Regulation of renal proximal tubule Na-K-ATPase by prostaglandins

Abstract

Prostaglandins (PGs) play a number of roles in the kidney, including regulation of salt and water reabsorption. In this report, evidence was obtained for stimulatory effects of PGs on Na-K-ATPase in primary cultures of rabbit renal proximal tubule (RPT) cells. The results of our real-time PCR studies indicate that in primary RPTs the effects of PGE₂, the major renal PG, are mediated by four classes of PGE (EP) receptors. The role of these EP receptors in the regulation of Na-K-ATPase was examined at the transcriptional level. Na-K-ATPase consists of a catalytic α-subunit encoded by the ATP1A1 gene, as well as a β-subunit encoded by the ATP1B1 gene. Transient transfection studies conducted with pHβ1-1141 Luc, a human ATP1B1 promoter/luciferase construct, indicate that both PGE₁ and PGE₂ are stimulatory. The evidence for the involvement of both the cAMP and Ca²⁺ signaling pathways includes the inhibitory effects of the myristolylated PKA inhibitor PKI, the adenylate cyclase (AC) inhibitor SQ22536, and the PKC inhibitors Gö 6976 and Ro-32-0432 on the PGE₁ stimulation. Other effectors that similarly act through cAMP and PKC were also stimulatory to transcription, including norepinephrine and dopamine. In addition to its effects on transcription, a chronic incubation with PGE₁ was observed to result in an increase in Na-K-ATPase mRNA levels as well as an increase in Na-K-ATPase activity. An acute stimulatory effect of PGE₁ on Na-K-ATPase was observed and was associated with an increase in the level of Na-K-ATPase in the basolateral membrane.

prostaglandins (PGs) are involved in mediating the response of the kidney to changes in Na⁺ balance, by regulating renal ion transport, glomerular filtration, renin release, and blood flow (9, 32, 61). The ultimate effect of PGs is on blood pressure (45). The specific effects of PGs on the nephron have been difficult to define by means of renal clearance studies, because PGs are potent vasodilators (61). Thus many of the previous studies with PGs were conducted with isolated, perfused tubules (9). These studies have been concerned with the acute effects of PGs (9).

The results of the studies with isolated, perfused tubules have clearly indicated that PGs do indeed affect transport in different nephron segments, although each nephron segment responds to PGs in a distinctive manner (9). For example, a 15-min incubation with PGE₂ inhibited Na⁺ transport across the rabbit cortical collecting tubule (CCT) while having no effect on the rabbit medullary thick ascending limb (15, 35). A consequence of the inhibition of Na⁺ transport in the CCT was natriuresis. The acute inhibition of Na⁺ reabsorption in the CCT is associated with a reduction in Na-K-ATPase activity (57), unlike the acute stimulatory effect of PGs on distal tubule Na-K-ATPase (57). In contrast to the extensive studies of acute PG effects, studies of chronic PG effects have not been conducted with isolated nephron segments, due to their limited viability.

For this reason, we have studied long-term effects of PGs on growth and transport using kidney epithelial cell culture systems. When studying the Madin-Darby canine kidney (MDCK) cell line, we observed that both PGE₁ and 8-Br-cAMP increased the activity of the Na-K-ATPase (67). The Na-K-ATPase consists of both an α-subunit (with catalytic activity) and a β-subunit, involved in insertion of Na-K-ATPase in the basolateral membrane. The level of both the α- and the β-subunits increased in PGE₁ and 8-Br-cAMP-treated MDCK cells. In addition, the level of the mRNAs for the α1- and the β1-subunit also increased (65). However, the level of β1-subunit mRNA increased to a larger extent than α1-subunit mRNA. Thus in our initial studies, the regulation of transcription of the β1-subunit gene (ATP1B1) was examined in detail, and the signal transduction pathways involved in β-subunit gene transcription were defined (65). Evidence was obtained for the involvement of PKA and PKC.

However, it was unclear whether the signal transduction pathways involved in the regulation of ATP1B1 transcription are unique to MDCK cells, which serve as a distal tubule model, or whether tubule epithelial cells in other nephron segments are subject to similar types of control. The renal proximal tubule (RPT) is of particular interest in these regards because this nephron segment is involved in the response of the kidney to changes in Na⁺ status (3). A number of hormones and other effectors mediate the response of the kidney to such changes, including PGs (21).

To study the effects of PGs on the RPT, we have employed a primary rabbit RPT cell culture system, which closely resembles normal RPTs in the animal (62, 63). The primary RPT cells possess a polarized morphology, as well as transport systems which are involved in polarized solute transport, including an apical Na⁺/glucose cotransport system (SGLT1) (14, 54), a basolateral PAH transport system (OAT1) (14, 30, 36, 72), and a Na⁺/phosphate cotransport system (Npt2) (70). In addition, the cultures respond to parathyroid hormone (PTH) (14), insulin (69), dopamine (31), and angiotensin II (ANG II) (29) in a manner that is characteristic of RPT cells. ANG II stimulates the synthesis of PGs (1), including PGE₂, PGF_2α, and 6-keto-PGF_1α (1).

In this report, effects of PGs on Na- K-ATPase in primary RPT cell cultures are examined. To determine whether the observed increase in the Na-K-ATPase β1-subunit following treatment with PGE₁ can be explained by transcriptional regulation, transient transfection studies were conducted with pH β1-1141 Luc, a human ATP1B1/luciferase construct (22). The involvement of both PKA and PKC in signaling has been examined. In addition, the effects of arachidonic acid (AA) and its metabolites, as well as norepinephrine and dopamine, have been studied because these effectors, like PGs, are produced in response to changes in Na⁺ balance. To evaluate the physiological significance, both the acute and chronic effects of PGs on Na-K-ATPase activity have been examined.

MATERIALS AND METHODS

Materials.

Hormones, human transferrin, PGs, and other chemicals were from Sigma (St. Louis, MO). PKC inhibitors were from Calbiochem (La Jolla, CA). PG receptor agonists and antagonists were from Cayman Chemical (Ann Arbor, MI). DMEM, Ham's F12 medium (F12), soybean trypsin inhibitor, and lipofectamine were from Invitrogen (Carlsbad, CA). Class IV collagenase was from Worthington (Freehold, NJ). Mouse monoclonal antibodies against the Na-K-ATPase α1 (sc-21712)-, β1-subunit (sc-21713) and goat anti-mouse horseradish peroxidase conjugate were from Santa Cruz Biotechnology (Santa Cruz, CA). Immun-Star HRP Substrate, nitrocellulose, acrylamide, and other electrophoresis reagents were from Bio-Rad (Hercules, CA). Streptavidin-Sepharose beads were from Amersham. The Galacto-Star System was obtained from Applied Biosystems (Bedford, MA). Both the p-SV β-galactosidase (pSVβgal; β-gal) plasmid and Reporter Lysis Buffer were from Promega (Madison, WI). The Prism 5 program was obtained from GraphPad Software (San Diego, CA). ⁸⁶Rb⁺ (>1 μCi/μg) was obtained from PerkinElmer (Billerica, MA). Transwells were obtained from Corning (Corning, NY).

Primary RPT cultures.

The basal medium for primary RPT cells consists of a 1:1 mixture of DMEM and Ham's F12 medium (F12), pH 7.4, containing 15 mM HEPES, 20 mM Na⁺ bicarbonate, penicillin (92 U/ml), and kanamycin (0.01%; DMEM/F12) (14). Immediately before use, the medium was further supplemented with bovine insulin (5 μg/ml), human transferrin (5 μg/ml), 50 nM hydrocortisone (Medium RK1), and, if appropriate, other factors (14). Water used for medium and growth factor preparations was purified using a Milli-Q deionization system. Iron oxide in 0.9% NaCl was prepared, sterilized in an autoclave, and diluted with PBS before use (18).

To initiate primary rabbit kidney proximal tubule (RPT) cell cultures (14, 36), a kidney from a male New Zealand White rabbit (2–2.5 kg) was perfused via the renal artery, first with PBS, and subsequently with sterile 0.5% (wt/vol) iron oxide. New Zealand White rabbits (4–5 lb) were euthanized by a protocol using CO₂, which has been received and approved by the Institutional Animal Care and Use Committee of the State University of New York at Buffalo. The renal cortex was sliced, homogenized with a sterile Dounce homogenizer (loose pestle), and the homogenate (consisting of nephron segments) was sequentially passed through a 253- and an 83-μm mesh. The RPTs and the glomeruli on the 83-μm mesh were transferred into a 50-ml conical tube containing DMEM/F12 and a sterile stir bar. The glomeruli (with bound iron oxide) were removed with the stir bar, resulting in a preparation of tubules that stained positively for γ-glutamyltranspeptidase (γGT; Taub M, unpublished observations), a distinctive RPT marker.

Tubules to be utilized for primary cultures were treated with 0.05 mg/ml collagenase in DMEM/F12 containing 0.5 μg/ml soybean trypsin inhibitor for 2 min at 23°C. The tubules were washed twice by centrifugation, resuspended in DMEM/F12, and plated in 35-mm dishes containing Medium RK-1. The medium was changed the day after plating and every 2 days thereafter.

Transient transfection studies.

Primary cultures were cotransfected with pHβ1-1141 Luc, a human ATP1B1 promoter/luciferase construct (1 μg), and the pSVβgal plasmid (0.2 μg) to correct for transfection efficiency. Transfections were carried out utilizing lipofectamine in antibiotic-free DMEM/F12 supplemented with 5 μg/ml insulin and 5 μg/ml transferrin. The next day, the medium was changed to antibiotic-containing DMEM/F12 supplemented with 5 μg/ml insulin and 5 μg/ml transferrin. The cultures were incubated for 2 h in a 5% CO₂-95% air humidified environment and then treated with appropriate effectors for 4 h (unless otherwise indicated). Four cultures were used per condition to maintain accuracy of determinations of ±10–20%. At the end of the incubation, the monolayers were solubilized in Reporter Lysis Buffer and spun for 1 min at 13,600 g in a microcentrifuge. The cell lysates were then assayed for luciferase and β-galactosidase (β-gal) activity.

To measure luciferase activity, an aliquot of the cell lysate was placed in 0.1 ml of a buffer containing 20 mM tricine, 1.07 mM (MgCO₃)·4Mg(OH)₂, 2.67 mM MgSO₄, 0.1 mM EDTA, 33.3 mM dithiothreitol, 270 μM coenzyme A, 470 μM luciferin, and 530 μM ATP. The light units emitted were determined using a Packard Tricarb 4530 scintillation counter with the coincidence circuit turned off. As a control in these studies, luciferase activity was determined as a function of light emitted per unit of luciferase activity using purified firefly luciferase (Sigma). Luciferase determinations were made within light units in which activity was linear as a function of increasing luciferase activity (which was obtained with activity measurements giving no more than 3.1 × 10⁶ cpm). Luciferase determinations were made with 3- to 6-day-old cultures.

To determine β-gal activity, aliquots of cell lysates were incubated in a reaction buffer containing the Galacton-Star substrate (a component of the Galacto-Star System). The light emitted was measured as described above. β-Gal activity was expressed as a fraction of the activity in control transfected cultures. As a control, β-gal activity was determined as a function of increasing levels of purified β-gal activity. β-Gal determinations were after a 90- to 120-min incubation, during which time the activity was constant. β-Gal determinations were conducted within a range, in which the activity was constant as a function of increasing β-gal activity (for activity levels giving measurements of up to 2 × 10⁶ cpm).

Each luciferase determination was normalized with respect to β-gal activity. For each condition a luciferase value was obtained by calculating the average of quadruplicate determinations (±SE). This average value was compared with the control value (obtained with cultures maintained with insulin and transferrin alone). Both a one-way ANOVA and the Newman-Keuls multiple comparison test (using Prism 5 software) were employed to determine the statistical significance of the results. Differences were determined to be significant when P < 0.05. The effects of regulatory molecules on luciferase and β-gal activity were determined in at least two independent culture sets.

Real-time PCR.

Primary RPT cell cultures were treated with appropriate effector molecules. RNA was purified from the cultures using an RNA-4PCR kit (Ambion). Genomic DNA was first removed using TURBO DNAase I (Ambion), and then cDNA was synthesized from the purified RNA using reverse transcriptase. Transcripts were amplified in a Bio-Rad I Cycler using iQ Sybr Green Supermix, as well as 5 μM forward and reverse primers complementary to appropriate templates on the cDNA. After amplification, C_t values were calculated using the Bio-Rad IQ5 program. The relative mRNA levels were quantitated using the Pfaffl method (50). GAPDH mRNA was used as an internal control in the calculations of relative mRNA levels.

The primers were designed by using the Primer-BLAST Primer designing tool on the NCBI website and synthesized by Invitrogen. Included among the cDNAs used for primer design are rabbit EP1 (Accession Number AF043491.1, PTGER1), AGAGGCCCAATCGGTTAAGT, AGCTTCATTTACCCTCGCAA; rabbit EP2 (Accession Number NM_001082659, PTGER2), CTGCAGCATCTGGAAGATCA, CATGGGCTCCTTGAGGATAA; rabbit EP3 (Accession Number NM_001122935.1, PTGER3), TAATAGCTGTTCGCCTGGCT, GACGTTCCTGTTCTCACGCT; rabbit EP4 (Accession Number NM_001082259.1, PTGER4), CACCTACATGAAGGGCCAGT, GGCTGTAGAAGTAGGCGTGG; rabbit ATP1A1 (Accession Number AF235024, ATP1A1_RABIT), TGACTCTCCTGCTCTGAAG, CACAATGGAAGCGAAGTTATC; rabbit ATP1B1 (Accession Number AF204927, ATP1B1_RABIT), ACTGGCAAGCGAGATGAAG, ATGGTGAGGTTGGTGAACTG; and rabbit GAPDH (Accession Number L23961, RABGLY3PHO), GCCCTCAATGACCACTTTGT, TCATGACAAGGTAGGGCTCC.

Preparation of cell lysates from cultures on polystyrene.

To determine the effect of exogenous PG and other effectors on Na-K-ATPase levels, primary RPT monolayers were grown in Medium RK-1 containing 10⁻⁶ M indomethacin to inhibit endogenous PG biosynthesis. The cultures were treated on a daily basis for 4 days with effector molecules, also in the presence of indomethacin. Cell lysates were prepared by first washing the RPTs twice with ice-cold PBS at 4°C and then solubilizing the monolayers in lysis buffer (20 mM Tris, pH 7.5, 1% Triton X-100, 120 mM NaCl, 1 mM EDTA, 1 mM PMFS, 1 μg/ml aprotinin, 1 μg/ml leupeptin, 1 μg/ml pepstatin, 1 mM Na⁺ orthovanadate, and 1 mM NaF). The cell lysates were transferred from the culture dishes into microfuge tubes, using a rubber policeman.

Cell surface biotinylation studies.

Rabbit kidney proximal tubules were plated on Transwell inserts (24-mm, 0.4-μm polycarbonate membranes, no. 3412, Corning) in Medium RK1 at twice the density used for cultures on polystyrene. The medium was changed initially after days 3 and 7 in culture and daily thereafter. Cell surface biotinylation studies were conducted after 10–12 days.

To conduct cell surface biotinylation studies, the medium was changed in Transwell cultures. Two hours later, the monolayers were incubated for 30 min in the presence or absence of effector molecule (PG). The monolayers were then placed on ice, washed, first with ice-cold DMEM/F12 and then twice with PBS supplemented with 0.1 mM CaCl₂ and 1.0 mM MgCl₂ (PBS-CaMg). The cultures were subsequently incubated at 4°C for 20 min. with NHS-ss-biotin (0.5 mg/ml) in biotinylation buffer (10 mM triethanolamine, pH 9.0, 2 mM CaCl₂, and 150 mM NaCl) with gentle horizontal motion. This was followed by a second 20-min incubation in biotinylation buffer, two washes with ice-cold PBS-Ca-Mg containing 100 mM glycine, and a final 20-min incubation at 4°C in PBS-Ca-Mg containing 100 mM glycine. The monolayers were then quickly washed with PBS-Ca-Mg. The filters were transferred to 35-mm dishes, and the monolayers were solubilized in 1 ml of lysis buffer (1% Triton X-100, 150 mM NaCl, 5 mM EDTA, 50 mM Tris, pH 7.5) for 1 h at 4°C. Cell lysates were removed from the filters with a rubber policeman and centrifuged at 14,000 g for 10 min at 4°C.

Aliquots of the cell lysates (100–250 μg of protein) were used for the purification of biotinylated cell surface proteins and for protein determinations (Bradford method). Cell lysates were incubated overnight with streptavidin-Sepharose beads. The beads were washed three times with lysis buffer, twice with a high-salt wash buffer (lysis buffer containing 0.1% Triton X-100 and 500 mM NaCl), and once with 10 mM Tris, pH 7.5. Proteins were eluted by heating 5 min at 100°C in SDS-PAGE sample buffer.

Western blot analysis.

Na-K-ATPase in cell lysates was subjected to Western blot analysis. Samples of the streptavidin precipitates and whole cell lysates (equalized with regard to protein by the Bradford method) were separated by electrophoresis through 7.5% SDS/polyacrylamide gels, along with parallel samples of purified dog Na-K-ATPase and molecular weight markers. The proteins in the gel were transferred to nitrocellulose using a Trans-Blot Apparatus (Bio-Rad). Nitrocellulose blots were blocked for 1 h in Tris-buffered saline (TBS) containing 0.1% (vol/vol) Tween 20 (TTBS), followed by a 2-h incubation in TTBS containing primary antibody (either an anti-Na-K-ATPase α- or β-subunit monoclonal antibody). The blots were washed (5× with TTBS, 5 min/wash) and then incubated for 45 min with a secondary antibody (goat anti-mouse horseradish peroxidase conjugate). After washing (6× with TTBS, 15 min/wash), bands were visualized following a 5-min incubation in Immun-Star HRP Luminol/Enhancer by exposure to Bio Max MS2 Film. Na-K-ATPase α- and β-subunits were identified by their comigration with purified dog Na-K-ATPase and with molecular weight markers.

As controls for equal loading, blots of biotinylated proteins were subsequently incubated for 1 h with mouse anti-biotin-alkaline phosphatase conjugate. Bands were visualized by incubation in a reaction mixture containing 0.33 mg/ml nitroblue tetrazolium (NBT) and 0.17 mg/ml 5-bromo-4-chloro-3-indolyl phosphate (BCIP) in 0.1 M NaCl and 0.05 N MgCl₂. The X-ray films and blots were scanned with a Bio-Rad Scanning Densitometer, and the relative intensity of bands was quantitated using the Quantity One Program.

Na-K-ATPase assays in vesicle preparations.

The Na-K-ATPase activity in cell lysates was measured by the method of Forbush (25) as previously described by Taub et al. (67). To summarize, primary RPT monolayers in 100-mm dishes were washed twice with sonication buffer (140 mM NaCl, 10 mM KCl, 10 mM HEPES, pH 7.4) and removed with a rubber policeman. The cells were disrupted by sonication at 4°C using three 5-s blasts. The sonicated material was diluted fivefold into preincubation buffer (10 mg/ml BSA, 2 mM EDTA, 25 mM HEPES, pH 7.0 and 0.6 mg/ml SDS) and incubated for 10 min at 23°C. The samples were then diluted fivefold using 3 mg/ml BSA in 25 mM HEPES, pH 7.0 (to reduce the SDS concentration) and maintained at 4°C. Aliquots of this diluted material were added to an assay mixture containing 120 mM NaCl, 25 mM KCl, 4 mM Na₂ATP, 4 mM MgCl₂, 1 mM EDTA and 60 mM Tris·HCl, pH 7.5, and incubated for 10 min at 37°C. Determinations were in triplicate, both in the absence and in the presence of 1 mM ouabain. The inorganic phosphate released from ATP was assayed using a modification of the Baginski method (5, 25). In each cell culture condition, the specific Na-K-ATPase activity was determined by calculating the difference in inorganic phosphate released in the presence and absence of 1 mM ouabain. Determinations were standardized with respect to protein content by the Bradford method (10).

⁸⁶Rb⁺ uptake studies.

To study chronic effects of PGE₁, rubidium (Rb⁺) uptake studies were conducted with confluent monolayers of primary RPT cells on 35-mm dishes, as previously described for MDCK cells (67). To summarize, primary RPT cells were grown to confluence in Medium RK1. The cultures were then maintained in experimental medium (DMEM/F12 supplemented with 5 μg/ml insulin, 5 μg/ml transferrin, and other appropriate supplements). The medium was changed on a daily basis. On the day of the uptake experiment, the medium was once again changed, preincubating cells in either the presence or absence of 1 mM ouabain for 4 h at 37°C in a 5% CO₂-95% air humidified environment. Immediately before the uptake period, the monolayers were washed twice at 23°C with 2 ml of uptake buffer (10 mM Tris, pH 7.4, and 140 mM NaCl). The monolayers were then incubated for 5 min at 23°C in uptake buffer containing 1 mM RbCl, and 6 μCi/ml ⁸⁶Rb⁺. Ouabain (1 mM) was also present in the uptake buffer of cultures that had been preincubated with ouabain. At the end of the uptake period, the cultures were rapidly washed three times with ice-cold wash buffer (10 mM Tris, pH 7.4, and 100 mM MgCl₂), and lysed in 0.2 N NaOH. The cell lysates were neutralized with HCl. One half of the cell lysate was used for scintillation counting, and the remainder was used for protein determinations, following the Bradford Method (10).

The labeled Rb⁺ uptake was determined in each of the cultures using a scintillation counter. The quantity of Rb⁺ uptake (in nmol) was determined from the determination of the specific activity of the labeled uptake buffer. Each Rb⁺ uptake determination was standardized with regard to protein. All uptake determinations were in triplicate and corrected for zero time uptake. Values are averages ± SE.

To study acute effects of PGE₁ on Na-K-ATPase activity, ouabain-sensitive ⁸⁶Rb⁺ uptake was determined by a modification of the method of Feraille et al. (20). To summarize, monolayers of primary RPTs were washed twice with K⁺-free DMEM/F12 buffered with 20 mM HEPES, pH 7.4, but lacking bicarbonate. The cells were then preincubated for 30 min at 23°C in the same medium supplemented with 5 μg/ml insulin, 5 μg/ml transferrin, and 1 μM indomethacin, in either the presence or absence of 1 mM ouabain. Subsequently, agonists were added, followed by a second 30-min incubation period. ⁸⁶RbCl was then added to each culture (6 μCi/dish). After a 15-min incubation, the uptake was stopped by cooling on ice, removal of the culture medium, and three washes with ice-cold wash buffer (150 mM choline chloride, 1.2 mM MgSO₄, 1.2 mM CaCl₂, 2 mM BaCl₂, and 5 mM HEPES, pH 7.4). The cultures were lysed in 1% (wt/vol) Na⁺ deoxycholate. Radioactivity in the lysate was measured in a scintillation counter, and protein by the Bradford method, as described above. All uptake determinations were conducted in triplicate, and corrected for zero time uptake. To determine the ouabain-sensitive component of Rb⁺ uptake in each condition, ouabain-insensitive Rb⁺ uptake was subtracted from total Rb⁺ uptake (in the absence of ouabain) Values are averages ± SE.

Statistics.

Values obtained in the results of transient transfection studies, Rb⁺ uptake studies, and Na-K-ATPase assays are averages ± SE of replicates in the same experimental condition. Similarly, in real-time PCR, the values obtained for relative mRNA levels are averages of four replicate samples ± SE. In Western blot analyses, the relative intensity of bands on blots (determined using the Quantity One program) are averages ± SE of two or replicate bands compared with the intensity of bands in the control condition on the same blot.

RESULTS

Effects of PGs and cAMP on ATP1B1 gene transcription.

To determine whether expression of the Na-K-ATPase β1-subunit gene (ATP1B1) is regulated by PGs in primary RPT cells, transient transfection studies were conducted with pHβ1-1141 Luc (22). Figure 1A shows stimulatory effects of an 18-h incubation with either 1.4 μM PGE₁ (5.1 ± 0.5-fold) or 1 mM 8-Br-cAMP (4.7 ± 0.8-fold) on luciferase gene expression. Figure 1B compares the effects of 0.14 μM PGE₁, PGE₂, and PGF₂α following an 18-h incubation. The results indicate that both PGE₁ and PGE₂ were stimulatory to an equivalent extent (5.1 ± 0.5- and 6.9 ± 1.7-fold, respectively). The stimulatory effects of PGE₁ and PGE₂ were significantly higher than the effect of PGF₂α (3.0 ± 0.2-fold). These results are consistent with the involvement of EP receptors (which have a high affinity for PGE₁ and PGE₂), rather than FP receptors. The concentration dependence of the PGE₂ effect was then examined, because PGE₂ is the major PG synthesized in the kidney (9). Figure 1C shows a maximal effect of PGE₂ was obtained at 0.28 μM.

Fig. 1.

Effects of prostaglandins on transcription. Primary renal proximal tubule (RPT) cells were transfected with pHβ1-1141 Luc and pSVβgal, as described in materials and methods. A: effect of PGE₁ and 8-Br-cAMP. Transfected primary RPT cells were incubated for 18 h in either DMEM/F12 supplemented with 5 μg/ml insulin and 5 μg/ml transferrin (control), control medium with 1.4 μM PGE₁, or control medium with 1 mM 8-Br-cAMP. B: effect of 1.4 μM PGE₁, PGE₂, and PGF_2α. Primary RPT cells were incubated for 18 h either in control medium or in control medium further supplemented with 0.14 μM PGE₁, PGE₂, or PGF_2α. C: effect of PGE₂ concentration. Transfected primary RPT cells were incubated for 18 h with PGE₂ at concentrations ranging from 0 to 1.4 μM. At the end of the 18-h incubation, the cultures were lysed. After the luciferase and β-galactosidase activity was determined in each cell lysate, the luciferase activity was standardized for transfection efficiency. In each condition, the fold-stimulation of luciferase activity was compared with the control. Values are averages ± SE of quadruplicate determinations. *P < 0.05 vs. control.

Expression of EP receptors in primary RPT cell cultures.

PGE₁ and PGE₂ stimulate ATP1B1 transcription through a number of signaling pathways, which are initiated by the interaction of PGE₁ and PGE₂ with specific PGE receptors. In the kidney, PGE₁ and PGE₂ interact with four different classes of G protein-coupled EP receptors (40), including EP2 and EP4, which are coupled to adenylate cyclase (AC) via Gs, EP3, which is coupled to AC via Gi, and EP1, which is coupled to PLC via Gq. Although all four EP receptors are present in the kidney (11), the EP receptors present in the RPT have not been clearly defined. The receptors, which are responsible for mediating PGE₁ and PGE₂ action in the RPT (and thus the possible signaling pathways involved), had not yet been defined. Thus real-time PCR was employed to quantitate the level of expression of the mRNAs for the four EP receptors in primary RPT cells and purified RPTs. The relative level of expression of each mRNA was compared with that in the medulla, a previously studied tissue (13).

Figure 2A shows that a significant level of each of the four EP receptor mRNAs was present in primary RPTs as well as purified proximal tubules. While the level of EP2 and EP3 mRNA was equivalent to that of purified proximal tubules (1.0 ± 0.04- and 1.04 ± 0.22-fold, respectively), the level of EP1 and EP4 mRNA was significantly lower (0.71 ± 0.03- and 0.30 ± 0.03-fold of the level in purified proximal tubules, respectively). In both primary RPTs and purified RPTs, the level of all EP receptors was significantly lower than in the medulla, including in particular EP3 (Fig. 2B).

Fig. 2.

Expression of EP receptors in RPT Cells. A: expression of EP1, EP2, EP3, and EP4 receptor mRNA in primary RPT cells was compared with the levels in purified RPTs as described in materials and methods. B: expression of EP1, EP2, EP3, and EP4 receptor mRNA in primary RPT cells and purified RPTs was compared with the levels in the medulla. Values are averages ± SE of triplicate determinations.

Involvement of Gs-coupled EP receptors.

To determine whether Gs-coupled EP2 and EP4 receptors mediate the PGE response, transient transfection studies were conducted using EP2- and EP4-specific agonists and antagonists over a 4-h incubation. The EP2-specific agonist butaprost, as well as the EP2- and EP4-specific agonists PGE₁-OH and 16,16 dimethyl PGE₂ were used. As shown in Fig. 3A, both butaprost (0.1 μM) and PGE₁-OH (0.5 μM) were stimulatory. Similarly, as shown in Fig. 3B, 16,16 dimethyl PGE₂ (0.25 μM) (16, 17) caused a 6.1 ± 0.6-fold stimulation (equivalent to that of PGE₁ and PGE₂). While the results with butaprost clearly indicate that EP2 receptors can mediate a stimulatory effect on ATP1B1 transcription, the results with PGE₂-OH and 16,16-dimethyl PGE₂ could also be explained by EP4 receptor-mediated events.

Fig. 3.

Effects of EP2 and EP4 receptor agonists and antagonists on transcription. A: effects of butaprost and PGE₁-OH. B: effects of 0.28 μM PGE₁, PGE₂, and 16–16 dimethyl PGE₂. C: effect of L161, 982 and MF191. D: effect of PKI and SQ22536. Values are the averages ± SE of quadruplicate determinations. *P < 0.05.

To determine whether EP4 receptors are indeed involved, two highly specific EP4 antagonists, L161, 982 and MF191, were employed at concentrations known to specifically inhibit PGE action via EP4 receptors (42, 44). Figure 3C shows that 0.25 μM L161, 982 and 0.01 μM MF191 both inhibit the PGE₁ stimulation by 81 ± 3 and 83 ± 7%, respectively.

To further evaluate whether Gs-coupled EP2 and EP4 receptors are involved, we examined the effects of PKI (a myristolyated PKA inhibitor) and the AC inhibitor SQ22536 on PGE₁ stimulation. As shown in Fig. 3D, 10 μM PKI and 0.3 mM SQ22536 inhibited PGE₁ stimulation by 39 ± 3 and 48 ± 0.5%, respectively. The stimulatory effect of 8-Br-cAMP (shown in Fig. 1 above) is also consistent with these results.

Involvement of Gq-coupled EP1 receptors.

To determine whether Gq-coupled EP1 receptors are also involved, the effects of several EP1-specific agonists and antagonists were examined. Figure 4A shows that the EP1 agonist 17-phenyl trinor PGE₂ was significantly stimulatory at concentrations ranging from 5 μM (3.1 ± 0.3-fold) to 20 μM (4.4 ± 0.3-fold). In contrast, two EP1 receptor antagonists, SC51089 (1 μM) and AH6809 (1 μM) (16), were significantly inhibitory to PGE₁ stimulation (Fig. 4B). Thus these results suggest that both the activation of PLC and AC are involved in mediating PGE stimulation.

Fig. 4.

Effects of EP1 agonists and antagonists on transcription. A: effect of 17-phenyltrinor PGE₂. B: effect of 1 μM AH 6809 and 1 μM SC51089 either in the presence or absence of 1.4 μM PGE₁. *P < 0.05 vs. control.

Role of Ca²⁺ and PKC.

Consequences of the activation of PLC include an increase in intracellular Ca²⁺ and the activation of PKC. Consistent with the involvement of PKC is the observation (shown in Fig. 5A) that 12-O-tetradecatetradecanoyl phorbol 12-myristate-13 acetate (TPA) was stimulatory between 10⁻⁹ and 10⁻⁷ M. A maximal effect of this PKC activator was obtained at 10 ⁻⁹ M (3.2 ± 0.5-fold control).

Fig. 5.

Effects of 12-O-tetradecatetradecanoyl phorbol 12-myristate-13 acetate (TPA) and PKC inhibitors on transcription. A: effect of TPA. Primary RPT cell cultures were incubated for 4 h with either 0, 10⁻⁹ M, 10⁻⁸ M, or 10⁻⁷ M TPA. B: effect of Gö 6976 on PGE₁ stimulation. Primary RPT cell cultures were incubated for 4 h with either 10⁻⁹ M TPA, 1.4 μM PGE₁, 1.4 μM PGE₁+Gö 6976, 10⁻⁹ M TPA+40 nM Gö 6976, or no further supplement (control). C: effect of Ro-32-0432 on PGE₁ Stimulation. Primary RPT cells were incubated for 4 h with either 1.4 μM PGE₁, 10⁻⁹ M TPA, 25 nM Ro-32-0432+1.4 μM PGE₁, 25 nM Ro-32–0432+10⁻⁹ M TPA, 25 nM Ro-32-0432, or no further supplement (control). *Differences were statistically significant when P < 0.05.

To determine whether PGE₁ stimulation depends upon the activation of PKC, the PKC inhibitor Gö 6976 was employed. As shown in Fig. 5B, 40 nM Gö 6976 reduced PGE₁ stimulation by 71 ± 4% (such that a 1.6 ± 0.2-fold stimulation was obtained with +Gö 6976 and PGE_1, compared with 4.8 ± 0.5-fold with PGE₁ alone). Figure 5B shows that 40 nM Gö 6976 similarly reduced the stimulatory effect of 10 ⁻⁹ M TPA. A 1.8 ± 0.2-fold stimulation was obtained with TPA and Gö 6976, compared with a 2.8 ± 0.3-fold stimulation with TPA and a 1.5 ± 0.1-fold stimulation with Gö 6976. Gö 6976 (at 40 nM) inhibits PKC-α, PKC-β1, and PKC-μ. Ro-32-0432, which inhibits PKC-α and PKC-β1 at 25 nM, similarly reduced TPA stimulation (Fig. 5C). Thus these results support the hypothesis that the PGE₁ stimulation of ATP1B1 gene transcription is dependent upon the activation of PKC-α and/or PKC-β1, a consequence of signaling through EP1 receptors. However, because the inhibition produced by the PKC inhibitors was only partial, the results are also consistent with the involvement of other classes of EP receptors.

EP1 receptor activation has been observed to cause an increase in intracellular Ca²⁺, which contributes to the activation of PKC (17). To determine whether such an increase in intracellular Ca²⁺ is sufficient to affect ATP1B1 gene transcription, the effects of 1) ionomycin, a Ca²⁺ ionophore which increases cytoplasmic Ca²⁺; 2) BAPTA-AM, a Ca ²⁺ chelator which reduces cytoplasmic Ca²⁺; and 3) TMB-8, an agent which inhibits Ca²⁺ efflux from intracellular stores (9), were examined. Figure 6 A shows that 1 μM ionomycin causes a 2.3 ± 0.5-fold stimulation. The stimulatory effect of ionomycin was reduced in the presence of 2.5 × 10 ⁻⁵ M BAPTA-AM (to 1.4 ± 0.2-fold of control). In contrast, as shown in Fig. 6B, 10⁻⁴ M TMB-8 was stimulatory after a 19-h incubation. Thus the effects of agents that perturb intracellular Ca²⁺ on ATP1B1 transcription depend upon their specific underlying mechanisms of action.

Fig. 6.

Effect of agents that modulate Ca²⁺ metabolism on transcription. Primary RPT cell cultures were transiently transfected as described in materials and methods. Subsequently, effects of agents that modulate Ca²⁺ metabolism were examined. A: effect of ionomycin and BAPTA-AM. Primary RPT cell cultures were incubated for 4 h either with DMEM/F12 supplemented with 5 μg/ml insulin and 5 μg/ml transferrin (control) or in the control condition further supplemented with either 1 μM ionomycin, 2.5 × 10⁻⁵ M BAPTA-AM individually, or ionomycin and BAPTA in combination. B: effects of TMB-8 and ionomycin. Primary RPT cell cultures were incubated for 19 h with either 10⁻⁶ M TMB-8 or 1 μM ionomycin. *Significantly different from the control, P < 0.05.

Actions of endogenously produced renal effector molecules.

The RPT produces PGs and other AA metabolites in response to changes in dietary Na⁺ (33, 34). Thus the effect of AA on transcription was examined as a function of AA concentration. An 18-h incubation period was employed, so as to allow for substantial metabolism of AA. Figure 7A shows that AA caused a significant stimulation between 10⁻¹⁰ and 10⁻⁷ M, with a maximal effect at 10⁻⁹ M. To determine whether the stimulatory effect of AA depends upon its metabolism by cyclooxygenase (COX), the COX inhibitor indomethacin was employed. Figure 7B shows that 10⁻⁶ M indomethacin reduced the stimulatory effect of 10⁻⁹ M AA by 83 ± 2% (such that only a 2.7 ± 0.2-fold stimulation was obtained in the presence of indomethacin and AA, compared with a 12 ± 2-fold stimulation obtained with AA alone). Thus the majority of the stimulatory effect obtained by the addition of AA could be attributed to AA metabolites generated by COX.

Fig. 7.

Effects of arachidonic acid (AA) and its metabolites on transcription. A: effect of AA concentration. Primary RPT cells were transiently transfected with pHβ1-1141 Luc and pSVβgal. Subsequently, the transiently transfected primary RPT cells were incubated for 18 h with AA at concentrations ranging from 10⁻¹⁰ to 10⁻⁷ M. B: effect of indomethacin. Primary RPT cells were preincubated for 30 min in either the presence or absence of 10⁻⁶ M indomethacin. Subsequently, the cultures were incubated for 18 h with either 10⁻⁹ M AA, 10⁻⁹ M AA with 10 ⁻⁶ M indomethacin, 10⁻⁶ M indomethacin alone, 1.4 μM PGE₂, or with no further supplement (control). Cultures were lysed, and luciferase determinations were made as described in the legend to Fig. 1. *Differences were statistically significant when P < 0.05.

Na-K-ATPase in the RPT is regulated by 20-HETE and 5,6 EET (34). Figure 8A shows that a 3.4-fold stimulation was obtained in response to 10⁻⁶ M 20-HETE, an AA metabolite generated by cytochrome P-450 ω-hydroxylase. Figure 8B shows that a 2.2-fold inhibition was obtained in response to 10⁻⁶ M 5,6 EET, an AA metabolite generated by cytochrome P-450 epoxygenase. Thus at least three different pathways of AA metabolism affect ATP1B1 gene transcription in the RPT.

Fig. 8.

Effect of 20-HETE and 5,6-EET on transcription. A: effect of 20-HETE. Primary RPT cells were incubated for 4 h with either 1 μM 20-HETE, 1.4 μM PGE₁, 1 mM 8-Br-cAMP, or control (no further supplement). B: effect of 5,6-EET. Primary RPT cells were treated with either 1 μM 5,6 EET, 0.070 μM PGE₁, or 0.70 μM PGE₁ for 4 h. Values were significantly higher than control when *P < 0.05 or lower than the control when **P < 0.05 vs. AA.

The kidney produces catecholamines in response to changes in dietary Na⁺, including norepinephrine and dopamine (3, 27). To determine whether norepinephrine and dopamine affect ATP1B1 transcription, transient transfection studies were conducted using both 3- and 8-day-old cultures. Figure 9A shows that 10 ⁻⁶ M norepinephrine was stimulatory in both 3- and 8-day-old cultures (causing a 2.2 ± 0.3- and a 2.1 ± 0.4-fold stimulation, respectively). This observation is consistent with the observed stimulatory effect of norepinephrine on Na-K-ATPase in the RPT. Figure 9A also shows a stimulatory effect of 10 μM dopamine (1.8 ± 0.1-fold in both days 3 and 8 in culture). This latter observation contrasts with the acute inhibitory effect of dopamine on RPT Na-K-ATPase, which results in natriuresis.

Fig. 9.

Effect of norepinephrine and dopamine on transcription. Primary RPT cells were transiently transfected with pHβ1-1141 Luc and pSVβgal, after either 3 or 8 days in culture, as described in materials and methods. Subsequently, the primary cultures were incubated for 4 h in control medium supplemented with 5 μg/ml insulin, 5 μg/ml transferrin, 10⁻⁶ M indomethacin, and either 10⁻⁶ M norepinephrine, 10 μM dopamine, or no further supplement. Cells were lysed, and luciferase activity was determined as described in the legend to Fig. 1. Values are the averages ± SE of quadruplicate determinations. Four independent experiments were conducted.

Chronic and acute effects of PGs on the Na-K-ATPase.

To determine whether PGs affect the level of Na-K-ATPase, primary RPT cells were incubated in either the presence or absence of 1.4 μM PGE₁ for 4 days. Indomethacin was present so as to prevent endogenous PG production. Figure 10, A and B, shows that the level of the α-subunit increased by 4.2 ± 0.6-fold, while the level of the glycosylated and unglycosylated β-subunit increased by 3.3 ± 0.5- and 4.6 ±1.0-fold, respectively. Figure 10 C shows that in parallel cultures, the level of α-and β-subunit mRNA had increased by 2.2 ± 0.34- and 5.1 ± 0.4-fold, respectively.

Fig. 10.

Chronic and acute effects of PGE₁ on the Na-K-ATPase. A: chronic effect of PGE₁ on the level of the Na-K-ATPase in primary RPT cell cultures. Primary RPT cell cultures were grown to confluence in Medium RK-1 in the presence of 10 ⁻⁶ M indomethacin. Subsequently, the cultures were maintained for 4 days in Medium RK-1 containing 10⁻⁶ M indomethacin further supplemented in either the presence or absence of 0.7 μM PGE₁. The medium was changed daily. At the end of the 4-day incubation, the level of the α- and β-subunits of the Na-K-ATPase was determined by Western blot analysis. The level of α- and β-subunit mRNA was determined by real-time PCR. A: blots of the α- and β-subunits as well as β-actin are illustrated. B: results of scanning densitometry of α- and β-blots. Values are averages of duplicate determinations. C: results of determinations of α- and β-subunit mRNA are shown. Values are averages ± SE of triplicate determinations. D: initial rate of Rb⁺ uptake was determined in both the presence and absence of 1 mM ouabain. Before the uptake study, the primary RPT monolayers were maintained for 4 days in DMEM/F12 supplemented with 5 μg/ml insulin, 5 μg/ml transferrin, 10⁻⁶ M indomethacin, and either 0.7 μM PGE₁ or no further supplement (control). The difference between total Rb⁺ uptake, and ouabain-insensitive Rb⁺ uptake (the ouabain-sensitive component) was then calculated. Values are averages ± SE of triplicate determinations. E: activity of the Na-K-ATPase in cell lysates was determined in primary RPTs monolayers which were maintained for 4 days in DMEM/F12 supplemented with 5 μg/ml insulin, 5 μg/ml transferrin, 10⁻⁶ M indomethacin, and either 0.35 μM PGE₁ and 0.125 mM IBMX, 0.125 mM IBMX alone, or no further supplement (control). Values are averages ± SE of triplicate determinations.

To determine whether the PGE-mediated increase in the level of Na-K-ATPase is associated with an increase in Na-K-ATPase activity, Rb⁺ uptake studies were conducted. Primary RPT cell cultures were first incubated for 4 days either in medium containing 250 ng/ml PGE_1. or with no further supplement (control), followed by Rb⁺ uptake studies. As shown in Fig. 10D, the initial rate of Rb⁺ uptake increased by 1.8 ± 2-fold in the presence of PGE₁. Figure 10D also shows that this observed increase in the Rb⁺ uptake rate could be explained by an increase in the ouabain-sensitive component of Rb⁺ uptake (2.0 ± 0.3-fold relative to control).

Because increases in Rb⁺ uptake may be the result of increased intracellular Na⁺, rather than an increase in the level of Na-K-ATPase, Na-K-ATPase activity was also determined in membrane vesicle preparations. In these studies, cultures were treated 4 days with 125 ng/ml PGE₁ and 0.125 mM IBMX, a cAMP phosphodiesterase inhibitor, to further increase intracellular cAMP levels. Figure 10E shows that the Na-K-ATPase activity in membrane preparations derived from cultures treated with PGE₁ and IBMX (30.7 ± 0.5 μmol·mg protein⁻¹·h⁻¹) was significantly higher than in membrane preparations obtained from control cultures maintained in either the presence or absence of IBMX (1.7 ± 0.03- and 2.4 ± 0.04-fold increase, respectively). Thus the observed increase in the rate of Rb⁺ uptake in PGE₁-treated cultures could indeed be explained by an increase in Na-K-ATPase activity.

Acute incubations with effector molecules have been observed to change the level of basolateral Na-K-ATPase without affecting the overall cellular level of Na-K-ATPase (23). To study such acute effects, primary RPT cells were cultured on Transwell membranes. Following a 30-min incubation with either 1) PGE₁, 2) butaprost, or 3) 17-phenyltrinor PGE₂, basolateral proteins were biotinylated and purified using streptavidin-Sepharose. Figure 11 shows the results of a Western blot analysis of the biotinylated proteins. The level of the α-subunit increased by 5.4-fold in cultures incubated with 1.4 μM PGE₁, 3.9-fold in cultures incubated with the EP2 agonist butaprost, and 2.6-fold in cultures incubated with the EP1 agonist 17-phenyltrinor PGE₂.

Fig. 11.

Acute effect of PGE₁ on the α-subunit. A: primary RPT cells on Transwell membranes were cultured in Medium RK-1 containing 1 μM indomethacin. Following a 30-min incubation with either 1.4 μM PGE₁, 50 μM butaprost, or 50 μM 17-phenyltrinor PGE₂, the basolateral surface of the primary RPT cells was biotinylated. Samples were precipitated using streptavidin beads, separated by SDS-PAGE, and transferred to nitrocellulose. The level of the α-subunit of the Na-K-ATPase was determined by Western blot analysis, as described in materials and methods. B: primary RPT cells in 35-mm dishes were cultured in Medium RK-1 containing 1 μM indomethacin. The monolayers were incubated for 30 min at 23°C with either 1.4 μM PGE₁, 0.125 mM IBMX, 1.4 μM PGE₁+0.125 mM IBMX, or no further supplement (control), followed by a 15-min ⁸⁶Rb⁺ uptake period, as described in materials and methods. Values are averages of triplicate determinations ± SE.

To determine whether the acute effects of PGE₁ on basolateral Na-K-ATPase result in a change in Na-K-ATPase activity, ⁸⁶Rb⁺ uptake studies were conducted. Figure 11B shows that following a 30-min incubation with 1.4 μM PGE₁, ouabain-sensitive Rb⁺ uptake increased 43 ± 2%. A similar increase was observed with 0.125 mM IBMX. However, in the presence of both 1.4 μM PGE₁ and 0.125 mM IBMX the rate of ouabain-sensitive Rb⁺ uptake decreased to 24 ± 16% of the control level. These results indicate that PGE₁ does indeed cause an acute increase in Na-K-ATPase activity in primary RPTs in addition to chronic regulation at the transcriptional level. However, the observed decrease in ouabain-sensitive Rb⁺ uptake in the presence of IBMX as well as PGE₁, suggests that acute regulation in response to PGE₁ is dependent upon the cells' cAMP status.

DISCUSSION

PGE₂ is the major metabolite of COX in all sections of the nephron, including the RPT (9). PGs produced by RPTs, as with all the nephron segments, act as autocrine factors that regulate transport and metabolism in the same cells that produce these autocoids (13). In this report, the effects of endogenous PGs have been studied in a primary proximal tubule (RPT) cell culture system that produces PGs in vitro in a manner similar to intact RPTs which are present in vivo (1). RPTs in vivo produce PGs, of which >90% is PGE₂, followed by PGF_2α, and 6-keto-PGF_1α (9). Similarly, PGE₂ is produced to the largest extent by the primary RPT cells (5 ng·mg protein⁻¹·h⁻¹), followed by PGF_2α (1 ng/h) and 6-keto-PGF_1α (0.6 ng/h) (1). Thus the primary cultures produce enough PGE₂ within 5 h to stimulate Na-K-ATPase β-subunit gene expression sevenfold, unlike the case with PGF_2α. Moreover, PGE₂ has been observed to play a significant role in regulating renal Na⁺ reabsorption in the kidney (9, 12, 32, 59). PGE₁ has been found to act in a virtually identical manner to PGE₂, interacting with the same set of receptors (17). For this reason, the effects of PGE₁ as well as PGE₂ have been studied.

Previously, G protein-coupled receptors have been identified in the kidney, which are specific for PGs (12). PGE₁ and PGE₂ in particular interact with a specific subset of these G protein-coupled receptors, the EP receptors (12). Four subtypes of EP receptors have been identified (11). EP1 receptors are primarily in the collecting duct, where they inhibit Na⁺ and water reabsorption via a Ca²⁺-coupled mechanism. Despite the relatively low levels of renal EP2 receptors, EP2 receptor knockout mice have salt-sensitive hypertension. In contrast, EP3 receptor knockout mice are relatively unaffected in their renal concentrating ability. This finding is surprising in light of the proposed involvement of EP3 receptors in mediating the inhibitory effects of PGs on vasopressin-stimulated salt and water absorption. Gs-coupled EP4 receptors are expressed primarily in the glomerulus (12). In addition, EP4 receptors have been proposed to be present at lower levels in individual nephron segments, including the proximal tubule (12).

Investigations with intact animals indicate that PGs have profound effects on renal function. However, subsequent to these studies, experiments were conducted with isolated, perfused tubules, because a number of the observed effects on reabsorption in intact animals may be the indirect consequence of PG effects on renal blood flow (9, 24). However, only acute effects were studied with isolated, perfused tubules, and precise molecular mechanisms still need to be defined. In vitro cell culture systems are more amenable to the study of molecular mechanisms than other model systems.

Previously, we studied the chronic effects of PGs on the Na-K-ATPase in MDCK cells, a model of the distal tubule (65). Our results indicate that PGE₁ and PGE₂ stimulate Na-K-ATPase activity in MDCK cells, which could be explained at least in part by transcriptional control (65, 67). However, it was unclear whether PGs have similar effects on cells in other nephron segments. Moreover, the physiological significance of these studies was unclear, because the MDCK cell line may have changed during its establishment in culture.

Thus in this report we examine the effects of PGs on transcription of the ATP1B1 gene in primary RPT cells, which have just been removed from the animal. Four major results obtained from these studies include the observations that 1) both primary RPT cell cultures and purified RPTs express EP1, EP2, EP3, and EP4 receptors, albeit to a lower extent that in the medulla. 2) Gq-coupled EP1 receptors as well as Gs-coupled EP2 and EP4 receptors mediate the stimulatory effects of PGE₁ and PGE₂ on ATP1B1 transcription in primary RPTs. These results are based upon the observed inhibitory effects of the EP1 antagonists SC51089 and AH6809, as well as the EP4 antagonists L161, 982 and MF191, and in addition, stimulatory effects of the EP1 agonist 17-phenyl trinor PGE₂, the EP2 agonist butaprost, and the EP2/EP4 agonists PGE₁-OH and 16,16 dimethyl PGE₂. 3) Other renal effector molecules known to regulate transport in the RPT stimulate ATP1B1 transcription, including norepinephrine and dopamine. 4) Chronic treatments with PGE₁ and PGE₂ result in an increase in the levels of the mRNAs for the Na-K-ATPase, as well as in the total level of cellular Na-K-ATPase. Acute treatments result in an increase in the level of Na-K-ATPase in the basolateral membrane without an increase in the overall level of cellular Na-K-ATPase. Both chronic and acute treatment result in an increase in N-,K-ATPase activity.

The evidence that PGE₁ and PGE₂ affect primary RPTs via mechanisms similar to those reported previously in MDCK cells includes our results with the EP1 antagonists SC51089 and AH6809 (17, 37, 40) and the EP2/EP4 agonist 16,16 dimethyl PGE₂ (16). In addition, the involvement of Gs-coupled EP2/EP4 receptors was indicated by the ability of PKAI to inhibit the PGE₁ stimulation, as previously described in MDCK cells (65). Unlike EP2 and EP4 receptors, Gq-coupled EP1 receptors are coupled to PLC. A consequence of the activation of PLC (16) is the generation of diacylglycerol and inositol 1,4,5-trisphosphate, an elevation in intracellular Ca²⁺, and the activation of PKC. The stimulatory effects of ionomycin and TMB-8 substantiated the involvement of intracellular Ca²⁺. TMB-8 causes a decrease in cytoplasmic Ca²⁺. However, at the same time TMB-8 (like ionomycin) increases the level of Ca²⁺ in intracellular stores, which could explain its stimulatory effects. The involvement of PKC was supported by the inhibitory effects of Gö 6976 (27, 51) and Ro-32-0432 (8, 71), also as previously described in MDCK cells (65), a distal tubule model. Thus PGE₁ and PGE₂ apparently regulate ATP1B1 transcription by similar cAMP- as well as Ca²⁺-mediated mechanisms in primary RPT cells as well as MDCK cells.

A number of lines of evidence indicate that similar PGE-mediated effects on transcription of Na-K-ATPase also occur in the RPT in vivo. PGE₂ infusion into the rat renal interstitium has been reported to enhance Na⁺ reabsorption by superficial proximal convoluted tubules (39). PGE₂ is produced in the kidney in response to low doses of ANG II (49) and thus has been proposed to elicit effects on Na⁺ reabsorption which are similar to those elicited by low doses of ANG II. Included among the mechanisms by which ANG II increases Na⁺ reabsorption in the proximal convoluted tubules is through its stimulatory effects on Na-K-ATPase (3). Such stimulatory effects of ANG II on Na-K-ATPase in the RPT contribute to the Na⁺ retention that occurs in rats with experimental heart failure (68). In contrast, the distal tubule exhibits decreased Na-K-ATPase activity in this condition (68). Thus the PGE-mediated effects on transcription of Na-K-ATPase in the distal tubule are elicited under a distinct set of physiological conditions in vivo.

In addition to its role as a mediator in ANG II action, PGE₂ is also involved in reversing the inhibitory effects of PTH on solute reabsorption in the rabbit proximal straight tubule (19). Included among the effects of PTH on the RPT is the inhibition of the Na-K-ATPase (38) as well as the Na⁺-P_i cotransport system (7) and the Na⁺/H⁺ antiport system (47). Although effector molecules like PTH also regulate Na-K-ATPase in the distal nephron, differences in mechanisms have been reported (55).

Our results indicate that the responses of primary RPTs and MDCK cells to PGE₁ and PGE₂ are mediated by signal transduction pathways that are similar, albeit not identical. The EP4 antagonists L161, 982 and MF191 inhibited the stimulatory effect of PGE₁ in primary RPT cells, unlike the case with MDCK cells (unpublished observations, Taub M). Similarly, Sauvant et al. (56) obtained evidence for the involvement of EP4 receptors in mediating the stimulatory effects of PGE₂ on basolateral organic anion transport in opossum kidney cells, another RPT model system. The investigations of Lopes et al. (43) provide additional evidence for a role of PGs in transport regulation in the RPT. The studies of these investigators indicate that the acute inhibitory effect of Ang-(1–7) on Na⁺-ATPase in the RPT is mediated by endogenous PGE₂ (43), Finally, the investigations of Jensen et al. (35) indicate that phosphate transport by mouse proximal convoluted tubules is inhibited by PGE₂.

PGE₂ has additional effects on transport in other nephron segments. In the collecting duct, activation of EP1 by 17-phenyl-trinor PGE₂ has been observed to cause inhibition of Na⁺ and water reabsorption, while activation of EP3 by MB-28767 had no effect (11, 12). However, EP3 receptors have been localized to the medullary thick ascending limb, where they may exert their effects, although EP2 receptor mRNA has been detected along the nephron, the cAMP-mediated effects of PGE₂ on transport in the CD are not mimicked by the EP2 agonist butaprost and have instead been attributed to EP4 receptor activation (12). EP2 receptors have a clearly defined role as vasodilators, which may explain the hypertension which occurs in EP2 knockout mice (12). In contrast, EP4 receptors have been proposed to activate salt and water absorption along the nephron (11).

Because of these other effects of PGE₂ in other regions of the nephron, it is important that our primary RPTs do indeed consist of a population of purified proximal tubule cells. Included among our previous results which support the contention that the primary RPTs consist of a highly enriched population of differentiated proximal tubule cells are our transport studies. The initial rate of p-AH uptake by basolateral membrane vesicles obtained from the primary cultures and purified rabbit kidney proximal tubules was equivalent (72). Similarly, the initial rate of [¹⁴C]-labeled α-methyl-d-glucopyranoside (α-MG) uptake in the primary RPT cell cultures is comparable to that observed in rabbit kidney cortical slices (6, 14). The kinetic properties of transport by the Na⁺/glucose cotransport system (SGLT1) and the Na⁺-P_i cotransport system (Npt2) similarly resemble those defined for SGLT1 and Npt2, respectively, in LLC-PK₁ cell cultures (52, 70). The observation that the level of SGLT1 remained constant after passaging (64), and the expansion of clonal isolates of RPTs following their immortalization with SV40 early region genes (66), also supports the contention that the primary cultures consist of a highly purified population of proximal tubule cells.

In addition, our results with apical membrane proteins indicate that the primary RPTs are highly enriched with proximal tubule cells. The primary RPTs express γ-GT, alkaline phosphates, and leucine aminopeptidase. Over 90% of the primary RPTs express γ-GT, as do the purified proximal tubule preparations, In addition, all five of our immortalized RPT clones, like the primary RPTs, express these apical membrane proteins (14).

Finally, the hormone responsiveness of the primary cultures is consistent with the contention that these cultures consist of a highly purified population of RPTs. The primary cultures possess PTH-sensitive adenylate cyclase activity, while lacking responsiveness to arginine vasopressin and calcitonin. In contrast, the renal cortex exhibits a significant response to arginine vasopressin and calcitonin due to a small proportion of cells derived from the distal tubule and loop of Henle (46).

The Na-K-ATPase consists of an α-subunit with catalytic activity, as well as a β-subunit required for the integration of Na-K-ATPase into the plasma membrane (26). The β-subunit has been proposed to be a limiting factor in the formation of intact α/β heterodimers (excess α-subunits being rapidly degraded) (26). Indeed overexpression of the β1-subunit has been observed to result in a corresponding increase in the expression of endogenous α in a number of cell systems, including Xenopus laevis oocytes (58) and Moloney sarcoma virus-transformed MDCK cells (53). However, recently, Laughery et al. (41) have reported that overexpression of flag-tagged β1 in MDCK cells does not lead to a significant change in expression of either endogenous α- or β1-subunits of Na-K-ATPase. In this report, the observed increase in transcription of the β1-subunit is associated with an increase in the level of both the α- and β-subunits in primary RPT cell cultures. In addition, α- as well as β-subunit mRNA levels increase in primary RPTs in response to PGE. Thus additional studies are necessary to evaluate whether an increase in α-subunit gene transcription also plays a role in regulating the level of Na-K-ATPase in the proximal tubule.

Our results indicate that β-subunit gene expression increases to a larger extent than α-subunit gene expression both in primary RPT cells and in MDCK cells. The results that we obtained by means of real-time PCR indicate that in primary RPTs treated with 0.7 μM PGE₁, the level of β-subunit mRNA increased to a twofold higher extent than the level of α-subunit mRNA. We obtained a similar result previously through a Northern blot analysis of 8-Br-cAMP-treated primary RPTs (30). A consequence of the increased expression of the Na-K-ATPase α- and β-subunit mRNA levels in PGE-treated primary RPTs is an increase in the level of the Na-K-ATPase, as well as in Na-K-ATPase activity. An increase in the initial rate of ouabain-sensitive Rb⁺ uptake was observed in primary RPTs treated with PGE₁ for 4 days. Because the observed increase in the rate of Rb⁺ uptake may possibly reflect an increase in intracellular Na⁺, rather than an increase in the number of Na-K-ATPases, Na-K-ATPase activity was also determined in crude membrane vesicle preparations. In our chronic studies, a similar increase in Na-K-ATPase activity was observed in the studies with membrane vesicle preparations and intact monolayers (in which ⁸⁶Rb⁺ was employed). In our acute studies, an increase in the rate of ouabain-sensitive Rb⁺ uptake was similarly observed in PGE₁-treated cultures. As in the case with the chronic studies, the increased rate of ouabain-sensitive Rb⁺ uptake may also reflect an increase in intracellular Na⁺, in addition to the observed increase in the level of Na-K-ATPase in the basolateral membrane.

The results of our Western blot analysis of Na-K-ATPase α- and β-subunits indicates that the observed increase in unglycosylated β-subunit in PGE₁-treated primary RPTs was only slightly higher (10%) than the increase in α-subunit. The increase in the level of glycosylated β-subunit did not differ significantly than the observed increase in the α-subunit. Similarly, in MDCK cells we previously reported a sevenfold increase in β-subunit mRNA and a twofold increase in α-subunit mRNA mediated by PGE₁ concentrations (65). Nevertheless, the increase in the α-subunit level did not differ significantly from the increase in the level of the β-subunit in MDCK cells. However, consistent with β-subunit being limiting are our preliminary results with MDCK cells that indicate a knockdown of the β-subunit with β-subunit small interference RNA results in a reduction in the α-subunit level.

PGs are modulators of renal microvascular hemodynamics and renin release as well as tubular salt and water reabsorption (11). PG biosynthesis by COX occurs in all sections of the nephron, including the RPT (9, 11). Thus the stimulatory effect of AA on transcription by pHβ1-1141 Luc can very likely be attributed to autocrine effects of endogenously produced AA metabolites, including PGE₂. Indeed, the stimulatory effect of PGE₂ was inhibited >80% by the COX inhibitor indomethacin. However, our results indicate that AA metabolites, including metabolites generated by cytochrome P-450 (such as 20-HETE) also contribute to the stimulatory effect of AA (60).

In addition, our results indicate that several other intrarenal factors, norepinephrine and dopamine, regulate ATP1B1 transcription in primary RPT cells. Previously, both norepinephrine and epinephrine were observed to cause an acute increase in renal Na-K-ATPase activity in the RPT, characterized by a redistribution of Na-K-ATPase from the cytosolic to basolateral compartment (3). Norepinephrine and epinephrine are produced in two distinct locales, in response to distinct stimuli. While epinephrine is produced by the adrenal gland in response to stress, norepinephrine is produced by renal nerves when dietary Na⁺ is low. When dietary Na⁺ is reduced for a lengthy time, chronic regulatory mechanisms also come into play (2). The consequence is an increase in transcription of the ATP1B1 gene in response to norepinephrine, as we have observed here.

Unlike norepinephrine, dopamine, is produced in the RPT from l-DOPA by an aromatic l-amino acid decarboxylase (4). Following the biosynthesis of dopamine in the RPT (presumably when dietary salt intake increases), dopamine acts via dopamine D1-like receptors so as to acutely inhibit the RPT Na-K-ATPase. In contrast, dopamine has a chronic stimulatory effect on ATP1B1 gene transcription in primary RPTs. This stimulation may possibly be the consequence of signaling through dopamine D2-like receptors. Indeed, Narkar et al. (48) have reported a dopamine D2-like receptor-mediated stimulation of Na-K-ATPase in rat RPTs treated with the D2 agonist bromocriptine. Similarly, a stimulatory effect of dopamine on Na-K-ATPase has been observed in lung alveolar epithelial cells and has been attributed to a stimulation of ATP1B1 gene transcription (28).

Similarly, chronic effects of PGs on Na-K-ATPase in primary RPTs differ from previously reported acute effects in collecting duct cells, both in vitro and in vivo (15). In primary RPTs, a stimulatory effect of PGE₁ and PGE₂ was observed following acute as well as chronic treatment. These results may accurately reflect long-term as well as short-term regulation in the RPT in response to increases in the levels of PGs, and changes in Na⁺ balance (59). During short-term fluctuations in Na⁺ balance, the tubulogomerular feedback mechanism stabilizes renal Na⁺ secretion (59). Long-term changes in Na⁺ balance are characterized by an altered rate of renin secretion, as well as changes in the level of ANG II and COX2 in the macula densa (59). Increases in ATP1B1 gene transcription in the RPT cells may be one of the consequences of such changes in renal cortical COX2 activity, in addition to other long-term physiological changes.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grant 1R01HL6976-01 to M. Taub.

DISCLOSURES

No conflicts of interest are disclosed by the authors.