These results suggest that Q141K SNPs may associate with a lower expression level, and S441N SNPs may affect both the expression level and cellular localization. It is possible that subjects with these polymorphisms may have lower expression level of BCRP protein and, consequently, a reduced ability to export these substrates.

Metabolic capacities for debrisoquin, sparteine, mephenytoin, nifedipine, and midazolam, which are substrates of polymorphic CYP2D6, CYP2C19, and CYP3A, have been reported to exhibit, in many cases, remarkable interindividual and ethnic differences. These ethnic differences are partly associated with genetic differences. In the case of the drug transporter ABCB1/MDR1, interindividual differences in its transporter activities toward various clinical drugs are also attributed to several ABCB1/MDR1 genetic polymorphisms. In this review, the existence and frequency of various low-activity alleles of drug metabolizing enzymes as well as populational drug metabolic capacities are compared among several different races or ethnicities. Distribution of nonsynonymous ABCB1/MDR1 SNPs and haplotype frequency in various races are summarized, with the association of nonsynonymous SNPs with large functional alterations as a rare event.

We performed comprehensive haplotyping of ABCB1/MDR1 gene blocks using 49 genetic polymorphisms, including seven novel ones, obtained from 145 Japanese subjects. The ABCB1/MDR1 gene was divided into four blocks (Blocks -1, 1, 2, and 3) based on linkage disequilibrium analysis of polymorphisms. Using an expectation-maximization based program, 1, 2, 8, and 3 haplotype groups (3, 12, 32, and 18 haplotypes) were identified in Blocks -1, 1, 2, and 3, respectively. Within Block 2, haplotype groups *1, *2, *4, *6, and *8 reported by Kim and colleagues (Clin Pharmacol Ther 2001; 70:189-199) were found, and additional three groups (*9 to *11) were newly defined. We analyzed the association of haplotypes with the renal clearance of irinotecan and its metabolites in 49 Japanese cancer patients given irinotecan intravenously. There was a significant association of the *2 haplotype in Block 2, which includes 1236C>T, 2677G>T and 3435C>T, with a reduced renal clearance of those compounds. Moreover, tendencies of reduced and increased renal clearance were also observed with *1f in Block 2 and *1b in Block 3, respectively. These findings suggest that the P-glycoprotein encoded by ABCB1/MDR1 in the proximal tubules plays a substantial role in renal exclusion of drugs and, moreover, that block-haplotyping is valuable for pharmacogenetic studies.

Human constitutive androstane (or active) receptor (hCAR), a member of the nuclear receptor superfamily NR1I3, regulates the expression of several genes that are mainly involved in the metabolism of endogenous and xenobiotic compounds (e.g., CYP2B6, CYP3A4, and UGT1A1). We found four novel splice variants in the ligand-binding domain (LBD) of hCAR (NCBI reference sequence, NM_005122; designated SV0 herein). The variants designated SV1 and SV2 contained in-frame 12-and 15-base pair (bp) insertions, respectively. SV3 carried both of the insertions, and SV4 contained an in-frame 117-bp deletion. The insertion site of SV1 is located in the ␣6 helix of hCAR LBD, which makes up the ligand-binding cavity, and that of SV2 is located in the highly conserved loop between helices ␣8 and ␣9. SYBR Green real-time reverse transcription-polymerase chain reaction analysis of each splice variant revealed that the hepatic expression of SV2 was almost comparable with that of SV0 (approximately 40%), whereas other variants accounted for 6 to 10% of the total hCAR transcripts. In the reporter gene assays employing the phenobarbital-responsible enhancer module (PBREM) from CYP2B6 and UGT1A1 genes, the splice variants, except for SV1, were inactive, whereas SV1 transactivated the CYP2B6 PBREM but not the UGT1A1 PBREM reporter. A nuclear translocation assay in rat hepatocytes revealed that all the splice variants lack the responsiveness toward phenobarbital and 6-(4-chlorophenyl)imidazo[2,1-b][1,3]thiazole-5-carbaldehyde O-(3,4-dichlorobenzyl)oxime (CITCO) in terms of the ligand-dependent nuclear translocation. Further characterization, such as the identification of specific ligands, will help elucidate physiological implication of these hCAR splice variants.

Three non-synonymous single nucleotide polymorphisms (SNPs) in the CYP3A4 gene were found in 34 cell lines derived from Japanese individuals. These three SNPs (T185S, L293P, and T363M)(1) have been previously reported, but little is known about the effect that these polymorphisms, especially T185S, have on catalytic activity. We measured testosterone hydroxylation in wild-type CYP3A4 and these three variants using a mammalian expression system. Testosterone 6beta-, 2beta-, and 15beta-hydroxylations by the variant CYP3A4 forms T363M (<40%) and T185S (<60%) were reduced as compared with the wild-type in transient expression assays. L293P was similar to the wild-type in testosterone 6beta- and 2beta-hydroxylase activities. Western blot analysis confirmed lower amounts of CYP3A4 protein in the T363M and T185S variants than in the wild-type. Interestingly, Northern blot analysis showed no significant difference among mRNA levels between the wild-type and variants. These results suggest that the T363M and T185S substitutions in CYP3A4 affect either protein expression or stability. These established cell lines provided useful CYP3A4 SNP information in the Japanese.