Requirement for heme in the activation of purified guanylate cyclase by nitric oxide.

Guanylate cyclase activity was purified to apparent homogeneity from rat liver (7700-fold) and bovine lung (8600-fold) soluble fractions by ammonium sulfate precipitation, DEAE-cellulose chromatography, agarose gel filtration and isoelectric focussing. The purified enzymes did not contain heme and did not respond to NO, nitroprusside or NO-cysteine in the absence of exogenous hematin. By contrast, preformed NO-hemoglobin increased enzyme activity 10-12-fold or 60-80-fold when 4 mM MnCl2 or 4 mM MgCl2, respectively, were employed as the metal ion co-factor. Addition of hematin to the enzyme preparations restored responsiveness to NO, nitroprusside or NO-cysteine to levels seen with NO-hemoglobin. Partial purification of guanylate cyclase from the soluble fraction of bovine lung (2400-fold) by ammonium sulfate precipitation, DEAE-cellulose chromatography, agarose gel filtration and high pressure liquid chromatography (HPLC) resulted in a preparation which contained endogenous heme as indicated by absorbance at 436 nm and responded to NO, nitroprusside and NO-cysteine in the absence of added hematin. By contrast, guanylate cyclase purified from the hepatic supernatant by the identical procedure, did not contain detectable absorption due to heme and did not respond or responded poorly to NO, nitroprusside or NO-cysteine in the absence of exogenous hematin. Analogous to hepatic guanylate cyclase purified by isoelectric focussing, the HPLC purified hepatic enzyme was activated 14-fold by NO-hemoglobin in assays which contained 4 mM MnCl2 and 60-fold in assays with 4 mM MgCl2. Further, addition of hematin to the HPLC purified enzyme restored responsiveness to NO, nitroprusside and NO-cysteine to levels seen with NO-hemoglobin. These effects of hematin were specific for hematin and were not mimicked by albumin, sucrose or dithiothreitol. Moreover, the failure to observe stimulation of purified hepatic guanylate cyclase was not explained by a shift in the concentration response relationship between NO and guanylate cyclase activity. Several observations indicated that neither NO-thiol complexes nor [Fe(CN)5NO]-3 were the proximate moieties responsible for activation of guanylate cyclase by nitroprusside and related agents, as has been previously suggested. These results strongly support the proposal that activation of guanylate cyclase by NO and related agents specifically requires formation of an NO-heme complex.