Phosphorylation Of Phosphorylase Kinase Research Articles

A new heparin-inhibited and polyamine-activated protein kinase from bovine kidney

Two casein kinases, casein kinase-1 (CK-1) and casein kinase-2 (CK-2), have been characterized from many sources. In this study we describe the properties of a third casein kinase, designated casein kinase-3 (CK-3). CK-3 (Mr 32 000) is readily separated from CK-2 by gel filtration and from CK-1 by hydroxyapatite chromatography. CK-3 phosphorylates several proteins, including phosphorylase kinase. Phosphorylation of phosphorylase kinase by CK-3 results in a 10-fold enzyme activation. CK-3 is activated by spermine and inhibited by heparin, ADP, and divalent metal ions (Mn2+, Zn2+). Heparin inhibition of the kinase is reversed by spermine. The physical and regulatory properties of CK-3 are very similar to CK-1, suggesting that these kinases may be closely related.

Inhibition by calmodulin of the cAMP-dependent protein kinase activation of phosphorylase kinase.

Calmodulin is shown to inhibit both the activation and phosphorylation of phosphorylase kinase by cAMP-dependent protein kinase. Maximal inhibition of both processes was approximately 66% at the highest calmodulin concentration tested (5.5 microM). It was found that the inhibition of phosphorylation was calcium-dependent, reversible by trifluoperazine, and specific for the beta subunit of phosphorylase kinase with no significant inhibition of phosphorylation of the alpha subunit. This inhibitory activity of calmodulin appears to be due to an interaction between calmodulin and the substrate, phosphorylase kinase. This finding implies either that the site of exogenous calmodulin interaction with phosphorylase kinase is at the beta subunit or that this interaction results in a conformational change of phosphorylase kinase that inhibits the interaction between cAMP-dependent protein kinase and the beta subunit of phosphorylase kinase. The beta subunit may contain a regulatory site that is recognized by either protein kinase or calmodulin. These findings further substantiate the role of the beta subunits in the activation of phosphorylase kinase and provide an additional example of substrate-directed control of phosphorylation.

The Hormonal Control of Glycogen Metabolism

Inhibitor-1 has been shown to be phosphorylated in skeletal muscle in vivo. In normal fed animals the degree of phosphorylation was 31 +/- 7% and this value increases to 70 +/- 12% following an intravenous injection of adrenaline. The results imply that the phosphorylation of inhibitor-1 may be equally as important as the phosphorylation of phosphorylase kinase in elevating the levels of phosphorylase a. The role of inhibitor-1 in metabolism is discussed.

The modulator-dependent protein kinase. A multifunctional protein kinase activatable by the Ca2+-dependent modulator protein of the cyclic nucleotide system.

A protein kinase which depends on the simultaneous presence of Ca2+ and the modulator protein for its histone phosphorylation activity has been demonstrated in rabbit skeletal muscle and partially purified. The purified enzyme was not activated by cAMP, cGMP, or incubation with trypsin. Nor was the enzyme inhibited by the protein inhibitor of cAMP-dependent protein kinase. In addition to histone, myosin light chains and phosphorylase kinase served as substrates for the protein kinase, and their phosphorylation also depended on the presence of Ca2+ and the modulator protein. The phosphorylation of phosphorylase kinase was accompanied with a marked activation of the enzyme. The results suggest that the protein kinase has multiple functions and may be involved in the mediation of Ca2+ effects in many biological processes. It is proposed that this enzyme be designated as the modulator-dependent protein kinase. The modulator-dependent protein kinase may be identical to the myosin light chain kinase; chicken gizzard light chain kinase has been shown activatable by the modulator protein (Dabrowska, R., Sherry, J. M. F., Aramatorio, D. K., and Hartshorne, D. J. (1978) Biochemistry 17, 253-258).

Effect of Mg2+ concentration on the cAMP-dependent protein kinase-catalyzed activation of rabbit skeletal muscle phosphorylase kinase.

Phosphorylase kinase was found to be activated and phosphorylated at 10mM Mg2+ by the cAMP-dependent protein kinase-catalyzed reaction ot much higher levels than observed previously when reactions were carried out in 1 to 2 mM Mg2+ (Cohen, P. (1973) Eur. J. Biochem. 34, 1; Hayakawa, T., Perkin, J.P., and Krebs, E.G. (1973) Biochemistry 12, 574). That the reaction at 10 mM Mg2+ is protein kinase-catalyzed is supported by several observations: (a) the reaction is facilitated by the addition of protein kinase; (b) the reaction depends on cAMP when protein kinase holoenzyme is uded; (c) the reaction is not inhibited by 1 mM ethylene glycol bis(beta-aminoethyl ether) N,N'-tetraacetate which is known to inhibit autoactivation and autophosphorylation of phosphorylase kinase; and (d) the protein inhibitor of protein kinase inhibits this reaction. The phosphorylation and activation of phosphorylase kinase seem to occur in two phases. At low Mg2+ only the first phase is manifested and involves the incorporation of 2 mol of phosphate, 1 mol into each of Subunits A and B. At high Mg2+ additional sites are phosphorylated almost exclusively on Subunit A, with phosphate incorporation approaching the final level of 7 to 9 mol. Enzyme activity at high Mg2+ is 2 to 3 times higher than that observed when activation is studied at low Mg2+. The observation that both casein and type II histone are phosphorylated to the same extent at 1 mM and 10 mM Mg2+ suggested that high Mg2+ may be altering the conformation of phosphorylase kinase thus rendering more phosphorylation sites accessible to protein kinase. Since the phosphorylation of phosphorylase kinase by either the protein kinase-catalyzed or autocatalytic reaction can result in the incorporation of 7 to 9 mol of phosphate, the finding that only about seven sites become phosphorylated by both mechanisms acting together suggest that activation by these two mechanisms may involve common phosphorylation sites.

Separation of Two Phosphorylase Kinase Phosphatase Activities in Rabbit Skeletal Muscle

The 20-40-fold activation of phosphorylase kinase catalysed by cyclic AMP-dependent protein kinase in the presence of cyclic AMP, ATP and Mg2+ ions in vitro (Walsh et al., 1968) represents one of the events leading to the stimulation of skeletal muscle glycogenolysis by adrenalin. However, recent experiments have demonstrated that this reaction is far more complex than had originally been supposed, and the information which led to this conclusion is summarized below. (1) Phosphorylase kinase was shown to be formed from three types of polypeptide chain, termed a, band y, and the smallest active species of the enzyme had the structure (aby)4. Incubation of phosphorylase kinase with the cyclic AMP-dependent protein kinase and ATP-Mg led to phosphorylation at two unique serine residueslabyunit. The fist mol rapidly entered the /I subunit and paralleled the rise in activity, whereas the second mol was incorporated five times more slowly into the a subunit after a short lag period, without any further effect on activity (Cohen, 1973; Cohen et al., 1975). (2) After phosphorylation of phosphorylase kinase in vitro, the preparation was shown to contain trace endogenous phosphorylase kinase phosphatase activity capable of dephosphorylating both the a and b subunits (Cohen & Antoniw, 1973). (3) The phosphorylation of the a subunit was found to enhance the rate of dephosphorylation of the b subunit and accompanying inactivation of the enzyme by a factor of 50. The results showed that the reversible activation of phosphorylase kinase was associated exclusively with the reversible phosphorylation of the subunit, but that there might also be a novel mechanism for actively regulating the reversal of hormonal activation, associated with the phosphorylation of the a subunit (Cohen & Antoniw, 1973). (4) Intravenous injection of adrenalin was shown to activate phosphorylase kinase 5-10-fold in skeletal muscle in vivo, and was accompanied by phosphorylation of the same two sites on the a and subunits that were labelled in vitro (Yeaman & Cohen, 1974,1975). This showed that the two functional changes associated with the phosphorylation of each site were control mechanisms that operated in vivo. Here we present an extension of these findings and demonstrate that there are in fact two distinct phosphorylase kinase phosphatases in skeletal muscle, one specific for the a subunit, and one specific for the

Demonstration of the Control of Rabbit Skeletal-Muscle Phosphorylase Kinase Activity in vivo by Adenosine 3′5′-Cyclic Monophosphate-Dependent Protein Kinase

Rabbit skeletal-muscle phosphorylase kinase has an activity at physiological pH (6.8) only 1-2% of that observed at the pH optimum (8.2), but may be activated @fold at pH6.8 after incubation with cyclic AMP-dependent protein kinase, cyclic AMP and ATP-MgZ+ in vitro (Walsh et al., 1971 ; Cohen, 1973). Two molecules of covalently bound phosphate are incorporated during the reaction. The phosphorylation at the primary site on the 8-subunit parallels the increase in activity (Hayakawa et al., 1973; Cohen, 1973), whereas the slower phosphorylation of a second site, on the a-subunit, controls the rate of dephosphorylation of the 8-subunit catalysed by phosphorylase kinase phosphatase (Cohen & Antoniw, 1973). The specificity of this reaction in vitro has been confirmed by the isolation of a nine-residue tryptic phosphopeptide from the 8-subunit (Cohen, 1974) and a 39-residue tryptic phosphopeptide from the a-subunit (P. Cohen, unpublished work). Although phosphorylase kinase activity previously has been shown to increase after an intravenous injection of adrenalin (Posner et al., 1965; Drummond et al., 1969), this need not necessarily be a consequence of the activation of cyclic AMP-dependent protein kinase. Phosphorylase kinase activity can also be stimulated by two further mechanisms in vitro, by a calcium-dependent phosphorylation of phosphorylase kinase by itself (Walsh et al., 1971) and by proteolysis (Huston & Krebs, 1968). In the present paper, phosphorylase kinase was activated by an intravenous injection of adrenalin, and purified through the (NH4)2S04 step by the procedure previously described for the non-activated enzyme (Cohen, 1973). The pH6.8/8.2 activity ratio in the muscleextract was0.22 incontrast with thevalueof0.03f0.01 observedin theabsenceof adrenalin. NaF (50m~) was included in the purification to inhibit phosphorylase kinase phosphatase, but there was nevertheless a slight decline of the activity ratio to 0.16 during the 8 h isolation. Phosphorylase kinase (1 .Oms), which had been phosphorylated in vitro at both the aand 8-subunit sites, was then added to phosphorylase kinase activated tn vivo (200mg), and the aand 8-subunit phosphopeptides were isolated by using the trace of radioactivemarker protein to locate the position of the peptides and to calculate the yield at each step.

Control of Phosphorylase Activity in a Muscle Glycogen Particle: II. ACTIVATION BY CALCIUM

Abstract The regulation of enzymatic activity of several enzymes involved in glycogen breakdown has been investigated in a skeletal muscle fraction containing a protein-glycogen complex and elements of the sarcoplasmic reticulum. In this fraction, phosphorylase is entirely in its inactive b form since phosphorylase kinase itself is totally inactive while phosphorylase phosphatase is fully active. Addition of Mg-ATP and Ca2+ triggers an immediate of phosphorylase (b to a conversion) resulting from kinase activation; no occurs with Mg-ATP alone. As soon as all the ATP has been consumed, phosphorylase a is rapidly reconverted to the b form by the phosphatase, and the over-all process can be repeated many times by successive readditions of ATP; this reaction cycle is referred to as flash activation of phosphorylase. The Ca2+ of kinase is reversed by chelation of the metal ion and, therefore, not the result of proteolytic attack by the calcium-dependent kinase-activating factor. Half-maximum of kinase in this system requires 2 x 10-6 m free Ca2+ (in contrast to approximately 10-7 m calcium for purified kinase solutions), i.e. the same Ca2+ concentration needed to trigger muscle contraction. Activation by Ca2+ resulted in a 13-fold increase in affinity of phosphorylase kinase for phosphorylase b. No evidence was obtained that it was mediated or accompanied by a phosphorylation of phosphorylase kinase. Only slight (20%) and delayed of endogenous phosphorylase b was produced by Mg-ATP and 10-5 m cyclic adenosine 3',5'-monophosphate in the absence of added Ca2+, although a 6-fold increase in kinase activity was measured in the usual assay system (at high dilution of phosphorylase kinase and in the presence of purified phosphorylase b). Disruption of the protein-glycogen complex by α-amylase digestion increased the affinity of phosphorylase kinase for Ca2+ 10-fold to the same level observed with the purified enzyme. Readdition of glycogen reversed this effect. Addition of 1 mm glucose-6-P suppressed the flash activation of phosphorylase both by a direct inhibition of the phosphorylase kinase reaction and, indirectly, by increasing the utilization of ATP. Furthermore, activated phosphorylase produced during the reaction was itself partially inhibited by glucose-6-P indicating that phospho-dephospho hybrids (rather than fully phosphorylated phosphorylase a) had been produced. Such hybrids were generated only in the intact (not in the disrupted) protein-glycogen complex.

Inactivation of Glycogen Synthetase and Activation of Phosphorylase Kinase by Muscle Adenosine 3',5'-Monophosphate-dependent Protein Kinases

Rabbit skeletal muscle glycogen synthetase I has been purified and obtained essentially free of phosphorylase, phosphorylase kinase, and glycogen synthetase kinase. Using the purified glycogen synthetase as substrate, it was determined that two separable adenosine 3',5'-monophosphate (cyclic AMP)-dependent protein kinase fractions from skeletal muscle each catalyze the conversion of glycogen synthetase I to glycogen synthetase D; activity changes during the reaction correlate closely with the uptake of phosphate by glycogen synthetase. These same protein kinase fractions also catalyze the cyclic AMP-dependent phosphorylation and activation of phosphorylase kinase. The specific activities of the two protein kinase fractions towards phosphorylation of glycogen synthetase and phosphorylase kinase were enriched equally during purification. Further evidence that phosphorylase kinase phosphorylation and glycogen synthetase phosphorylation are common activities of the same enzyme was obtained by studying heat inactivation of the protein kinase, destabilization by cyclic AMP, inhibition by a protein inhibitor, and the cyclic nucleotide specificity. Evidence is presented that the cyclic AMP-dependent protein kinase acts directly on glycogen synthetase I and not through a second kinase as occurs in the phosphorylase activation system. Rabbit skeletal muscle glycogen synthetase I sediments as a single peak with an s20, w of 14.3. Preliminary studies indicate that the enzyme probably exists as a tetramer composed of subunits having a molecular weight between 90,000 and 100,000. On conversion of the I form to the D form, the enzyme incorporates 1 mole of phosphate per 91,000 g of protein.

An Adenosine 3′,5′-Monophosphate-dependant Protein Kinase from Rabbit Skeletal Muscle

Abstract A protein kinase that catalyzes an adenosine 3',5'-monophosphate (cyclic AMP)-dependent phosphorylation of casein and protamine has been purified from rabbit skeletal muscle. The Km values of cyclic AMP for these reactions are 1 x 10-7 and 6 x 10-8 m, respectively. The protein kinase markedly increases the rate of the cyclic AMP-dependent activation and phosphorylation of phosphorylase kinase by ATP.