According to the latest virus classification released by the International Committee on Taxonomy of Viruses (ICTV), the family Narnaviridae belongs to positive single stranded RNA viruses (+ssRNA), and now only includes the genus Narnavirus (https://ictv. global/taxon omy). The viral genomes of family Narnaviridae have not been reported to encode the coat protein, therefore there is no typical viral particles [1]. Since it was named Saccharomyces cerevisiae narnavirus 23S (ScanV-23S) by Hillman et al. in 2011 after the first discovery of 23S RNA virus in yeast in 1998 [2], many Narnavirus have been found in protozoa and filamentous fungi [3-5]. With the further studies, the Narnavirus have been found to affect the interactions of host with other organisms [6]. In recent years, regarding the classification of +ssRNA viruses, the most peculiar Narnavirus type is splipalmivirus because no viruses with divided RdRP domains are found in other members of the kingdom Orthornavirae [7], which RNA segment numbers vary from two to seven [7-10].
Beauveria bassiana Vuillemin is an entomogenous fungus, which is used as insect pest biocontrol agent worldwide[11]. Previous studies have shown that B. bassiana could be infected by mycoviruses, with some viruses had no influence on it, while others could affect its biological characteristics and pathogenicity to insect pests [12,13]. So far, it has been reported that double stranded RNA viruses (dsRNA) were universal in B. bassiana [14,15], but there are few records of ssRNA mycovirus, especially Narnavirus, although a single stranded small Narna-like virus, Beauveria bassiana small Narna-like virus (BbSNLV) was reported by Kotta-Loizou and Coutts in 2017 [15]. In the present study, we identified a novel +ssRNA virus with typical characteristic of splipalmiviruses belonging to Splipalmiviridae (likely a new family),and suggested temporarily nominating it as Beauveria bassiana splipalmivirus (BbSpV1).
The B. bassiana strain GusB1 with the original host Ostrinia nubilalis Hubner, was identified and provided by Professor Georgy Lednev, from the Russia Plant Protection Institute, Saint Petersburg, Russia. The strain GusB1 was stored in sterile water in the form of conidia at 4℃. When extracting dsRNA, the strain was cultured on potato glucose agar (PDA) medium in darkness at 26℃ for seven to ten days until the conidia appeared. The RNA extraction was performed according to description of Morris and Dodds [16]. The extracted RNA was treated with DNase I (TaKaRa, Dalian, China), and analyzed with 1% agarose gel electrophoresis, with no obvious degradation was found. Then, the samples were sent to Novogene Bioinformation Technology Co., Ltd. (Beijing, China) for the construction of the Next Generation Sequencing (NGS) library using the HiSeq-PE150 platform. After sequencing, 3.8 Gb of raw data was obtained and analyzed using SPAdes (V3.13.1) to obtain RNA sequences. Firstly, the clean data were matched to the genome sequence of B. bassiana ARSEF 2860 (Genbank No. ADAH00000000), and the unmatched sequence were used to identify the genome sequence of the virus. All sequences of +ssRNA1, +ssRNA2, and +ssRNA3 were obtained through rapid amplification of the cDNA end (RACE) [17]. In detail, after the electrophoretic separation on 1% agarose gel and extraction, reverse transcription was performed with primer RACE3RT (CGATCGATCATGATGGATGCNNNNNNNNN), and PCR amplification was performed with primer RACE3 (CGATCGATCATGATGGATCGC) to obtain the sequence. Sequence comparisons were performed using the Local Sequence Alignment Search Tool (BLAST) from the National Center for Biotechnology Information (NCBI) in the United States. Using NCBI's ORF Finder (https://www.ncbi.nlm.nih.gov/orffinder/) and Conservative Domain Search Service (CD Search) (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) to predict ORFs and determine conservative structural domains [18]. Phylogenetic analysis was conducted using the maximum-likelihood (ML) method in MEGA X (https://www.mega.com/) and bootstrap method for 1000 replicates [19].
The viral genome of BbSpV1 consists of three +ssRNA sequences with lengths of 2149 bp (Genbank no. OR737626), 2133 bp (Genbank no. OR737627), and 1266 bp (Genbank no. OR737628), respectively (Fig.1A), which complete sequences have been submitted to the GenBank database with accession number PRJNA1026732. The genomic structure of BbSpV-1 is shown in Fig.1B. Potentially, the ssRNA1 and ssRNA2 each encodes a single protein, while the ssRNA3 encodes two hypothetical proteins. The length of 5' and 3' untranslated regions (5'- UTR and 3' - UTR) of the three +ssRNAs are all 79-97 nucleotides and 52-133 nucleotides (Figure 1B), with relatively conservation (Figure 1C). The 5'- UTR of three fragments is predicted to be a stable stem-loop structures (Fig. 1D).
Fig. 1 Genomic characterization of single stranded RNA segments extracted form Beauveria bassiana. (A) Purified +ssRNA was electrophoresed in a 1% agarose gel. Lane M, DNA marker. Lane 1 and 3, +ssRNA from strain GusB1, treated without DNase I (-). Lane 2 and 4, +ssRNA from strain GusB1, treated with DNase I (+). (B) Schematic representation of the BbSpV1 +ssRNA genome structures. (C) Comparison of the terminal sequences of +ssRNAs, the 5′ and 3' terminal conserved sequences of +ssRNAs, with black, gray, and light gray backgrounds indicating nucleotide similarity of not less than 100%, 75%, and 50%, respectively. (D) Predicted secondary structure of the BbSpV-1 5′-UTRs with the lowest energy.
The ORF (nt 79 to 2016) of +ssRNA1 (Fig. 1B), nominated ORF1, encodes a 645 aa of protein (73.07 kDa), with the highest BLASTx identity of 66.61% and 56.32% to the RNA dependent RNA polymerase (RdRp) of Plasmopara viticola lesion associated narnavirus 5 (PVaNarn5) (GenBank No. QIR30284.1, Query coverage =93%, E-value =0) and Diplodia seriata splipalmivirus 1(DsSpV1) (GenBank No. UOK20175.1, Query coverage =98%, E-value =0), respectively. The 5' and 3' UTRs of +ssRNA1 were 78 bp and 133 bp in length, respectively (Fig.1B). The virus was nominated Beauveria bassiana splipalmivirus 1 (BbSpV1) (Fig. 2A).
Fig. 2 (A) Phylogenetic analysis based on RdRp amino acids sequences of BbSpV1 and other members of the family Narnaviridae. Bootstrap values higher than 50% are shown. The maximum-likelihood (ML) tree was constructed in MEGA X with 1000 bootstrap replicates. BbSpV-1 is indicated by a red triangle. (B) Alignment of some polymycoviruses illustrate the conserved RdRp motifs I, V to VI.
The ORF (nt 97 to 2019) of +ssRNA2, nominated ORF2, encoded a 640-aa protein (73.66 kDa), with the highest BLASTX identity of 62.09% and 47.42% to the RdRp of Suillus luteus narnavirus 4 (SlNV4) (GenBank No. WLK77412.1, Query coverage = 97%, E-value=0) and Diplodia seriata splipalmivirus 1(DsSpV1) (GenBank No. WLK20176.1, Query coverage = 98%, E-value=0), respectively (Fig. 2A). The 5' and 3' UTRs of +ssRNA2 were 96 bp and 114 bp in length, respectively (Fig.1B).
The +ssRNA3 genome encodes two ORFs (Fig.1B), named ORF3 (nt 94-621) and ORF4 (nt 750-1214), which are separated by 129 nucleotides. The ORF3 encodes a 175 aa protein (19.91 kDa), with the highest amino acid sequence homology of 43.2% (GenBank No. UAW09580.1, Query coverage=95%, E-value=9e-29) to +ssRNA3 of Aspergillus flavus narnavirus 1 (AfNV1) from Aspergillus flavus. The ORF4 encodes a 154 aa protein (16.58 kDa), with the highest amino acid sequence homology of 63.7% and 50.94% to the hydrophilic protein of AfNV1 (GenBank No. BED 98308.1, Query coverage=94%, E-value=5e-60) and Cryphonectria naterciae fusagravirus 1 (CnFGV1) (GenBank No. BCX 55511.1, Query coverage=99%, E-value=3e-42), respectively. The 5' and 3' UTRs of +ssRNA3 were 93 bp and 52 bp in length, respectively (Fig.1B).
The only Narna-like virus BbSNLV has been found in B. bassiana, which harbors a single stranded RNA encoding a RdRp [14], while there has no report of Narnavirus with multiple segments in B. bassiana, although some Narna-like viruses with multiple segments have been identified from other fungus, for instance, four Narna-like viruses isolated from Sclerotinia sclerotiorum possessed two to four segments [20]. The Aspergillus tennesseensis narnavirus 1, belongs to the well-established genus Narnavirus, possesses of five segments, both the ssRNA1 and ssRNA2 of which encoded two different RdRp motifs, respectively [21]. Here, BbSpV1 was identified from B. bassiana with three RNA segments, notably, it has two RdRp motifs, which was the typical characteristic of splipalmiviruses. To estimate the phylogenetic relationships, the split RdRP sequences encoded by RNA1 and RNA2 of BbSpV1 and that of the other splipalmiviruses, as well as the representative members of the family Narnaviridae were used to construct phylogenetic trees. The results showed that among the viruses of two proposed splipalmiviral genera [7], BbSpV1 was placed in a clade with high branch support values of 85% (for +ssRNA1 RdRp) and 100% (for +ssRNA2 RdRp) with duasplipalmiviruses (Fig 2A). The conserved motifs (I, V to VI) with RdRps of other Narnavirus were found in the ORF1 of BbSpV1 (Fig. 2B). Therefore, taxonomically, the virus BbSpV1 belongs to a potential new family Splipalmiviridae.