Seedlings of N. benthamiana and V. unguiculata cv. Pusa Komal (cowpea) were raised in potting-mix in 10 cm pots in an environmentally controlled growth room with an average temperature of 24 ± 2°C, average humidity of 65% and 16/8 h of light and dark periods. GBNV was isolated from a tomato plant showing leaf and bud necrosis, which was confirmed by ELISA with the polyclonal antibody to GBNV and subsequently sequencing the NP gene. The virus culture was maintained in both cowpea and tomato host through sap inoculation.

A full-length NSs gene was amplified by RT-PCR from the above GBNV culture and cloned into pGEM T-Easy vector. The sequence of NSs was submitted to the GenBank with the accession number, OK105104. The NSs gene was further sub-cloned in the L4440 vector² having a T7 promoter at both ends. The recombinant vector was then used to transform the Escherichia coli strain HT115 for the production of dsRNA. To optimize the conditions for dsRNA generation, experiments with the different concentrations of IPTG (0.01, 0.1, 0.4, and 1.0 mM), different induction period after IPTG treatment (30 min, 1 h, 4 h, and overnight) and with different enzyme treatments [RNase A alone in buffer containing 300 mM sodium acetate, 10 mM Tris–HCl pH 7.5 and 5 mM EDTA (Thermo Fisher Scientific, United States) or without buffer; DNase I alone in buffer containing 10 mM Tris–HCl pH 7.5, 2.5 mM MgCl₂, 0.5 mM CaCl₂ (New England Biolabs, United Kingdom); both RNase A and DNase I together and without any enzyme treatment] were evaluated and the best combination of these factors were utilized for the production of dsRNA.

Purification of dsRNA from the bacterial culture was followed with some modification of the protocol as described by Posiri et al., 2013. Briefly, the IPTG-induced bacterial cells from 100 ml culture were harvested and used for the purification of dsRNA. The cell pellet was re-suspended in 1X PBS containing 70% v/v ethanol and incubated at 4°C for 30 min. After incubation, the cell pellets were collected by centrifugation at 6,000 × g for 10 min at 4°C. Then, the bacterial cells were re-suspended in 2.0 ml of 150 mM NaCl and incubated at 4°C for 1 h and centrifuged at 6,000 × g for 10 min at 4°C. To the supernatant, four units of RNase free DNase I and 20 μg of RNase A were added at 37°C for 30 min to remove DNA and ssRNA. The dsRNA was then precipitated by adding 500 μl of absolute ethanol. The pellet was air-dried, dissolved in deionized water, and the quantity was estimated spectrophotometrically (NanoDrop, Thermo Fisher Scientific, United States). The dsRNA preparation was stored at −20°C for further use. The quality of the purified dsRNA was judged by visualizing the expected dsRNA band on agarose gel. To examine the double stranded nature of the purified preparation, RNase treatment was performed in high and low salt concentration (Libonati and Sorrentino, 1992). Further, to confirm that the obtained band was of dsRNA from the NSs gene, nucleic acid hybridization assay (Northern/dot blot) was conducted with a DIG-labeled probe to the full-length NSs gene, which was prepared using DIG-High Prime DNA Labeling and Detection Starter Kit I (Roche, F. Hoffmann-La Roche Ltd., Switzerland).

In order to know, the entry of dsRNA into the plant system, the dsRNA aqueous solution containing 0.01% Celite 545 (BDH, England) was applied at the rate of 5.0 μg of dsRNA/plant by gentle rubbing on the adaxial surface of cowpea and N. benthamiana leaves. The excess dsRNA present on the leaf surface was removed by washing 4–5 times with distilled water. To detect the dsRNA in the different washes, each washout-water was collected separately, lyophilized to approximately 20 μl volume, and loaded in 1.5% agarose gel. The dsRNA treated leaves, after thorough and repeated washing, were used to isolate RNA by using TRIzol (Thermo Fisher Scientific)³. To understand the stability of the entered dsNSs in the plant tissues, cDNA of NSs was prepared using the total RNA isolated from the treated leaves with the reverse primer of NSs (BM1251: 5′ataagcttttactctggcttcacaatga3′) using Revert Aid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA, United States). The stability pattern of dsNSs in the plant tissue was analyzed in the local leaves (treated) of cowpea as well as both the local and systemic (non-treated) leaves of N. benthamiana by semi-quantitative reverse transcriptase PCR (semi-qRT-PCR) with the primer pair, BM1260: 5′gacagatgcagagggaaatg3′; BM1251: 5′ataagcttttactctggcttcacaatga3′ at 15, 28, and 35 PCR cycles. The PCR conditions were: initial denaturation at 95°C for 3 min, followed by each cycle of denaturation at 95°C for 30 s, annealing at 52°C for 30 s, and extension at 72°C for 30 s. The final extension at 72°C was allowed for 5 min. The GAPDH and actin genes (Reddy et al., 2016) were used as internal controls for cowpea and N. benthamiana, respectively.

The challenge-inoculation of dsNSs-treated plants with GBNV was performed as described by Rai et al., 2020. The GBNV inoculated N. benthamiana leaves showing initiation of mottling and bending symptoms were used to prepare inoculum in the extraction buffer at a ratio 1:6 (1.0 g of leaves in 6 ml of extraction buffer) in a pre-chilled mortar and pestle. The extraction buffer was composed of 0.1 M sodium phosphate buffer, pH 7.0, containing 0.15% sodium sulfite, and 0.002% beta-mercaptoethanol. The test plants were pre-dusted lightly with Celite powder and then 100 μl of the inoculum was applied gently on the surface of each leaves. The inoculated plants were sprayed with distilled water and maintained in a controlled environment room for symptom expression.

The effectiveness of dsNSs against GBNV was assayed in three different treatment combinations. (T1) dsNSs treatment followed by GBNV (dsNSs – GBNV): in this combination, 5.0 μg of dsNSs dissolved in deionized water containing 0.01% Celite was applied by gentle rubbing on the adaxial surface of leaves and 1-day post-application of dsRNA, plants were challenged with GBNV through mechanical sap inoculation by the method as described above. (T2) Succeeding application of dsNSs and GBNV (dsRNA + GBNV): in this combination, 5.0 μg of dsNSs solution containing 0.01% Celite was similarly applied on the leaves and allowed for 10–15 min and then plants were challenged by mechanical sap inoculation with GBNV. (T3) GBNV inoculation followed by dsNSs treatment (GBNV – dsNSs): in this combination, leaves were first sap inoculated with GBNV, and at 1-day post-inoculation (dpi), 5.0 μg of dsNSs solution containing 0.01% Celite was applied. For the above experiment, two fully expanded primary leaves of cowpea and two leaves of 1-month-old N. benthamiana plants were used for treating with dsRNA and GBNV inoculation. The experiment with these three treatment combinations was repeated three times on N. benthamiana and cowpea var. Pusa Komal. The experimental plants were maintained for further observations and the number of chlorotic and necrotic lesions in local and systemic leaves were counted at regular intervals in cowpea and the systemic disease progress was monitored for both N. benthamiana and cowpea.

To assess the impact of dsNSs treatment on symptom expression by GBNV, disease rating scales were developed on the basis of symptom severity grades by assigning the numerical values of 0–4 and 0–8 for N. benthamiana and cowpea, respectively (Table 1). The disease severity index (DSI) was assessed based on the disease incidence, which was calculated as [Σ (Grade assigned × number of infected plants)/Total grade × total number of plants] × 100. The relative progression of the disease till the death of the dsNSs-untreated GBNV inoculated plants was recorded and calculated using the area under the disease severity curve (AUDSC), following the standard method (Campbell and Madden, 1990; Bag et al., 2014) with the formula Y=∑i=1n-1[(Xi+Xi+1⁢)/2](t_{i+ 1}−t_i) where Y is the AUDPC, X_i is the disease incidence of the i^th evaluation and X_i+1 is the disease incidence of the subsequent evaluation; (t_i+1-t_i) is the number of days between two subsequent evaluations.

The viral load with or without dsNSs treatment was analyzed using semi qRT-PCR as well as qPCR. On the onset of disease symptoms at 9 dpi in N. benthamiana and at 6 dpi in cowpea, RNA was isolated from the fresh leaf samples. cDNA was prepared with 1.0 μg of RNA sample and random hexamers by using Revert Aid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA, United States). Semi qPCR was performed using 1.0 μl of cDNA as a template. For performing qPCR analysis, 1.0 μl of cDNA was used as template and a portion of NP gene of GBNV was amplified using specific primers (PF: 5′GACAGGTCTGGCACCAATTA3′ and PR: 5′GGCTACTTTGCAAACCTGTTC3′) with 1× KAPA SYBR^® FAST qPCR Master Mix (Roche, F. Hoffmann-La Roche Ltd., Switzerland). The actin and GAPDH genes were used as an endogenous control gene for N. benthamiana and cowpea, respectively. The relative viral load in the dsNSs-treated plants was calculated using the double C_t method (Livak and Schmittgen, 2001) comparing with the viral load as 1 unit in the untreated plants (calibrator). Three technical replicates were used in all the qPCR analysis.

For comparison of disease severity between the three combinations of dsRNA with GBNV treatments and the control (only GBNV inoculation), one-way ANOVA test was performed. Homogeneity of variance test (Levene’s statistic) was performed using SPSS package 20.0 to check the equality of variance in three datasets as the experiments were executed independently. Differences in the variance of the treatments were analyzed by Brown–Forsythe statistics.

For qPCR analysis, all the treatment combinations were evaluated as independent experiments and the data were statistically analyzed by ANOVA considering p-value < 0.05 as significant. The significance of the data within the dataset is also analyzed by performing student’s t-test at p < 0.1 and p < 0.05 in N. benthamiana and cowpea, respectively.

The clone of NSs gene contained 1,320 nucleotides and shared up to 97.5% sequence identity with that of other isolates of GBNV available in the GenBank. The clone carrying dsNSs in L4440 vector was subjected to the various parameters for the expression of dsRNA in the bacterial culture, which revealed incubation of bacterial culture with 0.4 mM IPTG at 37°C for 4 h induced maximum expression of the expected approximately 1.46 kb dsNSs. Purification of dsRNA from bacteria yielded approximately 1.5–2.0 mg of dsNSs per liter of bacterial culture. The dsRNA from the bacterial cells when was treated with RNase A in high salt, the approximately 1.46 kb band was not digested; however, it was digested with RNase A in low salt condition indicating double stranded nature of 1.46 kb band (Figure 1A). Further, the approximately 1.46 kb band was of NSs gene origin was confirmed through both Northern blot as well as dot blot analyses, which showed strong hybridization signal with the DIG-labeled probe to the NSs clone (Figure 1B).

To know the entry of dsRNA in the leaf tissues, the excess dsRNA present on the leaf surface was removed by repeated washing. It was found that after 4–5 washes, no dsRNA on the leaf surface could be detected in the agarose gel electrophoresis. Therefore, while detecting dsRNA in the plant tissue by semi qRT-PCR, the treated leaves were thoroughly washed five times before isolating the total RNA from the plant tissues. In N. benthamiana plants, dsNSs was tested by semi qRT-PCR with the NSs primers in the treated as well as in the newly developed leaves at 1, 3, and 7 days post-treatment. At the 15th cycle of semi qPCR, no detection of NSs gene was obtained; however, at 28th cycle of semi qRT-PCR, 118 nucleotide fragment of NSs was detected at all the three-time points in the treated leaves, interestingly, and NSs could be detected in the systemic leaves at 7 days post-treatment (Figure 2B). In cowpea, NSs was detected at the site of treatment till 5 days post-treatment but not at 6 or 7 days post-treatment at 28 cycles of semi qRT-PCR (Figure 3B).

The effectiveness of the topically applied dsNSs of GBNV for the prevention of GBNV infection was tested in 30 days old N. benthamiana plants containing 4–5 well-developed leaves (n = 4 plants in each of the three experiments). The dsNSs non-treated control plants of N benthamiana inoculated with GBNV showed no local lesion on the inoculated site; however, chlorotic patches were observed in these leaves at 9 dpi. Subsequently, systemic symptoms like veinal chlorosis, necrosis of newly developed leaves, and bending of the shoot with wilting of plants from the upper plant part were observed. Finally, the infected plant collapsed and died. To estimate the DSI each of the sequentially advancing symptoms like no symptom, chlorotic patches, bending of the shoot, wilting of leaves, and death of plants were graded as 0, 1, 2, 3, and 4, respectively (Table 1). The GBNV inoculated plants, where dsRNA was not applied, reached the severe disease grade of 3–4 by 9–16 days. Whereas the treatment combination-T2 (dsNSs + GBNV: where the dsRNA application was followed by the virus inoculation immediately), and the treatment combination-T3 (GBNV – dsNSs: where the dsRNA was applied at 1.0 dpi) showed a low disease severity grade of 0–2. The treatment combination-T1: dsNSs – GBNV, where the dsRNA was applied a day before GBNV inoculation, showed higher disease severity compared to that in the treatment combination-T2 and -T3 (Figure 2A). The DSI in the dsRNA-treated plants ranged between 50 and 66.7 indicating a reduction in the disease severity of about 43.3–50% in the dsRNA-treated plants compared to the non-treated controls. The AUDSC showed significantly less AUDSC values (669–924) in the treatment combination-T2 and -T3, where dsRNA treatment was conducted just before GBNV inoculation and a day post-GBNV inoculation, respectively, compared to the non-treated control which showed AUDSC value of 1378 (Table 2). Disease severity in GBNV treatment was significantly higher than the other three treatments (p < 0.05) indicated other three treatments were effective against viral infection. Although, homogeneity of variance test was significant (p < 0.05), but the differences in the treatments based on Brown–Forsythe statistics as the assumption of equal variances has been violated was found to be significant indicated differences really existed among the treatments.

In semi qRT-PCR, of all the three combinations, the combination-T2 and -T3 showed a marked reduction of NP gene amplification compared to the control or the combination-T1 (Figure 2C). The qRT-PCR results with NP gene primers showed a significant reduction in the viral load in plants treated with dsRNA compared to the non-treated plants (Figure 2D). The maximum reduction of viral load (12.5 fold) was found in the treatment combination-T2: dsNSs + GBNV followed by the treatment combination-T3 and i, which were 3.57 and 1.81 fold, respectively (Figure 2D). In the treatment combination-T2, which showed the maximum reduction of the viral load, the plants survived for a longer period of time compared to the non-treated ones (Figure 4A). In the three independent experiments of three application combinations of dsRNA, plants survived upto 23–25 dpi, indicating an 8–10 days delay in the death of the plants due to GBNV infection.

The efficacy of dsNSs was further examined in cowpea cv. Pusa Komal plants (n = 4, 10, and 3 plants in the three experiments). The primary leaves of cowpea cv. Pusa Komal inoculated with GBNV exhibited the chlorotic lesions at 4–6 dpi followed by necrotic lesions (Figure 3A), which coalesced, the inoculated leaves wilted and fell by 8–12 dpi. Subsequently, the plants developed systemic symptoms like necrotic lesions on newly developed leaves and necrosis on the petiole and stem. Eventually, the plants collapsed and died by 14–16 dpi. To determine the disease severity index, the level of the disease that is no-symptom to local necrosis and subsequent systemic necrosis followed by the death of plants were graded from 0 to 8 point scale (Table 1). The three treatment combinations of dsNSs and challenge with GBNV as carried out for N. benthamiana were also examined for cowpea. The dsNSs application on the two primary leaves of each seedling (5.0 μg/plant) in all the three treatments showed a significant reduction in the number of local lesions in cowpea seedlings (Figure 3D). The treatment combination-T2: dsNSs + GBNV showed a significant reduction of local lesions (1–2/seedling) compared to the non-treated plants (15–25 lesions/plant) (Figure 3A). The treatment combination T2 and T3 also resulted in a significant reduction of the number of lesions (3–5 lesions/plant) (Figures 3A,D). The dsRNA treatment combinations showed a significant effect on the reduction of local symptoms, which had a great implication on the restriction of in planta spread of the virus and survival of plant as when the non-treated plants died by 15–16 dpi due to the combined effect of local and systemic disease pressure, the dsRNA-treated plant survived with the development of new leaves (Figure 4B). The DSI, as well as AUDSC at 6 dpi in cowpea, were significantly low (DSI: 9.7–84.4 versus 100; AUDSC: 86.6–427.4 versus 544.1) in all the three combinations of dsRNA treatment as compared to the non-treated control (Table 2). The least DSI and AUDSC were obtained from the treatment combination-T2: dsNSs + GBNV. The presence of GBNV was detected by RT-PCR in all the dsRNA treatment combinations following the virus inoculation; however, the intensity of amplification was relatively less compared to the control especially in the case of the treatment combination T2 and T3 (Figure 3C). The GBNV load in the dsRNA-treated plants as judged by qRT-PCR was also significantly less compared to the non-treated control plants (Figure 3E).