The role of Nedd4-1 WW domains in binding and regulating human organic anion transporter 1

Abstract

Human organic anion transporter 1 (hOAT1), expressed at the basolateral membrane of kidney proximal tubule cells, mediates the active renal secretion of a diverse array of clinically important drugs, including anti-human immunodeficiency virus therapeutics, antitumor drugs, antibiotics, antihypertensives, and anti-inflammatories. We have previously demonstrated that posttranslational modification of hOAT1 by ubiquitination is an important mechanism for the regulation of this transporter. The present study aimed at identifying the ubiquitin ligase for hOAT1 and its mechanism of action. We showed that overexpression of neural precursor cell expressed, developmentally downregulated (Nedd)4-1, an E3 ubiquitin ligase, enhanced hOAT1 ubiquitination, decreased hOAT1 expression at the cell surface, and inhibited hOAT1 transport activity. In contrast, overexpression of the ubiquitin ligase-dead mutant Nedd4-1/C867S was without effects on hOAT1. Furthermore, knockdown of endogenously expressed Nedd4-1 by Nedd4-1-specific small interfering RNA reduced hOAT1 ubiquitination. Immunoprecipitation experiments in cultured cells and rat kidney slices and immunofluorescence experiments in rat kidney slices showed that there was a physical interaction between OAT1 and Nedd4-1. Nedd4-1 contains four protein-protein interacting WW domains. When these WW domains were inactivated by mutating two amino acid residues in each of the four WW domains (Mut-WW1: V210W/H212G, Mut-WW2: V367W/H369G, Mut-WW3: I440W/H442G, and Mut-WW4: I492W/H494G, respectively), only Mut-WW2 and Mut-WW3 significantly lost their ability to bind and to ubiquitinate hOAT1. As a result, Mut-WW2 and Mut-WW3 were unable to suppress hOAT1-mediated transport as effectively as wild-type Nedd4-1. In conclusion, this is the first demonstration that Nedd4-1 regulates hOAT1 ubiquitination, expression, and transport activity through its WW2 and WW3 domains.

organic anion transporters (OATs) play an essential role in body absorption, distribution, and elimination of a diverse array of clinically important drugs, including anti-human immunodeficiency virus therapeutics, antitumor drugs, antibiotics, antihypertensives, and anti-inflammatories (1, 21, 28, 31, 33). A thorough understanding of the regulation of these transporters is a crucial step in assessing their impact in drug efficacy under the normal and pathophysiological conditions.

OAT isoforms are mainly expressed in epithelial tissues such as the kidney, liver, brain, and placenta. In the kidney, OAT1 and OAT3 are localized at the basolateral membrane of proximal tubule cells and are responsible for moving anionic drugs across the basolateral membrane into proximal tubule cells for subsequent exit across the apical membrane into the urine for elimination (1, 21, 28, 31, 33). As a cell membrane transporter, the amount of OAT at the cell surface is critical for its transport activity. We have previously demonstrated that the posttranslational modification of human (h)OAT1 by ubiquitination is an important mechanism for the regulation of this transporter (35). Ubiquitination of hOAT1 resulted in decreased expression of hOAT1 at the cell surface and consequently a decreased transport activity.

Ubiquitination is the modification of the target substrate by ubiquitin conjugation. Ubiquitin is a highly conserved and small (8.5 kDa) regulatory protein that forms an isopeptide bond between its COOH-terminal glycine and a lysine residue on the target protein. Moreover, the ubiquitin molecule itself can form chains of different lengths through its internal lysine residues. Therefore, there are different types of ubiquitin conjugation of a substrate: monoubiquitination, where one single ubiquitin is conjugated to one single lysine on the substrate; multiubiquitination, where several monoubiquitin molecules are conjugated to multiple lysine residues on the substrate; or polyubiquitination, where an extended polyubiquitin chain is conjugated to the lysine on the substrate.

Ubiquitination is accomplished through sequential actions of the E1, E2, and E3 enzymes. E3 ubiquitin ligases are the enzymes responsible for the transfer of ubiquitin to a specific protein substrate. The neural precursor cell expressed, developmentally downregulated (Nedd)4 family of E3 ubiquitin ligases consists of nine members, with Nedd4-1 being the prototype. Recently, two highly homologous members of the Nedd4 family, Nedd4-1 and Nedd4-2, have been implicated in the ubiquitination of many mammalian transporters and channels, including the epithelial Na⁺ channel (12, 26, 38), dopamine transporter (27, 30), and amino acid transporter 2 (9). Structurally, members of the Nedd4 family have three features: a C2 (Ca²⁺/lipid binding) domain at the NH₂-terminus, a catalytic HECT domain at the COOH-terminus, and two to four WW domains in the middle region. The C2 domain is involved in membrane targeting. The HECT domain functions through its catalytic cysteine residue to form a thioester intermediate with the COOH-terminus of ubiquitin, and it subsequently transfers ubiquitin to its substrate. The WW domain contains ∼40 amino acids in length and two conserved tryptophan (W) residues. The WW domain has been widely recognized for its role in recognizing and interacting with specific protein substrate (4, 25, 30).

In the present study, we demonstrated that Nedd4-1 is an important ubiquitin ligase for hOAT1 ubiquitination, expression, and transport activity. Furthermore, we identified that two of the four WW domains of Nedd4-1 are critically involved in its binding and regulation of hOAT1.

MATERIALS AND METHODS

Materials.

Monkey kidney COS-7 cells and human embryonic kidney (HEK)-293T cells were purchased from the American Type Culture Collection (Manassas, VA). [³H]p-aminohippuric acid (PAH) was purchased from Perkin-Elmer (Waltham, MA). The membrane-impermeable biotinylation reagent sulfo-NHS-SS-biotin, streptavidin-agarose beads, and protein G-agarose beads were purchased from Pierce (Rockford, IL). cDNA for human hemagglutinin-Nedd4-1 was purchased from Addgene (Cambridge, MA), a resource deposited by the laboratory of Joan Massague (39). Mouse anti-Myc antibody (9E10) was purchased from Roche (Indianapolis, IN). Mouse anti-ubiquitin antibody P4D1 was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit anti-Nedd4-1 antibody and mouse anti-Nedd4-1 antibody were purchased from Abcam (Cambridge, MA) and R&D Systems (Minneapolis, MN), respectively. Nedd4-1 small interfering (si)RNA oligonucleotides (identification number s9415, Silencer Select) and negative control siRNA (catalog number 4390843, Silencer Select) were purchased from Ambion (Grand Island, NY). All other reagents were purchased from Sigma-Aldrich (St. Louis, MO).

Cell culture and transfection.

Parental COS-7 cells, COS-7 cells stably expressing hOAT1-Myc, or HEK-293T cells stably expressing human OAT1-Myc were cultured in DMEM supplemented with 10% FBS at 37°C in 5% CO₂. Myc was tagged to the COOH-terminus of hOAT1 for the immunodetection of hOAT1 (17). Transfection with plasmids/siRNAs was carried out using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. Cells were harvested following 48 h after transfection for further experiments.

Site-directed mutagenesis.

Nedd4-1 ligase-dead mutant Nedd4-1/C867S and Nedd4-1 WW domain mutants were generated using a QuickChange site-directed mutagenesis kit from Agilent Technologies (Santa Clara, CA) following the manufacturer's instructions. The sequences of the mutants were confirmed by the dideoxy chain termination method.

Transport measurements.

Cells were plated in 48-well plates. Transport measurements were carried out at room temperature. For each well, uptake solution was added. The uptake solution consisted of PBS/Ca²⁺/Mg²⁺ (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na₂HPO4, 1.4 mM KH₂PO4, 0.1 mM CaCl₂, and 1 mM MgCl₂, pH 7.3) and [³H]PAH (20 μM). At the time points indicated, the uptake process was stopped by aspirating the uptake solution and rapidly washing the cells with ice-cold PBS solution. Cells were then solubilized in 0.2 N NaOH, neutralized in 0.2 N HCl, and aliquotted for liquid scintillation counting.

Cell surface biotinylation.

The cell surface expression level of hOAT1 was examined using the membrane-impermeable biotinylation reagent sulfo-NHS-SS-biotin. Cells were plated in six-well plates. Each well of cells was incubated with 1 ml NHS-SS-biotin (0.5 mg/ml in PBS/Ca²⁺/Mg²⁺) in two successive 20-min incubations on ice with very gentle shaking. The reagent was freshly prepared for each incubation. After biotinylation, each well was briefly rinsed with 3 ml PBS/ Ca²⁺/Mg²⁺ containing 100 mM glycine and then incubated with the same solution for 30 min on ice to ensure complete quenching of the unreacted sulfo-NHS-SS-biotin. Cells were then lysed on ice for 30 min in 400 μl lysis buffer [10 mM Tris·HCl, 150 mM NaCl, 1 mM EDTA, 0.1% SDS, and 1% Triton X-100 with 1/100 protease inhibitor mixture and 20 mM N-ethylmaleimide (NEM)]. Cell lysates were cleared by centrifugation at 16,000 g at 4°C. Streptavidin-agarose beads (40 μl) were then added to the supernatant to isolate cell membrane proteins. hOAT1 (tagged with Myc at its COOH-terminus) was detected in the pool of surface proteins by SDS-PAGE and immunoblot analysis using anti-Myc antibody 9E10.

Ubiquitination assay.

To inhibit proteasomal degradation of ubiquitinated hOAT1, cells were treated with 100 μM N-acetyl-Leu-Leu-norleucinal for 2 h before cell harvest. Cells were then lysed with ubiquitination lysis buffer [20 mM Tris·HCl (pH 7.5), 1% Triton X-100, 2 mM EDTA, and 25 mM NaF] with freshly added 1% of proteinase inhibitor cocktail and 20 mM NEM. hOAT1 was then immunoprecipitated with anti-Myc antibody followed by immunoblot analysis with anti-ubiquitin antibody P4D1.

Preparation of rat kidney slices.

Sprague-Dawley rats (200–250 g, male) were euthanized by CO₂ inhalation, and the kidneys were immediately placed into freshly oxygenated ice-cold saline. Tissue slices (<0.5 mm, 5–10 mg wet wt) were cut with a Stadie-Riggs microtome and kept in oxygenated ice-cold saline afterward until homogenized and lysed.

All animal experiments were conducted following guidelines described in the guide for the care and use of laboratory animals (Association for Assessment and Accreditation of Laboratory Animal Care) as well as requirements established by the animal protocol approved by the Rutgers Institutional Animal Care and Use Committee.

Coimmunoprecipitation.

Cells or rat kidney slices were lysed with immunoprecipitation lysis buffer [10 mM Tris·HCl (pH 7.5), 10 mM NaCl, 0.5% Triton X-100, 2 mM EDTA, and 10% glycerol] with freshly added 1% proteinase inhibitor cocktail and 20 mM NEM. Cell lysates were precleared with protein G-agarose beads to reduce nonspecific binding at 4°C for 1.5 h. Anti-Myc antibody (1:100) was incubated with the appropriate volume of protein G-agarose beads at 4°C for 1.5 h. The precleared protein sample was then mixed with antibody-bound protein G-agarose beads and underwent end-over-end rotation at 4°C overnight. Proteins bound to the protein G-agarose beads were eluted with urea buffer containing β-mercaptoethanol and analyzed by immunoblot analysis with the indicated antibodies.

Immunofluorescence analysis.

Paraffin-embedded rat kidney slices were kindly provided by Dr. Ping L Zhang (Department of Anatomic Pathology, William Beaumont Hospital, Royal Oak, MI). For immunodetection of target proteins, sections were rehydrated through serial washes in xylene and graded alcohol solutions. Nonspecific binding was blocked using blocking buffer (1.5% goat serum + 1.5% BSA) for 1 h at room temperature. Primary antibodies [goat anti-mouse Nedd4-1 (1:50) and goat anti-rabbit OAT1 (1:100)] were diluted in blocking buffer and incubated with the kidney slices overnight at 4°C. After a wash in PBS, secondary antibody [Alexa fluor 633 goat anti-mouse IgG (H+L) (1:500) or Alexa fluor 555 goat anti-rabbit IgG (H+L) (1:1,000)] were incubated with the kidney slices at room temperature for 1 h. After a wash, coverslips were mounted on slides with ProLong Gold antifade mountant solution for image acquisition and analysis. Samples were visualized with a Zeiss LSM-510 laser scanning microscope (Carl Zeiss, Thornwood, NY).

Electrophoresis and immunoblot analysis.

Protein samples were resolved on 7.5% SDS-PAGE minigels and electroblotted on to polyvinylidene difluoride membranes. Blots were blocked for 1 h with 5% nonfat dry milk in PBS-0.05% Tween 20, washed, and incubated overnight at 4°C with appropriate primary antibodies followed by horseradish peroxidase-conjugated secondary antibodies. Signals were detected by the SuperSignal West Dura Extended Duration Substrate kit (Pierce). Nonsaturating, immunoreactive protein bands were quantified by scanning densitometry with the FluorChem 8000 imaging system (Alpha Innotech, San Leandro, CA).

Data analysis.

Each experiment was repeated a minimum of three times. Statistical analysis from multiple experiments was performed using Student's paired t-tests. P values of <0.05 were considered as significant.

RESULTS

Effect of Nedd4-1 on hOAT1 ubiquitination.

We examined whether Nedd4-1 is an ubiquitin ligase for hOAT1. hOAT1-expressing COS-7 cells were transfected with cDNAs for wild-type Nedd4-1 or for the ubiquitin ligase-dead mutant Nedd4-1/C867S. The ligase-dead mutant was unable to transfer ubiquitin to its target protein (24, 38). Transfected cells were then lysed, and hOAT1 was immunoprecipitated by anti-Myc antibody (Myc was tagged to hOAT1) followed by immunoblot analysis with anti-ubiquitin antibody. As shown in Fig. 1A, the ubiquitin-immunoreactive signal (control lane) displayed a smeary band at 180 kDa, ∼100 kDa larger than the size of hOAT1 (∼80 kDa). Given that each ubiquitin molecule is ∼8.5 kDa, hOAT1 is most likely to be polyubiquitinated or multiubiquitinated. Overexpression of Nedd4-1 resulted in enhanced hOAT1 ubiquitination by 36.8 ± 9.0% compared with that in control, whereas the ligase-dead mutant Nedd4-1/C867S had no stimulatory effect. Moreover, the differences in ubiquitination were not due to the differences in the amount of hOAT1 immunoprecipitated as a similar amount of hOAT1 was immunoprecipitated in all samples under these conditions (Fig. 1C). Taken together, these data suggest that the ligase activity of Nedd4-1 is required for hOAT1 ubiquitination.

Fig. 1.

Neural precursor cell expressed, developmentally downregulated (Nedd)4-1 enhanced human organic anion transporter (hOAT)1 ubiquitination. A: hOAT1-expressing COS-7 cells were transfected with cDNAs for wild-type Nedd4-1 or for the ubiquitin ligase-dead mutant Nedd4-1/C867S. Transfected cells were then lysed, and hOAT1 was immunoprecipitated (IP) with anti-Myc antibody followed by immunoblot (IB) analysis with anti-ubiquitin antibody (anti-Ub). B: densitometry plot of results from A as well as from other repeat experiments. Values are means ± SE; n = 3. *P < 0.05. C: the blot from A was reprobed with anti-Myc antibody to determine the amount of hOAT1 immunoprecipitated. It is important to note that the hOAT1 detected by anti-Myc antibody at ∼80 kDa mainly reflected nonubiquitinated hOAT1 as the signals for ubiquitinated hOAT1 spread out in a wide range (centered at ∼180 kDa) and therefore were relatively weak.

As an independent approach, we used a siRNA strategy to abrogate endogenous Nedd4-1 and evaluated the role of Nedd4-1 in hOAT1 ubiquitination. As shown in Fig. 2A, top, endogenous Nedd4-1 expression was effectively reduced in Nedd4-1 siRNA-transfected cells compared with that in scrambled siRNA-transfected negative control cells. When the same blot was reprobed with anti-β-actin, it was clear that the housekeeping protein β-actin was not affected under these conditions (Fig. 2A, bottom). We then proceeded to examine the role of Nedd4-1 in hOAT1 ubiquitination. As shown in Fig. 2B, top, the ubiquitination of hOAT1 was much reduced in cells transfected with Nedd4-1-specific siRNA compared with that in control cells. Moreover, the differences in hOAT1 ubiquitination were not due to the differences in the amount of hOAT1 immunoprecipitated, as evident when the same immunoblot was reprobed with anti-Myc antibody (Myc was tagged to hOAT1). A similar amount of hOAT1 was immunoprecipitated in both samples under these conditions (Fig. 2B, bottom). These results once again confirm the important role of Nedd4-1 in hOAT1 ubiquitination.

Fig. 2.

Knockdown of endogenous Nedd4-1 decreased hOAT1 ubiquitination. A: hOAT1-expressing COS-7 cells were transfected with negative/scrambled control small interfering (si)RNA or with Nedd4-1-specific siRNA. siRNA effectiveness was then tested by probing the lysis sample with anti-Nedd4-1 antibody. The same immunoblot was reprobed by β-actin antibody to determine the amount of the housekeeping protein marker β-actin. B: hOAT1-expressing COS-7 cells, transfected with negative/scrambled control siRNA or with Nedd4-1-specific siRNA, were lysed, and hOAT1 was immunoprecipitated (IP) with anti-Myc antibody followed by IB analysis with anti-ubiquitin antibody (anti-Ub). The same immunoblot was reprobed by anti-Myc antibody to determine the amount of hOAT1 immunoprecipitated. The experiments were performed at least three times.

Effect of Nedd4-1 on hOAT1 expression.

Next, we investigated the role of Nedd4-1 in hOAT1 expression both at the cell surface and in total cell lysates. The surface expression of hOAT1 was evaluated through a biotinylation assay, as described in materials and methods. As shown in Fig. 3A, overexpression of Nedd4-1 resulted in a decrease in hOAT1 expression at the cell surface by 40.9 ± 5.7% compared with that in control cells. Total expression of hOAT1 was also decreased in these cells by 50.9 ± 6.7% compared with that in control cells (Fig. 3C), whereas overexpression of the ligase-dead mutant Nedd4-1/C867S was without any significant effect (Figs. 3, A and C). Such a change in hOAT1 expression was not due to the general perturbation of cellular proteins as the expression of the cell protein marker β-actin was not affected under these conditions (not shown).

Fig. 3.

Effect of Nedd4-1 on hOAT1 expression. A: expression of hOAT1 at the cell surface. hOAT1-expressing COS-7 cells were transfected with cDNAs for wild-type Nedd4-1 or for the ubiquitin ligase-dead mutant Nedd4-1/C867S. Cell surface biotinylation was performed. Biotinylated/cell surface proteins were separated with streptavidin beads and analyzed by IB analysis with anti-Myc antibody. B: densitometry plot of results from A as well as from other repeat experiments. Values are means ± SE; n = 3. *P < 0.05. C: expression of hOAT1 in total cell lysates. hOAT1-expressing COS-7 cells were transfected with cDNAs for wild-type Nedd4-1 or for the ubiquitin ligase-dead mutant Nedd4-1/C867S. Transfected cells were lysed, and the amount of hOAT1 was measured by IB analysis with anti-Myc antibody. D: densitometry plot of results from A as well as from other repeat experiments. Values are means ± SE; n = 3. *P < 0.05

Effect of Nedd4-1 on hOAT1 transport activity and transport kinetics.

As a cell membrane transporter, the amount of hOAT1 at the cell surface is critical for its transport activity. As described above (Fig. 3), Nedd4-1 reduced hOAT1 expression at the cell surface. In this experiment, we explored whether the altered surface expression translated into a hOAT1 functional change. As shown in Fig. 4A, overexpression of Nedd4-1 significantly suppressed hOAT1-mediated transport of [³H]PAH, a prototypical substrate for hOAT1, by 31.8 ± 5.5% compared with that in control cells, whereas overexpression of the ligase-dead mutant Nedd4-1/C867S was without any significant effect. To examine the mechanism of Nedd4-1-induced inhibition of hOAT1 activity, we determined hOAT1-mediated [³H]PAH uptake at different substrate concentrations. An Eadie-Hofstee analysis of the derived data (Fig. 4B) showed that transfection of Nedd4-1 resulted in a decrease in the maximal transport velocity of hOAT1 (775.19 ± 32.08 pmol·mg⁻¹·3 min⁻¹ in control cells and 606.16 ± 43.56 pmol·mg⁻¹·3 min⁻¹ in cells transfected with Nedd4-1), with no significant change in the substrate-binding affinity K_m of the transporter (86.24 ± 5.62 μM in control cells and 85.07 ± 4.96 μM in cells transfected with Nedd4-1). These results indicate that Nedd4-1 is an important regulator for hOAT1 function.

Fig. 4.

Effect of Nedd4-1 on hOAT1 transport activity and kinetics. A: hOAT1-expressing COS-7 cells were transfected with cDNAs for wild-type Nedd4-1 or for the ubiquitin ligase-dead mutant Nedd4-1/C867S. Three-minute uptake of p-aminohippuric acid ([³H]PAH; 20 μM) was then performed at room temperature (uptake was linear up to 10 min). Uptake activity was expressed as a percentage of the uptake measured in control cells. Data represent uptake into hOAT1-transfected cells minus uptake into mock cells (parental COS-7 cells). Values are means ± SE; n = 3. *P < 0.05. B: hOAT1-expressing cells were transfected with Nedd4-1 or control vector. Three-minute uptake of [³H]PAH was measured at concentrations of 5–200 μM. Values are means ± SE; n = 3. V, velocity; S, substrate concentration.

Nedd4-1 interaction with hOAT1.

Nedd4-1 has been shown to interact either directly or indirectly with its target proteins (6, 22), and, thus, we assessed whether there was an association between Nedd4-1 and hOAT1 through a coimmunoprecipitation assay. Nedd4-1 cDNA or control vector were transfected into hOAT1-expressing COS-7 cells. hOAT1 was then immunoprecipitated by anti-Myc antibody (Myc was tagged to hOAT1) followed by immunoblot analysis with anti-Nedd4-1 antibody. As shown in Fig. 5A, in hOAT1-expressing cells, a strong Nedd4-1 signal was detected in hOAT1 immunoprecipitates in Nedd4-1-transfected cells compared with that in control vector-transfected cells. The Nedd4-1 signal was not detected in parental cells, where hOAT1 was not expressed. These results indicate that there was a direct and specific association between Nedd4-1 and hOAT1. Moreover, the differences in the amount of hOAT1-associated Nedd4-1 were not due to the differences in the amount of hOAT1 immunoprecipitated as the amount of hOAT1 pulled down by anti-Myc antibody was similar in both control vector-transfected and Nedd4-1-transfected cells (Fig. 5B). Similar results were obtained when these experiments were performed in HEK-293T cells (Fig. 5, C and D), suggesting that the association of hOAT1 with Nedd4-1 is not cell type specific but is rather a general feature of this transporter.

Fig. 5.

Interactions between Nedd4-1 and hOAT1 in cultured cells. A: interactions between Nedd4-1 and hOAT1 in COS-7 cells. hOAT1-expressing COS-7 cells were transfected with control vector or cDNA for wild-type Nedd4-1. Transfected cells were lysed, and hOAT1 was then immunoprecipitated (IP) with anti-Myc antibody followed by IB analysis with anti-Nedd4-1 antibody. Parental cells, which do not express hOAT1, were also transfected with Nedd4-1 as a negative control. B: the blot from A was reprobed with anti-Myc antibody to determine the amount of hOAT1 immunoprecipitated. C: interactions between Nedd4-1 and hOAT1 in human embryonic kidney (HEK)-293T cells. hOAT1-expressing HEK-293T cells were transfected with control vector or cDNA for wild-type Nedd4-1. Transfected cells were lysed, and hOAT1 was then immunoprecipitated (IP) with anti-Myc antibody followed by IB analysis with anti-Nedd4-1 antibody. D: the blot from C was reprobed with anti-Myc antibody to determine the amount of hOAT1 immunoprecipitated.

The physical interaction between hOAT1 and Nedd4-1 was further examined in rat kidney slices, where both OAT1 (Fig. 6A, samples 1–5) and Nedd4-1 (Fig. 6B, samples 1–5) are endogenously expressed. Rat kidney slices were lysed, and OAT1 was then immunoprecipitated with anti-OAT1 antibody or with IgG (as negative control) followed by immunoblot analysis with anti-Nedd4-1 antibody. As shown in Fig. 6C, Nedd4-1 was detected in OAT1 immunoprecipitates in all rat kidney samples (samples 1–4). No amount of Nedd4-1 was detected when OAT1 was immunoprecipitated with negative control IgG instead of hOAT1-specific antibody (rat kidney sample 5). These results provide physiological relevance of the interaction between OAT1 and Nedd4-1.

Fig. 6.

Interaction between Nedd4-1 and OAT1 in rat kidney slices. A: total expression of OAT1 in rat kidney slices. The kidney slices from rats (n = 5) were lysed, and 30 μg protein was loaded for IB analysis. The expression of OAT1 was detected by anti-OAT1 antibody. B: total expression of Nedd4-1 in rat kidney slices. The kidney slices from rats (n = 5) were lysed, and 30 μg protein was loaded for IB analysis. The expression of Nedd4-1 was detected by anti-Nedd4-1 antibody. C: interactions between Nedd4-1 and OAT1 in rat kidney slices. The kidney slices from rats (n = 5) were lysed, and OAT1 was then immunoprecipitated (IP) with anti-OAT1 antibody or with rabbit IgG (as negative control) followed by IB analysis with anti-Nedd4-1 antibody.

Immunolocalization of OAT1 and Nedd4-1.

The physiological relevance of the interaction between OAT1 and Nedd4-1 was further investigated by examining the cellular distribution of OAT1 and Nedd4-1 in rat kidney slices through immunofluorescence microscopy. Nedd4-1 was detected using anti-Nedd4-1 antibody combined with Alexa fluor 633-conjugated secondary antibody (red color). OAT1 was detected using anti-OAT1 antibody combined with Alexa fluor 555-conjugated secondary antibody (green color). As shown in Fig. 7B, OAT1 was expressed in the basolateral membrane of kidney proximal tubule cells, consistent with previous reports (2, 29), and Nedd4-1 partially colocalized with OAT1 (Fig. 7C, yellow color).

Fig. 7.

Immunolocalization of OAT1 and Nedd4-1 in rat kidney slices. Paraffin-embedded rat kidney sections were rehydrated and incubated with specific antibodies as described in materials and methods. Fluorescence images were taken for Nedd4-1 (A, red) and OAT1 (B, green). The merged image of Nedd4-1 and OAT1 is shown as orange/yellow (C). Bar = ∼20 μm.

Nedd4-1 interactions with hOAT1 through its WW2 and WW3 domains.

Members of the Nedd4 family have been shown to recognize and to interact with protein substrate through specific protein-protein interacting WW domains (6, 13). Nedd4-1 has four WW domains (WW1–WW4). We mutated two amino acid residues in each of the four WW domains (Mut-WW1: V210W/H212G, Mut-WW2: V367W/H369G, Mut-WW3, I440W/H442G, and Mut-WW4: I492W/H494G, respectively). Mutation of these amino acids has been previously proven to inactivate these domains and, therefore, affect the ability of Nedd4-1 to bind its substrate. However, inactivation of these WW domains does not affect the ligase activity of Nedd4, which is solely dependent on the single conserved cysteine residue in the HECT region (25, 30). The mutant cDNAs were transfected into hOAT1-expressing cells. As shown in Fig. 8A, similar amounts of wild-type Nedd4-1 and its mutants were detected in total cell lysates, indicating that the transfection efficiencies of cDNAs encoding Nedd4-1 and its mutants were comparable. The association of hOAT1 with wild-type Nedd4-1 and its WW domain mutants was then determined through coimmunoprecipitation experiments. Interestingly, a much less amount of Mut-WW2 and Mut-WW3 was detected in hOAT1 immunoprecipitates compared with that of wild-type Nedd4-1, Mut-WW1, and Mut-WW4 (Fig. 8B), indicating that less amount of Mut-WW2 and Mut-WW3 was associated with hOAT1. Furthermore, the differences in the amount of Nedd4-1 mutants detected in hOAT1 immunoprecipitates were not due to the differences in the amount of hOAT1 immunoprecipitated as a similar amount of hOAT1 was immunoprecipitated in all samples under these conditions (Fig. 8D). These results suggest that WW2 and WW3 domains play an important role in the binding of Nedd4-1 to hOAT1.

Fig. 8.

Interactions between Nedd4-1 WW domain mutants and hOAT1. A: two amino acid residues in each of the four WW domains of Nedd4-1 were mutated (Mut-WW1: V210W/H212G, Mut-WW2: V367W/H369G, Mut-WW3: I440W/H442G, and Mut-WW4: I492W/H494G, respectively). The mutant cDNAs were then transfected into hOAT1-expressing COS-7 cells. Transfected cells were lysed, and the expression of the mutants was determined by IB analysis with anti-Nedd4-1 antibody. B: mutant-transfected cells were lysed, and hOAT1 was then immunoprecipitated (IP) with anti-Myc antibody followed by IB analysis with anti-Nedd4-1 antibody. C: densitometry plot of results from B as well as from other repeat experiments. Values are means ± SE; n = 3. *P < 0.05. D: the blot from B was reprobed with anti-Myc antibody to determine the amount of hOAT1 immunoprecipitated.

Effect of WW domain mutants of Nedd4-1 on hOAT1 ubiquitination.

The above experiments (Fig. 8) revealed that mutations at WW2 and WW3 domains of Nedd4-1 significantly interrupted the binding of Nedd4-1 to hOAT1. To examine whether hOAT1 ubiquitination was affected by such mutations, hOAT1-expressing cells were transfected with cDNAs for wild-type Nedd4-1 or Nedd4-1 WW domain mutants. hOAT1 was then immunoprecipitated by anti-Myc antibody (Myc was tagged to hOAT1) followed by immunoblot analysis with anti-ubiquitin antibody. As shown in Fig. 9A, there was a significant decrease in hOAT1 ubiquitination in Mut-WW2- or Mut-WW3-transfected cells compared with that in wild-type Nedd4-1-, Mut-WW1-, and Mut-WW4-transfected cells. Moreover, the differences in the amount of hOAT1 ubiquitination were not due to the differences in the amount of hOAT1 immunoprecipitated as a similar amount of hOAT1 was immunoprecipitated in all samples under these conditions (Fig. 9C). These data indicate that decreased binding affinity of Mut-WW2 and Mut-WW3 to hOAT1 leads to decreased hOAT1 ubiquitination.

Fig. 9.

Effect of Nedd4-1 WW domain mutants on hOAT1 ubiquitination. A: hOAT1-expressing COS-7 cells were transfected with cDNAs for wild-type Nedd4-1 or for Nedd4-1 WW domain mutants (Mut-WW1, Mut-WW2, Mut-WW3, or Mut-WW4). Transfected cells were then lysed, and hOAT1 was immunoprecipitated (IP) with anti-Myc antibody followed by IB analysis with anti-ubiquitin antibody (anti-Ub). B: densitometry plot of results from A as well as from other repeat experiments. Values are means ± SE; n = 3. *P < 0.05. C: the blot from A was reprobed with anti-Myc antibody to determine the amount of hOAT1 immunoprecipitated.

Effect of WW domain mutants of Nedd4-1 on hOAT1 transport activity.

To examine the functional consequence of WW domain mutations of Nedd4-1, we measured hOAT1-mediated transport of [³H]PAH in hOAT1-expressing cells transfected with cDNAs for wild-type Nedd4-1 or for its WW domain mutants. As shown in Fig. 10, overexpression of wild-type Nedd4-1, Mut-WW1, or Mut-WW4 had similar inhibitory effects on hOAT1-mediated transport. However, when cells overexpressed Mut-WW2 or Mut-WW3, the inhibitory effect was significantly reversed. The level of suppression of hOAT1-mediated transport activity by these mutants correlated with the extent of their binding to hOAT1.

Fig. 10.

Effect of Nedd4-1 WW domain mutants on hOAT1 transport activity. hOAT1-expressing COS-7 cells were transfected with cDNAs for wild-type Nedd4-1 or for Nedd4-1 WW domain mutants (Mut-WW1, Mut-WW2, Mut-WW3, or Mut-WW4). Transfected cells were then measured for the 3-min uptake of [³H]PAH (20 μM) at room temperature. Uptake activity was expressed as a percentage of the uptake measured in control cells. Data represent uptake into hOAT1-transfected cells minus uptake into mock cells (parental COS-7 cells). Values are means ± SE; n = 3. *P < 0.05

DISCUSSION

hOAT1 plays a pivotal role in the therapeutic efficacy of many drugs. The transport activity of hOAT1 critically depends on its amount at the cell surface. We have previously demonstrated that hOAT1 at the cell surface is not static (34). Instead, it constitutively internalizes from and recycles back to the cell surface. Ubiquitination of hOAT1 is a prerequisite for hOAT1 internalization (17). An increase in hOAT1 ubiquitination leads to an increase in the rate of hOAT1 internalization. The internalized hOAT1 then targets to the lysosome for degradation. As a result, the amount of hOAT1 at the cell surface is reduced and its transport activity is suppressed. Our present study demonstrated that Nedd4-1 is an important ubiquitin enzyme involved in this process.

Our present study was carried out in heterologous cells (monkey kidney COS-7 cells and human kidney HEK-293T cells) and in rat kidney slices. Both COS-7 cells and HEK-293T cells are of kidney origin and do not endogenously express any of the OAT isoforms, which allowed us to study transfected OAT1 without the interference from other OATs. Indeed, these cells have been widely used as model systems for studying the regulatory mechanisms underlying many kidney transport processes (3, 10, 15, 16, 18–20, 23, 32, 34, 37). Our experiments in rat kidney slices (Figs. 6 and 7) further confirmed the physiological relevance of our work, which will pave the way for the future investigation focusing on determining whether the same mechanisms are operative in vivo.

Members of the Nedd4 family of ligases contain a ∼40-kDa HECT domain at the COOH-terminus. The conserved active cysteine within the HECT domain could form a thioester intermediate with the ubiquitin molecule and catalyze the formation of an isopeptide bond between the activated ubiquitin and ε-amino groups of lysine on the substrate (11). Mutation of the conserved cysteine in the HECT domain to serine/alanine/glutamine could have a dominant negative effect on the ubiquitination and the subsequent function of the substrate, which could be a useful tool when studying the specific involvement of the ubiquitin ligase activity of Nedd4 family members. Indeed, our present study established that the ubiquitin ligase activity of Nedd4-1 is critical for its regulation of hOAT1 function as overexpression of Nedd4-1 significantly enhanced hOAT1 ubiquitination, decreased hOAT1 expression at the cell surface, and inhibited hOAT1 transport activity. In contrast, overexpression of the ubiquitin ligase-dead mutant Nedd4-1/C867S was without any effect on hOAT1 (Figs. 1, 3, and 4).

Our present study also suggests that the effect of Nedd4-1 on OAT1 occurs through a direct interaction between these two proteins as evidenced by our coimmunoprecipitation experiments, in which the immunoprecipitation of one protein resulted in the pulling down of another protein (Figs. 5 and 6). This observation was made not only in cultured cells but also in rat kidney slices, where both Nedd4-1 and OAT1 were endogenously expressed. Furthermore, the colocalization of OAT1 and Nedd4-1 in rat kidney slices was confirmed by our immunofluorescence experiment (Fig. 7), demonstrating the physiological significance of the interaction between Nedd4-1 and OAT1.

WW domains are well recognized as protein-protein interaction domains that bind proline-rich sequences in target proteins (7, 14, 30) through Van der Waals contacts (5). Nedd4-1 contains four WW domains. These domains have been shown to play critical roles in the binding of Nedd4-1 to its substrate, despite having differential selectivity and affinity toward its target proteins. For example, the WW1 domain of Nedd4-1 does not bind epithelial Na⁺ channel subunits but all other WW domains do, albeit at varying affinities (6, 8). The WW1 and WW2 domains of Nedd4-2, a Nedd4-1 homolog, are critical for the ubiquitination of an intestinal apical Ca²⁺ entry channel, the transient receptor potential vanilloid 6 channel (36). Our present study also uncovered a considerable level of selectivity in the binding affinity of the WW domains of Nedd4-1 toward hOAT1. WW2 and WW3 domains but not WW1 and WW4 domains were found to be necessary for the binding of Nedd4-1 to hOAT1. When the WW2 or WW3 domain in Nedd4-1 was inactivated through site-directed mutation, the binding affinity of these Nedd4-1 mutants (Mut-WW2 and Mut-WW3) for hOAT1 was weakened (Fig. 8), their ability to ubiquitinate hOAT1 was decreased (Fig. 9), and their inhibitory effect on hOAT1-mediated transport was significantly diminished compared with that of wild-type Nedd4-1 (Fig. 10). It should be noted that hOAT1 does not have conventional WW domain-binding (L/P)PXY motifs. Yet, we were able to detect strong interactions between Nedd4-1 and hOAT1, indicative of an interaction through noncanonical sequences. This is in line with the previous reports showing that Nedd4-1 has physical association with the β₂-adrenergic receptor (22) and mammalian Na⁺/H⁺ exchanger 1 (24), both of which do not contain the conventional WW domain-binding motifs.

Taken together, our results reveal, for the first time, that Nedd4-1 is an important regulator for hOAT1 ubiquitination, expression, and transport activity (Fig. 11). The WW2 and WW3 domains of Nedd4-1 are critical for its interaction with hOAT1. Our study provides important insights into the understanding of the molecular and cellular basis of hOAT1 regulation.

Fig. 11.

Nedd4-1 binds and regulates hOAT1 through its WW2 and WW3 domains.

GRANTS

This work was supported by National Institute of General Medical Sciences Grants R01-GM-079123 and R01-GM-097000 (to G. You).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

D.X. and G.Y. conception and design of research; D.X., H.W., C.R.G., Z.P., P.L.Z., and J.Z. performed experiments; D.X. and G.Y. analyzed data; D.X. and G.Y. interpreted results of experiments; D.X. and G.Y. prepared figures; D.X., H.W., C.G., Z.P., PL.Z., J.Z., and G.Y approved final version of manuscript; D.X. and G.Y. edited and revised manuscript; G.Y. drafted manuscript.