Viruses belonging to the recently established genus Coguvirus also infect plants and, although this genus has not been included yet in the family Phenuiviridae, many structural and molecular features and phylogenetic relationships with other phenuiviruses support its classification in this family. With respect to nsRNA viruses, there is only one preliminary report of tomato spotted wilt virus in this host ; however, later studies failed to transmit TSWV to grapevine. During the screening of two selections of Garan dmak and Muscat rose grapevines via HTS, partial sequences with homology to apple rubbery wood viruses 1 and 2 were identified. ARWV1 and ARWV2 are two novel nsRNA viruses with tri-segmented genomes recently discovered in apple trees displaying rubbery wood symptoms in Canada. Based on sequence identity and phylogenetic analysis, ARWV1 and ARWV2 were proposed to be representative members of a new genus, tentatively named Rubodvirus, within the family Phenuiviridae. Here, we report the natural occurrence and discovery of two novel nsRNA viruses, blueberry container size named grapevine Muscat rose virus and grapevine Garan dmak virus , in grapevine plants. Even more, this is the first evidence for the natural occurrence of phenui-like viruses in grapevine.
Later, reverse transcription PCR -based assays for the specific detection of both viruses were designed to investigate the prevalence in different grapevine populations. Phylogenetics and transmission of GMRV and GGDV were also investigated.In 2015, a new selection of grapevine, cultivar Garan dmak, was received as dormant cuttings from Armenia under a USDA Animal and Plant Health Inspection Service controlled import permit for inclusion in the Foundation Plant Services collection. Cuttings were propagated under mist and later transferred to single pots; all this within an insect-proof greenhouse enclosure. Six months after bud break, leaf tissue from the Garan dmak plant was collected for HTS as part of the plant certification program at FPS. Additionally, a Muscat rose grapevine at the USDA National Clonal Germplasm Repository but originated from Argentina was sampled and analyzed by HTS during a separate study about the genetic diversity of grapevine leafroll-associated virus 3. At the time of sampling, both grapevine cultivars did not display any known symptom associated with virus infection. Total nucleic acid extracts were prepared using a MagMax Plant RNA Isolation kit as per manufacturer’s protocol, but adjusting the amount of plant material based on the following molecular analysis. For RT-PCR, 0.2 g of tissue was homogenized in 2 mL of guanidine isothiocyanate lysis buffer PVP-40 using a Homex grinder. In the case of HTS, 0.7 g of tissue was homogenized in 7 mL of guanidine isothiocyanate lysis buffer. Subsequently, the quality of TNA extracts was verified using an 18S ribosomal RNA assay.
HTS analysis was performed as described by Al Rwahnih et al.. Briefly, aliquots of TNA from source samples were subjected to rRNA depletion and complementary DNA library construction. Later, cDNA libraries were sequenced using the Illumina NextSeq 500 platform using asingle-end 75-bp regime. Illumina reads were demultiplexed and adapter trimmed prior to analysis using Illumina bcl2fastq v2.20.0.422. Trimmed reads were subsequently de novo assembled into contigs using SPAdes v3.13. Generated contigs were compared against the viral database of the National Center for Biotechnology Information using tBLASTx. Novel virus contigs were initially identified by their conserved protein domains. Open reading frames were annotated using HMMER v3.1 to look for Pfam protein domains associated with viruses infecting land plants. In the case of Phenuiviridae-like contigs, the large RNAs were annotated with the bunyavirus RNA-dependent RNA polymerase domain. The medium RNAs were annotated with the viral movement protein domain. The small RNAs were annotated with the Tenuivirus/Phlebovirus nucleocapsid protein domain. The association between the small and the medium RNAs was investigated by BLASTn sequence similarity of their 50 ends. These contig sequences were subsequently confirmed by BLASTx hits to the NCBI nucleotide database which produced top hits to different accessions of ARWV2 and ARWV1. To complete the 50 end of each RNA segment present in GMRV and GGDV, the SMARTer RACE 5 0 /3 0 Kit was employed following the instructions provided by the manufacturer. In the case of the 30 ends, the methodology described by Navarro et al. was used, which involved the addition of a poly tail to the RNA template.In the last few years, as a consequence of the increasing application of HTS, many novel nsRNA viruses have been identified, most of which are from invertebrates. Thus, the classification of these viruses has been recently reassessed, with plant-infecting viruses now classified in the order Bunyavirales, Serpentovirales and Mononegavirales .
Recently, the tentative genus Rubodvirus has been officially proposed to classify two novel nsRNA viruses from apple trees, with a suggested species demarcation criteria of <95% aa identity for the RdRp. In the present study, HTS allowed the identification of two novel plant-infecting viruses with segmented nsRNA genome, GMRV, and GGDV, which are also the first nsRNA viruses identified and transmitted in grapevine. Although these viruses infect the same host species, the aa sequence identity between the putative proteins encoded by their genomic RNAs is always below 75% , indicating that GMRV and GGDV are two different viruses. Their genomes encode proteins showing the highest sequence identity with ARWV1 and ARWV2. GGDV and GMRV also share other traits with these viruses, including the number of genomic components, limited to three nsRNAs, identical terminal nucleotides in the genomic RNAs, and the lack of any ORF coding for glycoproteins. Moreover, close phylogenetic relationships between these four viruses are supported by the ML phylogenetic trees reported here, in which, independently of the considered protein , they always clustered in the same clade, which is significantly separated from all the other nsRNA viruses included in the analyses. Altogether these data support the classification of GMRV and GGDV as two novel species in the tentative genus Rubodvirus. When the other bunyavirales are considered, the ML phylograms inferred from the RdRps or the NPs show a close phylogenetic relationship between rubodviruses and coguviruses, which together with the arthropod-infecting LLV, form a superclade nested at a basal node, in closer proximity to arthropod-infecting viruses than other plant-infecting viruses. These data are consistent with the hypothesis, previously advanced for the coguviruses, that all the members of this superclade evolved from a common ancestor virus infecting arthropods. In this evolutionary scenario, the acquisition of the MP gene appears to be the key step in the adaptation of the ancestor virus to plants, an event that likely happened through the typical modular genome evolution process proposed for most eukaryotic viruses. However, the clustering of rubodviruses and coguviruses in two distant clades, observed in the ML tree inferred with MPs , supports the independent acquisition of the MP gene by the ancestor of the viruses included in these two taxa. These data are consistent with the hypothesis that the adaptation of invertebrate-infecting nsRNA ancestor viruses to plants happened several times through independent events during the evolutionary history of nsRNA viruses infecting plants. It is worthy of note that rubodviruses, laulaviruses, and coguviruses, although phylogenetically related, have divergent genome structures and gene expression strategies. In fact, the members of the first two genera have a genome composed of three monocistronic nsRNAs encoding different proteins;RdRp, NP, and putative MP in the case of rubodviruses, and RdRp, NP and a protein of unknown function in the case of laulaviruses. Instead, coguviruses have a bipartite genome consisting of one nsRNA encoding the RdRp , and one ambisense RNA , in which the ORFs encoding the NP and the MP are separated by a long intergenic region .
It has been shown that such an IR is AU-rich and self-complementary, thus assuming in both polarity strands a compact conformation containing a long hairpin predicted to serve as a transcription termination signal during the expression of viral genes. Taking this into consideration, growing raspberries in container the question arises of how viruses with such different genomic organizations may have evolved from the same ancestor virus. In this respect, it can be speculated that a recombination event between the viral and the vc strand of two genomic RNAs with long, AU rich, and almost identical 50 UTRs could generate ambisense RNAs containing an IR similar to those of coguviruses. Since nsRNA viruses with genomic RNAs showing structural features compatible with this possibility were not known previously, such a possibility appeared unlikely. However, the very long, AU-rich, and highly conserved 50 UTRs reported here for RNA 2 and RNA 3 of GMRV and GGDV, and also observed in the corresponding RNAs of ARWV1 and ARWV2, are the first clear evidence that nsRNA viruses with the structural features compatible with this evolutionary scenario may exist. Based on these considerations, the possibility that the bipartite genomes of coguviruses originated from a tripartite ancestor with genomic RNAs containing 50 UTRs similar to those observed in the rubodviruses appears feasible. No glycoprotein is encoded by rubodviruses , a feature previously reported also for coguviruses that, according to electron microscopy observations, are flexuous, non-enveloped viruses. In contrast, glycoproteins are expressed by most nsRNA plant viruses transmitted by arthropods . The lack of glycoprotein in the genome of rubodviruses and coguviruses opens the question on the existence of vectors, if any, involved in their transmission. In this respect, it is worthy of note that ophioviruses and varicosaviruses, which are plant-restricted or transmitted by fungi, also do not code for any glycoprotein. Vegetative propagation has been proposed as the prevalent transmission mechanism for ARWV1 and ARWV2. Whether, this is also the case for GMRV and GGDV needs further investigation. Most plant viruses code for viral suppressor proteins counteracting the plant antiviral defense mechanisms based on RNA silencing. Further studies are needed to ascertain whether one or more of the three proteins encoded by GGDV and GMRV and other rubodviruses may interfere with RNA silencing, thus showing multifunctional role, as already reported for VSRs of other viruses. GMRV and GGDV were identified in two different grapevines cultivars, Muscat rose and Garan dmak, that were tested by HTS; moreover, both viruses were found in association with other viruses and viroids. Interestingly, no obvious symptoms were observed in the two grapevines. Although infectivity of both viruses was ascertained by graft-transmission, only a Cabernet franc grapevine infected by GGDV developed symptoms post-grafting; further HTS analysis on this indicator plant reveals the presence of GRSPaV, GYSVd-1, and HSVd. Therefore, it was not possible to ascertain whether GGDV is associated with symptoms and additional studies on its pathogenicity are needed, likewise GMRV. An initial survey using several accessions of grapevine located in three different collections in California resulted in the identification of two and one plants infected by GGDV and GMRV, respectively; like the original sources of the viruses , these grapevines were symptomless. A more extensive survey, including other grapevine-growing regions in California and the USA, is necessary to determine the real distribution of these novel viruses. In that sense, the detection method developed in this study could be useful for virus testing and certification programs. Finally, during the review process of this manuscript, two novel mycoviruses related to coguviruses and rubodviruses were reported, which extends the host range of phenui-like viruses.Molecular networking1 , introduced in 2012, was one of the first data organization approaches to visualize the relationships between tandem mass spectrometry fragmentation spectra. In molecular networking, relationships between similar MS/MS spectra are visualized as edges. As MS/MS spectral similarity implies chemical structural similarity , chemical structural information can thus be represented as a network and chemical relationships can be visualized. This approach forms the basis for the web-based mass spectrometry infrastructure, Global Natural Products Social Molecular Networking which sees ~200,000 new accessions per month. Molecular networking has successfully been used for a range of applications in drug discovery, natural products research, environmental monitoring, medicine, and agriculture. To tap into the chemistry of complex samples through metabolomics, a subset of MS/MS spectra can be annotated by spectral library matching or by using in silico approaches. While molecular networking facilitates the visualization of closely related molecules in molecular families, the inference of chemical relationships at a dataset-wide level and in the context of diverse sample metadata requires complementary representation strategies. To address this need, we developed an approach that uses fragmentation trees and machine learning to calculate all pairwise chemical relationships.