Certain fish have the remarkable capability of euryhalinity, being able to withstand large variations in salinity for indefinite periods. Using the highly euryhaline species, silver sea bream (Sparus sarba), as an experimental model, some of the molecular processes involved during ion regulation (Na+-K+-ATPase), cytoprotection [heat shock protein (hsp) 70], and growth (somatotropic axis) were studied. To perform these studies, seven key genes involved in these processes were cloned, and the tissue-specific expression profiles in fish adapted to salinities of 6 parts per thousand (ppt; hypoosmotic), 12 ppt (isoosmotic), 33 ppt (seawater), and 50 ppt (hypersaline) were studied. In gills, the transcriptional and translational expression profiles of Na+-K+-ATPase α- and β-subunit genes were lowest in isoosmotic-adapted fish, whereas in kidneys the expression of the β-subunit increased in seawater- and hypersaline-adapted groups. The hsp70 multigene family, comprising genes coding for heat shock cognate (hsc70), inducible heat shock protein (hsp70), and a heat shock transcription factor (hsf1), was found to be highly upregulated in gills of seawater- and hypersaline-adapted fish. In liver, hsc70 expression was lowest in isoosmotic groups, and in kidneys the hsp70 multigene family remained unchanged over the salinity range tested. The regulation of the somatotropic axis was studied by measuring pituitary growth hormone expression and liver IGF-I expression in salinity-adapted fish. The expression amounts of both genes involved in the somatotropic axis were highest in fish maintained at an isoosmotic salinity. The results of this study provide new information on key molecular processes involved in euryhalinity of fish.
- sodium pump
- heat shock protein 70
the euryhaline capability of fish has evolved continuously from most teleost taxonomic groups, and this could be a reason as to why fish can be found inhabiting almost all aquatic systems (32). As euryhaline fish adapt to different salinities they have to regulate a number of key physiological functions that are necessary for homeostasis. Of critical importance is the maintenance of ion regulation, since in low-salinity conditions fish would be prone to severe ion depletion and increased water entry, whereas in seawater or hypersaline conditions fish have to contend with the possibility of excess ion intrusion and osmotic water loss. Also, conditions of varying salinity impose an osmotic stress on fish, and as a result processes involved in growth and development can be severely restricted (12). It is necessary that euryhaline fish must possess a series of key molecular responses to cope with demands for ion regulation control, stress response, and growth maintenance during salinity adaptation. The Sparidae (sea bream) have been proven to display the capacity to survive in environments of large variations in salinity (26, 51, 58), and as such this group of fish serves as an ideal model for studies concerning gene expression during salinity adaptation.
Fundamental to osmoregulation and ion exchange is Na+-K+-ATPase, an enzyme found abundantly in osmoregulatory organs such as gill and kidney. The role of Na+-K+-ATPase in osmoregulatory organs of fish is to actively pump K+ into and Na+ out of a cell, against a concentration gradient (19, 24), and the activity of this enzyme also provides the driving force for several other transport processes associated with osmoregulatory function (36, 40). The structure of the Na+-K+-ATPase enzyme molecule is heterodimeric, consisting of a catalytic α-subunit and a glycosylated β-subunit, both of which are encoded for by separate genes. The Na+-K+-ATPase α-subunit contains all of the catalytic domains necessary for a functional enzyme (33, 43), and the Na+-K+-ATPase β-subunit is critical for maintaining the stability and proper membrane orientation of the enzyme (41, 44). The underlying mechanism of Na+-K+-ATPase regulation, during salinity adaptation, is complex and could involve regulation at subunit transcriptional, subunit translational, and functional enzyme formation. In addition, Na+-K+-ATPase subunit genes may also be regulated differently in osmoregulatory tissues such as gill and kidney, and this may reflect the different roles of Na+-K+-ATPase in these tissues. A few studies have examined the expression of Na+-K+-ATPase genes during salinity adaptation or seawater transfer. Expression of the Na+-K+-ATPase α- and β-subunit genes was highest in seawater-acclimated European eels compared with those maintained in freshwater (6, 7). In brown trout, Na+-K+-ATPase α-subunit expression was increased during seawater transfer (34) and was found to be highest during both freshwater and hypersaline acclimation of European sea bass (25). Increased environmental salinity also caused an elevation in two isoforms of the Na+-K+-ATPase α-subunit in tilapia gills (18).
Alterations in salinity can result in aquatic organisms becoming osmotically stressed, and as a consequence of such stress cellular ion regulation can be adversely affected, eventually causing disruption to protein synthesis and protein damage (12). In situations such as osmotic stress, proteins belonging to the heat shock protein (hsp) families are rapidly synthesized (23). The major hsp family induced during stress is the hsp70 family, and these proteins perform an important cytoprotective function helping to refold and repair damaged proteins (20). The hsp70 multigene complex consists of a cognate type (hsc70) and an inducible type (hsp70). In cells, hsc70 is found to be expressed constitutively, and during exposure to stress the expression of this gene generally remains unchanged or slightly upregulated, as previously reported for silver sea bream (10). During stressful conditions, resulting in protein damage, the hsp70 gene is highly induced, and the transcription of hsp70 is mediated via the binding of heat shock factor 1 (HSF1) to a heat shock promoter (56). From studies on bacteria and plants, it has been implied that the hsp70 family may protect cells against the damaging effects of osmotic stress (22, 54), and this family of proteins may also play a critical role in salinity tolerance of fish when exposed to osmotic stress, such as alterations in environmental salinity (13, 12). A clear understanding of how the hsp70 multigene family is regulated, when fish are exposed to different salinities, would aid our understanding as to the plasticity of the stress response and may shed some insights into how certain euryhaline fish are capable of withstanding large variations in salinity, without detrimental cellular protein damage.
The most important system for controlling fish growth and development is the somatotropic [growth hormone (GH)-insulin-like growth factor I (IGF-I)] axis (3). Studies concerning this axis have mainly used salmonids as experimental models (2, 15, 16), and it is apparent that a functional GH-IGF-I axis does exist and fits the somatomedin hypothesis (14), whereby GH promotes increased hepatic IGF-I expression. Environmental factors are known to affect the somatotropic axis in fish (2, 12, 15), and it has been proposed that measurements concerning this axis could provide for an integrated signal, indicative of favorable or unfavorable conditions for fish growth (2). Such an approach has been used for chinook salmon whereby higher levels of IGF-I were found in fish maintained at a warm temperature and fed more food than those maintained at an ambient temperature and fed less food (2). An upregulated somatotropic axis was found in tilapia maintained in seawater compared with tilapia maintained in freshwater, and this axis was further augmented upon methyltestosterone treatment (46). Also, salinity-adapted juvenile black sea bream had higher hepatic IGF-I mRNA expression levels during chronic isoosmotic salinity adaptation (12), an effect that was correlated with an enhanced growth rate (26).
In the present study the euryhaline silver sea bream (Sparus sarba) was used as an experimental model to investigate molecular processes involved in ion regulation (Na+-K+-ATPase), stress response (hsp70 multigene family), and growth (somatotropic axis). The tissue-specific expression of genes involved in these processes was profiled from silver sea bream adapted to a wide range of salinities. To our knowledge, this is the first study to comprehensively address the expression of key genes involved in salinity-adaptive processes from a euryhaline fish.
MATERIALS AND METHODS
Experimental fish and salinity adaptation.
Silver sea bream (S. sarba) weighing between 150 and 200 g were purchased from a local fish farm and were randomly divided into four groups (n = 7) and maintained in separate seawater [33 parts per thousand (ppt)] tanks. Fish were maintained in these tanks for 1 mo before salinity adaptation. In groups that were designated for low-salinity adaptation, salinity was reduced via a gradual dilution of seawater with dechlorinated tap water, over a period of 1 wk, until the final experimental salinities were reached, i.e., 6 ppt (hypoosmotic) and 12 ppt (isoosmotic). Hypersaline water was obtained by evaporating seawater to a salinity of 50 ppt, in a separate tank, away from the main culture tanks. Hypersaline conditions were obtained by gradual flushing of seawater with hypersaline water, over a period of 1 wk until a final salinity of 50 ppt was obtained. The water in each culture tank was fully aerated, and the temperature ranged from 22 to 24°C throughout the experimental period. Fish were fed daily with a formulated diet (58), and the culture period of fish maintained at hypoosmotic, isoosmotic, seawater, and hypersaline conditions was 1 mo.
Cloning of Na+-K+-ATPase subunit and hsp genes.
Silver sea bream gill, kidney, and liver cDNA libraries were constructed using a SMART cDNA library construction kit (Clontech). Approximately 106 plaques of each library were transferred to Hybond NX plaque lifts (Amersham Pharmacia Biotech) and prepared for screening according to instructions supplied with plaque lifts. For Na+-K+-ATPase subunit gene isolation, plaque lifts from gill and kidney libraries were prehybridized for 4 h, at 60°C in Rapid-Hyb buffer (Amersham), and then hybridized at 65°C for 16 h with [32P]dCTP-labeled Na+-K+-ATPase α- or β-subunit gene clones (11). For hsp70 gene studies, plaque lifts from gill, kidney, and liver libraries were hybridized with either a DNA fragment previously identified as an hsc70 clone from sea bream (10) or a 457-bp hsp70 DNA fragment isolated using specific PCR primers designed from the zebrafish hsp70 gene (45). After hybridization, plaque lifts were washed two times in 2× SSC (1.8% wt/vol sodium chloride, 0.9% wt/vol trisodium citrate)-0.1% SDS for 30 min at 65°C and then one time in 0.1% SSC-0.1% SDS for 30 min at 65°C. Membranes were exposed to X-ray film (Kodak) overnight, and positive plaques were isolated and subjected to further rounds of screening. After three rounds of screening, 10 putative clones for each gene, from each of the libraries, were selected and converted to plasmid by in vivo excision using Escherichia coli strain BM25.8 (Clontech). Plasmids were cycle sequenced using and ABI PRISM dye terminator kit with reaction products run on an ABI 310 Genetic Analyzer (Perkin-Elmer). The clones were sequenced on both strands, and sequence data were analyzed using the Basic Local Alignment Search Tool Program (1). The complete gene sequences for cloned genes have been submitted to the GenBank under accession numbers AY553205 (Na+-K+-ATPase α-subunit), AY553206 (Na+-K+-ATPase β-subunit), AY436786 (hsc70), and AY436787 (hsp70).
HSFI fragment isolation.
With the use of degenerate oligonucleotides, constructed from highly conserved regions of the hsf1 DNA-binding domain (AFLTKLWTLV and DDTEFQHP), a sea bream hsf1 DNA fragment was isolated. For this purpose, total RNA was extracted from gill, kidney, and liver tissues using an RNeasy mini kit (Qiagen), treated with DNase I, and then quantified spectrophotometrically. For first-strand cDNA synthesis, 1 μg total RNA from each sample was added to a reaction mix (20 μl) containing 0.5 μg oligo(dT) primer (Pharmacia), 2 μl dithiothreitol (0.1 M), 1 μl dNTP mix (10 mM; Pharmacia), 4 μl reaction buffer, and 1 μl Superscript II RT (200 U/μl; GIBCO-BRL). First-strand cDNA synthesis was allowed to proceed at 42°C for 1 h, after which time the reaction was incubated at 70°C for 15 min. For amplification of hsf1, PCR reactions (50 μl) containing 2 μl first-strand cDNA, 0.2 μl Taq DNA polymerase (5 U/μl; Promega), 5 μl MgCl2 (25 mM), 5 μl reaction buffer, 0.5 μl dNTP mix (10 mM), and 1 μl of each primer (50 μM) were prepared. PCR amplification was performed using a PTC-I00 thermal cycler (MJ Research) with cycle parameters of 94°C for 1 min, 45°C for 1 min, 72°C for 1 min, and a final extension of 72°C for 4 min. Reaction products were analyzed on a 1.4% wt/vol agarose gel and visualized by ethidium bromide staining. Putative gene fragments were subcloned using a Topo TA cloning kit (InVitrogen) and cycle sequenced as previously described. The hsf1 DNA-binding domain sequence has been deposited on the GenBank under accession number AY609319.
GH and IGF-I genes.
Clones for GH were isolated from pituitary first-strand cDNA in the same manner as described above, using primers that were previously employed in silver sea bream developmental studies (9), and the complete reading frame for GH was then obtained by a combination of 5′- and 3′-random amplification of cDNA ends using a SMART cDNA amplification kit (Clontech). Liver first-strand cDNA was used for isolation of an IGF-I gene fragment, in the same manner as above, using primers that were previously described (9, 12). The complete sequence for GH and the sequence of the IGF-I DNA fragment (∼40% of full length gene) have been deposited on the GenBank under accession numbers AY553207 and AY553208, respectively.
RT-PCR analysis of Na+-K+-ATPase subunits, hsc70, hsp70, hsf1, GH, and IGF-I.
Total RNA was extracted from gill, kidney, liver, and pituitaries, and first-strand cDNA was synthesized in the same manner as previously described. PCR amplification of first-strand cDNA was performed with a series of oligonucleotide primers designed from the nucleotide sequences of cloned genes. As a normalization control for each RT-PCR, primers specific for 18S rRNA were used (21). All primers were synthesized by Genset (Singapore) and had the following sequences: Na+-K+-ATPase α-subunit, 5′-AAGGCTATCCCTAAGGGGGTGGG-3′ (sense) and 5′-CATGTCAGTTCCCAGGTCCATAC-3′ (antisense); Na+-K+-ATPase β-subunit, 5′-TTCATCGGGACCATCCAAGCCAT-3′ (sense) and 5′-GCCGCCGATACCGTATACTTGAT-3 (antisense); hsc70, 5′-ATCAGTGATGACGACAA-3′ (sense) and 5′-TGACCCCCCCCCAGGGGC-3′ (antisense); hsp70, 5′-ATCAGTGAGGAGGACAAA-3′ (sense) and 5′-CTGGGAGCCGCTTCCTGC-3′ (antisense); hsf1, 5′-CCCCAGTGGAACCAGCTTCCATG-3′ (sense) and 5′-GGATGTTGGAATTCCGTGTCATC-3′ (antisense); GH, 5′-CTGGGCGTCTCTTCTCAGCCGAT-3′ (sense) and 5′-TGCCACCGTCAGGTAGGTCTCCA-3′ (antisense); IGF-I, 5′-AGTGCGATGTGCTGTATC-3′ (sense) and 5′-CAGCTCACAGCTTTGGAA-3′ (antisense); 18S, 5′-GCCAAGTAGCATATGCTTGTCTC-3′ (sense) and 5′-AGACTTGCCTCCAATGGATCC-3′ (antisense).
PCR amplification was performed as previously described with cycle parameters of 94°C for 1 min, 55°C for 1 min, 72°C for 1 min, and a final extension of 72°C for 4 min. To ensure amplification was at the midpoint of the linear phase of amplification, for each gene, preliminary RT-PCR reactions were performed. A single PCR product, of expected size, was obtained for each gene of interest, and these were subcloned into pCRscript plasmid vector (Stratagene) and cycle sequenced to confirm identity. No PCR products were detected from negative controls (reactions without RT added), and all PCR samples were stored at 4°C before further analysis.
Semiquantification of mRNA transcripts.
To confirm specificity of each RT-PCR and to establish stringent hybridization conditions for subsequent analysis, an aliquot (10 μl) was taken from a number of representative samples, and Southern blots were made according to described procedures (50). Blots were rinsed in 6× SSC, and the DNA was fixed to the membrane by ultraviolet cross linking. For hybridization, purified cDNA fragments of each gene were radiolabeled using a Rediprime II random labeling kit (Amersham) and used for membrane hybridization in Rapid-Hyb buffer (Amersham) at 55°C for 16 h. The membranes were then washed two times with a 2× SSC-0.1% SDS solution for 30 min, one time in 0.1× SSC-0.1% SDS at 65°C for 15 min, air-dried for 15 min, and then autoradiographed at −80°C using Hyperfilm (Amersham). From preliminary Southern hybridizations, it was established that probes were specific for corresponding amplified fragments from RT-PCR reactions. For semiquantification of transcripts, samples were analyzed together in a single hybridization using DNA dot blots that were prepared using a Bio-Dot microfiltration manifold (Bio-Rad). PCR amplification products (10 μl) of each sample were diluted 5-, 10-, 50-, and 100-fold to test for the linearity of detection during subsequent scanning procedures. Samples were prepared and blotted according to instructions supplied with Hybond-N+ membrane (Amersham). Membranes were hybridized and washed as described above and exposed to storage phosphor screens (Molecular Dynamics) for 3 h at room temperature, after which time the screens were scanned using the Storm PhosphorImaging system with ImageQuant software (Molecular Dynamics) for quantification of amplified fragment. The abundance of each specific gene fragment was normalized to the corresponding 18S abundance in all samples.
Preparation of Na+-K+-ATPase α- and β-subunit polyclonal antibodies and immunoanalysis.
A 576-bp region of the sea bream Na+-K+-ATPase α-subunit gene and a 558-bp region of the Na+-K+-ATPase β-subunit gene (11) were amplified with PCR primers containing restriction enzyme sites for BamH I and Hind III. After amplification, the fragments were incubated in restriction enzyme mix at 37°C overnight and then ligated, in frame, into the expression vector pQE30 (Qiagen). Recombinant proteins were prepared using a Qiaexpress type IV kit (Qiagen), according to the supplier's instructions. The handling of rabbits, which were used for raising antibodies, was performed by trained research personnel, under license from the Hong Kong SAR Government [Animals (Control of Experiments) Ordinance Chapter 340]. For each antibody raised, 100 μg recombinant protein was mixed with 1 ml Freund's complete adjuvant and then administered intramuscularly into the hindleg region. After 3 wk, a second injection was given in the same manner as the first, and then after a further 3 wk a final injection of 100 μg recombinant protein in Freund's incomplete adjuvant was given. After the final injection, 1 ml of blood was collected weekly and assayed for antibody titer. Serum was collected from rabbits within 3–4 wk after the final injection and purified for IgG antibodies using a Hi Trap antibody purification system (Amersham). To confirm the specificity of each antibody, a set of preliminary experiments using immunoblots of gill and kidney total protein was probed with either polyclonal serum or polyclonal serum that had been premixed with 500 μg recombinant protein overnight to block epitope-specific antibodies. No immunoreaction was detected for blots probed with blocked antibodies. For protein extraction, gill and kidney samples were homogenized for 1 min in 2 ml extraction buffer (4 M urea, 0.5% wt/vol SDS, 10 mM EDTA, and 2 mM phenylmethylsulfonyl fluoride), using an Ultra-Turrax T25 rotor stator homogenizer. The samples were then transferred to 1.5-ml microcentrifuge tubes, incubated at 94°C for 10 min, sonicated for 10 min, and then centrifuged at 10,000 g for a further 10 min. Total protein was quantified using the dye-binding method of Bradford (4), and immuno dot blots were used to assess protein samples for subunit amounts using previously described procedures (12). The antibody dilution used for α- and β-polyclonal antibodies were 1:4,000, and the secondary antibody dilution used was 1:10,000. A Lumi-Imager workstation (Roche) was used for dot-blot exposure and development, and for each sample the optical density (OD) × area (mm2) was quantified using Lumi-Analyst 3.1 software (Roche).
Na+-K+-ATPase enzyme assay.
Samples of gill filaments and trunk kidney tissue were homogenized in ice-cold SEID buffer (150 mM sucrose, 10 mM EDTA, 50 mM imidazole, and 0.1% wt/vol sodium deoxycholate) and then centrifuged at 5,000 g for 1 min. An aliquot (50 μl) was used for Na+-K+-ATPase enzyme assay at 25°C according to the method of McCormick (39), and enzyme activity was calculated as specific activity (μmol NADH·min−1·mg protein−1).
Normalized transcript abundance for each gene studied, Na+-K+-ATPase subunit protein amounts, and Na+-K+-ATPase activity were subjected to a one-way ANOVA to test for significance, followed by a Student-Newman-Keul's test (Jandel scientific) to delineate significance between groups. Significant differences were accepted at P < 0.05.
Gill and kidney Na+-K+-ATPase expression.
With the use of an RT-PCR assay coupled with radioisotope hybridization methodology, the expression of both Na+-K+-ATPase α- and β-subunits was studied in gills and kidney of sea bream adapted to different salinities. The expression of both Na+-K+-ATPase subunit genes, in gills, was significantly lowest in sea bream adapted to an isoosmotic salinity, and the highest expression was found in gills of hypersaline-adapted sea bream (Fig. 1). To study the translated amount of Na+-K+-ATPase subunits, polyclonal antibodies were prepared and used for immunoblot detection. It was found that both Na+-K+-ATPase α- and β-subunit proteins were lowest in gills of sea bream adapted to an isoosmotic salinity, and the profile of subunit translation generally followed that of subunit transcription (Fig. 2). Transcriptional and translational studies of Na+-K+-ATPase expression were also performed on kidney samples taken from salinity-adapted sea bream. In the kidney, the amounts of Na+-K+-ATPase α-subunit mRNA remained relatively unchanged, whereas levels of Na+-K+-ATPase β-subunit mRNA were increased in seawater and hypersaline-adapted sea bream (Fig. 3). The translated amount of Na+-K+-ATPase α-subunit protein was also unchanged over the salinity range tested, whereas the Na+-K+-ATPase β-subunit increased in seawater- and hypersaline-adapted fish kidney (Fig. 4). To complete these studies, Na+-K+-ATPase enzyme activity was measured in both gills and kidneys of salinity-adapted sea bream. Gill Na+-K+-ATPase activity was lowest in isoosmotic-adapted sea bream, whereas in kidneys Na+-K+-ATPase was low in hypoosmotic- and isoosmotic-adapted sea bream but increased in seawater- and hypersaline-adapted sea bream (Fig. 5).
Tissue-specific expression of the hsp70 multigene family.
In the present study, the expression profiles of members of the hsp70 multigene family (hsc70, hsp70, and hsf1) were studied in liver, kidney, and gill of salinity-adapted sea bream. The expression profiles of the hsp70 multigene family in liver showed that both hsp70 and hsf1 expression remained unchanged over the salinity range tested, whereas hsc70 expression was found to be highest in hypersaline-adapted sea bream and significantly lowest in isoosmotic-adapted sea bream (Fig. 6). The expression profiles of members of the hsp70 multigene family, in kidneys, were not significantly different over the salinity range tested (Fig. 7). In gills, the expression levels for hsc70, hsp70, and hsf1 were not changed between hypoosmotic and isoosmotic conditions but increased during seawater and hypersaline adaptation (Fig. 8).
Somatotropic axis gene expression.
The expression of GH was found to be at the highest amount in pituitaries of isoosmotic-adapted sea bream. The levels of GH mRNA were lowest in sea bream maintained at hypoosmotic and hypersaline conditions (Fig. 9). The expression of liver IGF-I was at a significantly high level in isoosmotic-adapted sea bream but decreased substantially during hypersaline adaptation (Fig. 9).
The present study was performed to provide new insights into the molecular regulation of euryhalinity in fish. Silver sea bream was used as the model animal for molecular studies, since it has been previously established that this fish is an excellent osmoregulator capable of withstanding large variations in salinity (58). Previous studies on silver sea bream have shown that exposure to altered salinity environments initially cause transient changes in serum ions and tissue chemistry that stabilize after a few days (29). To understand the physiological responses associated with salinity acclimation, studies on sea bream have used chronic acclimation periods (12, 26, 51, 57, 58). In the present study, a 1-mo acclimation period was used, since we have previously established that silver sea bream are fully acclimated to changes in salinity environments within this time (28). For the parameters measured in the present study, data are not available as to the exact time when changes occur, although it is likely that these would change within the 1st wk of salinity acclimation, since previous studies on silver sea bream have demonstrated that branchial osmoregulatory functions change and stabilize within a few days of salinity transfer (28, 29). Fundamental to osmoregulation and ion exchange, in fish, is Na+-K+-ATPase, a membrane-bound heterodimeric enzyme that is abundantly present in osmoregulatory organs such as gills and kidneys. The molecular mechanisms involved in the formation of functional Na+-K+-ATPase during salinity adaptation of euryhaline fish could involve regulation at subunit transcriptional and subunit translational levels, enzyme phosphorylation, and membrane insertion. In addition, Na+-K+-ATPase subunit genes may also be regulated differentially in osmoregulatory tissues, during salinity adaptation, and this would reflect the different functions of Na+-K+-ATPase in these tissues. In the present study, the molecular processes involved in Na+-K+-ATPase expression were studied in gill and kidney only; however, the importance of intestinal Na+-K+-ATPase, in particular during water uptake, should not be overlooked.
Na+-K+-ATPase in gill.
With the use of RT-PCR assays, for assessment of subunit gene transcriptional activity, it was found that the expression of sea bream gill Na+-K+-ATPase α- and β-subunit genes was lowest in fish adapted to isoosmotic salinity. The amounts of both subunits were also assessed using immunoassays, and it was found that the translated subunit amounts showed the same profile as subunit gene transcription, the with lowest levels at an isoosmotic salinity. Also, measurements of Na+-K+-ATPase activity demonstrated the lowest levels in gills of isoosmotic-adapted sea bream. Molecular expression studies focused on Na+-K+-ATPase regulation at subunit transcriptional/translational levels have yet to be reported for other euryhaline fish, but it was demonstrated that European sea bass maintained at 15 ppt salinity had a lower abundance of gill Na+-K+-ATPase α-subunit compared with those in freshwater and hypersaline conditions (25). Presently, two models of gill Na+-K+-ATPase regulation during environmental salinity adaptation, in fish, have been proposed. The first and most widely reported is the “diadromid paradigm” model whereby a positive correlation between increasing salinity and elevated Na+-K+-ATPase occurs, and this model has been proven to fit killifish, tilapia, brown trout, rainbow trout, European eel, medaka, and Atlantic salmon (49). Over the past few years, contradictions to the well-established diadromid paradigm model have emerged, since Na+-K+-ATPase activity in several fish species is not positively correlated with environmental salinity (32). The gill Na+-K+-ATPase activity of some of those fish that do not fit into the diadromid paradigm model may instead be regulated via the numerical osmotic gradient between the external water and internal blood (25). In such situations a “U” shape of Na+-K+-ATPase activity is found whereby the lowest activity occurs as a consequence of the minimal requirement for ion exchange in conditions of isoosmotic (or near-isoosmotic) salinities. Data taken from the present study on sea bream show that gill Na+-K+-ATPase subunit transcription, translation, and enzyme activity all were lowest at isoosmotic salinity, probably because of the minimal requirement for active ion transport. From the molecular and biochemical data obtained in this part of the study, it is clear that silver sea bream does not fit the diadromid paradigm model and therefore should be placed alongside fish that display Na+-K+-ATPase regulation via an osmotic gradient between environmental salinity and blood.
Na+-K+-ATPase in kidney.
The fish kidney is also critical in maintaining internal homeostasis, and the molecular regulation of Na+-K+-ATPase activity was also studied in this organ from salinity-adapted sea bream. It was found that Na+-K+-ATPase α-subunit expression remained relatively unchanged, whereas the Na+-K+-ATPase β-subunit increased during seawater and hypersaline adaptation. The activity of Na+-K+-ATPase followed the same profile as the β-subunit, with an increase in activity in seawater- and hypersaline-adapted fish. The processes involved during subunit assembly and the establishment of stoichiometric Na+-K+-ATPase complexes, in fish kidney, are unknown, but it can be seen from the results of the present study that the Na+-K+-ATPase α-subunit is expressed at a relatively constant and adequate level in sea bream adapted between hypoosmotic to hypersaline conditions. Presently, it is difficult to account for the lack of correlation between α-subunit expression and Na+-K+-ATPase activity, seen in seawater- and hypersaline-acclimated sea bream, although an explanation may be related to the molecular approaches used. The RT-PCR assay employed in this study was specific for a single α-subunit gene and would not have been able to detect any different α-isoforms that may have been induced in seawater and hypersaline conditions. Different α-subunit isoforms have yet to be reported in fish kidneys. It would also appear that the increased expression of the Na+-K+-ATPase β-subunit is critical during the requirement for increased kidney Na+-K+-ATPase enzyme upon exposure to seawater or hypersaline conditions. The importance of the Na+-K+-ATPase β-subunit as a rate-limiting factor for the formation of Na+-K+-ATPase enzyme molecules has been proven from mammalian studies (17, 41). Presently, there is a paucity of data regarding comprehensive molecular studies on kidney Na+-K+-ATPase expression during salinity adaptation, although heterologous probing of brown trout kidney total RNA showed no change in expression of the Na+-K+-ATPase α-subunit during seawater transfer (34). The kidney Na+-K+-ATPase enzyme activity profile that was found during salinity adaptation is similar to that reported for black sea bream (26) and gilthead sea bream (51). These results would indicate a general phenomenon across the Sparidae and point to a key role for kidney Na+-K+-ATPase probably related to the production of concentrated urine and/or the requirement for ion exchange in environments of increased salinity.
hsp family expression in liver.
Alterations in salinity can result in aquatic organisms becoming osmotically stressed, and as a response to such stress cytoprotective proteins belonging to the hsp70 family can be induced. A clear understanding of how the hsp70 multigene family is regulated would aid our understanding as to the haloplasticity of the stress response and may provide some insights into how certain euryhaline fish can tolerate wide variations in salinity. Analysis of liver samples demonstrated that the expression of hsc70 was at a low level in isoosmotic-adapted sea bream and considerably increased in hypoosmotic and hypersaline conditions. The expression of liver hsp70 remained unchanged in sea bream liver over the salinity range tested. The regulatory mechanisms that control hsc70 expression in fish are not well understood, but emerging evidence suggests a key role for certain hormones in lowering hsc70. It has been previously shown that the expression of a putative hsc70 gene was reduced in liver of silver sea bream, which were treated with GH (10). In the present study, it was found that elevated pituitary GH mRNA was greatly increased during isoosmotic salinity (as discussed in GH-IGF-I axis) and such an elevation in GH may be related to the reduced hsc70 expression, although at present this conjecture can only be regarded as causal since further in vitro mechanistic studies on GH-hsc70 expression are required. Although hsc70 is expressed constitutively in silver sea bream liver, it is also upregulated during acute temperature stress, since a 2.3-fold elevation was previously reported (10). The increased expression of hsc70 in both hypoosmotic and hypersaline conditions was similar to that previously found in heat-shocked silver sea bream and would suggest that this gene can be induced under different conditions of abiotic stress. Presently, no other studies have reported on hsc70 gene expression in salinity-acclimated fish. However, some comparison can be made with juvenile black sea bream, which had increased amounts of proteins belonging to the hsp70 family during chronic hypoosmotic and hypersaline acclimation (12). The cellular signals that regulate hsc70 induction in liver of sea bream during hypoosmotic and hypersaline acclimation are unknown at present but are unlikely to be related to changes in plasma osmolality or ionic disturbances. Previous studies on sea bream have shown that plasma osmolality remains stable at ∼340–350 mosmol/kgH2O, and serum ions were also unaltered during salinity acclimation (29, 26, 57, 59). Elements of the osmosensing pathway, such as the protein kinases (30, 38), could be part of the mechanism responsible for increased liver hsc70 expression during hypoosmotic and hypersaline acclimation. The increased hsc70 expression in hypoosmotic- and hypersaline-adapted fish indicates that there is greater requirement for protein folding and/or chaperone activity in liver of fish adapted to salinity extremes.
hsp family expression in kidney and gills.
The expression of the hsp70 multigene family was also studied in osmoregulatory tissues, and it was found that kidney expression levels for hsc70 and hsp70 remained unchanged, whereas both of these genes were found to be greatly increased in gills taken from seawater- and hypersaline-adapted fish. Presently, a time course study for gill hsc70 and hsp70 induction in sea bream has yet to be performed, although it is likely that such increases should occur rapidly, and it has been shown that hyperosmotic exposure of Atlantic salmon gills caused hsp70 induction within 12 h (53). The gills of fish are most important for respiration, nitrogen excretion, and ion exchange, and to perform these functions efficiently they have to be continuously ventilated with water. As a consequence of the regular passage of water, gills have to always contend with environmental perturbations, such as changes in temperature and salinity, as well as anthropogenic factors, such as pollutants. The external position and the functioning of gills in fish may explain why this tissue displayed such a high induction of both hsc70 and hsp70 compared with kidney (and liver), both of which are internal, and would not be directly exposed to environmental stress. The prominent elevation of gill hsc70 and hsp70 transcription in conditions of increased salinity (seawater and hypersalinity) suggests that there is a threshold of salt tolerance for fish gills, and once surpassed then mechanisms involved in stress protein upregulation are activated. Evidence to support the effect of increased NaCl, as an inducer for the hsp70 family, has been provided from in vitro studies. Exposure of MDCK cells to increased NaCl resulted in hsp70 induction (5), and incubation of isolated Atlantic salmon gill lamellae in media made hyperosmolar with NaCl also caused a prominent hsp70 induction (53). The increased elevation of hsc70 and hsp70 was observed after long-term acclimation of sea bream in seawater or hypersaline conditions, and this would suggest that gill cells require a continuously high amount of hsc70 and hsp70 expression to prevent irreversible gill cell damage. Whereas the major roles of hsc70 and hsp70 are related to preventing protein damage, recent studies have demonstrated that hsc70 is also involved in chloride ion transport via trafficking of a cystic fibrosis transmembrane conductance regulator (CFTR; see Refs. 8 and 47). Because CFTR is known to be upregulated in euryhaline fish that are exposed to increased salinity (37), it is possible that the elevated hsc70 expression, during seawater and hypersaline acclimation of sea bream, may be intimately linked with chloride ion transport. Expression of hsf1 was found to increase only in gills of seawater- and hypersaline-adapted sea bream, an effect that followed the same profile as hsp70. It is likely that transcription of hsp70 is regulated via hsf1, since recent studies using fish and mammalian models have provided conclusive evidence for a regulatory link between these two genes (42, 55, 60). An elevated hsf1 transcription would be a critical determining factor for maintaining sufficient hsp70, and as such the combined upregulation of both of these genes could be one reason as to why gills of euryhaline fish have the capability to maintain cellular functions, even under severe stress such as hypersaline exposure. In the present study, we have selected genes (hsp70, hsc70, and hsf1) that are primarily related to temperature stress. Given that the expression of these genes remained unchanged in gills of sea bream adapted to dilute salinities, it is possible that different genes, not related to temperature stress, are induced and play a key role in maintaining hyperosmoregulatory function. Evidence to support this conjecture has been provided from studies on goby fish where temperature and hypoosmotic salinity acclimation have been shown to induce different sets of proteins (31).
In fish, growth is continuous and highly dependent on environmental factors, including salinity, which has been widely reported to influence growth of many fish species (3). The most important system for controlling fish growth and development is the somatotropic (GH-IGF-I) axis, and the final part of this study concerned an investigation into the expression of the genes involved in this axis. Pituitary GH mRNA amount has been used as a molecular indicator of the influence of salinity on fish growth (46, 52), and, with the use of such measurements, it was found that sea bream pituitary GH mRNA was lowest in sea bream maintained at hypoosmotic and hypersaline conditions but substantially increased in those maintained at an isoosmotic salinity. With the use of immunocytochemical and ultrastructural methods, it has also been shown that gilthead sea bream adapted to brackish water display activation of GH cells compared with seawater-adapted fish (35). It would seem likely that pituitary GH mRNA and protein levels are increased in sea bream species at salinities of isoosmotic or near isoosmotic. Apart from the key role of GH in anabolic processes, this hormone has also been proven to be critical toward maintaining hypoosmoregulatory function in some fish species, performing a role as a seawater-adapting hormone (40, 48). However, a role for GH in maintaining hypoosmoregulatory function of silver sea bream was not apparent from the results of the present study because a gradual increase in GH mRNA would have been detected as environmental salinity increased. Instead, a progressive decrease was found as salinity adaptation moved from isoosmotic to hypersalinity. Previous studies on silver sea bream showed that GH treatment of fish adapted to either hypoosmotic or seawater conditions did not affect the expression of Na+-K+-ATPase subunit genes or modulate enzyme activity (11, 27). From the aforementioned studies on silver sea bream, it appears that GH may not be a critical determinant toward maintaining hypoosmoregulatory function during increased environmental salinity. Measurements of liver IGF-I mRNA demonstrated that the highest amounts occurred in isoosmotic salinity-adapted sea bream. This finding was the same as that reported in black sea bream (12) and would suggest a common response across this range of species. Expression of liver IGF-I was reduced in both hypoosmotic- and hypersaline-adapted sea bream, indicative of a reduction in growth regulation. A low abundance of hepatic IGF-I mRNA was also found to be correlated with a reduction of growth in salmon (16). The data from GH-IGF-I expression studies demonstrate this endocrine axis is modulated in different salinities and that there is an upregulation of this axis in isoosmotic-adapted sea bream. Isoosmotic conditions have been shown to be beneficial for increased growth in many fish species (3), including silver sea bream (58). From the data obtained in this study, it can be seen that expression of both GH and IGF-I genes correlated with enhanced growth of silver sea bream acclimated to isoosmotic salinity (58).
In conclusion, the results of the present study have demonstrated how certain processes are modulated in a euryhaline fish during salinity adaptation. It does appear that at an isoosmotic salinity minimal energy is expended for ion regulation across gill epithelium and at this same salinity the somatotropic axis is upregulated. Mechanisms of stress protein induction were also found to be modulated during salinity adaptation. Because the genes investigated in silver sea bream are highly conserved in most teleosts, the type of molecular analysis described in the present study can now be applied to many different fish species.
This research was supported by Earmarked Grants CUHK4168/99M, CUHK4252/00M, and CUHK4264/02M (Research Grants Council, Hong Kong) awarded to Dr. Norman Y. S. Woo.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society