BACKGROUND: Prostate-specific antigen (PSA) is the best tumor marker for diagnosis and prognosis of prostatic carcinoma. The secretion of PSA from LNCaP human prostate cancer cells is influenced by acute stimuli such as vasoactive intestinal peptide (VIP), growth hormone-releasing hormone (GHRH), and chronic stimuli like androgens. METHODS: To study the regulation of basal and VIP/GHRH or androgen-stimulated secretion from LNCaP cells, we used a superfusion system, which allowed us to simultaneously measure PSA gene expression, PSA secretion, and cAMP release from the same cancer cells. LNCaP cancer cells were also implanted orthotopically into nude mice. RESULTS: VIP (30 pM-3 nM), GHRH (3 nM-300 nM), and dihydrotestosterone (100 nM) induced a significant increase in PSA gene expression, PSA secretion, and cAMP release. The dose and time-dependent effects of peptides were manifested only in the presence of androgens. At the end of continuous stimulation of cells with 1 nM VIP for 2 hr, large amounts of stored immunoreactive PSA still remained in the cells. Adenylate cyclase activator, forskolin (FSK), significantly increased PSA secretion and gene expression, and potassium, which causes nonspecific depolarization of membranes, augmented gene expression, and secretion of PSA, but did not influence cAMP release. This suggests that PSA secretion is regulated by cAMP-dependent as well as cAMP-independent pathways. In superfusion system, stimulatory effects of VIP and GHRH on PSA secretion were inhibited by VIP antagonist JV-1-53, and less by GHRH antagonist JV-1-38. In cell cultures, JV-1-38 had a stronger inhibitory effect on proliferation, indicating an involvement of the recently discovered tumoral GHRH receptors in this process. In nude mice, with orthotopically implanted LNCaP cancer cells, GHRH antagonist JV-1-38 alone or androgen ablation by castration had no effect on tumor growth and PSA levels. However, castration combined with treatment with GHRH antagonist, significantly decreased tumor growth and PSA secretion. CONCLUSIONS: Our findings suggest that the secretion of PSA is regulated rather than constitutive, contrary to previous reports. In addition, the effect of GHRH and VIP antagonists on PSA secretion from prostate cancer cells is not correlated with their antiproliferative action.