High-Throughput Virome Surveillance Reveals Emerging Viral Threats to Cucumber Production in Taiwan

1. Introduction

Cucumber (Cucumis sativus L.) is one of the most important vegetable crops worldwide and represents a valuable component of protected vegetable production systems. Global cucumber production reached approximately 94.72 million tons in 2022, while Taiwan produced 47,083 tons from 1,955 ha, ranking 57th worldwide [1]. Benefiting from its favorable climate, Taiwan supports year-round cucumber cultivation, particularly in central and southern regions. Pingtung County accounts for nearly half of the national production, followed by Kaohsiung, Yunlin, and Changhua [2]. As a high-value vegetable crop, cucumber contributes more than US$30 million annually to Taiwan’s agricultural economy [3].

Insect pests and viral diseases frequently threaten the productivity and profitability of cucumber cultivation, the latter being particularly destructive because effective control measures are limited. In addition to causing direct feeding damage, insect vectors such as aphids, thrips, and whiteflies facilitate rapid virus dissemination within and between production systems. Consequently, viral epidemics can lead to substantial reductions in yield, fruit quality, and marketability, resulting in significant economic losses for growers.

More than 16 viruses infecting cucurbit crops have been reported in Taiwan. Major cucurbit viruses include the aphid-transmitted cucumber mosaic virus (CMV), papaya ringspot virus watermelon type (PRSV-W), and zucchini yellow mosaic virus (ZYMV); the thrips-transmitted melon yellow spot virus (MYSV) and watermelon silver mottle virus (WSMoV); and the whitefly-transmitted cucurbit chlorotic yellows virus (CCYV) and squash leaf curl Philippines virus (SLCuPV) [4,5,6,7]. Despite the widespread use of insect-proof netting and other protected cultivation practices, insect-transmitted viral diseases remain a major challenge in cucumber production systems. In particular, MYSV and CCYV are among the most prevalent and economically damaging viruses affecting greenhouse-grown cucumbers in Taiwan [6].

Effective disease management depends on accurate and timely pathogen diagnosis. However, cucurbit crops in commercial production fields are frequently infected by multiple viruses simultaneously, often resulting in complex symptom expression and making disease diagnosis difficult. Conventional diagnostic methods, including enzyme-linked immunosorbent assay (ELISA) and reverse transcription-polymerase chain reaction (RT-PCR), are valuable tools for detecting known viruses but are less effective for identifying novel, divergent, or unexpected pathogens. As global trade, climate change, and vector population dynamics continue to influence the emergence and spread of plant viruses. The need for comprehensive virus surveillance has become increasingly important for sustainable crop production.

Recent advances in high-throughput sequencing (HTS) technologies have provided powerful tools for the comprehensive characterization of plant virome. Illumina and Oxford Nanopore Technologies (ONT) are currently the most widely used sequencing platforms, offering complementary advantages in sequencing accuracy, read length, and field applicability [8]. HTS-based virome analysis enables the simultaneous detection of multiple viruses without prior knowledge of their presence and has substantially accelerated the discovery of emerging viruses in cucurbit crops. Examples include cucurbit vein banding virus [9], cucurbit cytorhabdovirus 1 [10], cucurbit chlorotic virus (CuCV), and chieh-qua endornavirus [11]. Beyond virus discovery, HTS has become an important tool for crop health monitoring, disease surveillance, germplasm and seed health testing, quarantine inspection, and the development of disease management strategies and virus-resistant cultivars [12].

In this study, severe viral disease outbreaks occurring in protected cucumber production facilities in Xizhou Township, Changhua County, central Taiwan, were investigated using Illumina-based HTS. Comprehensive virome analysis was conducted to characterize the viral communities associated with symptomatic cucumber plants and to identify potential emerging viral pathogens. In addition to detecting several endemic cucurbit viruses, we identified emerging viruses that had not previously been recognized in cucumber production systems of Taiwan. The results provide valuable information on the diversity and composition of cucumber-associated viral communities, demonstrating that virome-based surveillance offers crucial insights for improving disease diagnosis and integrated pest management in cucumber production.

2. Materials and Methods

2.1. Sample Collection

Two batches of symptomatic cucumber samples were collected from commercial net-house production systems in Xizhou Township, Changhua County, Taiwan. Batch 1 consisted of 34 leaf samples collected on August 10, 2022, whereas Batch 2 comprised 27 leaf samples and 3 fruit samples collected on December 10, 2025.

2.2. Total Rna Extraction

Total RNA was extracted from cucumber samples using the Total RNA Miniprep Purification kit (GMbiolab, Taichung, Taiwan) according to the manufacturer’s instructions. The purified RNA was stored at −80 °C until further use.

2.3. Hts Analysis

Equal amounts of total RNA (10 ng per sample) from the Batch 1 of 34 samples were mixed and then used to construct a random primer-based complementary DNA (cDNA) library using the Universal RNA-Seq Library Preparation kit with NuQuant (Tecan, Männedorf, Switzerland) according to the manufacturer’s instructions. The quantity and quality of the resulting cDNA library were assessed using a Qubit^® 2.0 Fluorometer with the dsDNA High Sensitivity kit (ThermoFisher Scientific, Waltham, MA, USA) and an Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA, USA). Paired-end sequencing and subsequent bioinformatics analyses were performed by Biotools Co., Ltd. (New Taipei City, Taiwan). Library sequencing was conducted on the NovaSeq 6000 platform (Illumina, San Diego, CA, USA). Ribosomal RNA reads were removed using the SILVA database [13]. The remaining reads were de novo assembled using Trinity software [14]. Redundant sequences were clustered and removed using CD-HIT-EST [15], generating a nonredundant set of contigs. The assembled contigs were subsequently subjected to BLASTn and BLASTx searches against the National Center for Biotechnology Information (NCBI) database for viral annotation and identification.

2.4. Genome Sequence Verification of Emerging Viruses

Total RNA extracted from cucumber samples that tested positive for emerging viruses was used as the template for viral genome verification. Primer pairs were designed based on assembled contigs and reference viral sequences (Supplementary Table S1) to amplify overlapping fragments spanning the complete viral genomes. First-strand cDNA synthesis was performed using SuperScript IV reverse transcriptase (SSIV RTase) (ThermoFisher Scientific, Waltham, MA, USA) with 2 μg of total RNA and 200 nM reverse primer at 50 °C for 10 min, followed by enzyme inactivation at 80 °C for 10 min. PCR amplification was carried out in a 25-μl reaction containing 2 μl of cDNA template, 1.25 U Blend Taq-Plus DNA polymerase (Toyobo, Osaka, Japan), 200 μM dNTPs, and 200 nM each of forward and reverse primers. The PCR program consisted of an initial denaturation at 94 °C for 3 min, followed by 35 cycles of 94 °C for 30 s, 58 °C for 1 min, and 72 °C for 3 min, with a final extension at 72 °C for 7 min. Amplified products were analyzed by electrophoresis on 1% agarose gels and eluted using the Micro-Elute DNA Clean/Extraction kit (GeneMark, Taichung, Taiwan) according to the manufacturer’s instructions. Eluted PCR products were cloned using the Invitrogen TOPO TA cloning kit (ThermoFisher Scientific, Waltham, MA, USA), ligated with the pCR2.1-TOPO vector. The ligation products were introduced into E. coli strain DH5α competent cells following the manufacturer’s protocol. Recombinant plasmids were purified from selected colonies and sequenced by Mission Biotech Co., Ltd. (Taipei, Taiwan) using an Applied Biosystems 3730XL DNA Analyzer (ThermoFisher Scientific, Waltham, MA, USA). Bidirectional Sanger sequencing was initially performed using M13 forward and reverse primers. Subsequently, sequence-specific primers were designed based on the obtained sequences to perform primer walking. For each amplicon, three independent clones were sequenced to ensure sequence accuracy.

2.5. Cucv Genome Sequence Analysis

The genome sequences of different CuCV isolates and representative members of the genus Crinivirus were retrieved from the NCBI GenBank database for comparative analyses (Supplementary Table S2). Pairwise nucleotide (nt) and amino acid (aa) sequence identities were calculated using Sequence Demarcation Tool Version 1.2 (SDTv1.2), which also generated color-coded identity matrices based on pairwise sequence alignments [16]. Multiple sequence alignments were performed using ClustalX version 2.1 [17]. Phylogenetic analyses were conducted using MEGA X [18]. Maximum-likelihood (ML) phylogenetic trees were constructed based on the appropriate nt substitution models [19,20], and branch support was evaluated using 1000 bootstrap replicates.

2.6. Virus Detection

Total RNA extracted from the collected cucumber samples was used as the template. Virus-specific primer pairs used to detect corresponding viruses: CBLV3900F/CBLV4576R to CBLV [6], Crini-hsp70-f/Crini-hso70-r to CCYV [21], CMV-CP-Nco-F/CMV-CP-Sac-R to CMV (our laboratory, unpublished), MY-N516f/MY-N774r to MYSV [22], WS-N572f/WS-N808r to WSMoV [23], TMVmpf/TMV3nr to TMV [24], and CuCV-R2-1F/1R to CuCV (designed in this study). Primer pairs c40723-1F/c40723-1R, c43195-1F/c43195-1R, and c44091-1F/c44091-1R were used to detect unclassified viruses. Primer sequences are listed in Supplementary Table S3.

MYSV and WSMoV were tested by SYBR Green I-based real-time RT-PCR as described previously [22]. Testing was conducted using the QuantStudio 1 Real-Time PCR System (Applied Biosystems, Thermo Fisher Scientific, Waltham, MA, USA). The reaction mixture (10 µL) included 2 µL total RNA, 5 µL KAPA SYBR^® FAST qPCR 2×Master mix (KAPA Biosystems Inc., Woburn, MA, USA), 50 U SAMscript reverse transcriptase (GMbiolab, Taichung, Taiwan), 10 U RNase inhibitor (GMbiolab, Taichung, Taiwan), and 100 nM of each primer. The reaction conditions were cDNA synthesis at 42◦C for 30 min, DNA polymerase activation at 95 °C for 5 min, followed by 35 cycles of 95 °C for 15 s, 65 °C for 30 s, and 75 °C for 20 s. Melting curve analysis was performed from 65 °C to 95 °C with a 0.3 °C increment. Reactions were considered positive if both an exponential increase in fluorescence and a specific melting peak were detected.

Other viruses were tested by conventional RT-PCR. Amplification was performed by VeritiPro Thermal Cycler (Applied Biosystems, Thermo Fisher Scientific, Foster City, CA, USA) using the One-step RT-PCR kit (GeneMark, Taichung, Taiwan) according to the manufacturer’s instructions. Each 25-µl reaction contained 2 µl of total RNA, 12.5 µl of 2× RT buffer (containing dNTPs, Mg2⁺ and enzyme stabilizer), 0.5 µl of enzyme mix (reverse transcriptase and Taq DNA polymerase), 0.5 µl each of forward and reverse primers (final concentration, 200 nM), 0.1 µl of RNase block, and nuclease-free water to the final volume. The RT-PCR program consisted of RT at 50 °C for 30 min, initial denaturation at 94 °C for 2 min, followed by 35 cycles of 94 °C for 30 s, 58 °C for 30 s, and 72 °C for 1 min, with a final extension at 72 °C for 7 min. Amplified products were analyzed by electrophoresis on 1% agarose gels. To verify virus identity, representative amplicons were randomly selected for Sanger sequencing. Sequencing was performed by Mission Biotech Co., Ltd. (Taipei, Taiwan) using an Applied Biosystems 3730XL DNA Analyzer (ThermoFisher Scientific, Waltham, MA, USA).

3. Results

3.1. Illumina Sequencing Data Analysis and Contig Assembly

Illumina sequencing generated 23807744 raw reads from the total RNA mixture of the Batch 1 samples. After removal of adaptor sequences, ambiguous reads, and low-quality reads, 23459635 clean reads were retained for downstream analysis. De novo assembly of the clean reads yielded 73518 contigs, with N50 and N90 values of 1708 and 341 bases, respectively (Table 1 and Figure 1).

3.2. Contig Annotation Against Viruses

The assembled contigs were subsequently subjected to BLAST searches against the NCBI database for viral annotation. A total of 22 contigs were annotated to nine viruses. Seventeen contigs showed 86.7%–99.9% base identities to the genomes of six viruses: CBLV, CCYV, chieh-qua chlorotic virus (also known as CuCV) [11], CMV, MYSV, and WSMoV, and cover almost their entire genome sequences. A short contig of 724 bases was closely related to tobacco mosaic virus (TMV), with 98.8% identity. Except for CuCV, the rest are endemic viruses. Additionally, four contigs were annotated to, but distinct from, two novel viruses: three to blackberry virus X (BVX), an unclassified virus of the order Tymovirales, and one to Sunn-hemp crinivirus (ShCV), an unclassified crinivirus. This suggested the possible existence of unknown viruses. The vial annotation results are summarized in Table 2.

3.3. Verification of the Presence of Emerging Viruses

The total RNA mixture used for HTS was tested for the presence of annotated emerging viruses by RT-PCR. CuCV and unknown viruses annotated to BVX and ShCV were tested using primers designed based on the corresponding contig sequences (Supplementary Table S3). Because the BVX-annotated contigs c43195_g2 and c43195_g1 shared head-to-tail overlapping sequences, forward and reverse primers c41395-1F and c41395-1R were designed from contigs c43195_g2 and c43195_g1, respectively, to assess if the two contigs belong to the same virus. The results showed positive for CuCV and the BVX-annotated unknown viruses (contigs c40723_g1, c43195_g2, and c43195_g1) (Figure 2), but negative for the ShCV-annotated unknown virus (contig c44091_g2) (data not shown). PCR products were sequenced to confirm their correctness. Additionally, the results also suggested that BVX-annotated contigs c40723_g1, c43195_g2, and c43195_g1 may be the same virus.

3.4. Clarification of the Bvx-Annotated Unknown Virus

Primers designed based on sequences of the three BVX-annotated contigs were used to successfully amplify four overlapping fragments covering the three contigs (Supplementary Table S1 and Figure 3). The overlapping fragments were cloned, and their sequences were determined by Sanger sequencing. By filling gaps, a 5394-base fragment containing an open reading frame (ORF) was obtained. BLAST searches against the NCBI database showed that the 5394-base sequence was best fit to BVX IPW_116 (accession number PQ317260), with 67.3% nt identity but only 12% coverage. The putative protein encoded from its ORF shared 37.6% aa identity and 97% coverage with the BVX polyprotein. The results demonstrated the presence of a novel virus in the cucumber field, tentatively named cucumber tymovirus (CTyV). Its sequence has been deposited in GenBank with accession number PZ836817.

3.5. Verification of Cucv Complete Genome Sequence

Because CuCV was first detected in Taiwan, its complete genome sequence was also verified by Sanger sequencing. Primers used in RT-PCR amplification were designed based on the annotated contig sequences and the available CuCV genome sequences in the GenBank database (Supplementary Table S1). Complete RNA1 and RNA2 sequences were obtained from overlapping fragments, as illustrated in Figure 4. The RNA1 and RNA2 of the virus, denoted XZ-C-1, were determined as 9157 nt and 7788 nt long, respectively, as similar structure and length to those of characterized CuCV isolates. Their sequences are nearly identical to the contigs c45922_g1 and c44091_g1 sequences, and have been deposited in GenBank with accession numbers PQ618563 and PQ618564.

The full-length nt sequences of XZ-C-1 RNA1 and RNA2 were aligned with those of other CuCV isolates, sharing high identities of 98.3%–99.8% and 99.3%–99.9%, respectively. By comparison with the whole genome sequences of representative cucurbit yellow stunting disorder virus (CYSDV) and CCYV isolates, 59.7%–66.8% nt identities were shown. The putative proteins encoded by RNA1 and RNA2 of XZ-C-1 were also closely related to those of other CuCV isolates, with 92.1%–100.0% aa identities (Supplementary Table S4). The results of pairwise comparison and phylogenetic analysis showed that XZ-C-1 clustered with CuCV isolates, but was distant from CYSDV and CCYV, and was therefore a CuCV isolate (Figure 5).

3.6. Virus Occurrences in the Same Cucumber-Producing Area in 2022 and 2025

Cucumber samples from Batch 1 were individually tested for six endemic viruses, CBLV, CMV, CCYV, MYSV, WSMoV, and TMV, and the two emerging viruses, CuCV and CTyV. All samples (100%) were positive for MYSV, but negative for TMV. Seven (20.6%) were positive for CuCV, four (11.8%) for CCYV, three (8.8%) each for CMV and CTyV, and one (2.9%) each for CBLV and WSMoV (Table 3). Fourteen samples (41.2%) were co-infected with two to four viruses. In conclusion, seven of the nine annotated viruses were detected.

Furthermore, a total of 30 symptomatic cucumber samples (Batch 2) including 27 leaf samples and 3 fruit samples were collected on December 10, 2025 to investigate the occurrences of the nine annotated viruses in the same cucumber-producing area. The same method was used for virus testing. None of all samples were positive for CMV, TMV, and the ShCV-annoated hypothetical virus. Of 29 samples (96.7%) were positive for WSMoV, followed by 27 (90.0%) for MYSV, 25 (83.3%) for CCYV, 16 (53.3%) for CuCV, 12 (40.0%) for CTyV, and five (16.7%) for CBLV (Table 3). All tested samples were co-infected with two to five viruses. In conclusion, six viruses were detected in this survey.

4. Discussion

Viral diseases are an important constraint to cucumber production in Taiwan, particularly in protected cultivation systems where favorable environmental conditions and continuous cropping can facilitate virus and vector spread. In this study, Illumina-based HTS provided a comprehensive overview of viruses associated with symptomatic cucumber plants in a commercial net-house production area in Xizhou Township, Changhua County. The analysis identified six previously reported viruses, CBLV, CCYV, CuCV, CMV, MYSV, and WSMoV, as well as a previously unrecognized virus, tentatively designated cucumber tymovirus (CTyV). This finding demonstrates that the viral complexity of cucumber production systems may be greater than indicated by routine pathogen-specific diagnostics.

The identification and characterization of the emerging CuCV is of particular significance for cucumber production in Taiwan. The complete genome of the CuCV XZ-C-1 isolate consisted of RNA1 and RNA2 and showed 98.3–99.8% and 99.3–99.9% nt identity, respectively, with previously reported CuCV isolates. Phylogenetic analysis further placed XZ-C-1 within the CuCV cluster and clearly separated it from CCYV, another cucurbit-infecting crinivirus prevalent in Taiwan. These results confirm the occurrence of CuCV in the investigated cucumber production system and expand the known distribution of this virus in Taiwan. To our knowledge, CCYV and CuCV cause similar interveinal chlorosis and yellowing symptoms in the lower leaves of cucurbit hosts [11,25]. Previously, we have developed a RT-PCR method to detect CCYV in the field [21]. In this study, we developed a CuCV-specific RT-PCR detection method to distinguish between them. The host range and distribution of CuCV in Taiwan will be further investigated to elucidate its impacts on the agricultural system.

The discovery of CTyV provides further evidence that emerging viruses can remain undetected when surveillance relies exclusively on established diagnostic targets. The initial contigs were best fit to BVX, an unclassified virus that is related to the family Betaflexiviridae of the order Tymovirales [26], but the assembled sequence showed only 67.3% nt identity over 12% of the BVX genome, indicating that CTyV is a distinct virus. The assembled 5,394-nt fragment appears to be a partial genome sequence of CTyV, whose complete genome sequence remains to be determined to clarify its taxonomic status. Additionally, its repeated detection in cucumber samples warrants further investigation of its host range, transmission, distribution, and potential effects on crop performance.

The detection of multiple viruses in the same production system is particularly relevant to disease management. In 2022, 41.2% of the tested samples were co-infected, whereas all samples collected in 2025 were co-infected with two to four viruses. Mixed infections in a plant are common and can produce overlapping symptoms and complicate field diagnosis, making symptom-based identification unreliable. Similar mixed infections have been previously reported in cucurbit production systems in Taiwan [6,27] and emphasize the value of HTS as a complementary diagnostic approach for detecting multiple viruses simultaneously. Importantly, HTS should not replace routine diagnostics but can serve as a discovery and surveillance platform, after which specific RT-PCR assays can be developed for rapid monitoring of newly recognized viruses.

The high detection rates of MYSV, 100% in 2022 and 90.0% in 2025, as well as our previous report [6], indicate it is the most dominant cucumber-infecting virus in Taiwan. A notable finding was the marked increase in the incidence of several viruses between the two sampling periods: WSMoV increased from 2.9% to 96.7%, CCYV increased from 11.8% to 83.3%, CuCV increased from 20.6% to 53.3%, CTyV increased from 8.8% to 40.0%, and CBLV increased from 2.9% to 16.7%. MYSV and WSMoV are associated with thrips transmission, whereas CCYV and CuCV are associated with whitefly-mediated transmission. Vector abundance may affect virus incidence. Thrips and whiteflies could be easily observed in the net houses. The high incidence of the viruses in 2025 indicates considerable and persistent insect vector and viral disease pressures in this production system.

In fact, two common aphid-transmitted cucurbit viruses, PRSV and ZYMV, were also tested in 2022 and 2025, but all cucumber samples were negative for them (data not shown). This supports the HTS analysis results. CMV can also be associated with aphid transmission. The low incidence of CMV, 8.8% in 2022, but no positives in 2025, is consistent with few aphids being found in the net houses.

The coexistence of viruses transmitted by different insect vectors has important implications for integrated disease management. Insect-proof netting alone may not provide sufficient protection when tiny insects such as thrips and whiteflies are present. Effective management should combine vector exclusion and regular virus surveillance. Such an integrated approach is particularly important in protected cultivation, where repeated cropping may facilitate the persistence and accumulation of viruses and their vectors.

The present study also highlights the practical value of HTS for agricultural disease surveillance; however, conventional RT-PCR to verify virus presence cannot be ruled out. For example, TMV and ShCV-related virus were annotated but could not be detected by RT-PCR, presumably due to assembly errors. However, HTS remains limited due to the limitations of the GenBank database; the possibility of unknown viruses still existing in the field cannot be ruled out.

The discovery of CuCV and CTyV from symptomatic cucumber plants allowed virus-specific primers to be developed and subsequently used to determine their occurrence in additional field samples. This discovery-to-detection workflow provides a practical strategy for incorporating emerging viruses into routine crop health monitoring. HTS-based surveillance can therefore complement conventional diagnostic methods by providing an early-warning system for newly emerging pathogens and supporting the development of targeted disease-management measures.

Overall, this study reveals a diverse and dynamic virus community associated with cucumber production in Taiwan and identifies CuCV and CTyV as emerging viruses requiring further attention. The high frequency of mixed infections and the substantial incidence of several viruses emphasize that cucumber disease management should move beyond single-pathogen diagnosis toward integrated, surveillance-based approaches. Incorporating HTS for virus discovery and periodic monitoring, followed by rapid virus-specific molecular diagnostics, could improve early detection and support more effective integrated disease management in protected cucumber production systems.