at SciVerse ScienceDirect

Insect Biochemistry and Molecular Biology 42 (2012) 902e910

Contents lists available

Insect Biochemistry and Molecular Biology

journal homepage: www.elsevier .com/locate/ ibmb

Characterization of a midgut-specific chitin synthase gene (LmCHS2) responsiblefor biosynthesis of chitin of peritrophic matrix in Locusta migratoria

Xiaojian Liu a, Huanhuan Zhang a, Sheng Li b, Kun Yan Zhu c, Enbo Ma a,*, Jianzhen Zhang a,b,**

aResearch Institute of Applied Biology, Shanxi University, Taiyuan, Shanxi 030006, Chinab Institute of Plant Physiology and Ecology, Shanghai Institutes of Biological Sciences, Chinese Academy of Sciences, Shanghai 200032, ChinacDepartment of Entomology, 123 Waters Hall, Kansas State University, Manhattan, KS 66506, USA

a r t i c l e i n f o

Article history:Received 2 July 2012Received in revised form8 September 2012Accepted 10 September 2012

Keywords:Chitin synthase 2Gene expression patternLocusta migratoriaPeritrophic matrixRNA interference

* Corresponding author. Tel.: þ86 351 7018871/701** Corresponding author. Research Institute of AppliTaiyuan, Shanxi 030006, China. Tel.: þ86 351 70187011981.

E-mail addresses: maenbo2003@sxu.edu.cn (E. Ma

0965-1748/$ e see front matter � 2012 Elsevier Ltd.http://dx.doi.org/10.1016/j.ibmb.2012.09.002

a b s t r a c t

Chitin, an essential component of peritrophic matrix (PM), is produced by a series of biochemicalreactions. Chitin synthase plays a crucial role in chitin polymerization in chitin biosynthetic pathway. Inthis study, we identified and characterized a full-length cDNA of chitin synthase 2 gene (LmCHS2) fromLocusta migratoria. The cDNA contains an open reading frame of 4569 nucleotides that encode 1523amino acid residues, and 76- and 373-nucleotides for 50- and 30-noncoding regions, respectively. Analysisof LmCHS2 transcript in different tissues of the locust by using real-time quantitative PCR indicated thatLmCHS2 was exclusively expressed in midgut and gastric caeca (a part of the midgut). The highestexpression was found in the anterior midgut with a decline of the transcript level from the anterior toposterior regions. During growth and development of locusts, there was only a slight expression in eggs,but the expression gradually increased from nymphs to adults. In situ hybridization further revealed thatLmCHS2 transcript mainly presented in the apical regions of brush border forming columnar cells ofgastric caeca. LmCHS2 dsRNA was injected to fifth-instar nymphs to further explore biological functionsof LmCHS2. Significantly down-regulated transcript of LmCHS2 resulted in a cessation of feeding anda high mortality of the insect. However, no visible abnormal morphological change of locusts wasobserved until insects molted to adults. After dissection, we found that the average length of midgutsfrom the LmCHS2 dsRNA-injected locusts was shorter than that of the control insects that were injectedwith dsGFP. Furthermore, microsection of midguts showed that the PM of the LmCHS2 dsRNA-injectednymphs was amorphous and thin as compared with the controls. Our results demonstrate thatLmCHS2 is responsible for the biosynthesis of chitin associated with PM and plays an essential role inlocust growth and development.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Chitin is an essential component of insect cuticular exoskeletonand tracheae, and significant chitin deposition occurs in peritrophicmatrix (PM) that lines the midgut epithelium (Merzendorfer andZimoch, 2003). The PM, semi-permeable matrix, is involved inthe protection of insect gut against food abrasion, and invasionof microorganisms and parasites, and many other functions(Terra, 2001; Khajuria et al., 2010).

6098; fax: þ86 351 7011981.ed Biology, Shanxi University,871/7016098; fax: þ86 351

), zjz@sxu.edu.cn (J. Zhang).

All rights reserved.

In insects, although there is a great diversity of PM structures, itwas typically categorized into two types (Types I and II) based onthe mode of PM formations. Type I PM is secreted by the entireepithelium lining midgut, whereas a small number of specializedcells called the cardia produce Type II PM. Type I PM is widespreadin insects and particularly prevalent in lepidopterans in whichforms a “felt-like” material with a thickness of 0.5e1.0 mm. Incontrast, Type II PM is more organized and contains one to threelaminated layers which are found in primitive orders (e.g., Der-maptera and Isoptera) (Shao et al., 2001; Kato et al., 2006). Bothtypes of PM contain chitin and proteins, which include proteins,glycoproteins, and proteoglycans (Terra, 2001). The chitin contentin PM generally accounts for 3e13% (Hegedus et al., 2009). Ultra-structural observations of PM suggest that chitin appears to forma flexible framework onto which the proteins are assembled toform a matrix structure (Wang and Granados, 2001).

X. Liu et al. / Insect Biochemistry and Molecular Biology 42 (2012) 902e910 903

Despite the PM is biologically important in insects, relativelylittle information is available about the chitin biosynthetic pathwayin insects. The last enzyme catalyzed chitin polymer biosynthesis ininsects is chitin synthasewhich converts UDP-N-acetylglucosamine(UDP-GlcNAc) to the growing chitin polymer (Cohen, 2001). Todate, two chitin synthase genes (CHS1 and CHS2, also referred toCHSA and CHSB, respectively) have been reported in insects. CHS1 isresponsible for chitin synthesis in cuticle and cuticular lining of theforegut, hindgut, and trachea, whereas CHS2 is dedicated to chitinsynthesis in the PM (Zimoch and Merzendorfer, 2002). They arelarge transmembrane proteins with a slightly acidic isoelectricpoints. Insect CHS1 is known to have a short alternative exons,which lead to the production of two splicing variants. Unlike CHS1,no alternative exons have been reported for CHS2 (Merzendorfer,2006). CHS1 and CHS2 can also be distinguished by expressionpatterns in different tissues of an insect. CHS1 is predominantlyexpressed in the epidermis and tracheal cells, whereas CHS2 isspecifically expressed in the midgut epithelial cells (Merzendorferand Zimoch, 2003). However, a recent study detected bothenzymes in newly formed compound eyes of Anopheles gambiaepupae by using immunohistochemical analysis (Zhang et al., 2012).

RNA interference (RNAi) is a phenomenon of down-regulationof gene expression triggered by double-stranded RNA (dsRNA) orsmall interfering RNA (siRNA). RNAi can be used as a powerful toolfor the rapid analysis of gene function in a variety of organisms. Thefirst clear demonstration that the two chitin synthase genes havedistinct functions was carried out in Tribolium castaneum (Arakaneet al., 2005, 2008). By injection of dsRNA synthesized based oneither of the two chitin synthase genes, they found that knock-down of TcCHS2 affected only midgut chitin without a significantimpact on total chitin, whereas down-regulation of TcCHS1 resul-ted in substantial loss of chitin only in the exoskeleton. In An.gambiae, the total chitin content was significantly reduced in thelarvae fed on chitosan/AgCHS1 dsRNA-based nanoparticles. Incontrast, the larvae fed on chitosan/AgCHS2 dsRNA-based nano-particles showed increased PM permeability and larval suscepti-bility to several chemicals (Zhang et al., 2010b). All these resultssuggest that there is a specialization in the function of the twochitin synthases genes. CHS1 is required exclusively for chitinsynthesis of cuticle and trachea, whereas CHS2 is only responsiblefor chitin synthesis of the PM-associated chitin in gut epithelialcells (Merzendorfer, 2006).

To date, most studies on insect chitin synthases have focused onchitin synthase 1 gene expressed by epidermal cells, and there islimited information on chitin synthase 2 gene. In our earlierresearch, RNAi of LmCHS1 in Locusta migratoria adversely affectsgrowth and development of nymphs (Zhang et al., 2010a). In thispaper, we report chitin synthase 2 gene (LmCHS2) from a major

Table 1Sequences of PCR primers used in RACE-PCR for amplification of the full-length cDNA, qPdsRNA synthesis for RNA interference of LmCHS2.

Application of primers Gene name Primer name Pr

cDNA cloning LmCHS2 Adapter CT50 RACE-R AG30 RACE-F CC

qPCR analysis LmCHS2 ECHS2F CAECHS2R CG

b-actin b-actin F CGb-actin R GC

In situ hybridization LmCHS2 CHS2F AGCHS2R CT

dsRNA synthesis LmCHS2 dsCHS2F TAdsCHS2R TA

GFP dsGFPF TAdsGFPR TA

hemimetabolous agricultural insect pest, L. migratoria, whichincludes: 1) isolation and sequencing of a full-length cDNA derivedfrom LmCHS2; 2) determinations of expression patterns of LmCHS2in different tissues and at different developmental stages; 3)subcellular localization of LmCHS2 mRNA by in situ hybridization;and 4) functional analysis of LmCHS2 by RNAi.

2. Materials and methods

2.1. Insects

L. migratoria were provided by Insect Protein Co., Ltd. of Can-gzhou City in China and reared using fresh wheat seedlings andbran in the laboratory. The eggs were then incubated in a growthchamber (Yiheng, China) at 30 �C under a relatively humidity of 50%and a photoperiod of 14 h light and 10 h dark cycle. Insects ofdifferent developmental stages were synchronized for samplecollections.

2.2. Isolation and sequencing of cDNA encoding CHS2

A cDNA fragment of LmCHS2 encoding chitin synthase 2 wasfirst identified from Migratory Locust EST Database (http://locustdb.genomics.org.cn/). Total RNA was extracted from themidgut of fifth-instar nymphs using RNAiso Plus reagent (TaKaRa,China). mRNA was then isolated using PolyATtract mRNA isolationsystems (Promega, Madison, WI). cDNA was synthesized from 1 mgmRNA using the Smart Race cDNA Amplification Kit (Clontech,Mountain View, CA). To obtain the full-length cDNA of LmCHS2,RACE-PCR was performed using Advantage 2 PCR Enzyme System(Clontech). The sequences of the gene-specific primers for RACEPCR are presented in Table 1. The PCR products were analyzed on 1%agarose gel, purified using Gel Mini Purification Kit (TIANGEN,China), subcloned into pGEM-T Easy Vector (Promega), and thensequenced completely from both directions.

2.3. Analysis of LmCHS2 cDNA and its deduced amino acidsequence

The amino acid sequence of LmCHS2 was translated using on-line tools of ExPASy website (http://www.expasy.org/tools/).Other sequence analysis tools to predict its molecular mass,isoelectric point, transmembrane helices, and N-glycosylation siteswere also obtained from ExPASy website. Multiple amino acidsequence alignment of all known insect chitin synthases wascarried out using ClustalW (http://www.ebi.ac.uk/Tools/msa/clustalw2/). The phylogenetic tree was generated by MEGA 4.

CR for gene expression analysis, RNA probe preparation for in situ hybridization, and

imer sequence (50e30) Size (bp)

AATACGACTCACTATAGGGCAAGCAGTGGTATCAACGCAGAGTCCTGTACCTTTGGAGATTCCCTTG 3229TGGCTCTGGGAGGCTAAGAATGCG 1537GCCTTCCGCATAGACAACT 150GCCATCATAACCAATGAATAAGCACAGTCAAAGAGAGGTA 156TTCAGTCAAGAGAACAGGATGGCACGTATCACCAAGGAC 360TGGTCCTCATTCCTTCCAATACGACTCACTATAGGGAGGTACAGGCTGAGCAGGAA 188ATACGACTCACTATAGGGCATCAGTGGACTTTCTCGCAATACGACTCACTATAGGGGTGGAGAGGGTGAAGG 712ATACGACTCACTATAGGGGGGCAGATTGTGTGGAC

X. Liu et al. / Insect Biochemistry and Molecular Biology 42 (2012) 902e910904

2.4. Tissue-specific expression of LmCHS2 using real-timequantitative PCR

LmCHS2- and b-actin-specific primers used for real-time quan-titative PCR (qPCR) analysis are shown in Table 1. b-actin was usedas a reference gene. For tissue-specific gene expression studies, 10tissues, including integument, foregut, midgut, hindgut, gastriccaeca, Malpighian tubules, fatbodies, muscles, wing and trachea,were dissected from fifth-instar nymphs on day 1. Total RNA wasisolated from each sample using RNAiso� Plus (TaKaRa, Japan).cDNAwas synthesized from 1.5 mg of total RNA treated with RNase-free DNase (Promega) using M-MLV reverse transcriptase (Prom-ega). qPCR was carried out in a 20-ml reaction containing 3 ml ofa 10-fold diluted cDNA, 0.4 mM of each primer and 12.5 ml SYBRPremix Ex Taq� II (TaKaRa, Japan) on Applied Biosystems 7300Real-Time PCR system (Applied Biosystems, USA). The relativeexpression of each gene was determined using cycling parametersof an initial denaturation at 95 �C for 10 s, followed by 40 cycles of95 �C for 5 s, and 60 �C for 34 s. A melt curvewas evaluated for eachqPCR experiment to confirm the amplification specificity. All theseexperiments were performed using three biological replicates, eachwith two technical replicates. The 2�DDCT method was used tocalculate the relative levels of LmCHS2 transcript in differenttissues.

2.5. Expression of LmCHS2 in different regions of midgut

To further analyze differential expression of LmCHS2 throughoutthe midgut. The midgut was dissected from fifth-instar nymphs onday 1 and divided into three equal segments corresponding toanterior, median, and posterior regions. Total RNA was isolatedfrom each sample from 15 midguts. All these experiments wereperformed using three biological replicates. Other methods werethe same as described in Section 2.4.

2.6. Developmental expression of LmCHS2

Eggs, midgut and gastric caeca dissected from first, second,third, forth and fifth-instar nymphs and adults were used for totalRNA isolation. The cDNA templates were used for determining thestage-dependent expression of LmCHS2. The relative expression ofLmCHS2 was analyzed by using qPCR as described in Section 2.4.

2.7. In situ hybridization

To prepare the RNA probes for LmCHS2, a pair of specific primerswas designed as shown in Table 1. After the resulting PCR product of360 bp was purified and ligated into the pGEM-T vector (Promega),both strands of the recombinant plasmid were sequenced.Digoxigenin-labeled RNA probes were generated by in vitro tran-scription from the plasmid using Dig RNA-Labeling Kit (Roche,Germany), and SP6 and T7 RNA polymerase to produce the senseand antisense strands, respectively. Subsequently, a dot-blot assaywas used to estimate the appropriate concentration of the probesfor in situ hybridization.

The gastric caeca from day 1 of the fifth-instar nymph weredissected in ice-cold phosphate buffered saline (PBS), fixed in 4%paraformaldehyde at 4 �C overnight, dehydrated through anethanol series and xylene, and embedded in paraffin. The sampleswere sectioned at 5 mm. The in situ hybridization was performedfollowing the protocol described by Zimoch and Merzendorfer(2002) and the slides were viewed under an OLYMPUS BX51 andphotography was taken with an OLYMPUS digital camera.

2.8. Functional analysis of LmCHS2

RNAi was carried out to study the function of LmCHS2. Ahomologous region with most sequence divergence between thecDNAs of LmCHS1 and LmCHS2 was selected for gene-specificdsRNA synthesis. The nucleotide sequence identity between thetwo genes is 58%. Specific primers, with T7 RNA polymerasepromoter sequence at the 50-end, were designed corresponding tothe dsRNA region. PCR was performed using cDNA frommidguts offifth-instar nymphs to prepare the template for dsRNA synthesis.The resulting fragment was subcloned and sequenced to confirm itsidentity. dsRNA of LmCHS2 and GFPwere prepared according to themethod of Zhang et al. (2010a). The sequences of the primers usedfor dsRNA synthesis and transcript analysis are shown in Table 1.

Aliquots of 10 mg LmCHS2 or GFP dsRNA were injected into thedorsal side between the second and third abdominal segments offifth-instar nymphs on day 2 using a manual microinjector (Ningbo,China). Each group consisted of 50 individual nymphs and theexperiment was carried out in three replicates. Following theinjections, nymphs were maintained in a growth chamber at 30 �C,and carefully observed for any visible abnormalities and mortality.To ensure that the down-regulation of transcripts was specific forLmCHS2, total RNA was isolated from days 5, 6 and 7 of fifth-instarnymphs (abbreviated as N5D5, N5D6 and N5D7, respectively) afterLmCHS2 dsRNA or GFP dsRNA injection. The integuments ormidguts dissected from three insects were pooled for each RNAextraction. cDNA synthesis and qPCR were performed to determinethe relative expression of LmCHS1 or LmCHS2 using the samemethods as described in Section 2.4.

The whole gut from individual nymph was dissected in ice-coldphosphate buffered saline (PBS) from days 5, 6 and 7 of fifth-instarnymphs after the injection of LmCHS2 dsRNA or GFP dsRNA. A totalof 10 whole guts were dissected from the LmCHS2 dsRNA or GFPdsRNA injected locusts in each replicate and scanned using EPSONPERFECTION V700 PHOTO (Indonesia). The lengths of the midgutswere measured using a vernier caliper. In addition, 10 midgutsdissected from the locusts in day 3 after injected with LmCHS2dsRNA or GFP dsRNA were fixed in 4% paraformaldehyde at 4 �Covernight to prepare paraffin sections as described in Section 2.7.Samples were stained with hematoxylin and eosin (H & E). Calco-fluor White (Sigma, Germany) was used to detect chitin. Rehy-drated samples were rinsed in phosphate-buffered saline (PBS;20 mM KH2PO4, 20 mM NaH2PO4, 0.15 M NaCl, buffered to pH 7.3),incubated for 90 min in PBS containing 0.01% (w/v) Calcofluorwhite (Sigma), 0.1% Triton-X 100 and 2% BSA and washed threetimes for 30 min with PBS. The fluorescence of Calcofluor wasvisualized using an Olympus BX51 fluorescence microscope.Photography was taken with an OLYMPUS digital camera. Thechitin content was also measured based on the method of Zhangand Zhu (2006).

3. Results

3.1. Characterization of LmCHS2 cDNA and deduced amino acidsequences

An LmCHS2 cDNA fragment was obtained fromMigratory LocustEST Database (accession number: LM_GL5_006268) and RACE-PCRwas used to amplify the 50 and 30-end regions of its full-lengthcDNA. By assembling the cDNA fragments from Locust EST Data-base and RACE-PCR, a full-length cDNA (GenBank accessionnumber: JQ901491) of LmCHS2 was finally obtained. The LmCHS2cDNA contains 5018 nucleotides consisting of an open readingframe of 4569 nucleotides, a 76-nucleotide 50-untranslated region(UTR) and a 373-nucleotide 30-UTR. The encoded protein contains

X. Liu et al. / Insect Biochemistry and Molecular Biology 42 (2012) 902e910 905

1523 amino acids with calculated molecular mass (MM) of about174 kDa and isoelectric point (pI) of 6.65. Like other CHSs, LmCHS2was predicted to contain three domains, including an N-terminaldomain (residues 1e544) with 7 transmembrane helices, a putativecatalytic domain (residues 545e881) in the middle showing high

Fig. 1. Nucleotide and deduced amino acid sequences of LmCHS2 cDNA from Locusta migrato(TAA) is indicated in *. The putative polyadenylation signals (AATAAA) are underlined. The pThe five potential N-glycosylation sites by PROSCAN are boxed. The amino acid sequence of th(EDR and QRRRW), which are suggested to be involved in catalytic function, are in white withJQ901491). (For interpretation of the references to color in this figure legend, the reader is

sequence identity with those from other insects, and a C-terminaldomain (residues 882e1523) with additional 7 transmembranehelices (Fig. 1). The signature sequences (QRRRW and EDR) forchitin synthases, were also found in LmCHS2. Five potential N-glycosylation sites were predicted using NetNGlyc 1.0 software

ria. The numbers on the right are for the deduced amino acid sequence. The stop codonutative transmembrane segments predicted by TMHMM Server v. 2.0 are shaded gray.e putative catalytic domain is inwhite with black background. The signature sequencesblue background. The sequences have been deposited in GenBank (accession number:referred to the web version of this article.)

IN FG MG HG GC MT FB MU WI TR

Re

la

tiv

e e

xp

re

ss

io

n (fo

ld

)

0

30

60

90

120

150

180

bb

bb b b b b

a

a

Fig. 3. qPCR analysis of tissue-dependent expression of LmCHS2 in fifth-instar nymphs(day 1). The constitutively expressed b-actin gene was used as internal control. Thetissues include integument (IN), foregut (FG), midgut (MG), hindgut (HG), gastric caeca(GC), Malpighian tubules (MT), fatbodies (FB), muscle (MU), wing (WI) and trachea(TR). Data are expressed as means � SD of three biological replications. The data oftissue-dependent expression are shown as fold changes as compared with the tissue

X. Liu et al. / Insect Biochemistry and Molecular Biology 42 (2012) 902e910906

from the ExPASy Proteomics website. These N-glycosylation sitesare distributed throughout the sequence, suggesting that theprotein is glycosylated.

3.2. Phylogenetic analysis of CHS2

A phylogenetic tree was generated using MEGA 4 after the full-length amino acid sequences of all known insect CHSs were alignedusing ClustalW (Fig. 2). Insect CHSs can be grouped into two classes,CHS1 and CHS2. The CHS that was derived from our locust cDNAanalysis belongs to CHS2 (i.e., LmCHS2). LmCHS2 has the highestidentity (47%) to the coleopteran T. castaneum CHS2 based ondeduced amino acid sequences.

3.3. Tissue-specific expression patterns of LmCHS2

qPCR was carried out to analyze the expression patterns ofLmCHS2 in different tissues of fifth-instar nymphs. Our resultsshowed that the transcript of LmCHS2was specifically expressed inmidgut and gastric caeca (a part of the midgut) and no detectableexpression in integument, foregut, hindgut, Malpighian tubules,fatbodies, muscle, wings and trachea (Fig. 3).

Fig. 2. Phylogenetic analysis of LmCHS1, LmCHS2 and other known insect CHSs. Thetree was constructed based on the CHS amino acid sequences. Values at clusterbranches indicate results of the boostrap analysis. CHSs are from Lucilia cuprina (Lc),Drosophila melanogaster (Dm), Drosophila pseudoobscura (Dp), Anopheles gambiae (Ag),Aedes aegypti (Aa), Anopheles quadrimaculatus (Aq), Manduca sexta (Ms), Spodopterafrugiperda (Sf), Spodoptera exigua (Se), Plutella xylostella (Px), Choristoneura fumiferana(Cf), Ostrinia furnacalis (Of), Ectropis oblique (Eo), Tribolium castaneum (Tc), Apis mel-lifera (Am), and Locusta migratoria (Lm). GenBank accession numbers are as follows:LcCHS1 (AAG09712), DmCHS1 (NP_524233), DmCHS2 (NP_524209), DpCHS1(XP_001359390), DpCHS2 (XP_001352881), AgCHS1 (XP_321337), AgCHS2(XP_321951), AaCHS1 (XP_001651163), AaCHS2 (EAT46081), AqCHS1 (ABD74441),MsCHS1 (AAL38051), MsCHS2 (AAX20091), SfCHS2 (AAS12599), SeCHS1 (AAZ03545),SeCHS2 (ABI96087), PxCHS1 (BAF47974), CfCHS1 (ACD84882.1), OfCHS1 (ACB13821),OfCHS2 (ABX46067), EoCHS1 (ACA50098), TcCHS1 (AAQ55059), TcCHS2 (AAQ55061),AmCHS1 (XP_395677), AmCHS2 (XP_001121152), LmCHS1 (GU067730) and LmCHS2(JQ901491).

showing the lowest expression which is ascribed an arbitrary value of 1. Differentletters on the bars of the histogram indicate statistically significant difference(P < 0.05, Fisher’s LSD test; n ¼ 3).

3.4. Expression patterns of LmCHS2 in different regions of midgut

Our qPCR analysis of the anterior, median, and posterior regionsof themidgut showed the highest expression level of LmCHS2 in theanterior midgut but with a decline from the anterior to the poste-rior regions (Fig. 4). The decreased expression was drastic from theanterior to the posterior regions of the midgut.

3.5. Developmental expression patterns of LmCHS2

The expression profiles of LmCHS2 during L. migratoria devel-opment were determined in eggs, midgut and gastric caeca

The midgut of the 5th instar nymph

Anterior Median Posterior

Re

la

tiv

e e

xp

re

ss

io

n (fo

ld

)

0

1

2

3

4

5

6

7

a

b

c

Fig. 4. qPCR analysis of LmCHS2 transcript in different regions of midguts dissectedfrom fifth-instar nymphs (day 1). The constitutively expressed b-actin gene was usedas internal control. Data are expressed as means � SD of three biological replications.The data of LmCHS2 expression in different regions of the midgut are shown as foldchanges as compared with the region showing the lowest expression which is ascribedan arbitrary value of 1. Different letters on the bars of the histogram indicate statis-tically significant difference (P < 0.05, Fisher’s LSD test; n ¼ 3).

X. Liu et al. / Insect Biochemistry and Molecular Biology 42 (2012) 902e910 907

dissected from first, second, third, forth and fifth-instar nymphsand adults by using qPCR (Fig. 5). Our studies showed that LmCHS2expression was not detected in eggs; however, LmCHS2 expressionwas found to gradually increase from first to fifth-instar nymphs,and reach the highest in the first day of adults. It appeared thatLmCHS2 is expressed during the periods when locusts are activelyfeeding.

3.6. In situ hybridization

In order to visualize the distribution of LmCHS2 transcript ingastric caeca, we performed in situ hybridization using paraffinsection of gastric caeca of the fifth-instar nymph. Our resultsshowed that LmCHS2 transcript was mainly detected in the apicalregions of brush border forming columnar cells (Fig. 6) as comparedwith the control hybridization using digitonin-labeled sense RNA.

3.7. Functional analysis of LmCHS2

RNAi was performed using LmCHS2-specific dsRNA to reveal thefunction of LmCHS2 in L. migratoria development and molting.LmCHS2 or GFP dsRNA was synthesized in vitro and injected intofifth-instar nymphs (2 days old). As shown in Fig. 7A, only thetargeted mRNA (LmCHS2) was greatly down-regulated and therewas no significant decrease in the transcript level of the non-targetgene (LmCHS1). These results indicate that our dsRNA-mediatedgene silencing for LmCHS2 was gene-specific. In total of 40 fifth-instar nymphs injected with LmCHS2 dsRNA, 20 died before theydeveloped into the adult stage, 10 molted to adults but finally diedafter 1e3 days, and 10 survived. The total mortality is 75%.However, no visible morphological change was observed in thelocusts injected with the dsRNA of LmCHS2.

However, when we examined the guts dissected from days 5, 6and 7 of the fifth-instar nymphs (abbreviated as N5D5, N5D6 andN5D7) that were injected with LmCHS2 or GFP dsRNA, we foundthat the midguts that were dissected from the nymphs injectedwith dsRNA of LmCHS2 contained virtually no food, whereas themidguts of the control nymphs injected with dsRNA of GFP genewere completely full with food. Furthermore, the average length ofthe midguts from the control nymphs was 9.0 � 1.5 mm (n ¼ 30),

EG N1 N2 N3 N4 N5 AD

Re

la

tiv

e e

xp

re

ss

io

n (fo

ld

)

0

20

40

60

80

100

d

c

bc

ab

ab

a

a

Fig. 5. qPCR analysis of LmCHS2 transcript in eggs (EG); first- (N1), second- (N2), third-(N3), forth- (N4) and fifth-instar (N5) nymphs; and adults (AD) of L. migratoria. Theconstitutively expressed b-actin gene was used as internal control. Data are expressedas means � SD of three biological replications. Mean expression in each stage is shownas fold change as compared with the stage showing the lowest expression which wasascribed an arbitrary value of 1. Different letters on the bars of the histogram indicatestatistically significant difference (P < 0.05, Fisher’s LSD test; n ¼ 3).

whereas the average length of the midguts from the LmCHS2dsRNA-injected nymphs was only 4.5 � 0.5 mm (n ¼ 30), repre-senting a 50% reduction (P < 0.001, Student’s t test) in the length ofthe midguts when the expression of LmCHS2 was suppressed byRNAi. Furthermore, the gastric caeca was smaller in the LmCHS2dsRNA-injected nymphs than those of the control nymphs (Fig. 7B).

Because our results of tissue-specific expression patterns indi-cated that LmCHS2 was specifically expressed in the midgut andgastric caeca, the highmortality associated with the decrease in theLmCHS2 transcript level apparently was due to a reduction of chitinbiosynthesis in the midgut after fifth-instar nymphs were injectedwith LmCHS2 dsRNA. To test this hypothesis, we analyzed theinfluence of the gene silencing for LmCHS2 on the PM formation.Themidgut paraffin sections were prepared from the nymphs threedays after the injection of LmCHS2 or GFP dsRNA. The sections werestained with hematoxylin and eosin (H & E) or Calcofluor White.We found that the midguts from the control nymphs injected withGFP dsRNA contained a fully developed PM. The PM appeared to bewell structured, clearly separating the food from the epithelial cells.In contrast, the midguts from the nymphs injected with LmCHS2dsRNA showed a disrupted structure of PM and contained littlefood. In some of these nymphs, the PM was even invisible, sug-gesting a possible loss of PM or significantly reduced PM (Fig. 7C).

4. Discussion

In the past decade, research hasmade some significant advancesin better understanding of chitin synthases in insects, particularlythrough molecular cloning and functional analyses in several insectorders such as Diptera, Lepidoptera, Coleoptera, Orthoptera andHymenoptera (Zhang et al., 2010a, 2012). To date, various insectspecies are known to possess two different chitin synthases, CHS1and CHS2, which are mainly responsible for the syntheses of chitinassociated with the cuticular exoskeleton and tracheae, and thatassociated with PM that underlies the epithelial cells of the midgut,respectively (Shao et al., 2001). However, relatively little informa-tion is available about the PM chitin synthesis in insects.

We previously reported molecular characteristics of the chitinsynthase 1 gene (LmCHS1) from hemimetabolous insect,L. migratoria (Zhang et al., 2010a). In this paper, we report the full-length cDNA sequence of a new chitin synthase from the sameinsect species. Based on our phylogenetic analysis of all knowninsect chitin synthases, the chitin synthase gene that we revealedfrom this study was classified to CHS Group 2. Therefore, wedesignated this gene as LmCHS2. LmCHS2 is relatively shorter thanLmCHS1 and encodes a protein with a predicted pI of 6.65. Theslightly more acidic pI than the predicted pI of 6.89 for LmCHS1 isconducive to its function in the PM. Like other chitin synthases,LmCHS2 is a large (174 kDa) and complex transmembrane proteincharacterized by multiple transmembrane segments. In Manducasexta, purification of MsCHS2 from the larval midgut revealeda trimeric chitin synthase complex (Maue et al., 2009). Hence, theability to form oligomers may facilitate the formation of pores inthe membrane through which the nascent chitin polymers aretranslocated. The isolation of LmCHS2 cDNA provided us with anopportunity to study the expression patterns and biological func-tions in L. migratoria.

LmCHS2 was specifically expressed in the midgut and gastriccaeca (Fig. 3), and such an expression pattern is consistent withthose of class 2 chitin synthase genes of other insect speciesincluding Drosophila melanogaster (Gagou et al., 2002), M. sexta(Hogenkamp et al., 2005), T. castaneum (Arakane et al., 2004),Ostrinia furnacali (Qu et al., 2011), Spodoptera exigua (Kumar et al.,2008) and Spodoptera frugiperda (Bolognesi et al., 2005). Such anexpression pattern is also in agreement with the hypothesis that

Sense-CHS2

Anti-CHS2

A A'

B B'

250

250100

100

Fig. 6. Localization of LmCHS2 transcript in gastric caeca of fifth-instar nymph of L. migratoria. In situ hybridization of the paraffin sections of the gastric caeca was performed witha digoxigenin-labeled antisense RNA probe that was complementary to a part of the LmCHS2 coding region. The arrow indicates the localization of LmCHS2 transcript. (A) and (A0)show the results of hybridization by using the sense probe of LmCHS2; (B) and (B0) show the results of hybridization by using the anti-sense probe of LmCHS2.

X. Liu et al. / Insect Biochemistry and Molecular Biology 42 (2012) 902e910908

LmCHS2 is responsible for biosynthesis of the PM-associated chitinin themidgut. It has been known that gastric caeca are connected tothe foregut and a supplementary structure of midgut to increasethe surface area for digestion as well as nutrient absorption (Akpanand Okorie, 2003). Therefore, LmCHS2 mRNA found in the gastriccaeca is likely to serve the same function as that found in themidgut. Our qPCR analysis clearly demonstrated that LmCHS2 washighly expressed in the anterior midgut, but tapered off in themedian and posterior parts of the midgut. A similar result was alsoobserved for MsCHS2 in M. sexta as evaluated by northern blot andRT-PCR analysis (Hogenkamp et al., 2005). Our results indicatedthat L. migratoria has Type I PM; however, the anterior midgut mayplay a more important role in chitin biosynthesis than the rest ofthe midgut.

To further localize LmCHS2 transcript in gastric caeca, in situhybridization of the paraffin sections of gastric caeca from fifth-instar nymphs was performed using a digoxigenin-labeled anti-sense RNA probe that was complementary to a specific region of theLmCHS2 cDNA. The transcript of LmCHS2was localized in the apicalregions of brush border forming columnar cells with strong signals,whereas the hybridization in the controls using digitonin-labeledsense RNA showed little signals (Fig. 6). In M. sexta, MsCSH2 tran-script was enriched in the cytoplasm of columnar cells (Zimoch andMerzendorfer, 2002).

We further examined stage-dependent expression patterns ofLmCHS2 transcript from eggs to adults by qPCR. Because LmCHS2 isspecifically expressed in the midgut and gastric caeca, we usedthese two tissues for our analysis. As demonstrated in Fig. 5,LmCHS2 was expressed throughout the feeding stages of nymphsand adults. By contrast, no detectable expressionwas found in eggs.L. migratoria is a typical hemimetabolous insect that develops fromegg to nymph and then directly to adult without going through anintermediate pupal stage. In holemetabolous insects like D. mela-nogaster (Gagou et al., 2002), M. sexta (Hogenkamp et al., 2005),T. castaneum (Arakane et al., 2004), O. furnacali (Qu et al., 2011), S.exigua (Kumar et al., 2008) and S. frugiperda (Bolognesi et al., 2005),CHS2 mRNA was expressed in the course of the intermolt stages.However, in phases of starvation during the instar molt or the

prepupal wandering stage, CHS2 mRNA is down-regulated. In themosquito Aedes aegypti, in situ hybridization of the midgut samplesshowed that AaCHS2 mRNA increased following a blood meal(Ibrahim et al., 2000). Through these studies, it is clear that CHS2mRNA levels gradually increased with the food demanding to formPM surrounding the food bolus.

In this study, we revealed the role of LmCHS2 in insect devel-opment by using RNAi. As shown in Fig. 7A, the injection of dsRNAofLmCHS2 substantially suppressed the level of LmCHS2 transcriptwithout significant effect on the level of LmCHS1 transcript. TheRNAi-mediated suppression of LmCHS2 transcript ultimately resul-ted in a high mortality of the insect. Such mortality was apparentlycaused by the starvation due to reduced chitin biosynthesis, whichis strongly supported by our several findings. First, the length ofmidguts dissected from the nymphs on days 5, 6 and 7 after theinjection of LmCHS2 dsRNA was significantly reduced as comparedwith those of the control nymphs injected with GFP dsRNA.Secondly, the size of the gastric caeca dissected from the nymphsinjectedwith LmCHS2 dsRNAwas also significantly reduced. Thirdly,the PM of the LmCHS2 dsRNA-injected nymphs appeared amor-phous and thin as compared with the controls. In some individuals,the PMwas even absent. Fourthly, the chitin content of the midgutsdissected from LmCHS2 dsRNA-injected nymphs was below thelimit of detection in our chitin content assay (data not shown). Incontrast, chitin content in themidguts from the control nymphs canbe easily detected (0.012 mg/midgut). Finally, amore direct evidencewas that the midguts dissected from the nymphs injected withLmCHS2 dsRNA virtually contained no food.

All these results support our notion that RNAi-mediatedsuppression of LmCHS2 transcript leads to a reduced biosynthesisof chitin to form the PM in the midgut. The loss of the PM or the PMwith disrupted structure due to insufficient chitin contents ulti-mately affects insect feeding. Thus, the integrity of the PM is criticalfor maintaining normal physiological function for insect growthand development as observed in other insect species (Khajuriaet al., 2010). Our findings are consistent with the results fromT. castaneum. Larvals treated with TcCHS2 dsRNA exhibited little orno chitin in their PM and a dramatic shrinkage in larval size due to

dsGFP

dsCHS2

0

1

2

3

4

5

6

N5D5 N5D6 N5D7

Re

lativ

e e

xp

re

ss

ion

(fo

ld)

B

N5D5 N5D6 N5D7

dsGFP dsCHS2 dsGFP dsGFPdsCHS2 dsCHS2

10 mm

PyGc

0

0.5

1

1.5

2

N5D5 N5D6 N5D7

Re

lativ

e e

xp

re

ss

ion

(fo

ld)

ALmCHS2LmCHS1

NS

NS

NS

*

*

*

dsGFP

dsCHS2

250100

250100

C

PM

Mg

Fig. 7. Effect of LmCHS2 dsRNA injected into fifth-instar nymphs on day 2 on the transcript levels of two LmCHS genes and development of L. migratoria. (A) Analysis of relativetranscript levels of LmCHS2 and LmCHS1 after LmCHS2 dsRNA injection on days 5, 6 and 7 of fifth-instar nymphs (abbreviated as N5D5, N5D6 and N5D7) using qPCR. Control insectswere injected with equivalent amount of GFP dsRNA. b-actin gene was used as internal control. Data are shown as means � SD from three independent experiments. * showsa significant difference between the nymphs injected with LmCHS2 dsRNA and those injected with GFP dsRNA (P < 0.05, T-test); NS indicates no significant difference between thetwo groups. (B) The change of midgut and gastric caeca after the injection of LmCHS2 dsRNA in fifth-instar nymphs on days 5, 6 and 7. The arrows show gastric caeca (Gc), pylorus(Py) and midgut (Mg). (C) Effect of the injection of LmCHS2 dsRNA on the formation of peritrophic matrix (PM) in nymphs 3 days after the injections by showing the hematoxylinand eosin (H & E) stained midguts from nymphs injected with LmCHS1 dsRNA and GFP dsRNA. The arrow shows the PM of the midgut from a control insect with fully formed PMlining the midgut lumen. However, there was no PM formation in the midgut from the insect injected with LmCHS2 dsRNA.

X. Liu et al. / Insect Biochemistry and Molecular Biology 42 (2012) 902e910 909

the cession of feeding (Arakane et al., 2005, 2008). In Ae. aegyptiand An. gambiae, the PM is disrupted when the levels of CHS2transcripts are suppressed. Such a PM disruption often leads toinsect mortality or increased susceptibility to chemical insecticides(Kato et al., 2006; Zhang et al., 2010a). In addition, insect CHS2 also

play other important roles as reported in several insect species. Forexample, TcCHS2 played important roles during embryonic andadult development in the red flour beetle. Specifically, the femalebeetles injected with TcCHS2 dsRNA are not able to lay their eggs(Arakane et al., 2008).

X. Liu et al. / Insect Biochemistry and Molecular Biology 42 (2012) 902e910910

Recently, transgenic plants engineered to produce hairpindsRNAs in vivo have been reported for protecting the plants fromthe feeding damage of herbivorous insects (Baum et al., 2007;Gordon and Waterhouse, 2007; Mao et al., 2007). However, iden-tification of a suitable gene for transgenic RNAi plants is important,as Baum et al. (2007) observed, not all dsRNAs may effectivelysuppress the expression of target genes, and not all dsRNAs areeffective in killing insects. Nevertheless, our results suggest thatLmCHS2 could serve as a good candidate gene for developing RNAi-based technologies for locust control due to the sequence speci-ficity of LmCHS2 dsRNA coupledwith its ability to suppress the genecritical for insect survival. Because chitin biosynthesis is absent inhigher animals and LmCHS2 is critically important for maintainingthe integrity of PM in insects, results from this study are expected tohelp researchers develop effective and environmentally friendlystrategies for insect pest control by attacking chitin biosynthesis ininsects.

Acknowledgments

This work was supported by National Basic Research Program ofChina (2012CB114102), National Natural Science Foundation ofChina (Grant No. 30970410), International Cooperation andExchange Program (30810103907), Science and TechnologyResearch Project of Shanxi Province (20110311010), Program for TopYoung Academic Leaders of Higher Learning Institutions of Shanxi(TYAL), China Postdoctoral Science Foundation (Grant No.20100480642 and 201104297) and Shanghai Postdoctoral ScienceFoundation (Grant No. >11R21417300).

References

Akpan, B.E., Okorie, T.G., 2003. Allometric growth and performance of the gastriccaeca of Zonocerus variegatus (L.) (Orthoptera: Pyrgomorphidae). Acta Entomol.Sinica 46, 558e566.

Arakane, Y., Hogenkamp, D.G., Zhu, Y.C., Kramer, K.J., Specht, C.A., Beeman, R.W.,Kanost, M.R., Muthukrishnan, S., 2004. Characterization of two chitin synthasegenes of the red flour beetle, Tribolium castaneum, and alternate exon usage inone of the genes during development. Insect Biochem. Mol. Biol. 34, 291e304.

Arakane, Y., Muthukrishnan, S., Kramer, K.J., Specht, C.A., Tomoyasu, Y.,Lorenzen, M.D., Kanost, M., Beeman, R.W., 2005. The Tribolium chitin synthasegenes TcCHS1 and TcCHS2 are specialized for synthesis of epidermal cuticle andmidgut peritrophic matrix. Insect Mol. Biol. 14, 453e463.

Arakane, Y., Specht, C.A., Kramer, K.J., Muthukrishnan, S., Beeman, R.W., 2008. Chitinsynthases are required for survival, fecundity and egg hatch in the red flourbeetle, Tribolium castaneum. Insect Biochem. Mol. Biol. 38, 959e962.

Baum, J., Bogaert, T., Clinton, W., Heck, G., Feldmann, P., Ilagan, O., Johnson, S.,Plaetinck, G., Munyikwa, T., Pleau, M., Vaughn, T., Roberts, J., 2007. Control of cole-opteran insect pests through RNA interference. Nat. Biotechnol. 25, 1322e1326.

Bolognesi, R., Arakane, Y., Muthukrishnan, S., Kramer, K.J., Terra, W.R., Ferreira, C.,2005. Sequences of cDNAs and expression of genes encoding chitin synthaseand chitinase in the midgut of Spodoptera frugiperda. Insect Biochem. Mol. Biol.35, 1249e1259.

Cohen, E., 2001. Chitin synthesis and inhibition: a revisit. Pest Manage. Sci. 57,946e950.

Gagou, M.E., Kapsetaki, M., Turberg, A., Kafetzopoulos, D., 2002. Stage-specificexpression of the chitin synthase DmeChSA and DmeChSB genes during theonset of Drosophila metamorphosis. Insect Biochem. Mol. Biol. 32, 141e146.

Gordon, K.H.J., Waterhouse, P.M., 2007. RNAi for insect-proof plants. Nat. Biotechnol.25, 1231e1232.

Hegedus, D., Erlandson, M., Gillott, C., Toprak, U., 2009. New insights into peri-trophic matrix synthesis, architecture, and function. Annu. Rev. Entomol. 54,285e302.

Hogenkamp, D.G., Arakane, Y., Zimoch, L., Merzendorfer, H., Kramer, K.J.,Beeman, R.W., Kanost, M.R., Specht, C.A., Muthukrishnan, S., 2005. Chitin syn-thase genes in Manduca sexta: characterization of a gut-specific transcript anddifferential tissue expression of alternately spliced mRNAs during development.Insect Biochem. Mol. Biol. 35, 529e540.

Ibrahim, G.H., Smartt, C.T., Kiley, L.M., Christensen, B.M., 2000. Cloning and char-acterization of a chitin synthase cDNA from the mosquito Aedes aegypti. InsectBiochem. Mol. Biol. 30, 1213e1222.

Kato, N., Mueller, C.R., Fuchs, J.F., Wessely, V., Lan, Q., Christensen, B.M., 2006.Regulatory mechanisms of chitin biosynthesis and roles of chitin in peritrophicmatrix formation in the midgut of adult Aedes aegypti. Insect Biochem. Mol.Biol. 36, 1e9.

Khajuria, C., Buschman, L.L., Chen, M.S., Muthukrishnan, S., Zhu, K.Y., 2010. A gut-specific chitinase gene essential for regulation of chitin content of peritrophicmatrix and growth of Ostrinia nubilalis larvae. Insect Biochem. Mol. Biol. 40,621e629.

Kumar, N.S., Tang, B., Chen, X., Tian, H., Zhang, W., 2008. Molecular cloning,expression pattern and comparative analysis of chitin synthase gene B in Spo-doptera exigua. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 149, 447e453.

Mao, Y., Cai, W., Wang, J., Hong, G., Tao, X., Wang, L., Huang, Y., Chen, X., 2007.Silencing a cotton bollworm P450 monooxygenase gene by plant-mediatedRNAi impairs larval tolerance of gossypol. Nat. Biotechnol. 25, 1307e1313.

Maue, L., Meissne, D., Merzendorfer, H., 2009. Purification of an active, oligomericchitin synthase complex from the midgut of the tobacco hornworm. InsectBiochem. Mol. Biol. 39l, 654e659.

Merzendorfer, H., 2006. Insect chitin synthases: a review. J. Comp. Physiol. B 176,1e16.

Merzendorfer, H., Zimoch, L., 2003. Chitin metabolism in insects: structure,function and regulation of chitin synthases and chitinases. J. Exp. Biol. 206,4393e4412.

Qu, M., Liu, T., Yang, J., Yang, Q., 2011. The gene, expression pattern and subcellularlocalization of chitin synthase B from the insect Ostrinia furnacalis. Biochem.Biophys. Res. Commun. 404, 302e307.

Shao, L., Devenport, M., Jacobs-Lorena, M., 2001. The peritrophic matrix of hema-tophagous insects. Arch. Insect Biochem. Physiol. 47, 119e125.

Terra, W.R., 2001. The origin and functions of the insect peritrophic membrane andperitrophic gel. Arch. Insect Biochem. Physiol. 46, 47e61.

Wang, P., Granados, R.R., 2001. Molecular structure of the peritrophic membrane(PM): identification of potential PM target sites for insect control. Arch. InsectBiochem. Physiol. 47, 110e118.

Zhang, J.Z., Liu, X.J., Zhang, J.Q., Li, D.Q., Sun, Yi., Guo, Y.P., Ma, E.B., Zhu, K.Y., 2010a.Silencing of two alternative splicing-derived mRNA variants of chitin synthase 1gene by RNAi is lethal to the oriental migratory locust. Insect Biochem. Mol.Biol. 40, 824e833.

Zhang, X., Zhang, J.Z., Zhu, K.Y., 2010b. Chitosan/double-stranded RNA nanoparticle-mediated RNA interference to silence chitin synthase genes through larvalfeeding in the African malaria mosquito (Anopheles gambiae). Insect Mol. Biol.19, 683e693.

Zhang, X., Zhang, J., Park, Y., Zhu, K.Y., 2012. Identification and characterization oftwo chitin synthase genes in African malaria mosquito, Anopheles gambiae.Insect Biochem. Mol. Biol. 42, 674e682.

Zhang, J.Z., Zhu, K.Y., 2006. Characterization of a chitin synthase cDNA and itsincreased mRNA level associated with decreased chitin synthesis in Anophelesquadrimaculatus exposed to diflubenzuron. Insect Biochem. Mol. Biol. 36,712e725.

Zimoch, L., Merzendorfer, H., 2002. Immunolocalization of chitin synthase in thetobacco hornworm. Cell. Tissue Res. 308, 287e297.