Evolution and Phylogeny of Soybean Mosaic Virus Based on 143 Complete Genomes

Abstract

Soybean mosaic virus (SMV) of the genus Potyvirus is an important virus in cultivated soybeans. Here, we obtained 7 SMV genomes from soybean germplasms using RNA sequencing and conducted a comprehensive evolutionary and phylogenetic study of 143 SMV genomes derived from 10 plant species and 12 countries. The phylogenetic tree we constructed using coding DNA sequences revealed the existence of nine clades of SMV isolates/strains. Recombination analysis revealed 76 recombinant events and 141 recombinants in total. Clades 1 and 3 contain the most common SMV pathotypes, including G1 through G7, which are distributed worldwide. Clade 2 includes several Chinese SMV pathotypes. The SMV isolates were further divided into two groups. The SMV isolates in the first group, including clades 8 and 9, were identified from Pinellia and Atractylodes species, whereas those in the second group (clades 1 through 7) were mostly found in cultivated soybeans. The SMV polyprotein undergoes positive selection, whereas most mature proteins, except for the P1 protein, undergo negative selection. The P1 protein of SMV isolates in group 1 may be highly correlated with host adaptation. This study provides strong evidence that recombination and plant hosts are powerful forces driving the genetic diversity of the SMV genome.

Keywords: diversity; evolution; genome; phylogeny; soybean mosaic virus.

Figure 1

Information for 143 SMV genomes and the 10 identified clades. Pie charts display information for each country where SMV genomes were obtained (A) and the host plants from which SMV genomes were obtained (B). The bar graph shows the number of SMV genomes based on the year of identification (C). Information for the 10 clades of 143 SMV genomes that were identified using phylogenetic analysis (D).

Figure 2

Phylogenetic tree of 143 SMV genomes constructed based on complete coding DNA sequences, generated using the maximum likelihood method. A total of 143 sequences with 9805 sites covering whole coding DNA sequences were used for phylogenetic tree construction using IQ-TREE with GTR + F+I + G4 for the best-fit model. The accession number and isolate (strain) name for each SMV genome are indicated. Both SH-aLRT and UFBoot support values are indicated on the branches. A total of 10 clades were found, and they are shown in 10 different colors. The isolate Uraria (GenBank LC037232.1) in clade 10 was used as an outgroup. Detailed information for individual SMV genomes can be found in Table S1.

Figure 3

Composition of SMV isolates in the three major clades that became apparent when the isolates were analyzed based on their country of origin and host plant. The number of SMV isolates in each of the three major clades (clades 1 through 3) based on country of origin (A) and host plant (B) are indicated.

Figure 4

Phylogenetic network tree of 143 SMV genomes. The network tree was rooted using the complete genome sequences of 143 SMV genomes and contains 2 groups (Group 1 and Group 2) of SMV genomes (A). Group 1 is shown in a green oval in the network tree and was then further magnified (B). Group 2 is shown in a pink oval in the network and then was further magnified (C) The seven known SMV strains (G1–G7) are shown in the light blue colored box. The accession numbers and isolate names are also indicated. Using the nexus alignment file in the supplemental file S1, the network tree can be visualized in the SplitsTree program.

Figure 5

Recombination breakpoint distribution plots for the SMV genome. SMV genome organization showing individual protein regions (A). Recombination breakpoint distribution plot (B) and respective statistical log converted p-value (C). Locally and globally significant breakpoint clusters were identified using 95% and 99% permutation tests. Recombination rate plot (D).

Figure 6

Number of recombinant events and recombinants in individual SMV proteins or genomes.

Figure 7

Genetic diversity of SMV proteins or genomes in different groups of SMV. Nucleotide diversity (Pi) (A) and average number of segregating sites (θw) (B) of individual SMV proteins or genomes in the three groups. ‘All’ indicates all 143 SMV genomes, and group 1 and group 2 include 6 and 137 SMV genomes, respectively.

Figure 8

Average dN/dS ratio of SMV proteins in three groups. ‘All’ indicates the 143 SMV genomes, whereas group 1 and group 2 include 6 and 137 SMV genomes, respectively.