ABCMdb

PMID: 16702730

Maekawa K, Itoda M, Sai K, Saito Y, Kaniwa N, Shirao K, Hamaguchi T, Kunitoh H, Yamamoto N, Tamura T, Minami H, Kubota K, Ohtsu A, Yoshida T, Saijo N, Kamatani N, Ozawa S, Sawada J
Genetic variation and haplotype structure of the ABC transporter gene ABCG2 in a Japanese population.
Drug Metab Pharmacokinet. 2006 Apr;21(2):109-21., [PubMed]

Sentences

No. Mutations Sentence Comment

10 ABCG2 p.Ser13Leu ABCG2 p.Gly354Arg In addition to 9 previously reported nonsynonymous single nucleotide polymorphisms (SNPs), 2 novel nonsynonymous SNPs, 38CÀT (Ser13Leu) and 1060GÀA (Gly354Arg), were found with minor allele frequencies of 0.3z. Login to comment 17 ABCG2 p.Arg482Thr ABCG2 p.Arg482Thr ABCG2 p.Arg482Gly ABCG2 p.Arg482Gly ABCG2 p.Gln141Lys ABCG2 p.Gln141Lys ABCG2 p.Gln141Lys ABCG2 p.Gln141Lys ABCG2 p.Gln141Lys ABCG2 p.Gln141Lys ABCG2 p.Val12Met ABCG2 p.Val12Met ABCG2 p.Val12Met ABCG2 p.Val12Met In vitro studies have also indicated that a number of anticancer drugs are good substrates for ABCG2: e.g. topotecan, an irinotecan metabolite, 7-ethyl-10-hydroxycamptothecin (SN-38), and its glucuronide conjugate, SN-38G.810) Indeed, inhibition of the murine ABCG2 homologue, Bcrp 1, increases the bioavailability of topotecan when orally administered to mdr1aW1b- decient mice.11) In a clinical study, coadministration of topotecan with GF120918, a dual inhibitor for ABCG2 and P-glycoprotein, was shown to markedly increase the bioavailability and systemic exposure of topotecan.12) The cloning of ABCG2 from drug-selected cell lines revealed that acquired amino acid substitutions at residue 482 (Arg482Gly and Arg482Thr) of ABCG2 resulted in marked alterations in substrate recognition and transport ability.13) Thereafter, naturally occurring genetic variations in ABCG2 have been extensively examined in various ethnic populations1421) because they were expected to explain interindividual dierences in oral bioavailability and clearance of ABCG2 substrate drugs.22) Two nonsynonymous polymorphisms, 34GÀA (Val12Met) and 421CÀA (Gln141Lys), were detected at relatively high frequencies in most ethnic groups including Caucasians, Asians, and Africans.1416,1821,23) Both polymorphisms were reported to be associated with reduced protein expression in vitro andWor the increased sensitivity of the expressed cells toward several anticancer drugs although conicting data were also reported.16,2426) The expression of ABCG2 protein in placenta was signicantly lower in homozygotes with the 421A alleles than in those with the 421C alleles, while 34GÀA (Val12Met) did not aect ABCG2 protein expression.23) However, in intestinal samples, no association was found between the ABCG2 protein levels and the 421CÀA (Gln141Lys) genotype.18) A pharmacokinetic study showed that 421A (Gln141Lys) was unlikely to inuence the in vivo disposition of irinotecan in European Caucasian cancer patients.27) On the other hand, diomotecan pharmacokinetics were signicantly aected by the 421A genotype.28) To explain these inconsistencies, the elucidation of the haplotype structure of ABCG2 would be helpful; however, only limited information is available for the linkage disequilibrium (LD) prole and haplotype structure of this gene.20,21) Also, to facilitate future pharmacogenetic studies on ABCG2 genetic variations, haplotype analysis using its high-density SNPs found in a large number of samples is warranted. Login to comment 70 ABCG2 p.Val12Met The observed allele frequencies were all in Hardy-Weinberg equilibrium (pÀ0.05) except for 34GÀA (Val12Met) and IVS2 {36AÀG (p0.028 for both variations). Login to comment 73 ABCG2 p.Ser13Leu ABCG2 p.Gly354Arg Two novel nonsynonymous SNPs, 38CÀT (Ser13Leu) and 1060GÀA (Gly354Arg), were heterozygous in dierent patients at an allele frequency of 0.003. Login to comment 80 ABCG2 p.Gln141Lys ABCG2 p.Val12Met However, marked ethnic dierences in the allele frequencies were observed with |1203ä |1200delCTCA, 34GÀA (Val12Met), 421CÀA (Gln141Lys), and some intronic variations. Login to comment 85 ABCG2 p.Gln141Lys ABCG2 p.Val12Met ABCG2 p.Ser441Asn ABCG2 p.Phe489Leu ABCG2 p.Phe431Leu 115Haplotype Structure in Human ABCG2 (from |1836 to |1175 bp upstream of the translational start site) of the basal promoter,30) and was suggested to inuence irinotecan pharmacokinetics.31) The frequencies of two well-known nonsynonymous SNPs, 34GÀA (Val12Met) and 421CÀA (Gln141Lys), were 0.192 and 0.319 in our study, which were comparable to those in Chinese (0.204 and 0.2220.350, respectively).20,27) However, the frequencies were much higher than those in Caucasians (0.020.065 and 0.080.15), African-Americans (00.09 and 00.05), and a Swedish population (0.02 and 0.1).18,19,21,23,27) Of other relatively rare nonsynonymous SNPs, 376CÀT (Gln126X), 1291TÀC (Phe431Leu), 1322GÀA (Ser441Asn), 1465TÀC (Phe489Leu), and 1515delC (Phe506SerfsX4) were already detected in a Japanese population by Itoda et al.17) andWor Kobayashi et al.,23) but not found in other ethnic groups. Login to comment 86 ABCG2 p.Arg160Gln ABCG2 p.Arg160Gln Recently, the two nonsynonymous SNPs, 479GÀA (Arg160Gln) and 1723CÀT (Arg575X), have been reported by Bosch et al.21) Arg160Gln and Arg575X were found as heterozygotes in single Asian and Caucasian subjects, respectively. Login to comment 87 ABCG2 p.Arg160Gln Therefore, the distribution of 479GÀA (Arg160Gln) might be restricted in Asians, while 1723CÀT (Arg575X) is likely to be detected both in Caucasians and Japanese at low frequencies. Login to comment 89 ABCG2 p.Ile206Leu ABCG2 p.Asn590Tyr ABCG2 p.Asp620Asn ABCG2 p.Pro269Ser On the other hand, several nonsynonymous SNPs reported in other ethnic groups were not detected: 805CÀT (Pro269Ser) found in Chinese at a 0.037 frequency,20) 1858GÀA (Asp620Asn) in undened (combined) ethnicities14) (0.011) and in a Dutch population21) (0.005), 616AÀC (Ile206Leu) in Hispanics (0.100), and 1768AÀT (Asn590Tyr) in Caucasians (0.010).18) Thus, these SNPs are either ethnic-specic or rare. Login to comment 98 ABCG2 p.Val12Met As previously reported in various ethnic groups,14,18) perfect LD (r2 1.0) was observed between 34GÀA (Val12Met) and IVS2{ 36AÀG. Login to comment 100 ABCG2 p.Ser441Asn Because the two rare nonsynonymous variations, 1515delC (F506SfsX4) and 1322GÀA (Ser441Asn), were found in the same patient, they were statistically estimated to be linked with each other. Login to comment 101 ABCG2 p.Gln141Lys Strong LDs (r2 À0.7) were observed between |262CÀT and 1098GÀA (Glu366Glu) (r2 0.798), between 421CÀA (Gln141Lys) and IVS12{49GÀT (r2 0.709), and among IVS11{20AÀG, IVS1321CÀT, and IVS15{110CÀT (r2 À0.720). Login to comment 115 ABCG2 p.Gln141Lys No close associations were observed (r2 º0.70) between the variations across the blocks except for one pair of SNPs, 421CÀA (Gln141Lys) and IVS12{49GÀT, Fig. . Login to comment 130 ABCG2 p.Gln141Lys ABCG2 p.Gln141Lys ABCG2 p.Val12Met ABCG2 p.Val12Met ABCG2 p.Arg160Gln ABCG2 p.Arg160Gln ABCG2 p.Ser13Leu ABCG2 p.Ser13Leu ABCG2 p.Gly354Arg ABCG2 p.Gly354Arg In Block 1, seven haplotype groups (*1 to *7) were inferred, and the groups of *2 to *7 harbored non-synonymous SNPs, 421CÀA (Gln141Lys) (*2), 34GÀA (Val12Met) (*3), 376CÀT (Gln126X) (*4), 38CÀT (Ser13Leu) (*5), 479GÀA (Arg160Gln) (*6), and 1060GÀA (Gly354Arg) (*7). Login to comment 134 ABCG2 p.Gln141Lys ABCG2 p.Val12Met The haplotype-tagging SNPs (htSNPs) that were able to resolve the 8 common haplotypes were the following 7 variations: IVS199GÀA, 34GÀA (Val12Met), IVS293TÀC, 376CÀT (Gln126X), 421CÀA (Gln141Lys), IVS6217AÀG, and IVS6204CÀT. Login to comment 141 ABCG2 p.Ser441Asn ABCG2 p.Phe489Leu ABCG2 p.Phe431Leu The thick lines represent the combinations with frequencies over 10z, and the thin lines represent the combinations with frequencies of 1.0 to 9.9z. 118 Keiko MAEKAWA et al. haplotype harboring nonsynonymous SNPs, 1465TÀC (Phe489Leu) (*2), 1291TÀC (Phe431Leu) (*3), 1322GÀA (Ser441Asn)W1515delC (Phe506SerfsX) (*4), and 1723CÀT (Arg575X) (*5), respectively. Login to comment 152 ABCG2 p.Ser13Leu ABCG2 p.Gly354Arg Eleven nonsynonymous variations, including two novel ones (Ser13Leu and Gly354Arg), were found. Login to comment 154 ABCG2 p.Gln141Lys ABCG2 p.Val12Met ABCG2 p.Val12Met Mizuarai et al. showed that 34GÀA (Val12Met) was associated with reduced drug resistance in polarized LLC-PK1 cells, which might be caused by its impaired apical membrane localization.25) In contrast, several groups did not nd any signicant eects of Val12Met on the protein expression levels as well as drug resistance using stable and transient mammalian expression systems.16,24,26) According to Imai et al. and Kondo et al.,16,24) the Gln141Lys substitution resulted in decreased protein expression and reduced drug resistance. Login to comment 155 ABCG2 p.Ser441Asn These results are inconsistent with those obtained by Mizuarai et al. and Morisaki et al., in which the reduced drug resistance was not caused by the decreased protein expression.25,26) Kondo et al. have shown that the Ser441Asn variant was not localized to apical membranes, but remains intracellular in the transfected LLC-PK1 cells,24) suggesting its reduced activity. Login to comment 158 ABCG2 p.Phe489Leu ABCG2 p.Phe431Leu ABCG2 p.Arg160Gln ABCG2 p.Ser13Leu ABCG2 p.Gly354Arg The functional eects of the other ve nonsynonymous SNPs (Ser13Leu, Arg160Gln, Gly354Arg, Phe431Leu, and Phe489Leu) have not yet been characterized. Login to comment 159 ABCG2 p.Arg160Gln ABCG2 p.Ser13Leu ABCG2 p.Gly354Arg Using the PolyPhen program (http:WW www.bork.embl-heidelberg.deWPolyPhenW) to predict the functional eect of these amino acid substitutions, three substitutions, Ser13Leu, Arg160Gln, and Gly354Arg, were estimated to cause possible alterations in protein function based on the PSIC (position specic independent count) prole score. Login to comment 160 ABCG2 p.Arg160Gln Notably, Arg160Gln is located in the functionally important ATP-binding region between the Waker A (amino acids 8089) and Walker B (amino acids 206210) motifs. Login to comment 162 ABCG2 p.Gln141Lys ABCG2 p.Val12Met Except for Val12Met, Gln126X, and Gln141Lys, the allele frequencies of eight nonsynonymous SNPs were less than 0.01, and these low frequency variations do not largely contribute to the overall heterozygosity of ABCG2; however, they might have clinical importance. Login to comment 174 ABCG2 p.Gln141Lys Nevertheless, the frequency (31.9z) of the haplotype *2a in Block 1 harboring 421CÀA (Gln141Lys) was comparable to their counterpart (20.4z). Login to comment 176 ABCG2 p.Gln141Lys Therefore, the substitution itself of Gln141 to Lys is likely to be responsible for the reduced expression of ABCG2 protein in placenta as demonstrated by Kobayashi et al.23) Furthermore, Chinese and Japanese share several common Block 2 haplotypes, *1a, *1b, *1c, and *1dW *1e. Login to comment 188 ABCG2 p.Val12Met ABCG2 p.Val12Met In a Swedish population, |1203ä |1200delCTCA was reported to be linked with 34GÀA (Val12Met), the representative SNP in the Block1 *3 group.19) Due to the high (0.54) and low (0.02) allele frequencies of |1203ä|1200delCTCA and 34GÀA (Val12Met), respectively, the Block |1 *1bBlock 1 *3 combination is not predominant in the Swedish population. Login to comment 189 ABCG2 p.Val12Met Zhou et al. suggested that |1203ä |1200delCTCA might inuence pharmacokinetic parameters of irinotecan.31) On the other hand, the functional signicance of 34GÀA (Val12Met) is not fully elucidated as described above.16,2426) In this context, the major combination in Japanese, Block |1 *1bBlock 1 *3a, should be carefully considered in pharmacogenetic studies in Japanese. Login to comment