This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Genomic characterization and gene expression analysis of four hepcidin genesin the redbanded seabream (Pagrus auriga)

Beatriz Martin-Antonio, Rosa Ma Jimenez-Cantizano, Emilio Salas-Leiton, Carlos Infante,Manuel Manchado*

IFAPA Centro El Toruno, CICE, Junta de Andalucıa, Camino Tiro de Pichon s/n, 11500 El Puerto de Santa Marıa, Cadiz, Spain

a r t i c l e i n f o

Article history:Received 30 October 2008Received in revised form21 January 2009Accepted 21 January 2009Available online 4 February 2009

keywords:HepcidinRedbanded seabreamInnate immune systemAntimicrobial peptide

a b s t r a c t

Hepcidin antimicrobial peptides (HAMPs) are key molecules of the innate immune system againstbacterial infections and in iron metabolism. In this study we report the molecular cloning and genomiccharacterization of four HAMP genes (referred to as HAMP1, HAMP2, HAMP3 and HAMP4) in the red-banded seabream (Pagrus auriga). All these genes possessed the eight characteristic cysteine residuesinvolved in protein folding. No canonical sequence for convertase-mediated processing of the HAMP3propeptide was identified. At the genomic level, all four HAMP genes consisted of two introns and threeexons. Phylogenetic analysis revealed that HAMPs could group in two main clusters with HAMP2, HAMP3and HAMP4 belonging to the more complex and diversified HAMP2-like group of acanthopterygians.Quantitation of mRNA levels in adult tissues showed that HAMP1 was ubiquitously expressed, HAMP2mainly in kidney, spleen and intestine, whereas HAMP3 and HAMP4 in liver. During development, HAMP2and HAMP3 were expressed at a high level in embryos. Moreover, the expression levels of the four HAMPgenes increased between 5 and 15 days after hatching when larvae started external feeding. Inductionexperiments with lipopolysaccharide revealed significant changes in gene expression of the four HAMPgenes in kidney, liver and spleen. However, expression profiles differed in magnitude and time courseresponse. HAMP1 mRNAs increased rapidly in kidney at 1 h p.i. whereas HAMP2 did later at 24 h.Moreover, HAMP4 transcripts increased more than 5000-fold in liver whereas HAMP2 mRNAs droppedsignificantly in spleen at 3 h p.i. All these data suggest that HAMPs are involved in the response againstbacterial infections although additional functions in iron regulation and embryogenesis in fish should beconsidered.

! 2009 Elsevier Ltd. All rights reserved.

1. Introduction

Hepcidin antimicrobial peptides (HAMPs), also termed LEAP-1,are key molecules of the innate immune system. They have beenidentified only in vertebrates including mammals, birds, amphib-ians and fish [1–3]. They comprise approximately 20–25 aminoacids (aa) in length with six to eight evolutionary highly conservedcysteines [3–5] although a novel type with only four residues hasbeen recently reported [6]. In fish, they can act both as antimicrobialpeptides and as iron-regulatory hormones. In vitro inhibitory assayshave shown that HAMPs possess antimicrobial activity againsta wide spectrum of Gram-negative and Gram-positive bacteria[5,7–9]. In addition, a significant induction of HAMP gene expres-sion after bacterial challenge, lipopolysaccharide (LPS) injection or

vaccination has been widely documented (reviewed in ref. [2]).With respect to iron homeostasis, HAMPs promotemacrophage ironretention and diminish intestinal iron absorption. To exert thisfunction, HAMP transcript levels increase in iron-overloaded fish[8,10,11] and reduce under severe anemia [12]. This regulation ofHAMP gene expression has been associated to a defensive mecha-nism to limit the iron available for bacterial growth [13].

HAMPs comprise a multigene family in fish. Their genomicorganization is quite conserved with three exons and two introns.However, phylogenetic analyses have shown that HAMPs can bedivided into two paralogous lineages referred to as HAMP1-like andHAMP2-like [3]. HAMP1-like group is present in all fish species, andis considered as the orthologue of the mammalian sequences. Incontrast, HAMP2-like group is detected only in acanthopterygiansrepresenting amore complex and diversified group that might haveevolved under positive Darwinian selection [14]. For instance, fivesequences with two HAMP2-like types (WF2 and WF3a) weredescribed in Pseudopleuronectes americanus [4]. Three genes with

* Corresponding author. Tel.: !34 95 601 1315; fax: !34 95 601 1324.(M. Manchado).

E-mail address: manuel.manchado@juntadeandalucia.es (M. Manchado).

Contents lists available at ScienceDirect

Fish & Shellfish Immunology

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

1050-4648/$ – see front matter ! 2009 Elsevier Ltd. All rights reserved.doi:10.1016/j.fsi.2009.01.012

Fish & Shellfish Immunology 26 (2009) 483–491

Author's personal copy

two HAMP2-like sequences (TH1–5 and TH2–2) were reported inOreochromis mossambicus [8]. Finally, seven HAMP2-like cDNAshave been identified in Acanthopagrus schlegelii [15]. All these genesshowed different expression patterns in adult tissues and duringlarval development [4,8]. In all cases, HAMP1- and HAMP2-likegenes activated their gene expression after bacterial or LPS treat-ments [4,5,8,15].

The redbanded seabream Pagrus auriga (Valenciennes, 1843) isa member of the Sparidae family that inhabits the Eastern Atlanticfrom Portugal to Angola, including the south-western Mediterra-nean, Madeira and the Canary Islands [16]. The large drop in catchesas the result of overfishing bespeaks this species as a target foraquaculture diversification. Little information is available about theinnate immune system and the defence mechanisms that thisspecies displays against bacterial infections. In the present study, wehave identified four different HAMP genes. cDNA sequences andgenomic organization have been characterized. A phylogeneticanalysis was carried out in order to establish orthologies. Finally,gene expression analyses in tissues, larvae and after lipopolysac-charide (LPS) injectionwere performed to elucidate the possible roleof HAMPs in the response against bacteria in redbanded seabream.

2. Materials and methods

2.1. Source of fish and experimental rearing conditions

All experimental larvae were obtained from fertilized eggscollected after natural spawning from a captive wild broodstockheld at Centro IFAPA El Toruno aquaculture facilities (El Puerto deSanta Marıa, Cadiz, Spain). Eggs were incubated at a density of2000 eggs l"1 in 300 l cylinder tanks with gentle aeration and onewater exchange every 2 h. Temperature and salinity during allexperiments were 20

#C and 38 ppt, respectively. Newly hatched

larvae were transferred to two 700 l tank at an initial density of45–50 larvae l"1 with a continuous light intensity of 600–800 lux.Larval rearing conditions were similar to those reported in ref. [17].Pools of larvae from 2 to 21 days after hatching (DAH) (n $ 3) werecollected, washed with DEPC water, frozen in liquid nitrogen, andstored at "80

#C until analysis. Embryos were taken approximately

at 15 h after fertilization.Juvenile redbanded seabream individuals (average weight $

90 % 5 g; n $ 3) were euthanized by immersion in 2-phenox-yethanol. Liver, intestine, stomach, spleen, kidney, skin, heart, gilland brain tissues were rapidly dissected, frozen in liquid nitrogenand stored at "80

#C until use.

For induction experiments, redbanded seabream specimens(average weight $ 40 % 2 g) were intraperitoneally injected withLPS (30 mg kg body weight"1; Escherichia coli serotype 0111: B4,Sigma Aldrich). Controls were injected with PBS. Prior to manipu-lation, specimens were anaesthetized with phenoxyethanol(50 ppm). After injection, animals recovered normally without anysigns of stress. Three specimens for each treatment were eutha-nized (100 ppm) and sampled at 1, 3, 6, 24 and 48 h post-inocula-tion (p.i.). Kidneys were rapidly dissected, frozen in liquid nitrogen,and stored at "80

#C until use. To study the response to LPS in

different tissues a second experiment was carried out. LPS doses,manipulation conditions and sample collection were as describedabove. In this case, animals (average weight $ 10 g) were eutha-nized at 3 h p.i. and kidney, spleen and liver tissues (n $ 3) weredissected for further analysis.

2.2. Molecular cloning of HAMPs

For HAMP1, a partial cDNA sequence was obtained aftersuppression subtractive hybridization (SSH) using the PCR-select"

cDNA subtraction kit (Clontech). Conditions were as describedpreviously [18]. Animals from the tester group were inoculatedwith LPS (30 mg kg body weight"1; E. coli serotype 0111: B4; SigmaAldrich) and sampled at 3 and 24 h after injection. Controls (driver)were injected with PBS and sampled at the same time.

For HAMP2, HAMP3 and HAMP4, partial sequences wereobtained after RT-PCR using total RNA from kidney and liver tissuesof animals stimulated with LPS. For PCR amplification, the degen-erate primer pairs Hepcons1/Hepcons2 and Hepcons3/Hepcons2were employed (Table 1). PCR conditions involved denaturing at94

#C for 2 min, followed by 35 cycles of 30 s at 94

#C, 15 s at 52

#C

and 1 min at 72#C. PCR amplified products were purified using the

Nucleofast96 PCR plates (Macherey Nagel) and subsequentlycloned with the TOPO TA Cloning kit for sequencing (Invitrogen).Positive clones were sequenced using the BigDye Terminator v3.1Cycle Sequencing kit in an ABI3130 Genetic Analyzer (AppliedBiosystems). DNA sequences were analyzed using Seqman v5.06(DNASTAR, Madison, WI, USA).

To obtain the complete cDNA sequences of HAMP genes, the 30

and 50 RACE Kits (Invitrogen) were employed. Conditions were thesameas describedpreviously [18]. Primers used in the 30 RACEwere:PauHepc13 and PauHepc15 for HAMP1; PauHepc21 for HAMP2;PauHepc31 for HAMP3; PauHepc41 for HAMP4 (Table 1). Primersemployed in the 50 RACE were: PauHepc22, PauHepc24, PauHepc26for HAMP2; PauHepc32, PauHepc34, PauHepc36 for HAMP3; Pau-Hepc46, PauHepc42, PauHepc44 for HAMP4 (Table 1). Primers weredesigned using Oligo v6.89 software (Medprobe). PCR conditionsinvolved denaturing at 94

#C for 2 min, followed by 35 cycles of 30 s

at 94#C, 15 s at 65

#C and 1 min at 72

#C. PCR products sequencing

and sequence analysis were performed as described above.For genomic amplification, a muscular portion of a redbanded

seabream specimen was excised and kept at "80#C. Total genomic

DNA was isolated from 150mg of the tissue using the FastDNA kit(Bio101 Inc., Vista, CA, USA) for 40 s and speed setting 5 on theFastprep FG120 (Bio101 Inc.) following the manufacturer’s protocol.PCR conditions involved denaturing at 94

#C for 2 min, followed by

30 cycles of 30 s at 94#C, 15 s at 60

#C, and 1 min at 72

#C. Primers

used were PauHepc13/PauHepc12 for HAMP1, Hepcons3/PauHepc22for HAMP2 and Hepcons3/PauHepc32 for HAMP3. PCR productswere cloned and sequenced as described above. Ten random clonesfor each gene were analyzed. The HAMP4 genomic sequence wasidentified unintentionally after analysis of HAMP2 clones.

2.3. Molecular characterization of HAMPs

Alignments of sequences were carried out using the MegAlignv7.2.1 program from the LASERGENE software suite (DNASTAR). Forphylogenetic analysis, sequences were retrieved from GenBank/EMBL/DDBJ or from ENSEMBL database for Oryzias latipes, Takifugurubripes and Tetraodon nigroviridis. The bestfit model of sequenceevolution (JTT ! I ! G) was determined using the ProtTest version1.3 [19]. Phylogenetic analysis was carried out using the PHYLIPpackage [20] as follows: a bootstrap analysis was carried out usingSEQBOOT (1000 replicates), and data were then analyzed bymaximum likelihood (PHYML), which generated 1000 trees.The consensus phylogenetic tree was subsequently obtained(CONSENSE). Trees were drawn using the TreeView program v1.6.2.

Isoelectric point was calculated using the ProtParam tool (http://cn.expasy.org/cgi-bin/protparam). Signal peptides were identifiedusing SignalP 3.0 (http://www.cbs.dtu.dk/services/SignalP).

2.4. RNA isolation and gene expression analysis

Homogenization of larvae (n $ 3 independent pools/sampling)and juvenile tissues (n $ 3 specimens) was carried out in the

B. Martin-Antonio et al. / Fish & Shellfish Immunology 26 (2009) 483–491484

Author's personal copy

Fastprep FG120 instrument (Bio101) using Lysing Matrix D (Q-BioGene) for 40 s at speed setting 6. Total RNA was isolated from50 mg of tissues or pools of larvae using the RNeasy Mini Kit(Qiagen). For skeletal muscle, heart and skin the RNeasy FibrousTissue Mini Kit (Qiagen) was utilized. All RNA isolation procedureswere carried out in accordancewith themanufacturer’s protocol. Inall cases, total RNAwas treated twice with DNase I using the RNase-Free DNase kit (Qiagen) for 30 min. RNA sample quality waschecked on an agarose gel, and quantification was spectrophoto-metrically accomplished. Total RNA (1 mg) from each sample wasreverse-transcribed using the iScript" cDNA Synthesis kit(Bio–Rad). Real-time PCR analysis was carried out using an iCycler(Bio–Rad). Reactions were done in a 25 mL volume containing cDNAgenerated from 10 ng of original RNA template, 300 nM each ofspecific forward and reverse primers (Table 1), and 12.5 mL of iQ"SYBR Green Supermix (Bio–Rad). Matching oligonucleotide primerswere designed using the Oligo v6.89 software (Medprobe). Theamplification protocol used was as follows: initial denaturation andenzyme activation for 3.5 min at 95

#C, followed by 40 cycles of

95#C for 15 s, optimal annealing temperature for 30 s (see Table 1)

and 72#C for 30 s. Each assay was done in duplicate. For normali-

zation of cDNA loading, all samples were run in parallel with thehousekeeping gene b-actin (Accession No. AB455098; Table 1) or18S rRNA [21]. RelativemRNA expressionwas determined using the2("DDCt) method [22]. Comparisons between groups were made byone-way analysis of variance, followed by a Tukey test for identi-fication of the statistically distinct groups. Significant differenceswere accepted at P < 0.05.

3. Results

3.1. cDNA cloning and genomic characterizaton of HAMP genes

Four HAMP cDNA sequences were identified in redbandedseabream. A partial sequence encoding the HAMP1 was detectedafter sequence analysis of a subtracted cDNA library using the SSHtechnique. Partial sequences corresponding to the HAMP2, HAMP3and HAMP4 genes were obtained after RT-PCR amplification using

degenerate primers (see Materials and methods). Complete HAMPcDNA sequences were achieved by 50 and 30 RACE-PCR.

HAMP1 encoded a transcript of 778 nucleotides (nt) [AccessionNo. AB440775] that contained a 50-untranslated region (201 nt)followed by an open reading frame (ORF) of 270 nt (Table 2). The30-untranslated region was 304 nt long, and included a canonicalpolyadenylation signal (AATAAA, 756–761) and a short oligo-A tail.For HAMP2, a cDNA of 607 nt was identified [Accession No.AB440776]. The ORF comprised 279 nt, and was preceded by a 90-nt 50-untranslated region. The 30 UTR length was 235 nt andincluded a canonical polyadenylation signal (AATAAA, 580–585).

HAMP3 cDNA possessed an ORF of 264 nt [Accession No.AB440777]. The 50 and 30-untranslated regions were 85 and 66 ntlong, respectively. HAMP4 comprised 609 nt in length [AccessionNo. AB440778] with an ORF of 285 nt. The 50 and 30 UTR lengthwere 90 and 231 nt, respectively. HAMP4 included a canonicalpolyadenylation signal (AATAAA, 592–597), whereas HAMP3possessed the sequence motif ATTAAA (399–404).

The three typical HAMP domains (signal peptide, prodomain andmature peptide) were identified in the deduced amino acidsequences of the four HAMP cDNAs from redbanded seabream. Asignal peptidase cleavage site was detected between Ala24 and Val25

or Asp25 (Fig. 1A). The prodomain ranged between 40 and 50 aa inlength. All except HAMP3 possessed the RX(K/R)R motif typical ofpropeptide convertases. In addition, two putative processing sites,RHKR andRRRR,were recognized inHAMP4. Mature peptide sizewascomprised between 24 and 31 aa, and the four HAMPs contained theeight conserved cysteine residues. Isoelectric points differed amongHAMPs, especially in the mature peptide (Table 2). HAMP2 andHAMP4possessed similar values in thewholepeptide andprodomainregion. HAMP3 showed the lowest pI values in all cases.

All four HAMP genes were constituted by three exons and twointrons (Fig. 1B). Exon 1 contained 50 UTR, the signal peptide andpart of prodomain sequence. Exon 2 encoded only a part of theprodomain, and its size (78 nt) was conserved in the four genes.Exon 3 included the final part of the prodomain, mature peptideand 30 UTR. Introns were small ranging between 93 and 166 nt inlength. All of them were flanked by the canonical ‘GT . AG’sequences.

3.2. Phylogenetic analysis

The phylogenetic analysis of prepropeptide HAMP sequencesrevealed that fish sequences clustered in two main groups: onecontaining HAMPs from acanthopterygians and non-acan-thopterygians fish (also known as HAMP1-like group), and anotherincluding only sequences from acanthopterygians (also termedHAMP2-like group) (Fig. 2). This latter group included HAMP2,HAMP3 and HAMP4 from redbanded seabream. These threepeptides clustered separately in different branches but only HAMP3was supported by significant bootstrap value (100). HAMP2 andHAMP4 appeared as the most phylogenetically related.

3.3. Gene expression in tissues from juvenile redbanded seabream

In order to establish the expression patterns of the four HAMPtranscripts in tissues,mRNAswerequantified indifferentorgans fromredbanded seabream specimens (Fig. 3). HAMP1 was ubiquitouslyexpressed in the nine tissues studied with slightly more abundantmRNA levels in kidney and spleen. In contrast, HAMP2, HAMP3 andHAMP4 showed tissue-specific expression patterns although theirtranscripts could be detected at a lower level in all tissues. WhereasHAMP2 transcripts were quite abundant in kidney, spleen andintestine, HAMP3 and HAMP4were mainly expressed in liver.

Table 1List of primers used in this study.

Target Primers Sequence (50 / 30) Fragment size (Ta)

HAMPs Hepcons1 RTTGCAGTYRCMGTGRCMSTCGTHepcons2 RTTGYTMASTGSCTCYTCCRGCTCYTGHepcons3 ATGAAGRCATTCAGYRTTGC

ACTB Pauactin1 CAGTCGTGACCTCTTCTTCCCCCTGT 96 bp (68#C)

Pauactin2 GGTCTGGCACCCCTTGTAACCCCATCHAMP1 PauHepc13 TGCAGTGACACTCGTGCTCGCCTTT 82 bp (68

#C)

PauHepc14 CTCCTCCAGCTATTGCACCCCGTTCPauHepc15 CAATGGAATCGTGGATGATGCCGAGTPauHepc12 GCAGTAACCGCAGCCCTTGTTGG

HAMP2 PauHepc21 CCAGAGGAGTCCTGGAAGATGCCGTA 93 bp (68#C)

PauHepc22 ACAGCAACGACAGCAAGTGCGACAPauHepc24 CTGCTTCTGCTGGCATACGGCATCPauHepc26 TCATTGGCTCCTCCAGCTCTTGCACC

HAMP3 PauHepc31 TGCCATTTGACATCAGACAGAAGCCTCA 107 bp (68#C)

PauHepc32 CAATCCTCACTGGCAGCACATTCCACPauHepc34 CTGCACTTTATGAGGCCGCTATGAGGPauHepc36 GCTGCATCGTCATTGCTTATTTCTTCCTC

HAMP4 PauHepc45 TACTCTTAAACACATTTCCAGCTTTCAATC 157 bp (58#C)

PauHepc46 TTTATTGATCACACACTTTAGTACATGAAGPauHepc41 TTCACTGAGGTGCAAGAGCAGGAGGAGCPauHepc42 TGTCTGTTGTTATACGGCATCTTCCAGGACPauHepc44 CATCTCTTCATGTGCAGCAACTGGACTGT

Amplicon size and annealing temperature are indicated for those pairs used in real-time PCR.

B. Martin-Antonio et al. / Fish & Shellfish Immunology 26 (2009) 483–491 485

Author's personal copy

3.4. Gene expression during larval development

Expression patterns of the four HAMP genes in redbandedseabream larvae are depicted in Fig. 4. HAMP genes showed distinctexpression patterns. HAMP1 mRNAs increased significantlybetween 5 and 9 DAH. Thereafter, mRNAs decreased althougha peak at 19 DAH was identified. HAMP2 and HAMP3 showed thehighest mRNA levels in embryos decreasing progressively duringlarval development. Interestingly, a gene expression peak at 9 DAHwas identified for HAMP2, HAMP3 and HAMP4 genes. The lattergene also showed significant higher mRNA levels at 5 DAH.

3.5. Gene expression after LPS inoculation in kidney

To determine the role of HAMPs in the response against bacteria,redbanded seabream specimens were injected intraperitoneallywith LPS (Fig. 5). Expression levels were quantified in kidney.HAMP1 transcripts increased rapidly at 1 h p.i. recovering thesteady-state levels between 6 and 24 h p.i. HAMP3 and HAMP4mRNAs peaked later at 6 h p.i. In contrast, HAMP2 transcriptsdecreased significantly at 1 h p.i increasing later at 24 h p.i.

To study if this response was comparable among tissues,expression patterns for the four HAMP genes were evaluated inliver, kidney and spleen at 3 h after LPS injection. HAMP1, HAMP3and HAMP4 were significantly induced in all three tissues (Fig. 6).The liver-specific expressed genes, HAMP3 and HAMP4, exhibitedthe highest induction values in this organ (417- and 5565-fold,respectively). Transcription of HAMP2was not activated in any case,and even a significant drop in expression levels was measured inspleen.

4. Discussion

HAMPs are important cysteine-rich bioactive peptides of theinnate immune system. They play a pivotal role against bacterialinfections and iron homeostasis. In the present study, we haveidentified four different HAMP genes in the sparid redbandedseabream. Their genomic organization, three exons and twointrons, as well as the relative position and length of introns werequite conserved among them, and with respect to other fish andmammals [2]. All HAMP cDNAs possessed a 24 aa signal peptide,a variable prodomain and a mature peptide with eight cysteines

Table 2Molecular characteristics of HAMP genes from P. auriga.

cDNA 50 UTR 30 UTR Whole PI

Whole Prodomain Propeptide Mature Cysteines

HAMP1 778 201 304 90 6.69 4.80 6.85 8.51 8HAMP2 607 90 235 93 8.84 4.57 8.94 10.07 8HAMP3 418 85 66 88 4.14 3.76 4.04 6.68 8HAMP4 609 90 231 95 8.42 4.71 8.50 9.60 8

Sizes for cDNA, 50 UTR and 30 UTR are indicated in bp. Whole peptide length is in aa. Isoelectric points for whole peptide, prodomain, propeptide and mature peptide as well asthe number of cysteines are shown.

Fig. 1. Molecular characterization of HAMP gene from P. auriga. (A) Alignment of HAMP amino acid sequences. Dots indicate identity, and hyphens represent indels. Signal peptide isboxed. The predicted prodomain cleavage site is indicated by an arrow. Conserved cysteines in the mature peptide are shaded. (B) Genomic organization of HAMP genes. In DNA,numbers indicate size in bp. In the mRNA, figures denote size in bp for 50 and 30 untranslated regions and in amino acids for signal peptide, prodomain and mature peptide.

B. Martin-Antonio et al. / Fish & Shellfish Immunology 26 (2009) 483–491486

Author's personal copy

required for disulphide bridges and protein structure. The typicalRX(K/R)R motif of propeptide convertases [23] was present in allexcept HAMP3 (RQKP). The identification of a non-canonical motifhas also been reported in other seabreams such as Sparus aurata(RHWK) [9] and A. schlegelii (As-hepc5, RHNR; As-hepc6, RRWR;As-hepc7, RSKT) [15] and in O. mossambicus (TH2–2, RQKH) [8]. Allof these hepcidins belong to the more diversified acanthopterygianHAMP2-like family [3], and they could be functional. For instance,transcription of the HAMP from S. aurata increases after LPS treat-ments and possesses antibacterial activity for four out of eightstrains studied [9]. In contrast, HAMP TH2–2 increases its expres-sion after iron dextran treatments, and does not possess

antimicrobial activity against any bacteria assayed. In the red-banded seabream, HAMP3 transcription is also induced after LPStreatments although its low pI (6.68), not optimum for an antimi-crobial activity, suggests a limited function as antimicrobialpeptide.Whereas the actual cleavage site for fish hepcidins remainsto be experimentally determined, these HAMPs could have gainedadditional functions during evolution such as in anemia, hypoxia,and inflammation [2].

Gene duplications represent a major evolutionary force invertebrate organisms [24,25]. Most gene duplicates are lost orsilenced during evolution. Only some functional groups involved inprotein binding, signal transduction, transcription, development,

Fig. 2. Phylogenetic relationships among HAMPs from a wide range of vertebrates using the maximum likelihood method. HAMPs from Homo sapiens (Hsa) and Mus musculus(Mmu) were used as outgroups to root the tree. Only bootstrap values higher than 60% are indicated on each branch. The scale for branch length (0.1 substitutions/site) is shownbelow the tree. Abbreviations are: Acanthopagrus schlegelii (Asc); Danio rerio (Dre); Dicentrarchus labrax (Dla); Fundulus heteroclitus (Fhe); Ictalurus punctatus (Ipu); Ictalurus furcatus(Ifu); Gadus morhua (Gmo); Lithognathus mormyrus (Lmo); Morone chrysop (Mch); Oreochromis mossambicus (Omo); Oryzias latipes (Ola); Paralichthys olivaceus (Pol); Pseudo-pleuronectes americanus (Pam); Pseudosciaena crocea (Pcr); Perca fluviatilis (Pfl); Tetraodon nigroviridis (Tni); Takifugu rubripes (Tru); Salmo salar (Ssa); Sparus aurata (Sau); Soleasenegalensis (Sse); Scophthalmus maximus (Sma). Accession number for each sequence is indicated. The specific name of HAMP gene is also indicated if available. HAMP1 sequencesfor T. rubripes and T. nigroviridis were taken from Hilton and Lambert [3] and for O. mossambicus from Huang et al. [8]. HAMPs from P. auriga are shaded.

B. Martin-Antonio et al. / Fish & Shellfish Immunology 26 (2009) 483–491 487

Author's personal copy

DNA binding, receptor activity, ion transport, and protein modifi-cation are preferentially retained [24]. The functional divergence ofduplicates occurs through three main processes: neofunctiona-lization, in which one copy acquires an entirely new functionwhereas the alternative copy maintains the original function;subfunctionalization, in which each descendant copy adopts part ofthe tasks of the ancestral gene; subneofunctionalization in whicha rapid subfunctionalization is followed by prolonged andsubstantial neofunctionalization [26–28]. In these processes, thepositive Darwinian selection plays a crucial role accelerating thefixation of advantageous mutations. In vertebrates three rounds oflarge-scale gene duplications (referred to as 1R, 2R, and 3R or fish-specific genome duplication) have been identified [29,30]. Phylo-genetic analysis of HAMP sequences indicates that HAMP2-likegroup would have appeared after the 3R genome duplication (sinceit has not been identified in tetrapods until now). Most duplicates

have been lost in the non-acanthopterygian fish retaining only theHAMP1 lineage. Fixation of HAMP2-like genes in acanthopterygianscould be favored by the radiation of teleosts in different marine andbrackish environments and the operation of positive Darwinianselection as recently reported for HAMPs in pleuronectiformes andperciformes [14]. Also, an adaptive positive evolution to the polarenvironment has been reported in notothenioid fishes bearinga novel four-cysteine HAMP2-like variant [6]. However, additionalHAMP2 lineage-specific duplications in acanthopterygians associ-ated to neofunctionalization and subfunctionalization forces arerequired to explain the high complexity of this cluster. Thishypothesis is supported by the identification of three HAMP2-likesequences (HAMP2, HAMP3, HAMP4) in redbanded seabream withdifferent expression patterns in tissues and during larval develop-ment as well as the different kinetic properties after LPStreatments.

Fig. 3. Relative expression levels of HAMP genes in different tissues from P. auriga. Data were expressed as the mean fold difference (mean % SEM, n $ 3) from the calibrator group(skin). Data were normalized using the 18S rRNA. Values with the same superscript are not significantly different (P > 0.05).

Fig. 4. Relative expression levels of HAMP genes during larval development in P. auriga. Expression values were normalized to those of b-actin. Data were expressed as the meanfold difference (mean % SEM, n $ 3) from the calibrator group (one DAH). Values with the same superscript are not significantly different (P > 0.05).

B. Martin-Antonio et al. / Fish & Shellfish Immunology 26 (2009) 483–491488

Author's personal copy

This study shows that HAMPs are differentially expressedduring larval development. Although the four genes could bedetected in embryos, HAMP2 and HAMP3 exhibited the highestexpression levels. The early detection of mRNAs in fish eggs is not

rare and it has been associated to transcripts storedmaternally thatwould enable the development in these first stages. However, genesthat are highly expressed in embryos, such as insulin-like growthfactor 2, generally play an essential role in embryogenesis [31,32].

Fig. 5. Relative expression levels of HAMP genes after LPS injection. Expression values were normalized to those of b-actin. Transcripts were quantified using total RNA fromhead kidney (n $ 3). Data are expressed as the mean fold change (mean % SEM) from the calibrator group (PBS 1 h). Expression values corresponding to the control (PBS), andLPS-injected specimens are indicated by light-gray and dark-gray bars, respectively. Bars with different letters indicate statistically significant differences (P < 0.05).

Fig. 6. Relative expression levels of HAMP genes in tissues after LPS injection. Expression values were normalized to those of b-actin. Transcripts were quantified at 3 h. Data areexpressed as the mean fold change (mean % SEM) from the calibrator group (liver PBS 3 h). Bars with different letters indicate statistically significant differences (P < 0.05).

B. Martin-Antonio et al. / Fish & Shellfish Immunology 26 (2009) 483–491 489

Author's personal copy

Hepcidin null mice develop iron overload [33], while animals thatoverexpress hepcidin develop severe hypochromic anemia [34]. Inzebrafish, the mutant weissherbst (ferroportin1-deficient) die earlyin life with severe hypochromic anemia. This ferroportin1 isrequired for the transport of iron from maternally derived yolkstores to the circulation [35]. Hepcidin can directly bind to ferro-portin to provoke its internalization and degradation and reducecellular iron export (reviewed in ref. [2]). Hence, it is probably thatHAMP2 and HAMP3 could be necessary to regulate iron levelsduring embryogenesis in this species.

HAMP1 and HAMP4 increased their expression levels laterduring larval development peaking between five and nine DAH.Similar results were described in P. americanus for HAMP WF1 andHAMP WF3a, peaking between five and 15 DAH [4], and in Ictaluruspunctatus showing higher expression levels in posthatched larvaethan in embryos [36]. In P. auriga, this age corresponds to the stage3 or exotrophic I [17]. During this period occurs an intenseorganogenesis, the yolk reserves and oil globules become exhaus-ted and mucous cells appear in the intestine. Moreover, hepaticsinusoids appear with blood inside and the hematopoietic tissueproliferates in the kidney. Under this scenario, we can postulatethat the commencement of external feeding would require a higherlevel of antimicrobial defenses to cope with this new hostileenvironment. Moreover, the induction of HAMP2 and HAMP3 at day9 DAH also suggests that they would be involved in the regulationof iron metabolism when intestine and liver become completelyfunctional.

HAMPs were detected in all adult tissues in the redbandedseabream although they showed distinct expression profiles. TheHAMP3 and HAMP4 transcripts were highly abundant in liver. TheHAMP2 mRNAs were expressed mainly in kidney, spleen andintestine whereas the HAMP1 was detected at a similar level in alltissues studied. These differences in the expression patterns amongHAMP paralogs make comparisons difficult across species. Forinstance, if we only consider HAMP1-like sequences, the HAMP sal1from Salmo salar was expressed ubiquitously at relatively highlevels in the liver, blood and muscle; whereas the HAMP sal2 wasbarely detectable in the gill and skin [4]. In Paralichthys olivaceus,the HAMP Hep-JF2 was detected in gill, liver, heart, kidney,peripheral blood leucocytes, spleen and stomach [5]. In I. punctatus,the HAMP showed the highest expression in liver and spleen, fol-lowed by gill and intestine [36]. In Psetta maxima, HAMPwas highlyabundant in liver, heart, head kidney, spleen, skin and gill [37]. In O.mossambicus, the HAMP TH2–3 mRNA was mainly detected in theliver [8]. This complex situation can be applied also to the acan-thopterygians HAMP2-like sequences although in this case most ofthem were expressed at high level in liver [4,8,9,15]. However, inmost of these expression studies the coexistence of multipleparalogous HAMP genes and the design of specific gene primers indivergent regions have not been taken into consideration. Theincreasing number of EST in the public databases and the identifi-cation of conserved clusters in acanthopterygians offer a powerfultool for further investigation of HAMP evolution, their tissuedistribution and the degree of function conservation across species.

HAMPs are important components of the innate immunesystem. Different stressors such as LPS or bacteria can activate theirexpression [4,8–10,37,38]. However, the HAMP responses seem tobe tissue-, time- and infection-specific. For example, in Moronechrysops, the HAMP was induced by Streptococus iniae in liver [38].However, in I. punctatus, Edwardsiella ictaluri failed to induce HAMPexpression in liver although it did in spleen and head kidney [36]. Inthe seabream S. aurata, Vibrio anguillarum induced HAMP tran-scription in liver, spleen and peritoneal exudate leucocytes only at4 h but not at 72 h. However, transcripts increased non-signifi-cantly in head kidney [9]. Our data also support tissue-, time- and

gene-specific responses after LPS treatment. Whereas the HAMP1was activated early at 1 h after LPS inoculation, the HAMP2 tran-scripts decreased at this time although they increased later at 24 hin kidney. Moreover, all HAMPs except HAMP2 activated theirtranscription in the three tissues studied at 3 h p.i. with HAMP3 andHAMP4 showing the highest induction values in liver. The particularexpression patterns of HAMP2 with limited responses against LPSand higher transcripts in organs involved in iron homeostasissuggest that HAMP2 could mainly function as an iron-regulatoryhormone.

In conclusion, four HAMP genes have been identified andcharacterized in the redbanded seabream. Phylogenetic analysisseparated two main clusters revealing the HAMP2-like group as themost recent and complex evolutionary group present only inacanthopterygians. Real-time analyses revealed that the fourHAMPs exhibited different expression profiles in tissues, embryosand larvae and after injection with LPS suggesting additionalfunctions besides the antimicrobial activity. Although the existenceof additional HAMP genes in the genome of P. auriga cannot beexcluded, a better knowledge of the expression profiles of thesefour genes identified under different feeding regimes and growthconditions in larvae and adults, will undoubtedly contribute toimprove the processes of fish husbandry with benefits for theaquaculture industry.

Acknowledgements

This study was funded by Project RTA2005-00028-C02 fromINIA (Subprograma Nacional de Recursos y Tecnologıas Agrarias encooperacion con las Comunidades Autonomas) from the Spanishgovernment.

References

[1] Segat L, Pontillo A, Milanese M, Tossi A, Crovella S. Evolution of the hepcidingene in primates. BMC Genomics 2008;9:120.

[2] Shi J, Camus AC. Hepcidins in amphibians and fishes: antimicrobial peptides oriron-regulatory hormones? Dev Comp Immunol 2006;30:746–55.

[3] Hilton KB, Lambert LA. Molecular evolution and characterization of hepcidingene products in vertebrates. Gene 2008;415:40–8.

[4] Douglas SE, Gallant JW, Liebscher RS, Dacanay A, Tsoi SC. Identification andexpression analysis of hepcidin-like antimicrobial peptides in bony fish. DevComp Immunol 2003;27:589–601.

[5] Hirono I, Hwang J-Y, Ono Y, Kurobe T, Ohira T, Nozaki R, et al. Two differenttypes of hepcidins from the Japanese flounder Paralichthys olivaceus. FEBS J2005;272:5257–64.

[6] Xu Q, Cheng CH, Hu P, Ye H, Chen Z, Cao L, et al. Adaptive evolution of hepcidingenes in antarctic notothenioid fishes. Mol Biol Evol 2008;25:1099–112.

[7] Lauth X, Babon JJ, Stannard JA, Singh S, Nizet V, Carlberg JM, et al. Bass hep-cidin synthesis, solution structure, antimicrobial activities and synergism,and in vivo hepatic response to bacterial infections. J Biol Chem 2005;280:9272–82.

[8] Huang PH, Chen JY, Kuo CM. Three different hepcidins from tilapia, Oreo-chromis mossambicus: analysis of their expressions and biological functions.Mol Immunol 2007;44:1922–34.

[9] Cuesta A, Meseguer J, Esteban MA. The antimicrobial peptide hepcidin exertsan important role in the innate immunity against bacteria in the bony fishgilthead seabream. Mol Immunol 2008;45:2333–42.

[10] Rodrigues PN, Vazquez-Dorado S, Neves JV, Wilson JM. Dual function of fishhepcidin: response to experimental iron overload and bacterial infection insea bass (Dicentrarchus labrax). Dev Comp Immunol 2006;30:1156–67.

[11] Fraenkel PG, Traver D, Donovan A, Zahrieh D, Zon LI. Ferroportin1 is requiredfor normal iron cycling in zebrafish. J Clin Invest 2005;115:1532–41.

[12] Hu X, Camus AC, Aono S, Morrison EE, Dennis J, Nusbaum KE, et al. Channelcatfish hepcidin expression in infection and anemia. Comp Immunol MicrobiolInfect Dis 2007;30:55–69.

[13] Verga Falzacappa MV, Muckenthaler MU. Hepcidin: iron-hormone and anti-microbial peptide. Gene 2005;364:37–44.

[14] Padhi A, Verghese B. Evidence for positive Darwinian selection on thehepcidin gene of Perciform and Pleuronectiform fishes. Mol Divers2007;11:119–30.

[15] Yang M, Wang KJ, Chen JH, Qu HD, Li SJ. Genomic organization and tissue-specific expression analysis of hepcidin-like genes from black porgy (Acan-thopagrus schlegelii B). Fish Shellfish Immunol 2007;23:1060–71.

B. Martin-Antonio et al. / Fish & Shellfish Immunology 26 (2009) 483–491490

Author's personal copy

[16] Bauchot ML, Hureau JC. Sparidae. In: Whitehead PJP, Bauchot ML, Hureau JC,Nielsen J, Tortonese E, editors. Fishes of the North-Eastern Atlantic and theMediterranean. Bungay, UK: UNESCO; 1986. p. 902–3.

[17] Sanchez-Amaya MI, Ortiz-Delgado JB, Garcıa-Lopez A, Cardenas S,Sarasquete C. Larval ontogeny of redbanded seabream Pagrus auriga Valen-ciennes, 1843 with special reference to the digestive system. A histological andhistochemical approach. Aquaculture 2007;263:259–79.

[18] Jimenez-Cantizano RM, Infante C, Martin-Antonio B, Ponce M, Hachero I,Navas JI, et al. Molecular characterization, phylogeny, and expression of c-typeand g-type lysozymes in brill (Scophthalmus rhombus). Fish ShellfishImmunol 2008;25:57–65.

[19] Abascal F, Zardoya R, Posada D. ProtTest: selection of best-fit models of proteinevolution. Bioinformatics (Oxford, England) 2005;21:2104–5.

[20] Felsenstein J. PHYLIP – phylogeny inference package (Version 3.2). Cladistics1989;5:164–6.

[21] Ponce M, Infante C, Funes V, Manchado M. Molecular characterization andgene expression analysis of insulin-like growth factors I and II in the red-banded seabream, Pagrus auriga: transcriptional regulation by growthhormone. Comp Biochem Physiol B Biochem Mol Biol 2008;150:418–26.

[22] Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2("ddCT) method. Methods 2001;25:402–8.

[23] Nakayama K. Furin: a mammalian subtilisin/Kex2p-like endoproteaseinvolved in processing of a wide variety of precursor proteins. Biochem J1997;327:625–35.

[24] Blomme T, Vandepoele K, de Bodt S, Simillion C, Maere S, Van de Peer Y. Thegain and loss of genes during 600 million years of vertebrate evolutionGenome Biol 2006;7:R43.

[25] Lespinet O, Wolf YI, Koonin EV, Aravind L. The role of lineage-specific genefamily expansion in the evolution of eukaryotes. Genome Res 2002;12:1048–59.

[26] Lynch M, O’Hely M, Walsh B, Force A. The probability of preservation ofa newly arisen gene duplicate. Genetics 2001;159:1789–804.

[27] Rastogi S, Liberles DA. Subfunctionalization of duplicated genes as a transitionstate to neofunctionalization. BMC Evol Biol 2005;5:28.

[28] He X, Zhang J. Rapid subfunctionalization accompanied by prolonged andsubstantial neofunctionalization in duplicate gene evolution. Genetics2005;169:1157–64.

[29] Meyer A, Van de Peer Y. From 2R to 3R: evidence for a fish-specific genomeduplication (FSGD). Bioessays 2005;27:937–45.

[30] Hoegg S, Brinkmann H, Taylor JS, Meyer A. Phylogenetic timing of the fish-specific genome duplication correlates with the diversification of teleost fish.J Mol Evol 2004;59:190–203.

[31] Gabillard JC, Rescan P-Y, Fauconneau B, Weil C, Le Bail P-Y. Effect of temper-ature on gene expression of the Gh/Igf system during embryonic developmentin rainbow trout (Oncorhynchus mykiss). J Exp Zool 2003;298A:134–42.

[32] Funes V, Asensio E, Ponce M, Infante C, Canavate JP, Manchado M. Insulin-likegrowth factors I and II in the sole Solea senegalensis: cDNA cloning andquantitation of gene expression in tissues and during larval development. GenComp Endocrinol 2006;149:166–72.

[33] Nicolas G, BennounM, Devaux I, Beaumont C, GrandchampB, KahnA, et al. Lackof hepcidin gene expression and severe tissue iron overload in upstreamstimulatory factor 2 (USF2) knockout mice. Proc Natl Acad Sci USA2001;98:8780–5.

[34] Nicolas G, Bennoun M, Porteu A, Mativet S, Beaumont C, Grandchamp B, et al.Severe iron deficiency anemia in transgenic mice expressing liver hepcidin.Proc Natl Acad Sci USA 2002;99:4596–601.

[35] Donovan A, Brownlie A, Zhou Y, Shepard J, Pratt SJ, Moynihan J, et al. Positionalcloning of zebrafish ferroportin1 identifies a conserved vertebrate ironexporter. Nature 2000;403:776–81.

[36] Bao B, Peatman E, Li P, He C, Liu Z. Catfish hepcidin gene is expressed in a widerange of tissues and exhibits tissue-specific upregulation after bacterialinfection. Dev Comp Immunol 2005;29:939–50.

[37] Chen SL, Li W, Meng L, Sha ZX, Wang ZJ, Ren GC. Molecular cloning andexpression analysis of a hepcidin antimicrobial peptide gene from turbot(Scophthalmus maximus). Fish Shellfish Immunol 2007;22:172–81.

[38] Shike H, Lauth X, Westerman ME, Ostland VE, Carlberg JM, Van Olst JC, et al.Bass hepcidin is a novel antimicrobial peptide induced by bacterial challenge.Eur J Biochem 2002;269:2232–7.

B. Martin-Antonio et al. / Fish & Shellfish Immunology 26 (2009) 483–491 491