Characterization by Small RNA Sequencing of Taro Bacilliform CH Virus (TaBCHV), a Novel Badnavirus

Abstract

RNA silencing is an antiviral immunity that regulates gene expression through the production of small RNAs (sRNAs). In this study, deep sequencing of small RNAs was used to identify viruses infecting two taro plants. Blast searching identified five and nine contigs assembled from small RNAs of samples T1 and T2 matched onto the genome sequences of badnaviruses in the family Caulimoviridae. Complete genome sequences of two isolates of the badnavirus determined by sequence specific amplification comprised of 7,641 nucleotides and shared overall nucleotide similarities of 44.1%‒55.8% with other badnaviruses. Six open reading frames (ORFs) were identified on the plus strand, showed amino acid similarities ranging from 59.8% (ORF3) to 10.2% (ORF6) to the corresponding proteins encoded by other badnaviruses. Phylogenetic analysis also supports that the virus is a new member in the genus Badnavirus. The virus is tentatively named as Taro bacilliform CH virus (TaBCHV), and it is the second badnavirus infecting taro plants, following Taro bacilliform virus (TaBV). In addition, analyzes of viral derived small RNAs (vsRNAs) from TaBCHV showed that almost equivalent number of vsRNAs were generated from both strands and the most abundant vsRNAs were 21 nt, with uracil bias at 5' terminal. Furthermore, TaBCHV vsRNAs were asymmetrically distributed on its entire circular genome at both orientations with the hotspots mainly generated in the ORF5 region.

Fig 1. Genome organization of Taro bacilliform CH virus (TaBCHV).

The putative ORFs of TaBCHV are indicated by rectangles, domains identified within ORF 3 are shown (A), contigs obtained from samples T1 (C1– C5) and T2 (C'1–C'9) are presented by black lines (A1), and fragments F1–F8 amplified from the first cycle of PCR (A2) and F'1–F'7 amplified from the second cycle of PCR (A3) are represented by arrows. The genome organization of Taro bacilliform virus (TaBV) (B) is outlined to show its difference with that of TaBCHV.

Fig 2. Comparison of amino acid sequences of domains highly conserved in the polyproteins encoded by ORF3 of badnaviruses and a tungrovirus.

The virus names and the positions of starting amino acid are indicated before each sequence. Identical (*) and conserved (:) amino acids are marked.

Fig 3. Neighbor-joining phylogenetic trees of badnaviruses generated from the full genomic sequences (A) and putative amino acid sequences of ORF3 (B).

The phylogenetic trees were rooted by using the genome sequence of Rice tungro bacilliform virus (RTBV) (A) and the polypeptide of RTBV (B). Branch lengths are proportional to genetic distances. Numbers at the nodes of the branches represent bootstrap values (1000 replicates).

Fig 4. Size distribution of vsRNAs derived from TaBCHV-1 and TaBCHV-2 (A) and the relative frequency of 5' terminal nucleotide of 21- and 22-nt vsRNA (B).

Blue and red bars indicate sense and antisense vsRNAs respectively.

Fig 5. The distribution of 21- and 22-nt vsRNAs on the genomes of TaBCHV-1 and TaBCHV-2.

The bars above the axis represent sense reads; those below represent antisense reads.

Fig 6. The silencing hot spots of TaBCHV-1(A) and TaBCHV-2(B) genomes and the predicated secondary structure around the vsRNA at 6432 nt (C).

The start and stop positions of the vsRNA₆₄₃₂ are marked by bold arrows.