Introduction

Magnaporthe oryzae, a filamentous heterothallic ascomycete, is the most destructive pathogen of rice worldwide and the principal model organism for determining the molecular basis of fungal disease of plants. M. oryzae depends on its asexual spores (conidia) for disease establishment in nature. The conidia colonize leaves and produce necrotic lesions, or attack other parts of the plant at any stage of growth [10, 15]. Mycoviruses have been reported in many fungi [1, 4, 5]. Most mycoviruses found in filamentous fungi have double-stranded RNA (dsRNA) genomes; however, some novel single-stranded RNA (ssRNA) and single-stranded DNA (ssDNA) mycoviruses also have been reported [2, 18]. Four distinct viruses in the family Totiviridae and Chrysoviridae have been found in M. oryzae: Magnaporthe oryzae virus 1 and 2 (MoV1 and MoV2), belonging to the family Totiviridae, and Magnaporthe oryzae chrysovirus 1-A and 1-B (MoCV1-B and MoCV1-A), belonging to the family Chrysoviridae. The complete sequences of these viruses have been determined [8, 12–14, 17]. In this study, we report a novel mycovirus from strain QSP5 that was isolated from Hubei province in China. Sequence comparison and phylogenetic analysis revealed that MoV2 is closely related to members of the genus Victorivirus, which mainly infect filamentous fungi.

Materials and methods

The M. oryzae strain QSP5 was isolated from a lesion of a rice leaf collected in Hubei province, China, in 2011. The strain was stored on filter paper at −20 °C and cultured on potato dextrose agar (PDA) at 28 °C. dsRNA was extracted from the mycelia of strain QSP5 using cellulose (Sigma-Aldrich, Dorset, England) column chromatography [6, 7] and was detected using electrophoresis on 1.0 % agarose gels. Two dsRNA elements present in strain QSP5 were designated as Magnaporthe oryzae virus 3 (MoV3) and Magnaporthe oryzae chrysovirus 1-C (MoCV1-C) (Fig. 1a). To obtain the full dsRNA sequence of MoV3 in strain QSP5, the following methods were used: first, the purified dsRNA was used as a template for reverse transcription PCR (RT-PCR) amplification using a cDNA Synthesis Kit (Fermentas) with tagged random primer dN6 (5′-CGATCGATCATGATGCAATGCNNNNNN-3′) and a single specific primer (5′-CGATCGATCATGATGCAATGC-3′), which was based on the tagged random primer-dN6. The amplified PCR products were sequenced, analyzed and assembled using DNAMAN version 5.2.9 and the BLASTP program on the NCBI website [9, 16]. Second, regions corresponding to gaps in the sequences were obtained by RT-PCR using specific primers. Third, to obtain the 5′- and 3′-terminal sequence of dsRNA, an adaptor, pC3-T7loop (P-GGATCCCGGGAATTCGGTAATACGACTCACTATATTTTTATAGTGAGTCGTATTA-OH), was added to the terminal sequence of the purified dsRNA segments. The dsRNA with pC3-T7loop was then used as a template for RT-PCR using the specific primer pC2 (5′-CCGAATTCCCGGGATCC-3′), which was based on pC3-T7loop. Sequence analysis, alignments, phylogenetic analysis and identification of protein motifs was done by using the DNAMAN, CLUSTALX, MEGA5.0 and BLAST programs at the National Center for Biotechnology Information website (http://blast.ncbi.nlm.nih.gov/Blast.cgi) [3, 9].

Fig. 1
figure 1

(a) dsRNAs in M. oryzae strain QSP5. Total dsRNA extracted from mycelia of strain QSP5. dsRNA samples were fractionated on a 1.0 % agarose gel and stained with ethidium bromide. The positions of host genomic DNA and viral dsRNAs (MoV3, MoCV1-C) of strain QSP5 are indicated. Lane M, Hind III-digested λ DNA marker. (b) Diagrammatic representation of the genomic organization of MoV3. (c) Amino acid sequence alignment of RdRps between MoV3 and selected viruses (GaRV-L1, MoV2, CmRV, HvV190S and MoV1) of the genus Victorivirus. Conserved motifs are motif I to VIII, asterisks represent identical residues, and colons represent similar residues

Full size image

Sequence data

The full-length cDNA sequence of MoV3 has been deposited in the GenBank databases with accession number KP893140. The complete genome of MoV3 in strain QSP5 was 5181 bp, with a GC content of 61 %. Sequence analysis indicated that MoV3 has two overlapping open reading frames (ORF1 and ORF2). The 5′-proximal ORF1 (nt 283-2634) encodes a putative coat protein (CP) of 783 amino acid residues with a molecular weight of 80,939 Da that shows high similarity to the CPs of viruses in the family Totiviridae, particularly to those of Gremmeniella abietina RNA virus L1 (GaRV-L1, 73 % identity) [11], Magnaporthe oryzae virus 2 (MoV2, 67 % identity) [8], and Coniothyrium minitans RNA virus (CmRV, 63 % identity) [3]. Moreover,there is an Ala/Gly/Pro (A/G/P)-rich region near the C-terminal sequence (amino acid position from Gly711 to Gln783) of MoV3 CP, which is an unique molecular feature of members of the genus Victorivirus in the family Totiviridae [19]. The 3′-proximal ORF2 (nt 2631-5108) encodes a putative RNA-dependent RNA polymerase (RdRp) of 824 amino acid residues with a molecular weight of 90,506 Da, which showed a high similarity to the RdRps of viruses in the family Totiviridae, such as GaRV-L1 (58 % identity), MoV2 (52 % identity) and CmRV (49 % identity). According to the ICTV rules for species demarcation in the genus Victorivirus, sequence identity less than 60 % in either the CP or the RdRP protein indicates that the viruses belong to different species [19]. Although the CP sequence of MoV3 shares 67 % identity with MoV2, the RdRp sequence shares only 52 % identity with MoV2, which is less than the 60 % cutoff for species delimitation. Thus, MoV2 could be a new species in the genus Victorivirus. The stop codon of ORF1 overlaps the start codon of ORF2 with the tetranucleotide sequence AUGA, which is characteristic of the genus Victorivirus of the family Totiviridae. The 5′ untranslated region (UTR) and 3′-UTR contains 282 and 73 nucleotides, respectively (Fig. 1b).

In addition, a multiple sequence alignment of the deduced amino acid sequences of the RdRps of MoV3 and five other victoriviruses (GaRV-L1, MoV2, CmRV, HvV190S and MoV1) in family Totiviridae revealed that MoV3 has eight conserved motifs that are typical of members of the family Totiviridae. In addition, the eight conserved regions exhibited high sequence similarity among these viruses (Fig. 1c).

Phylogenetic analysis of the RdRps between MoV3 and selected viruses in the families Totiviridae and Chrysoviridae showed that the putative RdRp of MoV3 clusters closely with the members of genus Victorivirus (GaRV-L1, MoV2, CmRV, HvV190S and MoV1). Furthermore, phylogenetic analysis of the CP of MoV3 and that of viruses in the families Totiviridae and Partitiviridae also indicated that the putative CP of MoV3 clusters with members of genus Victorivirus (Fig. 2).

Fig. 2
figure 2

Phylogenetic analysis of the RdRp and CP genes of MoV3 and related viruses in the families Totiviridae, Chrysoviridae and Partitiviridae. The unrooted phylogenetic tree was constructed by the neighbor-joining method using the program MEGA 5.0. Bootstrap values (1000 replicates) are shown at the nodes, and the scale bar (0.1) corresponds to genetic distance. The position of MoV3 is indicated by a red star. The sequences of the RdRp and CP were obtained from the GenBank database. The mycoviruses are as follows (abbreviation; GenBank accession no.): Gremmeniella abietina RNA virus L1 (GaRV-L1; AAK11656.1), Coniothyrium minitans RNA virus (MoV2; YP_001649206.1), Coniothyrium minitans RNA virus (CmRV; YP_392467.1), Helminthosporium victoriae virus 190S (HvV190S; NP_619670.2), Magnaporthe oryzae virus 1 (MoV1; P_122352.1), Leishmania RNA virus 1-1 (LRV1-1; NP_041191.1), Leishmania RNA virus 2-1 (LRV2-1; NP_043465.1), Trichomonas vaginalis virus 1 (TvV1; ABF57713.1), Trichomonas vaginalis virus 2 (TvV2; AED99806.1), Trichomonas vaginalis virus 4 (TvV4; AED99794.1), Trichomonas vaginalis virus 3 (TvV3; AED99800.1), Saccharomyces cerevisiae virus L-A (ScV-L-A; NP_620495.1), Saccharomyces cerevisiae virus L-BC (ScV-L-BC; NP_042581.1), Cryphonectria nitschkei chrysovirus 1 (CnCV1; ACT79256.1), Verticillium dahliae chrysovirus 1 (VdCV1; ADG21213.1)

Full size image

Our present results showed that MoV3 is a new member of the genus Victorivirus of the family Totiviridae. Our report provides preliminary evidence that two distinct viruses, MoV3 and MoCV1-C, from the families Totiviridae and Chrysoviridae, could exist in the same host, which is similar to what has been reported previously for Helminthosporium victoriae, which can be co-infected by the victorivirus Helminthosporium victoriae virus 190S and the chrysovirus Helminthosporium victoriae virus 145S [5].