Translational control by hemin is due to binding to cyclic AMP-dependent protein kinase

Our previous work [Proc. Natl, Acad. Sci. USA (1977) 74, 1463-1467, 3326-3329] is consistent with the view that (a) the hemin-controlled inhibitor of protein synthesis in reticulocyte lysates (active eIF-2 kinase) is formed by phosphorylation of proinhibitor (inactive eIF-2 kinase) catalyzed by cyclic AMP-dependent protein kinase (ATP-protein phosphotransferase; EC 2.7.1.37), and (b) hemin prevents this conversion by blocking the interaction of cyclic AMP with the kinase's regulation subunit, thereby rendering the enzyme inactive. We now show that hemin blocks cyclic AMP binding because it itself binds specifically to the regulatory subunit. This binding is noncompetitive with respect to cyclic AMP. Whereas unlabeled hemin can displace bound [3H]hemin as well as cyclic [3H]AMP, unlabeled cyclic AMP can displace bound cyclic [3H]AMP but not [3H]hemin. This suggests that cyclic AMP and hemin bind to different sites on the protein and that hemin binding affects cyclic AMP binding in an allosteric manner.