INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

REGULATION OF THE ACTIVITY of proximal renal tubule sodium-phosphate cotransporters is a critical mechanism for the maintenance of total body phosphate homeostasis. Two types of sodium-phosphate cotransporters have been described on the apical membrane of renal proximal tubule. Type II transporters, which are regulated by physiological, dietary, and hormonal stimuli (18), have been cloned from the kidneys of several species (3-5, 8, 14, 16, 23-27). The derived amino acid sequences of these proteins demonstrate over 80% identity and nearly 90% similarity to each other but little sequence similarity to other known proteins, including other sodium-coupled cotransporter proteins (28).

Parathyroid hormone (PTH) is a major hormonal regulator of sodium-phosphate cotransport in proximal renal tubules. Stimulation of the PTH receptor causes activation of two signaling pathways, adenylyl cyclase and phospholipase C, resulting in activation of protein kinase A and protein kinase C (6). Selective activation of either protein kinase results in inhibition of sodium-dependent phosphate cotransport, similar to the effect of PTH itself; however, the specific roles of these two signaling pathways in the physiological regulation of phosphate transport have not been determined. Studies in rat kidney show that PTH inhibition of sodium-phosphate cotransport is accompanied by downregulation of the apical membrane expression of NaPi-2, the rat type II sodium-phosphate cotransporter (10, 17, 20). The precise molecular mechanism by which NaPi-2 expression is downregulated is unknown; however, evidence suggests that the cotransporter proteins undergo endocytosis and are incorporated into endosomal vesicles (13) through a process that does not require protein synthesis (15) but that may involve small molecular weight G proteins of the rho/rac family (22). Recovery of sodium-phosphate cotransport after withdrawal of PTH does require protein synthesis, and we have recently demonstrated that the cotransporter proteins incorporated into endosomal vesicles undergo degradation (20).

One model used to study regulation of proximal tubule phosphate transport is opossum kidney (OK) cells, a continuous cell line derived from opossum kidney (1, 21). These cells retain many properties of proximal tubules, including expression of PTH receptors coupled to adenylyl cyclase and phospholipase C and sodium-dependent phosphate transporters regulated by PTH and other physiological stimuli. The effect of PTH on expression of NaPi-4, the OK cell type II sodium-phosphate cotransporter, has not been studied. Because sodium-phosphate cotransport is inhibited by PTH in OK cells similarly to rat kidney, we hypothesized that PTH would also downregulate the membrane expression of NaPi-4. To test this hypothesis, we developed polyclonal antisera directed against the carboxy-terminal 12-amino acid sequence of NaPi-4 in rabbits and used these antisera to detect expression of NaPi-4 in OK cell membranes by immunoblot and in intact OK cells by confocal imaging. Additionally, we studied the effect of selective activation of protein kinase A and protein kinase C on NaPi-4 expression.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Materials. Wild-type OK cells were a generous gift from Dr. Steven Scheinman (Health Sciences Center, Syracuse, NY). PTH-(1-34) was obtained from Bachem (Philadelphia, PA). PTH-(3-34) was obtained from Peninsula Laboratories (Belmont, CA). 8-Bromo-cAMP (8-BrcAMP) and phorbol 12-myristate 13-acetate (PMA) were purchased from Sigma (St. Louis, MO). Peroxidase-labeled goat anti-rabbit immunoglobulin G (IgG) was obtained from Vector Laboratories (Burlingame, CA). Fluorescein isothiocyanate-coupled goat anti-rabbit polyclonal IgG was obtained from Molecular Probes (Eugene, OR). [³²P]phosphoric acid was purchased from ICN Biomedicals (Irvine, CA).

Cell culture. Wild-type OK cells, passages 82-89, were grown as monolayers in 175-cm² plastic flasks (Falcon), six-well dishes, or eight-well chamber coverglass slides (Nunc) in Eagle's medium with Earle's salts (GIBCO-BRL; Life Technologies, Grand Island, NY), supplemented with 10% heat-inactivated fetal calf serum, 4 mM glutamine, 100 µg/ml streptomycin, and 100 IU/ml penicillin in a humidified 5% CO₂-95% air environment at 37°C. They were fed three times per week and split 1:4 once per week by brief trypsinization and dispersal. Cells were used for experiments at 100% confluence.

Peptide antibody production. A peptide identical to the carboxy-terminal 12-amino acid sequence of NaPi-4 (CGVLSQHNATRL) was generated and conjugated to keyhole limpet hemocyanin (Genosys Biotechnologies, Woodlands, TX). Peptide (100 µg) was mixed in Freund's complete adjuvant and injected subcutaneously into New Zealand White rabbits. The rabbits were given booster injections in incomplete Freund's adjuvant on a monthly basis. Before the first injection, 50 ml of blood was drawn, and serum was separated and frozen at 70°C to be used for preimmune testing. After immunization, 50 ml of blood was drawn, and the serum was separated and frozen on a thrice-monthly basis.

Membrane preparation. OK cells grown in 175-cm² flasks were suspended in phosphate-buffered saline (PBS), washed twice in buffer containing 10 mM triethanolamine (TEA) and 140 mM NaCl, pH 7.4, and resuspended in buffer containing 250 mM sucrose, 20 mM tris(hydroxymethyl)aminomethane hydrochloride (Tris · HCl), pH 7.5, 1.5 mM MgCl₂, 1 mM ATP-Na₂, 1 µM leupeptin, 2 µg/ml soybean trypsin inhibitor, 3 mM benzamidine, and 1 mM phenylmethylsulfonyl fluoride (PMSF). The cells were lysed by nitrogen cavitation at 325 psi for 30 min or by sonication three times for 5 s each. To remove nuclei, the lysate was centrifuged at 4,000 rpm for 45 s, and the supernatant was filtered through cheesecloth. Then the supernatant was centrifuged at 15,000 rpm for 15 min. The pellet was resuspended in buffer containing 20 mM Tris · HCl and 1 mM EDTA, snap frozen in liquid nitrogen, and stored at 70°C until use. Membranes prepared by this method retain PTH-stimulated adenylyl cyclase activity (data not shown).

OK cells grown to 100% confluence in six-well trays were washed with serum-free medium and incubated in medium containing specific agonists for the designated period of time. The cells were then washed with calcium- and magnesium-free PBS and with HEPES-KOH (5 mM HEPES, pH 7.4). Ice-cold HEPES-KOH (500 µl) containing 4 mM EDTA and 1 mM PMSF was added to the wells, and the cells were scraped into microcentrifuge tubes and homogenized by repeated aspiration into a 1-ml syringe through a 20-gauge needle. The cell lysates were centrifuged at 2,000 rpm for 10 min at 4°C to remove nuclei. The pellet was discarded, and the supernatant was centrifuged at 15,000 rpm for 40 min at 4°C. The resulting pellet was suspended in 100 µl of 50 mM mannitol with 10 mM HEPES-Tris (pH 7.2) and frozen at 70°C until used for immunoblot assays.

Membrane digestion with glycosidase. Twenty-five microliters of 5× SDS-Laemmli sample buffer was added to 100 µl of membrane protein and boiled for 5 min. To this solution, 275 µl of buffer containing 30 mM NaPO₄, pH 7.2, 14.5 mM sodium azide, 72.7 mM EDTA, and 0.7% Nonidet P-40 was added, and the solution was boiled for another 2 min. To this solution, 16 µl of a 6 U/120 µl solution of N-glycosidase F was added, and the preparation was incubated at 37°C for 24 h. One hundred and four microliters of this preparation was loaded into the SDS-PAGE lane, along with equal quantities of undigested membrane preparation.

Immunoblot assay. OK cell membranes (50-100 µg/assay) were solubilized in Laemmli sample buffer, subjected to 10% SDS-PAGE, and transferred electrophoretically to either nitrocellulose (Trans-Blot; Bio-Rad Laboratories, Hercules, CA) or polyvinylidene difluoride (PolyScreen; DuPont-NEN, Boston, MA). The membrane was incubated in 5% milk in 20 mM Tris, 50 mM NaCl, and 0.05% Tween 20 (TTBS) at room temperature for 1 h to inhibit nonspecific binding and then incubated overnight at 4°C in antisera for the NaPi-4 peptide at a 1:2,000 dilution in 5% milk in TTBS. Location of specific antibodies was detected by incubation with peroxidase-labeled goat anti-rabbit IgG at a 1:10,000 dilution in 5% milk in TTBS, followed by development with enhanced chemiluminescence (Renaissance; DuPont-NEN). The bands imaged by chemiluminescence were analyzed by densitometry. The films were scanned using a Personal Densitometer SI (Molecular Dynamics), and the areas of interest were identified and quantitated by integrated software (ImageQuant, Molecular Dynamics).

Confocal imaging. OK cell monolayers were trypsinized, seeded lightly onto eight-well chamber slides, and incubated overnight in serum-free medium containing agonist. On the morning of the assay, the cells were washed in calcium- and magnesium-free Hanks' balanced salt solution (HBSS) and fixed in 4% paraformaldehyde in Ca/Mg-free HBSS for 1 h at room temperature. The cells were rinsed, incubated in rabbit anti-NaPi-4 antisera 1:100 in Ca/Mg-free HBSS for 1-2 h at room temperature, rinsed again, and incubated in fluorescein isothiocyanate-coupled goat anti-rabbit IgG 1:500 in HBSS for 1 h at room temperature in the dark. Fluorescence was visualized by confocal microscopy (Meridian Laboratories, Okemos, MI). With online computer software (InSight Plus; Meridian Laboratories), individual cells were optically sectioned horizontally from the base to the apex. With the fluorescence pattern for each section, a three-dimensional picture of each cell was constructed, and the total fluorescence for each cell was calculated.

Phosphate uptake assays. Phosphate transport was measured by determination of radiolabeled phosphate uptake into OK cell monolayers, as previously described (11). OK cells were seeded onto 96-well plates and grown to 100% confluence 4-5 days after seeding. Twenty-four hours prior to the assay, the cells were serum deprived by washing in serum-free medium. The assay was initiated by washing the cells in serum-free medium, followed by incubation in serum-free medium containing agonist for the designated period at 37°C. The serum-free medium was then removed, and the cells were washed three times with transport medium consisting of (in mM) 137 NaCl, 5.4 KCl, 2.8 CaCl₂, 1.2 MgSO₄, and 0.1 KH₂PO₄. Phosphate uptake was initiated by the addition of transport medium containing ³²P-radiolabeled phosphoric acid. Uptake was continued for 10 min at room temperature, after which the cell monolayers were washed three times with ice-cold medium which differed from transport medium in that N-methylglucamine was substituted for sodium chloride and ³²P was omitted. We have previously demonstrated that phosphate uptake in OK cells is linear for up to 20 min (data not shown); therefore, 10-min phosphate uptake reflects activity of the apical sodium-phosphate cotransporter (17). The cells were solubilized in 125 µl 0.5% Triton X-100 for 90 min at room temperature. A 100-µl aliquot from each well was aspirated into a scintillation vial to which scintillant was added and analyzed by liquid scintillation spectroscopy. We have previously determined that over 90% of the phosphate uptake in OK cells is sodium dependent (data not shown). Therefore, no correction was made for sodium-independent phosphate uptake. Nonspecific binding was determined and subtracted from each separate assay. Each assay on OK cells was performed in sextuplicate and averaged, and the mean was considered as a single data point. The data are expressed as percent inhibition compared with control uptake.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Characterization of carboxy-terminal peptide antibody. Immunoblot analysis of OK cell membrane proteins separated by 10% SDS-PAGE, transferred to a nylon membrane, and probed with NaPi-4 peptide antisera is shown in Fig. 1. A 1:2,000 dilution of antisera identified several bands (middle lane). Prominent bands included a broad-based band spanning 110-140 kDa, two bands at 43-44 kDa, and a band at 50 kDa. Fainter bands were visible at 55-60, 40, 38, 30, and 25 kDa. Incubation of OK cell membranes with preimmune sera identified only a 33-kDa band (left lane). Incubation of OK cell membranes with immune serum that had been preincubated with the immunizing peptide at a concentration of 100 µg/ml prevented identification of the broad 110- to 140-kDa band but failed to prevent identification of the other bands (right lane). To confirm the specificity of the peptide blocking, we also performed immunoblot analysis using antisera that had been preincubated with an unrelated peptide of similar size and compared these results with immunoblots performed with antisera alone and antisera preincubated with cognate peptide. This results are shown in Fig. 2. The left two lanes represent immunoblots of membranes using antisera at a 1:2,000 dilution. The middle two lanes represent immunoblots of membranes using antisera that have been preincubated with unrelated peptide 100 µg/ml. The right two lanes represent immunoblots of membranes using antisera that have been preincubated with cognate peptide 100 µg/ml. The lanes marked "U" are undigested membrane preparations. The lanes marked "D" are membranes that have been digested with N-glycosidase (discussed below). Figure 2 shows that the 110- to 140-kDa band, which is identified by the antisera alone, is still identified by the antisera when they are preincubated with an unrelated peptide but not when they are preincubated with the cognate peptide. Thus the 110- to 140-kDa band is specifically identified by antisera directed against the 12-amino acid carboxy terminus of NaPi-4.

View larger version (56K): [in this window] [in a new window]   Fig. 1.   Immunoblot analysis of opossum kidney (OK) cell membranes, using polyclonal antisera to NaPi-4 carboxy-terminal peptide. OK cell membrane protein (50 µg/lane) was subjected to 10% SDS-PAGE, transferred electrophoretically to polyvinylidene difluoride (PVDF), and blotted against indicated rabbit antisera (1:2,000), followed by incubation in goat anti-rabbit IgG (1:10,000 dilution). Immunoblot was developed by chemiluminescence. Lanes are from 1:2,000 dilutions, with immune sera preincubated for 30 min with immunizing peptide 100 µg/ml.

View larger version (73K): [in this window] [in a new window]   Fig. 2.   Effect of digestion with N-glycosidase on NaPi-4 expression. OK cell membranes were incubated with 2 U/ml N-glycosidase for 24 h prior to gel electrophoresis. Twenty-five micrograms of digested (D) or undigested (U) membrane protein were loaded onto 10% SDS-PAGE gels, as described above, and transferred electrophoretically to PVDF. Immunoblots were blocked with 5% milk in TTBS followed by incubation in primary antibody (1:2,000 dilution) overnight and incubation in goat anti-rabbit IgG (1:10,000 dilution). Immunoblot was developed by chemiluminescence. Left lanes were incubated in specific antisera alone, middle two lanes were incubated in antisera that had been preincubated with a nonrelated peptide 100 µg/ml, right lanes were incubated in antisera that had been preincubated with 100 µg/ml immunizing peptide.

The apparent molecular mass of 110-140 kDa is higher than the 65- to 70-kDa molecular mass predicted by the amino acid sequence of NaPi-4. The relatively broad spread of the band identified at 110-140 kDa suggests that the protein undergoes posttranslational modification, such as glycosylation, and the predicted amino acid sequence for NaPi-4 shows several potential sites for glycosylation. To determine whether the 110- to 140-kDa band represented a glycosylated protein, we performed two maneuvers. First, we preincubated isolated membranes for 24 h with N-glycosidase, which cleaves proteins at sites of N-glycosylation, prior to subjecting the membranes to SDS-PAGE and performing immunoblot analysis. Figure 2 (left two lanes) shows the effect of digestion with endoglycosidase H on the density of the 110- to 140-kDa band. Equal quantities of protein from digested (D) and undigested (U) membrane preparations were loaded onto the gel. The intensity of the 110- to 140-kDa band from the digested protein preparation is markedly reduced compared with the undigested preparation. Additionally, a band at the molecular size of ~70 kDa appears much more prominently in the digested preparation, likely representing the nonglycosylated form of the protein. The comparison of digested to undigested membrane preparations was performed four times with identical results. To confirm these data, we performed immunoblot analysis of membranes from cells that had been grown in medium containing 2 µg/ml tunicamycin to inhibit posttranslational glycosylation. As shown in Fig. 3, membranes from cells grown in tunicamycin showed a marked decrease in the intensity of the 110- to 140-kDa band and an increase in the intensity of a band at ~70 kDa, compared with membranes from control cells. This assay was performed three times with the same results.

View larger version (126K): [in this window] [in a new window]   Fig. 3.   Effect of tunicamycin on NaPi-4 expression. OK cells were grown for 24 h in medium containing vehicle (Control) or 2 µg/ml tunicamycin prior to membrane preparation. Fifty micrograms of membrane protein were loaded, subjected to 10% SDS-PAGE, tranferred electrophoretically to PVDF, and blotted with rabbit antisera (1:2,000) in 5% milk in TTBS overnight at 4°C. Immunoblot was then washed and incubated with goat anti-rabbit IgG (1:10,000) in 5% milk in TTBS for 1 h, and the blot was developed by chemiluminescence.

To determine the cellular distribution of the proteins identified by peptide antibody, OK cells grown on coverslips were fixed and stained with antisera, and the bound antisera were visualized with fluorescein-conjugated secondary antibody using confocal microscopy. As shown in Fig. 4, the antisera directed against the carboxy-terminal peptide recognized a protein that has a predominantly apical distribution in intact OK cell monolayers (Fig. 4A). Similar to what has been reported with rat renal proximal tubule cells (18), there was significant variability in the cell to cell staining for NaPi-4. Preincubation of the antisera with immunizing peptide resulted in markedly decreased fluorescence in OK cell monolayers, confirming the specificity of the peptide antibody for the apical membrane protein (Fig. 4B). Preincubation of the antisera with an unrelated peptide, identical to the peptide used in the immunoblot experiments, did not decrease the intensity of the fluorescence (data not shown).

View larger version (82K): [in this window] [in a new window]   Fig. 4.   Confocal imaging of intact OK cells using polyclonal antisera to NaPi-4 carboxy-terminal peptide. OK cell monolayers were grown on coverslips, washed, and fixed in 4% paraformaldehyde for 1 h at room temperature. The cells were rinsed and incubated with NaPi-4 antisera 1:100 for 1 h at room temperature, followed by incubation with fluorescein-conjugated goat anti-rabbit IgG (1:500) for 1 h in the dark (A). B: cells that were incubated with NaPi-4 antisera (1:100) preincubated for 30 min with 1 mg/ml immunizing peptide.

Response to PTH and phosphate deprivation. The expression of type II sodium-phosphate cotransporters in rat proximal tubule cells is downregulated by exposure to PTH and upregulated by low-phosphate diet (12), corresponding to a decrease and an increase in phosphate transport, respectively. PTH decreases and low-phosphate medium increases sodium-phosphate cotransport similarly in OK cells (1, 17, 21). We therefore performed immunoblots on membranes of OK cells that had been exposed to 10⁷ M PTH for 24 h or to a very low phosphate medium for 72 h prior to membrane formation to determine whether the expression of NaPi-4 was also regulated by these stimuli. Figure 5 shows comparison of immunoblots of OK cells membranes under control conditions (left lane), after a 24-h pretreatment with PTH 10⁷ M (middle lane), and after a 72-h exposure to a very low phosphate (0.1 mM) medium (right lane). Membranes from the PTH-treated cells showed a 60% reduction in the density of the 110- to 140-kDa band (n = 10), consistent with downregulation of the membrane expression of NaPi-4. Conversely, membranes from cells exposed to a low-phosphate medium showed a twofold increase in the density of the same band (n = 4), indicating upregulation of the membrane expression of NaPi-4.

View larger version (K): [in this window] [in a new window]   Fig. 5.   Immunoblot analysis of OK cell membranes derived from cells grown under control conditions (left), after 24 h of treatment with 107 M parathyroid hormone (PTH, middle), or after 72 h of incubation in a low-phosphate (0.1 mM) medium (right). One hundred micrograms of membrane protein/lane were subjected to 10% SDS-PAGE, transferred electrophoretically to nitrocellulose, and blotted against NaPi-4 peptide antibody (1:2,000) overnight, followed by incubation with goat anti-rabbit IgG (1:10,00) for 1 h. Immunoblot was developed by chemiluminescence.

The decrease in membrane expression of NaPi-4 in PTH-treated cells was confirmed by confocal imaging of OK cells. Figure 6 shows sagittal and coronal views of single OK cells under control conditions (Fig. 6A) or after pretreatment with 10⁷ M PTH for 24 h (Fig. 6B). These confocal images demonstrate the apical localization of the fluorescence and the marked attenuation after pretreatment with PTH.

View larger version (147K): [in this window] [in a new window]   Fig. 6.   Confocal imaging of intact OK cells under control conditions (top) and after treatment for 24 h with 107 M PTH (bottom). Cells were grown on coverslips, fixed in 4% paraformaldehyde for 1 h at room temperature, washed, and incubated with NaPi-4 peptide antisera (1:100) at room temperature for 1 h, followed by incubation with fluorescein-conjugated goat anti-rabbit IgG (1:500) for 1 h in the dark. Cells were scanned and optically sectioned horizontally using a laser confocal microscope. Fluorescence was calculated, and a three-dimensional representation of the cell was generated using computer interface software to produce a coronal (left) and sagittal (right) image of each cell.

Effect of 8-BrcAMP and PMA on NaPi-4 expression. PTH activates protein kinase A and protein kinase C, and selective activation of either protein kinase inhibits phosphate uptake similar to the effect of PTH. Figure 7 shows inhibition of phosphate uptake by 10⁷ M PTH, 10⁴ M 8-BrcAMP, a direct activator of protein kinase A, and 10⁶ M PMA, a direct activator of protein kinase C, after 4 and 24 h of incubation. As can be seen, the three agonists produced similar reduction in phosphate uptake at 4 h. PTH decreased phosphate uptake by 35.8 ± 0.7%, 8-BrcAMP decreased phosphate uptake by 45.3 ± 0.8%, and PMA decreased phosphate uptake by 44.1 ± 2.3%. After 24 h, inhibition of phosphate uptake by PTH and 8-BrcAMP was sustained at the same level, 39.4 ± 1.5% and 41.5 ± 1.5%, respectively. On the other hand, after a 24-h treatment with PMA, sodium-dependent phosphate uptake had not only returned to control values but exceeded control uptake values. To determine whether selective activation of protein kinase A or C also regulated expression of NaPi-4, immunoblot analysis was performed on membranes derived from OK cells which had been pretreated for up to 24 h with 10⁷ M PTH, 10⁴ M 8-BrcAMP, or 10⁶ M PMA. As can be seen in Fig. 8, PTH and 8-BrcAMP produced a progressive decrease in NaPi-4 expression, which was detectable by 2 h and maximal by 4 h. After 24 h, 8-BrcAMP decreased the density of NaPi-4 in OK cell membranes by 91 ± 1.9% (n = 3). On the other hand, PMA had little effect on the expression of NaPi-4 for the entire 24 h of incubation, decreasing the density of NaPi-4 by 9.2 ± 3.1% (n = 3).

View larger version (29K): [in this window] [in a new window]   Fig. 7.   Inhibition of phosphate uptake by PTH, 8-BrcAMP (8Br), and phorbol 12-myristate 13-acetate (PMA). OK cell monolayers in 96-well plates were incubated in medium containing vehicle, 107 M PTH (PTH), 104 M 8-BrcAMP, or 106 M PMA for either 4 or 24 h. Phosphate uptake was initiated by addition of transport medium containing radiolabeled [32P]phosphoric acid. After a 10-min uptake period, the cells were washed and solubilized with 0.5% Triton X-100, and an aliquot from each well was counted by liquid scintillation spectroscopy. Each assay was performed in sextuplicate, averaged, and considered a single data point. Results are expressed as %inhibition phosphate uptake compared with control conditions. * P < 0.01, phosphate uptake significantly different than control. Graph represents n = 4 separate experiments. Control phosphate uptake was 435.1 ± 7.4 pmol Pi · min1 · mg protein1.

View larger version (47K): [in this window] [in a new window]   Fig. 8.   Time course of PTH, 8-BrcAMP (8-Br), and PMA effect on OK cell membrane expression of NaPi-4. OK cell monolayers in 6-well trays were incubated with 107 M PTH, 104 M 8-BrcAMP, or 106 M PMA for 0, 2, 4, 6, 8, or 24 h. Equal quantities (50-100 µg/lane) of membrane protein were subjected to 10% SDS-PAGE, transferred electrophoretically to PVDF, and exposed to NaPi-4 peptide antisera (1:2,000) in 5% milk in TTBS, followed by incubation with goat anti-rabbit IgG (1:10,000) in 5% milk in TTBS. Blots were developed by chemiluminescence. Each time course is representative of n = 3 for each agonist.

These data suggest that PTH-stimulated protein kinase A is responsible for downregulation of NaPi-4, whereas PTH-stimulated protein kinase C inhibits phosphate uptake through another mechanism. To test this hypothesis, we compared the NaPi-4 staining on membranes derived from cells that had been treated with 10⁷ M PTH-(1-34), which activates protein kinase A and protein kinase C, and membranes from cells which had been treated with 10⁶ M PTH-(3-34), a fragment that activates protein kinase C only and not protein kinase A (2). Figure 9 shows an immunoblot representing these data; the two lanes on the left are membranes from control cells, the two lanes in the middle are membranes from cells treated with PTH(1-34), and the lanes to the right are membranes from cells treated with PTH-(3-34) for 2 h. PTH-(1-34) but not -(3-34) downregulated expression of NaPi-4. To confirm that PTH-(3-34) inhibits phosphate uptake, we performed phosphate uptake after a 2-h incubation with PTH-(3-34). Figure 10 shows the dose-response curve. At 10⁶ M, PTH-(3-34) inhibits sodium-dependent phosphate uptake to a comparable degree as PTH-(1-34) after a 2-h incubation. Thus the discrepancy in NaPi-4 staining cannot be accounted for by differential ability to inhibit phosphate uptake. These data are consistent with the hypothesis that PTH-stimulated protein kinase C inhibits phosphate uptake but not NaPi-4 expression.

View larger version (76K): [in this window] [in a new window]   Fig. 9.   Comparison of the effect of PTH-(1-34) and PTH-(3-34) on NaPi-4 expression. OK cell monolayers were incubated in vehicle, 107 M PTH-(1-34), or 106 M PTH-(3-34) for 2 h followed by membrane preparation. Fifty micrograms of each preparation were subjected to 10% SDS-PAGE, transferred electrophoretically to PVDF, and blotted with antisera (1:2,000) in 5% milk in TTBS overnight at 4°C. Blot was then incubated in (1:1,0,000) dilution of goat anti-rabbit IgG for 1 h at room temperature and developed by chemiluminescence. Results are for two separate experiments for each condition.

View larger version (12K): [in this window] [in a new window]   Fig. 10.   Inhibition of phosphate uptake by PTH-(3-34). OK cell monolayers in 96-well trays were incubated in vehicle or increasing concentrations of PTH-(3-34) for 2 h at 37°C. Phosphate uptake was initiated by addition of transport medium containing radiolabeled [32P]phosphoric acid. After a 10-min uptake period, the cells were washed and solubilized with 0.5% Triton X-100, and an aliquot from each well was counted by liquid scintillation spectroscopy. Each assay was performed in sextuplicate, averaged, and considered a single data point. Results are expressed as %inhibition phosphate uptake compared with control conditions for 3 separate assays.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

OK cells are widely used as a model of proximal renal tubules, especially for the investigation of PTH signaling and regulation of sodium-phosphate cotransport. Previous studies have documented that regulation of sodium-phosphate cotransport in OK cells is similar to that seen in intact mammalian models such as the rat. In the rat, PTH inhibition of phosphate transport correlates with a decrease in the expression of the type II sodium-phosphate cotransporter, as demonstrated by immunohistochemistry (10). OK cells express a type II sodium-phosphate cotransporter, which was recently cloned and shown to have significant similarity and identity to other type II sodium-phosphate cotransporters (23). The similarities in PTH inhibition of phosphate uptake and in expression of type II sodium-phosphate cotransporters in rat renal proximal tubule and OK cells suggested to us that the mechanisms employed by PTH to regulate sodium-phosphate cotransport in OK cells would be similar to the mechanisms seen in rat. Specifically, we asked whether PTH inhibition of phosphate uptake would correspond to a decrease in the membrane expression of the cotransporter protein. To measure membrane expression of NaPi-4, we developed a peptide antibody for use in immunoblot and confocal microscopic analysis of OK cells.

Our results show that a polyclonal antibody raised against a peptide identical to the 12-amino acid carboxy-terminal sequence of NaPi-4 identifies a broad band of 110-140 kDa, which is recognized by neither preimmune sera nor by sera incubated with immunizing peptide. Incubation of the antisera with an unrelated peptide does not block recognition of the 110- to 140-kDa band, confirming the specificity of the antisera. The estimated molecular size and the appearance on immunoblot correspond well to results seen in rat with studies of NaPi-2 (18). The diffuse nature of the band is presumably secondary to differential glycosylation of cotransporter proteins, as has been demonstrated for NaPi-2 (9). The putative amino acid sequence for NaPi-4 predicts several potential glycosylation sites. Treatment of OK cell membranes with N-glycosidase results in loss of the high-molecular-weight band recognized by the NaPi-4 peptide antisera. Similar data are obtained when OK cells are pretreated with tunicamycin, which inhibits glycosylation. Further evidence that our antisera are specific for NaPi-4 is furnished by the confocal microscopy studies. Intact OK cells exposed to antisera exhibit an apical localization of the fluorescence, as would be expected for NaPi-4. This protein is not visualized when the cells are exposed to preimmune sera. The apical fluorescence is also markedly reduced when cells are exposed to antisera that has been preincubated with immunizing peptide. These data confirm that our antisera specifically recognize NaPi-4 in OK cells.

To determine whether regulation of sodium-phosphate cotransport correlates with regulation of NaPi-4 expression, we examined the effect of PTH and low phosphate on NaPi-4 expression. With immunoblot analysis and confocal imaging, we showed that after 24 h of exposure, PTH decreases NaPi-4 expression by 60%. On the other hand, growth in a low-phosphate medium, a stimulus for increased phosphate uptake, produces a twofold increase in NaPi-4 expression. These data confirm that regulation of NaPi-4, the type II sodium-phosphate cotransporter in OK cells, is similar to regulation of the homologous sodium-phosphate cotransporter in rat.

Although it has been recognized that PTH stimulates two signaling pathways and that activation of either pathway alone inhibits phosphate uptake, the role of each pathway in PTH regulation of phosphate transport has not been determined. Therefore, we also studied the effect of protein kinase A and protein kinase C on regulation of the membrane expression of NaPi-4. Our studies demonstrate that protein kinase A activation decreases expression of NaPi-4 to even a greater extent than PTH; however, protein kinase C activation has almost no effect. These results suggest that PTH-stimulated protein kinase A inhibits phosphate uptake by decreasing NaPi-4 protein expression and that PTH-stimulated protein kinase C may inhibit sodium-dependent phosphate uptake by decreasing NaPi-4 function. The fact that PTH-(3-34), a PTH fragment which stimulates protein kinase C but not protein kinase A, also does not inhibit NaPi-4 expression strengthens this conclusion. Because the polyclonal sera used to identify NaPi-4 recognize a carboxy-terminal epitope of the protein, an alternative explanation for our data is possible. Protein kinase A activation could result in an alteration of the carboxy-terminal portion of the protein, which inhibits function of NaPi-4 without altering expression of the protein but which prevents recognition of the protein by the antibody. Protein kinase C could activate a pathway that alters another portion of the protein, allowing continued recognition of the protein by the antibody but still inhibiting function of NaPi-4. Our data do not allow us to distinguish between these two possibilities. The recent report by Pfister et al. (20) showing downregulation by PTH of NaPi-2 and NaPi-4 using both carboxy-terminal and amino-terminal antisera suggests that this latter explanation is unlikely. The significance of the difference in protein kinase A and protein kinase C regulation of NaPi-4 remains speculative, but several explanations could be forwarded. The activation of these two pathways could simply represent redundant mechanisms to ensure that this physiologically important effect of PTH can be carried out. Alternatively, at different concentrations or duration of exposure to PTH, the effects of protein kinase A and protein kinase C may differ in their predominance. For example, at lower PTH concentrations or with transient exposure to PTH, the protein kinase C effect may predominate, whereas, with higher or more chronic increases in PTH concentrations, the protein kinase A effect may predominate. Another explanation for the presence of two pathways is that one or the other pathway may be preeminent during different conditions such as aging, growth, or regeneration (6, 19). The availability of two different pathways with differing mechanisms of controlling NaPi-4 may allow better fine tuning of the effect of PTH.