Genetic polymorphisms in uridine diphospho-glucuronosyltransferase 1A1 andassociation with breast cancer among African Americans

We examined the role of constitutional genetic variation at the UDP-glucuro nosyltransferase (UGT) 1A1 locus in breast cancer susceptibility. The UGT1A 1 enzyme is a major UGT involved in estradiol glucuronidation. To date, fou r UGT1A1 variant alleles characterized by a variation in the number of TA f rom five through eight repeats in the atypical TATA box region have been de scribed in the African-American population. Functional analyses of the tran scriptional activity in breast and Liver cells revealed that the transcript ion activation of a reporter gene is inversely correlated with the number o f repeats. Reverse transcription-PCR analysis confirmed the expression of U GT1A1 in human liver in the hepato-carcinoma cell line HepG2 and provided e vidence of the expression of UGT1A1 in breast cancer tissue, where a positi ve signal was observed in 11 of 12 breast cancer cell lines tested. The pop ulation-based case-control study involved 200 women with breast cancer and 200 female controls of African ancestry, We postulated that breast cancer c ases might have a higher prevalence of low activity allele-containing genot ypes than controls (alleles presenting seven and eight repeats in the A(TA) nTAA motif of the TATA box). The age-adjusted odds ratio (OR) for breast ca ncer comparing women with seven and eight allele-containing genotypes versu s 5/5, 5/6, and 6/6 genotypes was 1.8 [95% confidence interval (CI), 1.0-3. 1; P = 0.06] in premenopausal women and 1.0 (95% CI, 0.5-1.7; P = 0.9) in p ostmenopausal women. The observed 1.8-fold elevated risk in premenopausal w omen with invasive breast cancer is highly suggestive of a possible Interac tion between UGT genotype and hormones. Additional analyses suggested a str onger association of UGT1A1 genotype with estrogen receptor (ER)-negative b reast cancer. Among premenopausal women, the association was stronger for E R- breast cancer (OR, 2.1; 95% Cl, 1.0-4.2; P = 0.04) than ER+ breast cance r (OR, 1.3; 95% Cl, 0.6-3.0; P = 0.5). The OR was slightly stronger among w omen who used oral contraceptives, and the association remained null in pos tmenopausal women, regardless of whether they took hormone replacement ther apy. Our current findings suggest that further investigations are warranted to elucidate the role of UGT1A1 in breast cancer risk.