Role of internalization in AT 1A receptor function in proximal tubule epithelium

Transcription

1 Am J Physiol Renal Physiol 282: F623 F629, 2002; /ajprenal Role of internalization in AT 1A receptor function in proximal tubule epithelium THOMAS THEKKUMKARA AND STUART L. LINAS Department of Medicine, University of Colorado Health Sciences Center, Denver Health Medical Center, Denver, Colorado Received 12 April 2001; accepted in final form 5 October 2001 Thekkumkara, Thomas, and Stuart L. Linas. Role of internalization in AT 1A receptor function in proximal tubule epithelium. Am J Physiol Renal Physiol 282: F623 F629, 2002; /ajprenal Angiotensin II (ANG II), acting through angiotensin type I (AT 1) receptors on apical and basolateral surfaces of proximal tubule epithelial cells, increases sodium reabsorption in proximal tubules. Apical and basolateral receptors internalize after exposure to ANG II, but the role of internalization in receptor signaling and transport is not well defined. To determine the role of receptor internalization in ANG II-mediated receptor signaling and sodium transport, we stably expressed full-length and truncated AT 1A receptors in opossum kidney cells. After stimulation with ANG II, wild-type receptors on apical and basolateral surfaces rapidly internalized, inhibited adenylate cyclase, and increased transcellular sodium transport. Truncation of the cytoplasmic tail of the AT 1A receptor (TL314) resulted in receptors that were expressed on apical and basolateral surfaces but did not internalize, inhibit adenylate cyclase, or increase sodium transport. Because the cytoplasmic tail contains putative G protein coupling sites, mutant receptors that leave G protein interaction sites intact were designed. Cells expressing the truncation (TK333) or deletion (Del ) also failed to internalize. When ANG II was added to basolateral surfaces of TK333 or Del , adenylate cyclase activity was inhibited and sodium transport was increased. In contrast, apical addition of ANG II was not associated with decreases in adenylate cyclase or increases in sodium transport. In conclusion, internalization pathways are important for AT 1A receptor function in polarized proximal tubule epithelial cells. Apical AT 1A receptors internalize before they interact with G proteins and signal camp. In contrast, basolateral AT 1A receptors interact with G proteins and signal camp without internalizing. sodium transport; angiotensin; kidney function; hemodynamics ANGIOTENSIN II (ANG II) has multiple effects on the cardiovascular system that are mediated through regulation of renal sodium reabsorption. Although ANG II-dependent changes in sodium reabsorption are mediated by ANG II-dependent increases in systemic and renal vascular tone and aldosterone secretion, ANG II directly regulates proximal tubule sodium reabsorption (17, 24). The proximal tubule effects of ANG II are transduced by angiotensin type 1 (AT 1 ) receptors. In Address for reprint requests and other correspondence: S. L. Linas, Dept. of Medicine, Box 4000, Denver Health Medical Center, 777 Bannock St., Denver, CO ( Stuart.Linas@UCHSC.edu). rats, two isoforms of AT 1 receptors have been identified: AT 1A and AT 1B (13). All the effects of ANG II in the proximal tubule have been shown to be mediated by AT 1A or AT 1B receptors (9). For example, volume regulation is normal in AT 1A knockout mice (6) but severely impaired in angiotensinogen knockout mice (19). These receptors are expressed on the apical and basolateral membranes of the proximal tubule (4, 5, 10). ANG II binding to apical or basolateral AT 1A receptors results in inhibition of adenylate cyclase and increases in transcellular sodium reabsorption (16, 27). Apical and basolateral membranes of the proximal tubule are composed of different lipids, signaling enzymes, and transporters. For example, basolateral membranes contain adenylate cyclase and sodium-potassium-atpase, whereas apical membranes contain phospholipase (PL) A 2 (3) and the NHE 3 isoform of the sodium/hydrogen exchanger. Apical and basolateral AT 1A receptors undergo internalization (2, 7). Earlier studies in cultured proximal tubules indicate that pharmacological agents that block cellular pathways of internalization decreased apical receptor-dependent increases in PLC, PLA 2, and sodium transport (3, 22). In contrast, internalization blockers had no effect on basolateral AT 1A receptor function (22). Because the pharmacological inhibitors of internalization interfere with cellular structures that regulate internalization, it was not known whether 1) AT 1A receptors, rather than cell structures, regulate internalization-dependent signaling pathways and 2) internalization was required for apical and basolateral AT 1A receptor function. The AT 1A receptor has been cloned, and the domains, which regulate receptor internalization, have been identified by mutation analysis. When internalizationmutant receptors were expressed in a number of different cells under nonpolarizing conditions, no identifiable changes in AT 1A receptor function were reported (30). For example, noninternalizing AT 1A mutants activate PLC and desensitize in transfected Chinese hamster ovary cells grown under nonpolarizing conditions. Because internalization pathways are not used for AT 1A receptor function under nonpolarizing condi- The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact /02 $5.00 Copyright 2002 the American Physiological Society F623

2 F624 AT 1A RECEPTORS IN KIDNEY tions and because earlier pharmacological studies indicated that internalization pathways were critical to apical AT 1A receptor-mediated increases in sodium transport in cultured polarized proximal tubule epithelium, the goal of this study was to determine the role of internalization on AT 1A receptor signaling and sodium transport in polarized tissue. Because rat proximal tubule cells express AT 1A and AT 1B receptors, to determine AT 1A receptor function without interference by other subtypes, we utilized an immortalized proximal tubule cell line, opossum kidney (OK), which does not normally express angiotensin receptors. OK cells were transfected with wild-type and mutant AT 1A receptors to answer our questions. MATERIALS AND METHODS Materials. OK cells were obtained from Dr. John Raymond (Medical University of South Carolina). Dulbecco s modified Eagle s medium-ham s F-12 nutrient mixture, trypsin, geneticin, bovine calf serum, and antibiotics were obtained from Life Technologies (Gaithersburg, MD). Tissue culture dishes, flasks, plastic ware, and Transwell cell culture inserts were purchased from Costar (Cambridge, MA). Radioiodinated ANG II was purchased from New England Nuclear (Boston, MA). ANG II was obtained from Sigma Chemical (St. Louis, MO). Candesartan was a generous gift from Astra Pharmaceutical (Wayne, PA). The camp assay kit was purchased from Amersham Life Science (Arlington Heights, IL). Taq polymerase and PCR reagents were obtained from Perkin- Elmer (Norwalk, CT). All other molecular biology-grade chemicals were obtained from US Biochemical (Cleveland, OH), Fisher Scientific (Pittsburgh, PA), or Sigma Chemical. Stable expression of the rat AT 1A receptor in OK epithelial cells. The cloning and isolation of a genomic clone for the rat AT 1A receptor, construction of the full-length receptor (coding for 359 amino acids), and construction of COOH-terminal truncated (TL314 and TK333) and deleted (Del ) receptors into the prc/cmv eukaryotic expression vector have been previously described (28, 30). Stable expression of the full-length AT 1A receptor in OK cells has also been previously described (27). To develop mutant receptor-expressing cell lines, 50% confluent OK cells were transfected with 1 2 g of plasmid DNA using the Lipofectamine method as described by the manufacturer (Life Science). Neomycin (G418)-resistant individual clones were isolated, and cell surface receptor expression was determined by 125 I-ANG II binding studies. Cells stably expressing the receptor were maintained under a selection pressure of 200 g/ml G418. Multiple clones expressing full-length and mutant receptors were isolated, and passages 6 30 were used. The results for individual clones were also confirmed in multiple clones. ANG II receptor binding and internalization studies. Receptor transfected and nontransfected OK cells were grown to confluence on 6.5-mm-diameter, 0.4- m-pore permeable Transwell membrane supports for differential binding to apical or basolateral surfaces (21). Confluence was confirmed by the development of a transepithelial resistance and barrier to transport of [ 3 H]inulin (21). Assay buffer consisted of 50 mm Tris, ph 7.4, 100 mm NaCl, 5 mm MgCl 2, 0.25% bovine serum albumin, and 0.5 mg/ml bacitracin. Binding assays were performed in triplicate. For each experiment, cells were washed twice with ice-cold PBS, and 30 l of binding buffer or unlabeled ANG II were added to apical or basolateral surfaces, and then 270 l of binding buffer containing nm 125 I-ANG II were added. Equilibrium binding was performed for 180 min at 4 C for apical or basolateral surface binding. Incubations were terminated by rapid removal of incubation solutions and addition of ice-cold PBS. Cells were washed three times with PBS and dissolved in 200 l of 0.2 N NaOH. Solutions were transferred to mm disposable tubes, or inserts were placed directly into tubes, and radioactivity was determined using a Packard Auto-Gamma counter. Results are expressed as specific binding. Specific binding was defined as total binding minus nonspecific binding (in the presence of unlabeled 1 M ANG II). Nonspecific binding was 5% of total binding. Values were normalized to the amount of protein determined using the Bio-Rad DC protein assay. Results were analyzed using the computer software GraphPad Prism, and binding constants were determined as described elsewhere (26). To determine the rate of apical and basolateral receptor internalization, cells were grown to confluence on membrane inserts and assayed as described previously (28). Briefly, cells were washed twice with PBS, covered with binding buffer, and incubated at 37 C for 15 min. 125 I-ANG II, in 30 l of binding buffer, was added to apical or basolateral buffer (300 l total volume) in a final concentration of 0.05 nm, and incubation continued at 37 C for 2, 5, 10, and 20 min. At each time point, inserts were chilled on ice, and radioactivity associated with noninternalized receptors was removed by two 40-s washes with 5 mm ice-cold acetic acid in 150 mm NaCl, ph 2.5. Radioactivity in the acid-sensitive and -insensitive fractions was measured as described above. The percentage of internalized receptors was plotted against time, curves were analyzed, and the half time to reach maximal level of internalization was determined using GraphPad Prism. Measurement of camp accumulation. Cells were grown in 24-well Transwell plates in Dulbecco s modified Eagle s medium-ham s F-12 nutrient mixture supplemented with 10% bovine calf serum, as previously described (27). Cells were washed twice with assay buffer containing 50 mm Tris, ph 7.4, 100 mm NaCl, 6 mm Na 2HPO 4,4mMKH 2PO 4, 0.25% bovine serum albumin, and 0.5 mg/ml bacitracin, and apical or basolateral surfaces were exposed to ANG II for 7 min in the same buffer (250 l) with 5 mm 3-isobutyl-1-methylxanthine, a phosphodiesterase inhibitor. At the end of each reaction, cells were treated with 250 l of 0.02 N HCl at 4 C for 20 min. The solution was collected, and camp concentrations were determined by radioimmunoassay as previously described for rat proximal tubule epithelial cells in culture (23). Total protein was estimated by Bio-Rad DC protein assay, and camp accumulation was expressed as picomoles per milligram of protein. Measurement of sodium flux. Unidirectional apical-to-basolateral sodium flux was determined by a modification of methods previously described by our laboratory (22). Cells were grown to confluence on polycarbonate membrane inserts. After cells were rinsed with PBS, 500 l of preincubation buffer containing 136 mm NaCl, 5 mm NaHCO 3,5mM KCl,4mMNa 2HPO 4, 1.0 mm MgSO 4, 5.0 mm urea, 5.4 mm glucose, and 2 mm glutamine (296 mosm, ph 7.4) were added to the apical side and 3 ml of the buffer were added to the basolateral side. The cells were incubated at 37 C in the presence of 5% CO 2 for 1 h. Preincubation buffer was removed, and inserts were placed in the transport chamber. 22 Na (l Ci/ml) was added to the apical chamber in transport buffer, which contained 128 mm NaCl, 5 mm Na HCO 3,5 mm KCl, 4 mm Na 2HPO 4, 1 mm MgSO 4, 5 mm urea, 5.4 mm glucose, 2 mm glutamine, and 20 mm HEPES (305 mosm, ph 7.4). Basolateral surfaces were exposed to the same buffer without 22 Na. Initiation of transport was defined as the time

3 AT 1A RECEPTORS IN KIDNEY F625 Fig. 1. Time dependence of ANG II-stimulated receptor internalization from apical (AP) and basolateral (BL) surfaces of opossum kidney (OK)/angiotensin type 1A (AT 1A) cells. OK cells stably expressing AT 1A receptors were grown to confluency. Confluency was confirmed by development of transepithelial resistance and barrier to transport of [ 3 H]inulin. 125 I-ANG II was added to apical or basolateral buffers and incubated for 2 30 min at 37 C. At indicated times, surface-bound and internalized 125 I was determined by acid washing. Internalization was calculated by expressing acid-resistant (internalized) radioactivity as a percentage of acid-resistant plus acid-susceptible (total binding) radioactivity. Values are means of triplicate determinations from 3 separate experiments. when apical buffer was added. Aliquots of 100 l were removed from the basolateral buffer 1 20 min after initiation of transport. In preliminary studies, we found that 22 Na transport was linear for up to 30 min after addition of agonist. 22 Na was quantified by a liquid scintillation counter (Packard Tri-Carb), and 22 Na transport from apical to basolateral buffer was determined. Statistics. Values are means SE. Comparison between two groups was made by unpaired Student s t-test and for more than two groups by one-way analysis of variance, followed by Tukey s test for comparison of group means. Statistical significance is defined as P cdna encoding wild-type and mutated AT 1A receptors were subcloned into the prc/cmv mammalian expression vector and transfected into OK cells. Colonies resistant to G418 were selected and propagated. In preliminary studies, we sought to determine the minimum sequences required for cell surface expression of AT 1A receptors. Transfection with TL306 failed to yield cell surface AT 1A receptors, as indicated by kinetic analysis of binding: 10 fmol/mg for apical and basolateral surfaces. In contrast, transfection with TL314 resulted in stable expression of AT 1A receptors. As shown in Table 1, wild-type and TL314-mutated receptors bound ANG II with high affinity on apical and basolateral cell surfaces. On apical membranes, the number of binding sites tended to be less ( to ) and the binding affinity lower ( to nm) in the mutated than in the wild-type receptor ( fmol/mg and nm), although these values did not reach statistical significance. On basolateral membranes, the number of binding sites was similar ( to 1,065 87) and the binding affinities were similar ( to nm) in mutated and wild-type receptors ( fmol/mg and nm). The level of cell surface receptor expression and binding affinity are comparable to values obtained for membrane preparations of rat proximal tubules (4, 5). To determine whether wild-type and mutated AT 1A receptors were internalized, we measured the appearance of acid-resistant radioactivity after addition of 125 I-ANG II at 37 C. Figure 1 demonstrates internalization of wild-type and TL314-mutated receptors from apical and basolateral surfaces. As we reported earlier, 70% of apical and basolateral wild-type receptors were internalized within 20 min (27): 70 5 and 82 9% on apical and basolateral surfaces, respectively. In contrast, after 20 min, 15% of TL314-mutated apical (10 2%, P 0.01) and basolateral (14 3%, P 0.01) receptors were acid resistant and internalized. These data suggest that the cytosolic tail of the AT 1A receptor is required for AT 1A receptor internalization from apical and basolateral sites. To determine whether mutated AT 1A receptors were functional, we measured AT 1A receptor coupling to adenylate cyclase (Fig. 2A) and to sodium transport (Fig. 2B). Figure 2 indicates that ANG II inhibited adenylate cyclase when added to apical (39 7% vs. no additions) or basolateral (42 8% vs. no additions) surfaces of cells transfected with wild-type receptor. There were time-dependent decreases in adenylate cyclase that were prevented by prior addition of the AT 1A receptor blocker candesartan (95 3 and 93 4% for RESULTS Fig. 2. ANG II-mediated inhibition of camp (A) and stimulation of transcellular sodium transport (B) inat 1A receptor-transfected OK cells. OK cells stably expressing AT 1A receptor constructs were grown to confluency on inserts. Apical and basolateral receptors were exposed to 100 nm ANG II. camp or apical-to-basolateral 22 Na transport was determined as described in MATERIALS AND METHODS. Values are means SE of triplicate determinations from 3 separate experiments.

4 F626 AT 1A RECEPTORS IN KIDNEY apical and basolateral surfaces, respectively, n 4in each group) and were maximal after 10 min. In contrast, ANG II did not inhibit adenylate cyclase from apical (94 6%) or basolateral (106 8%) sites in cells transfected with TL314-mutated receptor. Furthermore, higher concentrations of ANG II and longer duration of ANG II exposure were not associated with reductions in camp. As reported earlier, ANG II caused increases in transcellular sodium transport when added to apical (197 16%) or basolateral (214 18%) surfaces of cells transfected with the AT 1A receptor (27). Increases in sodium transport were blocked by prior addition of candesartan: and 111 3% for apical and basolateral surfaces, respectively [n 4 in each group, P not significant (NS) vs. no additions]. In contrast, addition of ANG II did not result in increases in sodium transport when added to apical (106 14%) or basolateral (109 11%) surfaces of cells transfected with TL314-mutated receptors. The results indicated that removing the cytosolic tail of the AT 1A receptor resulted in decreases in G protein coupling and sodium transport. Because truncation mutation of the entire cytosolic tail altered G protein coupling, as well as internalization, we utilized two additional mutated receptors that do not contain putative G protein coupling sites: TK333 and Del As shown in Table 1, both of these mutated receptors expressed high-affinity binding sites for ANG II. Figure 3 depicts internalization data for mutated receptors. Compared with wild-type receptors (64 4 and 72 5% for apical and basolateral surfaces, respectively), both mutants displayed marked decreases in the amount internalized: for TK333, 32% for the apical surface (P 0.01) and 27% for the basolateral surface (P 0.01) and for Del , 30% for the apical surface (P 0.01) and 24% for the basolateral surface (P 0.01) after 20 min. Decreases in internalization from apical, but not basolateral, sites resulted in functional defects of apical, but not basolateral, AT 1A receptors. Figure 4 indicates that basolateral exposure to ANG II resulted in reduction in camp in wild-type AT 1A receptors (45 6% of Fig. 3. Percentage of ANG II-stimulated receptor internalization from apical and basolateral surfaces of OK cells stably expressing wild-type, TK333, or Del AT 1A receptors. 125 I-ANG II was added to apical and basolateral buffers, and buffers were incubated for 30 min at 37 C. Internalized ANG II was calculated as described in Fig. 1 legend. Values are means of triplicate determinations from 3 experiments in each group. Fig. 4. ANG II-mediated inhibition of camp (A) and stimulation of transcellular sodium transport (B) inat 1A receptor-transfected OK cells. OK cells stably expressing AT 1A receptor constructs were grown to confluency on inserts. Apical and basolateral receptors were exposed to 100 nm ANG II. camp or apical-to-basolateral 22 Na transport was determined as described in MATERIALS AND METHODS. Values are means SE of triplicate determinations from 3 separate experiments. no ANG II) as well as mutant AT 1A receptors: 41 5 and 56 5% for TK333 and Del , respectively (both P NS vs. wild-type AT 1A ). Figure 4B indicates that basolateral ANG II increased sodium transport by % in wild-type AT 1A receptors and % and % in TK333 and Del mutants, respectively (both P NS vs. wild-type AT 1A ). In contrast (Fig. 4A), apical ANG II-dependent decreases in camp did not occur with either mutant receptor: 42 6% for wild-type AT 1A receptor and and % for TK333 and Del , respectively (both P 0.01 vs. wild-type AT 1A ). In association with camp responses, Fig. 4B indicates that apical ANG II-dependent increases in sodium flux did not occur with either mutant receptor: % for wild-type AT 1A receptor and 97 7 and 94 13% for TK333 and Del , respectively (both P 0.01 vs. wild-type AT 1A ). DISCUSSION The results of our study indicate that 1) the COOHterminal tail of the AT 1A receptor contains sites that mediate agonist-dependent internalization, inhibition of adenylate cyclase, and increases in transcellular

5 AT 1A RECEPTORS IN KIDNEY F627 sodium transport and 2) receptor internalization is required for apical, but not basolateral, AT 1A receptor inhibition of adenylate cyclase and increases in transcellular sodium transport. The cytosolic tail of a number of cell surface receptors often regulates internalization. Whereas some receptors use a single segment, e.g., a NPXY motif, in the tail to drive internalization ( 2 -adrenergic low-densitylipoprotein receptors) (1, 31), other receptors use two motifs, one more proximal and the other distal, in the tail for internalization. Earlier studies indicate that two distinct regions in the cytosolic tail of the AT 1A receptor are important for internalization. These motifs are at Leu 316 to Tyr 319, hydrophobic and aromatic residues on a putative -helix, and Ser 335 to Leu 337, which contains an ST domain (11, 12, 29). AT 1A receptors are internalized in polarized epithelial as well as nonpolarized tissues. Apical and basolateral receptor internalization occur at different rates (2). Our data indicate that apical and basolateral AT 1A receptor internalization are regulated by the same motifs. When the and sites were removed (TL314 truncation), neither apical nor basolateral receptors internalized. Moreover, when the ST motif alone was removed (TK333 truncation), internalization was reduced by 80%. The ST site alone was not sufficient to drive internalization, because Del caused an 80% reduction in internalization. These data indicate that the and motifs are required for internalization in epithelial tissues. The data also indicate that the cellular apparatus that interacts with AT 1A receptors is the same for apical and basolateral receptors and that specific motifs on the receptor drive internalization and interaction with signal transduction pathways. We also found that, in contrast to wild-type receptors, the TL314 truncation failed to inhibit adenylate cyclase or increase sodium transport when ANG II was added to the apical or the basolateral surface. We considered the possibility that this finding resulted from the aforementioned decreases in internalization but rejected this interpretation, because in earlier studies we showed that noninternalizing basolateral receptors were able to increase PLC and sodium transport (21, 22). Alternatively, the TL314 truncation could have been rendered nonfunctional as a result of a decrease in AT 1A receptor coupling to G proteins, rather than decreases in internalization. A number of investigators have reported that the proximal region of the AT 1A receptor cytosolic tail is important for G protein interactions. For example, Shirai et al. (25) found that a synthetic peptide containing residues bound GTP analogs; Kai et al. (14) reported that a residue inhibited ANG II-dependent GTPase activation, and Franzoni et al. (8) showed that the stretch between 306 and 320 was required to maintain the three-dimensional structure for AT 1A receptors to interact with G proteins. The G protein coupling the AT 1A receptor to signaling pathways has been identified. For example, G q couples the receptor to PLC activation. Sano et al. (20) demonstrated that deletion mutations T309, T312, T313, and T314, but not T317, failed to activate PLC. They concluded that Tyr 312, Phe 313, and Leu 314 were essential for coupling and activation of G q.g i couples AT 1A receptors to inhibition of adenylate cyclase. However, there is limited information on the domain of the receptor that mediates this interaction. Whereas Shirai et al. (25) found that residues were able to activate GTPase activity of G 1 proteins, there are no reports on adenylate cyclase activity. Because the TL314 truncation failed to decrease adenylate cyclase while the Del decreased adenylate cyclase, our data suggest that the domain may be important for G i, as well as G q, coupling. The results also provide insight into AT 1A receptor targeting to apical and basolateral membranes. Whereas internalization and/or G protein coupling domains regulate targeting of some receptors, AT 1A receptor cell surface expression does not depend on these mechanisms. In this regard, the TL314 deletion mutation targeted to apical and basolateral surfaces (Table 1), even though the mutant did not internalize or signal. In other studies (data not shown), we found no expression of the TL306 mutation on apical or basolateral surfaces. These data indicate that sites between L306 and L314 regulate targeting. Our study further defines the role of adenylate cyclase in mediating ANG II-dependent increases in proximal tubule sodium reabsorption. In microperfused proximal tubules, Liu and Cogan (18) showed that ANG II decreased camp formation and increased sodium transport and that exogenous camp prevented this effect. Earlier we demonstrated that ANG II decreased adenylate cyclase and increased sodium transport in proximal tubule epithelial cells in culture and that exogenous camp blocked this effect as well (23). However, because both studies used high concentrations of exogenous camp, alternative interpretations were possible. Our present studies using a loss-offunction strategy indicate that camp decreases transduce AT 1A receptor-mediated sodium transport in the proximal tubule. Mutations that uncoupled basolateral receptors from adenylate cyclase (TL314) prevented ANG II-mediated increases in sodium transport, whereas mutations that did not uncouple basolateral AT 1A receptors from adenylate cyclase (TK333 and Del ) resulted in inhibition of adenylate cyclase and increase in sodium transport. The results also indicate that apical AT 1A receptors internalize before interacting with G 1, whereas basolateral AT 1A receptors can activate G 1 without internalization. In this regard, noninternalizing basolateral AT 1A receptors containing the G 1 site (TK333 and Del ) decreased adenylate cyclase and increased sodium transport, but noninternalizing apical AT 1A receptors were unable to inhibit adenylate cyclase or increase sodium transport. Earlier studies suggested that apical AT 1A receptors internalize before activating signaling pathways, including PLC and PLA 2 (3, 21). However, all of these studies utilized pharmacological agents, which were potentially nonspecific and, fur-

6 F628 AT 1A RECEPTORS IN KIDNEY thermore, disrupted cellular internalization pathways, rather than the receptor. Our studies indicate that internalization pathways are driven by motifs on the receptor. The cellular internalization pathways and the mechanism by which AT 1A receptors interact with these pathways are not known. Many receptors interact with an adapter/clathrin pathway and require dynamin or -arrestin to internalize (15, 32). Neither of these pathways may be important for AT 1A receptor internalization (32). The role of alternative pathways, such as the ubiquitin or caveolar pathways, in AT 1A receptor internalization is not known. AT 1A receptor internalization does not appear to be required for AT 1 receptor function in nonpolar tissue (30). For example, noninternalizing AT 1 receptors activate PLC, calcium, mitogen-activated protein kinase, and other pathways (28). In nonpolar tissue, G proteins and signaling enzymes are in the same membrane as AT 1A receptors. In contrast, in the proximal tubule, adenylate cyclase is located in the basolateral membrane and is not apposed to apical AT 1A receptors. Our results indicate that apical AT 1A receptors communicate with basolateral adenylate cyclase and that this occurs by an internalization pathway. Either the receptor is internalized before interacting with G 1 or G 1 is internalized after interacting with AT 1A receptors. It is not known whether G 1 is present in apical membranes or the specific G 1 that interacts with AT 1A receptors. Because adenylate cyclase is present in basolateral membranes of proximal tubule epithelium, it seems most reasonable to suggest that apical AT 1A receptors are internalized before they interact with G 1. In summary, the results indicate that, in contrast to nonpolar cells, internalization pathways are important for AT 1A receptor signaling and function in polarized proximal tubule epithelial cells. Apical and basolateral AT 1A receptor expression does not require receptor internalization. Apical AT 1A receptors undergo internalization before interacting with G proteins and inhibiting camp accumulation and increasing transcellular sodium transport. Basolateral AT 1A receptors interact with G proteins and inhibit camp accumulation and increase sodium transport without internalizing. This work was supported by funds provided by Astra Zeneca (S. L. Linas) as well as National Heart, Lung, and Blood Institute Grant HL (T. Thekkumkara). REFERENCES 1. Barak LS, Tiberi M, Freedman NJ, Kwatra MM, Lefkowitz RJ, and Caron MG. A highly conserved tyrosine residue in G protein-coupled receptors is required for agonist-mediated 2- adrenergic receptor sequestration. J Biol Chem 269: , Becker BN, Cheng HF, Burns KD, and Harris RC. Polarized rabbit type 1 angiotensin II receptors manifest differential rates of endocytosis and recycling. Am J Physiol Cell Physiol 269: C1048 C1056, Becker BN, Cheng HF, and Harris RC. Apical ANG IIstimulated PLA 2 activity and Na flux: a potential role for Ca 2 -independent PLA 2. Am J Physiol Renal Physiol 273: F554 F562, Brown GP and Douglas JG. Angiotensin II binding sites on isolated rat renal brush border membranes. Endocrinology 111: , Brown GP and Douglas JG. Angiotensin II-binding sites in rat and primate isolated renal tubular basolateral membranes. Endocrinology 112: , Cervenka L, Mitchell KD, Olivero MI, Coffman TM, and Navar LG. Renal function in the AT 1A knockout mouse during normal and volume-expanded conditions. Kidney Int 56: , Chiao H, Kohda Y, Craig L, Housini I, and Star RA. -Melanocyte-stimulating hormone protects against renal injury after ischemia in mice and rats. J Clin Invest 99: , Franzoni L, Nicastro G, Pertinhez TA, Tato M, Nakaie CR, Paiva ACM, Schreier S, and Spisni A. Structure of the C- terminal fragment of the rat angiotensin II AT 1A receptor and its relevance with respect to G protein coupling. J Biol Chem 272: , Harris RC, Becker BN, and Cheng HF. Acute and chronic mechanisms for regulating proximal tubule angiotensin II receptor expression. J Am Soc Nephrol 8: , Harrison-Bernard LM, Navar LG, Ho MM, Vinson GP, and El-Dahr SS. Immunohistochemical localization of ANG II AT 1 receptors in adult rat kidney using a monoclonal antibody. Am J Physiol Renal Physiol 273: F170 F177, Hunyady L, Bor M, Balla T, and Catt KJ. Identification of a cytoplasmic Ser-Thr-Leu motif that determines agonist-induced internalization of the AT 1 angiotensin receptor. J Biol Chem 269: , Hunyady L, Bor M, Baukal AJ, Balla T, and Catt KJ. A conserved NPLFY sequence contributes to agonist binding and signal transduction but is not an internalization signal for the type 1 angiotensin II receptor. J Biol Chem 270: , Iwai N and Inagami T. Identification of two subtypes in the rat type I angiotensin II receptor. FEBS Lett 298: , Kai H, Alexander RW, Ushio-Fukai M, Lyons PR, Akers M, and Griendling KK. G protein binding domains of the angiotensin II AT 1A receptors mapped with synthetic peptides selected from the receptor sequence. Biochem J 332: , Lefkowitz RJ. G protein-coupled receptors. III. New roles for receptor kinases and -arrestins in receptor signaling and desensitization. J Biol Chem 273: , Linas SL. Role of receptor-mediated endocytosis in proximal tubule epithelial function. Kidney Int 61 Suppl: S18 S21, Liu FY and Cogan MG. Angiotensin II stimulation of hydrogen ion secretion in the rat early proximal tubule: modes of action, mechanism, and kinetics. J Clin Invest 82: , Liu FY and Cogan MG. Angiotensin II stimulates early proximal bicarbonate absorption in the rat by decreasing cyclic adenosine monophosphate. J Clin Invest 84: 83 91, Okubo S, Nimura F, Matsusaka T, Fogo A, Hogan BL, and Ichikawa I. Angiotensinogen gene null-mutant mice lack homeostatic regulation of glomerular filtration and tubular reabsorption. Kidney Int 53: , Sano T, Ohyama K, Yamano Y, Nakagomi Y, Nakazawa S, Kikyo M, Shirai H, Blank JS, Exton JH, and Inagami T. A domain for G protein coupling in carboxyl-terminal tail of rat angiotensin II receptor type 1A. J Biol Chem 272: , Schelling JR, Hanson AS, Marzec R, and Linas SL. Cytoskeleton-dependent endocytosis is required for apical type 1 angiotensin II receptor-mediated phospholipase C activation in cultured rat proximal tubule cells. J Clin Invest 90: , Schelling JR and Linas SL. Angiotensin II-dependent proximal tubule sodium transport requires receptor-mediated endocytosis. Am J Physiol Cell Physiol 266: C669 C675, Schelling JR, Singh H, Marzec R, and Linas SL. Angiotensin II-dependent proximal tubule sodium transport is mediated by camp modulation of phospholipase C. Am J Physiol Cell Physiol 267: C1239 C1245, 1994.

7 AT 1A RECEPTORS IN KIDNEY F Schuster VL, Kokko JP, and Jacobson HR. Angiotensin II directly stimulates sodium transport in rabbit proximal convoluted tubules. J Clin Invest 73: , Shirai H, Takahashi K, Katada T, and Inagami T. Mapping of G protein coupling sites of the angiotensin II type 1 receptor. Hypertension 25: , Swillens S. How to estimate the total receptor concentration when the specific radioactivity of the ligand is unknown. Trends Pharmacol Sci 13: , Thekkumkara TJ, Cookson R, and Linas SL. Angiotensin (AT 1A) receptor-mediated increases in transcellular sodium transport in proximal tubule cells. Am J Physiol Renal Physiol 274: F897 F905, Thekkumkara TJ, Du J, Dostal DE, Motel TJ, Thomas WG, and Baker KM. Stable expression of a functional rat angiotensin II (AT 1A) receptor in CHO-K1 cells: rapid desensitization by angiotensin II. Mol Cell Biochem 146: 79 89, Thomas WG, Baker KM, Motel TJ, and Thekkumkara TJ. Angiotensin II receptor endocytosis involves two distinct regions of the cytoplasmic tail: a role for residues on the hydrophobic face of a putative amphipathic helix. J Biol Chem 270: , Thomas WG, Thekkumkara TJ, Motel TJ, and Baker KM. Stable expression of a truncated AT 1A receptor in CHO-K1 cells. The carboxyl-terminal region directs agonist-induced internalization but not receptor signaling or desensitization. J Biol Chem 270: , Trowbridge IS, Collawn JF, and Hopkins CR. Signal-dependent membrane protein trafficking in the endocytic pathway. Annu Rev Cell Biol 9: , Zhang J, Ferguson SSG, Barak LS, Menard L, and Caron MG. Dynamin and -arrestin reveal distinct mechanisms for G protein-coupled receptor internalization. J Biol Chem 271: , 1996.