Biological control of eutypa dieback of grapevines: interactions between the pathogen and fungal antagonists
Sharmini John
Thesis submitted for the degree of Doctor of Philosophy at the lJniversity of Adelaide
School of Agriculture and Wine Faculty of Sciences
May 2003 DECLARATI.N""" """rrr
PUBLICATIONS AND CONFERENCE PROCEEDINGS vl
CHAPTER I.INTRODUCTION...... 1
CHAPTER 2. LITERATURE REVIE\ry...... 4
2.1. Introduction. .4
2.2,The pathogen 5
2.2.I . Historic al background 5
2.2.2. T axonomy and nomenclature 6
2.2.3. Biology of E.lata .8
2.2.3. 1. Disease cycle..... 8
2.2. 3.2. Histopatholo gy 10
2.2.4. Symptoms 10
2.2.5. Variability of the pathogen. .. .. 11
2.2.6. Environmental factors that affect disease development.... 12
13 2.2.6.1. Temperature and relative humidity. . . . . ' 2.2.6.2. Rainfall t4
2.3. The host plant. """'15
2.3.1. Yield loss...... r6
Z.4.Management of eutypa dieback.... """"'16
2.4.I. Cultural practices t6
2.4.2. Chemical control t7
2.4.3. Biological control. . . ..
2.4.3.1. The needfor biological control 18
2.4.3.2. Microbes tested against eutypa dieback 19
.20 2.4.3.3. Trichoderma spp. in biological control. '.... '
2.4.3.4. Mechanisms of antagonism by Trichoderma spp. 22
2.5. Other trunk diseases.... .""""""26
2.6. Summary .27
CHAPTER 3. GENERAL MATERIALS AND METHODS. ..28
3.2. The pathogen.. .29
3.2.I. Isolates of E. lata 29 3 .2.2. Extr actions of ascospores ..30
3.2.3. Y iability testing of ascospores. .30
3.3. The antagonists. .32
3.3.1. Strains...... ,,32
3.3.2. Commercial formulations ,.JJ
3.3.3. Enumeration of colony forming units (CFU) .33
3.4. Re-isolation of fungi from wood. .35
3.4.1. Re-isolation of pathogen 35
3.4.2. Re-isolation of antagonist. . .36
3.5. Maintenance of isolates and cultures...... ""37
3.6. Statistical analysis 37
CHAPTER 4. INTERACTION STUDIES IN VITRO .38
4.2.Materials and Methods...... "40
4.2.1. Mode of inhibition .40
4. 2. 1. 1 . Antibiosis by volatile metab olite s . . . .40
4.2. 1. 2. Antibio sis by non-v olatile metabolite s 4l 4.2. 1. 3. Parasitism... .43
4.2.2.Interactions on cane segments. . . .43
4.2.2.1. Colonisation of wood byE.lata in the presence of T.haruianum - experiment I 44
4.2.2.2. Colonisation of wood by E.lata in the presence of T.harzian'tm - experiment 2 .45
4.2.2.3. Colonisation of wood byE.latain the presence ofT.haruianum -
experiment 3 ... 46
4.2.3. Microscopy .46
4.2.3.1. Processing of woodfor SEM 41
4. 2. 3. 2. Ino culation of g amma- irradiate d w o o d .47
4.2.3.3. Inoculation of canes Sfown in rochuool pieces in the laboratory...... 48
4.2.3.4. Simultaneous co-inoculation of canes grown in rochuool pieces in the Iaboratory 49
4.3.I. Mechanisms of inhibition in vitro .49
4.3.1.1. Inhibition by volatile metabolites .49
4. 3. 1 .2. Inhibition by non-v olatile metab olite s 50 4. 3. l. 3. Inhibition by parasitism... .51
4.3.2.Interaction studies on cane segments .51
4.3.2. t. Interactions between pathogen and antaSonist - experiment I . -.. '...... '..51
4.3.2.2. Interactions betyveen pathogen and antagonist - experiment 2. '.. ' ' ' '...... 52
4.3.2.3. Interactions between pathogen and antagonist - experiment 3 .52
4.3.3. Microscopy 6l
4. 3. 3. 1 . Int eraction in gamma-irradiated w ood...... 6l
4.3.3.2. Interaction in cuttings Srown in rockwool pieces in the laboratory... . . ' "61
4.3.3.3. Interactions in simultaneously co-inoculated cuttings grown in
roclantool pieces in the laboratory. 62
CHAPTER 5. GLASSHOUSE EXPERIMENTS...... 76
5.1.Introduction...... """'76
5.2. Materials and Methods.. ""78
5.2.1. Experiment 5.1. Colonisation of cuttings by antagonists. 18
5.2.2.Experiment 5.2.Effect of T. harzianum oî infection of cuttings by E. lata. 80 5.2.3.Experiment 5.3. Effect of antagonists on infection of cuttings by E. lata. '. 81
5.2.4.Experiment 5.4. Protection of pruning wounds from infection by E.lata 82
5.2.5. Experiment 5.5. Protection of pruning wounds with Trichoseal@
5.2.6.Experiment 5.6. Protection by prior inoculation with antagonist...'...... '.....84
5.3.1. Experiment 5.1. Extent of colonisation by antagonists
5.3.2.Experiment 5.2.Protection of cuttings by T' harzianum .94
5.3.3. Experiment 5.3. Protection of cuttings by antagonists .95
5.3.4. Experiment 5.4. Protection of pruning wounds...... 97
5.3.5. Experiment 5.5. Pruning wound treatments 99
5.3.6. Experiment 5.6. Colonisation of cuttings by pathogen in the presence of antagontst 105
CHAPTER 6. FIELD EXPERIMENTS...... 115
6.1. Introduction...... 115
6.2. Materials and methods...... "'117 6.2.I. Pruning wound trials. tt7
6.2. 1. 1. Nuriootpa 1 ... tt7
6.2.1.2. Nuriootpa 2
6.2. 1.3. Eden Valley...... 118
6.2.1.4. Waniparinga I 119
6.2. l. 5. Watiparinga 2.. 119
6.2.2. Trichodowel trial. r20
6.2.3.Injection trial t2l
6.3. Results ""L22
6.3.I. Nuriootpa L,2 and Eden Valley... 122
6.3.2. Warriparin ga I, 2 124
6.3.3. Results of Trichodowel trial 130
6.3.4. Results of injection trial 131
6.4. Discussion... ""131
CHAPTER 7. PRE,LIMINARY STTJDIES OF WOUND R8SPONS8...... I45
7.1..Introduction...... ""145 1.2.Materials and Methods.. """"148
148 7 .2.I.Inoculation and harvest of canes..... '
149 1 .2.2. Detection of lignin
7.2.3. Detection of suberin. .149
149 7 .2.4. Detection of phenolic compounds
.150
7.3.1. Lignin. 150
7.3.2. Suberin. ..150
7.3.3. Phenolic compounds.. 151
7.4. Discussion...... """"161
CHAPTER 8. GENERAL DISCUSSION 165
165 8.1.. Introduction......
8.2. Summary of findings... """"""165
S.3. Implications of findings and future research. . . " " " " "166 APPENDTX 1""""' """""""217
APPENDIX 3 .22L ABSTRACT
Biological control of eutypa dieback of grapevines using Trichoderma harzianum was investigated in laboratory, glasshouse and field experiments. Fusarium lateritium, a fungus known to be an effective antagonist of E. lata, was also used in some experiments.
T. harzianuLø inhibited mycelial growth and germination of ascospores of E' lata by antibiosis on potato dextrose agar medium. The three strains investigated inhibited mycelial growth by production of both volatile and non-volatile antibiotics, although the degree of inhibition varied between strains of the antagonist and between isolates of the pathogen. The non-volatile antibiotics had a fungistatic effect on some isolates of E. Iata
and a fungicidal effect on others. Scanning electron microscopic examination of co-
inoculated gamma-irradiated grapevine cane segments and of co-inoculated l-year-old-
canes placed in water-saturated rockwool in the laboratory showed hyphae with loss of
turgor and collapse, abnormal swelling, winding and parallel growth.
Significant reduction in infection by E. lata was detected in vitro when autoclaved or
gamma-inadiated canes were inoculated with mycelial plugs or spores of both pathogen
and antagonist. When pruning wounds on l-year-old canes of cultivar Shiraz in the
glasshouse were treated with spores of T. harzianum, then challenged 2 and 7 days later
with mycelial plugs of E. lata, infection by E. lata was reduced significantly at both
times. The pathogen was recovered from 13-38{,o andO-257o of the canes treated with
the antagonist and challenged 2 andT days later, respectively, with the pathogen. E. lata ii was recovered from all of the controls at both times of treatment. Furthermore, in 12 weeks T. harzianurz colonised the canes up to 10 cm below the point of inoculation.
In the vineyard, five trials were established to test pruning wound treatments over 3 years using the cultivars Cabernet Sauvignon, Shiraz, Rondella and Palomino' Treatment of pruning wounds with Z. harzianum or F. lateritium protected vines from infection by
ascospores of E. lata, in most experiments, when the wounds were challenged24 hours
or 14 days after treatment with the antagonist. In most of the pruning wound trials, the
application of spore suspensions of the antagonists significantly (P<0.001) reduced
infection by the pathogen. The percentage recovery of E. lata from the control vines that
had been inoculated with E. lata alone was low in four of the five trials. Also,
colonisation of vines by T. harzianum was studied following insertion of T. harzianum-
impregnated wooden dowels (Trichodowels@) into holes drilled 30 cm above ground
level into trunks of vines of cultivar Nyora. T. harzianum was re-isolated from four of
the seven vines 20 months after inoculation and had grown 18 cm above the point of
inoculation in two of the vines during this time'
Results suggested thatT. harzianum has potential in the control of eutypa dieback of
grapevines when used as a pruning wound treatment. While results of experiments with
the Trichodowels@ were encouraging, there is a need for more detailed studies of their
efficacy in preventing infection of vines by E' lata. lll
Declaration
This thesis contains no material which has been accepted for the award of any other degree or diploma in any University or other tertiary institution and, to the best of my knowledge and belief, contains no material previously published or written by another person, except where due reference has been made in the text'
I give consent to this thesis being made available for loan and photocopying when
deposited in the University Library.
olfll,,(u= lv
ACKNO\ryLEDGEMENTS
I sincerely thank my supervisors, Dr Eileen Scott, Dr Trevor Wicks and Dr John Hunt' for their valuable guidance, helpful criticism and advice during the course ofthis project.
I particularly wish to thank Dr Scott for her efforts in maintaining a friendly working environment in the laboratory and within the research group.
I also wish to acknowledge:
I Agrimm Technologies Ltd, New Zealand and the University of Adelaide for funding
this project.
t Dr Mary Cole, Monash University for supplying E. lata isolates M280, M295 and
M302.
I Ms Lyn Waterhouse for assistance with Scanning Electron Microscopy and Mr John
Terlet and the staff of Adelaide Microscopy for their support during the time I spent
in that facility.
t Steritech, Victoria for assisting with gamma-irradiation of grapevine wood and
Trichoprotection@ products'
I The Cooperative Research Centre for Viticulture (CRCV) and the Grape and Wine
Research and Development Corporation (GWRDC) National Grapevine Trunk
Diseases Project for allowing the use of trial sites at Nuriootpa and Eden Valley, in
collaboration with Dr Mette Creaser.
t Ms Nari Anderson and Ms Karolina Pniewska, for technical assistance during the
latter phase of this Project. I Ms Michelle Lorimer and Ms Helena Oakey (BiometricsSA) for statistical advice and
assistance.
I Members of laboratories N105 and N107, especially Dr Evelina Facelli, Mr Peter
Crisp and Mr Richard Lardner for their friendship and help. Special thanks also to Dr
Belinda Rawnsley and Dr Sandra Savocchia who befriended and helped me initially
when I joined the department.
I My parents and family for their encouragement and support throughout the years. a My husband, Roshan John, for his help, encouragement and patience during the
course of this study.
I Finally, most importantly, God who enabled me to undertake and complete this
project. v1
PUBLICATIONS AND CONFERE,NCE PROCEEDINGS
John, S., E.S. Scott, T. Wicks, J. Hunt, 2003. Towards biological control of eutypa
dieback of grapevines. 8th International Congress of Plant Pathology. Abstracts of
Offered Papers. Christchurch, New Zealand. p' 43
John, S., E.S. Scott, T.J. Wicks, J.S. Hunt, 2003. Studies of interactions between Eutypa
lata and Trichoderma harzianum. 3'd International Workshop on Grapevine Trunk
Diseases. Lincoln University, New Zealand. p. 33.
John, S., E.S. Scott, T.J. Wicks, J.S. Hunt, 2003. Studies of interactions between Eutypa
gi (submitted)' I at a and T r i ch o de rma har zi anum. Phytop athol o a Mediterranea
John, S., E.S. Scott, T. Wicks, J. Hunt, 2001. Interactions between E. lata and
Trichoderma sp. 13ù Biennial Australasian Plant Pathology Society Conference. Cairns,
Australia. p. 338.
John, S., R. Lardner, E. Scott, B. Stummer, T. Wicks,200l. Eutypa dieback: research on
biological control and diagnostics. The Australian Grapegrower and Winemaker. 449a:
73-75 1
CHAPTER 1. INTRODUCTION
Eutypa dieback is a destructive canker disease caused by the ascomycetous fungus
Eutypa lata (Pers.: Fr.) Tul. & C. Tul. (syn. Eutypa armeniacae Hansf. & M. V. Carter)
(Moller and Kasimatis, 1978; Moller and Kasimatis, 1981a). This disease affects woody plants in 88 species within 28 families and is widely distributed in the temperate regions around the globe (Bolay and Carter, 1985; Carter,l99l). Grapevine (yilts spp'), apricot
(Prunus armeniaca L.), almond (P. dulcls (Mill.) Webb), apple (Malus domesticq
Borkh.) and sweet cherry (Prunus cerasus L.) are some of the hosts of agricultural importance.
This disease is most prevalent in older vines and in areas that receive an annual rainfall of more than 500 mm (Wicks and Hall, 1997). The disease is usually rare until a vineyard reaches the age of 10-12 years after which the incidence increases until almost all the vines are infected at 20 years (Duthie et aL,1991; Munkvold and Marois, 1995).
Eventually the disease kills the vines and, therefore, reduces the longevity of the vineyards.
Significant yield losses afe caused by eutypa dieback in grape-growing areas throughout rhe world (Carter, 1991). Wicks and Hall (1997) have suggested that in
Australia up to 60 7o of the vines may be affected in some old, elite vineyards' Estimates made in North America show yield losses of 30-627o (Munkvold et al., t994)' In South
Australia the disease is common in Barrosa Valley, Coonawarra, Clare Valley, Eden
Valley and Mclaren Vale areas. Wicks and Davies (1999) estimated yield loss of more 2 than $2,800 per hectare in Shiraz, in Eden Valley, where 477o of the vines showed symptoms of infection.
Treating wounds with fungicides or wound sealants is the preferred method of
control, since the pathogen enters its host through pruning wounds (Moller and
Kasimatis, 1978; Moller and Kasimatis, 1981b). Eradicative sanitation methods are not
always successful because of the broad host spectrum of the pathogen (Carter, 1991).
Benomyl has been reported to be effective against E. lata (Munkvold and Marois,
1993b), but this fungicide was never registered in Australia as a grapevine wound
protectant. The manufacturers withdrew benomyl from the market in20D2. Also, pruning
wounds may remain susceptible for 4 weeks (Munkvold and Marois, 1994), hence
chemical treatments may not protect the wound for the entire period of susceptibility.
Biological control agents capable of colonising pruning wounds and the tissues below
the wounds may provide better long-term protection than fungicides.
Biological control of E. lata on grapevines has been demonstrated using F' Iateritium,
Cladosporium herbarum and Bacillus subtilis (Ferreira et al., 1991; Munkvold and
Marois, 1993a), however, none of these have yet been developed into commercial
products. Trichoprotection@ products, manufactured by Agrimm Technologies Ltd, New
Zealand, containing seven strains of Trichoderma harzianum) have been registered for
protection against the silver leaf pathogen Chondrostereum purpureum on stone fruit
trees since 1991, and have been suggested to be effective against E. Iata on grapevines
(Hunt, lg99). The potential for biological control of E. hta using T. harzianum was
investigated in this project. 3
The aims of this study were to: (1) elucidate mechanism/s of inhibition of E. lataby
T. harzianurn in vitro; (2) investigate interactions of pathogen and antagonist on grapevine tissues in the laboratory; (3) study inhibition of E. Iata by T. harzianum \n grapevine cuttings in the glasshouse; (4) investigate the potential of T. harzianum as a biological agent to prevent infection by E.latain the vineyard, with emphasis on pruning wound treatments and (5) study the wound response of the host to understand interactions between host, pathogen and antagonist' 4
CIIAPTER 2. LITERATURE REVIEW
2.1. Introduction
Eutypa dieback may be managed by treating pruning wounds with fungicides and delaying pruning till late in the dormant season (Moller and Kasimatis, 1980; Moller and
Kasimatis, 198lb; Petzoldt et a1.,1981). These management practices, however, are only partially effective. Sanitation pruning is usually used in premium wine-growing areas' but is labour intensive and expensive. In view of the threat of eutypa dieback to the sustainability of South Australian vineyards and the possibility of the development of fungicide resistant strains of the pathogen, there is an urgent need to develop a suite of effective control measures including cultural, chemical and biological control methods.
The literature on eutypa dieback and related trunk diseases of grapevines is reviewed in this chapter by dividing the relevant information into four main categories. The first section reviews the literature on the pathogen in terms of its historical background, taxonomy and nomenclature, biology, genetic variability, symptoms of the disease and the environmental factors that affect disease development. The second section deals with the host. In the third section the various control practices are discussed with emphasis on
research using Trichoderma spp. in biological control. Other trunk diseases of
grapevines are reviewed in the fourth section. 5
2.2. The pathogen
2,2.1. Historical background
Eutypa dieback was initially recognized in the 1920s in Australia as a disease of apricot trees. In the South Australian Journal of Agriculture in 1924 and 1925, it was reported that old apricots in many parts of the southern district were losing big limbs and this was attributed to "root trouble" (Carter, 1991). Dowson (1931) provided the first detailed description of the disease, stating that it was a definite dying back of both young and old branches, with fungal mycelium invading the dead or dying tissue. The presence of this fungus was thought to cause the trees to produce gum to such an extent that the sap conducting vessels of the wood became clogged and the water supply was affected, resulting in the death of branches. The infection could be traced to old pruning wounds
(Dowson, 1931). Harris (1932) observed the same symptoms in the Barossa district of
South Australia, and called the condition "gummosis". Adam (1938) confirmed the tentative diagnosis of Samuel (1933) of the conidial stage of the pathogen as
Cytosporina sp. However, the mode of reproduction and dissemination remained obscure until 1956, when the sexual stage, Eutypa armeniacae Hansf. & Carter, was identified at the'Waite Agricultural Research Institute in South Australia (Carter, 1957a).
Eutypa armeniacae was first reported as a saprophyte on grapevine in 1957 in
Adelaide, South Australia (Carter, 1957b). A decade later, Moller et aI. (1968) found it
on dead arms of grape in California. Meanwhile, in Europe, the term "apoplexy" was
widely used to describe any disease manifest by sudden wilting of leaves of apricot trees,
which was found to be widely distributed and increasing in frequency. No single 6 organism had been consistently associated with the disease and, therefore, it was attributed to numerous and diverse physiological and pathological factors (Carter, 1991).
Reports of dieback symptoms in apricot trees in the 1970s in Spain, Italy, Greece'
Bulgaria and Libya confirmed the presence of E. armeniacae in Europe and the
Mediterranean region but not a single record of the pathogen existed in the UK until
1985 (Carter, 1991). Meanwhile, Carter and Price (1974) confirmed the pathogen in apricot trees in South Africa.
Eutypa dieback of grapevines was confused with other fungal disorders because of similarities of symptoms that had long been attributed to other pathogens such as
Phomopsis viticola Sacc.. Research by Moller and Kasimatis (1978) revealed that E. armeniacae was the pathogen causing dieback symptoms on grapevine. The pathogen is now commonly known as E.lata (Carter, 1991).
2.2.2. Taxonomy and nomenclature of the pathogen
E. lata belongs to the Ascomycota and is classified in the single family Diatrypaceae of the order Diatrypales. The fungi grouped in this family produce asci with thin walls, long stalks and small, iodine-positive apical rings, in which pale brown or yellow allantoid ascospores are formed. Ascospores afe 6.2 - 11 x 1.5 - 2ltm in size and aseptate. The asci are contained in perithecia, which usually have furrowed ostioles
(Carter, 1991). The perithecia ate produced in stromata on dead host tissue'
The teleomorph of E. lata develops in the wood as extensive sffomata. The ostioles
emerge separately, projecting 50 - 150 pm, are rounded or conical, never cruciform, and
tz1 - 180 pm in diameter. The single layered ascomata are closely spaced or in contact, 7 sometimes compressed together. These are spherical or ovoid in shape (300 -) 400 - 600
(- 700) pm in diam. with short necks. The anamorph is found as conidiomata on natural substrates. Conidia are produced in white to cream translucent masses on agar, single celled, filamentous, curved towards the obtuse apex and straight towards the truncate base, (14-) 20-35 (-40) pm long xl- 1.3 ¡"rm thick whereas on natural substrates, conidia are (20-) 30- 45 (-50) U,m in length (Carter, l99l). The anamorph of E. lata is now placed in the genus Libertella (Carter, 1991). The ascospores are responsible for the dissemination of the pathogen and conidia may act only as spermatia (Carter, 1991).
However, some researchers observed conidial germination and have suggested that these can propagate disease (Belarbi and Mur, 1983; Ju et al.,l99l).
E. lata is one of the five species contained in the genus Eutypa, which was first established by Tulasne and Tulasne (Carter, 1991). Glawe and Rogers (1982) described the anamorphs of six members of the family Diatrypaceae in detall. E. lata and E. armeniacae were retained as separate species but it was noted that these two species differed very little or not at all in their morphological characteristics (Glawe and Rogers,
1982; Carter, l99l). More recently, De Scenzo et al. (1999) used molecular evidence to suggest that these are two different pathogenic species. Both species were capable of infecting native and cultivated hosts in California. 8
2.2.3. Biology of Eutypalata
2.2.3.1. Disease cycle
Development of eutypa dieback (Figure 2.1) begins in grapevine when the ascospores enter the plant via exposed xylem vessels of fresh pruning cuts (Carter, 1960; Carter,
1965; Moller and Kasimatis, 1978). The ascospores germinate in the vascular tissues, and hyphae proliferate slowly within the vessels and later colonise the associated elements of the functional wood. The progress of the disease in grape is slow and no symptoms are seen for one or two seasons, after which foliar symptoms are manifested
(Moller and Kasimatis, 1978). A canker is usually apparent by the third or fourth season after infection and many more growing seasons may elapse as the disease progresses and eventually kills the affected arm or cordon (Pearson and Goheen, 1988). Once death occurs, it takes several more years before perithecia are formed on the dead infected wood (Flaherty et al., Igg2). Ascospores are generally released during and soon after rainfall and are windborne to infect fresh wounds, in spring, sunìmer and autumn. Rain is necessary for the release of the ascospores from the perithecia and their entry into the exposed ends of the xylem vessels (Pearson and Goheen, 1980). According to Carter
(1957a), a minimum rainfall of 2 mm is necessary to initiate the liberation of ascospores when the stromata are dry. Wind is necessary for the aerial transport of ascospores.
Studies by Pearson and Goheen (1988) suggested that ascospores are able to travel 50-
100 km. Generally, dispersal occurs during overcast weather when solar radiation is
minimal (Ramos et al.,I975b). 9
Figure 2.L. Disease cycle of eutypa dieback (Flaherty et al., 1992).
DEAO I¡IFECTED WOOÐ
ËUTYPA TIEBÀTK If'IF€CTËÛ GR¡,FçVIHE
Ffi UITI NG ËODIE$ (PERITHEçIA) PRODUCËÐ IN OLD DÊÃD WOOD
CANKER f
SPORES PRODUT.ED IN SPRING, SUIóMER ANþ FRESH I,SROE AUTUMN SP0RES ÇËftMlþÌÅTÊ WÜUND 1þ¡' XYLEM VESSTLS In $PORE$ DISCHARGED IN WËT WEATHER 10
2. 2. 3. 2. Hístop atholo gy
Infection is initiated when the ascospores are washed into fresh wounds on grapevine
wood more than ayear old. Ramos et al. (1975a) demonstrated that a single ascospore is
sufficient to initiate infection. According to Moller & Kasimatis (1980), wounds that are
made on 1 year-old wood are less susceptible to infection than those made on older
wood. Generally, viable spores germinate within 12-24 h at the optimal temperature of
20-25"C, usually 2 mm or more beneath the wound surface (Carter, 1991). Symptoms
develop 2 to 3 years after infection, when the fungus has colonised the vascular tissue of
the arm and trunk of the vine (Moller and Kasimatis, 1981a; Duthie et aI., 1991). The
pathogen releases phytotoxic compounds, one of which has been isolated and identified
as eutypine, 4-hydroxy-3-(3-methyl-3-butene-1-ynyl) benzaldehyde (Tey-Rtlh et al.,
1q¡gl). The toxin was believed to be responsible for the expression of disease symptoms
(Mauro et a1.,1988; Tey-Rulh ¿/ al., l99l). However, a recent report suggests that the
fungus produces a suite of potentially toxic secondary metabolites that may cause the
symptoms (Molyneux et a1.,2002).
2.2.4. Symptoms
Symptoms are seldom observed in vines younger than 6 years of age (Moller and
Kasimatis, 1980). The disease is obvious in early spring when the new season's shoots
appeff stunted, defonned and discoloured.
Generally, the young leaves are small, chlorotic, often cupped with tattered margins
(Moller and Kasimatis, 1981a). Brown speckles develop on leaves and the margins
appear scorched (Pearson and Goheen, 1988; Flahetty et al., 1992: Emmett and 1l
Magarey, Igg4). The shoots usually have "zig-zagged" internodes. The disease first appears on one or two spurs, but becomes more extensive with each passing year, killing the infected arm and eventually the entire grapevine. Formation of pruning wound cankers is an important diagnostic feature of dieback disease (Moller and Kasimatis,
198la; Duthie et a1.,1991). The dead tissue surrounds old pruning wounds and appears as a wedge-shaped necrotic zone in the cross-section of the wood (Flaherty et al., 1992).
Creaser and Wicks (2002) reported that in severely affected vines the wood symptoms might progress into the trunks below ground level but do not extend into the root system.
Flower clusters tend to shrivel and die on severely affected shoots. Clusters on shoots that are affected by the disease may have a mixture of large and small berries' Usually, symptoms are detected only after the fungus has become well established in the grapevine tissue. This makes disease management difficult.
2.2.5. Variability of the pathogen
E. lata exists as a range of genotypes that differ in virulence to individual hosts. This is expected since the pathogen is an ascomycete which relies exclusively on ascospores for its propagation and dissemination (Carter, 1991). Péros and Berger (1994) showed that isolates differ in their ability to causes symptoms. They demonstrated that some
isolates that colonised grapevine cuttings did not induce foliar symptoms. Pétos et al.
(Igg7) also reported that E. lata isolates show a large variation in pathogenicity on
cuttings in the geenhouse, but qualitative differences in pathogenicity of isolates did not
correspond to the presence of symptoms in the vineyard. Differences in pathogenicity
have been noticed among E. lata isolates from the same stroma/perithecium (Cattet et 12 al., !985; Rumbos, 1987; Péros and Berger, 1994) and even among single spore-derived isolates of the same ascus (English et a1.,1983). Research has also revealed differences in pathogenicity between isolates from different geographic locations (Ramos et al.,
I975a; Péros and Berger, 1994). Furthermore, genetic diversity in E. lata has been demonstrated recently using random amplified polymorphic DNA (RAPD) markers
(Péros and Berger, 1999: Péros et a1.,1999). Variation has also been observed in cultural ftaits (Glaw e et al., 1982; English et al., 1983; Rumbos, 1987; Peros and Berger, 1994)'
This diversity has to be taken into account when devising disease control strategies.
Differences in virulence of E. lata have been demonstrated on apricots (Ramos et al.,
1975a; Carter et a\.,1985).
Carter (1991) has also mentioned the possibility of hypovirulence transmission between isolates, which might render virulent strains ineffective when they are co- inoculated with an avirulent strain. He considered that hypovirulence could be a dominant character in E. Iata. Apart from Carter's observations, not much is known about interactions between the different pathotypes of E. lata. This aspect needs further investigation, as hypovirulence could be useful in devising effective management strategies.
2.2.6. Environmental factors that affect disease development
It has been established that a two-fold process is involved in the release,
dissemination and deposition of the propagules of the pathogen (Carter, 1965; Moller
and Carter, 1965). Ascospore release occurs after rainfall and spores are disseminated
and deposited by wind. Subsequent showers of rain aid in the redistribution of the 13 inoculum. The viable ascospores that reach the infection courts germinate, penetrate the vascular system ancl grow into woody tissue, where they are protected from the ever- changing environmental conditions. The major environmental factors that affect infection are temperature, relative humidity and rainfall'
2.2.6.1. Temperøture and relatíve humidity
The time required for germination of ascospores is directly related to temperature.
Carter (1991) showed that germination of freshly discharged ascospores was maximal at
Z0-25"C on agar within ll-12 h, whereas spores required more than 100 h to germinate at 2"C. Munkvold and Marois (1995) reported that the optimum conditions for germination and growth of E. lata are22-25"C and at leastg0% relative humidity (RH).
According to Carter (1951a), if temperatures ale below 15oC, the minimum RH must be maintained for longer periods of time for germination to occur. Apricot trees have been shown to become resistant to infection by E. Iata when the mean maximum daily temperature was 20oC and RH 60 7o + 8 but were susceptible at the dormant temperature of 3oC (Ramos et al., L975a). Similar observations have been made in California, where infection by E. lata was less on gtapevines pruned towards the end of winter, than on those pruned in early winter (Tresse et a1.,1982). Tresse et al. (1980) also demonstrated
that E, lata infecfs pruning wounds on grapevines under controlled conditions at -1oC to
+loc, although infection under these conditions was infrequent. Their freeze and thaw
tests also suggested that ascospores can germinate during winter and early spring, where
temperatures alternate between above freezing and below freezing (e'g' -20"C)
temperatures. l4
2.2.6.2. Rainfall
Liberation of ascospores begins within 3 h of rainfall, continues during the rainy
period, and ceases after rainfall ends and stromata become dry (Pearson, 1980). Ramos ¿/
al. (1975b) demonstrated that a minimum rainfall of l.2l Íìm was necessary for
ascospore discharge. Tresse et al. (1980) reported the presence of octads of airborne
ascospores after the stromata were exposed to a minimum of approximately 2 mm of
rainfall at temperatures above 0"C. They also reported that free water (vine trunk
wetness) maintained by prolonged rain resulted in continued ascospore dispersal fot 24
h. The numbers of ascospore octads were high in the spring and declined in summer. In
sub-freezing weather conditions during winter, when there was no rainfall, no ascospores
were trapped (Tresse et a1.,1980)'
Usually, eutypa dieback occurs abundantly in areas of mean annual rainfall of more
than 600 mm, but is unlikely to develop where rainfall is less than 250 mm (Pearson and
Goheen, 1938). Perithecia are exceedingly rare and the incidence of the disease is low in
regions where the mean annual rainfall is below 219 mm (Carter, 1957a)' Similar
observations of the relationship between mean annual rainfall and the distribution and
occurrence of the perithecial stage have been made by Ramos et al. (I975b)' Pearson
(1980) suggested that ascospore release may also be triggered by natural snowmelts'
Ramos et at. (1975b) demonstrated that wind direction and speed also have an effect on
numbers of airborne ascospores. Many ascospores were observed in the atmosphere
when srrong winds (8 - 11 km/h) followed rain, but light wind conditions (2 km/h)
following rain did not favour the aerial transport of ascospores (Ramos et aI., 1975b). 15
Rainfall and wind speeds are, therefore, important factors that need to be taken into account when pruning, to minimise the risk of infection.
2.3. The host plant
Grapes belong to the family Vitaceae, genus Vitis. The Eurasian species, Vltis vinifura, comprises the majority of genetic material (apart from rootstocks) used for viticulture in Australia and in many other grape-growing countries (Dry and Gregory
1988; Pearson and Goheen, 1988). V. vinifura has become very diverse with time and about 5,000 cultivars exist today. Of these, only a small proportion are used commercially. Cabernet Sauvignon, Chardonnay, Chenin Blanc, Grenache, Merlot,
Riesling, Shiraz and Sultana are some of the grape varieties in commercial use in
Australia.
Cultivars vary in their susceptibility to the disease (Mauro et a1.,1988; Carter, I99l;
Tey-Rulh et al., 1991). However, Chapius et al. (1,998) reported that there was no difference in susceptibility to infection Ay ø. torobetween cultivars, although under field conditions they differ in symptom expression. Further investigations are necessary to establish whether these differences in susceptibility are due to physiological or other factors. Also, the role of plant cell defence reactions in preventing or limiting invasion by E. Iata in the tolerant cultivars is not yet known. Furthermore, information is lacking
on the effect that the toxins produced by the pathogen have on the various cultivars of
grapevine. l6
2.3.1. Yield loss
E. lata causes severe yielcl losses in grapevines around the globe. Yield loss estimates varied from 62 to 947o in severely affected vines on North America, while in moderately affected vines losses ranged from 19 to 5O7o (Johnsen and Lunden, 1985; Munkvold and
Marois, 1994; kelan et al., 1999). According to Irelan et al. (1999)' a poll in the
California grape growing regions revealed that, of all the vineyard disease management practices during the period 1996-99, eutypa disease control was the most expensive, with the disease prevention costs being 72%o greater than that for powdery mildew, the next closest ranking disease. French research indicates that within the 44,000 cognac- producing stocks examined in the Charentes vineyard, 207o of the mature gtapevines were affected by the dieback disease in 1988. In the Bordeaux area, loss of 80 - 9O Vohas been reported in some vineyards (Dubos, 1987a)'
In Australia, yield losses have been estimated to range from 0.6 kg to 9 kg per vine depending on the severity of the disease and losses may be in excess of $ 2,800 per hectare, as reported for Shiraz in the Eden Valley (Wicks and Davies , 1999, see Chapter
1).
2. .Management of eutypa dieback
2.4.L. Cultural practices
Research indicates that special attention has to be paid to the timing of pruning.
Pruning vines in late winter when ascospore production is low reduces the risk of
infection (Ramos et a1.,1975b). Petzoldt et aI. (1981) found that vines pruned towards t7 the end of the dormant season are less susceptible to infection by E. Iata and that susceptibility of the wounds declines more rapidly in early spring compared to early winter. It has also been recommended that, where possible, top working of vines should be done in dry weather. Research is currently taking place in California to evaluate the influence of different canopy training systems in the management of the disease
(Burnham, 1998;Lake et a1.,1998).
Other recommendations for management of eutypa dieback include detection and removal of diseased arms in early spring before healthy shoots obscure diseased shoots, removal and burning of affected plant debris and removal of diseased stumps that are above soil level. Remedial surgery of diseased arms is carried out by making a series of saw cuts until there is no evidence of stained wood. This will have a better chance of success if it is carried out before the disease has progressed extensively through the grapevine (Anon., 1997; Flaherty et al., L992; Emmett and Magarey, 1994).
Reconstructing vines is yet another management practice. There are different ways of reworking existing vines which are not sevetely infected. These methods include layering, cutting the trunk l0 - 20 cm below any sign of infection and then training up a healthy shoot from the base of the trunk and removal of the infected trunk when the new shoots start producing and grafting.
2.4.2. Chemical control
Of the various wound protectants tested, benomyl was the most effective fungicide in protecting apricots from infection by E. lata (Moller and Carter,1969).Moller et al.
(Ig71) subsequently reported benomyl (Benlate@) to be more effective than wound 18
sealants in protecting pruning wounds on apricots from infection by E. lata. Further research confirmed benomyl to be effective in preventing infection by E. Iata of apricot and grape pruning wounds (Moller and Kasimatis, 1980; Pearson, 1982; Gendloff et al.,
1983). Benomyl was the industry standard for prevention of eutypa dieback in grapevines until 2001, when it was withdrawn from the market by the manufacturer (E' I.
Du Pont de Nemours & Co.). Munkvold and Marois (1993b) tested a range of chemicals against eutypa dieback and found flusilazole (Nustar@) to be effective. Significant reduction in infection was reported when grapevine pruning wounds were treated with
NECTEC RrM, â paste containing l7o propiconazole plus 27o imazilll and inoculated with a mycelial slurry of E. lata the day after treatment (kelan et al., 1999)' However, none of the above chemicals are registered for control of eutypa dieback in Australia.
2.4.3. Biological control
2.4.3.1. The needfor biological control
Biological control is defined as "the reduction of inoculum density or disease - producing activities of a pathogen or parasite in its active or dormant state, by one or more organisms, accomplished naturally or through manipulation of the environment, host, or antagonists or by mass introduction of one or more antagonists" (Baker and
Cook, t974).
Interest in biological control as a viable practice in modern agriculture has accelerated
in recent years. There is an increase in public awareness of the potential ecological and
health hazards posed by the use of pesticides (Burge, 1988). The emergence of fungicide 19
resistance in pathogen populations and the non-durable nature of fungicides throughout the susceptible period of the wounds (Munkvold and Marois, 1994) have further encouraged the quest for alternative bio-control methods. Furthermore, there are no chemicals registered currently for the control of eutypa dieback of grapevines and cultural practices are only partially effective in controlling the disease (Munkvold et aI., lgg4). Hence, biological control may be an effective component of an integrated management strategy for this destructive disease.
2.4.3.2. Microbes tested agøinst eutypa diehack
Biological control of eutypa dieback was first achieved by treating pruning wounds in apricot trees with macro-conidia of F. Iateritiunt (Cafier, l97l; Carter andPtice, t974:
Carter and Price, lg75). The mode of action was mainly inhibition by antibiosis, attributed to a non-volatile diffusible metabolite produced in amounts proportional to the age of the colonies (Carter and Price, 1974). Very few authors have reported mycoparasitic activity of F. lateritium on E. lata. However, Vajna (1986) observed F lateritium to parasitise E. lata in dual cultures. According to Irelan et al' (1999), the
Australian strain of F. lateritium which was benomyl resistant did not perform well in the control of E. lata in experiments conducted by another research group in a different geographic location. Irelan et al. (1999) suggested a thorough screening of biocontrol
agcnts for uniformity of activity against a range of E. lata isolates to ensure reliable
disease control.
Ferreira et at. (199I) demonstrated effective control of E. lata, both in the field and in
vitro, using a strain of Bacillus subtilis and suggested antibiosis as the mode of 20 antagonism. Cladosporium herbarum and F. Iateritium inhibited infection by E- lata in the vineyard when applied to pruning wounds (Munkvold and Marois, L993a). A French group has suggested that microbes which naturally colonise grapevine wounds at moderate temperatures, notably a Rhodotorula sp., may reduce infection of E. Iata by competition (Chapius et al., 1998).
Biological control of eutypa dieback may be a cost-effective and environmentally friendly approach to disease management. However, while control of eutypa dieback using other microbes has been achieved at the research level none have been commercially developed and evaluated in commercial situations.
2.4.3.3. Trichoderma spp. in biological control
Trichodermc spp.have been used extensively as biological conffol agents for diseases in many different crops, including cotton, lettuce, onions, peas' grapes, plums and apples, caused by pathogens such as Pythium, Phytophthora, Rhizoctonia, Fusarium,
Sclerotinia and Botrytis spp. (Sivan and Chet, 1989; Nelson, 1991). Z. harzianum, T' viride and Z. hamatum have been used in the biological control of the various plant pathogens mentioned above. A preparation of ?. hamatum comprising a wheat-bran/peat mixture controlled disease caused by Pythium aphanidermatum in pea, cucumber and tomato, and by R. solani and Sclerotium rolfsii in beans in the glasshouse (Sivan and
Chet, 1982). Sivan et al. (1984) also reported control of P. aphanidermatum using an
isolate of T. harzianum. Control of fruit rot of tomato caused by R' solani has been
demonstrated using T. harzianum in field conditions (Strashnov et a1.,1985). Conidia of
T. harzianurz prevented co-inoculated conidi a of Botrytis cinerea from infecting newly 2l opened strawberry flowers in laboratory, glasshouse and field conditions (Hjeljord et al.,
2001). T. harzianum has been successfully used against B. cinerea, a serious pathogen on grapevines, under conditions of low disease incidence (Elad, 1994;Hatman et al., 1996).
Dry rot of apples caused by natural infection of B. cinerea was controlled by T. harzianum (Tronsmo and Ystass, 1980). T. harzianum, T. longibrachiatum and T. stromaticum were reported to reduce witches' broom disease while T. virens was shown to reduce black pod disease in cocoa (Krauss and Soberanis,2OO2).
Trichoderma spp. have been used successfully in curative and prophylactic treatments for various diseases in tree crops. Silver leafdisease ofpeach caused by Chondrostereum purpureum was cured when 7. viride was introduced into the trees in a paste of glycerol and barley flour, or as a spore suspension (Dubos and Ricard, 7974). Trichoderma spp. introduced to freshly-cut stumps of Eucalyptus diversicolor prevented infection by
Armillaria luteobubalirza (Nelson et al., 1995). Infection of wounds on red maple by
Fomes connatus was reduced when the fresh wounds were painted with chopped T. harzianum hyphae in glycerol (Smith et al.,l98I).
Commercial preparation s of Trichoderma spp. are now available. T. virens is available as GlioGardrM for the control of seedling diseases of ornamental and bedding plants, while T. hazianur¿ is sold as F- stoprM to control several soil borne plant pathogenic fungi. BINAB TrM is a corìmeïcial product containing T. harzianum and T. polysporum, available to control wood decay (samuel, tq96; Agios, 1997). A New
Zealand-based company, Agrimm Technologies Ltd, has formulated a lange of
biological products containing T. harzianunr. Trichoseal@ and Trichospray@ are pruning
wound applications, which are designed to control wood decay fungi. Other products 22
such as Trichoject@ and Trichodowels@ are Trichoderma formulations registered with the
New Zealand Pesticides Board for use in the control of silver leaf (Chondrostereum purpureum) on pip and stone fruit trees and Armillaria root rots of kiwi fruit. The manufacturer suggests that these products could also be used to control eutypa dieback of grapevines (Hunt, l9g9), and a New Zealand registration has recently been approved (J.
Hunt, pers. com.).
2.4.3.4. Mechanisms of antagonism by Trichoderma spp.
The three major mechanisms proposed to explain antagonistic interactions between
Trichodermd spp. and other fungi are antibiosis, mycoparasitism and competition
(Hjeljord and Tronsmo, 1998).
The various metabolites produced by Trichoderma spp. were found to inhibit many plant pathogens. Dennis and Webster (l97la; 1971b) found that many isolates of
Trichoderma produced volatile and non-volatile antibiotics. The metabolites produced by
Trichodermq spp. are classified as alkyl pyrones, isonitriles, polyketides, peptaibols, diketopiperazines, sequiterpenes and steroids (Ghisalberti and Sivasithamparam, l99l:
Howell, 1998). The coconut-scented compound 6-pentyl-pyrone, which belongs to the group of alkyl pyrones, was first identified in isolates of T. viride (Collins and Halim,
1972). Bisby (1939) previously noted this peculiar odour from cultures of T. viride' 6-
pentyl-pyrone has now been isolated also from T. harzianum, T. koningii and T.
harrnatum (Simon et al., 1988; Ghisalberti and Sivasithamparam, 1991)' This volatile
metabolite shows antibiotic activity against several plant pathogens (Claydon et aI.,
1987; Ghisalberti et al.,1990). Various isonitriles have been isolated from ?' harzianum, 23
T. hamatum, T. koningii, T. polysporum and T. viride (Fujiwara et aI., 1982). Isonitrin A is effective against both bacteria and fungi while isonitrin D showed activity against fungi only. Harzianolide, which belongs to the polyketide group and was isolated from T. harzianum, inhibited growth of the take-all fungus (Gaeumannomyces graminis) and suppressed take-all disease of wheat in the glasshouse (Almassi et al. l99L)' T. koningii has also been reported to produceharzianolide (Dunlop et a1.,1989).
Non-volatile, chloroform-soluble antibiotics such as trichodermin and dermadine, and peptide antibiotics, such as alamethicine and suzukacilline, which were initially extracted from cultures of Z. viride, were found to be active against a rùnge of fungi and bacteria
(Dennis and Webster l97la). The sesquiterpene metabolite, heptelidic acid, which was isolated from Z. viride, T. virens and T. koningii, showed antibiotic activity against P. ultimum and R. solani (Howell et al., 1993). The steroid, viridin, first isolated from T. viride (Brian and McGowan, 1945) was reported to inhibit germination of spores of many fungi (Brian and Hemming, 1945). Viridin was also reported to inhibit growth of
R. solani and P. ultimum and germination of sclerotia of Sclerotiurn rolfsii (Lumsden e/ al., 1992b). Gliotoxin and gliovirin produced by T. virens have been implicated in the biocontrol of soil borne fungi such as R. solani, Phytophthora spp. and Pythium ultimum
(Weindling,794l; 'Wright, 1952; Howell and Stipanovic, 1983; Lumsden ¿/ al., I992a;
Wilcox et al., 1992; Howell et at., 1993). It is clear from these reports that the metabolites produced by T. harzianum have an antagonistic effect on many plant
pathogens. However, reports on the effects of the metabolites produced by T. harzianum
on E. lata are scarce. Hence, it would be useful to determine if the T. harzianun strains
used in this study produce antibiotics that may inhibit E. lata. 24
Many authors have cited mycoparasitism as a mechanism of antagonism by
Trichoderma spp. (Boosalis, 1964; Harman et a1.,1980; Chet and Baker, l98l;Elad et aI., 1983c), Mycoparasitic activity has been commonly observed in T' harzianum, T' hamatum and. T. koningii. Some destructive soft rot and wilt pathogens of vegetables and field crops, such as S. rolfsii, R. solani and S. sclerotiorum, are parasitised by
Trichoderma spp. Trichoderma spp. were observed to parasitise Phytophthora cinnamomi (Elad, et al., I983c; Trutman and Keane, 1990; Chambers and Scott' 1995).
Mycoparasitism of R. solaniby Trichoderma spp. was observed by Elad et al' (1983c).
Mycoparasitism generally involves four successive stages. (1) Chemotropic growth of antagonist towards the host fungus (Chet et a1.,1981). (2) Recognition between host and the antagonist. Lectins, which are sugar-binding proteins, have been shown to be involved in the host-mycoparasite relationship between T. harzianum and Rhizoctonia solani (Elad et al., L983a). Specific lectin activity has been reported between Sclerotiurn rotfsii and conidia of T. hamatumT-244 (Barak et a1.,1985). (3) Attachment and coiling of hyphae of the antagonist around the host. Trichoderma hyphae usually attach to the host by forming appressoria or hook-like structures (Harman et al', 1981; Elad et al.,
1983a; lnbar et al., 1996). (4) Penetration of host hyphae by the mycoparasite.
Trichoderma penetrates by secreting lytic enzymes, which degrade the cell walls of the
host fungus (Hadar et al.,1979a;Hadar et aI.,1919b; Chet and Baker, 1981;Elad et al',
1983c; Baker, 1987). Chitinase, glucanase and protease are the main enzymes involved
in degradation of fungal cell walls (Elad et al., 1982; Inbar and Chet, 1995).
Mycoparsitism is an important phenomenon in biological control. However, little
information is available on parasitic interactions between T. harzianum and E. Iata and 25 the relative importance of mycoparsitism in the vine. Therefore, it is necessary to investigate this aspect of biological control.
Competition is an important aspect of biological control. It occurs when the pathogen and the antagonist demand more of the same resource than is immediately available'
Disease control may be achieved if the growth of the antagonist results in reduction of the pathogen population or inoculum production (Hjeljord and Tronsmo, 1998).
Trichodermd spp. are considered to be aggressive competitors because of their rapid rate of growth, prolific conidiation and the ability to utilise a range of substrates. Biological control by competition by Trichodermd spp. to control fungi such as Botrytis cinerea and
Sclerotinia spp. in apple, strawberry and cucumber has been demonstrated (Tronsmo and
Dennis, 1971; Tronsmo and Raa, 1917; Elad et al., 1993). These pathogens opportunistically invade dead or senescing plant tissues as a nutrient base before colonising healthy tissues. Trichoderma spp. sprayed on grape flowers during blossom delay colonisation by B. cinerea and, thereby, reduce disease levels in the berries
(Dubos, 1987b; Harman et al., 1996). Seed treatment of cereal and vegetable crops with
Trichodermd spp. reduced germination of sporangia of P. ultimum by competing for the root exudates that stimulate germination of the pathogen (Ahmad and Baker, 1988;
Harman and Nelson, lgg4). Early colonisation of fresh wound sites by T. viride applied
in sprays or via pruning shears has been used to control C. purpureum (Grosclaude et al.,
1973; Corke, 1974). Application of T. harzianum to grapevine pruning wounds may,
therefore, prove useful in preventing infection by E' lata' 26
2.5. Other trunk diseases
Other major trunk diseases of gtapevine include esca and Petri disease (formerly called black goo decline). The occurrence of these two diseases was confirmed in
Australia only recently (Pascoe, 1999). Esca is a complex disease of grapevine, caused by wood rotting basidiomycetes in association with Phaeomoniella chlan'rydospora
(formerly known as Phaeaoacremonium chlamydosporum). Symptoms of esca develop due to interactions between several factors (Mugnai et al., 1999).It has been suggested that esca is caused by a succession of fungi in diseased vines following initial colonisation by E. lata and Cephalosporium sp. (Chiarappa, 2000). Three species of fungi, Fomitiporia punctata, P. chlamydospora and P' aleophilum are thought to be associated with esca of grapevines, the latter two being suspected of acting as precursors for wood decay (Larignon and Dubos, lg97). However, Sparapano et al' (2000) demonstrated that F. punctata alone colonised the wood and caused esca in grapevines.
Esca symptoms include "apoplexy", "tiger stripe" leaf symptoms and black blotches on fruit widely known as "black measles" symptoms. Affected cordons show a soft, creamy, white heart rot. P. chlamydosporum is always found surrounding the heart rot (Pascoe,
1999). Although esca is currently not a major disease in Australia, the incidence and
severity of esca may increase in the next few decades since P. chlamydosporumhasbeen
commonly observed in newly planted vines (Pascoe and Cottral, 2000).
Petri disease affects both young and mature vines, but decline is rapid in young vines
that may show stunting and produce chlorotic leaves and thin stems. Wood symptoms
involve black streaks in the heartwood and a glistening black exudate from cut vessels
(pascoe, lggg). Petri disease may occur in newly planted vineyards and is caused by 27
Phaeomoniella chlamyd.ospora (Morton, 2000; Pascoe and Cottral, 2000)' These diseases need to be taken into account in studies ofthe efficacy ofbiological control.
2.6. Summary
Eutypa dieback, which is a global problem, affects older vineyards and the grape industry worldwide faces the risk of losing vineyards to this disease. The pathogen enters the vines through large pruning wounds. Symptoms may not be detected immediately as the pathogen proliferates very slowly in the host tissues. The foliar symptoms are thought to be due to toxins produced by E. Iata in the wood and translocated to the foliage. Diseased vines show reduced vigour and declining productivity, and eventually die. Disease development and dissemination are influenced by environmental factors such as rainfall, temperature and relative humidity'
The conventional methods discussed in this review are inadequate to manage this destructive disease. Hence, biological control may be a useful component of an effective disease control strategy. Very little is known about interactions between Trichoderma spp. and Eutypa lata in grapevine wood. It is imperative that the interactions between the pathogen and the antagonists are thoroughly investigated and well understood in order to formulate an integrated pest management strategy, comprising conventional and biological control methods, to ensure the long-term sustainability of grape and wine
production. 28
CHAPTER 3. GENERAL MATERIALS AND METHODS
The materials and methods described in this chapter are those which have been used often in the experiments which are described in detail in the relevant chapters.
3.1. The host
Various cultivars of grapevine were used in these investigations. Shiraz was used in all laboratory experiments while both Shiraz and Chardonnay were used in the glasshouse experiments, except in the preliminary investigations. The field trials involved older cultivars such as Exotic, Ribier, Nyora, Rondella and Palomino, as well as the contemporary cultivars Shiraz and Cabernet Sauvignon. Exotic, Ribier and Nyora were grown in the Alverstoke vineyard, Waite Campus; these vines were considered expendable by the vineyard manager. Rondella and Palomino were grown in an abandoned vineyard in Warriparinga, close to the Flinders University of South Australia, while Shiraz was grown in a corrunercial vineyard at Eden Valley and Cabernet
Sauvignon \üas grown in research vineyards at Nuriootpa, South Australia. The cultivars were assigned to the various experiments, depending on their availability, as given in
Table 3.1.
In the glasshouse trials, l-year-old cuttings of cultivars Chardonnay and Shiraz
obtained from the Riverland Vine Improvement Committee, Barmera, South Australia
were used. The cuttings were pretreated with hot water at 50 + 1"C for 30 min. These
two-node cuttings were allowed to stand in tap water overnight after the third basal bud
was removed, before they were subjected to the various treatments, described in Chapter 29
5. These cuttings were inserted into rockwool pieces (4 x 4 cm Grodan blocks, supplied by Home Hydro, Glen Osmond, South Australia) which were saturated with water, and arranged on beds of vermiculite in trays (29 x 34 cm). The cuttings were maintained in the glasshouse in natural light at 5-35"C and were watered using tap water every other day. The cuttings in rockwool pieces were transferred to pots (20 cm diam.) filled with
University of California (UC) potting mix (Baker, 7957) after 6 to 8 weeks, in the same layout as in the trays. These transfers were necessary to accommodate vigorous vine growth in the glasshouse. In glasshouse experiments 4,5,6 and 7,the cuttings inserted into the rockwool pieces were placed directly into UC potting mix in 2O cm diam. pots.
The water-saturated rockwool pieces provided sufficient moisture for root intiation. Tap water was used to water the plants, every 2 days, and the cuttings were treated with
ThriverM nutrient solution (Arthur Yates and Co. Ltd; Homebush, NSW, Australia) every
6 weeks. The cuttings were maintained in the glasshouse as above. The l-year-old cuttings used in all the experiments were IO-20 mm in diameter.
3.2. The pathogen
3.2.l.Isolates of E. hta
The isolates of E. Iatawhích were used in experiments are listed in Table 3.2'The
cultures were incubated at 22 to 25"C in darkness for 5 to 7 days and mycelial plugs
were obtained from the margins of the colonies (Carter, 1991)' 30
3.2.2, Extraction of ascospores
Pieces of infected dead wood (5-10 cm long) with stromata of E.lata were washed in tap water and then soaked in Reverse Osmotic (RO) water for t h (Carter, 1991)' They were transferred to 70 mm wide and 145 mm deep sterile polycarbonate tubs with polypropylene lids (MagentarM, GA7) and allowed to discharge ascospores overnight'
The ascospores discharged from the perithecia were collected in 1 ml sterile distilled water (SDW) and serially diluted to the desired concentration. Spores were counted using a light microscope (I-nitz wetzlar, Orthoplan 871288) with the aid of a haemocytometer (Improved Neubauer B. S. 748).
3.2.3. Viability testing of ascospores
Viability was tested when ascospores were used as inoculum in the experiments. The spore suspensions prepared as described in section 3.2.2 were diluted to 103 spores/ml and 0.1 ml of this suspension was pipetted on to potato dextrose agar (PDA, Difco) or
2To water agar (WA, Bitek Agar, Difco) plates. These plates were incubated at 23"C in the dark for 2-4 days and germinating spores were counted by examining the plates in natural light. The percentage of germination was found tobe IOOVo in all instances. 31
TABLE 3.L. List of grapevine cultivars used in this project'
Type of Experiment Cultivar Age
experiment
Laboratory Test 1 Shiraz 1-yr-old canes
Laboratory Test 2 Shiraz 1-yr-old canes
Laboratory Test 3 Shiraz 1-yr-old canes
Glasshouse Experiment 1 Shiraz 1-yr-old canes
Glasshouse Experiment 2 Chardonnay, Shiraz 1-yr-old canes
Glasshouse Experiment 3 Chardonnay 1-yr-old canes
Glasshouse Experiment 4 Shiraz 1-yr-old canes
Glasshouse Experiment 5 Shiraz 1-yr-old canes
Glasshouse Experiment 6 Shiraz 1-yr-old canes
Glasshouse Experiment 7 Chardonnay I yr-old canes
Field Nuriootpa lu Cabernet Sauvignon 23 yr-old vines
Field Nuriootpa 2u Cabernet Sauvignon 16-yr-old vines
Field Eden Valleyu Shiraz 27-yr-old vines
Field Warriparinga 1D Rondella > 60-yr-old vines
Field Warriparinga 2o Palomino > 60-yr-old vines
Field Trichodowel trialD Nyora 22-yr-old vines
Field Injection trialD Ribier, Exotic 22-yr-old vines
n Pruned routinely using the "Finger and thumb" method (T' Gherlach, pers. com., 2000). o Spu, prunrd routinely. 32
TABLE 3.2. List of isolates of E. lata'
Strain Source Origin
M2g0u'o Grapevine wood Victoria
M302u Grapevine wood Victoria
}'{zg5u Grapevine wood South Australia
CS-Ba.1.99112/06 Grapevine wood South Australia
(Cabernet sauvignon)
CS-8a.2.99/12106 Grapevine wood South Australia
(Cabernet sauvignon)
CS-8a.3.99112106 Ascospores (Cabernet South Australia
sauvignon)
u Strain from F. M. Cote, Monash University, Victoria.
b pers' Designated virulent on the basis of pathogenicity to micropropagated grapevines in vitro (M. Cole, com.).
3.3. The antagonists
3.3.1. Strains
Three strains of T. harzianum provided by Agrimm Technologies Ltd were used in
the initial experiments to ascertain the mechanisms of antagonism. These are designated
strains I,2 and3 in this thesis, but have since been re-named AGl, AG2 and AG3 by the
manufacturer. Strain 1 was used in all subsequent experiments on the recommendation of
Agrimm Technologies Ltd. Also, F. lateritium was used in some experiments (Carter,
1983). The isolates ,were cultured on PDA or 116 strength Czapek Dox agar (CDA' JJ
Difco) at22-23"C in the dark. Mycelial plugs were taken from the margins of 3-4-day- old cultures.
Spore suspensions were prepared using 5-7-day-old cultures. The culture plates were flooded with SDW, gently scrubbed with a sterile wire-loop, transferred to a sterile polycarbonate container and the concentration of the spore suspension was ascertained with the aid of a haemocytometer (Improved Neubauer B. S. 748) and the light microscope (I-eitz wetzlar, Orthoplan 871288). The spore suspension was diluted with
SDW to obtain a concentration of 10e spores/ml in the case of T. harzianum or IO6 spores/ml for F. lateritiun't.
3.3.2. Commercial formulations
The commercial formulations supplied by Agrimm Technologies Ltd contained seven strains of T. harzianum, including the three strains mentioned in section 3.3.1. The different types of formulations used in the glasshouse and field investigations and the experiments in which they were used are listed in Table 3.3.
3.3.3. Enumeration of colony forming units (CFU)
Malt agar (MA) was prepared by dissolving 50 g of malt extract agar (Oxoid) in 1 L
of RO water and autoclaving at lzl"C for 20 min. The preparation of sterile diluent,
used to suspend the commercial Trichoseal@ or Vinevax@ formulation and to prepare the
dilution series of the suspended product, is described below. The method of enumeration,
based on that provided by Agrimm Technologies Ltd, is also given below. 34
A. Preparation of sterile diluent:
Bitek agar (Difco, 0.2 Ð was dissolvedin2 L of RO water and autoclaved at lzl"C for 20 min. Tween 20 (Sigma, Polyoxyethylene Sorbitan Monolaurate) and concentrated
HCl, 50 ¡rl of each, were added to each litre of the autoclaved agar solution. A series of dilutions was prepared aseptically as described in the Appendix 1.
B. Plating:
applied to each A 250 ¡,ll droplet of each dilution of each commercial formulation was of three plates of MA and spread out by gently rolling the plates from side to side. The plates were sealed with Parafilm@ and incubated at23"C for 72 h. Individual colonies on the plates were counted by examining them under natural light. The cfu/g was calculated as shown below.
C. Conversion into CFU/g:
1. The mean numbers of colonies on the replicate plates were calculated for each
dilution.
2. CFU/ml for each dilution = mean of each dilution which was multiplied by 4.
3. The values obtained at step 2 were multiplied by the corresponding dilution factor for
each dilution.
4. The values obtained at step 3 were multiplied by 5 X 103 to obtain the number of
CFU/g of product (J. Hunt, pers. com., 1999).
The CFU assay was carried out twice to confirm that Trichoseal spray@ and Vinevax@
yielded CFU of T. harzianurn, as stated by the manufacturer. The CFU/g values obtained
for Trichoseal spray@ and Vinevax@ were I x 1010 Czu/g and 32 x 108 CFU/g, 35 respectively. The figure stated on the label for Trichoseal spray@ was "at least 11 X 106
CFU/g" and for Vinevax@ "not less than of 5 x 108 cfu/g".
Table 3.3. List of commercial formulations
Formulations Strains Experiments involved
(D Trichoseal 7 strains of T. harzianum Glasshouse trials 4, 5 and 6
Trichoseal spray'or 7 strains of T. harzianum Pruning wound trials in the
@ * v lnevax field
Trichodowels@ 7 strains of T. harzianum Trichodowel trial
(9 Trichoject 7 strains of T. harzianum Injection trial
* was marketed by Agrimm Technologies Ltd in 2002 to be used in place of and
Trichoseal sprayt. This formulation was used only in the Warriparinga 2 Trial.
3.4. Re-isolation of fungi from wood
3.4.1. Re-isolation of pathogen
The bark was stripped from the canes and the I-2 cm cane segments were surface
sterilised in 2$7o sodium hypochloride (NaOCl) containing a drop of Tween 80 (Sigma,
Polyoxyethylene Sorbitan Monooleate) on a rotary shaker for 12 min. and washed twice
with SDW. The cane pieces were then split longitudinally and cut into 5-10 mm chips'
The chips were transferred to a medium selective for E.lala (EUSM), the composition of
which is described in the Appendix 2.The plates were incubated in the dark at23"C fot
7-14 days.If any one of the chips yielded E. lata, the cane was considered to be infected. 36
E. lata mycelium was identified by comparing the colour and colony morphology with th at of a known culture of similar age (Petzol dt et al., 1 98 I ; Carter , l99l; Munkvold and
Marois, Igg3). When cultures were further incubated for 4-8 weeks, some isolates produced pale yellow or orange droplets, which contained the characteristic conidiomata with curved conidia (Carter, 1991). However, not all isolates developed conidiomata.
3,4.2. Re-isolation of antagonists
The canes were surface sterilised and cultured as described in section3.4.l. However, the wood chips were either plated on acidified PDA (APDA) or CDA plates (see
Appendix 2) and,incubated at23"C in the dark for 3-7 days to re-isolate T. harzianum.In the case of wood shavings collected by drilling into the trunks of vines, as in the case of the Trichodowel trial, the shavings were only washed thrice in SDW and plated on
APDA in clumps. Re-isolation of Z. harzianum was also carried out on EUSM in the simultaneous testing of E. lata and T. harzianum, when separate re-isolations were not possible due to limited availability of wood tissue. Re-isolation of F. Iateritium was carried out as for re-isolation of T. harzianum lrom wood chips, and the chips were plated on CDA medium (see Appendix 2) and incubated at 23"C in darkness fot 3-1
days. T. harzianum and F. lateritium could be identified by their colony morphology on
culture plates. 37
3.5. Maintenance of isolates and cultures
All fungal isolates were stored at 5oC as culture plates. T. harzianum and E. IatL were stored on PDA while F. lateritium was stored on CDA. Long-term storage of E' Iata isolates was in SDW as mycelial plugs in McCartney bottles at room temperature
(Boesewinkle, 197 6; Carter, 199 1).
3.6. Statistical analysis
All analyses were perfornìed in GENSTAT for windows, 5th edition (Lawes
Agricultural Trust, Rothamsted, England). Analysis of variance (ANOVA) was carried out to determine the significant differences between the treatments. LSD values at the
57o probability level were used to separate the means. For data comprising the presence or absence response, ANOVA techniques were not appropriate since the data were not normally distributed. Hence, such data were modeled using one form of generalised linear model (GLM), assuming a binomial distribution and a logit link as advised by Ms
Lorimer, BiometricsSA. This method is also known as logistic regression (McCullagh
and Nedler, 1989). 38
CHAPTER 4. INTERACTION STTJDIES IN VITRO
4.L. Introduction
The high degree of ecological adaptability shown by Trichodermq spp. and their amenability to cultivation on inexpensive substrates make these fungi attractive candidates for applications in biological control. To become successful biological control
adaptability but must 4gents, the isolates of Trichoderma not only must show ecological
also have suitable antagonistic characteristics against the specific pathogen. According to
one "school of thought", isolates of potential bio-protectants need to be screened in the
laboratory before applications are tested in the glasshouse or field. However, some
researchers prefer to screen potential biological control agents first in pot or field trials
and, if the antagonists perform efficiently in these conditions, then mechanisms of
inhibition and relevant in vitro studies are carried out in the laboratory (Blakeman, 1988;
Faull, 1988). Also, the mechanism(s) of biological control has been used to selectively
isolate potential biological control agents (Cook et al., 1,997).In this project, laboratory
experiments were first carried out to investigate mechanism/s of inhibition of E. lataby
T. harzianum.
Antagonistic interactions between Trichoderma spp. and other fungi have been
classsified into three main categories: antibiosis, mycoparasitism and competition (see
section 2.4.3.3). These mechanisms are not mutually exclusive and antagonistic activity
of a biocontrol agent could fall into one or more of these categories. Biological control of
B. cinerea on grapes was achieved by T. harzianum through mycoparasitism and 39 competition for nutrients (Dubos, 1987b). Antibiotics and hydrolytic enzymes that are proclnced together by Trichoderma spp. have been shown to act synergistically in mycoparasitism (Di Pietro et al., 1993; Schirmböck et al., 1994). Moreover, some antagonistic interactions may not fall into any of the three classical categories. For example, T. harzianum indftectly controlled the decay fungus, Fomes connatlts, in red maple by replacing the pioneer fungus Phialophora melinii which renders the wood susceptible to the decay fungus by reducing the phenolic constituents of maple sapwood
(Smith et al., 1981). More recently, Zimmand et al. (1996) have suggested that Z.
harzianum T39 antagonises B. cinerea by reducing the amount of pectin-degrading
enzymes produced by the pathogen.
The antimicrobial compounds produced by Trichodermq spp. constitute a diverse
group of secondary metabolites with respect to structure and function. These metabolites rüebster,l97la; comprise both volatile and non-volatile compounds (Dennis and Dennis
and Webster,l97Ib), as described in section 2'4.3'4.
Mycoparasitism, the direct attack of one fungus on another, usually results in death of
the host fungus if Trichoderma spp. are used as the antagonist (Barnett and Binder,
lg73). Biological control due to mycoparasitism by Trichoderma spp. is reviewed in
section 2.4.3.4.
Competition occurs when two or more microorganisms demand more of the same
resource than is immediately available (see section 2.4.3.3). The three major mechanisms
of antagonism are reviewed in section 2.4.3'3.
The investigations reported in this chapter were directed towards studying interactions
between E. lata and T. harzianum in the laboratory on agar medium and in sterilised 40 cane segments to ascertain the mechanism(s) of inhibition. Scanning electron microscopy was used to examine interactions in gamma-irradiated cane segments and living canes.
4.2. Materials and Methods
4.2.1. Mode of inhibition
4.2.1.1. Antibiosis by volatile metabolites
The method of Dennis and Webster (1971b) was used to investigate the effect of volatile antibiotics produced by strains l, 2 and 3 of T. harzianum on E. lata isolates
M280 and CS-Ba.1. Plugs of T. harzianun't (8 mm diam.) were placed in the centre of 90 mm Petri dishes containing 20 ml PDA and incubated at22-25"C in darkness. After 2 days, the bases of fresh plates of PDA inoculated with 8 mm diam. plugs of E. lata in the central position were inverted over the bases of the plates with the 2-day-old antagonists, taped with Parafilm@ and incubated in the same conditions for a further 5 days. Controls consisted of plates of E. lata inverted over PDA plates inoculated with sterile PDA plugs. There were eight replicates per treatment combination. For each test plate, the colony diameter of E. lata was measured (the diam. of the plug of inoculum, 8 mm, was
subtracted), and the results were analysed using ANOVA (see section 3.6).
The effect of volatile metabolites on the germination of ascospores of E. Iata was also
tested. Ascospores were extïacted from wood containing stromata as described in section
the 3.3.2 and20 ¡t"l droplets, each containing 500 ascospores in SDW, were placed on
surface of each PDA plate. The droplets were spread evenly over the surface of the PDA 4l and the bases were inverted over the bases of plates that had been inoculated 2 days previously with 8-mm-diam. plugs of the three strains of antagonists, applied individually, and sealed with Parafilm@. Controls consisted of PDA plates containing the same number of ascopores inverted over PDA plates inoculated 2 days previously with
8-mm-diam. plugs of sterile PDA and sealed with Parafilm@. After 2 days of incubation at 22-25" C in the dark, the number of germinating spores that had formed colonies was counted using a dissecting microscope lWild@; magnification X 50). There were eight replicates per treatment combination. The ascospores were too dense to count accurately.
This problem was rectified in the next experiment (section 4.2.L2) by further diluting the
ascospore suspension so that a 10 ¡ll droplet contained only 25 ascospores. However, this
experiment was not repeated with further dilutions, since there was a total inhibition of
germination by the volatile metabolites produced by all three strains of the antagonist.
4. 2. 1. 2. Antib io s i s by no n- v olatile m e tab o lit e s
This experiment was based on the modified method of Dennis and Webster (I97Ia)
used by Chambers (1993). The effect of non-volatile metabolites produced by strains 1, 2
and 3 of T. harzianum on six isolates of E. lata was investigated (M280, M295,M302,
CS-Ba. 1, CS-8a.2, CS-Ba.3).
Sterile uncoated cellophane discs (80 mm diam.; Australia Cellophane, Victoria) were
placed in each 90 mm diam. Petri dish, containing 2O ml PDA. The cellophane discs
were first immersed in boiling water for 10 min. and then autoclaved at lzl"C for 20
min. Plugs of mycelia (8 mm diam.) of the antagonists were each cut from the edge of an
actively growing culture on PDA and placed in the centre of each cellophane disc. Sterile 42 plugs of PDA of the same dimensions served as controls. The plugs were incubated at
22-25"C in the dark. After 2 days the plug of mycelium and agar was removed together with the cellophane, replaced by a plug of E. lata (8 mm diam.) and the plate incubated in the same conditions. There were eight to ten replicates per treatment combination. The colony diameters of E. lata isolates were measured after 4 days, and the diam. of the plug of inoculum (i.e. 8 mm) subtracted before analysis of the data. In a subsequent repeat experiment, the colony diameter of the slower growing isolates M295, M.302, CS-
Ba.2 and CS-8a.3 was measured after 6 days. Plugs which showed no growth of
mycelium after 2 weeks were transferred to fresh PDA plates, incubated in the same
conditions for a further 2 weeks and observed for signs of mycelial growth to determine
whether effects were fungistatic or fungicidal. The results were subjected to ANOVA
(see section 3.6).
In addition, the effect of non-volatile metabolites produced by the three strains of Z'
harzianum on germination of ascospores of E. lata was investigated. The antagonist was
grown from a plug of inoculum on the surface of a cellophane disc exactly as described
above. After 2 days of incubation the mycelial plugs of the antagonist or PDA plugs and
the cellophane were removed and the plates were inoculated with a LO ¡tl droplet
containing 25 ascospores as described in section 4.2.1.L and incubated in the dark at22'
25"C. After 3 to 6 days the number of germinated spores was counted as described in
section 4.2.1J. There were eight replicates per treatment combination. ANOVA was
used to analyse the results (see section 3.6). 43
4.2.1.3. Parasitism
Parasitic interactions were investigated in dual cultures (Dennis and'Webster, L97lc;
Chambers and Scott, 1995). A sterile strip of uncoated cellophane (80 x 30 mm;
Australia Cellophane, Victoria), prepared as described in section 4.2.1.1, was placed in the centre of each 90 mm diam. Petri-dish containing 20 ml PDA. Plugs (8 mm diam.) of
E. Iata isolate M280 or CS-8a.1 were placed on one end of the strip,20 mm from the edge of each plate. The controls received sterile PDA plugs. The plates were incubated at
22"C for 3 days in darkness. A plug of T. harzianum (8 mm diam.) was transferred to the cellophane strip 50 mm from the E. lata p\tg and incubated in the dark for 3 to 4 days.
The three strains of ?. harzianum that were tested for volatile and non-volatile metabolites were also tested for parasitism. There were five replicates per treatment combination. After 3 to 4 days, 5 x 5 mm pieces of cellophane were cut from the zone of interaction, mounted on slides, stained with ammonical congo red (Chambers and Scott,
1995) and observed using the compound microscope (I-nitz Wetzlar Orthoplan; magnification x 400).
4.2.2. Interactions on cane segments
Inter-nodal grapevine cane segments of cultivar Shiraz were sterilised by autoclaving or gamma-irradiation to eliminate competition from resident organisms. 44
4. 2. 2. 1. C olonis ation of w o o d by E, lata in the pr e s e nc e of T . hatzianum -
experiment 7
This experiment was based on the method used by Mercer and Kirk (1984a) to test
biological treatments for the control of decay in tree wounds. Inter-nodal segments (60
mm long) of l-year-old canes of cultivar Shiraz were split into two halves, bark was
removed, and the canes sterilised either by autoclaving at I2I" C for 20 min. or by
gamma irradiation. kradiation was carried out by Steritech, Victoria, Australia at 25
kgray. Bark was removed from the canes before sterilisation. Each of the split segments
was then placed in the centre of a Petri dish (90 mm diam.) lined with three layers of
sterile Whatman No. 42 ashless filter paper discs (80 mm diam). The filter papers were
first autoclaved at 121"C lor 20 min and moistened with SDV/. Plugs (5 mm diam.) of
the virulent E. lata isolate M280 were placed on the cane segment2-3 mm from one end
and the canes were incubated at 23-25"C in the dark. After 5 days, a 5-mm diam. plug of
strain 1 of T. harzianum (or a plug of PDA of the same dimension in controls) was
placed 2-3 mm from the other end of the cane segment. E. lata was applied before Z.
harzianum to allow the slow growing pathogen to become established in the wood. There
were eight replicates per treatment combination. The co-inoculated segments were
incubated for a further 3 days in the same conditions, after which the extent of growth of
E. lata (i.e. the distance between the inoculum plug and the hyphal front) in the presence
and absence of T. h.a.rziarutm was measured. The results were analysed using ANOVA
(see section 3.6). 45
4.2.2.2. Colonisation of wood hy E.lata in the presence of T.harzianum -
experim,ent 2
This experiment was based on the method used by Munkvold and Marois (1993a) to identify possible biological control agents for E. lata. Canes of cultivar Shiraz were cut into 10 mm long segments, the bark was removed and the canes autoclaved at I2loC fot
20 min. then embedded upright in 2 7o water agar in 140 mm diam. Petri dishes. There were 10 segments in each of four dishes. One of the following four treatments was administered to the 10 cane segments in one of the four Petri dishes so that there was one plate of 10 segments per treatment. The plates were incubated at 23oC fot 24 h in darkness.
Treatment 1 = 5 mm sterile PDA plug (control)
Treatment 2 = I0 ¡rl droplet of SDW (control)
Treatment 3 = 5 mm plug of strain 1 of T, harzianum
Treatment 4 = 1,500 spores of strain 1 of ?. harzianum in a 10 ¡,tl droplet of SDW
The treatments were applied to the upper freshly cut surface of the cane segments. After
24hthecanes were inoculated with 5 mm plugs of E.lata isolate M280 (see Table 3.2).
The mycelial plugs of the antagonist and the plugs of PDA were removed aseptically from the canes that received Treatments 1 and 2 before these were inoculated with the pathogen. The segments were incubated for a further 10 days. Re-isolation of E lata was carried out by cutting each segment into 10 chips after surface sterilisation in 2.5 Vo
NaOCI as described in section 3.4.1. Statistical analysis was deemed unnecessary since there was complete inhibition of colonisation of canes by E. lata in the presence of Z'
harzianum. 46
4.2.2.3. Colonisation of wood by E.lata in the presence of T. harzianum -
experiment 3
This experiment was based on the method of Rolshausen and Gubler (1998). Bark was removed from l-year-old canes of cultivar Shiraz and 30 mm long segments were cut from these canes. Each segment was placed into a McCartney bottle containing 3 ml of SDW and autoclaved at l2l"C for 20 min. The cane segments were treated with a 25
(strain ¡rl SDW droplet containing 1,000 ascospores or conidia of E. lata or T. harzianum
1), respectively, in the following way. There were 10 replicates per treatment combination.
Treatment I = E. lata ascospores + T. harzianurn (sttain 1) conidia
Treatment 2 = T. harzianum (strain 1) conidia only (control)
Treatment 3 = E. Iata ascospores only (control)
Treatment 1 involved application of a25 ¡rl droplet of spore suspension of each fungus at the same time. The inoculated segments were incubated at 23"C for 4 weeks, after which surface sterilisation and re-isolation of E. lata and T. harzianum on EUSM were carried out as described in sections 3.4.1. and 3.4.2.
4.2.3. Microscopy
Interactions between E. lata and T. harzianurn were investigated in l-year-old canes
of cultivar Shiraz using the Field Emission Scanning Electron Microscope (SEM)
(Philips Field Emmission Scanning Electron Microscope XL30). Initially, uninoculated
canes were checked for the presence of endophytic fungi by microscopic examination 47
((I-eitz Werzlar Orthoplan; magnification x 400). Cane segments (1-1.5 cm long) that were sterilised by gamma-irradiation at 25 kgray, as described in section 4.2.2.1, were used. Also, interaction in living canes was studied using cuttings grown in rockwool pieces (4 x 4 cm Grodan blocks) in the laboratory. The specimens were processed for
SEM as described in section 4.2.3,L
4.2.3.1. Processing of woodfor SEM
The samples were fixed overnight in SEM fixative comprising4To paraformaldehyde;
1.257o glutaraldehyde in phosphate buffered saline (PBS, see Appendix 2); 47o sucrose; pH 7.2. Next, the samples were washed twice for 30 min. in two changes of washing buffer (PBS + 47o sucrose) and post-fixed in I7o OsO+ in PBS for t h, then dehydrated through an ethanol series of 707o,907o, 95To and lOOTo ethanol in water' The samples were dehydrated for 20 min. thrice at each concentration of ethanol and at the final stage of l¡07o ethanol, after the three changes of 20 min., dehydration was carried out for an additional t h. The dehydrated segments were then immersed in a 1:1 mixture of 1007o ethanol and I007o acetone for 15 min. and transferred to lOOTo acetone for another 15 min. Thereafter, the specimens were dried in the critical point drier (Bal-tec 030)' mounted on metal stubs, coated (3 nm in thickness) with gold and palladium and viewed
under the microscope at an accelerating voltage of 10 kV.
4. 2. 3. 2. I no c uløtio n of g ømm a- irr adiat e d w o o d
Cane segments were inoculated with a 5 mm diam. plug of E. lata strain M280 at one
of the cut ends and incubated at 23"C in darkness in 9 mm diam. Petri-dishes lined with 48 sterile Whatman No. 42 ashless filter paper discs (80 mm diam). The filter papers were first sterilised by autoclaving at L2I"C for 2O min. and then moistened with SDW. After
4 days, the cane pieces were inoculated with plugs (5 mm diam.) of T. harzianum strain
1 at the other end and incubated for a further 3 days as before. The filter paper lining was moistened again with SDW at this stage. Controls were inoculated with M280 or Z. harzianum alone. There were three replicates per treatment combination. After 3 days the inoculated pieces were split open and processed for SEM as described in section 4-2.3.L.
4.2.3.3. Inoculation of canes grown in rockwool pieces in the laboratory
Single-node canes of cultivar Shiraz were allowed to stand in water overnight after the second basal bud was removed. The cuttings were pruned2-4 cm above the bud and inoculated on the freshly cut surfaces with 5 mm diam. plugs of M280 or ascospores of
E. lata (1,000 spores in a 25 ¡rl droplet of SDW). The canes were grown in rockwool pieces in a water-saturated condition on the laboratory bench at approximately 22"C in natural light. After 2 days, the canes that had been inoculated with M280 were either inoculated with 5 mm diam. mycelial plugs of T. harzianum (strain 1) or with a 25 ¡t"l droplet of SDW containing 1,000 conidia of T. harzianum (strain 1) and the canes that had been inoculated with ascospores of E. Iata were inoculated with conidia of strain 1.
The controls were inoculated with mycelial plugs of M280 alone, ascospores of E. lata
alone, mycelial plugs or conidia of T. harzianurn (strain 1) alone. The inoculum was
covered with Parafilm@ at each stage. There were three replicates per treatment
combination. The canes were harvested after 2 weeks and 1 cm long segments below the
point of inoculation were excised split open and processed for SEM (section 4.2.3.4). 49
4,2,3.4. Simultaneous co-inoculation of canes g,rown in rockwool pieces in the
laboratory
This was similar to the experiment described in section 4.2.3.3, except the pruning wounds were simultaneously treated with spore suspensions of E lata and T. harzianum
(strain 1) prepared in SDW, both of which were introduced in a25 ¡tl droplet containing
1,000 spores. The controls consisted of treatments with either ascospores of E.lata alone or conidia of T. harzianum alone. The canes were harvested after 30 min., 6 h, 22 h, 7 days, 14 days and 21 days. There were three replicates per treatment combination. Cane segments (1 cm long) below the point of inoculation were excised, split into two and processed for SEM as described in section 4.2.3.1'
4.3. Results
4.3.1. Mechanisms of inhibitionin vitro
4.3.1.1. Inhibition by voløtile ntetabolites
The volatile antibiotics produced by all three strains of Z. harzianum significantly
(P<0.001) reduced growth of both isolates of E. Iata tested (Figure 4.1). However, there
was significant variation in the extent of the growth reduction of E. lata in the presence
of volatile metabolites produced by the three strains (P<0.001). For example, the volatile
metabolites produced by strain 2 were the most effective in reducing the growth of
isolate M280. The volatile antibiotics produced by strain 3 inhibited isolate CS-8a.1
most strongly, while those produced by strains I and2 had similar effects on this isolate. 50
Also, the volatile metabolites of the three strains of Z. harzianum totally inhibited germination of ascospores of E. lata (Table 4.1). Incubation was carried out for 2 days as germination was assessed on the basis of colonies visible using the dissecting microscope. Incubation for more than2 days was not attempted as two replicates of each of the E. lata cultures treated with strains 1 and 2 and all eight replicates treated with strain 3 were contaminated with Z. harzianum after 2 days. A coconut odour was emitted from culture plates of the three strains of Z. harzianum'
4. 3. 1. 2. I nhib itio n by no n -v ol øtile m e tab olit e s
The three strains of T. harzianum significantly (P<0.001) inhibited growth of all six isolates of E. lata (Figures 4.2 and 4.3). However, total inhibition of growth was not observed for isolates M280, CS-Ba.1 and CS-Ba.2by any of the three strains of Z. harzianum while CS-Ba.3 was completely inhibited only by strain 1. The mycelial growth of isolates M295 and M302 was completely inhibited by the three strains of ?. harzianum (Figure 4.3). No growth was observed even when the plugs of isolate M295 and M302 were transferred to fresh PDA medium and incubated for a further 2 weeks in the dark at 22-25"C. therefore the effect was deemed fungicidal. The non-volatile metabolites produced by strains 1,2 and 3 had a fungistatic effect on the remaining four isolates of E. Iata. The data (Figures 4.2 and 4.3) for the two experiments were analysed
separately.
The effect of the non-volatile antibiotics produced by the three strains of the
antagonist was not significantly different for isolates M280 or CS-Ba.1 (Figure 4.2)'
However, the non-volatile antibiotics produced by strain 1 were significantly more 51 effective (P<0.001) in reducing growth of isolate CS-8a.3 than those produced by the other two strains (Figure 4.3). In the case of isolate CS-Ba.2, all three strains of ?' harzianum were equally effective in reducing growth. These results suggest that the degree of inhibition by the three strains of Z. harzianum may differ for a given isolate of
E.lata.
Germination of ascospores of E. lata was significantly inhibited by the non-volatile metabolites produced by all three strains of T. harzianum (P<0.001). The non-volatile metabolites of strains 2 and 3 totally inhibited germination (Figure 4.4).
4.3.1.3. Inhihition by parasitism
There was no evidence of mycoparasitisn in dual cultures on PDA when the interaction zones of dual cultures were examined using the light microscope (Leitz
Wetzlar Orthoplan). Distinguishing the hyphae of the pathogen from the antagonist was not possible, since both were of the same dimensions and showed similar branching patterns.
4.3.2.Interaction studies on cane segments
4.3.2.1. Interactions hetween pathogen and antagonist - experiment I
In the absence of T. harzianum, E. lata grew an average of 42 mm along the
autoclaved cane segments, significantly further than along the gamma-irradiated cane
segments (38 mm). There was no difference in growth of the pathogen on autoclaved and
gamma-irradiated co-inoculated canes (Figure 4.5). Growth of E. lata was slightly, but 52 significantly, retarded on both autoclaved canes and gamma-irradiated canes when mycelial plugs of T. harzianum weÍe introduced 5 days after E. lata (P=0.023).
4.3.2,2. Interactions between pøthogen and antagonist ' experiment 2
E. Iata colonised all of the control cane segments which had been treated with sterile
PDA plugs or SDW. E. lata was not re-isolated from any of the autoclaved canes which had been inoculated 24 h prev\ously with spores or mycelial plugs of T. harzianurn
(Figure 4.6).
4.3.2.3. Interøctions between pathogen and øntagonist' experiment 3
The pathogen and antagonist were isolated from all of the appropriate control segments (Figure 4.D. f. harzianum was isolated from 907o of co-inoculated canes and
E. Iata from l07o of the canes. This difference was statistically significant (P<0.001) according to the f - t"st. None of the cane segments yielded both fungi when cultured on
EUSM. 53
Figure 4.1. Inhibition of growth of two isolates of E. lata due to volatile antibiosis by three strains of Z. harzianum on PDA. Mycelial plugs of T. harzianum were grown on pDA for 2 days in the dark at 22-25"C. Bases of fresh plates of PDA inoculated with E. lata were inverted over the bases of plates containing the antagonist and incubated for 5 days. Y-axis denotes the average colony diam. (minus the inoculum plug) for eight replicates per treatment combination. Bars denote standard elrors.
45 40 E Ê 35 t! (! 30 u¡ 25 o trM280 20 Ë ICS-Ba.1 .g t, 15 c 10 -9o o 5
0 Control strain 1 strain 2 strain 3 Treatment with 7. harzianum 54
Table 4.L. Inhibition of germination of ascospores of E. lata by volatile metabolites produced by three strains of T. harzianum. Mycelial plugs of T. harzianunx were grown on pDA for 2 days in the dark at 22-25"C. A 2O ¡.tl droplet containing 500 ascospores was spread on a fresh PDA plate and the base of the plate was inverted over the base of the plate containing the 2-day-old antagonist and incubated for another 2 days. There were eight replicates for each treatment combination.
Treatment Average number of germinated
ascospores
PDA plug + ascospores of E. hta >100u
(control)
Strain 1 + ascospores of E.lata 0
Strain 2 + ascospores of E. lata 0
Strain 3 + ascospores of E.lata 0
Estimates only, as germlings were crowded on the plate 55
Figure 4.2. Inhibition of growth of two isolates of E. lata due to non-volatile antibiosis by three strains of T. harzianum on PDA. Mycelial plugs of T. harzianum weÍe grown on cellophane discs on PDA for 2 days in the dark at 22-25"C. The cellophane discs and mycelial plugs were removed and replaced by plugs of E. Iata, and plates incubated for another 4 days. Y-axis denotes the average colony diam. (minus the inoculum plug) of the eight replicates of E. latareceiving each treatment combination. Bars denote standard elTors
40
35 E E 30 G (! 25 t¡¡ 20 o tr M280 Ë 15 .9 ICS-8a.1 tl 10 tr -9 5 oo 0 Control strain 1 strain 2 strain 3 Treatment 56
Figure4.3. Inhibitionof gtowthof fourisolates of E. lataduetonon-volatileantibiosis bythreestrains of T.harzianumonPDA.Mycelialplugsof T.harzianumwere grownon cellophane discs placed on surface of PDA for 2 days in the dark at 22-25"C. The cellophane discs and mycelial plugs were removed and replaced by plugs of E. lata, and plates incubated for another 6 days. Y-axis denotes the average colony diam. (minus the inoculum plug) of the five replicates of E. lata receiving each treatment combination.
Bars denote standard errors.
45
40 T E35 .o i' T M295 r.i ZS TM3O2 o trCS-8a.2 Ë20 .g trCS-8a.3 3rsç o€10 5
0 Control strain 1 strain 2 strain 3 Treatment with L harzianum 57
Figure 4.4. Inhibition of germination of ascospores of E. lata due to non-volatile metabolites produced by three strains of T. harzianum. Mycelial plugs of T. harzianum were grown on cellophane discs placed on PDA for 2 days in the dark at 22-25"C. 'lhe cellophane discs and mycelial plugs were removed and a 10 ¡r.l droplet containing 25 ascospores were spread on the PDA and incubated for another 3-6 days. Y-axis denotes the average percentage of ascospores germinated in the eight replicates per each treatment. Bars denote standard errors.
s (! 1 00 G 90 u¡ 80 o o 70 o o 60 CL at o 50 o (!o Æ o 30 Ê, o 20 o 10 .= E 0 o o Control strain 1 strain 2 strain 3 Treatment with L hatzianum 58
Figure 4.5. Growth of E. lata on sterilised canes in the presence of T. harzianum,based on the method of Mercer and Kirk (1984a). Sterilised cane segments of cv. Shiraz were inoculated at opposite ends with 5 mm plugs of E. lata and T. harzianum or PDA and incubated at 23-25"C in the dark. E. Iata was applied 5 days prior to T. harzianum.
Growth of E. lata was measured 3 days after the antagonist was applied. Y-axis denotes the average growth of the eight replicates of E. lata receiving each treatment combination. Bars denote standard efrors.
50 o 45 G 40 N F 35 t- tt5o c,) 30 Ëa 25 oÂl< õE 20 s u 15 o 10 E 5 =o o 0 T1 T2 T3 T4 Treatment
Tl'= E. tataM2SO + PDA (autoclaved cane) - control
T2 = E. tataM2SO + PDA (gamma-irradiated cane) - control
T3 = E. tataM280 + T. harzianum strain 1 (autoclaved cane)
T4 = E. tataM2SO + T. harzianum sttain 1 (gamma-irradiated cane) 59
Figure 4.6. Growth of E. lata on autoclaved canes in the presence of T. harzianum, based on the method of Munkvold and Marois (1993a). Autoclaved cane segments of cv.
Shiraz wsre inoculated with spores or mycelial plugs of T. harzianum and incubated in darkness at 23oC. Mycelial plugs of E. lata were introdtced 24 h later and the segments were incubated for a further 10 days.
100 90 80 870 H60G
$soo oÐ40 oð30 Ê20 10 0 T1 r2 T3 14 Treatment
T1 = PDA plug + plug of E.lata M280 (control)
T2 = l0 ¡rl droplet of SDW + plug of E. lata M280 (control)
T3 = Plug of T. harzianum (strain 1) + plug of E. lata M280
T4 = 10 ¡rl droplet of SDW containing 1,500 spores of T. harzianum (strain 1) + plug of
E.lata M280
There were 10 replicates per treatment combination. 60
Figure 4.7. Colonisation of autoclaved canes by E. lata in the presence of spores of Z. harzianum. Autoclaved cane segments of cv. Shiraz were co-inoculated with a 25 ¡t'l droplet containing 1,000 spores each of E. lata and T. harzianum simultaneously and incubated at23"C in darkness. Re-isolation of pathogen and antagonist was carried out 4 weeks later.
1 00
s 90 80 .!2c o cD 70 G tr (E 60 tr o I Pathogen ctì 50 o E trAntagonist (E 40 CL o 30 o 20 o o o 10 É, 0 T1 T2 T3 Treatment
Tl = E. lata ascospores + T. harzianum (strain 1) spores
T2 = T. harzianum (strain 1) spores alone (control)
T3 = E. Iata ascospores alone (control)
There were 10 replicates per treatment combination. 61
4.3.3. Microscopy
The mycelium of E. lata was similar in dimensions and branching patterns to that of
T. harzianum,hence distinguishing the mycelia of the two species in co-inoculated canes using microscopy was not possible. Both the pathogen and the antagonist were observed to grow not only in the vessels but also in the pith parenchyma cells of both the gÍìÍìma- irradiated canes and the canes grown in rockwool pieces. No hyphae were observed in the un-inoculated canes, although structures resembling rod-shaped bacteria were observed occasionally, predominantly in the vessels'
4.3.3.1. Interaction in gamma-inadiated wood
Turgid and healthy hyphae were observed in the cane pieces that had been inoculated with M280 alone (Figure 4.8 A) or T. harzianum alone (Figure 4.8 B). SEM of co- inoculated segments showed hyphae that were flaccid and collapsed (Figure 4.8 C).
Similar observations were made for all three replicates.
4.3.3.2. Interaction in cuttings grown in rockwool pieces in the laborøtory
The controls, which received E. lata alone or T. harzianum alone, showed healthy,
turgid hyphae. Loss of turgor and abnormal swelling of hyphae were observed in cuttings
co-inoculated with mycelial plugs of E. lata M280 and 2 days later with spores of T.
harzianum (Figures 4.9 and 4.9 B). Hyphae were observed to have lost turgidity and to ^ have shrivelled in the cuttings co-inoculated with mycelial plugs of both pathogen and
antagonist (Figure 4.9 C). Abnormal swelling and collapse of hyphae were observed in
the cuttings co-inoculated with spores of the pathogen and antagonist (Figure 4.9 D, 4.9 62
E and 4.9Ð. Furthermore, parallel growth and winding or coiling of hyphae were also observed in this material. This may indicate early stages of parasitism.
4,3.3.3. Interactions ín simultaneously co-inoculated cuttings grown in rockwool
pieces in the laboratory
Spores of T. harzianum and E. lata were observed to have germinated and produced mycelia in the controls 22 h after inoculation and interacting mycelia were first observed
7 days after inoculation. There were no obvious signs of interactions in cuttings harvested 30 min., 6 h and 22 h after inoculation. Healthy and turgid hyphae were observed at all times of harvest in cuttings that were inoculated with ?. harzianum or E' lata alone (Figures 4.10 A and 4.10 B). In the co-inoculated cuttings, abnormal swelling and collapse of hyphae were generally observed in the cuttings harvested aftet 7, L4 and
21 days (Figures 4.10 C, 4.10 D and 4.10 F). Also, parallel growth and looping or coiling of hyphae were observed in the cuttings (Figures 4.10 E and 4.10 F). This may be an indication of early stages of parasitic interactions. 63
Figures 4.84. - 4.8C. SEM of gamma-irradiated and inoculated cane segments. Pink affows indicate healthy hyphae of E. lata or T. harzianum. Figure 4.84. Turgid hyphae of E. lata in a xylem vessel adjacent to pith parenchyma in cane segment inoculated with
E. lata alone. Figure 4.88. Turgid hyphae of T. harzianum growing through a pit of a xylem vessel in cane segment inoculated with T. harzianum alone. Figure 4.8C. Flaccid hyphae in co-inoculated cane segments.
Figures 4.9^ - 4.9F. SEM of cuttings inoculated first with E. lata and 2 days later with
T. harzianu¡ø. Pink affows indicate healthy hyphae of E.lata or T. hørzianum. Figure
4.94. Flaccid hyphae in cutting co-inoculated with mycelial plugs. Figure 4.98.
Abnormal swelling of hyphae in cutting co-inoculated with mycelial plugs' Figure 4.9C.
Shrivelled hypha in cutting co-inoculated with mycelial plugs. Figure 4.9D. Abnormal swelling of hyphae in cutting co-inoculated with spores. Figure 4.9ß'. '4.9F. I-oss of turgidity and parallel growth of hyphae in cutting co-inoculated with spores.
Figures 4.104. - 4.10F. SEM of cuttings simultaneously inoculated with spores of E.
Iata and T. harzianum. Pink affows indicate healthy hyphae. Figure 4.104. Turgid
hyphae in cutting treated with E. lata alone. Figure 4.108. Healthy hyphae of T.
harzianumin pith parenchyma in cutting treated with T. harzianum alone. Figure 4.10C.
Hyphae showing coiling, abnormal swelling and collapse in co-inoculated cutting.
Figure 4.10D. Rupture and collapse of hyphae in co-inoculated cutting. Figure 4.108.
Parallel growth of hyphae in co-inoculated cutting. Figure 4.10F. Coiling and abnormal
swelling of hyphae in co-inoculated cutting. 64
Figure 4.84. Figure 4.88.
Figure 4.8C.
t
I lt
It Um 65
Figure 4.9^. Figure 4.98. I -!
.{rfi;lfe Flaccid
Abnormal swelling
Figure 4.9C. Figure 4.9D.
Shrivelled hypha
Abnormal swelling
/
Figure 4.9E. Figure 4.9F.
Flaccid Flaccid hyphae
10 Um 66
Figure 4.104. Figure 4.108.
G
?
Figure 4.10C. Figure 4.10D. I 7 ì Collapse Rupture
Coiling
Abnormal
Figure 4.108. Figure 4.10F.
ã Parallel growth
f Coiling \ Abnormal 67
4.2. Discussion
Mycelial growth of E. lata on PDA was inhibited by antibiosis by all three strains of
T. harzianum blult there was no evidence of mycoparasitism when interaction zones of dual cultures were examined using light microscopy. There are many reports of inhibition of plant pathogenic fungi by antibiotic compounds produced by Trichoderma spp. (Fravel, 1988; Ghisalberti et al., 1990; Ghisalberti and Sivasithamparam, l99I)'
Volatile metabolites produced by Trichoderma spp. have been shown to inhibit various fungi (see section 2.4.3.4). The three strâins used here produced a coconut odour, previously characterized as 6-n-pentyl-2F-pytan-2-one (Claydon et al', 1987), and shown to inhibit mycelial growth of Ophiostoma ulmi, Botrytis cinerea (Merlier et al.,
1985) and R. solani (Dennis and Webster,lgTla). Other volatile antibiotics may include isocyanide compounds (Ghisalberti and Sivasithamparam, 1991). Chambers and Scott
(1995) reported inhibition of P. cinnamomi and P. citricola by an odourless, volatile antibiotic that was produced by T. pseudokoningii, and partial suppression of growth of
S. rolfsii was caused by volatile inhibitors produced by T. virens (Maiti et al., 1991)'
Volatile antibiotics alone produced by Trichodermd spp. were thought to be responsible
for inhibition of the wood decay fungus Lentinus lepideus (Bruce et aI., 1984). The
present study also suggested that the volatile metabolites produced by the three strains of
T. harzianummightbe sufficient to inhibit E. lata. However, further testing with more
isolates of E. lata would be necessary before a conclusion is reached.
The volatile metabolites produced by the three strains of T. harzianum inhibited
mycelial growth of the two isolates of E. lata (Figtxe 4.1) and had a fungistatic effect on
both. Also, volatile metabolites prevented germination of ascospores of E' Iata (Table 68
4.1). However, incubation for more than 2 days to observe germination was not possible since the bases of plates inoculated with ascospores and inverted over the bases containing T. harzianum weÍe contaminated after 2 days. Hence, whether or not these volatile metabolites had a fungicidal effect on the ascospores of E. hta is unknown.
In the experiments designed to assess production of non-volatile metabolites, it was assumed that the inhibition of growth of mycelium or of germination of ascospores of E. lata was caused by metabolites of T. harzianurn, which had diffused through the cellophane membrane into the PDA. Nutrient depletion by the growth of T. harzianum was considered unlikely to account for the inhibition of E. lata since PDA is rich in nutrients and incubation was only for 2 days (Dennis and Webster, 1971a). The cellophane overlay method has also been used to demonstrate anta3onism by non- volatile metabolites produced by Trichoderma spp. on other fungi such as F. annosus,
Fusarium oxysporum and Phytophthora spp. (Dennis and Webster, l97la; Chambers,
lgg3). Several authors have used non-volatile antibiotics in culture filtrates to demonstrate inhibition by antibiosis without the possibly confounding effects of nutrient depletion. Sivan et al. (1984) showed inhibition of growth of Pythium aphanidermatum by culture filtrate of T. harzianum. Similarly, Chambers and Scott (1995) demonstrated
the ability of culture filtrates of T. hamcttum, T. pseudokoningii and T. virens to inhibit P.
cinnamomi and P. citricola. However, the culture filtrate method was not used in the
present study to demonstrate antibiosis. Carter and Price (1974) demonstrated in a dual
culture experiment that non-volatile antibiotics produced by 6-10 day-old F. lateritium
cultures on Czapek-dox minerals plus yeast extract medium (57o nutrient strength)
significantly inhibited germination and mycelial growth of E. lata. Their findings further 69
suggest that depletion of nutrients in the culture medium was unlikely to contribute towards inhibition of the pathogen.
Schirmböck et al. (1994) suggested that the group of peptaibols produced by T. harzianum was ïesponsible for the antagonistic activity against B. cinerea and Fusarium oxysporum f . sp. phaseoll. Reduction in radial growth of S. rolfsii has been attributed to the trichorzianines produced by T. harzianum (Correa et at. 1'995). In the present study, identification of the anti-microbial compounds produced by the three strains of T. harzianum \ilas not attempted, and this warrants investigation as part of the evaluation of the potential of T. harzianum to control E. lata'
The non-volatile metabolites produced by the three strains had a fungistatic effect on some isolates of E. lata and a fungicidal effect on others. Also, the degree of inhibition bythethreestrainsof T. harzianumvariedwiththedifferentisolates of E'lqta (Figures
4.2 and 4.3). This observation suggested that the isolates of E. lata may differ in sensitivity to the diffusible metabolites produced by the three strains of Z. harzianum.
This is in agleement with previous work, which demonstrated that isolates of a pathogen may differ in sensitivity towards antagonists (Berg and Ballen, 1994; Mazzola et al.,
lgg5). This feature in E. Iatct is not surprising given the high genetic diversity of this pathogen (Péros et a1.,1997; Pèros andBerger, 1999; Péros et al., L999). Gibbs (1967)
showed that F. annosus isolates from different geographical locations varied in their
sensitivity to antibiotics produced,by Trichoderma spp. Schmidt et al. (2001a) observed
that a strain of Erwinia herbicola inhibited mycelial growth of the five isolates of E. lata
mainly by the hydrolases such as chitinase, protease and cellulase they produced on
grapevine wood while a sixth isolate was only weakly inhibited by these metabolites. 70
The difference in the degree of inhibition by the three strains of T. harzianurn observed in the present study suggests that the three strains may have produced different antibiotics or enzymes (Kullnig et al., 2000) or the same metabolites at varying concentrations. Research has shown that strains belonging to the same species group can differ physiologically in their production of metabolites (Moubashet, 1963; Dennis and
'Webster, l97la). Mercer and Kirk (19S4a) demonstrated variation in the antagonistic activity of three isolates of T. viride towards Chondrostereum purpureum, in that, of the three isolates of T. viride tested, one strongly inhibited the pathogen, one slightly inhibited and the other one stimulated the mycelial growth of the pathogen'
Ascospores did not germinate in the presence of diffusible metabolites produced by strains 2 and3 of T, harzianum and less than 57o of spores germinated in the presence of metabolites produced by strain 1. In the controls, germination was slightly more than gOTo (Figure 4.4). Carter and Price (1974) demonstrated inhibition of germination of ascospores of E. lata by non-volatile metabolite(s) produced by F' Iateritium. They showed that the degree of inhibition by the non-volatile antibiotics produced by 6-day- old cultures of F. lateritium varied from 100 to 7O7o in the absence of nutrients, whereas tp to 967o of untreated E. lata ascospores germinated after 24 h of incubation. However,
the inhibition was not always permanent, as incubation over 24hled to an increase in the
percentage of germination of ascospores of E. lata.In a separate time-course experiment
they also showed that inhibition of germination of ascospores of E. lata increased with
the age of F. Iateritium colony, over 6 successive days' They observed that the
germination of ascospores decreased rapidly during the first 2 days and reached l7o at6
days. Ferreira et al. (I99I),likewise, observed reduction in ascospore germination in 7L dual cultures with Bacillus subtilis. Similarly, Pachenari and Dix (1980) found that low molecular weight substances from Gliocladium roseum inhibited the germination of conidia and subsequent mycelial growth of Botrytis allii.
In the experiment based on the method of Munkvold and Marois (1993a), complete inhibition of growth and establishment of E. lata mycelia was observed when canes were first inoculated with spores or mycelial plugs of the antagonist (Figure 4.6). Although the
E. lata isolate used was known to be virulent, it was unable to establish in any of the replicates in the presence of the antagonist. In this experiment, E. lata was introduced24 h after the antagonist. This time interval may have been sufficient fot T. harzianum to colonise the canes and have a competitive advantage over the pathogen, which was introduced in the form of mycelial plugs. Conidia of T. harzianum getminate and initiate mycelial extension in 14 to 18 h at optimal temperature and nutrient conditions (Lifshitz
et al., 1986; Hjelj ord et at., 2000). Munkvold and Marois (1993a) reported reduction in
infection of 807o or more by ascospores of E. lata when applied to autoclaved cane
segments embedded upright in water agar pre-inoculated 48 h previously with conidia of
F. lateritium and Cladosporiurn herbarum.They also reported that I%o of the 348 fungal
isolates they had screened prevented germination of ascospores of E lata in wood.
E. lata was not re-isolated from autoclaved cane segments that were inoculated
simultaneously with equal concentrations of spores of the antagonist and the pathogen
(Figure 4.7). While the reasons for this are not known, it is possible that conidia of Z.
harzianum spores inhibited further development of the germlings of E. lata that may
have germinated ahead of the conidia. At the optimum temperature of 20-25"C, freshly
discharged ascospores were shown to germinate on agar in ll-12 h (Carter, 1991). The 72 failure of E. lata to colonise the canes in this particular experiment could be attributed to the depletion in Oz and increase in COz levels in the microenvironment since the plates were sealed with Parafilm@, which is likely to have reduced gaseous diffusion' Increase in respiration during germination of conidia and ascospores has been previously reported
(Martin and Nicolas, 1970; Rosen et a1.,1974).
An obvious criticism of these experiments involving wood is that gowth conditions for T. harzianunl and E. lata were optimised by maintaining high humidity and moisture content of the wood and by eliminating competition by the indigenous microflora.
Munkvold and Marois (1994) reported the microflora on fresh grapevine pruning wounds tobe as low as 102- 103cfu initially, butthis increasedto 106within 3-28 days.
Furthermore, since the wood tissues were killed by sterilisation, the natural resistance reactions such as accumulation of lignin, suberin and phenolics, and formation of tyloses
would not have taken place (Schmidt et a1.,2001a). Interactions between pathogen and
antagonist may differ somewhat depending on the experimental conditions. For example,
Mercer and Kirk (1984a) showed T. viride and T. koningii to be equally effective in
controlling wood decay fungi, such as Stereum hirsutum, Ganoderma applanatum and C.
purpureum, when applied as mycelial plugs to autoclaved beech wood strips placed on
2To malt agar and on beech wood strips placed on glass beads in Petri dishes containing
distilled water. However, they found other antagonists, such as F. Iateritium and
Cryptosporiopsis fasiculatø, to be less effective against these pathogens on wood than
they had been on agar alone. Reduced antagonistic activity on autoclaved wood has also
been reported by others (Toole, I97l; Ptatt, 1982). Nevertheless, tests on sterilised
wood, generally autoclaved, have proved valuable in the selection of biological agents 73 for the control of wood pathogens (Mercer and Kirk, 1984b; Ferreira et al., 1990;
Munkvold and Marois, !993a; Hutchinson et a1.,1994; Schmidt et a1.,2001a).
Although T. harzianum is well known for its mycoparasitic activity, the SEM observations suggested that inhibition on gamma-irradiated canes and cuttings grown in rockwool in the laboratory was primarily due to antibiosis (Chet and Baker, 1981; Chet et a1.,1981;Elad et aL,1983c). Abnormal swelling, collapse and shrivelling of hyphae were frequently observed in the gamma-irradiated and living co-inoculated canes, which may be attributed to antibiosis. Lifshitz et at. (1986) reported abnormal hyphal swelling in Pythium sp. that was grown in dual culture plates with Z. hq.rzianum in water agar medium before mycoparasitism occurred.
Parallel growth and coiling were observed occasionally in the co-inoculated cuttings.
These features may indicate early stages of mycoparasitic activity of Trichoderma spp', as reported by many researchers (Weindling, 1932; Dennis and Webster, 797Lc; Wells ¿r al., 1972; Hennis and Chet, 1975; Tronsmo and Dennis, 1978; Harman et al., l980i
Harman et a1.,1981; Elad et a1.,1983b; Elad et al., 1983c; Trutmann and Keane, 1990;
Benhamou and Chet, 1993; Chambers and Scott, 1995; Benhamou and Chet,1997). The lytic enzymes chitinase, glucanase and proteases have been reported to be responsible for
degradation of fungal cells by mycoparasitic T. harzianum (Elad et al., t982; Chérif and
Benhamou, 1990; Benhamou and Chet, 1993; Harman et aI., 1993). Some authors have
demonstrated the synergistic effect on antifungal activity by antibiotics and the lytic
enymes involved in parasitism. Kullnig et al. (2000) showed that enzyme diffusion from
Trichoderma atroviride \ryas necessary before the antagonist could parasitise R. solani.
Di Pietro et al. (1993) reported that an endochitinase and gliotoxin isolated from Z. 74 virens did not have a significant effect on germination of conidia of B.cinere¿ when either was applied alone, whereas, when the antibiotic and the enzyme were applied in combination there \üas a 957o inhtbition of germination. A similar effect was observed on B. cinerea conidia due to trichorazianines and the lytic enzymes endochitinase, chitobiosidase and glucanase produced by T. harzianum (Schirmböck et al., 1994).
Lorito et at. (1996) used this concept of synergism to obtain high levels of inhibition of
B. cinerea and Fusarium ucysporum tsing seven different antibiotics and two fungicides in all combinations with eight enzymes from fungi, bacteria or plants. However, the relationship between antibiotics and lytic enzymes of T. harzianum and their synergistic effects on E. lata was not investigated in the present study. Further investigation of this aspect may be valuable in future studies.
Although coiling and winding \üere observed in cane segments and cuttings using
SEM, they were not detected in the in vitro study of dual cultures on PDA using light microscopy (sections 4.3.1.2; 4.3.3.1; 4.3.3.2 and 4.3.3.3). However, parasitism of E. lataby F. lateritiunx oî agar medium has been reported by Vajna (1986). Nutrient stress is thought to be necessary for the expression of some cell wall degrading enzymes involved in mycoparasitism by Trichoderma spp. (Lorito, 1998; Mach et aI', 1999). The conditions in wood segments in this case may have been conducive for parasitic interactions between T. harzianum and E. Iata compared to the nutrient-rich PDA
medium. Furthermore, increased antibiotic activity has been reported to impede parasitic
interactions between UV-induced mutants of T. hqrzianum and P. ultimum, R. solani and
F. oxysporum (Graeme-Cook and Faull, 1991). Also, Howell and Stipanovic (1995)
showed that gliotoxin-deficient mutants of T. viride which had lost antibiotic activity 75
against R. solani were as efficient as the parental strains in controlling disease of cotton seedlings induced by R. solani, obviously by a mechanism other than antibiosis. Hence, if strain 1 of T. harzianum was a "high" antibiotic producing strain, it may not readily have parasitised E. lata in the conditions tested. This hypothesis, however, needs further testing.
In summary, inhibition of germination and mycelial growth of E. lata on PDA by L harzianum was primarily by antibiosis. SEM suggested that inhibition in grapevine tissue was mainly by antibiosis but that parasitism may have occurred in cuttings. Spores and mycelial inoculum of T. harzianum were capable of inhibitin g E. lata in grapevine cane segments in laboratory conditions when co-inoculated or when the pathogen was introduced before or after the antagonist. The potential of the antibiotics produced by 7' harzianum to control E. lata should be evaluated and tested on a wide range of isolates of E. lata in future studies. Also, the relationship between the antibiotic activity of Z. harzianum and the lytic enzymes involved in mycoparasitism needs to be investigated.
The results of experiments carried out in this chapter indicated that T. harzianum has potential in the biocontrol of E. lata on grapevines. 76
CIIAPTER 5. ANTAGONISM OF E. I^ATA IN PLANTA AND COLONISATION
OF CANES BY PATHOGEN AND ANTAGONIST
5.1. Introduction
In planta tests in glasshouse conditions are considered by most researchers to be a prerequisite for field experiments. If the candidate organism is unable to control the pathogen or the disease in the glasshouse, it is unlikely to be able to do so in the variable environmental conditions in the field.
The ability of various biological agents to prevent or reduce infection of wounds by fungal pathogens has been evaluated in the glasshouse. Spraying of roses grown in pots, soon after pruning, with a spore suspension of the saprophyte Ulocladium atrum reduced infection by B. cinerea,by reducing sporulation of the pathogen occuring naturally on senescing plant material in the glasshouse (Kohl and Gerlagh, 1999). Efficacy of stem
wound applications of U. atrum against infection by B. cinereø of tomatoes grown in the
glasshouse was shown to be equivalent to the control achieved by treatment with
fungicides (Fruit et al.,Ig9g). Wound applications of 7. harzianum and other fungi such
as Aureobasidium pullulans and Cryptococcus qlbidus reduced the incidence of stem
infection of cucumber plants by B. cinerea in the glasshouse (Dik et al., 1999). Non-
pathogenic strains of Fusarium spp. protected wounds on tomatoes against B. cinerea in
the glasshouse (Decognet et al., 1999; Trottin et a1.,2001)'
Wound protection using biological agents has been successful in many tree species'
Treatment of pine stumps with Peniophora gigantea to protect against infection by 77
Fornes annosus is widely practised in Britain (Rishbeth, 1963). Application of P. gigantea and Polysporus adustu.s to spruce stumps has also shown potential to protect against infection by Stereum sanguinolentum (Rishbeth, 1973). }JalT et al. (L986) demonstrated that colonisation of stems of seedlings of silver and Norway maples by V. dahliae, the causal agent of wilt, was reduced when stem wounds were inoculated with isolates of Bacillus subtilis prior to the introduction of the pathogen.
Boirie and Pons (1984) showed that the product, Phior-PrM, based on T. harzianum, when daubed on wounds or sprayed at the time of pruning prevented infection of fruit trees by C. purpureum.However, Spiers and Brewster (1997) did not find treatment of wounds on willow and peach trees with T. viride to prevent infection by C. purpureum.
Infection of pruning wounds on grapevines and apricots by E. lata has been prevented by the introduction of various fungi and bacteria (Carter and Price, I974;Feneita et al.,
1991; Munkvold and Marois,I993a). Carter and Price (I974) showed that inoculation of freshly pruned sapwood of apricots with F. lateritium gave significant protection against infection by E. lata. They also reported that F. lateritiun was restricted mainly to the sapwood within 2 cm of the pruned surface 6 months after inoculation. Treatment of grapevine pruning wounds with B. subtilis significantly reduced infection by E. lata
when the pathogen was introduced to the wounds 4h after the antagonist (Ferreira et al.,
1991). Munkvold and Marois (1993a) demonstrated biological control of E. lata on
grapevines by treating pruning wounds with F. lateritium and Cladosporium herbarum,
2 or 14 days prior to the introduction of the pathogen. Furthermore, Munkvold and
Marois (1993a) found T. viride to be moderately effective in reducing infection of
grapevine pruning wounds by E. lata and inconsistent in comparison with F. Iateritium 78 and C. herbarum. However, apart from these reports of biological control of E. Iata, information is lacking on the interactions between E.lata and fungal antagonists. Further information might lead to optimising biological protection of pruning wounds on grapevines, which would facilitate the development of commercial biological control products.
The objectives of the experiments described in this section were to test antagonism of
T. harzianum against E. lata in planta, to evaluate different methods of introduction of the antagonist into grapevine cuttings and the extent of subsequent colonisation. F lateritium, which is a known antagonist of E. lata, was used for the sake of comparison in some experiments (Carter and Price, l9l4; Carter, 1983). The technique used to inoculate the canes under the upper bud in some of the experiments was adapted from the method used by Péros and Berger (1994) to test the pathogenicity of E' lata isolates on grapevine canes.
5.2. Materials and methods
5.2.1. Experiment 5.1. Colonisation of cuttings by antagonists
The aim of this experiment was to study colonisation of cane cuttings by the
antagonists when introduced as mycelial plugs or spore suspensions on pruning wounds
made above the upper bud or into holes drilled under the upper bud. Cane cuttings of
Shiraz were allowed to stand in tap water overnight after the third basal bud was
removed (see section 3.1;Table 3.1) then inoculated as follows. 79
Pruning wounds were made on 96 cuttings, 4 cm above the upper bud and the cut surface was inoculated ("top inoculation") with one of the following treatments:
Treatment I = F.Iateritium spores @0 ¡tl droplet)
Treatment 2 = mycelial plugs (5 mm diam.) of F. lateritium onCDA
Treatment 3 = T. harzianum (strain 1) spores ØO ttl)
Treatment 4 = mycelial plugs (5 mm diam.) of T. harzianum (sftain 1) on PDA
Treatment 5 = SDW ØO ttl)
Treatment 6 = sterile PDA plug (5 mm diam')
The concentration of the spore suspensions of both F. lateritium and T. harzianurø used in this experiment was 10e spores/ml (see section 3.3.1).
Another 96 cuttings were inoculated via holes (5 mm diam.) drilled 3 cm below the upper bud, the treatments listed above were administered ("middle inoculation") into the holes and the wound covered with Parafilm@.
The cuttings were harvested at two times; there were eight replicates per treatment for
each of the "top" and "middle inoculated" positions at each harvest. The experiment was
a split plot design with no blocks. Twelve canes were arranged in each of 16 trays
(section 3.1). Within a tray, each cane received one of the 12 combinations of treatment
(T1 - T6) and position of inoculation. The cuttings were maintained in the glasshouse as
described in section 3.1. Half the replicates were harvested 10 weeks after inoculation
(96 canes), and re-isolation of the antagonists was carried out (see section 3.4.2) from the
discoloured tissue (i.e. 3-5 mm below the wound) as well as the healthy tissue (i.e. up to
2 cm below the point of inoculation after removal of discoloured tissue). Transverse
sections were also cut by hand using double-edged Gillettet Silu"t Blue razor blades, 80 from three replicates of each treatment combination, just beyond the margin of the discoloured tissues. The sections were mounted in cotton blue in lactophenol and observed with the light microscope (Leitz Orthoplan 871288) for the presence of spores or mycelia within the xylem vessels.
The other 96 canes were harvested 20 weeks after inoculation. In this case, re- isolation of the antagonists was from the discoloured tissues at the wound site as well as from wood at I, 2 and, 4 cm below the point of inoculation for the "top" and "middle inoculated" canes (see section 3.4.2). Additionally, re-isolation was carried out at l, 2 and 4 cm above the point of inoculation for the "middle inoculated" cuttings.
Logistic regression was performed on the results (see section 3.6) for re-isolation of the antagonists from different positions on the canes 10 and 2O weeks after inoculation.
The control treatments with SDW and sterile PDA plugs were not included in the analysis since isolation yielded no fungi, confirming that the canes were not contaminated with the antagonist.
5.2.2.Experiment 5.2. Effectof T. harzinnum on infection of cuttingsby E.latø
The aim of this experiment was to study the effect of T. harzianum oî infection from mycelial plugs of E. lata in grapevine cuttings when introduced at a distance from the
antagonist on the same day. Holes (5 mm diam.) were drilled 1 cm below the upper bud
in 60 canes each of cultivars Chardonnay and Shiraz (see section 3.1, Table 3.1)' The
holes in 30 canes of each cultivar were inoculated with 5 mm diam. plugs of T.
harzianum (strain 1) while the other 30 canes of each cultivar were inoculated with plugs
of sterile PDA of the same dimensions. Holes of 3 mm in diam. were drilled I cm below 81
the upper bud in another 60 canes each of cultivars Chardonnay and Shiraz and the holes in 30 of the canes of each cultivar were inoculated with 40 ¡tl of spore suspension (10e spores/ml) of T. harzianum strain 1 (see section 3.3.1). The other 30 canes of each cultivar were inoculated with 40 ¡t"l of SDW.
On the same day, each cane was pruned 4 cm above the upper bud and a mycelial plug (5 mm diam.) of E. lata isolate M280 placed on the cut surface, then covered with
Parafilm@. The experiment was set up in a split plot design with no blocks. There were
10 replicates per treatment combination. There were 30 trays, each containing eight canes representing the eight treatment combinations. The canes were harvested 11, 14 and 20 weeks after inoculation. The discoloured portion from top of the cane (3-5 mm) was discarded and re-isolation of E. lata from 1 cm segments below the point of inoculations was carried out as described in section 3.4.1.
5.2.3.Experiment 5.3. Effect of antagonists on infection of cuttings by E. løta
The aim of this experiment was to study the effect of T. harzianum and F. lateritium on infection by E. lata in planta when introduced at a distance from the antagonist at three different time intervals. Mycelial plugs (5 mm diam.) of T. harzianum (strain I) or
F. Iateritium were inserted into holes, 5 mm diam., drilled 2.5-3 cm below the upper buds of 64 cuttings of cultivar Chardonnay (see section 3. 1, Table 3.1). Plugs of sterile
PDA of the same dimensions were appliedto 32 canes as controls. A further 32 canes
were treated with Benlate@, instead of inoculation with T. harzianum or F. lateritium.
All cuttings were pruned 4 cm above the upper bud on Day 0 and those designated
fungicide-treated controls were immediately painted with Benlatet (ZS g/L). The pruning 82 wounds on all cuttings were inoculated with plugs of E.lata isolate M280 or sterile PDA
(5 mm diam.) I andl days after pruning and treatment with the antagonists or Benlate@'
There were eight replicates for each of the 12 treatment combinations. The inoculum of the pathogen and antagonists was covered with Parafilm@ The cuttings were maintained in the glasshouse as described in section 3.1. The experiment was a split plot design with no blocks. Twelve canes were affanged in each of 16 trays. Within a tray, each cane received one of the 12 treatment combinations. The canes were harvested 11 weeks and
8 months after the introduction of antagonists or the treatment with Benlate@. The discoloured wounded portion (3-5 mm) was discarded and 1 cm segments from the top of the canes were excised for re-isolation of E. lata as described in section 3 '4.I .
5.2.4.Experiment 5.4. Protection of pruning \ilounds from infection by E. lata
This experiment was designed to investigate the ability of T. harzianutn strain 1, applied as a spore suspension in SDW and in Trichoseal@ base, to prevent infection of pruning wounds by E. lata. Cane cuttings of cultivar Shiraz were pruned 4 cm above the upper bud and the following treatments were administered to the wounds using 25 mm flat varnish brushes (see section 3.1; Table 3.1). Separate brushes were used for each treatment to avoid cross-contamination.
Treatment I (T1) - T. harzianum (strain 1) in Trichoseal@ base (100 g/L)
Treatment 2 (T2) - T. harzianum (strain 1) spore suspension (10e spores/ml SDW)
Treatment 3 (T3) = gamma-iffadiated T. harzianurn (strainl) in Trichoseal@ base
(100s/L)
Treatment4(T4)=SDW 83
plugs (5 mm diam.) of E. lata isolate M280 were introduced to the wounds 0,2 andT days after the above treatments were administered, and covered with Parafilm@. Each treatment combination, which was randomly assigned to the canes in the glasshouse in a
12 x B (row x column) design, was replicated eight times. The treatment combinations were applied in a factorial affangement with four types of treatment with the antagonist or SDW control and three different times for the introduction of the pathogen. The canes were harvested 12 weeks after the application of treatments T1- T4, and re-isolation of E lata was performed (see section 3.4.1) from the wood 2 cm below the point of inoculation after the discoloured wound (3-5 mm) was discarded.
5.2.5.Experiment 5.5. Protection of pruning wounds with Trichoseal@
This experiment was similar to Experiment 4, except that two additional treatments were included and E. lata isolate M280 was applied 2 and 7 days after the antagonists.,
The additional treatments were (i) Trichoseal@ and (ii) gamma-irradiated Trichoseal@ as the corresponding control (see Table 3.3). The treatment combinations were applied in a factorial arangement with six types of treatment with the antagonist and two different
times of application of the pathogen. Re-isolation of E lata was carried out as above 12
weeks after the pruning wounds were treated with the antagonist (see section 3.4.1)'
This experiment was repeated but with re-isolation of E. lata and T. harzianum
carried out simultaneously (see sections 3.4.1 and3.4.2) at regular distances along the
length of the canes by excising 2 cm segments up to a distance of 10 cm below the point
of inoculation. Re-isolation along the whole length of the cane was carried out only for
cuttings inoculated on Day 2. The data for the two trials were analysed separately' 84
Logistic regression was performed for the isolation data 2 cm below the point of inoculation for the repeated trial.
5.2,6. Experiment 5.6. Protection by prior inoculation with antagonist
The aim of this experiment was to investigate the ability of T. harzianum strain I to protect grapevine cuttings from E. lata when inoculated prior to, and at a different position from, the pathogen. Holes (5 mm diam.) were drilled into 30 cuttings of cultivar
Chardonnay 1.5 cm below the upper bud. Plugs of T. harzianum strain 1 (5 mm diam') were inserted into the holes in 20 canes while plugs of sterile PDA were inserted into 10 canes. Inoculation sites were covered with Parafilm@ (see section 3.1; Table 3.1). The canes were arranged in a 10 x 3 (row x column) design in rockwool pieces in pots as described in section 3.1. After 7 days, 5-mm diam. holes were drilled I cm below the upper bud and inoculated with 5 mm diam. plugs of E. lata isolate M280 or sterile PDA.
Ten of the 20 canes that had been treated with plugs of T. harzianum teceived plugs of E.
lata whtle the other 10 canes were treated with plugs of sterile PDA. The 10 canes which
were initially treated with PDA plugs were also inoculated with E. lata. All treatments
were randomly administered to the canes. There were 10 replicates for each of the three
treatment combinations. The canes were harvested 12 weeks after the introduction of Z.
harzianum. The canes were cut into 2 cm pieces along the whole length and re-isolation
of E.lata and T. harzianum was carried out simultaneously (see sections3.4.2 and 3.4.1)
to estimate the extent of colonisation of wood by the pathogen and the antagonist. 85
Logistic regression (see section 3.6) was performed on the results to evaluate the effect of T. harzianum on E. Iata when the pathogen and antagonists were in close proximity to each other but inoculated at different points.
5.3. Results
5.3.1. Experiment 5.1. Extent of colonisation by antagonists
Microscopic examination of transverse sections of canes did not reveal spores or mycelium of the antagonists within the xylem vessels.
Re-isolation of antagonists from the discoloured tissue:
Both antagonistic fungi were re-isolated from the discoloured tissue from the "top" and "middle inoculated" canes, at 10 and 20 weeks. However, T. harzianum appeared not to have persisted in, or to have declined in, the discoloured tissue when applied as
spores. Neither fungus was re-isolated from the un-inoculated controls (Tables 5.1-5'3).
The logistic regression (see section 3.6) performed for the results for re-isolation of
antagonists from the discoloured tissues indicated that there was a significant interaction
(P=0) between the treatments and the position of inoculation (i.e. "top" or "middle"
inoculation). Also, there was a significant interaction (P=0.03) between the time of
harvests (i.e. 10 or 20 weeks) and the position of inoculation. The predicted probabilities
of re-isolating antagonists from the discoloured wound tissues at the "top inoculated"
canes were high, 1.00 for canes treated with mycelium of F. lateritiunt and mycelium of
T. harzianun, whereas canes that were treated with spores of T. harzianum did not yield 86 the antagonist (Table 5.4). The predicted probability of re-isolating F. lateritium ftom discoloured tissues of "top inoculated" canes was 0.75. Furthennore, the probabilities of re-isolating the antagonists from the discoloured tissues were similar for the "top" and
"middle inoculated" canes, 032 and 0.66 respectively, harvested 10 weeks after inoculation (Table 5.5). The antagonists were more likely to be isolated from dead tissues of canes that were "top inoculated" and harvested 20 weeks after inoculation than from "middle inoculated" canes, predicted probabilities being 1.00 and 0.56 respectively.
Re-isolation of antagonists at 1 cm below the point of inoculation:
T. harzianuLz persisted 1 cm below the point of inoculation at2} weeks in the cuttings that were "top-inoculated" only when applied as plugs of mycelia, wheteas it could not be re-isolated at this distance from cuttings treated with spores (Table 5.2). In the
"middle inoculated" canes, however, T. harzianum was re-isolated at this distance in more than half the cuttings that were treated with spores whereas it persisted in all the cuttings treated with the plugs of mycelia at 20 weeks (Table 5.3). F. lateritium persisted at 1cm in60-lOO7o of the "top" and "middle inoculated" cuttings (Tables 5.2and5'3).
Neither fungus was re-isolated from the un-inoculated controls. The logistic regression
analysis (see section 3.6) for re-isolation of antagonists 20 weeks after inoculation
indicated that there was no significant interaction between the effects of position of
inoculation ("top" and "middle") and the treatments (P=0.06). However, the main effects
of treatment (P=0) and position of inoculation (P=0.01) were significant. The
probabilities for re-isolating the antagonists at 1 cm below the point of inoculation for
the treatments with spores and mycelium of F. lateritium and mycelium of T. harziqnum 87
were high, being 0.88, 0.94 and 1.00, respectively, compared to those treated with spores of T. harzianum which was 0.31 (P=0) (Table 5.6). Also, the probability for re-isolating the antagonist 1 cm below the point of inoculation for the "middle inoculated" canes was significantly higher, at 0.87 (P=0.01), than that for the "top inoculated" canes, 0.69
(Table 5.7).
Table 5.L. Experiment 5.1: Number of canes, of eight, yielding antagonists from the discoloured and living tissues (2 cm below point of inoculation) 10 weeks after application of inoculum to pruning wounds ("top inoculation") or below the upper bud
("middle inoculation").
ttTop inoculation" "Middle inoculation"
Treatment Discoloured Living tissue Discoloured Living tissue tissue tissue
l. F. lateritium 1 7 8 8
spores
2. F. lateritium 8 8 8 8 myceliun 3.7. harzianum 0 8 8 8 strain I spores
4. T. harzianum 8 8 8 8
strain 1 mycelium
5. SDW 0 0 0 0
6. PDA plug 0 0 0 0 88
Table 5.2. Experiment 5.1: Number of "top inoculated" canes, of eight' yielding antagonists at various distances below point of inoculation after 2O weeks.
Treatment Discoloured 1 cm below 2 cm below 4 cmbelow tissue
l. F. Iateritium 5 6 6 8
spores
2. F. Iateritium 8 8 8 8 myceliun 3.7. harzianum 0 0 0 0 strain 1 spores
4.7. harzianum 8 8 8 7 strain I myceliun 5. SDW 0 0 0 0
6. PDA plug 0 0 0 0
Table 5.3. Experiment 5.1: Number of "middle inoculated" canes, of eight, yielding antagonists at various distances above and below the point of inoculation after 20 weeks.
Treatment Discolour 1cm 2cm 4cm Lcm 2cm 4cm -ed tissue below below below above above aboYe
I. F. lateritium 8 8 8 8 7 8 6
spores
2. F. lateritium 7 1 8 7 8 6 myceliun 3.T. harzianum 2 5 5 8 8 8 strain 1 spores
4. T. harzianum I 8 8 8 8 8 6 strain I mycelium
5. SDW 0 0 0 0 0 0 0
6. PDA plug 0 0 0 0 0 0 0 89
Table 5.4, Experiment 5.1: Predicted probabilities of re-isolating antagonists, irrespective of the harvest times, from the discoloured wound tissues of "top" and
"middle inoculated" canes when each treatment was applied to eight replicates.
Predicted probabilities Treatment Top Middle
9 L F. lateritium sporcs (40 ¡tl droplet, 10 0.75 1.00 spores/ml)
2. F.Iateritiummycelial plugs (5 mm diam.) 1.00 o.94
3. T. harzianum (straín 1) spores @0 ¡t"l droplet, 0.00 0.62
10e spores/ml).
4. Treatment 4 = mycelial plugs of Z. 1.00 0.56 harzianum strain 1 (5 mm diam.).
Trearment 5 ØO ttl SDV/) and Treatment 6 (sterile PDA plug) were not included in the analysis, as no fungi were isolated from the cane tissues.
Table 5.5. Experiment 5.1: Predicted probabilities of re-isolating the antagonists from the discoloured tissues of "top" and "middle inoculated" canes harvested 10 and 20
weeks after inoculation.
Predicted probabilities at the two Harvests
Position of inoculation 10 weeks 20 weeks "Top inoculation" o.72 1.00 "Middle inoculation" 0.66 0.56 90
Re-isolation of antagonists at 2 cm below the point of inoculation:
T. harzianum and F. lateritiurø persisted 2 cm below the point of inoculation in almost all of the cuttings at 10 weeks (Table 5.1). Howevet, T. harzianum was not recovered at this distance after 20 weeks from the cuttings that were "top inoculated" with spores, while more than 50 7o of the cuttings that were "middle inoculated" with the spores yielded T. harzianum (Tables 5.2 and 5.3). F. lateritium persisted in the majority of the "top" and "middle" inoculated cuttings at 20 weeks, while none of the controls yielded the antagonists (Tables 5.2 and 5.3).
The logistic regression model indicated a significant interaction between the time of harvest (10 or 20 weeks after inoculation) and the treatments (P=0.04). Furthermore, the main effect of the position of inoculation ("top" or "middle inoculation") was significant
(P=0). The probabilities for re-isolating the antagonists 2 cm below point of inoculation were high (P=0.04) for canes that were treated with the spores and mycelium of F lateritium and mycelium of T. harzianu¡n¿ when the canes were harvested after 10 and 20 weeks (Table 5.8). The probabilities of antagonists existing at 2 cm below point of inoculation were significantly higher (P=0) for the "middle inoculated" canes than for those which were "top inoculated" (Table 5.9). 9t
Table 5.6. Experiment 5.1: Predicted probabilities of re-isolating antagonists from wood that was lcm below the point of inoculation when canes were harvested 20 weeks after inoculation and each treatment was administered to eight replicates.
Treatment Probabilities
I. F.lateritium spores (a0 ¡A droplet, 10' 0.88 spores/ml).
2. F.lateritiummycel\al plugs (5 mm diam.) 0.94
3. T. harzianum (strain 1) spores @0 ¡.r'l 0.31
droplet, 10e spores/ml).
4. mycelial plugs of T. harzianutn strain 1 (5 1.00 mm diam.)
in the Treatmenr 5 @0 ¡tl SDV/) and Treatment 6 (sterile PDA plug) were not included analysis, as no fungi were isolated from the cane tissues.
Table 5.7. Experiment 5.1: Predicted probabilities of re-isolating antagonists from 1 cm below the point of inoculation for the "top" and "middle inoculated" canes when these
were harvested 20 weeks after inoculation.
ttTop inoculationtt "Middle inoculation" 0.69 0.87 92
Table 5.8. Experiment 5.1: Predicted probabilities of re-isolating the antagonists at2 cm below the point of inoculation for canes harvested 10 and 20 weeks after inoculation.
There were eight replicates per treatment.
Treatment 10 weeks 20 weeks
L F. lateritium spores @O ¡t"l droplet, I spores/ml). 0.94 0.87 2. F.Iateritiummycelial plugs (5 mm diam.) 1.00 1.00
3. T. harzianum strain 1 spores @O ¡tl droplet, 1 1.00 0.31 spores/ml).
4. Mycelial plugs of T. harzianum sttaín 1 (5 mm 1.00 1.00 diam.)
Treatment 5 Ø0 ttl SDU/) and Treatment 6 (sterile PDA plug) were not included in the analysis, as no fungi were isolated from the cane tissues.
Table 5.9. Experiment 5.1: Predicted probabilities for re-isolating the antagonists at 2 cm below point of inoculation from canes that had been inoculated at the "top" and
"middle positions".
ttTop inoculation" "Middle inoculation" 0.83 0.95
Re-isolation of antagonists from 4 cm below the point of inoculation:
At20 weeks both the antagonists persisted at 4 cm below the point of inoculation in
the majority (> 60 7o) of the cuttings (Tables 5.2 and 5.3) but, T. harzianurø did not
persist in the cuttings at this distance when applied as spores. None of the un-inoculated
controls yielded the fungus. 93
The regression model indicated a significant interaction between the position of inoculation and the ffeatments (P=0). The probability of re-isolating the antagonists from canes treated with spores and mycelium of F. lateritium and mycelium of T. harzianum was high for canes inoculated at both positions, while the probability of re-isolating T' harzianum 4 cm below the point of inoculation in canes that had been inoculated at the top with spores of T. harzianum was low (Table 5'10).
Re-isolation of antagonists at 1 cm above point of inoculation:
Both fungi persisted 1 cm above the point of inoculation at 20 weeks in 85-1007o of the cuttings (Table 5.3). The logistic regession performed on the results for re-isolation of antagonists indicated there was no significant effect of the treatments (P=0.41). The model predicted a high probability of 0.97 for re-isolating both antagonists at I cm above the point of inoculation.
Re-isolation of antagonists at 2 cm above point of inoculation:
Both fungi were re-isolated from 60-f0o7o of the cuttings at 20 weeks (Table 5.3).
There were no significant treatment effects (P=0.11) and the probability of re-isolating
the antagonists at 2 cm above the point of inoculation was 0.94 for all treatments.
Re-isolation of antagonists at 4 cm above point of inoculation:
Both fungi persisted 4 cm above the point of inoculation at 20 weeks in more than
607o of the cuttings (Table 5.3). There was no significant difference between the four
treatments (P=0.11). The probability of re-isolating the antagonists 4 cm above the point
of inoculation was 0.89. 94
Table 5.10. Experiment 5.1: Probabilities for re-isolating antagonists at 4 cm below the point of inoculation from the "top" and "middle inoculated" canes when these were harvested 20 weeks after inoculation. There were eight replicates per treatment'
Treatment ttTop inoculationtt "Middle inoculation" \. F.lateritium spores @0 ¡tl 1.00 1.00 droplet, 10e spores/ml). 2. F. Iateritium mycelial plugs 1,00 0.88 (5 mm diam.).
3. T. harzianum (strain l) 0.00 1.00
spores @0 ¡t"ldroplet, 10e spores/ml).
4. mycelial plugs of Z. 0.88 1.00
harzianum strain 1 (5 mm diam.).
included in the Treatment 5 Ø0 tt"l SDW) and Treatment 6 (sterile PDA plug) were not analysis, as no fungi were isolated from the cane tissues.
5.3.2. Experiment 5.2. Protection of cuttings by T. harziønurn
E. Iata was re-isolated from all segments of canes of Chardonnay and Shiraz
inoculated with the pathogen plus mycelial plugs or spores of T. harzianum, PDA or
SDW on the same day, 1 cm below the upper bud, regardless of time of harvest.
Statistical analysis, therefore, was not warranted' 95
5.3.3. Experiment 5.3. Protection of cuttings by antagonists
E. lata was recovered from the majority of the canes (60-lo07o of the cuttings) treated with the antagonist 1 and 7 days before the pathogen, at both times of harvest (Table
harvest 5. 1 1). All cuttings treated with E. Iata alone also yielded the pathogen at the two times. However, E. lata was not isolated from any of the cuttings treated with antagonists alone nor from those treated with E. lata andBenlate@ and harvested after 11 weeks. E. lata was isolated from some of the Benlate@-treated cuttings 8 months after inoculation (Table 5. 1 1).
According to the logistic regression analysis there were no interactions at the 57o level. The main effects of treatment (P=0) and harvest times (P=0.04) were significant and, therefore, were examined separately. The difference between the two times of inoculation of E. lata after the antagonists (i,e. Day 1 and Day 7) was not significant
(P=0.21). This indicated that application of T. harzianum or F. lateritium I or 7 days before E. lata did not prevent colonisation of the canes by the pathogen (Table 5.12).
Benlate@ was the only treatment that inhibited colonisation by E. lata. The probability of re-isolating E. Iata from the canes 8 months after inoculation was significantly higher
(P=0.04) than at 11 weeks (Table 5.13). 96
Table 5.L1. Experiment 5.3: Number of canes, of eight, yielding E. lata from 1 cm segments excised from the top of canes after 11 weeks or 8 months. The pathogen was introduced I or 7 days after application of the antagonists or Benate@.
Harvested at LL weeks Harvested at I months
Day 1 Day 7 Day L Day 7
T. harzianum strain l+ E 8 8 8 8 IataM280 F. lateritium + E. lata 6 8 8 7 M280
PDA plug + E. Iata 8 8 8 8 M280 Benlate + E. lataM280 0 0 1 J
T. harzianum strain 1+ 0 0 0 0 PDA
F. lateritium + PDA 0 0 0 0
Table 5.12. Experiment 5.3: The probability of re-isolation of E. lata from canes treated with the antagonists (eight replicates per treatment) when harvested 12 weeks after inoculation with the antagonists as estimated by the logistic regression model.
Treatment Probability of re-isolation of E.lata
T. harzianum strain I+ E.lata M280 1.00u F. Iateritium + E.lataM2SO 0.91u PDA plug + E.lata M280 1.00u Benlate + E.Iata M280 0.13b
Probabilities with the same letters (tre not significantly dffirent. 97
Table 5.L3. Experiment 5.3: The probability of canes (eight replicates per treatment) yielding E. lata at each time of harvest as estimated by the regression model across all treatments
ll Weeks 8 months 0.72" 0.gOd
Probabilities with the same letters are not significantly dffirent.
5.3.4. Experiment 5.4. Protection of pruning wounds
E. lata was recovered from all the replicates of the controls inoculated with the pathogen 2 and 7 days after treatment with the antagonist. E. lata was recovered in all of the cuttings inoculated with the pathogen on Day 0 except that it was re-isolated from only three of eight cuttings treated with gamma-irradiated T. harzianum sttain | \n
Trichoseal@ base. Also, the pathogen was recovered from one of eight and two of eight cuttings inoculated with E lata 2 days after treatment with 7. harzianum strain 1 in
Trichoseal@ base and T. harzianum strain 1 spores, respectively, and vice versa ftom cuttings inoculated with E lataT days after the antagonist (see Appendix 3).
The logistic regression indicated that there were significant interactions (P<0.001) between the effects of treatment and day at the 57o level. T. harzianum (sttain 1), both in
the form of spore suspension in SDW and in the Trichoseal@ base, reduced (P<0.001)
colonisation of the canes by E. lata when the pathogen was introduced after 2 ot 7 days
after the antagonist (Table 5.I4). Introducing the pathogen at the two different times
(Day 2 and Day 7) did not have a significant impact on infectionby E. lata (Table 5.14)'
Colonisation by E. lata was not reduced compared to the controls (SDSD when the
pathogen and antagonist were applied on the same day. However, the canes treated with 98 ganrma-iradiated T. harzianum strain 1 in Trichoseal@ showed significant increase in infection by E. lata (P<0.001), when introduced 2 or J days after the antagonist compared to those treated with Z. harzianum strain 1 in Trichoseal@ and inoculated with
E.lata on the same day.
Table 5.L4. Experiment 5.4: Predicted probabilities of infection by E. lata of canes as estimated by the logistic regression model. E. lata was introduced 0, 2 and 7 days aftet the antagonist and the canes were harvested 12 weeks later. There were eight replicates per treatment.
Probabilities that canes are infecteùby E.løta
Day of T1 T2 T3 T4 inoculation of E. htø
0 0.99" (0.00) 0.99'(0.00) 0.43" (0.19) 0.999 (o.oo)
2 0.13" (0.12) 0.38d (0.17) 0.991(o.oo) 0.99c (0.00)
7 0.25b (0.15) 0.13d (0.12) 0.99'(o.oo) 0.99s (0.00)
Probabilities with the same letters in each column are not significantly dffirent. Standard error values
in parentheses.
Treatment 1 (T1) = T. harzianum (strain 1) in Trichoseal@ base (100 g/L)
Treatment 2 (TZ) - T. harzianum (strain 1) spore suspension (10e spores/ml)
Treatment 3 (T3) = Gamma-irradiated T. harzianurn (strain 1) in Trichoseal@ base
(100 g/L) (control)
Treatment 4 (T4) = SDW (control) 99
5.3.5. Experiment 5.5. Pruning wound treatments
Recovery of E. lata was 1007o in the controls that were treated with non-viable T. harzianum or SDW while infection by E. Iata was reduced in canes that were treated with Z. harziqnum in both repetitions of the experiment (Table 5.15). Logistic regression was performed for the results of re-isolations of E. lata from wood 2 cm below the pruning wound to estimate the efficacy of the preventive treatments on infection by E. lata.In the first trial, there was no interaction between the effects of ffeatment and the day of introduction of E. lata (p=9.936). The three treatments containing viable ?' harzianumpropagules reduced (P<0.001) infection by E. Iata as shown in Table 5.16, irrespecrive of the day (Day 2 and Day 7) of introduction of E. lata (P=0.372) when compared to the control treatments of the first trial.
Analysis of data of the repeat trial of experiment 5 indicated that there were no interactions between the treatment and the day effects (P=0.877). Also, the difference in the response to infection by E lata when the pathogen was introduced on Day 2 andDay
7 was not significant (P=0.216).
Results for isolation of E. lata lrom 2-10 cm below the point of inoculation from
canes treated with the pathogen 2 days after the antagonist in the second trial are shown
in Figure 5.1. E. lata was isolated from 1007o of the canes treated with SDW, but from
less than 307o of the canes treated with the antagonist. The differences in the
probabilities of re-isolation of E. lata (for 2 cm below the point of inoculation) from
canes treated with viable antagonist and the control treatments (Table 5.16) were highly
significant (P<0.001), with the probabilities being 0.81, 0.81 and 0.94 respectively. 100
Re-isolation of T. harzianum was carried out only for the cuttings inoculated with the pathogen on day 2. Many of the wood samples (87.5Vo) treated with Trichoseal@ containing the 7 strains of Z. harzianum yielded T. harzianurn at a distances of up to and including 10 cm below point of inoculation, 12 weeks after the pruning wounds were treated (Figure 5.D. f. harzianunx strain 1 was re-isolated from canes treated with viable
T. harzianum at 2 to 10 cm below the point of inoculation but the frequency of recovery generally decreased with distance from the point of inoculation. T. harzianum was not recovered from any of the controls treated with gamma-irradiated T. harzianum strain I in Trichoseal@ base, gamma-irradiated Trichoseal@, or SDW' 101
Table 5.L5. Experiment 5.5: Number of canes, of eight, yielding E. lata aftet 12 weeks in the first and repeated trials. E. lata was introduced2 and Tdays after the antagonist and the canes were harvested 12 weeks later.
Treatment First trial Second trial E.løta M280 E.latø M280 E.lata M280 E.lata M280 inoculated inoculated inoculated inoculated on Day 2 on Day 7 on Day 2 on Day 7
rLU t ncnoseal 2 5 2 1
Gamma-irradiated 8 8 8 8
r@ I ncnoseal
T. harzianum J 4 2 1
strain 1 in
Trichoseal@ base
Gamma-irradiated 8 8 8 8 T. harzianum strain I in
Trichoseal@ base
Spore suspension 3 2 1 0 of T. harzianum
strain 1
SDW 8 8 8 8 r02
Table 5.16. Experiment 5.5: Predicted probabilities that canes were infected with E. lata irrespective of timing of inoculation with E. lata, as estimated by the reglession model.
E. lata was re-isolated,2 cm below point of inoculation. There were eight replicates per treatment. The canes were harvested 12 weeks after the treatments were applied to the pruning wounds.
Treatment Probabilities for first Probabilities for trial second trial Trichoseal 0.44b Q.n) 0.19" (0.10)
Gamma-irradiated Trichosealt 0.99^ (0.00) 1.ood (o.oo)
T. harzianum strain 1 in 0.44b Q.rz) 0.13' (0.08)
Trichoseal@ base
Gamma-irradiated T. 0.99" (0.00) o.g4d (0.06) harzianum strain 1in
Trichoseal@ base
Spore suspension of 7. 0.31b (0.12) 0.06" (0.06)
harzianum strain 1 o SDW 0.99" (0.00) 1.oo (o.oo)
Probabilities with the same letters are not significantly dffirent. Standard error values in parentheses. 103
Figure 5.L. Experiment 5.5: The extent of colonisation of canes by E. lata isolate M280 in trial 2. E. Iata was re-isolated from canes inoculated with the pathogen 2 days after the antagonist and harvested after 12 weeks.
100 880 G S^^.ou TT1 u¡ aT2 E trT3 r40 ElT4 o o Ë20 rt-
0 2cm 4cm 6cm 8cm 10cm Distance below point of inoculation
Treatment 1 (T1) - T. harzianum strain 1 in Trichoseal@ base
Treatment 2 (T2) = Spore suspension of T. harzianum sffain I
Treatment 3 (T3)= Trichoseal@
Treatment4G4)=SDW
There were eight replicates for each treatment. The Y-axis denotes the percentage of
canes that yielded E. lata. Recovery of E. lata from canes treated with the controls,
gamma-irradiated Trichoseal@ and gamma-irradiated T. harzianum strain 1 in
Trichoseal@ base (controls) are not included as the results were similar for SDW. 104
Figure 5.2. Experiment 5.5: Extent of colonisation of canes by T. harzianunrintnal2.T. harzianum strain lwas re-isolated from canes inoculated with the pathogen 2 days after the antagonist and harvested after 12 weeks.
E roo
ç¡ 'ñ 80 G F60 l11 q¡ È lr2 Ë40o trT3 E 320 0) Eoo É, 2cm 4cm 6cm 8cm 10 cm Distance below point of inoculation
Treatment 1(T1) = T' harzianum súain 1 in Trichoseal@ base
Treatment 2 (T2) = Spore suspension of T. harzianum strain I
Treatment 3 (T3) = Trichoseal@
There were eight replicates per treatment combination. The Y-axis denotes the percentage of canes that yielded T. harzianum. Canes treated with SDW, g¿unma-
irradiated T. harzianum strain 1 in Trichoseal@ base and gamma-irradiated Trichoseal@
(controls) are not included in the graph as T. harzianunl was not re-isolated from any of
the replicates. 105
5.3.6. Experiment 5.6. Colonisation of cuttings by pathogen in the presence of
antagonist
Canes that were treated with E. lata alone were colonised by the pathogen 5 cm above and.7 cm below the point of inoculation in 12 weeks. Colonisation of canes by E. lata was reduced to 0-3OEo in the presence of T. harzianum strain 1 (Figure 5.3). Canes that were treated with T. harzianum alone yielded T. hørzianum lrom 407o of wood segments
5.5 cm above and from 807o of the wood segments at 6.5 cm below the point of inoculation in 12 weeks (Figure 5.4). However, at 5.5 cm above the point of inoculation only 2O7o of the canes treated with E. lata alone were colonised by the pathogen. Also both fungi grew further below than above the point of inoculation (Figures 5.3 and 5'4).
The logistic regression performed on the results indicated that there was a significant difference between treatments (P<0.001). E. lata was recovered from all the canes treated with the pathogen alone and T. harzianum significantly reduced (P<0'001) infection by E. Iata when the fungi were applied to the same cane (Table 5.t7).
Table 5.L7. Experiment 5.6: Predicted probabilities of recovering E. lata as estimated by the logistic regression model. The cuttings were harvested 12 weeks after inoculation with T. harzianum. There were 10 replicates for each treatment.
Treatment Probability E. lata M280 + T. harzianum strain I (0.14)
E.Iata M280 + PDA plug 1 (0.00) PDA plug + T. harzianum strain I 0.00" (0.00)
Probabilities with the same letters are not significantly dffirent. Standard error vqlues in parentheses 106
Figure 5.3. Experiment 5.6: Isolation of E. lata from cuttings inoculated with isolate
M280 7 days after T. harzianum strain 1 and harvested 12 weeks later. Both fungi introduced as mycelial plugs into holes drilled 0.5 cm apart under the upper bud.
100 I 80 s60G r¡¡ lT1 o Elr2
þ¿oo o (¡) É, 20 I
0 5.0 3.0 lnoc. 3.0 below 5.0 below 7.0 below above above Point Distance from the point of E. Iata inoculation (cm)
Treatment 1 (T1) = T. harzianum strain I+ E.lata isolate M280
Treatment 2 (T2) = PDA plug + E.lata isolate M280
There were 10 replicates for each treatment. E. lata was not isolated from canes that were treated with T. harzianum and PDA plug (Treatment 3, control). to7
Figure 5.4. Experiment 5.6: Isolation of T. harzianum from cuttings inoculated with E. lata isolate M280 7 days after T. harzianum strain 1 and harvested 12 weeks later. Both fungi introduced as mycelial plugs into holes drilled 0.5 cm apart under the upper bud.
100 E^ S80 G cÈ60 lT1 ]- 40 TT3 o à o920 o É,0o 5.5 3.5 lnoc. 2.5 4.5 6.5 above above Point below below below Distances above and below the point of inoculation of T. harzianum (cm)
Treatment 1 (T1) - T. harzianum + E. lata
Treatment 3 (T3) - T. harzianurn + PDA plug
There were 10 replicates for each treatment. T. harzianurn was not isolated from canes
that were treated with E lata and PDA plugs (Treatment 2, control). 108
5.4. Discussion
The antagonists colonised the wood and survived in the tissues at least 10-20 weeks after inoculation of the wounds with spore suspension or mycelial plugs of F. lateritium and mycelial plugs of T. harzianum strain I (sections 5.3.1). However, colonisation of canes by T. harzianuru in most instances was low when the fungus was applied as spores
(section 5.3.1). The canes used in this experiment were not sterile and, therefore' may have carried both epiphytic and endophytic micro-organisms. Aspergillus sp., Alternaria alternata, Aureobasiclium pullulans, Candida SP.' Epicoccum purpurascens, F' lateritium, Penicillium spp., Phomopsis viticola, Rhodotorula sp. andT. harzianumhave been found in grapevine canes (Ferreira et aI., 1990; Munkvold and Marois, 1'993ai,
Chapius et al., 1998) whereas, Pseudomonas and Bacillus spp. are coÍlmon on grapevines (Ferreira et al., 1991; Munkvold and Marois, I993a). Such organisms may have inhibited germination of the spores of 7. harzianum. A highly concentrated inoculum has been identified as one of the factors necessary for effective biological control by Trichoderma spp. (Tronsmo and Dennis,7917; d'Enfert, 1997). The inoculum applied, 4 X lO4 spores of T. harzianum in the 40 ¡ll droplet, may have been insufficient to overcome competition from the resident microbiota. Another reason may have been that the volume of inoculum was too small. Further investigations, involving larger volumes and higher concentration of spores, are warranted.
T. harzianum was re-isolated from the discoloured wound tissue (section 5.3.1)' This
may be because T. harzianum is a saprophyte (Klein and Eveleigh, 1998) and, therefore,
is likely to be able to survive on the dead tissues of the pruning wounds. As effective 109 wound colonisers, Trichoderma spp. have been used in woody plants as wound protectants (see section 2.4.3.4).
Spores, unlike mycelia, require time to germinate before colonising the plant tissue'
Furthermore, mycelial inoculum in PDA plugs is less likely to desiccate and undergo nutrient stress compared to spores in 40 ¡tl SDW. Hence, the probability of T. harzianum establishing itself in the pruning wounds and tissues below in the grapevine cuttings was greater when the inoculum was applied as mycelial plug (section 5.3.1).
F. lateritiuffi, F. roseum and some strains of F. solani and F' oxysporurn, are saprophytes (Baker and Cook, 1974: Campbell, 1989). F. lateritium, when used as a wound treatment, has been effective against various pathogens, for example, Cytospora cincata, one of the causal organisms of canker and dieback in apricots (Rozsnyay et al.,
1991). F. lateritium was an effective antagonist of Nectria galligenø on wounds of apple shoots (Grabowski and Schwartz, 1997). Application of F. lateritium and other wound colonisers, such as T. viride, Phoma macrostomum, Cylindrocarpon magnusianum,
Cryptococcus laurentii and E. purpurascerzs, reduced necrosis in apple shoots due to pezicula alba, when the antagonists colonised the wounds before the pathogen. Various combinations of these fungi inhibited P. alba more effectively than did individual
isolates (Grabowski, lggg). Fusarium spp. are reported to be effective colonists of
pruning wounds on grapevines (Carter and Price, 1975; Ferreira et a1.,1990; Munkvold
and Marois, 1993a). F. lateritiu¡nz in this study colonised the living tissues up to 4 cm
above and below the point of inoculation and persisted for at least 10-20 weeks (section
5.3.1). Carter and Price (1974) re-isolated F. lateritiurn from wood up to 2 cm below the
pruned surface of apricot branches 6 months after inoculation in the field. However, 110 colonisation by F. Iateritium or T. harzianum of living tissue below pruning wounds has not been investigated in grapevines, therefore, further experiments are necessary to study this aspect in the field.
The failure of T. harzianum to reduce infection by E. lata in Experiment 5.2 may have been due to the distance (5 cm) between the points of inoculation of the antagonist and the pathogen, and lack of prior establishment of the antagonist (sections 5'2.2 and 5'3.2)'
Prior colonisation of plant tissues by Trichoderma spp. was necessary for the inhibition of infection by B. cinerea and Sclerotinia spp. (Whipps, 1987; Sutton and Peng, 1993)'
O'Neill et al. (1996) showed that application of T. harzianum to tomato stem pieces 8 h prior to the introduction of B. cinerea reduced the rate of infection and sporulation of the pathogen. The need to introduce T. harzianum in advance of the pathogen for biological control has been reported in other host-pathogen systems (Newhook, 1951;Wood, 1951;
Hong et al., 1998). For example, Hong et al. (1998) demonstrated that an isolate of
Rhodotorula sp. and three isolates of Trichoderma spp. gave good control of brown rot of plums when the antagonists were applied 12 h prior to the pathogen, Monilinia
for the fructicola. Other researchers have used the concept of allowing a time lag antagonist to colonise the host before introducing the pathogen to achieve successful biological control in various antagonist-pathogen systems (Scheffer, 1983; O'Brien,
1984; Murd och et al., 1984; Hall et at., 1986). Nonetheless, effective biological control
of B. cinerea in wounded tissues of lettuce and tomatoes has been demonstrated
following simultaneous inoculation of T. harzianum and B. cinerea, although the conidia
of T. harzianutn do not germinate before those of B. cinerea (Wood, 1951; O'Neill et
al., 1996). Hjeljord et at. (2001) suggested that antagonism in this case was due to an 111 inhibitory micro-environment at the site of inoculation as a result of depletion of oxygen by respiring conidia of T. harzianum.
In Experiment 5.3, the probability of re-isolating E. lata 8 months after inoculation was significantly higher than at 11 weeks (P=0.04). This observation implies thatT. hqrzianum and F. Iateritium had failed to grow up the wood towards the pruning wound and colonise the cane sufficiently to inhibit the pathogen. It is likely that E. lata had colonised more of cane that was used for re-isolation in 8 months than in 11 weeks, as it progressed slowly from the site of the wound into the woody tissues below and once it became established in the cane it could not be inhibited by T. harzianltm or F. Iateritium.
The fungitoxic metabolites produced by T. harzianum and F. lateritium may be responsible for inhibition of the E. lata in planta, This aspect of biological control in grapevines has not been researched in detail yet. The application of the antagonists (7. harzianum strain 1 and F. lateritium) in Experiment 5.3, 6.5-7.0 cm below the pathogen, was ineffective in reducing infection by E. Iata even when they were applied 7 days prior to the pathogen (sections 5.2.3 and 5.3.3). The titre of fungitoxic metabolites produced by the antagonists, at this distance, may have been too low to inhibit the growth of E. lata. O'Brien (1984) proposed a similar explanation for the failure of Streptomyc¿s spp' to control Dutch elm disease when stems of saplings of American elm were inoculated in the glasshouse.
Treatment with viable inoculum of T. harzianum did not reduce colonisation by E
lata when the pathogen and antagonist were introduced on the same day (section 5'3'4).
However, re-isolation of the pathogen was poor in the corresponding control treatment
comprising gamma-irradiated T. harzianurz (strain 1) in Trichoseal@ base (section 5.3.4). t12
It is possible that certain compounds formed during gamma-irradiation may have been toxic to the ascospores of E.lata and may have broken down within 2 days when applied to the surface of wounds. Further experiments are required to test this hypothesis'
Application of T. harzianum, both in the form of spore suspension and Trichoseal@ formulation, to pruning wounds reduced infection by E. lata in cuttings in the glasshouse when the antagonist was introduced2 or 7 days prior to the pathogen (sections 5.3-4 and
5.3.5). E. Iata was introduced to pruning wounds as mycelial plugs whereas T' harzianum was introduced as spores in the pruning wound experiments. T. harzianum spores require at least 14 h to complete germination (Lifshitz et a1.,1986;Hjeljotd et al''
2000). Hence, it is likely that T. harzianum was unable to reduce colonisationby E. lata when the pathogen was introduced on the same day, as the antagonist had not yet established in the cane tissues. There was no difference in the incidence of infection by
E.latawhen the pathogen was introduced2 or 7 days after the antagonist (sections 5.3.4 and 5.3.5). This suggests that 2 days was a sufficient interval fot T. harzianum to colonise the surface of the wound and underlying tissues of the l-year-old canes used in the pruning wound experiments. The nutrients supplied in the Trichoseal@ formulation did not appeff to enhance antagonism of E. lata by T. harzianum, since there was no significant difference between the response to spore suspensions of T. harzianum in
SDW and in Trichoseal@ base. In conffast, access to exogenous nutrients has been identified as one of several factors necessary for effective biological control by
Trichoderma isolates (Papavizas, 1985; Nelson et ql., 1988). The composition of
nutrients in the Trichoseal@ base is not known. The percentage and rate of germination,
hyphal extension and sporulation of Trichodermø isolates are reduced in nutrient-poor 113 conditions (Ko and Lockwood, 1967; Steiner and Lockwood, 1969; Martin and Nicolas,
1970; Danielson and Davey, 1973; Mitchell and Dix, 1975; Beagle-Ristaino and
Papavizas, 1985). However, the effect of nutrient availability on mechanisms of antagonism such as mycoparasitism and antibiosis is ambiguous. The availability of simple sugars or favoured substrates necessary for rapid growth of T. harzianum has been found to repress the enz)¡mes involved in mycoparasitism (Garcia et aL, 1994).
Nutrient stress may be a prerequisite for the secretion of certain cell wall-degrading enzymes (Lorito, 1998; Mach et aI., 1999). On the other hand, high C: N ratios have been reported to enhance antibiotic production in many Trichodermd spp. (Howell,
1998). Therefore, further investigations to study the effects of a range of nutrients on establishment and metabolite production of T. harZianum aÍe warranted.
In Experiment 5.5, T. harzianurn was recovered from more than 8O7o of the canes that were treated with Trichoseal@ (comprising 7 strains of T. harzianum) at 10 cm below the point of inoculation (Figure 5.2) whereas the recovery of the antagonist from cuttings treated with spore suspension of T. harzianum andT. harzianum strain 1 in Trichoseal@ base was less than 307o. This observation suggested that strain 1 may not be a good colonist of wood compared to the other six strains in Trichoseal@. Further trials are necessary to identify and incorporate strains that readily colonise wood as these may be useful in protecting grapevines from infection by E. Iata. Also, more research is required to optimise the method of delivery of the antagonist into grapevines as well as improve
the nutrient base to enhance rapid colonisation of wood.
In Experiment 5.6, infection by E. lata was significantly reduced (P<0.001) when the
pathogen was inserted into holes drilled 0.5 cm above the point of insertion of, and 7 II4 days after, T. harzianum strain 1 (section 5.3.6). It is likely that the antagonist grew up the wood, colonised the tissues, and thereby inhibited E. Iata when introduced 7 days later close to and above the position of inoculation of the antagonist.
T. harzianum and F. lateritium were isolated frequently from canes 1-4 cm above the point of inoculation 20 weeks after inoculation (section 5.3.1). This observation implies that inoculation of the antagonists into the trunks may be able to protect vines from infection of pruning wounds by E. Iata, and this may be useful in developing prophylactic treatments other than pruning wound applications of the antagonists.
In summary, T. harzianum colonised not only the wounds, but also the living tissues of the grapevine cuttings (sections 5.3.1,5.3.5 and 5.3.6). Application of T' harzianumto pruning wounds 2 days prior to the pathogen, inhibited infection by E. lata applied as mycelial plugs (5.3.4 and 5.3.5). T. harzianum inhibited infection by E. Iata in cuttings when applied as mycelial plugs close to the inoculation point of E. lata (section 5.3'6).
The results obtained in these glasshouse experiments suggested that T. harzianun't may be able to inhibit colonisation of grapevines by E. lata' r15
CHAPTER 6. CONTROL OF EUTYPA DIEBACK IN THE FIELD
6.1. Introduction
Biological control of E. lata in controlled conditions has been demonstrated using various micro-organisms (Vajna, 1986; Chapius et al., 1998; Schmidt et aI',200La)'
However, only a few reports on successful biological control of eutypa dieback of grapevine in field conditions are available to date (Ferreira et al., 1991; Munkvold and
Marois, I993a). Carter and Price (1974) demonstrated control of eutypa dieback of apricots using F. lateritium, and this fungus has also been reported to control dieback of grapevines (Munkvold and Marois, 1993a). Recent results of bioassays with gtapevine segments suggest that F. lateritium, either alone or in combination with benomyl, may provide good control of eutypa dieback of grapevines (McMahan et a1.,2001).
Generally, only a small proportion (57o) of potential biocontrol agents selected from screening tests in the laboratory and glasshouse have been capable of controlling the disease in rapidly fluctuating field conditions (Powell and Faull, 1991). Trichoderma spp., however, have been reported to be effective in the field against a wide range of diseases affecting economically important crops (Papavizas, 1985; Tronsmo, 1986; Chet,
1987). This may be attributed to the different mechanisms of antagonism exhibited by the various isolates of TrichocJerma (Chet et al., 1998; Hjeljord and Tronsmo, 1998;
Tronsmo and Hjeljord, 1998). The fast growth rates of Trichoderma species and their
general ability to tolerate a wide range of environmental conditions help the fungus to
persist in the field (Tronsmo and Hjeljord, 1998). These characteristics also allow Lt6
Trichodermø spp. to compete with other fungi and to survive in different ecological niches
As laboratory and glasshouse-based investigations showed T. harzianum to have potential in the biological control of E. lata, experiments were carried out in the field to ascertain the efficacy ofthis organism in natural conditions.
pruning wounds on grapevines serve as entry points to E.lata ascospores (Moller and
Kasimatis, 1978). Hence, pruning wound applications involving commercial products
(see section 3.3.2) such as Trichoseal ,pruytor Vinevax@, as well as spore suspensions of T. harzianum strun 1, were tested. An isolate of F. lateritium (Carter and Price, 1974) was also tested in these pruning wound trials for the sake of comparison. Control of E.
Iataby trunk injections of T. harzianurz as Trichoject@ (see section 3.3.2) was tested in the field. Furthermore, the ability of T. harzianum to persist and grow within the vines in the field, when introduced into trunks of vines in the form of Trichodowels@ (see section
3.3.Z),was investigated in this study. The benefit envisaged in using the Trichodowels@ was the protection of vines from infection by E. lata over a long period of time. Such long-term protection would be unattainable by chemical treatment. lr7
6.2. Materials and methods
6.2.1. Pruning wound trials
6.2.1.1. Nuriootpø 7
This trial, which comprised a completely randomized design, was carried out in early
August 2000 in Nuriootpa, where 25 healthy vines (cultivar Cabernet Sauvignon, see
Table 3.1) were selected and ten l-year-old canes per vine were pruned to the two-bud level through the third node. The following treatments were painted on to the wounds
(spurs), which were 10-20 mm in diameter, using 25 mm flat varnish brushes, within an hour of pruning. Each of the five treatments was randomly assigned to five vines (i'e. there were 50 spurs per treatment). Spore suspensions of the antagonists were prepared as described in section 3.3.1. and were transported to the field in polycarbonate tubs
(Magenta@ GA7). Each treatment was applied with a separate brush to avoid cross- contamination between treatments.
Treatment 1 (T1) = T. harzianum (strain 1), spore suspension (10e spores/ml SDW)
Treatment 2 (T2) = F. lateritium, spote suspension (106 spores/ml SDW)
Treatment 3 (T3) = Trichoseal sp.ay@ (100 g/10 L)
Treatment 4 (T4) = RO water
Treatment 5 (T5) = RO water - no E latainoculum
E. lqta ascospores were applied to half the number of pruning wounds on each vine
(i.e. five wounds per vine, 125 wounds in total) 1 day after the above ffeatments were
applied, while the other half were inoculated with E. lata ascospores on Day 14. Each 118 pruning wound received a IO ¡tl droplet containing 500 ascospores. Wounds designated
T5 did not receive the ascosporic inoculum. After inoculation, the spurs were lightly
misted with RO water. The spurs were harvested 11 months after the treatments
containing the antagonists were applied and re-isolation of E. lata was carried out as
described in section 3.4.1., except that instead of using 2.57o NaOCl, the bark was
removed and the spurs were dipped in 7O7o ethanol and flamed. As this trial and those
described in sections 6.2.1.2. and 6.2.I.3 were carried out in collaboration with Dr Mette
Creaser, the protocol for surface sterilisation was modified to match her standard
protocol to maintain uniformity.
6.2.1.2. Nuríootpa 2
This trial (see Table 3.1) was similar in all aspects to that described in section 6.2.1.1
except that it was carried out in mid-August 2000.
6.2.1.3. Eden Valley
A third repetition of the trial (see Table 3.1) was conducted in Eden Valley in early
August 2000. It was similar to the trials described above except that cultivar Shiraz was
used and there were six vines (30 vines in total) and 60 spurs per treatment. tt9
6.2.1.4. Warripøringa 1
The trial was carried out in late August 2001 at the Waniparinga site (see Table 2.1), in a completely randomized design with some improvements to the methods used in the trials described above. Pruning of l-year-old canes was performed on 25 vines (cultivar
Rondella, see Table 3.1) with 20 healthy canes per vine selected for this purpose' yielding 500 wounds. The canes were treated as described in section 6'2'I.1, with each of the five treatments being randomly assigned to five vines. Half the number of wounds on each vine (250 wounds in total) was inoculated with E. lata ascospores on Day 1, while the other 250 wounds were inoculated on Day 14.
The treated spurs ,were lightly misted with SDW before being inoculated with a 25 ¡t"l droplet containing 500 ascospores. The spurs were harvested 9 months after ffeatment with the antagonists and re-isolation of E. Iata was carried out exactly as described in section 3.4.I., using NaOCl.
6.2. I. 5. Warrip øringa 2
The trial described in section 6.2.t.4. was repeated in early July 2002, when there was no sap flow in the plants. There were 25 vines (cultivar Palomino, see Table 3.1), but half the number of pruning wounds (i.e. 250 wounds in total). Instead of Trichoseal spray@, the replacement product Vinevax@ *us used as Treatment 3 (T3) (see Table 3.3).
The spurs were harvested 10 weeks after the treatments were administered to the canes.
Re-isolations were carried out as described in section 3.4.1. 120
6.2.2. Trichodowel trial
Holes were drilled into the trunks of 14 vines (cultivar Nyora, see Table 3.1) in July
2000, using 5 mm diam. drill bits at 30 cm above ground level, so that each vine had two holes (5 mm diam.) side by side at this height. Trichodowels@ were inserted (two per vine) into seven of the vines, while gamma-irradiated Trichodowels@ were inserted into the remaining vines, as controls. The dowel treatments were randomly assigned to the vines. Wood samples were collected every 4 months for a period of 20 months, by drilling into the trunks of these vines at different positions above the point of inoculation
(2-3 mm, 3 cm and 6 cm on the first harvest, and 3 cm, 6 cm, 18 cm and the crown of the vine at subsequent harvests). The crown was defined as the central point at which the trunk separated into cordons on either side. The drill bit was sterilised by spraying with
J07o ethanol and wiping with Kleenex@ facial tissues between samples to avoid cross- contamination. The trunk also was sprayed with 70Vo ethanol and wiped in the same way to disinfect the bark. The wood shavings were collected into polycarbonate tubs
(Magenta@ GA7) and transported to the laboratory where they were stored at 5oC for up to 3 days. The shavings were plated aseptically (see section 3.4.2) on APDA. Each sample was plated in clumps of 10 (five clumps per plate), incubated at23"C in darkness
for 3-4 days and observed for growth of T. harzianum, which was identified by colony morphology as well as spore and hyphal characteristics under the microscope (Leitz
Orthoplan 871288).
Holes (5 mm diam.) were drilled 3.5 cm above the dowel insertion point and
inoculated with 5 mm plugs of E. lata strain M280 (see section 3.2.I) or plugs of sterile
PDA of the same dimensions, 16 months after the vines rwere treated with L2L
Trichodowels@, i.e. in Novemb er 200I. Four of the Trichodowel@-treated vines received plugs of M280 and three received plugs of PDA as the control treatment. Similarly, four
of the vines treated with gamma-irradiated Trichodowels@ received plugs of M280 while
three of the same received sterile PDA plugs. The inoculation points were covered with
parafilm@. Fourteen weeks after inoculation with E. Iata, holes (5 mm diam.) were
drilled into the trunks just above the point of insertion of the M280 or PDA plug' The
wood shavings were washed thrice with SDW, plated out on EUSM in clumps of 10 per
sample (five clumps per plate) , and E. lata was re-isolated as described in section 3.4.1'
Furthermore, 16 months after the vines were treated with Trichodowels@, five l-year-
old canes per vine \üere pruned to the two-bud level through the nodes and the spurs
were lightly misted with SDW. Each pruned spur on vines that had been treated with
the spurs plugs of M280 received a25 ¡tl droplet containing 500 viable ascospores, while
on vines treated with the plugs of PDA receiv ed a 25 ¡rl droplet of SDW. The wounds
were covered with Parafilm@ after the spurs had absorbed the droplets. The canes were
harvested 14 weeks after inoculation with E. lata and re-isolation of E. Iata was carried
out as described in section 3.4.1.
6.2.3.Injection trial
Holes were drilled (6 mm diam.) into the trunks of 22 vines (cultivars Ribier and
Exotic, see Table 3.1) at 60 cm above ground level. The vines were injected with 10 ml
of Trichoject@ 115 vines) or SDW (seven vines), using Chemjet@ plastic spring-loaded
syringes provided by Agrimm Technologies Ltd. The holes in the ffunks of 15 vines,
which had been injected with Trichoject@, were sealed with Trichodowels@ as t22 recoÍìmended by the manufacturer. The holes in the seven vines treated with SDW were sealed with autoclaved wooden dowels. One day after the treatments were injected into the vines, five l-year-old canes on each vine were pruned and the spurs were lightly misted with SDV/. The pruning wounds (spurs) on seven of the 15 Trichoject@-treated vines were then inoculated with 5 mm diam. plugs of E. lata strain M280 while the wounds on the other eight vines treated with Trichoject@ were inoculated with 5 mm diam. pDA plugs. The plugs were covered with Parafilm@ to avoid desiccation' The spurs on the seven vines which had been injected with SDW, were also inoculated with plugs of M280. The same procedure was carried out on day 14 after pruning another five canes on each of the vines. The vines were observed for symptoms of eutypa dieback in the following spring, 9 months after the treatments were injected into the vines. The spurs were harvested 15 months after the application of Trichoject@ and re-isolation of E lata was carried out as described in section 3.4.I. There was a loss of 337o of the inoculated spurs due to vandalism.
6.3. Results
6.3.1. Nuriootpa 1,2 and Eden Valley
The three pruning wound trials carried out at Nuriootpa and Eden Valley in the year
2000 showed variable results. From the logistic regression performed for the trial
Nuriootpa 1, there was no interaction between the treatment effect and the day effect
(p=0.679), indicating that the vines receiving the treatments did not respond differently if
inoculated with the pathogen on day 1 or day 14. Both the main effect of the treatment 123
(P<0.001) and the main effect of day (P=0.014) were significant at the 57o level. The treatments comprising the antagonists (T1, T2,T3) did not significantly (Þ0.05) reduce infection of the wounds by E. lata (Figure 6.1). The recovery of E. lata from spurs which received neither pathogen nor antagonist (T5) indicated that there had been some natural inoculum in the vineyard. The incidence of infection was lower for spurs inoculated with the pathogen 14 days (P=0.014) after the antagonists (or controls) than those inoculated 1 day after the antagonists, but not significantly so'
The logistic regression for Nuriootp a 2 indicated no interaction between the treatment and the day effects (P=0.648). The two treatments with the spore suspensions of the antagonists (T1 and T2) significantly reduced infection by E. lata (P<0.001) when compared to the control treatment (T4), while the Trichoseal spray@ did not, in the conditions imposed in this experiment (Figure 6.2). Also, the difference between the two days (Day 1 and Day 14) was not statistically significant (P=0.065), although the trend was similar to that exhibited in Nuriootpa 1. Infection due to natural inoculum was observed in this trial too.
The Eden Valley trial yielded results generally similar to the above two trials. There was no interaction between treatment and day effects (P=0.002), while both the main effect of treatment (P=0.002) and that of the day (P=0.005) were significant at the 57o level. The overall incidence of infection was low, but the spore suspensions with the
anragonists (T1 and T2) significantly reduced infection (P=0.002) by E' lata when
compared with the control treatment (T4) (Figure 6.3). Trichoseal tp.ayt had no effect
on the incidence of infection. The percentage of infection, by E. lata, of spurs treated on 124
Day 1 was significantly greater (P=0.005) than that on Day 14. There was no evidence that natural infection occurred in this vineyard.
6.3.2. Warriparinga l, 2
These two trials were carried out after the completion of the three performed in 2000, with some modifications to increase the incidence of infection by E. lata of spurs that received the inoculated control treatment (T4)'
Logistic regfession performed for the trial Warriparinga 1 yielded no interactions between the treatment and day effects (P=0.318). The main effects of day (P<0.001) and treatment (P<0.001) were significant at the 57o level. In this experiment, all three treatments with the antagonists (T1, T2 and T3), including Trichoseal spray@, significantly reduced colonisation (P <0.05) by the pathogen (Figure 6.4)' This was not so in the pruning wound trials of 2000. The percentage of infection on Day 14 was significantly lower (P<0.001) than that of Day 1. Furthermore, none of the spurs which received T5 were colonised by E. lata.
The subsequent trial, Waniparinga 2, was carried out early in the winter of 2002, when there was no sap flow in the vines. E. lata was re-isolated from 777o of the spurs which received T4 (Figure 6.5). There was no interaction between treatment and day
(P=0.428) at the 5Tolevel when logistic regression was performed. The three treatments
(Tl, T2 and T3) significantly reduced colonisation (P<0.001) compared to the control
treatment (T4). Also, colonisation by E. lata onDay 14 was low (P<0.001) compared to
Day 1. E. lata was not re-isolated from any of the spurs which received T5 in2O02. t25
Figure 6.1. Percentage of spurs yielding E. lata when re-isolations \ryere carried out in the field trial, Nuriootpa 1, established in August 2000.
Nuriootpa 1
70 guo s50 T G tr¡ 40 tr Dayl o I Dayl 4 oà30 ãro tr10o
0 T1 T2 T3 T4 T5 Treatment
There were five vines per treatment with 10 wounds per vine and the pathogen was applied, as 500 ascospores in 10 ¡ll SDW, I or 14 days after the antagonist (or water):
Treatment I (T1) = T. harzianum sttain 1 (10e spores/ml SDSD + E. lata
Treatment 2 (T2) = F.lateritium (lO6 spores/ml SDSD + E.lata
Treatment 3 (T3) = Trichoseal sptuy@ (100 g/10 L) + E.lata
Treatment 4 (T4) = RO water + E.lata
Treatment 5 (T5) = RO water only - (no E latainocuhtm)
There were 25 replicates per treatment combination. Bars denote standard effors. The Y-
axis denotes the proportion of spurs, of 25, from which E.lata was re-isolated' 126
Figure 6.2. Percentage of spurs yielding E. lata in the field trial, Nuriootpa 2, established in August 2000.
Nuriootpa 2
40 ^35I 3o i'$zs u¡ I Dayl :o20 lDay14 ãrs
o8ro É, 5 0 T1 T2 T3 T4 T5 Treatment
There were five vines per treatment with 10 wounds per vine and the pathogen was applied, as 500 ascospores in 10 ¡rl SDW, I or 14 days after the antagonist (or water):
Treatment 1 (T1) - T. harzianum strain 1 (10e spores/ml SDS/) + E. Iata
Treatment 2 (T2) = F. lateritium (106 spores/ml SDUD + E. lata
Treatment 3 (T3) = Trichoseal spray@ (100 g/10 L) + E.Iata
Treatment 4 (T4) = RO water + E.lata
Treatment 5 (T5) = RO water only - (no E lata inocuhtm)
There were 25 replicates per treatment combination. Bars denote standard errors. The Y-
axis denotes the proportion of spurs, o125, from which E. lata was re-isolated. r2'7
Figure 6.3. Percentage of spurs yielding E. lata in the Eden Valley field trial, established in August 2000.
Eden Valley
30 gru Ëro tr¡ lDayl b15 I tr Day14
å,0o Éso
0 T1 T2 T3 T4 T5 Treatment
There were six vines per treatment with 10 wounds per vine and the pathogen was applied, as 500 ascospores in 10 ¡ll SDW, 1 or 14 days after the antagonist (or water):
Treatment 1 (Tl) - T. harzianum strain 1 (10e spores/ml SDUD + E. lata
Treatment 2 (T2) = F. lateritium (106 spores/ml SDW) + E. lata
Treatment 3 (T3) = Trichoseal sptay@ (100 g/10 L) + E.lata
Treatment 4 (T4) = RO water + E.Iata
Treatment 5 (T5) = RO water only - (no E. latainoculum)
There were 30 replicates per treatment combination. Bars denote standard errors. The Y-
axis denotes the proportion of spurs, of 30, from which E. lata was re-isolated' r28
Figure 6.4. Percentage of spurs yielding E. lata in the field trial, Warriparinga 1, established in August 2001.
Warriparinga 1
50 45 8¿o s35 lÙ 30 uj I Dayl b25 tr Day14 ãzo ()ã15 Ë10 5 0 T1 T2 T3 T4 T5 Treatment
There were five vines per treatment with 20 wounds per vine and the pathogen was applied, as 500 ascospores in25 ¡.I"LSDW, I or 14 days after the antagonist (or water):
Treatment 1 (Tl) = T. harzianum strain 1 (10e spores/ml SDUD + E. lata
Treatment 2 (T2) = F.laterütium (106 spores/ml SDSD + E. lata
Treatment 3 (T3) = Trichoseal spray@ (100 g/10 L) + E.lata
Treatment 4 (T4) = SDW water + E. lata
Treatment 5 (T5) = SDW only - (no E. latainoculum)
There were 50 replicates per treatment combination. Bars denote standard effors. The Y-
axis denotes the proportion of spurs, of 50, from which E. Iata was re-isolated. t29
Figure 6.5. Percentage of spurs yieldiîg E. lata in the field trial, Warripatinga 2, established in July 2002.
Warriparinga 2
90 80 Ð70rO Ëuo uì 50 I Dayl Day14 Boo E o930 ?, 20 É, 10 0 T1 Í2 T3 T4 T5 Treatment
There were five vines per treatment with 10 wounds per vine and the pathogen was applied, as 500 ascospores in25 ¡tl SDW, I or 14 days after the antagonist (or water):
Treatment 1 (T1) = T. harzianum sttain 1 (10e spores/ml SDW) + E' Iata
Treatment 2 (T2) = F.lateritium (106 spores/ml SDUD + E' lata
Treatment 3 (T3) = Vinevax@ (100 g/10 L) + E.lata
Treatment 4 (T4) = SDW water + E.lata
Treatment 5 (T5) = SDW only - no E' lata inoculum
There were 25 replicates per treatment combination. Bars denote standard erors. The Y-
axis denotes the proportion of spurs, of 25, from which E. lata was re-isolated. 130
6.3.3. Results of Trichodowel trial
Trichodermd spp. were not isolated from woody tissues taken from any of the control vines. T. harzianum was re-isolated from all the vines, 4 months after they had been inoculated with Trichodowels@, at a distance of 2-3 mm and 3 cm above the point of insertion of Trichodowels@ , whereas at a distance of 6 cm, T. harzianum was re-isolated from only six of the seven vines (see Table 6.1). In 8 months T. harzianum had migrated a distance of 18 cm in one of the seven vines and in 12 months it was re-isolated from two of the vines at this particular distance. Moreover, the antagonist was re-isolated from most of the vines (6/7-717) at3-6 cm above the dowel insertion point over the first 12 months of the experiment. After 20 months, T. harziønum was re-isolated at 3 cm from the inoculation site on three of the vines. T. harzianum had spread to the crown of one vine in 20 months. All the vines appeared to be healthy throughout the trial.
Infection resulting from inoculation with mycelial plugs of E. lata was significantly less (P<0.072by Fisher's exact probability test) in the Trichodowel@ -treated vines than in the control vines which had been treated with gamma-irradiated Trichodowels@ lTable
6.2). None of the vines that had received Trichodowels@ prior to inoculation with the pathogen yielded E. lata. Also, E. lata was not re-isolated from the control vines that were treated with plugs of PDA.
E. lata was re-isolated from only one of the 20 canes that were pruned and the cut
surface treated with ascospores of E. lata (Table 6.3), while none of the 20 canes that
were inoculated with ascospores of E. Iata following treatment with sterile dowels
yielded E. lata. The difference between these two treatments was not significant. 131
Furthermore, E. lata was not recovered from any of the canes that had received sterile water (controls).
6,3.4. Results of injection trial
When the vines were assessed for symptoms 9 months after inoculation, none showed any symptoms of dieback. E. lata was recovered from 52-607o of spurs injected with
SDW before inoculation (Figure 6.6). The Trichoject@ treatment did not reduce
(P=0.288) infection of the spurs by E. lata in comparison to those on vines injected with
SDW. Also, there was no difference (P=0.28) in the percentages of spurs infected between the two different days on which E. lata was applied (Day 1 and 14 after
Trichoject@). E. Iata was not isolated from controls treated with Trichoject@ and PDA plugs.
6.4. Discussion
E. lata,was recovered less frequently from spurs that received the E. latainoc:ulum on
day 14 than those inoculated on day I in three of the five pruning wound trials and a
similar trend was observed in the other two. Many authors have reported increase in
efficacy of biological control agents with time. Carter (1983) achieved effective
biological control of E. lata on apricots by introducing F. Iateritium to pruning wounds 6
days prior to the application of the pathogen. Likewise, Munkvold and Marois (1993a)
reported effective control using Cladosporium herbarum and F. lateritium on grapevines
when application of ascospores of E. latawas delayed for 14 days after application of the
antagonists to pruning wounds. t32
Table 6.1. Number of vines (cultivar Nyora) from which T. harzianum was recovered when sampled every 4 months at various distances above the point of insertion of
Trichodowels@ (seven replicate vines per treatment).
Time after Distance above Number of vines from which inoculation dowel inoculation T. harzianurn was re-isolated point Trichodowel@ Gamma -treated vines -irradiated Trichodowel@- treated vines
4th Month 2-3 mm 7 0 3cm 7 0 6cm 6 0
18 cm Not tested Not tested Top Not tested Not tested
8th Month 3cm 6 0 6cm 7 0
18 cm 1 0 Top 0 0
12th Month 3cm 7 0 6cm 6 0
18 cm 2 0 Top 0 0
16th Month 3cm 4 0 6cm 2 0
18 cm 2 0 Top 0 0
20th Month 3cm J 0 6cm 4 0
18 cm 2 0
Top 1 0 133
Table 6.2. Proportion of vines (cultivar Nyora) from which E. lata was re-isolated when wood samples were collected 14 weeks after the trunks were inoculated with 5 mm plugs of E. lata or sterile PDA (controls). The trunks were treated with Trichodowels@ or sterile dowels 16 months prior to the introduction of the pathogen.
Introduction of Treatment of trunks with Proportion of vines
antagonist into the plugs of E. luta or sterile yielding E.lata 14 weeks
trunks in July 2000 PDA in November 200L after inoculation
r r(8) I rrcnooowel Isolate M280 014
(9 Trichodowel PDA plug 013
Gamma-irradiated Isolate M280 3t4
Trichodowel@ Si gnificant at (P <0.O7 2)
Gamma-irradiated PDA plug ot3
Trichodowel@ 134
Table 6.3. Proportion of l-year-old canes yielding E. lata 14 weeks after they were pruned and inoculated with ascospores of E. Iata or SDW. The trunks were treated with
Trichodowels@ or sterile dowels 16 months prior to the introduction of the pathogen.
Introduction of Treatments applied 16 months after Proportion of canes
antagonist into dowels in November 2001 yielding E.lata
the trunks in after 14 weeks
Juty 2000 E.lata or sterile E.lata or SDW to
PDA into trunks pruning wounds
Trichodowel G) Isolate M280 Ascospores U20
Trichodowel (9 PDA plug Sterile water 0/15
Sterile dowel Isolate M280 Ascospores 0/20
Sterile dowel PDA plug Sterile water 0/15 135
Figure 6.6. Percentage of spurs yielding E. lata from vines injected with (cultivars
Ribier and Exotic) Trichoject@ or SDW (controls).
90 EBo *zo i60 T bs0 IDay 1 140 T lDay14 I å30 820 É, 10 0 T1 r2 T3 Treatment
There were 22 vines with 10 wounds per vine (five wounds were made on Day 1 and five on Day 14) and the pathogen was applied 1 or 14 days after the antagonist (or water):
Treatment 1(T1) = Trichoject@ + E.lata M280 plug --+ administered to 7 vines
Treatment 2(T2) = SDW + E.lata M280 plug --+ administered to 7 vines
Treatment 3(T3) = Trichoject@ + PDA plug ---+ administered to 8 vines
Tlrere were 35 r'eplicate wounds for T1 and T3 and 40 replicates for T2. Bars denote
standard effors. The Y-axis denotes the proportion of spurs from which E. lata was re-
isolated from each sample of 35 or 40 replicates receiving each treatment combination. 136
The antagonists require time to colonise the wound sites before they are able to impair colonisation by the pathogen. Also, wounds undergo healing as they age, and this renders them less susceptible to the pathogen. This inverse relationship between the age of the pruning wounds and susceptibility to infection by canker pathogens has been reported for many cfops, such as almond, apple, apricots, grape and peach (Carter and Moller,l97O;
Ramos et al., I975a;Kile, 1916; Petzoldt et al., 1981; Wicks et al., 1983; Biggs and
Miles, 1988; Doster and Bostock, 1988; Biggs, 1989).
In the four trials performed in late winter, low incidence of infection by E. lata (less than 507o) was observed in the spurs that received the control treatment of ascospores of
E. lata alone. In the second trial at Wa:riparinga,777o of the spurs that received E. lata alone (inoculated controls) yielded the pathogen. This was the only trial that was performed in early winter. Sap flow was evident in the vines treated in the late winter trials, and the fluid exuding from the xylem vessels at the wounds soon after pruning may have caused physical flushing of the ascospores. This would have impeded the entry of ascospores into vessels and establishment of E. lata in the tissues below the surface of the wound to some extent (Munkvold and Marois, 1995). Also, the increase in metabolic activity of the vines with the rising temperatures towards the end of the dormant season, may have accelerated the host wound response. This may involve necrosis of parenchyma cells, accumulation of phenolic compounds (Hart and Shrimpton, 1979;
Shain, 1979;Biggs, 1987), the formation of papillae, lignituber, tyloses, corklayers and
deposition of gums (Agrios, 1997). Temperature is known to affect the accumulation of
phenolics in wounded woody plants. Low temperatures after wounding have been linked
to longer duration of susceptibility to wound pathogens in woody plants (Biggs, 1986a; r37
Doster and Bostock, 1988). Furthermore, the xylem vessels in the host plant may become clogged with tyloses and polysaccharides (Kile, 1976; Peatce and Rutherford,
1981; Pearce and Holloway, 1984; Biggs, 1987). The desiccation of the pruning wound
tissues is yet another wound response (Shain, 1979). However, the critical moisture
content that is required for the inhibition of E. Iata is not yet known. The decline in
wound susceptibility to E. lata towards the end of dormancy has been reported on
apricots, by Ramos et aL (I975a), and on grapevines by various researchers in California
and France (Petzoldt et al., I98l; Trese et al., 1982; Chapíts et al., 1998).
In these experiments, pruning wounds were exposed to natural inoculum in addition
to artificial inoculum. Infection due to natural inoculum on the spurs that were sampled
was lower than in the control treatments in Nuriootpa 1 and Nuriootpa 2, while it was
absent in the other three trials. Failure of natural inoculum to reach the infection courts
when the wounds were susceptible to infection may have been one of the reasons for this
observation (Ramos et al., I975b; Petzoldt et a1.,1981; Wicks et al., 1983; Biggs and
Miles, 1988; Biggs, 1989). Naturally occurring ascospores which may have landed on
the susceptible pruning wounds may have failed to germinate in sub-optimal conditions
(Carter, 1991). Also, the host wound response may have inhibited the establishment of
any germlings (Kile, 1976: Hart and Shrimpton, 1919; Shain, I9l9; Pearce and
Rutherford, 1981; Pearce and Holloway, L984; Biggs, 1987)'
Treatment of pruning wounds with spores of T. harzienum and F. lateritium resulted
in reduced frequency of re-isolation of E. lata compared to the inoculated control
treatment in four of the five trials. The findings of this study, therefore support previous
repofts on biological control of various pathogens of tree crops using F. lateritium as 138 wound treatments. (see section 5.4). Rapid colonisation of the pruning wounds by ?. harzianum and F. lateritium may have been the reason for effective control of E. Iata.
Fresh pruning wounds are rich in nutrients (Anderson and Brodbeck, 1989). Since these two antagonists arrived at the wounds before E. Iata, they had the potential to preempt colonisation by the pathogen and possibly inhibited it by competing with it for nutrition, space and moisture (see section 2.4.3.3).
A dosage of 106 macroconidia per ml of F. Iateritium painted on apricot wounds gave substantial protection from infection by E. lata in Australia (Carter and Price, 1974).
This dose of 106 macroconidia/ml of F. Iateritium was equally effective on the grapevine wounds in the present study. On the contrary, Gendloff et al. (1983) were not successful in reducing infection of grapevines by E. lata using an Australian strain of F. Iateritium in USA. This failure was attributed to the low inoculum dosage (1,000 macroconidia per wound) and the inability of the Australian strain of F. lateritium to adapt to the cold conditions in Michigan where the trials were carried out. Munkvold and Marois (I993a) demonstrated reduction in infection by E. lata by painting grapevine wounds with 108 propagules per ml of T. viricle or F. lateritium, w\th the latter being more effective' The concentration of 10e spores/ml of T. harzianum used in this study was possibly sufficient to have filled the exposed vessels of the grapevine pruning wounds so that the probability of a single ascospore of E. lata entering a vessel unoccupied by the
antagonist was reduced. Hence, establishment of E. lata in the spurs treated with T.
harzianum was low in most of the trials. Nuriootpa trial 1 was the exception, in which
treatment of pruning wounds with spores of F. lateritium and T. harzianum wàs not as
effective as in the other four trials. This exception may have reflected deficiencies in r39 technique, as it was the first trial conducted. Alternatively, it may be that T. harz.ianum and F. lateritium colonised the wounds poorly due to the presence of naturally occurring wound colonisers in these vines. At the end of the dormant season there were xylem exudates, which contain carbohydrates, amino acids and organic acids, from the pruning wounds that may have promoted rapid growth of wound microflora which, in turn, may have competed with the antagonists and impaired their ability to reduce infection by E.
Iata (Anderson and Brodbeck, 1989; Munkvold and Marois, 1993a; Chapius et al',
1998). Nonpathogenic wound colonisers may also have contributed to the overall reduced incidence of infection by E. lata in the four trials that were conducted in late winter. The possible influence of the wound response was discussed above'
Trichoseal spray@ did not reduce infection by E. lata when compared to the inoculated control treatment in the trials performed in the year 2000. The viability of the batch of product used in the collaborative trial in year 2000 was not tested. It is possible that the batch of Trichoseal ,pruyt used in year 2000 had reduced viability, perhaps due to conditions during transit from New Zealand.
Infection by E. lata was less in spurs treated with spore suspension of T. harzianum strain 1 in SDW than those treated with Trichoseal spray@ in all the pruning wound trials except in Warriparinga 2, which was carried out in early winter. T. harzianum isolates require 14 to 18 hours to complete germination and initiate mycelial extension even in
optimal conditions (Lifshitz et a1.,1986; Hjelj ord et at.,2000). The nutrient base used in
the product may have caused unintended stimulation of the wound microflora which may
have been present during the end of winter or even stimulated the pathogen itself before
the T. harzianum strains in the Trichoseal ,pruyt formulation were able to germinate 140
(Kelley, 1976; Harman et al., 1981; Washington et al., 1999). This may have contributed to the difference in the performance of Trichoseal spray@ and spore suspension of T. harzianum in the pruning wound trials performed at the end of the dormant season.
Some of the conditions imposed in these experiments may have influenced the overall incidence of infection by E. Iata. For example, the concentrations of inoculum used in the pruning wound trials (500 ascospores/lO ¡rl and 500 ascospores/Z5 ¡t'l) wete artificially high. These concentrations were used to enhance infection of spurs by E' lata and to facilitate the detection of differences between the effects of treatments. The droplet size was increased from 10 to25 ¡tl in the trials conducted in 2001 and 2OO2to provide additional moisture for germination of ascospores. The inoculum of 500 ascospores per droplet was less than the concentrations used previously by other researchers, however, it yielded detectable rates of colonisation (Petzoldt et al., l98li
Munkvold and Marois, 1993a; Munkvold and Marois, 1993b; Chapius, et al. 1998)' ln contrast, Carter and Moller (1971) estimated that 10-100 ascospores per wound were sufficient, and suggested that the natural deposition of ascospores in field conditions was unlikely to exceed 10 per wound (Carter, 1991). Hence, anything above this range was considered an unrealistic overload. Ramos et al. (1975a) showed that a single ascospore of E. lata was adequate to cause infection and obtained a remarkable 7l7o colonisation following inoculation of single ascospores to fresh pruning wounds on apricots.
Therefore, future investigations of the ability of T. harzianum to protect pruning wounds
should involve smaller amounts of inoculum of the pathogen than were used here'
ideally, 1-100 ascospores per wound. L4l
The ability of T. harzianum to grow above the point of inoculation on the trunk of the vines over a period of 2O months was demonstrated in the experiment with
Trichodowels@. The antagonist had colonised two of the seven Trichodowel@-treated vines at a distance of 18 cm above the point of inoculation 12 months after the treatment was applied. Also, one vine showed colonisation at the crown by T. harzianum 2O months after the treatment. This persistence of T. harzianum in the Trichodowel@-treated vines may assist in the prevention of infection by E. Iata. T. harzianurn persisted in the vines 20 months after treatment with Trichodowels@ in varying numbers of the vines at the different distances above the point of inoculation (Table 6.1). Most authors have tested the persistence of Trichodermd spp. at the site of, or just above or below the point of, inoculation (Pottle et al., 1977; Smith et al', 1979; Mercer and Kirk, 1984b)' Z' harzianurn was reported by Smith et at. (1979) to persist in the wood for one year after inoculation while pottle et al. (1977) reported persistence 31 months after inoculation.
Also, Mercer and Kirk (1984b) demonstrated persistence of T. viride in wounds on beech trees 4 years after inoculation. Furthermore, Mercer and Kirk, (1984b) reported the horizontal growth of Trichoderma sp. into beech wood, but research on the vertical spread of Trichoderma spp. in wood is scarce. While results for the colonisation and persistence of T. harzianum in vines in the present study were encouraging, further
experiments with Trichodowels@ should be conducted on vines of a range of cultivars at
various stages of maturity, with inoculation at different time of the year and at different
points along the trunks of the vines, to ascertain the ability of T. harzianum to colonise
and persist in grapevines. 142
In a small experiment performed on Trichodowel@-treated vines, establishment of E. lata from plugs inserted into trunks was observed in three of the four vines treated with gamma-irradiated Trichodowels@, whereas none of the four vines that were treated with
Trichodowels@ and inoculated with plugs of E. lata yielded the pathogen 14 weeks after inoculation. This observation provides a preliminary indication that T. harzianum inhibited the establishment E. lata, however, further trials with more replicates, isolates and cultivars are necessary to test this possibility.
The low incidence of infection of spurs inoculated with ascospores of E. lata in the
Trichodowel trial could, again, be due to physiological activity of the plant in early summer when this trial was conducted and to increased competition from the epiphytic microflora. Further trials with more replicates, established in early winter, are necessary to study the effect of Trichodowels@ on infection of spurs by ascospores of E' lata. A similar technique (see section 2.4.3.3) has been used by Dubos and Ricard (1974). While they reported effective control of the disease using this technique, it involved up to 12 holes in the trunk or the major limbs of the tree in a spiral pattern to ensure even distribution of the spores, whereas in the Trichodowel@ experiment only two holes were drilled side by side. Also, with the Trichodowel@ trial, the preventive effect of the Z. harzianum on the establishment of E. lata was tested, whereas Dubos and Ricard (1974) tested the curative effective of T. viride on silver leaf disease. The incidence of Dutch elm disease was reduced using a method similar to the Trichodowel@ experiment by introducing Trichoderma pellets into elm trees (Ricard' 1983).
Although the infection by E. lata of spurs on vines injected with T. harzianum was
slightly less than on those injected with SDW, the difference was not statistically t43 significant, largely due to the reduced number of replicates caused by vandalism. The conditions imposed in this experiment was artificial in that natural infection occurs by the introduction of ascospores to fresh pruning wounds (Moller and Kasimatis, 1978), whereas here mycelial plugs of a virulent isolate (M280) were introduced to circumvent the issue of hypovirulence (Carter, 1991). Significant reduction in the presence of wood decay fungi was reported when beech trees were injected with spore suspensions of T. viride, F. Iateritium and Cryptosporiopsis fasiculata (Mercer and Kirk, 1984b).
However, these reports did not investigate the effect of injected antagonists on the establishment of the pathogen when inoculated at a distance from the antagonist, as was tested here. Other authors have reported successful prophylactic control of Dutch elm disease using injections of trees with bacterial suspensions (Scheffer, 1983; Murdoch ¿/ al. 1984). Murdoch et al. (1984) were able to prevent Dutch elm disease by injecting
American elm trees with Pseudomonas fluorescens. The preventive treatments with P. fluorescens v/ere more effective than the therapeutic treatment and P' fluorescens reduced the mortality and symptom development one year after treatment' Further investigation of Trichoject@ as a treatment to prevent or cure eutypa dieback of grapevines is warranted.
The frequency of re-isolation of E. lata from the spurs that were inoculated one day
after injection did not differ from that of spurs inoculated 14 days later. This may be due
to the fact that the age of the wounds which was the same on day 1 and day 14, hence
these wounds were equally susceptible to infection by E. lata (Carter and Moller, 1970;
Ramos et al., 1975a; Munkvold and Marois 1993a). Further investigations with larger
delays (more than 14 days) between the time of injection of the antagonists and the r44 introduction of the pathogen are necessary to evaluate the effect of injections of T. harzianum on the incidence of infection by E.lata.
The absence of foliar symptoms on the vines in the injection trials 9 months after inoculation with E. Iata is in agreement with a report by Moller and Kasimatis (1978) that the appearance of visible symptoms in grapevines may take 3 years or more' Hence, similar trials should be established and assessed over a period of 3 years or more if foliar symptoms are to be used to evaluate the effect of treatment with Trichoject@ on eutypa dieback of grapevines.
In summary, treatment of pruning wounds with spore suspensions of Z' harzianum and F. lateritium consistently reduced infection by E. Iata, although the effect was not always statistically significant. Treatment with Trichoseal spray@ or Vineua*@ *u, effective in two of the five trials. Poor establishment and colonisation of pruning wounds by E. lata was a limitation of the four trials established at the end of winter. T. harzianum introduced in the form of Trichodowels@ was able to colonise, grow and persist in the vines for more than ayear.
Research is required to ascertain the mechanism(s) of antagonism of E. Iata in grapevines in field conditions. Also, longer-term trials are needed to establish the duration of protection afforded by pruning wound treatment, and treatment with
Trichodowel s@ and Trichoj ect@. t45
CHAPTER 7. PRELIMINARY STUDIES OF \ryOUND RESPONSE
7.1. Introduction
The study of the wound response of grapevine canes could allow an improved understanding of the reactions of the host-plant to wounding and invasion by E. lata and
T. harzianum.'lhe plant cell wall is the first mechanical barrier that has to be overcome by fungi to ensure pathogenesis (Pearce and Rutherford, 1981; Morris et a1.,1989). Cell walls comprise complexes of carbohydrates, proteins, lignin, cutin, suberin and certain inorganic compounds (Muhlethaler, 1967', Spiro, 1970; Cleland, 1971; Kolattukudy,
1980; Kolattukudy et a1.,1981; Showalter,Igg3). The appearance and properties of the cell walls may alter in response to microbial invasion. These changes may include deposition of suberin and callose, lignification, impregnation with oxidized phenols such as melanin and accumulation of calcium, silicon or sulphur. As a result, there may be an increase in the strength and thickness of cell walls, enhancement of resistance to enzymic
attack and direct toxicity of wall plecufsors, such as phenols, to pathogens' These host reactions may impede penetration of cells by the pathogen (Lucas, 1998)' Abiotic factors and other types of injury, including mechanical wounding, may also induce these cell- wall changes. Defence-related compounds can be detected qualitatively by histochemical staining.
Lignin is a complex three-dimensional polymer that is deposited in secondary walls
(Talmadge et al., 1973).It is resistant to degradation by most microorganisms and is
thought to be formed generally in response to penetration by microbes and mechanical
injury (Kirk, l97l; Heitefuss and William 1976; Benhamou et a1.,2000). Increased t46
lignification was reported in Japanese radish inoculated with Peronospora parsitica and
Alternaria japonica and in potatoes inoculated with Phytophthora infestans (Asada and
Matsumato, 196l; Asada and Matsumato, 1972: Henderson and Friend, 1919)'
Accumulation of lignin was induced in cells of wheat leaves by filamentous fungi and
the abiotic inducer mercuric chloride (Ride and Pearce, 1979 Pearce and Ride, 1980).
Also, Southerton and Deverall (1990) have reported increase of lignin in cells
undergoing a hypersensitive response in a wheat cultivar which expressed a highly
specific resistance to Puccinia recondita f .sp. tritici. Lignification was one of the host
reactions that was associated with the containment of Leptosphaeria maculans that
causes stem canker in oilseed rape (Hammond and Lewis, 1987). In many host-pathogen
interactions, deposition of lignin is thought to provide a substantial barrier to pathogen
ingtess (Benhamou and Bélanger, 1998; Benhamou, et al., 2000; Hammerschimidt,
2000)
Suberin is a polymeric compound attached to the walls of cells of the periderm,
including wound periderm, on aerial parts of plants and the endodermis (Dean and
Kolattukudy, 1977; Espelie and Kolattukudy, I979a; Espelie and Kolattukudy, I979b)'
Periderms may form naturally or be due to wounds. In woody plants, the epidermis is
initially replaced by the first periderm and subsequently by layers of consecutive
periderms (Esau, 1953; Bostock and Stermer, 1989). Formation of suberised tissues has
been related spatially and terrrporally to inhibition of fungal colonisation (Biggs, 1984;
Biggs et al., 1984; Biggs, 1986b). Functions of suberin include: barrier to diffusion of
enzymes or toxins of the pathogen into living tissues, structural barrier to pathogen
ingress and biochemical barrier to the microorganisms due to the large amounts of 147
phenolics incorporated into the suberin polymer (Kolattukudy, 1980; Kolattukudy, 1981;
Ouellette, 1981; Pearce and Rutherford, 1981; Biggs, 1985a; Biggs, 1985b; Espelie er al., 1986; Biggs and miles, 1988; Kolattukudy, 1989). A rapid increase in the number of layers of suberised cells has been reported in potato tubers that were inoculated with
Verticillium dahliae (Vaughn and Lulai, I99Ia). A reduction in the incidence of canker caused by Leucostoma sp. in peach trees was attributed to suberisation around bark wounds (Biggs and Miles, 1985).
The appearance of fluorescent compounds in tissues is considered to be due to the presence of phenolic compounds that accumulate in host-plant tissues to contain the development of the pathogen (Cohen et al., 1990). Autofluorescence of plant cells is an early response of the host to fungal infection (Kidger and Carver, 1981; Kunoh et al.,
1982; Kunoh et al., 1983; Kunoh et al., I985a; Kunoh et al., 1985b)' Tomato cell cultures inoculated with Verticittium albo-atrum showed an increase in wall-bound phenolics (Bernards and Ellis, 1991). Accumulation of phenolics has also been reported in carnation stems infected with Fusarium oxysporum f. sp. dianthi (Niemann and
Baayer, 1988; Niemann et al., l99la; Niemann et aI., 1991b). High levels of phenolic compounds were reported in a variety of maize that was resistant to Helminthosporium maydis (Angra-Sharma and Sharma, 1994).
The anatomy and composition of wood in grapevine is similar to that of most woody
species except, in Vitis spp., the phloem in the woody tissues remains functional for
more than one growing season (Esau, 1948). The cells do not die in winter, but retain the
same characteristics of active elements that they showed when they first differentiated
from the cambium. 148
Experiments described in this chapter involved the detection of accumulation of lignin, suberin and phenolic compounds in grapevine cuttings following the introduction of the pathogen and antagonist into the host together or alone vla pruning wounds. The main objective of this study was to investigate the role of T. harzianum in stimulating the wound response to protect grapevine tissues from infection by E. lata.
7 .2. Materials and Methods
T.2,l.Inoculation and harvest of canes
Single-node cuttings of cultivar Shiraz were placed in water overnight after the second basal bud was removed. The cuttings were inoculated with 5 mm plugs of ?. harzianum (strain 1) or E lata (isolate M280) alone, or T. harzianum plus E lata by making a fresh pruning cut 4 cm above the node. In the co-inoculated canes E. Iata was introduced 2 days after T. harzianum, by removing the plug of T. harzianum and replacing it with a plug of E. lata. The plugs of inoculum rwere covered with Parafilm@.
Also, cuttings were pruned and left un-inoculated. There were three replicates per
Íeatment. The cutttings were grown in rockwool pieces (4 x 4 cm Grodan blocks, see section 3.1) in a water-saturated condition in the laboratory. The canes were harvested 2
and 7 days after treatment with E. lata and2-cm segments below the point of inoculation
were excised ancl placecl in chloral hydrate (50 g in 2O ml distilled water; BDH
Chemicals, England) for 3 to 5 days (Ruzin, 1999). After discarding 0.2 mm of the tissue
immediately below the wound transverse sections of the segments were cut using
double-edged Gillette silver blue razor blades@. Sections were stained for lignin and r49
suberin as described in sections 7.2.2 and 7.2.3 and viewed using a light microscope
(l-eitz wetzlar, Orthoplan 871288). A fluorescent microscope (Olympus BH2-RCFA
246056) fitted with a 100W high-pressure mercury burner and BP 495 exciter-b.arrier fluorescence filters was used to detect phenolic compounds in non-stained sections
(section 7.2.4).
7.2.2. Detection of lignin
The transverse tissue sections were placed in phloroglucinol solution, incubated for 30 min. then mounted in lactoglycerol. Phloroglucinol solution was prepared by dissolving
0.1 g phloroglucinol (BDH Chemicals, England) in 16 ml of concentrated HCI and 100 ml of 95Vo ethanol. Phloroglucinol- HCI stains lignified walls red (Ruzin, 1999).
7,2.3. Detection of suberin
The transverse sections were placed in 5O7o ethanol for a few seconds and then incubated in a solution of Sudan Black B for 5-10 min. The sections were differentiated with 50Vo aqueous ethanol for 1 min. and mounted in glycerin. The solution of Sudan
Black B was prepared by dissolving 0.07 g of Sudan Black B (BDH Chemicals,
England) in 100 ml of 707o ethanol (Ruzin, 1999). Sudan Black B stains suberin black
(Pearse, 1960).
7.2.4. Detection of phenolic compounds
The un-stained tissue sections were mounted in lactophenol and observed for
autofluorescence (see section 7 .2.1). 150
7.3. Results
The cuttings treated with the two fungi did not differ from the uninoculated cuttings in terms of the presence of lignin, suberin and phenolics. The same was true for the two harvest dates. Hyphae were observed in the sections taken from the inoculated cuttings harvested after 7 days, but not in those harvested after 2 days.
7.3.1. Lignin
Intense red staining of walls of vessels and of most cells in the inner cortex region indicated the presence of lignin (Figures 7 .l and 7.2). The ray parenchyma cells and the pith parenchyma cells did not stain as intensely as the cells of the cortex (Figures 7.1,7.
2 and 7.3). The sclerenchymatous bundles in the outer cortex next to the epidermis also stained red (Figure 7.4). Adjacent to these bundles of sclerenchyma, strands of sclerenchyma (four to five layers in thickness) showing a positive reaction to lignin were interspersed with collenchyma cells that did not test positive for lignin (Figure 7'4).
7.3.2. Suberin
Suberin was observed in the cortical cells and in the walls of the xylem vessels
(Figure 7.5). Intense blue or black staining of outer cortical cells was observed (Figures
7.6 and 7.7). The walls of pith parenchyma cells close to the cortex and the sclerenchymatous bundles in the outer cortex wele stained blue, black or brown with
Sudan Black B (Figures 7.8 and 7.9). Furthennore, black, brownish-black or blue
staining of the lumina of cortical cells was observed (Figure 7.10). 15l
7 .3,3. Phenolic compounds
Blue autofluorescence was observed in the walls of cells in the cortex and pith
(Figures 7.11 and 7.14). The inner walls of the vessels autofluoresced blue while the outer walls of the vessels and cells surrounding the vessels showed yellow autofluoresence (Figure 7.II). Ray parenchyma cells generally produced blue autofluorescence (Figure 7.11). Also, yellow autofluoresence was observed in the bundles of sclerenchyma in the outer cortex, in the continuous band of cells below the bundle of sclerenchyma (Figure 7.12) and in the strands of sclerenchyma adjacent to this band of cells (Figurel.I3). The lumina of pith parenchyma cells that were close to the cortex showed blue and yellow autofluoresence (Figure7.l4). 152
Figures 7.1-7.4. Transverse sections of l-year-old canes of cultivar Shiraz inoculated
with Z. harzianum (strain 1), or co-inoculated and harvested after 2 ot 7 days and stained
with phloroglucinol-HCl to detect lignin. Figure 7.L. Intense red staining of walls of
vessels and most cells of the cortex in canes inoculated with ?' harzianum alone and
harvested after 7 days. White arrow indicates less intensely stained ray parenchyma cells.
Blue arrow shows the slightly stained pith parenchyma cells. Bar = 160 ¡rm. Figure 7.2.
Most of the vessels and cortical cells show deposition of lignin in cane inoculated with T.
harzianum alone and harvested after 7 days. White affow indicates the ray parenchyma
cells showing slight staining. Bar = 100 ¡rm. Figure 7.3. Difference in deposition of
lignin in cortex and pith parenchyma cell walls in cane inoculated with Z. harzianum and
harvested after 2 days. White affows indicate slight staining of pith parenchyma cells and
blue arrows indicate intense staining of cortex. Bar = 100 ¡rm. Figure 7.4. Staining of
the thick cell walls of the bundles of sclerenchyma in the outer cortex region (blue
arrows) and the parenchyma in the outer cortex shows no deposition of lignin (black
arrows). Strands of sclerenchyma (geen arrows) below the bundles of sclerenchyma
show intense staining and are interspersed by collenchyma in the outer cortex. The cane
was co-inoculated and harvested after 2 days. Bar = 100 ¡rm. €çI 154
7.3
{, ,'! I -\ '1" {
"\
.-+ ìt ii
)
I
I I 155
Figures 7.5-7.10. Accumulation of suberin in l-year-old canes of cultivar Shiraz uninoculated or inoculated with T. harzianum (strain l), E. lata or co-inoculated, and harvested 2 or 7 days later. Suberin detected by staining transverse sections with Sudan
Black B. Figure 7.5. Accumulation of suberin indicated by blue to brownish-black staining in the cortical cells and walls of xylem vessels in cutting inoculated with E' Iata alone and harvested after 7 days. Figures 7.6 and7.7. Blue and black staining of lumen and cell walls of outer-cortex cells. The black ¿ürows indicate deposition of suberin in the bundles of sclerenchyma in the outer cortex. The sections were from cuttings that were co-inoculated and harvested after 2 days and from cuttings inoculated with E. lata alone and harvested 7 days later, respectively. Figure 7.8. Pith parenchyma cells
(indicated by black arrows) showing deposition of suberin in cutting inoculated with Z. harzianum alone and harvested 7 days later. Figure7.9. Blue staining of the bundles of sclerenchyma in the outer cortex of canes inoculated with r' harzianum alone and harvested after 7 days (indicated by white arrows). Figure 7.10. Brownish-black or blue staining of lumen of cortical cells in canes inoculated with E. lata alone and harvested 2 days later. Bars = 100 Pm. 156
.a.-
t
t
útl t I
J-a -ì 'i¿? r+ r57
a
!. I 158
Figures 7,ll-7.14. Detection of phenolic compounds in l-year-old canes of cultivar
Shiraz by autofluoresence of the transverse sections. The canes were uninoculated or inoculated with E. lata or T. harzianum alone or co-inoculated and harvested after 2 or J days later. Figure 7.11. Autofluorescence of cells of cortex. Inner walls of vessels show blue autofluoresence while the outer walls of vessels and the cells surrounding the vessels emit yellow autofluoresence. The section was obtained from a cane inoculated with E. lata alone and harvested after 7 days. Figure l.l2.Yellow autofluorescence in the bundles of sclerenchyma and in the band of cells below in cane that was inoculated with T. harzianum alone and harvested 7 days later. Figure 7.13. Strands of sclerenchyma showing yellow autofluoresence in the outer cortex region. The section was cut from cane inoculated with Z. harzianum alone and harvested 7 days later. Figure
7.14. Lumina of pith parenchyma cells close to the cortex showing yellow autofluoresence. Bars = 40 pm. -
r t"¿
6çr 09r 161
7.4. Discussion
The results of these histochemical tests suggested that, apart from the normal response of the host plant to pruning, there was no change in accumulation of lignin, suberin or phenolic compounds in the presence of either the pathogen or the antagonist' There were no obvious differences in the distribution of lignin, suberin or phenolic compounds in wood harvested 2 and J days after inoculation. These observations implied that, in the conditions imposed in these investigations, T. harzianum did not induce the accumulation of wound response compounds in grapevine.
A large proportion of the cortical cells tested positive for lignin and suberin (section
L3.1 and,7 .3.2). Almost all the cells of the woody tissues of the 1-year-old canes showed the presence of phenolic compounds (section 1.3.3). Likewise, many authors have reported the accumulation of lignin, suberin and phenolic wound response compounds in woody plants after mechanical wounding (e.g. Pearce and Rutherford, 1981; Pearce and
Holloway; 1984; Doster and Bostock, 1988; Biggs, 1987)'
Cell wall depositions such as lignin may be preformed or induced in the plant tissues.
However, infection-induced lignin can be different from the preformed polymer in healthy plants (Asada and Matsumoto, 1912; Ride, 1975). This difference in lignin, however, may not have been detected in this study using the phloroglucinol-HCl test
alone. Other stains or tests, such as toluidine blue-O, chlorine-sulfite and the modified
chlorine-sulfite test, have been useil by resealchers to detect additional lignin (Ride,
1975; Southerton and Deverall, 1989; Southerton and Deverall, 1990). Lignin polymers
devoid of the cinnamaldehyde groups may not react with phloroglucinol. Southerton and
Deverall (1989; 1990) noted that additional lignin formed in the incompatible r62
interactions between Puccinia recondita f. sp, triticl in wheat cultivars could not be detected with the chlorine-sulfite test because it was not rich in syringic gloups.
Lignification is generally suppressed in compatible interactions but is predominant in resistant interactions (Hijwegen, 1963; Ride,I9l8; Ride, 1980; Bird and Ride, 1981;
Beardmore et al., 1983; Coffey and Cassi dy, 1984; Dean and Kuc, 1987; Cadena-Gomez and Nicholson, 1987; Benhamou et a1.,2000; Kumudini and Shetty, 2OO2).In this study, neither T. harzianum îor E. Iata increased accumulation of lignin in Shiraz. Hence, further testing using alternative staining techniques, a range of grapevine cultivars and strains of T. harzianum and E. lata may be necessary before a conclusion on interactions between T. harzianum, grapevine and E. lata could be drawn in terms of lignification.
Also, biochemical assays for lignin, its precursors, or enzymes in the biosynthetic pathway should be explored.
Suberin has been reported to occur not only in cortical cells close to the epidermis, but also in the inner cortex in xylem parenchyma, fibres, tyloses and vessel linings of various woody plants (Ouellette, 1978; Robb et a1.,1979; Ouellette, 1980; Ouellette, 1981; Robb et al., 1982;Pearce and Holloway, 1984). Also, Biggs (1987) reported suberin in xylem parenchyma and vessels following mechanical wounding. The presence of suberin in the outer and inner cortex in the present study is in agreement with these findings (section
7.3.2). However, the lack of difference in accumulation of suberin in the uninoculated
grapevine canes and those inoculated with Z. harzianum and E. lata is in contrast to
observations with other host-microbe systems. Simard et al. (2001) reported the
formation of ligno-suberized tissues as an anatomical defence mechanism in jack pines
resistant to the fungal pathogen Gremmeniella abietina. Jabaji-Hare et al. (1999) r63
repofted that binucleate Rhizocronla (BNR) induced accumulation of suberin in bean hypocotyls. Suberin has been reported to accumulate generally in the walls of cells.
However, in the present study, the lumina of grapevine cortical cells tested positive with
Sudan Black B. This observation is in agreement with that of Kasiamdari (2001), who detected suberin in the lumina of cells of the endodermis of mung bean roots infected with BNR. The presence of suberin in the lumina of grapevine cuttings may provide some degree of protection against microbial invasion, although it did not prevent colonisation by E lata andT. harzianum.
In addition to their role in the formation of structural barriers, phenolic compounds may be fungitoxic in their own right (Keen and Littlefield, 1979; Vance et aI., 1980;
Haars et al., 1981; Conti et al., 1986; Goodwin et al., l98l; Ismail et al., 1987)' The blue autofluoresence observed in the grapevine cells may have been due to phenolic compounds (section 7.3.3). Similar observations have been attributed by various authors to the accumulation of phenolic compounds (Pearce and Rutherford, 1981:.Mouzeyat et at., 1993). Yellow autofloresence was also observed in the grapevine cells (section
7.3.3), and this phenomenon has also been attributed to phenolics. Ride and Pearce
(1979) demonstrated that, in wheat leaves, lignin that was formed in response to infection by non-pathogenic fungi showed yellow fluoresence. This was markedly shifted to blue-green fluoresence (which resembled the characteristic fluoresence of the xylem tissues) when other material, which urasked the true colour of lignin, was extracted by pre-treating the tissue with alkali. Ride and Pearce (1979) also suggested that this difference in colour in autofloresence of lignin might be due to the differences
in the lignin polymer itself. The grapevine tissues that tested positive for lignin were r64
found to autofluoresce yellow in this study, possibly due to the phenolic component of the lignin polymer. Yellow autofluoresence was also reported by Stockwell and Hanchey
(1983) in Rhizoctonia solani- infected bean stem tissues and by Jabaji-Hare et al. (1999) in cells within and around lesions due to R. solani in bean hypocotyls. It has also been associated with the accumulation of phenolic compounds in response to inoculation by
Plasmoparaviticola, in the stomatal cells and the cell walls around the necrotic stomata in the downy mildew-resistant grapevine Vitis rotundiþlia (Dai et al., 1995). Blue autofluoresence has been associated with the presence of suberin (Pearce and Rutherford,
1981: Parker and Koller, 1998; Jabaji-Hare et al., 1999). The phenolic (aromatic or lignin-like) domain of the suberin polymer is thought to be responsible for its ability to autofluoresce in UV light (Vaughn and Lulai, 1991b; Lulai and Morgan, 1992).
Autofluoresence of both cell walls and cell lumina observed in the present study is in agleement with observations made by other authors (Cohen and Eyal, 1988; Cohen et al.,
1990; Cordier et al., t996).
In summary, these preliminary investigations indicated that T. harzianurz (strain 1) did not boost the wound response in terms of accumulation of lignin, suberin or phenolic compounds in Shiraz. Further investigations with different cultivars of grapevines and different strains of the antagonist, as well as the use of alternative staining techniques and quantitative biochemical analysis of the defence-related compounds, would be necessary before a conclusion could be reached. 165
CHAPTER 8. GENERAL DISCUSSION
8.1. Introduction
The aim of this study was to evaluate the potenti al of T. harzianum to control infection of grapevines by E. lata. The interactions between the pathogen and antagonist were investigated to understand the mechanisms which may be involved in biological control of the dieback pathogen.
8.2. Summary of findings
Laboratory studies indicated that antagonism was mainly by antibiosis. Volatile and non-volatile antibiosis by the three strains of Z. harzianum significantly inhibited mycelial growth and germination of ascospores of E. lata on PDA. Non-volatile antibiotics produced by 2-day-old cultures of T. harzianurn strains 2 and 3 on PDA completely inhibited the germination of ascospores. The non-volatile metabolites had a fungistatic effect on some isolates and a fungicidal effect on other isolates of E. lata
(chapter 4).
Application of T. harzianum strain 1 as spore suspension and in Trichoseal@ base, as
well as Trichoseal@ with seven strains of T. harzianl,olx, rcduced infection of pruning
wounds by E. lata in cuttings grown in the glasshouse when the antagonist was
introduced at least 2 d,aysprior to the pathogen. The nutrients in the Trichoseal@ base did
not enhaíce the biological control activity of T. harzianum. These glasshouse t66
experiments suggested that the antagonist should be applied to the pruning wounds or at least in close proximity to the pathogen to reduce colonisation by E. lata (Chapter 5)'
Spore suspensions of Z. harzianum strain 1 protected pruning wounds from infection by E. lata in the majority of the field trials. The commercial formulation containing seven sffains of T. harzianum and nutrients did not protect pruning wounds from infection by ascospores of E. lata in most of the trials established late in winter.
However, in the pruning wound trial established early in winter, the commercial formulation was more effective than the spore suspension of T. harzianum sÚain l.
T. harzianurø introduced into trunks of vines in the form of Trichodowels@ persisted in the vines for more than a year and colonised the woody tissues up to 18 cm above the point of dowel insertion. Furthermore, inoculation of vines with Trichodowels@ inhibited the establishment of E. lata when the pathogen was applied as plugs of mycelium to the trunks 16 months after treatment with Trichodowels@ (Chapter 6).
T. harzianum did not boost host defence by increasing the accumulation of lignin,
suberin and phenolic substances (Chapter 7).
8.3. Implications of fÏndings and future research
E. lata is a random-mating species with a high degree of genetic diversity (Péros and
Berger, Iggg). Hence, the sensitivity of isolates to a particular antibiotic or combinations
of antibiotics produced by the antagonist, may vary. Further research is necessary to
identify the antibiotics produced by the seven strains of T. harzianum used in the
commercial formulations of the Trichoprotection@ products, and these antibiotics need to
be tested in different proportions both in vitro and in planta against a range of isolates of r61
E. Iata.The inclusion of seven strains of T. harzianum in the Trichoprotection@ products is likely to be advantageous in view of the variable response of the isolates of E. lata to the metabolites of T. harzianum. Furthermore, variability in pathogenicity among isolates of E. lata has been reported when isolates obtained from different geographic locations (Ramos et al., 1975a; Péros and Berger, 1994), from the same perithecial stroma (Carter et a1.,1985; Rumbos, 1987; Péros and Berger, 1994), and even from the same ascus (English et a\.,1983), were compared. Therefore, further trials using a wide range of isolates of E. lata are necessary to test these Trichoprotection@ products in the glasshouse and in the field in different locations around Ausffalia.
The performance of different strains of a biological control agent may vary considerably in different geographic locations, due to variation in the environmental conditions, such as temperature and moisture (Stabb et al., 19941,Handelsman and Stabb,
1996; Bull et al., 1997). Hence, broadening the spectrum of effective strains of T. harzianumused in the Trichoprotection@ products may be beneficial in terms of effective control of E. Iata in various environmental conditions. The effect of environment on biological control, however, was not investigated in this study.
In 12 weeks,T. harzianuntcolonised cuttings in the glasshouse up to a distance of 10 cm below the point of inoculation (see section 5.3.5). Strain 7 of T. harzianum in SDW
colonised more than 2O7o of the cuttings at this distance and, in cuttings treated with
strain 1 in Trichoseal@, colonisation was observed in more than307o of the cuttings at 10
cm below the point of inoculation after 12 weeks. More than 807o of the cuttings treated
with Trichoseal@ containing seven strains of T. harzianurn yielded the antagonist at 10
cm below the point of inoculation after 12 weeks (see section 5.3.5). Also, in the 168
Trichodowel@-treated vines 7. harzianum colonised wood at 18 cm above the point of inoculation in two of the seven vines in 12 months (see section 6.3.3). The lack of adverse effects in the tissue suggest that T. harzianum may exist in the vines as an
"endophyte". Further investigation is needed to determine if T. harzianum can prevent infection by E. lata after it has established itself as an "endophyte" within the vine.
The results of the pruning wound experiments in both the glasshouse and fieltl showed that the additives and/or nutrients in Trichoseal@ and Trichoseal sptay' formulations did not enhance antagonism of E. lata by T. harzianum in comparison to the treatment with spore suspension of T. harzianurz in SDW. However, in the field trial established early in winter, the application of Trichoseal ,pruyt to pruning wounds was more effective than was the treatment with a spore suspension of T. harzianum in SDW.
The formulation, especially nutrient amendments, may need to be reviewed to promote establishment of T. harzianum,if the Trichoprotection@ products are to be used to protect wounds on grapevines from E. lata at all times of the year. Various authors have demonstrated the importance of additives in biological control. Schmidt et al' (2001a) showed that the antagonistic activity of B. subtilis and E. herbicola was greafer in grape wood when the bacteria were re-suspended and applied in fresh nutrient broth than when
applied in MgSOa solution. Likewise, the antagonistic activity of Bacill¿rs sp' against P.
aphanidermatum in cucumber fruit was stimulated by the addition of fresh nutrient
medium (Smith et a1.,1993). Furthermore, in investigations on biological control of E
lata, peptone has been used as an amendment to preparations of antagonists such as T.
viride, F.lateritium, C. herbarum andB. subtilis to ensure rapid increase in populations
of the antagonists on pruning wounds (Ferreira et al., 1991; Munkvold and Marois, 169
1993a). Glucose and peptone increased in vitro inhibition of E. Iataby B. subtilis and E. herbicola, respectively, and peptone was reported to increase the antagonistic performance of B. subtilis on autoclaved grape wood (Schmidt et aI., 2001b).
Polysaccharides such as alginate and methylcellulose have been used as additives to preparations of biocontrol agents (Kloepper and Schroth, 1981; Lewis and Papavizas,
1991). Tronsmo (1986) reported that methylcellulose not only improved adhesion of propagules of Trichoderma spp. to the treated surface but was also utilised as a nutrient source by the antagonist. Methylcellulose, peptone and glucose were reported to stimulate the production of volatile antifungal compounds by B. subtills (Fiddaman and
Rossall, Igg4). Schmidt et al. (2001b) reported that methylcellulose increased the antagonistic activity of B. subtills against E. lata. Reports are available on the positive effects of nitrogen and phosphate salts and manganese ions on antagonistic performance of B. subtllls (Gupta and Utkhede, 1987; I-eifert et al', 1995). On the other hand, additives may also have adverse effects on biocontrol activity. Alginate, carob and phosphate supplements had a negative effect on biological control of E. lata by B.
subtilis (Schmidt et al. 2001b). Phosphates repressed production of herbicolin by E. herbicola and many other antibiotics (Martin and Demain, 1980; Greiner and
Winkelmann, 1991). Hence, the additives used in the Trichoprotection@ products could be tested in different environmental and climatic conditions to optimise their beneficial effects on the biocontrol activity of T. harzianum towards E. lata. Furthermore, in
addition to the effect that the additives have on the antagonist, their influence on the
pathogen also must be considered. Many authors have reported a growth-inducing effect
of additives on necrotrophic pathogens, which may use these additives as nutrients r70
(Pottle et al., 1977; Tronsmo, 1986; Schmidt et al' 2OOlb). In the present study, the possibility that E. Iata may have utilised the additives was suggested as a reason for the poor performance of Trichoseal .p.uyt in some of the pruning wound trials (see section
6.4). Therefore, in vitro and in planta screening is necessary to eliminate the possibility of utilisation of the additives in Trichoseal@, Trichoseal spray@ or Vinevax@ by E. loto, or even by the resident microflora which, in turn, may compete with T. harzianurn. Since initially, Trichoprotection@ products were not developed exclusively for the control of eutypa dieback, investigation of these aspects is important to ensure consistent results in the vineyard.
Superior biocontrol strains of Z. harzianum and other Trichoderm4 spp. have been obtained not only through selection from nature (Chet and Baker, 1981; Sivan and Chet,
1986; Smith et al., 1990; Elad et al., 1993) but also after mutation or selection for enhanced resistance to fungicides (Ahmad and Baker, 1987; Tronsmo, 1989). Currently, recombinant protoplast techniques are being used to engineer genetically modified organisms or improve existing qualities of microorganisms (Papavizas, 1987) and transformation techniques are well developed for various fungi, including Trichoderma spp. (Mach and Zellinger, 1998). Hence, it may be beneficial to improve the T. harzianum strains used in the commercial formulations to increase production of metabolites, in such combinations that would enhance inhibition of a wide range of
isolates of E. lata. However, genetically modified organisms would not be accepted in all
countries nor in organic farming systems, a likely market for biological control agents.
Strains could also be manipulated to exhibit resistance to fungicides used to protect
pruning wounds, so that biological and chemical agents could be used in combination. L7I
For example, Carter and Price (1975) selected a strain of F. Iateritiurnwith resistance to benomyl, with the intention that the fungicide could protect the wound tissues while F. lateritium became established. Furthermore, it may be useful to evaluate the other six strains of T. harzianum in the Trichoproducts@ for mycoparasitic activity on E. lata in planta and, thereafter, either genetically modify existing strains to be aggressive mycoparasites, or select new aggressive mycoparasitic strains of T. harzianum and incorporate these into commercial products. Trichoprotection@ products containing aggressive mycoparasitic strains may be effective as a prophylactic treatment, and could also be used to inhibit further development of E latathathas already become established in the vines (Hjeljord and Tronsmo, 1998).
In one of the glasshouse experiments, T. harzianum (strain 1) and F. lateritium falled to gtow up towards the pruning wounds and colonise the canes sufficiently to prevent establishment of the pathogen in the cuttings, even when the antagonists were introduced
7 days prior to, but at a distance, from the pathogen (see sections 5.2.3 and 5'3.3). This result implied that once E. lata became established in the cuttings, inhibition by T. harzianum (strain 1) or F. lateritium was not possible although in the various laboratory tests significant inhibition of E. lata was observed when the pathogen was introduced prior to the antagonist and in the same position or close to the antagonist. Hence, in the subsequent glasshouse and field experiments, preventive use of the antagonists was investigated. It is possible that once E. lata becomes established within the grapevine tissues, it is better able to compete with the antagonists, since the grapevine tissues are
not the natural habitat of the antagonists. This hypothesis could be tested by introducing 172 the antagonists into grapevine cuttings at various time intervals after the pathogen and assessing colonisation of canes by E. lata.
Further experiments in the field are necessary to study the protective effect of
Trichodowels@ against eutypa dieback of grapevines. T. harzianum may prevent infection of trunks by E. lata if the antagonist has the opportunity to colonise the vines before the arrival of the pathogen (see section 6.3.3). Howevet, it may be necessary for
T. harzianum to grow and colonise woody tissues close to the potential infection courts of the pathogen. Trichodowels@ inserted into holes drilled in a spiral pattern along the trunk well before pruning, possibly 3 to 4 months before pruning, may facilitate rapid colonisation by, and uniform distribution of, T. harzianum throughout the trunk to protect the vines from infection by E. lata via pruning cuts.
The development of molecular markers for both T. harziønum and E. lata would be useful in tracking these organisms in the vine. The conventional method of re-isolations to detect fungi in wood is tedious and time consuming. PCR primers have been developed for the detection of E. lata in France and California (kelan et al., 1999;
Lecomte et a1.,2000). Whether these primers identify Australian isolates of E. lata is not known. Currently, two approaches are being pursued in Australia to develop DNA based methods capable of detecting E. lata in grapevines: (1) SCAR (Sequence Characterised
Amplified Region) markers (2) DNA probes developed from DNA sequences that are
specific to E. lata (John e/ al., 2O0l; Lardner et aL, 2OO3). DNA markers have been
developed for Trichoderma spp. for taxonomical purposes. Restriction Fragment Length
Polymorphism (RFLPs) patterns of DNA have been produced, representing the
cellobiohydrolase I-encoding gene cbhl from different species of Trichoderma that were t73 species-specific (Morawetz et at., 1992; Kubicek et aL, 1996). RFLPs were used to assess variation between isolates of T. harzianum that colonised mushroom compost
(Muthumeenakshi et al., 1994). PCR finger printing has been used by Schlick et al.
(Igg4) to identify mutants induced by gamma-irradiation and patented strains of T. harzianum.In addition, Bowen et at. (1996) have succeeded in differentiating an isolate of T. harzianum that was an antagonist of Sclerotinia sclerotiorum from other closely related Trichodermd spp. using RFLP analysis. This development is considered to be of immense value since it may enable monitoring of strains of T. harzinnurn in trials to assess biological control ability in the field.
Future research should include the investigation of induced resistance by T. harzianum in terms of deposition of lignin, suberin and callose and the synthesis of pathogenesis-related proteins (PRP) in a range of cultivars of grapevines. Calederón ¿r at. (1993) reported increased production of benzoic acid and resveratrol, a potent phytoalexin, in grapevine cell cultures challenged with an elicitor, cellulase (Onozuka T-
10). The elicitor was obtained from an isolate of T. viride that was an effective biocontrol agent of B. cinerea in grapevines. Barber and Ride (1988) observed that the cellulase from Z. viride elicited lignification of wounded wheat leaves. Benzoic acid and/or citric acid or ascorbic acid have been reported to be effective in controlling infection of various crops by B. cinerea (Elad, 1992). Research is required to study the relationships between cultivar and wound susceptibility, accumulation of lignin, suberin,
callose, phenolics and PRP, infection by E. lata and biocontrol activity of T. harzianum
in grapevines. r74
In the short term, nutrient amendments to Trichoprotection@ products and the inclusion of a broad spectrum of T. harzienum strains in these formulations may be promising lines of work to follow. Genetic manipulation to produce aggressively mycoparasitic strains of T. harzianum and research into induced resistance of grapevines may be long term strategies worth investigating, albeit with the caveat about acceptability of genetically modified organisms, as discussed above.
In suÍìmary, biological control of eutypa dieback of grapevines based on T. harzianum has the potential to be effective. However, successful commercial use of T. harzianum is dependent upon the appropriate formulation in terms of additives, nutrients and T. harzianum strain composition of the Trichoprotection@ products. Such modification in the Trichoprotection@ products would give the antagonists a competitive advantage over E. lata and epiphytes that are present on grapevines and, thereby, ensure consistent and reliable biological control of E' lata' t75
REFERENCES
Adam, D.8., 1938. A progress report on gummosis (dieback) disease in South Australia apricot trees. Journal of the Department of Agriculture of South Australia. 42: 14-29.
Agrios, G.N., 1997. Plant Pathology. Academic Press, San Diego, USA.
Ahmad, J.S., R. Baker, 1987. Rhizosphere competence of Trichoderma harzianum.
Phytopathology. 77 : 1 82- 1 89.
Ahmad, J.S., R. Baker, 1988. Implications of rhizosphere competence of Trichoderma harzianumon carbon substrates. Canadian Journal of Microbiology. 34:694-696.
Almassi, F., E.L. Ghisalberti, M.J. Narbey, K. Sivasithamparam, 1991. New antibiotics from strain s of Trichoderma harzianum. Journal of Natural Products. 54: 396-
402.
Anderson, P.C., B.V. Brodbeck, 1989. Chemical composition of xylem exudate from bleeding spurs of Vitis rotundiþtia noble and Vitis hybrid Suwanee in relation to pruning dates. American Journal of Enology and Viticulture. 40: 155-160.
Angra-Sharma, R., D.K. Sharma, 1994. Biochemical and histological studies on susceptible and resistant maize leaves infected by Helminthosporium maydis. Plant
Pathology. 43: 912-97 8.
Anon., lgg7, IPM Viticulture Research to Practice Training Workshop Manual. Department of Natural Resources and Environmental Agriculture, Victoria.
Asada, Y., I. Matsumoto, 1967. Formation of lignin in the root tissues of Japanese radish affected by Alte rnaria j aponic a. Phytopathology. 57 : 1339 -1343. 176
Asada, Y., I. Matsumoto, L972. The nature of lignin obtained from downy mildew infec ted Japane se radi sh roots. Phytop atho lo gy zeitschrift. 7 3 : 08 -21 4.
Baker, K.F., 1957. The U. C. system for producing healthy container-grown plants. California Agricultural Experiment Station Manual' 23 : 332pp'
Baker, K.F., 1987. Evolving concepts of biological control of plant pathogens. Annual Review of Phytopathology. 25: 67 -85.
Baker, K.F., R.J. Cook, 1974. Biological Control of Plant Pathogens. W. H. Freeman and Company, San Francisco, USA.
Barak, R., Y. Elad, D. Mirelman, I. Chet, 1985. Lectins a possible basis for specific recognition in the interaction of Trichoderma and Sclerotium rolfsii. Phytopathology 75: 458-462.
Barber, M.J., J.P. Ride, 1988. A quantitative assay for induced lignification in wounded wheat leaves and its use to survey potential elcitors of the response.
Physiological and Molecular Plant Pathology. 32: 1 85 - 1 97.
Barnett, H.L., F.L. Binder,1973. The fungal host-parasite relationship. Annual Review of Phytopathology. ll: 27 3-292.
Beagle-Ristaino, J.8., G.C. Papavizas, 1985. Survival and proliferation of propagules of Trichoderma spp. and Gliocladium virens in soil and in plant rhizospheres. 75:.729- 732.
Beardmore, J., J.P. Ride, J.W. Granger, 1983. Cellular lignification as a factor in the hypersensitive resistance of wheat stem rust. Physiological Plant Pathology. 22:2O9-23O. r77
Belarbi, 8., G. Mur, 1983. Observations sur la germination des conidies ou stylospores du champignon Eutypa armeniacae. I-e Progrés Agriculture et Viticultute.24; 636-637 .
Benhamou, N., R.R. Bélanger, 1998. Benzothiadiazole-mediated induced resistance to Fusarium oxysporum f. sp. radicis-lycopersicl in tomato. Plant Physiology. 118: l2O3- t2r2.
Benhamou, N., I. Chet, 1993. Hyphal interaction between Trichoderma harzianum and Rhizoctonia solani: Ultra-structure and goldcytochemistry of the mycoparasitic process.
Phytopathology. 83 : 1062-107 I.
Benhamou, N., I. Chet, 1997. Cellular and molecular mechanisms involved in the interaction between Trichoderma harzianum and Pythium ultimum. Applied and Environmental Microbiology. 63 : 2095 -2099.
Benhamou, N., S. Gagne, D. Le Gere, L. Debbi, 2000. Bacterial-mediated induced resistance in cucumber: beneficial effect of the endophytic bacterium Serratia plymuthicaon the protection against infection by Pythium ultimum. Phytopathology. 90: 45-46.
Berg, G., G. Ballen, 1gg4. Bacterial antagonists to Verticillium dahlia¿ Kleb.. Journal of
Phytopathology. 141 : 99-1 10'
Bernards, M.4., B.E. Etlis, L991. Phenylalanine ammonia-lyase from tomato cell
cultures inoculated withVerticiltium albo-atum. Plant Physiology.91: 1494-1500'
Biggs,4.R., 1984. Boundary zone formation in peach bark in response to wounds and
Cyt o sp ora leuc o stoma infection. Canadian Journal of Botany. 62: 2814-2821'
Biggs, 4.R., 1,985a. Detection of impervious tissue in tree bark with selective
histochemistry and fl uorescence microscopy. Stain Technology . 60: 299 -304. 178
Biggs, 4.R., 1985b. Suberized boundary zones and the chronology of wound response in tree bark. Phytopathology. 75: 1191-1195.
Biggs, 4.R., 1986a. Prediction of lignin and suberin deposition in boundary zoîe tissue of wounded tree bark using accumulated degtee days. Journal of the American Society for Horticulture Science. 11 1: 757 - 760.
Biggs, 4.R., 1986b. Comparative anatomy and host response of two peach cultivars inoculated with Leucostoma cincta and L. persoonii. Phytopathology ' 76: 905-912.
Biggs, 4.R., 1987. Occurrence and location of suberin in wound reaction zones in xylem of 1 7 tree species. Phytopathology . 7 7 : 7 I8-7 25'
Biggs, 4.R., 1989. Temporal changes in the infection court after wounding of peach bark and their association with cultivar variation in infection by Leucostoma persoonü.
Phytopathology 7 9 : 627 -630.
Biggs, 4.R., W. Merill, D.D. Davis, 1984. Discussion: Response of bark tissues to injury and infection. Canadian Journal of Forest Research. 14: 351-356.
Biggs, 4.R., N.N. Miles, L985. Suberin deposition as a measure of wound response in peach bark. Hortscience. 20: 903-905.
Biggs, 4.R., N.W. Miles, 1988. Association of suberin formation in un-inoculated wounds with susceptibility to Leucostoma cincta and L. persoonii in various peach cultivars. Phytopathology. 78: IO7 0-IO7 4.
Bird, P.M., J.P. Ride, 1981. The resistance of wheat to Septoria nodorum: fungal development in relation to host lignification. Physiological Plant Pathology' 19:289-299. t79
Bisby, G.R., 1939. Trtchoderma viride Pers. ex. Fr. and notes on Hypocrea. Transactions of the British Mycological Society' 23: 149-168.
Blakeman, J.P., 1988. Competitive antagonism of air-borne fungal pathogens. In: Fungi in Biological Control Systems. (ed. M. N. Burge). Manchester University Press, Manchester, UK.
Boesewinkel, H.J., L976. Storage of fungal cultures in water. Transactions of the British Mycological Society. 66: 183-185.
Boirie, M., B.J. Pons, 1984. Phior-P, a new fungicide based on Trichoderma harzianum
Revue Suisse de Viticulture, d'Arboriculture et d'Horticulturae. 16: 199-200'
Bolay, 4., M.V. Carter, 1985. Newly recorded hosts of Eutypa lata (= E. armeniacae) in Australia. Plant Protection Quarterly. I: I0-I2.
Boosalis, M.G., lgí .Hyperparasitism. Annual Review of Phytopathology. 2:363-373
Bostock, R.M., B.A. Stermer, L989. Perspectives on wound healing in resistance to pathogens. Annual Review of Phytopathology. 27:343- 371.
Bowen, J.K., S.C. Fransicevic, R.N. Crowhurst, M.D. Templeton, A. Stewart, L996. Differentiation of a specific Trichoderma biological control agent by restriction fragment length polymorphism analysis. New Zealand Journal of Crop and Horticulture Science. 24:207-217.
Brian, P.W., H. Hemming, 1945. Gliotoxin, a fungistatic metabolite oT T.richoderma viride. Annals of Applied Biology. 32:2lI-220.
Brian, P.W., J.C. McGowan, 1945. Viridin: A highly fungistatic substance produced by Trichoderma viride. Nature. 156: I44. 180
Bruce,4., W.J. Austin, B. King, 1984. Control of growth of Lentinus lepideusby volatiles fromTrichoderma. Transactions of the British Mycological Society' 82:423-
428.
Bull, C.T., J.P. Stock, J.L. Smilanick, 1997. Pseudomonas syringa¿ strains ESC-10 and ESC-ll survive in wounds on citrus and control green and blue mold of citrus. Biological Control. 8: 81-88.
Burge, M.N., 1988. The scope of fungi in biological control. In: Fungi in Biological Control Systems. (ed. M. N. Burge). Manchester University Press, Manchester, UK.
Burnham, T.J., 1998. Pruning and training are keys to Eutypa control. Grape Grower April:p.4.
Cadena-Gomez, G., R.L. Nicholson, 1967. Papilla formation and associated peroxidase activity: A non-specific response to attempted fungal penetration of maize. Physiological and Molecular Plant Pathology. 3I: 5t-67.
Calderón, 4.A., J.M. Zapata, R. Munz, M.A. Pedréno, A. Ros Barceló' 1993. Resveratrol production as a part of the hypersensitive-like response of grapevine to an elicitor from Trichode rma viride. New Phytologist. 124: 45 5 -463.
Campbell, R., L989. Biocontrol of Microbial Plant Pathogens. Cambridge University
Press, Cambridge, UK.
Carter, M.V., 1957a. Eutypa armeniacae Hansf. & Carter, sp. nov., an airborne vascular pathogen of Prunus armeniacae L. in Southern Australia. Australian Journal of Botany' 5:21-35.
Carter, M.V., 1957b. Vines aid spread of "gummosis". Journal of the Department of Agriculture of South Australia. 6O:482-483. 181
Carter, M.V., L960. Further studies on Eutypa armeniacae Hansf. & Carter. Australian
Journal of Agricultural Research. 1 1: 498-505.
Carter, M.V., 1965. Ascospore deposition of Eutypa armeniaca¿. Australian Journal of Agricultural Research. 16: 825-836.
Carter, M.V., 1971. Biological control of Eutypa armeniacae. Australian Journal of Experimental Agriculture and Animal Husbandry. ll: 687-692.
Carter, M.V., 1983. Biological control of Eutypa armeniacae 5. Guidelines for establishing routine wound protection in commercial apricot orchards. Australian Journal of Experimental Agriculture and Animal Husbandry. 23:429-436.
Carter, M.V., 1991. The status of Eutypa lata as a pathogen. C.A.B. International, Wallingford, UK.
Carter, M.V., A. Bolay, I. Rumbos, 1985. Variation in the pathogenicity of Eutypa lata (= E. armeniacae). Australian Journal of Botany. 33 36I-366.
Carter, M.V., W.J. Moller, 1970. Duration and susceptibility of apricot pruning wounds to infection by Eutypa armeniacae. Australian Journal of Agricultural Research. 2I:915-920
Carter, M.V., ril.J. MollerrlgTl. The quantity of inoculum required for infection of apricot and other Prunus species by Eutypa armeniacae. Australian Journal of Experimental Agriculture and Animal Husbandry . I I : 684-686.
Carter, M.V., T.V. Price,l974. Biological control of Eutypa armeniacae 2. Studies of the interaction between Eutypa armeniacae and Fusarium lateritium and their relative r82 sensitivities to benzimidazole chemicals. Australian Journal of Agicultural Research. 25: lO5-119.
Carter, M.V., T.V. Price, 1975. Biological control of Eutypa armeniacae 3. A' comparison of chemical, biological and integrated control. Australian Journal of
Agricultural Research. 26: 537 -543.
Chambers, S.M., 1993. Phytophthora root rot of chestnut. PhD thesis, University of Adelaide, Adelaide, Australia.
Chambers, S.M., E.S. Scott, 1995. In vitro antagonism of Phytophthora cinnamomi and P. citricola by isolates of Trichodernxa spp. and Gliocladium virens. Journal of
Phytopathology. 143: 41 l-411 .
Chapius, L., L. Richard, B. Dubos, 1998. Variation in susceptibility of grapevine pruning wounds to infection by E. lata in South-'Western France. Plant Pathology. 47:
463-472.
Chérif, M., N. Benhamou, L990. Cytochemical aspect of chitin breakdown during the parasitic action of aTrichoderma sp. on Fusarium oxysporum f. sp. radicis lycopersici. Phytopathology. 80: 1406-1414.
Chet, 1., L987. Trichoderma: Application, mode of action and potential as a biocontrol agent of soilborne plant pathogenic fungi. In: Innovative Approaches to Plant Disease Control. (ed. I. Chet). John Wiley and Sons, New York, USA.
Chet, I., R. Baker, 1981. Isolation and biocontrol potential of Trichoderma hamatum
from soil naturally suppressive of Rhizoctonia s olani. Phytopathology. 7 I: 286-290.
Chet, I., N. Benhamou, S. Haran, L998. Mycoparastism and lytic enzymes. In: Trichoderma and Gliocladium: Enzymes, Biological Control and Commercial 183
Applicarions. Vol. 2. (ed. G. E. Harman & C. P. Kubicek). Taylor and Francis Ltd, London, UK.
Chet, I., G.E. Harman, R. Baker, 1981. Trichoderma hamatum: Its hyphal interactions with Rhizoctonia solani and Pythium spp' Microbial Ecology. 7:29-38'
Chiarappa ,L, 2000. Esca (black measles) of grapevine. An overview. Phytopathologia
Mediterranea. 39: 1 1- 1 5.
Claydon, N., M. Allan, J.R. Hanson, A.G. Avent, 1987, Antifungal alkyl pyrones of Trichoderma harzianum. Transactions of the British Mycological Society. 88: 503-513.
Cleland, R., 197L. Cell wall extension. Annual Review of Plant Physiology.22: 197- 222.
Coffey, M.D., D.S.M. Cassidy, 1984. Peroxidase activity and induced lignification in rusted flax interactions varying in their degee of incompatibility. Canadian Journal of Botany. 62:134-14I.
Cohen, Y., H. Eyal, 1988. Epifluorescence microscopy of Sphaerotheca fuliginea race 2 on susceptible and resistant genotypes of Cucumis melo.Phytopathology' 18: 144-148.
Cohen, Y., H. Eyal, J. Hanania, L990. Ultrastructure, autofluorescence, callose deposition and lignification in susceptible and resistant muskmelon leaves infected with the powdery mildew fungus Sphaerotheca fuliginea. Physiological and Molecular Plant Pathology. 36: l9I-204.
Collins, R.P., A.F. Halim, 1972. Characterization of the major aroma constituent of the fungus Trichoderma viride (Pers.). Journal of Agriculture and Food Chemistty.2O: 437- 438" 184
Conti, G.G., M. Bassi, D. Maffï, A.M. Bocci, 1986. Host-parasite relationship in a susceptible and a resistant rose cultivar inoculated with Sphaerotheca panosa. ll. Deposition rates of callose, lignin and phenolics in infected or wounded cells and their possible role in resistance. Journal of Phytopathology. lI7: 312-320'
Cook, D.W.M., P.G. Long, S. Ganesh, H.F. Neilson, 1997. Attachment microbes antagonistic against Botrytis cinerea - biological control and scanning electron microscope studies in vivo. Annals of Applied Biology. l3l:213-226.
Cordier, C., S. Gianinazzi, V. Gianinazzi-Pearson, 1996. Colonisation patterns of root tissues by Phytophthora nicotianae var. parasitica related to reduced disease in mycorrhizal tomato. Plant and Soil. 185: 223-232.
Corke, A.T.K., 1974. The prospect of biotherapy in trees infected by silver leaf. Journal of Horticulture Science. 49: 391-394.
Correa, 4., S. Rebuffat, B. Bodo, M.F. Roquebert, J.B. Dupont, L, 1995. In vitro inhibitory activity of trichorzianines of Sclerotium rolfsii Sacc. Cryptogamie Mycologie. 16:185-190.
Creaser, M.L., T.J. Wicks, 2002, Infections extend below ground in vines severely affected by eutypa dieback. The Australian Grapegrower and Winemaker. 46Ia:76-77.
Dai, G.H., C. Andary, L. Mondolot-Cosson, D. Boubals, L995. Histochemical studies on the interactions between three species of grapevine, Vitis vinifura, V. rupestris and V. rotundiþlia and the downy mildew fungus Plasmopara viticola. Physiological and Molecular Plant Pathology. 46: 177-I88.
Danielson, R.M., C. B. Davey, L973. Effects of nutrients and acidity on phialospore germination of Trichoderma in vllro. Soil Biology and Biochemistry. 5:511-524. 185 d'Enfert, C.r lg97, Fungal spore gennination: Insights from the molecular genetics of
Aspergillus nidulans and Neurospora crassa. Fungal Genetics and Biology.2l: 163-172.
De Scenzo, R.4., S.R. Engel, G. Gomez, E.L. Jackson, G.P. Munkvold, J. Weller' N.A. Irelan, 1999. Genetic analysis of Eutypa strains from California supports the presence of two pathogenic species. Phytopathology. 89: 884-893.
Dean,8.8., P.E. KolattukudyrlgTT. Biochemistry of suberization: Incorporation of (1- 14) oleic acid (1-14c)-acetate into aliphatic components in potato tuber disks (Solanum tuberosum). Plant Physiology . 59: 48-54.
Dean, R.4., J. Kuc, 1987. Rapid lignification in response to wounding and infection as a mechanism for induced systemic protection in cucumber. Physiological and Molecular
Plant Pathology. 7 9: 107 1 - 1078.
Decognet, V., P. Nicot, J.C.V. Lenteren, 1999. Effects of fungicides on a Fusarium sp. biological control agent of Botrytis cinerea stem infections in the perspective of an integrated management of fungal diseases in greenhouse tomatoes. International Organisation for Biological and Integrated Control of Noxious Animals and Plants-West Palaearctic Regional Section. Working Group on "Integrated control in glasshouses", Yol.22. Brest, France. pp.49-52.
Dennis, C., J. Webster, l97la. Antagonistic properties of species groups of Trichoderma. I. Production of non-volatile antibiotics. Transactions of the British
Mycological Society. 57 : 25-39.
Dennis, C., J. Webster, 1971b. Antagonistic properties of species goups of Trichoderma. II. Production of volatile antibiotics. Transactions of the British
Mycological Society. 57 : 47-48. 186
Dennis, C., J. Webster, l97lc. Antagonistic properties of species groups of Trichoderma. IIIr Hyphal interaction. Transactions of the British Mycological Society. 57:363-369.
Di Pietro, 4., M. Lorito, C.K. Hayes, R.M. Broadway, G.E. Harman, 1993' Endochitinase from Gliocladium virens: isolation, characterization and synergistic antifungal activity in combination with gliotoxin. Phytopathology. 83: 308-313'
Dik, 4.J., G. Koning, J. Kohl, 1999. Evaluation of microbial antagonists for biological control of Botrytis cinerea stem infection in cucumber. European Journal of Plant
Pathology. 105: 1 15-122.
Doster, M.4., R.M. Bostock, 1988. Quantification of lignin formation in almond bark in response to wounding and infection by Phytophthora species. Phytopathology. 78: 473-
417.
Dowson, \ry.J., 1931. The dieback disease of apricots. Preliminary note. Tasmanian
Journal of Agriculture. 2: L65 -166.
Dry, P.R., G.R. Gregory, 1988. Grapevine varieties. In: Viticulture, Vol. 1. Resource. (ed. B.G. Coombe & P. R. Dry). Australian Industrial Publishers Pty Ltd, Adelaide, Australia.
Dubos, 8., 1987a. Mise au point sur les maladies de dépérissement dans le vignoble français. Le Progrés Agriculture et Viticulture. 104: 135-140'
Dubos, 8., 1987b. Fungal antagonism in aerial agrobiocenoses. In: Innovative Approaches to Plant Disease Control. (ed. I. Chet). John Wiley and Sons, New York, USA. 187
Dubos, 8., J.L. Ricard, 1974, Carctive treatment of peach trees against silver leaf disease (Stereum purpureum) with Trichoderna viride preparations. Plant Disease Reporter. 58: 147-150.
Dunlop, R.'!il., A. Simon, K. Sivasithamparam, E.L. Ghisalberti, 1989. An antibiotic from Trichoderna koningii active against soilborne pathogens. Journal of Natural Products. 52:61-74.
Duthie, J.4., G.P. Munkvold, J.J. Marois, 1991.. Relationship between age of vineyard and incidence ofeutypa dieback. Phytopathology' 81: 1183-1184.
Elad, Y.rlgg2. The use of antioxidants (free radical scavengers) to control grey mould (Botrytis cinerea) and white mould (Sclerotinia sclerotiorum) in various crops. Plant
Pathology. 4I: 417 -426.
Elad, Y,, lgg4. Biological conffol of grape gray mould by Trichoderma harzianum. Crop Protection. 13: 35-38.
Elad, Y., R. Barak, I. Chet, 1983a. Possible role of lectins in mycoparasitism. Journal of Bacteriology. 154: l43l-I435.
Elad, Y., R. Barak, I. Chet, Y. Henis, 1983b. Ultrastructural studies of the interaction between Trichoderma spp. and plant pathogenic fungi. Phytopathology Zeitschrift. 107: 168-175.
Elad, Y., I. Chet, P. Boyle, Y. Henis, 1983c. Parasitism of Trichoderma spp' oî Rhizoctonia solani and Sclerotium rolfsü. Phytopathology. 73: 85-88.
Elad, Y., I. Chet, Y. Henis, 1982. Degradation of plant pathogenic fungi by Trichoderma harzianum. Canadian Journal of Microbiology. 28: 7t9-725. 188
Elad, Y., G. ZimandrY. Zaqs, S. Zuriel, I. Chet, 1993. Use of Trichoderma harzianum in combination or alteration with fungicides to control cucumber grey mould (Botrytis cinerea) under commercial greenhouse conditions. Plant Pathology.42:324-332'
Emmett, R.W., P.A. Magatey, 1994. Eutypa dieback. In: Diseases and Pests. Grape Production Series 1. (ed. P. Nicholas, P. Margarey & M. Wachtel). Winetitles, Cowandilla, South Australia.
English, H., J.R. Davis, J.M. Ogawa, F.J. Schick' 1983. Variation in Eutypa arrneniacae and discovery of its ascigerous stage in California's Central Valley.
Phytopathology. 7 3 : 95 8 (abstr.).
Esau, K., 1948. Phloem structure in grapevine and its seasonal changes. Hilgardia. 18: 2t1-296.
Esau, K., L953. Plant Anatomy. John Wiley and Sons Inc., New York, USA.
Espelie, K.8., V.R. Franceshie, P.E. Kolattukudy, 1986. Immunocytochemical localisation and time course of appearance of an anionic peroxidase associated with suberisation in wound-healing potato tuber tissue. Physiological and Molecular Plant
Pathology. 8l: 487 -492.
Espelie, K.E., P.E. Kolattukudy, 1979a. Composition of the aliphatic components of
suberin from the bundle sheath of Zea mays leaves. Plant Science Letters. L5 225-230.
Espelie, K.E., P.E. Kotattukudy, 1979b. Composition of aliphatic components of suberin of the endodermal fraction from the first internode of etiolated sorghum
seedlings. Plant Physiology. 63 : 433-435. 189
Faull, J.L., 1988. Competitive antagonism of soil-borne plant pathogens. In: Fungi in Biological Control Systems. (ed. M. N. Burge). Manchester University Press, Manchester, UK.
Ferreira, J.H.S., F.N. Matthee, A.C. Thomas, 1990. Fungi associated with dieback and pruning wounds of grapevines in South Africa. South African Journal for Enology and Viticulture. lO: 62-66.
Ferreira, J.H.S., F.N. Matthee, A.C. Thomas, 199L. Biological control of Eutypa lata on grapevine by an antagonistic strain of Bacillus subtilis. Phytopathology. 81: 283-287 '
Fiddaman, P.J., S. Rosall, 1994. The production of antifungal volatiles by Bacillus subtilis. Journal of Applied Bacteriology' 74: 119-126'
Flaherty, D.L., P.L. Christensen, T.W. Lanini, J.J. Marois, P.A. Phillips' L.T. Wilson, 1992. Grape Pest Management. The Regents of the University of California, Division of Agriculture and Natural Resources. California, USA'
Fravel, D.R., 1988. Role of antibiosis in the control of plant diseases. Annual Review of
Phytopathology. 26: 7 5-91.
Fruit, L., P. Nicot, J.V. Lentern, 1999. Biological control of Botrytis cinerea on tomato stem wounds with Ulocladium atrum. International Organisation for Biological and Integrated Control of Noxious Animals and Plants-West Palaearctic Regional Section. Working Group on "Integrated control in glasshouses", Vol. 22.Brest, France. pp. 81-84.
Fujiwara,4., T. Okutla, S. Masuda, Y. Shiomi, C. Miyamoto, Y. Sckine, M. Tazoe, M. Fujiwarar1982. Fermentation, isolation and characterization of isonitrile antibiotics.
Agricultural and Biological Chemis try. 46: 1 803- 1 809. 190
Garcia, I., J.M. Lora, J. De La Crtz, T. Benitez, A. Llobell, J.A. Pintor-toror 1994. Cloning and characterization of a chitinase (CHIT 42) cDNA from the mycoparasitic fungus, Trichoderma harzianum. Current Genetics. 27 : 83 -89 .
Gendloff, E.H., D.C. Ramsdell, C.L. Burton, 1983. Fungicidal control of Eutypa armeniacae infecting Concord grapevine in Michigan. Plant Disease. 61:754-756.
Genstat 5 committeerlgST. Reference Manual. Clarendon Press, Oxford, UK.
Ghisalberti,8.L., M.J. Narbey, M.M. Dewan, K. Sivasithamparam, 1990. Variability among strains of Trichoderma harzianum in their ability to reduce take-all and produce pyrones. Plant and Soil. 121:287-29L
Ghisalberti, 8.L., K. Sivasithamparam, 199L. Antifungal antibiotics produced by Trichodermd spp.. Soil Biology and Biochemistry. 23: I0l1-1020.
Gibbs, J.N., 1967. The role of host vigour in the susceptibility of pines to Fomes annosus. Annals of Botany.31: 803-815.
Glawe, D.4., J.D. Rogers, 1982. Observations of the anamorphs of six species of
Eutypa and Eutypella.Mycotaxon. 14: 334-346.
Glawe, D.4., C.B. Skotland, W.J. Moller, lgS2.Isolation and identification of Eutypa armeniacae from diseased grapevines in Washington state. Mycotaxon. 16: 123-132.
Goodwin, J.R., J.W. Mansfïeld, P. Darby, 1987. Microscopical studies of resistance to powdery mildew disease in hop cultivar Wye Target. Physiological Plant Pathology. 24:
33-42.
Grabowski, M., 1999. Research on utilisation of chosen isolates of fungi in biological control of Pezicula alba (Gtunthr.). Folia Horticulturae.It:29-35. 191
Grabowski, M., E. Schwartz, 1997. Research on possible application of chosen fungi in biological control of Nectria galligena (Bres.) on apple shoots. Phytopathologia Polonica. t4:75-81.
Graeme-Cook, K.4., J.L. Faull, 1991. Effect of ultra-violet induced mutants of Trichoderma harzianum with altered production on selected pathogens in vivo. Canadian
Journal of Microbiology. 37 : 659-664.
Greiner, M., G. Winkelmann, 199L. Fermentation and isolation of herbicolin A, a peptide antibiotic produced by Erwinia herbicola strain A 111. Applied Microbiology and Biotechnology. 34: 565-569.
Grosclaud€, C., J. Ricard, B. Dubos, 1973. Inoculation of Trichoderma viride spores via pruning shears for biological control of Stereum purpureum on plum tree wounds. Plant Disease Reporter. 57:25-28.
Gupta, V.K., R.S. Utkhede, 1987. Nutritional requirement for production of antifungal substance by Enterobacter aerogenes and Bacillus subtilis antagonists of Phytophthora cactorum. Journal of Phytopathology. I20: 143-153'
Haars, 4., I. Chet, A. Hüttermann, 1981. Effect of phenolic compounds and tannin on growth and laccase activity of Fomes annosus. European Journal of Forest Pathology. II 67-76.
Hadar, Y., I. Chet, Y. Henis, 1979a. Biological control of Rhizoctonia solani damping off with wheat bran culture of Trichoderma harzianum. Phytopathology. 69: 64-68.
Hadar, 8., Y. Elad, S. Ovadia, Y. Hadar, 1979b. Biological and chemical control of Rhizoctonia solani in carnation. Phytoparasitica. 7:55' r92
Hall, T.J., L.R. Schreiber, C. Leben, 1986. Effects of xylem-colonising Bacillus spp. on Verticillium wilt in maples. Plant Disease. 70: 52I-524.
Hammerschmidt, R., 2000. Changing susceptibility into resistance. Physiological and Molecular Plant Pathology. 57: 1-3.
Hammond, K.8., B.G. Lewis, 1987. Variation in stem infection caused by aggressive and non-aggressive isolates of Leptosphaeria maculans on Brassica napus var. oleiþra.
Plant Pathology. 36: 53-65.
Handelsman, J., E.V. Stabb, 1996. Biocontrol of soilborne plant pathogens. The Plant Cell. 8: 1855-1869.
Harman, G.8., I. Chet, R. Baker, L980, Trichoderma hamatum. Effects on seed and seedling disease induced in radish and pea by Pythium spp. or Rhizoctonia solani. Phytopathology. 70: I 167 -Il7 2.
Harman, G.8., I. Chet, R. Baker, 1981. Factors affecting Trichoderma hamatum applied to seeds as a biocontrol agent. Phytopathology ' 7l: 569-572.
Harman, G.E., C.K. Hayes, M. Lorito, R.M. Broadway, A. Di Pietro, C. Peterbauer' A. Tronsmo, 1993. Chitinolytic enzymes of Trichoderma harzianumi purification of chitobiosidase and endochitinase. Phytopathology. 83: 313-3 18.
Harman, G.8., B. Latorre, E. Agosin, R.S. Martin, D.G. Riegel, P.a. Nielsen' a. Tronsmo, R.C. Pearson, 1996. Biological and integrated control of Botrytis bunch rot of grape usingTrichoderma spp. Biological Control' 11:259-266.
Harman, G.8., E.B. Nelson, 1994. Mechanisms of protection of seed and seedlings by biological seed treatment: implications for practical disease control. Monograph No. 57. British Crop Protection Council, Canterbury, UK' pp'283-292. 193
Harris, J.8., 1932. Dieback of apricot trees in the Barrosa district. Journal of the Department of Agriculture of South Australia. 35: 1394-1395.
Hart, J.H., D. M. Shrimpton, 1979. Role of stilbenes in resistance of wood decay
Phytopathology. 69: 1 138-1 143.
Heitefuss, R., P.H. William, 1976. Physiological Plant Pathology. Springer-Veylag, Berlin, Germany.
Henderson, S.J., J. Friend, 1979.lncrease in PAL and ligninJike compounds as race- specific resistance response of potato tubers to Phytophthora infestans. Phytopathology
Zeitschrift. 94: 323 -33I.
Henis, Y., I. Chet, 1975. Microbial control of plant pathogens. Advances in Applied
Microbiolo gy. 19: 85-1 1 1.
Hijwegen, 1963. Lignification as a mechanism of induced systemic resistance in cucumber. Physiological Plant Pathology. 20: 6l -7 I.
Hjetjord, L.G., A. Stensvand, A. Tronsmo, 2000. Effect of temperature and nutrient stress on the capacity of commercial Trichoderma products to control Botrytis cinerea
and Mucor piriþrmis in greenhouse strawberries. Biological Control. 19: 149-16O.
Hjeljord, L.G., A. Stensvand, A. Tronsmo, 2001. Antagonism of nutrient activated conidia of Trichoderma harzianum (øtroviride) P1 against Botrytis cinerea.
Phytopathology. 9 1 : ll7 2-ll80.
Hjeljord, L., A. Tronsmo, 1998. Trichoderma and Gliocladium in biological control: An overview. In: Trichoderma and Gliocladium: Enzymes, Biological Control and Commercial Applications. Vol. 2. (ed. G. E. Harman & C. P. Kubicek). Taylor and Francis Ltd, London, UK. t94
Hong, C.X., T.J. Michailides, B.A. Holtz,1998. Effects of wounding, inoculum density
and biological control agents on post harvest brown rot of stone fruits. Plant Disease. 82: t2r0-1216.
Howell, C.R., 1998. The role of antibiosis in biocontrol. In: Trichoderma and Gliocladium: Enzymes, Biological Control and Commercial Applications. Vol. 2. (ed. G. E. Harman & C. P. Kubicek). Taylor and Francis Ltd, London, UK'
Howell, C.R., R.D. Stipanovic, 1983. Gliovirin, a new antibiotic ftom Gliocladium virens and its role in the biological control of Pythium ultimum. Canadian Journal of Microbiolo gy. 29: 321-324.
Howell, C.R., R.D. Stipanovic, R.D.Lumsden, 1993. Antibiotic production by strains of Gliocladium virens and its relation to the biocontrol of cotton seedling diseases. Biocontrol Science and Technology. 3: 435-441.
Howell, C.R., R.D. Stipanovic, 1995. Mechanisms in the biocontrol of Rhizoctonia solani-indtced cotton seedling disease by Gliocladium virens: antibiosis.
Phytopathology. 85 : 469 -47 2.
Hunt, J.S., 1999. Wood-invading fungi: protective management in the vineyard. The
Australian Grapegrower and Winemaker ' 426a: 125-126.
Hutchinson, L.J., P. Chakrvarty, L.M. Kawchuk, Y. Hiratzuka, 1994. Phoma etherigei sp. nov. from black galls and canker of trembling aspen (Populus tremuloides) and its potential role as a bioprotectant against the aspen decay pathogen Phellinus tremulae. Canadian Journal of Botany. 72:1424-1431.
Inbar, J., I. Chet, 1995. The role of recognition in the induction of specific chitinases during mycoparasitism by Trichoderma harzianum. Microbiology. 141: 2823-2829. 195
Inbar, J., A. Menendez, I. Chet, 1996. Hyphal interaction between Trichoderma harzianum and, Sclerotinia sclerotiorum and its role in biological control' Soil Biology and Biochemistry. 28 7 57 -7 63.
Irelan, N., D. Guler, R. Descenzo, 1999. Efficacy testing of Eutypa, chemical and biological control candidates with DNA-based diagnostics. Practical Winery and Vineyard. JanuaryÆebruary: pp. 47 -56.
Ismail, M.K., A.-A.M. Salama, M.I.A. Ali, S.A.-8. Ouf' 1987. Effect of some phenolic compounds on spore germination and germ-tube length of Aspergillus fumigatus and
F us arium ocysp orum f . sp. ly c op ersici' Cryptogamie Mycologie' 8 : 5 1 -60.
Jabaji-Hare, S., H. Chamberland, P.M. Charest, 1999. Cell wall alteration in hypocotyls of bean seedlings protected from Rhizoctonia stem canker by a binucleate
Rhizoctonia isolate. Mycological Research' 1 03 : 1 03 5 - 1 043'
John, S., R. Lardner, E. Scott, B. Stummer, T. Wicks,2001. Eutypa dieback: research on biological control and diagnostics. The Australian Grapegrower and Winemaker. 449a:73-75.
Johnsen, D.4., J.D. Lunden, 1985. Eutypa dieback in Washington State: Disease incidence and effect on yield. Phytopathology.T5:1370 (abstr.).
Ju, Y.M., D.A. Glawe, J.D. Rogers, 199L. Conidial germination in Eutypa armeniacae and selected other species of Diatrypaceous fungi. Mycotaxon . 4:3lI-32O.
Kasiamdari, R.S., 200l.Interactions between arbuscular mycorrhizal fungi and other root-infecting fungi. PhD thesis, University of Adelaide, Australia. 196
Keen, N.T., L.J. Littlefield, 1979. 'lhe possible association of phytoalexin with resistance gene expression in flax to Melamspora lini. Physiological Plant Pathology. 14: 265-280.
Kelley, W.D., 1976. Evaluation of Trichoderm harzianum-impregnated clay granules as a biocontrol of Phytophthora cinnamomi causing damping off of pine seedlings.
Phytopathology. 66: LO23-1027 .
Kidger, 4.L., T.L.N. Carver, 1981. Autofluorescence in oats infected by powdery mildew. Transactions of the British Mycological Society. 76: 405-409.
Kile, G.4., 1976. The effect of pruning and of time since pruning upon changes in apple sapwood and its susceptibility to invasion by Trametes versicolor. Phytopathology
Zeitschrift. 87 : 231-240.
Kirk, T.K., 1971. Effects of microorganisms on lignin. Annual Review of
Phytopathology. 9: 1 85-210.
Klein, D., D.E. Eveleigh, 1998. Ecology of Trichoderma and Gliocladium. In:
Trichoderma and Gliocladium: Basic Biology and Genetics. Vol. 1. (ed. G. E. Harman &
C. P. Kubicek). Taylor and Francis Ltd, London, UK.
Kloepper, J.!Y., M.N. Schroth, L981. Development of a powder formulation of rhizobacteria for inoculation of potato seed pieces. Phytopathology. 7l:590-592.
Ko, W., J.L. Lockwood, 1967. Soil fungistasis: relation to fungal spore nutrition. Phytopathology. 57: 894-901.
Kohl, J., M. Gerlagh, 1999. Biological control of Botrytis cinerea in roses by the antagonist Ulocladium atrum. Procceedings of the 51'tlnternational Symposium on Crop Protection. Part II, Vol. 64. Gent, Belgium.pp. 441-445. t91
Kolattukudy, P.8., 1980. Cutin, suberin and waxes. In: The Biochemistry of Plants. (ed. P. K. Stumpf & E. E. Conn). Academic Press, New York, USA.
Kolattukudy, P.8., K.E. Espelie, C.J. Soliday, 1981. Hydrophobic layers attached to cell walls. Cutin, suberin and associated waxes, Plant Carbohydrates II. Vol. l3B' In: Encyclopedia of Plant Physiology: New Series (ed. W. Tanner &' F. A. Loewus).
Springer-V erlag, B erlin, Germany.
Kolattukudy, P.8., G.K. Podilla, R. Mohan, 1989. Molecular basis of the early events in plant fungus interaction. Genome. 3I:342-349.
Krauss, U., lV. Soberanis, 2002. Effect of fertilization and biocontrol application frequency on cocoa pod diseases. Biological Control. 24: 109O-2112.
Kubicek, C.P., U.M. Bolzlbauer, W. Kovacs, K. Kuhls, E. Lieckfeldt' T. Borner, G.H. Samuels, 1996. Cellulase formation by species of Trichoderma sect. Longibrachiatum and of Hypocrea spp. with anamorphs referable to Trichoderma sect.
Lon gibrachi atum. Fun gal Genetics and B iolog y . 2O : 105 - I I 4.
Kullnig, c., R.L. Mach, M. Lorito, c.P. Kubicek,2000. Enzyme diffusion from Trichoderma atroviride (= T. harzianum P1) to Rhizoctonia solani is a prerequisite for triggering of Trichoderma ech42 gene expression before mycoparasitic contact. Applied and Env ironmental Microbi olo gy . 66 : 2232 -223 4 .
Kumudini, 8.S., H.S. Shetty , 2002. Association of lignification and callose deposition with host cultivar resistance and induced systemic resistance in pearl millet to
S clerosp ora graminic olø. Australasian Plant Pathology. 3l : 157 -164.
Kunoh, H., K. Kunoh, H. Ishizaki, L985. Cytological studies of the early stages of powdery mildew in barley and wheat. xi. Autofluorescence and haloes at penetration 198
sites of appressoria of Erysiphe graminis hordei and Erysiphe pisi on barley coleoptiles. Canadian Journal of Botany.63: 1535-1539.
Kunoh, H., K. Yamamori, H. Ishizaki, L982. Cytological studies of early stages of powdery mildew in barley and wheat. viii. Autofluorescence at penetration sites of Erysiphe graminis hordei on living barley coleoptiles. Physiological Plant Pathology. 21:
373-379.
Kunoh, H., K. Yamamori, H. Ishizaki, 1983. Cytological studies of early stages of powdery mildew in barley and wheat. ix. Effect of various inorganic salts on the occu11ence of autofluorescence at penetration sites of appressoria of Erysiphe graminis hordei. Canadian Journal of Botany. 6I:218L-2185'
Lake, C.8., R.K. Striegler, P.S. Verdegaal, C.4., Ingles, N. Irelan, J. Price, 1998. Influence of training systems of Cabernet Sauvignon on Eutypa dieback. Lodi- Woodbridge Wine Grape Commission Research / IPM update. California, USA.
Lardner, R., N. Mahoney, R. Molyneux, B. Stummer, E. Scott,2003. Early diagnosis of eutypa dieback of grapevines. 8ù International Congress of Plant Pathology. Abstracts of Offered Papers. Christchurch, New Zealand. p'77'
Larignon, 8., B. Dubos, 1997. Fungi associated with esca disease in grapevine.
European Journal of Plant Pathology. 103: 147-757 '
Lecomte, P., J.-P. Péros, D. Blancard, N. Bastien, C. Délye, 2000. PCR assays that identify the grapevine dieback fungus Eutypa lata. Applied and Environmental
Microbiolo gy . 66: 447 5-4480. 199
Leifert, c., H. Li, s. chidburee, s. workman, D. sigee, H.A.S. Epton, A. Harbour, 1995. Antibiotic production and biocontrol activity by Baccillus subtilis, CL 27 and Bacillus pumilus CL45. Journal of Applied Bacteriology. TS: 97-108.
Lewis, J.4., G.C. Papavizas, L99L. Biocontrol of plant diseases: the approach for tomorow. Crop Protection. 10: 95-105.
Lifshitz, R., M.T. Windham, R. Baker, 1986. Mechanisms of biological control of pre- emergence damping off of pea seed treatment with Trichoderma spp. Phytopathology. 76:720-725.
Lorito, M., 1998. Chitinolytic enzymes and their genes' In: Trichoderma and Gliocladiunt: Enzymes, Biological Control and Commercial Applications. Vol' 2. (ed. G. E. Harman & C. P. Kubicek). Taylor and Francis Ltd, London, UK.
Lorito, M., S.C. Woo, M.D. D'Ambrosio, G.E. Harman, C.K. Hayes, C.P. Kubicek' F. Scala, 1.996. Synergistic interactions between cell wall degrading enzymes and membrane affecting compounds. Molecular Plant-Microbe Interactions. 9: 206-213.
Lucas, J.4., L998. Plant Pathology and Plant Pathogens. Blackwell Science, London, UK.
Lulai, 8.C., \ry.C. Morgan, 1992. Histochemical probing of potato periderm with neutral red: a sensitive cytofluorochrome for the hydrophobic domain of suberin. Biotechnic and Histochemistry. 67: 185-195.
Lumsden, R.D., J.C. Locke, S.T. Adkins, J.F. walter, c.J. Ridout, l992a.Isolation and localization of the antibiotic gliotoxin produced by Gliocladium virens from alginate prill in soil and soilless media. Phytopathology' 82: 230-235' 200
Lumsden, R.D., C.J. Ridout, M.E. Vendermia, D.J. Harrison, R.M. Walters, J.F. Walters, lg92b. Charccteization of major secondary metabolites produced in soilless mix by a formulated strain of the biocontrol fungus Gliocladium virens. Canadian Journal of Microbiology. 38: I274-1280.
Mach, R.L., C.K. Peterbauer, K. Payer, S. Jaksits, S.L. Woo, S. Zeilinger, C.M. Kullnig, M. Lorito, C.P. Kubicek, L999. Expression of two major chitinase genes of Trichoderma (T. harzianum Pl) is triggered by different regulatory signals. Applied and Environmental Microbiology. 65: 1858-1863.
Mach, R.L., S. Zeilinger, 1998. Genetic transformation of Trichoderma and Gliocladium. In: Trichoderma and Gliocladium: Enzymes, Biological Control and Commercial Applications. Vol. 2. (ed. G. E. Harman & C' P. Kubicek). Taylor and Francis Ltd, London, UK.
Maiti, D., B. Dasgupta, C. Sen, 1991. Antagonism of Trichoderma harzianum and Gliocladium virens isolates to Sclerotium rolfsii and biological control of ground nut and betelvine. Journal of Biological Control. 5: 105-109.
Martin, J.F., A. L. Demain, 1980. Control of antibiotic synthesis. Microbiological Reviews. 44:230-251
Martin, J.F., G. Nicolas, 1970. Physiology of spore germination in Penicillium notatum and Trichoderma lignorum. Transactions of the British Mycological Society. 55: I4l-
148.
Mauro, M.C., V. Vaillant, P. Tey-Rulh, Y. Mathieu, J. Fallot, f988. ln vitro study of the relationship betwe en Vitis viniftra and Eutypa lata (Perc.: Fr.) Tul. I. demonstration of toxic compounds secreted by the fungus. American Journal of Enology and
Viticulture. 39 : 2OO -203. 20r
Mazzola, M., D.K. Fujmoto, L.S. Thomashow, R.J. Cook, 1995. Variation in sensitivity of Gaeumannomyces graminis to antibiotics produced by fluorescent Pseudomonas spp. and effect on biological control of take-all of wheat. Applied and Environmental Microbiology. 6l: 2554-2559.
McCullagh, P., J.A. Nedler, 1989. Generalized Linear Models. Chapman and Hall, London, UK.
McMahan, G., M. Yeh, M.N. Marshall, M. Olsen, S. Sananikone, J.Y. Wu' D.E' Block, J.S. Vander Gheynst, 2001. Characterising the production of a wild-type and benomyl resistant Fusarium lateritium for biocontrol of Eutypa lata on grapevine. Journal of Industrial Microbiology and Biotechnol ogy.26: 151-155.
Mercer, P.C., S.A. Kirk, 1984a. Biological treatments for the control of decay in tree wounds-1. Laboratory tests. Annals of Applied Biology. I04:2Il-219.
Mercer, P.C., S.A. Kirk, 1984b. Biological treatments for the control of decay in tree wounds-2. Field tests. Annals of Applied Biology. I04:221-229.
Merlier, A.M.O., J.M. Boire, J.B. Pons, M.C. Renaud, 1985. Strain of Trichoderma harzianum, its isolation, its culture, peptides or compounds produced by this strain and application of this strain and these peptides or the product produced by the culture process as a means for biological control in the form of an agticultural fungicide. European Patent Application EP 124388. Chemical Abstracts. 183747r:102.
Mitchell, C.P., N.J. Dix, 1975. Growth and germination of Trichodermd spp. under the influence of soil fungistasis. Transactions of the British Mycological Society. 64: 235- 24t.
Moller, W.J., M.V. Carter, 1965. Production and dispersal of ascospores in Eutypa armeniacae. Australian Journal of Biological Sciences' 18: 67-80. 202
Moller, W.J., M.V. Carter, 1969. A preliminary observation on apricot dieback prevention with chemicals. Plant Disease Reporter. 53:828-829'
Moller, W.J., H. English, J.R. Davis, 1968. Eutypa armeniaccte on grape in California. Plant Disease Reporter. 52:75L
Moller, W.J., A.N. Kasimatis, 1978. Dieback of grapevines caused by Eutypa armeniac ae. Plant Disease Reporter. 62: 25 4-258'
Moller, W.J., A.N. Kasimatis, 1980. Protection of grapevine pruning wounds from eutypa dieback. Plant Disease. 64: 278-280'
Moller, W.J., A.N. Kasimatis, 198La. Further evidence that Eutypa armeniacae- not Phomopsis viticola - incites dead arm symptoms on grape. Plant Disease' 65: 429-43t.
Moller, \ry.J., A.N. Kasimatis, 1981b. Fungicide protects grapevines ftom Eutypa' California Agriculture. 65: 8.
Moller, W.J., D.E. Ramos, R.R. Sanborn, 1977. Eutypa dieback in California apricot orchards: Chemical control studies. Plant Disease Reporter. 61: 600-604'
Molyneux, R.s., N. Mahoney, P. Bayman, R.Y. Wong, K. Meyer, N. Irelan, 2002. Eutypa dieback in grapevines: Differential production of acetylenic phenol metabolites by strains of Eutypa lata. Joûrnal of Agricultural andFood Chemistry. 50: 1393-L399'
Morawetz, R., F. Gruber, R. Messner, C.P. Kubicek, 1992. Presence, transcription and translation of cellobiohydrolase genes in several Trichoderma species. Current Genetics. 2I:3t-36.
Morris, S.C., M.R. Forbes-Smith, F.M. Scriven, 1989. Determination of optimum conditions for suberization, wound periderm formation, cellular desiccation and 203
pathogen resistance in wounded Solanum tuberosum tubers. Physiological and Molecular Plant Pathology. 35: 171-I90.
Morton, L., 2000. Viticulture and grapevine declines: lessons of black goo.
Phytopathologia Mediterranea. 39: 59 -67 .
Moubasher, 4.M., 1963. Effect of duramycin on some isolates of Trichoderma viride. Nature. 200:492.
Mouzeyar, T., D. Tourvielle de Labrouhe, F. Vear, 1993. Histopathological studies of resistance of sunflower (Helianthus annuus L.) to downy mildew (Plasmopara
halstedii). Journal of Phytopathology. I39: 289-297 .
Mugnai, L., a. Graniti, G. Surico, 1999. Esca (black measles) and brown wood- streaking: two old elusive diseases of grapevines. plant Disease. g3: 404-41g.
Muhlethaler, K., 1967. Ultrastructure and function of plant cell walls. Annual Review of Plant Physiology. 18: l-24.
Munkvold, G.P., A. Duthie, J.J. Marois, 1994. Reductions in yield and vegetative growth of grapevines due to eutypa dieback. phytopathology. 84: lg6-192.
Munkvold, G.P., J.J. Marois, 1993a. Efficacy of natural epiphytes and colonizers of grapevine pruning wounds for biological control of eutypa dieback. Phytopathology. 83: 6724-6729.
Munkvold, G.P., J.J. Marois, 1993b. The effect of fungicides on Eutypa lata germination, gtrowth and infection of grapevines. plant Disease. 77: 50-54. 204
Munkvold, G.P., J.J. Marois, 1994. Factors associated with the variation in susceptibility of grapevine pruning wounds for biological control of eutypa dieback.
Phytopathology. 83 : 624-629.
Munkvold, G.P., J.J. Marois, L995. Factors associated with variation in susceptibility of grapevine pruning wounds to infection by Eutypa lata.Phytopathology. 85:249-256.
Murdoch, C.\ü., R.J. Campana, N.E. Hoch, t984. On the biological control of
Ceratocystis ulmi with Pseudomonas fluorescens. Phytopathology. 74: 805 (abstr.).
Muthumeenakshi, S., P.R. Mills, A.E. Brown, f).4. Seaby, 1994. Intraspecific molecular variation among Trichoderma hqrzianum isolates colonising mushroom compost in the British Isles. Microbiology. 140:769-777 .
Nelson, 8.8., 1991. Current limits to biological control of fungal pathogens. In: Handbook of Applied Mycology: Soil and Plants. (ed. D. K. Arora, B. Rai, D. G. Mukerji, & G. R. Knudsen). Marcel Dekker,Inc., New York, USA.
Nelson, E.8., G.E. Harman, G.T. Nash, 1988. Enhancement of Trichoderma-indtced biological control of Pythium seed rot and pre-emergence damping off of peas. Soil Biology and Biochemistry. 20: 145-150.
Nelson, E,.8., M.H. Pearce, N. Malajczuk, L995. Effects of Trichoderma spp. and ammonium sulphamate on establishment of Armillaria luteobubalina on stumps of
Euc alyptus div ersic olor. Mycological Research. 99 : 957 -962.
Newhook, F.J., 1951. Microbiological control of Botrytis cinerea Pers. II. Antagonism by fungi and Actinomycetes. Annals of Applied Biology. 38: 185-202. 205
Niemann, G.J., R.P. Baayer, 1988. Involvement of phenol metabolism in resistance of Dianthus caryophyllus L. to Fusariurn oxysporum f . sp. dianthl. Netherlands Journal of
Plant Pathology. 94: 289-301.
Niemann, G.J., A. Van der Bii, B. Brandt-de Boer, J.J. Boon, R.P. Bayen' 1991a. Differential response of four carnation cultivars to races 7 and2 of Fusarium oxysporum
f . sp. dianthl and to Phialophora cinerscens. Physiological Plant Pathology' 38: 117-136.
Nieman, G.J., A. Van der Kerk, W.M.A. Niessen, K. Versluis, 1991b. Free and cell- wall bound phenolics and other constituents from healthy and fungus-infected carnation (Dianthus caryophyllu.r L.) stems. Physiological and Molecular Plant Pathology. 38: 417-432.
O'Brien, J.G., 1984. Assessment of Streptomyces spp. from elms for biological control of Dutch elm disease. Plant Disease. 68: 104-106'
O'Neill, T.M., A. Niv, Y. Elad, D. Shtienberg, 1996. Biological control of Botrytis cinerea on tomato stem wounds with Trichoderma harzianum.Etropean Journal of Plant Pathology. 102 635-643.
Ouellette, G.8., 1978. Unusual cell wall layers in elm parenchyma of secondary xylem. Canadian Journal of Botany. 56:2109-2113'
Ouellette, G.8., 1980. Occurrence of tyloses and their ultrastructural differentiation from similarly configured structures in American elm infected by Ceratocystis ulmi.
Canadian Journal of Botany. 58: 1056-1073.
Ouellette, G.8., 1981. Ultrastructural cell wall modifications in secondary xylem of American elm surviving the acute stage of Dutch elm disease: fibres. Canadian Journal
of Botany. 59 : 2425 -2438. 206
Pachenari, 4., N.J. Dix, 1980. Production of toxins and wall degrading enzymes by Gliocladium roseurn. Transactions of the British Mycological Society. 74: 56I-566.
Papavizas, G.C., 1985. Trichoderma and Gliocladium. Biology, ecology and potential for biological control. Annual Review of Phytopathology. 23: 23-54'
Papavizas, G.C., 1987. Genetic manipulation to improve the effectiveness of biocontrol
fungi for plant disease control. In: Innovative Approaches to Plant Disease Control. (ed. I. Chet). John Wiley and Sons, New York, USA'
Parker, D.M., W. Koller, 1998. Cutinase and other lipolytic esterases protect bean leaves from infection by Rhizoctonia solani. Molecular Plant-Microbe Interactions. 11: 5t4-522.
Pascoe, 1., lggg. Grapevine trunk diseases- black goo decline, esca, eutypa dieback and
others. The Australian Grapegrower and Winemaker ' 429:24-28.
Pascoe, I., E. Cottral, 2000. Developments in grapevine trunk diseases research in Australia. Phytopathologia Mediterranea. 39 : 68-7 5.
Pearce, R.8., P.J. Holloway, 1984. Suberin in the sapwood of oak (Quercus robur L.), its composition from a compartmentalization barrier and its occuffence in tyloses in
undecayed wood. Physiological Plant Pathology' 24: 7l-81.
Pearce, R.8., J.P. Ride, 1980. Specification of induction of the lignification response in wounded leaves. Physiological Plant Pathology' t6: 197-204.
Pearce, R.8., J. Rutherford, 198L. A wound-associated suberized barrier to the spread of decay in sapwood of oak (Quercus robur L.). Physiological Plant Pathology' 19: 359- 369. 207
Pearse, A.G.E., 1960. Histochemistry, Theoretical and Applied. J. & A. Churchill Ltd, London, UK.
Pearson, R.C., 1980. Discharge of ascospores of Eutypa armeniaca¿ in New York. Plant
Disease. 64: 17L-114.
Pearson, R.C., 1982. Protection of grapevine pruning wounds from infection by Eutypa
arn'teniacae in New York. American Journal of Enology and Viticulture' 33: 51-53'
Pearson, R.C., A.C. Goheen, L988. Compendium of Grape Diseases. American Phytopathological Society Press, St, Paul, Minnesota, USA.
péros, J.-P., G. Berger, lgg4. A rapid method to assess the aggressivenes of E. lata isolates and the susceptibility of grapevine cultivars to eutypa dieback. Agronomie. 14: 5t5-523
Péros, J.-P., G. Berger, F. Lahogte, 1997. Variation in pathogenicity and genetic structure in the Eutypa lata population of a single vineyard. Phytopathology. 87: 799-
806.
péros, J.-P., G. Berger, 1999. Diversity within natural progenies of the grapevine dieback fungus Eutypa lata. Cunent Genetics. 36: 301-309.
Péros, J.-P., I. Jamaux-Despreaux, G. Berger, I). Gerba, 1999. The potential importance of diversity in Eutypa lata and co-colonising fungi in development of grapevine dieback. Mycological Research' 103: 1385-1390.
Petzoldt, C.H., W.J. Moller, M.A. Sall, 1981. Eutypa dieback of grapevine: Seasonal differences in infection and duration of susceptibility of pruning wounds
Phytopathology. 7 1 : 540-543. 208
Pottle, \ry.H., A.L. Shigo, R.O. Blanchard, 1977. Biological control of wound hymenomycetes by Trichoderma harzianum. Plant Disease Reporter. 6l:687-690.
Powell, K.4., J.R. Faull, 1991.. Commercial approaches to the use of biological control
agents. In: Biotechnology of Fungi for Improving Plant Growth. (ed. J. M. Whipps & R. Lumsden). Cambridge University Press, Cambridge, UK.
Pratt, J.F,., 1982. Fomes annostls butt rot of Sitka spruce: Observations of the distribution of Cytosporiopsis abietina (Petrak) in the stems of rotted trees. Forestry. 55:
183.
Ramos, 8.8., W.J. Moller, H. English,l975a. Susceptibility of apricot pruning wounds to infection by Eutypa armeniacae. Phytopathology. 65: 1359-1364.
Ramos, E,.E,., \ry.J. Moller, H. EnglishrlgTSb. Production and dispersal of ascospores of Eutypa armeniacae in California. Phytopathology' 65: t364-137t.
Ricard, J.L., 1983. Field observation on the biocontrol of Dutch elm disease with Trichoderma viride pellets. European Journal of Forest Pathology. 13: 60-62.
Ride, J,P., 1975. Lignification in wounded wheat leaves in response to fungi and its possible role in resistance. Physiological Plant Pathology. 5: L25-I34.
Ride, J.P., 1978. The role of cell wall alterations in resistance to fungi. Annals of Applied Biology. 89: 302-306.
Ride, J.P., 1980. The effect of induced lignification on the resistance of wheat cell walls to fungal degradation. Physiological Plant Pathology' 16: I87-196' 209
Ride, J.P., R.B. Pearce, 1979. Lignification and papilla formation at sites of attempted penetration of wheat leaves by non-pathogenic fungi. Physiological Plant Pathology. 15:
79-92.
Rishbeth, J., 1963. Stump protection against Fomes annosus. IlL lnoculation with Peniophora gigantea. Annals of Applied Biology' 52:63-77.
Rishbeth, J.r1973. Problems involved in monoculture and intensive forest management. Second International Congress of Plant Pathology, Vol. II. p. 839 (abstr.).
Robb, J., J.D. Brisson, L. Busch, B.C. Lu, L979. Ultrastructure of wilt syndrome caused by Verticillium dahliae. vii. Correlated light and transmission elecffon microscope identification of vessel coatings and tyloses. Canadian Journal of Botany. 57: 822-834.
Robb, J., A. Smith, L. Busch, 1982, Wilts causedby Verticillium species. A cytological survey of vascular alterations in leaves. Canadian Journal of Botany. 60 825-837 .
Rolshausen, P.8., W.D. Gubler, 1998. A new assay for screening fungicides for control of eutypa dieback of grapevine. Phytopathology. 88: 576
Rosen, D., M. Edelman, E. Galun, D. Danon, 1974. Biogenesis of mitochondria in Trichoderma viride: Structural changes in mitochondria and other spore constituents during conidium maturation and germination. Journal of General Microbiology. 83: 31- 49.
Rozsnyay, 2.D., M. Hevesi, Z. Klement, L. Vajna, 199L. Biological control against canker and dieback diseases of apricots. Acta Horticulturae. 293:525-532'
Rumbos, I.C., 1987. Effects of pathotypes and time of year on the susceptibility of apricot to the infection by Eutypa lata.Phytopathologia Mediterranea.2l:85-90. 2ro
Ruzin, S.E., 1999. Plant microtechnique and microscopy. Oxford University Press, New York, USA.
Samuel, G., L933. Untitled report. Journal of the Department of Agriculture of South Australia. 36: 979-980.
Samuel, G.J., 1996 . Trichoderma; a review of biology and systematics of the genus. Mycological Research. lO0: 923 -935.
Scheffer, R.J., 1983. Biological control of Dutch elm disease by Pseudomonøs species. Annals of Applied Biology. 103:21-30.
Schirmböck, M., M. Lorito, Y.-L. Wang, C.K. Hayes,I. Arisan'Atac, F. Scala' G.E. Harman, C.P. Kubicek,1994. Parallel formation and synergism of hydrolytic enzymes and peptaibol antibiotics, molecular mechanisms involved in the antagonistic action of Trichoderma harzianum against phytopathogenic fungi. Applied and Environmental Microbiology. 60: 4364-4370.
Schlick, 4., K. Kuhls, W. Meyer, E. Lieckfeldt, T. Börner, K. Messner, 1994. Fingerprinting reveals gaûìma-ray induced mutations in fungal DNA: implication for identification of patent strains of Trichoderma harzianum. Canent Genetics .26:74-78.
Schmidt, C.S., D. Lorenz, G.A. Wolf, 2001a. Biological control of the gtapevine dieback fungus Eutypa lata l'. screening bacterial ahtagonists. Journal of Phytopathology. 149:427-437.
Schmidt, C.S., D. Lorenz, G.A. Wolf' J. Jäger, 2001b. Biological control of the grapevine dieback fungus Eutypa lata II: Influence of formulation additives and transposon mutagenesis on the antagonistic activity of Bacillus subtilis and Erwinia herbicola. Journal of Phytopathology' I49 437-446. 2rl
Shain, 1., lg7g. Dynamic responses of differentiated sapwood to injury and infection.
Phytopathology. 69: 1143-1147 .
Showalter, 4.W., lgg3, Structure and function of plant cell wall proteins. The Plant Cell.5:9-23.
Simard, M., D. Rioux, G. Lalflamme, 2001. Formation of ligno-suberized tissues in jack pine resistant to the European race of Gremmeniella abietina' Phytopathology' 91: tt28-1r40.
Simon,4., R.\ry. Dunlop, E.L. Ghisalberti, K. Sivasithamparam' 1988' Trichoderma koningü produces a pyrone compound with antibiotic properties. Soil Biology and
Biochemistr y . 20 263 -264.
Sivan, 4., I. Chet, 1982. Biological control of Pythium by Trichoderma'
Phytoparasitica. 10: 1 18 (abstr.)'
Sivan, 4., I. Chet, 1986. Biological control of Fusarium spp.in cotton, wheat and muskmelon by Trichoderma harzianum. Journal of Phytopathology' 116:39-47 '
Sivan, 4., I. Chet, 1989. The possible role of competition between Trichoderma harzianum and Fusarium oxysporum on rhizosphere colonization. Phytopathology. 79:
198-203
Sivan, 4., y. Elad, I. Chet, 1984. Biological control effects of a new isolate of Trichoderma harzianum on Pythium aphanidermatum. Phytopathology . 7 4: 498-501 .
Smith, K.T., R.O. Blanchard, W.C. Shortle, 1979. Effect of spofe load of Trichoderma harzianum oî wood-invading fungi and volume of discoloured wood associated with Acer rubrum. Plant Disease Reporter.63: l07O-lO7l' 2t2
Smith, K.T., R.O. Blanchard, W.C. Shortle, 1981. Postulated mechanism of biological control of decay fungi in red maple wounds treated with Trichoderma harzianum.
Phytopathology. 7 1 : 496-498.
Smith, V.L., W.F. Wilcox, G.E. Harman, 1990. Potential for biological control of Phytophthora root and crown rots of apple by Trichoderma and Gliocladium spp. Phytopathology. 80: 880-885.
Smith, K.P., M.J. Havey, J. Handelsman, 1993. Suppression of cottony leak of cucumber with Bacillus cereus strain UW85. Plant Disease. TT; 139-142.
Southerton, S.G., B.J. Deverall, 1989. Histological studies of the expression of Lr 9, Lr 20 and Lr 28 alleles for resistance to leaf rust fungi in wheat. Physiological Plant Pathology. 36: 483-494.
Southerton, S.G., B.J. Deverall, L990. Histological and chemical evidence for lignin accumulation during the expression of resistance to leaf rust fungi in wheat.
Physiological and Molecular Plant Pathology. 36: 483-494.
Sparapano, L., G. Bruno, A. Graniti, 2000. Infection of grapevines by some fungi associated with esca. L Fomitiporia punctata as a wood-rot inducer. Phytopathologia
Mediterran ea. 39 : 46-52.
Spiers, 4.G., D.T. Brewster, 1997. Evaluation of chemical and biological treatments for control of Chondrostereum purpureum infection of pruning wounds in willows, apples and peaches. New zealand Journal of Horticulture Science.25: 19-31.
Spiro, R.G., 1970. Glycoproteins. Annual Review of Biochemistry. 39: 599-638. 213
Stabb,8.V., L.M. Jacobson, J. Handelsman,lgg4.Zwittermycin A producing strains of Bacillus cereus from diverse soils. Applied and Environmental Microbiology' 68: 4404-4412.
Steiner, G.W., J.L. Lockwood, 1969. Soil fungistasis: sensitivity of spores in relation to germination time and size. Phytopathology. 59: 1084-1092'
Stockwell, V., P. Hasnchey, L983. The role of the cuticle resistance to Rhizoctonia
s olani. Phytopathology . 7 3 : I 640 - 1642.
Strashnov, Y., Y. Elad, A. Sivan, Y. Rudich, I. Chet, 1985. Control of Rhizoctonia solani fruit rot of tomatoes by Trichoderma harzianum. Crop Protection. 4:359-364.
Sutton, J.C., G. Peng, 1993. Biocontrol of Botrytis cinerea in strawberry leaves
Phytopathology. 83 : 615-621.
Talmadge, K.W., K. Keegstra, W.D. Bauer, P. Albersheim, 1973. The structure of plants cell walls. L The macromolecular components of the walls of suspension-cultured
sycamore cells with a detailed analysis of the pectic polysaccharides. Plant Physiology' 5l 258-173
Tey-Rulh, P., I. Philippe, J. Renaud, G. Tsoupras, P. de anglis, J. Fallot' R. Tabacchi, lggl. Eutypine, a phytotoxin producedby Eutypa lata, the causal agent of
dying-arm disease of grapevine. Phytochemistry. 30: 47 I -47 3 .
Toole, 8.R., 1971. Interaction of mold and decay fungi on wood in laboratory tests.
Phytopathology. 61 : t24-125.
Trese, 4.T., C.L. Burton, D.C. Ramsdell, L980. Eutypa armeniacae in Michigan vineyards: Ascospore production and survival, host infection and fungal growth at low
temperatures. Phytopathology. 7 O: 7 88-7 93. 214
Trese, 4.T., C.L. Burton, D.C. Ramsdell, 1982. Effects of winter and spring pruning and post inoculation cold weather on infection of grapevine by Eutypa armeniacae
Phytopathology. 7 2: 438-440.
Tronsmo, 4., 1986. Use of Trichoderma spp. in biological control of necrotropic pathogens. In: Microbiology of the Phyllosphere. (ed. N. J. Fokkema &. J' Van de Heuvel). Cambridge University Press, Cambridge, UK'
Tronsmo, 4., 1989. Effect of fungicides on growth of Botrytis cinerea, Trichoderma viride andTrichoderma harzianum. Norwegian Journal of Agriculture Sciences' 3: 151-
156.
Tronsmo, 4., C. Dennis, 1977. The use of Trichoderma species to control strawberry fruit rots. Netherlands Journal of Plant Pathology. 83: 449-455.
Tronsmo, 4., C. Dennis, 1978. Effect of temperature on antagonistic properties of Trichoderma species. Transactions of the British Mycological Society. 71:469-414.
Tronsmo, 4., G.L. Hjeljord, 1998. Biological control with Trichoderma species. In: Plant-Microbe Interactions and Biological Control. (ed. G. J. Boland &. D' L' Kuykendall). Marcel Dekker,Inc., New York, USA.
Tronsmo, 4., J. Raa, 1977. Antagonistic action of Trichoderma pseudokoningii against the apple pathogen Botrytis cinerea. Phytopathology Zeitschrift. 89: 216-220'
Tronsmo, 4., J. Ystaas, 1980. Biological control of Botrytis cinerea on apple. Plant Disease. 64: 1009.
Trottin, C.Y., C. Fournier, J.M. Leyre, P. Nicot, V' Decognet' M' Bardin, C' Romitic, 2001. Protected tomato: Biological control of Botrytis cinerea and Oidium 2t5 neolycopersici. Pépiniéristes Horticulteuurs Maraîchers. Revue Horticole. 46 (Supplement): xiii-xv.
Trutman, P., P.J. Keane, 1990. Trichoderma koningii as a biological control agent for Sclerotinia sclerotiorum in Southern Australia. Soil Biology and Biochemistry. 22: 43-
50.
Vajna, L., 1986. Fusarium lateritium Nees. Ex. Link as a parasite and host in inter- fungal hyphal interaction. Journal of Phytopathology: 157-L64.
Vance, C.P., T.K. Kirk, R.T. Sherwood, 1980. Lignification as a mechanism of disease resistance. Annual Review of Phytopathology. 18: 259 -288.
Vaughn, S.F., E.C. Lulai, l99la. The involvement of mechanical barriers in the resistance response of a field-resistant and a field-susceptible potato cultivar to Verticillium dahliae. Physiological and Molecular Plant Pathology. 38: 455-465.
Vaughn, S.F., E.C. Lulai, 1991b. Comparison of fluorescent stains for the detection of suberin in potato periderm. American Potato Journal. 68:667-674'
Washington, N.S., S. Engleitner, G. Boontjes, N. Shanmuganathan, 1999. Effect of fungicides, seaweed extracts, tea tree oil and fungal agents on fruit rot and yield in strawberry. Australian Journal of Experimental Agriculture. 39 : 487 -494.
Weindling, R., 1932. Trichoderma lignorum as a parasite of other soil fungi
Phytopathology. 22: 837 -845.
Weindling, R., 194L. Experimental consideration of the mold toxins of Gliocladium and Trichodermø. Phytopathology. 31: 991-1003. 216
Wells, H.D., D.K. Bell, C.A. Jawarsk, 1972. Efficacy of Trichoderma harzianurÍ7 as a biological control agent for Sclerotium rolfsii. Phytopathology.62: 442-447 '
Whipps, J.M., 1987. Behaviour of fungi antagonistic to Sclerotinia sclerotiorum on plant tissue segments. Journal of General Microbiology.133: 1495-1501.
Wicks, T., K. Davies, 1999. The effect of Eutypa on yield. The Australian Grapegrower and Winemaker. 406a: 15-16.
Wicks, T.J., B. Hall, 1997. Eutypa dieback, a serious disease. The Australian Grapegrower and Winemaker. 405: 6l-62.
Wicks, T.J., D. Volle, T.C. Lee, 1983. Effect of fungicides in infection of apricot and cherry pruning wounds inoculated with Chondrostereum purpureum. Australian Journal of Experimental Agriculture and Animal Husbandry . 23 : 9 I -94.
Wilcox, W.F., G.E. Harman, D. Pietro, 1992. Effects of gliotoxin on growth, sporulation and zoospore motility of seven Phytophthora spp. in vitro. Phytopathology. 82: lI2L
Wood, R.K.S., 1951. The control of diseases of lettuce by the use of antagonistic organisms. 1. The control of Botrytis cinerea Pers. Annals of Applied Biology' 38:203- 216.
Wright, J.M., 1952. Production of gliotoxin in unsterilized soil. Nature. I70:673-614
Zimand, G., Y. Elad, I. Chet, 1996. Effect of Trichodenna harzianunt on Botrytis cinerea pathogenicity. Phytopathology. 86: 125 5 - 126O. 217
APPENDIX 1
Preparation of dilution series
1. 10 mg of Trichoseal spray@ /Vinevax@ was suspended in 100 ml of
sterile diluent and dissolved by shaking for 1 min.------Solution A
2. 1 ml from solution A was transferred to 99 ml sterile diluent------
Solution B (Dilution factor - 10')
3. 1 ml of Solution B was added to 49 ml of sterile diluent
Dilution 1 (Dilution factor - 5 x 103)
4. 1 ml of Solution B was added to 99 ml of sterile diluent
Dilution 2 (Dilution factor - 104)
5. 10 ml of Dilution 2 was added to 15 ml of sterile diluent
Dilution 3 (Dilution factor = 2.5 x 104)
6. 10 ml of Dilution 2 was added to 40 ml of sterile diluent-----
Dilution 4 (Dilution factor = 5 x 104)
1. 10 ml of Dilution 2 was added to 65 ml of sterile diluent-----
Dilution 5 (Dilution factor - 7.5 x 104)
8. 10 ml of Dilution 2 was added to 90 ml of sterile diluent-----
Dilution 6 (Dilution factor - 10s)
9. 10 ml of Dilution 6 was added to 15 ml of sterile diluent-----
Dilution 7 (Dilution factor - 2.5 x 10s)
10. 10 ml of Dilution 6 was added to 40 ml of sterile diluent-----
Dilution 8 (Dilution factor - 5 x 10s) 2t8
11. 10 ml of Dilution 6 was added to 65 ml of sterile diluent-----
Dilution 9 (Ditution factor - 7.5 x 10s)
12.10 ml of Dilution 6 was added to 90 ml of sterile diluent-----
Dilution 10 (Dilution factor = 106)
The dilution series was prepared aseptically. 219
APPENDIX 2
Preparation of Potato Dextrose Agar (PDA)
39 g of PDA (Difco) was dissolved in 1 L of RO water and autoclaved at lzI"C for 2O min.
Preparation of Eutypa Selective Medium (EUSM)
PDA prepared as described above was amended with 100 ¡lglml of streptomycin sulfate,
50 p{m\ of chlortetracycline HCI and 5 ¡tglml of dicloran (Munkvold and Marois,
1993a).
Preparation of Acidified Potato Dextrose Agar (APDA)
A drop of 85.6Vo lactic acid was placed in each plate with a sterile 1ml pipette before
PDA was poured.
Preparation of Czapek Dox Agar (CDA)
50 g of Czapek Dox liquid medium compound (Oxoid) and 2.5 g of Bitek Agar (Difco) were dissolved in 1 L of RO water and autoclaved at l2l"C for 20 min.
Preparationof 2Vo Water Agar (WA)
2O g of Bitek Agar (Difco) was dissolved in 1 L of RO water and autoclaved at 121"C for
20 min. 220
Preparation of Phosphate Buffered Saline (PBS) (0.06M' PIJ=7.2)
8.5 g of NaCl, 0.7 g of NazHPO¿ and 0.16 g of KHzPO¿ were dissolved in 1 L of RO water 22r
APPENDIX 3
Experiment 4: Number of canes, of eight, yielding E. lata 12 weeks after the antagonist.
E. lata was introduced 0, 2 and,7 days after the antagonist and the canes were harvested
12 weeks later.
Number of canes infected bY E. lata
Day of T. harzianum in T. harzianum Gamma-irradiated SDW E.lata Trichoseal@ base (10e spores/ml) T. hørzíanum in inocula (100 s/L) Trichoseal@ base -tion (100 s/L)
0 8 8 J 8
2 I 2 8 8 8 1 2 1 8