*To whom correspondence should be addressed.†Present address: BASF Corporation, 26 Davis Drive, PO Box 13528, Research Triangle Park, NC 27709, USA.‡E-mail: email@example.com
Sexuality in the oomycete Plasmopara viticola, the causal agent of grapevine downy mildew, was studied using isolates from five populations from North America and Europe. Leaf discs of Vitis vinifera cv. ‘Chardonnay’ were inoculated with either individual single-sporangiophore isolates, or in all possible pairwise combinations of 25 isolates from New York State, USA. The occurrence of oospores in leaf discs indicated that the pathogen was heterothallic with two mating types, P1 and P2 in a ratio of 11 : 14 for this population. Heterothallism was confirmed when three representative isolates of each mating type from New York were coinoculated with each of 40 isolates from populations of P. viticola from Michigan, Missouri (USA), Germany and Italy. For each isolate tested, oospores formed with either test isolates of P1 or test isolates of P2 mating types, indicating that the isolates were exclusively P1 or P2 only. For these same isolates, no oospores formed as a result of self-crosses. The ratio of P1 : P2 mating types for all isolates in the study was 27 : 38, statistically equivalent to a 1 : 1 ratio according to χ2 analysis (P = 0·68).
Grapevine downy mildew is a disease of both great economic and historic importance (Viennot-Bourgin, 1981; Hewitt & Pearson, 1988). It is particularly destructive in viticultural regions with warm, wet conditions during the growing season, including parts of Europe and North America. The causal agent, Plasmopara viticola, has been the subject of study since its introduction into European vineyards in the 1880s (Lafon & Bulit, 1981), yet many aspects of its basic biology remain unknown.
While conducting a series of laboratory studies on fungicidal control of P. viticola, it was impossible to obtain oospores when single-sporangiophore isolates of the fungus were inoculated onto unsprayed grape leaves, suggesting that the fungus might be heterothallic. A subsequent review of the literature revealed that although the role of oospores in the survival and epidemiology of this pathogen has been well documented (Lafon & Bulit, 1981; Hewitt & Pearson, 1988), the nature of sexuality in P. viticola was unknown and remained ambiguous to date (Michelmore et al., 1988; Ronzon-Tran Manh Sung & Clerjeau, 1988). Therefore, the objective of this study was to determine whether P. viticola is a heterothallic fungus, and if so, to characterize the mating system.
Materials and methods
Collection and maintenance of P. viticola isolates
A total of 65 single-sporangiophore isolates was used for this study. The origin and nature of these isolates is summarized in Table 1. The 25 isolates comprising the New York (NY) population were collected in 1997 and previously characterized with regard to strobilurin fungicide sensitivity (Wong & Wilcox, 2000). Ten isolates each from populations from Michigan and Missouri, USA, and Milan, Italy (denoted as AS, JM, and PC groups, respectively) were obtained as previously described (Wong & Wilcox, 2000) from sporulating leaves, collected and shipped (via overnight delivery service) by the sources identified in Table 1 during the 1999 growing season. One single-sporangiophore isolate was obtained per infected leaf. Samples from a German population (GH from Reinhessen) arrived as leaf tissue that had been collected in 1995 and which contained oospores that had been allowed to mature under natural conditions by placing infected leaves in nylon mesh bags and overwintering them on the vineyard floor. Subsequently, the leaves were coarsely ground, and stored at 4°C in plastic bags. To obtain isolates from this material, surface-sterilized 10-mm diameter leaf discs, excised from glasshouse-grown Vitis vinifera cv. ‘Chardonnay’ vines, were floated, abaxial side down, in a suspension containing 50 mL of distilled H2O and 10 g of the ground, overwintered leaves. Leaf discs were then placed onto 15 mL of RAP agar [1.5% granulated agar (DIFCO, Detroit, MI, USA), amended with 30 μg mL−1 rifampicin, 150 μg mL−1 sodium ampicillin, and 5 μg mL−1 pimaricin (Sigma, St Louis, MO, USA)]Wong & Wilcox (2000), contained in 60-mm diameter Petri dishes, with the abaxial surface exposed, and maintained at room temperature (20–22°C) with 12-h periods of alternating fluorescent light and darkness from a 40 W cool-white fluorescent bulb, ≈ 45 µEm−2s−1 (Philips, Sommerset, NJ, USA). Lesions and sporangia developed between 3 and 7 days after removal from the oospore suspension, permitting 10 single-sporangiophore isolates to be obtained, one per infected leaf disc. Isolates were maintained on surface-sterilized leaf discs, with weekly transfers to new tissue, using the methods previously described (Wong & Wilcox, 2000).
Table 1. Origin of Peronospora viticola isolates
New York, USA
Dynamics of oospore formation
In initial experiments, selected isolates from the NY group were crossed with each other by coinoculating surface-sterilized 10-mm-diameter leaf discs, maintained on RAP agar in 60-mm-diameter Petri dishes, with all 15 possible pairings of isolates DM−15, AR−1, AR-2, FC-3, FC-4, and FC-5. Leaf discs and inocula were produced according to published methods (Wong and Wilcox, 2000), and six replicate discs were coinoculated on their abaxial surface with individual 10-µL droplets containing 1 × 104 sporangia per mL of each paired isolate (one drop in the centre of each leaf disc). Single, discrete, lesions, with abundant sporangia, formed after 5 days incubation at room temperature (20–22°C) with alternating 12-h periods of fluorescent light and dark. Leaf discs were subsequently transferred to a 12°C growth chamber (Percival, Perry, LA, USA), and incubated for 14 days with alternating 12-h periods of fluorescent light and darkness from a Philips 20 W cool-white fluorescent bulb, ≈ 20 µEm−2s−1, as suggested by previous studies on oospore formation (Gehmann & Staudt, 1986; Ronzon-Tran Manh Sung & Clerjeau, 1988). Subsequently, leaf discs from individual crosses were transferred separately to 15-mL screw-top centrifuge tubes, covered with 10 mL of 95% ethanol (heated in a 90°C hot water bath), and incubated for at least 72 h at room temperature(20–22°C) to clear the leaf discs. Oospores were detected by mounting the leaf discs in distilled H2O and observing the tissue with phase-contrast microscopy ( × 100).
Two isolate pairs, one that produced oospores under these test conditions (AR−1 × DM−15), and one that did not (DM−15 × FC-4), were employed further to determine the timing of oospore formation. For each isolate pairing, 10 sets of three replicate leaf discs per Petri dish were prepared, inoculated and incubated as described above. After the initial 5-day incubation at room temperature, one set of leaf discs from each isolate pairing was cleared with ethanol as described above. The remaining sets were moved to a 12°C growth chamber, and one set for each isolate pairing was removed and cleared either 2, 4, 6, 8, 10, 12, 14, 16, 18 or 20 days afterwards, as described above. The number of oospores formed in each leaf disc was quantified by counting the total number of oospores formed in the entire leaf disc, with the aid of a microscope at × 100 magnification. This experiment was repeated three times and the means of the replicate treatments were analysed by analysis of variance (anova; General Linear Model, Minitab vs. 12·10, State College, PA, USA). Data were normalized by using the log (y + 1) transformation, where y = the number of oospores detected in the leaf discs.
Mating type determination
To identify mating types in the NY population, all possible pairings were performed for each of the 25 isolates, including individual isolates crossed to themselves, for a total of 325 crosses. For each cross, six replicate leaf discs were coinoculated with the appropriate isolate(s), incubated for 5 days at room temperature, and transferred to a 12°C growth chamber, as described above. After 21 days incubation in the growth chamber, leaf discs were cleared in ethanol and oospores were detected (but not quantified) with light microscopy, as described above. Matings were considered positive if one or more oospores were detected in the entire leaf disc. This experiment was repeated three times to confirm the results of the test. To test for the potential of delayed homothallism, similar to that observed for the powdery mildew fungus, Uncinula necator (Gadoury & Pearson, 1991), additional sets of six leaf discs, each inoculated with only one individual isolate, as described above, were incubated for approximately 90 days at 12°C after the initial 5 days incubation at room temperature to allow for disease development. These leaf discs were cleared with ethanol and oospore formation detected as described above.
To test the remaining populations, the 40 isolates from groups AS, JM, OP and PC(Table 1) were crossed in pairwise combinations with each of six test isolates of the NY population, or they were self-crossed without the use of a test isolate. For each cross, six replicate leaf discs were prepared, inoculated and incubated as described above. The experiment was performed twice to confirm the results of the test.
Nature and timing of oospore formation
The six test isolates from the NY population of P. viticola used in the initial experiments fell into two distinct mating compatibility groups (mating types): group 1 (DM−15, FC-3, FC-4) and group 2 (AR−1, AR-2 and FC-5). Pairings of isolates from within either group did not result in any observed oospore formation. However, all pairings of isolates between groups resulted in oospore formation. Oospores were thick-walled, and between 30 and 50 µm in diameter, consistent with those of P. viticola previously described in the literature (Lafon & Bulit, 1981) (Fig. 1). Formation occurred in discrete clusters throughout the infected tissue, with higher concentrations visible near the vasculature of the leaf tissue (secondary, but not tertiary leaf veins) and at the centre of the lesion. No partially formed oospores were visible in the samples examined.
When the formation of oospores over time was examined using isolates AR−1 and DM−15, no oospore formation in leaf discs was detected until 8 days of incubation at 12°C. Relatively few oospores were detected at this time, but the number of oospores increased dramatically after an additional 2 days, with no significant difference detected between leaf discs incubated for any period between 10 and 20 days (P = 0·85) (Fig. 2). Oospore densities were highly variable, ranging from 23 to 686 for discs incubated 10–20 days at 12°C. No oospores were detected in the leaf discs coinoculated with isolates DM−15 and FC-4.
Mating type determination
Consistent with the initial experiment, all 25 isolates from the NY population could be separated into two distinct mating types in the ratio of 11 : 25. These were designated as P1 and P2 (Fig. 3). Coinoculation of isolates between the two groups always resulted in oospore formation, but not among crosses within groups. In almost all compatible matings, all six replicate leaf discs contained oospores; in no cases were oospores detected in fewer than three of the discs. Oospore formation was abundant in almost all cases, consistent with the numbers observed in previous experiments. For noncompatible interactions, no oospores were detected in any of the six replicate leaf discs. In all cases, 100% of the leaf discs exhibited symptoms of downy mildew infection 5 days after inoculation. The results from these crosses were identical for each of the three repeated experiments. Within these, the infrequent compatible interaction with fewer than six leaf discs containing oospores was not specific to pairs of isolates, suggesting that this was linked to the methodology rather than to genetic factors.
No oospores were detected in leaf discs inoculated with a single isolate when discs were examined at 21 or 90 days incubation at 12°C, confirming the absence of delayed homothallism in the NY population.
When the test isolates from the NY population were crossed with the other 40 isolates from populations AS, GH, JM and PC, oospores were formed in all cases, with each of the 40 isolates successfully crossing with all three of either the P1 or P2 test isolates but never with test isolates from both groups (Fig. 4). No oospores were detected when single isolates were self-crossed. Results from each individual pairing were identical when the experiment was repeated.
The ratio of P1 : P2 mating types identified for all 65 single-sporangiophore isolates in this study was 27 : 38, statistically equivalent to a 1 : 1 ratio as determined by χ2 analysis (P = 0·68).
The results clearly show that P. viticola is a heterothallic fungus with a mating system that is predominantly, if not exclusively, bipolar. The sample size used in this study was large enough to represent 95% of the sampled population with > 95% confidence (Leung et al., 1993). Thus, it is possible that a small fraction of individual isolates within the species do exhibit a different mating type or some form of homothallism (including secondary homothallism or delayed homothallism), although we were not able to detect them. A more detailed and expansive study including (but not limited to) the visualization of mating structures in the leaf, the analysis of mating types for progeny recovered from mature oospores formed in test crosses, and the development of molecular markers to identify mating types, would lead to a more complete understanding of sexuality of this fungus. Nevertheless, the results presented here provide a sound foundation for further exploration of sexuality in P. viticola.
Sexuality within the Peronosporaceae is far from being understood, with only six species having been characterized to date (Table 2). Further studies on other downy mildew fungi would help to understand more clearly their mating systems, in addition to their basic biology, and biological and taxonomical relationship to each other (Hall, 1996).
Table 2. Characterized mating systems in the Peronosporaceae
+, indicates the presence of this mating system for the species.
In addition to expanding the relatively meagre body of knowledge concerning sexuality within the Peronosporaceae, the results reported here may have practical implications with respect to future research into the genetics of P. viticola and control of grapevine downy mildew. Topics on which these results impact might include the potential, persistence or heritability of resistance to fungicides such as the phenylamides (Leroux & Clerjeau, 1985; Hewitt & Pearson, 1988) and strobilurins (Wong & Wilcox, 2000), and the potential for pathogenic specialization among P. viticola isolates relative to individual grapevine genotypes.
The authors would like to thank Drs P. Cortesi, G. Hill, J. Moore, A. Schilder and M. Seidel for their assistance in collecting P. viticola used for this study; J. A. Burr, L. E. Hoffman and D. G. Riegel for technical assistance; and Professor E. E. Butler and Dr D. M. Gadoury for insight and advice.