Lycopene, selenium, vitamin E and prostate cancer

Epidemiologic studies suggest that tomato and lycopene consumption and serum lycopene concentrations are inversely related to prostate cancer occurrence. The oldest, and possibly most cited potential mechanism of action, is that lycopene increases gap junction communication by increasing connexin 43 (Cx43) levels. To better characterize this mechanism, we utilized Cx43 +/+ and Cx43 -/- mouse embryonic fibroblasts (MEF) and two different treatment schedules (72-hour and 24-hour lycopene treatment). 72-hour treatment was more effective than the 24-hour treatment in decreasing cell growth, but only in Cx43 +/+ MEFs. Treatment with the gap junction communication inhibitor, 18β-glycyrrhetinic acid, failed to counteract this decrease in cell numbers. These results suggest that lycopene metabolic and/or oxidation products may be responsible for the decrease in cell growth, and this inhibition of growth may involve the gap junction-independent effects of Cx43. Despite the excitement surrounding lycopene, in our prostate cancer models lycopene alone has not significantly decreased prostate cancer development or progression. Therefore, we evaluated the effects of dietary lycopene (250 ppm), selenium (Se-methylselenocysteine, 2 ppm), and vitamin E (γ-tocopherol, 200 ppm) alone, and in combination, on the growth of androgen-dependent Dunning R3327-H rat prostate adenocarcinomas. AIN-93G diets containing these micronutrients were prefed for 4-6 weeks prior to tumor implantation by subcutaneous injection. The tumors were then allowed to grow for -18 weeks. Multiple linear regression analysis revealed that selenium consumption resulted in a highly significant decrease in final tumor areas and tumor weights, while lycopene and γ-tocopherol consumption did not significantly alter tumor growth. There were no significant interactions among nutrient combinations. Selenium consumption did not significantly alter serum testosterone or dihydrotestosterone levels, tumor proliferation, or apoptosis rates. Finally using Real-Time PCR we wanted to examine whether alterations in androgen gene (androgen receptor, probasin) and/or selenium-associated protein genes (selenium binding protein 2, selenoprotein 15, selenoprotein P) in the normal prostate or Dunning prostate tumors could explain selenium's decrease in tumor growth. Selenium consumption had no effect on gene expression but castration or finasteride-treatment did lead to expression alterations, suggesting that these genes are androgen-responsive.