Rabbit skeletal muscle contains two phosphorylase kinase isozymes arising from the two different muscle types, the white and the red muscle (Jennissen, H. P.,.
Communication
THE JUURNAI.OF BIOLOGICALCHEMISTRY
Vol. 255. N o 23. Issue of December 10, pp. 11102-11105, 1980 Printed In U.S. A.
long beenestablished to bea Ca“-dependent enzyme in Differential Interactionof Rabbit skeletal muscle (5, 6). In addition, Cohen et al. (4) have also Skeletal Muscle Phosphorylase shown that phosphorylase kinase may be further activatedby Kinase Isozymes with Calmodulin* calmodulin in the presence of Ca2+.These observations have (Received for publication, September 2, 1980)
Rajendra K. Sharma, StanleyW. Tam*, David M. Waismang, and JerryH. Wang From the Departmentof Biochemistry, Faculty of Medicine, University of Manitoba, Winnipeg, Manitoba, CanadaR3EOW3
Rabbit skeletal muscle contains two phosphorylase kinase isozymes arising from the two different muscle types, the white and the red muscle (Jennissen, H. P., and Heilmeyer, L. M. G. (1974) FEBS Lett. 42, 77-80). The two phosphorylase kinase isozymes could be separated by affinity chromatography on a calmodulinSepharose 4B column. In media containing high concentrations of Ca2+,about 2 m, both isozymes were bound to the affinity column. When the column was eluted with a buffer containing 0.2 m~ Ca2+,the red muscle isozyme was eluted, whereas white muscle isozyme was eluted from the column byan ethylene glycol bisu-aminoethyl ether) N,N,N’,lV”tetraacetic acid. The purified white muscle isozyme can be distinguished from the red muscle isozyme by its ability to inhibit calmodulin-dependent cyclic nucleotide phosphodiesterase. The two isozymes are also regulated differently by calmodulin. Both isozymes contain tightly bound P., Burchell, A., calmodulin as asubunit(Cohen, Foulkes, J. G., Cohen, P. T. W., Vanaman, T. C., and Nairn, A. C. (1978) FEBS Lett. 92, 287-293), which renders the enzyme sensitive to Ca2+.Only the white muscle isozyme can be activated by exogenous calmodulin.
been c o n f i i e d and extended (7-11). During the course of a study on the possible existence of calmodulin binding proteins in rabbitskeletal muscle, we noticed that phosphorylasekinase is one of theproteins exhibiting specific and reversible bindingto calmodulin-Sepharose 4B column.’ The present studywas initiated to characterize this interaction. The result indicates that the two different isozymes of phosphorylase kinase exhibit differential interactions with the calmodulin-Sepharose 4B column. This differential interaction is reflected in the differential regulation of the isozymes by calmodulin. EXPERIMENTAL PROCEDURES
The nonactivated rabbit skeletal muscle phosphorylase kinase was prepared by the method of Cohen (1).The kinase preparation was stored frozen at -70°C in small aliquots. Phosphorylase b was prepared from frozen rabbit skeletal muscle according to the method of Fischer and Krebs (12). Calmodulin and calmodulin-deficient cyclic nucleotide phosphodiesterase were prepared from bovine brain according to procedures described previously (13,14).Calmodulin-Sepharose 4B conjugates were prepared essentially as described (14). Phosphorylase kinase activity was determined byfollowing incorporation into phosphorylase as described (15). One unit of phosphorylase kinase activity was defined as the amount of the enzyme catalyzing the incorporation of 1 pmol of phosphate into phosphorylase blmin at 30°C. Inhibition of cyclic nucleotide phosphodiesterase by proteins capable of binding calmodulin was carried out as previously described (16). Protein concentration was assayed by the method of Bradford (17) using bovine serum albumin as the standard. RESULTS
Fig. 1 shows that, when a sample of purified rabbit skeletal muscle phosphorylase kinase was applied to a calmodulinSepharose 4B column in the presence of millimolar concen90% of the applied protein bound Rabbit skeletal muscle phosphorylase kinase has been pu- trations of Ca2+, more than rified to apparent homogeneity in two different laboratories to thecolumn. The small amount of protein that came through (Fig. 1, Peak A ) had little or nophosphorylase kinase activity (1, 2). The purified enzyme, however, contains two isozymes which originate from the two different muscle types, red and and represented a minor contaminant of the enzyme sample. white muscles (3). Early studies indicated that the phospho- When thecolumn was eluted with a buffer containing 0.2 mM Ca2+, an additional small protein peak (Fig. 1, Peak B ) aprylasekinase isozyme fromwhitemuscle containedthree types of subunits; a , /3, and y of molecular weights 145,000; peared in the eluent, which was associated with high phos128,000; and 45,000 (1-3), respectively. The red muscle phos- phorylase kinase activity.The majorityof the applied protein and phosphorylase kinase activity, however, was eluted from phorylase kinase was composed of subunits a‘, /3, and y , a’ the column by an elutionsolution containing 2.0 mM ethylene having a molecular weight of 133,000 (3). Using the purified enzyme consisting of predominantly the glycol bi@-aminoethyl ether)N,N,N’,N’-tetraacetic acid to chelate Ca2+(Fig. 1, Peak C ) . isozyme from white muscle, Cohen et al. (4) have discovered The two phosphorylase kinase fractions from the affinity recently that phosphorylase kinase contains calmodulin as an column(Fig. 1, Peaks B and C ) , along withthe original additional subunit, the S subunit, which may be the Ca”sensitizing factor in this enzyme. Phosphorylase kinase has phosphorylase kinase sample, were analyzed by 5%polyacrylamide gel electrophoresis in the presence of sodium dodecyl * This work was supported by Grant MT-2381 from the Medical sulfate (Fig. 2). The presence of protein bands corresponding Research Council and by The Muscular Dystrophy Association of to botha and a’ subunits in the electrophoretic patternof the Canada. The costs of publication of this article were defrayed in part original enzyme sample was consistent with the notion that by the payment of page charges. This article musttherefore be hereby rabbit skeletal muscle phosphorylase kinase purified by the marked “advertisement” in accordance with 18 U.S.C. Section 1734 established method (1)contained isozymes from both red and solely to indicate this fact. of a than a’ white muscle (3). The much greater quantity $ Predoctoral Fellow of The Muscular Dystrophy Association of Canada. subunit indicated that the predominant isozyme in the enzyme
0 Presentaddress,Department of Medicine, Yale University School of Medicine, New Haven, CT 06510.
11102
’ S. W. Tam and J . H. Wang, unpublished results.
Skeletal Muscle Phosphorylase Kinase Isozymes Buffer A t
2.0
0.2mM Co"
L
1.6
2 8
1.2-
Z
d
p< m
2.0mM EGTA
0.0 0.4
-
20
80
40 FRACTION NUMBER
FIG. 1. Rabbitskeletal muscle phosphorylasekinase (30 mg/ml) in Buffer A (50 IUM glycerophosphate, 10% sucrose, 2 mM EDTA, 5 m~ MgAc2,lO mM 2-mercaptoethanol) plus 2 m~ CaC12 was applied to a calmodulin-Sepharose 4B column (1.5 X 15 cm) pre-equilibrated with the samebuffer. The column was first eluted with the equilibrating buffer, then with Buffer A plus 0.2 mM Ca2+,and finally with Buffer A plus 2 mM ethylene glycol bis(Paminoethyl ether)N,N,N',N'-tetraacetic acid (EGTA),as indicated in the figure by arrows.
0 B C
'*
)Q
d-
a"
6-
11103
calmodulin. As is shown inFig. 3, both theoriginial phosphorylase kinase sample and the purified white muscle isozyme inhibited the calmodulin-dependent cyclic nucleotide phosphodiesterase reaction. Incontrast,the isozyme from red muscle did not possess inhibitory activity toward the phosphodiesterase reaction. Identical result was obtained when the phosphodiesterase reactionswere carried out in the presence of 2 mM Ca2' instead of0.1 mM Ca2+ as used in the experiment shown in Fig. 3. Unlike other calmodulin binding proteins, the inhibitory phosphorylase kinase samples did not bring the calmodulin-activated phosphodiesterase to basal its level. The reason for this is not clearat present. Since phosphorylase kinase samples appeared to activate calmodulinthe deficient phosphodiesterase to a small extent, it suggested that these samples might contain small amounts of free calmodulin. Using the enzyme preparation containing mostly the white muscle isozyme, Cohen et al. (4) have shown that phosphorylase kinase contains calmodulin as a subunit, 8 subunit. The purified white andred muscle isozymes were analyzed for the existence of 8 subunit by sodium dodecyl sulfate-gel electrophoresis on 10% polyacrylamide gel. Fig. 4 shows that both samples containedS subunit which exhibited a mobility identical with that of bovine brain calmodulin. The gels were overloaded with protein so that theS subunit could be easily visualized. Densitometric analysis of the gels indicated that the ratiosof staining intensitiesof y to S in theoriginal sample, the red muscle, and whitemuscle isozymes were 3.75,3.8, and 3.77, respectively. The 6 subunit could be isolated from phosphorylase kinase by heating of the enzyme in a boiling water bath (4). The isolated S subunit from both isozymes could activate cyclic nucleotide phosphodiesterase withidentical potency. Results presented so far suggest that isozymes of phosphorylasekinasefrom redandwhite musclesshowdifferent interaction withcalmodulin.While both isozymes contain endogenous calmodulin as S subunit, only the white muscle
7-
w -
-
.
FIG. 2. Original purified sample of phosphorylase kinase(0) and the pooled fractions B (B) and C ( C ) from the affinity chromatography (Fig. 1) were analyzed by sodium dodecyl sulfate-gel electrophoresis on5% polyacrylamide gel. About 25 pg of the protein sample was applied on each gel.Electrophoresiswas carried out at 4 mA/gel.
preparation was of the white muscle type (3). In contrast to the original sample, samples from the calmodulin-Sepharose 4B column appeared to represent pureisozymes of phosphorylase kinase. Protein Peaks B and C (Fig. 1) contained red and white muscle isozymes of phosphorylase kinase, respectively. The observation that the two isozymes of phosphorylase kinase could be separated by affinity chromatography on the calmodulin-Sepharose 4B column suggests that the two forms of the enzyme interact with calmodulin differently. Proteins which are capable of binding calmodulinhave been found to antagonize the activation of cyclic nucleotide phosphodiesterase by calmodulin (16, 18). Thus, the ability of a protein to inhibit the calmodulin-dependent cyclic nucleotide phosphodiesterase reactionis suggestive of its association with
6
N -
O.*
0
t \ 1
0
2
0
3
0
4
0
PHOSPHORYLASEKINASE
5
0
6
0
(pg/ml)
FIG. 3. Cyclic nucleotide phosphodiesterase reactions containing 0.012 units/ml of the enzyme and 40 ng/ml of calmodulin were carried out in the presence of various amounts of and ) the the original sampleof phosphorylase kinase (M affinity chromatography-purified red muscle (M and ) white muscle isozymes (A-A) of phosphorylase kinase. Or reactions of calmodulin-deficient cyclic nucleotide phosphodiesterase (0.012 unit/ml) were carried out in the presence of various amounts of original phosphorylasekinase (M red ) muscle , isozyme (M), and white muscle isozyme (A-A).
Skeletal Muscle PhosphorylaseKinase Isozymes
11104
was obtained when 2 mM instead of 0.1 mM CaC12was used on the enzyme reactions. The activation by exogenous calmodulin depends on the presence of Ca”. DISCUSSION
Jennissen and Heilmeyer (3) originally demonstrated the existence of red and white muscle-specific isozymes of phosphorylase kinase. More recently, it was shown (19, 20) that cardiac muscle phosphorylase kinase contained a’ instead of a subunit. The present reportsuggests that thesetwo isozymes of phosphorylase kinase have different regulatory properties. Only the whitemuscle isozyme is found capable of binding to calmodulin-Sepharose 4B column a t normal range of cellular Ca2+concentrations, antagonizing the activation of phospho6 diesterase by calmodulin, and responding to the activationby exogenous calmodulin. However, further studies are needed to define exactly the nature of the differential interaction of the two isozymes with calmodulin. I t is possible that under other conditions the red muscle isozyme may also respond to the activation by exogenous calmodulin. Although the two isoz-ymes show differential response to exogenous calmodulin, they both contain equal amounts of FIG.4. Original sample of phosphorylase kinase(0)and the the endogenous calmodulin, 6 subunit. The observation sugred muscle isozyme ( B ) and white muscle isozyme ( C ) were gests strongly that the endogenous and the exogenous calanalyzed by sodium dodecyl sulfate-gelelectrophoresis on10% modulin-mediated Ca2+ activations of phosphorylase kinase polyacrylamide gels. Protein (118 p g ) was applied to each gel. have different regulatory significance. The difference in the regulatory properties of the two isozymes of phosphorylase kinase may reflect the different physiological and metabolic needs of the two muscle types. Since the white (fast) muscle is for fast contraction, it is expected to show a very efficient response to stimulation. Thus, the activation of phosphorylase kinase by exogenous calmodulin may be designed specifically for efficient response to a sudden upsurge of Ca2+ concentration. On the other hand, the Ca2’ regulation by endogenous calmodulin may be a more general regulatorymechanism for phosphorylase kinase. Since theonly difference insubunit structuresof the white and red muscle isozymes are in a and a‘ subunits (3), the differential interaction of the isozymes with exogenous calmodulin may suggest that theexogenous calmodulin binds to subunit a. The observation that phosphorylase kinase lost its ability to bind calmodulin after limited proteolysis (9) is in agreement with such a hypothesis. However, the possibility that the a subunit, instead of being the calmodulin-binding unit, exertsspecific effects on theconformation of the calmodulin binding subunit cannot be excluded a t present. The observation that phosphorylasekinase does not markTIME (min.) edly activate cyclic nucleotide phosphodiesterase isin contrast FIG. 5. Time course of phosphorylase kinase using either with the previous report ( 4 ) that native phosphorylasekinase the purified white muscle isozyme ( A ) or the red muscle iso- is capable of activating myosin light chain kinase. The result zyme (B). The reaction media contained 0.025 M glycerophosphate, may suggest that the two enzymes, phosphodiesterase and 0.025 M Tris-HC1 (pH 7.0), 1 mM [y-.”P]ATP, 5 mM MgAcr, 4 m g / d of phosphorylase 6, 15 mM 2-mercaptoethanol andin the presence of myosin kinase, interact at different domains of calmodulin. either 2 mM ethylene glycol bis(/l-aminoethylether)N,N,N’,N’-tet- However, such a suggestion is not supported by the finding that myosin light chain kinase antagonizes the activation of 0.1 mM CaCI? plus 1 raacetic acid (A-A), 0.1 mM CaCL (A-A), nM calmodulin (W), or 0.1 mM CaCl?plus 10 nM calmodulin phosphodiesterase by calmodulin (21). Thus, additional stud(M). ies are needed to elucidate the natureof the difference in the interactions of the two enzymes by phosphorylase kinase. The differential interaction of phosphorylase kinase isoisozyme appears capable of binding exogenous calmodulin in zymes with calmodulin can beused to separate these isozymes. the normal rangeof cellular Cat+ concentration. The method has been used recently to obtain pure red muscle The differential interaction of phosphorylase kinase iso- type phosphorylase kinase from bovine skeletal muscle (21). zymes with calmodulin appears to reflect the different mechanisms of regulation of these enzyme forms by calmodulin. Acknowledgment-We thank Mr.ErwinWirch for excellent asFig. 5 shows that both isozymes are dependent on Ca” for sistance throughoutthe course of this work. activity, presumably due to the presence of the 6 subunit, the REFERENCES endogenous calmodulin. However, only the isozyme of white muscle was activated by exogenous calmodulin.Similar result 1. Cohen. P. (1973) Eur. J. Biochem. 34, 1-14
- /
Skeletal Muscle Phosphorylase Isozymes Kinase
11105
2. Hayakawa, T., Perkins, J. P., and Krebs, E. G . (1973) Biochem- 11. Shenolikar, S., Cohen, P. T. W., Cohen, P., Mairn, A. C., and istry 12, 574-580 Perry, S. V. (1979) Eur. J . Biochem. 100,329-337 3. dennissen, H. P., and Heilmeyer, L. M. G. (1974) FEBS Lett. 42, 12. Fischer, E. H., and Krebs, E. G . (1958) J. Biol. Chem. 231,65-71 77-80 13. Sharma, R. K., and Wang, J. H. (1979) Adu. Cyclic Nucleotide 4. Cohen, P., Burchell, A,. Foulkes, J. G., Cohen, P. T. W., Vanaman, Res. 10, 187-198 T. C., and Nairn, A. C.(1978) FEBS Lett. 92, 287-293 14. Sharma, R. K., Wang, J. H., Wirch, E., and Wang, J. H. (1980) J. 5. Ozawa, E., Hosoi, K., andEbashi, S . (1967) J. Biochem. (Tokyo)Biol. Chem. 255,5916-5923 61, 531-533 15. Singh, T. J., and Wang, J. H. (1977) J . Biol. Chem. 252, 625-632 6. Brostrom, C. D., Hunkeler, F. L., and Krebs, E. G. (1971) J . Biol. 16. Wang, J. H., and Desai, R. (1977) J . Biol. Chem. 252,4175-4184 Chem. 246, 1961-1967 17. Bradford, M. M. (1976)Anal. Biochem. 72, 248-254 7. Waisman, D. M. (1979) Ph.D. thesis, University of Manitoba, 18. Klee, C.B., and Krinks, M. H. (1978) Biochemistry 17, 120-126 Winnipeg, Manitoba, Canada 19. McCullough, T. E., and Walsh, D. A. (1979) J. Biol. Chem. 254, 8. Walsh, K. X., Millikin, D. M., Schlender, K. K., and Reimann, E. 7336-7344 M. (1980) J. Biol. Chem. 255,5036-5042 20. Burchell, A,, Cohen, P. T. W., and Cohen, P. (1976) FEBS Lett. 67, 17-21 9. DePaoli-Roach, A. A,, Gibbs, J. B., and Roach, P. J. (1979)FEBS Lett. 105, 321-324 21. Wang, J. H., Tam, S. W., Sharma, R. K., and Lewis, W. G. (1981) 10. Cohen, P., Picton, C., and Klee, C. (1979) FEBS Lett. 104, 25-30 Cold Spring Harbor Symp. Quant. Biol., in press
