A novel narnavirus isolated from the wheat stripe rust fungus Puccinia striiformis f. sp. tritici

P. striiformis strain SCSN-12 was isolated from a urediniospore pustule infecting wheat in Sichuan province, China, in 2015. This P. striiformis strain was propagated on the susceptible wheat variety Mingxian 169 as described previously [8], and the urediospores were harvested and stored at 4 °C until further use [12]. dsRNA was obtained from 0.5 g of urediniospores by selective absorption to a cellulose powder CF-11 column containing 16% ethanol [8]. The extracted dsRNAs were further treated with DNase I and S1 nuclease (Takara) to remove genomic DNA and ssRNA as described previously (Fig. 1A) [7]. A cDNA library of the purified dsRNA sample was constructed using an NEBNext^®Ultra™ RNA Library Prep Kit (Illumina) according to the manufacturer’s protocol. To obtain the terminal sequences of the dsRNA, the 5’ and 3’ termini were amplified by rapid amplification of cDNA ends (RACE kit, Invitrogen) [13]. The amplified PCR product was cloned and sequenced by the Sanger method. The sequence was assembled using SeqMan from the Lasergene sequence analysis package (DNASTAR). The complete nucleotide sequence of PsNV1 has been submitted to the GenBank database under the accession number MN756800. The putative ORF was identified using the ORF Finder program (http://​www.​ncbi.​nlm.​gov/​ gorf/gorf.html). A protein domain search was performed using the Conserved Domain Database (CDD) (http://​www.​ncbi.​nlm.​nih.​gov/​Structure/​cdd/​wrpsb.​cgi). Multiple sequence alignment of the protein sequences were performed with ClustalW 2.0 [14] and DNAMAN 7.0 software (Lynnon Biosoft, USA). A phylogenetic tree was generated by the maximum-likelihood (ML) method in MEGA 7.0 software with 1000 bootstrap replicates [15].

The full-length cDNA sequence of the PsNV1 genome is 2,340 nt in length with a GC content of 50.04%. The 5′ and 3′ untranslated regions (UTRs) of PSNV1 are 34 and 80 nt in length, respectively (Fig. 1B). Using the Unafold program (http://​unafold.​rna.​albany.​edu), two potential stem-loop structures were predicted, and the initial ΔG values of 5′ and 3′ UTRs were -20.30 kcal/mol and -40.20 kcal/mol, respectively (Fig. 1C). Using the standard genetic code, the PsNV1 genome contained a single ORF encoding a 741-amino-acid (aa) protein with a molecular mass of 81.8 kDa (Fig. 1B).

A sequence search using the BLASTp tool showed that this 81.8-kDa protein had significant sequence similarity to the RNA-dependent RNA polymerases (RdRps) of narnaviruses. Fusarium poae narnavirus 1 (FpNV1) [16] was the best match (29.6% identity), followed by Saccharomyces 23S RNA narnavirus (26.96% identity) and Saccharomyces 20S RNA narnavirus (24.9% identity) [9]. Furthermore, a CDD search and multiple protein sequence alignments indicated that the RdRp domain includes six conserved motifs (Fig. 2A) that are characteristic of members of the genus Narnavirus. These results suggest that P. striiformis strain SCSN-12 was infected by a novel narnavirus, PsNV1.

To investigate the relationship between PsNV1 and other mycoviruses, a molecular phylogenetic tree was constructed based on the RdRp amino acid sequences of PsNV1, other narnaviruses, mitoviruses, and plant ourmiaviruses. The phylogenetic tree revealed that PsNV1 grouped together with FpNV1 as a sister branch of narnaviruses, forming a distinct clade. Therefore, the phylogenetic tree clearly placed PsNV1 in a separate cluster containing FpNV1 and unclassified narnaviruses (Fig. 2B). To our knowledge, this is the first report of the full-length nucleotide sequence of a novel narnavirus in the obligate parasitic fungus P. striiformis.