VIEW ARTICLE

Genetics

Local and Regional Variation in Populations of Fusarium oxysporum from Agricultural Field Soils. D. J. Appel, Graduate student researcher, Department of Plant Pathology, University of California, Berkeley 94720; T. R. Gordon, associate professor, Department of Plant Pathology, University of California, Berkeley 94720. Phytopathology 84:786-791. Accepted for publication 29 April 1994. Copyright 1994 The American Phytopathological Society. DOI: 10.1094/Phyto-84-786.

Populations of Fusarium oxysporum were sampled from two agricultural fields in Maryland. Among 197 isolates, 56 vegetative compatibility groups (VCGs) and 25 mitochondrial DNA (mtDNA) haplotypes were identified. Fifty-four VCGs, representing 97% of the VCG diversity and 92% of the mtDNA haplotypes, were found among 78 isolates that were not pathogenic to muskmelon. The remaining two VCGs included 115 isolates of F. o. melonis: 40 isolates in VCG 0131 (races 0, 1, and 2) and 75 in 0134 (races 1 and 1,2). This is the first report from North America of race 1,2, for which there is presently no resistance in muskmelon. The two VCGs associated with virulent isolates also included four nonpathogenic isolates. Only one VCG and four mtDNA haplotypes overlapped between the two locations sampled. Sixty nonpathogenic VCGs of F. oxysporum previously identified from the San Joaquin Valley of California, were paired with representative isolates of the 54 VCGs from Maryland; only one VCG was shared between these populations. Based on parsimony analysis of the mtDNA data, the Maryland nonpathogens were more closely related to F. o. melonis than to the California nonpathogens. One nonpathogen shared the mtDNA haplotype of VCG 0131 but lacked pathogenicity to muskmelon. The other nonpathogen within VCG 0131 and two in VCG 0134 differed in mtDNA haplotype from the pathogens within these groups and were not closely related to them, except for their shared vegetative compatibility phenotype. Two nonpathogen VCGs from Maryland and eight from California shared the same mtDNA haplotype as VCG 0131, but none shared the mtDNA haplotype associated with VCG 0134. This result may indicate that VCG 0131 has resided longer in North America than VCG 0134. Based on mtDNA, both pathogen VCGs were more closely related to a nonpathogen VCG than to each other. Therefore, isolates of F. o. melonis do not form a distinct phylogenetic group as the forma specialis designation implies. However, the recent appearance of the new races in North America is most likely the result of an introduction from another continent and not the result of evolution of isolates associated with VCG 0131 or indigenous nonpathogens into new races of F. o. melonis.