First author: United States Department of Agriculture–Agricultural Research Service, 1636 East Alisal St., Salinas, CA 93905; and second author: Plant Pathology and Microbiology, University of California, Riverside 92521.
Although Phytophthora cinnamomi is heterothallic, there are few instances of successful crossing in laboratory experiments, and analysis of field populations indicates a clonally reproducing population. In the absence of sexual recombination, the ability to monitor mitochondrial haplotypes may provide an additional tool for identification of clonal isolates and analysis of population structure. To determine mitochondrial haplotypes for this species, seven mitochondrial loci spanning a total of 6,961 bp were sequenced for 62 isolates representing a geographically diverse collection of isolates with A1 and A2 mating type. Three of the regions were primarily intergenic regions between trnG and rns, rns and nad3, and nad6 and cox1, while the remaining loci spanned cox2, nad9, rps10, and secY coding regions and some of the flanking spacer regions. In total, 45 mitochondrial haplotypes were identified (75% of the total isolates examined) with differences due to single-nucleotide polymorphisms (SNPs, totaling 152 bp) and length mutations (17 indels >2 bp representing a total of 910 bp in length). SNPs were the predominate mutation in the four coding regions and their flanking intergenic regions, while both SNPs and length mutations were observed in the three primarily intergenic regions. Some of the length mutations in these regions were due to addition or loss of unique sequences while others were due to variable numbers of subrepeats (in the trnG-rns region, there were 3 to 12 copies of a 24-bp subrepeat sequence that differentiated 17 haplotypes). Network analysis of the haplotypes identified eight primary clades, with the most divergent clade representing primarily A1 isolates collected from Papua New Guinea. The isolate grouping in the network corresponded to mating type and previously published isozyme classifications, with three exceptions: a haplotype representing an A1 mating type (H29) was placed well within the A2 mating type haplotype grouping, one haplotype (H26) had isolates with two isozyme classifications, and one isozyme group was represented on separate network clades, suggesting that recombination has occurred in the past. Among the 62 isolates examined, several examples were identified of isolates recovered from different geographic regions having the same mitochondrial haplotype, suggesting movement of isolates via plant material. Analysis of the data set to determine whether fewer loci could be sequenced to classify haplotypes indicated that the trnG-rns and rns-nad6 loci would classify 87% of the haplotypes identified in this study, while additional sequencing of the nad9 or secY loci would further differentiate the remaining six haplotypes. Based on conservation of gene order in Phytophthora spp., the trnG-rns locus should be useful for mitochondrial haplotype classification in other species, as should the cox2, nad9, rps10, and secY loci. However, the rns-nad3 and nad6-cox1 loci span regions that can have a different gene order in some Phytophthora spp.