C. R.

Howell
USDA/ARS Southern Plains Agricultural Research Center

Mechanisms Employed by Trichoderma Species

in the Biological Control of Plant Diseases:
The History and Evolution of Current Concepts
Fungal species belonging to the genus control. The selected research papers cited certain circumstances T. lignorum might
Trichoderma are worldwide in occurrence in this article were chosen because they act as a competitor for nutrients with R.
and easily isolated from soil, decaying illustrate what has been learned about solani, he much favored mycoparasitism as
wood, and other forms of plant organic mechanisms involved in biocontrol with the principal mechanism for biocontrol.
matter. They are, for the most part, classi- Trichoderma species. Two years later, Weindling (32) reported
fied as imperfect fungi, in that they have that a strain of T. lignorum produced a
no known sexual stage. Rapid growth rate Mycoparasitism and “lethal principle” that was excreted into the
in culture and the production of numerous Antibiotic (Toxin) Production surrounding medium, allowing parasitic
spores (conidia) that are varying shades of Those who are familiar with members of activity by the biocontrol agent. In 1941
green characterize fungi in this genus. The the genus Trichoderma know that one of (33), he characterized the “lethal princi-
reverse side of colonies is often uncolored, the most salient characteristics of the group ple”, demonstrated that it was toxic to both
buff, yellow, amber, or yellow-green, and is their ability to parasitize other fungi R. solani and Sclerotinia americana, and
many species produce prodigious quanti- (Fig. 1). It is therefore not surprising that named it gliotoxin (Fig. 2). Subsequently,
ties of thick-walled spores (chlamydo- Weindling (31) ascribed biocontrol by T. it has been demonstrated that the fungus
spores) in submerged mycelium (8). The lignorum of citrus seedling disease, incited that produced gliotoxin was not T.
potential of Trichoderma species as bio- by Rhizoctonia solani, to mycoparasitism. lignorum, but Gliocladium virens (30), a
control agents of plant diseases was first Weindling described in detail the myco- species that has recently been renamed
recognized in the early 1930s (31), and in parasitism of R. solani hyphae by the hy- Trichoderma virens (27).
subsequent years, control of many diseases phae of the biocontrol agent, including In the years following this seminal
has been added to the list (1,3,5,7,9,11,19, coiling around pathogen hyphae, penetra- work, many instances of successful biocon-
23,29,34,37,40). This has culminated in the tion, and subsequent dissolution of the host trol with Trichoderma species have been
commercial production of several Tricho- cytoplasm. This phenomenon occurred ascribed to the mechanisms of mycopara-
derma species for the protection and regardless of the supply of external nutri- sitism and/or antibiosis (1,3,5,11,17,20,34).
growth enhancement of a number of crops ents to the host or mycoparasite. Although The importance of antibiotics for bio-
in the United States (24), and in the pro- he considered the possibility that under control activity was demonstrated in sev-
duction of Trichoderma species and mix-
tures of species in India, Israel, New Zea-
land, and Sweden (D. R. Fravel, personal
communication).
One of the most interesting aspects of
the science of biological control is the
study of the mechanisms employed by
biocontrol agents to effect disease control.
Past research indicates that the mecha-
nisms are many and varied, even within the
genus Trichoderma. In order to make the
most effective use of biocontrol agents for
the control of plant diseases, we must
understand how the agents work and what
their limitations are. We can then develop
effective means of culturing, storing, ap-
plying, and utilizing biocontrol agents so
that we harness their best effort for disease

Dr. Howell’s address is: USDA/ARS Southern

Plains Agricultural Research Center, 2765 F&B
Road, College Station, TX 77845
E-mail: [email protected]

Publication no. D-2002-1028-01F

This article is in the public domain and not copy-
rightable. It may be freely reprinted with custom-
ary crediting of the source. The American Phyto- Fig. 1. Penetration and haustoria formation within the large hyphae of Rhizoctonia
pathological Society, 2003. solani by the smaller hyphae of Trichoderma virens.

4 Plant Disease / Vol. 87 No. 1

eral studies. In 1983, Howell and Stipano- tant strains of T. virens (G6-2, -15, and root. The difficulty in viewing competition
vic (15) isolated and described a new -57) that were unable to parasitize R. so- through rhizosphere competence as a major
antibiotic, gliovirin, from Gliocladium lani (Fig. 4). However, antibiotic biosyn- mechanism in biological control is that
(Trichoderma) virens (GV-P) that was thesis did not differ from the parent strains. strains of T. koningii that are excellent root
strongly inhibitory to Pythium ultimum and When compared for biocontrol efficacy colonizers exhibit little or no biocontrol
a Phytophthora species, but not to R. so- against R. solani–incited cotton seedling activity against R. solani on cotton seed-
lani, Thielaviopsis basicola, Phymato- disease, the mycoparasitic-deficient mu- lings (C. R. Howell, unpublished). Al-
trichum omnivorum, Rhizopus arrhizus, or tants were just as effective as the parent though competition through rhizosphere
Verticillium dahliae. Gliovirin also was not strains. These results indicated that myco- competence may not be among the princi-
inhibitory to the bacteria Bacillus thurin- parasitism was not a major mechanism in pal mechanisms that drive biological con-
gensis and Pseudomonas fluorescens. They the biocontrol of this particular disease. trol, it is certainly a valuable adjunct to
demonstrated that mutants unable to syn- Later studies also raised questions about those that do. One concept that is associ-
thesize the antibiotic lost the capacity to the role of the antibiotic gliotoxin pro- ated with competition and rhizosphere
control Pythium damping-off of cotton. duced by T. virens as a mechanism in bio- competence, the replacement of endoge-
However, a mutant (GV-1) with enhanced control of cotton seedling disease incited nous fungi on the root surface (10), can be
gliovirin production was no more effective by R. solani. In one instance (16), mutants difficult to demonstrate. Trichoderma spe-
than the wild type in controlling the dis- of T. virens that were deficient in gliotoxin cies are often able to suppress the growth
ease (Fig. 3). In another instance, Lifshitz biosynthesis were just as effective in con- of endogenous fungi on an agar medium
et al. (19) showed that control of Pythium trolling the disease as parent strains (Fig. and therefore mask their presence. A case
species on peas by T. harzianum (T-12) and 5). In other studies (13,14), a mutant of T. in point is that of T. virens–treated root
T. koningii (T-8) was not due to either my- virens deficient for both mycoparasitism segments taken from soil heavily infested
coparasitism or competition. They ascribed and gliotoxin biosynthesis (G6-5) still with propagules of Macrophomina phaseo-
biocontrol to the production of a toxic retained biocontrol efficacy equal to that of lina, the pathogen that causes charcoal rot
factor by the biocontrol agent in the sper- the parent strain (G-6) against both P. ulti- in a wide range of crops. When these seg-
mosphere that inhibited growth of the mum and R. solani. These results indicate ments are plated on an agar medium at
pathogen. Lumsden et al. (23) found that that neither mycoparasitism nor antibiosis room temperature, only T. virens grows
suppressive activity of T. virens (GL-21) to is required for biocontrol of cotton seed- from the root. However, if these same cul-
damping-off of zinnias, incited by both R. ling diseases incited by these pathogens. tures are incubated at 40°C, a temperature
solani and P. ultimum, was correlated with at which T. virens will not grow, the patho-
production of the antibiotic gliotoxin by Competition gen grows readily from many parts of the
the biocontrol agent. Further, Wilhite et al. and Rhizosphere Competence root system (Fig. 7). This may also occur
(35) used mutation to demonstrate that loss If mycoparasitism and antibiosis are not with other Trichoderma species and other
of gliotoxin production in T. virens mutants the principal mechanisms in the biocontrol pathogens, but it is not easily demonstrated
(G20-M36, M37, M50, and M96) had an process, what is? One mechanism that has because growth of the biocontrol agent
adverse effect on the efficacy of these gained adherents in recent years is that of can’t be suppressed without suppressing
strains as biocontrol agents of Pythium competition through rhizosphere compe- the pathogen also. One method of demon-
damping-off. The difficulty of correlating tence. Rhizosphere competence is impor- strating pathogen replacement on the roots
loss of antibiotic biosynthesis with a de- tant because a biocontrol agent cannot by biocontrol agents might be through the
crease in biocontrol activity, using mutants compete for space and nutrients if it is use of fungicide-resistant strains of the
generated by ultraviolet light, is that the unable to grow in the rhizosphere. Tricho- pathogen and culture of infested roots on a
roles of silent mutations in the process are derma species, either added to the soil or medium amended with fungicide suppres-
not considered. applied as seed treatments, grow readily sive only to the biocontrol agent.
Other studies utilizing the techniques of along with the developing root system of
genetic manipulation of microbes have the treated plant (9,14,40). This can be Enzymes
brought some of the assumptions about the shown easily by simply plating surface- More recent research into the possible
respective roles of mycoparasitism and sterilized root segments from treated plants mechanisms involved in biological control
antibiotic production in biological control on an agar medium (Fig. 6). After a suit- by Trichoderma species has led to several
into question. In 1987, Howell (12) used able incubation period, the fungus can be
ultraviolet light irradiation to produce mu- seen growing from virtually all parts of the

Fig. 3. Growth inhibition of Pythium

ultimum by the Trichoderma virens–
Fig. 2. Growth inhibition of Rhizoctonia solani by the Trichoderma virens–produced produced antibiotic gliovirin: A, parent
antibiotic gliotoxin: A, gliotoxin-amended medium, and B, nonamended medium. strain, and B, gliovirin-deficient mutant.

Plant Disease / January 2003 5

alternative explanations for successful of the plant pathogen. These enzymes Metcalf and Wilson (25) described the
biocontrol. One idea that has been ad- function by breaking down the polysaccha- colonization of onion roots, infected with
vanced is that enzymes such as chitinases rides, chitin, and b-glucans that are respon- Sclerotium cepivorum, by T. koningii (Tr5).
and/or glucanases produced by the biocon- sible for the rigidity of fungal cell walls, Hyphae of the biocontrol agent penetrated
trol agent are responsible for suppression thereby destroying cell wall integrity. into infected epidermal and cortical tissue
of the root to destroy the hyphae of the
pathogen, with little or no damage to unin-
fected plant tissue. The authors ascribed
this biocontrol phenomenon to production
of endo- and exo-chitinases by T. koningii.
Baek et al. (2) disrupted or over-expressed
the gene coding for chitinase (cht42) in T.
virens (Gv29-8). Transformants with re-
duced enzyme activity or over-expression
of the enzyme gave significantly decreased
or enhanced biocontrol activity, respec-
tively, against R. solani–incited cotton
seedling disease. However, the differences
were not great and probably indicate that
other factors were also at work. In a
similar study, Woo et al. (36) also dis-
rupted chitinase (ech42) activity in T. har-
zianum (P1) and showed reduced biocon-
trol activity against Botrytis cinerea on
bean leaves. However, the transformant
was as effective as the wild type against P.
ultimum and exhibited enhanced activity
against R. solani when compared with the
control. The authors concluded that reac-
tions between T. harzianum strains and
various fungal hosts were based on differ-
ent mechanisms. This, again, indicates that
Fig. 4. Mycoparasitism of Rhizoctonia solani by Trichoderma virens: A, parent strain factors other than chitinase activity are
coiling around host hyphae, and B, mycoparasitic-deficient mutant with no coiling or important to the biocontrol process. The
penetration of host hyphae.
possible role of chitinolytic enzymes in
biocontrol is further supported by the work
of Lorito et al. (22), who transferred the
gene encoding endochitinase from T. har-
zianum (P1) into tobacco and potato and
demonstrated a high level and broad spec-
trum of resistance against a number of
plant pathogens. Recently, Bolar et al. (4)
demonstrated enhanced resistance to apple
scab incited by Venturia inaequalis in
transgenic apple plants that had been trans-
formed with genes encoding both endo-
and exo-chitinases from T. atroviride (P1).
Finally (C. M. Kenerley, personal commu-
nication), a gene from T. virens (Gv29-8)
Fig. 5. Trichoderma virens strains cultured on potato dextrose agar amended with a encoding a 42-kDa endochitinase under
Bacillus subtilis lawn. G-6, G-11, and G-20 are parent strains. G6-22 and -151, G11-2 control of the CaMV 35S promoter was
and -4, and G20-92 and -67 are gliotoxin-deficient mutants. transformed into the cotton cultivar Coker
312. Transgenic lines were screened for
fungal chitinase activity, and several high
producers were identified. Homozygous
lines were generated that retained high
enzyme activity and expression of this
gene in the roots, shoots, and leaves. Seed-
lings of these lines were screened against
the seedling pathogens Rhizoctonia solani
and Thielaviopsis basicola, and detached
leaves were screened against Alternaria
alternata. Selected transgenic lines demon-
strated significant levels of resistance
against all three pathogens when compared
with the parental control and a commercial
cotton cultivar.
Another interesting concept related to
Fig. 6. Trichoderma virens cultured from tap and secondary roots of cotton plants enzyme biosynthesis as a mechanism in the
grown from T. virens–coated seed in field soil. biocontrol process is that advanced by Elad

6 Plant Disease / Vol. 87 No. 1

and Kapat (7) and Kapat et al. (18), who oxysporum. Synergism occurred in all induced synthesis of much higher concen-
suggested that biocontrol of B. cinerea by cases, but the level depended on the anti- trations of the terpenoids desoxyhemigos-
T. harzianum (T39) might be due, in part, fungal activity of the enzyme. Synergism sypol (dHG), hemigossypol (HG), and
to the actions of T. harzianum–produced was lower when the enzyme was added gossypol (G) in developing roots than
proteases that inactivate the hydrolytic after the antifungal compound, indicating those found in untreated controls. Gossypol
enzymes produced by B. cinerea on bean that cell wall degradation was needed in was toxic only at high levels, but the path-
leaves. The protease enzymes break down order to establish the interaction. way intermediates dHG and HG were
hydrolytic enzymes into peptide chains strongly inhibitory to the cotton seedling
and/or their constituent amino acids and Induction of Defense pathogen, R. solani, at much lower concen-
thereby destroy their capacity to act on Responses in Plants trations. Trichoderma species were much
plant cells. The authors demonstrated that Another mechanism proposed to explain more resistant to cotton terpenoids than
protease solutions produced by the biocon- biocontrol activity by Trichoderma species were seedling disease pathogens. Biocon-
trol fungus on bean leaves partially deacti- is that of induction of resistance in the host trol activity against R. solani was highly
vated hydrolytic enzymes and reduced plant by treatment with the biocontrol correlated with induction of terpenoid syn-
disease severity by 56 to 100% when the agent. This concept is supported by the thesis in cotton roots by Trichoderma spe-
solutions were used to treat leaves infected work of Yedidia et al. (37), who demon- cies, even among strains of T. virens that
with the pathogen. They also showed that strated that inoculating roots of 7-day-old were deficient for mycoparasitism and
the addition of protease inhibitors to drops cucumber seedlings in an aseptic hydro- antibiotic production. In addition to terpe-
containing mixed conidia of the two fungi, ponic system with T. harzianum (T-203) noid synthesis, treatment of cotton roots
applied to bean leaves, nullified the bio- spores to a final concentration of 105 per with T. virens also induced significantly
control effect of the T. harzianum conidia. ml initiated plant defense responses in both higher levels of peroxidase activity than
Protease production by T. harzianum has the roots and leaves of treated plants. They that found in control roots. Peroxidase
also been associated with biocontrol of the also demonstrated that hyphae of the bio- activity and terpenoid levels in seedling
root-knot nematode Meloidogyne javanica control fungus penetrated the epidermis hypocotyls were not significantly different
on tomato plants. Sharon et al. (29) showed and upper cortex of the cucumber root. The from those found in the controls. In this
that tomato plants treated with the biocon- plant response was marked by an increase case, plant defense responses appeared to
trol agent (T-203) and grown in nematode- in peroxidase activity (often associated be confined to the root system.
infested soil exhibited a drastic reduction with the production of fungitoxic com-
in root galling when compared with the pounds), an increase in chitinase activity, Metabolism
control. A transformed strain of T. har- and the deposition of callose-enriched wall of Germination Stimulants
zianum (P2) containing multiple copies of appositions on the inner surface of cell One unique mechanism employed by
a gene-encoding protease reduced root walls. Increased enzyme activities were Trichoderma species to effect biological
galling better than wild-type strains, and it observed in both roots and leaves. Interest- control that does not fit neatly into any of
was the only strain able to penetrate egg ingly, the plant defense became muted with the categories previously mentioned was
masses and the eggs inside. The authors time and began to resemble a symbiotic recently discovered. In a study of the
suggested that improved proteolytic activ- mycorrhizal association. Later, Yedidia et mechanisms involved in the biocontrol of
ity might be important for biological con- al. (38) showed that inoculation of cucum- preemergence damping-off of cotton seed-
trol of nematodes. ber roots with T. harzianum (T-203) in- lings incited by P. ultimum and/or Rhizopus
Finally, Migheli et al. (26) showed that duced an array of pathogenesis-related oryzae, Howell (13) found that control by
transformants of T. longibrachiatum proteins, including a number of hydrolytic T. virens (G6, G6-5) or protoplast fusants
(CECT2606), over-expressing a gene en- enzymes. Plants treated with a chemical of T. virens/T. longibrachiatum (Tvl-30,
coding b-1,4-endoglucanase, were slightly inducer (2,6-dichloroisonicotinic acid) of Tvl-35) was due to metabolism of germi-
more effective in the biocontrol of P. ulti- disease resistance displayed defense re- nation stimulants released by the cotton
mum on cucumber than the wild type. They sponses that were similar to those of plants seed. These compounds normally induced
concluded that a mixture of several en- inoculated with the biocontrol agent. pathogen propagules to germinate. Disease
zymes might be necessary for efficient cell In another study, Howell et al. (14) dem- control could be effected by wild-type
wall lysis. onstrated that seed treatment of cotton with strains or by mutant strains that were defi-
The concept of enzyme biosynthesis as a biocontrol preparations of T. virens (G-6, cient for mycoparasitism, antibiotic pro-
mechanism of biocontrol has been ex- G-11, G6-5) or application of T. virens duction, and induction of terpenoid synthe-
panded to include synergism between en- culture filtrate to cotton seedling radicles sis in cotton roots. If, however, pathogen
zymes and antibiotics. Di Pietro et al. (6)
studied the synergistic effects of endochiti-
nase and gliotoxin on conidial germination
of B. cinerea. They found that treatment of
B. cinerea conidia with the combination
was much more effective than treatment
with the enzyme or antibiotic alone.
Schirmbock et al. (28) noticed a similar
effect on conidial germination and hyphal
elongation of B. cinerea when the fungus
was treated with combinations of hydro-
lytic enzymes and peptaibols produced by
T. harzianum (ATCC 36042). Neither the
enzymes nor the antibiotics alone were as
effective as a combination of the two. Lo-
rito et al. (21) further expanded this con-
cept by combining a number of antifungal Fig. 7. Cultures of cotton roots from plants grown in Macrophomina phaseolina–
compounds with several kinds of hydro- infested soil and incubated at 25°C: A, roots from nontreated seed showing M. phase-
lytic enzymes and applying them to olina, and B, roots from Trichoderma virens–treated seed showing only the biocontrol
propagules of B. cinerea and Fusarium agent. Incubation of B at 40°C shows only M. phaseolina growing from cotton roots.

Plant Disease / January 2003 7

 on to isolate potential biocontrol agents
were, at best, only marginally effective.
Many strains were isolated because they
were observed parasitizing plant pathogens
or inhibiting their growth in culture. The
enzymes and antibiotics produced by
Trichoderma species that appear to be in-
volved in biocontrol are strongly influ-
enced by the substrate on which the fungus
is grown, and conditions in the laboratory
probably occur only rarely in nature or not
at all. Temperature also has a profound
effect on the production and activities of
enzymes and antibiotics associated with
biocontrol by Trichoderma species. What
occurs at 78°F in a petri dish may not oc-
cur at all in the soil around a germinating
seed at 60°F. The presence of other mem-
bers of the soil microflora also may influ-
ence biocontrol activity by inhibiting the
growth and development of the biocontrol
agent or by metabolizing its enzymatic
and/or antibiotic products. This may not
entirely negate the biocontrol activities of
Trichoderma species, but it might well
Fig. 8. Effect of seed exudates cultured with Trichoderma virens preparations on limit their efficacy in terms of the length of
germination of Rhizopus oryzae sporangiospores: A, exudate cultured with dead T. time that they are effective and the distance
virens preparation stimulates spore germination and mycelial growth of R. oryzae, from the biocontrol focal point that they
while B, exudate cultured with live T. virens preparation does not stimulate spore exhibit influence.
germination.
Given the above considerations, perhaps
the best method for obtaining a potential
biocontrol agent might be one where the
propagules were induced to germinate by cient, treatment with T-22 did not result in candidate Trichoderma species is isolated
artificial means, none of the above treat- increased yields of grain or silage. How- from areas of the plant and soil where it is
ments gave effective control of the disease. ever, the nitrogen required for optimum expected to function in disease control, and
The importance of metabolism of stimu- yields was lower for T-22 treated plants where it is growing under conditions of
latory compounds by the biocontrol agent than for the untreated controls. Harman temperature, moisture, and nutrient avail-
is further supported by the fact that cotton (10) later reported a strong interaction ability that approximate those found in
cultivars that do not produce pathogen between T-22 and the nitrogen-fixing bac- nature. Isolates obtained in this manner can
propagule stimulants during germination terium Bradyrhizobium japonicum. Treat- then be screened for biocontrol efficacy.
are virtually immune to the disease. Again, ment of soybeans with the combination Since many of the mechanisms that have
artificial induction of pathogen propagule stimulated root growth more effectively proven to be most important in biological
germination rendered these cotton cultivars than did either alone. Theoretically, the control do not lend themselves to assay in
susceptible to disease. Also, seed exudates combination of a nitrogen-fixing bacterium a petri dish, the screening procedure will
cultured with dead T. virens biocontrol and a fungus that enables the plant to util- probably entail treatment of the seed, soil,
preparation stimulated pathogen propa- ize nitrogen more efficiently should de- or plant with the biocontrol agent. This can
gules to germinate, while exudates cultured crease even further the nitrogen fertilizer be followed by culture of the treated host
with live biocontrol preparation did not requirements of the crop. More recently, plant in a pathogen-infested environment
(Fig. 8). Yedidia et al. (39) showed that treatment of until disease symptoms manifest them-
cucumber plants in soil with T. harzianum selves. The level of symptom expression
Adjunct Mechanisms (T-203) resulted in large increases in root can be used as a measure of the disease
While perhaps not of primary impor- area and cumulative root length, and sig- control efficacy of the biocontrol agent.
tance as mechanisms in biological control, nificant increases in dry weight, shoot The results of many years of research
Trichoderma species exhibit other charac- length, and leaf area over that of the un- with Trichoderma species as biocontrol
teristics during interactions with host treated control. In hydroponic culture, T- agents has shown that not all the mecha-
plants that may contribute to disease resis- 203 inoculated cucumber roots contained nisms and characteristics deemed neces-
tance or tolerance. These characteristics significant increases in Cu, P, Fe, Zn, Mn, sary for optimum biocontrol are found in
manifest themselves by increases in plant and Na. In the shoots of these plants, the the same organism. Very often those strains
root and shoot growth, resistance to biotic concentrations of Zn, P, and Mn were in- that have the capacity to produce enzymes
and abiotic stresses, and changes in the creased. The authors considered that im- and antibiotics that are associated with
nutritional status of the plant. These phe- provement of plant nutrition was directly biocontrol are not the ones that have good
nomena have been thoroughly reviewed by related to the general beneficial growth storage qualities or function well at tem-
Harman (9), who showed that seed treat- effect on the root system of inoculation perature and moisture levels where patho-
ment of corn planted in low nitrogen soil with T. harzianum. gens flourish. Therefore, hybridization of
with T. harzianum (T-22) resulted in plants different strains or species will be required
that were greener and larger in the early Conclusions in order to combine these beneficial char-
part of the growing season. At maturity, the One of the things we have learned acteristics. The production of hybrids from
treated plants had larger stem diameters through research on biocontrol mecha- Trichoderma species, many of which have
and increased yields of grain and silage. In nisms employed by Trichoderma species is no known sexual stage, will entail the use
fields where nitrogen levels were suffi- that the screening methods developed early of transformation or protoplast fusion in

8 Plant Disease / Vol. 87 No. 1

Fig. 9. Cotton seedling disease control field trial: A, preemergence damping-off skips C. R. Howell
occur in the row planted with untreated seed, while B, seed treatment with Tricho-
derma virens results in a complete and even stand. Photograph courtesy R. H. Dr. Howell is a research plant
Garber, Shafter, CA. pathologist with the USDA/ARS at
the Southern Plains Agricultural
Research Center in College
order to obtain strains with optimal sets of Hayes, C. K., Harman, G. E., and Aldwinckle, Station, TX. He received his B.S.
characteristics. Once produced and H. S. 2000. Expression of endochitinase from in biological sciences from Cali-
Trichoderma harzianum in transgenic apple fornia State Polytechnic University
screened, hybrids may yield strains with increases resistance to apple scab and reduces
expanded host, temperature, and moisture in 1962, and a Ph.D. in plant
vigor. Phytopathology 90:72-77.
pathology from Washington State
parameters, and they may yield strains with 5. Chet, I. 1987. Trichoderma-Application,
University in 1967. He then accep-
better storage qualities than those exhibited mode of action, and potential as a biocontrol
ted a position with the USDA/ARS
by the parents. agent of soilborne pathogenic fungi. Pages
137-160 in: Innovative Approaches to Plant at the Texas A&M Experiment
The mechanisms employed by biocon- Disease Control. I. Chet, ed. John Wiley & Station in Lubbock, where he
trol agents to effect biological control of Sons, New York. worked on Verticillium wilt of
plant diseases are many and complex, and 6. Di Pietro, A., Lorito, M., Hayes, C. K., Broad- cotton. In 1970, he transferred to
their use varies with the kind of biocontrol way, R. M., and Harman, G. E. 1993. Endo- the Cotton Pathology Research
agent, pathogen, and host plant involved in chitinase from Gliocladium virens: Isolation, Unit in College Station. He con-
the interaction. Mechanisms are also influ- characterization, and synergistic antifungal tinued work on cotton wilts for 8
activity in combination with gliotoxin. Phyto- years; then he began a research
enced by the soil type, by the temperature, pathology 83:308-313. program on the biological control
pH, and moisture of the plant and soil envi- 7. Elad, Y., and Kapat, A. 1999. The role of of cotton seedling diseases with
ronment, and by other members of the Trichoderma harzianum protease in the bio- bacteria and fungi. His work has
microflora. What we observe and define as control of Botrytis cinerea. Eur. J. Plant
been concerned with the mechan-
biocontrol may be the culmination of a Pathol. 105:177-189.
isms employed by bacteria and
8. Gams, W., and Bisset, J. 1998. Morphology
number of different mechanisms working and identification of Trichoderma. Pages 3-34 fungi to effect seedling disease
synergistically to achieve disease control. in: Trichoderma & Gliocladium. Vol. 1. G. E. control, and with the effect of
Our knowledge of the complexity of these Harman and C. P. Kubicek, eds. Taylor and environment on the activities of
systems is currently limited by our ability Francis, London. biocontrol agents. His current re-
to perceive them, and a great deal of re- 9. Harman, G. E. 2000. Myths and dogmas of search is focused on discovering
search will have to be undertaken in order biocontrol: Changes in perceptions derived the mechanisms in Trichoderma
from research on Trichoderma harzianum T- species that are associated with
to fathom exactly what is taking place 22. Plant Dis. 84:377-393. cotton seedling disease control,
during the biocontrol process. As with so 10. Harman, G. E. 2001. Microbial tools to im- and on transferring beneficial traits
many other aspects of science, basic prove crop performance and profitability and from different sources into se-
knowledge about the mechanisms involved to control plant diseases. Pages 4-1–4-14 in: lected strains to optimize bio-
in the biocontrol process will be of im- Int. Sympos. Biol. Control Plant Dis. New control efficacy.
Century-Mode Action Application Technol.
mense value to those intent on developing 11. Howell, C. R. 1982. Effect of Gliocladium
new methods for utilizing biocontrol virens on Pythium ultimum, Rhizoctonia so-
agents (Fig. 9). lani, and damping-off of cotton seedlings. Gliovirin, a new antibiotic from Gliocladium
Phytopathology 72:496-498. virens, and its role in the biological control of
12. Howell, C. R. 1987. Relevance of mycopara- Pythium ultimum. Can. J. Microbiol. 29:321-
Literature Cited sitism in the biological control of Rhizoctonia 324.
1. Aluko, M. O., and Hering, T. F. 1970. The solani by Gliocladium virens. Phytopathology 16. Howell, C. R., and Stipanovic, R. D. 1995.
mechanism associated with the antagonistic 77:992-994. Mechanisms in the biocontrol of Rhizoctonia
relationship between Corticium solani and 13. Howell, C. R 2002. Cotton seedling preemer- solani-induced cotton seedling disease by
Gliocladium virens. Trans. Br. Mycol. Soc. gence damping-off incited by Rhizopus Gliocladium virens: Antibiosis. Phytopathol-
55:173-179. oryzae and Pythium spp. and its biological ogy 85:469-472.
2. Baek, J. M., Howell, C. R., and Kenerley, C. control with Trichoderma spp. Phytopathol- 17. Howell, C. R., Stipanovic, R. D., and Lums-
M. 1999. The role of extracellular chitinase ogy 92:177-180. den, R. D. 1993. Antibiotic production by
from Trichoderma virens Gv29-8 in the bio- 14. Howell, C. R., Hanson, L. E., Stipanovic, R. strains of Gliocladium virens and its relation
control of Rhizoctonia solani. Curr. Genet. D., and Puckhaber, L. S. 2000. Induction of to the biocontrol of cotton seedling diseases.
35:41-50. terpenoid synthesis in cotton roots and control Biocontrol Sci. Technol. 3:435-441.
3. Bliss, D. E. 1951. The destruction of Armil- of Rhizoctonia solani by seed treatment with 18. Kapat, A., Zimand, G., and Elad, Y. 1998.
laria mellea in citrus soils. Phytopathology Trichoderma virens. Phytopathology 90:248- Effect of two isolates of Trichoderma har-
41:665-683. 252. zianum on the activity of hydrolytic enzymes
4. Bolar, J. P., Norelli, J. L., Wong, K.-W., 15. Howell, C. R., and Stipanovic, R. D. 1983. produced by Botrytis cinerea. Physiol. Mol.

Plant Disease / January 2003 9

 Plant Pathol. 52:127-137. vorum in white rot affected onion roots by 33. Weindling, R. 1941. Experimental considera-
19. Lifshitz, R., Windham, M. T., and Baker, R. Trichoderma koningii. Plant Pathol. 50:249- tion of the mold toxin of Gliocladium and
1986. Mechanism of biological control of 257. Trichoderma. Phytopathology 31:991-1003.
preemergence damping-off of pea by seed 26. Migheli, Q., Gonzalez-Candelas, L., Dealessi, 34. Wells, H. D., Bell, D. K., and Jaworski, C. A.
treatment with Trichoderma spp. Phytopathol- L., Camponogara, A., and Ramon-Vidal, D. 1972. Efficacy of Trichoderma harzianum as
ogy 76:720-725. 1998. Transformants of Trichoderma longi- a biocontrol for Sclerotium rolfsii. Phytopa-
20. Lo, C.-T., Nelson, E. B., Hayes, C. K., and brachiatum overexpressing the b-1,4-endo- thology 62:442-447.
Harman, G. E. 1998. Ecological studies of glucanase gene egl1 show enhanced biocon- 35. Wilhite, S. E., Lumsden, R. D., and Straney,
transformed Trichoderma harzianum strain trol of Pythium ultimum on cucumber. Phyto- D. C. 1994. Mutational analysis of gliotoxin
1295-22 in the rhizosphere and on the phyl- pathology 88:673-677. production by the biocontrol fungus Gliocla-
loplane of creeping bentgrass. Phytopathol- 27. Rehner, S. A., and Samuels, G. J. 1994. Tax- dium virens in relation to suppression of Py-
ogy 88:129-136. onomy and phylogeny of Gliocladium ana- thium damping-off. Phytopathology 84:816-
21. Lorito, M., Woo, S. L., D’Ambrosio, M., lyzed by large subunit ribosomal DNA se- 821.
Harman, G. E., Hayes, C. K., Kubicek, C. P., quences. Mycol. Res. 98:625-634. 36. Woo, S. L., Donzelli, B., Scala, F., Mach, R.,
and Scala, F. 1996. Synergistic interaction be- 28. Schirmbock, M., Lorito, M., Wang, Y. L., Harman, G. E., Kubicek, C. P., Del Sorbo, G.,
tween cell wall degrading enzymes and mem- Hayes, C. K., Arisan-Atac, I., Scala, F., Har- and Lorito, M. 1999. Disruption of the ech42
brane affecting compounds. Mol. Plant-Mi- man, G. E., and Kubicek, C. P. 1994. Parallel (endochitinase-encoding) gene affects biocon-
crobe Interact. 9:206-213. formation and synergism of hydrolytic en- trol activity in Trichoderma harzianum P1.
22. Lorito, M., Woo, S. L., Fernandez, I. G., Col- zymes and peptaibol antibiotics, molecular Mol. Plant-Microbe Interact. 12:419-429.
lucci, G., Harman, G. E., Pintor-Toros, J. A., mechanisms involved in the antagonistic ac- 37. Yedidia, I., Benhamou, N., and Chet, I. 1999.
Filippone, E., Muccifora, S., Lawrence, C. B., tion of Trichoderma harzianum against phyto- Induction of defense responses in cucumber
Zoina, A., Tuzun, S., and Scala, F. 1998. pathogenic fungi. Appl. Environ. Microbiol. plants (Cucumis sativus L.) by the biocontrol
Genes from mycoparasitic fungi as a source 60:4364-4370. agent Trichoderma harzianum. Appl. Environ.
for improving plant resistance to fungal path- 29. Sharon, E., Bar-Eyal, M., Chet, I., Herra- Microbiol. 65:1061-1070.
ogens. Proc. Natl. Acad. Sci. USA 95:7860- Estrella, A., Kleifeld, O., and Spiegel, Y. 38. Yedidia, I., Benhamou, N., Kapulnik, Y., and
7865. 2001. Biological control of the root-knot Chet, I. 2000. Induction and accumulation of
23. Lumsden, R. D., Locke, J. C., Adkins, S. T., nematode Meloidogyne javanica by Tricho- PR proteins activity during early stages of
Walter, J. F., and Ridout, C. J. 1992. Isolation derma harzianum. Phytopathology 91:687- root colonization by the mycoparasite Tricho-
and localization of the antibiotic gliotoxin 693. derma harzianum strain T-203. Plant Physiol.
produced by Gliocladium virens from alginate 30. Webster, J., and Lomas, N. 1964. Does Biochem. 38:863-873.
prill in soil and soilless media. Phytopathol- Trichoderma viride produce gliotoxin and vir- 39. Yedidia, I., Srivastva, A. K., Kapulnik, Y., and
ogy 82:230-235. idin? Trans. Br. Mycol. Soc. 47:535-540. Chet, I. 2001. Effect of Trichoderma har-
24. McSpadden Gardener, B. B., and Fravel, D. 31. Weindling, R. 1932. Trichoderma lignorum as zianum on microelement concentrations and
R. 2002. Biological control of plant patho- a parasite of other soil fungi. Phytopathology increased growth of cucumber plants. Plant
gens: Research, commercialization, and appli- 22:837-845. Soil 235:235-242.
cation in the USA. Plant Health Progress. On- 32. Weindling, R. 1934. Studies on a lethal prin- 40. Zhang, J., Howell, C. R., and Starr, J. L. 1996.
line, publication doi:10.1094/PHP-2002- ciple effective in the parasitic action of Suppression of Fusarium colonization of cot-
0510-01-RV. Trichoderma lignorum on Rhizoctonia solani ton roots and Fusarium wilt by seed treat-
25. Metcalf, D. D., and Wilson, C. R. 2001. The and other soil fungi. Phytopathology 24:1153- ments with Gliocladium virens and Bacillus
process of antagonism of Sclerotium cepi- 1179. subtilis. Biocontrol Sci. Technol. 6:175-187.

10 Plant Disease / Vol. 87 No. 1