CFTR-associated ligand is a negative regulator of Mrp2 expression

Abstract

The multidrug resistance-associated protein 2 (Mrp2) is an ATP-binding cassette transporter that transports a wide variety of organic anions across the apical membrane of epithelial cells. The expression of Mrp2 on the plasma membrane is regulated by protein-protein interactions. Cystic fibrosis transmembrane conductance regulator (CFTR)-associated ligand (CAL) interacts with transmembrane proteins via its PDZ domain and reduces their cell surface expression by increasing lysosomal degradation and intracellular retention. Our results showed that CAL is localized at the trans-Golgi network of rat hepatocytes. The expression of CAL is increased, and Mrp2 expression is decreased, in the liver of mice deficient in sodium/hydrogen exchanger regulatory factor-1. To determine whether CAL interacts with Mrp2 and is involved in the posttranscriptional regulation of Mrp2, we used glutathione S-transferase (GST) fusion proteins with or without the COOH-terminal PDZ binding motif of Mrp2 as the bait in GST pull-down assays. We demonstrated that Mrp2 binds to CAL via its COOH-terminal PDZ-binding motif in GST pull-down assays, an interaction verified by coimmunoprecipitation of these two proteins in cotransfected COS-7 cells. In COS-7 and LLC-PK1 cells transfected with Mrp2 alone, only a mature, high-molecular-mass band of Mrp2 was detected. However, when cells were cotransfected with Mrp2 and CAL, Mrp2 was expressed as both mature and immature forms. Biotinylation and streptavidin pull-down assays confirmed that CAL dramatically reduces the expression level of total and cell surface Mrp2 in Huh-7 cells. Our findings suggest that CAL interacts with Mrp2 and is a negative regulator of Mrp2 expression.

the multidrug resistance-associated protein 2 (Mrp2, ABCC2) is a member of the ATP-binding cassette (ABC) transporter superfamily that transports a wide variety of organic anions across the apical membrane of epithelial cells (1, 12, 26). In the liver, Mrp2 is localized at the canaliculi of hepatocytes and plays an important role in bile formation and detoxification by transporting bile salt conjugates, bilirubin-glucuronides, and drug conjugates into bile (8, 21). In humans, mutations of the MRP2 gene lead to impaired targeting and expression of functional Mrp2 protein at the canalicular membrane, causing Dubin-Johnson syndrome, characterized by conjugated hyperbilirubinemia and defects in the excretion of several endogenous and xenobiotic compounds (14, 22).

Mrp2 expression on the hepatocyte canalicular membrane is a highly regulated but poorly understood process. In addition to regulation by nuclear receptors, such as FXR and CAR, the synthesis, apical targeting, and localization of Mrp2 protein are also regulated by a variety of factors, such as bile acids, cyclic AMP, hyperosmolarity, and oxidative stress, which promote rapid exocytic insertion or endocytic internalization of Mrp2 to/from the plasma membrane (8, 19, 24). Growing evidence suggests that interacting proteins play important roles in this dynamic recycling of Mrp2 between the apical membrane and endosomal subapical compartments. Radixin, a member of the ezrin-radixin-moesin (ERM) family of cytoskeletal proteins that cross-links F-actin cytoskeleton to membrane transporters, has been shown to interact with Mrp2. Deletion of radixin in mice results in selective loss of Mrp2 from the canalicular membrane, leading to hyperbilirubinemia (17). In rat hepatocytes, small interfering RNA-induced suppression of radixin results in dislocation of Mrp2, as well as other apical transporters (27). The sodium/hydrogen exchanger regulatory factor-1 (NHERF-1), also known as ERM-binding phosphoprotein 50 (EBP50), a scaffolding protein that contains two PDZ domains at its NH₂-terminus and a COOH-terminal domain that binds the ERM proteins, also interacts with Mrp2 via PDZ interaction (10, 18, 23, 29). Our laboratory previous reported that Mrp2 protein is reduced in the liver and kidney of NHERF-1 knockout mice compared with wild-type (WT) mice, although Mrp2 mRNA remains unchanged (18). Thus NHERF-1 positively regulates Mrp2 expression by posttranscriptional mechanisms.

To better understand the regulation of Mrp2 by protein-protein interactions, we investigated the role of another protein, the cystic fibrosis transmembrane conductance regulator (CFTR)-associated ligand (CAL), in Mrp2 expression. CAL, also known as GOPC (Golgi-associated PDZ and coiled-coil motif-containing protein), PIST (PDZ domain protein interacting specifically with TC10), and FIG (fused in glioblastoma), contains a PDZ domain and two coiled-coil domains (5, 20, 31). It is ubiquitously expressed in various tissues and well conserved in evolution. In acinar cells, CAL was located primarily at the Golgi apparatus, colocalizing with trans-Golgi network markers (5). CAL has been reported to interact with transmembrane proteins, including ion channels, such as the CFTR, and adhesion molecules, such as cadherin 23, via its PDZ domain and has been shown to reduce cell surface expression of these proteins by increasing lysosomal degradation and intracellular retention (5, 6, 30). Based on the structural similarity between Mrp2 and CFTR, we hypothesized that CAL interacts with Mrp2 and negatively influences the expression of Mrp2 by posttranscriptional regulation.

MATERIALS AND METHODS

Animals.

NHERF-1^−/− mice bred into a C57BL/6 background for six generations were introduced from the laboratory of Dr. Edward J. Weinman (Departments of Medicine and Physiology, University of Maryland School of Medicine, Baltimore, MD) and were previously described (25). WT C57BL/6 mice were purchased from the Jackson Laboratory (Bar Harbor, ME). Age-matched (13–17 wk) male mice were used for the present study. Animals were housed in a temperature- and humidity-controlled room under a light cycle with free access to food and water. All experiments involving animals were carried out in accordance with the Guide for the Care and Use of Laboratory Animals, as adopted and promulgated by the National Institutes of Health and were approved by Yale University's Institutional Animal Care and Use Committee.

Collagen sandwich-cultured hepatocytes.

Hepatocytes were isolated from male Sprague-Dawley rat liver by collagenase perfusion, as described previously from this laboratory (2). Cells were seeded onto collagen-coated glass coverslips in William's E medium with the addition of 5% fetal bovine serum, 10 mmol/l HEPES buffer, 2 mmol/l l-glutamine, 1 μmol/l dexamethasone, 4 mg/l insulin, 100 U/ml penicillin, 100 μg/ml streptomycin, and 10 mg/l gentamicin. Cells were overlaid with gelled collagen 24 h after seeding.

Plasmid constructs and expression of recombinant proteins.

Hemagglutinin (HA)-tagged full-length human CAL (GenBank accession no. AF45008) expression plasmid was a gift from Dr. Bruce A. Stanton (Dartmouth Medical School, Hanover, New Hampshire). Full-length rat Mrp2 (rMrp2) (Abcc2) cDNA (GenBank accession no. NM_012833) cloned into expression vector pcDNA3 was generously provided by Dr. Dietrich Keppler (German Cancer Research Center, Heidelberg, Germany). Full-length rat NHERF-1/EBP50 cDNA (GenBank accession no. AF154336, a gift from Dr. R. Brian Doctor, University of Colorado Health Sciences Center, Denver, CO) and full-length rat bile salt export pump (rBsep) cDNA (kindly provided by Dr. Bruno Stieger, University Hospital, Zurich, Switzerland) were subcloned into pcDNA3 vector. rMrp2 cDNA encoding the COOH-terminal intracellular portion with or without the PDZ-binding motif was amplified by PCR and cloned into pGEX-3X vector (Amersham Pharmacia Biotech) for expression of glutathione S-transferase (GST)-rMrp2 fusion protein in transformed E. coli cells. GST and GST-rMrp2 fusion protein were purified with glutathione Sepharose 4B beads (GE Healthcare/Amersham, Piscataway, NJ), according to the manufacturer's instructions. Purified GST and GST-rMrp2 fusion protein were dialyzed against 50 mM Tris·HCl, pH 7.5, 1 mM EDTA, and 0.1% 2-mercaptoethanol and were quantified by Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA).

Cell culture and transfection.

Human embryonic kidney (HEK)-293 and COS-7 cells from ATCC and Huh-7 cells from the laboratory of Dr. Yung-Chi Cheng (Department of Pharmacology, Yale University School of Medicine) were maintained in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA), supplemented with 10% fetal bovine serum and antibiotics. LLC-PK1 cells were bought from ATCC and maintained in Medium 199 (Invitrogen), supplemented with 10% fetal bovine serum and antibiotics. Cells were kept in a 37°C, 5% CO₂, humidified incubator. Cells in six-well plates (BD Falcon, Franklin Lakes, NJ) were transfected with plasmid DNA using Lipofectamine 2000 (Invitrogen; for HEK-293, LLCPK-1, and Huh-7 cells) or Trans-IT-LT1 (Mirus Bio LLC, Madison, WI; for COS-7 cells), in Opti-MEM I Reduced Serum Medium (Invitrogen), according to the manufacturer's instructions.

GST pull-down assay and coimmunoprecipitation.

Twenty-four hours following transfection, HEK-293 cells were washed twice with ice-cold Dulbecco's phosphate-buffered saline (DPBS) and lysed at 4°C in lysis buffer containing 25 mM Tris·HCl, pH 7.2, 150 mM NaCl, 5 mM MgCl₂, 0.5% NP-40, 1 mM DTT, 5% glycerol, Halt protease inhibitor, and Halt phosphatase inhibitor cocktails (Thermo Scientific, Rockford, IL), and the cell lysates were cleared by centrifugation at 12,000 g for 10 min at 4°C. For GST pull-down assay, 200 μl of cleared cell lysates were incubated overnight with 2.5 μg of purified GST or GST-rMrp2 fusion protein at 4°C. The samples were then supplemented with glutathione Sepharose 4B beads and incubated for an additional 2 h at 4°C. For coimmunoprecipitation (co-IP) experiments, 200 μl of the cleared lysates were incubated overnight with anti-HA agarose beads from Pierce HA Tag IP/co-IP kit (Thermo Scientific) at 4°C. After centrifugation, the beads were washed extensively with TBS-T buffer containing 25 mM Tris·HCl, pH 7.2, 150 mM NaCl, and 0.05% Tween 20. The GST pull-down or co-IP complex were eluted with two times reducing sample buffer and subjected to SDS-PAGE and immunoblotting.

In vitro N-glycosidase F and endoglycosidase H digestion of cell lysates.

Lysates of transfected COS-7 cells were digested with peptide N-glycosidase (PNGase) F and endoglycosidase H (Endo H; New England Biolabs, Ipswich, MA), according to the manufacturer's instructions without boiling the samples.

Biotinylation and streptavidin pull-down of Huh-7 cell surface proteins.

Twenty-four hours following transfection, Huh-7 cells in poly-l-lysine-coated six-well plate were washed three times with ice-cold DPBS with Ca²⁺ and Mg²⁺ (DPBS-Ca²⁺/Mg²⁺) and incubated with EZ-Link Sulfo-NHS-SS-Biotin (Thermo Scientific; 1 mg/ml in DPBS-Ca²⁺/Mg²⁺; 1 ml/well) at 4°C for 1 h. Cells were then quenched with ice-cold DPBS-Ca²⁺/Mg²⁺ containing 100 mM glycine and washed with DPBS-Ca²⁺/Mg²⁺ before being lysed with 250 μl/well lysis buffer containing 50 mM Tris·HCl, pH 7.4, 150 mM NaCl, 1% Triton X-100, 1 mM EDTA, Halt protease inhibitor, and Halt phosphatase inhibitor cocktails. The cell lysates were cleared by centrifugation at 12,000 g for 10 min at 4°C. Approximately 150 μg of the cleared and quantified lysates were incubated overnight with prewashed streptavidin agarose beads (Thermo Scientific) at 4°C. After centrifugation, the beads were washed four times with the lysis buffer and once with DPBS-Ca²⁺/Mg². The streptavidin pull-down complex was eluted with two times reducing sample buffer and subjected to SDS-PAGE and immunoblotting.

Immunoblotting and densitometry analysis.

Protein samples were separated by SDS-PAGE and transferred to polyvinylidene difluoride membranes (Bio-Rad Laboratories, Hercules, CA). The membranes were blocked with 5% nonfat dry milk in TBS-T containing 10 mM Tris·HCl, pH 8.0, 150 mM NaCl, and 0.1% Tween 20 and then incubated overnight with the appropriate primary antibodies. After extensive washes with TBS-T, the membranes were incubated with goat anti-rabbit or goat anti-mouse IgG conjugated to horseradish peroxidase for 1 h at room temperature. After further washes, immune complexes were detected by SuperSignal West Pico Chemiluminescent Substrate or SuperSignal West Femto Maximum Sensitivity Substrate (Thermo Scientific). For membrane stripping, membranes were incubated with Restore PLUS Western Blot Stripping Buffer (Thermo Scientific), according to the manufacturer's instructions. Membranes were then washed three times with TBS-T and reblotted. Densitometry scanning of protein bands was performed using TotalLab TL100 software (Nonlinear Dynamics, Newcastle Upon Tyne, UK).

Antibodies.

Rabbit polyclonal anti-GOPC (CAL) antibody was kindly provided by Dr. Ryoji Yao (JFCR Cancer Institute, Tokyo, Japan). Monoclonal antibody M2III-6 against Mrp2 (catalog no. ALX-801-016-C250) was purchased from Enzo Life Sciences (incorporating BIOMOL and Alexis, Plymouth Meeting, PA). Anti-HA polyclonal antibody (catalog no. 631207) was purchased from Clontech (Clontech Laboratories, Mountain View, CA). Polyclonal antibody against NHERF-1 (catalog no. AB5487) was purchased from Millipore (Billerica, MA). Polyclonal antibody anti-SPGP against Bsep (catalog no. PC-064) was purchased from Kamiya Biomedical (Seattle, WA). Mouse monoclonal antibodies against Hsp70 (catalog no. 610607) and TGN38 (catalog no. 610898) were purchased from BD Transduction Laboratories (BD Biosciences, San Jose, CA). Mouse monoclonal antibody against Na⁺-K⁺-ATPase-α₁ (C464.6) (catalog no. sc-21712) was from Santa Cruz Biotechnology (Dallas, TX). Monoclonal antibody against Gapdh (catalog no. G8795) was purchased from Sigma-Aldrich (St. Louis, MO).

Immunofluorescence microscopy.

Immunofluorescence microscopy was performed as previously described (27). Briefly, isolated rat hepatocyte in collagen gel sandwich culture were fixed with 4% paraformaldehyde and incubated with primary antibody for 2 h at room temperature, followed by secondary antibodies (Alexa 488/568, Molecular Probes, Eugene, OR) for 1 h. Images were acquired using a Zeiss LSM 510 confocal microscope (Thornwood, NJ) and processed with Adobe PhotoShop (San Jose, CA).

Statistical analysis.

All data were analyzed using the Student t-test and are expressed as means ± SD. A P value of <0.05 was considered significant.

RESULTS

CAL is localized at the trans-Golgi network of rat hepatocytes.

CAL has been detected in various tissues, including the liver (5, 31). However, it is unknown if CAL is present in hepatocytes where hepatic Mrp2 is exclusively expressed. To examine the expression and subcellular localization of CAL in hepatocytes, immunofluorescence microscopy was performed on cultures of isolated rat hepatocytes using affinity-purified anti-GOPC (CAL) antibody. As shown in Fig. 1A, CAL was detected in the perinuclear area and intracellular compartments of hepatocytes and colocalized with the trans-Golgi network marker, TGN38, whereas colocalization of Mrp2 and CAL on the canalicular membrane was not detected (Fig. 1B). These results confirmed that CAL is expressed as a Golgi-associated protein in hepatocytes.

Fig. 1.

Subcellular localization of CAL in rat hepatocytes. Rat hepatocytes were grown for 3 days in collagen sandwich-culture as described in materials and methods. Cells were then fixed with 4% paraformaldehyde and subjected to immunofluorescence microscopy detection. A: cells were labeled with polyclonal antibody against GOPC (CAL; red) and monoclonal antibody against the trans-Golgi network marker, TGN38 (green), before being stained by TO-PRO-3 for the nucleus (blue). Colocalization was visualized by image overlay (yellow). Note the intensive labeling of CAL in the perinuclear area and intracellular compartments and the colocalization of CAL and TGN38 (arrowheads). B: cells were labeled with polyclonal antibody against GOPC (CAL; red) and monoclonal antibody against Mrp2 (green), a marker of hepatocytes. Mrp2 was detected at the canalicular membrane, and its colocalization with CAL was not detected. Scale bar: 10 μm.

NHERF-1^−/− mice have increased hepatic CAL expression, accompanied by reduced membrane expression of Mrp2.

Our laboratory's previous studies showed that NHERF-1^−/− mice have reduced Mrp2 expression in the canaliculi and membrane-enriched fractions of the liver, indicating that NHERF-1 positively regulate Mrp2 expression in mouse hepatocytes (18). To examine whether CAL is involved in the reduction of Mrp2 expression in NHERF-1^−/− mouse liver, we performed immunoblotting on whole cell lysates of liver tissues from WT and NHERF-1^−/− mice. Our results showed that, compared with the WT, NHERF-1^−/− mice have an increased CAL protein expression in the hepatic whole cell lysates (Fig. 2), suggesting that CAL might adversely affect Mrp2 membrane expression in mouse hepatocytes.

Fig. 2.

Expression of CAL is significantly increased in the liver of NHERF-1^−/− mice. Representative immunoblots of whole cell lysates demonstrate increased expression of CAL protein in NHERF-1^−/− mouse liver. Numbers represent the relative ratio of CAL protein bands in wild-type (WT) and NHERF-1^−/− mice by densitometry analysis. Data are normalized to Gapdh. The amount of protein from the WT is set as 1. Values are means ± SD of 4 mice in each group. *P < 0.05.

Mrp2 binds to CAL via the COOH-terminal PDZ-binding motif in GST pull-down assays.

Mrp2 possesses a class I PDZ-binding motif at its intracellular COOH-terminus (HTEL in rat/mouse Mrp2, STKF in human MRP2) that has been shown to mediate its interaction with PDZ proteins (10, 18). To determine whether CAL and Mrp2 interact in vitro, plasmids expressing GST fusion proteins encoding rMrp2 COOH-terminus with or without the PDZ-binding motif were constructed (Fig. 3A). The purified GST or GST-rMrp2 fusion proteins were used as bait in GST pull-down assays with lysates of HEK-293 cells transfected with HA-CAL. Immunoblotting analysis demonstrated that HA-CAL was detected in the pull-down complex when the PDZ-binding motif of Mrp2 was included in the bait, but was not detected in the pull-down complex by GST alone or GST fusion proteins without the PDZ-binding motif (Fig. 3B), indicating that rMrp2 interacts with CAL, and the interaction was mediated through its COOH-terminal PDZ binding motif.

Fig. 3.

Mrp2 binds to CAL via the COOH-terminal PDZ-binding motif in GST pull-down assays. A: diagram of GST-rat Mrp2 COOH-terminus constructs for GST pull-down assays. a: Full-length rat Mrp2 consists of three membrane-spanning domains (in green) and an intracellular COOH-terminal tail (amino acid 1255–1541, in yellow). b: Construct 1 encodes GST (in orange) and amino acid 1255–1541 of rMrp2, including the COOH-terminal PDZ-binding motif that is composed of 4 amino acids (red asterisks). c: Construct 2 encodes GST and amino acid 1255–1538 of rMrp2 with deletion of the last 3 amino acids. d: Construct 3 encodes GST and amino acid 1255–1534 of rMrp2 with deletion of the last 7 amino acids. B: GST pull-down assays. HEK-293 cells were transfected with Lipofectamine 2000, and lysates of cells transfected with HA-CAL were incubated with GST control or GST-rMrp2 fusion protein encoded by the constructs shown in A. The samples were then supplemented with glutathione Sepharose 4B beads. Both the pull-down complex (top) and the unbound fraction (bottom) were immunoblotted with anti-HA antibody. Note the presence of HA-CAL in the pull-down complex when the COOH-terminus PDZ-binding motif of Mrp2 was included in the GST-Mrp2 bait, but was not detected in the pull-down complex when the COOH-terminus PDZ-binding motif was deleted in (Construct 1 vs. Construct 2 and Construct 3). The bottom panel confirms the presence of HA-CAL in the unbound fraction of all of the pull-down assays. Data are representative of three independent experiments.

Mrp2 but not Bsep coprecipitates with HA-CAL in cotransfected COS-7 cells.

To validate our GST pull-down assay results, we tested whether Mrp2 also interacts with CAL when these two proteins are coexpressed in cells. COS-7 cells were cotransfected with plasmids expressing full-length rMrp2 and HA-CAL. The cell lysates were precipitated with anti-HA agarose beads, and the pull-down complexes were analyzed by immunoblotting with anti-Mrp2 antibody. As shown in Fig. 4, Mrp2 coprecipitated with HA-CAL from cotransfected cells, but was not precipitated from cell lysates cotransfected with the Mrp2 plasmid and a vector control. To further determine the specificity of the interaction between Mrp2 and CAL, COS-7 cells were cotransfected with HA-CAL and a DNA construct encoding the rBsep, another liver-specific ABC transporter protein that functions at the apical/canalicular membrane of hepatocytes but does not possess PDZ-binding motifs. As expected, rBsep did not coprecipitate with HA-CAL from the cotransfected cells. These results demonstrated that CAL associates with Mrp2, but not Bsep, in cotransfected COS-7 cells.

Fig. 4.

Mrp2 but not Bsep coprecipitates with HA-CAL in cotransfected COS-7 cells. COS-7 cells were cotransfected with plasmids expressing full-length rMrp2 or rBsep and HA-CAL using Trans-IT-LT1. The cell lysates were precipitated with anti-HA agarose beads, and the pull-down complexes were analyzed by immunoblotting (IB) with anti-Mrp2 or anti-Bsep antibody. Note that Mrp2 was detected only in the co-IP complex from cells cotransfected with plasmids expressing both Mrp2 and HA-CAL (arrow), but was not detected from cells cotransfected with the Mrp2 plasmid and a vector control. In contrast, rBsep was absent in the co-IP complex, although it was detected in the cell lysates. IP, immunoprecipitation. Data are representative of three independent experiments.

CAL reduces the expression level of mature Mrp2 in cotransfected COS-7 and LLC-PK1 cells.

To determine whether CAL plays a role in Mrp2 expression, we cotransfected COS-7 cells with fixed quantity of rMrp2 construct and different molar ratios of HA-CAL or NHERF-1 constructs. Immunoblotting analysis showed that, in COS-7 cells cotransfected with Mrp2 and a vector control, only a high-molecular-mass band (∼200 kDa) of Mrp2 was detected (Fig. 5A, transfection 2). In contrast, when cells were cotransfected with Mrp2 and HA-CAL, Mrp2 was expressed as both the high-molecular-mass and low-molecular-mass (∼150 kDa) bands (Fig. 5A, transfections 3 and 4).

Fig. 5.

CAL reduces the expression level of mature Mrp2 in cotransfected COS-7 and LLC-PK1 cells. COS-7 (A and B) were cotransfected with fixed quantity of plasmid expressing full-length rMrp2 and different molar ratios of plasmids expressing HA-CAL or NHERF-1 to maintain equal quantity of total plasmid DNA in each transfection. Cells were transfected with Trans-IT-LT1 and were incubated for 24 h before harvest. Whole cell lysates were extracted and subjected to immunoblotting (IB). Mrp2 and Hsp70 (served as the loading control) were detected with their respective antibodies. Note that only a high-molecular-mass band (∼200 kDa) of Mrp2 was detected in the lysates of cells cotransfected with Mrp2 and a vector control (A, transfection 2). In contrast, both the high-molecular-mass band and a lower molecular-mass band (∼150 kDa) of Mrp2 were detected in cells cotransfected with Mrp2 and HA-CAL (A, transfection 3 and transfection 4). C: similar results were obtained with LLC-PK1 cells transfected with Lipofectamine 2000. B: 8 μg of whole cell lysates of transfected COS-7 (from transfection 2 and transfection 3 in A) were digested in vitro with PNGase F or Endo H without boiling the samples and then subjected to IB. Note that, while the high band of Mrp2 (∼200 kDa) shifted to the low band (∼150 kDa) of Mrp2 after PNGase F digestion, but remained unchanged after Endo H digestion, the low-molecular-mass band was sensitive to both PNGase F and Endo H treatment. Data are representative of 4 independent experiments.

To further investigate the nature of these two bands of Mrp2, we performed in vitro PNGase F and Endo H digestion of cell lysates of transfected COS-7 cells. As shown in Fig. 5B, the high-molecular-mass band was sensitive to PNGase F but resistant to Endo H. It shifted to the low band of Mrp2 after PNGase F digestion, but remained unchanged after Endo H digestion. This indicates that the high-molecular-mass band represents the fully glycosylated, mature form of Mrp2 (32). In contrast, the low-molecular-mass band was sensitive to both PNGase F and Endo H treatment, indicating that this band represents the core-glycosylated form of the protein. Densitometry analysis of the bands indicates that this effect of CAL on Mrp2 expression was dose dependent, with expression of the mature form of Mrp2 reduced to ∼60% of total Mrp2 in COS-7 cotransfected with equal quantities of the Mrp2 and HA-CAL constructs compared with cells cotransfected with Mrp2 and a vector control (Fig. 5A, transfection 3 vs. transfection 2). In addition, when cells were cotransfected with Mrp2 and NHERF-1 constructs, only the mature form of Mrp2 was detected, and the expression level of Mrp2 was comparable to that in cells transfected with Mrp2 and a vector control (Fig. 5A, transfections 2 and 5). Similar effects of CAL and NHERF-1 on Mrp2 were also detected in cotransfected LLC-PK1 cells, a polarized kidney epithelial cell line (Fig. 5C). These findings suggest that CAL, but not NHERF-1, reduces the expression level of mature Mrp2 in cotransfected COS-7 and LLC-PK1 cells.

CAL downregulates total as well as cell surface expression of Mrp2 in cotransfected Huh-7 cells.

To further determine whether CAL affects Mrp2 expression in physiological relevant cell types, we examined the expression of total as well as cell surface Mrp2 in cotransfected Huh-7, a human hepatoma cell line, by surface biotinylation and streptavidin pull-down assays. Interestingly, unlike the results obtained with COS-7 and LLC-PK1 cells, Mrp2 was expressed as a single band of ∼200 kDa in Huh-7 cells cotransfected with Mrp2 and HA-CAL (Fig. 6A). However, compared with cotransfection of Mrp2 and a vector control, cotransfection of HA-CAL significantly decreased the total as well as cell surface Mrp2 expression in Huh-7 cells in a dose-dependent manner (P < 0.05; Fig. 6B, transfections 1, 2, and 3), indicating that CAL also negatively regulates the expression of Mrp2 in Huh-7 cells. In contrast, the level of Mrp2 expression remained basically unchanged when cotransfected with NHERF-1 in these cells (Fig. 6, transfections 1 and 4).

Fig. 6.

Cotransfection with CAL but not NHERF-1 dramatically decreased total (A) and cell surface (B) Mrp2 expression in Huh-7 cells. Huh-7 cells were cotransfected with fixed quantity of plasmid expressing full-length rMrp2 and different molar ratios of plasmids expressing HA-CAL or NHERF-1 to maintain equal quantity of total plasmid DNA in each transfection. Cells were transfected for 24 h, and the cell surface proteins were labeled with membrane-impermeable EZ-Link Sulfo-NHS-SS-Biotin and pulled down by streptavidin agarose beads, as described in materials and methods. The beads were washed extensively, and the pull-down complex was eluted with SDS-PAGE sample buffer and subjected to immunoblotting (IB) with antibodies against Mrp2 and Na⁺-K⁺-ATPase, which was used as the loading control. The protein bands (left) were analyzed by densitometry scanning (right), and the relative ratio of Mrp2 to Na⁺-K⁺-ATPase was calculated with the ratio in the transfection with Mrp2 and a vector control (transfection 1) set at 1. Note that, compared with cells cotransfected with Mrp2 and a vector control (transfection 1), both total Mrp2 in the whole cell lysates (A) and cell surface Mrp2 (B) were significantly reduced in cells cotransfected with Mrp2 and HA-CAL (transfection 2 and transfection 3). In contrast, when cotransfected with NHERF-1, Mrp2 expression was not significantly changed (transfection 4). Values are means ± SD. Data are representative of 4 independent experiments. *P < 0.05 vs. transfection 1.

DISCUSSION

In the present study, we report a novel interaction between Mrp2 and a Golgi-associated protein, CAL, via its PDZ-binding motif and demonstrate that CAL negatively regulates Mrp2 expression in cotransfected cells. As a multispecific organic anion transporter, Mrp2 plays a predominant role in bile formation and elimination of both endogenous and xenobiotic conjugates. Expression of functional Mrp2 is dependent on its correct synthesis, maturation, sorting, and trafficking from the endoplasmic reticulum (ER) and Golgi to the cell surface, as well as its retention on the plasma membrane. Mrp2 has at least three putative N-glycosylation sites, two of which are located in the NH₂-terminus and the third one in the second extracellular loop of the COOH-terminus (21). Numerous studies have shown that glycosylation plays a critical role in Mrp2 trafficking. In patients with Dubin-Johnson syndrome, an autosomal recessive disorder characterized by chronic conjugated hyperbilirubinemia, mutated MRP2 proteins fail to traffic to the apical membrane but are retained in the ER of hepatocytes. These mutant proteins are less glycosylated and sensitive to endoglycosidase (9, 15, 16). In sandwich-cultured rat hepatocytes, during the re-differentiation process of the cells, only the fully N-glycosylated form of Mrp2 was associated with functional activity; inhibition of glycosylation by tunicamycin decreases the molecular mass and simultaneously impairs the trafficking of Mrp2 to the canalicular membrane (7, 32). Here we show that, when HA-CAL was coexpressed in COS-7 and LLC-PK1 cells, Mrp2 was detected as both mature and immature forms, with the fully glycosylated mature form of Mrp2 reduced to ∼60% of total Mrp2 compared with cells cotransfected with Mrp2 and a vector control (Fig. 5A). As a Golgi-associated protein, CAL has been shown to interact via its two putative coil-coiled domains with Golgi resident proteins such as Golgin-160 (11). The Golgi apparatus is a complex organelle in the secretory pathway that plays a key role in the posttranslational modification and trafficking of proteins and lipids. It seems likely that, under physiological conditions, CAL associates with nascent Mrp2 and tethers it to the trafficking complex in the Golgi stacks to facilitate the maturation and movement of Mrp2 from ER to Golgi and to the plasma membrane. Overexpressed CAL may compete for the binding to immature Mrp2 with proteins that are involved in posttranslational modification, such as glycosyltransferases. This inhibits further maturation of Mrp2 and favors overretention of the CAL-Mrp2 complex in the Golgi, leading to reduced packaging of Mrp2 into secretory transport vesicles destined for the cell surface. This notion is supported by our co-IP assays showing that the immature, lower molecular mass Mrp2 is the major component of Mrp2 in the co-IP complex pulled down by HA-CAL (Fig. 4). It is noteworthy that cotransfection of CAL increased the abundance of the low-molecular-mass band of rBsep as well, although rBsep did not coprecipitate with HA-CAL in cotransfected COS-7 cells (Fig. 4). Interestingly, unlike Mrp2, which was expressed as a single high-molecular-mass band when CAL was not cotransfected, rBsep was expressed as both a high- and a low-molecular-mass band, even when CAL was not cotransfected. Currently, it is not clear how the expression of rBsep is affected by the overexpression of CAL in COS-7 cells. Given the fact that four asparagine residues are N-glycosylated in the first extracellular loop of rBsep, it is plausible that glycosylation plays a more important role in the maturation of rBsep, and overexpression of rBsep or CAL in transfected COS-7 cells prevents proper glycosylation and maturation of the newly synthesized protein, resulting in increased immature form of rBsep in these cells.

Our data obtained with Huh-7 cells indicate that CAL also adversely affects Mrp2 expression in hepatocytes (Fig. 6). To determine whether CAL reduces the expression of Mrp2 in vivo, we attempted to label cell surface Mrp2 in collagen sandwich-cultured rat hepatocytes by biotinylation and streptavidin pull-down assays, but access of biotin to the canalicular Mrp2 protein was blocked by the top collagen gel (data not shown). Nevertheless, CAL was detected in the trans-Golgi network of isolated rat hepatocytes, cells that expresses abundant Mrp2 in the apical membrane (Fig. 1A), which is consistent with its subcellular distribution reported in earlier studies (5, 31).

NHERF-1 is a scaffolding protein that is abundantly expressed in the subplasma membranes of polarized epithelial cells and has been shown to interact with numerous membrane proteins (28). Our laboratory previously reported that Mrp2 binds to NHERF-1 via its COOH-terminal PDZ binding motif, and Mrp2 protein is reduced in the liver of NHERF-1^−/− mice compared with WT mice, although Mrp2 mRNA remains unchanged (18), indicating that NHERF-1 positively regulates Mrp2 expression by posttranscriptional mechanisms. Interestingly, in the present study, we also observed an increase of CAL expression in NHERF-1^−/− mouse liver, accompanied by decreased expression of Mrp2 in the membrane fractions (Fig. 2). These findings suggest that the reduction of Mrp2 expression in the liver of NHERF-1^−/− mice could be caused by the combination of NHERF-1 deficiency and CAL overexpression. Given the fact that both NHERF-1 and CAL binds to the same COOH-terminal PDZ-binding motif of Mrp2, it is plausible that CAL competes with NHERF-1 in binding to Mrp2 in the subplasma membrane compartments, affecting translocation of Mrp2 from the intracellular pool to the plasma membrane. Alternatively, CAL might also affect the endocytic recycling of Mrp2 by promoting the lysosomal degradation of endocytosed Mrp2. CAL has been reported to bind syntaxin 6, a member of the soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) protein family that is involved in protein trafficking between the trans-Golgi network and endosomal system (3, 4). The CAL-syntaxin 6 complex mediates lysosomal trafficking and degradation of CFTR (4). In addition, CAL has been shown to interact with cell surface receptors, such as pathogen receptor CD46, and Beclin 1, a key component of the core autophagic machinery, linking phagosomes to the lysosomes for degradation (13).

In summary, we reported that Mrp2 specifically interacts with PDZ protein CAL, a Golgi-associated protein, via its PDZ-binding motif. Overexpression of CAL resulted in reduced expression of Mrp2. Our findings provide evidence for the importance of PDZ proteins with counterregulatory roles in the posttranscriptional regulation of Mrp2 expression and have implications for new therapeutic strategies for Mrp2-related diseases, such as cholestasis, as well as mechanisms for the posttranscriptional regulation of other membrane proteins.

GRANTS

This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants DK-R37-25636 and DK-P30-34989.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

M.L., C.J.S., and K.H. performed experiments; M.L. and C.J.S. analyzed data; M.L., C.J.S., and J.L.B. interpreted results of experiments; M.L. and C.J.S. prepared figures; M.L., C.J.S., and J.L.B. drafted manuscript; M.L., C.J.S., and J.L.B. edited and revised manuscript; M.L., C.J.S., K.H., and J.L.B. approved final version of manuscript.