RESUMEN
ABSTRACT Thirty-nine isolates of Fusarium oxysporum were collected from tomato plants displaying wilt symptoms in a field in California 2 years after F. oxysporum f. sp. lycopersici race 3 was first observed at that location. These and other isolates of F. oxysporum f. sp. lycopersici were characterized by pathogenicity, race, and vegetative compatibility group (VCG). Of the 39 California isolates, 22 were in VCG 0030, 11 in VCG 0031, and six in the newly described VCG 0035. Among the isolates in VCG 0030, 13 were race 3, and nine were race 2. Of the isolates in VCG 0031, seven were race 2, one was race 1, and three were nonpathogenic to tomato. All six isolates in VCG 0035 were race 2. Restriction fragment length polymorphisms (RFLPs) and sequencing of the intergenic spacer (IGS) region of rDNA identified five IGS RFLP haplotypes, which coincided with VCGs, among 60 isolates of F. oxysporum from tomato. Five race 3 isolates from California were of the same genomic DNA RFLP haplotype as a race 2 isolate from the same location, and all 13 race 3 isolates clustered together into a subgroup in the neighbor joining tree. Collective evidence suggests that race 3 in California originated from the local race 2 population.
RESUMEN
ABSTRACT Wheat heads showing symptoms of Fusarium head blight were collected from four commercial fields in Zhejiang Province, China, an area where epidemics occur regularly. A total of 225 isolates were subjected to population-level analyses using restriction fragment length polymorphism (RFLP) as markers. Diagnostic RFLP markers established that all isolates belonged to Fusarium graminearum lineage 6. Nine polymorphic probes were hybridized to all isolates, resulting in 65 multilocus RFLP haplotypes (MRH). Probing with the telomeric clone pNla17, which reveals differences among isolates in the hypervariable subtelomeric region, differentiated the 65 MRH further into 144 clones. Mean gene diversity for the four field populations was similar, ranging from H = 0.306 - 0.364 over the nine RFLP loci for clone-corrected data. High levels of gene flow were inferred from a low level of population subdivision among all field populations, indicating that they were part of the same population. Pairwise linkage disequilibrium measures did not unequivocally support a random mating population, because one-third of locus pairs were significantly different from the null hypothesis of no-association between alleles. We speculate therefore that sexual recombination may not be frequent and that high levels of genotypic diversity may be maintained by relatively low selection pressure acting on a highly diverse population.