Abstract

Stem end rot of kiwifruit caused by Botrytis cinerea has been responsible for significant postharvest storage losses. Current disease control strategies in New Zealand orchards are based upon preharvest applications of dicarboximide fungicide. However, the emergence of fungicide resistance in Botrytis populations and increasing consumer demands for reduced pesticide residues on fruit has highlighted the need for alternative disease control strategies. Biological control of Botrytis in kiwifruit was investigated in field studies in 1991/92. An isolate of Trichoderma viride (RH strain) applied to staminate kiwifruit flowers significantly reduced the incidence of Botrytis on such flowers from 73% to 55%. The level of control was similar to that achieved with the fungicide, iprodione (47%). Botrytis contamination of the hairy external fruit surface at harvest was identified in epidemiology studies as the primary source of inoculum for fruit infection. Applications of T. viride and iprodione applied to fruit prior to harvest significantly reduced the number of viable Botrytis propagules on the fruit surface from 1354 to 45. Bacillus subtilis and T. harzianum (isolate T39) did not reduce the incidence of Botrytis flower infection in the spring or the number of Botrytis propagules on the fruit surface at harvest. The preharvest application of iprodione was the only treatment that significantly reduced both the number of viable Botrytis propagules on the fruit surface at harvest and the incidence of stem end rot in coolstorage. A critical examination of biological control in kiwifruit highlighted the need for a much greater understanding of the ecology and epidemiology of Botrytis in kiwifruit blocks. Strategies to improve the selection and timing of applications of biological control agents (BCA’s) in kiwifruit canopies in the future are discussed.

Introduction

Stem end rot of kiwifruit (Actinidia deliciosa (A. Chev.) C.F Liang et A.R. Ferguson cv. ‘Hayward’) caused by B. cinerea has been responsible for significant postharvest storage losses. In 1994, the cost of these losses was estimated by the New Zealand kiwifruit industry to be $7 million (Anon 1994). Current disease control strategies are based upon applications of benzimidazole and dicarboximide fungicides, but resistance to both chemical groups has now been reported (Pak 1994). The emergence of fungicide resistance and increasing consumer demands for reduced residues on fruit emphasises the need for alternative disease control strategies.

Biological control, as an alternative strategy, has the advantage of greater public acceptance and reduced environmental contamination (R.A. Hill pers. comm.). The use of biological control agents (BCA’s) to control plant diseases has been investigated by researchers in a wide variety of host/pathogen systems. Unfortunately, success in the laboratory has rarely led to successful implementation in the field (Wilson et al. 1991). Biological control of Botrytis was reported over forty years ago (Wood 1951; Newhook 1957) and in the last decade the majority of research has concentrated on the Trichoderma spp. (Dubos 1992; Gullino 1992). Biological control of Botrytis on kiwifruit tissues was first reported by Menzies et al. (1989), but the results were variable. In laboratory studies, the bacterium Bacillus subtilis successfully inhibited Botrytis colonisation of staminate kiwifruit flowers (Franicevic 1993). Postharvest applications of isolates of T. viride and Cladosporium cladosporoides to kiwifruit picking wounds significantly reduced Botrytis stem end rot (Harvey et al. 1991). Complete control of Botrytis stem end rot was achieved by applying a selected isolate of T. viride to the kiwifruit picking wound immediately after fruit harvest (Hill 1992). In Italy, postharvest applications of a yeast species to freshly harvested fruit significantly increased the proportion of healthy kiwifruit after artificial inoculation with Botrytis and coolstorage for 15 days (Testoni et al. 1993).

Botrytis, like all post-harvest diseases, has a preharvest component and studies have indicated that a reduction in the size of Botrytis populations in vines during the growing season reduced Botrytis development and Botrytis stem end rot in coolstorage. The objective of this research was to evaluate the efficacy of selected BCA’s to reduce Botrytis colonisation of senescing floral tissues and reduce the level of Botrytis contamination on the fruit surface at harvest. In the second part of this paper we report on parallel epidemiological studies and discuss how this knowledge will be used to devise a new biological control strategy for Botrytis in kiwifruit.

Material and Method

Field application of BCA’s (1991)

Two Trichoderma species and an isolate of B. subtilis, with reported activity against Botrytis, were selected for evaluation in kiwifruit vines (Table 1).

A field experiment was established at the Hawkes Bay Research Centre, Hastings, in a block of kiwifruit (cv. ‘Hayward’) trained on the pergola growing system. Experimental plots were single female vines (canopy area=12.5 m²) and were separated by buffer plots of single untreated female vines within rows and one metre of male vine foliage between rows. Treatments were arranged in a randomised block design with four replicate blocks and treatment means were compared with the water controls using a t-test.

BCA preparation

Trichoderma harzianum (Trichodex, 1x10⁷ spores/gm as a wettable powder) was supplied by NuFarm New Zealand Ltd. The T. viride isolate was derived from 10-14 day old cultures grown on malt dextrose agar plates incubated at laboratory temperature (20°C ±2°C). A suspension of B. subtilis was prepared from a Nutrient Yeast Dextrose Broth culture which had been continuously shaken on an orbital shaker and incubated at 30°C ±2°C for four days.

Table 1: Source of BCA’s and application rates to kiwifruit vines (1991/92)

Treatments were applied with a handheld pressurized pump (5L Cambrian sprayer), when 60% of the female flowers were open and just prior to harvest. Prior to vine treatment, water was applied to all experimental plots via overhead sprinklers to increase the relative humidity and assist BCA survival. Iprodione (Rovral™ FLO, 250g/litre) was applied at 1.5ml/litre. Three days after the treatment applications, a Botrytis spore suspension, derived from malt extract agar plates was adjusted to 5 x 10⁵ spores/ml and applied (flowering period only) with a handheld pressurized pump (Cambrian) so that all floral surfaces were wetted. To avoid potential desiccation in the canopy, all treatments were applied in the evening and after vines had been thoroughly watered as described above.

Petal fall assessment

Forty eight hours after Botrytis application, forty fruitlets (1.5cm diameter) with senescent floral tissue (petals and anthers) were collected at random at the petal fall stage of development and incubated at 22°C (±3°C) in humidity chambers for five days to simulate disease conducive conditions. Floral tissues were assessed visually for the incidence and severity of Botrytis sporulation using a sporulation severity score (1 = nil sporulation, 2 = slight, 3 = moderate, 4 = profuse, 5 = complete sporulation over the entire floral surface).

A mesh bag containing 20, 1cm thick slices of kiwifruit with profuse Botrytis sporulation was suspended in the canopy during the growing season to provide Botrytis inoculum at the centre of each plot.

Harvest Assessment

The level of Botrytis contamination of the fruit surfaces at harvest was determined by sampling, at random, ten fruit from each plot. Each fruit was individually washed in 10 ml sterile distilled water plus 0.05% Tween 20, and a 0.1 ml aliquot was plated onto each of four replicate plates of a Botrytis selective medium (Kerssies 1990). Plates were incubated at room temperature (22°C ±2°C) for 12 days and the results were expressed as the number of viable Botrytis propagules/fruit. At harvest maturity (soluble solids content=6.2%), 100 fruit per plot were sampled at random then jostled for 20 seconds in clean dry canvas picking bags to simulate standard commercial practice. Fruit were then packed into standard export kiwifruit trays (#36 count), sealed with plastic liners, coolstored at 0°C and the incidence of stem end rot assessed visually after eight and 13 weeks coolstorage.

Epidemiology studies (1991-1994)

In the first year of this investigation, prunings on the ground (prunings), senescing male flowers, senescing petals adhering to fruitlets (attached petals), wind-blown senescent shoots in the vine (blowouts) and green leaves with necrosis (necrotic leaves) were identified as potential sources of Botrytis inoculum (Elmer et al. 1993). In two kiwifruit orchards in the Motueka region, subsamples of all potential sources of Botrytis inoculum were collected at random from ten female plots (two vines/plot) per orchard at key growth stages during the growing season. Host tissues were then incubated in high humidity chambers for five days and the area of sporulation measured. The number of spores from host tissues was estimated from regression equations described in Elmer et al. (1993). The number of spores produced on all host tissues sampled within plots was a measure of the size of the Botrytis population and was expressed as total potential spore production (TPSP/plot). From the same experimental plots, the number of viable Botrytis spores contaminating the fruit surface at harvest was also determined on three sets of five fruit of similar size. External contamination of the fruit surface was determined as described in Elmer et al. (1995).

In the 1994 growing season, the effect of removal of Botrytis inoculum from female vines during the growing season on Botrytis development was assessed. Details of the methods were published in Elmer et al. (1995). Data were subjected to analysis of variance and means were compared using a t test (1991-92 data) and the least significant difference (LSD) test (1993-94 data).

Results

Biological control

The background level of flower infection in the uninoculated plots was 26% (Table 2). In the absence of chemical or BCA treatment, the incidence of flower infection was high (73%) after flowers were sprayed with an aqueous suspension of Botrytis (water control). In comparison to the water control, applications of iprodione and T. viride significantly (P<0.05) reduced the incidence of Botrytis on senescent floral tissues when applied prior to pathogen inoculation (Table 2). Applications of B. subtilis and T. harzianum did not reduce the incidence or severity of Botrytis colonisation on floral tissues. Iprodione was the only treatment that significantly reduced both the incidence (P<0.05) and severity (P<0.01) of Botrytis.

Table 2: Effect of biological control agents on the incidence and severity of Botrytis infection of female kiwifruit flowers - Hawkes Bay, 1991

Botrytis inoculum contaminating the fruit surface at harvest was identified as an important primary source of inoculum for infection of the picking wound during the harvesting process (Elmer et al. 1993; 1995). Preharvest T. viride and iprodione treatments significantly (P<0.05) reduced the number of viable Botrytis propagules contaminating the external fruit surfaces at harvest, but only iprodione significantly (P<0.05) reduced the incidence of stem end rot (Table 3).

Epidemiology Studies (1991-1994)

Source of inoculum
In 1991/92 growing season several potential sources of Botrytis inoculum were identified in kiwifruit orchards (Elmer et al. unpublished). In more detailed studies (1993), potential sources of Botrytis inoculum were quantified in two kiwifruit blocks at key growth stages during the growing season (Elmer et al. 1994). Early in the season (first female flower opening), the Botrytis population in experimental blocks was low and the primary source of inoculum was prunings on the ground. At full bloom, the Botrytis population increased slightly and senescing male flowers were the primary source of inoculum.

Table 3: Effect of preharvest BCA applications on external fruit contamination by Botrytis at harvest and stem end rot in coolstorage (1991/92)

By petal fall, the Botrytis population increased significantly (P<0.05) and this coincided with an abundance of senescing floral tissues at that time. By the mid-fruit assessment, the Botrytis populations declined significantly (P<0.05) and windblown senescent shoots were the primary source of inoculum. At harvest, green leaves with necrosis were the primary source of Botrytis inoculum. There was a significant (P<0.05) relationship (correlation coefficient=0.55) between the size of the Botrytis population in individual plots and the number of viable Botrytis propagules contaminating the external fruit surfaces. A similar result was also detected in later studies in 1993/94. In a separate study in 1993, kiwifruit orchards with high numbers of viable Botrytis spores (mean=10,500/fruit) on the fruit surface at harvest had significantly (P<0.05) more stem end rot in coolstorage (3%) compared to fruit from orchards where the number of viable Botrytis propagules on the fruit surfaces at harvest was low (mean=1,900/fruit, stem end rot incidence=0.4%).

Table 4: The effect of removal of Botrytis inoculum on subsequent Botrytis development - Nelson, 1994

The number of viable Botrytis propagules on the fruit surface at harvest was significantly (P<0.05) greater on fruit sampled from plots where sources of Botrytis inoculum were left in the kiwifruit canopy, compared to plots where inoculum sources were removed at regular intervals (Table 4). In addition, stem end rot incidence was also significantly (P<0.05) lower in the plots where Botrytis inoculum was removed compared to the control treatment.

Discussion

These results indicate that applications of T. viride (RH strain) reduced the incidence of Botrytis at flowering and the number of viable Botrytis propagules on the fruit surface at harvest. The level of Botrytis control was similar to that achieved with a standard recommended fungicide (iprodione). The preharvest fungicide application was the only treatment which significantly reduced the incidence of stem end rot in this experiment in this season. The lack of control provided by the T39 isolate of T. harzianum in this study was in contrast to reports of significant reductions of Botrytis in other host systems (Dik et al. 1995; Y. Elad pers. comm.). Several factors may account for the results obtained with T39 in the kiwifruit system. The concentration of T39 applied (3.2x10⁵ spores/ml) in these studies (1991) was low relative to that which is now recommended as a result of later investigations (Elad et al., 1993) and the concentration of T39 on susceptible host tissues may be critical for successful suppression of Botrytis. Variable control of Botrytis with Trichoderma spp. in the Botrytis/grape system was explained because of the poor survival of Trichoderma spp. in the grape phylloplane in the warmest part of the growing season (Gullino et al. 1995). Low temperatures (<18°C) were also a factor in the reduced efficacy of T39 in some situations (Elad pers. comm.). Applications of B. subtilis did not reduce Botrytis flower infection, the number of viable Botrytis propagules on the fruit surface or stem end rot incidence. Bacterial survival on the phylloplane is very sensitive to environmental fluctuations (Knudsen and Hudler 1987) and B. subtilis desiccation on tissue surfaces in the canopy may have occurred.

Epidemiology studies carried out in parallel with the BCA experiments suggested there was a need to improve both the type of BCA selected in screening experiments and the timing of BCA applications. As a result of these and other investigations it is clear that low Botrytis populations in the canopy reduces the probability of Botrytis contamination of the picking wound and subsequent stem end rot in coolstorage. This is supported by research which showed significant (P<0.001) correlations (r=0.832) between stem end rot incidence and the number of Botrytis propagules artificially applied to fruit surfaces at harvest (Elmer et al. 1995). When suppression of Botrytis sporulation was simulated by the removal of necrotic tissues, fruit contamination at harvest and stem end rot incidence were significantly reduced. Therefore, we conclude that a new strategy for biological control in the Botrytis/kiwifruit system is required. In 1992 and 1993 Botrytis epidemics developed rapidly during the petal fall growth stage when conditions were favourable for Botrytis development. This increase coincided with an abundance of senescing floral tissues at that time. Selection of BCA’s in the future should concentrate on the ability of the antagonist to rapidly colonise senescing floral tissues and suppress Botrytis sporulation. After petal fall, there is often an abundance of senescent and necrotic tissues in the canopy in the form of wind-blown senescent shoots. These tissues have the capacity to act as an "inoculum bridge" where Botrytis survives as saprophytic mycelium within necrotic tissue, thereby escaping desiccation during the summer months. When conditions are favourable for Botrytis sporulation these tissues provide inoculum for infection of senescing leaves and new necrotic tissues in the canopy. If there are sufficient quantities of these tissues in the canopy then Botrytis populations have the capacity to build-up rapidly in the preharvest period (Pak and Manning 1994). BCA’s applied during fruit growth (post flowering) and in the preharvest period must have the ability to rapidly colonise necrotic leaf tissues, survive environmental fluctuations and suppress Botrytis sporulation.

The ability to survive interrupted wet periods was a major selection criterion for successful antagonism of Botrytis on necrotic host tissues exposed to field conditions (Köhl et al. 1993).

The suppression of Botrytis sporulation on necrotic tissues with antagonists capable of surviving interrupted wet periods represents a relatively new approach to biological control (Köhl et al. 1992; Sutton and Peng 1993). One major advantage of this strategy is the long interaction period between the pathogen and saprophytic antagonist (Fokkema 1993). Experimentation to screen saprophytes capable of surviving interrupted wet periods and suppressing Botrytis sporulation on kiwifruit tissues is now in progress.

The RH strain of T. viride significantly reduced the number of viable Botrytis propagules on the fruit surface at harvest but did not reduce the incidence of Botrytis stem end rot. A potent anti-fungal metabolite was isolated from the RH strain of T. viride and was identified as 6-pentyl-alpha-pyrone (6PAP). When 6AAP, a synthetic derivative of 6PAP, was applied to the picking wounds of freshly harvested fruit no stem end rot developed in coolstorage (Hill 1994). Both the synthetic 6AAP and naturally derived 6PAP extracted from Trichoderma spp. are already approved by the United States Federal Drug Administration (FDA) as food additives. A combination of preharvest suppression of Botrytis in the orchard and postharvest applications of 6AAP or 6PAP may allow kiwifruit producers in New Zealand a real opportunity to control Botrytis biologically without the need for fungicide applications.

Acknowledgements

We are grateful to the New Zealand Kiwifruit Marketing Board and the Foundation for Research Science and Technology for assistance with funding this programme. We also thank David Manktelow, Nick Grbavac, Monika Walter, Chris Morgan (technical assistance) and Dr Alison Stewart (editorial comments).

References

Anon. 1994. Botrytis fruit losses. KIWIFLYER: A publication of the New Zealand Kiwifruit Marketing Board Newsletter, October, 1994.

Dik, A.J., Köhl, J., Fokkema, N.J., Elad, Y. and Shtienberg, D. 1995. Biological control of Botrytis cinerea in cucumber and tomato. European Journal of Plant Pathology. 13th International Plant Protection Congress, Abstracts: 587.

Dubos, B. 1992. Biological control of Botrytis: state of the art. pp. 169-178 in Verhoeff, K., Malathrakis, N.E. and Williamson, B. (eds), Recent advances in Botrytis research. Pudoc Scientific Publishers, Wageningen.

Elad, Y., Zimand, G., Zaqs, Y., Zurial, S. and Chet, I. 1993. Use of Trichoderma harzianum in combination or alternation with fungicides to control cucumber grey mould (Botrytis cinerea) under commercial greenhouse conditions. Plant Pathology 42: 324-332.

Elmer, P.A.G., Boyd-Wilson, K.S.H., Cook, D.W., Gaunt, R.E., Frampton, C.M. and Pyke, N.B. 1993. Sources of Botrytis cinerea inoculum in kiwifruit orchards and the relationship between total inoculum potential, external fruit contamination and stem end rot of kiwifruit. Sixth International Plant Pathology Congress, Montreal, Canada, August 1993.

Elmer, P.A.G., Boyd-Wilson, K.S.H., Whelan, H.G., Pyke, N.B., Gaunt, R.E. and Frampton, C.M. 1994. Sources of Botrytis in kiwifruit vines and spore contamination of the fruit surface at harvest. Proceedings of the 5th National Research Conference of the New Zealand Kiwifruit Marketing Board 5: 19-21.

Elmer, P.A.G., Gaunt, R.E, Boyd-Wilson, K.S.H., Pyke, N.B. and Fermaud, M. 1995. External contamination of kiwifruit by Botrytis cinerea: An important source of inoculum for fruit infection. Proceedings of 48th Plant Protection Conference 48: 95-100.

Fokkema, N.A. 1993. Opportunities and problems of control of foliar pathogens with micro-organisms. Pesticide Science 37: 411-416.

Franicevic, S.C. 1993. Biological control of Botrytis cinerea and Sclerotinia sclerotiorum on kiwifruit. Ph.D Thesis, School of Biological Sciences, University of Auckland

Gullino, L.M. 1992. Control of Botrytis rot of grapes and vegetables with Trichoderma spp. pp. 125-132 in Tjamos et al (eds), Biological Control of Plant Diseases. Plenum Press, New York.

Gullino, L.M., Monchiero, M. and Garibaldi, A. 1995. Biological control of Botrytis cinerea of grapevine: Critical analysis of the results. IOBC WPRS Bulletin 18 (3): 84-86.

Harvey, I.C., Pyke, N.B., Elmer, P.A.G. and Balasubramanian, R. 1991. The biological control of Botrytis stem end rot of kiwifruit. 2nd International Kiwifruit Symposium, Palmerston North, New Zealand. (Poster).

Hill, R.A. 1992. Can kiwifruit diseases be controlled biologically? Proceedings of the 4th National Kiwifruit Conference (supplement) 4.

Hill, R.A. 1994. Novel disease control methods. Proceedings of the 5th National Research Conference of the New Zealand Kiwifruit Marketing Board. 5: 22-24.

Kerssies, A. 1990. A selective medium for Botrytis cinerea to be used in a spore trap. Netherlands Journal of Plant Pathology 96: 247-50.

Knudson, G.R. and Hudler, G.W. 1987. Use of a computer simulation model to evaluate a plant disease biocontrol agent. Ecological Modelling 35: 45-62.

Köhl, J., Molhoek, W.M.L., Van der Plas C.H., Kessel, G.J.T. and N.J. Fokkema 1992. Biological control of Botrytis leaf blight of onions: Significance of sporulation suppression. pp 192-196 in Verhoeff, K., Malathrakis, N.E. and Williamson, B. (eds), Recent Advances in Botrytis Research. Pudoc Scientific Publishers, Wageningen.

Köhl, J., Van der Plas, C.H., Molhoek, W.M.L. and Fokkema, N.J. 1993. Drought tolerance as a major selection criterium for antagonists of Botrytis spp. in Fokkema, N.J., Köhl, J. and Elad, Y. (eds). Biological Control of Foliar and Post-Harvest Diseases. Bulletin OILB/SROP 16 (11): 169-171.

Menzies, J.G., Kempler, C., Boland, G.J. and Inglis, G.D. 1989. Biological control of Botrytis cinerea on kiwifruit, 1988. Biological and Cultural Tests for Control of Plant Diseases. 4: 7.

Newhook, F.J. 1957. The relationship of saprophytic antagonism to control of Botrytis cinerea Pers. on tomatoes. New Zealand Journal of Science and Technology 38: 473-481.

Pak, H. 1994. Resistance of Botrytis to fungicides in kiwifruit orchards. 5th National Research Conference of the New Zealand Kiwifruit Marketing Board. Rotorua, July 1994. (Poster).

Pak, H.A. and Manning, M.A. 1994. Predicting Botrytis storage rots. Proceedings of the 5th National Research Conference of the New Zealand Kiwifruit Marketing Board. 5: 16-18.

Testoni, A., Aloi, C., Mocioni, M. and Gullino, M.L. 1993. Biological control of Botrytis rot of kiwi fruit. in Fokkema, N.J., Köhl, J. and Elad, Y. (eds). Biological Control of Foliar and PostHarvest Diseases. Bulletin OILB/SROP 16 (11): 95-98.

Sutton, J.C. and Peng, G. 1993. Biocontrol of Botrytis cinerea in strawberry leaves. Phytopathology 83: 615-621.

Wilson, C.L., Wisniewski, M.E., Biles, C.L., McLaughlin,R., Chalutz, E. and Droby, S. 1991. Biological control of postharvest diseases of fruits and vegetables: Alternatives to synthetic fungicides. Crop Protection 10: 172-177.

Wood, R.K.S. 1951. The control of diseases of lettuce by the use of antagonistic organisms. I. The control of Botrytis cinerea Pers. Annals of Applied Biology 38: 203-216.