ENVIROMENTAL MICROBIOLOGY (2010) RNAi mediated knockdown of a Spodoptera frugiperda trypsin like...

RNAi-mediated knockdown of a Spodoptera frugiperdatrypsin-like serine-protease gene reduces susceptibilityto a Bacillus thuringiensis Cry1Ca1 protoxinemi_2259 2894..2903

Lianet Rodríguez–Cabrera1,Damian Trujillo–Bacallao1, Orlando Borrás–Hidalgo2,Denis J. Wright3 and Camilo Ayra–Pardo1*1Environmental Biotechnology Department, Centre forGenetic Engineering and Biotechnology (CIGB), Havana10600, Cuba.2Plant Functional Genomic Department, Centre forGenetic Engineering and Biotechnology (CIGB), Havana10600, Cuba.3Division of Biology, Department of Life Sciences,Faculty of Natural Sciences, Imperial College London,Silwood Park campus, Ascot, Berkshire SL5 7PY, UK.

Summary

SfT6 has been identified in a subtracted cDNA libraryof Spodoptera frugiperda larval midgut transcripts asa serine-protease gene downregulated within 24 h ofintoxication with Bacillus thuringiensis Cry1Ca1protein. In the present study, the specific role of SfT6during Cry1Ca1 intoxication was investigated byRT-PCR and in vivo RNA interference. Quantitativereal-time RT-PCR analysis showed SfT6 mRNA levelsin the midgut tissue were significantly reduced afterinjecting or feeding 4th-instar larvae with specificlong-size dsRNA. Gut juice-mediated in vitro protoxinactivation and susceptibility for Cry1Ca1 were inves-tigated in Sft6-knockdown larvae and compared withcontrol treated with nonspecific dsRNA. Our resultsdemonstrate SfT6 plays a determinant role in Cry1Ca1toxicity against S. frugiperda since a decreasedexpression caused a reduced protoxin activation bylarval gut juice and reduced susceptibility of insectsto toxin in bioassays. We propose SfT6 downregula-tion occurring at the early stages of Cry1Ca1 intoxi-cation is part of a complex and multifaceted defensivemechanism triggered in the insect gut to withstand B.thuringiensis pathogenesis.

Introduction

As in many other lepidopteran species, the fall armyworm,Spodoptera frugiperda (J.E. Smith) (Lepidoptera: Noctu-idae) is known to be armed with a battery of midgutenzymes for digestion of dietary proteins where serine-proteinases, particularly those with trypsin and chymot-rypsin specificities, are predominant (Ferreira et al.,1994a,b). Serine-proteinases are also determinants ofBacillus thuringiensis (Bt) Cry toxins insecticidal potencyby affecting their solubilization and activation in the alka-line and reducing environment of the midgut of intoxicatedinsects (Oppert, 1999; Lightwood et al., 2000). Solubilizedprotoxins of lepidopteran active Bt Cry1-class are 130–140 kDa in size that midgut serine-proteinases transforminto activated toxins of 55–65 kDa after removing approxi-mately 600 residues from the C-terminus and the first 28N-terminal residues (Schnepf et al., 1998). Activatedmonomers of the Cry toxins then cross the peritrophicmatrix barrier to bind specific membrane receptor(s) onepithelial target cells, where further proteolysis may takeplace before toxin oligomers are formed and inserted intothe membrane to open lytic pores (Bravo et al., 2002,2004).

A significant threat to Bt-based insect control is thepotential of key pests to develop resistance to the Crytoxins. Changes to Cry toxin binding sites on insectmidgut membranes have been the most commonlyreported type of Bt resistance mechanism (Ferré and VanRie, 2002; Sayyed et al., 2004; Heckel et al., 2007), butchanges in the pattern of trypsin- and chymotrypsin-likeproteinases have also been found in some Bt-resistantinsect strains suggesting a major role in resistance via thealteration of Cry protoxins activation (Oppert et al., 1994,1997; Forcada et al., 1996; Sayyed et al., 2001; Khajuriaet al., 2009). Oppert and colleagues (1997) found that astrain of Indianmeal moth (Plodia interpunctella) resistantto Bt var. entomocidus HD-198 processed the protoxin ata slower rate than the parental susceptible strain. Thisresistant strain lacked a major gut proteinase involved inthe activation of Cry protoxins, and the proteinase mecha-nism was responsible for about 90% of the total resis-tance to Cry1Ab (Herrero et al., 2001). In other studies,

Received 19 October, 2009; accepted 12 April, 2010. *Forcorrespondence. E-mail [email protected]; Tel. (+53) 72716022; Fax (+53) 7 2731779.

Environmental Microbiology (2010) 12(11), 2894–2903 doi:10.1111/j.1462-2920.2010.02259.x

© 2010 Society for Applied Microbiology and Blackwell Publishing Ltd

mailto:[email protected]

midgut enzymes from a strain of Helliothis virescens resis-tant to Bt var. kurstaki HD-73 were reported to activate theprotoxin more slowly and degrade the toxin faster thanenzymes from a susceptible strain (Forcada et al., 1996;1999).

The multiple physiological and biochemical changesdisplayed by Bt-resistant insects, such as enzymatic alter-ations in gut serine-proteinases, are generally acceptedas adaptive responses of midgut cells after prolongedperiods of exposition to sublethal Bt toxin doses (Oppertet al., 1997; Loseva et al., 2002; Candas et al., 2003).Transcriptional studies on insect midgut tissue followingsublethal exposure to Bt toxins have shown a multifacetedresponse involving up- and downregulation of midgutgenes associated with defence-, stress- and metabolic-related activities (van Munster et al., 2007; Rodríguez-Cabrera et al., 2008). Interestingly, a S. frugiperda midgutexpression sequence tag EST-SFFC3 (GenBank acces-sion no. EL618668) with homology to insect serine-proteinases has been identified among the genesdownregulated within the first 24 h of specific S.frugiperda–Bt Cry1Ca1 sublethal interaction (Rodríguez-Cabrera et al., 2008). According to the analysis of sub-tracted midgut cDNA libraries, EST-SFFC3 wasapparently the most if not the only S. frugiperda serine-protease affected during the early stages of Bt-Cry1Ca1intoxication. It is well known serine-proteinase genesfamily from lepidopteran insects is diverse and canrespond in different ways under stress conditions (Bownet al., 1997; Brioschi et al., 2007). Considering the role ofthese digestives enzymes at initial steps of the Cry1 toxicpathway and to some insect Bt-resistance mechanisms,the specific downregulation of EST-SFFC3 could indicatea major role of this particular serine-protease for theinsecticidal activity of Bt Cry1Ca1 in S. frugiperda.

Here we used RNA interference (RNAi) in S. frugiperdato determine the specific role of the EST-SFFC3 corre-sponding serine-proteinase gene, referred to as SfT6,during the Cry1Ca1 intoxication. The predicted aminoacids sequence of SfT6 full-length cDNA showed highidentity with annotated lepidopteran trypsins. Transcrip-tion profiling studies revealed SfT6 transcripts werepresent in feeding stages (larvae and adults) and absentin embryonic and pupal stages. Exposure of 4th-instarlarvae to Cry1Ca1 caused a drastic and sustaineddecrease in SfT6 transcript accumulation 15 min afterapplying the protoxin whereas mRNA levels from other S.frugiperda trypsins stayed invariable. RNAi-mediatedSfT6 silencing by injection or by feeding larvae with spe-cific dsRNA was found to cause a reduction in both the gutjuice-mediated protoxin activation ability and the suscep-tibility to Cry1Ca1 in bioassays compared with controlinsects treated with unspecific dsRNA derived from theEscherichia coli b-D-glucuronidase uidA gene.

Results

Cloning and sequence analysis of SfT6

The full-length cDNA sequence of SfT6 was amplified byRACE procedures and analysed. The cDNA (GenBankaccession no. FJ940726) was 904 bp in length and con-tained a 765 bp open reading frame (ORF) spanningnucleotides 57 to 821. The ORF encoded a protein con-sisting of 254 amino acids with a calculated molecularmass for the mature protein of 25.1 kDa and isoelectricpoint of 8.6.

Multiple alignment and phylogenetic relationship analy-sis of predicted amino acids for SfT6 mature polipeptidedenoted that the cloned fragment belongs to a trypsin-likeserine-proteinase gene (Fig. 1). The phylogenetic treeshowed that the S. frugiperda SfT6-predicted polypeptideis most similar to a Spodoptera litura trypsin-like enzyme(91% identity) and in the same branch as trypsin-likededuced proteins from most lepidopteran annotatedsequences. Specific alignment of SfT6 with partialsequences of four S. frugiperda trypsin-like serine-proteinases found in the integrative database SPODO-BASE (http://bioweb.ensam.inra.fr/spodobase/; Nègreet al., 2006) yielded high identity (95%) with trypsindeduced from ESTs cluster sf1f00035-3-1 and only a mar-ginal identity of 20% with those contained in clusterssf1f04035-3-1, sf1f02950-3-1 and sf1f04740-5-1.

Monitoring the expression of SfT6 gene by RT-PCR

To expolore the developmental expression pattern ofSfT6 gene, specific mRNA levels were assessed in eachS. frugiperda developmental stage, including eggs, alllarvae stages, pupae and adults. First, a set of oligo-nucleotides specific for spliced SfT6 mRNA was testedby PCR on S. frugiperda genomic DNA and midgutcDNA and was found that the fragment of expected sizeis produced only from the cDNA template (data notshown). RT-PCR showed that SfT6 transcripts are pro-duced during actively feeding stages of the insect life(larvae and adult) and are totally absent in embryonicand pupal stages (Fig. 2A). The accumulation rate ofSfT6 dramatically increased during the progression fromneonates to sixth larval instar suggesting an importantrole of this protease for food digestion during larvalgrowth and development.

Besides detecting SfT6 transcripts, RT-PCR analysisalso demonstrated transcript accumulation for trypsinssf1f02950-3-1 and sf1f04035-3-1 but not for sf1f04740-5-1 in this laboratory strain (Fig. 2B). Among the threetranscripts, SfT6 showed the highest production rate innormal conditions and it was the only visibly affected overthe first 24 h of larval exposure to the Cry1Ca1 protoxin.SfT6 mRNA levels were shown to rapidly decline by

RNAi of insect trypsin reduces susceptibility to a Bt toxin 2895

© 2010 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 2894–2903

http://bioweb.ensam.inra.fr/spodobase/

15 min of intoxication, although a basal level was stilldetected after 24 h exposure to the toxin (Fig. 2B).

In vivo RNAi-mediated SfT6 transcript suppressionas measured by quantitative one-step RT-PCRanalysis (qRT-PCR)

In order to minimize any off-target effect during SfT6RNAi, the gene fragment used for dsRNA synthesis com-prised the first 567 bp of the sequence where a 19 bpconsecutive identical cDNA sequences with trypsinssf1f04035-3-1 and sf1f02950-3-1 was not found (Fig. S1).For RNAi by feeding, first attempts administering dsRNAto young larval stages either on the diet or as a droplet-feeding assay were unsuccessful. In vitro dsRNA stabilityexperiments in the presence of S. frugiperda midgut juiceshowed a complete degradation of SfT6 dsRNA within30 min of incubation, possibly by the action of midgutdsRNA-degrading alkaline nucleases, but it was insignifi-

cant when digestive juice from 24 h starved newly moltedfourth instar larvae was used in the reaction (Fig. 3).Stress conditions such us starvation joined to larvalmolting tend to make the insect gut pH lower and moreuniform along the entire midgut affecting the activity andproduction of most alkaline gut–digestive enzymes(Wieczorek et al., 1999). However, when younger newlymolted larvae from first and second instars that haveessentially their midgut without food were subjected tothe 24 h starvation treatment, most of them were tooweak for further manipulation and a high percent of mor-tality occurs in mocked larvae during droplet–feedingexperiments. To avoid the above problems, newly molted4th-instar S. frugiperda larvae were chosen for SfT6 func-tional studies.

The efficacy of SfT6-knockdown was investigated 48 hafter the corresponding RNAi treatment by qRT-PCRanalysis on larval midgut total RNA. The results revealedthat both RNAi by injection and RNAi by feeding with

Anopheles gambiae str. PEST (XP_317170)

Anopheles stephensi (AAB66878)

Bombyx mandarina (ABU49589)

Apis mellifera (XP_397087)

Drosophila mojavensis (XP_002003182)

Manduca sexta (CAM84320)

Ostrinia nubilalis (OnT2) (AAR98920)

Drosophila melanogaster (NP_525112)


Galleria mellonella (AAK81696)

Plodia interpunctella (PiT2c) (AAF24227)


Plodia interpunctella (PiT2a) (AAC36247)

Plodia interpunctella (PiT2b) (AAF24225)

Agrotis ipsilon (AAF74736)

Sesamia nonagrioides (AAT95353)

Bombyx mori (NP_001040350)

Spodoptera litura (ABU96714)

Spodoptera frugiperda (SfT6) (ACR25157)

Heliothis virescens (T2a) (ABR88242)

Helicoverpa armigera (CAA72962)

Fig. 1. Phylogram showing the relationship between S. frugiperda SfT6 (FJ940726) and selected insect trypsin-like proteins as obtained usingPhyML and TreeDyn programs. SfT6 is highlighted in bold. Deduced amino acid sequence for SfT6 protein and sequences encoded by relatedgenes were phylogenetically aligned using the Muscle vs. 3.7 program. GenBank accession numbers are reported in parentheses. Bootstrapvalues greater than 50 are indicated along branches. The scale bar indicates the evolutionary distance.

2896 L. Rodríguez–Cabrera et al.


specific dsRNA-SfT6 produce a significant reductionof the SfT6 expression in midgut tissue (Fig. 4). Inboth cases, SfT6 transcript levels decreased in morethan 90% of control being slightly more prominentwhen the dsRNA was administered orally. Also, targetspecificity of RNAi silencing was confirmed by theadministration of the nonspecific dsRNA-Gus thatshowed no effect on the SfT6 levels (Fig. 4). In addition,no amplification of the fluorescent signal was detected ina negative control sample containing S. frugiperdagenomic DNA as template, confirming the specificity forspliced mRNA of the designed SfT6 primers (data notshown).

Effects of SfT6-silencing on S. frugiperda larvalresponse to Cry1Ca1 protoxin

Table 1 shows mortality (LC50, concentration that causesdeath in 50% of the population) and molting inhibition

(MIC50, concentration that cause molt inhibition in 50% ofthe population) data of Cry1Ca1 protoxin on second andfourth larval instars of this laboratory strain. Fourth instarlarvae were about 60-fold less sensitive in terms of LC50

than second instar larvae, and only 20-fold less sensitivein terms of their MIC50, which shows that Cry1Ca1 is stilltoxic against fourth instar larvae of S. frugiperda.

Since the invasive nature of RNAi by injection thatcaused larval death in up to 16% of control larvae, dsRNAdroplet-feeding larvae were preferred for SfT6 functionanalysis during Bt Cry1Ca1 pathogenesis. A larval growthinhibition (GI) assay using a protoxin concentration of2 mg cm-2, which approximated to the LC25 and MIC70 ofCry1Ca1 against fourth instar stage larvae, was usedinstead of a dose-mortality response since weight reduc-tion is a more sensitive parameter than the LC50 for lesssusceptible stages (MacIntosh et al., 1990). Cry1Ca1 tox-icity expressed as percentage of growth inhibition (%GI)was determined in dsRNA-treated and untreated larvae48 h after droplet-feeding assay. No differences in %GIvalues were found between larvae that were fed withbuffer only (mocked) or nonspecific dsRNA-Gus (Fig. 5).In contrast, larvae from the group fed with dsRNA-SfT6were ~30% less affected by the Cry1Ca1 protoxin in termsof %GI at the same tested concentration.

Fig. 2. RT-PCR analysis of S. frugiperda trypsins.A. SfT6 transcript accumulation at different stages of the S.frugiperda life cycle. N, neonates; L1–6, the six larval instars of S.frugiperda; PP, pre-pupae; P, pupae; M, moths; E, embryo.B. Analysis of SfT6, sf1f02950-3-1 and sf1f04035-3-1 mRNA levelsat indicated time (h) post-intoxication with 2 mg cm-2 of Bt Cry1Ca1protoxin. rRNA refers to ribosomal RNAs.

Table 1. Toxicity of Bt Cry1Ca1 protoxin to different larval stages of S. frugiperda determined after 8 days at 30°C.

Larval stage LC50a Slope (� S.E.) MIC50

b Slope (� S.E.) nc

2nd instar 0.529 (0.304–0.921) 0.57 � 0.282 0.028 (0.017–0.047) 0.73 � 0.257 2804th instar 30.01 (14.90–58.33) 0.43 � 0.349 0.550 (0.239–1.266) 0.81 � 0.424 280

Fiducial limits (95%) are giving in parenthesis.a. LC50, concentration that causes death in 50% of the population expressed as mg of toxin per cm2 of diet surface.b. MIC50, concentration that causes molt inhibition in 50% of the population (mortality plus larvae failing to molt to next instar) expressed as mgof toxin per cm2 of diet surface.c. Number of larvae used in bioassay, including control.

Fig. 3. in vitro stability of dsRNA-SfT6 in the midgut juice of S.frugiperda. 100 ng of dsRNA-SfT6 in 20 ml of delivery buffer (lane1) were incubated at 28°C during 1 h with 5% (v/v) gut juice fromnon-starved (lane 2) and 24 h starved (lane 3) newly molted fourthinstar larvae before loading onto 0.8% agarose gels. Lane 4,GeneRuler 100 bp Plus DNA ladder (Fermentas GmbH, St.Leon-Rot, Germany).



Based on toxicity results we chose to examine the par-ticipation of SfT6 in Cry1Ca1 protoxin activation throughin vitro processing assays using larval gut juice from dif-ferent RNAi treatments. After 1 h of Cry1Ca1 incubationwith gut juice from larvae fed with nonspecific dsRNA-Gus, a single band of ~58 kDa was generated that did notdiffer from that obtained with gut juice from mocked(buffer-fed) larvae (Fig. 6, lanes 1 and 3). On the contrary,

a moderate but clear decrease in protoxin activation ratewas found when midgut juice from dsRNA-SfT6-fedlarvae was used in the reaction (Fig. 6, lane 2). In thissample, the midgut juice failed to completely process theCry1Ca1 protoxin to the ~58 kDa product, the apparentactivated form of Cry1Ca1 toxin in S. frugiperda, but pro-duced in addition, two major partially processed productsof ~60 and ~66 kDa (Fig. 6, lane 2).

Discussion

In the present work, the full-length cDNA sequence ofSfT6, a 25 kDa trypsin-like proteinase from S. frugiperdamidgut tissue, was cloned and its role during larval intoxi-cation with Bt Cry1Ca1 protoxin investigated. Here wedemonstrated, by using RT-PCR and in vivo RNAi silenc-ing, that SfT6 serine-proteinase is specifically involved inthe Cry1Ca1 toxic pathway in S. frugiperda. Besides, invitro protoxin processing and insect bioassays on RNAi-treated 4th-instar larvae showed that SfT6 plays an impor-tant role in protoxin activation in the gut and hence in thelarval susceptibility against Cry1Ca1.

RNAi has been used for investigating a number ofgenes in different insect species of the order Lepidoptera(Bellés, 2010). In S. frugiperda, RNAi silencing has beenpreviously used to investigate the role of allatoregulatingneuropeptide genes in the regulation of hemolymph juve-nile hormone titres during the insect life stages (Meyering-Vos et al., 2006; Griebler et al., 2008). Griebler andcolleagues (2008) found that injection and droplet feedingof dsRNA in freshly molted penultimate larvae caused asimilar effect in the reduction of the specific transcriptlevels whereas we have found the efficiency of RNAi forSfT6 in fourth instar larvae was slightly greater whendsRNA was applied orally compared with injection

Fig. 4. Reduction of SfT6 expression in larval midgut after RNAisilencing. Midgut total RNA (100 ng) pooled from 10 S. frugiperdafourth instar larvae 48 h after injecting with 5 mg dsRNA larva-1 ordroplet feeding with 3 mg dsRNA larva-1 of either dsRNA-SfT6 ordsRNA-Gus was used to measure SfT6 gene expression byqRT-PCR. The results are expressed as the percentage of control(untreated with dsRNA) after normalization with the reference geneb-actin. Bars represent means of two independent replicates � SE.

Fig. 5. Toxicity of Bt Cry1Ca1 protoxin (2 mg cm-2) to S. frugiperdafourth instar larvae droplet-fed delivery buffer (DB) solutionscontaining 3 mg of dsRNA from either SfT6, gusA or DB only. Barsshow the means � SE of two independent experiments. In eachexperiment, 24 larvae were used per treatment. The means werecompared using a Kruskal–Wallis test with Dunn’s post-test. The Hstatistic was 61.09. Treatments not sharing a common letter weresignificantly different (P < 0.05).

Fig. 6. SDS-12% PAGE analysis of Cry1Ca1 products from theprotoxin activation process with 0.5 mg of S. frugiperda midgut juiceproteins from larvae fed with: gusA-dsRNA (lane 1), SfT6-dsRNA(lane 2) and buffer only (lane 3). Sizes are expressed in kDa. PT,protoxin only; MW, broad-range molecular weight marker (NEB).



(Fig. 4). We infer RNAi by feeding would provide withimmediate access to SfT6-producing midgut cells of ahigh amount of specific dsRNA to trigger the RNAimachinery. In our study, a 24 h starvationtreatment of newly molted fourth instar larvae previousto droplet feeding was shown to increase in vitro thestability of dsRNA in midgut juice. In Lepidoptera,dsRNA alkaline nucleases are restricted to the gut juicewhere they are produced at high levels only during thefeeding periods mainly for digestive purpose but theycould also play a defensive role against viral dsRNA(Koga et al., 1969). In previous studies, RNAi by feedingin third instar larvae of Epiphyas postvittana and Plutellaxylostella was preceded by a 24 h and 48 h starvationtreatment respectively (Turner et al., 2006; Bautista et al.,2009). Starvation has also been shown to facilitatedsRNA uptake in several Drosophila cell lines (Clemenset al., 2000). Spodoptera frugiperda newly molted firstand second instars were affected with the starvation treat-ment and it was not possible to assay the effects ofSfT6-knockdown on these highly Cry1Ca1-susceptiblelarval stages. Further purification of S. frugiperda midgutdsRNA nucleases will allow the investigation of their enzy-matic properties to be used in the design of RNAi experi-ments in younger larval instars.

In the in vitro processing assays described in thispaper, SfT6-silencing could not prevent that a smallamount of the 58 kDa activated Cry1Ca1 form was pro-duced after 1 h of incubation with corresponding midgutjuice. The reasons for this partial protoxin activation couldbe SfT6 that is still present at low levels in SfT6-knockdown insects but also the sf1f04035-3-1 andsf1f02950-3-1 trypsins that were not targeted by the RNAitreatment (Fig. S2). Proteolytic activation of Cry protoxinsis known to occur in a processive manner where morethan one midgut protease with different specificity isinvolved acting at different level (Schnepf et al., 1998).Though we cannot entirely exclude the latter two trypsinsfrom participating in Bt Cry1Ca1 processing and toxicity,the fact that only SfT6 transcripts dropped after protoxinexposition and that SfT6-knockdown affected both pro-toxin proteolysis and insect susceptibility confirms a spe-cific role for this particular serine protease during theintoxication pathway of Cry1Ca1 in S. frugiperda. SfT6specific downregulation could represent a putative defen-sive mechanism to reduce active toxin amounts reachingtarget cells providing with a survival advantage for insectsfeeding on Cry1Ca1-contaminated diets. Unprocessedprotoxins cannot pass the peritrophic matrix and can berapidly eliminated by the insect during the gut clearanceprocess.

That reduction of SfT6 transcription affects Cry1Ca1protoxin activation and toxicity against S. frugiperdaraises the possibility for a Bt-resistance mechanism

based on alterations in SfT6 gene expression. Accordingto the phylogenetic analysis described in this paper, SfT6orthologs in P. interpunctella and Ostrinia nubilalis arePiT2a and OnT25 respectively. The expression of thesetwo genes has been investigated in Bt-resistant insectstrains. In the first case, Zhu and colleagues (2000)reported major differences between Bt-susceptible andBt-resistant P. interpunctella strains in the expressionrates of trypsin genes (PiT) using quantitative RT-PCR. Infourth instar larvae of the resistant HD198r strain, themRNA levels of 25 kDa PiT2a trypsinogen were 1.6-foldlower than in the susceptible unselected RC688s strain.Whereas, Cry1Ab resistance in O. nubilalis was not attrib-uted to OnT25 but to OnT23 trypsin gene that was foundat lower levels in the resistant strain (Li et al., 2005).These authors reported that observed differences inOnT25 expression between Bt-resistant and susceptiblestrains were not statistically significant.

Taken together, the results here demonstrate that adecrease in SfT6 expression levels leads to a reducedCry1Ca1 protoxin-activating ability of the S. frugiperdalarval gut juice and therefore reduced susceptibility ofinsects in bioassays. The specific downregulation of SfT6in response to Bt Cry1Ca1 intoxication together withslower protoxin processing and decreased susceptibilityof SfT6-knockdown larvae confirmed a major role for thistrypsin-like protease during the toxic pathway of Cry1Ca1and its potential for becoming a target in a proteinase-mediated Bt-resistance mechanism. Experiments are inprogress to obtain a Cry1Ca1-resistant S. frugiperdastrain to examine the expression of SfT6 for a moreprecise definition of its role in resistance.

Experimental procedures

Insects, total RNA isolation, Bt toxin preparation

A population of S. frugiperda has been maintained at CIGB,Havana for more than 50 generations. Larvae were reared onartificial diet based on maize flour (Ayala and Armas, 1987) at30°C, 80% RH and a 16:8 h light : dark cycle.

For total RNA isolation, samples of 15 mg were taken fromeggs, neonates, larval midgut, pre-pupae, pupae and adultabdominal cavity. When dissection was needed, insectsamples were pre-chilled on ice during 10 min. RNA wasisolated using the spin or vacuum (SV) total RNA isolationsystem (Promega, Madison, WI, USA). RNA quality andintegrity was verified by 1% agarose gel electrophoresis.RNA concentration was determined at 260 nm in aGeneQuant (Amersham Pharmacia, Amersham, UK) andsubsequently adjusted to a final concentration of 1 mg ml-1.

Parasporal crystals of the Cry1Ca1 Bt protoxin were pre-pared (Ayra–Pardo et al., 2007) and solubilized in 50 mMNa2CO3 – 10 mM dithiothreitol (DTT) (pH 10.5) at 37°C for60 min. Solubilized Cry1Ca protoxin was run on SDS – 10%PAGE (Laemmli and Favre, 1973) and shown by CoomassieBrilliant Blue staining to give a major product of c. 130 kDa.



The amount of the Cry1Ca1 protoxin was quantitated byscanning densitometry.

Cloning of the SfT6 gene and sequence analysis

The cDNA sequence of EST-SFFC3 was used for cloningthe full-length ORF of SfT6 with a BD SMART RACE cDNAAmplification Kit (Clontech, Palo Alto, CA, USA) followingthe protocol provided by the manufacturer. Briefly, 5′- and3′-RACE-ready first-strand cDNAs were produced inreverse transcription reactions from isolated total RNA oflarval midguts and used to PCR amplify overlapping 5′- and3′-RACE end fragments. Gene-specific primers (GSP) weredesigned based on the EST-SFFC3 nucleotide sequence.The primers GSP1-SfT for 5′-RACE PCR and GSP2-SfTfor 3′-RACE PCR were 5′-ACCACCAGAGTCACCCTGGCACTGG-3′ and 5′-GAAACAACTACGCTACCCGCGG-3′respectively. For cloning 5′- and 3′-RACE products, eachfragment was purified using the PCR Clean-Up System(Promega), ligated into the pGEM-T Easy vector (Promega)and the resultant recombinant plasmids DNA transformedinto high efficiency JM109 E. coli competent cells(Promega). Recombinant colonies were randomly pickedand cultured in Luria–Bertani medium containing ampicillinat 100 mg l-1, followed by plasmid DNA extraction using theWizard Plus SV Minipreps DNA Purification System(Promega). Characterization of cloned fragments wasaccomplished by PCR using GSP1-SfT and GSP2-SfTprimers together in the same reaction. Several PCR-positiveclones were DNA sequenced for each RACE fragment. Thefull-length ORF of SfT6 was amplified from larval midgutcDNA by PCR using primers (F) 5′-ATGCGTGTCCTCGCTTGCTTGG-3′ and (R) 5′-CTTATGCATTGGAAGAGATCCAGGCAG-3′. The PCR fragment was cloned into thepGEM-T Easy vector to produce the pGEM-SfT6 and DNAsequenced.

Multiple sequence alignments and phylogenetic recon-structions were performed using the Phylogeny.fr platform(http://www.phylogeny.fr/; Dereeper et al., 2008). The pre-dicted amino acid sequence of SfT6 gene was compared withthose in public databases and all sequences with significantrelatedness were identified. In order to simplify the phyloge-netic reconstruction, a total of 17 insect sequences moreclosely related of those identified by BLAST search were setup in the final analysis. The presence of a signal peptide waspredicted using Signal P program (Bendtsen et al., 2004) andremoved from the sequences for a better alignment.Sequences of mature proteins were aligned with Muscle vs.3.7 (Edgar, 2004) and ambiguous positions removed withGblocks (vs. 0.91b; Castresana, 2000) with a minimum blockof five and allowed gap positions equal to half. The phyloge-netic tree was reconstructed using the maximum-likelihoodmethod implemented in the PhyML program (vs. 3.0 aLRT;Guindon and Gascuel, 2003) together with bootstrap analysisusing 100 replicates. Graphical representation and edition ofthe phylogenetic tree were performed with TreeDyn vs. 198.3(Chevenet et al., 2006).

Toxicity assays

For Bt Cry1Ca1 toxicity against different larval stages, dupli-cate diet overlay bioassays were carried out using six toxin

concentrations and 40 newly molted larvae for each concen-tration and the control. Assay plates were incubated for 8days at 30°C, 80% RH and a photoperiod of 16:8 (light/dark).Mortality and molting inhibition were determined after 8 days,with mortality recorded as larvae that failed to respond whenprodded with a blunt, sterile probe and molt inhibitionrecorded as larvae that failed to molt to the next larval instarincluding dead larvae. The effective dose estimates (LC50 andMIC50) were calculated with PROBIT (Finney, 1952).

For SfT6 function analysis, a GI assay was settled usingdsRNA-treated and untreated fourth instar larvae 48 h afterdroplet feeding. Pre-weighed larvae from each treatmentwere allowed to eat on diet surface-contaminated or not with2 mg cm-2 of Cry1Ca1 protoxin, a dose equivalent to LC25 andMIC70 of this protoxin against fourth instar stage (Table 1).Every test was performed with 24 droplet-fed larvae/RNAitreatment in two independent replicates. Larvae werere-weighed 48 h later and the percentage of GI calculated as:%GI = [1 – (RGt/RGc)] x 100, where RGt and RGc representthe relative growth in the presence of toxin (RGt) or in thecontrol (RGc) respectively. Relative growth (RG) was calcu-lated as RG = [(W1 – W0)/W0], where W0 and W1 are theinitial and final weight of the larva respectively (Herrero et al.,2002). The %GI means among treatments were compared bya Kruskal–Wallis test with Dunn’s post-test using the Graph-Pad Prism software vs. 4.00 for Windows (San Diego, CA,USA).

Expression analysis by RT-PCR

The SfT6 mRNA abundance was estimated by RT-PCR indifferent life stages of S. frugiperda and in midguts fromlarvae exposed to 2 mg Cry1Ca1 protoxin cm-2 (section 2.7).Midguts were pooled from batches of three larvae dissectedat 0, 0.25, 1.5 and 24 h post intoxication respectively. Aprimer set was designed to anneal and produce cDNA ampli-cons from spliced mRNAs, avoiding any amplification fromcontaminating genomic DNA. To this end, the structuralgenomic organization of the partial 5′-end region of SfT6 wasdetermined by sequencing purified S. frugiperda genomicDNA (Promega Wizard SV genomic DNA purification system)using the reverse primer 5′-AAGTTCCACGAGTACAGCAGAG-3′. The sequenced DNA fragment (GenBank acces-sion no. GQ374470; 280 bp) comprised the first intron andthe two adjacent exons in the genomic sequence of SfT6. Aforward primer (5′-CTCTTAGCTGTGGTAGCAGCCG-3′)was then designed that hybridize to the 3′-end of the firstexon and to the first three nucleotides of the 5′-end ofthe subsequent exon. Together, forward and reverseprimers amplified a cDNA fragment of 113 bp and wereused in RT-PCR of total RNA samples (1 mg) with the aid ofthe AccessQuick RT-PCR system (Promega). For transcrip-tion analysis of SPODOBASE S. frugiperda trypsinssf1f02950-3-1, sf1f04035-3-1 and sf1f04740-5-1 under thethree different post-intoxication time, amplicons of ~150 bpwere expected by using gene-specific RT-PCR primers (F)5′-CTAGCGTAGAAATTCTGGAG-3′ (R) 5′-TACAAGGGACCTCCGAAGTC-3′ (F) 5′-GAAGACCCAGCCCGGCTACC-3′(R) 5′-AAGCCCCAGGAGACGACACC-3′ and (F) 5′-GGAGTCTGGACTGGTGTCTGG-3′ (R) 5′-GGTGTTACGGTAGGAAGATCC-3′ respectively. As a control, RT-PCR of a



http://www.phylogeny.fr/

150 bp internal fragment of b-actin, primers (F) 5′-CGGTATCGTGCTGGACTCCGGTG-3′ and (R) 5′-GAGTAACCCCTCTCGGTGAGGATC-3′ was used. Reaction productswere run in agarose gels and visualized using ethidiumbromide (0.5 mg l-1).

dsRNA synthesis

To generate a SfT6 template for dsRNA synthesis, a fragmentcomprising the first 567 bp of the ORF was PCR amplifiedfrom pGEM-SfT6 (section 2.2) with primers (F) 5′-ATGCGTGTCCTCGCTTGCTTGG-3′ and (R) 5′-GGTGATACCGCGGGTAGCGTAG-3′, and subcloned into the uniqueEcoRV restriction site of the multiple cloning region of Litmus29 plasmid (New England BioLabs, Beverly, MA, USA). ForRNAi control experiments with nonspecific dsRNA, a frag-ment from the E. coli b–D–glucuronidase (uidA) gene(GenBank accession no. CP001509) spanning 58–669nucleotides was produced by PCR with primers5′-GCATTCAGTCTGGATCGCGAAAAC-3′ (forward) and5′-CACCTGTTGATCCGCATCACGCAG-3′ (reverse) andsubcloned into the EcoRV-digested Litmus 29 vector. Thesource for uidA gene was the pPIC-GUS plasmid (Ayra-Pardo et al., 1998). Since the Litmus 29 polylinker region isflanked by two T7 promoters in opposite orientations, single-strand sense RNA (sRNA) and antisense RNA (asRNA) ofcloned fragments were generated in vitro using the T7RiboMAX Express RNAi System (Promega). Two different T7RNA Pol templates, for each Litmus 29-SfT6 and Litmus29-Gus, were prepared by linearizing with XhoI and XbaIrestriction endonucleases that do not cut SfT6 or uidA inserts,purified using the PCR Clean-Up System (Promega) andtranscribed in separated reactions. To produce dsRNA, equalamounts of corresponding sRNA and asRNA were mixed,heated to 65°C, and annealed by slow cooling over 4 h fol-lowed by DNase treatment for 15 min at 37°C. The dsRNAwere then precipitated with ice-cold ethanol in the presenceof 0.3 M sodium acetate, pH 5.4. The dsRNA pellets werewashed with 75% ethanol and resuspended in DEPC-treatedwater to a final concentration of 5 mg ml-1. The dsRNA wasquantified at 260 nm. The dsRNA were stored at –20°C untiluse.

dsRNA injections and feeding

Seventy-two early fourth instar larvae were immobilized at4°C to a piece of Styropor with two crossed needles at eachlarval edge. In this position, larvae were injected once with a10 ml Hamilton microsyringe, needle G 30 (Hamilton AG,Bonaduz, Swiss) into the base of a pseudopodium with 1 ml ofspecific or nonspecific dsRNA solution (5 mg ml-1, N = 24each) or 1 ml nuclease free water (N = 20) for the control.Individuals showing sign of hemolymph leakage after with-drawal of the needle were discarded. The larvae stayed in thefixed position for 1 h before placing back to the growingchamber into individual wells of 24-well plates provided ofartificial diet. Knockdown analysis by quantitative real-timeRT-PCR was conducted 48 h post injection. The experimentwas repeated twice.

For dsRNA feeding, three groups of 24 early fourth instarlarvae were used. Larvae were individually placed into each

well of 24-well plates to avoid cannibalism and starved for24 h. Two groups of larvae were then fed with a 5 ml drop of3 mg dsRNA from either SfT6 or uidA, respectively, in deliverybuffer (DB, 10 mM Tris-Cl, pH 7.5; 10 mM EDTA). The thirdgroup of larvae were fed with DB as a control. After 3 h,droplet-fed larvae were placed back into a well of a 24-wellplate provided of artificial diet. The experiment was repeatedtwice.

qRT-PCR of RNAi experiments

Midgut SfT6 mRNA levels were investigated 48 h afterdsRNA administration by qRT-PCR using a QuantiTectSYBR Green RT-PCR kit (QIAGEN GmbH, Hilden,Germany) and a Rotor-Gene 3000 real-time cycler (CorbettResearch, Sydney, Australia). Larvae were dissected andmidguts pools of 10 larvae from each RNAi treatments wereused to isolate total RNA. The forward and reverse primersmentioned earlier and used for SfT6 RT-PCR were used forthe analysis. The real-time one-step amplification andanalysis was performed in triplicates in PCR reactions of25 ml final volumes containing 7.5 pmol of each primers and100 ng of total RNA as template. The b-actin amplicon wasobtained by using b-actin primers and used as internal ref-erence to normalize the SfT6 transcripts in dsRNA-treatedand control samples. The amplification efficiencies of bothSfT6 and reference genes were determined with the aid ofstandard curves generated by serial dilutions of correspond-ing cDNAs. The relative levels of SfT6 mRNA in dsRNA-treated and control groups were expressed as ‘MeanNormalized Expression’ data using Q-Gene software(http://www.gene-quantification.de/qgene.zip; Muller et al.,2002).

In vitro processing of Cry1Ca1 protoxin

Midgut juice isolation and in vitro processing of Bt Cry1Ca1protoxin were essentially as previously described (Ayra-Pardo et al., 2007). Solubilized protoxin (10 mg) in 50 ml of thesolubilization buffer was incubated at 37°C for 60 min in thepresence of 0.3 mg of midgut juice proteins (0.5 mg ml-1) fromdroplet-fed larvae of the different RNAi treatments. Proteoly-sis was stopped by adding phenylmethanesulfonyl fluoride(1 mM, final concentration), and the samples were centri-fuged at 20 000 r.p.m. for 20 min. In order to determine theextent of toxin proteolysis, 10 ml samples from the solublefractions were subjected to SDS-12% PAGE. Gels were visu-alized with Coomassie Brilliant Blue staining.

Acknowledgements

This work was partially supported by Grants C/3579-2 fromthe International Foundation for Sciences (IFS, Stockholm,Sweden) and 3031-243 from the Cuban Ministry of Scienceand Technology (CITMA). The authors are grateful to Ms Y.Fernández and Mr R. Maura for insect manipulation andmicroinjection and to Ms M. Ponce for her excellent technicalassistance. The authors thank to anonymous reviewers andeditor for useful suggestion.



http://www.gene-quantification.de/qgene.zip

References

Ayala, J., and Armas, J. (1987) Metodología para la críacontinua de Spodoptera frugiperda J.E Smith (Lep., Noc-tuidae). Centro Agrícola 14: 88–90.

Ayra–Pardo, C., Montejo–Sierra, I.L., Vázquez–Padrón, R.I.,and García–Martínez, C. (1998) b-D-glucuronidase genefrom Escherichia coli is a functional reporter in the methy-lotrophic yeast Pichia pastoris. Lett Appl Microbiol 29: 278–283.

Ayra–Pardo, C., Davis, P., and Ellar, D.J. (2007) The muta-tion R423S in the Bacillus thuringiensis hybrid toxin CryAACslightly increases toxicity for Mamestra brassicae L.J Invertebr Pathol 95: 41–47.

Bautista, M.A., Miyata, T., Miura, K., and Tanaka, T. (2009)RNA interference-mediated knockdown of a cytochromeP450, CYP6BG1, from the diamondback moth, Plutellaxylostella, reduces larval resistance to permethrin. InsectBiochem 39: 38–46.

Bellés, X. (2010) Beyond Drosophila: RNAi in vivo and func-tional genomics in insects. Annu Rev Entomol 55: 111–128.

Bendtsen, J.D., Nielsen, H., von Heijne, G., and Brunak, S.(2004) Improved prediction of signal peptides: SignalP 3.0.J Mol Biol 340: 783–795.

Bown, D.P., Wilkinson, H.S., and Gatehouse, J.A. (1997)Differentially regulated inhibitor-sensitive and insensitiveprotease genes from the phytophagous insect pest, Heli-coverpa armigera, are members of complex multigenefamilies. Insect Biochem 27: 625–638.

Bravo, A., Sanchez, J., Kouskoura, T., and Crickmore, N.(2002) N-terminal activation is an essential early step in themechanism of action of the Bacillus thuringiensis Cry1Acinsecticidal toxin. J Biol Chem 277: 23985–23987.

Bravo, A., Gómez, I., Conde, J., Muñoz-Garay, C., Sánchez,J., Miranda, R., et al. (2004) Oligomerization triggersbinding of a Bacillus thuringiensis Cry1Ab pore-formingtoxin to aminopeptidase N receptor leading to insertion intomembrane microdomains. Biochim Biophys Acta 1667:38–46.

Brioschi, D., Nadalini, L.D., Bengtson, M.H., Sogayar, M.C.,Moura, D.S., and Silva-Filho, M.C. (2007) General up regu-lation of Spodoptera frugiperda trypsins and chymotrypsinsallows its adaptation to soybean proteinase inhibitor. InsectBiochem 37: 1283–1290.

Candas, M., Loseva, O., Oppert, B., Kosaraju, P., and Bulla,L.A., Jr (2003) Insect resistance to Bacillus thuringiensis:alterations in the indianmeal moth larval gut proteome. MolCell Proteomics 2: 19–28.

Castresana, J. (2000) Selection of conserved blocks frommultiple alignments for their use in phylogenetic analysis.Mol Biol Evol 17: 540–552.

Chevenet, F., Brun, C., Bañuls, A.L., Jacq, B., and Christen,R. (2006) TreeDyn: towards dynamic graphics and anno-tations for analyses of trees. BMC Bioinformatics 7:439.

Clemens, J.C., Worby, C.A., Simonson-Leff, N., Muda, M.,Maehama, T., Hemmings, B.A., et al. (2000) Use of double-stranded RNA interference in Drosophila cell lines todissect signal transduction pathways. Proc Natl Acad SciUSA 97: 6499–6503.

Dereeper, A., Guignon, V., Blanc, G., Audic, S., Buffet, S.,Chevenet, F., et al. (2008) Phylogeny.fr: robust phyloge-netic analysis for the non-specialist. Nucleic Acids Res 36:W465–W469.

Edgar, R.C. (2004) MUSCLE: a multiple sequence alignmentmethod with reduced time and space complexity. BMCBioinformatics 5: 113.

Ferré, J., and Van Rie, J. (2002) Biochemistry and genetics ofinsect resistance to Bacillus thuringiensis. Annu RevEntomol 47: 501–533.

Ferreira, C., Capella, A.N., Sitnik, R., and Terra, W.R. (1994a)Digestive enzymes in midgut cells, endo- and ectoper-itrophic contents and peritrophic membranes ofSpodoptera frugiperda (Lepidoptera) larvae. Arch InsectBiochem 26: 299–313.

Ferreira, C., Capella, A.N., Sitnik, R., and Terra, W.R. (1994b)Properties of the digestive enzymes and the permeability ofthe peritrophic membrane of Spodoptera frugiperda (Lepi-doptera) larvae. Comp Biochem Physiol 107A: 631–641.

Finney, D.J. (1952) In Probit Analysis, 2nd ed. Cambridge,UK: Cambridge University Press.

Forcada, C., Alcácer, E., Garcerá, M.D., and Martínez, R.(1996) Differences in the midgut proteolytic activity of twoHeliothis virescens strains, one susceptible and one resis-tant to Bacillus thuringiensis toxins. Arch Insect BiochemPhysiol 31: 257–272.

Forcada, C., Alcacer, E., Garcerá, M.D., Tato, A., and Mar-tínez, R. (1999) Resistance to Bacillus thuringiensisCry1Ac toxin in three strains of Heliothis virescens: pro-teolytic and SEM study of the larval midgut. Arch InsectBiochem 42: 51–63.

Griebler, M., Westerlund, S.A., Hoffmann, K.H., andMeyering–Vos, M. (2008) RNA interference with the allato-regulating neuropeptide genes from the fall armywormSpodoptera frugiperda and its effects on the JH titer in thehemolymph. J Insect Physiol 54: 997–1007.

Guindon, S., and Gascuel, O. (2003) A simple, fast andaccurate algorithm to estimate large phylogenies bymaximum likelihood. Syst Biol 52: 696–704.

Heckel, D.G., Gahan, L.J., Baxter, S.W., Zhao, J.-Z., Shelton,A.M., Gould, F., et al. (2007) The diversity of Bt resistancegenes in species of Lepidoptera. J Invert Pathol 95: 192–197.

Herrero, S., Oppert, B., and Ferré, J. (2001) Different mecha-nisms of resistance to Bacillus thuringiensis toxins in theindianmeal moth. Appl Environ Microbiol 67: 1085–1089.

Herrero, S., Borja, M., and Ferré, J. (2002) Extent of variationof the Bacillus thuringiensis toxin reservoir: the case of thegeranium bronze, Cacyreus marshalli Butler (Lepidoptera:Lycaenidae). Appl Environ Microbiol 68: 4090–4094.

Khajuria, C., Zhu, Y.C., Chen, M.S., Buschman, L.L., Higgins,R.A., Yao, J., et al. (2009) Expressed sequence tags fromlarval gut of the European corn borer (Ostrinia nubilalis):exploring candidate genes potentially involved in Bacillusthuringiensis toxicity and resistance. BMC Genomics 10:286.

Koga, K., Hara, A., and Satoru, A. (1969) Correlation ofnucleolytic activities with the stage of growth of the silk-worm. J Insect Physiol 15: 799–806.

Laemmli, U.K., and Favre, M. (1973) Maturation of the headof bacteriophage T4. J Mol Biol 80: 575–599.



Li, H., Oppert, B., Higgins, R.A., Huang, F., Buschman, L.L.,Gao, J.R., et al. (2005) Characterization of cDNAs encod-ing three trypsin-like proteinases and mRNA quantitativeanalysis in Bt-resistant and -susceptible strains of Ostrinianubilalis. Insect Biochem 35: 847–860.

Lightwood, D.J., Ellar, D.J., and Jarrett, and P. (2000) Role ofproteolysis in determining potency of Bacillus thuringiensisCry1Ac d-endotoxin. Appl Environ Microbiol 66: 5174–5181.

Loseva, O., Ibrahim, M., Candas, M., Koller, C.N., Bauer,L.S., and Bulla, L.A., Jr (2002) Changes in protease activityand Cry3Aa toxin binding in the Colorado potato beetle:implications for insect resistance to Bacillus thuringiensistoxins. Insect Biochem 32: 567–577.

MacIntosh, S.C., Stone, T.B., Sims, S.R., Hunst, P.L., Green-plate, J.T., Marrone, P.G., et al. (1990) Specificity and effi-cacy of purified Bacillus thuringiensis proteins againstagronomically important insects. J Invertebr Pathol 56:258–266.

Meyering–Vos, M., Merz, S., Sertkol, M., and Hoffmann, K.H.(2006) Functional analysis of the allatostatin-A type gene inthe cricket Gryllus bimaculatus and the armywormSpodoptera frugiperda. Insect Biochem Mol Biol 36: 492–504.

Muller, P.Y., Janovjak, H., Miserez, A.R., and Dobbie, Z.(2002) Processing of gene expression data generated byquantitative real-time RT-PCR. BioTechniques 32: 1372–1380.

van Munster, M., Prefontaine, G., Meunier, L., Elias, M.,Mazza, A., Brousseau, R., et al. (2007) Altered geneexpression in Choristoneura fumiferana and Manducasexta in response to sublethal intoxication by Bacillus thu-ringiensis Cry1Ab toxin. Insect Mol Biol 16: 25–35.

Nègre, N., Hotelier, T., Volkoff, A.N., Gimenez, S., Cousser-ans, F., Mita, K., et al. (2006) SPODOBASE: an EST data-base for the lepidopteran crop pest. Spodoptera BMCBioinf 7: 322.

Oppert, B. (1999) Protease interactions with Bacillus thuring-iensis insecticidal toxins. Arch Insect Biochem 42: 1–12.

Oppert, B., Kramer, K.J., Johnson, D.E., MacIntosh, S.C.,and McGaughey, W.H. (1994) Altered protoxin activationby midgut enzymes from a Bacillus thuringiensis resistantstrain of Plodia interpunctella. Biochem Biophys ResCommun 198: 940–947.

Oppert, B., Kramer, K.J., Beeman, R.W., Johnson, D., andMcGaughey, W.H. (1997) Proteinase-mediated insectresistance to Bacillus thuringiensis toxins. J Biol Chem272: 23473–23476.

Rodríguez-Cabrera, L., Trujillo-Bacallao, D., Borrás-Hidalgo,O., Wright, D.J., and Ayra-Pardo, C. (2008) Molecular char-acterization of Spodoptera frugiperda–Bacillus thuringien-sis Cry1Ca toxin interaction. Toxicon 51: 681–692.

Sayyed, A.H., Gatsi, R., Kouskoura, T., Wright, D.J., andCrickmore, N. (2001) Susceptibility of a field-derived, Bacil-lus thuringiensis-resistant strain of diamondback moth to invitro activated Cry1Ac toxin. Appl Environ Microbiol 67:4372–4373.

Sayyed, A.H., Raymond, B., Ibiza-Palacios, M.S., Escriche,B., and Wright, D.J. (2004) Genetic and biochemical char-acterization of field evolved resistance to Bt toxin Cry1Ac inthe diamondback moth. Appl Environ Microbiol 70: 7010–7017.

Schnepf, E., Crickmore, N., Van Rie, J., Lereclus, D., Baum,J., Feitelson, J., et al. (1998) Bacillus thuringiensis and itspesticidal crystal proteins. Microbiol Mol Biol Rev 62: 775–806.

Turner, C.T., Davy, M.W., MacDiarmid, R.M., Plummer, K.M.,Birch, N.P., and Newcomb, R.D. (2006) RNA interference inthe light brown apple moth, Epiphyas postvittana (Walker)induced by double-stranded RNA feeding. Insect Mol Biol15: 383–391.

Wieczorek, H., Grüber, G., Harvey, W.R., Huss, M., andMerzendorfer, H. (1999) The plasma membrane H+-V-ATPase from tobacco hornworm midgut. J BioenergBiomembr 31: 67–74.

Zhu, Y.C., Oppert, B., Kramer, K.J., McGaughey, W.H., andDowdy, A.K. (2000) cDNA sequence, mRNA expressionand genomic DNA of trypsinogen from the indianmealmoth, Plodia interpunctella. Insect Mol Biol 9: 19–26.

Supporting information

Additional Supporting Information may be found in the onlineversion of this article:

Fig. S1. Multiple alignment of the nucleotide sequences ofSfT6 and deduced S. frugiperda trypsin-like genes from theSPODOBASE ESTs clusters sf1f04035-3-1 and sf1f02950-3-1. The alignment only comprised the SfT6 region used fordsRNA generation.Fig. S2. qRT-PCR analysis of mRNA levels from different S.frugiperda trypsins in untreated and dsRNA-SfT6 fed larvae.For each determination, midgut total RNA (100 ng) was iso-lated from a pool of 10 dissected larvae. The detected tran-script amounts were normalized to the reference gene b-actinwith the aid of Q-gene software. Bars represent means of twoindependent replicates � SE.

Please note: Wiley-Blackwell are not responsible for thecontent or functionality of any supporting materials suppliedby the authors. Any queries (other than missing material)should be directed to the corresponding author for thearticle.



ENVIROMENTAL MICROBIOLOGY (2010) RNAi mediated knockdown of a Spodoptera frugiperda trypsin like...

Documents

Transcript of ENVIROMENTAL MICROBIOLOGY (2010) RNAi mediated knockdown of a Spodoptera frugiperda trypsin like...