A novel single-stranded RNA virus isolated from a phytopathogenic filamentous fungus, Rosellinia necatrix, with similarity to hypo-like viruses

Abstract

Here we report a biological and molecular characterization of a novel positive-sense RNA virus isolated from a field isolate (NW10) of a filamentous phytopathogenic fungus, the white root rot fungus that is designated as Rosellinia necatrix fusarivirus 1 (RnFV1). A recently developed technology using zinc ions allowed us to transfer RnFV1 to two mycelially incompatible Rosellinia necatrix strains. A biological comparison of the virus-free and -recipient isogenic fungal strains suggested that RnFV1 infects latently and thus has no potential as a virocontrol agent. The virus has an undivided positive-sense RNA genome of 6286 nucleotides excluding a poly (A) tail. The genome possesses two non-overlapping open reading frames (ORFs): a large ORF1 that encodes polypeptides with RNA replication functions and a smaller ORF2 that encodes polypeptides of unknown function. A lack of coat protein genes was suggested by the failure of virus particles from infected mycelia. No evidence was obtained by Northern analysis or classical 5'-RACE for the presence of subgenomic RNA for the downstream ORF. Sequence similarities were found in amino-acid sequence between RnFV1 putative proteins and counterparts of a previously reported mycovirus, Fusarium graminearum virus 1 (FgV1). Interestingly, several related sequences were detected by BLAST searches of independent transcriptome assembly databases one of which probably represents an entire virus genome. Phylogenetic analysis based on the conserved RNA-dependent RNA polymerase showed that RnFV1, FgV1, and these similar sequences are grouped in a cluster distinct from distantly related hypoviruses. It is proposed that a new taxonomic family termed Fusariviridae be created to include RnFV1 and FgV1.

Keywords: Rosellinia necatrix; fusarivurs; hypovirus; novel mycovirus; ssRNA virus; transcriptome shotgun assembly.

Figure 1

Gel electrophoretic profiles of dsRNAs isolated from R. necatrix NW10. DsRNAs purified from R. necatrix W779 infected with Rosellinia necatrix megabirnavirus 1 (RnMBV1) (lane W779), R. necatrix NW10 infected with Rosellinia necatrix fusarivirus 1 (RnFV1), and R. necatrix W1118 infected with Rosellinia necatrix quadrivirus 1 (RnQV1) (lane W1118) were analyzed in parallel by 1.2% agarose gel electrophoresis. A GeneRuler 1-kb DNA ladder (Thermo Scientific) was also electrophoresed (Marker).

Figure 2

Genomic organization of RnFV1-NW10 and comparison with those of FgV1 and related virus-like sequences. The RnFV1 genome is 6286 nts in length excluding the poly(A) tail with two ORFs. The colored boxes and lines represent open reading frames (ORFs) and non-coding sequences, respectively. The boxes in ORF1 represent conserved domains including RNA-dependent RNA polymerase (RdRp, red) and RNA helicase (Hel, green) domains. The sizes of the 5′-UTR, inter-ORF regions, ORF1 and ORF2 are shown above or below the diagram. Regions used for probes in Northern blotting (Figure 4) are indicated by thick arrows: probe 1, 1440–2180 nt; probe 2b, 5471–6100 nt. The genetic organizations of FgV1 and the assembled contigs of the related virus-like sequences from the transcriptome shotgun assembly (TSA) or expressed sequence tags (ESTs) libraries (see Figure S1 in Supplementary Material) are also similar to RnFV1. Putative transmembrane (TM) and coiled-coil domains (CCD) are denoted by thick pink solid and thick orange dashed lines, respectively (see Figures S2 and S3 in Supplementary Material). The possible undetermined sequences are shown by light-gray dashed lines. Arrowheads indicate the inter-ORF regions and the nt numbers above them denote space. “Fs” at ShTSA_fusa and SiTSA1_fusa denote internal stop codons. A scale bar (1-kb increment) is presented at the bottom.

Figure 3

Maximum likelihood (ML) phylogeny of RnFV1-NW10 and its related virus and virus-like sequences. Phylogenetic trees were constructed using PhyML 3.0 based on the multiple aa sequence alignment of the RNA-dependent RNA polymerase (RdRp) (A) and RNA helicase (Hel) domains (B), as shown in Figure S4A and B, respectively, in Supplementary Material. The cured alignments were subjected to phylogenetic analysis. The selected best-fit models for each alignment are shown at the bottom. Red and green shaded circles with dashed line: fusariviruses proposed in this study. Gray circles: the members of the genera Alphahypovirus and Betahypovirus proposed by Yaegashi et al. (2012). The related virus-like sequences denoted by asterisks were also included in the analyses. GenBank/Refseq accession numbers of the sequences are listed in Table 1 with the addition of PPV (plum pox virus-NAT; D13751), Sclerotinia sclerotiorum hypovirus 2 isolate 5472 (SsHV2; NC_022896), ShTSA_hypo (Sclerotinia homoeocarpa TSA sequence; JW820052) and StTSA_hypo (S. trifoliorum TSA sequence; JP556778). Asterisks show the virus-like sequences. The branch support values were estimated by the approximate likelihood ratio test (aLRT) with a SH-like algorithm (Anisimova and Gascuel, 2006) (only values greater than 0.9 are shown).

Figure 4

Northern blotting (A,B) and 5′ RACE analyses (C) of RT60-2/RnFV1. Total RNA fractions (0.5 μg/lane), obtained from virus-infected RT60-2/RnFV1 (+) (lane 2) and virus-free RT60-2 (−) (lane 1), were electrophoresed under denaturing conditions and subjected to Northern blotting as described in Materials and Methods (A,B). In vitro synthesized, 6.3-kb genome-sized (lane 3, FL, 10 ng) and 1.7-kb shorter-sized transcripts (lane 4, Short, 0.1 ng) were analyzed in parallel (A,B). The short transcript corresponds to positive-sense RNA spanning ORF2 plus the 3′-UTR region (see Figure 5A for in vitro transcription start site G at position 4652) and mimics hypothetical subgenomic RNA. Two cDNA probes specific for regions of ORF1 (probe 1) and ORF2 (probes 2b) (see Figure 2 for positions) were used in B and A, respectively. Faint bands detected at a similar position to the short transcript by each probe (shown by white arrows), these are unlikely to be subgenomic RNA (see the text and Figure 4C). Classical 5′-RACE, entailing primer-specific cDNA synthesis, dC-tailing, and PCR amplification, was conducted aiming at subgenomic RNA detection (C). The template was 0.5 μg of total RNA from virus-infected RT60-2 (lane 1). As controls, 0.5 μg of total RNA from virus-free RT60-2 without (−, lane 7) or with ten-fold serial dilutions of short (St) 1.7 kb RNA, ranging from 2.6 ng to 2.6 pg (lanes 3–6), was analyzed in parallel. Two viral specific primers 5′-CTTGTATTGCTTGCGGGAGGCTTCC-3′ (map positions 5102-5078) and 5′-TCAAGCGAAGAGAGTTCAATTC-3′ (map positions 4989-4968) were used for cDNA synthesis and DNA amplification. Similarly 10 ng of genome-sized transcript (lane FL) was used in this assay. PCR products were examined by 1.5% agarose gel electrophoresis.

Figure 5

Comparison of the inter-ORF sequences and their flanking regions between RnFV1-NW10 and the selected virus-like sequences (A) or within FgV1 (B). The upper-case bold letters in red and blue indicate the stop codon of the upstream ORF and the start codon of the downstream ORF, respectively. The in-frame stop or start codons in the inter-ORF regions are denoted by bold, lower-case italic letters except for the in-frame downstream AUG codon of SiTSA1_fusa. The ORF3 start-codon that overlaps the ORF2 stop codon in FgV1 is underlined. Asterisks, dots, and — indicate identical and similar nucleotides, and gaps, respectively. All potential start and stop codons are highlighted by pink and light blue, respectively. The G at position 4652 for in vitro transcription start site is indicated (see Figures 4A,B).

Figure 6

Colony morphology of RnFV1-infected and virus-free isogenic strains. RnFV1-NW10 was transferred from R. necatrix NW10 to somatically incompatible fungal strains (RT60-2, RT37-1, and RT45-1) via co-culturing in the presence of zinc chloride. Infection of the fungal recipient strains by RnFV1 was confirmed by RT-PCR (A). Total RNA fractions, isolated from RT60-2, RT37-1, and RT45-1 before (−) and after (+) virus transmission, were used in RT-PCR. The donor strain NW10 was also included in the analysis. An oligo (dT) was used as a primer for cDNA synthesis, while a forward primer 5′-CATTCGCAAGAGCCTGAGC-3′ (map positions 5471-5489) and a reverse primer 5′-AGCCAACGCCTAACAAGCAC-3′ (map positions 6100-6081), were used as primers for DNA amplification. The genetic backgrounds of recipient fungal strains were examined by UP-PCR using three primer sets (AS4, AS15, and AS4+AS15) (B). All recipient isolates (RT60-2/RnFV1) manifested UP-PCR profiles identical to their original strain, RT60-2 (lane R), and not to NW10 (lane N), confirming lateral virus movement. Abbreviations (N and R) below each gel indicate the banding pattern for their parental strains. Distinguishing banding patterns of two parental strains are boxed in red in each UP-PCR gel. A GeneRuler 1 kb DNA ladder (Thermo Scientific, M) was also electrophoresed. As an example, virus-uninfected (RT60-2) and virus-infected (RT60-2/RnFV1) fungal strains were grown on PDA and Vogel's medium for 6 days in the dark and photographed (C).