Data from epidemiological studies has suggested that vitamin D deficiency may promote prostate cancer, although the mechanism is not understood. We have previously demonstrated the presence of vitamin D receptors (VDR) in three human prostate carcinoma cell lines (LNCaP, PC-3, and DU-145) as well as in primary cultures of stromal and epithelial cells derived from normal and malignant prostate tissues. We have also shown that 1,25-dihydroxyvitamin D3 [1,25-(OH)2D3] can elicit an antiproliferative action in these cells. In the present study we compared the biological actions of 1,25-(OH)2D3 to those of a series of natural vitamin D3 metabolites and several synthetic analogs of vitamin D3 known to exhibit less hypercalcemic activity in vivo. In ligand binding competition experiments, we demonstrated the following order of potency in displacing [3H]1,25-(OH)2D3 from VDR: EB-1089 > 1,25-(OH)2D3 > MC-903 > 1,24,25-(OH)3D3 > 22-oxacalcitriol (OCT) > 1 alpha,25-dihydroxy-16-enecholecalciferol (Ro24-2637) > 25-hydroxyvitamin D3, with EB-1089 being approximately 2-fold more potent than the native hormone. No competitive activity was found for 25-hydroxy-16,23-diene-cholecalciferol. When compared for ability to inhibit proliferation of LNCaP cells, MC-903, EB-1089, OCT, and Ro24-2637 exhibited 4-, 3-, and 2-fold greater inhibitory activity than 1,25-(OH)2D3. Interestingly, although OCT and Ro24-2637 exhibit, respectively, 10 and 14 times lower affinity for VDR than 1,25-(OH)2D3, both compounds inhibited the proliferation of LNCaP cells with a potency greater than that of the native hormone. The relative potency of vitamin D3 metabolites and analogs to inhibit cell proliferation correlated well with the ability of these compounds to stimulate prostate-specific antigen secretion by LNCaP cells as well as with their potency to induce the 25-hydroxyvitamin D3-24-hydroxylase messenger RNA transcript in PC-3 cells. In conclusion, these results demonstrate that synthetic analogs of vitamin D3, known to exhibit reduced calcemic activity, can elicit antiproliferative effects and other biological actions in LNCaP and PC-3 cell lines. It is noteworthy that although binding to VDR is critical for 1,25-(OH)2D3 action, the analog data indicate that additional factors significantly contribute to the magnitude of the biological response. Finally, the strong antiproliferative effects of several synthetic analogs known to exhibit less calcemic activity than 1,25-(OH)2D3 suggest that these compounds potentially may be useful as an additional therapeutic option for the treatment of prostate cancer.