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

【摘要】  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 HCO 3 - secretory fluxes and to raise the pH of -intercalated cells. Acid incubation caused reversal of polarity of net HCO 3 - transport from secretion to absorption, comprised of a 40% increase in H + secretion and a 75% decrease in HCO 3 - secretion. The ET B receptor antagonist BQ-788, as well as the NO synthase inhibitor, N G -nitro- L -arginine methyl ester ( L -NAME), attenuated the adaptive decrease in HCO 3 - secretion by 40%, but only BQ-788 inhibited the adaptive increase in H + secretion. There was no effect of inactive D -NAME or the ET A receptor antagonist BQ-123. Both BQ-788 and L -NAME inhibited the acid-induced inactivation (endocytosis) of the apical Cl - /HCO 3 - exchanger. The guanylate cyclase inhibitor LY-83583 and cGMP-dependent protein kinase inhibitor KT-5823 affected HCO 3 - transport similarly to L -NAME. These data indicate that signaling via the ET B receptor regulates the adaptation of the CCD to metabolic acidosis and that the NO guanylate cyclase component of ET B receptor signaling mediates downregulation of Cl - /HCO 3 - exchange and HCO 3 - secretion.

【关键词】  tubule microperfusion endothelin receptor antagonist nitric oxide synthase inhibitor cyclic GMP cyclic GMPdependent 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 ET B 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 ET B 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 HCO 3 - 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 HCO 3 - 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 N G -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 NH 4 Cl 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 + /HCO 3 - 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 HCO 3 -, and this net flux is made up of a small H + secretory flux that is outstripped by a much larger HCO 3 - secretory flux ( 26, 33 ). The adaptation to acidosis is associated with both a modest increase in H + secretory flux and a large decrease in HCO 3 - secretory flux, with the resultant sum of fluxes being the secretion of net protons ( 26, 33 ); that is, a reversal in polarity of net HCO 3 - 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 NaHCO 3, 2.5 K 2 HPO 4, 2 CaCl 2, 1.2 MgSO 4, 5.5 D -glucose, 1 trisodium citrate, 4 sodium lactate, and 6 L -alanine, 290 ± 2 mosmol/kgH 2 O, and gassed with 94% O 2 -6% CO 2, 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 NaHCO 3 (GIBCO, BRL, Gaithersburg, MD), Burg's solution, and dissection solution (Burg's solution with 25 mM NaHCO 3 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 HCO 3 - 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 HCO 3 - using a Nanoflo (WPI, Sarasota, FL) ( 26, 33, 35, 36 ). Net HCO 3 - was calculated as J HCO3 = (C O - C L ) x (V L / L ), where C O and C L are the HCO 3 - concentrations of perfused and collected fluid, respectively, V L is the rate of collected fluid, and L is the length of the tubule (mm) ( 26, 33 ). When HCO 3 - transport ( J HCO3 0, there is net HCO 3 - absorption; when J HCO3 is <0, there is net HCO 3 - secretion. To distinguish between unidirectional H + and HCO 3 - secretion after net HCO 3 - transport is measured, luminal Cl - was replaced by gluconate ( 21 ). In this maneuver, HCO 3 - is not secreted, thereby uncovering the unidirectional H + secretory flux. Subtracting this H + secretory flux from the net bicarbonate transport flux reveals the HCO 3 - 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, ET B receptor antagonist) ( 8 ), BQ-123 (1-10 µM, Sigma, ET A 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 ET B 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 + /HCO 3 - flux ratios and the acid-induced changes in net HCO 3 - 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 HCO 3 - 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 HCO 3 - secretion. After 3-h incubation at pH 6.8, the solutions were restored to pH 7.4 and HCO 3 - 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 HCO 3 - 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 -intercalated cells, because HCO 3 - secretion is simultaneously inhibited in the -intercalated cells. Figure 1 shows that the mean rate of net HCO 3 - 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 HCO 3 - 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 HCO 3 - 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 HCO 3 - 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 HCO 3 - 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 + /HCO 3 - 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 HCO 3 - secretion). Right : HCO 3 - 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.

Table 1. Basal and acid incubation net and H + /HCO 3 - fluxes

Fig. 2. Inhibition of ET B 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 HCO 3 - 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 ET B 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 ET B receptor signaling blocks adaptive changes in HCO 3 - secretion induced by acid incubation. The ratio of acid-incubated/basal HCO 3 - 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.

Adaptive changes in H + /HCO3- secretion fluxes induced by low pH are inhibited by ET B 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 + /HCO 3 - secretory fluxes induced by incubation at low pH. Acid incubation with BQ-788 failed to reverse the polarity of the net HCO 3 - flux ( Table 1 ); that is, net HCO 3 - secretion was reduced to no significant HCO 3 - 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 HCO 3 - 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 HCO 3 - secretion (-7.98 ± 0.35 to -4.00 ± 0.15 pmol·min -1 ·mm -1, P < 0.01); the acid-incubated HCO 3 - 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 ET A receptor antagonist used at 1-10 µM, did not prevent the changes in H + secretion and HCO 3 - secretion resulting from the incubation at pH 6.8. Acid incubation with BQ-123 resulted in a reversal of polarity of net HCO 3 - flux that was comparable to CCDs treated with vehicle control ( Table 1 and Figs. 2 - 4 ). Because the response to acid incubation in the presence of BQ-788 resulted in a larger HCO 3 - 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 ET B, but not ET A, 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 HCO 3 - 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 HCO 3 - 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 HCO 3 - flux was significant ( P < 0.01), the acid-induced reversal of polarity of HCO 3 - 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 HCO 3 - 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 HCO 3 - 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 HCO 3 - 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 HCO 3 - 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 HCO 3 - secretory flux was significantly increased over vehicle, indicating a failure to adaptively reduce HCO 3 - 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 HCO 3 - flux ( Table 1 and Figs. 2 - 4 ), similar to what was observed with vehicle control. These data show that NOS inhibition via L -NAME diminishes the downregulation of HCO 3 - 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 ET B receptor antagonist, BQ-788 ( Table 1 and Fig. 2 ), so that the end result is not a reversal in polarity of HCO 3 - transport, but a reduction to zero net transport.

ET B receptor antagonism and L -NAME block the effect of acidosis on chloride-dependent cell pH changes. To confirm the changes in HCO 3 - transport, we examined -intercalated cell pH changes in response to removal of Cl - from the lumen or bath. Previously, we showed that -intercalated cells express apical Cl - /HCO 3 - exchangers and basolateral Cl - conductances ( 21, 25 ). Figure 5 A 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. 5 B ), 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 - /HCO 3 - exchange and basolateral Cl - exit would be appropriate during metabolic acidosis.

Fig. 5. ET B receptor signaling is required for loss of luminal Cl - /HCO 3 - 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 - /HCO 3 - exchange (0.46 ± 0.02 to 0.46 ± 0.01 pH U, P = NS; Fig. 5 C ) and in sensitivity to basolateral Cl - removal (-0.44 ± 0.03 to -0.44 ± 0.04, P = NS; Fig. 5 D ). Similarly, L -NAME added to the pH 6.8 incubation clearly prevented the reduction in apical Cl - /HCO 3 - exchange (0.47 ± 0.03 to 0.45 ± 0.03 pH U, P = NS; Fig. 5 E ) and in sensitivity to basolateral Cl - removal (-0.48 ± 0.04 to -0.51 ± 0.05 pH U, P = NS; Fig. 5 F ). These results confirm that ET B receptor signaling and NO synthesis are required for the adaptive decrease in HCO 3 - secretion (presumably by endocytosis of apical Cl - /HCO 3 - exchangers) induced by incubation at low pH.

cGMP generation and subsequent activation of PKG are required for acid-induced changes in H + /HCO 3 - 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 HCO 3 - 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 HCO 3 - flux in response to low pH, and there was no net HCO 3 - 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 HCO 3 - 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 HCO 3 - 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 HCO 3 - 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 + /HCO 3 - fluxes. KT-5823 substantially prevented the reversal of polarity of HCO 3 - 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 HCO 3 - 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 HCO 3 - secretion that would be anticipated in response to metabolic acidosis.

Taken together, these data indicate that the adaptive decrease in HCO 3 - secretion in response to acidosis is mediated in large part by ET-ET B -NO-guanylate cyclase-PKG signaling pathway. In contrast, the adaptive increase in H + secretion, which may be directly stimulated by ET-ET B 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 HCO 3 - flux from net secretion to net absorption ( 21, 26 ). This model recapitulates the pathophysiological condition of 3 days of NH 4 Cl 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 -intercalated cells, and decreased HCO 3 - secretion by -intercalated cells ( 23, 24, 33 ). The rabbit CCD contains both - and -intercalated cells, and both types are active under normal conditions. Previously, we showed that luminal bafilomycin, by inhibiting the apical H + -ATPase of -intercalated cells of CCDs from normal rabbits, reduces H + secretion and thereby makes the net HCO 3 - 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 HCO 3 - 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 HCO 3 - secretion in CCDs from normal rabbits, but after they are incubated at low pH, the HCO 3 - secretion rate is greatly diminished, such that when the offsetting H + secretion by -intercalated cells is inhibited by bafilomycin, there is virtually no net HCO 3 - 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 ET B receptor signaling in response to low pH mediated the decreased HCO 3 - secretion as well as the adaptive increase in H + secretion. It is not surprising that the ET B 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 HCO 3 - secretion but also the increased H + secretion in the distal tubule of rats given dietary acid ( 42 ). Thus ET, via the ET B receptor, regulates adaptive changes in H + /HCO 3 - 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 HCO 3 - reabsorption and develop metabolic acidosis ( 41 ). Mice lacking inducible NOS, but not endothelial NOS, also show reduced proximal HCO 3 - reabsorption ( 40 ). These studies show that inducible and nNOS stimulate proximal HCO 3 - 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 -intercalated cell HCO 3 - 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 -intercalated cells. It is possible that ET-ET B 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 -intercalated cell to acidosis. Based on the analysis of HCO 3 - flux and cell pH changes in response to luminal Cl - removal, we suggest that cGMP activation of PKG stimulates endocytosis of apical Cl - /HCO 3 - exchangers, leading to a decrease in HCO 3 - 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).

【参考文献】
  Ahn KY, Mohaupt MG, Madsen KM, and Kone BC. In situ hybridization localization of mRNA encoding inducible nitric oxide synthase in rat kidney. Am J Physiol Renal Fluid Electrolyte Physiol 267: F748-F757, 1994.

Bhowmick N, Narayan P, and Puett D. The endothelin subtype A receptor undergoes agonist- and antagonist-mediated internalization in the absence of signaling. Endocrinology 139: 3185-3192, 1998.

Bouley R, Breton S, Sun T, McLaughlin M, Nsumu NN, Lin HY, Ausiello DA, and Brown D. Nitric oxide and atrial natriuretic factor stimulate cGMP-dependent membrane insertion of aquaporin 2 in renal epithelial cells. J Clin Invest 106: 1115-1126, 2000.

Cannon C, van Adelsberg J, Kelly S, and Al-Awqati Q. Carbon-dioxide-induced exocytotic insertion of H + pumps in turtle bladder luminal membrane: role of cell pH and calcium. Nature 314: 443-446, 1985.

Garcia NH, Pomposiello SI, and Garvin JL. Nitric oxide inhibits ADH-stimulated osmotic water permeability in cortical collecting ducts. Am J Physiol Renal Fluid Electrolyte Physiol 270: F206-F210, 1996.

Garcia NH, Stoos BA, Carretero OA, and Garvin JL. Mechanism of the nitric oxide-induced blockade of collecting duct water permeability. Hypertension 27: 679-683, 1996.

Hanafy KA, Krumenacker JS, and Murad F. NO, nitrotyrosine, and cyclic GMP in signal transduction. Med Sci Monit 7: 801-819, 2001.

Hercule HC and Oyekan AO. Cytochrome P450 omega/omega-1 hydroxylase-derived eicosanoids contribute to endothelin(A) and endothelin(B) receptor-mediated vasoconstriction to endothelin-1 in the rat preglomerular arteriole. J Pharmacol Exp Ther 292: 1153-1160, 2000.

Herrera M and Garvin JL. Endothelin stimulates endothelial nitric oxide synthase expression in the thick ascending limb. Am J Physiol Renal Physiol 287: F231-F235, 2004.

Hirata Y, Emori T, Eguchi S, Kanno K, Imai T, Ohta K, and Marumo F. Endothelin receptor subtype B mediates synthesis of nitric oxide by cultured bovine endothelial cells. J Clin Invest 91: 1367-1373, 1993.

Kitakaze M, Node K, Takashima S, Asanuma H, Asakura M, Sanada S, Shinozaki Y, Mori H, Sato H, Kuzuya T, and Hori M. Role of cellular acidosis in production of nitric oxide in canine ischemic myocardium. J Mol Cell Cardiol 33: 1727-1737, 2001.

Kohan DE. Endothelin synthesis by rabbit renal tubule cells. Am J Physiol Renal Fluid Electrolyte Physiol 261: F221-F226, 1991.

Laghmani K, Preisig PA, Moe OW, Yanagisawa M, and Alpern RJ. Endothelin-1/endothelin-B receptor-mediated increases in NHE3 activity in chronic metabolic acidosis. J Clin Invest 107: 1563-1569, 2001.

Mohaupt MG, Elzie JL, Ahn KY, Clapp WL, Wilcox CS, and Kone BC. Differential expression and induction of mRNAs encoding two inducible nitric oxide synthases in rat kidney. Kidney Int 46: 653-665, 1994.

Moncada S and Higgs A. The L -arginine-nitric oxide pathway. N Engl J Med 329: 2002-2012, 1993.

Ohuchi T, Yanagisawa M, and Gariepy CE. Renal tubular effects of endothelin-B receptor signaling: its role in cardiovascular hemeostasis and extracellular volume regulation. Curr Opin Nephrol Hypertens 9: 435-439, 2000.

Ortiz PA and Garvin JL. Autocrine effects of nitric oxide on HCO 3 - transport by rat thick ascending limb. Kidney Int 58: 2069-2074, 2000.

Ortiz PA and Garvin JL. Role of nitric oxide in the regulation of nephron transport. Am J Physiol Renal Physiol 282: F777-F784, 2002.

Pedoto A, Caruso JE, Nandi J, Oler A, Hoffmann SP, Tassiopoulos AK, McGraw DJ, Camporesi EM, and Hakim TS. Acidosis stimulates nitric oxide production and lung damage in rats. Am J Respir Crit Care Med 159: 397-402, 1999.

Plato CF, Pollock DM, and Garvin JL. Endothelin inhibits thick ascending limb chloride flux via ET B receptor-mediated NO release. Am J Physiol Renal Physiol 279: F326-F333, 2000.

Plato CF, Stoos BA, Wang D, and Garvin JL. Endogenous nitric oxide inhibits chloride transport in the thick ascending limb. Am J Physiol Renal Physiol 276: F159-F163, 1999.

Satlin LM and Schwartz GJ. Cellular remodeling of HCO 3 - -secreting cells in rabbit renal collecting duct in response to an acidic environment. J Cell Biol 109: 1279-1288, 1989.

Schnermann J, Lorenz JN, Briggs JP, and Keiser JA. Induction of water diuresis by endothelin in rats. Am J Physiol Renal Fluid Electrolyte Physiol 263: F516-F526, 1992.

Schuster VL. Function and regulation of collecting duct intercalated cells. Annu Rev Physiol 55: 267-288, 1993.

Schwartz GJ. Plasticity of intercalated cell polarity: effect of metabolic acidosis. Nephron 87: 304-313, 2001.

Schwartz GJ, Barasch J, and Al-Awqati Q. Plasticity of functional epithelial polarity. Nature 318: 368-371, 1985.

Schwartz GJ, Tsuruoka S, Vijayakumar S, Petrovic S, Mian A, and Al-Awqati Q. Acid incubation reverses the polarity of intercalated cell transporters, an effect mediated by hensin. J Clin Invest 109: 89-99, 2002.

Stoos BA, Carretero OA, and Garvin JL. Endothelial-derived nitric oxide inhibits sodium transport by affecting apical membrane channels in cultured collecting duct cells. J Am Soc Nephrol 4: 1855-1860, 1994.

Takemoto F, Uchida S, Ogata E, and Kurokawa K. Endothelin-1 and endothelin-3 binding to rat nephrons. Am J Physiol Renal Fluid Electrolyte Physiol 264: F827-F832, 1993.

Terada Y, Tomita K, Nonoguchi H, and Marumo F. Polymerase chain reaction localization of constitutive nitric oxide synthase and soluble guanylate cyclase messenger RNAs in microdissected rat nephron segments. J Clin Invest 90: 659-665, 1992.

Tojo A, Guzman NJ, Garg LC, Tisher CC, and Madsen KM. Nitric oxide inhibits bafilomycin-sensitive H + -ATPase activity in rat cortical collecting duct. Am J Physiol Renal Fluid Electrolyte Physiol 267: F509-F515, 1994.

Tomita K, Nonoguchi H, Terada Y, and Marumo F. Effects of ET-1 on water and chloride transport in cortical collecting ducts of the rat. Am J Physiol Renal Fluid Electrolyte Physiol 264: F690-F696, 1993.

Tsuruoka S and Schwartz GJ. Adaptation of rabbit cortical collecting duct HCO 3 - transport to metabolic acidosis in vitro. J Clin Invest 97: 1076-1084, 1996.

Tsuruoka S and Schwartz GJ. Metabolic acidosis stimulates H + secretion in the rabbit outer medullary collecting duct (inner stripe) of the kidney. J Clin Invest 99: 1420-1431, 1997.

Tsuruoka S, Schwartz GJ, Wakaumi M, Nishiki K, Yamamoto H, Purkerson JM, and Fujimura A. Nitric oxide production modulates cyclosporin A-induced distal renal tubular acidosis in the rat. J Pharmacol Exp Ther 305: 840-845, 2003.

Tsuruoka S, Swenson ER, Petrovic S, Fujimura A, and Schwartz GJ. Role of basolateral carbonic anhydrase in proximal tubular fluid and bicarbonate absorption. Am J Physiol Renal Physiol 280: F146-F154, 2001.

Ujiie K, Yuen J, Hogarth L, Danziger R, and Star RA. Localization and regulation of endothelial NO synthase mRNA expression in rat kidney. Am J Physiol Renal Fluid Electrolyte Physiol 267: F296-F302, 1994.

Van Adelsberg J and Al-Awqati Q. Regulation of cell pH by Ca 2+ -mediated exocytotic insertion of H + -ATPases. J Cell Biol 102: 1638-1645, 1986.

Wang H, Carretero OA, and Garvin JL. Nitric oxide produced by THAL nitric oxide synthase inhibits TGF. Hypertension 39: 662-666, 2002.

Wang T. Role of iNOS and eNOS in modulating proximal tubule transport and acid-base balance. Am J Physiol Renal Physiol 283: F658-F662, 2002.

Wang T, Inglis FM, and Kalb RG. Defective fluid and HCO 3 - absorption in proximal tubule of neuronal nitric oxide synthase-knockout mice. Am J Physiol Renal Physiol 279: F518-F524, 2000.

Wesson DE. Endogenous endothelins mediate increased distal tubule acidification induced by dietary acid in rats. J Clin Invest 99: 2203-2211, 1997.

Wesson DE and Dolson GM. Endothelin-1 increases rat distal tubule acidification in vivo. Am J Physiol Renal Physiol 273: F586-F594, 1997.

Wesson DE, Simoni J, and Green DF. Reduced extracellular pH increases endothelin-1 secretion by human renal microvascular endothelial cells. J Clin Invest 101: 578-583, 1998.

Yanagisawa M, Kurihara H, Kimura S, Tomobe Y, Kobayashi M, Mitsui Y, Yazaki Y, Goto K, and Masaki T. A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature 332: 411-415, 1988.

Yasoshima K, Satlin LM, and Schwartz GJ. Adaptation of rabbit cortical collecting duct to in vitro acid incubation. Am J Physiol Renal Fluid Electrolyte Physiol 263: F749-F756, 1992.

作者单位：1 Department of Pharmacology, Jichi Medical School, Tochigi, Japan; and 2 Department of Pediatrics, University of Rochester School of Medicine, Rochester, New York