Genotype-phenotype relations in glutathione-S-transferases and the role of vegetable consumption in a Dutch sigmoidoscopy-based population

The large intestine has a relatively low glutathione-S-transferase (GST) expression and high occurrence of neoplasia. Both genetic and environmental factors are thought to influence the functionality of the GST system. GST genotype-phenotype relations were investigated in 94 Dutch patients scheduled for sigmoidoscopy. Subjects were free of bowel inflammation and colorectal cancer. They kept a 3-d dietary record, ending at the time of endoscopy, and filled out a general questionnaire on other lifestyle factors. Functional polymorphisms in 4 GST isoforms were assessed in DNA isolated from blood, among them single-nucleotide polymorphisms in GSTP1 (A313G, resulting in an amino acid substitution) and in GSTA1 (C-69T, part of a functional haplotype). GST [pi] and -[alpha] isoenzyme levels were measured in rectal tissue by Western blotting. Total GST activity was measured spectrophotometrically in rectal tissue and white blood cells using 1-chloro-2,4-dinitrobenzene as a substrate. Results from all laboratory assays were normalized to protein content. Statistical analyses were adjusted for age and sex. Rectal GST activity showed a significant correlation with GST[pi] level (partial [r.sup.2] = 0.18). Activity differed between GSTP1 A313G genotypes (P < 0.001), the AG and GG genotype showing a mean of -30 and 60 nmol x [min.sup.-1] x mg [protein.sup.-1] lower GST activity, respectively. The same trend was seen for GST activity in lymphocytes but not in leukocytes. Consumption of allium vegetables was positively associated with rectal GST [pi] expression (P < 0.001). GST[alpha] level did not demonstrate any correlation with rectal GST activity or with GST activity in lymphocytes or leukocytes. GST[alpha] level differed among GSTA1 C-69T genotypes (P < 0.001), the CC genotypes showing the highest, the CT genotypes intermediate, and the TT genotypes no expression.