Distinguishing Galactomyces citri-aurantii from G. geotrichum and characterizing population structure of the two postharvest sour rot pathogens of fruit crops in California.

A growth assay in lemon juice and polymerase chain reaction amplifications using newly designed species-specific primers from endopolygalacturonase and β-tubulin genes rapidly differentiated isolates of the morphologically similar fruit sour rot pathogens Galactomyces citriaurantii and G. geotrichum. Isolates of both species were collected from agricultural soils and decaying fruit at locations within and outside California, including worldwide locations, and were used in population genetic studies based on amplified fragment length polymorphic (AFLP) DNA markers. For all four geographically defined subpopulations (three counties of California and locations outside California) among 97 isolates of G. citri-aurantii and for the two subpopulations (origin within or outside California) among 35 isolates of G. geotrichum, the proportion of polymorphic loci and haplotypic diversity was high. In total, 82 unique haplotypes were identified for G. citri-aurantii for the four subpopulations and, of these, 80 haplotypes were unique among subpopulations. For G. geotrichum, 25 unique haplotypes were identified among the two subpopulations and no haplotype was shared. Indices of genetic differences (F(ST)) between subpopulations within each species were all low (e.g., 0.038 for G. geotrichum and 0.085 to 0.226 for G. citriaurantii), indicating a low level of genetic differentiation. Following clone correction, mating type segregation ratios for G. citri-aurantii did not significantly (P > 0.1) deviate from a 1:1 ratio for all four subpopulations or the entire population. Tests of the index of association (I(A)) and parsimony tree-length permutation tests (PTLPT) supported a random mating structure for clone-corrected data for the Kern, Tulare, and Ventura County subpopulations and the null hypothesis of random mating could not be rejected. Additionally, PTLPT also supported random mating for the "outside of California" population. For G. geotrichum, random mating was only tested using I(A) and PTLPT and the null hypothesis of random mating was not rejected (P > 0.05) using clone-corrected data. Further evidence that sexual recombination likely occurs in both species of Galactomyces was the lack of grouping consistency in the unweighted pair-group method with arithmetic mean clustering of AFLP data. A high confidence based on bootstrap values was obtained for only a few of the nodes in each of the two trees. A mixed reproduction system with an out-crossing sexual mating system and a prolific asexual phase is proposed for both species.