Regulation by Glucocorticoids of Expression and Activity of rBSC1, the Na+-K+(NH4 +)-2Cl−Cotransporter of Medullary Thick Ascending Limb*

To assess whether glucocorticoids regulate rBSC1, the apical Na+-K+( NH4 +)-2Cl−cotransporter of kidney medullary thick ascending limb ( MTAL ), studies were performed in normal rats, adrenalectomized ( ADX ) rats, and ADX rats infused with dexamethasone for 6 days. The effects of dexamethasone on rBSC1 were also studied in vitro using isolated rat MTAL segments. Cotransport activity was estimated by intracellular pH measurements; rBSC1 protein was quantified in MTAL crude membranes by immunoblotting analysis, and mRNA was quantified by quantitative reverse transcription-polymerase chain reaction. The abundance of rBSC1 protein and mRNA increased in ADX rats infused with dexamethasone compared with ADX rats (p < 0.04). In addition, application of dexamethasone for 1–3 h to MTALs caused rBSC1 protein and mRNA abundance and cotransport activity to significantly increase in a hyperosmotic medium (450 mosmol/kg of H2O) containing 0.7 nmarginine vasopressin, which is an in vitro experimental condition that resembles the in vivo MTAL environment. Results obtained in various media and with 8-bromo-cAMP indicated that stimulation of rBSC1 expression by glucocorticoids required interactions between glucocorticoid receptor- and cAMP-dependent factors. Up to 100 nm d-aldosterone had no effect on cotransport activityin vitro. Thus glucocorticoids directly stimulate MTAL rBSC1 expression and activity, which contributes to glucocorticoid-dependent effects on the renal regulation of acid-base balance and urinary concentrating ability.

Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport is responsible for the apical step of NaCl and ammonia transport by the thick ascending limb (TAL) 1 of the nephron. NaCl and ammonia absorption without water by the medullary TAL (MTAL) causes transepi-thelial concentration differences of these solutes, which constitutes the "single effects" responsible for NaCl and ammonia accumulation in the renal medulla. This is critical both to the level of renal medullary hyperosmolality and thus to the urinary concentrating ability of the kidney and to urinary ammonia excretion and thus to the renal regulation of acid-base balance (1,2). The MTAL apical Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransporter (BSC1 (bumetanide-sensitive cotransporter) or NKCC2 (Na ϩ -K ϩ -Cl Ϫ cotransporter)) was recently cloned from rat (3), mouse (4,5), rabbit (6), and human (7) kidneys. The transporter protein has been localized at the apical membrane of the TAL as well as at the macula densa (8 -10). BSC1 was recently shown to be up-regulated by chronic saline loading (9), restriction of water intake and arginine vasopressin (AVP) administration (11), and metabolic acidosis (12) and down-regulated by potassium depletion (13). However, the stimuli and cellular mechanisms of these adaptations of rBSC1 expression were not specified in the latter in vivo studies.
The presence of specific glucocorticoid receptors (GR) in the MTAL has been demonstrated by binding, immunological, and mRNA detection methods (14 -17). A number of studies have suggested that in vivo glucocorticoid administration acts on the MTAL, but little is known about the direct effects of glucocorticoids on the functions of this nephron segment. Dexamethasone has been shown to stimulate within a few hours the Na ϩ /K ϩ -ATPase activity of MTALs incubated in vitro (18,19). It must be pointed out that glucocorticoids have long been known to contribute to renal urinary concentrating ability through (at least in part) maintenance of medullary hyperosmolality (20) and to the increased urinary excretion of NH 4 ϩ in response to metabolic acidosis (21) and that both processes are greatly dependent on Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport activity in the MTAL, as pointed out above. These considerations prompted us to design the present study to assess whether glucocorticoids affect rBSC1 expression in the MTAL. To this end we have measured the effects of a 1-3 h in vitro application of glucocorticoids on rBSC1 transport activity and protein and mRNA abundance. We have also determined the abundance of rBSC1 protein and mRNA after in vivo glucocorticoid administration. The results show that rBSC1 expression in the MTAL is up-regulated by glucocorticoids through interactions between GR-and cAMP-dependent pathways.

EXPERIMENTAL PROCEDURES
In Vivo Studies-Male Harlan Sprague-Dawley rats weighing 250 -300 g were allowed free access to standard rat chow and drinking solution up to the time of the experiments. Rats were adrenalectomized (ADX) under light ether anesthetization and given 0.9% NaCl in distilled water as drinking solution during 6 days before the experiments. Some ADX rats were administered dexamethasone at 1.2 g/100 g of body weight/day, a dose that is known to restore normal glucocorticoid activity, delivered by continuous infusion through subcutaneously implanted Alzet micro-osmotic pumps (Alza Corporation, Palo Alto, CA) during 6 days (ADX ϩ Dexa); these ADX ϩ Dexa rats also drank 0.9% NaCl in distilled water. Control rats from the same shipments were sham operated and drank normal water or 0.9% NaCl as drinking solution. After anesthetization by sodium pentobarbital, the kidneys were rapidly removed and cut into thin slices along the corticopapillary axis, and under a dissecting microscope the inner stripe of outer medulla of each slice was excised and cut into uniform small pieces, which were used for immunoblotting of membrane proteins and mRNA determinations.
In Vitro Studies: Suspension of Rat MTAL Tubules-The method used to isolate MTAL fragments in suspension has been previously described (22). We have established by light and electron microscopy that this suspension was made almost exclusively of MTALs (Ն95%), occasional thin limbs, and rare outer medullary collecting tubules, with no isolated cells or proximal tubules (12,22,23). MTAL suspensions were prepared from ADX rats that were given 0.9% NaCl in drinking distilled water for 6 days. Samples destined for rBSC1 protein and mRNA quantification were incubated for 2 h in the presence of dexamethasone as the glucocorticoid hormone or vehicle. Samples destined for measurement of intracellular pH (pH i ) to estimate the Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport activity were loaded with 2Ј,7Ј-bis(2-carboxyethyl)-5(6)-carboxyfluorescein acetoxymethyl ester, and measurements of Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport activity were performed, as described previously (12,24), after 1-3 h of incubation in the presence of dexamethasone or vehicle. In brief, samples of 2Ј,7Ј-bis(2-carboxyethyl)-5(6)-carboxyfluorescein acetoxymethyl ester-loaded MTALs, preincubated in solution A or C (Table I), were diluted in the fluorometer cuvette in 2 ml of solution B or D (Table I), and pH i was monitored at 37°C; then Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport activity was assessed by determining the bumetanide-sensitive component of the cell acidification caused by abrupt exposure to 4 mM NH 4 Cl in the presence of 10 mM barium and 1 M amiloride to block NH 4 ϩ carriers other than Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport (12,24). We have previously demonstrated that the initial rate of NH 4 ϩ -induced cell acidification (dpH i /dt, calculated as described previously (24)) is not significantly affected by changes in the activities of pH i regulatory mechanisms such as Na ϩ /H ϩ antiport (24). However, we have determined in the present study, by a previously described method (23,24), that the total Na ϩ /H ϩ exchange activity of MTAL fragments in suspension was not affected by 10 nM dexamethasone (data not shown). Thus the initial rate of NH 4 ϩ -induced cell acidification in the presence of 10 mM barium and 1 M amiloride will be hereafter referred to as Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport activity. Crude membranes from the inner stripe of outer medulla or from MTAL suspensions were prepared for immunoblotting studies in the following way. Tissues were homogenized in a medium composed of 125 mM sucrose, 12 mM Trizma (Tris base) (pH 7.4), 0.1 mM 4-(2-aminoethyl)-benzenesulfonyl fluoride, and 5 g/ml leupeptin. These homogenates were centrifuged at 1000 ϫ g for 5 min in a Beckman GS-6KR centrifuge with a GH-3.7 rotor, and the supernatants were further centrifuged at 200,000 ϫ g for 60 min in a Beckman L-70 Ultracentrifuge with a 70 TI rotor. The membrane pellets were suspended in the above medium and stored at Ϫ80°C until use.
Electrophoresis and Immunoblotting of Membrane Proteins-Semiquantification of membrane protein amounts was performed as described previously (12). In brief, the membranes were solubilized at ambient temperature for 20 min in Laemmli medium containing (final concentrations) 62.5 mM Tris-HCl (pH 6.8), 5% SDS, 100 mM dithiothreitol, and 10% glycerol. SDS-polyacrylamide gel electrophoresis was performed with solubilized membranes (15-25 g of protein) and prestained molecular weight markers (Sigma) on 7.5% polyacrylamide minigels (Mini Protean II, Bio-Rad). Protein was subsequently transferred electrophoretically from the gels to nitrocellulose membranes (Mini Trans Blot Module, Bio-Rad). Equal loading and transfer efficiency were systematically checked by Ponceau red staining of the nitrocellulose membranes. Exposition of the membranes to an anti-rBSC1 polyclonal antibody (8,12), to an anti-␤-actin mouse monoclonal antibody (Sigma-Aldrich Fine Chemicals), and then to the second antibodies (peroxidase-linked anti-rabbit Ig and anti-mouse Ig (Bio-Rad)) and quantification of each band were performed as described previously (12). Quantification of ␤-actin was used as an additional control to check equal loading and transfer efficiency in the nitrocellulose membranes.
RNA Extraction, Reverse Transcription, and Polymerase Chain Reaction-Total RNA (RNA tot ) was extracted from aliquots of the inner stripe of outer medulla or from MTAL suspensions with use of the SV total RNA isolation system kit (Promega). A competitor RNA that differed from the wild rBSC1 mRNA by a 116-base deletion of the latter was obtained, and quantitative reverse transcription-polymerase chain reaction was performed exactly as described previously in detail (12). Amounts of rBSC1 mRNA are expressed in amol per 100 ng of RNA tot . Note that both the rBSC1 protein and mRNA determination methods employed in the present work were designed to detect determinants common to the published rBSC1 isoforms that take place in the rat MTAL (12).
Statistics-Results are expressed as the mean Ϯ S.E. Statistical significance between experimental groups was assessed by Student's paired or unpaired t test or by 1-way analysis of variance completed by a t test using the within-groups residual variance of the analysis of variance, as appropriate.

RESULTS
In Vivo Studies-There was no difference in the plasma concentrations of sodium and chloride between three control (145.7 Ϯ 0.3 mM and 98 Ϯ 1, respectively), three ADX (142.7 Ϯ 1.5 and 97 Ϯ 2, respectively), and three ADX ϩ Dexa rats (145.3 Ϯ 0.7 and 96 Ϯ 2, respectively); as expected, the plasma potassium concentration was higher in ADX (5.1 Ϯ 0.5 mM) and ADX ϩ Dexa (5.0 Ϯ 0.2 mM) than in control rats (4.4 Ϯ 0.1 mM). As shown in Table II, the abundance of rBSC1 protein was lower in ADX than in control rats that drank normal water, which was of borderline significance (p Ͻ 0.06), or 0.9% NaCl (p Ͻ 0.009); however, there was no difference in rBSC1 mRNA abundance. It should be emphasized that, when compared with the normal condition in rats, adrenalectomy is a complex condition in which several factors may have had opposing effects on rBSC1 expression, as discussed below. Thus, in another experimental series, results obtained from 5 ADX rats were compared with those obtained from 5 ADX ϩ Dexa rats. As shown in Fig. 1, dexamethasone administration increased rBSC1 protein abundance in crude membranes of the inner stripe of outer medulla by ϳ91% (191 Ϯ 23 arbitrary units in  (Fig. 2). These results establish that glucocorticoid administration enhances rBSC1 expression in the MTAL.
In Vitro Studies-To assess wether glucocorticoids directly stimulate rBSC1 expression in the MTAL in vitro, tubule fragments were incubated in experimental media in the presence of 10 nM dexamethasone or vehicle. In an attempt to recreate the in vivo MTAL environment, we performed in vitro experiments using a moderately hyperosmotic medium (ϳ450 mosmol/kg of H 2 O obtained by adding 50 mM NaCl plus 50 mM urea to isoosmotic medium; solution C in Table I). In addition, this hyperosmotic medium contained 0.7 nM AVP or 0.5 mM 8-bromo-cAMP. Indeed, the MTAL is surrounded in vivo by the hyperosmotic interstitial medium of the inner stripe of outer medulla of the kidney and is submitted to tonic influences by cAMP-generating peptide hormones such as AVP, glucagon, and calcitonin (25). In the hyperosmotic medium containing 0.7 nM AVP, within 2 h of incubation dexamethasone increased the abundance of rBSC1 protein and mRNA. rBSC1 protein abundance increased from 100 Ϯ 4 arbitrary units in control to 137 Ϯ 12 arbitrary units (p Ͻ 0.02; Fig. 3), whereas rBSC1 mRNA abundance increased from 1.1 Ϯ 0.6 amol/100 ng of RNA tot in control to 2.5 Ϯ 0.5 amol/100 ng of RNA tot (p Ͻ 0.04; Fig. 3). There was no difference in ␤-actin protein abundance (100 Ϯ 14 versus 95 Ϯ 30 arbitrary units; n ϭ 6 for both; NS). The increases in rBSC1 mRNA and protein were accompanied with stimulation of Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport activity. Dexamethasone increased the cotransport activity by ϳ25% (p Ͻ 0.04; Fig. 4) within 1-3 h of incubation in the hyperosmotic medium containing 0.7 nM AVP. Stimulation of the cotransport activity by dexamethasone was also observed when 0.5 mM

TABLE II
Quantification of rBSC1 protein and mRNA abundance in the inner stripe of outer medulla of control and ADX rats Values are the mean Ϯ S.E. In series I, ADX rats were compared to control rats that drank normal water (Sham). In series II, ADX rats were compared to control rats that drank 0.9% NaCl (Sham ϩ NaCl). The abundance of rBSC1 protein was assessed in duplicate in each rat. NS, not significant; ND, not done. 8-bromo-cAMP was added in place of AVP in the hyperosmotic medium (the dpH i /dt was Ϫ0.86 Ϯ 0.05 versus Ϫ0.66 Ϯ 0.03 pH unit/min in controls; n ϭ 18 and 20, respectively; p Ͻ 0.002). By contrast, dexamethasone had no effect on NH 4 ϩ -induced dpH i /dt in the presence of 0.1 mM bumetanide, which blocks the Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport activity (Ϫ1.20 Ϯ 0.04 versus Ϫ1.17 Ϯ 0.08 pH unit/min in controls; n ϭ 6 and 7, respectively; NS). Furthermore, dexamethasone-induced stimulation of cotransport activity in the hyperosmotic medium containing AVP was abolished by 20 M actinomycin D or 20 M cycloheximide (Fig. 4), which are inhibitors of transcription and protein synthesis, respectively.
To further study the interactions between dexamethasone and AVP or the experimental medium, dexamethasone was applied for 2 h to MTALs incubated in an isoosmotic medium (ϳ300 mosmol/kg of H 2 O; solution A in Table I) containing 0.7 nM AVP or in AVP-free hyperosmotic and isoosmotic media. In the isoosmotic medium containing 0.7 nM AVP, dexamethasone again strongly increased both rBSC1 protein abundance from 100 Ϯ 9 arbitrary units in control to 212 Ϯ 18 arbitrary units (p Ͻ 0.0004; Fig. 5) and mRNA abundance from 7.3 Ϯ 0.4 amol/100 ng of RNA tot in control to 21.0 Ϯ 3.4 amol/100 ng of RNA tot (p Ͻ 0.04; Fig. 5). There was no difference in ␤-actin protein abundance (100 Ϯ 3 arbitrary units in controls versus 107 Ϯ 6 arbitrary units; n ϭ 6 for both; NS). By contrast, in the AVP-free hyperosmotic medium, dexamethasone decreased rBSC1 protein abundance from 100 Ϯ 1 arbitrary units in control to 62 Ϯ 5 arbitrary units (p Ͻ 0.0001; Fig. 6) but did not affect rBSC1 mRNA abundance (4.5 Ϯ 1.0 amol/100 ng RNA tot in control versus 4.5 Ϯ 1.1 amol/100 ng RNA tot ; NS; Fig. 6). There was no difference in ␤-actin protein abundance (100 Ϯ 4 arbitrary units in controls versus 96 Ϯ 7 arbitrary units; n ϭ 6 for both; NS). Finally, in the AVP-free isoosmotic medium, dexamethasone decreased the abundance of rBSC1 protein and mRNA in MTAL suspensions within 2 h. rBSC1 protein abundance decreased from 100 Ϯ 4 arbitrary units in control to 74 Ϯ 8 arbitrary units (p Ͻ 0.02; Fig. 7), whereas rBSC1 mRNA abundance decreased from 2.8 Ϯ 0.3 amol/100 ng of RNA tot in control to 2.1 Ϯ 0.4 amol/100 ng of RNA tot , which was of borderline significance (p Ͻ 0.06; Fig. 7). There was no difference in ␤-actin protein abundance (100 Ϯ 8 versus 80 Ϯ 24 arbitrary units; n ϭ 4 for both; NS). The decreases in rBSC1 mRNA and protein abundance were accompanied by inhibition of Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport activity. Indeed, 10 nM dexamethasone decreased the cotransport activity by ϳ27% within 1-3 h of incubation (p Ͻ 0.001; Fig. 8). Dexamethasone had no effect on NH 4 ϩ -induced dpH i /dt in the presence of 0.1 mM bumetanide (Ϫ1.05 Ϯ 0.06 versus Ϫ1.09 Ϯ 0.05 pH unit/min in controls; n ϭ 12 for both; NS). Furthermore, dexamethasoneinduced inhibition of cotransport activity in the isoosmotic medium was abolished by 20 M actinomycin D or 20 M cycloheximide (Fig. 8). Thus the stimulating effects of dexamethasone required the presence of AVP in both hyperosmotic and isoosmotic media.
To validate the changes in dpH i /dt in the presence of 10 mM barium plus 1 M amiloride described above as reflecting the effects of glucocorticoids on the Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport activity, the following control experiments were performed. Dexamethasone had no effect on the cell-buffering capacity (76 Ϯ 6 versus 80 Ϯ 4 mmol of H ϩ /pH unit/liter; NS) or on the cell volume (0.39 Ϯ 0.01 versus 0.39 Ϯ 0.01 nl/mm of tubule length; NS); cell buffering capacity and cell volume were estimated exactly as described previously (24). We have also checked that, after up to 3 h of incubation, Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport activity was inhibited by the hyperosmotic medium (from Ϫ0.59 Ϯ 0.07 pH unit/min in the isoosmotic medium to Ϫ0.43 Ϯ 0.04 pH unit/min; p Ͻ 0.05) and was stimulated by 0.5 mM 8-bromo-cAMP in the isoosmotic medium (from Ϫ0.62 Ϯ 0.09 pH unit/min in controls to Ϫ0.95 Ϯ 0.02 pH unit/min; p Ͻ 0.01). Finally, 10 and 100 nM d-aldosterone had no effect on the Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport activity (Fig. 9).

DISCUSSION
This study is the first, to our knowledge, in which possible effects of glucocorticoids on rBSC1 expression in the MTAL were assessed both in vivo and in vitro. Comparing the level of rBSC1 expression in ADX rats to that in normal rats to assess the effects of glucocorticoid deficiency can hardly be achieved satisfactorily because adrenalectomy is a complex condition with respect to rBSC1 expression. For example, adrenalectomy is associated with increased circulating AVP concentrations probably because of impaired cardiac function (26), and AVP administration and cardiac insufficiency, even with normal AVP levels, both have been shown to strongly stimulate rBSC1 expression in the MTAL (11,27). In addition, NaCl administration, which was used in ADX rats to minimize urinary NaCl losses, has also been shown to stimulate rBSC1 expression (9). With these issues in mind, we observed that the abundance of rBSC1 protein in ADX rats tended to be lower as compared with rats given normal water (p Ͻ 0.06) and was significantly lower as compared with rats given 0.9% NaCl as drinking solution, like ADX rats (p Ͻ 0.009); however, there was no significant difference in rBSC1 mRNA abundance in both experimental series. On the other hand, supplementing ADX rats with dexamethasone appears to be a better means of assessing the effects of glucocorticoids on rBSC1. Glucocorticoid administration to ADX rats strongly stimulated rBSC1 mRNA and protein expression in the MTAL as compared with ADX rats. Thus these results establish that glucocorticoids enhance rBSC1 expression when administered in vivo. Furthermore, when dexamethasone was directly applied to MTALs in vitro, rBSC1 mRNA and protein expression and cotransport activity were stimulated in a hyperosmotic medium containing AVP, an experimental condition that resembles the natural MTAL environment. Strong stimulation of rBSC1 expression by dexamethasone was observed in an AVP-containing isoosmotic medium as well, and a glucocorticoid-induced increase in Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransport activity also occurred when 8-bromo-cAMP was used in place of AVP in a hyperosmotic medium. The stimulating effects of dexamethasone were not seen in AVP-and cAMP-free media. Thus it is clear that stimulation of rBSC1 expression by dexamethasone required interactions with cAMP-dependent factors. Because several peptide hormones such as AVP, calcitonin, and glucagon stimulate the MTAL adenylyl cyclase, it is very likely that MTAL cells in vivo are chronically subjected to the influences of cAMP-generating peptide hormones (25). The in vitro effects of dexamethasone on cotransport activity were abolished by actinomycin D or cycloheximide and were not observed with d-aldosterone. This indicates that GR activation and attendant effects on gene transcription and translation were responsible for the changes in rBSC1 expression and activity. The present finding that glucocorticoids physiologically stimulate rBSC1 expression and activity would be consistent with the observations that dexamethasone also stimulates Na ϩ /K ϩ -ATPase activity in the MTAL (18,19). Both of these effects would provide the mechanism for glucocorticoid-dependent stimulation of NaCl and NH 4 ϩ transport by the TAL. The intracellular mechanisms by which glucocorticoids exerted their effects in the MTAL were not investigated in the present study. As mentioned above, GR activation in vitro appears to be able to stimulate as well as to inhibit rBSC1 expression and activity depending on the presence or absence of cAMP. The results obtained in the various experimental media suggest that several interactions between GR activation and osmolality-and cAMP-dependent factors take place in the MTAL to physiologically enhance rBSC1 expression. Thus a number of intracellular events, such as altered rBSC1 gene transcription, mRNA decay, translation efficiency, and membrane trafficking of rBSC1 protein, may have combined to explain our results. Regulation of accessory protein expression may also have occurred. It is well known that the GR can interact with several other proteins through protein-protein interactions or with transcription factors at the level of the promoters of regulated genes, which may be controlled by cAMP. In particular, glucocorticoid-induced GR activation was shown, through interactions with the transcription factor AP1 at the level of a composite glucocorticoid response element named plfG, to activate or inhibit transcription depending on the c-Jun-c-Jun or c-Jun-c-Fos composition of AP1 (28). Further work is needed to address these issues in MTAL cells.
Thus the present results establish that in vivo glucocorticoid administration as well as in vitro glucocorticoid application to freshly harvested MTALs in media containing AVP stimulates rBSC1 expression in rat MTAL, as manifested by an increase in rBSC1 mRNA, protein, and transport activity, which required interactions with cAMP-dependent pathways. Because the Na ϩ -K ϩ (NH 4 ϩ )-2Cl Ϫ cotransporter is a major MTAL apical NaCl and NH 4 ϩ carrier, these observations contribute to explain the role of glucocorticoids in the ability of the kidney to concentrate the urine (20) and in the adaptive increase in urinary NH 4 ϩ excretion in response to metabolic acidosis (21). With respect to urinary NH 4 ϩ excretion, a glucocorticoid-induced increase in MTAL ammonia absorption would be complementary to the known stimulating effect of glucocorticoids on ammonia production by the proximal tubule (29). It must be emphasized that we have recently established that rBSC1 expression in the MTAL is enhanced during chronic metabolic acidosis (12) and that adrenal glucocorticoid production is known to increase in response to acid loading (30 -32). This suggests that glucocorticoids act in both the proximal tubule and MTAL to increase urinary NH 4 ϩ excretion in response to metabolic acidosis. In addition, we have also shown that in vitro incubation of MTALs in an acid medium enhances rBSC1 mRNA and protein abundance and cotransport activity (12). Thus the direct effects of an acid pH and of glucocorticoids would add to fully explain the stimulation of rBSC1 expression in the MTAL by metabolic acidosis (Ref. 12 and present study). Further work is needed to test this hypothesis. Otherwise, it is worth noting that, if glucocorticoids stimulate rBSC1 expression in the cortical TAL also, sodium chloride absorption without water should be enhanced along the entire TAL, which would contribute to explaining the permissive role of glucocorticoids in the renal elimination of a water load (33). The effects of glucocorticoids on rBSC1 described in the present study may thus explain, at least in part, the well known inability of the kidney to maximally concentrate or dilute the urine during adrenal insufficiency.