Fruit and vegetable intake and UGT1A1 genotypes: Sex differences in UGT1A1 activity as measured by bilirubin.

B59

UDP-glucuronosyltransferase (UGT) 1A1 conjugates bilirubin, as well as compounds that have been associated with cancer risk, such as estradiol and 2-amino-1-methyl-6-phenylimidazo[4,5-f]-pyridine (PhIP), facilitating their metabolism and excretion. The UGT1A1*28 polymorphism results in lower promoter activity due to 7 thymine-adenine (TA) repeats compared to the more common 6 TA repeats. We showed previously, in an observational study using 3-day food records (Peterson et al., J Nutr, 2005), and in a controlled feeding study (in preparation), that serum bilirubin concentrations (a marker of UGT1A1 activity) were lower in individuals homozygous for the [TA]7-repeat allele (7/7) when exposed to fruits and vegetables, whereas no effect was seen in individuals with the 6/6 and 6/7 genotypes. Our objective here was to determine whether such genotype-diet interactions could be detected using food frequency questionnaires (FFQs) from 267 healthy, non-smoking men (n=135) and women (n=132), aged 20-40 y, that provided blood samples for genotyping and serum bilirubin measurements. Daily servings of total fruits and vegetables and groups of plant foods shown to increase UGT activity (cruciferous vegetables, citrus fruits, and soy foods) were analyzed in separate multivariate linear models to determine their association with natural logarithmic-transformed serum total and indirect (unconjugated) bilirubin concentrations. After adjusting for age and UGT1A1 genotype, mean total and indirect serum bilirubin levels were significantly lower among men that consumed 1.5 or more servings/d of total fruit and vegetables compared to those whose consumed less (p<.03 for both). There was a statistically significant interaction between UGT1A1*28 genotype and soy product intake; however, the effect differed by sex. Women with the 7/7 genotype that consumed 0.01-0.33 servings/d had higher mean total [1.9 mg/dL (1.4-2.5)] and indirect [1.7 mg/dL (1.2-2.3)] bilirubin concentrations compared to those that consumed none [0.9 mg/dL (0.6-1.2) for total and 0.8 mg/dL (0.6-1.1) for indirect bilirubin], after adjusting for age and total fruit and vegetable intake (p=.001 for both interactions). In contrast, men in this group had lower mean total [1.4 mg/dL (1.0-2.0) and indirect [1.2 mg/dL (0.8-1.8)] bilirubin concentrations compared to those that consumed none [2.1 mg/dL (1.6-2.7) for total and 1.9 mg/dL (1.5-2.5) for indirect bilirubin (p<.05 for both interactions)]. No effects were seen in individuals with the 6/6 and 6/7 genotypes. These results support the hypothesis that fruits and vegetables may affect UGT1A1 activity, particularly for individuals with the 7/7 genotype; however, there appear to be sex differences. In addition, FFQs may be adequately sensitive to detect interactions between diet and UGT1A1 genotype. This work was supported by NIH grants R01CA92288 and R25 CA94880.