A

bscsptobwcrEaeia©

K

1

abta

ef

0d

Molecular Immunology 44 (2007) 722–731

Cloning and characterization of a novel C-type lectin fromZhikong scallop Chlamys farreri

Hao Wang a,b, Linsheng Song a,∗, Chenghua Li a,b, Jianmin Zhao a,b,Huan Zhang a,b, Duojiao Ni a,b, Wei Xu a,b

a Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, PR Chinab Graduate School, Chinese Academy of Sciences, Beijing 100039, PR China

Received 25 March 2006; received in revised form 24 April 2006; accepted 25 April 2006Available online 14 June 2006

bstract

C-type lectin is a family of Ca2+ dependent carbohydrate-recognition proteins which play crucial roles in the innate immunity of invertebratesy mediating the recognition of host cells to pathogens and clearing microinvaders as a pattern recognition protein (PRP). The cDNA of Zhikongcallop Chlamys farreri C-type lectin (designated CFLec-1) was cloned by expressed sequence tag (EST) and RACE techniques. The full-lengthDNA of CFLec-1 was 1785 bp, consisting of a 5′-terminal untranslated region (UTR) of 66 bp and an unusually long 3′ UTR of 1040 bp witheven polyadenylation signal sequences AATAAA and a poly(A) tail. The CFLec-1 cDNA encoded a polypeptide of 221 amino acids with autative signal peptide of 15 amino acid residues and a mature protein of 206 amino acids. Analysis of the protein domain features indicated aypical long-form carbohydrate-recognition domain (CRD) of 130 residues in the CFLec-1 deduced amino acid sequence. The expression patternf CFLec-1 transcripts in healthy and bacterial challenged scallops was studied by semi-quantitative RT-PCR. mRNA transcripts of CFLec-1 coulde mainly detected in the tissues of haemocytes, gill, gonad and mantle of unchallenged scallops, whereas the expression of CFLec-1 transcriptsas increased in all the tested tissues after heat-killed Vibrio anguillarum challenge. The temporal expression of CFLec-1 mRNA in haemolymph

hallenged by Micrococcus luteus and V. anguillarum was both up-regulated and reached the maximum level at 8 and 16 h post stimulation,espectively, and then dropped back to the original level. In order to investigate its immune functions, CFLec-1 was recombined and expressed inscherichia coli BL21(DE3)-pLysS as a fusion protein with thioredoxin. The recombinant CFLec-1 agglutinated bacteria E. coli JM109 in vitro,

2+

nd the agglutination was Ca dependent which could be inhibited by EDTA. But it did not agglutinate M. luteus, Candida lipolytica and animalrythrocytes including rabbit, rat, mouse, chicken, human group A, human group B, human group O. Meanwhile, the recombinant CFLec-1 couldnhibit the growth of both E. coli JM109 and M. luteus, but no inhibition activity against V. anguillarum. These result indicated that CFLec-1 wasconstitutive and inducible PRP which was involved in the reorganization and clearance of invaders in scallop.2006 Elsevier Ltd. All rights reserved.

; Rec

npcr

eywords: Chlamys farreri; C-type lectin; Innate immunity; mRNA expression

. Introduction

Innate immunity in both vertebrates and invertebrates isttracting a significant amount of attention. Invertebrates are

elieved to lack adaptive immunity and they rely exclusively onheir innate immune response to efficiently defend themselvesgainst a variety of pathogens (Loker et al., 2004). Innate immu-

∗ Corresponding author at: Institute of Oceanology, Chinese Academy of Sci-nces, 7 Nanhai Rd., Qingdao 266071, PR China. Tel.: +86 532 82898552;ax: +86 532 82880645.

E-mail address: lshsong@ms.qdio.ac.cn (L. Song).

sagaMipto

161-5890/$ – see front matter © 2006 Elsevier Ltd. All rights reserved.oi:10.1016/j.molimm.2006.04.015

ombinant expression; Antibacterial activity; Bacteria agglutination

ity is designed to recognize a few highly conserved structuresresent in many different microorganisms. The structures arealled pathogen-associated molecular patterns (PAMPs), whichepresent conserved molecular patterns that are essential for theurvival of the microbes, including lipopolyssacharide (LPS)nd peptidoglycan (PGN) from bacteria, �-1,3-glucan from fun-al cells, bacterial DNA, double-stranded RNA from virusesnd the related molecules (Janeway and Medzhitov, 2002;edzhitov and Janeway, 2002). Recognition of PAMPs in the

nnate immune system is mediated by a group of proteins namedattern recognition proteins (PRPs) or pattern recognition recep-ors (PRRs). PRPs recognize and bind to the conserved patternsn the surface of invading microorganisms. There are six sub-

Immu

gtbC2

tnisnVtptBvcptoaCl((t(llpC

moltv2pmfst

tCftmd(wstls

2

2c

iw1

eelwstwr8ptt4eh

2

stgysrif

2

ptcCcvc1a(l5

H. Wang et al. / Molecular

roups of PRP in invertebrate, peptidoglycan recognition pro-ein (PGRP), thioester-containing protein (TEP), Gram-negativeinding protein (GNBP), mulidomain scavenger receptor (SCR),-type lectin (CTL), and galectin (GALE) (Christophides et al.,002).

Lectin is one of the PRPs which can recognize and bind toerminal sugars on glycoproteins and glycolipids, and plays sig-ificant roles in nonself recognition and clearance of invadersn invertebrate immunity (Yu and Kanost, 2003). Based on theirtructures and functions, lectins are broadly classified as cal-exin, C-, L-, P-, I-, R- and S-type lectins (Savan et al., 2004;asta et al., 2004). C-type animal lectins represent an impor-

ant recognition mechanism for extensive oligosaccharides androteins in the serum and extracellular matrix, and for pro-eins and lipids on the surface of cells (Drickamer, 1988).inding of specific sugar structures by these lectins mediatesarious biological events, such as cell–cell adhesion, serum gly-oprotein turnover and innate immune responses to potentialathogens (Drickamer and Taylor, 1993). The members of C-ype lectin superfamily are defined as proteins that have at leastne carbohydrate-recognition domain (CRD) of ∼130 aminocid residues (Drickamer, 1988; Drickamer and Taylor, 1993).-type lectins are divided further into seventeen subgroups: (I)

ecticans, (II) the ASGR group, (III) collectins, (IV) selectins,V) NK receptors, (VI) the macrophage mannose receptor group,VII) Reg proteins, (VIII) the chondrolectin group, (IX) theetranectin group, (X) polycystin 1, (XI) attractin, (XII) EMBP,XIII) DGCR2, (XIV) the thrombomodulin group, (XV) Bim-ec, (XVI) SEEC, and (XVII) CBCP. Except group VII C-typeectins, which consist of only one CTLD and a short N-terminaleptide, the C-type lectins in other groups have one to severalTLDs in their multi-domain structures (Gready, 2005).

C-type lectins have been well-studied in vertebrates, andany C-type lectin genes have been identified and cloned. Even

nly a few genes have been cloned and the information is stillimited (Gready, 2005). C-type lectins in invertebrates are provedo be involved in various biological responses, for instance acti-ation of the proPO system (Yu et al., 1999; Yu and Kanost,000; Yu et al., 2005), antibacterial activity (Suzuki et al., 1990),romotion of phagocytosis (Luo et al., 2005), and nodule for-ation (Koizumi et al., 1999). To our knowledge, the molecular

eatures and functional studies of C-type lectins in invertebratestill remain deficient, and no C-type lectin genes have been iden-ified from marine mollusks to date.

Zhikong scallop Chlamys farreri is one of the most impor-ant species cultured widely in the northern coastal provinces inhina. Since the summer in 1997, a large-scale mortality of C.

arreri scallop has caused catastrophic losses to scallop aquacul-ure. Understanding the immune defense mechanisms of scallop

ay contribute to the development of management strategies forisease control and long-term sustainability of scallop farmingSong et al., 2006). The main objectives of the present studyere (1) to clone the full-length cDNA of the C-type lectin from

callop C. farreri, (2) to study the expression pattern includingemporal expression and tissue specific expression of the C-typeectin, and (3) to understand the roles of C-type lectin in thecallop’s immune responses.

petu

nology 44 (2007) 722–731 723

. Materials and methods

.1. Scallops, immune challenge and haemolymphollection

The scallop C. farreri was collected from a commercial farmn Qingdao, Shandong Province, China. Individuals with a fresheight of 24 ± 3 g were cultured in filtered aerated seawater at8–20 ◦C for 2 weeks before processing.

For the bacterial challenge experiment, 700 scallops weremployed and kept in fourteen aerated tanks (50 individuals inach tank). A 50 �l of live Vibrio anguillarum or Micrococcusuteus resuspended in 0.1 mol l−1 PBS (pH 6.4, O.D.600 = 0.4)as injected into the adductor muscles of scallops. Untreated

callops and scallops injected with 50 �l of PBS were used ashe blank and control group, respectively. The injected scallopsere returned to seawater tanks and fifteen individuals were

andomly sampled from each group at the time point of 2, 4, 6,, 16, and 32 h post injection. The haemolymph (about 0.5 mler individual) was collected using a syringe from the adduc-or muscles. The haemolymph from five scallops were pooledogether as one sample, and immediately centrifuged at 800 × g,◦C for 10 min to collect the haemocytes. Three replicates weremployed for each time point. Total RNA was extracted fromaemolymph samples using TRIzol reagent (invitrogen).

.2. cDNA library construction and EST analysis

A cDNA library was constructed from the whole body of thecallop challenged by V. anguillarum using the ZAP-cDNA syn-hesis kit and ZAP-cDNA GiapackIII gold cloning kit (Strata-ene). Random sequencing of the library using T3 primerielded 6935 successful sequencing reactions. BLAST analy-is of all the EST sequences revealed that one clone (clone no.scag0 005323; length: 475 bp) was homology to the previouslydentified C-type lectin, and this EST sequence was selected forurther cloning of the full-length cDNA of CFLec-1.

.3. Cloning the cDNA full length of CFLec-1 gene

Three gene specific primers, sense primer S1 and reverserimer R1 and R2 (Table 1) were designed based on the ESTo clone the full-length cDNA of CFLec-1 by rapid amplifi-ation of cDNA ends (RACE). PCR reaction to get 3′ end ofFLec-1 was performed in a PTC-100 programmable thermalontroller cycler (MJ research) by using sense primer S1 andector reverse primer T7 (Table 1) in a 25 �l reaction volumeontaining 2.5 �l 10 × PCR buffer, 2 �l dNTP (2.5 mmol l−1),.5 �l MgCl2 (25 mmol l−1), 1 �l of each primer (10 �mol l−1)nd 15.8 �l PCR-grade water, 0.2 �l (5 U �l−1) Taq polymeraseTakara) and 1 �l of cDNA mix. The PCR program was as fol-ows: 94 ◦C for 5 min, followed by 31 cycles of 94 ◦C for 45 s,8 ◦C for 30 s and 72 ◦C for 1 min. After the final cycles, sam-

les were incubated for a further 10 min at 72 ◦C for the finalxtension. The 5′ end of the gene was cloned with nested-PCRechnique. The PCR reaction was performed by the same vol-me described above for the 3′ amplification. For the first round

724 H. Wang et al. / Molecular Immu

Table 1Oligonucleotide primers used in the experiments

Primer name Sequence

CFLec-1S1 (forward) 5′-ACCTCCCGAGTCGTGTTT-3′R1 (reverse) 5′-ACACGACTCGGGAGGTTTCT-3′R2 (reverse) 5′-GGGCAGGTGTAGTCGTAGTAGC-3′R3 (reverse) 5′-TCACAGTGGATCGGTTGG-3′

�-actinAF (forward) 5′-TATGCCCTCCCTCACGCTAT-3′AR (reverse) 5′-GCCAGACTCGTCGTATTCCT-3′

Vector primersT3 (forward) 5′-AATTAACCCTCACTAAAGGG-3′T7 (reverse) 5′-GTAATACGACTCACTATAGGGC-3′

Sequencing primerM13-47 (forward) 5′-CGCCAGGGTT TTCCCAGTCACGAC-3′RV-M (reverse) 5′-GAGCGGATAACAATTTCACACAGG-3′

RecombinationReS (forward) 5′-CGCGGATCCTATATTTCAAAGCATGGT-

ACGGGAGATG-3′

PwFupa5a

pcfiLtta

2

ptfRap(

2

oqw

cOirwsa

MRtcoffm(2oPaCawbtf

2b

tTTtbfasEaGP(D

2p

aR

ReR (reverse) 5′-CGCGCGGCCGCTTAGCGTACAATCTCG-CAGATATAGAAC-3′

CR, 1 �l cDNA mix was used as template, and the primers usedere vector sense primer M13 and reversed primer R1 (Table 1).or the second round, 1 �l of the first round PCR product wassed as template, and a vector sense primer T3 and reversedrimer R2 (Table 1) were used. PCR reaction was performeds following: 94 ◦C 5 min, followed by 31 cycles of 94 ◦C 45 s,8 ◦C (first round) or 62 ◦C (second round) for 30 s, 72 ◦C 1 minnd 72 ◦C 10 min for the final extension.

The PCR products were gel-purified and cloned into theMD18-T simple vector (Takara). After transformed into theompetent cells of E. coli JM109, the recombinants were identi-ed through blue–white color selection in ampicillin-containingB plates. The positive clones were sequenced in both direc-

ions with sequencing primer M13-47 or RV-M (Table 1), andhe resulting sequences were verified and subjected to clusternalysis.

.4. Sequence analysis

The similarity searches were performed with the BLASTrogram at the National Center for Biotechnology Informa-ion (http://www.ncbi.nlm.nih.gov/blast). The protein motifseatures were predicted by Simple Modular Architectureesearch Tool (http://smart.embl-heidelberg.de/) (Letunic etl., 2006). Multiple alignment of the scallop CFLec-1 waserformed with the ClustalW Multiple Alignment programhttp://www.ebi.ac.uk/clustalW/) and optimized manually.

.5. Tissue specific expression of CFLec-1

The mRNA expression of CFLec-1 in different tissuesf healthy and challenged scallops was measured by semi-uantitative RT-PCR. The V. anguillarum in logarithmic phaseas heated at 65 ◦C for 15 min, and the heat-killed cells were

Itcp

nology 44 (2007) 722–731

ollected and resuspended in PBS to the concentration of.D.600 = 0.4. 50 �l of the bacterial suspension was injected

nto the adductor muscle of a scallop. The injected scallop wereeturned to seawater and cultured for another 8 h. Total RNAas extracted from different tissues of healthy and challenged

callops, respectively, including haemolymph, kidney, gill, hep-topancreas, adductor muscle, gonad and mantle.

The first-strand synthesis was carried out based on Promega-MLV RT Usage information using the DNase-treated totalNA as template. Reactions were incubated at 37 ◦C for 1 h,

erminated by heating at 95 ◦C for 5 min. Two CFLec-1 gene spe-ific primers S1 and R3 (Table 1) were used to amplify a productf 627 bp. A set of �-actin primers (Table 1) served as a controlor amount and quality of cDNA. All PCR reactions were per-ormed according to the following protocol: 1 �l of cDNA wasixed with 2 �l dNTPs (2.5 mmol l−1), 0.2 �l Taq polymerase

5 U �l−1), 1 �l of each gene specific primer (10 �mol l−1),.5 �l 10 × PCR buffer, 1.5 �l Mg2+ (25 mmol l−1) and 15.8 �lf PCR water. The PCR reactions were performed in a PTC-100rogrammable Thermal Controller (MJ research) with 5 mint 94 ◦C followed by 23 cycles (for �-actin) or 30 cycles (forFLec-1) of 94 ◦C for 30 s, 58 ◦C for 30 s, 72 ◦C for 45 s andfinal extension step at 72 ◦C for 10 min. The PCR productsere separated in 2% agarose gel and stained with ethidiumromide. To confirm the specificity of RT-PCR amplification,he RT-PCR products were purified from the gel and submittedor sequencing.

.6. The temporal mRNA expression of CFLec-1 afteracterial challenge

The expression of scallop CFLec-1 in haemocytes after bac-erial challenge was measured by semi-quantitative RT-PCR.he templates for RT-PCR were prepared as described above.he PCR reactions were performed in a PTC-100 programmable

hermal controller (MJ research) with 5 min at 94 ◦C followedy 23 cycles (for �-actin) or 28 cycles (for CFLec-1) of 94 ◦Cor 30 s, 58 ◦C for 30 s, 72 ◦C for 45 s and a final extension stept 72 ◦C for 10 min. After amplification, the PCR products wereeparated in 2% agarose gel and stained with ethidium bromide.lectrophoretic image and the densities of target bands werenalyzed using the Quantity one 1D analysis software in theel Doc 2000 System (Bio-Rad). The data obtained from RT-CR analysis were subjected to one-way analysis of varianceone-way ANOVA) followed by an unpaired, two-tailed t-test.ifferences were considered significant at P < 0.05.

.7. Recombinant expression of CFLec-1 and proteinurification

PCR fragment encoding the mature peptide of CFLec-1 wasmplified using Promega Taq polymerase with specific primerseS and ReR (Table 1). For the convenience of cloning, a BamH

site was added to the 5′ end of ReS and a Not I site was added

o the 5′ end of ReR after the stop codon. The PCR fragment wasloned into pMD18-T simple vector (Takara), and digested com-letely with restriction enzymes BamH I and Not I (NEB), and

Immu

t3wiataiItsfiNmggs(v

2

ro(5ontbbtgLc

2

tpMb2y2i2CwmDCTtb

maAT12stiw

3

3

ytclpcTwcos

iasl1pifIwTo

3

atsar3Sh

H. Wang et al. / Molecular

hen cloned into the BamH I/Not I sites of expression vector pET-2a (Novagen). The recombinant plasmid (pET-32a-CFLec-1)as transformed into E. coli BL21(DE3)-pLysS (Novagen). Pos-

tive clones were screened by PCR reactions with primers ReSnd ReR, and confirmed by nucleotide sequencing. Positiveransformants were incubated in 1 l of LB medium (50 �g ml−1

mpicillin and 34 �g ml−1 chloramphenicol) at 37 ◦C with shak-ng at 220 rpm until the culture reached O.D.600 of 0.5–0.7.PTG was then added to the medium at a final concentra-ion of 1 mmol l−1, and incubated for another 5 h under theame conditions. The recombinant CFLec-1 protein was puri-ed by nickel affinity chromatogramphy MagExtractor His-TagPK-700 (Toyobo) under native condition as described by theanufacturer. The purified protein was dialysed against TBS-

lycerol buffer (50 mmol l−1 Tris–HCl, 100 mmol l−1 NaCl, 5%lycerol, pH 7.5) at 4 ◦C overnight. The resultant protein waseparated electrophoretically on 12% SDS-polyacrylamide gelSDS-PAGE) according to the method of Laemmli (1970) andisualized with Coomassie brilliant blue R250.

.8. Antibacterial activity assay

The lysoplate assay was performed to assess the antibacte-ial activity of recombinant CFLec-1 according to the methodf Minagawa et al. (2001). The medium containing bacterialO.D.600 = 0.1) was poured onto 90 mm plates. Wells (diameter:mm) were cut into the freshly poured plates after solidificationf the agar. Approximately 10 �g recombinant protein and theegative control sample were dropped into individual wells inhe agar plates. The bacterial strains used to detect the antimicro-ial spectrum of recombinant CFLec-1 included Gram-negativeacteria E. coli JM109, V. anguillarum; and Gram-positive bac-eria M. luteus. Among the tested microorganisms, vibrio wasrown and assayed in marine broth 2216 at 25 ◦C and others inB agar at 30 ◦C. The TBS-glycerol buffer was used as negativeontrol.

.9. Bacteria agglutination and hemagglutination study

The bacteria agglutination assay was performed accordinghe method described by Yu et al. (1999). 4′-6-Diamidino-2-henylindole (DAPI) stained E. coli JM109, V. anguillarum,. Luteus, and Candida lipolytica were suspended in TBS

uffer (50 mmol l−1 Tris–HCl, 100 mmol l−1 NaCl, pH 7.5) at.5 × 109 cells ml−1 (for bacteria) or 2.5 × 108 cells ml−1 (foreast). A 10 �l of bacteria or yeast suspension was added to5 �l recombinant CFLec-1 solution (dissolved in TBS contain-ng 10 mmol l−1 CaCl2, final concentration of ∼80 �g ml−1), or5 �l BSA solution (dissolved in TBS containing 10 mmol l−1

aCl2, final concentration 1 mg ml−1) as control. The mixturesere incubated at room temperature for about 45 min. To deter-ine whether the agglutination of bacteria requires calcium,API-labeled E. coli JM109 was incubated with recombinant

FLec-1 (∼80 �g ml−1) in TBS-EDTA buffer (50 mmol l−1

ris–HCl, 100 mmol l−1 NaCl, 1 mmol l−1 EDTA, pH 7.5) underhe same condition described above. Cells were then observedy fluorescence microscopy (Carl Zeiss).

mCda

nology 44 (2007) 722–731 725

The hemagglutination assay was performed according to theethod described by Luo et al. (2005). Erythrocytes from the

nimals including rabbit, rat, mouse, chicken, human group, human group B, human group O were washed thrice withBS-Ca buffer (50 mmol l−1 Tris–HCl, 100 mmol l−1 NaCl,0 mmol l−1 CaCl2, pH 7.5), trypsinized, and then suspended in% TBS-Ca buffer. A 25 �l of two-fold serial diluted rCFLec-1olution in TBS-Ca buffer was mixed with 25 �l of the ery-hrocyte suspension in 96-well microtiter plate. The plate wasncubated for 1 h at room temperature and the hemagglutinationas observed under a microscope (Carl Zeiss).

. Result

.1. Isolation of full length cDNA of CFLec-1

Sequencing of Zhikong Scallop cDNA library with T3 primerielded 6935 EST sequences, which were clustered into 689 con-igs and 2191 singletons. Blastx analysis revealed that a 475 bplone (clone no: rscag0 005323) was similar to vertebrate C-typeectins. Based on the sequence of this EST, three gene specificrimers (S1, R1, and R2) were designed to clone the full-lengthDNA of CFLec-1. By 3′ RACE technique with primers S1 and7, a 1285 bp fragment was amplified. The product of 5′ RACEas 322 bp. A 1785 bp nucleotide sequence representing the

omplete cDNA sequence of CFLec-1 cDNA was obtained byverlapping the two fragments mentioned above with the ESTequence.

The full-length cDNA sequence of CFLec-1 was depositedn GenBank under accession no. DQ209290, and the deducedmino acid sequence was shown in Fig. 1. The completeequence of CFLec-1 cDNA contained a 5′-terminal untrans-ated region (UTR) of 66 bp, an unusually long 3′ UTR of040 bp with seven AATAAA polyadenylation signal sites and aoly(A) tail, and an open reading frame (ORF) of 666 bp encod-ng a polypeptide of 221 amino acids. The N-terminus had theeatures consistent with a signal peptide as defined by the PSORTI program analysis (National Institute for Basic Biology, Japan)ith a putative cleavage site located after position 28 (ASQ-YI).he deduced mature protein was of 206 amino acids with a pIf 5.12 and a theoretical mass of 23.49 kDa.

.2. Homology analysis and characterization of CFLec-1

CFLec-1 sequence was blast against the database of NCBInd no strictly similar sequences were found, which indicatedhat CFLec-1 was a novel protein. The deduced amino acidequence of CRD region in CFLec-1 shared 28% identitynd 47% positivity with the predicted macrophage mannoseeceptor in Gallus gallus, which contains eight CTLDs, and1% identity with C-type lectin receptor C from Atlantic salmonalmo salar. According to the Blastp analysis, several proteinsarboring CRDs (Fig. 2) were picked out to perform the align-

ent analysis. All these proteins contain at least one long-formRD, which has four disulfide-bonded cysteine residues thatefine the CRD and two additional cysteine residues at themino terminus. The conserved residues in these proteins were

726 H. Wang et al. / Molecular Immunology 44 (2007) 722–731

Fig. 1. Nucleotide and decuced amino acid sequences of CFLec-1. The amino acid residues in the mature protein are assigned positive numbers, and those in signalpeptide are assigned negative numbers. Conserved cysteine residues that define the short-form C-type lectin domain are marked with �, whereas two extra cysteiner ng speb left m

sbsts

S

fs(

esidues in the long-form are marked with �. The EPD motif for the ligand-bindiy double underline. The nucleotides and amino acids are numbered along the

hown in Fig. 2 and residues that presumed to participate in theinding of carbohydrate and Ca2+ were marked. The sequenceimilarity and the common structure features strongly suggested

hat CFLec-1 should be a new member of the C-type lectinuperfamily.

The protein domain features of CFLec-1 were analyzed usingMART program, which indicated the presence of one long-

tttb

cificity was boxed. The first polyadenylation sequence AATAAA was indicatedargin.

orm CRD at the C-terminus. There were several signatureequences of the C-type lectin family: the four cysteine residuesCys104, Cys177, Cys193, Cys202) involved in the formation of

he CRD internal disulfides bridges were completely conserved;he other two cysteine residues (Cys74, Cys85) at N-terminus andhe EPD motif (Glu169–Pro170–Asp171) for determining ligand-inding specificity.

H. Wang et al. / Molecular Immunology 44 (2007) 722–731 727

Fig. 2. Multi sequence alignment of the CRD of CFLec-1 with other genes’ CRDs. Salmon: salmon C-type lectin receptor (AAT77222); mouse: mouse C-type lectind toxoct wo ex

3

ewmstwalg

3b

ola6f

Fsc

cetelgsgamtCtcant difference in CFLec-1 gene expression at 2, 4, 6, and 8 hpost injection in the M. luteus challenged group and from 4to 32 h post injection in the V. anguillarum challenged group(P < 0.05).

omain family 4, member a2 (NP 036129); Rat: RAT Fc receptor (AAH78749);hat define the short-form C-type lectin domain were marked with �, whereas t

.3. Tissue distribution of CFLec-1 mRNA

RT-PCR was employed to determine the tissue specificxpression of the CFLec-1 mRNA. The CFLec-1 transcriptsere mainly detected in the tissues of gill and gonad, andarginally detectable in haemolymph and mantle, while there

eemed to be no signal in the kidney, hepatopancreas and adduc-or muscle of healthy scallops. However, in scallops injectedith heat-killed V. anguillarum, the CFLec-1 mRNA levels in

ll the detected tissues were higher than those in the unchal-enged scallops, especially in the tissues of haemolymph, gill,onad and mantle (Fig. 3).

.4. The temporal expression of CFLec-1 mRNA afteracterial challenge

The temporal expression of CFLec-1 gene after M. luteusr V. anguillarum challenge was shown in Fig. 4. In the M.

uteus challenged group, there was an up-regulation in the rel-tive expression level of CFLec-1 in haemolymph. At 2 andh after bacterial challenge, there was an 11.3-fold and 18.3-

old increase in the relative abundance of the CFLec-1 mRNA

ig. 3. . Tissue distribution of CFLec-1 transcript revealed by RT-PCR. �-Actinerved as an internal control. (a) Blank tissues; (b) heat-killed V. anguillarumhallenged tissues. HE: haemolymph; HP: hepatoancreas.

FlCwTa

ara: toxocara C-type lectin Tc-ctl-4 (AAD31000). Conserved cysteine residuestra cysteine residues in the long-form were marked with �.

ompared with control. At 8 h post injection, the CFLec-1 genexpression reached the maximum level and was 25.4-fold higherhan that observed in control scallops. As time progresses, thexpression of CFLec-1 transcript dropped back to the originalevel at 16 h post injection. In the V. anguillarum challengedroup, no significant difference was observed in the expres-ion level of CFLec-1 among the blank, control and challengedroup. At 4, 6, and 8 h after bacterial challenge, there wassignificant increase in the relative abundance of CFLec-1RNA. At 16 h, the expression increased 15.7-fold and reached

he highest level. As the time progresses, the expression ofFLec-1 transcription dropped at 32 h. An unpaired, two-tailed

-test with the control and challenged groups showed signifi-

ig. 4. The temporal expression of CFLec-1 mRNA in hemolymph after M.uteus and V. anguillarum stimulation. Data was expressed as the ratio of theFLec-1 mRNA to the �-actin mRNA. The relative CFLec-1 expression levelas determined for each group and values were shown as means ± S.E., n = 3.he unstimulated scallops and scallops injected with PBS were used as the blanknd control groups.

728 H. Wang et al. / Molecular Immu

Fig. 5. (A) SDS-PAGE analysis of the recombinant CFLec-1 in E. coliBL21(DE3)-pLysS. After electrophoresis, the gel was visualized by CoomassiebCad

3

fabawTl

3

r

etawl

3

rEhaCtT(lrm

d

4

taedJdiaswsi

isat2fs2il

tsw

rilliant blue R250. Line M: protein molecular standard; line 1: pET-32a-FLec-1, not induced; line 2: pET-32a-CFLec-1, IPTG induced. (B) SDS-PAGEnalysis of the purified recombinant CFLec-1. line M: protein molecular stan-ard; line 1: purified recombinant CFLec-1.

.5. Recombinant expression of CFLec-1 in E. coli

The recombinant plasmid pET-32a-CFLec-1 was trans-ormed and expressed in E. coli BL21(DE3)-pLysS as describedbove. After IPTG induction, the whole cell lysate was analysedy SDS-PAGE and the thioredoxin fusion protein rCFLec-1 gavedistinct band with a molecular weight of ∼40 kDa (Fig. 5),hich was in accordance with the predicted molecular mass.he rCFLec-1was purified under native condition, and then dia-

yzed against TBS-glycerol buffer.

.6. Antibacterial activity study

From the radius of the antimicrobial zone, it was found thatecombinant CFLec-1 harbored remarkable in vitro inhibitive

fCcE

nology 44 (2007) 722–731

ffect on tested Gram-positive bacteria M. Luteus. The radius ofhe antimicrobial zone was about 11 mm. While the lytic activitygainst Gram-negative bacteria E. coli JM109 was relativelyeak. The radius of the antimicrobial zone was about 7 mm. No

ytic effect was observed on V. anguillarum.

.7. Agglutination assay

Bacteria and yeast were used to test the agglutination ofCFLec-1 toward the surface of different microorganisms. When. coli JM109 was incubated with rCFLec-1 at a concentrationigher than 10 �g ml−1, large agglutination could be observed,nd the size of aggregates of bacteria increased with higherFLec-1 concentration. No agglutination could be observed in

he BSA control. When the E. coli JM109 was incubated withBS-EDTA, the bacteria agglutination was completely inhibited

Fig. 6). No obvious agglutination could be observed when M.uteus and C. lipolytica were incubated with rCFLec-1. Theseesults indicated that rCFLec-1 mainly recognizes the surfaceolecules of Gram-negative bacteria.For the hemagglutination assay, no obvious agglutination was

etected on all examined animal erythrocytes (data not shown).

. Discussion

Distinguishing self from nonself is an important process inhe immune response. In acquired immune system of vertebrates,ntibody molecules and T cell receptors can recognize a vari-ty of antigens and function as recognition molecules to triggerifferent immune responses (Janeway, 1989; Janeway, 1992;aneway and Medzhitov, 2002; Yu et al., 2005). Invertebrateso not have antibodies and, therefore they have to rely on theirnnate immune response to efficiently defend themselves againstvariety of pathogens (Loker et al., 2004). Recognition of non-

elf materials in innate immune is mediated by a group of PRPs,hich could recognize and bind to different molecules on the

urface of invading microorganisms and then trigger a series ofmmune responses.

C-type lectins are a group of PRPs, which are involved in var-ous pathways in the innate immune system (Weis et al., 1998),uch as recognition, neutralization and clearance of pathogens,ctivation of prophenoloxidase, nodule formation and opsoniza-ion (Koizumi et al., 1999; Luo et al., 2005, 2003; Yu et al., 1999,005; Yu and Kanost, 2000). However, molecular features andunctional studies of C-type lectin in marine invertebrates aretill deficient compared with those of vertebrates (Luo et al.,005). To our knowledge, only a few C-type lectins have beendentified in marine invertebrate and no information about mol-usk C-type lectin is available.

In the present study, the first mollusk cDNA encoding a C-ype lectin (designated CFLec-1) was identified from Zhikongcallop, C. farreri by EST and RACE techniques. Consistentith other known long-form C-type lectins, CFLec-1 contained

our conserved cysteine residues involved in the formation of theRD internal disulfides bridges, two additional disulfide-bondedysteine residues at the N-terminus and the ligand-binding motifPD (Day, 1994; Drickamer, 1988; Drickamer and Taylor,

H. Wang et al. / Molecular Immunology 44 (2007) 722–731 729

F rCF(

1smeEistaiCswwD1Aod

paoulsrrp

sr1CcC

Clghsai

otlfi1ailir

ig. 6. The agglutination of Gram-negative bacteria E. coli JM109 induced by1 mg ml−1); (D) rCFLec-1 (∼80 �g mL−1) in TBS-EDTA.

993). The CRDs with EPN motif binds mannose (or similarugar has the equatorial 3-OH and 4-OH), whereas the QPDotif binds galactose or similar sugar with the axial 3-OH and

quatorial 4-OH (Weis et al., 1998). However, CFLec-1 had anPD motif on the second Ca2+ binding site, which was also found

n C-type lectin receptors (SCLRA and SCLRB) from Atlanticalmon (Soanes et al., 2004). With Asp in the place of Asn athe third position, a hydrogen bond donor was replaced with ancceptor, which perhaps influenced the specific recognition abil-ty towards carbohydrate. Another distinct feature of the CRD inFLec-1 was the substitution of two residues in the highly con-

erved WND sequences. In CFLec-1, the Trp and Asn residuesere replaced by Met and Ile, respectively. Similar phenomenaere also reported in Caenorhabditis elegans (Drickamer andodd, 1999), sea urchin Anthocidaris crassispina (Giga et al.,987), barnacle Megabalanus rosa (Takamatsu et al., 1994), andtlantic salmon S. salar (Soanes et al., 2004). The substitutionf highly conserved residues in CFLec-1 suggested the greativergence of C-type lectin in marine invertebrates.

In addition, the 3′-UTR region of CFLec-1 contained fourentameric sequences of AUUUA which had been proved to betypical class of adenylate uridylate-rich elements (AREs) in

ther animals. The AREs, an important cis-acting element, wassually found in early response genes that regulated cell pro-iferation and responses to exogenous. The changes in mRNA

tability mediated by AREs are important processes that areequired for transient responses such as cellular growth, immuneesponse, cardiovascular toning and external stress-mediatedathways. A common trait of the ARE-mRNAs is that they are

c1so

Lec-1. (A) rCFLec-1 (∼20 �g ml−1); (B) rCFLec-1 (∼80 �g ml−1); (C) BSA

hort-lived, and thus they will degrade rapidly once their criticaloles in gene regulation ceased (Bakheet et al., 2001; Chen et al.,995; Lagnado et al., 1994). The existence of potential AREs inFLec-1 indicated that the turnover rate of CFLec-1 was strictlyontrolled. This deduction was consistent with the result of theFLec-1 expression in haemocytes after bacterial challenge.

Semi-quantitative RT-PCR analysis demonstrated thatFLec-1 was moderately expressed in the gill and gonad, very

ow expressed in hemocyte and mantle in the unchallengedroup. After heat-killed V. anguillarum challenge, remarkablyigher expression levels could be observed in all detected tis-ues. These results suggested that CFLec-1 was a constitutivend inducible acute-phase protein that could play a crucial rolen mucosal defense.

A clear time-dependent pattern of CFLec-1 expression wasbserved in haemocytes after bacterial challenge. There werewo distinct phases in the immune responses of scallop to M.uteus challenge. During the first phase, corresponding to therst 2–8 h post injection, the relative expression of CFLec-mRNA was up-regulated and reached the maximum level

t 8 h post injection. As time progresses, 16 and 32 h postnjection, the expression of CFLec-1 dropped to the originalevel. The result suggested that CFLec-1 could be markedlynduced by M. luteus stimulation and perhaps involved in theecognition and clearance of invaded bacteria. This result was

onsistent with the antibacterial assay which proved rCFLec-had the inhibitive activity towards M. luteus in the in vitro

tudy. In the V. anguilarum challenged group, the expressionf CFLec-1 was also up-regulated, but the expression pattern

7 Immu

wtMag

FltoidcweaimOhnlaoaas

hpmcgE1(paaEitCotWBot1iiGiriV

A

thtdSHpfa

R

B

C

C

D

D

D

D

D

D

F

G

G

J

J

30 H. Wang et al. / Molecular

as different from the M. luteus challenged group. As a whole,he expression level in V. anguilarum was lower than that in

. luteus challenged group, and the highest level occurredt 16 h post injection rather than 8 h in M. luteus challengedroup.

C-type lectins exist widely in vertebrates and invertebrates.or example, more than 100 human proteins containing C-type

ectin like domains have been identified. Approximately half ofhem were supposed to function as C-type CRDs, and only somef them were Ca2+-dependent (Dodd and Drickamer, 2001). Innvertebrates, most of the identified C-type lectins were Ca2+-ependent. The rCFLec-1 could agglutinate Gram-negative E.oli JM109, and the agglutination was calcium dependent whichas same to some of the other invertebrate C-type lectins. How-

ver, rCFLec-1 could not agglutinate Gram-negative bacteria V.nguillarum, Gram-positive bacteria M. Lutes, yeast C. lipolyt-ca, and all examined animal erythrocytes, including rabbit, rat,

ouse, chicken, human group A, human group B, human group. The C-type lectin immulectin-1 identified from tobaccoorn worm could agglutinate Gram-positive bacteria, Gram-egative bacteria and yeast. The newly reported shrimp C-typeectin PmLec had a strong hemagglutinating and bacterial-gglutinating activity. Both of the C-type lectins described hadpsonic effect that enhanced haemocyte phagocytosis (Luo etl., 2005; Yu et al., 1999). The difference of the agglutinationbility perhaps related to the EPD motif on the Ca2+ bindingite 2.

In addition to the agglutination activity, some C-type lectinsave been reported to have the antibacterial activity. For exam-le, the C-type lectin purified from Tunicate Polyandrocarpaisakiensis displayed strong antibacterial activity even at the

oncentration of 1 �g ml−1 (Suzuki et al., 1990). The con-lutinin from rat demonstrated antibacterial activity against. coli and S. typhimurium in vitro (Friis-Christiansen et al.,990). In the present study, the recombinant protein of CFLec-1rCFLec-1) displayed remarkable inhibitive effect on Gram-ositive bacteria M. lutes and relatively weak lytic activitygainst Gram-negative bacteria E. coli JM109. The antibacterialctivity of rCFLec-1 was also observed when it was expressed in. coli strain BL21(DE3). Firstly, the rCFLec-1 was expressed

n BL21(DE3), but the bacteria stopped growing after the addi-ion of IPTG, and the culture became clear. It suggested thatFLec-1 was toxic to the host bacteria. The same result wasbtained in the recombinant expression of immulectin-1 fromobacco hornworm in E. coli strain XL1-blue (Yu et al., 1999).

hen the host stain for rCFLec-1 expression was changed toL21(DE3)-pLysS, which could inhibite the “leaky expression”f the insert gene, the inhibitive effect on bacteria growth andhe cell lysing did not happen again (Dubendorff and Studier,991). These results indicated that CFLec-1 not only recognizedts corresponding ligands but also participated in clearance ofnvading bacteria. Interestingly, rCFLec-1 did not agglutinateram-negative bacteria V. anguilarum, and it did not either

nhibit the growth of V. anguilarum. Further investigation isequired to find if it is related to the virulence or pathogenic-ty of V. anguillarum, and the susceptibility of scallop to. anguillarum.

J

K

nology 44 (2007) 722–731

cknowledgements

The authors would like to thank Miss Jing Ren, other labora-ory members and Prof. Xiao-qiang Yu for technical advice andelpful discussions, and Prof. Zhaolan Mo for kindly providinghe Vibrio anguillarum Strain, and Dr. Luping Zhang from Qing-ao health hospital for kindly providing human erythrocytes.equencing of the scallop cDNA library was conducted at theuada Genomic Center (Beijing, China). This research was sup-orted by 863 High Technology Project (No. 2002AA626020)rom the Chinese Ministry of Science and Technology, China,nd a grant (No. 40276045) from NSFC to Dr. Linsheng Song.

eferences

akheet, T., Frevel, M., Williams, B.R., Greer, W., Khabar, K.S., 2001. ARED:human AU-rich element-containing mRNA database reveals an unexpect-edly diverse functional repertoire of encoded proteins. Nucleic Acids Res.29, 246–254.

hen, Q., Adams, C.C., Usack, L., Yang, J., Monde, R.A., Stern, D.B., 1995.An AU-rich element in the 3′ untranslated region of the spinach chloroplastpetD gene participates in sequence-specific RNA-protein complex forma-tion. Mol. Cell Biol. 15, 2010–2018.

hristophides, G.K., Zdobnov, E., Barillas-Mury, C., Birney, E., Blandin, S.,Blass, C., Brey, P.T., Collins, F.H., Danielli, A., Dimopoulos, G., Hetru,C., Hoa, N.T., Hoffmann, J.A., Kanzok, S.M., Letunic, I., Levashina, E.A.,Loukeris, T.G., Lycett, G., Meister, S., Michel, K., Moita, L.F., Muller, H.M.,Osta, M.A., Paskewitz, S.M., Reichhart, J.M., Rzhetsky, A., Troxler, L.,Vernick, K.D., Vlachou, D., Volz, J., von Mering, C., Xu, J., Zheng, L.,Bork, P., Kafatos, F.C., 2002. Immunity-related genes and gene families inAnopheles gambiae. Science 298, 159–165.

ay, A.J., 1994. The C-type carbohydrate recognition domain (CRD) superfam-ily. Biochem. Soc. Trans. 22, 83–88.

odd, R.B., Drickamer, K., 2001. Lectin-like proteins in model organisms:implications for evolution of carbohydrate-binding activity. Glycobiology11, 71R–79R.

rickamer, K., 1988. Two distinct classes of carbohydrate-recognition domainsin animal lectins. J. Biol. Chem. 263, 9557–9560.

rickamer, K., Dodd, R.B., 1999. C-Type lectin-like domains in Caenorhabditiselegans: predictions from the complete genome sequence. Glycobiology 9,1357–1369.

rickamer, K., Taylor, M.E., 1993. Biology of animal lectins. Ann. Rev. CellBiol. 9, 237–264.

ubendorff, J.W., Studier, F.W., 1991. Controlling basal expression in aninducible T7 expression system by blocking the target T7 promoter withlac repressor. J. Mol. Biol. 219, 45–59.

riis-Christiansen, P., Thiel, S., Svehag, S.E., Dessau, R., Svendsen, P., Ander-sen, O., Laursen, S.B., Jensenius, J.C., 1990. In vivo and in vitro antibacterialactivity of conglutinin, a mammalian plasma lectin. Scand. J. Immunol. 31,453–460.

iga, Y., Ikai, A., Takahashi, K., 1987. The complete amino acid sequence ofechinoidin, a lectin from the coelomic fluid of the sea urchin Anthocidariscrassispina. Homologies with mammalian and insect lectins. J. Biol. Chem.262, 6197–6203.

ready, A.N.Z.A.J.E., 2005. The C-type lectin-like domain superfamily. FEBSJ. 272, 6179–6217.

aneway Jr., C.A., 1989. Introduction: T-cell:B-cell interaction. Semin.Immunol. 1, 1–3.

aneway Jr., C.A., 1992. The immune system evolved to discriminate infectiousnonself from noninfectious self. Immunol. Today 13, 11–16.

aneway Jr., C.A., Medzhitov, R., 2002. Innate immune recognition. Ann. Rev.Immunol. 20, 197–216.

oizumi, N., Imamura, M., Kadotani, T., Yaoi, K., Iwahana, H., Sato, R., 1999.The lipopolysaccharide-binding protein participating in hemocyte noduleformation in the silkworm Bombyx mori is a novel member of the C-

Immu

L

L

L

L

L

L

M

M

S

S

S

S

T

V

W

Y

Y

Y

H. Wang et al. / Molecular

type lectin superfamily with two different tandem carbohydrate-recognitiondomains. FEBS Lett. 443, 139–143.

agnado, C.A., Brown, C.Y., Goodall, G.J., 1994. AUUUA is not sufficient topromote poly(A) shortening and degradation of an mRNA: the functionalsequence within AU-rich elements may be UUAUUUA(U/A)(U/A). Mol.Cell Biol. 14, 7984–7995.

aemmli, U.K., 1970. Cleavage of structural proteins during the assembly ofbacteriophage T4. Nature 227, 680–685.

etunic, I., Copley, R.R., Pils, B., Pinkert, S., Schultz, J., Bork, P., 2006. SMART5: domains in the context of genomes and networks. Nucleic Acids Res. 34,D257–D260.

oker, E.S., Adema, C.M., Zhang, S.M., Kepler, T.B., 2004. Invertebrateimmune systems—not homogeneous, not simple, not well understood.Immunol. Rev. 198, 10–24.

uo, T., Yang, H., Li, F., Zhang, X., Xu, X., 2005. Purification, characterizationand cDNA cloning of a novel lipopolysaccharide-binding lectin from theshrimp Penaeus monodon. Dev. Comp. Immunol.

uo, T., Zhang, X., Shao, Z., Xu, X., 2003. PmAV, a novel gene involved invirus resistance of shrimp Penaeus monodon. FEBS Lett. 551, 53–57.

edzhitov, R., Janeway Jr., C.A., 2002. Decoding the patterns of self and nonselfby the innate immune system. Science 296, 298–300.

inagawa, S., Hikima, J., Hirono, I., Aoki, T., Mori, H., 2001. Expressionof Japanese flounder C-type lysozyme cDNA in insect cells. Dev. Comp.Immunol. 25, 439–445.

avan, R., Endo, M., Sakai, M., 2004. Characterization of a new C-type lectinfrom common carp Cyprinus carpio. Mol. Immunol. 41, 891–899.

oanes, K.H., Figuereido, K., Richards, R.C., Mattatall, N.R., Ewart, K.V.,2004. Sequence and expression of C-type lectin receptors in Atlantic salmon(Salmo salar). Immunogenetics 56, 572–584.

Y

nology 44 (2007) 722–731 731

ong, L., Zou, H., Chang, Y., Xu, W., Wu, L., 2006. The cDNAcloning and mRNA expression of a potential selenium-binding proteingene in the scallop Chlamys farreri. Dev. Comp. Immunol. 30, 265–273.

uzuki, T., Takagi, T., Furukohri, T., Kawamura, K., Nakauchi, M., 1990. Acalcium-dependent galactose-binding lectin from the tunicate Polyandro-carpa misakiensis. Isolation, characterization, and amino acid sequence. J.Biol. Chem. 265, 1274–1281.

akamatsu, N., Heishi, M., Muramoto, K., Kamiya, H., Shiba, T., 1994. Cloningand analysis of the gene encoding lectin from the acorn barnacle Megabal-anus rosa. Gene 150, 407–408.

asta, G.R., Ahmed, H., Odom, E.W., 2004. Structural and functional diver-sity of lectin repertoires in invertebrates, protochordates and ectothermicvertebrates. Curr. Opin. Struct. Biol. 14, 617–630.

eis, W.I., Taylor, M.E., Drickamer, K., 1998. The C-type lectin superfamilyin the immune system. Immunol. Rev. 163, 19–34.

u, X.Q., Gan, H., Kanost, M.R., 1999. Immulectin, an inducible C-type lectinfrom an insect, Manduca sexta, stimulates activation of plasma prophenoloxidase. Insect Biochem. Mol. Biol. 29, 585–597.

u, X.Q., Kanost, M.R., 2000. Immulectin-2, a lipopolysaccharide-specificlectin from an insect, Manduca sexta, is induced in response to Gram-negative bacteria. J. Biol. Chem. 275, 37373–37381.

u, X.Q., Kanost, M.R., 2003. Manduca sexta lipopolysaccharide-specificimmulectin-2 protects larvae from bacterial infection. Dev. Comp. Immunol.

27, 189–196.

u, X.Q., Tracy, M.E., Ling, E., Scholz, F.R., Trenczek, T., 2005. A novel C-type immulectin-3 from Manduca sexta is translocated from hemolymphinto the cytoplasm of hemocytes. Insect Biochem. Mol. Biol. 35, 285–295.