Melothria scabra, commonly known as mouse melon or Mexican sour gherkin, is an annual vine plant belonging to the family Cucurbitaceae (Chomicki et al., 2020; Schaefer and Renner, 2011). It is native to Mexico and Central America but is now cultivated in various regions around the world, including but not limited to Africa, Asia, Europe, and North America (Chomicki et al., 2020; Roberts et al., 2018). M. scabra is popular for its unique appearance and flavor, and is used in a variety of culinary applications such as fresh food, salads, cocktails, and marinades. The phytochemical and biological components of M. scabra have been shown to possess various medicinal and pharmacological properties, including antioxidant, anti-diabetic, anti-hypoglycemic, and anti-cancer activities (Kamaruddin et al., 2021; Roberts et al., 2018). Additionally, it has a high nitrogen-to-carbon ratio, which is beneficial for improving muscle performance and aiding in recovery (Roberts et al., 2018). M. scabra also exhibits biopesticide effects against mosquitoes (Ahmed et al., 2019).
Since 2011, M. scabra has been successfully introduced to several regions in China, including Hangzhou, Guangzhou, Chengdu, Xi’an, Shandong, and Pingdingshan (Tao et al., 2013). However, large-scale cultivation and promotion have not yet been achieved. In China, M. scabra is often found scattered along roadsides and in fields, where it is commonly regarded as a weed. This ecological role raises the possibility of M. scabra serving as a reservoir or intermediate host for crop viruses, although no associated plant viruses have been documented thus far. Therefore, identifying potential viral diseases in M. scabra is of significant importance.
Potyviruses (genus Potyvirus, family Potyviridae) represent the largest group of plant-infecting RNA viruses (Gibbs et al., 2020; Revers and Garcia, 2015; Yang et al., 2021). Many potyviruses are agriculturally significant, causing devastating epidemics and substantial yield losses in crops worldwide (Inoue-Nagata et al., 2022; Rybicki, 2015; Scholthof et al., 2011; Yang et al., 2021). This genus infects a wide range of crop species, and several potyviruses are listed among the top ten most economically and scientifically important plant viruses (Moury and Desbiez, 2020; Rybicki, 2015; Scholthof et al., 2011). Two particularly destructive potyviruses affecting cucurbit crops are zucchini tigre mosaic virus (ZTMV) and zucchini yellow mosaic virus (ZYMV) (Lecoq and Desbiez, 2012; Romay et al., 2014). ZTMV primarily infects cucurbit crops such as bitter melon, zucchini, pumpkin, chieh-qua, wax gourd, ridge gourd, sponge gourd, and snake gourd (Peng et al., 2021; Romay et al., 2014; Zhou et al., 2022). It was first reported in Guadeloupe in 1982 (Romay et al., 2014) and has since been detected in Asia, Europe, the Americas, and islands in the Caribbean and Indian Ocean (Abdalla and Ali, 2018; Peng et al., 2021; Zhou et al., 2022, 2024). ZTMV was first reported in China in 2014 (Zhou et al., 2024). Currently, only 12 full-length ZTMV genome sequences are available in GenBank database: seven from China, three from France, and two from the Americas (Peng et al., 2021; Romay et al., 2014; Zhou et al., 2024). ZYMV poses a severe threat to crops such as cucumber, pumpkin, squash, melon, zucchini, watermelon, and rockmelon, particularly in tropical and subtropical regions (Ali et al., 2024). Infected plants exhibit a range of symptoms, including yellowing, blistering, mosaic patterns, necrosis, stunting, vein clearing, and leaf deformation (Ali et al., 2024; Coutts et al., 2011; Lecoq and Desbiez, 2012; Metwally et al., 2024; Yang et al., 2021). Experimentally, ZYMV has been shown to infect other species such as Bryonia dioica, Cyclanthera pedata, Ecballium elaterium, Momordica balsamina, M. rostrata, and Zehneria scabra through mechanical inoculation (Csorba et al., 2004). ZYMV is transmitted non-persistently by over 26 aphid species (Katis et al., 2006; Khanal and Ali, 2019), through seeds in some hosts (Simmons et al., 2011), and mechanically via contact (Simmons et al., 2013).
The genomes of ZTMV and ZYMV are both positive-sense single-stranded (+ss) RNA, approximately 10.3 kb and 9.6 kb in length, respectively, with a 3′ terminal poly(A) tail and a VPg protein at the 5′ end (Ali et al., 2024; Metwally et al., 2024; Zhou et al., 2024). Translation of the single major open reading frame (ORF) and a ribosomal frameshifting product, PIPO, results in the synthesis of a long polyprotein (Inoue-Nagata et al., 2022). This polyprotein is subsequently cleaved by viral proteases into 10 mature proteins: P1 (protein 1), HC-Pro (helper component-protease), P3 (protein 3), 6K1 (6-kDa peptide 1), CI (cylindrical inclusion), 6K2 (6-kDa peptide 2), VPg (viral protein linked to the genome), NIa-Pro (nuclear inclusion A-protease), NIb (nuclear inclusion B), and CP (coat protein) (Maghamnia et al., 2018; Romay et al., 2014).
In recent years, integrated diagnostic technologies, such as high-throughput sequencing (HTS), have opened up vast opportunities for advancing plant virus disease management strategies (Reuter et al., 2015; Villamor et al., 2019). HTS methods, including RNA sequencing (RNA-Seq) and small RNA sequencing (sRNA-Seq), have become widely used to uncover novel viruses and explore virus diversity (Maina et al., 2023; Massart et al., 2019).
In the present study, we first studied the presence of potential viruses in M. scabra leaves exhibiting symptoms indicative of viral infection through metatranscriptome and sRNA-Seq. High-throughput data analysis of the symptomatic sample revealed two virus-related contigs, showing 98.26% nucleotide similarity to ZTMV (100% query coverage) and 80.15% nucleotide similarity to ZYMV (93% query coverage), respectively. The full-length genome sequences of ZTMV and ZYMV isolates in M. scabra, designated ZTMV-ms and ZYMV-ms, were obtained through a combination of reverse transcription PCR (RT-PCR), rapid amplification of cDNA ends (RACE) and Sanger sequencing. Furthermore, the small RNA profiles of ZTMV-ms and ZYMV-ms were also characterized. Our study represents the first documented instance of ZTMV and ZYMV infecting M. scabra and provides pioneering insights into the plant virus pathogens of this species.
Materials and Methods
Sample collection and total RNA extraction
In July 2023, about five leaves exhibiting yellowing and mosaic-like symptoms were collected from a single M. scabra plant in a vegetable plot in Sanming City, Fujian Province, China. To ensure high-quality RNA, only fresh leaves were selected. The leaf samples were immediately frozen in liquid nitrogen and stored at −80°C to preserve RNA quality for subsequent follow-up research. Total RNA was extracted using TRIzol Reagent (Accurate Biology, Guangzhou, China) following the manufacturer’s instructions. After assessing the RNA’s quality and quantity, the samples were sent to Novogene (Beijing, China) for both metatranscriptomic (RNA-Seq) and sRNA-Seq.
RNA-Seq and small RNA library preparation and sequencing
The sequencing library preparation and sequencing were performed by Novogene in Beijing, China. In short, as for RNA-Seq, total RNA extracted from leaves showing suspected viral symptoms was subjected to rRNA depletion for library preparation. The resulting RNA library was then sequenced on the Illumina Novaseq 6000 platform in a paired-end mode with a read length of 150 base pairs (bp) on both ends (150 bp × 2). For sRNA-Seq, the library was prepared with the TruSeq Small RNA Library Prep Kit (Illumina, San Diego, CA, USA) and sequenced on the Illumina HiSeq 4000 platform using the SE50 protocol to obtain single-end sRNA reads.
De novo assembly of RNA-Seq dataset and virus identification
The raw reads were processed using fastp to remove adapter sequences, low-quality reads, and those shorter than 36 nucleotides (Chen et al., 2018). The remaining clean reads were subjected to de novo assembly using the Trinity assembler (version 2.8.5) (Haas et al., 2013). Following assembly, the assembled contigs were then subjected for sequence identification and annotation via BLASTn and BLASTx against the National Center for Biotechnology Information (NCBI) NT and NR databases (https://www.ncbi.nlm.nih.gov) (Altschul et al., 1990). Additionally, a prebuilt RdRP Hidden Markov Model database was searched using the hmmsearch tool within HMMER (http://hmmer.org/) to identify potential highly divergent viruses. ORFs were analyzed using online server tool ORFfinder (www.ncbi.nlm.nih.gov/projects/gorf/). Conserved domains were identified using NCBI’s Conserved Domain Search tool (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi).
Determination of the full-length genome of the virus
To validate the presence of the putative virus contigs obtained through assembly, primer sets were first designed for RT-PCR detection (M-MLV Reverse Transcriptase, Promega, Madison, WI, USA; 2×Taq Master Mix, Vazyme, Nanjing, China) using RNA extracted from a single plant (Supplementary Table 1). To obtain the complete genomes of ZTMV-ms and ZYMV-ms, primers flanking the terminal regions of the assembled contigs were designed to determine the 5′ ends using a 5′ RACE kit (cat. No. 18374-058, Invitrogen) and the 3′ ends by RT-PCR (Supplementary Table 1). The entire genomes were then amplified using specific overlapping primers (Supplementary Table 1). Viral cDNAs were synthesized from total RNA in a 10 μL reaction, followed by PCR amplification in a 50 μL reaction volume using a Bio-Rad S1000 thermal cycler (Bio-Rad Laboratories, Life Science Group, Hercules, CA, USA). PCR products were purified using the Multifunction DNA Purification Kit (cat. No. DH103, Guangdong Ardent Biomed Co., Ltd., Guangdong, China), ligated into the pMD 19-T Vector (cat. No. 6013, Takara, Otsu, Japan), and sequenced to obtain at least three positive clones. The full-length genome sequences of ZTMV-ms and ZVMV-ms were assembled and deposited in the NCBI GenBank database under accession numbers PQ720520 and PQ720521, respectively.
Pairwise alignment, phylogenetic analysis
Pairwise sequence identity of the viral genomes and encoded proteins was calculated using fast alignment with default parameters in DNAMAN version 10.0 software (Lynnon Biosoft, Montreal, Canada). To explore the evolutionary relationships among different isolates of ZTMV-ms and ZYMV-ms from diverse hosts, the related sequences of other ZYMV or ZTMV isolates from different hosts were retrieved from the NCBI GenBank database. These sequences (polyprotein coding sequence) were then aligned using MAFFT (Katoh and Standley, 2013). A phylogenetic tree was constructed based on the maximum likelihood method using IQ-TREE (v1.6.6) (Nguyen et al., 2015). The nucleotide substitution models TIM2 + F + R2 for ZTMV-ms and GTR + F + R3 for ZYMV-ms, selected by ModelFinder (Kalyaanamoorthy et al., 2017), were applied for phylogenetic analysis. The reliability of the tree topology was assessed using 1,000 bootstrap replicates.
Characterization of virus-derived small interfering RNAs
Raw sRNA data were first assessed using FastQC, then processed with the Cutadapt tool (v1.16) (Martin, 2011) to trim adapters and remove low-quality reads. Only clean reads ranging from 18 to 30 nucleotides without ambiguous nucleotides (“N”) were retained and further clustered into unique reads using an in-house Python script. The clean sRNA reads were then aligned to the ZYMV-ms or ZTMV-ms virus reference genomes (accession nos. PQ720521 and PQ720520) using Bowtie (Langmead et al., 2009), allowing a maximum of one mismatch. The characterization of virus-derived small interfering RNAs (vsiRNAs) included analyzing their size distribution, nucleotide frequency at the 5′ end, genome distribution, and polarity through mapping of sRNAs to the corresponding viral genomes.
Results
RNA-Seq, small RNA sequencing, and virus identification
To uncover potential viral pathogens associated with the yellowing and mosaic symptoms in M. scabra leaf samples (Fig. 1A and B), total RNA was extracted from pooled symptomatic leaves (2-3 leaves) for both RNA-Seq and sRNA-Seq sequencing. The RNA-Seq dataset generated approximately 45,923,644 raw reads. After quality trimming, approximately 45,110,762 clean reads were retained for sequence assembly. For the sRNA-Seq, 11,206,344 clean reads were retained from a total of 18,416,170 raw reads after rigorous quality control. The clean RNA-Seq reads were assembled into larger contigs and subsequently searched against the NCBI NR/NT database using BLASTx/BLASTn to identify putative viral contigs. As a result, two long contigs, measuring 10,465 nt and 9,609 nt, were annotated as ZTMV and ZYMV, respectively. Specifically, The ZTMV-related viral contig showed 98.8% similarity to its best match QGV13524.1 (polyprotein of ZTMV-CX1) in a BLASTx search, and 98.26% similarity to MK988416 (genome of ZTMV-CX1) in a BLASTn search. In contrast, the sequence divergence between the ZYMV-related viral contig and its best match was notably higher. The ZYMV-related viral contig showed only 87.1% similarity to its best hit CAD56800.1 (polyprotein of ZYMV-WM) in BLASTx search, and 80.15% similarity to its best hit AJ515911.1 (genome of ZYMV-WM) in BLASTn search. The species demarcation criteria for the family Potyviridae by the International Committee on Taxonomy typically involves complete ORF sequences with <76% (nt) and <82% (aa) identity. Thresholds for other protein cleaved are usually 58% (P1) or 74-78% (Inoue-Nagata et al., 2022). Based on this standard, the ZTMV-related viral contig is clearly classified as a taxonomic unit at the ZTMV virus species level. Although the sequence divergence between the ZYMV-related viral contig and its best match is near the threshold for defining a new species, it still falls within the current criteria for species demarcation in the genus Potyvirus of the family Potyviridae. Therefore, the ZYMV-related viral contig identified in M. scabra is still considered an isolate of the known ZYMV virus. Accordingly, the ZTMV and ZYMV virus isolates identified in M. scabra were named ZTMV-ms and ZYMV-ms, respectively. These two viruses identified in M. scabra represent the first documented virus infection in this plant species.
Virus detection and the recovery of the full-length genome of ZTMV-ms and ZYMV-ms
To test for the presence of ZTMV-ms and ZYMV-ms, two specific primer pairs designed from the newly assembled genomes were synthesized for RT-PCR amplification of ZTMV-ms and ZYMV-ms. Positive genomic signals for ZTMV-ms and ZYMV-ms were detected in M. scabra samples (Supplementary Fig. 1). Furthermore, amplification of both the 5′ and 3′ termini of ZTMV-ms and ZYMV-ms showed that specific bands were generated with the corresponding primers (Fig. 2A and B, Supplementary Table 1). Once the sequences of the 5′ and 3′ ends were established, six or seven overlapping primer pairs (as listed in Supplementary Table 1), each designed to amplify fragments of approximately 1,300-1,400 bp and covering the entire genome, were used to amplify genome fragments of ZTMV-ms and ZYMV-ms by RT-PCR. The amplification results confirmed that all primer sets produced specific bands of the expected lengths (Fig. 2A and B).
The genome organization analysis of ZTMV-ms and ZYMV-ms
Based on the Sanger sequencing of the amplification product from the overlapping RT-PCR and RACE, the full genomes of ZTMV-ms and ZYMV-ms were successfully assembled. The full genome of ZTMV-ms, is 10,331 nt long, excluding the 3′ poly(A) tail. It features a 85-nt-long 5′ untranslated region (UTR) and a 187-nt-long 3′ UTR (GenBank accession no. PQ720520) (Fig. 3A). The genome of ZTMV-ms encodes a polyprotein of 3,352 aa, which is cleaved into 10 mature proteins at nine putative sites located at amino acid positions 547, 1,004, 1,350, 1,402, 2,037, 2,094, 2,283, 2,521, and 3,058 (Fig. 3B). The mature proteins include: P1 (547 aa, 62.81 kDa), HC-Pro (457 aa, 51.80 kDa), P3 (346 aa, 40.36 kDa), 6K1 (52 aa, 6.00 kDa), CI (635 aa, 71.49 kDa), 6K2 (57 aa, 6.41 kDa), VPg (189 aa, 21.34 kDa), NIa-Pro (238 aa, 26.76 kDa), NIb (537 aa, 61.59 kDa), and CP (294 aa, 34.01 kDa) (Fig. 3A). Furthermore, the putative P3N-PIPO (72 aa, 8.38 kDa, nt 3,554-3,771) was also identified (Fig. 3A).
Meanwhile, the full genome of ZYMV-ms, excluding the 3′ poly(A) tail, is 9,602 nt long with a GC content of 48.8%. It encodes a large polyprotein of 3,082 aa, comprising 10 genes located between 142 nt and 211 nt in the 5′ and 3′ untranslated regions, respectively (GenBank accession no. PQ720521) (Fig. 3C). Consistent with other potyviruses, nine highly conserved proteolytic cleavage sites at positions 312, 768, 1,114, 1,166, 1,800, 1,853, 2,043, 2,286, and 2,803 (Fig. 3B) were identified within the polyprotein. These sites yield 10 mature proteins: P1 (312 aa, 35.92 kDa), HC-Pro (456 aa, 52.07 kDa), P3 (346 aa, 40.24 kDa), 6K1 (52 aa, 5.82 kDa), CI (634 aa, 70.94 kDa), 6K2 (53 aa, 6.17 kDa), VPg (190 aa, 21.87 kDa), NIa-Pro (243 aa, 27.01 kDa), NIb (517 aa, 59.69 kDa), and CP (279 aa, 31.42 kDa) (Fig. 3C). Additionally, the putative P3N-PIPO, encoding an 84-aa protein (9.88 kDa), was identified between nt 2,900 and 3,153 within a highly conserved GAAAAAA sequence (Fig. 3C).
To assess potential differences in conserved protein cleavage sites among ZTMV-ms, ZYMV-ms, their closest relatives, and other potyvirus (represented by Bean common mosaic virus gardenia isolate, BCMV-nszz), we predicted and compared the cleavage sites in the polyproteins of ZTMV-ms, ZTMV-CX1, ZYMV-ms, ZYMV-WM, and BCMV-nszz. Our analysis revealed that, despite notable nucleotide differences between ZYMV-ms and ZYMV-WM, the two ZYMV isolates showed complete consistency in their conserved protein cleavage sites. Similarly, ZTMV-ms and ZTMV-CX1 exhibited identical cleavage sites. However, while conserved cleavage sites were largely consistent across virus species (ZTMV, ZYMV, BCMV), partial inconsistencies were observed at specific sites, including 6K1/CI, CI/6K2, 6K2/VPg, and VPg/NIa-Pro.
We also performed a detailed analysis of the nucleotide and amino acid identities between ZTMV-ms and ZTMV-CX1 (MK988416), as well as ZYMV-ms and another ZYMV-WM isolate (AJ515911). For ZTMV-ms and ZTMV-CX1, the amino acid sequence identities of their 10 mature proteins ranged from 95.61% to 100%, while the nucleotide sequence identities ranged from 97.01% to 99% (Supplementary Table 2). In contrast, the sequence divergence between ZYMV-ms and ZYMV-WM was much greater. The nucleotide regions encoding P1, HC-Pro, and 6K1 exhibited relatively low similarities (63.57%, 75.66%, and 78.21%, respectively). The nucleotide identities of the other ORFs ranged from 78.55% to 79.42%, with CP showing the highest similarity (95.10%). At the amino acid level, P1 and PIPO had similarities below 70% (59.29% and 69.05%, respectively), while the other ORFs ranged from 85.26% to 97.13% (Supplementary Table 3). Overall, ZTMV-ms and ZYMV-ms exhibited relatively high divergence at both the nucleotide and amino acid levels.
Additionally, when aligning the clean RNA-Seq reads to the ZTMV-ms and ZYMV-ms viral genomes, 1,548,772 reads (3.43% of total reads, with an average sequencing depth of 21,963×) were mapped to ZTMV-ms, and 11,515,975 reads (25.53% of total reads, with an average sequencing depth of 175,233×) were mapped to ZYMV-ms. The complete viral genomes of ZTMV-ms and ZYMV-ms could be covered by RNA-Seq reads with high abundance on average (Fig. 3A and C). The exceptionally high average sequencing depth indicates a very high viral load in the sample.
Phylogenetic analysis of ZYMV-ms and ZTMV-ms isolates
Although ZTMV-ms and ZYMV-ms are isolates of the ZTMV and ZYMV species, respectively, the evolutionary relationship between these isolates and others within the same species remains a question for further investigation. To investigate the evolutionary relationship of ZTMV-ms and other ZTMV isolates identified from different plant hosts, a midpoint-rooted phylogenetic tree of ZTMV-ms along with other ZTMV isolates, was constructed by a maximum likelihood algorithm using IQ-TREE program. The multiple sequence alignment of polyprotein coding sequence was carried out by MAFFT. The final dataset used to construct evolutionary tree comprised 14 polyprotein coding sequences for ZTMV. All ZTMV isolates in the phylogenetic tree were derived from Cucurbitaceae plants. The closest relative to ZTMV-ms is MK988416, an isolate from melon collected in Zhengzhou, China (Fig. 4). The method used to construct the phylogenetic tree for ZYMV-ms was similar to that employed for other ZYMV isolates. A total of 139 sequences with complete polyprotein coding sequence were included in the analysis. While most sequences were from Cucurbitaceae plants, some exceptions were noted, such as KX421104.1 (isolated from Sesamum indicum) and MG967620.1 (isolated from soybean). The phylogenetic tree revealed that ZYMV-ms does not cluster with other ZYMV isolates, and there is no significant bootstrap support for a close evolutionary relationship between ZYMV-ms and its neighboring clades (Fig. 5). This indicates a significant evolutionary distance between ZYMV-ms and other known ZYMV isolates. Furthermore, the phylogenetic tree reveals that, in general, isolates originating from the same host tend to exhibit close genetic relationships. This finding aligns with the substantial sequence divergence observed between ZYMV-ms and other ZYMV isolates (Fig. 5).
Characterization of ZYMV-ms and ZTMV-ms isolates derived vsiRNAs
The RNA silencing pathway plays a critical role in antiviral defense by generating 21-24 nt vsiRNAs, which mediate antiviral RNA interference (RNAi). As a result, profiling the vsiRNAs provides valuable insights into the virus-host interaction. To characterize the vsiRNAs in M. scabra, we constructed an sRNA library using the same sample previously employed for RNA-seq. This library contained 11,206,344 clean reads, of which 793,650 (~7.08%) and 2,437,842 (~21.75%) reads were mapped to the ZTMV-ms and ZYMV-ms genomes, respectively. In both cases, the majority of vsiRNAs originated from both polarities of the ZTMV-ms and ZYMV-ms genomes, with predominant peaks at 21-nt and 22-nt, the latter being the most frequent (Fig. 6A and B). Polarity analysis revealed that the vsiRNAs were almost equally derived from the positive-sense (55% and 54%) and antisense (45% and 46%) genomic sequences in M. scabra (Fig. 6C), with a slight preference for the sense strand. It has been shown that the 5′-terminal nucleotides partially determine the preference of argonaute (AGO) proteins for sRNAs in Arabidopsis thaliana (Mi et al., 2008). Accordingly, we analyzed the distribution of 5′-terminal nucleotides among the sequenced ZTMV-ms and ZYMV-ms-derived siRNAs. In both cases, uridine (U) was the most abundant nucleotide in ZYMV-ms or ZTMV-ms vsiRNAs, while guanine (G) represented the least abundant nucleotide (Fig. 6D). Regarding the genomic regions from which vsiRNAs were generated, our results showed that vsiRNAs were distributed across the entire viral genome, although specific regions were identified as hotspots for vsiRNA production. Notably, the dominant 22-nt vsiRNAs were evenly spread across the genome in both ZTMV-ms and ZYMV-ms (Fig. 7A and B).
Discussion
ZTMV and ZYMV, both belonging to the genus Potyvirus in the family Potyviridae, are among the most economically important viruses impacting cucurbit crops globally (Ali et al., 2024; Zhou et al., 2024). ZTMV was first reported in Guadeloupe in 1982 (Romay et al., 2014) and has since been found across three continents (Asia, Europe, and America), as well as on several Caribbean and Indian Ocean islands (Abdalla and Ali, 2018; Peng et al., 2021; Romay et al., 2014; Zhou et al., 2022, 2024). In China, ZTMV was first documented in g2014 and has rapidly emerged as a prevalent pathogen, significantly impacting cucurbit production (Zhou et al., 2024). This virus affects various key cucurbit crops, including pumpkin, zucchini, and melon. Currently, only 13 full-length ZTMV genome sequences are available in the GenBank database. ZYMV, on the other hand, was first identified in the 1970s, simultaneously in Italy from zucchini (Lisa et al., 1981) and in France from muskmelon (Lecoq et al., 1981). Since then, it has spread to Asia, Africa, the Americas, and beyond. Its primary hosts are members of the Cucurbitaceae family, including cucumber, pumpkin, and melon (Ali et al., 2024; Metwally et al., 2024). Infected plants often exhibit severe symptoms and substantial yield loss (Ali et al., 2024; Coutts et al., 2011; Lecoq and Desbiez, 2012; Metwally et al., 2024; Yang et al., 2021). ZYMV is transmitted through various means, including aphids (Gal-On, 2007; Katis et al., 2006; Khanal and Ali, 2019), mechanical inoculation (Simmons et al., 2013), and seed transmission (Simmons et al., 2011). The ongoing advancement of HTS has led to the identification of numerous ZYMV isolates, with over 100 ZYMV sequences now available in the GenBank database.
In this study, both ZTMV-ms and ZYMV-ms were simultaneously identified in different leaves collected from the same host plant, M. scabra, marking the first documented viral infection in this species. The proportion of ZTMV-ms and ZYMV-ms was relatively high in both transcriptome RNA-Seq and sRNA-Seq datasets, resulting in a correspondingly high average sequencing depth. Analysis of the small RNA dataset further revealed that both ZTMV-ms and ZYMV-ms can trigger the host’s RNAi-mediated antiviral response. The complete genomes of ZTMV-ms and ZYMV-ms identified in M. scabra were determined using a combination of RT-PCR, RACE, and Sanger sequencing technologies. The experimentally validated full-genome sequences of ZTMV-ms and ZYMV-ms have been deposited in the NCBI GenBank database under accession numbers PQ720520 and PQ720521, respectively. In short, our study not only documents the first plant virus infection in M. scabra, but also expands the known host range of both ZTMV and ZYMV.
The polyprotein sequences of ZTMV-ms and ZYMV-ms were used for phylogenetic analysis. The phylogenetic analysis of ZTMV-ms aligns well with the results of BLASTx/BLASTn searches in the NCBI NR/NT databases. ZTMV-ms is most closely related to ZTMV-CX1, which also presents the best match in the BLASTx/BLASTn search results. However, for ZYMV-ms, the evolutionary analysis revealed a substantial genetic distance between ZYMV-ms and all currently known ZYMV isolates, preventing its assignment to any existing ZYMV clade. Our study provides a very sequence-divergent ZYMV and enriches the genetic diversity of ZYMV isolates. Moreover, M. scabra is a common weed in farmland ecosystems, making it a potential reservoir for plant viruses that could spread to other crops, particularly cucurbit crops. Whether the potyviruses carried by M. scabra can infect other Cucurbitaceae plants and pose a threat to their production is a significant scientific question that warrants further investigation. However, the transmissibility and pathogenicity of these two virus isolates require more detailed and systematic studies to fully understand their potential impact.
The existence of these complete sequence will enhance our understanding of the genome variation, evolution, and diversity among different ZTMV and ZYMV isolates. Our molecular characterization revealed that, whether for ZTMV-ms or ZYMV-ms, comparisons of the amino acid or nucleotide sequence identities of the 10 mature proteins with their best BLASTn/BLASTx hits showed a general trend: proteins near the C-terminus are more conserved, while those near the N-terminus exhibit higher variability. This pattern is particularly pronounced in ZYMV-ms. These findings are consistent with our previous observations in BCMV (Xu et al., 2024). Further analysis of the polyprotein of ZTMV-ms revealed the presence of several known sequence motifs associated with systemic movement and aphid transmission, including the KLSC motif (aa positions 579-601), the CSC motif (838-840), the PTK motif (856-858), and the DAG motif (3,065-3,067) near the N-terminus of the CP. Similarly, for ZYMV-ms, we identified the KLSC (aa positions 364-367), CCC (602-604), and PTK (620-622) motifs in HC-Pro, as well as a DAG motif (2,812-2,814) near the N-terminus of the CP (Atreya et al., 1991; Maia et al., 1996). The presence of these motifs suggests that both ZTMV-ms and ZYMV-ms, identified in M. scabra, may be capable of aphid-mediated transmission. This further underscores the importance of investigating whether these viruses can be transmitted to other Cucurbitaceae plants through aphids.
We also characterized the features of vsiRNA of both ZTMV-ms and ZYMV-ms, which were generated by the host’s RNAi pathway against viral infections. In this study, we found that vsiRNAs of both ZTMV-ms and ZYMV-ms were mainly 21 nt and 22 nt, with 22 nt being the most abundant. These properties of ZTMV-ms and ZYMV-ms vsiRNAs are consistent with the model of plant vsiRNA biogenesis in which viral long dsRNA precursors are processed into 21- and 22-nt vsiRNAs by Dicer-like 4 (DCL4) and DCL2, respectively (Diaz-Pendon et al., 2007; Ding, 2023). Hence, it can be inferred that DCL2 and DCL4 play a key role against ZTMV-ms and ZYMV-ms. Small RNA deep sequencing has demonstrated that the 21-nt class of vsiRNAs produced by DCL4 is the most dominant species of vsiRNAs produced that target RNA viruses in plants. DCL2-dependent 22-nt vsiRNAs accumulate at much lower levels when DCL4 is functional and seem to be less effective at mediating antiviral RNAi than 21-nt vsiRNAs (Guo et al., 2019). However, for the BCMV isolates we previously identified in Gardenia, the primary peak of the vsiRNA was at 21 nt, followed by 22 nt (Xu et al., 2024). This inconsistency suggests that the primary Dicer enzyme involved in RNAi-based antiviral defense may vary among species. Alternatively, it could indicate that the expression or activity of DCL2 in M. scabra is significantly higher than that of DCL4. And our study revealed a preferential occurrence of uracil residues at the 5′-termini of ZTMV-ms and ZYMV-ms vsiRNAs, respectively. At the 5′ end of small RNAs, the initial nucleotide typically regulates the sorting of small RNAs into specific AGO complexes in plants (Mi et al., 2008). In this study, we observed that the vsiRNAs of ZTMV-ms and ZYMV-ms showed a preference for ‘U’ and ‘C’ at their 5′ ends, suggesting the involvement of AGO1 or AGO5 in their recruitment and sorting (Mi et al., 2008). Furthermore, analysis of the polarity and genomic distribution of ZTMV-ms and ZYMV-ms vsiRNAs revealed that they were relatively evenly distributed across the viral genome, with a comparable preference for both the sense and antisense strands. These findings imply that the vsiRNAs are primarily derived from the dsRNA replication intermediates of ZTMV-ms and ZYMV-ms.
In conclusion, this study is the first to report that ZYMV and ZTMV can infect the novel host M. scabra, representing the first documented case of viral infection in this species. The full-length genomes of ZYMV-ms and ZTMV-ms were successfully recovered using a combination of RT-PCR, RACE, and Sanger sequencing. This research enhances our understanding of the genetic diversity, evolution, and epidemiology of ZYMV and ZTMV isolates that infect cucurbit plants. Furthermore, it contributes to the development of more effective diagnostic tools and virus control strategies.
PDF Links
PubReader
ePub Link
Full text via DOI
Download Citation
Supplement
Print
