Regular Article Potential Cholestatic Activity of Various Therapeutic Agents Assessed by Bile Canalicular Membrane Vesicles Isolated from Rats and Humans

Drug Metab. Pharmacokin. 18 (1): 16–22 (2003).

Masato HORIKAWA1,YukioKATO1,2,CharlesA.TYSON3 and Yuichi SUGIYAMA1,2 1Graduate School of Pharmaceutical Sciences, University of Tokyo, Tokyo, Japan 2Core Research for Evolutional Science and Technology (CREST), Japan Science and Technology Corporation, Kawaguchi, Japan 3Toxicology Laboratory, SRI International, Menlo Park, California, USA

Summary: The active transport of solutes mediated by the bile salt export pump (BSEPWABCB11) and multidrug resistance associated protein-2 (MRP2WABCC2) are thought to involve bile acid-dependent and -independent bile formation, respectively. To evaluate the potential of therapeutic agents as inhibitors of such transporters on bile canalicular membranes, we examined the inhibition of the primary active transport of typical substrates by 15 drugs, clinically known to cause cholestasis in canalicular membrane vesicles. The inhibition by most of the compounds in rat canalicular membrane vesicles (CMVs) was minimal or observed at much higher concentrations than obtained in clinical situations. However, cloxacillin, cyclosporin A and midecamycin inhibited BSEP, and cyclosporin A and midecamycin inhibited MRP2 with an inhibition constant close to the clinical concentration. By comparing the inhibition potential between rat and human CMVs, the inhibition of BSEP- and MRP2-mediated transport by midecamycin and cyclosporin A was relatively similar whereas the inhibitory eŠect on BSEP-mediated transport by cloxacillin and glibenclamide was more marked in humans than in rats. These results sug- gest that the majority of cholestasis-inducing drugs have a minimal inhibitory eŠect on rat BSEP and MRP2 although species diŠerences in inhibitory potential should be considered, especially in the case of BSEP.

Key words: cholestasis; biliary excretion; MRP2; BSEP

as well as the active excretion of glutathione (GSH) Introduction mediated by canalicular multispeciˆc organic anion Continuous bile formation is an important function transporterWmultidrug resistance associated protein-2 of the mammalian liver. It is an osmotic secretory proc- (MRP2WABCC2). ess that is driven by the concentration gradient of bile Cholestasis may occur via interference with the salts and other biliary constituents across the bile unidirectional transport of biliary constituents across canaliculi.1) The transport of solutes from blood to bile the canalicular membrane because of hereditary or is driven by several active transport systems in the plas- acquired impairment as a secondary consequence of ma membranes of both the basolateral (sinusoidal) and structural damage to hepatocytes and the bile duct. apical (canalicular) surfaces of hepatocytes. The active Recently, hereditary defects of speciˆc transporter transport of solutes across the latter membrane is molecules in canaliculi have been identiˆed as playing a thought to be the rate-limiting step in bile formation roleinthepathogenesesofcertainformsofcholestasis. under physiological conditions.1) Strautnieks et al. have reported that progressive familial Bile secretion has been traditionally divided into two intrahepatic cholestasis-2 (PFIC-2) is caused by a lack components. One is the bile acid-dependent bile ‰ow of BSEP.2) It has also been reported that a hereditary which is related to the excretion of bile salts mediated by mutation of MRP2 leads to hyperbilirubinemia with the bile salt export pump (BSEPWABCB11) in canalicu- cholestasis known as Dubin-Johnson syndrome.3) li. The other is the bile acid-independent bile ‰ow, In addition to these hereditary forms, cholestasis involving the passive excretion of inorganic electrolytes induced by drugs and other xenobiotic substances is a

Received; August 20, 2002, Accepted; January 8, 2003 To whom correspondence should be addressed: Prof. Yuichi SUGIYAMA,Ph.D.,Graduate School of Pharmaceutical Sciences, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan. Tel. ＋81-3-5841-4770, Fax. ＋81-3-5841-4766, E-mail: sugiyama＠mol.f.u- tokyo.ac.jp

16 Potential of Various Therapeutic Agents as Inhibitors of BSEP and MRP2 17 common problem in clinical medicine. A great deal of (ALP) and gamma-glutamyl transpeptidase (g-GTP). eŠort has been devoted to investigate the mechanisms of The activity of ALP was 536±57 (CMVs) and 8.89± cholestasis, although the pathogenesis of drug-induced 0.61 nmolWminWmg protein (homogenate) for the rats cholestasis remains unknown. Recently, several (mean±SE of 5 preparations, enrichment: 60.3±6.4) cholestatic compounds have been found to interact with and 848±58 (CMVs) and 10.8±0.8 nmolWminWmg pro- the export of bile acids and some organic anions at the tein (homogenate) for the human (mean±SE of 3 canalicular ATP-dependent transporters including preparations, enrichment: 78.6±5.4). The activity of g- BSEP and MRP2. The immunosuppressant, cyclospo- GTP was 264±61 (CMVs) and 5.43±0.40 nmolWminW rin A, has been found to inhibit both BSEP and MRP2 mg protein (homogenate) for the rats (enrichment: 48.5 in vitro.4) Similar mechanisms involving transporters ±11.2) and 4196±532 (CMVs) and 119±7nmolWminW were recently postulated for rifamycin, rifampicin and mg protein (homogenate) for the human (enrichment: glibenclamide.5) Therefore, it would be desirable in drug 35.2±4.4). The activity of CMVs used in the present screening and evaluation processes to exclude any com- study was also checked by measuring the ATP-depen- pounds which exhibit cholestatic potential by inhibiting dent uptake of standard substrates, [3H]TCA and transport across the bile canalicular membrane. [3H]DNP-SG. ATP-dependent uptake was obtained by Isolated canalicular membrane vesicles (CMVs) have subtraction of the uptake in the presence of AMP been used to evaluate the transport systems on the (ATP-independent) from that in the presence of ATP. membranes in vitro. As far as certain series of The activities of CMVs used in the present study under compounds which are reported to induce cholestasis are control conditions were as follows. [3H]TCA (1 mM, concerned, it has not been systemically examined yet incubation time 1min at 379C); ATP-dependent: 121± whether or not transporter inhibition on bile canalicular 24, ATP-independent: 4.78±2.55 (rats), ATP-depen- membrane involves cholestasis. Therefore, in the dent: 14.4±1.6, ATP-independent: 5.02±1.53 mLWmin present study, examining the possible involvement of Wmg protein (human), respectively. [3H]DNP-SG transporter inhibition in drug-induced cholestasis, we (1 mM, incubation time 2 min at 379C); ATP-depen- have investigated the inhibitory eŠects of 15 com- dent: 98.2±12.7, ATP-independent: 3.55±1.36 (rats), pounds, which cause cholestasis in clinical situations, on ATP-dependent: 5.81±0.30, ATP-independent: 2.01± both BSEP and MRP2 using CMVs. In addition, we 0.28 mLWminWmg protein (human), respectively. A have also evaluated the possibility of species diŠerences major part of uptake clearance of CMVs in the presence in the inhibitory eŠects of several compounds between of ATP was ATP-dependent both in rats and human. rats and humans. UptakestudybyCMVs: The inhibition studies for BSEP and MRP2 using various organic anions were Materials and Methods performed as follows: amoxycillin, chlorpromazine, Materials: Amoxycillin, chlorpromazine, cloxacil- cloxacillin, colchicine, cyclosporin A, doxycycline, lin, colchicine, cyclosporin A, doxycycline, gliben- glibenclamide, ketoconazole, midecamycin, minocy- clamide, ketoconazole, midecamycin, minocycline, cline, quinidine, sulpiride, tetracycline, ticlopidine and quinidine, sulpiride, tetracycline, ticlopidine, ver- verapamil were dissolved in dimethyl sulfoxide and apamil, ATP, AMP, creatine phosphate, and creatine diluted with 10 mM Tris-250 mM sucrose buŠer (pH phosphokinase were purchased from Sigma (St.Louis, 7.4). The ˆnal concentration of dimethyl sulfoxide in MO). [3H]Taurocholate ([3H]TCA, 3.47 mCiWnmol) and each assay mixture was 0.3z except for cyclosporin A [3H]glutathione (50.0 mCiWnmol) were purchased from (0.5z). The inhibition study was performed as reported New England Nuclear (Boston, MA). [3H]S-(2,4- previously.7) dinitrophenyl)-glutathione (DNP-SG, 50.0 mCiWnmol) Data analysis: The data analysis was performed as was synthesized by the method described by Saxena and reported previously.8) Brie‰y, the inhibition constants 6) Henderson. All other chemicals were commercial (Ki) for the uptake of TCA and DNP-SG by CMVs were products of analytical grade. obtained by ˆtting data to the following equations: Isolation of CMVs from human and rat liver: The V V ＝1 (1＋I K )Eq(1) human liver sample (from a female Caucasian, aged 21, (＋I) W (－I) W u W i who died from head trauma) was provided by SRI Inter- where V(＋I) and V(－I) represent the ATP-dependent national (Menlo Park, CA) and its shipment to SRI was transport velocity in the presence or absence of an inhi- coordinated by the National Disease Research Inter- bitor, respectively, and Iu is the inhibitor concentration change (Philadelphia, PA). CMVs from one human and in the medium ( mM). This equation was derived based six male Sprague-Dawley rats (250¿320 g, Japan SLC, on the assumption of competitive inhibition and the fact Hamamatsu, Japan) were prepared as described previ- that the TCA and DNP-SG concentrations (1 mM) were 7) ously. The purity of the prepared CMVs was checked much lower than the Km values (12.6 and 20.6 mMfor by determining the activity of alkaline phosphatase TCA and 21.0 and 193 mM for DNP-SG in rats and 18 Masato HORIKAWA, et al. humans, respectively).7) The ˆtting was performed by an minocycline and midecamycin inhibited DNP-SG iterative nonlinear least-squares method with a MULTI uptake in a concentration- dependent manner (Fig. 1C). 9) program, to obtain estimates of Ki. The input data The Ki values obtained are shown in Table 1.TheKi were weighted as the reciprocal of the square of the values for inhibition of TCA uptake by cyclosporin A observed values. and glibenclamide were similar to the previously The clinically relevant concentrations of inhibitors published data (0.8 and 8.6 mM, respectively).17) were estimated from the following equation.10–12) When the relative ratio of the clinical concentration to the K value was calculated based on the maximum f ･I Ãf ･(I ＋F ･Dose･k Q )Eq(2) i u in u max a a W h unbound concentration of each compound in circulat- where fu represents the plasma unbound fraction of the ing plasma (fu･Imax ), the obtained ratios were less than inhibitor. Imax is the reported value for the maximum 0.05 (no inhibition) except for the inhibition of TCA concentration in the systemic circulation and Iin is the uptake by cloxacillin (Table 1). When the relative ratio maximum inhibitor concentration at the inlet to the of the clinical concentration to the Ki value was calculat- liver. Fa represents the absorbed fraction of the inhibi- ed based on the maximum unbound concentration at the tor, assumed to be 1. The doses of inhibitors ad- inlet to the liver (fu･Iin), most of ratios were still close to ministered were those given in publications. The term, 0.05 except for the inhibition of TCA uptake by clox- ka, is the absorption rate constant and we chose acillin, cyclosporin A and midecamycin and that of 0.1 min－1 taking the maximum gastric emptying time DNP-SG uptake by cyclosporin A and midecamycin

(10 min) into consideration. Qh represents the hepatic (Table 1). blood ‰ow rate in humans (1610 mLWmin).12) To obtain an insight into any species diŠerence in Statistical analysis: The results are shown as mean such inhibition potential, the compounds with a ratio ±SE for the number of determinations. Dunnett's post- less than 0.014 were selected and their eŠect on the hoc test was used to determine the signiˆcance of diŠer- uptake of TCA and DNP-SG in human CMVs was ences between the mean of two groups, with Pº0.05 as examined (Fig. 2). Here, the chosen concentration of the minimum levels of diŠerence. each compound was the Ki (half-inhibitory concentration) observed in rats. The TCA uptake in the presence Results of midecamycin and cyclosporin A was 51.3 and 29.0z Compounds selected in the present study: All the of the control, showing inhibition comparable to that compounds (except colchicine) have been reported to exhibited by rat CMVs, whereas that in the presence of produce clinical cholestasis. Cyclosporin A has been cloxacillin and glibenclamide was much lower than demonstrated to induce cholestasis in clinical situations, 50z, suggesting greater inhibition in human CMVs causing an increase in serum bilirubin and bile acids and (Fig. 2). Chlorpromazine, midecamycin and cyclospo- a decrease in plasma bromosulfophthalein clearance.13) rin A exhibited slight or minimal inhibition (80¿100z Chlorpromazine has been also shown to cause chole- of control) of DNP-SG uptake in human CMVs (Fig. 2). static jaundice with an incidence of 1¿2z in clinical Discussion situations.14) Amoxycillin, cloxacillin, doxycycline, glibenclamide, ketoconazole, midecamycin, minocy- Cholestasis caused by several therapeutic drugs has cline, quinidine, sulpiride, tetracycline, ticlopidine and been suggested to involve the inhibition of transporters verapamil have been reported to cause cholestatic localized on the bile canalicular membranes. Bohme et symptoms in clinical situations (at least 1¿15 case al. have reported that cyclosporin A inhibits the ATP- reports each) as shown by liver function tests andWor dependent transport of TCA and leukotriene C4 (LTC4) liver biopsies. As for colchicine, no cholestasis has been with Ki values of 0.2 and 3.4 mM, respectively, in rat described in clinical situations, although it reduces bile CMVs, and that the bile ‰ow rate is reduced to 50z of ‰ow with leakage of lactate dehydrogenase into bile in baseline after cyclosporin A injection (25 mgWkg) in rats and increases both serum glutamic acid oxaloacetic vivo in rats.4,33) Funk et al. have also demonstrated that transaminase (SGOT) and alkaline phosphatase (ALP) troglitazone, an insulin-sensitizing drug for type 2 non- in patients.15,16) insulin-dependent diabetes mellitus, and its major Inhibition of ATP-dependent uptake of TCA and metabolite, troglitazone sulfate, inhibit ATP-dependent

DNP-SG by CMVs: The inhibitory eŠects of several TCA transport with a Ki of 1.3 and 0.23 mM, respec- compounds on the ATP-dependent uptake of TCA and tively, in rat CMVs and i.v. injection of troglitazone DNP-SG by rat CMVs are shown in Fig. 1.TheTCA (25 mgWkg) increases plasma bile acid concentrations in uptake was inhibited by chlorpromazine, cloxacillin, vivo.17) Thus, at least some cholestatic agents inhibit the colchicine, doxycycline, glibenclamide, cyclosporin A, primary active transport mediated by BSEP andWor midecamycin, quinidine and verapamil (Fig. 1A, B), MRP2. This led us to systematically examine the inhibi- whereas chlorpromazine, colchicine, cyclosporin A, tion potential of 15 compounds, reported to exhibit the Potential of Various Therapeutic Agents as Inhibitors of BSEP and MRP2 19

Fig. 1. Inhibition of the primary active transport of TCA and DNP-SG by various therapeutic agents in rat CMVs‚ CMVs were incubated with 1 mM TCA (A and B, incubation time 1 min) or DNP-SG (C, incubation time 2 min) in the presence of the indicated concentration of each compound at 379C, respectively. The ATP-dependent uptake was calculated by subtracting the uptake in the absence of ATP from that in its presence and shown as normalized values with respect to control values. Each data point represents the mean±SE of 3 experiments from 6 preparations.

Table 1. Copmarison of the inhibition constants (Ki ) of therapeutic drugs for BSEP and MRP2 with their clinical unbound concentrations

a b c d Ki,BSEP Ki,MRP2 dose fu･Imax fu･Iin compounds route References ( mM) (mg) ( mM)

amoxycillin À1000 À1000 p.o. 500 11.2 71.9 18 chlorpromazine 506 258 p.o. 800 0.0245 4.22 18, 19 cloxacillin 74.8 À1000 i.v. 1000 43.2 20 colchicine 539 975 p.o. 0.5 0.0120 0.0711 21, 22 cyclosporin A 1.09 1.96 p.o. 245 0.0501 0.936 18, 23 doxycycline 530 À1000 p.o. 100 0.898 3.32 18, 24 glibenclamide 5.29 À100 p.o. 25 0.0121 0.0750 25, 26 ketoconazole À100 À100 p.o. 200 0.0601 0.294 18 midecamycin 154 140 p.o. 1200 1.98 79.8 27 minocycline À1000 400 p.o. 100 0.874 3.39 24 quinidine 542 À1000 p.o. 400 0.260 4.39 18, 28 sulpiride À1000 À1000 i.m. 200 26.4 29, 30 tetracycline À1000 À1000 p.o. 250 1.81 13.1 18 ticlopidine À100 À100 p.o. 250 0.0311 1.07 18 verapamil 92.5 À1000 p.o. 80 0.0213 0.720 31, 32 a Ki for BSEP was obtained from the inhibition of ATP-dependent uptake of TCA by rat CMVs. b Ki for MRP2 was obtained from the inhibition of ATP-dependent uptake of DNP-SG by rat CMVs. c Maximum inhibitor concentration in circulating plasma. d Maximum inhibitor concentration at the inlet to the liver was calculated from Eq (2). 20 Masato HORIKAWA, et al.

Fig. 2. Inhibition of the primary active transport of TCA and DNP-SG by therapeutic agents in human CMVs‚ CMVs were incubated with 1 mM TCA (A, incubation time 1 min) or DNP-SG (B, incubation time 2 min) in the presence of each compound at a concentration equal to the Ki values obtained in rat CMVs. The ATP-dependent uptake was calculated by subtracting the uptake in the absence of ATP from that in its presence. Each data point represents the mean±SE of 3 experiments from 1 preparation.

cholestatic activity, on BSEP andWor MRP2 in present activities, enrichment factors, sideness and yields of study. CMVs were basically similar between rats and humans.7) Among the compounds examined, chlorpromazine, Because these values in the present studies were almost cloxacillin, colchicine, cyclosporin A, doxycycline, identical to previous data,7) the species diŠerence found glibenclamide, midecamycin, quinidine and verapamil in the present study was considered to be an intrinsic inhibited BSEP with a Ki of 1.09–542 mM, whereas diŠerence and independent of characterization. We also chlorpromazine, colchicine, cyclosporin A, midecamy- reported a diŠerence in the Km for the primary active cin and minocycline inhibited MRP2 with a Ki of transport of organic anions and TCA between human 1.96–975 mM(Table 1). All the other compounds only and rat CMVs,7) supporting a possible species diŠerence minimally inhibit both transporters (Table 1). In addi- in the a‹nity of primary active transport systems. It has tion, by comparing the obtained Ki with fu･Imax andWor been shown that the homologies in the amino acid se- fu･Iin, for most of the compounds, except cloxacillin quences of BSEP and MRP2 between rats and humans for BSEP, inhibition of the transporters seems to be are about 82 and 78z, respectively.34,35) The homologies observed at much higher concentrations than those seen of active sites, which interact with inhibitors, are still in clinical situations. Thus, despite their cholestatic unknown, however, and it is possible that these diŠer- activity, most of the therapeutic drugs produce only ences in amino acid sequence may result in a large diŠer- minimal inhibition of BSEP and MRP2, implying that ence in transporter inhibition. As for cloxacillin and the inhibition by the metabolites of parent compounds, glibenclamide, both compounds were reported to as in case of troglitazone, or other mechanisms for their increase serum bilirubin and alkaline phosphatase, cholestatic activity need to be considered. leading to clinical jaundice. In addition, liver biopsy To discuss in more quantitative terms the relationship specimens from jaundiced patients treated with these between the inhibition of these transporters and the compounds revealed the cholestatic ability of these observed cholestasis, further information is needed. In agents.36,37) Needless to say, to assess the degree of inhi- particular, the inhibitory activity of these therapeutic bition potential using human CMVs more precisely, a drugs may exhibit a species diŠerence between rats and comparison between the Ki values and clinically relevant humans. As a ˆrst step to obtain such information, the concentrations is needed. Therefore, to examine the inhibition in human CMVs was also examined for the possible involvement of BSEP inhibition in their compounds showing inhibitory eŠects in rat CMVs. As cholestatic activity, further quantitative analysis should far as the inhibition of DNP-SG transport was con- be performed. cerned, the inhibitory activity of the three compounds Considering that BSEP and MRP2 are localized on examined was relatively weak in humans, whereas the bile canalicular membranes, the estimation of the inhi- inhibition of TCA transport by cloxacillin and gliben- bition of these transporters should be performed using clamide was more marked in humans than in rats the hepatic unbound concentration rather than the plas- (Fig. 2), suggesting a possible species diŠerence for ma unbound concentration. Therefore, the concentra- BSEP. We previously reported that the marker enzyme tive uptake of these therapeutic drugs in human hepato- Potential of Various Therapeutic Agents as Inhibitors of BSEP and MRP2 21 cytes should also be examined to improve the estimation 7)Niinuma,K.,Kato,Y.,Suzuki,H.,Tyson,C.A., of inhibition. Stieger et al. have reported that the Weizer,V.,Dabbs,J.E.,Froehlich,R.,Green,C.E. cholestatic estrogen-metabolite, estradiol-17 b- and Sugiyama, Y.: Primary active transport of organic glucuronide, inhibits BSEP function in CMVs from anions on bile canalicular membrane in humans. Am. J. Physiol., 276: G1153–1164 (1999). normal rat liver and BSEPWMRP2-coexpressing vesicles, but not in CMVs from MRP2-deˆcient GY TR- 8) Horikawa, M., Kato, Y. and Sugiyama, Y.: The poten- W tial for an interaction between MRP2 (ABCC2) and rats or vesicles expressing only BSEP,5) suggesting that various therapeutic agents: Probenecid as a candidate estradiol-17 b-glucuronide exhibits trans-inhibition of inhibitor of the biliary excretion of irinotecan metabo- BSEP after its secretion into bile canaliculi by MRP2. lites. Drug Metab. Pharmacokin., 17: 23–33 (2002). This type of inhibition should also be considered in 9) Yamaoka, Y., Tanigawara, Y., Nakagawa, Y. and Uno, addition to cis-inhibition for a better understanding of T.: A pharmacokinetic analysis program (MULTI) for the detailed mechanism governing transporter inhibi- microcomputer. J. Pharmacobio.-Dyn., 4: 879–885 tion. (1981). In conclusion, here we have examined the inhibition 10) Ito,K.,Iwatsubo,T.,Kanamitsu,S.,Nakajima,Y.and potential of a series of therapeutic drugs, producing Sugiyama, Y.: Quantitative prediction of in vivo drug clinical cholestasis, on BSEP and MRP2. Although clearance and drug interactions from in vitro data on metabolism, together with binding and transport. Annu most of the drugs have only a minimal inhibitory eŠect Rev Pharmacol Toxicol., 38: 461–499 (1998). on rat BSEP and MRP2, inhibition of BSEP at clinical- 11) Komatsu, K., Ito, K., Nakajima, Y., Kanamitsu, S., ly relevant concentrations of cloxacillin as well as its Imaoka,S.,Funae,Y.,Green,C.E.,Tyson,C.A., potent inhibition in human CMVs suggests the possible Shimada, N. and Sugiyama, Y.: Prediction of in vivo involvement of BSEP inhibition in cholestasis. Further drug-drug interactions between tolbutamide and various quantitative analysis assessing the inhibition of human sulfonamides in humans based on in vitro experiments. transporters is needed to clarify the importance of trans- Drug Metab Dispos., 28: 475–81 (2000). porter inhibition in the cholestatic activity of other ther- 12) Ito, K., Iwatsubo, T., Kanamitsu, S., Ueda, K., Suzuki, apeutic compounds, including cyclosporin A, gliben- H. and Sugiyama, Y.: Prediction of pharmacokinetic clamide, and midecamycin. alterations caused by drug-drug interactions: Metabolic interaction in the liver. Pharmacol. Rev., 50: 387–411 References (1998). 13) Cadranel, J. F., Erlinger, S., Desruenne, M., Luciani, 1) Trauner,M.,Meier,P.J.andBoyer,J.L.:Molecular J., Lunel, F., Grippon, P., Cabrol, A. and Opolon, P.: pathogenesis of cholestasis. New Engl. J. Med., 339: Chronic administration of cyclosporin A induces a 1217–1227 (1998). decrease in hepatic excretory function in man. Dig. Dis. 2) Strautnieks, S. S., Bull, L. N., Knisely, A. S., Kocoshis, Sci., 37: 1473–1476 (1992). S.A.,Dahl,N.,Arnell,H.,Sokal,E.,Dahan,K., 14) Watson, R. P. G., Olomu, A., Clements, D., Waring, R. Childs, S., Ling, V., Tanner, M. S., Kagalwalla, A. F., H., Mitchell, S. and Elias, E.: A proposed mechanism Nemeth, A., Pawlowska, J., Baker, A., Mieli-Vergani, for chlorpromazine jaundice-defective hepatic sulphoxi- D.,Freimer,N.B.,Gardiner,R.M.andThompson,R. dation combined with rapid hydroxylation. J. Hepatol. J.: A gene encoding a liver-speciˆc ABC transporter is 7: 72–78 (1988). mutated in progressive familial intrahepatic cholestasis. 15) Crocenzi,F.A.,Sisti,A.,Pellegrino,J.M.,andRoma, Nat. Genet., 20: 233–238 (1998). M. G.: Role of bile salts in colchicine-induced hepato- 3) Paulusma,C.C.,Kool,M.,Bosma,P.J.,ScheŠer,G. toxicity. Implication for hepatocellular integrity and L.,terBorg,F.,Scheper,R.J.,Tytgat,G.N.,Borst,P., function. Toxicology, 121: 127–142 (1997). Baas,F.andElferink,R.P.:Amutationinthehuman 16) Bruns, B. J.: Colchicine toxicity. Australasian annals of canalicular multispeciˆc organic anion transporter gene Medicine., 17: 341 (1969). causes the Dubin-Johnson syndrome. Hepatology, 25: 17) Funk, C., Ponelle, C., Scheuermann, G. and Pantze, 1539–1542 (1997). M.: Cholestatic potential of troglitazone as a possible 4) Bohme,M.,Buchler,M.,Muller,M.andKeppler,D.: factor contributing to troglitazone-induced hepatotoxici- DiŠerential inhibition by cyclosporins of primary-active ty: In vivo and in vitro interaction at the canalicular bile ATP-dependent transporters in the hepatocyte canalicu- salt export pump (Bsep) in the rat. Mol. Pharmacol., 59: lar membrane. FEBS lett., 333: 193–196 (1994). 627–635 (2000). 5) Stieger, B., Fattinger, K., Madom, J., Kullak-Ublick, A. 18) Hardman,J.G.,Limbird,L.E.,MolinoŠ,P.B.and and Meier, P. J.: Drug- and estrogen-induced cholestasis Rudden,R.W.:Goodman & Gilman's The Pharmaco- through inhibition of the hepatocellular bile salt export logical basis of Therapeutics 10th edition, McGraw-Hill, pump (Bsep) of rat liver. Gastroenterology, 118: 422–430 Health Professions, New York, 2001, pp 1924–2023. (2000). 19) Rivera-Calimlim, L., Griesbach, P. H. and Perlmutter, 6) Saxena, M. and Henderson, G. B.: ATP-dependent R.: Plasma chlorpromazine concentrations in children eŒux of 2,4-dinitrophenyl-S-glutathione. J. Biol. with behavioral disorders and mental illness. Clin. Phar- Chem., 270: 5312–5319 (1995). macol. Ther., 26: 114–121 (1979). 22 Masato HORIKAWA, et al.

20) Vinge, E., Nergelius, G., Nisson, L. G. and Lidgrem, L.: 30) Bressolle, F., Bres, J. and Mourad, G.: Pharmacokinet- Pharmacokinetics of cloxacillin in patient undergoing ics of sulpiride after intravenous administration in hip or knee replacement. Eur. J. Clin. Pharmacol. 52: patientwithimpairedrenalfunction.Clin. Pharmacoki- 407–411 (1997). net., 17: 367–373 (1989). 21) Trnavska, Z., Kuchar, M., Rejholec, V. and Trnavsky, 31) Eldon, M. A., Battle, M. M., Voigtman, R. E. and K.: The binding of cholchicine and its derivatives to Colburn, W. A.: DiŠerences in oral verapamil absorp- bovine and human serum albmine and the human tion as a function of time of day. J. Clin. Pharmacol., plasma. Pharmacology, 18: 123–127 (1979). 29: 989–993 (1989). 22) Thomas, G., Girre, C., Scherrmabnn, P. and Steimer, J. 32) Abernethy, D. R., Wainer, I. W., Longstreth, J. A. and L.: Zero-order absorption and linear disposition of oral Andrawis, N. S.: Stereoselective verapamil disposition colchicine in healthy volunteers. Eur. J. Clin. Phar- and dynamics in aging during racemic verapamil macol., 37: 79–84 (1989). administration. J. Pharm. Exp. Ther., 266: 904–911 23) Lindholm, A.: Factors in‰uencing the pharmacokinetics (1993). of cyclosporin in man. Ther. Drug Monit., 13: 465–477 33) Bohme, M., Muller, M., Leier, I., Jedlitschky, G. and (1991). Keppler, D.: Cholestasis caused by inhibition of the 24) Saivin, S. and Houin, G.: Clinical pharmacokinetics of adenosine triphosphate-dependent bile salt transport in doxycycline and minocycline. Clin. Pharmacokinet., 15: rat liver. Gastroenterology, 107: 255–265 (1994). 355–366 (1988). 34) Taniguchi, K., Wada, M., Kohno, K., Nakamura, T., 25) Olsen, K. M., Kearns, G. L. and Kemp, S. F.: Glyburide Kawabe, T., Kawakami, M., Kagotani, K., Okumura, protein binding and the eŠect of albumin glycation in K.,Akiyama,S.andKuwano,M.Ahumancanalicular children, young adults, and older adults with diabetes. J. multispeciˆc organic anion transporter (cMOAT) gene is Clin. Pharmacl., 35: 739–745 (1995). overexpressed in cisplatin-resistant human cancer cell 26) Leif, G., Per-Henrik, G., Svante, S., Carola, S., Karl- lines with decreased drug accumulation. Cancer Res., 56: Johan, T., Frej, F. and Arne, M.: Comparison of phar- 4124–4129 (1996). macokinetics, metabolic eŠect and mechanisms of action 35) Strautnieks, S. S., Bull, L. N., Knisely, A. S., Kocoshis, of glyburide and glipizide during long-term treatment. S.A.,Dahl,N.,Arnell,H.,Sokal,E.,Dahan,K., Diabetes Care., 10: 671–678 (1987). Childs,S.,Ling,V.,Tanner,M.S.,Kagalwalla,A.F., 27) Periti, P., Mazzei, T., Mini, M. and Novelli, A.: Clinical Nemeth, A., Pawlowska, J., Baker, A., Mieli-Vergani, pharmacokinetic properties of the macrolide an- D.,Freimer,N.B.,Gardiner,R.M.andThompson,R. tibiotics—EŠect of age and various pathophysiological J.: A gene encoding a liver-speciˆc ABC transporter is states (Part I). Clin. Pharmacokinet., 16: 193–214 mutated in progressive familial intrahepatic cholestasis. (1989). Nat. Genet., 20: 233–238 (1998). 28) Knapp, V. W. and Lechleitner, P.: Relative bioavailabil- 36) Goland,S.,Malnick,S.D.H.,Gratz,R.,Feldberg,E., ity and in‰uence of a quinidine retard-formulation on Geltner, D. and Sthoeger Z. M.: Severe cholestatic blood-pressure and ECG in healthy volunteers (author's hepatitis following cloxacillin treatment. Postgrad. Med. trials). Arzneimittelforschung, 31: 1482–1486 (1981). J., 74: 59–60 (1998). 29) Bressolle,F.,Bres,J.,Blanchin,M.D.andGomeni,R.: 37) Carratu, R., Parisi, P. and Frullone, S.: Glibenclamide- Sulpiride pharmacokinetics in humans after intramuscu- associated reversible cholestasis. Eur. J. Med., 1: lar administration at three dose levels. J. Pharm. Sci., 441–443 (1992). 73: 1128–1136 (1984).