Effector Gene Suites in Some Soil Isolates of Fusarium oxysporum Are Not Sufficient Predictors of Vascular Wilt in Tomato

Seventy-four Fusarium oxysporum soil isolates were assayed for known effector genes present in an F. oxysporum f. sp. lycopersici race 3 tomato wilt strain (FOL MN-25) obtained from the same fields in Manatee County, Florida. Based on the presence or absence of these genes, four haplotypes were defined, two of which represented 96% of the surveyed isolates. These two most common effector haplotypes contained either all or none of the assayed race 3 effector genes. We hypothesized that soil isolates with all surveyed effector genes, similar to FOL MN-25, would be pathogenic toward tomato, whereas isolates lacking all effectors would be nonpathogenic. However, inoculation experiments revealed that presence of the effector genes alone was not sufficient to ensure pathogenicity on tomato. Interestingly, a nonpathogenic isolate containing the full suite of unmutated effector genes (FOS 4-4) appears to have undergone a chromosomal rearrangement yet remains vegetatively compatible with FOL MN-25. These observations confirm the highly dynamic nature of the F. oxysporum genome and support the conclusion that pathogenesis among free-living populations of F. oxysporum is a complex process. Therefore, the presence of effector genes alone may not be an accurate predictor of pathogenicity among soil isolates of F. oxysporum.