INTRODUCTION

Arginine vasopressin (AVP)¹ acts on its target cells through binding to V₁ and V₂ receptors (1). The V₂ receptor-mediated activation of adenylate cyclase increases the intracellular cAMP content, which in turn activates protein kinase A. This mechanism accounts for the rapid effects of AVP on water and sodium transport in the renal cortical collecting duct (CCD). The classical short-term effect of AVP on sodium transport (2-6) consists of translocation of amiloride-sensitive sodium channels from intracellular pools to the apical membrane, promoting an increase in sodium entry (7). The increased sodium entry in turn induces a coordinate rise in the activity of the basolateral Na⁺/K⁺-ATPase (8-10).

Activation of the cAMP-dependent signaling cascade can also lead to the regulation of gene expression through cAMP-responsive transcription factors and DNA-binding proteins. This regulation consists of either a stimulation or a repression of the transcription of target genes (reviewed in Ref. 11). At least three cAMP-responsive DNA-binding proteins have been described: CREB (cAMP response element-binding protein), CREM (cAMP response element modulator), and ATF-1 (activating transcription factor 1). Within the nucleus, these proteins bind to DNA sequences that are typified by a consensus cAMP response element (CRE), thus modulating the transcription of target genes (11). Therefore, we decided to examine whether AVP, in addition to its rapid effects on the activity of the Na⁺ channel and Na⁺/K⁺-ATPase, also induces a long-term increase in transepithelial sodium transport through transcriptional effects. Indeed, it has been shown that after several days of treatment with AVP, the Na⁺/K⁺-ATPase activity of rat CCD increases (8). In A6 cells, used as an amphibian model of mammalian CCD, oxytocin (an analog of AVP) increases the biosynthesis of both ₁ and ₁ subunits of Na⁺/K⁺-ATPase (12).

In this study, evidence is provided that AVP induces an increase in transepithelial sodium transport in the RCCD₁ rat CCD cell line after 5-6 h, which involves an increase in both the steady-state mRNA level and the level of de novo protein synthesis for certain subunits of rENaC and Na⁺/K⁺-ATPase. This effect likely plays a role in the long-term regulation of renal sodium reabsorption and may cooperate with corticosteroid hormones to ensure sodium and water homeostasis.

EXPERIMENTAL PROCEDURES

Experiments were performed on the RCCD₁ rat cortical collecting duct cell line (13) between passages 14 and 45. Cells were grown on filters, glass slides, or Petri dishes coated with collagen (type 1 from rat tail, Institut Jacques Boy, Reims, France). The culture medium contained 1:1 Ham's F-12/Dulbecco's modified Eagle's medium, 14 mM NaHCO₃, 2 mM glutamine, 5 × 10⁸ M dexamethasone, 3 × 10⁸ M sodium selenite, 5 µg/ml transferrin, 5 µg/ml insulin, 10 µg/ml epidermal growth factor, 5 × 10⁸ M triiodothyronine, 10 units/ml penicillin/streptomycin, 20 mM HEPES, pH 7.4, and 2% fetal bovine serum (Life Technologies, Inc., Les Ulis, France) and was changed every other day.

RCCD₁ cells were grown on Snapwell filters (Costar Corp., World Precision Instruments) coated with collagen, and experiments were performed as described previously (13). Briefly, filters were incubated overnight in minimum medium composed of 1:1 Ham's F-12/Dulbecco's modified Eagle's medium, 14 mM NaHCO₃, 2 mM glutamine, 10 units/ml penicillin/streptomycin, and 20 mM HEPES, pH 7.4. They were then mounted onto a voltage clamp system (Costar Corp., WPI). Cells were bathed on each side with 8 ml of minimum medium thermostated at 37 °C and circulated by a gas lift (95% O₂ and 5% CO₂ mixture). This voltage current clamp was used to measure short circuit current I_sc (µA·cm²) by clamping transepithelial potential V_T to 0 mV for 1 s.

Transepithelial ²²Na transport was determined across RCCD₁ cells. Cells were prepared as for electrophysiological studies. That is, RCCD₁ cells were grown on Snapwell filters coated with collagen while high V_T and transepithelial resistance R_T were recorded. They were then incubated overnight in minimum medium before mounting in diffusion chambers. Each side was bathed with 5 ml of minimum medium thermostated at 37 °C and circulated by the gas lift. After a stabilization period (30 min), 5 µCi of ²²Na (4.8 mCi/ml; Amersham Corp.) was added at the apical side of the cells. 5 µl of medium was then collected at different time intervals from the basolateral side. After a control period, cells were either treated or not with 10⁸ M AVP. Radioactivity was counted using a -scintillation counter (Wallac).

RCCD₁ cells were grown on glass slides covered with collagen. Cells were fixed for 15 min in 4% paraformaldehyde with 5 mM MgCl₂ and then kept at 4 °C in 70% ethanol. In situ hybridization was performed as described previously (14, 15). Quantification was performed using an image analyzer (Optilab, Graftek, Grenoble, France) to determine the difference between signals obtained from antisense probe as compared with sense probe. Linearized rat ₁- and ₁-ATPase cDNAs subcloned into either Bluescribe or Bluescript were used to synthesize antisense and sense cRNA probes. The 3-noncoding sequence (which presents little homology to other isoforms) of ₁- and ₁-ATPase cDNAs was used. The ₁-ATPase probe corresponds to nucleotides 3096-3696. The ₁-ATPase probe corresponds to nucleotides 1229-1600. Parts of the 3-untranslated regions of the , , and  subunit cDNAs of rENaC subcloned into the Bluescript vector were used (corresponding to nucleotides 2185-2775 for the Na⁺ channel  subunit, nucleotides 2150-2463 for the Na⁺ channel  subunit, and nucleotides 2470-2911 for the Na⁺ channel  subunit) (16, 17). After linearization, ³⁵S-labeled RNA probes were synthesized using T7, SP6, or T3 polymerase. ³⁵S-Labeled uridine 5-triphosphate (1000 Ci/mmol) was obtained from Amersham Corp., and the other reagents (adenosine, guanosine, and cytosine 5-triphosphates; ribonucleasin; dithiothreitol; and RNA polymerases) were from Promega.

The RNase protection assay was performed directly on RCCD₁ cells grown on Transwell filters (Costar Corp.) and lysed with a guanidium thiocyanate solution (4 M guanidium thiocyanate, 25 mM sodium citrate, pH 7, 0.5% Sarcosyl, and 0.1 M -mercaptoethanol). The protocol was as described previously (16). Signal quantification was performed using an image analyzer. Care was taken to analyze nonsaturated signals. The glyceraldehyde-3-phosphate dehydrogenase mRNA signal used as an internal standard permitted loading differences on the gel to be taken into account. Antisense cRNA probes were synthesized using [³²P]UTP (15 TBq/mmol; Amersham Corp.) with a Promega riboprobe kit. The cDNAs used for ₁- or ₁-ATPase and the Na⁺ channel , , and  subunits were the same as for in situ hybridization. The glyceraldehyde-3-phosphate dehydrogenase probe was synthesized using a cDNA inserted in a Bluescript plasmid (16).

For the Na⁺ channel, polyclonal anti-rENaC , , and  subunit antibodies were raised against a fusion protein in rabbit as described (17). Competition experiments were performed to assess the specificity of the antibodies (see Fig. 4, A-C). For Na⁺/K⁺-ATPase, the polyclonal anti-rat kidney ₁ subunit antibody NK (18) and the polyclonal anti-rat kidney ₁ subunit antibody SpETb1 (19) were used. Antibodies were used at a saturating dilution (1:40). After different elapses of treatment with or without 10⁸ M AVP, cells grown on Transwell filters were labeled with [³⁵S]methionine (37.5 Bq/mmol; Amersham Corp.) for 1 h in both the presence and absence of the hormone, and immunoprecipitation was performed according to Shore and Nelson (20) with some minor modifications. Briefly, cells were scraped off the filters and extracted in ice-cold lysis buffer (1% Triton X-100, 10 mM Tris-HCl, pH 7.5, 20 mM NaCl, 25 mM KCl, 2 mM EDTA, 0.1 mM dithiothreitol, and 0.5 mM phenylmethylsulfonyl fluoride) for 20 min. Proteins were further extracted by adding SDS (2% final concentration) for another 20 min at 37 °C. Cell lysate was diluted with 6 volumes of 0.1% SDS, 1% deoxycholate, 0.5% Triton X-100, 20 mM Tris-HCl, pH 7.5, 120 mM NaCl, 25 mM KCl, 5 mM EDTA, and 0.1 mM dithiothreitol) and frozen until assayed. [¹⁴C]Ovalbumin (0.5 mCi; Amersham Corp.) was added to the samples prior to assay as a standard and was coprecipitated with a polyclonal anti-ovalbumin antibody (1:100; Nordic Immunological Labs). Although not designed to control for ³⁵S labeling or protein isolation, [¹⁴C]ovalbumin incorporation did allow loading differences on the gel to be taken into account. Protein extracts were precleared with a Staphylococcus aureus slurry (Pansorbin, Calbiochem). Antibodies against ovalbumin, the Na⁺ channel, or the Na⁺/K⁺-ATPase subunits were added to the supernatant, and the mixture was incubated overnight at 4 °C with end-over-end rotation. Immunoprecipitates were incubated with protein A-Sepharose CL-4B beads (Pharmacia Biotech Inc.) at 4 °C for 1 h with end-over-end rotation. The beads were then washed. To detect cell-labeled proteins, samples of eluted immunoprecipitates were submitted to polyacrylamide gel electrophoresis (7.5%). Gels were fixed with 10% methanol and 30% acetic acid in water and submitted to intensifying treatment (Intensify A and B, NEN Life Science Products) before vacuum drying. Autoradiography of gels was carried out on Kodak X-Omat AR-5 film, and the signal was quantitated using an image analyzer. Care was taken to analyze nonsaturated signals. The results were normalized to the signal obtained with ovalbumin.

To determine the number of Na⁺ channels present in the apical membrane of the cells, the specific binding of [³H]phenamil, a high affinity ligand of the Na⁺ channel (21, 22), was examined in RCCD₁ cells grown on Transwell filters. After treatment with 10⁸ M AVP for different lengths of time, both total and nonspecific [³H]phenamil binding were determined by replacing the apical medium with 20 µl of minimum medium to which 10⁷ M [³H]phenamil (30 Ci/mmol, 0.74 Ci/liter) or 10⁷ M [³H]phenamil and 10⁵ M amiloride were added. After incubation, cells were washed six times with ice-cold phosphate-buffered saline, and radioactivity was determined. Specific [³H]phenamil binding was the difference between total and nonspecific binding and is expressed as pmol/mg of protein.

Results are expressed as means ± S.E. Statistical analysis was performed using Student's t test for paired or unpaired data according to the experiments. Slopes of regression lines were compared using the Student test for equal tied variances as described by Snedecor and Cochran (23) using the GraphPad Prism program (IBM).

RESULTS

The effect of 10⁸ M AVP on I_sc was determined in RCCD₁ cells. Fig. 1A shows the effect of 10⁸ M AVP on I_sc after different incubation times (paired batches of cells). When cells were kept under control conditions, I_sc remained stable whatever the incubation time (5 min to 24 h). In contrast, 5 min of treatment with 10⁸ M AVP resulted in a large increase in I_sc (~80% increase). After 4 h of treatment with the hormone, no more difference could be observed between control and AVP-treated cells. After 7.5 h of treatment, a significant increase (80%) could be observed in AVP-treated cells compared with control cells. This effect remained significant after 10 h of treatment and was no longer detectable at 24 h.

5

8

[View Larger Version of this Image (28K GIF file)]

Experiments were also performed to test the influence of the Na⁺ channel blocker amiloride on the long-term effect of AVP on I_sc (Fig. 1B). In the absence of AVP, I_sc remained nearly stable for 3 h and then declined. In experiments in which 10⁸ M AVP was added to the basolateral medium, I_sc first increased after 5 min of treatment. A second significant increase in I_sc occurred after 5-6 h, with a progressive increase up to at least 7.5 h. When 10⁵ M amiloride was added to the apical side of the cells 4.5 h after the addition of AVP, the AVP-induced increase in I_sc was prevented. The slopes of the three regression lines corresponding to each condition for the 5-7.5-h period (control: y = 0.25x + 2.06; AVP: y = 0.23x + 0.98; and AVP/amiloride: y = 0.12x + 0.65) were significantly different (p < 0.001, AVP versus AVP/amiloride and AVP versus control). This argues that sodium transport is involved in the long-term effect of AVP. In addition, the fact that 10⁵ M amiloride did not completely block the delayed increase in I_sc induced by the hormone suggests that other transport pathways might be involved in this long-term effect of AVP.

To confirm the influence of sodium transport on the long-term effect of AVP, the effect of the hormone on net transepithelial ²²Na transport was examined (Fig. 1C). In control experiments, a small and gradual increase in cumulative sodium transport could be observed (n = 20; r = 0.97; y = 26.98x + 53.26). In AVP experiments, ²²Na gradually appeared in the basolateral medium before the addition of the hormone (n = 4; r = 0.85; y = 51.00x + 17.50). The slope of the regression line corresponding to this period was not statistically different from that of the control. Once AVP was added, an initial increase in the rate of ²²Na accumulation was observed (n = 11; r = 0.96; y = 77.46x + 82.42), followed by a further increase ~5.5 h after the addition of the hormone (n = 7; r = 0.95; y = 193.71x  524.83). The slopes of the regression lines corresponding to these two periods were statistically different from those of the control (p < 0.001, AVP versus control for the two periods).

Fig. 2 shows the effect of 1 µM actinomycin D or 2 µM cycloheximide on the AVP-induced increase in I_sc. The inhibitors were added in both the apical and basolateral media 30 min before the addition of 10⁸ M AVP. The results show that the long-term effect (7.5 h) of AVP was prevented by preincubating cells with actinomycin D. It should be noted that cycloheximide, administered for 7.5 h, completely abolished I_sc in the presence as well as absence of AVP, attesting to its drastic effect on protein synthesis.

8

[View Larger Version of this Image (16K GIF file)]

Effect of AVP on mRNAs Encoding the , , and  Subunits of the Na⁺ Channel and the ₁ and ₁ Subunits of Na⁺/K⁺-ATPase

The effect of 10⁸ M AVP on mRNAs encoding the different subunits of both the sodium channel and Na⁺/K⁺-ATPase was tested either by the RNase protection assay (Fig. 3) or by in situ hybridization (Table I). An example of the RNase protection assay carried out with the ₁ subunit of Na⁺/K⁺-ATPase is shown in Fig. 3A. A significant increase in ₁ subunit mRNA could be observed as early as 1 h after AVP addition. Then the mRNA level remained stable up to 24 h. No modification of ₁ subunit mRNA could be observed (Fig. 3C). Concerning the Na⁺ channel, the steady-state level of rENaC  subunit mRNA was not modified by 10⁸ M AVP (1-24 h) (Fig. 3B). The level of mRNA encoding the  subunit of rENaC was increased as early as 1 h after the addition of 10⁸ M AVP, and statistical significance was reached after 3 h. Then the mRNA level remained stable up to 24 h. Although different attempts were made to detect the mRNA encoding the  subunit of rENaC (in particular by using different probes), no signal was observed in these experiments. However, in situ hybridization experiments (Table I) showed that a 24-h treatment with 10⁸ M AVP resulted in a significant increase in the amount of mRNA encoding the  subunit of rENaC. It is noteworthy that the increases in the amount of mRNA encoding the  subunit of rENaC or the ₁ subunit of Na⁺/K⁺-ATPase observed when cells were grown on glass slides and the mRNAs were detected by in situ hybridization (Table I) were larger than when cells were grown on porous substrate and the mRNAs were detected by the RNase protection assay (Fig. 3). The difference might be due either to technical conditions or, more probably, to culture conditions.

8

[View Larger Version of this Image (33K GIF file)]

Table I. Effect of AVP on the amount of mRNA encoding the different subunits of the Na+ channel and Na+/K+-ATPase: influence of actinomycin D and amiloride The amounts of mRNA that encode the , , and  subunits of the Na+ channel and the 1 and 1 subunits of Na+/K+-ATPase were examined using in situ hybridization in RCCD1 cells grown on glass slides and treated (AVP) or not treated (control) with 108 M AVP for 24 h. The effect of 1 µM actinomycin D and 105 M amiloride on the AVP-induced increase in the level of mRNA encoding the Na+ channel  and  subunits and the Na+/K+ ATPase 1 subunit was examined. Cells were treated for 24 h with AVP, actinomycin D + AVP, or amiloride + AVP, or were not treated (control). Each point is the mean value of 7-14 determinations. Subunit Control AVP Control AVP Act. Da + AVP Control AVP Ami + AVP Na+/K+-ATPase 1 8.4  ± 0.9 12.9  ± 1.0b 17.1  ± 1.6 23.6  ± 1.6c 14.9  ± 1.7 24.5  ± 1.3 30.2  ± 2.0c 30.1  ± 2.1c Na+/K+-ATPase 1 4.5  ± 1.4 4.7  ± 0.8 ND ND Na+ channel 14.3  ± 0.5 13.3  ± 0.9 ND ND Na+ channel 1.1  ± 0.1 2.2  ± 0.1b 0.4  ± 0.1 1.5  ± 0.1d 1.0  ± 0.2c 0.5  ± 0.1 1.4  ± 0.3b 1.2  ± 0.1d Na+ channel 1.0  ± 0.3 2.7  ± 0.6c 0.5  ± 0.1 1.5  ± 0.4c 0.6  ± 0.1 2.0  ± 0.3 4.7  ± 0.5d 5.6  ± 0.2d a Act. D, actinomycin D; Ami, amiloride; ND, not determined. b,c,d p < 0.01, p < 0.05, and p < 0.001, experimental (AVP, actinomycin D + AVP, and amiloride + AVP) versus control, respectively.

To test whether the AVP-induced increase in the mRNAs encoding the Na⁺/K⁺-ATPase ₁ subunit and the rENaC  and  subunits was due to an increased transcription rate rather than to a modification of the stability of mRNAs, the effect of 1 µM actinomycin D was examined. The results are given in Table I. The addition of 1 µM actinomycin D blocked the effect of AVP on mRNA accumulation for both the ₁ subunit of Na⁺/K⁺-ATPase and the  and  subunits of the Na⁺ channel. This effect might thus correspond to a transcriptional effect of AVP rather than to an effect on mRNA stability. Indeed, actinomycin D blocks the transcription of DNA and the synthesis of mRNA by binding to the guanosine residues, but does not markedly affect post-transcriptional steps that lead to mRNA degradation (24). In contrast, the presence of 10⁵ M amiloride did not prevent the AVP-induced increase in accumulation of mRNA encoding the rENaC  and  subunits or the Na⁺/K⁺-ATPase ₁ subunit, suggesting a Na⁺-independent increase in the rate of transcription.

The rates of synthesis of the , , and  subunits of rENaC and the ₁ and ₁ subunits of Na⁺/K⁺-ATPase were determined by immunoprecipitation using specific antibodies for each subunit. For the epithelial sodium channel, the specificity of the anti-, anti-, and anti- subunit antibodies was tested in preliminary experiments. The results are given in Fig. 4 (A-C). A signal was obtained with the immune antiserum (arrows), whereas no signal could be observed with the preimmune serum. In addition, for each antibody, the signal was displaced only by the fusion protein used to generate the antibody. Fig. 4D shows an example of immunoprecipitation carried out with the anti- subunit antibody in RCCD₁ cells, and Fig. 4 (E and F) summarizes the results. No modification in the rate of synthesis of the rENaC  subunit was observed at 1, 3, 6, or 24 h treatment with 10⁸ M AVP (Fig. 4E). On the contrary, a significant increase in the rate of synthesis of the rENaC  subunit was observed at 3 h, with no further increase at 1, 6, or 24 h (Fig. 4E). For the rENaC  subunit, a significant increase was observed after 1 h of treatment; thereafter, no difference could be observed between control and AVP-treated cells (Fig. 4F). Concerning Na⁺/K⁺-ATPase, no modification in the rate of synthesis of the ₁ subunit was observed (Fig. 5D). In contrast, a significant increase in the rate of synthesis of the ₁ subunit was present as early as 1 h after AVP treatment; thereafter, the rate of synthesis returned to control values (Fig. 5, C and D).

8

[View Larger Version of this Image (34K GIF file)]

8

[View Larger Version of this Image (38K GIF file)]

The effect of 10⁸ M AVP on the number of Na⁺ channels present in the apical membrane of the cells was determined by specific [³H]phenamil binding studies. Results are given in Fig. 6. 10 min after the addition of the hormone, a small but not statistically significant increase in specific [³H]phenamil binding was observed. 4 h after AVP addition, no difference in specific [³H]phenamil binding between control and AVP-treated cells was detected. In contrast, after 7.5 h, AVP induced a significant increase (~60%) in specific [³H]phenamil binding compared with the control.

8

[View Larger Version of this Image (16K GIF file)]

DISCUSSION

AVP, which mainly modulates water permeability in the kidney, also exerts short-term effects on renal Na⁺ transport (4-6). However, very limited information is available on the putative long-term effects of this hormone. The cAMP pathway has been shown to induce transcriptional effects in different systems, besides its short-term action as a second messenger in several cell functions. In particular, the protein kinase A-induced phosphorylation of nuclear proteins, such as CREB or CREM, could be responsible for transcriptional effects via CREs present in gene promoters (11).

In this study, we have evaluated the possibility of long-term regulation of sodium transport by AVP in a renal cell line with properties of the cortical collecting duct (13). AVP was used at 10⁸ M. Indeed, we have previously shown that it produces a maximal short-term response on short-circuit current in RCCD₁ cells (13). However, this high concentration stands in marked contrast to the normal physiological range of AVP concentration (1-100 pM) that produces a maximal physiological effect both in vivo and in isolated CCD segments perfused in vitro. The reason for the difference in sensibility between in vivo tissue and cultured cells remains to be determined. In addition, in view of the high concentration used, one cannot exclude possible binding of AVP to other related receptors, such as oxytocin receptors. In tight epithelia, such as the renal collecting duct, two main transporters are involved in transepithelial sodium reabsorption: the apical amiloride-sensitive epithelial sodium channel and the basolateral Na⁺/K⁺-ATPase (25). The sodium channel is composed of several subunits, three of which (, , and ) have been recently cloned (26-28). Na⁺/K⁺-ATPase is formed by two subunits ( and ). According to experimental evidence, only the ₁ and ₁ isoforms are expressed in rat CCD (15, 29, 30).

We have shown that AVP augments the synthesis of certain subunits of both transporters at the mRNA and protein levels, resulting in a delayed stimulation of transepithelial sodium transport. This is likely to constitute a novel regulatory pathway of AVP in transporting epithelia. It appears that the different subunits of rENaC and Na⁺/K⁺-ATPase are differentially affected by AVP treatment. The  and  (but not ) subunits of rENaC and the ₁ (but not ₁) isoform subunit of Na⁺/K⁺-ATPase are up-regulated at both the mRNA and protein levels. In view of the inhibitory action of actinomycin D on mRNA accumulation and the Na⁺ transport process, this is likely to reflect an increase in the transcription rate of these subunits. The AVP-induced increase in both the transcription and translation rates occurs within 1-3 h. Interestingly, de novo protein synthesis was not observed at 6 or 24 h of hormonal exposure, despite sustained high levels of the corresponding mRNAs. This fact could be accounted for by the existence of an additional regulatory translation process. Such translational regulation of mRNAs has been reported in other systems (31). It may involve the interaction of regulatory proteins with the 5- or 3-untranslated regions of the mRNAs or sequestration of mRNAs in messenger ribonucleoprotein particles inaccessible to translation (31). The magnitude of the observed changes in these mRNA levels and in de novo protein synthesis corresponds closely with the resulting effects on transepithelial sodium transport and the number of [³H]phenamil-binding sites. The time course of the increase in the steady-state level of mRNAs and in the level of de novo protein synthesis accords well with the AVP-dependent modulation of sodium transport rate over time.

While such transcriptional effects of AVP on sodium transporters have not heretofore been reported in the literature, other information is available on the transcriptional effects of corticosteroid hormones (which are the main regulators of renal sodium reabsorption) and, in particular, aldosterone. Interestingly, differential effects of aldosterone on the different subunits of the sodium channel (32-35) or of Na⁺/K⁺-ATPase (29, 36) have been reported. The magnitude of the observed aldosterone-related changes is in general relatively small (30-80%), as is the magnitude of the AVP effects reported in this study. Thus, the expression of mRNAs encoding the sodium channel and Na⁺/K⁺-ATPase seems to be under the control of both AVP and aldosterone, i.e. the two main hormones that control sodium reabsorption and homeostasis. Although still a subject of debate, the regulation of the different types of subunits may be hormone-specific.

The mechanism of the transcriptional regulation of sodium transporters by AVP remains to be established. One can hypothesize that regulation implies CREs present in the promoter region of the sodium transporter genes. It has been shown that nuclear proteins, like CREB or CREM, could be phosphorylated by protein kinase A and thus could act as transcription factors after binding to CREs (11, 37). One or two CREs have been described in the promoter region of the gene that encodes the ₁ subunit of Na⁺/K⁺-ATPase (38-40), whereas no CRE has been identified in the ₁ subunit promoter (41). The promoter regions of the different subunits of the epithelial Na⁺ channel have not yet been documented, except for one region of the promoter of the  subunit that has been shown to contain a CRE sequence (42), a finding compatible with our results. In the promoter region of the gene that encodes the ₁ subunit of Na⁺/K⁺-ATPase, a glucocorticoid-responsive element has also been described. This glucocorticoid-responsive element could be responsible for the effects of corticosteroid hormones on the synthesis of Na⁺/K⁺-ATPase. Synthesis of the ₁ subunit of Na⁺/K⁺-ATPase could thus be regulated by both corticosteroids and AVP. A synergistic action of aldosterone and AVP has been described for apical Na⁺ channel (7) or basolateral Na⁺/K⁺-ATPase (9) activities. Whether a joint effect of the two hormones also occurs at the transcriptional level, thereby increasing the biosynthesis of these proteins, remains to be determined.