- Email: [email protected]
Phosphorylation of a 43 kdaltons protein from rat heart by a calmodulin-dependent protein kinase
3ournal of Molecular
and Cellular Cardiologv
(1983)
RAPID
15, 6 l-65
COMMUNICATION
Phosphorylation of a 43 kdaltons by a Calmodulin-Dependent
Protein Protein
from Rat Kinase
Heart
(Received 13 August 1982) The Ca2+-dependent phosphorylation of proteins has been recognized as a major regulatory mechanism of biological processes. In the heart, protein kinases that are activated by Ca’“+ include phosphorylase kinase, myosin light chain kinase, phospholamban kinase [review in It], and the kinases responsible for phosphorylation of endogenous proteins in thee membrane [1P] and soluble [6] fractions of the cell. All of these Ca2+-dependent enzymes require the presence, either as an enzyme subunit or as a cofactor, of calmodulin, a Ca2+binding protein which is involved in various other CaZ+ -requiring reactions or processes [review in 31. We demonstrate here the presence, in the rat heart, of a soluble calmodulindependent protein kinase which seems different from those already described in this tissue. The substrate for this enzyme is a 43 kdaltons protein, present in the same soluble fraction.
All the preparative procedures were carried out at 0 to 4°C unless otherwise stated. Ventricular tissue (total: 10 g) was immediately excised from stunned male Wistar rats (400 g), minced finely with scissors, and homogenized in 3 vol. of 40 mM Tris-HCl (pH 7.6), 0.25 M sucrose, 4 mM EDTA, 1 rn~ dithiothreitol, 0.5 mM of phenylmethylsulfonylfluoride and benzamidine (buffer A), with 3 passes in a glass-teflon homogenizer. The homogenate was centrifuged at 50 000 x g for 45 min and the supernatant was purified by ion exchange chromatography on DEAE-sephacel, with a O-O.5 M KC1 gradient. Calmodulin-activated protein kinase activity exerted on endogenous substrates was assayed by incubating the eluted fraction in the presence of (y-““P) ATP, 10-s M calmodulin and 0.1 mM Ca2+. The peak of activity, eluted from 40 to 120 mM KCl, was pooled, analyzed by polyacrylamide gel electrophoresis, and referred to as fraction I. Figure 1 (lane 1) shows that fraction I was heterogenous. Fraction I was incubated in the presence of (y-““P) ATP, in the different conditions described in the legend to Figure 1, and analyzed by polyacrylamide gel electrophoresis and autoradiography. A protein of 43 kdaltons was the main substrate phosphorylated in the presence of calmodulin (Figure 1, lane 2). a2Pi incorporation was quantitated by densitometric scanning of the autoradiogram and approximately 90% of the protein-bound s2Pi was found to be incorporated in that protein. Fraction I was free from endoOOZZ-2828/83/01006
1 + 05 $03.00/O
genous calmodulin, which was eluted from the DEAE-sephacel column at a higher ionic strength. As shown on Figure 1, lane 3, the phosphorylation process was absolutely dependent on the presence of calmodulin in the incubation medium. Incub;ations were carried out as described in the legend to Figure 1, in the presence of 5.10~is to 1.1O-6 M calmodulin and the resultant 43 kdaltons phosphorylation level was quantitated by densitometric scanning of the autoradiogram (ntot shown). The apparent Km for calmodulin of this kinase activity was approximately .?1.10-~ M. The 43 kdaltons protein phosphorylation, normally observed in the presence of 0.1 mM Casf and 1O-6 M calmodulin, was entirely prevented by either the addition of 2 mM EGTA (Figure 1, lane 4), or of 1O-4 M trifluoperazine (Theraplix), an inhibitor of the biological effects of calmodulin (not shown). We concluded that a Ca2+ and calmodulin-dependent protein kinase which phosphorylates a 43 kdaltons protein was present in fraction I. We made use of the calmodulin-dependency of this kinase, referred in the text to as “43 K kinase”, to purify it further by calmodulin-affinity chromatography. The fraction I was dialyzed against 40 mM Tris-HCl (pH 7.6), 50 rnM NaCl, 0.2 mM CaCl,, 3 mM Mg acetate, 1 mM dithiothreitol, 0.5 mM phenylmethyl sulfonyl fluoride and benzamidine (buffer B) and chromatographed on a calmodulin-sepharose 4B column equilibrated in buffer B (Figure 2, (A)). The fractions were 0 1983 Academic
Press Inc.
(London)
Limitled
62
Rapid
Communication:
C. Delcayre
et al. Autoradiograph
I
Mr
2
(k)
3
5
4
7
6
8
Colmodulin
EGTA
-
-
+
-
-
-
-
FIGURE 1. Phosphorylation of rat heart proteins under various incubation conditions. The conditions are indicated below the lanes. The fraction I from DEAE-Sephacel chromatography of rat heart homogenate supernatant was analyzed by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (SDS) on a 7-15% acrylamide gel using the buffer system of Laemmli [7]. The proteins were visualized by Coomassie blue coloration (lane 1). The phosphorylation experiments were then carried out by incubation of fraction I (20 ~1) in 40 rnM Tris-HCl (pH 7.6), 1 rnM dithiothreitol, 10 rnM Mg acetate, 0.5 rnM (y-““P) ATP at 80 cpm/pmol 0.1 mM CaCl,, at 30°C for 10 min, in a final volume of 100 ~1. 1OW M calmodulin prepared according to [I] was present in incubation medium, unless otherwise indicated. 2 rnM ethylene glycol bis (B-amino-ethyl ether) N,N N’,N’ tetraacetic acid (EGTA) was added when indicated. The reaction was stopped either by putting an aliquot onto a Whatman 3 MM filter paper disc which was immersed and washed in 10% trichloroacetic acid and dried for scintillation counting, or by addition of a volume of Laemmli buffer [7] containing 1 y0 SDS and boiling for 2 min for electrophoresis. The phosphorylated substrates were revealed by autoradiography of the dried gel with intensifying screens (lanes 2 to 8). 4 mg/ml of isolated beef cardiac myosin light chains (MLC), 0.6 mg/ml of rabbit skeletal muscle actin (AC) prepared according to [I31 or 2 mg/ml of rat cardiac troponin (Tn) prepared according to [II] were included to the incubation medium, as indicated (lanes 5 to 8). The molecular weights were determined by comparison with molecular weight markers (Bio-Rad Laboratories), and, are indicated by the arrows (K = kdaltons).
assayed for protein kinase activity and analyzed by SDS-polyacrylamide gel electrophoresis as described in the legend to Figure 1. Fraction II was not retained on the affinity column. As seen in Figure 2 (B), lane 1, the protein pattern of this fraction was quite similar to that of the fraction I (see Figure 1, lane 1). Fraction II contained 99.5% of the proteins loaded on the column. Non-specifically bound proteins were removed by washing the column with buffer B made 0.2 M NaCl (fraction III, Figure 2 (A)), and then the specifically calmodulin-bound proteins were eluted with buffer B made 0.2 M NaCl and 2 mM EGTA (fraction IV, Figure Z(A)). The protein pattern
of this fraction is shown in Figure 2 (B), lane 2. The major band of about 65 K daltons represented 72% of the total proteins of fraction IV, as measured by densitometric scanning of the gel. The minor bands observed at about 120 and 35-37 kdaltons represented respectively 7% and 8% of the total proteins of this fraction. Each of the fractions II and IV were incubated in the presence of (y-““P) ATP, in the conditions described in the legend to Figure 1, and analyzed by polyacrylamide gel electrophoresis and autoradiography. Although a minor radioactive band at the level of 43 kdaltons was occasionally evidenced in fraction II (Figure 2 (B), lane. 3),
Calmodulin-Dependent
Kinases
in Rat
Heart
:A)
-!
NaCl
0.2M,
ca2+
0.2
NaCl mM
0.2M
EGTA
2mM
III I
I
+
1 ‘hr. 0
.
5
I
t I
IO
15
20
Y2mlN
)
Gel
W I
II
0.5
30 ml ----++
Fraction Fraction
volume
AuToradiograph 2
3 M,
Fraction
25
f---
IV
4
5
6
( k )
II
IV
II +Iv
FIGURE 2. Affinity chromatography of the calmodulin-dependent kinase activities from rat heart. (A) The fraction I (see Figure 1) was purified by chromatography on calmodulin-Sepharose 4B as described in the text. Protein concentrations were determined by the method of Bradford [Z]. (B) The fraction II and IV &ted from the column were analyzed by electrophoresis as described in the legend to Figure 1, and the proteins were visualized by Coomassie blue (lane 1) or silver nitrate [8] (1 ane 2) coloration. The total amounts of protein applied to the gel were 30 Fg (lane 1) and 0.5 yg (lane 2). The column fractions were also incubated separatdy or together in the conditions described in the legend to Figure 1, in the presence (lanes 3 to 5) or absence (lane 6) of 1Om6 M calmodulin.
64
Rapid
Communication:
no protein was obviously phosphorylated in both fractions (lanes 3 and 4). However, when fractions II and IV were mixed (vol/vol) and incubated in the same way with (y-““P) ATP, the 43 kdaltons protein was phosphorylated, when 10m6 M calmodulin was also added (Figure 2 (B), lanes 5 and 6). Therefore, the substrate of 43 K kinase, i.e. the 43 K protein had no affinity for calmodulin and was thus not retained by the affinity column, whereas the enzyme could only be eluted from the column with 2 mM EGTA. We have thus demonstrated the presence in fraction IV of a calmodulindependent protein kinase able to phosphorylate a 43 kdaltons protein. This fraction contained as revealed by the electroseveral proteins, phoresis (Figure 2 (B), lane 2). As the purification procedure was that usually used for the purification of most of the calmodulin-binding proteins, we could suppose that fraction IV contains several calmodulin-dependent protein kinases. These observations raised two questions: (i) could the 43 K kinase be identified with a calmodulin-dependent protein kinase already described in the literature? (ii) what is the nature of the 43 kdaltons protein? In order to answer the first question, we compared the biochemical properties of two well described soluble calmodulin-activated protein kinases, i.e. myosin light chain kinase and phosphorylase kinase, with the properties that we observed for the 43 K kinase. The myosin light chain kinase was evidenced in fractions I and IV by the phosphorylation of added bovine cardiac myosin light chains. In fraction IV, this activity was CaZ+ and calmodulin-dependent (not shown). Myosin light chain kinase has been purified by different laboratories and it has a molecular weight of 85 or 95 kdaltons in bovine heart [review in 151 and 65 kdaltons in dog heart [IO]. This enzyme could thus be the major protein of 65 K present in fraction IV. We observed that, in fraction I (Figure 1, lanes 5 and 6), the characteristics of phosphorylation of both 43 kdaltons protein and myosin light chains clearly differed: around 50% of the myosin light chain phosphorylating activity was calmodulin-independent, while, in the same assay, the 43 kdaltons protein phosphorylating activity was entirely calmodulindependent. Loss of calmodulin-dependency is the result of a partial proteolysis of the myosin light chain kinase [ 151. It is unlikely that the sensitivity of myosin light chain kinase to proteolysis would differ depending on the substrate. On the other hand, myosin light chain kinase has been until now recognized as highly specific for myosin light
C. Delcayre
et al.
chains [15]. For these reasons, myosin light chain kinase does not seem to be responsible for the phosphorylation of the 43 kdaltons protein. Phosphorylase kinase has been described by Cohen et al. [5] as a calmodulin-activated enzyme. However, it is very unlikely that it might be responsible for phosphorylation of the 43 K protein, as: (i) cardiac isoenzyme of phosphorylase kinase is not inhibited by trifluoperazine [9] and (ii) it does not bind to a calmodulin-sepharose column in our conditions, i.e. 0.2 mM Ca2+ [12], and indeed the phosphorylase kinase subunits rx’ and p (135 and 128 kdaltons respectively) were not evidenced in the electrophoregram of fraction IV (Figure 2 (B), lane 2). Katoh et al. [6] described in the guinea-pig heart, a mitochondrial protein of 44 kdaltons, the phosphorylation of which is partially inhibited in the absence of calmodulin. With the aim to avoid lysis of subcellular organelles, and thus contamination of the soluble fraction by membranes or mitochondrial components, we used an isotonic extraction buffer (buffer A). We observed no change in our 43 K kinase activity when extraction was carried out in hypotonic conditions (buffer A without sucrose), and after centrifugation of the homogenate so obtained at 200 000 x g for 30 min (not shown). Thus, the 43 K kinase activity seems to be of cytoplasmic origin, and could not be related to an already known activity. Concerning the nature of the 43 kdaltons protein, various possible substrates for the 43 K kinase, having molecular weights around 43 K, were tested by incubation with fraction I. Figure 1 (lanes 7 and 8) shows that skeletal muscle actin (45 kdaltons in our electrophoresis system), and rat cardiac troponin T (41 kdaltons) were phosphorylated but did not comigrate with the unknown 43 kdaltons phosphorylated protein. Bovine heart creatine kinase (40 kdaltons) (Sigma Chemical Co.) was not phosphorylated. It appears that the 43 K kinase substrate is neither actin, troponin T, nor creatine kinase. In conclusion, we show here the existence of a new calmodulin-dependent phosphorylating system in rat heart. Together with the works of Katoh et al. [S] and Schulman et al. [11] who observed other substrate-kinase pairs in guineapig and rat heart, our data provide additional evidence supporting a mediating role for calmodulin in heart muscle. Further purification and characterization of the 43 kdaltons protein should lead to a better understanding of its involvement in regulation of myocardial function.
Calmodulin-Dependent
Kinases
in Rat Heart
65
Acknowledgements The authors would like to express their thanks to Drs K. Schwartz and discussions, Dr P. S. Cassidy for reading the manuscript and M. Albaret assistance. They also thank P. Bouveret from the laboratory for providing gift of Theraplix laboratories, Paris, France.
Claude
Delcayre,
Fraqoise
B. Swynghedauw for helpful and P. Cagnac for secretarial actin. Trifluoperazine was a
Marotte
and
Lydie
INSERM Unitt! 127, H@ital 41, bld de la Chapelle, F-75010 KEY
WORDS:
Calmodulin
dependent
kinases;
Phosphorylation;
Heart
phosphorylated
Rappaport Lariboisihe, Paris, France
proteins.
REFERENCES 1. 2. 3. 4. 5. 6.
7. 8. 9. 10. 11.
12. 13. 14. 15.
AUTRIC, F., FERRAZ, C., KILHOFFER, M. C., CAVADORE, .I. C. & DEMAILLE, J. G. Large scale purification and characterization of calmodulin from ram testis. Biochim Biophys Acta 631, 139-147 (1980). BRADFORD, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 721, 248-254 (1976). CHEUNG, W. Y. Calmodulin plays a pivotal role in cellular regulation. Science 207, 19-27 (1980). COHEN, P. The role of protein phosphorylation in neural and hormonal control of cellular activity. Nature 296, 613-620 (1982). COHEN, P., PICTON, C. & KLEE, C. B. Activation of phosphorylase kinase from rabbit skeletal muscle by calmodulin and troponin. FEBS Lett 104, 25-30 (1979). KATOH, N., WRENN, R. N., WISE, B. C., SHOJI, M. & Kuo, J. F. Substrate proteins for calmodulin-sensitive and phospholipid sensitive Ca 2+-dependent protein kinases in heart, and inhibition of their phosphorylation by palmitoyl carnitine. Proc Nat1 Acad Sci USA 78, 4813-4817 (1981). LAEMMLI, U. K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 (1970). MERRILL, C. R., GOLDMAN, D., SEAMAN, S. A. & EBERT, M. H. Ultrasensitive stain for proteins in polyacrylamide gels shows regional variation in cerebrospinal fluid proteins. Science 211, 1437-1438 (1981). PICTON, C., KLEE, C. B. & COHEN, P. The regulation of muscle phosphorylase kinase by calcium ions, calmodulin and troponin-C. Cell Calc 2, 281-294 (1981). RAPPAPORT, L. & ADELSTEIN, R. S. Phosphorylation of canine cardiac muscle myosin light chain kinase. Abstract. Fed Proc 39, 2041 (1980). SCHULMAN, H. & GREENGARD, P. Ca2+-dependent protein phosphorylation system in membranes from various &sues, and its activation by “calcium-dependent regulator”. Proc Nat1 Acad Sci USA 75,5432-5436 (1978). SHARMA, R. K., TAM, S. W., WAISMAN, D. M. & WANG, J. H. Differential interaction of rabbit skeletal muscle phosphorylase kinase isozymes with calmodulin. J Biol Chem 255, 11102-l 1105 (1980). SPUDICH, J. A. & WATT, S. The regulation of rabbit skeletal muscle contraction. J Biol Chem 246,48664871 (1971). TSUKUI, R. & EBASHI, S. Cardiac troponin. J Biochem (Tokyo) 73, 1119-l 121 (1973). WALSH, M. P. Calmodulin-dependent myosin light-chain kinases. Cell Calc 2, 333-352 (1981).
and Cellular Cardiologv
(1983)
RAPID
15, 6 l-65
COMMUNICATION
Phosphorylation of a 43 kdaltons by a Calmodulin-Dependent
Protein Protein
from Rat Kinase
Heart
(Received 13 August 1982) The Ca2+-dependent phosphorylation of proteins has been recognized as a major regulatory mechanism of biological processes. In the heart, protein kinases that are activated by Ca’“+ include phosphorylase kinase, myosin light chain kinase, phospholamban kinase [review in It], and the kinases responsible for phosphorylation of endogenous proteins in thee membrane [1P] and soluble [6] fractions of the cell. All of these Ca2+-dependent enzymes require the presence, either as an enzyme subunit or as a cofactor, of calmodulin, a Ca2+binding protein which is involved in various other CaZ+ -requiring reactions or processes [review in 31. We demonstrate here the presence, in the rat heart, of a soluble calmodulindependent protein kinase which seems different from those already described in this tissue. The substrate for this enzyme is a 43 kdaltons protein, present in the same soluble fraction.
All the preparative procedures were carried out at 0 to 4°C unless otherwise stated. Ventricular tissue (total: 10 g) was immediately excised from stunned male Wistar rats (400 g), minced finely with scissors, and homogenized in 3 vol. of 40 mM Tris-HCl (pH 7.6), 0.25 M sucrose, 4 mM EDTA, 1 rn~ dithiothreitol, 0.5 mM of phenylmethylsulfonylfluoride and benzamidine (buffer A), with 3 passes in a glass-teflon homogenizer. The homogenate was centrifuged at 50 000 x g for 45 min and the supernatant was purified by ion exchange chromatography on DEAE-sephacel, with a O-O.5 M KC1 gradient. Calmodulin-activated protein kinase activity exerted on endogenous substrates was assayed by incubating the eluted fraction in the presence of (y-““P) ATP, 10-s M calmodulin and 0.1 mM Ca2+. The peak of activity, eluted from 40 to 120 mM KCl, was pooled, analyzed by polyacrylamide gel electrophoresis, and referred to as fraction I. Figure 1 (lane 1) shows that fraction I was heterogenous. Fraction I was incubated in the presence of (y-““P) ATP, in the different conditions described in the legend to Figure 1, and analyzed by polyacrylamide gel electrophoresis and autoradiography. A protein of 43 kdaltons was the main substrate phosphorylated in the presence of calmodulin (Figure 1, lane 2). a2Pi incorporation was quantitated by densitometric scanning of the autoradiogram and approximately 90% of the protein-bound s2Pi was found to be incorporated in that protein. Fraction I was free from endoOOZZ-2828/83/01006
1 + 05 $03.00/O
genous calmodulin, which was eluted from the DEAE-sephacel column at a higher ionic strength. As shown on Figure 1, lane 3, the phosphorylation process was absolutely dependent on the presence of calmodulin in the incubation medium. Incub;ations were carried out as described in the legend to Figure 1, in the presence of 5.10~is to 1.1O-6 M calmodulin and the resultant 43 kdaltons phosphorylation level was quantitated by densitometric scanning of the autoradiogram (ntot shown). The apparent Km for calmodulin of this kinase activity was approximately .?1.10-~ M. The 43 kdaltons protein phosphorylation, normally observed in the presence of 0.1 mM Casf and 1O-6 M calmodulin, was entirely prevented by either the addition of 2 mM EGTA (Figure 1, lane 4), or of 1O-4 M trifluoperazine (Theraplix), an inhibitor of the biological effects of calmodulin (not shown). We concluded that a Ca2+ and calmodulin-dependent protein kinase which phosphorylates a 43 kdaltons protein was present in fraction I. We made use of the calmodulin-dependency of this kinase, referred in the text to as “43 K kinase”, to purify it further by calmodulin-affinity chromatography. The fraction I was dialyzed against 40 mM Tris-HCl (pH 7.6), 50 rnM NaCl, 0.2 mM CaCl,, 3 mM Mg acetate, 1 mM dithiothreitol, 0.5 mM phenylmethyl sulfonyl fluoride and benzamidine (buffer B) and chromatographed on a calmodulin-sepharose 4B column equilibrated in buffer B (Figure 2, (A)). The fractions were 0 1983 Academic
Press Inc.
(London)
Limitled
62
Rapid
Communication:
C. Delcayre
et al. Autoradiograph
I
Mr
2
(k)
3
5
4
7
6
8
Colmodulin
EGTA
-
-
+
-
-
-
-
FIGURE 1. Phosphorylation of rat heart proteins under various incubation conditions. The conditions are indicated below the lanes. The fraction I from DEAE-Sephacel chromatography of rat heart homogenate supernatant was analyzed by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (SDS) on a 7-15% acrylamide gel using the buffer system of Laemmli [7]. The proteins were visualized by Coomassie blue coloration (lane 1). The phosphorylation experiments were then carried out by incubation of fraction I (20 ~1) in 40 rnM Tris-HCl (pH 7.6), 1 rnM dithiothreitol, 10 rnM Mg acetate, 0.5 rnM (y-““P) ATP at 80 cpm/pmol 0.1 mM CaCl,, at 30°C for 10 min, in a final volume of 100 ~1. 1OW M calmodulin prepared according to [I] was present in incubation medium, unless otherwise indicated. 2 rnM ethylene glycol bis (B-amino-ethyl ether) N,N N’,N’ tetraacetic acid (EGTA) was added when indicated. The reaction was stopped either by putting an aliquot onto a Whatman 3 MM filter paper disc which was immersed and washed in 10% trichloroacetic acid and dried for scintillation counting, or by addition of a volume of Laemmli buffer [7] containing 1 y0 SDS and boiling for 2 min for electrophoresis. The phosphorylated substrates were revealed by autoradiography of the dried gel with intensifying screens (lanes 2 to 8). 4 mg/ml of isolated beef cardiac myosin light chains (MLC), 0.6 mg/ml of rabbit skeletal muscle actin (AC) prepared according to [I31 or 2 mg/ml of rat cardiac troponin (Tn) prepared according to [II] were included to the incubation medium, as indicated (lanes 5 to 8). The molecular weights were determined by comparison with molecular weight markers (Bio-Rad Laboratories), and, are indicated by the arrows (K = kdaltons).
assayed for protein kinase activity and analyzed by SDS-polyacrylamide gel electrophoresis as described in the legend to Figure 1. Fraction II was not retained on the affinity column. As seen in Figure 2 (B), lane 1, the protein pattern of this fraction was quite similar to that of the fraction I (see Figure 1, lane 1). Fraction II contained 99.5% of the proteins loaded on the column. Non-specifically bound proteins were removed by washing the column with buffer B made 0.2 M NaCl (fraction III, Figure 2 (A)), and then the specifically calmodulin-bound proteins were eluted with buffer B made 0.2 M NaCl and 2 mM EGTA (fraction IV, Figure Z(A)). The protein pattern
of this fraction is shown in Figure 2 (B), lane 2. The major band of about 65 K daltons represented 72% of the total proteins of fraction IV, as measured by densitometric scanning of the gel. The minor bands observed at about 120 and 35-37 kdaltons represented respectively 7% and 8% of the total proteins of this fraction. Each of the fractions II and IV were incubated in the presence of (y-““P) ATP, in the conditions described in the legend to Figure 1, and analyzed by polyacrylamide gel electrophoresis and autoradiography. Although a minor radioactive band at the level of 43 kdaltons was occasionally evidenced in fraction II (Figure 2 (B), lane. 3),
Calmodulin-Dependent
Kinases
in Rat
Heart
:A)
-!
NaCl
0.2M,
ca2+
0.2
NaCl mM
0.2M
EGTA
2mM
III I
I
+
1 ‘hr. 0
.
5
I
t I
IO
15
20
Y2mlN
)
Gel
W I
II
0.5
30 ml ----++
Fraction Fraction
volume
AuToradiograph 2
3 M,
Fraction
25
f---
IV
4
5
6
( k )
II
IV
II +Iv
FIGURE 2. Affinity chromatography of the calmodulin-dependent kinase activities from rat heart. (A) The fraction I (see Figure 1) was purified by chromatography on calmodulin-Sepharose 4B as described in the text. Protein concentrations were determined by the method of Bradford [Z]. (B) The fraction II and IV &ted from the column were analyzed by electrophoresis as described in the legend to Figure 1, and the proteins were visualized by Coomassie blue (lane 1) or silver nitrate [8] (1 ane 2) coloration. The total amounts of protein applied to the gel were 30 Fg (lane 1) and 0.5 yg (lane 2). The column fractions were also incubated separatdy or together in the conditions described in the legend to Figure 1, in the presence (lanes 3 to 5) or absence (lane 6) of 1Om6 M calmodulin.
64
Rapid
Communication:
no protein was obviously phosphorylated in both fractions (lanes 3 and 4). However, when fractions II and IV were mixed (vol/vol) and incubated in the same way with (y-““P) ATP, the 43 kdaltons protein was phosphorylated, when 10m6 M calmodulin was also added (Figure 2 (B), lanes 5 and 6). Therefore, the substrate of 43 K kinase, i.e. the 43 K protein had no affinity for calmodulin and was thus not retained by the affinity column, whereas the enzyme could only be eluted from the column with 2 mM EGTA. We have thus demonstrated the presence in fraction IV of a calmodulindependent protein kinase able to phosphorylate a 43 kdaltons protein. This fraction contained as revealed by the electroseveral proteins, phoresis (Figure 2 (B), lane 2). As the purification procedure was that usually used for the purification of most of the calmodulin-binding proteins, we could suppose that fraction IV contains several calmodulin-dependent protein kinases. These observations raised two questions: (i) could the 43 K kinase be identified with a calmodulin-dependent protein kinase already described in the literature? (ii) what is the nature of the 43 kdaltons protein? In order to answer the first question, we compared the biochemical properties of two well described soluble calmodulin-activated protein kinases, i.e. myosin light chain kinase and phosphorylase kinase, with the properties that we observed for the 43 K kinase. The myosin light chain kinase was evidenced in fractions I and IV by the phosphorylation of added bovine cardiac myosin light chains. In fraction IV, this activity was CaZ+ and calmodulin-dependent (not shown). Myosin light chain kinase has been purified by different laboratories and it has a molecular weight of 85 or 95 kdaltons in bovine heart [review in 151 and 65 kdaltons in dog heart [IO]. This enzyme could thus be the major protein of 65 K present in fraction IV. We observed that, in fraction I (Figure 1, lanes 5 and 6), the characteristics of phosphorylation of both 43 kdaltons protein and myosin light chains clearly differed: around 50% of the myosin light chain phosphorylating activity was calmodulin-independent, while, in the same assay, the 43 kdaltons protein phosphorylating activity was entirely calmodulindependent. Loss of calmodulin-dependency is the result of a partial proteolysis of the myosin light chain kinase [ 151. It is unlikely that the sensitivity of myosin light chain kinase to proteolysis would differ depending on the substrate. On the other hand, myosin light chain kinase has been until now recognized as highly specific for myosin light
C. Delcayre
et al.
chains [15]. For these reasons, myosin light chain kinase does not seem to be responsible for the phosphorylation of the 43 kdaltons protein. Phosphorylase kinase has been described by Cohen et al. [5] as a calmodulin-activated enzyme. However, it is very unlikely that it might be responsible for phosphorylation of the 43 K protein, as: (i) cardiac isoenzyme of phosphorylase kinase is not inhibited by trifluoperazine [9] and (ii) it does not bind to a calmodulin-sepharose column in our conditions, i.e. 0.2 mM Ca2+ [12], and indeed the phosphorylase kinase subunits rx’ and p (135 and 128 kdaltons respectively) were not evidenced in the electrophoregram of fraction IV (Figure 2 (B), lane 2). Katoh et al. [6] described in the guinea-pig heart, a mitochondrial protein of 44 kdaltons, the phosphorylation of which is partially inhibited in the absence of calmodulin. With the aim to avoid lysis of subcellular organelles, and thus contamination of the soluble fraction by membranes or mitochondrial components, we used an isotonic extraction buffer (buffer A). We observed no change in our 43 K kinase activity when extraction was carried out in hypotonic conditions (buffer A without sucrose), and after centrifugation of the homogenate so obtained at 200 000 x g for 30 min (not shown). Thus, the 43 K kinase activity seems to be of cytoplasmic origin, and could not be related to an already known activity. Concerning the nature of the 43 kdaltons protein, various possible substrates for the 43 K kinase, having molecular weights around 43 K, were tested by incubation with fraction I. Figure 1 (lanes 7 and 8) shows that skeletal muscle actin (45 kdaltons in our electrophoresis system), and rat cardiac troponin T (41 kdaltons) were phosphorylated but did not comigrate with the unknown 43 kdaltons phosphorylated protein. Bovine heart creatine kinase (40 kdaltons) (Sigma Chemical Co.) was not phosphorylated. It appears that the 43 K kinase substrate is neither actin, troponin T, nor creatine kinase. In conclusion, we show here the existence of a new calmodulin-dependent phosphorylating system in rat heart. Together with the works of Katoh et al. [S] and Schulman et al. [11] who observed other substrate-kinase pairs in guineapig and rat heart, our data provide additional evidence supporting a mediating role for calmodulin in heart muscle. Further purification and characterization of the 43 kdaltons protein should lead to a better understanding of its involvement in regulation of myocardial function.
Calmodulin-Dependent
Kinases
in Rat Heart
65
Acknowledgements The authors would like to express their thanks to Drs K. Schwartz and discussions, Dr P. S. Cassidy for reading the manuscript and M. Albaret assistance. They also thank P. Bouveret from the laboratory for providing gift of Theraplix laboratories, Paris, France.
Claude
Delcayre,
Fraqoise
B. Swynghedauw for helpful and P. Cagnac for secretarial actin. Trifluoperazine was a
Marotte
and
Lydie
INSERM Unitt! 127, H@ital 41, bld de la Chapelle, F-75010 KEY
WORDS:
Calmodulin
dependent
kinases;
Phosphorylation;
Heart
phosphorylated
Rappaport Lariboisihe, Paris, France
proteins.
REFERENCES 1. 2. 3. 4. 5. 6.
7. 8. 9. 10. 11.
12. 13. 14. 15.
AUTRIC, F., FERRAZ, C., KILHOFFER, M. C., CAVADORE, .I. C. & DEMAILLE, J. G. Large scale purification and characterization of calmodulin from ram testis. Biochim Biophys Acta 631, 139-147 (1980). BRADFORD, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 721, 248-254 (1976). CHEUNG, W. Y. Calmodulin plays a pivotal role in cellular regulation. Science 207, 19-27 (1980). COHEN, P. The role of protein phosphorylation in neural and hormonal control of cellular activity. Nature 296, 613-620 (1982). COHEN, P., PICTON, C. & KLEE, C. B. Activation of phosphorylase kinase from rabbit skeletal muscle by calmodulin and troponin. FEBS Lett 104, 25-30 (1979). KATOH, N., WRENN, R. N., WISE, B. C., SHOJI, M. & Kuo, J. F. Substrate proteins for calmodulin-sensitive and phospholipid sensitive Ca 2+-dependent protein kinases in heart, and inhibition of their phosphorylation by palmitoyl carnitine. Proc Nat1 Acad Sci USA 78, 4813-4817 (1981). LAEMMLI, U. K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 (1970). MERRILL, C. R., GOLDMAN, D., SEAMAN, S. A. & EBERT, M. H. Ultrasensitive stain for proteins in polyacrylamide gels shows regional variation in cerebrospinal fluid proteins. Science 211, 1437-1438 (1981). PICTON, C., KLEE, C. B. & COHEN, P. The regulation of muscle phosphorylase kinase by calcium ions, calmodulin and troponin-C. Cell Calc 2, 281-294 (1981). RAPPAPORT, L. & ADELSTEIN, R. S. Phosphorylation of canine cardiac muscle myosin light chain kinase. Abstract. Fed Proc 39, 2041 (1980). SCHULMAN, H. & GREENGARD, P. Ca2+-dependent protein phosphorylation system in membranes from various &sues, and its activation by “calcium-dependent regulator”. Proc Nat1 Acad Sci USA 75,5432-5436 (1978). SHARMA, R. K., TAM, S. W., WAISMAN, D. M. & WANG, J. H. Differential interaction of rabbit skeletal muscle phosphorylase kinase isozymes with calmodulin. J Biol Chem 255, 11102-l 1105 (1980). SPUDICH, J. A. & WATT, S. The regulation of rabbit skeletal muscle contraction. J Biol Chem 246,48664871 (1971). TSUKUI, R. & EBASHI, S. Cardiac troponin. J Biochem (Tokyo) 73, 1119-l 121 (1973). WALSH, M. P. Calmodulin-dependent myosin light-chain kinases. Cell Calc 2, 333-352 (1981).