Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 May 15;25(7):e14111. doi: 10.1111/1755-0998.14111

An Innovative Binding‐Protein‐Based dsRNA Extraction Method: Comparison of Cost‐Effectiveness of Virus Detection Methods Using High‐Throughput Sequencing

Mamadou L Fall 1,✉, Dong Xu 1, Pierre Lemoyne 1, Geneviève Clément 2, Peter Moffett 2, Christophe Ritzenthaler 3
PMCID: PMC12415820  PMID: 40370068

ABSTRACT

Viral diseases represent a threat to global food production. Managing the impact of viruses on crop production requires the ability to monitor viruses, study their ecology and anticipate outbreaks. Double‐stranded RNA (dsRNA) sequencing is a well‐established and reliable method of detecting viruses and studying virome‐host interactions and ecology. Compared to total RNA extraction, dsRNA extraction eliminates the majority of host RNAs, improving the recovery of viral RNAs. In this study, we developed and evaluated a novel dsRNA extraction method for high‐throughput sequencing (HTS) applications based on the Flock House virus (FHV) B2 protein (B2‐based method), and compared its performance with that of established cellulose‐based and DRB4‐based methods (commercial kit), as well as total RNA extraction techniques. The electrostatic properties of B2 have been instrumental in developing a bead‐free and resin‐free dsRNA extraction method. The B2‐based method demonstrated high viral read recovery, achieving proportions exceeding 20% in most samples, and provided better dsRNA purity with less low weight molecule co‐extracted RNA than the DRB4‐based method and cellulose‐based methods. Despite producing overall fewer total reads than the DRB4‐based method, the B2‐based enrichment for viral‐derived dsRNA was better, with a higher percentage of viral reads, making it effective in virome profiling. Furthermore, it had an excellent detection specificity (0.97) and a good detection sensitivity (0.71), minimising false positives and false negatives. In addition, the B2‐based method proved to be highly cost‐effective, with a per‐reaction cost of $4.47, compared to $35.34 for the DRB4‐based method. This method offers a practical solution for laboratories with limited resources or for large‐scale sampling for viral ecology studies. Future improvements to the B2‐based method should focus on optimising sensitivity to Vitivirus species and developing scalable, automated workflows for high‐throughput viral detection.

Keywords: capture sequencing, dsRNA, grapevine virome, HTS, viral ecology, Viromics

1. Introduction

Viral diseases pose a substantial threat to global food production, requiring effective monitoring and control strategies (Scholthof 2007; Jones 2009). Unlike bacterial or fungal infections, plant pathogenic viruses currently have no practical treatments, putting emphasis on early detection and area‐wide surveillance in order to mitigate the impact of these infections (Hull 2002). Understanding viral ecology and proactive monitoring enables timely mitigation measures to be implemented to reduce the damage caused by viral outbreaks (Carisse et al. 2017). The advent of second‐generation and third‐generation sequencing technologies has made high‐throughput sequencing (HTS) a routine tool in the surveillance and viral ecology studies. Four nucleic acid extraction methods—virion‐associated nucleic acid (VANA), small RNA, total DNA and/or RNA (metagenomic and metatranscriptomic) and double‐stranded RNA (dsRNA)—are commonly used to capture and sequence viral sequences in diverse types of environmental samples, including plants and soil (Roossinck 2014; Fall et al. 2020; Javaran et al. 2023; Poursalavati, Larafa, et al. 2023; Poursalavati, Javaran, et al. 2023). VANA sequencing tends to favour DNA and enveloped RNA viruses, potentially skewing the representation of the viral community (Poursalavati, Larafa, et al. 2023; Hillary et al. 2022). This bias can result in an incomplete understanding of the viral landscape and ecology, leading significant viral threats to be overlooked. Small RNA sequencing offers a highly sensitive and cost‐effective approach for plant virus detection by leveraging the host's production of virus‐derived small interfering RNAs (siRNAs), which can constitute up to 30% of total small RNAs in infected plants. It simplifies sample preparation compared to other methods, allowing faster processing without sequence‐independent amplification. However, its limitations include the requirement for active viral replication, potential host‐specific biases in siRNA size distribution, and the need for complex bioinformatics analysis to accurately assemble and interpret viral sequences (Wu et al. 2015). Metagenomic and metatranscriptomic methods require deep sequencing, which is costly and is influenced by the presence of large‐genome organisms, complicating the identification of viruses in complex environmental samples. In contrast, dsRNA sequencing has a number of advantages over total DNA and RNA sequencing (Javaran et al. 2023; Decker et al. 2019). First, dsRNA serves as the genomic RNA of many viruses or as a viral replication intermediate, making it a highly specific target that reduces background noise from host RNA and DNA, which enhances the sensitivity and specificity of virus detection. Detection specificity is crucial because it reduces the complexity of the sequencing data, allowing for the more straightforward identification and characterisation of viral genomes (Javaran et al. 2023). Additionally, dsRNA sequencing is less affected by the presence of ribosomal RNA (rRNA), which dominates total RNA and can obscure viral sequences in metatranscriptomic analyses (Gaafar et al. 2021). The dsRNA method eliminates the need for rRNA depletion, streamlining library preparation and reducing both costs and processing time (Roossinck 2014). Another advantage of dsRNA sequencing is its ability to capture a range of RNA and DNA viruses, including low‐abundance viruses, which are often missed by total RNA sequencing due to competition from the more abundant host RNA (Javaran et al. 2023; Poursalavati, Larafa, et al. 2023; Roossinck 2012). This capability is particularly important in comprehensive virome analysis, where detecting low‐abundance viruses can provide insights into viral ecology and epidemiology. Finally, the enrichment of viral dsRNA also increases sequencing depth, leading to better genome coverage (Javaran et al. 2023).

Despite these advantages, using dsRNA extraction methods for HTS virus detection presents challenges, including complexity, labour intensiveness and high cost, which can limit throughput and increase the risk of errors. Also, these extraction methods may generate dsRNA of suboptimal quality or in insufficient quantities, thereby compromising the sensitivity and accuracy of HTS applications (Roossinck 2012). Contamination with other nucleic acids species is also a major concern, as it may impede accurate virus detection and quantification (Roossinck 2014). High costs of reagents and consumables, as well as labour intensiveness can also represent barriers for some laboratories. In response to some of these limitations, the Plant Viral dsRNA Enrichment Kit from MBL Life Science, launched in the early 2010s, offers an effective solution for enriching viral dsRNA from plant samples, improving subsequent virus detection and characterisation with HTS (MBL Life Science, 2019) (Atsumi et al. 2015). However, this kit is rather expensive, and can be difficult to obtain, particularly in North America. Initially, we sought to develop a similar tool, but this proved to be less cost‐effective than other methods (Rott et al. 2024). We have since overcome these challenges by developing a novel B2‐based method for dsRNA extraction using a dsRNA‐binding protein from the Flock House virus (FHV). The B2 protein exhibits unique physical and biochemical properties that enable its high‐affinity and sequence‐independent binding to dsRNA. Structurally, B2 functions as a dimer, forming a four‐helix bundle that interacts specifically with one face of an A‐form RNA duplex (Chao et al. 2005; Lu et al. 2005; Lingel et al. 2005). The dimeric structure allows B2 to bind two successive minor grooves and the intervening major groove of the dsRNA, without disrupting the RNA's overall conformation (Chao et al. 2005). Key interactions occur through hydrogen bonding and electrostatic forces with the ribose‐phosphate backbone of the RNA. Residues such as lysine and arginine play critical roles in stabilising these interactions, while van der Waals contacts, involving residues like cysteine and methionine, further enhance binding specificity (Chao et al. 2005). Biochemically, B2 exhibits a remarkable binding affinity, with a dissociation constant (Kd) of approximately 1 nM, enabling it to effectively coat longer dsRNAs (Chao et al. 2005). Also, a critical aspect of B2's function is its pH‐dependent binding and dissociation mechanism. Changes in pH alter the ionisation states of amino acid side chains, modifying the electrostatic properties of the protein. For example, a decrease in pH increases the protonation of basic residues, amplifying positive charges on the B2 surface, which strengthens interactions with the negatively charged phosphate groups of dsRNA. Conversely, these protonation‐driven dynamics also induce localised conformational flexibility, which may reduce binding affinity by destabilising protein‐RNA complexes at lower pH (Parvez et al. 2024). This dual effect of pH highlights a balance between electrostatic attraction and structural adaptability, which is necessary for B2's role in binding dsRNA. These properties make B2 a highly versatile and efficient tool for targeting dsRNA within a pool of diverse nucleic acids types. Building on these properties, we developed the B2‐based dsRNA extraction method that leverages B2's ability to form stable dsRNA‐B2 complexes. These complexes can be efficiently separated using B2's pH‐dependent binding and dissociation properties, combined with straightforward centrifugation steps, to optimise dsRNA extraction. We carried out a comparative analysis to evaluate this method's cost‐effectiveness against that of four existing nucleic acid extraction methods (two total RNA extraction and two proven dsRNA extraction methods) for virus detection using the Illumina MiSeq HTS platform. Our results demonstrate that among the methods tested, the DRB4 dsRNA method (commercial kit) exhibited the highest accuracy, followed by the B2‐based dsRNA and cellulose‐based methods. Notably, the B2‐based dsRNA method emerged as the most affordable option, providing significant advantages in large‐scale applications. The ability to cost‐effectively detect a wide range of plant viruses is important when aiming at implementing effective disease management and mitigation strategies.

2. Methods

2.1. Validation Plants and Sampling

Grapevine shoots were collected from various vineyards in southern Quebec, Canada (Table 1). A certified virus‐free plant cutting of Pinot noir CL MARIAFELD M3 (PN_M3_1349), originally from Switzerland, was generously provided by the Canadian Food Inspection Agency (CFIA) in Sidney, British Columbia, Canada. The cuttings were propagated in Pro‐Mix Pro soil under greenhouse conditions. The Chaudo_PN9 sample was produced by grafting a virus‐free Chardonnay plant (sourced from Ontario, Canada, and provided by CFIA) as the rootstock with a Grapevine leafroll‐associated virus 3 (GLRaV‐3) infected Pinot noir scion. The newly grafted plants and other propagated plants were maintained under natural lighting in a greenhouse with temperatures in the range of 18°C–30°C and relative humidity levels of 60%–85%.

TABLE 1.

Grapevine samples, origin and varieties used in this study.

Sample name Origin (city/province, country) Variety
CO3_36 Frelighsburg, Canada Vidal
TM6_75 Frelighsburg, Canada Vidal
DSJ3_2 Quebec, Canada Pinot noir
BacPN4_2 Quebec, Canada Pinot noir
Bac9_2 Quebec, Canada Pinot noir
FN_1 Quebec, Canada Frontenac noir
P_GPgV Frelighsburg, Canada Vidal
Chaudo_PN9 Ontario, Canada Chardonay + Pinot noir
CA15_6 Frelighsburg, Canada Vidal
Comb_9 — Combined all the above 9 varieties
PN_M3_149 Unknown, Switzerland Pinot noir
Open in a new tab

Mature leaves were collected and immediately ground to a powder in liquid nitrogen using a mortar and pestle. For the Chaudo_PN9 sample, leaves were collected from branches well clear of the graft union. To evaluate the methods used, a tenth sample, referred to as ‘Mix’, was created by combining one gram of tissue powder from each of the nine virus‐infected samples (CO3_36, TM6_75, DSJ3_2, BacPN4_2, Bac9_2, FN_1, P_GPgV, Chaudo_PN9 and CA15_6). The Mix was stored at −80°C for subsequent analyses as 1‐g aliquots in 50 mL Falcon tubes.

2.2. Total RNA Extraction

2.2.1. Cetyltrimethylammonium Bromide‐Based (CTAB) Extraction Method

This method involved a cetyltrimethylammonium bromide (CTAB) protocol, using a lysis buffer containing 100 mM Tris‐HCl, 25 mM EDTA (pH 8.0), 2 M NaCl, 4% CTAB and 4% β‐mercaptoethanol. Frozen grapevine powder was mixed with the hot lysis buffer at 65°C for 2 min, and the total RNA was precipitated at −20°C for 1 h after the addition of 0.3 M sodium acetate and two volumes of 100% ethanol. The precipitated RNA was then resuspended in RNase‐free water and treated with DNase I for 20 min to remove contaminating DNA. DNase I was removed with a single chloroform extraction, followed by RNA precipitation and resuspension as described above.

2.2.2. Spectrum‐Based (SPEC) Extraction Method

The Sigma‐Aldrich Spectrum Plant Total RNA Kit (MilliporeSigma, MA, USA) was used following the manufacturer's procedure. The eluted total RNA was also treated with DNase I to ensure the removal of any residual DNA before downstream analysis.

2.3. dsRNA Extraction

2.3.1. B2‐Based Extraction Method

The B2 protocol utilised a polyhistidine‐tagged recombinant protein approach. A DNA fragment (225 nt) from the FHV genome encoding the N‐terminus 75 amino acids of the B2 gene (X77156.1) was synthesised and cloned into the pLATE31 vector (Cloning Kit pLATE31, #K1261, Thermo Fisher Scientific, MA, USA) and transformed into chemically competent E. coli BL21(DE3) cells as per the manufacturer's instructions (Data S1). Positive transformants were expanded for B2 protein expression in lysogeny broth (LB) medium supplemented with ampicillin and induced with isopropylthio‐β‐galactoside (IPTG). The polyhistidine‐tagged B2 protein was purified using a HisTrap HP column (Millipore Sigma, USA) and desalted with a HiPrep Sephacryl column (Millipore Sigma, USA). To isolate dsRNA, 10% bovine serum albumin (BSA) rinsed tubes (0.5 and 5‐mL tubes) were used in all steps. Subsequently, 200 μL of binding buffer (BB) (20 mM Tris‐HCl, 10 mM EDTA [pH 8.0], 180 mM NaCl, 1% BSA) containing 40 μg of the total RNA extracted using the CTAB method was supplemented with 80 μg of purified B2 protein. The B2‐total RNA mixture was shaken at 350 rpm at 37°C for 10 min, and then transferred to 3.8 mL of pre‐cooled treatment buffer (TB) (20 mM Tris‐HCl, 10 mM EDTA [pH 8.0], 240 mM NaCl, 1% BSA) and shaken at 200 rpm at 10°C for 30 min. The solution was then centrifuged at 18,000 g for 10 min to collect the B2‐dsRNA complexes. The pellet was washed with 1 mL wash buffer (WB) (20 mM Tris‐HCl, 10 mM EDTA ([pH 8.0]), 80 mM NaCl), and centrifuged. The B2‐dsRNA complex was dissociated by adding 72 μL of 2.5 M LiCl, followed by incubation on ice for 3 min. After vortexing, 150 μL of anhydrous alcohol and 3 μL of glycogen (Thermo Fisher Scientific) were added, and the mixture was chilled at −20°C for 1 h before centrifugation at 18,000 g for 10 min. Lastly, the pellet was washed twice with cold 70% alcohol, air‐dried, and the dsRNA was resuspended in sterile water for subsequent analysis (Figure 1).

FIGURE 1.

FIGURE 1

Open in a new tab

Graphical representation of the method using a double‐stranded RNA (dsRNA) binding protein (B) to extract dsRNA from total RNA. The process involves three different buffer solutions: Binding buffer (BB), treatment buffer (TB) and washing buffer (WB) and one dissociation solution (LiCL, ETHOH).

2.3.2. DRB4‐Based Extraction Method

The Plant Viral dsRNA Enrichment Kit (MBL, Japan) was used following the manufacturer's instructions. First, 40 μg of total RNA extracted using the CTAB protocol was processed, and the final dsRNA was eluted in the elution buffer provided. To remove potential contaminants, the dsRNA was treated with chloroform, and the supernatant was precipitated with 0.3 M sodium acetate and two volumes of alcohol.

2.3.3. Cellulose‐Based (CELL) Extraction Method

For this protocol, 1 g of powdered leaf tissue per sample was used. Frozen leaf tissue (2 g) is homogenised with 8 mm glass beads in a 50 mL centrifuge tube, followed by extraction with buffer (18 mL) and 2‐mercaptoethanol (200 μL), shaken for ≥ 30 min at 400 rpm, and centrifuged at 1000 g for 1 min at 10°C. Potassium acetate (5.8 M, 18 mL) is added, mixed and centrifuged at 15,000 g for 30 min at 10°C. The supernatant (25 mL) is filtered through three layers of cheesecloth, mixed with 100% isopropanol (20 mL), and centrifuged at 1000 g for 30 min at 4°C. The pellet is re‐suspended in STE‐18 buffer (45 mL), centrifuged at 15,000 g for 30 min at 4°C, and purified using Sigmacell cellulose type 101 (0.5 g) with shaking at 400 rpm for ≥ 20 min. After multiple wash steps with STE‐18 and centrifugation at 15,000 g, the pellet is air‐dried and re‐suspended in 1 × STE buffer (5 mL). Three sequential elutions yield 15 mL of dsRNA solution, which is treated with T1 RNase (1000 units) and Turbo DNase (30 units) at 37°C for 30 min. The dsRNA is further purified with cellulose, washed with 70% ethanol, and precipitated using sodium acetate (40 μL, 3 M) and ethanol (1 mL). After centrifugation at 15,000 g for 30 min at 4°C and final ethanol washes, the dsRNA is re‐suspended and combined into a final volume of 25 μL (Fall et al. 2020). Following extraction, DNase I and RNase T1 treatments were applied to remove contaminated DNA and ssRNA. The digested samples were precipitated with 0.3 M sodium acetate and 70% alcohol. The washed pellet was suspended in 50 μL of autoclaved Milli‐Q purified water.

2.4. cDNA Preparation

Total RNA was first treated with the QIAseq FastSelect‐rRNA Plant Kit (Qiagen, USA) to remove rRNA, following the manufacturer's instructions. The rRNA‐depleted total RNA and dsRNA samples were then denatured by heating at 99°C for 5 min, followed by rapid cooling in ice water for 5 min. Reverse transcription was performed using SuperScript III (Invitrogen, MA, USA) for first‐strand synthesis and Klenow DNA polymerase I (NEB, MA, USA) for second‐strand synthesis. The resulting double‐stranded cDNA (dscDNA) was purified using Mag‐Bind TotalPure NGS beads (Omega Bio‐tek, GA, USA). The dscDNA was quantified using a Qubit 4 fluorometer (Thermo Fisher Scientific, USA) and adjusted to a concentration of 0.2 ng/μL.

2.5. Library Preparation and Illumina Sequencing

One nanogram of dscDNA was used for library preparation with the Nextera XT DNA Library Preparation Kit (Illumina, CA, USA), following the manufacturer's instructions. The quality of the libraries was assessed using the High Sensitivity DNA Reagents Kit with the Agilent 2100 Bioanalyzer system (Agilent, CA, USA). The libraries were then pooled in equimolar amounts to a final concentration of 4 nM. The denatured pooled libraries were diluted to 12 pM before loading onto the reagent cartridge of the MiSeq Reagent Kit v3. Sequencing was conducted with 2× 301 cycles using paired‐end reads on the MiSeq System (Illumina).

2.6. PCR Validation

Total RNA was reverse‐transcribed, and dscDNA was synthesised. The dscDNA was then purified using beads, following the previously described protocols. Purified dscDNA was quantified with a Qubit 4 fluorometer, and the concentrations were adjusted to 2 ng/μL. A virus‐free plant (PN_M13_1439) was used as the negative control (NC). Three technical replicates were prepared for each sample. Primers (Table S1) were used at a final concentration of 0.4 μM, with 1.4 ng of dscDNA per 10 μL reaction. qPCR was conducted using Bio‐Rad SsoAdvanced Universal SYBR Green Supermix in a Bio‐Rad CFX96 Touch Real‐Time PCR Detection System. The reaction conditions included initial incubation at 55°C for 10 min, followed by heating to 94°C for 5 min. Next, 40 cycles of DNA denaturation at 94°C for 15 s, primer annealing at 55°C for 30 s, and elongation of the target region at 72°C for 30 s were performed. After the qPCR cycles, the reaction was incubated at 72°C for 5 min, followed by a melt curve analysis. The melt curve consisted of 60 cycles, starting at 65°C, with a 0.5°C increment per cycle and a ramp rate of 0.5°C per second.

2.7. Data Analysis

The dsRNA‐based virus detection tool was developed and validated using the guidelines and principles described in Cardwell et al. (2023) and endorsed by the Diagnostic Assay Validation Network (DAVN). The raw sequencing data associated with this study has been deposited at the Sequence Read Archive (SRA, SUB14844148) of NCBI of NIH, USA: Bioproject PRJNA1197386, biosamples from SAMN45803673 to SAMN45803683.

2.8. Bioinformatics and Statistical Analysis

The raw FASTQ files from the MiSeq sequencer were demultiplexed and assessed for sequencing quality using FastQC (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Adapter sequences and low‐quality reads (below a quality score of 30) were removed using Trimmomatic V.0.32 (Bolger et al. 2014). The paired FASTQ files were then imported into two different pipelines designed for virus detection. The first pipeline, Virtool, is a tool developed by Rott et al. (2017), which maps sequencing reads directly to reference genomes from a customised plant virus database. The second pipeline, SOVAP, developed by our team (Poursalavati, Larafa, et al. 2023), assembles short reads into viral contigs, clusters these contigs, and then maps them to reference genomes from two distinct databases (NCBI and IMG/VR) (Figure 2). These pipelines use different approaches to virus detection: Virtool maps raw reads directly to reference genomes using BLASTn, while SOVAP assembles reads into contigs before mapping them and utilises both BLASTn and BLATx. As a result, any errors they produce are likely to be of a different nature.

FIGURE 2.

FIGURE 2

Open in a new tab

Graphical representation of the workflow, from nucleic acid extraction to bioinformatic processing and data analysis, used to handle the results from each of the five nucleic acid extraction methods. Two pipelines, SOVAP and Virtool, were employed, each using different approaches to process raw data from the MiSeq sequencer.

To validate the results from both pipelines, PCR results (see previous section) were used based on the following criteria for positive virus detection: (1) Cq<mean Cq of NC‐2× standard deviation; (2) Absence of Cq values for NC, Cq < 35; and (3) for positive samples, the merge curve and melt peak were checked. A true detection (true positive) required a positive PCR result and detection by at least one of the pipelines, while a false detection (false negative) required a negative PCR result and non‐detection by at least one pipeline (Figure 2). For each method, the total number of reads versus the number of viral reads was plotted, and the relative abundance of viral species was determined by calculating transcripts per million (TPM) (Krinos et al. 2024; Liang et al. 2019; Puente‐Sanchez et al. 2020; Ghaly et al. 2024). To compare the different methods in terms of viral read yield, the Kruskal–Wallis test was performed since the data distribution was not normal. A pairwise Wilcoxon test was used as a post hoc analysis to identify which methods differed significantly from the others. Heatmaps showing the relative abundance of each viral species in the samples were generated for each extraction method. To compare the performance of the five extraction methods, receiver operating characteristic (ROC) analysis was performed to calculate specificity (the proportion of healthy plants correctly identified as virus‐free), sensitivity (the proportion of virus‐infected plants correctly identified) and overall accuracy (the proportion of total plants correctly classified as either virus‐free or virus‐infected) (Fall et al. 2018; Fall and Carisse 2022). To aid in interpreting the ROC analysis, each detection event was assigned a unique random number using the runif function in R, and all detection events were plotted in a coordinate system. In addition, the cost per reaction for each extraction method was calculated prior to sequencing, excluding manpower expenses.

3. Results

3.1. dsRNA Purification Efficiency

Three optimised buffer solutions are integral to this process: the BB initiates dsRNA‐B2 complex formation, the TB stabilises the complexes, and the LiCl and alcohol solutions help to release dsRNA from B2. This approach overcomes the need for beads or resins, reducing complexity, cost and reliance on additional materials. The centrifugation steps further ensure the efficient separation and purification of dsRNA, improving yield while minimising contaminants like small RNA fragments (Figure 1). To evaluate the performance of the new B2‐based method, its efficiency was compared with four other nucleic acid extraction methods using a structured bioinformatics and data analysis approach. The evaluation employed two distinct pipelines, SOVAP and Virtool, which process raw sequencing data using different analytical frameworks. Both pipelines yielded identical virome profiles for each sample, detecting the same viruses. Additionally, PCR validation and ROC analysis were integrated into the assessment to ensure robust and comprehensive performance evaluation (Figure 2). The dsRNA profiles obtained using the three different extraction methods (cellulose, Plant Viral dsRNA Enrichment Kit [DRB4‐based], and B2‐based method) reveal key differences in efficiency and yield (Figure 3). The cellulose method (Figure 3A) produced a range of migration profiles. It produced faint and high molecular weight (HMW) dsRNA bands despite the application of RNase T1 after extraction, as well as small RNA fragments and impurities (lower bands). The DRB4‐based method (Figure 3B) generated overall strong signals, particularly in the low MW RNAs likely in the pooled sample (lane 10), but also showed substantial contamination from small fragments. The B2‐based method (Figure 3C) yielded the greatest amount of HMW dsRNA, with clear and strong bands across multiple samples, and less small RNA contamination. The virus‐free control (lane 11) in all panels did not display any HMW dsRNA bands, confirming the specificity of the extractions. These results indicate that, while the Plant Viral dsRNA Enrichment Kit effectively enriched dsRNA, the B2‐based method provided the highest dsRNA yield with less small RNA contamination (Figure 3).

FIGURE 3.

FIGURE 3

Open in a new tab

dsRNA profiles obtained with different extraction methods. Panel A shows dsRNA extracted using the cellulose method; Panel B, dsRNA extracted using the Plant Viral dsRNA Enrichment Kit (DRB4‐based); and Panel C, dsRNA extracted with the B2‐based method. The gel includes the following samples: (M) XLarge DNA Ladder RTU (GeneDireX Inc.); (1) CO3_36; (2) TM6_75; (3) BacPN4_2; (4) DSJ3_2; (5) Bac9_2; (6) FN_1; (7) P_GPgV; (8) Chaudo_PN9; (9) CA15_6; (10) Comb_9 extract from an equal weight mixture of samples 1–9; and (11) PN_M3_1349 (virus‐free control). The dsRNA samples were run on a 1% agarose gel with 1× SB buffer at 100 V for 30 min and stained with GelRed (EMD Millipore).

3.2. Viral Read Recovery Efficiency

The comparison between the total number of reads and viral reads in 10 different samples using five nucleic acid extraction methods (B2, CELL, DRB4, SPEC and CTAB) is presented in Figure 4. The B2‐based method provided a total number of reads ranging from 1.4 × 104 to 6.1 × 105, with viral read percentages between 4% and 65%. Eight out of 10 samples had viral read percentages exceeding 20% of the total number of reads. A similar pattern was found for the CELL method, with total reads ranging from 5.7 × 103 to 6.9 × 105 and viral read percentages from less than 4% to 94%; eight out of 10 samples showed viral read percentages exceeding 45% of total reads. The DRB4 method provided total reads ranging from 1.4 × 103 to 9.8 × 106 and viral read percentages from 1% to 73%; eight out of 10 samples had viral read percentages exceeding 20% of total reads. The SPEC method generally provided high total read counts, ranging from 4.1 × 105 to 1.4 × 106, but viral reads made up a smaller proportion, with percentages between 5% and 32%; eight out of 10 samples showed viral read percentages exceeding 20% of total reads. Lastly, for the CTAB method, the total read counts ranged from 2.0 × 105 to 4.1 × 106, with a span of 23%–40% for viral read percentages. Half of the samples showed viral read percentages between 23% and 30% and the rest between 32% and 40% (Figure 4).

FIGURE 4.

FIGURE 4

Open in a new tab

Comparison of total and viral reads from different samples using five nucleic acid extraction methods (B2, CELL, DRB4, SPEC and CTAB): The bar charts represent the total number of reads (red) and viral reads (green) for each sample, with the viral read percentages indicated by a dashed line. The box plots (red boxes edged in aqua blue for total reads and green boxes edged in red for viral reads) illustrate the distribution of total and viral reads from all extraction methods, emphasising the differences in viral read recovery between techniques. The lower boundary of each box represents the 25th percentile, the line inside the box indicates the median, and the upper boundary represents the 75th percentile. Whiskers (aqua blue for total reads and yellow for viral reads) extend above and below the box, marking the 90th and 10th percentiles. The yellow dashed line indicates the mean for viral reads, while outliers are shown as yellow circles for viral reads and grey squares for total reads.

Following the non‐parametric analysis, comparisons between methods in terms of viral read count were not significant (p > 0.050), except for the CTAB method, which yielded significantly more viral reads than the B2 method (p < 0.043). The box plot at the bottom right of Figure 4 summarises total reads and viral reads for all RNA extraction methods, with each method showing a distinct distribution of total and viral reads. For the B2‐based and CELL methods, both total read counts were relatively low compared to other methods, with the mean for viral reads near the 25th percentile and a few outliers present. The DRB4 method had the highest total read counts, with the median and mean viral reads near the 75th percentile and the total read count showing substantial variability and multiple outliers. For the SPEC method, total reads were low, with viral reads close to the 25th percentile and a few outliers visible. Lastly, the CTAB method exhibited moderate read recovery, with total reads distributed across a wider range, and mean viral reads situated around the 50th percentile, with fewer outliers compared to the other methods (Figure 4).

3.3. Viral Composition and Abundance

Graphical representations of viral composition and abundance in each sample are shown in Figures 5 and 6, respectively. Following the criteria for true detection of a given virus, the number of viruses across the 10 samples ranged from zero (virus‐free, sample PM_M3_1249) to six distinct viral species (sample TM6_75). In total, 10 different viral species were detected, including Grapevine rupestris stem pitting‐associated virus (GRSPV), Grapevine leafroll associated virus 2 (GLRaV‐2), Grapevine leafroll associated virus 3 (GLRaV‐3), Grapevine virus H (GVH), Grapevine virus B (GVB), Grapevine virus E (GVE), Grapevine Pinot gris virus (GPGV), Grapevine red blotch virus (GRBV), Grapevine red globe virus (GRGV) and Nepovirus lycopersici (Tomato ringspot virus, ToRSV). At least two distinct viral species were detected in each sample; the composite sample, which was a combination of all samples except the virus‐free one, contained all expected viral species except GRGV. GRGV was present in one sample (Chaudo_PN9, CO3_36), with a low percentage of transcripts per kilobase million (0.038% of TPM) (Figures 5 and 6; Figure S1). The most abundant viral RNA across all samples and methods was GRSPV, followed by GLRaV2 and GLRaV3. Vitivirus species (GVB, GVB and GVE) were the least abundant (0.003% < TPM ≥ 0.048%) (Figure 6 and Figure S1). In the composite sample (Comb_9), which contained nine distinct viral species with different titers, the DRB4‐based method was the only method to detect two of the three Vitivirus species and, out of the nine viruses present, only failed to detect GVH; the B2‐based method missed three out of the nine viruses (the three vitiviruses); the CELL method missed four out of nine; the CTAB method missed five out of nine; and the SPEC method missed four out of nine (Figure 6 and Figure S1).

FIGURE 5.

FIGURE 5

Open in a new tab

Virome composition of the 11 samples used to assess the efficiency of the five nucleic acid extraction methods compared in this study. The graph represents a total of 10 distinct viral species that were detected: Grapevine rupestris stem pitting‐associated virus (GRSPV), Grapevine leafroll‐associated virus 2 (GLRaV‐2), Grapevine leafroll‐associated virus 3 (GLRaV‐3), Grapevine virus H (GVH), Grapevine virus B (GVB), Grapevine virus E (GVE), Grapevine Pinot Gris virus (GPGV), Grapevine red blotch virus (GRBV), Grapevine Red Globe virus (GRGV) and Nepovirus lycopersici (Tomato ringspot virus, ToRSV). Note that the virus icons in the figure are symbolic and do not reflect the actual morphology of the virions.

FIGURE 6.

FIGURE 6

Open in a new tab

Heatmap showing the abundance of each virus in five extraction methods (B2‐based, CELL, DRB4, CTAB and SPEC). In this figure, “Expected” refers to the true detection of a virus, based on the criteria outlined in the main text and Figure 2; B2 represents the B2‐based dsRNA extraction method; CELL, the cellulose‐based dsRNA extraction method; DRB4 method, the DRB4‐based extraction method (commercial kit, MBL); CTAB, the cetyltrimethylammonium bromide‐based RNA extraction method; and SPEC, the total RNA extraction method using the Spectrum Kit. Viral abundance is expressed as a percentage of transcripts per kilobase million (TPM).

3.4. Performance and Cost‐Effectiveness of the Methods

A ROC analysis was used to calculate the specificity (the proportion of healthy plants correctly identified as virus‐free), sensitivity (the proportion of virus‐infected plants correctly identified) and overall accuracy of each method, shown in Figure 7. In terms of specificity, the B2‐based and CELL methods had the highest value (0.97), followed by CTAB and SPEC (0.95), and then DRB4‐based (0.92). For sensitivity, DRB4‐based had the best performance (0.81), followed by the B2‐based method and CELL (0.71); the CTAB and SPEC methods had the lowest sensitivity, with values of 0.67 and 0.5, respectively (Figure 8). The overall accuracy values were 0.86 for DRB4‐based, 0.83 for the B2‐based method, 0.83 for CELL, 0.81 for CTAB, and 0.76 for SPEC (Figure 7). The B2‐based method had by far the lowest pre‐sequencing cost per reaction ($4.47/sample), followed by CELL and DRB4‐based, at $35.14/sample and $35.54/sample, respectively. The SPEC method had the highest pre‐sequencing cost ($74.90/sample), followed by the CTAB method ($73.10/sample) (Figure 8). The time required to process a sample up to the library preparation step was estimated for each method. The minimum times recorded were 117 min for B2‐based, 145 min for CELL‐based, 90 min for DRB4‐based, 66 min for SPEC‐based and 138 min for CTAB‐based methods. However, direct comparisons of time efficiency should be made with caution. Some dsRNA‐based methods include cooling periods during incubation or centrifugation, which extend the total processing time. Therefore, these cooling periods allow for parallel experiments, meaning the actual hands‐on time is lower than the total duration.

FIGURE 7.

FIGURE 7

Open in a new tab

Graphical representation of the results of the receiver operating characteristic (ROC) analysis. Specificity refers to the proportion of healthy plants correctly identified as virus‐free, while sensitivity reflects the proportion of virus‐infected plants correctly identified. The area under the ROC curve (AUC) represents overall accuracy. To, TN, FP and FN denote true positive, true negative, false positive and false negative events, respectively. Each detection event was assigned a random unique value for plotting in the coordinate system. B2 refers to the B2‐based dsRNA extraction method; CELL, the cellulose‐based method; DRB4 method, the DRB4‐based commercial kit; CTAB, the cetyltrimethylammonium bromide‐based RNA extraction method; and SPEC, total RNA extraction using the Spectrum Kit.

FIGURE 8.

FIGURE 8

Open in a new tab

Breakdown of the costs associated with each of the five extraction methodologies. The total cost per reaction was calculated by compiling the ingredients and quantities of materials consumed in each method. B2 refers to the B2‐based dsRNA extraction method; CELL, the cellulose‐based method; DRB4 method, the DRB4‐based commercial kit; CTAB, the cetyltrimethylammonium bromide‐based RNA extraction method; and SPEC, total RNA extraction using the Spectrum Kit. Rtn stands for reaction.

4. Discussion

Double‐sequencing is a well‐established and proven method for detecting viruses (both DNA and RNA) and profiling the virome across various types of environmental samples (Diaz‐Ruiz and Kaper 1978; Dodds and Jordan 1984). Compared to total RNA sequencing, it reduces the need for host RNA depletion and/or deep sequencing, making it an attractive choice for large‐scale virome studies (Poursalavati, Larafa, et al. 2023). However, dsRNA extraction is more labour‐intensive and skill‐dependent than total RNA extraction methods (Roossinck 2014). This challenge aligns with observations by other authors (Gaafar et al. 2021; Roossinck et al. 2015), who stressed the importance of efficient dsRNA extraction in improving the sensitivity and accuracy of HTS. The introduction of the DRB4‐based kit in the late 2010s helped reduce the complexity of dsRNA extraction, making it as user‐friendly as total RNA extraction (Atsumi et al. 2015). However, this kit is not widely available, especially in North America, and can be difficult to procure (Rott et al. 2024). In Rott et al. (2024), the authors compared resin‐based and B2‐based dsRNA extraction methods, the latter using a binding protein (B2) derived from the FHV; they concluded that both methods could potentially yield better results with further optimisation. Both methods involve several steps, including nuclease treatment, and require around 6 h to process 30 samples, at an average cost of $6.75 per sample. In this study, we significantly improved and evaluated a novel B2‐based method for virus detection using HTS (Fall and Xu 2024), comparing it to other established techniques.

The performance of the B2‐based dsRNA extraction method is rooted in the unique biochemical properties of the B2 protein, which demonstrates high‐affinity and sequence‐independent binding to dsRNA through interactions with the ribose‐phosphate backbone (Chao et al. 2005). By developing three solutions, we overcome the need for beads or other matrices, reducing complexity and cost. Additionally, centrifugation steps are made to separate and purify the dsRNA, improving its yield while minimising contaminants like LWM RNA fragments. Contrary to existing methods that rely on dsRNA‐binding molecules coupled with beads or resins (Rott et al. 2024), our approach simplifies workflows by leveraging the biochemical properties of the B2 protein. Our results demonstrated that the B2‐based method provided significant improvements in dsRNA yield, specificity, and cost‐effectiveness compared to the other approaches, including the widely used cellulose‐based and DRB4‐based extraction methods.

The B2‐based method demonstrated better dsRNA recovery, as shown by consistent and strong dsRNA bands, with less LWM RNA contamination compared to the cellulose and DRB4‐based methods (Figure 3). While strong HMW dsRNA bands were observed in the DRB4‐based method, it was also more contaminated with LMW co‐extracted RNA species, which could interfere with downstream analyses—a challenge previously noted (Figueiredo Prates et al. 2023). In contrast, the B2‐based method provided higher dsRNA purity, which is critical for reducing background noise and avoiding the sequencing of unwanted co‐extracted RNAs; this results in overall better viral read recovery across samples, highlighting the efficiency of this method. Although the B2‐based method yielded fewer total reads than the DRB4‐based and SPEC methods, it recovered a good percentage of viral reads (often exceeding 20%) in most samples (Figure 4). This performance is comparable to the well‐established cellulose‐based dsRNA extraction method (CELL), which is known for its ability to extract high quantities of dsRNA enriched and less small RNAs, a common issue in the metagenomic and metatranscriptomic approaches (Gaafar et al. 2021; Cobbin et al. 2021). The B2‐based method's ability to detect multiple viral species with varying abundances shows its potential in comprehensive virome ecology studies (Figures 5 and 6). The performance analysis revealed that the B2‐based method presented a great specificity and sensitivity, with an overall accuracy of 0.83, which is close to the 0.86 accuracy obtained with the DRB4‐based method (Figure 7). However, one limitation of the B2‐based method, as observed in the composite sample analysis, was its reduced ability to detect Vitivirus species (except GVH) compared to the DRB4‐based method (Figure S1). This may be due to the lower propensity of these viruses to produce HMW dsRNA, to poor replication efficiency, or to suboptimal purification due to their phloem restriction (Caruso et al. 2022; Samarfard et al. 2020).

The B2‐based, DRB4‐based, and CELL methods all exhibited strong viral read percentages. In contrast, the methods using total RNA extraction (CTAB and SPEC), despite the use of ribodepletion, generally produced high total read counts but displayed lower viral read percentages, indicating that they capture a significant amount of non‐viral RNA (co‐extracted RNAs). These latter methods were less efficient in detecting low‐abundance viruses, likely due to the dominance of host RNA, further emphasising the advantages of dsRNA purification methods for virome profiling. These results align with previous studies highlighting the advantages of dsRNA‐based methods for viral detection in plants. Several studies (Gaafar et al. 2021; Roossinck et al. 2015; Rott et al. 2024) have demonstrated that dsRNA is a valuable target for virus detection, due to the fact that it is an intermediate product of viral replication. However, despite the high sensitivity and accuracy of some dsRNA extraction methods, their higher cost and tendency to capture LWM RNA contaminants are significant drawbacks. The B2‐based method addresses these issues and represents a cost‐effective alternative. This method represents a significant advancement in virus detection, combining good sensitivity and specificity with cost‐effectiveness. The minimum time required to process a single sample up to the library preparation step varied by method, with SPEC‐based (66 min) and DRB4‐based (90 min) being faster than B2‐based (117 min), CTAB‐based (138 min) and CELL‐based (145 min). However, direct comparisons should be made with caution, as some dsRNA‐based methods include cooling periods during incubation or centrifugation. While these steps extend the total processing time, they enable parallel experiments, effectively reducing the hands‐on time. Notably, the B2‐based method bettered the other methods in cost‐effectiveness, with a significantly lower cost per reaction ($4.47) compared to $35.54 for the DRB4‐based method and over $70.00 for both the SPEC and CTAB methods (Figure 8). This is particularly important for laboratories with limited resources or large‐scale viral studies, where high‐cost dsRNA extraction kits may not be feasible. The affordability of the B2‐based method makes it a viable option for virome surveillance, where numerous samples must be processed efficiently and economically. The use of the B2 method requires the expertise and availability of protein production. However, given the stability and size of the B2 protein, production is not limiting. Alternatively, B2 protein production may be outsourced to protein production platforms.

Future improvements for the B2‐based method could focus on increasing its sensitivity to detect Vitivirus species, possibly by optimising extraction conditions or combining several binding proteins together (ex. DRB4 and B2). Additionally, testing this method on a broader range of sample types and expanding its use to include other plant species and viromes will be essential in validating its generalisability. Developing a more scalable version of the B2‐based method for automated workflows could further enhance its utility in high‐throughput viral detection, ultimately contributing to more efficient viral surveillance and management strategies.

Author Contributions

Conceptualisation, M.L.F. and D.X.; methodology, D.X., M.L.F., G.C., P.M.; validation, D.X. and M.L.F.; formal analysis, M.L.F. and D.X.; investigation, M.L.F., D.X. and P.L.; resources, M.L.F.; data curation, M.L.F.; writing – original draft preparation, M.L.F.; writing – review and editing, P.L., D.X., C.R., P.M., G.C.; visualisation, M.L.F.; supervision, M.L.F.; project administration, M.L.F.; funding acquisition, M.L.F. and P.M. All authors have read and agreed to the published version of the manuscript.

Disclosure

Patent: The B2‐based dsRNA extraction method described in this manuscript is patented (WO2024086949A1). All the rights belong to the government of Canada.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1. Cloning of B2 Functional Domain, PCR Protocols and Primer Sequences used for the validation of detected viral species.

MEN-25-e14111-s002.docx (37.4KB, docx)

Figure S1. Occurrence (presence or absence) of each of the 10 viral species detected across 11 samples using five different nucleic acid extraction methods. B2 refers to the B2‐based dsRNA extraction method; CELL, the cellulose‐based method; DRB4‐method, the DRB4‐based commercial kit; CTAB, the cetyltrimethylammonium bromide‐based (CTAB) RNA extraction method; and SPEC, total RNA extraction using the Spectrum Kit. Viral species: Grapevine rupestris stem pitting‐associated virus (GRSPV), Grapevine leafroll‐associated virus 2 (GLRaV‐2), Grapevine leafroll‐associated virus 3 (GLRaV‐3), Grapevine virus H (GVH), Grapevine virus B (GVB), Grapevine virus E (GVE), Grapevine Pinot Gris virus (GPGV), Grapevine red blotch virus (GRBV), Grapevine Red Globe virus (GRGV) and Nepovirus lycopersici (Tomato ringspot virus, ToRSV).

MEN-25-e14111-s001.pdf (1.1MB, pdf)

Acknowledgements

The authors gratefully acknowledge Abdonaser Poursalavati and Joel Lafond‐Lapalme for their help and assistance in bioinformatics data processing. We are also grateful to Eric Courchesne and his team for their support, as well as their management of the experimental vineyard. In addition, we would like to express our sincere gratitude to the AAFC Office of Intellectual Property and Commercialisation team and the greenhouse team (Karine Fréchette) at the Saint‐Jean‐sur‐Richelieu Research and Development Centre for their support. Finally, we would like to acknowledge the administrative support that we received from the teams (in particular, those led by Vicky Toussaint and Mélanie Maheux) at the Saint‐Jean‐sur‐Richelieu Research and Development Centre. Special thanks to Elizabeth McFarlane, Public Services and Procurement Canada, for the English editing support of this manuscript. Open Access funding provided by the Gouvernement du Canada Agriculture et Agroalimentaire Canada library.

Handling Editor: Tatiana Giraud

Mamadou L. Fall and Dong Xu are co‐first authors.

Funding: This research was funded by Agriculture and Agri‐Food Canada under its Genomics Research and Development Initiative (GRDI) and Science Supporting an Innovative and Sustainable Sector programme: J‐002375, J‐002869, J‐0001792.

Data Availability Statement

The raw sequencing data associated with this study has been deposited at the Sequence Read Archive (SRA, SUB14844148) of NCBI of NIH, USA: Bioproject PRJNA1197386, biosamples from SAMN45803673 to SAMN45803683. For any questions, please contact the corresponding author.

References

  1. Atsumi, G. , Sekine K. T., and Kobayashi K.. 2015. “A New Method to Isolate Total dsRNA.” Methods in Molecular Biology 1236: 27–37. [DOI] [PubMed] [Google Scholar]
  2. Bolger, A. M. , Lohse M., and Usadel B.. 2014. “Trimmomatic: A Flexible Trimmer for Illumina Sequence Data.” Bioinformatics 30: 2114–2120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Cardwell, K. F. , Harmon C. L., Luster D. G., et al. 2023. “The Need and a Vision for a Diagnostic Assay Validation Network.” PhytoFrontiers 3: 9–17. [Google Scholar]
  4. Carisse, O. , Fall M. L., and Vincent C.. 2017. “Using a Biovigilance Approach for Pest and Disease Management in Quebec Vineyards.” Canadian Journal of Plant Pathology. [Google Scholar]
  5. Caruso, A. G. , Bertacca S., Ragona A., Matić S., Davino S., and Panno S.. 2022. “Evolutionary Analysis of Grapevine Virus A: Insights Into the Dispersion in Sicily (Italy).” Agriculture 12, no. 6: 835. 10.3390/agriculture12060835. [DOI] [Google Scholar]
  6. Chao, J. A. , Lee J. H., Chapados B. R., Debler E. W., Schneemann A., and Williamson J. R.. 2005. “Dual Modes of RNA‐Silencing Suppression by Flock House Virus Protein B2.” Nature Structural & Molecular Biology 12: 952–957. [DOI] [PubMed] [Google Scholar]
  7. Cobbin, J. C. , Charon J., Harvey E., Holmes E. C., and Mahar J. E.. 2021. “Current Challenges to Virus Discovery by Meta‐Transcriptomics.” Current Opinion in Virology 51: 48–55. [DOI] [PubMed] [Google Scholar]
  8. Decker, C. J. , Steiner H. R., Hoon‐Hanks L. L., et al. 2019. “dsRNA‐Seq: Identification of Viral Infection by Purifying and Sequencing dsRNA.” Viruses 11, no. 10: 943. 10.3390/v11100943. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Diaz‐Ruiz, J. R. , and Kaper J. M.. 1978. “Isolation of Viral Double‐Stranded RNAs Using a LiCl Fractionation Procedure.” Preparative Biochemistry 8: 1–17. [DOI] [PubMed] [Google Scholar]
  10. Dodds, J. A. M. , and Jordan T. J.. 1984. “Plant Viral Double‐Stranded RNA.” Annual Review of Phytopathology 22: 151–168. [Google Scholar]
  11. Fall, M. L. , Boyse J. F., Wang D., Willbur J. F., Smith D. L., and Chilvers M. I.. 2018. “Case Study of an Epidemiological Approach Dissecting Historical Soybean Sclerotinia Stem Rot Observations and Identifying Environmental Predictors of Epidemics and Yield Loss.” Phytopathology 108: 469–478. [DOI] [PubMed] [Google Scholar]
  12. Fall, M. L. , and Carisse O.. 2022. “Dynamic Simulation for Predicting Warning and Action Thresholds: A Novelty for Strawberry Powdery Mildew Management.” Agricultural and Forest Meteorology 312: 108711. [Google Scholar]
  13. Fall, M. L. , and Xu D.. 2024. Isolation and Purification of Double Stranded RNA. (Bureau WIPOI ed., vol. WO2024086949A1), 22. Canada Agriculture and Agri‐Food Canada. [Google Scholar]
  14. Fall, M. L. , Xu D., Lemoyne P., Moussa I. E. B., Beaulieu C., and Carisse O.. 2020. “A Diverse Virome of Leafroll‐Infected Grapevine Unveiled by dsRNA Sequencing.” Viruses 12, no. 10: 1142. 10.3390/v12101142. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Figueiredo Prates, L. H. , Merlau M., Ruhl‐Teichner J., Schetelig M. F., and Hacker I.. 2023. “An Optimized/Scale Up‐Ready Protocol for Extraction of Bacterially Produced dsRNA at Good Yield and Low Costs.” International Journal of Molecular Sciences 24, no. 11: 9266. 10.3390/ijms24119266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Gaafar, Y. Z. A. , Westenberg M., Botermans M., et al. 2021. “Interlaboratory Comparison Study on Ribodepleted Total RNA High‐Throughput Sequencing for Plant Virus Diagnostics and Bioinformatic Competence.” Pathogens 10: 1174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Ghaly, T. M. , Focardi A., Elbourne L. D. H., et al. 2024. “Exploring Virus‐Host‐Environment Interactions in a Chemotrophic‐Based Underground Estuary.” Environmental Microbiomes 19: 9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hillary, L. S. , Adriaenssens E. M., Jones D. L., and McDonald J. E.. 2022. “RNA‐Viromics Reveals Diverse Communities of Soil RNA Viruses With the Potential to Affect Grassland Ecosystems Across Multiple Trophic Levels.” ISME Communications 2, no. 34: 34. 10.1038/s43705-022-00110-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hull, R. 2002. “Control and Uses of Plant Viruses.” In Matthews' Plant Virology, 675–741. Academic Press. [Google Scholar]
  20. Javaran, V. J. , Poursalavati A., Lemoyne P., Ste‐Croix D. T., Moffett P., and Fall M. L.. 2023. “NanoViromics: Long‐Read Sequencing of dsRNA for Plant Virus and Viroid Rapid Detection.” Frontiers in Microbiology 14: 1192781. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Jones, R. A. 2009. “Plant Virus Emergence and Evolution: Origins, New Encounter Scenarios, Factors Driving Emergence, Effects of Changing World Conditions, and Prospects for Control.” Virus Research 141: 113–130. [DOI] [PubMed] [Google Scholar]
  22. Krinos, A. I. , Bowers R. M., Rohwer R. R., McMahon K. D., Woyke T., and Schulz F.. 2024. “Time‐Series Metagenomics Reveals Changing Protistan Ecology of a Temperate Dimictic Lake.” Microbiome 12: 133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Liang, Y. , Wang L., Wang Z., et al. 2019. “Metagenomic Analysis of the Diversity of DNA Viruses in the Surface and Deep Sea of the South China Sea.” Frontiers in Microbiology 10: 1951. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lingel, A. , Simon B., Izaurralde E., and Sattler M.. 2005. “The Structure of the Flock House Virus B2 Protein, a Viral Suppressor of RNA Interference, Shows a Novel Mode of Double‐Stranded RNA Recognition.” EMBO Reports 6: 1149–1155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lu, R. , Maduro M., Li F., et al. 2005. “Animal Virus Replication and RNAi‐Mediated Antiviral Silencing in Caenorhabditis elegans .” Nature 436: 1040–1043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Parvez, F. , Amin Z., Sangpal D., and Chugh J.. 2024. “Role of pH in Modulating RNA–Protein Interactions in TRBP2‐dsRBD2: An Interplay Between Conformational Dynamics and Electrostatic Interactions.” Journal of Physical Chemistry B 128: 12698. [DOI] [PubMed] [Google Scholar]
  27. Poursalavati, A. , Javaran V., Laforest‐Lapointe I., and Fall M.. 2023. “Soil Metatranscriptomics: An Improved RNA Extraction Method Toward Functional Analysis Using Nanopore Direct RNA Sequencing.” Phytobiomes 7: 42–54. [Google Scholar]
  28. Poursalavati, A. , Larafa A., and Fall M. L.. 2023. “dsRNA‐Based Viromics: A Novel Tool Unveiled Hidden Soil Viral Diversity and Richness.” bioRxiv. 10.1101/2023.05.10.540251. [DOI]
  29. Puente‐Sanchez, F. , Garcia‐Garcia N., and Tamames J.. 2020. “SQMtools: Automated Processing and Visual Analysis of Omics Data With R and Anvi'o.” BMC Bioinformatics 21: 358. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Roossinck, M. J. 2012. “Plant Virus Metagenomics: Biodiversity and Ecology.” Annual Review of Genetics 46: 359–369. [DOI] [PubMed] [Google Scholar]
  31. Roossinck, M. J. 2014. “Metagenomics of Plant and Fungal Viruses Reveals an Abundance of Persistent Lifestyles.” Frontiers in Microbiology 5: 767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Roossinck, M. J. , Martin D. P., and Roumagnac P.. 2015. “Plant Virus Metagenomics: Advances in Virus Discovery.” Phytopathology 105: 716–727. [DOI] [PubMed] [Google Scholar]
  33. Rott, M. , Xiang Y., Boyes I., et al. 2017. “Application of Next Generation Sequencing for Diagnostic Testing of Tree Fruit Viruses and Viroids.” Plant Disease 101: 1489–1499. [DOI] [PubMed] [Google Scholar]
  34. Rott, M. E. , Ghoshal K., Lerat S., et al. 2024. “Improving Grapevine Virus Diagnostics: Comparative Analysis of Three dsRNA Enrichment Methods for High‐Throughput Sequencing.” Journal of Virological Methods 329: 114997. [DOI] [PubMed] [Google Scholar]
  35. Samarfard, S. , McTaggart A. R., Sharman M., Bejerman N. E., and Dietzgen R. G.. 2020. “Viromes of Ten Alfalfa Plants in Australia Reveal Diverse Known Viruses and a Novel RNA Virus.” Pathogens 9: 214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Scholthof, K. B. 2007. “The Disease Triangle: Pathogens, the Environment and Society.” Nature Reviews. Microbiology 5: 152–156. [DOI] [PubMed] [Google Scholar]
  37. Wu, Q. , Ding S. W., Zhang Y., and Zhu S.. 2015. “Identification of Viruses and Viroids by Next‐Generation Sequencing and Homology‐Dependent and Homology‐Independent Algorithms.” Annual Review of Phytopathology 53: 425–444. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1. Cloning of B2 Functional Domain, PCR Protocols and Primer Sequences used for the validation of detected viral species.

MEN-25-e14111-s002.docx (37.4KB, docx)

Figure S1. Occurrence (presence or absence) of each of the 10 viral species detected across 11 samples using five different nucleic acid extraction methods. B2 refers to the B2‐based dsRNA extraction method; CELL, the cellulose‐based method; DRB4‐method, the DRB4‐based commercial kit; CTAB, the cetyltrimethylammonium bromide‐based (CTAB) RNA extraction method; and SPEC, total RNA extraction using the Spectrum Kit. Viral species: Grapevine rupestris stem pitting‐associated virus (GRSPV), Grapevine leafroll‐associated virus 2 (GLRaV‐2), Grapevine leafroll‐associated virus 3 (GLRaV‐3), Grapevine virus H (GVH), Grapevine virus B (GVB), Grapevine virus E (GVE), Grapevine Pinot Gris virus (GPGV), Grapevine red blotch virus (GRBV), Grapevine Red Globe virus (GRGV) and Nepovirus lycopersici (Tomato ringspot virus, ToRSV).

MEN-25-e14111-s001.pdf (1.1MB, pdf)

Data Availability Statement

The raw sequencing data associated with this study has been deposited at the Sequence Read Archive (SRA, SUB14844148) of NCBI of NIH, USA: Bioproject PRJNA1197386, biosamples from SAMN45803673 to SAMN45803683. For any questions, please contact the corresponding author.

Articles from Molecular Ecology Resources are provided here courtesy of Wiley

RESOURCES