Available online at www.sciencedirect.com

www.elsevier.com/locate/ygcen

General and Comparative Endocrinology 155 (2008) 227–233

Cloning of sole proopiomelanocortin (POMC) cDNA and theeffects of stocking density on POMC mRNA and growth

rate in sole, Solea solea

F. Palermo a, M. Nabissi a, G. Cardinaletti a, E. Tibaldi b,G. Mosconi a, A.M. Polzonetti-Magni a,*

a Dipartimento di Scienze Morfologiche e Biochimiche Comparate, Universita degli Studi di Camerino,

via Gentile III da Varano, 62032 Camerino (MC), Italyb Universita di Udine, Dipartimento di Scienze della Produzione Animale, via S. Mauro 2, 33010 Pagnacco, Udine, Italy

Received 7 November 2006; revised 14 March 2007; accepted 1 May 2007Available online 8 May 2007

Abstract

Proopiomelanocortin (POMC) is an important gene implicated in different functions, such as the stress response of the hypothala-mus–pituitary–adrenal axis. The aim of the present study was to determine whether farming conditions, such as stocking density, canbe considered a powerful stressor influencing in turn the growth rate in juvenile fish. Thus, POMC cDNA expression was investigatedduring adaptation to farming conditions in sole (Solea solea), as a model for studying the effects of rearing densities on stress response;different stocking densities (50, 100, and 250 animals/m2) were applied and, after 7 and 21 days, the fishes were examined for body weightand plasma cortisol levels as indicators of stress. In addition, proopiomelanocortin was cloned and sequenced from the brain of sole,allowing semi-quantitative RT-PCR to be performed to evaluate POMC mRNA expression in brain tissue. There was a significantincrease in cortisol levels in fish reared at high stocking densities of 250/m2 compared to fish reared at control densities of 100 and50/m2, in both experimental times, i.e., 7 and 21 days. The high stocking densities were also found to decrease the specific growth rateof fish. Moreover, it was demonstrated that the highest stocking density induced a significant decrease in sole POMC mRNA expression.It is concluded that POMC and cortisol are both involved in the stress response due to high rearing densities, during which cortisol mayserve as a negative regulator of POMC.� 2007 Elsevier Inc. All rights reserved.

Keywords: Solea solea POMC cDNA; POMC mRNA; Stocking density; Growth rate

1. Introduction

Proopiomelanocortin (POMC) is a precursor proteinthat undergoes posttranslational processing to yield severalbiologically active peptides. The functional ones, detectedin a variety of tissues, were generated through a series ofordered proteolytic cleavage of POMC by prohormoneconvertases. POMC processing in the anterior pituitaryresults in an N-terminal peptide, adrenocorticotropic

0016-6480/$ - see front matter � 2007 Elsevier Inc. All rights reserved.

doi:10.1016/j.ygcen.2007.05.003

* Corresponding author. Fax: +39 0737 402738.E-mail address: alberta.polzonetti@unicam.it (A.M. Polzonetti-Magni).

hormone (ACTH), and beta-lipotropin (beta-LPH). Inthe intermediate lobe of the pituitary, each polypeptide isfurther cleaved at dibasic amino acids and yields melano-cyte stimulating hormones (MSH), alpha-MSH, beta-MSH, and gamma-MSH, corticotropin-like intermediatelobe peptide (CLIP), and beta-endorphin (beta-EP).POMC mRNA sequence and its protein structure havealready been well documented in fish species (Karsi et al.,2005), including the data on zebrafish (Gonzalez-Nunezet al., 2003; Hansen et al., 2003).

POMC-derived peptides play important physiologicalroles in the stress response (Mosconi et al., 2006), duringwhich activation of the hypothalamus–pituitary–interrenal

228 F. Palermo et al. / General and Comparative Endocrinology 155 (2008) 227–233

axis (HPI) results in production and secretion of cortico-tropin-releasing factor (CRF) in the hypothalamus, whichin turn stimulates the production of MSH, ACTH andbeta-EP. ACTH induces the elevation of plasma cortisollevels, and plays a key role in the restoration of homeosta-sis (Mosconi et al., 2006). Many studies have evaluated theeffects of stress on cortisol secretion and POMC-derivedpeptides (Balm and Pottinger, 1995; Mosconi et al., 1998;Rotllant et al., 2001; Flik et al., 2002); moreover, POMCmRNA expression during stress has been reported fromcarp (Arends et al., 1998), and rainbow trout (Winbergand Lepage, 1998).

The issue of fish welfare in aquaculture is one of growinginterest, and stocking density is considered one of the mainfactors that can influence fish homeostasis, gene expression(Gornati et al., 2004), and growth rate (Schram et al.,2006); crowding, in fact, is considered a powerful stressorwhich activates the neuroendocrine cascade related withthe stress responses.

The sole (Solea solea) is an interesting new species forMediterranean aquaculture as it is a highly valued specieswith a large market; in this context, the role of POMC infish adaptation to farming conditions was investigated inthe present study. For that purpose, a cDNA coding forPOMC was cloned and sequenced from S. solea, and thePOMC mRNA changes were related with those of plasmacortisol levels in order to answer the question of whetherfarming conditions, such as stocking density, can be con-sidered a powerful stressor influencing in turn the growthrate in sole juveniles.

2. Material and methods

2.1. Animals

The animals utilized were obtained from previous spawning of S. solea

broodstock maintained in captivity for two years in an Italian fish farm,Orbetello Pesca Lagunare s.r.l. (Comune di Orbetello GR; 42�26 01300N,11�12 03000E, Localita Santa Liberata). Sole juveniles (204 animals), 195days post-hatch (dph), were reared in fish-farm conditions under naturalphotoperiod and water temperature, in 4.65-m3 circular tanks at a stock-ing density of 250 animals/m2; the mean weight was found to be3.64 ± 1.06 g. Animal manipulation was performed according to the rec-ommendations of the University Ethical Committee and under the super-vision of the authorized investigators.

2.2. Experimental design

The stocking densities were 50, 100, and 250 animals/m2; we used thefish-farm stocking density (250 animals/m2) as our experimental control;the first experiment was performed for 7 days and the second for 21 days.The experiments were carried out in 3 different circular tanks (4.65 m3),and, according to the experimental periods, in each tank the fish werereared in duplicate in 0.26-m2 cages (51 · 51 · 51 cm; 500 lm Ø), using13, 26, and 65 fish for 50, 100, 250 animals/m2 stocking density, respec-tively. The tanks were connected with a closed seawater circulation system,equipped with UV and biological filters, and illuminated with artificiallight (at simulated natural photoperiod 11:13 LD) provided by two neontubes of 40 W each. Water quality (oxygen, salinity, pH, ammonia andnitrites) was monitored every 4 days throughout the experimental period.Fish were fed with commercial dry pellets (Gemma micro 0.75, Trouw

Aquaculture, France) twice a day with a feeding rate of 1% body weightper day. The tanks were inspected daily for mortalities and dead fisheswere removed immediately from the tanks after detection.

At the end of the experiments, the fish from each cage were anesthe-tized with 300 ppm of 2-phenoxyethanol (Sigma) and weighed, the bloodwas collected in heparinized syringes by cardiac puncture, centrifuged(1500g for 15 min at 4 �C), and the plasma was stored at �20 �C untilbeing assayed. The brain tissues including pituitaries were harvested,immediately frozen in liquid nitrogen and stored at �80 �C for molecularbiology experiments.

2.3. Evaluation of specific growth rate (SGR)

All the fishes were weighed at the beginning of each experiment. Then,after 7 and 21 days, they were anesthetized and the body weight wasrecorded before the sampling to determine the specific growth rate(SGR = (ln(Wt) � ln(W0)) · 100/T), where Wt = weight (g) variable atday 7 (first experiment) and at day 21 (second experiment); W0 = weight(g) variable at day 1; T = number of days.

2.4. Cortisol assay

Plasma cortisol levels were analyzed by an enzyme immunoassay (EIA)method (Cayman, USA), using a specific anti-cortisol antibody. The assaysensitivity was 12 pg/ml, and the inter- and intra-assay coefficients of var-iation were 6.5 and 4.2%, respectively. Plasma samples were previouslyextracted with methylene chloride (Cold Spike Extraction) as describedby the supplier. To validate cortisol assay for sole, parallelism betweenthe standard curve and serial dilutions of plasma samples was established(data not shown).

2.5. RNA extraction

Total RNA was extracted from brain tissues (n = 7 for each experi-mental group) using Trizol RNA isolation reagent (Gibco/BRL) accord-ing to the manufacturer’s instructions. DNase digestion (45 min, 37 �C,2 U; Ambion, Austin, TX) was performed to eliminate genomic DNAcontamination. To confirm the absence of genomic DNA, the sameamount of total RNA as that used for the RT-PCR was added to theRT reaction mixture without reverse transcriptase and was subsequentlyamplified to confirm the absence of genomic contamination. Quantityand purity of the RNA was assessed spectrophotometrically at 260 and280 nm, and integrity was confirmed by electrophoresis through 1%agarose gels stained with ethidium bromide.

2.6. Rapid amplification of cDNA ends (5 0/3 0 RACE)

The 3 0 RACE-PCR was used to generate full-length cDNA in the 3 0

direction using 3 0 RACE System for Rapid Amplification of cDNA Ends(Roche Diagnostic). First-strand cDNA synthesis was initiated at thepoly(A)+RNA using the oligo(dT) anchor primer (Roche Diagnostic) in25 ll of solution containing: 10 mM dNTP, 400 U RevertAid TM HMinus M-MuLV Reverse, 10· RT buffer and 20 U RNase inhibitor.The RT reaction mixture was incubated at 42 �C for 90 min, followedby 10 min at 92 �C. The cDNA was purified using High Pure PCR purifi-cation Kit (Quiagen), following the manufacturer’s instructions. An ali-quot of cDNA was amplified using 12.5 mM of 3 0 anchor primer and12.5 mM of POMC fish 2 fw primer (5 0-ATG TGT CCT GCG TGGCTA TT-3 0) in a 25-ll PCR mixture containing: 12.5 ll of 2· PCR mix(Promega). PCR was carried out for 40 cycles as follows: 94 �C for 30 sfor denaturing; 47 �C for 25 s for primer annealing, and 72 �C for 30 sfor extension, followed by 1 cycle of 7 min at 72 �C for extension. The finalPCR product (500 bp) was electrophoresed in 1.5% agarose gels, and frag-ments were isolated, ligated, cloned into pGEM-T Easy Vector (Promega),and sequenced.

Primers used for the different amplifications are shown in Table 1.

Table 1Sequences of primers used for amplification and sequencing of POMCcDNA

Primer name Sequences

POMC fish 2 fw 5 0-ATG TGT CCT GCG TGG CTA TT-30

POMC Solea 1 5 0-TAA CCG CGT CTC TCC TGA CT-30

POMC Solea 2 5 0-CTG AGG TCA GAG CGA CAC AG-3 0

POMC Solea 3 fw 5 0-AGA GAG GGC AGT GTG GAG AA-3 0

POMC Solea 3 re 5 0-GTC AGC TCG TCG TAG CGT TT-30

F. Palermo et al. / General and Comparative Endocrinology 155 (2008) 227–233 229

For 5 0 RACE-PCR, cDNA from brain mRNA was obtained using 5 0

POMC Solea 1 (5 0-TAA CCG CGT CTC TCC TGA CT-3 0), and cDNAswere polyadenylated at the 3 0 end using terminal transferase (Roche). Thepolyadenylated cDNAs were then amplified by PCR using 12.5 lM of oli-go(dT)-anchor primer and 12.5 lM of POMC Solea 2 primer (5 0-CTGAGG TCA GAG CGA CAC AG-30) in a 25 ll PCR mixture containing12.5 ll of 2· PCR master mix (Promega). The thermal profile for the 5 0

RACE reaction consisted of an initial 95 �C denaturation step for 5 minfollowed by 40 cycles of denaturation (94 �C for 30 s), annealing (57 �Cfor 25 s), extension (72 �C for 30 s) and a final extension step at 72 �Cfor 7 min. An aliquot of the PCR was analyzed by electrophoresis in a1.5% agarose gel in 0.5-strength TAE buffer (Tris–HCL 20 mM, pH 7.5,acetic acid 20 mM, and EDTA 0.5 mM, pH 8.0). PCR products were thenpurified using Minelute gel Extraction Kit (Qiagen), and �50 ng of PCRproducts were used to be cloned with pGEM-T vector system (Promega)and transformed into competent cells. After selection, clones werescreened for appropriate size inserts using restriction endonuclease Not I(Promega). Plasmid DNA was extracted using QIAprep Spin Miniprepkit (Qiagen). Extracted plasmid DNA was sequenced by using an auto-mated DNA sequencer ABI PRISM 310 (Perkin-Elmer).

2.7. Relative-quantitative reverse transcription-polymerase chain

reaction

Ten micrograms of total RNA were used to synthesize a first strandcDNA; the reaction mixture was made up to a final volume of 20 ll, con-taining 50 mM oligo(dT)18, 2 ll of 10· M-MuLV RT reaction buffer, 1 llribonuclease inhibitor (10 U), 4 ll dNTPs (2.5 mM each dATP, dCTP,dGTP and dTTP), and 100 U M-MuLV Reverse Transcriptase (Ambion,Austin, TX). The reaction was incubated at 44 �C for 90 min. Initial pilotexperiments were performed to determine the linear range of PCR and theratio of competimers to primers for the 18 S rRNA. The reactions werecarried out in a final volume of 25 ll containing: 12.5 ll PCR mastermix 2· (Promega), 12.5 lM Solea POMC 3 sense primer (5 0-AGA GAGGGC AGT GTG GAG AA-3 0), 12.5 lM Solea POMC 3 antisense primer(5 0-GTC AGC TCG TCG TAG CGT TT-3 0; Invitrogen Life Technolo-gies, Milan, Italy), 18 S primers:competimers (3/7). Cycling parameterswere as follows: 95 �C for 5 min, followed by 40 cycles of 95 �C for 30 s,50 �C for 20 s, and 72 �C for 30 s, with a final extension step of 72 �Cfor 7 min. POMC primers and 18 S primers:competimers generated frag-ments of 500 and 324 base pairs (bp), respectively. Gel images were sub-sequently captured using a digital camera EDAS 290 (Kodak), and thereaction products were analyzed using 1 D Image Analysis software(Kodak). Normalization was carried out by dividing the average valueof POMC by the average value of 18 S RNA in each group.

2.8. Statistical analysis

All data were analyzed using Sigma Stat 3.1 (Systat software, Inc).Data distributions were first examined for their fit to a normal distributionand homogeneity of variance using Kolmogorov–Smirnov and Levenemedian tests. One-way Analysis of Variance (ANOVA) was used to com-

pare results between groups, followed by Tukey post hoc test. Differencesbetween means were considered significant when P < 0.05. Results areexpressed as means + SE of data.

3. Results

3.1. Cloning of sole POMC gene

Multiple PCR strategies 3 0 RACE, RT-PCR, and 5 0

RACE were used in tandem to clone and sequence the solePOMC cDNA (Accession No. DQ845108, Fig. 1).Sequence analysis revealed that the full-length sole POMCcDNA is 738 bp in length.

This cDNA encodes, in length, a 5 0 UTR of 79 nucleo-tides, a 3 0 UTR of 98 nucleotides, and an open readingframe of 558 nucleotides. The open reading frame encodesa predicted protein of 186 residues. As shown in Fig. 1, theopen reading frame can be subdivided into functionaldomains commonly found in gnathostome POMCs: aputative signal sequence region (M1–V20), an N-terminalregion (T21–A83), an ACTH sequence an (A86– V120), agamma-LPH region (Q123–S159) a beta-MSH region(D143–S159) and a beta-endorphin sequence (Y162–A186). A comparison of the amino acid sequence of soleand other POMC sequences in fish species, showing theregions containing the highly conserved residues, is pre-sented in Fig. 2. The deduced amino acid sequence for solealpha-MSH has 93% sequence identity with that of Veras-per moseri, and the deduced amino acid sequence for solebeta-MSH has 79% sequence identity with that ofV. moseri. Furthermore, within the open reading framesof sole and V. moseri POMCs, there is 66% identity atthe nucleotide level.

3.2. Evaluation of POMC mRNA changes

Fig. 3A shows the representative image of the relative-quantitative RT-PCR of S. solea POMC (515 bp PCRproduct). The POMC mRNA was evaluated in the brainof 7 fish for each experimental group. Fig. 3B shows thatPOMC mRNA levels found in the brain of fish at stockingdensities of 50/m2 for 7 days were significantly (P < 0.05)higher than that found in those maintained at a stockingdensity of 250/m2. A similar trend of POMC mRNA wasfound in the brain of fish maintained at the same stockingdensities for 21 days.

3.3. Plasma cortisol changes

Cortisol was measured in the plasma of 7 fish for eachexperimental group. Fig. 4 shows that cortisol titers foundin the plasma of fish at stocking densities of 50/m2 for 7days were significantly (P < 0.05) lower than that foundin those maintained at a stocking density of 100/m2 and250/m2. When the experiment lasted 21 days, fish rearedat stocking densities of 50/m2 and 100/m2 had significantly

Fig. 1. Cloning of a full-length sole (Solea solea) POMC cDNA. The deduced amino acid sequence of sole cDNA POMC is shown in bold above thenucleotide sequence.

230 F. Palermo et al. / General and Comparative Endocrinology 155 (2008) 227–233

(P < 0.05) decreased plasma levels of cortisol compared tofish maintained at the higher stocking density (250/m2).

3.4. Changes in body weight

The relation between specific growth rate (SGR) andstocking density (SD) in fish/m2 is presented in Fig. 5. Itwas found that stocking density had no effect on mortality.Statistical analysis revealed an overall significant (P< 0.001) decrease of specific growth rate of fish at highstocking densities in both experimental times, 7 and 21days. The specific growth rate of fish at stocking densitiesof 50/m2 for 7 days was significantly (P < 0.05) higher thanthat found in those maintained at stocking densities of 250/m2; moreover, there was a significant difference (P < 0.05)between the specific growth rate of fish at a stocking den-sity of 100/m2 and animals reared at one of 250/m2. A sim-ilar trend of SGR was found when the experiments lastedfor 21 days. In this case, significant differences (P < 0.05)

in specific growth rate were found between a stocking den-sity of 50/m2 and both the higher stocking densities, 100/m2 and 250/m2; on the contrary, no significant differencesin specific growth rate were revealed between fish main-tained at stocking densities of 100/m2 and 250/m2.

4. Discussion

POMC derived peptides play important roles in differentphysiological function, and they have been found to beinvolved in fish stress response (Mosconi et al., 2006).Many recent studies have demonstrated that a typical fea-ture of several neuropeptide and polypeptide hormone pre-cursors is the presence of multiple copies of the same endproduct flanked by sets of paired-basic amino acid proteo-lytic cleavage sites. The evolution of these types of precur-sors is the apparent result of duplication of DNA regionsthat code for the biologically active polypeptide (Doreset al., 1999).

Fig. 2. Alignment of the predicted amino acid sequence of sole (Solea solea) POMC with other POMC sequences of fish species, Verasper moseri

(AB051424), Paralichthys olivaceus (AF191593) and Thunnus obesus (AB020971). Gaps introduced for optimal alignment are indicated by dashes. (*)Identical or conserved residues in all sequences, (:) conserved substitutions and (.) semi-conserved substitutions.

POMC

18S

250/m2 100/m2 50/m2MW MW 250/m2 100/m2 50/m2

POMC

18S

(7d) (21d)

0

0.5

1

1.5

2d17dtime (days)

PO

MC

/18S

(ar

bit

rary

un

its) a ab b a ab b

250/m2

100/m2

50/m2

A

B

Fig. 3. (A) Image representative of the relative-quantitative RT-PCR ofSolea solea POMC (515 bp PCR product). (B) POMC expression levelsmeasured by relative-quantitative RT-PCR in Solea solea reared at 250,100, and 50 fish/m2 over two periods (7 days; 21 days). The 18S RNA(324-bp PCR product) was used as control standard. For the differentgroups, bands were quantified by image analysis, while values (means+ SE; n = 7) are represented by histograms. Stocking densities that share adifferent letter for a given time are significantly different from each other(P < 0.05).

0

1

2

3

4

5

7d 21d time (days)

pla

sma

cort

iso

l (n

g/m

l)

a a ba b b

250/m2

100/m2

50/m2

Fig. 4. Changes in plasma cortisol levels in Solea solea reared at 250, 100and 50 fish/m2 over two periods (7 days; 21 days). Data representmeans + SE; n = 9. Stocking densities that share a different letter for agiven time are significantly different from each other (P < 0.05).

0

1

2

3

7d 21dtime (days)

SG

R (

%B

W/d

)

250/m2

100/m2

50/m2

a b ba a b

Fig. 5. The effect of stocking density (SD) on the specific growth rate(SGR) of Solea solea reared at 250, 100 and 50 fish/m2 over two periods (7days; 21 days). Data represent means + SE; n = 13. Stocking densities thatshare a different letter for a given time are significantly different from eachother (P < 0.05).

F. Palermo et al. / General and Comparative Endocrinology 155 (2008) 227–233 231

The sole POMC cDNA that we cloned has end productregions flanked by sets of basic amino acids, an ACTHsequence with an alpha-MSH sequence, a beta-MSHsequence, and a beta-EP sequence; thus it is very similarto POMC subtype ‘‘A’’ that Takahashi et al. (2005) found

in barfin flounder (V. moseri). It is important to emphasizethe absence of the gamma-MSH sequence, because this fea-ture sets sole POMC apart from the general organizational

232 F. Palermo et al. / General and Comparative Endocrinology 155 (2008) 227–233

scheme for this precursor protein. However, further studiesare needed to clarify the exact duplication event that gaverise to the gamma-MSH sequence.

The gamma-MSH sequence and gamma-MSH-relatedend products, flanked by sets of paired basic amino acidproteolytic cleavage sites, are present in the pituitaries oftetrapods (Dores et al., 1997; Douglass et al., 1984; VanStrien et al., 1996). Among non-tetrapod vertebrates, thepresence of gamma-MSH sequences is more variable; infact, this sequence is absent in jawless fish, like the lamprey,but it is present in gnathostomes, like cartilaginous fish (thespiny dog-fish) and stem ray-finned fish like sturgeon(Amemiya et al., 1997; Alrubaian et al., 1999) and gar(Dores et al., 1997). In teleost fish like salmon and trout(Kawauchi et al., 1981; Kitahara et al., 1988; Salbertet al., 1992; Okuta et al., 1996), carp (Arends et al.,1998), or tilapia (Lee et al., 2000), the entire gamma-MSH sequence has been deleted. These observations sug-gest that the gamma-MSH sequence was present in ances-tral gnathostomes, but in all lineages of extant ray-finnedfishes this POMC region was accumulating progressivemutations that resulted in the elimination of the gamma-MSH sequence from teleost POMC (Dores et al., 1999).

In the present study, the effects of stocking density onPOMC gene expression were also investigated, by measur-ing POMC mRNA levels together with plasma cortisol, thekey hormone related with stress response and, in turn, withthe specific growth rate. In fact, Schram et al. (2006), dem-onstrated that stocking density influences growth rate inthe Dover sole (S. solea), using sole juveniles stocked atincreasing densities from 8% up to 195% bottom coverage.Even if the stocking densities here considered were lowerthan that applied in the case of the Dover sole, the specificgrowth rate was influenced in both experimental times, 7and 21 days.

Fish reared at high stocking densities of 250/m2 showedhigher plasma cortisol concentrations than fish reared atdensities of 100/m2 and 50/m2. This result agrees with sev-eral other studies that show a similar cortisol response pat-tern in different fish species (Rotllant et al., 1997; Monteroet al., 1996; Mosconi et al., 1998, 2006). The increase inplasma cortisol levels contributes to stress-related growthsuppression (Pickering, 1990), and our data demonstratethat the elevation of cortisol levels was closely correlatedwith a significant specific growth rate reduction in fishreared at a density of 250/m2, both for 7 days and 21 days.Moreover, our results indicate that a control mechanismfor the regulation of POMC mRNA expression was acti-vated in the sole according to the different stockingdensities.

In a general stress response, ACTH release increasescortisol circulating concentrations; an elevation in ACTHlevel could be an indicator of an elevated POMC mRNAexpression. In contrast, it was found that the higheststocking density induced a significant decrease in solePOMC mRNA transcriptional rate. Expression of thePOMC gene is positively regulated by CRF and is under

negative feedback control by glucocorticoids through glu-cocorticoid responsive elements (Autelitano et al., 1989;Eberwine et al., 1987). Numerous studies have previouslydemonstrated that cortisol may exert negative feedbackeffects on the CRF-synthesizing neurons of the nucleuspreopticus (NPO) in teleosts (Fryer and Peter, 1977;Olivereau and Olivereau, 1989). The negative feedbackeffects of cortisol on CRF gene expression are mediatedby glucocorticoid receptors (GR) and can significantlyreduce the stress-induced activation of the HPI axis (Ber-nier et al., 1999). Several studies support the down regu-lation of GR by cortisol in fish (Maule and Schreck,1991; Pottinger, 1990; Shrimpton and Randall, 1994).Parallel to a cortisol mediated suppression of CRF syn-thesis, several other feedback effects of cortisol on theHPI axis may contribute to the regulation of cortisolsecretion; for example, among these feedback loops, Fryeret al. (1984) showed the direct inhibitory effect of cortisolon ACTH secretion from the goldfish pituitary; therefore,the decrease of POMC mRNA can be considered anadaptive response of the fish to farm stocking densityconditions.

In conclusion, our study provides, for the first time, thegenomic structure of the POMC cDNA in the teleost S.

solea, of high commercial interest for Mediterranean aqua-culture; moreover, evidence was also found that high rear-ing densities affect POMC gene expression, whose levels inthe brain seem to be down regulated by high peripheralcortisol levels.

Acknowledgments

This research was supported by Ministero delle PoliticheAgricole e Forestali (MIPAF), and Agenzia Regionale perlo Sviluppo e l’Innovazione nel settore Agro-forestale dellaregione Toscana (ARSIA). We are very grateful to Dr.Roberto Salvatori (Orbetello Pesca Lagunare) for his careand maintenance of the experimental fish.

References

Alrubaian, J., Danielson, P., Fitzpatrick, M., Schreck, C., Dores, R.M.,1999. Cloning of a second proopiomelanocortin cDNA from thepituitary of the sturgeon, Acipenser transmontanus. Peptides 20, 431–436.

Amemiya, Y., Takahashi, A., Dores, R.M., Kawauchi, H., 1997. Sturgeonproopiomelanocortin has a remnant of gamma-melanotropin. Bio-chem. Biophys. Res. Commun. 230, 452–456.

Arends, R.J., van der Gaag, R., Martens, G.J., Wendelaar Bonga, S.E.,Flik, G., 1998. Differential expression of two pro-opiomelanocortinmRNAs during temperature stress in common carp (Cyprinus carpio

L.). J. Endocrinol. 159 (1), 85–91.Autelitano, D.J., Snyder, L., Sealfon, S.C., Roberts, J.L., 1989. Dopamine

D2-receptor messenger RNA is differentially regulated by dopaminer-gic agents in rat anterior and neurointermediate pituitary. Mol. Cell.Endocrinol. 67, 101–105.

Balm, P.H.M., Pottinger, T.G., 1995. Corticotrope and melanotropePOMC-derived peptides in relation to interrenal function during stressin rainbow trout (Oncorhynchus mykiss). Gen. Comp. Endocrinol. 98,279–288.

F. Palermo et al. / General and Comparative Endocrinology 155 (2008) 227–233 233

Bernier, N.J., Lin, X., Peter, R.E., 1999. Differential expression ofcorticotropin-releasing factor (CRF) and urotensin I precursor genes,and evidence of CRF gene expression regulated by cortisol in goldfishbrain. Gen. Comp. Endocrinol. 116, 461-47.

Dores, R.M., Smith, T.R., Danielson, P., Marra, L.E., Youson, J.H.,1997. Deciphering post translational processing events in the pituitaryof a neopterygian fish: cloning of a gar proopiomelanocortin cDNA.Gen. Comp. Endocrinol. 107, 401–413.

Dores, R.M., Sollars, C., Danielson, P., Lee, J., Alrubaian, J., Joss, J.M.,1999. Cloning of a proopiomelanocortin cDNA from the pituitary ofthe australian lungfish, Neoceratodus forsteri: analyzing trends in theorganization of this prohormone precursor. Gen. Comp. Endocrinol.116, 433–444.

Douglass, J., Civelli, O., Herbert, E., 1984. Polyprotein gene expression:generation of diversity of neuroendocrine peptides. Ann. Rev.Biochem. 53, 665–715.

Eberwine, J.H., Jonassen, J.A., Evinger, M.J., Roberts, J.L., 1987.Complex transcriptional regulation by glucocorticoids and corticotro-pin-releasing hormone of proopiomelanocortin gene expression in ratpituitary cultures. DNA 6, 483–492.

Flik, X.J., Stouthart, F.A., Spanings, R.A., Lock, J.C., Fenwick,Wendelaar Bonga, S.E., 2002. Stress response to waterborne Cuduring early life stages of carp, Cyprinus carpio. Aquat. Toxicol. 56,167–176.

Fryer, J.N., Peter, R.E., 1977. Hypothalamic control of ACTH secretionin goldfish. III. Hypothalamic cortisol implants studies. Gen. Comp.Endocrinol. 33, 215–225.

Fryer, J., Lederis, K., Rivier, J., 1984. Cortisol inhibits the ACTH-releasing activity of urotensin I, CRF and sauvagine observed withsuperfused goldfish pituitary cells. Peptides 5, 925–930.

Gonzalez-Nunez, V., Gonzalez-Sarmiento, R., Rodriguez, R.E., 2003.Identification of two proopiomelanocortin genes in zebrafish (Danio

rerio). Brain Res. Mol. 120, 1–8.Gornati, R., Terova, G., Vigetti, D., Prati, M., Saroglia, M., Bernardini,

G., 2004. Effects of population density on seabass (Dicentrarchus

labrax L.) gene expression. Aquaculture 230, 229–239.Hansen, I.A., To, T.T., Wortmann, S., Burmester, T., Winkler, C., Meyer,

S.R., Neuner, C., Fassnacht, M., Allolio, B., 2003. The pro-opiomel-anocortin gene of the zebrafish (Danio rerio). Biochem. Biophys. Res.Commun. 303, 1121–1128.

Karsi, A., Waldbieser, G.C., Small, B.C., Wolters, W.R., 2005. Genomicstructure of the proopiomelanocortin gene expression during acute low-water stress in channel catfish. Gen. Comp. Endocrinol. 143, 104–112.

Kawauchi, H., Takahashi, A., Abe, K., 1981. Gamma-Melanotropin isnot present in an N-terminal peptide of salmon proopiomelanocortin.Int. J. Peptide Protein Res. 18, 223–227.

Kitahara, N., Nishizawa, T., Iida, K., Okazaki, H., Andoh, T., Soma, G.,1988. Absence of a g-melanocyte-stimulating hormone sequence inproopiomelanocortin mRNAof chum salmon, Onchorhynchus keta.Biochem. Biophys. Res. Commun. 91b, 365–370.

Lee, J., Danielson, P., Sollars, C., Alrubaian, J., Balm, P., Dores, R.M.,2000. Cloning of a Neoteleost (Oreochromis mossambicus) POMCcDNA reveals a deletion of the gamma-MSH region and most of thejoining peptide region: Implications for POMC processing. Peptides20, 1391–1399.

Maule, A.G., Schreck, C.G., 1991. Stress and cortisol treatment changedaffinity and number of glucocorticoid receptors in leucocytes and gillof coho salmon. Gen. Comp. Endocrinol. 84, 83–93.

Montero, D., Izquierdo, M., Tort, L., Robaina, L., Vergara, J.M., 1996.High stocking density produces crowding stress altering some phys-iological and biochemical parameters in gilthead seabream, Sparus

aurata, juveniles. Fish Physiol. Biochem. 20, 53–60.Mosconi, G., Cardinaletti, G., Carotti, M., Palermo, F., Soverchia, L.,

Polzonetti-Magni, A.M., 2006. Neuroendocrine mechanisms regulat-ing stress response in cultured teleost species. In: Reinecke, M.,Zaccone, G. (Eds.), Fish Endocrinology, vol. 2. IBH PublishingCompany, Oxford.

Mosconi, G., Gallinelli, A., Polzonetti-Magni, A.M., Facchinetti, F.,1998. Acetyl salmon endorphin-like and interrenal stress response inmale gilthead sea bream, Sparus aurata. Neuroendocrinology 68, 129–134.

Okuta, A., Ando, H., Ueda, H., Urano, A., 1996. Two types of cDNAsencoding proopiomelanocortin of sockeye salmon, Onchorhynchus

nerka. Zool. Sci. 13, 421–427.Olivereau, M., Olivereau, J., 1989. Quantitative changes of CRF-like

immunoreactivity in the brain and in the pituitary of teleost fish.Peptides 9, 1261–1267.

Pickering, A.D., 1990. Stress and the suppression of somatic growth inteleost fish. Prog. Clin. Biol. Res. 342, 473–479.

Pottinger, T.G., 1990. The effect of stress and exogenous cortisol onreceptor-like binding of cortisol in the liver of rainbow trout,(Oncorhynchus mykiss). Gen. Comp. Endocrinol. 78, 194–203.

Rotllant, J., Balm, P.H., Perez-Sanchez, J., Wendelaar-Bonga, S.E., Tort,L., 2001. Pituitary and interrenal function in gilthead sea bream(Sparus aurata L., Teleostei) after handling and confinement stress.Gen. Comp. Endocrinol. 121, 333–342.

Rotllant, J., Pavlidis, M., Kentouri, M., Abad, M.E., Tort, L., 1997. Non-specific responses in the red porgy Pagrus pagrus after crowding stress.Aquaculture 156, 279–290.

Salbert, G., Chauveau, I., Bonnec, G., Valotaire, Y., Jego, P., 1992. Oneof the two trout proopiomelanocortin messenger RNAs potentiallyencodes new peptides. Mol. Endocrinol. 6, 1605–1612.

Schram, E., Van der Heul, J.W., Kamstra, A., Verdegem, M.C.J., 2006.Stocking density-dependent growth of Dover sole (Solea solea).Aquaculture 252, 339–347.

Shrimpton, J.M., Randall, D.J., 1994. Downregulation of corticosteroidreceptors in gills of coho salmon due to stress and cortisol treatment.Am. J. Physiol. 267, 432–438.

Takahashi, A., Amano, M., Itoh, T., Yasuda, A., Yamanome, T.,Amemiya, Y., Sasaki, K., Sakai, M., Yamamori, K., Kawauchi, H.,2005. Nucleotide sequence and expression of three subtypes ofproopiomelanocortin mRNA in barfin flounder. Gen. Comp. Endo-crinol. 141, 291–303.

Van Strien, F.J., Devreese, B., Van Beeumen, J., Roubos, E.W., Jenks,B.G., 1996. Biosynthesis and processing of the N terminal part of pro-opiomelanocortin in Xenopus laevis: characterization of gamma-MSHpeptides. J. Neuroendocrinol. 7, 807–815.

Winberg, S., Lepage, O., 1998. Elevation of brain 5-HT activity, POMCexpression, and plasma cortisol in socially subordinate rainbow trout.Am. J. Physiol. 274, 645–654.