Topical Spray of dsRNA Induces Mortality and Inhibits Chilli Leaf Curl Virus Transmission by Bemisia tabaci Asia II 1

Chilli leaf curl virus (ChiLCV; genus: Begomovirus), transmitted by Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) in a persistent-circulative manner, is a major constraint in chilli production. The present study demonstrates for the first time that a topical spray of naked double-stranded RNA (dsRNA) on chilli plants causes mortality and inability to acquire and transmit ChiLCV in B. tabaci. dsRNA targeting heat shock protein 70 (hsp70) and fasciclin 2 (fas2) of B. tabaci Asia II 1 was first assessed under controlled conditions through oral delivery. Hsp70 and fas2 dsRNA resulted in up to 82.22% and 72% mortality of B. tabaci and around 12.4- and 8.5-fold decreases in mRNA levels, respectively, 24 h post-ingestion. ChiLCV copies in hsp70 dsRNA-fed B. tabaci steadily decreased with an increase in dsRNA concentration and were undetectable at a higher concentration of dsRNA. However, ChiLCV copies significantly increased in fas2 dsRNA-fed B. tabaci. Transmission of ChiLCV by B. tabaci was completely inhibited post-24 h feeding on hsp70 dsRNA at 3 μg/mL. Naked hsp70 dsRNA was topically sprayed on ChiLCV-infected chilli plants like an insecticide. 67.77% mortality of B. tabaci, 4.6-fold downregulation of hsp70 mRNA, and 1.34 × 1015-fold decreased ChiLCV copies in B. tabaci were recorded when adults were exposed to the dsRNA-treated plants under semi-field conditions. Foliar application of naked dsRNA reduced the ChiLCV transmission by 75% without any visible symptoms in the inoculated plants. A total of 2 consecutive sprays of dsRNA provided significant protection to B. tabaci for up to 20 days under semi-field conditions.


Introduction
Silverleaf whitefly [Bemisia tabaci (Gennadius), Hemiptera: Aleyrodidae] is a phloemfeeding hemipteran insect that has been reported to infest over 600 plant species worldwide [1][2][3]. B. tabaci has been considered a complex of at least 46 morphologically indistinguishable cryptic species [4][5][6]. B. tabaci adults are about 1 mm long with small triangular-shaped bodies. The body is yellowish with distinct hyaline wings dusted with white powdery wax. Adults B. tabaci can be distinguished from other whiteflies by the position in which the wings are held over the body. The wings are held close to the body and tent-like in B. tabaci. It sucks phloem sap, leaving the affected plants extremely weak, and secretes a honeydew onto the surface of the leaves that promotes the growth of sooty mold fungus [7]. Besides direct damage caused by feeding, B. tabaci transmits more than a hundred begomoviruses, carlaviruses, criniviruses, cytorhabdoviruses, ipomoviruses, poleroviruses, and torradoviruses ( [8][9][10][11][12][13][14][15][16]. Among them, begomoviruses represent 90% of the viruses transmitted by B. tabaci. Begomovirus species have become widespread in Central America, the Dominican Republic, Israel, Mexico, Trinidad, and across South East Asia, including Cambodia, Indonesia, India, and Thailand [17]. These viruses can cause an estimated yield loss of 50-90% in tomatoes and other crops, including beans, cassava, chilli, cotton, cucurbits, eggplant, papaya, and potatoes [18,19].

Whitefly Population
A homogeneous isofemale line of B. tabaci was raised from a single adult female and has been maintained on eggplants (var. Navkiran, Mahyco) at the Advanced Centre for Plant Virology, Indian Agricultural Research Institute (IARI), New Delhi, since 2015. The population was characterized by sequencing the mitochondrial cytochrome oxidase subunit I (mtCOI) gene ( Table 1). The genotype or cryptic species of the B. tabaci population was confirmed based on Bayesian Inference phylogeny, considering a genetic divergence cutoff of 4% as described by Rehman et al. [6]. The population was maintained under controlled environmental conditions at 28 ± 2 • C, 60 ± 10% relative humidity, and a 16 h light-8 h dark photoperiod.

Virus Culture
The inoculum of ChiLCV was collected from a pure culture maintained at the Advanced Centre for Plant Virology, IARI, New Delhi, India. ChiLCV was maintained in chilli (var. Preeti, Nunhems) by B. tabaci inoculation under insect-proof conditions. The identity of the virus was further confirmed by amplifying the DNA-A component in PCR with primer pairs Begomo F and Begomo R [44] (Table 1) and sequencing.
Primers targeting the dsRNA segment were designed in NCBI primer blast (https: //www.ncbi.nlm.nih.gov/tools/primer-blast/ (Accessed on 4 June 2021) ( Table 1). The primers AG137F-AG138R (for hsp70) and AG283F-AG284R (for fas2) were validated in a gradient PCR. The dsRNA stretches were amplified from isofemale B. tabaci Asia II 1 DNA in PCR with initial denaturation at 94 • C for 5 min, 30 cycles of 94 • C for 30 s, 56 • C for 30 s, and 72 • C for 30 s and a final extension step at 72 • C for 10 min. The PCR products were cloned into an L4440 plasmid vector between two T7 promoters. The recombinant plasmids were transformed into RNase III mutant Escherichia coli HT115 cells. Total RNA from the recombinant E. coli HT115 cells was extracted using Trizol reagent (Invitrogen, CA, USA) following the manufacturer's protocol, and dsRNA was purified following Ahn et al. [45] with modifications. In brief, the recombinant E. coli HT115 cells were incubated for 12 h with continuous shaking at 37 • C followed by induction of T7 promoter with 1M isopropyl-β-D-1-thiogalactopyranoside (IPTG). The bacterial cells were harvested in a 1.5 mL microcentrifuge tube by centrifugation at 10,000× g for 5 min and resuspended in 1 mL Trizol. The cell suspension was mixed by vortexing and kept at room temperature for 5 min. The cell suspension was mixed with 200 µL chloroform, vortexed for <10 s, and incubated at room temperature for 10 min. The mixture was then centrifuged at 16,000× g for 10 min at 4 • C. The upper clear aqueous phase was transferred to a fresh tube and 0.8 volume of ice-chilled isopropanol was added. The solution was mixed properly and incubated at 4 • C for 10 min. The mixture was again centrifuged at 16,000× g for 10 min at 4 • C, and the supernatant was discarded. The pellet was washed with 70% ethanol, air-dried, and resuspended in 30 µL nuclease-free water. Total RNA isolated from recombinant E. coli HT115 cells was mixed with 1X RNA loading dye (Thermo Fisher Scientific, MA, USA), heated at 70 • C for 5 min, and visualized on 2% agarose gel stained with GoodView (BR Biochem, New Delhi, India). The total RNA was then treated with 1 unit of DNase I, RNase-free (Thermo Fisher Scientific) and 1 unit of RNase A, DNase-, and protease-free (Thermo Fisher Scientific) in the presence of 500 mM sodium chloride and incubated for 1 hr at 37 • C to eliminate the DNA and single-stranded RNA contaminants. The enzymes were inactivated by chloroform extraction, and the remaining dsRNA was resuspended in nuclease-free water. The purified dsRNA was quantified in a spectrophotometer (NanoDrop 2000, Thermo Fisher Scientific), and the integrity of the specific dsRNA was confirmed on 2% native agarose gel stained with GoodView and visualized in a gel documentation system (Maestrogen, Xiangshan District, Taiwan). The purified hsp70 and fas2 dsRNA were individually and orally delivered to B. tabaci adults following the method of Upadhyay et al. [40] with modifications. In brief, an artificial diet was prepared by mixing 20% sucrose and 5% yeast extract in sterile distilled water and autoclaved. The artificial diet was supplemented with 1.0, 2.0, and 3.0 µg/mL of dsRNA and sandwiched between 2 layers of UV-sterilized stretched Parafilm M membranes on the open mouth of a 15 mL cylindrical pet bottle (3.5 cm diameter and 16 cm height). A hole was made on the wall of the bottle and covered with muslin cloth for ventilation. A batch of 30 adults of B. tabaci was released in each bottle to feed on the dsRNA mixed with the artificial diet for 24 h. Sterile distilled water instead of dsRNA was served as control. After 24 h of feeding, percent mortality was calculated and compared with untreated control. Three replicates for each concentration were maintained and repeated twice. The mean mortalities in each treatment were corrected by normalizing mortality in the control set. Mean differences among the categories were separated by Tukey's test at a confidence interval of 95% using XLSTAT 2014.5.03. The surviving B. tabaci from several such replicates were used for determining the relative expression of hsp70 and fas2 mRNA and assessing the ChiLCV acquisition and transmission efficiency post-dsRNA feeding as described below.

Estimating Relative Expression of hsp70 and fas2 mRNA
Relative expression of B. tabaci hsp70 mRNA post-dsRNA exposure was estimated by RT-qPCR assay following the 2 −∆∆C T method [48]. The β-actin gene served as endogenous control. The primer pairs (AG204F-AG205R for hsp70; AG283F-AG284R for fas2; and AG177F-AG178R for β-actin) used in RT-qPCR are listed in Table 1. After 24 h of has70 and fas2 dsRNA feeding at 1.0, 2.0, and 3.0 µg/mL as described above, the surviving B. tabaci were collected from each of the treatments separately. About 30 surviving B. tabaci in 3 replicates were used for each of the treatments and doses for estimating the relative expression of the target genes. The B. tabaci adults were crushed in 1 mL Trizol reagent within microcentrifuge tubes using a hand automizer. Total RNA was isolated as described earlier. Total RNA was quantified in a spectrophotometer (NanoDrop 2000, Thermo Fisher Scientific), and complementary DNA was synthesized using the FIREScript RT cDNA synthesis kit (Solis BioDyne, Tartu, Estonia) with 1.0 µg template RNA for each set. The reaction mixture contained 1X RT reaction buffer, 1.0 µg template RNA, 5.0 µM oligo dT primer, 500 µM dNTP mix, 10 units of FIREScript RT, and 1 unit of RiboGrip RNase inhibitor. The reverse transcription was carried out in a thermocycler (T100, Bio-Rad, CA, USA) at 42 • C for 60 min, followed by enzyme inactivation at 85 • C for 5 min. The relative RT-qPCR assay was carried out in an Insta Q48M real-time PCR (Himedia, Mumbai, India) with 20 µL reaction mixture containing 10 µL of 1X Maxima SYBR green master mix, 10 µM ROX passive reference dye, 10 pmole each forward and reverse primer (AG204F-AG205R for hsp70; AG283F-AG284R for fas2; and AG177F-AG178R for β-actin), and 2 µL template cDNA. Thermal cycling was performed as initial denaturation at 94 • C for 5 min, 30 cycles of 94 • C for 30 sec, 56 • C for 30 s, and 72 • C for 30 s. Since SYBR Green I dye binds non-specifically to any double-stranded DNA, a dissociation or melting stage was carried out after every reaction to determine the specificity of the amplicons based on the melting curve. The RT-qPCR was performed with three biological and two technical replicates. The fold change in expression was normalized by excluding the changes in cycle threshold (C T ) value of the endogenous control, β-actin. Log 2 fold change value was calculated, and relative expression of mRNA was determined by normalizing the log 2 2 −∆∆C T values of the dsRNA-treated samples with untreated control [48]. Statistical analysis and preparation of graphs were carried out in Microsoft Excel 2016.

Quantification of Virus Copies in B. tabaci and Transmission of ChiLCV
A portion of B. tabaci that survived post-feeding of dsRNA at 1.0, 2.0, and 3.0 µg/mL as described above was allowed to feed on ChiLCV-infected chilli plants (var. Preeti) for 24 h. About 50 surviving B. tabaci in 3 replicates were used for each of the treatments and doses. ChiLCV copies acquired after 24 h feeding by dsRNA-exposed B. tabaci were estimated by absolute quantification in qPCR. A standard curve of ChiLCV was prepared using a clone of partial ChiLCV CP gene in a pJET1.0 vector (Thermo Fisher Scientific). Then, 10-fold serial dilutions (5 × 10 2 to 5 × 10 −5 ng) of linearized plasmid were amplified in qPCR as described below. The standard curve was prepared by plotting a linear regression curve with log 10 DNA dilutions on the X-axis and C T values on the Y-axis. Three replicates of each dilution were used in preparing the standard curve. The statistical analysis and preparation of graphs were carried out in Microsoft Excel 2016.
After 24 h of acquisition, DNA was isolated using CTAB extraction buffer [49] from B. tabaci (30 per replicate) exposed to different doses (1.0, 2.0, and 3.0 µg/mL) of hsp70 and fas2 dsRNA separately. Briefly, B. tabaci adults were crushed in 1 mL CTAB extraction buffer (2% CTAB, 100 mM Tris-HCl pH 8.0, 20 mM EDTA pH 8.0, 1.4 M NaCl, and 2 µL β-mercaptoethanol). The homogenate was heated in a dry bath at 65 • C for 30 min, and an equal volume of chloroform-isoamyl alcohol (24:1) was added to it. The mixture was vortexed for <10 s, incubated at room temperature, and centrifuged at 16,000× g for 15 min. The upper aqueous phase was transferred to a fresh microcentrifuge tube, and an equal volume of isopropanol was added to precipitate the DNA. The solution was then centrifuged at 16,000× g for 15 min to pellet down the DNA and washed with 70% ethanol. The pellet was air-dried and dissolved in 30 µL sterile distilled water followed by qPCR with the ChiLCV-specific primers AG149F and AG150R [47] ( Table 1). The qPCR was carried out in an Insta Q48M real-time PCR with a 20 µL reaction mixture as described above. Thermal cycling was performed as initial denaturation at 94 • C for 5 min, 30 cycles of 94 • C for 30 s, 53 • C for 30 s, and 72 • C for 30 s. Melting curve analysis was carried out after each reaction to check the specificity of the amplicons. Each treatment was comprised of three biological and two technical replicates.
The mean C T values obtained in qPCR were fitted into the standard curve, and the resulting concentration was used for the calculation of virus copy number in Microsoft Excel 2016 using the following formula. Mean differences among the mean virus copies were separated by Tukey's test at a confidence interval of 95% using XLSTAT 2014.5.03.

Virus copy number
where N = number of viral copies, x = amount of amplicon in ng, and n = length of linearized plasmid DNA. For the ChiLCV transmission experiment, B. tabaci surviving different doses (1.0, 2.0, and 3.0 µg/mL) of hsp70 and fas2 dsRNA exposure were allowed to feed on ChiLCVinfected chilli plants for 24 h. After acquisition feeding, they were released onto healthy chilli plants (var. Preeti) at the 4-6 leaf stage for 24 h of inoculation feeding and eliminated thereafter. Two adult females per plant were released. All the plants were maintained under insect-proof conditions and monitored for symptom development. ChiLCV infection in inoculated plants was confirmed by PCR 35 days post-inoculation. The transmission efficiency was calculated as the percent of plants infected by ChiLCV post-24 h inoculation feeding by dsRNA-treated B. tabaci. B. tabaci without any exposure to dsRNA was used as a control. For each treatment, three biological replicates were used, and each replicate contained five plants. Tukey's test was used to test significant differences in transmission efficiency. p-values less than 0.05 were considered statistically significant.

Topical Spray of hsp70 dsRNA under Contained Semi-Field Conditions
A 10-fold higher dose of dsRNA (30 µg/mL) than the controlled condition assay was used to test its efficacy in a contained experiment under semi-field conditions where plants were covered by insect-proof nets and kept in the open air. The dsRNA was not artificially fed or applied directly on B. tabaci. Instead, naked hsp70 dsRNA was topically sprayed on the ChiLCV-infected chilli plants (var. Preeti) like an insecticide in an insect-proof net house. Sterile distilled water was sprayed as a control. The plants were allowed to air-dry, and 100 virus-free B. tabaci adults were released on each dsRNA-treated ChiLCV-infected plant after 24 h of dsRNA spray. A total of 3 biological replicates, each containing 10 plants, were maintained. Percent mortality of B. tabaci 24 h after release on dsRNA-treated plants was calculated by normalizing the mean mortality in the control set. About 30 surviving B. tabaci in 3 replicates were used for determining the hsp70 mRNA levels in RT-qPCR as described above. The other portion of surviving B. tabaci was used for transmission of ChiLCV and was released onto 4-6 leaf stage healthy chilli plants (5 adults/plant) for 24 h of inoculation feeding. All the plants were maintained in insect-proof conditions and tested in PCR for ChiLCV-infection 35 days post-inoculation. A total of 5 biological replicates containing 12 plants each were used, and transmission efficiency was calculated as the percent of plants infected by ChiLCV. Tukey's test was performed to determine the significant differences in mean mortality and transmission efficiency among the categories at a confidence interval of 95%.

Stability of hsp70 dsRNA in Leaf Tissue
To assess the stability of the dsRNA on the topically sprayed plants, apical leaf tissue was collected at 1, 3, 6, 24, and 48 h post-dsRNA spray in 3 replicates. Plants sprayed with sterile distilled water served as control. Total RNA was isolated from 100 mg of plant tissues using Trizol reagent as described above. cDNA was synthesized using hsp70 dsRNA-specific primer (AG137F and AG138R), and the presence of hsp70 dsRNA was checked in PCR as described above. The experiment was repeated twice.

Persistent Efficacy of hsp70 dsRNA to Eradicate B. tabaci
The persistency in eradicating B. tabaci population by topical spray of hsp70 dsRNA was evaluated in a contained experiment under semi-field conditions. Chilli plants were topically sprayed with hsp70 dsRNA at 30 µg/mL. B. tabaci were released on the dsRNAtreated plants after 24 h of dsRNA spray. At every 24 h, old B. tabaci adults were manually eliminated, and fresh adults were released to check the persistency in the efficacy of dsRNA. A total of 30 adults per plant were released each and every time, and percent mortality was recorded at 24 h post each release. This was continued until the mortality percentage decreased significantly. Upon significant decrease in mortality after the first spray, the second application of dsRNA was undertaken. The mortality percentage was recorded at 24 h intervals after the second spray, as described above. Sterile distilled water was sprayed in place of dsRNA as untreated control. Three biological replicates, each containing five plants, were used to calculate the percent mortality.

Characterization of B. tabaci and Begomovirus
A homogeneous population of B. tabaci developed from a single adult female was characterized by sequencing the mtCOI gene. PCR amplification of B. tabaci mtCOI with C1-J-2195 and L2-N-3014 primers (Table 1) showed an expected amplicon of~860 bp on an agarose gel. The nucleotide (nt) sequence showed 99.99% homology in BLASTn analysis with other B. tabaci Asia II 1 sequences in NCBI. The sequence can be retrieved using the GenBank Accession No. MT920041. Bayesian inference phylogeny considering genetic divergence cutoff of 4% revealed that the population belonged to the cryptic species B. tabaci Asia II 1 (data not presented).
The identity of the virus was confirmed by sequencing the DNA-A component from infected chilli plants. PCR amplification of the DNA-A using Begomo F and Begomo R primer produced a 2.7 kb product visualized on 1% agarose gel. Bidirectional sequencing of the cloned products produced a 2763 nt sequence comprising complete DNA-A that showed 100% nucleotide identity to ChiLCV isolates upon BLASTn analysis. The sequence can be retrieved by GenBank Accession No. OM513903.

Effect of dsRNA on Mortality and mRNA Expression of B. tabaci under Controlled Conditions
The efficacy of hsp70 and fas2 dsRNA was first evaluated under controlled laboratory conditions. Significantly higher mortality of B. tabaci adults was recorded in hsp70 dsRNA than fas2 dsRNA when orally delivered under controlled laboratory conditions (Figure 2a). Mortality of 67.77% of B. tabaci was observed at 24 h upon feeding on 1.0 µg/mL hsp70 dsRNA. The mortality of B. tabaci significantly increased with an increase in dsRNA concentration. The mean mortality was 70.0% when hsp70 dsRNA was fed to B. tabaci adults at 2.0 µg/mL, and it increased significantly up to 82.22% at 3.0 µg/mL hsp70 dsRNA (Figure 2b). Oral delivery of fas2 dsRNA at 1.0 µg/mL exhibited 39.33% mean mortality after 24 h (Figure 2b). Significantly higher mortality (62 and 72%) of B. tabaci adults was recorded when 2.0 and 3.0 µg/mL fas2 dsRNA was orally delivered to B. tabaci under controlled laboratory conditions. At all doses, the mortality of B. tabaci post-fas2 dsRNA exposure was significantly lower than the hsp70 dsRNA at p < 0.05. However, no morpho-deformities were observed in killed B. tabaci adults either with hsp70 or fas2 dsRNA.
RT-qPCR analysis showed that hsp70 dsRNA significantly reduced the hsp70 mRNA expression levels in treated B. tabaci. Hsp70 mRNA levels in adult B. tabaci were decreased by 5.43-fold with respect to an endogenous control gene, β-actin, post-24 h feeding on hsp70 dsRNA (Figure 2c). The mRNA expression levels were significantly reduced with an increase in hsp70 dsRNA concentration. An exposure to hsp70 dsRNA at 2.0 µg/mL for 24 h significantly reduced the hsp70 mRNA expression by 11.32-fold and up to 12.85-fold when fed with 3.0 µg/mL hsp70 dsRNA. The fas2 mRNA expression levels were declined by 5.48-fold post-24 h feeding on fas2 dsRNA at 1.0 µg/mL. The mRNA expression levels post-hsp70 and -fas2 dsRNA feeding were almost similar at 1.0 µg/mL. The mRNA expression levels were significantly reduced up to 8.53-fold with an increase in fas2 dsRNA concentration. The reduction in mRNA expression level was significantly higher post-hsp70 dsRNA feeding compared to fas2 dsRNA at 2.0 and 3.0 µg/mL. The primer pairs for hsp70, fas2, and β-actin did not produce any secondary peaks in the RT-qPCR melting curve analysis that indicated the specificity of the reactions (Figure 2d,e).

Effect of dsRNA on Virus Acquisition and Transmission by B. tabaci under Controlled Conditions
After 24 h of dsRNA feeding, the adult B. tabaci were allowed to feed on ChiLCV-infected plants for 24 h. A standard curve of ChiLCV was prepared in qPCR using ten-fold dilutions of a clone of the partial ChiLCV CP gene. The standard curve of ChiLCV showed a coefficient of correlation (R 2 ) of 0.995 and high amplification efficiency near to 100%, indicating optimal conditions for absolute quantification (Figure 3a). The virus copies acquired by the B. tabaci were quantified by fitting the mean C T values obtained in qPCR into the standard curve. The specificity of the reactions was confirmed by a single peak at 81 • C in the melting curve analysis of qPCR (Figure 3b). The virus copy number acquired by hsp70 dsRNA-treated B. tabaci was significantly less than the untreated control. When B. tabaci was fed with sterile distilled water instead of hsp70 dsRNA, the mean ChiLCV copy number was 2.22 × 10 15 , whereas it decreased by 35.18-fold (6.31 × 10 13 copies) post-exposure to hsp70 dsRNA at 1.0 µg/mL (Figure 3c). The virus copy number in treated B. tabaci further decreased with a gradual increase in hsp70 dsRNA concentration. At 2.0 µg/mL hsp70 dsRNA, ChiLCV copy numbers were 6.14 × 10 10 in B. tabaci i.e., 3.62 × 10 4 -fold lower than the untreated control. The virus copies were undetectable in B. tabaci with a further increase in hsp70 dsRNA concentration to 3.0 µg/mL. The retention of ChiLCV might be completely ceased at a higher concentration of hsp70 dsRNA or beyond the detection limit using qPCR. In contrast, the ChiLCV copy number acquired by fas2 dsRNA-treated B. tabaci was significantly higher than the untreated control ( Figure 3d). When B. tabaci was fed with sterile distilled water instead of fas2 dsRNA, the mean ChiLCV copy number was 3.19 × 10 11 , whereas the virus copies increased significantly by 139.18-fold (4.44 × 10 13 copies) at 1.0 µg/mL of fas2 dsRNA feeding. However, the mean virus copies were estimated to be 3.02 × 10 13 and 2.14 × 10 13 after exposure to 2.0 and 3.0 µg/mL fas2 dsRNA, respectively. Although significantly higher numbers of virus copies were recorded in all fas2 dsRNA doses compared to the control, higher doses of fas2 dsRNA recorded a decline in virus copies compared to the lower doses. As the virus copies in B. tabaci increased post-fas2 dsRNA feeding, it was not assessed further for inhibiting transmission of ChiLCV and topical application.
A similar pattern was recorded in the transmission of ChiLCV from infected to healthy chilli plants by hsp70 dsRNA-fed B. tabaci. A 100% transmission of ChiLCV was recorded when B. tabaci was not exposed to hsp70 dsRNA in the control set. All the inoculated plants produced strong leaf curl and vein-clearing symptoms and tested ChiLCV positive in PCR. Although there was a significant decrease in ChiLCV copies in B. tabaci exposed to 1.0 µg/mL hsp70 dsRNA, all the test plants were found to be infected with ChiLCV in PCR testing at 35 days post-inoculation. Probably even a lower concentration of ChiLCV in B. tabaci was sufficient to make 100% transmission. However, the severity of the leaf curl symptoms in inoculated plants was less than the untreated control. The transmission of ChiLCV was decreased to 40% when B. tabaci were fed with 2.0 µg/mL hsp70 dsRNA for 24 h (Figure 3e). The inoculated plants that tested positive in PCR produced mild leaf curl symptoms. Interestingly, there was no transmission of ChiLCV when B. tabaci was exposed to 3.0 µg/mL hsp70 dsRNA. All the inoculated plants were without any leaf curl symptoms ( Figure 3f) and tested ChiLCV-free in PCR at 35 days post-inoculation. Probably virus transmission by B. tabaci completely inhibited at a higher concentration of hsp70 dsRNA.

Effect of Topical Spray of hsp70 dsRNA on Mortality and Virus Transmission by B. tabaci under Semi-Field Conditions
In a contained experiment, the efficacy of topically sprayed naked hsp70 dsRNA was tested under semi-field atmospheric conditions. The daily temperature fluctuated between 20 • and 35 • C with 40-60% RH. dsRNA was not directly delivered to B. tabaci. Instead, it was sprayed on ChiLCV-infected chilli plants, and B. tabaci adults were released on the sprayed plants. Considering the exogenous degradation of dsRNA under natural atmospheric conditions, a ten-fold higher dose of dsRNA (30 µg/mL) than the controlled condition assay was used. A 67.77% mortality of adult B. tabaci was recorded post-24 h exposure to dsRNA-treated plants (Figure 4a). A 4.6-fold decrease in hsp70 mRNA expression levels of B. tabaci exposed to dsRNA-treated plants was also noted in RT-qPCR with respect to the endogenous control gene, β-actin (Figure 4b). Specific peaks of hsp70 and β-actin in the melting curve analysis without any secondary peaks confirmed the specificity of the reactions (Figure 4c).
The surviving B. tabaci adults post-exposure to dsRNA-treated plants were further assessed for ChiLCV acquisition and transmission efficiency. A 1.84 × 10 8 -fold decrease in ChiLCV copies was observed in B. tabaci post-24 h exposure to ChiLCV-infected plants that were sprayed with hsp70 dsRNA at 30 µg/mL (Figure 4d). The melt curve of ChiLCV amplicon in qPCR produced specific peaks without any secondary peaks that confirmed the specificity of the reactions (Figure 4e). When a portion of the same B. tabaci population was used for inoculation of ChiLCV in healthy chilli plants, a 75% decrease in transmission efficiency was confirmed in PCR test (Figure 4f) with no visible symptoms in the inoculated plants up to 50 days post-inoculation (Figure 4g). Probably, the virus titer was too low in inoculated plants to produce strong visible symptoms, whereas 100% transmission and severe leaf curl symptoms were noted in plants inoculated by unexposed B. tabaci. No transmission of ChiLCV was recorded in the plants mock-inoculated with untreated virus-free B. tabaci.

Stability and Persistent Efficacy of hsp70 dsRNA in Eradicating B. tabaci Population
The stability of hsp70 dsRNA in leaf tissue was assessed through RT-PCR using hsp70 specific primers. A specific amplicon of~130 bp was produced from the leaf tissues collected 1, 3, and 6 h post-spray of hsp70 dsRNA (Figure 5a).
To check the persistency in eradicating B. tabaci population by the topical spray of hsp70 dsRNA under contained semi-field conditions, fresh B. tabaci adults were released on the dsRNA-treated chilli plants at 24 h intervals. After the first topical spray, mortality of 58.89% was recorded at 2 days. Significant mortality (56.67%) persisted up to 8 days after the first spray (Figure 5b). The mortality decreased to 13.33% at 10 days after the first spray. A second spray was done on the 10th day to check the effect of multiple sprays. After the second topical spray of dsRNA, the mortality percentage again rose to 75% at two days. The highest mortality (86.67%) of B. tabaci was recorded 4 days after the second spray. Significantly higher mortality was sustained up to 8 days after the second application and thereafter started decreasing. Overall, 2 consecutive sprays provided substantial protection against fresh releases of B. tabaci for up to 20 days.

Discussion
Several management strategies have been adopted around the world, with insecticides as the core component for the management of the B. tabaci-ChiLCV complex. However, the use of chemical insecticides has led to several detrimental effects like the development of insecticide-resistant B. tabaci population, destruction of beneficial organisms, environmental contamination, and health hazards [50][51][52]. There is a need for an alternative eco-friendly approach in managing B. tabaci-ChiLCV. RNAi or post-transcriptional gene silencing (PTGS) was introduced to study the functionality of specific genes through knocking down target gene mRNA [53]. However, in no time, it gained recognition as a potential therapeutic agent to control insect pests [54,55]. With the advancements in next-generation sequencing and transcriptomics technology, several genes have been identified as potential targets to control B. tabaci and its viruses [33][34][35][36][37][38][39][40][56][57][58]. However, most studies are limited to the oral feeding of the dsRNA to B. tabaci through artificial diet or genetic modification of the host plants [40,43,59,60]. None of these RNAi molecules have been tested for their efficacy or found effective under field conditions, limiting their large-scale adoption. RNAi using dsRNA constructs is not suitable for field conditions due to rapid degradation and a lack of reliable delivery agents. In the present study, for the first time, we demonstrated the efficacy of a naked dsRNA against an insect pest by spray-on application under semi-field conditions. The dsRNA construct targeting the hsp70 gene of B. tabaci not only induced mortality in B. tabaci but also inhibited the transmission of ChiLCV from infected to healthy chilli plants by B. tabaci.
Hsp70 and fas2 were found to be significantly upregulated in viruliferous B. tabaci along with several other genes [30,35,57]. Gene regulatory networking showed these genes were enriched with higher degrees of interactions [35]. Hsp70 transcripts increased upon ingestion of tomato yellow leaf curl virus (TYLCV) and squash leaf curl virus (SLCV) in DNA microarray [31]. The role of hsp70 in begomovirus transmission is well studied [30,31,57,60,61]. TYLCV CP and Hsp70 interacted in vitro and co-localized within midgut epithelial cells [30,31]. The fas2 mRNA expression level was upregulated in viruliferous B. tabaci by 2.6 and 4.518-fold in the RNA-Seq and RT-qPCR assay, respectively. The Fas2 orthologue in mammals (NCAM) is known to serve as a receptor for the rabies virus [39]. Hence, we targeted to knock down these two genes, hsp70 and fas2, to induce immunity to ChiLCV infection in B. tabaci. Conserved 128 and 150 nt stretches in the hsp70 and fas2 genes, respectively, were found to be specific to B. tabaci without any cross-reactivity to humans, mice, birds, butterflies, bees, ants, and plants. Each produced a putative siRNA of 21 nt. dsRNA was synthesized by in vivo transcription to harvest large quantities of dsRNA at a low cost. Bacterially expressed dsRNA is economical and convenient, as reported in several studies [62][63][64]. Before evaluating the efficacy of topically sprayed dsRNA, we tested it under controlled laboratory conditions. Both hsp70 and fas2 dsRNA were orally delivered to B. tabaci adults by mixing with an artificial diet in a feeding setup. Similar types of artificial feeding setup were reported to be efficient in the oral delivery of dsRNA [40][41][42]. A significant knockdown of the B. tabaci population was recorded at 24 h, which gradually increased with an increase in dsRNA concentration. Up to 82.22% mortality of B. tabaci was recorded after 24 h feeding on 3.0 µg/mL hsp70 dsRNA. Similarly, fas2 dsRNA at 3.0 µg/mL caused 72% mortality of B. tabaci 24 h post-feeding. The dsRNA exposure downregulated the hsp70 and fas2 mRNA expressions by 12.85-and 8.53-fold, respectively. The expression of an endogenous control gene was unregulated post-dsRNA exposure. This indicated the specificity of the dsRNA constructs to target genes. Hsp belongs to the multifunctional molecular chaperone family involved in the aggregation of damaged proteins, transportation, assembly, and disassembly of multi-structured units under stressed conditions [36,65]. The mortality of B. tabaci post-dsRNA feeding may be due to the loss of function via depletion of hsp70 mRNA, which might interrupt the normal biological processes in B. tabaci. In Drosophila melanogaster, Fas2 is involved in synaptic development and growth [37,38]. It is also known to function in intracellular signaling pathways that involve mitogen-activated protein kinase (MAPK) and regulate intracellular calcium levels [38]. However, the function of fas2 in B. tabaci is uncharacterized. Kanakala et al. [60] reported twisting of wings in B. tabaci upon silencing hsp70. However, we did not record any morphological deformities in treated B. tabaci. Hsp70 and fas2 might not be involved in the morphogenesis of B. tabaci, or longer exposure to dsRNA is required to produce such deformities.
The persistent circulative transmission of ChiLCV by B. tabaci indicates that the virus particles reach the midgut after ingestion. The virions cross the midgut barrier to become circulative in the hemolymph and accumulate in the primary salivary glands [30]. Hsp70, along with other molecular chaperones, plays an important role in the translocation of viral proteins besides its function in viral replication, assembly, and disassembly of the viral proteins inside the hosts [66][67][68][69][70]. Hsp70 has been found to interact with TYLCV CP in vitro and co-localize within midgut epithelial cells [31]. In the present study, oral delivery of hsp70 dsRNA significantly reduced ChiLCV copies in B. tabaci Asia II 1 and its transmission to healthy chilli plants as well. The resistance to ChiLCV in B. tabaci was enhanced at higher concentrations of hsp70 dsRNA. ChiLCV copies in B. tabaci were imperceptible when B. tabaci fed on 3 µg/mL hsp70 dsRNA for 24 h. The transmission of ChiLCV from infected to healthy plants by B. tabaci Asia II 1 was completely inhibited at 3 µg/mL hsp70 dsRNA treatment under controlled conditions. Our findings are consistent with Kanakala et al. [60], who reported the deleterious effect of B. tabaci hsp70 dsRNA on TYLCV transmission. The TYLCV titer decreased in B. tabaci Middle East Asia Minor 1 (MEAM 1) after feeding on hsp70 dsRNA-expressing tomato plants. Transmission of TYLCV also dropped by 12% post-silencing of hsp70 [60]. In the present study, the inability of B. tabaci Asia II 1 to retain and transmit ChiLCV post-silencing of hsp70 indicates that it might also play an important function in ChiLCV infection. However, Gotz et al. [31] reported an increase in TYLCV transmission upon blocking Hsp70 by anti-Hsp70 antibody in B. tabaci MEAM1. It was hypothesized that the Hsp70-TYLCV interaction mediates the degradation of virions. Downregulation of hsp70, hsp40, and hsp20 resulted in 3.1-, 1.5-, and 1.2-fold increases in cotton leaf curl virus (CLCuV) titer within B. tabaci. Significantly increased transmission efficiency of CLCuV was noted in B. tabaci Asia II 1 when hsp70 was silenced [61]. The difference may be due to the variation in virus species and B. tabaci cryptic species. The interactions of B. tabaci with begomoviruses are not conserved and alter with cryptic species of B. tabaci and begomovirus species [71,72]. B. tabaci MEAM1 has two genes (Bta03000 and Bta02903) annotated as hsp70, and their amino acid sequences differ by 97.55% [56]. They might also exhibit differential responses to begomovirus infection. Unlike hsp70 dsRNA treatment, an increase in the ChiLCV copies in B. tabaci was recorded post-fas2 dsRNA feeding. The Fas2 orthologue in mammals (NCAM) is known to serve as a receptor for the rabies virus. Although NCAM promotes the penetration of the virus in cells, it suppresses virus replication via induction of Interferon-ß [39], which is mainly involved in innate immunity against viral infection. Upregulation of fas2 transcripts in B. tabaci post-ChiLCV infection might be due to the innate immune response against the virus infection [35]. We hypothesized that fas2 has a negative regulatory role in ChiLCV infection, and thus, knocking down fas2 increased the virus titer in B. tabaci. Hence, silencing fas2 was not considered to induce resistance to ChiLCV despite its efficacy in eradicating B. tabaci.
Although RNAi has emerged as a promising alternative to suppress crop pests [73,74], there are limitations in the application of dsRNA as a 'spray-on' technology for field use. Reduced stability in the extracellular environments of dsRNA contributes to poor RNAi response. The naked dsRNA quickly degrades in higher temperatures and UV radiation under natural environmental conditions. Besides, several insect species, mainly lepidopteran insects, have been observed to be refractory to RNAi due to degradation and poor intracellular transport of exogenous dsRNA [75,76]. Nucleases are the major factors limiting the efficacy of dsRNA [75][76][77]. In the first phase of the study, hsp70 dsRNA showed high potential in eradicating B. tabaci and inhibiting ChiLCV transmission. In the next phase, the dsRNA construct was assessed under natural environmental conditions by spraying like an insecticide. Adult B. tabaci are highly mobile, so spraying on adult flies was not preferred. Hence, naked hsp70 dsRNA without any delivery agent was topically sprayed onto the foliage of the ChiLCV-infected chilli plant under contained semi-field conditions. B. tabaci adults were released on the dsRNA-sprayed plants to evaluate the dsRNA's efficacy in eradicating B. tabaci and inhibiting the transmission of ChiLCV. Considering the extracellular degradation of dsRNA and delivery through the host plant, a 10-fold higher dose than the controlled assay was used for topical application. A 67.77% mortality with a 4.6-fold decrease in hsp70 mRNA levels in B. tabaci exposed to the dsRNA-sprayed plants was noted after 24 h of release. Further, 1.84 × 10 8 -fold decreased ChiLCV copies and 75% reduced transmission of ChiLCV indicated the efficiency of topically sprayed hsp70 dsRNA in managing both the virus and its vector. Although the efficacy of topically applied naked dsRNA against plant viruses is known [78][79][80], this is the first evidence of its efficacy against a hemipteran insect like B. tabaci.
Topical application of dsRNA could provide resistance for 12 to 21 days against plant viruses [78,[80][81][82]. In the present study, dsRNA was detected in the leaf tissues up to 6 h post-application. Significant mortality of B. tabaci on the treated plants persisted up to 8 days and was further sustained up to 20 days with a second spray on the 10th day. Protection of the plants during the early growth stage is crucial in the management of begomoviruses. Begomovirus infection at a 5-leaf stage in tomatoes reduced yield by 95% [83]. Infection of ChiLCV at the seedling stage leads to a considerable reduction in plant height, internodal length, and fruiting. Hence, a 20-day shield by 2 consecutive sprays of naked hsp70 dsRNA would protect the crucial phase of crop growth from the invasion of B. tabaci and ChiLCV and reduce yield losses. The foliar spray of hsp70 dsRNA was effective under semi-field conditions where the daily temperature fluctuated between 20-35 • C with 40-60% RH. The efficacy and persistency of topically sprayed naked dsRNA at a wide temperature range endorse its potentiality for field uses as a spray-on technology.

Conclusions
In conclusion, the spray-on application of naked dsRNA targeting hsp70 of B. tabaci provides resistance to both ChiLCV and its vector, B. tabaci. This is a novel alternative to hazardous insecticides and may be further assessed on a large scale for its efficacy under real-field conditions.