Introduction

Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) is a pest of global importance that causes significant crop loss as a direct pest and more importantly as a vector of Geminiviruses [1]. The emergence of novel genetic groups in B. tabaci increased the risk of transmission of Geminiviruses to wide varieties of crops worldwide. It is reported that B. tabaci infests more than 900 species of plants of agricultural, fiber, vegetables, and ornamentals and also a dreaded vector of approximately 111 Geminiviruses [2]. It is also the world’s number one invasive species among the 100 invasive ones according to the International Union for Conservation of Nature and Natural resources (IUCN, http://www.iucn.org/). In addition, B. tabaci excretes copious amount of honeydew that promotes the growth of saprophytic fungi affecting photosynthesis [3]. Chemical management of B. tabaci has become increasingly ineffective due to the development of high levels of resistance to insecticides that include organophosphates, carbamates, and pyrethroids [4, 5]. Therefore, there is an urgent need to harness modern molecular technique as an alternative, where ribonucleic acid interference (RNAi) holds immense potential in developing species-specific insecticides [6–8].

RNAi is the sequence-specific posttranscriptional gene silencing mechanism, discovered more than a decade ago in Caenorhabditis elegans [9]. RNAi has shown its potential in establishing the functions of genes as well as in pest and disease management [10]. RNAi is induced either by endogenous or exogenous double-stranded RNA (dsNA) molecule, which destroys the cognate mRNA, thus affecting the physiology of the organism [9, 11]. To achieve gene silencing, the cognate dsRNA is delivered through various methods such as through diet, soaking, microinjection, feeding bacteria-expressed dsRNA, and dsRNA-expressing transgenic plants [11, 12]. However, dsRNAs delivered either as a spray or through transgenic plants are more pertinent to the field level pest management practices.

RNAi has been successfully used in the management of insect pests by silencing a wide range of genes including gut proteins in tsetse fly [13], molting genes in Tribolium catstaneum [14], aquaporins in aphids [15], digestive and caste regulatory genes in Reticulitermes flavipes [16], olfactory receptors in Apolygus lucorum [17], and extensively in other insect pests [12]. In this study, we have silenced glutathione S-transferase (GST) of the sigma class and evaluated its potential as pest management candidate in B. tabaci. GST is a cytolytic enzyme found in all eukaryotic organisms that participates in a number of different reactions which are important in defense against oxidative damage both by oxygen and free radicals [18]. However, GST has also been implicated in insecticide resistance in whiteflies [19].

In this study, we assessed the feasibility of diet-delivered dsRNA in silencing and persistence of the gene knockdown in two different time intervals. Additionally, we employed less invasive method of dsRNA administration (parafilm feeding method) to increase the ease of experimentation with improved whitefly survival.

Materials and Methods

Insect Culture

B. tabaci (=MEAM-1 genetic group, biotype B , Bemisia argentifolli Bellows and Perring) was utilized from the continuous culture maintained at the Division of Entomology, IIHR, Bangalore, maintained on cotton, Gossypium hirsutam at 28 °C, 45 % RH, and 14:10 photoperiod [11].

Cloning of GST

Total RNA from ten adult flies was extracted employing the RNeasy Mini Kit (Qiagen, Netherlands) following the manufacturer’s protocol. DNA contaminant was removed by treating with RNase-free DNase I (Ferments Life Sciences, USA) and purified by Phenol/chloroform (25:24) extraction [20]. The RNA concentration was estimated using Nanodrop ND-1000 (Thermo Scientific, Belgium) and on 2 % agarose gel electrophoresis. First-strand complementary DNA (cDNA) was prepared from 1 μg RNA employing PrimeScript™ First-Strand cDNA Synthesis Kit (TaKaRa) following the manufacturer’s protocol. Subsequently, cDNA was diluted (1:5) for PCR amplification with gene-specific primers (Table 1). PCR was performed in a final volume of 25 μl containing Taq buffer, 2.5 mm dNTP, 10 pmol of forward and reverse primers, 1 U TaqDNA polymerase, 3 μl of diluted cDNA, and water to make up the volume. The reaction condition was 94 °C for 5 min, followed by 35 cycles of 94 °C for 30 s, 54 °C for 30 s, 72 °C for 45 s, and final extension of 72 °C for 5 min. A negative control containing no DNA template was also included.

Table 1 List of primers employed in gene amplification, dsRNA synthesis, and real-time quantitative PCR (RT-qPCR)
Full size table

Amplified PCR product were eluted using gel extraction kit (Machery Nagel, Germany) and cloned in to general purpose cloning vector (PT257R/T, Thermo Scientific, USA), and three clones were sequenced in both the directions.

Sequence and Phylogenetic Analyses

Sequences assembly (both forward and reverse) were performed in BioEdit [21] and analyzed for homology with other sequences from insects in NCBI-BLAST, and the final sequences were submitted to NCBI-GenBank (Table 1). The annotated GST sequences from B. tabaci and 27 insect species were aligned using MUSCLE [22]. The alignment was used to construct the maximum likelihood (ML) trees using RaxML version 7.26 [23] under GTR + I + G model estimated by ProtTest [24]. To assess the branch support, 500 bootstrapping were performed using CIPRES interface [25].

dsRNA Synthesis

An off-target minimized region of 416 bp was identified for GST employing dsCheck (http:dscheck.RNAi.jp/) [26]. The T7 RNA polymerase promoter sequences (5′-TAATACGACTCACTATAGGG-3′) were tailed to the 5′ of the gene-specific primers that were employed for dsRNA synthesis. DNA templates for dsRNA were synthesized using PCR with similar reagents and PCR conditions, expect for DNA templates (1:100 diluted plasmid clones of the gene) and annealing temperature (65 °C for 45 s). The PCR products were resolved in 1.2 % agarose gel and purified using Nucleospin Extract II kit (Macherey Nagel, Germany). One microgram of the eluted DNA template was used for in vitro transcription using the T7 Ribomax™ express RNAi system (Promega, USA), precipitated with isopropanol and resuspended in nuclease-free water, followed by quantification in Nanodrop™ 1000 (ThermoScientific, USA). The integrity was analyzed by agarose gel electrophoresis (1.5 %) and kept at −70 °C until further use. dsRNA was diluted with artificial diet [27] to yield with three different concentrations, viz., 0.25, 0.50, and 1.0 μg/μl.

Insect Bioassay

A diet-based dsRNA delivery technique as previously reported [27–29] was followed in this study. The above diet was prepared in molecular biology grade water, filter-sterilized (0.22 μm) prior to the mixing of dsRNA at three concentrations, viz., 0.25, 0.50, and 1.0 μg/μl, and sandwiched between the two layers of UV-sterilized parafilm and one stretched on the inner surface of the cap. About 40 newly emerged adults of B. tabaci were taken in each tube, allowing them to feed on diet alone for 6 to 8 h, and then, healthy live 20 insects were transferred to a new tube with the diet containing dsRNA and the mortality was recorded daily. There were five treatments, such as diet alone (blank control), the diet containing dsRNA for Lac-Z (negative control), and the diets containing dsGSTs. All the treatments were replicated five times, and a total of 100 adults in each treatment were used.

RT-qPCR

Total RNA extracted from ten adults and converted to cDNA as described previously. cDNA was diluted tenfold of which 5 μl was used for PCR amplification employing gene-specific primers (Table 1). Real-time PCR was performed employing SYBR Green Jump Start™ Taq ReadyMix™ (Sigma-Aldrich, USA) according to the manufacturer’s protocol in a Light Cycler 480 II (Roche Applied Science, Switzerland). All the RT-qPCR assays were performed according to the MIQE guidelines [30]. Actin was used as an internal control gene for normalization, which was found to be the most suitable reference gene in a previous study (unpublished data). All the assays were triplicated, and also, no template controls were included. The relative gene expression data were analyzed using 2−ΔΔCT method as described by Livak and Schmittgen [31], and normalized values were expressed as percent silencing and compared to negative controls.

Data Analyses

The data is presented as mean ± SE and were analyzed by using t test employing GraphPad Prism v.5 (GraphPad Software, Inc., USA) at P < 0.05.

Results

Molecular Cloning and Sequence Analyses

A full-length GST (609 bp) was amplified employing the cDNA prepared from the total RNA from adult B. tabaci and submitted to NCBI-GenBank (accession number KJ913696). The BLAST results of the sequence indicated 100 % match to the previously reported GST sequences from B. tabaci. The deduced sequences of the cDNA consisted of 203 amino acid residues with a calculated molecular mass (MM) and isoelectric point (PI) of 50.09 kDa and 5.19, respectively, employing ExPASy Proteomics Website (http://web.expasy.org/). TMHMM Server v. 2.0 (http://www.cbs.dtu.dk/services/TMHMM-2.0/ [32] could not predict any transmembrane helix from our GST sequence. Similarly, no signal peptide motif was identified by SignalP 4.0 Server (http://www.cbs.dtu.dk/services/SignalP/) (Supplementary Fig. 1) as a result of no cleavage site evident.

Phylogenetic Analyses

NCBI-GenBank database search identified 27 sigma-class GSTs, and the retrieved sequences exhibited perfect match with GST sequence of B. tabaci from this study implemented in the MUSCLE program fixed in the MEGA 6.0 [22]. The alignment was used to construct a phylogenetic tree using MEGA 6.0. Neighbor-joining, maximum parsimony, ML, and minimum evolution methods generated trees with similar topology and bootstrap values; thus, only the ML tree is presented here (Fig. 1). The overall sequence homology shared between the BtGST and known sigma-class GSTs from other species of insects was about 49 to 65 % with highest homology to Acyrthosiphon pisum. The evolutionary relationship of GST proteins derived from B. tabaci and 28 other species of insects were evaluated. The ML trees for the GSTs of B. tabaci grouped with various aphid species viz., Acyrthosiphon pisum, Aphis citricidus, Aphis gossypii, and Kaburagia rhusicola that are nearest to the order Hemiptera.

Fig. 1
figure 1

Maximum likelihood (ML) tree showing phylogenetic relationship of the newly identified sigma-class GST from Bemisia tabaci (BtGST) with other corresponding sequences from different insects. Species name of insects with NCBI-GenBank accession numbers mentioned in the tree with indicated bootstrap values >50 %. Apart from sigma class, other classes of GSTs viz., theta, mu, zeta, omega, delta, and epsilon from Bombyx mori and Sarcoptes scabiei used as outgroup

Full size image

Insect Bioassay

The administration of dsRNA (Fig. 2) for GST showed significant levels of mortality due to silencing. The mortality data recorded in different time intervals such as 24, 36, 48, 60, and 72 h for three concentrations of cognate dsRNA viz., 0.25, 0.50, and 1.0 μg/μl and are given in the Supplementary Table 1. In each treatment, 20 adult whiteflies were used and the mortality recorded at different time intervals as independent experiments. Diet-delivered dsRNA for GST produced mortality ranging from 3.0 to 77 % (F = 1.335, df = 4, 20, P < 0.05) (Fig. 3). Mortality was observed in treated (non-target gene, dsLacZ) and untreated (diet alone) controls, which was in the range of 3.0–20.00 and 2.0–21.00 %, respectively. However, our results indicated that diet-delivered dsRNA for LacZ did not show any impact on whitefly mortality, since the similar percent mortality was observed in the untreated control (Figs. 3 and 4).

Fig. 2
figure 2

Collection of whiteflies and bioassays: diet pouch with diet were prepared using sterilized stretched parafilm on individual caps of specimen tubes (A1 and A2). Adult flies were collected on cabbage, Brassica oleracea plants and transferred into 30 ml specimen tubes (A3 and A4). Control live white flies feeding on the sandwiched diet (48 h). Diet pouch with dsRNA were prepared using sterilized stretched parafilm on individual caps of specimen tubes (B1 and B2). Dead white flies upon the treatment inside the tubes (48 h) (side view) (B3 and B4)

Full size image
Fig. 3
figure 3

The percent mortality of whiteflies after diet-mediated dsGST ingestion. Mortality rate was counted in every 24, 36, 48, 60, and 72 h. The values are represented as the average of five biological replications. Error bars represent the standard deviation value. Asterisks above the bars indicate that the values were significantly different (P < 0.05, Tukey’s t test, n = 5)

Full size image
Fig. 4
figure 4

Percent silencing of whiteflies after diet-mediated dsGST delivery. Expression was monitored for 2 days. Sampling was done on 24 and 48 h. The values are represented as the average of three biological replications. β-Actin was used as an internal reference gene. The results were analyzed using 2−∆∆CT method. Error bars represent the standard deviation of 2−∆∆CT value. Asterisks above the bars indicate the values were significantly different (P < 0.001, Tukey’s t test, n = 3)

Full size image

Target Gene Silencing

Real-time PCR efficiency was calculated for both GST (target) and Actin (internal control) as 1.936 (96.8 %) and 1.942 (97.1 %), respectively. Melt curve analyses of the amplicons of these genes exhibited single peak that indicated single product, which was reconfirmed by cloning, sequencing, and BLAST analyses of the same.

Our results showed downregulation of BtGST transcript in comparison with both controls on different time intervals across various dsRNA concentrations (Fig. 4). The percent silencing of BtGST varied from 7.84 % (0.25 μg/μl–24 h) to 77.43 % (1.0 μg/μl–72 h) compared to the dsLacZ and untreated whiteflies (Fig. 4). Percent silencing is directly proportional to the dsRNA the concentration (Fig. 4) and mortality (Figs. 3 and 4).

Discussion

The GST is a large family of multifunctional enzymes that play a vital role in the detoxification of both endogenous and xenobiotic compounds [33]. GSTs primarily catalyze the conjugation of the thiol group of reduced glutathione to the electrophilic centers of lipophilic compounds rendering them more water soluble and can be rapidly excreted [34, 35]. Majority of studies on insect GSTs have been focused on their role in detoxification of insecticides and allelochemicals and oxidative stress response [35–37]. GSTs are primarily classified into two major groups based on their location within the cell, i.e., microsomal and cytosolic. A third group of GSTs, the kappa class, is located in mammalian mitochondria and peroxisomes [38, 39]. Later, phylogenetic analyses of insect GST genes showed that it can be broadly classified into different classes, viz., sigma, delta, and epsilon, and majority of the remaining cytosolic insect GSTs are members of the zeta, theta, and omega classes [40–42]. Specifically, GSTs have been shown their potential in developing insecticide resistance in B. tabaci [5], ticks [43, 44], mites [45], etc. In this study, we sequenced BtGST and showed that it is phylogenetically more closely related to sigma class (insect class II) and more divergent from other classes of GSTs such as mu, epsilon, delta, zeta, and theta (Fig. 1).

In this study, we also demonstrated that diet-mediated delivery of dsGST can successfully inhibit the translation of the same in B. tabaci. Oral delivery of dsRNA has enormous potential applications in the management of agricultural insect pests. Diet-mediated dsRNA-induced gene silencing has been previously evidenced in a wide variety of agricultural insect pests of the order Lepidoptera [46–49], Diptera [13, 50–53], Orthoptera [54], Hymenoptera [55], Hemiptera [15, 56–65], Coleoptera [63, 66–69], and Isoptera [16]. So far, dsRNA delivery to various insect has been accomplished through oral feeding, microinjection, soaking, and transgenics [12, 70, 71, 46]. Among which, microinjection is the most commonly used method, but the disadvantages are (i) laborious, (ii) requirement of expertise and equipments, and (iii) mortality rates will be higher due to the invasive nature. Hence, we employed alternate less invasive and easy method of diet-mediated dsRNA delivery to assess the effect of RNAi in B. tabaci. In this study, we demonstrated that the ingestion of dsRNA can bring about the gene silencing of target genes in B. tabaci. A reduction in the BtGST gene transcripts resulted in mortality of B. tabaci during the bioassay. Thus, our study showed that BtGST downregulation mediated by dsRNA ingestion was lethal to B. tabaci and revealed the role of GST in cell signaling for cell proliferation required for growth and development of this hemimetabolous insect.

The effect of diet-delivered dsRNA on BtGST transcript levels in B. tabaci has been directly proportional to the dose (Fig. 4). Similar results were observed in Sogatella furcifera [72], Nilapravata lugens [73], Aedes aegypti [51], etc. However, certain other studies have also observed that increasing dsRNA concentration beyond an optimal dose did not improve extent of gene silencing, although the optimal concentration of dsRNA may vary for the insect species in question, mode of cognate dsRNA delivery, and the developmental stages of insects [15, 12]. Our percent silencing ranged 7.84–21.43, 16.67–30.85, 25.15–42.65, 37.77–57.94, and 52.95–77.43 % for BtGST in 24, 36, 48, 60, and 72 h, respectively. Thus, our data is comparable to that of the previous studies conducted in many other insects, where the percent silencing typically ranged from 30 to 80 % with either mortality or observable phenotypic changes [13, 16, 48, 51, 52, 57, 64, 74].

It is known that few factors are critical for exogenous dsRNA-mediated gene silencing in insects such as physiological importance of the target genes, concentration of cognate dsRNA, frequency of administration, and presence or absence of key core components for systemic RNAi machinery such as systemic interference defective (sid-1) and RNA-dependent RNA polymerase (RdRp), which are less understood in insects [75, 76]). As far as target genes are concerned, our results showed that BtGSTs could potentially serve as a target for developing novel insect pest management strategy. Different target genes resulted in various levels of silencing with either mortality or phenotypes [12, 14]. Our results showed that after 72 h of ingestion of dsGST, it resulted in 77 % mortality, while 21 and 20 % mortality was observed in untreated and dsLacZ, respectively. It has been previously reported that about 70 % silencing was evident after hemocoelic injection of dsRNA in B. tabaci [11] as against 29–97 % mortality [29], which supports the results obtained in the present study.

Depletion of 7.84 % (0.25 μg/μl–24 h) to 77.43 % (1.0 μg/μl–72 h) of BtGST transcripts though was not high as compared to the previous studies on Spodoptera litura, where 95 % reduction of amino peptidase transcript was evident. It could be due to the pooled samples of B. tabaci that were used to construct the overall efficiency of dsGST. Both mortality data and percent silencing of BtGST indicated that the extent of mortality and silencing was directly proportional to the concentration of dsRNA administered. Thus, gene knockdown by diet-mediated delivery of dsGST in B. tabaci will be undoubtedly an invaluable tool in the management of this invasive insect pest.

In conclusion, we have demonstrated the utility of the exogenously administered cognate dsRNA for GST of B. tabaci, a notorious vector of Geminiviruses and direct pests of wide varieties crops worldwide. We have also designed a simple, inexpensive, high-throughput method of dsRNA administration, which resulted in mortality due to downregulation of BtGST. Mortality was evaluated across three concentrations of diet-delivered dsGST of sigma class with highest mortality of 77.43 % for 1.0 μg/μl after 72 h, while mortality in control was 21 % under the same conditions. Based on the present study, we conclude that oral delivery of cognate dsRNA through diet or transgenic crops expressing dsRNA has potential applications for whitefly management and deserve further research.