CCK-8 (l-aspartyl-l-tyrosyl-l-methionylglycyl-l-tryptophyl-l-methionyl-l-aspartyl-l-phenylalaninamide hydrogen sulfate ester), a derivative of the gastrointestinal peptide hormone cholecystokinin, is specifically taken up into human hepatocytes by the organic anion transporter OATP1B3 (OATP8). So far it was unknown which transporter mediates the excretion of CCK-8 into bile. Double-transfected Madin-Darby canine kidney strain II cells, expressing recombinant human OATP1B3 in the basolateral membrane together with human ABCC2 (multidrug resistance protein 2, MRP2) in the apical membrane, represent a valuable model system to study vectorial transport. The importance of an appropriate filter support for optimized protein localization and substrate transport was demonstrated by the comparison of filter pore densities of 2 × 106 and 1 × 108 per cm2. At the high pore density, immunofluorescence microscopy showed an intense OATP1B3 signal in the basolateral membrane of all cells, and 82 ± 8% of cells expressed ABCC2 in the apical membrane. Uptake and efflux of radiolabeled CCK-8 in the double-transfected cells grown at high pore density was enhanced 3.5- and 5.6-fold, respectively, compared with cells grown at lower pore density. Higher transport rates were also observed with [3H]bromosulfophthalein. The high-affinity ATP-dependent transport of CCK-8 by ABCC2 was directly demonstrated in ABCC2-containing membrane vesicles with a Km value of 8.1 μM. The uptake by OATP1B3 and hence the vectorial transport of CCK-8 was inhibited by cyclosporin A (Ki 1.2 μM) and by MK571 [(3-(3-(2-(7-chloro-2-quinolinyl)ethenyl)phenyl) ((3-dimethylamino-3-oxopropyl)thio)methyl)thiopropanoic acid] (Ki 0.6 μM); the respective Ki values for the ABCC2-mediated transport were 24 and 8.5 μM. Thus, using an optimized filter support, we demonstrate vectorial transport of CCK-8 by OATP1B3 and by the apical export pump ABCC2.
The vectorial transport of substances from blood via hepatocytes into bile is mediated by different transporters in the sinusoidal and canalicular membrane and is important for the elimination of many endogenous and xenobiotic substances. One large family of solute carriers comprises the organic anion transporters of the OATP family (Hagenbuch and Meier, 2004). OATP1B3, encoded by SLCO1B3 (formerly termed OATP8; former gene symbol SLC21A8), is located in the basolateral membrane of human hepatocytes (König et al., 2000). In the canalicular (apical) membrane, ABCC2 (multidrug resistance protein 2, MRP2), a member of the ABCC subgroup of the ABC superfamily, is responsible for ATP-dependent transport of anionic conjugates (König et al., 1999).
Cholecystokinin, a peptide hormone released postprandially from the intestine, has a broad range of biologic activities. It stimulates contraction of the gallbladder, release of pancreatic enzymes, and intestinal motility (Mutt, 1980). The C-terminally sulfated octapeptide cholecystokinin-8 (CCK-8, sincalide; l-aspartyl-l-tyrosyl-l-methionylglycyl-l-tryptophyl-l-methionyl-l-aspartyl-l-phenylalaninamide hydrogen sulfate ester), a small derivative of cholecystokinin, was recently shown to be a selective substrate of OATP1B3 (Ismair et al., 2001). OATP1B1 has 80% amino acid identity with OATP1B3 (König et al., 2000) but is not able to mediate CCK-8 transport (Ismair et al., 2001). CCK-8 is efficiently excreted into rat bile (Gores et al., 1986a,b). However, the molecular basis of this transport of CCK-8 across the canalicular membrane remained unknown.
The vectorial transport of endogenous and xenobiotic substances has been studied recently by use of double-transfected Madin-Darby canine kidney strain II (MDCKII) cells expressing a basolateral uptake transporter and an apical efflux pump (Cui et al., 2001; Sasaki et al., 2002, 2004). In this study, we used MDCKII cells expressing the human hepatic uptake transporter OATP1B3 and the ATP-dependent export pump ABCC2 to examine the vectorial transport of CCK-8. With these double-transfected cells and inside-out membrane vesicles from ABCC2-expressing MDCKII cells, we demonstrate that CCK-8 is a high-affinity substrate for ABCC2. Cyclosporin A and the quinoline derivative MK571 [(3–3-(7-chloro-2-quinonlinyl)ethenyl)phenyl)(3-dimethylamino-3-oxopropyl)thio)methyl)thio)propanoic acid] were identified as inhibitors of both OATP1B3-mediated uptake and ABCC2-mediated efflux of CCK-8. An additional aim of this work was to improve the system of the double-transfected cells as a tool to study vectorial transport of endogenous and xenobiotic substances. We demonstrate the enhancement of transport rates by growth of the cells at much higher pore densities of the cell culture inserts than used previously.
Materials and Methods
Antibodies and Chemicals. The polyclonal antibodies SKT and EAG5 were raised in rabbits against human OATP1B3 (König et al., 2000) and ABCC2 (Büchler et al., 1996), respectively. The mouse monoclonal antibody M2III-6 against ABCC2 was purchased from Alexis Biochemicals (San Diego, CA). Alexa Fluor 488-conjugated goat anti-mouse antibody was from Molecular Probes (Eugene, OR) and Cy3-conjugated goat anti-rabbit antibody was from Jackson ImmunoResearch Laboratories Inc. (West Grove, PA). Other chemicals were commercially available and were obtained at the highest degree of purity. MK571 and cyclosporin A were obtained from Alexis Biochemicals and Sigma-Aldrich Chemie GmbH (Taufkirchen, Germany), respectively. [3H]Cholecystokinin-8 sulfate ([3H]CCK-8) (2.5–3.9 TBq/mmol) was purchased from Amersham Biosciences UK, Ltd. (Little Chalfont, Buckinghamshire, UK). [3H]Bromosulfophthalein ([3H]BSP) (0.6 TBq/mmol) was obtained from Hartmann Analytic (Braunschweig, Germany).
Cell Culture and Transfection. MDCKII cells were cultured in minimum essential medium (Sigma-Aldrich Chemie GmbH) containing 10% fetal calf serum (Biowest, Nuaillé, France), 100 U/ml penicillin, and 100 μg/ml streptomycin at 37°C and 5% CO2. MDCKII cells were transfected with the respective plasmid (pcDNA3.1(+)-ABCC2 and pcDNA3.1/Hygro(–)-OATP1B3) (Cui et al., 1999; Letschert et al., 2004). After geneticin and hygromycin selection, single colonies were screened for OATP1B3 and ABCC2 protein expression by immunofluorescence microscopy and immunoblot analysis. Due to the fact that each cell line is unique, we normalized the cell lines in this study on the basis of their OATP1B3 expression level determined by immunoblotting. Since the most frequent single nucleotide polymorphism resulting in the amino acid exchange S112A exhibits no functional differences compared with the so-called reference sequence (NM_019844) (Letschert et al., 2004), we used single OATP1B3-expressing cells with this polymorphism in the present study to obtain OATP1B3 levels similar to those of the double transfectants. MDCKII cells were grown on cell culture inserts to confluence for 3 days and induced with 10 mM sodium butyrate for 24 h prior to analysis to obtain higher levels of the recombinant proteins (Cui et al., 1999). The polyethylene terephthalate cell culture inserts (ThinCert, 24 mm diameter, pore size 0.4 μm) and the tissue culture multiwell plates (Cellstar) were obtained from Greiner Bio-One GmbH (Frickenhausen, Germany). Low pore density inserts of 2 × 106 pores per cm2 and inserts with a high pore density of 1 × 108 pores per cm2 were used. MDCKII cells transfected with the empty pcDNA3.1 vector served as a negative control in all experiments.
Immunoblot Analysis. Membrane proteins, obtained after cell lysis in hypotonic buffer, were pelletized by centrifugation at 20,800g for 60 min, diluted with sample buffer, and incubated at 37°C for 30 min prior to their separation on 4% stacking and 7.5% and 10% resolving SDS polyacrylamide gels for ABCC2 and OATP1B3, respectively. Immunoblotting was performed using a tank blotting system from Bio-Rad Laboratories GmbH (Munich, Germany) and enhanced chemiluminescence detection (PerkinElmer Life and Analytical Sciences, Boston, MA). The primary polyclonal antibodies SKT (König et al., 2000) and EAG5 (Schaub et al., 1999) were diluted 1:3000 in Tris-buffered saline/Tween 20 (10 mM Tris-HCl, pH 8.0, 150 mM NaCl, 0.05% Tween 20). The secondary antibody was a horseradish peroxidase-conjugated goat anti-rabbit IgG (Bio-Rad) at a 1:2000 dilution.
Immunofluorescence Microscopy. MDCKII cells were grown and butyrate-induced as described above. The cells were fixed for 30 min with 2% paraformaldehyde in phosphate-buffered saline (PBS) and permeabilized for 30 min in 1% Triton X-100 in PBS. The cells were incubated with the primary antibodies for 1.5 h at room temperature. After washing three times with PBS, cells were incubated with the secondary antibodies for 1.5 h at room temperature. All antibodies were diluted in PBS as follows: SKT and M2III-6 1:50 and Alexa Fluor 488-conjugated goat anti-mouse IgG and Cy3-conjugated goat anti-rabbit IgG 1:200. Filter membranes were removed from their plastic support and mounted onto glass slides with 50% glycerol in PBS. Immunofluorescence pictures were taken with a confocal laser-scanning microscope (Carl Zeiss GmbH, Jena, Germany). For quantification of ABCC2-expressing cells, five different pictures, each containing 50 to 120 cells, were counted, and the percentage of ABCC2-positive cells relative to the total number of cells was determined.
Transport Assays. MDCKII cells were grown on cell culture inserts and induced with butyrate as described above. For transport measurements, the cells were washed with uptake buffer (142 mM NaCl, 5 mM KCl, 1 mM K2HPO4, 1.2 mM MgSO4, 1.5 mM CaCl2, 5 mM glucose, and 12.5 mM HEPES, pH 7.3). Subsequently, 1 ml of uptake buffer was added to the apical compartment, and 1.5 ml of uptake buffer containing the 3H-labeled substrate was added to the basolateral compartment. After 15 min, the buffer from the apical compartment was collected. The cells were washed three times with cold uptake buffer and solubilized with 2 ml of 0.2% SDS in water. The radioactivity of the buffer from the apical compartment and of the lysate was determined by liquid scintillation counting, and the appropriate protein concentration was determined by bicinchoninic acid assay. For kinetic analyses of OATP1B3-mediated uptake, the single-transfected cells were grown on multiwell plates (Cellstar, Greiner Bio-One GmbH) instead of the cell culture inserts (ThinCert, Greiner Bio-One GmbH) because MDCKII cells grow as nonpolarized cells when cultured on multiwell plates. Transport of CCK-8 into membrane vesicles was measured by the rapid filtration method as described (Keppler et al., 1998).
Expression and Localization of Human OATP1B3 and ABCC2 in Stably Transfected MDCKII Cells. The protein expression of human OATP1B3 and ABCC2 in the stably transfected MDCKII cells was verified by immunoblot analyses (Fig. 1). OATP1B3 was detected in MDCKII-OATP1B3 and in MDCKII-OATP1B3-ABCC2 cells (Fig. 1A). ABCC2 was detected in MDCKII-ABCC2 and in OATP1B3-ABCC2 cells (Fig. 1B). The vector-transfected control cells did not show any OATP1B3 or ABCC2 protein.
The differences in protein expression and sorting of cells cultured in different cell culture inserts were analyzed by immunofluorescence and confocal laser scanning microscopy of double-transfected MDCKII cells (Fig. 2). Cells grown on low-density pore cell culture inserts showed a positive lateral staining for OATP1B3 in all cells, but ABCC2, located in the apical membrane domain, was only weakly detectable in some cells (vertical section; Fig. 2A) and middle view of this area (Fig. 2C). When the identical cell clone was grown on high-density pore cell culture inserts, a stronger signal of OATP1B3 and ABCC2 (vertical section; Fig. 2B) and middle view of this area (Fig. 2D) was detected. Furthermore, some basal staining of OATP1B3 was detectable, and the cells appeared higher when observed in the vertical section (Fig. B). The percentage of cells with detectable ABCC2 properly localized to the apical membrane varied between immunofluorescence analyses and was approximately 3 times higher in cells cultured on the high-density pore cell culture inserts (82 ± 8%).
Influence of Different Pore Density Cell Culture Inserts on Transport of CCK-8 and Bromosulfophthalein. The transcellular transport of CCK-8 and of bromosulfophthalein, an established substrate for OATP1B3 and ABCC2, was estimated on the different cell culture inserts (Fig. 3). After incubation of the cells with 4 nM [3H]CCK-8 for 15 min, the radioactivity inside the cells and in the apical compartment was measured. In general, based on the relative ratio of intracellular accumulation, i.e., uptake and transcellular transport, the low-density pore cell culture inserts provide a system for functional transport studies. However, the cultivation of cells on high-density pore cell culture inserts showed higher absolute values and higher relative ratios between control cells, OATP1B3, and OATP1B3-ABCC2 cells. The total transport of CCK-8 was significantly enhanced using the high-density pore cell culture inserts (Fig. 3, A and B). A summary of the data is shown in Table 1. The relative uptake ratio of the single-transfected MDCKII cells (OATP1B3) compared with control cells was enhanced 7.1 times (40.3 versus 5.7) with CCK-8 as substrate. The uptake and transcellular transport of the double-transfected MDCKII cells (OATP1B3-ABCC2) were enhanced by the factor 3.5 (11.5 versus 3.3) and 5.6 (15.2 versus 2.7), respectively. The intracellular accumulation of CCK-8 using the double-transfected cells was lower than in the single-transfected cells, suggesting that CCK-8 is a substrate for ABCC2-mediated efflux. High-performance liquid chromatography analyses indicated that the excreted CCK-8 was not metabolized by the cells (data not shown).
The influence of different pore densities on the absolute and relative transport rates was also tested with the standard substrate bromosulfophthalein (Fig. 3, C and D). Again, uptake and transcellular uptake rates were higher with MDCKII cells grown on high-density pore cell culture inserts than grown on low-density pore cell culture inserts (Table 1).
The leakage of the cell layers growing on low- and high-density pore cell inserts was measured using 1 μM[3H]inulin and indicated an average leakage of 1 and 2%, respectively (data not shown). Comparison of the corresponding control cells and stably transfected cells indicated no significant differences.
Transport of CCK-8 into the Apical and the Basolateral Compartment. The specific uptake and transcellular transport of CCK-8 was verified by transport assays performed in both directions. 3H-Labeled CCK-8 was applied at the same concentration as in the standard transport assays (B → A) into the apical chamber, and the radioactivity in the basal chamber was determined (A → B, Fig. 4). Control cells, ABCC2-, and OATP1B3-ABCC2-expressing cells showed similar values as the vector-transfected control cells in the transport of CCK-8 from the apical to the basolateral chamber.
Time Dependence of CCK-8 Transport in Stably Transfected MDCKII Cells.Figure 5 shows the time course of the uptake and transcellular transport of CCK-8 in the transfected MDCKII cells. After 5 min, the total transport by the single- and the double-transfected cells was in the same range. The single-transfected cells (OATP1B3) showed a higher intracellular amount of CCK-8 at all time points. The lower intracellular accumulation in double-transfected cells (OATP1B3-ABCC2) was caused by the rapid and efficient export of CCK-8 by ABCC2.
In our system of OATP1B3-expressing MDCKII cells grown on multiwell plates, the Km value for the OATP1B3-mediated CCK-8 transport was 11.3 ± 0.7 μM (S.D., n = 3). The Km value determined on high pore density filter supports, also measured for 1 min to obtain initial transport rates, was 21.4 ± 4.8 μM. Studies using oocytes had indicated a Km value of 11.1 ± 2.9 μM (Ismair et al., 2001).
Transport of CCK-8 into Inside-Out Membrane Vesicles. The transport of CCK-8 by ABCC2 was confirmed by transport assays using membrane vesicles of ABCC2-expressing MDCKII cells (Fig. 6, A and B). Comparison of vesicles from the control cells (circles) with those from ABCC2-expressing cells (triangles) indicated no significant differences in the absence of ATP. However, ABCC2-mediated CCK-8 transport in the presence of ATP was enhanced 7-fold. ATP-dependent ABCC2-mediated transport at different CCK-8 concentrations indicated a Km value of 8.1 ± 0.5 μM.
Inhibition of CCK-8 Transport by Cyclosporin A and MK571. Cyclosporin A and MK571, previously established as competitive inhibitors of ABCC2-mediated transport (Kamisako et al., 1999; Leier et al., 2000), were tested for their inhibitory action on the vectorial transport of CCK-8. The uptake of 4 nM [3H]CCK-8 was partially inhibited by 10 μM cyclosporin A in single-transfected (OATP1B3) and double-transfected (OATP1B3-ABCC2) MDCKII cells, whereas the transcellular transport was reduced to control values (Fig. 7).
Single-transfected cells (OATP1B3) showed a concentration-dependent inhibition of the CCK-8 uptake by MK571 (Fig. 8). The uptake of CCK-8 by the OATP1B3-expressing cells was inhibited by 1 μM MK571 to approximately 60% of the value in the absence of MK571 (Fig. 8A). Double-transfected cells (OATP1B3-ABCC2) confirm the inhibitory effects on the transcellular transport (Fig. 8B). Control cells and ABCC2-expressing cells did not change their transport rate, reflecting that these values are due to background (data not shown). MK571 was also able to inhibit the transport of the prototypic substrate bromosulfophthalein by OATP1B3 (data not shown).
Further inhibition studies were performed using OATP1B3-expressing cells and inside-out membrane vesicles of ABCC2-expressing cells to examine the kinetic properties of both inhibitors (Table 2). The IC50 values for cyclosporin A were 1.8 μM for the CCK-8 transport (5 μM CCK-8) by OATP1B3 (whole cells) and 45.3 μM for the ATP-dependent transport of CCK-8 (7 μM) by ABCC2 (membrane vesicles). The respective Ki value for cyclosporin A for OATP1B3-mediated CCK-8 transport was 1.2 and 24.0 μM for the ABCC2-mediated CCK-8 transport (Table 2). Corresponding experiments using MK571 indicated Ki values for OATP1B3-mediated CCK-8 transport of 0.6 μM and of 8.5 μM for the ABCC2-mediated CCK-8 transport (Table 2).
The carrier-mediated hepatic uptake of the gastrointestinal peptide hormone derivative CCK-8 has been demonstrated in studies on the substrate specificity of OATP1B3 expressed in Xenopus laevis oocytes (Ismair et al., 2001). Earlier studies with radiolabeled cholecystokinin peptides in the isolated perfused rat liver and in primary cultured rat hepatocytes indicated that CCK-8 undergoes biliary excretion (Gores et al., 1986a,b, 1989). Our system of double-transfected MDCKII expressing human OATP1B3 and human ABCC2 has been a useful tool to study the substrate specificity of ABCC2 and the transcellular transport of specific substrates. In our earlier transport studies, we used cell culture inserts with a pore size of 0.4 μm and with a pore density of 4 × 106 per cm2 (Cui et al., 2001; Letschert et al., 2004). The growth conditions and the respective cell shape seemed to be important for the phenotype of ABCC2-expressing cells. The comparative studies in this work showed that the identical cell clone exhibited different expression patterns, depending whether the cells were grown on high- or low-density pore cell culture inserts. The high-density pore cell culture inserts were also useful for immunofluorescence studies. The cells grown on high-density pore inserts showed to some extent basal OATP1B3 localization, in addition to its lateral localization, and a much higher percentage of ABCC2-positive cells (Fig. 2). An inhomogenous expression pattern of ABCC2 protein in different cell batches has also been observed in cells expressing endogenous ABCC2 (Prime-Chapman et al., 2004). When cultured on high-density pore cell culture inserts, culture medium is provided through approximately 200 pores (0.4-μm diameter each) per cell compared with approximately four pores per cell in case of the low-density pore cell culture inserts. The importance of the nutrient supply through fenestrae (0.16-μm diameter) in endothelial cells of the liver as a regulator of the metabolic function of hepatocytes has been discussed earlier (Gatmaitan et al., 1996).
Comparison of the data with single transfectants (OATP1B3) with those obtained with double transfectants (OATP1B3-ABCC2) indicated that the total uptake by the double transfectants is much higher (Fig. 5). This is in line with the fact that ABCC2 has a high affinity to CCK-8, thus efficiently effluxing the intracellular CCK-8. With this substrate, the efflux system seems to be rate-determining for excretion. The efficient transcellular transport of CCK-8 was also observed by Gores and coworkers in the isolated perfused rat liver (Gores et al., 1986a,b, 1989). Vesicle studies using membranes from MDCKII cells expressing ABCC2 demonstrated the efficient ABCC2-mediated transport of CCK-8 with a Km value of 8.1 ± 0.5 μM.
Our present work indicates that the uptake of CCK-8 by OATP1B3 is inhibited by both cyclosporin A and MK571 (Figs. 7 and 8, respectively). Cyclosporin A has been known as an inhibitor of OATP1B3-mediated transport (Letschert et al., 2004) and ABCC2-mediated transport (Kamisako et al., 1999). The quinoline derivative MK571 has been widely used as a specific inhibitor for ATP-dependent transporters, including ABCC1 (Leier et al., 1994) and ABCC2 (Leier et al., 2000). MK571 has now been identified, additionally, as an inhibitor of OATP1B3 in this work. The Ki value of cyclosporin A for the transport of CCK-8 by OATP1B3 was 1.2 μM, and MK571 inhibited CCK-8 transport by OATP1B3 with a Ki value of 0.6 μM (Table 2).
The Ki value of cyclosporin A for the transport of CCK-8 by ABCC2 was 24 μM. A similar Ki value (21 μM) has been described for the inhibition of transport of monoglucuronosyl bilirubin by ABCC2 (Kamisako et al., 1999). MK571 inhibited CCK-8 transport by ABCC2 with a Ki value of 8.5 μM. Earlier studies on the inhibitory effect of cyclosporin A and MK571 on leukotriene C4 transport, using vesicles from ABCC2-expressing cells, showed Ki values of 4.7 and 13.1 μM, respectively (Chen et al., 1999).
The OATP1B3-mediated CCK-8 transport was more sensitive to cyclosporin A than the ABCC2-mediated transport (Table 2). Thus, the more sensitive site of action of cyclosporin A in the transcellular transport was the inhibition of the uptake. MK571 also showed a stronger inhibition of OATP1B3 than of ABCC2-mediated CCK-8 transport, with Ki values of 0.6 and 8.5 μM, respectively (Table 2).
Because of the easy handling of double-transfected cells growing on cell culture inserts compared with other methods for measurement of ABCC2-mediated transport (e.g., insideout membrane vesicles), this system may serve in medium- to high-throughput screening efforts for the identification of possible substrates and inhibitors of both transporters. The improved assay conditions as well as the identification of CCK-8 as an additional excellent substrate for these double-transfected cells contribute to the usefulness of this in vitro system for measurements of vectorial transport.
We thank Dr. H. Spring for expert help in confocal laser scanning microscopy, Dr. J. König for many stimulating discussions, and D. Keller for excellent technical support. We are grateful to Dr. T. Takeuchi, Kagoshima University, and to the Mochida Memorial Foundation for Medical and Pharmaceutical Research, Tokyo, Japan, for encouragement and support of Dr. M. Komatsu's studies in Heidelberg, Germany.
This work was supported in part by grants from the Deutsche Forschungsgemeinschaft (Ko 2120/1-1) and the Bundesministerium für Bildung und Forschung through the program on Systems Biology (31P3111).
ABBREVIATIONS: OATP, organic anion-transporting polypeptide; MRP2, multidrug resistance protein 2 (ABCC2); CCK-8, sulfated cholecystokinin octapeptide (l-aspartyl-l-tyrosyl-l-methionylglycyl-l-tryptophyl-l-methionyl-l-aspartyl-l-phenylalaninamide hydrogen sulfate ester); MDCKII, Madin-Darby canine kidney strain II; BSP, bromosulfophthalein; MK571, (3-(3-(2-(7-chloro-2-quinolinyl)ethenyl)phenyl)((3-dimethylamino-3-oxopropyl)thio)methyl)thiopropanoic acid; PBS, phosphate-buffered saline; CsA, cyclosporin A.
- The American Society for Pharmacology and Experimental Therapeutics