulsedfield gel electrophoresis of В. cereus ATCC 11778 Notl fragments, and (B) autoradiography of blotted
Notl fragments. The fragments were separated at 60 s pulse time at 150 mA for 20 h (I, II, III and IV), or at 30 s pulses
(V). The probes used for hybridization were as indicated in (B).
ю
O'
262 KOLST0 and CARLSON

P LC
Abr В
Be 37 PDH

Be 37

FIG. 4. Comparison o f maps-of one B. thuringiensis and four B. cereus strains.

four B . c e r e u s strains [6 , 9], as well as one B . t h u r i n g ie n s is strain [10]. When these

maps were compared (Fig. 4), it was observed that: (1) the B . t h u r i n g i e n s i s strain
appeared to be as closely related to the four B . c e r e u s strains as the four B . c e r e u s
strains were to each other; and (2 ) the differences between the strains may be
explained by inversions/deletions/insertions.

3.3. Haemolytic activity and enterotoxin production

When the haemolytic activity and the enterotoxin production were measured
in 12 B . t h u r i n g i e n s i s strains and 25 B . c e r e u s strains, most of the strains were
haemolytic, and more than half of both the strains were positive for enterotoxin
(Table I).

TABLE I. HAEMOLYTIC ACTIVITY AND ENTEROTOXIN PRODUCTION

BY THE B . c e r e u s AND B . t h u r in g i e n s i s STRAINS

B. cereus B. thuringiensis

Positive Negative Positive Negative

Haemolytic activity 23 2 1 0 2

Enterotoxin production 18 7 9 3
 IAEA-SM-327/46 263

4. CONCLUSIONS

From these studies it was concluded that:

(1) The DNA from strains of B . t h u r i n g i e n s i s and В . c e r e u s shows a characteristic

fingerprint pattern when their Notl fragments are analysed by pulsed field gel
electrophoresis. The pattern appears to be specific for the individual strains.
Although a few strains were closely related, no typical B . t h u r i n g i e n s i s or
B . c e r e u s pattern was discovered.
(2) B . t h u r i n g i e n s i s strains cannot be distinguished from B . c e r e u s strains by
DNA fingerprint analysis. Our results are in keeping with data from other
laboratories, indicating that B . t h u r i n g i e n s i s strains are insecticidal variants of
B . cereu s.
(3) Nine out of 12 B . t h u r i n g i e n s i s strains produce enterotoxins that may cause
diarrhoea in humans. This information should be taken into consideration when
B . t h u r i n g i e n s i s is used as a biopesticide.

REFERENCES

[1] ARONSON, A .I., BECKMAN, W ., DUNN, P., Bacillus thuringiensis and related
insect pathogens, Microbiol. Rev. 50 (1986) 1-24.
[2] HÓFTE, H., WHITELEY, H.R., Insecticidal crystal proteins of Bacillus thuringien­
sis, Microbiol. Rev. 53 (1989) 242-255.
[3] TURNBULL, P.C.B., Bacillus cereus toxins, Pharmacol. Ther. 13 (1981) 453-505.
[4] CHRISTIANSEN, A ., NAIDER, A .S ., NILSSON, I., WADSTRÔM, T.,
PETTERSON, H .-E ., Toxin production by Bacillus cereus dairy isolates in milk at low
temperatures, Appl. Environ. Microbiol. 55 (1989) 2595-2600.
[5] SMITH, C .L ., KLCO, R.S., CANTOR, C.R ., “ Pulsed-field gel electrophoresis and
the technology of large DNA molecules” , Genome Analysis: A Practical Approach
(DAVIES, K.E., Ed.), IRL Press, Oxford (1988) 41-72.
[6 ] KOLST0, A .-B., GR0NSTA D , A ., OPPEGAARD, H., Physical map of the Bacillus
cereus chromosome, J. Bacteriol. 172 7 (1990) 3821-3825.
[7] SCHWARTZ, D ., CANTOR, C.R ., Separation of yeast chromosome-sized DNAs by
pulsed field gradient electrophoresis, Cell 37 (1984) 67-75.
[8 ] SAMBROOK, J., FRITSCH, E.F., MANIATIS, T ., Molecular Cloning: A Labora­
tory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (1989).
[9] CARLSON, C ., GR0NSTAD, A. KOLST0, A .-B ., Physical maps of the genomes of
three Bacillus cereus strains, J. Bacteriol. 174 11 (1992) 3750-3756.
[10] CARLSON, C ., et al., (submitted for publication).
POSTER PRESENTATION

IAEA-SM-327/23P

MOLECULAR GENETICS OF INSECTS

Oogenesis and maternal control
o f early development

C. MALVA, F. GRAZIANI
Istituto Internazionale di Genetica e Biofísica,
Consiglio Nazionale delle Ricerche,
Naples,
Italy

The central interest of our research concerns the molecular bases of maternal
control of development in D r o s o p h i l a m e l a n o g a s t e r and other insect species. We are
studying a group of female semi-sterile and maternal effect mutants, located in
region 32 of the standard salivary gland map of the second chromosome of
D r o s o p h i l a . Besides detailed genetic analyses to define the phenotypic aspects of the
mutation, we analysed the structure of ovaries and early embryos produced by
homozygous mutant females using DAPI or propidium iodide staining. At the
molecular level we isolated by chromosome walking 250 kb of DNA from region
32D-E, performed a restriction enzyme site polymorphism analysis in w i l d t y p e and
mutant stocks and identified a copia like b l o o d transposon inserted in region 32E of
the a b o chromosome. We demonstrated that the a b o phenotype is strictly correlated
with the presence of the b l o o d transposon. We identified transcripts on the two sides
of the transposon insertion site and isolated corresponding cDNA clones. From
sequence analysis of the cDNAs and the 10 kb genomic region, we identified a gene
whose putative product has an average 60% identity with the guanylate cyclase/atrial
natriuretic factor receptors (sea urchin, rat, mouse, human). The data obtained by
all the different approaches appear to indicate that this group of genes is involved
in general processes such as cell proliferation and/or cell migration. As part of the
study of the maternal effect mutants in region 32, we identified molecularly a vitel­
line membrane protein (VMP) gene, called V M P 3 2 E , whose regulatory sequences
were dissected by using in vitro mutagenesis coupled with P element mediated germ
line transformation.

ACKNOWLEDGEMENTS

This work was supported by Consiglio Nazionale delle Ricerche Target Pro­
jects on Biotechnology and Bioinstrumentation (C. Malva) and Genetic Engineering
(F. Graziani) and is part of a FAO/IAEA Agreement (No. 5116/CF).

265
OPERATIONAL PROGRAMMES
(Session 4)

Chairman
L.E. LACHANCE
FAO/IAEA
 IAEA-SM-327/28

GENETIC CONTROL OF COTTON INSECTS

The pink bollworm as a working program m e

R.T. STATEN
Plant Protection and Quarantine,
Phoenix Methods Development Center,
Animal and Plant Health Inspection Service,
United States Department of Agriculture,
Phoenix, Arizona

R.W. ROSANDER
Plant Protection and Quarantine,
Animal and Plant Health Inspection Service,
United States Department o f Agriculture,
Sacramento, California

D.F. KEAVENY
Control and Eradication Division,
California Department of Food and Agriculture,
Bakersfield, California

United States of America

Abstract

GENETIC CONTROL OF COTTON INSECTS: THE PINK BOLLWORM AS A WORK­

ING PROGRAMME.
Establishment of a continuous population has been prevented over a 24 year period in
the San Joaquin Valley, USA, through continuous, daily in-season release of sterile pink
bollworms based on an extensive trap monitoring programme. A post-harvest crop destruction
ordinance and occasional use of pheromones as disruptants were the only other factors used
by programme management, except in 1990. In 1990, the programme used a conventional
insecticide on 280 acres (113 ha) out of 1.18 million acres (477 546 ha) of cotton. During the
four year period 1986-1989, a management system was explored using a high rate pheromone
disruption system and sterile insects. Major reductions in conventional insecticide usage,
while maintaining extremely low populations, were evident in this semi-isolated valley of
southern California. It is hoped that this will provide a model for a future large scale test on
up to 2 0 0 0 0 acres (8100 ha) of cotton.

1. INTRODUCTION

The use of genetic control mechanisms for cotton insects has been explored to
varying degrees with several insects, including the tobacco budworm, H e l i o t h i s
v i r e s c e n s (F.), and the boll weevil, A n t h o n o m is g r a n d i s Boheman. Their use and

269
270 STATEN et al.

actual application on a large scale have been limited, however, to the pink bollworm,
(Saunders). This paper covers that usage and its potential
P e c t in o p h o r a g o s s y p ie lla
for expansion.
It is important to have a historical perspective of the pink bollworm problem
in the United States of America. At the present time, it is a problem in the western
USA, restricted to the irrigated desert areas. Pink bollworms are occasionally
recovered from states east of Texas , with only one endemic population on wild cotton
in the southern extreme of Florida. The Floridian population has total geographical
isolation from any commercial cotton growing area in the USA.
The pink bollworm is most destructive in Arizona, southern California and the
adjacent northwestern Mexican desert cotton growing areas. Its introduction and
establishment in central Arizona in the mid-1950s and into the Colorado River Basin
of western Arizona, southern California and northwestern Mexico in 1965 have led
to the development of a major pest problem with extreme side effects on the whole
ecosystem. This is, of course, exacerbated by à long growing season which, when
pushed for maximum production, results in crop longevity, giving pink bollworm
tremendous population potential. The extensive use of broad spectrum conventional
pesticides, including the newer synthetic pyrethroid compounds, has resulted in
expensive control systems with secondary pest problems. The only major area in the
western cotton belt where the impact of the pink bollworm has not been felt by the
grower is in the central California San Joaquin Valley. In this area, the establishment
of the pink bollworm has been prevented through the use of sterile insect release
technology, minor use of pheromones as a disruptant and adequate cultural control.
This paper briefly outlines the details and provides programme statistics for .that
project. In addition, it discusses efforts to integrate sterile release technology with
pheromone systems and optimum cultural practices in the heavily infested southern
California desert.

2. THE SAN JOAQUIN EXCLUSION PROGRAMME

The San Joaquin Valley has never had a known continuous endemic population
of the pink bollworm, even after the spread of this insect into the Colorado River
desert growing areas in 1965 and 1966. The San Joaquin Valley is approximately
175 miles (282 km) (direct line) from any generally infested, areas of southern
California. The entire 175 miles is a host free zone. This expanse of desert and
mountains has formed a p a r t i a l barrier to pink bollworm movement. Movement
within the desert region between the generally infested southern California deserts
and the San Joaquin Valley has been documented by Stern and Savacherian [1] and
by trap lines maintained by programme personnel in the mountain passes between
the desert cotton growing areas and the San Joaquin Valley. Extensive movement
into the desert area has been documented early in the spring, late in the summer and
 IAEA-SM-327/28 271

TABLE I. ANNUAL SUMMARY (1985-1989)

OF THE MATING PROPENSITY OF STERILE
PINK BOLLWORM FEMALES SHIPPED TO
THE SAN JOAQUIN VALLEY, AS DETER­
MINED IN THE LABORATORY

% mated
Year
(48 h)

1985 43

1986 70

1987 81

1988 76

1989 8 6

1990 85

1991 87

throughout the fall. Movement over mountain passes into the adjacent San Joaquin
Valley has been documented in the fall. Programme trap data further provide evi­
dence of movement within the San Joaquin Valley, since late season increases in
detection have coincided with captures in the desert trap lines. All of these data have
been gathered by using gossyplure baited traps placed along highways in the deserts
between cotton growing areas and in field traps.
In 1968, a programme was initiated to protect the 500 000 ha of cotton in the
San Joaquin Valley by releasing irradiated pink bollworm moths. Initially, the
released moths were irradiated with 25 krad of gamma rays, but since 1973 the dose
has been lowered to 20 krad. When sterile insects are treated at this dosage, they
will produce eggs but most of these progeny do not survive to reach the adult stage.
These few adults are infertile. Thus, the main purpose of lowering the dosage to
2 0 krad was to obtain more competitive males rather than to employ F! sterility.

Sterile insects are now reared in Phoenix, Arizona, using techniques or modifi­
cations of techniques described by Stewart [2]. Attention to detail and constant
refinements of these rearing processes since 1 9 7 1 have improved the q u a n t i t y a n d
quality of insects available for shipment.
The increase in quality is illustrated in Table I, which summarizes quality
control mating tests carried out by programme personnel. These data come from
samples taken each day of shipment (seven days a week) throughout the season.
272 STATEN et al.

TABLE IL ANNUAL PRODUCTION OF THE

PHOENIX PINK BOLLWORM REARING
FACILITY, 1970-1991

Year Total moths x 106

1970 135

1971 1 2 0

1972 113

1973 99

1974 41

1975 154

1976 177

1977 192

1978 301

1979 385

1980 372

1981 401

1982 497

1983 594

1984 579

1985 489

1986 702

1987 735

1988 804

1989 826

1990 809

1991 813

Three samples of 20 pairs are held in cages in the laboratory for 48 h before
dissection. An additional set of samples are checked for mating on arrival in the
release zone. Those females which contain spermatophores are considered mated.
Table I shows an increased mating propensity between 1985 and 1991 after small
procedural improvements were made in shipping. Mean mating levels of samples
taken directly from shipment containers (0 h) did not exceed 3.3% for any year.
 IAEA-SM-327/28 273

Laboratory space was increased by approximately 15% during the 1970-1989

period. All other production increases were technical. The increases in production
from 1973 to 1991 are illustrated in Table II.
The Phoenix United States Department of Agriculture (ÚSDA)/California
Cotton Growers funded facility was the only provider of sterile insects for the San
Joaquin Valley from 1970 to 1977 and from 1983 to the present. It has also provided
significant numbers of pink bollworms for research work, including sterile insect
research. These trials included those in the Moapa Valley of Nevada [3], St. Croix
[4] and in the Coachella Valley of California (reported here). Pheromone isolation
and identification projects [5, 6 ], were also supported.
From 1977 to 1982, a commercial ‘ satellite’ pink bollworm rearing facility was
operated for eight months of each year. This facility also provided moths for the
San Joaquin Valley.
Sterile insects are shipped at 40 ± 5°F (4.44 ± -15°C ) from Phoenix,
Arizona, to Bakersfield, California, where they are transferred (still in a chilled
condition) to aircraft and released at approximately 500 ft (152 m) above the cotton
fields. The time of collection from eclosion chambers to the time of release is from
24 to 48 h. Insects are held in an immobile state by chilling during this time.
The sterile and native populations are currently monitored with pheromone
traps baited with gossyplure in a controlled release formulation. The Delta trap [7]
has been used since 1974.
With the advent of gossyplure and its availability in 1974, the conduct of this
programme has become much more precise. For this reason, this paper discusses
programme operations since that time. A summary of pertinent information is given
in Table III.
The number of traps has varied from 12 850 to 33 711. Traps have been placed
at average densities varying between one trap per 39.5 acres (15.99 ha) to one trap
per 79.8 acres (32.3 ha). Traps are serviced weekly. The data produced by the traps
is used to schedule the release of sterile insects. From 1974 to 1991, the area
requiring release of sterile insects ranged from 4.6 to 29.5% of the total San Joaquin
cotton production area.
Each year the season starts with releases on cotton in sections (640 acres
(259 ha)) which had moth finds from the previous year. Thus, the percentage of the
total acres released varies. As appropriate (depending upon native moth capture rates
for the previous year), buffer zones are also released around these sections. Moth
releases are controlled at the section level (640 acres). As new, native (unmarked)
pink bollworms are found, the releases are adjusted accordingly.
Moth release typically begins before the first flower buds are present on the
cotton plants (pin square) and continue with daily shipments of moths throughout the
season. Releases of sterile insects are normally terminated each year during the last
two weeks of October or. the first two weeks of November.
TABLE III. GENERAL STATISTICS FOR THE SAN JOAQUIN PROGRAMME, 1974-1991
4^

No. of acres3 No. of traps Acres3 per Maximum acres3 Per cent Mean No. of steriles
Year
mapped x 1 0 0 0 maintained x 1 0 0 0 trap x 1 0 0 released x 1 0 0 acres3 released per acre3 released

1974 1275 32 39.5 101 7.9 366

1975 8 6 6 2 2 40.0 117 13.1 1 287
1976 1130 27 41.0 90 7.9 2 141
1977 1353 33.7 40.0 400 29.5 1 030
1978 1501 19.6 76.3 355 26.6 1 286
1979 1625 2 0 .6 78.8 254 15.6 2 512

STATEN et al.
1980 1463 2 0 .8 70.2 177 1 2 .1 2 880
1981 1451 19.4 74.8 352 24.3 2 251
1982 1326 16.9 78.3 147 1 1 .0 5 260
1983 1005 12.9 78.2 . 132 13.1 4 452
1984 1392 17.6 79.0 42 6 .1 13 329
1985 1336 17.2 77.9 59 4.6 8 167
1986 1061 13.3 79.8 44 4.1 14 930
1987 1 2 1 2 19.9 60.8 63 4.6 10 864
1988 1368 22.7 60.4 71 5.2 10 655
1989 1128 19.2 58.8 104 9.2 7 258
1990 1184 19.5 60.7 129 1 0 .8 5 438
1991 1103 18.4 59.9 140 1 2 .6 5 806

3 1 acre = 0.405 ha.

TABLE IV. STERILE MOTH RELEASE AND TOTAL MOTH CAPTURE DATA IN THE SAN JOAQUIN VALLEY

Sterile moths Per cent sterile No. of sterile Non-sterile moths No. of sterile insects
Year
released x 1 0 6 moths recovered moths recovered recovered (native) captured per native

1974 37.01 0.76 282 897 437 647

1975 150.11 1 .0 1 1 528 260 245 623.8
1976 173.64 0.63 1 229 742 1474 834
1977 412.17 0.40 1 677 900 7402 2 2

1978 455.99 0.09 429 063 69 621.8

1979 636.98 0.09 545 295 754 723

IAEA-SM-327/28
1980 510.49 0 .1 1 566 170 4492a 126
1981 794.26 0 .1 1 864 861 677 127.7
1982 772.12 0.13 1 041 280 1 2 0 8 677.3
1983 586.77 0.18 1 057 735 863a 1 225.6
1984 571.50 0.38 2 144 018 351a 6 108.3
1985 483.74 0.09 434 739 160 2 717.1
1986 660.35 0.05 352 825 62 5 690.7
1987 670.07 0.15 1 057 925 294 3 598.7
1988 754.87 0.16 1 188 335 891 1 335.7
1990 701.50 0.27 1 943 863 3239b 600.14
1991 812.97 0.45 3 6 6 8 337 263 10 105.6

a Larvae finds of four larvae: four larvae found in respective years.

275
b In 1990, two fields had a cumulative average peak of 15.2% boll infestation.
276 STATEN et al.

The number of sterile moths released has increased steadily since 1974
(Table IV). Major improvements in rearing technology and management systems in
1975 represented a significant improvement in programme operations. A very
dramatic degree of improvement in rearing is seen when one compares the numbers
of moths shipped to the San Joaquin Valley in 1975 compared with 1974. Shipments
increased from 37.01 x 106 moths in 1974 to 150.11 x 105 in 1975, and
173.6 x 10° in 1976. Although native unmarked populations were low early in
1974, it became apparent that a major problem existed as the season progressed. The
total sterile moth captures of 282 897 was 647 times higher than the non-sterile
recovery, but an acceptable ratio1 of 60 sterile: 1 native moth at the individual field
level could not be maintained when native moth capture rates became very great in
August and September. It was very fortunate that improvements were brought on line
before the 1975 season. The production in 1975 was able to cover the entire
‘infested’ area adequately from the beginning of the season.
Relatively high native moth capture (1474) late in the season in 1976 followed
tropical storms originating off Baja California, Mexico and resulted in a decision to
augment sterile moth production at the USDA facility by outside contract production
for the 1977 season. This contract provided 28% of the 412 million moths used in
the San Joaquin Valley in 1977 and provided an average of 39% of all production
until the 1983 season. In 1983, the 586 million moths shipped were provided exclu­
sively by the USDA facility in Phoenix. '
In 1980, we captured the highest number of native moths in programme
history., Unlike Í974, however, coverage of the entire area where moths were
detected was achievable throughout the season. Major efforts were also made
through grower meetings throughout this area to optimize cultural control through
rapid-xrop destruction after harvest and early plough down of host material.
The following year, 24% of the Valley received sterile insects. This rate is high
in comparison with the preceding few years of release, but with good winter mortal­
ity; between the 1980 and 1981 season, it was apparently adequate.
In 1990, following a warm dry winter in which those cotton plants left standing
wère not frozen, populations developed predominantly in two areas within the
Valley. For the first and only time, a conventional insecticide was used against a
larval population. This population was large enough in two fields to be measured
with conventional sampling procedures. Guthion was applied on 28 September and
3 October on 280 acres (113 ha)- The crop in these fields.was chemically terminated
rapidly following these treatments and. cultural measures of the harshest nature rela­
tive to the pink bollworm were token. It must be emphasized that this was, the only
time in the 24 years that sterile insects have been released in the San Joaquin Valley
that a conventional insecticide was used against the pink bollworm. The only other

1 Through experience, the ratio of 60 sterile moths: 1 native moth has been established
as à working, sáfe ratio. " ' ;
 lÀEÀ-SM-327/28 277

in-season treatment available to programme personnel has been the use of

gossyplure as a mating disruptant. Its use has never exceeded 0.2% of the total
acreage, or 4.6% of any year’ s release area. Our problem was not repeated in 1991.
To date in 1992, native moth capture rates are at a very low level.
When one tries to assess this programme, with its massive number of traps and
the extensive amount of data they generate, it is apparent that any analysis would be
highly speculative in nature. There are no check plots and/or adjacent identical
infested areas.
It should also be noted that until the present time, we have used resources (the
sterile moth) on an as-available basis. We are now constructing in the Animal and
Plant Health Inspection Service (APHIS), with the co-operation of Texas A&M
University, a heuristics based expert system to help the decision maker utilize the
extensive database that this project generates. We acknowledge that although the
project has its limitations, two pertinent points have not been missed by the cotton
industry. The first is that there is no detectable ongoing pink bollworm population
in the San Joaquin Valley, and second, that our rearing technology has improved
significantly and consistently over the years.

3. INTEGRATION OF THE STERILE INSECT !SYSTEM •

WITH PHEROMONE DISRUPTION

Once the pink bollworm spread into the Colorado River Basin, the entire cotton
pest management system changed drastically. Cotton growers were determined to
continue maximizing yields and were quick to utilize regularly scheduled insecticide
application. This philosophy continued, essentially unabated,, for approximately
twenty years. In the most severely infested areas (such as in the Imperial Valley),
grower leadership is now seeking other remedies. In addition to working with sterile
insect technology, our laboratory, the Agricultural Research Service and industrial
co-operators have worked extensively in the development of pheromone communica­
tion disruption. The large scale field testing of the Mitsubishi pink bollworm rope
in 1985 [8 ] in Imperial Valley provided a breakthrough. This .breakthrough led to
the carrying out of an ‘area wide’ management project in the Coachella Valley on
its restricted acreage.
The Coachella Valley is reasonably isolated during the main growing season
and is measurably affected by migration only during storm conditions that include
surface winds from the southeast, usually late in the season. In 1986, we started
working in a co-operative venture with the growers of the area and their pest control
advisor (M. Grummet of Foster Gardner Chemical Co.). Our first step in 1986 was
to treat the entire valley with the high rate pink bollworm rope system, at 30 g of
active ingredient (AI) per ácre in one application at the six to eight leaf cotton
development stage. Our results were so encouraging that between 1987 and 1989 we
 К)
'J
00

TABLE V. CONTROL STRATEGIES AND CONVENTIONAL INSECTICIDE USE PATTERNS FOR PRE-MANAGEMENT
AND AREA WIDE MANAGEMENT SYSTEMS IN THE COACHELLA VALLEY

Fields treated Mean No. of No. of fields treated Cumulative conventional

Treatment system
with pheromone conventional insecticides with insecticides insecticide treatments

STATEN et al.
1985 insecticide only
(pre-trial data) 7.27 56/57 414

1986 pheromone and insecticides 31/31 1.8 17/31 56

1987 sterile insects, pheromone
and insecticides 17/21 1.03 7/27 28

1988 sterile insects and pheromone 4/31 0 0/31 0

1989 sterile insects, pheromone

and insecticides 3/23 1.9 18/23 44
TA B L E V I. S U M M A R Y OF B O LL D A T A F R O M TH E C O A C H Ë L L A V A L L E Y PINK B O L L W O R M MANAGEM ENT
S T U D Y , S H O W IN G T O T A L L A R V A E PER SA M P LE A N D N U M B E R O F BOLLS SA M P LE D

1986 1987 1988 1989

Sample No. of bolls Total No. of bolls Total No. of bolls Total No. of bolls Total
week sampled larvae sampled larvae sampled larvae sampled larvae

IAEA-SM-327/28
1 50 3 370 0

2 500 3 500 0 726 0 560 0

3 960 2 1700 0 1 2 0 0 0 720 0

4 1180 0.9 2400 7 1600 1 960 0

5 1520 1 2400 33 2 0 0 0 0 1520 0

6 1520 3 2500 34 2160 0 1600 1

7 1600 0 2700 11 2320 0 1760 0

8 1520 1 2700 6 2480 5 1564 18

9 1600 1 2700 5 2480 5 1860 39

1 0 .1760 43 1350 25 2480 2 1960 39
11 . 2 2 0 0 138 1350 41 2480 0 1240 47

279
280 STATEN et al.

TABLE VIL DATA FROM BOLL SURVEYS SELECTED RANDOMLY FROM

FIELDS IN IMPERIAL VALLEY, CA, PARKER VALLEY, AZ, AND ALL
COACHELLA FIELDS

No. of larvae per 100 bolls

Date Imperial Valley3 •Coachella Parker

1-5 July — 0 —

6-14 July 7.84 0 —

17-21 July 8.84 0 —

24-28 July 9.5 0.06 3.6

31 July - 4 August 11.40 0 5.4

7-11 August ■- 15.34 1.15 11.05

14-18 August 25110 1.99 16.2

21-25 August 30.7 1.98 18.86

28 August T 1 September 34.34 3.8 21.89

3 -6 September . — — 23.02

10-13 September — — 16.98

24-27 September — — 14.15

a Data adjusted from Boll Box Survey to equate with standard cutting procedures [4].

started releases of 500 sterile pink bollworms per day. This was done in connection
with a minimum shortening of the growing season. Under this system, sterile insect
release started at the three to four leaf cotton stage. If a ratio of 60:1 sterile to native
insects was not achieved in a given field at a six leaf stage, that field received a
pheromone treatment. Thus, while not all fields would receive pheromone, all fields
received sterile insects. If the chemical control thresholds were exceeded at any time
in the season, the grower and his pest control advisor were expected to utilize
chemical control to complete the growing season. This treatment was on a field by
field basis, as needed.
Pink bollworm sterile and native populations were monitored with Delta traps
at 2 traps per 40 acres (16.19 ha) and with samples of 80-100 bolls collected per
field each week. The bolls were then returned to the laboratory for examination
under large lighted 3 X magnification lenses. Samples were taken from the second
week in July through August.
 IAEA-SM-327/28 281

M a n a g e m e n t t r i a l r e s u l t s : The overall treatment strategies, pheromone usage

and insecticide use patterns are shown in Table V. Most important, we achieved
major reductions in conventional pesticide usage in the four years that we operated
this programme with pheromone and combined pheromone and sterile insect strate­
gies, as shown in the table.
Before 1986, the majority of cotton growers in the Valley used expensive mul­
tiple chemical pesticide applications for control of the pink bollworm even though
other crops, such as grapes, dates, citrus and vegetables, were considered to be more
important and were planted on the better land. Cotton is usually grown on more
marginal ground at as low a cost as possible. In the most successful year, 1988, no
conventional organic insecticides were used. In 1988, pink bollworm movement
from Imperial Valley to Coachella Valley was not significant until the last weeks of
August, as detected by traps in the desert between the two valleys. In 1989, Imperial
Valley terminated its cotton by ceasing all irrigation in early to mid-August. All the
130 fields were defoliated chemically by 1 September, with only 12 exceptions.
Movement was very apparent across the desert from the last weeks of July and
throughout August. Our control with minimal insecticide in Coachella Valley was
still acceptable, as illustrated by both the pesticide use patterns (Table V) and by low
larval populations (Table VI).
It should be noted that, although use of conventional insecticides was being
restricted significantly, the overall population levels were held at drastically reduced
levels. This is particularly true in comparison to the larval populations in Imperial
Valley and Parker Valley (Arizona) in 1989. This should be noted when comparing
our data with those from co-operative monitoring projects in Imperial Valley [9] and
a Parker, Arizona, project (Table VI) [10].
The larvae population in Table VII are from boll samples taken all season long
in 45 fields in Parker, Arizona, 55 fields in the Imperial Valley, California, and in
our test fields in Coachella. In Imperial and Parker, all fields were treated heavily
with conventional insecticides during this time.

4. SUMMARY AND FUTÜRE PROSPECTS

The twenty-five year history of the San Joaquin Valley Exclusion Project is
considered by the California cotton industry to be a major success. The industry
funds their programme at more than 85 % of its total cost and has opted to spend
US $2.1 million for two new rearing buildings, with options for major expansion in
order to continue this programme.
In addition, these same growers and the USDA are pursuing an expanded trial
most likely to be in the Imperial Valley. This valley is ideal as a target for expanding
these techniques becaiise it normally has much larger acreage than Coachella and is,
as a cotton growing area, under extreme duress. Normal cotton production of
282 STATEN et al.

140
m

1 Jul. 15JuL 29 Jul. 12 Aug. 26 Aug. 9 Sep. 23 Sep. 7 Oct.

W eeksam pled

FIG. 1. Comprehensive larval infestation in the Imperial Valley before and after reduction
in season length expressed as larvae per 100 bolls ( ■ : 1990; 13 ; 1989).

40 000-80 000 acres (16 188-32 376 ha) has been the rule over past years. Cotton
production has slipped to approximately 4000 acres (1618 ha) largely due to the
extreme cost of treating pink bollworm infestations and a new biotype of the white-
fly, Bemisia. In addition, growers in this, community have opted to drastically
shorten the growing season, which has had a marked impact on the pink bollworm
population. Our laboratory, in co-operation with the Imperial County Commissioner
of Agriculture and the Agricultural Research Service, monitored the pink bollworm
population extensively in 1989 before any winter cultural programme could affect
the population of that year. Populations were thus mentioned in an identical fashion
in the season of 1990, following a mandatory fall termination in 1989. The popula­
tions have been drastically reduced, as evidenced by adult moth counts in Delta traps
and boll infestation data (Fig. 1) [11]. In addition, test releases in 1990 clearly
indicated that a 500 sterile moth per acre per day release was feasible.
The importance of these approaches and the necessity of finding a better
management scenario are imperative in today’s ecologically concerned society. We
can no longer live with major insecticide usage for pink bollworm and its associated
secondary pest problems. The introduction of a new biotype of whitefly has further
drastically affected the entire agricultural ecosystem. In addition, we cannot keep
ahead of insecticide resistance for key pests such as the pink bollworm and the white-
fly now found in cotton. We must reduce environmental hazards from pesticide use,
including drift and residue. The pink bollworm problem, with today’ s knowledge,
can only be dealt with on an organized, area wide, rational basis.
 IAEA-SM-327/28 283

REFERENCES

[1] STERN, V ., SAVACHERIAN, V ., Long range dispersal of pink bollworm into the
San Joaquin Valley, California Agrie. 32 (1978) 4-5 .
[2] STEWART, F.D., “ Mass rearing the pink bollworm Pectinophora gossypiella” ,
Advances and Challenges in Insect Rearing (KING, E.G ., LEPPLA, N .C ., Eds),
Agricultural Research Service, United States Department of Agriculture, Washington,
DC (1984).
[3] STATEN, R.T., unpublished.
[4] HENNEBERRY, T.J., KEAVENY, D .F., Suppression of Pink Bollworm by Sterile
Moth Releases, Agricultural Research Service, United States Department of Agricul­
ture, Phoenix, AZ (1985).
[5] HUMMEL, H.E., et al., Clarification of the chemical status of the pink bollworm sex
pheromones, Science 181 (1973) 873-875.
[6 ] BIERL, B .A ., BEROZA, М ., STATEN, R.T., SONNET, P.E., ADLER, V .E ., The
pink bollworm sex attractant, J. Econ. Entomol. 67 (1974) 211-216.
[7] FOSTER, R .N ., STATEN, R .T ., MILLER, E ., Evaluation of traps for pink bollworm,
J. Econ. Entomol. 70 (1977) 289-291.
[8 ] STATEN, R.T., et al., Pink bollworm (Lepidoptera: Gelichiidae): Large scale field
trials with a high rate gossyplure formulation, J. Econ. Entomol. 80 (1987) 1267-1271.
[9] WEDDLE, D ., CHU, C .C ., STATEN, R.T., HENNEBERRY, T.J., BIRDSALL, S.
(unpublished).
[10] STATEN, R.T., ANTILLA, L., MYERS, F., CROUT, R. (unpublished).
[11] CHU, C .C ., et al., “ Pink bollworm: Populations two years following initiation of a
short-season cotton system in the Imperial Valley, C A ” , Beltwide Cotton Production
(Proc. Conf. Nashville, TN, 1992) (BROWN, J., Ed.), National Cotton Council of
America, Memphis, TN (1992).
 IAEA-SM-327/29

IMPLEMENTATION OF THE
STERILE INSECT RELEASE PROGRAMME
TO ERADICATE THE CODLING MOTH,
Cydia pom onella (L.) (LEPIDOPTERA: OLETHREUTIDAE),
IN BRITISH COLUMBIA, CANADA

V.A. DYCK, S.H. GRAHAM

Agriculture Canada Research Station,
Summerland

K.A. BLOEM
Okanagan-Kootenay Sterile Insect Release Program,
Osoyoos

British Columbia,
Canada

Abstract
IMPLEMENTATION OF THE STERILE INSECT RELEASE PROGRAMME TO ERADI­
CATE THE CODLING MOTH, Cydia pomonella (L.) (LEPIDOPTERA: OLETHREUTI-
DAE), IN BRITISH COLUMBIA, CANADA.
The sterile insect release (SIR) programme to eradicate the codling moth, Cydia
pomonella (L.), from the Okanagan region by the year 2000 has begun. The SIR programme
includes about 8000 ha of apple and pear trees. In many orchards, the cessation of insecticidal
sprays for codling moth control should permit apples to be grown without pesticide applica­
tions during the fruit development period, a major environmental and economic benefit.
Research done by M .D. Proverbs and colleagues over twenty years has established techniques
for rearing, sterilizing and releasing codling moths. However, the SIR costs estimated from
a pilot project were more than twice those of chemical sprays to control the pest. Nevertheless,
recent studies have shown that a programme would be economical if only the minimum
required number of moths was released; if the eradication area was treated in steps and if
reinfestation was prevented. The British Columbia Fruit Growers’ Association helped to
develop an implementation plan which included a budget, a revenue scheme and a political
and administrative framework. The plan was approved by the municipal governing bodies in
the region, as well as the Provincial and Federal Governments. Enabling legislation was
passed in 1989. Funds for equipment and a rearing facility to produce about five million moths
per week were provided by the two senior governments, and the municipal governing bodies
will collect property taxes and parcel taxes based on the orchard area to cover the operational
costs. The first phase of the programme, wild population reduction, started in 1992; the
second, sterile moth release, will begin in Í994, and the third, prevention of reinfestation, will
start in 1997. Recent improvements in the rearing procedures will increase efficiency and
production security, and reduce worker health hazards. The integrated pest management
systems in apples and pears may require some changes when the moth release phase of SIR
begins.

285
286 DYCK et al.

1. INTRODUCTION

Thirty years of research and planning have finally developed into the sterile
insect release (SIR) programme to eradicate the codling moth, C y d i a p o m o n e l l a (L.),
by the year 2000 from the Okanagan region of south-central British Columbia,
Canada. This region includes about 8000 ha of apple and pear trees, mostly in
commercial orchards. Besides the Okanagan Valley, portions of other valleys and
districts are included, e.g. the Similkameen Valley, Salmon Arm and Crestón, to
incorporate all the areas which grow commercial fruit and exchange fruit and fruit
containers.
The codling moth is regarded in this region as the most important pest of
apples, and as a major pest of pears. Avoiding codling moth dàmage, and reducing
the use of expensive hazardous insecticides, are of great monetary and environmental
benefit to growers and consumers. There is potential for the development of pesticide
resistance in the pest population, and restrictions on pesticide use are becoming more
common. In many orchards, it is expected that the cessation of chemical sprays for
codling moth control will permit apples to be grown without spraying any pesticides
during the. fruit development period, and such apples could have a marketing advan­
tage. A. SIR programme will reduce the need for increasingly costly, foreign made
chemical insecticides, while encouraging local employment.

2. BACKGROUND RESEARCH

Research done over twenty years by M.D. Proverbs and colleagues at Agricul­
ture Canadà’ s research station in Summerland has established the techniques for
rearing, sterilizing and releasing codling moths [1, 2]. Related research has been
conducted in the Yakima Agricultural Research Laboratory (United States Depart­
ment of Agriculture) in the neighbouring state of Washington, USA [3]. In the pilot
project (1976-1978) [2]. that was the culmination of Proverbs’, research, about
500 ha o f apple and pear orchards in the Similkameen Valley were treated, first with
insecticides to reduce the wild moth population, and then with reared sterile moths
to eradicate the pest. The technology was effective in rearing and releasing the sterile
moths, which appeared to eradicate the wild population, at least in some local areas.
The project was considered to be a field success, and received strong grower
support.
In the pilot project, about 1 000 000 moths were reared each week on a saw­
dust based artificial diet [2, 4]. About 10% of the moths were put in oviposition
cages to mate and lay eggs for the next production cycle, while the remaining 90% ,
both sexes in approximately equal numbers, were released after exposure to
 IAEA-SM-327/29 287

35 krad1 gamma radiation in a 60Co irradiator. This dose induced about 92% steril­
ity in the male moths and complete sterility in the females. The diet included Calco
red dye, which made the internal organs turn red. (The organs of the wild moths are
yellow in colour, and therefore the origin of the moths caught in field traps can be
determined by t h is c o l o u r difference.) Chilled moths were released via a specially
designed dispenser mounted on dune buggies that were driven through the orchards.
The desired sterile:wild ratio of 40:1 was, on average, exceeded.

3. ECONOMIC STUDIES

After the pilot project was completed, analysis of the annual costs showed that
these were more than twice those of chemical sprays to control the pest [2]. Such
a high price for SIR, despite its environmental benefits, was considered unacceptable
to growers and precluded implementation. However, there were those in research
and amongst the growers (especially organic growers in the Similkameen Valley)
who hoped that another review of SIR, which would include a wide range of eco­
nomic options and a long term view, might yield a positive result. Agriculture
Canada, notwithstanding the unfavourable economic implications of the pilot
project, decided to fund several studies by agricultural economists to review codling
moth SIR and reassess its commercial feasibility [5-7].

3.1. The Holm reports [5, 6]

A complete review of the strategy, methodology and costs of SIR was made.
Various scenarios for a possible future programme were considered. The concept of
three years of moth release followed by three years of monitoring [2 ], with the
expectation that the cycle would be repeated, was abandoned in favour of assuming
that eradication would occur after three release years and that post-release protection
activities against reinfestation could maintain the pest free status. To reduce capital
costs, a small rearing facility was proposed which could produce enough moths to
service only one-sixth of the total area. At three years of release for each of the six
zones, 18 years would be required to eradicate the moth from the whole area. To
further reduce costs, only the minimum required number of moths estimated by
Proverbs [ 8 ] , plus 25% to cover the rearing risks, would be produced. The SIR costs
were, compared with the chemical control costs, and no environmental costs or
savings were considered per se because of the difficulty of estimating such figures.
The encouraging conclusion of this study was that a commercial eradication
programme would be feasible economically in the long run if changes in the strategy
and operations were made.

1 1 rad = 1.00 X 10"2 Gy.

288 DYCK et al.

The second report [6 ] described sensitivity analyses of the various risks and
uncertainties in a SIR programme, assessing what impact changes in the various
components of the programme would have on its final economic feasibility. The
major aspects considered were the risk of contamination or disease reducing moth
production, the release ratio and, therefore, the number of moths to be reared in rela­
tion to the somewhat unknown size of the wild population, the method of moth
release, the cost of chemical sprays, the risk of other pests requiring chemical
sprays, and the risk of reinfestation after eradication. The results of the analyses indi­
cated that the number of moths required for field release, and the cost of chemical
sprays, were the most important aspects in assessing the cost benefits of SIR over
insecticidal treatments. Ground release of moths was more cost effective than aerial
release with rented helicopters. Other aspects had only a minor impact on SIR eco­
nomics, but the programme would become uneconomical if the apple maggot became
established and required frequent sprays for control, since the codling moth would
be controlled partially by these sprays.

3.2. The Hansen report [7]

This report continued the analysis of a commercial SIR programme. A larger

production facility and larger zones (three zones instead of six) provided a more eco­
nomical result than in the previous model, and also shortened the time period needed
to treat all the orchards. A benefit-cost ratio between 1.1 and 1.4 was expected. The
cost of chemical sprays, and the average number of sprays applied per year, were
reduced. Motorcycles with sidecars were chosen as the most cost effective vehicles
for ground release.

4. IMPLEMENTATION PLAN - THE DEBIASIO REPORT [9]

Since the effectiveness of SIR technology had been demonstrated in the pilot
project [2 ], the economic studies had given positive predictions about the financial
robustness of a codling moth SIR programme, and the benefits to apple and pear
growers were both monetary and environmental, the British Columbia Fruit
Growers’ Association decided to investigate the feasibility of implementing a
programme over the whole area represented by its members. The growers obtained
a grant from the Federal and Provincial Governments to hire a consultant to prepare
an implementation plan. The plan was to recommend practical operating procedures
that included economies of scale, to show the time relationships between the different
activities, to review programme costs, to develop a revenue scheme, to prepare a
budget for each year of operation, and to propose a political and administrative
framework for the programme.
 IAEA-SM-327/29 289

A committee of growers, government entomologists, economists and horticul-

turalists and municipal government officials met frequently with DeBiasio to guide
the development of the implementation plan. ■
DeBiasio found that the proposed programme was very strong economically,
with a benefit-cost ratio of 1 .8 8 and an internal rate of return of 2 2 %.
The key recommendations in the implementation plan were as follows:

(1) To requ est fu n d s fr o m F ed eral and P r o v in c ia l a g e n c ie s fo r c a p ita l

ex p e n d itu r es .
(2) T o o b ta in re v e n u e f o r o p e r a t io n a l c o s ts th ro u g h lo c a l p r o p e r ty a n d p a r c e l
ta x es. Since all the members of the communities would benefit somehow from
a SIR programme, it was considered appropriate to tax all the properties in the
SIR programme area as well as to tax all apple and pear orchards through the
parcel tax.
(3) T o c o l l e c t p r o p e r t y t a x e s f o r e i g h t y e a r s o n l y to help cover the costs in the
initial years of the programme, especially the first five years, when costs
would be high and parcel taxes would not yet have been collected in all the
areas. The property tax rate would vary from year to year according to the
need for revenue, but it would be rather low — possibly ranging from Can.
$0.01 to Can. $0.26 per Can. $1000 of property value (not including
improvements).
(4) T o b a s e p a r c e l ta x es o n th e n u m b e r o f h e c t a r e s p la n t e d to a p p le s a n d p e a r s .
The parcel tax would commence when the sterile moths were being released,
and continue indefinitely. Through the parcel tax, the growers would pay by
far the largest part of the operational cost. The proposed parcel tax rate
(Can. $100/ha in 1988 dollars) was intentionally set lower than the average
cost of spraying chemicals (Can. $126/ha) so that growers would realize an
economic advantage. (The spray cost was based on 2.5 sprays per season for
the codling moth, and included the cost of insecticide, the operating cost of
application equipment and owner labour cost.) DeBiasio proposed that the
moths be released for three years in one-half of the area (Zone 1), and then
in the second half (Zone 2) for the next three years, each zone containing about
4000 ha of orchard. (A natural break between orchard areas was identified as
the line between zones.) The Can. $ 100/ha parcel tax should be levied for
six years to growers receiving the moths first (Zone 1, the area where the
codling moth causes the most damage), and for three years to growers receiv­
ing the moths last (Zone 2). After that, when eradication would have been
achieved, the rate would be reduced progressively to pay indefinitely the cost
of maintaining the area free of codling moth.
(5) T o o b ta in s p e c i a l le g is la tio n fr o m t h e P r o v in c ia l G o v e rn m en t t o e s t a b lis h th e
O k a n a g a n - K o o t e n a y S t e r i l e I n s e c t R e l e a s e B o a r d w it h in t h e p o l i t i c a l a n d
a d m in istra tiv e fr a m e w o r k o f r e g io n a l d is tr ic ts , units of local municipal govern-
290 DYCK et al.

ment covering both the rural and urban areas. Five regional districts
(Okanagan-Similkameen, Central Okanagan, North Okanagan, Columbia-
Shuswap and Central Kootenay) would work together, each district appointing
one member of its board of elected officials to represent it on the SIR Board.
Other people representing various interested parties, such as grower organiza­
tions and Federal, Provincial and municipal agricultural departments, should
be included as non-voting Board Members. The SIR Board would be empo­
wered to levy and collect the property and parcel taxes through the regional
district taxation system. Also, the Board should have authority to operate and
administer the' SIR programme, to enter property as required, and to enforce
the programme activities.
(6 ) T o c o n s t r u c t a r e a r i n g f a c i l i t y t o p r o d u c e m o t h s f o r 4 0 0 0 h a a t a t im e . Two
release zones, instead of three or six, would enable all the growers to receive
the benefits of the programme within a reasonable time. However, making two
zones instead of one would still allow for a relatively small, and therefore rela­
tively inexpensive, rearing facility. The facility design provided was only a
sketch based on enlargement of the existing research facilities, with some
rearrangements to improve the sanitation and material flow and to reduce the
costs.

5. SIR PROGRAMME PHASES

There are three distinct phases of the programme: prerelease sanitation, rear­
ing and release of sterile moths, and protection against reinfestation. For a given
zone, the three phases follow each other in time, and operations in one zone are
independent of those in the other zone. The current programme schedule is as
follows:
Z on e Г. 1992-1993, prerelease sanitation; 1994-1996, rearing and release of
sterile moths; 1997, protection against reinfestation.
Z one 2: 1995-1996, prerelease sanitation; 1997-1999, rearing and release of
sterile moths; 2 0 0 0 , protection against reinfestation.

5.1. Prerelease sanitation (two years)

Since the efficiency of SIR technology increases when the number of sterile
moths becomes much greater than that of wild moths, the goal of prerelease sanita­
tion is to reduce the wild moth population as much as possible using other control
methods, primarily the application of insecticides.
Apple and pear trees are located, mapped and then categorized as abandoned,
non-commercial or commercial orchard trees. Abandoned trees provide a refuge for
 IAEA-SM-327/29 291

codling moths and must be cut down. Few abandoned trees exist in Zone 1 because
of the low rainfall, but in the wetter Zone 2 there is a significant number of aban­
doned trees.
Non-commercial trees are found primarily in urban areas, in backyards and in
very small orchards. It is difficult to monitor the wild population size in these trees,
so control actions need to be prophylactic rather than remedial. The recommended
methods of control include well timed chemical sprays, blossom or fruit removal
early in the season, and placement of corrugated cardboard strips around the tree
trunks and limbs to trap the mature larvae searching for cocooning sites.
Commercial orchard trees are monitored for moth density by using sex phero­
mone traps. The SIR programme employs monitors travelling in all terrain vehicles
(ATVs) to count the moths caught in the traps each week. Trap data and control
recommendations are given weekly to orchardists to enable them to apply appropri­
ate and timely control tactics.
Release of marked sterile male moths into orchards followed by sex phero­
mone trap counts of marked and wild moths, especially if done at the peak emergence
period of a brood, provide a measure of the sterile:wild moth ratio. This ratio indi­
cates how low the wild population density is, and is a measure of the readiness of
an orchard for full scale sterile moth release. Observations in 1990 and 1991 in
several Similkameen Valley orchards using this method showed that some orchards
were ready, but many were not.
At harvest, field samples of fruit can estimate the level of damage incurred,
especially damage that indicates mature larvae have exited and will overwinter in the
orchard. Such estimates permit a projection of the number of moths that will emerge
in the first brood the following year. Again, such observations in the Similkameen
Valley showed that some orchards had low codling moth populations, but others not.
On the basis of the number of sterile moths scheduled for release in the SIR
programme, to maintain a sterile:wild ratio of >40:1, an overwintering larval
density of <100 larvae/ha the previous winter may be acceptable. The damage level
that would generate such a larval density is hard to determine because of the variable
crop sizes.and the kind of control tactics used, but codling moth damage levels of
0.1-0.3% may be acceptable, depending on the particular orchard. Many growers
who spray insecticides appear to achieve codling moth damage levels of 0 .2-0.3%
[10]. Proverbs advocated a very low prerelease damage level of 0.05% [2]; such a
level is desirable if it can be achieved with existing control technology.

5.2. Rearing and release of sterile moths (three years)

The production goal of the SIR programme during peak production is 5.2 mil­
lion moths per week derived from 10 500 trays of diet. After emergence, the male
and female moths will be collected and then released from ATVs within one or two
days in orchards. Moths for 1 ha will be released while the ATVs travel 300 m
292 DYCK et al.

between the tree rows. Releases will be carried out twice a week in each orchard,
and continue for about 20 weeks to cover both codling moth broods. Sex pheromone
traps, set up periodically for one night only in selected orchards, will catch both
sterile and wild male moths and provide an estimate of the sterile:wild ratio.
The new rearing facility will produce 4 000 000 moths per week for release.
If 4000 ha are treated twice a week, then each hectare of orchard will receive
500 moths at each release, and over a 20 to 25 week season up to 25 000 moths.
Since up to 36 500 moths/ha were released in the pilot project [2], admittedly achiev­
ing high sterile:wild ratios, it is very important that the current SIR programme
emphasizes prerelease sanitation as well as moth production and quality in order to
achieve the goal of eradication within the targeted time period.
If 100 larvae/ha overwinter in an orchard, only about 50 larvae will survive
the winter and emerge as adults during the spring brood. If temperatures are hot,
up to 2 0 moths ( 1 0 males, since the sex ratio is 1 : 1 ) might emerge within one week.
If 500 sterile male moths are released per hectare per week, then the sterile:wild ratio
is a satisfactory 50:1 in the week of the brood when keeping a high ratio would be
the most difficult; in other weeks, the ratio should be even higher.
A mathematical model [11], based on the sterile insect technique work of
E.F. Knipling, indicates that an average sterile:wild ratio of 40:1 in the first brood
leads to extinction of the wild population in a 4000 ha zone by the end of the third
brood (1.5 years) if certain assumptions are true. These assumptions are: that the
natural population growth rate in the first brood each year is 5, and in the second
brood 1 0 ; that sterile male competitiveness is the same as in wild males; that male
and female sterility are both 100%; and that the sex ratio is 1:1. In the SIR
programme, the radiation dose that may be used, 35 krad, causes only 92% sterility
in male moths. As a result, there will be more surviving eggs and larvae in each
brood than expected, and this will cause some fruit damage but, since the mortality
of these larvae will be higher than usual and most will be males [ 1 2 ], the number
of female moths produced in each brood will be similar to what the model predicts.
Also, ? ! sterility of the few adult progeny from sterile male matings with fertile
females will reduce the threat to population growth, since the progeny will be
virtually sterile [ 1 2 ]; they will behave like released sterile moths.
Releases in urban areas will be done from special release boxes that allow the
moths to escape but prevent birds from entering the boxes. Experimental releases in
the city of Penticton indicated that placement of boxes every two blocks should
provide a good distribution of moths.

5.3. Protection against reinfestation (indefinitely)

Three major activities will be required after eradication — monitoring for the
presence of wild moths, releasing sterile moths at border sites to prevent invasion
of wild moths, and controlling the transfer of infested fruit containers.
 IAEA-SM-327/29 293

Monitoring with sex pheromone traps must be continued indefinitely in the

whole eradication area. Should even one wild moth be trapped, immediate action
could be needed to eradicate the possible new infestation. Therefore, the colony
would have to be maintained so that sterile moths could be released at any time.
There is one major corridor of reinvasion into the Okanagan region, i.e. the
border area at Osoyoos, where US orchards lie adjacent to Canadian orchards. To
prevent fertile moths in the USA from flying into Canada, the programme will
release sterile moths on the border, or even in the nearby US orchards, to create a
zone that is free of wild moths. There may be other corridors of reinvasion at the
periphery of the SIR area, and if wild moths are found entering the eradication zone
through these corridors, they would also have to be treated with sterile moths. One
mitigating factor that will slow down any reinvasion process is the sedentary habit
of female moths. While males fly some distance, perhaps kilometres, females usually
fly only short distances, perhaps metres, in their lifetime. Reinvasion is unlikely in
the Okanagan region because of the narrow host range of the codling moth (virtually
only found on apples and pears, which are irrigated crops in a dry region), and the
narrow, deep valleys in which these crops are grown.
Considerable transfer of large fruit bins, used to transport >fruit from the
orchard to the packing house, occurs in the Okanagan region. This transfer could
involve long distances, since fruit may be moved from one valley to another. If
codling moth larvae emerge from infested apples after harvest, the larvae may spin
cocoons in the bin crevices and remain there in diapause all winter. Bins are placed
in orchards during the spring and summer after they have been emptied to await
harvest, and any larvae would then continue development and adults emerge, ready
to infest trees. This threat of reinfestation via bin transfer can be eliminated if the
original source of the bins is identified and if they are returned to their source. Heat
treatment of bins to kill any resident larvae may be necessary if the transfer of
infested bins cannot be prevented, but the cost of treatment could be high.

6 . GETTING STARTED (1988-1992)

Following the completion of DeBiasio’ s report [9] on SIR: programme

implementation, the leaders of the British Columbia Fruit Growers’ Association met
several times to obtain clarification and: to discuss the implications of the
programme. After careful consideration, they agreed that the proposed implementa­
tion plan was acceptable.
In the next step the concurrence of all local governments (in the SIR
programme area) in the five regional districts was obtained. In 1989, the Provincial
Government passed Bill 75, which provided a legal basis for the regional districts
to collect taxes for and operate the programme. The SIR Board was formed. A
request for capital funds was made to the Federal and Provincial Governments.
294 DYCK et al.

In 1990, the Provincial Government requested three consultants to evaluate the

programme plans. Two consultants, W.H. Sudlow with experience in screwworm
eradication and G.G. Rohwer with experience in Mediterranean fruit fly eradication,
reviewed the SIR field operations plans. Both scientists generally approved of the
plans. The third consultant, G.G. Hartley, who had experience in rearing many
different insect species, reviewed the plans for the rearing facility. Hartley’s report
recommended significant changes to the building design and rearing operations,
especially in relation to sanitation, contamination control, production security,
quality control and worker safety and health. A redesign of the building by members
of the SIR Board indicated that the cost of the new design would be much greater
than that of the old, and that this could create problems in obtaining capital funds.
In 1991, the Federal and Provincial Governments agreed to provide funds for
preliminary work on the building design so that more accurate estimates of the
construction costs could be made. An engineering firm was contracted to do the
work. The preliminary cost estimate for the facility was about Can. $ 8 million, a
good deal more than DeBiasio’s [9] estimate of Can. $3 million. The SIR Board and
the two governments eventually set the capital budget at Can. $7.7 million. By the
end of 1991, the capital funds were assured. Construction of the rearing, facility
began in 1992 in Osoyoos, at the extreme southern end of the Okanagan Valley in
Canada; it should be operational by early 1993. Additionally, in 1992 some staff
were hired and the first year of prerelease sanitation was initiated in Zone 1.

7. IMPROVEMENTS IN SIR TECHNOLOGY

The technology used during the pilot project [2] is the basis of the commercial
SIR programme. However, some modifications in the technology have been made
to improve prerelease sanitation, quality control, production security and the work­
ing environment, and to reduce costs.

7.1. Prerelease sanitation

To assist in controlling the codling moth, pheromone mating disruption was

tested on an experimental basis in some, especially organic, orchards; the moth
populations were usually greatly reduced [13].
Considerable automation and use of electronic equipment have been introduced
into programme operations. Trap data are recorded in the field with electronic data
collectors and transmitted to the computer, at programme headquarters by modem.
Staff can then immediately analyse the field data, both through database software and
a geographical information system that permits mapping of the moth infestations.
 IAEA-SM-327/29 295

7.2. Rearing

To improve the quality and competitiveness of the reared sterile moths, Hutt
[14] reared codling moths under daily fluctuating temperatures. He found that more
male moths reared under fluctuating temperatures than under constant temperatures
were caught in the sex pheromone traps during cool evenings. We confirmed this
finding in a similar experiment in the Okanagan region, and therefore moths destined
for release in spring or autumn should be reared under fluctuating temperatures. In
another of our experiments during warm weather, however, just as many constant
temperature reared sterile moths as released wild moths were caught in the traps,
indicating that constant temperature rearing should be acceptable in the summer.
This last experiment also showed that reared moths responded to the traps as well
as wild moths, indicating that the flight and pheromone responses of the reared males
should be satisfactory for a SIR programme.
To maintain vigour in the laboratory colony, and prior to starting a colony in
the new SIR rearing facility, infested apples were collected from commercial
orchards in the Okanagan region in 1992. For about six weeks, the male moths
reared out of these apples were added to oviposition cages, and the same number of
colony males removed, to increase the heterogeneity of the colony.
The moths that will form the nucleus of the SIR colony should be free of the
codling moth granulosis virus. Efforts are being made to create a virus free colony
which will be transferred to the rearing facility.
The competitiveness of reared sterile male moths with wild males in mating
wild females appears to be satisfactory in laboratory tests.,
A modification made to the existing oviposition cages [15] reduced by 50% the
time needed to remove the paper egg sheets each day. Water wicks with reservoirs
inside the cages to provide the moths with water for the duration of oviposition took
less time to install than inserting wet wicks into the cages every day. A series of
experiments revealed the appropriate number of moths to place in these cages in
order to obtain the number of eggs that would lead to optimum production in the next
brood.
Two methods of rearing will probably be used at the SIR facility: (1) the usual
sawdust diet held under long day length for continuous rearing, and (2 ) a soft diet
held under short day length for larvae that will spin cocoons outside the diet, enter
diapause and then be stored in the cold until a time of year when large numbers of
moths are required.
In the pilot project [2], moths that were emerging at large in an emergence
room were collected by attracting them to UV lighted cages on the floor. In an
attempt to avoid the health hazards of entering a room with moth scales in the air,
to make moth collection less labour intensive and to collect high quality moths, the
concepts of this collection system and that used by Stewart [16] for pink bollworm
296 DYCK et al.

moths were combined in a prototype automated moth collector that was built and
tested, and then further developed for the new facility.
In, collaboration with the engineering firm, the new rearing facility was
designed to incorporate many features that have improved the quality control, rear­
ing efficiency and production security, and reduced the rearing costs and health and
safety hazards to workers.

8 . INTEGRATED INSECT PEST MANAGEMENT SYSTEMS

The big change for these systems in apples and pears will be the elimination
of chemical insecticides for codling moth control. The most commonly sprayed
chemical at present is azinphosmethyl, a broad spectrum organophosphate. Cessation
of these sprays, applied after petal fall until near harvest, will have an impact on the
control of other insect pests. Also, the application of insecticides for other insects
could kill the released sterile moths.
It is expected that the major pests in apple that could require sprays in the sum­
mer are the eyespotted bud moth, S p i l o n o t a o c e l l a n a (D. & S.), the white apple leaf-
hopper, T y p h l o c y b a p o m a r i a McAtee, and two species of two generations-a-year leaf
rollers. Reports from the pilot project [2], work done by Madsen and Downing [17]
and Madsen [18], and recent experiments at Summerland indicated that few summer
pest problems are likely to require sprays in SIR apple orchards, especially if spring
sprays have been applied. Should sprays be needed to save the crop, chemicals such
as azinphosmethyl, diazinon and endosulfan will kill some released codling moths,
and therefore releases should be stopped for a few days after treatment. This will
have little effect on SIR, since wild moths will also be killed by the sprays. An
approach to controlling other pests would be to apply chemical treatments in the
spring as a first priority and, if needed, biological control agents that would not harm
the released moths, such as B a c i l l u s t h u r in g i e n s i s ( B t ) , during the summer.

REFERENCES

[1] PROVERBS, M .D ., “ Sterile insect technique in.codling moth control” , Sterile Insect
Technique and Radiation in Insect Control (Proc. Symp. Neuherberg, 1981), IAEA,
Vienna (1982) 85-99.
[2] PROVERBS, M .D ., NEWTON, J R., CAMPBELL, C.J., Codling moth: A pilot
program of control by sterile insect release in British Columbia, Can. Entomol. 114
(1982) 363-376.
[3] HUTT, R.B., WHITE, L .D ., Status of Sterile Insect Technique Investigations on
Codling Moth at the Yakima Agricultural Research Laboratory, Rep. SEA ARM-W-9,
United States Department of Agriculture, Oakland, CA (1979).
 IAEA-SM-327/29 297

[4] BRINTON, F.E., PROVERBS, M .D ., CARTY, B.E., Artificial diet for mass produc­
tion of the codling moth, Carpocapsa pomonella (Lepidoptera: Olethreutidae), Can.
Entomol. 101 (1969) 577-584.
[5] HOLM, W .R ., An Evaluation of the Commercial Cost of a Sterile Insect Release
Control Program for Codling Moth in British Columbia, Agriculture Canada, New
Westminster, BC (1985).
[6 ] HOLM, W .R ., Analysis of the Risks and Costs of a Sterile Insect Release Program for
the Control of Codling Moth in the Okanagan Region of British Columbia, Agriculture
Canada, New Westminster, BC (1986).
[7] JECK, S., HANSEN, J., Economics of Codling Moth Control by Sterile Insect
Release: Benefit-Cost Approach, Agriculture Canada, New Westminster, BC (1987).
[8 ] PROVERBS, M .D ., “ Codling moth control by the sterility principle in British Colum­
bia: Estimated cost and some biological observations related to cost” , The Sterile-
Insect Technique and its Field Applications (Proc. Panel Vienna, 1972), IAEA, Vienna
(1974) 81-88.
[9] DEBIASIO, D ., Codling Moth Sterile Insect Release Study, British Columbia Fruit
Growers’ Association, Kelowna, BC (1988).
[10] HOLDER, D ., “ Pack-out analysis of size, colour and cullage” , Apples — Cultural
Practices to Meet Present Market Needs (Proc. 18th Annu. BCFGA Horticultural
Forum, Penticton, 1986) 24-29.
[11] WEIDHAAS, D .E., Rnipling Life Lotus Sheet for SIT, computer program on floppy
disk (1990).
[12] PROVERBS, M .D ., NEWTON, J.R., Some effects of gamma radiation on the
reproductive potential of the codling moth, Carpocapsa pomonella (L.) (Lepidoptera:
Olethreutidae), Can. Entomol. 94 (1962) 1162-1170.
[13] JUDD, G.J.R., GARDINER, M .G .T ., THOMSON, D.R., Management of codling
moth with pheromones, Br. Columbia Orchardist 15 3 (1992) 12-15.
[14] HUTT, R.B., Codling moth (Laspeyresia pomonella (Lepidoptera: Olethreutidae)):
Improving field performance of mass-reared males, Can. Entomol. I l l (1979)
661-664.
[15] PROVERBS, M .D ., LOGAN, D .M ., A rotating oviposition cage for the codling moth,
Carpocapsa pomonella, Can. Entomol. 102 (1970) 42-49.
[16] STEWART, F.D., “ Mass rearing the pink bollworm, Pectinophora gossypiella” ,
Advances and Challenges in Insect Rearing (KING, E.G ., LEPPLA, N .C ., Eds),
Agricultural Research Service, United States Department of Agriculture, New Orleans,
LA (1984) 176-187.
[17] MADSEN, H.F., DOWNING, R.S., Integrated control of the fruit-tree leaf roller,
Archips argyrospilus (Walker), and the eye-spotted bud moth, Spilonota ocellana
(Denis & Schiffermuller), J. Entomol. Soc. Br. Columbia 65 (1968) 19-21.
[18] MADSEN, H .F ., Integrated control of the fruit-tree leaf roller and the white apple leaf-
hopper in British Columbia, J. Econ. Entomol. 62 (1969) 1351-1353.
 IAEA-SM-327/30

ADVANCES IN SHEEP BLOWFLY

GENETIC CONTROL IN AUSTRALIA

G.G. FOSTER, G.L. WELLER, W.J. JAMES,

K.M. PASCHALIDIS, L.J. McKENZIE
Division of Entomology,
Commonwealth Scientific and Industrial
Research Organization,
Canberra,
Australia

Abstract

ADVANCES IN SHEEP BLOWFLY GENETIC CONTROL IN AUSTRALIA.

Economic analysis indicates that the benefit: cost ratio of eradicating the sheep blowfly
(SBF), Lucilia cuprina, from Australia would be highly favourable. One strategy would be
to combine trapping with either the genetically impaired female technique (GIFT) or the sterile
male technique (SMT), both of which use genetic sexing systems. GIFT strains are complex
genetic sexing systems that contain chromosome rearrangements and other genetic mutations.
Mass reared females are killed or debilitated because they are homozygous for recessive muta­
tions. Released males transmit the mutations and rearrangements by mating with field females.
The genetic death caused by partial sterility of the rearrangements and homozygosis for the
mutations in the field is sufficient to cause population collapse, leading to eradication. GIFT
strains currently held contain sex linked translocations, eye colour mutations, homozygous-
viable pericentric inversions and temperature sensitive lethal mutations. Field trials of GIFT
have demonstrated the successful suppression of target populations. However, the mass rear­
ing methods used during the trials are not suitable for large scale use. There is a need for
collaboration by entomologists and engineers on R&D to develop cost effective rearing
systems. An idea under consideration is a modular facility capable of simultaneously rearing
more than one pest species. This could initially be used for SBF eradication, then for the sterile
insect technique (SIT) against other major pests such as fruit flies, codling moth and heliothis,
and be available for SIT campaigns against any incursions by exotic pests.

1. INTRODUCTION

The sheep blowfly (SBF), L u c i l i a c u p r i n a , is an important myiasis pest of

sheep in Australia. The annual costs of prevention and production losses average
A $135 million [1].
Insecticide usage against SBF has led to the evolution of resistance to cyclo-
dienes, organophosphates and carbamates [2]. Resistance to the currently used tria-
zines has not yet been reported.

299
300 FOSTER et al.

The spectre of insecticide resistance and the success of the sterile insect tech­
nique (SIT) against the screwworm in the United States of America led to the initia­
tion of genetic and ecological research into the genetic control of SBF [3].
A review of the ecological studies is beyond the scope of this paper. Neverthe­
less, their importance in the development of genetic control must be emphasized.
They have established a quantitative basis for planning and assessing genetic and
other control measures [4, 5]. They have permitted the use of ecological approxima­
tions [6 ] in the assessment of trials, the design of release strains and the evaluation
of tactics. Models based on these studies should save many millions of dollars in
large scale control programmes.
Until recently, the genetic control programme was guided by the assumption
that, while eradication would be feasible in isolated areas, this goal could be exces­
sively ambitious for the entire continent. Therefore, we concentrated on developing
systems suitable for long term suppression of SBF populations rather than
eradication.
Recently, simulations combining genetic and ecological models suggested that
these suppression systems were also capable of eradication [7]. Moreover, at low
release rates they could be much more effective than SIT. Thus, both long term
suppression and eradication are currently being considered as feasible options.
The paper summarizes these and other developments which have made the
genetic control of SBF more commercially attractive.

2. THE GENETIC METHODS AVAILABLE

The genetic methods available for SBF include SIT, the sterile male technique
(SMT) and the genetically impaired female technique (GIFT). Genetic sexing sys­
tems form the basis of both SMT and GIFT.
GIFT was formerly called the field female killing (FK) system [7]. This system
combines chromosome rearrangements with eye colour and other mutations, such
that mass reared females are killed or debilitated. The mutations and rearrangements
are inherited in the offspring of field females mated by released males (Fig. 1),
causing genetic death, leading to population collapse [7, 8 ].
The systems constructed in SBF contain homozygous-viable pericentric
inversions on the same chromosomes as the eye colour mutations, opposite the sex
linked translocation which carries the wild type eye colour genes [8 ]. One strain also
contains a temperature sensitive lethal ( t s l) mutation.
The inversions stabilize the strain against genetic breakdown and contribute to
genetic death in GIFT systems because crossing over within heterozygotes generates
inviable products [8 ].
 IAEA-SM-327/30 301

FIELD RELEASED
FEMALE MALE
(NN) (Tl)

Partially sterile and

Normal carries eye mutations

50% sterility

I
MALE FEMALE RELEASED
OFFSPRING OFFSPRING MALE
(TN) (IN) (Tl)

Partially sterile Partially sterile

Partially sterile and and carries
carries eye mutations eye mutations

up to 90% sterility

MALE FEMALE
OFFSPRING OFFSPRING
(TI.TN) (И, IN)
Combined effect of
sterility and blindness
causes up to 98% death

All partially sterile Most (II) and some (IN)

and many carry females blind. Rest of (IN)
eye mutations partially sterile and
carry eye mutations

FIG. 1. Proposed GIFT genetic control system: transfer of genes and chromosomes into a
field population. N represents the normal autosome set; I represents the autosome set
containing homozygous-viable inversions and eye mutations on both chromosomes 3 and 5
(heterozygotes fertile in males, semi-sterile in females), passed to the daughters o f released
males; T represents the sex linked (Y;3;5) translocation (semi-sterile), passed to the sons.
302 FOSTER et al.

The SMT would use a simpler genetic sexing system, involving a translocation
between the Y chromosome and only one autosome, plus a single eye colour muta­
tion, a t s l mutation and a pericentric inversion.
Both SIT and SMT would involve the sterilization of males by irradiation
before release, whereas GIFT males cause genetic death without irradiation. It is
possible that with SIT or SMT, released males would suffer from radiation induced
loss of competitiveness, whereas GIFT males would not. This question has not been
investigated adequately in SBF.
SMT and GIFT should be cheaper to implement than SIT because females can
be eliminated using the t s l mutations at an early stage of mass rearing, thereby
reducing both capital outlays and operating costs.
In addition, stringent safeguards against the escape of fertile flies from a
rearing operation, as required with SIT, would be less necessary with GIFT or SMT.
The female flies in these cases would be debilitated and unable to survive in the wild.
Any males escaping the factory would be either identical to those being released
(GIFT) or at least generate partial sterility in their daughters (SMT).

3. SIMULATION OF GENETIC CONTROL

The GENCON simulation programs [6 - 8 ] were developed for a variety

of uses.

3.1. Comparison of genetic and sterile male alternatives

Simulations predict that at high release rates SIT should cause more rapid sup­
pression than GIFT [7]. On the other hand, at low release rates GIFT should give
more rapid suppression and earlier eradication than SIT (Fig. 2).
However, if releases cease before eradication, the predicted recovery of target
populations is more rapid following SIT than with GIFT [7].
The better performance of GIFT strains at low release rates or when releases
are interrupted is due to generation in the target population of males similar to the
mass reared GIFT males (Fig. 1).

3.2. Assessment of the potential of inversions for genetic control

One edition of GENCON [8 ] was designed to evaluate GIFT systems contain­

ing pericentric inversions before investing the resources necessary to make these
rearrangements [9]. The simulations predicted that inversions would give both faster
suppression initially and greater persistence of suppression if the releases were inter­
rupted than GIFT systems without inversions.
 IAEA-SM-327/30 303

Generations

Generations

FIG. 2. Simulations o f autocidal control using SIT or GIFT strains carrying a T(Y;3;5) trans­
location. Initial population (one million) flies o f each sex, with the rate o f increase in each
generation (eight per year) influenced by simulated seasonal conditions and density [8]. GIFT
strains: Т/St, St — standard (non-inversion) chromosomes, heterozygous for three eye colour
mutations; T/In,St — chromosome 3 inversion only, heterozygous for two eye mutations;
T/Ih,In — chromosome 3 and 5 inversions, heterozygous for two eye mutations. Release rates:
(a) 0.6 million males per generation; and (b) 0.3 million males per generation.
304 FOSTER et al.

Simulations to illustrate the different potentials of SIT (or SMT) and GIFT are
presented in Fig. 2. At a release rate of 0.6 million males per generation (Fig. 2(a)),
SIT results in eradication within.two years; GIFT strains with two, one or no
inversions take progressively longer. At half this release rate (Fig. 2(b)), SIT does
not give eradication but the GIFT strains do, with the rate of suppression increasing
with the number of inversions.

3.3. Analysis of field trial data

Field trials of GIFT carried out in the 1970s did not achieve suppression of
target populations. Initially, this was attributed to an inappropriate release method
[10], but simulations using an analytical edition of GENCON suggested that these
trials were also influenced by the immigration of wild flies, density dependent effects
[6 ] and lower survival of the sons of released males in the field, possibly due to the
dieldrin resistance mutation they carried [ 1 1 ].

4. GENETIC STUDIES AND STRAIN DEVELOPMENT

4.1. Male recombination

Genetic sexing strains deteriorate following genetic exchange in males and

selection favouring recombinant genotypes. High male recombination frequencies in
two Y linked translocation strains led to the theory that breakage of the Y chromo­
some caused increased male recombination [12]. However, studies in other species
did not appear to follow this rule [13].
A reinvestigation of this problem in SBF showed that the frequency of male
recombination was independent of the presence or type of translocations [14].
Variability in genetic backgrounds seems to be the most likely cause of the differ­
ences in the male recombination frequencies seen in some crosses. We concluded
that strain stability in our sexing systems could not be improved by the selection of
particular types of translocation.

4.2. Use of chromosomal inversions to stabilize strains

The best cure for male recombination appears to be that adopted by mosquito
geneticists, i.e. including inversions in genetic sexing systems [15].
Inversions can be used in several ways, depending on the specific use intended
for the sexing system [16]. We chose, homozygous-viable pericentric inversions,
since these have several advantages over other types of inversion and can be used
in both GIFT and SMT strains. One advantage is that the inversions and translocation
can be on separate chromosomes, and can readily be changed if necessary.
 IAEA-SM-327/30 305

The products of recombination within heterozygous, pericentric inversions are

usually inviable. Thus, including such an inversion in a release strain ensures the
selective elimination of the recombinant products in males.
In heterozygous females, recombination within a pericentric inversion can
cause up to 50% semi-sterility. In GIFT systems, all the daughters of released males
are heterozygous for the inversions and are therefore partially sterile, increasing the
level of genetic death available.
Including a pericentric inversion on both of the non-translocated autosomes in
a GIFT strain causes up to 75% semi-sterility in the daughters of released males,
giving maximum genetic death rates of approximately 98% when combined with two
eye colour mutations and the translocation (Fig. 1).
The fertility of mass reared strains is not affected by the inversions. In
heterozygous males, the frequency of recombination is too low to cause observable
sterility. In homozygous females, fertility is not affected because crossing over does
not generate inviable products.

4.3. Isolation of homozygous viable inversions

A search for homozygous-viable pericentric inversions on chromosome 3

yielded four, one of which is suitable for inclusion in GIFT systems [9].
Small scale field trials demonstrated that released males carrying this inversion
were as competitive at mating released sentinel females as released non-inversion
males [17].
A similar search on chromosome 5 yielded four large pericentric inversions
from 6139 screened chromosomes, of which one is homozygous viable [17].

4.4. Isolation of new genetic sexing translocations

All thé sexing systems tested' so far have descended from a translocation
constructed to carry the dieldrin resistance mutation R d l to permit early killing of
females in mass rearing [18].
. However, dieldrin treatment proved harmful to the released males [19] ; the R d l
gene also reduces fitness in the field [11]. Hence, a search was undertaken for suit­
able translocatións without this gene.
GIFT translocations involve the Y chromosome and two aütosomes, each
carrying the wild type allele of a selected eye colour mutation: chromosome 3 (white)
plus either chromosome 5 (topaz) or 6 (yellow). They must have a higher fertility
than the 25 % usually associated with three chromosome translocations (a fertility of
approximately 50% gives a reàsonable compromise between suppressive potential in
the field and fecundity in mass rearing) [7].
For SMT genetic sexing systems, selection of translocatións with a fertility
approaching 1 0 0 % should be the goal.
306 FOSTER et al.

More than two hundred translocations were screened over a three year period
[17]. First, two chromosome translocations were screened for near centromeric
break points. Experience suggests that these are more likely to be highly fertile than
other translocations. We assume that this is related to the positions of the break
points with respect to the euchromatic male meiotic pairing sites [2 0 , 2 1 ].
These were then screened for high fertility (70-100%). Two highly fertile
translocations have been retained for possible use in SMT systems. These and several
other translocations were also reirradiàted and screened genetically for inclusion of
a third chromosome in the rearrangement. These were then screened as above, first
cytologically and then for 40-60% fertility.
Four translocations which fit the selection criteria for GIFT were saved. One
has given encouraging results in preliminary field mating tests [17].

4.5. Isolation of tsl mutations

Temperature sensitive lethal mutations, which die at one temperature but sur­
vive at another, can be used to eliminate females from mass reared strains prior to
release.
The mutagen ethyl methanesulphonate (EMS) was used to treat flies, since this
chemical gives a higher proportion of t s l mutations than mutagens such as radiation
or alkylating agents [2 2 ].
Treated males homozygous for the chromosome 3 white eye mutation were
crossed to a balancer chromosome [9] in a scheme designed to make the treated
chromosomes homozygous. In the final generation of this scheme, eggs were
incubated for 24 h at 35 °C and then reared at 27 °C. If no lethal mutation had been
generated on the treated chromosome, 1/3 of the offspring would be white eyed and
2/3 wild type. The absence .of white eyed offspring indicated a lethal gene on
chromosome 3. Such families were then screened for survival when hatched
at 27°C. '
Of the 318 chromosomes tested, 62 lethals were found, two of which were t s l.
One t s l was too weak to be useful, and the remaining one was saved for incorporation
with a Y autosome translocation into a genetic sexing strain [17].
This strain gives 99% males when the eggs are incubated for 24 h at 35° С and
67 % males at 27°C, compared with 57% males in a n o n - t s l strain with this transloca1
tion. Male production averages 27% less at 35°C than at 27°G, which may indicate
a dominant effect of the t s l [17]. This is being investigated further.
We are mapping the t s l mutation genetically and will try to cross it on to an
inversion bearing chromosome. If this is not possible, it may be necessary to isolate
a new t s l, starting with an existing homozygous viable inversion. ■
 IAEA-SM-327/30 307

5. FIELD TRIALS

In the 1980s, two field trials demonstrated the suppression of SBF populations
using GIFT. In these trials, migration was reduced by better geographical barriers
than in previous trials, and suppréssion of target populations to low levels was
achieved [23, 24].
In a mainland trial (240 km2) conducted in 1984-1985, a genetic death rate
averaged 50% suppressed populations to below the spring emergence level for
the entire season, unlike nearby areas where the fly densities increased by factors
of 50-80 [23, 24].
In an island trial (40 km2) carried out in 1985-1986, genetic death rose to
90% after six months of releases [23, 24]. The population was suppressed to
undetectable levels after eight months and remained at very low levels for nearly
one year after releases ceased [23].
A larger island trial (1990 km2) was conducted in 1989-1991 in the Bass
Strait. A substantial reduction in the target population occurred in 1989-1990 [25],
but problems in the mass rearing programme [26] reduced the release numbers at
critical times, preventing useful suppression in 1990-1991.
The main causes of irregular mass rearing production included inadequate
facilities and equipment and lack of appropriate criteria for controlling the critical
processes. This has led to recognition of the need for collaboration by entomologists
and engineers to develop large scale rearing systems.

6 . ECONOMIC CONSIDERATIONS AND COMMERCIAL PROSPECTS

Economic analysis indicates that when the technical problems with mass rear­
ing are solved, the benefit:cost ratio of genetic control of SBF will be highly favour­
able [1]. Analysis of genetic control as a potential business enterprise suggests that
it could be attractive to private investors, who would be paid a fee for the SBF eradi­
cation service [27].
Considerable interest has been shown by government and private enterprise in
development of genetic methods for insect eradication commercially. One approach
under discussion would be to first eradicate SBF from Tasmania. This is currently
seen as an economically justifiable operation on its own, but it could also be used
to develop improved rearing technology suitable for Australia wide eradication.

6.1. Tactical and technical considerations

Broadly speaking, there are three potentially cost effective approaches to the
genetic control of SBF: (1) suppression without eradication as a specific goal;
(2 ) local eradication protected by a barrier zone and quarantine measures; and
(3) total eradication from Australia.
308 FOSTER et al.

The above strategies are not mutually exclusive. For example, total eradication
would involve a series of local eradications, with a moving barrier between the
eradicated and; infested areas. Moreover, if a suppression campaign achieved local
eradication, the goals could then be modified, either to maintain the eradicated status
with a barrier or to proceed to total eradication.
The initial approach adopted will influence decisions such as the choice
between SIT/SMT or GIFT, and whether genetic control is centrally organized or
decentralized.
GIFT was designed as a low cost suppression system, requiring fewer released
males than SIT or SMT. Because GIFT does not require radiation equipment, it
could be implemented either as a single large scale operation (e.g. 500-1000 million
males per week) or several small scale local operations.
The SIT/SMT options, on the other hand, would require irradiation equipment,
which could be prohibitively expensive if installed in a large number of smaller scale
operations. These options would necessitate centrally organized, large scale rearing
and release operations.

6.2. Economic considerations

A GIFT suppression campaign using ,dispersed small scale rearing facilities

could be implemented with relatively low capital costs for each plant, but lost oppor­
tunities for economies of scale would mean greater operating costs than with a central
high technology system.
Economic sensitivity assessments suggest that eradication in as short a time as
possible, while initially more capital intensive, would provide a more cost effective
result than an indefinite suppression programme [1]. In this scenario, SMT would
probably be the most profitable genetic control system to adopt.
However, under the current economic circumstances in Australia, the large
amount of money (e.g. A $20 million [27]) required to build suitable rearing facili­
ties before any benefit is obtained is seen as too adventurous by some politicians and
rural industry groups.
Therefore, it may be expedient to begin with small scale suppression
programmes using GIFT strains, possibly at the local government or farmer
co-operative level. This approach could transfer significant economic and environ­
mental benefits to the rural community, until such time as improved technology
and investor confidence permit the implementation of a large scale eradication
programme.

6¿3. Combining trapping with genetic control

It is possible that combining trapping with an autocidal technique could be

highly cost effective. Trapping is effective in suppressing SBF populations [28]. It
 IAEA-SM-327/30 309

Generations

FIG. 3. Simulations o f autocidal control with concurrent trapping removing 50% o f the flies
per generation. The strains and ecological parameters are as in Fig. 2, with a release rate
o f 0.3 million males per generation.

should be possible to remove half of the SBF population each generation [29], but
traditional trapping methods are too costly. However, recent progress in developing
a synthetic lure for SBF [30] may change this situation.
Simulations suggest that trapping would not eliminate a population by itself,
but would greatly accelerate the rate of eradication by autocidal methods (Fig. 3).
In these simulations, the strains and numbers released are identical to those used in
Fig. 2(b). Trapping + SIT predicts eradication within one year, compared with.no
eradication without trapping. Trapping + GIFT predicts eradication times of about
1.5 years, down from 3.5-5 years without trapping.

6.4. Multispecies mass rearing facilities

An idea currently receiving consideration is a mass rearing facility capable of

simultaneous rearing o f more than one pest species. If this concept is technically fea­
sible, it should be economically more cost effective than a single species rearing
plant whose utility would cease on achievement of eradication. The costs of and
benefits accruing from its construction could be spread over several sectors of the
Australian rural industry. Moreover, an operational facility with trained personnel
would be able to respond quickly if an eradicated pest were to reinvade Australia.
This concept has arisen partly out of concern for ensuring Australia’s
preparedness against possible incursions by the Old World screwworm fly,
310 FOSTER et al.

This species and the New World screwworm fly, C o c h l i o m y i a

C h r y s o m y a b e z z ia n a .
h o m in iv o r a x , have each been detected in Australia once within the last five years,
although neither incursion has resulted in establishment [31, 32].
Both the Mediterranean fruit fly, C e r a t i t i s c a p i t a t a (in Western Australia) [33]
and the Queensland fruit fly, B a c t r o c è r a t r y o n i (mainly in eastern Australia) [34],
have been or are currently the subject of SIT programmes aimed at local eradication.
An outbreak of B . t r y o n i in Western Australia was eradicated in 1990 [35].
Other potential candidates for autocidal control in Australia are the codling
moth, C y d i a p o m o n e l l a , and the heliothis species, H e l i c o v e r p a a r m í g e r a and
H . p u n c t i g e r a , both of which attract heavy insecticide usage on various crops.
A multispecies rearing facility would probably be used for SBF eradication in
the first instance, then for SIT against other major pests. At the same time it would
be available for rapid implementation of a SIT campaign if either screwworm species
became established in this country.

ACKNOWLEDGEMENTS

This research was supported financially by the Australian Wool Corporation,

Project CEC4. The authors would also like to thank D.G. Bedo for performing the
cytological analyses and R.J. Mahon and K.G. Wardhaugh for permission to cite
unpublished results.

REFERENCES

[1]. KINO, R .L ., McSHANE, R .N ., PAULEY, J.R., An Investigation o f the Economic

Feasibility o f the CSIRO’ s Autocidal Techniques for the Genetic Control o f the Sheep
Blowfly in Australia, Report to the Australian W ool Corporation by the Department o f
Primary Industry, Tasmania (1992).
[2] HUGHES, P.B., M cKENZIE, J.A., “ Insecticide resistance in the Australian
sheep blowfly, Lucilia cuprina: Speculation, science and strategies” , Combating
Resistance to Xenobiotics: Biological and Chemical Approaches (FORD, M .G .,
HOLLOM AN, D .W ., K H AM BAY, B .P.S., SAWICKI, R .M ., Eds), Ellis Horwood,
Chichester (1987) 162-177.
[3] WHITTEN, M .J., et al., “ Current status o f genetic control o f the Australian sheep
blowfly, Lucilia ciiprina (Wiedemann) (Diptera: Calliphoridae)” , Proc. X V Int. Con­
gress on Entomology, Washington, D C, 1976, pp. 129-139.
[4] W ARDHAUGH, K .G ., M ORTON, R ., The incidence o f flystrike in sheep in relation
to weather conditions, sheep husbandry and the abundance o f the Australian. sheep
blowfly, Lucilia cuprina (Wiedemann) (Diptera: Calliphoridae), Aust. J. Agrie. Res.
■ 41 (1990) 1155-1167. .
 IAEA-SM-327/30 311

[5] VO G T, W .G ., M ORTON, R ., Estimation o f population size and survival o f sheep

blowfly , Lucilia cuprina, in the field from serial recoveries o f marked flies affected by
. weather, dispersal and age-dependent trappability, Res. Pop. Ecol. 33 (1991) 141-163.
[6] FOSTER, G .G ., SMITH, P.H ., Genetic control o f Lucilia cuprina: Analysis o f field
trial data using simulation techniques, Theor. Appl. Genet. 82 (1991) 33-43.
[7] FOSTER, G .G ., VO G T, W .G ., WOODBURN, T .L ., SMITH, P .H ., Computer simu­
lation o f genetic control: Comparison o f sterile males and field-female killing systems,
Theor. Appl. Genet. 76 (1988) 870-879.
[8] FOSTER, G .G ., Simulation o f genetic control: Homozygous-viable pericentric inver­
sions in field-female killing systems, Theor. Appl. Genet. 82 (1991) 368-378.
[9] FOSTER, G .G ., WELLER, G .L ., BEDO, D .G ., Homozygous-viable pericentric
inversions for genetic control o f Lucilia cuprina, Theor. Appl. Genet. 82 (1991)
681-689.
[10] VO G T, W .G ., W OODBURN, T .L ., FOSTER, G .G ., Ecological analysis o f field trials
conducted to assess the potential o f sex-linked translocation strains for genetic control
o f the Australian sheep blowfly, Lucilia cuprina (Wiedemann), Aust. J. Biol. Sci. 38
(1985) 259-273.
[11] M cKENZIE, J.A ., Selection at the dieldrin resistance locus in overwintering popula­
tions o f Lucilia cuprina (Wiedemann), Aust. J. Zool. 38 (1990) 493-501.
[12] FOSTER, G .G ., M ADDERN, R .H ., MILLS, A .T ., Genetic instability in mass-rearing
colonies o f a sex-linked translocation strain o f Lucilia cuprina (Wiedemann) (Diptera:
Calliphoridae) during a field trial o f genetic control, Theor. Appl. Genet. 58 (1980)
169-175.
[13] ROSSLER, Y ., Recombination in males and females o f the Mediterranean fruit fly
(Diptera: Tephritidae) with and without chromosomal aberrations, Ann. Entomol. Soc.
Am. 75 (1982) 619-622. . ■
[14] FOSTER, G .G ., WELLER, G .L ., CLARK E; G .M ., Male crossing over and genetic
sexing systems in the Australian sheep blowfly, Lucilia cuprina, Heredity 67 (1991)
365-371.
[15] CURTIS, C .F ., A K IY A M A , J., DAVIDSON, G ., A genetic sexing system in
Anopheles gambiae species A , M osq. News 36 (1976) 492-498.
[16] FOSTER, G .G ., Chromosomal inversions and genetic control revisited: The use o f
inversions in sexing systems for higher Diptera, Theor. Appl. Genet. 81 (1991)
619-623.
[17] FOSTER, G .G ., WELLER, G .L ., JAMES. W .J., PASCHALIDIS, K .M .,
McKENZIE, L.J., unpublished data.
[18] W HITTEN, M .J., FOSTER, G .G ., Genetical methods o f pest control, Annu. Rev.
Entomol. 20 (1975) 461-476.
[19] SMITH, P .H ., K ON O V ALO V, C ., FOSTER, G .G ., KERR, R .W ., Assessment o f the
quality o f mass-reared Lucilia cuprina (Wiedemann) (Diptera: Calliphoridae) males
treated with dieldrin as larvae in a female-killing procedure, Bull. Entomol. Res. 71
(1981) 1-10.
[20] FOSTER, G .G ., M ADDERN, R .H ., Segregation and pairing o f compound fifth-
chromosomes in Lucilia cuprina males, Genet. Res. 46 (1985) 149-168.
312 FOSTER et al.

[21] BEDO, D .G ., Specific recognition and differential affinity o f meiotic X -Y pairing sites
in Lucilia cuprina (Diptera: Calliphoridae) males, Chromosoma 95 (1987) 126-135.
[22] SUZUKI, D .T ., et al., Temperature sensitive mutations in Drosophila melanogaster.
I. Relative frequencies among gamma-ray and chemically induced sex-linked recessive
lethals and semilethals, Proc. Natl. Acad. Sci. USA 57 (1967) 907-912.
[23] M AHON, R.J., unpublished data.
[24] FOSTER, G .G ., unpublished data.
[25] M AHON, R.J., STORRS, М ., W A R D H A U G H ,.K .G ., unpublished data.
[26] FOSTER, G .G ., FEEHAN, J.E., PASCHALIDIS, К Ж , unpublished data.
[27] STONE, J., Flies by the billion (and other pests): Mass rearing o f insects for control
programs, CMPS&F News, Summer 1991 (1991) 20-22.
[28] M ACKERRAS, M .J., FULLER, M .E ., AUSTIN, K .M ., LEFROY, E.H ., Sheep
blowfly investigations: The effect o f trapping on the incidence o f strike in sheep,
J. Council Sci. Indus. Res. Aust. 9 (1936) 153-162.
[29] VO G T, W .G ., personal communication.
[30] URECH, R ., ibid., “ Field performance o f synthetic lures for the Australian sheep
blowfly, Lucilia cuprina (Diptera: Calliphoridae)” , Pest Control and Sustainable
Agriculture, Proc. 5th Australian Conference on Applied Entomological Research
(1992) (in press).
[31] RAJAPAKSA, N ., SPRADBERY, J.P., Occurrence o f the old world screw-worm fly,
Chrysomya bezziana on livestock vessels and commercial aircraft, Aust. Vet. J. 66
(1989) 94-96.
[32] SEARSON, J., SANDERS, L ., DAVIS, G ., TW EDDLE, N ., THORNBER, P.,
Screw-worm fly miyasis in an overseas traveller — case report, Communie. Dis. Intelli­
gence (Commonwealth Department o f Health, Canberra) 16 (1992) 239.
[33] FISHER, K .T ., HILL, A .R ., SPROUL, A .N ., Eradication o f Ceratitis capitata
(Wiedemann). (Diptera: Tephritidae) in Carnarvon, Western Australia, I. Aust.
Entomol. Soc. 24 (1985) 207-208.
[34] HORTICULTURAL POLICY COUNCIL, The impact o f fruit flies on Australian
horticulture, HPC Industry Report No. 3 (1991).
[35] DEPARTMENT OF AGRICULTURE, WESTERN AUSTRALIA, Queensland Fruit
Fly Eradication Campaign (SPROUL, A .N ., BROUGHTON, S., M ON ZU , N ., Eds)
(1992). ;
 IAEA-SM-327/31

ERRADICACION DEL GUSANO BARRENADOR

DEL NUEVO MUNDO

J.M. IRASTORZA, C. BAJATTA,

J. ORTEGA, S.J. MARTINEZ
Comisión México-Americana para
la Erradicación del Gusano Barrenador del Ganado,
México DF, México

Abstract-Resumen

ERADICATION OF THE NEW W O RLD SCREW WORM.

Historically, the New World Screwworm has been one o f the most destructive and
costly pests, attacking both domestic and wild warm blooded animals, including humans. Its
original habitat was the tropical and subtropical areas o f the southern United States o f
America, M exico, Central America, two thirds o f South America and the islands o f the
Caribbean. In 1972, the Governments o f M exico and the United States o f America set up a
commission to eradicate the parasite from northern and western M exico, down to the Isthmus
o f Tehuantepec, and to establish a “ sterile fly barrier” in that zone. For the purpose o f eradi­
cation the commission used the combined action o f two basic systems — production and
dispersion o f sterile flies, and field operations. In 1986, in view o f the progress made, it
extended its screwworm eradication activities to the Yucatán Peninsula and the countries o f
Central America. Thus, more than 90% o f Guatemala’ s territory is now free from this pest.
Belize was declared free from it in June 1992. Eradication operations are in progress in
Honduras and El Salvador. In addition, it is planned to conclude agreements with Costa Rica,
Nicaragua and Panama. M exico had officially been declared screw worm-free in February
1991, but there was an outbreak which it is hoped will have been brought under control by
the end o f 1992. Mention should also be made o f the activities which have been carried out
to eradicate the screwworm in the Libyan Arab Jamahiriya. As a result o f the joint eradication
efforts in that region, the Libyan Arab Jamahiriya was declared free from this pest in June
1991. Hence the New World Screwworm eradication programme has basically been success­
ful in the United States o f America, M exico, Central America and North Africa.

ERRADICACION DEL GUSANO BARRENADOR DEL NUEVO M UNDO.

El gusano barrenador del Nuevo Mundo ha sido históricamente uña de las plagas más
destructivas y costosas que atacan a los animales de sangre caliente, tanto domésticos como
silvestres, e incluso al hombre. Originalmente se encontraba distribuido en las áreas tropicales
y subtropicales del sur de los Estados Unidos de América, M éxico, Centroamérica, dos
terceras partes de Sudamérica y las islas del Caribe. En 1972, los gobiernos de M éxico y de
los Estados Unidos de América crearon una comisión con el objeto de erradicar el parásito
del norte y oeste de M éxico, hasta el Istmo de Tehuantepec y de establecer una “ barrera de
moscas estériles’ ’ en esa zona. Para la erradicación, la comisión ha utilizado la acción conjunta
de dos sistemas fundamentales: la producción y dispersión de moscas estériles y las opera­
ciones de campo. Debido a los progresos conseguidos, en 1986 extendió sus actividades de

313
314 IRASTORZA et al.

erradicación del gusano barrenador a la Península de Yucatán y a los países centroamericanos.

Así, Guatemala está libre de la plaga en más del 90% de su territorio. Belice se declaró libre
de la plaga en junio de 1992. En Honduras y El Salvador ya sé están efectuando operaciones
de erradicación. También se proyecta establecer convenios con Costa Rica, Nicaragua y
Panamá. M éxico fue declarado oficialmente libre del gusano barrenador en febrero de 1991,
pero se produjo un brote que se espera haber controlado antes de concluir 1992. Cabe destacar
también las actividades que se realizaron para erradicar el brote del gusano barrenador en la
Jamahiriya Arabe Libia. Los esfuerzos conjuntos para la erradicación en esa región dieron
com o resultado que en junio de 1991 se declaró al país libre de la plaga. Así pués, el Programa
de erradicación del gusano barrenador del Nuevo Mundo ha logrado progresos fundamentales
en los Estados Unidos de América, M éxico, Centroamérica y norte de Africa.

1. INTRODUCCION

El gusano barrenador del Nuevo Mundo (GBNM) es producto del ciclo

biológico de la mosca cuyo nombre científico es C o c h i l i o m y i a h o m n i v ó r a x
(Coquerel). El ciclo de vida del parásito dura aproximadamente 21 días. La mosca
hembra ya fecundada pone sus huevecillos en las heridas abiertas de animales o seres
humanos, donde después eclosionan y las larvas de ellos emergidas penetran desga­
rrando el tejido, y se alimentan del líquido que se forma. A los siete días, las larvas
caen al suelo, donde se entierran para convertirse en pupas. Después de siete días
a dos meses, las moscas adultas emergen para completar el ciclo biológico.
Originalmente este parásito se encontraba distribuido en las áreas tropicales y
subtropicales del sur de los Estados Unidos de América, México, Centroamérica,
dos terceras partes de Sudamérica y las Islas del Caribe.
Para que el parásito pueda infestar, es indispensable la existencia de una
herida, siendo las más comunes el ombligo de los recién nacidos y las producidas
por castración, trasquila y mareaje. Los daños causados por el GBNM son enormes,
siendo los principales la baja de producción, retardo en el crecimiento, deterioro de
las pieles, susceptibilidad a infecciones secundarias e incluso la muerte de los
animales. Además de los animales domésticos, también se ven afectados los
silvestres de sangre caliente, así como el hombre.
La manera de eliminar el parásito fue descubierta en los años treinta, cuando
se observó que la mosca podía ser esterilizada sexualmente utilizando radiaciones,
ya que la hembra se aparea una sola vez en su vida y al hacerlo con un macho estéril
los huevecillos depositados por ella no nacen, haciendo posible la erradicación.
En 1961 quedó eliminado el parásito del, sureste de los Estados Unidos de
América. A pesar de los esfuerzos desarrollados, el GBNM continuaba causando
daños en las regiones de los Estados Unidos de América y de México, por las
constantes migraciones de insectos fértiles a lo ancho de la frontera. Ante esto, los
 IAEA-SM-327/31 315

ganaderos de ambos países solicitaron a sus respectivos gobiernos el apoyo para

eliminar la plaga, por lo cual el 28 de agosto de 1972 se formó la Comisión México-
Americana para la Erradicación del Gusano Barrenador del Ganado.

2. ERRADICACION DEL GBNM

Para la erradicación del parásito, la Comisión ha utilizado la acción conjunta

de los dos sistemas fundamentales siguientes.

2.1. Producción y dispersión de moscas estériles

En 1976 inició su producción la Planta de Moscas Estériles de Chiapa de

Corzo, Chiapas (México), que sustituyó a la anteriormente ubicada en Mission,
Texas (Estados Unidos de América).
El proceso implantado ha sido diseñado para asegurar una eficiente producción
y esterilización de insectos en la cantidad y calidad requeridas. Actualmente la Planta
produce 300 millones de insectos por semana, ya que así lo requiere técnicamente
el Programa.
Una vez llevada a cabo la esterilización sexual por radiación con 137Cs
durante un tiempo comprendido entre 1'49" y 1'52" a 7000 rads, se procede al
empaque y envío de los insectos a los Centros de distribución, para su dispersión en
las zonas infestadas por medio de aviones, s i g u i e n d o patrones previamente
establecidos.
Actualmente, debido a la magnitud del Programa, se tiene ubicado el Centro
de distribución en el Aeropuerto de Tuxtla Gutiérrez, Chiapas (México), así como
Centros de dispersión en Tenosique, Tabasco (México) y en Retalhuleu y Puerto
Barrios (Guatemala).

2.2. Operaciones de campo

Las operaciones de campo consisten en la inspección de los animales, a fin de

detectar y curar todo tipo de heridas en ellos producidas, así como en el reporte y
envío para su diagnóstico de las larvas colectadas, con el consiguiente tratamiento
de moscas estériles.
Las rutas de inspección se establecen basándose en la biología del parásito, ya
que deben realizarse en circuitos con una duración de hasta 21 días, con el propósito
de eliminar el riesgo de la.presentación de “ reciclajes” de la plaga.
Para proteger de reinfestaciones las zonas ya libres, es necesario mantener un
control sobre las movilizaciones de los animales; debido a lo anterior, el Programa
estableció en 1984 un sistema de estaciones de inspección y cuarentena, consistente
en tres estaciones permanentes localizadas en las rutas pecuarias del Istmo de
316 IRASTORZA et al.

Tehuantepec, y apoyadas por otros puntos de inspección temporales dentro de las

áreas aún infestadas.
Todos los animales que cruzaban por las estaciones permanentes en dirección
este-oeste eran sometidos a inspección minuciosa para detectar la presencia del
GBNM, y bañados con una solución larvicida. Cuando se detectaba algún animal
infestado o con heridas que podían infestarse, se ponía en cuarentena hasta lograr
su curación.
En los ocho años de actividad de las Estaciones cuarentenarias se inspecciona­
ron 2 688 829 animales, de los cuales se pusieron en cuarentena 1063, detectándose
175 parasitados con el GBNM. Cabe mencionar que en ese período se inspecciona­
ron 10 703 animales provenientes de otros países, detectándose 6 infestados, a los
que se les impidió la entrada en el país.
La implantación de las Estaciones cuarentenarias contribuyó definitivamente a1
evitar la presencia de brotes del GBNM en las zonas ya libres.
Para el mejor desarrollo de las operaciones antes señaladas, ha sido necesario
utilizar como apoyo los medios masivos de comunicación, y establecer lazos de
coordinación con los organismos oficiales y privados relacionados con la actividad
pecuaria.

3. PROGRESOS Y RESULTADOS

Debido a los progresos del Programa, en 1986 se modificó el convenio que dio
origen a la Comisión México-Americana para la Erradicación del Gusano
Barrenador del Ganado, con la finalidad de extender las actividades de erradicación
a la Península de Yucatán y a los países centroamericanos.
Como resultado de todas las acciones citadas, en julio de 1990 se presentó el
último caso positivo en México, por lo cual el 25 de febrero de 1991, México fue
declarado libre del GBNM.
La erradicación de la plaga en territorio mexicano se obtuvo después de
19 años de intenso trabajo y con una inversión de 600 millones de dólares de los
Estados Unidos, lográndose como beneficio un ahorro para la ganadería de México
del orden de 82,6 millones de dólares anuales.
Como consecuencia de la erradicación del parásito en México, durante todo
1991 se realizaron, actividades de sobrevigilancia epizootiológica, con el propósito
de seguir contando con el apoyo de los productores; así también, se estrechó la
coordinación con las diferentes dependencias del sector agropecuario.
De esta manera, el 31 de diciembre de 1991, la Comisión terminó las
actividades de campo y de cuarentena en México, quedando a cargo de la sobre-
vigilancia la Comisión México-Estados Unidos para la Prevención de la Fiebre
Aftosa y Otras Enfermedades Exóticas.
 IAEA-SM-327/31 317

Sin embargo,- después de 18 meses del último caso, el 22 de enero de 1992 se

presentó una muestra positiva en el Estado de Campeche, por lo cual se ha aplicado
un programa operativo de emergencia para su control.
Actualmente se han comunicado 41 casos positivos, localizados en los Estados
de Campeche, Chiapas, Tabasco, Veracruz y Tamaulipas; en este último Estado se
detectó el último caso, el 30 de septiembre, en el Municipio de Aldama.
El origen del brote viene causado posiblemente por la introducción ilegal de
animales desde países centroamericanos aún infestados.
Las acciones en el operativo se han repartido, siendo el SINESA (Sistema
Nacional de Emergencia en Salud Animal) el responsable de la inspección, detec­
ción, muestreo; diagnóstico y control de las movilizaciones de los animales, mientrás
que la Comisión México-Americana para la Erradicación del Gusano Barrenador del
Ganado se ocupa de la producción y dispersión de las moscas estériles requeridas,
del monitoreo y tampeo entomológico en las zonas afectadas y de los apioyos técnicos
colaterales.
Actualmente la Comisión está dispersando un promedio de 100 millones de
insectos por semana sobre las áreas reinfestadas de México.
Con las actividades señaladas se espera establecer un pronto control y erradica­
ción del brote en beneficio de la salud y producción pecuaria nacional.
En lo que respecta a las operaciones en los países centroamericanos, se han
logrado avances sustanciales. Así, Guatemala está libre de la plaga en más del 90%
de su territorio y se considera que para fines de 1992 lo estará en su totalidad.
En cuanto a Belice, no se han registrado casos positivos desde octubre de 1991,
por lo que también se declaró libre de la plaga el 21 de junio de 1992.
De la misma manera, en Honduras y El Salvador ya se están efectuando opera­
ciones de erradicación, con los correspondientes envíos de moscas estériles desde la
Planta de Tuxtla Gutiérrez, Chiapas (México).
Tanto en el Programa para Belice como en el de Honduras y El Salvador, se
utiliza la técnica de la “ mosca aletargada” , consistente en transportar y dispersar
a granel el insecto, a temperaturas bajas y con control de la humedad; esta técnica
elimina la caja y su empaque tradicionales, reduciendo los costos de manejo y disper­
sión consecuentemente.
El Programa tiene proyectado establecer convenios con el resto de los países
centroamericanos, tal como se ha hecho en 1992 con Nicaragua, y se hará con Costa
Rica y Panamá en 1993. En este último país se establecerá una barrera de moscas
estériles en el Tapón del Darien para evitar migraciones de insectos fértiles desde
Sudamérica.
También se tiene planeada la construcción de una nueva planta productora de
moscas estériles, la cual podría ubicarse en Costa Rica o en Panamá, de acuerdo con
las facilidades que se otorguen. Dicha planta sustituiría a la que actualmente se tiene
en Chiapas (México).
318 IRASTORZA et al.

Es importante destacar las actividades que se realizaron para erradicar el brote

del GBNM presentado en la Jamahiriya Arabe Libia, causado por la introducción de
animales infestados. Para la erradicación, la FAO solicitó a la Comisión los apoyos
técnicos correspondientes, consistentes en asesoría, capacitación, estudios
entomológicos y suministro de moscas estériles.
En consecuencia, entre diciembre de 1990 y octubre de 1991 se enviaron a la
Jamahiriya Arabe Libia un total de 1322 millones de moscas estériles, por vía aérea
directa de Tuxtla Gutiérrez a Trípoli. Se empezó por dispersar 3,5 millones por
semana, pasando después a 28 millones, para llegar en su última etapa a 40 millones
semanalmente. Los esfuerzos conjuntos dieron como resultado que el último caso en
este país se presentara en abril de 1991, por ló cual se le declaró libre de la plaga
el 22 de junio de ese año.
El Programa de erradicación del gusano barrenador del Nuevo Mundo ha
logrado progresos fundamentales en la lucha contra el parásito en los Estados Unidos
de América, México, Centroamérica y norte de Africa. Sin embargo, para alcanzar
la meta contemplada, es necesaria la continua participación de todas las instituciones
y personas involucradas, lo que deberá redundar en beneficio de la salud y de la
economía pecuarias.
 IAEA-SM-327/32

ERADICATION OF THE NEW WORLD SCREWWORM

FROM THE LIBYAN ARAB JAMAHIRIYA

D.A. LINDQUIST*, M. ABUSOWA**, W. KLASSEN*

* Insect and Pest Control Section,

Joint FAO/IAEA Division of Nuclear Techniques
in Food and Agriculture,
•International Atomic Energy Agency,
Vienna

** Libyan Veterinary Services,

Tripoli,
Libyan Arab Jamahiriya

Abstract
ERADICATION OF THE NEW WORLD SCREWWORM FROM THE LIBYAN ARAB
JAMAHIRIYA. •
The New World screwworm, Cochliomyia hominivorax (Coquerel), invaded North
Africa in the late 1980s. It became established in the Libyan Arab Jamahiriya, where it was
first detected in 1988 and confirmed in 1989. This devastating pest of livestock became the
target of a major eradication programme costing approximately US $80 million. The basis of
the eradication programme was the sterile insect technique. A total of 1400 million sterile
screwworm flies from Mexico were released over a 40 000 km2 area between December
1990 and October 1991. Total eradication was achieved. The eradication programme included
a very strong quarantine effort to prevent expansion of the infested area in the Libyan Arab
Jamahiriya. Animals in the infested area were inspected and wounds treated about every
three or four weeks. Trapping of adults determined the presence or absence of adult
screwworm flies. A major information programme supported the eradication effort.

1. INTRODUCTION

The New World screwworm (NWS), C o c h l i o m y i a h o m i n i v o r a x (Coquerel), is

a native animal parasite of the New World. It is the most serious livestock insect pest
in'the New World and has bèen the target of eradication programmes in Mexico, the
United States of America, some of,-the Caribbean islands and, currently, Central
America. [1, 2]. . ‘ ,
The NWS is an obligate parasite of warm blooded animals. The larval stage
attacks only living flesh. A gravid female NWS fly is attracted to the wounds of

319
320 LINDQUIST et al.

warm blooded animals, where she deposits eggs in clutches of about 20-200
(Fig. 1). The eggs are laid along the edge of the wounds, which can be as small as
a tick bite. The navel of newborn is a favourite site for deposition. Also, management
practices such as dehorning, castration and branding are excellent NWS oviposition
sites. In NWS endemic areas, management practices are to a large extent regulated
by NWS seasonal populations. Thus, births, castration, dehorning, branding, etc.,
are timed for when the NWS population is at its seasonal low period.
Individual livestock owners can effectively reduce or eliminate losses resulting
from NWS attack. However the cost is very high, particularly for labour. During
the season of the year when NWS populations are sufficiently large to result in losses
of livestock, each animal must be inspected individually at least two times per week
and all wounds, whether NWS infested or not, must be treated with the appropriate
insecticide to kill the worms in the infested wound and to prevent an infestation
(Figs 2 and 3).
The eggs that are laid on the wound hatch within 24 h and the larvae feed on
the living flesh and exudate from the damaged flesh. Feeding within the host con­
tinues for five to six days during which time the larvae enlarge the wound signifi­
cantly. The odours emanating from the wound áre attractive to gravid females, who
may lay additional eggs and thus greatly aggravate the situation. When mature, the
larvae leave the wound and burrow into the ground and pupate. During warm

FIG. 1. A wound which is susceptible to screwworm attack.

 IAEA-SM-327/32 321

FIG. 3. Spraying a flock o f sheep with an insecticide to control the screwworm.

322 LINDQUIST et al.

weather (20 ° C or higher), the pupal period is about 8 d. The adult flies which emerge
from the pupae will then mate within a few days and the females deposit eggs on
wounds. The total life-cycle requires about 21 d during good weather conditions.
The cost of controlling the NWS by individual livestock producers has been
estimated at US $5-10 per animal per year. As stated previously, the major part of
this cost is associated with labour; however, the cost of insecticides is also signifi­
cant. Although there are a number of insecticides available which kill screwworm
larvae, Coumaphos is widely used because of its effectiveness against larvae, some
residual activity and lack of repellent activity against the adult female fly. This latter
feature can be quite important, since if the adult fly is repelled by the insecticide she
will deposit her eggs on another wound which has not been treated with insecticide.
The NWS attacks wildlife and in south Texas the mortality rate of newborn
White tailed deer ranges from 25-80% each year prior to the eradication of the pest.
People living in endemic NWS areas have learned to live with the pest by protecting
wounds and by frequent inspection of wounds, primarily of children. The use of
insecticide is widespread for this purpose.

2. ERADICATION OF THE SCREWWORM FROM THE

LIBYAN ARAB JAMAHIRIYA

The NWS was discovered in the area around Tripoli, Libyan Arab Jamahiriya,
in early 1988 when unusually severe myiasis cases in sheep were reported. The pest
was identified in late 1988, confirmed by the Food and Agriculture Organization of
the United Nations (FAO) in early 1989 and an eradication programme was planned
and implemented [3-8].
It is not known exactly when the NWS was introduced into the Libyan Arab
Jænahiriya, but it probably occurred prior to 1988; it could have been 1987, or
perhaps it occurred as early as 1986. The confirmation that the NWS was established
in the Libyan Arab Jamahiriya resulted in termination of livestock imports (primarily
sheep) from the New World to that country. It also resulted in the termination of
most livestock movement from this country into neighbouring countries.
The NWS infestation in the Libyan Arab Jamahiriya undoubtedly resulted from
immature stages being transported in live animals, or viable pupae being transported
in association with live animals. Sheep are the most commonly imported animal and
thus the probable carrier of the disease. However, a NWS infected dog, zoo animal
or even a human could have been the source of the infestation.
Nearly all of the live animal imports into the Libyan Arab Jamahiriya are by
ship. Air transport has been rarely used, but would certainly be the most likely mode
of import because of the short time needed for transport and the relatively short time
of the egg and larvae stages in the host animal, about seven to eight days. The short
time that the parasite is in the egg or larvae stage and on and in the host animal
 IAEA-SM-327/32 323

precludes an infested animal being placed on a ship in the New World and the animal
being removed from the ship in the Libyan Arab Jamahiriya with the same infesta­
tion. The transit time by ship exceeds the seven to eight days that the immature stages
are in or on the host animal. Thus, there seem to be two possibilities: (1) the adults
from the parent generation mated and infested wounded animals on board ship and
it was the Fj generation which arrived in the Libyan Arab Jamahiriya; (2) pupae of
the parent generation were in a suitable pupation media on board ship and this media,
with the NWS pupae, was removed in that country. Since the infestation was origi­
nally centred around Tripoli, it is reasonable to assume that the infestation originated
from the Tripoli port area.
Mating and genetic studies were conducted in the USA to ensure that the
Mexican sterile NWS was sexually compatible with the Libyan NWS strain. These
data showed that the two strains would readily mate and produce viable offspring and
that the mitochondrial DNA of the two strains was similar [9]. Further, the mito­
chondrial DNA of the Libyan strain was quite similar to other strains available in
the laboratory which originated from some of the Caribbean islands and Central
American countries; Studies to determine the geographical origin of NWS strains
have not been conducted and thus it is not possible to pinpoint the country of origin
for the Libyan NWS infestation.
Prior to the establishment of the NWS in the Libyan Arab Jamahiriya, the
myiasis resulting from NWS attack was not an International Office of Epizootics
reportable disease. This has changed and NWS myiasis is now a category В
reportable disease.
The establishment of a NWS population in North Africa was a major concern
to national and international organizations because of the fear that it would spread
throughout North Africa, migrate down the Nile River to other parts of Africa and
spread throughout the Middle East as well as southern Europe. Because of this con­
cern, a major eradication campaign was determined to be the only feasible approach
to solving the problem. Fortunately, extensive knowledge was available from the
New World on NWS eradication. The NWS had been eradicated first from southeast
USA and then from the southwest. This was followed by eradication from Mexico
and some of the Caribbean islands. The current programme has the objective of
eradicating the pest from Central America and Panama and the establishment of a
barrier at the Darien Gap to prevent reinfestation. In addition, plans are being devel­
oped to eradicate the pest from those Caribbean islands which are still infested.
The cost of NWS eradication in the Libyan Arab Jamahiriya can be rather eas­
ily calculated. The cost of loss of exports from the New World to North Africa is
much more difficult to calculate and at present no data are available. Furthermore,
because of the wide publicity about NWS infestation in this country it is likely that
exports of live animals from NWS endemic countries in the New World have been
reduced because of fear of a repeat infestation in other parts of the world. Again,
no data are available.
324 LINDQUIST et al.

The decision to initiate a NW S eradication campaign in the Libyan Arab Jama­

hiriya was made jointly by F A O and the Government o f the Libyan Arab Jamahiriya.
Once this decision was made, intensive planning was undertaken in order to estimate
the costs o f the eradication campaign and to prepare the technical details
required [10]. The only eradication technology available was that used for the past
twenty-five or more years in the New W orld and involved the use o f the sterile insect
technique (SIT) in combination with surveillance and quarantine and supported by
an intensive information campaign. It was fortunate that there was extensive
knowledge and there were people available with experience in NW S eradication. By
utilizing these resources, the eradication campaign was implemented late in 1990,
completed in October o f 1991 and confirm ed in June o f 1992. The speed with which
the eradication was achieved and the fact that the spread o f the N W S did not occur
as was originally feared, at least partially because o f effective quarantine activities,
greatly reduced the cost from the original planning figures. The programme took
advantage o f the 1990-1991 co o l winter in that country and sterile flies were
dispersed prior to the time when the overwintering generation would emerge and
start the spring NW S population increase. Only six NW S cases were reported during
1991 as compared with over 12 000 during 1990 [3 -5 ].
During the period mid-Decem ber 1990 to mid-October 1991, 1300 million
sterile NW Ss were dispersed in the NW S infested area o f the Libyan Arab Jama­
hiriya, which was approximately 28 000 km 2. During the actual eradication cam­
paign, two dispersals o f sterile NW Ss were made each week over the entire infested
area. This resulted in a high ratio o f sterile to wild NW Ss. One release per week
is not effective for NW S eradication. The sterile flies were released in predetermined
grid patterns with weekly effective flight lanes 2 km apart [3, 6, 7].
The sterile NW Ss were reared in Tuxtla Gutierrez, M exico, sterilized in the
late pupal stage (Fig. 4 ), packaged in adult emergence/dispersal boxes (1600 sterile
pupae per box) and sent by air freight to the Libyan Arab Jamahiriya. During the
early part o f the programme, the boxes o f sterile NWSs were sent by temperature
controlled truck from Tuxtla Gutierrez to M exico City where they were placed on
a scheduled airline flight to Frankfurt, Germany, and then by charter aircraft to the
Libyan Arab Jamahiriya. Later in the programme, direct charter flights were flown
from Tuxtla Gutierrez to Tripoli (Figs 5 -7 ).
Appropriate temperature control space was provided in the country, as well as
staff to handle the incoming sterile NW Ss. The release o f the sterile NW Ss was by
Twin Otter aircraft flying at about 250 km/h at an elevation o f 500 m. Releases were
made during the early morning to avoid excessive heat during the other parts o f the
day (Fig. 8).
About one hundred teams, each consisting o f two individuals with a vehicle
and the necessary treatment and sampling supplies, conducted extensive surveillance
(Fig. 9). These teams were assigned specific areas and inspected all livestock within
that area every 2 1 -2 8 d. Larvae samples collected were sent to the Central Veteri-
 IAEA-SM-327/32 325

FIG. 5. Loading boxed and palletized sterile flies onto a charter aircraft at Tuxtla Gutierrez,
Mexico. The plane will fly directly to Tripoli, with a stop in Bermuda for fuel and a crew
change. The flight takes about 14 h.
326 LINDQUIST et al.

FIG. 6, Unloading boxed, palletized, sterile screwworm flies at Tripoli International Airport.

FIG. 7. Sterile Fly Distribution Centre, Tripoli International Airport, showing boxed sterile
screwworm flies which had just arrived from Mexico and were being placed in temperature
controlled trailers.
 IAEA-SM-327/32 327

FIG. 8. Loading boxed sterile screwwormflies onto dispersal aircraft at Tripoli International
Airport. Each box contains 1600 sterile flies which will be released in the predetermined grid
over the infested area.

FIG. 9. A surveillance team treating the wound o f a camel with insecticide to prevent
screwworm infestation.
328 LINDQUIST et al.

FIG. 11. Inspecting a sheep for screwworm infestation at a quarantine station.

 IAEA-SM-327/32 329

FIG. 12. After passing through the quarantine station, a farmer is given literature describing
the programme and the reasons for the quarantine stations.

nary Laboratory in Tripoli for identification. Approximately eighty traps were

utilized to confirm the dispersal o f sterile NW Ss throughout the infested area,
as well as to estimate the ratio o f sterile to wild NW Ss (Fig. 10).
Eleven quarantine stations were located at the periphery o f the infested area
to prevent the spread o f the disease into neighbouring uninfested areas (Figs 11
and 12). Fortunately, the infested area was semi-isolated by the Mediterranean Sea
in the north, the desert in the south and quite dry and harsh areas with little livestock
to the east and west. This semi-isolation and the effectiveness o f the quarantine
prevented a significant spread o f the disease.
Climatic conditions in a belt about 15-25 km wide along the coast were such
that though the NW S could easily overwinter, south o f this area the climatic condi­
tions did not permit NW S overwintering. Again this helped in preventing the spread
o f the disease.

3. CONCLUSIONS

The NW S has been successfully eradicated from the Libyan Arab Jamahiriya.
A large part o f the credit for the success o f this campaign should g o to the effective
support provided by the Government o f the Libyan Arab Jamahiriya, which included
cash paid on time, the assignment o f a technically competent and dedicated staff (a
total o f about four hundred), the very high priority that the Government placed on
330 LINDQUIST et al.

successful NWS eradication, which resulted in the many usual bureaucratic problems
being solved instantly and the fact that all concerned set aside politics so that the
programme could succeed.
The eradication programme which was successfully implemented in this coun­
try cost a total o f about US $80 000 000. A n econom ic analysis clearly demonstrated
the benefits o f this expenditure [11]. Approximately half o f this was paid for by the
Government o f the Libyan Arab Jamahiriya in cash and in kind, while the remainder
was from a number o f donors. The eradication programme was planned and
implemented by F A O and was supported by the IA E A , International Fund for
Agricultural Development and the United Nations Development Programme. In
addition, major support was received from the M exican-A m erican Commission for
the Eradication o f Screww orm , the organization which is responsible for screwworm
eradication in M exico and is the only source o f sterile NW Ss.

REFERENCES

[1] G R A H A M ,О.H ., E d ., Eradication of the Screwworm from the United States and M e x ­
ico (Proc. Symp. San Antonio, TX, 1985), Vol. 62, Entomological Society of America,
Lanham, M D (1985) 1-66.
[2] W O R L D A N I M A L R E V I E W , Special Issue (October 1991).
[3] LINDQUIST, D.A., A B U S O W A , М., H A L L , M.J.R., The N e w World Screwworm
fly in Libya: A review of its introduction and eradication, Med. Vet. Entomol. 6 (1992)
2- 8 .
[4] C U N N I N G H A M , E.P., A B U S O W A , М., LINDQUIST, D.A., SID A H M E D , A. E.,
V A R G A S - T E R Á N , М., The North African Screwworm Eradication Programme,
Invertebr. Repr. Dev. 22 1-3 (1993) 103-108.
[5] LINDQUIST, D.A., A B U S O W A , М., Eradicating the N e w World Screwworm from
the Libyan Arab Jamahiriya, IAEA, Bull. No. 4 (1992) 17-24.
[6] LIND Q U I S T, D.A., A B U S O W A , М., K L A S S E N , W., “Eradication of the N e w
World Screwworm in the Libyan Arab Jamahiriya”, I A E A Yearbook 1992, IAEA,
Vienna (1992) B7-B18.
[7] The N e w World Screwworm Eradication Programme, North Africa 1988-1992, FA O ,
R o m e (1992).
[8] V A N D E R V L O E D T , A.M.V., K L A S S E N , W., The development and application of
the sterile insect technique (SIT) for N e w World Screwworm eradication, World Anim.
Rev. (1991) 42-49.
[9] T A Y L O R , D.B., H A M M A C K , L., R O E H R D A N Z , R.L., Reproductive compatibility
and mitochondrial D N A restriction site analysis of N e w World Screwworm,
Cochliomyia hominivorax, from North Africa and Central America, Med. Vet.
Entomol. 5 (1991) 145-151.
[10] A Programme for the Eradication of the N e w World Screwworm from North Africa,
Joint F A O / I A E A Division of Nuclear Techniques in Food and Agriculture, IAEA,
Vienna (1990).
[11] G R I N D L E , J., Economic Impact of N W S Eradication from North Africa, Technical
Report SECNA/INT/001/MVL, F A O , R o m e (1991).
 IAEA-SM-327/70

EFFECTIVE CONTROL OF THE MEDITERRANEAN

FRUIT FLY BY GENETIC SEXING MALE ONLY
STERILE INSECT TECHNIQUE RELEASES
DURING 1989-1990*
Y . N IT Z A N **, Y . RÔSSLER
Citrus Agrotechnical Services,
The Israel Cohen Institute for B iological Control,
Rehovot,
Israel

A .P . E CO N O M O P O U L O S***
F A O /IA E A Entomology Unit,
A gen cy’ s Laboratory Seibersdorf,
International A tom ic Energy A gency,
Vienna

Abstract
EFFECTIVE C O N T R O L O F T H E M E D I T E R R A N E A N FRUIT F L Y B Y G E N E T I C S E X ­
I N G M A L E O N L Y S T E R I L E I N S E C T T E C H N I Q U E R E L E A S E S D U R I N G 1989-1990.
For two successive years, the Mediterranean fruit fly, Ceratitis capitata (Wiedemann),
was effectively controlled by a combination of bait spraying onto boards suspended on host
trees in May, followed by weekly releases of about one million g a m m a sterilized genetic sex-
ing males till the end of the year. The treated area, Kibbutz Gvulot in southern Israel, included
about 500 ha of citrus, mango and backyard fruit trees. In both years, the female/male ratio
declined to extremely low levels in November-December, suggesting that the wild population
had been driven to near extinction. This was apparent from fruit infestation as well, which
was kept to levels below 0.1% in both years as compared with similar or even higher levels
in the control area, which was treated by air bait spraying about weekly from September till
the end of the year.

Research carried out with the support of the I A E A under Research Contract
No. 5339/R1/TC.
" Present address: Department of Plant Protection and Inspection, Ministry of
Agriculture, Bet-Dagan, Israel.
Present address: Department of Biology, University of Crete and Institute of
Molecular Biology and Biotechnology, P.O. Box 1470, GR-71110 Heraklion, Crete, Greece.

331
332 NITZAN et al.

1. IN TROD U CTION

The Mediterranean fruit fly (m edfly), Ceratitis capitata, is the major fruit pest
in many countries around the world, including Israel. The only effective control
relies on the wide and indiscriminate use o f pesticides (such as poisoned bait sprays).
Sterile insect releases o f the medfly were tested in various countries, including
Israel [1], and relied always on the joint release o f sterilized males and females due
to the lack o f a sound method o f genetic sexing. The construction o f ‘ genetic sexing’
strains o f the medfly [2 -5 ] permitted the testing o f an all male release programme,
which should reduce the cost o f handling and release, eliminate oviposition punctures
by the released females [6] and enhance the efficacy o f the released males [7].
The all male release method was previously tested on the island o f Procida
(Italy) in a field cage [7] and in a limited field test [8]. It was also tested in a field
cage in Hawaii [9].
The present test lasted for two entire seasons. Its main objective was to evalu­
ate the feasibility o f the all male release method as a practical control technique in
the field and compare it with currently used malathion bait aerial applications.

2. M A TE RIALS A N D M ETHODS

2.1. Test site

The test was carried out in the northern N egev. This is a semi-arid region in
the south o f Israel. The groves were relatively isolated from neighbouring cultivated
groves by a savannah-like plain. Releases were made over a 500 ha area in the
groves o f the Kibbutz Gvulot on citrus, mango orchards and backyard trees (apricots,
fejoa, etc.). The control area, in the Kibbutz Zeelim , located approximately 6 km
from the test area, included similar sized groves o f citrus and mango (Fig. 1).
The conventional medfly control method was used in Zeelim (aerial sprays o f
malathion baits). These were applied, on the basis o f trimedlure (T M L ) trap surveys,
almost every ten days in 1989. In 1990, the malathion bait applications were
governed by LADD® trap1 surveys (see Section 2 .2 ). The citrus orchards in Zeelim
received 12 applications (from September to Decem ber) and the mango orchards 15
applications (from July to O ctober), i.e. an average interval o f eight days between
applications.
The sterile insect technique (SIT) test area in Gvulot received in both test sea­
sons tw o applications o f malathion bait sprays, at the end o f April and M ay, one and
two weeks before the beginning o f the releases to reduce the density o f the wild popu­
lation o f medflies.

1 L A D D , Inc., Burlington, Vermont, USA.

 IAEA-SM-327/70 333

FIG. 1. (a) Test area, Kibbutz Gvulot and (b) control area, Kibbutz Zeelim ( о : trimedlure
trap; *: LADD trap).

2 .2 . Genetic sexing strain and releases

The T :Y (w p )3 0 c strain (álso called the ‘ Robinson’ strain) was used. The strain
was originally constructed by À .S. Robinson and stabilized during its mass rearing
in the A gen cy ’ s Laboratory at Seibersdorf, near Vienna [1]. The strain had brown
(wild type) pupae males and white pupae females. The flies used in the test were
mass reared at Seibersdorf [10], where the pupae were photoelectrically sexed and
gamma sterilized (95 G y ) before being shipped to Israel. W eekly shipments o f
approximately two million male pupae, one to two days before adult eclosion, were
air freighted under anoxia (in plastic bags) in cooled , insulated cardboard containers
and delivered directly to the SIT test area within 24 h from irradiation-packing. A
small sample o f pupae was sent to the Rehovot Laboratory upon arrival for quality
control tests. In 1989, the pupae were distributed in regular 53 cm X 37 cm paper
bags, with 5000 pupae per bag. W e inserted crumpled newspaper sheets sprayed with
a 10% sugar solution and two pieces o f rubber sponge (10 cm X 8 cm x 1 cm)
334 NITZAN et al.

soaked with the same sugar solution into each bag. The emerging adult flies were
kept in the bags for three days at a temperature o f 2 5 -2 7 .5 °C for sexual maturation.
The flies were then released at approximately 400 sites, uniformly spread in the test
area, and the bags were torn and hanged on the trees. Thirty-five consecutive weekly
releases from May to Decem ber 1989 were carried out.
In 1990, the paper bags were replaced with 32 cm x 17 cm X 40 cm flat-
bottom paper bags and the total number o f pupae per bag was reduced to 4500. The
number o f release sites was increased to 450. The releases were delayed for five
weeks as compared with 1989 and only 30 weekly releases were carried out from
June to Decem ber 1990. Quality control tests were carried out, both at Seibersdorf
and Rehovot, for pupal size, adult emergence, adult flight ability and strain break­
down (the occurrence o f brown pupae females and white pupae males).
The survival o f the released flies in the groves was checked by the m ark-
release-recapture method after the first release in the citrus grove in 1990. Following
first release, all the flies were collected weekly from the L A D D traps and checked
for dye traces (as mentioned in Section 2 .3).
Aerial release (by helicopter) was carried out once in 1990, in the first release,
in order to compare it with the ground release. Fluorescent dyes were used to
differentiate between the flies from the two release methods, which were carried out
simultaneously. ‘ A rc Y e llo w ’ dye (for aerial release) and ‘ Neon R ed’ dye (for
ground release) (D ay-G lo C olour, Cleveland, Ohio, USA) were used, with 2 g o f
dye per 1 L o f pupae, prior to the release. Equal numbers o f flies were released by
each method at 75 sites, uniformly spread in an isolated group o f groves (4500 flies
per site). As the test was preliminary in nature, we may only conclude that flies
released from the air reached the traps and were not less abundant than ground
released flies. It seemed that air release did not harm the flies.

2.3. Monitoring the medfly population

Adult male and female flies were counted in L A D D traps and the T M L conven­
tional traps, throughout the test period. The L A D D traps consist o f red plastic
spheres located in the middle o f a yellow rectangular plastic sheet and covered with
Tanglefoot. The trap attracts both sexes, whereas the T M L trap attracts males only.
In 1989, a total o f 21 L A D D traps and 10 T M L traps were placed in the SIT test
area (Gvulot) and 13 L A D D traps and 5 T M L traps were placed in the control area
(Zeelim ). Four additional T M L traps were placed outside the test area, at varying
distances, between Gvulot and the nearest cultivated areas. In 1990, two traps were
also placed in the area between the backyards and the citrus groves to intercept
migrating medflies.
In the 1989 test season, the T M L traps captured large numbers o f released
males and interfered with the SIT test. It was therefore decided not to use these traps
in 1990 and to rely only on the L A D D traps, which had been very effective in attract­
 IAEA-SM-327/70 335

ing both sexes o f the m edfly. Thus, in the 1990 test period, 19 L A D D traps were
placed in the SIT test area (Gvulot) and 10 L A D D traps were placed in the control
area (Zeelim ). A ll the traps (in 1989 and 1990) were inspected weekly with a binocu­
lar magnifier (2 .5 X ) (Tem om Hessen, Germany). The flies were sexed, counted and
removed from the trap.
T w o methods were used to distinguish between natural and released flies:

(1) The released flies were marked with fluorescent “ Neon R ed” dye, with 2 g
o f dye per 1 L o f pupae, prior to release. In 1989, this was done four times
during the release period: in May a single release; in July, Septem ber-October
and in Decem ber four successive releases were marked. In 1990, such marking
was done three times during the release period: in June a single release; in
Septem ber-October and in N ovem ber-D ecem ber four successive releases
were marked. The objective o f the successive markings was to eliminate the
possibility o f capturing released unmarked flies when the flies were checked
for colour marking. Flies were therefore recaptured for examination only four
days after the last release o f marked flies. The recaptured flies were sexed and
inspected under ultraviolet light. No m ore than 50 specimens o f each sex were
examined from each trap.
(2) It was assumed that released (irradiated) females can be distinguished from
normal and sexually mature females by their undeveloped ovaries. (The
method will not distinguish irradiated from one-day-old normal females.) The
ovaries o f all the live and dead females (where possible) caught by the L A D D
traps were dissected. Such dissections were carried out routinely every week
during the last three and a half months o f the test in 1989 and throughout the
test period in 1990.

2.4. Fruit infestation

In both release seasons we searched intentionally for fruits which seemed

infested by the m edfly. The fruits were either picked from the trees or collected from
under the trees. They were inspected in the laboratory under magnification. When
oviposition punctures were observed, the fruits were maintained for the emergence
o f larvae and pupation.
Inspections o f harvested fruits were carried out in 1989, during five picking
days in Novem ber 1989, Decem ber 1989 and the beginning o f January 1990. During
the 1990 test season, inspections were carried out in Decem ber 1990 and January
1991. W e checked both picked fruits from the picking bins and collected dropped
fruit from under the trees on the same dates.
Picking bins. W e screened the upper layer o f fruit in 1 0 -20 bins (each had a
surface area o f 1 m 2 and contained approximately 100-220 fruits). Fruits with sus­
pected ovipunctures were taken to the laboratory to verify the nature o f the ‘ ovipunc-
336 NITZAN et al.

ture’ . These fruits were then kept for possible emergence o f médfly larvae. The
‘ sterile stings’ , where no larvae emerged, and the fruit with live medfly infestation
were counted separately. Ten thousand individual fruits were checked in the test area
as well as in the control in both test seasons.
Dropped fruits. A ll the dropped fruits were collected from under ten randomly
selected trees on each picking day. Sterile stings and the emergence o f live larvae
were recorded. A yield o f approximately 500 fruits per tree was assumed and our
calculation o f the percentage o f infested fruits was based on that figure.

3. RESULTS A N D DISCUSSION

3.1. Quality of the irradiated and released flies

The quality o f the released flies was checked thoroughly in the mass rearing
facility at the A gen cy’ s Laboratory at Seibersdorf. For technical reasons, only part
o f these data was examined for 1989, while for 1990 all data were examined. Certain
aspects o f quality such as adult eclosion, the percentage o f white pupae and the per­
centage o f females among the shipped flies, were also checked in Rehovot (follow ing
the air journey o f the flies from Vienna to Israel).
Adult eclosion in the shipments arriving at Rehovot was a major concern dur­
ing both the 1989 and 1990 periods. It seemed to fluctuate greatly between shipments
(Figs 2 and 3). It dropped often to lows o f 30% and less. The data from Seibersdorf
(Figs 2 and 3) suggested that the air journey and handling contributed most to pupal
mortality and the decrease in adult eclosion (Fig. 4). In nine shipments (two in 1989
and seven in 1990), the differences between the Rehovot and Seibersdorf data were
between 50% and 70% ! Our conclusion is that in spite o f detailed instructions given
to airport and shipment company personnel, on certain dates the pupal shipments
were not handled correctly, mainly because o f ignorance as a result o f airport person­
nel changes.
The sorting o f the genetic sexing line was carried out at Seibersdorf using an
electro-optical sorting machine. Such sorting left between 2 and 5% o f white pupae
among the wild type batch, with occasional exceptions o f 10% white pupae early in
1989 (Fig. 4 ), and fluctuated between 1 and 7% in 1990 (Fig. 5). The inclusion o f
such a proportion o f white pupae (assuming that these are females) meant that at least
100-250 females were released at each release site, or a total o f 20 0 0 0 -1 0 0 000
females were released in the whole test area each week. These were all sterilized
females, o f course, but their interference with the activity o f the released males, as
well as their interference with our checking method, was obvious.
The occurrence o f females among the wild type pupae was a measure o f the
stability o f the genetic sexing line T :Y (w p )3 0 c. In 1989, there seemed to be a rather
accelerated decline and breakdown o f the line (Fig: 4) and the proportion o f females
 IAEA-SM-327/70 337

0 4 -0 5 0 8 -0 6 0 6 -0 7 0 3 -0 8 3 0 -0 8 2 8 -0 9 2 6 -1 0 2 3 -1 1 21 -1 2
2 5 -0 5 21 - 0 6 2 0 -0 7 1 6 -0 8 1 4 -0 9 1 2 -1 0 0 9 -1 1 0 7 -1 2
Date
FIG. 2. Effect o f airfreighting on adult eclosion o f the medfly, 1989 ( ■: Rehovot laboratory;
a ; Agency ’
.v Laboratory Seibersdorf).

Date

FIG. 3. Effect o f airfreighting on adult eclosion o f the medfly, 1990 ( я : Rehovot laboratory;
л. : Agency’s Laboratory Seibersdo>f).
338 NITZAN et al.

0 4 -0 5 0 8 -0 6 0 6 -0 7 0 3 -0 8 3 0 -0 8 2 8 -0 9 2 6 - 1 0 .2 3 -1 1 21 - 1 2
2 5 -0 5 21 - 0 6 2 0 -0 7 1 6 -0 8 1 4 -0 9 1 2 -1 0 0 9 -1 1 0 7 -1 2
Date

FIG. 4. Percentage o f white pupae andfemales in released medflies, 1989 ( я : white pupae;
a. : females).

FIG. 5. Percentage o f white pupae andfemales in released medflies, 1990 ( ■: white pupae;
à. : females).
 IAEA-SM-327/70 339

among the wild type flies reached 7% and higher early in the test period. The line
used in 1990 was much more stable and the proportion o f females fluctuated between
2 and 7 % throughout the season, with an exception o f 8 % at the beginning o f the
test (Fig. 5). It should be noted that 1 % o f the flies meant that at least 10 000 addi­
tional (though sterilized) females were released with the males every week.
The survival o f released flies was also monitored and it was found that these
flies survived for no m ore than three weeks in the orchard. The number o f marked
flies in the trap dropped from 297 to 21 to 2 in the three consecutive weeks follow ing
the release.

3.2. Trends of the medfly population

The trend o f the female population in the control (Zeelim ) and test (Gvulot)
areas is presented in Fig. 6 for 1989 and in Fig. 7 for 1990. W e transformed the
L A D D data to display the number o f females per trap per day. The transformation
was direct for Zeelim , where all the females were native. In Gvulot, we had to distin­
guish between the released and native females. The ratio between released and native
females was calculated from the data obtained by the mark-release-recapture
method, as well as the routine dissections o f ovaries. The data were transformed into
the total number o f females and the number o f native females per trap per day.
In 1989, there was a sharp increase in the number o f female catches in the con­
trol orchard in June and July, follow ed by a sharp decrease (ow ing to conventional
control operations) towards the end o f the test season (Fig. 6). In the SIT orchards
in Gvulot, a similar but much smaller increase was observed. This increase con­
tinued to the middle o f August and declined later, but did not reach the ‘ zero’ level
as in the Zeelim control area. The occurrence o f females in the SIT orchard (Gvulot)
could not be attributed only to the presence o f females in the released flies, but also
to the continuous presence o f native females in the orchard (Fig. 6).
In 1990, the female population in the control orchard seemed to be much better
controlled and showed a steady and continuous decline throughout the season until
it disappeared completely in Decem ber. The SIT orchard also harboured a relatively
small population o f females, with some increase towards the end o f October and
Novem ber (Fig. 7). The female population nevertheless consisted in these instances
largely o f released females.
The decline in the female population due to the release o f sterile males can also
be viewed through the change in the sex ratio (females/males) during the release
period (Table I). The number o f males per trap increased during the season (due to
the presence o f a large population o f released males). At the same time the proportion
o f females among the captured flies declined to a level which was close to the ‘ impu­
rity’ o f the released line, or even low er (Figs 4 and 5).
34 0 NITZAN et al.

Date

FIG. 6. Sterile insect technique test — Kibbutz Gvulot, 1989 (trend o f the female population)
( a : Kibbutz Zeelim; +; total, Gvulot; О : native, Gvulot).

' 1.2

1.1
1.0
0.9
■§ 0:8
ш
£ 0.7
ra
~ 0.6
’ a>
«..0.5
Ш
I 0.4
Ф
^ 0.3
0.2

0.1

0
14 Jun. 12 Jul. 09 Aug. 06 Sep. 03 Oct. 01 Nov. 29 Nov. 27 Dec.
Date

FIG. 7. Sterile insect technique test — Kibbutz Gvulot, 1990 (trend o f the female population)
( a : Kibbutz Zeelim; + : total, Gvulot; О• native, Gvulot).
 IAEA-SM-327/70 341

TAB LE I. A V E R A G E NU M BER (PER M O N T H ) OF M ALES PER

L A D D T R A P A N D THE SEX R A T IO IN CITRUS O R C H AR D S IN
KIBBU TZ G V U L O T

No. of Sex ratio,

Month
males/trap females/males

September 1989 12.0 0.18

October 1989 15.9 0.16
November 1989 56.2 0.03
December 1989 61.0 0.02

June 1990 0.20

OO
OO

July 1990 15.4 0.07

August 1990 36.8 0.03
September 1990 46.8 0.03
October 1990 .33.3 0.05
November 1990 33.0 0.03
December 1990 77.9 ' 0.01

3.3. Fruit infestation in 1989

Four thousand grapefruits were checked, in the SIT test area in Gvulot, in pick­
ing bins on 19 and 30 Novem ber 1989. O f these, 16 were found with sterile stings
and only 5 fruits with live medfly maggots; On the same dates, and; also ón
19 Decem ber 1989, 176 grapefruits were collected from under 30 trees and
inspected for medfly damage. T w o fruits were found with sterile stings and 33 with
live medfly maggots. (O f the 33 fruits, 29 were collected from a single tree on
19 Decem ber 1989.) Six thousand oranges o f the variety ‘ Shamouti’ were inspected
on 28 Decem ber 1989, 31 Decem ber 1989 and 1 January 1990. O f these, 92 fruits
were found with sterile stings and only 2 fruits with live medfly maggots. On the
same dates, 222 oranges were collected from under 30 trees and inspected for medfly
damage. O f these, four were found with sterile stings and six fruits with live medfly
maggots.
The control orchard in Zeelim included only grapefruit and M ineóla orange
hybrids. Eight thousand grapefruits were inspected on 13 Novem ber, 20 Novem ber
and 3 Decem ber 1989 and also on 2 January 1990. Ten fruits were found with sterile
stings and four with live medfly maggots. A total o f 329 fruits were collected from
342 NITZAN et al.

TABLE II. FRUIT INFESTATION IN THE SIT AND CONTROL ORCHARDS

Harvested fruit Tree infestation'

Per cent sterile Per cent live Per cent sterile Per cent live
Season
stmgs maggots stings maggots

1989 SIT area 1.1 0.07 0.06 0.130

Control 0.1 0.04 0.00 0.004

1990 SIT area 0.05 0.40 0.10 0.076

Control 0.21 0.59 0.15 0.210
a Based on an average number of 500 fruits per, tree.

under 40 trees and only 1 was found with live medfly maggots. The 2000 M ineóla
fruits which were inspected on 12 Decem ber 1989 and the 91 fruits collected from
under 10 trees on the same date showed no sign o f medfly damage.
The infestation rates in 1989 are presented in Table II. The harvested fruit
(grapefruit and oranges) at the SIT site (Gvulot) included 1.1% fruits with sterile
stings and 0 .07% fruits with live maggots. In the control orchard in Zeelim , only
0 .1 % o f the harvested fruit had sterile stings and 0.04% had live medfly maggots.
The infestation found in the dropped fruit was transformed to an estimated infestation
per tree. The average yield per tree consisted o f 500 fruits and the infested fruits
found on the ground under the trees was calculated to be part o f these 500 fruits.
Thus, the SIT site had an infestation rate o f 0.0 6 % o f sterile stings per tree and
0.13 % o f fruits with live medfly maggots per tree. I f the 29 fruits found under a sin­
gle tree are excluded, infestation o f 0.024 and 0 .0 4 % , respectively, are reached.
Similarly, the control orchard in Zeelim showed 0 .1 % sterile stings and 0.0 4 % live
medfly maggots in the harvested fruits, and no sterile stings and 0.004% live medfly
maggots in the fruit collected from under the trees. The conclusion was that the SIT
method was effective and produced very g ood medfly control.

3.4. Fruit infestation in 1990

M onitoring for medfly infestation in citrus fruit throughout the test, when
fruits that seemed infected were intentionally searched, showed a low rate o f infesta­
tion. The infestation started in October (5 .0 % ) and declined in Novem ber (3 .7 % )
and Decem ber (1 .5 % ).
Ten thousand grapefruits were checked in the SIT test area in Gvulot in picking
bins on 20 Decem ber 1989 and on 9 -1 2 January 1991. O f these, five were found
with sterile stings and 40 fruits with live medfly maggots. On the same dates, 353
 IAEA-SM-327/70 343

grapefruits were collected from under 50 trees and inspected for medfly damage.
Twenty-five fruits were found with sterile stings and 19 with live medfly maggots.
A total o f 10 000 grapefruits were inspected in the control orchard in Zeelim
on 12, 26 and 31 Decem ber 1990 and on 6 and 11 January 1991. Ten fruits were
found with sterile stings and four with live medfly maggots. A total o f 310 fruits
were collected from under 50 trees. Thirty-seven fruits were found with sterile stings
and 52 fruits with live medfly maggots.
The total infestation rate in 1990 is presented in Table П. The harvested fruit
(grapefruit and oranges) at the SIT site (Gvulot) consisted o f 0.05 % fruits with sterile
stings and 0 .4 % fruits with live maggots. In the control orchard in Zeelim , 0.21%
o f the harvested fruit had sterile stings and 0 .5 9 % live medfly maggots. The infesta­
tion found in the dropped fruit was transformed tó an estimated infestation per tree.
The average yield per tree consists o f 500 fruits and the infested fruits found on the
ground under the trees were calculated to be part o f these 500 fruits. Thus, the SIT
site had an infestation rate o f 0.1 % sterile stings per tree and 0.076% o f fruits with
live medfly maggots per tree. Similarly, the control area in Zeelim had an infestation
rateofO .15% sterile stings andO .21% live medfly maggots per tree. The SIT method
in 1990 was as effective, if not more so, as in 1989.

REFERENCES

[1] B U S C H - P E T E R S E N , E., K A F U , A., Stability of two mass-reared genetic sexing

strains of Ceratitis capitata (Diptera: Tephritidae) based on pupal color dimorphism,
Environ. Entomol. 18 (1989) 315.
[2] R O S S L E R , Y., Automated sexing of Ceratitis capitata: The development of strains
with inherited sex-limited pupal colour dimorphism, Entomophaga 24 (1979) 411.
[3] R O S S L E R , Y., “69-apricot”— A synthetic strain of the Mediterranean fruit fly, Cera­
titis capitata (Diptera: Tephritidae), with sex-limited pupal colour and eye colour mar­
ker, Entomophaga 25 (1980) 275.
[4] R O B I N S O N , A.S., V A N H E E M E R T , C., Ceratitis capitata: A suitable case for
genetic sexing, Genetica 58 (1982) 229.
[5] R O B I N S O N , A.S., Unexpected segregation ratios from male-linked translocations in
the Mediterranean fruit fly, Ceratitis capitata (Diptera: Tephritidae), Genetica 62
(1984) 209.
[6] M E L L A D O , L., “La técnica de machos esteriles en el control de la mosca del Mediter-
rano: Programas realizados en España”, Sterility Principle for Insect Control or Eradi­
cation (Proc. Symp. Athens, 1970), IAEA, Vienna (1971) 49-54.
[7] R O B I N S O N , A.S., CIRIO, U., H O O P E R , G.H.S., C A P P A R E L L A , М., Field cage
studies with a genetic sexing strain in the Mediterranean fruit fly, Ceratitis capitata,
Entomol. Exp. Appl. 41 (1986) 931.
344 NITZAN et al.

[8] CIRIO, U., C A P P A R E L L A , М., E C O N O M O P O U L O S , А.Р., “Control of medfly

(Ceratitis capitata Wied.) by releasing a mass-reared genetic sexing strain”, Fruit Flies
(Proc. 2nd Int. Symp. Colymbari, Crete, 1986) ( E C O N O M O P O U L O S , A.P., Ed.),
Elsevier, Amsterdam (1987) 515-522.
[9] McINNIS, D.O., W O N G , T.T.Y., T A M , S.Y.T., Mediterranean fruit fly (Diptera:
Tephritidae): Suppression efficiencies of unisexual and bisexual sterilized release popu­
lation in field cages, Ann. Entomol. Soc. A m . 79 (1986) 931.
[10] H O O P E R , G.H.S., Application of quality control procedures to large-scale rearing of
the Mediterranean fruit fly. Entomol. Exp. Appl. 44 (1987) 161.
 IAEA-SM-327/71

TSETSE STERILE INSECT TECHNIQUE

PROGRAMMES IN AFRICA
Review and analysis o f future prospects

E D . OFFORI
Forint Enterprise,
Achimota,
Ghana

Abstract
T S E T S E S T E R I L E I N S E C T T E C H N I Q U E P R O G R A M M E S IN AFRI CA : R E V I E W A N D
AN A L Y S I S O F F U T U R E PROSPECTS.
Following the successful eradication of the screwworm, Cochliomyia hominivorax
(Coql.), from the island of Curaçao and the southeastern United States of America, the sterile
insect technique (SIT) was tested against several other noxious insects, including the tsetse fly.
Between 1967 and 1987, experiments were conducted in Zimbabwe and the United Republic
of Tanzania (East Africa) and in Burkina Faso and Nigeria (West Africa) to test the feasibility
of the new technique in eradicating selected species of tsetse fly. For the Zimbabwe
programme, sterile flies were obtained from field collected pupae treated with the
chemosterilant tepaR. Complete eradication was not achieved, primarily because of insuffi­
cient sterile males emerging from the wild pupae. In later programmes in the United Republic
of Tanzania, Burkina Faso and Nigeria, flies were obtained from laboratory bred mass reared
colonies. Males were sterilized either as pupae (Tanzanian project) or as young adults using
g a m m a irradiation from a ^ C o or 137Cs source. Efforts currently in progress to apply the
technique to eradicate Glossina austeni from Zanzibar Island, United Republic of Tanzania,
have attracted considerable interest and funding from several international organizations. Past
and current tsetse SIT programmes are reviewed and future prospects of the technique in large
scale tsetse/trypanosomiasis programmes are discussed.

1. IN TR O D U C TIO N

A little over three decades ago, the screwworm , Cochliomyia hominivorax

(C oql.) was eradicated from the island o f Curaçao and the southeastern United States
o f Am erica [1], using a novel approach to pest management. Since then, the sterile
insect technique (SIT) has been shown to be effective against many tropical fruit
flies, several destructive moths, the bollw eevil, the stable fly, certain mosquitoes and
the tsetse fly, Glossina species [1].
The method o f choice for controlling tsetse flies has been the use o f insecti­
cides, applied over large areas o f tsetse infested country, using ground or aerial
spraying techniques. In northern Nigeria, for example, approximately 250 000 km2

345
346 OFFORI

o f savannah land, equivalent to about 30% o f the tsetse infested area, were cleared
o f tsetse by ground and limited aerial spraying o f insecticides [2]..
The use o f SIT in tsetse control is a recent introduction. Over the past 25 years,
the technique has been introduced into several tsetse infested African countries
through the training o f personnel already involved in national tsetse control activi­
ties, through scientists participating in IA EA Co-ordinated Research Programmes
and, more practically, through Joint F A O /IA E A Division o f Nuclear Techniques in
Food and Agriculture field operational technical co-operation projects, or through
bilateral or similar projects funded with donor contributions. Four examples are
described briefly to illustrate the approach adopted in each case, the progress
achieved and the lessons learned. It is important to stress that all the tsetse SIT pro­
grammes undertaken to date have been, essentially, research and development
projects aimed, primarily, at testing the feasibility o f the new concept in.a practical
field situation.

2. THE E A R L Y Y EARS

The story begins in Zim babwe in 1964, on an island in Lake Kariba. At that
time, the sterility principle for pest control had been tested, with positive results on
a number o f insects; much curiosity had been aroused with regard to its applicability
to the tsetse fly because o f the fly ’ s low reproductive capacity and its relatively low
population density in nature. Because very little information was available on tsetse
rearing and radiation sterilization, and the behaviour o f sterile flies released into the
natural environment, a research programme was called for [3].
Sponsored by the United States Agency for International Development
(U SA ID ), the project was conducted jointly by the Agricultural Research Service o f
the United States Department o f Agriculture, the then Agricultural Research Council
o f Central Africa and the then University o f Rhodesia. The target species was
Glossina morsitans morsitans, the most important transmitter o f trypanosomiasis in
eastern Africa. The species was most abundant in the selected field station in the
valley o f the Zambezi River near Lake Kariba: The objectives o f the project were
to develop methods o f rearing and sterilizing the target species, and to evaluate SIT
against the species in the field.
In the initial stages, attempts were made to rear the flies in large screen
cages covering as much as 10 000 ft2 o f natural habitat.1 Flies were allowed to
feed at will on cattle and bushpigs. Smaller cages o f 8 in X 8 in X 11 in
(22 cm x 22 cm x 28 cm ) were also used. Both types o f cage were stocked with
pupae and adults collected from the field in the project area. The large field cage

1 1 f t 2 = 9 .2 9 0 x 10~2 m 2 .
 IAEA-SM-327/71 347

design and the general rearing conditions meant that there was no climatic or other
control o f the rearing environment. In the end, the attempt to rear G. m. morsitans
in its natural habitat was unsuccessful. The next stage o f the experiment therefore
relied upon the use o f field collected materials.
Sterility was induced in the males by dipping field collected pupae in a
5 % solution o f the chemosterilant tepaR. Irradiation was not used because only
pupae o f unknown ages were available, which meant that mortality from the over­
exposure o f younger pupae to high doses o f gamma radiation could be excessive [3].
Prior to the release o f sterile flies, the original population (estimated at
3000 flies/m ile2) was reduced by two aerial applications o f dieldrin.2 Trays o f
chemosterilant treated pupae were placed at selected points in the test area. Within
20 months o f the operation, the population o f G. m. morsitans was reduced to below
the detectable level [3].
Complete eradication was not achieved for several reasons. The number o f
field collected pupae available for sterilization was limited. Because, o f the limited
number o f pupae, only a small number (1 .2 % o f any batch) o f treated pupae emerged
as sterile males each day. It was, therefore, necessary to leave the pupae in the field
for a long time in order to obtain an adequate number o f released sterile males. The
long exposure, in turn, meant that the chemosterilant became unstable and, there­
fore, failed to produce com plete sterility in the flies emerging.later [3]. In addition,
the pupae were o f unknown ages, therefore rather high mortality resulted from
exposing ‘ under age’ pupae to the chemosterilant. The test results emphasized the
need to obtain known age pupae and/or adults for sterilization. Uncertainties about
the reliability o f chemosterilization would require that gamma irradiation be used as
the preferred method o f inducing sterility..

3. THE PROJECT A T M K W A N JA R A N C H A N D T A N G A ,
UN ITED REPUBLIC OF T A N Z A N IA

The first large scale SIT field project involving the use o f laboratory reared
radiation sterilized tsetse flies was initiated towards the end o f 1971 in the United
Republic o f Tanzania. The Tanga based project was, in fact, the continuation o f
studies begun four years earlier in Zim babwe. The objective was to integrate the
release o f sterile males with limited applications o f insecticides to control the tsetse
population. The research would also demonstrate whether good quality flies could
be produced under African conditions for use in a SIT programme. The project was
jointly funded by the U SA ID and the Government o f the United Republic, o f
Tanzania.

2 1 m i l e 2 = 2 . 5 9 0 k m 2.
348 OFFORI

The target species, G. m. morsitans, was colonized in the laboratory using flies
from pupae collected in the field (Handeni), about 100 km from Tanga, the project
headquarters [4]. T he colon y, held in three separate facilities, was fed daily on goats
purchased locally and another group on rabbits imported from Europe. The newly
acquired goats were quarantined and treated for trypanosomiasis and worm infec­
tions before being used for fly feeding. From the 60 000 fly colony (45 000 females
and 15 000 males), approximately 6000 surplus males were available weekly for
sterilization and release [4]. A back-up colony o f some 10 000 females was main­
tained in the A gen cy’ s Laboratory at Seibërsdorf.
The target area, Mkwanja Ranch, 195 km 2 in size and located 100 km by
road from the production centre in Tanga, proved ideal all round for the study,
except that it was not strictly ‘ isolated’ . Therefore, an artificial barrier (1 km in
width) was constructed on the perimeter o f the project area'in an attempt to ensure
isolation and prevent fly immigration from the adjoining vegetation [5].
Detailed studies o f the natural tsetse fly population on the Mkwanja Ranch
were follow ed by two aerial applications o f endosulfan at 28-d intervals in order to
suppress the native population o f G. m. morsitans before the release o f sterile males.
The flies were sterilized by exposing late stage pupae to 11.8 krad o f gamma radia­
tion from a 137Cs sou rce.3 Pupae collected over a three or four day period were
irradiated as a group. Sterilized pupae were held refrigerated at 10°C while being
transported by road to the release site. Refrigeration ensured synchronous emer­
gence, so that a large number o f sterile males became available within a very short
time follow ing pupal release in the field. Releases were made twice a week, giving,
on average, 135 sterile males per square kilometre per month. During the 15 months
o f fly releases, a total o f 510 000 pupae were utilized at a cost o f US $0.22 per
pupa [4].
The main achievements o f this programme can be summarized as follow s:

(1) For the first time, a large, self sustaining colony o f tsetse flies was raised and
maintained under African conditions. The use o f local animals (goats) to mass
rear G. m. morsitans was clearly demonstrated, and the flies produced from
such a system were o f high quality. A major objective o f the experiment was
thus realized.
(2) With a combination o f limited application o f insecticides and release o f sterile
males, the natural population o f the target species o f tsetse was reduced, albeit
by only 81 % . Total eradication was not achieved because the target area was
reinvaded by flies from the adjoining vegetation. The barrier clearing became
overgrown with bush because o f unseasonably heavy rains during the project
period.

3 1 rad = 1.00 x 10'2 Gy.

 IAEA-SM-327/71 349

4. SIT A G A IN ST RIVERINE A N D S A V A N N A H TSETSE

IN B U RK IN A FASO

Experiments carried out in Burkina Faso between 1976 and 1984 aimed at
testing and then applying SIT for eradicating riverine and savannah tsetse flies.
In a five year research programme, Glossina palpalis gambiensis was eradi­
cated in a 32 km (100 km 2 area) riverine forest through combined application o f a
non-persistent insecticide, thiodan and the release o f laboratory reared sterile males
o f the target species [6]. The research involved testing the effectiveness o f Challier-
Laveissière traps, mechanical barriers (total clearing and burning) and application o f
persistent insecticides for re-enforcing the barriers.
A large scale programme initiated in 1981 was directed against two riverine
species o f tsetse, G. p. gambiensis and G. tachinoides, and one savannah species,
G. m. submorsitans. Jointly sponsored by the German Gesellschaft fur Technische
Zusammenarbeit (G T Z ) and the French Institut d ’ élevage et de médecine vétérinaire,
des pays tropicaux (IE M V T) and supported by the Government o f Burkina Faso, this
was the first large scale tsetse SIT project directed against eradication o f more than
one species at the same time. The project covered 3500 km 2 o f agropastoral land,
including approximately 500 linear kilometres o f riverine forest. Clearing tsetse flies
in the area would permit the development o f livestock in a 240 000 ha area o f the
Sideradougou agropastoral zone. The project headquarters and mass rearing facility
were located at B obo Dioulasso [7].
The flies were fed in vivo (on rabbits) or in vitro (on defibrinated or
heparinized bovine blood collected from the local abattoir). T o prevent bacterial
infection, the blood was treated with 50 krad gamma radiation in a 60C o source.
A t the peak o f production, the colonies contained approximately 150 000 G. p.
gambiensis, 115 000 G. tachinoides and 40 000 G. m. submorsitans [7].
Prior to fly release, a thorough entomological survey was undertaken over
the entire Sideradougou area. A ccess roads were first provided, and C hallier-
Laveissière traps were placed at 100 m intervals for a full day. In the savannah areas,
parallel transects were cut every 4 -5 km, and surveys similarly undertaken. T o iso­
late the experimental zone, the entire drainage system was protected by three perma­
nent barriers comprising traps and insecticide (deltamethrin) impregnated screens
placed 100 m apart; some were hung from tree branches. In dense forests, the
screens were placed 2 0 -3 0 m apart. In the m ore open country to the southeast, as
many as 1700 traps were deployed [7].
A total o f 7204 screens deployed over a two month period reduced the fly
population by 88% for G. p. gambiensis and 92% in the case o f G. tachinoides.
Adult males sterilized with 11— 13 krad gamma radiation were fed one blood
meal before being transported for release at the experimental site. Releases
were done during the rainy season, first on a. weekly basis but later fortnightly.
In all, more than 900 000 sterile males (713 000 G. p. gambiensis and
350 OFFORI

225 000 G. tachinoides) were released. Complete eradication was achieved within
four months.
The Burkina Faso project demonstrated for the first time the use o f insecticide
impregnated screens for tsetse population suppression and as strategic barriers
against reinvasion. A lso, for the first time sterile flies were fed in the laboratory
before being released. Both practices were later adopted and used with positive
results in the Nigerian project, BICOT.

5. BICO T, NIGERIA

The project Biological Control o f Tsetse Using the Sterile Insect Technique
(BICOT) was initiated in 1979 follow ing several preparatory meetings between the
Joint F A O /IA E A Division o f Nuclear Techniques in Food and Agriculture and o ffi­
cials o f the Government o f Nigeria. With financial contributions by Belgium,
Germany, Italy and Sweden, and in-kind contributions from the United Kingdom ,
the experiment was successfully concluded in 1986. The objective was to demon­
strate the use o f SIT in eradicating Glossina palpalis palpalis, and to investigate ways
o f integrating the technology into the regular tsetse control operations in Nigeria.
The target species, G. p. palpalis, was mass reared at the project headquarters
in V om , using both in vivo (guinea pigs) and in vitro (fresh bovine and freeze dried
porcine blood) feeding techniques. The bovine b lood was obtained from the local
abattoir, but the freeze dried porcine blood was imported from the A gen cy ’ s Labora­
tory at Seibersdorf. At peak production, the colony contained 180 000 females. A
back-up colony o f about 80 000 females was maintained at the A gen cy ’ s Laboratory
on an in vitro feeding system [8].
The project field area, located approximately 250 km by road south o f V om ,
is part o f the Lafia Agricultural Development Project, which included provision for
crop farming and livestock grazing. The target area, 1500 km2 in extent, was
characterized by three major river systems: Akuni, Achiba and Ehula-Ganye. These
were treated as separate operational areas. Extensive cultivation by subsistence
farmers provided an effective barrier to the south o f the project area. Otherwise,
no physical barriers such as vegetation clearing and bush burning were instituted;
however, the western boundary zone was secured by limited ground spraying with
dieldrin. The native fly population was suppressed using biconical traps and
deltamethrin impregnated blue screens. This approach, already demonstrated in
Burkina Faso, proved effective in reducing the population to as low as 5% o f the
original.
Males sterilized as 2 to 4-d-old adults with 120 Gy o f gamma radiation in a
60C o source were fed at least two blood meals before being released. Giving them
at least one blood meal lowered the probability o f their becom ing infected with
trypanosomes. Fly releases were done at weekly intervals from predetermined points
 IAEA-SM-327/71 351

2 km apart. Up to 800 sterile males were released at each point. Initially, a 3:1 ratio
o f sterile to native males was used. Later, the ratio was increased to 10:1. Survival
o f the sterile males was excellent in spite o f the 3 h transportation by car from V om
to the release site. O ver 25% o f the males were still alive two weeks after release.
Using the release ratio o f 10:1, eradication was achieved in 8 -1 2 months, i.e. by the
end o f 1986, in all the river systems within the 1500 km 2 project area.
T o prevent reinfestation o f the cleared area, insecticide impregnated blue
screens were placed at critical areas 100 m apart, on the project boundaries. The
interval was decreased to 50 m or less in areas o f denser vegetation [9]. Screens were
reimpregnated every three to four months. Frequent (fortnightly) visits were made
to inspect the screens and to replace the lost ones.

6. GEN E RA L C O M M E N TS

From the foregoing review, several lessons emerge that should benefit and
guide future tsetse SIT programmes.

6.1. Mass rearing

The difficulty in mass rearing tsetse flies was one o f the main obstacles to
applying SIT for tsetse eradication in the past. Since the days o f the Zimbabwe
programme, tsetse rearing on a large scale has becom e a routine activity. Fly colony
maintenance is now done using artificial feeding systems, with little or no reliance
on live animals. Thus, the expense involved in maintaining large colonies o f host
animals can be eliminated. This approach was demonstrated effectively in Burkina
Faso and Nigeria. The project in Burkina Faso also demonstrated the possibility o f
mass rearing and releasing more than one species o f tsetse at the same time. Since
1988, varying numbers o f six different species o f Glossina have been maintained
successfully on an artificial rearing system at the A gen cy’ s Laboratory at Seibers-
dorf. Clearly, the basic obstacle to application o f SIT for tsetse control or eradication
has been overcom e. What is needed is to refine rearing operations and to introduce
procedures that would further reduce operational costs.

6.2. Sterilization

With field collected pupae o f unknown age (as was the case in the Zimbabwe
programme), chemosterilization provides a means o f inducing sterility in emerging
adults without the high mortality associated with the gamma irradiation o f young
pupae. H ow ever, gamma radiation sterilization is safer, more accurate and much
simpler to administer than chemosterilization. M oreover, the programme in Burkina
Faso and Nigeria demonstrated that gamma sterilization o f the adult does not produce
352 OFFORI

any deleterious effect on the survival, behaviour or sexual performance o f sterile

males released in the field.

6.3. Prerelease population suppression

Suppression o f the native fly population was a major undertaking in all the pro­
grammes. W hile limited spraying o f insecticides was used in the initial programmes
to achieve suppression, the same result was obtained in later programmes using
insecticide impregnated targets or screens placed at vantage locations for a limited
time, prior to release o f the sterile males. In all cases, the screens/targets were e ffec­
tive in reducing the tsetse population by as much as 95 % .

6.4. „Barriers against reinvasion and réintroduction

Maintenance o f barriers is an essential part o f any tsetse eradication

programme, especially if the target area is relatively small, as was the case in the
four programmes described. Without a secure barrier system, reinvasion o f the
cleared area is almost a certainty. Thus, creating and maintaining barriers was con­
sidered important in all these programmes. In an island situation, a barrier as such
would not be necessary; however, réintroduction o f flies from outside sources must
be prevented through the imposition o f strict quarantine measures. In the vast open
savannah type o f situation, an artificial barrier may have to be created through com ­
plete removal o f trees and bushes, as was done in the Tanzanian (Mkwanja Ranch)
project. On the other hand, in the two W est African programmes (Burkina Faso and
Nigeria) insecticide impregnated screens served both for suppressing the native fly
population prior to the release o f sterile males and for. preventing reinvasion from
outside. In addition, in the Nigerian programme sterile males were used as a
‘ biological’ barrier to secure the various river systems follow ing eradication o f the
target species.
Whatever the method o f choice, the cost o f creating, maintaining and monitor­
ing the effectiveness o f the barriers can be high. Such a cost must be considered an
integral part o f the total cost o f a tsetse SIT programme.

7. CURREN T ACTIVITIES

The tsetse SIT programmes concluded to date were supported financially,

either through bilateral type arrangements (Zim babwe, the United Republic o f
Tanzania and Burkina Faso) or through extrabudgetary funds provided through an
international organization (Nigeria). Considering the total inputs for each o f the
projects described, it is obvious that similar funding would be necessary for future
programmes. Unfortunately, very few , if any, bilaterally assisted tsetse SIT pro­
 IAEA-SM-327/71 353

grammes have been initiated since the U SA ID project in the United Republic o f
Tanzania and the successful Germ an-French project in Burkina Faso. A s far as inter­
national organizations are concerned, it is pertinent to note that the F A O is support­
ing tsetse/trypanosomiasis control projects in several countries in East, Central and
West A frica. H ow ever, none o f these has an SIT component, except the one on
Zanzibar, United Republic o f Tanzania, where the IA E A is also involved.
The IA E A is currently supporting small to medium scale tsetse SIT projects
through its technical co-operation programme in several African countries. Activities
in Ghana, Mali, Nigeria, Uganda, the United Republic o f Tanzania and Zambia
include ecological studies o f the target species in a given country or locality,
establishment o f tsetse colonies and experimental feeding in vitro or, in some cases,
mass rearing in anticipation o f large field programmes.
In Zanzibar, the United Republic o f Tanzania, a major SIT programme is
planned, starting in 1993, to eradicate G. austeni, the only tsetse species on the island
and which is the sole transmitter o f animal trypanosomiasis on Zanzibar. For this
programme, the facility in Tanga is already being used to produce sterile flies for
release on Zanzibar. E cological studies have been initiated that involve the testing
o f various devices for trapping the target species, G. austeni, which has eluded
detection for many years. In the Jozani forest, one o f the main habitats o f the species,
baseline data on fly abundance and population structure are currently being assem­
bled. Both sticky panels and insecticide impregnated blue screens are being evaluated
for population sampling and population suppression. At the same time, an F A O field
team, supported by staff o f the Zanzibar Department o f Livestock Development, is
engaged in fly population suppression using insecticide (deltamethrin) ‘ spot-on’
treatment o f animals. Monitoring o f animal trypanosomiasis in the island’ s cattle
population is also in progress. Pilot releases have been initiated using sterile male
flies from the Tanga mass rearing facility. Results to date indicate that the sterile flies
adapt and disperse well within the forest. It is anticipated that a full scale SIT
programme will be in place in early 1993 and that eradication o f G. austeni will be
completed within tw o years. The programme is already benefitting from a substantial
financial input from F A O . Both the Organization o f Petroleum Exporting Countries
(OPEC) Fund and the International Fund for Agricultural Development (IF A D ) have
pledged financial support. Additional funding is being solicited from the Government
o f Belgium and other sources.

8. FU TU RE PROSPECTS FO R TSETSE SIT

It has been shown that when integrated with other methods, SIT could be an
effective tool for eradicating tsetse flies or for reducing tsetse populations to manage­
able levels if control is the objective. The difficulty o f maintaining viable laboratory
colonies o f tsetse has been overcom e, and mass rearing is now a routine operation.
354 OFFORI

A number o f African scientists and technicians have been trained in the use o f the
technology, and several administrators in A frica have been made aware o f the poten­
tial o f the technology in solving the tsetse/trypanosomiasis problem . Therefore, it is
reasonable to expect that requests for assistance will increase in the future to apply
SIT for tsetse/trypanosomiasis control. H ow ever, application o f SIT on a routine
basis for large scale tsetse population management will only be possible when certain
additional conditions have been satisfied.
First, governments o f the affected African countries must decide whether and
under what conditions to utilize the technique as part o f their tsetse/trypanosomiasis
control programme. In making this decision, governments will depend upon tsetse
specialists, preferably their own nationals, who are knowledgeable in all aspects o f
the problem . Therefore, it is important that tsetse scientists in A frica becom e more
than familiar with SIT.
Second, governments, having made the decision to utilize the technique, must
be willing to comm it resources, unreservedly, for implementing the programme.
This commitment is even more important should it becom e necessary to conduct a
regional tsetse control programme embracing several countries. In this connection,
it must be conceded that although 38 African countries are listed as ‘ affected by
tsetse’ , the priority rating o f the problem is not the same in all countries, for various
reasons.
Third, external (donor) support must be assured because o f the inability o f
some national governments to cope with the financial requirements o f tsetse eradica­
tion programmes. Unfortunately, donor countries have their own preferences for
countries to assist in A frica, and sometimes their own preferred tsetse control/
eradication approaches. Therefore, it is necessary to work towards the harmoniza­
tion o f the attitudes, approaches and objectives o f donors.
Fourth, it will be necessary to keep the cost o f SIT operations as low as
possible. (Knipling — quoted by Dame and Schmidt [3] — suggested that tsetse
pupae produced at US $0.05 per pupa would be econom ically acceptable for SIT
programmes.) Since one o f the main sources o f the high cost is fly production and
handling, labour requirements for mass rearing.must be reduced. Therefore, serious
consideration should be given to establishing regional rearing facilities. Such an
approach would take away the financial burden from individual countries setting up,
equipping and running mass production facilities. A well designed, robust produc­
tion facility, one for western and another to serve eastern African countries, would
be used by several countries on a cost sharing basis. It was recently recommended
by a group o f consultants meeting in Vienna that, in designing regional rearing facili­
ties, a ‘modular system’ should be considered as it has the advantage that it could
be expanded or contracted, depending upon circumstances. Furthermore, construc­
tion o f these mass rearing facilities has been adopted as a goal in the IA E A medium
term plan.
 IAEA-SM-327/71 355

9. CONCLUSIONS

Until the scourge o f trypanosomiasis is totally eliminated, tsetse control or

eradication will continue to be a major concern for all affected countries in A frica.
In the fight against tsetse flies, insecticide application will continue to feature promi-.
nently until other proven, more effective, alternatives becom e available.
Currently, traps and insecticide impregnated targets and screens are being
promoted by several organizations, including the F A O ; village communities and
individual stock.farmers are being encouraged to utilize these simple and inexpensive
devices to control tsetse flies. Although traps and targets have been shown to be
effective in reducing tsetse populations to a .very low level, the evidence to date is
that these methods are not effective for eradicating tsetse populations [8]. Therefore,
where eradication is the objective, a more effective approach would be required.
A lso, it should be borne in mind that tsetse populations could, possibly, develop
resistance to insecticides and even targets and screens.
The effectiveness o f SIT in eradicating tsetse populations has been demon­
strated. It is an environmentally acceptable approach for tsetse population manage­
ment. The technique can be used most effectively in combination with other
population reduction methods and, more importantly, in area wide tsetse manage­
ment programmes as opposed to field by field, operations that individual farmers
could undertake using tsetse traps and targets. In the field by field approach, there
is always the danger o f flies from an adjoining untreated farm or plot invading a
‘treated’ area. Using the area wide approach, an entire tsetse infested area covering
several livestock, farms, for example, could be effectively covered. Therefore, in
addition to being an effective tool for tsetse eradication, SIT should be considered
the method o f ch oice for eradicating tsetse populations in a given area, with little or
no deleterious effect on the environment. Currently, a major concern to address-in
order to make the technique routinely applicable is to keep costs to the minimum.
With improvements in mass rearing, including the introduction o f automation plus
detailed planning o f field programmes, the cost o f producing sterile flies and o f the
entire operation o f tsetse SIT programmes should be brought down to within
financially acceptable limits.
The number o f countries opting to. apply SIT for tsetse control may be expected
to increase in the future. Therefore, it would be more practical and less expensive
to plan and execute SIT programmes on a regional basis than on an individual
country basis. The approach could be modelled on the recently concluded Oncho­
cerciasis Control Programme in West A frica, which involved several countries o f the
subregion. Regional tsetse SIT programmes would indeed give a m ore realistic and
true meaning to the idea o f area wide control, for which SIT is suited.
The importance o f controlling or eradicating tsetse in order to eliminate
trypanosomiasis cannot be overemphasized. Since 1974, when the W orld F ood C on­
ference o f the United Nations mandated the F A O “ to launch ... a long-term
356 OFFORÏ

programme for the control o f African animal trypanosomiasis as a project o f high

priority” , considerable emphasis has been placed on the fact that tsetse flies occupy
approximately 10 million km2 o f the land surface in tropical and subtropical Africa.
The point has been stressed repeatedly in official F A O documents that over 70% o f
this area could be used to produce an additional 120 million head o f cattle if freed
o f tsetse. In other words, the presence o f tsetse in 70 million km2 o f land in Africa
is effectively robbing the continent o f 1.5 million tonnes o f meat per year [10].
The F A O Special Action Programme that was initiated follow ing the United
Nations mandate makes provision for drug treatment o f animal trypanosomiasis
cases, breeding o f trypanotolerant cattle and control o f the tsetse vector. In addition,
emphasis is placed on area development follow ing tsetse control or eradication.
Therefore, in some situations, for example, when area development means establish­
ing a cattle ranch in a tsetse infested area, tsetse eradication would not only be justi­
fied but would be the only means o f ensuring protection o f livestock that could be
at risk from trypanosomiasis.
Finally, a word on the role o f international organizations. The primary reason
for controlling or eradicating tsetse in vast areas o f the African continent is to
eliminate trypanosomiasis and thus create conditions for increased agricultural
production and sustainable rural development. This noble ideal can only be attained
with external financial support from individual donor countries and the international
community, because o f the generally poor econom ies o f most o f the affected
countries. The experiments carried out over the past 25 years to test the feasibility
o f SIT tsetse control/eradication were possible only because donor countries were
willing to assist recipient countries in A frica. It is true that two. United Nations
organizations, F A O and the W orld Health Organization, are already involved in
activities aimed at controlling animal and human trypanosomiasis, respectively.
Since vector eradication can lead to total elimination o f trypanosomiasis, perhaps the
programmes and approaches adopted by the F A O and W H O would be more effective
in eliminating trypanosomisis if some aspects o f their respective programmes were
to be harmonized with those o f other organizations, especially the IA E A , which is
responsible for the research, development and application o f SIT. The W H O played
a key role in the successful implementation o f the Onchocerciasis Control
Programme in West A frica. Perhaps it is not too much to expect other UN bodies
to play a leading role, not only in identifying regional tsetse SIT projects but also
in assisting the affected African countries in their planning, funding and execution.

REFERENCES

[1] B U S H L A N D , R.C., “Sterility principle for insect control: Historical developments

and recent innovations”, Sterility Principle for Insect Control or Eradication (Proc.
Symp. Athens, 1970), IAEA, Vienna (1971) 3-14.
 IAEA-SM-327/71 357

[2] D A V I D - W E S T , K.B., “Integrated tsetse control: Present and future programmes of

national and international organizations”, Integrated Tsetse Fly Control: Methods and
Strategies ( C A V A L L O R O , R., Ed.) (1978) 153.
[3] D A M E , D.A., S C H M I D T , C., The sterile-male technique against tsetse flies,
Glossina spp., Bull. Entomol. Soc. A m . 16 1 (1970) 24.
[4] W I L L I A M S O N , D.L. et al., Integration of insect sterility and insecticides for control
of Glossina morsitans morsitans Westwood (Diptera: Glossinidae) in Tanzania. I.
Production of tsetse flies for release, Bull. Entomol. Res. 73 (1983) 259.
[5] G A T E S , D.B. et al., Integration of insect sterility and insecticides for control of
Glossina morsitans morsitans Westwood (Diptera: Glossinidae) in Tanzania. III. Test
site characteristics and the natural distribution of tsetse flies, Bull. Entomol. Res. 73
(1983) 373.
[6] P O L I T Z A R , H., C U I S A N C E , D., “SIT in the control and eradication of Glossina
palpalis gambiensis", Sterile Insect Technique and Radiation in Insect Control (Proc.
Symp. Neuherberg, 1981), IAEA, Vienna (1982) 101-109.
[7] K A B O R E , I., B A U E R , B., “Integrated control against Glossines in Burkina Faso”,
Integrated Tsetse Fly Control: Methods and Strategies ( C A V A L L O R O , R., Ed.)
(1987) 71.
[8] O L A D U N M A D E , M.A. et al., “Eradication of Glossina palpalis palpalis (Robineau-
Desvoidy) (Diptera: Glossinidae) from agropastoral land in Central Nigeria by means
of the sterile insect technique, Sterile Insect Technique for Tsetse Control and Eradica­
tion (Proc. Final Research Co-ordination Mtg V o m , 1988), IAEA, Vienna (1990) 5-23.
[9] T A K K E N , W., “The sterile insect technique for tsetse eradication in Nigeria”,
Integrated Tsetse Fly Control: Methods and Strategies ( C A V A L L O R O , R., Ed.)
(1987) 83.
[10] FINELLE, P., “Programme for the control of African animal trypanosomiasis and
related development”, Isotope and Radiation Research on Animal Diseases and their
Vectors (Proc. Symp. Vienna, 1979), IAEA, Vienna (1980) 3-14.
POSTER PRESENTATION
IAEA-SM-327/16P

EVALUATION OF ORIENTAL FRUIT FLY

CONTROL USING THE STERILE INSECT TECHNIQUE
AT DOI ANG KHANG

M . S U T A N T A W O N G , P. K A O C H O N G ,
W . LIM O P A SM A N E E , M . L U A N G A P IC H A T K U L
O ffice o f Atom ic Energy for Peace,
Chatuchak, Bangkok,
Thailand

The Oriental fruit fly, Dacus dorsalis Hendel, is a major pest o f fruits in
Thailand. Its attack on fruits not only reduces the yield but also affects international
trade because o f the quarantine restrictions imposed by major importing countries.
Heavy reliance on insecticide applications has led to new problems such as a resur­
gence o f secondary pests, undesirable chemical residues and environmental contami­
nation. The sterile insect technique (SIT) is one o f the few highly specific control
methods which can be used to overcom e these problems. D oi Ang Khang was
selected for the pilot test as a model for larger operations. The 20 km 2 D oi Ang
Khang area located at the northwest tip o f the Chiang Mai provincial area (northern
Thailand) is highland country (1400 m above sea level) comprising semi-isolated
areas surrounded by hills and slopes. It is covered by regular plantations o f peach,
pear, persimmon and other fruit crops. The O ffice o f A tom ic Energy for Peace,
which has pioneered SIT activities against fruit flies at D oi Ang Khang since 1982,
has extended these activities until 1996.
Fruit fly pupae to be sterilized and released were reared in the mass rearing
facility in Bangkok. B efore irradiation, the pupae were marked with a fluorescent
dye two days before emergence. The marked pupae were packed into 800 m L
narrow polyethylene bags, which were then closed with rubber bands. The pupae
were sterilized at 90 Gy with gamma rays from a ^ C o source. The bags o f pupae
were packed into polystyrene containers and kept co o l with ice packs. Five million
sterilized puae were transported to D oi A ng Khang by train and vehicle. During the
period February to September, sterilized pupae were sent to the area and released
every week in 15 release huts. After the release o f the sterilized flies, the percentage
o f infested peaches decreased from 54.7% in 1984 to 17.1% in 1991. In 1991, the
cost o f fruit fly control using SIT was US $30 397, while the net return from the
increasing value o f the fruits was US $121 558.
In addition, the public has benefited from some o f the advantages accrued from
SIT, namely, reduced risks o f toxic damage and pollution.

359
STERILITY AND INSECT BEHAVIOUR
(Session 5)
Chairman
У.А. DYCK
Canada
 IAEA-SM-327/35

INTEGRATION OF INHERITED STERILITY

AND OTHER PEST MANAGEMENT STRATEGIES
FOR Heticovefpa zea
Status and potential

J.E. C AR PEN TER

Insect B iology and Population Management Research Laboratory,
Agricultural Research Service,
United States Department o f Agriculture,
Tifton, Georgia,
United States o f Am erica

Abstract
I N T E G R A T I O N O F I N H E R I T E D S T E RILITY A N D O T H E R P E S T M A N A G E M E N T
S T R A T E G I E S F O R Helicoverpa zea: S T A T U S A N D P O T E N T I A L .
The corn earworm, Helicoverpa zea (Boddie), is one of the most destructive pests of
field crops in the United States of America. Currently, control of H. zea is achieved almost
entirely through the use of synthetic insecticides. However, successful application of the
inherited sterility principle to a wild population of H. zea during a pilot test has encouraged
further development of this pest control strategy. Data from recent studies and population
models suggest that the full potential of inherited sterility as an area wide control strategy for
H. zea and other lepidopteran pests m a y be realized only when inherited sterility is integrated
with other suppression methods. Applications of singular, non-insecticidal methods for area
wide suppression of H. zea have had limited success and have been vulnerable to temporary
programme failures (e.g. inclement weather, reduced insect rearing output, etc.). A n
integrated approach would be horizontally diversified through the use of several strategies
which use different modes of action and, therefore, would be less vulnerable to temporary
programme failures than would a single strategy. Combining two or more control strategies
with additive effects may increase programme efficiency. However, the greatest benefit in
combining inherited sterility with other control strategies is the dynamic synergistic effects.
The demonstrated potential of inherited sterility to reduce the reproductive ability of H. zea
and the demonstrated capacity of inherited sterility to perform compatibly and synergistically
with other control strategies suggest that inherited sterility should be a major component of
future integrated approaches to managing H. zea.

1. IN TR O D U C TIO N

The c o m earworm, Helicoverpa zea (B oddie), is one o f the most destructive

pests o f field crops in the United States o f America. Currently, control o f H. zea
is achieved almost entirely through the use o f synthetic insecticides on a crop to crop
basis. Such a management strategy is costly, non-selective and only effective at the

363
364 CARPENTER

target site for a short period o f time. Insecticide resistance, the mounting concern
over pesticide pollution and the desire to effectively manage H. zea both within fields
and on an area wide basis have encouraged scientists to identify and develop new
methods o f control.
The potential for using inherited sterility as a component o f the area wide
management o f H. zea has been discussed by LaChance [1]. Knipling [2] demon­
strated the potential advantage o f inherited sterility over the sterile insect technique
(SIT) through the use o f population models. Carpenter et al. [3] found that sub-
sterilizing doses o f radiation induced deleterious effects in H. zea that were inherited
through the F2 generation. Subsequently, a series o f investigations was conducted to
further define and understand the effects o f inherited sterility in H. zea. Irradiated
(100 G y), laboratory reared moths were competitive with non-irradiated, laboratory
reared moths in attracting and securing mates under field conditions [4]. Females that
mated with irradiated males, and females that mated with non-irradiated males had
the same mating propensity and the same refractory period [5]. Although there was
a difference in mortality, between H. zea larvae from irradiated and non-irradiated
parents when they were reared in the laboratory, this mortality differential was
reduced when larvae were reared under natural conditions in the field [6]. Other
studies have revealed no interaction between inherited sterility and diapause in
H. zea [7].
A s a result o f these studies, a pilot test was conducted in small mountain
valleys in North Carolina (U SA ) to assess the influence o f released, substerilized
(100 Gy) males on wild H. zea populations and to measure the infusion rate o f
inherited sterility into the wild population. Results from this study revealed that the
number o f wild males captured per hectare was positively correlated with the
distance from the release site o f irradiated males. Analyses o f the seasonal population
curves o f wild H. zea males calculated from mark-recapture data suggested that
seasonal increases o f wild H. zea males were delayed and/or reduced in the mountain
valleys where the irradiated males were released. The incidence o f larvae with
chromosomal aberrations (progeny o f irradiated, released H. zea males [8]) collected
from the test sites during the growing seasons indicated that irradiated males were
very competitive in mating with wild females and were successful in producing F]
progeny, which further reduced the wild population.
Successful application o f the inherited sterility principle to a wild population
o f H. zea during this pilot test has encouraged further development o f this pest con ­
trol strategy. H owever, because H. zea is a highly m obile,xpolyphagous pest with
a wide geographical distribution, and because the technology required to rear H. zea
econom ically is incomplete, the conventional methodology employed in SIT pro­
grammes may benefit from certain modifications. Therefore, the development o f a
pest management approach in which inherited sterility is integrated as a primary
component with other pest control strategies is to be encouraged.
 IAEA-SM-327/35 365

2. C U RR E N T STATUS

Knipling [9] demonstrated the potential benefit o f combining sterile insects

with conventional control methods. Recently, laboratory studies have revealed that
inherited sterility is compatible with other methods o f pest control. In a study using
two insect strains and two insecticides, Carpenter and Young [10] found that progeny
from irradiated parents and progeny from non-irradiated parents demonstrated the
same level o f insecticide resistance. Because the ratio between irradiated and non-
irradiated insects is the most critical factor in regulating the efficacy o f SIT or
inherited sterility release programmes, insecticide applications during a continuous
release o f irradiated insects would benefit the release programme. Although the
insecticide would reduce the number o f both irradiated and non-irradiated insects,
it would not change the ratio. Therefore, subsequent to insecticide applications the
wild population would be at a low er level and the ratio o f irradiated to non-irradiated
insects would be increased by continual releases o f irradiated insects. Host plant
resistance is another method o f pest control that would reduce the rate o f increase
for both irradiated and non-irradiated insects, while simultaneously allowing the con­
tinual release o f irradiated insects to increase the ratio o f irradiated to non-irradiated
insects. Carpenter and Wiseman [11] investigated the effects o f inherited sterility and
host plant resistance on H. zea development, and found that larvae resulting from
irradiated males by non-irradiated female crosses were equally competitive with
normal larvae for all the parameters measured.
Additional laboratory and field studies have been initiated to measure the
compatibility o f inherited sterility and augmented natural enemies such as parasitoids
and viruses. Preliminary data indicate that parasitoids and viruses are compatible
with the inherited sterility method.

3. PERSPECTIVE

Integration o f inherited sterility and other pest control strategies into a

successful management approach is predicated upon the validity o f the follow ing:
(1) inherited sterility is effective at low release ratios; (2) inherited sterility is effec­
tive at different rates o f increase for the pest population; (3) other pest control
strategies do not negatively impact irradiated insects and their progeny more than
that o f the wild population; and (4) the inherited sterility component is flexible in
its application.

3.1. Influence of low release ratios and rates of increase

A computer model using data from Carpenter et al. [3] quantified the influence
o f different parameters on the predicted efficacy o f the inherited sterility method.
366 CARPENTER

Ratio (irradiated:non-irradiated)

FIG. 1. Per cent reduction in normal population growth at different release ratios of
irradiated (100 Gy) males.

Rate of increase per generation

FIG. 2. Relationship between the per cent reduction in normal population growth and the rate
o f increase per generation.
 IAEA-SM-327/35 367

Increases in the release ratio had the greatest effect on the percentage reduction in
normal population growth when the release ratio was less than 10:1 (Fig. 1). Even
a release ratio as low as 1;1 reduced the normal population growth by 30 % . These
model predictions were corroborated in a pilot test (unpublished data) in which
release ratios averaging less than 5:1 reduced the wild population o f H. zea by an
average ( ± S E ) o f 73.5 ± 1 4.9% .
The population model suggests that the efficacy o f inherited sterility is almost
constant when the rate o f increase per generation is tw ofold or higher (Fig. 2). This
is in contrast to the efficacy o f SIT which declines as the intrinsic rate o f increase
rises. The model also suggests that a substantial increase (3 % ) in treatment efficacy
o f inherited sterility occurs when the normal population is in decline (0.5 fold per
generation). Because the H. zea population normally declines follow ing the
em ergen çe-of the overwintering generation, the increase in treatment efficacy o f
inherited sterility revealed in the model provides an additional potential for managing
this pest population.

3.2. Impact of other pest control strategies

Available data from insecticide resistance and host plant resistance studies
suggest that the impact o f other pest control strategies will not be greater for
irradiated insects and their progeny than for wild insects. A s explained earlier,
conventional uses o f both insecticides and host plant resistance would increase the
ratio o f irradiated to non-irradiated insects, thus producing synergistic effects.
Furthermore, models developed by Knipling [12, 13] depicting different integration
scenarios suggest that combining inundative releases o f parasites with sterile insects
will yield both additive.and synergistic, effects. Although the parasite technique and
SIT have different modes o f action, the effectiveness o f SIT increases the ratio o f
adult parasites to adult hosts, and the effectiveness o f the parasites increases the ratio
o f sterile to fertile insects. Knipling [13] calculated that com bining sterile insects
with parasites theoretically could be 10 000 times m ore effective than i f each tech­
nique were used alone. W hile these figures are impressive, even greater suppression
could be expected i f parasite releases were combined with the inherited sterility
technique. Not only is inherited sterility in H. zea m ore effective than full sterility
in reducing population increases, but the inherited sterility technique produces sterile
F! larvae that would provide an increased number o f hosts for the parasites. There­
fore, the number o f parasites produced would increase even if the rate o f parasitism
remained the same (host density independent), and whether or not additional para­
sites were released.
368 CARPENTER

3.3. Flexibility of inherited sterility application

Flexibility in application o f the inherited sterility method has been described

by LaChance [1]. The level o f sterility and the number o f individuals in the Fj and
F2 generations are controlled by the dose o f radiation administered to the P! genera­
tion, and by whether only irradiated males or both irradiated males and females are
released. This flexibility should be enhanced by the integration o f other pest control
methods, thus allowing management strategies to.be m odified as seasonal changes
occur in the pest status o f H. zea and the host plant availability.
There are many different scenarios in which an integrated approach employing
inherited sterility could be envisioned to control H. zea populations. The release
o f partially sterile male and female H. zea would produce large numbers o f sterile
larvae that could be field reared on early season weeds or possibly whorl stage corn.
Parasites (native and/or released) could use these sterile larvae as hosts and, thereby,
substantially increase the parasite population for the next H. zea generation. A lso,
surviving sterile larvae would produce sterile H. zea adults for the next generation.
I f the econom ic injury level o f cultivated host plants indicated that additional larvae
were undesirable, then the dose o f radiation administered to H. zea could be
increased to a level that would reduce or eliminate the number o f progeny from
irradiated females, or releases could be limited to irradiated males.
Although inherited sterility is compatible with synthetic organic insecticides,
parasites and/or predators are not generally compatible with these insecticides. If
insecticides are needed to reduce H. zea infestations, insect growth regulators, bio-
logicals, or other formulations compatible with natural enemies should be consid­
ered. An alternative management scheme could be to establish H. zea host plants in
insecticide free areas adjacent to insecticide treated crops. Host plants could be artifi­
cially infested with H. zea larvae to provide parasites (native and/or released) with
an adequate supply o f hosts. I f the H. zea larvae used in the artificial infestations
were sterile (progeny o f irradiated parents), then non-parasitized larvae would not
contribute to the increase o f the wild population, but would produce sterile H. zea
adults for the next generation.
Resistant plant varieties, if available, would play an important role in an
integrated control programme involving inherited sterility. Resistant varieties with
antibiosis as the m ode o f action would not only provide some plant protection, but
also would reduce the number o f H. zea entering the next generation. Therefore, in
an irradiated H. zea release programme, resistant varieties could effectively increase
the ratio o f irradiated to non-irradiated H. zea.

4. DISCUSSION

Employment o f an integrated management approach for H. zea will no doubt

require consideration o f numerous econom ic, ecological, behavioural and logistical
 IAEA-SM-327/35 369

factors. Currently, this knowledge is incomplete. H ow ever, the model presented in

Fig. 1 provides som e unique insights into how different control strategies could be
com bined for greatest efficiency. Although the effectiveness o f inherited sterility
continues to increase as the ratio o f irradiated to non-irradiated insects increases, the
efficiency decreases quickly once a 10:1 ratio has been obtained. A similar loss o f
efficiency occurs in the parasite release technique when the parasite to host ratio
increases above 5:1. A ccording to these m odels, the econom ic feasibility o f com bin­
ing inherited sterility and parasite release techniques would be greatest when the ratio
o f irradiated to non-irradiated H. zea is < 10:1 and the ratio o f parasite to host is
< 5 :1 . A ccess to reliable efficiency curves for all the control strategies involved in
an integrated control programme would facilitate the optimal use o f programme
funds.
The full potential o f inherited sterility as an area wide control strategy for
H. zea and other lepidopteran pests may be realized only when inherited sterility is
integrated with other suppression methods. Applications o f singular, non-insecticidal
methods for area wide suppression o f H. zea have had limited success and have been
vulnerable to temporary programme failures (e.g. inclement weather, reduced insect
rearing output, etc.). A n integrated approach would be horizontally diversified
through the use o f several strategies which used different m odes o f action and,
therefore, would be less vulnerable to temporary programme failures than would
a single strategy. Combining two or m ore control strategies with additive effects
may increase programme efficiency. H ow ever, the greatest benefit in combining
inherited sterility with other control strategies is the dynamic synergistic effects.
The demonstrated potential o f inherited sterility to reduce the reproductive ability o f
H. zea and the demonstrated capacity o f inherited sterility to perform compatibly and
synergistically with other control strategies suggest that inherited sterility should be
a major component o f future integrated approaches to managing H. zea.

REFERENCES

[1] L A C H A N C E , L.E., Genetic Methods for the Control of Lepidopteran Species,

Ser. 28, Agricultural Research Service, United States Department of Agriculture,
Washington, D C (1985).
[2] KNIPLING, E.F., Suppression of pest Lepidoptera by releasing partially sterile males:
A theoretical appraisal, BioScience 20 (1970) 465-470.
[3] C A R P E N T E R , J.E., Y O U N G , J.R., S P A R KS , A.N., C R O M R O Y , H.L.,
C H O W D H U R Y , M.A., Corn earworm (Lepidoptera: Noctuidae): Effects of sub-
sterilizing doses of radiation and inherited sterility on reproduction, J. Econ. Entomol.
80 (1987) 483-489.
[4] C A R P E N T E R , J.E., S P AR K S , A.N., PAIR, S.D., C R O M R O Y , H.L., Heliothis zea
(Lepidoptera: Noctuidae): Effects of radiation and inherited sterility on mating competi­
tiveness, J. Econ. Entomol. 82 (1989) 109-113.
370 CARPENTER

[5] C A R P E N T E R , J.E., S P A R KS , A.N., C R O M R O Y , H.L., C o m earworm (Lepidop­

tera: Noctuidae): Influence of irradiation and mating history on the mating propensity
of females, J. Econ. Entomol. 80 (1987) 1233-1237.
[6] C A R P E N T E R J.E., Y O U N G , J.R., C R O M R O Y , H.L., SPAR K S, A.N., C o m ear­
w o r m (Lepidoptera: Noctuidae): Comparison of field survival of larvae from normal
and irradiated parents, J. Econ. Entomol. 80 (1987) 883-886.
[7] C A R P E N T E R , J.E., G R OS S , H.R., Interaction of inherited sterility and diapaiise in
the c o m earworm (Lepidoptera: Noctuidae), J. Econ. Entomol. 82 (1989) 1354-1357.
[8] C A R P E N T E R , J.E., Effect of radiation dose on the incidence of visible chromosomal
aberrations in Helicoverpa zea (Lepidoptera: Noctuidae), Environ. Entomol. 20 (1991)
1457-1459.
[9] KNIPLING, E.F., The Potential Role of the Sterility Method for Insect Population
Control with Special Reference to Combining this Method with Conventional Methods,
Ser. 33-98, Agricultural Research Service, United States Department of Agriculture,
Washington, D C (1964).
[10] . C A R P E N T E R , J.E., Y O U N G , J.R., Interaction of inherited sterility and insecticide
resistance in the fall army w o r m (Lepidoptera: Noctuidae), J. Econ. Entomol. 84 (1991)
25-27.
[11] C A R P E N T E R , J.E.,W I S E M A N , B.R., Effects of inherited sterility and resistant dent-
c o m silks on Helicoverpa zea (Lepidoptera: Noctuidae) development, J. Entomol. Sci.
(in press).
[12] KNIPLING, E.F., The Basic Principles of Insect Population Suppression and Manage­
ment, Agricultural Handbook No. 512, United States Department of Agriculture,
Washington, D C (1979).
[13] KNIPLING, E.F., Principles of Insect Parasitism Analyzed From N e w Perspectives:
Practical Implications for Regulating Insect Populations by Biological Means, Agricul­
tural Handbook No. 693, United States Department of Agriculture, Washington, D C
(1992).
 IAEA-SM-327/36

EVALUATION OF THE Fx STERILITY

TECHNIQUE FOR POPULATION
SUPPRESSION OF THE PINK BOLLWORM,
Pectinophora gossypiella (SAUNDERS)

Z .A . QURESHI, T. HUSSAIN, N. A H M ED
Atom ic Ènergy Agricultural Research Centré,
Tandojam, Sind,
Pakistan

Abstract
É V A L U A T I O N O F T H E F, ST E RILITY T E C H N I Q U E F O R P O P U L A T I O N S U P P R E S ­
S I ON O F T H E P I N K B O L L W O R M , Pectinophora gossypiella (SAUNDERS).
Field cage studies were carried out to evaluate the F, sterility technique for population
suppression of the pink bollworm, Pectinophora gossypiella (Saunders),' in cotton. For this
purpose, six field cages (1.8 m x 1.8 m x 1.8 m) were placed over cotton seedlings in the
field.. Five pairs of laboratory reared, untreated adult moths were released in all six cages dur­
ing the first week of August. In addition, on the same date 100 pairs.of adults, following irradi­
ation of the mature pupae at 100 Gy, were released in two cages and 250 such pairs of
irradiated adults in two other cages in ratios of 20:1 and 50:1 (irradiated:normal), respec­
tively. The results of per cent larval infestation in the rosette blooms in different field cages
indicated that larval infestation in the flowers was significantly lower (4.93%) in cages where
the moths were released at the 50:1 ratio than in cages where moths were released at the 20:1
ratio (6.71%). Larval infestation in the flowers was significantly highest (21.87%) in the con­
trol cages. Similarly, larval infestation in the green bolls was significantly reduced to the
siibeconomic level (5.23%) in the 50:1 cages compared with the 20:1 cages (8.10%).
However, larval infestation in the control cages exceeded 20%. These preliminary studies
indicate the potential of the F, sterility technique for effective control of the pink bollworm.'

1. IN TR O D U C TIO N

The feasibility o f using induced sterility for the control o f numerous e co ­

nomically important lepidopteran pests has been investigated [1]. The pink
bollw orm , Pectinophora gossypiella (Saunders), demonstrates typical characteristic
lepidopteran responses to gamma radiation . [2, 3]. •. ■
The possibility o f achieving insect population suppression by introducing
irradiated insects o f the same species was suggested b y . Serebrovskij [4].
Subsequently, Proverbs [5] demonstrated that the F, progeny o f the irradiated
codling moth was sterile. Since then, a number o f investigators have reported the
same phenomenon in other lepidopteran pests [2, 3, 6]. The theoretical models o f

371
372 QURESHI et al.

Knipling [1] for population suppression o f Lepidoptera by releasing partially sterile

males were confirm ed by Flint et al. [7], who reported that pink bollworm moths
irradiated at 10 krad and released in field cages were more effective in population
suppression than moths treated with higher doses o f gamma radiation.1 Bartlett and
Butler [8] found that the F! female progeny o f irradiated females (15 and 20 krad)
paired with untreated males were more than 90% sterile. Furthermore, releases o f
either sex sterilized at 10, 15 or 20 krad or both sexes sterilized at 10 krad suppressed
the pink bollworm population from 82.6 to 99 .9 % over two generations. Henneberry
and Clayton [9] found that the number o f progeny produced by laboratory reared
moths exposed to 15 krad or higher doses and paired with untreated females was
reduced by over 80% compared with those produced by untreated pairs. The
reproduction rate o f male and female pink bollw orm moths from 10 or 15 krad
irradiated pupae and mated to untreated females or males was reduced by 88% or
m ore [10]. In lepidopterous species, partial sterility in the treated P, males is
associated with high levels o f sterility in the F j progeny. Thus, fairly low doses o f
gamma radiation should probably be administered to the released males in order to
achieve an effective combination o f induced partial sterility and conditional lethal
mutations capable o f suppressing a native populàtion and, simultaneously, o f chang­
ing its genotype. This paper reports on the evaluation o f the F] sterility technique
for population suppression o f the pink bollw orm in field cages.

2. M A TE RIALS A N D M ETH ODS

Field cage studies were carried out to evaluate the efficiency o f irradiated
moths and their F] progeny for population suppression o f the pink bollworm . For
this purpose, six field cages ( 1 . 8 m x 1 . 8 m x 1.8 m) were placed over cotton
seedlings in the field. The native population o f pink bollw orm , if any, from
diapausing larvae in the cages was monitored by gossyplure baited traps and by
inspecting the cotton fruiting bodies weekly. After confirming that there was no
infestation in the cages, five pairs o f laboratory reared, untreated adult moths were
released in all six cages during the first week o f August. In addition, on the same
date 100 pairs o f adults, follow ing irradiation o f the mature pupae at 100 G y, were
released in two cages and 250 such pairs o f irradiated adults in two other cages at
ratios o f 20:1 and 50:1 (irradiated:normal), respectively. The remaining two cages
were kept as the control, with no irradiated moths. Subsequently, three releases o f
moths at the above mentioned ratios were made in the respective cages at three week
intérvals. The observations made on the rosette bloom s and larval infestation in the
green bolls were recorded at weekly intervals.

1 1 rad = 1.00 x 10"2 Gy.

 IAEA-SM-327/36 373

TABLE I. M E A N PER CEN T L A R V A L IN FESTA­

TION OF PINK B O L LW O R M IN THE FLOW ERS
AND GREEN BOLLS OF C O T TO N IN FIELD
CAGES

Release ratio Per cent larval infestation in

Normal :irradiated Flowers:green bolls

1:50 4.93b:5.24b
1:20 6.71b:8.57b
Control 21.87a:21.35a

Note: Means with the same letters are statistically non-

significant (P = 0.05).

3. RESULTS

3.1. Flower infestation (rosette bloom)

The results o f the pier cent larval infestation in rosette bloom s in different field
cages (Table I) indicated that larval infestation in the flowers was significantly lower
(4 .9 3 % ) in cages where the moths were released at the 50:1 ratio compared with
cages where they were released at the 20:1 ratio (6 .7 1 % ). Larval infestation in the
flowers was significantly the highest (2 1 .8 7 % ) in the control cages.
The results o f the larval population buildup in the different field cages, as
recorded for flowers by inspecting the rosette bloom s (Fig. 1), showed that the first
rosette bloom was recorded two weeks after the release o f normal moths in the con­
trol cages. How ever, rosette bloom s appeared after the third week o f release in cages
where the ratio o f irradiated . normal moths was maintained at 20:1 and 50:1.
Although the trend o f larval population buildup in the flowers was almost identical
in all the cages, the releases o f irradiated moths resulted in reduced moth population
development in the 20:1 and 50:1 cages. How ever, larval infestation in the 50:1
cages was significantly lower when compared with the rest o f the field cages.

3.2. Boll infestation

The data on larval infestation in green bolls in different field cages (Table I)
indicated that infestation was significantly reduced, to the su becon om ic, level
(5 .2 3 % ), in cages where a release ratio o f 50:1 moths, was maintained as compared
374 QURESHI et al.

©•— « — -& Control

Weeks after release

FIG. 1. Pink bollworm infestation in cotton flowers.

with the cages with a release ratio o f 20:1 moths (8 .1 0 % ). How ever, larval infesta­
tion in green bolls was significantly the highest (2 1 .9 8 % ) in the control check cages.
The results o f larval population buildup in mature green bolls in different cages
(Fig. 2) indicated that in cages receiving irradiated moths, follow ing irradiation o f
mature pupae, larval infestation appeared three weeks after the first release, whereas
in the control cages the infestation appeared after two weeks. The increase in larval
infestation was rapid in the control cages, reaching its peak seven weeks after the
first release o f moths. In the cages where a release ratio o f 20:1 was maintained, lar­
val infestation in green bolls was significantly low er than in the control cages.
Although the larval infestation trend in the green bolls after three releases o f moths
 IAEA-SM-327/36 375

©— в---------------------------- в Control
£ ъ ъ. 1:20 (Normal:¡rradiated)
в— в— в 1:50 (Normal:irradiated)

Weeks after release

FIG. 2. Pink bollworm infestation in the green bolls of cotton.

was almost identical in all the cages, larval infestation in the 50:1 cages was
significantly lower than in the 20:1 cages.

4. DISCUSSION

The greatest disadvantages o f using radiation sterilized male moths to control

lepidopteran populations have been the costs o f rearing and the failure o f the irradi­
ated males to compete successfully with the wild males [11]. How ever, data on
delayed larval infestation in field cages receiving irradiated moths suggest that the
irradiated male and female pink bollw orm moths were competitive in mating with
376 QURESHI et al.

non-irradiated, laboratory reared pink bollworm moths. Furthermore, male moths

treated with 100 Gy mated effectively and prevented the females from remating. The
males and females that emerged from the irradiated pupae elicited low er oviposition
responses when mated with either untreated or treated males. It could, therefore, be
inferred that radiation sterility in moths that emerged from the irradiated pupae was
comparable to F I sterility [2, 3, 9, 12].
The release o f partially sterilized moths at a ratio o f 50:1 at three week
intervals reduced larval infestation to the subeconom ic level (5 %) compared with the
cages where a release ratio o f 20:1 was maintained (9 % ). In the control cages, larval
infestation exceeded 2 0 % . The relatively reduced infestation in bolls in field cages
demonstrated that the number o f progeny from partially sterile moths was limited and
failed to cause significant damage to the crop. These results are in close conformity
with those o f Bariola et al. [13] and Flint et al. [7]. The reduced larval infestation
in both the flowers and the bolls indicated the additive effect o f induced and inherent
sterility in pink bollworm moths. Therefore, the releases o f irradiated moths should
be carried out more frequently than at three week intervals to ensure that sexually
active sterile or partially sterile males are always present to mate with the native
females. These preliminary studies indicate the potential o f the F ! sterility technique
for the effective control o f the pink bollworm . However, further detailed investiga­
tions are needed to evaluate the efficiency o f the F, sterility technique under field
conditions.

REFERENCES

[1] KNIPLING, E.F., Suppression of pest Lepidoptera by releasing partially sterile males,
Bioscience 20 (1970) 465-470.
[2] C H E N G , W.Y., N O R T H , D.T., Inherited sterility in the F, progeny of irradiated
male pink bollworms, J. Econ. Entomol. 65 (1972) 1272-75.
[3] G R A H A M , H.M., O U Y E , M.T., G A R C I A , R.D., D E L A R O S S A , H.H., Dosage of
g a m m a irradiation for full and inherited sterility in adult pink bollworms, J. Econ.
Entomol. 65 (1972) 645-650.
[4] SEREBRO V SK I J , A.S., O n the possibility of a new method for the control of insect
pests, Zool. Zh. 19 (1940) 618-630 (in Russian).
[5] P R O V E R B S , M.D., Progress on the use of induced sexual sterility for the control of
codling moth, Proc. Entomol. Soc. Ont. 92 (1962) 5-11.
[6] N O R T H , D.T., H O L T , G.G., Inherited sterility in progeny of irradiated male cabbage
looper, J. Econ. Entomol. 61 (1968) 928-931.
[7] FLINT, H.M., P A L M E R , D.L., B A R I O L A , L.A., H O R N , B., Suppression of popu­
lation of native pink bollworm in field cages by the release of irradiated moths, J. Econ.
Entomol. 67 (1974) 55.
[8] B A R T L E T T , A.C., B U T L E R , G.D., Jr., A model of pink bollworm populations and
its use for evaluating the effects of radiation, Southwest Entomol. 4 (1979) 216.
 IAEA-SM-327/36 377

[9] H E N N E B E R R Y , T.J., C L A Y T O N , T.E., Effects on reproduction of g a m m a irradi­

ated, laboratory reared pink bollworms and their F, progeny after matings with
untreated laboratory reared or native insects, J. Econ. Entomol. 74 (1981) 19.
[10] H E N N E B E R R Y , T.J., C L A Y T O N , T.E., Effects of g a m m a radiation on pink boll­
w o r m (Lepidoptera: Gelechiidae) pupae: Adult emergence, reproduction, mating and
longevity of emerged adults and their F, progeny, J. Econ. Entomol. 81 (1988) 322.
[11] N O R T H , D.T., H O L T , G.G., Population control of Lepidoptera: The genetic and
physiological basis, Manit. Entomol. 4 (1970) 53.
[12] L A C H A N C E , L.E., BELL, R.A., R I C H A R D , R.D., Effect of low doses of g a m m a
irradiation on reproduction of male pink bollworms and their F, progeny, Environ.
Entomol. 2 (1973) 653.
[13] B A R I O L A , L.A., B A R T L E T T , A.C., S T A T E N , R.T., R O S A N D E R , R.W.,
K E L L E R , J.C., Partially sterilized adult pink bollworms: Releases in cages and field
cause chromosomal aberrations, Environ. Entomol. 2 (1973) 173.
 IAEA-SM-327/37

EVALUATION OF THE
POTENTIAL CONTROL OF
THE EUROPEAN CORN BORER
(Ostrinia nubilalis Hb.) IN THE FIELD
BY RADIATION INDUCED Fx STERILITY
I.L R O SC A , A l. BARBU LESCU
Research Institute for Cereal and Industrial Crops,
Fundulea,
Romania

Abstract
EVALUATION OF T H E POTENTIAL C O N T R O L OF THE E U R O P E A N C O R N BORER
(Ostrinia nubilalis Hb.) IN T H E F I E L D B Y R A D I A T I O N I N D U C E D F! STERILITY.
The European corn borer (Ostrinia nubilalis Hb.) is considered to be the most important
pest of maize in Romania after panicle emergence. It covers the entire cropping area of the
country. Data collected over a five year period have indicated an average of 4 4 % plants
infested, 1.1 larvae per plant, 23 180 larvae/ha and 550 kg/ha or 7.5% losses in yield.
Depending on the geographical area, the climatic conditions, the structure of the culture and
the corn hybrids cultivated in spring, the E C B population can theoretically be estimated at
263-5845 larvae/ha. Of a total area of 3 x 106 ha c o m cultivated in Romania, it can reach
levels of between 789 and 17 535 x 106 larvae. A mass rearing technique suitable for
Romanian conditions has been developed. In Romania, the European c o m borer is attacked
by many natural enemies throughout its life; the role of parasites and predators is presented
in the paper. It is n o w possible to apply biological control by Triçhogramma spp. but the
results differ depending on the region in which the control is applied. Sex pheromone formula­
tion developed in Romania is less efficient, but it can be used to record the flight of the c o m
borer males. Control of the flies in small areas (1-2 ha) was attempted by mating disruption
and mass capture of males in sticky traps. The results show that under these conditions it is
not possible to control the pest. Experiments performed in field cages have demonstrated a
theoretical possibility of controlling the pest by releases of flies having inherited F, sterility.
Our data have demonstrated that the percentage of stems attacked by the pest and the number
of overwintering larvae per stem were correláted with the hatching percentage. Releases of
the F, generation with inherited sterility, in field cages, at different ratio of males with
inherited F, sterility to wild males demonstrate that this method of control can be used as
part of an integrated control system.

379
380 ROÇCA and BARBULESCU

1. INTRODUCTION

The European corn borer (Ostrinia nubilalis H b.) (ECB) is one o f the most
damaging pests o f maize. It covers an extensive area in almost all parts o f the world,
especially in the Northern Hemisphere, and is able to thrive in a wide range o f
climates, from equatorial to temperate zones. Its particular econom ic significance
has determined the intensive research being carried out on this insect, especially in
North America.
In Romania, ECB is most harmful after panicle emergence, and is encountered
in all corn culture areas. The losses induced have risen to 40% o f the grain yield [1].
Data collected over a five-year period indicated averages o f 44% attacked plants,
1.1 larvae per plant, 23 180 larvae/ha and 550 kg/ha (or 7 .5 % ) loss in yield [2]
(Table I).
Under the ecological conditions prevailing in Romania, the insect develops one
generation per year, except for areas in the south, where a second partial generation
has been recorded with a population that is less than 20% that o f the previous genera­
tion. The mass flight o f the overwintering generation o f moths com m only takes place
a few days before panicle emergence, the main attack taking place in tunnels inside
the stems. The attack on c o m is econom ically important, but the pest also feeds on
hemp and sorghum crops, as well as on various species o f spontaneous flora.
Because o f the particular significance o f ECB for the corn crop, a series o f
investigations has been undertaken in Romania with a view to preventing outbreaks
o f this pest using chemicals [3], biological means [4 -8 ] and resistant hybrids [9 -1 5 ].
In recent years, particular attention has been paid to the study o f synthetic sex
pheromones [16—19] and, since 1988, investigations have been carried out on male
sterilization using radiation [2 0 -2 3 ].

TA B L E I. IM PO R TAN CE OF EUROPEAN C O R N BORER FO R M A IZ E CROPS

IN R O M A N IA

Limiting values Average

Plants attacked 13.0-70.0% 44.0%

Larvae/plant 0 . 2- 2.2 1.1
Larvae/ha 2630-58 455 23 180
Yield reduction/ha 80-1326 kg 550 kg
Loss/larva reaching diapause 13.9-57.0 g 34.0 g
Increase in tunnelling in the stalk — stalk breakage
— incidence of stalk rot
 IAEA-SM-327/37 381

2. LIFE H ISTO R Y

The life history o f ECB has been studied in detail by many workers and its
biology is well known. The developmental rates are influenced by many factors,
including temperature and moisture.

3. M IG R A TIO N

The ECB is a poor flyer and until now large migrations have not been reported.
The moths leave the emergence sites in corn field debris and fly to places with dense
vegetation, usually grasses, where a g ood microclimate exists for feeding, resting
and mating.
After mating, the female moth leaves the mating site and deposits her eggs in
corn fields. This movement is possibly repeated. N ot enough experimental data are
available on the flying capacity o f the moths.

4. POPULATIO N

Depending on the geographical area, the climatic conditions, the structure o f

the cultures and the corn hybrids cultivated in spring, the ECB population has
theoretically been estimated at 2 6 3-5845 larvae/ha (on average, 2318). In all co m
cultivated areas (3 X 106 ha) it can rise to 7 8 9 -1 7 535 X 106/larvae. Such large
populations can be reduced by weather conditions and soil tillage.

5. M A SS R EA RIN G OF ECB

Since all investigàtions aimed at preventing ECB attack are conditioned by the
availability o f large numbers o f insects, mass rearing o f this insect is essential.
Consequently, a series o f experiments were performed on various diets, using
bean meal as the basic ingredient because it affords satisfactory growth o f the ECB
and because the evolution o f ECB is not influenced negatively by any o f the ingre­
dients added. The diet ingredients shown in Table II are o f practical importance
because they were used in all the investigations carried out on male sterilization.
These ingredients are com m on, readily available and inexpensive. Depending on the
extent o f the rearing facilities, the cost o f 1000 pupae ranged from 751 to 4130 lei
(US $1 = lei 430 (Oct. 1992)) (Table Ш ).
When evaluating the performance o f O. nubilalis moths reared for a varying
number o f generations on an artificial diet, it was found that the fecundity and
longevity values did not exhibit significant differences between generations. The
382 ROÇCA and BARBULESCU

TAB LE II. INGREDIENTS USED IN THE DIET FOR M ASS A R T IFICIAL

REARIN G OF Ostrinia nubilalis Hb (O N E B A T C H = 4500 kg)

Ingredient Amount %

Bean meal 372.0 g 8.20

Wheat bran 160.0 g 3.53
Brewers’yeast 136.0 g 3.00
Milk powder substitute for calves ... 106.0 g 2.34
Salt mixture for poultry 40.0 g 0.88
Sugar 133.0 g 2.93
Ascorbic acid 13.6 g 0.30
Sorbic acid . . 10.0 g 0.22
Acetic acid glacial 14.8 m L .0.33
Formaldehyde 8.4 m L 0.19
Agar 40.0 g 0.88
Water 3500.0 m L 77.20

results o f experiments carried out on generations 81 and 106 were similar to those
obtained iñ generations 8 and 9, respectively [24].
It is worth mentioning that mass rearing o f ECB has been running for 161
successive generations and is still in progress, unlike individual rearing, which
will be discontinued after nine successive generations because o f a degeneration in
induced sterility.
A mass rearing technique has been developed that fits the Romanian
conditions. On the basis o f this technique, using a diet com posed essentially o f
indigenous ingredients, insects can be reared continuously, so satisfying the
increasing requirements.
Since sufficient biological material is available, it was possible to perform
experiments on the sterilization o f O. nubilalis malts by irradiation.

6. D A M A G E LEVELS

Corn plants d o not yield well as a result o f ECB damage. The damage resulting
from leaf feeding is o f no relevance, but stalk tunnelling is o f great importance. In
Romania, the reduction in yield ranges between 80 and 1326 kg/ha (on average,
550 kg/ha).
 IAEA-SM-327/37 383

TA B L E III. CO ST OF ECB (Ostrinia nubilalis H b.) REA RE D A T F U N D U L E A ,

R O M A N IA (US $1 = lei 430 (O C T . 1992))

12 000 pupae/ 300 000 pupae/

month month
(lei) (lei)

Artificial diet 4 230 105 750

Wages 35 100 70 200
General expenses 8 115 39 145
C o m m o n expenses 2 110 10 185

Total 49 555 225 280

Cost/1000 pupae 4 130 (US $9.6) 751 (US $1.75)

7. C O N TR O L

Integrated control o f this pest is now being explored. New pest management
strategies should include the econom ic threshold levels, regular monitoring surveys,
biological control opportunities, etc.

8. C H E M IC A L C O N T R O L

Because the mass flight o f moths usually takes place a few days before panicle
emergence, the c o m is generally too tall when the decision is made to control ECB.
Equipment availability and the time required to complete application over large areas
often necessitate use o f aerial treatment, but this is very expensive. For this reason,
no chemical control o f ECB is carried out in Romania, as a result o f which no
insecticide resistance exists.

9. N A T U R A L C O N T R O L B Y B IO L O G IC A L AGENTS

In Romania, ECB is attacked by many natural enemies throughout its life. The
parasites and predators are presented in Table IV and Table V shows the most im por­
tant biological factors affecting populations o f O. nubilalis Hb.
Biological control o f ECB by Trichogramma spp. is theoretically possible in
Romania, but Table V I shows that control differs, depending on the region in which
384 ROSCA and BARBULESCU

T A B L E TV. PARASITES A N D PREDATORS OF Ostrinia hubilalis Hb. IN

R O M A N IA

Host stage Family No. of species Importance

Egg T richogrammatidae 4 0; + + +
Anthochoridae 2 0; +
Nabidae 4 0; +
Chrrysopidae 2 0; +

Larva Chalcididae 8 0; +
Ichneumonidae 14 + ;+ +
Tachinidae 5 + ;+ +
Chloropidae 8 0; +

Pupa Ichneumonidae 1 0

Note: 0: Unimportant.
+: O f small account.
+ + : Important.
+ + + : Of the first importance.

TA B L E V . PRIN CIPAL B IO LO G ICA L FAC TO R S AFFECTIN G POPULATIONS

OF Ostrinia nubilalis Hb. IN R O M A N IA

Affected stage of Limiting value Average

Biological factor
host (%) (%)

Trichogramma spp. Egg 0-75.0 7.8

Lydella thomsoni Hert. Larva 0-15.6 4.6
Sinophorus crassifemur Thom. Larva 0-5.0 1.8
Fungi Larva 13-47.2 31.6
Bacteria Larva 11.7-35.2 19.1
 IAEA-SM-327/37 385

TA B L E VI. C O N T R O L OF Ostrinia nubilalis Hb. B Y Trichogratnma maidis PINT

IN F U N D U L E A A N D T U R D A

Fundulea

Five weekly treatments

Control
(200 000/ha)

1989 1990 1991 1989 1990 1991

% attacked plants 49.7 28.1 57.5 37.9 28.0 40.8

Larvae/plant 0.61 0.21 1.1 0.52 0.19 0.8

Turda

_ , One treatment T w o treatments

(200 000/ha) (200 000/ha)

1988 1989 1988 1989 1988 1989

% attacked plants 93.3 74.6 49.0 24.1 34.3 10.1

it is applied. Trichogramma species has weak flying and searching activity and a
short period o f life. For these reasons, it is necessary to repeat the releases in corn
fields (7 m -7 m ) and no results are achieved when the temperature is over 3 1 °C in
the fields. The parasites must be applied manually; this type o f control is expensive
and the results are not precise.

10. P L A N T RESISTANCE

Selections o f inbred corn lines resistant to ECB have been made in Romania
over the past tw o decades. Today, many types o f hybrid are available that show an
intermediate resistance tolerance to O. nubilalis.
' At the Research Institute for Cereals and Industrial Crops at Fundulea, there
is an extensive breeding project in which germplasm resistant to ECB is used as
genetic stock for breeding purposes.
386 ROÇCA and BARBULESCU

11. PHEROMONES

Components o f various ratios and amounts o f the synthetic ECB sex phero­
mone were evaluated in the field. The results showed a relatively high number o f
males captured in the sticky traps throughout the flying period; this varied with the
location and the pheromone variant used. It is worth mentioning that in Romania both
cis (Z) and trans (E) pherotypes o f this species are found, but the latter are missing
in the central and northeastern parts o f the country.
The pherotype trans (E) is generally found in less numerous populations, but
the problem o f ECB in Romania lies in the cis (Z) pherotype.
Even though the synthetic sex pheromone formulation developed in Romania
is less efficient, it can be used to record the flight o f ECB males; thus flight curves
can be drawn for generations 1 and 2 (Fig. 1).
W e should stress that in Romania the first generation is particularly important,
occurring in the second half o f June; the second generation, specific to the southern
part o f the country, appears in August and is practically devoid o f damage potential.
There are a series o f theoretical models for using sex attractants for insect
control. In 1990 and 1991 at Fundulea, we tried to control ECB in small areas
( 1-2 ha) isolated in the forest by mating disruption and mass capture o f males in
sticky traps.
The results presented in Tables VII and VIII showed that it is not possible to
control ECB in small areas because we presume that mating o f the females did not
take place in the corn fields.
Sex pheromones are now used in ecological studies, mainly for migration and
evaluation or monitoring o f pest populations.

12. LAB ELLIN G A D U LTS

At present, ‘ Calco Red D y e ’ is used to investigate pest dynamics and to esti­

mate natural populations. This marker behaved very well in our trials by labelling
adults obtained from larval rearing on a regular diet to which the dye had been added.

13. USE OF IO NIZING R A D IA T IO N FO R PA R T IA L

STERILIZATION OF ECB

In an attempt to develop the fundamentals o f genetic control through F, steril­

ity as a component o f an integrated control programme for ECB at the Research
Institute for Cereals and Industrial Plants at Fundulea, investigations related to the
sterile insect release technique have been carried out within the framework o f an
IA E A Co-ordinated Research Programme, ‘ Radiation Induced F| Sterility in
Lepidoptera for Area-W ide Control’ .
 IAEA-SM-327/37 387

— Pherotype cis(Z)
■■■ Pherotype trans (E)
No. males/trap

Fundulea 1991

Fundulea 1990

Turda 1991
No.’ males/trap

Turda 1990

FIG. 1. Flight curves o f the ECB, registered by number o f males captured in sticky traps.
388 RO§CA and BARBULESCU

T A B L E VII. C O N T R O L O F Ostrinia nubilalis Hb. — M A S S T R A P P I N G ( M T )

No. of males/trap Attacked plants

Larvae/plant
(total males) (%)
Yeai
Control MTa Control MTa Control MTa

1990 30.2 11.4 23.3 22.8 0.17 0.23

(183)

1991 50.0 18.8 62.0 55.3 0.55 0.47

(301)

a Sixteen sticky traps/ha.

Note: Numerals in parentheses denote total number of males captured per year.

T A B L E VIH. C O N T R O L O F Ostrinia nubilalis Hb. — M A T I N G D I S R U P T I O N

Attacked plants
Larvae/plant
(%)
icai
Control M D a Control M D a

1990 18.8 18.0 0.2 0.25

1991 33.4 30.6 0.45 0.5

a One hundred lures/ha.

Application of this technique requires knowledge of the method, radiation dose

and biological parameters of the specimens being exposed to radiation in order to
ensure the required degree of substerility of the pest with the m i n i m u m detrimental
effect on the irradiated insects.
T h e insects used in experiments were derived from a laboratory strain and have
been reared according to individual and mass rearing techniques elaborated at
Fundulea.
Preliminary trials have established the role of pupae age and radiation dose on
the emergence rate of moths.
T h e experiment with irradiated pupae revealed that hatching was dependent on
their age at treatment and the radiation dose: the older the pupae, the less affected
 IAEA-SM-327/37 389

their hatching in comparison with the unirradiated check pupae. It should be noted
that adult emergence was negatively influenced in direct proportion to the radiation
dose [21 ].
As a result o f these preliminary trials, subsequent investigations used 6-d-old
pupae, from which the adults emerged Within the next 48 h.
The two sexes had different levels o f sterility at the same radiation dose; males
were more resistant to radiation than females, which had a higher degree o f sensitiv­
ity. Complete sterility o f the females was obtained with 30 krad, a d o se 1 at which
the males had only 86.5% sterility, i.e. partial sterility [22].
It is well known that pupal sexing is particularly difficult. Therefore, for sterile
insect releases it is necessary to irradiate both the male and female pupae together.
It is necessary to use partially sterile males because their competitiveness is close to
that o f hormale ones, on one hand, and completely sterile females, unable to generate
larvae which subsequently would attack corn plants, on the other. T o increase the
degree o f sterility at a lower rate o f radiation, the F, generation had to be used for
releases instead o f the P generation.
T o decrease the rate o f radiation, the P male genitors were irradiated with 10
or 15 krad as 6-d-old pupae. The males that emerged were crossed with normal
females. The offspring o f these crossings were inbred or backcrossed; F! observa­
tions were carried out on the number o f eggs/fem ale, the percentage egg hatch, the
adult lifespan and the percentage sterile couples.
The data in Tables IX and X showed no significant decrease in the number o f
eggs laid by one female in the P, F[ or F 2 generations after irradiation with 10 and
15 krad. Generally, the number o f eggs was a little lower when adults from the
P0 I X N were inbred.
W e failed to obtain a sufficiently large population to test in F 2 populations
from F] I x- I because hatching rose to 25 .4 % at 10 krad and 9 .1 % at 15krad;
however, the larvae obtained did not reach the adult stage.
In all the combinations tested in F j, hatching decreased with the increasing
radiation dose. The hatching percentage in Fj was lower in the three variants from
P0 I x N, when compared with the parent generation. Egg viability in F , was
significantly reduced in comparison with the control for both the doses applied,
indicating the existence o f recessive lethal genes within these populations. In F2, a
decrease in the percentage egg hatching was also noted in all the combinations tested,
indicating the inheritance o f harmful effects induced by males irradiated in the parent
generation. It is also remarkable that the variants analysed for F2, although more
fertile than those in F 1; still exhibited a noticeable effect o f lethalgenes. No
significant reduction in the adult lifespan was noted in P, F, or F2.
The percentage sterile couples increased in descendences derived from
irradiated males, reaching the highest values in inbred variants.'

1 1 rad = 1.00 X IQ'2 Gy.

390 ROÇCA and BARBULESCU

TAB LE IX . TRE A T M E N T OF Ostrinia nubilalis Hb. M A LE S W ITH A

SUBSTERILIZING DOSE OF 10 krad

Cross variant Longevity of adults

Egg masses/ Hatchability, Sterile
Generation
. Male Female female (%) Male Female pairs (%)

Po N x N(C) 7.6 90.5 8.9 7.7 4.2

I X N 7.7 67.5 7.5 .6.7 6.7

F. N x N (C) 7.4 88.0 6.9 7.0 5.7

A. x N 7.6 . 53.6 8.1 8.3 8.5
■N x A . 7.2 59.2 7.2 7.8. 3.2
A x A 6.0 25.4 6.1 6.8 8.2

f 2 N x N (C) 6.8 85.3 8.7 7.9 6.8

N x (A x; N) 7.2 70.1 8.2 8.1 8.1
N x (N x A) 6.9 80Л 8.3 7.8 8.4
(A x N) x N 6.7 64.0 7.9 8.0 8.9
(A X N) x (N X A) 7.1 45.2 8.1 7.6 7.1
(A x N) x (A x N) 6.5 38.3 7.8 7.9 12.8
(N X A) x N 7.3 68.4 8.5 8.1 7.5
(N x A) x (A x N) 7.0 40.3 8.7 8.2 8.6
(N x A) x (N x A) 6.7 32.8 8.1 8.2 14.3

N: non-irradiated or normal; I: irradiated; A: progeny from irradiated male x normal female

(I x N); C: control. . .

This experiment showed that the F! sterility obtained at 10 krad was lower
than that at 15 krad; consequently, for field cage releases 15 krad were used.
Experiments performed in 1990-1991 and in 1992, and carried out in
3 m x 2 m x 2 m field cages, have demonstrated a theoretical possibility for
controlling the pest by the release o f insects with an inherited F! sterility (Tables X I
and XII).
 IAEA-SM-327/37 391

TA B L E X . T R E A T M E N T OF Ostrinia nubilalis Hb. M A LE S W ITH A

SU BSTERILIZIN G DOSE OF 15 krad

Longevity of adults
Cross-variant
(d)
Egg masses/ Hatchability Sterile
Generation
Male Female female (%) Male Female pairs (%)

P N x N (C) 7.6 90.5 8.9 7.7 4.2

I x N 8.1 42.3 6.9 7.5 10.2

F, N x N (C) 7.4 88.0 7.5 7.0 5.7

A x N 7.5 42.8 7.2 7.5 18.7
N x A 8.0 40.7 8.2 6.7 15.2
A x A 7.1 9.1 6.0 5.9 . 22.5

f2 N x N (C) 7.7 88.7 7:4 7.2 4.3

N x (A x N) 6.9 60.3 8.5 7.9 10.5
N x (N x A) 7.3 64.7 7.7 .7.8 11.2
(A x N) x N 8.0 55.9 8.2 8.0 10.1
A) 7.5 59.2 8.1 8.0 12.3
Z
<

Z
X

X
X

(A x N) x (A x N) 7.8 33.5 7.9 7.3 18.7

(N x A) x N 6.9 60.8 8.9 7.9 8.0
(N x A) x (A x N) 8.7 53.2 7.5 7.8 10.8
(N.x A) x (N x A) 7.3 29.9 7.8 7.3 21.1

N: normal; I: irradiated; A: progeny from irradiated male x normal female (I x.N).

The data in these tables show no significant reduction in the number o f eggs
laid by irradiated females, except for variant F t X F ,. The hatching percentage was
lower in all the variants using Fj moths; this effect still persisted in F 2 and
disappeared in F4.
The percentage attacked stems and the number o f overwintering larvae per
stem were correlated with the hatching percentage.
392 ROÇCA and BARBULESCU

TAB LE X I. FIELD C A G E RELEASE TESTS - 1990

Cross-variant variant No. egg Egg Stems No. Larvae/

batches/ hatch damaged larvae/ ha
Male Female female (%) (%) stem (%)

Isolated Control 2.48 42.5 76 1.92 100.00

F,* N 2.7 20.3 4 0.04 2.08
N F,* 2.6 19.5 8 0.08 4.17
F,* F,* 1.9 3.7 0 0 0
F 2* ' N 2.8 25.7 36 0.6 31.25
f 2* f 2* 2.3 14.2 16 0.28 14.58
F* F4* 2.5 40.3 80 1.8 93.75

* : progeny from irradiated male (15 krad) x normal female; N : normal.

T A B L E XII. F I E L D C A G E R E L E A S E T E S T S - 1991

Cross-variant variant No. egg Egg Stems No. Larvae/

batches/ hatch damaged larvae/ ha
Male Female female (%) (%) stem (%)

Isolated Control 3.4 60.5 84 1.88 100.00

F,* N 3.5 28.7 8 0.12 6.3
N F,* 3.2 28.0 16 0.24 12.8
F,* F,* 2.5- 4.1 0 0 0
12F,* + 12N N 3.7 99.5 68 1.16 61.7
16F,*+ 8 N N 3.4 47.4 44 0.72 38.3
18F,*+ 6 N N 3.5 20.5 32 0.6 31.9
20F,*+ 5 N N 3.0 8.7 16 0.44 23.4

* : progeny from irradiated male (15 krad) x normal female; N : normal.

Releases o f the ECB F, generation in field cages demonstrated that F! sterility

can be used as a part o f an integrated control system along with other means o f
control.
 IAEA-SM-327/37 393

REFERENCES

[1] P A U L I A N , FL., B A R B U L E S C U , AL., M U S T E A , D., B E L U , V., PEIU, М., A con­

tribution to the knowledge of the biology and control of maize borer (Pyrausta nubilalis
Hb.) in the RPR, Ann. ICCA-Bucharest, Ser. В 29 (1961) 397-420.
[2] P A U L I A N , FL., et al., The evolution of European Corn Borer (Ostrinia nubilalis Hb.)
and damaging potential recorded between 1971-1975, R.S. Romania, Probl. Prot.
Plant. 3 (1976) 23-48.
[3] V O I N E S C U , I., B A R B U L E S C U , AL., Efficacy of granular insecticides in control of
corn borer (Ostrinia nubilalis Hb.), Ann. ICCPT-Fundulea 53 (1986) 383-386.
[4] G A L A N I , C., V O I N E S C U , I,, B A R B U L E S C U , AL., Efficacy of some micro­
biological products based on Bacillus thuringiensis1Berliner in control of corn borer
(Ostrinia nubilalis Hb.), Ann. ICCA-Bucharest, 16 (1979) 235-241.
[5] RO§ CA, I., B A R B U L E S C U , AL., Numerical limiting by biologic factors of c o m
borer (Ostrinia nubilalis Hb.) in Romania, St. Cere. Biol., Ser. Biol. Anim. 35 (1983)
32-35.
[6] RO§ CA, I., B A R B U L E S C U , AL., Attempts on biological testing of Bacillus thurin­
giensis products with Ostrinia nubilalis Hb., as a test species, Probl. Prot. Plant. 14
(1986) 133-140.
[7] RO§ CA, I.,B A R B U L E S C U , A L . , V O N I C A , I., “Preliminary dates regarding role of
biological factors in reducing populations of Ostrinia nubilalis Hb. in R.S. Romania”,
paper presented at 8th National Conf. on Plant Protection, Iasi, 1983.
[8] R O S C A , I., B A R B U L E S C U , AL., PISICA, C., V O N I C A , I., Spreading in Romania
and role of species Sinophorus crassifemur Thoms, and Lydella thomsoni Hert.,
parasite on grubs of Ostrinia nubilalis Hb., St. Cere. Biol., Ser. Biol. Anim. 36
(1984) 92-95.
[9] B A R B U L E S C U , AL., Studies conducted at Fundulea on maize resistance to Ostrinia
nubilalis Hb., Probl. Prot. Plant. 9 (1981) 373-380.
[10] B A R B U L E S C U , AL., S A R C A , T., The behaviour of sweet corn lines and hybrids to
the attack of Ostrinia nubilalis Hb., Probl. Prot. Plant. 8 (1980) 150-157.
[11] B A R B U L E S C U , AL., S A R C A T., Testing of some hybrid combinations between
maize lines tolerant to the European Corn Borer (Ostrinia nubilalis Hb.), Probl. Prot.
Plant. 11 (1983) 5-9.
[12] B A R B U L E S C U , AL., C O S M I N , O., S A R C A , T., BICA, N., N E G U T , C., Testing
for resistance to Ostrinia nubilalis Hb. of some maize lines (1975-1978), Probl. Prot.
Plant. 9 (1981) 36-38.
[13] B A R B U L E S C U , AL., C O S M I N , O., S A R C A , T., BICA, N., N E G U T , C., Testing
for resistance to Ostrinia nubilalis Hb. of some maize lines (1979-1981), Probl. Prot.
Plant. 10 (1982) 93-97.
[14] B A R B U L E S C U , AL., C O S M I N , O., N e w dates regarding behaviour of corn lines and
hybrids to the attacks of the borer (Ostrinia nubilalis Hb.), Ann. ICCPT-Fundulea 46
(1980) 341-346.
[15] B A R B U L E S C U , AL., C O S M I N , O., Maize inbred lines with some degree of
resistance to Ostrinia nubilalis Hb., Probl. Prot. Plant. 15 (1987) 301-306.
394 ROÇCA and BARBULESCU

[16] RO§ CA, I., “Research regarding sexual synthesis pheromones of moths injurious of
field crops”, paper presented at 8th National Symp. on Environmental Protection,
Constanfa, 1990.
[17] RO§ CA, I., et al., “Possibilities of using synthetic sexual pheromones in protection
of cereal and technical crop cultures”, paper presented at 9th National Conf. on Plant
Protection, Bucharest, 1985.
[18] R O Ç C A , I.,et al., Results and prospects in utilization of sex pheromones in cereals and
technical plants from Romania, Probl. Prot. Plant. 17 (1989) 297-312.
[19] RO§ CA, I., et al., Research on the behaviour of Ostrinia nubilalis by the use of phero­
mone technique, Rev. Roum. Biol. 35 2 (1990) 105-115.
[20] B A R B U L E S C U , AL., RO§ CA, I., Results Regarding Sterilization Through Irradiation
of Male Ostrinia nubilalis Hb., an Important Pest of C o m in Romania, report, Joint
F A O / I A E A Division of Nuclear Techniques in Food and Agriculture, IAEA, Vienna.
[21] RO§ CA, I., B A R B U L E S C U , AL., G a m m a radiation sterilization of Ostrinia nubilalis
Hb. A n important pest of maize crops in Romania, Rev. Roum. Biol. 34 2 (1989)
107-111.
[22] RO§ CA, I., B A R B U L E S C U , AL., Sterility inheritance in the irradiated European
C o m Borer, Ostrinia nubilalis Hb., Rev. Roum. Biol. 35 1 (1990) 27-30.
[23] RO§ CA, I., “Research regarding use of g a m m a rays as genetic control of European
Corn Borer (Ostrinia nubilalis Hb.)”, paper presented at 8th National Symp. on
Environmental Protection, Constanta, 1990.
[24] B A R B U L E S C U , AL., “Achievements obtained at Fundulea regarding mass rearing of
E C B ”, paper presented at E C B W G Symp. Varna, Sept. 1988.
 IAEA-SM-327/38

THE STERILE INSECT TECHNIQUE

AND TRANSMISSION OF INHERITED STERILITY
TO CONTROL THE DIAMONDBACK MOTH,
Plutella xylostella (L.), AND THE CABBAGE WEB WORM,
Crocidolomia binotalis ZELL.

S. SUTRISN O, M . H O E D A Y A
Centre for the Application o f Isotopes and Radiation,
National Atom ic Energy A gency,
Jakarta Selatan,
Indonesia

Abstract
T H E STERILE INSECT T E C H N I Q U E A N D T R A N S M I S S I O N O F INH E R I T E D
STERILITY T O C O N T R O L T H E D I A M O N D B A C K M O T H , Plutella xylostella (L.), A N D
T H E C A B B A G E W E B W O R M , Crocidolomia binotalis ZELL.
T w o major cabbage insect pests, the diamondback moth (DBM), Plutella xylostella
(L.), and the cabbage w e b w o r m ( C W W ) , Crocidolomia binotalis Zell., have been studied
intensively in relation to the possibility of insect control, either by using the sterile insect
technique (SIT) or by transmission of inherited sterility. Four hundred and fifty moths each
of P. xylostella and C. binotalis were sterilized with 0.30 and 0.40 kGy of g a m m a
radiation, respectively. The irradiated insects, released into two laboratory cages
(90 c m x 60 c m x 60 cm) containing 50 unirradiated moths, reduced the F, population
of P. xylostella and C. binotalis by 61.1 and 65.3%, respectively. In the field cage
(2 m X 2 m x 2 m), the F, population reductions in the respective insect species were 55.6
and 50.5%. Release of about 4500 irradiated C. binotalis into an experimental plot
(15 m x 10 m) containing 500 normal moths reduced the F, population by 41.02% in the
dry season and 50.55% in the rainy season. Release of about 5000 irradiated P. xylostella
into an experimental plot containing about 350 moths of a natural population resulted in a
reduction in egg hatch of 85.9% in the unreleased plot and 17.0% in the released plot.
Experiments on the F, sterility of P. xylostella and Ç. binotalis were also conducted to
explore the possibility of controlling these pests by using inherited sterility. In D B M , a
substerilizing dose of 200 G y resulted in considerable F, sterility. Release of Fj inherited
sterile D B M moths into population laboratory cages caused different levels of population
suppression, depending on the irradiation dose. A substerilizing dose of 200 G y induced
inherited F, sterility, which reduced the population by 2 2 % in D B M . In. C. binotalis, a
substerilizing dose of 0.275 kGy was sufficient to induce inherited sterility, because the levels
of sterility in the parent and in the F, generation were 65.82 and 84.64%, respectively.

395
396 SUTRISNO and HOEDAYA

1. INTRODUCTION •

The use o f insects to control populations o f their own kind through, the transfer
o f damaged genetic material represents a new approach that could prove to be useful
for the control o f a number o f major insect pests throughout the world. This autocidal
method o f control is now generally referred to as the sterile insect technique (SIT).
Experiments have been conducted under laboratory and field cage conditions to
explore the possibility o f using SIT for controlling two cabbage pests in Indonesia.
The diamondback moth, Plutella xylostella (L .), and the cabbage webworm ,
Crocidolomia binotalis Z e ll., are the most important pests o f cruciferous crops [1].
These insects attack many varieties o f cabbage. The other important insect pest o f
cabbage is the cabbage looper, Diachrysia (Plusia) orichalcea (Z .) [2]. The highest
econom ic losses o f cabbage crops caused by P. xylostella and C. binotalis occur
in the dry season whenever control measures using insecticides are not properly
applied [3]. Plutella xylostella is found on young cabbage plants, while C. binotalis
is found on young and old cabbage plants.
T o control these insects, farmers usually apply various types o f insecticide,
e .g . Ambush 2 EC, Basudin 60 EC, Dipel W P and Lannate L [4]. The use o f
insecticides tends to result in the development o f insect resistance. A strain o f
P. xylostella found at Lembang, West Java, has shown resistance to permethrine
(pyrethroid insecticide) (up to eleven fold) [5]. Increasing insect resistance to
insecticides stimulates farmers to apply a higher dose or more toxic insecticides.
The use o f the parasite Angitia cerophaga to control P. maculipennis was introduced
in Indonesia, but the results were not satisfactory, particularly in regions where
intensive use o f insecticides was practised [ 6].
The purpose o f this paper is to explore the possibility o f using SIT and
transmitted inherited sterility to control the major cabbage pests in Indonesia because
o f the inadequate control measures available.

2. M A TE RIALS A N D M ETH ODS

2.1. SITprogramme in P. xylostella and C. binotalis

Unsexed, 3 to 4-d-old pupae o f P. xylostella were irradiated with

0.30 kGy o f gamma rays in a “ Co Gamma Cell-220 irradiator. T w o cages
(90 cm x 60 cm x 60 cm ) were used in the laboratory test. Fifty (25 pairs)
unirradiated and 450 (225 pairs) irradiated moths (at a ratio o f 1:9) were placed in
one cage; only 50 (25 pairs) unirradiated moths were introduced into the second
cage. Moths that emerged at the larval stage were fed cabbage leaves. The test was
repeated five times.
 IAEA-SM-327/38 397

The field cage tests were conducted at the field station of the Horticultural
Research Institute in Cipanas, West Java. Two field cages (2 m x 2 m x 2 m)
containing 300 two-week-old cabbage crops were used to conduct experiments that
were similar to those carried out in the laboratory. The experiments were replicated
three times.
. The field tests were slightly different to those done in the laboratory and field
cage tests. They were carried out in the agricultural production area of Cipanas,
West Java. Each experimental unit (plot) (15 m x 10 m) contained 600 two-week-
old cabbage crops. The natural population was developed prior to the release of
irradiated moths in the plots. The irradiated moths were released into the plot at a
ratio of 14:1 to the natural population. Egg hatchability was observed by collecting
eggs randomly .
In the C. binotalis test, unsexed, 6-d-old pupae were exposed to 0 .4 0 kGy. The
conditions for the laboratory tests were the same as those used for P. xylostella.
The F , generation was measured by counting the larvae in weeks 1 and 2 after
release.
The field cage tests were also carried out in the Cipanas area. The conditions
for these tests were the same as those used in the P. xylostella laboratory test, except
that four-week-old cabbage crops were used.
The field tests were again conducted in the Cipanas area, but the experi­
mental side of the field was isolated to avoid reinfestation. Two experimental plots
(7 m X 7 m) containing 250 cabbage crops were used. The conditions were similar
to those of both the laboratory and field cage tests, but the number of insects used
was ten times higher than that used in either of the other tests.

2 .2 . Transmission of inherited sterility in P. xylostella and C. binotalis

Substerilizing doses of 0 .1 0 -0 .2 kGy were applied to study the level of

sterility of the irradiated pupae of P. xylostella and their subsequent generations
(F t and F 2). Other biological parameters such as the viability of the pupae and
adults and their fecundity were also observed. In addition, preliminary results of the
population reduction in P. xylostella, as affected by the release of Fi moths under
laboratory conditions, were observed.
In C. binotalis, 3 to 4-d-old male pupae were exposed to substerilizing doses
that ranged from 0.25 to 0.35 kGy. The 40 moths that emerged were crossed with
untreated females to observe the F ! sterility level. Female pupae were also observed
in order to obtain information on their potential for developing F ¡ sterility. The sub­
sterilizing doses used to obtain the inherited sterility in the females were much lower
than those used for the males, ranging from 0.075 to 0 .1 5 kGy. The inherited
sterility measured in the F ! generation was similar to that of the irradiated male
pupae.
398 SUTRISNO and HOEDAYA

TA BLE I. REDUCTION IN THE F ! POPULATION OF P. xylostella AND

C. binotalis AS AFFECTED B Y THE R ELEA SE OF STERILE MOTHS

No. of No: of
parents (P) Fj populations
Condition Reduction
Treated (%)
Control Unirradiated Irradiated Control Treated

P. xylostella
Laboratory8 50 50 + . 450 1141 ' 430 61.1
Field cageb 50 50 + 450 378 180 55.6
315 338 + 4827 466c 42.4

OO
s
Field plot

C. binotalis
Laboratory2 50 50 + 450 556 193 65.3
Field cageb 50 50 + 450 275 136 50.55
Field plot:
Dry season 500 500 + 4500 1219 719 41.02
Rainy season 500 500 + 4500 2216 1096 50.55

a Average of five tests.

b Average of three tests.
c First instar larvae.

3. RESULTS AND DISCUSSION

3 .1 . SIT in P. xylostella and C. binotalis

The effects of releasing sterile insects on the reduction in population are shown
in Table I. Release of 450 irradiated (sterile) moths into the laboratory cage
containing 50 unirradiated moths reduced the emerging Fj population from 1141 in
the control cage to 430 in the released cage. Release of irradiated moths into the
unirradiated population at a 9:1 ratio resulted , in a 61.1% reduction in the
F , generation.
In the field cage tests (Table I), the F ! populations of the control and released
cages were 378 and 180, respectively. This indicated that the reduction in the F,
population, as affected by the release of irradiated moths, was 55.6% . In the field
tests, the release of irradiated moths to the field population at a ratio of about 14:1
resulted in.a reduction in egg hatchability from 8 5.9 to 17%. The number of eggs
collected from the control and treated plots was 942 and 2743, respectively.
 IAEA-SM-327/38 399

However, the F] population was 809 larvae (85.9% of 942) in the control plot and
466 larvae (17% of 2743) in the treated plot. Therefore, the reduction in the F|
population was 42.40% .
In C. binotalis, the effects of releasing irradiated (sterile) insects on the
reduction in progeny were quite promising (Table I). By releasing 450 irradiated
into 50 unirradiated insects under laboratory and field cage conditions, the F, popu­
lation was reduced by 65.3 and 50.55% , respectively. Release of about 4500 irradi­
ated pupae into an experimental plot containing 500 unirradiated pupae reduced the
F , population by 41.12% in the dry season and 50.55% in the rainy season. The
difference in F , population reductions obtained under laboratory, field cage and
field conditions may have been due to an uncontrolled source of variation related
to the different test conditions.

3 .2 . Inherited sterility in P. xylostella and C. binotalis

The effects of substerilizing doses of gamma irradiation on the sterility of the

F , and F 2 generations in DBM are presented in Tables II and III. The use of
inherited sterility in lepidopteran insects showed its potential for suppressing popula­
tions [7]. In P. xylostella, the viability of the irradiated pupae was much lower than
that of. the unirradiated pupae (54.66 versus 74.33% ), as indicated in Table II.
Similar results were also found for moth emergence. The fecundity was not affected
by the doses of radiation. The average sterilities for F , males and females were
about 64% at 175 Gy and 63% at 200 Gy. The F2 sterility level was lower than that

TA BLE II. VIABILITY, FECUN D ITY AND STERILITY OF F , IRRADIATED

DBM M ALE PUPAE

Irradiation dose Viability (%)a Fecundity b Sterility (%)c

(Gy) Pupae Moths Male Female Male Female

0 74.33 71.66 1826.00 1826.00 12.21 12.21

(unirradiated)
100 58.66 57.66 1614.66 1954.00 41.52 42.86
125 44.66 41.00 1638.66 2083.66 55.01 56.18
150 54.66 52.66 1541.00 1781.33 55.19 53.58
175 46.00 38.66 1534.66 1404.00 54.55 74.09
200 54.66 49.66 1852.00 1656.00 56.81 71.74

a Average percentage of 100 neonatal larvae, three replications.

b Average of the number of eggs produced by ten pairs of each sex, three replications.
c Average of ten pairs of moths of each sex, three replications.
 400
TA BLE III. VIABILITY, FECUN D ITY AND STERILITY OF F 2 IRRADIATED DBM M ALE PUPAE

Irradiation dose Viability (%)a Fecundity b Sterility (%)c

с
(Gy) F, 9 x ui 9 F, o' x ui с
у F, cr x ui 9 F, 9 F, с? x ui 9 F, 9 x ui o'

Q
X
Pupae Moths Pupae Moths Male Female Male Female Male Female Male Female

SUTRISNO
0 74.33 69.66 1395 1395 4.32 4.32
(unirradiated)
100 76.66 75.00 63.66 58.00 1186.50 1683.50 951.00 1072.00 9.47 11.92 10.79 7.51

and HOEDAYA
125 61.00 51.00 56.00 50.66 1382.00 1113.00 1096.50 1222.00 19.56 9.88 15.55 19.36
150 69.66 65.00 47.66 43.33 1108.00 1147.50 977.00 1329.00 7.59 15.93 13.26 8.90
175 37.00 29.33 19.66 17.00 1139.50 1074.00 1015.00 696.50 17.06 33.65 15.55 41.85
200 62.33 56.33 68.00 60.66 1465.50 1119.00 1223.50 1376.00 8.82 13.18 9.33 14.69

a Average percentage of 100 neonatal larvae, three replications.

b Number of eggs produced by ten pairs of moths of each sex, two replications.
c Ten pairs of moths, two replications.

ui: unirradiated.
 IAEA-SM-327/38 401

TA BLE IV. PERCEN TAG E3 OF POPULATION REDUCTION IN DBM AS

A FFEC TED B Y THE R ELEA SE OF F , MOTHS OF IRRADIATED M ALE
PUPAE

Treatment Replication
Total Average
(No. of released moth pairs) 1 2 3 4

5 ui
5 ui + 45 i - 1 8.93 2.37 22.16 5.22 38.68 9.67
5 ui + 45 i — 2 14.43 16.91 17.66 29.73 77.73 19.43
5 ui + 45 i — 3 25.43 20.47 33.53 12.31 91.74 22.93

Calculated u u r x 100%
where u is the offspring population (moths) in the unreleased cage/check and R is the
offspring population (moths) in the released cage.

ui: unirradiated.
i — 1:irradiated with a dose of 100 Gy.
i — 2:irradiated with a dose of 150 Gy.
i — 3:irradiated with a dose of 200 Gy.

of F ! ; however, it was close to the sterility level of the irradiated parents

(Table III). Therefore, doses of 175 or 200 Gy can be considered for obtaining
inherited sterility in the F) progeny of P. xylostella. The results of the experiment
to show the effects of releasing F , sterile moths on population reduction in
P. xylostella under field cage conditions are shown in Table IV. Release of F]
moths originating from irradiated pupae at doses of 100, 150 and 200 Gy into
laboratory cages containing unirradiated moths at a ratio of 9:1 caused population
reductions of 9.6 7 , 19.43 and 22.9 3 % , respectively.
In C. binotalis, an experiment on the F] inherited sterility was conducted to
explore the possibility of its application in a control programme. The relationship
between the substerilizing doses and sterility are shown in Tables V and VI. A dose
of 0 .25 kGy resulted in 63.95% sterility in the irradiated male parents and about
81% (average) sterility in the F t generation (Table V). The higher the dose, the
higher the level of sterility in both the parents and the F , generation. A dose of
0.275 kGy was sufficient to obtain inherited sterility, because the levels of sterility
in the parents and in the F] generation were 65.82% and 8 4 .6 4 % , respectively.
402 SUTRISNO and HOEDAYA

TABLE V. EFFECTS OF SUBSTERILIZING DOSES ON THE INHERITED

STERILITY OF IRRADIATED MALE PUPAE OF C. binotalis

Dose P(, i x 9 ui F, уx
с 9 ui F, 9 x a ui
(kGy) No. of eggs Sterility No. of eggs Sterility No. of eggs Sterility
(%) (%) (%)

0 2384 4.43 1186 3.81 1180 5.21

0.250 2172 63.95 891 81.71 819 82.66
0.275 2084 65.82 819 83.78 794 84.64
0.300 1909 68.05 759 85.51 689 85.96
0.325 1870 70.86 636 88.52 607 89.46
0.350 1686 74.79 599 92.82 547 93.60

H S D : 1% 1.189 2.57 2.59

5% 1.38 1.87 1.89
CC 11.1 14.34 14.42

i: irradiated,
ui: unirradiated.
HSD: high significant difference.
CC: correlation coefficient.

T A B L E VI. E F F E C T S O F S U B S T E R I L I Z I N G D O S E S O N T H E I N H E R I T E D
S T E R I L I T Y O F I R R A D I A T E D F E M A L E P U P A E O F C. binotalis

Dose Pç i x o' ui F, o' x 9 ui F, ç x cr ui

(kGy) No. of eggs Sterility No. of eggs Sterility No. of eggs Sterility
(%) (%) (%)

0 1445 3.75 720 5.4 520 4.3

0.075 1121 28.6 467 15 486 11.6
0.1 875 54.75 385 30 290 6.9
0.125 743 74.65 341 40 285 3.6
0.15 626 79.6 296 54 246 4.4

HSD: 1% 22.44 20.4 5.51

5% 17.81 16.55 4.45
CC 23.42 15.63 26.9

i: irradiated,
ui: unirradiated.
 IAEA-SM-327/38 403

The irradiated female parents showed slightly different results (Table VI). The
F[ progeny of the irradiated males were more sterile than the parents, and the Fj
progeny of the irradiated females were partially sterile, but they were as sterile as
the irradiated female parents (Tables V and VI). The male progeny were still the
more sterile of the two sexes. These phenomena also occur in the cabbage looper,
Trichoplusia ni (Hubner) [8].

ACKNOWLEDGEMENTS

The authors are grateful to D. Sutardji for preparation of the manuscript.

REFERENCES

[1] K A L S H O V E N , L.G.E., Pest of Crops in Indonesia, P.T Ichtiar Baru Van Houve,
Jakarta (1974).
[2] S A S T R O S I S W O J O , S., S U K A R N A , D., O M O Y , T.R., “The effect of several kinds
of pyrethroid synthetic insecticide on caterpillar cabbage leaf feeder and parasitoid of
diadegma eucerophagà on cabbage fields”, Proc. 2nd Entomological Congress, 1983,
Jakarta (1983).
[3] S U D A R W O H A D I , S., The relationship between planting time of cabbage and popula­
tion dynamic of Plutella maculipennis Curt and Crocidolomia binotalis Z., Bull.
Penel. Hort. 33 (1975) 3-14.
[4] Pesticide for Agriculture and Forestry, Directorate of Food Crop Protection,
Indonesian Department of Agriculture, Jakarta (1987).
[5] S O E K A R N A , D., KILLIN, D., S U D A R W O H A D I , S., “Status of resistency of
diamondback moth caterpillar, Plutella maculipennis Curt, to pyrethroid insecticide”,
Proc. Symp. Entomology, Bandung, 1982, (1982).
[6] N.C.C.A.A. VOS, News of Agricultural Reseach Center, 134, Bogor, 3 (1953).
[7] KNIPLING, E.F., Suppression of pest Lepidoptera by releasing partially sterile males:
A theoretical appraisal, Bio. Sci. 20 8 (1970) 465.
[8] N O R T H , D.T., H O L T , G.G., “Inherited sterility and its use in population suppression
of Lepidoptera”, Application of Induced Sterility for Control of Lepidopterous
Populations (Proc. Panel Vienna, 1970), IAEA, Vienna (1971) 99.
 IAEA-SM-327/39

ESTEMLIDAD EN LA F!
DE Diatraea saccharalis (Fab.)>
LEPIDOPTERA: CRAMBIDAE
I. Efectos de las dosis subesterilizantes
sobre la reproducción y la competividad

J.C . GARCIA, J.R . GONZALEZ

Instituto de Investigaciones de Sanidad Vegetal,
Ministerio de la Agricultura i
Ciudad de la Habana,
Cuba

Abstract-Resum en

F, STERILITY O F Diatraea saccharalis (Fab.), L E P I D O P T E R A : C R A M B I D A E .

I. E F F E C T S OF SUBSTERILIZING DOSES O N REPRODUCTION AND
COMPETITIVENESS.
Female and mâle chrysalises of Diatraea saccharalis (Fab.) were irradiated at doses of
150, 200 and 300 Gy. The moths which emerged were mated with untreated individuals of
the opposite sex. Their mating capacity was not affected, nor was there any effect on the
mating capacity of the progeny of males irradiated at the first two doses down to the
F 3 generation, or on that of the F, of the irradiated females. Sterility was high where one of
the pair had been irradiated; the percentage of non-viable eggs was between 79.2 and 99.4
for irradiated parents and between 90.8 and 99.8 in the F, of irradiated males. There was
recovery of fertility in the F, of the treated females, the percentage of dominant lethal muta­
tions lying between 4.4 and 34.5. The F2 adult progeny of the initially irradiated males
retained significantly high percentages of non-viable eggs, but the absolute values were lower
than in the preceding generation, indicating some recovery of fertility. In the F3, the recov­
ery of fertility was even greater, the percentage of non-viable eggs approaching the level of
the control, except in the case of males whose ancestors were all from the line of irradiated
males. The action of radiation continued to be evident during post-embryonic development.
The percentages of adult formation in the F, of both irradiated males and females and in the
F2 of irradiated males were significantly lower than in the control; the F 3 larval generation
exhibited adult conversion values similar to the control. The moths from male chrysalises
irradiated at 200 and 300 G y were competitive when mated in the proportion of 1:1:1 with
untreated females and males. The competitiveness values calculated by the Fried method were
0.78 and 3.08, respectively, and, for the higher dose, the mating competitiveness calculated
by the method of direct observation of máting adults was 1.33.

E S T E R I L I D A D E N L A F, D E Diatraea saccharalis (Fab.), L E P I D O P T E R A :

C R A M B I D A E . I. E F E C T O S D E L A S DOSIS S U B E S T E R I L I Z A N T E S S O B R E L A
R E P R O D U C C I O N Y L A COMPETIVIDAD.
Se irradiaron crisálidas hembras y machos de Diatraea saccharalis (Fab.), con dosis
de 150, 200 y 300 Gy. Las polillas emergidas se aparearon con individuos del sexo opuesto

405
406 GARCIA y GONZALEZ

no tratados, no afectándose su capacidad para copular, ni la de la descendencia de los machos

irradiados con las dos primeras dosis hasta la generación F 3 y tampoco en la F, de las
hembras irradiadas. Fue alta la esterilidad de las parejas donde uno de sus miembros era irra­
diado; los porcentajes de huevos inviables (% HIV) se situaron entre 79,2 y 99,4% de los
parentales irradiados y entre 90,8 y 99,8% en la F, de los machos irradiados. La fertilidad
en la F, de las hembras tratadas se recuperó, estando entre 4,4 y 34,5 los porcentajes de
mutaciones letales dominantes. Los adultos descendientes F 2 de los machos inicialmente
irradiados mantuvieron significativamente altos los porcentajes de HIV, pero los valores abso­
lutos fueron menores que en la generación precedente, indicando una cierta recuperación de
la fertilidad. En los F 3 la recuperación de fertilidad fue aún mayor, tendiendo los porcentajes
de HIV al nivel del testigo, excepto en los machos cuyos antecesores fueron siempre de la
línea de machos irradiados. Durante el desarrollo post-embrionario continuó manifestándose
la acción de las radiaciones. Fueron significativamente menores que el testigo los porcentajes
de formación de adultos en la F, tanto de machos como de hembras irradiadas y en la F2 de
los machos irradiados; la generación larval F 3 presentó similares valores de conversión en
adultos que el testigo. Las polillas procedentes de crisálidas machos irradiadas con 200 y
300 G y fueron competitivas al ser apareadas en la relación 1:1:1 con hembras y machos no
tratados. Los valores de competitividad calculados por el método de Fried fueron 0,78 y 3,08
respectivamente, y en la dosis superior la competitividad para el apareamiento calculada por
el método de observación directa de adultos copulando fue de 1,33.

1. INTRODUCCION

Desde el primer informe sobre la trasmisión de esterilidad a la descendencia

F] de insectos machos irradiados aparecida en Carpocapsa pomonella [1] se ha
observado que ocurre el mismo fenómeno en muchos otros lepidópteros y hay crite­
rios de que la esterilidad en la F ] puede ser inducida en todas las especies de este
orden [2]. Siendo de particular interés el desarrollo de ésta como un método práctico
de control de lepidópteros [3], Walker et al. [4, 5] y Sanford [6] coincidieron en que
se presentaba alta esterilidad en la F j de los machos de Diatraea saccharalis, pero
hay diferencias en la respuesta a las dosis evaluadas por estos autores, planteando
ambos que podía haber interferencia en los resultados debido a factores genéticos,
nutritivos e incluso acción de patógenos. Por otra parte, Walker et al. [7] plantearon
que se podía obtener una moderada producción de adultos bastante competentes irra­
diando crisálidas pero que se reduce la competitividad severamente al aplicar la dosis
esterilizante, y se daña la esperma. Estudios realizaos sobre otras especies de lepi­
dópteros [2, 8, 9] han demostrado que los insectos irradiados con dosis subesterili-
zantes son competitivos sexualmente. Por todo lo anterior y debido a la importancia
que tiene por ser la plaga más extendida y peligrosa del cultivo de la caña de azúcar,
está incluida entre las que deben ser estudiadas para su control por método genético
[2 , 6 , 10].
 IAEA-SM-327/39 407

El objetivo de este trabajo fue determinar los efectos de la exposición de crisá­

lidas de Diatraea saccharalis a las radiaciones gamma, en el rango de dosis sub-
letales entre 150 y 300 G y, sobre la capacidad reproductiva y competitividad de las
polillas emergidas y sobre la capacidad reproductiva de su descendencia.

2. M A TE R IA LE S Y M ETO DO S

Los insectos procedían de una cría en laboratorio, alimentándose las larvas con
una dieta merídica [ 11 ].
Las crisálidas con más de cin co días de edad y previamente sexadas se coloca ­
ron en viales de vidrio de 3,0 cm de diámetro y 7 ,5 cm de altura tapados con tapones
de algodón y gasa, y se irradiaron en una fuente de ^ C o . La temperatura ambiente
durante la irradiación fue de 2 2 °C y el aire de la cámara no fue renovado durante
él período de irradiación. El procedimiento seguido fue aplicar primero la mitad de
la dosis, esperar dos horas, y aplicar entonces la segunda mitad. Las crisálidas duran­
te el período de recuperación se colocaban dentro de una incubadora a 27 ± 1°C .

2.1. Estudio de la capacidad reproductiva

Durante la mañana siguiente a la emersión de los adultos, se colocaban éstos

individualmente en las jaulas de apareamiento y ovoposición con un individuo del
sexo opuesto no irradiado com o pareja; los recipientes de apareamineto eran frascos
de vidrio de 10 cm de diámetro y 12,5 cm de altura con papel parafinado en su inte­
rior, cubierto el fondo con algodón y papel de filtro embebidos en agua.
L os papeles parafinados se revisaban diariamente, recortándose las porciones
con puestas de huevos, las que se ponían a incubar hasta la eclosión de las larvas.
Las larvas se contaban y transferían a los recipientes con dieta, además se contaban
los huveos que quedaban en “ cabezas negras” sin eclosionar y los huevos infértiles.
A l m orir cada hembra se determinaba la presencia de esperm atófoíos en la bursa
copulatrix.
A los trece días de eclosionadas las larvas se individualizaban en recipientes
de cría con dieta hasta su conversión en pupa; al emerger el adulto se formaban las
parejas con individuos de la cría, repitiéndose el proceso con la descendencia hasta
la F3, según el esquema que aparece en la Fig. 1 y cuya sim bología representa a los
individuos parentales irradiados y las generaciones subsecuentes, donde el número
indica la dosis con que se irradió la pupa que dió origen al adulto parental irradiado
y la primera letra su sexo (macho: M ; hembra: H ); en ese orden se le sigue
adicionando letras según el sexo de la generación filial que sea. Por ejem plo:
F2200M M H significa una hembra F 2 hija de un macho F! que era hijo de un macho
irradiado en estado de pupa con 200 Gy.
408 GARCIA y GONZALEZ

150M 150Н
200M 200Н
300M 300Н

150MM 150MH 150HM 150НН

200MM 200MH 200HM 200НН
300MM 300MH 300HM 300НН
'a , 9 cr 9
F2
150MMM 150ММН 150МНМ 150МНН
200MMM 200ММН 200МНМ 200МНН
300MMM 300ММН 300МНМ 300МНН

F3
150MMMM 150MMMH 150MMHM 150MMHH
200MMMM 20ÓMMMH 200MMHM 200MMHH
300MMMM 300MMMH 300MMHM 300MMHH

FIG. 1. Simbología y esquema utilizado para identificar las variantes evaluadas.

Las dosis de irradiación empleadas fueron 150, 200 y 300 Gy (16 G y/m in),
así com o un testigo no irradiado. En todos los casos se trataron machos y hembras;
para los parentales se tomaron 21 parejas y para las descendencias el número depen­
dió de la disponibilidad de individuos. Los indicadores evaluados fueron: huevos
puestos por cada hembra copulada, huevos fértiles en cabezas negras, larvas eclo­
sionadas, espermatóforos por hembra copulada, crisálidas y adultos form ados. Con
los primeros tres indicadores se calcularon los porcentajes de huevos con mutaciones
letales dominantes (M L D ), porcentajes de huevos inviables (HIV) y prom edio de
larvas eclosionadas, evaluándose estadísticamente por la prueba de rangos múltiples
de Duncan para el 5 % de probabilidad de error, previa transformación de los porcen­
tajes en .arcsen raíz cuadrada y el número de larvas eclosionadas en V x + 0,5;
con los demás indicadores se calcularon los porcentajes de cópula, de pupación, de
emersión de adultos y de conversión larva-adulto, comparándose con el testigo a
través del estadígrafo ts.

2.2. Estudio de la competitividad

Se usaron los adultos emergidos hasta 48 horas después de irradiadas las crisá­
lidas macho con dosis de 200 y 300 Gy (10,14 Gy/m in). Para calcular las competiti-
 IAEA-SM-327/39 409

vidad sexual (C f) se utilizó el método de Fried [12] evaluando la cantidad de larvas

emergidas en cada una de las tres combinaciones de apareamiento siguientes:

Número de machos Número de machos Número de hembras

irradiados normales normales
(cr CTl) (cr ctN) ( 9 9 N)

10 10 10
10 10
_ 10 10

Las jaulas de apareamiento fueron recipientes cilindricos plásticos de 25 cm

de alto y 24 cm de diámetro acondicionados com o los del experimento anterior y el
tratamiento a las puestas de huevos fue similar también. Se replicó 10 veces cada
com binación y se compararon las medias de larvas emergidas por hembra copulada
mediante la prueba de Duncan para un 5% de probabilidad de error, previa transfor­
mación a raíz cuadrada.
También se calculó la competitividad de los individuos irradiados con 300 Gy
por el método directo de adultos copulando (С о ), según la siguiente fórmula:

S/N

donde n es el número de machos normales que copularon, i es el número de machos

irradiados que copularon y N y S son las cantidades de machos normales e irradiados
utilizados respectivamente [13].
Para identificar el tipo de macho se coloreaba parte de un ala con solución de
verde brillante, aplicada con un pincel, unas veces a los irradiados y otras a los no
irradiados. En cada jaula se ponían 30 adultos (10 machos normales, 10 machos
irradiados y 10 hembras), se extraían con un vial de vidrio las parejas que se forma­
ban durante la noche hasta el siguiente día y se registraba el tipo de macho que había
copulado. El experimento se replicó ! 8 veces.

3. RESU LTAD O S Y DISCUSION

Los adultos emergidos de crisálidas de más de cin co días de edad e irradiadas

con dosis de 150, 200 y 300 Gy no presentaron afectada su capacidad para copular
con individuos del sexo opuesto no tratados, com o se observa en el Cuadro I. Esto
ha sido similar en otras especies de lepidópteros cuando son tratados con dosis sub-
esterilizantes [1 4 -1 6 ], no alterándose negativamente tampoco esta capacidad en la
descendencia Fj de las parejas donde las hembras fueron irradiadas y la descenden-
410 GARCIA y GONZALEZ

C U A D R O I. NU M ERO DE PAREJAS FORM ADAS POR M ACHOS O

H EM BRAS IR R A D IA D A S CON INDIVIDU OS DEL SEXO OPUESTO
CORRESPONDIEN TE N O IR R A D IA D O S. PORCENTAJES DE C O PU LA Y
SIGN IFICACION. E V A L U A C IO N DE SU DESCEND EN CIA

Número de Copularon
Generación Variante ts1 Significación2
parejas (% )

Parental testigo 21 71,4 — —

150M 21 66,6 0,334 NS

200M 21 57,1 0,971 NS
300M 21 57,1 0,971 NS
150H 20 75,0 0,258 NS
200H 20 65 ,0 0,442 NS
300H 18 94,4 - 2 ,0 3 0 S

F, testigo 23 47,8 —
150MM 24 50,0 0,717 NS
150MH 24 50,0 - 0 ,1 4 9 NS
200M M 24 58,3 - 0 .4 2 5 NS
200M H 14 50,0 - 0 ,1 2 8 NS
300M M 9 0 ,0 — —
300MH 1 100,0 — —
150HM 15 86,6 -2 ,6 1 1 S
150HH 15 73,3 - 1 ,5 9 4 NS
200HM 9 55,5 - 0 ,3 9 4 NS
200HH 13 61,5 - 0 ,7 9 6 NS
300HM 4 75 ,0 - 1 ,0 4 7 NS
300HH 8 62,5 - 0 ,7 2 2 NS

f2 testigo 24 79,1 — —
150M M M 10 80,0 - 0 ,0 5 5 NS
150M MH 7 71 ,4 0,419 NS •
200M M M 12 66,6 0,800 NS
200M M H 0 — — —
150MHM 4 75,0 0,184 NS
150MHH 10 70,0 0,561 NS
200M H M 13 69,2 0,662 NS
200M HH 20 90,0 - 1 ,0 0 6 NS

F3 testigo 25 84,0 — —
150M M M M 21 66,6 . 1,378 NS
150M MMH 25 56,0 2,218 S
200M M M M 25 80,0 0,369 NS
200M M M H ■ 25 84,0 0,000 NS
150M MHM 25 76,0 0,710 NS
150MMHH 25 84,0 0,000 NS

1 Significación para un 5 % de probabilidad de error (t student) con respecto al testigo en cada

generación.
2 NS: no significativo. S: significativo.
 IAEA-SM-327/39 411

C U A D R O II. PROM EDIOS DE FE C U N D ID A D , PORCENTAJES DE HUEVOS

IN VIABLES (HIV), PORCENTAJES DE M U TACIO N ES LETALES
D O M IN A N TE S (M L D ) Y LARVAS E CLO SIO N A D A S EN PAREJAS
F O R M A D A S POR UN M A C H O IR R A D IA D O Y U N A H EM BRA NO T R A T A D A
Y EN SU DESCEN D EN CIA

Fecundidad HIV M LD Larvas

Generación Variante
promedio (% ) (* ) eclosionadas

Parental testigo 190,1a 40,5b 15,7b • 112,2a

150M 172,6a 79,2a 66,7a 44,8ab
200M 134,6a 74,0a 64,9a 46,2ab
300M 139,6a 93,8a . 89,9a 8,6b

F, ■ testigo 345,5a 32,5c 7,7c 209,5a

150MM 175,0b 98,9ab 93,3a 4,1b
200M M 155,9b 99,8a 99,4a 1,0b
150MH 172,8b 90,8b 85,0b 18,0b
200M H 144,7b 91,9ab 89,3ab 15,8b

f2 testigo 414,4a 13,3c 5,0c 331,3a

150M M M 282,2ab 69,5ab 36,4b 90,1b
200M M M 150,8b 86,5ab 63,8ab 20,6c
150M MH , 283,4ab 44,2bc 28,8bc 151,8ab
200M H M 2 5 5 ,6ab 87,5ab 75,9ab 26,6b
150MHH 185,8b 93,1a 92,4a 25,2b
200M HH 415,2a 6 4 ,lab 62,5ab 122,6b

F3 testigo 20 7,2ab l,9 e l,6 e 199,1a

150M M M M 3,5c 93,2a 89,4a 0,3c
200M M M M 190,2ab 66,8b 59,4b 63,5b
150M M M H 149,9b 48,8bc 46,6bc 90,6b
200M M M H 154,7b 34,2cd 19,4cd 96,5b
150M M H M 252,5a 6 ,le 4,0de 234,6a
206M M H H 285,2a ll,6 d e 7,8de 245,3a

Letras iguales en la misma columna de cada generación indican que no hay diferencias
significativas para un 5% de probabilidad de error.

cia hasta la F 3 donde los machos de las parejas parentales fueron irradiados con las
dos dosis inferiores; en este último caso, la diferencia significativa que apareció en
F3150M M M H fue para el 5% y no para menores niveles de significación, por lo
que pensamos que esto no se debió a la acción de las radiaciones sobre los parentales
tres generaciones antes. En el Cuadro I vem os también que las nueve parejas forma­
412 GARCIA y GONZALEZ

das por los machos FÍ300M M no copularon y sólo una pareja de F^OOMH pudo
formarse, no siendo, posible en este caso hacer comparación estadística.
En el Cuadro П vemos que no hubo diferencias significativas con respecto al
testigo en le fecundidad de las hembras normales apareadas con machos irradiados,
pero la aparición de mutaciones letales dominantes y los porcentajes de huevos
in viables fueron significativamente mayores. También vemos que el prom edio de
larvas eclosionadas fue significativamente bajo en las parejas en que los machos
fueron irradiados con 300 G y, presentándose valores intermedios y no significativa­
mente diferentes cuando los machos estaban irradiados con 150 ó 200 Gy.
Por consiguiente, la aplicación de radiaciones gamma a las crisálidas machos
produjo uñ aumento de la esterilidad en las parejas formadas con hembras no trata­
das, lo que coincide con lo obtenido por Sanford [6] en dosis menores, siendo esto
mayor en la dosis de 300 Gy. Con respecto a la Fj descendiente de machos irra­
diados, el Cuadro II muestra que las fecundidades promedios tanto de las hembras
Fj 150MH y F]200M H com o de las hembras copuladas por los machos F] 150MM
y F, 200M M fueron significativamente inferiores al testigo y similares entre sí y los
% de HIV y de M L D fueron significativamente superiores a los del testigo, apare­
ciendo los valores mayores en los provenientes de los parejas donde el macho era
de la línea de los irradiados. Además, se observa que estos porcentajes fueron más
altos que los de la generación anterior (P150M у P200M ) y la cantidad de larvas
eclosionadas prom edio fue significativamente baja, incluso menor que los promedios
de sus respectivos progenitores. El aumento de la esterilidad en la generación Fj
procedente de machos irradiados ocurrió co m o ya había sido reportado para esta
especie [4 -6 ] y pára muchas otras especies de lepidópteros [ 8, 1 4 -16], sobre todo
la heredada por los machos [17]. La esterilidad fue más marcada en las parejas donde
el macho era descendiente de irradiados.
En la F 2 de la línea de los machos inicialmente irradiados se pudieron formar
por lo general pocas parejas de adultos (Cuadro II) e incluso ninguna de
F 2200M M H por no emerger hembras. Además no se consideraron en los análisis
los resultados de fecundidad, fertilidad, etc., de las tres parejas que copularon en la
variante F2150MHM por ser muy pequeña la muestra. En la generación F 2 se
observa una recuperación de la fecundidad prom edio al nivel del testigo (Cuadro II),
aunque este indicador en F2200M M M y en F2150MHH se mantuvo significativa­
mente diferente. Los % de HIV y de M L D se mantuvieron significativamente altos,
aunque se observa que los valores promedio fueron por lo general menores que los
de la generación precedente, e incluso no apareció diferencia significativa entre los
resultados de F2150MM H y el testigo. En cuanto al número prom edio de larvas
eclosionadas, vem os que aún fue significativamente bajo en la mayoría de todas las
variantes descendientes de irradiados con relación al testigo pero sus valores absolu­
tos mayores que los de la generación precedente [8, 15], lo que indica una cierta
recuperación de la fertilidad.
 IAEA-SM-327/39 413

C U A D R O Ш . PRO M ED IOS DE FE C U N D ID A D , PORCENTAJES DE

HU EVOS IN VIABLES (HIV), PORCENTAJES DE M U TA CIO N E S LETALES
D O M IN A N TE S (M L D ) Y LARVAS E C LO SIO N A D A S EN PAREJAS
F O R M A D A S POR U N A H E M BR A IRRADIADA\Y UN M A C H O NO T R A T A D O
Y EN SU DESCEN D EN CIA

Fecundidad HIV M LD Larvas

Generación Variante
promedio (%) ' (%) eclosionadas

Parental testigo 190,1a 40,5b 15,7b 112,1a

150H 114,8ab 93,0a 82,1a 12,0b
200H 120,7ab 99,4a 98,1a 2,1b
300H 108,2b 99,4a 98,2a 1,9b

F, testigo 345,5a 32,5a 7,7a • 209,5a

150HM 271,1a 74,2a 34,5a 67,1a
200H M 132,6b 40,2a 12,8a 43,3a
150HH 280,9a 20,9a 4,4a 191,1a
200HH 281,4a 3 0 ,7a 12,4a . 123,6a
300HH 20 5 ,5ab 58,8a 21,2a 96,4a

Letras iguales en la misma columna de cada generación indican que no hay diferencias
significativas para un 5 % de probabilidad de error.

En la F 3 las fecundidades prom edio de las variantes de esta generación no

fueron significativamente diferentes con respecto al testigo, excepto la de
F3150M M M M (Cuadro П) que fue anormalmente baja al ser comparada con las
otras variantes de su generación y con relación a la fecundidad prom edio de las gene­
raciones precedentes. En esta generación F 3 la recuperación de fertilidad fue aún
m ayor, tendiendo los % de HIV y de M L D al nivel del testigo [8], excepto en ¡os
machos cuyos antecesores fueron siempre de la línea de machos irradiados
(1 50M M M M y 20 0 M M M M ), donde se mantuvieron niveles significativamentes al­
tos de M L D y de huevos inviables puestos por las hembras que ellos copularon.
C on relación a la capacidad reproductiva de las hembras, vem os que su fecun­
didad al ser irradiadas con la dosis de 300 G y (Cuadro Ш ) fue significativamente
baja, tomando valores intermedios no estadísticamente diferentes de la fecundidad
de las hembras que recibieron dosis de 150 y 200 Gy. Los % de HIV y de M L D en
todas las variantes fueron significativamente altos y el prom edio de larvas eclosiona­
das, por tanto, muy pequeño. La fecundidad en la generación F] correspondiente
fue similar a la del testigo (Cuadro П) en las hembras de la cría apareadas con los
machos FjlSO H M , mientras que fue significativamente menor la de las copuladas
por F^OOHM. N o se incluyó en el análisis estadístico los resultados de F^OOHM
414 GARCIA y GONZALEZ

C U A D R O IV . PORCENTAJES DE PU PACIO N , EM ERSION Y CONVERSION

EN A D U L T O S DE LAS L A R V A S PROCEDENTES DE PAREJAS PAR EN ­
TALE S D O N D E UN O D E LOS M IEM BROS ES IR R A D IA D O Y DE LAS
L A R V A S DE LOS A D U L T O S DESCENDIENTES F, Y F2

Emersión de Conversión
Pupación
adultos larva-adulto
Generación Larvas
Variante
larval evaluadas % ts % ts % ts

F, testigo 350 90,8 93,1 84,6

150M 341 94,1 - 1 ,6 4 4 73,2 7,024* 68,9 4,935*
200M 182 81,3 3,059* 79,0 4,205* 64,2 5,183*
300M 44 75,0 2,706* 36,3 7,188* 27,2 7,723*

150H 235 77,8 4,329* 81,9 . 3,716* 63 ,8 5,729*

200H 55 83,6 1,507 •78,2 2,775* .65 ,4 3,097*
300H 33 90,9 -0 ,0 1 0 60,0 .4,383* 54,5 3,692*

f 2 testigo 350 98,2 96,2 94,6

150MM 21 90,4 1,624 84,2 -1 ,6
60 76,1 . 2,444*
150MH 74 51,3 10,012* 52,6 6,591* 27,0 12,331*
200M M 13 100,0 - 0 ,9 3 0 92,3 0,604 92,3 0,325
200M H 51 80,3 4,369* 87,8 1,950 70,5 4,511*

F3 testigo 280 98,9 81,9 81,0

150M MM 85 96,4 1,377 75,6 1,236 72,9 1,566
* O

150MMH 85 98,8 0,080 94,0

88-“

92,9 - 2 ,9 2 6 *
200M M M 85 98,8 0,080 88,0 - 1 ,3
89 87,0 - 1 ,3 2 6

* Significativo con respecto al testigo de esa generación y columna para un 5% de

probabilidad de error (t student).

ya que fueron muy pocas las.parejas que se pudieron formar. Las hembras F.j copu­
ladas por machos de la cría pusieron una cantidad de huevos no significativamente
diferente. Los démas indicadores que aparecen en el Cuadro Ш no dieron resultados
estadísticamente diferentes entre las variantes y el testigo. La irradiación de las crisá­
lidas hembras dió com o resultado que el porcentaje de infertilidad de las polillas
emergidas al copular con machos de la cría fuese muy alto, incluso mayor que el
correspondiente producido por los machos irradiados al copular con hembras de la
cría, lo que indica que las hembras fueron más sensibles a las radiaciones que los
 IAEA-SM-327/39 415

machos. Esto es similar a lo ocurrido en otros lepidópteros [15, 17], y la poca des­
cendencia adulta F] que se obtuvo de estas hembras irradiadas tuvo una recupera­
ción de su fertilidad [17], lo que también habría que tenerse en cuenta al elaborar
una estrategia de liberación en la lucha contra esta plaga.
Durante el desarrollo post-embrionario (Cuadro IV ) continuó manifestándose
la acción de las radiaciones, de manera que los porcentajes de conversión larva-
adulto fueron significativamene más bajos en la generación larval F ,, tanto de los
machos com o de las hembras parentales irradiadas, y continuó siendo así en la gene­
ración larval F 2 (excepto en F 2200M M que, por su muy escaso número, tuvo valo­
res similares) descendiente de los adultos F! de machos inicialmentes irradiados, lo

CUADRO V . R E L A C IO N ENTRE LA DOSIS DE R A D IA C IO N Y LA

CO M P E T IT IV ID A D S E X U A L DE Diatraea saccharalis E V A L U A D A POR EL
M E TO D O DE FRIED

Larvas eclosionadas por

Valor de
Dosis hembra copulada 1 Larvas
competitividad
(Gy) esperadas
Hn Hs He (Cf)

200 190,9a 69,2c 137,7b 130,0 0,78

300 165,3a 28,3c ' 61,8b 96,8 3,08

1 Transformación raíz cuadrada.

Hn cantidad promedio de larvas procedentes de СУce N :Ç Ç N .
Hs cantidad promedio de larvas procedentes de cal :ç Ç N .
He cantidad promedio de larvas procedentes de O ' O 'N :c r o l : ÇÇ N .
Letras diferentes en la misma fila indican que hay diferencias significativas para un 5 % de
probabilidad de error.

C U A D R O V I. C O M P E T IT IV ID A D EN EL A PA R E A M IE N TO DE Diatraea
saccharalis IR R A D IA D A CON 300 G y, E V A L U A D A POR EL M E TO D O
D IRECTO DE A D U L T O S C O P U L A N D O

Hembras copuladas por:

(media ± ES) Valor de competitividad
(Со)
Machos normales Machos irradiados

3,83 ± 0,25 5,11 ± 0,32 1,33

ES = error estandard.
416 GARCIA y GONZALEZ

que concuerda con los resultados para otros lepidópteros [ 8, 14, 15] y los obtenidos
por Sanford [6] en la generación larval F j. Los porcentajes de pupación, emersión
de adultos y de conversión larva-adulto de las larvas F 3 fueron similares o supe­
riores al testigo.
- En el Cuadro V se observa que el valor de la competitividad C f (calculada por
el método de Fried) de adultos irradiados en estado de crisálida con 200 Gy fue
cercano a la unidad, siendo la cantidad prom edio de larvas emergidas por hembra
copulada (He) similar a la cantidad esperada. Con la dosis de 300 Gy y el valor de
competitividad superior a uno, la cantidad prom edio de larvas emergidas fue aproxi­
madamente un tercio menor que la esperada. También se observa que en cada dosis
la cantidad prom edio de larvas emergidas fue significativamente menor cuando esta­
ban juntos en la jaula de apareamiento machos irradiados y normales y menor aun
cuando solo hábía machos irradiados, en todos los casos con hembras normales.
En general, estos resultados de competitividad obtenidos coinciden con los
reportados en Plodia interpunctella (Hubner) [18], Ostrinia fumacalis (Guen) [11]
y Heliothis virescens (F .) [19]. L os resultados de la observación directa de los adul­
tos copulando, irradiados con la dosis mayor, aparecen en el Cuadro VI. El prom e­
dio de machos irradiados que copularon fue mayor que el de los normales y,
correspondientemente, el valor de competitividad (С о) mayor que uno, lo que coin ci­
de con lo obtenido en el experimento anterior calculado por el número de larvas
eclosionadas.

4. CONCLUSIONES

Los efectos de las radiaciones y la esterilidad heredada en la reproducción de

D. saccharalis fueron generalmente similares a los descritos para otras especies de
lepidópteros.
Los machos adultos irradiados en estado de crisálida con 200 y 300 Gy fueron
competitivos sexualmente, en condiciones de laboratorio, con relación a los machos
no tratados.
La presencia de igual cantidad de machos irradiados con las dosis estudiadas
y de machos normales p rov ocó una significativa reducción del prom edio de larvas
emergidas por hembra copulada, siendo mayor esta reducción cuando se aplicó la
dosis de 300 Gy.

R E F E R E N C IA S

[1] P R O V E R B S , M.D., Progress in the use of induced sexual sterility for the control of
the codling moth, Carpocapsa pomonella L. (Lepidoptera: Olethreutidae), Proc. Ento-
mol. Soc. Ont. 92 (1962) 5-11.
 IAEA-SM-327/39 417

[2] L A C H A N C E , L.E., Genetic Methods for the Control of Lepidopteran Species. Status
and Potential, Rep. ARS-28, United States Department of Agriculture, Washington,
■DC (1985).
[3] Modern Insect Control: Nuclear Techniques and Biotechnology (Proc. Symp. Vienna,
1987), IAEA, Vienna (1988).
[4] W A L K E R , D.W., Q U I N T A N A , V., Inherited partial sterility among survivors from
irradiation experiments, J. Econ. Entomol. 61 (1968) 318-319,
[5] W A L K E R , D.W., Q U I N T A N A , V., T O R R E S , J., Genetic collapse of insect popula­
tions. Extinction of inbred and outbred lines in laboratory population of the sugarcane
borer, J. Econ. Entomol. 64 (1971) 660-667.
[6] S A N F O R D , J.W., Inherited sterility in progeny of irradiated male sugarcane borers,
J. Econ. Entomol. 69 (1976) 456-458.
[7] W A L K E R , D.W., Q U I N T A N A , V., P A D O V A N I , F., “Effect of g a m m a irradiation
on immature sugarcane borers”, Sterility Principle for Insect Control or Eradication
(Proc. Panel Athens, 1970), IAEA, Vienna (1971) 513-524.
[8] LI, Y.Y., Z H A N G , H.Q., L O U , H.Z., C H A N , C.D., “The inherited sterility of the
corn borer (Ostrinia fumacalis Guen.)”, M o d e m Insect Control: Nuclear Techniques
and Biotechnology (Proc. Symp. Vienna, 1987), IAEA, Vienna (1988) 403-411.
[9] M A S T R O , V.C., S C H W A L B E , C.P., “Status and potential of sterility for control of
noxious Lepidoptera”, ibid., pp. 15-40.
[10] “Recommendations of this panel”, Application of Induced Sterility for Control of
Lepidopterous Populations (Proc. Panel Vienna, 1970), IAEA, Vienna (1971)
157-162.
[11] G O N Z A L E Z , J.R., G A R C I A , J.C. de C E S P E D E S , E., Una nueva dieta artificial para
la cría de Diatraea saccharalis (Fab.), C. Tec. Agrie. Cañ. 7 3 (1987) 1-6.
[12] FRIED, М., Determination of sterile competitiveness, J. Econ. Entomol. 64 4 (1971)
869-872.
[13] T E R U Y A , T., Z U K E Y A M A , H., Sterilization of the melon fly Dacus curcubitae
Coquillet with g a m m a radiation: Effect of dose on competitiveness for irradiated males,
Appl. Entomol. Zool. 14 3 (1979) 241-244.
[14] B U G H I O , A.R., “Parental and inherited sterility induced by g a m m a radiation in male
moths of the maize borer, Chilo partellus (Swinhoe)”, M o d e m Insect Control: Nuclear
Techniques and Biotechnology (Proc. Symp. Vienna, 1987), IAEA, Vienna (1988)
413-421.
[15] C A R P E N T E R , J.E., Y O U N G , J.R., S P A R KS , A.N., C R O M R O Y , H.L.,
C H O W D H U R Y , M.A., Corn earworm (Lepidoptera: Noctuidae): Effects of substerili-
zing doses of radiation and inherited sterility on reproduction, J. Econ. Entomol. 80
(1987)483-489.
[16] N A B O R S , R.A., PLESS, C.D., Inherity sterility induced by g a m m a radiation in a
laboratory population of the European c o m borer, Entomol. Soc. A m . 74 6 (1981)
701-702.
[17] H E N N E B E R R Y , T.J., C L A Y T O N , T.E., Effects of g a m m a radiation on pink boll­
w o r m (Lepidoptera: Gelechiidae) pupae: Adult emergence, reproduction, mating, and
longevity of emerged adults and their F, progeny, J. Econ. Entomol. 81 1 (1988)
322-326.
418 GARCIA y GONZALEZ

[18] A H M E D , M . Y.Y., T I L T O N, E.W., B R O W E R , J.W., Competitiveness of irradiated

adults of the Indian meal moth, J. Econ. Entomol. 69 3 (1976) 349-352.
[19] W O L F E N B A R G E R , D.A., G U E R R A , A.A., Sexual competitiveness of “C o irra­
diated tobacco budworm, J. Econ. Entomol. 65 6 (1972) 1596-1600.
 IAEA-SM-327/56

ESTERILIDAD EN LA Ft
DE Diatraea saccharalis (Fab.),
LEPIDOPTERA: CRAMBIDAE
II. Dinámica de apareamiento y
efectos sobre su descendencia

J.R. G O N Z A L E Z , J.C. G A R C IA
Instituto de Investigaciones de Sanidad Vegetal,
Ministerio de la Agricultura,
Ciudad de la Habana,
Cuba

Abstract-Resumen

F, STERILITY O F Diatraea saccharalis (Fab.), L E P I D O P T E R A : C R A M B I D A E .

II. M A T I N G D Y N A M I C S A N D E F F E C T S O N P R O G E N Y .
The effect of irradiation of Diatraea saccharalis pupae at substerilizing g a m m a doses
on the competitiveness of adult males emerging from the irradiated pupae and on their sterile
progeny was evaluated on the basis of mating dynamics, mating duration and the length of
the pre-mating period, as well as the sex ratio and the variation of pupal weights in the
progeny. It is concluded that substerilizing g a m m a doses do not affect the indicators evaluated
and that in the progeny the sex ratio is altered in favour of males and F, pupal weights are
reduced significantly.

E S T E R I L I D A D E N L A F, D E Diatraea saccharalis (Fab.), L E P I D O P T E R A :

C R A M B I D A E . П. D I N A M I C A D E A P A R E A M I E N T O Y E F E C T O S S O B R E S U
DESCENDENCIA.
Se evaluó el efecto de la irradiación de pupas de Diatraea saccharalis con dosis
subesterilizantes de radiaciones g a m m a sobre la competitividad de los adultos machos
emergidos de las pupas irradiadas y sobre su descendencia estéril, mediante la evaluación de
la dinámica de apareamiento, la duración de la cópula y la duración del período de preaparea-
miento, así como la relación de sexos en la descendencia y la variación de los pasos pupales
en ésta. Se concluye que las dosis de radiaciones g a m m a subesterilizantes no afectan los indi­
cadores evaluados y que en la descendencia se produce una alteración en la relación de sexos
a favor de los machos y una significativa reducción de los pesos pupales en la Fj.

1. IN TR O D U C CIO N

La inducción de esterilidad en la descendencia de lepidópteros irradiados con

dosis subesterilizantes de radiaciones gamma constituye la base del método de
control de plagas denominado esterilidad herada o esterilidad en la Fj [1]. Este
fenómeno ha sido com probado en D. saccharalis [2 -4 ] y a partir de estos resultados

419
420 GONZALEZ y GARCIA

se desarrollaron experimentos para evaluar la competitividad de los adultos irra­

diados con las dosis que dieron lugar a ese efecto. La finalidad de esos experi­
mentos era evaluar la aplicabilidad de los tratamientos en un programa de control
o erradicación de la plaga en cuestión.

2. M A TE RIALES Y M ETO DO S

El material b iológico empleado fueron pupas de más de 5 días de edad,

procedentes de la cría del Instituto de Investigaciones de Sanidad Vegetal cubano,
mantenida con la dieta y m etodología vigentes [5, 6]. Se aplicó la dosis de 300 Gy
fraccionada con dos horas de intervalo entre cada media dosis, y las irradiaciones
se hicieron en una fuente de cobalto 60 con una tasa de dosis de 10 G y/m in y a
una temperatura de 22 a 2 5 °C .
Diariamente se colectaban los adultos emergidos y se colocaban en jaulas de
apareamiento diseñadas y construidas para facilitar las observaciones y para colectar
las parejas formadas durante las mismas. Estas consistían en armazones de alambrón
soldado de 4 mm con forma cilindrica de 40 cm de diámetro y 50 cm de altura,
recubiertas con tela “ marquisette” y provistas de mangas de lienzo para trabajar
en su interior. Todas las observaciones se llevaron a cabo de noche entre las 20.00
y las 8.00 horas del siguiente día, a una temperatura de 22 + 2 °C , con luz roja
para facilitarlas.
En las jaulas so colocaban los adultos recién emergidos en la relación 1:1:1
de machos irradiados, machos no irradiados y hembras no tratadas, empleándose
entre 6y 14 adultos de cada categoría por jaula. En total se evaluaron 164 adultos
de cada tipo en las 17 réplicas del experimento. Para diferenciar los machos de una
u otra categoría durante las observaciones se marcaban éstos con una disolución de
verde brillante al 1 %, alternándose las marcas en las diferentes réplicas del trabajo.
Con los datos tomados durante las observaciones se calcularon, tanto para los
machos irradiados com o para los controles, los siguientes indicadores:

— frecuencias de apareamiento en cada hora,

— duración de las cópulas,
— tiempo que tardaron los machos recién emergidos en aparearse.

C on los resultados se hicieron las distribuciones de frecuencias correspon­

dientes, comparándose éstas mediante pruebas de Ji cuadrado (p = 0,01).
Para la evaluación de la descendencia se tomaron aleatoriamente 9 parejas
donde el macho era irradiado y 7 parejas testigo no tratadas. Se recogieron las masas
de huevos, se incubaron a 30 ± 1°C y 80% de humedad relativa hasta la completa
emersión de las larvas, las cuales se contaron y transfirieron a viales de vidrio de
40 mi de capacidad, con 10 mi de dieta, cubiertos con tapones de algodón y gasa.
Del total de larvas vivas, a los 14 días se tom ó aleatoriamente el 30% y se transfirió
 IAEA-SM-327/S6 421

individualmente a viales con 10 ml de dieta fresca, que se mantuvieron en las

condiciones de temperatura y humedad antes descritas hasta la form ación de las
pupas.
Las pupas fueron colectadas diariamente, sexadas, contadas y pesadas,
comparándose los valores medios calculados mediante el estadígrafo “ z ” para un
1% de probabilidad de error.

3. R ESU LTAD OS Y DISCUSIÓN

En la Fig. 1 se presentan los resultados del estudio de la sincronía de

apareamiento de los machos irradiados y los normales con hembras no irradiadas.
El período de mayor actividad sexual en las condiciones descritas se produjo entre
las 20.00 horas y la 1.30 del siguiente día, existiendo una sincronía casi perfecta
entre ambos tipos de machos. La disminución de la actividad sexual fue coincidente
también, pues en ambos casos ésta se produjo pasadas las 2.00 horas del siguiente
día.
Con relación a las duraciones medias de las cópulas, tampoco se manifestaron
diferencias significativas entre los resultados de los machos irradiados y los no irra­
diados (la Fig. 2 muestra gráficamente estos resultados). Los apareamientos duraron
69,3 ± 12,1 minutos en los controles no tratados y 67,9 ± 1 1 , 9 minutos en las
parejas formadas por los machos tratados y las hembras no tratadas, lo que evidencia
que en este aspecto hay también un comportamiento similar.
En la Fig. 3 se presentan los resultados de la evaluación del tiempo que media
entre la emersión de los adultos y el inicio de las cópulas, cuando esta emersión se
produce de noche. N o se encontraron diferencias significativas en las distribuciones
de los apareamientos, en períodos de tiempo iguales entre los machos irradiados y
los no irradiados, por lo que ambos evidentemente poseen períodos de preaparea-
miento iguales.
Las evaluaciones realizadas por Walker et al. en la década del 60 [7] y por
Contreras Duran en 1980 [ 8] a D. saccharalis no abarcaron los aspectos contempla­
dos en este trabajo, ya que el primero centró su atención en las dosis esterilizantes
y sólo evaluó la mortalidad y las deformaciones en los adultos emergidos de las pupas
irradiadas con dosis esterilizantes, y el segundo en focó su trabajo sólo al aspecto
radiobiológico.
Por otra parte, el Cuadro I presenta los resultados de la evaluación de los
efectos de las radiaciones sobre la descendencia de los adultos irradiados durante el
estado de pupa. Entre éstos destaca la significativa variación de la relación de sexos
a favor de los machos, indicativa de una alta tasa de mortalidad de las hembras, com o
producto de la acción de las mutaciones letales recesivas que afectan preferentemente
a los individúes de ese sexo. El efecto en cuestión minimiza las posibles consecuen-
Frecuencia de
apareamiento

Horas de apareamiento

FIG. 1. Dinámica de apareamiento de los machos irradiados y no irradiados.

Frecuencia
de cópula

IAEA-SM-327/56
Duración media (min)

423
FIG. 2. Distribución de las duraciones medias de las cópulas de las parejas formadas.
424 GONZALEZ y GARCIA

Tiempo (horas)

FIG. 3. Distribución de los inicios de apareamiento de los adultos machos recién emergidos
con hembras no irradiadas recién emergidas.

cias negativas derivadas de la emersión de las hembras en la descendencia, en el

sentido de una hipotética competencia con la población natural.
El Cuadro II presenta los valores prom edio de los pesos de las pupas machos
y hembras de la Fi de los adultos machos irradiados y de los testigo. Tanto las pu­
pas macho com o las hembras de la F, testigo son significativamente más pesadas
que las pupas de los descendientes de los machos irradiados, lo que, evidentemente,
es una consecuencia del efecto de las radiaciones aplicadas a sus padres.
 IAEA-SM-327/56 425

-C U A D R O I. N U M ER O DE ADULTOS OBTENIDOS Y COM POSICION

SEXUAL DE LA D E SCEN D EN CIA F, DE LOS ADULTOS M A CH O S
PROCEDENTES DE PUPAS IR R A D IA D A S C O N 300 Gy DE RAD IAC IO N ES
GAMMA

Parejas Larvas Pupas Pupas Pupas Relación

evaluadas evaluadas macho hembra totales sexual

Machos
9 204 114 31 145 3,7
irradiados
Testigo 7 309 95 103 198 0,9

C U A D R O II. PESO DE LAS PUPAS M A C H O Y H EM BRA DESCENDIENTES

Fj DE LOS A D U L T O S M A C H O S IR R A D IA D O S EN EL E ST A D O D E PUPA

Pupas evaluadas Peso (X ± D.S.) m g

machos hembras Total machos hembras

Parentales
106 31 137 51,56 ± 14,2 81,00 ± 24,5
irradiados
Testigo 65 101 186 59,22 ± 17,7 96,31 ± 18,3

Estos resultados muestran una incidencia de los efectos de las radiaciones sobre
la descendencia de los adultos machos irradiados que se manifiesta a nivel somático,
lo que viene a sumarse a lo ya reportado en cuanto a su alta esterilidad.

4. CONCLUSION ES

La irradiación de pupas macho de D. saccharalis con dosis de 300 G y, según

el sistema de dosis fraccionadas y dos horas de intervalo entre cada media dosis,
no afecta la competitividad de los machos desde el punto de vista de la conducta,
y da lugar a que en la descendencia se manifiesten los siguientes efectos: mutaciones
recesivas ligadas al sexo, y una disminución significativa del peso de las pupas
tanto hembras com o machos en la descendencia estéril.
426 GONZALEZ y GARCIA

R E F E R E N C IA S

[1] La C H A N C E , L.E., “The induction of dominant lethal mutations in insects by ionizing

radiation and chemicals as related to the SIT”, Generation of Insect Vectors of Dise­
ases, Elsevier, Amsterdam (1967) 617-650.
[2] G A R C I A , J.C., G O N Z A L E Z , J.R., de C E S P E D E S , E., Evaluación de los efectos
subesterilizantes de las radiaciones g a m m a en D. saccharalis y su descendencia,
C. Tec. Agrie. Cañ. (en prensa).
[3] S A N F O R D , J.V., Inherited sterility by progeny of irradiated male sugarcane borers,
J. Econ. Entomol. 69 (1976) 456-458.
[4] W A L K E R , D.W., Q U N T A N A , V., Inherited partial sterility among survivors from
irradiation experiments, J. Econ. Entomol. 61 (1968) 318-319.
[5] G O N Z A L E Z , J.R., G A R C I A , J.C., de C E S P E D E S , E., Una nueva dieta artificial
para la cría de D. saccharalis, C. Tec. Agrie. Cañ. 7 2 (1987) 5-16.
[6] G O N Z A L E Z , J.R., G A R C I A , J.C., de C E S P E D E S , E., Metodología para la cría de
D. saccharalis en laboratorio, C. Tec. Agrie. Cañ. 6 2 (1987) 7-15.
[7] W A L K E R , D.W., Q U I N T A N A , V . , P O D O V A N I , F., Effect of g a m m a irradiation on
inmature sugar-cane borers, Proc. Panel Athens, F A O / I A E A (1970) 513-524.
[8] C O N T R E R A S - D U R A N , J.V., Efeitos da radiacao g a m m a no comportamento da broca
da cana de azúcar D. saccharalis, Tesis de grado presentada en la Escola Superior de
Agricultura Luis de Queiroz, Piracicaba, Sâo Paolo (1980) 70.
 IAEA-SM-327/40

PARTIAL STERILIZING RADIATION DOSE EFFECT

ON THE Fj PROGENY OF Spodoptera litura (Fabr.)
Growth, bioenergetics and reproductive competence

R .K . SETH, S.S. SEH GÁL

Department o f Z o o lo g y ,
University o f Delhi,
Delhi,
India

Abstract
P A R T I A L STERILIZING R A D I A T I O N D O S E E F F E C T O N T H E F, P R O G E N Y O F
Spodoptera litura (Fabr.): G R O W T H , B I O E N E R G E T I C S A N D R E P R O D U C T I V E
COMPETENCE.
Partial sterilizing g a m m a radiation doses (4-20 krad) were evaluated with respect to the
pre-imaginal and imaginai behaviour of a serious pest in Indian tropical regions, Spodoptera
litura (Fabr.), with a view to generating more competitive and sterile F, insects to be used
for the control of this pest. A radiosensitivity differential between males and females was
apparent in the P, generation, and male irradiation was considered better than female irradia­
tion for producing behaviourally more viable, but infertile, F, candidates when a moderate
sublethal dose of 13 krad was tested. It was pertinent to assess the viability of F, larvae
before studying adult behaviour. Various growth and developmental characteristics of
F, progeny of the treated moths were studied. A reduction in the growth indices and an
increase in the level of malformations observed in F, progeny were found to be dose depen­
dent. The nutritional profile and energy budget of F, progeny of the irradiated male moths
were ascertained. The irradiation impact was more pronounced in F, females than in
F, males, reflecting the better viability of F, males. Oviposition and egg viability were found
to be negatively correlated with irradiation when F, crosses (self-crosses and out-crosses)
were studied. F, progeny exhibited more sterility than their parent generation. Also,
F, males inherited more sterility with better reproductive competence than F, females.
Higher doses reduced male longevity, whereas the irradiation effect on mating frequency was
not significantly evident. According to the present investigations on F, bioenergetics and
reproductive performance, the feasibility of employing F, sterility by the administration of
substerilizing g a m m a doses to parent moths might be considered with further cautious
appraisal for this pest’ s management.

1. IN TR O D U C TIO N

The use o f radiation as a non-chemical tool has an impetus for controlling

lepidopteran pests, but it requires a modification strategy in the conventional sterile
insect technique. Owing to the enhanced level o f radiation damage and reduced

427
428 SETH and SEHGAL

mating competitiveness caused by the high ionizing doses required for complete
sterilization, the release o f partially sterile insects has been suggested [1 -3 ].
Spodoptera litura (Fabr.), a serious polyphagous pest in the Indian subcontinent, was
reported to have a negative correlation o f radiosusceptibility with age, and fully
sterile insects were found to possess a great deal o f somatic damage and poor
reproductive performance [4, 5]. The w ork reported on here was undertaken to
evaluate the effect o f substerilizing gamma radiation doses on the biology o f the
tobacco caterpillar (S. Litura) with a view to assessing the competence o f
progeny and the potential o f inherited sterility.

2. M A TE R IA L S A N D M ETH ODS

The nucleus culture o f S. litura was maintained on the leaves o f castor, Ricinus
communis, under controlled environmental conditions: 27 + 2 °C temperature,
75 ± 5% relative humidity and a 12 h light: 12 h dark regimen. The irradiation
facilities at the Institute o f Nuclear M edicine and Allied ¡Sciences, Delhi, were used
and the required doses o f gamma irradiation were applied to 0 to 24-h-old adult
moths from a “ C o source at a dose rate o f 780 rad/min . 1 Reproductive pairing o f
the moths was conducted in perspex-nylon cages. Adult moths were fed a 20%
honey solution and a fresh castor leaf was used as the ovipositional trap. Various
reproductive cháracteristics such as the total number o f eggs oviposited, the percent­
age egg hatch, the mating frequency (number o f spermatophores/mated fem ale), the
percentage mating and the longevity o f both sexes were observed for the treated and
normal crosses. The corrected sterility and control o f reproduction 2 were computed
in response to the reproductive performance o f the treated insects and their Fj
progeny involved in different in-crosses and out-crosses.
Growth and development were studied in terms o f the percentage pupae and
adult formation, the developmental period, the growth index, the degree o f morpho­
genetic deformities and the sex ratio. The nutritional profile and energy budget o f
sixth instar Fj S. litura larvae o f the treated male moths were evaluated [ 6, 7].
The data were subjected to appropriate statistical analyses for comparison o f
the means (x + standard error) computed from ten replicates [ 8, 9].

1 1 rad = 1.00 x 10“

2 Gy.
(Y _ vj
2 Control of reproduction is given by — ------ x 100, where V c denotes the
? Vc
viable number of eggs in the control and V, the viable number of eggs in the treated insects.
 SETH and SEHGAL

100

90

80

70

60

50

40

30

20

10

Set I Set II
(Progeny of T or) (Progeny of T Ç )
(at13krad) (at13krad)
2. ’production o f S. litura in the F¡ progeny o f irradiated male and female adults
• /> п I
? ; ■: egg hatch).
 IAEA-SM-327/40 431

3. RESULTS AND DISCUSSION

The reproductive behaviour o f S. litura treated as 0 to 24-h-old moths was

observed (Fig. 1). The adverse effect o f irradiation on the reproductive potential was
found to be greater for females than for males, as has also been pointed out in studies
o f S. frugiperda [1, 10]. Initially, the radiosensitivity o f male and female S. litura
regarding inherited sterility was assessed with respect to one gamma dosage o f
13 krad; the male moths were found to be better candidates for producing more
sterile and competitive F[ progeny (Fig. 2). H ence, studies on the substerilizing
dose range (4 -2 0 krad) were pursued for S. litura treated as male moths.
It was imperative to study the Fj viability in relation to growth, fo o d utiliza­
tion and energy efficiencies o f the larval stage, which might reflect the behavioural
potency o f F! adults. Table I shows the various growth and development charac­
teristics o f S. litura in the F ! progeny o f treated males. The percentage pupation at
4 krad was 79.2 ; it was reduced to 40% at 20 krad. Likewise, the effect o f treatment
was apparent on adult production, causing 69 .1 , 59 .6 , 49.3 , 36.4 and 2 0 .9 % adult
formation at 4, 7, 13, 16 and 20 krad, respectively. The post-em bryonic develop­
mental period o f F! progeny was delayed, resulting in a reduction in the growth
index, which exhibited a negative relationship with the gamma dosage. M orpho­
genetic deformities affecting the mouth parts, legs and wings were observed and the
malformation intensity was determined for male and female Fj moths. For instance,
a gamma dose o f 13 krad resulted in about 50% adult formation, with 21% mal­
formation in Fj males and 25% malformation in F] females. The sex ratio was
observed to be skewed in favour o f males in the F¡ progeny o f treated moths; this
effect was significant at 7 krad and higher doses.
The influence o f substerilizing doses on the dry matter nutritional profile, as
a function o f the physiological response to irradiation stress, was studied (Table П).
The phagoperiod (PP) was increased to 3.25 d in males and 3.47 d in females at a
dose o f 20 krad compared with the untreated moths (1.81 d in males and 1.94 d in
females). The growth rate (G R ) and consumption index (C l) values were higher in
normal males than in normal females, unlike the weight gain, food balance, approxi­
mate digestibility (A D ), efficiency o f conversion o f ingested food (ECI) and effi­
ciency o f conversion o f digested fo o d (E C D ), which were found to be higher in
unirradiated females than in unirradiated males. Interestingly, the impact o f irradia­
tion was m ore marked in F¡ females than in Fj males for all the nutritional
parameters studied. For instance, the G R was reduced from 0.580 at 0 krad to 0.322
at 20 krad in F] males, whereas it was 0.567 at 0 krad and decreased to 0.288 at
20 krad in Fj females. .
The radiosensitivity o f F! males and females was also ascertained in terms o f
the energy budget (Table 1П). The gross energy in food balance o f unirradiated
Fj insects was 1551 cal in males and 1765 cal in females, and showed a negative
correlation with irradiation, which resulted in a reduction in energy in food balance
432 SETH and SEHGAL

up to the level o f 25% in F] males and 30% in Fj females at 20 krad .3 The energy
stored in the body, the coefficient o f metabolizable energy (C M E ), the efficiency o f
storage o f ingested energy (ESI(E)) and the efficiency o f storage o f metabolizable
energy (ESM (E)) were determined in unirradiated and irradiated F] insects; the
adverse effects o f irradiation were more pronounced in F[ females than in F! males.
For instance, at 20 krad the gross energy stored in the body, ESI(E)) and ESM (E),
was reduced by 48, 38 and 31% in F, males compared with 58, 48 and 4 1 % ,
respectively, in Ft females with respect to the control.
In the progeny o f treated males, reproductive competence was studied in three
different crosses: F, cr X N Ç , N cr X F t Ç and F t cr X F! Ç (Table IV ). The
fecundity and fertility were affected by irradiation and the radiosusceptibility
increased at higher doses. The number o f eggs oviposited by normal females
(crossed with F , males) and F¡ females (crossed with normal males) was reduced in
relation to the ovipositional response elicited in a parental cross, but the irradiation
impact o f these out-crosses was less than that observed in F[ self-crosses. The
fertility measured in terms o f egg viability was markedly affected in the F t progeny
in comparison to that exhibited by parents treated with the substerilizing doses. The
effect o f radiation on Fj fertility was apparent in groups treated with doses as low
as 4 krad. A dose o f 13 krad, which caused about 50% fertility, leading to
39% sterility in the treated males, could induce 72 .4 % sterility in F! males and
56.5% sterility in F! females, whereas the sterility shown by F! self-crosses at this
dose was about 75 % . At a higher dose o f 16 krad, the inherited sterility in Fj males
was 78.8% and in F, females 6 6 .4 % , whereas Fj self-crosses at 16 krad exhibited
82.6% sterility. W hile comparing a particular pairing cross (e.g. F , cr x N Ç )
between radiation treatments, the sterility o f the Fj progeny increased with the
increase in radiation dose administered to the male parent. In other words,
F, progeny o f the 13 krad treated males were more sterile than the progeny o f
7 krad treated males. H ow ever, mating was more fertile when the F! females were
out-crossed than when the F! males were out-crossed. The percentage mating was
higher in F[ males than in F! females and a significant decrease in mating was
observed at higher doses. For instance, the mating success in F[ males was 75% at
13 krad and 52% at 20 krad. In F, progeny, the mating frequency was not signifi­
cantly affected, although male longevity was affected from 16 krad onwards.
Generally, the proportion o f non-productive Fj pairs increased as the radia­
tion dose increased. This could be attributed to the reduction in percentage mating
or to an insufficiency or absence o f sperm transfer, or failure o f the sperm to fertilize
the egg or a combination o f all three; this remains to be determined by further sperm
studies. The inherited sterility was o f a greater degree in the progeny o f the treated
males than in the treated Pj moths, and more competitive F, progeny with a sex

3 1 cal = 4.184 J. Text cont. on p. 440

T A B L E I. EFFECT OF G A M M A IR R A D IA T IO N O N T H E G R O W TH A N D D E V E L O P M E N T A L C H AR AC TER ISTIC S OF
S. litura IN F! PROGENY D ER IV ED FROM Р,Тст X N ? CROSS

Adult Developmental Malformation (% )

Dose Pupation formations period (d) Growth index Sex ratio
(krad) (% ) (% ) (first instar-adult) (N /D ) male: female
Male Female
(N) (D)

0 91.66 ± 3 .0 83.9a ± 2.95 22.78a ± 0.56 3.68a ± 0.05 5 .6 6.71 1:1.03

IAEA-SM-327/40
(control) ± 0 .6 0 ± 0 .7 1

4 79.2 ± 2 .0 69.16b ± 2.79 23.4ab ± 0.54 2.97b ± 0.22 16.6 • 19.0 1:0.85
± 1 .9 6 ± 2 .0 1

7 69.9 ± 2 .4 59.65c ± 2.25 24.32ab ± 0.62 2.45bc ± 0.12 17.91 2 3 .0 1:0.69

± 2 .0 2 ± 2 .1 1

13 56.7 ± 3.1 49.35d ± 2.20 25.09b ± 0.37 1.96c ± 0.04 21.09 25.01 1:0.57
± 1 .5 2 ± 1 .3 2

16 48.31 ± 2 .6 36.4e ± 2.13 25.31b ± 0.11 1.43d ± 0.07 32.6 40.22 1:0.38
± 2 .2 2 ± 1 .9 2

20 40.05 ± 2 . 1 1 2 0 .95f ± 2.29 25.98b ± 0.58 0.80e ± 0.03 39.15 14.50 1:0.35
± 2 .2 8 ± 2 .5 6

Note: Means followed by the same letter in columns are not significantly different (P < 0 .0 5 ; Duncan’ s multiple range test).

433
 434
T A B L E II. EFFECT OF G A M M A IR R A D IA T IO N ON T H E N U T R IT IO N A L PR O FILE OF S. Litura IN F j PROGENY
D E R IV E D FR O M T H E P,To- X N ç CROSS (FOR E X P L A N A T IO N OF A B B R E V IA T IO N S , SEE T E X T )

Food Weight Food Mean

Dose PP Faeces AD ECI ECD
ingested gain balance weight GR Cl
(krad) (d) (g) (% ) (% ) (% )
(g) (g) (g) (g)

Males

SETH
0 1.81 0.3744 0.1817 0.0807 0.1925 0.0777 0.580 2.74 51.77 21.57 44.31
+ 0 .0 6 ± 0 .0 2 0 ± 0 .0 1 8 ± 0 .0 0 5 ± 0 .0 1 4 ± 0 .0 0 4 ± 0 .0 2 4 ± 0 .1 8 ± 2 .1 9 + 0 .7 7 ± 3 .0 1

and SEHGAL
4 2.47 0.3696 0.1968 0.0802 0.1727 0.765 0.432 1.96 45.41 22.42 50.97
± 0 .1 2 ± 0 .0 2 9 ± 0 .0 1 1 ± 0 .0 0 4 ± 0 .0 2 1 + 0 .0 0 2 ± 0 .0 2 5 ± 0 .1 2 ± 1 .8 7 ± 1 .0 5 ± 3 .5 9

7 2.78 0.3499 0.2013 0.769 0.1485 0.0740 0.385 1.77 43.04 2 1 .96 52.92
+ 0 .1 5 ± 0 .0 2 1 ± 0 .0 1 9 ± 0 .0 0 5 ± 0 .0 1 1 ± 0 .0 0 5 ± 0 .0 2 2 ± 0 .1 2 ± 2 .9 3 ± 0 .7 2 ± 3 .4 6

13 2.87 0.3399 0.2227 0.0701 0.1 170 0.0703 0.362 1.75 34.54 20.92 61.80
± 0 .1 4 ± 0 .0 1 4 ± 0 .0 1 2 ± 0 .0 0 1 ± 0 .0 0 7 ± 0 .0 0 5 ± 0 .0 2 4 ± 0 .1 3 ± 1 .9 3 + 1.03 ± 3 .8 6

16 3.18 0.3395 0.2012 0.0615 0.1382 0.0567 0.344 1.85 40.85 18.73 47.35
± 0 .0 8 ± 0 .0 2 6 ± 0 .0 1 8 ± 0 .0 0 3 ± 0 .0 1 2 ± 0 .0 0 2 + 0 .0 2 0 ± 0 .0 6 ± 2 .7 6 ± 1 .4 9 ± 4 .8 0

20 3.25 0.3122 0.1693 0.0577 0.1427 0.0554 0.322 1.79 45.36 18.57 42.29
± 0 .1 0 ± 0 .0 1 4 ± 0 .0 0 8 ± 0 .0 0 3 ± 0 .0 1 1 ± 0 .0 0 4 ± 0 .0 2 6 ± 0 .0 9 ± 2 .5 1 ± 1 .0 8 ± 3 .2 6
TABLE II. (cont.)

Food Weight Food Mean

Dose PP Faeces AD ECI ECD
ingested gain balance weight GR CI
(krad) (d) (g) (% ) (% ) (% )
(g) (g) (g) (g)

Females

0 1.94 0.4167 0.1930 0.1018 0.2236 0.0923 0.567 2.37 53.20 25.27 50.11

IAEA-SM-327/40
± 0 .0 3 ± 0 .0 2 4 ± 0 .0 1 4 ± 0 .0 0 6 ± 0 .0 2 0 ± 0 .0 0 5 ± 0 .0 1 8 ± 0 .1 8 ± 2 .9 3 ± 2 .1 3 . ± 3 .9 8

4 2.6 4 0.3903 0.2029 0.0928 0.1873 0.0866 0.409 1.68 . 47.57 24.97 53.87
± 0 .1 1 ± 0 .0 2 9 ± 0 .0 1 4 ± 0 .0 0 3 ± 0 .0 1 6 ± 0 .0 0 2 ± 0 .0 1 8 ± 0 .0 8 ± 1 .9 4 ± 2 .2 9 ± 3 .8 9

7 3.11 0.3614 0.2246 0.0856 0.1367 0.0830 0.332 1.41 37.53 23.95 65.53
± 0 .0 9 ± 0 .0 1 6 ± 0 .0 0 9 ± 0 .0 0 5 ± 0 .0 1 0 ± 0 .0 0 5 ± 0 .0 1 4 ± 0 .0 7 ± 1 .6 2 ± 1 .4 9 ± 3 .1 2

13 3.20 0.3593 0.2073 0.0819 0.1519 0.0774 0.328 1.45 42.65 22.59 53.82
± 0 .1 2 ± 0 .0 2 8 ± 0 .0 1 9 ± 0 .0 0 7 ± 0 .0 1 1 ± 0 .0 0 5 ± 0 .0 1 3 ± 0 .0 4 ± 1 .8 4 ± 0 .5 2 ± 2 .9 7

16 0.21 0.3547 0.2157 0.0625 0.1390 0.0617 0.321 1.78 38.89 18.05 46.76
± 0 .1 3 ± 0 .0 3 2 ± 0 .0 1 8 ± 0 .0 0 4 ± 0 .0 1 5 ± 0 .0 0 4 ± 0 .0 1 7 ± 0 .0 6 ± 1 .7 5 ± 0 .9 5 ± 2 .3 3

20 3.47 0.3332 0.1902 0.0588 0.1429 0.0602 0.288 1.62 42.63 18.05 43.26
± 0 .1 4 ± 0 .0 1 4 ± 0 .0 0 6 ± 0 .0 0 2 ± 0 .0 0 9 ± 0 .0 0 2 ± 0 .0 1 9 ± 0 .0 8 ± 1 .1 7 ± 1 .2 7 ± 4 .0 1

435
 436
T A B L E III. EFFEC T OF G A M M A IR R A D IA T IO N ON TH E ENERG Y BU D G ET OF S. litura IN F j PROGENY
D E R IV E D FR O M T H E P,Tc7 X N ? CROSS (FOR E X P L A N A T IO N OF A B B R E V IA T IO N S , SEE T E X T)

Gross energy in Gross energy in , Gross energy in Gross energy

Dose CME ESI(E) ESM(E)
food ingested faeces food balance , stored in body
(krad) (% ) (% ) ‘ (% )
(cal) (cal) (cal) (cal)

Males

SETH
0 2301 750 .1551 524 67.40 22.77 33.78
±80 ±44 ±42 ±15 ± 2 .5 7 . ± 0 .8 6 ± 3 .7 0

and SEHGAL
4 2271 840 1431 513 63.01 22.58 35.84
±58 ±22 ±52 ±10 ± 2 .1 1 ± 1 .1 8 ± 2 .1 1

7 . 2150 880 1270 434 59.06 20.18 34.17

±52 ±47 ±37 ±13 ± 2 .1 3 ± 0 .6 2 ± 1 .1 1

13 2089 997 1092 350 52.27 16.75 32.05

±35 ±31 ±37 ±11 ± 1 .9 8 ± 0 .9 1 ± 1 .0 0

16 2086 921 1165 305 55.84 14.62 26.18

±49 ±46 ±31 ±17 ± 2 .2 4 ± 1 .1 2 ± 2 .0 2

20 1918 760 1158 270 60.37 14.07 23.31

±35 ±26 ±25 ±17 ± 1 .3 0 ± 0 .8 0 ± 1 .7 7
TABLE Ш. (cont.)

Gross energy in Gross energy in Gross energy in Gross energy

Dose CME ESI(E) ESM(E)
food ingested faeces food balance stored in body
(krad) (%) (%) (%)
(cal) M ) (cal) (cal)

Females

0 2561 796 1765 661 68.91 25.81 37.45

IAEA-SM-327/40
±89 ±35 ±53 ±22 ±2.22, ±2.42 ±4.06

4 2398 866 1532 594 63.88 24.77 38.77

±71 ±35 ±40 ±17 ±2.17' ±1.96 ±3.46

7 2221 982 1289 484 58.03 21.79 37.54

±40 ±31 ±35. ±13 ±1.60 ±1.32 ±1.09

13 2208 . 9 28 1280 409 57.97 18.52 31.95

±71 ±49 ±25 ±19 ±1.81 ±0.31 ±1.10

16 2179 988 1191 310 54.65 14.22 26.02

±40 ±46 ±38 ±14 ±1.97 ±0.76 ±1.30

20 2047 808 1239 ' 276 60.52 13.48 22.27

±35 ±53 ±29 ±13 ±1.25 ±0.78 ±2.06

437
T A B L E IV. EFFECT O F G A M M A IRRA D IA TIO N ON THE REPRO D U CTIV E COM PETEN CE OF S. litura (T R E A TE D AS £
Pj A D U L T S ) IN TH E F , GE N E RA TIO N 00

Fertility Life span (d) Corrected Control of

Dose Fecundity Mating Mating
Cross* (egg hatch) sterility reproduction
(krad) (eggs/female) frequency percentage Male Female
(%) (*) (%)

0 Ncr x N 9 1991 ± 76 85.4a ± 2.4 1.79 ± 0.13 94 ± 3.1 12.05 10.71 0 0

(control) +0.41 ±0.67
4 P,Tcr x N 9 1915 ± 68 73.96b ± 2.3 2.02 ± 0.16 83 ± 2.9 11.47 10.35 9.8 13.10
+0.50 ±0.34
Fjcr x N 9 1985 ± 94 61.27c ± 2.0 2.19 ± 0.22 85 ± 2.6 11.66 10.91 26.74 26.90 Vi
я
N ct x F,9 1655 ± 85 67.35bc ± 3.1 1.82 ± 0.21 78 ± 35
+0.37
11.08
±0.35
11.11 18.75 33.70 a
H

±0.48 3
±0.59 a
F, cr x F,ç 1568 ± 69 61.71c ± 1.8 1.70 ± 0.29 76 ± 9.2 11.30 9.75 27.40 39.75 ел
и
±0.61 ±0.53 BE
О
7 P,Ttf x N 9 1742 ± 69 64.90bc ± 2.6 2.29 ± 0.15 77 ± 2.5 10.70 10.14 20.85 30.64 t
±0.33 ±0.44
F,cr x N 9 1501 ± 70 49.22d ± 2.5 1.99 ± 0.24 76 ± 4.4 10.91 9.98 42.09 56.38
±0.31 ±0.71
Nor x F ,9 1344 ± 72 59.90c ± 2.2 1.80 ± 0.11 68 ± 1.9 11.64 9.91 27.74 51.70
±0.74 ±0.13
Fj o- x F, 9 1308 ± 70 47.lid ± 2.8 1.76 ± 0.19 62 ± 5.8 10.91 10.14 44.57 61.63
±0.71 0.73
13 Р,Тст x N 9 1592 ± 57 49.90d ± 2.3 1.98 ± 0.11 71 ± 2.4 10.27 9.68 39.14 51.26
±0.53 ±0.41
F,cy x N 9 1329 ± 59 23.39fg ± 1.8 1.96 ± 0.23 75 ± 3.5 10.80 9.29 72.48 81.62
±0.40 ±0.51
TABLE IV. (cont.)

Fertility Life span (d) Corrected Control of

Dose Fecundity Mating Mating
Cross* (egg hatch) sterility reproduction
(krad) (eggs/female) frequency percentage Male Female
(%) <%) (%)

Ncr x F, 9 1278 ± 142 35.98e ± 2.2 1.60 ± 0.33 62 ± 5.3 10.78 10.04 56.59 72:41
±0.35 ±0.31
F, a x F, Ç 1085 ± 100 21.26fg ± 2.1 1.85 ± 0.47 59 ± 7.6 9.85 10.01 74.98 85.63
±0.96 ±0.44

16 P,Tcr X N 9 1502 ± 35 40.21e ± 3.1 1.95 ± 0.18 68 ± 3.9 10.39 9.41 50.96 62.94
±0.36 ±0.57

IAEA-SM-327/40
F,cr x N 9 1311 ± 69 17.99g ± 2.1 1.89 ± 0.22 62 ± 4.4 10.50 9.59 78.85 86.06
±0.33 ±0.52
N ct x F, 9 1252 ± 67 27.80f ± 1.5 1.71 ± 0.17 54 ± 3.8 9.98 9.99 66.46 79.12 .
±0.17 ±0.41
F,cr x F ,9 1014 ± 37 14.7lg ± 1.2 2.Ó1 ± 0.63 51 ± 6.6 0.09 8.96 82.69 90.71

20 P ,T с
уX N 9 1251 ± 53 25.19f ± 2.3 2.36 ± 0.17 60 ± 4.0 9.49 9.43 69.28 83.75
±0.53 ±0.40
F ,cr x N 9 1215 ± 89 2.98h ± 0.4 1.61 ± 0.24 52 ± 5.1 9.90 9.50 96.49 97.86
±0.35 ±0.28
N o 1 x F, 9 1092 ±. 113 12.25g ± 0.6 1.33 ± 0.29 40 ± 7.1 9.95 9.53 85.22 91.97
±0.13 ±0.13
F íO" x F, 9 831 ± 94 0 1.60 ± 0.16 39 ± 5.8 9.06 9.22 100 100
±0.50 ±0.77

Note: Means followed by the same letter in the same test are not significantly different (P < 0.05; Duncan’
s multiple range test).

439
a Pi: parental generation; F,: first filial generation of T CT x N ç ; N: normal; T: treated.
440 SETH and SEHGAL

ratio skewed in favour o f males in the present findings are in accordance with the
observations made in other lepidopteran species [2, 11, 12]. Interestingly, F, males
showed better competence in terms o f larval and adult behaviour than F, females.
Further critical evaluations need to be conducted, both in the laboratory and in
the field, on the competence o f F! progeny, to select an appropriate dose and
to optimize the strategy o f using inherited sterility for the suppression o f this
lepidopteran pest.

REFERENCES

[1] KNIPLING, E.F., Suppression of pest Lepidoptera by releasing partially sterile males:
A theoretical appraisal, BioScience 20 (1970) 465-470.
[2] N O R T H , D.T., Inherited sterility in Lepidoptera, Annu. Rev. Entomol. 20 (1975)
167-182.
[3] C A R P E N T E R , J.E., et al., Fall armyworm (Lepidoptera:Noctuidae): Inheritance of
gamma-induced deleterious effects and potential for pest control, J. Econ. Entomol. 76
(1983) 378-382.
[4] SETH, R.K., Evaluation of Growth and Reproductive Bioactivities of Spodoptera
litura vis-à-vis G a m m a Irradiation and its Bioefficacy Modified by Alternate Host
Plants, P h D Thesis, Department of Zoology, Delhi University, Delhi (1985).
[5] SETH, R.K., S E H G A L , S.S., “Efficacy of g a m m a rays in modifying behavioural
traits of Spodoptera litura and their evâluation in F, generation”, paper presented at
18th Int. Congress on Entomology, Vancouver, BC, 1988.
[6] W A L D B A U E R , G .P ., The consumption and utilization o f fo o d by insects, A d v . Insect
Physiol. 5 (1968) 229-288.
[7] B H A T T A C H A R Y A , A.K., P A N T , N.C., Studies on host plant relationships:
Consumption and utilization profile in insects, Proc. Natl. Acad. Sci. India, Part В46
I, II (1976) 273-301.
[8] S N E D E C O R , G.W., C O C H R A N , W.G., Statistical Methods, Iowa State University
Press, Ames, IA (1967).
[9] D U N C A N , D.B., Multiple range and multiple F-tests, Biometrics 11 (1955). 1-42.
[10] C A R P E N T E R , J.E., et al., Fall armyworm (Lepidoptera:Noctuidae): Comparison of
inherited deleterious effects in the progeny from irradiated males and females, J. Econ.
Entomol. 79 (1986) 46-49.
[11] L A C H A N C E , L.E., Genetic Methods for the Control of Lepidopteran Species: Status
and Potential, Rep. ARS-28, Agricultural Research Service, United States Department
of Agriculture, Washington, D C (1983).
[12] M A S T R O , V.C., S C H W A L B E , C.P., “Status and potential of F, sterility for control
of noxious Lepidoptera”, M o d e m Insect Control: Nuclear Techniques and Biotech­
nology (Proc. Symp. Vienna, 1987), IAEA, Vienna (1988) 15-40.
 IAEA-SM-327/41

MATING BEHAVIOUR OF THE MELON FLY

Sexual selection and sperm competition

Y . ITÔ*
Laboratory o f Applied Entomology and Nem atology,
Nagoya University,
Chikusa, Nagoya

M . Y A M A G IS H I, H. KU BA
Okinawa Fruit Fly Eradication Project O ffice,
Naha, Okinawa

Japan

A bstract

M A T I N G B E H A V I O U R O F T H E M E L O N FLY: S E X U A L S E L E C T I O N A N D S P E R M
COMPETITION,
Melon fly females show strong mate choice and, after continuous mass releases of
sterile males, a genetic fraction, which leads females to reject the courtship of mass reared
males, will increase (a sterile insect technique (SIT) resistant strain). To prevent this increase,
as large a number of sterile males as possible should be released from the start of SIT. Last
male sperm precedence exists mainly due to the short lifespan of melon fly sperm. A long
duration of copulation is necessary for males to inhibit female remating, which reduces the
fitness of the first male. This information is important for the future use of SIT for insect
species that mate many times.

1. IN TR O D U C TIO N

The melon fly, Bactrocera (form erly Dacus) cucurbitae, is an important pest
o f cucurbit vegetables and many tropical fruits. A large project using the sterile
insect technique (SIT) to eradicate this species from the Southwestern Islands
(Okinawa and Amani), Japan, has been under way since 1972 and, as Yamagishi
et al. [ 1 ] report, almost complete success has been achieved.
Three topics are presented here which have been developed from our melon
fly project:

(1) The existence o f mate ch oice among melon fly females, and its ‘evolution’ in
nature during mass releases;

* Present address: Faculty of Science and Arts, Okinawa University, 555 Kokuba,
Naha, Okinawa 902, Japan.

441
442 ITÔ et al.

TAB LE I. M AJOR CONDITION S N E CE SSAR Y FO R THE SUCCESSFUL

E RA D IC ATIO N OF IN SECT PESTS USING SIT

(1) Mass rearing without a decrease in the vigour o f released males.

(a) Phenotypic effects: Decrease in vigour due to poor nutrition, high density, etc., in
mass rearing facilities.
(b) Genetic effects: Changes in behaviour because o f genetic changes that took place
during continued mass rearing (‘domestication’)

(2) Sterilization without a decrease in the vigour o f released males.

(3) No economic damage due to the released adults.

(4) Isolated distribution o f the target species.

(2 ) T w o aspects o f sperm competition in the melon fly, namely, the absence o f last
male sperm precedence and the inhibition o f female remating;
(3) The importance o f mating behaviour studies for improvements in SIT. and fruit
fly control.

Table I shows the major conditions that are necessary for the successful eradi­
cation o f a pest using SIT. For the melon fly, the second item mentioned above was
achieved; the harmful effect o f sterilization with gamma ray irradiation on the sur­
vival and dispersal o f males was negligible.
Item 1 above consists o f two aspects, phenotypic and genetic effects. In the
melon fly, the latter is far m ore important than the former (see, for example,
Ref. [2]). In the case o f the melon fly, damage to crops from released adults is
negligible and the target area, the Southwestern Islands, is very well isolated.

2. FEM ALE M A T E CHOICE A N D ITS EVO LU TIO N

Figure 1(a) [3, 4] shows the changes in mating competitiveness o f sterile

melon fly males under field conditions on K um e-zim a, from where the fly was eradi­
cated in 1977.
Evaluation o f the mating competitiveness o f mass reared and sterilized males
is usually carried out in laboratory cages. H ow ever, the results often do not reflect
mating competitiveness under field conditions. Iwahashi et al. [3] developed a
method which was used to measure field competitiveness in the Kume project.
B efore the start o f mass releases, the competitiveness value was moderately high and
acceptable (0.83 ), but the value decreased over the life o f the project. Near the end
o f the project, after the flies had been reared for 18 generations in the mass rearing
 IAEA-SM-327/41 443

C ommencement

CO
Q.
"D
<D
(5
e
о
CD
O)
S
С
a>
о
ф
CL

log (volume of space (cm3 )/fly)

FIG. L (a) Mating competitiveness o f mass reared and sterilized melon fly males measured
under field conditions (modified from Ref. [3]) ( о ; data taken from Kudaka-zima; • : data
taken from Kume-zima; ▲: competitiveness values measured in laboratory cages, (b) Rela­
tionship between the size o f the cage per fly and the percentage o f successful mating o f males
o f mass reared (33-34 generations ( о )) and wild ( • ) strains when caged together with mass
reared females (from Ref [4]).
444 ITÔ et al.

laboratory, the competitiveness value fell to only 0.2. These results possibly
originate from genetic effects. H owever, at this time the competitiveness values
obtained in laboratory cages were still high (see triangles in Fig. 1(a)).
Soemori et al. [4] reported the results o f experiments that suggest an explana­
tion for the discrepancy between laboratory and field data. They released individu­
ally marked flies into cages or rooms o f different sizes and recorded the matings in
the evening. Figure 1(b) shows the percentage o f released males that mated in rela­
tion to the size o f the experimental area. Males o f the wild strain could not mate well
in a small space, but males o f the laboratory strain could mate.
Iwahashi and Majima [5] found that melon fly males formed leks for mating.
In the evening, males aggregated on a non-host tree where each male established his

Experiment 1 -> Escape ($) ------- > Reject (?)

32 1
Courtship 39 7
male wing
Courtship —> vibration -----> Mount --------- > Copulation
type 37 5 4
45 7 0
Encounter-----
O-male 51
R-male 66
Non-courtship
type ------------- (Omitted)

Experiment 2 i- > Escape ($) -> Reject ($)

40 4
Courtship 52 10
male wing
Courtship - vibration Mount -> Copulation
type 47 7 3
62 10 0
Encounter ■
O-male 54
R-male 77
Non-courtship
type ■(Omitted)

FIG. 2. Mate choice by wild melon fly females collected on Okinawa-Hontô (Okinawa Island,
О females) to wild males (O males) and males of a mass reared strain (R males). The arrows
indicate the direction in which the behavioural sequence proceeds. The upper numerals indi­
cate the frequencies o f the transitions when О females encountered О males, while the lower
numerals indicate cases when О females encountered R males (redrawn from Refs [7, 8]).
 IAEA-SM-327/41 445

-> Escape ($) -------> Reject (?)

39 9
Courtship 38 15
male wing
Courtship —> vibration ♦ Mount > Copulation
type 51 12 3
57 19 4
Encounter
l-male 52
R-male 62
Non-courtship
type (Omitted)

FIG. 3. Mate choice by wild melon fly females, collected on Ishigaki-zima (I females), to
lshigaki wild males (I males) and mass reared males (R males) (from Ref [8]). For a full
explanation, see Fig. 2.

territory, usually on a single leaf. They then secreted sex pheromone to attract
females. When a female arrived, a territorial male courted her with special m ove­
ments and sounds. How ever, the female often rejected the courtship and flew away;
a female usually accepted copulation after visits to the territories o f several males.
This is evidence o f female mate ch oice in insects, which has been demonstrated only
in a few other animals, despite Darwin’ s [ 6] earlier suggestion.
Hibino and Iwahashi [7] compared the mating success o f males o f a wild and
a mass reared strain. Figure 2 [7, 8] shows the results o f field cage experiments
using flies o f a wild strain o f Okinawa-Hontô ( ‘ О flies’ ). The figure shows that even
when courted by wild males (O males), wild females (O females) accepted copula­
tion in only 4 o f 37 courtship trials, or 3 o f 47 trials. It is notable that О females
never accepted courtship from mass reared (R ) males. This may be caused by differ­
ences in the courtship song [9].
When Hibino and Iwahashi carried out these experiments, the melon fly on
Okinawa-Hontô was near extinction as a result o f the eradication project. Therefore,
О flies were subjected to strong selection pressure by the released sterile males.
Hibino and Iwahashi [ 8] carried out a similar experiment using wild flies taken
from Ishigaki-zima, where there had been no SIT programme. Figure 3 [ 8] shows
that females o f the Ishigaki wild strain (I females) accepted courtship from R males,
as well as by I males. Table II [ 8] shows the mean duration o f wing vibration from
the time o f the m ale-fem ale encounter to the escape o f the female. In О females there
was a significant difference between the duration o f encounters with О males and
with R males, but in I females the difference was insignificant.
There are two possible explanations for the differences in behaviour between
flies from the two islands. The first is that О flies and I flies had genetically different
446 ITÔ et al.

TAB LE II. M E A N AND STANDARD D E V IA T IO N OF D U RA TIO N (IN

SECONDS) OF W IN G VIBR ATIO N FRO M ENCO U N TER TO ESCAPE O R
M O U N T W H EN О FEM ALES ENCO U N TERED О M ALES O R R M ALES (I),
A N D W H EN I FEM ALES EN COUN TERED I M A LE S O R R M ALES (П).
N IN DICATES THE N U M BER OF SAM PLES [ 8]

Wing vibration from encounter to

Type of encounter
Escape N Mount N

(I)
0 females to 0 males 60 ± 105 40 146 ± 149 4
0 females to R males 25 ± 24 52 39 ± 31 10
Mann-Whitney U test P < 0.05 P < 0.05

(П)
I females to I males 50 ± 45 39 93 ± 76 9
I females to R males 50 ± 47 38 101 ± 117 15
Mann-Whitney U test NS NS

NS: not significant

courtship and acceptance characters and the courtship character o f R males was more
similar to that o f I males. The second is the ‘ SIT resistance hypothesis’ , which states
that although the range o f the wild male courtship character was narrow, the wild
female population was heterogeneous and contained individuals who accepted a
broad range o f male courtship characters. H ow ever, females that accepted sterilized
R male courtship could not reproduce. Thus, under strong selection pressure from
SIT, a female genotype that accepted the courtship o f mass reared males may have
becom e extinct.
T o test the tw o hypotheses, Hibino and Iwahashi [ 8] carried out an experiment
using О females and О and I males. There were no differences between the two,
which suggests that the mate ch oice o f О females changed to a rejection o f the court­
ship o f R males after being artificially selected by SIT. Thus, the authors made an
interesting discovery: evolution o f mate choice which had not been demonstrated
before in animals.
How can this problem be overcom e? Tsubaki and Bunroongsook [10] con ­
ducted a simulation experiment to estimate the effect o f the change in mate choice.
 IAEA-SM-327/41 447

Their mode] is an expansion o f a logistic model developed by Itô [11]. They showed
that the effect o f a reduction in mating competitiveness o f mass reared and sterile
males is far more important than the effect o f a change in female mate ch oice, and
that releases o f two to three times m ore sterile males than could be released in a non­
mate choice model can eradicate the target insect.

3. L A ST M A L E SPERM PRECEDENCE

Because melon fly males insert free sperm into the fem ales’ spermathecae, and
many females copulate more than once, sperm mixing may take place. Thus, sperm
competition might be an important problem for SIT.
T o examine this possibility, we used sterile males to measure the P2 values,
an index o f the last male sperm precedence in the melon fly. Figure 4 shows the
results obtained by Yamagishi et al. [12]. When the interval between the first and
second matings was short (1 -4 d ), there were no significant differences in the hatch-
ability o f eggs between the first norm al-second sterile (NS) and first sterile-second
normal (SN) matings. H ow ever, when the interval was 16 d or m ore, hatchability
was significantly higher when the second mating was normal than when the second
mating was sterile. Thus, last male sperm precedence could be detected
(P 2 > 0 .5 ). T w o processes, the consumption o f sperm during fertilization and the
outflow o f sperm at the time o f fertilization, or sperm mortality, could be a reason
for this change.
As melon fly females do not oviposit without stimuli from host fruit, oviposi­
tion can be manipulated by either placing or not having a pumpkin slice in the rearing
cups. In Fig. 4 , NS1 and SN1 are experiments where females oviposited during the
period between the first and second matings because we placed pumpkin slices in the
cups, while NS2 and SN2 are experiments where the females could not oviposit
because no pumpkin was present. There were no significant differences between the
hatchability values o f NS1 and NS2 or SN1 and SN2, suggesting that it was not
sperm loss owing to fertilization and the related processes, but sperm mortality that
was the major cause o f differences in P2 values.
Figure 5(a) is a sperm survival curve showing a decrease in the number o f
sperm in the spermathecae o f non-ovipositing females [13]. The curve in Fig. 5(b)
shows the change in P2 values expected from the survival curve [12]. The circles
and triangles are the observed P2 values. These values were about 0.5 (no last male
precedence) for short mating intervals, but increased with the increase in mating
intervals. There was no significant difference between the expected and observed
values ( x 2 test, P > 0.05).
Based mainly on knowledge o f the long lifespan o f sperm in the spermathecae
o f social insects, entomologists have often considered that sperm may survive for a
448 ITÔ et al.

ñ . . .* i i .i . l l i

100

50 .... 48SN2
— 48 NS2
1mi
L .U - l I l- i- U o lU - ]

-25. 25 50
Days after second mating

FIG. 4. Mean hatchability of eggs laid by melonflyfemales mated with normal (N) and sterile
(S) males (from Ref. [12]). NS: Mated first with sterile and then with normal males; SN: first
sterile and second normal mating. The numbers on the graphs indicate the mating interval,
in days. N SI, SN1: Females that could oviposit after thefirst mating. NS2, SN2: Females that
could oviposit only after the second mating.
 IAEA-SM-327/41 449

Intervals between two matings (log of days)

FIG. 5. (a) Change in the number of sperm infemale spermathecae without oviposition (from
Ref. [13]). (b) P 2 values in relation to intervals between two matings. The curve shows the
expected values calculated from the sperm survival rate shown in (a) (from Ref. [12]).

long time in spermathecae (see, for example, papers in Ref. [14]). The Yamagishi
et al. report [12] is the first case where the short lifespan o f sperm is the main reason
for the high P2 values.

4. INHIBITION OF FEMALE REMATING

In the melon fly, sperm transfer is completed after 4 h o f copulation, but melon
fly copulations continue for more than 8 h. To examine the reason for this long dura­
tion o f copulation, Kuba and Itô [15] studied the effects o f the duration o f copulation
450 ITÔ et al.

,0-0-0- o-o-o-cr
.o-o-o-cr0-
0- 0
м- ■
/
17 •
> A -A -A -A
/
Duration of
Type of males copulation (h)
Mated normal 8
Virgin normal 8
Mated sterile 3
Mated sterile 8
Virgin sterile 3
Virgin sterile 8
I___ I
10 15 25
Days after first copulation

FIG. 6. Cumulative remating curves of melon fly females mated with virgin and previously
mated normal and sterile males. Thé mated sterile males lack sperm (from Ref. [14]).

on the inhibition o f female remating. Virgin normal males, virgin sterile males and
normal and sterile males that had already mated four times were used. Gamma ray
irradiation not only induces dominant lethal mutations in sperm which exists in the
testes o f melon fly pupae, but it also completely destroys spermatogonia. Therefore,
sterile males that have mated three or four times have no sperm, and are good sub­
jects for the evaluation o f the effects o f sperm.
Figure 6 shows changes in the cumulative percentage o f remated females that
copulated the first time for 3 or 8 h with different types o f males. Three hours o f
copulation did not inhibit female remating, but 8 h did. Copulation not only with vir­
gin sterile males, but also with spermless sterile males inhibited female remating at
the same rate as normal matings. Our recent experiment showed that 4 h o f copula­
tion did not inhibit female remating, but 6 and 8 h did.
In Drosophila, although secretion from male accessory glands is the main fac­
tor that inhibits female remating, the existence o f sperm in spermathecae also inhibits
female remating. In the melon fly, however, sperm is not an important factor inhibit­
ing remating. We suggest that the transfer o f an accessory gland substance after 4 h
o f copulation is the main factor inhibiting female remating. Determining the impor­
tance o f the accessory gland substance and its chemical structure, asChen et al. [16]
did for Drosophila, are the next subjects o f our study.

5. CONCLUSIONS

Melon fly researchers in Japan have developed studies which are important not
only for future improvement o f sterile insect release programmes and in fruit fly con­
 IAEA-SM-327/41 451

trol, but also important in understanding the evolution o f sexual selection. The
results given in this paper show that more basic ecological and ethological studies
are necessary to improve the genetic control o f insect pests.

REFERENCES

[1] Y AM AG ISH I, М ., K AK IN O H AN A , H ., K U B A , H ., K O H A M A , T ., N A K A M O TO ,
Y ., SOKEI, Y ., KINJO, K ., Paper IA E A -S M -327/4, these Proceedings.
[2] N AK AM O R I, H ., Behavioral and ecological studies on sexual competitiveness in the
melon fly, Dacus cucurbitae Coquillett, mass production, Bull. Okinawa Pref. Agrie.
Exp. Station Suppl. No. 2 (1988) 1-64.
[3] IW AH ASH I, O ., ITÔ, Y ., SH IYOM I, М ., A field evaluation o f the sexual competi­
tiveness o f the sterile melon fly, Dacus (Zeugodacus) cucurbitae, Ecol. Entomol. 8
(1983) 4 3 -4 8 .
[4] SOEMORI, H ., TSU KAG U CHI, S ., N A K A M O R I, H ., Comparison o f mating ability
and mating competitiveness between mass-reared and wild strains o f the melon fly,
Dacus cucurbitae Coquillett (Diptera: Tephritidae), Jpn J. Appl. Entomol. Zool. 24
(1980) 146-250.
[5] IW AH ASH I, O ., M AJIM A, T ., Lek formation and male-male competition in the
melon fly, Dacus cucurbitae Coquillett (Diptera: Tephritidae), Appl. Entomol. Zool.
21 (1986) 7 0 -7 5 .
[6] D AR W IN , C ., The Descent o f Man, and Selection in Relation to Sex, Appleton,
New York (1871).
[7] HIBINO, Y ., IW AH ASH I, O ., Mating receptivity of wild type females for wild type
males and mass-reared males in the melon fly, Dacus cucurbitae Coquillett (Diptera:
Tephritidae), Appl. Entomol. Zool. 24 (1988) 152-154.
[8] HIBINO, Y ., IW A H ASH I, O ., Appearance o f wild females unreceptive to sterilized
males on Okinawa Island in the eradication programme o f the melon fly, Dacus cucur­
bitae Coquillett (Diptera: Tephritidae), Appl. Entomol. Zool. 26 (1991) 265-270.
[9] K A N M IY A , K ., N A K A G A W A , K ., T A N A K A , A ., K A M IW A D A , H ., Comparison
o f acoustic properties o f tethered flight sounds for wild, mass-reared, and irradiated
melon flies, Dacus cucurbitae Coquillett (Diptera: Tephritidae), Appl. Entomol. Zool.
22 (1987) 85 -9 7.
[10] TSUBAKI, Y ., BUNROONGSOOK, S., Sexual competitive ability o f mass-reared
males and mate preference in wild females: Their effects on eradication o f melon flies,
Appl. Entomol. Zool. 25 (1990) 45 7-46 6.
[11] ITÔ, Y . , A model o f sterile insect release for eradication o f the melon fly, Dacus cucur­
bitae Coquillett, Appl. Entomol. Zool. 12 (1977) 303-312.
[12] Y AM AG ISH I, M ., ITÔ, Y ., TSUBAKI, Y ., Sperm competition in the melon fly, Bac-
trocera cucurbitae (Diptera: Tephritidae): Effects o f sperm “ longevity” on sperm
precedence, J. Insect Behav. 5 (in press).
[13] TSUBAKI, Y ., YAM A G ISH I, М ., “ Longevity” o f sperm within the female of the
melon fly, Dacus cucurbitae (Diptera: Tephritidae), and its relevance to sperm compe­
tition, J. Insect Behav. 4 (1991) 24 3-25 0.
452 ITÔ et al.

[14] SMITH, R .L. (Ed.), Sperm Competition and the Evolution o f Animal Mating Systems,
Academic Press, London (1984).
[15] KU BA, H ., ITÔ, Y ., Remating inhibition in Bactrocera ( = Dacus) cucurbitae (Díp­
tera: Tephritidae): Copulation with spermless males inhibits female remating, J. Ethol­
ogy 10 (1992) (in press).
[16] CHEN, P .S., et al., A male accessory gland peptide that regulates the reproductive
behavior of the female D. melanogaster, Cell 54 (1988) 29 1-29 8.
 IAEA-SM-327/42

FIRST FIELD ASSESSMENT OF THE DISPERSAL

AND SURVIVAL OF MASS REARED STERILE
MEDITERRANEAN FRUIT FLY MALES OF
AN EMBRYONAL, TEMPERATURE SENSITIVE
GENETIC SEXING STRAIN
J. HENDRICHS*, V. WORNOAYPORN*,
B.I. KATSOYANNOS**, K. GAGGL*

* FAO/IAEA Entomology Unit,

Agency’ s Laboratory Seibersdorf,
International Atomic Energy Agency,
Vienna

** Laboratory o f Applied Zoology and Parasitology,

University o f Thessaloniki,
Thessaloniki,
Greece

Abstract

FIRST FIELD ASSESSMENT OF THE DISPERSAL A N D SU RVIVAL OF MASS

REARED STERILE MEDITERRANEAN FRUIT FLY MALES OF A N EM BR YON AL,
TEMPERATURE SENSITIVE GENETIC SEXING STRAIN,
Functional temperature sensitive lethal (tsl) genetic sexing strains o f the Mediterranean
fruit fly (medfly) have been recently developed at the Agency’ s Laboratory at Seibersdorf.
Unlike previous medfly genetic sexing strains, these ‘ second generation’ sexing strains are
stable under mass rearing conditions and allow females to be killed at an early stage of
development. Over the past year, successful efforts to mass rear these strains have shown that
they are amenable to routine large scale rearing with only minor adjustments in éxisting
production facilities. In addition, results of detailed greenhouse behavioural assessments have
confirmed that the behaviour and mating activity o f tsl males is comparable overall not only
to that o f males o f the pupal colour sexing strain that was successfully used in the field, but
also to that o f wild males. Field assessment o f the most promising tsl strains is the next step
in the evaluation o f these flies. Before embarking on large pilot evaluations, however, one
more aspect o f tsl male behaviour had to be tested: dispersal and survival under orchard condi­
tions. The results o f the first preliminary field assessment o f mass reared sterile tsl males
carried out in citrus orchards in Chios, Greece, in September 1992, are reported. Sterile
tsl males were recaptured during all ten days o f trapping after the initial release. Even though
the numbers recaptured declined rapidly, mortality rates were comparable with similar
dispersal studies of bisexual medfly strains. Furthermore, and in spite o f windy weather,
tsl males dispersed not only downwind, but also upwind more than the minimum o f 100 m
required to cover the 200 m distance between flight lines during conventional aerial sterile fly
releases. Based on the results obtained to date, the first pilot test using tsl males will take place
in the oasis o f Tozeur, Tunisia, starting in 1993.

453
454 HENDRICHS et al.

1. IN TROD U CTION

The benefits o f releasing only males in fruit fly control programmes based on
the sterile insect technique (SIT) are manyfold [1]. However, pupal colour strains,
the only genetic sexing strains o f the Mediterranean fruit fly (medfly), Ceratitis
capitata (Wied.), currently available, are not applicable in mass rearing facilities
because females must be reared to the pupal stage and mechanical separation o f
pupae by colour is inexact, slow and expensive.
Recently, temperature sensitive lethal ( tsl) genetic sexing strains o f the medfly
have been developed at the Agency’ s Laboratory at Seibersdorf [1]. Unlike the previ­
ous sexing strains, these ‘ second generation’ sexing strains are (a) stable under mass
rearing conditions and (b) allow females to be killed at an early (embryonal) stage.
This represents an important breakthrough because both o f these attributes were con­
sidered indispensable for genetic sexing strains with any potential to replace bisexual
strains in sterile medfly production facilities.
Over the past year, the main procedures developed to mass rear these tsl strains
have proved amenable to routine large scale rearing, with only minor adjustments
in existing facilities. Standard mating tests, as well as detailed greenhouse
behavioural assessments, have confirmed that the behaviour and mating activity o f
tsl males is comparable with that o f the pupal colour sexing strain (successfully used
in a large pilot trial in Israel [2]), as well as with that o f wild males (unpublished
data).
Field assessment o f the most promising tsl strains is the next step in the evalua­
tion o f these flies. Pilot trials o f tsl strains are planned for various sites in the
Maghreb countries o f North Africa. Starting in 1993, the first o f these pilot tests is
planned for the oasis o f Tozeur, in southern Tunisia. Before embarking on large pilot
evaluations, however, one more aspect o f tsl male behaviour had to be assessed,
namely dispersal and survival under orchard conditions. W e report here on the
results o f the first preliminary assessment o f the dispersal and survival o f mass reared
sterile tsl males under field conditions.

2. MATERIALS AND METHODS

The field study was carried out during the first half o f September 1992 on the
island o f Chios, Greece. Minimum daily temperatures ranged between 16 and 21 °C
and maximum temperatures between 27 and 33 °C. During the short time available
it was only possible to carry out one test lasting ten days. Typical o f the Mediterra­
nean region, property sizes on Chios are small and orchards are interspersed between
houses and gardens. Under these conditions, the best site available for a sufficiently
large dispersal area included various contiguous citrus orchards belonging to differ­
ent owners (Fig. 1). These were separated from each other by 3-5 m high stone
 IAEA-SM-327/42 455

NW

Predominant
winds

Traps
Release point
Citrus orchards
Houses
Open fields and
gardens

Streets
3 - 5 m high
stone walls

100 m 210 m

FIG. 1. Distribution of a grid of 35 Jackson traps in various contiguous citrus orchards in

Chios, Greece, to assess the dispersal and survival of sterile males of tsl strain Vienna-42
(1-61) under field conditions.

walls, which apparently protect trees from strong winds. The citrus fruits, mostly
orange and mandarin, were largely at an immature stage. A 30 m X 30 m grid o f
35 Jackson traps was distributed over the central orchards and a few traps were
placed inside the outer orchards (Fig. 1).
The males released originated from tsl strain Vienna-42 (T(Y;5)1-61), which
had been held for nine generations iinder mass rearing conditions at the Agency’ s
Laboratory at Seibersdorf. To eliminate female embryos, eggs were subjected to a
heat treatment o f 32°C for 48 h o f incubation in aerated water. Thereafter, they were
seeded at a rate o f 1.4 mL/kg o f larval diet. Pupae o f the first three larval collection
456 HENDRICHS et al.

days were irradiated in air one day before emergence with a dose o f 90 Gy. Subse­
quently, they were marked with a fluorescent dye and packed in anoxia in 1.0 L plas­
tic bags and immediately shipped in a cardboard container to Chios. The total
transport time from Vienna to Chios was 12 h. Upon arrival at Chios, the pupae were
placed in paper bags at a rate o f approximately 25 mL o f pupae/release bag. The
paper bags had previously been painted internally with sugar and provided with rest­
ing surfaces for emerging males. The bags were held under field conditions and, two
to three days after emergence, they were opened during the afternoon to release the
sterile males from one central release point. Control paper bags were retained to
determine the average per cent o f emergence, which was 86.5. Based on these data,
approximately 240 000 males were released.
Traps were first installed 23 h after release o f the flies. Then, during each o f
ten days, the traps were installed and removed after 1 h between 1700 and 1800 h.
As in previous dispersal studies [3], this short exposure period was an attempt to
ensure that the traps themselves would not bias the movement o f the males. Each
trap was installed and removed sequentially in the same order in the predetermined
grid pattern so that all traps were exposed for 1 h daily. Initially, traps were baited
with fresh, slow release trimedlure plugs. After each trap removal, the inserts
covered with Tanglefoot were observed under ultraviolet lamps. The head o f each
unmarked fly was squashed for signs o f fluorescent dye and the number o f marked
and unmarked males was recorded. The traps were provided with new sticky inserts
on the following day, just before installation.
Starting with the fly release and in parallel with the entire trapping period, con­
trol males were held in a rectangular open mesh field cage ( l m x 0 . 5 m x 0.5 m).
They were held under orchard conditions and provided with water and a mixture o f
sucrose and yeast hydrolysate. The number o f flies in the field cage corresponded
to one paper bag. Dead males in the cage were counted and removed daily. At the
end o f the ten day trapping period, the number o f males remaining in the cage was
recorded.

3. RESULTS

The population o f wild flies remained relatively constant throughout the trap­
ping period, despite the near absence o f mature fruit in the citrus orchards (Table I).
Most o f these wild flies originated apparently from sour oranges which had ripened
the previous month. Only during the first trapping day were wild fly captures high,
a normal situation at the initiation o f trapping with a dense trap grid. The tsl males
were recaptured in the citrus orchards during each o f the ten days o f trapping follow­
ing the initial release (Table I). The numbers recaptured, however, declined very
rapidly (Fig. 2). As a result, only on the first two days o f trapping were their num­
bers higher than those o f wild flies (Table I). The daily rate o f disappearance o f tsl
 IAEA-SM-327/42 457

TABLE I. AVERAGE DAILY CAPTURE ( ± S D a) OVER TEN DAYS OF WILD

AND STERILE MALES OF tsl STRAIN VIENNA-42 (1-61) CAUGHT IN A
TRAP GRID OF 35 JACKSON TRAPS IN CITRUS ORCHARDS IN CHIOS,
GREECE

Average No. of Average No. of

Day recaptured tsl males captured wild males Ratio L :W C
per trap per hourb per trap per hourb

1 127.8 (171.8) 70 .4 (83.2) 1.815

2 119.6 (131.7) 45.9 (40.5) 2.606

3 26.7 (23.5) 43.5 (42.0) 0.614

4 9 .0 (13.8) 43.1 (48.6) 0.209
5 3.7 (4.0) 43.7 (40.5) 0.085
6 1.8 (2.0) 34.6 (32.0) 0.052

7 1.4 (1.6) 47.1 (35.2) 0.030

8 0.5 (0.7) 46.9 (30.6) 0.011
9 0.5 (0.8) 50.0 (30.3) 0.010
10 0.2 (0.4) 56.1 (37.7) 0.004

a SD: standard deviation.

b Traps were deployed only 1 h per day.
c L: Laboratory; W : wild.

Days since release

FIG. 2. Comparison of sterile males of tsl strain Vienna-42 (1-61) recaptured in Jackson
traps up to ten days after release with control males of the same strain and shipment surviving
infield cages with food and water. In the equation N, = Nge'^, ц is the rate of disappear­
ance (mortality and emigration, or only mortality in the case of the cages), and N 0 the esti­
mated initial population, i.e. not the total released population, but rather the fraction of the
population equal to thefractional trapping rate (---------- : field trap catches;---------- : field cage
control).
458 HENDRICHS et al.

TABLE П. RATIO OF DOWNWIND TO UPWIND CAPTURES

(IN RELATION TO THE RELEASE POINT) OF STERILE MALES OF tsl
STRAIN VIENNA-42 (1-61) AND WILD MALES OVER TEN DAYS IN CITRUS
ORCHARDS IN CHIOS, GREECE

Ratio o f downwind to upwind captures

Day
Wind speed
Sterile tsl
(Beaufort scale)
males E Wild males2

1 1.81 2.57 4

2 1.16 1.12 5

3 1.20 1.23 6

4 2.48 1.63 6

5 0.65 1.39 7

6 0.44 0.93 5

7 0.34 1.18 3

8 0.83 0.83 3

9 0.55 0.96 2

10 0.45 0.97 3

a No significant interaction (P = 0.64) strain versus trapside over time (two way analysis of
variance (A N O V A)).

males, either due to mortality or emigration, was 0.68. This rate was at least ten
times higher than that o f the flies in the control field cage (0.06). However, in the
latter case the rate includes only mortality and not emigration or recapture.
Continuous winds from the northwest prevailed during the whole ten day
recapture period following release. Wind speeds ranged from a light or gentle breeze
(2 or 3 on the maritime Beaufort scale) up to a moderate gale (7 on this scale =
50-61 km/h). The tsl males.dispersed in spite o f the wind not only downwind, but
also upwind. In fact, the ratio o f downwind to upwind recaptures decreased through­
out the study, and from days five to ten, a majority o f recaptures occurred on the
upwind side o f the initial central release point (Table П). The ratio o f downwind to
upwind captures for wild flies followed a similar pattern.
The percentage distribution o f average recaptures per trap at various distances
from the release point is shown in Fig. 3. Although tsl males were recaptured in all
traps deployed starting from the first day o f trapping, only 0.6% o f recaptures per
trap occurred initially in the most distant traps compared with 43 % recaptured at the
 IAEA-SM-327/42 459

60
Day 2

40

20
о.
to
к»
*—»
© 60
Day 4
CL

ф . 40
СО
Е 20

■о
О
Day 6
3 60
О.
со
Ф 40

20

Day 8
Ъ 60
ъ
Ф 40
О)

с 20
0)
о
(D
0-
60

40

20

0 30-45 60-75 90-105120-150 215-260

Distance from release point (m)

FIG. 3. Percentage distribution of recaptured sterile males of tsl strain Vienna-42 (1-61) per
trap and distance from the release point.

central release site. However, the percentage o f recapture per trap at the release site
declined rapidly to 16 on day four, 9 on day six, and finally 0 on days eight and ten.
At the outlying traps on the other hand, the percentage o f recapture per trap increased
to 13 on day four, 30 on day six, 47 on day eight and finally 66 on day ten.

4. DISCUSSION

To date, a number o f very thorough field assessments o f the dispersal and sur­
vival o f sterile medflies have been carried out in different regions o f the world [3-8].
460 HENDRICHS et al

Although these studies did not involve male only releases, they all report rapid
declines in the numbers o f released sterile flies. On average, the last recaptures o f
released medflies in these studies have been on day eight, when released populations
were monitored until they had reached a practically undetectable level. This rapid
disappearance o f flies occurred even where the recapture area covered a 1 km2 non­
host orchard [3], and fly emigration was insignificant, as evidenced by the few flies
that reached the outer perimeter o f the traps.
In our dispersal study, the rate o f disappearance was again high, although tsl
males were recaptured during each o f the ten trapping days. The percentage o f recap­
ture o f tsl males was approximately 4, which is considerable, taking into account that
the traps were deployed daily for only 1 h in the afternoons. Not all o f the high rate
o f disappearance o f tsl males from the trapping area was due to mortality within the
release area. The percentage o f males recaptured in the most distant traps in the out­
lying orchards increased throughout the study (Fig. 3). In addition, tsl males
probably disperse farther than males o f bisexual strains. In various field studies
[2, 8, 9], it was shown that sterile males range much wider in search o f wild females
when released alone than when they are released together with females. This advan­
tage o f sexing strains, where only sterile males are released, results in considerable
increases in the effectiveness o f SIT [9].
Although emigration and recapture account for some fly losses, they do not
account for most o f the tenfold difference in survival between the tsl males in the
orchard and those in the protected control field cage. Unlike released tsl males, tsl
males in the control cage had access to food and water. The dissection o f recaptured
tsl males showed, however, that they had the same type o f light green dense sub­
stance in their crops as wild males, probably an indication that sterile males were
finding and feeding at the same natural food sites as wild males.
Probably a more important factor to consider here is mortality due to predation
as a considerable number o f cases o f predations on the released flies were noted in
the release area. Particularly apparent was predation by yellowjacket wasps (Vespula
germánico) [10]. This may account to a large extent for the enormous difference in
survival between the tsl males in the orchard and those in the protected control field
cage. During observations o f sterile medflies released on a mango tree, Baker and
van der Valk [11] also observed rapid declines in fly numbers immediately after
release. They found no evidence that the physical effect o f heavy rain might reduce
numbers. However, although it was not measured, they suggested that heavy preda­
tion, including by wasps, might account for the precipitous fall in fly numbers.
We quantified our observations o f predation losses using a field cage
(2.2 m high X 3 m diameter) with a fruit bearing citrus tree. A release bag contain­
ing tsl males was opened within the field cage. The entrance o f the field cage
remained open (approximately 0.5 m wide) to allow access to foraging wasps. On
each o f the three occasions this study was replicated [12], wasps entered the open
cage in large number's and within 5-7 h killed all o f the more than 1000 flies that
 IAEA-SM-327/42 461

had been released. Under natural conditions, the escape o f the flies would have been
less constrained than in the cage, and predation losses would probably have been less
drastic. However, the study confirmed that predation losses can be very large and
suggested .that predaceous wasps are the most likely explanation for the much more
rapid decline in the numbers o f males released in the field than in the closed control
field cage.
In another study within the same field cage (unpublished data), we recorded
the foraging o f individual wasps to compare predation on wild and sterile males. We
found that although wild males suffer less, they still suffer considerable losses from
wasp predation. This is also confirmed by the observation that a large majority o f
the wild males captured live for behaviour studies after the first trapping day were
still sexually immature. However, sterile males were on average three to four times
as likely as wild males to be captured by wasps [12, 13]. Particularly vulnerable to
predation were young tsl males and those engaged in pheromone calling and lekking
activities. In view o f the high predation mortality, and because the reproductive
behaviour o f released sterile males is still increasing in intensity three days after the
normal 3-d-old release, there would be advantages to releasing flies that are older
than those currently released (Ref. [11]; Hendrichs et al., unpublished data).

5. CONCLUSIONS

The results o f this field test should be considered to be preliminary as only one
release was carried out and the recapture area was not sufficiently large to distinguish
accurately between mortality and emigration. Even so, findings on the survival and
dispersal o f tsl males are comparable with the most thorough field evaluations o f
bisexual strains reported in the literature. Flies seemed able to locate food and water
and to respond to trimedlure. Furthermore, and in spite o f windy weather, they dis­
persed not only downwind, but also upwind more than the minimum o f 100 m
required to cover the 200 m distance between flight lines during conventional aerial
sterile fly releases. Larger field trials using these strains seem warranted, although
consideration should be given to releasing sterile males that are older than those
currently released to reduce predation losses before the flies attain their full sexual
maturity.

ACKNOWLEDGEMENTS

The authors wish to thank L. Lang and H. Gensthaler for technical help, H.
Baumgartner for preparing the figures, and R. Gingrich, W. Klassen, D .A . Lind­
quist and J. Richards for supporting this work. Also thanked is E. Katsoyannos for
her hospitality and S. Kokkinakis, I. Kokkinakis, A. Melekos and G. Maurogiannis
for allowing us to work in their orchards.
462 HENDRICHS et al.

REFERENCES

[1] F R A N Z , G ., K E R R E M A N S , P., Paper IA E À -S M -327/20, these Proceedings.

[2] N IT Z A N , Y ., RÓSSLER, Y ., ECO NO M O PO U LO S, A .P ., Paper IA E A -S M -327/70,
these Proceedings.
[3] P L A N T , R .E ., C U N N IN G H A M , R .T ., Analyses o f the dispersal o f sterile M editerra­
nean fru it flies (Diptera: Tephritidae) released from a point source, Environ. Entomol.
20 (1991) 1493-1503.
[4] B A K E R , P.S., C H A N , A .S .T ., Identification o f tephritid fru it fly dispersal. Guidelines
fo r a sterile release programme, J. A pp l. Entomol. 112 (1991) 410-421.
[5] W O N G , T .T .Y ., et al., Mediterranean fru it fly : Dispersal o f w ild and irradiated and
untreated laboratory-reared males, Environ. Entomol. 11 (1982) 339-343.
[6] B A K E R , P.S., C H A N , A .S .T ., J IM E N O -Z A V A L A , M .A ., Dispersal and orientation
o f sterile Ceratitis capitata and Anastrepha ludens (Tephritidae) in Chiapas, J. Appl.
Ecol. 23 (1986) 27-38.
[7] SERG HIO U, C ., S Y M M O N S , P., Ecology o f the Mediterranean fru it fly in Cyprus,
United Kingdom M in istry o f Overseas Development-Cyprus M in istry o f Agriculture
(1974).
[8] B L O E M , K .A ., B L O E M , S., C H A M B E R S , D .L ., Field assessment o f quality:
Release-capture o f mass-reared Mediterranean fru it flies (Diptera: Tephritidae) o f
different sizes, Environ. Entomol. (in press).
[9] M cIN N IS , D .O ., T A M , S., G R AC E, C ., M IY A S H IT A , D ., Population suppression
and sterility rates induced by variable sex-ratio, sterile-insect releases o f Ceratitis
capitata (Diptera: Tephritidae) in H aw aii, Ann. Entom ol. Soc. A m . (in press).
[10] H E N D R IC H S , J., K A T S O Y A N N O S , B ., W O R N O A Y P O R N , V ., Odour-mediated
foraging by yellowjacket wasps (Hymenoptera: Vespidae): Predation on leks o f
pheromone-calling Mediterranean fru it fly males (Diptera: Tephritidae), Oecologia
(submitted fo r publication).
[11] B A K E R , P.S., V A N D E R V A L K , H ., D istribution and behavior o f sterile M editerra­
nean fru it flies in a host tree, J. Appl. Entomol. 114 (1992) 67-76.
[12] H E N D R IC H S , J., et al., Field cage comparison o f predation losses due to foraging
yellowjacket wasps, Vespula germanica L. (Hymenoptera: Vespidae) in lekking w ild
and sterile mass reared males o f an embryonal, temperature sensitive genetic sexing
strain o f the Mediterranean fru it fly , Ceratitis capitata W ied. (Diptera: Tephritidae) (in
preparation).
[13] H E N D R IC H S , J., et al., Rapid quality control method to measure predator avoidance
in w ild and mass reared males o f an embryonal, temperature sensitive genetic sexing
strain o f the Mediterranean fru it fly (unpublished).
 IAEA-SM-327/43

RECENT PROGRESS IN THE DEVELOPMENT

OF ATTRACTANTS FOR MONITORING
THE MEDITERRANEAN FRUIT FLY
AND SEVERAL Anastrepha SPECIES

R.R. HEATH, N.D. EPSKY

Insect Attractants, Behavior and Basic Biology Research Laboratory,
Agricultural Research Service — South Atlantic Area,
United States Department o f Agriculture,
Gainesville, Florida,
United States o f America

Abstract
RECENT PROGRESS IN THE DEVELOPM ENT OF ATTR ACTAN TS FOR M ONITOR­
ING THE M EDITERRANEAN FRUIT F L Y A N D SEVERAL Anastrepha SPECIE S.
Recent developments in the analytical methods used for collecting and analysing
semiochemicals that are attractive to fruit flies are reviewed. These efforts have resulted in
increased knowledge o f the nature o f the pheromonal components released by males o f the
Mediterranean fruit fly (medfly), Ceratitis capitata (Wiedemann), and males o f several
Anastrepha species. A recently developed bioassay system is described that permits the testing
o f various substrates for biological activity in a flight tunnel, while simultaneously collecting
a portion o f the volatiles from the attractive source for subsequent chemical identification and
quantification. This system is currently being used to re-examine the attractiveness o f various
protein based baits for pest fruit flies. It was found, for example, that an increase in the pH
of the protein bait Nulure significantly increases the attraction o f female medflies to this bait.
Similarly, pheromone research using the described systems for the bioassay and collection of
attractive pheromonal components from the medfly and from several Anastrepha species are
under way, and the results from this research are presented.

1. INTRODUCTION

The availability o f effective lures for monitoring fruit flies is o f paramount

importance for control and eradication programmes. Control o f the Mediterranean
fruit fly (medfly), Ceratitis capitata (Wiedemann), and several Anastrepha spp. is
o f considerable economic importance worldwide. Currently used lures to monitor
for medfly include trimedlure [1] and hydrolysed protein baits [2]. Although
trimedlure is effective in attracting male medflies, it is only weakly attractive to
female medflies. Protein baits used in glass McPhail traps attract both male and
female medflies and several Anastrepha species; however, this monitoring method
is cumbersome to deploy and the effectiveness o f these traps is reportedly low [3].

463
464 HEATH and EPSKY

Thus, there is considerable interest in developing improved lures, especially

synthetic lures, that would attract female fruit flies.
This paper reports on recent developments both in analytical methods used for
collecting semiochemicals that are attractive to fruit flies and in bioassay methods
used for determining the attractiveness o f these chemicals in flight tunnel tests. Also
presented are the results o f recent efforts to improve protein based attractants and
to develop synthetic pheromone attractants for the medfly and several Anastrepha
spp. fruit flies.

2. METHODS TO COLLECT AND BIO ASSAY FRUIT FLY ATTRACTANTS

2.1. System to collect fruit fly semiochemicals

To fully understand insect chemical communication systems and to maximize

the potential development and use o f volatile semiochemicals produced by insects
and plants, the chemicals emitted must be collected and analysed both qualitatively
and quantitatively. The collection process must be conducted in a manner that will
minimize both the stress placed on the organism being studied, as well as the collec­
tion o f unwanted background organic compounds from the surrounding environ­
ment. An illustration o f the system recently designed by us is shown in Fig. 1 and
a complete description o f the system has been published [4]. Unique to this system
is that inlet air is purified by passage through three 7-cm-diameter, 0.5-cm-thick
charcoal filter discs. The system consists o f four main components: the air filtration
unit, the volatile collection unit, the collector trap and the multiport collector base.
The volatile collection chamber is constructed o f glass and the tubing in the system
is stainless steel to minimize contamination. Manual switching for the collection o f
volatiles on one o f the three collector traps or purge is accomplished using a four
way, multiport valve. A flow meter set at 1 L/min is used to control the amount o f
air pulled by a vacuum pump through the collector traps or to purge the system.
Volatiles are collected on traps using Super-Q® (Altech Assoc. Inc., Deer­
field, IL, USA) as the adsorbent. Super-Q traps are prepared by packing approxi­
mately 20 mg o f the adsorbent in 4 cm long x 4.0 mm i.d. glass tubes and the
material is contained between two stainless steel frits.

2.2. Flight tunnel bioassay system

A bioassay system has been developed in our laboratory that permits the testing
o f various substrates for biological activity in a flight tunnel, while simultaneously
collecting a portion o f the volatiles from the attractive source for subsequent chemi­
cal identification and quantification. The bioassay system consists o f four main sub­
components designed for specific tasks and capable o f varying certain parameters
 IAEA-SM-327/43 465

FIG. 1. A system designed for the collection of volatiles released from plants. The system.
components are as follows: an air filtration system (AFS), a multiport collector base (MCB),
a volatile collection chamber (VCC) and a volatile collector trap (VCT). ;

between experiments (Fig. 2). These components include the bioassay flight tunnel,
the air flow system and air delivery system, the insect isolation trap and, finally, the
volatile collection system. The air flow system contains sheets o f cloth that have been
infused with charcoal material, as described for the volatile collection system. This
material purifies the air entering the tunnel, which facilitates pheromone collection
at various points inside the flight tunnel with minimal contamination.
This system was used to determine the periodicity o f response o f female med­
flies to the pheromone released by male flies. Studies conducted with female factory
and feral medflies indicated that these flies were highly responsive to odours from
live males in the flight tunnel bioassay, with an average o f 50-75% o f the tested
females trapped. The periodicity o f response is shown in Fig. 3. The system
described should be o f general utility in the determination o f the attraction o f pest
fruit flies to suspected attractants, and is currently being used to test the response
o f females to synthetic pheromone blends.

3. PROTEIN BAIT ATTRACTANTS

Prior to the isolation o f the ‘ food’ attractant chemicals from various protein
sources, it is paramount to determine what chemical conditions are required for
attractiveness. Although considerable efforts have been made on improving protein
 To exhaust blower
ADS AFS HT VCS

FIG. 2. The wind tunnel bioassay system, which is comprised of a bioassay flight tunnel (BFT), an air flow system (AFS), an air delivery
system (ADS), two insect isolation traps (IITs), a multiport collector base (MCB), a volatile collection system (VCS) and a volatile collector
trap (VCT).
 IAEA-SM-327/43 467

i— — i
9-1 0 10—11 11-1.2 12—1 . 2-3
Time of bioassay (h)

FIG. 3. The percentage offemale medflies responding to volatiles from five male medflies
during the day. The lines indicate the standard error (n = 12).

baits, these efforts have been largely empirical. W e recently determined that the
attractiveness o f Nulure (Miller Chemical and Fertilizer C o., Hanover, PA, USA)
for medflies can be ‘programmed’ on the basis o f pH. In tests conducted in
Guatemala, we compared solutions o f torula yeast plus borax pellets, 10% Nulure
plus 1, 3, 5 and 10% borax, and a control (water) in McPhail traps. Every 2-3 d,
flies were removed from the traps, the pH o f the bait recorded, recycled bait replaced
in the traps, and the traps moved to a new position sequentially. Trapped flies were
sorted by sex and species, and the number per trap recorded. The effect o f bait treat­
ment was analysed by one way analysis o f variance followed by Duncan’ s mean
separation test (P = 0.05). Based on five replicates, Nulure at a pH o f 8.5 captures
significantly more female medflies than both Nulure at a lower pH and torula yeast
pellets (Fig. 4). These findings allow the determination o f chemicals emitted from
Nulure having various degrees o f attractiveness and allow us to determine either the
amount of/or ratio o f differences o f potentially attractive chemicals released from
protein baits.

4. MEDFLY PHEROMONE

Baker et al. [5] reported the identification o f nine compounds emitted by male
medflies. However, no behavioural data to substantiate pheromonal activity were
46 8 HEATH and EPSKY

Female medflies

В С D E F
Treatment

Male medflies

Treatment

FIG. 4. The percentage of female and male medflies trapped with various protein baits.
A : Water (not shown); treatment B: five torula yeast pellets; C: Nulure + 1 % borax; D ;
Nulure + 3 % borax; E: Nulure + 5 % borax; F; Nulure + 10% borax; The effect ofbait treat­
ment was analysed by one way analysis of variance followed by Duncan’s mean separation
test (P = 0.05). n = 90 for each treatment.

provided. Subsequently, Jang et al. [6] determined the electroantennogram responses

o f male and female medflies to these nine compounds and to an additional 56 com ­
pounds collected from air passed over male medflies. In 1991, we reported on the
qualitative and quantitative analyses o f three principal components o f the volatiles
emitted by laboratory reared and wild male medflies during photophase [7]. The
attractiveness o f the pheromone blend at four release rates was investigated in field
trials conducted with wild medfly populations in Guatemala [7]. In subsequent trials
 IAEA-SM-327/43 469

Control Three component blend

Five male equivalents

Control ' Four component blend

Five male equivalents

FIG. 5. Response of female medflies to a three component and four component synthetic
pheromone blend (n = 16). ,

using the flight tunnel bioassay system described earlier, it appears that additional
components may result in a significantly improved lure for female medflies. A com­
parison between the feral and factory reared female medfly response to a control
equivalent to five caged males versus the previously field tested three-component
blend containing ethyl-(£)-3-octenoate, geranyl acetate and E, £-a-farnesene is
shown in Fig. 5. It is apparent that the three-component blend is considerably less
attractive. W e determined that the addition o f delta 1 pyrolline results in a considera­
ble increase in attractiveness (Fig. 5). Although we have improved the attractiveness
o f our previously field tested three-component blend, additional tests are being
conducted to delineate a blend that will be equal and perhaps better than that emitted
by male medflies.
470 HEATH and EPSKY

5. Anastrepha suspensa PHEROMONE

Research that began in 1972 and which has continued to date.has not yet
resulted in a pheromone based trap for the Caribbean fruit fly (caribfly), Anastrepha
suspensa (Lóew). The most current information on the pheromone components can
be found in a recent article by Rocca et al. [8] and references therein. Initial studies
on the chemical nature o f the pheromone extracted from the abdomens o f sexually
mature male caribflies resulted in the identification o f (Z)-3-nonenol and
(Z, Z)-3,6-nonadienol. Subsequent investigations o f the abdominal extracts o f male
flies resulted in the identification o f two additional components, the lactones, ahas-
trephin (írans-hexahydro-írans-4, 7a-dimethyl-4-vinyl-2-(3H)-benzofuranoñe) and
epi-anastrephin (irans-hexahydro-di-4, 7a-dimethyl-4-vinyl-2-(3H)-benzofuranone).
Laboratory bioassays o f these compounds showed that all were individually attrac­
tive to females, but a blend o f all four components was the most attractive to
females [9]. A synthetic mix o f the four components, however, failed to attract flies
in field trials [10]. A fifth component, a macrolide (E ,E )A , 8-dimethyl-3,
8-decadien-10-olide), was identified, synthesized and has been named ‘ suspen-
solide’ . Additionally, /З-bisabolene, ocimene and (E, £)-a-farnesene have been
reported as volatiles emitted by the caribfly .
Admittedly, the chemistry o f the male caribfly pheromone is complex. Analy­
sis o f some o f the pheromonal components is further complicated by their thermal
lability. For example, the identification o f suspensolide was not completed until 1988
and the identification o f (£ , E)-a-farnesene not until 1992. Elucidation o f these com ­
ponents in earlier research was hampered by the lack o f appropriate analytical
methodologies now available to researchers. Not all components o f the volatile blend
produced by calling males have been proved to have pheromonal activity. W e have
begun studies to test the female response to the putative male pheromone compo­
nents. We were able to obtain high rates o f attraction o f females to caged males using
the flight tunnel bioassay system. The testing o f synthetics is awaiting the completion
o f the synthesis o f some o f the pheromonal components.

6. O T H E R Anastrepha FRUIT FLY PHEROMONES

Considerable research on the pheromone chemistry o f Anastrepha ludens

(Loew), the Mexican fruit fly, has been conducted (Ref. [8] and references therein).
Again, this research has not yet resulted in a pheromone based trap. It is interesting
to note that there is cpnsiderable overlap between the pheromone components
produced by A. ludens and A. suspensa (Fig. 6). Analysis o f compounds from
A. fraterculus suggests that this pheromonal system may be similar to that of. the
caribfly and the Mexican fruit fly. This commonality does not appear to extend to
the other Anastrepha species, however. In some o f our recent investigations o f the
 IAEA-SM-327/43 471

A В C D E F G H

A. suspensa Y Y Y Y Y Y Y

A. ludens N Y Y Y N Y Y

A. obliqua N Y N N N N N

A. striata N N N N N N N
? ? ? ? ?
A. fraterculus

A. serpentina N N N N N N • N' ■

FIG. 6. Comparison of volatiles emitted from several species of male Anastrepha. 'Y'
indicates the presence of the compound, ‘N ’ indicates the absence of the compound and
'? ’ indicates that the compounds are likely present (A: ocimene; B: (E, E)-a-famesene;
C: (Z)-3-nonenol; D : (L, Z )-3, 6-nonadienol; E: suspensolide; F: ¡i-bisabolene; G : anastre-
phin; H : epi-anastrephin).

pheromonal compounds released by virgin male Æ obliqua, A. striata and A. serpen­

tina, we were unable to detect many o f the compounds found to be released by
A. suspensa and A. ludens.

1. CONCLUSIONS

The success in developing better attractants for fruit flies will require consider­
able efforts in understanding the biology o f the insect, especially as it relates to the
attractiveness o f the semiochemical being investigated. Additionally, a great deal o f
effort will be required to isolate, identify, synthesize and formulate the attractive
semiochemicals for use in the field. An integrated approach between, the chemistry
o f attractive chemicals and an understanding o f the insect response to the semio­
chemicals should provide a significant series o f excellent lures for the detection, and
perhaps even control, o f many economically important fruit flies.

A CKNO W LED GEM ENTS

The authors would like to thank the numerous scientists and technicians who
have made this research possible. Significant contributions by D. Baker, C. Calkins,
472 HEATH and EPSKY

B. Dueben, A. Fritz, P. Landolt, A. Manukian, J. Sivinski, T. Proveux (Agricul­

tural Research Service (ARS), United States Department o f Agriculture (USDA),
Gainesville, FL), T. Bakèr, J. Millar (University o f California, Riverside, CA),
D. Robacker (ARS, USDA, Weslaco, TX ), E. Jang (ARS, USDA, Hilo, HI),
K. Bloem, S. Bloem, D. Chamber and A ’ Guzman (Animal and Plant Health
Inspection Service (APHIS), Guatemala) are gratefully acknowledged. This research
was supported in part by the California Department o f Food and Agriculture under
Agreement No. 90-0621.

REFERENCES

[1] BERO ZA, М ., A LE XA N D E R , B .H ., STEINER, L .F ., M ITCHELL, W .C .,

M IY ASH ITA, D .H ’ , New'synthetic lures for the male melon fly, Science 131 3406
(1960) 1044-1045.
[2] McPHAIL, М ., Protein lures for fruitflies, J. Econ. Entomol. 32 6 (1939) 75 8 -7 6 1 .
[3] CALKINS, C .O ., SCHROEDER, W .J ., CHAMBERS, D .L ., The probability of
detecting the Caribbean fruit fly, Anastrepha suspensa (Loew) (Diptera:Tephritidae)
with various densities o f McPhail traps, J. Econ. Entomol. 77 1 (1984) 198-201.
[4] H EATH , R .R ., M A N U K IA N , A ., Development and evaluation o f systems to collect
volatile semiochemicals from insects and plants using a charcoal-infused medium for
air purification, J. Chem. Ecol. 18 7 (1992) 1209-1226.
[5] BAKER, R ., HERBERT, R .H ., G RANT, G .G ., Isolation and identification o f the sex
pheromone o f the Mediterraneáñ fruit fly, Ceratitis capitata (W ied.), J. Chem. Soc'.
Chem. Commun. X X (1985) 824-825.
[6] JANG, E .G ., LIGHT, D .M ., FLATH , R .A ., N A G A T A , J.T ., M O N , T .R ., Elec-
troantennogram responses o f the Mediterranean fruit fly, Ceratitis capitata, to
identified volatile constituents from calling males, Entomol. Exp. Appl. 50 1 (1989)
7 -1 9 .
[7] H EATH , R .R ., et al., Analysis, synthesis, formulation, and field testing o f the prin­
cipal components o f the male Mediterranean fruit fly pheromone, J. Chem. Ecol. 17
, 9 (1991) 1925-1940.
[8] R O C C A, J.R., N ATIO N , J.L., STREKOWSKI, L ., BATTISTE, M .A ., Comparison
o f volatiles emitted by male Caribbean and Mexican fruits flies, J. Chem. Ecol. 18 2
(1992) 22 3-24 4.
[9] N ATIO N, J.L ., The sex pheromone biend o f Caribbean fruit fly males: Isolation, bio­
logical activity and partial chemical characterization, Environ. Entomol. 4 1 (1975)
2 7 -3 0 .
[10] N ATIO N, J.L., “ The role o f pheromones in the mating system o f Anastrepha fruit
flies” , World Crop Pests, V ol. ЗА, Fruit Flies, Their Biology, Natural Enemies and
Control (ROBINSON, A .S ., HOOPER, G ,, Eds), Elsevier, New York (1989)
189-205.
 IAEA-SM-327/33

EFFECT OF THE SEX RATIO

ON EGG COLLECTION AND HATCH
OF THE MEDITERRANEAN FRUIT FLY
(DIPTERA: TEPHRITIDAE)

A.P. ECONOMOPOULOS*
FAO/IAEA Entomology Unit,
Agency’ s Laboratory Seibersdorf,
International Atomic Energy Agency,
Vienna

Abstract
EFFECT OF THE SEX RATIO ON EGG COLLECTION A N D H ATCH OF THE
M EDITERRANEAN FRUIT FLY (Diptera: Tephritidae).
When the female:male ratio in the adult cages o f artificially reared Mediterranean
fruit fly, Ceratitis capitata (Wiedemann), was increased from 50:50 to 80:20 (the initial
number o f total adults was always the same) a 50% overall increase in the fertilized eggs
collected per cage occurred during the first 2 weeks o f adult life. Smaller increases occurred
when the sex ratios were between 50 :50 and 80:20, and there was a decrease (compared with
the 80:20 sex ratio) when the sex ratio was 90:10 because the males needed several days to
fertilize all, or nearly all o f the females. At high female:male ratios, there was no increase
in the eggs sticking on the oviposition net, in spite o f the increased egg laying per surface
unit o f oviposition net.

1. INTRODUCTION

In area wide sterile insect technique programmes, hundreds o f millions o f

insects may be mass produced and released weekly. To reduce rearing and release
costs, rearing technique innovations are made. For example, genetic sexing strains
o f the Mediterranean fruit fly, Ceratitis capitata (Wiedemann) (Diptera:
Tephritidae), have been developed to separate the sexes so that only males are
released [1]. When the sexes are easily separated, it is possible to make any desired
sex ratio in the mass rearing adult cages. If the number o f eggs collected per cage
could be substantially increased, the rearing costs could be reduced significantly.
In the present work, a study was made o f the effect o f different female:male ratios
on egg collection and fertilization in two different adult density experiments.

* Present address: Department of Biology, University o f Crete, and Institute of

Molecular Biology and Biotechnology, P.O: Box 1470, GR-71110 Heraklion, Crete, Greece.

473
474 ECONOMOPOULOS

T A B L E I. E G G S C O L L E C T E D PER C A G E P E R D A Y F R O M
M E D IT E R R A N E A N F R U IT F L IE S K E P T A T S E V E N D IF F E R E N T
S E X R A T IO S
(Insect density at start, 0.3 flies/cm3 cage volume, or 0 .7 flies /cm 2 cage surface;
oviposition net surface/cage, 66.5 cm2)

Sex ratio Eggs collected per cage per experiment day2

(% females) Days 4, 5, 6, 7, 8 Days 15, 22

90 18 289 ± 1254a 3450 ± 668a

80 17 119 ± 882a 4032 ± 574a
70 14 208 ± 886b 1866 ± 483b
60 11 489 ± 1323c 2405 ± 839b
50 10 609 ± 586c 2409 ± 316b
40 9 006 ± 579d 2441 ± 510b
. 30 7 035 ± 839e 2370 ± 268b

a Four cage replications per treatment. Eggs laid on days other than those indicated were
removed but not counted. For each column, means with the same letter do not differ
- significantly (P > 0 .0 5 ; Duncan’ s multiple range test [3]).

T A B L E П. PER C E N T H A T C H A B IL IT Y O F E G G S O F M E D IT E R R A N E A N
F R U IT FLIE S K E P T A T S E V E N D IF F E R E N T S E X R A T IO S
(Same experiment as in Table I)

Sex ratio Per cent egg hatcha

(% females) Days 4, 5, 6, 7, 8 Days 15, 22

90 59.7 ± 2.3d 6 4.4 ± 5.6b

80 . 75.4 ± 2.2b 80.7 ± 4.3a
70 . 68.7 ± 9.5c 66.5 ± 7.8b
60 80.6 ± 3.0ab 62.4 ± 8.6b
50 80.4 ± 1.8ab 62.1 ± 6.2b
40 78.5 ± 2.0ab 61.4 ± 5.1b
30 83.2 ± 0.8a 61.9 ± 3.5b

a Four cage replications per treatment. Eggs laid on days other than those indicated were
removed but not counted. For each column, means with the same letter do not differ
significantly (P > 0 .0 5 ; Duncan’ s multiple range test [3]).
 IAEA^SM-327/33 475

TABLE Ш. EGGS COLLECTED PER CAGE PER D A Y FROM

MEDITERRANEAN FRUIT FLIES KEPT A T FIVE DIFFERENT
SEX RATIOS
(Insect density at start, 0.16 flies/cm3 cage volume, or 0.33 flies/cm2
cage surface; oviposition net surface/cage, 184 cm2)

Sex ratio Eggs collected per cage per experimental daya

(% females) Days 1, 2, 3, 4, 5, 6, 8, 10 Days 15, 17, 22, 24, 29, 31

90 13 223 ± 420a 2677 ± 639a

80 11 972 ± 686b 2231 ± 665ab
70 10 644 ± 183c 2352 ± 204ab
60 '9 125 ± 171d 2059 ± 329ab
50 7 248 ± 190e 1679 ± 381b

a Four cagé replications per treatment. Eggs laid on days other than those indicated were
removed but not counted. For each column, means with the same letter do not differ
significantly (P > 0 .0 5 ; Duncan’ s multiple range test [3]).

TABLE IV. PER CENT HATCHABILITY OF EGGS OF MEDITERRANEAN

FRUIT FLIES KEPT AT FIVE DIFFERENT SEX RATIOS
(Same experiment as in Table III)

Sex ratio Per cent egg hatch3

(% females) Days 1, 2, 3, 4, 5, 6, 8, 10 Days 15, 17, 22, 24, 29, 31

90 65.6 ± 2.7d 32.4 ± 4.1c

80 74.4 ± 0.4c 4 6.7 ± 5.2b
70 75.2 ± 0.8c 54.2 ± 2.8a
60 79.2 ± 1.4b 55 .2 ± 1.3a
50 81.2 ± 1.1a 53.8 ± 4.2a

a Four cage replications per treatment. Eggs laid on days other than those indicated were
removed but not counted. For each column, means with' the same letter do not differ
significantly. (P > 0 .0 5 ; Duncan’ s multiple range test [3]).
476 ECONOMOPOULOS

2. M A TE RIALS A N D M ETHODS

The insects used were from a colony established in the IAEA Laboratory
at Seibersdorf in October 1983 from pupae that had been collected from guava in
the Sohag Governorate in Egypt. The rearing procedures were as described by
Hooper [2]. Small plastic cages (11 cm x 11 cm x 15.3 cm) were used, with four
cages (replicates) per sex ratio treatment in both experiments.
Two successive experiments were run. In the first, the adult flies were kept
for 3 1/2 weeks in cages at an initial density o f 0.3 flies/cm3 (600 flies per cage).
A 9 cm diameter hole, covered by a 0.1 mm2 hole synthetic oviposition net (the
total oviposition surface per cage was 66.5 cm 2) was prepared for oviposition in
one o f the two small sides (11 cm x 11 cm) o f the cage. Sex ratios o f 30, 40, 50,
60, 70, 80 and 90% females were compared. Insect mortality, egg collection
(eggs inserted through the net and dropped on moistened filter paper plus those that
stuck on the oviposition net) and egg hatch (100 eggs from those dropped on the
filter paper) were examined per cage on the experiment days (adult day minus 2)
shown in Tables I and II. The cages were kept at 25 ± 2°C , 55-65% RH and a
12 h photoperiod. The light intensity was about 2800 lux inside the cages.
In the second experiment, the adult flies were kept for 4 1/2 weeks in cages
at an initial density o f 0.15 flies/cm3 (300 flies per cage). An 8 cm x 11.5 cm
rectangular hole, covered by an oviposition net, was prepared for oviposition on
each o f the two opposite large sides (11 cm x 15.3 cm) o f the cage (the total
oviposition surface per cage was 2 X 92 cm 2 = 184 cm2). Sex ratios o f 50, 60,
70, 80 and 90% females were compared. Again, insect mortality, egg collection
and egg hatch (from 500 eggs that had dropped) were examined on the experiment
days shown in Tables III and IV. The cages were again kept at 25 ± 2°C and
55-65 % RH but with a 14 h photoperiod.
In both experiments, the first experiment day was usually day 3 o f adult life.
Adults were o f variable size, having been collected as larvae on day 2 o f their
emigration from the diet to the pupation medium. The flies were provided with
water and solid adult food containing three parts sucrose to one part hydrolized
yeast.
Data were analysed by analysis o f variance (ANOVA), and significant differ­
ences between the means (P < 0.05) were detected by Duncan’ s multiple range
test [3].

3. RESULTS

In both experiments, there were no significant differences in the longevity

o f either sex when compared among the different sex ratios tested. When the
longevity o f the two sexes was compared, for the different sex ratio treatments o f
 IAEA-SM-327/33 477

the second experiment, male longevity was always shorter than female longevity,
by 12.5-18.5% .
Tables I and Ш show the egg collection rates per cage in the two experiments.
The total egg collection per cage per day increased by 72 and 82% when the sex
ratio was changed from 50 to 90% females in the first and second experiment,
respectively. The collection o f eggs per cage per day increased, usually significantly
(P < 0.05), in all successive increases in the female:male ratio. In the second part
o f both experiments, a significant increase in egg collection again occurred when the
lower female:male ratios were compared with the higher ones. Nevertheless, the
differences were not as striking as in the first part o f the experiments, and often no
differences could be detected between successive increases in the female:male ratios.
Tables П and IV show the percentage eggs fertilized in the different sex ratios
o f the two experiments. In the first part o f the experiments, egg hatch decreased
by 26 and 19% when the sex ratio was changed from 50 to 90% females in the first
and second experiment, respectively. In the first experiment, the decrease occurred
only in the higher female.male sex ratios, while in the second experiment the
decrease was small, but constant, in successive increases in the female:male ratio.
In the second part o f the first experiment, no significant decrease in egg hatch was
observed with successive increases in the female:male ratio, while in the same period
o f the second experiment a striking decrease in egg hatch was observed in the two
highest female:male ratios.

TABLE V. PER CENT INCREASE IN FERTILIZED EGGS COLLECTED

PER CAGE WITH INCREASE IN FEM ALE:MALE RATIO OF
MEDITERRANEAN FRUIT FLIES KEPT A T TWO DIFFERENT DENSITIES
AND OVIPOSITION SURFACE SIZES

Increase in Per cent increase in fertilized eggs/cagea

females Experiment 1 Experiment 2
(* ) (Tables I, П) (Tables III, IV)

3 0 -4 0 20.8 —

40 - 50 20.7 —
5 0 -6 0 8 .6 22.8
6 0 -7 0 5 .4 ■ 10.8
7 0 -8 0 32.2 11.3
8 0 -9 0 -1 5 .4 - 2 .6

a Only for days 4 , 5, 6, 7, 8 and 1, 2, 3, 4 , 5, 6, 8, 10 o f Experiments 1 and 2 , respectively.

478 ECONOMOPOULOS

Table V combines egg collection and fertilization and shows the effect o f the
sex ratio on the overall collection o f fertilized eggs during the first 2 weeks o f the
experiments. In both experiments, when the female:male ratio was changed from
50:50% to 80:20% , there was an overall increase in the fertilized eggs, by about
50% . A further increase in the female:male ratio, to 90:10% , caused a decrease in
the fertilized eggs.
A detailed study o f egg hatch in the first experiment showed that only 1-2 days
were needed for the female:male ratios o f 60:40, 70:30 and 80:20 to reach
the percentage egg hatch o f the 50:50 sex ratio. The corresponding time for the
90:10 female:male ratio was 4 days. Also, after the second week o f adult life, the
percentage eggs that hatched with the 90:10 female:male ratio decreased much
faster than in the other sex ratios, corresponding to the increase in percentage
males per cage. Similar changes also occurred in the second experiment.
The collection o f eggs per cage was 26-46% higher in the first experiment for
the corresponding sex ratio treatments than for the second experiment (Tables I, III).
There were no significant differences in the percentage o f eggs that stuck on the
oviposition net among the different sex ratios in both periods-of the two experiments.

4. DISCUSSION

The major economic species o f tephritids (fruit flies) are in the subfamilies
o f Dacinae and Trypetinae. In Dacinae, the sexual behaviour has been studied
mostly in the genus Dácus, where the males are polygamous but the females do
not remate frequently. In Trypetinae, different mating frequencies have been found
for females o f the different species o f Anastrepha, Ceratitis and Rhagoletis [4]. In
the Mediterranean fruit fly, males are polygamous while females generally are not,
depending on the amount o f sperm stored in their spermathecae [5]. Among
laboratory reared flies, o f the females that remated (about 60% o f the total), 54%
remated once, 25% twice, 12% three times and the remaining between four and six
times. The amount o f sperm stored in the spermathecae declined considerably only
after 20-25 days following initial mating, and remating increased with the decrease
in stored sperm. This suggests that one successful mating can fertilize most o f the
eggs produced in an average female’ s life and certainly for the 2 weeks that the
flies are kept in mass rearing facilities.
Following the recent successful efforts to develop genetic sexing strains in the
Mediterranean fruit fly [1], the present results clearly show that high female:male
ratios can be used in the mass rearing o f the fly. Female:male ratios o f around 80:20
are recommended, since they result in the highest possible collection o f fertilized
eggs per cage during the first 2 weeks o f adult life. A further increase in the
 IAEA-SM-327/33 479

proportion o f females does not increase the collection o f fertilized eggs because
it takes several days for the low number o f males to achieve fertilization o f the
females.
Although in Experiment 1 there were twice as many females as in Experi­
ment 2, egg collection was only 26-46% higher. As has been reported [6], appar­
ently this resulted from the longer duration o f light in Experiment 2 (14 versus
12 hours, respectively), and the reduced oviposition surface and increased insect
density in Experiment 1. The finding that there was no increase in eggs sticking on
the oviposition net as a result o f increased egg laying per unit o f oviposition surface
in the highest female:male ratios confirms that sticking depends on fly activity,
which results in dirt on the net [7, 8], and not the number o f eggs laid. Finally, the
reduced survival rates o f males compared with females for all the sex ratios tested
indicate that reproduction is not a major survival stress for the female. Competition
among the polygamous males to secure matings could be one o f the factors, among
others, which reduced the male lifespan. Nevertheless, it is noted that no significant
difference was observed in the present study in male survival between treatments
o f low and high female:male ratio, i.e. high and low male mating competition,
respectively.
In conclusion, to achieve a substantial reduction in the mass rearing costs
o f genetic sexing strains, a ratio o f 80% females to 20% males in the cages is
recommended. This results during the first two weeks o f adult life in about a 50%
increase in fertilized eggs collected per cage compared with cages containing a
50:50 ratio. The possible negative effects o f high female:male ratios on the quality
o f flies, in particular their sexual behaviour, should be examined.

ACKNOW LEDGEM ENTS

The author wishes to thank D. Lindquist and R. Gingrich for supporting this
work, L. Lang for technical help, and H. Baumgartner for assistance in statistical
analysis o f the data.

REFERENCES

[1] Genetic Sexing o f the Mediterranean Fruit Fly (Proc. Final Research Co-ordination
Mtg, Colymbari, Crete, 1988), IAE A, Vienna (1990).
[2] HOOPER, G .H .S ., Application o f quality control procedure to large-scale rearing of
the Mediterranean fruit fly, Entomol. Exp. Appl. 44 (1987) 161-167.
[3] STEEL, R .G .D ., TORRIE, J.H ., Principles and Procedures o f Statistics, McGraw-
Hill, New York (I960).
480 ECONOMOPOULOS

[4] B A T E M A N , M .A ., et al., “ Fruit Flies” , Studies in Biological Control

(DELUCCHI, V .L ., Ed.), International Biological Programme 9, Cambridge Univer­
sity Press, London (1976) 11-49.
[5] N A K A G A W A , S ., FARIAS, G .J., SU D A , D ., C U N N IN G H AM , R .T .,
CHAMBERS, D .L ., Reproduction o f the Mediterranean fruit fly: Frequency of
mating in the laboratory, Ann. Entomol. Soc. Am . 64 4 (1971) 949-950.
[6] ECONOMOPOULOS, A .P ., BRU ZZO N E, N .D ., Mass rearing o f the Mediterranean
fruit fly (Diptera: Tephritidae): Continuous light in the adult stage, J. Econ. Entomol.
82 5 (1989) 1482-1489.
[7] N A D E L, D .J., “ Current mass-rearing techniques for the Mediterranean fruit fly” ,
Sterile Male Technique for Control o f Fruit Flies (Proc. Panel Vienna, 1969), IAE A,
Vienna (1970) 13-18.
[8] ECONOMOPOULOS, A .P ., JUDT, S., Artificial rearing o f the Mediterranean fruit
fly (Diptera: Tephritidae): Size of oviposition holes, J. Econ. Entomol. 82 2 (1989)
66 8-674.
POSTER PRESENTATION

1AEA-SM-327/1P

R E C O V E R Y O F FE R TILIT Y IN IR R A D IA TE D
Epestia cautella (W A LK E R)
(LEPIDOPTERA: PH YCITIDAE)

H. MAKEE
Atomic Energy Commission,
Damascus,
Syrian Arab Republic

I. INTRODUCTION

The recovery o f damaged sperm o f Ephestia cautella irradiated with 10 krad1

was determined when: (1) irradiated males were repeatedly mated with hew unirradi­
ated females; and (2) irradiated males were held after treatment for a period o f time
before allowing them to mate with unirradiated females.

2. MATERIAL AND METHODS

2.1. Effect o f repeated mating on the mating ability

and fertility o f irradiated males

Twenty-nine 1-d-old males were irradiated with 10 krad at a dose rate o f

4584 rad/min. After treatment, the males paired singly with 2-d-old females. The
females were removed after 24 h and subsequent oviposition observed. Fresh 2-d-old
females were introduced to the males for 24 h. The same procedure was followed
for four successive days. Males in the control groups were mated in the same manner
as the irradiated males. Every two days, all the eggs laid by each female in each
group were counted and a sample taken to determine the percentage egg hatch! The
females were dissected after death and examined for the presence o f spermatophores.

2.2. Effect o f delayed mating on the mating ability

and fertility o f irradiated males

Four different groups o f irradiated males were mated at various intervals

(0, 1 ,2 and 3 d) after treatment. In each group, the males were paired, singly, with

1 1 rad = 1.00 x 1 0 '2 Gy.

481
482 POSTER PRESENTATION

2-d-old females for 24 h, after which the males were removed and oviposition
observed. Four groups o f unirradiated males aged 1, 2, 3 and 4 d old were taken
as controls. The percentage egg hatch and mating ability were determined as
described in Section 2.1.

3. RESULTS AND DISCUSSION

3.1. Effect of repeated mating

The mating ability o f irradiated and unirradiated males was very high in the
first copulation; after that it started to decline. At any copulation, the mating ability
o f irradiated males did not differ significantly from that o f the control (Table I).
It is possible that in late mating the males were unable to mate because they
had already depleted their supply o f sperm and accessory gland secretions during the
early mating [1].
The results show that both irradiated and unirradiated males were most fertile
in the first mating, but then their fertility declined during the following mating.
There was no significant difference in the fertility o f irradiated males in the second,
third and fourth matings. The fertility o f these males was markedly lower than the
control males, regardless o f the number o f matings involved (Table I).
It has been known that mature sperm are the most radioresistant stage o f sper­
matogenesis: spermatids are less resistant and would suffer greater damage. This
could explain the higher fertility in early mating than in late mating.

TABLE I. EFFECT OF REPEATED MATING ON MATING ABILITY AND

FERTILITY OF IRRADIATED Ephestia cautella MALES (10 krad)

Time o f mating after treatment (d)

Parameter Male group
1 2 3 4. .

% mating Control . (20/20) 100a (20/20) 80b (20/20). 100a (7/20) 35c
Irradiated (26/26) 100a (23/26) 88a (19/26) 73ab (10/26) 38c

Mean % Control 44.8a 37.5a 20.8b 13.1b

fertility Irradiated. 19.9c 7.8d 9.7d 3. Id

Note: Numbers followed by the same letter are not significantly different.
Numbers in parentheses denote (number o f mated males)/(number o f tested males).
 POSTER PRESENTATION 483

T A B L E II. EFFECT OF D E L A Y E D M A T IN G ON M A TIN G A B IL IT Y A N D

FER TILITY OF IR R A D IA T E D Ephestia cautella M ALES (10 krad)

Time o f mating after treatment (d)

Parameter Male group
0 1 2 3

% mating Control (25/26) 96a (24/28) 86a (22/27) 81a (17/28) 61b
Irradiated (23/24) 96a (18/25) 72a (19/25) 76a (13/26) 50b

Mean % Control 41.1a 40.3a 33.9a 36.7a

fertility Irradiated 12.6b 17.6b 17.7b 11.2b

Note: Numbers followed by the same letter are not significantly different.
Numbers in parentheses denote (numbers o f mated males)/(numbers o f tested
males).

3.2. Effect o f delayed mating

When the irradiated males were paired on the same day o f treatment, their mat­
ing ability was higher than at any other time. The mating ability o f the control males
declined steadily with age. There was no significant difference in the mating ability
o f irradiated and unirradiated males (Table II). Irradiation appears, therefore, to
have no effect on the ability o f the male to mate and only age is important in the
reduction in male mating ability [2].
When irradiated males were mated at various intervals after treatment, their
fertility did not change significantly. A very similar result was recorded in the con­
trol males (Table II).
Thus, the age or storage o f sperm in the reproduction system for various inter­
vals before mating did not have any effect on male fertility.
These results are very important in applying the sterile male technique against
E. cautella.

REFERENCES

[1] T A Y L O R , O .R ., Relationship o f multiple mating to fertility in Atteva punctella, Ann.

Ent. Soc. Am . 60 (1967) 58 3-59 0.
[2] М А КЕЕ, H ., Studies on the Sterile Male Technique for the Control o f the Tropical
Warehouse Moth, Ephestia cautella, PhD Thesis, Reading University, Reading (1989).
BIOCONTROL
(Session 6)

Chairman

S. BARBOSA
FAO
 IAEA-SM-327/44

PROTEIN HYDROLYSATE COMPONENTS

ATTRACTIVE TO TEPHRITIDS

R. TERANISHI, S. KINT, R .A , FLATH, D.M . LIGHT

Western Regional Research Center,
Agricultural Research Service,
United States Department o f Agriculture,
Albany

S.B. OPP, K.M . REYNOLDS, W. HAMERSKY

Department o f Biological Sciences,
California State University,
Hayward

California,
United States o f America

Abstract

PROTEIN H YD R O LYSATE COMPONENTS ATTR ACTIVE TO TEPHRITIDS.

Volatiles from protein hydrolysates have for a long time been known to attract
tephritids. Many volatiles from protein hydrolysates have previously been idëntified, biit no
highly attractive materials have been determined. There have been few studies on the very low
boiling components, other than ammonia. Because protein hydrolysate is more attractive to
tephritids at alkaline rather than at slightly acidic conditions, vapours from Nu-Lure insect bait
(NLIB) at pH4.5 and pH8.5 were examined by gas chromatography/mass spectrometry
(G C /M S). Compounds identified by G C /M S did not evoke very high responses, therefore
attention was focused on ammonia. Because ammonia is a gas at ambient temperatures and
at atmospheric pressure, a slow release system was devised. Ammonia was tested with walnut
husk flies in California and was found to be primarily attractive to the female flies.

1. INTRODUCTION

Even though volatiles from protein hydrolysates have for a long time been
known to attract tephritids [1-10], attractants accounting for most o f the activity have
not been chemically identified,, although there are indications that some components
have been found to be somewhat active. The standard methods o f distillation and
extraction have been applied, and many compounds have been identified
[4, 5, 10]. Because such distillation and extraction methods did not yield very active
isolates, it was decided to look at components with very low boiling points,
specifically ammonia. • ¡ . .

487
488 TERANISHI et al.

It is well known that when the pH o f protein hydrolysate is adjusted to higher

pH values from the pH o f about 4.5 at which most protein hydrolysates are received,
the attractancy to tephritids is increased considerably [4, 10]. Although ammonia has
been mentioned [3, 4, 1-, 9, 10], no definite explanation has been given as to what
compound or compounds is/are responsible for the increased attractancy as the pH
is increased. No strict correlation between increased rate o f ammonia release to
increased attractancy has been found [7]. This paper describes some o f our investiga­
tions o f ammonia.

2. MATERIALS AND METHODS

The protein hydrolysate used in this study was Nu-Lure insect bait (NLIB),
provided by the California Department o f Food and Agriculture, arranged by
R.T. Cunningham, Tropical Fruit and Vegetable Research Laboratory, Agricultural
Research Service (ARS), United States Department o f Agriculture (USDA),
Honolulu, Hawaii.
The ammonium carbonate used, certified ACS grade, was purchased from the
Fisher Scientific Company.
Cotton wicks were purchased from the Richmond Dental Cotton Company,
Charlotte, North Carolina.
The gas chromatographs used were Hewlett Packard models 5830A and 5840A
fitted with flame ionization detectors. The mass spectrometer used was a Finnigan
M AT 4500 quadrupole fitted with a gas chromatograph. Cross-linked bonded methyl
silicone columns (DB-1, 60 m X 0.32 mm i.d., 0.25 fim film, J&W Scientific, Inc.)
and DB-W AX columns (same dimensions and source as the silicone columns)
were used.
The isolation and chemical identification methods used were those developed
for flavour studies by the group at the Western Regional Research Center, ARS,
USDA, Albany, California.
The bioassay methods used were those developed at the Department o f Biologi­
cal Sciences, California State University, Hayward, California.

3. RESULTS AND DISCUSSION

The efficacy o f ammonia as an attractant to tephritids has been debated

[3, 4, 7, 9]. One o f the reasons for conflicting results is perhaps that, because
ammonia has a boiling point o f —33.4°C at atmospheric pressure, it is very difficult
to sustain an even, controllable supply o f this material over a period long enough
to observe insect responses accurately for days or weeks. To overcome this problem
o f accurate release o f low boiling compounds, a slow release system was designed.
 IAEA-SM-327/44 489

Diffusion Time

Hours

Cotton wick

Days

Wick in vial

Weeks

Small hole in cap

FIG. 1. Diagrams illustrating the reduction in diffusion by limiting the direction and reducing
the orifice for the escape of ammonia.

In perfumery, slow release is achieved, by dissolving a desired odorant in a

high molecular weight substance, called a fixative. In this manner, vapour pressure
is reduced, and slow evaporation is achieved. No simple fixative seems feasible for
ammonia because it has such a low boiling point and is so polar.
The other principle which can be utilized involves reducing diffusion. Figure 1
illustrates diagrammatically how diffusion can be controlled. If an aqueous saturated
ammonium carbonate solution is absorbed on a cotton wick, because diffusion can
take place in many directions, ammonium carbonate is completely evaporated in a
few hours. If the cotton wick is placed in a vial, diffusion is restricted to one direc­
tion. Vapours have only one escape path — out o f the mouth o f the vial . In this situa­
tion, ammonia may take a few days to dissipate. If the escape path is limited to a
small hole in the cap o f the vial, it may take weeks for the ammonia to evaporate
completely.
Because ammonium carbonate is composed o f a weak acid and a weak base,
it dissociates to yield ammonia, carbon dioxide and water. One molecule o f ammo­
nium carbonate yields two. molecules o f ammonia.
Loss o f weight due to the evaporation o f ammonium carbonate is simply
measured by weighing the bottle containing this material every day. If a. bottle con­
taining 30 g o f ammonium carbonate, with a hole (6.25 mm) drilled in its cap, loses
490 TERANISHI et al.

0.3 g/d, this bottle will provide ammonia vapours for 100 d. To indicate the amount
volatilized per second in terms o f moles, the weight loss is divided by the formula
weight o f ammonium carbonate, and the result is the number o f moles o f ammonium
carbonate dissociated per day. Because there are two moles o f ammonia given o ff
per mole o f ammonium carbonate, the moles o f ammonia emitted are double the
moles o f ammonium carbonate evaporated. To obtain the amount per second, one
simply divides by the number o f seconds in a day.
From simple experimentation and from simple calculations, the amount o f
ammonia given o ff can be accurately measured and easily calculated. With an inex­
pensive source available, with controlled and measurable amounts dispensed, opti­
mum conditions for efficiency o f trapping with ammonia can now be systematically
approached. A report on bioassays with the walnut husk fly (Rhagoletis completa
Cresson) is currently in preparation. However, a brief discussion o f the effect o f
ammonia seems necessary to illustrate the importance o f ammonia in attracting
tephritids. The observations reported here are with the walnut hiisk fly.
Walnut husk flies were captured on Scentry Multigard traps baited with various
lure substances during the summer o f 1992 in an unsprayed English walnut orchard
at the Ardenwood Historic Farm, East Bay Regional Parks, Newark, California.
These were wild flies found under natural, field conditions. Four materials were used
as lures: (1) bird faeces (fresh chicken droppings mixed with an equal volume o f
water); (2) NLIB at pH4.5; (3) NLIB at pH8.5; and (4) ammonia from powdered
ammonium carbonate sprinkled on the sticky material on the traps or from
ammonium carbonate dispensed with slow release systems. Water baited traps were
used as controls. Each trap-lure combination was replicated five times for a total o f
25 traps hung for continuous five-day periods during mid-season. Flies were
removed and separated by gender from each trap daily.
Traps baited with bird faeces captured fewer flies than other materials
tested and did not differ much from the water controls (Fig. 2). On the first day o f
observations, more than three times the number o f walnut husk flies were captured
on traps baited with NLIB at pH8.5 than on traps baited with NLIB at pH4:5.
Moreover, twice the number o f flies were captured on ammonium carbonate baited
traps than on traps baited with NLIB at pH8.5. Both ammonium carbonate and NLIB
at pH8.5 baited traps captured a far greater number o f flies on the first day than they
did on the remaining two days. This observation can be explained by the substantial
decrease in ammonia concentration by the second day.
Ammonium carbonate baited traps captured significantly more female walnut
husk flies than NLIB at pH4.5 or at pH8.5 (Fig. 3). Capture o f females continued
at a higher rate with use o f systems capable o f controlled slow release o f ammonia
versus the application o f powdered ammonium carbonate (Fig. 4).
No significant differences in the capture o f males were seen among the
traps baited with the three substances from the first day to the end o f the five
trapping days. Equally low numbers o f males and females were captured daily with
 IAEA-SM-327/44 491

Date (in 1992)

FIG. 2. Captures of walnut husk flies on yellow panel traps baited with various substances
showing the rapid decline in attractiveness of ammonium carbonate and N U B at pH8.5.
Each bar represents the total number offlies captured on five replicates of each treatment.

Lure substances

FIG. 3. First'day captures of female versus male walnut husk flies on yellow panel traps
baited with N U B atpH4.5, N U B atpH8.5 orammonium carbonate. Each bar represents the
total number of flies of each sex captured on five replicates of each treatment
(fâ: females; ■ / males).
492 TERANISHI et al.

4 0 0-i

350-
"D

l , . I —r i
18 Aug. 19 Aug. 20 Aug. 21 Aug. 22 Aug.

Date (in 1992)

FIG. 4. Captures of walnut husk flies on yellow panel traps baited with 2 g of powdered
ammonium carbonate versus a saturated solution of ammonium carbonate. Each bar
represents the total number of flies captured on five replicates of each treatment
( 0 : ammonium carbonate powder; ■ ; ammonium carbonate tube).

NLIB at pH4.5. Perhaps an explanation for the low rate o f capture o f males with
protein hydrolysate volátiles is that females need protein food to provide for the
development o f the eggs they carry, whereas male adults have no such urgent need
for protein food.

4. SUMMARY

The efficacy o f ammonia as an attractant to tephritids has been demonstrated.

Ammonia from ammonium carbonate attracts more walnut husk flies than NLIB at
pH8.5. The incréasëd attractancy when NLIB is made basic may be explained by the
increase in the ammonia concentration. The loss o f attraction to NLIB at pH8.5 after
one day may be explained by the decrease in the concentration o f ammonia. Ammo­
nia is a good attractant for females but not for males. A system for the slow release
o f ammonia has been devised by utilizing simple, inexpensive and commercially
available chemicals and equipment to deliver controllable and easily measured
amounts o f ammonia for periods o f weeks and months. The use o f ammonia and the
slow release systems can be easily applied for the survey and control o f tephritids.
 IAEA-SM-327/44 493

REFERENCES

[1] STEINER, L .F ., Fruit fly control in Hawaii with poison-bait sprays containing protein
hydrolysates, J. Econ. Entomol. 45 (1952) 83 8-843.
[2] STEINER, L .F ., Bait sprays for fruit fly control, Agrie. Chem. 10 (1955) 3 2 -4 3 and
113-115.
[3] B A T E M A N , M .A ., M O RTO N , T .C ., The importance of ammonia in proteinaceous
attractants for fruit flies (Family. Tephritidae), Aust. J. Agrie. Res. 32 (1981) 88 3-89 0.
[4] M O R TO N, T .C ., B A TE M A N , M .A ., Chemical studies on proteinaceous attractants
for fruit flies, including the identification o f volatile constituents, Aust. J. Agrie. Res.
32 (1981) 90 5-91 6.
[5] BUTTERY, R .G ., LING, L .C ., TERANISHI, R ., M O N , T .R ., Insect attractants:
Volatiles o f hydrolyzed protein insect baits, J. Agrie. Food Chem. 31 (1983) 68 9-69 2.
[6] M AT SU M O T O , K .E ., BUTTERY, R .G ., FLATH, R .A ., MON, T .R .,
TERANISHI, R ., “ Protein hydrolysate volatiles as insect attractants” , Bioregulators
for Pest Control (HEDIN, P .A ., Ed.), ACS Symposium Series No. 276, American
Chemical Society, Washington, DC (1985) 35 3-366.
[7] M A ZO R , М ., GOTHILF, S ., G A L U N , G ., The role o f ammonia o f the attraction of
females in the Mediterranean fruit fly to protein hydrolysate baits, Entomol. Exp. Appl.
43 (1987) 2 5 -2 9 .
[8] TERANISHI, R ., et al., “ Recent developments in chemical attractant for Tephritid
fruit flies” , Allelochemicals: Role in Agriculture and Forestry (W ALLER , G .R ., Ed.),
ACS Symposium Series No. 330, American Chemical Society, Washington, DC (1987)
4 3 1-43 8.
[9] D R EW , R .A .I., F A Y , H .A .C ., Comparison o f the roles o f ammonia and bacteria in
the attraction o f Dacus tryoni (Froggatt) (Queensland fruit fly) to proteinaceous suspen­
sions, J. Plant Prot. Tropics 5 (1988) 127-130.
[10] F L A T H , R .A ., M A T S U M O T O , K .E ., B IN D E R , R .G ., C U N N IN G H A M , R .T .,
M O N , T .R ., E ffect o f pH on the volatiles o f hydrolyzed protein insect baits, J. A grie.
Food Chem. 37 (1989) 814-819.
 IAEA-SM-327/45

ECOHORMONES FOR THE

MANAGEMENT OF FRUIT FLY PESTS
Understanding plant-fruit fly-predator interrelationships

Keng Hong TAN

School o f Biological Sciences,
University o f Sciënce, Malaysia,
Penang,
Malaysia

Abstract
ECOHORM ONES FOR THE M A N A G E M E N T OF FRUIT F LY PESTS: U ND ERSTAN D ­
ING PLA N T-FRU IT FLY -PR ED ATO R INTERRELATIONSHIPS.
The flowers o f several plant species o f Bulbophyllum (Orchidaceae) and Spàthiphyllum
(Araceae) that release methyl eugenol (ME) or other phenylpropanoids, and several varieties
of Dendrobium anosmum (Orchidaceae) that secrete raspberry ketone (RK) are attractive to
fruit' flies, which actively consume the ecohormone (fragrance) released. Ocimum sanctum
stores large quantities o f M E , which is released only when some part o f the plant is damaged.
When the male melon fly, Bactrocera cucurbitae, feeds on RK, the ecohormone is
sequestered. Methyl eugenol, when consumed by male flies, is converted to other phenyl-
propanoid(s) for storage. Bactrocera dorsalis (taxon A ) produced coniferylalcohol (CF);
B. dorsalis (taxon B) produced three compounds: CF, 2-allyl-4,5-dimethoxyphenol (com­
pound I) and Z-3,4-dimethoxycinnamyl alcohol (Z compound П); and B. umbrosa produced
three compounds: compound I, E and Z compound II and 3,4-dimethoxy-hydroxyallyl
benzene. Coniferyl alcohol and compound I are attractive to conspecific males, while CF
attracts conspecific females and stimulates ovipositor extrusion at close range. The strong
attractancy o f fruit flies to certain lures can now be explained by demonstrating: (1) an anti­
predation mechanism in several species of fruit fly through the endogenous production and
secretion o f allomone(s); and (2) chemical(s) responsible for male aggregation in a lek forma­
tion, after pharmacophagy. Compulsive feeding o f M E by male flies produced a feeding deter­
rent effect on the house lizard, Hemidactylus frenatus. Lizards did not attempt to feed on fruit
flies after initial exposure to flies that had fed on M E. On the other hand, sequestration of
RK may not play an anti-predation role in the melon fly, but functions as a male aggregation
pheromone. The anti-predation mechanism in the melon fly is the result o f the endogenous
synthesis o f ethyl-hydroxybenzoate and 1,3-nonandiol in the male rectal gland at sexual matu­
rity. Most mature male flies undergo reflex ejaculation o f rectal gland content when they are
under stress, e.g. when they are being anaesthetized with carbon dioxide or when they are
being held by a feather forceps.

495
496 TAN

1. INTRODUCTION

Certain species o f tephritid fruit flies are amongst the most important global
pests infesting fruit and vegetable crops. In the tropics and subtropics, many species
o f Bactrocera (Dacus) are major, or potentially important, pests. Most Bactrocera
species are strongly attracted to male lures, either methyl eugenol (ME) or cuelure
(CL)/raspberry ketone (rheosmin/RK) [1-3]. Since the discovery in 1912 that
B. diversus and B. zonatus were attracted to citronella oil [4], attempts have been
made to explain the strong attractancy o f fruit fly males to lures. The attractancy o f
Dacinae flies to, and their compulsive feeding on, certain male lures are unique
characteristics o f tephritid fruit flies; the reason for such a phenomenon is not fully
understood [1, 3]. These lures play a central role in the chemical ecology o f fruit
flies. Methyl eugenol was first shown to be a strong attractant for fruit fly species
in 1915 [5] and has been used successfully in monitoring and male annihilation pro­
grammes [6], and in estimating population and survival rates o f native males in two
species [7, 8] and fly movements between ecosystems [9]. In Malaysia, dacine fruit
flies are serious pests o f fruits and vegetables: two species o f B. dorsalis complex
(taxa A and B)-and B. umbrosa, which are strongly attracted to ME, ME attracted
species; B. cucurbitae, B. caudatus and B. tau, which are attracted to CL and RK,
CL/RK attracted species; and B. latifrons (not attracted to latilure, alpha-ionol,
under tropical conditions) and B. arecae, which are not attracted to the known lures.
The biology, ecology, behaviour arid control o f economically important fruit fly spe­
cies have been reviewed elsewhere [1-3]. In this overview, because o f space limita­
tion, emphasis is placed on the central role o f fruit fly attractants, secreted as
ecohormones, in understanding the complex interrelationships between plant, fruit
fly (Bactrocera species) and predator.

2, CLASSIFICATION OF ECOHORMONES

Semiochemicals are chemicals (usually volatile) that affect the behaviour o f a

receptive animal at submicrogram or nanogram levels. They may be broadly divided
into two main groups: ecohormones, which are released or secreted naturally into
the environment, and para-ecohormones, which are animal or plant constituents not
secreted naturally, or synthetic chemicals. An ecohormone with intraspecies activity
is known as a pheromone. An ecohormone with interspecies activity which benefits
the releaser is known as an allomone; one which benefits the receiver is a kairomone;
and one which benefits both releaser and receiver species is a syпотопе. Therefore,
an ecohormone may act as a pheromone as well as an allomone, a kairomone or a
synomone. Many ecohormones involved in insect communication may be exploited
as agents for the surveillance and monitoring o f pests and/or pest control within an
insect pest management programme.
 IAEA-SM-327/45 497

3. P L A N T -F R U IT F L Y RELATIONSHIP

Plants form the focal point o f the ecological activities o f fruit flies, including
host finding, oviposition, development, adult feeding and pharmacophagy (consump­
tion o f non-essential chemicals), courtship and mating behaviour, and protective
behaviour against predators. Many o f these behavioural patterns are mediated via
chemical reception and some have been extensively reviewed [1-3].

3.1. Plant ecohormone and para-ecohormone

About 20 plant species from 16 families are known to contain either ME or

RK; their role as kairomones in fruit fly ecology has been discussed elsewhere [3].
Here, we show that the chemicals do not function as kairomones, but actually as syn-
omones or allomones. Several species and varieties o f plants in Malaysia that contain
either o f the chemicals are reported on here.
Five species o f Bulbophyllum (Orchidaceae), section Sestochilus (with rather
large flowers, either solitary or in a subumbellate inflorescence), Bu. cheiri,
Bu. ecomotum, Bu. macranthum, Bu. patens and Bu. vinaceum, attracted the Orien­
tal fruit fly, B. dorsalis, complex species and the Atrocarpus fruit fly, B. umbrosa,
which responds to ME [10]. The fruit fly orchid, Bu. cheiri, has a solitary flower
containing eugenol, compound I and large amounts o f ME [11] that are secreted as
fragrance, acting as a synomone. The flies o f ME-attracted species compulsively
feed on the secreted ME and in return help in pollination [10]. In nature, a flower
o f a fruit fly orchid in bloom is usually covered with fruit flies. The flower stops
producing the fragrance soon after pollination. It has a hinged lip, which moves in
response to the fly’ s weight, thus helping the attracted fruit fly to remove the
pollinarium and produce cross-pollination. The presence o f compound I is interest­
ing; it may function as an allomone against vertebrates (see Section 5) that can
damage the flower. In the other Bulbophyllum species mentioned, the flowers also
release a fragrance, which probably contains ME. These species are currently await­
ing chemical analysis. Fruit flies may be the sole pollinator for these species o f
orchid.
The flowers o f two species o f Spathiphyllum (Araceae), S. cannaefolium and
S. commutatum, attracted fruit flies. In Australia, floral spikes, bearing pollen, o f
S. cannaefolium have been shown to emit benzyl acetate and ME which attracted
fruit flies [11]. However, the spike o f the S. cannaefolium variety found in Malaysia
contains eugenol, which is secreted [11]. Eugenol has a weak attraction for the ME
attracted species. The flower fragrance attracts fruit flies and Trígona wasps, which
help in pollination. Varietal differences in ecohormone also occur in the flowers o f
Couroupita guianensis (Lecythidaceae). A variety has been reported to contain
ME [3]; however, the variety found in Penang contains eugenol (ME not detected)
498 TAN

that attracts a few fruit flies, which thus play a role along with other insects ( Trígona
spp.) as pollinators.
The orchid flower o f Dendrobium superbum (synonymous with D. Anosmum
(Orchidaceae)) has been reported to contain 4-phenyl-2-butanone [12]. Re­
examination o f the floral chemical constituents showed that all three varieties, one
with white flowers, one with purplish flowers and one with white sepals and petals
and purple lip flowers, contain 4-(4-hydroxyphenyl)-2-butanone (RK). Cuelure,
when consumed through pharmacophagy by the male melon fly, is stored as RK in
the rectum. If the fly feeds directly on RK, the ecohormone is sequestered [13]. Male
melon flies, B. cucurbitae, have been observed to feed around the lip, petal and sepal
o f the Dendrobium flowers. It has not yet been observed that the melon fly acts as
a pollinating agent for the orchid flower.
Healthy Ocimum sanctum plants, normally free from insect attack, do not
attract the fruit fly, except when some part o f the plant is damaged [14]. Two varie­
ties o f О. sanctum (Labiatae), one with a green calyx and white corolla, the other
with a violet calyx and corolla, contain relatively large quantities o f ME — more than
60% o f the essential oil. Hence, ME, probably stored in numerous glands (well dis­
tributed on both the upper and lower leaf surfaces and having no pores, as seen under
a scanning electron microscope), is released only upon tissue damage. This may be
the plant’ s defence mechanism against foraging insects because ME is toxic to some
phytophagous insects. Here, ME is a para-ecohormone that acts as an allomone-like
deterrent upon release after plant damage. We believe that more varieties and species
o f Ocimúm (e.g. O. basilicum, currently under study) with such a mechanism will
be discovered.
Piper betel is normally free from insect damage. Healthy plants do not attract
fruit flies. However,'when a leaf is squashed in a cage containing ME attracted fruit
fly species, many males are attracted. The leaf contains two phenyl propanoids —
eugenol and hydroxychavicol [10]. It also contains numerous glands without pores.
The chemicals act as a para-ecohormone in deterring phytophagous insects.

3.2. Plant kairomone

Female fruit flies probably locate host fruits through chemical as well as visual
cues. The odour o f fruit acting as a kairomone may attract female fruit flies to
oviposit in the fruit, resulting in fruit damage caused by larvál development. Infesta­
tion o f 18 host fruit species by different fruit fly species has been reported [15]. A
recent study using ripe host fruits (without ovipuncture) from 14 different species
suspended under the canopy o f a non-host tree in a village ecosystem showed that
most female B. dorsalis (taxon B) flies (caught after landing) preferred to alight on
banana ( Musa paradisiaca), starfruit (Averrhoa carambola) and papaya ( Carica
papaya) (unpublished data from our laboratory). Female fruit flies are attracted to
fruit odour, but the chemical nature o f the kairomone has not been identified.
 IAEA-SM-327/45 499

4. FRU IT F L Y -F R U IT F L Y RELATIONSHIP

4.1. Male response to lures

The number o f B. dorsalis flies attracted to lure traps in the field showed two
peaks in a daily cycle: a high peak between 0800 and 1000 h and a low peak between
1600 and 1800 h. Similar daily cycle peaks have been observed for B. umbrosa.
Under subtropical conditions, different species have a different peak response in the
daily cycle [3].
Conflicting data exist about the effect o f age between 1 and 14 days on respon­
siveness to lures [3]; However, in Malaysia, all major pest species respond to lures
at sexual maturity. In our wind tunnel studies, B. dorsalis (taxa ¡A and В),
В. umbrosa and В. cucurbitae began to respond to their respective lures after 5 d,
with a high percentage o f males responding at 12-14 d after adult eclosion. The
response o f the native male o f B. dorsalis (taxon B) (marked after emergence) to ME
traps set up in a village ecosystem showed that the first marked male trapped was
7 d old. The majority o f marked males caught were between 10 and 16 d after eclo­
sion [16]. . -

4.2. Fruit fly pheromone

Analysis o f the male fruit fly rectal gland after pharmacophagy showed
that ME was converted to other phenylpropanoids for storage. Bactrocera
dorsalis (taxon A ) produced coniferyl alcohol (CF); B. dorsalis (taxon B)
produced three compounds: CF, 2-allyl-4,5-dimethoxyphenol (compound I) and
Z-3,4-dimethoxycinnamyl alcohol (Z compound П) [17]; and B. umbrosa produced
three compounds: compound I, E and Z compound II and 3,4-dimethoxy-
hydroxyallyl benzene. Coniferyl alcohol and compound I are attractive to conspecific
males while CF attracts conspecific females and. stimulates ovipositor extrusion at
close range [14]. The rectal compounds were released into the air during dusk, which
coincides with the fruit fly courtship period [17]. The male rectal gland is known
to be the source o f the sex pheromone in fruit flies [18]. The three fruit fly species
have been observed to form lek. Therefore, CF and compound I act as a male aggre­
gation pheromone. Controversy currently exists over the possible interbreeding o f
taxa A and В (to be placed as separate species) flies in the wild. They were interbred
in the laboratory, and intermediate morphological forms between the two were col­
lected from the wild [19]. The ability to interbreed may be caused by the presence
o f CF, acting as sex pheromone, in the pheromone system o f the two taxa, provided
the sexes are in close proximity.
500 TAN

5. FRUIT FLY-PREDATOR RELATIONSHIP

It has now become possible to explain the strong attractancy o f fruit flies to
certain lures by demonstrating an anti-predation mechanism in several species o f
fruit fly through the endogenous production and secretion o f an allomone after phar-
macophagy. Compulsive feeding o f ME by male flies produced a feeding deterrent
effect on the house lizard, Hemidactylus frenatus. Lizards did not attempt to feed on
fruit flies after initial exposure to flies that had fed on ME [20]. Some lizards would
rather starve than feed on fruit flies (demonstrated in a no choice feeding experiment
over a two to three week period). Lizards learn to avoid preying on fruit flies, proba­
bly through the characteristic yellow markings on the flies. It has been shown that
compound I was a stronger fèeding deterrent than ME against sparrows [21].
Besides functioning as a male aggregation, pheromone, RK may not play an
anti-predation role in B. cucurbitae. The anti-predation mechanism in the melon fly
is achieved at sexual maturity [22], which coincides with the endogenous synthesis
o f ethyl-hydroxybenzoate and 1,3-nonandiol in the male rectal gland. The rectal
compounds were not detected in the first week after eclosion, but they increased
significantly with age, starting from two weeks old [23]. The mechanism probably
involves endogenously synthesized chemicals. The chemical 1,3-nonandiol reduced
the consumption o f treated house flies but did not directly deter predation. It may
play a role in conjunction with other components o f the rectal gland [23]. Further
investigation is necessary to determine the actual rectal component(s) involved as
allomones in the anti-predation mechanism o f the male melon fly.
Sexually mature male fruit flies congregate at dusk and produce a smoke-like
substance originating from the rectum during courtship [24]. This substance, when
sprayed onto the máte, may also protect females from lizards, especially during
copulation, which normally lasts from dusk to dawn. The female flies may be pro­
tected through automimicry, because they resemble the males. Furthermore, both
B. latifrons and B. arecae may be Batesian mimics to gain protection from predators.
This is because the two species resemble B. dorsalis and they are much less abun­
dant. However, their status as Batesian or Mullerian mimics is under review.
Sexually mature males hâve an escape mechanism. They have been observed
to release droplets during escape when disturbed. Spontaneous ejaculation o f rectal
content has been observed in sexually mature males when under stress, e.g. when
they are being anaesthetized with carbon dioxide or when they are being held by a
feather forceps [20, 22]. Such behaviour may distract a potential predator and thus
help the male flies to escape.

6. CONCLUSIONS

Phenylpropanoids in plants may act as ecohormones or para-ecohormones

against insects. Owing to the special relationship that exists between fruit flies and
 IAEA-SM-327/45 501

several species o f Bulbophyllum orchids through synomonal fragrance, care should

be taken in the implementation o f fruit fly eradication program m es to avoid causing
the extinction o f the wild orchids. Plants make use o f the para-hormones as a protec­
tive m echanism against phytophagous insects.
Th e abundance o f plant species containing M E m ay affect the performance o f
the attractant in m ale annihilation, monitoring and surveillance techniques.
The pherom one systems o f other fruit fly species should, be investigated.
Fem ale attractants such as C F should be further investigated and developed for use
in the management o f fruit fly pests.
Th e male m elon fly ’ s ability to deter predation with attainment o f sexual matu­
rity shows a different anti-predation mechanism from that achieved by male B. dor­
salis (taxon B ). T he. latter needs to feed on M E and to convert the chemical to its
analogues in the rectal gland in order to deter lizard predation. This indicates that
M E attracted species m ay be considered m ore advanced than the C L/RK-attracted
species, because the form er spend less metabolic energy in producing allom ones.
It is suggested that suppression o f the anti-predation m echanism through the
m ale annihilation technique m ay leave the fem ale fruit flies unprotected against pre­
dators, which could further reduce the fruit fly population. Em ploying such a
m echanism in an integrated fruit fly management program m e warrants further
investigation.

ACK NO W LED GEM ENTS

Th e author wishes to thank W . Klassen o f the Joint F A O /I A E A D ivision o f

Nuclear Techniques in Food and Agriculture for arranging a financial grant. H e is
also grateful to R . Nishida and Y . C . Toon g for many years o f collaborative research,
and to A . L am b , Department o f Agriculture, Sabah, for providing information arid
many o f the orchid specimens. The research grant provided by the M alaysian
G overnment R & D Geran M P K S N /IR P A 1 -0 7 -0 6 -0 3 -0 3 is greatly appreciated.

REFERENCES

[1] FLETCHER, B .S ., The biology o f Dacinae fruit flies, Annu. Rev. Entómól. 32 (1987)
115-134.
[2] M E TC ALF, R .L ., Plant volatiles as insect attractants, CRC Crit. Rev. Plant Sci. 5 3
(1987) 25 1 -3 0 1 . ' ‘
[3] M E TC ALF, R .L ., Chemical ecology o f Dacinae fruit fliés (Diptera: Tephritidae),
Ann. Entomol, Soc. Am . 83 (1990) 1017—1030.
[4] H O W LETT, F .M ., The effects o f oil o f citronella on two species o f Dacus, Trans. R.
Entomol. Soc. 60 (1912) 41 2-41 8.
502 TAN

[5] H OW LETT, F .М ., Chemical reactions o f fruit flies, Bull. Entomol. Res. 6 (1915)
. 29 7-30 5.
[6] STEINER, L .F ., et al., Eradication o f the oriental fruit fly from the Mariana Islands
by the methods o f male annihilation and sterile insect release, J. Econ. Entomol. 50
(1970) 50 8-509.
[7] T A N , K .H ., Estimation o f native populations o f male Dacus spp. by Jolly’s stochastic
method using a new designed attractant trap in a village ecosystem, J. Plant. Prot.
Tropics 2 (1985) 87 -9 5.
[8] T A N , K .H ., JAAL, Z ., Comparison o f male adult population densities o f the Oriental
and Artocarpus fruit flies, Dacus spp. (Diptera: Tephritidae), in two nearby villages
in Penang, Malaysia, Res. Popùl. Ecol. 28 (1986) 85 -8 9.
[9] T A N , K .H ., SERIT, М . , Movements and population sizes o f native male adult Dacus
dorsalis and Dacus umbrosus (Diptera: Tephritidae) in three ecosystems in Tanjung
Bungah, Penang, Malaysia, J. Plant Prot. Tropics 5 (1988) 17-21.
[10] TO ONG, Y .C ., T A N , K .H ., “ Fruit fly attractants from local plants” , paper presented
at symposium on Tropical Fruit Flies, Kuala Lumpur, Malaysia, 1992.
[11] •LEW IS, J .A ., et.al., Volatile compounds from the flowers o f Spathiphyllum cannaefo-
lium, Phytochemistry 27 (1988) 2755-2757.
[12] FLATH , R. A . , O H IN AT A, K . , Volatile components of the orchid Dendrobium super-
bum RChb., J. Agríe. Food Chem. 30 (1982) 841-842.
[13] NISHIDA, R ., IW AH ASH I, O ., TAN, K .H ., Accumulation o f Dendrobium
(Orchidaceae) flower fragrance in the rectal glands by males o f the melon fly, Dacus
cucurbitae (Tephritidae), J. Chem. Ecol. (in press).
[14] T A N , K .H ., “ Chemical ecology of Malaysian Dacinae fruit flies — Plant-fruit fly -
predator interrelationships” , paper presented at 9th Annual Mtg o f the Int. Soc. of
Chem. Ecologists, Kyoto, Japan, 1992.
[15] TAN, K .H ., LEE; S .L ., Species diversity and abundance o f Dacus (Diptera:
Tephritidea) in,five ecosystems o f Penang, West Malaysia, Bull. Entomol. Res. 72
(1982) 70 9-716.
[16] T A N , K .H ., KIRTON, L .G ., SERIT, М ., “ Age response o f Dacus dorsalis (Hendel)
to methyl eugenol in (a) a wind tunnel and (b) traps set in a village, and its implication
in population estimation” , Fruit Flies (Proc. 2nd Int. Symp. Colymbari, Greece, 1986)
(ECONOM OPOULOS, A .P ., Ed.), Elsevier Science Publishers, Amsterdam (1987)
4 2 5-43 2.
[17] NISHIDA, R ., et al., Accumulation o f phenylpropanoids in the rectal glands o f males
o f the Oriental fruit fly, Dacus dorsalis, Experientia 44 (1988) 53 4-53 6.
[18] FLETCHER, B .S ., The storage and release o f a sex pheromone by the Queensland fruit
fly, Dacus tryoni (Diptera: Trypetidae), Nature (London) 219 (1968) 63 1-632.
[19] K H O O , S .G ., Y Ó N G , H .S ., Department o f Zoology, University o f Malaysia, Kuala
Lumpur, Malaysia, personal communication, 1992.
[20] T A N , K .H ., NISHIDA, R ., “ Fruit fly allomones against predation by lizard” , Proc.
Nad IRPA Seminar, Kuala Lumpur, Malaysia, Vol. 1 (1992) 23 7-238.
[21] NISHIDA, R ., FU KAM I, H ., Sequestration o f distasteful compounds by some phar-
macophagous insects, J. Chem. Ecol. 16 (1990) 151-164.
 IAEA-SM-327/45 503

[22] T A N , К .H ., “ Effect o f sexual maturity in male Bactrocera cucurbitae on lizard” ,

paper presented at symposium on Tropical Fruit Flies, Kuala Lumpur, Malaysia, 1992.
[23] NISHIDA, R ., et a l., Volatile components o f male rectal glands o f the melon fly, Dacus
cucurbitae Coquillett (Diptera: Tephritidae), Appl. Entomol. Zool. 25 (1990) 105-112.
[24] K U B A , H ., SOKEI, Y . , The production o f pheromone clouds by spraying in the melon
fly, Dacus cucurbitae Coquillett (Diptera: Tephritidae), J. Ethol. 6 (1988) 105-110.
 IAEA-SM-327/47

CONTROL OF THE CHINESE CITRUS FLY,

Dacus citri (CHEN), USING
THE STERILE INSECT TECHNIQUE

Huasong W ANG, Heqin ZHANG

Institute for Application o f Atomic Energy,
Chinese Academy o f Agricultural Sciences,
Beijing,
China

Abstract

CONTROL OF THE CHINESE CITRUS F L Y , Dacus citri (CHEN), USING THE STERILE
INSECT TECHNIQUE.
Artificial rearing o f the Chinese citrus fly, Dacus citri (Chen), is described. It was
found that the appropriate irradiation stage for sterilization treatment is at the last pupal phase,
one or two days before emergence, and that 9 krad o f irradiation is a suitable dose for steriliz­
ing D . citri. A total o f 56 000 and 95 000 irradiated sterile males o f D . citri were released
in the Zhonglian orchard (about 34 ha) in Huishui County, Guizhou Province, in 1987 and
1989, respectively. The release ratio o f sterile to native fruit flies was 12.5:1 and 45 :1. The
percentage o f oranges damaged by D . citri dropped from 7.5 to 0.005.

1. INTRODUCTION

It is known that Dacus citri.(Chen) only occurs in China. This pest was first
recorded in two provinces during the 1940s [1]; it was then found in four other
provinces during the 1960s [2]. It occurs in all the citrus producing regions in China
(except for Zhejiang and Guandong Provinces), ranging climatically from temperate
to subtropical to tropical zones. Dacus citri is highly injurious to sweet orange and
other citrus species, even though it produces only one generation per year. In some
orchards, infestation o f fruits by this pest has been estimated to amount to more than
50% [3]. During the 1950s, infestation in some counties in Sichen Province was
higher than 80%. On the basis o f a recent report from the Huaihua Plant Station in
Hunan Province, infestation in some areas is still as high as 80%. In 1987, the
infestation in Bangong township, Lodian County, Guizhou Province, was 22% ; in
one o f these orchards it rose; to 100%.
Over the last decade, China has made great economic progress. The living
standards and purchasing power have increased and the demand for vegetables and
fruits has become greater. To meet this increasing demand, large areas o f orchard
have been cultivated, especially for orange production. The area covered by citrus

505
506 WANG and ZHANG

orchards rose to 1 million hectares in 1989 and the annual output o f citrus fruit
increased to 4.56 million tonnes, from 0.9 million tonnes in 1980. Following the
expansion o f citrus orchards, great efforts have been made to protect citrus fruit and
the damage caused by D. citri has been reduced. However, the losses caused by
D. citri are still heavy. Therefore, it is necessary to find new ways o f coping with
D. citri.

2. MATERIALS AND METHODS

2.1. Artificial rearing

Wild pupae collected from Lodian County, Guizhou Province, southwest

China, were placed in sand, soil or sawdust. The influence o f temperature, the water
content o f the sand, etc., on the survival and emergence rates o f the overwintered
pupae were measured.
Paired adults were fed on a complementary nutrient and water in a glass jar;
cellophane and an agar ball covered with gauze were also added for oviposition.
Incubation o f the eggs was carried out in culture dishes. The eggs were placed
on the filter paper, which was saturated with physiological saline.
The artificial larval diet put in the container (a diameter o f 28 cm and a height
o f 15 cm) had a thickness o f 2 cm for rearing.

2.2. Effect o f radiation on the sterility o f D. citri

One or two days before emergence the pupae were irradiated with 6, 9, 12 and
15 krad o f gamma irradiation at a dose rate o f 89.5 rad/min.1
The irradiated adults were placed in cages as mating crosses o f SÇ x Scr,
SÇ x N a , Ser x N 9 and N Ç x N cr. Each irradiation treatment was repli­
cated four or five times and each cage contained five to ten pairs o f adults covered
with a citrus tree branch.
The mating time, lifespan and damage done to the citrus fruits, etc., were
recorded.

2.3. Release test

The release o f irradiated sterile Chinese citrus flies was conducted in the
Zhonglian orange orchard in Huishui County, Guizhou Province, southwest China,
which covers about 34 ha and has 4800 orange trees. The orange orchard is isolated

1 1 rad = 1.00 X 10"2 Gy.

 IAEA-SM-327/47 •' 507

by the surrounding hills, where there are neither citrus trees nor wild hosts o f D. citri
such as trifoliate orange.
The irradiated sterile pupae were mixed with fluorescent powder in containers
and then placed in the cages. After the female adults were removed, the irradiated
sterile male adults labelled with fluorescent powder were released. Six to ten sites
evenly distributed in the orchard were chosen as the releasing locations. Traps,
baited with a solution o f brown sugar, hydrolytic protein and brewer’ s yeast, were
used to capture the wild adults and recapture the released, labelled, sterile adults o f
D. citri.

3. RESULTS AND DISCUSSION

3.1. Artificial rearing

The survival and emergence rates o f the overwintered pupae were 92.66 and
78.10% in the sand compared with 82.64 and 75.90% in the soil and 0 and 0% in
the sawdust, respectively. Therefore, fine sand was considered to be a favourable
substance for preserving the pupae. This also proved that the 10-15% water content
in the fine sand was appropriate, and that the temperature should not drop below 0°C
for several days.
The peak o f emergence was at about 12—14 h. Mating occurred at least by day
7 after emergence, days 12—17 on average, and was most concentrated at 12-14 h
under 100-1000 lux o f illumination.
Dacus citri is a polygamous fly. Under single pair rearing o f a total o f 55 pairs,
60% mated once, 34.55% mated twice and 5.54% mated three times. The preovipo-
sition period ranged between 23 and 69 d, most often 33-42 d. On average, the
fem ale laid 81.6 eggs three times within 10 d.
The best adult complementary nutrient’ was soybean powder + brown sugar
+ brewer’ s yeast. ■ 1
Most o f the females laid eggs on the surface o f the glass jar; only a few eggs
were laid on the cellophane and the gauze net covering the jar or the agar ball. Most
o f the eggs needed 40-60 d to complete development.
The artificial larval diet was selected from 31 formulas. Its ingredients were
the powder o f wheat germ, the powder o f carrot, sucrose, wheat bran and brewer’ s
yeast, juice o f the green orange, agar, citric acid, methylparaben, etc. The pércent-
age pupae obtained was 80.14. The larvae stage was about 66-79 d. Mature larvae
o f a lustrous light yellow colour pupated in the fine sand or soil.

3.2. Effect of radiation on sterility

A series o f tests indicated that there was no significant influence o f the radia­
tion treatments (6—15 krad) on the emergence rate and lifespan o f the adults (Tables I
508 WANG and ZHANG

T A B L E I. IN F L U E N C E S O F D IF F E R E N T
R A D IA T I O N D O S E S O N T H E M A T I N G R A T E O F T H E
C H IN E S E C IT R U S F L Y , D . citri

Treatment Mating ratés ( % ) b

Crosses3
(krad)
1985 1986 Average

S cr X N 9 395 420 408**

6 Na X S9 310 300 305*
So- X S 9 330 315 323*

Scr X N 9 324 305 315*

9 N o- X S 9 310 315 313*
So- X S 9 330 375 353**

So- X N 9 208 105 157

12 N o- X S 9 200 207 . 204
So- X S 9 110 100 105*

So- X N 9 132 120 126

15 N o- X S 9 180 253 217
So- X S 9 100 93 97*

Control Scr X N 9 216 198 207

a S: sterile; N: normal.
b ** and * denote significantly different at P = 0.01 and
P = 0.05 (LSD), respectively, compared with the control.

T A B L E П. IN F L U E N C E O F D IF F E R E N T R A D IA T I O N D O S E S O N T H E
L IF E S P A N S O F C H IN E S E C IT R U S F L Y A D U L T S 3

Treatment Lifespan o f males (d) Lifespan o f females (d)

(krad)
1985 1986 Average 1985 1986 Average

6 35.67 45.96 40.82 A. 43.71 50.42 48.87 A

9 38.00 49.38 ■ 43.69 A 45.83 50.78 48.32 A
12 37.87 42.51 40.19 A 42.33 43.31 42.82 A
15 38.71 43.29 41 .00 A 44.77 45.12 44.95 A

Control 35.42 44.91 40.17 A 32.95 45.73 39.34 A

a Means followed by the same letter are not significantly different at P = 0.01, LSD.
TA B L E Ш . M A T IN G TIM ES OF IR RA D IA TED CHINESE CITRUS FLIES

Treatment 0800 qqqq ю оо цоо 1200 1300 1400 1500 1600 1700 1800 1900 2000 Total
and crosses

6 krad

S9 x N ct 2 6 10 18 11 17 15 22 8 7 4 120
№ x S ct 2 9 20 17 20 21 22 10 24 15 6 2 168
SÇ x S ct 6 5 18 16 17 14 16 18 9 6 1 126

9 krad

IAEA-SM-327/47
.S Ç x Ncr 2 7 8 13 13 11 19 20 13 13 6 1 126
N Ç x Sa 1 6 8 12 14 10 18 17 15 13 7 1 122
S9 x So- 2 5 11 21 15 15 19 16 21 15 7 3 152

12 krad

S9 x N o- 2 4 8 16 8 8 8 3 3 60
N9 x So- 2 3 2 4 6 9 6 5 5 42
S9 x So- 2 7 5 3 5 3 3 2 30

15 krad

S9 x N cf 1 6 6 10 13 14 18 14 9 8 2 101
N9 x Scr 3 5 3 11 7 12 3 2 2 48
S9 x S ct 4 8 5 4 5 2 1 7 1 37

'ontrol 1 2 7 7 13 9 12 7 10 7 3 1 79

509
 510
TABLE IV. RELATION BETWEEN THE IRRADIATION DOSES AND THE CONTROL EFFECTS

Total No. No. of Dissected fruits No. of No. of Damage

Treatment Total No.
of pierced pierced fruits damaged fruits damaged fruits rate
and crosses of fruits
fruits dissected No. of eggs No. of larvae dissected checked (%) .

6 krad
S Ç X NO" 31 17 12 13 6 7 9.92 32.00
N 9 x Sо 41 20 15 4 5 6.67 16.27
S ç x Sa 29 23 15 2 1 1.53 5.28

9 krad

WANG
S ç x N o- 42 24 17 0 0 0
N 9 x So- 28 9 6 0 0 0

and Z H A N G
S 9 x So- 23 15 12 0 0 0

12 krad
S 9 x No- 13 8 5 0 0 0
N 9 x So- 22 13 12 0 0 0
S 9 x So- 16 3 2 0 0 0

15 krad
S 9 x Ñcr 23 8 6 0 0 0
N 9 x So- 27 15 9 0 0 0
S9 x So 7 0 0 0 0 0

Control 34 17 12 1 70 9 12.75 37.50

Note: (1) The number of damaged fruits dissected is the number of fruits containing eggs and larvae from the dissected fruits.
(2) The number of damaged fruits checked is the number of damaged fruits dissected times the total number of pierced fruits divided by
the number of pierced fruits dissected.
 IAEA-SM-327/47 511

and П), and that the mating time o f the irradiated D. citri was not changed
(Table Ш ). H ow ever, for both males and females treated with 6 and 9 krad, the mat­
ing rate increased significantly. Some control occurred at 6 krad, but complete con­
trol did not occur in the mating crosses until'9 'krad was applied (Table IV ).
Therefore, 9 krad o f irradiation (for pupae irradiated one or two days before
emergence) was determined as the appropriate sterile irradiation dose and irradiation
phase [4].

3.3. Releasing test

Using the labelling and recapture method, the longest dispersal distance o f
D. citri was found to be about 1500 m in a favourable wind direction. It appears that
movement and transportation o f damaged citrus fruits containing the larvae o f
D. citri are the main ways o f spreading infestation.
From the number o f wild and treated adults o f D. citri captured at different
sites in the orange orchard and the times at which they were captured, it was deduced
that D. citri prefers flying and is active in a breeze. A lso, it was proved that irradi­
ated sterile adult D. citri still retained the behaviour o f native populations in the field.
A total o f 56 000 and 95 000 sterile males irradiated with 9 krad were released
at a ratio o f 12.5:1 and 45:1 (sterile: native fruit flies) in 1987 and 1989, respec­
tively. The percentage o f oranges damaged by D. citri dropped from 7.5 to 0.005
during the same period [5].
In 1989, about 48 000 irradiated sterile female D. citri removed from the
Zhonglian orange orchard were released in another small orange orchard, about
1 ha. The percentage o f damaged oranges dropped from 0.55 to 0.003. Although the
release ratio was not obtained, it appeared that release o f irradiated sterile females
was effective, although many o f the orange fruits were scarred because the females
still produced oviposition marks.
Because preliminary control o f D. citri using the sterile insect technique has
been successful, a larger release programme is to be carried out in all orange
orchards with about 100 000 orange trees throughout Huishui County during the next
five years.

REFERENCES

[1] C H E N , F.J., W O N G , F.P., Study on citrus maggot in Jiangjin county, Agrie. Sci. 1
(1943) 46-59.
[2]S U N , Z. Y., ‘‘The study and control of Dacus citri’ ’
, Chinese Plant Protection Science,
Scientific Publishing House, Beijing (1961).
[3] C H E N , S.H., T w o new Dacinae from Szechwan, Sinensia 11 1, 2 (1940) 131-135.
 WANG and ZHANG

LIU, Q.R., Z H A O , C.D., Effect of radiation on the sterility of Chinese citrus fly, Acta
Agrie. Nucl. Sinica 6 1 (1991) 46-50.
W A N G , H.S., et al., Control of Chinese citrus fly Dacus citri by male sterile tech­
nique, Acta Agrie. Nucl. Sinica 5 3 (1990) 135-138.
 IAEA-SM-327/48

Bacillus thuringiensis ENDOTOXINS

ACTIVE AGAINST Chilo partellus
AND Glossina morsitans morsitans

E .O . OSIR, G .N . M A G O M A ,
M .W . V U N D L A , E. K E N Y A
International Centre o f Insect Physiology and E cology,
Nairobi

Abstract

Bacilliis thuringiensis E N D O T O X I N S A C T I V E A G A I N S T Chilo partellus A N D Glossina

morsitans morsitans.
Bacillus thuringiensis crystal endotoxins were isolated by centrifugation on linear
sucrose gradients. Analysis of the crystals by gel electrophoresis revealed that the major c o m ­
ponent of the Chilo partellus active crystal endotoxin was a protein of M r ~ 130 kilodalton.
The Glossina morsitans morsitans active crystal endotoxin gave a major protein band of
M r ~ 120 kilodalton. Upon solubilization under alkaline p H and reducing conditions, the
C. partellus and G. m. morsitans crystal endotoxin yielded protoxins of M r ~ 63 and
M r ~ 64 kilodalton, respectively. Activation of the C. partellus protoxin with bovine
trypsin resulted in no apparent change in the molecular weight. However, treatment with
bovine chymotrypsin or C. partellus midgut homogenate resulted in a shift in the mole­
cular weight of the protoxin to a toxin of M r ~ 60 kilodalton. Similarly, treatment of
G. m. morsitans protoxin with bovine trypsin gave a toxin of M r — 62 kilodalton, but
bovine chymotrypsin gave a toxin of M r ~ 60 kilodalton. Staining with periodic acid
Schiff reagent revealed that both the crystal endotoxins were glycosylated. The carbohydrate
moieties were of the high mannose type, as shown by staining with fluorescein isothio-
cyanate conjugated-concanavalin A. Rabbit antibodies against C. partellus protoxin cross­
reacted with the G. m. morsitans toxin.

1. IN TROD U CTION

The use o f synthetic chemical pesticides for the control o f insect pests and
disease vectors poses a great risk to human health and the environment [1]. Further,
selected mortality o f the more susceptible genotypes follow ing repeated application
o f insecticides has resulted in the rapid development o f resistance among several
insect pests and disease vectors [2]. Consequently, the last half o f this century has
seen efforts to develop alternative pest management strategies that are non-polluting
and environmentally benign. One such strategy that has attracted much interest
involves use o f the naturally occurring bacterium Bacillus thuringiensis (Bt), as a
biopesticide. Bacillus thuringiensis (Berliner) is a gram positive soil bacterium that

513
514 OSIR et al.

synthesizes a parasporal, proteinaceous, crystalline inclusion during the sporulation

cycle [3]. These inclusions consist o f proteins known as delta-endotoxins or insectici­
dal crystal proteins (ICPs). Upon ingestion by a susceptible insect, the crystals are
solubilized in the insect midgut to release the ICPs. The ICPs are then activated by
midgut proteases to form the toxins. The toxins then interact with midgut epithelium,
disrupting membrane integrity and leading ultimately to insect death [4].
Although much work has been reported on the use o f Bt as a biopesticide, there
is still little information on the molecular basis o f the insecticidal properties o f these
toxins. O f particular interest also is the understanding o f the molecular basis for
selective toxicity o f the toxins.
This paper reports on the physical-chem ical properties o f two Bt strains active
against the econom ically important crop pest Chilo partellus and the disease vector
Glossina morsitans morsitans. Understanding the properties o f these endotoxins is
considered an essential first step in elucidating their mode o f action.

2. M A TE R IA L S A N D M ETH ODS

2.1. Experimental insects and bacteria

Adult tsetse flies (G. m. morsitans W estw ood) and third instar С. partellus
larvae were supplied by the Insect and Animal Breeding Unit o f the International
Centre o f Insect Physiology and E cology (ICIPE).
Bacillus thuringiensis strain M F4B/2 was obtained from the ICIPE M icrobial
Bank. This strain was originally isolated in Kenya and found to be active against
C. partellus [5]. The Bt (Tikki) strain active against adult G. m. morsitans was iso­
lated from dead or dying flies at the Kenya Veterinary Research Laboratories,
Kabete [6]. Strain M F4B/2 was grown in a m odified nutrient broth medium [7]. The
Tikki strain was grown in nutrient broth (O xoid Ltd, Basingstoke, Hants, United
Kingdom ). After a starter culture (220 rev./m in, 2 8 °C , 24 h), the culture was
aliquoted into media and incubated in a shaker incubator (220 rev./m in, 2 8 °C ,
72 h).

2.2. Isolation of the endotoxin crystals

The M F4B/2 strain crystals were isolated as follow s. The spores, crystals and
cell debris were pielleted from a 72-h culture by centrifugation (10 000g, 10 min,
4 °C ). The resulting pellet was first washed with distilled water ( x 3), layered on
to a linear sucrose gradient (4 0 -7 0 % ) and then centrifuged (8000g, 1 h, 15°C ) in
a Beckman M odel L5-50 ultracentrifuge. The resulting interphases were washed free
o f sucrose by centrifugation (5000g, 20 min, 4 °C ) using distilled water. The inter­
phase samples were checked for crystals m icroscopically after staining with Sm irnoff
 IAEA-SM-327/48 515

stain [8]. The interphase with the most crystals was layered on to a liñeár sucrose
gradient (4 0 -7 0 % ) and centrifuged (7000g, 1 h, 15°C ) as described above. The
resulting single interphase from this step was washed with water and stored
at - 2 0 ° C .
The relatively large Tikki crystals were isolated by single step, low speed
centrifugation (1100g, 10 min, 4 °C ) o f the 72-h culture. The crystals that settled at
the bottom o f the tube were then washed ( x 3) with distilled water and stored at 4 °C .

2.3. Solubilization and activation of the endotoxin crystals

The M F4B /2 crystals were dispersed in 50m M Na 2C 03 .H Cl (pH 10.5) con ­

taining 20m M dithithreitol (D T T ) and incubated (3 h, 3 7 °C ). A fter incubation,-the
suspension was centrifuged (10 000g, 10 min, 4 °C ) in a M inifiige. The resulting
supernatant and the pellet fractions were stored at - 2 0 ° C . The Tïkki crystals were
solubilized in 50m M Na2C 0 3.H Cl (pH 9.5) containing lOmM D T T for 30 h at
3 7 °C . After incubation, the suspension was centrifuged (10 000g, 10 min, 4 °C ).
The supernatant fraction and the pellet fraction were stored at - 2 0 ° C .

2.4. Enzymatic activation

The supernatant fraction resulting from solubilization o f the M F 4B /2 crystals

was extensively dialysed against 0 .1 M T ris-H C l buffer, pH 8.5, and incubated with
com m ercial bovine trypsin (5 :1 , wt/wt) or bovine chymotrypsin (5 :1 , wt/wt) at 3 7 0С
for 1 h. A fter incubation, the mixture was centrifuged (10 000g, 10 min, 4 °C ) in
a M icrofuge. The same procedure was repeated using the solubilized Tikki protoxin.

2.5. Protein determination

Protein estimation was carried out using the Pierce Bicinchoninic açid (B C A )
protein assay method. The absorbances were measured at 562 nm using a M odel
D U -50 Beckman spectrophotometer. Bovine serum albumin (B SÁ) was used as the
protein standard.

2.6. Polyacrylamide gel electrophoresis

Electrophoresis on denaturing polyacrylamide gels was carried out according

to Laemmli [9]. Gradients were cast using a BRL gradient maker. The low and high
molecular weight standards were obtained from Bio-Rad. The gels were stained for
protein with Coom assie Brilliant blue in 50% methanol/9 % acetic acid and destained
in 50% methanol/9 % acetic acid.
516 OSIR et al.

2.7. Carbohydrate analysis

The presence o f carbohydrate moieties on the M F4B/2 and Tikki endotoxins

was determined using the periodate Schiff stain [10] and fluorescein isothiocyanate
conjugated-concanavalin A (F IT C -C on A ) [11]. In the latter case, the protein
samples were transferred electrophoretically on to nitrocellulose paper using the
N ova-B lot system (Pharmacia). The nitrocellulose paper was then soaked in
concanavalin A buffer (50m M NaCl, 50m M T ris-H C l, Im M CaCl2, Im M M gC l2,
pH 7.0) prior to staining with F IT C -C on A .

2.8. Production of antibodies and immunological procedures

The toxin samples ( — 1.0 m g) were emulsified in complete Freund’ s adjuvant

and injected subcutaneously at several sites into a New Zealand rabbit. A booster
injection ( ~ 0 . 5 mg) in Freund’ s incomplete adjuvant was administered intra­
muscularly four weeks later. After a further tw o weeks, the rabbit was bled through
the main ear artery and the serum prepared as previously described [ 12 ].
Double radial immunodiffusion [13] was performed on glass plates using
1% agarose in phosphate buffered saline (PBS: 0 .2 M NaCl, 0 .02M sodium phos­
phate, 0.02 % NaN 3, pH 6.7). The samples were allowed to diffuse for 24 h at room
temperature in a moist chamber. The plates were then extensively washed in PBS
to remove the unreacted proteins and then dried using filter paper. Staining for pro­
tein was.carried out using Coom asie Brilliant blue, as described in Section 2.6.

3. RESULTS

3.1. Isolation and properties of the endotoxin crystals

The C. partellus (M F 4B /2) active endotoxin crystals were separated from the
spores and cell debris by centrifugation in a linear sucrose gradient. The crystals
form ed a distinct band in the low er half o f the centrifuge tube. The cell debris formed
tw o bands in the upper half o f the tube, while the spores form ed a pellet at the bottom
o f the tube. The purity o f the crystals obtained by this procedure was ascertained by
Sm irnoff staining. In the case o f the G. m. morsitans (Tikki) active crystals, separa­
tion was achieved by a simple procedure involving centrifugation o f the crystal,
spore and cell debris mixture at a low speed. Under these conditions, the large Tikki
crystals settled at the bottom o f the tube, leaving the spores and cell debris in the
supernatant fraction.
The M F4B/2 crystals were analysed by SDS-polyacrylam ide gel electro­
phoresis (PAGE) and found to contain only one major polypeptide of
M r ~ 130 kilodalton (Fig. 1). The crystal endotoxin could be solubilized under
 IAEA-SM-327/48 517

1 2 3 4 5
3
h
щ
----- 200

m «-----116
& ------ 97

$4 ----- 66
1И& . 9
9 <9

<¡B ------ 43

SÉ
FIG. 1. Polyacrylamide gel electrophoresis of MF4B/2 crystal proteins. The protein samples
were separated by dissociating gradient PAGE (4-15%), as described in Section 2. (I) Native
crystal; (2) insoluble crystal (pellet); (3) solubilized crystal (protoxin); (4) toxin (trypsin
treated protoxin); (5) bovine trypsin; and (6) molecular weight markers (Bio-Rad) (in
kilodalton).

5
to — 200

— 116
97 — — 97
67 — — 67.

45 — — 45

31 —

21 —

14 —
/

FIG. 2. Polyacrylamide gel electrophoresis o f Tikki crystal proteins. The protein samples
were separated by gradient SDS-PAGE, as described in Section 2. (1) Low molecular weight
markers; (2) native crystal; (3) insoluble crystal (pellet); (4) solubilized crystal (protoxirt);
and (5) high molecular weight markers (Bio-Rad) (in kilodalton):
518 o sm et al.

1 2 3 4 5 6

f - 20°

31 — • шт» :

WmlÊlÊÊÊImÊÊKlÊËSIU
21 — ■ НИМ'

lillIB lllllillllS lI^ ílllllilB eilli^ B i

FIG. 3. Polyacrylamide gel electrophoresis showing activation o f the Tikki protoxin. The
protein samples were separated by gradient SDS-PAGE, as described in Section 2. (1) Low
molecular weight markers (Bio-Rad); (2) protoxin; (3) toxin (trypsin treated protoxin);
(4) toxin (chymotrypsin treated protoxin); (5) toxin (chymotrypsin + trysin treated protoxin);
and (6) high molecular weight markers (in kilodalton).

high pH and reducing conditions to form the protoxin. Analysis by S D S -P A G E

showed the M F4B /2 protoxin to be com posed o f a single major protein band
(M r ~ 63 kilodalton) (Fig. 1). Similarly, analysis o f the T ik k i crystals by S D S -
PAG E gave only one major polypeptide band (M r ~ 120 kilodalton) (Fig. 2).
H ow ever, compared with the M F4B/2 crystals, the T ik k i crystals required a longer
solubilization time. The protoxin form ed by this procedure had a molecular weight
o f M r ~ 64 kilodalton (Fig. 2).
Enzymatic activation o f the M F4B/2 and T ik k i protoxins to the toxins gave
interesting results. Treatment o f the M F4B /2 protoxin with bovine trypsin resulted
in no apparent change in the molecular weight o f the protoxin (Fig. 1). Treatment
with either G . m . m o r sita n s or S p o d o p te r a e x e m p ta midgut homogenates also gave
similar results (data not shown). How ever, when the protoxin was treated with either
com m ercial chymotrypsin or C . p a r te llu s midgut homogenate, a toxin of
M r ~ 60 kilodalton was form ed (data not shown).
In the case o f T ik k i protoxin (M r — 64 kilodalton), activation with com m er­
cial bovine trypsin gave a toxin o f M r ~ 62 kilodalton (Fig. 3). On the other hand,
comm ercial bovine chymotrypsin or a mixture o f trypsin and chymotrypsin gave a
toxin o f M r ~ 60 kilodalton (Fig. 3). Similarly, G . m . m o r s ita n s midgut hom o­
genate also gave a toxin with M r ~ 60 kilodalton (data.not shown).
 IAEA-SM-327/48 519

FIG. 4. Staining for carbohydrates. The protein samples were separated by gradient SDS-
PAGE and the gel stained using the PAS stain, as described in Section 2. The arrow indicates
the position of the protoxin band.

\ ‘
x ill illl l

FIG. 5. Double radial immunodiffusion. The centre well had the antibodies against the
MF4B/2 protoxin (x), while the peripheral wells had the following samples: (a) MF4B/2
protoxin; (b) Tikki protoxin; and (c) Ae. aegypti active Bt protoxin.
520 OSIR et al.

3.2. Carbohydrate moieties of the endotoxins

The presence o f carbohydrate moieties on the M F4B/2 protoxin was tested by

staining the S D S -P A G E gels with the PAS stain. A s shown in Fig. 4, the protoxin
reacted with the stain. The same result was obtained with the T ik k i protoxin. Further,
F IT C -C on A also reacted with both the M F4B/2 and T ik k i protoxins (data not
shown).

3.3. Immunological studies

Antibodies to both the M F4B/2 and T ik k i protoxins were prepared in rabbits.

In double radial immunodiffusion studies, the antibodies against the M F4B/2
protoxin showed very slight cross-reactivity with the T ik k i protoxin (Fig. 5).
H ow ever, no cross-reactivity was detected with protoxin from another B t strain
active against A e d e s a e g y p ti (Fig. 5 ). In a related experiment, antibodies against the
T ik k i protoxin cross-reacted strongly with the protoxin o f M F4B/2 (data not shown).

4. DISCUSSION

The lepidopteran stalk borer, C. p a r te llu s , is an important pest o f maize and

sorghum in many parts o f the tropics. Infestations usually result in severe crop
losses. Similarly, tsetse flies are the vectors o f trypanosomes, the causative agents
o f African trypanosomiasis [14]. Current control measures for both C. p a r te llu s and
G . m . m o r s ita n s rely largely on a number o f chemical insecticides available in a vari­
ety o f formulations. How ever, many o f these compounds are also toxic to a wide
range o f insects and non-target organisms such as ants, spiders and parasitoids, as
well as some predatory birds, which are o f great value in regulating pest populations.
An alternative to chemicals is use o f naturally occurring pathogens for pest control.
The most useful organisms to date are a group o f toxin producing bacteria, B t [15].
An attractive set o f virtues o f B t is that it is readily degraded in the environment,
has no harmful effects on non-target organisms and, so far, only one or two cases
o f resistance in the field have been reported [16]. Recent w ork at the ICIPE led to
the isolation of several B t strains, two of which are active against
C. p a r te llu s and G . m . m o r s ita n s [5]. Development o f these strains for field applica­
tion is currently under way.
A major drawback in the use o f B t as a biopesticide is the rather narrow range
o f hosts susceptible to a particular isolate. T o promote the widespread use o f B t, it
would be important to increase the host range, possibly by genetic manipulations.
H ow ever, before this can be achieved, it is essential that there should be a clear
understanding o f the molecular basis for selective toxicity and o f the genetic basis
for toxin production. This understanding could also provide a means o f tailoring
toxins o f high potency for specific pests. The two strains mentioned above constitute
 IAEA-SM-327/48 521

the basis for such studies in the project we are currently undertaking. This paper
presents some initial findings on the biochem ical characterization o f the endotoxins
derived from the two strains.
Although several procedures o f isolating Bt crystals have been reported
[17, 18], a linear sucrose gradient was found to be most suitable for isolating the
M F4B/2 crystals. O ver 90% o f the crystals form ed a discrete interphase in the lower
half o f the centrifuge tube. This was in contrast to the relatively large Tikki crystals,
which had a higher density than that o f the spores and consequently settled to the
bottom o f the tube during low speed centrifugation. These observations indicated that
for each strain a specific procedure for isolating crystals has to be established. The
freshly prepared M F 4B /2 and Tikki crystals yielded a single major polypeptide band
on S D S -P A G E , with other minor bands occasionally appearing. The intensity and
number o f these minor bands increased with storage o f the crystals, probably due
to endogenous proteases associated with the crystals. Solubilization converted the
crystals into protoxins o f M r ~ 63 kilodalton in both cases. Compared with the
Tikki crystals, the M F4B /2 crystals were more completely solubilized under high
pH conditions and in the presence o f a reducing agent. V ery little solubilization was
achieved in the absence o f the reducing agent in both cases. Enzymatic conversion
o f the protoxins into the toxins gave interesting results. Compared with the Tikki
protoxin, treatment with trypsin did not appear to alter appreciably the molecular
weight o f the M F4B /2 protoxin. Further, chymotrypsin caused a shift in the m olecu­
lar weight o f the M F4B /2 and Tikki protoxins. In the case o f M F 4B /2, chymotrypsin
gave results similar to those o f C. partellus midgut homogenate.
Many Bt toxins have been reported to be glycosylated [4]. In this study, both
the M F4B/2 and Tikki protoxins were shown to be glycosylated by a general carbo­
hydrate stain. Furthermore, both proteins reacted with the lectin concanavalin A ,
suggesting the presence o f N linked high mannose carbohydrate chains. How ever,
it is possible that the О linked glycosyl residues are also present on the protoxins.
The exact role o f the carbohydrate moiety on these toxins remains unknown.
H ow ever, it has been proposed that they may play a role in insect toxicity [19].
Although M F4B/2 and Tikki had many similarities, they appeared to be im munologi-
cally distinct. H ow ever, it should be noted that the im munological relationship
between the protoxins was tested using the relatively insensitive technique o f
immunodiffusion. It will be interesting to repeat the tests using the more sensitive
technique o f immunoblotting. Other studies aimed at characterizing these two
Bt toxins more fully are under way.

ACKNOWLEDGEMENTS

This work was supported by grants from the United States A gency for Interna­
tional Development and the OPEC Fund.
522 OSIR et al.

REFERENCES

[1] H U TS O N , H .D ., ROBERTS, T .R . (Eds), Insecticides, W ile y, New Y o rk (1985)

10- 11.
[2] B R O W N , A .W .A ., “ Epilogue: Resistance as a factor in pesticide management” , Proc.
X V International Congress, Washington, D C (1977) 816-824.
[3] H O FTE, H ., W H IT E L E Y , H .R ., Insecticidal crystal proteins of Bacillus
thuringiensis, M icro b io l. Rev. 53 (1989) 242-255.
[4] G IL L , S.S., C O W LES, A .W ., P IE T R A N T O N IO , P .V ., The mode o f action o f
Bacillus thuringiensis endotoxins, Annu. Rev. Entomol. 37 (1992) 615-636.
[5] BR O W N BR ID G E, М ., Isolation o f new entomopathogenic strains o f Bacillus
thuringiensis and Bacillus sphericus, Israel J. Entomol. 23 (1989) 109-113.
[6] O TIE N O , L .H ., K O TE N G O , P., International Centre o f Insect Physiology and
Ecology, N airobi, personal communication.
[7] M A G O M A , G .N ., OSIR, E .O ., International Centre o f Insect Physiology and
Ecology, N airobi, unpublished data. '
[8] S M IR N O FF, W .A ., A staining method fo r differentiating spores, crystals and cells o f
Bacillus thuringiensis (Berliner), J. Invert. Pathol. 4 (1962) 384-386.
[9] L A E M M L I, U .K ., Cleavage o f structural proteins during the assembly o f the heads o f
bacteriophage T 4, Nature (London) 227 (1970) 680-685.
[10] K A P IT A N Y , R .A ., ZEBR O W SKI, E .J., A high resolution PAS stain fo r polyacryl­
amide gel, Anal. Biochem. 56 (1973) 361-369.
[11] F U R L A N , М ., PERRET, В .A ., B EC K, E .A ., Staining o f glycoproteins in poly­
acrylamide and agarose gels w ith fluorescent lectins, Anal. Biochem. 96 (1979)
208-214.
[12] OSIR, E .O ., LA B O N G O , L .V ., U N N IT H A N , G .C ., A high molecular weight
diapause-associated protein from the stem-borer, Busseola fusca: Purification and
properties, Arch. Insect Biochem. Physiol. 11 (1989) 173-187.
[13]. O U C H T E R LO N Y , O ., “ Antigen-antibody reaction in gels” , Handbook o f Im m uno­
diffusion and Immunoelectrophoresis, Ann A rb o r Science Publishers, Ann A rb o r, M I
(1958).
[14] H O A R E , C .A ., The Trypanosomes o f Mammals, Blackwell Scientific Publishers,
O xford (1972).
[15] BURGES, H .D . (E d.), M icro bia l Control o f Pests and Plant Diseases, 1970-1980,
Academic Press, London (1981).
[16] K IR C H , K ., SC H M U TTE R E R , H ., L o w efficacy o f Bacillus thuringiensis (Ber.)
form ulation i n . controlling the diamondback moth, Plutella xylostella (L .), in the
Philippines, J. A ppl. Entom ol. 105 (1988) 249-255.
[17] IB A R R A , J.E., F E D E R IC I, G ., Isolation o f a relatively .non-toxic 65-kilodalton
protein inclusion from the parasporal body o f Bacillus thuringiensis subsp. israelensis,
J. Bacteriol. 65 (1986) 527-533.
[18] T H O M A S , W .E ., E L L A R , D .J ., Bacillus thuringiensis var. israeliensis crystal delta-
endotoxin: Effects on insect and mammalian cells in vitro , J. Cell Sci. 60 (1983)
181-197.
[19] M U T H U K U M A R , G ., NIC KER SO N , K .W ., The glycoprotein toxin o f Bacillus
thuringiensis subsp. israeliensis indicates a lectin-like receptor in the larval mosquito
gut, A ppl. Environ. M icro b io l. 53 (1985) 2650-2655. .
 IAEA-SM-327/49

NATURAL FACTORS
AS POTENTIAL INSECTICIDES

H. BUEDS, C . P YC K E ; A . DE LOOF
Z oolog ical Institute,
Katholieke Universiteit Leuven,
Leuven, .
Belgium

Abstract
NATURAL FACTORS AS POTENTIAL INSECTICIDES.
In order to reduce environmental pollution, it is o f great interest to find alternative
methods for controlling insect pests. With the progress made in the isolation and identification
o f peptides and endogenous toxins from insects, the question can be raised whether or not
these natural factors are potentially useful as insecticides. The aim o f the present study was
to test certain toxins and myotropic peptides isolated from insects for their usefulness as insec­
ticides. Over twenty neuropeptides isolated from different insect species have been isolated
and identified in the laboratory. To date, tests have been carried out on the influence o f four
neuropeptides on the food intake o f Lj larvae o f Mamestra brassicae. Also, the crude haemo-
lymph, as well as some o f the purified fractions o f the Colorado potato beetle, have been
tested. At least one o f the neuropeptides and some o f the compounds present in the haemo-
lymph and the purified fraction D have a negative influence on the development o f L, larvae
o f M. brassicae after oral intake.

1. IN TR O D U C TIO N

T o counteract the progressing resistance towards some com m ercial insecti­

cides and to low er the environmental pollution caused by them, efficient and environ­
mentally safe methods have to be developed continuously. The use o f insect growth
regulators designed from juvenile hormone was a first success in this respect. Espe­
cially in recent years, our knowledge o f the endocrine system o f insects has increased
tremendously, particularly as the result o f the identification o f rather a large number
o f neuropeptides. Furthermore, progress has also been made in the isolation and
identification o f endogenous toxins from insects. The question can be raised whether
or not peptides are potentially useful as insecticides. For the time being, large scale
synthesis o f peptides is far too expensive a process to be com m ercially viable for the
production o f insecticides. In addition, investigations on whether such peptides are
active upon oral uptake still have to be carried out. H ow ever, the use o f these
peptides and protein toxins, provided they result in lethal effects follow ing oral

523
524 BUEDS et al.

uptake, might perhaps be realized by generating genetically transformed plants into

which genes coding for peptides/proteins have been introduced.
Notable results have been attained with the insect toxins produced by Bacillus
thuringiensis. The insecticidal activity resides in crystalline inclusion bodies
produced during sporulation o f the bacteria, which are com posed o f proteins specifi­
cally toxic against a variety o f insects. Different strains o f B. thuringiensis differ in
their spectra o f insecticidal activity. M ost are active against Lepidoptera, but some
strains specific to Diptera and Coleoptera have been identified.
The feasibility o f engineering plants that defend themselves against lepi-
dopteran insects that are sensitive to the B. thuringiensis Berliner insect toxin has
been shown by Vaeck et al. [1]. H ow ever, some species belonging to the Noctuidae,
such as Heliothis and Spodoptera, an important group o f pest insects, are less sensi­
tive to com m on strains o f B. thuringiensis. In this regard, it might be essential to
find other natural factors that cause lethal effects follow ing oral intake. For transfor­
mation, however, the problem is to find out which proteins or peptide genes are also
suitable for transfection into plants.
The aim o f the first phase o f our study was to identify peptides and proteins
which, upon ingestion, cause lethal effects in a relatively short period.

2. TESTED PRODU CTS

2.1. Neuropeptides

In our laboratory, over twenty new insect neuropeptides were isolated and fully
sequenced. Then they were synthesized. So far, we have tested four such synthetic
peptides that were originally isolated from Locusta migratoria and Sarcophaga bul-
lata. During isolation, the biological activity was monitored by observing the effect
o f fractions on changes in the frequency or amplitude o f spontaneous contractions
o f the cockroach (Leucophaea maderae) proctodeum (hindgut). This bioassay is an
excellent detection method for isolating peptides from different insect species.
Three o f the four peptides tested showed m yotropic effects. The one isolated
from head extracts o f Sarcophaga was a myoinhibiting peptide. Their amino acid
sequences were:

(1) Thr-Asp-Val-Asp-H is-Val-Phe-Leu-Arg-Phe-N H 2

(from head extracts o f S. bulldta [2]),
(2) A la-P ro-G ln-A la-G ly-Phe-Tyr-G ly-V al-A rg-N H 2
(from head extracts o f L. migratoria [3]),
(3) Gly-Phe-Lys-Asn-Val-Ala-Leu-Ser-Thr-Ala-Arg-G ly-Phe-N H 2
(from accessory glands o f the males o f L. migratoria [4]),
(4) Ala-Ala-Leu-G ln-G lu-Lys-Ala-G ly-Ser-Ile-V al-A rg-Tyr-N H 2
(from the Malphigian tubules o f L. migratoria [5]).
 IAEA-SM-327/49 525

2 .2 . Insect toxins

W e also tested toxic proteins present in the haemolymph o f the Colorado potato
beetle for possible lethal effects on Mamestra larvae. This haemolymph contains four
different proteinaceous toxins [6]. Some are toxic to mammals and others to insects.

2 .3 . Sim m ondsin

Finally, we carried out some orienting experiments with Simmondsin, a gluco-

side isolated from the seeds o f the jo jo b a plant, Simmondsia califomica. This
compound strongly reduces appetite and food intake in mammals. A ccording to some
endocrinologists, it may perhaps exert its action by affecting gastrin
synthesis/release. Since insects also have gastrin like peptides, it was tempting to
find out whether or not this compound could exert an effect on fo o d intake in this
insect species.

3. CHOICE OF TEST IN SECT SPECIES

A ll the natural factors were tested on fo o d intake and lethal effects in the
phytophageous noctuid Mamestra brassicae L. This species occurs in large parts o f
Europe and Asia and can cause considerable damage to crops. O f extreme im por­
tance, however, for this species is that an artificial diet is available so that the insects
can be reared all year round. A s mentioned in Secton 1, some species o f Noctuidae
are less sensitive to com m on strains o f B. thuringiensis, so we hoped that our tested
natural factors may be effective.

4. E XPER IM E N TAL SET-UP

A s the larvae hatched from the egg they were divided into two groups o f
20 larvae each. One group was fed a treated diet, the other group a normal diet. The
compounds to be tested were dissolved in water and small cubes o f artificial diet
were covered with a given amount o f the solution. The larvae were checked daily
for repellent and lethal effects.

5. RESULTS

Preliminary results indicate that none o f the tested natural factors has a repel­
lent effect on the larvae. Only one o f the neuropeptides (Ala-Pro-Gln-Ala-Gly-Phe-
T yr-G ly-V al-A rg-N H 2), when administered orally, has a lethal effect within five
 1 2 3 4 5 6 7

_^C NP HL _b_ D ^_S

Day

FIG. I. Survival of the L¡ larvae of M. brassicae following treatment with: crude

haemolymph (HL) (500 ng/larva), purified fraction D (D) (1 mg/larva), neuropeptide
(Ala-Pro-Gln-Ala-Gly-Phe-Tyr-Gly-Val-Arg-NH2) (NP) (17.5 ng/larva), Simmondsin (S)
(500 ¡¿g/larva) and control (C).

days o f emergence. The crude haemolymph, as well as one o f the purified fractions,
have a deleterious effect on the larvae o f Mamestra. Simmondsin was also lethal for
the larvae. Figure 1 shows the survival o f the L] larvae o f M. brassicae treated with
the various natural factors.

6. DISCUSSION

At this point it is clear that at least some o f the natural factors tested are lethal
for the L , larvae o f M. brassicae. H ow ever, none o f the factors was totally pure,
with the exception o f Simmondsin. For the active neuropeptide, there was a possibil­
ity the impurities still present after synthesis are responsible for the lethal effects.
How ever, the preliminary results o f the feeding experiments with the purified
neuropeptide gave the same results.
 IAEA-SM-327/49 527

Further research is being carried out to find out if the alimentary canal is the
primary target o f the active compounds. Preliminary results with larvae treated with
the active fraction isolated from the crude haemolymph o f the C olorado potato beetle
indicate that these compounds disturb the structure o f the alimentary epithelium.
Additional research is required before w e can draw conclusions about the possibili­
ties for practical use o f all these compounds.

ACKNOWLEDGEMENTS

The research was sponsored by the Instituut ter bevordering van onderzoek in
nijverheid en landbouw (IW O N L ) o f Belgium; we thank M . Cokelaere for providing
the Simmondsin sample.

REFERENCES

[1] VAECK, M ., et al., Transgenic plants protected from insect attack, Nature (London)
328 (1987) 33-37.
[2] FONAGY, A ., et al., Isolations, primary structure and synthesis o f Neomyosupressin,
a myoinhibiting neuropeptide from the grey fleshfly, Neobelleria bullata, Insect
Biochem. (submitted).
[3] SCHOOFS, L., et al., Locustatachykinin in and IV: Two additional insect
neuropeptides with homology to peptides o f the vertebrate tachykinin family, Regul.
Peptides 31 (1990) 199-212.
[4] PAEMEN, L., et al., Lom -AG-М уotropin: A novel myotropic peptide from the male
accessory glands o f Locusta migratoria, Peptides 12 (1990) 7-10.
[5] T IP S , A ., K ath oliek e U niversiteit L euven , personal com m unication.
[6] AMELINCKX, М ., Katholieke Universiteit Leuven, personal communication.
 IAEA-SM-327/50

QUANTITATIVE ANALYSIS
OF THE FATE OF A PESTICIDE
AFTER ITS APPLICATION TO INSECTS

R. T Y K V A
Institute o f Organic Chemistry and Biochemistry,
C zechoslovak Academ y o f Sciences,
Prague

B. B E N N ETTO V Á
Institute o f Entom ology,
C zechoslovak A cadem y o f Sciences,
Ceské Budéjovice

Czechoslovakia

Abstract
QUANTITATIVE ANALYSIS OF THE FATE OF A PESTICIDE AFTER ITS APPLICA­
TION TO INSECTS.
Utilization o f biologically active compounds applied in insect pest management has
been studied by a complex method using carbamate juvenoid radiolabelled in different posi­
tions with ,4C and/or 3H. The technique enables the penetration, distribution and excretion
o f the applied compound and/or its metabolites to be followed quantitatively.

1. IN TR O D U C TIO N

Radiolabelling o f insect growth regulators [1, 2] is a highly sensitive and effec­

tive technique for studying their fate, which is also closely related to environmental
protection. From this point o f view , the relation between the applied and utilized
doses is o f special importance. Knowledge o f the sites o f interaction o f the inves­
tigated compound with the insect tissues and quantitative analysis o f all radioactive
compounds, including expired C 0 2, make a comparison o f the applied and utilized
doses possible.
The results obtained with this technique prove that quantitative analysis o f the
fate o f a pesticide after its application to insects is possible. A t present, this technique
is also being applied in the IA E A programme on tsetse fly control.

529
530 TYKVA and BENNETTOVÁ

2. MATERIALS AND METHODS

The flesh fly, Sarcophaga bullata Parker, was used in the experiments. Larvae
were reared on b eef liver and pupariated in sawdust. One-day-old sugar fed females
were employed. Since the strain o f Sarcophaga used is partly autogenous, no liver
was offered during the experiments.
Previously synthesized [3] carbamate juvenoid (signature W 328) was labelled
separately at the Institute o f Nuclear B iology and Radiochemistry, Prague, in two
different positions with 14C using chemical synthesis, and independently with 3H by
the catalytic exchange solution-gas method. The radiochemical purity after prepara­
tion [4] was higher than 9 9 % . The stability was examined during storage and
chromatographic purification was carried out before application, if necessary. The
labelled positions are summarized in Fig. 1.
An acetone solution (5 /xL) o f labelled W 328 was topically applied to the upper
part o f the thorax. The flies were narcotized at given time intervals and then sepa­
rated into three main parts (head, thorax and abdomen) to measure the macrodistri-
bution o f radioactivity. Washing was carried out three times in a 5 m L solution o f
chloroform -m ethanol (1 :1 ). The body parts were dissolved in 1 m L o f scintillation
cocktail (Backman, BTS 450), then 15 m L ó f toluene scintillator were added. A
liquid scintillation spectrometer, Backman LS 6000 SE, was used; the standard
deviation was less than 1% in all the measurements.

i . 'л , ' " 0

° x ° 1 ^ •
r S - C H 2- ^ - 0 - C H 2- C H 2- N H - C - 0 - E t

! iv

Specific activity
Labelling No.
(GBq/mmol)

Benzyl-3H 82.3 I
Benzyl-3H 323 II
Benzyl-3H 557 III
Carbonyl-HC 0.185 IV
Benzene-U-u C 2.294 V

FIG. 1. Labelling o f the applied carbamate juvenoid W328 (molecular weight = 363.4).
 IAEA-SM-327/50 531

T o determine the microdistribution o f radioactivity, the flies were sectioned at

- 2 0 ° C in a cryostat. The section thickness varied from 40 to 400 цт, which is an
infinitely thick layer for 14C ; usüally, 260 /im thick whole body sections are used.
They were mounted on a m icroscopic slide lightly smeared with glycerol-album in.
For the radioactivity distribution, a computer controlled arrangement o f special
semiconductor detectors [5] having a measuring slit o f 1 mm x 1 mm was used.
For the overall balance, ten flies were kept in a gas chamber lined with filter
paper. A ir (2 5 -3 0 m L/s) first passed through the gas washing bottles filled with
sodium hydroxide or water, then through the chamber, and finally through two
bottles filled with 100 m L o f 2M ethanolamine in methylcelosolve.
The radiochromatographic samples were prepared by extraction o f sonified
tissues using m ethanol-chloroform (1 :1 ). The supernatant was cleaned up as
described in R ef. [6]. A fter H PLC, standard detection using a U V /V IS detector at
230 nm and subsequent determination in a flow through detector fitted with a solid
scintillator were carried out [4].

3. RESULTS A N D DISCUSSION

The dependence o f the barrier effect o f the cuticle on the applied dose, is
demonstrated in Table I, expressed as a comparison o f the radioactivity recovered
by surface washing to the radioactivity applied [7]. It follow s from the measured
values that the percentage o f applied radioactivity recovered increased with increas­
ing dose. F or example, the routes o f surface decontamination after application o f
0.081 fig o f tritiated W 328 differed considerably, depending on whether they were
applied alone or simultaneously with 14C . In the latter case, the values for both the
radiotracers were comparable, although their positions in the molecule differed and
the radioactivity applied differed considerably. A lso, the time dependence o f the
radioactivity recovered differs according to the dose applied.
T o appraise the influence o f this barrier effect on the radioactivity levels in the
body, we compared the results obtained with different doses o f the same label
(Table П). Although the relative radioactivity recovered by washing is strongly
dependent on the applied dose, body radioactivity after washing is roughly propor­
tional to the applied radioactivity. Within two orders o f magnitude o f the applied
dose (0 .2 0 6 -1 3 .7 /¿g per fly), the percentage o f applied radioactivity measured
within the whole body was only slightly different during a period o f more than
two weeks. Therefore, with increasing input o f radioactivity in the range o f a b io­
logically active dose, the radioactivity o f a certain tissue could be increased, i f neces­
sary. A ccording to bur measurements o f the detailed radioactivity distribution
(Table Ш ), the fluctuations were partly due to excretion by the gut. The effects o f
washing were mostly evident on the thorax, because this was the site on which the
label was applied.
532 TYKVA and BENNETTOVÁ

T AB LE I. R ELATIV E R A D IO A C T IV IT Y (% OF THE APPLIED V A L U E )

R EC O V E R E D FRO M THE B O D Y SU RFACE A FT E R APPLICA TIO N OF
DIFFERENT DOSES OF DIFFEREN TLY LAB ELLED W 328 (THE M EAN S OF
FIVE FLIES)

Labelled W328 1 П + IV Ш

Dose (jig per female) 0.81 0.081 + 18.8 0.081

Washed off per female H-3 H-3 C-14 H-3

Day (% ) (ng) (% ) (ng) (%) (m) (%), (ng)

1 22.4 181 67.9 68 . 63.1 11.9 9.7 7.9

3 6.4 52 38.7 32 38.2 9.2 3.7 3.0
6 3.2 26 25.9 21 20.1 3.8 2.0 1.6
7 1.9 15 18.0 15 18.2 3.4 1.3 1.1

Note: The value o f mass (g) was calculated from the appropriate specific activities o f the labels
(see Fig. 1), assuming their surface stability.

TAB LE П. COM PAR ISO N OF THE R A D IO A C T IV IT Y (% OF THE APPLIED

V A L U E ) INSIDE THE B O D Y W ITH THAT W A SH E D OFF FRO M ITS
SU RFACE AFTE R APPLICA TIO N OF ( 14C RING) W 328 DOSES W ITHIN
THREE ORDERS OF M A G N ITU D E (THE M EAN S OF SIX FLIES)

Dose (fig per female) 0.206 L72 13.7

Day Washed off Body Washed off Body Washed off Body

1 6.5 6.1 23.8 10.4 48.2 8.5

3 4.1 4.3 7.1 7.8 33.6 8.0
6 3.2 3.3 6.4 8.4 18.5 7.9
10 1.0 3.7 1.4 4.8 5.9 5.6
17 0.6 3.0 0.8 4.2 5.0 4.2
 IAEA-SM-327/50 533

T A B L E Ш . INFLUENCE OF W A SH IN G ON THE R E L A T IV E R A D IO ­
A C T IV IT Y (% OF THE APPLIED V A L U E ) OF THE M A IN B O D Y PARTS
A FT E R A PP L IC A T IO N O F 24.6 fig O F ( 14C RIN G ) W 328 (THE M E AN S OF
THREE FLIES)

Head Thorax Abdomen

Day
+ - + - + -

1 0.3 4.0 2.1 65 2.3 9.0

2 0.3 1.3 2.1 32 2.7 8.2
3 0.25 1.2 2.0 32 4.4 8.5
6 0.18 0.6 1.9 14 2.0 0.8
10 0.10 0.3 1.3 12 1.2 3.8
14 0.08 0.2 1.0 12 0.3 2.5

Note: + denotes washed; — denotes unwashed.

TAB LE IV . C O U N TIN G R ATES M E ASU R ED IN THE

W H O LE B O D Y B Y THE SEM ICO N D U CTO R TO PO -
G RAPH IC TECH NIQUE O N D A Y 4 A FT E R A PP L IC A ­
TION O F 12.4 kBq O F (C A R B O N Y L -14C ) W 328 PER
FLY

Tissue Counts/min (SD ± 3 % )

Thoracic cuticle 201

Head cuticle 63
Ovaries 47
Gut 46
Abdomen cuticle 24
Thoracic muscles 20
Brain 19
534 TYKVA and BENNETTOVÁ

TAB LE V . O V E R A L L B A L A N C E O N D A Y 7 A FT E R APPLICATION OF
198.3 kBq OF (C A R B O N Y L - 14C ) W 328 PER F L Y

Sample Per cent applied activity

Ten unwashed flies 47.7

Expired C 0 2 6.2
Excrements, cage surface contamination and water 31.0
Sugar 2.9

Total 87.8

8 2U AO

Time (min)
FIG. 2. Standard UV and radioactivity detection after high pressure liquid chromatography
o f an aliquot o f the whole body extract from three flies on day 7 after application o f 12.4 kBq
o f (carbonyl-14C) W328 per fly.
 IAEA-SM-327/50 535

Correspondingly, the microdistribution o f the relative radioactivity (Table IV)

shows relatively high values in the thoracic cuticle application site. The thoracic
muscles contain much less radioactivity. A n accumulation was found in the oocytes
already four days after application.
From the measured values o f radioactivity distribution (Tables П- I V ) it
follow s that within the range o f biologically active doses the low er values are prefer­
able for effective utilization o f the applied compound with low surface contamina­
tion. The applied com pound is probably transferred through the cuticle into other
distant body parts, with a certain amount penetrating to a haemolymph [ 8]. During
penetration, there is a possibility o f carbamate juvenoid metabolism in the cuticle,
as has been demonstrated for organophosphorus insecticides [9]. The appropriate
experiments are in progress.
Using (carbonyl- 14C ) labelling, decarboxylation o f the applied W 328 could be
analysed. In such a way, an overall balance was evaluated (Table V ).
The radioactivity distribution within the insect body determines, in all cases;
only the sites o f interaction o f the applied juvènoid with the appropriate insect
tissues, but not the chemical character o f the label. A lso, the com position o f the
washings could be analysed.
High pressure liquid chromatography analysis is used to distinguish between
the applied compound and its labelled metabolites or degradation products. Figure 2
demonstrates such a separation, from which it was calculated that 36% o f the radio­
activity o f the original W 328 was still present (retention time tR = 36.6 min).
Besides, two main radiopeaks were found, one o f which probably corresponds to
keton (1 6 .9 % ) and the other lies in the region o f the polar fractions (2 2 .7 % ). Further
radiochromatographic experiments are now in progress.

ACKNOW LEDGEM ENTS

The authors wish to thank T. Elbert and V . Vlasáková from the Institute o f
Nuclear B iology and Radiochemistry, Prague, for the radiochromatographic
analyses.

REFERENCES

[1] M A T S U M U R A , F., Toxicology of Insecticides, 2nd edn, Plenum Press, N e w York

(1985).
[2] L A N G L E Y , P.A., H O W L , V., G O U C H I , H., Regulation of reproduction in
Rhodnius prolixus by the juvenile hormone mimic pyriproxyten, Entomol. Exp. Appl.
57 (1990) 271-279.
[3] W I M M E R , Z., STREINZ, L., R O M A Ñ U K , М., Preparation of carbamate derivates
of (2-(4-hydroxybenzyl)-l-cyklohexanone with a juvenoid activity, Collect. Czech.
Chem. Co m mun. 50 (1985) 2453-2456.
536 TYKVA and BENNETTOVÁ

[4] T Y K V A , R., C E R N Ÿ , B., E L B E R T , T., M A T U C H A , М., V L A S Á K O V Á , В.,

. “Testing of [4-methyl-3H]-labelled ethyl 2-{4-[(l,4-dioxaspiro[4,5]dec-6-yl)methyl]
phenoxyjethyl carbamate”, Pesticide Analysis and Chemistry (Proc. Conf. Ceské
Budéjovice) (TRÍSKA, J., Ed.), Czechoslovak Academy of Sciences, Ceské
Budéjovice (1991) 91-99.
[5] T Y K V A , R., M A C H Á C K O V Á , I., K R E K U L E , J., JISL, R „ A topographic method
for studying uptake, translocation and distribution of inorganic ions using two
radiotracers simultaneously, J. Exp. Bot. 43 (1992) 1083-1087.
[6] B E N N E T T O V Á , В., STREINZ, L., T Y K V A , R., W I M M E R , Z., “In vivo meta­
bolism of a carbamate juvenoid in the flesh fly, Sarcophaga bullata: Determination of
metabolites”, Insect Chemical Ecology (Proc: Conf. Tábor, 1990) (H R D Ÿ , I., Ed.),
Academia, Prague (1991) 501-504.
[7] L A N G L E Y , P.A., F E L T O N , T., S T A F F O R D , K., G O U C H I , H., Formulation of
pyriproxyfen, a juvenile hormone mimic, for tsetse control, Med. Vet. Entomol. 4
(1990) 127-133.
[8] G E R O L T , P., M o d e of entry of contact insecticides, J. Insect Physiol. 15 (1969)
563-580.
[9] F O N T A N, A., Z E R B A , E., Integumental esteratic activity in Triatoma infestons and
its contribution to the degradation of organophosphorus insecticides, Comp. Biochem.
Physiol. 79C (1984) 183-188.
 IAEA-SM-327/51

USE OF JUVENOIDS FOR SUPPRESSION

OF INSECT REPRODUCTION

F. SEH NAL
Faculty o f Biological Sciences,
University o f South Bohemia,
Ceské Budëjovice,
Czechoslovakia

Abstract

USE O F JUVENOIDS F O R SUPPRESSION O F INSECT REPRODUCTION.

Juvenoids are bio-analogues of the juvenile hormones that are vital regulators of insect
development and reproduction. Insects treated with juvenoids often perish owing to a block
of embryogenesis or metamorphosis, but certain treatments cause sterility by disrupting
gametogenesis and egg development in otherwise normal adults. Thousands of structurally
diversified juvenoids are known, but only five compounds are currently on the market and a
few others with promising properties are in the stage of laboratory testing. High species speci­
ficity and low environmental stability are the major obstacles in the commercialization of juve­
noids. Juvenoids are used to suppress cockroaches, fleas and ants in households, to protect
some stored commodities, to control mosquitoes and certain other dipterans, to defeat tor-
tricids and homopterans in orchards, and to increase the silk yield in sericulture. Other appli­
cations, such as against various homopterans and against tsetse flies¿ are being developed. A
combination of juvenoid treatments with biological control and their incorporation into sterile
insect technique programmes deserve more attention.

1. IN TR O D U C TIO N

Insect development and reproduction are under hormonal control and are ham­
pered when the hormonal regulation is disturbed. The practical potential o f artifi­
cially induced hormonal imbalance was recognized by W illiams [1], who discovered
that topical applications o f juvenile hormone (JH) extracts caused developmental
defects with lethal consequences. Since then, the structure o f five JHs has been eluci­
dated. Juvenile hormones seem to be unique to insects, in which they control
metamorphosis, reproduction, and several other processes and functions. There are
differences in the role o f JHs in various insect groups [2]. For example, in Lepidop-
tera, metamorphosis ensues after a drop in JH titre and egg formation often occurs
in pupae when JH remains low . By contrast, in cyclorrhaphous Diptera, metamor­
phosis appears to begin in the last larval instar in the presence o f JH, and egg produc­
tion requires a high titre o f this hormone.

537
538 SEHNAL

The effects o f JHs are mimicked by juvenoids, whose structure is often only
vaguely related to the known hormones. A few juvenoids were isolated from plants
and several thousand were synthesized. Examples o f the structural variety o f ju ve­
noids can be found in Refs [3, 4] and in a number o f other publications. Juvenoids
have also been linked to inert moieties such as sugars and fatty acids to obtain juveno-
gens from which the hormonally active component is released upon enzymatic
cleavage in the target insect. Research efforts were further directed to substances that
interfere either with the production or with the action o f JH [5]. These ‘ anti-JH
agents’ , as well as the juvenogens, seem to be far from practical utilization.
Arguments for and against the use o f juvenoids in insect pest management have
been put forward for the last twenty years [6 , 7]. It is agreed that juvenoids suffer
two major disadvantages. First, the activities o f juvenoids are invariably species
specific and some insect groups are difficult to affect with any compound [ 8 , 9].
Second, the desired effects in most pest control programmes are death or sterility,
both o f which occur days or weeks after the juvenoid treatments. Owing to these
limitàtions, progress in juvenoid commercialization has been slow and to date only
five compounds are available on the market. The flaws o f juvenoids, however, are
compensated for by their selectivity, which makes them safe to most non-target spe­
cies, and their biodegradability, which prevents accumulation o f residues in the food
chain.
Available knowledge and experience indicate that juvenoids are not likely to
becom e a general means o f controlling insect pests. On the other hand, in certain
cases their use appears to be the method o f choice. The aim o f the present paper,
in which the effects and the current use o f juvenoids are briefly reviewed, is to
encourage considerations on combining juvenoid treatments with radionuclear,
molecular and genetic techniques o f pest control. Owing to lack o f space the refer­
ences listed are limited to reviews and selected reports published within the last few
years.

2. V A R IE T Y OF JUVENOID EFFECTS

Juvenoids have been reported to affect all the developmental stages o f insects
[ 10 ], but usually only applications to one or two stages are o f practical significance
in any given species. The most suitable stage must be identified experimentally. The
follow ing paragraphs illustrate the variety o f effects that can be encountered after
juvenoid treatments. The suppression o f reproduction is emphasized.

2.1. Inhibition of embryogenesis

Juvenoids applied to the eggs either directly or via the maternal organism
inhibit embryogenesis. The effects range from the blockage o f cleavage to the hatch­
 IAEA-SM-327/51 539

ing o f slightly defective larvae that are bound to perish. Death usually occurs soon
after hatching, but tobacco hornworm larvae eclosing from fenoxycarb treated eggs
sometimes survive until the second instar [ 1 1 ].

2.2. Effects on polymorphism

Treatments o f larvae in polym orphic species cause a shift towards a certain

morph. Thus, juvenoids interfere with the development o f casts in social insects, the
seasonal polymorphism o f aphids and phase dimorphism in locusts, lepidopterans
and others. These effects have a potential for practical exploitation, but so far they
have not been included in any application scheme.

2.3. Inhibition of moulting

Juvenoids inhibit the production of insect moulting hormones, the

ecdysteroids, and this results in a temporal or permanent blockage in moulting.
Moult is prevented in the larvae o f most endopterygotes when they reach a body size
incompatible with further larval development [12]. Such larvae often respond to
juvenoids by continuous feeding without moulting, and may attain enormous body
sizes. Eventually they perish, but pupation may take place when the juvenoid is
disposed o f. Precisely timed and dosed juvenoid application is used in sericulture to
increase the size o f silkworm larvae, which then produce more silk [13].
In the German cockroach and certain scale insects, juvenoids inhibit moulting
and cause death in first instar larvae.

2.4. Interference with developmental diapause

The life cy cle o f many insects includes a facultative or obligatory diapause that
occurs in various developmental periods and is controlled by differential hormonal
mechanisms. Juvenoids have been reported to terminate early embryonic diapause,
while our experience indicates that they have no effect on diapause at the very end
o f embryonic development when the fully form ed first instar larvae are prevented
from hatching. The effects o f juvenoids on diapause in early larval instars have not
been examined, except for a report on diapause termination in a planthopper [14] .
Diapause in an early part o f the last larval instar, i.e. before the programming for
metamorphosis takes place (such diapausing larvae undergo in som e species station­
ary larval moults), was induced and maintained with juvenoids in several endop­
terygotes. The effect o f juvenoids is variable in the case o f diapause occurring late
in the last larval instar; diapause induction is facilitated but in some species juvenoids
cannot revert stimulation o f development by environmental cues. In contrast, ju ve­
noids invariably terminate pupal diapause. Precisely dosed treatments induce con­
tinuation o f normal development, whereas high doses bring about morphogenetic
defects, as described in Section 2.5.
540 SEHNAL

2 .5 . Derangement of metamorphosis

Inhibition o f morphogenesis in metamorphosing insects is the most frequently

described effect o f juvenoids. In a few insect groups, juvenoid applications to the
penultimate or newly ecdysed last instar larvae inhibit any development towards the
adult form . Such insects ecdyse into an extra larval instar and may undergo metamor­
phosis one or more instars later. M ore frequently, however, the ‘ supernumerary’
larvae are defective and perish. Juvenoid treatments performed in the course o f the
last instar cannot revert the m orphological changes that have already taken place but
inhibit their further progress: the insects moult to intermediate forms between larva
and adult in exopterygotes or between larva and pupa in endopterygotes. Similarly,
juvenoid applications to pupae cause development o f pupal-adult intermediates. The
‘ intermediate’ character o f insects that ecdyse after a juvenoid treatment may be
manifested only internally. The insects may appear as normal adults but their flight
capacity, mating ability or fertility are impaired. A s with all other juvenoid effects,
the maximum extent o f metamorphosis inhibition varied in different insects.
Morphogenetic effects on larval-pupal transformation are absent in cyclorrhaphous
Diptera and probably in some other groups as well.

2.6. Stimulation of reproduction

Stimulation o f egg development is an important function o f JH in many insect

species and it is readily mimicked with juvenoids. Owing to this effect, the inhibition
o f imaginai differentiation described above is occasionally associated with preco­
cious egg formation. Juvenoids also compensate for the lack o f JH that causes imagi­
nai diapause. In the bug Eurygaster integriceps, which has an obligatory imaginai
diapause, the adults becom e sensitive to juvenoids only a couple o f weeks after emer­
gence. Except for this refractory phase, juvenoid treatments o f diapausing adults
induce reproduction and eliminate other diapause syndromes. For example, termina­
tion o f diapause in certain water bugs and carabid beetles is associated with regenera­
tion o f the flight muscles, while in bark beetles and crickets these muscles are
induced to degenerate.
Imaginai diapause is often terminated with juvenoid doses that cause derange­
ments o f reproduction when administered to the non-diapausing adults (see
Section 2.7). N o explanation has been offered for this interesting phenomenon.
Induction o f normal reproduction in an unsuitable season, however, would have
detrimental consequences for the pest population. Such a premature termination o f
diapause with juvenoids was proposed as a control strategy for the pear psylla [15].
 IAEA-SM-327/51 541

2 .7 . Inhibition of reproduction

2 .7 .1 . Indirect effects on reproduction

M ost o f the juvenoid effects listed above have indirect consequences for
reproduction. O bviously no progeny are produced when the treated insects die from
morphogenetic defects. The supernumerary larvae and intermediate form s, which
ecdyse after treatment at the metamorphosing stages, can survive for a long time,
but incomplete development o f the reproductive system prevents their reproduction.
A n exception is found in viviparous aphids that deliver their parthenogenic progeny
via a rupture in the body wall. In most insects, however, reproduction is impaired
when juvenoids cause even a minor m orphological abnormality in the emerged
adults, such as incomplete genitalia rotation in the males or distorted ovipositor in
the females o f cyclorrhaphous Diptera. The interference o f juvenoids with polym or­
phism, moulting and diapause may also inhibit reproduction. For example, juvenoids
suppress the reproductive potential o f termite cdlonies by causing a shift from the
development o f reproductives to that o f soldiers.

2.7.2. Direct reproductive failures

Administration o f juvenoids to the last instar larvae, pupae or adults causes

defects in the differentiation o f gametes and follicular cells. The differentiation is
inhibited in the same manner as any other morphogenetic process and, in cases when
it is separated from the morphogenesis o f other body tissues in time or by threshold
sensitivity to juvenoids, one can obtain sterile, but in all other respects normal,
adults. Inhibition o f spermatogenesis by juvenoid treatment o f larvae has been
achieved in a number o f species. Both under in vivo and in vitro conditions it has
been demonstrated that spermiogenesis is more readily blocked than spermatogene­
sis. Differentiation o f oogonia begins in larvae rather exceptionally and its blockage
with juvenoids has been described only in the fall w ebw orm [16]. In this species,
meiosis in spermatogonia begins in the fourth larval instar and mature sperm are
form ed in the seventh (last) instar. The oogonia enter meiotic prophase in the sixth
instar and diversify into oocytes and trophocytes in the second half o f the last instar.
Juvenoid applications to larvae do not affect the proliferation o f stem cells and
primary gametogonia, but at least in males they prevent the last round o f mitoses by
which the cysts o f synchronously developing gametocytes are form ed. In both sexes,
juvenoids inhibit the onset o f meiosis and block further progress o f gametogenesis
at various steps. Gametocytes whose development is blocked degenerate. N o
gametes are form ed in either sex when webworm larvae receive the juvenoid
hydroprene at a rate o f 10 ^g/specim en in each from the third to the seventh instar.
In the case o f the fall w ebw orm , such specimens moult into larval-pupal intermedi-
542 SEHNAL

ates and perish. In the German cockroach, however, it is possible to cause permanent
sterility in externally normal adults by administering juvenoids to the larvae [17].
Juvenoids also inhibit the differentiation and function o f follicular cells and this
results in the formation o f faulty egg chambers. The chambers either fuse to contain
m ore than one oocyte or remain unusually small, and/or they appear normal but
either vitellogenesis or chorion and m icropyle formations are defective. Imperfect
egg chambers are usually resorbed but occasionally they sustain and non-developing
eggs are laid. Their sterility, however, may also be due to juvenoid deposition in the
egg and consequent blockage o f embryogenesis (see Section 2 .1 ). The latter case
occurs frequently after juvenoid applications to adults. Females are obviously more
susceptible to this effect, but treated males may transfer juvenoids to the eggs during
mating. High doses o f juvenoids administered to the reproducing females o f the fire-
brat, Thermobia domestica, cause irreversible degeneration o f entire о varioles. Such
an effect has not been reported for any other insects and exemplifies great differences
in the sterilizing potential o f juvenoids for different species. In many cases, the cause
o f reproductive failure is not precisely known and may be unrelated to gonads. For
example, reduction in fecundity in Spodoptera litura has been attributed to juvenoid
interference with a humoral stimulation o f oviposition [18].

3. PRACTICAL USE OF JUVENOIDS

3.1. Commercially available compounds

Z oecon Corporation (now Sandoz Crop Protection Ltd) o f California com m er­
cialized three esters o f 3,7,ll-trim eth yl-dodeca-2,4-dien oic acid m ore than a decade
ago. The juvenoid kinoprene was introduced under the name Enstar for the control
o f homopterans on ornamental plants in greenhouses (applied at 0 .1 g /m 2; these and
all follow ing data refer to the active ingredient). T w o other juvenoids had a much
broader spectrum o f application. By 1985, the juvenoid methoprene was formulated
as Altosid to control mosquitoes (280 g/ha) and hom flies (0.02% in mineral blocks
for a feed through application); as Apex against sciarid flies in mushroom cultures
(1 g /m 2); as Kabat for application in tobacco storehouses (10 ppm sprays), and as
Dianex and Diacon to protect stored peanuts (0.0207 g /m 3 in aerosol formulation
and 1.12 g/100 kg peanuts in spray). Bait formulations Pharorid (Z oecon ) and Lafa-
rex (Lachema, Czechoslovakia) proved very effective against the pharaoni ant. The
third juvenoid o f Z oecon Corporation, the hydroprene, was developed for use
against fleas (Precor or, in combination with a pyrethroid, Raid Flea Killer II Plus)
and cockroaches (G encor, or Gencor Plus in a mixture with Permethrin).
Dr. R. Maag Ltd (Switzerland) developed for commercial use the juvenoid
fenoxycarb, which is characterized by the phenoxyphenoxy moiety and a carbamate
functional group. Fenoxycarb is used under the name Insegar (Ciba Geigy, Switzer-
 IAEA-SM-327/51 543

land) to control tortricids and homopterans in orchards (the doses vary from
0.25 to 1 kg/ha with one to three sprays per season), fire ants in the households
(5 -1 0 m g/colon y), mosquito larvae in water reservoirs (0 .0 3 -0 .1 ppm ), and some
stored product pests (4 -1 0 ppm). Fenoxycarb is now also manufactured in Japan, but
its broad use in agriculture is being threatened by reports that fenoxycarb contamina­
tion has prevented cocoon spinning in com m ercially reared silkworms in certain
parts o f Italy and France [19].
The fifth com m ercially available juvenoid, named pyriproxyfen, is produced
by the Sumitomo Chemical Company Ltd, Japan. A s Sumilarv, it is particularly
recommended for the control o f fly development in waste products (0.2 g /m 2) and
o f mosquitoes in water reservoirs (0.01 ppm ). G ood results have also been obtained
with the use ô f pyriproxyfen against aphids, psyllids, white flies and scales [ 20 - 22 ],
and also against the triatomid bug, Rhodnius prolixus [23].
There are at least two juvenoids o f interest in the testing stage. A compound
resembling fenoxycarb, ethyl 2 -{4-[(l,4-dioxaspiro[4,5]dec-6-yl)m ethyl]ph enoxy}
ethylcarbamate, was discovered at the Czechoslovak A cadem y o f Sciences.
It exhibits high activities on several insect groups, psyllids in particular [24].
A structurally unrelated com pound, 4-chloro-5-(6-chloro-3-pyridylm ethoxy)-2-
(3,4-dichlorophenyl)-pyridazin-3(2H )-one, Nissan Chemical Industries, Japan,
shows very promising effects on leafhoppers [15, 25].

3.2. Further perspectives of juvenoid exploitation

Juvenoids harbour a potential for use with other environmentally friendly tech­
niques o f sustained pest suppression. A combination o f juvenoids with the use o f
bioagens, such as parasitoids and predators, is an emerging trend in some pest con ­
trol schemes. Such an approach is possible because juvenoids are often destructive
for the pests but are safe for their natural enemies [26, 27]. There are indications
that juvenoid treatments may even enhance the population growth o f some parasi­
toids; w e observed that treatment o f the eggs o f the sun bug, Eurygaster integriceps,
extends by several times the period in which they can be infested with the parasitoid
Azolcus grandis.
It is possible to envisage the use o f juvenoids as part o f various sterile insect
technique programmes. For instance, the high ovicidal activity o f both fenoxycarb
and pyriproxyfen on the codling moth could be exploited for suppressing the natural
population before sterile insect release. Pyriproxyfen has also been identified as an
effective tsetse sterilant that acts by inhibiting metamorphosis in the offspring o f
treated flies [28]. Luring adults to traps in which they becom e contaminated with
pyriproxyfen was devised as a method for achieving a considerable population reduc­
tion in tsetse [29]. This approach should certainly be considered in tsetse eradication
schemes.
544 SEHNAL

Juvenoids have successfully been tested as a quarantine treatment o f fruits [30]

and this offers a possibility for using them as a supplement to com m odity irradiation.
In summary, exploitation o f juvenoids for enhancing the efficacy o f radio-
nuclear, molecular and genetic techniques deserves attention.

REFERENCES

[1] W I L L I A M S , C.M., The juvenile hormone of insects, Nature (London) 178 (1956)
212-213.
[2] S E H N A L , F., “Growth and life cycles’ ’, Comprehensive Insect Physiology,
Biochemistry and Pharmacology ( K E R K U T , G.A., GILBERT, L.I., Eds), Vol. 2,
Pergamon Press, Oxford (1985) 1-86.
[3] S L A M A , K., R O M A N U K , М., S O R M , F., Insect Hormones and Bioanalogues,
Springer-Verlag, Vienna (1974).
[4] H E N R I C K , C.A., “Juvenile hormone analogues: Structure-activity relationships”,
Insecticidal Modes of Action (COATS, J.R., Ed.), Academic Press, N e w York (1982)
315-402.
[5] S T A A L , G., Anti-juvenile hormone agents, Annu. Rev. Entomol. 31 (1986) 391-429.
[6] E D W A R D S , J.P., M E N N , J.J., “The use of juvenoids in insect pest management”,
Chemie der Pflanzenschutz- und Schàdlingsbekàmpfungsmittel ( W E G L E R , R., Ed.),
Springer-Verlag, Berlin (1981) 185-214.
[7] B O W E R S , W.S., “Prospects for the use of insect growth regulators in agriculture”,
Advances in Invertebrate Reproduction (HOSHI, М., Y A M A S H I T A , O., Eds),
Vol. 5, Elsevier, Amsterdam (1990) 365-382.
[8] S T A A L , G., Insect growth regulators with juvenile hormone activity, Annu. Rev.
Entomol. 20 (1975) 417-460.
[9] S E H N A L , F., “Action of juvenoids on different groups of insects”, The Juvenile Hor­
mones (GILBERT, L.I., Ed.), Plenum Press, N e w York (1976) 301-322.
[10] S E H N A L , F., “Juvenile hormone analogues”, Invertebrate Endocrinology
( D O W N E R , R.G.H., L A U F E R , H., Eds.), Vol. 1, Alan R. Liss, N e w York (1983)
657-672.
[11] M A S N E R , P., A N G S T , М., D O R N , S., Fenoxycarb, an insect growth regulator with
juvenile hormone activity: A candidate for Heliothis virescens (F.) control on cotton,
Pesticide Sci. 18 (1987) 89-94.
[12] GEBLIC, I., S E H N A L , I., Failure of some individuals of Spodoptera littoralis
(Lepidoptera) to produce superlarvae, Acta Entomol. Bohemoslov. 83 (1986) 161-170.
[13] AKAI, H., K I M U R A , K., KIUCHI, М., S H I B U K A W A , A., Increase of silk produc­
tion by repeated treatments with juvenile hormone analogue, J. Serie. Sci. Jpn 54
(1985) 297-299.
[14] M I Y A K E , T., H A R U Y A M A , H., MITSUI, T., S A K U R À I , A., Effects of a new
juvenile hormone mimic, NC-170, on metamorphosis and diapause of the small brown
planthopper, Laodelphax striatellus, J. Pesticide Sci. 17 (1992) 75-82.
[15] K R Y S A N , J.L., “Premature termination of reproductive diapause by a juvenile hor­
mone mimic: A possible control strategy for pear psylla”, Advances in Invertebrate
 IAEA-SM-327/51 545

Reproduction (HOSHI, М., Y A M A S H I T A , O., Eds), Vol. 5, Elsevier, Amsterdam

(1990) 387-392.
[16] B U R O V , V.N., K O Z H A N O V A , N.I., P R A U A , I.I., S H I N Y A E V A , L.I., S E H N A L ,
F., Influence of juvenoid treatment of early larval instars on the gametogenesis in
Hyphantria cunea, Cytologia 30 (1988) 1467-1471 (in Russian).
[17] K O D R I K , D., S E H N A L , F., “Inhibition of reproduction in Blatella germanica by
hydroprene S applied in food”, Regulation of Insect Reproduction ( T O N N E R , М.,
S O L D Á N , T., B E N N E T T O V Á , B., Eds), Vol. 4, Academia, Prague (1989) 237-250.
[18] H A T A K O S H I , М., H I R A N O , М., “Effects of S-71639 on the reproduction of Spodop­
tera litura", Advances in Invertebrate Reproduction (HOSHI, М ., Y A M A S H I T A , O.,
Eds), Vol. 5, Elsevier, Amsterdam (1990) 429-434.
[19] M A U C H A M P , B., INRA, Versailles, personal communication.
[20] M c M U L L E N , R.D., “A report on pyriproxyfen and fenoxycarb for control of pear
psylla, Psyllapyricola Foerster”,Advances in Invertebrate Reproduction (HOSHI, М.,
Y A M A S H I T A , O., Eds), Vol. 5, Elsevier, Amsterdam (1990) 399-402.
[21] Y A M A M O T O , H., K A S A M A T S U , K., “Effects of a juvenile hormone mimic,
S-71639, on the greenhouse whitefly, Trialeurodes vaporariorum, and the green peach
aphid, Myzus persicae, in a greenhouse”, ibid., pp. 393-398.
[22] H A T A K O S H I , М ., S H O N O ; Y., Y A M A M O T O , H., H I R A N O , М ., Effects of
juvenile hormone analog pyriproxyfen on Myzus persicae and Unaspis yanonensis,
Appl. Entomol. Zool. 26 (1991) 412-414.
[23] L A N G L E Y , P.A . , H O W L , V., O O U C H I , H., Regulation of reproduction in Rhodnius
prolixus by the juvenile hormone mimic pyriproxyfen, Entomol. Exp. Appl. 57 (1990)
271-279.
[24] N O V Á K , K., Z E L E N Ÿ , J., S E H N A L , F., “Activities of a new juvenoid on the
representatives of Homoptera”, Chemical Insect Ecology ( H R DY , I., Ed.), Academia,
Prague (1991) 457-460.
[25] M I Y A K E , T., H A R U Y A M A , H . , O G U R A , T., MITSUI, T., S A K U R A I , A., Effects
of insect juvenile hormone NC-170 active on metamorphosis, oviposition and embryo­
genesis in leafhoppers, J. Pesticide Sci. 16 (1991) 441-448.
[26] P E L E G , B.A., Effect of a new phenoxy juvenile hormone analog on California red
scale (Homoptera: Diaspididae), Florida wax scale (Homoptera: Coccidae) and the
ectoparasite Aphytis holoxanthus DeBache (Hymenoptera: Aphelinidae), J. Econ.
Entomol. 81 (1988) 88-92.
[27] N A G A I , K., Effects of a juvenile hormone mimic material 4-phenoxyphenyl-
(RS)-2-(2-pyridyloxy) propyl ether, on Thripspalmi K a m y (Thysanoptera: Thripidae),
and its predator Orius sp. (Hemiptera: Anthocoridae), Appl. Entomol. Zool. 25 (1990)
199-204.
[28] L A N G L E Y , P.A., F E L T O N , T., O O U C H I , H., Juvenile hormone mimics as effective
sterilants for the tsetse fly, Glossina morsitans morsitans, Med. Vet. Entomol. 2 (1988)
29-35.
[29] H A R G R O V E , J.W., L A N G L E Y , P.A., Sterilizing tsetse (Diptera: Glossinidae) in the
field: A successful trial, Bull. Entomol. Res. 80 (1990) 397-403.
[30] Y O K O Y A M A , V.Y., M I L L E R , G.T., Potential of pyriproxyfen as a quarantine treat­
ment for codling moth and oriental fruit moth (Lepidoptera: Tortricidae), J. Econ.
Entomol. 84 (1991) 942-947.
RESEARCH AND DEVELOPMENT ON THE TSETSE FLY
(Session 7)

Chairman
B. BENNETTOVÁ
Czechoslovakia
 IAEA-SM-327/S2

AGE DEPENDENT SAMPLING BIASES

IN TSETSE FLIES (Glossina)
Problems associated with estimating mortality
from sample age distributions*

J.W . H A R G R O V E
Insect Pest Management Initiative,
Tsetse and Trypanosomiasis Control Branch,
Harare,
Zimbabwe

Abstract
A G E D E P E N D E N T S A M P L I N G BIASES IN T S E T S E FLIES (Glossina): P R O B L E M S
ASSOCIATED WITH ESTIMATING MORTALITY FROM SAMPLE AGE
DISTRIBUTIONS.
For a closed (island) population of G. morsitans morsitans Westwood, the probability
per week of capturing females on ox fly rounds was about 0.3 in the first week of life, less
than 0.2 for 27 to 35-d-old flies and greater than 0.4 for flies more than 80 d old. For open
populations, the relative changes in capture probability were measured from the ovarian
age distributions of trap and ox fly round samples. They were used (with the island data) to
show that the age dependent sampling bias of traps for female G. m. morsitans increased
more than sixfold over the first 80 d of life. The age dependent bias for G. pallidipes
Austen taken from odour baited traps is probably at least as serious as for G. m. morsitans.
Estimates of daily mortality from the mark-recapture studies were always (up to 20 times)
higher than estimates from ovarian age samples taken at the same times. The mortalities
recalculated from samples adjusted for sampling biases were closer to, but still lower than,
the mark-recapture estimates. Odour baited targets are successful in controlling tsetse popu­
lations, despite the relatively low probability of treating young females. If sterilants instead
of insecticides were used on the targets, young females could be treated indirectly via treated
males, which transfer the sterilant to virgin females during copulation.
\

1. IN TR O D U C TIO N

Attempts to control tsetse flies (Glossina) in much o f A frica rely increasingly

on the use o f odour baited targets [1]. Because o f their low natural birth rate, a popu­
lation can be eradicated by superimposing, on the natural death rate, a mortality due
to the targets such that the total mortality is greater than about 3.5 % per day [2 -4 ]
by an amount that depends on the level o f pre-adult mortality and the desired rate

* Research carried out with the support of the I A E A under Research Contract
No. 3031/RB.

549
550 HARGROVE

o f population reduction. In deciding on the required target density, it is necessary

to know the natural death rate o f the adult population. Changes in mortality during
the eradication campaign will also produce information on the performance o f the
targets under different circumstances, and may warn about operational problems or
indicate points at which the overall mortality decreases as a consequence o f the
density dependent processes.
Mark-recapture methods [5, 6] can be used to measure mortality [7, 8] for
closed populations but, under m ore natural open condition it is difficult to separate
the effects o f mortality and emigration, and the methods are generally com plex and
time consuming. Under certain conditions [9], mortality can be estimated more
simply from sample age structures, one condition being that the sample should be
random with respect to age or, if not, that the age bias be known and the sample
corrected accordingly.
The ovarian age structure [10] o f female tsetse flies Glossina morsitans
morsitans W estw ood and G. pa.llid.ipes Austen depends on the system used to capture
the flies [11]. T w o sampling methods are o f importance in the current context:
( 1 ) stationary odour baited traps, since these are widely used to survey tsetse popula­
tions and to collect samples for analysis o f the ovarian age structure [3, 12]; and
( 2 ) m obile ox fly rounds, since these were used in an experiment where tsetse
mortalities could be estimated simultaneously by mark-recapture and by analysing
sample ovarian age structures [7]. It is also the only technique for which estimates
have been produced o f the absolute probability o f capture o f tsetse flies o f different
ages [13]. These studies are used here to estimate the age dependent sampling bias
o f odour baited traps and its effect on estimates o f mortality in adult tsetse flies.

2. M ETH ODS

2.1. Estimating mortality from ovarian age data in tsetse flies

In the past, mortalities have been estimated from ovarian age data by non­
linear regression [3, 12], but the method assumes equal variance in each o f the n¡
(the number o f sampled flies in age class i). It can be shown that the maximum
likelihood estimate for the mortality is obtained by solving, for <p, the equation

x 2( l - ¥>4) - X j(l + v + <p2 + <p3) + 4 ^ 4x 3 = 0 (1)

where
7 7 7

*1 = 2 ni; x2 = D in¡; and Хз = Y j ni

i= l i=l i= 1
 IAEA-SM-327/52 551

and where <p = exp (ц + X) fo r mortality p and growth rate X. (Flies in ovarian
category zero have generally been excluded from such analyses, since they are
judged to be severely underrepresented in field samples.) The mortality ц can then
be estimated if the growth rate X is known; the latter is obtained from changes in
population levels estimated, for the present study, by multiple mark-recapture
[5, 6].
The com plex form o f Eq. (1) is due to the pooling o f older flies in ovarian
categories 4 -7 . Thus, ovarian category 4 contains flies from age categories 4, 8 , 12,
etc., which cannot be separated by the dissection technique alone; analogous
problems exist for ovarian categories 5, 6 and 7. The older age classes can be
partitioned approximately on the basis o f their wing fray [13]; this technique is used
in Section 3.

2.2. Estimating age dependent sampling bias

For a population o f N flies, suppose n¡ are in age category i and that for a
sample o f M j flies (using technique j ) m¡ j are in age category i. Then we define the
bias ( b j j ) for age category i and capture system j by

by = (my/MjVOii/N) (2)
Conversely, if the b¡ j are known, the n¡ are estimated by

n¡ = ( т у Н ) /( М г Ьу) (3)

Note that the population age distribution can be estimated even i f N is unknown.
It also follow s that the biases o f the two different sampling systems are related by

b ¡,2 = (m i, 2/M 2) ( M 1/m ijl) b iji (4)

3. RESULTS

In a study o f a closed (island) population o f G. m. morsitans [13], the absolute

age dependent bias was estimated fo r female flies uniquely marked and released at
birth and recaptured on o x fly rounds. The probability o f capture per week (the
approximate interlarval period on the island) can be described by a quadratic function
o f age (Fig. 1(a)). This type o f experiment has so far not been possible to perform
in open populations. H ow ever, the island results (Fig. 1(a)) can be used in conjunc­
tion with ovarian age samples taken from open populations to estimate the age depen­
dent biases for the latter situation.
552 HARGROVE

0.8 г 25
(a) (b )
S' 20
¡t1 0.6 С
ф
5
ce I 15
x>
o
0.4 Ф

It
Q. ф
<0 0.2 о
o ф 5

0.0
0
_i_____ i_____ i------- i_
4 8 12 . 16 20
CL

I
0 1 2 3 4 5 6 7
Ovarian age Ovarian category

(d)

4 8 12 16 20
Ovarian age Ovarian age
(e) w

<a a} 4
ш m

_J___________I___________I___________L - _l_____ I_____ I_____ I_____ I-------- 1--------L-

0 4 8 12 16 20 0 1 2 3 4 5 6 7
Ovarian age Ovarian category

FIG. 1. Age dependent sampling biases in female G. m. morsitans. (a) The relationship,
for flies recaptured on an ox fly round [13], between the age and probability o f capture
(± standard deviation) in successive seven-day periods, (b) Frequency by ovarian category
for flies caught in odour baited traps (solid bars) and on ox fly rounds (open bars) [11].
(c) As for (b), but flies in ovarian categories 4-7 distributed to older categories according
to wing fray [13]. (d) Estimate of true age distribution using information from (a) and (c).
(e) Estimated age dependent sampling bias for flies caught in odour baited traps calculated
using the estimated true distribution (d) and the trap sample (c). (f) As for (e), but the
distributions in (c) and (d) are pooled prior to calculation to give only ovarian categories 0-7,
as in (b).
 IAEA-SM-327/52 553

<0
XI

70 . 60 90 100 .1 1 0 120 130

Week of e x p e r im e n t

FIG. 2. Population levels and mortality rates for female G. m. morsitans and G. pallidipes
(all filled symbols) on Antelope Island, Lake Kariba, Zimbabwe [7]. (a) Weekly mark-
recapture estimates o f the total population numbers (logarithmic scale). Unes are fitted by
regression to estimate growth rates over times shown, (b) Estimates o f mortality resulting
from: mark-recapture (circles), analysis o f original ovarian age distribution (squares) and
o f 'corrected' distributions (triangles).
554 HARGROVE

Note that Fig. 1(a) shows the absolute probability o f capture at various ages,
rather than the bias. The latter is obtained by multiplying each o f the probabilities
(Pÿ) by the constant factor

where к is the number o f age classes and the n¡ and m¡j are defined in Section 2.2.
Figure 1(b) shows the percentage in each ovarian age class for samples o f
G. m. morsitans taken from odour baited traps and o x fly rounds in a study on an
open population [11]. Flies in ovarian categories 4 -7 can be partitioned, using the
wing fray o f individual flies, to produce an approximate sample age distribution
(Fig. 1(c)). The true age distribution (Fig. 1(d)) can then be estimated (from
Eq. (3)), using the ox fly round sample age distribution (Fig. 1(c)) and the capture
probability function (Fig. 1(a)).
The age dependent sampling bias for the traps (Figs 1(e) and 1 (f)) is now
estimated (from Eq. (4)), using the estimated true age distribution (Fig. 1(d)) and the
trap sample age distribution (Fig. 1(c)). The bias changes more radically with age
for traps than for the ox fly. rounds (see Figs 1(a) and 1(e), noting the tenfold differ­
ence in scale), increasing approximately linearly with age for most o f the fly ’ s life.
The vital parameters of island populations of G. m. morsitans and
G. pallidipes [7] have been estimated by mark-recapture (Fig. 2). During this
experiment, three ovarian age samples o f trapped flies were also obtained. The daily
mortalities estimated by mark-recapture were up to five times higher than those
from age structure analysis for G. m. morsitans and up to 20 times higher for
G. pallidipes (Fig. 2 (b)). This could be explained by a bias in trap samples towards
old flies. The mortalities estimated from age structures ‘ corrected’ using the bias
estimates (Fig. 1(d)) were indeed higher than the originals, but still substantially
low er than the mark-recapture estimates in each case (Fig. 2(b)).

4. DISCUSSION

The differences between mortality estimates from mark-recapture and age

distribution analysis are only partly explained by an age dependent trapping bias o f
the order estimated in Section 3. The remaining difference is not easy to explain.
The bias function used has, perforce, been derived by a circuitous route and involves
many assumptions regarding the equality o f behaviour o f the tsetse flies at three
different localities and times in Zimbabwe. The trapping systems used on the island
[7] and at Rekomitjie [11] were also somewhat different. It may be that the bias
was even more pronounced on the island than assumed here.
 IAEA-SM-327/52 555

Alternatively, mortality may be overestimated by the mark-recapture method.

How ever, theoretical arid computer simulations show that the method used results
in mortality estimates with biases that are less than 5% o f the mean [ 8, 14]. Since
the changes in the capture probability o f G. m. morsitans females on o x fly rounds
wére also small (Fig. 1(a)), there is no reason to suppose that the mark-recapture
estimates o f mortality are seriously in error as a result o f the changing probabilities
o f recapture by this system during the fly ’ s life.
Part o f the discrepancy might be due to the fact that mark-recapture mortalities
were averages for all adult flies, whereas the estimates derived from ovarian
dissections ignored the first ovarian category, which has above average mortalities
[13]. In addition, one cannot be certain that the island population had a stable age
distribution at the time the samples for ovarian ageing were collected.
At the practical level, the results suggest that odour baited targets are success­
ful in controlling tsetse populations, despite the fact that the (important) younger
elements o f the female population are treated with a relatively low probability.
This difficulty could be overcom e, and the technique thereby enhanced, by using
sterilants instead o f insecticides on the target [15]. The young elements o f the female
population could then be treated indirectly via the males, which pick up the sterilant
from the target and transfer it to virgin females during copulation.

ACKNOWLEDGEMENT

The Government o f Zim babwe is grateful for the continuing support given by
the Joint F A O /IA E A Division o f Nuclear Techniques in F ood and Agriculture,
IA E A .

REFERENCES

[1] VALE, G.A., etal., Odour-baited targets to control tsetse flies, Glossina spp.
(Diptera: Glossinidae), in Zimbabwe, Bull. Entomol. Res. 78 (1988) 31-49.
[2] HARGROVE, J.W., Tsetse dispersal reconsidered, J. Animal Ecol. 50 (1981)
351-373.
[3] ROGERS, D.J., etal., Local variations in the population dynamics o f Glossina
palpalis palpalis (Robineau-Desvoidy) (Diptera: Glossinidae). I. Natural population
regulation, Bull. Entomol. Res. 74 (1984) 403-423.
[4] HARGROVE, J.W., Tsetse: The limits to population growth, Med. Vet. Entomol. 2
(1988) 203-217.
[5] JOLLY, G.M ., Explicit estimates from capture-recapture data with both death and
immigration — stochastic model, Biometrika 52 (1965) 225-247.
[6] SEBER, G.A.F., A note on the multiple recapture census, Biometrika 52 (1965)
249-259.
556 HARGROVE

[7] VALE, G.A., et al., Field trials o f baits to control populations o f Glossina morsitans
morsitans Westwood and G. pallidipes Austen (Diptera: Glossinidae), Bull. Entomol.
Res. 76 (1986) 179-193.
[8] HARGROVE, J.W., “ Population estimation from mark-recapture data: Equations
for a pooled mark system and for pooled data, with applications to a study on island
populations o f tsetse flies in Zimbabwe” , Sterile Insect Technique for Tsetsë Control
and Eradication (Proc. Panel Vóm, 1988), IAEA, Vienna (1990) 79-114.
[9] CAUGHLEY, G., Analysis o f vertebrate populations, Wiley, Chichester (1977)
234 pp.
[10] SAUNDERS, D.S., Age determination for female tsetse flies and the age composition
o f samples o f G. pallidipes Aust., G. palpalis fuscipes Newst. and G. brevipalpalis
Newst., Bull. Entomol. Res. S3 (1963) 579-595.
[11] HARGROVE, J.W., Ovarian ages o f tsetse flies (Diptera: Glossinidae) caught from
mobile and stationary baits in the presence and absence o f humans, Bull. Entomol.
Res. 81 (1991) 43-50.
[12] VAN.SICKLE, J., PHELPS, R.J., Age distributions and reproductive status o f
declining and stationary populations o f G. pallidipes Austen (Diptera: Glossinidae) in
Zimbabwe, Bull. Entomol. Res. 78 (1988) 51-61.
[13] HARGROVE, J.W., Age-dependent changes in the probabilities o f survival and
capture o f the tsetse, Glossina morsitans morsitans Westwood, Insect Sci. Applic. 11
(1990) 323-330.
[14] HARGROVE, J.W., BORLAND, C.H., Pooled population parameter estimates from
mark-recapture data, Biometrics (accepted for publication).
[15] HARGROVE, J.W., LANGLEY, P.A., Sterilizing tsetse (Diptera: Glossinidae) in the
field: A successful trial, Bull. Entomol. Res. 80 (1990) 397-403.
 IAEA-SM-327/53

DETECTION AND IDENTIFICATION

OF TISSUE SPECIFIC LECTINS
OF THE TSETSE FLY, Glossina tachinoides
Midgut lectin activity with
lipopolysaccharide binding specificity*

P. V O L F
Department o f Parasitology,
Charles University,
Prague

L. GRUBH O FFER, M , M U SK A
Institute o f Parasitology,
Czechoslovak Academ y o f Sciences,
Ceské Budëjovice

Czechoslovakia

Abstract
D E T E C T I O N A N D I D E N T I F I C A T I O N O F T I S SU E SPECIFIC L E C T I N S O F T H E
T S E T S E FLY, Glossina tachinoides: M I D G U T LECTIN ACTIVITY W I T H
L I P O P O L Y S A C C H A R I D E B I N D I N G SPECIFICITY.
Lectin that agglutinates human and animal red blood cells (RBCs) was demonstrated in
midgut extracts of Glossina tachinoides. The highest haemagglutination titres were against
pig and rabbit RBCs. Treatment of rabbit R B C s with pronase, trypsin, neuraminidase,
bromelain, glutaraldehyde and periodate reduced the agglutination titres. The lectin is specific
for amino, methyl and deoxy derivates of glucose, amino and methyl derivates of mannose,
D-galactosamine, N-acetylneuraminic acid and trehalose. In addition, very high reactivity
against the lipopolysaccharide of E. coli К 235 was found. Lectin is secreted to the midgut
lumen. It consists of a 27 kilodalton protein component that is not glycosylated. Sandwich
ELI S A permits quantification of lectin in tissue samples.

1. IN TR O D U C TIO N

In insects, lectins take part in the humoral and cellular immune mechanisms.
In vectors o f infectious diseases, they may also react with the glycosylated com p o­
nents o f transmitted pathogens [1]. Several species o f Glossina possess haemolymph
and gut lectins that are important for the life-cycle o f transmitted trypanosomes, e.g.

* This research was carried out with the support of the I A E A under Research Contract
No. 6804/RB.

557
558 VOLF et al.

in Glossina morsitans morsitans, the lectins limit the susceptibility to trypanosomes

and serve as signalling factors for maturation o f Trypanosoma congolaise [2].
Differences between species (strains) o f flies in lectin secretion might be responsible
for the species specific (or strain specific) barriers which exist between trypano­
somes and vectors [3].
This paper is aimed mainly at identification o f the binding specificity o f the
midgut lectin o f G. tachinoides and its structural characterization.

2. M A TE R IA L S A N D M ETH ODS

Adult tsetse flies, G. tachinoides, obtained as puparia from the A gen cy’ s
Laboratory at Seibersdorf, were maintained on a mixture o f cattle/pig blood.
The midgut was removed from teneral and non-teneral flies o f both sexes,
repeatedly homogenized, frozen and thawed three times at - 7 0 ° C . The gut extracts
were then centrifuged at 3000g for 10 min and the supernatant stored at —7 0 °C .
In a haemagglutination activity (H A ) test, four types o f human ( A l , A 2, В
and O) and seven types o f animal red blood cells (RBCs) (rabbit, sheep, cattle, pig,
goat, mice and chicken) were used. Rabbit RBCs were used native or treated with
pronase, trypsin, bromelain, periodate and glutaraldehyde. An extract of
G. tachinoides midgut was serially diluted in haemagglutination buffer T N (10 mM
T ris-H C l, 0.15M NaCl, pH 7.2) on microtitration plates. The same volume (50 /¿L)
o f erythrocyte suspension was added to each well. The reciprocal value o f the highest
dilution with a positive reaction was designated a titre.
In the haemagglutination inhibition test (H IT), the inhibitors (saccharides and
glycoconjugates) were serially diluted in TN buffer. The midgut extract was adjusted
to contain 1.5 units o f haemagglutination activity (H AU ) and the same volume
(50 p,L) o f 1 % suspension o f rabbit or pig erythrocytes was added to each well. The
lowest concentration o f an inhibitor blocking 1.5 H AU was designated a 50% inhibi­
tion concentration.
Antibodies directed against the H A o f G. tachinoides midgut were raised in
rabbits and m ice according to R ef. [4]. A 1% suspension o f RBCs was agglutinated
by midgut extract, washed three times in TN buffer and injected subcutaneously. The
gammaglobulin fraction was isolated from the rabbit hyperimmune serum using
caprylic acid according to R ef. [5].
Electrophoresis in p'olyacrylaminè gel in the presence o f sodium dodecyl sul­
phate (S D S-P A G E ) was carried out in a discontinuous system according to R ef. [6]
on a 10-15% linear gradient slab gel (thicknéss 1.5 mm) under reduced conditions.
The gel, with markers o f molecular masses (L M W kit, Pharmacia, Uppsala,
Sweden), was stained with Coomassie Brilliant Blue R-250.
Transfer o f proteins to the NC membrane was carried out according to
R ef. [7]. The NC membrane was rinsed in T-PBS (PBS, pH 7.4, 0.05% Tween 20).
 IAEA-SM-327/53 559

Immunoblats were incubated for 2 h in 5 % skim milk (O xoid, London, United King­
dom ), washed in T-PBS and incubated for 2 h with rabbit antiserum to
G. tachinoides H A and then for 1 h with horseradish peroxidase-cOnjugated swine
anti-rabbit immunoglobulins (S E V A C , Prague, Czechoslovakia).
In affinoblotting, NC strips blocked with 3% BSA (Serva, Heidelberg, Ger­
many) were incubated with peroxidase labelled lectins (diluted 1:100) for 2 h. The
lectins from the jack bean (C on A ), garden pea (P S A ), peanut (P N A ), soybean
(SBA) and wheat germ (W G A ) obtained from Lectinola (Prague) were labelled with
horseradish peroxidase (Serva, Heidelberg), as described in Ref. [8].
The peroxidase reaction product was developed by substrate solution with
0.6m M 3,3'-diam inobenzidine.
For indirect im munofluorescence assay, the midgut tissue was fixed in
3% paraformaldehyde at 4 °C and embedded in JB-4 resin (Polysciences) in ВЕЕМ
capsules. Sections (3 p m thick) were incubated with 3% BSA in PBS overnight at
4 °C , then with rabbit antibodies directed against G. tachinoides H A diluted'in 3%
BSA for 30 min and with fluorescein-conjugated swine anti-rabbit immunoglobulins
(S E V A C , Prague) for 30 min at 3 7 °C . Sections post-stained with Evans blue were
photographed using a Jenalumar (Zeiss, Jena, Germany) fluorescent m icroscope.
Sandwich ELISA was developed for the lectin measurement. Polystyrene
microtitration plates (G A M A , Ceské Budëjovice, Czechoslovakia) were incubated
overnight with 60 /xL o f rabbit gammaglobulins directed against G. tachinoides HA
(20 /ug/mL) in 0.1M carbonate buffer (pH 9:6). Midgut extracts (50 juL) were
serially diluted in PBS-Tw and incubated for 2 h at 3 7 °C . Then the plates were
incubated for 90 min with m ice antiserum directed against G. tachinoides HA
(diluted 1:1000) at 3 7°C and for 60 min with peroxidase-labelled swine anti-mice
immunoglobulins (S E V A C , Prague). Orthophenylendiamine in phosphate-citrate
buffer (pH 5.5) served as a substrate and the plates were read at 492 nm.

3. RESULTS

Extract o f teneral and non-teneral G. tachinoides midgut agglutinated rabbit,

m ice, pig and human A1 RBCs. The haemagglutination titres ranged from 1:10 to
1:10.240. The highest midgut extract titres were detected against rabbit and pig
RBCs. Treatment o f rabbit RBCs with pronase, neuraminidase, bromelain,
glutaraldehyde and periodate reduced the agglutination titres.
The binding specificity o f G. tachinoides H A was studied using the haemag­
glutination inhibition test with native rabbit and pig RBCs (Table I). Eight carbo­
hydrates caused the inhibition o f agglutination at concentrations o f 250m M ór below .
These sugars comprised D-galactosamine, D-glucosamine, D-mannosamine,
2-deoxyglucose, methyl-a-D-mannoside, m ethyl-a-D-glucoside, N-acetylneuraminic
acid and trehalose. O f the glycoconjugates used, the most effective inhibitors were
560 VOLF et al.

TA B L E I. INHIBITION OF THE M ID G U T H A EM A G G LU TIN A TIO N

A C T IV IT Y (H A ) OF T E N E R A L FLIES (TF) A N D N O N -TE N E RA L FLIES
(N T F ), G. tachitïoides, USING RABBIT A N D PIG ERY TH RO CY TES

Concentration required for Inhibition of H A

Inhibitor 50 rabbit erythrocytes pig erythrocytes

(TF) (NTF) (NTF)

Carbohydrates (mM)

D-galactosamine 62 250 250

D-glucosamine 62 250 У
D-mannosamine 62 125 y
Trehalose 125 125 y
2-deoxy-D-glucose У У 125
Methyl-a-D-glucopyranoside У У 250
Methyl-a-D-mannopyranoside У y 125
N-acetylneuraminic acid x x 25

Glycoconjugates (/xg/mL)

Fetuin 39 156 10
Fetuin desialized 156 39 5
Bovine submaxillary mucin 156 78 39
Bovine submaxillary mucin
desialized z z 20
Ovomucoid z z 312
Ovalbumin z z 156
Gelatine z 30 10
Heparin z z 1
LPS E. coli K235 2 10 5
LPS S. typhosa 625 312 z

Note: LPS: lipopolysaccharide; x = > lOOmM; y = > 2 5 0 m M ; z = > 62 5 (¿g/mL.

Without inhibition effect in HIT using both rabbit and pig RBCs: D-glucose,
D-galactose, D-mannose, L-fucose, D-arabinose, D-tagatose, N-acetyl-D-glucosamine,
N-acetyl-D-galactosamine, N-acetyl-D-mannosamine, lactose, maltose, fructose,
cellobiose, mellibiose, stachyose, dulcit, inositol, hyaluronic acid, laminarin and
peroxidase.
 IAEA-SM-327/53 561

А В C D

FIG. 1. SDS-PAGE analysis o f G. tachinoides midgut extracts from unfed flies (A) and from
flies four days after feeding (B). Immuno- and affinoblotting analyses o f the same extracts
(C, D) with: (a) rabbit antiserum directed against the haemagglutination activity of
G. tachinoides midgut; (b) control rabbit serum; (c) lectin Con A ; and (d) lectin PSA.
562 VOLF et al.

FIG. 2. Indirect immunofluorescence assay of G. tachinoides midgut with rabbit antibodies

directed against the G. tachinoides midgut lectin activity.

lipopolysaccharide (LPS) from E. coli К 235 (with RBCs o f both species) and
glycoprotein heparin (with pig RBCs only). On the other hand, only a weak inhibi­
tion occurred with Salmonella typhosa LPS (Table I). Using rabbit RBCs, no
qualitative differences were found between the binding specificity o f midgut extracts
from teneral and non-teneral flies o f both sexes (Table I).
S D S -P A G E and blotting techniques were used for the identification and
characterization o f lectin components. Only one polypeptide subunit o f 27 kilodalton
was detected using immunoblotting with rabbit antibodies directed against the
haemagglutination activity o f G. tachinoides midgut (Fig. 1). The 27 kilodalton
protein component did not react with the plant lectins in affinoblotting.
 IAEA-SM-327/53 563

Protein concentration (/xg/mL)

FIG. 3. Sandwich ELISA with G. tachinoides midgut extract. Standard curve. The optical
density (OD 492) reading was plotted against the protein concentration in the sample.

In an indirect immunofluorescence assay, rabbit serum directed against H A o f

the G. tachinoides midgut extract reacted with the inner surface o f the midgut
epithelium and with the peritrophic membrane (Fig. 2). Control negative rabbit sera
did not react with tsetse midgut tissue.
In sandwich ELISA, the optical density o f the samples was dependent (non-
linearly) on the protein concentration (Fig. 3).

4. DISCUSSION

In the present study, high haemagglutinin activity was found in G. tachinoides

midgut. Contrary to results reported for midgut lectin o f G. m. morsitans [9], lectin
activity was detected also in teneral flies, although in the low er titres.
The specificity o f the carbohydrate binding activity was very similar to that o f
G. tachinoides hindgut lectin reported by Ingram and M olyneux [10] against human
RBCs. D-glucosamine, D-mannosamine, D-galactosamine and 2-deoxyglucose were
effective inhibitors in both cases. In our study, inhibition occurred also with tre­
halose (using rabbit RBCs only), sialic acid, methyl mannoside and methyl glucoside
(using pig RBCs only). In contrast, the agglutination activity o f hindgut lectin against
human RBCs was inhibited by low concentrations o f N-acetyl-D-galactosamine and
L-fiicose [10].
564 VOLF et al.

The G. m. morsitans midgut lectin responsible for the stimulation o f trypano­

some maturation [2] was also inhibited with amino, N-acetyl and methyl derivates
o f glucose and mannose [10]. This suggests that G. tachinoides midgut lectin could
play a similar role in the life-cycle o f the transmitted trypanosomes.
The LPS o f E. coli К 235 was the most effective inhibitor o f G. tachinoides
midgut lectin. A s LPS is a typical surface component o f the cell wall o f gramnegativé
bacteria, the strong specificity o f the Glossina lectin suggests its possible role i n .
defence against microbial infections. The lectin is secreted to the midgut lumen,
possibly in order to combat the microorganisms contaminating the ingested blood.
Great differences in the inhibitory effects o f LPS from two bacteria species may
reflect the differences in LPS composition known to occur within gramnegative bac­
teria [ 1 1 ].
The LPS binding proteins have also been found in other insects, namely in the
haemolymph o f Periplaneta americana [12] and the midgut o f Triatoma infestans
[13].
The G. tachinoides lectin consists o f a 27 kilodalton protein component(s) that
is(are) not glycosylated. T o the best o f our knowledge, the structural characteriza­
tions o f tsetse lectin have previously been reported only by Ingram and M olyneux
[14] and Stiles et al. [15]. Lectin o f G. fuscipes fuscipes haemolymph with a native
molecule o f 710 kilodalton comprised approximately ten non-covalently linked
70 kilodalton subunits; G. palpalis midgut agglutinin has a subunit o f 67 kilodalton.
The preliminary results obtained with sandwich ELISA suggest that the method
is a promising tool for quantifying tsetse lectins. W e are now working on other assay
modifications which should im prove its sensibility and permit evaluation o f antigenic
similarities o f lectins in different populations or species o f tsetse flies.

REFERENCES

[1] I N G R A M , G.A., M O L Y N E U X , D.H., Insect lectins: Role in parasite-vector interac­

tions, Lectin Rev. 1 (1991) 103-127.
[2] W E L B U R N , S.C., M A U D L I N , I., Lectin signalling of maturation of Trypanosoma
congolense infections in tsetse, Med. Vet. Entomol. 3 (1989) 141-145.
[3] W E L B U R N , S.C., M A U D L I N , I., ELLIS, D.S., Rate of trypanosome killing by lec­
tins in midgut of different species and strains of Glossina, Med. Vet. Entomol. 3 (1989)
77-82.
[4] Y E A T O N , R.W., “Occurrence of nonlymphatic hemagglutinins in arthropods and
their possible functions”, Hemocytic and Humoral Immunity in Arthropods
(GUPTA, À.P., Ed.), Wiley, N e w York (1986) 505-516.
[5] R U S S O , C., C A L L A G A R O , L., L A N Z A , E., F E R R O N E , S., Purification of IgG
monoclonal antibody by caprylic acid purification, J. Immunol. Methods 65 (1983)
216-271.
[6] L A E M M L I , U:K., Cleavage of structural proteins during the assembly of the head of
bacteriophage T4, Nature (London) 227 (1970) 680-685.
 IAEA-SM-327/53 565

[7] T O W B I N , H., S T E A H L I N , T., G O R D O N , J., Electrophoretic transfer of proteins

from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications,
Proc. Natl. Acad. Sci. U S A 76 (1979) 4350-4354.
[8] G R U B H O F F E R , L., G U I R A K H O O , F., HEINZ, F.X., K U N Z , C., “Interaction of
tick-borne encephalitis virus protein E with labelled lectins”, Lectins: Biology,
Biochemistry, Clin. Biochem. 7 (1990) 313-319.
[9] M A U D L I N , I., W E L B U R N , S.C.,Lectin mediated establishment of midgut infections
of Trypanosoma congolense and Trypanosoma brucei in Glossina morsitans, Trop.
Med. Parasitol. 38 (1987) 167-170.
[10] I N G R A M , G.A., M O L Y N E U X , D.H., Sugar specificities of anti-human ABO( H )
blood group erythrocyte agglutinins (lectins) and haemolytic activity in the haemolymph
and gut extracts of three Glossina species, Insect Biochem. 18 (1988) 269-279.
[11] M O R R I S O N , D.C., R A Z I U D D I N , S., “Lipopolysaccharides and endotoxin”,
Immunopharmacology (SIROIS, P., R O L A - P L E S Z C Y N S K I , М., Eds), Elsevier,
Amsterdam (1982) 169-197.
[12] J O M O R I , T., K U B O , T., N A T O R I , S., Purification and characterization of
lipopolysaccharide-binding protein from haemolymph of the American cockroach
Periplaneta americana, Eur. J. Biochem. 190 (1990) 201-206.
[13] G R U B H O F F E R , L., H Y P S A , V., Institute of Parasitology, Ceské Budéjovice, private
communication.
[14] I N G R A M , G.A., M O L Y N E U X , D.H., Lectins (haemagglutinins) in the haemolymph
of Glossina fuscipes fuscipes: Isolation, partial characterization, selected physico­
chemical properties and carbohydrate-binding specificities, Insect Biochem. 20 (1990)
13-27.
[15] STILES, J.K., et al., Identification of midgut trypanolysin and trypanoagglutinin in
Glossina palpalis ssp. (Diptera:Glossinidae), Parasitology 101 (1990) 369-376.
 IAEA-SM-327/54

INFLUENCE COMBINEE DE LA BASSE TEMPERATURE

ET DE L’IRRADIATION DURANT
LA DERNIERE PHASE PUPALE SUR
LA CAPACITE VECTORIELLE DES MALES DE
Glossina palpalis gambiensis

A . DJITEYE
Laboratoire central vétérinaire,
Bamako, Mali

B. B AU ER
Centre de recherches sur les trypanosomoses animales,
B obo-D ioulasso, Burkina Faso

A . V A N D E R V L O E D T f , U , F E L D M A N N , M .J.B . V RE YSE N
D ivision mixte F A O /A IE A ,
A gence internationale de l ’énergie atomique,
Vienne

Abstract-Résumé

COMBINED INFLUENCE OF LOW TEMPERATURE AND IRRADIATION DURING

THE LAST PUPAL PHASE ON THE VECTORIAL CAPACITY OF Glossina palpalis
gambiensis MALES.
Glossina palpalis gambiensis pupae 25, 28 or 30 days old irradiated under ambient con­
ditions with doses o f 60 Gy, 80 Gy or 100 Gy (1 Gy = 100 rad) were exposed (or not
exposed) for five days at 15 °C (or incubated) before or after irradiation. The adult flies,
whether in the teneral state or not, were fed for three successive days on animals infected with
Trypanosoma vivax, T. congolense or T: b. brucei at the peak o f parasitemia. As o f the tenth
day (post-infection), the rates o f infection with T. vivax for flies in the teneral and the non-
teneral state were, respectively: 98.9% and 97.9%, for the control group and 96.8% and
95.7% for those incubated and irradiated with 100 Gy on day 30. As o f the twentieth day,
the percentages o f infection with T. congolense were, respectively, for the different states
(teneral and non-teneral): 77.5% and 34.7% for the controls, 71.3% and 54.2% for the
incubated group, 83.9% and 50.0% for those irradiated with 80 Gy on day 30, and 58.2%
and 64.8% for those irradiated with 100 Gy on day 30. The rates o f infection o f the flies with
T. b. brucei as o f the twenty-fifth day were, respectively, for flies infected in the teneral state
and in the non-teneral state: 91.1% and 54.7% for the controls, 77.0% and 44.9% for the
incubated group, 51.3% and 35.0% for those irradiated with 80 Gy on day 30 and 73.1% and
69:5% for those irradiated with 80 Gy on day 28 and incubated. The percentages o f sterility

567
568 DJITEYE et al.

in males irradiated with various doses are: 7 1 % (60 G y on day 25), 7 5 % (incubated and
60 G y on day 25), 8 2 % (incubated and 60 G y on day 28), 100% (80 G y on day 28) and 100%
(100 G y on day 30).

I NFLUENCE C O M B I N E E D E L À BASSE T E M P E R A T U R E ET D E L ’ IRRADIATION

D U R A N T L A DERNIERE PHASE PUPALE S UR LA CAPACITE VECTORIELLE DES
M A L E S D E Glossina palpalis gambiensis.
Des pupes, âgées de 25, 28 ou 30 jours, de Glossina palpalis gambiensis irradiées dans
les conditions ambiantes à la dose de 60 Gy, 80 G y ou 100 G y (1 G y = 100 rad), ont été
exposées (ou non) durant 5 jours à 15 °C (ou incubées) avant ou après irradiation. Les mouches
adultes sont alimentées, à l’ état ténéral ou à l’
état non ténéral, pendant trois jours successifs
sur des animaux infectés par Trypanosoma vivax, T. congolense ou T. b. brucei au pic de la
parasitémie. A partir du 10e jour (post-infection), les taux d’ infection des mouches par
T. vivax à l’état ténéral ou à l’état non ténéral sont respectivement: 98,9% et 97,9% pour le
lot témoin, 96,8% et 95,7% pour les incubées et irradiées à 100 G y au jour 30. A partir du
20e jour, les pourcentages d ’ infection par T. congolense sont respectivement pour les
différents états (ténéral et non ténéral): 77,5 et 34,7 pour les témoins; 71,3 et 54,2 pour les
incubées; 83,9 et 50,0 pour les irradiées à 80 G y au jour 30; 58,2 et 64,8 pour les irradiées
à 100 G y au jour 30. Les taux d ’ infection des glossines par T. b. brucei à partir du 25e jour
sont respectivement pour les mouches infectées à l’ état ténéral et à l’
état non ténéral: 91,1%
et 54,7% pour les témoins; 77,0% et 44,9% pour les incubées; 51,3% et 35,0% pour les
irradiées à 80 G y au jour 30; 73,1% et 69,5% pour les irradiées à 80 G y au jour 28 et
incubées. Les pourcentages de stérilité des mâles irradiés aux différentes doses sont:
71 (60 G y au jour 25), 75 (incubées et 60 G y au jour 25), 82 (incubées et 60 G y au jour 28),
100 (80 G y au jour 28), 100 (100 G y au jour 30).

1. IN TROD U CTION

Le Laboratoire central vétérinaire de Bamako envisage d ’utiliser la technique

de l ’ insecte stérile, complétée par des traitements insecticides pour lutter contre la
mouche tsé-tsé dans la zone agro-pastorale de Tienfala-Baguineda, et de déterminer
à quelles conditions son éradication serait possible. L ’ entreprise est réalisable car il
n’ existe qu’une seule espèce de glossine (Glossina palpalis gambiensis) et la zone
à traiter est isolée par des barrières naturelles. Il est programmé de lâcher des mâles
stériles nourris et marqués sur place, issus de pupes irradiées en provenance du
Centre de recherches sur les trypanosomoses animales (C R T A ) de Bobo-Dioulasso.
Après des études préparatoires à l ’ Unité d ’ entomologie de la Division mixte
F A O /A IE A (Laboratoire de Seibersdorf), des séries d ’ expériences ont été realisées
au C R T A . L ’ objectif de ces expériences a été de déterminer une dose optimale
d ’ irradiation des pupes et de connaître l ’ impact de la conservation à basse tempéra­
ture ainsi que celui de l ’ état de réplétion des adultes avant l ’ infection sur le risque
de la transmission des différentes espèces de trypanosomes pathogènes.
 IAEA-SM-327/54 569

2. MATERIELS ET METHODES

Des pupes de Glossina palpalis gambiensis âgées de 25, 28 ou 30 jours ont été
irradiées à l ’ aide d ’un irradiateur gamma ( 137Cs) dans les conditions ambiantes aux
doses de 60 G y, 80 G y ou 100 Gy (1 Gy = 100 rad). Elles ont été également
exposées (ou non) durant 5 jours dans un incubateur du type LKB M ini Coldlab réglé
à 15 + 1 ,5 °C et 90% d ’ humidité relative, immédiatement avant ou après
l ’ irradiation.
Après éclosion, les mouches mâles sont groupées dans des cages du type
Roubeau à raison de 2 5 -3 0 individus par cage et alimentées, durant 5 min, pendant
trois jours successifs sur des animaux infectés: une chèvre pour T. vivax et des lapins
pour T. congolense et T. brucei, tous au pic de la parasitémie. Dans chaque lot, deux
catégories de mouches, dont le nombre varie de 100 à 120, sont infectées à différents
états de réplétion: à l ’ état ténéral avec trois repas infectants aux jours 2 , 3 et 4 post­
émergence ou à l ’ état non ténéral avec deux repas ordinaires in vitro sur membrane
de silicone aux jours 2 et 3 post-ém ergence, suivis de trois repas infectants aux
jours 4, 5 et 6.
Dans le cas spécifique de T. vivax, deux lapins ont été infectés sans succès
avec la souche Zaria 81/486/699. Le premier a éliminé l ’ infection (2 trypanosomes/
champ) avant l ’émergence des mouches, et les quelque 1000 mouches nourries sur
le second, avec un p ic de parasitémie de 8 trypanosomes/champ, n’ ont pas été
infectées. Une chèvre sahélienne a été directement inoculée par voie intraveineuse
par une seconde souche (Banan 8 5 /C R T A /7 8 , précédant une administration de
Dexamethason.
La parasitémie la plus élevée a été de 80 trypanosomes/champ, mais la chèvre
est morte avant la fin de l ’ expérience (20e jou r post-infection, 40 trypanosomes/
champ, hématocrite = 17% ). C ’est pour cette raison que certains lots ont reçu 1 ou
2 repas infectants et que d ’ autres n’ en ont pas reçu du tout. Après les trois repas
infectants, toutes les mouches sont alimentées tous les jou rs, durant 5 min, sur
membrane à silicone avec du sang de bœ uf hépariné, irradié et enrichi de glucose
et d ’ adénosine triphosphate, et gardées dans une salle à la température de 2 4 -2 6 °C
avec 7 5 -8 5 % d ’ humidité relative.
La mortalité est vérifiée avant les repas infectants et suivie tous les 5 jours.
Les mouches infectées par T. vivax sont disséquées et examinées à l ’ aide d ’ une loupe
binoculaire et d ’ un m icroscope, à partir du 10e jou r; celles qui sont infectées par
T. congolense et T. brucei sont examinées respectivement à partir des jours 20 et
25 après le premier repas infectant.
L ’étude de l ’ effet des différents traitements sur la fertilité a concerné 16 lots
de mouches, constitué chacun de 30 femelles âgées de 2 jours, accouplées durant
3 jours avec 30 mâles âgés de 7 jours. Après séparation, la production des pupes et
les avortements sont suivis chez les femelles jusqu’au 3 5 e jour.
570 DJITEYE et al.

T A B L E A U I. DONNEES SUR L ’ EM ERGENCE DE Glossina palpalis

gambiensis APRES L ’ A C T IO N COM BINEE DE L ’IR R A D IA T IO N ET DE
L ’ EXPOSITION DES PUPES A BASSE TEM PE RA TU RE (m ars-mai 1990,
nombre total = 36 285 pupes)

Emergence

Séries Traitement Pupes Total Mâles Durée pupilison (j)

(Nb) (%) (%) Femelles Mâles

A Témoins 3444 95,6 : 53,4 28,8 31,2

B 15°C J20-25 3701 94,2 50,0 32.7 34,6

35.8 37,8

C 60 G y J25 3189 91,2 47,5 28,2 30,1

31,2 33,0

D 60 G y J28 2969 92,6 50,8 • 27,8 29,9

29,0 31,3

E 80 G y J28 3146 91,9 48,6 27,8 29,9

31,1 33,2

F 80 G y J30 3012 91,2 49,0 * 29,8

31,3

G 100 G y J30 2216 92,6 49,6 * 29,1*

H 15°C J20-25 2596 83,2 47,9 32,8 34,5

/60 G y J25 34,1 36,2

I 15°C J20-25 2651 87,5 49,6 32,5 34,3

/60 G y J28 34,0 36,2

J 15°C J20-25 2504 92,4 52,2 32.8 34,5

/80 G y J28 33.9 36,2

К 80 G y J28 2646 82,9 50,3 29,0 34,4

/15°C J28-33 32,6 35,8

L 80 G y J30 2691 84,0 45,1 * 30.8

/15°C J30-35 33.9

M 100 G y J30 1920 80,5 48,0 * 30,8

/15°C J30-35

*: Une partie importante des mouches a émergé avant le second traitement.

J: Jours post-émergence. Basse température: 15°C.
Irradiation: 60, 80 ou 100 G y (1 G y = 100 rad).
Note: Les premiers nombres figurant dans les colonnes 6 et 7 indiquent le nombre de jours
pour les pupes pondues en mars 1990, les seconds se réfèrent au mois de mai.
 IAEA-SM-327/54 571

Pour les tests de compétitivité, des mâles traités, incubés ou irradiés ont été
marqués et mis dans une cage d ’ éclosion avec des témoins (25:25, à 4 reprises) en
présence du même nombre de femelles vierges. Les couples form és ont été
immédiatement prélevés à l ’ aide d ’ un tube à essais, puis comptabilisés.

3. RESU LTATS

3 .1 . E m ergence des glossines (Tableau I)

Le taux m oyen d ’émergence des glossines est de 9 0 ,7 6 % . Les pourcentages

observés dans les différents lots sont: 95,6 pour les témoins, 94,2 pour les incubées,
91.8 pour les irradiées, 92,4 pour les incubées et irradiées et 82,6 pour les irradiées
et incubées. L ’ incubation ou exposition à 15°C durant 5 jours augmente la durée de
la vie pupale d ’environ 4 jou rs; les moyennes enregistrées au m ois d ’ avril 1990 sont
respectivement pour les femelles et les mâles: 28,8 et 31,2 jours pour les témoins,
contre 33,3 et 35,6 jours pour les incubées. Des variations ont été constatées suivant
les mois de ponte, en effet chez les incubées, la vie pupale a une durée moyenne de
35.8 et 37,8 jours respectivement pour les femelles et les mâles issus de pupes
pondues au mois de mars, contre 32,7 et 3 4,6 jours au mois de mai. D ’une manière
générale, les femelles émergent 2 jours avant les mâles, et l ’ irradiation provoque une
émergence massive chez les pupes âgées.

3 .2 . Infections par les espèces de trypanosom es

3 .2 .1 . Trypanosoma vivax (Banan 8 5 /C R T A /7 8 , Tableau П)

Les infections par cette espèce se limitent à la trompe des glossines: au labre
et à l ’ hypopharynx. En général, les mâles nourris à l ’ état ténéral sur l ’ animal infecté
par T. vivax n ’ont pas un taux d ’ infection différent d ’une manière significative de
ceux qui ont pris leurs repas infectants à l ’ état non ténéral.
Les taux observés dans le lot témoin sont respectivement 9 8 ,8 % et 9 7 ,9 % ,
contre 96,7 % et 95 ,6% dans le lot des pupes exposées à 15°C du jou r 20 au jou r 25
et irradiées à 80 Gy au 2 8 e jou r. Cependant, les taux d ’ infection suivants 67,3%
(ténéral) et 71,2 % (non ténéral) observés dans le lot des pupes irradiées à 80 Gy au
jou r 28 et incubées sont inférieurs d ’ une manière significative aux taux des lots
précédents. L ’ action de maintenir les pupes à basse température (à 15°C durant
5 jours) après irradiation semble diminuer le taux d ’ infection à T. vivax par rapport
au témoin et au lot des pupes incubées et irradiées. Nous avons également constaté
que le taux d ’ infection à T.vivax et la charge parasitaire augmentent avec le nombre
de repas infectants, et que les mouches peuvent perdre l ’ infection avec le temps.
572 DJ1TEYE et al.

T A B L E A U П. T A U X D ’INFECTION D E Glossina palpalis gambiensis

P A R Trypanosoma vivax (Banan 85/C R T A /78)

Survie Infection au
Repas
Effectifs au Dissections labre et à
Traitement Etat infectants
(Nb) jour 10 (Nb) l’hypopharynx
(Nb)
(Nb) (Nb) (% )

Témoins Ténéral 99 3 89 88 87 98,9

Non ténéral 100 . 3 98 97 95 97,9

15°C J20-25 Ténéral _ ._ _ _ _

Non ténéral 100 . 1,-2 9.1 90 79 87,8

60 Gy J25 Ténéral 100 1-2 94 93 80 86,0

Non ténéral 95 1-2 91 . 90 82 91,1

60 Gy J28 Ténéral _ _ _ _
Non ténéral 100 1 90 90 77 85,5

80 Gy J28 Ténéral 99 1-3 95 86 66 76,7

Non ténéral 100 1-2 93 86 79 91,8

80 Gy J30 Ténéral _ _ _ ■
_
Non ténéral 60 1 51 51 39 76,5

100 Gy J30 Ténéral _ — _ _

Non ténéral 100 1 83 80 57 71,2

15°C J20-25 . Ténéral 99 3 87 82 71 86,6

/6 0 Gy J25 Non ténéral 100 3 87 86 59 68,6

15°C J20-25 Ténéral 98 3 .8 9 85 82 96,5

/6 0 Gy J28 Non ténéral 100 3 96 92 90 97,8

15 °C J20-25 Ténéral 100 3 94 93 90 96,8

/8 0 Gy J28 Non ténéral 104 3 94 93 89 95,7

80 Gy J28 Ténéral 100 3 92 92 62 67,4

/1 5 °C J28-33 Non ténéral 100 3 82 80 57 71,2

— : Les observations n’ont pas été faites, la chèvre étant morte avant la fin de l ’expérience.
J: Jours post-émergence. Basse température: 15°C.
Irradiation: 60, 80 ou 100 Gy (l Gy = 100 rad).
 IAEA-SM-327/54 573

3.2.2. Trypanosoma congolense (Karankasso 83/C R T A /5 7 , Tableau Ш )

L e développement de cette espèce com m ence au niveau de l ’ instestin moyen

de la mouche avant d ’envahir la trompe (le labre et l ’ hypopharynx). En général, les
mâles qui ont pris leurs premiers repas infectants à l ’ état ténéral ont un taux d ’ infec­
tion supérieur à celui des mâles infectés à l ’ état non ténéral. Les pourcentages
observés dans les différents lots sont respectivement: 77,5 et 34,6 pour les témoins,
7 1,2 et 54,2 pour les incubées, 59,7 et 23,5 (80 G y au jou r 28), 83,8 et 5 0,0 (80 Gy
au jou r 30). En revanche, les 58,2% et 64 ,8 % trouvés chez les pupes irradiées à
100 G y au jou r 30 ne diffèrent pas d ’ une manière significative (écart réduit = 0,76).
La majeure partie (plus de 8 0% ) des infections à T. congolense restent
bloquées au niveau l ’ intestin m oyen, n ’ atteignant donc pas la trompe. Une
étude comparative a révélé que l ’ espèce de glossine riveraine (G. p. gambiensis:
89,74 % , 70/78) s’ infecte autant et même un peu plus que l ’ espèce de savane
(G. morsitans submorsitans: 7 9 ,4 8 % , 62/7 8 ); la différence n ’ est pas significative,
écart réduit = 1,77. Cependant, G. p. gambiensis est un mauvais vecteur de
T. congolense, car seules 27,77% et 7 ,1 4 % des infections initiales.de l ’ intestin
atteignent respectivement le labre et l ’ hypopharÿnx, contre 91,93% et 82,25% chez
G. m. submorsitans (écart réduit = 8,40 pour le labre).

3.2.3. Trypanosoma brucei brucei (F a ra k ob a /8 0 /C R T A /l, Tableau IV)

Les trypanosomes de cette espèce se multiplient d ’ abord dans l ’ intestin moyen

de la glossine, avant de migrer vers les glandes salivaires et l ’hypopharynx. A l ’ état
ténéral les mâles de G. p. gambiensis contractent plus aisément les infections à
T. brucei qu’à l ’état non ténéral. Les taux d ’ infection observés dans les différents
lots sont respectivement: 9 1 ,1 % et 54,6% pour les témoins, 77 ,0 % et 4 4 ,8 % pour
les incubées, 74 ,2% et 2 4 ,2 % (60 G y au jou r 25), 6 3 ,7 % et 50 ,0 % (incubées et
60 Gy au jou r 25). En revanche, chez les mouches irradiées à 100 G y au jo u r 30,
celles qui sont infectées à l ’ état ténéral ont un taux d ’ infection de 3 4 ,1 % qui ne
diffère pas d ’une manière significative de celui des mouches infectées à l ’ état non
ténéral (3 5 ,7 % ); écart réduit = 0,19.
Sur 799 infections de l ’ intestin m oyen dues à T. brucei, 284 (3 5 ,5 % ) et 62
(7 ,7 % ) seulement ont évolué respectivement au niveau des glandes salivaires et de
l ’ hypopharynx. Les taux de survie au 2 5 e jou r des mâles infectés à l ’ état ténéral
sont de 85,8% pour les incubées, de 59,0% pour les irradiées et de 77 ,5 % pour les
incubées et irradiées.
574 DJITEYE et al.

T A B L E A U Ш . T A U X D ’ INFECTION DE Glossina palpalis gambiensis

P A R Trypanosoma congolense (Karankasso 83/C R T A /57)

Taux d!in fection

Effectif Survie J20 Dissections Intestin Labre
Traitement Etat
(Nb) (Nb) (% ) (Nb) moyen
(% ) (% )

Témoins Ténéral 100 97 97,0 80 77,5 1,2

Non ténéral 100 83 83,0 75 34,7 1,3

15°C Ténéral 99 91 91,9 87 71,3 2,3

J20-25 Non ténéral 99 91 91,9 83 54,2 3,6

60 Gy Ténéral 125 79 63,2 72 66,7 2,8

J25 Non ténéral 121 90 74,4 73 37,0 4,1

60 Gy Ténéral 120 6 2 ". 51,7 ' 61 75,4 3,3

J28 Non ténéral 120 77 64,1 62 66,1 16,1

80 Gy Ténéral 100 92 92,0 72 59,7 1,4

J28 Non ténéral 163 116 71,2 85 23,5 3,5

80 Gy Ténéral 110 90 81,8 62 83,9 14,5

J30 Non ténéral 100 52 52,0 52 50,0 7,7

100 Gy Ténéral 120 101 84,2 79 58,2 10,1

J30 Non ténéral 110 72 65,4 54 64,8 18,5

Basse température: 15°C ; J: jours post-émergence.

Irradiation: 60, 80 ou 100 Gy (1 Gy = 100 rad).
 IAEA-SM-327/54 575

T A B L E A U IV . T A U X D ’IN FECTION D E Glossina palpalis gambiensis

P A R Trypanosoma brucei brucei (Farakoba 83 /C R T A /5 7 )

Taux d’ infection
Effectif Survie J25 Dissections Intestin Glandes
Traitement Etat
(Nb) (Nb) (% ) (Nb) moyen salivaires
(% ) (% )

Témoins Ténéral 99 85 85,9 79 91,1 25,3

Non ténéral 102 83 81,4 75 54,7 10,7

15 °C J20-25 Ténéral 99 83 83,8 74 77 ,0 12,2

Non ténéral 100 81 81,0 78 44,9 6,4

60 Gy J25 Ténéral 117 75 64,1 66 74,2 19,7

Non ténéral 117 76 64,9 70 24,3 1,4

60 Gy J28 Ténéral 120 52 43,3 47 57,4 27,6

Non ténéral 120 50 41,7 50 38,0 28 ,0

80 Gy J28 Ténéral 116 63 54,3 50 66,0 18,0

Non ténéral 139 58 41,7 52 34,6 7,7

80 Gy J30 Ténéral 120 78 65,0 78 51,3 35,9

Non ténéral 105 40 38,1 40 35,0 12,5

100 Gy J30 Ténéral 120 82 68,3 82 34,1 15,8

Non ténéral 120 57 47,5 56 35,7 .17,8

15°C J20-25 Ténéral 120 84 70,0 69 63,8 30,4

/60 Gy J25 Non ténéral 120 61 50,8 60 50,0 16,7

15 °C J20-25 Ténéral 120 102 85,0 67 62,7 26,9

/6 0 Gy J28 Non ténéral 120 110 91,7 72 • 44,4 18,0

15°C J20-25 Ténéral 120 93 77,5 65 67,7 29,2

/8 0 Gy J28 Non ténéral 120 88 73,3 75 41,3 22,7

80 Gy J28 Ténéral 120 55 45,8 52 73,1 23,1

/1 5 °C J28-33 Non ténéral 120 62 51,7 59 69,5 23,7

80 Gy J30 Ténéral — — — —
/1 5 °C J30-35 Non ténéral 75 36 48,0 36 50 ,0 16,7

100 Gy J30 Ténéral _ — — —

/1 5 °C J30-35 Non ténéral 58 18 31,0 18 50 ,0 11,1

Basse température: 15°C . J: Jours post-émergence.

Irradiation: 60, 80 ou 100 Gy (1 Gy = 100 rad).
576 DJITEYE et al.

TA B L E A U V . FERTILITE DES FEM ELLES DE Glossina palpalis gambiensis

INSEMINEES PAR DES M A LE S IRRADIES A U STAD E PUPAL E T/O U
EXPOSES A BASSE TEM PE RA TU RE (Accouplement: 30 mâles/30 femelles)

Traitement Survie . Pupes Avortements Taux de

femelles stérilité

Femelles Mâles J25 J35 Total Par Total Par (% )

(Nb) (Nb) (Nb) femelle (Nb) femelle

Témoin Témoin 27 26 52 2,0 6 0,2 0 ,0

15 °C Témoin 29 29 66 2,3 5 0,2 -3 5 ,0
J20-25

Témoin 15 °C J20-25 24 23 42 1,8 7 0,3 9,0

15°C 15°C J20-25 29 29 65 2,2 4 0Д -1 2 ,0
J20-25

Témoin 60 Gy J25 29 29 17 0 ,6 47 1,6 71,0

Témoin 15°C J20-25 29 28 14 0,5 64 2,3 75,0
/60 Gy J25

Témoin 60 Gy J28 28 28 18 0 ,6 62 2,2 68,0

Témoin 15°C J20-25 25 25 9 0,4 63 2,5 82,0
/60 Gy J28

Témoin 80 Gy J28 22 ■ 21 0 0,0 66 3,1 100

Témoin 15°C J20-25 30 30 7 0,2 67 2,2 88,5
/80 Gy J28
Témoin 80 Gy J28 24 24 1 0 ,4 65 2,7 9 8,0
/1 5 °C J28-33

Témoin 80 Gy J30 26 26 6 0,3 64 2,5 87,0

(N = 22) /23

Témoin 80 Gy J30 19 18 2 0,1 49 2,7 94,5

15°C J30-35

Témoin 100 Gy J30 26 25 0 0 ,0 53 2,1 100

60 Gy J25 Témoin 23 23 0 0 ,0 29 2,3 100

80 Gy J28 Témoin ■ 25 24 0 0 ,0 33 1,4 100

Basse température: 15°C . J: Jours post-émergence.

Irradiation: 60, 80 ou 100 Gy (1 Gy = 100 rad). % de stérilité ( -) = plus fertile que le témoin.
26/23: 26 femelles dont 23 inséminées.
Note: Le signe - dans la neuvième colonne signifie une fertilité supérieure à celle du témoin.
 IAEA-SM-327/54 577

3.3. Effets de l’incubation et/ou de l’irradiation sur la fertilité des mouches

(Tableau V )

Les femelles de G. p. gambiensis non traitées (témoins) ou incubées,

accouplées avec des mâles témoins, ont produit en moyenne 2 à 2,27 pupes par
femelle. Les mâles incubés ont donné avec.les femelles témoins ou incubées, respec­
tivement, 1,82, et 2,24 pupes par fem elle. Les femelles inséminées par des mâles
irradiés au stade pupal ont donné un maximum de 0,64 pupes/fem elle avec la dose
de 60 Gy au 2 8 e jou r. La production des pupes a été nulle chez les femelles
accouplées avec les mâles irradiés à 80 Gy au jou r 28 ou à 100 Gy au jou r 30. Toutes
les femelles irradiées (60 Gy au jou r 25, ou 80 Gy au jou r 28) sont devenues
complètement improductives après 2 ,26 ou 1,37 avortements/femelle.
Les pourcentages de stérilité des mâles irradiés aux différentes doses sont:
71 (60 G y J25), 75 (incubés et 60 Gy J25), 68 (60 Gy J28), 82 (incubés et 60 Gy
J28) , 100 (80 Gy J28), 88,5 (incubés et 80 Gy J28), 100 (100 Gy J30).
Les tests de compétitivité réalisés entre les mâles traités et les témoins ont
révélé que les premiers sont très combatifs pour la recherche des fem elles. Les pour­
centages d ’ élection enregistrés pour les mâles traités sont: 53 incubées, 57 (60 Gy
J25), 51 (60 Gy J28), 47 (80 G y J28), 50 (80 Gy J30) et 52 (100 Gy J30).

4. CONCLUSION

Les résultats obtenus montrent que les mâles de G. p. gambiensis irradiés au

stade pupal ont une longévité leur permettant de transmettre toutes les espèces de
trypanosomes pathogènes. L ’action de nourrir les mâles avant de les lâcher réduit
considérablement les risques d ’ infections par T. congolense et T. b. brucei, et les
rend plus compétitifs, car ils vont directement à la recherche des femelles.
La technique de lâchers de mâles stériles est une méthode d ’ avenir qui est déjà
compétitive du point de vue coût grâce à la maîtrise parfaite de l ’ élevage en masse
des glossines. Il n’ y a pas de doute que la lutte génétique demeure la seule méthode
capable d ’ éradiquer une population de glossines sans porter préjudice à
l ’ environnement.
Le lâcher et la recapture des femelles irradiées sont les moyens les plus fiables
pour la détection d ’ une population résiduelle de glossines et la détermination très fine
de ses limites de répartition.
La lutte génétique est envisageable d ’ ores et déjà dans les pays ou régions qui
ne possèdent ni insectarium, ni source gamma, mais disposent d ’ un modeste
laboratoire nécessaire pour Г éclosion des pupes et la maintenance des adultes in vivo
avant les lâchers.
Le maintien des pupes à basse température durant le transport est sans
inconvénient. La voie est déjà bien ouverte pour les lâchers de mâles stériles sans
578 DJITEYE et al.

risque de transmission de la trypanosomiase. Des recherches sur le délai, la durée

et l ’ intensité de l ’ exposition à basse température pourront probablement réduire
l’ infection par T. vivax, et l ’ augmentation du nombre de repas avant le lâcher des
mâles stériles rendrait très peu probable les infections par T. congolense et T. brucei.
En effet, trois repas pris sur membrane avant un seul repas infectant a réduit le taux
d ’ infection de G. p. gambiensis par T. b. brucei à 2 6 ,6 6 % , contre 59,61% (ténéral
et un seul repas infectant) et 91,13% (ténéral et 3 repas infectants).

REMERCIEMENTS

Nous réitérons nos très sincères remerciements à M M . Hans Blix, directeur

général de l ’ A IE A , D. A . Lindquist de la D ivision mixte F A O /A IE A , W . Klassen,
ch ef de la section de la lutte contre les insectes et autres ravageurs, R. Gingrich, ch ef
de l ’ Unité d ’ entom ologie, E. O ffori et M . l ’ Annunziata de l ’ A IE A , G. Duvallet et
P. Clausen du CRT A .

BIBLIOGRAPHIE

CLAIR, M ., POLITZAR, H ., CUISANCE, D ., L A F A Y E , A ., Observations sur un essai

préliminaire de lâchers de mâles stériles de Glossina palpalis gambiensis (Haute-Volta), Rev.
Elev. Méd. Vét. Pays trop., 2 9 4 (1976) 34 1-35 1.

M A K U M Y À V IR I, A .M ., VAN DER V LO ED T, A ., Trypanosoma congolense : taux

d’ infection de Glossina palpalis palpalis élevée en laboratoire, Rev. Méd. Vét. 140 3 (1989)
2 2 1 -2 2 4 . '

M O LO O , S .K ., Cyclical transmission o f pathogenic Trypanosoma species by gamma-

irradiated sterile male Glossina morsitans morsitans, Parasitology 8 4 (1982) 28 9-296.

M O LO O , S .K ., K U T U Z A , S .B ., Vectorial capacity o f gamma-irradiated sterile male

Glossina morsitans centralis,, G. austeni and G. tachinoides for pathogenic Trypanosoma
species, Insect Sci. Applic • 5 5 (1984) 41 1-414.

T A Z E , Y ., CUISANCE, D ., POLITZAR, H ., CLAIR, М ., SELIN, E ., Essais de détermi­

nation de la dose optimale d’ irradiation des mâles de Glossina palpalis gambiensis
(Vanderplank, 1949) en vue de la lutte biologique par lâchers de mâles stériles dans la région
de Bobo-Dioulasso (Haute-Vol ta), Rev. Elev. Méd. Vét. Pays trop. 3 0 3 (1977) 2 6 9-27 9.

V A N DER VLO ED T, A ., BARNOR, H ., Effects o f ionizing radiation on tse-tse biology.

Their relevance to entomological monitoring during integrated control programmes using the
sterile insect technique, Insect Sci. Applic. 5 5 (1984) 43 1-437.
 IAEA-SM-327/55

REARING TSETSE FLIES FOR USE

IN STERILE INSECT TECHNIQUE
VECTOR CONTROL PROGRAMMES

U. F E L D M A N N *
F A Q /IA E A Entom ology Unit,
A gen cy ’ s Laboratory Seibersdorf,
International A tom ic Energy A gency,
Vienna

Abstract
REARING TSETSE FLIES FOR USE IN STERILE INSECT TECHNIQUE VECTOR
CONTROL PROGRAMMES.
In past years, considerable progress has been achieved in the mass rearing o f tsetse flies
for use in vector control programmes. This was made possible through the introduction o f the
membrane feeding'system and the development o f strict quality control procedures for differ­
ent developmental stages o f tsetse and for various aspects that are relevant to insect produc­
tion, such as blood diet quality screening. The maintenance o f large tsetse fly colonies, e.g.
more than 100 000 producing ¡females that’ may provide more than 12 000 reproductively
sterile males and equal numbers o f surplus female material, is feasible without major efforts.
Calculations o f labour requirements and the costs involved for different mass rearing systems
can now be conducted. Depending on the purpose for which the insects are reared, different
types o f mass rearing facilities may be erected (a small stationary .insect factory, large breed­
ing centre or mobile production plant). Sexually sterile tsetse from a mass rearing system may
be used for different reasons in tsetse control or eradication campaigns, namely for eradication
or control through the sterile insect technique, for ecological monitoring o f a target tsetse
population, particularly if it is under advanced control, and for transtaxon control o f closely
related species (or subspecies). Bottlenecks for a larger scale field use o f mass reared tsetse
remain for the time being: (a) the relatively long process o f strain adaptation to mass rearing
conditions, (b) laborious sexing procedures o f mature insecis upon emergence and after mating
and (c) unavailability o f simple genetic tools to collect baseline information on the suitability
o f a mass reared strain to combat an identified target population and to assess the degree of
isolation o f a target population.

Present address: Insect and Pest Control Section, Joint F A O /IA E A Division o f
Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency,
Wagramerstrasse 5, P.O . Box 100,' A -1400 Vienna, Austria.

579
580 FELDMANN

1. IN TROD UCTION

Tsetse flies are larviparous insects with a very low reproductive capacity.
During its three to four month lifespan, a female produces on average only five to
seven offspring, hence the low natural population density o f tsetse. In spite o f this,
many tsetse fly species are effective transmitters o f pathogenic Trypanosoma species,
and Glossina control in many regions o f A frica is a prerequisite for sustainable
agricultural production. The use o f conventional area wide control methods, namely
aerial spraying o f insecticides, has been reduced in the past years because o f (a) rela­
tively high expense; (b) insufficient effectiveness against tsetse fly populations in
densely vegetated habitats; and (c) environmental concerns resulting in legislative
restrictions. Currently more favoured methods o f tsetse control are field-by-field
approaches that involve the participation o f local communities for the application o f
insecticides in combination with attractive devices. Although involvement o f the
inhabitants o f the region that is affected by trypanosomiasis is generally beneficial
for the smooth conduct o f vector control operations, any insufficiently controlled use
o f insecticides and trypanocides by individual farmers bears a considerable risk o f
inappropriate application and may promote the development o f resistance among the
trypanosome parasites and tsetse flies. The sterile insect technique (SIT) does not
share the environmentally detrimental aspects o f conventional area wide control pro­
grammes, because o f its specificity against only the target pest insect. H ow ever, the
success o f SIT depends largely on the regular supply o f sufficient and competitive
sexually sterile insects from a mass rearing facility. This paper briefly describes the
status and prospects for the mass rearing o f tsetse flies and their use in vector control
programmes,

2. REA RIN G OF TSETSE FLIES

2 .1 . M eth ods o f rearing tsetse flies

Both sexes o f the tsetse fly are obligatory haematophagous. A variety o f host
animal preferences has been reported for different species [ 1] and there are distinct
habitat requirements among various species [2]. Holding methods for tsetse flies in
the laboratory have to reflect natural demands. These include the maintenance o f an
adequate breeding room climate, the provision o f a suitable b lood diet and the
appropriate manipulation o f different immature and adult stages for optimal colony
female insemination and satisfactory fly performance. Until a few years ago, the
maintenance o f tsetse flies under laboratory conditions required the permanent avail­
ability o f host animals for in vivo feeding. Nevertheless, fly colonies o f considerable
size were held using rabbits, goats or guinea pigs [3 -5 ]. H ow ever, the maintenance
o f a host animal colony for in vivo feeding proved to be problematic and expen-
TA B L E I. H O LD IN G CONDITIONS FOR TSETSE F L Y SPECIES C O L O N IZE D A T THE F A O /IA E A E N T O M O L O G Y UNIT,
SEIBERSDORF

Pupae incubation and em ergen ce o f flies H andling o f y o u n g flies P rod u cin g fem ales
D iet
Species
Incubation, M ean fem ale Rate o f % b ovin e: Sexing H old in g F eeding A g e at 2 d m ating (d) H oldin g Feeding
con ditions pupal p eriod em ergence m ales porcin e m ethod upon con ditions regim en con ditions regim en
M ales F em ales
(d ) (% ). (% ) E M and SEP

G. tachinoides 2 3°C , 32 9 3 .7 4 8 .4 7 5 :2 5 Im m ob ilized 2 4°C , M6 8-10 2 -3 - 2 3 .5 ° C , M4

8 2 % r . h .a ( + 0 .1 m M at + 4 ° C 8 2 % r.h. .80% r.h.
A T P )b

G. p. palpalis 2 4°C , 34 91 .3 4 2 .8 7 5 :2 5 Im m ob ilized 2 4 °C , M6 8-10 2 -3 2 4 °C , M5

88% r.h. ( + lm M at + 4 ° C 88% r.h. 88% r.h.
ATP)

IAEA-SM-327/55
G. f. fuscipes 2 4°C , 36 9 1 .4 5 0.5 7 5 :2 5 Im m ob ilized 2 4 °C , M5 8-10 2 -3 2 4 °C , M5
8 5 % r.h. ( + lm M at + 4 ° C 8 5 % r.h. 8 5 % r.h.
ATP)

G. austeni 2 4°C , 34 9 6 .0 51.1 100% b ovin e Im m ob ilized 2 4 °C , M6 8-10 7 -8 2 4 °C , M5

8 0 % r.h. ( + Im M at + 4 ° C 8 0 % r.h. 8 0 % r.h.
(increased r.h. ATP)
b e fo re E M )

G. pallidipes 2 4°C , . 33 7 7.8 5 0 .5 100% b ovin e Im m obilized 2 4 °C , M5 . 8-10 7 -8 2 4 °C , M5

7 5 % r.h. ( + Im M at + 4 ° C 7 5 % r.h. 7 5 % r.h.
ATP)

G. m. submorsitans 2 4°C , 33 8 9.6 3 2 .0 7 5 :2 5 Im m ob ilized 2 4°C , M5 8- 1 0 . 2 -3 2 4 °C , ' M5 .

7 5 % r.h. ( + Im M at. + 4 ° C 7 5 % r.h. 7 5 % r.h.
A T P )’

G. brevipalpis 2 4°C , 36 94 .7 4 7 .7 7 5 :2 5 Im m ob ilized 2 4°C , M5 8-10 2 -3 2 4 °C , M5

' 8 0 % r.h. ( + Im M at + 4 ° C 8 0 % r.h. 8 0 % r.h.
ATP)

LU
a r.h .: relative hum idity. 00
b A T P : adenosine triphosphate.
582 FELDMANN

Producing fem âfes

Pupae incubation

Young females and m ales

FIG. 1. Colonization of 100 000female G. tachinoides at the FAO/IAEA Entomology Uriit,

Seibersdorf. Flow chart of routine colony maintenance.
 IAEA-SM-327/55 583

sive [ 6]. The development o f artificial membranes [7] and their successive m odifica­
tion for use with different tsetse fly species have greatly facilitated and reduced the
cost for the laboratory maintenance o f Glossina. T o breed a colony o f 30 000 G. pal­
palis palpalis, a host animal colony o f 1000 guinea pigs (4 0 -4 5 % thereof are
feeders, the rest being breeding stock, weaners or sick animals) needs to be main­
tained [6]. The same colony size, fed a quality tested diet using the iri vitro feeding
regimen, consumes 10-15 L o f b lood per week. The basic tsetse handling procedures
and conditions o f incubation o f different developmental stages, as conducted at the
F A O /IA E A Entom ology Unit at the A g e n cy ’s Laboratory at Seibersdorf, are
summarized in Table I. The colonization procedures for mass rearing 100 000
G. tachinoides to produce surplus pupae for use in a SIT project in A frica are illus­
trated in Fig. 1.

2.2. Status of mass rearing

Self-sustaining colonies o f most Glossina species that are o f econom ic im por­

tance can be reared under defined laboratory conditions using the membrane feeding
system [8 -1 0 ]. This not only provides standard fly materials for toxicological, phys­
iological or behavioural studies to develop more effective insecticide formulations,
im prove attractive devices for tsetse control or conduct basic research, but is also
the basis for mass producing high quality insects for release in tsetse control or eradi­
cation programmes in A frica. Figure 2 shows the number o f Glossina spp. pupae
supplied by the F A O /IA E A Entomology Unit to technical co-operation projects in
A frica and to collaborative researchers over past years. Table П [6, 8, 11-13] sum­
marizes the achievements o f membrane fed tsetse colonies that were mass reared to
provide excess fly material for use in vector control programmes.

2.3. Aspects of quality control

Quality assurance needs to be maintained for various stages o f tsetse fly pro­
duction [14, 15].

2.3.1. Quality o f the host animals and blood diet

The health o f animals kept for feeding tsetse requires permanent care. For
smaller scale in v ivo feeding, suitable animals can be bought upon demand. Before
use as hosts to feed tsetse, they require quarantine observation and treatment against
diseases, particularly an effective trypanocidal treatment in case the animals
originate from an area with endemic trypanosomiasis. The feed mixture arid vitamin
supplementation have to reflect the special holding purpose o f the animals as blood
donors. In order not to interfere with tsetse colony performance, the host animal
feeds need to be free o f insecticides, antibiotics [16] or other toxins for tsetse. Limits
 10 000 000 _ m Glossina fuscipes fuscipes
Ui
00
0 Glossina brevipalpis 4b

Щ Glossina pallidipes
1 000 000
0 Glossina austeni
^ Glossina palpalis palpalis
100 000 Ш Glossina tachinoides
No. of pupae supplied

10 000

1000

100

10

E CTJ >, >» -о Е о <в

3 с л С о
(0 CO я ф л "О ¡5 с
D) > E 1Z ф D) О) э ®
Ф о О ф О. N
CD <0 a> N Ф
о О сс (О
.с Ï
о СО
ТЗ ч -
Ф о
ф
N
О с
э
Country
F/G. 2. Tsetse fly pupae supplied by the F A O / I A E A Entomology Unit to I A E A technical co-operation projects and to collaborative researchers.
TAB LE II. IN V IT R O FED Glossina spp. M ASS R E A R E D FO R THE PRO VISIO N OF E XCESS FLIES F O R USE IN V E C T O R
C ONTROL PROGRAM M ES

Location Institute Species Colony size Source Year

Burkina-Faso C R TA, G. palpalis gambiensis 159 415 ‘Rapport d ’ activité’ 1983

Bobo-Dioulasso G. tachinoides 71 068 ‘ Rapport d ’activité’ 1983

IAEA-SM-327/55
G. submorsitans 15 822 ‘ Rapport d ’activité’ 1983

A ll o f the above tsetse species 330 000 Ref. [8] 1984

N igeria B IC O T, Vom G. p. palpalis 139 427 Ref. [6] 1986

G. tachinoides 87 409 Ref. [11] 1991

United Republic o f Tanzania T T R I, Tanga G. austeni 89 418 Ref. [12] Sep. 1992

Austria F A O /IA E A Entomology U n it, G. p. palpalis 80 000 Ref. [13] 1990

Seibersdorf G. tachinoides 130 740 1990

G. austeni 43 650 Oct. 1992

585
586 FELDMANN

for minimum weight and maximum fly challenge per host animal need to be estab­
lished and strictly observed, e .g . for guinea pigs used as host animals, a 550 g mini­
mum weight, a maximum challenge with 200 tsetse colony females at a time and their
use for a maximum o f twice per week as ‘ feeder’ . The maintenance o f large numbers
o f tsetse flies may require a self-sustaining host animal colony with established stan­
dards for colony performance (mortality and female fecundity) and for the condition
o f individual animals (offspring and feeder weight) [6].
Small amounts o f b lood for in vitro feeding o f tsetse can be collected under
sterile conditions from a known and healthy host that is not under any medication
treatment. Once the microbial cleanliness o f such b lood has been confirm ed, it can
be offered to tsetse. Large tsetse colonies require a system o f b lood collection,
microbial decontamination and diet testing with a quantifiable result, i.e. a diet qual­
ity control factor [14]. The system o f diet quality control at the F A O /IA E A Entomol­
ogy Unit includes a bioassay with a small group o f test females [17] and microbial
screenings at different stages o f diet processing (Fig. 3). Thus, only an uncontami­
nated diet with sufficient nutritional quality is offered to the mass reared colony.

2.3.2. Immature stages o f tsetse

The offspring quality is usually monitored routinely by weight one day after
pupation or by size (width class, Ref. [18]). The appropriateness o f incubation con ­
ditions for tsetse fly pupae can be determined by means o f a weekly emergence con­
trol, i.e. the pupal period, per cent emergence rate and sex ratio are monitored from
a group o f 100 pupae. I f the emergence rate is below 85 % , all non-emergent pupae
are dissected in order to determine the developmental stage as an indication o f
whether any interference with ‘ normal’ intrapuparial development may have
occurred. For many Glossina spp., an increased number o f weak females is recorded
if the peak o f female emergence occurs before day 32 post-larviposition. Occasion­
ally, at least once per month, some males from the weekly emergence control should
be dissected in order to determine the development o f their testes and sperm m obil­
ity. In particular, incubation temperatures above 2 7 °C may affect spermatogenesis
or sperm maturation. An adjustment o f climatic conditions is usually sufficient to
achieve an appropriate incubation o f immature stages. Species with rather waxy
puparia, such as G. austeni, require slightly increased humidity during the last third
o f the pupal period (Fig. 4). Without changing the climatic settings o f the breeding
room , a suitable microclimate can be created in the vicinity o f the pupae by placing
moist sponges below slightly m odified pupal dishes in the standard emergence cages.

2.3.3. Adult flies

There is currently no automatic method to identify the sex o f flies upon emer­
gence or for separating males and females after routine colony female mating. The
 IAEA-SM-327/55 587

Collection of 150-200 L fresh blood

In vitro diet quality control

1 kGy irradiation

Bacteriological test
(contaminated diet is discarded)

Colony feeding к

FIG. 3. Production of colony blood and diets.

 100.0 —i

90.0 -

80.0 -
Per cent emergence

70.0 -

Emergence under humid incubation

Emergence in breeding room as of day 20 after iarviposition
60.0 -
at 82% r. h.

x = 76.9 ±5.2% x = 90.3 ± 2.7%

50.0 - t = 17.273 > t 0 да, = 3.387 .

40.0
Dec. 1988 Feb. 1989 Mar. 1989 May 1989

Day of larviposition

FIG. 4. Development of the adult emergence rate in the Glossina austeni colony in relation to relative humidity during the late pupal stage
(r.h.: relative humidity).
 IAEA-SM-327/55 589

age of colony units (weeks)

FIG. 5. Glossina tachinoides under mass rearing conditions. The graph shows the percentage
daily mortality in relation to the age of the female colony units (— ; colony mean;
E M : emergence).

age of colony units (weeks)

FIG. 6. Glossina tachinoides under mass rearing conditions. The graph shows female fecun­
dity in relation to the age of the female colony units (—— : colony mean; E M : emergence).
590 FËLDMANN

âge of colony units (weeks)

FIG. 7. Glossina tachinoides under mass rearing conditions. The graph shows the percentage
of produced small (i.e. А-class) pupae in relation to the age of the female colony units
(-— : colony mean; E M : emergence).

manipulation o f young tsetse is very laborious but permits observations on the

appearance, feeding response, activity and early mortality during the first days after
emergence. Thereafter, the performance o f tsetse colony females is quantified by the
determination o f (a) the survival or mortality, (b) the productivity or fecundity and
(c) the offspring weight or size. The regular collection o f data from weekly form ed
colony female units provides a means to establish quality standards on the above
parameters for the:entire colony and for different female ages. Figures 5 -7 illustrate
the age dependent pattern o f female mortality, fecundity and quality o f offspring
production for a G. tachinoides colony under mass rearing conditions. In order to
trace the sources for increased female mortality (if above 1 .2 % per day on average),
it is necessary to (i) differentiate between the types o f dead females (i.e. starved flies,
flies with undigested b lood in the abdomen and females with a larva that pupated
in utero), and (ii) analyse the mortality in separate colony female units for patterns
that may be age dependent, caused by inappropriate handling or related to the hold­
ing position in the breeding room (humidity and temperature gradients). Concerning
the mean colony female fecundity, a value o f 0 .6 0 pupae per surviving female per
ten days can be regarded as the minimum limit. L ow female fecundity can be caused
by different factors [14], such as bad insemination (use o f premature males, inap­
propriate incubation o f pupae, inadequate mating regim en), non-suitable climatic set­
 IAEA-SM-327/55 591

tings o f the holding room , stress (overcrow ding, rough handling, inadequate
illumination), intoxication or bacterial contamination. In many cases, when low
female fecundity is recorded, an increased number o f abortions (expelled at the early
or late pregnancy stages) are also observed. Factors causing inferior performance o f
colony females need immediate correction in order to prevent detrimental effects on
colony size or the supply o f excess pupae to field projects.

2.3.4. Dispatched insect material

The quality o f dispatched pupae and the suitability o f methods used for packing
and transport can be monitored as described above for immature stages. The quality
o f sexually sterile males can be assessed under laboratory conditions by means o f
mating competitiveness and survival tests in large cages or with flight activity simu­
lations in specially designed flight mills.

2.4. Labour and cost for rearing tsetse

Investigations were conducted recently at the F A O /IA E A Entom ology Unit on

the labour requirements to maintain a 100 000 female G. tachinoides colon y. Using
the current colonization procedure, i.e. for young flies and producing flies, an M 6
and an M 4 feeding regimen, respectively (membrane feeding on M onday, Tuesday,
Thursday and Friday), and maintenance o f colony female units for 18 weeks,
approximately 140 labour hours o f experienced technicians are required. O f this
time, 45% (or 63 h) is spent on the manipulation o f young flies and 30% (or 42 h)
is needed for blood collection, diet processing, quality control and feeding o f colony
flies. The remaining time is spent on pupae handling, removal o f dead females from
holding cages, data collection and evaluation, washing, cleaning and sterilizing o f
materials and other routine w ork. This is a reduction by 50 h when com pared with
the labour needed for colony maintenance in early 1991, when fly producing females
were still fed five times per week and when old female colony units were disposed
o f at the age o f 13 weeks instead o f 17-18 weeks. Other costs for the colonization
o f tsetse at the F A O /IA E A Entom ology Unit include buildings, equipment and
materials and are summarized in Table Щ as they are actually incurred at
Seibersdorf.

3. USE OF L A B O R A T O R Y P RO D U CE D TSETSE
IN V E C T O R C O N T R O L P RO G R AM M E S

There may be either continuous demand for sexually sterile insects throughout
the year until eradication o f the target pest insect has been achieved, or flies are
T A B L E III. COSTS FO R M A SS REARIN G 100 000 FEM ALE Glossina tachinoides AS IN C U R R E D IN 1991-1992 A T THE
F A O /IA E A E N T O M O L O G Y UNIT, SEIBERSDORF (SALARIES E X C L U D E D )

1. Facilities
Amortization Total cost
No. Item Cost per unit time Cost per year per pupa
(US $) (years) (US $) (US $)

30 m 2 Fly holding/breeding room for producing females

25 m 2 Feeding room

45 m2 Handling room

10 m2 For pupae incubation

10 m2 Fly holding/breeding room for young flies

lO m 2 Washing area

10 m 2 Blood processing

10 m 2 Store

150 m 2 Total 1 250 187 500 25 7 500 0.0029

TA BLE III. (cont.)

2. Equipment
Amortization Total cost
No. Item Cost per one Total cost time Cost per year per pupa
(US $) (US $) (years) (US $) (US $)

16 Fly holding trolleys 550 8 800 15 587

5 Air conditioners 3 000 15 000 10 1 500

3 Humidifying systems (incl. air ventilation systems) 3 500 10 500 10 1 050

IAEA-SM-327/55
3 Climate control systems 1 500 4 500 10 450

50 Pupae holding trays 4 200 5 40

100 Emergence cages 20 2 000 5 400

5000 S C -11 fly holding cages for young females 2 .50 12 500 5 2 500

1600 M C -20 fly holding cages for producing females 5 8 000 5 1 600

2500 Male holding cages 2.50 6 250 5 1 250

100 Silicone membranes 20 2 000 1 2 000

100 Aluminium feeding trays 20 2 000 10 200

3 Washing machines (1 each for cages and feeding trays) 1 000 3 000 5 600

2 Heat sterilizing ovens 1 200 2 400 10 240

1 Pupal sorting machine (quality control) 7 000 7 000 10 700

593
 594
T AB LE III. (cont.)

Amortization Total cost

No. Item Cost per one Total cost time Cost per year per pupa
(US $) (US $) (years) (US $)

1 Pupae counting machine 3 500 3 500 10 350

3 Fly chillers 1 850 5 550 5 1 110

1 Laminar air flow bench 15 000 15 000 15 1 000

FELDMANN
3 Electric blood stirrers 800 2 400 15 160

3 Blood collection sets 480 1 440 5 288

4 Freezing cabinets (500 L) 1 000 4 000 10 400

2 Refrigerators 600 1 200 10 120

1 Set shelves, benches, tables, chairs 5 000 5 000 10 500

122 240 17 045 0.0066

1 Gamma radiation source 160 000 160 000 10 16 000

33 045 0.0129
(including the
gamma source)
T A B L E III. (cont.)

Total cost
Cost per year per pupa
(U S $ ) (US $)

IAEA-SM-327/55
3. Supplies

Blood, chemicals, glassware 20 000 0.0078

4. Overhead costs

Running overhead costs at the Agency’ s Laboratory 74 460 0.0290

depend on the number of staff involved in each project

Total costs 135 005 0.0526

L /l
40
Ui
596 FELDMANN

released on a seasonal basis if other control methods cannot be used due to the inac­
cessibility o f the control area, e .g . in the rainy season. Depending on the nature and
location o f the field programme, different types o f rearing facilities may be erected,
such as:

— A stationary insect factory for national campaigns with a relatively small treat­
ment area,
— A large breeding centre for the provision o f sterile insects to several neigh­
bouring countries (regional programmes),
— M obile insect factories (ship or land based container system) that can be shifted
as the release area changes in accordance with the expansion o f the eradication
zone.

In addition to procedures for quality assurance o f the released insects, the long
distance transfer o f tsetse from a rearing facility may require the co-operation o f
various relevant agencies, such as veterinary, quarantine and customs authorities,
brokers and importation firms, airlines and other transport businesses.

3.1. Sterile males for SIT

‘ Conventional’ means o f tsetse control, e.g. the use o f insecticides or trapping,

are usually more efficient at the beginning o f a treatment campaign than later, when
the density o f the target population is lower. The SIT has a converse efficiency pat­
tern: it becom es more efficient with progressive reduction o f the target insect popula­
tion. Sexually sterile tsetse fly males can therefore be more efficiently utilized when
the phase o f release follow s an initial period, e .g . three months, o f ‘ conventional’
population suppression. Sterile males need to disperse and behave like native fertile
males. The release location, interval and the numbers o f males released are chosen
such that throughout the fly habitat the sterile males outnumber the fertile wild males
by at least 10 to 1. This ratio must be maintained for at least three to four generation
periods (one tsetse generation period is approximately 50 d). Thus, in the dense
riverine fly habitat o f the southern Guinea vegetation zone, the eradication o f Glos­
sina p. palpalis requires during the release phase some one hundred sterile males per
linear kilometre o f riverine forest per week. This number o f males can be supplied
by a tsetse colony o f 1000 females.

3.2. Sterile females for sterile virgin female ecological monitoring

The collection o f data on tsetse population density and dynamics is important

for appropriate decision making in vector control operations. During the advanced
stage o f tsetse control, the density o f the target population may be below the level
detectable by trapping; furthermore, the collection o f information requires extensive
trapping and is laborious and relatively expensive. A t this stage the release o f labora-
 IAEA-SM-327/55 597

tory reared, sexually sterile virgin females into the target habitat and their recapture
and dissection (sterile virgin female ecological monitoring (S V FE M )) may have
technical and econom ic advantages.
Preliminary results from tests with various tsetse fly species colonized at
Seibersdorf (i.e. G. p. palpalis, G. tachinoides, G. f fuscipes, G. austeni, G. pal-
lidipes and G. brevipalpis) provide evidence that:

(a) Female Glossina spp. can be radiation sterilized sexually (ovary atrophy) by
means o f a 60 G y gamma treatment one week before expected emergence.
(b) There is evidence that such radiation sterilized pupae may be shipped over a
long distance without affecting their emergence (provided the packing and
transport conditions and duration are adequate).
(c) Sexually sterile virgin females emerging from 60 Gy gamma ray treated pupae
survive well and indicate no difference in mating behaviour and receptivity to
insemination when compared with fertile virgin females.
(d) Sterile virgin females may be released into the fly habitat o f the target popula­
tion and be recaptured later. Inseminated recaptured females prove the
presence o f native males (i.e. the prevalence o f a wild population).

F or the detection o f a relic tsetse target population, the time interval between
release o f the sterile females and their recapture may be extended, which increases
the chances o f sexual contact between released sterile females and native males, thus
amplifying the monitoring results. In order to explore the feasibility and econom ics
o f SVFEM as a detector technique, releases and recaptures o f virgin sterile tsetse
females have to be conducted before, during and after vector control operations.

3.3. Tsetse for hybridization with closely related species or subspecies

The genetic implications o f the hybridization o f different species or subspecies

have been reviewed [19]. In the morsitans group o f tsetse, evidence for conspecific
sperm selection by the female for egg fertilization, however, presents an obstacle for
using the SIT for trans-specific eradication. For species o f the palpalis group, similar
evidence for conspecific selection o f sperm has not been found.

3.3.1. Closely related subspecies in the palpalis group

Laboratory studies o f G. palpalis palpalis and G. palpalis gambiensis indicated

that there is no preference for selective inter- or intrasubspecific mating [20]. ‘ Fer­
tile’ intersubspecific crosses result in a reduced emergence rate among the F, o ff­
spring and à sex ratio distortion. Furthermore, most Fem ales are sterile (aspermia).
M orphometrical studies on the male hypopygium in the field provide evidence for
hybridization o f G. p. palpalis and G. p. gambiensis. In natural habitats, where the
possibility o f contact between the two subspecies exists, only a very narrow corridor
598 FELDMANN

o f hybridization has been detected. This limited extent o f the hybridization zone is
another indication o f the occurrence o f hybridization in the field, as it is probably
caused by the reduced emergence and the sex ratio distortion among hybrid offspring
and the sterility o f hybrid males.

3.3.2. Closely related species in the palpalis group

Preliminary results from laboratory investigations on the mating behaviour o f

G. p. palpalis and G. f. fiiscipes [21] suggest the feasibility o f using sterile, labora­
tory reared G. p. palpalis males and females for the control o f G. f. fuscipes-.

(a) G. f. fuscipes males, if given the ch oice o f mating G. f. fuscipes or G. p. pal­

palis females, express a preference to mate with females o f the other taxon.
(b) G. p. palpalis males mate with females o f both taxa at random. H ow ever, the
abdomen o f G. / . fuscipes females that were mated by G. p. palpalis males
show severe wounds and approximately 40% o f such females die within few
days after mating as a result o f secondary microbial infections.

The above results suggest the use o f laboratory produced sexually sterile flies
from one species for the control o f its own and two other species or subspecies, i.e.
G. p. palpalis, G. p. gambiensis and G. f. fuscipes. The econom ic implications o f
the mass rearing o f tsetse flies, i.e. producing only one species in the laboratory to
combat at least three in the field, are obvious.

4. BOTTLENECKS A N D R & D REQUIREM ENTS FOR L A R G E R SCALE

FIELD USE OF M A SS REA RE D TSETSE

The establishment o f a small laboratory colony from wild fly material using
fresh sterile whole b lood with the membrane feeding system is relatively simple for
most tsetse fly species, provided there are adequate maintenance conditions.
How ever, the transfer o f such a strain to mass rearing procedures, such as the adap­
tation to (a) immobilization at + 4 ° C for sex recognition, (b) mass processed b lood
diet, or (c) higher female densities in larger cages, usually lasts about two years and
may cause undesired selective pressure. Information is required on the genetic diver­
sity and the fate o f symbionts during this process o f adaptation. Efforts should aim
at a reduction o f the adaptation period to the mass rearing system.
Procedures for morphological sex recognition and separation, or the develop­
ment o f a genetic sexing technique without losses in fly performance would double
the number o f males that can currently be supplied by a tsetse colony and is the
prerequisite for large scale mass production o f sexually sterile tsetse flies for use in
vector control projects. Once this has been achieved, the mass rearing o f tsetse can
be automated.
 IAEA-SM-327/55 599

Basic information is needed on the genetic backgrounds o f the target popula­

tions and mass reared strains in order to determine ( 1 ) the suitability o f the colony
fly material for release and ( 2) the existence o f genetically isolated fly populations
or the gene flow between neighbouring populations. The latter is very important to
assess the risk o f reinfestation o f a determined eradication zone by fly populations
in the border areas. I f necessary, methods that permit the introduction o f the genetic
background o f the wild target strain into the mass reared strain need to be elaborated.
Although it is known that the current procedures o f fly handling prior to release
reduce almost to zero the probability o f a sterile male becom ing infected, future
im proved, mass reared tsetse strains will have to be refractory to trypanosome infec­
tion. The interrelationship between trypanosomes, symbionts and RLO s and tsetse
has to be further investigated in order to incorporate principles for refractoriness into
mass reared strains.
The demand for sterile tsetse may fluctuate during the year and the insect
production may be in excess o f or below demand. Biological materials that are
produced for pest insect control should be storable like conventional insecticides.
Existing knowledge on the preservation o f other insects [22] should be reviewed for
their applicability to produce in advance and stockpile tsetse. Methods o f shipping
tsetse over a long distance from regional breeding centres to control projects are also
expected to benefit from this research.
The efficiency o f SVFEM would increase with enhanced attractivity o f
released virgin females to relic native males (colours, manipulation o f the short range
sex pheromone). First, however, detailed field studies are proposed to obtain further
information on the feasibility and econom ic aspects o f using sterile virgin females
as detectors o f tsetse populations before, during and after vector control operations.
Sexually sterile tsetse from a mass production facility can be used for eradica­
tion or control in combination with or in support o f other regional or field-by-field
applied methods o f trypanosomiasis control. Provided the research topics identified
above are sufficiently covered and the main bottlenecks eliminated, the use o f tsetse
flies from a mass production facility has considerable potential for the efficient and
specific control o f several econom ically important Glossina species. The proposed
erection o f regional mass breeding facilities close to international airports would
permit the supply o f sexually sterile tsetse to projects in neighbouring countries upon
demand.

REFERENCES

[1] WEITZ, B., The feeding habits o f Glossina, Bull. World Health Organ. 28 (1963)
711-729.
[2] BUXTON, P. A ., The Natural History o f Tsetse Flies, Mem. No. 10, School ó f Tropi­
cal Medicine and Hygiene, London (1955).
600 FELDMANN

[3] N ASH , T .A .M ., JORDAN, A .M ., B O YLE, J .A ., The large scale rearing o f Glossina

austeni (Newst.) in the laboratory. IV. The final technique, Ann. Trop. Med. Parasitol.
62 (1968) 336-341.
[4] W IL LIAM SO N , D .L ., et al., Integration o f insect sterility and insecticides for control
o f Glossina morsitans morsitans Westwood (Diptera: Glossinidae) in Tanzania. I.
Production o f tsetse flies for release, Bull. Entomol. Res. 73 (1983) 259-265.
[5] V A N DER V LO E D T, A .M .V ., “ Recent advances in tsetse mass rearing with particu­
lar reference to Glossina palpalis palpalis (R ob.-D esv.) fed in vivo on guinea pigs” ,
Sterile Insect Technique and Radiation in Insect Control (Proc. Symp. Neuherberg,
1981), IA E A , Vienna (1982) 223-253.
[6] OLADUNM ADE, M .A ., et al., “ Eradication of Glossina palpalis palpalis
(Robineau-Desvoidy) (Diptera: Glossinidae) from agropastoral land in central Nigeria
by means o f the sterile insect technique” , Sterile Insect Technique for Tsetse Control
and Eradication (Proc. Final Research Co-ordination Mtg Vom, Nigeria, 1988), IAE A,
Vienna (1990) 5 -2 3 .
[7] BAUER, B ., W E TZEL, H .W ., A new membrane for feeding Glossina morsitans
(Westwood) (Diptera: Glossinidae), Bull. Entomol. Res. 65 (1976) 319-326.
[8] CUISANCE, D ., et al., “ La campagne de lutte integrée contre les glossines dans la
zone pastorale d’ accueil de Sidéradougou (République du Burkina-Faso)” , Proc. 18th
O AU /STR C (ISCTRC) Meeting Harare, Zimbabwe, 1985, Doc. No. 604 (1986)
33 4-34 3. ■
[9] G A O , M .K ., CH ALO , О ., BAKULI, B ., Laboratory maintenance o f tsetse flies on
membranes, Res. Training Newsl. 5 2 (1990) 16-18.
[10] FELD M A N N , U ., et al., “ Tsetse fly mass rearing: Colony management, deployment
of sterile flies, related research and development” , Tsetse Control, Diagnosis and
Chemotherapy using Nuclear Techniques (Proc. Seminar Muguga, Kenya, 1991),
IA E A -T E C D O C -643, IA E A , Vienna (1992) 167-180.
[11] D E N G W A T , L ., PARKER, A ., personal communication.
[12] G A O , M .K ., personal communication.
[13] V A N DER VLO ED T, A .M .V ., et al., “ Research and development in the IAEA
Laboratory at Seibersdorf in support o f BICOT for the eradication o f Glossina palpalis
palpalis", Sterile Insect Technique for Tsetse Control and Eradication (Proc. Final
Research Co-ordination Mtg Vom , Nigeria, 1988), IA E A , Vienna (1990) 2 5 -2 9 .
[14] FELD M A N N , U ., ‘ ‘ Guidelines for the rearing o f tsetse flies using the membrane feed­
ing technique” , Manual on Insect Rearing, International Centre for Insect Physiology
and Ecology, Nairobi (in press).
[15] F ELD M A N N , U ., “ Some quality control parameters used in the rearing o f tsetse
flies” , ibid.
[16] N OGGE, G ., Sterility in tsetse flies ( Glossina morsitans Westw.) caused by loss o f
symbionts, Experientia 32 (1976) 995.
[17] V A N DER V LO E D T, A .M .V ., unpublished.
[18] ZELGER, R ., RUSS, K ., Puppentrennmaschine mit zwei Walzenpaaren, Z . Angew.
Zool. 63 3 (1976) 257.
[19] GOODING. R .H ., Tsetse genetics: A review, Quaest. Entomol. 20 (1984) 89-1 28 .
 IAEA-SM-327/55 601

[20] VR EYSEN , V A N DER V LO ED T, A .M .V ., The effect o f intersubspecific

hybridization and gamma radiation on the reproductive biology o f Glossina palpalis
palpalis (Robineau-Desvoidy) and Glossina palpalis gambiensis Vanderplank, Ann.
Soc. Belg. Méd. Trop. 20 (1990) 145-158.
[21] F E LD M A N N , U ., BARNOR, H ., Laboratory investigations on the interspecific
mating behaviour o f Glossina palpalis palpalis (Robineau-Desvoidy) and Glossina
fuscipes fuscipes (Newstead) for the trans-taxon use o f sexually sterile tsetse (in
preparation).
[22] LEE, R .E ., Jr., DENLINGER, D .L . (Eds), Insects at Low Temperature, Chapman and
Hall, London (1991).
 IAEA-SM-327/57

H Y B R ID IZ A T IO N O F Glossina swynnertoni
W IT H S U B S P E C IE S O F Glossina morsitans
( D IP T E R A : G L O S S IN ID A E )
Implicationsfor use of hybrid sterility
and satyrsfor genetic control of tsetse
R .H . G O O D IN G
Department o f Entom ology,
University o f Alberta,
Edmonton, Alberta,
Canada

Abstract

HYBRIDIZATION OF Glossina swynnertoni W ITH SUBSPECIES OF Glossina morsitans

(DIPTERA: GLOSSINIDAE): IMPLICATIONS FOR USE OF HYBRID STERILITY AN D
SATYRS FOR GENETIC CONTROL OF TSETSE.
Adult Glossina swynnertoni Austen that emerged from puparia collected during 1989
and 1991 near Makuyuni, United Republic o f Tanzania, and their progeny were hybridized
with G. morsitans centralis Machado, G. morsitans morsitans Westwood, or G. m. submorsi-
tans Newstead. All the G. swynnertoni females were receptive and mated readily, regardless
of whether they were placed with conspecifics or members o f a subspecies o f G. morsitans.
Glossina swynnertoni males attempted to mate with females o f all three subspecies o f
G. morsitans; the order o f increasing success in pair formation was G.m. centralis,
G. m. morsitans and G. m. submorsitans . About 50% o f the G. swynnertoni females were
fertilized by conspecific males, about 25% by G. m. centralis and none were fertilized by
G. m. morsitans or G. m. submorsitans. Glossina swynnertoni males fertilized from 35 %
(G. m. submorsitans) to 68% (G. m. morsitans) o f their mates. All the F, hybrid males were
sterile, but most o f the F, hybrid females were fertile and could be backcrossed to either
parental taxon. Recurrent backcrossing produced sterile and fertile males, and an X chromo­
some marker gene was used to demonstrate that a major cause o f male hybrid sterility was
an incompatibility o f the sex chromosomes o f G. swynnertoni with those o f G. m. centralis
and G. m. morsitans.

1. IN TR O D U C TIO N

The genus Glossina Wiedemann consists o f 31 species and subspecies arranged

in three subgenera [ 1 ,2 ] . Adult tsetse flies are haematophagous and are the vectors
o f trypanosomes throughout most o f sub-Saharan A frica. The subgenus Glossina
s. str. includes seven species and subspecies and it is usually accepted that within
this subgenus Glossina swynnertoni Austen is closely related to G. morsitans sensu

603
604 GOODING

lato W estw ood [1, 3; 4 ], and it has been suggested that G. swynnertoni recently
evolved from G. morsitans as a species that is adapted to drier and m ore open
savannah [5].
Many species or subspecies o f tsetse will hybridize with members of. closely
related taxa. From a historical perspective, the most interesting species is
G. swynnertoni. This species was one o f the first to be studied in hybridization
experiments in the laboratory [3, 6- 8] and in the field [9], and it was the first
species o f insect against which genetic control methods were used [ 8, 10].
Vanderplank reduced the number o f G. swynnertoni in an isolated population by
releasing G. m. centralis W estw ood at that site [ 8, 10]. He interpreted the reduction
in the G. swynnertoni population as the result o f hybrid sterility, but Ribeiro [11]
interpreted this reduction as being the result o f ‘ satyrist’ behaviour, i.e. the tendency
o f males o f one taxon to mate with females from another taxon. Ribeiro [11] p ro­
posed that satyrs may be useful in the genetic control o f insects. This suggestion has
been challenged, at least for use against subspecies o f G. morsitans, by the demon­
stration that polyandrous females o f this species generally use, the sperm from their
ow n taxon only [ 12 ].
The objectives o f this study were to determine the hybridization relationships
between G. swynnertoni and the three subspecies o f G. morsitans and to use this
hybridization data to determine the potential o f hybrid sterility as a genetic control
agent for use against populations o f G. swynnertoni or subspecies o f G. morsitans.

2. M A TE RIALS A N D M ETH ODS

Puparia o f G. swynnertoni Austen were collected in the Naitolia Ranch, Rift

Valley Seed Company Ltd, about 20 km south o f Makuyuni (at the junction o f high­
ways A104 and B142) in the northern part o f the United Republic o f Tanzania during
August 1989 (about 700 puparia) and during late February and early March 1991
(about 800 puparia). The puparia were transported by road (to Nairobi, Kenya, in
1989 and to Arusha, United Republic o f Tanzania, in 1991) and then by air to
Edmonton, Canada. In 1989, approximately 300 adults emerged and in 1991
approximately 250 adults emerged. In 1989, deformed adults were com m on and
mortality was very high among the young flies, but in 1991 only a few adults were
deformed. The flies were maintained in the laboratory at 2 5 °C and 6 5 -8 0 % relative
humidity by feeding six or seven days per week on rabbits.
In a preliminary experiment in 1989, G. swynnertoni males that had mated with
G. swynnertoni females always attempted to mate with the females o f all three
G. morsitans subspecies, but were least likely to be accepted by G. m. centralis and
most likely to be accepted by G. m. submorsitans. In a subsequent experiment,
19 virgin male G. swynnertoni that were one to two weeks old were individually
placed with a female o f one G. morsitans subspecies; if the male was rejected, it was
placed with a female o f another subspecies. Each female that rejected a G. swynner-
 IAEA-SM-327/57 605

toni male was receptive when placed with a male o f its own subspecies. Males were
used up to five times, at 4 8 -7 2 h intervals. Each day, G. swynnertoni males were
placed with females in the order G. m. centralis, G. m. morsitans and finally
G. m. submorsitans.
In the 1991 experiments, males were at least seven days old and females were
three or four days old when tested. With the exception o f 14 males that were mated
twice with G. swynnertoni, all the males emerging from field collected G. swynner­
toni were mated once only. H ow ever, the 15 laboratory reared G. swynnertoni males
used to establish the hybrid pedigrees were mated four times, at 4 8 -7 2 h intervals.
The first mate was a G. swynnertoni female, after which one third o f the males
were placed with females in the order G. m. centralis, G. m. morsitans and
G. m. submorsitans', one third in the order G. m. morsitans, G. m. submorsitans and
G. m. centralis; and one third in the order G. m. submorsitans, G. m. centralis
and G. m. morsitans.
Females that became pregnant were maintained, by the procedures mentioned
above, to obtain offspring for further experiments. Hybrid and backcross females
were mated with males from one o f their parental taxa and maintained to obtain
offspring for further fertility tests. Recurrent backcrosses (to G. swynnertoni or
to the appropriate G. morsitans subspecies) were accomplished with the crosses
G. m. centralis. X G. swynnertoni, G. m. morsitans X G. swynnertoni and
G. m. submorsitans X G. swynnertoni. (Throughout this paper the taxon of. the
female is specified first in an inter-taxon cross.)
Each hybrid or backcross male was mated with two to four females from his
parental G. morsitans subspecies. Forty-eight hours after mating, each o f the test
females was dissected and the spermathecae examined (at 160 X ) for motile sperm.
Those males that passed motile sperm were mated with several m ore G. morsitans
subspecies females and each female was observed daily until she became pregnant,
or for 28 d, at which time the spermathecae were examined for motile sperm.
Only one marker gene (Pgm, the locus for phosphoglucomutase) was available
to distinguish the X chrom osom es o f G. swynnertoni from those o f G. m. centralis
and G. m. morsitans. The electrophoretic procedure used [13] gave phosphogluco­
mutase (PG M ) bands with the follow ing migrations, relative to the bromophenol blue
marker: 0 .5 8 , 0.62 and 0.65 for G. m. centralis, 0.58 and 0.62 for G. m. morsitans,
and 0.55 and 0 .59 for G. swynnertoni. The origins o f the Y chrom osom es were
established on the basis o f the taxon o f the sire.

3. RESULTS

3 .1 . M ating beh aviou r

Glossina swynnertoni males always attempted to mate with females o f G. swyn­

nertoni, G. m. morsitans, G. m. centralis and G. m. submorsitans. In the 1989
606 GOODING

TAB LE I. RESULTS OF PAIRING G. swynnertoni W ITH G. swynnertoni O R

W ITH G. morsitansa

Insemination arid pregnancy0

Mated by
G. swynnertonib Mates of
G. swynnertoni
Taxon tested G. swynnertoni
Females Males females
males

G. swynnertonid 100 (65) 100 (65) 25/53/61 25/53/61

G. m. centralis 100 (10) 43.5 (23) 2/3 /8 4 /4 /1 0

G. m. morsitans 100 (10) 100 (11) 0 /2 /7 8/8/11

G. m. submorsitans 100 (10) 100 (16) 0/6 /10 5 /8 /16

a The adult G. swynnertoni used in these experiments had emerged from puparia collected
in the United Republic o f Tanzania in 1991 and shipped to the University o f Alberta,
Edmonton, Canada.
b The numbers in these two columns are the percentage pairs that copulated (the number of
pairs tested is in parentheses).
c The numerals in the last two columns are: No. o f pregnant/No. o f inseminated/No. o f mated
females alive 28 d after mating.
d Data for the same 65 pairs appear in all the columns.

experiments, o f 19 G. swynnertoni males that were placed with females from one,
two or three G. morsitans subspecies in 1 d, four mated with G. m. centralis,
15 mated with G. m. morsitans and 14 mated with G. m. submorsitans. Three males
were rejected by G. m. centralis, then by G. m. morsitans and then accepted by
G. m. submorsitans. On one occasion, a male was rejected by the females o f all three
G. morsitans subspecies; this male mated with two G. m. submorsitans and one
G. m. morsitans on other days.
In the 1991 experiments, all the G. swynnertoni females accepted their mates,
regardless o f whether these were conspecifics or members o f a G. morsitans
subspecies (Table I). M ore than half o f the G. m. centralis females rejected
G. swynnertoni males that emerged from field collected puparia, but the rejected
males were accepted by G. m. morsitans or G. m. submorsitans (Table I). Similar
results were obtained using males from the first generation o f a laboratory colony
o f G. swynnertoni (Table П).
 IAEA-SM-327/57 607

TA B L E П. RESULTS OF PAIRIN G L A B O R A T O R Y REA RE D G. swynnertoni

M A L E S 3 W ITH FEM A LES OF G. m. centralis, G. m. morsitans A N D
G. m. submorsitans

Inseminated Offspring
Per cent mating and produced during
Taxon o f females (No. tested) pregnantb three months

G. m. centralis 47 (15) 4 /4 /7 Four males: two females (6)c

G. m. morsitans 93 (15) 9 /9 /1 4 Four males: nine females (16)

G. m. submorsitans 100 (15) 7/7 /15 Six males: one female (7)

a Males were from the first generation that was laboratory reared, and each had mated with
a G. swynnertoni female before being tested with a G. morsitans female.
b The numerals are: No. o f pregnant/No. o f inseminated/No. o f mated females alive 28 d
after mating.
c The number o f puparia is given in parentheses.

3.2. Insemination and fertilization of G. swynnertoni

and G. morsitans females

In the 1989 experiments, none o f the females that mated with G. swynnertoni
became pregnant. O f 12 G. swynnertoni females examined, only three were insemi­
nated, and none o f the 65 females o f the various G. morsitans subspecies were
inseminated. A t that time, the three G. swynnertoni males that were still alive were
dissected; all had motile sperm in their testes. It is not known why the G. swynnertoni
in the 1989 sample were unable to fertilize females.
In the 1991 experiments, only 41% o f the G. swynnertoni females were ferti­
lized by conspecifics, but 28 o f the 36 females (7 8 % ) that did not becom e pregnant
were inseminated (Table I). None o f the G. swynnertoni females that mated with
G. m. morsitans or G. m. submorsitans became pregnant, but two o f those mating
with G. m. centralis did (Table I).
Polling the data on males emerging from field collected puparia (Table I) with
the data from males o f the first generation o f the laboratory colony (Table П) showed
that G. swynnertoni males fertilized 47% o f G. m. centralis, 60% o f G. m. morsitans
and 39% o f G. m. submorsitans.
608 GOODING

TAB LE III. FERTILITY OF F, HYBRIDS PRO DU CED BY CROSSING

G. swynnertoni (Gs)a W ITH G. m. centralis (Gmc), G. m. morsitans (Gmm) A N D
G. m. submorsitans (Gms)

Taxon Males Testing o f hybrid femalesb

Maternal taxon Paternal taxon

Female Male Fertile Sterile0
male male

Gs Gmc 0 2 1/1/2 1/1/1

Gmc Gs 0 2d 5/5/5 1/1/1

Gmm Gs 0 11 8/8/10 6 /6 /6
О

Gs
00

Gms 1/2/2 2 /2 /2

“ The data are from field collected and laboratory reared G. swynnertoni and include F,
hybrids o f the pedigrees Gmc/Gs, Gmm/Gs and Gms/Gs.
b The numerals are: No. o f pregnant/No. o f inseminated/No. o f mated females alive 28 d
after mating.
c The criterion for sterility was the inability to inseminate a female o f the appropriate
G. morsitans subspecies.
d Three other males died when less than one week old.
e One other male died when less than one week old.

3 .3 . Fertility o f h yb rid and back cross flies

A ll the F] hybrid males obtained by crossing G. swynnertoni with the

G. morsitans subspecies were, sterile, but most o f the F [ hybrid females could be
fertilized by backcrossing to either parental taxon (Table III).

3.3.1. Pedigree Gmc/Gs

This pedigree began with four G. m. centralis females fertilized by

G. swynnertoni. Three o f the four Fj males died when less than one w eek old; the
fourth was sterile. One F, female (F! female A ) was backcrossed to G. swynnertoni
(Fig. 1) and the other F, female (Fj female B) was backcrossed to G. m. centralis
(Fig. 2).
Recurrent backcrossing to G. swynnertoni produced fertile females in back-
cross generations B x ^ i) to Bx 3(s) (Fig. 1), but switching to recurrent backcrossing
to G. m. centralis produced both fertile and non-fertilized females in backcross
generations B x j(i) to B x(s/c) (Fig. 1). Four o f these non-fertilized females were not
 IAEA-SM-327/57 609

inseminated; the fifth was not checked for sperm. Fertile males were not found in
the first three backcrosses, but did occur among Bx 4(s) and Bx 4( j/c ) (Fig. 1).
M ost o f the females produced by recurrent backcrossing to G. m. centralis
were fertile (Fig. 2 ), with only two non-inseminated females found among the
Bx 3(c) females. Fertile backcross males occurred among the second to fourth back-
cross generations (Fig. 2).
A ll but two o f the 39 backcross males that were placed with the females from
all three G. morsitans subspecies were accepted by the females from all three subspe­
cies. The exceptions (Bx 3(s) and Bx 4(s) males, Fig. 1) were rejected by the
G. m. centralis females.
O f the 37 males that mated with the females from all three G. morsitans
subspecies, two failed to fertilize G. m. centralis, four did not fertilize
G. m. morsitans, two did not fertilize G. m. submorsitans and one fertilized
G. m. submorsitans only. The remaining 28 males fertilized the females from all
three G. morsitans subspecies.
A ll 15 backcross males that had an X from one taxon and a Y from another
were sterile (Table IV ). Fertile and sterile individuals occurred among the 76 males
that had both sex chrom osom es from the same taxon.

3.3.2. Pedigree Gmm/Gs

This pedigree began with nine G. m. morsitans females fertilized by

G. swynnertoni that produced 16 puparia. A ll four F, males were sterile.
Four F! females were fertilized by backcrossing to G. m. morsitans, but only
seven backcross (B xj) males and five Bx! females were obtained. T w o Bx! males
died when less than one w eek old and the others were sterile. A ll the B x[ females
were mated with G. m. morsitans males, but none became pregnant and only two
were inseminated.
The remainder o f pedigree Gmm/Gs is shown in Fig. 3. In each backcross
generation, there were both fertilized and non-fertilized females. A m ong the latter,
some were inseminated (one each o f Bx 3(wi), Bx 2(s) and Bx 3(s)) and others were not
(one each o f Bx 2(w ), В х 3(/я), B xj(s) and Bx 2(s)). Only one o f the 16 B x j(í) males
(B x, male 13) was fertile and no further fertile males were found until Bx 3 (Fig. 3).
During these experiments, seven males (Bx 3 and BX4) were placed with
females from all three subspecies o f G. morsitans. The males were always accepted
by G. m. morsitans and G. m. submorsitans, but three males were rejected by
G. m. centralis. Another four Bx 3 males that were rejected by G. m. centralis
mated with G. m. submorsitans, but were never offered G. m. morsitans as mates.
Alm ost all o f the males were sterile, regardless o f the taxonomic origin o f their
sex chrom osom es (Table IV ). Only two fertile males were tested with all three
subspecies o f G. morsitans', one male fertilized the females from all subspecies and
the other male fertilized G. m. morsitans and G. m. submorsitans only.
 Gmc 9 9 Gs crcr
(four fertilized)

610
I
F, ç A x Gs cr
I____________ _________________ I

Bx, (s) Ç Gs cr Bx, (s)o’ o' Bx, (s) 9 9 Gmc сy a

(one fertilized) (four sterile) (two fertilized,
two not fertilized)
-I_____

G O O D IN G
Bx2 (s) Ç 9 x Gs <у a Bx2 (s) 9 Bx2 (s/c) 9 9 x Gme o'er No c c r
(two fertilized) (one fertile) (three fertilized,
one not fertilized)

Bx3 (s) Ç ç x Gs a <y Bx3 (s) cr a Bx3 (s/c) 9 9 . > Gmc cr о• Bx3 (s/c) a a
(three fertilized) (three fertile) (two fertilized, (nine sterile)
two not fertilized)

Bx4 (s) cr cr Bx4 (s/c) cr с

(three fertile, (four sterile)
four sterile)

FIG. 1. Portion of pedigree Gmc/Gs that involved recurrent backcrossing to G. swynnertoni. (Gmc: G. m. centralis; Gs: G. swynnertoni.)
(See text for which of these females were inseminated.)
 Gmc 9 9 x Gs crcr
(four fertilized)

F, 9 В Gmc a
___I

Bx, (c) 9 9 Gmc a o' Bx, ( c ) a a Bx, (c) 9 9 Gmc a cr

(five fertilized) (two sterile) (one fertilized)

Bx2 (c) 9 9 Gmc crcr Bx2 (с) оr a Bx2 (c) 9 9 Gs a a Bx2 (c) a
(17 fertilized) (four fertile, (two fertilized) (one sterile)
ten sterile)

Bx3 (c) 9 9 x Gmc e ra Bx3 (c) a a Bx3 (c/s) a a

(17 fertilized, (11 fertile, (three sterile,
two not fertilized) 13 sterile) two died young)

Bx4 (c) a a
(five fertile,
five sterile)

FIG. 2. Portion of pedigree Gmc/Gs that involved recurrent backcrossing to G. m. centralis. (Gmc: G. m. centralis; Gs: G. swynnertoni.)
(See text for which of these females were inseminated.)
 Gmm 9 9 G s o' c
(nine fertilized)

612
I------------------
F, 9 9 Gs o ' c
(five fertilized)
I________

Вх, (s) 9 9 x Gs cr cr Bx, (s)o’ o’ Bx, (s) or 13 Gmm 9 9

(eight fertilized, (15 sterile) (three fertilized)
one not fertilized)
I______

G O O D IN G
Bx2 (s) 9 9 x Gs crcr Bx2 (s) a о' Bx2 (m) 9 9 > Gs o'er Bx2 (m) cr cr
(12 fertilized, (12 sterile sterile) (four fertilized, (six sterile)
two not fertilized) one not fertilized)

Bx3 (s) 9 9 x Gs e ra Bx3 (s) or a Bx3 (m) 9 9 > Gs cr o' Bx3 (m) cr
(nine fertilized, (one fertile, (three fertilized, (one sterile)
one not fertilized) eight sterile) two not fertilized)
I________

Bx4 (s) cr cr
(three fertile,
four sterile)

FIG. 3. Portion of pedigree Gmm/Gs that involved recurrent backcrossing to G. swynnertoni. (Gmm: G. m. morsitans; Gs: G. swynnertoni.)
(See text for which of these females were inseminated.)
 IAEA-SM-327/57 613

T A B L E IV. INFLUENCE OF THE

ORIGIN OF THE SEX C H RO M OSO M ES
O N THE FER TILITY OF THE B A C K -
CROSS M A L E S 3

No. o f males
Originb of
X Y Fertile Sterile

5 S 3 • 16

m s 2 ' 10

с с 36 30

s s 3 7

c s 0 5

s c 0 10

a Those in the upper part o f the table are from the

pedigree Gmm/Gs and those in the lower part
are from the pedigree Gmc/Gs.
b The taxonomic origin o f X was based on the
electrophoresis o f PGM ; the origin o f Y was
based on the taxon o f the father.

3.3 .3. Pedigree Gm s/Gs

This pedigree began with seven G. m. submorsitans females fertilized

by G. swynnertoni, producing seven puparia. Five Fj males were sterile; a
sixth male died when less than one week old. The only Fj female was fertilized by
G. swynnertoni but produced only one sterile male.

4. DISCUSSION

The G. swynnertoni females used were those that emerged from field collected
puparia. Under laboratory conditions, they mated readily with conspecifics and with
the males o f all three subspecies o f G. morsitans. Similarly, G. swynnertoni males
attempted to mate with conspecifics and with the females from laboratory colonies
o f G. m. centralis, G. m. morsitans and G. m. submorsitans, but were frequently
rejected by G. m. centralis. The reluctance o f G. m. centralis females to copulate
614 GOODING

with G. swynnertoni was unéxpected, since it had previously been reported that
G. m. centralis and G. swynnertoni mated non-assortatively [9, 10]. The only reports
suggesting any difficulties in mating G. m. centralis females with G. swynnertoni
were the results o f Corson [6], who was able to get only 12 o f 18 G. m. centralis
females to mate with G. swynnertoni, and the observation by Potts [7] that “ ... the
inter-specific couplings were undertaken with greater reluctance than the intra-
specific ones ... ” . The present results indicated that the tendency o f females o f all
three subspecies o f G. morsitans to mate with G. swynnertoni is inversely related to
the geographical proximity o f the G. swynnertoni and G. morsitans subspecies. This
finding, although tentative, helps to clarify the ambiguous statement made by
Vanderplank [3] that G. swynnertoni and the subspecies o f G. morsitans “ ... will
intermate, though the degree to which they d o so varies with the geographical
proximity o f the species . . . ” .
It has been proposed [14] that asymmetries in mating behaviour indicate the
direction o f evolution, with the derived species having females that accept males
o f both derived and ancestral taxa, while females o f the ancestral taxon accept
conspecifics only. I f this hypothesis is applicable to tsetse, it would indicate that
G. swynnertoni is derived from G. m. centralis, a proposal that is consistent with
an earlier suggestion [5] that G. swynnertoni evolved from G. morsitans as a result
o f its ability to inhabit more open types o f bush than are occupied by G. morsitans.
In any case, the data suggested that premating barriers have begun to develop
between G. m. centralis and G. swynnertoni, but that there is little to prevent mating
o f G. swynnertoni to either G. m. morsitans or G. m. submorsitans. I f the mating
barriers between G. m. centralis females and G. swynnertoni males exist under field
conditions, they would impose limitations on the use o f G. swynnertoni males for
the control o f G. m. centralis.
With respect to the fertility o f interspecific crosses, the results o f the present
experiments were similar to most o f the results o f Vanderplank [3]. In the present
study, G. m. centralis, G. m. morsitans and G. m. submorsitans males fertilized 25,
0 and 0% o f G. swynnertoni; corresponding figures from the previous study were
6 and 0 % , with the third cross not reported. Similarly, in the present study
G. swynnertoni males fertilized 47% o f G. m. centralis, 68% o f G. m. morsitans
and 39% o f G. m. submorsitans; the corresponding figures from the previous study
were 24, 58 and 0% (the latter being based on two pairs).
The most striking effect o f the hybridization o f females o f G. morsitans was
the suppression o f their reproductive capacity. In our colonies o f G. morsitans,
females normally average six to eight puparia during the three months that they are
maintained. How ever, females that were hybridized with G. swynnertoni produced
an average o f 1.0 (G. m. submorsitans), 1.5 (G. m. centralis) and 1.8 puparia/female
(G. m. morsitans) during three months (Table II). A ll the females that were scored
as pregnant had produced their first offspring within 28 d o f mating, i.e. each had
produced a mature larva as a result o f the first or second ovulation. The low number
 IAEA-SM-327/57 615

o f offspring produced thereafter suggests that the females were sterilized as a result
o f their first pregnancy. Such an effect would have a profound effect on reducing
the size o f a population.
The ability o f G. m. centralis males to mate with G. swynnertoni and to ferti­
lize a small percentage o f these females indicates that these males could serve as
satyrs or as a means o f introducing hybrid male sterility into a G. swynnertoni popu­
lation. Either would have the effect o f reducing the reproductive capacity o f the
G. swynnertoni population. This was, in fact, accomplished by Vanderplank with the
marked reduction in or eradication o f a G. swynnertoni population in the United
Republic o f Tanzania [ 8, 10]. The results o f the present study do not shed light on
whether the success o f this control operation was solely the result o f hybrid sterility
(as claimed by Vanderplank [ 8, 10]) or the effect o f satyrism (as suggested by
Ribeiro [11]).
The ability o f G. m. morsitans and G. m. submorsitans males to inseminate,
but not fertilize, G. swynnertoni suggests that these males may be useful as satyrs
for the control o f G. swynnertoni. T o explore this possibility, it will be necessary
to determine whether polyandrous G. swynnertoni females, use the sperm o f
conspecifics only (as is the case with G. morsitans subspecies [12]) and to establish
whether interspecific mating can occur in nature.
A ll the Fj hybrid males were sterile and failed to inseminate their mates.
These results contrast with those o f Vanderplank [10], who found that about 80%
o f the Fj hybrid males from the G. swynnertoni/G. m. centralis crosses were able
to inseminate, but not fertilize, females. The discrepancy in our results may be due
to differences in the conditions under which we maintained the puparia and adults.
H ow ever, three backcross males were found (one ВхД с) male from pedigree
Gmc/Gs and two Bx 3(s) males from pedigree Gmm/Gs) that each inseminated the
first female with which it mated; however, none o f the females that were subse­
quently mated (six, six and four) became pregnant, and'only one o f these females
was inseminated.
M ost o f the F! hybrid females could be fertilized by backcrossing to either
parental taxon. The occcurrence o f sterile females among the first three backcross
generations was sporadic. Although the number o f flies in each pedigreee was low ,
the occurrence o f sterile backcross females was most com m on in the Gmm/Gs
pedigree, and in that portion o f the Gmc/Gs pedigree in which backcrossing
was switched from G. swynnertoni to G. m. centralis. I f genetic control o f a
G. m. centralis population were attempted by repetitive release o f G. swynnertoni
males, one would expect some o f the backcrosses to be o f this ‘ switch’ type and this
phenomenon, whatever its genetic basis, would contribute to a m ore rapid decline
in the population.
Sterile males predominated among the first three backcross generations and
made up approximately half the males o f the fourth backcross generation. A m ong
the backcross males, for which the origin o f the sex chrom osom es was determined
616 GOODING

(Table IV ), there were 42 fertile and 53 sterile males whose X and Y chromosomes
came from the same taxon, and two fertile and 25 sterile males whose X and
Y chromosomes came from two different taxa (x 2 (with Yates correction) = 10 801,
1 degree o f freedom, P < 0;005). Thus, a major cause o f hybrid male sterility is
an incompatibility o f sex chromosomes from two different taxa. This is consistent
with previous reports that X /Y incompatibility in tsetse is a major cause o f hybrid
male sterility [15]. How ever, it is clear that since 58% o f the males that had X and
Y chrom osom es from one taxon were sterile, there must be other factors that
influence hybrid sterility. Furthermore, the fertility o f hybrid males was not tested
with females from both parental taxa, and there may be maternally inherited sterility
factors that will cause an asymmetry in the reproductive success o f backcross males
[15]. Such factors could further influence the effectiveness o f backcross males in
reducing the reproductive potential o f a population.
The results reported here are preliminary and are based on two samples o f
G. swynnertoni taken from a population against which control operations had been
carried out, and were also dependent on the use o f inbred laboratory colonies o f
G. morsitans subspecies. Interpretation o f the results must be tempered also by the
fact that a relatively low percentage o f G. swynnertoni females were fertilized by
conspeciflc males.. The above notwithstanding, the results suggest that in certain
combinations there is a possibility for control o f populations o f G. swynnertoni and
o f the subspecies o f G. morsitans by the effects o f satyrism and by interspecific
hybridizations that might occur by introducing members o f one taxon into the habitat
o f another.

ACKNOW LEDGEM ENTS

The author wishes to thank D .F .P . Ringo, Director o f the Tropical Pesti­

cides Research Institute (TPRI), Arusha, United Republic o f Tanzania, for his c o ­
operation in facilitating this w ork, and S. M bise, P. Macha, R. D oriye, H. Matechi,
I. Ramadhani and other TPRI staff for collecting tsetse puparia. A lso thanked is
B .M . Rolseth, w ho assisted in both the field and laboratory aspects o f this study.
The work was financed, in part, by a Natural Sciences and Engineering Research
Council o f Canada grant (Grant N o. A -3900) and was performed as part o f a
Research Agreement (N o. 5309/C F ) with the IA E A . This paper is dedicated to
the memory o f André Van der Vloedt, a tireless student o f tsetse biology and a very
good friend.
 IAEA-SM-327/57 617

REFERENCES

[1] JORDAN, A .M ., “ Systematics” , Tsetse: The Future for Biological Methods in

Integrated Control (LAIRD, M ., Ed.), International Development Research Centre,
Technical Report No. 077C , International Development Research Centre, Ottawa
(1977) 13-22.
[2] G O U TEU X, J.P., Une nouvelle glossine du Congo: Glossina (Austenina) frezili sp.
nov. (Diptera: Glossinidae), Trop. Med. Parsitol. 38 (1987) 97 -1 00 .
[3] VAN D ER PLA N K , F .L ., Experiments in cross-breeding tsetse-flies ( Glossina species),
Ann. Trop. Med. Parasitol. 42 (1948) 131-152.
[4] GOODING, R .H ., et al., Genetic variation in Glossina brevipalpis, G. longipennis and
G. pallidipes, and the phenetic relationships of Glossina species, Med. Vet. Entomol.
5 (1991) 165-173.
[5] POTTS, W .H ., The distribution o f tsetse-flies in Tanganyika Territory, Bull. Entomol.
Res. 28 (1937) 129-148.
[6] CORSON, J.F., A note on tsetse flies, J. Trop. Med. Hyg. 25 (1932) 9 7 -9 8 .
[7] POTTS, W .H ., Tsetse hybrids, Nature (London) 154 (1944) 6 0 6-60 7.
[8] VAN D ERPLA N K, F .L ., Hybridization between Glossina species and suggested new
method for control o f certain species o f tsetse, Nature (London) 154 (1944) 60 7-60 8.
[9] JACKSON, C .H .N ., Pairing o f Glossina morsitans Westwood with G. swynnertoni
Austen (Diptera), Proc. R. Entomol. Soc., Ser. A 2 0 (1945) 106.
[10] VAN D ER PLA N K, F .L ., Experiments in the hybridisation o f tsetse-flies ( Glossina,
Diptera) and the possibility o f a new method o f control, Trans. R. Entomol. Soc.,
London 98 (1947) 1-18.
[11] RIBEIRO, J .M .C ., Can satyrs control pests and vectors? J. Med. Entomol. 25 (1988)
43 1-44 0.
[12] GOODING, R .H ., “ Genetic studies on Glossina morsitans and Glossina palpalis
related to genetic control” , Tsetse Control, Diagnosis and Chemotherapy Using
Nuclear Techniques (Proc. Seminar Muguga, Kenya, 1991), IAE A-T EC D O C -634,
IA E A , Vienna (1992) 151-165.
[13] GOODING, R .H ., et al., Electrophoretic comparisons of pheromotypes o f the dingy
cutworm, Feltia jaculífera (Gn.) (Lepidoptera: Noctuidae), Can. J. Zool. 70 (1992)
79 -8 6.
[14] KANESHIRO, K .Y ., Sexual selection and direction o f evolution in the biosystematics
o f Hawaiian Drosophilidae, Annu. Rev. Entomol. 28 (1983) 161-178.
[15] GOODING, R .H ., Postmating barriers to gene flow among species and subspecies o f
tsetse flies (Diptera: Glossinidae), Can. J. Zool. 68 (1990) 1727-1734.
 IAEA-SM-327/77

THE VARIOUS VECTOR CONTROL ACTIVITIES

OF THE CENTRE DE RECHERCHES
SUR LES TRYPANOSOMES ANIMALES (CRTA)

B. B AU E R , S. A M SLE R
Centre de recherches sur les trypanosomes animales (C R T A ),
Bobo-Dioulasso,
Burkina Faso

Abstract

THE VARIOUS VECTOR CONTROL ACTIVITIES OF THE CENTRE DE

RECHERCHES SUR LES TRYPANOSOM ES AN IM ALES (CRTA).
The role o f the Centre de recherches sur les trypanosomomes animales (CRTA) in anti­
tsetse control will become more important because the centre will soon become an institution
with a regional mandate. In the laboratory, studies are being carried out on olfactory attrac­
tants to riverine species, for instance from reptiles. The persistence and efficiency of
pyreth'roids are being tested, after impregnation on the tissues used for traps and targets, and
after application on cattle. The rearing o f ticks is also beginning and CRTA is continuing to
rear different species o f tsetse fly that can be used by other countries or organizations for their
sterile insect technique programmes. In the field, many experiments are being carried out to
improve traps and targets (shapes, colours, attractants, etc.) and entomological surveys are
being carried out regularly in co-ordination with other programmes.

1. IN T R O D U C TIO N ,

The Centre de recherches sur les trypanosomes animales (C R T A ) is in a transi­

tional stage. Since it is financed by the European Econom ic Community, France and
Italy, it is soon to becom e an institution with a regional mandate that covers, apart
from vector control, various other aspects, such as the epidem iology and econom ics
o f parasitic diseases and their control, animal production and husbandry systems,
including characterization o f trypanotolerant cattle, resistance o f crossbreeds, train­
ing o f national staff at different levels, and extension activities in the field.
This transformation has com e about at the request o f the five M ember States
o f the Communauté économ ique du bétail et de la viande (C E B V ), namely Benin,
Burkina Faso, Côte d ’ ivoire, Niger and T ogo. The name o f the new centre will be
the Centre international de recherche-développement sur l ’élevage en zone sub­
humide (CIRDES).

619
620 BAUER and AMSLER

2. RESEARCH A CTIVITIES OF THE V E C T O R C O N T R O L UNIT

IN THE L A B O R A T O R Y

2.1. Isolation and identification of new odours

None o f the substances attractive for savannah species such as Glossina morsi­
tans or G. pallidipes (l-o cte n -3 -o l, acetone or the phenolic fractions o f urine) has
proved to be effective against G. p. gambiensis. Therefore, part o f our research is
oriented towards other host animals o f this tsetse species. Bearing in mind that more
than 50% o f the feed o f G. p. gambiensis consists o f bloodmeals taken from reptiles,
it is worthwhile having a closer look at the mechanisms that are responsible for the
attractiveness o f these hosts to this tsetse. In fact, monitor lizards ( Varanus niloti-
cus), crocodiles ( Crocodylus niloticus and Osteolaemus tetraspis) and snakes
(Python sebae) are being evaluated for their attraction to G. p. gambiensis and
G. tachinoides. Large scale bioassays with flies o f known age — C R T A still disposes
o f about 150 000 producing females o f G. p. gambiensis, G. tachinoides and
G. morsitans submorsitans — allow rapid appraisal o f the attraction o f the various
reptiles. The preferred feeding sites on the reptiles are also being recorded. Promis­
ing compounds or mixtures are then assayed in a field station to evaluate their ability
to influence the effectiveness o f a standard capture technique.

2.2. Effectiveness and persistence of pyrethroids

Servicing o f the targets or traps accounts for approximately 30% o f the total
costs o f a tsetse campaign based on the use o f such devices. The incentive is to reduce
the number o f visits required per year for servicing. The possible ways o f achieving
this are: identification o f pyrethroids with maximum persistence; increase in the
amount o f insecticide used for treatment and selection o f the most suitable tissue —
in general, synthesis tissues have proved to be more suitable than natural materials,
e.g . synthetic cotton. After impregnation, the tissues are constantly exposed to sun­
light and rain. Bioassays are conducted at regular intervals with defined numbers o f
flies o f known age. So far, results have indicated that ‘ washing o f f , i.e. a decrease
in the effectiveness o f an insecticide due to rainfall, is less deleterious than the effects
o f sunlight itself. The maximum persistence o f certain deltamethrin formulations was
found to exceed nine months, approaching a situation where the effectiveness o f a
pyrethroid would last as long as the tissues themselves, hence making servicing
unnecessary.

2.3. Treatment of cattle with acaricides/insecticides

Pyrethroids, originally conceived to combat ticks, have shown remarkable

potential against tsetse, leading to a situation where the strategic deployment o f
 IAEA-SM-327/77 621

targets/traps and the treatment o f cattle constitute the main pillars in rural areas for
tsetse control. Currently, large scale bioassays are being conducted to evaluate all
the comm ercially available pyrethroids. Flies o f known age and reproductive state
are released in the presence o f a treated zebu and the follow ing parameters are
recorded: the persistence o f a given product, the mortality rates, the knock down
effects and their duration, the percentage o f flies that failed to feed, as well as the
effects on reproductive performance. Every second day, the treated animal is
exposed to sun for about 3 h and thereafter rinsed with 50 L o f water in order to
provide a realistic view o f the respective efficacies. O f all the products tested so far,
none has shown any decrease in persistence when being rinsed with water even 2 h
after treatment and thereafter every second day. On the other hand, sunlight has a
deleterious effect on certain formulations. Some o f the products show a distinct per­
sistence, exceeding 90 d. A manual is in preparation that will summarize all the
results and provide technical advice on the rational and strategic use o f acari-
cides/insecticides for the simultaneous control o f ticks and tsetse.

2.4. Rearing of tick species

Rearing o f Amblyomma variegatum and Boophilus geigyi has started on live

hosts. Priority is being given to monitoring and assessing the development o f
resistance to acaricides/insecticides. Indiscriminate and incoherent use o f insecti­
cides during the rainy season is likely to enhance the probability o f a development
towards resistance. Although chlorinated hydrocarbons (D D T , lindane, etc.) have
basically been used up to now , resistance against pyrethroids may build up as well.
The incidence o f tickborne diseases and their econom ic impact will be evaluated in
the field.

2.5. Improvements in the transport of sterile insects

The sterile insect technique has proved its efficacy in large scale eradication
campaigns in the pastoral zones o f Sideradougou, Burkina Faso, and V om , Plateau
State, Nigeria. C R T A /C IR D E S , with its regional mandate, is prepared to deliver
sterilized tsetse flies to any project in the region i f required. Until recently, the fragil­
ity o f the biological material has been a serious obstacle to the supply o f tsetse over
long distances. In general, either late stage puparia or young adult flies have been
transported. There are now . signs that storage o f puparia at about 15°C does not
adversely affect the emergence rate or the subsequent viability o f the newly emerged
flies. Therefore, supply o f puparia over longer distances and taking several days may
becom e feasible.
622 BAUER and AMSLER

3. IM PRO VEM EN T IN THE EFFECTIVENESS

OF TARGETS A N D TRAPS

The shape and colour o f the devices, apart from olfactory compounds, are
requiring further w ork to define the optimum system. Further, it should be borne
in mind that the system should be affordable and sustainable for the rural communi­
ties concerned. Comparisons between different trapping systems are being conducted
in a field station in the presence o f the tsetse species mentioned in Section 2.

4. LA R G E SCALE TSETSE C O N TR O L

Large scale tsetse control is being carried out by integrating the periodic treat­
ment o f livestock with a pyrethroid in combination with the use o f traps and targets
in habitats inaccessible for treated cattle.
An immediate improvement in the overall health o f the herd after a few acarici-
dal treatments is rapidly being perceived by the livestock owner, thus facilitating his
further collaboration and his taking over o f the costs o f the treatments. The prerequi­
sites o f any intervention are thorough entomological and epidem iological surveys
providing justification for interventions. O f increasing importance are socioeco­
nomic analyses o f the various situations. The main aspects should relate to a
benefit/cost analysis o f the control in areas o f agropastoral land use and a compara­
tive cost analysis o f the different methods o f control. The incidence o f African
animal trypanosomosis (A A T ) is regularly monitored in sentinel herds. Consecutive
entomological surveys, and dissection o f captured flies to determine their physiologi­
cal age and infection rate permit constant appraisal o f the effects o f the tsetse cam­
paign. Ideally, this should allow assessment o f trypanosomosis challenge.
However, in certain areas the results obtained so far indicate that incongruities
exist between the incidence o f A A T and tsetse challenge. M ore investigations are
required to elucidate the dynamics o f the tsetse-trypanosom e-livestock relationship.
The vectorial capacity o f the different tsetse species must be better understood.
Apparently, this capacity is low in G. p. gambiensis in an area (Samorogouan)
where a tsetse campaign started about tw o years ago based on the treatment o f the
cattle population. Apparently, there exists a relic population o f this species that does
not seem to be affected by the tsetse campaign, presumably because it is partly feed­
ing on reptiles. Can we afford to neglect this population because o f its apparent low
vectorial capacity and its present behaviour? D o we have to deal with other tsetse
species (G. m. submorsitans were abundant before the campaign) that have a density
below a detectable level and which could be responsible for an incidence o f 2 0 -2 5 %
A A T in the sentinel herds?
 IAEA-SM-327/77 623

It is evident that the actual methodology needs to be adjusted to fit the existing
situation. Additional research is required, for instance, to identify bloodm eals or
D N A probes for detecting infections in tsetse flies. Tracking o f low density tsetse
flies demands an improvement in the trapping devices; identification o f new olfac­
tory compounds could further enhance the response o f the flies, thereby improving
the trapping systems.
QUARANTINE
(Session 8)

Chairm an

D .A . L IN D Q U IS T
F A O /IA E A
 IAEA-SM-327/58

IRRADIATION AS A QUARANTINE TREATMENT

OF AGRICULTURAL COMMODITIES
AGAINST ARTHROPOD PESTS

N .W . HEATH ER
Entomology Branch,
Department o f Primary Industries,
Indooroopilly, Queensland,
Australia

Abstract

IRRADIATION AS A Q UAR ANTINE TREATM EN T OF AG RICU LTURAL C O M M O DI­

TIES AG AIN ST ARTHROPOD PESTS.
The purpose o f quarantine treatments is to eliminate, as far as possible, the risks of
introduction or establishment o f exotic pests in countries or regions where they do not already
occur. Treatments may be applied to host commodities traded commercially or carried by
travellers. Ionizing radiation is very effective when used as a quarantine treatment to disinfest
fresh, dried or processed fruits, grains and other plant materials. It is highly effective in killing
or inactivating arthropod pests, leaves no residues, and at the low doses required it can be
used on most commodities without affecting the quality. The most important single pest group
o f quarantine importance internationally is arguably fruit flies in fresh fruits and vegetables.
More than thirty species o f fruit flies are recognized as serious quarantine pests. D ose-
mortality studies with irradiation have shown that doses o f 7 5 -1 5 0 Gy prevent adult emer­
gence. At two Task Force Meetings on Irradiation as a Quarantine Treatment convened by
the International Consultative Group on Food Irradiation, a generic dose o f 150 Gy was
recommended against any fruit, based on extensive research data for these pests. Most fruits
are relatively unaffected by quarantine disinfestation treatments o f 100-300 Gy but some, for
example, avocado, appear to be intolerant. Other pests o f quarantine importance for which
irradiation is an appropriate disinfestation treatment include certain moths (Lepidoptera),
beetles (Coleoptera), bugs (Homoptera), flies (Diptera), thrips (Thysanoptera) and mites
(Acariña). The Task Force Group also recommended that a generic treatment o f 300 Gy,
based on the inability to perpetuate the species, would be appropriate for any pest other than
fruit fly. This was derived from extensive research on the codling moth, Cydia pomonella
(L .), and the mango seed weevil, Stemochaetus mangiferae (Fabricius), with supporting
results on eleven other pests from six orders.

1. IN TROD U CTION

The purpose o f quarantine treatments against arthropod pests is to reduce, as

far as is practicable, the risk o f introduction o f species into areas where they d o not
already occur. Treatments are conventionally applied to bulk produce traded

627
628 HEATHER

com m ercially, but they are also applicable to items carried by travellers. Used as a
quarantine disinfestation treatment, irradiation is highly efficacious and has other
advantages for fruits, grains and agricultural produce which might be infested with
unwanted pests.
Fruit flies, family Tephritidae, in fresh fruits or vegetables are arguably the
most important quarantine pest o f agricultural commodities [1]. The orders Acariña,
Thysanoptera, Coleoptera, Lepidoptera and Diptera contain further pests o f quaran­
tine importance [2 ] against which irradiation can constitute a useful, and sometimes
unique, method o f disinfestation.
The potential o f irradiation to disinfest agricultural and horticultural produce
has been recognized since early in this century, although most o f the research has
been done in the latter half. The availability o f gamma radiation sources, particularly
60C o, has given rise to many dose-m ortality studies on insect pests and on the
effects o f irradiation on the organoleptic qualities o f fruit and other commodities.
A major conference on the radiation disinfestation o f food and agricultural
products [3], held in Hawaii in 1983, focused attention on irradiation as a quarantine
disinfestation treatment. This was follow ed in 1986 and 1991 by Task Force
meetings at Chiang M ai, Thailand, and Bethesda, Maryland, United States o f
America, respectively, convened by the F A O /IA E A sponsored International Consul­
tative Group on Food Irradiation (ICGFI) to review irradiation as a quarantine treat­
ment. Reports o f these Task Force meetings [4, 5] identified the pests and host
commodities for which irradiation was an effective method o f disinfestation. They
recommended generic and specific dosages o f irradiation which would ensure
quarantine security based upon the appropriate criteria for effectiveness.
A major factor motivating interest in irradiation for quarantine purposes was
the cancellation o f registration o f the fumigant ethylene dibromide (EDB) in the
USA [6]. This fumigant was very effective against fruit flies, econom ical to purchase
and apply, and logistiçally very flexible, requiring only a gas tight room or tent.
W hile it remained available, alternative methods, including irradiation, were greatly
disadvantaged. The loss o f EDB highlighted the risk to any treatment giving rise to
residues and led to heightened interest in the physical residue free treatments, cold ,
heat and irradiation.
A prerequisite to the use o f irradiation as a quarantine treatment for agricul­
tural commodities was approval o f its use on foodstuffs. The joint F A O /W H O C odex
Alimentarius Commission, General Standard for Irradiated Foods [7], recommended
approval o f doses up to 10 kGy as safe for use on any food . Subsequently, the United
States Food and Drug Administration [ 8] approved doses o f up to 1 kGy for dis­
infestation o f fresh fruits and vegetables. Many other countries follow ed suit and it
is estimated that the process has been approved in a majority o f the world based on
population. Finally, in 1989 the United States Department o f Agriculture’ s Animal
and Plant Health Inspection Service amended its quarantine regulations to permit use
o f irradiation as a quarantine treatment [9].
 IAEA-SM-327/58 629

In this paper, a review is made o f the background regulatory requirements

which all quarantine disinfestation treatments must meet, together with research
demonstrating efficacy against quarantine pests, specific considerations which apply
to irradiation, and monitoring o f the dosage required to kill arthropod pests while
avoiding any damage to the comm odity.

2. FRUIT FLIES

The ultimate requirement o f a quarantine treatment is to prevent the establish­

ment o f pests in new areas. This can be accomplished by killing the insect at the time
o f treatment or shortly afterwards, preventing its development to a reproductive
stage and preventing reproduction through physical deformity or infertility.
For fruit flies, the Chiang Mai Task Force Group [4] recommended that the
criteria for acceptance o f a proposed quarantine treatment could be based on the
dosage to prevent larval survival, pupation, emergence o f adults capable o f flight or
o f adults capable o f reproduction. It was recognized that this could result in the
presence o f properly treated, but still living, insects in fruit at a post-treatment
inspection [10]. Although not compromising quarantine security, they would present
a problem for inspection authorities who would be uncertain o f whether the treatment
had been applied properly.
The problem has almost certainly been overestimated for fruit flies. The levels
o f infestation present in export quality comm ercial fruit need to be very low , regard­
less o f quarantine requirements. M ost markets have a nil tolerance for this type o f
pest based on quality, and consignments found to contain infested units could be
expected to be rejected on this basis. This raises the principle that quarantine disin­
festation treatments should not be used as a part o f pest management needed to
produce a quality product. For insect pests which do not affect the fruit quality, the
problem is more valid. In a subsequent section, the means o f identifying insects
which have received a lethal reproduction inhibiting dose o f irradiation, but are still
alive at the time o f inspection, are discussed.
The efficacy required o f a quarantine treatment is usually defined in terms o f
the. mortality induced in the pest. Other effects, including prevention o f reproduction
through physical defects or sterility, are more difficult to measure but can be equiva­
lent. Mortality is usually quoted as a percentage or a probit value [10], e.g.
99.9968% is probit 9, and should have a statistical confidence level, usually 95%
[11]. These give rise to the levels o f testing required by quarantine authorities to
demonstrate treatments experimentally, e.g. to demonstrate probit 9 at the 95% con ­
fidence level requires tests against 93 616 pests without survivors [12].
Although some o f the variation could be attributed to differences among
species, differences in age and stage as well as differences in experimental technique
could account for most o f the variation exhibited. Because o f the nature o f gamma
630 HEATHER

TA B L E I. EFFECTS OF EXPOSU RE OF FRUIT F L Y L A R V A E TO

G A M M A R A D IA T IO N O N PREVENTION OF SUBSEQUENT A D U L T
EM ERGENCE [5, 17-30]

Dose for
Species Host Refs
no emergence (Gy)

Anastrepha ludens (Loew) Grapefruit 50 [17]

(Mexican fruit fly) Mango 100 [18]

A. obliqua (Macquart) Mango 100 [18]

(West Indian fruit fly)

A. serpentina (Wiedemann) Mango 100 [18]

(Sapodilla fruit fly)

A. suspensa (Loew) Grapefruit 25 [19]

(Caribbean fruit fly) Grapefruit 154 [20]
Mango (Florida, USA) 17)5 [21]
Mango (Dominican Republic) 25 [21]
Mango (Haiti) 80 [21]
Carambola 50 [22]
Grapefruit . 40 [23]
Ceratitis capitata (Wiedemann) Papaya 25 -7 5 [24]
(Mediterranean fruit fly) Mango , 150 [18]
Bactrocera cucurbitae (Coquillett) Pumpkin 150 [23]
(Melon fly) Mixed 100 [24]
B. dorsalis (Hendel) Avocado 100 [24]
(Oriental fruit fly) Mixed 100 [24]
Mango 150 [25]
Mango 100 [26]
Mango 100 [26]
Mango 150 [27]
B. jarvisi (Tryon) Mango 75-101 [15]
(Jarvis’ fruit fly)
B. tryoni (Froggatt) Mango 75-101 [15]
(Queensland fruit fly) Apple 75 [28]
Tomato 50 [28]
Cherry 75 [28]
Avocado 75 [28]
Orange 75 [28]
B. zonaias (Saunders) Guava 50 [29]
(Guava fruit fly) 50 [29]
55 [29]
Rhagoletis indiffererts (Curran) Cherry 97 [30]
(Western Cherry fruit fly)
 IAEA-SM-327/58 631

radiation, it is virtually impossible to administer a uniform dose to any experimental

sample o f significant volume. A lso, use o f statistical regression analyses to predict
the dose required for high efficacies such as probit 9 can lead to overestimation. The
efficacy targeted and the criteria for effectiveness have a bearing on the results.
In general, exposure o f eggs or larvae in fruit to a dose o f 150 G y or less
prevented emergence o f adults at an efficacy o f probit 9 [1]. This was adopted as
the generic dose applicable to all fruit flies o f the family Tephritidae. M uch greater
doses would be required to cause probit 9 mortality o f the stage treated, increasing
the likelihood o f damage to the fruit [13]. W here a low er dose has been shown to
achieve probit 9 efficacy, it would be used; for example, for the Queensland fruit
fly, Bactrocera tryoni (Froggatt), a dose o f less than 100 G y, possibly 75 G y, has
been shown to be effective [14, 15].
There are sound statistical data for 11 fruit fly species o f major quarantine
importance showing the dose to prevent adult emergence [16]. These are listed in
Table I [5, 1 7-30], together with the fruit on which the efficacy was demonstrated.
The fly species are from five genera with geographical distribution in Australia,
Asia, North Am erica, South Am erica and Europe. These results provide strong
assurance o f the efficacy o f the generic dose against any fruit fly o f the family
Tephritidae.

3. OTH E R INSECTS

The 1986 Chiang M ai ICGFI Task Force Meeting on Irradiation as a Quaran­

tine Treatment [4] recommended á generic dose o f 300 Gy for insects other than fruit
fly. On available evidence, this dose was judged to provide quarantine security
against any stage o f any insect pest using the criteria o f prevention o f ‘ ‘emergence
o f normal adults from treated eggs, larvae or pupae or to sterilise any adults
emerging from treated larvae or pupae” . A review presented at the 1991 Bethesda
Task Force Meeting [31] identified insects from five orders and mites from two
groups against which a disinfestation treatment with irradiation had been shown to
provide quarantine security at doses o f 300 Gy or less. These included three species
o f Lepidoptera, two o f Coleoptera and two o f Acariña. Effective quarantine doses
ranged from 80 to 300 Gy (Table П) [5, 15, 3 2 -4 2 ].
With the exception o f the codling moth, Cydia pomonella (Linnaeus), and the
mango seed weevil, Stemochaetus mangiferae (Fabricius), for which comprehensive
mortality data were available [33, 38], these pests were not shown to have been
killed to very high levels o f treatment efficacy. This was due to the scope o f the
experiments and does not reflect on the treatment. It is a measure either o f the lesser
quarantine importance o f most o f these pests or o f the early stage o f research.
H ow ever, all o f the results supported the adequacy o f 300 Gy as a generic treatment.
632 HEATHER

TAB LE II. EFFECTS OF EXPOSU RE OF A D U L T S A N D JUVENILES

OF INSECTS A N D A LLIE D FORM S (O TH ER TH A N FRUIT FLIES) TO
G A M M A IR RA D IA TIO N O N PREVENTION OF SUBSEQUENT
A D U L T EM ERGENCE O R REPRO DU CTIO N [5, 15, 3 2 -42]

Effective
Pest species Host Refs
dose (Gy)

Lepidoptera

Clepsis spectrana (Treitschke) Rose 200 [32]

(rose leaf roller)

Cydia pomonella (Linnaeus) Apple, walnuts 139-177 [33, 34]

(codling moth)

Epiphyas postvittana (Walker) Apple 199 [35, 36]

(lightbrown apple moth)

Coleoptera

Asynonchus cervinus (Boheman) Citrus 150 [37]

Stemochaetus mangiferae (Fabricius) Mango 300 [15, 38]

(mango seed weevil)

H e m ip te ra -H o m o p te ra

Brachycorynella asparagi (Mordvilko) Asparagus 100 [39]

Quadraspidiotus pemiciosus (Comstock) Apple 300 [40]

(San José scale)

Diptera

Liriomyza trifolii (Burgess) Chiysanthemum, 80 [32]

(serpentine leaf miner) tomato

Thysanoptera

Frankliniella pallidus (Steele) Flowers 100 [32]

Acariña

Brevipalpis destructor (Baker) Grapes 300 [41]

Tetranychus urticae (Koch) Fruits, flowers 300 [42]
 IAEA-SM-327/58 633

Mites responded in a similar fashion to insects. From the responses o f mites

as a group to irradiation it can be expected that 300 Gy will prove adequate. In addi­
tion to the mites given in Table II, a comprehensive study on the mould mite,
Tyrophagus putrescentiae (Schrank), and the bulb mite, Rhyizoglyphus echinopus
(Fumouze and Robin) [42], showed that mortality or adult sterility resulted from
doses o f 260 G y. The only exceptional response o f mites came from a preliminary
study on cut flowers by W it and de Vrie [32]. It can therefore be expected that the
physiological response o f mites to irradiation will be similar to that o f insects.
Whereas quarantine restrictions on the entry o f fruit fly host material almost
invariably invoke treatments o f the order o f probit 8.7 (9 9 .9 9 % ) or probit 9
(9 9.9 968% ), the efficacy required for other pests varies. A n example o f a pest for
which a low er efficacy would be satisfactory is against the eggs o f the Fuller rose
beetle, Asynonchus cervinus (Boheman), on citrus fruit, where the requirement is
that there are no live eggs on fruit at inspection. The condition can be met by a treat­
ment o f considerably low er efficacy than probit 9. A similar situation can apply to
thrips and mites on cut flowers.
Irradiation is also highly appropriate for the disinfestation o f stored products,
although the justification for quarantine treatments against such cosmopolitan insects
is questionable. Legume weevils (bruchids), Callosobruchus spp. and others, can be
disinfested at doses as low as 70 Gy [43]. Cereal grains can be disinfested o f the
com m on grain beetles Sitophilus, Rhyzopertha, Tribolium and other species at doses
below 200 G y, but information on the moths Sitotroga, Ephestia, Plodia and other
species is less precise. The dosages reported often appear to have been chosen
arbitrarily and have possibly been influenced by operational factors related to
the irradiation source. An important aspect o f stored grain pests is that they can be
irradiated with an electron beam source directed at a grain flow [44].

4. M O D E O F A C T IO N OF IR R A D IA T IO N

A ll insect pests are susceptible to control by irradiation. The tissues most sensi­
tive to irradiation are those containing a high proportion o f dividing cells. During
the immature development o f insects, there are periods within each stage when cell
division is highly active. This could explain the difference in response o f insect
developmental stages with age [45]. In adults, the two most important sites o f cell
division appear to be the gonads and gut cells. Disruption o f the form er can cause
temporary or permanent sterility, and o f the latter, death by starvation. A t higher
doses, other tissues may also be damaged, leading to more rapid mortality [46].
Particular sites o f damage include the enzyme secretion systems, epidermal lipid
secretion systems and the nervous system, especially the supraoesophageal ganglion,
which exercises hormonal control over development [47, 48].
634 HEATHER

5. EFFECTS O N COM M OD ITIES

Because quarantine disinfestation treatments against arthropod pests involve

very low doses o f irradiation, most agricultural commodities are essentially
unaffected. The quality attributes o f fruit, food and other agricultural products were
reviewed in depth at the 1983 conference on radiation disinfestation ,o f food and
agricultural products held in Hawaii [3]. Subsequent trials in Queensland [13, 49,
50] showed that lychee, mangoes (Kensington), capsicum, broccoli, chínese
cabbage, melons, peaches, nectarines, passion fruit and tomatoes withstood a
(generic) 300 Gy treatment without ill effects. Banana, carambola, navel orange,
rambutan, persimmon, papaw, zucchini and rockmelons were relatively unaffected
at a low er (fruit fly) dose o f 75 Gy. Fruit which did not tolerate disinfestation were
custard apple and avocado (except, possibly, cv Hass). Strawberries are well known
for their high tolerance to irradiation (2 kG y), although this far exceeds any require^
ment for quarantine disinfestation. These trials assessed fruit on the basis o f external
colour, firmness, injury, internal colour, soluble solids, pH, acidity, vitamin С ,
sugars, acids, eating quality and disease.
Cereal grains are another major agricultural comm odity which can be irradi­
ated for quarantine purposes. Adverse effects on baking properties have been
reported, but at insect or mite disinfestation doses o f 200 Gy it is doubtful if irradia­
tion o f grains would have significant effects [51]. Spices are frequently irradiated at
high doses to reduce their bacterial count, but this is done for health, not quarantine,
purposes.

6. O PE RA TIO N AL CON SIDERATIONS

6.1. Doses

Although the minimum doses for quarantine security may be very low , in prac­
tice a much higher dose may have to be given to ensure that all o f the product being
irradiated receives at least the minimum dose. Factors o f 2:1 are com m on where
boxes o f fruit are to be irradiated; 3:1 can be needed. Careful attention to irradiation
geometry is therefore required where the product to be irradiated is susceptible to
damage. Fruits range in tolerance but almost all can be damaged at or above twice
the generic dose o f 300 Gy [13, 49, 50]. W here this is a problem , consideration can
be given to combination treatments such as cold storage and irradiation [52].

6.2. Markers

Physical irradiation markers are a topic outside the scope o f this paper.
H ow ever, it would be o f advantage to quarantine authorities if it could be determined
 IAEA-SM-327/58 635

for insects found alive at an inspection follow ing irradiation whether they had been
irradiated. Studies on the Mediterranean fruit fly [48] and the Queensland fruit fly
[52] have shown that the supraoesophageal ganglion fails to develop normally in
irradiated larvae. Development can be expressed as a ratio o f the size o f the ganglion
compared with a structure such as the crop. Another difference reported was the
failure o f ovaries and testes to develop and the loss o f yellow colouration, e .g . o f
humeral calli, in flies emerging from the pupae o f Queensland fruit flies irradiated
at quarantine doses [53].
Biochemical diagnosis is also possible. A study using the technique o f poly­
acrylamide gel electrophoresis on irradiated larvae o f the Oriental fruit fly,
Bactrocera dorsalis (Hendel), has shown reliable detectable differences in the
protein band profiles [54]. This technique holds considerable promise as a relatively
rapid diagnostic method.

6.3. Security after treatment

Irradiation does not confer any residual protection to disinfested produce, so

measures must be implemented to ensure quarantine security after treatment. Where
the product can be irradiated in insect secure packaging, e.g. fruit cartons, the only
security measure necessary is separation o f the unirradiated and irradiated stock. If
the product is irradiated as individual pieces, e .g . fruit, then the whole facility must
be made insect secure after treatment. Regulatory authorities will probably be
involved in certification o f the product.

6.4. Source options

Ionizing energy can be applied as gamma radiation, X radiàtion or as electron

beams. Commercial facilities using all o f these methods are now in operation. As
far as the effects on arthropods are concerned, there should be no difference in effect
between gamma and X rays. The choice between gamma rays ( 60C o) and X rays
(electric machines) becom es a: comm ercial decision. H ow ever, for electron beam
radiation, penetration is much less than either gamma or X rays and there is no
certainty that the doses needed would be exactly the same. M odern linear accelera­
tors possess the capability o f operating in either the electron beam or X ray mode.
The choice o f the type o f irradiation can then depend on the product to be irradiated.

REFERENCES

[1] BURDITT, A.K., Jr., Effectiveness of irradiation as a quarantine treatment against

various fruit flies, FAO Plant Protection Bull. 39 (1991) 102-103.
[2] HEATHER, N.W., Irradiation as a quarantine treatment for insects other than fruit
flies, FAO Plant Protection Bull. 39 (1991) 103-105.
6 36 HEATHER

[3] MOY, J.H., Ed., Radiation Disinfestation of Food and Agricultural Products (Proc.
Conf. Honolulu, 1983), University of Hawaii at Manoa, Honolulu (1985).
[4] INTERNATIONAL ATOMIC ENERGY AGENCY, Irradiation as a Quarantine Treat­
ment, International Consultative Group on Food Irradiation, Task Force Meeting,
Chiang Mai, IAEA, Vienna, 1986 (internal report).
[5] INTERNATIONAL ATOMIC ENERGY AGENCY, Irradiation as a Quarantine Treat­
ment, International Consultative Group on Food Irradiation, Task Force Meeting,
Bethesda, MD, IAEA, Vienna, 1991 (internal report).
[6 ] UNITED STATES ENVIRONMENTAL PROTECTION AGENCY, Ethylene
dibromide, Fed. Regist. 48 (1983) 46234-46243.
[7] CODEX ALIMENTARIUS COMMISSION, Codex General Standard for Irradiated
Foods and Recommended International Code of Practice for the Operation of Radiation
Facilities Used for the Treatment of Foods, Vol XV, FAO, Rome (1984).
[8 ] UNITED STATES FOOD AND DRUG ADMINISTRATION, Irradiation in the
production, processing and handling of food: Final rule, Fed. Regist. 51 (1986)
13376-13399.
[9] UNITED STATES DEPARTMENT OF AGRICULTURE, Use of irradiation as a
quarantine treatment for fresh fruit of papaya from Hawaii, Fed. Regist. 54 (1989)
387-393.
[10] KANESHIRO, K.Y., etal., Gamma radiation treatment for disinfestation of the
Mediterranean fruit fly in California grown fruit. I. Stone fruits, Proc. Hawaiian
Entomol. Soc. 24 (1983) 245-259.
[11] FINNEY, D.J., Probit Analysis, 3rd edn, Cambridge University Press, London (1971)
333 pp.
[12] CHEW, V., OUYE, M .T., “ Statistical basis for quarantine treatment schedule and
security” , Radiation Disinfestation of Food and Agricultural Products (Proc. Conf.
Honolulu, 1983) (MOY, J.H., Ed.), University of Hawaii at Manoa, Honolulu (1985)
70-74.
[13] McLAUCHLAN, R.L., etal., Ionising Energy Treatment for Fresh Horticultural
Produce, Lychee and Mango, Trial Ш , Jan.-Feb. 1988, Department of Primary Indus­
tries Project Report, Brisbane, Queensland (1990).
[14] RIGNEY, C.J., WILLS, P.A. “ Efficacy of gamma irradiation as a quarantine treat­
ment against Queensland fruit fly ” , Radiation Disinfestation of Food and Agricultural
Products (Proc. Conf. Honolulu, 1983) (MOY, J.H., Ed.), University of Hawaii at
Manoa, Honolulu (1985) 116-120.
[15] HEATHER, N.W., CORCORAN, R.J., BANOS, C., Disinfestation of mangoes with
gamma irradiation against two Australian fruit flies (Diptera: Tephritidae), J. Econ.
Entomol. 84 (1991) 1304-1307.
[16] BURDITT, A.K., Jr., “ Effectiveness of irradiation as a quarantine treatment against
various fruit fly species” , Irradiation as a Quarantine Treatment, International Consul­
tative Group on Food Irradiation, Task Force Meeting, Bethesda, MD, IAEA, Vienna,
1991 (internal report).
[17] BROWNELL, L.E., YUDELOVIÇH, М ., “ Effect of radiation on Mexican fruit fly
eggs and larvae in grapefruit” , Radioisotopes and Radiation in Entomology (Proc.
Symp. Bombay, 1960), IAEA, Vienna (1962) 193-202.
 IAEA-SM-327/58 63 7

[18] BUSTOS, M .E., ENKERLEIN W ., TOLEDO, J., REYES, J., CASIMIRO, A.,
“ Irradiation of mangoes as a quarantine treatment” , Use of Irradiation as a Quarantine
Treatment o f Food and Agricultural Commodities (Proc. Research Co-ordination Mtg
Kuala Lumpur, 1990), IAEA, Vienna (1992) 77-90.
[19] BURDITT, A.K., Jr., et al., Low Dose Irradiation as a Treatment for Grapefruit and
Mangoes Infested with Caribbean Fruit Fly Larvae, Agricultural Research Results,
Rep. ARR-S-10, Agricultural Research Service, United States Department of Agri­
culture, New Orleans, LA (1991) 1-9.
[20] VON WINDEGUTH, D .L., Effects of gamma irradiation on the mortality of the
Caribbean fruit fly in grapefruit, Proc. Fla. State Hortic. Soc. 95 (1982) 235-237.
[21] VON WINDEGUTH, D .L., Gamma irradiation as a quarantine treatment for
Caribbean fruit fly infested mangoes, Proc. Fla. State Hortic. Soc. 99 (1986) 131-134.
[22] VON WINDEGUTH, D .L ., GOULD, W.'P., Gamma irradiation as a quarantine treat­
ment for Florida grapefruit infested with Caribbean fruit flies, Fla. Entomol. 73 (1989).
[23] VON WINDEGUTH, D .L., GOULD, W.P., Gamma irradiation followed by cold
storage as a quarantine treatment for Florida grapefruit infested with Caribbean fruit
fly, Fla. Entomol. 74 (1990) 242-247.
[24] BALOCK, J.W., et al., Gamma radiation as a quarantine treatment for Hawaiian fruit
flies, J. Econ. Entomol. 59 (1966) 202-204.
[25] KOMSON, P. et al., “ Irradiation as a quarantine treatment for the oriental fruit fly,
Dacus dorsalis Hendel, in mangoes” , Modem Insect Control: Nuclear Techniques and
Biotechnology (Proc. Symp. Vienna, 1987), IAEA, Vienna (1988) 319-324.
[26] MANOTO, E.C., et al., “ Effects of gamma radiation on the insect mortality and fruit
quality o f Philippine ‘Carabao’ mangoes” , Use o f Irradiation as a Quarantine Treat­
ment of Food and Agricultural Commodities (Proc. Research Co-ordination Mtg Kuala
Lumpur, 1990), IAEA, Vienna (1992) 91-116.
[27] THOMAS, P., RAKHALAR, G.W., Disinfestation of fruit flies in mango by irradia­
tion, Curr. Sci. (1975) 775-776.
[28] JESSUP, A.J., RIGNEY, C.J., M ILLAR, A., SLOGGETT, R.F., QUINN, N.M .,
“ Gamma irradiation as a commodity treatment against the Queensland fruit fly in fresh
fruit” , Use of Irradiation as a Quarantine Treatment of Food and Agricultural Com­
modities (Proc. Research Co-ordination Mtg Kuala Lumpur, 1990), IAEA, Vienna
(1992) 13-42.
[29] HAQUE, H., AHMAD, R., Effects of ionising radiation on Dacus zonatus fruit fly
eggs and larvae in situ, Pakistan J. Sci. 19 (1967) 233-238.
[30] BURDITT, A.K., Jr., HUNGATE, F.P., Gamma irradiation as a quarantine treatment
for cherries infested by Western Cherry fruit fly (Diptera: Tephritidae), J. Econ.
Entomol. 81 (1988) 859-862.
[31] HEATHER, N.W., “ Review of irradiation as a quarantine treatment for insects other
than fruit flies” , Use of Irradiation as a Quarantine Treatment of Food and Agricultural
Commodities (Proc. Research Co-ordination Mtg Kuala Lumpur, 1990), IAEA, Vienna
(1992) 203-218.
[32] WIT, A.K.H , DE VRIE, М., “ Possibilities for irradiation to control insects and mites
in cut flowers after harvest” , Irradiation as a Quarantine Disinfestation Treatment,
Report to the 1st Meeting of the Co-ordinated Research Project, Chiang Mai, 1986, per­
sonal communication.
638 HEATHER

[33] BURDITT, A .K ., Jr., HUNGATE, F.P., Gamma irradiation as a quarantine treatment

for apples, infested by codling moth (Lepidoptera: Tortricidae), J. Econ. Entomol. 82
(1989) 1386-1390.
[34] BURDITT, A .K ., Jr., Gamma irradiation as a quarantine treatment for walnuts infested
with codling moths (Lepidoptera: Tortricidae), J. Econ. Entomol. 79 (1986)
1577-1579.
[35] BATCHELOR, T.A., O’DONNELL, R.L., ROBY, J.J:, “ Irradiation.as a quarantine
treatment for ‘Granny Smith’ apples infested with Epiphyas postvittana (Walk.) (light-
brown apple moth) stages” , Irradiation as a Quarantine Disinfestation Treatment,'
Report to the 1st Meeting of the Co-ordinated Research Project, Chiang Mai, 1986, per­
sonal communication.
[36] DENTENER, P.C., WADDELL, B.C., BATCHELOR, T.A., “ Disinfestation of
lightbrown apple moth: A discussion of three disinfestation methods” , Managing
Postharvest Horticulture in Australia (BEATTIE, B.B., Ed.), Occasional Publication
No. 46, Australian Institute o f Agricultural Science, Melbourne (1990) 166-177.
[37] JOHNSON, J.A.,. et al., Gamma radiation as a quarantine treatment for Fuller rose
beetle eggs (Coleoptera: Curculionidae) on citrus fruit, J. Econ. Entomol. 83 (1990)
905-909.
[38] SEO, S.T., et al., Mango weevil: Cobalt-60 gamma irradiation of packaged mangoes,
J. Econ. Entomol. 67 (1974) 504-505.
[39] HALFHILL, J.E., Irradiation disinfestation o f asparagus spears contaminated with
Brachycorynella asparagi (Mordvilko) (Homoptera: Aphididae), J. Econ. Entomol. 81
. (1988) 873-876.
[40] SANCHEZ, F., “ Irradiation quarantine, treatment to sterilise Quadraspidiotus
pemiciosus and Neotermes chilensis” , Report of the Final Research Co-ordination
Meeting on the Use of Irradiation as a Quarantine Treatment of Food and Agricultural
Commodities, Kuala Lumpur, 1990, Food Irradiat. Newsl. 15 1 (1991) 51.
[41] ARELLANO, F.S., “ Irradiation treatment to sterilise Naupactus xanthographus and
Brevipalpus chilensis females in table grapes” , Irradiation as a Quarantine Treatment,
International Consultative Group on Food Irradiation, Task Force Meeting, Orlando,
IAEA, Vienna, 1988 (internal report).
[42] IGNATOWICZ, S., BRZOSTEK, G., Use o f irradiation as a quarantine treatment for
agricultural products infested by mites and insects, Radiat. Phys. Chem. 35 (1990)
263-267.
[43] BROWER, J.H., TILTON, E.W., “ The potential of irradiation treatment for insects
infesting stored-food commodities” , Radiation Disinfestation of Food and Agricultural
Products (Proc. Conf. Honolulu, 1983) (MOY, J.H., Ed.), University of Hawaii at
Manoa, Honolulu (1985) 75-86.
[44] ZAKLADNOI, G.A., Mukomol’no Elevat. Kombi, Promyshl 6 (1981) 29.
[45] BALOCK, J.W., et al., Effects o f gamma radiation on various stages of three fruit fly
species, J. Econ. Entomol. 56 (1963) 42-46.
[46] SHIPP, E., “ Ionising treatment of food — disinfestation entomology” , Regional
Workshop on Commercialisation of Ionising Energy Treatment, of Food, Lucas
Heights, 1985, Lecture II, Australian Atomic Energy Commission, Lucas Heights,
New South Wales (1985).
 IAEA-SM-327/S8 639

[47] NATION, J.L., SMITTLE, B.J., “ A proposal to investigate physiological ways to

determine when insects have been irradiated with ionising irradiation” , Irradiation as
a Quarantine Treatment of Mites, Nematodes and Insects Other than Fruit Fly, Interna­
tional Consultative Group on Food Irradiation, Task Force Meeting, Gainesville,
IAEA, Vienna, 1992 (internal report).
[48] RAHMAN, R., et al., Irradiation as a quarantine treatment against Ceratitis capitata
(Diptera: Tephritidae): Anatomical and cytogenetic changes in mature larvae after
gamma irradiation, J. Econ. Entomol. 83 (1990) 1449-1454.
[49] McLAUCHLAN, R.L., etal., Ionising Energy Treatment for Fresh Horticultural
Produce, Mandarins and Other Produce, Trials I and II, May-Jul. 1987, Department
of Primary Industries Project Report, Brisbane, Queensland (1988).
[50] McLAUCHLAN, R.L., etal., Ionising Energy Treatment for Fresh Horticultural
Produce — Capsicum, Persimmon, Cucumber, Zucchini, Nectarine and Peach, Trials
IV, V and VII, Apr.-Dec. 1988, Department o f Primary Industries Project Report,
Brisbane, Queensland (1991).
[51] WOOTEN, М ., “ Application of gamma irradiation to cereals and cereal products” ,
Regional Workshop on Commercialisation of Ionising Energy Treatment of Food,
Lucas Heights, 1985, Lecture 14, Australian Atomic Energy Commission, Lucas
Heights, New South Wales (1985).
[52] JESSUP, A.J., M ILLAR, A., “ The identification of irradiated immature lifestages of
Queensland fruit fly (Dacus tryoni Froggatt)” , Report to the Final Research Co­
ordination Mtg, Kuala Lumpur, 1990.
[53] LIN, CHAO SEN, HEATHER, N.W., “ Experiment on irradiation of lychee for
Queensland fruit fly, Bactrocera tryoni (Froggatt)” , Proc. X IX Int. Congress of Ento­
mology, Beijing, 1991 (1992) 590 (abstract only).
[54] YULO-NAZAREA, M .T., etal., Polyacrylamide gel electrophoresis as a detection
method for irradiated oriental fruit fly, Nucleus X X IX (1991) 47-52.
 IAEA-SM-327/60

A BIOCHEMICAL MARKER FOR DETECTION

OF IRRADIATED PUPAE OF THE
ORIENTAL FRUIT FLY (Dacus dorsalis)
M .T . Y U L O -N A Z A R E A , E .C . M A N O T O
Department o f Science and Technology,
Philippine Nucleár Research Institute,
Diliman, Quezon City,
Philippines

A bstract

A BIOCHEMICAL MARKER FOR DETECTION OF IRRADIATED PUPAE OF THE

ORIENTAL FRUIT FLY (Dacus dorsalis).
•A biochëmical method for detécting a radiation sensitive protein marker in pupae of
gamma irradiated larvae of the Oriental fruit fly (Dacus dorsalis) using SDS-polyacrylamide
gel electrophoresis (SDS-PAGE) is presented. Third instar larvae were irradiated at 100 Gy
using а мСо dry gamma cell. SDS-PAGE protein profile analysis o f pupae o f the irradiated
larvae revealed the absence of a specific protein band (ascribed to a gene product, Gs, of a
putative radiation sensitive marker gene) in comparison to the protein profile o f unirradiated
control larvae. Absence o f the Gs protein band from the electrophoretic gel profile was
observed only in two-dayrold pupae (day 3 post-irradiation) and even at a dose as low as
40 Gy. Thus, irradiation o f immature (two-day-old) larvae at 100 Gy resulted in the sub­
sequent absence of the Gs protein only after day 8 post-irradiation (three-day-old pupae), by
which stage the pupae o f the unirradiated larvae had already started to synthesize the Gs
protein, as seen by the SDS-PAGE gel profile. The absence of this Gs protein band in the
SDS-PAGE profile of pupae o f the irradiated larvae could be used as a biochemical marker
for establishing the efficacy o f irradiation o f the Oriental fruit fly.

1. IN TR O D U C TIO N

The quarantine security level for fresh agricultural products imposed by

importing countries usually involves the minimum radiation dose that will prevent
insects from emerging into viable adults. This is the criterion for establishing the
efficacy o f irradiation as a quarantine treatment [1]. A 300 G y dose is the minimum
dose required to treat any insect stage for such quarantine security [2]; for member
species o f the Tephritidae fam ily, 100 Gy is the minimum dose [2].
Experimental data have shown that as a quarantine treatment, a dose o f 100 Gy
prevented the Oriental fruit fly, Dacus dorsalis, from emerging into an adult capable
o f flight [3]. The tolerance o f fruit flies to radiation increases with age; this sensitiv­
ity to radiation depends on the species, the age and the stage o f development as well

641
642 YULO-NAZAREA and MANOTO

as the dose rate applied. Mature, third instar larvae are more resistant to a low dose
o f 25 Gy than larvae at the first and second instar stages [4].
Detection methods for irradiated foods are required to determine whether food
has been irradiated effectively, i.e. whether the larvae or pupae found in a fo o d pack­
age will emerge as viable adult flies. Several detection methods have been presented
as part o f the Programme on Analytical Detection Methods for Irradiation Treatment
o f Foods (A D M IT ). A few are acceptable but, to date, no detection method has been
adopted as an internationally agreed upon protocol for establishing the efficacy o f
irradiation treatment o f foods [1].
Separation and identification o f macromolecules using polyacrylamide gel
electrophoresis (PAGE) show some potential for adaptation as an analytical tech­
nique for detecting radiation induced changes in protein biosynthesis. Our prelimi­
nary studies [5] on the electrophoretic profile o f proteins extracted from D. dorsalis
at different stages o f development showed changes in certain proteins as the insect
evolved into the adult stage. Distinct banding patterns o f the proteins were observed
in the electrophoretic profile, at each stage o f development. Thus, each profile could
be used as an electrophoretic signature for a particular stage o f insect develop­
ment [5].
One o f the recommendations made at the International Consultative Group on
Food Irradiation (ICGFI) Task Force Meeting held in 1991 in Bethesda, Maryland,
U SA , was to develop a rapid, practical technique for determining whether insects
have been irradiated adequately [2]. T o address this problem , this paper presents a
biochemical detection method for gamma irradiated larvae o f the Oriental fruit fly
using SD S -P A G E .

2. M A TE RIALS A N D M ETH ODS

Acrylam ide, bis-acrylamide, ammonium persulphate, T R IZ M A base, glycine,

TEM E D (tetramethyl ethylenediamine), bromophenol blue, Coomassie Brilliant
blue, molecular weight markers (bovine serum albumin, chicken albumin, urease)
were obtained from Sigma. Glycerine CP was purchased from a local chemical com ­
pany. 2-mercaptoethanol was obtained from Fluka. N aH P 04, E D T A , KC1,
methanol .and acetic acid were purchased from Baker Chemicals. Sodium dodecyl
sulphate was obtained from Aldrich Chemicals. The homogenizer used was Dounce
homogenizer purchased from V W R (U SA ). A Bio-Rad Mini-Protean II electropho­
resis apparatus,was used employing a Buchler power supply.

2.1. Irradiation

Dacus dorsalis larvae (5 to 6-day-old (third instar)) were irradiated at the

multipurpose gamma irradiation facility o f the Philippine Nuclear Research Institute
 IAEA-SM-327/60 64 3

using a 60C o dry gamma cell. Immature D. dorsalis larvae (two-day-old) were like­
wise irradiated in the same facility. A fter irradiation, the larvae were transferred to
vials containing coconut coir dust and allowed to continue their life-cycle at room
temperature.

2.2. Homogenization

Five pieces o f post-irradiated larvae/pupae were homogenized in 0 .4 mL

buffer (lOm M N aH P04, 400m M KC1, Im M E D T A ), pH 7.6, in the cold using a
Dounce homogenizer. The homogenate was centrifuged at 1500g for 5 min at 8 °C
and the soluble fraction was transferred to a clean test tube.

2.3. SDS-PAGE

A 100 /xL sample buffer (3.3% SDS, 10% glycerine, 0 .1 M tris) was added to
100 fiL o f soluble fraction. 2-mercaptoethanol (10 /xL) and 0.1 % bromophenol blue
(10 fiL) were pipetted into the sample in buffer solution and heated at 8 0 -9 0 ° С for
1 min or longer. The Laemmli method for slab gel electrophoresis was follow ed,
with some modifications [6]. The 10-15 /xL sample was applied to each electrophore­
sis well. A 7 .5 % separating gel was used, with a 3 .5 % stacking gel. The running
buffer was 0.025M tris, 0 . 19M glycine. The proteins were electrophoresed at 25 m A
for about 1 h or until the blue dye was approximately 2 cm above the low er end o f
the gel. Gel was stained in Coomassie Brilliant blue in 7% H A c for 1-2 h and
destained in 7% H A c, 5% M eO H overnight, or until the blue dye background was
removed.

3. RESULTS A N D DISCUSSION

Mature (six-day-old) larvae irradiated at 100 Gy and ‘ electrophoresed’ at day

3 post-irradiation (two-day-old pupae) showed the absence o f a protein band when
compared with the unirradiated pupae o f the same age. This band, which we desig­
nated the Gs protein, only appeared at the start o f day 3 post-irradiation (two-day-old
pupae); it had evidently not been synthesized at an earlier stage (Fig. 1).
It has been shown that although larvae irradiated at a dose o f 100 G y g o into
pupation, they fail to emerge into adult flies capable o f flight [7]. A t this dose, a
probit 9 (99.9968% mortality) quarantine security is reached for D. dorsalis [2]. In
our studies, we observed delayed pupation o f irradiated third instar larvae; day 3
post-irradiated larvae (tw o-day-old pupae) exhibited electrophoretic profiles showing
the absence o f the Gs protein band (Fig. 1).
T o determine when the putative gene coding for the Gs protein is turned on
for synthesis and how this is affected by gamma irradiation, immature, two-day-old
644 YULO-NAZAREA and MANOTO

■—
д В дШ ш М
ш ш л ттШ Я
■■■■■■I
■
■ в
Щр
¡¡1 11
Шш

ЯЯЯЯШЯШШШ

f ,-^ r '* ( * *
■ИЯИИИмИИШ
ЯИИИИИ»
В

F/G. /. 77ie SDS-PAGE protein profile o f seven-day-old larvae (A, В) one-day-old pupae
(C, D) and two-day-old pupae (E, F) at day 1, day 2 and day 3 post-irradiated third instar
larvae o f D. dorsalis, respectively. Dose is 100 Gy. The irradiated larvae are A, С and E,
and the unirradiated control larvae are B, D and F. The arrow in F indicates the Gs protein.

larvae were irradiated at 10, 25, 50 and 75 Gy. A t day 4 post-irradiation, the elec­
trophoretic profile showed identical protein patterns for both the irradiated and the
unirradiated larvae at all the radiation doses tested. When irradiation o f the immature
larvae reached 100 G y, the S D S -P A G E profile o f six- and seven-day-old larvae
(days 4 and 5 post-irradiation) did not differ from the protein profiles o f larvae irradi­
ated at low er doses. The GS protein band was definitely absent in the electrophoretic
profiles derived from both the irradiated and unirradiated larvae; at days 4 and 5
post-irradiation, the insect had not yet turned into a pupa and so the protein profiles
were still those o f a larva. This suggests that at this early stage, when the insect was
still in its larval stage, the putative gene coding fo r the Gs protein had not yet been
transcribed. How ever, in one-day-old pupae (day 6 post-irradiation) the elec­
trophoretic profile showed a faintly stained band o f Gs protein, which became a
 IAEA-SM-327/60 645

darkly stained band in three-day-old pupae (day 8 post-irradiation). In contrast, this

Gs band was definitely missing in the pupae o f irradiated larvae o f the same age
(Fig. 2). Experimental results suggest that at least one mutational event had taken
place at the level o f the genome and that the putative gene susceptible to such damage
(and coding for the Gs protein) was responsible for the synthesis o f a primary or
secondary gene product necessary to transform pupae into adult flies. It has been
demonstrated [4] that in the D. dorsalis life-cycle, immature larvae are. the insect
stage most sensitive to irradiation at a dose o f 100 Gy or higher.
The electrophoretic protein profile o f three-day-old pupae from irradiated
(2 5 -5 0 G y) larvae at the third instar stage and tested for Gs protein already showed
the absence o f the Gs protein band at a low dose o f 40 G y (Fig. 3). The Gs protein
bands from pupae o f larvae irradiated at doses o f 25, 30 and 35 Gy were still faintly

А В C D E F

FIG. 2. The SDS-PAGE protein profile o f seven-day-old larvae (A, B), one-day-old pupae
(С, D) and three-day-old pupae (E, F) at day 5, day 6 and day 8 post-irradiated immature
(two-day-old) larvae o f D. dorsalis, respectively. Dose is 100 Gy. The irradiated larvae
are B, D and F, and the unirradiated control larvae are A, С and E. The arrow in E indicates
the Gs protein.
646 YULO-NAZAREA and MANOTO

А В С D E F G

FIG. 3. The SDS-PAGE protein profile o f three-day-old pupae (day 4 post-irradiation) o f the
third instar larvae o f D. dorsalis irradiated at various doses. The irradiated larvae are В
(25 Gy), С (30 Gy), D (35 Gy), E (40 Gy), F (45 Gy) and G (50 Gy) and the unirradiated
control larvae are A. The arrow in A indicates the Gs protein.

present (compared with the electrophoretic profiles o f the pupae from unirradiated
larvae). This signifies that some form o f damage (D N A lesion) had already been sus­
tained by the Gs gene at a low dose o f 25 G y, although this dose may not be sufficient
to disable (transcriptionally) the putative Gs gene completely.
The apparent molecular weight o f the Gs protein, by linear regression, ranges
between 105 and 115 kilodalton. This protein is detectable only at the pupal and adult
stages o f the D. dorsalis life-cycle and may be an obligatory protein in the maturation
process towards the pupal stage and, beyond, into adult flies capable o f flight.
M anoto and Resilva [7] have shown that in larvae exposed to 100 G y, pupation
takes place but not emergence into adults In our studies, the S D S -P A G E profile o f
pupae from the irradiated larvae showed clearly that these pupae do not emerge into
adult flies because o f the ‘ missing’ protein component (Fig. 1). It was apparent that
 IAEA-SM-327/60 647

a certain gene-enzyme system had been disrupted by gamma radiation and thus cellu­
lar production o f certain key proteins had becom e inoperative, which explains why
a certain protein band found in tw o-day-old control pupae was missing in two-day-
old pupae (day 3 post-irradiation) o f irradiated larvae. It is speculated that this partic­
ular Gs protein is necessary for the emergence o f pupae into adults capable o f flight.

4. CONCLUSION S

Studies on the S D S -P A G E profile o f proteins from pupae o f irradiated larvae

showed distinctly that a specific protein, designated as Gs, which is not present in
the early stages o f development (egg and larva), is needed in the transformation
process o f pupae into adult flies capable o f flight. Absence o f this putative Gs protein
from the pupal stage was observed when the larvae were irradiated at a dose as low
as 40 Gy. Our studies showed that the gene coding for this protein may have sus­
tained a mutational lesion upon exposure o f the larvae to a gamma dose as low as
40 Gy. Thus, as a result o f radiation at this dose, this protein was found to be absent
in the S D S -P A G E profile o f tw o-day-old pupae o f the irradiated larvae. This can
be taken as p ro o f that the irradiation o f the pest, even at the immature larval stage,
will prevent its emergence into an adult fly capable o f flight. The absence o f the Gs
protein band in the S D S -P A G E protein profile o f pupae therefore shows good poten­
tial as a biochem ical marker for establishing the efficacy o f irradiation o f fruit flies.
It is also an indicator that can be used by the quarantine services o f a fruit importing
country. The test, as implemented here, need not use S D S -P A G E ; for example,
some type o f colou r reaction could be used that is more specifically designed to
detect directly the absence o f the Gs protein in the soluble protein fraction o f the
pupae. This type o f direct test would require further developmental studies, which
are currently under way at our laboratory.

ACKNOWLEDGEMENTS

The authors express their special thanks to A .D . Nazarea for his support,
invaluable discussions and critical review o f the manuscript. They would also like
to thank L .C . Cobar for technical assistance and D. Villamater for irradiation o f the
fruit flies.

REFERENCES

[1] IN TER N ATIO N A L ATO M IC ENERGY AGENCY, First Research Co-ordination

Meeting o f the Research Co-ordination Programme on Analytical Detection Methods
for Irradiation Treatment o f Foods (A D M IT ), Warsaw, 1991, Food Irradiat. Newsl.
IS 1 (1991) 27-41.
648 YULO-NAZAREA and MANOTO

[2] IN TER N ATIO N A L A TO M IC ENERGY AGENCY, Irradiation as a Quarantine Treat­

ment o f Fresh Fruits and Vegetables, Report o f the International Consultative Group
on Food Irradiation, Bethesda, M D , IA E A , Vienna, 1991 (internal report).
[3] MANOTO, E., Effect o f gamma radiation on insect mortality and fruit quality o f
Philippine mangoes (cv Carabao on Manila Super), Food Irradiat. Newsl. 15 1
(1991) 45.
[4] VUAYSEGARAN, S., Gamma irradiation as a quarantine treatment o f carambola,
papaya, mango and guava, Food Irradiat. News. 15 1 (1991) 45.
[5] N AZAREA, T .Y ., COBAR, M .L ., M AN O TO , E., BILOG, G., Polyacrylamide gel
electrophoresis as a detection method for irradiated oriental fruit fly , Nucleus 29 (1991)
47-52.
[6] L A E M M L I, U .K ., Cleavage o f structural proteins during the assembly o f the head o f
Bacteriophage T4, Nature (London) 227 (1970) 680-685.
[7] MANOTO, E., RESILVA, S., Effects o f gamma radiation on mortality o f the oriental
fru it fly, Dacus dorsalis Hendel, when inoculated in Manila Super mango, Philipp.
Nucí. J. 6 (1989) 453-458.
POSTER PRESENTATION

IAEA-SM-327/19P

IRRADIATION AS A QUARANTINE TREATMENT

FOR EUROPEAN RED MITE EGGS ON APPLES*

S. IG N A T O W IC Z
Department o f Applied Entom ology,
Agricultural University o f Warsaw,
Warsaw,
Poland

Viable eggs o f the European red mite, Panonychus ulmi K och (Acari: Tetrany-
chidae), on apples have been the concern o f several importing countries, and exports
require pre-shipment (i.e. quarantine) treatment to eliminate the live eggs [1]. The
follow ing treatments are suggested: fumigation with chemicals, m odified
atmosphere, cold treatment and ionizing radiation.
The objectives o f this study were to determine the density o f European red mite
eggs on apples, and to determine the dose o f gamma radiation required to prevent
development o f diapausing eggs o f P. ulmi.
Apples with eggs o f European red mite were treated with ^ C o gamma radia­
tion at a dose rate o f 140 Gy/m in. After the treatment, the calyx end and stem cavity
o f each apple were ringed with white glycerine prior to incubation. Apples with eggs
were incubated in darkness at 2 4 °C for a period o f 14 d. After this period, the apples
were examined and the eggs and larvae counted.
Usually, 7 0 -9 0 % o f the apples were infested with eggs o f P. ulmi. The more
infested apples in the sample, the more eggs were laid per apple. Distribution o f mite
eggs on apples was very characteristic. Aggregations o f eggs were located in the
calyx ends m ore often than in the stem cavities.
Preliminary experiments indicate that gamma radiation applied at 0 .4 -0 .6 kGy
inhibited completely the hatchability o f eggs. About 60% o f control eggs hatched.
The effects o f low er doses o f gamma radiation on the viability o f diapausing
eggs are presented in Table I. It is seen that wintering eggs o f P. ulmi are very
susceptible to irradiation. A dose o f 100-125 G y caused com plete inhibition o f egg
hatchability. This dose is suggested for quarantine treatment o f apples infested with
diapausing eggs o f the European red mite. A few larvae hatched at doses higher than
100 G y. H ow ever, these larvae died soon after treatment without moulting into the
next developmental stage.

This research was carried out with the support o f the IA E A under Research
Contract No. 6942/RB.

649
650 POSTER PRESENTATION

T A B L E I. H A T C H A B IL IT Y OF W IN T E R IN G EGGS OF TH E EUROPEAN
RED M IT E TR EATED W IT H G A M M A R A D IA T IO N

Apple Dose No. of eggs % hatched % dead

variety (kGy) treated eggs eggs

Bancroft 0.0 428 66.4 33.6

0.125 395 0.0 100.0
0.25 430 0.5 99.5

Boiken 0.0 672 55.1 44.9

0.1 682 0.0 100.0
0.2 384 0.3 99.7
0.3 308 0.0 100.0
0.4 558 0.0 100.0

Apples are among those fresh fruits and vegetables which are relatively
tolerant to irradiation stress at doses below 1 kGy. A t these doses, only m inimum
detrimental effects o f radiation on apples have been observed.
The embryonic stage o f an animal is a tim e o f extreme radiosensitivity [2, 3]
and mites are no exception. However, the specific stage o f embryonic development
involved determines the observed radiosensitivity [4]. Usually, as the embryo
achieves greater development, its resistance increases dramatically. Taking this into
consideration, as w ell as the results which indicate the high susceptibility o f the
wintering eggs o f P. ulmi to irradiation, one may conclude that these diapausing eggs
|r e at the early stages o f embryogenesis, before termination o f diapause.

ACKNOW LEDGEM ENT

The support received from the IA E A is gratefully acknowledged.

REFERENCES

[1] LIDSTER, P.D., SANFORD, K.H., McRAE, K.B., “ Effects o f modified atmosphere
storage on overwintering populations of the apple rust mite and European red mite eggs,
HortScience 16 (1981) 328-329.
 POSTER PRESENTATION 651

[2] BROWER, J.H ., “ Combined effects o f egg age and radiation dosage on egg hatch o f
Tenebrio molitor (Coleoptera: Tenebrionidae), Can. Entomol. 104 (1972) 141-147.
[3] TTLTON, E .W ., BROWER, J.H., “ Radiation effects on arthropods” , Preservation o f
Food by Ionizing Radiation, CRC Press, Boca Raton, F L (1983) 269-316.
[4] IG NATO W ICZ, S., SYSIAK, M :, “ Sterilization o f the bulb mite, Rhizoglyphus
echinopus (F. et R.) (Acarida: Acaridae), with gamma radiation” , Proc. 5th Int.
Workshop on Stored-Products Protection, Bordeaux, 1990, Vol. 2 (1990) 1211-1220.
 CHAIRMEN OF SESSIONS

Session 1 W . K LASSEN F A O /IA E A

Session 2 J. O A K E SH O TT Australia
A .S . ROBINSON United Kingdom
Session 3 Y . ITÔ Japan
Session 4 L .E . L A C H A N C E F A O /IA E A
Session 5 V .A . D Y C K Canada
Session 6 S. B A R BO SA FAO
Session 7 B. B E N N ETTO V Á Czechoslovakia
Session 8 D .A . LINDQU IST F A O /IA E A

SECRETARIAT OF THE SYMPOSIUM

W . KLASSEN Scientific Secretary (F A O /IA E A )

S. B A R BO SA Scientific Co-Secretary (F A O )
H. PROSSER Symposium Organizer (IA E A )
P. H O W A R D -K IT T O
R .F . KELLEHER Proceedings Editors (IA E A )
G .V . RAM E SH
A . HAM ENDE French Editor
L . HERRERO Spanish Editor
 LIST OF PARTICIPANTS

(The affiliations given are those that pertained at the time o f the Symposium)

Aksoy, S. Yale University,

700 LEPH, 60 College Street,
New Haven, CT 06525,
United States o f America

Al-Shawan, A .M .F . Al-Hassa Regional Agricultural Research Center,

P.O. Box 43, Hofouf,
Al-Hassa 31982,
Saudi Arabia

Amsler, S. Centre de recherches sur les

trypanosomoses animales (CRTA),
B.P. 454,
Bobo-Dioulasso 01,
Burkina Faso

Atkinson, P.W. Division o f Entomology,

Commonwealth Scientific and Industrial
Research Organization,
G.P.O. Box 1700,
Canberra ACT 2601,
Australia

Attathorn, T. Department o f Entom ology,

Kasetsart University,
Kamphaengsaen Campus,
Nakom Pathom 73140,
Thailand

Barbosa, S. Food and Agriculture Organization o f

the United Nations (FАО /AGP),
Vie delle Terme di Caracalla
1-00100 Rome,
Italy,

Bennettová, В. Institute o f Entomology,

Czechoslovak Academy o f Sciences,
Branisovská 31,
CS-37005 Ceské Budéjovice,
Czechoslovakia

655
656 LIST OF PARTICIPANTS

Blanchetot, A. Department o f Biochemistry,

University o f Saskatchewan,
Saskatoon, Saskatchewan S7N OWO,
Canada

Blümel, S. Federal Agency o f Plant Protection,

Trunnerstrasse 1-5,
A - 1020 Vienna,
Austria

Bueds, H. Zoological Institute,

Katholieke Universiteit Leuven,
Naamsestraat 59,
B-3000 Leuven,
Belgium

Carpenter, J.E. Insect Biology and Population Management

Research Laboratory,
Agricultural Research Service,
United States Department o f Agriculture,
P.O. Box 748,
Tifton, GA 31793-0748,
United States o f America

Chadenga, V. Tsetse and Trypanosomiasis Control Branch,

P .O . B ox 8283,
Causeway,
Zimbabwe

Cortez Martinez, D. Ministry o f Agriculture,

P.O. Box 2335-2312, Telcor,
Managua,
Nicaragua

Crampton, J.M. Wolfson Unit o f Molecular Genetics,

Liverpool School o f Tropical Medicine,
Pembroke Place,
Liverpool L3 5QA,
United Kingdom

de Azeredo-Espin, A .M .L . Departamento Genética e Evoluçâo,

Centro de Biologia Molecular e Enga Genética,
Universidade Estadual de Campinas,
Caixa Postal 6109-CEP 13081,
Campinas, Sâo Paulo,
Brazil
 LIST OF PARTICIPANTS 657

De Clercq, R. Research Station for Nematology and Entomology,

Ministry o f Agriculture,
Burg. Van Gansberghelaan 96,
B-9820 Merelbeke,
Belgium

Djiteye, A. Laboratoire central vétérinaire,

B.P. 2295,
Bamako,
Mali

Dyck, V .A . Agriculture Canada Research Station,

Summerland,
British Columbia VOH 1Z0,
Canada

Economopoulos, A .P. Department o f Biology,

University o f Crete and Institute o f
Molecular Biology and Biotechnology,
P.O. Box 1470,
GR-71110 Heraklion, Crete,
Greece

El-Mayasse, I. Département de entomologie,

Institut de recherches agronomiques,
с/o Mr. Mounzer,
P.O. Box 11-8281,
Beirut,
Lebanon

Feldmann, U. Agency’ s Laboratory,

International Atomic Energy Agency,
Wagramerstrasse 5, P.O. Box 100,
A-1400 Vienna,
Austria

Foster, G.G. Division o f Entomology,

Commonwealth Scientific and Industrial
Research Organization,
G .P.O . Box 1700,
Canberra AC T 2601, .
Australia
6 58 LIST OF PARTICIPANTS

Franz, G. Agency’ s Laboratory,

International Atomic Energy Agency,
Wagramerstrasse 5, P.O. Box 100,
A-1400 Vienna,
Austria

García, J.C. Instituto de Investigaciones de Sanidad Vegetal,

Calle Union No. 56 e/Johnson y milagros,
10 de Octubre,
Havana,
Cuba

Gasperi, G. Dipartimento di Biología Animale,

Université di Pavia,
Piazza Botta, 9,
1-27100 Pavia,
Italy .

Ghannam, M. Abdul R. Al-Hassa Regional Agricultural Research Center,

P.O. Box 43,
Hofouf,
Al-Hassa 31982,
Saudi Arabia

Gooding, R.H. Department o f Entomology,

University o f Alberta,
Edmonton, Alberta T6G 2E3,
Canada

Hamann, H.J. Internationales Büro,

Gesellschaft fflr Strahlen-
und Umweltforschung mbH München,
Ingolstadter Landstrasse 1,
D-W -8042 Neuherberg,
Germany

Handler, A .M . Insect Attractants, Behavior, and Basic

Biology Research Laboratory,
Agricultural Research Service,
United States Department o f Agriculture,
1700 SW 23rd Drive,
Gainesville, FL 32608,
United States o f America
 LIST OF PARTICIPANTS

Hargrove, J.W. Insect Pest Management Initiative,

Tsetse and Trypanosomiasis Control Branch,
P.O. Box 8283,
Causeway,
Zimbabwe

Haymer, D. Department o f Genetics and Molecular Biology,

1960 East West Road,
University o f Hawaii,
Honolulu, HI 96822,
United States o f America

Heath, R.R, Insect Attractants, Behavior, and Basic

Biology Research Laboratory,
Agricultural Research Service,
United States Department o f Agriculture,
1700 SW 23rd Drive,
Gainesville, FL 32608,
United States o f America

Heather, N .W . Department o f Primary Industries,

Agricultural Research Laboratories,
80 Meiers Road,
Indooroopilly, Queensland 4068,
Australia

Hendrichs, J. Agency’ s Laboratory,

International Atomic Energy Agency,
Wagramerstrasse 5, P.O. Box 100,
A-1400 Vienna,
Austria ,

Hôbaus, E. Federal Agency o f Plant Protection,

Trunnerstrasse 1-5,
A - 1020 Vienna,
Austria

Hussain, T. Pakistan Atomic Energy Commission,

P.O. Box 1114,
Islamabad,
Pakistan
660 LIST OF PARTICIPANTS

Huybrechts, R. Zoological Institute,

Katholieke Universiteit Leuven,
Naamsestraat 59,
B-3000 Leuven,
Belgium

Ignatowicz, S. Agricultural University o f Warsaw,

ul. Nowoursynowska 166,
PL-02-766 Warsaw,
Poland

Irastorza, J.M. Comisión México-Americana para la

Erradicación del Gusano Barrenador del G anado,.
Apartado Postal M -2890, Leibnitz No. 11,
M exico DF 06000,
M exico

Itô, Y. Faculty o f Science and Arts,

Okinawa University,
555 Kokuba, Naha,
Okinawa 902,
Japan

Kafu, A .A . P.O. Box 2933,

Tripoli,
Libyan Arab Jamahiriya

Kerremans, P. Agency’ s Laboratory, ,

International Atomic Energy Agency,
Wagramerstrasse 5, P.O. Box 100,
A-1400 Vienna,
Austria

Kolste, A.-B. Biotechnology Centre o f Oslo,

University o f Oslo,
P.O. Box 1125 Blindern,
N-0316 Oslo 3,
Norway

LaChance, L.E. P.O. Box 5571,

Fargo, ND 58105,
United States o f America
 LIST OF PARTICIPANTS

Leopold, R .A . Biosciences Research Laboratory,

Agricultural Research Service,
United States Department o f Agriculture,
P.O. Box 5674,
State University Station,
Fargo, ND 58105,
United States o f America

Lindquist, D .A . Joint FAO /IAEA Division o f Nuclear

Techniques in Food and Agriculture,
International Atomic Energy Agency,
Wagramerstrasse 5, P.O. Box 100,
A - 1400 Vienna,
Austria

Макее, H. Atomic Energy Commission,

P.O. Box 6091,
Damascus,
Syrian Arab Republic

Malacrida, A. Dipartimento di Biologia Animale,

Université di Pavia,
Piazza Botta, 9,
1-27100 Pavia,
Italy

Malva, С. Istituto Intemazionale di Genetica e Biofísica,

Via Guglielmo Marconi, 10,
1-80125 Naples,
Italy

Maree, F. Institute o f Entomology,

Czechoslovak Academy o f Sciences,
Branisovská 31,
CS-37005 Ceské Budéjovice,
Czechoslovakia

Matioli, S.R. Departamento de Biologia (IBUSP),

Instituto de Biociencias,
Universidade de Sâo Paulo,
Caixa Postal 11.461,
05499 Sâo Paulo, SP,
Brazil
6 62 LIST OF PARTICIPANTS

Maudlin, I. Tsetse Research Laboratory,

Deparment o f Veterinary Medicine,
University o f Bristol,
Langford, Bristol BS18 7DU,
United Kingdom

Mialhe, E. Laboratory o f Malaria Research,

National Institutes o f Health,
Building 4, Room B2-37,
Bethesda, M D 20892,
United States o f America

Miyatake, T. Fruit Flies Laboratory,

Okinawa Prefectural Agricultural Experiment Station,
4-222, Sakiyama-cho, Naha,
Okinawa 903,
Japan

Moazami, N. Biotechnology Department,

Iranian Research Organization for Science
and Technology,
P.O. Box 15815-3538,
Tehran,
Islamic Republic o f Iran

Mouches, С. Laboratoire d ’écologie moléculaire,

University de Pau et des Pays de l ’ Adour,
F-64000 Pau,
France

Oakeshott, J. Division o f Entomology,

Commonwealth Scientific and Industrial
Research Organization,
G .P.O . Box 1700,
Canberra A C T 2601,
Australia

Offori, E.D. P.O. Box 47,

Achimota,
Ghana

Osir, E.O. International Centre o f Insect Physiology and Ecology,

P.O. Box 30772,
Nairobi,
Kenya
 LIST OF PARTICIPANTS

Pham, Van Lap Department o f Genetics,

Faculty o f Biology,
University o f Hanoi,
90 Nguyen Trai,
Hanoi,
Viet Nam

Polesny, F. Federal Agency o f Plant Protection,

Trunnerstrasse 1-5,
A - 1020 Vienna,
Austria

Polito, L.C. Dipartimento di Genetica, Biologia Generale

e Molecolare,
Facolta di Scienze,
Università di Napoli,
Federico II, via Mezzocannone, 8,
1-80134 Naples,
Italy

Rabson, R. O ffice o f Basic Energy Sciences,

ER-17, GTN,
United States Department o f Energy,
Washington, DC 20545,
United States o f America

Reyes Flores, J. Permanent Mission o f M exico,

Tiirkenstrasse 15,
A-1090 Vienna,
Austria

Ridgway, R.L. Beltsville Agricultural Research Center,

Agricultural Research Service,
United States Department o f Agriculture,
Room 108, Building 402, BARC-East,
Beltsville, M D 20705,
United States o f America

Robinson, A.S. Institute o f Molecular Biology and Biotechnology,

P.O. Box 1527,
Heraklion,
G R-71110 Crete,
Greece
6 64 LIST OF PARTICIPANTS

Ro§ca, 1.1. Research Institute for Cereal and Industrial Crops,

Fundulea, 8264, Com. Fundules,
Jud. Cáláragi,
Romania

Russ, K. Federal Agency o f Plant Protection,

Trunnerstrasse 1-5,
A-1020 Vienna,
Austria

Sehnal, F. Faculty o f Biological Sciences,

University o f South Bohemia,
Branisovská 31,
CS-37005 Ceské Budéjovice,
Czechoslovakia

Seth, R.K. Department o f Zoology,

University o f Delhi,
Delhi 110007,
India

Staten, R.T. Phoenix Methods Development Center,

Animal and Plant Health Inspection Service,
United States Department o f Agriculture,
4125 E. Broadway Road,
Phoenix, A Z 85040,
United States o f America

Sutantawong, M. O ffice o f Atomic Energy for Peace,

Vibhavadi Rangsit Road,
Chatuchak, Bangkok 10900,
Thailand

Sutrisno, S. Center for the Application o f Isotopes

and Radiation,
National Atomic Energy Agency,
P.O. Box 7002 JKSKL,
Kebaroran Lama,
Jakarta Selatan 12070,
Indonesia

Tan, Keng-Hong School o f Biological Sciences,

Science University o f Malaysia,
11800 Penang,
Malaysia
 LIST OF PARTICIPANTS

Teranishi, R. Western Regional Research Center,

Agricultural Research Service,
United States Department o f Agriculture,
800 Buchanan Street,
Albany, C A 94710,
United States o f America

Thomas, G.D. United States Department o f Agriculture,

University o f Nebraska,
305 Plant Industry Building,
Lincoln, NE 68583-0938,
United States o f America

Tykva, R. Institute o f Organic Chemistry

and Biochemistry,
Czechoslovak Academy o f Sciences,
Flemingovo 2,.
CS-16610 Prague 6,
Czechoslovakia

Underdal, B.J. Norwegian College o f Veterinary Medicine,

P.O. Box 8146 — Dep.,
N-033 Oslo 1,
Norway

Volf, P. Department o f Parasitology,

Charles University,
Vinicna 7,
CS-12844 Prague 2,
Czechoslovakia

Vulsteke, V. Zoological Institute,

Katholieke Universiteit Leuven,
Naamsestraat 59,
B-3000 Leuven,
Belgium

Waiyaki, B. United Nations Environment Programme, ,

P.O. Box 30552,
Nairobi,
Kenya
666 LIST OF PARTICIPANTS

Wang, Huasong Institute for the Application o f Atomic Energy,

Chinese Academy o f Sciences,
P.O. Box 5109,
100094 Beijing,
China

Yamagishi, M. Okinawa Prefectural Fruit Fly

Eradication Project Office,
123 Maji, Naha,
Okinawa 902,
Japan

Yulo-Nazareá, M .T. Department o f Science and Technology,

Philippine Nuclear Research Institute,
Commonwealth Avenue,
Diliman, Quezon City,
Philippines
 AUTHOR INDEX
Abusowa, М .: 319 Gasperi, G .: 251
Ahm ed, H .: 371 Giordano, E .: 215
Amsler, S.: 619 González, J .R .: 405, 419
Artiaco, D .: 215 G ooding, R .H .: 167, 603
Atkinson, P .W .: 17, 93 Graham, S .H .: 285
Attathom, T .: 125 Graziani, F .: 153, 265
Attathom, S.: 125 Grubhoffer, L .: 557
Bajatta, C .: 313 Guglielmino, C .R .: 251
Bandi, C .: 251 Hamersky, W .: 487
Barbulescu, A l.: 379 Handler, A .M .: 79
Baruffi, L .: 251 Hargrove, J .W .: 549
Bauer, B .: 567, 619 Haymer, D .: 149
Bennettová, В .: 529 Heath, R .R .: 463
Bensaadi, N .: 133 Heather, N .W .: 627
Blanchard, P.: 133 Hendrichs, J.: 453
Blanchetot, A .: 167 Hoedaya, М .: 395
Bloehm, K .A .: 285 Tan, Keng Hong: 495
Bueds, H .: 523 H owells, A .J .: 93
Carlson, C .: 257 Hussain, T .: 371
Carpenter, J.E .: 363 Huybrechts, R .: 115, 121
Christian, P.: 17 Ignatowicz, S .: 649
Coates, C .J.: 93 Inscoe, M .N .: 3
Crampton, J .M .: 33 Irastorza, J .A .: 299
Damiani, G .: 251 Isanont, B .: 125
de Azeredo-Espin, A .M .L .: 161 Itô, Y .: 441
De L oof, A .: 523 James, W .J .: 299
Djiteye, A .: 567 Janssen, I.: 121
D os Santos, P .: 175 Kakinohana, H .: 49
D yck, V .A .: 285 Kaochong, P .: 359
Econom opoulos, A .P .: 331, 473 Karama, S.: 133
Epsky, N .D .: 463 Katsoyannos, B .I.: 453
Feldmann, U .: 567, 579 Keaveny, D .F .: 269
Flath, R .A .: 487 Kenya, E .: 513
Foster, G .G .: 299 Kerremans, P .: 187
Franz, G .: 187 K injo, K .: 49
Furia, М .: 215 Kint, S.: 487
Gaggl, К .: 453 Klassen, W .: 319
García, J.C .: 405, 419 Kohama, T .: 49
Gargiulo, G .: 153 K olste, A .-B .: 257

667
668 AUTHOR INDEX

Kuba, H.: 49, 441 Qureshi, Z.A.: 371

Laughinghouse, A.: 145 Reynolds, K.M.: 487
Leopold, R.A.: 63, 225 Ridgway, R.L.: 3
Light, D.M.: 487 Robben, J.: 115
Limopasmanee, W.: 359 Robinson, A.S.: 101, 235
Lindquist, D.A.: 319 Rosander, R.W.: 269
Luangapichátkul, М.: 359 Ro§ca, I.I.: 379
Magoma, G.N.: 513 Róssler, Y.: 331
Макее, H.: 481 Saccone, G.: 215
Malacrida, A.R.: 251 Salvado, J.C.: 133
Malva, С.: 153, 265 Schoofs, L.: 115
Manoto, Е.С.: 641 Sehgal, S.S.: 427
Maree, F.: 243 Sehnal, F.: 537
Martínez, S.J.: 313 Seth, R.K.: 427
Matioli, S.R.: 175 Sokei, Y.: 49
Maudlin, I.: 195 Staten, R.T.: 269
Mclnnis, D.: 149 Sutantawong, М.: 359
McKenzie, L.J.: 299 Sutrisno, S.: 395
Mende, H.A.: 93 Teranishi, R.: 487
Mialhe, E.: 145 Thorpe, K.W.: 3
Milani, R.: 251 Torti, C.: 251
Miller, L.H.: 145 Traut, W.: 243
Miyataka, T.: 201 Tremblay, E.: 153
Morgante, J.S.: 175 Tykva, R.: 529
Mouches, С.: 133 Van der Vloedt, A.: 567
Muska, М.: 557 Van Brussel, М.: 115, 121
Nakamoto, Y.: Varricchio, P.: 153
Nitzan, Y.: 331 Vito, P.: 215
O’Brochta, D.A.: 93 Volckaert, G.: 115
Oakeshott, J.: 17 Volf, P.: 557
Offori, E.D.: 345 Vreysen, M.J.B.: 567
Opp, S.B.: 487 Vulsteke, V.: 115, 121
Ortega, J.: 313 Vundla, M.W.: 513
Osir, E.O.: 513 Wang, Huasong: 505
Paschalidis, K.M.: 299 Welbum, S.C.: 195
Peluso, I.: 215 Weller, G.L.: 299
Pennacchio, F.: 153 Wornoayporn, V.: 453
Pham, Van Lap: 235 Yamagishi, М.: 49, 201, 441
Pinkerton, A.C.: 93 Yulo-Nazarea, М.T.: 641
Polito, L.C.: 215 Zhang, Heqin: 505
Руске, C.: 523
 INDEX OF PAPERS AND POSTERS BY NUMBER
Papers
IA E A -S M -3 2 7 / Page IA E A -S M -3 2 7 / Page

1 ................................... .............. 3 33 ................................... .............. 473

2 ................................... .............. 17 35 ................................... .............. 363
4 ................................... .............. 49 36 ................................... .............. 371
5 ................................... .............. 33 37 ...................................
.............. 379
6 ................................... .............. 70 38 ................................... .............. 395
7 ................................... .............. 93 39 ................................... ............. 405
8 ................................... .............. 101 40 ................................... .............. 427
10 ................................... .............. 115 41 ................................... .............. 441
11 ................................... .............. 121 42 ................................... .............. 453
12 ................................... .............. 125 43 ................................... .............. 463
13 ................................... .............. 133 44 ................................... .............. 487
14 ................................... .............. 145 45 ................................... .............. 495
15 ................................... .............. 149 46 ................................... .............. 257
16 ................................... .............. 153 47 ................................... .............. 505
17 ..........................:........ .............. 161 48 ................................... .............. 513
18 ................................... .............. 167 49 ................................... .............. 523
19 ................................... .............. 175 50 ................................... .............. 529
20 ................................... .............. 187 51 ................................... .............. 537
21 ................................... .............. 195 52 ................................... .............. 549
22 ................................... .............. 201 53 ................................... .............. 557
23 ................................... .............. 215 54 ................................... .............. 567
24 ................................. .............. 225 55 ................................... .............. 579
25 ................................... .............. 235 56 ................................... .............. 419
26 ................................... .............. 243 57 ................................... .............. 603
27 ................................... ............... 63 58 ................................... .............. 627
28 ................................... .............. 269 60 ................................... .............. 641
29 ................................... .............. 285 66 ................................... .............. 251
30 ................................... .............. 299 70 ................................... .............. 331
31 ................................... .............. 313 71 .............................. .............. 345
32 ................................... .............. 319 77 .............................. .............. 619

Posters
IA E A -S M -3 2 7 / Page IA E A -S M -3 2 7 / Page

IP .......................................... 481 19P ............................................ 649

16P ......................................... 359 23P ............................................ 265

669
 HOW TO ORDER IAEA PUBLICATIONS
☆ ☆ In the United States o f Am erica and Canada, the exclusive sales agent for
IAEA publications, to whom all orders and inquiries should
be addressed, is:
UNIPUB, 4611-F Assembly Drive, Lanham, MD 20706-4391, USA

☆ ☆ In the follow ing countries IAEA publications may be purchased from the sources
listed below, or from major local booksellers. Payment may
be made in local currency or with UNESCO coupons.

ARGENTINA Com isión Nacional de Energía Atóm ica, A venida del Libertador 8250,
RA-1429 Buenos Aires
AUSTRALIA H unter Publications, 58A Gipps Street, C ollingw ood, V ictoria 3066
BELGIUM Service C ourrier UNESCO, 202, Avenue du Roi, B-1060 Brussels
CHILE Com isión C hilena de Energía Nuclear, Venta de Publicaciones,
A m unategui 95, C asilla 188-D, Santiago
CHINA IAEA Publications in C hinese:
C hina N uclear Energy Industry Corporation, Translation Section,
P.O. Box 2103, Beijing
IAEA Publications o ther than in Chinese:
C hina National Publications Im port & E xport C orporation,
D eutsche A bteilung, P.O. Box 88, Beijing
FRANCE O ffice International de Docum entation et Librairie, 48, rue Gay-Lussac,
F-75240 Paris C edex 05
GERMANY UNO -Verlag, Vertriebs- und Verlags Gm bH , Dag H am m arskjold-Haus,
Poppelsdorfer Allee 55, D-53115 Bonn
HUNGARY Librotrade Ltd., Book Im port, P.O. Box 126, H-1656 B udapest
INDIA Oxford Book and S tationery Co., 17, Park Street, C alcutta-700 016
Oxford Book and S tationery Co., Scindia House, New Delhi-110 001
ISRAEL YO ZM O T Literature Ltd., P.O. Box 56055, IL-61560 Tel Aviv
ITALY Librería S cientifica Dott. Lucio di Biasio “ A E IO U ",
Via Coronelli 6, 1-20146 Milan
JAPAN Maruzen C om pany, Ltd, P.O. Box 5050, 100-31 Tokyo International
NETHERLANDS M artinus N ijhoff International, P.O. Box 269, NL-2501 AX The Hague
Swets and Z e itlin g e r b.v., P.O. Box 830, NL-2610 SZ Lisse
PAKISTAN M irza Book Agency, 65, Shahrah Quaid-e-Azam , P.O. Box 729, Lahore 3
POLAND Ars Polona, Foreign Trade Enterprise,
Krakow skie Przedm ieácie 7, PL-00-068 W arsaw
ROMANIA llexim , P.O. Box 136-137, Bucharest
RUSSIAN FEDERATION M ezhdunarodnaya Kniga, Sovinkniga-EA,
Dim itrova 39, SU-113 095 Moscow
SLOVAK REPUBLIC Alfa Publishers, H urbanovo námestíe 3, SQ-815 89 Bratislava
SOUTH AFRICA Van Schaik Bookstore (Pty) Ltd, P.O. Box 724, Pretoria 0001
SPAIN Díaz de Santos, Lagasca 95, E-28006 M adrid
Díaz de Santos, Balm es 417, E-08022 Barcelona
SWEDEN AB Fritzes Kungl. H ovbokhandel, Fredsgatan 2, P.O. Box 16356,
S-103 27 Stockholm
UNITED KINGDOM HMSÓ, P ublications C entre, A gency Section,
51 Nine Elm s Lane, London SW 8 5DR
YUGOSLAVIA Jugoslovenska Knjiga, Terazije 27, P.O. Box 36, YU-11001 Belgrade

☆ ☆ Orders (except for customers in Canada and the USA) and requests for information
may also be addressed directly to:
Sales and Promotion Unit
ÿ International Atomic Energy Agency
£ Wagramerstrasse 5, P.O. Box 100, A-1400 Vienna, Austria
INTERNATIONAL ATOMIC ENERGY AGENCY
VIENNA
ISBN 92-0-000293-5
ISSN 0074-1884
INTERNATIONAL ATOMIC ENERGY AGENCY
VIENNA
ISBN 92-0-000293-5
ISSN 0074-1884