Activity of the angiotensin converting enzyme (ACE) is usually measured in vitro by estimation of products cleaved by the enzyme from synthetic substrates. These substrates have affinities for ACE different from the natural substrate angiotensin I, and insensitive detection systems necessitate millimolar substrate concentrations while physiological angiotensin I concentrations are in the picomolar range. A new assay for ACE activity measurement was developed which reliably quantitates femtomoles of generated angiotensin II in plasma from angiotensin I added at a 17 pmol/mL concentration. The production of high affinity monoclonal antibodies against angiotensin II (Kd = 7 x 10(-11) mol/L) allowed a quantitative trapping (and thus protection from degrading enzymes) of angiotensin II generated during the incubation step and subsequent radioimmunoassay by simple dilution with labelled angiotensin II. Using 40 microL plasma, the detection limit was 20 fmol/mL/min. Normal human plasma has an ACE activity of 335 +/- 83 fmol/mL/min (mean +/- SD). Precision was characterized by coefficients of variation of less than or equal to 11% both within-assay and between-assays. Accuracy of the new method was established by comparing ACE activity with the ratio of plasma angiotensin II/angiotensin I in plasma obtained from normal volunteers 0.5 to 24 h after oral administration of 20 mg enalapril. The percentage of ACE inhibition indicated by both methods was almost identical (r = 0.93, n = 60, P less than .001). Since the latter ratio appears to reflect in vivo ACE activity, these results indicate that accurate measurement in vitro of ACE activity in vivo has been achieved.