Asexual evolution and forest conditions drive genetic parallelism in Phytophthora ramorum

Authors: YUZON, JENNIFER - University Of California, Davis; TRAVADON, RENAUD - University Of California, Davis; MALAR C, MATHU - Council Of Scientific And Industrial Research (CSIR); TRIPATHY, SUCHETA - Council Of Scientific And Industrial Research (CSIR); RANK, NATHAN - Sonoma State University; MEHL, HEATHER - University Of California, Davis; RIZZO, DAVID - University Of California, Davis; COBB, RICHARD - University Of California, Davis; SMALL, CORINN - University Of California, Davis; TANG, TIFFANY - University Of California, Davis; MCCOWN, HALEY - University Of California, Davis; GARBELOTTO, MATTEO - University Of California; Kasuga, Takao

Submitted to: Microorganisms
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/18/2020
Publication Date: 6/22/2020
Citation: Yuzon, J., Travadon, R., Malar C, M., Tripathy, S., Rank, N., Mehl, H., Rizzo, D., Cobb, R., Small, C., Tang, T., McCown, H., Garbelotto, M., Kasuga, T. 2020. Asexual evolution and forest conditions drive genetic parallelism in Phytophthora ramorum. Microorganisms. 8(6). https://doi.org/10.3390/microorganisms8060940.
DOI: https://doi.org/10.3390/microorganisms8060940

Interpretive Summary: Phytophthora ramorum is an invasive and devastating plant pathogen that causes sudden oak death in coastal forests in the western United States and ramorum blight in nursery ornamentals and native plants in various landscapes. P. ramorum in California propagates clonally, yet the pathogen displays diverse phenotypic variation. In this research we sequenced genomes of 43 P. ramorum isolates from California and characterized distinctive genome variations such as single nucleotide polymorphism and structural variation. By use of the observed mutations, phylogenetic relationship among the isolates was reconstructed.

Technical Abstract: Parallel evolution at the genetic level was considered unlikely because of the number of states a mutation can follow. However, increasing evidence in genetic and evolutionary studies show that parallelism at the genotypic level is more frequent than previously believed. Here, we present the example of a recently introduced, asexually propagating pathogen, P. ramorum NA1. Though originally from a sexually reproducing population, P. ramorum has colonized the Pacific Coast and has propagated asexually in its adopted land at least since 1994. Despite P. ramorum’s reproductive strategy, previous genomic analyses have identified a considerable amount of genetic diversity in the form of Structural Variants (SVs). To test if tSVs contribute to future generations of the pathogen, we first reconstructed the phylogeny of the NA1 population in California. 716 Single Nucleotide Variants (SNVs) from 43 P. ramorum samples were tested for any discordances in a phylogenetic network analysis. Very few discordances were found and most branches were bifurcating. Moreover, using a maximum distance of 1000kb to calculate r2, the population had high linkage disequilibrium with a median and mean of 1.0 further supporting the asexual hypothesis. Once we were confident in the SNV’s for the phylogenetic reconstruction, we identified SVmutations (5- 802,123 bp in length) using read depth, split read, and paired end read methods and reconstructed the phylogenetic relationships between 43 individuals from California. 158 regions with 140 amplifications, 13 translocations, and 5 loci with amplifications and translocations or deletions were identified. Loci that had at least two SVs found in the population were kept. To know if SVs arose by recombination between isolates (meiotic recombination or Horizontal Gene Transfer (HGT)) or by somatic mutations, SV DNA sequence alignments were studied for recombination between isolates. SV alignments show separation between haplotypes except for cases of gene conversion where haplotypes cluster by isolate. Moreover, very few SNV were identified in SVs. Therefore, it is unlikely that SV arise by HGT but rather by somatic mutations. With SNVs, phylogenetic relationships were reconstructed, and dwelling times and number of state transitions in were estimated for SVs in a joint analysis. With the fixed tree from the joint analysis, the full history of SVs were estimated for each SV. There were 150 parallel mutations, 1 transmissive SV, 6 parallel and transmissive SVs, and one that was neither (1 amplification and 1 deletion). From the joint analysis, the ratio of dwelling times to mutations for each SV was calculated. 22 SV loci had significantly higher dwelling times vs. number of mutations. 6 SVs with higher dwelling times-mutations were parallel and transmissive compared to other SV loci. Because these SVs are transmissive, parallel mutations and have higher dwelling times over mutations, this would suggest selection for these SVs. By describing the histories of SNVs and SVs, we are able to describe an unusual pattern in the asexual evolution of P. ramorum NA1. Genetic variation arises by somatic mutations that occur in independent lineages. Moreover, a subset of these parallel mutations are also transmissive and are potentially under positive selection.