Developmental and Comparative Immunology 44 (2014) 163–172

Contents lists available at ScienceDirect

Developmental and Comparative Immunology

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

PmPPAF is a pro-phenoloxidase activating factor involved in innateimmunity response of the shrimp Penaeus monodon

0145-305X/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.dci.2013.12.007

⇑ Corresponding authors. Address: Fisheries College, Guangdong Ocean Univer-sity, Zhanjiang, Guangdong, PR China. Tel.: +81 18319121352.

E-mail addresses: scb248@163.com (C.-B. Sun), siuming573@sina.com (S.F.Chan).

Tracy H.T. Ma a,b, John A.H. Benzie c, Jian-Guo He d, Cheng-Bo Sun a,⇑, Siuming F. Chan a,⇑a Fisheries College, Guangdong Ocen University, Zhanjiang, PR Chinab School of Biological Sciences, The University of Hong Kong, PR Chinac University College Cork, Cork, Ireland and WorldFish, Jalan Batu Maung, Batu Maung, 11960 Bayan Lepas, Penang, Malaysiad School of Life Sciences, Zhongshan University, PR China

a r t i c l e i n f o a b s t r a c t

Article history:Received 26 August 2013Revised 2 December 2013Accepted 3 December 2013Available online 15 December 2013

Keywords:Prophenoloxidase activating systemPhenoloxidase activityPenaeus monodon

One of the major steps in the innate immune response of shrimp includes the activation of serine protein-ases of the pro-phenoloxidase pathway by the prophenoloxidase activation enzyme (PPAF). In this study,the cDNA encoding a serine proteinase homologue (SPH) with prophenoloxidase activating activity ofPenaeus monodon (PmPPAF) was cloned and characterized. PmPPAF cDNA consists of 1444 nucleotidesencoding a protein with 394 amino acid residues. The estimated molecular weight of PmPPAF is43.5 kDa with an isoelectric point of 5.19. PmPPAF consists of a signal peptide, a CLIP domain and a car-boxyl-terminal trypsin-like serine protease domain. It is highly similar to the masquerade-like protein 2A(61% similarity) of the crayfish Pacifastacus leniusculus, other serine proteases (42.9–67% identity) of P.monodon, and the PPAF of the crab (61% similarity). Unlike other SPH of P. monodon, which expressmainly in the hemocytes, PmPPAF transcripts were detected in the hemocytes, eyestalk, hypodermis, gill,swimming leg and brain. Similar to the crab PPAF, PmPPAF transcript level is high in shrimp at the premoltstages and PmPPAF expression is up-regulated in shrimp infected with white spot syndrome virus(WSSV). Gene silencing of PmPPAF decreased expression of a prophenoloxidase-like gene and injectionof Anti-PmPPAF antibody causes a decrease in PO activity. Taken together, these results provided evidencethat PmPPAF is a serine proteinase homologue, and is involved in the pro-PO activation pathway of theshrimp innate immune system.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Invertebrates lack an adaptive immune system and have to relyon an innate immune system to combat invading pathogens. Themajor immune responses in invertebrates include (1) the genera-tion of antimicrobial peptides mediated by toll-like receptors; (2)coagulation of hemolymph; (3) formation of melanin and (4) acti-vation of a complementation-like system mediated by lectins.When pathogens such as bacteria or virus enter the host, the hostwill respond by a complex system of innate defense mechanisms todetoxify and/or remove these pathogens. Entrance of a pathogen tothe invertebrate host can trigger the pro-phenoloxidase activatingsystem that consists of endogenous trypsin-like serine protein-ase(s) (also known as prophenoloxidase-activating factor(s)(PPAF). These PPAF then convert prophenoloxidase (pro-PO) to

the active form of phenoloxidase (PO) and causes melanization ofpathogens (Cerenius and Söderhäll 2004; Söderhäll et al., 1994).

CLIP domain serine proteinase is one of the most studiedenzyme groups in animals (Jiang and Kanost 2000). They exhibitdiverse functions including digestion, complement activation,embryonic development, reproduction, zymogen activation, proen-zyme and prohormone regulation, and innate immunity Melaniza-tion is one of the innate defense mechanisms in crustaceans inwhich multiple steps occur leading to the sequestration and killingof invading microorganisms/pathogens, and it is regulated by acascade of serine proteinases that ultimately activates propheno-loxidase (PPO) (Cerenius and Söderhäll, 2004). As a group, prophe-noloxidase activating factors are proteins belonging to the serineproteinase (SP)/serine proteinase homologs (SPH) family.

For arthropod defense, various responses such as blood coagu-lation, melanocyte encapsulation, induction of anti-microbialpeptide synthesis, cytokine activation, embryo protection, repro-duction, digestion and fertilization (Kanost et al., 2001; Jiang andKanost, 2000; Gorman and Paskewitz, 2001; Kawabata et al.,1996) are carried out by serine proteinase and serine proteinasehomologs. In insects, several proPO-activating proteinases have

164 T.H.T. Ma et al. / Developmental and Comparative Immunology 44 (2014) 163–172

been purified and characterized (Jiang and Kanost, 2000; Lee et al.,2002; Satoh et al., 1999; Wang et al., 2001; Wang and Jiang,2004a). In Holotrichia diomphalia, the prophenoloxidase activatingfactor-II (PPAF II) serves as a cofactor of serine proteinase PPAF-Iand is involved in the activation of the ProPO system (Kwonet al., 2000). These studies suggest that the prophenoloxidase acti-vation system is regulated and controlled by a cascade of similarenzymes.

In crustaceans, many SP and SPHs have been characterized re-cently and they are also found to have different functions. Forexample, in the crayfish Pacifastacus lenisculusis they are involvedin pathogen adhesion (Huang et al., 2000); the factor D of horse-shoe crab Tachypleus tridentatus is reported to have anti-microbialactivity (Kawabata et al., 1996) and the PPAF of the crab Callinectessapidus can control and regulate molting (Buda and Shafer, 2005).Several recent studies reported the cloning and characterizationof serine protease homologues in P. monodon (i.e., PmMasSPH,PmSP, PmcSPH), most of which were proposed to have immune re-lated functions, although few studies demonstrated prophenoloxi-dase functions for these SP/SPH (Amparyup et al., 2007; Lin et al.,2006).

Given the importance of SP and SPH in the immune defensemechanism of cultivated shrimp, understanding their role inshrimp innate immune mechanisms may provide important in-sights for the control of shrimp disease. The objective of this studywas to clone and characterize a SPH potentially involved in the in-nate immunity of P. monodon. RNA interference was employed tostudy the function of the putative prophenoloxidase-activating en-zyme by direct injection of dsRNA in juvenile P. monodon and anti-body against PmPPAF was injected to antagonize the effect ofPmPPAF.

2. Materials and methods

2.1. Animals

Black tiger shrimp P. monodon (15–20 g) were purchased fromlocal seafood markets and acclimated in aerated artificial sea-water at 25 �C for 3 days. They were used for RNA preparation,virus, antibody and dsRNA injection experiments as describedbelow.

2.2. Cloning full length cDNA of PmPPAF

Hypodermal tissue was dissected from the carapace and totalRNA was extracted (Chomczynski and Sacchi, 1987). Total RNA(5 g) was used in the first strand cDNA synthesis by reverse tran-scription (RT) reaction. For RT, the reaction mix (50 ll) consistedof1 �MMLV buffer, 2.5 lg oligo(dT)12 primer, 2 mM dNTP, 20Uof RNA guard (Invitrogen, USA) and 50U MMLV-RT, incubated at37 �C for 3 h and terminated at 70 �C for 15 min. For RT-PCR,degenerate primers (Fig. 1) were designed based on the highly con-served nucleotide sequences of serine proteinase from P. monodon(i.e., PmMasSPH, PmSPH and PmSP, Table 1). The primers of PmPPAF-F1 and PmPPAF-R1 were used to amplify a partial cDNA of SP andcDNA from P. monodon hypodermis cDNA. An aliquot (1 ll) ofthe RT mix was used as templates for PCR (GeneAmp PCR system2700, Applied Biosystems, Forster, CA, USA). For PCR, the reactionmix (pH = 9) consisted of 50 mM KCl, 1% Triton X-100, 2.5 mMMgCl2, 2.5U Taq polymerase, 0.25 mM NTPs, and 0.5 M of each pri-mer. The PCR condition was 35 cycles of denaturation at 94 �C for1 min, annealing at 60 �C for 1 min, and extension at 72 �C for1 min, followed by a 7-min extension at 72 �C and cooling to4 �C. PCR product was analyzed on a 1.5% agarose gel and sub-cloned into the TA-cloning vector (Promega, USA). DNA sequencing

for the subclones was performed to obtain the partial PmPPAFcDNA.

For cloning the 50 and 30 cDNA of PmPPAF, a rapid amplificationof cDNA ends (RACE) kit (Roche, Germany) was used. Gene specificprimers PmPPAF-R2 and PmPPAF-R3 (Fig. 1; Table 1) were designedand used in 50 RACE whereas gene specific primers PmPPAF-F2 andPmPPAF-F3 were used for the 30 end of the cDNA. Both 50 RACE and30 RACE were performed according to the manufacturer’s instruc-tion manual. Similar to the RT-PCR cloning, RACE products weresubcloned into a TA vector (Promega, USA) and DNA sequencedetermination was performed using the dideoxynucleotide chaintermination method on a DNA sequencer (Model 373A, AppliedBiosystem). Full-length cDNA of PmPPAF was obtained from the se-quences of the overlapping clones. The reconstructed cDNA and se-quence was analyzed and compared using the BLASTX searchprogram (http://blast.ncbi.nlm.nih.gov/Blast.cgi) with GenBankdatabase search.

Amino acid sequence of PmPPAF was aligned and comparedwith other serine protease homologs from P. monodon, crayfish,crab and insects using Clustal W. The P.monodon serine proteinasehomologues used in the comparison were (PmPPAF, Genbank#EU082201: PmSP, Genbank# AY372186; Pmc-SPH, Genbank#AY600627, PmSPH516, Genbank# DQ403191; PmSPH509, Gen-bank# DQ916148; PmMasSPH, GenbanK# DQ455050) and othersequences included in this study are: Marsupenaeus japonicus ser-ine proteinase homologue (MsSP, GenBank# BAD34945); Litopena-eus vannamei serine proteinase (LvSP, GenBank# AAQ92356), thecrab C. sapidus phenoloxidase activating factor (CsPPAF, GenBank#AAS60227); the insect Manduca sexta serine proteinase-like protein1 (MsSP, GenBank# AAM69352); Anopheles gambiae serine prote-ase (AgSP, GenBank# CAB90818); Drosophila melanogaster Tryp-sin-like serine protease (DmSP, GenBank# AAF52904); thecrayfish Pacifastacus leniusculus masquerade-like protein (PlsMas-II, GenBank# AY861652); T. tridentatus factor D (TtppA, GenBank#BAA13312); Dermacentor variabilis factor D-like protein (DvSP,GenBank# AAN78224).

For phylogenetic analysis, tree construction used the neighbor-joining method, based on the amino acid alignment (Clustal W) ofpublished serine protease or serine protease homologue sequencesfrom the shrimp, crab, and insects, performed using program MEGAversion 4 (Tamura et al., 2007).

2.3. Expression study of PmPPAF

For Northern blot analysis of PmPPAF transcripts distribution,total RNAs from various tissues were prepared by the guanidinethiocyanate method (Chomczynski and Sacchi, 1987). RNA samples(20 lg) were analyzed on 1.2% formaldehyde agarose gels andtransferred to nylon N+ membrane (Amersham, USA). The blotwas hybridized in 50% formamide, 5� SSC, 0.5% SDS, 5� Denhardt’ssolution containing DIG-labeled non-radioactive probe of PmPPAFat 42 �C overnight. The DIG-labeled probe (Roche, Germany) wasderived from a partial cDNA fragment of PmPPAF (from AA37 toAA355). After hybridization, the membrane was washed twice in2� SSC, 0.1% SDS for 15 min, and then washed twice at 0.5� SSC,0.1% SDS at 58 �C for 15 min. After incubating the membrane inantibody solution containing anti-DIG-AP conjugated, chemilumi-nescent substrate CDP-star was added and the membrane was ex-posed to X-ray film. For RT-PCR, 1 lg of total RNAs was used in thereverse transcription (RT) reaction for synthesis of the first-strandcDNA as described above. RT reaction (50 ll) was performed in1 �M-MLV Reaction Buffer, 2.5 lg of oligo(dT)12 primer, 2 mMdNTP, 20U of RNA guard (Invitrogen, USA) and 50U of MMLV-RTat 37 �C for 3 h and terminated at 70 �C for 15 min. The conditionsfor PCR (20 ll) were the same as described above. PCR productswere analyzed on 1.5% agarose gels.

PmPPAF-F1

PmPPAF-F2

PmPPAF R2

PmPPAFR1

AA37 AA355

PmPPAF R3

PmPPAF F4 PmPPAF R4

RT-PCR

5’ RACE

3’ RACE

ORF

PmPPAF F3

5’ 3’Sp CLIP domain Trypsin-like serine proteinase domain1 16 30 99 128 379

PmPPAF

(a)

(b)

Fig. 1. Structure and schematic diagram showing the cloning strategies for the PmPPAF of P. monodon. (a) Structural organization of the shrimp PmPPAF. PmPPAF consists of asignal peptide sequence (Sp), a CLIP domain region and a trypsin-like serine protease domain. The numbers below the PmPPAF schematic structure show the amino acidresidue positions; (b) A cDNA fragment spanning the CLIP domain and the trypsin region was obtained by RT-PCR with degenerate primers, primer PmPPAF-F1 and primerPmPPAF-R1. Gene-specific primers were designed for 50 (PmPPAF-R2 and PmPPAF-R3 primer) and 30 RACE (PmPPAF-F2 and PmPPAF-F3 primer) and the full-length cDNA wasobtained from the overlapping RACE clones. Furthermore, primers F4 and R4 were used to amplify the cDNA corresponding to open reading frame (ORF) for finalized sequenceconfirmation. Supplement. Nucleotide and deduced amino acid sequence of PmPPAF. The CLIP-domain is shaded in light green color and the conserved cysteine residuesforming the clip domain are boxed. The sites of the catalytic triad of serine proteinase are circled. The signal peptide is underlined and bolded. The polyadenylation signal isshown in italics and underlined. The start and stop codons are in bold letters. The serine protease domain (AA residues 128-379) is shaded in dark grey color.

Table 1Serine proteases reported in P. monodon.

Name Expression sites Proposed functions Number of clip domains Triad Accession number

PmPPAF Hypodermis, gill, nerve tissues Molting and Immunity 1 H D G EU082201PmSP (2005) Only hemocytes – 1, pseudo-clip H D G AY372816Pm c-SPH (2006) Mainly hemocytes Cell adhension 1 H D G AY600627Pm SPH 516 (2007) Hemocytes and gills Binding protein of YHV 1 H D G DQ403191Pm SPH 509 (2007) – – 1 H D G DQ916148Pm Mas-like SPH(2007) Hemocytes Increased expression after infection of Vibrio 1 H D G DQ455050

T.H.T. Ma et al. / Developmental and Comparative Immunology 44 (2014) 163–172 165

2.4. PCR expression studies for PmArgonaute, PmPPO1, and PmPPO-II

Primers for PCR amplification of the genes are shown in Table 1.As a control to estimate the expression of PmPPAF, the expressionof the house keeping b-actin gene (GenBank Accession#EF087977) was analyzed by RT-PCR using gene specific primers(i.e., PmActinF: 50-atgtgtgacgacgaggatcttg-30 and PmActinR 50-aaacttcaccggtagtacgacct-30; (Fig. 7c). Diluted cDNA was amplifiedin a 30-l reaction containing 0.5 M free nucleotides, 0.2 M primer,and 1U of Taq DNA polymerase in the provided PCR buffer. Theappropriate cDNA dilutions were empirically determined for eachgene to ensure that signals were derived only from the exponentialphase of amplification. PCR was performed with a DNA ThermalCycler programmed for initial denaturation of 1 min at 94 �C, fol-lowed by 30 cycles of 45 s at 94 �C, 45 s at 60 �C, and 1 min 30 sat 72 �C, and a final extension time of 5 min at 72 �C. Aliquots(20 l) of the PCR were subjected to electrophoresis in 1% agaroseethidium bromide containing gel. Intensity of the bands was quan-tified with the use of Image J software (http://rsb.info.nih.gov/ij/in-dex.html). Normalization of PCR results was performed by analysisthe P. monodon house keeping gene-actin from the same cDNA

samples. In addition to PmPPAF, we have also analyzed the expres-sion of other P. monodon genes that may be involved the melaniza-tion pathway. These genes include the Argonaute gene thatfunctions in the RNA interference pathway, and the PPO1 andPPO II genes potentially involved in innate immunity of the shrimp.The expressions of these genes were monitored using RT-PCR andgene specific primers as shown in Table 1.

2.5. Functional study of PmPPAF by RNA interference

To prepare DNA template for dsRNA synthesis, a partial PmPPAFcDNA fragment was amplified by PCR using T7 promoter linkedprimers (Table 1). PCR was performed using similar temperatureprofiles as described from the above. PCR products were analyzedby 2% agarose gel electrophoresis. The PCR product was purified bythe GENECLEAN� II Kit (BioGene, California, USA). For PmPPAFdsRNA synthesis, 1–2 lg of T7 promoter linked PmPPAF partialcDNA was used as template and the transcription reaction was per-formed with T7 Megascript RNAi Kit (Ambion, USA) according tothe manufacturer’s recommendations. In 20 ll reaction, DNA tem-plate was mixed with nuclease-free water, 2 ll 10� T7 Reaction

166 T.H.T. Ma et al. / Developmental and Comparative Immunology 44 (2014) 163–172

Buffer, 2 ll ATP solution, 2 ll CTP solution, 2 ll GTP solution, 2 llUTP solution and 2 ll T7 Enzyme Mix. The mixture was incubatedat 37 �C for 18 h. During the transcription, the two RNA strandswere hybridized to form dsRNA. For dsRNA injection, juvenileshrimp of 16–20 g were used and 50 l (in 1� phosphate bufferedsaline) of PmPPAF dsRNA (3 lg) was injected into shrimp throughthe arthrodial membrane of the periopods by a Hamilton syringe.The controls received either a non-specific dsRNA (i.e., a DrosophiladsRNA provided by the Megascript RNAi kit, Ambion) or PBS injec-tion. At different time intervals, shrimp were sacrificed and totalRNA were extracted from the hypodermis and hemolymph sam-ples were collected for later assay of the prophenoloxidase activi-ties. The relative level of PmPPAF expression in the hypodermiswas used to indicate the effect of RNAi between the treatmentand control groups.

2.6. Effect of Anti-PmPPAF antibody injection on PO activity

A synthetic peptide (i.e., LRTEYSEDDNVEEKH) derived from theN-terminal end of PmPPAF, corresponding to amino acid residues148–161 was coupled to a carrier protein, rabbit serum albuminand used as an antigen to inject 2 rabbits for antibody production.All procedures and handling of animals for this process were re-viewed and approved by the University Laboratory Animal Careand Use Committee, and were performed in accordance with theGuidelines for the Care and Use of Laboratory Animals. 100 lg ofthe RSA-conjugated peptide was used to inject New Zealand WhiteRabbits at three different locations. The injections were repeatedthree times at 2-week intervals. One week after the fourth injec-tion, anti-serum against PmPPAF was prepared from the rabbitwhole blood. To study the influence of anti-PmPPAF on the activa-tion of proPO, 40 lg of hemocyte lysate supernatants (HLS) waspre-incubated with 4 lg of anti-serum or pre-immune serum at25 �C for 40 min. After the pre-incubation, PO activity assay waspreformed as described below.

2.7. PO activity assay

Hemocyte lysate supernatants (HLS) were prepared fromshrimp hemocytes. The hemocytes were washed and resus-pended in CAC buffer (10 mM sodium cacodylate, 450 mM NaCl,20 mM CaCl2, pH 7.0). The cell suspensions were homogenizedwith a sonicator (output 5, duty cycle 50%) and centrifuged at20,000g for 40 min at 4 �C. The resulting HLS were used forthe PO activation assay. The protein concentration of HLS weredetermined and adjusted to 250 lg/ml by the Bradford method,using bovine serum albumin as a standard (Bio-Rad Protein As-say kit), before the enzyme activity assay. For PO activity assay,L-3, 4-dihydroxyphenylalanine (L-dopa) (Sigma, USA) was usedas substrate. 50 ll of HLS was incubated with 50 ll of elicitor,sodium dodecyl sulfate (SDS) (0.5 mg/ml, Sigma), at 37 �C for15 min. 50 ll of L-dopa (5 mg/ml) was then added and incubatedat room temperature (25 �C) for 5 min, 150 ll of CAC buffer wasthen added to dilute the reaction. The absorbance of optical den-sity at wavelength of 490 nm was measured. For the negativecontrol groups, CAC buffer was used instead of elicitor in theincubation.

2.8. Effect of WSSV challenge on PmPPAF expression shrimp

To study if PmPPAF is involved in immune responses inshrimp, we challenged shrimp separately with white spot syn-drome virus (WSSV). The shrimp (P. monodon) used for the chal-lenge experiment were purchased from a local seafood market.The procedure for the extraction of WSSV for the challengeexperiment was described earlier (Zhao et al., 2007). Briefly, the

shrimp were acclimated at 25 �C in artificial seawater (30 partsper thousand) and muscle from WSSV infected shrimp washomogenized in 1� PBS (1:10 ratio). The extract was centrifugedat 6000 rpm for 5 min. The presence of the WSSV virus in the ex-tract was confirmed by PCR (Zhao et al., 2007;), the supernatantwas used for intramuscular injection. Supernatant (50 ll pershrimp) from WSSV infected shrimp homogenate was injectedinto the shrimp through the arthrodial membrane. At differenttime intervals, total RNA from hemocytes of the virus-injectedshrimp was prepared for Northern blot analysis. Control shrimpwere injected with healthy shrimp homogenate.

3. Results

3.1. Cloning and molecular characterization of PmPPAF

Using degenerate primers (PmPPAF-F1 and PmPPAF-R1) span-ning the CLIP domain and trypsin-like domain of the other P.monodon serine protease, a partial cDNA was amplified by RT-PCR (Fig. 1; Table 1). DNA sequencing determination confirmedthat this partial cDNA encoded a protein with high similarity toserine protease. Subsequently, 50 RACE primers (PmPPAF-R2 andPmPPAF-R3) were designed and a partial cDNA encoding the sig-nal peptide and part of the CLIP domain was amplified by 50 RACE(Fig 1). Similarly, 30 RACE primers (PmPPAF-F2 and PmPPAF-F3)were designed and a partial cDNA encoding the 30 end of thecDNA were cloned. The complete PmPPAF cDNA and the deducedamino acid sequence of PmPPAF were re-constructed (Fig. 1a).The full-length PmPPAF cDNA was 1444 bp in length with anopen reading frame (ORF) of 1131 bp (Fig. 2). The deduced openreading frame of SP cDNA was 394 amino acids. The start (ATG)and stop codon (TAA) are located at positions 69–71 and 1251–1253 bp of the cDNA, respectively. A potential polyadenylationsignal, AATAAA, is located at positions 1408–1414 bp. As pre-dicted by the SignalP software (http://www.cbs.dtu.dk/services/SignalP), residues 1–16 represent the putative signal peptide se-quence. Therefore, the mature SPH is composed of 378 aminoacids residues (Supplementary data). The estimated molecularmass of PmPPAF mature protein is 43.5 kDa with a predicted pIof 5.19.

Following the signal peptide sequence is a linker region(residues 17–29) joining the signal peptide and the putative CLIPdomain (AA30–AA99). In CLIP domain of PmPPAF, six cysteine res-idues were identified (Cys30, Cys37, Cys43, Cys91, Cys98, andCys99). By comparing the canonical CLIP domains with other SP,it is predicted that these cysteines will form three pairs of disulfidebonds. The other conserved domain is in the carboxyl-terminal se-quence (residues 128 through 379) consisting of 9 conserved cys-teine residues. Most of these cysteines also align perfectly well inmost of the P. monodon serine proteinases. However, the serine res-idue of the typical catalytic triad (i.e., His179, Asp229, and Ser330)found in most serine proteinases is modified to His179, Asp229and Gly330 in PmPPAF (Fig. 3). Therefore, like other serine prote-ases identified (i.e., PmMasSPH, PmcSPH, and PmSP), PmPPAF canbe considered as a serine proteinase homologue.

BlastX search analysis revealed that PmPPAF had the highest ami-no acid similarity with the serine proteinase like protein (PlSP-IIc;GenBank# ACB41380) of the crayfish P. leniusculus (61% similarity).When amino acid gaps are allowed in the alignment, PmPPAFshowed 61–67% similarity to PPAFs of the shrimp L. vannamei (Gen-bank# AFW98993) and Fenneropenaeus chinensis (Genbank#AFW98986). Similarly, within the P. monodon, PmPPAF shared 67%sequence similarity with PmMasSPH (Genbank# ABI78947), PmcSPH(Genbank# AY600227) and PmSP (Genbank AAQ93679). Whencompared to other decapods, PmPPAF showed high sequence

PmPPAF --------------CNNNRDICVPYYLCRDGIVITDGSGLIDIRIKPRLGGNNTSATRSGTPEPARSSATEFISECPSFLDVCCTDP-TRKPPPPPKYVAQ---CGRRNEEGVNARIVDFPmMasSPH ---VQPVPGSQTGQQTGNQIAGPQGCICLPVNQRCPYDGSSGSTSGAGVFDVRIVNRPG----------AGGECGIPGQKICCPPGVRPARPVRPPPVAPQPVPIQRSGCGGQNPIPYRQPmSPH ---VDPVPG---GISGGGNVVLPPGCFCIPVNQVCPFR-----------TEVRVGFRP-----------VVERC--PDQKECCSS----IGATTLPPTSPLGS------CGRQS--VVRDPmSPH509 --------------CNNGLGVCVPYYLCNEGNVITDGAGLIDIRFGSSKKSNDTSTRSS--------------SDCPQFLDVCCTNPNPPDVVTPAPYTPR---CGKRNSQGFDVRITGFPmSPH516 --------------CNNGLGVCVPYYLCNEGNVITDGAGLIDIRFGSSKKSNDTSTRSS--------------SDCPQFLDVCCTNPNPPDVVTPAPYAPR---CGERNSQGFDVRITGFCrayfish NPVERVVEGGNYAICNNGEGSCVPYYLCREGDIVTDGAGLIDIRFGG----NTTVTRSS--------------SECPQFLDVCCNNPQTVVPPTPIPYTSD---CGRRNPQGVNARILGFCrab ------------QVCRGGAGVCVPYYLCQDDKVNTDGAGIIDIRTGS---------------------------ECANFLDVCCTNPTGPVSTTQXPFTSR---CGNRNYNGIDVRIQGFApis ---DTILGTENRPKQPETNCECVPYYLCHNGSILENGTGIIDIRLGFDDESVTKHRTSR------------ALGQCENYLDVCCKPPDRIEKVTPPPVEKKGT-CGQRHPQGVGFRITGDLimulus ----GISSRVGNPQSGFGNCECVPYYLCKDNNIIIDGSGLLDPRKKPVASKEPKLSARLGP---------EGPSGCGPFH-VCCIAP---ETSTVKPYTHQ---CGFRNVNGINKRILSPPmSP ---------------CTS-------LELARQKSNPSLQLFFGTVLG---------------------------KELLKFTERCCNS--------LIPSEEE---CGISEPTGPIT-----LvSP --------------SCASRSDCGGYLSLARRRSDPSLKPFFDTNLG---------------------------RELLDFTTRCCKSP-------LIPSEQQ---CGFPEPSGRLS-----

* * PmPPAF PA-NQAQFGEFPWMAAVLRTEYSEDNVEEKHYVCGGSLIHPQVVLTAAHCVYNI-NPT TLNVRLGEWDTQK--AYELYPHQDRNVGYVVTHREFNH----INLFNDFALLFLDSPV--ELPmMasSPH AQYAEATFGEYPWMVVVLD--------FGDGYKGGGVLVAPDWVLTAAHKVYNE--R SLKVRLGEHNVRQRQDHPNYAHLEVPIDRIIIHPNFDN-----QALLNDVALLHLAQPVNVNQPmSPH_ ITHGPALFGELPWMTMVLN--------GRGSYVAGGALISSEWVLTAAHRIRNQ---R NLIVRLGELDFSKPQDSPQYTHRDVPIDNIIVHPQFNS----QTLANDVALLHLSRPVYTAIPmSPH509 KD-NEAQFAEFPWMTAILR-VEKVGKKELNLYVCGGSLIHPSIVLTAAHCVHSK---AAS SLKTRFGEWDTQK--TYERYPHQDRNVISVKIHPNYNS--GALYNGFALLFLDSPA--TLPmSPH516 KD-NEAQFAEFPWMTAILR-VEKVGKKELNLYVCGGSLIHPSIVLTAAHCVHSK---AAS SLKTRFGEWDTQK--TYERYPHQDRNVISVKIHPNYNS--GALYNDFALLFLDSPA--TLCrayfish KD-NQAQFGEFPWMIAVLRQEEVVVDKPVNLYVCGGSLIHPSVVLTAAHCVASW---DAG VLKVRAGEWDTQR--TYELFPHQDRNVAKVVVHQGYKS--GPLFNDFALLFLDQPF--ELCrab QG-NETQVAEFPWMTAVLK-KEVVSGEEINLYLCGGSLIHPSIVLTAAHCVDKH-TSP HLRVRLGEWDTQN--EYEPY-DQDRDVATVVIHPDFNP----SNLHNDYALLYLQTPA--DLApis VN-GEAQFGEFPWMVAIIK-EENIGEEKLNVYQCGGSLIHKQAVLTAAHCQ-------AS ELRIRAGEWDTQT--KNEIFPHQDRNVQNVIIHENFAG--GTLYNDFAILILSEPL--NLLimulus NGKDLSEFGEWPWQGAVLK-----VEGKVNIFQCGAVLIDSYHLLTVAHCVYKFTLENAF PLKVRLGEWDTQN--TNEFLKHEDYEVEKIYIHPKYDDERKNLWDDIAILKLKAEV--SFPmSP ---DLQRRGAWPWYAALSP-----YPGDEFLVFCGGTLITRRHVLTGAHCLVLP--NSLT LKYARLGEYDLNV-------KNEANYVELGFARYEEAGYDSN--RDIAIITLDSDV--EFLvSP ---DLQRRGAWPWFASLSA-----TSGSNFRPISGGSLITRRHVLTTGHTLLGH--ISQQ PKYARLGEYDLSV-------ENDAPYIELAILRFQEAGYNADNTRDVGVITLERDV-—VF

* PmPPAF APNVDTVCLPEQGQTFD--GSYCWATGWGKDRFGKEGVFQNVLKEVELPIVSQYDCQESL RTTRLGKLFKLHP-SFTCAGGIAGVDTCTGDGGSPLVCLGLGNS--YVQTGIVAWGIG-C PmMasSPH YPHIGTACLPSPGQIFN--GQTCWVTGWGKDAFETNGNFQEILKEVDVPIVDSFRCQASL QQTRLGLSFLLNQQSFICAGGIAGKDACTGDGGSPLVCP--TQN-GWTVVGLVAWGIG-C PmSPH APHIGAVCLPSQGQIFQ--GRKCVVSGWGGDPNIPGNAFQNLLRVVEVPMVDPFACQQRL GTARLGANFTLDQTSFVCAGGVEGNDACTGDGGSPLVCL--NDNRSWTLVGLVAWGLG-C PmSPH509 APNVDTVCLPQANQKFD--YDTCWATGWGRDKFGKEGEFQNILKEVALPVVPNHDCQNGL RTTRLGSFFQLHN-SFMCAGGQQGIDTCKGDGGSPLVCEAVAGSGVYVQAGIVAWGIG-C PmSPH516 APNVDTVCLPQANQKFD--YDTCWATGWGRDKFGKEGEFQNILKEVALPVVPNHDCQNGL RTTRLGSFFQLHN-SFMCAGGQQGIDTCKGDGGSPLVCEAVAGSGVYVQAGIVAWGIG-C Crayfish APNVDTLCLPNQDQNLL--GVECWATGWGKDRFGKEGEFQNVLKKIKLGLTPNDKCQAAL RTTRLGKFFVLDK-SFACAGGEAGLDTCKGDGGSPLMCQVSPNK--YVQAGIVAWGIG-C Crab SRNVDVICLDNAPQILNP-QHDCLVTGWGKDRFGKKGIFQNVLKKIDLPYVXHGKCQAAL RTTRLGKFFILDK-SFLCAGGEAGKDSCSGDGGSPLVCLDKTKT-QYVQVGIVAWGIG-C Apis MENVDLVCLPERNTVFD--GTRCFASGWGKDKFGKEGHYQVILKKINLPVVPHNQCQDLL RKTRLGKYFVLDS-SFICAGGESDKDTCKGDGGSPLVCPSNSNPNIYLQAGIVAWGIG-C Limulus GPHIDTICLPNNQEHFA--GVQCVVTGWGKNAY-KNGSYSNVLREVHVPVITNDRCQELL RKTRLSEWYVLYE-NFICAGGESNADSCKGDGGGPLTCWRKDGT--YGLAGLVSWGIN-C PmSP NDFIRAACLSFNYEDEDFVNQHLAVVGYG---HTSDGRSSSVPVAAVVPVVDLATCKQKY HNSDR------ITDSVICAGN--GNDACLGDGGSPLNYFDVNTS-RFYVVGIVSFGSNTCLvSP NDFIRPVCLPFNYKDEDFVNQHLAVVGYG---RTSDGKLSSLPVAAVVPVVDLATCRQKY KNNHR------ITDSVICAGG--YNDACTGDSGGPLHYFDVNSK-RFYVVGXVAFGATLC

PmPPAF GEDNVPGVYANVPAFVNWIKDNVNSILLQEFNYDNKNYW------PmMasSPH AQGNVPGVYVNIPNMMDFIRQFVRF--------------------PmSPH AQREVPGVYVNVASYTNFIRQFVRF--------------------PmSPH509 GEQGVPGVYADVGYASDWIQTEAN-IGLASLYSIQGYDWDYGRFIPmSPH516 GEQGVPGVYADVGYASDWIQTEAN-IGLASLYSIQGYDWDYGRFICrayfish GEGGIPGVYANVPYASKWIKDTSN-SILSELKVTVGNYWDYIPS- Crab GTSNIPGVYADVLYGYDWIVTEAD----KLLAGPIVDYWQYV---Apis GEGGIPGVYANVASVRDWIDEQMAFYNLDNTVYQAQKIN------Limulus GSPNVPGVYVRVSNYLDWITKITG--------RPISDYWPRS---PmSP GLSGIPGGYTRVGAYLGWINDTISNDPV-----------------LvSP GQSDLPGGYTRVGAFLDWINDAIISDTT-----------------

Fig. 2. Multiple alignments of the single clip domain from selected arthropod serine protease homologues using the Clustal X program. Sequences included in this study are:P. monodon PmPPAF (GenBank# EU082201); PmMasSPH; Masquerade-like serine proteinase homologue (GenBank# DQ455050); PmSPH509 (GenBank# DQ916148);PmSPH516 (GenBank# DQ403191); PmcSPH (GenBank# AY600627); PmSP serine proteinase (GenBank# AAQ93679). The putative proteolytic activation cleavage site is shownby a downward arrow. Gaps (–) were introduced to maximize the alignment of the clip domain cysteines and the SPH catalytic triad is indicated by open circles above thedark thick line which shows the trypsin-like region of the PmPPAF; the amino acid residues forming the substrate specificity pocket in catalytically active serine proteases areindicated by an⁄.

T.H.T. Ma et al. / Developmental and Comparative Immunology 44 (2014) 163–172 167

identity to the prophenoloxidase activating enzyme of the crab C.sapidus (Genbank# AAS60227: 61% similarity and other invertebrateserine proteinase including the mosquito Aedes aegypti serine prote-ase (Genbank# XP_001655706: 58% similarity), Bombyx mori mas-querade-like serine proteinase homologue (Genbank#NP_001037053), T. tridentatus limulus factor D (Genbank# 192/344: 55%) and African malaria mosquito A. gambiae (GenBank#Z49813).

To elucidate the structural features of PmPPAF, selected serineproteases of P. monodon were aligned and compared. This analysisrevealed that PmPPAF has a much higher degree of similarity withother serine proteinase of P. monodon (Fig. 2). In all the P. monodonserine proteinase aligned, the cysteine residues are conserved de-spite some variation in the number of total cysteine residues ineach protein (Fig. 2). A phylogenetic tree of PmPPAF with selectedserine proteinase and serine proteinase homologues of other ani-mals (Fig. 3) revealed that PmPPAF shows a much closer evolution-ary relationship with the three SPH (i.e., PmMasSPH, PmSPH509,PmSPH516) of P. monodon. PmPPAF is more closely related to thecrab CsPPAF and some insect SPs (i.e., AgSP, MsSP, DmSP). However,other P. monodon serine proteinase homologues (i.e., PmcSPH,PmSP) were distantly related to PmPPAF and clustered separatelyin two different groups. For example, PmcSPH is closely related tothe crayfish PlcMas-II and PmSP is more closely related to the M.japonicus MjSP and the L. vannamei LvSP (Fig. 3).

3.2. Tissue distribution and expression studies of PmPPAF

PmPPAF transcripts was detected in the hypodermis (Hd),hemocytes (Hc), gill (Gi), swimming leg (Sl), brain (Br) and eye-stalk (Es) but not in hepatopancreas (Hp), heart (Ht), muscle(Mu), thoracic ganglion (Tg) and ventral nerve cord (Vn)(Fig. 4a). The apparent larger size of the PmPPAF transcript de-tected in the brain and hypodermis is due to the overall slowmobility of the total RNA of these two tissues (Fig. 4a). Of allthese tissues inspected, the eyestalk had the smallest quantityof PmPPAF transcript. A similar tissue expression pattern was ob-served using RT-PCR (data not shown). To study the expressionpattern of the PmPPAF in shrimp during the molting cycle, themolt stage of the shrimp were determined by pleopod setogene-sis (Chan et al., 1988), Northern blot result indicated that thePmPPAF transcript level was low during the intermolt stage, de-crease further in early premolt but the level gradually increaseto a much higher level towards the late premolt (stage D3)(Fig. 4b). The results suggest that PmPPAF expression is correlatedwith the molt cycle and may be important for the immune de-fense during the late molt cycle stage of the shrimp.

For shrimp injected with WSSV, there was a gradual decrease inthe expression of PmPPAF from 4 to 24 h thereafter increasing to ahigher level than the controls at 48 h post-injection (Fig. 5c).Although there was a slight variation in overall PmPPAF gene

PmMasSPH523

PmSPH509

PmSPH526

PmPPAF

CsPPAF

AgSP

MsSP

DmSP

PmcSPH

PlcMas-II

TtppA

DvSP

MjSP

PmSP

LvSP

0.1

10099

100

90

99

78

56

87

100

99

9480

Fig. 3. Phylogenetic analysis of representative serine proteinases or serine proteinase homologues of the P. monodon and other arthropods. The phylogenetic tree wasconstructed by the neighbor-joining method (MEGA 4), based on the amino acid alignment (Clustal W) of individual SPH sequences from selected crustaceans and insects. P.monodon sequences include PmPPAF (GenBank# EU082201); PmMasSPH; Masquerade-like serine proteinase homologue (GenBank# DQ455050); PmSPH509 (GenBank#DQ916148); PmSPH516 (GenBank# DQ403191); PmcSPH (GenBank# AY600627); PmSP serine proteinase (GenBank# AAQ93679). Sequences used in this study are theinsects M. sexta serine proteinase-like protein 1 (MsSP, GenBank# AAM69352); A. gambiae serine protease (AgSP, GenBank# CAB90818); D. melanogaster Trypsin-like serineprotease (DmSP, GenBank# AAF52904); D. variabilis factor D-like protein (DvSP, GenBank# AAN78224). Other sequences included are from the shrimp M. japonicus serineproteinase homologue (MjSP, GenBank# BAD34945); L. vannamei serine proteinase (LvSP, GenBank# AAQ92356), the crab C. sapidus phenoloxidase activating factor (CsPPAF,GenBank# AAS60227); the crayfish P. leniusculus low mass masquerade-like protein (PlsMas-II, GenBank# AY861652); T. tridentatus factor D (TtppA, GenBank# BAA13312);The bootstrap support is shown for each branch; scale bar indicates 10% sequence divergence.

(a)

(b)

(c)

M Es Hd Hc Hp Ht Gi Mu Sl Br Tg Vn

C D0-1 D1-2 D2-3 D3

PmPPAF expression after healthy injectionPmPPAF expression after WSSV injection

a

bb

c

c

a aa a

Fig. 4. Expression study of PmPPAF (a) Tissue distribution of PmPPAF transcript(arrow) using Northern blot analysis. Top panel shows the PmPPAF transcriptdetected by the Dig-labeled probe. The lanes, from left to right, are 1 kb marker (M),eyestalk (Es), hypodermis (Hd), hemocyte (Hc), hepatopancreas (Hp), heart (Ht), gill(Gi); muscle (Mu), swimming leg (Sl), brain (Br), thoracic ganglion (Tg) and ventralnerve cord (Vn). Lower panel shows the rRNA of the corresponding samples stainedby ethidium bromide. (b) Expression pattern of PmPPAF in intermolt (c) and premolt(D0–D3) stages of P. monodon. Top panel shows the Northern blot signal and lowerpanel shows the rRNA staining of the same sample (c) Bar diagram showinghypodermis in vivo time course expression of PmPPAF after WSSV infection ascompared to the healthy shrimp extract injection (N = 8 with 3 triplicates for eachanimal). Y-axis show the percentage of expression as compared to the control(assigned as 100%) and same letter on the top of the bar indicate no significantdifferent (P < 0.05) for the expression level.

168 T.H.T. Ma et al. / Developmental and Comparative Immunology 44 (2014) 163–172

expression in shrimp injected with healthy shrimp extract, the dif-ference was not significant.

3.3. Effect of PmPPAF gene silencing by RNA interference

Based on our previous experience in shrimp RNA interferenceexperiments (Tiu and Chan 2007), we injected 3.0 g/gm of PmPPAFdsRNA in shrimp collecting samples at 24 and 48 h after injection.There was no significant difference in mortality in shrimp injectedwith the PmPPAF dsRNA and the control animals. When shrimp wereinjected with PmPPAF dsRNA, a decrease in PmPPAF transcript level to<5% was observed in 24 h reducing further by 48 h. However, therewas no change in the PmPPAF transcript level in the uninjected con-trol (Fig. 5a). In addition to decrease in PmPPAF transcript level,Northern blot results also indicated the presence of residual dsRNAhybridized to the DNA probe suggesting the stability of this dsRNAin shrimp (Fig. 5a).

To study the potential functions of PmPPAF in shrimp innateimmunity, the expression of PPO genes, important enzymes inthe PO immune response cascade, were analyzed in dsRNA injectedshrimp, using probes for two different PPO genes. The first PPOprobe was obtained from the cloning of the PPO gene reportedby Sritunyalucksana et al. (1999). The PPO gene (tentatively calledPmPPO2) used was newly cloned PPO from our laboratory (Supple-mentary data). No significant difference in PmPPO1 expression le-vel in PmPPAF dsRNA injected shrimp was observed relative to thecontrols at 24 and 48 h post-injection (Fig. 6a). However, a progres-sive decrease in PmPPO2 expression level was observed, with lim-ited effect at 24 h post injection and a stronger reduction at 48post-injection (Fig. 6b). In contrast to the PmPPO1 and PmPPO2genes, the expression of other genes involved in the regulation ofRNA interference (i.e., PmArgonaute) decreased in shrimp in thefirst 8 h and then increased from 16 to 48 h (Fig. 6c).

The phenoloxidase activity of the PmPPAF dsRNA injected shrimpwas reduced by an estimated 26% in hemolymph samples relative tothose of the positive controls (Fig. 7a). PO activity is the same forboth the anti-PmPPAF injected group and the pre-immune serum in-jected group when no elicitor is added. However, when the elicitor is

0

0.2

0.4

0.6

0.8

1

1.2

Con-24h Con-48h ds-24h ds-48h

Rela

tive

Inte

nsity

of P

mPO

A ex

pres

sion

Control dsPPAF

24h 24 h 48h 48h 24h 24h 48h 48h

Rel

ativ

e In

tens

ity o

f Pm

PPAF

exp

ress

ion

Fig. 5. Gene knockdown of PmPPAF by RNAi. The expression of PmPPAF in shrimp was monitored at 24 and 48 h in shrimp after the injection of dsRNA (PmPPAF). (a) Northernblot analysis of the PmPPAF transcript level in control and dsPPAF injected shrimp. Each lane is a sample from an individual shrimp. The arrow indicates the PmPPFA transcriptand the smear in the PmPPAF lanes represents degrading dsRNA; (b) histogram showing the PmPPAF transcript level. A significant decrease in PmPPAF RNA transcript level isobserved in shrimp injected with dsRNA for PmPPAF.

24 48 24 48 –ve

PmPPO 1

PmPPO 2

Control dsPmPPAF

00.10.20.30.40.50.60.7

Con-24h Con-48h ds-24h ds-48h

Rel

ativ

e In

tens

ity

PmPPO 2Control dsPmPPAF

00.10.20.30.40.50.60.70.80.9

Con-24h Con-48h ds-24h ds-48h

Rel

ativ

e In

tens

ity

PmPPO 1

0

1

2

3

4

5

Control 4 6 8 16 24 48Time after injection (Hours)

Rel

ativ

e in

tens

ity o

f Pm

Ago

exp

ress

ion

PmAgo

β-actin

Con 4 6 8 16 24 48

24 48 24 48 –ve

(a)

(b)

(c)

β-actin

β-actin

PmAgo

Fig. 6. Effect of PmPPAF gene silencing on the expression of (a) PmPPO1; (b) PmPPO2; and (c) PmArgonauat and b-actin gene expression. Left: RT-PCR detection of expressionof the genes in the hypodermis at 24 and 48 h after dsRNA (PmPPAF) injection. Right: histograms showing the expression levels of the gene (N = 8).

T.H.T. Ma et al. / Developmental and Comparative Immunology 44 (2014) 163–172 169

added, an obvious decrease in PO activity was observed in theshrimp injected with Anti-PmPPAF antibody (Fig. 7b).

4. Discussion

4.1. Molecular characterization of PmPPAF

Although there are many SPH and SP identified and cloned in P.monodon, it is interesting to know that PmPPAF is more similar to

the Masqueradel-like protease of the crayfish P. leniusculus (Leeand Söderhäll, 2001). The crayfish Masqueradel-like protease is amulti-functional pattern recognition protein that has cell adhesionand opsonic activities to gram-negative bacteria as well as to yeast.Whether PmFFAF has similar pattern recognition properties as thecrayfish requires further examination. Moreover, both PmPPAF andPlMas-II also lack the glycine-rich domain that other P. monodonSPH have. PmPPAF contains a highly conserved N-terminal CLIPdomain and a C-terminal proteinase domain similar to most

0

0.04

0.08

0.12

0.16

0.2

Control Treatment negative control

Phen

olox

idas

e A

ctiv

ity(4

90nm

)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

Without elicitor With elicitor (SDS)

Phen

olox

idas

e ac

tivity

Anti-POA serum

Pre-immune serum

(a)

(b)

Fig. 7. Effect of PmPPAF dsRNA injection and antibody injection on phenoloxidase activity. (a) Hemocyte PO activity decreases in shrimp injected with dsRNA for PmPPAF ascompared to the control and is higher relative to the negative control group. Sample size N = 8 for each treatment. (b) Decrease in PO activity in shrimp injected with anti-PmPPAF antibody.

170 T.H.T. Ma et al. / Developmental and Comparative Immunology 44 (2014) 163–172

arthropod serine proteinases. Compared with other serine protein-ase homologues of P. monodon, the mature PmPPAF is much smallerdue to the shorter N-terminal end which lacks a long stretch of gly-cine-rich repeats compared with other SPH of P. monodon. The gly-cine rich region is thought to have anti-bacterial function thoseSPHs (Amparyup et al., 2007; Lin et al., 2006; Jimenez-Vega et al.,2005). Absence of this domain suggested that PmPPAF may nothave an anti-bacterial function but more work is needed to confirmthis. The CLIP domains are features found only in arthropod serineproteinases. They are cleaved from the proteinase domain uponactivation of the pro-protein but remain attached by disulfidebridges. The CLIP-domain of PmPPAF has six cysteine residues in-volved in disulfide bridge formation similar to the CLIP domainof PPAF from the crab C. sapidus (Buda and Shafer, 2005). The func-tion of the CLIP domain in arthropod serine proteinases are not yetwell established but has been postulated to have multiplefunctions in arthropod immune responses and in regulating thefunction of the enzymes through their interaction with other pro-teins. Additionally, the CLIP domain may exhibit anti-bacterialactivity towards gram-positive bacteria as described for the serineproteinase of the crayfish (Huang et al., 2000). Unlike the PmPPAF,one SP of P. monodon (Jimenez-Vega et al., 2005), an SPH of M.

japonicus (Rattanachai et al., 2005) and an SP of the crab Seratusserrata have only 4–5 instead of six conservative cysteines. Thesemodified secondary structures have been named pseudo-CLIP do-mains (Jimenez-Vega et al., 2005) and are expected to form only2 Cys disulfide bridges. These results also suggested that the pres-ence of other serine proteinases with a pseudo-CLIP domain withfunctions in an innate defense system.

The C-terminal serine proteinase domain of PmPPAF shares sig-nificant sequence homology with those of typical arthropod PPAFsincluding the PPAF of C. sapidus (Buda and Shafer, 2005), the SP-like protein 4 of M. sexta (Yu et al., 2002), masquerade-like SPHof B. mori and PPAF of Tenebrio molitor (Lee et al., 2002). PmPPAFalso shares common catalytic sites (i.e., His, Asp, and Gly) andstructural characteristics with other SPHs. Because of the substitu-tion of an essential active serine residue, PmPPAF is predicted tohave weak or no enzyme activity. SPH that lack enzyme activitydue to a modified proteinase domain have been suggested to medi-ate protein–protein interaction or to act as antagonists of serineproteinase to regulate and control that enzyme activity. In the phy-logenetic tree, PmPPAF is more closely related to non-catalyticPPAFs rather than to the active serine proteinase. PmPPAF is closestto C. sapidus PPAF among the SPH that form only three disulfide

T.H.T. Ma et al. / Developmental and Comparative Immunology 44 (2014) 163–172 171

bridges. Currently in P. monodon, there are at least 6 CLIP domainSP and/or SPH cloned and characterized (Sriphaijit et al., 2007;Amparyup et al., 2007; Jimenez-Vega et al., 2005; Lin et al.,2006). Based on the phylogenetic tree results and the large numberof PmSPH and PmSP identified so far, we predict that there aremore serine proteinases involved in immunity shrimp yet to beidentified and cloned.

4.2. Expression study of PmPPAF

The diverse functions of the serine proteinase and/or serine pro-teinase homologues mean their expression pattern in other organ-isms is highly diverse. Despite the sequence similarity betweenPmPPAF and PmMasSPH, their expression patterns are quite differ-ent. PmPPAF is expressed in many tissues including the hypoder-mis, hemocytes, gill, swimming leg and brain PmMasSPH isexpressed only in the hemocyte (Amparyup et al., 2007). Thisbroad tissue distribution pattern of PmPAFF is in contrast to somecrustaceans SP/SPH which were mainly detected in the hemocytesand not in the hypodermis or other nervous tissue (Amparyupet al., 2007). The difference in expression pattern suggested thatthey may have a different function. Although only fewer RNA sam-ples were collected from different molt stages of P. monodon, thehigher expression level of PmPPAF at the premolt stage is similarto that of the PPAF of the crab (Buda and Shafer, 2005.). The factthat PmPPAF can also be detected in other tissues suggests thatPmPPAF may have a broader function such as involvement in themelanization process.

4.3. Potential role in fine-tuning the melanization pathway

The melanization pathway consists of a cascade of enzymaticreactions controlled by serine proteinase and their homologues.Since the by-products of the melanization pathway may be poison-ous, a mechanism is required to regulate harmful by-product accu-mulation. We speculate that the evolution of serine proteinases(with high PO activation activity) and serine proteinase homologue(with low activation activity) may serve as a control to regulate theaccumulation of these harmful by-products. At present, many ser-ine proteinase homologues of P. monodon have been identified.Several SPHs have been reported and their functions in the activa-tion of the proPO system have been demonstrated and discussed(Amparyup et al., 2013). Although there is a report on the latestcloning of a PmClipSP2 P. monodon, manual sequence alignmentof PmPPAF with PmClip2 revealed only a <50% overall similarity.Furthermore, the catalytic Triad in PmClipSP2 is His, Asp, and Serand PmClipSP2 is expressed mainly in the hemocyte. This suggeststhat PmClipSP2 and PmPFAF may have different functions in P.monodon. For example, the newly reported serine proteinasehomologues have their catalytic triad modified so that they maynot have a high PO activating ability. If both enzymes are presentin the hemocyte, it would be interesting to know if the expressionof either one will affect the function of the other. Moreover, fromthe small increase in PO activity after dsRNA and anti-PmPPAFinjection studies, it is predicated that the PO activating functionof PmPPAF would be low.

Serine proteinase and/or serine proteinase homologue (PmPPAFin this study) may act to fine-tune the melanization process or reg-ulate the accumulation of toxic intermediate metabolites harmfulto the animals. In the insect M. sexta, three prophenoloxidase acti-vating proteinases (PAP) have been isolated and their cDNAscloned that can directly activate proPO. PAP-1 was originally iso-lated from cuticle, whereas PAP-2 and PAP-3 are from hemocytes(Yu et al., 2002). Two SPH molecules have also been isolated fromM. sexta plasma and cloned (Yu et al., 2002). Both SPHs can serve ascofactors for each of the PAPs. Two non-catalytic cofactor proteins

are known from the beetle T. molitor as well, one designated PPAF(Kwon et al., 2000) and the other SPH (Lee et al., 2002). In short, itis logical to predict that shrimp also have a large number of serineproteinases or serine protease homologues that are involved in theProphenoloxidase activating system/pathway. In summary, wehave cloned and characterized the PmPPAF cDNA from P. monodon.Given the substitution of an essential active serine residue, PmPPAFis predicted to have weak or little enzyme activity. SPHs that lackenzyme activity due to the modified proteinase domain have beensuggested to mediate protein–protein interaction and control theirenzyme activity. The results of this study suggested that PmPPAF isa weak-catalytic enzyme and an immune-inducible gene associ-ated with the pro-phenoloxidase activation pathway and may beinvolved in the fine-tuning of the PO activating enzymes.

Acknowledgements

The author wishes to thanks the two anonymous reviewers fortheir help to improve the writing of the manuscripts. This researchis supported in part by the Hong Kong Research Grant Council (toSMC) and by a start up fund of the Guangdong Ocean University, PRChina.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.dci.2013.12.007.

References

Amparyup, P., Jitvaropas, R., Pulsook, N., Tassanakajon, A., 2007. Molecular cloning,characterization and expression of a masquerade-like serine proteinasehomologue from black tiger shrimp Penaeus monodon. Fish Shellfish Immunol.22, 535–546.

Amparyup, P., Promrungreang, K., Charoensapsri, W., Sutthangkul, J., Tassanakajon,A., 2013. A serine proteinase PmClipSP2 contributes to prophenoloxidasesystem and plays a protective role in shrimp defense by scavenginglipopolysaccharide. Dev. Comp. Immunol. 41, 597–607.

Buda, E.S., Shafer, T.H., 2005. Expression of a serine proteinase homologueprophenoloxidase-activating factor from the blue crab, Callinectes sapidus.Comp. Biochem. Physiol. B 140, 521–531.

Cerenius, L., Söderhäll, K., 2004. The prophenoloxidase-activating system ininvertebrates. Immunol. Rev. 198, 116–126.

Chan, S.M., Rankin, S.M., Keeley, L.L., 1988. Characterization of the molt Stages inPenaeus vannamei: setogenesis and hemolymph levels of total protein,ecdysteroids, and glucose. Biol. Bull. 175, 185–192.

Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acidguanidinium thiocyanate–phenol–chloroform extraction. Anal. Biochem. 162,156–159.

Gorman, M.J., Paskewitz, S.M., 2001. Serine proteases as mediators of mosquitoimmune responses. Insect Biochem. Mol. Biol. 31, 257–262.

Huang, T., Wang, H., Lee, S.Y., Johansson, M.W., Söderhäll, K., Cerenius, L., 2000. Acell adhesion protein from the crayfish Pacifastacus leniusculus, a serineproteinase homologue similar to Drosophila masquerade. J. Biol. Chem. 275,9996–10001.

Jiang, H., Kanost, M.R., 2000. The clip-domain family of serine proteinases inarthropods. Insect Biochem. Mol. Biol. 30, 95–105.

Jimenez-Vega, F., Vargas-Albores, F., Söderhäll, K., 2005. Characterisation of a serineproteinase from Penaeus vannamei haemocytes. Fish Shellfish Immunol. 18,101–108.

Kanost, M.R., Jiang, H., Wang, Y., Yu, X.Q., Ma, C., Zhu, Y., 2001. Hemolymphproteinases in immune responses of Manduca sexta. Adv. Exp. Med. Biol. 484,319–328.

Kawabata, S., Tokunaga, F., Kugi, Y., Motoyama, S., Miura, Y., Hirata, M., Iwanaga, S.,1996. Limulus factor D, a 43-kDa protein isolated from horseshoe crabhemocytes, is a serine protease homologue with antimicrobial activity. FEBSLett. 398, 146–150.

Kwon, T.H., Kim, M.S., Choi, H.W., Joo, C.H., Cho, M.Y., Lee, B.L., 2000. A Masquerade-like serine proteinase homologue is necessary for phenoloxidase activity in thecoleopteran insect, Holotrichia diomphalia larvae. Eur. J. Biochem. 267, 6188–6196.

Lee, S.Y., Söderhäll, K., 2001. Characterization of a pattern recognition protein, amasquerade-like protein, in the freshwater crayfish Pacifastacus leniusculus. J.Immunol. 166, 7319–7326.

Lee, K.Y., Zhang, R., Kim, M.S., Park, J.W., Park, H.Y., Kawabata, S., Lee, B.L., 2002. Azymogen form of masquerade-like serine proteinase homologue is cleaved

172 T.H.T. Ma et al. / Developmental and Comparative Immunology 44 (2014) 163–172

during pro-phenoloxidase activation by Ca2+ in coleopteran and Tenebrio molitorlarvae. Eur. J. Biochem. 269, 4375–4383.

Lin, C.Y., Hu, K.Y., Ho, S.H., Song, Y.L., 2006. Cloning and characterization of a shrimpclip domain serine protease homologue (c-SPH) as a cell adhesion molecule.Dev. Comp. Immunol. 30, 1132–1144.

Rattanachai, A., Hirono, I., Ohira, T., Takahashi, Y., Aoki, T., 2005. Peptidoglycaninducible expression of a serine proteinase homologue from kuruma shrimp(Marsupenaeus japonicus). Fish Shellfish Immunol. 8, 39–48.

Satoh, D., Horii, A., Ochiai, M., Ashida, M., 1999. Prophenoloxidase-activatingenzyme of the silkworm, Bombyx mori. J. Biol. Chem. 274, 7441–7453.

Söderhäll, K., Cerenius, L., Johansson, M.W., 1994. The prophenoloxidase activatingsystem and its role in invertebrate defence. Ann. N Y Acad. Sci. 712, 155–161(under review).

Sriphaijit, T., Flegel, T.W., Senapin, S., 2007. Characterization of a shrimp serineprotease homolog, a binding protein of yellow head virus. Dev. Comp. Immunol.31, 1145–1158.

Sritunyalucksana, K., Cerenius, L., Söderhäll, K., 1999. Molecular cloning andcharacterization of prophenoloxidase in the black tiger shrimp, Penaeusmonodon. Dev. Comp. Immunol. 23, 179–186.

Tamura, K., Dudley, J., Nei, M., Kumar, S., 2007. MEGA4: molecular evolutionarygenetic analysis (MEGA) software versions 4.0. Mol. Biol. Evol. 24, 1596–1599.

Tiu, S.H., Chan, S.-M., 2007. The use of recombinant protein and RNA interferenceapproaches to study the reproductive functions of a gonad-stimulatinghormone from the shrimp Metapenaeus ensis. FEBS J. 274, 4385–4395.

Wang, Y., Jiang, H., 2004a. Purification and characterization of Manduca sextaserpin-6: a serine proteinase inhibitor that selectively inhibitsprophenoloxidase-activating proteinase-3. Insect Biochem. Mol. Biol. 34, 387–395.

Wang, R., Lee, S.Y., Cerenius, L., Söderhäll, K., 2001. Properties of theProphenoloxidase activating enzyme of the freshwater crayfish, Pacifastacusleniusculus. Eur. J. Biochem. 268, 895–902.

Yu, X.Q., Jiang, H., Wang, Y., Kanost, M.R., 2002. Nonproteolytic serine proteinasehomologs are involved in prophenoloxidase activation in the tobaccohornworm Manduca sexta. Insect Biochem. Mol. B. 32, 197–208.