© 2007 Plant Management Network.
Addressing the Gaps in our Knowledge of Grapevine Downy Mildew for Improved Forecasting and Management
Megan M. Kennelly, Department of Plant Pathology, Kansas State University, Manhattan KS 66506 (formerly, Department of Plant Pathology, Cornell University); David M. Gadoury and Wayne F. Wilcox, Department of Plant Pathology, New York State Agricultural Experiment Station, Geneva 14456; Peter A. Magarey, South Australian Research and Development Institute, Loxton, South Australia 5333; and Robert C. Seem, Department of Plant Pathology, Cornell University, New York State Agricultural Experiment Station, Geneva 14456
Kennelly, M. M., Gadoury, D. M., Wilcox, W. F., Magarey, P. A., and Seem, R. C. 2007. Addressing the gaps in our knowledge of grapevine downy mildew for improved forecasting and management. Online. Plant Health Progress doi:10.1094/PHP-2007-0726-03-RV.
The complex interactions of Plasmopara viticola with environment and host make grapevine downy mildew an ideal candidate for disease forecasting. However, a forecasting model is only as good as the knowledge used to build it, and DMCast is no exception. We addressed some knowledge gaps concerning this disease: (i) initial timing and span of primary infection; (ii) survival of the lesions and sporangia; and (iii) critical period of fruit susceptibility. Experiments revealed that, though emerging shoots are susceptible earlier than previously thought, primary infection frequently occurs near the confluence of a specific host phenological stage and certain weather conditions. Primary infection also may trigger new epidemics later in the season than was traditionally hypothesized. Lesions declined with repeated sporulation cycles but, contrary to prior reports, not age alone. Sporangia died within 8 h on dry, warm days but retained high viability on cooler days. With controlled inoculations, we determined that in the New York climate, fruit of several cultivars (Chardonnay, Riesling, Concord, and Niagara) become resistant to infection by 2 to 3 weeks post-bloom. These studies have clarified several knowledge gaps and long-held assumptions that have direct implications for improving disease forecasting and disease management.
Plasmopara viticola, the causal agent of grapevine downy mildew, has complex interactions with the environment and with its host. Environmental factors such as temperature, relative humidity, surface wetness, and light can affect germination, infection, incubation, sporulation, and survival. The risk of disease can vary with cultivar, the plant organ infected, and host phenology. The disease can be sporadic in its appearance or severity across regions, but individual vineyards within a region may be at high risk for loss in any given year due to local conditions. These complex interactions and hit-or-miss pattern of occurrence make downy mildew seemingly difficult to predict. However, the combined complexity and specificity also makes grapevine downy mildew an ideal candidate for forecasting. Computer-based forecast programs can integrate many variables to identify critical periods of high host susceptibility, favorable environment, and high inoculum dose: the components of the disease triangle. DMCast is one such forecasting model (21), and there are many others (3,4,5,10,18,19,23).
Downy mildew has been studied for over a century, since it first emerged as a threat to the European wine industry in the late 1880s, and much is known about this disease. However, many key questions about the disease have remained unanswered, and there are little to no data to support some long-held assumptions regarding the biology of the pathogen and the epidemiology of the disease. These questions include: (i) when during the growing season does primary infection begin? (ii) how long does primary inoculum last? (iii) how does the productivity of lesions change over time? (iv) how long do sporangia survive in the field? and (v) when are the fruit most susceptible to infection? Any forecasting model is only as good as its biological base, and these knowledge gaps have limited our ability to forecast and manage this disease.
When does primary infection begin?
Plasmopara viticola survives the winter as oospores in fallen leaf litter in vineyard soils. In the spring, with soil wetness and adequate temperatures, oospores germinate to produce sporangia that liberate zoospores which are splashed by rain into the canopy to cause the first infections. We developed a set of simple rules from 18 years of historical data to forecast initial oosporic infection. The rules were based upon temperature, rainfall, and host phenology thresholds: rainfall > 2.5 mm coincident with temperatures > 11°C after vines had reached Eichorn and Lorenz (E&L) growth stage 12 (5 to 6 leaves unfolded). It has been assumed that vines were not susceptible before this stage (23). We evaluated the guidelines in six regional vineyards of the highly susceptible cultivar Chancellor. Chancellor clusters display a distinct and diagnostic epinasty after infection by P. viticola (Fig. 1). We scouted the vineyards for symptoms every 1 to 7 days, starting before E&L stage 12, and collected weather data from nearby weather stations (13). We then analyzed whether the thresholds correctly identified the date of first infection. To determine whether shoots are susceptible before E&L stage 12, we inoculated shoots at various stages as early as bud break (13).
In 24 of 29 site/year combinations ranging from 1981-2004, the thresholds accurately identified the first primary infections; we observed symptoms of downy mildew as expected following the satisfaction of the thresholds. However, in 2003, infections occurred 7 days before the expected date at E&L stage 12. Our parallel inoculation experiments clearly demonstrated that shoots are susceptible almost as soon as they begin to grow. Thus, the typical infection of the host at E&L stage 12 reflects the usually coincident timing of oospore maturity with this host phenological stage under vineyard conditions. Host susceptibility, per se, does not limit infection before E&L stage 12, and in exceptional years (like 2003), infection can and does occur earlier. However, the guidelines we developed were surprisingly accurate, and with a minor adjustment of the phenological stage, can provide useful and effective warnings of initial oosporic infection that could be used to time initial fungicide applications.
How long into the season do oospores cause infection?
Although it has been reported that oospores mature and germinate over a period of weeks or even months (21,22), many forecasting models, including DMCast, assume that, after the initial oosporic infection event, the epidemic is driven by secondary infections caused by sporangia. However, recent studies in Europe (7) and the US (11) suggest that oospores have an even longer period of maturation, and may indeed be more of a driving force for an epidemic than previously thought. We used a trap plant assay with seedlings exposed to oospore infested soil to determine the period of oospore-based (primary) infection, and were able to detect high levels of new primary infections throughout the growing season (13). In fact the highest disease incidence (> 75%) occurred towards the end of the season, in September. This is important because the conditions for primary infection (rainfall > 2.5mm and temperature > 11°C) are substantially different from the more complicated, and rarer, conditions required for secondary infection, such as the requirement for a night with nearly 100% relative humidity. Thus, a forecasting model that does not include the possibility of protracted oosporic infection might miss important infection periods.
How does the productivity of lesions change over time?
Individual lesions, or “oilspots,” can sporulate multiple times. Young lesions produce a dense field of sporulation (Fig. 2A) but on older lesions, as the tissue becomes necrotic, sporulation occurs only on the margins and total spore production is greatly reduced (Fig. 2B and 2C). There were few data to quantitatively describe this process, so we examined the role of lesion age and sporulation cycles on productivity. We established specific generations of lesions and induced sporulation at various scheduled times, and collected and counted sporangia (13).
Contrary to prior reports (9), the age of lesions did not affect their potential sporangial production during the first 3 weeks after infection (13,17). Lesions that sporulated the first time 3 weeks post inoculation were equally productive as those that first sporulated 1 week post inoculation. Additionally, sporangia from 3-week-old lesions were equally viable as those from younger lesions. In contrast, the number of times that lesions sporulated had a marked effect on productivity. Lesions sporulated abundantly one to three times, and then sporangial production declined sharply (Fig. 3) (13,17). In conditions favorable to sporulation, lesions can quickly become inactive. In drier conditions, they retain high potential to produce sporangia, and this might explain how downy mildew can re-emerge quickly after a dry spell when wet conditions return.
How long do sporangia survive in the field?
In addition to the reduced productivity of lesions over time, the sporangia of P. viticola decline in viability. Sporangia have been reported to die within 15 min of direct sunlight (24), or to survive as long as 14 days in a 10°C moist chamber (8). Many forecasting models, including DMCast (21) use data from a single laboratory study (2) to estimate sporangial survival under vineyard conditions.
We monitored sporangial viability under field conditions (13,17). We inoculated potted vines, induced them to sporulate, and placed the potted vines in the vineyard canopy under various conditions. We collected sporangia at intervals throughout the day and assessed their viability. During warm, dry days (maximum temperature approximately 30°C), nearly all sporangia died within 8 h of exposure (Fig. 4). On cloudier days, with higher humidity, sporangial viability remained at nearly 100% after more than 24 h exposure (Fig. 4) (13,17).
When are the fruit most susceptible?
Grape berries are known to develop ontogenic, or age-related, resistance to downy mildew. Plasmopara viticola infects through stomata, which, in fruit, become non-functional and develop into lenticels (1), and it is assumed that this removal of the infection court prevents disease (20). However, exactly when and to what degree this resistance develops was poorly understood. Furthermore, there were conflicting reports on whether the rachis and berry stems remain susceptible. Finally, there is wide variation in the relative susceptibility of specific organs (i.e., leaves, berries, and stems) among cultivars. For example, the cultivar Delaware has highly susceptible foliage but fruit are immune. The converse is true for the cultivar Chancellor. Our objective was to determine the temporal distribution of susceptibility in two V. vinifera cultivars, Chardonnay and Riesling, and two V. labrusca cultivars, Concord and Niagara.
In Geneva, NY, we inoculated fruit clusters of the above cultivars at several growth stages, from prebloom to 6 weeks postbloom (12). We repeated the studies in Loxton, South Australia, using Chardonnay and Riesling vines. Two weeks after inoculation we assessed fruit clusters for sporulation and symptoms of necrosis on berries, pedicels, and the main rachis (12).
Symptom development depended upon cluster age at time of inoculation (12,14,15,16). On clusters inoculated at prebloom and bloom, berries, pedicels, and the rachis supported abundant sporulation (Fig. 1) (12). On clusters inoculated at later stages, pedicels continued to support sporulation (Fig. 5A) or necrosis and berries on those pedicels eventually became discolored or necrotic (Fig. 5B) (12). Some discolored berries that failed to support sporulation still harbored the pathogen. Conversion of the stomata to lenticels post-infection prevented sporulation. However, when incisions were made in the epidermis, the pathogen was able to emerge through the incisions and sporulate (Fig 5C) (12). This is particularly interesting because P. viticola is not known to sporulate through wounds. Berry susceptibility decreases markedly in NY (Fig. 6). The transition to resistance begins at about 1 week (100 cumulative degree-days, base 10°C) after bloom, and there is high variability in the susceptibility in this transitional time. By 2 to 3 weeks (200 to 300 degree-days) postbloom, berries and pedicels are resistant and clusters do not develop any symptoms (12,14,15,16). Consequently, we can now deploy fungicides in a way that better reflects the changes in fruit susceptibility, and focus our management on the period when the fruit are most at risk.
In Loxton, the period of susceptibility was longer (12). In contrast to climates like in NY (with cold winters, and where bloom of grapevines is highly synchronous), in warm climates such as Loxton’s the bloom period is protracted (6). The phenological heterogeneity at bloom leads to a longer window of susceptibility, as later-blooming clusters remain susceptible. This climate-based phenological heterogeneity is currently being modeled, and will be incorporated into models of fruit susceptibility, allowing our model to be used in a variety of climates.
Through a critical evaluation of our biological base of knowledge, we were able to make substantial improvements to a forecasting model for grapevine downy mildew. The model now more accurately reflects (i) when initial infection occurs, (ii) the protracted duration of oosporic infection, and (iii) the dynamics of lesion longevity, sporangial viability, and ontogenic resistance of the host. The most recent version of DMCast has been made available for use by Cornell University through the New York State IPM Program website.
This research was partially funded by the Viticulture Consortium-East through a grant to Cornell University, NYSAES, under Agreement #34360-7382 and the New York State Wine and Grape Foundation, and from the Riverland Winegrape Industry Council in South Australia.
1. Bessis, R. 1972. Etude de l'évolution des stomates et des tissus peristomatiques du fruit de la vigne. C.R. Acad. Sc. Paris 274:2158-2161.
2. Blaeser, M., and Weltzein, H. 1978. Die Bedeutung von Sporangienbildung, -ausbreitung und -keimung fur die Epidemiebildung von Plasmopara viticola. Zeitschrift fur Pflanzenkrankheiten und Pflanzenschutz 85:155-161.
3. Bleyer, G. 1997. A strategy for the controlled management of Plasmopara viticola. Viticult. Enolog. Sci. 52:168.
4. Cortesi, P., and Hill, G. K. 1991. Simulation of grapevine downy mildew epidemics and control with the P.R.O. model. Pages 74-81 in: Proc. of the First Intn'l Workshop on Grapevine Downy Mildew Modeling. D. M. Gadoury and R. C. Seem, eds. Geneva, NY.
5. Fouassier, S., Magnien, C., and Jacquin, D. 1997. MILVIT: A model for the asexual phase of grapevine downy mildew–Results of 4 years of validation. Viticult. Enolog. Sci. 52:169-171.
6. Gadoury, D. M., Seem, R. C., Kennelly, M. M., and Wilcox, W. F. 2003. Climate-based temporal heterogeneity in flowering and the distribution of ontogenic resistance to major fruit diseases of grapevine. Phytopathology 93:S28.
7. Gobbin, D., Pertot, I., and Gessler, C. 2003. Identification of microsatellite markers for Plasmopara viticola and establishment of high throughput method for SSR analysis. Eur. J. Plant Pathol. 109:153-164.
8. Gregory, C. T. 1915. Studies on Plasmopara viticola (downy mildew of grapes). Pages 126-150 in: Proc. of the Intn'l Cong. Vitic., 12-13 July 1915. San Francisco, CA
9. Hill, G. K. 1989. Effect of temperature on sporulation efficiency of oilspots caused by Plasmopara viticola (Berk. & Curt. ex de Bary) Berl. & de Toni in vineyards. Viticult. Enolog. Sci. 44:86-90.
10. Huber, B., Bleyer, G., and Kassemeyer, H. H. 1998. Verification of the Freiburg model against Plasmopara viticola by field trials from 1993-1997. Pages 18-19 in: Proc. of the Third Intn'l Workshop on Grapevine Downy and Powdery Mildew, at Loxton, South Australia. P. A. Magarey, S. A. Thiele, K. L. Tschirpig, R. W. Emmett, K. Clarke, and R. D. Magarey, eds. SARDI Research Report Series.
11. Kennelly, M. M., Eugster, C., Gadoury, D. M., Smart, C. D., Seem, R. C., Gobbin, D., and Gessler, C. 2004. Contributions of oosporic inoculum to epidemics of grapevine downy mildew (Plasmopara viticola). Phytopathology 94:S50.
12. Kennelly, M. M, Gadoury, D. M, Wilcox W. F., Magarey P. A., and Seem R.C. 2005. Seasonal development of ontogenic resistance to downy mildew in grape berries and rachises. Phytopathology 95:1445-1452.
13. Kennelly, M. M., Gadoury, D. M., Wilcox, W. F., Magarey, P. A., and Seem, R. C. 2007. Primary infection, lesion productivity, and survival of sporangia in the grapevine downy mildew pathogen, Plasmopara viticola. Phytopathology 97:512-522.
14. Kennelly, M. M., Seem, R. C., Gadoury, D. M., Wilcox, W. F., and Magarey, P. A. 2002. Refinement of DMCast, a predictor of grapevine downy mildew (Plasmopara viticola). Phytopathology 92:S41.
15. Kennelly, M. M., Seem, R. C., Gadoury, D. M., Wilcox, W. F., and Magarey, P. A. 2002. Refinement of DMCast, a predictor of grapevine downy mildew. Proc. of the 4th Int. Workshop on Grapevine Downy and Powdery Mildew, Napa, Calif., 30 Sept.-4 Oct., 2002. UC Davis Press.
16. Kennelly, M. M., Seem, R. C., Gadoury, D. M., Wilcox, W. F., and Magarey, P. A. 2003. Integration of lesion productivity and ontogenic resistance of fruit into a warning system for grape downy mildew (Plasmopara viticola). Phytopathology 93:S44.
17. Kennelly, M. M., Seem, R. C., Gadoury, D. M., Wilcox, W. F., and Magarey, P. A. 2004. Survival of grape downy mildew (Plasmopara viticola) sporangia and lesions under field conditions. Phytopathology 94:S50.
18. Madden, L. V., Ellis, M. A., Lalancette, N., Hughes, G., and Wilson, L. L. 2000. Evaluation of a disease warning system for downy mildew of grapes. Plant Dis. 84:549-554.
19. Magarey, P. A., Wachtel, M. F., Weir, P. C., and Seem, R. C. 1991. A computer-based simulator for rational management of grapevine downy mildew (Plasmopara viticola). Plant Prot. Quar. 6:29-33.
20. Muller, K., and Sleumer, S. 1934. Biologische Untersuchungen über die Peronosporakrankheit des Weinstockes, mit besonderer Berücksichtigung ihrer Bekämpfung nach der Inkubationskalendermethode. Landwirt. Jahrbücher 79:509-576.
21. Park, E. W., Seem, R. C., Gadoury, D. M., and Pearson, R. C. 1997. DMCast: A prediction model for grape downy mildew development. Viticult. Enolog. Sci. 52:182-189.
22. Ronzon-Tran Manh Sung, C., and Clerjeau, M. 1988. Techniques for formation, maturation, and germination of Plasmopara viticola oospores under controlled conditions. Plant Dis. 72:938-941.
23. Rosa, M., Genesio, R., Gozzini, B., Maracchi, G., and Orlandini, S. 1993. PLASMO: A computer program for grapevine downy mildew development and forecasting. Comput. Elect. Agric. 9:205-215.
24. Zachos, D. G. 1959. Recherches sur la biologie et l'épidémiologie du mildiou de la vigne en Grece. Ann. Inst. Phytopathol. Benaki 2:193-335.