Parathyroid hormone-related protein - Stanniocalcin antagonism in the regulation of bicarbonate secretion and calcium precipitation in a marine fish intestine

Bicarbonate secretion in the intestine (duodenum) of marine fish has been suggested to play a major role in the regulation of calcium availability for uptake. However, while the end process may lead to carbonate precipitation, the regulation of transport of calcium and/or bicarbonate may actually result in the fine-tuning of calcium availability for transport. To test this hypothesis sea bream (Sparus auratus) dudodenal preparations were mounted in Ussing type chambers and the effect of parathyroid hormone related protein (PTHrP) and stanniocalcin 1 (STC 1) on the control of intestinal bicarbonate secretion and calcium transport was analysed. As expected PTHrP increased net calcium uptake, as a result of an increase of calcium uptake without changes in calcium efflux. In contrast, purified sea bream STC 1 caused a minor decrease of calcium uptake, and a two to three fold increase in calcium efflux. As a result STC 1 was able to invert the calcium flux from net calcium uptake to net calcium loss, which is in keeping with its known actions as a hypocalcemic factor. Furthermore, both PTHrP and STC 1 regulate intestinal bicarbonate secretion. PTHrP increased calcium uptake and simultaneously reduced the single factor which induces calcium precipitation, bicarbonate secretion. In contrast, STC 1 while reversing the calcium net flux to make it secretory, promoted intestinal bicarbonate secretion, both actions directed to decrease the calcium gradient across the epithelium and promote immobilization in the form of bicarbonate in the intestinal lumen. Taken together our results provide robust evidence to support an antagonistic action of PTHrP and STC 1 in the fine control of movements of both calcium and bicarbonate in the intestine of seawater fish. INTRODUCTION Marine teleosts live in an environmental osmolality of circa 1000mOsm.kg while keeping their internal milieu at approximately 350 mOsm.kg. As part of the regulated osmoregulatory process, high rates of water ingestion have been described for marine fish (8). The fluid imbibed is processed along the intestinal canal with a primary step of net ion assimilation in the esophagus, that allows water absorption in the intestine by a process driven by NaCl via a Na-K-2Clcotransporter (24). In addition to water ingestion and absorption, the formation of carbonate aggregates in the intestine has been proposed to be central to the preservation of body fluid homeostasis in marine fish (21). Thus, calcium reaches the intestine at concentrations ranging from 7-15 mM (9, 39), which upon precipitation in the form of CaCO3 driven by HCO3 secretion lowers intestinal fluid osmolality between 15-25 mOsm.kg and favours water absorption (13). While in higher vertebrates duodenal HCO3 secretion is considered one of the most important defence mechanism against acid injury in the duodenum, the intestine of fish is able to produce a highly alkaline microenvironment that appears even in unfed fish (39), indicating that this process is dissociated or independent from digestion. This high alkalinity is the result of HCO3 secretion, and has the capacity to precipitate divalent cations (35, 37-39). Intracellular hydration of CO2 in the intestine epithelial cell plays an important role in the generation of HCO3 for apical secretion (14, 15). However, if bicarbonate is routed via the transcellular pathway, the organization of bicarbonate secretion entails a dual component of bicarbonate, one basolateral for internalization in the epithelial cell and the second to accomplish the actual secretion to the lumen of the intestine (21). Regardless of how it is generated, secreted bicarbonate immobilizes calcium in the form of calcium carbonate aggregates. As such, in addition to the proposed idea of water availability and enhancement of intestinal water absorption, bicarbonate secretion has also been suggested to participate in calcium metabolism (39). Accordingly, in order to prevent excessive calcium load, bicarbonate secretion would inhibit calcium uptake in the intestine by decreasing or annulling luminal calcium availability by precipitation. A lack of calcium absorption as a consequence has been proposed (37). However, this is not the case and net intestinal calcium absorption, has been shown to take place in the intestine of seawater fish by regulated processes (9, 30). In keeping with a calcium and bicarbonate cross regulation, the endocrine control of one or the other may play a crucial role in the substrate availability for carbonate aggregate formation. A model for calcium regulation, which incorporates a single endocrine factor, Stanniocalcin (STC), has been proposed in fish. It is justified on the basis of the nearly endless calcium pool available in seawater, the absence of parathyroid glands in fish, and the presence of fish specific Corpuscles of Stannius which produce STC a hypocalcemic or anti-hypercalcemic factor (3, 28, 34, 36). However, this model is thrown into doubt by: 1) the recent identification of Parathyroid hormone (PTH) and Parathyroid hormone related protein (PTHrP) in fish tissues and, 2) their demonstrated role in calcium regulation as hypercalcemic factors (2, 4, 9-11, 19, 25). In addition to its hypercalcemic nature PTHrP was shown to regulate drinking rates in one of our previous studies in the sea bream larvae (19) in seawater, a role in keeping with a potential regulation of imbibed fluid processing. Despite its physiological relevance (17, 38) and potential importance for other models of biomineralization, the process of bicarbonate secretion in seawater fish intestine has not been fully characterized at the mechanistic level (21), and no information is available at the regulatory endocrine level. We hypothesize that calcium and bicarbonate transport combine into a functional system and that the same endocrine factors that control calcium movements, also regulate directly or indirectly bicarbonate secretion in the intestine. The present study was designed to characterise calcium and bicarbonate transport in the intestinal epithelium in vitro and their regulation by PTHrP (hypercalcemic factor) and STC (hypocalcemic factor). MATERIALS AND METHODS Peptides and chemicals. The (1–34)PTHrP (2) from puffer fish was synthesized by Genemed Synthesis, Inc (San Francisco, CA). All chemicals were of the highest grade and obtained from Sigma-Aldrich (Madrid, Spain) unless stated otherwise. Animals Sea bream (Sparus auratus) were obtained as fry or juveniles from the stock of Ramalhete Experimental Marine Station (University of Algarve) and kept for at least 1 month before experimentation/sacrifice. Fish were held in 600 L seawater tanks (density <8 kg fish/tank) with flowing seawater (salinity 38‰; water temperature 18– 24oC), under natural photoperiod and fed twice a day (9 a.m. and 6 p.m.) with 1% body weight commercial dry pellets (PROVIMI, Portugal). All fish were fasted for 48 hours before experimental manipulations. For tissue collection fish were anaesthetized with 2-phenoxyethanol (1 ml.l water, Sigma-Aldrich, Madrid) and sacrificed by decapitation. All animal manipulations were carried out in compliance with the Guidelines of the European Union Council (86/609/EU) and Portuguese legislation for the use of laboratory animals. Purification of STC In vitro culture of corpuscles of Stannius. Corpuscles of Stannius were collected from 60 (0.8 to 1kg body weight) sea water adapted sea bream (anesthetised with 1 ml.l with 2-phenoxyethanol) and placed in several wells of a sterile 48-well plate with 0.2 ml each of modified Cortland’s saline culture medium: NaCl 160 mmol.l, KCl 2.55 mmol.l, CaCl2 1.56 mmol.l, MgSO4 0.93 mmol.l, NaHCO3 17.85 mmol.l, NaH2PO4 2.97 mmol.l, and glucose 5.55 mmol.l, pH 7.8, when equilibrated with a 99% O2 / 1% CO2 gas mixture. The saline was supplemented with 10 μl.ml of vitamins (MEM 100x Vitamins, Sigma-Aldrich), 20 μl.ml essential amino acids (MEM 50x, Sigma-Aldrich), 10 μl.mlnon essential amino acids (MEM 100x, SigmaAldrich), 10 μl.ml antibiotic (penicillin 10,000 IU.ml; streptomycin 10,000 UG.ml, GIBCO, Scotland) and 20 μl.ml L-glutamine (200 mM, Sigma-Aldrich). Plates were introduced into a chamber with a controlled gas atmosphere 99% O2 / 1% CO2 and gently shaken throughout their incubation at 22oC for 24 hours. After the culture, the corpuscles of Stannius and culture media were stored separately at -20oC for further processing. SDS-PAGE and Western-blot. Culture medium or homogenates of 3-4 corpuscles were fractionated by SDS-PAGE (15% polyacrylamide) under reducing conditions at 100 V and blotted onto ECL membranes (Hybond ECL, Amersham Biosciences, USA). The transfer was carried out in a vertical tank transfer system (Mighty Small Hoefer, Amersham Pharmacia) for 60 min at 100mA. The membrane was incubated with blocking solution (10% (w/v) milk; 0.1% (v/v) tween-20 in Tris buffered saline (TBS), overnight at 4oC. Subsequently, the membrane was incubated for 1.5 hr at room temperature with constant agitation with the primary antisera (anti-trout STC, kindly provided by Professor G Flik, Radboud University, Nijmegen, Holland) diluted 1:1000 in TBS. Excess antisera was removed and the membranes were washed 3 x 10 minutes in TBS /Tween 20 prior to addition of the secondary antibody (anti-rabbit IgG-peroxidase conjugate 1:2500; GE Healthcare, UK). Excess secondary antisera was removed and membranes were washed 3 x 10 minutes in TBST and developed in 3,3’ diaminobenzidine (0.2 mg/ml DAB in 100 mM Tris, pH 7.5) and H2O2 (0.003%) colour enhanced with Nickel (0.40 mg. ml NiCl2). Mass spectrophotometry of proteins. The putative band for STC identified by positive Western blot was excised from the gel and protein identification was conducted by the Centro de Genómica y Proteómica (Facultad de Farmacia, UCM, Spain). The proteins were subjected in-gel to trypsin digestion and the molecular weights of the peptides and amino acid sequences were analyzed by MALDI-TOF (Matrix Assisted Laser Desorption Ionization (MALDI) tandem Time-of-Flight (TOF) mass spectrophotometer), using the PostSource Decay (PSD) technique. Unambiguously identified peptides were further examined by MS/MS fragmentation. The resulting amino acid sequences were identified both by peptide mass fingerprinting and MS/MS fragmentation. Database searches using the peptide mass fingerprint integrated with tandem mass spectrometry (MS/MS) was performed using the MASCOT program (http://www.matrixscience.com) and BLASTP 2.2.15. Protein separation with continuous-elution electrophoresis. After confirmation of identity, culture medium from several fish (15 ml) were pooled and concentrated with Ultrafree-15 centrifugal filter devices with a molecular cut-off of 5 kDa (Millipore, Bedford, UK) and run on SDS–PAGE (12% polyacrylamide gels). Purification and analytical separation of proteins from concentrated culture medium was carried out by preparative SDS–PAGE under denaturing condition with continuous-elution electrophoresis using a Model 491 Prep Cell (Bio-Rad, Portugal). The resolving gel was composed of 15% polyacrylamide in 0.375 M Tris–HCl, pH 8.8. The stacking gel was 4% polyacrylamide in 0.125 M Tris–HCl, pH 6.8. The dimensions (height, mm) of the resolving gel and the stacking gel were approximately 54 mm and 28 mm, respectively. Electrophoresis was carried out using 0.025M Tris–HCl; 0.19 M glycine; 0.1% SDS, pH 8.3 as the running buffer and samples (3 ml of concentrated culture medium) were mixed with 1.5 ml sample buffer (100 mM Tris–HCl, pH 6.8; 200 mM DTT; 4% SDS; 0.2% bromophenol blue; 20% glycerol), heated at 100 oC for 5 min, centrifuged briefly, and run at a constant current of 20 mAmps over approximately 18 h maintaining a constant temperature (18 oC). Proteins fractionated on the polyacrylamide gel were allowed to migrate off the end of the gel into the elution chamber and collected as 2.5ml fractions at 8 oC. Fraction collection was initiated when the ion front (bromophenol blue) had migrated out of the gel. Every 10th fraction was analyzed by SDS–PAGE (15% polyacrylamide gels) and stained with silver nitrate in order to determine which fractions contained the proteins of interest (identified on the basis of size). Additional Western blots were performed to identify first the range and second the individual fractions reacting with the anti-STC antibody. Fractions containing the putative hormone, i.e., giving a single clear band, were pooled. Concentration of pooled fractions and substitution of buffer was achieved by means of ULTRAFREE-15 centrifugal filter devices with ultrafiltration membranes (MILLIPORE) against 20 mM Tris-HCl, pH 7.5 by several steps of centrifugation at 1900g 4oC for 40 minutes. Short circuit current measurements The general methods used for Ussing chamber experiments have previously been explained (9), with the exception that the short circuit current vales were presented as absolute values, not considering the sign of the current injected. Here the current is presented considering the voltage referenced to the apical side of the preparation. . Sea bream (80-100g) were sacrificed by decapitation and the anterior intestine (the region corresponding to the duodenum and comprising an homogenous region of about 3 cm in length distal to the pyloric caeca) was carefully dissected out, and transferred to chilled freshly prepared and gassed (10 mM HCO399:1 O2/CO2) basolateral saline (see Table 1 for composition). The intestinal portions were defatted, cleaned with fresh saline and opened longitudinally to produce a flat sheet. The mucosal crypts were separated to minimize the unstirred boundary layer, and mounted (with apical and basolateral sides identified) by pinning over the circular aperture (0.5 or 1 cm) of an Ussing half-chamber between 2 parafilm gaskets to minimize edge damage. The Ussing chamber was assembled and 3 to 5 ml of basolateral saline was added to each hemi-chamber. The saline in the chambers was gassed with a 99:1 O2/CO2 mix to provide oxygenation, good mixing by gas lift and pH control to 7.8. Temperature was maintained between 21-22oC throughout the experiments. The preparations were left to stand for at least 60 min or until a steady basal measurement of bioelectrical variables was achieved. Measurement of bioelectrical variables was performed in symmetric conditions to avoid non-zero junction potentials using the saline compositions described in Table 1. Short-circuit current (Isc, μA.cm) was monitored by clamping of epithelia to 0 mV. All bioelectrical variables were monitored by means of Ag/AgCl electrodes (with tip asymmetry <1mvolt) connected to either side of the Ussing chamber with 3 mm bore agar bridges (3 M KCl in 3% agar). Depending on the clamp amplifier used epithelial resistance Rt (Ω.cm) was automatically calculated from voltage deflections after injecting bipolar 200 msec pulses of 50 μAmp every minute or manually calculated by (Ohm’s law) using the current deflections induced by a 2 second 1 mV pulse every minute. Recording of variables was performed by means of a microcomputercontrolled voltage/current Clamp Electronic with automatic correction for fluid resistance and electrode voltage asymmetry (KMSCI, Aachen, Germany) or a DVC1000 voltage clamp amplifier (WPI, Sarasota, US). Intestinal bicarbonate secretion (BCS) Duodenum preparations were mounted as described before with the following exceptions: a) both apical and basolateral sides of the preparation received 3 to 5 ml of the corresponding saline either apical or basolateral (Table I) to simulate in vivo conditions; b) the gas/HCO3 and Hepes (4-(2-Hydroxyethyl)piperazine-1ethanesulfonic acid sodium salt, N-(2-Hydroxyethyl)piperazine-N′-(2-ethanesulfonic acid) sodium salt) was modified in the basolateral chamber (Table I for details),,while the apical side received 100% O2.; c) the preparations in the Ussing type chamber were current clamped to 0 μA/cm. For the purpose of BCS measurement, preparations were left to achieve a transepithelial steady state voltage (generally around 30 min) at which point the saline was replaced in both chambers and left undisturbed for 1.5 hours and then the apical saline replaced and the hormones incorporated into the basolateral chamber to give a final concentration of 1 μg.ml. The preparation was left undisturbed for an additional period of 1.5 hours and a final collection of apical saline performed at this point. Bicarbonate secretion in response to basolateral hormones was measured using a combination of 10 mM HCO3/1% CO2 in O2 in the basolateral saline. A single preparation was used for two consecutive measurements one in the absence (control period) and one in the presence (hormone effect) of hormones. Whole samples of apical saline for either controlor hormone-treated periods were transferred to small titration vessels with a magnetic stirrer and constant gassing (O2). The samples were gassed for 30 min to remove CO2 and titrated to pH 3.8 with 10mM HCl, an additional gassing period of 20 min was applied to remove any remaining CO2 and the sample was back titrated to its original pH with 10mM NaOH. BCS was calculated from the difference in the volume of HCl and NaOH needed for either titration and considering the molarity of the titrant, the time elapsed and the Ussing chamber opening. BCS is expressed as as nmol.h.cm Intestinal calcium fluxes The Ussing chamber (0.5 cm opening) was assembled exactly as described for BCS with asymmetrical saline (basolateral 10 mM HCO3/1% CO2 in O2) and gas in the apical and basolateral half-chambers. Preparations were left in saline for 30 minutes, followed by a further 15 min in new, freshly gassed saline. For the measurement of calcium uptake Ca (CaCl2 0.2μCi NEN, Life sciences Products) was added, to the apical (uptake) or basolateral (efflux) sides and left to mix for 15 min after addition of tracer. Saline (100μl) from the cold half-chamber was collected and replaced by 100μl fresh saline (time 0) followed by similar procedures at 30 min intervals for the duration of the experiments comprising 2 consecutive periods of 1.5h in the absence (control period) or the presence of either STC or (1-34) PTHrP on the basolateral side. Samples (100μl) were also collected from the Ca labelled saline at time 0 and at the end of the experiments for calculation of specific activities. All radiotracer experiments were performed under current clamp to 0 μA/cm. The quality of experiments was checked for initial velocity and linearity of flux measurements. At the end of the experimental period to confirm the physical integrity of the tissue the volume of one of the half-chambers was removed to verify that no fluid moved from the full to the empty half-chamber. Unidirectional fluxes were calculated according to the following equations: [1] Calcium uptake Jin (nmol.cm.h)=Δ[Ca]Bl(1/SAAp)[VolumeBl]/[(Time)(Area)], where Δ[Ca]Bl represents the increase in radioactivity on the basolateral halfchamber and SAAp represents the apical side specific activity (cpm.nmol). [2] Calcium efflux Jout (nmol.cm.h)=Δ[Ca]Ap(1/SABl)[VolumeAp]/[(Time)(Area)], where Δ[Ca]Ap represents the increase in radioactivity on the apical half-chamber and SABl represents the basolateral side specific activity (cpm.nmol-). [3] Calcium net flux Jnet(nmol.cm.h)= Jin Jout Statistics Results are shown as mean ± standard error of the mean (SEM) unless otherwise stated. After assessing normality and homogeneity of variances differences between groups were established by paired Student’s t-test or one-way analysis of variance followed by the post hoc Bonferroni test to identify significantly different groups. All statistical analysis were performed using the statistical package SPSS (SPSS Inc, Chicago, IL, USA). Groups were considered significantly different at p<0.05, unless stated otherwise. RESULTS Sea bream STC purification A single STC immunoreactive band of 25 Kd was identified by Western blot of the medium used for culture of sea bream corpuscles of Stannius in vitro (Figure 1). The sequences obtained from the mass spectrometry profile of the fragmented protein corresponded to 89 amino acids. A MASCOT blast search of the amino acid sequences obtained revealed it shared greatest similarity with STC 1. Sequence comparison shows a high degree of similarity between the partial sea bream STC 1 sequences and those from fish of diverse orders (Figure 1). One of the features of the STC family is the presence of 11 cysteine residues important for protein dimerization and after alignment of the sea bream tryptic peptides cysteine residues were identified in positions 99, 115 and 171, the latter is important for protein dimerization. Characteristic sites for protein kinase c phosphorylation (residues 102-104) and casein kinase II phosphorylation (residues 36-39 and 142145) were also identified in the sequence of the tryptic peptides. Stannius glands from 60 sea bream yielded 1,750 μg of secreted STC. Bioelectrical characterization of sea bream intestine under voltage clamp The sea bream duodenum was mounted in Ussing chambers, with the same saline (Table I) bathing the apical and basolateral sides, generates a small Potential Difference (PD) of 1.13±0.12 mvolts (mucosa reference). To voltage clamp the tissue to 0 mvolts the injection of a small, but consistent current (short circuit current, Isc) of around -9 μA.cm was required (Table II). Control preparations sustained constant Isc (μA.cm,Figure 2A) and Gt (mS.cm ,Figure 2B) for the duration of the experimental periods. To test the effects of (134)PTHrP and STC 1 in Isc, the hormones were applied to the basolateral side of the preparation as a single dose of 1 μg.ml and the Isc was followed up to 1 hour after addition. The application of (1-34)PTHrP basolaterally had a small effect on Isc, with a significant decrease (more negative) after 40 min of application (Figure 2C). No change was observed in Gt in response to (1-34)PTHrP (Figure 2D). The application of sea bream STC 1 resulted in a slow but consistent increase/inversion of the Isc (p<0.01, one-way ANOVA) from 25 min after addition up to the end of the recording period (Figure 2E). In addition Gt, was slightly, but significantly, increased after the application of basolateral STC 1 (Figure 2F). Current clamped sea bream intestine Basal variables of sea bream duodenum under current clamp conditions are shown in Table III. Manipulation of bicarbonate availability in the basolateral saline significantly modifies apical bicarbonate secretion in the sea bream duodenum (Figure 3). The combination of 10 mM Hepes and O2 (0 CO2, 0 mM HCO3) results in an apical bicarbonate secretion of 183±26 nmol.cmh which is not significantly different from the secretion obtained in the presence of 5 mM Hepes 5 mM HCO3and 0.3% CO2. In contrast, apical bicarbonate secretion was significantly higher (p<0.01, one-way ANOVA) in the presence of 10 mM HCO3with 1% CO2. The latter combination of saline was used in the basolateral chamber for all further experimentation under current clamp, thus allowing transcellular HCO3 movements to be represented in the total bicarbonate secretion measured. The application of basolateral (1-34)PTHrP (1 μg.ml) resulted in a significant decrease in bicarbonate secretion, to values less than half those of the control period (Figure 4). In contrast, the application of basolateral STC 1 (p<0.01, One-way ANOVA; 1 μg.ml) resulted in a more than two-fold significant increase in bicarbonate secretion (p<0.01, One-way ANOVA). In control preparations, bicarbonate secretion was determined in consecutive periods in the absence of hormones showing no significant variation, confirming the specificity of the hormonal effects (Figure 4). Calcium fluxes determined in current clamped sea bream duodenum are shown in Figure 5. Application of basolateral (1-34)PTHrP (1μg.ml) caused a significant increase in calcium uptake (p<0.01, One-way ANOVA) and a slight but not significant decrease of calcium efflux, with the net result of an about 5 fold increase in calcium uptake (Figure 5). In turn, application of basolateral STC 1 (1 μg.ml) had no effect in calcium uptake and a induced a significant increase of about two-fold in calcium out-flux (p<0.01, One-way ANOVA), which resulted in a significant decrease of net calcium transport (p<0.01, One-way ANOVA) which becomes outside directed (Figure 5). DISCUSSION The present study establishes for the first time the role of the calcitropic hormones STC 1 and PTHrP in both the regulation of calcium movements and bicarbonate secretion in the intestine of marine fish. These results support the hypothesis that calcium regulating factors may have, in addition to their calcitropic actions, a role in the general osmoregulatory process of marine fish. Assuming the premise that the formation of CaCO3 aggregates in the intestine drives fluid processing and subsequent water absorption (13) the endocrine regulation of bicarbonate secretion by calcitropic factors may have a central role in this process. The short-term effects of both PTHrP and STC 1 in the regulation of calcium transport are in good agreement with previous studies. PTHrP achieves its hypercalcemic action in the intestine by the combination of a 3 fold increase in calcium uptake and a slight increase of calcium efflux that results in a 3 fold increase of net calcium uptake (Figure 5). The results obtained are in agreement with previous in vivo (2, 10, 11, 19) and in vitro results (9) and indicates a mechanism consistent with hypercalcemic actions of PTHrP in fish. The effect of purified sea bream STC 1 on calcium transport in vitro demonstrates its bioactivity (Figure 5), and indicates a mechanism for its hypocalcemic action demonstrated in other in vivo and in vitro models (3, 28, 34, 36). In our in vitro system, the intestinal absorptive calcium pathway was unaffected by application of STC 1 (Figure 5), which is surprising since among the targets of STC 1 are the apical epithelial calcium channels, the limiting step for calcium internalization in the cell, as revealed by the stimulation of epithelial channels in morpholino knockdown of STC 1 in zebra fish embryos (31). A possible explanation for this divergence would be the relatively minor role of the intestine in freshwater zebra fish larvae compared to its importance in the marine sea bream (18) where the extra intestinal component of calcium uptake varies as a function of salinity from >90% at 0-10% seawater to 40-50% in full strength seawater. In other model systems such as the mammalian intestine, the addition of STC 1 to in vitro intestinal preparations induces a net decrease in calcium absorption, which surprisingly is the result of simultaneous increases in both absorptive and secretory calcium pathways (22). The hypocalcemic effect of STC 1 in the intestine of the sea bream in vitro is not achieved by inhibition of the uptake component, but instead by a 3 fold stimulation of calcium efflux (Figure 5). As a consequence, the net calcium transport becomes negative and the intestine becomes primarily secretory in relation to calcium (Figure 5). This result is surprising as it happens against a 5-fold concentration gradient (Table I) and is achieved by an as yet unidentified molecular mechanism, although the secretory pathway Ca-ATPase (SPCA) is the most likely candidate. The SPCA functions in models where calcium transfer for secretion is functionally important such as the mammary gland (32) and where the epithelial cells must transport large amounts of calcium against a large concentration gradient which is similar to the situation in the fish intestine. In addition, the SPCA is present in all the segments of the gastrointestinal system of mammals (33), where calcium is primarily processed. While calcium efflux has been routinely described in several studies, the calcium secretory pathway remains uncharacterized both at a mechanistic and regulatory level in fish. Database searches with the mammalian SPCA sequence (NCBI Accession No: NM_001001487) identified expressed sequence tags with high homology in piscine species such as the euryhaline killifish (Fundulus heteroclitus, Accession No: DR441449 and DR441450), euryhaline rainbow trout (Oncorhynchus mykiss, Accession No: DT957416 and DT956043, the freshwater stickleback (Gasterosteus aculeatus, Accession No: DT957416 and DT956043), the euryhaline medaka (Oryzias latipes, Accession No: DK011462) and the fresh water zebra fish (Danio rerio, Accession No: EH541307). This further supports the notion that this pathway is also active in fish. The sea bream intestine secretes bicarbonate at a rate in the range of 185 to 450 nmol.cmh depending on the combination of CO2/HCO3 in the basolateral saline (Figure 3). In the saline used for most of the experiments in this study (1%CO2/ 10 mM HCO3) roughly 50% of the bicarbonate secretion is likely driven via transcellular pathways; while the other 50% is likely produced by hydration of CO2 in the epithelial cell (as shown in preparations devoid of CO2/HCO3). The values of intestinal bicarbonate secretions obtained in this study in the sea bream are well within the range of those described for other marine species using pH-stat methods (1, 14). Validation of the robustness of the preparation of sea bream intestine was provided by consecutive 1.5h measurements with virtually the same bicarbonate secretion rates (Figure 4). In addition, bicarbonate secretion in the sea bream intestine is hormonally regulated as shown by the effects of PTHrP and STC 1. Accordingly, while the dose used (1 μg/ml) PTHrP significantly reduces bicarbonate secretion STC 1 addition raises bicarbonate secretion between 2 and 3 fold (Figure 4). To our knowledge this is the first report on the effect of calcitropic hormones in intestinal bicarbonate secretion in vertebrates. The mechanisms responsive to both PTHrP and STC 1 in the regulation of calcium and bicarbonate secretion so far remain unknown. The key calcium transport mechanisms involved in the calcium transport cascades (5-7, 20, 26, 27) i.e. apical epithelial calcium channels (31) and basolateral Na/Ca exchanger and CaATPase could adjust bicarbonate secretion by indirect regulation of calcium availability. In addition, there is a long list of potential candidates involved in bicarbonate transport in mammalian models such as: Na/H Exchangers, HATPase, Na-HCO3 co-transporters, anion exchangers, carbonic anhydrase and even the apical cystic fibrosis transmembrane regulator (29). In the marine fish intestine only recently have some of the key elements of bicarbonate secretion been functionally confirmed: H-ATPase (V-type), Na/H exchanger 3, Na-HCO3 cotransporter 1 and two carbonic anhydrase isoforms (15-17). In addition, the presence and functional analysis (in Xenopus oocytes) of apical Slc26a6A and Slc26a6B (putative Cl/HCO3 exchangers) and basolateral NBCe1 (Na-HCO3 co-transporter) confirms the apical mechanisms and the existence of transcellular bicarbonate movement (21). On the other hand, the timing and Isc reversal obtained in voltage clamp experiments in response to STC (Figure 2), resembles the Cystic fibrosis transmembrane regulator (CFTR) response to ionomycin obtained in the intestine of seawater Fundulus heteroclitus (23). Considering the relationship between CFTR and the SLC26 family of chloride-bicarbonate exchangers in higher vertebrates (12) it would be tempting to suggest a regulatory action of STC on CFTR to explain enhanced Isc in voltage clamp experiments and bicarbonate secretion in the present