The catalytic (C) subunit of cAMP-dependent protein kinase interacts with two classes of inhibitors. The regulatory (R) subunits, types I and II, associate to form an inactive holoenzyme complex that is activated in response to cAMP. The C-subunit is also inhibited by small heat-stable protein kinase inhibitors (PKI's). Inhibition by both PKI and R(I)-subunit requires the synergistic high-affinity binding of MgATP. The stabilizing effect of ATP was quantitated by using analytical gel chromatography. Both the type I holoenzyme and the C.PKI complex in the presence of MgATP show apparent K(d)'s for subunit association that are below 0.1 nM, while in the absence of MgATP the apparent K(d)'s are 125 nM and 2.3 muM, respectively, for the two complexes. In the absence of MgATP both complexes also can be dissociated readily and, hence, activated by salt-induced dissociation. Under physiological salt concentrations, salt-induced dissociation would be substantial in the absence of the high-affinity binding of MgATP. In both complexes, the ATPase activity of the free C-subunit is abolished. The off rates for MgATP also indicate that the type I holoenzyme is more stable than the C-PKI complex. The off rate (t1/2) for MgATP from the C.PKI complex is 17 min, while the off rate for the type I holoenzyme is 11.7 h. When the C-PKI complex is incubated with R(I)-subunit in the presence or absence of MgATP, the C-subunit preferentially reassociates with the R(I)-subunit, forming holoenzyme. In contrast, free PKI cannot compete for the C-subunit when it is part of a holoenzyme complex. These in vitro results suggest that the formation of the C.PKI complex may not be a dead-end pathway. Instead, PKI could function as a shuttle mechanism for eventually returning the C-subunit to a holoenzyme complex.
