Heterologous downregulation of vasopressin type 2 receptor is ...

Am J Physiol Renal Physiol 304: F553–F564, 2013. First published December 12, 2012; doi:10.1152/ajprenal.00438.2011.

Heterologous downregulation of vasopressin type 2 receptor is induced by transferrin Richard Bouley,1 Paula Nunes,1 Billy Andriopoulos Jr.,1 Margaret McLaughlin,1 Matthew J. Webber,1 Herbert Y. Lin,1 Jodie L. Babitt,1 Thomas J. Gardella,2 Dennis A. Ausiello,1 and Dennis Brown1 1

Nephrology Division, MGH Center for Systems Biology, Program in Membrane Biology, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; and 2Endocrine Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts Submitted 4 August 2011; accepted in final form 8 December 2012

Bouley R, Nunes P, Andriopoulos Jr. B, McLaughlin M, Webber MJ, Lin HY, Babitt JL, Gardella TJ, Ausiello DA, Brown D. Heterologous downregulation of vasopressin type 2 receptor is induced by transferrin. Am J Physiol Renal Physiol 304: F553–F564, 2013. First published December 12, 2012; doi:10.1152/ajprenal.00438.2011.— Vasopressin (VP) binds to the vasopressin type 2 receptor (V2R) to trigger physiological effects including body fluid homeostasis and blood pressure regulation. Signaling is terminated by receptor downregulation involving clathrin-mediated endocytosis and V2R degradation. We report here that both native and epitope-tagged V2R are internalized from the plasma membrane of LLC-PK1 kidney epithelial cells in the presence of another ligand, transferrin (Tf). The presence of iron-saturated Tf (holo-Tf; 4 h) reduced V2R binding sites at the cell surface by up to 33% while iron-free (apo-Tf) had no effect. However, no change in green fluorescent protein-tagged V2R distribution was observed in the presence of bovine serum albumin, atrial natriuretic peptide, or ANG II. Conversely, holo-Tf did not induce the internalization of another G protein-coupled receptor, the parathyroid hormone receptor. In contrast to the effect of VP, Tf did not increase intracellular cAMP or modify aquaporin-2 distribution in these cells, although addition of VP and Tf together augmented VP-induced V2R internalization. Tf receptor coimmunoprecipitated with V2R, suggesting that they interact closely, which may explain the additive effect of VP and Tf on V2R endocytosis. Furthermore, Tf-induced V2R internalization was abolished in cells expressing a dominant negative dynamin (K44A) mutant, indicating the involvement of clathrincoated pits. We conclude that Tf can induce heterologous downregulation of the V2R and this might desensitize VP target cells without activating downstream V2R signaling events. It also provides new insights into urine-concentrating defects observed in rat models of hemochromatosis. G protein-coupled receptor; endocytosis; antidiuretic hormone; kidney epithelial cells THE TYPE 2 VASOPRESSIN RECEPTOR (V2R) is a G protein-coupled receptor (GPCR) mainly localized in the kidney, where it plays a major role in water reabsorption (5, 7, 8, 37, 43, 55, 64). When the peptide hormone vasopressin (VP) binds V2R in the collecting duct principal cells, it triggers an increase of intracellular cAMP, leading to the phosphorylation and membrane accumulation of the aquaporin-2 water channel (AQP2), which increases the water permeability of collecting ducts, allowing urine concentration to occur (10, 32, 43, 55, 64). Unlike many GPCRs, V2R signaling is not desensitized by phosphorylation by its downstream effector PKA (homodesensitization), nor by kinases activated by other GPCRs (heterodesensitization) (6, 34). Instead, termination of agonist-stimulated V2R

Address for reprint requests and other correspondence: R. Bouley, MGH Center for Systems Biology, Program in Membrane Biology/Nephrology Division, Simches Research Center Massachusetts General Hospital, 185 Cambridge St., Boston, MA 02114 (e-mail: [email protected]). http://www.ajprenal.org

signal transduction requires V2R internalization and degradation, as shown by our group and others (11, 29, 39, 51, 73). Several studies have investigated potential cross-talk between ligandstimulated V2R and other receptors (6, 34, 49). Some reported long-term effects of either ANG II, endothelin, and prostaglandin E2 on the intracellular abundance of V2R protein at the level of gene expression (40, 60, 71, 72). Recently, VP, via its action on V1R, was reported to affect V2R promoter activity (30) while Sarmiento et al. (54) suggested that an alternative splice variant of V2R downregulated the level of functional V2R at the plasma membrane. No short-term cross-talk mechanisms regulating the termination of V2R or the modulation of V2R trafficking have been described to date. Most hormone-binding GPCRs internalize only, or mainly, in the presence of their cognate ligands, but many other receptors for nonhormonal ligands such as nutrients, including the transferrin (Tf) receptor and the LDL receptor, are constitutively internalized even in the absence of their ligands (4, 15, 41). Tf is an apoprotein of 80 kDa that can bind 1 or 2 iron ions (partially saturated- or holo-Tf, respectively). An increase in the level of Tf iron saturation increases its affinity for its receptor (TfR1). TfR1 is a dimer that constitutively recycles between the plasma membrane and endosomal compartments by internalization via clathrin-coated pits in the absence of ligand (17, 28). Binding of Tf increases the rate of TfR1 internalization, although the level of TfR1 at the membrane remains constant due to a concomitant increase in recycling rates (1, 69). In contrast, while VP binding also increases V2R internalization, the majority of VP-bound V2R complexes are degraded in lysosomal compartments and only a small fraction is recycled back to the cell surface (11, 51). In this study, we investigated an unexpected initial observation that green fluorescent protein-tagged V2R (V2R-GFP) internalization is stimulated by the presence of iron-bound Tf. We show that this is a specific effect that occurs without triggering the canonical downstream signaling pathway of V2R. These findings represent the first example of heterologous downregulation of V2R and suggest a novel connection between iron metabolism and body water homeostasis. MATERIALS AND METHODS

Cell culture and transfection. Unless otherwise stated, all chemicals were purchased from Sigma (St. Louis, MO), and all cell culture reagents were purchased from Invitrogen (Carlsbad, CA). FLAG-V2R and the COOH-terminus GFP-tagged V2R (V2R-GFP) were stably transfected into LLC-PK1a cells that were cultured in DMEM, 10% FBS, 1 mg/ml neomycin as previously described (LLC-FLAG-V2R and LLC-V2R-GFP) (11, 58, 73). LLC-PK1 porcine kidney epithelial cells untransfected and stably transfected with human parathyroid hormone (PTH) receptor (PTHR) (LLC-PTHR-GFP) were also cultured in DMEM, 10% FBS, with or without 1 mg/ml neomycin,

1931-857X/13 Copyright © 2013 the American Physiological Society

F553

F554

DOWNREGULATION OF V2R BY TRANSFERRIN

respectively. The expression of FLAG-V2R and V2R-GFP resulted in the replenishment of “normal” V2R expression levels in LLC-PK1a cells, a subclone that has a low level of endogenous V2R expression. LLC-PK1 cells stably expressing c-myc-tagged AQP2 (c-myc AQP2; LLC-AQP2 cells) were grown as previously described (33). LLCV2R-GFP cells were transfected with c-myc-tagged TfR1 in pcDNA6/B (a gift from Dr. Nancy Andrews, Duke University, NC). Briefly, 80% confluent LLC-V2R-GFP cells were incubated for 20 h in the presence of both 4 ␮g of c-myc-tagged TfR1-encoding plasmid and 15 ␮l of lipofectamine 2000. Twenty-four hours after transfection, cells were incubated in selection medium (DMEM, 10% FBS, 5 ␮g/ml blasticidin). Resistant colonies were isolated by dilution and several clones were analyzed for TfR1 protein expression by immunofluorescence. [3H]-VP, [125I]-Tf, and [125I]-PTH binding assays. Evaluation of cell surface [3H]-VP binding on LLC-V2R-GFP cells pretreated with either VP or Tf was performed. LLC-V2R-GFP cells were plated at a density of 50,000 cells/well in 24-well plates 72 h before experiments. Before each experiment, cells were washed three times in DMEM alone to avoid interference with FBS components such as endogenous hormone or bovine Tf, then all incubations were performed in DMEM only. Cells were preincubated for 1 to 4 h with 0.3 ␮M iron-free (apo-human Tf), partially iron-saturated human Tf or with completely iron-saturated human Tf (holo-Tf) diluted in DMEM without FBS. Dose-effect binding assays were performed by incubating cells 4 h with increasing amounts of holo-Tf (0.01 to 10 ␮M). Binding assays were also performed on cells incubated for 4 h in the presence of 1 ␮M ANG II, 1 ␮M BSA, or 1 ␮M atrial natriuretic peptide (ANP) diluted in DMEM only. After incubation, ligands were removed by two acid washes (50 mM sodium citrate, pH 5.0, 0.2 mM NaH2PO4, and 90 mM NaCl) and the acidity was neutralized by two washes with cold PBS. [3H]-VP binding on LLC-PK1 and LLC-FLAG-V2R cells pretreated with or without holo-Tf was also performed. [3H]-VP binding assays were performed as previously described (12). A similar method was used to investigate the direct [3H]-VP binding displacement from V2R in the presence of increasing amounts of holo-Tf. Briefly, [3H]-VP (44.0 Ci/mmol; PerkinElmer, Boston, MA) was diluted in 20 mM sodium phosphate buffer, pH 7.4, 1 mg/ml glucose, 1 mM CaCl2, 1 mM MgCl2, 3.5 mM KCl, 137 mM NaCl, 1 mM tyrosine, 1 mM phenylalanine, and 0.5% BSA. Nonspecific binding was determined in the presence of excess unlabeled VP (1 ␮M). After incubation (3 h, 4°C), washed cells were solubilized in 500 ␮l of 0.1 N NaOH. Radioactivity was quantified with a Tricarb 2200 CA liquid scintillation analyzer (Packard). Binding assays were also performed to investigate the effect of holo-Tf on the recovery of VP binding sites at the cell surface. LLC-V2R-GFP cells were pretreated with or without cycloheximide (35 ␮M) for 1 h before the incubation (4 h) with Tf (0.3 mM), VP (1 mM), or both ligands simultaneously. The ligands were removed by three washes using acid wash buffer containing 50 mg/ml of deferoxamine mesylate. LLC-V2R-GFP cells were washed in DMEM without serum three times and then incubated at 37°C for 30 to 360 min in the presence or absence of cycloheximide. After the recovery period, [3H]-VP binding assays were performed as described above. The effect of Tf on [3H]-VP binding sites was also studied in cells infected for 16 h with a recombinant adenovirus expressing the dominant negative dynamin 2/K44A mutant (a gift of Dr. Jeffrey Pessin, University of Iowa). Adenovirus was purified as previously described (62). The effect of holo-Tf on [125I]-PTH binding site endocytosis was investigated by dose-displacement assays. [125I]PTH binding site assays on LLC-PTHR-GFP cells pretreated with or without holo-Tf were performed using similar conditions as described above. Cells plated in 24-well plates were washed three times in DMEM alone. Cells were preincubated for 4 h with or without 0.3 ␮M holo-Tf diluted in DMEM without FBS. After incubation, cells were washed with acidic buffer and subsequently neutralized with cold PBS before the binding assay was performed. [125I]-PTH binding assays

were performed as previously described except that ligands were diluted in the same binding buffer as described above (58). Dosedisplacement assays where [125I]-PTH (0.1 nM) was displaced by increasing amounts of PTH (0.01 to 1,000 nM) were performed. Maximal binding site densities in LLC-PTHR-GFP cells were determined by nonlinear regression curve fit using Prism (Graphpad software, La Jolla, CA). The effect of VP on [125I]-Tf binding site endocytosis was also investigated by dose-displacement assays. [125I]-Tf binding site assays were performed on LLC-V2R-GFP cells pretreated with or without VP for 4 h at 37°C in DMEM without FBS. After incubation, cells were washed with acidic buffer containing deferoxamine mesylate (50 mg/ml) and subsequently neutralized with cold PBS before the [125I]-Tf binding assay was performed. Dosedisplacement assays were performed in the presence of [125I]-Tf (8 to 10 nM; PerkinElmer, Waltham, MA). The radioligand was displaced by increasing amounts of holo-Tf (0.1 to 1,000 nM). Maximal binding site densities were determined using by nonlinear regression curve fit using Prism. cAMP assay. Intracellular cAMP levels were measured with a Biotrak kit in the constant presence of 3-isobutyl-1-methyl-xanthine (1 mM), a phosphodiesterase inhibitor (Amersham Biosciences, Piscatway, NJ) as previously described (9, 12). Intracellular cAMP accumulation was evaluated in LLC-V2R-GFP cells after incubation with 1 ␮M VP and 0.3 ␮M holo-Tf. All cAMP assays were performed in triplicate. Immunocytochemistry. LLC-V2R-GFP, LLC-PTHR-GFP, or LLCAQP2 cells were plated on 12 ⫻ 12-mm glass coverslips (Fisher Scientific, Pittsburgh, PA). The cells were incubated 4 h at 37°C in DMEM with or without VP (1 ␮M), ANG II (1 ␮M), ANP (1 ␮M), or BSA (1 ␮M). Cells were also incubated for different periods of time in the presence of rhodamine-tagged Tf (Tf-rho) or unlabeled holo-Tf (0.3 ␮M). After incubation, cells were rinsed with cold PBS and fixed in PBS containing 4% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA) and 5% sucrose for 20 min at room temperature. The fixed cells were rinsed three times in PBS and mounted in Vectashield (Vector Labs, Burlingame, CA). LLC-AQP2 cells were stained with a monoclonal anti-c-myc antibody as previously described (9) before being mounted in Vectashield and examined using a Zeiss Radiance 2000 confocal microscope (Carl Zeiss MicroImaging, Thornwood, NY). Endocytosis of V2R in the presence of VP and holo-Tf was investigated on LLC-V2R-GFP cells grown on coverslips. Cells were incubated with vasopressin fluorescent ligand (VP-TMR, 1 ␮M) (18) for 4 h at 4°C in the presence or absence of holo-Tf (0.3 ␮M). After the incubation period, cells were warmed up to 37°C for different period of times. After 3, 5, 10, 15, or 30 min at 37°C, cells were fixed and mounted as described above. More than six images of the cells at each time point in both conditions were acquired using a Nikon A1R confocal microscope (Nikon Instruments, Melville, NY). Intracellular and plasma membrane VP-TMR fluorescence intensities were analyzed using Volocity image analysis software (PerkinElmer). Electrophoresis and Western blot analysis. The effect of holo-Tf and VP on V2R degradation was examined by Western blot analysis. LLC-V2R-GFP cells were incubated for 4 h with several concentrations of VP (0 to 1 ␮M) in the presence or absence of holo-Tf (0.3 ␮M). Protein electrophoresis and Western blotting were performed as previously described (11). In brief, ⬃20 ␮g of protein were solubilized in RIPA buffer (Boston Bioproducts), 4 mM EDTA, and protease inhibitor cocktail (Roche Diagnostics GmbH) and lysates were separated on 4 –12% Bis-Tris-PAGE gels (Invitrogen) and then transferred onto PVDF membranes (Bio-Rad, Hercules, CA). The presence of V2R-GFP was detected using a polyclonal rabbit anti-GFP antibody (0.4 ␮g/ml; Molecular Probes/Invitrogen) and revealed using an Amdex goat anti-rabbit IgG-horseradish peroxidase (1:100,000 dilution; Amersham, Little Chalfont, UK). Proteins were visualized using a Western Lightning chemiluminescence reagent plus system (PerkinElmer Life Sciences). For reblotting, acid-stripped membranes were incubated with a mouse anti-pan-actin antibody (0.2 ␮g/ml;

AJP-Renal Physiol • doi:10.1152/ajprenal.00438.2011 • www.ajprenal.org


Chemicon International, Temecula, CA) and used as loading controls. All Kodak BioMax XAR films (Fisher Scientific) were scanned and band intensities were quantified using IPLab software (BD Biosciences, San Jose, CA). Immunoprecipitation. LLC-V2R-GFP cells transfected with c-myctagged Tf were grown to 80% confluency. After incubation with holo-Tf (0.3 ␮M) for various periods of time, cells were solubilized for 15 min at 4°C in 50 mM Tris·HCl, pH 7.4, 1% NP-40, 0.25 mM sodium deoxycholate, 150 mM NaCl, 1 mM EDTA buffer containing 2 mM sodium orthovanadate, 1 mM sodium fluoride, and protease inhibitor cocktail. Samples were centrifuged at 14,000 g (Legend Micro 17R, Thermo Scientific, Rockford, IL). After determination of protein concentration by BCA (Thermo Scientific), lysates containing 250 ␮g of solubilized proteins were precleared by 1-h incubation with 50 ␮l of protein A (Thermo Scientific). Precleared lysate supernatants were then incubated overnight with 0.5 ␮g of anti-GFP polyclonal antibody. After 12 h, new protein A (50 ␮l) was added for 1 h and then samples were washed four times in lysis buffer. Proteins were diluted in SDS-PAGE buffer and heated to 70°C before separation by SDSPAGE and transfer onto a nitrocellulose membrane as described above. The presence of TfR1 was detected using a mouse monoclonal anti-TfR1 antibody (0.5 ␮g/ml; Invitrogen) and revealed using goat anti-mouse IgG-horseradish peroxidase (0.16 ␮g/ml; Jackson ImmunoResearch, West Grove, PA). Proteins were visualized using Western lightning chemiluminescence reagent plus system (PerkinElmer Life Sciences, Boston, MA). For reblotting, acid-stripped membranes were incubated with a mouse monoclonal anti-GFP antibody (0.04 ␮g/ml; Santa Cruz Biotechnology, Santa Cruz, CA) and used as loading controls. Chemiluminescence signals were captured on Kodak BioMax XAR films (Fisher Scientific). Electron microscopy and immunogold labeling. The gold-Tf probe was produced as described by Neutra et al. (42). Briefly, the pH of a solution of 10-nm colloidal gold (Ted Pella, Redding, CA) was adjusted to 0.5 pH units above the isoelectric point of Tf, and 2.5 ␮g of holo-Tf per 1 ml of the gold colloidal solution (pH 6.0) were added to stabilize the gold probe. The stabilized Tf-conjugated gold solution was diluted to a final concentration of 0.01% with aqueous polyethylene glycol solution (PEG; MW 20,000). The mixture was centrifuged at 105,000 g for 90 min. The soft part of the pellet that contains stable gold-Tf was washed twice with 6.7 mM phosphate buffer pH 6.0 and PEG 0.01%. The gold-Tf was stored in Hanks balanced salt solution containing 20 mM HEPES and 1 mM ferric ammonium citrate until the immunogold labeling assay was performed. LLC-FLAG-V2R cells were grown in 60-mm dishes. Cells were incubated 3 h at 4°C in DMEM containing gold-Tf and mouse anti-FLAG antibody (Sigma). During the last hour of incubation, goat anti-mouse IgG coupled to 15-nm gold particles (Ted Pella) was added. After incubation, cells were washed twice in DMEM. Cells were incubated in DMEM medium containing 1 mM ferric ammonium citrate in the absence or presence of VP (1 ␮M). After 30 or 60 min at 37°C, treated cells were fixed for 1 h in 2% glutaraldehyde in 0.1 M cacodylate buffer. The cells were scraped, postfixed/stained in 1% osmium tetroxide (Electron Microscopy Sciences, Fort Washington, PA) in 0.1 M cacodylate buffer, and then dehydrated in a series of ascending graded alcohols and embedded in Epon (Ted Pella). Ninety-nanometer sections were cut on a Reichert ultramicrotome (Depew, NY) and collected on formvar-coated grids. The cells were stained with 2% uranyl acetate, rinsed in distilled water, dried, examined, and photographed at 80 kv with a JEOL 1011 electron microscope (Tokyo, Japan) equipped with an AMT digital camera (Danvers, MA). RESULTS

Holo-Tf induces V2R internalization. Confocal microscopy showed that V2R-GFP was mainly expressed at the plasma membrane under basal conditions (Fig. 1A) and internalized in

F555

vesicles localized throughout the cytoplasm in the presence of either VP (1 ␮M) or holo-Tf (0.3 ␮M; see white arrows, Fig. 1, B and C, respectively, n ⫽ 10). To confirm the effect of Tf on V2R-GFP, radioligand binding assays (Fig. 1D) were performed on LLC-V2R-GFP cells. The presence of VP (1 ␮M) or holo-Tf reduced the level of VP binding sites at the plasma membrane in a time-dependent manner. After 2 h, [3H]-VP binding sites were reduced by more than ⬃80% in the presence of VP (Fig. 1D, open bars) while holo-Tf reduced [3H]-VP binding sites at the plasma membrane by ⬃33% of their original level (*P ⬍ 0.05; Fig. 1D, filled bars). An increase of holo-Tf (0.001 to 10 ␮M) is associated with a reduction of [3H]-VP binding sites on the plasma membrane (Fig. 1D, inset). This result suggests that the holo-Tf effect on [3H]-VP binding site membrane density is dose-dependent. The role of the iron saturation level of Tf was also investigated by [3H]-VP radioligand binding assay (n ⫽ 3; Fig. 1E). Ironsaturated Tf (holo-Tf) reduced [3H]-VP binding sites at the plasma membrane by 33 ⫾ 3% (n ⫽ 3), partially saturated Tf by 17 ⫾ 6%, and iron-free Tf (apo-Tf) by only 7 ⫾ 3%. To ensure that the GFP molecule did not produce an artifactual interaction, the effect of holo-Tf on [3H]-VP binding sites at the plasma membrane was additionally studied in both untransfected LLC-PK1 cells and in LLC-PK1 cells stably transfected with FLAG-tagged V2R. The presence of holo-Tf (0.3 ␮M, open bar) reduced [3H]-VP binding site at the surface by 12 ⫾ 6 and 44 ⫾ 24% (n ⫽ 4), respectively (Fig. 1F). Furthermore, we studied whether holo-Tf has a direct effect on [3H]-VP binding on V2R and performed a dose displacement of [3H]-VP by increasing amounts of holo-Tf. This was performed at 4°C to prevent receptor internalization. This result showed that [3H]-VP was not displaced from V2R even in the presence of 16 times more holo-Tf (Fig. 1G, n ⫽ 3), suggesting that Tf did not interfere directly with [3H]-VP binding to the V2R. Finally, we investigated [3H]-VP binding site recovery after holo-Tf treatment in the absence of VP (Fig. 1H) or in the presence of VP (Fig. 1G). Cells were treated with cycloheximide to avoid newly synthesized V2R-GFP interference. However, in the absence of new V2R-GFP synthesis, the plasma membrane [3H]-VP binding site density was slowly reduced (4.9 ⫾ 1.2% per hour, n ⫽ 3). Therefore, the effect of cycloheximide was corrected at each individual time point to distinguish the recycling of V2R due to holo-Tfn. All data were expressed as a percentage of the maximum specific [3H]-VP binding site density measured in the absence of cycloheximide. The [3H]-VP binding site densities increased in untreated and holo-Tf-treated conditions (Fig. 1H). The difference in the recovery at each time point between the cycloheximide-treated (open bars) and cycloheximide- and holoTf-treated cells (filled bars) was not significant using 2-way ANOVA analysis. The effect of holo-Tf on [3H]-VP binding site recovery after VP treatment (Fig. 1I) was also investigated. Here, again holo-Tf did not significantly affect the rate of VP binding site recovery. [3H]-VP binding site density increased by 10% after either VP treatment (open bars) or VP and holo-Tf treatment (filled bars). These results show the limited effect of holo-Tf on recycling of V2R to the cell surface. V2R internalization is specifically due to the presence of holo-Tf. To confirm that Tf-induced internalization of V2R was specific, we compared the effect of different ligands such as albumin (BSA), ANP, and angiotensin II (ANG II) on the number of VP binding sites expressed at the plasma membrane by immunocytochemistry and binding assays. The presence of these ligands for over 4 h had no significant effect on V2R


F556


internalization (Fig. 2, A, B, C, respectively). This result is supported by the [3H]-VP binding assay performed in the presence of high amounts of BSA, ANP, and ANG II (Fig. 2D). Tf-induced V2R internalization is not, therefore, a random event that is also associated with endocytosis of these other ligands. The effect of holo-Tf on the internalization of receptors other than V2R was also investigated (Fig. 3). Holo-Tf was added to cells expressing PTHR-GFP. In the presence of 1 ␮M PTH, the PTHR-GFP was internalized (Fig. 3B), but not in the presence of 0.3 ␮M holo-Tf (Fig. 3C). Similar results were

observed by binding assays. Both holo-Tf-treated and nontreated cells showed similar plasma membrane PTH binding site densities (103,406 ⫾ 28,424 vs. 127,198 ⫾ 26,107 binding sites per cell, respectively, means ⫾ SD, n ⫽ 3). In addition, we investigated the effect of VP on Tf receptor endocytosis in LLC-V2R-GFP cells. A Scatchard plot analysis of the [125I]-Tf dose displacement showed similar [125I]-TF binding site densities in either control and VP-treated cells (100,418 ⫾ 40,766 vs. 113,607 ⫾ 61,326 binding sites per cell, respectively, means ⫾ SD, n ⫽ 3). This result suggests that the presence of



F557

Fig. 2. BSA, atrial natriuretic peptide (ANP), and ANG II have no significant effect on V2R-GFP internalization. Immunocytochemistry of LLC-V2R-GFP cells shows that treatment for 4 h with 1 ␮M BSA (A), ANP (B), or ANG II (C) did not affect the baseline plasma membrane localization of V2R-GFP (n ⫽ 3, see also Fig. 1A). D: [3H]-VP radioligand binding assay on LLC-V2R-GFP cells treated for 4 h with 1 ␮M BSA, ANP, or ANG II at 37°C. Subsequently, cells were incubated at 4°C for 3 h with [3H]-VP (9 nM), and the bound radioactivity was quantified as an indication of plasma membrane abundance of V2R-GFP. The effect of holo-Tf (0.3 ␮M) is shown for comparison. Nonspecific binding was determined in the presence of 1 ␮M unlabeled VP. In contrast to holo-Tf, treatment with BSA, ANP, and ANG II did not significantly reduce the density of [3H]-VP binding sites. Each column represents the means ⫾ SD of triplicate samples. Similar data were obtained in 3 separate experiments (bar ⫽ 5 ␮m; t-test, *P ⬍ 0.05, compared with control). NS, not significant.

VP, added to the cells for 4 h, did not significantly affect [125I]-Tf binding site density. The combination of immunocytochemistry and binding assays supports the specificity of Tf in inducing V2R internalization. Holo-Tf and V2R are colocalized. The subcellular localization of V2R and TfR1 was investigated by electron microscopy using double immuno-gold labeling (Figs. 4, A to D). Under basal conditions, gold-labeled proteins were expressed at the plasma membrane (Fig. 4, A and B). V2R and TfR1 were observed in close proximity (Fig. 4, A and B, inset). In the

presence of holo-Tf, clusters of gold particles representing V2R-GFP as well as TfR1 were observed in intracellular vesicular compartments after 30 or 60 min, suggesting that TfR1 and V2R-GFP share common compartments after internalization (Fig. 4, C and D). These results were supported by gold-labeling quantification (Fig. 4, E and F, respectively). Both quantifications show a similar profile. We observed no internalization at 4°C while more than 50% of the gold-V2R (filled bar) and Tf receptor (open bar) were internalized when cells were warmed to 37°C. There was a significant increase in

Fig. 1. Vasopressin (VP) and transferrin (Tf) induce internalization of stably expressed vasopressin type 2 receptor (V2R)-green fluorescent protein (GFP) in LLC-PK1 cells. Confocal microscopy studies showed that while V2R-GFP is mainly localized at the plasma membrane under control conditions (A), 4-h incubation in the presence of 1 ␮M VP (B) or 0.3 ␮M iron-saturated Tf (holo-Tf; C) induced internalization of V2R-GFP (white arrows) stably expressed in LLC-PK1 cells (bar ⫽ 5 ␮m). The extent of internalization was greater with VP than with holo-Tf. Significant membrane V2R remains visible after holo-Tf, but not after VP treatment. VP and Tf reduce plasma membrane binding of [3H]-VP in LLC-V2R-GFP, LLC-FLAG-V2R, and untransfected LLC-PK1 cells. LLC-V2R-GFP cells were incubated in the presence of holo-Tf (0.3 ␮M; filled bars) or VP (1 ␮M; open bars) for 0 to 240 min (4 h) at 37°C. Subsequently, cells were incubated at 4°C for 3 h with 9 nM [3H]-VP, and the bound radioactivity was quantified as an indication of plasma membrane abundance of V2R-GFP. Nonspecific binding was determined in the presence of 1 ␮M unlabeled VP. Both ligands caused a significant, time-dependent loss of V2R from the cell surface, but the effect was greater with VP than holo-Tf, consistent with the fluorescence data (B and C). A dose-response experiment was performed in LLC-V2R-GFP cells incubated in the presence of increasing amounts of holo-Tf (0.001 to 10 ␮M). The reduction in [3H]-VP binding sites at the plasma membrane is dose-dependent (D, inset). Binding assays were performed as described above with holo-Tf and different iron-saturated Tf (E). Saturated Tf (holo-Tf) reduced [3H]-VP binding more than partially and noniron-bound Tf (apo). Holo-Tf treatment (open bars) of both LLC-PK1 and LLC-FLAG-V2R cell lines showed a reduction of [3H]-VP binding compared with controls (filled bars) using the binding assay as described above (F). Furthermore, direct dose-displacement binding assays at 4°C between [3H]-VP and an increasing amount of holo-Tf (0.3 to 9.6 ␮M) do not affect the level of [3H]-VP binding sites at the plasma membrane (G). The effect of holo-Tf on the replenishment rate of [3H]-VP binding sites at the surface of cycloheximide-treated LLC-V2R-GFP cells was investigated in cells treated 4 h in the absence (open bars) or in the presence of holo-Tf (0.3 ␮M; filled bars; H). After incubation, ligand was washed out, and cells were incubated for 0 to 360 min (6 h) at 37°C. The [3H]-VP binding assay was performed after the incubation as described above. The [3H]-VP binding density slowly increased in both conditions. The effect of holo-Tf was also studied in the presence of VP (I). Cells were treated in the presence of VP alone (1 ␮M; open bars) or with holo-Tf (0.3 ␮M; filled bars). Here, we also observed an increase of [3H]-VP binding site density over time under both experimental conditions. Each column represents the means ⫾ SD of 4 independent experiments done in triplicate (*P ⬍ 0.05, compared with control, 2-way, or 1-way ANOVA, Bonferroni post hoc test). The dose-response curve is an average of 3 independent experiments performed in triplicate (means ⫾ SE). AJP-Renal Physiol • doi:10.1152/ajprenal.00438.2011 • www.ajprenal.org

F558


Fig. 3. Tf does not induce internalization of parathyroid hormone (PTH) receptor (PTHR)-GFP stably expressed in LLC-PK1 cells. LLC-PTHR-GFP cells were treated 4 h with either 1 ␮M PTH or 0.3 ␮M rhodamine-tagged Tf (Tf-rho). PTHR-GFP (green) was localized at the plasma membrane under basal conditions (A) and was internalized in the presence of PTH (B). Tf-rho did not affect the location of PTHR-GFP at the plasma membrane (C) even though Tf-rho (red) was internalized, and it became located throughout the cytoplasm. This image is representative of 3 independent experiments (bar ⫽ 5 ␮m). This observation was confirmed using a quantitative [125I]-PTH ligand binding assay (see RESULTS).

gold-V2R internalization in the presence of VP, but VP did not affect Tf internalization. An interaction between both receptors was supported by immunoprecipitation assays (Fig. 4G). After immunoprecipitation of V2R-GFP using a polyclonal anti-GFP antibody, TfR1 was detected by Western blot using a TfR1 antibody. Immunoprecipitation in the absence of the polyclonal anti-GFP antibody showed no TfR1, confirming the specificity of the immunoprecipitation. Interestingly, we observed an increase in the degree of coimmunoprecipitation of V2R and TfR1 after 30 and 60 min in the presence of holo-Tf (n ⫽ 10). The intensity of the TfR1 band was normalized in relation to the intensity of GFP bands detected using a monoclonal antibody against GFP. This corrected for the variation of V2RGFP expressed in cells or the stability of the V2R-GFP in the presence of Tf. The quantification revealed a 50% increase in V2R/TfR1 association by coimmunoprecipitation in the presence of Tf (Fig. 4G; n ⫽ 3). Holo-Tf-induced V2R internalization involves clathrin-coated pits. We investigated the mechanism involved in the holo-Tfinduced downregulation of V2R. [3H]-VP binding assays showed that holo-Tf-induced V2R-GFP internalization was abolished when cells were infected with an adenovirus encoding a dominant negative dynamin 2 (K44A) mutant (Fig. 5), suggesting that holo-Tf-induced V2R is a dynamin-dependent mechanism. We also investigated whether a synergistic effect on V2R internalization existed between holo-Tf and VP by quantitative Western blot analysis as previously described (11) (Fig. 6A, top). The presence of Tf alone did not induce V2R degradation (Fig. 6A, lane 1 vs. 2), which was observed in the presence of VP (lane 3 to 6) as we reported previously (11). The reduction of mature V2R-GFP band intensity in the presence of VP correlated with an increase in the 46-kDa degradation product, suggesting that V2R was at least partially degraded (Fig. 6A, lane 1 vs. lane 3 to 6). This result suggests that V2R-GFP internalized due to Tf exposure is not degraded and that VP ligand occupancy may be necessary for lysosomal targeting of the receptor. Nevertheless, the reduction in the mature V2R band appeared to be more extensive in the presence of both VP and holo-Tf (Fig. 6A, lanes 7 to 10). This was supported by quantification of the intensity of the mature V2R band from several experiments (Fig. 6B) performed in the presence (open bars) or in the absence of holo-Tf (filled bars; n ⫽ 5). In the presence of Tf and increasing amounts of VP, the intensity of the mature V2R bands was significantly lower than the initial control value under all conditions. (P ⬍ 0.05, 2-way ANOVA, Bonferroni post hoc test). This effect cannot be completely ex-

plained by the only slight increase of V2R endocytosis observed in the presence of holo-Tf (Fig. 6C). Ligand-bounded V2R internalized a little more rapidly in the presence of holo-Tf than in the absence of holo-Tf (0.73 ⫾ 0.08 vs. 0.52 ⫾ 0.16%/min, P ⫽ 0.051). Thus, while Tf alone does not result in detectable V2R degradation, it seems to have a synergistic effect in the presence of VP, resulting in increased degradation in the presence of both ligands. Holo-Tf does not stimulate the cAMP-signaling pathway. Next, we investigated whether the effect of holo-Tf on V2R internalization triggered its canonical downstream signaling pathway. First, the intracellular cAMP level in LLC-V2R-GFP cells was evaluated. In contrast to VP, which triggers a significant increase of intracellular cAMP, incubation of cells with holo-Tf did not have a significant effect on intracellular cAMP concentration (Fig. 7A). Intracellular cAMP was also not affected in cells incubated with ANG II, ANP, or BSA (data not shown). In addition, LLC-PK1 cells stably expressing AQP2, the cAMP/VP-sensitive water channel (LLC-AQP2 cells), were incubated with holo-Tf (Fig. 7). Under baseline conditions, AQP2 was located in scattered intracellular vesicles (Fig. 7B), whereas VP shifted the localization of AQP2 from the cytoplasm to the plasma membrane (Fig. 7C) as we previously reported (9). The presence of holo-Tf did not affect the localization of AQP2 compared with baseline conditions (Fig. 7B). Using this well-characterized AQP2-recycling model to study the interacting effect of drugs with the V2R signaling pathway (9, 33, 62), we, therefore, confirmed that Tf-induced V2R downregulation does not affect AQP2 trafficking, recycling, or baseline distribution. DISCUSSION

In this study, we report that internalization of V2R from the plasma membrane of LLC-PK1 cells occurs upon their exposure to holo-Tf. We found that saturated iron-bound Tf induces the downregulation of cell surface V2R via a dynamin-dependent mechanism. However, this internalization process does not initiate the canonical cAMP signaling cascade that is associated with V2R stimulation by vasopressin. Nor does it result in the increased V2R degradation that is seen upon VP-induced V2R downregulation (11, 51). While such heterologous downregulation of the V2R by holo-Tf is a new finding, Sponsel et al. (61) showed that high-iron concentration and the presence of Tf reduce by 20% the expression of



F559

Fig. 4. Internalization of holo-Tf and V2R followed by electron microscopy and coimmunoprecipitation. Colocalization of Tf (10-nm gold particles) and V2R-GFP (15-nm gold particles) in LLC-FLAG-V2R cells was shown by gold labeling (A to D). Under basal conditions, Tf and V2R-GFP were located at the plasma membrane both separately and together (A and B, insets). In the presence of Tf for 30 and 60 min, clusters of V2R-GFP were internalized into vesicles of various shapes and sizes (C and D). These images are representative of 3 experiments (bars ⫽ 500 nm). Quantification of immunogold-labeled V2R and gold-conjugated Tf was performed (E and F, respectively). Gold-V2R (filled bar) is internalized when cells are warmed to 37°C. This effect is more pronounced in the presence of VP. The number of gold-Tf particles (open bar) is reduced at the plasma membrane at 37°C but is not affected by the presence of VP. The quantification was performed on 6 whole cells (means ⫾ SD). One-way ANOVA was performed between 4°C and other conditions (*P ⬍ 0.05 or #P ⬍ 0.001). The 37°C control and 37°C ⫹ VP conditions are significantly different (*P ⬍ 0.05). LLC-V2R-GFP cells expressing TfR1c-myc were used for an immunoprecipitation assay. V2R-GFP was immunoprecipitated using an anti-GFP polyclonal antibody (G). TfR1-c-myc was detected under basal conditions (lane 2) but not in immunoprecipitations performed in the absence of GFP antibody (lane 1). The TfR1 band intensity increased when cells were exposed to Tf for 30 and 60 min before performing the immunoprecipitation experiment (lanes 3 and 4, respectively). Immunoprecipitated GFP was detected using a monoclonal anti-GFP antibody (G, middle). Quantification of TfR1 band intensity was normalized to the corresponding GFP band intensity (G, bottom). This result is the average of 3 independent experiments (means ⫾ SE, *P ⬍ 0.05, 1-way ANOVA).


F560


Fig. 5. Tf-induced V2R internalization is dynamin-dependent. An [3H]-VP binding assay on cells that were or were not infected with dominant-negative Dyn 2 (K44A) adenovirus was performed. The reduction of [3H]-VP binding sites in the presence of holo-Tf (0.3 ␮M; open bars) was abolished in cells infected with Dyn 2 (K44A) mutant virus. Dyn 2 (K44A) mutant virus did not affect untreated cells (filled bar). This result is representative of 3 independent experiments performed in triplicate (means ⫾ SD; t-test, *P ⬍ 0.05, compared with control).

␤1-integrin in LLC-PK1 cells. Furthermore, cell surface muscarinic K⫹ channel transporter activity is also reduced in the presence of holo-Tf but not in the presence of apo-Tf (46). Taken together, these data suggest that holo-Tf may have

effects on surface receptors and channels that extend beyond its role in iron transport processes. Many GPCRs, including the V2R, are internalized from the plasma membrane by a clathrin-mediated mechanism (2, 14, 21, 24, 68). Similarly, the TfR1 is also internalized in clathrincoated pits. Our study confirms the previously reported endosomal colocalization of V2R and Tf (38, 50) but we also report here a new phenomenon whereby the presence of holo-Tf induces downregulation of V2R in LLC-PK1 cells in the absence VP. This effect is dose- and time-dependent, and it requires the presence of 10 to 100 times more holo-Tf than its estimated affinity for the human Tf receptor (16, 19, 25, 59). Because the TfR1 recycles constitutively in most cells even in the absence of its ligand, our findings suggest that the presence of Tf is required to induce association of the two receptors, leading to V2R internalization in the absence of VP. Indeed, we found that there was a concentration-dependent association of TfR1 and V2R by immunoprecipitation. Furthermore, the high levels of Tf required for the effect on V2R internalization probably reflect the fact that the human Tf used in our studies has a lower affinity for the porcine TfR1 in LLC-PK1 cells. Indeed, previous work showed that the level of iron reuptake in cells was different with the use of homo or heterologous sources of Tf (65, 67). Despite the high amino acid homology (70%) between porcine and human Tfs, porcine Tf has a lower affinity toward human TfR1 (3, 65, 66). The effect of Tf on the number of VP cell surface binding sites eventually reached a steady state, which may be due to various factors. First, the system may have reached an equi-

Fig. 6. Tf and VP have a synergistic effect on internalization of plasma membrane V2R-GFP. Western blot analysis of LLC-V2R-GFP cell lysates using an anti-GFP antibody showed that while the mature V2R band intensity was reduced in the presence of VP (filled bars; B), this event was even more pronounced in the presence of both VP and Tf (open bars; B) as shown by quantification of the band intensity representing the mature V2R (A; *P ⬍ 0.05, **P ⬍ 0.01, 2-way ANOVA, Bonferroni post hoc test). This figure is representative of 5 independent experiments (means ⫾ SE). VP-TMR (seen as yellow spots due to overlap with V2R-GFP) was observed mainly at the plasma membrane of LLC-V2R-GFP cells after 3 min at 37°C (C). VP-TMR-labeled V2R endocytosis is observed 30 min in the absence (D) or in the presence of holo-Tf (E). Quantification of the immunocytochemistry data (F) shows that VP-TMR-labeled V2R endocytosis is slightly increased in the presence of holo-Tf over time. Each time point is the means of 3 different experiments. Each experiment time point has been done in 60 cells (*P ⬍ 0.05, 2-way ANOVA, Bonferroni post hoc test; bar ⫽ 5 ␮m). AJP-Renal Physiol • doi:10.1152/ajprenal.00438.2011 • www.ajprenal.org


F561

Fig. 7. Tf-induced V2R internalization does not trigger the canonical V2R downstream signaling pathway. The presence of Tf for 4 h did not modify the level of intracellular cAMP in LLC-V2R-GFP cells (A; filled bars). In contrast, VP (A; open bars) triggered a significant intracellular cAMP accumulation. The VP/cAMPsensitive water channel [aquaporin-2 (AQP2)] was located in intracellular vesicles of LLC-PK1 cells under basal conditions (B) or in the presence of Tf (D), whereas VP (10 nM) induced AQP2, a marked plasma membrane accumulation of AQP2 (C). The intracellular cAMP evaluation is the average or representative of 3 experiments (means ⫾ SD). The images are representative of 3 independent experiments (bar ⫽ 5 ␮m).

librium between endocytosis and exocytosis of the V2R. Our study suggests that holo-Tf does not affect the rate of V2R membrane recovery after downregulation. Second, the effect may be reduced over time due to a reduction in the amount of free iron in the medium. This reduction may result from the release of human and residual bovine Tf, in the form of apo-Tf, from cells during the incubation period. It is well-known that the level of Tf iron saturation is important for cell iron reuptake. While TfR1 internalized with either holo or apo-Tf (19), previous studies show that apo-Tf is three to five times less efficient in cellular iron reuptake than holo-Tf (31, 66). Therefore, a difference in efficiency of binding to TfR1 could explain the different effect of apo and holo Tf on V2R internalization. Our binding assays suggest that holo-Tf-induced V2R downregulation is not due to an allosteric effect of direct Tf binding to the V2R as observed in the case of holo-Tf and insulin-like growth factor-binding protein-3 (70). Our results also show that VP does not affect TfR1 plasma membrane density, and we propose that an interaction between V2R and TfR1/holo-Tf is the most likely explanation for our observations. The absence of V2R downregulation when other receptors are internalized or the lack of effects of apo-Tf on the level of membrane [3H]-VP binding sites suggests a more complex mechanism than a simple increase in fluid-phase micropinocytosis as suggested in other studies for the TfR1 (1). However, while we favor the receptor interaction hypothesis to explain

our data, we cannot rule out other indirect mechanisms to account for our findings. For example, Saint-Marie et al. (52) suggested that generation of a calcium signal occurs after Tf binds to TfR1 and showed that while intracellular calcium levels affect TfR1 recycling, they do not modify endocytosis of TfR1 via dynamin-dependent pathways. They concluded that this calcium-generating effect might be due to an unidentified membrane protein interacting with TfR1. Our finding that Tf-induced V2R endocytosis is inhibited by the dominant negative dynamin K44A mutant also supports the conclusion that a clathrin-mediated process is involved in the internalization of these receptors. The TfR1 also physically interacts with the TCR complex and this plays a role in TfR-induced signal transduction in T cells (53). It is possible that the V2R/TfR1 interaction may play a role in the accumulation of intracellular calcium but this remains to be determined. Under basal conditions, the iron in DMEM and the residual Tf in cells may result in the basal level of coimmunoprecipitation seen between the V2R and TfR1. Importantly, however, the extent of this coimmunoprecipitation increases significantly in the presence of holo-Tf. These data suggest a close interaction, which may result in a temporary dimer that modifies the conformation of the V2R, leading to its internalization. Such hetero-oligomerization among GPCRs has been shown in the past (23, 47). Oligomerization between the V1a and V2R illustrates that receptor activation in its heterodimeric state determines the fate of the complex (63). In contrast, ho-


F562


modimerization between the V2R and a peptide mimicking its third intracellular loop also affected its signaling characteristics (26). Several studies also showed that V2R homo-oligomerization rescues receptor signaling or binding (56, 57). This study sheds some light on the cellular mechanisms of Tf-induced V2R internalization. We showed that this process is dynamin-dependent. This result was not surprising since both receptors, V2R and TfR1, internalize using a clathrin- and dynamin-dependent process (12, 13, 20, 22, 27, 44, 48). V2R can, therefore, internalize in a similar manner via clathrin-coated pits upon either VP- or Tf-induced internalization. We also demonstrated that holo-Tf and VP act synergistically on V2R internalization and degradation in the presence of low concentrations of VP but V2R degradation is not stimulated by Tf alone. This implies that the association of the VP ligand with the V2R either before or during internalization determines the ultimate fate of the V2R-VP complex—i.e., lysosomal degradation. However, internalized TfR1 is sorted into the recycling endosome and the accompanying V2R is not degraded. This result may explain the steady state that is achieved between the recycled V2R at the membrane and the newly internalized V2R. Thus, Tf action may create a pool of V2R that may recycle to the membrane more rapidly, while in the presence of VP, the V2R complex is instead targeted to the perinuclear late endosomal compartment where it is degraded (11). The physiological role of Tf-induced V2R downregulation remains under investigation, and the role of V2R as signaltransduction molecule for Tf is still elusive (52, 53). In addition to its cellular role, the heterologous downregulation of V2R by holo-Tf may be involved in diabetes insipidus in humans and in animal models of hemochromatosis. Hemochromatosis is an iron overload disorder where Tf iron saturation is inappropriately elevated from a normal value of 17% up to 95% (35, 74). Our findings raise the possibility that the downregulation of cell surface V2R in the presence of holo-Tf might be involved in the impaired urinary concentrating ability in these conditions in animals and sometimes in humans (36, 45, 74). In this respect, we show here a clear dissociation between V2R internalization by Tf and the downstream effects that are normally associated with V2R activation. In particular, intracellular cAMP was not elevated upon Tf-induced V2R downregulation, and the AQP2 water channel was not mobilized to the plasma membrane as is usually the case following VP-induced activation of V2R signaling. In this respect, Saint-Marie et al. (52) suggested that Tf is not involved in intracellular signal transduction events involving any other receptors that utilize the clathrin-mediated endocytic pathway. The existence of cell biological cross talk between VP and Tf in downregulation of the V2R with and without activation of the urinary concentrating mechanism suggests that circulating Tf levels might be an important factor in determining body fluid homeostasis. Future in vivo studies will be needed to address this issue. ACKNOWLEDGMENTS We thank Jeremy Roy and Anilkumar Nair for help with the data quantification. GRANTS This work was supported by National Institutes of Health (NIH) Grants PO1DK38452 (to D. Brown and D. A. Ausiello) and DK96586 (to D. Brown). R. Bouley received an investigator award from the National Kidney Founda-

tion. P. Nunes was supported by a Doctoral Level Postgraduate Scholarship from National Sciences and Engineering Research Council. J. L. Babitt was supported by NIH Grants K08 DK-075846 and RO1 DK063016. Dr. Lin is supported by NIH Grants ROI DK069533 and ROI 071837. Dr. T. J. Gardella is supported by NIH Grant DK11794. The Microscopy Core facility of the MGH Program in Membrane Biology receives additional support from the Boston Area Diabetes and Endocrinology Research Center (Grant DK57521) and the Center for the Study of Inflammatory Bowel Disease (Grant DK43351). DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: R.B., B.A., H.Y.L., J.L.B., T.J.G., D.A.A., and D.B. conception and design of research; R.B., P.N., B.A., M.M.M., and M.J.W. performed experiments; R.B., P.N., B.A., M.M.M., J.L.B., and D.B. analyzed data; R.B., P.N., M.M.M., M.J.W., J.L.B., T.J.G., and D.B. interpreted results of experiments; R.B. and M.M.M. prepared figures; R.B., P.N., M.J.W., H.Y.L., D.A.A., and D.B. drafted manuscript; R.B., D.A.A., and D.B. edited and revised manuscript; R.B. and D.B. approved final version of manuscript. REFERENCES 1. Ajioka RS, Kaplan J. Intracellular pools of transferrin receptors result from constitutive internalization of unoccupied receptors. Proc Natl Acad Sci USA 83: 6445–6449, 1986. 2. Bachvarov DR, Houle S, Bachvarova M, Bouthillier J, Adam A, Marceau F. Bradykinin B(2) receptor endocytosis, recycling, and downregulation assessed using green fluorescent protein conjugates. J Pharmacol Exp Ther 297: 19 –26, 2001. 3. Baldwin GS. Comparison of transferrin sequences from different species. Comp Biochem Physiol B 106: 203–218, 1993. 4. Bider MD, Spiess M. Ligand-induced endocytosis of the asialoglycoprotein receptor: evidence for heterogeneity in subunit oligomerization. FEBS Lett 434: 37–41, 1998. 5. Birnbaumer M. Vasopressin receptor mutations and nephrogenic diabetes insipidus. Arch Med Res 30: 465–474, 1999. 6. Birnbaumer M, Antaramian A, Themmen AP, Gilbert S. Desensitization of the human V2 vasopressin receptor. Homologous effects in the absence of heterologous desensitization. J Biol Chem 267: 11783–11788, 1992. 7. Birnbaumer M, Seibold A, Gilbert S, Ishido M, Barberis C, Antaramian A, Brabet P, Rosenthal W. Molecular cloning of the receptor for human antidiuretic hormone. Nature 357: 333–335, 1992. 8. Boone M, Deen PMT. Physiology and pathophysiology of the vasopressin-regulated renal water reabsorption. Pflügers Arch 456: 1005–1024, 2008. 9. Bouley R, Breton S, Sun T, McLaughlin M, Nsumu NN, Lin HY, Ausiello DA, Brown D. Nitric oxide and atrial natriuretic factor stimulate cGMP-dependent membrane insertion of aquaporin 2 in renal epithelial cells. J Clin Invest 106: 1115–1126, 2000. 10. Bouley R, Hasler U, Lu HA, Nunes P, Brown D. Bypassing vasopressin receptor signaling pathways in nephrogenic diabetes insipidus. Semin Nephrol 28: 266 –278, 2008. 11. Bouley R, Lin HY, Raychowdhury MK, Marshansky V, Brown D, Ausiello DA. Downregulation of the vasopressin type 2 receptor after vasopressin-induced internalization: involvement of a lysosomal degradation pathway. Am J Physiol Cell Physiol 288: C1390 –C1401, 2005. 12. Bouley R, Sun TX, Chenard M, McLaughlin M, McKee M, Lin HY, Brown D, Ausiello DA. Functional role of the NPxxY motif in internalization of the type 2 vasopressin receptor in LLC-PK1 cells. Am J Physiol Cell Physiol 285: C750 –C762, 2003. 13. Bowen-Pidgeon D, Innamorati G, Sadeghi HM, Birnbaumer M. Arrestin effects on internalization of vasopressin receptors. Mol Pharmacol 59: 1395–1401, 2001. 14. Bremnes T, Paasche JD, Mehlum A, Sandberg C, Bremnes B, Attramadal H. Regulation and intracellular trafficking pathways of the endothelin receptors. J Biol Chem 275: 17596 –17604, 2000. 15. Brown MS, Anderson RG, Goldstein JL. Recycling receptors: the roundtrip itinerary of migrant membrane proteins. Cell 32: 663–667, 1983. 16. Brown PJ, Molloy CM, Johnson PM. Transferrin receptor affinity and iron transport in the human placenta. Placenta 3: 21–28, 1982.


DOWNREGULATION OF V2R BY TRANSFERRIN 17. Carter LL, Redelmeier TE, Woollenweber LA, Schmid SL. Multiple GTP-binding proteins participate in clathrin-coated vesicle-mediated endocytosis. J Cell Biol 120: 37–45, 1993. 18. Chen S, Webber MJ, Vilardaga JP, Khatri A, Brown D, Ausiello DA, Lin HY, Bouley R. Visualizing microtubule-dependent vasopressin type 2 receptor trafficking using a new high-affinity fluorescent vasopressin ligand. Endocrinology 152: 3893–3904, 2011. 19. Ciechanover A, Schwartz AL, Dautry-Varsat A, Lodish HF. Kinetics of internalization and recycling of transferrin and the transferrin receptor in a human hepatoma cell line. Effect of lysosomotropic agents. J Biol Chem 258: 9681–9689, 1983. 20. Collawn JF, Stangel M, Kuhn LA, Esekogwu V, Jing SQ, Trowbridge IS, Tainer JA. Transferrin receptor internalization sequence YXRF implicates a tight turn as the structural recognition motif for endocytosis. Cell 63: 1061–1072, 1990. 21. Dale LB, Bhattacharya M, Seachrist JL, Anborgh PH, Ferguson SS. Agonist-stimulated and tonic internalization of metabotropic glutamate receptor 1a in human embryonic kidney 293 cells: agonist-stimulated endocytosis is beta-arrestin1 isoform-specific. Mol Pharmacol 60: 1243–1253, 2001. 22. Damke H, Baba T, Warnock DE, Schmid SL. Induction of mutant dynamin specifically blocks endocytic coated vesicle formation. J Cell Biol 127: 915–934, 1994. 23. Devi LA. Heterodimerization of G protein-coupled receptors: pharmacology, signaling and trafficking. Trends Pharmacol Sci 22: 532–537, 2001. 24. Dupre DJ, Chen Z, Le Gouill C, Theriault C, Parent JL, RolaPleszczynski M, Stankova J. Trafficking, ubiquitination, and downregulation of the human platelet-activating factor receptor. J Biol Chem 278: 48228 –48235, 2003. 25. Fuchs H, Gessner R. Iodination significantly influences the binding of human transferrin to the transferrin receptor. Biochim Biophys Acta 1570: 19 –26, 2002. 26. Granier S, Terrillon S, Pascal R, Demene H, Bouvier M, Guillon G, Mendre C. A cyclic peptide mimicking the third intracellular loop of the V2 vasopressin receptor inhibits signaling through its interaction with receptor dimer and G protein. J Biol Chem 279: 50904 –50914, 2004. 27. Hansen SH, Sandvig K, van Deurs B. Internalization efficiency of the transferrin receptor. Exp Cell Res 199: 19 –28, 1992. 28. Iacopetta BJ, Rothenberger S, Kuhn LC. A role for the cytoplasmic domain in transferrin receptor sorting and coated pit formation during endocytosis. Cell 54: 485–489, 1988. 29. Innamorati G, Le Gouill C, Balamotis M, Birnbaumer M. The long and the short cycle. Alternative intracellular routes for trafficking of G protein-coupled receptors. J Biol Chem 276: 13096 –13103, 2001. 30. Izumi Y, Nakayama Y, Mori T, Miyazaki H, Inoue H, Kohda Y, Inoue T, Nonoguchi H, Tomita K. Downregulation of vasopressin V2 receptor promoter activity via V1a receptor pathway. Am J Physiol Renal Physiol 292: F1418 –F1426, 2007. 31. Jandl JH, Katz JH. The plasma-to-cell cycle of transferrin. J Clin Invest 42: 314 –326, 1963. 32. Katsura T, Gustafson CE, Ausiello DA, Brown D. Protein kinase A phosphorylation is involved in regulated exocytosis of aquaporin-2 in transfected LLC-PK1 cells. Am J Physiol Renal Physiol 272: F817–F822, 1997. 33. Katsura T, Verbavatz JM, Farinas J, Ma T, Ausiello DA, Verkman AS, Brown D. Constitutive and regulated membrane expression of aquaporin 1 and aquaporin 2 water channels in stably transfected LLC-PK1 epithelial cells. Proc Natl Acad Sci USA 92: 7212–7216, 1995. 34. Klein U, Muller C, Chu P, Birnbaumer M, von Zastrow M. Heterologous inhibition of G protein-coupled receptor endocytosis mediated by receptor-specific trafficking of beta-arrestins. J Biol Chem 276: 17442– 17447, 2001. 35. Kom GD, Schwedhelm E, Nielsen P, Boger RH. Increased urinary excretion of 8-iso-prostaglandin F2alpha in patients with HFE-related hemochromatosis: a case-control study. Free Radic Biol Med 40: 1194 –1200, 2006. 36. Krysiak R, Kobielusz-Gembala I, Okopien B. Recurrent pregnancyinduced diabetes insipidus in a woman with hemochromatosis. Endocr J 57: 1023–1028, 2010. 37. Kwon TH, Hager H, Nejsum LN, Andersen ML, Frøkiær J, Nielsen S. Physiology and pathophysiology of renal aquaporins. Semin Nephrol 21: 231–238, 2001. 38. Lakadamyali M, Rust MJ, Zhuang X. Ligands for clathrin-mediated endocytosis are differentially sorted into distinct populations of early endosomes. Cell 124: 997–1009, 2006. 39. Lutz W, Sanders M, Salisbury J, Kumar R. Internalization of vasopressin analogs in kidney and smooth muscle cells: evidence for receptor-

40.

41.

42.

43. 44.

45.

46. 47. 48. 49. 50. 51. 52.

53.

54.

55. 56. 57.

58.

59. 60.

F563

mediated endocytosis in cells with V2 or V1 receptors. Proc Natl Acad Sci USA 87: 6507–6511, 1990. Machida K, Wakamatsu S, Izumi Y, Yosifovska T, Matsuzaki T, Nakayama Y, Kohda Y, Inoue T, Saito H, Tomita K, Nonoguchi H. Downregulation of the V2 vasopressin receptor in dehydration: mechanisms and role of renal prostaglandin synthesis. Am J Physiol Renal Physiol 292: F1274 –F1282, 2007. Mohammad S, Baldini G, Granell S, Narducci P, Martelli AM, Baldini G. Constitutive traffic of melanocortin-4 receptor in Neuro2A cells and immortalized hypothalamic neurons. J Biol Chem 282: 4963– 4974, 2007. Neutra MR, Ciechanover A, Owen LS, Lodish HF. Intracellular transport of transferrin- and asialoorosomucoid-colloidal gold conjugates to lysosomes after receptor-mediated endocytosis. J Histochem Cytochem 33: 1134 –1144, 1985. Nielsen S, Frøkiær J, Marples D, Kwon TH, Agre P, Knepper MA. Aquaporins in the kidney: from molecules to medicine. Physiol Rev 82: 205–244, 2002. Oakley RH, Laporte SA, Holt JA, Barak LS, Caron MG. Association of beta-arrestin with G protein-coupled receptors during clathrin-mediated endocytosis dictates the profile of receptor resensitization. J Biol Chem 274: 32248 –32257, 1999. Okumura A, Kondo K, Hirai C, Nishimura H, Tamai H, Kawarazaki F, Ichikawa M, Mizuno M, Oiso Y, Yamamoto M. Nephrogenic diabetes insipidus associated with hemochromatosis. Am J Kidney Dis 40: 403–406, 2002. Park KT, Kang D, Han J, Park JY, Hur CG, Hong SG. Reduction of muscarinic K⫹ channel activity by transferrin in ischemic rat atrial myocytes. Korean J Physiol Pharmacol 7: 333–339, 2003. Park PS, Filipek S, Wells JW, Palczewski K. Oligomerization of G protein-coupled receptors: past, present, and future. Biochemistry 43: 15643–15656, 2004. Pfeiffer R, Kirsch J, Fahrenholz F. Agonist and antagonist-dependent internalization of the human vasopressin V2 receptor. Exp Cell Res 244: 327–339, 1998. Phillips PA, Grant SL, Davidson AF, Risvanis J, Stephenson J, Gow CB. Epidermal growth factor antagonizes vasopressin in vivo and in vitro. Kidney Int 45: 1028 –1036, 1994. Puthenveedu MA, von Zastrow M. Cargo regulates clathrin-coated pit dynamics. Cell 127: 113–124, 2006. Robben JH, Knoers NV, Deen PM. Regulation of the vasopressin V2 receptor by vasopressin in polarized renal collecting duct cells. Mol Biol Cell 15: 5693–5699, 2004. Sainte-Marie J, Lafont V, Pecheur EI, Favero J, Philippot JR, Bienvenue A. Transferrin receptor functions as a signal-transduction molecule for its own recycling via increases in the internal Ca2⫹ concentration. Eur J Biochem 250: 689 –697, 1997. Salmeron A, Borroto A, Fresno M, Crumpton MJ, Ley SC, Alarcon B. Transferrin receptor induces tyrosine phosphorylation in T cells and is physically associated with the TCR zeta-chain. J Immunol 154: 1675– 1683, 1995. Sarmiento JM, Anazco CC, Campos DM, Prado GN, Navarro J, Gonzalez CB. Novel down-regulatory mechanism of the surface expression of the vasopressin V2 receptor by an alternative splice receptor variant. J Biol Chem 279: 47017–47023, 2004. Sasaki S, Noda Y. Aquaporin-2 protein dynamics within the cell. Curr Opin Nephrol Hypertens 16: 348 –352, 2007. Schoneberg T, Yun J, Wenkert D, Wess J. Functional rescue of mutant V2 vasopressin receptors causing nephrogenic diabetes insipidus by a co-expressed receptor polypeptide. EMBO J 15: 1283–1291, 1996. Schulz A, Grosse R, Schultz G, Gudermann T, Schoneberg T. Structural implication for receptor oligomerization from functional reconstitution studies of mutant V2 vasopressin receptors. J Biol Chem 275: 2381–2389, 2000. Shimizu N, Dean T, Tsang JC, Khatri A, Potts JT Jr, Gardella TJ. Novel parathyroid hormone (PTH) antagonists that bind to the juxtamembrane portion of the PTH/PTH-related protein receptor. J Biol Chem 280: 1797–1807, 2005. Shindelman JE, Ortmeyer AE, Sussman HH. Demonstration of the transferrin receptor in human breast cancer tissue. Potential marker for identifying dividing cells. Int J Cancer 27: 329 –334, 1981. Sonntag M, Wang MH, Huang MH, Wong NL. Endothelin upregulates the expression of vasopressin V2 mRNA in the inner medullary collecting duct of the rat. Metabolism 53: 1177–1183, 2004.


F564


61. Sponsel HT, Alfrey AC, Hammond WS, Durr JA, Ray C, Anderson RJ. Effect of iron on renal tubular epithelial cells. Kidney Int 50: 436 –444, 1996. 62. Sun TX, Van Hoek A, Huang Y, Bouley R, McLaughlin M, Brown D. Aquaporin-2 localization in clathrin-coated pits: inhibition of endocytosis by dominant-negative dynamin. Am J Physiol Renal Physiol 282: F998 – F1011, 2002. 63. Terrillon S, Barberis C, Bouvier M. Heterodimerization of V1a and V2 vasopressin receptors determines the interaction with beta-arrestin and their trafficking patterns. Proc Natl Acad Sci USA 101: 1548 –1553, 2004. 64. Valenti G, Procino G, Tamma G, Carmosino M, Svelto M. Minireview: aquaporin 2 trafficking. Endocrinology 146: 5063–5070, 2005. 65. van Bockxmeer FM, Morgan EH. Comparative aspects of transferrinreticulocyte interactions: membrane receptors and iron uptake. Comp Biochem Physiol A Comp Physiol 71: 211–218, 1982. 66. Verhoef NJ, Kester HC, Noordeloos PJ, Leijnse B. Comparative studies on the interaction of various transferrins and rat bone marrow cells. Int J Biochem 10: 595–602, 1979. 67. Verhoef NJ, Kremers JH, Leijnse B. The effect of heterologous transferrin on the uptake of iron and haem synthesis by bone marrow cells. Biochim Biophys Acta 304: 114 –122, 1973. 68. von Zastrow M, Kobilka BK. Ligand-regulated internalization and recycling of human beta 2-adrenergic receptors between the plasma mem-

69.

70.

71.

72.

73.

74.

brane and endosomes containing transferrin receptors. J Biol Chem 267: 3530 –3538, 1992. Wauben-Penris PJ, Strous GJ, Van der Donk HA. Transferrin receptors of isolated rat seminiferous tubules bind both rat and human transferrin. Biol Reprod 35: 1227–1234, 1986. Weinzimer SA, Gibson TB, Collett-Solberg PF, Khare A, Liu B, Cohen P. Transferrin is an insulin-like growth factor-binding protein-3 binding protein. J Clin Endocrinol Metab 86: 1806 –1813, 2001. Wong NL, Sonntag M, Tsui JK. Attenuation of renal vasopressin V2 receptor upregulation by bosentan, an ETA/ETB receptor antagonist. Metabolism 52: 1141–1146, 2003. Wong NL, Tsui JK. Angiotensin II upregulates the expression of vasopressin V2 mRNA in the inner medullary collecting duct of the rat. Metabolism 52: 290 –295, 2003. Yi X, Bouley R, Lin HY, Bechoua S, Sun TX, Del Re E, Shioda T, Raychowdhury MK, Lu HA, Abou-Samra AB, Brown D, Ausiello DA. Alix (AIP1) is a vasopressin receptor (V2R)-interacting protein that increases lysosomal degradation of the V2R. Am J Physiol Renal Physiol 292: F1303–F1313, 2007. Zhou XJ, Vaziri ND, Pandian D, Wang ZQ, Mazowiecki M, Liao SY, Oveisi F. Urinary concentrating defect in experimental hemochromatosis. J Am Soc Nephrol 7: 128 –134, 1996.
