of chemistry no. 7501-7509, 1981 m purification …purification and characterization of smooth...

THE JOURNAL OF BIOLOGICAL CHEMISTRY

Printed m U.S.A. Vol. 256, No. 14, Issue of July 25. pp. 7501-7509, 1981

Purification and Characterization of Smooth Muscle Myosin Light Chain Kinase*

(Received for publication, January 6, 1981, and in revised form, March 27, 1981)

Robert S. Adelstein and Claude B. Klee From the Cardiology Branch, National Heart, Lung, and Blood Institute and the Laboratory of Biochemistry, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20205

Smooth muscle myosin light chain kinase was purified from turkey gizzards. The enzyme was extracted from washed myofibrils and the final step of purification was affinity chromatography using calmodulin coupled to Sepharose 4B. The purified enzyme was characterized with respect to its physical, chemical, and kinetic properties. It has a molecular weight of 130,000 by sodium dodecyl sulfate polyacrylamide gel electrophoresis and 124,000 by sedimentation equilibrium centrifugation under nondenaturing conditions. It is an asymmetric molecule with a Stokes radius of 75 A, a sedimentation coefficient of 4.45 S, and a frictional coefficient of 1.85.

Smooth muscle myosin light chain kinase is dependent on the calcium-binding protein calmodulin for activity. It has an apparent for calmodulin of lo-’ M and binds 1 mol of calmodulin/mol of myosin kinase in the presence of calcium. The binding of calmodulin increases the sedimentation coefficient from 4.45 S to 5.05 S, and the Stokes radius from 75 A to 79 A, and does not alter the frictional coefficient.

The enzyme has a K,,, for ATP and the 20,000-dalton light chain of smooth muscle myosin of 50 PM and 5 p ~ , respectively. It phosphorylates the 20,000-dalton light chain of smooth muscle myosin more rapidly than the equivalent light chain from cardiac and skeletal muscles. It does not phosphorylate histones, a-casein, phosphorylase kinase, or phosphorylase b at a significant rate.

Smooth muscle myosin light chain kinase catalyzes the transfer of the y-phosphate from ATP to smooth muscle myosin. This reaction plays a major role in regulating smooth muscle contraction (for reviews, see Refs. 1 and 2). Smooth muscle myosin is a hexamer composed of two “heavy” polypeptide chains (200,000 daltons) and two pairs of “light” chains (20,000 and 15,000 daltons). The reversible phosphorylation of seryl residue 13 of the 20,000-dalton light chain (3), has a marked effect on the actin-activated ATPase activity of smooth muscle myosin (4-10). Actin activates the Mg-ATPase activity of phosphorylated but not of unphosphorylated smooth muscle myosin. Moreover, experiments carried out with skinned muscle fibers (11) as well as intact muscles (12- 14) confirm the role of myosin phosphorylation in smooth muscle contraction. These experiments do not rule out a possible role for other regulatory mechanisms (15), in addition to myosin phosphorylation.

Smooth muscle myosin light chain kinase is dependent on * The costs of publication of this article were defrayed in part by

the payment of page charges. This article must therefore be hereby marked “aduertisernent” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Ca2+ for its activity (16). Dabrowska et al. showed that the Ca2’ requirement of smooth muscle myosin kinase is mediated by the calcium-binding protein, calmodulin (17), and Yagi et al. (18) demonstrated that the skeletal muscle enzyme was calmodulin-dependent. The enzyme is inactive in the absence of either Ca2’ or calmodulin. Myosin light chain kinases from cardiac muscle (19-21) and non-muscle cells (22-25) share this requirement for Ca2+ and calmodulin, as do a number of other unrelated proteins (for reviews, see Refs. 26-29).

Smooth muscle myosin light chain kinase is a substrate for the catalytic subunit of CAMP-dependent protein kinase (30). Phosphorylation of smooth muscle myosin kinase in the absence of bound calmodulin results in a decrease in myosin kinase activity (30, 31).

A previous report on smooth muscle myosin light chain kinase isolated from chicken gizzards described an enzyme of 105,000 daltons (17). Employing a number of proteolytic inhibitors we report in this paper a procedure for isolating milligram quantities of a homogeneous smooth muscle myosin light chain kinase with a molecular weight of 130,000. Data are presented on the physical, chemical, and kinetic properties of this enzyme in the presence and absence of bound calmod- din.

EXPERIMENTAL PROCEDURES

Materials All chemicals were reagent grade, unless specified otherwise.

Deionized water was used throughout. Skeletal muscle troponin C was a &t of Dr. Paul L e a h (Boston Biomedical Research Institute). Phosphorylase b (rabbit skeletal) was the gift of Dr. Erwin Reimann (Medical College of Ohio). Histones V-S, 11-A, and a-casein were purchased from Sigma.

Proteolytic inhibitors were obtained from the following sources: pepstatin A and leupeptin were from Vega Biochemicals, phenylmethylsulfonylfluoride, N-tosylphenylalanine chloromethyl ketone, a-N-benzoyl-L-arginine methyl ester, soybean trypsin inhibitor, and diisopropyl fluorophosphate were from Sigma. Methane sulfonic acid, 4 N, containing 0.2% of 3-(2-aminoethyl) indole, 6 N HCl, dansyl chloride, and dimethylsuberimidate dihydrochloride were products of Pierce Chemical Co. N-[methyl-’4C]Dansyl chloride (56 mCi/mmol) was purchased from Amersham Co.

Bovine brain calmodulin was purified as described previously (32). [‘4C]Calmodulin was prepared by guanidination with O-[I4C]methyl- isourea (33); the modified protein contained 1 mol of homoarginine/ mol and had a specific activity of 120,000 cpm/nmol.

Methods All procedures were carried out at 4 “C unless specified otherwise. Preparation of the Isolated Phosphorylatable Light Chain from

Smooth, Cardiac, and Skeletal Muscle Myosin-The isolated 20,000- dalton light chain of smooth muscle myosin was the substrate used for most of the experiments described in this paper. It was prepared from 500 g of washed myofibrils (described under “Results”) which can be used for preparation of both the 20,000-dalton myosin light chain, as well as myosin kinase. To prepare the isolated myosin light

7501

7502 Smooth Muscle Myosin Light Chain Kinase

chains, actomyosin was extracted from the washed myofibrils using 4 volumes of: 40 mM Tris-HCl (pH 7.2 at 4 "C), 40 mM KCl, 2 mM EGTA,' 10 mM ATP, and 100 mg/liter of streptomycin sulfate. After 20 min at 4 "C, the mixture was sedimented at 13,000 X g for 45 min. While monitoring the pH (7.2), the supernatant fluid was made 20 mM with respect to CaClz by the addition of 1 M CaClZ and the resulting precipitate of actomyosin was sedimented at 13,000 X g for 45 min. The precipitate was suspended in about 100 ml of cold water (4 "C) and solid guanidine hydrochloride was added, to make it 5 M with respect to guanidine (final total volume = 250 ml). To precipitate the myosin heavy chains, 250 ml of cold (4 "C) water, followed by 1 liter of cold (4 "C) 95% ethanol were added, according to the procedure of Perrie and Perry (34).

The resulting precipitate was sedimented for 30 min at 13,000 x g and the supernatant fluid was concentrated, by removing the ethanol with a rotary evaporator, to approximately 300 ml. The aqueous solution containing guanidine hydrochloride, was dialyzed against a t least three changes of 8 liters each of 1 mM EDTA, 0.1 m~ EGTA adjusted to pH 7.0, and lyophilized. The lyophilized powder was dissolved in 300 ml of 20 m~ Tris-HC1 (pH 8.0), 20 m~ KCl, and 75 mg/liter of phenylmethylsulfonylfluoride. The solution of isolated light chains is still contaminated with appreciable M ) amounts of calmodulin. Removal of calmodulin was by chromatography on a column of DEAE-Sephacel. An example of the separation of calmodulin from myosin light chains is shown in Fig. 1.

The light chains of cardiac and skeletal muscle myosin were prepared from the respective myosins as previously outlined (34). Final purification of the isolated light chains and removal of calmodulin, was carried out as outlined above for the smooth muscle myosin light chains.

Preparation of Myosin-Smooth muscle myosin was prepared using the same initial steps outlined above for the isolated light chains, through the precipitation of actomyosin by addition of 1 M CaC12 to a final concentration of 20 mM. The resulting precipitate was dissolved in 10 ml of 20 mM Tris.HC1 (pH 7.3), 0.5 M KC1, 1.0 mM EDTA, 0.5 mM EGTA, and 1 mM dithiothreitol to give a protein concentration of 10 to 15 mg/ml. This sample was applied to a column (2.5 X 90 cm) of Sepharose 6B which was eluted with the same buffer. Just prior to chromatography the sample was made 5 mM with respect to MgATP. Myosin was identified by assaying the column for K+- EDTA-activated ATPase activity as previously described (35). The pooled peak of myosin activity was concentrated by addition of solid ammonium sulfate to 50% of saturation. The sample was dialyzed against and stored in 20 mM Tris.HC1 (pH 7.3), 100 mM KCl, 1 mM EDTA, 0.5 mM EGTA, 1 m~ NaNs, and 1 mM dithiothreitol. The final protein concentration was 6 mg/ml.

Assays for Myosin Light Chain Kinase Activity-Myosin light chain kinase activity was assayed in a volume of 0.1 ml of 20 mM Tris. HCl (pH 7.3), 10 mM MgCl,, 0.2 mM CaC12 (in excess over EGTA) or 2 mM EGTA, 0.1 mM [y3'P]ATP (0.5 Ci/mmol), 0.2 mg/ml of smooth muscle myosin light chains, a t 24 "C. Unless indicated otherwise, M calmodulin was included, and the assay was initiated by addition of the kinase or a dilution of the kinase in bovine serum albumin.

Since the isolated myosin light chain preparation was often a mixture of the 20,000 (phosphory1atable)- and 15,000-dalton light chains (approximately 50% of each), the Concentration of substrate sites available for phosphorylation was based on the maximum amount of phosphate that could be introduced into a given solution of light chains. The extent of myosin light chain phosphorylation was confirmed by urea-glycerol polyacrylamide gel electrophoresis, which separates the phosphorylated and unphosphorylated species of the 20,000-dalton light chain (34).

The assay was terminated by addition of an aliquot of a solution of trichloroacetic acid-sodium pyrophosphate to give a final concentration of 10% trichloroacetic acid, 2% sodium pyrophosphate. After Millipore fdtration and washing, the fdter was counted in a Searle Mark I11 scintillation spectrometer following immersion in 15 ml of Aquasol (Amersham) (36). Time of the assay varied from 15 s (for lo-" M kinase) to 5 min (for M kinase) at 24 "C.

It is important to ascertain that the rate of incorporation of '*P into the isolated light chain of myosin is linear with respect to time. It was found that as long as only 20% or less of the total available light chains in the incubation mixture underwent phosphorylation,

I The abbreviations used are: EGTA, [ethylene-bis(oxyethy1- enenitri1o)ltetraacetic acid; SDS, sodium dodecyl sulfate; CaM, calmodulin; dansyl, 5-dimethylaminonaphthalene-l-sulfonyl.

8 8 ' 7 I n

1 ' I d l I h"., J

0 0 1 * _". ~ """

XI 40 80 sa TUBE NUMBER

FIG. 1. Resolution of smooth muscle myosin light chains from calmodulin by ion exchange chromatography. The light chain ( L C . ) fraction, prepared as outlined under "Methods" (300 ml; approximately 0.3 mg/ml) was applied to a column (2 X 10 cm) of DEAE-Sephacel previously equilibrated with 20 m~ KC1.20 mM Tris. HCl (pH 8.0), and 75 mg/liter of phenylmethylsulfonylfluoride. The light chains were eluted with a linear gradient (total volume 400 ml) created by using equal amounts of 20 m~ KC1 and 0.5 M KC1 in the buffer described above. The flow rate was 80 ml/h and 6-ml fractions were collected. Phosphorylatable light chains were identified by assaying 30-pl aliquots with myosin kinase in the presence (0) and absence (0) of calmodulin. Calmodulin (W) was identified by the ability of a given fraction to activate myosin kinase.

the rate of incorporation did not deviate significantly from linearity. Background counts, using this particular concentration of [Y-~'P]ATP were found to be 3000 to 5000 dpm. The time of the assays was usually selected so that at least 40,000 dpm were incorporated into the light chains, where 40,000 dpm = 5 to 10% of the total available substrate.

When turkey gizzard smooth muscle myosin was assayed as a substrate for myosin kinase, the concentration of 20,000-dalton myosin light chain in the assay mixture was 15 p ~ . The concentration of the isolated light chain used for direct comparison of the rate of phosphorylation of the isolated light chain with that of intact myosin was 18 p ~ . The conditions for the assay were those described above, except that the KC1 concentration was 50 mM.

The concentrations of the other substrates assayed with myosin kinase were as follows: histone 11-A, 50 @a; histone V-S, 114 p ~ ; phosphorylase b, 5.3 FM and 14 PM; a-casein, 20 p ~ . The following light chain concentrations were used in determining the K,,, of myosin kinase for different light chains: 20,000-dalton light chain of smooth muscle myosin, 2 to 16 ELM; 18,500-dalton rabbit skeletal muscle myosin light chain, 2 to 12 p ~ ; 20,000-dalton canine cardiac myosin light chain, 3 to 13 p ~ .

Amino Acid Analyses-The enzyme was dialyzed overnight against 1000 volumes of 0.02 M Tris-HCI (pH 8.0) containing 0.05 M NaCl and 0.2 mM dithioerythritol. Aliquots of 250 pl of the protein solution (absorbance of 0.5 to 0.7 at 280 nm) were transferred to hydrolysis tubes containing one crystal of phenol and 25 nmol of norleucine in order to correct for recovery upon acid hydrolysis and subsequent chromatography. (Recoveries of norleucine were 97 to 99%). Hydrol- yses in 6 N HC1 were carried out at 110 "C for 12, 22, 45, and 72 h. Analyses were performed as previously described (32). Except when indicated otherwise, the values represent the average of the four determinations. Performic acid oxidation was performed on samples which had been dialyzed extensively against 0.05 M (NH,)HCOa and lyophilized to dryness to remove (NH4)HCOa. Tryptophan determinations were performed according to the method of Simpson et al. (37). Recoveries of tryptophan from N-acetyl tryptophanamide or lysozyme were between 88 to 95%. With short hydrolysis times, the proline values were anomalously high probably because of the presence of cysteine which coelutes with proline under these conditions.

The NHz-terminal residue determinations were performed by the dansyl chloride procedure described by Weiner et al. (38) using [I4C]dansyl chloride (30,000 cpm/nmol). Glycerol gradient sedimentation experiments were performed as previously described (39).

SDS-Polyacrylamide Gel Electrophoresis-Gel electrophoresis was performed in 5 to 15% or 7.5 to 20% polyacrylamide gradient slab gels (40) using the buffer system of Laemmli (41). For the determination of M,, a 0.1% SDS, 7.5% polyacrylamide gel was used.

Ultracentrifugal Analysis-Sedimentation velocity and equilib-

Smooth Muscle Myosin Light Chain Kinase 7503

rium experiments were performed with a Beckman model E ultracen- trifuge equipped with a photoelectric scanner (42). The protein concentration was monitored at 278 nm. The solvent was 0.04 M Tris. HCl (pH 8.0), containing 0.1 M KCl, 0.2 mM dithiothreitol, and 0.05 mM EDTA. Sedimentation velocity runs were at 60,000 rpm at 23 "C. The protein concentration was 0.35 to 0.5 mg/ml. Sedimentation equilibrium experiments were carried out a t 12,000 and 15,000 rpm at 16 "C using the meniscus depletion technique of Yphantis (43). The protein concentration was 0.04 mg/ml. The integrity of the protein after centrifugation was tested by SDS-polyacrylamide gel electrophoresis. The data were corrected to solvent with the density and viscosity of water at 20 "C using the partial specific volume of 0.73 for myosin kinase, determined as described in the legend to Table 11, and 0.71 for calmodulin (44). The data were analyzed by means of nonlin- ear least squares curve-fitting using the MLAB program on a DEC- 10 Computer. The best fit was obtained with a single component system except in the case of the centrifugation of a mixture of myosin kinase and calmodulii in the presence of EGTA. In this case, a base- line correction had to be included in the program. This correction was due to the presence of calmodulin which did not sediment with myosin kinase in EGTA. The concentration of calmodulin, however, was too low to obtain independent values of the M, of the two proteins assuming a two-component, noninteracting system.

Protein Determinations-Protein concentrations were determined by the method of Lowry et al. (45), as well as by UV absorption for calmodulin, € 2 7 7 - = 3300 (32). The UV absorption spectra were recorded on a Cary model 118 spectrophotometer.

RESULTS

Purification of Myosin Kinase from Turkey Gizzards: Preparation of Myofibrils-Fresh gizzards are obtained from turkeys killed the same day. The lobes of smooth muscle are freed of all fat and connective tissue and placed in a beaker immersed in ice. The muscle is ground in an electric meat grinder fitted with a plate with 1-mm holes, resulting in a finely ground muscle mince. The muscle is weighed (10 average gizzards = 500 g of ground muscle) and suspended in 3 to 4 volumes of wash: 20 m~ Tris-HC1 (pH 6.8 at 4 "C), 40 mM KC1 (NaC1 can be used in place of KC1 throughout the preparation), 1 mM MgClz, 1 m~ dithiothreitol, 5 m~ EGTA, 75 mg/liter of phenylmethylsulfonylfluoride, 100 mg/liter of streptomycin sulfate, and 0.05% Triton X-100. The mince is suspended with a Sorvall Omm-Mixer in a 2-quart Mason jar at two-thirds full speed, activated three times for 5 s each. The suspension is homogenized with two passes in a Glenco glass/Teflon homogenizer (250 ml), using a Sears % inch electric drill to turn the pestle. The mixture is sedimented at 15,000 X g for 15 min. The pellet is resuspended in the same volume of wash solution, in the absence of Triton with the aid of the Sorvall Omni-Mixer. The suspension is again sedimented for 15 min at 15,000 X g and the supernatant is again discarded. This step is repeated once more, by which time the myofibrils have lost most of their red color.

Kinase Extraction-The pellet is resuspended in 3 to 4 volumes of 40 m~ Tris. HCl (pH 7.5), 60 mM KC1, 25 mM MgClZ, 1 mM dithiothreitol, 5 mM EGTA, 75 mg/liter of phenylmethylsulfonylfluoride, 100 mg/liter of streptomycin sulfate, 1 mg/liter of leupeptin, 1 mg/liter of pepstatin A, 10 mg/liter of N-tosylphenylalanine chloromethyl ketone, a-N- benzoyl-L-arginine methyl ester, soybean trypsin inhibitor, and M diisopropyl fluorophosphate (diisopropyl fluorophosphate is added in a hood with extreme care). The pellet is suspended in the Omni-Mixer, homogenized with one pass through the Glenco homogenizer, and sedimented for 30 min at 15,000 X g. The supernatant fluid is filtered through glass wool and the pellet is discarded. Solid (NHJZSO4 is slowly added to the supernatant and the protein that precipitates between 0 and 40% saturation is sedimented at 15,000 X g for 15 min and discarded. The supernatant fluid is made 60% saturated with respect to ammonium sulfate and sedimented

at 15,000 X g for 30 min. This pellet is homogenized by hand in a glass/glass homogenizer in a buffer of 15 mM Tris.HC1 (pH 7.5), 0.5 M NaC1, 1 mM EDTA, 5 mM EGTA, 1 mM dithiothreitol, and 75 mg/liter of phenylmethylsulfonylfluoride, 100 mg/liter of streptomycin sulfate, 1 mg/liter leupeptin, 1 mg/liter of pepstatin A, 10 mg/liter of N-tosylphenylalanine chloromethyl ketone, a-N-benzoyl-L-arginine methyl ester, soybean trypsin inhibitor, and M diisopropyl fluorophosphate and is dialyzed at least 3 h against 10 volumes of the same buffer. The 40 to 60% ammonium sulfate fraction was dissolved in approximately 60 ml of buffer, with a resulting protein concentration of about 25 mg/ml.

Column Chromatography Steps-The 40 to 60% ammonium sulfate fraction is applied to a column (5 X 87 cm) of Sephacryl S-300 (Fig. 2). Thirty ml of a sample (25 mg/ml) was applied at one time, requiring 2 columns when starting with 500 g of ground muscle. A single peak of myosin kinase activity elutes from the column with a Kay = 0.31. The peak is pooled and dialyzed against 4 liters of 20 m~ Tris HC1 (pH 7.8), 1 mM EGTA, 1 mM dithiothreitol, and 75 mg/liter phenylmethylsulfonylfluoride.

Following dialysis, the pooled Sephacryl S-300 peaks (from two columns, 225 ml, 1.5 mg/ml) were chromatographed on a column (2.0 X 20 cm) of DEAE-Sephacel. The myosin kinase activity elutes at a conductance of 10 to 13 mmho at 4 "C (Fig. 3). A single peak of myosin kinase activity is pooled (150 ml, 0.7 mg/ml) and dialyzed against 1 liter of 20 mM Tris-HCl (pH 7.8), 50 mM NaC1, 1 mM EGTA, and 1 mM dithiothreitol.

The final step of purification is affinity chromatography on a column of bovine brain calmodulin coupled to Sepharose- 4B (Fig. 4). Just prior to chromatography, enough Ca2+ and Mg2+ are added to the sample to give a final free concentration of 0.5 mM CaZC and 3 m~ Mg2+. Occasionally, a small peak of myosin kinase (indistinguishable in its properties from that eluted with EGTA) is eluted with the buffer containing 0.2 M NaCl and 0.5 mM Ca2+. The major peak of kinase activity is eluted when 2 mM EGTA is introduced into the eluting buffer (Fig. 4). The peak of myosin kinase activity is pooled, divided into 0.2-ml aliquots, and kept frozen at -70 "C until ready to use. Approximately 30 to 50 mg of purified kinase can be prepared from 500 g of ground gizzards. Table I summarizes the recovery and activity of myosin kinase at the various steps.

7 '2 - 2.6

2 4

- 2 2

2 0

18

-

~

-

- 1 6 i :14 1 2 I - 1 0

8

6

4

1 2 Y

400 5m Wl 700 Bm 9J) 1 m O llm 1 m O 1W 1400

VOLUME ELUTED lmll

FIG. 2. Gel filtration of smooth muscle myosin kinase. Thirty ml (25 mg/ml) of the 40 to 60% ammonium sulfate fraction was applied to a column (5 X 87 cm) of Sephacryl S-300 equilibrated and eluted in a solution of 15 mM Tris-HCl (pH 7.5), 0.5 M NaCl, 1 mM EDTA, 5 m~ EGTA, 1 mM dithiothreitol, and 75 mg/liter of phenylmethylsulfonylfluoride. The column was eluted at a flow rate of 200 ml/h. Fractions were assayed for myosin kinase activity (0- - -0) as outlined under "Methods." LC., light chain.


R II

TUBE NUMBER

FIG. 3. Ion exchange chromatography of the pooled Sephac- ryl S-300 peak of myosin kinase. Two hundred twenty-five ml of myosin kinase (1.5 mg/ml) was applied to a column (2 X 20 cm) of DEAE-Sephacel equilibrated with 20 m~ Tris.HC1 (pH 7.8), 1 mM EGTA, 1 mM dithiothreitol, and 75 mg/liter of phenylmethylsulfonylfluoride. The kinase was eluted at a flow rate of 60 ml/h with a 1200-ml linear NaCl gradient (20 to 600 mM NaCl) made up in the equilibration buffer. Aliquots of the 15-ml fractions were assayed for myosin kinase (0- - -0) (see “Methods”) and myosin phosphatase (H) activity (47) L C . , light chain.

i f I

VOLUME ELUTED lmll

FIG. 4. Calmodulin-affinity chromatography of myosin kinase. P.pproximately 50 ml (0.7 mg/ml) of the pooled peak following DEAE-Sephacel chromatography was applied to a 7-ml column of calmodulin-Sepharose 4B equilibrated with 40 mM Tris. HCl (pH 7.2), 50 mM NaC1, 0.2 m~ CaCh, 3 mM MgC12, and 1 m~ dithiothreitol. The column was washed with 3 volumes of equilibrating buffer and then with 2 to 3 volumes of the same buffer made 0.2 M with respect to NaC1. The kinase was eluted by introducing 2 m~ EGTA into the buffer containing 0.2 M NaC1. The assay for kinase activity (0- - -0) is not linear with respect to 32P incorporation. L C . , light chain.

Comments on the Purification Procedure.-The fist part of this procedure, preparation of the myofibrils and extraction of the kinase from the myofibrils, has previously been pub- lished (4, 46). The washes discarded in preparation of the myofibrils were found to contain approximately 5% of the total kinase activity. Varying the manner in which the myofibrils are homogenized may lead to an increased loss of enzyme during these steps. The presence of 25 m~ MgCl2 is critical for extraction of the kinase from the washed myofibrils.

The 3-h dialysis step inserted between the precipitation with ammonium sulfate and molecular sieve chromatography improved the loading on the Sephacryl S-300 column. Se- phacryl S-300 chromatography has been substituted for Seph- arose 6B, which gives equally good results, but requires slower flow rates. Following ion exchange chromatography the major proteins that can be identified following SDS-polyacrylamide gel electrophoresis are: myosin kinase (M, = 130,000) and tropomyosin (M, = 43,000 and 38,000). As indicated on the elution profile (Fig. 3), phosphatases capable of dephosphor-

ylating the 20,000-dalton light chain of myosin elute just following the peak of myosin kinase activity. These enzymes, which have been purified (47), cannot be identified on the gel and are removed by affinity chromatography on a calmodulin- Sepharose column.

Including EGTA in the buffer during the initial steps serves to dissociate myosin kinase from calmodulin and prevents Ca2+-dependent proteolysis. To minimize the possibility of proteolysis, Ca2+ and M e are added immediately prior to calmodulin-affinity chromatography and not during the dialysis. The purified kinase can be stored without proteolysis or loss of activity at -70 “C for at least 6 months. Storage at -30 “C results in a slow loss of activity (about 30% over a 3- week period).

Physical and Chemical Properties of Myosin Kinase- Myosin kinase isolated as described above was shown to be homogeneous by gel electrophoresis under native and denaturing conditions. More than 80% of the enzyme activity subjected to electrophoresis under native conditions co-migrated with the single Coomassie blue staining component (30). Gel electrophoresis in the presence of sodium dodecyl sulfate also revealed the presence of a single polypeptide, with a M , of 130,000 * 6,000 estimated by comparison with proteins of known M, (Fig. 5).

Table I1 summarizes some of the physical properties of the purified enzyme. The molecular weight of the enzyme determined by the sedimentation equilibrium method of Yphantis (43), under nondenaturing conditions, was 124,000 -+ 2,200. Myosin kinase is therefore a monomer. The low sedimentation coefficient ( s ~ o . = 4.45 S) indicates that the protein exhibits some degree of asymmetry. The Stokes radius of 75 A, calculated on the basis of the sedimentation coefficient, the U, and the molecular weight, agrees with the value of 71 A, estimated from the elution pattern of the enzyme from Sephacryl S-300 gel filtration columns run in the presence of globular proteins with known Stokes radii (data not shown, method given in Ref. 48). The relatively large Stokes radius and the low sedimentation coefficient are consistent with a markedly asymmetric shape (f/fo = 1.85) (Table 11).

The amino acid composition of myosin kinase is shown in Table 111. The protein contains many charged residues, lysine, aspartic and glutamic acid, and it has a relatively high content of cysteine and tryptophan. The latter is responsible for the characteristic UV-absorption spectrum of the protein with marked shoulder at 290 nm (Fig. 6) and a relatively large

of 11.35. The content of tyrosine and tryptophan determined chemically is very similar to that determined spectro- scopically by comparison with the spectrum of an equimolar solution of N-acetyl tyrosinamide and N-acetyl tryptophanamide in methanol (49). The amino acid composition of myosin kinase isolated from chicken gizzard (17) is also shown in Table 111. The two proteins have a similar amino acid

TABLE I Summary: purification of turkey gizzard myosin light chain kinase

(based upon starting with 500 g of ground gizzards)

Volume concentra- Specific tion activity activity Protein

ml mg/ml (to- p o l / r n i n pmol/mg/ tal mg) VV mm

Extract supernatant 1500 3 (4500) 450 (100) 0.10 40-6056 ammonium sul- 60 25 (1500) 405 (90) 0.27

Sephacryl S-300 225 1.5 (338) 371 (82) 1.1 DEAE-Sephacel 150 0.7 (105) 231 (51) 2.2 Calmodulin affinity col- 20 1.5 (30) 180 (40) 6

fate fraction

umn


20,000

- - 94.” - 68,”

” - 43.” - 39.”

1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0

MIGRATION INTO GEL, rnrn FIG. 5. Determination of M, of myosin kinase by 0.1% SDS-7

M% polyacrylamide gel electrophoresis (41). Myosin kinase ( M K ) was run alone and in the presence of M, standards. From the top: myosin heavy chain; /3 and p’ subunit of RNA polymerase, P- galactosidase, phosphorylase b, bovine serum albumin, ovalbumin, soybean trypsin inhibitor. The graph shows a plot of the molecular weights of these standards (with the exception of P-galactosidase), as used to determine the molecular weight of myosin kinase: 130,000.

TABLE I1 Physical properties of myosin light chain kinase

M, SDS-gel electrophoresis 130,000 f 6,000” Sedimentation equilibriumb 124,000 * 2,200’

Sedimentation coefficient (szo.,,,) 4.45 s Stokes radius

Calculated from s;lo.,,,, M,, 17 75 A Gel filtration 71 A

Frictional coefficient ( f / fdd 1.85 Partial specific volume (ne € 3 “rn

0.73 11.40 f 0.2

” Range, including results from different gel systems (see Fig. 8). Sedimentation equilibrium and velocity centrifugations were as

‘Calculated as indicated under “Methods” * standard error of

” The frictional coefficient was calculated from Wyman and Ingall’s

‘The partial specific volume was calculated on the basis of the

‘The extinction coefficient is based on protein concentration de-

described under “Methods.”

fitting.

nomogram (64).

amino acid composition (65).

termined by amino acid analyses as described under “Methods.”

composition although the lower values reported for the enzyme isolated from chicken muscle reflect its smaller size, 105,000 (17, 50). Some residues are significantly different in content. The low valine content in the chicken enzyme may be explained by incomplete hydrolysis. It seems, however, that the aspartyl, prolyl, and phenylalanyl residues are less abundant in the chicken protein. These differences may be explained by the smaller size of the chicken gizzard enzyme. The difference in molecular weight between the two enzymes is due to the chicken enzyme having undergone proteolysis (see “Discussion”).

Attempts to identYy the NH2-terminal residue of turkey gizzard myosin kinase were unsuccessful, suggesting that the enzyme has a blocked NH2 terminus. Phenol-OH and elysyl

NH2 group were quantitatively labeled but no other dansyl- ated amino acids were detected by autoradiography of the chromatogram. Cysteine and tryptophan would not have been detected by the method used.

Kinetic Properties of Myosin Kinuse-Table IV summarizes some of the kinetic properties of turkey gizzard myosin kinase. The K, values for ATP and for the isolated 20,000- dalton light chain of turkey gizzard myosin are 50 p~ and 5 p ~ , respectively. These values were measured at saturating

TABLE I11 Amino acid composition of smooth musle myosin light chain

kinase Turkev g i u a r d ” Chicken aizzard*

~~

LYS 115.1 102

’4% 42.1 38 ASP 109.8 92

His 13 16

Thr‘ 68.5 61 Ser‘ 86.4 86 Glu 155.2 139 Pro 58.0 47 GlY 64.2 71 Ala 85.1 73 Vald 82.6 52 Met 21.7 20 Ile“ 51.6 44 Leu 66.9 59 Tyr 24.7 23 Phe 32.1 27 Trp‘ 14.6 ND‘ CY# 27.9 ND

M R W ~ 11 1.58 Total residues 1120 950

The data are expressed for a M, of 125,000. Amino acid composition of chicken gizzard myosin kinase based

Values obtained after extrapolation to zero time hydrolysis. Values obtained after 72-h hydrolysis. Tryptophan determinations were done according to the method

on a M, of 105,000 (17).

of Simpson et al. (37) and are corrected for 91% recovery. ’ND, not determined. .e Cysteine was determined as cysteic acid after performic acid

oxidation and the values were normalized according to the content of glutamic acid.

MRW = mean residue weight.

12 1 I 1 1 1

2 4 0 2 6 0 2 8 0 3 0 0 3 2 0 WAVELENGTH (nml

FIG. 6. U V absorption spectrum of myosin kinase. The UV absorption spectrum of the protein solution (0.5 mg/ml) was measured in 0.02 M Tris-HCI (pH 8.0) containing 0.05 M NaCl and 0.2 mM dithioerythritol. Protein concentration was determined by amino acid analysis as described under “Methods.” The absorbance is shown for a 1 8 solution of protein.


concentrations of CaZ+ ( M) and calmodulin ( M). The VmaX obtained by extrapolation of a number of reciprocal plots were 10 to 30 pmol of Pi transferred/mg of kinase/min (Table IV). The higher values were obtained with fresh enzyme.

The concentration of calmodulin needed to obtain 50% activation of myosin kinase (10"' M in the assay mixture) was approximately lo-' M. At concentrations below lo-' M kinase, this value was not significantly affected by the concentration of kinase. The concentration of Ca2+ in these experiments was

M. No activation was observed when skeletal muscle troponin C to M) which resembles calmodulin in its primary structure (51) was substituted for calmodulin. This is in agreement with the report of Walsh et al. (52).

Turkey gizzard smooth muscle myosin kinase is capable of phosphorylating the isolated 20,000-dalton light chain of canine cardiac myosin and the 18,500-dalton light chain of rabbit skeletal muscle myosin. However, the rates of phosphorylation of these substrates are, respectively, one-f ih and one-thir- tieth the rate at which the enzyme phosphorylates the smooth muscle light chain (see Table V). These differences in the rates of phosphorylation are not due to major differences in the K , values of the enzyme for the various substrates (see Table IV).

The rate at which smooth muscle myosin kinase phosphorylates the isolated light chain of turkey gizzard myosin was found to be similar to the rate at which it phosphorylates the light chain of the native myosin (Table V). The phosphorylation rate of the isolated light chain was 8 pmol/mg/min and that for the intact myosin molecule was 4 pmol/mg/min. The

TABLE IV Kinetic properties of turkey gizzard myosin light chain kinase The data for ATP, smooth muscle myosin light chains, and V,,,

are derived from reciprocal plots. The K , for canine cardiac and rabbit skeletal muscle myosin light chains were derived from reciprocal plots performed at the same time as those for turkey gizzard smooth muscle myosin light chains. The K0.s for calmodulin was determined at a kinase concentration of 10"" M.

K,,, ATP 50 p~ K , myosin light chains

Turkey smooth muscle 5 PM Canine cardiac muscle 8 P Rabbit skeletal muscle 11 pM

K0.5 calmodulin 1 nM Vmax 10-30 pnol/mg kinase/min

TABLE V Specificity of smooth muscle myosin light chain kinase

Substrate Relative rate of phosphorylation"

%

20,000-dalton smooth muscle myosin light 100

Smooth muscle myosin 50 20,000-dalton cardiac muscle myosin light 20

18,500-dalton skeletal muscle myosin light 3.3

Phosphorylase 6 Phosphorylase kinase Histone 11-A <0.2 Histone V-S a-Casein

chain

chain

chain

The details of the assay are given under "Methods." The relative rates of phosphorylation for the various light chains were obtained from Vmax of reciprocal plots. The concentrations of light chains used in comparing smooth muscle myosin and the isolated 20,000-dalton light chain of smooth muscle myosin were 15 and 18 p ~ , respectively. The concentrations of the other potential substrates are given under "Methods."

enzyme does not catalyze phosphorylation of a number of other potential substrates (see Table V).

Ca2+-dependent Interaction of Myosin Kinase with Cal- modulin-The stoichiometry of the myosin kinase-calmodulin interaction was determined by three independent methods: titration of myosin kinase activation by calmodulin, determination of the molecular weight of the complex after cross- linking with dimethylsuberimidate (53) and characterization of the complex by sucrose gradient (39), or sedimentation equilibrium centrifugation.

Fig. 7 is a titration curve showing the effect of increasing amounts of calmodulin on the activity of myosin kinase. The abcissa indicates the molar ratio of calmodulin to myosin kinase. The figure shows that 1 eq of calmodulin activates 1 eq of myosin kinase at a myosin kinase concentration of 10"

When myosin kinase is incubated in the presence of ['4C]calmodulin, Ca", and dimethylsuberimidate, the M, of the protein determined by SDS-polyacrylamide gel electrophoresis was increased from 125,000 to 145,000 (Fig. 8, gels A and B ) indicating the formation of a one to one complex between the two proteins. The 145,000-dalton cross-linked polypeptide was labeled as shown by autoradiography (Fig. 8, gel B'). If EGTA instead of Ca'+ was present during the incubation with the cross-linking reagent (gel A), the M, of myosin kinase was 125,000 and no radioactivity was detected on the gel except that associated with the 17,000 M , band, corresponding to ['4C]calmodulin (gel A'). During the incubation with the cross-linking reagent high molecular weight aggregates were also formed, probably due to disulfide bond formation since the protein contains a large amount of cystei- nyl residues. These large complexes (Mr = 250,000 to 300,000) were also associated with ['4C]calmodulin when Ca2+ was present during the reaction (gel B').

The existence of a Ca2+-dependent, stoichiometric complex between myosin kinase and ['4C]calmodulin was confimed by co-sedimentation of the two proteins in gradients of glycerol (Fig. 9). In the absence of Ca", the two proteins identified by enzyme activity, radioactivity, and SDS-polyacrylamide gel electrophoretic patterns are clearly separated, sedimenting according to their apparent sedimentation coefficients (4.5 and 1.9 S, respectively) (Fig. 9A). When Ca2+ is added to the gradient solutions (0.2 m ~ ) , 50% of the labeled calmodulin co- migrated with myosin kinase (Fig. 9B). Only a small increase in the amount of bound calmodulin was observed when the latter was included in the gradient solutions at 4 X lo-* M and

M.

1.05 .-

E 0.41 . F

1 I 1

0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 CALMODULIN/MYOSIN KINASE

FIG. 7. Titration of myosin kinase activity using increasing amounts of calmodulin. The concentration of myosin kinase was

M and assays were carried out under initial rate conditions in the presence of increasing amounts of calmodulin at 0 "C. The arrow indicates that at a molar ratio of approximately 1:1, calmodulin/ myosin kinase, there is no further increase in activity with increasing amounts of calmodulin.


3 x M (Fig. 9, C and D). Since a significant amount of calmodulin (more than 50%) is bound to myosin kinase even in the absence of calmodulin in the gradient, the binding is too tight to allow determination of a binding constant by this method. The fraction of calmodulin associated with myosin kinase corresponds to 1 molar equivalent since the two proteins were applied to the gradient in a 2:l molar ratio of calmodulin to myosin kinase.

The interaction of the two proteins was also analyzed by sedimentation equilibrium experiments. As shown in Table VI, when myosin kinase and calmodulin are sedimented in the presence of Ca2+, a single component with a molecular weight of 141,000 was detected. There was no evidence of dissociation even at the lowest concentrations detectable (10" M). When EGTA was present, the molecular weight of the enzyme was not significantly different from that of the enzyme in the absence of calmodulin (Table 11). Under the conditions of the experiments, even at concentrations greater than M at the bottom the cell, no interaction between the two proteins was observed in the presence of EGTA. In the presence of Ca", on the other hand, the complex was very tight. As indicated above, the calmodulin-myosin kinase complex had a slightly larger sedimentation coefficient as expected on the basis of its molecular weight. No large change in the conformation of the protein was observed since the frictional coefficient and the Stokes radius were not significantly different from those measured in the absence of calmodulin (Tables I1 and VI).

0-

MLCK, CaM - MLCK. 125 K-

-0

- 145 K

p4C]-CaM 1)11

EGTA ca2+ EGTA Ca2+

A B A' B'

FIG. 8. Cross-linking of myosin kinase with ["C]calmodulin. Myosin light chain kinase (MLCK) (0.5 mg/ml) was dialyzed overnight against 0.2 M triethanolamine-HCI (pH 8.5) containing 1 mM MgCI2 and 0.1 mM dithiothreitol a t 4 "C. ["C]Calmodulin was added to aliquots of 50 pI of kinase to give a concentration of IO-% M, MgClz was added to give a concentration of 2 mM, and either EGTA (gels A and A') or CaC12 (gels B and B') were added to give a concentration of 1 mM or 0.5 mM, respectively. The cross-linking reaction was started by addition of 10 pl of a fresh solution containing 25 pg of dimethylsuberimidate in 0.2 M triethanolamine buffer (pH 8.5). The samples were incubated at 25 "C for 4 h. The reaction was terminated by addition of 20 pl of a 0.05 M Tris. HCI (pH 8) solution in 6 M urea, 0.5 M dithiothreitol, and 1% SDS, and boiling for 1 min. The samples were subjected to SDS-polyacrylamide gel electrophoresis in a 7% to 15% gradient of acrylamide (40). All gels contained 20 pg of myosin kinase and 8 pg of ['4C]calmodulin; gels A and B were stained with Coomassie blue and gels A' and B' are the corresponding autoradi- ograms. Molecular weight markers were the polymers of lactic dehydrogenase and catalase a5tained after cross-linking, as described above.

FRACTION NUMBER

FIG. 9. Sedimentation of myosin kinase and calmodulin in a gradient of glycerol. A solution containing 0.6 nmol of myosin kinase and 1.3 nmol of ['4C]calmodulin in a final volume of 50 pl was applied to the top of 10 to 40% gradients of glycerol (total volume, 3.8 ml) in 50 mM Tris-HCI (pH 8.0). 50 mM KCI, 3 mM MgCI2, 0.5 mM dithiothreitol, and 1.0 mM EGTA (gradient A) or 0.2 m~ CaCL (gradients B to D). Gradients C and D, in addition, contained 4 X IO-" M and 2.5 X 10" M [14C]calmodulin, respectively. The gradients B to D were sedimented at 55,000 rpm for 24 h at 4 "C in an SW 60 rotor (Beckman model LS 65 centrifuge). Gradient A was sedimented 17 h at 50,000 rpm. Five-drop fractions were collected to measure radioactivity and enzyme activity. The s values were calculated by comparison with that of lactic dehydrogenase (as shown by the arrow) added to the mixture of myosin kinase and calmodulin and that of free calmodulin. The recovery of enzyme and calmodulin varied between 80 and 90% of the material applied to the gradient. Direction of sedimentation was from right to left. LC., light chain.

TABLE VI Caz+-dependent interaction of myosin light chain kinase and

calmodulin Ca" ECTA

Sedimentation coefficient" 5.05 S 4.45 s M, (sedimentation equilib- 141,000 f 600 124,000 f 3,000

Stokes radius (A) 79 75 Frictional coefficient (f/fo) 1.85 f 0.03 1.85 f 0.03 Partial specific volume (t7) 0.727 0.731 4% nm 10.10 11.40

(sm.d

rium")

" A mixture of myosin kinase (2.6 X IO-" M) and calmodulin (2.8 X M) were analyzed by sedimentation velocity experiments in the

presence of 0.5 mM CaCL or 0.2 mM EGTA as described under "Methods." For sedimentation equilibrium experiments, the protein concentration was lowered to 0.3 X M. The protein concentration along the cell was measured at 278 nm where contribution by calmodulin absorption is negligible compared to that of myosin kinase (0.00s and 0.307, respectively).

" The 6 and absorption coefficients were calculated on the basis of the amino acid composition and absorption coefficient of the two proteins assuming a one to one complex. M , values were calculated as indicated under "Methods" f standard error of fitting.

Thus, myosin kinase and calmodulin form tight, one-to-one stoichiometric complexes even in the absence of substrate, but only in the presence of Ca". These experiments also confm that the kinase is monomeric under native conditions and exists as a heterodimer (myosin kinase-calmodulin) in the presence of Ca".

DISCUSSION

Smooth muscle myosin kinase was isolated from turkey gizzards and purified 60-fold from washed myofibrils. Approx- imately 30 to 50 mg of myosin kinase can be isolated from 500


g of ground gizzards. Based on the recovery of enzyme activity, the content of myosin kinase is approximately 150 mg/kg of turkey gizzard muscle which corresponds to a concentration of approximately 1.2 X lop6 M. For comparison, the total (free and bound) calmodulin concentration of chicken gizzard muscle is approximately 1.1 X M (54). How much of the calmodulin is bound to other proteins and how much is free is not known at present.

Size-The purified myosin kinase appears to be a homogeneous protein as judged by SDS-polyacrylamide gel electrophoresis and sedimentation equilibrium and sedimentation velocity centrifugation. The molecular weight of myosin kinase is 130,000 k 6,000 as determined from its mobility in SDS-polyacrylamide gel electrophoresis. This molecular weight contrasts with that reported for myosin kinase isolated from rabbit skeletal muscle 80,000 (55), bovine cardiac muscle 92,000 (21), human platelets 105,000 (22, 23), but not bovine brain 130,000 (24). In the case of the skeletal and cardiac muscle, a myofibrillar-bound form of the myosin kinase has been partially purified and appears to have a molecular weight of approximately 150,000 by gel electrophoresis (56). The apparent discrepancy in the molecular weight between the myosin kinase isolated from chicken gizzards (105,000) (17) and turkey gizzards has been resolved and was apparently due to proteolysis of the chicken enzyme. Both enzymes appear to have a molecular weight of 130,000 by SDS-polyacrylamide gel electrophoresis.*

The molecular weight of myosin kinase under nondenaturing conditions was determined by sedimentation equilibrium centrifugation and found to be 124,000 both in the absence of calmodulin, and in the presence of calmodulin and EGTA (i.e. conditions under which calmodulin does not bind to myosin kinase). This is in agreement with the molecular weight determined under denaturing conditions and suggests that myosin kinase is a monomer in the absence of bound calmodulin.

In the presence of Ca*+-calmodulin, the molecular weight of myosin kinase measured by sedimentation equilibrium centrifugation increased from 124,000 to 141,000 (Mr calmodulin = 16,500). This is consistent with the active species of myosin kinase being composed of a single subunit of both proteins. The 1:l stoichiometry is also supported by cross-linking experiments using dimethylsuberimidate and [14C]calmodulin and by titrating the enzyme with increasing amounts of calmodulin.

Shape-Smooth muscle myosin kinase is markedly asymmetric in shape. The Stokes radius calculated from sedimentation velocity, molecular weight, and partial specific volume is 75 A. This is in good agreement with the value of 71 A, obtained by gel filtration. Table VI shows that when myosin kinase binds calmodulin there is no major change in the Stokes radius. The unusual shape of the kinase may help orient it in relation to the myosin molecule, which is also markedly asymmetric.

Substrate Specificity-Similar to other myosin kinases (23), the smooth muscle enzyme is specific in that it phosphorylates the 20,000-dalton light chain of myosin much more rapidly than other substrates. Although the K , values of smooth muscle myosin kinase for the isolated phosphorylatable light chains of skeletal and cardiac muscle are approximately the same as the K , for the 20,000-dalton light chain of smooth muscle, the heterologous light chains are phosphorylated at significantly slower rates (see Table V). With respect to the markedly different rates of phosphorylation of the skeletal and smooth muscle myosin light chains, it is of interest to

* M. P. Walsh and D. J. Hartshorne, personal communication.

note that there are reported differences in the amino acid sequences around the phosphorylatable serine residue (3,57).

The isolated 20,000-dalton light chain of turkey gizzard smooth muscle myosin is phosphorylated about twice as fast as myosin purified from the same source. In contrast using myosin kinase, myosin and the isolated myosin light chain from chicken gizzard smooth muscle Mrwa and Hartshorne (58) reported that the light chain was phosphorylated 10 times more rapidly than intact myosin.

Interaction of Ca2+-Calmodulin and Myosin Kinase-One of the main functions of calmodulin in smooth muscle appears to be the regulation of myosin kinase activity. Since phosphorylation of myosin plays a major role in smooth muscle contraction (1,2,4-13), the mechanism of this regulation is of interest. Myosin kinase shows an absolute dependence on Ca*'-calmodulin for activity. Moreover, the affinity of Ca2'- calmodulin for smooth muscle myosin kinase is strong (apparent K o . ~ = lo-' M). This is similar to the value reported for skeletal muscle myosin kinase (59) and also for cyclic nucleotide phosphodiesterase (60, 61).

How Ca2+ and calmodulin interact to regulate smooth muscle myosin kinase activity in vivo is not presently known. Indeed, an important determinant of this interaction, the concentration and distribution of free calmodulin is unknown. One model that has been suggested for a number of calmodulin-dependent enzymes involves a two-step mechanism: 1) Ca" + calmodulin % Ca2'-calmodulin; 2) Ca'+-calmodulin + enzyme (inactive) % Ca"-calmodulin-enzyme (active) (for a review, see Ref. 27). Recently, it has been shown that calcium f i s t binds to three or four of the binding sites on calmodulin and the Cas-calmodulin or Ca4-calmodulin then binds to and activates CAMP phosphodiesterase (44, 62). This mechanism may be applicable to myosin kinase (see e.g. skeletal muscle myosin kinase, Ref. 59).

It is worth considering another mechanism, which is suggested by the ability of calmodulin to bind calcium at multiple sites. Above we have presented in vitro evidence indicating that [14C]calmodulin does not bind to myosin kinase in the absence of Ca*+. On the basis of the calmodulin and myosin kinase concentrations used in the sedimentation velocity and cross-linking experiments, the apparent K d for calmodulin, in the presence of EGTA must be greater than about 5 X M. On the other hand, it is possible that at M free Ca*+, the concentration in resting smooth muscle, Ca2' may be bound to one or two of the calcium binding sites on calmodulin. The species Gal-calmodulin and/or Can-calmodulin might then be capable of binding to myosin kinase, without activating the enzyme. This mechanism for activating myosin kinase, assuming the enzyme was bound to its substrate, would ody require a rise in free ca2+ (e.g. M to M) to initiate phosphorylation of myosin. The mechanism, which is limited by diffusion of Ca2' (rather than that of Ca2+-calmodulin) has been suggested for certain calmodulin-dependent enzymes by Klee and Haiech (63).

A related problem deals with the inactivation of myosin kinase and the strong affinity of myosin kinase for Ca*'- calmodulin. As mentioned above, in the absence of Ca2+, the binding of calmodulin to these enzymes is extremely weak. Based on kinetic considerations, one might expect that the dissociation of the Ca2+-calmodulin complex from myosin kinase would be very slow. However, preliminary experiments have shown that the removal of Ca" by addition of EGTA results in a rapid loss ( 4 0 s) of enzyme activity, suggesting that the initial step in inactivating myosin kinase is the dissociation of Ca2' from the Ca"-calmodulin-enzyme complex. This mechanism for inactivation was described by Klee


and Haiech for the calmodulin-dependent enzyme, phosphodiesterase (63).

Previous work has shown that the activity of myosin kinase can also be modulated by phosphorylation. Myosin light chain kinase is a substrate for CAMP-dependent protein kinase (30) and phosphorylation of smooth muscle myosin kinase, in the absence of bound calmodulin, decreases the affinity of the enzyme for calmodulin (31). Future studies of the interaction of the purified enzyme, with Ca", calmodulin and CAMP- dependent protein kinase should help to increase our under- standing of how myosin kinase can participate in the regulation of smooth muscle contraction.

Acknowledgments-We are indebted to C. Robert Eaton, William Anderson, Jr., and James M. Miles for excellent technical assistance. We wish to thank Exa Murray for her expert editorial assistance and Dr. M. Lewis for his advice in analyzing the centrifugation data.

REFERENCES 1. Small, J. V., and Sobieszek, A. (1980) Znt. Reu. Cytol. 64,241-306 2. Adelstein, R. S., and Eisenberg, E. (1980) Annu. Reu. Biochem.

3. Jakes, R., Northrop, F., and Kendrick-Jones, J. (1976) FEBS Lett.

4. Sobieszek, A,, and Small, J. V. (1977) J. Mol. Biol. 112, 559-576 5. Sherry, J . M. F., Gorecka, A., Aksoy, M. O., Dgbrowska, R., and

6. Chacko, S., Conti, M. A., and Adelstein, R. S. (1977) Proc. Natl.

7. Onishi, H., and Watanabe, S. (1979) J. Biochem. (Tokyo) 85,457-

8. Lebowitz, E. A,, and Cooke, R. (1979) J. Biochem. (Tokyo) 85,

9. Huszar, G., and Bailey, P. (1979) Am. J. Obstet. Gynecol. 135,

10. DiSalvo. J.. Gruenstein. E.. and Silver. P. (1978) Proc. SOC. EXD.

49,921-956

70, 229-234

Hartshorne, D. J. (1978) Biochemistry 17,4411-4418

Acad. Sci. U. S. A . 74, 129-133

472

1489-1494

718-726

11.

12.

13.

14.

15. 16.

17.

18.

19.

20. 21.

22.

23.

24.

25.

26.

Biol. Med. 158,410-414

503-506

loo00

254,4954-4956

Hoar, P. E., Kerrick, W. G. L., and Cassidy, P. (1979) Science 204,

De Lanerolle, P., and Stull, J. T. (1980). J. Biol. Chem. 255,9993-

Barron, J . T., Birany, M., and Bkany, K. (1979) J. Biol. Chem.

Dillon, P. F., Aksoy, M. O., Driska, S. P., and Murphy, R. A.

Ebashi, S. (1980) Proc. R. SOC. Lond. B. Biol. Sci. 207, 259-286 Frearson, N., Focant, B. W. W., and Perry, S. V. (1976) FEBS

Lett. 63, 27-32 Dabrowska, R., Aromatorio, D., Sherry, J. M. F., and Hartshorne,

D. J . (1977) Biochem. Biophys. Res. Commun. 78, 1263-1272 Yagi, K., Yazawa, M., Kakiuchi, S., Ohshima, M., and Uenishi, K.

(1978) J. Biol. Chem. 253, 1338-1340 Walsh, M. P., Vallet, B., Autric, F., and Demaille, J . G. (1979) J.

Biol. Chem. 254, 12136-12144 Rappaport, L., and Adelstein, R. S. (1979) Fed. Proc. 38, 337a Wolf, H., and Hofmann, F. (1980) Proc. Natl. Acad. Sci. U. S. A.

Dabrowska, R., and Hartshorne, D. J . (1978) Biochem. Biophys.

Hathaway, D. R., and Adelstein, R. S. (1979) Proc. Natl. Acad.

Hathaway, D. R., Adelstein, R. S., and Klee, C. B. (1981) J. Biol.

Yerna, M-J., Dabrowska, R., Hartshorne, D. J., and Goldman, R.

Cheung, W. Y. (1980). Science 207, 19-27

(1981) Science 211,495-497

77, 5852-5855

Res. Commun. 85,1352-1359

Sci. U. S. A . 76, 1653-1657

Chem. 256, in press

D. (1979) Proc. Natl. Acad. Sci. U. S. A. 76, 184-188

27. Wolff, D. J., and Brostrom, C. 0. (1979) Adu. Cyclic Nucleotide

28. Wang, J. H., and Waisman, D. M. (1979) Curr. Top. Cell. Regul.

29. Klee, C. B., Crouch, T. H., and Richman, P. G. (1980) Annu. Reu.

30. Adelstein, R. S., Conti, M. A,, Hathaway, D. R., and Klee, C. B.

31. Conti, M. A,, and Adelstein, R. S. (1981) J. Biol. Chem. 256,

32. Klee, C. B. (1977) Biochemistry 16, 1017-1024 33. Klee, W. A,, and Richards, F. M. (1957) J. Biol. Chem. 229,489-

3 4 . Perrie, W. T., and Perry, S. V. (1970) Biochem. J. 119,31-38 35. Adelstein, R. S., Pollard, T. D., and Kuehl, W. M. (1971) Proc.

36. Scordilis, S. P., and Adelstein, R. S. (1978) J. Biol. Chem. 253,

37. Simpson, R. J., Neuberger, M. R., and Liu, T.-Y. (1976) J. Biol.

38. Weiner, A. M., Platt, T., and Weber, K. (1972) J. Biol. Chem.

39. Richman, P. G., and Klee, C. B. (1978) J. Biol. Chem. 253,6323-

40. Nee, C. B., and Krinks, M. H. (1978) Biochemistry 17, 120-126 41. Laemmli, U. K. (1970) Nature 227,680-685 42. Schachman, H. K., and Edelstein, S. J. (1966) Biochemistry 5,

43. Yphantis, D. A. (1964) Biochemistry 3, 297-317 44. Crouch, T. H., and Klee, C. B. (1980) Biochemistry 19,3692-3698 45. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J .

46. Sobieszek, A., and Bremel, R. D. (1975) Eur. J. Biochem. 55,49-

47. Pato, M. D., and Adelstein, R. S . (1980) J. Bwl. Chem. 255,6535-

48. Laurent, T. C., and Killander, J . (1964) J. Chromatog. 14, 317-

49. Edelhoch, H. (1967) Biochemistry 6, 1948-1954 50. Dabrowska, R., Sherry, J . M. F., Aromatorio, D., and Hartshorne,

D. J. (1978) Biochemistry 17, 253-258 51. Watterson, D. M., Sharief, F., and Vanaman, T. C. (1980). J. Biol.

Chem. 255,962-975 52. Walsh, M. P., Vallet, B., Cavadore, J-C., and Demaille, J . G.

(1980) J. Biol. Chem. 255,335-337 53. Davies, G. E., and Stark, G. R. (1970) Proc. Natl. Acad. Sci. U.

S. A . 66,651-656 5 4 . Grand, R. J . A., Perry, S. V., and Weeks, R. A. (1979) Biochem.

55. Nairn, A. C., and Perry, S. V. (1979) Biochem. J. 179,89-97 56. Walsh, M. P., Guilleux, J-C., and DemaiLle, J. G. (1981) Adu.

57. Matsuda, G., Suzuyama, Y., Maita, T., and Umegane, T. (1977)

58. Mrwa, U., and Hartshorne, D. J . (1980) Fed. Proc. 39, 1564-1568 59. Blumenthal, D. K., and Stull, J . T. (1980) Biochemistry 19,5608-

5614 60. Klee, C. B., Crouch, T. H., and Krinks, M. H. (1979) Biochemistry

18, 722-729 61. La Porte, D. C., Toscano, W. A,, Jr., and Storm, D. R. (1979)

Biochemistry 18,2820-2825 62. Huang, C. Y., Chau, V., Chock, P. B., Wang, J. H., and Sharma,

R. K. (1981) Proc. Natl. Acad. Sci. U. S. A . 78, 871-874 63. Klee, C. B., and Haiech, J. (1980). Ann. N . Y. Acad. Sci. 356,43-

54 64. Edsall, J . T. (1953) in The Proteins (Neurath, H., and Bailey K.,

eds) Vol. 1, Part B, pp. 549-726. Academic Press, New York 65. Cohn, E. J., and Edsall, J. T. (1943) Proteins, Amino Acids and

Peptides, p. 375, Van Nostrand and Reinhold Co., New York

Res. 11, 27-88

15,47-107

Biochem. 49,489-515

(1978) J. Biol. Chem. 253, 8347-8350

3178-3181

504

Natl. Acad. Sci. U. S. A . 68, 2703-2707

9041-9048

Chem. 251, 1936-1940

247,3242-3251

6326

2681-2705

(1951) J. Biol. Chem. 193, 265-275

60

6538

330

J. 177,521-529

Cyclic Nucleotide Res. 14, in press

FEBS Lett. 84, 53-56

of chemistry no. 7501-7509, 1981 m purification …purification and characterization of smooth...

Documents