Portal de Periódicos da CAPES

Silencing of Midgut Aminopeptidase N of Spodoptera litura by Double-stranded RNA Establishes Its Role asBacillus thuringiensis Toxin Receptor

2002; Elsevier BV; Volume: 277; Issue: 49 Linguagem: Inglês

10.1074/jbc.c200523200

1083-351X

Raman Rajagopal, Sivakumar Swaminathan, Neema Agrawal, Pawan Malhotra, Raj K. Bhatnagar,

Entomopathogenic Microorganisms in Pest Control

Insecticidal crystal proteins of Bacillus thuringiensis bind to receptors in the midgut of susceptible insects leading to pore formation and death of the insect. The identity of the receptor is not clearly established. Recently a direct interaction between a cloned and heterologously expressed aminopeptidase (slapn) from Spodoptera lituraand the Cry1C protein was demonstrated by immunofluorescence andin vitro ligand blot interaction. Here we show that administration of slapn double-stranded RNA to S. litura larvae reduces its expression. As a consequence of the reduced expression, a corresponding decrease in the sensitivity of these larvae to Cry1C toxin was observed. The gene silencing was retained during the insect's moulting and development and transmitted to the subsequent generation albeit with a reduced effect. These results directly implicate larval midgut aminopeptidase N as receptor for Bacillus thuringiensis insecticidal proteins. Insecticidal crystal proteins of Bacillus thuringiensis bind to receptors in the midgut of susceptible insects leading to pore formation and death of the insect. The identity of the receptor is not clearly established. Recently a direct interaction between a cloned and heterologously expressed aminopeptidase (slapn) from Spodoptera lituraand the Cry1C protein was demonstrated by immunofluorescence andin vitro ligand blot interaction. Here we show that administration of slapn double-stranded RNA to S. litura larvae reduces its expression. As a consequence of the reduced expression, a corresponding decrease in the sensitivity of these larvae to Cry1C toxin was observed. The gene silencing was retained during the insect's moulting and development and transmitted to the subsequent generation albeit with a reduced effect. These results directly implicate larval midgut aminopeptidase N as receptor for Bacillus thuringiensis insecticidal proteins. The bacterium Bacillus thuringiensis(Bt) 1The abbreviations used are: Bt, Bacillus thuringiensis; APN, aminopeptidase N; BBMV, brush border membrane vesicles; NBT, nitro blue tetrazolium; BCIP, 5-bromo-4-chloro-3-indolyl phosphate toluidine; Cry, crystalline inclusion protein; dsRNA, double-stranded RNA; RNAi, RNA interference; LMW, low molecular weight; IB, inclusion bodies; siRNA, small interfering RNAs; sRNA, sense RNA; asRNA, antisense RNA; DEPC, diethyl pyrocarbonate; RT, reverse transcriptase; nt, nucleotide(s) produces insecticidal crystal proteins, which upon ingestion by susceptible larvae get activated in the midgut, interact with specific receptor and form pores in the epithelium, resulting in the death of the larvae (1Schnepf E. Crickmore N. Van Rie J. Lereclus D. Baum J. Feitelson J. Zeigler D.R. Dean D.H. Microbiol. Mol. Biol. Rev. 1998; 62: 775-806Google Scholar). Understanding the mechanism of action of Bt toxin and development of resistance in insects is fundamental in sustaining the use of Cry proteins in integrated pest management. One of the mechanisms of resistance development is an alteration in the binding ability and/or a decrease in the population of receptor molecules, which bind Bt toxin in the insect midgut (2Gould F. Annu. Rev. Entomol. 1998; 43: 701-726Google Scholar). There have been intense efforts to characterize the nature of this receptor. As a result of several independent experiments employing ligand blot analysis and fluorescent labeling of insecticidal proteins, cadherin and aminopeptidase N (APN) have emerged as main putative receptor molecules (Ref. 3Jenkins J.L. Dean D.H. Setlow J.K. Genetic Engineering Principles and Methods. 22. Kluwer Academic/Plenum Publishers, New York2000: 33-54Google Scholar and references there in). While the role of a receptor molecule in mediating the effect of Cry toxin is acknowledged, the identity of this receptor is still being worked out. Aminopeptidase N from Manduca sexta was the first molecule to be tentatively identified as a Cry toxin-binding protein (4Knight P.J.K. Crickmore N. Ellar D.J. Mol. Microbiol. 1994; 11: 429-436Google Scholar, 5Sangadala S. Walters F.S. English L.H. Adang M.J. J. Biol. Chem. 1994; 269: 10088-10092Google Scholar), and APN is the most extensively studied putative receptor, having been identified and isolated subsequently from other lepidopteran insect pests. Independently, a 210-kDa cadherin-like protein from M. sexta was shown to interact with Cry1Ab toxin (6Vadlamudi R.K. Weber B. Ji I. Ji T.H. Bulla L.A. J. Biol. Chem. 1995; 270: 5490-5494Google Scholar) and later its presence and toxin interaction was also demonstrated from another insect, Bombyx mori (7Nagamatsu Y. Koike T. Sasaki K. Yoshimoto A. Furukawa Y. FEBS Lett. 1999; 460: 385-390Google Scholar). Relative abundance of APN in the posterior midgut (8Carroll J. Wolfersberger M.G. Ellar D.J. J. Cell Sci. 1997; 110: 3099-3104Google Scholar) and lower binding constants of Cry toxin toward cadherin as compared with APN (9Jenkins J.L. Dean D.H. BMC Biochemistry. 2001; 2: 12Google Scholar) raised apprehension about the role of APN as a receptor for Bt toxin in the insect midgut. Moreover, a recent report shows that high levels of resistance to the Bt toxin, Cry1Ac, inHeliothis virescens is due to disruption of a cadherin superfamily gene by retroposon-mediated insertion (10Gahan L.J. Gould F. Heckel D.G. Science. 2001; 293: 857-860Google Scholar). Althoughin vitro experiments such as toxin-induced increase in the86Rb+ efflux from lipid vesicles reconstituted with APN and reduction in the inhibition of short circuit current (ISC) for Cry1Ac following the release of APN from midgut membrane by phosphatidylinositol phospholipase C treatment provide support for the role of APN as a receptor, the vital in vivo evidence for receptor characterization is lacking (5Sangadala S. Walters F.S. English L.H. Adang M.J. J. Biol. Chem. 1994; 269: 10088-10092Google Scholar,11Lee M. You T. Young B. Cotrill J. Valaitis A. Dean D. Appl. Environ. Microbiol. 1996; 62: 2845-2849Google Scholar). RNA interference (RNAi) is a process of dsRNA-mediated gene silencing in which only the mRNA cognate to dsRNA is specifically degraded (12Fire A. Xu S.Q. Montgomery M.K. Kostas S.A. Driver S.E. Mello C.C. Nature. 1998; 391: 806-811Google Scholar, 13Kennerdell J.R. Carthew R.W. Cell. 1998; 95: 1017-1026Google Scholar, 14Lee N.S. Dohjima T. Bauer G. Li H. Li M.J. Ehsani A. Salvaterra P. Rossi J. Nat. Biotechnol. 2002; 20: 500-505Google Scholar). Recently, we reported isolation of a 2.8-kb APN-encoding geneslapn (AF320764), from the midgut of the polyphagous pest,Spodoptera litura, and its expression in the insect cells by a baculovirus expression vector. By in vitro ligand blot analysis and immunofluorescence toxin binding studies, we demonstrated a direct interaction between the expressed receptor and Cry1C protein (15Agrawal N. Malhotra P. Bhatnagar R.K. Appl. Environ. Microbiol. 2002; 68: 4583-4592Google Scholar). Here we report that dsRNA-mediated silencing of the aminopeptidase N gene results in increased resistance of S. litura larvae to Cry1C protein, thereby demonstrating a functional role for this protein in Cry protein-mediated toxicity. S. litura larvae were reared on fresh castor leaves (Ricinus communis) under a photoperiod of 14:10 h (light:dark), 70% relative humidity, and 27 °C. Midguts from 6th instar 1st day larvae were dissected in DEPC-treated water, snap-frozen into liquid nitrogen, and stored at −70 °C. Total RNA was isolated from the midgut tissue using TRIzol reagent (Invitrogen) according to manufacturer's protocol. The amount of RNA was quantitated spectrophotometrically at 260 nm. Inclusion bodies (IB) of Cry1C toxin were prepared as reported earlier by Lee (16Lee M.K. Milne R.E. Ge Z. Dean D.H. J. Biol. Chem. 1992; 267: 3115-3121Google Scholar). The amount of the toxin was quantitated densitometrically by resolving the IB on SDS-PAGE. Toxin amounts from 1000 to 10,000 ng was diluted in 10 mm Tris, pH 7.5, and 10 μl of each concentration was coated on both the sides of a castor leaf disc (area = 3.8 cm2). The toxin-coated leaf disc was air-dried and placed in a well of a 12-well tissue culture plate (Nunc Inc). One 6th instar 1st day larva was released on each well and exposed to the toxin treatment for 24 h. After 24 h the larva was transferred to fresh castor leaves (without toxin). Mortality was recorded after 4 days and the LC50 value calculated by Probit analysis using the software Indo Stat (Indostat Services, Hyderabad, India). Ten larvae were tested for each treatment and the bioassay replicated three times. LC50 values were also determined for neonate larvae ofS. litura by coating castor leaf discs with different doses of Cry1C toxin and scoring the mortality in each treatment after 72 h. 10 neonate larvae were released on each leaf disc, and each treatment was replicated five times. The dsRNA was prepared following the procedure prevoiusly described by us (17Malhotra P. Dasaradhi P.V.N. Kumar A. Mohmmed A. Agrawal N. Bhatnagar R.K. Chauhan V.S. Mol. Microbiol. 2002; 45: 1245-1254Google Scholar). A 756-bp fragment, from amino acid 609 to 861, was amplified from the S. litura APN gene and subcloned in pGEM-Te. The cloned fragment was amplified using vector-specific universal and reverse primers (Promega). The PCR product was purified using PCR purification kit (Qiagen GmbH) and used as a template to generate sense RNA (sRNA) and antisense RNA (asRNA) using T7 and SP6 RNA polymerases (Ambion), respectively. To make dsRNA, equal amounts of sRNA and asRNA were mixed, heated to 65 °C, and annealed by slow cooling over 4 h followed by DNase (Invitrogen) treatment for 15 min at 37 °C. The dsRNA was extracted with phenol:chloroform and precipitated overnight with ice-cold ethanol in the presence of 0.3m sodium acetate at −20 °C. The pellet was washed with 75% ethanol and resuspended in 50 μl of DEPC-treated water. Varying doses of dsRNA were injected intrahemocoelically into early 5th instar S. litura larvae using a microapplicator (KDS 200, KD Scientific Inc., New Hope, PA). After 48 h, the insect midguts were dissected, and total RNA was extracted. The quantity of RNA in each treatment was normalized by amplifying the β-actin gene in each treatment to equal intensity after 20 cycles of RT-PCR using the Titan one-tube RT-PCR kit (Roche Diagnostics Gmbh). The amplification regimen was as follows: reverse transcription at 43 °C for 35 min followed by 20 PCR cycles of denaturation at 94 °C for 30 s, re-annealing at 51 °C for 30 s, and extension at 72 °C for 30 s followed by a final extension of 10 min at 72 °C. Using this normalized amount of RNA as template, the slapntranscript in different treatments was compared by amplifying a 756-bp part of the slapn gene. Before loading on 1% agarose gel the RT-PCR products were treated with 1 μg of RNase (Qiagen GmbH) at 37 °C for 10 min to eliminate template RNA, since it hinders the correct estimation of the β-actin gene product in the gel. The gels were photographed with Polaroid 667 black and white print film and scanned for net intensity of each RT-PCR product using the software Kodak 1D, version 2.0. BBMVs were prepared from 6th instar midgut by following the protocol of Wolfersberger (18Wolfersberger M.G. Luthy P. Maurer A. Parenti P. Saachi V.F. Giordana B. Hanozet G.M. Comp. Biochem. Physiol. A. 1987; 86: 301-308Google Scholar). One microgram of the BBMV protein was resolved by 7.5% SDS-PAGE and electrotransfered to nitrocellulose membranes at 50 mA for 2 h at 4 °C. The blot was blocked in 3% bovine serum albumin in 1× phosphate-buffered saline and then incubated with 1:50,000 dilution of anti-APN antibodies. Subsequently, the blot was incubated with alkaline phosphatase-conjugated goat anti-rabbit IgG and developed with NBT-BCIP substrate. Low molecular weight RNA was isolated from midgut of insects injected with α-32P-labeled dsRNA following the protocol of Hamilton and Baulcombe (19Hamilton A.J. Baulcombe D.C. Science. 1999; 286: 950-952Google Scholar) and resolved on 12% acrylamide 7 m urea gel in 1× TBE buffer (10.8 g of Tris, 5.5 g of boric acid, 4.0 ml of 0.5 m EDTA, pH 8.0, made up to 1 liter with water). The gel was dried and exposed to a Kodak Bio-Max MR autoradiographic film overnight at −70 °C. The film was developed to detect the occurrence of radiolabeled siRNA. To evaluate the functional role of aminopeptidase N in the insecticidal activity of Cry1C in vivo, we sought to specifically inhibit the expression of slapn by its corresponding dsRNA. The reported routes of dsRNA delivery in lower eukaryotes such as Caenorhabditis elegans are soaking, feeding, or injection of dsRNA solution into the worm (12Fire A. Xu S.Q. Montgomery M.K. Kostas S.A. Driver S.E. Mello C.C. Nature. 1998; 391: 806-811Google Scholar, 20Tabara H. Grishok A. Mello C. Science. 1998; 282: 430-431Google Scholar), while dsRNA in Drosophila melanogaster was introduced by injecting into eggs. D. melanogaster embryos hatched from dsRNA-injected eggs displayed nearly 75% gene-silencing, which reduces to 3% with maturity (21Misquitta L. Paterson B.M. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 1451-1456Google Scholar). Preliminary experiments to introduce dsRNA into neonate larvae of S. litura by soaking them in dsRNA solution or by feeding through diet were unsuccessful, since no reduction in transcript levels was detected. Subsequently, varying amounts of dsRNA were injected into hemolymph of 5th instar larvae (2 μl/insect) using a microapplicator (KDS 200, KD Scientific Inc.). After 48 h of injection of dsRNA the transcript abundance was estimated by relative RT-PCR. An injection of 4 μg of dsRNA resulted in a 95% reduction over control of midgut slapn transcript levels (Fig. 1 a, lanes 1 and 2). The consequence of decrease in slapn transcript on the susceptibility of S. litura larvae against Cry1C protein was examined by performing toxicity assays. Initially, the assays were performed on freshly moulted, 6th instar larvae (injected with 2 μl of DEPC-treated water in 5th instar), a developmental stage corresponding to the growth status of 5th instar larvae after 48 h of injection. These larvae were released on 3.8-cm2 castor leaf discs coated with various concentrations of Cry1C protein (1–10 μg). Mortality was recorded after 4 days and LC50 (50% lethal concentration) value calculated by probit analysis. The calculated LC50 value was 5.642 μg/disc with the regression equation, y = 5.1921x-14.478, and the fiducial limits lie between 5.355 and 6.025 μg. Larvae injected with dsRNA and released on castor leaf discs coated with 6 μg of Cry1C displayed reduction in the mortality rate. The increase in tolerance for dsRNA-injected larvae to Cry1C protein correlated with a comparable (70%) increase in the larvae that pupated. On the other hand, none of the control larvae could reach pupation (TableI).Table IReduction in S. litura's toxicity to Cry1C by dsRNATreatmentEffect on 6th instar larvaeEffect on neonate larvae in next generation, mortality %MortalityPupation%%Control61.0a00.0a65.0adsRNA injected15.0 (75.0)b70.0 (70.0)b45.0 (30.0)bslapn dsRNA was injected into 5th instar, 1st day larvae at 4 μg/larvae. Control larvae were injected with DEPC water of equal volume. On moulting to 6th instar, the larvae were transferred to individual leaf discs (3.8 cm2) coated with 6.0 μg of Cry1C protein. After 24 h, the larvae were transferred to fresh leaves. The experiment was replicated on three different occasions with 50 larvae each. Mortality was recorded 4 days after toxin application. In a separate experiment, 50 larvae injected with slapn dsRNA were allowed to grow without being exposed to Cry1C, pupate, and mate among themselves. The neonates from this treatment were transferred to castor leaf discs (3.8 cm2) coated with 250 ng of Cry1C and compared with neonates emerging from normal insects. Ten larvae of each treatment were released on toxin-coated leaf discs and replicated 15 times. The mortality data were subjected to Students t test and values superscripted by a and b are significantly different at 95% confidence limits. Values in parentheses indicate percent reduction in mortality over control. Open table in a new tab slapn dsRNA was injected into 5th instar, 1st day larvae at 4 μg/larvae. Control larvae were injected with DEPC water of equal volume. On moulting to 6th instar, the larvae were transferred to individual leaf discs (3.8 cm2) coated with 6.0 μg of Cry1C protein. After 24 h, the larvae were transferred to fresh leaves. The experiment was replicated on three different occasions with 50 larvae each. Mortality was recorded 4 days after toxin application. In a separate experiment, 50 larvae injected with slapn dsRNA were allowed to grow without being exposed to Cry1C, pupate, and mate among themselves. The neonates from this treatment were transferred to castor leaf discs (3.8 cm2) coated with 250 ng of Cry1C and compared with neonates emerging from normal insects. Ten larvae of each treatment were released on toxin-coated leaf discs and replicated 15 times. The mortality data were subjected to Students t test and values superscripted by a and b are significantly different at 95% confidence limits. Values in parentheses indicate percent reduction in mortality over control. To examine the effect on the expression of APN upon dsRNA delivery, BBMVs were prepared by the differential MgCl2 method from the midgut of 6th instar larvae and resolved by 7.5% SDS-PAGE. Probing the BBMV proteins with anti-APN antibodies revealed nearly 80% reduction in APN expression in dsRNA-injected larvae as compared with control larvae (Fig. 1 b). Thus the reduction in the expression of aminopeptidase correlates well with the reduction ofslapn transcript levels and the decreased sensitivity to Cry1C proteins (Fig. 1 a). Fifth instar larvae injected with slapn dsRNA were reared up to pupation and bred into the next generation. Analysis of neonate larvae for abundance of slapn transcript revealed 60% reduction in slapn levels (Fig. 1 a, lanes 3 and 4). Neonate larvae of the F1 generation withslapn silencing displayed resistance to Cry1C toxin compared with untreated neonate larvae (Table I). The LC50 value of Cry1C on untreated neonate larvae was 200 ng/3.8 cm2. An important step in RNAi-mediated gene silencing is the formation of 21–25-nt siRNAs, which target and degrade mRNA of the target gene (22Zamore P.D. Tuschl T. Sharp P.A. Bartel D.P. Cell. 2000; 101: 25-33Google Scholar). In the present study, the generation of siRNA was investigated by extracting LMW RNA (19Hamilton A.J. Baulcombe D.C. Science. 1999; 286: 950-952Google Scholar) from the midgut of larvae injected with α-P32-labeled dsRNA. By resolving the LMW RNA preparation on PAGE, a distinct band corresponding to 25 nt is observed (Fig.1 c). Occurrence of a 25-mer oligonucleotide establishes that RNAi pathway in larvae is similar to that observed in other organisms. Thus in the present study, we have demonstrated that the dsRNA-mediated silencing of slapn in whole S. litura larvae resulted in the decrease in the amount of APN expressed in the epithelial membrane of midgut cells, which in turn enabled the larvae to tolerate lethal concentrations of the Cry1C protein. By performingin vivo experiments we are able to provide a direct evidence for the role of APN as a Bt toxin receptor in the midgut of insects. The silencing of the S. litura apn by introducing cognate dsRNA into larvae demonstrates that RNAi functions in whole larvae and also that the midgut columnar cells can take up dsRNA molecules injected in the hemocoel. This result shows that like in the plant kingdom, there is a "systemic" effect of RNAi in animals too, in organs away from the point of delivery of dsRNA. Also, we show that RNAi effect by dsRNA administration is heritable in the next generation, as comparable with the interference obtained by using "hairpin-loop" RNA (23Kenderdell J.R. Carthew R.W. Nat. Biotechnol. 2000; 18: 896-898Google Scholar). Here, we have used RNAi as a technique to probe the function of a protein as a toxin receptor, thereby implying that it can be used to explore the functional role of different proteins involved in various host-pathogen interactions.