The aim of the present study was to investigate whether a pathway for conversion of angiotensin I (ANG I) to angiotensin II (ANG II) other than that via angiotensin-converting enzyme (ACE) is present in the smooth muscle of the human detrusor. Isolated detrusor strips from 11 patients were contracted by ANG I (1 mu M) in the absence or presence of enalaprilat (10 mu M), soybean trypsin inhibitor (STI, 200 mu g/ml), or both. The metabolic activity in detrusor membranes from four patients was studied separately using Hip-Gly-Gly or ANG I as a substrate, with or without various protease inhibitors. The contractile response to ANG I (1 mu M) was depressed by enalaprilat from 66 +/- 22 (mean c SD) to 39 +/- 13% of the K+ (124 mM)-induced response (P < 0.01, n = 11), and the combination of enalaprilat and STI resulted in a further reduction in contractile amplitude to 25 +/- 14% (P < 0.01 vs. K+, and P < 0.05 vs. enalaprilat alone) and a significantly slower developing contraction with a time to peak of 3.7 +/- 1.7 vs. 1.1 +/- 0.3 min for ANG I alone (P < 0.01). In detrusor membranes, a low ACE activity, inhibitable by captopril, was demonstrated by the formation of hippuric acid (0.70 nmol.min(-1).mg protein(-1)) from the synthetic ACE substrate, Hip-Gly-Gly. However, the conversion of ANG I (166 nmol.min(-1).mg protein(-1)) to ANG II was not affected by ACE inhibition, while serine protease inhibitors, e.g., STI and chymostatin, completely prevented ANG II formation. These results indicate the presence of an alternative pathway for ANG II formation in the human detrusor, involving one or several serine protease(s).