Abstract
It has recently become evident that maintenance of ionic homoeostasis in euryhaline salmonids involves a reciprocal shift in expression of two isoforms of the gill Na+,K+-atpase α-subunit when the surrounding salinity changes. The present study investigated the regulation of this shift between the α1a (freshwater (FW) isoform) and the α1b (seawater (SW) isoform) by cortisol, Gh, prolactin (Prl) and Igf 1. Injection with cortisol into FW salmon increased α1a expression, while Gh had no effect. Conversely, both cortisol and Gh stimulated α1b expression, and a significant synergy was observed. igf1 expression was increased by Gh in both gill and liver, and inhibited by cortisol in the liver. Gill igf1 and gh receptor expression increased in response to cortisol. Injection with Prl into SW salmon compromised their hypo-osmoregulatory performance, selectively reduced the expression of the α1b isoform and decreased enzymatic Na+,K+-atpase activity in the gill. Cortisol and Prl reduced gill and liver igf1 expression, and both hormones stimulated gill igf1 receptor expression. In a short-term experiment with incubation of FW gill cell suspensions, cortisol stimulated α1a and α1b expression, while Igf1 stimulated only α1b. The data elaborate our understanding of Prl and Gh as being antagonists in the control of gill ion regulation, and support a dual role for Gh involving endocrine and paracrine Igf1 action. Gh and Prl may be the decisive stimuli that direct cortisol-aided mitochondrion-rich cell development into either secretory or absorptive types.
Introduction
The gill epithelium of euryhaline teleosts has an extraordinary functional plasticity, as it is able to interchange between modes of ion uptake in freshwater (FW) and ion excretion in seawater (SW). In both environments, mitochondrion-rich cells (MRCs) and their adjoining accessory and pavement cells are fundamental to compensatory ion transport for which the driving force is delivered by the Na+,K+-atpase and in the FW mode possibly complemented by an apical V-type H+-ATPase (see Hirose et al. 2003). The recent discovery of different Na+,K+-atpase α-1 isoforms in the salmonid teleost gill (α1a, α1b and α1c; Richards et al. 2003) has offered new insight into transepithelial ion transport in the gill. The mRNA expression levels of these isoforms have been investigated during salinity acclimation in several salmonids, and it is remarkable how salinity induces a reciprocal switch between α1a and α1b isoforms, α1a being augmented in FW and α1b in SW (Richards et al. 2003, Mackie et al. 2005, Bystrianski et al. 2006). This has led to the interesting suggestion that α1a is a FW isoform driving ion uptake, whereas α1b is a SW isoform driving ion excretion. Recently, this hypothesis has gained support based on structural and thermodynamical considerations (Jorgensen 2008). We demonstrated the specific localization of the α isoform transcripts in the gill (Madsen et al. 2009) and recognized α1b as a good marker for SW type MRC and their putative precursor cells deeper in the filament, while the possibility that α1a plays a dual role in both FW and SW could not be excluded.
An array of signalling mechanisms mediates the physiological changes necessary to counter changes in environmental salinity. Characteristically for salmonid teleosts, transfer to SW induces increased plasma Gh and insulin-like growth factor-I (Igf1) along with reduced prolactin (Prl) levels (Avella et al. 1990, Young et al. 1995, Poppinga et al. 2007). There are reports that cortisol levels increase temporarily after both hyper- and hypo-osmotic challenges (see McCormick 2001), reflecting its role in the general stress–response. More specific roles associated with activation of ion-uptake and ion-secretory mechanisms respectively have also been assigned to cortisol (see Evans et al. 2005). An important role of the Gh–Igf1 axis and cortisol in promoting SW tolerance and increasing overall gill Na+,K+-atpase activity has been clearly documented in salmonids (Björnsson et al. 1987, Madsen 1990, McCormick 1996, Seidelin et al. 1999). It was early on established that Prl has an ion-retaining effect in teleosts (Pickford & Phillips 1959). Several years later, it was demonstrated that Prl opposes SW acclimation and the effect of Gh when injected into trout (Madsen & Bern 1992, Seidelin & Madsen 1997). Thus, Prl and Gh appear as antagonists with respect to control of osmoregulatory status. In FW, cortisol may stimulate gill ion uptake (Laurent & Perry 1990), possibly in combined action with Prl (Zhou et al. 2004), but the molecular details of the interaction of these osmoregulatory hormones are not yet clear.
The main aim of the present study was to clarify the effects of Gh, Igf1, Prl and cortisol on the expression level of the assumed FW- and SW-type Na+,K+-atpase isoforms α1a and α1b respectively. In addition to analysing the effects of these hormones separately, especially the interaction with cortisol of the three peptide hormones was intriguing. The effect of cortisol was tested in two-factorial injection experiments with Gh in FW salmon, with Prl in SW salmon and with Igf1 directly in isolated gill cells from FW salmon in vitro. The Na+,K+-atpase β1-subunit that is needed for membrane targeting and functional maturation of the Na+,K+-atpase in general and the Na+,K+,2Cl− cotransporter (Nkcc)-1a, a marker of secretory MRCs in salmon gill (Mackie et al. 2007), were analyzed in parallel with α1a and α1b. In order to approach the mechanism of hormonal actions and interactions, we examined the expression levels of gh receptor type I (ghr1) and igf receptor type I (igfr1) in the gill together with expression of igf1 in both liver and gill.
Materials and Methods
Fish and sampling procedure
One-year-old Atlantic salmon (mixed sex Salmo salar, Danish Centre for Wild Salmon, Randers, Denmark) were kept in outdoor tanks with flowing tap water at environmental temperature and exposed to natural daylight through smoltification. In July, they were transferred to indoor conditions in 500 l glass fibre tanks at 14 °C at 12 h light:12 h darkness cycle. During standard sampling, the fish were anaesthetized in buffered MS222 (3-aminobenzoic acid ethyl ester) and blood was collected with a heparinized syringe from the caudal vessels, after which the fish was killed and further tissues were excised. In injection experiments, the fish were anaesthetized in buffered MS222 before injection. Upon sampling, one second gill arch was dry frozen immediately in liquid nitrogen for RNA extraction, and the other second arch was frozen in sucrose–EDTA-imidazole buffer (300 mM sucrose, 5 mM Na2EDTA, 50 mM imidazole, pH 7.3) for analysis of Na+,K+-atpase enzymatic activity. One first arch was dissected into 3 mm blocks and frozen on dry ice in optimal cutting temperature compound (Miles Inc., Eikhart, IN, USA) for further analysis by in situ hybridisation (ISH). The liver was excised and frozen in liquid nitrogen for RNA extraction. The experimental protocols were approved by the Danish Animal Experiments Inspectorate and in accordance with the European convention for the protection of vertebrate animals used for experiments and other scientific purposes (#86/609/EØF).
Experimental protocols
Three two-factorial experiments were set up to investigate the effects of and interaction between 1) cortisol and Gh, 2) cortisol and Prl and 3) cortisol and Igf1 on osmoregulatory and endocrine targets in gill and liver. All hormone doses were chosen to be within the range commonly used in studies of osmoregulatory endocrinology in fish. Based on studies of hormone dynamics and clearance values in a variety of fish species (Hirano & Utida 1971, Weisbart et al. 1987, Sakamoto et al. 1990, Duan & Hirano 1991, Mancera et al. 1994, Shrimpton & Randall 1994), the chosen doses were expected to give temporarily supra-physiological serum levels and metabolic half-lives in the range of 3–12 h.
Effect of cortisol and Gh
FW-acclimated fish were used in order to analyse the hormonal induction of hypo-osmoregulatory mechanisms in fish being in a hyper-osmoregulatory condition. In September, four groups of fish (15–25 g) were injected intraperitoneally with either saline (0.9% NaCl, 0.4% BSA), 4 μg cortisol/g (F; Na-hydrocortisone hemisuccinate; Sigma), 0.2 μg recombinant trout Gh/g (Novozymes Biopharma AU Ltd, Adelaide, Australia) or a combination of 4 μg cortisol and 0.2 μg Gh/g. Injections were repeated 2 days later, and the experiment terminated by sampling on the following day.
Effect of cortisol and Prl
In this experiment, fish (30–40 g) acclimated to SW (2 weeks in October) were used in order to analyse the hormonal induction of hyper-osmoregulatory mechanisms in fish being in a hypo-osmoregulatory condition. Four groups were injected intraperitoneally with either saline, 4 μg F/g, 0.2 μg salmon Prl/g or a combination of 4 μg cortisol and 0.2 μg salmon Prl/g. The Prl was purified from chum salmon pituitaries and a gift from Dr S Moriyama (Kitasato Univ., Japan). Injections were repeated after 2 days, and the experiment terminated by sampling 1 day later.
Direct effect of cortisol and Igf1
In this experiment, we used gill cell suspensions from FW-acclimated fish (60–80 g) in order to analyze the direct effect of expected hypo-osmoregulatory hormones in gills from fish being in a hyper-osmoregulatory condition. In December, gill cell suspensions were prepared as described by Tipsmark & Madsen (2001). The cells were incubated 12 h in salmon Ringer's solution equilibrated with 99% O2/1% CO2 (in mM: 140 NaCl, 15 NaHCO3, 3.5 KCl, 1.5 CaCl2, 1.0 NaH2PO4, 0.8 MgSO4, 10.0 d-glucose and 5.0 N-2-hydroxyethyl-piperazine propanesulphonic acid; 500 IU penicillin/ml; 250 μg/ml streptomycin; 4 mg/ml BSA; osmolality: 305 mOSM; pH 7.8) with either vehicle, 10 μg F/ml, 0.1 μg Igf1/ml (recombinant human Igf1 was supplied by Chiron Corporation, Emeryville, CA, USA) or a combination of 10 μg cortisol and 0.1 μg Igf1/ml. After incubation, cells were isolated by centrifugation (5000 g, 1 min) and sonicated in TRIzol solution (Invitrogen) for extraction of mRNA.
Analytical techniques
Plasma osmolality and Na+,K+-atpase activity
Plasma osmolality was measured by freezing point depression osmometry (Osmomat 030; Gonotec, Berlin, Germany). Gill Na+,K+-atpase activity was measured according to McCormick (1993) using a microplate reader (SPECTRAmax PLUS, Molecular Devices, Sunnyvale, CA, USA) and normalized to protein content of the homogenate (Lowry et al. 1951).
Extraction of RNA and cDNA synthesis
Total RNA was extracted by the TRIzol procedure (Invitrogen) according to the manufacturer's recommendation and using maximally 100 mg of tissue or cells per ml TRIzol. Total RNA concentrations were determined by measuring A260 in duplicate with a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). A260/A280 varied from 1.8 to 2.0. One microgram RNA was treated with 0.7 unit RQ1 DNase (Promega) for 30 min at 37 °C followed by addition of RQ DNase stop solution and 10 min at 65 °C to inactivate the DNAse. cDNA was synthesized by reverse transcription carried out on 1 μg DNAse-treated RNA with oligo(dT)15 primers using DyNAmo cDNA synthesis kit (Finnzymes, Espoo, Finland), according to the manufacturer's recommendations. At the end, the cDNA was diluted to a total volume of 50 μl with milli-Q H2O.
Primers
Primer3 software was used to design primers detecting the deduced sequences and checked using NetPrimer software (Premier Biosoft International, Palo Alto, CA, USA). Primer sequences were tested for non-specific product amplification and primer–dimer formation by analysis of melting curve analysis, and amplicon size was verified by agarose gel electrophoresis. Na+,K+-atpase and elongation factor 1a (ef1a) primers have been reported previously (Madsen et al. 2009). Additional primers used were (5′–3′): igf1 (forward, ACTGTGCCCCTGTCAAGTCT; reverse, CTGTGCTGTCCTACGCTCTG; based on acc. no.
Real-time QPCR
Quantitative PCR analysis was carried out on a Mx3000p instrument (Stratagene, La Jolla, CA, USA) with SYBR Green detection using standard software settings with adaptive baseline for background detection, moving average and amplification-based threshold settings with built-in FAM/SYBR filter (excitation wavelength: 492 nm; emission wavelength: 516 nm). Reactions were carried out with a cDNA amount equivalent of 20 ng total RNA, 150 nM forward and reverse primer, 1× SYBR Green JumpStart (Sigma) in a total volume of 25 μl. Cycling conditions were 95 °C for 30 s and 60 °C for 60 s in 50 cycles. Melting curve analysis was carried out routinely with 30 s for each 1 °C interval from 55 to 95 °C. For all primer sets, a pooled cDNA sample was diluted and analyzed by QPCR to establish amplification efficiency (Ea). This was in the range 1.85–2. For normalization of gene expression, ef1a was used in accordance with Olsvik et al. (2005). Relative copy numbers of the target genes were calculated as EaCt, where Ct is the threshold cycle number and Ea is the amplification efficiency. Normalized units were obtained by dividing relative copy number of target genes with relative copy number of ef1a. Negative control reactions using DNase-treated total RNA from representative samples were used to analyze carry-over of genomic DNA. In all samples and with all primers, there was no genomic contamination.
In situ hybridisation
The dissected gill specimens were stored and desiccated at −80 °C until further processing. Frozen sections (8–10 μm) were cut on a cryostat at −14 °C, transferred to Superfrost Plus glass slides (Erie Scientific Company, Portsmouth, NH, USA) and dried for 1 h at 37 °C. The sections were then transferred to −80 °C until further processing. ISH was performed using the protocol described by Madsen et al. (2009), which is based on the method by Lambertsen et al. (2001). In short, gill sections were incubated with 3 pmol/ml alkaline phosphatase (AP)-labelled synthetic 28-mer cDNA probes recognising the α1a and α1b isoforms of the Na+,K+-atpase α-subunit (Madsen et al. 2009). Hybridisation conditions were: 35% formamide, 4× saline sodium citrate buffer (4× SSC: 30 mM NaCl, 3 mM sodium citrate), 1× Denhardt's solution (0.2 mg/ml Ficoll, 0.2 mg/ml polyvinyl pyrrolidon, 0.2 mg/ml BSA), 10% dextran sulphate, 10 μg/ml single-stranded salmon sperm DNA overnight at 37 °C. Post-hybridisation washes were: 3×30 min changes of 3× SSC (22.5 mM NaCl, 2.3 mM sodium citrate, pH 8.0) at 37 °C followed by two 10 min rinses in post-hybridisation buffer (0.01 M Tris–HCl, 0.15 M NaCl, pH 9.5) at room temperature. Finally, the specimens were incubated in freshly made AP-developer buffer (0.32 mg/ml nitro blue tetrazolium (Sigma), 0.17 mg/ml 5-bromo-4-chloro-3-indolyl phosphate (Sigma), 0.1 M Tris–HCl, 0.1 M NaCl and 0.05 M MgCl2.6H2O, pH 9.5) for 1–2 days at room temperature until appropriate colour deposition was observed. After a rinse in distilled water (30 °C, 30 min) to stop the colorimetric reaction, coverslips were mounted using Aquatex (Merck). Images were taken at 250 and 400× magnification using a Leica microscope (DM R, Leica Microsystems Wetzlar GmbH, Wetzlar, Germany) mounted with a Leica digital camera (DC200).
Statistical analysis
FW and SW control values were compared using a Student's t-test. In the injection experiments, a two-factorial ANOVA was used to analyze for overall effects of hormones and their interaction. When interaction between hormones was significant, multiple comparisons were performed. Differences between means were assessed by Tukey's hydroxysteroid dehydrogenase multiple comparison test or Dunn's post test, as appropriate. When required, transformation of data was done to meet the ANOVA assumption of homogeneity of variances as tested by Bartlett's test, and if this was unattainable, Kruskal–Wallis nonparametric one-way ANOVA was employed. All statistical analyses were performed using GraphPad Prism 4.0 software (San Diego, CA, USA), and significant differences were accepted when P<0.05. Asterisks indicate significance level (*P<0.05; **P<0.01; ***P<0.001).
Results
Effect of salinity
FW and SW control values from the two injection experiments are compared in Table 1. Plasma osmolality was not significantly different, whereas all other parameters showed major differences with respect to salinity. Gill Na+,K+-atpase activity, α1b, β1, nkcc, igf1, igfr1, ghr and liver ghr1 mRNA levels were all higher in SW than in FW. Gill α1a and liver igf1 mRNA were lower in SW than in FW.
Comparison of plasma osmolality, gill Na+,K+-atpase activity and relative mRNA expression levels in freshwater- (FW) and seawater (SW)-acclimated Atlantic salmon. All mRNA expression levels are calculated relative to expression of elongation factor-1a (ef1a). Ratio is calculated as the ratio between SW and FW value
| Parameter | FW | SW | Ratio | P value* | |
|---|---|---|---|---|---|
| Osmolality | 313.8±6.7 | 299.2±6.7 | 0.95 | 0.14 | |
| Gill Na+,K+-atpase | 2.10±0.12 | 4.24±0.45 | 2.0 | 0.0003 | |
| Gill mRNA | α1a | 9.31±1.78 | 0.40±0.03 | 0.04 | 0.0002 |
| α1b | 0.57±0.11 | 3.72±0.45 | 6.5 | <0.0001 | |
| β1 | 0.45±0.03 | 0.80±0.09 | 1.8 | 0.001 | |
| nkcc1a | 0.065±0.006 | 1.12±0.12 | 17 | <0.0001 | |
| igf1 | 0.0023±0.0004 | 0.022±0.009 | 9.6 | <0.03 | |
| igfr1 | 0.0004±0.0001 | 0.0108±0.0017 | 27 | <0.0001 | |
| ghr | 0.051±0.007 | 0.10±0.02 | 2.0 | <0.03 | |
| Liver mRNA | igf1 | 12.3±1.7 | 1.02±0.22 | 0.08 | <0.0001 |
| ghr | 2.47±0.40 | 8.16±1.03 | 3.3 | 0.0001 | |
*P value is obtained from a two-tailed Student's t-test. FW, n=8; SW, n=7.
Effect of cortisol and Gh in FW fish
Gh induced a significant decline in plasma osmolality of ∼10 mOsm/kg (Fig. 1A). Overall, cortisol stimulated α1a, β1 mRNA and gill Na+,K+-atpase activity, but no significant effect of Gh was observed on these parameters (Fig. 1B, D and F). Both cortisol and Gh stimulated gill α1b and nkcc1a expression (Fig. 1C and E) with a significant synergistic interaction. Gill igf1 expression was stimulated by Gh (P<0.001), while cortisol had no effect (Fig. 2A). Both gill igfr1 and ghr1 mRNA levels were stimulated by cortisol, while Gh depressed the igfr1 level (Fig. 2B and C). Hepatic igf1 was inhibited by cortisol and stimulated by Gh, and the effects were additive (interaction: P>0.05; Fig. 2D). The α1a transcript was most abundant in large cells in the interlamellar (IL) filament region and occasionally on the lamellae in gills of FW control fish (Fig. 3). No major effect on this localization pattern was observed after treatment with F and/or Gh. The α1b transcript was mostly present in small cells below the surface of the IL filament region in FW control fish (Fig. 4). The cellular staining intensity, size and abundance increased markedly after F+Gh treatment.
Effects of cortisol (F) and Gh on plasma osmolality (A) and expression of α1a (C), α1b (D), β1 (D), nkcc1a (E) and Na+,K+-atpase activity (F) in the gill of FW salmon. Asterisks next to F or Gh indicate significant overall effect. Asterisks next to F×Gh indicate significant interaction, in which case different letters above bars indicate significant difference between groups. Values are means±s.e.m. (n=8–10).
Citation: Journal of Endocrinology 203, 2; 10.1677/JOE-09-0281
Effects of cortisol (F) and Gh on expression of gill igf1 (A), gill igfr1 (B) and gill ghr1 (C) and liver igf1 (D) in FW salmon. Asterisks next to F or Gh indicate significant overall effect. Values are means±s.e.m. (n=8–10).
Citation: Journal of Endocrinology 203, 2; 10.1677/JOE-09-0281
In situ hybridisation of gill cryosections using an AP-conjugated 28-mer antisense probe detecting the α1a isoform of the Na+,K+-atpase α-subunit. Con, control; F, cortisol; Gh+F, Gh+cortisol. Bar represents 100 μm. Full colour version of this figure available via http://dx.doi.org/10.1677/JOE-09-0281.
Citation: Journal of Endocrinology 203, 2; 10.1677/JOE-09-0281
In situ hybridisation of gill cryosections using an AP-conjugated 28-mer antisense probe detecting the α1b isoform of the Na+,K+-atpase α-subunit. Con, control; F, cortisol; Gh+F, Gh+cortisol. Bar represents 100 μm. Full colour version of this figure available via http://dx.doi.org/10.1677/JOE-09-0281.
Citation: Journal of Endocrinology 203, 2; 10.1677/JOE-09-0281
Effect of cortisol and Prl in SW fish
Injection with Prl induced a very significant increase in plasma osmolality of ∼50 mOsm/kg (Fig. 5A). Prl induced a decrease in gill α1b and nkcc1a expression, while no effect of cortisol on these transcripts was observed (Fig. 5C and E). Overall, cortisol stimulated gill β1 mRNA levels, while no effect on α1a was observed (Fig. 5B and D). An overall decrease in gill Na+,K+-atpase activity was observed in response to Prl (Fig. 5F). Gill igf1 expression was decreased by cortisol, while Prl had no effect (Fig. 6A). Gill igfr1 was stimulated by both Prl and cortisol, while ghr1 was inhibited by cortisol (Fig. 6B and C). A nonparametric ANOVA showed that hepatic igf1 was inhibited by Prl, while cortisol had no effect (Fig. 6D). The α1a transcript was localised in small cells below the surface of the IL filament and occasionally in the lamellar epithelium of SW control fish (Fig. 7). F treatment stimulated the number of positive cells, whereas there was no obvious effect of Prl or F+Prl treatment. The α1b transcript was abundant in cells in the IL filament and occasionally on the lamellae (Fig. 8). Prl treatment induced weaker staining and smaller cells.
Effects of cortisol (F) and Prl on plasma osmolality (A) and expression of α1a (B), α1b (C), β1 (D), nkcc1a (E) and Na+,K+-atpase activity (F) in the gill of SW salmon. Asterisks next to F or Prl indicate significant overall effect. Values are means±s.e.m. (n=8–10).
Citation: Journal of Endocrinology 203, 2; 10.1677/JOE-09-0281
Effects of cortisol (F) and Prl on expression of gill igf1 (A), gill igfr1 (B), gill ghr1 (C) and liver igf1 (D) in SW salmon. Asterisks next to F or Prl indicate significant overall effect. The data in panel D were subject to nonparametric analysis, and groups marked with different letters are significantly different. Values are means±s.e.m. (n=8–10).
Citation: Journal of Endocrinology 203, 2; 10.1677/JOE-09-0281
In situ hybridisation of gill cryosections using an AP-conjugated 28-mer antisense probe detecting the α1a isoform of the Na+,K+-atpase α-subunit. Con, control; F, cortisol; Prl, prolactin; F+Prl, cortisol+prolactin. Bar represents 100 μm. Full colour version of this figure available via http://dx.doi.org/10.1677/JOE-09-0281.
Citation: Journal of Endocrinology 203, 2; 10.1677/JOE-09-0281
In situ hybridisation of gill cryosections using an AP-conjugated 28-mer antisense probe detecting the α1b isoform of the Na+,K+-atpase α-subunit. Con, control; F, cortisol; Prl, prolactin; F+Prl, cortisol+prolactin. Bar represents 100 μm. Full colour version of this figure available via http://dx.doi.org/10.1677/JOE-09-0281.
Citation: Journal of Endocrinology 203, 2; 10.1677/JOE-09-0281
Effect of cortisol and Igf1
Incubation of FW gill cells with cortisol for 12 h stimulated α1a, α1b and β1 mRNA (Fig. 9A–C). In contrast, igf1 only stimulated α1b of the three Na+,K+-atpase genes examined. Both cortisol and Igf1 stimulated gill cell igf1 expression (Fig. 9D). igfr1 was stimulated by cortisol, an effect that was blocked by igf1 (Fig. 9E; antagonistic effect: P<0.05). ghr1 mRNA levels were also stimulated by cortisol, an effect that was significantly antagonized, albeit not blocked by Igf1 (Fig. 9F; P<0.01).
Effects of cortisol (F) and Igf1 on expression of α1a (A), α1b (B), β1 (C), igf1 (C), igfr1 (E) and ghr1 (F) in isolated gill cells from FW salmon. Asterisks next to F or Igf1 indicate significant overall effect. Asterisks next to FxIgf1 indicate significant interaction, in which case different letters above bars indicate significant difference between groups. Data are pooled from two separate experiments, and values are means±s.e.m. (n=8).
Citation: Journal of Endocrinology 203, 2; 10.1677/JOE-09-0281
Discussion
Changes in environmental salinity induce intricate molecular and cellular changes in the gill of euryhaline teleosts. As a fundamental part of the response in salmonids, a distinct switching occurs between Na+,K+-atpase α isoforms 1a and 1b (Richards et al. 2003), which may change the functional properties of the enzyme (Jorgensen 2008) in order to maintain the appropriate ionic gradients across the gill. The primary focus of our study was the endocrine and paracrine regulation of this particular isoform switch, and the experiments were performed under both hypo- and hyper-osmotic conditions in anticipation of maximal effects of the applied hormonal treatments. Accordingly, the initial expression level of α1a was higher in FW than in SW, and the level of α1b and nkcc1a was higher in SW than in FW, reflecting the characteristic ion-regulatory modus of the gill epithelium in the two situations.
Hormonal effect on α1a and α1b isoforms in the gill
Several studies have shown mineralocorticoid actions of cortisol in teleosts (see Mommsen et al. 1999). Cortisol is known to stimulate gill MRC proliferation and overall gill Na+,K+-atpase activity, and it has been puzzling for many years how cortisol in doing so may both facilitate acclimation to hyper- and hypo-osmotic conditions. The discovery by Richards et al. (2003) that two isoforms of the Na+,K+-atpase α-subunit contribute to overall enzymatic activity in the gill has led to the hypothesis, that α1a is primarily expressed in MRCs responsible for absorption of monovalent ions (and possibly Ca2+), while α1b is the dominant isoform in secretory-type MRCs. This has gained support from ISH analyses of transcript localisation in the salmon gill (Madsen et al. 2009). In the present study, both isoforms were stimulated by cortisol in vivo and in vitro. This explains how cortisol may simultaneously facilitate both FW and SW acclimation by increasing the general ion transport capacity of the gill. A salinity-specific effect of cortisol is, however, only attained by interaction with additional hormones: Gh or Prl. The present data show a very significant synergy with Gh specifically on the α1b isoform, which was confirmed by a strong increase in cellular ISH staining intensity and abundance of large α1b-positive cells in the FW salmon gill. A similar synergism is seen with respect to the nkcc1a transcript level, which is specifically involved in ion secretion of SW MRCs (Wu et al. 2003), thus underlining the specific importance of α1b in this cell-type. The data nicely explain in molecular terms how cortisol and Gh in salmonids exert a strong synergistic effect on the ability to regain osmotic balance after FW to SW transfer (Madsen 1990, McCormick 1996). In earlier studies, the functional effect of cortisol and Gh was associated with stimulatory effects of the two hormones on overall gill Na+,K+-atpase activity, which was either synergistic (Madsen 1990) or additive (McCormick 1996). The lack of interaction on overall Na+,K+-atpase activity in the present experiment and in a previous study by McCormick (1996) is likely to adhere to masking by the dominant gill isoform in FW α1a, which was insensitive to Gh treatment but stimulated by cortisol. In addition, the present lack of interaction on enzyme activity may be due to the short duration of the experiment (3 days).
In the SW experiment, Prl induced a substantial osmotic imbalance underlining its hyper-osmoregulatory action. Our data suggest that this is causally related to a selective reduction of α1b isoform expression in the Na+,K+-atpase, while the α1a isoform is unaffected. Consequently, gill Na+,K+-atpase activity and the secretory capacity of the gill decreased. The anti-SW effect is further substantiated by the parallel downregulation of the secretory nkcc1a and the much weaker ISH-staining intensity in gills from Prl-treated fish. IL filament cells staining positive for α1b expression were much reduced in size and retrieved from the mucosal surface of the epithelium. Taken together, the data suggest that Prl impairs the functional status of secretory MRCs, thereby supporting histological evidence in SW tilapia gill and opercular membrane, where Prl induced a changed morphology of the putative ion-secretory α-type MRC into the typical FW MRC morphology (Herndon et al. 1991, Pisam et al. 1993). The observed effect of Prl may either be due to anaplasia of existing SW-type MRCs, transformation of SW-type MRCs into FW-type MRCs or inhibition of recruitment of SW-type MRCs (Fig. 10). Transformation into FW-type MRCs does not seem likely from the present data, as there was no evidence that the α1a isoform was stimulated by Prl.
Tentative model for hormonal regulation of Na+,K+-atpase isoform expression in mitochondrion-rich cells (MRC) in salmon gill epithelia. Stimulation is indicated by arrows and inhibition by block-ended lines. Two types of differentiated MRCs (or their associated precursor cells) are suggested. One difference between these is the abundance of the Na+,K+-atpase α1a isoform in the ‘FW-type’ versus the dominance of the α1b isoform in the ‘SW-type’ MRC. Upon the appropriate endocrine/paracrine stimulation, these may develop into fully functional ionocytes (‘FW-type’ or ‘SW-type’). The possibility of transformation between the functionally differentiated MRCs is currently speculative. Cortisol is a stimulator of both isoforms of the Na+,K+-atpase α-subunit; however, only in the SW-type cell does cortisol interact in synergy with Gh and Igf1 (local and systemic). In addition, cortisol sensitizes this cell type by increasing Gh and Igf1 receptor levels. Prl interferes with this development by blocking the expression of α1b and also by suppressing hepatic igf1 production.
Citation: Journal of Endocrinology 203, 2; 10.1677/JOE-09-0281
Endocrine and paracrine mechanisms of action
All the prerequisites for a dual-effector mode-of-action of Gh in regulating gill MRC development are present in salmonids. MRCs are responsive to Gh, they express the igfr, they express igf1 in response to Gh, and Igf1 and its receptor colocalise with the Na+,K+-atpase enzyme in MRCs (Reinecke et al. 1997, Tipsmark et al. 2007). Radioreceptor assays have shown specific Gh binding to gill membranes (Gray et al. 1990), but the precise cellular localisation of the Ghr in gills still awaits examination. Gill Na+,K+-atpase activity is stimulated by both Gh and IgfI (Madsen & Bern 1992, McCormick 1996), and here we show explicitly that this is due to specific stimulation of the α1b isoform of the pump. In the isolated-cell experiment, the α1a isoform expression was solely stimulated by cortisol, while the SW-type, α1b, was sensitive to both cortisol and Igf1. This partially explains the selective effect of Gh on α1b found in the injection experiment, where endocrine and paracrine igf1 expression was elevated. However, the strong synergistic effect of Gh and cortisol cannot be explained by these findings, but is likely to result from Gh-enhanced sensitivity to Igf1 (Madsen & Bern 1992) and cortisol (Shrimpton & McCormick 1998), or cortisol enhanced sensitivity to Igf1 and Gh as suggested by the current data.
The effect of environmental salinity on additional elements of the Gh–Igf1 axis of the salmon was further investigated. igf1 expression, which, as expected, was much higher in the liver than the gill, responded to SW with a tenfold increase in the gill and a tenfold decrease in the liver. The increase in the gill confirmed the reports in rainbow trout by Sakamoto & Hirano (1993) and Poppinga et al. (2007), and suggests paracrine actions of Igf1 to be involved in modification of ion transport in the gill. The decrease in the liver was surprising and is at variance with Poppinga et al. (2007), but may be explained by our additional observation that cortisol has a strong inhibitory effect on hepatic igf1 expression. Cortisol was not measured in the present experiment but is known to increase when FW salmonids are transferred to SW (see McCormick 2001). Additional factors such as the time course of sampling and nutritional status may also add variance when endocrine responses are compared between experiments. The ghr1 was also expressed at higher levels in the liver than in the gill and showed an increase in the liver in response to SW. This is in accordance with Poppinga et al. (2007) and with increased total ghr number after SW transfer in rainbow trout (Sakamoto & Hirano 1991). On the contrary, igf1 receptor type I mRNA was found in the gill but not detected in liver. The absence in liver agrees with Greene & Chen's (1997) observation in juvenile and adult trout, but is at variance with early stages of shi drum (Radaelli et al. 2003) and barramundi (Drakenberg et al. 1997), where hepatic expression was reported. Gill igfr1 expression increased very significantly in response to SW, suggesting elevated sensitivity to igf1. In trout and striped bass, igfr also increased after SW transfer although lasting only for 1–2 days (Poppinga et al. 2007, Tipsmark et al. 2007). Altogether, these data support our conception of involvement of the Gh–Igf1 axis in SW acclimation of the salmon.
Gh injection elevated both liver and gill igf1 mRNA, and thus parallels findings in injection experiments in rainbow trout (Sakamoto & Hirano 1993) and in vitro data from coho salmon hepatocytes (Duan et al. 1993). Meanwhile, cortisol inhibited igf1 expression in the liver both with and without exogenous Gh induction. This effect is in full agreement with in vitro observations on primary culture of salmon hepatocytes, where dexamethasone blocks Gh-induced igf1 expression possibly due to decreased ghr expression (Pierce et al. 2005). In the gill, cortisol had either no effect (in vivo, FW) or a stimulatory effect (in vitro, FW) on igf1 expression, which agreed well with the elevated ghr1 levels induced by the steroid. The sensitivity of the gill to Gh and its secondary effector, Igf1, are thus likely to be elevated by cortisol in FW since both ghr1 and also the igfr1 were stimulated by the steroid. Therefore, the strong synergistic effect of cortisol and Gh on α1b expression may relate to elevated sensitivity to Gh and Igf1 in putative SW MRCs or precursor cells (Fig. 10). The synergy may be further reinforced by enhanced cortisol sensitivity found in Atlantic salmon gill upon Gh treatment (Shrimpton & McCormick 1998). Curiously, in the SW gill, the effect of cortisol is totally different: igf1 and ghr1 mRNA levels decrease, and α1b and nkcc are not affected.
One possible mode of action of Prl may be via reduced igf1 expression in the liver assuming systemic Igf1 is a requirement for sustaining secretory mode of the gill. However, Prl receptors are expressed in salmon gill (Nilsen et al. 2008), and a direct effect of Prl could also be involved. In previous injection studies performed in hyper-osmoregulating trout in FW, Prl antagonized the hypo-osmoregulatory effect induced by cortisol and Gh (Madsen et al. 1995, Seidelin & Madsen 1997). In these experiments, Prl had no isolated effect on gill Na+,K+-atpase activity but solely opposed the stimulatory action of Gh and F. The Prl receptor has been localised in FW MRCs (tilapia: Sandra et al. 2000), and it has been hypothesized that Prl works together with cortisol to promote development/maintenance of FW MRC (McCormick 2001). The present data gave no such indication, as neither Prl alone nor in interaction with cortisol stimulated the alleged FW α1a isoform of the Na+,K+-atpase. Instead, Prl had a negative effect on elements of SW-type MRCs. Taken together, Gh and Prl may be the decisive stimuli that direct cortisol-aided MRC development into either secretory or absorptive types. The effects of Gh/Prl may occur at the level of stem cell differentiation or proliferation of smaller differentiated progenitor cells (deeper in the epithelium) into large functional MRCs in contact with the apical surface.
As summarized in Fig. 10, the present data give functional support to a dual-effector mode of action of Gh on establishment of a hypo-osmoregulatory modus of the salmon gill. Cortisol stimulates the general ion transport capacity of the gill, reflected by its stimulation of both α1a and α1b isoforms of the Na+,K+-atpase. Through a synergistic interaction with cortisol, Gh specifically potentiates equipment of the gill with secretory transport proteins. This action is in part mediated by paracrine/endocrine igf1 and abolished by Prl. Even though the present data do not preclude that FW and SW MRCs may transformed back and forth upon the appropriate stimuli, future studies should focus on clarifying the mechanisms responsible for early stem cell differentiation into MRCs in the gill.
Declaration of interest
There is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
The research was supported by grants from the Danish Natural Research Council (272-06-0526), the Novo Nordisk Foundation (10584) and a Fellowship grant from the Carlsberg Foundation (2007-01-0213).
Acknowledgements
We would like to thank Annette Duus and Annie Bach for excellent technical assistance and Dr Shunsuke Moriyama from School of Marine Biosciences at Kitasato University for the generous gift of pituitary chum salmon prolactin. Recombinant human Igf1 was a gift from Chiron Corporation, Emeryville, CA, USA.
References
Avella M, Young G, Prunet P & Schreck CB 1990 Plasma prolactin and cortisol concentrations during salinity challenges of coho salmon, Oncorhynchus kisutch, at smolt and post-smolt stages. Aquaculture 91 359–372.
Björnsson BT, Yamauchi M, Nishioka RS, Deftos LJ & Bern HA 1987 Effects of hypophysectomy and subsequent hormonal replacement therapy on hormonal and osmoregulatory status of coho salmon, Oncorhynchus kisutch. General and Comparative Endocrinology 68 421–430.
Bystrianski JS, Richards JG, Schulte PM & Ballantyne JS 2006 Reciprocal expression of gill Na+/K+-ATPase α-subunit isoforms α1a and α1b during seawater acclimation of three salmonid fishes that vary in their salinity tolerance. Journal of Experimental Biology 209 1848–1858.
Drakenberg K, Carey G, Mather P, Anderson A & Sara VR 1997 Characterization of an insulin-like growth factor (IGF) receptor and the insulin-like effects of IGF-I in the bony fish, Lates calcarifer. Regulatory Peptides 69 41–45.
Duan C & Hirano T 1991 Plasma kinetics of growth hormone in the Japanese eel, Anguilla japonica. Aquaculture 95 179–188.
Duan C, Duguay SJ & Plisetskaya EM 1993 Insulin-like growth factor (IGF-I) mRNA expression in coho salmon, Oncorhynchus kisutsch: tissue distribution and effects of growth hormone/prolactin family proteins. Fish Physiology and Biochemistry 11 371–379.
Evans DH, Piermarini PM & Choe KP 2005 The multifunctional fish gill: dominant site of gas exchange, osmoregulation, acid–base regulation, and excretion of nitrogenous waste. Physiological Reviews 85 97–177.
Gray ES, Young G & Bern HA 1990 Radioreceptor assay for growth hormone in coho salmon (Oncorhynchus kisutch) and its application to the study of stunting. Journal of Experimental Zoology 256 290–296.
Greene MW & Chen TT 1997 Temporal expression pattern of insulin-like growth factor mRNA during embryonic development in a teleost, rainbow trout (Onchorynchus mykiss). Molecular Marine Biology and Biotechnology 6 144–151.
Herndon TM, McCormick SD & Bern HA 1991 Effects of prolactin on chloride cells in opercular membrane from seawater-adapted tilapia. General and Comparative Endocrinology 83 283–289.
Hirano T & Utida S 1971 Plasma cortisol concentration and the rate of intestinal water absorption in the eel, Anguilla japonica. Endocrinologia Japonica 18 47–52.
Hirose S, Kaneko T, Naito N & Takei Y 2003 Molecular biology of major components of chloride cells. Comparative Biochemistry and Physiology 136B 593–620.
Jorgensen PL 2008 Importance for absorption of Na+ from freshwater of lysine, valine and serine substitutions in the α1a-isoform of Na,K-ATPase in the gills of rainbow trout (Oncorhynchus mykiss) and Atlantic salmon (Salmo salar). Journal of Membrane Biology 223 37–47.
Lambertsen KL, Gregersen R, Drøjdahl N, Owens T & Finsen B 2001 A specific and sensitive method for visualization of tumor necrosis factor in the murine central nervous system. Brain Research Protocols 7 175–191.
Laurent P & Perry SF 1990 The effects of cortisol on gill chloride cell morphology and ionic uptake on the freshwater trout, Salmo gairdneri. Cell and Tissue Research 259 429–442.
Lowry OH, Rosebrough NJ, Farr AL & Randall RJ 1951 Protein measurement with the Folin reagent. Journal of Biological Chemistry 193 265–275.
Mackie PM, Wright PA, Glebe BD & Ballantyne JS 2005 Osmoregulation and gene expression of Na+/K+-ATPase in families of Atlantic salmon (Salmo salar) smolts. Canadian Journal of Fisheries and Aquatic Sciences 62 2661–2672.
Mackie PM, Ballantyne JS, McCormick SD & Wright PA 2007 Na+/K+/2Cl− cotransporter FTR, gill expression after seawater transfer in smolts (0+) of different Atlantic salmon (Salmo salar) families. Aquaculture 272 625–635.
Madsen SS 1990 The role of cortisol and growth hormone in seawater adaptation and development of hypoosmoregulatory mechanisms in sea trout parr (Salmo trutta trutta). General and Comparative Endocrinology 79 1–11.
Madsen SS & Bern HA 1992 In vitro effects of insulin-like growth factor 1 on gill Na+,K+-atpase in Coho salmon, Oncorhynchus kisutch. Journal of Endocrinology 138 23–30.
Madsen SS, Jensen MK, Nohr J & Kristiansen K 1995 Expression of Na+,K+-ATPase in the brown trout Salmo trutta: in vivo modulation by hormones and seawater. American Journal of Physiology 269 R1339–R1345.
Madsen SS, Kiilerich P & Tipsmark CK 2009 Multiplicity of expression of Na+,K+-atpase α-subunit isoforms in the gill of Atlantic salmon (Salmo salar): cellular localisation and absolute quantification in response to salinity change. Journal of Experimental Biology 212 78–88.
Mancera JM, Péres-Fígares JM & Fernández-Llebrez P 1994 Effect of cortisol on brackish water adaptation in the euryhaline gilthead seabream (Sparus auratus L.). Comparative Biochemistry and Physiology 107A 397–402.
McCormick SD 1993 Methods for nonlethal gill biopsy and measurements of Na+,K+-atpase activity. Canadian Journal of Fisheries and Aquatic Sciences 50 656–658.
McCormick SD 1996 Effects of growth hormone and insulin-like growth factor I on salinity tolerance and gill Na+,K+-atpase in Atlantic salmon (Salmo salar): interaction with cortisol. General and Comparative Endocrinology 101 3–11.
McCormick SD 2001 Endocrine control of osmoregulation in teleost fish. American Zoologist 41 781–794.
Mommsen TP, Vijayan MM & Moon TW 1999 Cortisol in teleosts: dynamics, mechanisms of action, and metabolic regulation. Reviews in Fish Biology and Fisheries 9 211–268.
Nilsen TO, Ebbesson LO, Kiilerich P, Björnsson BT, Madsen SS, McCormick SD & Stefansson SO 2008 Endocrine systems in juvenile anadromous and landlocked Atlantic salmon (Salmo salar): seasonal development and seawater acclimation. General and Comparative Endocrinology 155 762–772.
Olsvik PA, Lie KK, Jordal AE, Nilsen TO & Hordvik I 2005 Evaluation of potential reference genes in real-time RT-PCR studies of Atlantic salmon. BMC Molecular Biology 6 Art. No. 21.
Pickford GE & Phillips JG 1959 Prolactin, a factor in promoting survival of hypophysectomised killifish in fresh water. Science 130 454–455.
Pierce AL, Fukada H & Dickhoff WW 2005 Metabolic hormones modulate the effect of growth hormone (Gh) on insulin-like growth factor-I (IGF-I) mRNA level in primary culture of salmon hepatocytes. Journal of Endocrinology 184 341–349.
Pisam M, Auperin B, Prunet P, Rentier-Delrue F, Martial J & Rambourg A 1993 Effects of prolactin on α and β chloride cells in the gill epithelium of the saltwater adapted tilapia Oreochromis niloticus. Anatomical Record 235 275–284.
Poppinga J, Kittilson JK, McCormick SD & Sheridan MA 2007 Effects of somatostatin on the growth hormone-insulin-like growth factor axis and seawater adaptation of rainbow trout (Oncorhynchus mykiss). Aquaculture 273 312–319.
Radaelli G, Domeneghini C, Arrighi S, Bosi G, Patruno M & Funkenstein B 2003 Localization of IGF-I, IGF-I receptor, and IGFBP-2 in developing Umbrina cirrosa (Pisces: Osteichthyes). General and Comparative Endocrinology 130 232–244.
Reinecke M, Schmid A, Ermatinger R & Loffing-Cueni D 1997 Insulin-like growth factor I in the teleost Oreochromis mossambicus the tilapia: gene sequence, tissue expression, and cellular localization. Endocrinology 138 3613–3619.
Richards JG, Semple JW, Bystriansky JS & Schulte PM 2003 Na+/K+-ATPase α-isoform switching in gills of rainbow trout (Oncorhynchus mykiss) during salinity transfer. Journal of Experimental Biology 206 4475–4486.
Sakamoto T & Hirano T 1991 Growth hormone receptors in the liver and osmoregulatory organs of rainbow trout: characterization and dynamics during adaptation to seawater. Journal of Endocrinology 130 425–433.
Sakamoto T & Hirano T 1993 Expression of insulin-like growth factor I gene in osmoregulatory organs during seawater adaptation of the salmonid fish: possible mode of osmoregulatory action of growth hormone. PNAS 90 1912–1916.
Sakamoto T, Ogasawara T & Hirano T 1990 Growth hormone kinetics during adaptation to hyperosmotic environment of rainbow trout, Salmo gairdneri. Journal of Comparative Physiology 160B 1–6.
Sandra O, Le Rouzic P, Cauty C, Edery M & Prunet P 2000 Expression of the prolactin receptor (tiPrl-R) gene in tilapia Oreochromis niloticus: tissue distribution and cellular localization in osmoregulatory organs. Journal of Molecular Endocrinology 24 215–224.
Seidelin M & Madsen SS 1997 Prolactin antagonizes the seawater-adaptive effect of cortisol and growth hormone in anadromous brown trout (Salmo trutta). Zoological Science 14 249–256.
Seidelin M, Madsen SS, Byrialsen A & Kristiansen K 1999 Effects of insulin-like growth factor-I and cortisol on Na+,K+-atpase expression in osmoregulatory tissues of brown trout (Salmo trutta). General and Comparative Endocrinology 113 331–342.
Shrimpton JM & McCormick SD 1998 Regulation of gill cytosolic corticosteroid receptors in juvenile Atlantic salmon: interaction effects of growth hormone with prolactin and triiodothyronine. General and Comparative Endocrinology 112 262–274.
Shrimpton JM & Randall DJ 1994 Downregulation of corticosteroid receptors in gills of coho salmon due to stress and cortisol treatment. American Journal of Physiology 267 R432–R438.
Tipsmark CK & Madsen SS 2001 Rapid modulation of Na+/K+-ATPase activity in osmoregulatory tissues of a salmonid fish. Journal of Experimental Biology 204 701–709.
Tipsmark CK, Luckenbach JA, Madsen SS & Borski RJ 2007 IGF-I and branchial IGF receptor expression and localization during salinity acclimation in striped bass. American Journal of Physiology 292 R535–R543.
Weisbart M, Chakraborti PK, Gallivan G & Eales JG 1987 Dynamics of cortisol receptor activity in the gills of brook trout, Salvelinus fontinalis, during seawater adaptation. General and Comparative Endocrinology 68 440–448.
Wu YC, Lin LY & Lee TH 2003 Na+,K+,2Cl− cotransporter: a novel marker for identifying freshwater- and seawater-type mitochondria-rich cells in gills of the euryhaline tilapia, Oreochromis mossambicus. Zoological Studies 42 186–192.
Young G, Mccormick SD, Bjornsson BT & Bern HA 1995 Circulating growth hormone, cortisol and thyroxine levels after 24 h seawater challenge of yearling coho salmon at different developmental stages. Aquaculture 136 371–384.
Zhou B, Kelly SP & Wood CM 2004 Response of developing cultured freshwater gill epithelia to gradual apical media dilution and hormone supplementation. Journal of Experimental Zoology 301A 867–881.
