INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

DETOXIFICATION OF endo- and xenobiotics is a major function of the liver. Hepatic uptake of many of these amphipathic compounds is mediated by polyspecific organic anion transporting polypeptides (Oatp) that have been cloned from liver, brain, and kidney (20). The first member of the Oatp gene family of membrane transporters (Oatp1) has been isolated from rat liver and shown to mediate Na⁺-independent saturable transport of sulfobromophthalein (BSP) [Michaelis-Menten constant (K_m) of 1.5-3.3 µM] and taurocholate (K_m of 19-50 µM) when expressed in Xenopus laevis oocytes (11, 16) and in transfected HeLa cells (12, 30, 32). Oatp1 represents an ~80-kDa glycoprotein that in addition to the basolateral plasma membrane of hepatocytes is also localized at the apical membranes of kidney proximal tubule (S3 segment) (3) and choroid plexus epithelial cells (2). The endogenous substrates as well as the transport mechanism are not definitively elucidated yet. In stably transfected HeLa cells, Oatp1 has been shown to function as a taurocholate/HCO₃ exchanger (30), whereas in Xenopus laevis oocytes Oatp1-mediated taurocholate transport was transstimulated by reduced glutathione (19). Besides BSP and taurocholate, Oatp1 has been shown to transport also estradiol-17-glucuronide in transfected HeLa cells (13, 30, 32). In addition, with the use of the Xenopus laevis expression system, a wide variety of structurally unrelated compounds have been suggested to be transported by Oatp1, including the steroid conjugates estrone-3-sulfate (K_m of ~4.5 µM) (5), the neutral steroids aldosterone (K_m of ~15 nM), cortisol (K_m of ~13 nM), and ouabain (K_m of 1,700 µM) (5), the thrombin inhibitor CRC-220 (K_m of ~30 µM) (7), and the mycotoxin ochratoxin A (K_m of ~ 17 µM) (14). So far, the kinetic parameters of only three substrates have been determined in a mammalian expression system (13, 30, 32). Furthermore, the kinetics of Oatp1-mediated uptake of bile salts has only been determined for taurocholate (16). This gap of knowledge is closed in this study with the use of stably transfected Chinese hamster ovary (CHO) cells to determine the kinetic parameters of several bile salts and various nonbile acid endo- and xenobiotics in a mammalian expression system. The results demonstrate that similar to the Na⁺-taurocholate cotransporting polypeptide (Ntcp) (31), Oatp1 also exhibits the highest affinity for the dihydroxy bile salt taurochenodeoxycholate among all bile salts tested. Furthermore, the steroid conjugates, including the newly characterized dehydroepiandrosterone sulfate (DHEAS), have been identified as substrates with the highest affinities for Oatp1, supporting the concept that these conjugates represent important endogenous substrates of Oatp1.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Materials [³H]DHEAS (16 Ci/mmol), [³H]taurocholic acid (2.6 Ci/mmol), [³H]cholic acid (13.2 Ci/mmol), [³H]estrone-3-sulfate (49.0 Ci/mmol), [³H]estradiol-17-D-glucuronide (49 Ci/mmol), and [³H]ouabain (20.5 Ci/mmol) were obtained from DuPont NEN (Boston, MA). [³⁵S]BSP (4.1 Ci/mmol) was kindly provided by A. W. Wolkoff of Albert Einstein College of Medicine (Bronx, NY), CRC-220 by W. Stüber of Behringwerke (Marburg, Germany), [³H]ochratoxin A by E. Petzinger of Justus Liebig-Universität (Giessen, Germany), and [2-³H]taurochenodeoxycholate (0.5 Ci/mmol) and [2-³H]tauroursodeoxycholate (0.5 Ci/mmol) by A. W. Hofmann and C. D. Schteingart of University of California at San Diego (La Jolla, CA). All cell culture media and reagents were obtained from Life Technologies (Paisley, UK). All other chemicals and reagents were of analytical grade and were readily available from commercial sources.

Antibody production, immunofluorescence, and Western blotting. The cDNA coding for the last 40 amino acids of Oatp1 (11) was PCR amplified using the following primers: 5'-GACATTGACTCTTCAGCAACTG-3' (corresponding to nucleotides 1977-1998) and 5'-CTGTTCATGGCCTTGAACAGG-3' (corresponding to nucleotides 2135-2115). The blunted PCR product was cloned into the Asp700-cut pMAL-c2. After sequencing was performed to verify the correct in-frame subcloning, the fusion protein between Oatp1 and the maltose binding protein of E. coli was isolated and a rabbit was immunized as described (34). The antibody raised did not cross-react with Oatp2 (21) as verified in separate in vitro expression and in vivo localization experiments (data not shown). Immunofluorescence was performed as described previously (34). SDS-PAGE and Western blotting were performed according to standard procedures (17, 26).

Stable transfection of Chinese hamster ovary cells with Oatp1. The complete coding region of Oatp1 was cut out from the original plasmid (11) using Sal I and Hind III. This region was blunt ended and then subcloned into the Stu I-digested and dephosphorylated pCMV vector-1 (31). This construct was introduced into CHO cells by electroporation, and stably transfected cells were selected by adding G418 to the culture medium. From the resulting transfected cell pool, single clones were isolated with the use of cloning cylinders and tested for Na⁺-independent taurocholate uptake. Clone CHO-03 exhibited the highest transport activity of taurocholate and was selected for use in all further experiments.

Cell culture. CHO cells were grown in DMEM supplemented with 10% FCS, 2 mM L-glutamine, 50 µg/ml L-proline, 100 U/ml penicillin, 100 µg/ml streptomycin, and 0.5 µg/ml Fungizone (amphotericin B) at 37°C with 5% CO₂ and 95% humidity. Selective medium contained additional 400 µg/ml G418 sulfate (Geneticin).

Uptake studies in CHO cells. Determination of Na⁺-independent uptake of potential substrates for Oatp1 was performed as described (31). For some experiments, expression of Oatp1 was induced by incubation of the cells for 24 h with culture medium supplemented with 5 mM butyrate as described (22). For determination of the kinetic parameters, the linear range of uptake was first determined for each substrate individually. Transport was then measured at a time point well within this linear range (usually 20-30 s), and net uptake values used for the calculation of the kinetic parameters were obtained by subtracting the uptake values obtained with wild-type CHO-K1 cells from values obtained with stably transfected CHO-03 cells. Because preliminary experiments did not demonstrate any Na⁺ dependency, all uptake experiments were performed in choline chloride-containing solutions.

Determination of protein concentration. Protein concentrations were determined using the bicinchoninic acid protein assay kit (Pierce, Rockford, IL) (33).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The antiserum generated against the fusion protein of the COOH-terminal 40 amino acids of Oatp1 and the maltose binding protein of E. coli was tested by Western blot analysis with rat liver basolateral and canalicular membrane vesicles. As shown in Fig. 1 (lane 1), the antiserum reacted with an antigen in the basolateral membrane fraction, yielding a single broad band with an apparent molecular mass of 81 ± 6 kDa (mean ± SD of 6 independent determinations). The signal was virtually absent in the canalicular lane (Fig. 1, lane 2), confirming the selective localization of Oatp1 to the basolateral plasma membrane domain of hepatocytes (3). The specificity of the signal was tested by preincubation of the antiserum with the fusion protein (Fig. 1, lanes 3 and 4). The results demonstrate that the raised antiserum is specific for a single 81-kDa basolateral liver plasma membrane protein.

View larger version (62K): [in this window] [in a new window]   Fig. 1.   Characterization of antiserum raised against organic anion transporting polypeptide (Oatp1) in rat liver basolateral and canalicular membrane vesicles. Seventy-five micrograms of rat liver basolateral (lanes 1 and 3) or canalicular (lanes 2 and 4) membrane vesicles were separated using 7.5% SDS-PAGE and subsequently transferred to nitrocellulose. Western blot was incubated with a 1:2,000 dilution of the Oatp1 antiserum (lanes 1 and 2) or with Oatp1 antiserum preabsorbed with the fusion protein used to raise the antiserum (lanes 3 and 4). Bound antibodies were visualized using 125I-labeled protein A. Molecular mass standards are indicated on the right (in kDa).

To localize the native Oatp1 in intact rat liver, the antiserum was used on cryosections for immunofluorescence studies. As demonstrated in Fig. 2 and supporting the Western blot results from Fig. 1, Oatp1 immunoreactivity is restricted to the basolateral plasma membrane of hepatocytes. No immunostaining could be detected at the canalicular domain (Fig. 2). In addition, biliary epithelial cells were immunonegative for Oatp1 (data not shown).

View larger version (97K): [in this window] [in a new window]   Fig. 2.   Immunofluorescent localization of Oatp1 in intact rat liver. The 0.5- to 1.0-µm cryosections were incubated as described (35). Immunoreactivity is restricted to the basolateral membrane, whereas the canalicular domain (arrows) is immunonegative. Bar = 25 µm.

To prove that the immunopositive basolateral hepatocyte antigen indeed represents the native Oatp1, we next investigated whether expression of cell surface immunopositivity is associated with Na⁺-independent transport of BSP. As demonstrated in Fig. 3, CHO cells stably transfected with Oatp1 cDNA (CHO-03) exhibited immunopositive surface staining as well as Na⁺-independent BSP uptake. In contrast, wild-type CHO-K1 cells were immunonegative and showed only minimal BSP uptake. These results demonstrate that the immunopositive protein represents the functionally active rat liver Oatp1.

View larger version (76K): [in this window] [in a new window]   Fig. 3.   Correlation of Oatp1 expression and Na+-independent sulfobromophthalein (BSP) uptake in stably transfected Chinese hamster ovary (CHO) cells. Control CHO cells (CHO-K1, A) or Oatp1-expressing CHO cells (CHO-03, B) were grown to confluency on coverslips and treated as described in MATERIALS AND METHODS. On separate dishes, 2 µM [35S]BSP uptake was measured, after a 24-h incubation in 5 mM sodium butyrate, for 5 min in a choline chloride medium. Uptake values, bottom, represent means ± SD of triplicate determinations.

In a previous study, CHO cells stably expressing Ntcp showed a ~10-fold stimulation of Na⁺-dependent uptake of bile salts when gene expression was induced by sodium butyrate (31). As illustrated in Fig. 4, sodium butyrate exerted a similar inducing effect on the expression of Oatp1 in stably transfected CHO-03 but not in wild-type CHO-K1 cells. Thus sodium butyrate increased the maximal velocity (V_max) value for Oatp1-mediated estrone-3-sulfate uptake ~10-fold, whereas the K_m value remained unchanged (Table 1). Because high level expression is important for the correct delineation of the substrate specificity of a given transport system, we performed all subsequent transport studies under butyrate-induced conditions. On the basis of previous initial uptake activities (20), we next determined the kinetics of a variety of established and new Oatp1 substrates. As demonstrated in Table 1, BSP and estradiol-17-glucuronide exerted the highest affinities for Oatp1 (K_m of ~3 µM) among all substrates tested. These results are similar to previous studies in Xenopus laevis oocytes (11, 16) and in transfected HeLa cells (13), respectively. Second were taurocheno- and tauroursodeoxycholate, DHEAS, and estrone-3-sulfate (K_m values of 5-13 µM) (Table 1). The identification of DHEAS as a new substrate of Oatp1 (Fig. 5) supports the concept that endogenous steroid conjugates are important physiological high-affinity substrates of Oatp1 (13). A third group of substrates with K_m values between 29 and 57 µM included ochratoxin A, taurocholate, cholate, glycocholate, and CRC-220 (Table 1). Finally, the studies in stably transfected CHO-03 cells confirmed the low affinity of Oatp1 for ouabain (K_m of ~3,000 µM) (5). These studies in stably transfected CHO-03 cells confirm the polyspecific substrate spectrum of Oatp1 and identify dihydroxy bile salts and DHEAS as new high-affinity Oatp1 substrates.

View larger version (10K): [in this window] [in a new window]   Fig. 4.   Butyrate induction of Oatp1-mediated estrone-3-sulfate uptake into CHO cells. Wild-type CHO-K1 cells (A) and Oatp1-expressing CHO-03 cells (B) were incubated for 24 h in the presence () or absence () of 5 mM sodium butyrate. For uptake measurements, cells were incubated with 15 µM [3H]estrone-3-sulfate at 37°C for the indicated time periods in a choline chloride medium. Data points represent means ± SD of triplicate determinations.

View larger version (10K): [in this window] [in a new window]   Fig. 5.   Kinetics of Oatp1-mediated dehydroepiandrosterone sulfate (DHEAS) uptake in stably transfected CHO-03 cells. Wild-type CHO-K1 cells () or Oatp1-expressing CHO-03 cells () were incubated with increasing concentrations of [3H]DHEAS at 37°C for 20 s in a choline chloride medium. Kinetic parameters were calculated using a nonlinear curve fitting computer program according to the Michaelis-Menten equation with the net Oatp1-mediated uptake values () (values obtained with wild-type CHO-K1 cells subtracted from values obtained with stably transfected CHO-03 cells). Means ± SD of triplicate determinations are given. Curve represents the fitted model. Km, Michaelis-Menten constant.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Previous studies in Oatp1-expressing Xenopus laevis oocytes and in transfected HeLa cells have provided evidence that Oatp1 mediates transport of a wide variety of amphipathic compounds (5, 11-13, 16, 20, 32). Among several anionic, neutral, and even cationic substrates identified, only the anions BSP, taurocholate, and estradiol-17-glucuronide have been characterized in a mammalian expression system (12, 13, 32). In this study, we directly correlate surface expression of Oatp1 and multispecific transport of several anionic as well as neutral compounds in a stably transfected mammalian CHO cell line (Figs. 3-5).

The antibody generated against the COOH-terminal end of Oatp1 recognized an ~81-kDa antigen at the basolateral membrane of hepatocytes (Figs. 1 and 2), which is similar to previous studies using a different antibody raised against a 13-amino acid peptide close to the COOH-terminal end (3). The antigen used to generate our new antibody encompasses the last 40 amino acids of Oatp1. The specificity for Oatp1 is indicated by an amino acid identity of the epitope of 70% to Oatp3 (1), 61% to OAT-K1 (29), and only 56% to Oatp2 (21). Furthermore, the specificity of the antibody for Oatp1 is supported by the observation of 1) basolateral immunopositivity in hepatocytes (Fig. 2), which do not express Oatp3 (1), and 2) a distinct distribution of Oatp2 in liver (24) and choroid plexus (8). The positive correlation between immunostaining of the plasma membrane and transport of BSP in the CHO-03 cells (Fig. 3) definitively shows that the basolateral protein recognized by the antiserum indeed represents functionally active Oatp1.

Because high level expression of a transport protein is required for correct delineation of its substrate specificity and kinetic transport parameters (31), we treated the stably transfected CHO-03 cells with 5 mM sodium butyrate. Similar to Ntcp-expressing CHO 9-6 cells (31), butyrate induction also resulted in a 10-fold increase of Oatp1-mediated estrone-3-sulfate uptake (Fig. 4; Table 1). This butyrate-induced expression level of Oatp1 in CHO-03 cells is about fivefold higher than in hepatocytes, as estimated on the basis of the apparent V_max values for Na⁺-independent cholate uptake (Table 1; Ref. 4). A similar increase in Oatp1 expression is also evident in comparison to stably transfected HeLa cells (30, 32). Hence, butyrate-induced CHO-03 cells were routinely used to determine the kinetics of a variety of presumptive Oatp1 substrates not previously tested in a mammalian cell system. On the basis of the ratio of V_max to K_m, the best transport substrate of Oatp1 was BSP followed by the dihydroxylated bile salts taurourso- and taurochenodeoxycholate, the steroid conjugates estradiol-17-glucuronide, estrone-3-sulfate, and DHEAS, the mycotoxin ochratoxin A, the trihydroxylated bile salts cholate, glycocholate, and taurocholate, and the thrombin inhibitor CRC-220 (Table 1). Similar to previous studies in Xenopus laevis oocytes, the cardiac glycoside ouabain exhibited by far the lowest affinity for Oatp1 among all substrates tested (Table 1). Our results indicate that dihydroxy bile salts and steroid 3 and 17 conjugates represent important endogenous substrates of Oatp1. This conclusion is further supported by the identification of DHEAS as a new high-affinity endogenous substrate of Oatp1 (Fig. 5; Table 1). DHEAS is also transported by the human OATP with an apparent K_m value of ~7 µM (15). Because the concentration of DHEAS in human blood plasma reaches 10 µmol/l and because OATP is widely distributed in the human brain, it is possible that OATP plays an important role in the intracerebral distribution and action of DHEAS in humans (15). In the rat, the physiological role of Oatp1 in the disposition of DHEAS and other steroid conjugates is most probably concentrated in the liver and kidney (3, 25), since cerebral expression of Oatp1 is confined to the apical portion of choroid plexus epithelial cells only (2).

Besides endogenous substrates, Oatp1 also mediates transport of the mycotoxin ochratoxin A (Table 1, Ref. 14), which is a frequent contaminant of food and animal chow (14, 27). Its principal mechanism of action is inhibition of protein synthesis by competition with phenylalanine. The main target organ for ochratoxin A toxicity is the kidney, whereas the liver is less frequently affected (27), most likely because in the liver the toxin is rapidly glucuronidated, sulfated, and excreted into bile (28). In the kidney, ochratoxin A is reabsorbed in the proximal straight tubule, resulting in toxic intracellular concentrations of the mycotoxin in kidney epithelial cells (6). Interestingly, renal reabsorption of ochratoxin A can be partially inhibited by BSP (6), which represents a classical high-affinity substrate of Oatp1 (11) (Table 1). Moreover, a significant portion of peritubular uptake of ochratoxin A is probenecid sensitive and p-aminohippurate insensitive (9). These characteristics of renal ochratoxin A transport could be explained by Oatp1-mediated transport (16), since Oatp1 is also localized at the brush-border membrane of the late proximal straight tubule (S3) (3). Thus, in addition to uptake into hepatocytes, Oatp1 might play a role in the renal reabsorption of ochratoxin A and thus significantly contribute to the overall nephrotoxicity of this mycotoxin. Whether Oatp1 and/or other members of the Oatp gene family of membrane transporters are also involved in hepatic and/or renal transport of other mycotoxins such as microcystin is not yet known and remains to be investigated.

In conclusion, the present study proves the multispecific nature of Oatp1-mediated amphipathic substrate transport in a mammalian cell system. In addition to the substrates identified previously and in this study, recent evidence indicates that Oatp1 can also transport glutathione conjugates, leukotriene C₄, and certain dipeptidic drugs (10, 19, 23). In one of these studies, leukotriene C₄ exhibited a very high affinity for Oatp1 (K_m of ~0.27 µM), indicating that Oatp1 mediates leukotriene C₄ uptake into hepatocytes under physiological conditions (18, 19). Furthermore, reduced glutathione efflux has been proposed as a new driving force for Oatp1-mediated substrate uptake into hepatocytes (19). Hence, the importance of Oatp1 as a polyspecific organic anion and drug transporter is increasingly recognized. Furthermore, additional members of the Oatp gene family of membrane transporters have been cloned and shown to exhibit partially overlapping substrate specificities with Oatp1 (1, 20). Thus exact delineation of the transport characteristics of each Oatp in high expression mammalian cell systems is important in defining the physiological and pathophysiological roles of individual Oatps in the normal and diseased body.