Diversity of Epidemic Populations of Gibberella zeae from Small Quadrats in Kansas and North Dakota.

ABSTRACT Gibberella zeae (anamorph Fusarium graminearum) causes Fusarium head blight (FHB) of wheat and barley and has been responsible for several billion dollars of losses in the United States since the early 1990s. We isolated G. zeae from the top, middle, and bottom positions of wheat spikes collected from 0.25-m(2) quadrats during severe FHB epidemics in a single Kansas (KS) field (1993) and in a single North Dakota (ND) field (1994). Three amplified fragment length polymorphism (AFLP) primer pairs were used to resolve 94 polymorphic loci from 253 isolates. Members of a subset of 26 isolates also were tested for vegetative compatibility groups (VCGs). Both methods indicated high levels of genotypic variability and identified the same sets of isolates as probable clones. The mean number of AFLP multilocus haplotypes per head was approximately 1.8 in each population, but this value probably underestimates the true mean due to the small number of samples taken from each head. Isolates with the same AFLP haplotype often were recovered from different positions in a single head, but only rarely were such apparently clonal isolates recovered from more than one head within a quadrat, a pattern that is consistent with a genetically diverse initial inoculum and limited secondary spread. The KS and ND samples had no common AFLP haplotypes. All G. zeae isolates had high AFLP fingerprint similarity (>70%, unweighted pair group method with arithmetic means similarity) to reference isolates of G. zeae lineage 7. The genetic identity between the KS and ND populations was >99% and the estimated effective migration rate was high (Nm approximately 70). Tests for linkage disequilibrium provide little evidence for nonrandom associations between loci. Our results suggest that these populations are parts of a single, panmictic population that experiences frequent recombination. Our results also suggest that a variety of population sampling designs may be satisfactory for assessing diversity in this fungus.