Contrasting genetic structure between Magnaporthe grisea populations associated with the golf course turfgrasses Lolium perenne (perennial ryegrass) and Pennisetum clandestinum (kikuyugrass).

Gray leaf spot (GLS) disease of perennial ryegrass (Lolium perenne) and kikuyugrass (Pennisetum clandestinum) in golf courses in California was first noted in 2001 and 2003, respectively, and within 5 years had become well established. The causal agent of the disease is the fungus Magnaporthe grisea, which is known to consist primarily of clonal lineages that are highly host specific. Therefore, our objective was to investigate host specificity and population dynamics among isolates associated primarily from perennial ryegrass and kikuyugrass since the disease emerged at similar times in California. We also obtained isolates from additional hosts (tall fescue, St. Augustinegrass, weeping lovegrass, and rice) and from the eastern United States for comparative purposes. A total of 38 polymorphic amplified fragment length polymorphism makers were scored from 450 isolates which clustered by host with high bootstrap support (71 to 100%). Genetic structure between kikuyugrass and perennial ryegrass isolates differed significantly. Isolates from kikuyugrass were genotypically diverse (n = 34), possessed both mating types, and some tests for random mating could not be rejected, whereas isolates from perennial ryegrass were less genotypically diverse (n = 10) and only consisted of a single mating type. Low genotypic diversity was also found among the other host specific isolates which also only consisted of a single mating type. This is the first study to document evidence for the potential of sexual reproduction to occur in M. grisea isolates not associated with rice (Oryza sativa). Moreover, given the significant host specificity and contrasting genetic structures between turfgrass-associated isolates, the recent emergence of GLS on various grass hosts in California suggests that potential cultural practices or environmental changes have become conducive for the disease and that the primary inoculum may have already been present in the state, despite the fact that two genotypes associated with perennial ryegrass and St. Augustinegrass in California were the same as isolates collected from the eastern United States.

Significant diversity and potential problems associated with inferring population structure within the Cenococcum geophilum species complex.

Cenococcum geophilum is perhaps the most widely distributed and most recognized ectomycorrhizal fungus with a host range of more than 200 tree species from 40 genera of both angiosperms and gymnosperms. We conducted a phylogenetic analysis on a large collection of isolates (n=74) from North America and Europe based on glyceraldehyde 3-phosphate dehydrogenase (gpd). A subset of isolates (n=22) also was analyzed with the more conservative LSU-rDNA locus. Significant nucleotide diversity was detected (approximately 20%) in the gpd region and the LSU-rDNA analysis supported that the C. geophilum isolates studied were monophyletic but distinct from two isolates, Am5-1 and N2-10, which previously were used in population genetic studies of this species. These results suggest that Am5-1 and N2-10 are likely two undescribed species or even genera. Our results suggest that C. geophilum sensu lato is a species complex and support previous molecular, physiological and morphological studies that have shown significant diversity in C. geophilum. This study also revealed that caution is advised when conducting population genetic studies in C. geophilum due to the possibility of pooling unrelated isolates. This potential problem also has implications for other fungal taxa because cryptic species routinely have been found in recent years based on molecular data.