Molecular cloning and characterization of European seabass(Dicentrarchus labrax) and Gilthead seabream (Sparus aurata)complement component C3

I. Mauri a,*, N. Roher a, S. MacKenzie a, A. Romero d, M. Manchado c, J.C. Balasch a, J. Béjar b,M.C. Álvarez b, L. Tort aaDepartament de Biologia Cel$lular, Fisiologia Animal i Immunologia, Universitat Autònoma de Barcelona, O8193 Cerdanyola, Catalunya, SpainbDepartamento de Microbiología, Facultad de Ciencias, Universidad de Málaga, Campus Teatinos, 29071 Málaga, SpaincCIFAP “El Toruño”, Camino Tiro de Pichón s/n, 11500 El Puerto de Santa María, Spaind Instituto de Investigaciones Marinas, CSIC, Eduardo Cabello 6, 36208 Vigo, Spain

a r t i c l e i n f o

Article history:Received 30 November 2010Received in revised form10 March 2011Accepted 12 March 2011Available online 21 March 2011

Keywords:European seabassGilthead seabreamComplementViral and bacterial infectionAbsolute quantification

a b s t r a c t

We present the complete C3 cDNA sequence of Gilthead seabream (Sparus aurata) and European seabass(Dicentrarchus labrax) and its molecular characterization with a descriptive analysis of their structuralelements. We obtained one sequence for Gilthead seabream (gsbC3) which encodes a predicted proteinof 1656 amino acids, and two sequences for European seabass (esbC3_1 and esbC3_2) which encode twopredicted proteins of 1654 and 1587 amino acids respectively. All sequences present the characteristicstructural features of C3 but interestingly esbC3_2 lacks the anaphylotoxin domain and the cysteineresidue responsible for thiolester bond formation. Moreover, we have detected and quantified (by real-time PCR-based absolute quantification) specific isoform expression in European seabass depending onpathogen and density conditions in vivo. In addition, we have analyzed the tissue distribution pattern ofEuropean seabass and Gilthead seabream C3 genes under crowding stress and under pathologicalchallenges in vivo, and we have observed that crowding and infection status provoke changes inexpression levels, tissue expression pattern and C3 isoform expression balance.

! 2011 Elsevier Ltd. All rights reserved.

1. Introduction

For a long time, the complement system in mammals has beenregarded as a biological system that plays an important role in innateimmunity. However it has been also recognized a role of thecomplement as a modifier of acquired immunity [1e3]. This systemhas been mainly described in vertebrates but also in a wide phylo-genetic range of animals [4]. Over the past 10 years, analysis of DNAsequences revealed the presence of the complement genes in aninvertebrate deuterostome such as sea urchin [5] and in ascidia [6,7].In contrast, no complement gene was found in the genome ofDrosophila melanogaster [8] or Caenorhabditis elegans [9], suggestingthat the complement system was established in the deuterostomelineage. However, recent reports on the horseshoe crab C3, factor B(Bf) [10], coral C3 [11] and a sea anemone genome analysis indicatethat the complement system is of a much more ancient origin [12].Fish have well-developed complement systems that play importantroles in their innate immune responses. Bony and cartilaginous fish

complement systems show a close functional similarity to those ofmammals, with a few characteristic differences. In general, fishcomplement is more heat labile, has a lower optimal reactiontemperature (10e25 !C) [13,14] and has stronger antimicrobialaction. It is of great interest that C3 in teleost shows a high degree ofcomplexity that is not commonly seen in mammalian species [1].Complement is comprised of about 35 individual proteins [1]. Inmammals, activation of complement results in the generation ofactivated protein fragments that play a role in microbial killing,phagocytosis, inflammatory reactions, immune complex clearance,and antibody production [15]. Fish appear to possess activationpathways similar to those in mammals and the fish complementproteins identified thus far show many homologies to theirmammalian counterparts [16]. Because knowledge about comple-ment proteins, regulatory proteins, and complement receptors infishseems to be far to be complete, it is unclear whether all thecomplement functions that have been identified in mammals alsooccur in fish [17,18]. However, it has been clearly demonstrated thatfish complement can lyses foreign cells and opsonise foreign organ-isms for destruction by phagocytes [19]. There are also indicationsthat complement fragments participate in inflammatory reactions

* Corresponding author. Tel.: "34 93 5814127; fax: "34 93 5812390.E-mail address: Israel.Mauri@uab.cat (I. Mauri).

Contents lists available at ScienceDirect

Fish & Shellfish Immunology

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

1050-4648/$ e see front matter ! 2011 Elsevier Ltd. All rights reserved.doi:10.1016/j.fsi.2011.03.013

Fish & Shellfish Immunology 30 (2011) 1310e1322

[20]. Three pathways have been elucidated through which thecomplement cascade can be initiated: the classical, the alternativeand the lectin pathway [15,18]. All three pathways merge througha common intersection, the complement C3. The classical pathwaymediates specific antibody responses and it is initiatedby thebindingof antibodies to cell surface antigens. Subsequent binding of theantibody to complement C1q subunits of C1 result in catalyticallyactive C1s subunits. The two activated C1s subunits are then able tocatalyze the assembly of the C3 convertase (complement C4b2a)from complements C2 and C4 [21]. The alternative pathway does notrequire the action of antibodies to initiate the cascade, but is initiatedby foreign cell surface components. In the alternative pathwaycomplement, C3 undergoes spontaneous cleavage resulting incomplementBbinding toC3b.Diffusion of the Bf subunit results in anactive alternative pathwayC3convertase (C3bBb). C3bBb is stabilizedby binding to properdin prior to merging into the common pathwayandconversionofC3 [16]. The lectinpathway is similar to the classicalpathway. C1q is not involved in the lectinpathway and in contrast, anopsonin mannan binding protein (MBP) is involved in the initiationprocess [22e24]. Fish seem to possess multiple isoforms of severalcomplement proteins such as C3 and factor B. It has been hypothe-sized that the function of this diversity serves to expand the innateimmune recognition capacity and response [1]. For example, therainbow trout, a quasi-tetraploid species, contains at least 3 differentisoforms of C3 [25]. Multiple C3 isoforms have also been character-ized in the Gilthead seabream at protein level [26,27] while in thecommon carp up to eight C3 forms have been described [28]. Unlikemammals, it has been demonstrated that some species of teleostpossess multiple forms of functionally active C3 that may be theproducts of several genes and may vary in their ability to bind to

various surfaces. These various isoforms of C3 molecules have beenshown to differ in their function, as indicated by the differences intheir binding efficiencies to various complement-activating surfacessuch as erythrocytes, Escherichia coli or zymosan particles [26,27]. Inaddition on its major number of C3 isoforms, complement in fish isalso readily activated at low temperature. Moreover, the alternativepathway complement activity is 5-fold to 10-fold higher than inhigher vertebrates [14]. A diversified C3 repertoire, coupled witha higher titer and activity, may greatly expand the capacity ofcomplement as a form of innate immunity to defend fish againstmicroorganisms in the aquatic environment. Thismechanismconfersa survival advantage upon fish that have developed only limitedadaptive immunity [29]. Understanding the functions of complementinfish and the roles that the individual proteins, including thevariousisoforms, play in host defense is important not only for under-standing the evolution of this systembut also for the development ofnew strategies in fish health management.

2. Materials and methods

2.1. Fish and treatments

Adult Gilthead seabream (Sparus aurata) and adult Europeanseabass (Dicentrarchus labrax) of body weight of 30e40 g wereobtained from the Centro de Investigación y Formación Pesquera yAcuícola El Toruño (Puerto de Santa María, Cádiz, Spain). The fishwere acclimatized to laboratory conditions for 15 days beforeexperiments. Fishwere held in tankswith recirculatingwater undera photoperiod of 12 h light/12 h dark and natural conditions oftemperature (19e21 !C) and fed twice a day with a commercial diet(Skretting). For density experiments, fish were exposed to a crowd-ing stress by increasing the density of the tanks from 8e10 kg/m3

(hereon, low density conditions) to 50 kg/m3 (hereon, high densityconditions). After 15 days, fish were sampled (n # 10, using 0 h asa control) and challenged with Nodavirus (NNV) intraperitoneal(104 TCID50/ml) or with the bacteria Photobacterium damselaesubsp. piscicida, by bath immersion (105 cfu/ml, 1 h) [5], and then,were sampled (n # 10) at 24 h.

2.2. Tissue sampling

The fish were anaesthetized with 2-phenoxyethanol (Merck,France). Each fish was dissected to obtain the gills, the brain, theintestine and the liver. The samples were kept at $80 !C until use.All organs were used for the expression analysis and the liver wasalso used for C3 cloning.

2.3. RNA extraction and first-strand cDNA synthesis

Total RNA was isolated with TriReagent (Molecular ResearchInc.) following manufacturer’s instructions. RNA concentration wasquantified using a Nanodrop ND-1000 and RNA quality wasassessed using a Bioanalyzer 2100 with the RNA 6000 Nano Lab-Chip kit (Agilent Technologies). First-strand cDNA was synthesizedwith 2 mg of total RNA using SuperScript III reverse transcriptase(Invitrogen) and oligo-dT15 primer (Promega) in a 20 ml reactionvolume.

2.4. Complement C3 cloning

The cloning procedure for C3 was carried out analyzing the C3nucleotide and protein sequences available in public data bases(NCBI and TIGR). Alignments of the most representative C3sequences (Paralychtys olivaceus and Anarhichas minor) weredone using the ClustalX software [30] and specific primers were

Table 1Sequences of used primers.

Primer Sequence (50e30) Species

C3DorC3_Z1For TGGAGTGTCAAGACTGCACAGG BothDorC3_Z1Rev CAACAGCATATGGGTTGGTGAGG Both

C3 expressionSNiAnato_or_1f GGACGGTATGAGTGTGTTCTTC Sparus auratusSNiAnato_or_1r CAGGATCAGGGCTGTAGTTGT Sparus auratusSNiAnato_llob_1f ATGATGCTGGGCTGTTGTTC Dicentrarchus labraxSNiAnato_llob_1r GCACACATGTTCCTCCTCAA Dicentrarchus labrax

18S18S For CGAGCAATAACAGGTCTGTG Both18S Rev GGGCAGGGACTTAATCAA Both

30 RACEC3GSP3or1 CACCTGTTTGGGCATGGAGGAGCTG Sparus auratusC3GSP3llob1 TAACATGATCCACATGAC Dicentrarchus labraxAbriged anchor

primerGGCCACGCGTCGACTAGTACGGGIIGGGIIGGGIIG

Both

50 RACEC3GSP51_OR GTCTGGATGAAGATGTA Sparus auratusC3GSP52_OR GCTCCATGAGGGGTGTCACTGC Sparus auratusC3GSP51_LLOB CTAAGACGACTTTCTCTA Dicentrarchus labraxC3GSP52_LLOB GCTCCATACGGGGTGACAGTGCAA Dicentrarchus labraxAbriged anchor

primerGGCCACGCGTCGACTAGTACGGGIIGGGIIGGGIIG

Both

Universal anchorprimer

CUACUACUACUAGGCCACGCGTCGACTAGTAC

Both

C3 absolutequantification

Both isoformsQTOTllobF ACCAAAGAACTGGCAACCAC Dicentrarchus labraxQTOTllobR CTAGCAGTCGGTCAGGGAAC Dicentrarchus labraxAnato isoformAnatoBassF TGTTGTTTGAATGGCATAAAGG Dicentrarchus labraxAnatoBassR TCTCCTTACAACAATGCAGGAA Dicentrarchus labrax

I. Mauri et al. / Fish & Shellfish Immunology 30 (2011) 1310e1322 1311

designed in several conserved regions using the Primer 3 software[31]. An amplification, using C3 specific designed primers (Table 1),was performed using DorC3_Z1For and DorC3_Z7Rev, using 2 mlcDNA as template in a 50 ml PCR reactions with an Expand HighFidelity PCR System (Roche Applied Science), with a step of 94 !C2min, 30 cycles of 94 !C 15 s, 62 !C 30 s and 72 !C 2min, followed bya final step of 72 !C for 7 min. The PCR products were analyzed ina 0.8% agarose gel stained with ethidium bromide, excised andpurified using MinElute Gel Extraction Kit (Qiagen). PCR fragmentswere ligated into PCR-XL-TOPO (Invitrogen) and transformed inTOP10 E. coli competent cells. Plasmid DNA was isolated using theNucleospin Quickpure kit (Macherey-Nagel), digested with EcoRV(Promega) to verify the appropriate insert size and sequenced withM13Rev/M13For primers (Sistemas Genómicos, Spain). C3sequenceswere submitted to analytical sequencing, consisting in a 6times repetition of the sequencing procedure, with ABI PRISM BigDye Terminator Ready Reaction Cycle Sequencing 9700 (AppliedBiosciences) and were analyzed by DNA Analyzer 3730%. Two 30

RACE (30 Rapid Amplification of cDNA Ends, Invitrogen) reactionswere carried out, using gene specific primer C3GSP3or1 for Gilthead

seabream, and C3GSP3llob1 for European seabass, following themanufacturer’s instructions. Then, 50 RACE was performed usinggene specific primer 1 (C3GSP51_OR), and gene specific primer 2(C3GSP52_OR) for Gilthead seabream and gene specific primer 1(C3GSP51_LLOB) and gene specific primer 2 (C3GSP52_LLOB) forEuropean seabass. PCR products were cloned into pGEM-T EasyVector (Promega). Plasmid DNA was isolated using the NucleospinQuickpure kit, digested with EcoRI (Promega) and sequenced withSP6/T7 primers. RACE sequences were submitted to High Qualitysimple chain sequencing. All the primers used in this study weresynthesized at Sigma-Aldrich (Madrid, Spain) and are listed inTable 1.

2.5. Sequence analysis

The nucleotide and deduced amino acid sequences similaritieswere analyzed with BLAST using non-redundant nucleotidecollections database optimized for BLASTn algorithm at theNational Center for Biotechnology Information (NCBI) (http://blast.ncbi.nlm.nih.gov/Blast.cgi). The nucleotide and deduced amino acid

Fig. 1. A. Identity (%) between Gilthead seabream nucleotide sequence and representative C3 sequences using megablast algorithm. B. Identity (%) between European seabass_1nucleotide sequence and representative C3 sequences using megablast algorithm. C. Identity (%) between European seabass_2 nucleotide sequence and representative C3 sequencesusing megablast algorithm.

I. Mauri et al. / Fish & Shellfish Immunology 30 (2011) 1310e13221312

sequences of C3 cDNAs were analyzedwith ClustalX, Geneious 3.6.1and Mega 4 software, and their protein structures were predictedusing analysis SMART and PROSITE.

2.6. Multiple alignment and phylogenetic analysis

The deduced amino acid sequences fromGilthead seabream andEuropean seabass C3 cDNAs were compared with those from othervertebrate using ClustalX [30]. The selected species wereP. olivaceus (GenBank acc. no. AB021653.1), A. minor (GenBank accno. AJ309570.1), Oncorhynchus mykiss (GenBank acc. no. L24433.1),Gallus gallus (GenBank acc. no. NM205405.1), Austrelaps superbus(GenBank acc. no. DQ149984.1), Mus musculus (GenBank acc. no.BC043338.1) and Homo sapiens (GenBank acc. no. J04763.1). Theamino acid sequences for C3 complement component downloadedfrom the Genbank database were aligned with the obtained aminoacid sequences of Gilthead seabream and European seabass (gsbC3,esbC3_1 and esbC3_2) using the multiple sequence alignmentfeature implemented in the ClustalX [30] using default gap openand extension penalties. The alignment file was imported into thePHYML phylogenetic package [32]. A bootstrapped (1000 repli-cates) Maximum Likelihood Estimation (MLE) analysis was con-ducted on C3 sequences using the JTT substitutionmodel algorithm.

2.7. RT-PCR C3 expression analysis

2 mg of all extracted RNAs (pools of n # 6 individuals) werereverse transcribed as indicated previously. C3 was amplified (30cycles) as follows: 95 !C for 45 s, 60 !C for 45 s, 72 !C for 45 s,

followed by a final extension at 72 !C for 7 min. The PCR productswere analyzed by electrophoresis on a 1% agarose gel. The primersused in the expression analysis were SNiAnato-or-1f and SNiAnato-or-1r for Gilthead seabream and SNiAnato-llob-1f and SNiAnato-llob-1r for European seabass (see Table 1).

2.8. Absolute quantification by Q-PCR

Absolute Quantification by Q-PCR was used to determine therelative contribution of the European seabass C3 isoforms. Reactionswere conducted on a MyiQ Single Color Real-time PCR DetectionSystem (BioRad). For the specific amplification of the isoformsAnatoBassF and AnatoBassR primers were designed inside the ana-phylotoxin domain to amplify only the isoform with this domain(Table 1). The primers QTOTllobF and QTOTllobR (close to the ana-phylotoxin domain) were designed in order to amplify the totalnumber of transcripts of both isoforms. The copy number of thetranscript without anaphylotoxin was deduced subtractingthe anaphylotoxin isoform number of transcripts from the totalnumber of transcripts of both isoforms. In order to carry out absolutequantification, both C3 fragments were cloned into pGEM-T easyvector (Promega) and transformed into E. coli JM109 cells. Bacteriawere grown overnight in LBmediumwith ampicillin (50 mg/ml) andplasmids were purified using Nucleospin Plasmid Quickpure kit(Macherey-Nagel). Standard curves (Ct-Threshold cycle versus logcopy number) were constructed using the C3-pGEM purifiedplasmid. The plasmid copy number was determined using thefollowing equation: DNA (copy) # [6.02 % 1023 (copy/mol) % DNAamount (g)]/[DNA length (bp) % 660 (g/mol/bp)]. For the reaction,

Fig. 2. A. Domain graphical representation and identity (%) between Gilthead seabream amino acid sequence and representative C3 sequences using tblastx algorithm. B. Domainand graphical representation and identity (%) between European seabass_1 amino acid sequence and representative C3 sequences using tblastx algorithm. C. Domain and graphicalrepresentation and identity (%) between European seabass_2 amino acid sequence and representative C3 sequences using tblastx algorithm.

I. Mauri et al. / Fish & Shellfish Immunology 30 (2011) 1310e1322 1313

diluted plasmid (from 109 to 102 copies/ml) was mixed with 10 mMspecific Q-PCR primers, purewater and SYBRGreen (BioRad). Cyclingconditionswere: 5min initial denaturation at 95 !C, then40 cycles of10 s denaturation at 95 !C and 30 s binding and elongation atannealing temperature 60 !C, 1 min final elongation at 95 !C and1 min at 60 !C followed by a melt curve analysis, where thetemperature starts at 60 !C and increase 0.5 !C every 30 s until 95 !C.

3. Results

3.1. Full length cloning and identification of the C3 cDNA

We obtained by PCR amplification of cDNA from Gilthead seab-ream and European seabass two clones of 3601 bp and 3201 bprespectively. These amplifications were carried out using C3 specificprimersDorC3_Z1For andDorC3_Z7Rev (Table 1). Subsequent 50 and30 RACE on Gilthead seabream provided a set of clones that wereused to construct the final contig of 5594 bp named gsbC3 (acc. no:HM543456). In European seabass, after 50 and30 RACEwereobtainedtwo contigs. The first resulting contig, named esbC3_1, wascomposed of 5153 bp and showed the expected domain structure.The second contig composed of 4952 bp and named esbC3_2,showed the absence of anaphylotoxin domain (acc. no: HM563078and HM563079 respectively). One contig for Gilthead seabream(5594 bp) contained an opening reading frame (ORF) that started atthe position 46 and ended at position 5013 with 50 and 30 UTRs of45 bp and 581 bp, respectively. This contig encoded a predictedprotein of 1656 amino acids (Supplementary Fig. 1). Two contigswere obtained for European seabass composed of 5153 and 4952 bprespectively. The first contig (esbC3_1) with an ORF that started atthe position 1 and ended at position 4962 and the second contig

(esbC3_2) starting at position 1 and ending at position 4761 whichencodepredictedproteins of 1654and1587aminoacids respectively(Supplementary Figs. 2 and 3). The 30 UTRs of all three sequencesshowed the characteristic polyadenilation sequence (AATAAA)(Supplementary Figs. 1e3). The BLAST alignment of Gilthead seab-ream and European seabass C3 nucleotide sequences showed a highhomologywith awide number of fish species (Fig.1AeC). Regardingthe gsbC3, we obtained a high nucleotide identity and coveragewithA. minor and P. olivaceus and a low coverage with more distantspecies such as Pseudomonas flavescens (Fig. 1A). Regarding esbC3_1and esbC3_2 we observed a high nucleotide identity and coveragealso with A. minor and P. olivaceus (Fig. 1B and C).

3.2. Sequence analysis and multiple alignment

The analysis of the deduced protein sequences revealed theappearance of 8 different C3 specific domains in gsbC3 and esbC3_1sequences (Fig. 2A and B) and 7 domains in esbC3_2 sequence(Fig. 2C). Starting on the N-terminal region, the first observeddomain was a-2-macroglobulin domain, or MG2, the N-terminalregion of the a-2-macroglobulin family. The next domain was thea-2-macroglobulin_2 domain. After these domains we observedthe anaphylotoxin-like domain but this domain was not present intheesbC3_2. The followingdomainwasC-terminal regionof thea-2-macroglobulin family that belongs to the MEROPS proteinaseinhibitor family I39, clan IL. Thenwe observed the presence of a ‘baitregion’ and the presence of the thiolester region containing thehighly conserved motif GCGEQ. The next observed domain wasthe a-macroglobulin complement component. This domain coversthe complement component region of the a-2-macroglobulin familyrelated with complement components, and then the receptor-

Fig. 2. (continued).

I. Mauri et al. / Fish & Shellfish Immunology 30 (2011) 1310e13221314

binding domain (RBD) of a-2-macroglobulin proteins, located at theC-terminus with a protein binding molecular function. Finally, wefound the netrin domains UNC-6 and the C345C, homologous tocomplement protein family regions.

ThealignmentofC3predictedproteins showedahighand relevanthomology (Fig. 3).While theb-aprocessing signal (RRRR/RRKR)or thecleavage site (RS), are quite conserved the thiolester site (GCGEQ) thatis well conserved in the majority of the vertebrate species, appearsmodified in European seabass. In this case, a cysteine substitution byan aspartate was detected. Furthermore, the thiolester sites in gsbC3,and esbC3_1 and esbC3_2 are surrounded by well-conserved hydro-phobic amino acids as in other C3 and C4 proteins. Previous studiessuggested that Pro1007 and Pro1020 are necessary for stable thiolesterformation and His1126 determines the thiolester binding specificity[33]. Our analysis showed that Pro1007 is conserved and Pro1020 isreplaced by Leu in according to previous studies in teleost [34]. Thisconservation pattern was also observed in two other important resi-dues: His1126, also involved in thiolester bond formation and Glu1128,that was involved in binding specificity in humans [35].

3.3. Phylogenetic analysis

The deduced amino acid sequences of C3 cDNAs were comparedwith those from other vertebrates using ClustalX [30]. Our resultsshowed that C3was homologous to other C3 sequences: P. olivaceus,

A. minor, O. mykiss, G. gallus, A. superbus,M. musculus and H. sapiens(see GenBank acc. no in Materials and methods section). Thealignment (Fig. 3) showed a highly conserved pattern. On one handthe experimental species C3 sequences clustered together near ofthemost evolved fish P. olivaceus and far from the older groups suchas A. minor or O. mykiss. On the other hand the remote group withbirds (G. gallus) and reptilians (A. superbus) were clustered together,and the sameoccurredwith themammalian group (M.musculus andH. sapiens) distant from the whole group of fish (Fig. 4).

3.4. C3 expression analysis in Gilthead seabream and Europeanseabass C3 mRNA

The RT-PCR analysis showed that Gilthead seabream C3 wasexpressed mainly in liver but extrahepatic expression was alsoobserved (Fig. 5A). At 0 h under low density conditions we detectedhigh expression levels in liver and a weak expression in intestineand brain. Under high density conditions, we observed C3 expres-sion in all the tissues with the highest expression in the intestine. At24 h under bacterial challenge at both densities, we observed themain expression in liver. However, in intestine, a weak expressionwas detected under low density conditions in contrast to a higherexpression detected under high density conditions. At 24 h in thevirus challenged group, we observed that the expression wasmainly found in liver but weak expression was also detected in

Fig. 2. (continued).

Fig. 3. AeC. Multiple alignment of the generated Gilthead seabream and European seabass C3 molecules with other vertebrate selected C3 molecules. The conservation is rep-resented by the bar graph under the alignment. The coding regions are described over the alignment according to the general domain structure of the complement family. The firstdomain (A2M_N) was marked with ;. The second domain (A2M_N_2) was marked with +. The Cleavage Site (RRKR) was marked using a vertical line. The ANATO domain washighlighted with A. The C3 convertase site was marked with :. The fourth domain (A2M) was marked using C. The fifth domain (A2M_complement domain) was marked with .The H binding site was marked with . The sixth domain (Properdin Binding Site) was marked using . The conserved residues, the thiolester site and Factor I are boxed witha black box.

I. Mauri et al. / Fish & Shellfish Immunology 30 (2011) 1310e1322 1315

I. Mauri et al. / Fish & Shellfish Immunology 30 (2011) 1310e13221316

Fig. 3. (continued).

intestine and gills. In contrast, under high density conditions wefound a total inhibition of the C3 expression.

Regarding European seabass, C3 was found to be expressed in allexamined tissues (Fig. 5B). Two fragments, corresponding to eachisoform, were amplified. The largest fragment (698 bp), corre-sponding to esbC3_1, with the anaphylotoxin domain and the otherfragment (500 bp) corresponding to esbC3_2, without the anaphy-lotoxindomain. At 0 hunder lowdensityconditionsweobserved thehighest expression in the liver and in the intestine for both isoforms.The isoform without anaphylotoxin was the most abundant. Thebrain only expressed the isoformwithout anaphylotoxin domain. Athigh density this expression pattern changed and the abundance ofboth isoforms decreased in liver and in intestine. In contrast theexpression of both isoforms in brain was increased. At 24 h underbacterial challenge in the low density groups, we observed the samedistribution but the intestine expression decreased and the brainexpression increased. At high density, a decrease of the expression ofboth isoforms was observed, and a weak expression of the isoformwithout anaphylotoxin domain in gills was increased. Under viruschallenge, at 24 h, in the low density group, we observed a clearexpression in gills, liver and brain with a poor expression of bothisoforms in intestine. However, at high density conditions, theexpression of both isoforms in gills and intestine disappearedcompletely. Moreover, the expression of the anaphylotoxin isoformin liver, decreased and both isoforms decreased in brain.

3.5. Differential C3 isoform expression

In European seabass, we have detected by RT-PCR two differenttranscripts differing in the presence/absence of anaphylotoxindomain. These isoformswerequantified showingdifferentexpression

levels of each isoform depending on the experimental conditions(Fig. 6). In terms of isoformRNA abundancewe could observe at 0 h atlow density, a higher percentage of the esbC3_1 (isoform with ana-phylotoxin domain) in contrast to the esbC3_2 (isoform withoutanaphylotoxin domain). This patternwas inverted under high densityconditions. Under bacterial challenge at 24 h, the low density groupshowed a decrease in the percentage of esbC3_1 transcript (50%)compared with the control situation (0 h at low density). In contrast,the opposite pattern was observed at high density conditions underbacterial challenge. Under viral challenge, we observed a minorcontribution of esbC3_1 with respect to control situation, where anincreaseof its expressionwasobservedunderhighdensityconditions.

4. Discussion

Here we reported the cloning and the expression analysis ofGilthead seabream and European seabass complement C3 underdiverse husbandry situations. Until date no C3 DNA sequences forGilthead seabream and European seabass were available. Differentstudies had been carried out in salmonids [36e39] showing thatthe complement C3 is one of the best studied molecules [40]. Thereported C3 sequences were aligned with all the available proteinsequences and the alignment revealed the existence of 8 well-conserved domains in both species. The first domain was the MG2domain, similar to N-terminal region of a-2-macroglobulin family,associated to a molecular endopeptidase-inhibitor activity func-tion. The second domain was the a-2-macroglobulin_2 domain,with the ability to inhibit proteases from all catalytic classes. Thisinhibition is carried out by steric hindrances thatinduce confor-mational changes and affect the functionality of the protein [41]. Inthe third position, it has been found the anaphylotoxin-like domain

Fig. 3. (continued).

I. Mauri et al. / Fish & Shellfish Immunology 30 (2011) 1310e13221318

only found in the Gilthead seabream C3 molecule and in theEuropean seabass isoform 1 (esbC3_1). These protein fragments,homologous to a three-fold repeat present in fibulins [42], aregenerated enzymatically in serum during activation of complementmolecules C3, C4, and C5 and it has been described that they caninduce smooth muscle contraction [43]. The structure of the C3aanaphylotoxin module is a compact a-helical fold stabilized bythree disulphide bridges in the pattern Cys1e4, Cys2e5 andCys3e6. Interestingly, this important domain was not found in thesecond isoform of European seabass. Previous studies on carp,showed the existence of an isoformwithout anaphylotoxin domainand more similar to human C3 rather than to carp C3-Q2 isoform[28] indicating a possible different functional role of this isoform.After this, we observed the a-2-macroglobulin domain. The C-terminal region of the a-2-macroglobulin family [44e47] is alsoassociated to an endopeptidase-inhibitory function. After the a-2-macroglobulin domain, we found the presence of a ‘bait’ anda thiolester region indicating a similar protease inhibitor mecha-nism involved in the inactivation of the inhibitory capacity byreaction of the thiolester with small primary amines [48]. Thisa-macro-globulin thiolester bond-forming region is a short andhighly conserved region of proteinase-binding a-macro-globulinsand it contains the necessary cysteine and a glutamine for thethiolester bond formation, which is cleaved at the moment ofproteinase-binding. It mediates the covalent binding of thea-macro-globulin to the proteinase. The GCGEQmotif that is highlyconserved across the evolution was also present in both seabreamand seabass C3. However, a substitution of the cysteine (C) by anaspartate (D) in both C3 mRNA transcripts in European seabassseems to have no functional implications. It has been reported in

carp that this substitution could revert in a non-functional isoform[49], but previous results of our group demonstrate the full func-tionality of the complement hemolytic activity. Those conservedregions were mainly located near a potential post-translationalprocessing site (in the b-chain domain) and in the thiolester site(GCGEQ) along with important hydrophobic amino acids (in the a-chain domain) involved in the stable thiolester formation and in thebinding specificity. The differences concerning the presence of thethiolester site in European seabass lead us to hypothesize a possibledifference in terms of fine modulation. It has been demonstratedthat in mammals, individuals without a functional site, hada decreased response in terms of hemolytic activity but not in termsof protein stability [50]. In fish, this site is well conserved in themajority of the described species. However, several isoforms ofCyprinus carpio [28], or other components like C5a of Ginglymos-toma cirratum [51], have a serine-substitution maintaining thefunctionality. In conclusion, the amino acid substitution in thethiolester regions and the absence of the anaphylotoxin domain inthe European seabass isoform_2 did not seem to alter the inflam-matory responses or the functionality of the whole system.Previous studies have reported a powerful response of the C5afragment [52,53] and it could explain the normal response of thecomplement system. It has been reported that some teleost canhave multiple isoforms [25,27] and it has been hypothesized thatthis diversity in complement proteins would serve to expand theirinnate immune recognition repertoires [54]. By absolute quantifi-cation we have demonstrated that esbC3_2 was expressed in alltested conditions, suggesting a principal role of it in the comple-ment response to pathogens. That fact reveals the possibledependence of a C5a stronger response [53] mainly involved in

Fig. 4. Phylogenetic relationships of C3 selected species. Maximum likelihood phylogenetic tree generated from a CLUSTALW alignment and MEGA 4.0 program. The bootstrap with1000 replicates values was marked. The bar indicated the distance. The protein IDs are as follows: Sparus_aurata, HM543456; Dicentrarchus_labrax (noanato), HM563078;Dicentrarchus_labrax (anato), HM563079; Paralichthys_olivaceus, AJ309570.1; Anarhichas_minor, AB021653; Oncorhynchus_mykiss, L24433.1; Gallus_gallus, NM205405.1; Aus-trelaps_superbus, DQ149984.1; Mus_musculus, BC043338.1; Homo_sapiens, J04763.1.

I. Mauri et al. / Fish & Shellfish Immunology 30 (2011) 1310e1322 1319

leukocyte chemotaxis and in triggering the respiratory burst ofleukocytes [39]. In addition, we have found an isoform specificresponse depending on the tissue and pathogen, showing up therole of the complement in terms of acquired immunity [55]. Thepredicted protein also showed the presence of the A-macroglobulincomplement component (the complement component region ofthe a-2-macroglobulin family) and the receptor-binding domain(RBD) of a-2-macroglobulin proteins, located at the C-terminus.Finally, netrin domains UNC-6, and C345C, which belong to thecomplement protein family members, such as C3, C4, C5. Interest-ingly, these domains are homologous with the N-terminal domainsof tissue inhibitors of metalloproteinases (TIMPs) [56,57]. Thesedomains are present in a number of other proteins and are alsofound in cobra venom factor and in complement factors C3, C4 andC5. Its main function correspond to control Central Nervous System(CNS) motor axons [57]. The mechanism which led to the genera-tion of multiple C3 isoforms remains unknown. In Gilthead seab-ream has been described up to 5 protein isoforms in plasma [27]but our study showed the existence of a unique transcript in liver. InEuropean seabass, two different transcripts have been detected inliver but there are no available studies on serum isoformsdescription. Previous studies have shown the presence of tissular

Fig. 5. A. RT-PCR analysis of C3 mRNA expression in Gilthead seabream tissues (gills, liver, intestine and brain). Expression of 18S in each sample (pool n # 6) was used asa housekeeping gene. The size of amplified product was 695 bp for C3 and 212 bp for 18S. B. RT-PCR analysis of C3 mRNA expression in European seabass tissues (gill, liver, intestineand brain). Expression of 18S in each sample (pool n # 6) was added for normalization. The size of amplified products was 698 bp for C3_1, 500 bp for C3_2 and 212 bp for 18S.

Fig. 6. Absolute Quantification by Q-PCR of C3 mRNA expression in European Seabassliver. The results are expressed as the percentage of each isoform esbC3_1 (A, ana-phylotoxin) and esbC3_2 (NA, no anaphylotoxin). The letters on the bottom indicategroups of sampling: 0 h low # 0 h under low density (10 kg/m3), 0 h high # 0 h underhigh density (50 kg/m3), 24 h B low # 24 h under low density (10 kg/m3) plus bacterialchallenge, 24 h B high # 24 h under high density (50 kg/m3) plus bacterial challenge,24 h V low# 24 h under low density (10 kg/m3) plus viral challenge, 24 h V high # 24 hunder high density (50 kg/m3) plus viral challenge.

I. Mauri et al. / Fish & Shellfish Immunology 30 (2011) 1310e13221320

specific isoforms [58]. These isoforms may come from [59] post-transcriptional modifications at pre mRNA level or alternativesplicing in order to generate the diversified repertoire of proteinisoforms in some fish species [58,60] according to the tissue werethey are located. In this sense, we have cloned the hepatic C3, butfurther work must be carried out in order to find the extrahepaticsequences of C3. Moreover, the quantification of European seabassisoforms showed a specific response confirming the possible role interms of specific immunity of the C3 different isoforms. On onehand, a stronger esbC3_2 expression has been observed in front tothe viral challenge, suggesting a specific participation of thismolecule in front to a viral challenge. This observation is inagreement with previous studies in trout macrophages in our lab,demonstrating that the immune response to viral infections wasstronger than to bacterial infections, suggesting a esbC3_2 specificresponse to virus rather than to bacteria [61]. On the other hand,the esbC3_1 expression appeared stronger under bacterial chal-lenge, especially under high density conditions. This fact suggestsagain the hypothesis of a differential role of C3 isoforms dependingon the challenges applied to fish. Fig. 6 shows that viral treatmentor high density plus infection would differentially upregulateesbC3_2 isoform whereas the esbC3_1 isoform would be upregu-lated under non-specific challenges/stressors, since either highdensity or bacterial exposure induce differentially its expression. Inprevious works such stress-induced complement activation wasalready observed [62,63], and it was suggested that the immune-endocrine connection could play a role, and in particular hormonessuch as cortisol. In the present work, although cortisol levels couldnot be analyzed in all experimental animals, the available datareveals that both high density and challenge increased plasmacortisol by a rate of 20e40 fold the basal levels, which is in accor-dance with previous results on in vitro and in vivo stress challengesin fish species [64e67]. Therefore, as the present results indicatedifferential C3 isoform expression depending on the challenge, theresearch on the role of these functional C3 isoforms in fish sub-jected to stressors or challenges deserves further research.

Acknowledgments

This work has been supported by the Imaquanim project(Immunity of Aquatic Animals. UE 6th framework program) andBFU2009-07354 (Ministry of Science and Innovations) and SGR2009-554 (Generalitat de Catalunya). IM is an Imaquanim research asso-ciate, NR is a Ramon y Cajal Fellow (Ministry of Science andInnovation). LT, IM and NR belong to the Xarxa de Referència enAqüicultura de Catalunya.

Appendix. Supplementary data

The Supplementary data associated with this article can befound online, at doi:10.1016/j.fsi.2011.03.013.

References

[1] Oriol Sunyer J, Zarkadis IK, Lambris JD. Complement diversity: a mechanismfor generating immune diversity? Immunology Today 1998;19:519e23.

[2] Lange S, Magnadóttir B. Spontaneous haemolytic activity of Atlantic halibut(Hippoglossus hippoglossus L.) and sea bass (Dicentrarchus labrax) serum.Comparative Biochemistry and Physiology Part B: Biochemistry and MolecularBiology 2003;136:99e106.

[3] Ricklin D, Hajishengallis G, Yang K, Lambris JD. Complement: a key system forimmune surveillance and homeostasis. Nature Immunology 2010;11:785e97.

[4] Nonaka M, Yoshizaki F. Evolution of the complement system. MolecularImmunology 2004;40:897e902.

[5] Al-Sharif W, Sunyer O, Lambris J, Smith LC. HP1 Characterization of the cDNAencoding a protein homologous to the mammalian complement componentC3 which is expressed in sea urchin coelomocytes. Developmental andComparative Immunology; 21:147.

[6] Ji X, Azumi K, Nonaka M, Namikawa-Yamada C, Sasaki M, Saiga H, et al.Opsonic complement C3 in the solitary ascidian, Halocynthia roretzi. MolecularImmunology 1998;35:363.

[7] Nonaka M, Azumi K. Opsonic complement system of the solitary ascidian,Halocynthia roretzi. Developmental and Comparative Immunology 1999;23:421e7.

[8] Adams MD, Celniker SE, Holt RA, Evans CA, Gocayne JD, Amanatides PG, et al.The genome sequence of Drosophila melanogaster. Science 2000;287:2185e95.

[9] Consortium TCeS. Genome sequence of the nematode C. elegans: a platformfor investigating biology. Science 1998;282:2012e8.

[10] Zhu Y, Thangamani S, Ho B, Ding JL. The ancient origin of the complementsystem. The EMBO Journal 2005;24:382e94.

[11] Dishaw L, Smith S, Bigger C. Characterization of a C3-like cDNA in a coral:phylogenetic implications. Immunogenetics 2005;57:535e48.

[12] Nonaka M, Kimura A. Genomic view of the evolution of the complementsystem. Immunogenetics 2006;58:701e13.

[13] Sakai DK. Repertoire of complement in immunological defense mechanisms offish. Annual Review of Fish Diseases 1992;2:223e47.

[14] Oriol Sunyer J, Tort L. Natural hemolytic and bactericidal activities of seabream Sparus aurata serum are effected by the alternative complementpathway. Veterinary Immunology and Immunopathology 1995;45:333e45.

[15] Holland MCH, Lambris JD. The complement system in teleosts. Fish & ShellfishImmunology 2002;12:399e420.

[16] Fujita T, Endo Y, Nonaka M. Primitive complement systemerecognition andactivation. Molecular Immunology 2004;41:103e11.

[17] Kimura A, Ikeo K, Nonaka M. Evolutionary origin of the vertebrate bloodcomplement and coagulation systems inferred from liver EST analysis oflamprey. Developmental and Comparative Immunology 2009;33:77e87.

[18] Zarkadis IK, Mastellos D, Lambris JD. Phylogenetic aspects of the complementsystem. Developmental and Comparative Immunology 2001;25:745e62.

[19] Rooijakkers SHM, van Strijp JAG. Bacterial complement evasion. MolecularImmunology 2007;44:23e32.

[20] Raftos DA, Robbins J, Newton RA, Nair SV. A complement component C3a-likepeptide stimulates chemotaxis by hemocytes from an invertebrate chorda-teethe tunicate, Pyura stolonifera. Comparative Biochemistry and Physiology -Part A: Molecular & Integrative Physiology 2003;134:377e86.

[21] Boshra H, Li J, Sunyer JO. Recent advances on the complement system ofteleost fish. Fish & Shellfish Immunology 2006;20:239e62.

[22] Fujita T. Evolution of the lectin-complement pathway and its role in innateimmunity. Nature Reviews Immunology 2002;2:346e53.

[23] Endo Y, Takahashi M, Fujita T. Lectin complement system and patternrecognition. Immunobiology 2006;211:283e93.

[24] Matsushita M, Fujita T. The lectin pathway. Research in Immunology 1996;147:115e8.

[25] Sunyer JO, Zarkadis IK, Sahu A, Lambris JD. Multiple forms of complement C3in trout that differ in binding to complement activators. Proceedings of theNational Academy of Sciences of the United States of America 1996;93:8546e51.

[26] Sunyer J, Tort L, Lambris J. Structural C3 diversity in fish: characterization offive forms of C3 in the diploid fish Sparus aurata. Journal of Immunology 1997;158:2813e21.

[27] Sunyer JO, Tort L, Lambris JD. Diversity of the third form of complement, C3, infish: functional characterization of five forms of C3 in the diploid fish Sparusaurata. The Biochemical Journal 1997;326(Pt 3):877e81.

[28] Nakao M, Mutsuro J, Obo R, Fujiki K, Nonaka M, Yano T. Molecular cloning andprotein analysis of divergent forms of the complement component C3 froma bony fish, the common carp (Cyprinus carpio): presence of variants lackingthe catalytic histidine. European Journal of Immunology 2000;30:858e66.

[29] Song W-C, Sarrias MR, Lambris JD. Complement and innate immunity.Immunopharmacology 2000;49:187e98.

[30] Thompson J, Gibson T, Plewniak F, Jeanmougin F, Higgins D. The CLUSTAL_Xwindows interface: flexible strategies for multiple sequence alignment aidedby quality analysis tools. Nucleic Acids Research 1997;25:4876e82.

[31] Rozen S, Skaletsky H. Primer3 on the WWW for general users and for biologistprogrammers. Methods in Molecular Biology (Clifton, NJ) 2000;132:365e86.

[32] Guindon S, Lethiec F, Duroux P, Gascuel O. PHYML Onlineea web server forfast maximum likelihood-based phylogenetic inference. Nucleic AcidsResearch 2005;33:W557e9.

[33] Isaac L, Isenman DE. Structural requirements for thioester bond formation inhuman complement component C3. Reassessment of the role of thioesterbond integrity on the conformation of C3. Journal of Biological Chemistry1992;267:10062e9.

[34] Zarkadis IK, Sarrias MR, Sfyroera G, Sunyer JO, Lambris JD. Cloning andstructure of three rainbow trout C3 molecules: a plausible explanation fortheir functional diversity. Developmental and Comparative Immunology2001;25:11e24.

[35] Sunyer JO, Lambris JD. Evolution and diversity of the complement system ofpoikilothermic vertebrates. Immunological Reviews 1998;166:39e57.

[36] Mikrou A, Marioli D, Papanastasiou AD, Zarkadis IK. CR3 complementreceptor: cloning and characterization in rainbow trout. Fish & ShellfishImmunology 2009;26:19e28.

[37] Chondrou M, Papanastasiou AD, Spyroulias GA, Zarkadis IK. Three isoforms ofcomplement properdin factor P in trout: cloning, expression, gene organiza-tion and constrained modeling. Developmental and Comparative Immunology2008;32:1454e66.

I. Mauri et al. / Fish & Shellfish Immunology 30 (2011) 1310e1322 1321

[38] Chondrou MP, Mastellos D, Zarkadis IK. cDNA cloning and phylogeneticanalysis of the sixth complement component in rainbow trout. MolecularImmunology 2006;43:1080e7.

[39] Sunyer JO, Boshra H, Li J. Evolution of anaphylatoxins, their diversity andnovel roles in innate immunity: insights from the study of fish complement.Veterinary Immunology and Immunopathology 2005;108:77e89.

[40] Lambris JD. The multifunctional role of C3, the third component of comple-ment. Immunology Today 1988;9:387e93.

[41] Enghild JJ, Salvesen G, Thøgersen IB, Pizzo SV. Proteinase binding and inhi-bition by the monomeric alpha-macroglobulin rat alpha 1-inhibitor-3. Journalof Biological Chemistry 1989;264:11428e35.

[42] Pan TC, Sasaki T, Zhang RZ, Fässler R, Timpl R, Chu ML. Structure andexpression of fibulin-2, a novel extracellular matrix protein with multipleEGF-like repeats and consensus motifs for calcium binding. The Journal of CellBiology 1993;123:1269e77.

[43] Rotllant J, Parra D, Peters R, Boshra H, Sunyer JO. Generation, purification andfunctional characterization of three C3a anaphylatoxins in rainbow trout: rolein leukocyte chemotaxis and respiratory burst. Developmental and Compar-ative Immunology 2004;28:815e28.

[44] Huang W, Dolmer K, Liao X, Gettins PGW. NMR Solution structure of thereceptor binding domain of human a2-macroglobulin. Journal of BiologicalChemistry 2000;275:1089e94.

[45]. Xiao T, Decamp DL, Sprang SR. Structure of a rat alpha-macroglobulinreceptor-binding domain dimer. Protein Science 2000;9:1889e97.

[46] Szakonyi G, Guthridge JM, Li D, Young K, Holers VM, Chen XS. Structure ofcomplement receptor 2 in complex with its C3d Ligand. Science 2001;292:1725e8.

[47] Zanotti G, Bassetto A, Battistutta R, Folli C, Arcidiaco P, Stoppini M, et al.Structure at 1.44 Å resolution of an N-terminally truncated form of the ratserum complement C3d fragment. Biochimica et Biophysica Acta (BBA) -Protein Structure and Molecular Enzymology 2000;1478:232e8.

[48] Janssen BJC, Huizinga EG, Raaijmakers HCA, Roos A, Daha MR, Nilsson-Ekdahl K, et al. Structures of complement component C3 p.ovide insights intothe function and evolution of immunity. Nature 2005;437:505e11.

[49] Nakao M, Mutsuro J, Nakahara M, Kato Y, Yano T. Expansion of genes encodingcomplement components in bony fish: biological implications of the comple-mentdiversity. Developmental andComparative Immunology2003;27:749e62.

[50] Lamping N, Schumann RR, Burger R. Detection of two variants of complementcomponent C3 in C3-deficient guinea pigs distinguished by the absence andpresence of a thiolester. Molecular Immunology 2000;37:333e41.

[51] Graham M, Shin D-H, Smith SL. Molecular and expression analysis ofcomplement component C5 in the nurse shark (Ginglymostoma cirratum) andits predicted functional role. Fish & Shellfish Immunology 2009;27:40e9.

[52] Klos A, Tenner AJ, Johswich K-O, Ager RR, Reis ES, Köhl J. The role of the ana-phylatoxins in health and disease. Molecular Immunology 2009;46:2753e66.

[53] Köhl J. Anaphylatoxins and infectious and non-infectious inflammatorydiseases. Molecular Immunology 2001;38:175e87.

[54] Nakao M, Matsumoto M, Nakazawa M, Fujiki K, Yano T. Diversity ofcomplement factor B/C2 in the common carp (Cyprinus carpio): three isotypesof B/C2-A expressed in different tissues. Developmental and ComparativeImmunology 2002;26:533e41.

[55] Dodds AW, Matsushita M. The phylogeny of the complement system and theorigins of the classical pathway. Immunobiology 2007;212:233e43.

[56] Ishii N, Wadsworth WG, Stern BD, Culotti JG, Hedgecock EM. UNC-6, a lam-inin-related protein, guides cell and pioneer axon migrations in C. elegans.Neuron 1992;9:873e81.

[57] Bányai L, Patthy L. The NTR module: domains of netrins, secreted frizzledrelated proteins, and type I procollagen C-proteinase enhancer protein arehomologous with tissue inhibitors of metalloproteases. Protein Science 1999;8:1636e42.

[58] Dooley H, Flajnik MF. Antibody repertoire development in cartilaginous fish.Developmental and Comparative Immunology 2006;30:43e56.

[59] Wilson MR, Warr GW. Fish immunoglobulins and the genes that encode them.Annual Review of Fish Diseases 1992;2:201e21.

[60] Bengtén E, Clem LW, Miller NW, Warr GW, Wilson M. Channel catfishimmunoglobulins: repertoire and expression. Developmental and Compara-tive Immunology 2006;30:77e92.

[61] MacKenzie S, Balasch J, Novoa B, Ribas L, Roher N, Krasnov A, et al. Compar-ative analysis of the acute response of the trout, O. mykiss, head kidney to invivo challenge with virulent and attenuated infectious hematopoietic necrosisvirus and LPS-induced inflammation. BMC Genomics 2008;9:141.

[62] Tort L, Sunyer JO, Gómez E, Molinero A. Crowding stress induces changes inserum haemolytic and agglutinating activity in the gilthead sea bream Sparusaurata. Veterinary Immunology and Immunopathology 1996;51:179e88.

[63] Mauri I, Romero A, Acerete L, MacKenzie S, Roher N, Callol A, et al. Changes incomplement responses in Gilthead seabream (Sparus aurata) and Europeanseabass (Dicentrarchus labrax) under crowding stress, plus viral and bacterialchallenges. Fish & Shellfish Immunology 2011;30:182e8.

[64] Rotllant J, Ruane NM, Caballero MJ, Montero D, Tort L. Response to confine-ment in sea bass (Dicentrarchus labrax) is characterised by an increasedbiosynthetic capacity of interrenal tissue with no effect on ACTH sensitivity.Comparative Biochemistry and Physiology - Part A: Molecular & IntegrativePhysiology 2003;136:613e20.

[65] Acerete L, Balasch JC, Castellana B, Redruello B, Roher N, Canario AV, et al.Cloning of the glucocorticoid receptor (GR) in gilthead seabream (Sparusaurata): differential expression of GR and immune genes in gilthead seabreamafter an immune challenge. Comparative Biochemistry and Physiology Part B:Biochemistry & Molecular Biology 2007;148:32e43.

[66] Castillo J, Teles M, Mackenzie S, Tort L. Stress-related hormones modulatecytokine expression in the head kidney of gilthead seabream (Sparus aurata).Fish & Shellfish Immunology 2009;27:493e9.

[67] Castro R, Zou J, Secombes CJ, Martin SAM. Cortisol modulates the induction ofinflammatory gene expression in a rainbow trout macrophage cell line. Fish &Shellfish Immunology 2011;30:215e23.

I. Mauri et al. / Fish & Shellfish Immunology 30 (2011) 1310e13221322