Parathyroid Hormone-dependent Degradation of Type II Na+/Pi Cotransporters*

Parathyroid hormone (PTH) inhibits proximal tubular brush border membrane Na+/Picotransport activity; this decrease in the transport activity was found to be associated with a decrease in type II Na+/Pi cotransporter protein content in rat brush border membranes. In the present study we investigated the PTH-dependent regulation of the type II Na+/Pi cotransporter in opossum kidney cells, a previously established model to study cellular mechanisms involved in the regulation of proximal tubular Na+/Picotransport. We transfected opossum kidney cells with a cDNA coding for NaPi-2 (rat renal type II Na+/Pi cotransporter). This allowed the study of PTH-dependent regulation of the transfected NaPi-2 and of the corresponding intrinsic cotransporter (NaPi-4). The results show (i) that the intrinsic and the transfected cotransporters are functionally (transport) and morphologically (immunofluorescence) localized at the apical membrane, (ii) that the intrinsic as well as the transfected Na+/Pi cotransport activities are inhibited by PTH, (iii) that PTH leads to a retrieval of both cotransporters from the apical membrane, (iv) that both cotransporters are rapidly degraded in response to PTH, and (v) that the reappearance/recovery of type II Na+/Pi cotransporter protein and function from PTH inhibition requires de novo protein synthesis. These results document that PTH leads to a removal of type II Na+/Pi cotransporters from the apical membrane and to their subsequent degradation.

Renal proximal tubular P i reabsorption is acutely regulated by parathyroid hormone (PTH). 1 This effect involves inhibition of the brush border membrane sodium-dependent P i transport and is characterized by a decrease in the maximal transport rate (V max ) (1,2). Two different renal Na ϩ /P i cotransporters have been cloned, classified either as type I Na ϩ /P i cotransporter or as type II Na ϩ /P i cotransporter (3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14). Both are localized at the brush border membrane in proximal tubules. Recent data documented that physiologically and pathophysiologically altered brush border membrane Na ϩ /P i cotransport involves altered brush border expression of the type II Na ϩ /P i cotransporter (15)(16)(17).
In the present study we investigated the PTH-mediated regulation of the type II Na ϩ /P i cotransporter in opossum cells (OK cells); these cells have recently been shown to contain such a cotransporter (NaP i -4; Ref. 8). The validity of the opossum kidney cell model to study proximal tubular Na ϩ /P i cotransport and its regulation has been established (18 -23). With respect to PTH-dependent control of Na ϩ /P i cotransport activity, we have reported that the recovery from the PTH-mediated inhibition of Na ϩ /P i cotransport in OK cells is dependent on de novo protein synthesis. This latter observation led to the hypothesis that PTH might lead to the retrieval and degradation of the transporter (27).
The aims of the present study were 2-fold: (i) to study cellular/molecular mechanisms involved in PTH-dependent control of type II Na ϩ /P i cotransporters, (ii) to create by transfection an in vitro model that also permits the study of the PTH control of the rat type II Na ϩ /P i cotransporter. Obviously, the latter approach would then offer a tool to characterize the molecular determinants involved in such regulations. A prerequisite for this approach was the availability of antisera, permitting a distinction between intrinsic (NaP i -4) and transfected (NaP i -2) cotransporters.
The results obtained show that both intrinsic (NaP i -4) and transfected type II Na ϩ /P i cotransporters (NaP i -2, rat) are functionally (transport) and morphologically (immunofluorescence) located at the apical cell surface, are functionally inhibited in response to PTH addition, are retrieved in a PTH-dependent manner from the apical cell surface (immunofluorescence), and are subsequently degraded (Western blots). These data document that PTH control of the type II Na ϩ /P i cotransporters involves a step of membrane retrieval and degradation. Furthermore, the OK cell system should represent the ideal in vitro model to dissect the cellular/molecular mechanisms participating in regulation of this proximal tubular transport function, which is crucially involved in overall P i homeostasis.

EXPERIMENTAL PROCEDURES
Vectors-The generation of the vectors used (pLKneo and NaP i -2/ pLKneo) has been described previously (24,25). Both vectors code for a geneticin (G418) resistance under a SV40 promoter. In addition, the vector NaP i -2/pLKneo contains a cDNA coding for the rat renal type II Na ϩ /P i cotransporter (NaP i -2) under a dexamethasone-inducible promoter.
Monolayers on permeable filter supports were grown on Millicell-CM filter inserts (Millipore; 12-mm diameter, 0.45-m pore size) coated with a very thin film of rat tail collagen (R type, 0.5 mg/ml in 50% ethanol; Serva, Basel, Switzerland). Cells were seeded at approximately 1-2 ϫ 10 5 cells/filter and were re-fed with fresh medium every 12 h until reaching confluency 24 h after seeding. Transport studies were commenced 36 h after seeding. The tightness of confluent cell monolayers has been measured by resistance measurements as described previously (22).
For transfection, cells grown to a confluency of approximately 60% in 35 mm dishes (Nunc) were incubated for 16 h with 20 l of a 1:1 mixture of water and Lipofectin (Life Technologies) containing 10 g of either NaP i -2/pLKneo or the same amount of empty vector pLKneo. Afterward, cells were trypsinized, split at a ratio of 1:30, and grown in 150-mm dishes (Nunc) in medium containing, in addition, 400 g/ml active geneticin. After 1-2 weeks, colonies of geneticin-resistant cells were isolated by ring cloning, expanded, and analyzed for the expression of the NaP i -2 protein by immunoblotting (see below). For experimental purposes, transfected cells were used within 10 passages. Dexamethasone induction was performed by an incubation with 1 M dexamethasone (Sigma) for 20 h (10,000-fold stock, made in ethanol).
SDS-Polyacrylamide Gel Electrophoresis and Immunoblotting-Cells grown to confluency in 10-cm Petri dishes were incubated either with or without dexamethasone (1 M) for 20 h and washed twice with TBS (0.9% NaCl, 10 mM Tris-HCl, pH 7.4). 15 ml of TBS containing 4 mM EDTA and 1 mM phenylmethylsulfonyl fluoride was added, and the cells were scraped off the dish. The scraped cells were homogenized 5 times with a 20-ml syringe connected to a 20-gauge needle. This homogenate was centrifuged at 2000 rpm for 10 min at 4°C (Sorvall centrifuge, SS-34 rotor). The postnuclear supernatant was centrifuged at 16,000 rpm for 40 min at 4°C (Sorvall centrifuge, SS-34 rotor). The pellet corresponding to a crude membrane preparation was resuspended in 100 l of 50 mM mannitol, 10 mM Hepes-Tris (pH 7.2).
In experiments in which the PTH-mediated degradation of the Na ϩ /P i cotransporter was investigated, the total cell homogenate was centrifuged at 31,000 rpm (100,000 g) for 60 min at 4°C (Sorvall ultracentrifuge OTD 50B/T865 rotor) to ensure that all membranes were contained within the pellet. The pellet was resuspended in 200 l of 50 mM mannitol, 10 mM Hepes-Tris (pH 7.2). The protein concentration was determined by the Bio-Rad protein assay. 50 g of total protein were used for SDS-polyacrylamide gel electrophoresis (9%) and subsequent transfer to nitrocellulose (Schleicher & Schuell, Inc.; 0.45 m). Nonspecific binding was blocked by incubating the nitrocellulose at room temperature for 2 h in TBS (0.9% NaCl, 10 mM Tris-HCl, pH 7.4) containing 5% nonfat dry milk and 1% Triton X-100 (Blotto-TX-100, pH 7.4). The transfected NaP i -2 protein was detected using a polyclonal antiserum raised against the N terminus of the NaP i -2 protein (14) (antibody dilution, 1/4000). The intrinsic NaP i -4 protein was detected using a polyclonal antiserum raised against the C-terminal 12 amino acids of the published NaP i -4 sequence (8) (antibody dilution, 1/2000). Incubation with the primary antibody took place overnight at 4°C. The nitrocellulose was washed four times with TBS, 10% Blotto-TX-100 (pH 7.4) and incubated for 1 h with Blotto-TX-100 (pH 7.4) at room temperature. Then the nitrocellulose was incubated with a 1:10,000 dilution of an anti-rabbit IgG labeled with horseradish peroxidase (Amersham Life Science, Inc.) in Blotto-TX-100 (pH 7.4) for 2 h at room temperature. The nitrocellulose was washed four times with TBS, and the signals were detected by enhanced chemiluminescense (Amersham) according to manufacturer protocol using Kodak X-Omat AR films. For peptide protection assays, the corresponding antigenic peptide was included at a concentration of 100 g/ml. Broad range SDS-polyacrylamide gel electrophoresis molecular mass marker proteins (Bio-Rad) were run in parallel.
Immunofluorescence-NaP i -2-transfected and untransfected 3B/2 OK cells were grown to confluency on coverslips as well as on permeable filter supports, as described previously (22). After washing three times with PBS containing 0.5 mM MgCl 2 and 1 mM CaCl 2 , cells were fixed for 10 min at room temperature with PBS supplemented with 3% paraformaldehyde, washed three times with PBS, incubated 10 min with 20 mM L-glycine in PBS, and washed again (3 times) with PBS. Permeabilization was performed by an incubation for 30 min with PBS containing 0.1% saponin (PBS/saponin). After one wash with PBS/ saponin, cells were incubated with anti-NaP i -2 (14) or anti-NaP i -4 antiserum at a dilution of 1:100 in PBS/saponin for 1 h at room temperature and washed three times with PBS/saponin. Thereafter, the cells were incubated with a fluorescein isothiocyanate-conjugated IgG (Dakopatts, Denmark) (dilution, 1:50) and phalloidine rhodamine (Calbiochem) (dilution, 1:50) in PBS/saponin. After incubation for 30 min in the dark, cells were washed three times with PBS/saponin and once with PBS. Coverslips were mounted using Dako-Glycergel (Dakopatts) plus 2.5% 1,4-diazabicyclo-[2.2.2]octane (Sigma) as a fading retardant. Im-munofluorescence was revealed by confocal microscopy (Zeiss laser scan microscope 310; Zeiss, Oberkochen, Germany).
Scanning Electron Microscopy-OK cell monolayers grown on collagen-coated porous filter supports were prefixed with 0.25% glutaraldehyde in 0.16 M cacodylate buffer (pH 7.2) for 30 min at room temperature, postfixed with 2% glutaraldehyde in 0.16 M cacodylate buffer also for 30 min at room temperature, and washed three times with 0.16 M cacodylate buffer (pH 7.2) for 5-10 min. After that, the monolayers were osmicated with 1% OsO 4 in 0.16 M cacodylate buffer (pH 7.2) for 1 h at 37°C, washed three times with 0.16 M cacodylate buffer, dehydrated in an acetone series, and dried by the critical point method. The specimens were then examined in a scanning electron microscope 505 (Philips, Eindhoven).
Phosphate Uptake Measurements-Na ϩ -dependent and -independent transport of phosphate was determined in cells grown to confluency on either plastic dishes (35 mm; Nunc) or on permeant filter supports (8 mm), as described previously (22). Briefly, uptake solutions consisted of 137 mM NaCl, 5.4 mM KCl, 2.8 mM CaCl 2 , 1.2 mM MgSO 4 , 10 mM Hepes-Tris (pH 7.4), and 0.1 mM KH 2 32 PO 4 (1 Ci/ml). For Na ϩ -independant uptake, NaCl was replaced equimolarly by N-methyl-D-glucamine⅐HCl. Routine uptake on plastic dishes was performed at room temperature for 6 min and then stopped by washing the cells four times with ice-cold stop solution (137 mM NaCl, 10 mM Tris-HCl, pH 7.2). Cells were solubilized with 1% Triton X-100, and radioactivity was determined by liquid scintillation. Transport rates are expressed as nmol of P i taken up/mg of total cellular protein, which was determined by the Bio-Rad protein assay.
For transport assays on permeant filter supports, growth medium was aspirated, and both sides of the monolayer were gently rinsed twice in substrate-free uptake solution. Filter insert monolayers were then placed in a 24-well culture plate (Nunclon) for uptake measurements. Substrate-free uptake solution (500 l) was added to the appropriate filter insert compartment, and N-methyl-D-glucamine⅐HCl-uptake solution was added to the opposite compartment. Transport was initiated by mixing 50 l of the same uptake solution containing an 11-fold concentration of the desired radioactive substrate to the uptake solution already present in the filter insert compartment. Uptake was stopped as described above. Nonspecific binding (blanks) was assessed measuring zero time uptake in N-methyl-D-glucamine⅐HCl-uptake solution by starting uptake and immediately aspirating the uptake solution. Nonspecific binding was Ͻ10% that of radioactivity associated with any experimental point. Total radioactivity incorporated into the monolayer was measured by liquid scintillation counting of the whole filter insert.
PTH Incubation/Treatment with Cycloheximide-Incubation of OK cells with PTH has been described previously (22). Treatment of OK cells with cycloheximide to prevent protein synthesis has also been described previously (27).
Presentation of the Results-Statistical results are expressed as mean Ϯ S.E. for three dishes. Significance was accepted at p Ͻ 0.05. Experiments were repeated at least twice, and one representative experiment was choosen for presentation. The results presented concerning the transfected NaP i -2 were obtained with one single clone of NaP i -2-transfected 3B/2 OK cells. Qualitatively, the same results were obtained with two other clones of stably NaP i -2-transfected 3B/2 OK cells.

Characterization of OK Cells Stably Transfected with the Rat
Renal Type II Na ϩ /P i Cotransporter (NaP i -2)-To study transport function, transfected OK cells were grown to confluency on plastic Petri dishes and exposed to 1 M dexamethasone for 20 h to induce the expression of the transfected NaP i -2. As illustrated in Fig. 1, induction of NaP i -2 expression in transfected OK cells by dexamethasone led to an approximately 2-fold stimulation of the Na ϩ -dependent P i transport, whereas dexamethasone had no significant effect on the Na ϩ /P i cotransport activity in empty vector-transfected and in untransfected OK cells. Furthermore, dexamethasone had no effect in any of the tested cell lines on the Na ϩ -independent P i transport ( Fig.  1 and data not shown). Corresponding experiments have been carried out with NaP i -2-transfected OK cells grown to confluency on collagen-coated porous filter supports. By measuring the P i transport at the apical and basolateral membrane separately in induced and noninduced NaP i -2-transfected OK cells, it was found that the additional P i uptake, as observed in cells grown on Petri dishes (Fig. 1), is entirely restricted to the apical membrane (data not shown).
Transfected OK cells were also analyzed for expression of the NaP i -2 protein. Induction of NaP i -2-transfected OK cells by dexamethasone (1 M, 20 h) led to the expression of a protein with an apparent molecular mass of 95-120 kDa (Fig. 2). An excess of the corresponding antigenic peptide (100 g/ml) prevented the appearance of this band completely. The band seen above the NaP i -2 protein is unspecific: it is neither induced by dexamethasone nor protected by the antigenic peptide ( Fig.  2A). The intrinsic type II transporter protein (NaP i -4) was detected by the use of an antiserum directed against the C terminus of NaP i -4. Fig. 2B shows the NaP i -4-specific signal detected with membranes obtained from untransfected OK cells. Specificity of the anti-NaP i -4 antiserum was established by peptide protection assay with the corresponding antigenic peptide (100 g/ml; Fig. 2B). Comparing Fig. 2A with Fig. 2B shows that the intrinsic NaP i -4 and the transfected NaP i -2 have approximately the same molecular mass (95-120 kDa). Furthermore it is seen that the anti-NaP i -2 antiserum does not cross-react with the intrinsic NaP i -4 ( Fig. 2A, transfected cells in the absence of dexamethasone).
Expression of the type II transporter was also analyzed by immunofluorescence. Immunofluorescence pictures obtained by confocal microscopy showed that both the intrinsic NaP i -4 (Fig. 3A) and the transfected NaP i -2 (Fig. 3B) are localized at the apical membrane within distinct clusters of a diameter of about 1-2 m, whereas no type II Na ϩ /P i cotransporter-specific staining was seen at the basolateral membrane. The immunohistochemical staining of the transfected NaP i -2 as well as of the intrinsic NaP i -4 could be specifically abolished by the corresponding antigenic peptide (100 g/ml; data not shown). In Figs. 3A and 3B, the parallel staining for ␤-actin (a component of the microvillar cytoskeleton) is shown. It is apparent that NaP i -4-specific (Fig. 3A) and NaP i -2-specific (Fig. 3B) staining on the apical cell surface coincides with the ␤-actin staining. In addition to the above apical staining, ␤-actin is also present at the basolateral cell surfaces (Fig. 3A and 3B). Visualization of the apical surface of OK cells by scanning electron microscopy showed that microvilli are expressed at the apical surface, forming distinct clusters (Fig. 3C); the diameter of the clusters

in untransfected OK cells (A) and ␤-actin and NaP i -2 in dexamethasone-induced NaP i -2-transfected OK cells (B). It is seen that ␤-actin and NaP i -4 and ␤-actin
and NaP i -2 colocalize within distinct clusters at the apical membrane. Scanning electron microscopy demonstrates the presence of clustered microvilli at the apical surface of OK cells (C). The diameter of these clustered microvilli corresponds well with the diameter of the distinct clusters at the apical membrane seen in the immunofluorescence pictures in Fig. 3, A and B.   FIG. 1. Dexamethasone-induced expression of Na ؉ /P i cotransport in NaP i -2-transfected OK cells. OK cells (NaP i -2-transfected, empty vector-transfected, and untransfected OK cells) were grown to confluency on plastic Petri dishes and, where indicated (DEX, ϩ), were exposed to 1 M dexamethasone for 20 h. Dexamethasone-induced NaP i -2-transfected OK cells showed an approximately 2-fold stimulation of the Na ϩ /P i cotransport activity, whereas dexamethasone had no significant effect on the intrinsic Na ϩ /P i cotransport in empty vectortransfected and untransfected OK cells.
corresponds well with the diameter of the clusters seen by immunofluorescence double staining for ␤-actin and the corresponding type II Na ϩ /P i cotransporter. Therefore the data given in Fig. 3 (A-C) documents that intrinsic and transfected type II Na ϩ /P i cotransporters are predominantly expressed at the apical cell surface (most likely within microvilli).
PTH-dependent Regulation of Type II Na ϩ /P i Cotransporters- Fig. 4 summarizes the effect of PTH (10 Ϫ8 M; 4 h) on the Na ϩ -dependent P i transport in NaP i -2-transfected OK cells as well as in untransfected OK cells. PTH inhibited the Na ϩ -dependent P i transport activity in control cells by about 60%. A similar inhibitory effect of PTH was observed in NaP i -2-transfected cells that were not treated with dexamethasone. In cells expressing the NaP i -2 transporter (induced by dexamethasone), PTH also inhibited the additionally expressed Na ϩ /P i cotransport activity. Interestingly, the residual transport activity after PTH treatment was similar in all cells tested. This latter observation is in agreement with earlier studies (21,27) demonstrating a PTH-insensitive Na ϩ /P i cotransport activity in OK cells.
The effect of PTH on the type II Na ϩ /P i cotransporter protein content was investigated by immunoblotting. Fig. 5 shows that incubating OK cells for increasing times with PTH (10 Ϫ8 M) leads to a time-dependent decrease of NaP i -4 and NaP i -2 protein expressed in OK cells. We conclude that the whole protein was degraded due to PTH action and that both transfected (NaP i -2) and intrinsic transporter (NaP i -4) behave very similar, i.e. are degraded. The finding that PTH leads to the degradation of the type II Na ϩ /P i cotransporter is in agreement with our previous observation that the recovery of the Na ϩ /P i cotransport after PTH-mediated inhibition is dependent on de novo protein synthesis (27,28). To evaluate whether the recovery of the transport was due to de novo synthesis of type II Na ϩ /P i cotransporter, cell monolayers were exposed to 10 Ϫ8 M PTH for 4 h, PTH was removed, and cell monolayers were incubated with normal medium in the presence or absence of 20 M cycloheximide for 2 or 4 h. The results presented in Fig. 6 show that the reappearance of the transporter is dependent on de novo protein synthesis. It is also seen that 4 h after PTH removal, the reappearance of the transporter is not completed. Parallel measurements of the Na ϩ /P i cotransport showed that after 4 h of PTH removal, the recovery of the Na ϩ /P i cotransport is also incomplete. This is in agreement with our previous observation that complete recovery of Na ϩ /P i cotransport from PTH inhibition takes about 10 -12 h (28).
The retrieval of the intrinsic (NaP i -4) and the transfected cotransporter (NaP i -2) upon addition of PTH could also be documented by immunofluorescence. Consistent with the data presented above (Figs. 4 -6), we found that incubating OK cells with PTH (10 Ϫ8 M) leads to a time-dependent decrease of NaP i -4-dependent immunofluorescence and NaP i -2-dependent immunofluorescence at the apical membrane of OK cells (data not shown). After 4 h of PTH treatment (10 Ϫ8 M), the Na ϩ /P i cotransporter type II-specific immunofluorescence staining in untransfected (Fig. 7A) and NaP i -2-transfected OK cells (Fig. 7B) was virtually absent. Corresponding immunofluorescence pictures (double immunofluorescence) stained for ␤-actin clearly showed that under these conditions the microvilli were still present at the apical surface of OK cells. With the aim to detect the retrieved Na ϩ /P i cotransporters within intracellular vesicles, we have treated OK cells for increasing times with PTH (0, 0.5, 1, 2, and 4 h; data not shown). As mentioned above, we observed a time-dependent decrease of the NaP i -4-specific staining and NaP i -2-specific staining at the apical surface. However, we were not able to detect a significant amount of NaP i -4 protein or NaP i -2 protein within intracellular vesicles at any time points (data not shown). As already suggested by the experiments presented in Fig. 5, a delay between a PTHdependent membrane retrieval and intracellular degradation seems to be absent or minimal. FIG. 6. Reappearance of the NaP i -4 protein after PTH removal requires de novo protein synthesis. Confluent OK cell monolayers were incubated for 4 h with PTH (10 Ϫ8 M). Then the culture medium was removed, cells were repeatedly washed with normal medium to remove the PTH, and cells then incubated with medium with or without 20 M cycloheximide for 2 and 4 h, respectively. The above Western blot shows clearly that the reappearance of the transporter is dependent on de novo protein synthesis.

DISCUSSION
Previous studies on rats have suggested that inhibition of Na ϩ /P i cotransport in renal proximal tubules by PTH involves a retrieval of the type II Na ϩ /P i cotransporter (NaP i -2) from the brush border membrane (15). The recent cloning of the type II Na ϩ /P i cotransporter of OK cells (NaP i -4; Ref. 8) and the obvious limitations of an in vivo system to study molecular mechanisms prompted us to investigate the regulation of this transporter by PTH in OK cells. OK cells are a renal epithelial cell line that has been used successfully to investigate mechanisms involved in the control of proximal tubular Na ϩ /P i cotransport by PTH (21,27). In the present study, we stably transfected OK cells with a cDNA coding for the rat type II Na ϩ /P i cotransporter (NaP i -2) and investigated whether similar mechanisms are involved in the PTH-mediated regulation of the intrinsic (NaP i -4) and the transfected transporters (NaP i -2).
On immunoblots, appearance of the transfected (NaP i -2) and the intrinsic (NaP i -4) cotransporter proteins resembled each other closely. The broad staining pattern (95-120 kDa) of these transporters likely represents different degrees of glycosylation, which for the NaP i -2 protein, has recently been demonstrated (29). The expression of the rat type II Na ϩ /P i cotransporter was also demonstrated by transport experiments. Dexamethasone-induced NaP i -2-transfected OK cells exhibited an approximately 2-fold-stimulated transport activity compared with uninduced NaP i -2-transfected or control cells (untransfected and empty vector-transfected). Immunofluorescence experiments demonstrated an exclusive apical localization for both Na ϩ /P i cotransporters (intrinsic and transfected), that, in the case of the intrinsic NaP i -4, is in agreement with earlier transport studies performed with OK cells grown on permeant filter supports (22). Furthermore, immunofluorescence demonstrated a concentration of the intrinsic as well as of the transfected Na ϩ /P i cotransporters within clustered microvillar structures. It is suggested that such a distinct concentration of the Na ϩ /P i cotransporters within the microvilli may be due to a yet unknown interaction of these transporters with components of the microvillar cytoskeleton. Clearly no immunostaining related to these transporters was detected in the basolateral membrane, suggesting a specific apical sorting mechanism for the type II Na ϩ /P i cotransporters in OK cells. In contrast, we recently demonstrated that the NaP i -2, when transfected into Madin-Darby canine kidney cells, is expressed to equal amounts at the apical and basolateral membrane (25). The observed different sorting behavior of the same transporter in two different cell lines indicates that not only molecular determinants are decisive for a polarized sorting but that also the cellular context is of importance. This has also been documented for other proteins and cell systems. For example, it has been shown that aquaporin-2, when transfected into LLC-PK 1 cells, is, upon cAMP stimulation, inserted into the basolateral membrane rather than into the apical membrane as suggested from studies on native renal epithelia (30).
In previous studies it has been demonstrated that the intrinsic apical Na ϩ /P i cotransport activity in OK cells is inhibited by PTH (21,27). In the present study we extended this observation by showing that the inhibition of the Na ϩ /P i cotransport by PTH in OK cells is paralleled by the disappearance of the apical, Na ϩ /P i cotransporter-specific immunofluorescence staining within the microvilli. It is also shown that PTH leads to the degradation of the type II Na ϩ /P i cotransporter in OK cells. This finding is in agreement with our earlier observation demonstrating that the recovery of Na ϩ /P i cotransport activity from PTH inhibition in OK cells is dependent on de novo protein synthesis. Correspondingly, we found that the recovery of Na ϩ /P i cotransport activity from PTH inhibition was paral- leled on immunoblots by the reappearance of the type II Na ϩ /P i cotransporter protein. In the presence of cycloheximide, there was neither a recovery of transport activity nor a reappearance of the specific transporter protein. As documented by immunoblots and by immunohistochemical studies presented in this paper, PTH leads to an almost complete retrieval and degradation of the type II Na ϩ /P i cotransporter (Figs. 5 and 7). However, even after prolonged exposures to PTH, a "refractory" residual Na ϩ /P i cotransport activity is observed (Fig. 4 and Ref. 21). This implies that this residual activity (ϳ40% of total transport activity) is not related to the type II Na ϩ /P i cotransporter but could be associated with, for example, the type I transporter. Such conclusions are also valid for the Na ϩ /P i cotransport activity in rat brush border membranes; 2 h of PTH treatment leads to a much higher reduction of the NaP i -2 protein compared with the reduction in Na ϩ /P i cotransport activity (15).
The present study clearly indicates that the regulation of the type II Na ϩ /P i cotransporter in renal proximal tubules and in OK cells is very similar. One apparent difference exists: after PTH treatment of OK cells for various lengths (0.5, 1, 2 , and 4 h) we were not able to detect a significant transient intracellular staining specific for the type II Na ϩ /P i cotransporter. In rat proximal tubules, the degradation of the NaP i -2 protein seems to be delayed (compared with the OK cell system), permitting a visualization of an increased intracellular NaP i -2 protein content after short (15 min to 1 h) but not after prolonged treatments with PTH (15). Despite this difference, the similarities in the regulation of this transporter by PTH in OK cells and in rat renal proximal tubules are evident. Furthermore, in this report we demonstrated that the transfected (NaP i -2) and intrinsic type II Na ϩ /P i cotransporters (NaP i -4) are regulated in OK cells in the same way by PTH. OK cells are therefore a physiologically relevant in vitro system for the study of the regulation of the type II Na ϩ /P i cotransporter type II. They are a useful tool to dissect the molecular/cellular mechanisms involved in the PTH-mediated internalization and subsequent degradation of the transporter. Although it is tempting to assume a final breakdown of these Na ϩ /P i cotransporters within lysosomes, the mechanisms involved in internalization, trafficking, and subsequent degradation are completely unknown.
In summary, the present study shows (i) that the intrinsic (NaP i -4) and the transfected rat type II Na ϩ /P i cotransporter (NaP i -2) are functionally and morphologically localized at the apical membrane, (ii) that the intrinsic as well as the transfection-mediated (NaP i -2) Na ϩ /P i cotransport activities are inhib-ited by PTH, (iii) that PTH leads to the disappearance of both cotransporters from the apical membrane, and (iv) that both cotransporters are rapidly degraded in response to PTH. These results suggest that PTH leads to the endocytosis of the type II Na ϩ /P i cotransporters from the apical membrane and to their subsequent degradation, thereby leading to the down-regulation of the Na ϩ /P i cotransport.