Functional constraints on the constitutive androstane receptor inferred from human sequence variation and cross-species comparisons.

Members of the NR1I subfamily of nuclear receptors play a role in the transcriptional activation of genes involved in drug metabolism and transport. NR1I3, the constitutive androstane receptor (CAR), mediates the induction of several genes involved in drug response, including members of the CYP3A, CYP2B and UGT1A subfamilies. Large inter-individual variation in drug clearance has been reported for many drug metabolising enzyme genes. Sequence variation at the CAR locus could potentially contribute to variation in downstream targets, as well as to the substantial variation in expression level reported. We used a comparative genomics-based approach to select resequencing segments in 70 subjects from three populations. We identified 21 polymorphic sites, one of which results in an amino acid substitution. Our study reveals a common haplotype shared by all three populations which is remarkably similar to the ancestral sequence, confirming that CAR is under strong functional constraints. The level and pattern of sequence variation is approximately similar across populations, suggesting that interethnic differences in drug metabolism are not likely to be due to genetic variation at the CAR locus. We also identify several common non-coding variants that occur at highly conserved sites across four major branches of the mammalian phylogeny, suggesting that they may affect CAR expression and, ultimately, the activity of its downstream targets.

Haplotype structure of the UDP-glucuronosyltransferase 1A1 promoter in different ethnic groups.

Genetic variation in UDP-glucuronosyltransferase 1A1 (UGT1A1)expression has several important clinical implications. UGT1A1 basal transcription is affected by a polymorphic (TA)n repeat, and another important regulatory element is the phenobarbital-responsive enhancer module (PBREM) which might contain variants affecting inducible gene expression. We assessed the extent of linkage disequilibrium between the (TA)n polymorphism and variants in the PBREM and UGT1A1 promoter. We also investigated the relationship between PBREM-(TA)n haplotypes and the glucuronidation rate of the UGT1A1 substrate SN-38. DNAs from 83 human livers were genotyped for the (TA)n polymorphism and microsomes from the same livers were phenotyped for SN-38 glucuronidation. The (TA)n polymorphism was genotyped in 24 additional African-Americans included in the Human Variation Panel (Coriell Institute). A 606-bp region spanning the PBREM was sequenced in 81 liver and a subset of 22 Human Variation Panel DNAs and six variants were found. The -3279G T and -3156G A variants are common (0.39 and 0.30, respectively). -3279G T is more common in Caucasians than African-Americans (P = 0.001). In Caucasians, linkage disequilibrium was highly significant between sites -3279, -3156, and the (TA)n polymorphism (P < 0.0001). In contrast, in African-Americans, only marginal levels of significance were observed between (TA)n and -3279 (P = 0.02) and between -3279 and -3156 (P = 0.04). Ten promoter haplotypes were identified. Haplotype I is the most common (0.39), from which haplotype II (0.15) differs at position -3279. SN-38G formation rates were correlated with (TA)n genotypes. This study showed that (i) common promoter variants are in linkage disequilibrium and (ii) the haplotype structure of promoter is probably different between Caucasians and African-Americans.