J Biocontrol 2007 05 014

Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

Biological Control 42 (2007) 336–344

www.elsevier.com/locate/ybcon

Screening rhizobacteria for biological control of Ralstonia


solanacearum in Ethiopia
F. Lemessa *, W. Zeller
Institute for Biological Control, Federal Biological Research Centre for Agriculture and Forestry (BBA), Heinrichstr. 243,
64287 Darmstadt, Germany

Received 23 February 2007; accepted 21 May 2007


Available online 29 May 2007

Abstract

Bacterial wilt caused by Ralstonia solanacearum (Smith) has become a severe problem mainly on potato and tomato in Ethiopia and
no effective control measure is available yet. To explore possibilities for the development of biological control for the disease, 118 rhi-
zobacteria, most of them collected from Ethiopia, were screened against an Ethiopian R. solanacearum strain. On the basis of in vitro
screening, six strains (RP87, B2G, APF1, APF2, APF3, and APF4) with good inhibitory effect were selected for in planta testing in a
greenhouse. In the greenhouse, soil and tomato seedlings were treated with the antagonists and their effects studied. The study showed
that APF1 and B2G strains significantly reduced disease incidence and increased weight of tomato plants. Area under disease progress
curves (AUDPC) was reduced by 60% and 56% in plants inoculated with APF1 and B2G strains, respectively. Plant dry weight increase
in plants inoculated with APF1 and B2G strains was 96% and 75%, respectively. APF1 was found to be the most beneficial strain in
disease suppression and also growth promotion resulting in 63% dry weight increase compared to untreated control. The study revealed
that APF1 and B2G strains are promising strains whose effectiveness under field conditions and their mode of action should be
investigated.
 2007 Elsevier Inc. All rights reserved.

Keywords: Rhizobacteria; Biocontrol; Bacterial wilt; Ralstonia solanacearum; Bacillus subtilis; Streptomyces setonii; Fluorescent pseudomonads

1. Introduction rotation, and use of bactericides have met with only limited
success (Ciampi-Panno et al., 1989). Although disease
Ralstonia solanacearum (Smith) (Yabuuchi et al., 1995) resistance is an important component of integrated disease
is an important soilborne bacterial plant pathogen with a management, it is generally agreed that breeding for resis-
worldwide distribution and a wide host range of more than tance is not completely effective, producing only modest
200 species in 50 families (Hayward, 1991). Some of its eco- gains and often lacking stability and/or durability (Hay-
nomically important plant hosts include tomato, potato, ward, 1991; Boucher et al., 1992). Furthermore, the high
eggplant, pepper, tobacco, banana, chilli, and peanut variability of strains of R. solanacearum (Elphinstone,
(French and Sequeira, 1970). In Ethiopia R. solanacearum 1992) combined with the influence of environmental factors
is an important disease of potato and tomato (Yaynu, on host–pathogen interactions (Hayward, 1991) often
1989) and its importance is increasing from time to time. restricts the expression of resistance to specific regions.
To date, no effective control method has been developed Various recent studies have indicated that biological con-
for this wilt disease. Plant breeding, field sanitation, crop trol of bacterial wilt disease could be achieved using antag-
onistic bacteria (McLaughlin et al., 1990; Ciampi-Panno
*
et al., 1989). Toyota and Kimura (2000) have reported the
Corresponding author. Present address: Department of Crop Sciences, suppressive effect of some antagonistic bacteria on R. solan-
College of Agriculture and Veterinary Medicine, Jimma University, P.O.
Box 307, Jimma, Ethiopia. Fax: +251 471110934.
acearum. Moreover, Ciampi-Panno et al. (1989) has proved
E-mail address: lemessaf@yahoo.com (F. Lemessa). the use of antagonistic pathogens to be effective in control

1049-9644/$ - see front matter  2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.biocontrol.2007.05.014
F. Lemessa, W. Zeller / Biological Control 42 (2007) 336–344 337

of R. solanacearum under field condition. Potential biolog- ous study (Lemessa and Zeller, 2007), a representative
ical agents used to control bacterial wilt caused by R. solan- strain IBC Pot 4JU was used in the screening study. The
acearum include avirulent mutants of R. solanacearum strain was originally isolated from wilted potato from
(Dong et al., 1999), genetically engineered antagonistic bac- Jimma, Ethiopia, and identified as R. solanacearum by
teria (Kang et al., 1995), and some naturally occurring tomato bioassay and species specific PCR primers (759/
antagonistic rhizobacteria such as Bacillus spp. (Silveira 760). IBC Pot 4JU belongs to biovar II as determined
et al., 1995), Pseudomonas spp. (Guo et al., 2001), and according to Hayward (1964) and race 3 according to the
Streptomyces spp. (El Albyad et al., 1996). However, cur- standard set of Buddenhagen et al. (1962). It is pathogenic
rent interest in the possible release of genetically modified to potato, tomato, and eggplant but not to pepper and
microorganisms into the environment has raised concerns tobacco (Lemessa and Zeller, 2007). The strain was rou-
over issues of environmental health. Thus naturally occur- tinely cultured on casamino acids peptone glucose (CPG)
ring microorganisms remain the potential candidates. agar (Smith et al., 1995) and on tetrazolium chloride
Soilborne diseases have been controlled more recently (TTC) agar (Kelman, 1954) at 28 C for 48 h and tempo-
by means of certain beneficial bacteria that are indigenous rarily stored in sterile water at room condition.
to the rhizosphere of plants (Thomshaw, 1996). The rhizo-
sphere, representing the thin layer of soil surrounding plant 2.3. In vitro antagonistic activity
roots and the soil occupied by the roots, supports large and
metabolically active groups of bacteria (Villacieros et al., In vitro antagonism studies between rhizospheric bacte-
2003) known as plant growth promoting rhizobacteria ria and the pathogenic strain of R. solanacearum IBC Pot
(PGPR) (Kloepper et al., 1980). PGPR are known to rap- 4JU were carried out on KB and LB agar plates using chlo-
idly colonize the rhizosphere and suppress deleterious roform vapour (Ryan et al., 2004) and agar diffusion meth-
microorganisms as well as soilborne pathogens at the root ods (Mitchell and Carter, 2000).
surface (Rangajaran et al., 2003). These organism can also
be beneficial to the plant by stimulating growth (Bloemberg 2.3.1. Chloroform vapour method
and Lugtenberg, 2001). Candidate antagonistic bacteria were spotted on KB
Therefore, the aim of this study was to isolate rhizobac- and LB media and incubated for 48 h at 28 C. After
teria and screen them for in vitro and in planta antagonistic 48 h of incubation, the growing antagonists were killed
activity against R. solanacearum for biological control in by inverting Petri dishes over chloroform for 3 min. The
Ethiopia. test strain was cultured in CPG broth on rotary shaker
for 24 h and centrifuged at 10,000 rpm and cell pellets were
2. Materials and methods diluted in 0.85% (w/v) NaCl solution and adjusted to
108 cfu/ml. The plates in which antagonists were grown
2.1. Isolation of potential antagonistic bacteria were flooded with 2 ml cell suspension of the pathogen,
dried, and incubated for two or more days. The effective-
A total of 98 bacteria were collected from rhizosphere of ness of strains was evaluated by measuring the inhibition
potato, tomato, pepper, coffee, and maize plants from Jimma zones around antagonistic bacteria. The experiment was
and its surroundings in Ethiopia during March to May 2005. performed with a completely randomized design with three
For isolation from the rhizosphere, plant roots were gently replications and repeated twice.
washed twice in sterile water to remove adhering soil, and
then root sections of approximately 1 g were added to 2.3.2. Agar-diffusion test
200 ml sterile water in flasks and shaken on rotary shaker One-hundred microliters of R. solanacearum suspension
at 150 rpm for 30 min. Then serial dilutions of the root sus- containing 108 cfu/ml was spread on KB and LB plates and
pension were plated on King’s medium B agar (KB) (King four holes of 9 mm diameter punched into the agar. In
et al., 1954) and Luria-Bertani agar (LB) (Sambrook et al., these holes 30 ll suspension of each test antagonist
1989) and incubated at 28 C for 48 h. When the bacterial (109 cfu/ml) was added and the plates incubated at
colony appeared on the medium, representative isolates were 28 C for 48 h. Inhibition of R. solanacearum growth was
picked for antagonism study. In addition to the rhozospheric assessed by measuring the radius of inhibition zone (mm)
bacteria isolated from Ethiopia, 20 species of bacterial after incubation for 48 h at 28 C.
strains, which were isolated mainly from the rhizosphere of
potato, were procured from Rostock University, Germany, 2.4. Greenhouse evaluation of rhizobacterial isolates
and included in the screening study. For long-term preserva-
tion, bacteria were stored in 20% glycerol at 70 C. 2.4.1. Growth of plants
A potting medium that constitutes a mixture (1:3) of
2.2. Bacterial strain and culture conditions sand: commercial potting substrate (FRUHSTORFER
ERDE Typ LD 80; Industrie-Erdenwerk Archut, Lauter-
As the majority of Ethiopian R. solanacearum strains bach, Germany) was sterilized at 121 C for 20 min and
were identified as biovar II race three groups in our previ- filled in a sterilized plastic tray. Tomato c.v ‘Matina’ seed
338 F. Lemessa, W. Zeller / Biological Control 42 (2007) 336–344

was obtained from the HILD Samen GmbH, Marbach, where DC is disease of control and DT is disease of the
Germany. Seeds were surface sterilized with 2% sodium treatment group.
hypochlorite for 2 min (Guo et al., 2004), washed thor-
oughly with sterilized water and planted in plastic tray 2.4.4. Population dynamics of R. solanacearum in soil
filled with the sterilized potting medium. The plants were treated with antagonists
maintained in a greenhouse at temperatures of 24–28 C To assess the effect of antagonists on the population
and 75–90% relative humidity and seedlings were watered density of R. solanacearum in soil, 2 g of pathogen–antag-
with sterile water when necessary. onist infested soil samples were taken from each pot of
treatments at different intervals (0, 3, 6, and 9 weeks after
treatment) giving 12 g of soil per treatment. The soil was
2.4.2. Bioassay mixed thoroughly, and then 1 g was added to sterile dis-
Six strains of bacteria with the greatest inhibition in tilled water (1:9, wt/vol) and shaken for 30 min on a rotary
in vitro test were further tested in greenhouse on tomato shaker, serial dilutions were made, and 0.1 ml aliquots were
plants to evaluate their ability to control bacterial wilt in spread on the surface of a semi-selective SMSA medium
planta. For this purpose, the same potting mix used for (Englerbrecht, 1994). The medium is made of: 10 g bacto
raising tomato seedlings above was used after autoclaving. peptone (Difco), 5 ml glycerol, 1 g casamino acid (Difco),
The pathogen was prepared by culturing in CPG broth for 15 g bacto agar (Difco), and 1 L distilled water. This med-
48 h at 28 C and 150 rpm on rotary shaker. Cultures were ium was supplemented with 1% polymyxin B sulphate
centrifuged at 10,000 rpm (Beckman J2-21 M/E centrifuge, (Sigma), 1% crystal violet, 1% tetrazolium salt (Sigma),
USA) for 10 min at 10 C. Bacterial pellets were suspended 1% bacitracin (Sigma), 0.1% penicillin (Sigma), 1% chl-
in distilled water and adjusted to 108 cfu/ml. Four-hundred oramphenol (Sigma), and 1% cycloheximide (Sigma). After
grams of the sterilized potting medium was mixed with incubating plates at 28 C for 3 days, colonies of R. solan-
75 ml of R. solanacearum IBC Pot 4JU at 108 cfu/ml acearum were counted and cfu were calculated per gram
(1.2 · 106 cfu/g soil, dry weight) and placed in 12 cm (dry weight) of potting medium. Four replicates were pre-
diameter pots. One week after incorporation of the patho- pared for each sample. Since populations of bacteria
gen into the soil and one day before transplanting the test approximate a log normal distribution (Loper et al.,
plants, antagonists were incorporated in the soil at a rate of 1984), values were log transformed before analysis to nor-
50 ml per pot (400 g) at 109 cfu/ml. On the next day, four malize variance.
weeks old tomato seedlings raised in plastic tray were root
dipped in suspension (109 cfu/ml) of antagonistic bacteria 2.4.5. Plant weight and growth promotion assessment
for 60 min and transplanted into pathogen–antagonist mix- At the end of the experiment (2 months after transplant-
ture soil. Plants were kept in the greenhouse at 24–28 C ing), plants including the roots were harvested from the
and 75–90% relative humidity in 12 h light and 12 h dark pots and fresh weight recorded. Healthy plants were
conditions. The experiment was conducted two times with counted and uprooted separately and their weights
completely randomised design. Treatments were replicated recorded to measure growth promotion, compared with
five times with 12 plants per replication. the untreated control (Lim and Kim, 1997). For dry weight
measurement, plants were dried in an oven at 60 C for 3
2.4.3. Disease assessment days and weights evaluated for each treatment. Dry and
The percentage of diseased plants was scored separately fresh weights were used for data analysis.
per pot at different time points based on wilting symptoms
and the area under disease progress curves (AUDPC) was 2.4.6. Statistical analysis
calculated according to the mid point rule (Garrett and The data were subjected to analysis of variance using
Mundt, 2000) as: SAS version 8 (SAS Institute, 1999). Single and interaction
effects of factors were determined using the GLM proce-
X
n1
dure of SAS. Correlation analysis was carried out using
AUDPC ¼ ½0:5ðxi þ xiþ1 Þ½tiþ1  ti ;
the CORR procedure. Whenever significant interactions
i¼1
were observed between factors, the level of one factor
where xi is the percentage of disease incidence at ith assess- was compared at each level of the other factor. Mean val-
ment, ti is the time of the ith assessment in days from the ues among treatments were compared by the Tukey’s test
first assessment date and n is the total number of days at a = 0.05 level of significance.
disease was assessed. Because incidence (x) was expressed
in per cent and time (t) in days, AUDPC was expressed 2.5. Identification of the bacterial isolates
in %-days (Campbell and Madden, 1990). Biological
control efficacy was calculated according to Guo et al. Six bacterial strains with the highest inhibitory effect
(2004) as: in vitro were selected and used during the study. Two of
them were Streptomyces setonii RP87 (Millard and Burr)
BCE ¼ ½ðDC  DT Þ=DC   100%; and Bacillus subtilis B2G (Ehrenberg) that were obtained
F. Lemessa, W. Zeller / Biological Control 42 (2007) 336–344 339

from Rostock University, Germany, and included in our (Table 1). APF1 and APF2 exhibited significantly greater
study. The remaining four were strains from the collection inhibition on KB than on LB. Inhibitory effect of strains
from Ethiopia and taxonomically differentiated on the RP87, B2G, APF3, and APF4 was not significantly affected
basis of their reactions to standard biochemical tests from by type of media.
Bergey’s Manual of Systematic Bacteriology (Krieg and
Holt, 1984): pigmentation on KB medium, Gram stain, 3.2. Greenhouse experiment
oxidase and catalase test, starch hydrolysis, gelatine lique-
faction, and growth on carbon sources (fructose, sucrose, The in planta efficacy of selected antagonists for the con-
glucose, and galactose). trol of R. solanacearum wilt in tomato plants was evaluated
under greenhouse conditions and there were no significant
differences until 45 days after transplanting (Fig. 1). At 45
3. Results
days after transplanting, B. subtilis B2G and the pseudo-
monad APF1 and APF2 strains significantly reduced wilt
3.1. In vitro assays
incidence (P < 0.0001) compared to the control. At 57 days
after transplanting, only B. subtilis B2G and pseudomonad
A total of 118 rhizospheric bacteria were screened
APF1 strains significantly reduced disease incidence
against R. solanacearum strain IBC Pot 4JU and 23 strains
(P = 0.0007) compared to the control. Streptomyces setonii
had inhibitory effect that ranged from an average of
RP87 and the pseudomonad APF2 and APF3 did not sig-
0.5–11 mm radius of inhibition zone. Among those 23
nificantly differ from the control throughout the experi-
strains, 16 (70%) were fluorescent under UV light when cul-
ment. Generally, the pseudomonad APF1 and B. subtilis
tured on KB medium. Six strains with inhibition zone of
B2G strains reduced disease incidence at 45 days after
5–11 mm radius were selected and used for further study.
transplanting by 70% and 63% and at 57 days after trans-
Two of the strains were B. subtilis B2G and Streptomyces
planting by 53% and 52%, respectively.
setoni RP87 which were from Rostock University, Ger-
Wilt incidence in the form of AUDPC was also signifi-
many, and four (APF1, APF2, APF3, and APF4) were
cantly affected by treatments. Plants treated with B. subtilis
from the collection of Ethiopia; all were identified as fluo-
B2G, and pseudomonad APF1 and APF2 strains sustained
rescent pseudomonads based on the Bergey’s Manual. The
significantly lower AUDPC (P = 0.0022) compared to con-
four Ethiopian strains are currently maintained in the lab-
trol (Table 2). Though the strain S. setonii RP87 showed
oratory of Plant Pathology at College of Agriculture and
very high inhibition in in vitro test, it failed to reduce
Veterinary Medicine, Jimma University, Ethiopia.
disease in the in vivo biocontrol assay. Generally, poor
The results of in vitro tests indicated significant differ-
ences among antagonistic strains and between media (LB
and KB) used (Table 1). Methods of in vitro test (chloro-
form vapour and agar diffusion) did not significantly affect
60
*
50
growth inhibition. The two-way interaction, medium by * APF1
Wilt incidence (%)

antagonist, was significant (P < 0.0001), indicating that 40


the strains differed in inhibiting the pathogen depending B2G
ns
on type of medium. On LB medium, strains RP87 and 30 ns APF2
B2G exhibited significantly greater inhibition zones than APF3
the other four strains, while on KB significantly greater 20 APF4
inhibition was recorded by RP87, B2G, APF1, and APF2 10
RP87
Control
Table 1 0
In vitro inhibition (mm) of growth of Ralstonia solanacearum strain by 21 33 45 57
antagonistic bacterial isolates (Streptomyces setonii RP87, Bacillus subtilis Days after transplanting
B2G, and fluorescent pseudomonads APF1, APF2, APF3, and APF4) in
chloroform vapour and agar diffusion methods on Luria-Bertani (LB) and Fig. 1. Bacterial wilt symptom development expressed as wilt incidence on
King’s B (KB) media incubated at 28 C for 48 h in laboratory tomato plants treated by different bacterial antagonists (Streptomyces
setonii RP87, Bacillus subtilis B2G, and fluorescent pseudomonads APF1,
Antagonistic strains Chloroform vapour Agar diffusion
APF2, APF3, and APF4). The root system of 4-week-old plants were
LB KB LB KB dipped in 109 cfu/ml antagonistic bacterial suspensions or sterile water
RP87 10.7 aA 10.3 aA 10.0 aA 10.2 aA (control) for 60 min, then transplanted into autoclaved soil artificially
B2G 10.3 aA 10.0 aA 10.3 aA 10.1 aA infested with the pathogen (108 cfu/ml) and bacterial antagonists (109 cfu/
APF1 6.0 bB 12.1 aA 6.0 bB 12.3 aA ml) and grown in the greenhouse at 25–28 C for 60 days. Incidence of the
APF2 5.3 bB 13.0 aA 4.0 bB 12.0 aA disease was calculated by counting the number of plants showing typical
APF3 4.1 bB 6.0 bB 4.2 bB 6.7 bB diseases symptoms at 21, 33, 45, and 57 days after transplanting. Control
APF4 3.7 bB 5.3 bB 3.7 bB 5.7 bB plants grown in non-infested soil were free of symptoms and not included
in statistical analysis. ns and * indicate absence and presence of significant
Means within column (row) followed by the same lower (upper) case differences (Tukey’s test, a = 0.05) among treatments at a particular day
letters are not significantly different (Tukey, a = 0.05). after transplanting, respectively.
340 F. Lemessa, W. Zeller / Biological Control 42 (2007) 336–344

correlation (data not shown) was found between in vitro 7


antagonism and in planta suppression of disease. Biocon-
ns
trol efficacy as estimated by AUDPC ranged from 12.6% 6 a a

Population (log cfu/g) dry wt.)


ns
to 60.3% and the highest biocontrol efficacy was recorded ab
5 ab
by pseudomonad APF1 followed by B. subtilis B2G and
ab
pseudomonad APF2 (Table 2). 4
ab

Population densities of R. solanacearum (log cfu/g, dry


b RP87
weight) in the soil after incorporation of antagonists into 3
b B2G
potting medium is shown in Fig. 2. Up to 3 weeks after APF1
incorporation of antagonists into soil, population density 2 APF2
APF3
of R. solanacearum declined in all treatments including APF4
1
the control. Later, populations of the pathogen started to Control

increase, however, differently among the treatments. Six 0


weeks after soil treatment, population of the pathogen with 0 3 6 9
pseudomonad APF1 and B. subtilis B2G were significantly Weeks after treatment with antagonists
(P = 0.0018) lower than the control. However, 9 weeks Fig. 2. Changes in population density of Ralstonia solanacearum in the
after soil treatment population of the pathogen was signif- rhizosphere of tomato at different intervals after introduction of bacterial
icantly (P = 0.0002) lower only in soil treated with pseudo- antagonists (Streptomyces setonii RP87, Bacillus subtilis B2G, and
monad APF1 strain. Generally, per cent reductions of R. fluorescent pseudomonads APF1, APF2, APF3, and APF4) into the
potting medium under greenhouse conditions at 25–28 C. Potting
solanacearum population with pseudomonad APF1 and medium treated only with the pathogen served as control. Means with
B. subtilis B2G strains were 44% and 45%, respectively, 6 the same letter are not significantly different (Tukey’s test, a = 0.05). ns, no
weeks post-treatment and 35% and 26%, respectively, at significant differences among treatments.
9 weeks.
There were significant differences among treatments for
both dry and fresh weight of tomato plants (Fig. 3a and b).
100
Average dry weight of tomato was significantly higher Dry weight
(P < 0.0001) for plants treated with pseudomonad APF1 % Increase
80
and B. subtilis B2G and average fresh weight was signifi-
Dry wt. (g) / % increase

a
cantly higher (P < 0.001) for plants treated with strains 60
ab
abc
B. subtilis B2G and pseudomonad APF1 and APF2. High-
bcd bcd
est dry and fresh weight increase was recorded with pseu- 40 d
dc

domonad APF1 treatment: 96% and 81%, respectively.


Dry and fresh weight for plants treated with B. subtilis 20
B2G were 75% and 54% higher than the control, respec-
tively. No significant difference was observed between con- 0
RP87 B2G APF1 APF2 APF3 APF4 Control

-20
Antagonistic strains
Table 2
Effect of antagonistic bacterial strains (Streptomyces setonii RP87, Bacillus 450
a Fresh wt.
subtilis B2G, and fluorescent pseudomonads APF1, APF2, APF3, and 400 % Increase
Fresh wt. (g) / % increase

ab
APF4), on Ralstonia solanacearum wilt incidence expressed as area under 350 abc

disease progress curves (AUDPC) in tomato plant under greenhouse 300 bcd
dc
conditions
250 d d
Straina AUDPC (%-days) Biocontrol efficacy (%)
200
RP87 983.3 ab 12.6 150
B2G 493.6 b 55.6
100
APF1 441.6 b 60.3
APF2 499.9 b 50.0 50
APF3 632.6 ab 43.1 0
APF4 949.9 ab 14.5 RP87 B2G APF1 APF2 APF3 APF4 Control

Control 1111.6 a — Antagonistic strains


a
The root system of 4-week-old plants were dipped in 109 cfu/ml Fig. 3. Effect of antagonistic bacterial strains (Streptomyces setonii RP87,
antagonistic bacterial suspensions or sterile water (control) for 60 min, Bacillus subtilis B2G, and fluorescent pseudomonads APF1, APF2, APF3,
then transplanted into autoclaved soil artificially infested with the and APF4) on dry (a) and fresh (b) weight of tomato plants and per cent
pathogen (108 cfu/ml) and bacterial antagonists (109 cfu/ml) and grown increase compared to control under greenhouse conditions at 25–28 C.
in the greenhouse at 25–28 C for 60 days. Control plants grown in non- Dry weight was recorded after plants were dried in an oven at 60 C for 3
infested soil were free of symptoms and not included in statistical analysis. days. Data are the mean of five replicates with 12 plants per replications.
Means in column followed by the same letter are not significantly different Bars with the same letter are not significantly different according to
(Tukey’s test, a = 0.05). Tukey’s test (a = 0.05).
F. Lemessa, W. Zeller / Biological Control 42 (2007) 336–344 341

trol and S. setonii RP87 and pseudomonad APF3 and associated bacteria were isolated from the rhizospheres
APF4 strains in dry and fresh weights. Simple correlation mainly of potato, tomato, and pepper with the objective
analysis showed that there were negative and significant of selecting efficient antagonists against soilborne infection
associations between disease AUDPC and tomato dry with R. solanacearum, the causal agent of bacterial wilt.
and fresh weights. AUDPC was associated with fresh and In in vitro screening on LB and KB media, six strains
dry tomato weights with correlation coefficients r = 0.57 (S. setonii RP87, B. subtilis B2G, and fluorescent pseudo-
(P < 0.0001) and r = 0.34 (P = 0.0039) (n = 70), monads APF1, APF2, APF3, and APF4) with the highest
respectively. inhibitory effect were selected for further greenhouse condi-
The plant growth promotion efficiency of antagonistic tions. The in vitro test showed that the type of media used
isolates monitored by measuring plant biomass (fresh and affects expression of inhibition by these antagonists.
dry weight) showed variation among plants treated with Accordingly, the antagonistic strains S. setonii RP87,
antagonists and the untreated control. There were signifi- B. subtilis B2G, and fluorescent pseudomonad APF3 and
cantly higher dry (P = 0.0185) and fresh (P = 0.0273) APF4 showed similar results on both LB and KB media,
weights of tomato plants treated by pseudomonad APF1 while the fluorescent pseudomonad APF1 and APF2
compared with the untreated control (Table 3). However, strains produced significantly higher inhibition in KB as
there was no significant difference between control and other compared to LB (Table 1). This indicated that the mecha-
antagonistic strains. Treatment with pseudomonad APF1 nism of inhibition of pseudomonad APF1 and APF2
resulted in 63% and 72% increases in dry and fresh weights strains was affected by the constituents of the media. It is
of tomato, respectively, compared to the untreated control. known that LB is an iron rich medium (Hassett et al.,
1995), while KB is a medium deficient of iron (Lim and
3.3. Identification of bacterial strains Kim, 1997). Thus the main increase in inhibitory activity
of pseudomonad strains APF1 and APF2 in KB medium
For the antagonism study, six strains were used. Two of could be siderophore production that inhibited R. solana-
them were identified species S. setonii RP87 and B. subtilis cearum growth. This conclusion is supported by the obser-
B2G. Identification of the remaining four strains (APF1, vation that amendment of KB medium with iron reduced
APF2, APF3, and APF4) was carried out and all were the inhibitory activity of pseudomonad strains APF1 and
identified as fluorescent pseudomonads on the basis of APF2 (data not shown). This is in agreement with other
standard biochemical methods. They were Gram positive, observations that inhibition of growth on KB could be a
fluorescent under UV light, and oxidase, catalase, glucose, consequence of production of siderophores by Pseudomo-
fructose, sucrose, and galactose positive. Moreover, they nas strains (Lim and Kim, 1997; Blanco et al., 2004). The
liquefy gelatine but do not hydrolysis starch. Originally, type of culture medium strongly affects activity of antago-
RP87, B2G, APF1, and APF3 strains were isolated from nists by mediating production of substances that are
potato, APF2 from tomato, and APF4 from pepper responsible for inhibition (Montesinos et al., 1996; Chen
rhizospheres. et al., 2003). For instance, Duffy and Défago (1999) noticed
that addition of zinc increases antibiotic synthesis in pseu-
4. Discussion domonads. Also the yields of exotoxin A in P. aeruginosa
cultures were influenced by the concentration of iron in
Root associated bacteria are an important functional the culture medium (Bjorn et al., 1978). When the iron
group of beneficial bacteria used for control of soilborne concentration of the culture media was increased from
pathogens and plant growth promotion (Gamalero et al., 0.05 to 1.5 lg/ml, there was at least a 90% decrease in
2003; Rajkumar et al., 2005). In this investigation, root exotoxin A.

Table 3
Effect of antagonistic bacterial isolates (Streptomyces setonii RP87, Bacillus subtilis B2G, and fluorescent pseudomonads APF1, APF2, APF3, and APF4)
on tomato growth response (as dry and fresh weight) as compared to untreated control under greenhouse conditions
Strain Average dry weight Increase compared Average fresh weight Increase compared
per plant (g) to control (%) per plant (g) to control (%)
RP87 3.9 ab 3 30.8 bc 9
B2G 4.0 ab 5 38.7 bc 37
APF1 6.2 a 63 48.5 a 72
APF2 4.7 ab 24 42.0 abc 49
APF3 4.2 ab 11 35.2 bc 25
APF4 4.4 ab 15 32.6 bc 16
Control 3.8 b — 28.2 c —
Healthy plants from treatments that received antagonists and Ralstonia solanacearum were counted and uprooted separately and their weights compared
with the untreated control that received no antagonists and Ralstonia solanacearum. For dry weight measurement, plants were dried in an oven at 60 C for
3 days and weights evaluated for each treatment. Values within the same column followed by the same letter do not differ significantly (a = 0.05) according
to Tukey’s mean separation test.
342 F. Lemessa, W. Zeller / Biological Control 42 (2007) 336–344

On the other hand, the inhibitory activity of the strains population in the soil followed by B. subtilis B2G com-
S. setonii RP87, B. subtilis B2G, and pseudomonads APF3 pared to the control. The decline in population of the path-
and APF4 was equal in LB and KB (Table 1), and amend- ogen at the early stage cannot be attributed to the effect of
ment of KB with iron did not alter their activity signifi- treatments but to other factors such as death of some pro-
cantly. From these results it may be possible to speculate portion of the population when introduced to soil from the
that inhibitory activities of S. setonii RP87, B. subtilis laboratory. The population reduction observed in treat-
B2G, and pseudomonads APF3 and APF4 were not iron- ments with pseudomonad APF1 and B. subtilis B2G strains
dependent. Thus antibiotic production may play an active may be due to their antagonistic effects.
role in the inhibition of the pathogen by these strains. Ran Disease development on plants (Fig. 1) was closely
et al. (2005) suggested that when an antagonist is equally related to population dynamics of R. solanacearum in soil
effective in the absence and presence of iron, antibiotics (Fig. 2). Disease progressed more quickly in soil treated
seem to be the agent of activity. by pseudomonads APF3 and APF4, S. setonii RP87, and
Under greenhouse conditions, plants inoculated with the control. However, disease increased slowly in soil treated
antagonistic isolates pseudomonad APF1 and B. subtilis by B. subtilis B2G, and pseudomonad strains APF1 and
B2G significantly reduced disease compared to the control. APF2 strains. Wilt incidence was significantly reduced by
Both strains reduced AUDPC by 60% and 56%, respec- pseudomonad APF1 and B. subtilis B2G and was corre-
tively. Moreover, significantly greater amounts of biomass lated with a decrease of the population of the pathogen
(fresh and dry weight) compared to the control were in soil. Similar reports of pathogen decline in the soil due
obtained with these strains. Guo et al. (2004) reported com- to antagonists and disease reduction have been presented
parable R. solanacearum wilt disease reduction and yield (Pieterse et al., 2001; Szczech and Shoda, 2004).
increase of tomato plants after treatment by Bacillus spp. The antagonist S. setonii RP87 that produced the high-
and fluorescent pseudomonads. Priou et al. (2005) recorded est antagonism in vitro, did not significantly differ from the
80% reduction of the same disease on tomato plants under control in both wilt suppression and biomass production in
greenhouse conditions using Pseudomonas putida (Trevi- tomato. This lack of correlation between in vitro result and
san). Antagonistic Pseudomonas spp. were tested for their biological control in vivo has also been documented in
ability to suppress bacterial wilt in tobacco and some other studies on other pathogens (e.g., Ran et al., 2005;
showed promising results (Liu et al., 1999). Recently, Ran Rajkumar et al., 2005). Much of inconsistency in the per-
et al. (2005) reported suppression of bacterial wilt in Euca- formance of antagonistic bacteria has been attributed to
lyptus urophylla (Blake) by fluorescent Pseudomonas spp. variability in the physical and chemical properties within
There are several modes of action known for rhizobacte- the niches occupied by biocontrol agents, as well as the
ria applied for the control of plant diseases (Blanco et al., plant, that affect both colonization and expression of bio-
2004; Ran et al., 2005; Dwivedi and Johri, 2003). Pseudo- control mechanisms (Notz et al., 2002, Ryan et al., 2004).
monads exert a protective effect on the roots through Among the antagonists tested, fluorescent pseudomonad
antagonism towards phytopathogenic bacteria by produc- APF1 showed the most beneficial characteristics, as it con-
ing metabolites that include: lytic enzymes (Berg, 1996); sistently suppressed the R. solanacearum wilt and also pro-
plant hormones and other plant growth promoting sub- moted increased plant fresh and dry weight compared to
stances, e.g., auxins, indole-3-acetic acid, and gibberellins untreated control (Table 3). Plant growth by rhizobacteria
(Ramamoorthy and Samiyappn, 2001); siderophores may be associated with secretion of auxins, gibberellins,
(Dwivedi and Johri, 2003); and antibiotics (Dwivedi and and cytokinins (Ramamoorthy and Samiyappn, 2001)
Johri, 2003; Ran et al., 2005). Bacillus spp. is also known and suppression of deleterious microorganisms in the rhi-
to produce a wide range of secondary metabolites such as zosphere (Gamliel and Katan, 1993). The use of rhizo-
antibiotics, non-volatile and volatile compounds (Parke sphere bacteria for increasing yield and for crop
and Gurian-Sherman, 2001) and lytic enzymes (Frändberg protection is an attractive approach in the modern system
and Schnürer, 1994). in developing a sustainable agriculture.
The population dynamics studies of R. solanacearum in In conclusion, the fluorescent pseudomonad APF1
the soil showed that the amount of the recoverable patho- strain and B. subtilis B2G have proved to be consistently
gen was significantly affected by some of the applied antag- efficient in the control of R. solanacearum wilt disease in
onists. At early stage of incorporation of antagonists into a in planta biocontrol assay under greenhouse conditions.
the soil, population of R. solanacearum declined in all treat- Field studies should be undertaken to confirm the effective-
ments, including the control, until 3 weeks after treatment ness of the antagonistic strains under natural conditions
(Fig. 2). Later, however, significant differences could be and their modes of action studied.
observed among the treatments. In soil treated with fluo-
rescent pseudomonad APF1 and B. subtilis B2G strains, Acknowledgments
density of the pathogen declined significantly and started
to increase gradually. Conversely, in the presence of other We are very grateful to the Catholic Academic
strains and the control, population steadily increased. Exchange Service (KAAD), Germany, for financial sup-
Pseudomonad APF1 strain significantly minimized the port of the first author. We also thank the Network for
F. Lemessa, W. Zeller / Biological Control 42 (2007) 336–344 343

the Improvement of Potato and Sweetpotato Program in Garrett, K.A., Mundt, C.C., 2000. Host diversity can reduce potato late
East and Central Africa (PRAPACE) for financing the field blight severity for focal and general patterns of primary inoculum.
Phytopathology 90, 1307–1312.
research in Ethiopia. The experiments were conduct as part Guo, J.H., Guo, Y.H., Zhang, L.X., Qi, H.Y., Fang, Z.D., 2001.
of a PhD study by the first author in the Federal Biological Screening for biocontrol agents against cayenne pepper bacterial wilt.
Research Centre for Agriculture and Forestry (BBA), Insti- China J. Biol. Control 17, 101–106.
tute for Biological Control, Darmstadt, with attachment to Guo, J.H., Qi, H.Y., Guo, Y.H., Ge, H., Gong, L.Y., Zhang, L.X., Sun,
Hannover University in Germany. P.H., 2004. Biocontrol of tomato wilt by growth-promoting rhizobac-
teria. Biol. Control 29, 66–72.
Hassett, D.J., Schweizer, H.P., Ohman, D.E., 1995. Pseudomonas aeru-
References ginosa sodA and sodB mutants defective in manganese and iron
cofactored superoxide dismutase activity demonstrate the importance
Berg, G., 1996. Rhizobacteria of oil seed rape antagonistic to Verticillium of the iron-cofactored form in aerobic metabolism. J. Bacteriol. 177,
dahliae. J. Plant Dis. Protec. 103, 20–30. 6330–6337.
Bjorn, M.J., Iglewski, B.H., Ives, S.K., Sadoff, J.C., Vasil, M.L., 1978. Hayward, A.C., 1964. Characteristics of Pseudomonas solanacearum. J.
Effect of iron on yields of exotoxin A in cultures of Pseudomonas Appl. Bacteriol. 27, 265–277.
aeruginosa PA-103. Infec. Immun. 19, 785–791. Hayward, A.C., 1991. Biology and epidemiology of bacterial wilt
Blanco, J.M., Jurado, D.R., Hervas, A., Diaz, R.M.J., 2004. Suppression caused by Pseudomonas solanacearum. Ann. Rev. Phytopathol. 29,
of Verticillium wilt in olive planting stocks by root-associated 65–87.
fluorescent Pseudomonas spp. Biol. Control 30, 474–486. Kang, Y., Mao, G., Lu, C., He, L., 1995. Biological control of
Bloemberg, G.V., Lugtenberg, B.J.J., 2001. Molecular basis of plant bacterial wilt of tomato by extracellular protein defective mutant
growth promotion and biocontrol by rhizobacteria. Curr. Opin. Plant of Pseudomonas solanacearum. Acta Phytopathol. Sinica 22, 287–
Biol. 4, 343–350. 288.
Boucher, C.A., Gough, C.L., Arlat, M., 1992. Molecular genetics of Kelman, A., 1954. The relationship of pathogenicity in Pseudomonas
pathogenicity determinants of Pseudomonas solanacearum with special solanacearum to colony appearance on a tetrazolium medium. Phyto-
emphases on hrp genes. Ann. Rev. Phytopathol. 30, 443–461. pathology 44, 693–695.
Buddenhagen, I.W., Sequeira, L., Kelman, A., 1962. Designation of races King, E.O., Ward, M.K., Raney, D.E., 1954. Two simple media for the
of Pseudomonas solanacearum. Phytopathology 52, 726. demonstration of pyocyanin and fluorescein. J. Lab. Clin. Med. 44,
Campbell, C.L., Madden, L.V., 1990. Introduction to Plant Disease 301–307.
Epidemiology. Wiley, NY. Kloepper, J.W., Leong, J., Teintze, M., Schroth, M.N., 1980. Enhanced
Chen, W.Q., Morgan, D.P., Felts, D., Michailides, T.J., 2003. Antagonism plant growth by siderophores produced by plant growth promoting
of Paenibacillus lentimorbus to Botryosphaeria dothidea and biological rhizobacteria. Nature 268, 885–886.
control of panicle and shoot blight of pistachio. Plant Dis. 87, 359–365. Krieg, N.R., Holt, J.G., 1984. Bergey’s Manual of Systematic Bacteriol-
Ciampi-Panno, L., Fernandez, C., Bustamante, P., Andrade, N., Ojeda, ogy, vol. 1. Williams and Wilkins, Baltimore.
S., Conteras, A., 1989. Biological control of bacterial wilt of potatoes Lemessa, F., Zeller, W., 2007. Isolation and characterisation of Ralstonia
caused by Pseudomonas solanacearum. Am. Potato J. 66, 315–332. solanacearum strains from Solanaceae crops in Ethiopia. J. Basic
Dong, C., Zeng, X., Liu, Q., 1999. Biological control of tomato bacterial Microbiol. 47, 40–49.
wilt with avirulent bacteriocinogenic strain of Ralstonia solanacearum. Lim, H., Kim, S., 1997. Role of siderophores in biocontrol of Fusarium
J. Sci. China Agric. Univ. 20, 1–4. solani and enhanced growth response of bean by Pseudomonas
Duffy, B.K., Défago, G., 1999. Environmental factors modulating fluorescens GL20. J. Microbiol. Biotec. 7, 13–20.
antibiotic and siderophore biosynthesis by Pseudomonas fluorescens Liu, Q.G., Li, Z., Tang, Z., Zeng, X.M., 1999. Control of tobacco
biocontrol strains. Appl. Environ. Microbiol. 65, 2438–2439. bacterial wilt with antagonistic bacteria and soil amendment. China J.
Dwivedi, D., Johri, B.N., 2003. Antifungals from fluorescent pseudomo- Biol. Control 15, 94–95.
nads: biosynthesis and regulation. Curr. Sci. 85, 1693–1703. Loper, J.E., Suslow, T.V., Schroth, M.N., 1984. Log normal distribution
El Albyad, M.S., el Sayed, M.A., el Shanshoury, A.R., 1996. Effect of of bacterial populations in the rhizosphere. Phytopathology 74, 1454–
culture conditions on the antimicrobial activities of UV-mutants of 1460.
Streptomyces corchorusii and S. spiroverticillatus against bean and McLaughlin, R.J., Sequeira, L., Weingartner, D.P., 1990. Biocontrol of
banana wilt pathogens. Microbiol. Res. 151, 201–211. bacterial wilt of potato with an avirulent strain of Pseudomonas
Elphinstone, J.G., 1992. Virulence of isolates of Pseudomonas solanacea- solanacearum: interactions with root-knot nematodes. Am. Potato J.
rum from worldwide sources in resistant and susceptible tomato 67, 93–107.
cultivars. In: Prior, P., Bart, S., Leclercq, S., Darrasse, A., Anais, G. Mitchell, J.K., Carter, W.E., 2000. Modeling antimicrobial activity of
(Eds.), Resistance to bacterial wilt in tomato as discerned by spread of Clorox using an agar-diffusion test: a new twist on an old
Pseudomonas (Burholderia) solanacearum in the stem tissues. Plant. experiment. Bioscene 26, 9–13.
Pathol. 45, 720–726. Montesinos, E., Bonaterra, A., Ophir, Y., Beer, S.V., 1996. Antagonism of
Englerbrecht, M.C., 1994. Modification of a semi-selective medium for the selected bacterial strains to Stemphylium vesicarium and biological
isolation and quantification Pseudomonas solanacearum. ACIAR Bact. control of brown spot of pear under controlled environment condi-
Wilt News Lett. 10, 3–5. tions. Phytopathology 86, 856–863.
Frändberg, E., Schnürer, C., 1994. Chitinolytic properties of Bacillus Notz, R., Maurhofer, M., Dubach, H., Haas, D., Defago, G., 2002.
paubli K1. J. Appl. Bacteriol. 76, 361–367. Fusaric acid-producing strains of Fusarium oxysporum alter 2,4-
French, E.R., Sequeira, L., 1970. Strains of Pseudomonas solanacearum diacetylphloroglucinol biosynthetic gene expression in Pseudomonas
from Central and South America: a comparative study. Phytopathol- fluorescens CHAo in vitro and rhizosphere of wheat. Appl. Environ.
ogy 60, 506–512. Microbiol. 68, 2229–2235.
Gamalero, E., Lingua, G., Berta, G., Lemanceau, P., 2003. Methods for Parke, J.L., Gurian-Sherman, D., 2001. Diversity of the Burkholderia
studying root colonization by introduced beneficial bacteria. Agron- cepacia complex and implications for risk assessment of biological
omie 23, 407–418. control strains. Ann. Rev. Phytopathol. 39, 225–258.
Gamliel, A., Katan, J., 1993. Suppression of major and minor pathogens Pieterse, C.M.J., Van Pelt, J.A., Van wees, S.C.M., Ton, J., Leon-
by fluorescent pseudomonads in solarized soil and non-solarized soil. Kloosteriziel, K.M., Keurentjes, J.J.B., Verhagen, B.W.M., Knoester,
Phytopathology 83, 68–75. M., Van der Sluis, I., Bakker, P.A.H.M., Van Loon, L.C., 2001.
344 F. Lemessa, W. Zeller / Biological Control 42 (2007) 336–344

Rhizobacteria-mediated induced systemic resistance: triggering, sig- Silveira, E.B., Da, R., Mariano de, L.R., Michereff, S.J., 1995. Antago-
nalling and expression. Eur. J. Plant Pathol. 107, 51–61. nism of Bacillus spp. Against Pseudomonas solanacearum and effect on
Priou, S., Marquez, M., Gutarra, L., 2005. Biological control of bacterial tomato seedling growth. Fitopathol. Brasil 20, 605–612.
wilt of potato (Ralstonia solanacearum) using antagonistic Pseudomo- Smith, J.J., Offord, L.C., Holderness, M., Saddler, G.S., 1995. Genetic
nas putida strains. Phytopathology 95 (Supplement) S85. diversity of Burkholderia solanacearum (Synonym Pseudomonas solan-
Rajkumar, M., Lee, W.H., Lee, K.J., 2005. Screening of bacterial acearum) race 3 in Kenya. Appl. Environ. Microbiol. 61, 4263–
antagonists for biological control of Phytophthora blight of pepper. 4268.
J. Basic Microbiol. 45, 55–63. Szczech, M., Shoda, M., 2004. Biocontrol of Rhizoctonia damping-off of
Ramamoorthy, V., Samiyappn, R., 2001. Induction of defense related tomato by Bacillus subtilis combined with Burkholderia cepacia. J.
genes in Pseudomonas fluorscens treated chilli plants in response Phytopathol. 152, 549–556.
to infection by Colletotrichum capsici. J. Myc. Plant Pathol. 31, Thomshaw, L.S., 1996. Biological control of plant pathogens. Curr. Opin.
146–155. Biotech. 77, 343–347.
Ran, L.X., Liu, C.Y., Wu, G.J., van Loon, L.C., Bakker, P.A.H.M., 2005. Toyota, K., Kimura, M., 2000. Suppresision of Ralstonia solanacearum in
Suppression of bacterial wilt in Eucalyptus urophylla by fluorescent soil following colonization by other strains of R. solanacearum. Soil
Pseudomonas spp in. China. Biol. Control 32, 111–120. Sci. Plant Nutr. 46, 449–459.
Rangajaran, S., Saleena, L.M., Vasudevan, P., Nair, S., 2003. Biological Villacieros, M., Power, B., Sanchez-Contreras, M., Loret, J., Oruzebal,
suppression of rice diseases by Pseudomonas spp under saline soil R.I., Martin, M., Franandez-Pinas, F., Bouile, I., Whelan, C.,
conditions. Plant Soil 251, 73–82. Dowling, D.N., Rivilla, R., 2003. Colonization behaviour of Pseudo-
Ryan, A.D., Kinkel, L.L., Schottel, J.L., 2004. Effect of pathogen isolate, monas fluorescens and Sinorhizobium meloti in the alfalfa (Medicago
potato cultivar, and antagonist strain on potato scab severity and sativa) rhizosphere. Plant Soil 251, 47–54.
biological control. Bioc. Sci. Tech. 14, 301–311. Yabuuchi, E., Kosako, Y., Yano, I., Hota, H., Nishiuchi, Y., 1995.
Sambrook, J., Fritsch, E.F., Maniatis, T.A., 1989. Molecular Clonning: A Transfer of two Burkholderia and an Alcaligenes species to Ralstonia
Laboratory Manual, second ed. Cold Spring Harbor Laboratory gen. nov., Ralstonia solanacearum (Smith, 1986) comb. nov. Microbiol.
Press, Cold Spring Harbor Laboratory, NY. Immun. 39, 897–904.
SAS Institute Inc., 1999. SAS Institute Inc., SAS OnlineDoc, Version 8, Yaynu, H., 1989. Characteristics of isolates of Pseudomonas solanacearum
Cary, NC, USA. in Ethiopia. Eth. J. Agric. Sci. 11, 7–13.

You might also like