The crystal structure of the porcine heart catalytic subunit of cAMP‐dependent protein kinase in a ternary complex with the MgATP analogue MnAMP‐PNP and a pseudosubstrate inhibitor peptide, PKI(5‐24), has been solved at 2.0 A resolution from monoclinic crystals of the catalytic subunit isoform CA. The refinement is presently at an R factor of 0.194 and the active site of the molecule is well defined. The glycine‐rich phosphate anchor of the nucleotide binding fold motif of the protein kinase is a beta ribbon acting as a flap with conformational flexibility over the triphosphate group. The glycines seem to be conserved to avoid steric clash with ATP. The known synergistic effects of substrate binding can be explained by hydrogen bonds present only in the ternary complex. Implications for the kinetic scheme of binding order are discussed. The structure is assumed to represent a phosphotransfer competent conformation. The invariant conserved residue Asp166 is proposed to be the catalytic base and Lys168 to stabilize the transition state. In some tyrosine kinases Lys168 is functionally replaced by an Arg displaced by two residues in the primary sequence, suggesting invariance in three‐dimensional space. The structure supports an in‐line transfer with a pentacoordinate transition state at the phosphorus with very few nuclear movements.

The discovery of several hundred different protein kinases involved in highly diverse cellular signaling pathways is in stark contrast to the much smaller number of known modulators of cell signaling. Of these, the H series protein kinase inhibitors (1- Isoquinolinesulfonamide protein kinase inhibitors of the H series are among the most widely used inhibitors of Ser/Thr kinases and are indispensable in cellular and signal transduction research. Protein phosphorylation (central to cellular regulation) is mediated by the individual action of several hundred different protein kinases. Recent progress in the understanding of the highly complex cellular signaling networks depends largely on the availability and quality of specific agents that interfere with the pathways under investigation. The usefulness of a protein kinase inhibitor is defined by its ability to permeate cell membranes, its solubility, and its relative degree of specificity. Each year hundreds of publications describe work using isoquinolinesulfonamide inhibitors, despite the fact that the mode of specific inhibition by these compounds is not completely understood.(We have used the cAMP-dependent protein kinase (cAPK), 1 which binds many H series inhibitors, as a model system to investigate the factors governing inhibitor binding and specificity. Three of the most frequently used members of this class are H7, H8 (1), and H89 (2) (Fig. 1). They all act in competition to ATP but not to substrate. The highest selectivity and affinity is found with H89, which has a K i of 48 nM for cAPK, whereas H8 has a moderate affinity for both cGPK and cAPK, and H7 inhibits protein kinase C in addition to cGPK and cAPK (Table I). The inhibitors have enabled assignment of specific roles of these protein kinases in numerous regulatory interrelations (3-8). They are, however, not only valuable for the cell biologist. Pharmacologists are increasingly interested in the possibility of interfering with cellular signaling, especially in the area of anticancer drug discovery. There is evidence that isoquinolinesulfonyl inhibitors may be promising in this field (9 -14). Recently, crystal structures from several different protein kinases have been solved (for review see . The structures confirmed not only the high degree of structural conservation of the highly homologous catalytic kinase core (20) but simultaneously showed how subtle differences are responsible for individual properties of protein kinases. Most of the residues that are highly conserved or invariant in the protein kinase family line the active site, and many of them interact with ATP (21,22). Despite high similarity of K m values for ATP binding among protein kinases, the H inhibitors have remarkably different K i values. The crystal structures of kinase bound inhibitor molecules described here now show the mode of inhibitory action and the factors governing selectivity and will provide a firm basis for the design of new protein kinase inhibitors. EXPERIMENTAL PROCEDURESProtein Expression and Purification-Exp...

The results explain the high inhibitory potency of staurosporine, and also illustrate the flexibility of the protein kinase active site. The structure, therefore, is not only useful for the design of improved anticancer therapeutics and signaling drugs, but also provides a deeper understanding of the conformational flexibility of the protein kinase.

Protein kinases require strict inactivation to prevent spurious cellular signaling; overactivity can cause cancer or other diseases and necessitates selective inhibition for therapy. Rho-kinase is involved in such processes as tumor invasion, cell adhesion, smooth muscle contraction, and formation of focal adhesion fibers, as revealed using inhibitor Y-27632. Another Rho-kinase inhibitor, HA-1077 or Fasudil, is currently used in the treatment of cerebral vasospasm; the related nanomolar inhibitor H-1152P improves on its selectivity and potency. We have determined the crystal structures of HA-1077, H-1152P, and Y-27632 in complexes with protein kinase A (PKA) as a surrogate kinase to analyze Rho-kinase inhibitor binding properties. Features conserved between PKA and Rho-kinase are involved in the key binding interactions, while a combination of residues at the ATP binding pocket that are unique to Rho-kinase may explain the inhibitors' Rho-kinase selectivity. Further, a second H-1152P binding site potentially points toward PKA regulatory domain interaction modulators.