Glutathione S-transferases (GSTs) conjugate activated xenobiotics with glutathione; thus, GST induction may improve detoxification and excretion of potentially harmful compounds. Using a randomized cross-over design, we tested the hypothesis that, in humans, serum GST-alpha concentration (GST-alpha) and GST activity increase with vegetable consumption and that this effect is GSTM1 genotype dependent. Twenty-one men (10 GSTM1-null and 11 GSTM1+) and 22 women (15 GSTM1-null and 7 GSTM1+), nonsmokers, 20-40 years of age and not on medications, ate four 6-day controlled diets: basal (vegetable-free), and basal supplemented with three botanically defined groups of vegetables (i.e., brassica, allium, and apiaceous). Fasting blood samples, collected on the last 2 days of each feeding period, were analyzed for GST-alpha, serum GST activity [against 1-chloro-2,4-dinitrobenzene (CDNB) and 7-chloro-4-nitrobenzo-2-oxa-1,3-diazole (NBD-Cl)] and peripheral-lymphocyte GST-mu activity (against trans-stilbene oxide). The brassica, but not allium or apiaceous, vegetable diets (relative to the basal diet) increased GST-alpha by 26% (P = 0.005) and GST (NBD-Cl) activity by 7% (P = 0.02) in the GSTM1-null individuals, particularly the women. Apiaceous vegetable supplementation decreased GST-alpha in the GSTM1+ men (P = 0.03). Among the GSTM1+ women, both brassica and the allium diets increased GST-mu activity by 18% (P = 0.02) and 26% (P = 0.001), respectively. The vegetable diets had no effect on GST (CDNB) activity, irrespective of GSTM1 genotype or sex. These results demonstrate that GSTM1 genotype has a significant effect on GST responses to diet and that brassica vegetables are most effective at inducing GST-alpha, whereas both brassica and allium vegetables induce GST-mu. GST responses were more pronounced in women than men, but it is not clear from this study whether this is a dose-per-body-weight or a sex-specific effect.