Trigonella foenum graecum (fenugreek) is traditionally used to treat disorders such as diabetes, high cholesterol, wounds, inflammation, and gastrointestinal ailments. Recent studies suggest that fenugreek and its active constituents may possess anticarcinogenic potential. We evaluated the preventive efficacy of dietary fenugreek seed and its major steroidal saponin constituent, diosgenin, on azoxymethane-induced rat colon carcinogenesis during initiation and promotion stages. Preneoplastic colonic lesions or aberrant crypt foci (ACF) were chosen as end points. In addition, we assessed the mechanism of tumor growth inhibition of diosgenin in HT-29 human colon cancer cells. To evaluate the effect of the test agent during the initiation and postinitiation stages, 7-week-old male F344 rats were fed experimental diets containing 0% or 1% fenugreek seed powder (FSP) or 0.05% or 0.1% diosgenin for 1 week and were injected with azoxymethane (15 mg/kg body weight). Effects during the promotional stage were studied by feeding 1% FSP or 0.1% diosgenin 4 weeks after the azoxymethane injections. Rats were sacrificed 8 weeks after azoxymethane injection, and their colons were evaluated for ACF. We found that, by comparison with control, continuous feeding of 1% FSP and 0.05% and 0.1% diosgenin suppressed total colonic ACF up to 32%, 24%, and 42%, respectively (P < or = 0.001 to 0.0001). Dietary FSP at 1% and diosgenin at 0.1% fed only during the promotional stage also inhibited total ACF up to 33% (P < or = 0.001) and 39% (P < or = 0.0001), respectively. Importantly, continuous feeding of 1% FSP or 0.05% or 0.1% diosgenin reduced the number of multicrypt foci by 38%, 20%, and 36% by comparison with the control assay (P < or = 0.001). In addition, 1% FSP or 0.1% diosgenin fed during the promotional stage caused a significant reduction (P < or = 0.001) of multicrypt foci compared with control. Dietary diosgenin at 0.1% and 0.05% inhibited total colonic ACF and multicrypt foci formation in a dose-dependent manner. Results from the in vitro experiments indicated that diosgenin inhibits cell growth and induces apoptosis in the HT-29 human colon cancer cell line in a dose-dependent manner. Furthermore, diosgenin induced apoptosis in HT-29 cells at least in part by inhibition of bcl-2 and by induction of caspase-3 protein expression. On the basis of these findings, the fenugreek constituent diosgenin seems to have potential as a novel colon cancer preventive agent.