Mycoviruses are widespread and replicate in diverse filamentous fungi, yeasts and oomycetes [1,2,3]. Previous studies have shown that the genomes of mycoviruses are predominately double-stranded RNA (dsRNA), positive-sense single-stranded RNA (+ssRNA), or negative-sense single-stranded RNA (-ssRNA) [4]. However, mycoviruses with single-stranded DNA (ssDNA) genomes have been reported, such as Sclerotinia sclerotiorum hypovirulence-associated DNA virus 1 (SsHADV-1) and Fusarium graminearum gemytripvirus 1 (FgGMTV1) [5,6,7]. The +ssRNA mycoviruses are currently classified into nine families, including Alphaflexiviridae, Barnaviridae, Botourmiaviridae, Gammaflexiviridae, Hypoviridae, Endornaviridae, Narnaviridae, Deltaflexiviridae, and the recently proposed family "Mycotombusviridae" [8,9,10]. Members of the proposed family "Mycotombusviridae" possess non-segmented +ssRNA genomes, 2.6- 4.6 kb in length, that usually contain two noncontiguous open reading frames (ORFs), with the second ORF encoding an RNA-dependent RNA polymerase [11, 12].

Phoma matteucciicola is a filamentous fungus that causes leaf blight disease in Curcuma wenyujin, causing huge economic losses [13]. So far, only two mycoviruses, Phoma matteucciicola partitivirus 1 (PmPV1) and Phoma matteucciicola ourmia-like virus 1 (PmOLV1), have been reported in P. matteucciicola [14, 15]. In this study, a novel (+) ssRNA virus was isolated from P. matteucciicola strain LG-01 and identified. Based on its genome organization and phylogeny, the virus was named "Phoma matteucciicola RNA virus" (PmRV1), and it is proposed to be a new unassigned mycotombus-like virus.

Provenance of the virus material

The host of PmRV1, Phoma matteucciicola strain LG-01, was isolated from C. wenyujin showing symptoms of leaf blight disease Hainan, China, and identified as P. matteucciicola based on molecular phylogeny. LG-01 was cultured on potato dextrose agar (PDA) medium covered with a cellophane membrane at 28 °C for 10 days for dsRNA extraction. The dsRNAs of LG-01 were extracted from approximately 3.0 g of frozen mycelia according to the protocol described by Morris and Dodds [16]. Subsequently, the crude virus was treated with RNase-free S1 nuclease and DNase I (TaKaRa, Dalian, China), to remove contaminating nucleic acids. A cDNA library of purified PmRV1 dsRNA was then generated using tagged random dN6 primers (5'-CGATCGATCATGATGCAATGCNNNNNN-3') and Moloney murine leukaemia virus reverse transcriptase. To obtain the full-length PmRV1 cDNA, internal gaps between the initial sequences were obtained by RT-PCR using specific primers. The terminal sequences of PmRV1 dsRNA were obtained using rapid amplification of ligase-mediated rapid amplification of cDNA ends (RLM-RACE). All purified RT-PCR products were cloned into the pMD19-T vector (TaKaRa, Dalian, China) for Sanger sequencing. For each PCR product, at least three independent clones were sequenced in both directions by Sangon Biotech (Shanghai) Co., Ltd. The complete nucleotide sequence of PmRV1 was deposited in the GenBank database, with accession number MT590656. The amino acid (aa) sequence of the putative RNA-dependent RNA polymerase (RdRp) sequence of PmRV1 was aligned with other related viral RdRp aa sequences using the ClustalX2 program. A phylogenetic tree was created by the maximum-likelihood (ML) method using Molecular Evolutionary Genetic Analysis software version 7.0 (MEGA 7.0) with 1,000 bootstrap replicates.

Sequence properties

The full-length sequence of PmRV1 is 3432 nucleotides (nt) in length with two noncontiguous ORFs, separated by 115 nt and with 5' and 3' untranslated regions (UTRs) of 333 nt and 423 nt in length, respectively (Fig. 1a). Using the Mfold program (https://unafold.rna.albany.edu), the sequences at the PmRV1 5' and 3' UTR of PmRV1 were predicted to form two potential stem-loop structures, with initial ∆G values of -11.00 kcal/mol and -10.70 kcal/mol, respectively (Fig. 1b).

Fig. 1
figure 1

a Schematic representation of the genomic organization of PmRV1. The open reading frames (ORFs) and the untranslated regions (UTRs) are showed as open bars and single lines, respectively. The gray bar indicates the RdRp domain. The nucleotide positions of the initiation and termination codons are indicated under the ORFs. The molecular weights of the predicted proteins and the length of the non-coding sequence are shown above the ORFs. b Predicted secondary structures of the 5' (left) and 3' (right) termini of PmRV1. c Amino acid sequence alignment of PmRV1 RdRps motifs with those of the related mycotombus-like viruses. Horizontal lines above the alignment show the five motifs. Shaded areas show identical amino acid residues. Asterisks, colons, and dots indicate identical amino acid residues, conservative variations, and semi-conservative variations, respectively. Conserved abnormal GDN triplets that differ from the standard (+) ssRNA amino acid sequence are highlighted in red

Full size image

ORF1, starting at nt 334 and ending at nt 1419, encodes a 361-aa polypeptide with a predicted molecular mass of 39.6 kDa (Fig. 1a). A database search with BLASTp showed that the aa sequence of PmRV1 shares 67% sequence identity with a similar hypothetical protein of Periconia macrospinosa ambiguivirus 1 (PmAV1).

ORF2, starting at nt 1534 and ending at nt 3009, encodes a 491-aa protein with a predicted molecular mass of 54.8 kDa. Database searches showed that ORF2 was most closely related to the RdRps of PmAV1 (69.7% identity) and soybean leaf-associated ssRNA virus 2 (62.1% identity). In addition, a search of the conserved domain database and multiple protein sequence alignment indicated that ORF2 encodes a protein containing a conserved viral RdRp domain with five conserved motifs (I–V) in the subfamily RdRP_3 (pfam00998) (Fig. 1c).

To analyze the relationship between PmRV1 and other mycoviruses, we constructed a phylogenetic tree using the RdRp regions of PmRV1 and other related RNA viruses. The results showed that PmRV1 and nine other unclassified ssRNA mycoviruses form a well-supported clade that is separate from the insect and plant viruses (Fig. 2). Moreover, PmRV1 is closely related to nine other unclassified mycoviruses in sequence and genome organization, which clearly differ from members of the family Tombusviridae. All 10 of these viruses contain two noncontiguous ORFs, with ORF2 containing the RdRp domain. Additionally, the PmRV1 RdRp has a GDN triad in motif III versus the GDD motif found in +ssRNA viruses (Fig. 1c). It has been reported that modification of the GDD to GDN has a detrimental effect on enzymatic activity in +ssRNA viruses [17]. Based on our results, we concluded that PmRV1 is a new member of the recently proposed family "Mycotombusviridae".

Fig. 2
figure 2

Phylogenetic analysis of PmRV1 based on the amino acid sequences of the putative RdRp domains using the maximum-likelihood (ML) method with 1000 bootstrap replicates. The scale bar indicates a genetic distance of 0.05 amino acid substitutions per site. A red star shows the novel mycovirus PmRV1

Full size image