The role of adenylate cyclase (AC) in the maintenance of the hydroosmotic response to serosal hypertonicity (SH) in anuran urinary bladder is disputed. In this study, norepinephrine (NE) significantly reversed the hydroosmotic response of Rana temporaria bladders in hypertonic medium (330 mosmol/kgH2O). The reversal was inhibited by yohimbine but was unaffected by prazosin and propranolol, indicating that NE action was mediated via α2-adrenergic receptors. Preincubation of bladders with indomethacin did not interfere with the inhibitory action of NE, contraindicating a role for prostaglandins. The SH hydroosmotic response was abolished in the presence of 5-n-ethyl-N-isopropyl amiloride (EIPA), but the antidiuretic hormone (ADH) hydroosmotic response was not. EIPA inhibits Na+/H+, known to be activated by cell shrinkage. An investigation of the anionic requirement of the SH hydroosmotic response revealed that replacement of bath Cl− with the nonpermeable anion gluconate reversibly abolished this response. In contrast, the hydroosmotic response to ADH was unaffected by Cl− removal; however, when Cl− was absent, it was no longer augmented in hypertonic bath. The SH response was inhibited by the Cl− channel blocker 5-nitro-2-(3-phenylpropylamino)benzoate but not by the Na/K/2Cl inhibitor bumetanide. Our results show that not only the onset but also the maintenance of the SH hydroosmotic response is dependent on AC activity and does not differ in this respect to the ADH hydroosmotic response. The effect of modifying extracellular Cl− concentration, suggests that this anion, possibly functionally linked with Na+/H+ activity, may be involved in invoking the SH hydroosmotic response in anuran urinary bladder.
- water balance
- extracellular chloride
frogs and toads (anurans) are constantly under the threat of dehydration because of the high rate of evaporative water loss through their skin. With some exceptions (38), anurans do not hydrate orally; water balance is maintained by absorption of water (dermal drinking) through a specialized area of skin on the ventral surface, called the pelvic patch, and by reabsorption of fluid from the bladder (28). Transepithelial water movement in these tissues is through the granular cells, which are functionally analogous to the mammalian principal cell of the renal collecting duct. Although antidiuretic hormone (ADH) has been considered to regulate these events (9, 51), there are reports that anuran water balance can be maintained after ablation of the posterior pituitary (2, 55, 61).
ADH invokes a reversible increase in the permeability of the apical membranes of principal cells in the mammalian collecting duct and granular cells in anuran skin and bladder, allowing water movement across these tight epithelia. This hydroosmotic action of ADH is mediated via an increase in intracellular cAMP concentration (15, 22, 33). In the mammalian principal cell, ADH activation of V2 receptors causes a trafficking of vesicles, carrying the water channel protein aquaporin-2 to and from the apical membrane (39). In anuran granular cells, ADH causes a similar trafficking of vesicles (aggrephores) to and from the apical membrane (59), believed to carry water channels (7, 10, 37). The serosal hypertonicity (SH)-regulated water channel in anuran bladder remains to be cloned.
Bentley (4) described a reversible increase in water flux across toad bladder in response to hypertonic serosal bathing solutions that did not depend on the presence of ADH. Exposure of this tissue to SH results in the appearance of luminal membrane aggregates (14, 29) believed to be water channel proteins. SH responses in anuran bladder are related to a shrinkage effect (4), which led to the hypothesis that this hydroosmotic response is nonspecific and caused by concentration of intermediates involved in the ADH hydroosmotic response. However, researchers have shown that the onset (but not the maintenance) of the SH hydroosmotic response depends on an upregulation of adenylate cyclase (AC) activity, which can be pharmacologically modulated, indicating the involvement of a specific biological mechanism (48). The first aim of this study was to examine the role of AC in maintaining the SH hydroosmotic response in anuran bladder (Rana temporaria).
The serosal factor, or factors, involved in invoking the SH hydroosmotic response has yet to be elucidated. The response to SH is present but substantially lowered in sodium- and Ca2+-free and low-pH (6.6) serosal bathing medium (23). Similar “cationic” requirements are reported for the ADH-invoked hydroosmotic response (23). In contrast, no information exists on the effect of removal of the predominant medium anion, Cl−, on the SH hydroosmotic response. In light of this fact and the fact that activation of AC by halide ions (46), including Cl− (30), has been reported, the second aim of this study was to examine the importance of Cl− in the SH water response in anuran bladder.
In toad bladder, granular cell volume was initially reduced and thereafter returned to control values in the continued presence of hypertonic serosal perfusate (29). Cells, which have been osmotically shrunken, execute an acute response by a rapid uptake of NaCl (24). Na+/H+ exchangers (NHE) are known to be activated by cell shrinkage (5, 41). A third aim of this study was to investigate whether 5-n-ethyl-N-isopropyl amiloride (EIPA), a specific inhibitor of NHE, would have any effect on the functional response of anuran bladder to hypertonic serosal medium.
Male Rana temporaria, a frog indigenous to Ireland (supplied by Dennis McCarthy, Kells, County Meath, Ireland), were used in this study. Animals were killed by stunning, decerebration, and double pithing. Because bladders of R. temporaria are too small for catheterization, a modified version of Bentley's (3) gravimetric method for toad bladders was used. Bladders were always inflated with a hypotonic Ringer solution of tonicity dictated by the protocol [23 mosmol/kgH2O for the arginine vasopressin (AVP) protocol or 115 mosmol/kgH2O for the SH protocol; see below] by insertion of a fluid-filled syringe into the cloaca. The bladder was freed from its supporting mesenteric tissue, tied off with cotton thread, and dissected free to give a bilobed sac. This sac was immersed in a fixed volume (20 ml) of either isotonic (230 mosmol/kgH2O) or hypertonic (330 mosmol/kgH2O) Ringer solution, according to the protocol (see below). A hypertonic solution of 330 mosmol/kgH2O was used, as an increase of tonicity by 100 mosmol/kgH2O is within tolerated physiological values reported for the extracellular fluid of dehydrated anurans (54).
The test phase in all experiments was conducted with an osmotic gradient of ∼200 mosmol/kgH2O across the tissue. Water movement was detected by a change in the weight of the sac. Bladders were weighed every 5 min; weight change in milligrams was converted to microliters, taking a density of 1.0 for fluid lost. Water flow is expressed as microliters per minute. Water bath temperature was maintained at 19°C (25).
Results are expressed as means ± SE; n represents the number of independent experiments. Statistical analysis was carried out using unpaired Student's t-test with the exception of the data relating to Figs. 1B and ⇓⇓⇓⇓⇓⇓⇓9, A and B, where paired Student's t-test was employed as the bladders acted as their own controls. Differences were considered significant at P < 0.05.
Ringer solutions of varying tonicity were prepared in deionized water; pH was adjusted to 7.9 with HCl or NaOH to conform to reported values for anuran plasma pH (25).
Serosal (bath) solutions.
Isotonic Ringer (230 ± 5 mosmol/kgH2O) contained (in mmol/l) 114.5 Na+, 119.0 Cl−, 3.7 K+, 1.0 Mg2+, 2.5 HCO3−, 0.9 Ca2+, 3.5 HEPES, 0.5 phosphate, and 5.0 glucose.
Hypertonic Ringer (330 ± 5 mosmol/kgH2O) was prepared by addition of either 50 mM NaCl-SH (NaCl) or 100 mM mannitol-SH (mannitol) to the isotonic Ringer recipe.
Cl−-free solutions were prepared by equimolar substitution of gluconate salts for Cl− salts, in the standard isotonic or hypertonic recipes. Adjustment of pH to 7.9 was with H2SO4 instead of HCl. As gluconate chelates Ca2+ (11), Ca2+ concentration was raised to 4.5 mol/l to compensate for reduction in activity of this ion. Control experiments using standard isotonic and hypertonic solutions containing 4.5 mmol/l Ca2+ confirmed that, at this concentration, there was no observable effect of the ion on either of the hydroosmotic responses.
Solutions were prepared by reduction of NaCl concentration in the isotonic recipe to 6 mM (23 ± 3 mosmol/kgH2O) for AVP experiments or 56 mM (115 ± 5 mosmol/kgH2O) for SH experiments.
Because the sensitivity of anuran urinary bladder to ADH is altered by sequential increasing doses (16), classical dose-response curves are not possible. In separate bladder groups, responses were observed to 1, 2, 5, 10, 20, and 50 mU/ml AVP (unpublished observations). Maximal hydroosmotic responses to AVP were observed at 20 mU/ml.
HgCl2 was used at a final serosal bath concentration of 100 μM. Bumetanide, 5-nitro-2-(3-phenylpropylamino)benzoate (NPPB), and EIPA were used at bath concentrations of 100, 50, and 100 μM, respectively. Final serosal bath concentrations of other test agents were as follows: 50 μM norepinephrine (NE); 100 μM prazosin (a selective α1-antagonist); 100 μM yohimbine (a selective α2-antagonist); 100 μM propranolol (a nonselective β-antagonist); and 10 μM indomethacin.
All chemicals were supplied by Sigma-Aldrich, Ireland, with the exception of NE (Winthrop Laboratories). Osmolality was measured on a model 3D3 Advanced Instruments Osmometer.
AVP protocol (isotonic bath).
Luminal fluid was 23 mosmol/kgH2O; bath fluid was 230 mosmol/kgH2O. At time (t) = 40 min, bladders were placed in fresh isotonic bath containing AVP (20 mU/ml); transmembrane osmotic gradient was ≅200 mosmol/kgH2O. Washout was at t = 70 min, at which time bladders were placed in fresh isotonic bath.
SH protocol (hypertonic bath).
Luminal fluid was 115 mosmol/kgH2O; bath fluid was 230 mosmol/kgH2O. At t = 40 min, bladders were placed in hypertonic bath (330 mosmol/kgH2O); transmembrane osmotic gradient was ≅200 mosmol/kgH2O. Washout was at t = 70 min, at which time bladders were placed in fresh isotonic bath. Bladder exposure to SH and AVP combined was at t = 40 min; transmembrane osmotic gradient ≅200 mosmol/kgH2O.
Repeated exposure to a hydroosmotic agent.
In experiments where bladders were exposed to either AVP or SH for a second time, this was after a second control phase (t = 70–100 min) in fresh isotonic bath.
Hydroosmotic Response to SH and ADH
As seen in other anuran species, when R. temporaria bladder is exposed to either ADH or hypertonic bath (AVP and SH protocol respectively, cf. methods), the osmotic water permeability of the tissue increases, and water leaves the bladder along the transmural osmotic gradient, which is of the order of 200 mosmol/kgH2O in the test phase of both protocols (Fig. 1A). When this tissue is once again exposed to a hydroosmotic agent (Fig. 1B) be it AVP (20 mU/ml) or SH (330 mosmol/kgH2O), having been in fresh isotonic medium for 30 min, the osmotic permeability of the tissue again increases to a similar, if not slightly larger (but not statistically significant), degree. In separate bladder groups, first and second exposure exhibited mean fluid losses of 17.3 ± 2.8 and 20.2 ± 2.5 μl/min, 13.4 ± 2.7 and 15.8 ± 1.7 μl/min, 14.5 ± 2.7 and 15.4 ± 1.0 μl/min in isotonic bath + AVP, SH (NaCl), and SH (mannitol), respectively. In all three experimental groups, first and second exposure magnitudes were not statistically different (P > 0.05, paired t-test).
Determination of Whether Fluid Movement is Trans- or Paraepithelial
HgCl2 (100 μM) blocked fluid loss in hypertonic bathing medium (Fig. 2). Test bladders exhibited a hydroosmotic response with a mean flux of 20.4 ± 2.4 μl/min over the 30-min interval (t = 40–70 min) of the first exposure to SH. This value did not differ from that seen in control bladders (24.0 ± 5.9 μl/min) over the same time interval (P > 0.05, unpaired t-test). When the test bladders were again exposed to hypertonic medium, this time in the presence of HgCl2, the hydroosmotic response was significantly lower (2.1 ± 0.5 μl/min) than that seen in the control group (24.8 ± 4.8 μl/min; P < 0.01 unpaired t-test). HgCl2 did not disrupt the integrity of the epithelium, since analysis of test bladder fluid at the end of experiments showed that the transmural osmotic gradient was intact.
Investigation of a Role for Endogenous Prostaglandins in Invoking the SH Hydroosmotic Response
Test bladders were preincubated in isotonic medium containing indomethacin (10 μM) for 120 min (Fig. 3). The bathing medium was refreshed at t = 60 min to remove prostaglandins released by the bladder tissue into the bath (19, 62). Control bladders were preincubated in isotonic medium alone for 120 min. Weighings commenced at t = 90 min.
Mean fluid loss over the 30-min interval (t = 90–120 min) in isotonic bath was significantly higher in indomethacin-treated bladders than in the control group (11.9 ± 1.8 vs. 5.6 ± 1.8 μl/min; P < 0.04, unpaired t-test). Exposure to hypertonic bath for 30 min at t = 120–150 min (in the continued presence of indomethacin for the test group) showed similar large increases in osmotic water permeability: 39.3 ± 2.2 μl/min for the test group vs. 37.0 ± 2.9 μl/min for the control; P > 0.05, unpaired t-test. The SH response was reversed when bladders were placed in isotonic bath at t = 150 min (in the continued presence of indomethacin for the test group).
Determination of Effect of NE on the Developed Hydroosmotic Response in Hypertonic Bath
NE (50 μM) added to hypertonic (NaCl) bath at the peak of the hydroosmotic response (t = 115 min; 2nd exposure, after 30 min in isotonic bath) caused fluid loss to drop to basal levels: 7.9 ± 0.4 vs. 20.3 ± 4.0 μl/min in the control group over the time interval t = 120–140 min (P < 0.05, unpaired t-test; Fig. 4A).
A similar reversal (Fig. 4B) was seen when bladders undergoing a hyperosmotic response in hypertonic (mannitol) bath were exposed to NE (50 μM) at t = 115 min (2nd exposure, after 30 min in isotonic bath). Water permeability decreased in test bladders in comparison with control values over the time interval t = 120–140 min (5.7 ± 0.4 vs. 14.1 ± 1.2 μl/min, respectively, P < 0.005, unpaired t-test).
Because there is evidence of direct effects of mannitol on cell physiology (26, 31, 60) and because the increase of tonicity by addition of NaCl was more physiological, all subsequent investigations of the SH hydroosmotic response were with SH-NaCl bathing medium.
Determination of Receptor Type Involved in the Inhibitory Action of NE
Bladders (Fig. 5, A-C) exposed to SH alone or SH with addition of NE (50 μM) at the height of the hydroosmotic response to a second exposure to hypertonic bath (after a 30-min interval in fresh isotonic bath) functioned as control groups in a series of experiments to investigate the adrenergic receptor type involved in the NE reversal of the SH hydroosmotic response.
The selective α1-adrenergic receptor antagonist prazosin (100 μM) did not interfere with the reversal of the hydroosmotic response (Fig. 5A) by bath addition of NE (50 μM). Mean fluid loss over the 20-min interval (t = 120–140 min) after bladder exposure to NE was 5.7 ± 0.9 μl/min, which did not differ from the mean fluid loss (7.9 ± 0.4 μl/min) in bladders exposed to NE in its absence (P > 0.05; unpaired t-test; Fig. 5A).
In contrast, yohimbine (100 μM), which is a selective α2-adrenergic receptor antagonist, inhibited the reversal of the developed SH hydroosmotic response by NE (50 μM; Fig. 5B). Mean fluid loss over the 20-min interval after bath addition of NE at t = 115 min was 15.4 ± 2.6 μl/min, which did not significantly differ from the magnitude of the fluid loss (17.4 ± 1.7 μl/min) in the control group exposed to SH alone (P > 0.05, unpaired t-test).
Propranolol (100 μM), which blocks β-adrenergic receptors in a nonselective manner, had no effect on the reversal by NE (50 μM) of the hydroosmotic response (Fig. 5C). Mean fluid loss was equally decreased in the presence (10.3 ± 1.3 μl/min) and absence (7.9 ± 0.41.3 μl/min) of the β-blocker (P > 0.05, unpaired t-test).
Determination of Whether the Inhibitory Effect of NE via α2-Adrenergic Receptor Activation was Mediated Indirectly by Endogenous Prostaglandins
It was necessary to assess a possible role for prostaglandins in the inhibitory effect of NE on the developed SH hydroosmotic response. Bladders preincubated (120 min) with indomethacin (10 μM) exhibited a significant reversal of the hydroosmotic response when exposed to NE (50 μM) at the height of the response in hypertonic bath, in the continued presence of indomethacin (Fig. 5D). Mean fluid loss decreased to 6.1 ± 0.2 μl/min, which differed significantly from the mean value of 21.5 ± 2.1 μl/min recorded in control bladders (Fig. 5D) exposed to SH alone, in the continued presence of indomethacin, over the same time interval (P < 0.003, unpaired t-test). These findings contraindicate a role for prostaglandins and point to a direct effect of NE on the developed response by inhibition of AC activity.
Determination of the Anionic Requirements of SH Hydroosmotic Response
Bladders placed in “Cl−-free” hyperosmotic bath (Fig. 6A), in which medium Cl− had been substituted with the nonpermeant anion gluconate, showed a significantly different response to hypertonicity when compared with control bladders exposed to standard hypertonic bath; mean fluid loss over the 30-min exposure to SH was 3.6 ± 0.9 and 16.1 ± 1.8 μl/min, respectively (P < 0.005, unpaired t-test).
Analysis of luminal fluid at term of experiments in hyperosmotic Cl−-free gluconate bath (Fig. 6A) showed that Na+, K+, and Cl− concentrations and osmolality had not significantly changed from their corresponding values in the hypotonic fluid used to fill the bladders at the start of the experiments. Thus gluconate had not compromised junctional integrity, in agreement with earlier observations (21, 35, 57).
In a further set of experiments (Fig. 6B), bladders were exposed to hyperosmotic bath in which there had been a 50% replacement of Cl− with gluconate; mean fluid loss was 12.5 ± 2.5 μl/min. This response was significantly lower than the hydroosmotic control bladders in standard hypertonic bath (23.3 ± 2.0 μl/min; P < 0.01, unpaired t-test) but significantly larger than the response of bladders in Cl−-free bath (5.0 ± 0.7 μl/min; P < 0.005, unpaired t-test) over the 30-min exposure of the three groups to hypertonicity.
Determination of Whether the Abolition of the SH Hydroosmotic Response by Complete Cl− Removal Can be Reversed by Readdition of Cl− to the Bathing Medium
After a control period (t = 0–40 min) in standard isotonic bath, bladders were placed in Cl−-free hyperosmotic bath for 30 min (Fig. 6C). At t = 70 min, three bladders (test) were placed in standard hypertonic bath, while the remaining three (control) were placed in fresh Cl−-free hyperosmotic medium. With Cl− present in the bathing medium, test bladders had a substantial fluid loss within 10 min (t = 80 min) compared with the continued poor response seen in the control bladders (12.7 ± 2.5 vs. 4.7 ± 1.1 μl/min, respectively, P < 0.04, unpaired t-test).
Replacement of Cl− in serosal bath by the nonpermeant anion gluconate causes cell loss of Cl−, accompanied by potassium to maintain electroneutrality (35). The possibility that an indirect mechanism involving intracellular K+ concentration might be responsible for the abolition of the SH response in Cl−-free bath was contraindicated in experiments carried out to measure the hydroosmotic response of bladders exposed to “K+-free” hypertonic bath. Test bladders (n = 5) exposed to K+-free hypertonic bath at t = 40 min had a mean fluid loss of 17.9 ± 4.0 μl/min over 30 min, which was not significantly different (P > 0.05, unpaired t-test) from the hydroosmotic response (19.0 ± 3.0 μl/min) in the control group (n = 5) exposed to standard hypertonic bath over the same time interval (data only).
Determination of Cl− Requirement by the ADH Hydroosmotic Response
The hormone-invoked hydroosmotic response was not affected by the absence of bath Cl−. In isoosmotic conditions (Fig. 7A), the magnitude of the response to a maximal dose of AVP (20 mU/ml) in the test bladders was similar to that seen in the control group. Over the 30-min exposure to AVP, mean fluid loss was 15.4 ± 1.9 and 16.3 ± 1.3 μl/min, respectively (P > 0.05, unpaired t-test).
In Cl−-free hyperosmotic bath (Fig. 7B), the hydroosmotic response to AVP (20 mU/ml) was substantial. However, the augmentation of the response as seen in standard hypertonic bath was absent. Mean fluid loss was 25.3 ± 2.2 and 40.1 ± 2.5 μl/min, respectively (P < 0.005, unpaired t-test).
Examination of a Possible Role for a Cl− Transporter or Channel in the SH Response
Bumetanide (100 μM), which inhibits the Na/K/2Cl cotransporter, did not affect the hydroosmotic response to SH (Fig. 8). Mean fluid loss over the 30 min in hypertonic bath with bumetanide was 20.1 ± 3.9 μl/min compared with 23.9 ± 5.8 μl/min in control bladders exposed to SH alone (P > 0.05, unpaired t-test). By comparison, NPPB (50 μM), a Cl−-channel blocker, significantly reduced the hydroosmotic response compared with the control group (Fig. 8); mean fluid loss was 3.6 ± 0.9 μl/min (P < 0.01, unpaired t-test).
Effect of Inhibition of NHE on the SH Response
After a 30-min control period (t = 70–100 min) in isotonic bath, bladders were exposed for a second time to either SH (Fig. 9A) or SH + AVP (Fig. 9B), but now with EIPA (100 μM) in the bathing medium; these bladder groups therefore acted as their own controls. EIPA, which blocks the NHE, had an inhibitory effect on the response to SH alone (Fig. 9A); mean fluid loss was 3.2 ± 1.7 μl/min compared with a loss of 8.0 ± 2.1 μl/min seen upon a first exposure to SH, over a similar time interval (P < 0.005, paired t-test). The response to AVP (20 mU/ml) was no longer augmented in hypertonic bath when EIPA was present in the bathing medium (Fig. 9B). Mean fluid loss was 12.8 ± 1.0 μl/min compared with a mean loss of 29.7 ± 2.9 μl/min observed during the first exposure to SH (P < 0.005, paired t-test).
In agreement with earlier reports (4, 48), we show that anuran urinary bladder undergoes an increase in membrane water permeability when exposed to either ADH (AVP) or hypertonic bath (Fig. 1A). This response was reversed when bladders were placed in fresh isotonic bath and reoccurred with an equal magnitude when bladders were reexposed 30 min later to the hydroosmotic agent, be it AVP or SH (Fig. 1B).
There is evidence for permanent water channels in the basolateral membrane of frog bladder epithelial cells; these channels are blocked by the presence of mercurials (HgCl2 and P-chloromercuribenzene sulfonate) in the serosal bathing solution (58). Our functional studies with HgCl2 (Fig. 2) would support a transcellular route for water movement across the epithelium. Luminal fluid from the test group was still hypotonic at the end of the experiments, indicating that the junctional integrity had not been disrupted by the mercurial.
In contrast to a known inhibitory effect of prostaglandins on the ADH and SH hydroosmotic response in anuran bladder (40, 48), frog bladders exposed to low concentrations of PGE2 in isotonic medium are reported to undergo an increase in osmotic water permeability, without ADH (42). It was necessary to investigate whether a low endogenous prostaglandin synthesis, resulting from exposure of bladders to hypertonic bath (12), might be involved in invoking the hydroosmotic response.
Preincubation of toad urinary bladder in isotonic bath with indomethacin (10 μM) for 120 min is reported to inhibit PGE2 synthesis in toad urinary bladder (19). In test bladders (Fig. 3), the SH hydroosmotic response was not abolished by the depletion of bath prostaglandin but, rather, the magnitude of fluid loss was larger than that seen in experiments carried out under the standard SH protocol. Indomethacin significantly increased basal fluid loss (t = 90–120 min) in test bladders compared with control values (Fig. 3), in agreement with the finding of other researchers (1). Indomethacin is reported to inhibit phosphodiesterase activity (17); however, such an action is contraindicated by the fact that osmotic water permeability was decreased rapidly when the test group was placed in fresh isotonic bath at t = 150 min in the continued presence of indomethacin.
In anuran urinary bladder, NE is reported to inhibit the onset and maintenance of the ADH hydroosmotic response and, in addition, the onset of the hydroosmotic response to SH (48). We had similar findings (data not shown).
We show for the first time that the SH hydroosmotic response is reversed by NE, be it in SH-NaCl bath (Fig. 4A) or SH-mannitol bath (Fig. 4B). This reversal is inhibited by yohimbine (Fig. 5A) but not by prazosin (Fig. 5B) or propranolol (Fig. 5C), indicating that NE is acting via α2-adrenergic receptors. The α2-adrenergic receptor is a G protein-linked receptor known to inhibit AC (45). This reversal is also seen in tissue preexposed to indomethacin for 120 min (Fig. 5D), contraindicating an indirect role for prostaglandins (49) in the reversal by NE of the SH hydroosmotic response.
Using a volumetric method (7), Ripoche and coworkers (48) failed to demonstrate a reversal by NE of the hydroosmotic response in SH-mannitol (450 mosmol/kgH2O), whereas their results for NE addition postexposure of bladders to SH-NaCl (450 mosmol/kgH2O) were inconclusive. The authors concluded that, once initiated, the SH hydroosmotic response differed from that invoked by ADH; the latter depended on cAMP for its maintenance, whereas the former did not.
The failure of NE to reverse the hydroosmotic response in SH-mannitol, as noted by this group of researchers, may have been because of the high concentration (220 mM) of mannitol, in combination with the final tonicity (450 mosmol/kgH2O) of the bathing medium, used in the study (48). Under these conditions, mannitol may have interfered with the action of NE either directly at the level of receptor binding by the agonist or indirectly via intracellular events (31, 60).
In our hands, using the gravimetric method and a serosal hypertonic bathing solution (330 mosmol/kgH2O) within a physiological range, known to be tolerated in vivo (54), both types of SH hydroosmotic response were reversed by NE; this reversal is shown to be mediated by α2-adrenergic receptors and not to involve prostaglandins. Our findings indicate that the hydroosmotic response of anuran bladder to SH is similar in both onset and maintenance to that invoked by ADH; both stages depend on AC activity.
In the dehydrated anuran, the ability to selectively block hydroosmotic responses of the urinary bladder via neural pathways would be advantageous. When an external water source is available, rehydration would be by dermal “drinking,” and the reserve water in the bladder would not be compromised. There is evidence of dermal drinking with water present in the bladder (27). It has been shown that both NE and the specific β-agonist, isoprenaline, stimulate water flow across toad pelvic patch skin but fail to do so in the bladders of the same animal (13).
When Cl− is absent from the bathing medium (Fig. 6A), the SH response is abolished; by contrast, the ADH hydroosmotic response is unaffected under these conditions (Fig. 7, A and B). The hydroosmotic response to a maximal dose of AVP (20 mU/ml) is augmented in hypertonic bath (Fig. 7B). The finding that this augmentation does not occur in Cl−-free hyperosmotic bath (Fig. 7B) is consistent with a specific additive role of SH, resulting from an independent stimulation of water flow. It is intriguing to consider the possibility that two different isoforms of AC are involved in mediating the hormone-dependent (ADH) and the hormone-independent (SH) hydroosmotic responses in anuran urinary bladder. The existence of more than one isoform of AC in this tissue has already been suggested (53) based on the observation that the ADH-invoked natriferic and hydroosmotic responses can be differentially inhibited (36, 44). The two predominant cell types in anuran bladder are the mitochondria-rich (MR) cell and the granular cell (43). A role for MR cells (via intercellular movement of cAMP) in the response of granular cells to ADH has been proposed (20, 52). It is also possible that the AC activated in the SH hydroosmotic response is in a separate cytoplasmic compartment and that the response involves a separately regulated pool of water channels.
The causative agent or agents in the SH hydroosmotic response remain unknown. Our findings indicate an important role for Cl− in the SH hydroosmotic response. This may be a factor of the actual concentration of the anion at the serosal face of the tissue, since bladders exhibited a diminished hydroosmotic response when bath Cl− was reduced by 50% (Fig. 6B), whereas the response was abolished in Cl−-free hyperosmotic medium (Fig. 6B). Intracellular Cl− concentration may also be an important factor, since the SH response is restored within 10 min of exposure to standard hypertonic bath (t = 70 min; Fig. 6C). This suggests a need to increase intracellular Cl− concentration from its depleted levels,1 resulting in a normal distribution of Cl− across the membrane.
Examination of the effects of NPPB and bumetanide (Fig. 8) represents a preliminary investigation of the mechanisms of the Cl− dependence of the hydroosmotic response. The response is sensitive to serosal NPPB but not to serosal bumetanide, suggesting a role for a Cl− channel rather than a transporter in the response.
EIPA (100 μM), a specific inhibitor of NHE, caused a significant decrease in the hydroosmotic response (Fig. 9A) and significantly reduced the augmentation of the AVP response in hypertonic bath (Fig. 9B). These results are very similar to those seen in the absence of bath Cl− (Figs. 6A and 7B), raising the possibility that there may be a functional link between these two parameters, namely Cl− concentration and Na+/H+ activity, in the hydroosmotic response in this tissue.
The possibility that bath absence of Cl− and also inhibition of NHE might cause intracellular pH (pHi) to fall is contraindicated by the fact that, under both of these experimental conditions, the AVP response, which is known to be inhibited by intracellular acidification (8), is present (Figs. 7, A and B, and 9B). NHE activation in shrinkage situations has been attributed to a shift in the flux vs. pHi profile of the exchanger in an alkaline direction (5), resulting in an increase in pHi. In R. temporaria urinary bladder, an intracellular alkalization in hypertonic bath, resulting from activity of Na+/H+, might have a permissive effect on the activation of AC by another factor, such as the Cl− anion. Activation of AC by halide ions (46), including Cl− (30, 34, 56) has been reported.
In the mammalian collecting duct, ADH activation of V2 receptors is in the presence of a hypertonic extracellular fluid. Reports of a hormone-independent effect of hypertonicity in the mammalian system are scant. However, those that do exist indicate a potential independent input by hypertonicity on water and urea permeability (32, 50).
The results presented here add to the increasing evidence that the hormone-independent hydroosmotic response of anuran bladder in hypertonic bath involves specific biochemical steps and is not simply a concentration artifact resulting from cell shrinkage.
This work was supported by a research grant from the Health Research Board of Ireland.
Present address of A. T. Hanna-Mitchell: Renal-Electrolyte Div., Dept. of Medicine, Univ. of Pittsburgh, A1209 Scaife Hall, 3550 Terrace St., Pittsburgh, PA 15261 (e-mail: [email protected]).
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 © 2006 the American Physiological Society