Determining the relative importance of primary inoculum sources for Stemphylium leaf blight of onion in New York

Note: This section has been adapted from our scientific article published on the American Phytopathological Society (APS) journal 'Phytopathology', in the areas of pathogen population biology and epidemiology: Piñeros-Guerrero, N., Heck, D.W., Hay, F.S., and Pethybridge, S.J. 2025. Relative Contributions of Infected Transplants and Volunteers to the Population Biology of Stemphylium vesicarium in New York Onion Production. Phytopathology. https://doi.org/10.1094/PHYTO-06-25-0209-R

Results

1. To determine genetic diversity within and among S. vesicarium populations: 

Sample collection. Bare root transplants from each of the onion cultivars collected in 2022 (SV4643NT, Crockett, Cartier, Claudius, Oneida, Safrane, Delgado, and Hamilton) and 2023 (Cartier, Safrane, and Hamilton) were infected with S. vesicariumprior to planting, with incidences ranging from 11.2 to 100% and 97.8 to 100% in 2022 and 2023, respectively (Supplementary Fig. S1). S. vesicarium isolates were obtained from infected transplants (n = 162), volunteers (n = 95), and symptomatic onion plants from the main crops at mid- (n = 60) and late season (n = 220), resulting in 537 S. vesicarium isolates in 2022 (n = 217) and 2023 (n = 320). All isolates were identified as S. vesicarium by species-specific PCR (data not shown).

Gene diversity. Nei’s gene diversity (H_e) ranged from 0.636 (SvSSR_19) to 0.958 (SvSSR_09) with an average of 0.775 (Table 1). The locus SvSSR_09 had the highest number of alleles (N_a = 61), and the highest Simpson index (0.957) and Nei’s gene diversity (0.958; Table 1). The locus SvSSR_15 had the lowest number of alleles (N_a = 4) with a Nei’s gene diversity value of 0.666. Evenness (E₅) ranged from 0.541 to 0.892 in the SvSSR_26 and SvSSR_15 loci, respectively (Table 1).

Table 1. Locus summary statistics for nine microsatellites markers in Stemphylium vesicarium populations from New York onion fields in 2022 and 2023.

Genotypic diversity. The genotypic accumulation curve (Supplementary Fig. S2) showed a plateau with nine loci demonstrating these were sufficient to distinguish between individuals in these populations. Multiple alleles were found at each of the nine loci resulting in 399 MLGs from the 537 individuals (Tables 1 and 2). The number of MLGs in source populations ranged from 57 in main crop plants at mid-season to 156 in transplants. However, when comparing the genotype richness (eMLG) among source populations, the lowest was observed in the population from volunteer onions (eMLG = 42.9) while populations from main crop plants collected at mid-season had a higher genotype richness (eMLG = 57.0; Table 2). The population from volunteer onions also showed the lowest genotypic diversity based on the Shannon-Wiener (H’ = 3.587), Stoddart and Taylor’s (G’ = 28.4), and Simpson (λ = 0.964) indices (Table 2). The population from transplants had the highest genotype richness (eMLG = 59.1) and highest genotypic diversity indicated by the H’ (4.072), G’ (58.134), and λ (0.983) indices (Table 2). The highest Nei’s gene diversity was also observed in the population from transplants (H_exp = 0.747). In addition, evenness values for the population from transplants (E₅ = 0.990) indicated that MLGs were more evenly distributed than volunteers (E₅ = 0.775) and main crop plants at mid- and late season (E₅ = 0.887 and 0.974, respectively; Table 2). The genotypic diversity indices were lower in 2022 (H’ = 4.887, G’ = 87.040, λ = 0.986) than the 2023 (H’ = 5.134, G’ = 146.025, λ = 0.993) populations (Table 2). Populations collected in 2023 also had MLGs more evenly distributed (E₅ = 0.859) than those from 2022 (E₅ = 0.654; Table 2).

Table 2. Genetic diversity indices, clonal fraction, and indices of association for source and year Stemphylium vesicarium populations from New York onion fields in 2022 and 2023.

Clonal reproduction and linkage disequilibrium. Clonality levels within source populations varied between 0.037 to 0.347 with an average of 0.257. Populations from transplants and main crop plants collected at mid-season had the lowest clonal fractions (0.037 and 0.05, respectively), while those from volunteers had the highest clonality fraction (0.347). Populations from main crop plants collected at late season samplings had a higher clonal fraction (0.295) than those at mid-season (0.05; Table 2). A higher clonal fraction was observed in populations from 2022 (0.24) than 2023 (0.181). The index of association (I_A) and the standardized index of association (r̄_d) were significant (P < 0.001) for all S. vesicarium populations by source and year supporting deviation from the null hypothesis of linkage equilibrium (Table 2).

2. To determine the population structure of S. vesicarium in NY onion fields.

Population structure and genetic differentiation. The MSN based on Bruvo's distance clearly depicted an association between MLGs from volunteers and those collected from the main crop at mid- and late season with 20 MLGs shared among these populations (Fig. 1). The most frequent MLG (MLG.384) had 15 clonal individuals in S. vesicariumpopulations from volunteers (n = 10) and main crop plants at the late season sampling (n = 5; Fig. 1). The second most frequent MLG (MLG.325), with 13 clonal individuals, was also in S. vesicarium populations from volunteers (n = 3) and main crop plants collected at late season (n = 10). The MSN also showed that populations from transplants formed a distinct cluster containing most individuals in the same unique branch (Fig. 1). Populations of S. vesicarium from transplants only shared two of 156 MLGs (MLG.291 and MLG.385) with populations from the main crop plants at the late season sampling (Fig. 1). A few individuals from transplants were closely related to individuals from volunteers and main crop plants (Fig. 1). Only one MLG (MLG.167) was shared between S. vesicarium populations from transplants in 2022 and 2023 (Supplementary Fig. S3). Most of the MLGs (372 MLGs, 93.2%), were unique to a specific population, and only 27 MLGs (6.8%) were shared among two or more source populations (Fig. 1). According to year, 28 MLGs were shared between populations with MLG.384 being the most frequent (n = 13 and 2 in 2022 and 2023, respectively) and MLG.325 (n = 6 and 7 in 2022 and 2023, respectively; Supplementary Fig. S3).

AMOVA analysis with clone-corrected datasets detected significant genetic differentiation among S. vesicariumsource populations (Φ = 0.129, P = 0.001), with most of the genetic variation (87.1%) attributed within populations and 12.9% between populations (Table 3). AMOVA also detected significant genetic differentiation between populations according to year (Φ = 0.004, P = 0.012), with most of the genetic variation (99.6%) from within populations and only 0.42% between years (Table 3). Significant probability values (P < 0.05) rejected the null hypothesis of random mating between populations suggesting significant differentiation (Table 3).

Table 3. Analysis of molecular variance using clone corrected datasets of Stemphylium vesicarium populations from infected transplants, volunteers, and main crop plants in 2022 and 2023 from New York onion fields.

Pairwise genetic differentiation analyses (F_ST) confirmed the significant differentiation detected by AMOVA for source, but not for year populations. The F_ST values supported significant differentiation of S. vesicarium populations from transplants and volunteers (F_ST = 0.156, P < 0.01; Fig. 2). The high F_ST values between populations from transplants and the main crop at mid- (F_ST = 0.161) and late (F_ST = 0.182) season samplings revealed significant genetic differentiation (P < 0.01; Fig. 2). Moreover, low F_ST values between populations from volunteers and main crop plants from mid- (F_ST = 0.008, P = 0.146) and late season (F_ST = -0.0094, P = 0.999) samplings suggested no genetic differentiation between these populations (Fig. 2). A low F_ST value (-0.0008, P = 0.734) between 2022 and 2023 populations suggested no genetic differentiation. Although the AMOVA significance test determined populations to be significantly different by year, this analysis still attributed 99.6% of the total variation to variability within populations and only 0.42% to differences between populations (Table 3). When pairwise comparisons were performed by a combination of source and year populations (Supplementary Fig. S4), a low F_ST value (-0.0114, P = 0.999) between populations from transplants in 2022 and 2023 suggested no genetic differentiation. Similarly, a low F_ST value (-0.0475, P = 0.999) suggested that populations from volunteers between years were not significantly differentiated (Supplementary Fig. S4).

DAPC analyses of non-clone-corrected datasets using populations stratified by source (Fig. 3A) and year (Fig. 3B) also confirmed the AMOVA results for source but not year. S. vesicarium MLGs from volunteers and main crop plants from both samplings were aggregated in the scatter plot (Fig. 3A). The ellipses for the three clusters, showing the variability within each cluster, were mostly overlapped, demonstrating genetic similarity among MLGs within these populations (Fig. 3A). In contrast, MLGs of the population from transplants clustered together but were separated by vertical axis 2 from the other three populations, suggesting that MLGs from transplant populations were genetically distinct (Fig. 3A). When MLGs were clustered by combinations of source and year populations, those from transplants in both years overlapped (Supplementary Fig. S5A and S5B). DAPC density plots showed that populations from 2022 and 2023 were mostly overlapped, suggesting no genetic differentiation between these populations (Fig. 3B).

The STRUCTURE analysis predicted K = 2 as the optimal number of genetic clusters for both source and year populations (Supplementary Fig. S6A and S6B). According to source populations, STRUCTURE bar plots indicated low admixture across the populations, where most isolates from transplants presented a substantial ancestral contribution from the pink cluster (Fig. 4A). In contrast, most isolates from the other sources showed a substantial ancestral contribution from another cluster depicted by yellow (Fig. 4A). No distinct admixture patterns were observed in the STRUCTURE bar plots when analyzed by year (Fig. 4B). A combined analysis by source and year supported these results (Supplementary Fig. S7A and S7B).

Main_figures_First_look

Pineros_Supplementary Figures_First_look

Discussion

This study investigated the relative contributions of potential primary inoculum sources, infected transplants and volunteers, to SLB epidemics in NY onion fields. High levels of genetic diversity and clonal reproduction within S. vesicarium populations were depicted when populations were stratified by source and year. High levels of genetic differentiation were only observed among source populations but not between populations according to year.

Genotypic diversity analyses identified 399 distinct MLGs (74.3%) within 537 isolates, with the majority of MLGs (88.7%) represented by a single isolate, indicating high genotypic richness. According to source population, the highest genotypic richness, estimated by eMLGs, was found in populations from transplants, which also had high levels of genotypic diversity but low clonal fraction. The high levels of genotypic diversity observed in transplant populations might be attributed to differences in integrated disease management strategies, fungicide resistance, sampling timing, or sources of inoculum present in the region where transplants have been grown. Moreover, populations from main crop plants collected at mid-season in NY also showed high levels of genotypic richness. These findings were consistent with previous studies conducted by Heck et al. (2023) and Subedi et al. (2025), where S. vesicarium populations from NY onion fields were highly diverse showing high genotypic richness, with MLGs values of 96.2% (Heck et al. 2023) and 75% (Subedi et al. 2025). Chandel et al. (2024) and Hassan et al. (2021) reported similar results for S. vesicarium populations from onions in India using random amplified polymorphic DNA primers, depicting high levels of polymorphism among isolates.

 All S. vesicarium populations were predominantly clonal which significantly deviated from the null hypothesis of linkage equilibrium. Despite the suggestion of clonal reproduction in all populations, according to I_A and r̄_d, the clonal fraction was low in populations from transplants (0.037) and main crop plants collected at mid-season (0.05). Taken together, considering the high genotypic diversity and low clonal fraction for some populations, we hypothesize that S. vesicarium populations may not be entirely reproducing clonally and that recombination might still contribute to the lifecycle. The genus Stemphylium is comprised of both self-sterile (heterothallic) and self-fertile (homothallic) species (Inderbitzin et al. 2005). Heterothallic Stemphylium spp. means isolates contain one of two mating types, either MAT1-1or MAT1-2, in their MAT locus, while homothallic species contain a fusion of the two regions in the same genome (Debuchy and Turgeon 2006; Inderbitzin et al. 2005). However, self-fertile isolates of Stemphylium spp. carrying only the MAT1-1 gene have also been characterized (Inderbitzin et al. 2005). Sharma et al. (2020) found a single copy of both genes MAT1-1 and MAT1-2 fused in reverse orientation in two S. vesicarium isolates from NY onion fields suggesting homothallism. This mechanism allows an individual to not only produce offspring through outcrossing but also from haploid selfing or syngamy with genetically identical haploid cells (Billiard et al. 2012; Debuchy and Turgeon 2006). As a result, homothallic fungi often exhibit a population structure that resembles those of exclusively asexual or clonal reproduction, evidenced by strong linkage disequilibrium, even if sexual reproduction (haploid selfing) occurs (Billiard et al. 2012; Ekins et al. 2006). Moreover, the differences in clonality and genotypic diversity between populations from transplants and main crop plants at mid-season when compared to late season and volunteers, could have been influenced by sampling timing. Our results may also be biased towards clonal populations as sampling was conducted towards the end of the cropping season and late winter when asexual propagation was likely dominant (Hay et al. 2021, 2022). Selective pressure might favor the most well-adapted strains and potentially reduce overall diversity with each asexual cycle or clonal generation as the season progresses. These results were expected as S. vesicarium dispersal within the same cropping season relies on conidia dispersed by wind and rain (Hay et al. 2021), and insects including onion thrips (Thrips tabaci), generating multiple clonal generations (Leach et al. 2019). Significant linkage disequilibrium and highly diverse populations have also been observed in other studies including previous reports of S. vesicarium populations (Subedi et al. 2025) and other homothallic pathogens such as Sclerotinia sclerotiorum (Dunn et al. 2017; Lehner et al. 2017; Kamvar et al. 2017) and Stagonosporopsis citrulli (Brewer et al. 2015), where predominantly clonal populations also showed high genetic diversity.

 Low genetic exchange was observed between populations from transplants and the other sources evidenced by the low admixture from the STRUCTURE analysis. In contrast, similar admixture patterns were observed between years suggesting ongoing gene flow over time. MSN and DAPC analyses clearly display a relationship between individuals from volunteers and those from main crop plants at either sampling, with multiple shared MLGs. However, S. vesicariumpopulations from transplants were clustered separately from the other source populations suggesting some genetic isolation. The genetic structure observed for source populations but not by year, aligns with previous temporal studies of S. vesicarium (Heck et al. 2023; Subedi et al. 2025). This finding supports the capacity of S. vesicarium to overwinter in crop residue, soil or volunteers (Hay et al. 2022; Van der Heyden et al. 2022). The high levels of genetic differentiation estimated by F_ST between the transplant population and remaining populations support the findings from the structure analysis. These findings suggest that S. vesicarium populations from transplants do not contribute to SLB epidemics in NY onion fields and they are not persisting throughout the cropping season. This may be attributed to differences in local disease management strategies, fungicide sensitivities, environmental factors conducive to infection and disease development, and competition for resources with locally adapted populations (Croll and McDonald 2017; Zhan et al. 2002). Further studies, especially those assessing differences in fungicide sensitivity profiles in these isolates, should be conducted to explore the factors influencing this finding. However, our results clearly support the lack of genetic differentiation among S. vesicarium isolates from volunteers and those obtained from main crop plants collected at mid- and late season samplings suggesting that inoculum may be perpetuated through survival on volunteers (Abdelrhim et al. 2024; Van der Heyden et al. 2022).

High genetic diversity within S. vesicarium populations combined with significant linkage disequilibrium, and low genetic differentiation among sources and between years, may be attributed to multiple factors other than recombination. These factors include high mutation rates, some gene flow or genetic exchange among populations, long-distance dispersal of ascospores, and the ability of the pathogen to overwinter on crop residue and infect alternative crop hosts (Gossen et al. 2021; Hay et al. 2019, 2022; Jaimes et al. 2016; McDonald et al. 2023).