Mechanisms of autoinhibition in cyclic nucleotide-dependent protein kinases.

Cyclic AMP-dependent protein kinase (PKA) and cGMP-dependent protein kinase (PKG) are autoinhibited through multiple interactions between their respective regulatory and catalytic domains. A large portion of this autoinhibition occurs through interactions between residues within the catalytic domain and those within either a substrate-like sequence (-RRXSX-) or pseudosubstrate sequence (-RRXAX-) in the regulatory domains. These contacts effectively inhibit catalysis by blocking substrate binding. Particularly important contacts involve the P-2, P-3, and P+1 residues where either serine, which is potentially autophosphorylated, or alanine occupies the P0 position. The primary sequence is apparently less important for autoinhibition in PKGs than in PKAs, and a conserved serine at P+2 in PKGs is important for autoinhibitory contacts. Elements outside the substrate-related sequences also contribute to autoinhibition in both PKA and PKG. For example, synthetic peptides with relatively short pseudosubstrate sequences are weak inhibitors; while heat-denatured RII subunit does not inhibit catalytic subunit, it is still rapidly autophosphorylated; and truncated PKGs lacking the substrate-like sequence are still partially autoinhibited. Thus, capacity for autoinhibition of PKA or PKG is provided by contacts involving direct interactions with the catalytic site and by contacts that stabilize an inactive conformation.