IDENTIFICATION OF ELECTROSTATIC INTERACTION SITES BETWEEN THE REGULATORY AND CATALYTIC SUBUNITS OF CYCLIC-AMP-DEPENDENT PROTEIN-KINASE

Two classes of molecules inhibit the catalytic subunit (C) of the cycl ic AMP-dependent protein kinase (cAPK), the heat-stable protein kinase inhibitors (PKIs) and the regulatory (R) subunits. Basic sites on C, previously identified as important for R/C interaction in yeast TPK1 a nd corresponding to Lys213, Lys217, and Lys189 in murine C alpha, were replaced with either Ala or Thr and characterized for their kinetic p roperties and ability to interact with RI and PKI. rC(K213A) and rC(K2 17A) were both defective in forming holoenzyme with RI but were inhibi ted readily with PKI. This contrasts with rC(R133A), which is defectiv e in binding PKI but not RI (Wen & Taylor, 1994). Thus, the C-subunit employs two distinct electrostatic surfaces to achieve high-affinity b inding with these two types of inhibitory molecules even though all in hibitors share a common consensus site that occupies the active site c left. Unlike TPK1, mutation of Lys189 had no effect. The mutant C subu nits that were defective in binding RI, rC(K213A) and rC(K217A), were then paired with three RI mutants, rRI(D140A), rRI(E143A), and rRI(D25 8A), shown previously to be defective in recognition of C. Although th e mutations at Asp140 and Asp258 in RI were additive with respect to t he C mutations, rC(K213A) and rRI(E143A) were compensatory, thus ident ifying a specific electrostatic interaction site between RI and C. The results are discussed in terms of the RI and C crystal structures and the sequence homology between the yeast and mammalian enzymes.