Distinctively variable sequence-based nuclear DNA markers for multilocus phylogeography of the soybean- and rice-infecting fungal pathogen Rhizoctonia solani AG-1 IA

Abstract

A series of multilocus sequence-based nuclear DNA markers was developed to infer the phylogeographical history of the Basidiomycetous fungal pathogen Rhizoctonia solani AG-1 IA infecting rice and soybean worldwide. The strategy was based on sequencing of cloned genomic DNA fragments (previously used as RFLP probes) and subsequent screening of fungal isolates to detect single nucleotide polymorphisms (SNPs). Ten primer pairs were designed based on these sequences, which resulted in PCR amplification of 200-320 bp size products and polymorphic sequences in all markers analyzed. By direct sequencing we identified both homokaryon and heterokaryon (i.e. dikaryon) isolates at each marker. Cloning the PCR products effectively estimated the allelic phase from heterokaryotic isolates. Information content varied among markers from 0.5 to 5.9 mutations per 100 bp. Thus, the former RFLP codominant probes were successfully converted into six distinctively variable sequence-based nuclear DNA markers. Rather than discarding low polymorphism loci, the combination of these distinctively variable anonymous nuclear markers would constitute an asset for the unbiased estimate of the phylogeographical parameters such as population sizes and divergent times, providing a more reliable species history that shaped the current population structure of R. solani AG-1 IA.

Keywords: allelic discrimination; multilocus genotyping; polymorphisms; primer design.

Figure 1

Haplotype network of Rhizoctonia solani AG-1 IA for locus R44L, constructed using the statistical parsimony algorithm (Templeton et al., 1992) implemented by TCS (Clement et al., 2000), where haplotypes (H1-H13) form groups represented by circles; the area of each circle refers to the relative frequency of those haplotypes in the population, and the gray tones represent their geographical origin, as shown in the legend. A dot without denomination along the network indicates a putative haplotype not sampled from the population. Probable recombinant haplotypes, identified by sequence homoplasy, were removed from the network. Squares represent the nesting design following the rules proposed by Templeton (1987), which was used to test the geographical association of haplotypes, and was implemented by GeoDis (Posada et al., 2000).