Endothelin and nitric oxide mediate adaptation of the cortical collecting duct to metabolic acidosis

Shuichi Tsuruoka, Seiji Watanabe, Jeffrey M. Purkerson, Akio Fujimura, George J. Schwartz

Abstract

Endothelin (ET) and nitric oxide (NO) modulate ion transport in the kidney. In this study, we defined the function of ET receptor subtypes and the NO guanylate cyclase signaling pathway in mediating the adaptation of the rabbit cortical collecting duct (CCD) to metabolic acidosis. CCDs were perfused in vitro and incubated for 3 h at pH 6.8, and bicarbonate transport or cell pH was measured before and after acid incubation. Luminal chloride was reversibly removed to isolate H+ and HCO3 secretory fluxes and to raise the pH of β-intercalated cells. Acid incubation caused reversal of polarity of net HCO3 transport from secretion to absorption, comprised of a 40% increase in H+ secretion and a 75% decrease in HCO3 secretion. The ETB receptor antagonist BQ-788, as well as the NO synthase inhibitor, NG-nitro-l-arginine methyl ester (l-NAME), attenuated the adaptive decrease in HCO3 secretion by 40%, but only BQ-788 inhibited the adaptive increase in H+ secretion. There was no effect of inactive d-NAME or the ETA receptor antagonist BQ-123. Both BQ-788 and l-NAME inhibited the acid-induced inactivation (endocytosis) of the apical Cl/HCO3 exchanger. The guanylate cyclase inhibitor LY-83583 and cGMP-dependent protein kinase inhibitor KT-5823 affected HCO3 transport similarly to l-NAME. These data indicate that signaling via the ETB receptor regulates the adaptation of the CCD to metabolic acidosis and that the NO guanylate cyclase component of ETB receptor signaling mediates downregulation of Cl/HCO3 exchange and HCO3 secretion.

  • tubule microperfusion
  • endothelin receptor antagonist
  • nitric oxide synthase inhibitor
  • cyclic GMP
  • cyclic GMP-dependent protein kinase

endothelins (ET) were first described as vasoactive peptides that regulate regional vascular tone by signaling cells in the cardiovascular system via members of the G protein-coupled receptor family (15a, 45). Since the early 1990s, it has become apparent that ET also regulates sodium-water balance and pH homeostasis in the kidney by directly signaling renal epithelial cells. ET promotes water diuresis by acting to inhibit the hydroosmotic action of vasopressin (22). Several studies implicate ET in the regulation of acid-base homeostasis in the proximal tubule (41) as well as the distal nephron (42, 43). ET modulates NHE3 activity in the proximal tubule and thereby regulation of proton secretion by this segment (13). In the distal nephron, ET-1 is abundantly expressed by collecting ducts (12), as are ETB receptors (29). ET stimulates distal tubular acidification in the rat (42, 43), and the expression of ET in the kidney is stimulated by metabolic acidosis (44). Mice having a genetically disrupted ETB receptor experience a more severe acidosis than normal mice in response to acid loading (13). However, a detailed assessment of the regulation of proton vs. bicarbonate transport in the cortical collecting duct (CCD) is lacking.

In many cases, the effects of ET on ion transport processes in nephron segments are mediated by the generation of nitric oxide (NO) resulting from activation of a NO synthase (NOS) (9, 10, 17). In vivo studies suggest that NO induces a natriuresis and diuresis, which is mediated by activation of guanylate cyclase. In the proximal tubule, Wang et al. (41) reported that tubules from neuronal NOS knockout mice exhibited lower fluid and HCO3 absorption rates compared with tubules from wild-type mice. Thus NO produced by neuronal NOS in the proximal tubule is likely to stimulate fluid and HCO3 absorption. In the CCD NO inhibits Na+ absorption and vasopressin-stimulated osmotic water permeability (5, 17). The effect of NO on osmotic water permeability is blocked by guanylate cyclase and cGMP-dependent protein kinase (PKG) inhibitors, indicating that the effect of NO is mediated by activation of soluble guanylate cyclase and subsequent activation of PKG by cGMP (6, 17). This cascade results in inhibition of vasopressin-stimulated osmotic water permeability (6, 17).

In the thick ascending limb, endogenous NO inhibits chloride transport (20), and a comparable inhibition is observed using exogenous ET-1 (19). The ET effect is blocked by inhibiting NOS with NG-nitro-l-arginine methyl ester (l-NAME) (19), suggesting that the effects of ET-1 are mediated by NOS in the thick ascending limb of Henle's loop (9).

NO also regulates pH homeostasis in the distal nephron. In freshly isolated CCDs, NO donors decrease bafilomycin-sensitive H+-ATPase activity (31). Such inhibition is likely mediated by cGMP because cGMP analogs also inhibit H+-ATPase activity (31). However, this finding would lead to an expectation of decreased H+ secretion in the CCD, which could be life threatening in a setting of metabolic acidosis. In contrast, mice deficient in neuronal (n)NOS develop metabolic acidosis (41). In rats inhibition of NO by administration of l-NAME impairs urinary acid excretion after acute NH4Cl loading (35). CCDs taken from such treated rats showed that net bicarbonate absorption was reduced by 40%. These studies strongly suggest that NO is involved in the maintenance of acid-base homeostasis in the distal nephron; however, it has not been established whether NO regulates proton and/or bicarbonate flux by intercalated cells.

In this study, we examine whether the changes in H+/HCO3 secretion fluxes induced by acidosis are regulated by ET receptor signaling via the NO-guanylate cyclase pathway. We made use of in vitro acid incubation of CCDs that recapitulates the findings of 3 days of in vivo acid loading (21). CCDs taken from normal rabbits secrete net HCO3, and this net flux is made up of a small H+ secretory flux that is outstripped by a much larger HCO3 secretory flux (26, 33). The adaptation to acidosis is associated with both a modest increase in H+ secretory flux and a large decrease in HCO3 secretory flux, with the resultant sum of fluxes being the secretion of net protons (26, 33); that is, a reversal in polarity of net HCO3 transport.

METHODS

Animals.

Female New Zealand white rabbits (n = 49) weighing 1.6–2.5 (mean 2.04) kg were maintained on standard laboratory chow (Japan Clea) with free access to water (33). Animals were killed by intracardiac injection of 130 mg pentobarbital sodium after premedication with ketamine (44 mg/kg) and xylazine (5 mg/kg). Urine was obtained postmortem by bladder tap; urine pH averaged 7.96 ± 0.02 (SE, n = 49).

Microperfusion of CCDs.

CCDs were microdissected and microperfused as performed in this laboratory (26, 33). The average tubule length was 0.9 ± 0.1 mm. Equilibration and transport were performed using Burg's solution in the perfusate and bath, containing (in mM) 120 NaCl, 25 NaHCO3, 2.5 K2HPO4, 2 CaCl2, 1.2 MgSO4, 5.5 d-glucose, 1 trisodium citrate, 4 sodium lactate, and 6 l-alanine, 290 ± 2 mosmol/kgH2O, and gassed with 94% O2-6% CO2, yielding a pH of 7.4 at 37°C (26, 33, 34). Bath was continually exchanged at 14 ml/h by a peristaltic pump. Luminal perfusion rate was maintained at 1.5–2.1 nl/min.

Incubation for 3 h at low pH (pH 6.8 in both luminal and bathing solutions) has been previously described (21). Briefly, the luminal solution contained DMEM without NaHCO3 (GIBCO, BRL, Gaithersburg, MD), Burg's solution, and dissection solution (Burg's solution with 25 mM NaHCO3 replaced by NaCl) in a ratio of 3:2:4, respectively. The bathing solution was similar except that it also contained 30 U/ml penicillin, 30 μg/ml streptomycin, and 3.3% fetal calf serum (GIBCO, BRL) (21, 33, 46). Incubation at pH 6.8 in vitro yields a physiology comparable to 3 days of acidosis in vivo and reverses the polarity of HCO3 flux from net secretion to net absorption (21, 26, 33).

Bicarbonate transport.

Triplicate collections of 12–15 nl of tubular fluid were made under water saturated mineral oil and analyzed for HCO3 using a Nanoflo (WPI, Sarasota, FL) (26, 33, 35, 36). Net HCO3 was calculated as JHCO3 = (CO − CL) × (VL/L), where CO and CL are the HCO3 concentrations of perfused and collected fluid, respectively, VL is the rate of collected fluid, and L is the length of the tubule (mm) (26, 33). When HCO3 transport (JHCO3) is >0, there is net HCO3 absorption; when JHCO3 is <0, there is net HCO3 secretion. To distinguish between unidirectional H+ and HCO3 secretion after net HCO3 transport is measured, luminal Cl was replaced by gluconate (21). In this maneuver, HCO3 is not secreted, thereby uncovering the unidirectional H+ secretory flux. Subtracting this H+ secretory flux from the net bicarbonate transport flux reveals the HCO3 secretory flux.

Measurements were repeated after the 3-h incubation and compared with preincubation values. In most of the experiments, an agent was introduced in the bath for 30 min at pH 7.4 before being added to the pH 6.8 bathing solution for the 3-h incubation (21, 26). The agents included BQ-788 (1 μM, Sigma, ETB receptor antagonist) (8), BQ-123 (1–10 μM, Sigma, ETA receptor antagonist) (2), l-NAME (1 mM, Sigma, St. Louis, MO and Tokyo, Japan, NOS inhibitor) (39), d-NAME (1 mM, Sigma, inactive enantiomer control), LY-83583 (or 6-anilino-5,8-quinolinedione, 10 μM, Biomol, Plymouth Meeting, PA, guanylate cyclase inhibitor) (39), and KT-5823 (2 μM, Sigma, specific cell-permeant cGMP-dependent protein kinase inhibitor) (16).

Transepithelial voltage (mV) was measured using the luminal perfusion pipette as an electrode. The voltage difference between calomel cells connected via 3 M KCl agar bridges to perfusate and bath was measured with a high-impedance electrometer.

Cell pH studies.

Cell pH was measured by excitation ratio fluorometry (490 nm/445-nm excitation; 520-nm emission) using 5–10 μM BCECF (Molecular Probes, Eugene, OR) (26, 36). Fluorescence was detected in multiple intercalated cells and corrected for background (Photon Technology, London, Ontario). Movement and contaminating fluorescent signals were minimized by examining cells in focus close to the perfusion pipette and in the wall of the tubule. Duplicate readings were averaged in Burg's solution, after the reversible removal of luminal Cl and subsequently after the reversible removal of basolateral Cl. The sequence of readings was repeated in the same identified intercalated cells after 3-h incubation.

Agents were dissolved in 0.1% DMSO (vehicle) and added to the bathing solution 3–15 min before and during the 3-h incubation at pH 6.8. These agents included BQ-788 (1 μM, the ETB receptor antagonist) and l-NAME (1 mM).

Statistics.

Data are presented as means ± SE. Standard paired and unpaired comparisons were performed on spreadsheets using Excel 2003 (Microsoft, Bellvue, WA). One-way ANOVA, box plots, and post hoc Duncan and Scheffé's multiple comparison tests were used to examine the acid-incubated/basal H+/HCO3 flux ratios and the acid-induced changes in net HCO3 flux using NCSS 6.0 statistical software (Kaysville, UT). Significance was asserted if P < 0.05 for each multiple comparison test.

RESULTS

Incubation at pH 6.8 induces a reversal in polarity of net HCO3 flux.

To study the adaptation of intercalated cells, we utilized the model of in vitro acid incubation (21) and examined the effect of 3-h incubation of each CCD at pH 6.8 in both luminal and bathing solutions. In three CCDs taken from normal rabbits, the mean rate of bicarbonate transport before acid incubation was −3.18 ± 0.51 pmol·min−1·mm−1, indicating net HCO3 secretion. After 3-h incubation at pH 6.8, the solutions were restored to pH 7.4 and HCO3 transport was remeasured in the same tubule. After acid incubation the net flux was +3.07 ± 0.28 pmol·min−1·mm−1 (P < 0.05), indicating net H+ secretion, as has been shown previously (21). The transepithelial voltage became less negative after acid incubation, in keeping with a higher rate of net H+ secretion (−2.9 ± 0.1 to −2.4 ± 0.1 mV, P < 0.05).

To dissect out the unidirectional fluxes of HCO3 and H+ comprising the net flux, we incubated five CCDs at pH 6.8 and added a period in which luminal Cl was removed (and replaced by gluconate) to the pre- and postacid-incubation measurements. Bicarbonate transport in the absence of luminal Cl reflects the secretion of protons by α-intercalated cells, because HCO3 secretion is simultaneously inhibited in the β-intercalated cells. Figure 1 shows that the mean rate of net HCO3 transport was −3.81 ± 0.21 pmol·min−1·mm−1, which was comprised of an H+ secretion rate of 3.69 ± 0.16 pmol·min−1·mm−1 and a HCO3 secretion rate of −7.50 ± 0.36 pmol·min−1·mm−1 (Table 1). After 3-h incubation at pH 6.8 in the presence of vehicle, the net HCO3 flux reversed polarity to net H+ secretion (+3.27 ± 0.14 pmol·min−1·mm−1, P < 0.01); at 7.1 pmol·min−1·mm−1 (Fig. 2), the change in net flux from basal to acid incubated was large enough to result in a change in polarity from net HCO3 secretion to absorption. This adaptation was associated with a 39% increase in H+ secretory flux to 5.08 ± 0.23 pmol·min−1·mm−1 (P < 0.05; Table 1 and Fig. 3) and a 76% decrease in HCO3 secretory flux to −1.81 ± 0.11 pmol·min−1·mm−1 (P < 0.01); the acid-incubated flux was 24% of the basal flux (Table 1 and Fig. 4). Concomitant with the increase in H+ secretion, the transepithelial voltage tended to become less negative after acid incubation (−2.5 ± 0.1 to −2.1 ± 0.2 mV, P = 0.06).

Fig. 1.

Changes in H+/HCO3 flux induced by incubation of cortical collecting duct (CCD) at pH 6.8 (n = 5 CCDs). Left: net bicarbonate transport. Middle: H+ secretory fluxes obtained from luminal Cl removal (which inhibits HCO3 secretion). Right: HCO3 secretory fluxes obtained from the difference between the net and H+ secretion flux. Each flux is presented before and following acid incubation. *Significantly different (P < 0.01) from basal value.

Fig. 2.

Inhibition of ETB receptor signaling blocks adaptive changes in net bicarbonate flux induced by acid incubation. The difference between acid-incubated and basal net bicarbonate flux in pmol·min−1·mm−1 under various treatments is shown in box plots. Treatments include vehicle (n = 5), BQ-788 (n = 5), BQ-123 (n = 5), l-NAME (n = 4), d-NAME (n = 4), LY-83583 (n = 6), and KT-5823 (n = 6). The horizontal line represents the median, the top and bottom of the box represent the 75th and 25th percentiles, respectively, and the length of the box is the interquartile range; the T-shaped lines represent upper and lower adjacent values, wherein the former is the largest observation that is less than or equal to the 75th percentile plus 1.5 times the interquartile range and the latter is the smallest observation that is greater than or equal to the 25th percentile minus 1.5 times the interquartile range. The change in net bicarbonate flux is positive, indicating a shift from net secretion to zero net flux or to net HCO3 absorption. Because 1-way ANOVA was highly significant (P < 0.001), Duncan and Scheffé's multiple-comparison tests were performed. The change in vehicle net flux was significantly different from those due to BQ-788, l-NAME, LY-83583, and KT-5823, and the changes due to the inactive agents d-NAME and BQ-123 were also significantly different from these 4 treatments. The changes due to BQ-788, l-NAME, LY-83583, and KT-5823 were each significantly different from those due to vehicle, BQ-123, or d-NAME.

Fig. 3.

Antagonism of ETB receptor blocks increases in H+ secretion induced by low pH. The ratio of acid-incubated/basal H+ secretory fluxes under the various treatments is shown. The median and 75th and 25th percentiles are presented as in Fig. 2. The dashed horizontal line denotes the ratio of one, indicating no significant increase in H+ secretion. ANOVA was highly significant (P < 0.001) and the multiple range comparisons showed that the vehicle was significantly different from BQ-788, LY-83583, and KT-5823. The H+ secretory flux after BQ-788 treatment was significantly different from vehicle, l-NAME, d-NAME, and BQ-123; the BQ-788 ratio includes one within the interquartile range, indicating no significant increase in H+ secretion with acid incubation. LY-83583 and KT-5823 treatments were significantly different from vehicle, d-NAME, and BQ-123.

Fig. 4.

Inhibition of ETB receptor signaling blocks adaptive changes in HCO3 secretion induced by acid incubation. The ratio of acid-incubated/basal HCO3 secretory fluxes under the various treatments is shown. The median and 75th and 25th percentiles are presented as in Fig. 2. ANOVA was highly significant (P < 0.001) and the multiple range comparisons showed that vehicle was significantly different from BQ-788, l-NAME, LY-83583, and KT-5823 but was not different from d-NAME or BQ-123. d-NAME and BQ-123 were significantly different from BQ-788, l-NAME, LY-83583, and KT-5823. Each of the active treatments (BQ-788, l-NAME, LY-83583, and KT-5823) was significantly different from vehicle, BQ-123, and d-NAME.

View this table:
Table 1.

Basal and acid incubation net and H+/HCO3 fluxes

Adaptive changes in H+/HCO3− secretion fluxes induced by low pH are inhibited by ETB receptor antagonism.

Because ET secretion is induced by acidosis (42, 44) and ET receptor signaling regulates ion transport processes along the nephron (13, 32, 43), we examined the effect of ET receptor antagonism on the changes in H+/HCO3 secretory fluxes induced by incubation at low pH. Acid incubation with BQ-788 failed to reverse the polarity of the net HCO3 flux (Table 1); that is, net HCO3 secretion was reduced to no significant HCO3 transport (−4.03 ± 0.29 to 0.047 ± 0.16 pmol·min−1·mm−1, P < 0.01); at 4.1 pmol·min−1·mm−1, the change in net flux was significantly smaller than vehicle (Fig. 2). The reduction in net HCO3 secretion was not due to an offsetting increase in H+ secretion [3.95 ± 0.10 to 4.05 ± 0.20, P = not significant (NS); Table 1 and Fig. 3, significantly smaller than vehicle], but rather to a 49% reduction in HCO3 secretion (−7.98 ± 0.35 to −4.00 ± 0.15 pmol·min−1·mm−1, P < 0.01); the acid-incubated HCO3 secretory flux was 51% of the basal flux (Table 1 and Fig. 4) and significantly larger than vehicle. The lack of change in H+ secretion was associated with no significant change in transepithelial voltage (−2.13 ± 0.15 to −2.04 ± 0.14 mV, P = NS). BQ-123, an ETA receptor antagonist used at 1–10 μM, did not prevent the changes in H+ secretion and HCO3 secretion resulting from the incubation at pH 6.8. Acid incubation with BQ-123 resulted in a reversal of polarity of net HCO3 flux that was comparable to CCDs treated with vehicle control (Table 1 and Figs. 24). Because the response to acid incubation in the presence of BQ-788 resulted in a larger HCO3 secretory flux (P < 0.05) and a smaller H+ secretory flux (P < 0.05) than in the presence of vehicle or BQ-123 incubation (Table 1 and Figs. 3 and 4), we concluded that ETB, but not ETA, receptor signaling was critical for the adaptation of the CCD to acidosis.

Time control.

To ensure the stability of the isolated CCD preparation, we performed four timed control experiments with a 3-h incubation at pH 7.4; three of these experiments utilized 1 μM BQ-788 and the fourth used no inhibitor. Net bicarbonate secretion was unchanged by the incubation (or inhibitor): −3.89 ± 0.16 before and −3.83 ± 0.17 pmol·min−1·mm−1 after incubation (P = NS). There was no change in H+ secretory flux (3.44 ± 0.05 to 3.68 ± 0.21 pmol·min−1·mm−1, P = NS) or in HCO3 secretory flux (−7.33 ± 0.13 to −7.51 ± 0.29 pmol·min−1·mm−1, P = NS). In agreement with the stability of the H+ secretory flux, there was no change in transepithelial voltage (−2.1 ± 0.1 to −2.1 ± 0.1 mV, P = NS). Similar data have been reported previously (33).

Adaptive decrease in HCO3− secretion flux induced by low pH requires NO synthesis.

ET receptor signaling induces activation of NOS activity (9, 10). Therefore, we examined whether NO production was required for adaptation of the CCD to low pH. To detect the effect of NOS inhibition, we incubated five CCDs in 1 μM l-NAME just before and during the pH 6.8 incubation. Net HCO3 secretion before the acid incubation averaged −3.22 ± 0.22 pmol·min−1·mm−1 and after acid incubation was only slightly greater than zero, 0.59 ± 0.02 pmol·min−1·mm−1. Whereas the change in net HCO3 flux was significant (P < 0.01), the acid-induced reversal of polarity of HCO3 flux was inhibited by l-NAME. When the inactive d-NAME was employed in two CCDs, the full adaptation and reversal of polarity of the net HCO3 flux were apparent (preincubation: −2.59 and −3.74 pmol·min−1·mm−1; postacid incubation: 2.94 and 2.79 pmol·min−1·mm−1, respectively).

We further examined the effect of NOS inhibition on the adaptation to low pH (in vitro acidosis) by measuring HCO3 transport in the presence and absence of luminal Cl. Table 1 shows that acid incubation with l-NAME resulted in a significant reduction, but not a reversal in polarity, of net HCO3 transport (−3.66 ± 0.37 to 0.16 ± 0.19 pmol·min−1·mm−1, P < 0.01); at 3.8 pmol·min−1·mm−1, the change in net flux was significantly smaller than vehicle (Fig. 2). This attenuated adaptation was associated with a 24% increase in H+ secretory flux (3.17 ± 0.06 to 3.93 ± 0.08 pmol·min−1·mm−1, P < 0.01; Fig. 3) and a 44% decrease in HCO3 secretory flux (−6.84 ± 0.38 to −3.77 ± 0.12 pmol·min−1·mm−1, P < 0.01; Fig. 4); at 56%, the ratio of acid-incubated basal HCO3 secretory flux was significantly increased over vehicle, indicating a failure to adaptively reduce HCO3 secretion during acid incubation. The small increase in H+ secretion was not associated with a significant decrease in transepithelial voltage (−2.3 ± 0.1 to −2.1 ± 0.2 mV, P = NS). In contrast, for CCDs incubated at pH 6.8 with the inactive enantiomer d-NAME, there was a complete adaptive reversal of polarity of net HCO3 flux (Table 1 and Figs. 24), similar to what was observed with vehicle control. These data show that NOS inhibition via l-NAME diminishes the downregulation of HCO3 secretion that occurs in response to incubation at low pH. l-NAME also partially blocks the acid-induced increase in H+ secretion (24 vs. 37% for d-NAME; Table 1) but was not as effective as the ETB receptor antagonist, BQ-788 (Table 1 and Fig. 2), so that the end result is not a reversal in polarity of HCO3 transport, but a reduction to zero net transport.

ETB receptor antagonism and l-NAME block the effect of acidosis on chloride-dependent cell pH changes.

To confirm the changes in HCO3 transport, we examined β-intercalated cell pH changes in response to removal of Cl from the lumen or bath. Previously, we showed that β-intercalated cells express apical Cl/HCO3 exchangers and basolateral Cl conductances (21, 25). Figure 5A shows that when luminal Cl was removed, intercalated cell pH rose by 0.40 ± 0.04 pH U, but that after acid incubation, there was no alkalinization of cell pH with this maneuver (delta pH = −0.01 ± 0.03, P < 0.01). When bath Cl was removed before the incubation, intercalated cell pH fell by −0.42 ± 0.01 pH U (Fig. 5B), but after acid incubation, the change in pH was reduced (delta pH = −0.18 ± 0.05, P < 0.05). Such an adaptation to reduce apical Cl/HCO3 exchange and basolateral Cl exit would be appropriate during metabolic acidosis.

Fig. 5.

ETB receptor signaling is required for loss of luminal Cl/HCO3 exchange activity stimulated by low pH. Cell pH changes in response to Cl removal in basal state and after acid incubation are shown. A, C, and E: intercalated cell pH changes resulting from luminal Cl removal. B, D, and F: cell pH changes resulting from bath Cl removal. A and B: vehicle (n = 25 cells from 3 CCDs). C and D: BQ-788 (n = 40 cells from 5 CCDs). E and F: l-NAME (n = 66 cells from 7 CCDs). *Significantly different from basal value (P < 0.01).

BQ-788 added to the pH 6.8 incubation prevented the adaptive reduction in apical Cl/HCO3 exchange (0.46 ± 0.02 to 0.46 ± 0.01 pH U, P = NS; Fig. 5C) and in sensitivity to basolateral Cl removal (−0.44 ± 0.03 to −0.44 ± 0.04, P = NS; Fig. 5D). Similarly, l-NAME added to the pH 6.8 incubation clearly prevented the reduction in apical Cl/HCO3 exchange (0.47 ± 0.03 to 0.45 ± 0.03 pH U, P = NS; Fig. 5E) and in sensitivity to basolateral Cl removal (−0.48 ± 0.04 to −0.51 ± 0.05 pH U, P = NS; Fig. 5F). These results confirm that ETB receptor signaling and NO synthesis are required for the adaptive decrease in HCO3 secretion (presumably by endocytosis of apical Cl/HCO3 exchangers) induced by incubation at low pH.

cGMP generation and subsequent activation of PKG are required for acid-induced changes in H+/HCO3 fluxes.

Many of the effects of the signaling molecule NO are mediated by stimulation of NO-sensitive guanylate cyclase resulting in cGMP formation and the subsequent activation of cGMP-dependent protein kinase (PKG) (7, 15). We tested the effect on HCO3 transport of 10 μM LY-83583, an inhibitor of soluble guanylate cyclase, during acid incubation (Fig. 2). LY-8353 markedly attenuated the reversal of polarity of net HCO3 flux in response to low pH, and there was no net HCO3 flux (−3.76 ± 0.17 to 0.18 ± 0.09 pmol·min−1·mm−1, P < 0.01; Table 1); at 3.9 pmol·min−1·mm−1, the difference between acid-incubated and basal net flux was much smaller than vehicle or inactive agent (d-NAME or BQ-123, Table 1 and Fig. 2). There was a only a 16% increase in H+ secretion, smaller than seen with vehicle or inactive agent (3.25 ± 0.11 to 3.79 ± 0.17 pmol·min−1·mm−1, P < 0.05; Table 1 and Fig. 3), and a 49% reduction in HCO3 secretion (−7.01 ± 0.27 to −3.60 ± 0.23 pmol·min−1·mm−1, P < 0.01), such that at 51% the ratio of acid-incubated/basal HCO3 secretory flux was significantly increased compared with vehicle or inactive agent (Fig. 4). The increase in acid/basal H+ secretion and the decrease in acid/basal HCO3 secretion were smaller than those observed after acid incubation with vehicle (P < 0.05; Table 1).

The role of PKG in the acid-induced adaptation was tested by examining the effect of 2 μM KT-5823, a specific cell-permeant inhibitor of PKG, on H+/HCO3 fluxes. KT-5823 substantially prevented the reversal of polarity of HCO3 flux in response to low pH, so that there was no net flux after acid incubation (−4.22 ± 0.15 to 0.08 ± 0.05 pmol·min−1·mm−1, P < 0.01; Table 1); at 4.3 pmol·min−1·mm−1, the change in net flux from basal to acid incubated was significantly smaller than vehicle or inactive controls (Fig. 2). The increase in acid/basal H+ secretion and the decrease in acid/basal HCO3 secretion were smaller than those observed after acid incubation with vehicle or inactive controls (P < 0.05; Table 1 and Figs. 3 and 4). These results suggest that inhibitors of guanylate cyclase and PKG partially prevent the adaptive increase in H+ secretion and the decrease in HCO3 secretion that would be anticipated in response to metabolic acidosis.

Taken together, these data indicate that the adaptive decrease in HCO3 secretion in response to acidosis is mediated in large part by ET-ETB-NO-guanylate cyclase-PKG signaling pathway. In contrast, the adaptive increase in H+ secretion, which may be directly stimulated by ET-ETB signaling, tends to be less affected by the NO-signaling pathway.

DISCUSSION

To investigate the role of ET and NO in intercalated cells of the CCD, we made use of our in vitro acidosis model in which incubation of a CCD for 3 h at pH 6.8 induces a change in polarity of net HCO3 flux from net secretion to net absorption (21, 26). This model recapitulates the pathophysiological condition of 3 days of NH4Cl loading (in the drinking water) that results in metabolic acidosis in vivo (21, 25). The adaptation to acid treatment is accomplished by two adjustments: increased H+ secretion, presumably by existing and perhaps recruited α-intercalated cells, and decreased HCO3 secretion by β-intercalated cells (23, 24, 33). The rabbit CCD contains both α- and β-intercalated cells, and both types are active under normal conditions. Previously, we showed that luminal bafilomycin, by inhibiting the apical H+-ATPase of α-intercalated cells of CCDs from normal rabbits, reduces H+ secretion and thereby makes the net HCO3 flux more negative (higher rate of secretion under basal conditions). When bafilomycin was used after acid incubation, the basal and adaptive increases in H+ secretion were inhibited, such that net HCO3 transport was not significantly different from zero (33). The effect of inhibiting the H+-K+-ATPase was previously shown (33) to be much smaller than that of inhibiting the H+-ATPase in acid-adapted CCDs, indicating a limited role for the H+-K+-ATPase in adapting to metabolic acidosis. These data show that there is a substantial amount of HCO3 secretion in CCDs from normal rabbits, but after they are incubated at low pH, the HCO3 secretion rate is greatly diminished, such that when the offsetting H+ secretion by α-intercalated cells is inhibited by bafilomycin, there is virtually no net HCO3 transport. These data also show that there are partially offsetting H+ secretory fluxes in CCDs from normal rabbits but after acid incubation, the basal and adaptive increase in H+ secretion are entirely inhibited by a vacuolar H+-ATPase inhibitor. In this study, we extended these observations by demonstrating that ETB receptor signaling in response to low pH mediated the decreased HCO3 secretion as well as the adaptive increase in H+ secretion. It is not surprising that the ETB receptor antagonist entirely inhibited the adaptive increase in H+ secretion, because Wesson's group showed inhibition of B- but A-type ET receptors blunts not only the decreased HCO3 secretion but also the increased H+ secretion in the distal tubule of rats given dietary acid (42). Thus ET, via the ETB receptor, regulates adaptive changes in H+/HCO3 fluxes during metabolic acidosis.

NO has been found to play an important role in mediating salt and water transport, in addition to hemodynamics and tubuloglomerular feedback in the kidney (17). Less is known about how NO affects acid-base transport. Mice lacking nNOS have defective proximal tubule HCO3 reabsorption and develop metabolic acidosis (41). Mice lacking inducible NOS, but not endothelial NOS, also show reduced proximal HCO3 reabsorption (40). These studies show that inducible and nNOS stimulate proximal HCO3 reabsorption. Also, acidosis stimulates NO synthesis in rat lung (18) and canine heart (11), perhaps because the cellular acidosis increases cellular calcium levels (4, 38), and calcium activates NOS (15).

The CCD is known to express endothelial, inducible, and nNOS (1, 14, 30, 31, 37) and NO has important actions in the CCD. NO inhibits ADH-stimulated osmotic water permeability in the CCD (5, 6). Also, in M-1 cultured CCD cells, NO inhibits sodium transport at the apical membrane sodium channel (27). The present study adds modulation of β-intercalated cell HCO3 secretory fluxes in response to acidosis to the actions of NO on ion transport in the CCD. There are no reports of NO stimulating H+ secretion in the CCD; and furthermore, results of this study fail to demonstrate a pivotal role for NO guanylate cyclase in upregulating H+ secretion. However, Tojo et al. (31) showed that exogenous donors of NO inhibit bafilomycin-sensitive H+-ATPase activity in microdissected rat CCDs, suggesting that an inducible form of NOS downregulates H+ secretion by α-intercalated cells. It is possible that ET-ETB signaling to control H+ secretion works via another pathway that predominates over the NO-guanylate cyclase system.

Activation of soluble guanylate cyclase leading to the production of cGMP is a key function of NO as a signal generator. cGMP has several targets, with the major one being PKG (7). cGMP has recently been shown to cause the membrane insertion of aquaporin-2 in renal epithelial cells (3). Results presented in the present study demonstrate that activation of guanylate cyclase and subsequent stimulation of PKG are critical for the adaptation of the β-intercalated cell to acidosis. Based on the analysis of HCO3 flux and cell pH changes in response to luminal Cl removal, we suggest that cGMP activation of PKG stimulates endocytosis of apical Cl/HCO3 exchangers, leading to a decrease in HCO3 secretion in response to acidosis. Identification of substrates of PKG that directly or indirectly influence endocytic pathways is an important area for further investigation.

GRANTS

This work was supported in part by a Grant-in-Aid from the American Heart Association (0455829T) New York State Affiliate (G. J. Schwartz) and by grants from the Ministry of Education, Culture, Sports, Science and Technology, and Ministry of Health, Labor and Welfare of Japan (S. Tsuruoka).

Footnotes

  • 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.

REFERENCES

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 15a.
  17. 16.
  18. 17.
  19. 18.
  20. 19.
  21. 20.
  22. 21.
  23. 22.
  24. 23.
  25. 24.
  26. 25.
  27. 26.
  28. 27.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
View Abstract