Population Genomics Reveals Distinct Lineage of the Asian Soybean Rust Fungus Phakopsora pachyrhizi in the United States of America Unrelated to Brazilian Populations

Abstract

Asian soybean rust (ASR), caused by the obligate biotrophic fungus Phakopsora pachyrhizi, was first reported in the continental United States of America (USA) in 2004 and over the years has been of concern to soybean production in the United States. The prevailing hypothesis is that P. pachyrhizi spores were introduced into the United States via hurricanes originating from South America, particularly hurricane Ivan. To investigate the genetic diversity and global population structure of P. pachyrhizi, we employed exome-capture based sequencing on 84 field isolates collected from different geographic regions worldwide. We compared the gene-encoding regions from all these field isolates and found that four major mitochondrial haplotypes are prevalent worldwide. Here, we provide genetic evidence supporting multiple incursions that have led to the currently established P. pachyrhizi population of the United States. Phylogenetic analysis of mitochondrial genes further supports this hypothesis. We observed limited genetic diversity in P. pachyrhizi populations across different geographic regions, suggesting a clonal population structure. Additionally, this study is the first to report the F129L mutation in the Cytb gene outside South America, which is associated with strobilurin tolerance. This study provides the first comprehensive characterisation of P. pachyrhizi population structures defined by genetic evidence from populations across major soybean-growing regions.

Keywords: Phakopsoraceae; Pucciniales; fungicide resistance; lineages; population structure; soybean rust.

FIGURE 1

Sampling locations of Phakopsora pachyrhizi analysed by exome‐capture sequencing. Dots on the map represent the locations and numbers of the samples collected. The colour corresponds to the host where the samples/isolates originated from.

FIGURE 2

Evaluation of the gene space coverage by exome‐capture sequencing of Phakopsora pachyrhizi. Histograms representing the breadth (upper panel) and depth (lower panel) coverage of P. pachyrhizi genes by exome‐capture sequencing reads of isolate UFV02. The whole gene set (UFV02 “Gene Catalogue”) and a set of expressed genes were evaluated in (a) and (b), respectively. A gene was judged “captured” if the gene had a breadth coverage of over 0.8 and a depth coverage of over 5.

FIGURE 3

Mt‐haplotype network inferred from nucleotide polymorphisms in the mt genome of Phakopsora pachyrhizi. (a) Pie charts represent mt‐haplotypes and their geographic origins. The size of each node is proportional to the number of isolates sharing that mt‐haplotype. Connecting edges indicate genetic relationships, with each tick mark representing a single single‐nulceotide polymorphim (SNP) or InDel. (b) A non‐synonymous SNP (F129L) associated with resistance to quinone outside inhibitor (QoI) fungicides is marked by a blue vertical bar. Synonymous SNPs (no amino acid change) are marked by green vertical bars. Genotypes at SNP marker 16,801 (converted to KASP assay) in four mt‐haplotypes are shown below.

FIGURE 4

Population structure and phylogenetic relationships of Phakopsora pachyrhizi samples inferred from nuclear genome single‐nucleotide polymorphisms (SNPs). (a) Discriminant analysis of principal components (DAPC) using SNP markers from 53 samples, with principal components 1 and 2 plotted. (b) Probability of population membership is shown on the y‐axis for each sample, grouped in bins along the x‐axis. The stacked bar chart represents assigned population membership probabilities, with clusters distinguished by different colours. (c) Phylogenetic tree inferred from 33,634 SNP markers. Nodes with bootstrap support greater than 80% are marked with red dots.