at SciVerse ScienceDirect

Fish & Shellfish Immunology 33 (2012) 1276e1284

Contents lists available

Fish & Shellfish Immunology

journal homepage: www.elsevier .com/locate/ fs i

A four-domain Kunitz-type proteinase inhibitor from Solen grandis is implicated inimmune response

Xiumei Wei a, Jialong Yang b,*, Jianmin Yang a, Xiangquan Liu a,**, Meijun Liu a, Dinglong Yang a, Jie Xu a,Xiaoke Hu b

a Shandong Provincial Key Laboratory of Restoration for Marine Ecology, Shandong Marine Fisheries Research Institute, Yantai 264006, Chinab Yantai Institute of Coastal Zone Research, Chinese Academy of Sciences, 17 Chunhui Rd, Yantai 264003, China

a r t i c l e i n f o

Article history:Received 28 May 2012Received in revised form11 September 2012Accepted 11 September 2012Available online 26 September 2012

Keywords:Solen grandisKunitz-type proteinase inhibitorInnate immunityReal-time PCRInhibitory activity

* Corresponding author. Tel.: þ86 535 2109150.** Corresponding author. Tel.: þ86 535 6117033.

E-mail addresses: jlyang@yic.ac.cn (J. Yang), lxq68

1050-4648/$ e see front matter � 2012 Elsevier Ltd.http://dx.doi.org/10.1016/j.fsi.2012.09.015

a b s t r a c t

Serine proteinase inhibitor (SPI) serves as a negative regulator in immune signal pathway by restrainingthe activities of serine proteinase (SP) and plays an essential role in the innate immunity. In the presentstudy, a Kunitz-type SPI was identified from the mollusk razor clam Solen grandis (designated asSgKunitz). The full-length cDNA of SgKunitz was of 1284 bp, containing an open reading frame (ORF) of768 bp. The ORF encoded four Kunitz domains, and their amino acids were well conserved whencompared with those in other Kunitz-type SPIs, especially the six cysteines involved in forming of threedisulfide bridges in each domain. In addition, the tertiary structure of all the four domains adopteda typical model of Kunitz-type SPI family, indicating SgKunitz was a new member of Kunitz-type SPIsuperfamily. The mRNA transcripts of SgKunitz were detected in all tested tissues of razor clam, includingmuscle, mantle, gonad, gill, hepatopancreas and hemocytes, and with the highest expression level in gill.When the razor clams were stimulated by LPS, PGN or b-1, 3-glucan, the expression level of SgKunitzmRNA in hemocytes was significantly up-regulated (P < 0.01), suggesting SgKunitz might involved in theprocesses of inhibiting the activity of SPs during the immune responses triggered by various pathogens.Furthermore, the recombinant protein of SgKunitz could effectively inhibit the activities of SP trypsin andchymotrypsin in vitro. The present results suggested SgKunitz could serve as an inhibitor of SP involvingin the immune response of S. grandis, and provided helpful evidences to understand the regulationmechanism of immune signal pathway in mollusk.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Serine proteinase inhibitors (SPIs) have been widely found inmulticellular organisms, and play various and crucial roles innumerous biological processes, such as blood coagulation, inflam-mation, development, and activation of prophenoloxidase andcomplement system [1e6]. As suggested by their name, SPIsparticipate in these processes by limiting the excessive proteolyticactivity of serine proteases (SPs). Such precise regulation is quiteimportant, for once the balance between SPs and SPIs is broken, theenzymes activity out of control might be fatal to the cells ororganisms [7]. Based on the primary sequence, tertiary structureand mechanisms of binding motifs, SPIs have been classified into

08@163.com (X. Liu).

All rights reserved.

several families [8], and among of them, family of kazal, kunitz,serpin and a-macroglobulin have been well characterized [9,10].

Kunitz-type SPIs are present in a variety of animals, includingsea anemone, snails, ectoparasites, snake venom, horse crab, fly,tick and bovine [11e15]. This family of SPIs is characterized by oneor more Kunitz domain in their molecules. Each Kunitz domainconsists of approximately 60 amino acids forming a compact aþbstructure, stabilizing by three conserved disulfide bridges [16].Kunitz-type proteinase inhibitors exhibit inhibitory activity againstthe common SPs, such as trypsin, chymotrypsin or thrombin [17].The inhibitory activity toward SPs is closely associated with the P1amino acid of the reactive site and amino acid sequencesurrounding the region that interacts with the proteases [8]. Forexample, inhibitors with a lysine or arginine at their reactive sitescan inhibit trypsin or trypsin-like enzymes, while inhibitors withtyrosine, phenylalanine, leucine or methionine at the relevantposition tend to inhibit chymotrypsin and chymotrypsin-likeenzymes [8].

Table 1Primers used in the present study.

Primer Sequence (50e30)

Oligo(dT)-adapter GGCCACGCGTCGACTAGTACT17M13-F (forward) GTAAAACGACGGCCAGM13-R (reverse) CAGGAAACAGCTATGACCSgKunitz RTF (forward) GTGAAGGGGCTTTTCCGATGTSgKunitz RTR (reverse) TCCTTAGTAATGAAGCGGTTGGb-actin AF (forward) TGTACGCCAACACTGTCCTGTCb-actin AR (reverse) CATCGTATTCCTGCTTGCTGATCSgKunitz REF (forward) GATGCTTTTGACGTTGAAGTTAGGASgKunitz RER (reverse) TGCGCTCTTACATTCTCTTTCACAAT7 Primer TAATACGACTCACTATAGGG

X. Wei et al. / Fish & Shellfish Immunology 33 (2012) 1276e1284 1277

Several families of SPIs including Kunitz-type proteinaseinhibitors are believed to involve in the immune signal pathway bymodulating the proteolytic activities of SPs, owing to their prom-inent inhibitory abilities [9]. For example, Serpin-4 and Serpin-5from Manduca sexta function to regulate the proPO-activationpathway by inhibiting three hemolymph proteinases upstream ofthe pathway [6], while SPIs in Drosophila hemolymph are believedto regulate the activity of the cytokine spatzle, which is associatewith the activation of Toll signal pathway and the productionof antimicrobial peptides “Drosomycin” against Gram-positivebacteria and fungi [9,18].

In addition, some SPIs also serve as an inhibitor to the microbialproteinase and protect the animal from infection by pathogens [19].In recent years, increasing SPIs have been identified from marineinvertebrates, and most of them are proved to involve in theimmune response. For instance, the mRNA expression of manykazal-type and serpin-type SPIs from crab, shrimp or scallopare significantly up-regulated by infection of virus or bacteria[7,20e23], suggesting their crucial role in the immune defenseagainst invaders. However, the knowledge about the functionsmolluscan Kunitz-type SPIs performed in innate immunity is stilllimited. In present study, a full-length cDNA of Kunitz-type SPIs wasidentified frommollusk Solen grandis, and themain objectives were(1) to investigate its tissue distribution and the temporal expres-sion after stimulation of pathogen-associated molecular patterns(PAMPs), (2) to examine the inhibitory activity of the recombinantprotein toward trypsin and chymotrypsin, and (3) to betterunderstand the role Kunitz-type SPIs played in the immune systemof S. grandis.

2. Materials and methods

2.1. Razor clam

Natural captured healthy razor clams S. grandis with averageshell length of 85mm, fromYantai, Shandong Province, China, werecollected and maintained in the aerated seawater at 20e22 �C fora week before processing.

2.2. EST analysis and cloning of the full-length SgKunitz cDNA

A cDNA library was constructed with the whole body of a razorclam, using the SMART cDNA Library Construction Kit (Clontech).Random sequencing of the library using M13 primer yielded 2038successful sequencing reactions. BLAST analysis of the ESTsequences revealed that an EST of 788 bp was highly similar to theknown Kunitz-type proteinase inhibitor sequences. The corre-sponding clone was selected and completely sequenced usingprimer M13-F and M13-R (Table 1).

2.3. Sequence analysis

The cDNA sequence and deduced amino acid sequence ofSgKunitz was analyzed using the BLAST algorithm (http://www.ncbi.nlm.nih.gov/blast) and the Expert Protein Analysis System(http://www.expasy.org/). The protein domains were predictedwith the simple modular architecture research tool (SMART)version 4.0 (http://smart.embl-heidelberg.de/). Homology analysiswas conducted using the Ident and Sim Analysis provided at http://www.bioinformatics.org/sms/. Multiple sequence alignment ofSgKunitz was performed with the ClustalW Multiple Alignmentprogram (http://www.ebi.ac.uk/clustalw/). The presumed tertiarystructures of SgKunitz were established using the SWISS-MODELprediction algorithm (http://swissmodel.expasy.org/) [24,25] anddisplayed by PyMOL version 0.97. A phylogenetic tree was

constructed based on the deduced amino acid sequences of SgKu-nitz and other Kunitz-type SPIs from invertebrates and vertebratesby the neighbor-joining (NJ) algorithm using the MEGA4.1 soft-ware. To derive confidence value for the phylogeny analysis, boot-strap trials were replicated 1000 times.

2.4. Real-time PCR analysis of SgKunitz mRNA expression indifferent tissues

The total RNA from hemocytes, gill, gonad, muscle, hepatopan-creas and mantle were extracted from five adult razor clams asparallel samples using TRIzol reagent (Invitrogen). The first-strandsynthesis was carried out based on Promega M-MLV RT Usageinformation using the DNase I (Promega)-treated total RNA astemplate and oligo(dT)-adapter as primer (Table 1). The reactionwere performed at 42 �C for 1 h, and terminated by heating at 95 �Cfor 5 min. The cDNA mix was diluted to 1:100 and stored at �80 �Cfor subsequent SYBR Green fluorescent quantitative real-time PCR(RT-PCR).

Two gene-specific primers for SgKunitz, RTF and RTR (Table 1),were used to amplify a product of 106 bp from cDNA, and the PCRproduct was sequenced to verify the specificity of RT-PCR. Andb-actin gene was selected to normalize relative gene expression,for its convincible and statistically repeatable results. Two b-actinprimers, AF and AR (Table 1), were used to amplify a 213 bpfragment as an internal control to verify the successful tran-scription and calibrate the cDNA template for correspondsamples.

Real-time PCR amplification was carried out in an ABI 7300Realtime Thermal Cycler according to the manual (Applied Bio-systems). Dissociation curve analysis of amplification products wasperformed at the end of each PCR to confirm that only one PCRproduct was amplified and detected. After the PCR program, datawere analyzed using ABI 7300 SDS software V2.0 (Applied Bio-systems). To maintain consistency, the baseline was set automati-cally by the software. The 2�

OOCT method was used to analyzethe expression level of SgKunitz [26]. All datawere given in terms ofrelative mRNA expressed as mean � SE (N ¼ 4).

2.5. Expression analysis of SgKunitz after PAMPs stimulation

Two hundred razor clams were employed for the PAMPs stim-ulation experiment. The razor clams were randomly divided into 5groups and each group contained 40 individuals. Four groupsreceived an injection of 50 mL phosphate buffered saline (PBS,0.14 M sodium chloride, 3 mM potassium chloride, 8 mM disodiumhydrogenphosphate dodecahydrate, 1.5 mM potassium phosphatemonobasic, pH 7.4), LPS from Escherichia coli 0111:B4 (SigmaeAldrich, 0.5 mg ml�1 in PBS), PGN from Staphylococcus aureus(SigmaeAldrich, 0.8 mg ml�1 in PBS), glucan from Saccharomyces

X. Wei et al. / Fish & Shellfish Immunology 33 (2012) 1276e12841278

cerevisiae (SigmaeAldrich, 1.0 mg ml�1 in PBS) via the adductormuscle, respectively [27]. The group of untreated razor clams wasemployed as the blank. After the treatments, the razor clams werereturned to water tanks and 5 individuals were randomly sampledat 3, 6, 12, 24 and 48 h post-injection.

The hemolymphs were collected, and centrifuged at 600� g,4 �C for 10 min to harvest the hemocytes. RNA isolation, cDNAsynthesis and real-time PCR analysis were carried out as describedabove. All data were given in terms of relative mRNA expressed asmean � SE (N ¼ 4). The data were subjected to one-way analysis ofvariance (one-way ANOVA) followed by an unpaired, two-tailed t-test. Differences were considered significant at P < 0.05 andextremely significant at P < 0.01.

2.6. Recombinant expression of SgKunitz

The cDNA fragment encoding the mature peptide of SgKunitzwas amplified with specific primers REF and RER (Table 1). The PCRproducts were gel-purified and cloned into pEASY-E1 expressionvector with a His tag (Transgen, China). The recombinant plasmid(pEASY-E1-SgKunitz) was transformed into Trans1-T1 phage resis-tant chemically competent cell (Transgen). The forward positiveclones were screened by PCR using vector primer T7 and primerRER (Table 1), and further confirmed by nucleotide sequencing. Thevalid recombinant plasmid (pEASY-E1-SgKunitz) was extracted andtransformed into E. coli BL21 (DE3)-Transetta (Transgen). Positivetransformants were incubated in LB medium (containing50 mgml�1 ampicillin) at 37 �C with shaking at 220 rpm. When theculture mediums reached OD600 of 0.5e0.7, the cells were incu-bated for 4 additional hours with the induction of IPTG at the finalconcentration of 1 mmol L�1. The recombinant protein SgKunitz(designated rSgKunitz) was purified by a Ni2þ chelating Sepharosecolumn, pooled by elution with 400 mmol L�1 imidazole underdenatured condition (8 mol L�1 urea). The purified protein wasrefolded in gradient urea-TBS glycerol buffer (50 mmol L�1 TriseHCl, 50 mmol L�1 NaCl, 10% glycerol, 2 mmol L�1 reduced gluta-thione, 0.2 mmol L�1 oxide glutathione, a gradient urea concen-tration of 6, 5, 4, 3, 2, 1, 0 mol L�1 urea in each gradient, pH 8.0; eachgradient at 4 �C for 12 h). Then the resultant proteins were sepa-rated by reducing 15% SDS-polyacrylamide gel electrophoresis(SDS-PAGE), and visualized with Coomassie bright blue R250.The concentration of purified rSgKunitz was quantified by BCAmethod [28].

2.7. Protease inhibition assay

The protease inhibition assay was performed according to theprevious method [29]. In brief, 10 mL (50 mM Tris buffer, pH 8.0)containing 1 mg of rSgKunitz was incubated with 10 mL containing1 mg of trypsin (Sigma) or chymotrypsin (Sigma) at roomtemperature for 10 min in a flat bottomed microtiter plate. Then180 mL of an appropriate chromogenic substrate solution in 50 mMTris buffer (pH 8.0) was added, and the resultant proteolyticactivity was measured at A405 after 1 min of incubation. Forpositive controls, rSgKunitz was substituted with the samequantity of aprotinin (Sigma) from the bovine lung, which wasa well-known inhibitor against trypsin and chymotrypsin. Nega-tive control data were obtained without inhibitor. 100 mM ofN-Benzoyl-Phe-Val-Arg-p-nitroanilide hydrochloride (Sigma) andN-Succinyl-Ala-Ala-Pro-Phe-pnitroanilide (Sigma) were used forthe substrates of trypsin and chymotrypsin, respectively. Each datafrom a single experiment was performed in triplicate for statisticanalysis.

To determine the dose-dependent inhibition activity of SgKu-nitz, 10 mL buffer containing 0.25, 0.5, 1.0, 2.0 or 4.0 mg of rSgKunitz

was incubated with protease, respectively. And protease inhibitionassay was performed as described above.

3. Results

3.1. Characteristic of the SgKunitz full-length cDNA

One EST from S. grandis cDNA library was found homologous tothe previously identified Kunitz genes. The clone corresponding tothe EST was re-sequenced, and yielded a cDNA sequence of1284 bp. The cDNA sequence of SgKunitz was deposited in GenBankunder accession number JQ730814. The complete nucleotidesequence of SgKunitz cDNA consisted of a 50 terminal untranslatedregion (UTR) of 52 bp, a long 30 UTR of 464 bp with a poly (A) tail,and an open reading frame (ORF) of 768 bp (Fig. 1). The ORFencoded a polypeptide of 255 amino acids with a theoreticalisoelectric point of 8.13 and predictedmolecular weight of 28.6 kDa.SMART analysis showed that the amino acids 1 (M) w 17 (A)encoded a signal peptide, and amino acids 26 (R) w 79 (G), 84(P) w 137 (G), 142 (P) w 195 (K) and 200 (P) w 253 (K) encodeda Kunitz domain, respectively (Fig. 1).

3.2. Homologous, structure and phylogenetic character of SgKunitz

BLAST analysis revealed that SgKunitz shared significantsequence similarities with other Kunitz-type proteinase inhibitors,such as 46% identity with Sciaenops ocellatus tissue factor pathwayinhibitor 2 (ADM64310) and 46% with Sabellastarte magnificacarboxypeptidase inhibitor (CAK55547). Based on the multiplesequences alignment between SgKunitz and other five Kunitz-typeproteinase inhibitors with three or four Kunitz domains, it wasfound that the six cysteine residues involved in the formation of thethree internal disulfide bridges were well conserved in eachdomain (Fig. 2), while the putative reactive site P1 were diversity ineach Kunitz-type SPIs except for P1 residues in the second Kunitzdomain (Fig. 2).

The potential tertiary structures of four Kunitz domains inSgKunitz were established using the SWISS-MODEL predictionalgorithm, respectively (Fig. 3). The tertiary structures were quitesimilar to each other: all of them were consisted of two a-helices(a1ea2) and two b-strands (b1eb2), and the six cysteines whichformed three disulfide bridges werewell conserved among the fourdomains.

A phylogenetic tree was constructed using neighbor joiningmethod with 1000 bootstrap test based on the multiple alignmentsof SgKunitz and Kunitz-type SPIs from other animal, includinginvertebrate Kunitz-type SPIs and vertebrate tissue factor pathwayinhibitor 2. In the phylogenetic tree, SgKunitz clustered withKunitz-type SPIs from Ancylostoma caninum firstly, and thenformed a sister group with the same molecules from Latrodectushesperus, suggesting there was a close relationship betweenSgKunitz and other invertebrate Kunitz-type SPIs (Fig. 4).

3.3. The mRNA expression profile of SgKunitz in different tissues

The SYBR Green real-time PCR analysis was employed to studythe mRNA expression of SgKunitz with b-actin as internal control.In the healthy razor clams, the mRNA transcript of SgKunitz wasfound to be constitutively expressed in a wide range of tissues withdifferent levels, including mantle, gill, gonad, hemocyte, muscle,and hepatopancreas (Fig. 5). The SgKunitz mRNA was expressedwith higher level in gill and hepatopancreas, which was 8303.9 and3916.5-fold compared with that in muscle, while the expressionlevel in hemocyte, mantle and gonad were relative lower, and waslowest expressed in muscle (Fig. 5).

Fig. 1. Nucleotide and deduced amino acid sequences of SgKunitz. The nucleotides and amino acids are numbered along the left margin, and the putative signal peptide isunderlined. The start and stop codons, and the classical polyadenylation signal in the 30 UTR, are marked in bold. The conserved Kunitz domains are shadowed.

X. Wei et al. / Fish & Shellfish Immunology 33 (2012) 1276e1284 1279

3.4. Temporal expression of SgKunitz mRNA post PAMPs stimulation

The temporal expression of SgKunitz mRNAwasmonitored afterthe razor clams were stimulated by three PAMPs. In the LPS stim-ulation group, expression level of SgKunitz mRNAwas significantlyup-regulated (P < 0.01), and it reached the peak at 6 h post-stimulation, which was 3.1-fold compared with the blank group,and as time progressed, the expression level decreased gradually to0.4-fold of the origin level at 48 h post-injection (Fig. 6, A).

Compared with LPS, PGN stimulation could induce much higherexpression level of SgKunitz, which were 7.3 and 127.6-fold(P < 0.01) compared with that in blank group at 3 h and 6 h post-injection, respectively (Fig. 6, B). After razor clams were stimu-lated by b-1, 3-glucan, transcript level of SgKunitz was kept ona consecutive high level from6h to 48 h, and it reached themaximallevel at 48 h after stimulation (6.5-fold compared with blank group,P < 0.05) (Fig. 6, C). There was no significant change of SgKunitzexpression in the control group treated with PBS (Fig. 6, AeC).

Fig. 2. Multiple sequence alignment by ClustalW of SgKunitz with other Kunitz-type SPIs from Amblyomma variegatum (DAA34268), Latrodectus hesperus (ADV40281), Danio rerio(XP_688834), Homo sapiens (AAK13254) and Gallus gallus (XP_418662). Amino acid residues that are conserved in at least 70% sequences are shaded in dark, and similar amino acidsare shaded in gray. The conserved cystines are labeled with triangle, but in the forth domain are boxed. The P1 residues are labeled with circle.

X. Wei et al. / Fish & Shellfish Immunology 33 (2012) 1276e12841280

3.5. Recombination and purification of SgKunitz protein

After IPTG induction, the whole cell lysate of E. coli BL21 (DE3)-Transetta with pEASY-E1-SgKunitz was analyzed by SDS-PAGE, anda distinct band with molecular weight of w28 kDa was revealed,which was consistent with the predicted molecular mass (Fig. 7,line 3). After purification by a Ni2þ chelating Sepharose column,a unique band of w28 kDa representing the protein of rSgKunitzwas revealed (Fig. 7, line 4). The concentration of the purifiedrSgKunitz was 395 mg ml�1 according to the BCA assay.

3.6. Inhibitory effect of SgKunitz on proteolytic activity of protease

The inhibitory activities of rSgKunitz against trypsin andchymotrypsinwere tested. As a consequence, rSgKunitzwas capableof inhibiting the proteolytic activity of both proteases. When 1 mg ofrSgKunitzwas incubatedwith 1 mg of protease, it could inhibit about

Fig. 3. Predicted tertiary structures of the four Kunitz domains in SgKunitz. Red: a-helicesD: domain 4. (For interpretation of the references to color in this figure legend, the reader

57.2% and 43.1% of the activity of trypsin and chymotrypsin,respectively (Fig. 8, A). As a positive control, bovine aprotinin couldinhibit 88.1% and 86.3% of the activities, accordingly (Fig. 8, A).

The dose-dependent inhibition activity of SgKunitz againsttrypsin and chymotrypsin was also determined. The proteaseactivities were decreased corresponding with the increasing ofrSgKunitz concentration (Fig. 8, B), suggesting the inhibitionactivity of SgKunitz exhibited dose-dependent effect. When 4 mg ofrSgKunitz was incubated with 1 mg of protease, the proteolyticactivities of trypsin and chymotrypsin were inhibited 76.0% and56.9%, respectively (Fig. 8, B).

4. Discussion

Proteinase inhibitors play important roles in the innate immu-nity, either by regulating the host defense reaction, or aiming againstthe proteinases from the invadingmicroorganisms [8]. Four families

; blue: b-strands; yellow: disulfide bridges. A: domain 1; B: domain 2; C: domain 3;is referred to the web version of this article.)

blank 3 6 12 24 48

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

**

**

Rela

tive m

RN

A ex

pressio

n o

f S

gK

un

itz

Time post stimulation (hrs)

LPS PBS

**

blank 3 6 12 24 48

0

2

4

6

8

10

120

140

160

180

**

**

Re

la

tiv

e m

RN

A e

xp

re

ss

io

n o

f S

gK

un

itz

Time post stimulation (hrs)

PGN PBS

A

B

C

Fig. 4. Phylogenetic tree of SgKunitz with other Kunitz-type SPIs. The tree is con-structed by the neighbor-joining (NJ) algorithm using the Mega 4.1 program based onthe multiple sequence alignment by ClustalW. Bootstrap values of 1000 replicates (%)are indicated for the branches.

X. Wei et al. / Fish & Shellfish Immunology 33 (2012) 1276e1284 1281

of serine proteinase inhibitors including kazal, kunitz, serpin anda-macroglobulin are proved to involve in the immune response [9],and have been well characterized in mammals. However, little isknown about the Kunitz-type SPIs in mollusk, especially theirfunctions and regulative mechanisms in innate immunity.

In present study, a Kunitz-type SPI with four tandem Kunitzdomains, which were frequently observed in the Kunitz-type SPIfamily, was identified from marine mollusk S. grandis. There wasa canonical signal peptide at the N-terminus of deduced amino acidsequence, revealing that SgKunitz was a secreted protein, whichwas consistent with other Kunitz-type SPIs [30]. The amino acidsequence of SgKunitz shared highly similarities with Kunitz-typeSPIs from other animals. In the phylogenetic tree, SgKunitz kepta close evolutionary relationship with Kunitz-type SPIs from otherinvertebrate. In addition, the tertiary structure of SgKunitz adoptedthe typical model of this family, and all the three disulfide bridgeswere well conserved in the four Kunitz domains. The existence ofthe Kunitz domain and the conserved sequence and structuremotifs, together with the high sequence identity, strongly sug-gested that SgKunitz should be a member of Kunitz-type SPI family.Meanwhile, the similar hallmarks both in sequence and structurealso suggesting SgKunitz might perform the same function as otherKunitz-type SPIs [29,31].

musclegonad

mantle

hemocytes

hepatopancreas gill0

102030405060

2000

4000

6000

8000

10000

12000

Rel

ativ

e m

RN

A ex

pres

sion

of S

gKun

itz

Fig. 5. SgKunitz mRNA expression level in different tissues of S. grandis detected byreal-time PCR. SgKunitz transcript level in mantle, gill, gonad, hemocyte, muscle, andhepatopancreas were normalized to that of muscle. Vertical bars represent themean � SE (N ¼ 4).

blank 3 6 12 24 48

0

1

2

3

4

5

6

7

8

9

10

*

****

**

Relative m

RN

A e

xp

ressio

n o

f S

gK

un

itz

Time post stimulation (hrs)

glucan PBS

Fig. 6. Temporal expression of SgKunitz mRNA relative to b-actin was analyzed by real-time PCR in hemocytes after LPS, PGN, glucan and PBS challenge for 3, 6, 12, 24 and48 h. The values are shown as mean � SE (N ¼ 4). (*: P < 0.05, **: P < 0.01). A: LPSgroup; B: PGN group; C: glucan group.

Fig. 7. SDS-PAGE analysis of recombinant SgKunitz. Lane 1: protein molecular stan-dard; lane 2: negative control for recombinant SgKunitz (without induction); lane 3:induced recombinant SgKunitz; lane 4: purified recombinant SgKunitz.

Trypsin Chymotrypsin0.0

0.1

0.2

0.3

0.4

0.5

0.6

*

**

Pro

teol

ytic

act

ivit

y (4

05 n

m)

No inhibitor rSgKunitz Aprotinin

*

0.25 0.5 1 2 4

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

0.55

Pro

teol

ytic

act

ivit

y (4

05 n

m)

Concentration of rSgKunitz (µg)

Trypsin Chymotrypsin

A

B

Fig. 8. Inhibitory activity and its dose effect of rSgKunitz toward trypsin and chymo-trypsin. A: Inhibitory activity of rSgKunitz toward trypsin and chymotrypsin. 1 mg ofrSgKunitz was incubated with 1 mg of trypsin or chymotrypsin, then an appropriatechromogenic substrate solution was added, and the proteolytic activity was measuredat A405 after 1 min of incubation. Bovine aprotinin was used as positive control. Thevalues are shown as mean � SE (N ¼ 3). (*: P < 0.05); B: dose-dependent inhibitionactivity of SgKunitz. 0.25, 0.5, 1.0, 2.0 or 4.0 mg of rSgKunitz was incubated withprotease, and the inhibitory behavior was examined as described above.

X. Wei et al. / Fish & Shellfish Immunology 33 (2012) 1276e12841282

Detecting the tissue specific expression pattern of SgKunitz genewould benefit understanding its potential role in inhibitorybehavior and the functions in immune response. The Kunitz-typeSPI HlMKI from Haemaphysalis longicornis was expresseduniquely in midgut of the adult tick [32], while Rhipicephalus(Boophilus) microplus Kunitz-type SPI BmCI was expressed inhemocytes, ovary and salivary glands, but not fat body [30]. Unlikethe Kunitz-type SPIs in other invertebrates, SgKunitz exhibiteda wide range of tissues distribution with different levels. Consid-ering razor clams completely expose themselves to the aquaticenvironment with large number of pathogenic microorganism, thewide distribution of SgKunitz also suggested their important rolesin limiting the excessive proteolytic activity of serine proteasescausing by the continuing microbial invaders. For mollusk, hemo-cyte and hepatopancreas were two important immune-relatedtissues [33e35], while gill was the tissue that took the responsi-bility to seize oxygen and alga from the seawater, which frequentlycontact with pathogens [33]. Therefore, the fact that SgKunitz wasspecifically expressed in these tissues indicated SgKunitz mightcontribute to the immune system as a crucial molecule against theinfection by microbes.

The physiological function of Kunitz-type SPIs was setting toinhibit the SPs, and protecting tissues from the proteolysis byexcessive activities of SPs. Several isolated or recombinant Kunitz-type SPIs from invertebrates could effectively inhibit the activitiesof SPs [29,31,36]. For instance, Kunitz-type SPI purified fromhemolymph of Galleria mellonella larvae could inhibit the activitiesof trypsin, plasmin and kallikrein [29], while recombinant Kunitz-type SPIs of silkworm exhibit effective inhibitory activity towardchymotrypsin [36]. In present study, the recombinant SgKunitz wasproved to inhibit 57.2% and 43.1% of the activity of trypsin andchymotrypsin, respectively. Since the three disulfide bridgesforming by six conserved cysteines in each Kunitz domain werecrucial for the stability and activity of the molecule, and not all thecysteines could form disulfide bridges correctly on the processes ofrefolding, the fact that inhibitory activity of SgKunitz was relative

lower when compared with commercial inhibitor aprotinin wouldbe reasonable. Conventionally, the inhibitory specificity of Kunitzdomain could be speculated from the P1 residue in the reactive site,which was the second residue after the second conserved cysteineresidue. If P1 residue was lysine or arginine, it could inhibit trypsin,while if with tyrosine, phenylalanine, leucine or methionine at therelevant position, it tend to inhibit chymotrypsin [8]. In presentstudy, there were four Kunitz domains in SgKunitz, among of themthe P1 residue in Kunitz domain 1 was leucine, while the relevantresidues in Kunitz domain 2 and 3 were arginine and lysine,respectively, which indicated SgKunitz might inhibit the activity oftrypsin and chymotrypsin synchronously. Such deduction wasconfirmed by our subsequent result, which manifested therecombinant SgKunitz could inhibit the activities of both afore-mentioned enzymes, and this result was consisted with the Kunitz-type SPI Pr-mulgins-2 in Pseudechis australis [31]. However, exceptfor the aforementioned residues, the P1 position in Kunitz domain4 was glutamic acid that had not been reported before, whether itcould increase the inhibitory spectrum or not was still an unknownquestion that needed further illustration.

In arthropod, SPIs served as a negative regulator in the innateimmune signal pathway and eliminated the excessive activity of

X. Wei et al. / Fish & Shellfish Immunology 33 (2012) 1276e1284 1283

SPs at the appropriate time. The invading pathogens were detectedby the immune receptors targeting on the PAMPs, as a result severalkinds of SPs were activated, which would subsequently triggeredthe activation of prophenoloxidase cascade and Toll signal pathway[6,18,37]. Both aforementioned signal pathways were preciselyregulated by the SPIs, especially the serpin-type SPIs, and once thepathogens were finally eliminated, SPIs would deprive the exces-sive activity of SPs in the first time. To preliminarily unravel theregulation mechanism of Kunitz-type SPIs in the innate immunity,the temporal expression of SgKunitz mRNA in hemocytes, wasexamined by realtime RT-PCR after razor clams were stimulated bythree typical PAMPs. As observed in our study, all the three PAMPsincluding LPS from Gram-negative bacteria, PGN from Gram-positive bacteria and b-1, 3-glucan from fungi could significantlyup-regulated the expression of SgKunitz, suggesting SgKunitzmight involved in the processes of inhibiting the activity of SPsduring the immune responses triggered by various pathogens. Theexpression of SgKunitz exhibited a higher level after stimulation ofPGN or b-1, 3-glucan than that in LPS group, while the up-regulativerespondence toward b-1, 3-glucanwas later than other two PAMPs.This obvious difference might be a consequence of the activation ofdifferent immune signaling pathway responding to diverse PAMPsstimulation. In Drosophila, two kinds of antimicrobial peptideswere regulated by SPIs via different immune signaling pathways.Both Gram-positive bacteria and fungi were recognized by pepti-doglycan recognition protein (PGRP-L) or Gram-negative bacteriabinding protein (GNBP), and anti-fungi antimicrobial peptides“Drosomycin” was produced via the activation of Toll immunesignaling pathway, while Gram-negative bacteria was sensed byPGRP-S and subsequently triggered IMD immune signalingpathway to produce Gram-negative bacteria specific antimicrobialpeptides “Diptericin” [9,18,37]. Therefore, the diverse signalingpathways SPs and SPIs involved in might be the main reasonresponsible for the different expression pattern of SgKunitz towardthree PAMPs. In mollusk, though it was far fromwell understandingthe regulation of SPs- and SPIs-involved immune signalingpathway, a primitive Toll pathway had been proved to be existencein scallop and oyster [38,39]. In Zhikong scallop, all the facts thatexpression of pattern recognition receptors (CfPGRP-S1 andCfLGBP), SP (CfSP), SPI (CfKZSPI) and TLR (CfTLR) were significantlyup-regulated after the stimulation of PAMPs or pathogens[7,33,39e41], indicated that molluscan SPs and SPIs might alsoserve as modulators involving in the immune signaling pathwaylike that in Drosophila. Considering there was still no regulationmechanism illumining this pathway in mollusk, our results whichrevealed the expression patterns of Kunitz-type SPI toward PAMPsstimulation in S. grandis would provide helpful evidence tounderstanding the regulation mechanism of the immune signalpathway in marine mollusk.

Acknowledgments

We are grateful to all the laboratory members for their technicaladvice and helpful discussions. This research was supported byProgram of Agriculture Thoroughbred Project, Shandong province,China to Dr. Xiangquan Liu, NSFC funding (No. 31202025) andResearch Award Fund for Outstanding Young Scientists of Shan-dong Province (No. Y231041011) to Dr. Jialong Yang, and TaishanScholar position funding of Aquatic animal nutrition and feed toLimin Zhang.

References

[1] Davie EW, Fujikawa K, Kisiel W. The coagulation cascade: initiation, mainte-nance, and regulation. Biochemistry 1991;30:10363e70.

[2] Fuentes-Prior P, Noeske-Jungblut C, Donner P, Schleuning WD, Huber R,Bode W. Structure of the thrombin complex with triabin, a lipocalin-likeexosite-binding inhibitor derived from a triatomine bug. Proceedings of theNational Academy of Sciences of the United States of America 1997;94:11845e50.

[3] Hiemstra PS. Novel roles of protease inhibitors in infection and inflammation.Biochemical Society Transactions 2002;30:116e20.

[4] Konrad KD, Goralski TJ, Mahowald AP, Marsh JL. The gastrulation defectivegene of Drosophila melanogaster is a member of the serine protease super-family. Proceedings of the National Academy of Sciences of the United Statesof America 1998;95:6819e24.

[5] Reid KB, Porter RR. The proteolytic activation systems of complement. AnnualReview of Biochemistry 1981;50:433e64.

[6] Tong Y, Kanost MR. Manduca sexta serpin-4 and serpin-5 inhibit the proph-enol oxidase activation pathway: cDNA cloning, protein expression, andcharacterization. The Journal of Biological Chemistry 2005;280:14923e31.

[7] Wang B, Zhao J, Song L, Zhang H, Wang L, Li C, et al. Molecular cloning andexpression of a novel Kazal-type serine proteinase inhibitor gene from Zhi-kong scallop Chlamys farreri, and the inhibitory activity of its recombinantdomain. Fish & Shellfish Immunology 2008;24:629e37.

[8] Laskowski Jr M, Kato I. Protein inhibitors of proteinases. Annual Review ofBiochemistry 1980;49:593e626.

[9] Kanost MR. Serine proteinase inhibitors in arthropod immunity. Develop-mental and Comparative Immunology 1999;23:291e301.

[10] Polanowski A, Wilusz T. Serine proteinase inhibitors from insect hemolymph.Acta Biochimica Polonica 1996;43:445e53.

[11] Azzolini SS, Santos JM, Souza AF, Torquato RJ, Hirata IY, Andreotti R, et al.Purification, characterization, and cloning of a serine proteinase inhibitorfrom the ectoparasite Haematobia irritans irritans (Diptera: Muscidae).Experimental Parasitology 2004;106:103e9.

[12] Sasaki SD, Azzolini SS, Hirata IY, Andreotti R, Tanaka AS. Boophilus microplustick larvae, a rich source of Kunitz type serine proteinase inhibitors. Biochimie2004;86:643e9.

[13] Siddiqi AR, Zaidi ZH, Jornvall H. Purification and characterization of a Kunitz-type trypsin inhibitor from Leaf-nosed viper venom. FEBS Letters 1991;294:141e3.

[14] Tschesche H, Dietl T. The amino-acid sequence of isoinhibitor K form snails(Helix pomatia). A sequence determination by automated Edman degradationand mass-spectral identification of the phenylthiohydantoins. EuropeanJournal of Biochemistry/FEBS 1975;58:439e51.

[15] Wunderer G, Fritz H, Wachter E, Machleidt W. Amino-acid sequence ofa coelenterate toxin: toxin II from Anemonia sulcata. European Journal ofBiochemistry/FEBS 1976;68:193e8.

[16] Berndt KD, Guntert P, Orbons LP, Wuthrich K. Determination of a high-qualitynuclear magnetic resonance solution structure of the bovine pancreatictrypsin inhibitor and comparison with three crystal structures. Journal ofMolecular Biology 1992;227:757e75.

[17] Laskowski M, Qasim MA. What can the structures of enzyme-inhibitorcomplexes tell us about the structures of enzyme substrate complexes? Bio-chimica et biophysica acta 2000;1477:324e37.

[18] Ligoxygakis P, Pelte N, Hoffmann JA, Reichhart JM. Activation of DrosophilaToll during fungal infection by a blood serine protease. Science (New York,NY) 2002;297:114e6.

[19] Zheng QL, Chen J, Nie ZM, Lv ZB, Wang D, Zhang YZ. Expression, purificationand characterization of a three-domain Kazal-type inhibitor from silkwormpupae (Bombyx mori). Comparative Biochemistry and Physiology 2007;146:234e40.

[20] Li XC, Zhang RR, Sun RR, Lan JF, Zhao XF, Wang JX. Three Kazal-type serineproteinase inhibitors from the red swamp crayfish Procambarus clarkii and thecharacterization, function analysis of hcPcSPI2. Fish & Shellfish Immunology2010;28:942e51.

[21] Qian YQ, Li Y, Yang F, Yu YQ, Yang JS, Yang WJ. Two Kazal-type proteaseinhibitors fromMacrobrachium nipponense and Eriocheir sinensis: comparativeanalysis of structure and activities. Fish & Shellfish Immunology 2012;32:446e58.

[22] Wang S, Cui Z, Liu Y, Li Q, Song C. Identification and characterization ofa serine protease inhibitor (PtSerpin) in the swimming crab Portunus tritu-berculatus. Fish & Shellfish Immunology 2012;32:544e50.

[23] Wang ZH, Zhao XF, Wang JX. Characterization, kinetics, and possible functionof Kazal-type proteinase inhibitors of Chinese white shrimp, Fenneropenaeuschinensis. Fish & Shellfish Immunology 2009;26:885e97.

[24] Arnold K, Bordoli L, Kopp J, Schwede T. The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics2006;22:195e201.

[25] Kiefer F, Arnold K, Kunzli M, Bordoli L, Schwede T. The SWISS-MODEL Repos-itory and associated resources. Nucleic Acids Research 2009;37:D387e92.

[26] Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(T)(-Delta Delta C) method. Methods 2001;25:402e8.

[27] Wei X, Yang J, Yang D, Xu J, Liu X, Yang J, et al. Molecular cloning and mRNAexpression of two peptidoglycan recognition protein (PGRP) genes frommollusk Solen grandis. Fish & Shellfish Immunology 2012;32:178e85.

[28] Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD,et al. Measurement of protein using bicinchoninic acid. Analytical Biochem-istry 1985;150:76e85.

X. Wei et al. / Fish & Shellfish Immunology 33 (2012) 1276e12841284

[29] Lee JH, Kim CH, Shin YP, Park HJ, Park S, Lee HM, et al. Characterizationof Kunitz-type protease inhibitor purified from hemolymph of Galleriamellonella larvae. Insect Biochemistry and Molecular Biology 2010;40:873e82.

[30] Lima CA, Torquato RJ, Sasaki SD, Justo GZ, Tanaka AS. Biochemical charac-terization of a Kunitz type inhibitor similar to dendrotoxins produced byRhipicephalus (Boophilus) microplus (Acari: Ixodidae) hemocytes. VeterinaryParasitology 2010;167:279e87.

[31] Inagaki H, Kimoto H, Yamauchi Y, Toriba M, Kubo T. Functional character-ization of Kunitz-type protease inhibitor Pr-mulgins identified from NewGuinean Pseudechis australis. Toxicon 2012;59:74e80.

[32] Miyoshi T, Tsuji N, Islam MK, Alim MA, Hatta T, Yamaji K, et al. A Kunitz-typeproteinase inhibitor from the midgut of the ixodid tick, Haemaphysalis long-icornis, and its endogenous target serine proteinase. Molecular andBiochemical Parasitology 2010;170:112e5.

[33] Yang J, Qiu L, Wang L, Wei X, Zhang H, Zhang Y, et al. CfLGBP, a patternrecognition receptor in Chlamys farreri involved in the immune responseagainst various bacteria. Fish & Shellfish Immunology 2010;29:825e31.

[34] Yang J, Wang L, Zhang H, Qiu L, Wang H, Song L. C-type lectin in Chlamysfarreri (CfLec-1) mediating immune recognition and opsonization. PloS One2011;6:e17089.

[35] Zhang H, Wang H, Wang L, Song X, Zhao J, Qiu L, et al. A novel C-type lectin(Cflec-3) from Chlamys farreri with three carbohydrate-recognition domains.Fish & Shellfish Immunology 2009;26:707e15.

[36] He N, Fujii H, Kusakabe T, Aso Y, Banno Y, Yamamoto K. Overexpression inEscherichia coli and purification of recombinant CI-b1, a Kunitz-type chymo-trypsin inhibitor of silkworm. Protein Expression andPurification2004;38:9e16.

[37] Gubb D, Sanz-Parra A, Barcena L, Troxler L, Fullaondo A. Protease inhibitorsand proteolytic signalling cascades in insects. Biochimie 2010;92:1749e59.

[38] Zhang L, Li L, Zhang G. A Crassostrea gigas Toll-like receptor and comparativeanalysis of TLR pathway in invertebrates. Fish & Shellfish Immunology 2011;30:653e60.

[39] Wang M, Yang J, Zhou Z, Qiu L, Wang L, Zhang H, et al. A primitive Toll-likereceptor signaling pathway in mollusk Zhikong scallop Chlamys farreri.Developmental and Comparative Immunology 2011;35:511e20.

[40] Yang J, Wang W, Wei X, Qiu L, Wang L, Zhang H, et al. Peptidoglycan recog-nition protein of Chlamys farreri (CfPGRP-S1) mediates immune defensesagainst bacterial infection. Developmental and Comparative Immunology2010;34:1300e7.

[41] Zhu L, Song L, Mao Y, Zhao J, Li C, Xu W. A novel serine protease with clipdomain from scallop Chlamys farreri. Molecular Biology Reports 2008;35:257e64.