Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Many chiral compounds for which the enantiomers show different pharmacokinetics have been reported (Jamali et al., 1989). For renal excretion, stereoselective glomerular filtration dependent on stereoselective plasma protein binding was reported (Lee and Williams, 1990), and the stereoselective secretion of organic cations was suggested for several chemicals (Hsyu and Giacomini, 1985; Lima et al., 1985; Ofori-Adjei et al., 1986; Notterman et al., 1986). Although the stereoselective secretion of acidic chiral compounds was not reported until quite recently, we were able to clearly show that DBCA² was R-(+)-preferentially secreted in isolated perfused kidney (Higaki et al., 1994). (R)-(+)-DBCA has a CL_int,s value 3.3 times greater than that for (S)-()-DBCA during perfusion of racemic DBCA, and the intrinsic capability of secretion for (R)-(+)-DBCA was estimated to be 2 times larger than that for the antipode (Higaki et al., 1994). However, DBCA was R-(+)-predominantly N-monodemethylated in kidney as well as in liver (Higaki and Nakano, 1992); MBCA, which is much more available for renal secretion than DBCA, was generated in the perfusion system, and the interaction between these compounds in renal secretion complicated the analysis (Higaki et al., 1994). Renal tubular secretion is composed of three processes, i.e. uptake across the BLM of epithelial cells, intracellular movement, and secretion across the BBM. It remains to be determined which process is stereoselective. MBCA is also an acidic chiral compound, and its renal secretion is thought to be stereoselective, based on the results of an in vivo study of DBCA (Higaki et al., 1992). Moreover, MBCA remains virtually unmetabolized (Higaki et al., 1994), so its kinetics can be clearly evaluated. In the present study, we chose MBCA as a substrate and attempted to demonstrate more explicitly the stereoselective renal tubular secretion of an organic anion and to clarify the determining step of its stereoselective kinetics in an isolated kidney perfusion system, using the multiple-indicator dilution method.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Materials. MBCA, (R)-(+)-MBCA, (S)-()-MBCA, and 6,7-dichloro-2,3-dihydro-5-(pyrolidinosulfonyl)-2-benzofuran-carboxylic acid (internal standard) were synthesized at Shionogi Research Laboratories (Harada et al., 1987a,b). (R)-(+)-[¹⁴C]MBCA (1.18 MBq/mg) and (S)-()-[¹⁴C]MBCA (1.21 MBq/mg) were synthesized at Developmental Research Laboratories, Shionogi & Co., Ltd. (Osaka, Japan). Their radiochemical purities were >99% and >98%, respectively, as determined by HPLC (column, Nucleosil 5C₁₈, 4.6 × 150 mm; mobile phase, methanol/Pic.A, 3:7, v/v; flow rate, 1 ml/min; RI detector, Packard TRACE II). Their optical purities were >98%, as determined by HPLC (column, Nucleosil 5C₁₈, 4.6 × 150 mm; mobile phase, acetonitrile/water, 45:55, v/v; flow rate, 1 ml/min; detector, UV detector set at 223 nm) after their diastereomerization using (S)-()-1-(1-naphthyl)ethylamine. [³H]Inulin (11 MBq/mg) was purchased from American Radiolabeled Chemicals (St. Louis, MO). (S)-()--Methylbenzylamine and 1,1-carbonylbis-2-methylimidazole were purchased from Aldrich Chemical Co. (Milwaukee, WI). The optical purity of (S)-()--methylbenzylamine was 98%. BSA (fraction V) was obtained from Sigma Chemical Co. (St. Louis, MO). Creatinine, amino acids, mannitol, and Evans blue were purchased from Wako Pure Chemical Industries (Osaka, Japan).

Animals. Male Jcl:SD rats (body weight, approximately 300 g; Nippon Clea Co., Tokyo, Japan) were used in all experiments.

Preparation of Perfusate. KRB buffer solution, the composition of which was reported previously (Higaki et al., 1994), was used in all perfusion studies. The perfusate was prepared by mixing two solutions prepared independently, as described below. In KRB buffer (pH 7.4), which had been filtered through membrane filters (0.45 µm; Nihon Millipore Ltd., Tokyo, Japan), were dissolved amino acids, creatinine, and mannitol (solution A). Amino acids have been reported to be important to maintain the function of perfused kidneys (Epatein et al., 1982; Bekersky, 1983; Radermacher et al., 1991). Mannitol (final concentration, 3%) was added to increase the excreted urine volume for the determination (Hori et al., 1988). The solution was then filtered again through a 0.45-µm filter. BSA (fraction V; final concentration, 4.6%) was dissolved in another filtered KRB buffer, the solution was refiltered with an 8-µm filter (Nihon Millipore Ltd.) and with a 0.45-µm filter (solution B), and then solutions A and B were mixed (solution C). BSA was not dialyzed before use, because the contaminants were needed for the normal function of the perfused kidneys (Bekersky, 1983). Fresh bovine red blood cells were washed with the filtered KRB buffer and added to solution C at the concentration of 20% (v/v). Finally, the concentrations of components were adjusted as reported previously (Higaki et al., 1994). The resulting perfusate was delivered to the kidney through a filter (TF-AG25LS; Terumo Co., Tokyo, Japan), to remove the aggregate of red blood cells.

Preparation of Drug Solution. The drug solution for injection through the arterial cannula was prepared as follows. (R)-(+)-[¹⁴C]MBCA (final concentration, 222 kBq/500 nmol/ml) or (S)-()-[¹⁴C]MBCA (222 kBq/500 nmol/ml), [³H]inulin (592 kBq/50 mg/ml), Evans blue (2.46 mg/ml), and BSA (46 mg/ml) were dissolved in the filtered KRB buffer. Finally, the solution was subjected to ultrafiltration by using Centricut filters (W-MO, 0.45 µm; Kurashiki Bouseki, Osaka, Japan), and the filtrate was used as a drug solution.

Preparation of Isolated Perfused Kidneys. Isolated perfused rat kidneys were prepared according to the method described by Nishiitsutsuji-Uwo et al. (1967), with minor modification. Briefly, the abdomen of the rat was dissected along the midline, under diethyl ether anesthesia. The ureter of the right kidney was cannulated with polyethylene tubing (PE-10, 0.011 inch i.d. × 0.024 inch; Becton Dickinson and Co., Parsippany, NJ), and then heparin (200 units) and mannitol (100 mg) (in 1 ml of KRB buffer) were injected from the left femoral vein. After ligation of the left renal vein, the inferior vena cava was cannulated with polyethylene tubing (PE-240, 0.066 inch i.d. × 0.095 inch; Becton Dickinson). The arterial cannula (PE-60, 0.030 inch i.d. × 0.048 inch; Becton Dickinson) was inserted into the superior mesenteric artery and manipulated across the aorta into the right renal artery without interruption of renal blood flow. Immediately after this, delivery of the perfusate with a peristaltic pump (RP-VT3; Furue Science, Tokyo, Japan) was begun. The isolated kidney preparation was placed in a single-pass perfusion apparatus enclosed in a temperature-controlled environment maintained at 37°C. Preperfusion for approximately 10 min could stabilize the perfused kidney conditions with approximately 120 mm Hg perfusion pressure. The perfusate was aerated with 95% oxygen/5% carbon dioxide, and the perfusion pressure was monitored and recorded with a blood pressure-monitoring system (transducer, TP-400T; amplifier, AP-641G; recorder, WT-625G; Nihon Kohden Co., Tokyo, Japan) throughout the experiments.

Multiple-Indicator Dilution Study. Multiple-indicator dilution studies (Goresky et al., 1973; Itoh et al., 1986) were performed as follows. After a 10-min preperfusion, 100 µl of drug solution containing (R)-(+)-[¹⁴C]MBCA (222 kBq/500 nmol/ml) or (S)-()-[¹⁴C]MBCA (222 kBq/500 nmol/ml), [³H]inulin (592 kBq/50 mg/ml), Evans blue (2.46 mg/ml), and BSA (46 mg/ml) was bolus-injected from the arterial cannula. The outflow of perfusate from the venous cannula was collected every 1 sec for 10 sec, every 2 sec to 30 sec, every 5 sec to 60 sec, every 15 sec to 120 sec, and every 1 min to 15 min, and urine was collected every 1 min for 30 min.

Calculation of MTT_c Values. In this experimental system, the MTT_c values for the cannulae inserted into the renal artery and the inferior vena cava must be determined to obtain a precise estimation of MTT_kidney (Hori et al., 1988). Only the cannulae for the renal artery and inferior vena cava were perfused with the perfusate; 100 µl of Evans blue solution (2.46 mg/ml) was bolus-injected from the arterial cannula, and the outflow from the cannula for the inferior vena cava was collected every 1 sec for 20 sec. After centrifugation of the perfusate outflow, the concentration of Evans blue in the supernatant was determined as described in Analytical Procedures, and MTT_c was calculated. The natural logarithm of MTT_c was plotted against the plasma flow rate of the perfusate, and the regression line was obtained (fig. 1). Using the equation for the regression line, the MTT_c values were calculated for the plasma flow rates for the isolated kidney perfusion studies. The actual MTT_kidney value was obtained by subtracting the calculated MTT_c from the MTT_kidney value calculated from the experimental data.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Relationship between MTTc and plasma flow rate of perfusate. Circles, observed values; solid line, linear regression line, of which the slope and y intercept values were 0.1074 sec·min/ml and 2.5055 sec, respectively. The correlation was statistically significant at the probability level of 0.1% (r = 0.9745).

Protein Binding and Erythrocyte Partitioning Studies. The outflow "blood" perfusate was centrifuged at 3000 rpm for 10 min at 4°C (centrifuge, H-103RS; Kokusan, Tokyo, Japan), and aliquots of the supernatant (the plasma perfusate) were used to determine the total plasma concentration. The remaining supernatant was subjected to ultrafiltration. After the plasma perfusate was incubated at 37°C, about 1 ml was transferred to a Centrifree tube (Centrifree micropartition system; Amicon Inc., Beverly, MA). The tube was centrifuged at 1300g for 10-20 min at 37°C (H-103RS). Under these conditions, approximately 150 µl of filtrate was obtained. There was negligible adsorption of chemicals to the filtration system. The f_p values for chemicals were estimated directly from the ratio of the drug concentration in the filtrate to the total concentration.

Analytical Procedures. Determinations with Samples in Isolated Kidney Perfusion Studies. Total radioactivities in the outflow of plasma perfusate after centrifugation and in urine were measured with a liquid scintillation counter (Tri-Carb 2000CA; Packard Instrument Co., Downers Grove, IL), after the addition of Pico-fluor 40 (Packard). Radioactivity in the kidney, after solubilization with Soluene-350 (Packard) and addition of Pico-fluor 40, was determined with a liquid scintillation counter. Evans blue-BSA in the perfusate was determined at 610 nm, after dilution with distilled water.

Determination of Unlabeled MBCA. Concentrations of MBCA in the perfusate were determined after diastereomerization by the method previously reported, with minor modification (Nakano and Kawahara, 1991). Briefly, 1 ml of the plasma perfusate was mixed with 6 ml of methanol and 20 µg of internal standard. This was vortex-mixed for 10 min and centrifuged at 3000 rpm. The supernatant was removed to another test tube and evaporated. The residue was dissolved in 1 ml of 1 N HCl and extracted with 2 ml of diethyl ether by vortex-mixing and centrifugation. The organic solvent extracts were mixed with 0.5% 1,1-carbonylbis-2-methylimidazole in 400 µl of acetonitrile and maintained for 10 min at room temperature. The sample was mixed with 50 µl of 2% (S)-()--methylbenzylamine in acetonitrile and 100 µl of 2% acetic acid in acetonitrile and maintained for 30 min at room temperature. Twenty microliters of the reaction mixture were injected into an HPLC system. The lower limit of detection was 0.25 nmol/ml for MBCA enantiomers. The chromatographic system consisted of a model LC-6A HPLC pump, a model SIL-6B system controller, and a UV detector (Shimadzu, Kyoto, Japan) set at 254 nm. A Nucleosil C₁₈ column (150 mm × 4.6 mm i.d.; pore size, 120 Å; particle size, 5 µm; Chemcopacked; Macherey-Nagel, Duren, Germany) was used at room temperature. The mobile phase was tetrahydrofuran/methanol/water/acetic acid (30:10:60:1, by volume), delivered at 0.9 ml/min.

Determination of Glucose Levels. Glucose levels in the plasma perfusate and the urine were determined by the glucose oxidase method (Glucose B-Test Wako; Wako Pure Chemical Co.).

Determination of Sodium Levels. Sodium amounts in the plasma perfusate and the urine were measured by flame photometry (model 943; Instrumentation Laboratory Inc., Paderno Dugnano, Italy).

Data Analysis. The GFR was estimated by the inulin clearance calculated using eq. 1,

(1)

where AUC_inulin represents the AUC for inulin. The reabsorption percentage for glucose or sodium was obtained with eq. 2,

(2)

where C_p,in, C_u, and V_u indicate the plasma concentration of the inflow perfusate, the concentration in urine of glucose or sodium, and the volume of urine, respectively.

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Two-compartment, well-stirred model describing the renal tubular secretion of MBCA. Q, Cin, Cout, k1, and k2 represent renal plasma flow rate, inflow drug concentration, outflow drug concentration, rate constant for uptake across the BLM, and rate constant for secretion across the BBM, respectively. Vd,1 and Vd,cell represent distribution volumes for the extracellular space and epithelial cells, respectively.

Statistical Analysis. Statistical significance was estimated by using Student's t test. Results are expressed as mean ± SD.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Estimation of the Viability of Perfused Kidneys. The parameters that characterize the physiological function of perfused kidneys are summarized in table 1. Perfusate arterial pressure was maintained at approximately 120 mm Hg, by adjustment of the perfusate flow rate, in all studies. The flow rate was approximately 9 ml/min, as a plasma flow rate (table 1). Similar relationships between perfusion rate and pressure have been reported in studies using similar perfusates (Lieberthal et al., 1987; Hori et al., 1988; De Lannoy et al., 1989; Higaki et al., 1994). The GFR calculated from the urinary excretion of inulin was smaller (table 2) than those reported previously (Lieberthal et al., 1987; Hori et al., 1988; De Lannoy et al., 1989; Higaki et al., 1994), but the reabsorption of D-glucose and sodium was very efficient and the excretion rate of urine was regarded as appropriate (table 1), from comparison with other reports (Lieberthal et al., 1987; Hori et al., 1988; De Lannoy et al., 1989; Higaki et al., 1994). Considering these functional parameters, the viability of the perfused kidneys was judged to have been maintained throughout the perfusion studies.

Calculation of MTT_c Values. To precisely calculate MTT_kidney, the MTT_c value must be subtracted from the experimental value of MTT_kidney. Because MTT_c was reported to be dependent on the flow and independent of the substrate (Hori et al., 1988), the MTT_c for Evans blue-BSA was measured at eight flow rates. The logarithmic MTT_c values were plotted against the plasma flow rates (fig. 1), and their significant correlation (p < 0.001, r = 0.9745) and the equation describing their relationship were obtained as follows: ln(MTT_c) = 0.1074 · plasma flow rate + 2.5055. Using this equation, MTT_c was calculated for each perfusion study.

Protein Binding and Erythrocyte Partitioning. The f_p values of (R)-(+)- and (S)-()-MBCA were determined in the concentration range from 5 to 50 µM. The binding of both enantiomers to BSA was constant, and f_p values of 21.1 ± 2.7% and 9.8 ± 1.7% were obtained for the R-(+)- and S-()-enantiomers, respectively. The percentages of erythrocyte partitioning for (R)-(+)- and (S)-()-MBCA were 4.28 ± 1.69% and 3.68 ± 0.89%, respectively. It is reasonable to analyze the kinetics on the basis of plasma concentrations because of the limited partitioning of the enantiomers into erythrocytes.

Analysis of the Renal Excretion Kinetics of (R)-(+)- and (S)-()-MBCA. Fig. 3 shows typical patterns of the outflow from the renal vein and the urinary excretion rate-time curves after bolus injection of (R)-(+)-[¹⁴C]MBCA, (S)-()-[¹⁴C]MBCA, [³H]inulin, or Evans blue-BSA into the renal artery of the perfused kidney. Evans blue-BSA, which was not eliminated in the kidney, showed the highest concentration and rapidly decreased from the venous outflow. Although the venous outflow pattern for (S)-()-MBCA was similar to that for inulin, the concentrations of the R-(+)-enantiomer were apparently lower than those of inulin. The urinary excretion rates for both MBCA enantiomers were much faster than that for inulin, and the urinary concentration of the R-(+)-enantiomer was higher than that of its antipode, showing R-preferential secretion into the urine.

View larger version (39K): [in this window] [in a new window]   Fig. 3.   Typical patterns for renal vein outflow and urinary excretion rate-time curves for MBCA enantiomers. Upper, renal vein outflow curves; lower, urinary excretion rate-time curves for MBCA enantiomers. , MBCA enantiomers; , inulin; , Evans blue-BSA.

Effect of (R)-(+)-MBCA on the Elimination Kinetics of (S)-()-MBCA. To confirm stereoselective uptake across the BLM, we investigated which processes in the renal secretion of the S-()-enantiomer would be affected by the presence of the antipode. Fig. 4 shows semilogarithmic plots of the renal venous outflow patterns for (S)-()-MBCA with constant infusion of 0-200 µM (R)-(+)-MBCA. Although no effect of 10 µM (R)-(+)-MBCA was seen, 50 µM and especially 200 µM (R)-(+)-MBCA hastened S-()-enantiomer transit through the kidney. The urinary excretion rates were inhibited more markedly with increasing (R)-(+)-MBCA concentrations, and they dropped below those of inulin at 200 µM (R)-(+)-MBCA (fig. 5). In the simultaneous presence of 10, 50, or 200 µM (R)-(+)-MBCA, the f_p for (S)-()-MBCA was determined in the range of 5-50 µM. The f_p values obtained were 19.7 ± 4.5, 21.6 ± 5.5, and 23.9 ± 4.7% in the presence of 10, 50, and 200 µM (R)-(+)-MBCA, respectively. Although (R)-(+)-MBCA increased the f_p value for (S)-()-MBCA, the f_p value for the S-()-enantiomer was constant at each concentration of (R)-(+)-MBCA. The parameters calculated from the dilution patterns in figs. 4 and 5 are shown in fig. 6. The urinary extraction ratio and MTT_kidney values decreased with increases in the R-(+)-enantiomer concentration in the perfusate, although no change was observed in MTT_urine values (fig. 6A). The CL_int,s value for the S-()-enantiomer decreased significantly in the presence of the antipode, indicating that secretion of (S)-()-MBCA was significantly inhibited by the antipode (fig. 6A). In addition, the k₁, k₁/f_p, and V_d,cell values were found to be significantly decreased, which indicates that the two enantiomers competed in the processes of uptake across the BLM and intracellular distribution (fig. 6B). On the other hand, although _cell was decreased at 200 µM (R)-(+)-MBCA (fig. 6B), the profile did not correspond to the prominent changes in CL_int,s, k₁, k₁/f_p, and V_d,cell. Moreover, because _cell,sp and k₂ did not change significantly (fig. 6B), MBCA enantiomers are not thought to compete with each other in the processes of intracellular movement and secretion across the BBM. The results of the competitive studies support the stereoselective uptake of MBCA enantiomers across the BLM of renal epithelial cells.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Effects of (R)-(+)-MBCA on the renal vein outflow curves for (S)-()-[14C]MBCA after bolus injection. Each dilution pattern represents the semilogarithmic plot of the renal outflow of (S)-()-[14C]MBCA after bolus injection during the constant infusion of (R)-(+)-MBCA at different rates. , 0 µM; , 10 µM; , 50 µM; , 200 µM (R)-(+)-MBCA.

View larger version (35K): [in this window] [in a new window]   Fig. 5.   Effects of (R)-(+)-MBCA on the urinary excretion rate-time curves for (S)-()-[14C]MBCA after bolus injection. Each dilution pattern represents the renal outflow of (S)-()-[14C]MBCA after bolus injection during the constant infusion of (R)-(+)-MBCA at different rates. , (S)-()-MBCA; , inulin.

View larger version (30K): [in this window] [in a new window]   Fig. 6.   Effects of (R)-(+)-MBCA on the renal handling of (S)-()-MBCA. Results are expressed as the mean ± SD of at least three experiments. Statistically significant differences from results in the absence of (R)-(+)-MBCA are indicated as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Little has been known about the stereoselective renal tubular secretion of chiral organic anions (Jamali et al., 1989; Lee and Williams, 1990). One reason could be the lack of probes to estimate tubular secretion in in vivo studies. Stereoselective glucuronidation before secretion, as in the case of ibuprofen (Ahn et al., 1991), and the large contributions of biliary excretion and chiral inversion, as in the case of ketoprofen (Foster et al., 1988; Foster and Jamali, 1988), make it difficult to isolate the process of renal tubular secretion from whole-body kinetics and to assess it precisely. Even in the case of DBCA, for which we reported stereoselective renal tubular secretion (Higaki et al., 1994), it was necessary to carefully analyze the results of in vivo pharmacokinetic studies (Higaki et al., 1992) and isolated kidney perfusion studies (Higaki et al., 1994), because (R)-(+)-DBCA is predominantly metabolized to (R)-(+)-MBCA (Higaki and Nakano, 1992; Higaki et al., 1992, 1994) and (R)-(+)-MBCA efficiently inhibits the secretion of DBCA (Higaki et al., 1992, 1994). Because MBCA, a metabolite of DBCA, used in this study was not metabolized sequentially (Higaki et al., 1994), the kinetics of the MBCA enantiomers themselves in kidney could be investigated by performing isolated kidney perfusion studies.

A significant difference in the MTT_kidney values for inulin in the perfusion studies with (R)-(+)- and (S)-()-MBCA was observed (table 2). However, we judged that the difference in MTT_kidney values for inulin does not affect the kinetics of the MBCA enantiomers, because the MTT_kidney values for inulin are so small, compared with those of MBCA enantiomers, and the MTT_kidney value for (R)-(+)-MBCA is not smaller but larger than that for (S)-()-MBCA.

Our present study clarified the presence of R-(+)-predominant stereoselectivity in the process of uptake across the BLM in renal epithelial cells (table 5; fig. 6B), which suggests that each enantiomer of chiral organic and anionic drugs may have different pharmacokinetics. Particularly when enantiomers have different pharmacological and toxicological effects, it should be decided whether the drug is developed as an enantiomer or a racemate. If an enantiomer is selected, it should be decided which enantiomer is developed by considering stereoselective differences in renal secretion. In the case of DBCA, which has uricosuric, antihypertensive, and diuretic activities (Higaki and Nakano, 1992; Higaki et al., 1992), each enantiomer has a different pharmacological effect and both enantiomers and their metabolites, MBCA, inhibit the renal secretion and delay the elimination of one another (Higaki et al., 1992, 1994). Therefore, the racemate will be selected to maintain the pharmacological action for a longer time.

DBCA is the parent compound of MBCA; DBCA has been shown to be secreted in the proximal tubules of rats (Higaki et al., 1992), rabbits, dogs (Nakamura et al., 1990), and monkeys (Nakano and Kawahara, 1992), and probenecid inhibition of its secretion has been reported (Nakamura et al., 1990; Nakano et al., 1993). Moreover, the renal secretion of DBCA enantiomers is recognized to be inhibited by (R)-(+)-MBCA in isolated perfused kidneys (Higaki et al., 1994). Therefore, MBCA as well as DBCA would enter the epithelial cells across the BLM via the coupling system of the Na⁺-dicarboxylate cotransporter and the dicarboxylate-organic anion exchanger, as proposed by studies using PAH as a prototype of organic anions (Pritchard and Miller, 1991; Werner and Roch-Ramel, 1991; Makhuli et al., 1995), although additional studies are needed to clarify the details.

A significant difference in k₁/f_p ratios between MBCA enantiomers obviously indicates that the unbound molecules of the MBCA enantiomers are stereoselectively taken up across the BLM. However, it remains to be clarified whether the large V_d,cell value is dependent on active transport across the BLM into the cell or there is an intrinsically large capacity of distribution. In liver, ligandin, a subtype of glutathione-S-transferase, is thought to be important for the intracellular distribution and movement of organic anions, but its role is still unclear (Sugimoto et al., 1993; Elferink et al., 1995). No comparable evidence has been reported for kidney (Pritchard and Miller, 1993). We could not estimate the free fraction of MBCA enantiomers, particularly in the tubular epithelial cells. Therefore, additional studies may be needed for clarification of the intracellular distribution mechanism and secretion across the BBM.

_cell was reported to correspond to the mean time from uptake to secretion from epithelial cells (Hori et al., 1988). As shown in table 4, there was no significant difference in this parameter between (R)-(+)- and (S)-()-MBCA. _cell,sp, which is the mean time for molecules in the cytosol to exit across the BBM (Hori et al., 1991; He et al., 1991), was not significantly different between enantiomers (table 4). Moreover, the two mean time parameters were not affected by the presence of the antipode (fig. 6B). Therefore, the processes determining the values of _cell and _cell,sp were not thought to be stereoselective. For k₁ and k₁/f_p, significant differences between enantiomers (table 5) and decreases in the k₁ and k₁/f_p values for the S-()-enantiomer produced by the antipode were observed (fig. 6B). Nevertheless, _cell and _cell,sp were independent of the decrease in k₁ and were not changed, which may suggest that the process of uptake across the BLM is much more rapid and should not be reflected in _cell, as in the case of PAH (Saito et al., 1991). For the process of secretion across the BBM, there was neither a difference in the values for nor an interaction between the enantiomers, as shown by the values of k₂ (table 5; fig. 6B), from which the time taken for this process would not be thought to differ between MBCA enantiomers. These results suggest that the intracellular movement of MBCA might not be stereoselective.

The _cell value for (S)-()-MBCA was increased only in the presence of 200 µM (R)-(+)-MBCA (fig. 6B), which may be because the mean time for uptake across the BLM or for intracellular transport was exceptionally prolonged because of extensive inhibition by the high concentration of the antipode. Although Saito et al. (1991) reported that the increase in the _cell value for PAH in the presence of a high concentration of probenecid could be ascribed to inhibition of secretion across the BBM, our results showed that transport through the BBM should not be changed significantly (fig. 6B).

Several mechanisms have been proposed for transport through the BBM, and species differences have been reported for the transport systems (Pritchard and Miller, 1991; Ullrich, 1994). For rats, an antiport system with OH or Cl and a carrier-mediated transport system dependent on the membrane potential have been reported (Pritchard and Miller, 1991; Ullrich, 1994). However, this study does not offer positive evidence for the presence of carrier-mediated transport in the BBM, and transport through the BBM could be thought to be dependent on the intracellular concentrations of MBCA enantiomers, as in the case of PAH (Saito et al., 1991).

In conclusion, (R)-(+)-MBCA, rather than (S)-()-MBCA, was predominantly taken up across the BLM and distributed in epithelial cells. However, no stereoselective differences in the processes of intracellular movement and exit across the BBM were observed.