Double-Stranded RNA High-Throughput Sequencing Reveals a New Cytorhabdovirus in a Bean Golden Mosaic Virus-Resistant Common Bean Transgenic Line

Abstract

Using double-strand RNA (dsRNA) high-throughput sequencing, we identified five RNA viruses in a bean golden mosaic virus (BGMV)-resistant common bean transgenic line with symptoms of viral infection. Four of the identified viruses had already been described as infecting common bean (cowpea mild mottle virus, bean rugose mosaic virus, Phaseolus vulgaris alphaendornavirus 1, and Phaseolus vulgaris alphaendornavirus 2) and one is a putative new plant rhabdovirus (genus Cytorhabdovirus), tentatively named bean-associated cytorhabdovirus (BaCV). The BaCV genome presented all five open reading frames (ORFs) found in most rhabdoviruses: nucleoprotein (N) (ORF1) (451 amino acids, aa), phosphoprotein (P) (ORF2) (445 aa), matrix (M) (ORF4) (287 aa), glycoprotein (G) (ORF5) (520 aa), and an RNA-dependent RNA polymerase (L) (ORF6) (114 aa), as well as a putative movement protein (P3) (ORF3) (189 aa) and the hypothetical small protein P4. The predicted BaCV proteins were compared to homologous proteins from the closest cytorhabdoviruses, and a low level of sequence identity (15⁻39%) was observed. The phylogenetic analysis shows that BaCV clustered with yerba mate chlorosis-associated virus (YmCaV) and rice stripe mosaic virus (RSMV). Overall, our results provide strong evidence that BaCV is indeed a new virus species in the genus Cytorhabdovirus (family Rhabdoviridae), the first rhabdovirus to be identified infecting common bean.

Keywords: Phaseolus vulgaris; common bean; cytorhabdovirus; dsRNA high throughput sequencing; rhabdovirus.

Figure 1

Maximum likelihood tree of cowpea mild mottle virus (CPMMV) isolates. The tree was inferred using an alignment of the complete genomes of CPMMV available in GenBank. The tree is midpoint rooted for purposes of clarity, and the nodes indicated by a black dot have Shimodaira–Hasegawa-like branch test ≥0.9. The newly sequenced CPMMV (BR:GO:14) isolate is highlighted in red.

Figure 2

Maximum likelihood tree of alphaendornaviruses. The tree was inferred using an alignment of polyproteins from genomes of all members in the genus Alphaendornavirus available in GenBank. The tree is midpoint rooted for purposes of clarity, and the nodes indicated by a black dot have Shimodaira–Hasegawa-like branch test ≥0.9. The newly sequenced alphaendornaviruses are highlighted in red.

Figure 3

Maximum likelihood trees of comoviruses. The trees were inferred using all available genomes (RNA 1 and RNA 2) of members in the genus Comovirus. The trees are midpoint rooted for purposes of clarity, and the nodes indicated by a black dot have Shimodaira–Hasegawa-like branch test ≥0.9. The bean rugose mosaic virus isolate sequenced here is highlighted in red.

Figure 4

Genome organization and phylogeny of bean-associated cytorhabdovirus (BaCV). (A) BaCV genome structure showing the six open reading frames (ORFs) (N, P, P3, M, G, and L). (B) Complementary structure of the 5′ and 3′ ends in the BaCV, barley yellow striate mosaic virus (BYSMV), colocasia bobone disease-associated virus (CBDaV), and northern cereal mosaic virus (NCMC) genomes. (C) Conserved motif sequences in the intergenic regions. (D) Maximum likelihood tree inferred using an alignment of RdRp (L protein) from 17 available genomes of members in the genus Cytorhabdovirus. The tree was midpoint rooted for purposes of clarity, and the nodes indicate by a black dot have Shimodaira–Hasegawa-like branch test ≥0.9. BaCV is highlighted in red.