Interaction of a casein kinase (G-type) with a specific endogenous inhibitor. Possible target for the regulation of a cyclic nucleotide-independent protein kinase activity by polyamines

Eukaryotic translation-initiation factor eIF2beta binds to protein kinase CK2: effects on CK2alpha activity

eIF2 (eukaryotic translation-initiation factor 2) is a substrate and an interacting partner for CK2 (protein kinase CK2). Co-immuno-precipitation of CK2 with eIF2beta has now been observed in HeLa cells, overexpressing haemagglutinin-tagged human recombinant eIF2beta. A direct association between His6-tagged human recombinant forms of eIF2beta subunit and both the catalytic (CK2alpha) and the regulatory (CK2beta) subunits of CK2 has also been shown by using different techniques. Surface plasmon resonance analysis indicated a high affinity in the interaction between eIF2beta and CK2alpha, whereas the affinity for the association with CK2beta is much lower. Free CK2alpha is unable to phosphorylate eIF2beta, whereas up to 1.2 mol of phosphate/mol of eIF2beta was incorporated by the reconstituted CK2 holoenzyme. The N-terminal third part of eIF2beta is dispensable for binding to either CK2alpha or CK2beta, although it contains the phosphorylation sites for CK2. The remaining central/C-terminal part of eIF2beta is not phosphorylated by CK2, but is sufficient for binding to both CK2 subunits. The presence of eIF2beta inhibited CK2alpha activity on calmodulin and beta-casein, but it had a minor effect on that of the reconstituted CK2 holoenzyme. The truncated forms corresponding to the N-terminal or central/C-terminal regions of eIF2beta were much less inhibitory than the intact subunit. The results demonstrate that the ability to associate with CK2 subunits and to serve as a CK2 substrate are confined to different regions in eIF2beta and that it may act as an inhibitor on CK2alpha.

DNA binding activity of casein kinase II

Casein kinase II, an ubiquitous, oligomeric, messenger-independent protein kinase has previously been shown to concentrate in the nuclear compartment when cells are stimulated to proliferate. The present communication reports that purified mammalian CKII interacts with genomic DNA preparations in vitro. This interaction led to an apparent activation of the kinase, most likely explained by prevention of its aggregation and subsequent denaturation. Binding of CKII was optimum with double stranded DNA preparations; duplex lambda phage DNA exhibited at least two types of binding sites and the high affinity system (Kd approximately equal to 6 x 10(-13) M) represented a binding capacity of about 1 mol CKII per mol DNA. CKII-DNA interaction was stimulated in the presence of a polyamine and inhibited by heparin. Blotting experiments disclosed that DNA binds CKII through its alpha subunit. These observations are in line with the hypothesis that casein kinase II may be examined as a component in the transduction of the mitogenic signal from the cell membrane to the nucleus, in response to growth factors.

Changes in protein kinase activities in lamb adrenals at late gestation and early postnatal stages

Cytosols prepared from adrenal glands of ovine fetuses (110-144 days of gestation) and of newborn lambs (1-6 days post-partum) were analysed for their protein kinase activities. Two major peaks of casein kinase activities and two major peaks of histone kinase activities were observed in all cytosols of both fetal and neonatal adrenal glands. They were characterized as cAMP-independent casein kinases of type A and type G, and as cAMP-dependent histone kinase isoenzymes of type I and type II. The specific activity of each enzyme increased 2-fold between 118 days of gestation and 6 days post-partum. Casein kinase of the G type was 4-fold higher than casein kinase of the A type; in contrast, the mean ratio of type II to type I histone kinase activity varied between 5- and 12-fold. Infusion of ACTH1-24 (100 micrograms/day) for 5 days to 118- to 128-day-old ovine fetuses in utero increased their plasma corticosteroid levels and the responsiveness of their adrenal cells to stimulation by ACTH1-24 in vitro. In addition, such treatment doubled the specific activity of casein kinases A and G, but had no significant effect on cAMP-dependent histone kinase activities. In relation to current concepts of the role of protein kinases in adult adrenal cells, the present results suggest that casein kinase activities are involved in cell multiplication and differentiation in the fetal adrenal gland. In addition, they show that neither cytosolic histone kinase of type I nor that of type II is likely rate-limiting in the steroidogenic response to ACTH of ovine fetal adrenal cells.

Some properties of the nucleotide-binding site of troponin T kinase-casein kinase type II from skeletal muscle

Investigation of properties of skeletal muscle troponin T kinase (EC 2.7.1.37) has revealed that the enzyme belongs to the group of casein kinases of the second type. The enzyme consists of two subunits with apparent molecular weights of 44 000 and 26 000 and contains a protein with molecular weight of 39 000, which is probably the proteolytic fragment of the 44 000 subunit. The substrate specificity of troponin T kinase was tested, using 20 analogs of the nucleotide. The enzyme has a low substrate specificity toward the purine base and uses both ATP and GTP as substrates. Modification of the ribose ring does not influence the enzyme interaction with the nucleotide; however, the cleavage of ribose leads to a decrease of the enzyme-nucleotide interaction. Elimination of the gamma-terminal phosphate or its modification by bulky hydrophobic radicals do not affect this interaction. A comparison of the Ki values for different analogs suggests that the interaction of troponin T kinase with the nucleotide occurs via the binding of the purine base and the beta-phosphate group of the analog.

Protein kinases in normal human blood cells

Protein kinases active on basic and acidic artificial substrates were investigated in normal human erythrocytes, platelets, polymorphonuclear and mononuclear cells. These two types of protein kinases were partially purified by affinity chromatography, then assayed for their enzymatic activity using [gamma-32P]ATP or GTP as phosphoryl donor. Partially purified kinases active on acidic substrates were subjected to high-performance liquid chromatography (HPLC). Protein kinases active on basic substrates were analyzed by cellulose acetate electrophoresis of crude cellular extracts and the influence of 3'5' cyclic AMP was studied. Three forms of casein-phosvitin kinases could be distinguished according to their molecular weight (165 K, 38 K, and 31 K). The 165 K species, in contrast to the light species, can use GTP instead of ATP as phosphoryl donor and corresponds to the "so-called" casein kinase 2. This form is very sensitive to proteolysis and, when partial purification is performed without the addition of various antiproteolytic agents, it is degraded into 120-135 K and 105-115 K active species; this artefactual degradative process is especially active in platelet extracts. As many as eight different active bands of histone and protamine kinases can be separated by cellulose acetate electrophoresis, several of them being stimulated by cyclic AMP. Isozymic patterns of protein kinases, levels of activity on the different substrates, and utilization of ATP and GTP were found to be specific for each cell type. These results suggest the possibility of using protein kinases as markers for cell differentiation.

Polyamine-mediated protein phosphorylations: a possible target for intracellular polyamine action

Polyamines are well-known ubiquitous components of living cells. Although these polycations have been implicated in the regulation of major cellular functions such as DNA, RNA and protein synthesis occurring during cellular proliferation and/or differentiation processes, their mechanism of action at the molecular level has remained obscure. On the other hand, protein phosphorylation has emerged as a regulatory process of prime importance in cellular regulation. Data have recently been presented suggesting that polyamines may express at least part of their biological action through an effect upon selective protein phosphorylation systems. Two types of polyamine-sensitive protein kinases have been characterized in the last few years. The best known in molecular terms is the widespread casein kinase G (also termed casein kinase II), which represents a multifunctional protein kinase, at present classified as a messenger-independent activity. The other is a polyamine-dependent nuclear ornithine decarboxylase kinase characterized in Physarum polycephalum and several mammalian tissues. Both protein kinases are activated by polyamines in vitro at concentrations compatible with a physiological role, by a mechanism which most likely also involves an effect through the protein substrate conformation. Preliminary evidence suggests that both kinases may be implicated in the regulation of DNA-dependent RNA polymerase activities, although several other potential substrates have been suggested for casein kinase G. Another suggestion is that these kinases may also participate in the post-translational regulation of ornithine decarboxylase, the rate-limiting step in the polyamine biosynthetic pathway. A novel class of protein kinase activities may thus be defined as polyamine-mediated phosphorylation systems for which polyamines may function as intracellular messenger. Although their biological significance remains to be fully established, especially with regard to the definition of their specific intracellular target(s) and subsequent biological functions, these systems will be interesting to consider in future studies aimed at understanding the role of polyamines in cell regulation.

Reversibility of the phosphate transfer between ATP and phosphoproteins catalysed by a cyclic nucleotide independent (G type) casein kinase

Purified casein kinase G was found able to catalyse the synthesis of [gamma-32P]ATP in the presence of ADP, phosphocasein (previously 32P-labeled by the forward kinase reaction) and magnesium. Apparent Km values of approx. 0.5 mM for phosphocasein and 7.5 mM for ADP were calculated, these values indicating low affinities for the substrates as compared to those exhibited for casein and ATP in the forward reaction. The reverse casein kinase G activity appeared to prefer ADP and GDP as phosphate acceptors. Whereas the casein kinase G reverse reaction could be supported by casein, phosvitin and histone previously phosphorylated by the enzyme, the same proteins could not serve as a phosphate source when previously phosphorylated by the cAMP-dependent protein kinase. Forward and reverse casein kinase G reactions exhibited different optimal pH values (8.5 and 7.2, respectively) and a different sensitivity to Mg2+. Spermine, which activated the kinase activity, blocked the reverse reaction at millimolar concentrations. Although the biological significance of the casein kinase G reverse activity remains to be assessed in intact cell, the process may be useful as a tool in the characterization of phosphorylatable sites in phosphoproteins.

Identification of the catalytic subunit of an oligomeric casein kinase (G type). Affinity labeling of the nucleotide site using 5'-[p-(fluorosulfonyl)benzoyl]adenosine

Identification of the catalytic subunit of a G type [using guanosine 5'-triphosphate (GTP) as well as adenosine 5'-triphosphate (ATP) as phosphate donor], oligomeric, cyclic nucleotide independent casein kinase purified from bovine lung was carried out after reaction with 5'-[p-(fluorosulfonyl)-benzoyl]adenosine (FSBA) and isolation of the subunit components of the enzyme. FSBA exhibited the major characteristics of an affinity label reacting at the nucleotide (ATP, GTP) site of the casein kinase. FSBA acted as a competitive inhibitor of ATP (and GTP), led to complete inactivation of the enzyme in a reaction showing two kinetic steps, and became irreversibly bound to the protein. After being labeled with FSBA, the casein kinase (apparent molecular weight of 140 000) was separated into its two monomeric components of apparent molecular weights 38 000 (alpha) and 27 000 (beta), respectively, after sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Use of radioactive FSBA showed that specific affinity labeling was limited to the alpha casein kinase subunit. This result was in agreement with the fact that casein kinase activity was found associated with the alpha monomer after electrophoretic separation of the alpha and beta subunits. It may thus be concluded that the largest (alpha) subunit contains the catalytic site of the casein kinase G. Electrophoretic analysis of purified protein kinase under denaturing conditions suggested an alpha 3 beta 2 combination for an apparent molecular weight of 130 000-140 000. However, a maximum of 2 mol of FSBA could be specifically bound to the alpha subunit per mol of enzyme, with a concomitant complete inactivation. These data would be in agreement with an alpha 2 beta 2 subunit composition for casein kinase G, as proposed by other research groups for a similar type of protein kinase of different sources. These observations suggest that the alpha subunits are functionally similar, each of them containing a nucleotide (ATP, GTP) binding site. The possible role of the beta subunit in the enzyme activity remains to be established.

Catalytic and molecular properties of a highly purified G type casein kinase from bovine lung tissue

A preparation procedure has been worked out to obtain a highly purified G type (using GTP as well as ATP) casein kinase from large quantities of bovine lung tissue. It included ion-exchange (DEAE and phosphocellulose) and affinity (casein and ATP-Sepharose) chromatography combined with a flocculation step, and yielded an apparently homogeneous preparation with a 16% yield and a purification factor of more than 1400. The purified lung casein kinase used GTP (Km 16 microM) almost as well as ATP (Km 6.7 microM) and exhibited the major catalytic properties of the casein kinase G previously described in bovine adrenal cortex (Cochet, C., Job, D., Pirollet, F. and Chambaz, E.M. (1981) Biochim. Biophys. Acta 658, 191-201). Mg2+ (30-50 mM) and spermine (2 mM) were potent activators of lung casein kinase G activity, whereas the enzyme was inhibited by heparin and quercetin. The purified enzyme underwent self-phosphorylation in the presence of ATP or GTP, serine being the only target amino acid under these conditions, whereas both serine and threonine were phosphorylated by the enzyme in casein. Lung casein kinase G exhibited an apparent molecular weight between 140 000-160 000 upon gel filtration and appeared formed by the association of two different subunits upon SDS-polyacrylamide gel electrophoresis. The two subunits of Mr 38 000 (alpha) and 27 000 (beta) exhibited a 2:1 ratio upon quantitative scanning, suggesting an alpha 3 beta 2 combination in the oligomeric native enzyme structure. Peptide mapping of the two isolated subunits following 125I-labeling and papain digestion did not disclose any common fragment. The casein kinase catalytic activity was found associated with the alpha (38 kDa) enzyme subunit after recovery from gel electrophoresis in the presence of SDS, whereas the 27 kDa (beta) subunit was the major target of the enzyme self-phosphorylation reaction. alpha and beta subunits appeared strongly associated in the oligomeric enzyme and the possible role of the beta subunit in the casein kinase G activity remains to be examined. The purified casein kinase G, which can be obtained by the present procedure, should facilitate the study of the biological significance of this phosphorylation system in the intact cell.

Cyclic nucleotide independent casein kinase (G type) in bovine adrenal cortex: purification and properties of two molecular forms

Two soluble cyclic nucleotide independent protein kinase (ATP: protein-phosphotransferase, EC 2.7.1.37) activities have been purified from bovine adrenal cortex cytosol. Both purified enzymes exhibit the best affinity for acidic substrates such as casein and can use GTP as well as ATP as phosphoryl donor. They can thus be classified as casein kinase of the G type as previously proposed (Cochet C. et al., (1980) Endocrinology 106, 750-757). Whereas the two moieties could be separated using their different affinities toward a phosphocellulose resin, both purified enzymes appeared indistinguishable on the basis of several molecular and catalytic properties. Both G type casein kinase moieties have an identical sedimentation behavior (5.5 S in the presence of 0.5 M NaCl), yield similar patterns upon electrophoresis under denaturing conditions with three major protein components (42 000, 38 000 and 27 000), and show an ability to undergo self-phosphorylation mostly on the 27 000 component. Both enzymes have the same protein and nucleotide (ATP and GTP) substrate specificity, show similar increases in activity in the presence of polyamines and Mg2+ (optimum at 50 mM) and similar inhibition by NaCl above 0.2 M. The only difference between the two forms of casein kinase (i.e., affinity for phosphocellulose) could not be explained by a different degree of self-phosphorylation or by a limited proteolytic process during handling and purification. These results suggest that the two active moieties may represent isoenzymatic forms of the G type casein kinase activity in bovine adrenal cortex cytosol.

The ATP substrate site of a cyclic-nucleotide-independent protein kinase from porcine liver nuclei

The ATP substrate site of a second messenger-independent protein kinase of the type NII from porcine liver nuclei was mapped using a series of 30 ATP derivatives with modifications at the base, ribose or triphosphate moiety. Ki values for these derivatives were determined by competition with [gamma-32P]ATP; they range from 4 microM to 1.5 mM. For a comparison with data previously reported for the catalytic subunit of cAMP-dependent protein kinase I from rabbit skeletal muscle, the Ki values were transformed into delta delta values. These values are related to the Ki value of unsubstituted ATP and indicate the decrease of affinity caused by the different substitutions. With both enzymes the major binding affinity is derived from the interaction of the adenine base. The contributions of the two ribosyl OH groups are marginal and the triphosphate moiety interacts most strongly with its beta-phosphoryl group. Between the two enzymes the most striking differences, however, were observed for the specificity of the nucleobase interaction. While an unmodified N-6 amino group is required in the case of the cAMP-dependent protein kinase, the nuclear enzyme seems to tolerate extensive modification at this position, such as the introduction of a keto group or a bulky benzyl residue. Obviously, the ATP site of the nuclear kinase has an open cleft next to the N-6 of the adenine and binding of the adenine occurs by hydrophobic interaction without the formation of hydrogen bonds to any of the adenine nitrogens.