cAMP-dependent protein kinases I and II: divergent turnover of subunits

cAMP-dependent protein kinase subunits were isolated from livers of rats that had been subjected to biosynthetic labeling with radioactive leucine. By application of ligand and antibody affinity techniques pure regulatory (R I; R II) and catalytic (C) subunits could be obtained in high yields, which allowed measurement of the apparent degradation rate constants and half-lives following a double isotope labeling protocol. In this way marked differences of apparent half-lives of regulatory subunits R I (t1/2 = 31 h) and R II (t1/2 = 125 h) were observed. To avoid the negative influence of reutilization inherent in the decay experiments, specific radioactivities were determined after a short isotope pulse. This parameter, which under steady-state conditions reflects the fractional turnover rate of the subunits, was found to be different for all three protein kinase subunits. Relative to total liver protein, the ratios R I:R II:C corresponded to 3.9:0.6:2. Our data indicate that in each type of protein kinase isoenzymes regulatory and catalytic subunits turn over with similar rates. The type I isoenzyme, however, is renewed much faster than protein kinase II. Furthermore, our findings are consistent with the thesis that free subunits as generated by activation are more susceptible to degradation than the holoenzymes, leading under steady-state conditions to compensatory resynthesis. Since renewal of R I exceeded that of R II also in two other tissues, the elevated turnover of protein kinase I as an indicator of preferential activation appears to be a general phenomenon. The different turnover of the two isoenzymes, then, may relate to different cellular functions like modulation of enzyme activity vs. modulation of gene activity.     

Different phosphorylation mechanisms are involved in the activation of sucrose non-fermenting 1 related protein kinases 2 by osmotic stresses and abscisic acid

In Arabidopsis cell suspension, hyperosmotic stresses (mannitol and NaCl) were previously shown to activate nine sucrose non-fermenting 1 related protein kinases 2 (SnRK2s) whereas only five of them were also activated by abscisic acid (ABA) treatment. Here, the possible activation by phosphorylation/dephosphorylation of each kinase was investigated by studying their phosphorylation state after osmotic stress, using the Pro-Q Diamond, a specific dye for phosphoproteins. All the activated kinases were phosphorylated after osmotic stress but the induced phosphorylation changes were clearly different depending on the kinase. In addition, the increase of the global phosphorylation level induced by ABA application was lower, suggesting that different mechanisms may be involved in SnRK2 activation by hyperosmolarity and ABA. On the other hand, SnRK2 kinases remain activated by hyperosmotic stress in ABA-deficient and ABA-insensitive mutants, indicating that SnRK2 osmotic activation is independent of ABA. Moreover, using a mutant form of SnRK2s, a specific serine in the activation loop was shown to be phosphorylated after stress treatments and essential for activity and/or activation. Finally, SnRK2 activity was sensitive to staurosporine, whereas SnRK2 activation by hyperosmolarity or ABA was not, indicating that SnRK2 activation by phosphorylation is mediated by an upstream staurosporine-insensitive kinase, in both signalling pathways. All together, these results indicate that different phosphorylation mechanisms and at least three signalling pathways are involved in the activation of SnRK2 proteins in response to osmotic stress and ABA.     

Immunoelectron microscopic localization of catalytic and regulatory subunits of cAMP-dependent protein kinases in the parotid gland

The subcellular distribution of catalytic (C) and regulatory (RI and RII) subunits of cAMP-dependent protein kinases has been studied by electron microscopy immunocytochemistry. The C-subunit was localized in the inner membrane-matrix space of mitochondria, at the cytoplasmic face of the rough endoplasmic reticulum (rER), in the nucleolus and in peripheral heterochromatin regions. A C-specific immunoreactivity was also found in specific domains of the basal and basolateral plasma membranes of acinar and duct cells, in centrosomes and on keratin filaments which anchor in desmosomes. The RI- and RII-subunits showed a basically similar subcellular distribution. A remarkably high RII-immunoreactivity, in the absence of C-immunoreactivity, was demonstrated in the secretory granules. These results demonstrate the presence of cAMP-dependent protein kinases or their subunits in many subcellular organelles. They also indicate a role for cAMP-dependent protein kinases in the regulation of a number of basal cellular functions as well as their importance in functional and structural cell-cell interactions.     

Catalytic subunit of cAMP-dependent protein kinase is essential for cAMP-mediated mammalian gene expression

Cyclic AMP-stimulated mRNA levels in cultured rat hepatocytes were inhibited by three different inhibitors of cAMP-dependent protein kinase activity: (i) Rp-cAMPS, a cAMP analog with a sulfur substitution at the equatorial oxygen of the cyclic monophosphate; (ii) H8, an isoquinoline sulfonamide derivative; and (iii) PKI, a 20-amino acid synthetic peptide of the Walsh protein kinase inhibitor. These inhibitors specifically blocked the cAMP-stimulated increase in mRNA for tyrosine aminotransferase and phosphoenolpyruvate carboxykinase; they had no effect on the level of albumin mRNA which is not cAMP regulated. These results provide functional evidence that kinase activity involving protein phosphorylation is required in cAMP-mediated gene expression in mammalian cells.     

Post-translational modifications of the regulatory subunits of cAMP-dependent protein kinases during G1/S progression

During the G1/S transition of the cell cycle variations in the labelling by 8-N3-[32P]cAMP of the protein kinase A regulatory subunits RI and RII, used as a probe to monitor post-translational modifications that may regulate cAMP binding, were observed in synchronized HeLa cells. A decrease in 8-N3-[32P]cAMP labelling of RI, RII and RII phosphorylated by the catalytic subunit of PKA was correlated with the increased percentage of cells in phases G1. An increase in 8-N3-[32P]cAMP incorporated into the 54-kDa RII subunit during progression from G1 to S was correlated with an increase in intracellular cAMP. A transient increase in Mn-SOD activity was detected in cells arrested at the G1/S transition using two different techniques, suggesting that oxidative modulation of regulatory subunits by free radicals may modify cAMP binding sites during the cell cycle. Decreased photoaffinity labelling by 8-N3-[32P]cAMP of RI, RII and autophosphorylated RII subunits was found to be an inherent characteristic of PKA in the G1/S transition.     

Retinoylation of the cAMP-binding regulatory subunits of type I and type II cAMP-dependent protein kinases in HL60 cells.

Retinoylation (retinoic acid acylation) is a post-translational modification of proteins occurring in a variety of eukaryotic cell lines. There are at least 20 retinoylated proteins in the human myeloid leukemia cell line HL60 (N. Takahashi and T.R. Breitman (1990) J. Biol. Chem. 265, 19, 158-19, 162). Here we found that some retinoylated proteins may be cAMP-binding proteins. Five proteins, covalently labeled by 8-azido-[32P]cAMP which specifically reacts with the regulatory subunits of cAMP-dependent protein kinase, comigrated on two-dimensional polyacrylamide gel electrophoresis with retinoylated proteins of Mr 37,000 (p37RA), 47,000 (p47RA), and 51,000 (p51RA) labeled by [3H]retinoic acid treatment of intact cells. Furthermore, p47RA coeluted on Mono Q anion exchange chromatography with the type I cAMP-dependent protein kinase holoenzyme and p51RA coeluted on Mono Q anion exchange chromatography with the type II cAMP-dependent protein kinase holoenzyme. An antiserum specific to RI, the cAMP-binding regulatory subunit of type I cAMP-dependent protein kinase, immunoprecipitated p47RA. An antiserum specific to RII, the cAMP-binding regulatory subunit of type II cAMP-dependent protein kinase, immunoprecipitated p51RA. These results indicate that both the RI and the RII regulatory subunits of cAMP-dependent protein kinase are retinoylated. Thus, an early event in RA-induced differentiation of HL60 cells may be the retinoylation of subpopulations of both RI and RII.     

Characterization of small cAMP-binding fragments of cAMP-dependent protein kinases.

Monomeric cAMP-binding fragments of molecular mass 16,000 and 14,000 daltons were obtained by Sephadex G-75 chromatography of partially trypsin-hydrolyzed regulatory subunits of cAMP-dependent protein kinase isozymes I and II, respectively. The Stokes radii were 19.1 and 16.4 A, the frictional ratios were 1.15 and 1.03, and the sedimentation coefficients were 1.94 and 1.91 S for the 16,000- and 14,000-dalton fragments, respectively. The 16,000-dalton fragment retained specific cyclic nucleotide binding characteristics of the native protein. The specificity of cyclic nucleotide binding to the 14,000-dalton fragment (cAMP greater than cIMP = 8-bromo-cAMP = 8-oxo-cAMP greater than cUMP = cGMP) differed from that of the native subunit (cAMP = 8-oxo-cAMP greater than 8-bromo-cAMP greater than cIMP greater than cUMP = cGMP). The 14,000-dalton fragment bound nearly 1 mol of cAMP/mol of fragment. The binding exchange rate of cAMP was much faster for the 14,000-dalton fragment than for either of the native regulatory subunits or for the 16,000 dalton fragment. Although hemin inhibited cAMP binding to the native regulatory subunits and to the 16,000 dalton fragment, the molecule did not affect cAMP binding to the 14,000-dalton fragment. Both of the native regulatory subunits and the isolated 16,000- and 14,000-dalton fragments could be covalently labeled with the photoaffinity analog, 8-N3-[32P]cAMP. The 14,000-dalton fragment could not be phosphorylated and neither fragment could recombine with the catalytic subunit to inhibit its activity. The results indicate that the functional entities of the regulatory subunit other than cAMP binding are destroyed by trypsin. The properties of the 16,000-dalton fragment suggest that the intact cAMP-binding site is contained in a small trypsin-resistant "core" of the native regulatory subunit. The properties of the 14,000-dalton fragment imply that part of the binding site of the native regulatory subunit was slighlty modified or lost during preparation of this fragment.     

Characterization and comparison of membrane-associated and cytosolic cAMP-dependent protein kinases. Studies on human erythrocyte protein kinases.

Cyclic AMP-dependent protein kinase from human erythrocyte plasma membranes was solubilized with Triton X-100, partially purified, and systematically characterized by a series of physicochemical studies. Sedimentation and gel filtration experiments showed that the 6.6 S holoenzyme had a Stokes radius (a) of 5.7 nm and was dissociated into native 4.8 S cAMP-binding (a = 4.5 nm) and 3.2 S catalytic (a = 2.6 nm) subunits. A minimum subunit molecular weight of 48,000 was established for the regulatory subunit by photoaffinity labeling with 8-azido[32P]cAMP, sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and autoradiography. These data suggest an asymmetric tetrameric (R2C2) structure (Mr approximately equal to 160,000) for the membrane-derived enzyme. Membrane-derived protein kinase was characterized as a type I enzyme on the basis of its R subunit molecular weight, pI values (R, 4.9; holoenzyme, 5.75 and 5.95), dissociation by 0.5 M NaCl and 50 microgram/ml of protamine, 20-fold reduced affinity for cAMP in the presence of 0.3 mM MgATP, elution from DEAE-cellulose at low ionic strength, and kinetic and cAMP-binding properties. The physicochemical properties of the membrane protein kinase closely parallel the characteristics of erythrocyte cytosolic protein kinase I but are clearly dissimilar from those of the soluble type II enzyme. Moreover, regulatory subunits of the membrane-associated and cytosolic type I kinases were indistinguishable in size, shape, subunit molecular weight, charge, binding and reassociation properties, and peptide maps of the photoaffinity-labeled cAMP-binding site, suggesting a high degree of structural and functional homology in this pair of enzymes. In view of the predominant occurrence of particulate type II protein kinases in rabbit heart and bovine cerebral cortex, the present results suggest that the distribution of membrane-associated protein kinases may be tissue- or species-specific, but not isoenzyme-specific.     

Epac, in synergy with cAMP-dependent protein kinase (PKA), is required for cAMP-mediated mitogenesis

cAMP stimulates proliferation in many cell types. For many years, cAMP-dependent protein kinase (PKA) represented the only known cAMP effector. PKA, however, does not fully mimic the action of cAMP, indicating the existence of a PKA-independent component. Since cAMP-mediated activation of the G-protein Rap1 and its phosphorylation by PKA are strictly required for the effects of cAMP on mitogenesis, we hypothesized that the Rap1 activator Epac might represent the PKA-independent factor. Here we report that Epac acts synergistically with PKA in cAMP-mediated mitogenesis. We have generated a new dominant negative Epac mutant that revealed that activation of Epac is required for thyroid-stimulating hormone or cAMP stimulation of DNA synthesis. We demonstrate that Epac's action on cAMP-mediated activation of Rap1 and cAMP-mediated mitogenesis depends on the subcellular localization of Epac via its DEP domain. Disruption of the DEP-dependent subcellular targeting of Epac abolished cAMP-Epac-mediated Rap1 activation and thyroid-stimulating hormone-mediated cell proliferation, indicating that an Epac-Rap-PKA signaling unit is critical for the mitogenic action of cAMP.     

Mechanisms of cAMP-mediated gene induction: examination of renal epithelial cell mutants affected in the catalytic subunit of the cAMP-dependent protein kinase

The precise mechanistic role of the cAMP-dependent protein kinase (cAMP-PK) in cAMP-mediated gene induction remains unclear. Renal epithelial cell mutants were compared to the LLC-PK1 parental cell line for induction of the cAMP-responsive urokinase-type plasminogen activator (uPA) gene, as quantitated by the technique of mRNA solution hybridization. The FIB4 and FIB6 mutants, which possess less than 10% parental cAMP-PK catalytic (C) subunit activity, showed markedly diminished uPA mRNA induction in response to agents elevating intracellular cAMP such as the cAMP analogue 8-bromo-cAMP and the adenylate cyclase-stimulating hormones vasopressin and calcitonin. In contrast, the mutant cells responded to a similar or greater extent than the parental cells in terms of uPA mRNA induction following treatment with the Ca2+/phospholipid-dependent protein kinase activator phorbol 12-myristate 13-acetate (PMA). Elevation of intracellular cAMP was found to induce a translocation of the cAMP-PK C subunit from the perinuclear Golgi region to the nucleus in both parental and mutant cell lines, as shown by immunocytochemical techniques. Results argue for the role of the cAMP-PK C subunit activity and possibly nuclear translocation of the C subunit in cAMP-mediated uPA induction, which is mechanistically distinct from the PMA-stimulated response.     

cAMP-dependent protein kinase and phosphoproteins in mammalian mitochondria. An extension of the cAMP-mediated intracellular signal transduction

Evidence has been obtained for the occurrence of a cAMP-dependent serine protein kinase associated with the inner membrane/matrix of mammalian mitochondria. The catalytic site of this kinase is localized at the inner side of the inner membrane, where it phosphorylates a number of mitochondrial proteins. One of these has been identified as the AQDQ subunit of complex I. cAMP-dependent phosphorylation of this protein promotes the activity of complex I and mitochondrial respiration. A 5 bp duplication in the nuclear gene encoding this protein has been found in a human patient, which eliminates the phosphorylation site. PKA anchoring proteins have recently been identified in the outer membrane of mammalian mitochondria, which could direct phosphorylation of proteins at contact sites with other cell structures.     

A phenylalanine in peptide substrates provides for selectivity between cGMP- and cAMP-dependent protein kinases.

Bovine lung cGMP-binding cGMP-specific phosphodiesterase (cG-BPDE) is a potent and relatively specific substrate for cGMP-dependent protein kinase (cGK) as compared to cAMP-dependent protein kinase (cAK) (Thomas, M. K., Francis, S. H., and Corbin, J. D. (1990) J. Biol. Chem. 265, 14971-14978). A synthetic peptide, RKISASEFDRPLR (BPDEtide), was synthesized corresponding to the sequence surrounding the phosphorylation site in cG-BPDE. BPDEtide retained the cGK/cAK kinase specificity demonstrated by native cG-BPDE: the apparent Km of BPDEtide for cGK was 5-fold lower than that for cAK (Km = 68 and 320 microM, respectively). Vmax values were 11 mumol/min/mg for cGK and 3.2 mumol/min/mg for cAK. The peptide was not phosphorylated to a measurable extent by protein kinase C or by calcium/calmodulin-dependent protein kinase II. Thus, the primary amino acid sequence of the peptide substrate was sufficient to confer kinase specificity. Studies in crude tissue extracts indicated that BPDEtide was the most selective peptide substrate documented for measuring cGK activity. Peptide analogs of BPDEtide were synthesized to determine the contribution of specific residues to cGK or cAK substrate specificity. Substitution of a Lys for the amino-terminal Arg did not reduce cGK/cAK specificity; neither did the exchange of an Ala for the non-phosphorylated Ser nor the removal of the 3 carboxyl-terminal residues. A truncated BPDEtide (RKISASE) served equally well as substrate (Km approximately 90 microM) for both kinases. However, restoration of the Phe, to yield RKISASEF, reproduced the original cGK/cAK specificity for BPDEtide (Km = 120 and 480 microM, respectively), primarily by decreasing the affinity of cAK. Addition of a carboxyl-terminal Phe to the peptide RKRSRAE (derived from the sequence of the cGK phosphorylation site in histone H2B) or to the peptide LRRASLG (derived from the sequence of the cAK phosphorylation site in pyruvate kinase) also improved the cGK/cAK specificity by decreasing the affinity of cAK. These data suggested that the Phe in each substrate tested is a negative determinant for cAK.     

Parietal cell protein kinases. Selective activation of type I cAMP-dependent protein kinase by histamine

cAMP-dependent protein kinases have been characterized in parietal cells isolated from rabbit gastric mucosa. Both Type I and Type II cAMP-dependent protein kinase isozymes are present in these cells. Type II isozymes were detected in 900, 14,000, and 100,000 X g particulate fractions as well as 100,000 X g cytosolic fractions; Type I isozymes were found predominately in the cytosolic fraction. When parietal cells were stimulated with histamine, an agent that elevates intracellular cAMP content and initiates parietal cell HCl secretion, cAMP-dependent protein kinase activity was increased in homogenates of these cells as measured by an increase in the cAMP-dependent protein kinase activity ratio. Histamine activation of cAMP-dependent protein kinase was correlated with parietal cell acid secretory responses which were measured indirectly as increased cellular uptake of the weak base, [14C]aminopyrine. These results suggest that cAMP-dependent protein kinase(s) is involved in the control of parietal cell HCl secretion. The parietal cell response to histamine may be compartmentalized because histamine appears to activate only a cytosolic Type I cAMP-dependent protein kinase isozyme, as determined by three different techniques including 1) ion exchange chromatography; 2) Sephadex G-25 to remove cAMP and allow rapid reassociation of the Type II but not the Type I isozyme; and 3) 8-azido-[32P]cAMP photoaffinity labeling. Forskolin, an agent that directly stimulates adenylate cyclases, was found to activate both the Type I and Type II isozymes. Several cAMP-dependent protein kinases were also detected in parietal cell homogenates, including a Ca2+-phospholipid-sensitive or C kinase and two casein kinases which were tentatively identified as casein kinase I and II. At least two additional protein kinases with a preference for serine or lysine-rich histones, respectively, were also detected. The function of these enzymes in parietal cells remains to be shown.     

Different mechanisms for the induction of acetylcholinesterase in neuroblastoma cells.

Neuroblastoma cells, incliding clones selected for resistance to dibutyryl-cAMP or for their ability to survive and multiply at 40°C, were used to study differences in the induction of acetylcholinesterase activity by dibutyryl-cAMP and 5-bromodeoxyuridine. In nonselected neuroblastoma cells both of these compounds induced this enzyme activity. Actinomycin D inhibited induction by 5-bromodeoxyuridine but did not inhibit inducion by dibutyryl-cAMP. Enzyme activity in dibutyryl-cAMP-resistant cells was induced by 5-bromodeoxyuridine and not by dibutyryl-cAMP. In the temperature-resistant cells, induction by 5-bromodeoxyuridine was lower at 40 than at 37°C and induction by dibutyryl-cAMP was higher at 40 than at 37°C. This difference in induction at the two temperatures was associated with a higher inhibition of cell multiplication at 40°C by both compounds. The results indicate that 5-bromodeoxyuridine and dibutyryl-cAMP induce acetylcholinesterase activity in neuroblastoma cells by different mechanisms.     

Two different intrachain cAMP binding sites of cAMP-dependent protein kinases

The regulatory subunits of both isozymes of cAMP-dependent protein kinase bind 2 mol of cAMP/mol of monomer. cAMP dissociation studies indicate similar cAMP binding behavior for each isozyme. Each has two different intrachain cAMP binding components present in approximately equal amounts and the rate of cAMP dissociation is 5- to 10-fold slower from one site (Site 1) than from the other (Site 2). Equilibrium [3H]cAMP binding is inhibited by several competing cyclic nucleotides. Following equilibrium binding using saturating [3H]cAMP in the presence of competing nucleotide, the pattern of release of [3H]cAMP, monitored in the presence of an excess of nonradioactive cAMP, suggests site-specific selectivity of some of the cyclic nucleotides. As compared with cAMP, cIMP prefers Site 2 for both regulatory subunits, whereas N6, O2-dibutyryl-cAMP shows a similar preference only with isozyme II regulatory subunit. 8-Bromo-cAMP, 8-bromo-cGMP, and 8-azido-cAMP prefer Site 1 of both proteins. The results indicate that for each isozyme the two intrachain binding sites have different analogue specificities and cAMP dissociation rates. Site 1 or Site 2 of one isozyme has a similar but not identical cyclic nucleotide specificity and cAMP dissociation rate to the corresponding site of the other isozyme.