Antiandrogenic effects of novel androgen synthesis inhibitors on hormone-dependent prostate cancer.

We have found that in addition to being potent inhibitors of 17alpha-hydroxylase/C17,20-lyase and/or 5alpha-reductase, some of our novel androgen synthesis inhibitors also interact with the mutated androgen receptor (AR) expressed in LNCaP prostate cancer cells and the wild-type AR expressed in hormone-dependent prostatic carcinomas. The effects of these compounds on the proliferation of hormone-dependent human prostatic cancer cells were determined in vitro and in vivo. L-2 and L-10 are delta4-3-one-pregnane derivatives. L-35 and L-37 are delta5-3beta-ol-androstane derivatives, and L-36 and L-39 are delta4-3-one-androstane-derived compounds. L-2, L-10, and L-36 (L-36 at low concentrations) stimulated the growth of LNCaP cells, indicating that they were interacting agonistically with the mutated AR expressed in LNCaP cells. L-35, L-37, and L-39 acted as LNCaP AR antagonists. To determine whether the growth modulatory effects of our novel compounds were specific for the mutated LNCaP AR, competitive binding studies were performed with LNCaP cells and PC-3 cells stably transfected with the wild-type AR (designated PC-3AR). Regardless of AR receptor type, all of our novel compounds were effective at preventing binding of the synthetic androgen methyl-trienolone[17alpha-methyl-(3H)-R1881 to both the LNCaP AR and the wildtype AR. L-36, L-37, and L-39 (5.0 microM) prevented binding by >90%, whereas L-35 inhibited binding by 30%. To determine whether the compounds were acting as agonists or antagonists, LNCaP cells and PC-3AR cells were transfected with the pMAMneoLUC reporter gene. When luciferase activity was induced by dihydrotestosterone, all of the compounds were found to be potent inhibitors of transcriptional activity, and the pattern of inhibition was similar for both receptor types. However, L-2, L-10, and L-36 were determined to be AR agonists, and L-35, L-37, and L-39 were wild-type AR antagonists. When tested in vivo, L-39 was the only AR antagonist that proved to be effective at inhibiting the growth of LNCaP prostate tumor growth. L-39 slowed tumor growth rate in LNCaP tumors grown in male SCID mice to the same level as orchidectomy, significantly reduced tumor weights (P < 0.05), significantly lowered serum levels of prostate-specific antigen (P < 0.02), and significanty lowered serum levels of testosterone (P < 0.05). L-39 also proved to be effective when tested against the PC-82 prostate cancer xenograft that expresses wild-type AR. These results show that some of our compounds initially developed to be inhibitors of androgen synthesis also interact with the human AR and modulate the proliferation of hormone-dependent prostatic cancer cells. Therefore, compounds such as L-39, which have multifunctional activities, hold promise for the treatment of androgen-dependent prostate tumors.