Mapping adenosine cyclic 3',5'-phosphate binding sites on type I and type II adenosine cyclic 3',5'-phosphate dependent protein kinases using ribose ring and cyclic phosphate ring analogues of adenosine cyclic 3',5'-phosphate

A series of adenosine cyclic 3',5'-phosphate (cAMP) derivatives containing modifications or substitutions in either the 2',3',4', or 5' position or the phosphate were examined for their abilities to activate type I isozymes of cAMP-dependent protein kinase (PK I) from rabbit or porcine skeletal muscle and type II isozymes of cAMP-dependent protein kinase (PK II) from bovine brain and heart. The studies revealed that the activation of both PK I and PK II isozymes requires a 2'-hydroxyl group in the ribo configuration, a 3' oxygen in the ribo configuration, and a charged cyclic phosphate. The two isozymes appeared to differ in those portions of their respective cAMP-binding sites that are adjacent to the 4' position of the ribose ring and the 3' position, 5' position, and phosphate portion of the cyclic phosphate ring.     

Amino acid sequence of the regulatory subunit of bovine type II adenosine cyclic 3',5'-phosphate dependent protein kinase

Evidence is presented that establishes the amino acid sequence of the regulatory subunit of type II cAMP-dependent protein kinase from bovine cardiac muscle. Complementary sets of overlapping peptides were generated primarily by tryptic digestion and by chemical cleavage at methionyl residues. The analysis was augmented by chemical cleavage at a single tryptophanyl residue and at three of the four aspartyl-proline bonds. Several large fragments generated by limited proteolysis contributed to the proof of structure. The subunit is a single chain of 400 residues corresponding to a molecular weight of 45 004. An amino-terminal segment of about 100 residues is believed to include the region responsible for oligomeric association. The remainder of the molecule consists of two tandem homologous domains, each of which is thought to bind a single molecule of cAMP. Comparison of the three domains with corresponding regions of the type I isozyme, of the Escherichia coli catabolite gene activator protein, and of cGMP-dependent protein kinase indicates extensive regions of homology and as much as 50% identity with the sequence of an internal segment of the type I isozyme.     

Amino acid sequence of the regulatory subunit of bovine type I adenosine cyclic 3',5'-phosphate dependent protein kinase

The complete amino acid sequence of the regulatory subunit of type I cAMP-dependent protein kinase from bovine skeletal muscle is presented. The S-carboxymethylated protein was cleaved with cyanogen bromide to provide a complete set of nonoverlapping fragments. These fragments were overlapped and aligned by using peptides generated by proteolytic cleavage. The protein contains 379 amino acid residues corresponding to a molecular weight of 42 804. As in the type II regulatory subunit of cAMP-dependent protein kinase, a pattern of internal gene duplication is observed, which is consistent with two cAMP-binding domains. The two types of regulatory subunit from type I and type II kinase display similarities in domain substructure and in amino acid sequence, which provide a molecular basis for new insight into their regulatory roles. Detailed analyses of the homology of the regulatory subunits of type I and type II cAMP-dependent protein kinase and of similar relationships to cGMP-dependent protein kinase and Escherichia coli catabolite gene activator protein are presented in accompanying reports from this laboratory [Takio, K., Smith, S. B., Krebs, E. G., Walsh, K., & Titani, K. (1984) Biochemistry (second paper of three in this issue); Takio, K., Wade, R. D., Smith, S. B., Krebs, E. G., Walsh, K. A., & Titani, K. (1984) Biochemistry (third paper of three in this issue)].     

Amino acid sequence of the catalytic subunit of bovine type II adenosine cyclic 3',5'-phosphate dependent protein kinase

The 350-residue amino acid sequence of the catalytic subunit of bovine cardiac muscle adenosine cyclic 3',5'-phosphate dependent protein kinase is described. The protein has a molecular weight of 40 862, which includes an N-tetradecanoyl (myristyl) group blocking the NH2 terminus and phosphate groups at threonine-197 and serine-338. Seven methionyl bonds in the S-carboxymethylated protein were cleaved with cyanogen bromide to yield eight primary peptides. These fragments, and subpeptides generated by cleavage with trypsin, pepsin, chymotrypsin, thermolysin, and Myxobacter AL-1 protease II, were purified and analyzed to yield the majority of the sequence. The primary peptides were aligned by analyses of overlapping peptides, particularly of methione-containing tryptic peptides generated after in vitro [14C]methyl exchange labeling of methionyl residues in the intact protein.     

Conformation of Leu-Arg-Arg-Ala-Ser-Leu-Gly bound in the active site of adenosine cyclic 3',5'-phosphate dependent protein kinase

Studies utilizing NMR spectroscopy have shown that adenosine cyclic 3',5'-phosphate dependent protein kinase (A-kinase) probably binds Leu-Arg-Arg-Ala-Ser-Leu-Gly (peptide 1) in one of two extended coil conformations (A or B). The relative reactivities of a series of N-methylated peptides based on the structure of peptide 1 might, therefore, be related to how well each can assume the A or B conformation. From estimates of the magnitude of steric interactions that would be induced by N-methylation of an amide in peptide 1 that is locked in either conformation, the ability of each peptide to form that conformation was predicted. The ability of A-kinase to catalyze phosphorylation of the N-methylated peptides correlated well with the ability of each peptide to form conformation A, but not conformation B. In accord with these findings, the reactivity of an unreactive N-methylated peptide was partially restored by a second change, which allowed the peptide to assume conformation A. These results suggest that, when bound in the enzymatic active site, peptide 1 has a conformation that resembles structure A much more closely than structure B.     

Interaction of guanosine cyclic 3',5'-phosphate dependent protein kinase with lin-benzoadenine nucleotides

Using the activated cGMP-dependent protein kinase in the presence of the phosphorylatable peptide [[Ala34]histone H2B-(29-35)], we found that lin-benzoadenosine 5'-diphosphate (lin-benzo-ADP) was a competitive inhibitor of the enzyme with respect to ATP with a Ki (22 microM) similar to the Kd (20 microM) determined by fluorescence polarization titrations. The Kd for lin-benzo-ADP determined in the absence of the phosphorylatable peptide, however, was only 12 microM. ADP bound with lower affinity (Ki = 169 microM; Kd = 114 microM). With [Ala34]histone H2B-(29-35) as phosphoryl acceptor, the Km for lin-benzo-ATP was 29 microM, and that for ATP was 32 microM. The Vmax with lin-benzo-ATP, however, was only 0.06% of that with ATP as substrate [0.00623 +/- 0.00035 vs. 11.1 +/- 0.17 mumol (min.mg)-1]. Binding of lin-benzo-ADP to the kinase was dependent upon a divalent cation. Fluorescence polarization revealed that Mg2+, Mn2+, Co2+, Ni2+, Ca2+, Sr2+, and Ba2+ supported nucleotide binding to the enzyme; Ca2+, Sr2+, and Ba2+, however, did not support any measurable phosphotransferase activity. The rank order of metal ion effectiveness in mediating phosphotransferase activity was Mg2+ greater than Ni2+ greater than Co2+ greater than Mn2+. Although these results were similar to those observed with the cAMP-dependent protein kinase [Hartl, F. T., Roskoski, R., Jr., Rosendahl, M. S., & Leonard, N. J. (1983) Biochemistry 22, 2347], major differences in the Vmax with lin-benzo-ATP as substrate and the effect of peptide substrates on nucleotide (both lin-benzo-ADP and ADP) binding were observed.     

Interaction of "aza" and "deaza" analogs of adenosine cyclic 3', 5'-phosphate with some enzymes of adenosine cyclic 3', 5'-phosphate metabolism: evidence that the lone pair electrons of N-3 are involved in the binding of adenosine cyclic 3', 5'-phosphate to type II adenosine cyclic 3', 5'-phosphate-dependent protein kinase.

Five hetercyclic analogs of adenosine cyclic 3',5'-phosphate (cyclic AMP) were examined for their ability (1) to stimulate type II cyclic AMP-dependent kinases from bovine brain, bovine heart, and rat liver; (2) to serve as substrates for "high Km" (Km for cyclic AMP = 0.13-0.43 mM) cyclic nucleotide phosphodiesterases from bovine heart, rabbit kidney, and rat liver; and (3) to inhibit the hydrolysis of cyclic AMP catalyzed by "low Km" (Km for cAMP = 0.32-1.5 muM) cyclic nucleotide phosphodiesterases from bovine brain, bovine heart, dog heart, rabbit liver, rat brain and rat liver. The analogs all had a purine ring system which had been modified by replacement of a ring carbon with nitrogen or vice versa to yield 2-aza-cAMP (7-amino-4-beta-D-ribofuranosylimidazo [4,5-d] -v-triazine cyclic 3',5'-phosphate); 8-aza-cAMP (7-amino-3-beta-D-ribofuranosyl-v-triazolo-[4,5-d]-pyrimidine cyclic 3',5'-phosphate); 1 deaza-cAMP (7-amino-3-beta-D-ribofuranosylimidazo [4,5-b[pyridine cyclic 3',5'-phosphate); 3-deaza-cAMP (4-amino-1-beta-D-ribofuranosylimidazo[4,5-c]pyridine cyclic 3',5'-phosphate) and 7-deaza-cAMP (7-amino-4-beta-D-ribofuranosylpyrrolo[2,3-d]pyrimidine cyclic 3',5'-phosphate).     

Guanosine cyclic 3',5'-phosphate dependent protein kinase, a chimeric protein homologous with two separate protein families

The amino acid sequence of bovine lung cGMP-dependent protein kinase has been determined by degradation and alignment of two primary overlapping sets of peptides generated by cleavage at methionyl or arginyl residues. The protein contains 670 residues in a single N alpha-acetylated chain corresponding to a molecular weight of 76 331. The function of the molecule is considered in six segments of sequence which may correspond to four folding domains. From the amino terminus, the first segment is related to the dimerizing property of the protein. The second and third segments appear to have evolved from an ancestral tandem internal gene duplication, generating twin cGMP-binding domains which are homologous to twin domains in the regulatory subunits of cAMP-dependent protein kinase and to the cAMP-binding domain of the catabolite gene activator of Escherichia coli. The fourth and fifth segments may comprise one domain which is homologous to the catalytic subunits of cAMP-dependent protein kinase, of calcium-dependent phosphorylase b kinase, and of certain oncogenic viral protein tyrosine kinases. The regulatory, amino-terminal half of cGMP-dependent protein kinase appears to be related to a family of smaller proteins that bind cAMP for diverse purposes, whereas the catalytic, carboxyl-terminal half is related to a family of protein kinases of varying specificity and varying sensitivity to regulators. These data suggest that ancestral gene splicing events may have been involved in the fusion of two families of proteins to generate the allosteric character of this chimeric enzyme.     

A kinetic study of interactions of (Rp)- and (Sp)-adenosine cyclic 3',5'-phosphorothioates with type II bovine cardiac muscle adenosine cyclic 3',5'-phosphate dependent protein kinase

The stereoselectivity of the adenosine cyclic 3',5'-phosphate (cAMP) binding sites on the regulatory subunit of the type II bovine cardiac muscle cAMP-dependent protein kinase was investigated by examining the interactions of (Rp)- and (Sp)-adenosine cyclic 3',5'-phosphorothioates (cAMPS) with these sites. While activation of the holoenzyme and binding to the regulatory subunit of the type II kinase were observed for both of these diastereomers, there were significant differences between the interactions of the cAMPS isomers with the enzyme. In particular, the Sp isomer is more potent than the Rp species not only in the activation of reconstituted, as well as directly isolated, holoenzyme but also in the inhibition of [3H]cAMP binding to the regulatory subunit. A marked preference for the binding of the Sp isomer to site 2 in the regulatory subunit exists. Hydrogen bonding of a functional group on the regulatory subunit with preferential orientation toward the exocyclic oxygen rather than the sulfur of the thiophosphoryl residue may be involved in the observed selectivity of cAMPS binding and activation. In addition to our findings on the stereoselectivity of the binding of cAMPS to cAMP-dependent protein kinase, we have established a method for the reconstitution of holoenzyme from the purified subunits without subjecting the regulatory protein to denaturing conditions.     

Association of calmodulin with peptide analogues of the inhibitory region of the heat-stable protein inhibitor of adenosine cyclic 3',5'-phosphate dependent protein kinase

A 20-residue peptide analogue (IASGRTGRRNAIHDILVSSA) of the 8000-dalton heat-stable cAMP-dependent protein kinase inhibitor undergoes efficient calcium-dependent binding by calmodulin, with Kd approximately 70 nM when calcium is present. It is a potent inhibitor of smooth muscle myosin light chain kinase and of the calmodulin-dependent phosphatase activity of calcineurin. At concentrations above 3 microM, the peptide stimulates the basal activity of calcineurin. The native protein kinase inhibitor has no effect on the catalytic activity of myosin light chain kinase and is moderately inhibitory to both the calmodulin-dependent and -independent phosphatase activity of calcineurin. Competition experiments using excess concentrations of calcineurin and calmodulin suggest that the primary interaction of the native heat-stable inhibitor is with the catalytic subunit of protein kinase. Dansylcalmodulin exhibits only a weak interaction with the inhibitor. Observations on deletion peptides of the 20-residue analogue help to delineate the overlapping peptide binding specificities of the cAMP-dependent protein kinase [Scott, J. D., Glaccum, M. B., Fischer, E. H., & Krebs, E. G. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 1613-1616] and calmodulin. In both cases, the most effectively bound peptides contain the RTGRR sequence.     

Chemical mechanism of the adenosine cyclic 3',5'-monophosphate dependent protein kinase from pH studies

The pH dependence of kinetic parameters and inhibitor dissociation constants for the adenosine cyclic 3',5'-monophosphate dependent protein kinase reaction has been determined. Data are consistent with a mechanism in which reactants selectively bind to enzyme with the catalytic base unprotonated and an enzyme group required protonated for peptide (Leu-Arg-Arg-Ala-Ser-Leu-Gly) binding. Binding of the peptide apparently locks both of the above enzyme residues in their correct protonation state. MgATP preferentially binds fully ionized and requires an enzyme residue (probably lysine) to be protonated. The maximum velocity and V/KMgATP are pH independent. The V/K for Ser-peptide is bell-shaped with pK values of 6.2 and 8.5 estimated. The pH dependence of 1/Ki for Leu-Arg-Arg-Ala-Ala-Leu-Gly is also bell-shaped, giving pK values identical with those obtained for V/KSer-peptide, while the Ki for MgAMP-PCP increases from a constant value of 650 microM above pH 8 to a constant value of 4 mM below pH 5.5. The Ki for uncomplexed Mg2+ obtained from the Mg2+ dependence of V and V/KMgATP is apparently pH independent.     

Comparison of cyclic nucleotide specificity of guanosine 3',5'-monophosphate-dependent protein kinase and adenosine 3',5'-monophosphate-dependent protein kinase.

Guanosine 3',5'-monophosphate-dependent protein kinase (cyclic GMP-dependent protein kinase) and adenosine 3',5'-monophosphate-dependent protein kinase (cyclic AMP-dependent protein kinase) exhibited a high degree of cyclic nucleotide specificity when hormone-sensitive triacylglycerol lipase, phosphorylase kinase, and cardiac troponin were used as substrates. The concentration of cyclic GMP required to activate half-maximally cyclic dependent protein kinase was 1000- to 100-fold less than that of cyclic AMP with these substrates. The opposite was true with cyclic AMP-dependent protein kinase where 1000- to 100-fold less cyclic AMP than cyclic GMP was required for half-maximal enzyme activation. This contrasts with the lower degree of cyclic nucleotide specificity of cyclic GMP-dependent protein kinase of 25-fold when histone H2b was used as a substrate for phosphorylation. Cyclic IMP resembled cyclic AMP in effectiveness in stimulating cyclic GMP-dependent protein kinase but was intermediate between cyclic AMP and cyclic GMP in stimulating cyclic AMP-dependent protein kinase. The effect of cyclic IMP on cyclic GMP-dependent protein kinase was confirmed in studies of autophosphorylation of cyclic GMP-dependent protein kinase where both cyclic AMP and cyclic IMP enhanced autophosphorylation. The high degree of cyclic nucleotide specificity observed suggests that cyclic AMP and cyclic GMP activate only their specific kinase and that crossover to the opposite kinase is unlikely to occur at reported cellular concentrations of cyclic nucleotides.     

DNA binding by cyclic adenosine 3',5'-monophosphate dependent protein kinase from calf thymus nuclei.

Cyclic adenosine 3',5'-monophosphate (cAMP) dependent protein kinase and proteins specifically binding cAMP have been extracted from calf thymus nuclei and analyzed for their abilities to bind to DNA. Approximately 70% of the cAMP-binding activity in the nucleus can be ascribed to a nuclear acidic protein with physical and biochemical characteristics of the regulatory (R) subunit of cAMP-dependent protein kinase. Several peaks of protein kinase activity and of cAMP-binding activity are resolved by affinity chromatography of nuclear acidic proteins on calf thymus DNA covalently linked to aminoethyl Sephrarose 4B. When an extensively purified protein kinase is subjected to chromatography on the DNA column in the presence of 10(-7) M cAMP, the R subunit of the kinase is eluted from the column at 0.05 M NaCl while the catalytic (C) subunit of the enzyme is eluted at 0.1-0.2 M NaCl. When chromatographed in the presence of histones, the R subunit is retained on the column and is eluted at 0.6-0.9 M NaCl. In the presence of cAMP, association of the C subunit with DNA is enhanced, as determined by sucrose density gradient centrifugation of DNA-protein kinase complexes. cAMP increases the capacity of the calf thymus cAMP-dependent protein kinase preparation to bind labeled calf thymus DNA, as determined by a technique employing filter retention of DNA-protein complexes. This protein kinase preparation binds calf thymus DNA in preference to salmon DNA, Escherichia coli DNA, or yeast RNA. Binding of protein kinases to DNA may be part of a mechanism for localizing cyclic nucleotide stimulated protein phosphorylation at specific sites in the chromatin.     

In vivo activation of cyclic adenosine 3':5'-phosphate-dependent protein kinase in Chinese hamster ovary cells treated with N-6, O-2'-diburyl cyclic adenosine 3':5'-phosphate.

Treatment of Chinese hamster ovary cells with dibutyryl cyclic AMP, which results in a net increase of the intracellular cyclic AMP level, converts the epithelial-like cells to a fibroblast-like shape. Protein kinase activity in cells treated with 1 mM dibutyryl cyclic AMP show a 3-fold increase in V_max but no appreciable changes in the apparent K_m for ATP. When cells are treated with dibutyryl cyclic AMP, there is a time-dependent conversion of cyclic AMP-stimulable protein kinase to cyclic AMP-independent catalytic subunits, as demonstrated by Sephadex G-100 gel filtration. These experiments demonstrate the activation of the cyclic AMP-dependent protein kinase $\text{in vivo}$. This activation may lead to phosphorylation of certain cellular constituent(s) and thus may be involved in the observed morphological transformation.     

Binding of adenosine 3',5'-monophosphate dependent protein kinase regulatory subunit to immobilized cyclic nucleotide derivatives.

Several cyclic nucleotide derivatives with aminoalkyl side chains attached to the purine ring were synthesized and their interactions with adenosine 3',5'-monophosphate (cAMP) dependent protein kinase were studied before and after immobilization to CNBr-activated Sepharose 4B. The soluble N6-substituted derivatives were as effective as cAMP itself in activating protein kinase and were more effective than 8-substituted cAMP derivatives, whereas the 2-substituted cAMP derivatives and the cGMP derivatives were the least effective. All of the synthetic derivatives tested were poor substrates for beef heart phosphodiesterase being hydrolyzed at rates less than 2% for that of cAMP itself. Utilizing methodology developed to evaluate the affinity of protein kinase for immogilized cyclic nucleotides it was found that all of the immobilized cyclic nucleotides interacted with protein kinase in a biospecific manner as judged by the following criteria: (1) the immobilized cyclic nucleotides competed with cAMP for the binding sites on protein kinase; (2) the analogous spacer-arm did not compete; and (3) the effects of enzyme concentration, MgATP, and cleavage of the cyclic phosphate ring on the interactions of protein kinase with the immobilized cyclic nucleotides were the same as previously shown for free cAMP. In addition, the immobilized ligands were bound with the same order of effectiveness as the analogous soluble ligand. The observed Ka for the activation of 0.005 muM protein kinase by N6-H2N(CH2)2-cAMP was increased from 0.23 to 3 muM by the process of immobilization. This increase was unaffected by the coupling density and spacer-arm length. The observed Kb for 0.10 muM protein kinase binding to immobilized N6-H2N(CH2)2-cAMP was increased as the molecular sieving exclusion limit of the matrix used was decreased indicating that at least part of this decrease in apparent affinity upon immobilization is due to exclusion of the enzyme from a portion of the matrix and therefore of the immobilized ligand molecules.     

Limited proteolysis alters the photoaffinity labeling of adenosine 3',5'-monophosphate dependent protein kinase II with 8-azidoadenosine 3',5'-monophosphate

Photoaffinity labeling of the regulatory subunits of cAMP-dependent protein kinase with 8-azidoadenosine 3',5'-monophosphate (8-N3cAMP) has proved to be a very specific method for identifying amino acid residues that are in close proximity to the cAMP-binding sites. Each regulatory subunit contains two tandem cAMP-binding sites. The type II regulatory subunit (RII) from porcine heart was modified at a single site, Tyr-381 [Kerlavage, A., & Taylor, S.S. (1980) J. Biol. Chem. 255, 8483-8488]. When a proteolytic fragment of this RII subunit was photolabeled with 8-N3cAMP, two sites were covalently modified. One site corresponded to Tyr-381 and, thus, was analogous to the native RII. The other site of modification was identified as Tyr-196, which is not labeled in the native protein. Photoaffinity labeling was carried out in the presence of various analogues of cAMP that show a preference for one of the two tandem cAMP-binding sites. These studies established that the covalent modification of Tyr-381 was derived from 8-N3cAMP that was bound to the second cAMP-binding site (domain B) and that covalent modification to Tyr-196 was due to 8-N3cAMP that was bound to the first cAMP-binding site (domain A). These sites of covalent modification have been correlated with a model of each cAMP-binding site on the basis of the crystal structure of the catabolite gene activator protein (CAP), which is the major cAMP-binding protein in Escherichia coli.