Nuclear genome population genomics of the potato wart pathogen (Synchytrium endobioticum) reveals high-resolution relationships of global strains
TL;DR
A robust single-nucleotide polymorphism (SNP) discovery strategy is designed and applied to existing and novel sequence data sets that represent the global diversity in potato production systems and indicates that potato wart was introduced into Canada from a European source.
Abstract
ABSTRACT The chytrid Synchytrium endobioticum causes potato wart disease and is a regulated quarantine pathogen. Due to its obligate association with the potato host, obtaining pure genome sequences has been technically challenging. Previous population studies were limited to microsatellites or mitochondrial genomes, which capture only a small portion of the pathogen’s genetic diversity. To enable the first population genomic analyses of the nuclear genome of S. endobioticum, we designed and applied a robust single-nucleotide polymorphism (SNP) discovery strategy to existing and novel sequence data sets that represent the global diversity in potato production systems. We demonstrate that reliable nuclear genome data can be recovered from contaminated environmental samples, where S. endobioticum constitutes only ~5% of total reads. After stringent filtering, 74 high‑quality data sets for isolates from 10 countries and 12 pathotypes were retained. Phylogenetic analyses of genome-wide SNP alignments, some with up to 188 817 SNPs, identified three well-supported nuclear phylogroups that largely corresponded to mitochondrial haplogroups: (i) a basal Peruvian clade; (ii) a European/Canadian clade containing mostly pathotypes 1(D1), 2(G1), 6(O1), and 8(F1); and (iii) a clade with pathotypes 18(T1) and 38(Nevşehir). Using a subset of 13 327 homozygous SNPs, population structure analysis and phylogenetic networks confirmed three genetic clusters with limited admixture. Our analyses indicate that potato wart was introduced into Canada from a European source. Pathotype designations do not correspond to exclusive monophyletic lineages. Analysis of the AvrSen1 effector revealed that distinct loss-of-function variants can occur in multiple nuclear phylogroups, suggesting independent emergence of some virulence phenotypes. Isolates contain a reservoir of genetic variation, which may allow adaptation to changing host resistance pressures. This nuclear genome approach provides significantly greater resolution than previous molecular markers used for strain tracking. Furthermore, the solid phylogenetic framework developed here will support future effector-based pathotype inference and will aid regulators in surveillance and quarantine decisions for this high-risk pathogen.