Clade divination – phenotypic data facilitates epidemiological inferences from soybean dwarf virus sequence analysis

Abstract

Soybean dwarf virus (SbDV; family Tombusviridae, genus Luteovirus, species Luteovirus glycinis), transmitted solely by aphids in a persistent manner, can cause significant losses in cultivated Fabaceae. SbDV isolates are categorised into four strains; YP, YS, DP and DS. Yellowing (Y) and dwarfing (D) strains differ in their symptom expression and host range. P strains are transmitted by Acyrthosiphon pisum (pea aphid), and S strains are transmitted by Aulacorthum solani (foxglove aphid). Genetically, Y and D strains separate into two clades at every genomic region except for the N-terminal region of the read through domain (N-RTD) in which P and S strains separate. SbDV diversity in Australia has yet to be significantly investigated and so complete genome sequences were obtained from 41 isolates infecting cultivated Fabaceae in two regions of Australia (‘northeast’ and ‘southwest’). Phylogenetic analyses were conducted on the whole genome and the N-RTD of these sequences together with 50 sequences available on GenBank. At the whole genome level, isolates separated into D and Y clades. At the N-RTD level, with two exceptions, isolates separated into P and S clades. All south-west isolates were in the Y clade, whilst north-east isolates were in both the Y and D clades. All isolates sequenced were in the P group suggesting Ac. pisum is the primary vector of SbDV in these regions. Although biological evidence suggests S strain isolates are common in the southeast of Australia, just one previously sequenced Tasmanian isolate was available and was in the Y and S clades. Relevant biological data available for each isolate supported inferences made from phylogenetic clade. This study suggests that at least three of the four SbDV strains are present in Australia and is an example of the importance of continued collection of phenotypic data to provide more power to genetic analyses.