Inactive renin was partially purified from 4.5 liters of human plasma (502-fold, specific activity 0.8 X 10(-3) Goldblatt units/mg protein) and from 207 g renal cortex (103-fold, 52 X 10(-3) Goldblatt units/mg). In contrast to active renin, inactive renin from each source bound to Cibacron blue-agarose and was unable to bind to pepstatin-Sepharose. Both plasma and renal inactive renin had weaker affinity for anion-exchange resins than the active form, both bound to concanavalin A-Sepharose and were eluted with carbohydrate, and both bound tightly to hydrophobic gels. Each substance could be isolated in a completely inactive form during small-scale pilot studies, but "spontaneous" activation did occur, to a limited degree, during large-scale purification; this was possibly due to a plasma serine protease that fractionated with inactive renin during the initial purification steps. Both plasma and renal inactive renin were activated irreversibly by trypsin. Following activation, each substance lost it ability to bind to Cibacron blue-agarose. Each could be activated fully by acidification at 4 degrees C, but this activation was reversed during subsequent incubation at higher temperature and pH. There was no evidence of acid protease activity in either preparation. Activated inactive renin from both plasma and kidney were identical to partially-purified active renal renin in terms of pH optimum (pH 5.5-6.0) and reaction kinetics (Km 0.8-1.3 microM) with homologous angiotensinogen, noncompetitive inhibition by pepstatin (ki 2.5-3.5 microM), and an identical inhibition profile by monospecific antirenin antibodies. These results suggest that inactive renin from plasma and kidney may be the same substance and that their activated forms are similar to the endogenously produced active enzyme, consistent with the possibility that inactive renin is a precursor of circulating active renin.