Capilloviruses are members of the family Betaflexiviridae, which possess flexuous filaments particles 640-700 nm in length and a 6.5-7.4 kb, linear positive-sense ssRNA genome. The genomic RNA has two open reading frames (ORF) encoding a large replication-associated protein fused with the coat protein (CP) and a putative movement protein (MP), and is polyadenylated at its 3’ ends [1]. Currently, four virus species (Apple stem grooving virus, Cherry virus A, Currant virus A, and Mume virus A) are known to belong to the genus Capillovirus. They are known to occur mainly in woody plants, but a strain of ASGV has been reported in lily plants [2]. Moreover, no vectors have been clearly identified.
Gerbera plants (Gerbera spp.) are exceptionally important cut flowers in the current global floricultural trade. They are also widely grown in many countries as flowering potted or garden plants [3]. Several species of viruses were reported in gerbera plants worldwide including, Chrysanthemum stem necrosis virus (CSNV), Cucumber mosaic virus (CMV), Impatiens necrotic spot virus (INSV), Tobacco rattle virus (TRV), Tobacco ringspot virus (TRSV), and Tomato spotted wilt virus (TSWV) [4–9]. However, capilloviruses have not been identified in gerbera plants.
In August 2019, leaves (n=32) with viral symptoms such as chlorosis, ringspot, leaf distortion, yellowing and yellow spot were collected from G. jamsonii in seven greenhouses within Gyeongsangbuk-do, South Korea. To identify the causal agent of the virus, high-throughput RNA sequencing (HTS) was performed. An equal amount (10 mg) of the leaf was taken from the collected samples, pooled, and grounded using liquid nitrogen. The total RNA was extracted from the subsequently grounded powder using a Maxwell® 16 LEV Plant RNA Kit (Promega, Madison, USA). A cDNA library was constructed using a TruSeq Standard Total RNA with Ribo – Zero Plant (Illumina, San Diego, USA). Subsequently, paired-end sequencing was performed on 101 bp using the Illumina HiSeq 4000 platform (Macrogen, Seoul, Korea). The acquired reads were assembled de novo into transcript contigs using Trinity software as described previously [10]. The assembled contigs were subjected to BLASTx searches against the non-redundant NCBI protein database. These revealed a tentative novel capillovirus-like contig of 6908 nt that shares a 44.62% (query cover: 93%) amino acid (aa) sequence identity with the rubber tree virus 1 (RTV1) (GenBank no. QGR26011). To validate the presence of the novel capillovirus, total RNA was extracted from each sample (n = 32) using an easy-spin™ Total RNA Extraction Kit (iNtRON Bio, Seongnam, Korea). RT-PCR was then performed using a primer pair specifically designed against the acquired contig. Consequently, the novel capillovirus was confirmed in a sample showing yellow spots on leaf (Fig. 1). Then, cDNA was synthesized from the total RNA of the detected novel capillovirus using Oligo dT(18) as an initiator of the reaction, following which, PCR was performed with nine pairs of primer designed from the contig (Fig. 2, Supplementary Table 1). The obtained amplicon was cloned using an All in One™ PCR Cloning Kit (BioFact, Daejeon, Korea) and sequenced by Macrogen. To determine the complete genome sequence, the 5’ and 3’ ends were amplified using the 5’ and 3’ RACE System for Rapid Amplification of cDNA Ends (Invitrogen, Carlsbad CA, USA) with gene specific primers (Supplementary Table 1). The 5’ and 3’ RACE amplicons were cloned and sequenced in the same manner as other fragments. The resulting nucleotide sequences were assembled using the DNAMAN version 7.0.8.2 software (Lynnon Biosoft, Quebec, Canada). Additionally, the infected gerbera leaf tissue used sap inoculate gerbera plant which novel capillovirus was not infected and 15 species of indicator plants (Chenopodium amaranticolor, C. quinoa, Citrullus lanatus, Cucumis melo var. makuwa, Cu. sativus, Glycine max, Capsicum annuum, Datura stramonium, Nicotiana benthamiana, N. occidentalis, N. rustica, N. tabacum, Physalis angulata, Solanum lycopersicum). All plants were observed for 28 days after inoculation, but no symptoms were observed. Meanwhile, in RT-PCR assay, all indicator plants were negative, but only gerbera plant reacted positively. Therefore, it was confirmed that novel capillovirus was derived from the gerbera plant and that gerbera plant was included in the host range.
The complete genome of the virus consists of 6929 nucleotides (nt) of which 40.68% is G/C content (Fig. 2), while it is predicted, via the NCBI ORF Finder, to contain two ORFs like other known capilloviruses [1]. The 5’ and 3’ untranslated regions (UTR) are composed of 37 and 357 nt, respectively. The complete genome nucleotide shares a 68.54% identity (query: 22%) with the rubber tree capillovirus 1 (MN047299.1). The ORF1 (nt 38-6532) encodes a large replication-associated protein fused to the coat protein that is 2164 aa long. Additionally, it shares a 44.53% and 43.25% aa sequence identity (99% and 76% query cover) with a polyprotein in the rubber tree capillovirus 1 (QGR26011.1) and Currant virus A (YP_009229912.1), respectively. The ORF2 encodes a putative movement protein, 425 aa in length, which shares 40.00% and 41.85% aa sequence identity (97% and 52% query cover) with a movement protein of the rubber tree caillovirus 1 (QGR26012.1) and Mume virus A (QIM55854.1), respectively.
A simple modular architecture research tool (SMART) analysis revealed that the ORF1 encoded polyprotein contains four domains: methyltransferase (Mtr, Pfam01660; nt 161-1141), helicase (Hel, Pfam01443; nt 2669-3154), RNA-dependent RNA polymerase (RdRP, Pfam00978; nt 3527-4579, and coat protein (CP, Pfam05892; nt 5810-5815). In contrast, the ORF2 contains a single domain: movement protein (MP, Pfam01107; nt 4918-5475). The coat protein (CP) cistron of the genus Capilloviruses is located in the C-termiunal end of ORF1, and ORF2 (MP) is nested within ORF1 [1]. In this virus, the CP cistron was also located at the C-terminal end of ORF1 and it shares a 62.33% aa sequence identity (99% query cover) with a CP cistron in the rubber tree capillovirus 1 (QGR26011.1). In addition, the sequences, which appear as promoters (UUAGGU) and direct the subgenomic RNA (sgRNA) transcription of capilloviruses [11] were found at the 5' terminus (4841 nt and 5810 nt) of the MP and CP domains, respectively. The MP and CP of this virus are expected to be expressed through sgRNA transcription. Accordingly, the novel capillovirus was tentatively named “Gerbera virus A” (GeVA).
A phylogenetic tree was constructed using aa sequences of the polyprotein of GeVA along with 41 members of 10 genera in the family Betaflexiviridae (Supplementary Table 2). In the phylogenetic analysis, GeVA clearly placed with the capilloviruses. Also, pairwise comparisons of GeVA and 41 betaflexiviruses showed 31.4%-40.4% similarity in the nt of the complete genome, 37.9-48.2% for aa of the ORF1, and 5.2%-24.5% for aa of the movement protein (MP). The taxonomic position and sequence comparisons identified GeVA as a new member of the genus Capillovirus according to the ITCV species demarcation criteria [1].