Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

A novel methotrexate (MTX¹) derivative, MX-68 (N-[4-[(2,4-diamminopteridine-6-yl)methyl]-3,4-dihydro-2H-1,4-benzothiazin-7-yl]carbonyl-L-homoglutamic acid), was synthesized to avoid any polyglutamation with the aim of minimizing undesirable intracellular accumulation. MX-68 exhibits more potent antirheumatic activity than does MTX (Matsuoka et al., 1997; Mihara et al., 1996, 1997) and is subject to minimal metabolism or polyglutamation, unlike MTX. MX-68 is mainly excreted into bile and urine in rats. Our previous findings suggest that a saturable transport process plays a role in several membrane penetration processes involving this compound, including hepatic uptake on the sinusoidal membrane and biliary excretion on the bile canalicular membrane, renal secretion and reabsorption on the brush border membrane, and intestinal secretion in the gastrointestinal tract (Han et al., 1999). Therefore, to help understand its pharmacokinetics, it is important to characterize the transport system operating on MX-68 in each process. In the present study we focused on the mechanism of hepatobiliary transport of MX-68.

MX-68 is an organic anion with two carboxylic acid groups that are dissociated at physiological pH. For the hepatic uptake of organic anions, gene expression of several transport systems has been identified: these include an Na⁺-coupled secondary active transport system for taurocholic acid (TCA) (Ananthanarayanan et al., 1994; Hagenbuch and Meier, 1994), the organic anion transporting polypeptide family for the Na⁺-independent uptake of TCA and various nonbile acid organic anions such as bromosulfophthalein and estradiol 17-D-glucuronide (E2-17G) (Jacquemin et al., 1991; Kullak-Ublick et al., 1995; Noe et al., 1997; Abe et al., 1998), and the organic anion transporter family (Sekine et al., 1998). As far as the hepatic uptake of folate analogs is concerned, transport in isolated rat hepatocytes of 5CH₃-tetrahydrofolate (5CH₃-H4PteGlu), which is the major folate form in biological fluids, was reported to be independent of Na⁺ in the medium (Horne, 1990). This finding has been confirmed in transport studies using basolateral membrane vesicles (Horne et al., 1992). On the other hand, such Na⁺ dependence is controversial as far as the uptake of antifolate MTX is concerned; Na⁺ dependence has been reported in isolated hepatocytes (Horne et al., 1976; Gewirtz et al., 1980), but lack of Na⁺ dependence has also been reported in basolateral membrane vesicles (Horne and Reed, 1992).

On the bile canalicular membrane, at least four kinds of primary active transporters transport xenobiotics and endogenous compounds into bile: P-glycoprotein transporting mainly amphipathic organic cations and neutral compounds such as daunomycin and vinca alkaloids; canalicular bile salt export pump transporting bile acid such as taurocholate; mdr2 gene transporting phospholipids; and canalicular multispecific organic anion transporter (cMOAT) transporting organic anions such as 2,4-dinitrophenyl-S-glutathione (DNP-SG) and bilirubin glucuronides (Gatmaitan and Arias, 1995; Jedlitschky et al., 1997). We previously reported that the excretion of MTX into bile is greatly reduced in Eisai-hyperbilirubinemic rats (EHBRs), which have a hereditary deficiency in cMOAT (Masuda et al., 1997). Saturable kinetics, with ATP dependence, was found in its uptake by canalicular membrane vesicles (CMVs) obtained from Sprague-Dawley rats (SDRs), but not from EHBRs (Masuda et al., 1997). Thus, cMOAT seems to be predominantly involved in its biliary excretion. Biliary excretion of 5CH₃-H4PteGlu showed similar characteristics in its biliary excretion, and it has been identified as an endogenous cMOAT substrate in rats (Kusuhara et al., 1998).

In the present study, we attempted to investigate the hepatobiliary transport mechanism of MX-68. To this end, we examined the kinetics of membrane transport involved in the hepatic uptake and bile canalicular excretion of MX-68 using isolated rat hepatocytes and CMVs, respectively. To investigate the involvement of cMOAT in the latter membrane transport process, the biliary excretion of MX-68 was compared between SDRs and EHBRs.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. MX-68, [³H]MX-68 (6.60 MBq/µmol, 98.3% purity), and [¹⁴C]MX-68 (1.09 MBq/µmol, 98.3% purity) were synthesized at Fuji Gotemba Research Laboratories (Chugai Pharmaceutical Company, Shizuoka, Japan). [³H]TCA (128 MBq/µmol, 98.5% purity) and [³H]MTX (759 MBq/µmol, 98.6% purity) were purchased from PerkinElmer Life Science Products (Boston, MA). Folic acid and (6R,S)-5CH₃-H4PteGlu calcium salt were purchased from the laboratory of Dr. B. Schirck (Jona, Switzerland). TCA, MTX, carbonylcyanide p-(trifluoromethoxy)phenylhydrazone (FCCP), rotenone, 4,4'-diisothiocyanatostilbene 2,2'-disulfonic acid (DIDS), probenecid, and E2-17G were purchased from Sigma Chemical Corp. (St. Louis, MO). Collagenase was purchased from Wako Pure Chemical Industries (Osaka, Japan). All other chemicals and reagents were commercial products of analytical grade.

Animals. Male SDRs (Nisseizai, Tokyo, Japan) and EHBRs (Eisai Laboratories, Gifu, Japan), weighing 250 to 300 g, were used throughout the experiments. The animals had free access to water and food intake. This study was 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.

Isolated Hepatocytes Study.

Cell preparation Hepatocytes were isolated from male SDRs by the procedure of Baur et al. (1975). After isolation, the hepatocytes (2 mg of protein/ml) were suspended at 0°C in albumin-free Krebs-Henseleit buffer supplemented with 12.5 mM HEPES (pH 7.3). All studies were carried out in the presence of Na⁺ except for the studies of the Na⁺ dependence of substrate uptake. Cell viability was routinely checked by the trypan blue [0.4% (w/v)] exclusion test, and only those hepatocytes showing more than 95% viability were used. Protein concentrations were determined by using the method described by Bradford (1976), using the Bio-Rad protein assay kit with bovine serum albumin as a standard (Bio-Rad, (Hercules, CA).

Uptake study. Uptake of [³H]MX-68 was initiated by adding the substrate solution (0.1 ml) to the cell suspension (0.9 ml, 2 mg of protein/ml) that had been preincubated at 37°C for 5 min. At a designated time, the reaction was terminated by separating the cells from the medium using a centrifugal filtration technique (Schwenk, 1980). Briefly, 200-µl aliquots were placed into centrifuge tubes containing 50 µl of 2 N NaOH, covered by 100 µl of a mixture (density = 1.015) of silicone and mineral oils. The samples were then centrifuged for 15 s in a tabletop microfuge (10,000g, Beckman Instruments Inc., Fullerton, CA). Centrifugation pelleted the hepatocytes through the oil layer into the alkaline phase. After the cells had dissolved in the alkaline solution, the tube was sliced in two, and each compartment was transferred to a scintillation vial. The alkaline part was neutralized with 50 µl of 2 N HCl. Following addition of scintillation cocktail (Clear-sol II, Nacalai Tesque, Kyoto, Japan) to the vials, the radioactivity in the medium and cells was determined using a liquid scintillation spectrophotometer (LS 6000SE, Beckman Instruments). The uptake was normalized by dividing the uptake amount both by substrate concentration in the medium and amount of cellular protein, and is expressed as microliters per milligram of protein. The initial uptake velocity of MX-68 was calculated using a linear regression of points taken at 30 s and 90 s. The initial uptakes of [³H]TCA and [³H]MTX were also assessed by the same method.

Effect of Na⁺ in the medium. To determine the Na⁺-independent uptake, the study was performed in Krebs-Henseleit buffer after replacing the NaCl and NaHCO₃ with isotonic choline chloride and choline bicarbonate, respectively. Na⁺-dependent uptake was calculated by subtracting the initial uptake in the absence of Na⁺ from that in its presence.

Effect of pH in the medium. Uptake was initiated by the addition of hepatocytes (0.1 ml) suspended in Krebs-Henseleit buffer (pH 7.4) at 10 mg of protein/ml into Krebs-Henseleit buffer (0.9 ml) of appropriate pH containing [³H]MX-68, [³H]TCA, or [³H]5CH₃-H4PteGlu and following preincubation at 37°C for 5 min. The final pH of the incubation medium was checked after mixing and found to be 5.5, 6.5, or 7.5.

Inhibition studies. Cell suspension was preincubated with either metabolic inhibitors (2 µM FCCP or 30 µM rotenone) or anion exchanger inhibitor (100 µM DIDS) for 3 min at 37°C before adding MX-68. To determine the inhibitory effect of organic anions, either taurocholate, probenecid, or E2-17G was added to the cell suspension simultaneously with [³H]MX-68.

CMV Study.

Preparation of CMVs CMVs were prepared from male SDR and EHBR liver as described previously (Kobayashi et al., 1990). After suspending them in 50 mM Tris buffer (pH 7.4) containing 250 mM sucrose, the membrane vesicles were frozen in liquid N₂ and stored at 100°C until used. To check the purity of the prepared CMV, both Mg²⁺ ATPases and alkaline phosphatase were determined by the method of Schoner et al. (1967) and Yachi et al. (1989), respectively. The activity used in this study was also checked by measuring the ATP-dependent uptake of standard substrates, [³H]TCA (1 µM) and [³H]DNP-SG (1 µM: 0.1 µM labeled and 0.9 µM unlabeled), during a 2-min incubation at 37°C. Protein concentrations were determined as described previously (Bradford, 1976), using the Bio-Rad protein assay kit with bovine serum albumin as a standard. Since we previously reported that the ratio of inside-out vesicles in CMVs does not vary markedly between each CMV preparation (34.7 ± 0.9%; mean ± S.E. of seven CMV preparations) (Niinuma et al., 1999), we did not check this value in the present study.

Uptake study of MX-68 and DNP-SG by CMVs. The uptake study of [¹⁴C]MX-68 and [³H]DNP-SG was performed as reported previously (Niinuma et al., 1999). The transport medium (10 mM Tris, 250 mM sucrose, and 10 mM MgCl₂·2H₂O, pH 7.4) contained the substrates, 5 mM ATP, and an ATP-regenerating system (10 mM creatine phosphate and 100 µg/ml creatine phosphokinase). The transport reaction was started by rapidly mixing an aliquot of transport medium (16-18 µl) with the vesicle suspension (10 µg of protein in 2-4 µl). Transport was stopped by adding 1 ml of ice-cold buffer containing 250 mM sucrose, 0.1 M NaCl, and 10 mM Tris HCl (pH 7.4). The stopped reaction mixture was passed through a 0.45-µm filter (HAWP02500, Millipore Corp., Bedford, MA) and then washed twice with 5 ml of stop solution. Radioactivity retained on the filter and in the reaction mixture was mixed with scintillation cocktail (Clear-sol I, Nacalai Tesque) and determined using the LS 6000SE liquid scintillation counter. Uptake of ligands was normalized with respect to both the medium concentration of substrates and the amount of membrane protein. The ATP-dependent uptake was assessed by subtracting the uptake in the presence of AMP from that in the presence of ATP. For the inhibition study, the transport medium contained 1 µM [³H]DNP-SG and various concentrations of MX-68. The transport reaction was performed for 2 min.

Estimation of kinetic parameters. The kinetic parameters for MX-68 uptake by isolated hepatocytes and ATP-dependent MX-68 uptake by CMVs were calculated by fitting the data to the following equations, respectively:

(1)

(2)

where V₀ is the initial uptake rate of the drug (pmol/min/mg of protein), S is the drug concentration in the medium (µM), K_m is the Michaelis-Menten constant (µM), V_max is the maximum uptake rate (pmol/min/mg of protein), and P_dif is the nonspecific uptake clearance (µl/min/mg of protein). The fitting was performed by an iterative nonlinear least-squares method using a MULTI program (Yamaoka et al., 1981). The input data were weighted as the reciprocal of the observed values and the Damping Gauss Newton method was used as the fitting algorithm. The inhibition constant (K_i) of MX-68 for the ATP-dependent uptake of DNP-SG by CMVs was estimated by fitting the data, which were obtained in the inhibition study by varying the inhibitor (MX-68) concentration (I) with the substrate ([³H]DNP-SG) concentration kept at constant (1 µM), to the following equation:

(3)

where V_{0(+inhibitor)} and V_0(inhibitor) represent the initial uptake rate of [³H]DNP-SG in the presence and absence of MX-68, respectively. This equation was based on our previous finding that the substrate concentration is much lower than the K_m for DNP-SG (16-23 µM; Niinuma et al., 1997, 1999).

Statistical Analysis. For the analysis of the difference between two data sets, the test for equal variance (F-test) and a subsequent Student's t test were performed on the two means of the unpaired data. For the analysis of the other multiple data, Dunnett's test was performed. A p value of less than 0.05 was considered to be statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Time Profile and Concentration Dependence of MX-68 Uptake by Isolated Rat Hepatocytes. The uptake of MX-68 by isolated rat hepatocytes was linear up to 2 min and reached equilibrium by 10 min (Fig. 1A). The cell-to-medium concentration ratio at equilibrium was calculated to be approximately 2.6, considering the substrate concentration (1 µM) and intracellular volume (4.3 µl/mg of protein; Yamazaki et al., 1992). As shown in the inset of Fig. 1A, the MX-68 uptake was almost linear until 2 min, although there was a substantial y intercept. Therefore, in the following analysis, the uptake at both 30 and 90 s was determined to precisely estimate the initial uptake by subtracting the uptake at 30 s from that at 90 s. We assumed that the y intercept in the uptake profile mainly represents nonspecific adsorption to the cell surface because the y intercept of MX-68 did not exhibit clear saturation nor was it inhibited by other compounds. The initial uptake rate of MX-68 was determined at various substrate concentrations and shown as an Eadie-Hofstee plot in Fig. 1B. The uptake clearance (V₀/S) of MX-68 fell as the concentration increased, showing both saturable and unsaturable components. The kinetic parameters obtained were as follows: K_m = 2.15 ± 1.16 µM; V_max = 2.34 ± 0.99 pmol/min/mg of protein; and P_dif = 0.783 ± 0.055 µl/min/mg of protein (mean ± calculated S.D.). Since approximately 40% of the MX-68 uptake is unsaturable, the equilibrium cell-to-medium concentration ratio in the presence of an excess concentration of MX-68 concentration, i.e., which saturates the MX-68 uptake system, should be at least unity. Considering the high protein binding of MX-68 in the liver in vivo (unbound fraction in the liver is 0.1-0.2 at a liver concentration of 2.4-130 µM; Han et al., 1999), the actual cell-to-medium concentration ratio for unbound MX-68 at such a high concentration may be much less than unity. This hypothesis appears reasonable because such a value for a divalent anion should be approximately 0.05 if the membrane potential is assumed to be 40 mM.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Time profile (A) and substrate concentration dependence (B) of MX-68 uptake by isolated rat hepatocytes. A, hepatocyte suspension was incubated with [3H]MX-68 (1 µM), and its uptake was determined by a centrifugal filtration technique. Uptake was normalized both by the cellular protein and medium concentration of the substrate and expressed as microliters per milligram of protein. Inset, the expanded figure for the uptake until 2 min. B, Eadie-Hofstee plot describing the initial uptake rate of MX-68 at various concentrations (1-1000 µM). The fitted line based on eq. 1 is also shown. All data represent mean ± S.E. of three determinations in three preparations.

Effects of Na⁺ on the Initial Uptake of [³H]MX-68. Na⁺ replacement by choline had no effect on MX-68 uptake (Table 1), whereas a significant Na⁺ dependence was observed in the hepatic uptake of TCA and MTX. By comparing the uptake in the presence of Na⁺ with that in its absence, the Na⁺-dependent component in the TCA and MTX uptake was estimated to be approximately 70 and 50% of the total uptake, respectively.

Effects of pH on [³H]MX-68 Uptake by Hepatocytes. The effect of varying the extracellular pH on the initial rate of uptake of [³H]MX-68 was examined. As positive and negative control experiments, the effect of pH on the uptake of [³H]5CH₃-H4PteGlu and [³H]TCA was also investigated, respectively. The initial uptake rate of [³H]TCA was unaffected by the extracellular pH while the initial uptake of [³H]MX-68 and [³H]5CH₃-H4PteGlu increased as the extracellular pH fell from 7.5 to 5.5 (Fig. 2). The pK_a values for the - and -carboxylic groups and the N₁ position in the pteridine ring are 3.85, 4.76, and 5.5, respectively. Therefore, MX-68 is mostly in the dianion form at pH 7.5 (~99%). These pKa values indicate that the ratio of totally un-ionized form should be 0.17, 0.0036, and 4.0 × 10⁵% at pH 5.5, 6.5, and 7.5, respectively. Since the decrease in this ratio is much sharper than that in the uptake shown in Fig. 2A, the pH-dependent uptake cannot be simply explained by a change in the ratio of un-ionized form.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Effect of extracellular pH on the initial uptake rate of MX-68 (A), 5CH3-H4PteGlu (B), and TCA (C) by isolated rat hepatocytes. The initial uptake velocity of the substrate at 1 µM was determined by subtracting the uptake at 30 s from that at 90 s in media of different pH. Data represent the mean ± S.E. of three determinations in three preparations. *Significantly different from the uptake at pH 5.5 (p < 0.05).

Effects of Metabolic Inhibitors and an Anion Exchanger Inhibitor on the Initial Uptake of [³H]MX-68 by Hepatocytes. [³H]MX-68 uptake was reduced by the presence of FCCP (Fig. 3B), rotenone (Fig. 3B), or DIDS (Fig. 3A). The initial uptake of [³H]MX-68 in the presence of these reagents was about 40 to 50% of the control (Fig. 3, A and B), which was almost comparable with the uptake of [³H]MX-68 in the presence of an excess (200 µM) concentration of unlabeled MX-68 (Fig. 3C). In Fig. 3, the uptake was not normalized by the intracellular volume because we previously found that there was no significant difference in the intracellular volume, assessed as the uptake of ³H₂O, between control and rotenone-treated hepatocytes (Yamazaki et al., 1992).

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Effect of anionic compounds (A), ATP depletors (B), and folate and its analogs (C) on the initial uptake rate of MX-68 by isolated rat hepatocytes. The initial uptake velocity of MX-68 at 1 µM was determined by subtracting the uptake at 30 s from that at 90 s in the presence or absence of the compounds described. All values were normalized by the uptake in the absence of these inhibitors. Data represent the mean ± S.E. of three determinations in three preparations. *Significantly different from the control (p < 0.05).

Effects of Organic Anions and Folate Analogs on [³H]MX-68 Uptake by Hepatocytes. A bile acid, TCA, and organic anions, E2-17G and probenecid, inhibited [³H]MX-68 uptake at 100 µM (E2-17G) and 200 µM (TCA and probenecid) (Fig. 3A). The uptake was reduced to about half the control by the presence of these inhibitors. The inhibitory effect of various folate analogs was also assessed (Fig. 3C). MTX and 5CH₃-H4PteGlu produced comparable inhibition of [³H]MX-68 uptake, but this inhibitory effect was slightly less than that of unlabeled MX-68 (Fig. 3C). The inhibitory effect of 5 and 200 µM folic acid was minimal, and the uptake of [³H]MX-68 in the presence of folic acid was not significantly different from the control level (p > 0.1) (Fig. 3C).

Characterization of CMVs. The CMVs prepared from SDRs used in the present study exhibited ATP-dependent uptake of [³H]DNP-SG (160 ± 9.5 µl/mg of protein/2 min, mean ± S.E. of three preparations), while the CMVs prepared from EHBRs had approximately 1/40 the activity in terms of [³H]DNP-SG uptake (4.7 ± 0.5 µl/mg of protein/2 min, mean ± S.E. of three preparations). Both CMVs exhibited comparable activity with respect to ATP-dependent [³H]TCA uptake (140 ± 15 and 100 ± 7 µl/mg of protein/2 min for SDRs and EHBRs, respectively). The enrichment of the specific activity for canalicular marker enzymes in CMVs compared with liver homogenate was examined and found to be almost comparable between SDRs and EHBRs (97.6 ± 9.5 and 137 ± 11 for Mg²⁺ ATPases, 75.3 ± 6.9 and 98.2 ± 8.5 for alkaline phosphatase in CMVs prepared from SDRs and EHBRs, respectively). Since each CMV preparation used in the present study was obtained from five rat livers, we believe that any interindividual difference is minimized in the transport studies. In the following analysis, data were obtained as triplicate determinations in two CMV preparations, and a statistical analysis was performed on a total of six data sets.

Time Course and Saturation Kinetics of MX-68 Uptake by CMVs. Time profiles for the uptake of MX-68 by CMVs are shown in Fig. 4. In CMVs prepared from SDRs, ATP stimulation and overshoot phenomena were observed (Fig. 4A), while minimal ATP stimulation was observed in the uptake by CMVs prepared from EHBRs (Fig. 4B). Since a linear uptake of MX-68 by CMVs could be observed for at least 2 min (Fig. 4A), the kinetic profiles were checked at 2 min of incubation. The ATP-dependent uptake of MX-68, obtained by subtracting the uptake of MX-68 in the presence of AMP from that in the presence of ATP, showed saturation with a single saturable component (Fig. 5). The kinetic parameters for the ATP-dependent MX-68 uptake were as follows: K_m = 207 ± 62 µM and V_max = 1780 ± 190 pmol/min/mg of protein.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Time profiles for the uptake of MX-68 by CMVs prepared from SDRs (A) and EHBRs (B). The uptake of [14C]MX-68 (12.5 µM) by CMVs was determined by the rapid filtration technique in the presence of ATP () or AMP (). Data represent the mean ± S.E. of three determinations in two preparations.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Eadie-Hofstee plots for the ATP-dependent uptake of MX-68 by CMVs prepared from SDRs. Initial uptake velocity of MX-68 at various concentrations was determined after 2 min of incubation with CMVs. ATP-dependent uptake was calculated by subtracting the uptake in the presence of AMP from that in the presence of ATP. Data represent the mean ± S.E. of three determinations in two preparations. The fitted line based on eq. 2 is also shown.

Inhibition Study of the Uptake of [³H]DNP-SG by CMVs. The effect of MX-68 on the initial uptake of [³H]DNP-SG was examined. ATP-dependent uptake of [³H]DNP-SG was reduced in a concentration-dependent manner by unlabeled MX-68 (Fig. 6). The K_i of 162 µM obtained using eq. 3 was comparable with the K_m (207 µM) for ATP-dependent MX-68 uptake.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Effect of MX-68 on the ATP-dependent DNP-SG uptake by CMVs prepared from SDRs. Initial velocity of ATP-dependent uptake of DNP-SG (1 µM) in the presence of various concentrations of MX-68 was determined after 2 min of incubation with CMVs. Data represent the mean ± S.E. of three determinations in two preparations. The fitted line based on eq. 3 is also shown.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our previous experimental observations showed that biliary excretion of MX-68 is predominant compared with its urinary excretion in rats (Han et al., 1999). Therefore, the present study focused on the transport characteristics involved in the hepatobiliary transport of MX-68. Our present findings demonstrate that active transport systems are involved both in the uptake and excretion of MX-68 on sinusoidal and canalicular membranes, respectively.

The hepatocellular uptake of MX-68 should be an important factor in determining the degree of net biliary excretion clearance as a first step in hepatobiliary transport. Freshly isolated hepatocytes represent a useful model system for examining transport functions in the intact liver because hepatocytes in suspension maintain their membrane and metabolic integrity for several hours (Quistorff et al., 1973; Dickson and Pogson, 1977). Therefore, the hepatic uptake process of MX-68 was kinetically analyzed in isolated hepatocytes. The initial uptake of MX-68 by isolated hepatocytes was concentration-dependent and showed a high-affinity component with a K_m of 2.2 µM (Fig. 1), slightly lower than the K_m (5.9 µM) of MTX (Gewirtz et al., 1980). The cell-to-medium concentration ratio (2.6) at equilibrium was more than unity, and this uptake was reduced by treatment with ATP depletors (Fig. 3B). These results indicate that at least a part of the MX-68 uptake is governed by carrier-mediated active transport. We have previously determined the steady-state liver-to-plasma concentration ratio at various infusion rates of MX-68 in rats (Han et al., 1999). The concentration ratio thus obtained also exhibited saturation at an unbound plasma concentration around 0.3 to 3 µM (Han et al., 1999). This saturation is compatible with the saturation observed in the uptake of MX-68 by isolated rat hepatocytes (Fig. 1B). Therefore, we speculated that the uptake mechanism for MX-68 is localized on the sinusoidal membranes, although we cannot exclude the possibility that other mechanisms expressed on lateral and/or canalicular membranes also affect the present findings in isolated rat hepatocytes. By comparing V_max/K_m with P_dif, the contribution of an unsaturable component to MX-68 uptake under linear conditions can be calculated as being 42% of the total uptake (Fig. 1B), which is close to the uptake in the presence of metabolic inhibitors (approximately 40% of the control; Fig. 3B). Thus, carrier-mediated transport may account for approximately half of the total uptake of MX-68.

It has been reported that the hepatic uptake of many types of organic anions exhibits Na⁺ dependence. Here, we have investigated the Na⁺ dependence in hepatic uptake of MX-68 as well as that of TCA and MTX as control experiments. As shown in Table 1, Na⁺ dependence was found both in the uptake of TCA and MTX. The degree of the Na⁺-dependent portion of MTX uptake (approximately half the total uptake; Table 1) was comparable with a previous report that was also obtained using isolated rat hepatocytes (Horne et al., 1976; Gewirtz et al., 1980). On the other hand, the uptake of MX-68 was not reduced after replacing Na⁺ with choline (Table 1). These results suggest that the hepatic uptake system for MX-68 is, at least partially, different from that for MTX, although the molecular structures of these two antifolates are similar.

In addition to the Na⁺ gradient, a transmembrane H⁺ gradient is also known to energize the uptake of folates in isolated rat hepatocytes (Horne, 1990), intestinal brush-border membrane vesicles from rabbits (Schron et al., 1985) and humans (Said et al., 1987), and rat kidney brush-border membranes (Bhandari et al., 1988). Therefore, we examined the pH dependence of MX-68 transport into hepatocytes (Fig. 2). Our results indicate that the initial uptake of 5CH₃-H4PteGlu into hepatocytes is greater at pH 5.5 than at pH 7.5 (Fig. 2B). This result is compatible with the previous finding reported by Horne (1990). Under the same conditions, the uptake of MX-68 also increased as the extracellular pH fell from 7.5 to 5.5 (Fig. 2A). Thus, MX-68 is also taken up by an H⁺-dependent mechanism, although further studies are needed to identify the actual driving force.

To characterize the MX-68 uptake system, inhibition by anionic compounds was also examined (Fig. 3A). MX-68 uptake was reduced to approximately 50% of the control by TCA, E2-17G, or probenecid, suggesting that these compounds are also recognized by the MX-68 transporter. Similarly, an anion exchanger inhibitor (DIDS) inhibited MX-68 uptake, reducing it to half the control level (Fig. 3A). Considering that about half the MX-68 uptake was mediated by an unsaturable diffusion component (Fig. 1B), it is reasonable that only half the total MX-68 uptake was inhibited by these compounds. MX-68 uptake was also inhibited by MTX and 5CH₃-H4PteGlu in a concentration-dependent manner (Fig. 3C). Thus, the MX-68 uptake system has a broad specificity for organic anions, including folate analogs. The inhibitory effect of MTX and 5CH₃-H4PteGlu at 200 µM seemed to be slightly less than the same concentration of unlabeled MX-68 (Fig. 3C). Considering the K_m values for MTX and MX-68 uptake by rat hepatocytes, this concentration should be enough to almost completely saturate their own uptake systems. Therefore, such a small difference in inhibitory effect between MTX and MX-68 also suggests that the uptake system for MX-68 differs in some respect from that for MTX. On the other hand, minimal inhibition of MX-68 uptake was found in the presence of oxidized folate, folic acid (Fig. 3C). Thus, folic acid has a lower affinity for MX-68 transporter than MTX and 5CH₃-H4PteGlu. Gewirtz et al. (1980) reported that 100 µM folic acid was also unable to inhibit the uptake of 1 µM [³H]MTX in isolated hepatocytes, whereas 100 µM unlabeled MTX clearly inhibited [³H]MTX uptake. Thus, the major transport system of folic acid is different from MTX and MX-68. Concerning folate analog transporter, reduced folate carrier (RFC) has been identified as being able to transport both 5CH₃-H4PteGlu and MTX and has been shown to be expressed in several organs including the liver, kidney, and intestine (Dixon et al., 1994; Said et al., 1996; Brigle et al., 1997). Previous reports using RFC cRNA-injected Xenopus laevis oocytes and a cDNA-transfected cell line indicated that folic acid has a much lower affinity for RFC than 5CH₃-H4PteGlu and MTX (Dixon et al., 1994; Said et al., 1996). Considering the similarity in the chemical structure of MTX and MX-68, RFC might be a possible candidate for a common transport system for these two compounds, although further research must be performed to identify the MX-68 transporters.

cMOAT excretes amphiphilic organic anions into bile and is defective in EHBRs derived from SDR strains. We previously found that the in vivo biliary excretion of MX-68 is almost completely reduced in EHBRs compared with that in SDRs over a wide range of plasma and liver concentrations (Han et al., 1999). In the present study, we further investigated the in vitro biliary transport mechanism using CMVs to support the previous in vivo findings. This study suggests that cMOAT is predominantly responsible for the biliary excretion of MX-68 because 1) MX-68 uptake by CMVs prepared from SDRs exhibited ATP dependence, whereas minimal ATP dependence was observed in EHBR CMVs (Fig. 4); 2) ATP-dependent transport of MX-68 exhibited saturation (Fig. 5); and 3) ATP-dependent uptake of DNP-SG, a typical substrate for cMOAT, was inhibited by MX-68 in a concentration-dependent manner with the resultant K_i (162 µM) comparable with the K_m (207 µM) of ATP-dependent MX-68 uptake (Fig. 6). We have also obtained similar findings for the biliary excretion of MTX (Masuda et al., 1997). The K_m and V_max of MX-68 transport found in the present study was almost identical to those of MTX (Masuda et al., 1997). Therefore, the transport system on bile canalicular membrane seems to be basically similar for MX-68 and MTX. The steady-state hepatic MX-68 concentration (1.58-143 µM) observed in the previous in vivo study (Han et al., 1999) was still lower than the K_m for the ATP-dependent MX-68 uptake by CMVs found in the present study. Therefore, canalicular membrane transport of MX-68 in vivo should not be saturated to such an extent, even at the highest hepatic concentration. Thus, it is reasonable that CL_bile,h,u, defined as the ratio of the biliary excretion rate to the hepatic unbound concentration, did not fall as the hepatic concentration increased (Han et al., 1999).

In conclusion, the hepatic uptake of MX-68 on the sinusoidal membrane is mediated by Na⁺-independent active transport systems that also recognize other organic anions and folate analogs but are different in some respects from that for MTX. The biliary excretion of MX-68 via the bile canalicular membrane is mediated mainly by cMOAT with a similar affinity and transport capacity to MTX.