CopE and TLR6 RNAi-mediated tomato resistance to western flower thrips

Jelli VENKATESH ,Sung Jin KlM ,Muhammad lrfan SlDDlQUE ,Ju Hyeon KlM ,Si Hyeock LEE ,Byoung-Cheorl KANG

1 Department of Agriculture,Forestry and Bioresources,Research Institute of Agriculture and Life Sciences,Plant Genomics Breeding Institute,College of Agriculture and Life Sciences,Seoul National University,Seoul 08826,Republic of Korea

2 Department of Agricultural Biotechnology,Seoul National University,Seoul 08826,Republic of Korea

3 Vegetable Research Division,National Institute of Horticulture and Herbal Science,Rural Development Administration,Jeonju 55365,Republic of Korea

4 Research Institute of Agriculture and Life Sciences,Seoul National University,Seoul 08826,Republic of Korea

Abstract The western flower thrips (WFT;Frankliniella occidentalis) is a mesophyll cell feeder that damages many crops.Management of WFT is complex due to factors such as high fecundity,short reproduction time,ability to feed on a broad range of host plants,and broad pesticide resistance.These challenges have driven research into developing alternative pest control approaches for WFT.This study analyzed the feasibility of a biological control-based strategy to manage WFT using RNA interference (RNAi)-mediated silencing of WTF endogenous genes.For the delivery of RNAi,we developed transgenic tomato lines expressing double-stranded RNA (dsRNA) of coatomer protein subunit epsilon (CopE) and Toll-like receptor 6 (TLR6) from WFT.These genes are involved in critical biological processes of WFT,and their dsRNA can be lethal to these insects when ingested orally.Adult WFT that fed on the transgenic dsRNAexpressing tomato flower stalk showed increased mortality compared with insects that fed on wild-type samples.In addition,WFT that fed on TLR6 and CopE transgenic tomato RNAi lines showed reduced levels of endogenous CopE and TLR6 transcripts,suggesting that their mortality was likely due to RNAi-mediated silencing of these genes.Thus,our findings demonstrate that transgenic tomato plants expressing dsRNA of TLR6 and CopE can be lethal to F. occidentalis,suggesting that these genes may be deployed to control insecticide-resistant WFT.

Keywords: coatomer protein subunit epsilon (CopE),Frankliniella occidentalis,insect resistance, RNA interference,Tolllike receptor 6 (TLR6),tomato,transgenics

1.lntroduction

The western flower thrips (WFT),Frankliniella occidentalis,is a major insect pest that feeds on the field and horticultural crops worldwide and causes billions of dollars in crop damage (De Kogelet al.2015).WFT injure host plants by feeding on leaves and flowers using their strawshaped rasping-sucking-type mouthparts.They also act as vectors for destructive plant viruses such astomato spotted wilt virus(TSWV) andimpatiens necrotic spot virus(INSV) (Yudinet al.1986;Kirk and Terry 2003;Menget al.2018).Several insecticides have been used to controlWFT.However,due to its persistent nature,high reproduction rate,and short reproduction time,WFThas developed strong resistance to a broad range of insecticides (Espinosaet al.2002;Gaoet al.2012;Reitzet al.2020).Chemical control ofWFTcurrently yields only limited success.The use of comprehensive,integrated pest management (IPM) strategies,including crop rotation,gene pyramiding,refuge strategy,and other alternative management strategies,is essential for delaying insect resistance evolution through reduced selection pressure.

Recent European legislation has recognized the importance of sustaining crop ecosystems and implemented a remarkable change in pest management strategies to rely less on chemicals (Moudenet al.2017).Therefore,developing alternative pest management strategies is increasingly important to crop production(Wimmer 2005;Barzmanet al.2015;Bingsohnet al.2017;Moudenet al.2017).The use of transgenic plants expressing insect-lethal genes is considered a promising approach for insect pest management (Ferryet al.2006;Youniset al.2014).Enhanced resistance to WFT in crops such as tomato,potato,chrysanthemum,and alfalfa has been achieved through the introduction of serval insecticidal transgenes,including multidomain protease inhibitors,linalool synthase,and MLX56 defense protein (Thomaset al.1994;Outchkourovet al.2004;Yanget al.2013;Murataet al.2021).However,the potential interference of protease inhibitors with basic cell functions has limited their practical application for pest management.Another potential approach is targeting essential insect genes using RNA interference (RNAi),a post-transcriptional gene regulatory system in eukaryotes where a target gene is silenced in a sequence-specific manner.The knockdown of essential genes through RNAi offers an attractive control strategy for thrips,especially as part of an IPM approach,because of greater specificity than synthetic insecticides (Whyardet al.2015;Whitten and Dyson 2017).The insect midgut is designed to facilitate nutrient absorption with its large absorption area made up of microvilli with several active transporters,carriers,channels,and endocytosis apparatus (Hakimet al.2010;Deneckeet al.2018).These features make the insect midgut an important target as a potential site of double-stranded RNA (dsRNA) uptake.Indeed,RNAibased plant transgenesis,includingAgrobacteriummediated transient or stable expression of dsRNA,has been attempted as a pest control strategy mainly against Coleopteran,Hemipteran,and Lepidopteran species(Baumet al.2007;Maoet al.2007;Huvenne and Smagghe 2010;Zhaet al.2011;Xionget al.2013).Only a few studies have investigated the feasibility of using the RNAi approach for Thysanopteran pest control (Badillo-Vargaset al.2015;Whittenet al.2016).

In a previous study,external administration of dsRNA of several WTFs causing insect and mite mortality was identified,and the non-transgenic approach involved a leaf disc-mediated dsRNA delivery system,in which WFT ingested target dsRNAs along with leaf disc sap and cells(Hanet al.2019).A total of 57 genes were analyzed for WFT resistance,which determined that the dsRNA ofTolllike receptor 6(TLR6),apolipophorin(apoLp),coatomer protein subunit epsilon(CopE),andsorting and assembly machinery component 50(SAM50) resulted in high WFT mortality (Hanet al.2019).However,this approach is not feasible for practical field applications due to the limitations of large-scale production and stability maintenance of dsRNAs.Therefore,stable expression of dsRNA in plant cellsviatransgenesis is hypothesized as an efficient and reliable RNAi-based control method for WFT.

TLRs are type I membrane proteins that play an important role in the induction of innate immune responses and triggering of adaptive immune responses against microbial pathogens in various animals,including mice and insects (Tausziget al.2000;Weberet al.2003;Kawai and Akira 2007).Coatomer protein complex I(COP-I) plays an important role in vesicular transport between Golgi and endoplasmic reticulum (ER) and maintains Golgi-and ER-localized enzymes and chaperon proteins.Previous studies also highlighted COP-Is’ role in endosomal transport,lipid droplet homeostasis regulation,mRNA molecule transport,and nuclear envelope breakdown (Songet al.2021).Mutations in COP-I subunit proteins lead to abnormal trafficking from the Golgi to the ER and interferon signaling (Volpiet al.2018;Lepelleyet al.2020) and cause immune dysregulation (Volpiet al.2018).TheCopEgene encodes the epsilon (ε) subunit of COP-I,which can affect viral replication (Chenet al.2021).However,studies elucidating the roles ofTLR6andCopEin WTFs immune responses and growth and development have not been functionally characterized.

The present study designed inverted-repeat (IR) binary constructs that encode dsRNAs corresponding to the two key WFT genes,CopEandTLR6,to deliver RNAi.We expressedCopEandTLR6dsRNAs in transgenicSolanum lycopersicum‘Microtom’ plants,aiming to produce WFT-resistant tomato plants through the delivery of RNAi.WFT that fed on theCopEandTLR6RNAi Microtom transgenic lines showed significantly higher mortality than those that fed on wild-type control plants.Our results thus demonstrate the potential application ofCopEandTLR6dsRNA-mediated insect resistance in tomato,which may be adapted for other crop plants.

2.Materials and methods

2.1.Vector construction

TheCopEandTLR6sequences were amplified from the WFT and cloned into the pGEM-T vector (Promega,Madison,WI,USA),and the positive clones were confirmed by Sanger sequencing.To construct the RNAi vectors,we amplifiedCopEandTLR6sequences by PCR using the gene-specific primer sequences listed in Table 1.Amplicons were cloned into the entry vector pENTR/D/TOPO (Invitrogen,San Diego,CA,USA) and subsequently cloned into the pANDA destination vector by Gateway?(Invitrogen,Carlsbad,CA,USA).The genes cloned into pANDA were driven by the maizeubiquitinpromoter with the first intron andnopaline synthase terminator(nosT) (Fig.1-A).The RNAi vector harboredneomycin phosphotransferase I(NPTI) for bacterial selection using kanamycin,andneomycin phosphotransferase II(NPTII)andhygromycin phosphotransferase(Hpt) as plant selectable markers against kanamycin and hygromycin,respectively (Miki and Shimamoto 2004).The recombinant pANDA RNAi vectors carryingCopEandTLR6inverted repeats were transformed into theAgrobacterium tumefaciensstrain GV3101 through electroporation.

Fig.1 Diagram of the pANDA vector and molecular characterization of RNAi lines.A,partial diagram of the pANDA vector containing the Ubi promoter and inverted-repeat (IR) regions of CopE and TLR6 gene fragments.B,genomic DNA PCR of CopEtransgenic plants.C,genomic DNA PCR of TLR6-transgenic plants.M,molecular marker,arrows indicate 500 bp;WT,wild type;B,water blank;P,plasmid control.

Table 1 Primers used in this study

2.2.Plant material and Agrobacterium-mediated tomato transformation

Seeds ofS.lycopersicum‘Microtom’ were used for transformation.A singleAgrobacteriumcolony was inoculated into 20 mL of liquid LB medium containing 50 mg L-1kanamycin and 50 mg L-1rifampicin and incubated overnight in a shaking incubator at 28°C until the OD600reached 0.6.Then,theAgrobacteriumsuspension was centrifuged at 4 000 r min-1for 10 min with the A50T-8 rotor(Hanil SUPRA 22K,Seoul,Korea) at room temperature.The pellet was resuspended in sterile liquid 1/2 Murashige and Skoog (MS) medium containing 1.5% sucrose and 200 μmol L-1acetosyringone.Cotyledons from oneweek-old seedlings germinated and grown on 1/2 MS solid medium were excised in sterile conditions,sectioned transversely into two fragments,and incubated adaxial side down for one day on pre-culture medium (MS salts with vitamins,30 g L-1sucrose,1 mg L-11-naphthaleneacetic acid (NAA),1 mg L-1benzylaminopurine (BAP),and 8 g L-1agar,at pH 5.8).Pre-cultured explants were co-cultured in theAgrobacteriumsuspension for 20 min,then transferred to the same medium used for pre-culture and incubated for two days.Subsequently,explants were transferred to a shoot induction medium (MS with 30 g L-1sucrose,2 mg L-1trans-zeatin-riboside,0.1 mg L-1indole-3-acetic acid (IAA),20 mg L-1hygromycin,250 mg L-1carbenicillin,and 8 g L-1agar) for 3-4 weeks.Explants showing shoot induction were transferred to MS medium containing a reduced concentration oftrans-zeatin-riboside (1 mg L-1),0.1 mg L-1IAA,20 mg L-1hygromycin,and 250 mg L-1carbenicillin.Putative transgenic tomato shoots roughly 2 cm in height were cut and transferred to a rooting medium (MS with 30 g L-1sucrose,1 mg L-1IAA,10 mg L-1hygromycin,250 mg L-1carbenicillin,and 8 g L-1agar).Well-rooted plants were transferred into plastic pots containing coco peat-based potting mixture (Hanarum,Minong Fertilizer,Korea) and were kept in a growth room maintained at (24±2)°C and 16 h light/8 h dark cycle for hardening.The hardened plants were maintained under the same controlled conditions in a biosafety growth room.

2.3.Molecular characterization of transgenic tomato plants

Genomic DNA was isolated from young leaves of onemonth-old transformed plantlets and non-transformed wild-type plants using the CTAB method (Doyle and Doyle 1987).Putative transgenic plants were analyzed by polymerase chain reaction (PCR) to detect the integration of transgenes usingGUS-linker-specific (Miki and Shimamoto 2004) andCopEorTLR6-specific primers(Table 1).PCR was performed in a 25-μL reaction containing 50 ng of genomic DNA,0.2 mmol L-1dNTP mix,2 μmol L-1of each primer,one unit of homemadeTaqDNA polymerase,and PCR buffer.PCR conditions were as follows: initial denaturation at 94°C for 5 min,34 cycles of denaturation at 94°C for 30 s,annealing at 58°C for 30 s,and extension at 72°C for 45 s (GUS-linker) or 30 s(CopE/TLR6 genes),followed by a final extension at 72°C for 5 min.Agarose gel electrophoresis was performed to visualize gene-specific PCR products.

2.4.RNA isolation from tomato plants and reverse transcription (RT)-PCR

Total RNA from transgenic and WT control plants was isolated using the MG Total RNA extraction kit according to the manufacturer’s instructions (MG Med,Seoul,Korea).About 100 mg of leaf tissue was used for RNA isolation,which was then treated with DNase I and purified according to the manufacturer’s instructions (Qiagen,USA).Total RNA was quantified using a Nanodrop?ND-1000 spectrophotometer (BioTek,Winooski,VT,USA).A total of 1 μg of total RNA was used for first-strand cDNA synthesis using oligo-d(T) primers and EasyScript Reverse Transcriptase according to the manufacturer’s instructions (TransGen Biotech,Beijing,China).cDNA was diluted four times with sterile distilled water.GUSlinker-specific andCopEorTLR6-specific RT-PCR was performed in a 25-μL reaction containing 4 μL of diluted cDNA,0.2 mmol L-1dNTP mix,2 μmol L-1of each gene specific primer pair,one unit of homemadeTaqDNA polymerase,and PCR buffer.PCR conditions were the same as mentioned in the previous section.TomatoActin(SlActin) was used as the reference gene.

2.5.Northern blotting

Total microRNA was isolated from 100 mg of tomato leaf tissues using the mirPremier microRNA Isolation Kit(Sigma-Aldrich,St.Louis,MO,USA).Northern blotting was performed using the NorthernMaxTMkit (Thermo Fisher Scientific,USA) according to the manufacturer’s guidelines.TheCopEandTLR6probes used in the Northern blot analysis were synthesized by PCR using primers listed in Table 1.Purified PCR products were labeled using a Biotin DecaLabel DNA Labeling Kit(Thermo Fisher Scientific,USA).For each sample,2 μg of miRNA was separated on a denaturing 15% urea-PAGE gel and electro-blotted to a Hybond-N+membrane(Amersham Biosciences,NJ,USA).Hybridization was performed as described in the NorthernMaxTMKit(Thermo Fisher Scientific,USA).The Northern blot signal was detected using a Biotin Chromogenic Detection Kit (Thermo Fisher Scientific,USA) according to the manufacturer’s guidelines.

2.6.Thrips bioassay

An insecticide-susceptible WFT strain has been maintained on kidney bean (Phaseolus vulgaris)cotyledons,as described previously (Hanet al.2019).A flower stalk with 3-5 flowers and a few leaves from transgenic tomato expressing eitherCopEorTRL6hairpin RNA was used for the WFT bioassay.The bottom stem of the cut stalk was placed in water-soaked cotton,which was wrapped with a nitrile membrane and assembled into a screw cap of a 50-mL conical tube.Then,a bottom-cut 50 mL conical tube was assembled with the cap unit.Fifteen two-day-old female WFT were introduced into each bioassay unit,and thrips mortality was assessed every 12 h for five days (for T2generation)or every 24 h for three days (for T3generation).When observing mortality,the bioassay unit was disassembled and dead individuals were removed.The same bioassay was conducted with WT tomato as a control.Corrected mortality (%) was calculated as follows: 100×(Thrips mortality on transgenic tomato-Thrips mortality on wildtype tomato)/(100-Thrips mortality on wild-type tomato).

2.7.Analysis of CopE and TRL6 genes silencing in WFT

Following the thrips bioassay,total RNA was extracted from 10 thrips at 72 h post-treatment using TRI Reagent?(Molecular Research Center,Inc.,Cincinnati,USA)according to the manufacturer’s protocol.For the control treatment,10 thrips were collected at 72 h posttreatment and used for total RNA extraction.cDNA was synthesized from total RNA with oligo (dT)-primers using TB Green? PremixEx Taq? II (TaKaRa,Seoul,Korea),and quantitative real-time PCR (qPCR) was conducted in a 10-μL reaction mixture containing 10 ng of template cDNA,5 μmol L-1of forward and reverse primers (Table 1),and 5 μL of TB Green? PremixEx TaqII (Tli RNaseH Plus;TaKaRa,Seoul,Korea) using a LightCycler 96(Roche Inc.,Basel,Switzerland).Thermal cycling was conducted with a pre-incubation at 95°C for 1 min and 35 cycles of 95°C for 30 s,56°C for 15 s,and 72°C for 30 s.Melting curve analysis was conducted by increasing temperature from 45-95°C with a ramp rate of 0.2°C s-1.60S ribosomal protein L32(RPL32) was used as the reference gene.Relative expression levels of the target genes were calculated using the 2-ΔΔCTmethod (Pfafflet al.2001).

2.8.Statistical analysis

The bioassay was conducted using at least four replicates for each line.Data were transformed by arcsine transformation before performing the statistical analysis.Duncan’s multiple range test (P≤0.05) was performed for insect mortality.Survival curves of WFT were analyzed using PRISM GraphPad Software(version 9).Statistical comparison of the WFT survival curves was performed using the log-rank test.Six biological replicates were conducted for qPCR experiments.Duncan’s multiple range test (P≤0.05) was performed for qPCR data.

3.Results

3.1.Development and selection of WFT-resistant transgenic tomato plants

We developed two pANDA inverted-repeat (IR)constructs encoding dsRNAs corresponding to 496 and 498 bp fragments of theCopEandTLR6WFT genes,respectively.The inverted repeats were driven by the maizeubiquitin(Ubi) promoter and an intron to achieve high levels of dsRNA expression (Fig.1-A).We introduced theCopEandTLR6IR constructs into tomato usingAgrobacterium-mediated genetic transformation.

Nine and seven putative transgenicCopEandTLR6RNAi lines,respectively,were used for further molecular characterization.Transgene integration was confirmed by genomic DNA PCR usingGUS-linkerandCopEandTLR6gene-specific primers.Eight out of nineCopEand all sevenTLR6RNAi transgenic tomato lines harbored the transgene (Fig.1-B and C).GUS-linkerPCR-positive plants showed the presence of a 636-bp band on the agarose gel.Consistent with this result,all GUS-linkerpositive plants also showed the presence of 496 and 498 bp bands corresponding to theCopEandTLR6genes,respectively (Fig.1-B and C).These results indicate the presence of transgene integration into the genome and synthesis ofCopEandTLR6gene-specific transcripts.

3.2.Accumulation of CopE-and TLR6-specific siRNAs in transgenic tomato plants

Aβ-glucuronidase(GUS) gene linker of 920 bp facilitated the visualization of dsRNA synthesis in dsRNA transgenic lines (Miki and Shimamoto 2004).Therefore,to assess transgene expression in selected lines,we performed RT-PCR usingGUS-linker-specific primers.Three T1transgenic plants for each gene construct (CopEandTLR6) with the highest levels ofGUS-linkerexpression based on RT-PCR analyses were selected for Northern blot analysis (Fig.2-A).Using biotin-labeled DNA probes,we detected 21-23 nt siRNA hybridization signals in transgenic lines CP3,CP5,TL5,and TL7,indicating the expression ofCopEandTLR6dsRNA (Fig.2-B).No hybridization signal was observed in the WT plants.

Fig.2 Expression analysis of CopE and TLR6 RNAi T1 transgenic plants.A,RT-PCR of CopE and TLR6 transgenic plants.Transgene expression was confirmed by the presence of the GUS-linker amplicon.B,Northern blot analysis of CopE and TLR6 transgenic plants.For each line,2 μg of miRNA-enriched RNA was loaded.Non-transgenic wild-type (WT) plants were used as a control.P,plasmid control;B,water blank.

As revealed by RT-PCR analysis,T1plants showing the highest expression ofCopEandTLR6were self-fertilized to produce the T2generation.Leaves from the T2plants were analyzed for transgene segregation byGUS-linkerPCR.Based on transgene segregation in the T2progeny,homozygous transgenic plants were selected and selfpollinated to obtain T3seeds for use in further studies.These transgenic plants showed no obvious phenotypic differences compared with the WT tomato plants.

3.3.Enhanced resistance of transgenic CopE and TLR6 tomato plants to WFT

To determine whether the dsRNA-expressing transgenic tomato lines could cause WFT mortality,we performed WFT feeding bioassays on cut flower stalks (Fig.3-A).We infested T2/T3CopEandTLR6RNAi plants with WFT and monitored adult WFT mortality.In our initial bioassays,we used a single T2line expressing dsRNA ofCopE(line CP3) andTLR6(line TL5) along with WT plants.Compared with WT (36.4% mortality),significantly high mortality was recorded at 72 h post-feeding for WFT that fed on the CP3 (72.7%) and TL5 transgenic lines(54.5%) (Fig.3-B),though CP3 plants showed higher WFT mortality than TL5 plants.In T3bioassay analyses,two transgenic lines forCopE(lines CP3 and CP5) andTLR6(lines TL5 and TL7)were used (Fig.3-C).Similar to the T2bioassays,bothCopE(CP3 and CP5) andTLR6(TL5 and TL7) transgenic tomato plants caused increased WFT mortality.A higher WFT mortality rate was recorded inCopE-transgenic plants compared to those expressingTLR6.When WFT were fedCopE-transgenic tomato,they showed 72.7 and 78-100% mortality at 72 h post-treatment for T2and T3plants,respectively.ForTRL6-transgenic tomato,WFT mortalities were 54.5 and 77-99% for T2and T3plants,respectively,at 72 h posttreatment (Fig.3-B and C).From 60 h after feeding,WFT mortality was significantly higher in theCopEandTLR6lines than in the WT tomato plants.After 72 h after feeding,less than 50% of the WFT survived onCopEandTLR6T2lines,while the survival rate on the WT was about 70% (Fig.3-D).A similar trend of reduced survival rate of WTF was observed in WTF fed on T3transgenic lines (Fig.3-E).At 72 h after feeding,more than 90% of WTFs dead,while the survival rate of the WT was 90%(Fig.3-E).Overall,a similar trend of increased WFT resistance in both the T2and T3transgenic tomato plants expressingCopEandTLR6dsRNA confirms the stable inheritance ofCopEandTLR6transgenes.These results suggest that the survival rate of WFT onCopEandTLR6transgenic plants was significantly reduced compared with WT plants due to the RNAi-mediated silencing ofCopEandTLR6genes.

Fig.3 Thrips bioassay using CopE and TLR6 transgenic tomato plants.A,bioassay unit for evaluating the lethality of transgenic tomato plants against western flower thrips (WFT).B,T2 bioassay results showing the percent mortality of WFT over 120 h of feeding on T2 CopE and TLR6 transgenic tomato plants compared to wild type (WT).C,T3 bioassay results showing the percent mortality of WFT over 72 h of feeding on T3 CopE and TLR6 transgenic tomato plants compared to WT.D,survival curves of WTF fed on T2 CopE and TLR6 transgenic tomato plants compared to WT plants.Statistically significant differences were observed between WT and transgenic plants (CP3,P＜0.0002 and TL5,P＜0.0144) as determined by the log-rank test.E,survival curves of WTF fed on T3 CopE and TLR6 transgenic tomato plants compared to WT plants.Statistically significant differences were observed between WT and transgenic plants (P＜0.0001 for WT vs.CP3,CP5,TL5 and TL7) as determined by the log-rank test.Error bars indicate standard deviation (SD) of mean values (n=8 for the control and n=4 for transgenic lines).Different letters at each time point indicate significant differences between groups according to Duncan’s multiple range test (P≤0.05).

To determine the effect of feeding WFT dsRNAproducing transgenic tomato plants,we performed a qPCR analysis ofCopEandTRL6in WFT.The relative expression ofCopEwas significantly reduced,while the expression ofTRL6was slightly reduced in WFTfed transgenic tomato compared to WFT-fed WT tomato(P=0.03707 and 0.3165 forCopEandTRL6transgenic tomatoes,respectively) (Fig.4-A and B).Thus,these results indicate the dsRNA intake by WFT and successful gene silencing of endogenous genes.

Fig.4 Relative expression levels of CopE and TRL6 in WFT by qPCR.A,expression of CopE in WFT 72 h after feeding on wild-type tomato (WT-C) or CopE-transgenic tomato plants(T2).B,TLR6 expression in WFT 72 h after feeding on wild-type tomato (WT-T) or TRL6-transgenic tomato plants (T2).Error bars indicate the mean±SD of six biological replicates.Different letters indicate significant differences between groups according to Duncan’s multiple range test (P≤0.05).

4.Discussion

The RNAi approach has been commonly deployed in the field of entomology to investigate the function,regulation,and expression of genes and the mechanism of RNAi silencing in insects such as the fruit fly (Roignantet al.2003;Bischoffet al.2006;Milleret al.2008),red flour beetle (Tomoyasu and Denell 2004;Fujitaet al.2006;Minakuchiet al.2009;Parthasarathy and Palli 2009),and silkworm (Quanet al.2002;Ohnishiet al.2006;Hossainet al.2008).Knocking down essential insect genes by mechanical ingestion ofin vitrosynthesized dsRNA has been used to target Thysanopteran pests.Adult WFT microinjected with dsRNA targeting the vacuolarATP synthase subunit-Bgene exhibited increased mortality and reduced fecundity (Badillo-Vargaset al.2015).In another approach,symbiont-mediated RNAi (viadsRNA-expressing gut bacterial strain) targeting an essentialtubulingene caused high WFT larvae mortality (Whittenet al.2016).However,in many of those previous studies,dsRNA was directly injected into the organism or by administering dsRNA in the artificial feed mix,which is not a practical approach to control pests in the field.For efficient pest management,insects should be able to take up dsRNAs produced in the plant,such as through direct feeding and digestion in the insect gut.

The present study utilized an approach in which tomato plants were genetically engineered to express dsRNAs of essential insect genes and deliver resulting siRNAs to the WFT midgutviafeeding.The small regulatory RNAs in the plant tissues confer protection to host plants by silencing targeted endogenous insect gene transcripts.We selectedF.occidentalisendogenousCopEandTLR6for siRNA-mediated gene silencing,as these genes play an essential role in insect biology and could lead to lethal effects if disrupted (Hanet al.2019).CopE and other subunits form the coat protein complex I (COPI),which is involved in vesicle transport between the endoplasmic reticulum and thecis-Golgi complex (Bonifacino and Lippincott-Schwartz 2003;Hanet al.2019).TLRs are a class of pattern recognition receptors that play a crucial role in insect innate immunity against pathogens (Imler and Zheng 2004;Hanet al.2019).

We constructedCopEandTLR6RNAi plant expression vectors to integrateCopEandTLR6target sequences in sense and anti-sense directions,upon which expressionsCopEandTLR6dsRNAs would be produced,respectively.WFT that fed on cut flower stalks of dsRNA-producing T2/T3CopEandTLR6transgenic tomato plants showed increased mortality compared with WT plants.Moreover,WFT that fed on the transgenicCopEplants showed higher mortality than those feeding on theTRL6transgenic lines.Nevertheless,bothCopEandTLR6transgenic plants showed higher mortality than the WT plants.Consistent with these results,WFT fed on the transgenic leaves showed significantly reducedCopEexpression.AlthoughTRL6expression was not significantly affected (P=0.3165) 72 h postfeeding onTRL6transgenic plants,the mortality of WFT was increased remarkably.This suggests thatTRL6expression was knocked down at an earlier time point than 72 h post-feeding.

These results are in accordance with previous studies in which WFT growth was inhibited when insects fed on leaf discs soaked inCopEandTLR6dsRNA solutions(Hanet al.2019).Furthermore,dsRNA of components of the coat protein complexes,includingCOPIandCOPβ,have been proven effective against Colorado potato beetle(Zhuet al.2011) and phytophagous mite (Kwonet al.2013,2016).Our results demonstrate that transgenic expression of dsRNAs corresponding to insect-specific essential genes,CopEandTLR6can enhance tomato resistance to WFT and can potentially be employed as a strategy to protect plants from sucking insect pests.

5.Conclusion

In this study,we usedAgrobacterium-mediated transformation to develop stable transgenic tomato plants expressing dsRNAs of two essential insect genes,CopEandTLR6.Our results demonstrate thattransgenic expression ofCopEandTLR6dsRNAs can induce RNAi of their corresponding endogenous genes in WFT upon feeding on dsRNA-producing transgenic plants.RNAi silencing of endogenousCopEandTLR6accelerated WFT mortality compared to wild-type plants.Between the two RNAi plants,CopEdsRNA plants exhibited slightly greater WFT resistance compared withTLR6dsRNA plants.Our findings suggest that engineeredCopE/TLR6dsRNA lines may help manage WFT in the greenhouse and field crops,and this approach may lead to novel insect pest resistance applications.Further studies should be conducted to validate their efficacy in field conditions and test whether simultaneous RNAi silencing of two or more lethal genes may confer even greater pest resistance.Considering the important roles ofCopEandTLR6in insect biology,more research is warranted to determine the RNAi silencing effects ofCopE/TLR6on other biological parameters important in pest management,such as nymph development,female fecundity and adult longevity of WFT.

Acknowledgements

This research was supported by the Basic Science Research Program through the National Research Foundation (NRF),Ministry of Education,Korea(2021R1I1A1A01041938),a grant from the New Breeding Technologies Development Program,Rural Development Administration,Korea (PJ0165432022).Mr.Sung Jin Kim was supported in part by the BK21 Plus Program,Ministry of Education,Korea.

Declaration of competing interest

The authors declare that they have no conflict of interest.