Genistein treatment duration effects biomarkers of cell motility in human prostate

BACKGROUND: Long term dietary consumption of genistein by Chinese men is associated with decreased PCa metastasis and mortality. Short term treatment of US men with prostate cancer (PCa) with genistein decreases MMP-2 in prostate tissue. MEK4 regulates MMP-2 expression, drives PCa metastasis, and genistein inhibits MEK4, decreases MMP-2 expression and dietary dosing inhibits human PCa metastasis in mice. This study examines short- versus long-term treatment effects of genistein in humans and in vitro. METHODS AND FINDINGS: US men with localized PCa were treated on a phase II trial with genistein (N = 14) versus not (N = 14) for one month prior to radical prostatectomy. Prostate epithelial cells were removed from fresh frozen tissue by laser capture microdissection, and the expression of 12,000 genes profiled. Genistein significantly altered the expression of four genes, three had established links to cancer cell motility and metastasis. Of these three, one was a non-coding transcript, and the other two were BASP1 and HCF2. Genistein increased BASP1 expression in humans, and its engineered over expression and knockdown demonstrated that it suppressed cell invasion in all six human prostate cell lines examined. Genistein decreased HCF2 expression in humans, and it was shown to increase cell invasion in all cell lines examined. The expression of MMP-2, MEK4 and BASP1 was then measured in formalin fixed prostate tissue from N = 38 Chinese men living in China and N = 41 US men living in the US, both cohorts with localized PCa. MMP-2 was 52% higher in Chinese compared to US tissue (P < 0.0001), MEK4 was 48% lower (P < 0.0001), and BASP1 was unaltered. Treatment of PC3 human PCa cells in vitro for up to 8 weeks demonstrated that short term genistein treatment decreased MMP-2, while long term treatment increased it, both changes being significant (P<0.05) compared to untreated control cells. Long term genistein-treated cells retained their responsiveness to genistein’s anti-motility effect. CONCLUSIONS: Genistein inhibits pathways in human prostate that drive transformation to a lethal high motility phenotype. Long term treatment induces compensatory changes in biomarkers of efficacy. The current strategy of using such biomarkers after short term intervention as go/no-go determinants in early phase chemoprevention trials should be carefully examined.