INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE URINE-CONCENTRATING SYSTEM is regulated primarily by vasopressin (antidiuretic hormone), which increases both osmotic water and urea permeabilities in principal cells in the kidney collecting duct. The mechanisms regulating osmotic water permeability have been extensively studied (reviewed in Ref. 15). Aquaporin-2 (AQP2), the vasopressin-regulated water channel, is located in the apical membrane of collecting duct principal cells. Vasopressin increases osmotic water permeability by stimulating the insertion of AQP2-containing vesicles from the subapical region into the apical membrane. After the removal of vasopressin, AQP2 is removed from the apical membrane and remains within subapical vesicles until the cell is again stimulated by vasopressin.

Vasopressin also increases facilitated urea permeability in the perfused rat terminal inner medullary collecting duct (IMCD) (19). However, the mechanism by which vasopressin rapidly increases urea permeability is unknown. Although urea is a small molecule, it is highly polar and has a low permeability across lipid bilayers (7). There is physiological evidence that urea transport occurs by a facilitated transport pathway in the IMCD, and several cDNAs for facilitated urea transporters have now been cloned from kidney (UT-A) and erythrocytes (UT-B) (reviewed in Ref. 18).

One possible mechanism by which vasopressin could increase urea permeability is regulated trafficking of UT-A1, as suggested by comparison with the mechanism by which vasopressin regulates AQP2 [AQP2 and UT-A1 have similar immunolocalization patterns and both are regulated rapidly by vasopressin (13, 16, 19)]. However, Knepper and colleagues (8) tested for regulated trafficking of UT-A1 in the rat IMCD and showed that, in contrast to AQP2, vasopressin does not regulate UT-A1 via vesicular trafficking between the apical membrane and subapical vesicles.

The purpose of this study was to test an alternative mechanism by which vasopressin could increase urea permeability: that vasopressin, acting through cAMP, increases the phosphorylation of UT-A1 in the rat IMCD. The deduced amino acid sequence for UT-A1 contains several consensus sites for phosphorylation by protein kinase A (PKA), as well as protein kinase C or tyrosine kinase (11). Our results indicate that UT-A1 is phosphorylated by vasopressin acting through PKA.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

IMCD suspensions. Male Sprague-Dawley rats (National Cancer Institute, Frederick, MD), weighing 200-250 g, were fed a standard rat chow, kept in filter-top cages with autoclaved bedding, and received free access to normal drinking water. Rats were injected with furosemide (5 mg ip) 20-30 min before they were killed. After death, both kidneys were rapidly removed, whole inner medullas were excised, and the tissue was transferred into microcentrifuge tubes with 1 ml of dissecting buffer containing (in mM) 118 NaCl, 2 K₂HPO₄, 25 NaHCO₃, 1.2 MgSO₄, 2 CaCl₂, 5.5 glucose, and 5 sodium acetate, pH 7.4. Inner medullas were minced finely with a razor blade and put into digestion buffer containing (in mM) 118 NaCl, 5 KCl, 4 Na₂HPO₄, 25 NaHCO₃, 2 CaCl₂, 1.2 MgSO₄, 5.5 glucose, and 5 sodium acetate, as well as 2 mg/ml collagenase B, 0.5 mg/ml BSA, and 540 U/ml hyaluronidase. After a 30-min incubation at 37°C, DNase I was added to a final concentration of 0.001%, and incubation was continued for another 20 min. The suspension was periodically agitated to break up large tissue fragments and to facilitate the digestion process. After the incubation, the resulting suspension was transiently (10 s) centrifuged at 50 g, the supernatant was discarded, and the pellet was resuspended in suspension buffer (digestion buffer without collagenase, BSA, or hyaluronidase). This process was repeated two additional times with suspension buffer and one time with phosphate-free DMEM (GIBCO, Grand Island, NY). The pelleted tubules were resuspended and pooled in phosphate-free DMEM; a small aliquot was removed and checked for the integrity of the tubule suspension using a dissecting microscope, and tubules were redistributed evenly into individual sample microcentrifuge tubes. In general, the two kidneys from a single rat yielded sufficient tubules for about one sample; i.e., five rats yielded sufficient tubules to compare five to seven experimental permutations.

³²P labeling. IMCD suspensions were incubated in phosphate-free DMEM containing 0.1 mCi/ml [³²P]orthophosphate for 3 h at 37°C and gassed with 5% CO₂-95% air. At the end of the 3-h loading period, arginine vasopressin (10 or 100 nM, Sigma, St. Louis, MO), 8-(4-chlorophenylthio)-adenosine 3',5'-cyclic monophosphate (CPT-cAMP; 200 µM, Sigma), H-89 (5 µM, Calbiochem, La Jolla, CA), a specific peptide inhibitor (14---22 amide) of PKA (5 µM, Calbiochem), or 1-desamino-8-D-arginine vasopressin (dDAVP; Desmopressin, 10 or 100 nM, Sigma) was added as detailed in RESULTS. Unincorporated ³²P was removed by three washes with phosphate-free DMEM. Then, the IMCDs were lysed in 1 ml RIPA buffer (10 mM Tris · HCl, pH 7.4, 2.5 mM EDTA, 50 mM NaF, 1 mM Na₄P₂O₇ · 10H₂O, 1 mM phenylmethylsulfonyl fluoride; 1% Triton X-100, 10% glycerol, 1% deoxycholate, 1 µg/ml aprotinin, 0.18 mg/ml sodium orthovanadate) and sheared with a 26-gauge needle. After centrifugation at 14,000 g for 15 min to remove insoluble particulates, samples were incubated overnight with polyclonal anti-UT-A1 (12) at 4°C with gentle mixing; this antibody also detects UT-A2 and UT-A4. Immune complexes were precipitated with protein A-agarose (Pierce, Rockford, IL) for 2 h at 4°C; then, the pelleted beads were washed six times with RIPA and once with potassium-free phosphate-buffered saline. Washes were counted to ensure complete removal of unbound radiolabeled material. Laemmli-SDS-PAGE sample buffer was added directly to the pelleted beads and boiled; proteins were size-separated on two identical SDS-polyacrylamide gels. One gel was dried, and ³²P incorporation into UT-A1 was analyzed by autoradiography. The proteins on the other gel were transferred to polyvinylidene difluoride membrane, and the amount of immunoprecipitated UT-A1 protein was assayed by Western blot.

Statistics. All data are presented as means ± SE, and n is the number of rats. To test for the statistical significance between the results from two groups, Student's t-test or the Mann-Whitney U-test was used. To test more than two groups, ANOVA was used, followed by Tukey's protected t-test (20) to determine which groups' results were significantly different. The criterion for statistical significance was P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effect of vasopressin. There are two different glycosylated forms of UT-A1: 97 and 117 kDa (2). Both the 97-and 117-kDa UT-A1 proteins are phosphorylated in the absence of any added hormone (i.e., basal phosphorylation) and exhibit increased phosphorylation on treatment with 10⁸ M vasopressin (Fig. 1). After the addition of 10⁸ M vasopressin to the IMCD suspension, the phosphorylation of UT-A1 was significantly increased at 2 min, peaked at 5 min, and then remained increased for up to 30 min (Fig. 2). A significant increase in UT-A1 phosphorylation was evident with 10¹¹ M vasopressin, with a much larger effect with 10⁸ and 10⁷ M vasopressin (Fig. 3). A significant increase in the phosphorylation of UT-A1 was also observed with either 10⁸ or 10⁷ M dDAVP, a selective V₂-receptor agonist (Fig. 4).

View larger version (54K): [in this window] [in a new window]   Fig. 1.   Phosphorylation of urea transporter UT-A1 is stimulated by vasopressin. Rat inner medullas were collected, and inner medullary collecting duct (IMCD) suspensions were prepared as described in METHODS. After the radiolabel was loaded, IMCDs were incubated for another 10 min with 108 M vasopressin (arginine vasopressin; AVP); then, washed IMCDs were solubilized and UT-A1 was immunoprecipitated with an anti-COOH-terminal UT-A1 antibody. Left: Western blot showing equal proteins in control (AVP) and vasopressin-treated (+AVP) samples. Right: autoradiogram of the radiolabeled UT-A1 from the same immunoprecipitated proteins. Double-headed arrows highlight 97- and 117-kDa UT-A1 proteins.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Time course of vasopressin stimulation of UT-A1 phosphorylation. After the 3-h radiolabel-loading period, 108 M AVP was added at staggered times, and incubations were stopped with 3 rapid washes with phosphate-free DMEM, followed by solubilization in isolation buffer and RIPA as described in METHODS. The immunoprecipitated UT-A1 protein was analyzed for radiolabeled phosphate incorporation. A duplicate Western blot verified equal UT-A1 protein per lane (not shown). A: autoradiograph showing phosphorylation of the 97- and 117-kDa UT-A1 isoforms in response to incubation with AVP for 0, 2, 5, 10, 20, or 30 min. B: densitometry of 6 separate determinations. Increases in phosphorylation in response to vasopressin were statistically significant (*P < 0.001) at all time points.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Dose-response of UT-A1 phosphorylation to varying concentrations of vasopressin. After the 3-h radiolabel-loading period, different amounts of vasopressin (from 1011 to 107 M) were added to identical IMCD suspensions, and incubation was continued for another 10 min. IMCDs were processed and UT-A1 was immunoprecipitated as described in METHODS. A duplicate Western blot verified equal UT-A1 protein per lane (not shown). A: autoradiograph showing phosphorylation of UT-A1 in the presence of 1011, 1010, 108, and 107 M AVP. The far right lane shows control (ctrl) IMCDs that received no exogenous vasopressin. B: densitometry of 5 separate determinations. Increases in phosphorylation in response to vasopressin were statistically significant (*P < 0.001) at all doses tested.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   1-Desamino-8-[D-arginine] vasopressin (dDAVP) stimulates UT-A1 phosphorylation. After the radiolabel was loaded, IMCDs were incubated for another 10 min without (Ctrl) or with 108 M dDAVP; then, the washed IMCDs were solubilized and UT-A1 was immunoprecipitated with an anti-COOH-terminal UT-A1 antibody as described in METHODS. A duplicate Western blot verified equal UT-A1 protein per lane (not shown). A: representative autoradiogram showing phosphorylation of UT-A1 in IMCDs that received no dDAVP (Ctrl) or 108 M dDAVP. B: densitometry of 7 separate determinations of the effect of 108 M dDAVP, showing that the increase in UT-A1 phosphorylation is significant (*P < 0.002).

Role of PKA. Because vasopressin increases cAMP production in the rat terminal IMCD (21), we tested the effect of an exogenous cell-permeable cAMP analog, CPT-cAMP, and found that it significantly increased the phosphorylation of UT-A1 (Fig. 5). Next, IMCD suspensions were treated with the PKA inhibitor H-89. H-89 significantly blocked vasopressin's stimulation of UT-A1 phosphorylation (Fig. 6). This result was confirmed by using the PKA inhibitor 14---22 amide, a specific, cell-permeable peptide inhibitor of PKA (Fig. 6, lanes 4 and 5, n = 1).

View larger version (30K): [in this window] [in a new window]   Fig. 5.   Phosphorylation of UT-A1 is stimulated by 8-(4-chlorophenylthio)-adenosine 3',5'-cyclic monophosphate (CPT-cAMP). Rat inner medullas were collected and IMCD suspensions were prepared as described in METHODS. After the radiolabel was loaded, IMCDs were incubated for another 30 min with 200 µM CPT-cAMP; then, washed IMCDs were solubilized and UT-A1 was immunoprecipitated with an anti-COOH-terminal UT-A1 antibody. A: Western blot showing equal protein in the control () and CPT-cAMP-treated (+) samples. Right: autoradiogram of the radiolabeled UT-A1 from the same immunoprecipitated proteins. Double-headed arrows highlight 97-and 117-kDa UT-A1 proteins. B: densitometry of 5 separate determinations of the effect of CPT-cAMP, showing that the stimulation of UT-A1 phosphorylation is significant (*P < 0.002).

View larger version (31K): [in this window] [in a new window]   Fig. 6.   Inhibitors of protein kinase A (PKA) inhibit vasopressin-stimulated phosphorylation of UT-A1. After 3-h radiolabel loading, either H-89 (5 µM, n = 3) or the more specific peptide inhibitor PKA inhibitor 14---22 amide (5 µM, n = 1) was added to the IMCD suspension. After a 15-min preincubation, AVP was added and incubation was continued for another 20 min. IMCDs were processed and UT-A1 was immunoprecipitated as described in METHODS. A: autoradiograph of the UT-A1 proteins from control (Ctrl), H-89 alone, H-89 then AVP, AVP alone, and peptide inhibitor (Inhib.) then AVP. B: densitometry of 3 separate determinations of the effect of H-89, showing that H-89 blocks AVP-stimulated UT-A1 phosphorylation (*P < 0.05).

View larger version (39K): [in this window] [in a new window]   Fig. 7.   Calyculin and okadaic acid, inhibitors of protein phosphatases PP1 and PP2A, cause increased phosphorylation of UT-A1. After radiolabel loading for 3 h, either calyculin (5 or 50 nM) or okadaic acid (1 or 10 µM) was added to the IMCD suspension for 15 min. IMCDs were processed and UT-A1 was immunoprecipitated as described in METHODS. An autoradiograph of the UT-A1 proteins from these samples is shown. Left: control IMCDs (ctrl) and IMCDs that received 5 or 50 nM calyculin. Middle: control IMCDs (Ctrl) and IMCDs that received 1 or 10 µM okadaic acid. Right: densitometric analysis of calyculin (Caly; 50 nM, n = 10, *P < 0.001) and okadaic acid (OA; 10 µM, n = 4, *P < 0.002) treatments, showing that inhibition of serine/threonine phosphatases significant increases in UT-A1 phosphorylation.

View larger version (59K): [in this window] [in a new window]   Fig. 8.   Phosphorylation of UT-A1 does not occur on tyrosine residues. Two identical IMCD suspensions were incubated with antibodies to UT-A1 and phosphotyrosine (p-Tyr)-containing proteins (PY-20). Immunoprecipitated proteins were separated by SDS-PAGE and blotted to polyvinylidene difluoride membranes. Each blot contains, from left to right, UT-A1 immunoprecipitate, preimmunoprecipitation IMCD lysate, and PY-20 immunoprecipitate. Blots were probed with antibodies to UT-A1 (A) and PY-20 (B), the antibody that recognizes p-Tyr-containing proteins. Anti-UT-A1 did not recognize any proteins in the p-Tyr-protein pool, and PY-20 did not recognize immunoprecipitated UT-A1.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The major result of this study is that vasopressin, acting through PKA, increases the phosphorylation of UT-A1 in freshly isolated suspensions of rat IMCDs. The time course for vasopressin-mediated increases in UT-A1 phosphorylation matches that for vasopressin-mediated increases in facilitated urea permeability in perfused rat terminal IMCDs (14, 21, 23). These findings strongly suggest that phosphorylation of UT-A1 is a mechanism by which vasopressin increases facilitated urea permeability in rat terminal IMCDs in vivo.

The terminal IMCD is generally thought to express only V₂ receptors (5, 6), although one study did find evidence for V_1a receptors by RT-PCR (22). Previous studies showed that vasopressin stimulates urea permeability in perfused terminal IMCDs via the V₂ receptor by showing that dDAVP, a selective V₂ agonist, mimics the effect of vasopressin (21). The present study shows that dDAVP also mimics the effect of vasopressin to increase UT-A1 phosphorylation.

Plasma vasopressin levels generally range between 10¹¹ and 10¹⁰ M (4). These levels of vasopressin stimulate cAMP production in microdissected rat IMCDs (21). However, higher vasopressin levels (10⁸-10⁷ M) stimulate substantially higher levels of cAMP production (21) and UT-A1 phosphorylation (present study). We chose to use 10⁸ M vasopressin in the present experiments because this concentration 1) resulted in a maximal rate of cAMP accumulation (21); 2) has been used in a large number of the perfused tubule measurements of urea permeability (reviewed in Ref. 18); and 3) provided an optimal signal-to-noise for assessing inhibition of the PKA pathway. Although this vasopressin concentration is higher than that found in systemic plasma, vasa recta vasopressin levels could be higher than plasma levels due to countercurrent multiplication. Because terminal IMCDs are exposed to vasa recta rather than systemic plasma, it is possible that these higher vasopressin concentrations may be physiological in the deepest portions of the inner medulla.

We also found that inhibitors of PKA reduce both the vasopressin-stimulated and basal levels of phosphorylation of UT-A1. This result strongly suggests that PKA is phosphorylating UT-A1, both basally and in response to vasopressin stimulation. The reduction in basal phosphorylation by H-89 suggests that adenylyl cyclase may be constitutively phosphorylating UT-A1 in the IMCD. However, we cannot determine from the present studies whether PKA is directly phosphorylating UT-A1 or whether it phosphorylates UT-A1 indirectly by activating another kinase (or kinases), which then phosphorylates UT-A1.

PKA typically results in the phosphorylation of serine or threonine residues. The increase in UT-A1 phosphorylation by okadaic acid or calyculin is consistent with a serine or threonine phosphorylation site (1, 3, 9), and we showed that vasopressin does not phosphorylate a tyrosine residue. Future studies will be needed to determine the residues in UT-A1 that are phosphorylated by vasopressin.

Finally, the present study suggests a mechanism by which two vasopressin-regulated transport processes, water and urea transport, can be independently regulated: 1) urea permeability is regulated primarily by phosphorylation of UT-A1; whereas 2) osmotic water permeability is regulated primarily by the regulated trafficking of AQP2, although phosphorylation of AQP2 is important for its insertion into the apical membrane and the formation of tetramers (10). Atrial natriuretic factor inhibits vasopressin-stimulated osmotic water permeability, but not vasopressin-stimulated urea permeability, in perfused rat terminal IMCDs (17). This result could be explained if atrial natriuretic factor were to affect the trafficking of AQP2, but had no effect on phosphorylation, in response to vasopressin. Future studies will be needed to test this hypothesis.