Purification and Characterization of a Myosin Heavy Chain Kinase from Dict yostelium discoideum"

A Dictyostelium discoideum myosin heavy chain ki- nase has been purified 11,000-fold to near homogeneity. The enzyme has a M, = 130,000 as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and greater than 700,000 as determined by gel filtration on Bio-Gel A-1.Sm. The enzyme has a specific activity of 1 pmollmin-mg when assayed at a Dictyostelium myosin concentration of 0.3 mg/ml. A maximum of 2 mol of phosphate/mol of myosin is in- corporated by the kinase, and the phosphorylated amino acid is threonine. Phosphate is incorporated only into the myosin heavy chains, not into the light chains. The actin-activated Mg2+-ATPase of Dictyostelium myosin is inhibited 7040% following maximal phosphorylation with the kinase. The myosin heavy chain kinase requires 1-2 mM Mg2+ for activity and is most active at pH 7.0-7.5. The activity of the enzyme is not significantly altered by the presence of Ca2+, Ca2' and calmodulin, EGTA, CAMP, or cGMP. When incubated with Mg2+ and ATP, phosphate is incorpo- rated into the myosin heavy chain kinase, perhaps by autophosphorylation.

Myosins from many muscle and nonmuscle sources are phosphorylated in uiuo (reviewed in Refs. 1 and 2). Depending on the source of the myosin, the phosphate may be incorporated into either or both of the light chain and heavy chain subunits of the myosin. Phosphorylation of the light chains of myosin from smooth muscle tissue (3) or from mammalian nonmuscle sources (4-9) serves both to increase the actinactivated ATPase activity of the myosin and promote its assembly into bipolar filaments. The heavy chains of most mammalian nonmuscle myosins are also phosphorylated in uiuo (8, [10][11][12][13], but at present little is known concerning the effects of heavy chain phosphorylation on the properties of these myosins. The role of heavy chain phosphorylation in regulating myosin activity is better characterized for the myosins isolated from two lower eukaryotes, Dictyostelium discoideum (14) and Acanthumoeba castellanii (15)(16)(17)(18). In both cases, phosphorylation of the heavy chain inhibits the actinactivated ATPase activity of the myosin and its ability to polymerize into bipolar filaments. The heavy chain phosphorylation sites for Dictyostelium (19,20) and Acanthamoeba (21,22) myosin as well as for rabbit macrophage myosin (23) are located on the rodlike tail section of the molecules and are 100 nm or more in distance from the globular head regions *This work was supported by the Medical Research Council of Canada. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
' $ Recipient of a Medical Research Council of Canada scholarship.
which contain the actin-binding and ATPase sites. It has therefore been suggested that heavy chain phosphorylation acts by an intermolecular mechanism, perhaps by altering the packing of myosin molecules within a bipolar filament (24). At present little is known concerning the properties of any of the myosin heavy chain kinases, although such kinases have been partially purified from Acanthumoeba (16), Dictyostelium (25,26), and rat brain (27). We report here the purification to near homogeneity of a Dictyostelium myosin heavy chain kinase. The purified kinase incorporates 1 mol of phosphate into each of the Dictyostelium myosin heavy chains, thereby inhibiting the actin-activated ATPase activity of the myosin. The kinase elutes from gel filtration columns with an apparent molecular weight in excess of 700,000 but electrophoreses on SDS'-polyacrylamide gels as a single band of M , = 130,000 which, in the presence of Mg2' and ATP, is phosphorylated.

EXPERIMENTAL PROCEDURES
[y3'P]ATP was from ICN, and scintillation fluid (ACS 11) was from Amersham Corp. ATP, histone 2A, hydrolyzed casein, phosvitin, ovalbumin, bovine serum albumin for protein standards, imidazole (Grade III), and protease inhibitors were from Sigma. The reagent for the Bradford assay (33) was from Bio-Rad. Hog brain calmodulin was from Boehringer Mannheim. Smooth muscle myosin, the isolated light chains of smooth muscle myosin (28), and the catalytic subunit of smooth muscle phosphatase I (29) were prepared from turkey gizzard according to procedures previously described and were gifts of Dr. M. Pato (University of Saskatchewan). Platelet myosin was prepared from 7-day-old human platelets by a procedure similar to that described here for Dictyosteliurn myosin.
Preparation of Actomyosin Precipitate and Supernatant-Dictyostetium were grown in 15-liter carboys and harvested as previously described (30). The cells (300 g from two carboys) were washed once at room temperature in 1 liter of 20 mM KCI, 10 mM KP04, 1 mM MgCl', pH 6.5, and then were placed on ice. All remaining procedures were carried out at 0-4 "C. Cells were suspended in 600 ml of extraction buffer (30% sucrose, 25 mM sodium pyrophosphate, 10 mM imidazole, 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 10 mg/liter pepstatin, 2 mg/liter leupeptin, pH 7.5) and broken open by sonication. Diisopropyl fluorophosphate was then added to a concentration of 0.5 mM, and the extract was stirred for 30 min before being centrifuged in two Ti-45 rotors (1.5 h, 30,000 rpm). The clear supernatant was collected, added to 100 g of packed DE52-cellulose (Whatman) (equilibrated in 20 mM KC1, 10 mM imidazole, 5 mM sodium pyrophosphate, 1 mM dithiothreitol, pH 7.5), and stirred for 30 min after which the DE52 was removed by centrifugation. The DE52-treated extract was dialyzed overnight against 16 liters of 60 mM KCI, 10 mM triethanolamine, 1 mM EDTA, 2 mM mercaptoethanol, pH 7.5. The actomyosin precipitate which formed was collected by centrifugation and used for the preparation of Dictyostelium myosin, whereas the remaining extract was used for purification of the myosin heavy chain kinase.
Purification of Dictyostelium

1065
was washed once in 200 ml of dialysis buffer and then homogenized in 150 ml of G-buffer (5 mM imidazole, 0.4 mM ATP, 0.1 mM MgC12, 0.25 mM dithiothreitol, pH 7.5). Following centrifugation in a Ti-60 rotor (15 min, 20,000 rpm), the pellet was discarded; and the actomyosin, which remained in the supernatant, was precipitated by the addition of 40 mM NaCl and collected by centrifugation. The solubilization in G-buffer, followed by precipitation with salt, was repeated four times and removed most of the actin from the actomyosin. The myosin-enriched actomyosin was dissolved in 15 ml of 1 M KCl, 10% sucrose, 20 mM imidazole, 10 mM ATP, 2 mM MgC12, 1 mM dithiothreitol, pH 7.5, buffer, centrifuged in a Ti-70 rotor (1 h, 60,000 rpm), and loaded onto a 2.6 X 100-cm Sepharose CL-4B column (Pharmacia P-L Biochemicals) (equilibrated in 0.6 M KCl, 5% sucrose, 10 mM imidazole, 1 mM dithiothreitol, pH 7.0) into which has been prerun 100 ml of the ATP-containing buffer. The peak myosin-containing fractions were pooled, passed over a 10-ml column of Dowex 1-X2, and then dialyzed against 0.1 M KC1,20% sucrose, 10 mM imidazole, 1 mM dithiothreitol, pH 7.0. Any precipitate which formed was removed by centrifugation and represents myosin contaminated with a small amount of actin (usually 30-40% of the total myosin). The supernatant, containing myosin free of actin, was concentrated by dialysis against solid sucrose (yield: 10 mg/100 g of wet packed Dictyostelium) (Fig. 6). Myosin Heavy C h i n Kinase Purification-The extract remaining after removal of the actomyosin precipitate was applied at 200 ml/h to a 5 X 12-cm column of phosphocellulose (Whatman P-11) washed according to the manufacturer's instructions and equilibrated with 60 mM KC1, 10 mM imidazole, 1 mM dithiothreitol, pH 7.5. The column was washed with starting buffer until the absorbance at 280 nm dropped to baseline and then the flow rate was reduced to 60 ml/ h and the kinase was eluted with a 800-ml gradient to 0.36 M KC1 (Fig. 1). Fractions containing myosin heavy chain kinase activity were pooled, 1 mM EDTA was added, and the kinase was precipitated by the addition of 60% ammonium sulfate. The 0-60% precipitate was dissolved in 8-10 ml of buffer containing 20% sucrose, 0.3 M KCl, 10 mM imidazole, 0.2 mM EDTA, 1 mM dithiothreitol, pH 7.0, dialyzed overnight against this buffer, and then clarified by centrifugation in a Ti-70 rotor (1 h, 40,000 rpm) and chromatographed at 12 ml/h over a 1.6 X-100 cm Bio-Gel A-1.5 m column equilibrated in the same buffer (Fig. 2). Fractions containing kinase activity were pooled, dialyzed overnight against 20% sucrose, 0.1 M KCl, 50 mM KPO4, 2 mM dithiothreitol, pH 7.0, and applied at 10 ml/h to a 1.0 X 3-cm column of hydroxylapatite (Bio-Rad HT) equilibrated in the same buffer. Myosin heavy chain kinase was eluted with a 40-ml gradient to 0.2 M KPO, (Fig. 3). Kinase pooled from the hydroxylapatite column was dialyzed against 20% sucrose, 0.1 M KCl, 10 mM imidazole, 2 mM dithiothreitol, pH 7.0, and further purified over a 0.5-ml column of aminohexyl-Sepharose 4B (Pharmacia P-L Biochemicals) equilibrated in the same buffer. The myosin heavy chain kinase bound to the column and was eluted stepwise by increasing the KC1 concentration to 0.20 M.
Myosin Heavy Chain Kinase Assay-Unless stated otherwise, assays for myosin heavy chain kinase activity were performed at 25 'C in a reaction mixture containing 2 mM MgCl2, 10 mM imidazole, 0.5 mM [-y-32P]ATP (50 pCi/pmol) 1 mM dithiothreitol, pH 7.0 (Kbuffer). Dictyostelium myosin (2.0-2.5 mg/ml in 60% sucrose, 0.1 M KCl, 10 mM imidazole, 1 mM dithiothreitol, pH 7.0) was added to the assay mixture to give a final concentration of 0.3 mg/ml. Assays were initiated by addition of the kinase sample to the mixture, and the reaction was stopped by spotting 45-pl samples onto squares of Whatman No. 3MM filter paper which were then washed in 10% trichloroacetic acid, 1% sodium pyrophosphate as described (16) before being counted in ACS 11. To determine specific activities, samples were routinely taken after 30, 60, 90, and 120 s, and kinase concentrations were chosen so that 0.3-0.8 mol of phosphate/mol of myosin was incorporated by 120 s. Under these conditions, assays were linear with respect to time and enzyme concentration. Background activity was corrected for by performing identical assays but in the absence of myosin. To determine myosin heavy chain kinase activity in crude fractions, samples of the phosphorylation mixture were electrophoresed on 8% SDS-polyacrylamide gels. After staining and destaining the gel, the myosin heavy chain band was cut out and counted in ACS 11. An equivalent area of the gel for assays performed in the absence of myosin was counted to correct for background phosphorylation.
Preparation of Phspholylated Dictyostelium Myosin-Dictyostelium myosin (400 pl of 2.2 mg/ml in 60% sucrose, 0.1 M KC1, 10 mM imidazole, 1 mM dithiothreitol, pH 7.0) was diluted three times with K-buffer, and 100 pl of myosin heavy chain kinase (0.18 mg/ml) purified through the hydroxylapatite column step (200 nmol/min . mg specific activity) was added. After 20 min at 25 "C, 0.5 M KC1 and 5 mM ATP were added. The myosin was separated from kinase and ATP by passage over a 1 X 30-cm Sepharose CL-4B column, dialyzed, and concentrated as described under "Purification of Dictyostelium Myosin." Phsphoamirw Acid Analysis-Phosphorylated Dictyostelium myosin (2.2 mol of phosphate/mol of myosin) was dialyzed against 10 mM NH4HC03, lyophilized, and subjected to partial acid hydrolysis in 6 N HCl for 2 h at 110 "C in vacuo. The hydrolysate was lyophilized to remove HC1, mixed with authentic phosphoserine and phosphothreonine, and analyzed by electrophoresis at pH 1.9 as described (16).
Phosphatase MisceUuneous Methods-Protein concentrations were determined using the colorimetric assay of Bradford (33) with bovine serum albumin as a standard. The values determined for purified Dictyostelium myosin using the Bradford assay were 15% lower than values determined using the Markwell modification of the Lowry method (34) with bovine serum albumin as a standard. Discontinuous SDSpolyacrylamide gel electrophoresis was performed as described by Laemmli (35). Coomassie Blue-stained gels were scanned using an LKB 2202 Ultroscan laser densitometer. Peaks were cut out and weighed to determine areas. For autoradiography of 32P-labeled proteins and phosphoamino acids, dried gels and thin layer sheets were exposed to x-ray film (Kodak X-Omat AR-2) with an intensifying screen (Du Pont Cronex Lightning Plus). Acid-stable phosphate was determined following ashing of 0.3 mg of myosin using the method of Stull and Buss (36) as described by Collins and Korn (15).

RESULTS
Purification of Dictyostelium Myosin-Previous procedures for the purification of Dictyostelium myosin have depended on the use of KI to solubilize the actomyosin precipitate prior to gel filtration chromatography (26,37). If, however, the crude Dictyostelium extract is treated batchwise with DE52, the actomyosin precipitate which subsequently forms is readily solubilized with KC1 and ATP once the majority of actin has been removed. Dictyostelium myosin prepared by this method has ATPase activities similar to those reported by Kuczmarski and Spudich (14) for Dictyostelium myosin prepared using the KI treatment (Ca2+-ATPase: 1.0-1.2 pmol/ mine mg, maximal actin-activated ATPase: 0.15-0.25 pmol/ min . mg). Dictyostelium myosin prepared by the method reported here was used for all experiments in this paper.
Purification of a Dictyostelium Myosin Heavy Chain Kinase-Extraction of Dictyostelium in a buffer containing 30% sucrose and 25 mM sodium pyrophosphate solubilizes both Dictyostelium myosin and the myosin heavy chain kinase. Following dialysis to remove sucrose and pyrophosphate, an actomyosin precipitate was formed which was removed by centrifugation. No myosin heavy chain kinase activity was observed in the actomyosin precipitate. Quantitation of the specific activity of the kinase present in the initial extract and the actomyosin supernatant for several different preparations consistently gave values of 60-80 and 40-50 pmol/ minmg, respectively (Table I). These values were not altered if the phosphatase inhibitors sodium fluoride and sodium pyrophosphate were added to the standard assay mixture.
Although greater than 95% of the protein present in the actomyosin supernatant does not bind to phosphocellulose, no myosin heavy chain kinase activity is detectable in the flow through from this column. A single peak of kinase activity is eluted from the column at a KC1 concentration of 0.15 M (Fig. 1). The fractions were pooled so as to avoid the major peak of protein which eluted on the trailing edge of the myosin heavy chain kinase activity. Almost 70% of the total kinase activity present in the actomyosin supernatant is recovered in the phosphocellulose column pool (Table I). Following precipitation with 60% ammonium sulfate, the myosin heavy chain kinase activity was applied to a column of Bio-Gel A-1.5m and was recovered as a single peak of activity eluting close to the void volume of the column. The kinase activity elutes before thyroglobulin, a protein of M, = 670,000 (Stokes radius = 8.5 nm) (Fig. 2). It is unlikely that the early elution position of the myosin heavy chain kinase is due to nonspecific aggregation since the column is normally run in buffer containing 0.3 M KC1 and an identical elution position is obtained when buffer containing 0.6 M KC1 and 2 mM EDTA is used. Chromatography over a hydroxylapatite column (Fig. 3) yields a myosin heavy chain kinase fraction consisting of three major proteins (Fig. 4, lane E ) with M,, as estimated from SDSpolyacrylamide gels by plots of log M, uersus relative mobility for molecular weight standards, of 130,000, 105,000, and 100,000. Scanning densitometry indicates that these bands represent 25, 18, and 14%, respectively, of the total protein present. When fractions across the peak of kinase activity obtained from the hydroxylapatite column are electrophoresed on SDS-polyacrylamide gels, only the intensity of the 130,000-dalton band correlates with the amount of kinase activity present in each fraction (data not shown). The other protein bands are more prominent in either earlier or later fractions from the hydroxylapatite column, and none are present across the peak of kinase activity in a constant ratio to the 130,000-dalton band. It should be noted that due to the presence of nucleic acid, the absorbance at 280 nm gives a poor indication of the amount of protein present in fractions eluting from the hydroxylapatite column. Dictyosteliummyosin heavy chain kinase purified through the hydroxylapatite column step was not contaminated with detectable levels of either a protease or a protein phosphatase as judged by incubating the kinase with 32P-labeled Dictyostelium myosin.
When myosin heavy chain kinase pooled from the hydroxylapatite column was rerun over a 1 x 50-cm column of Bio-Gel A-1.5m, the kinase activity again eluted earlier than thyroglobulin and again was correlated with the appearance of the 130,000-dalton protein band (data not shown).
Following chromatography over aminohexyl-Sepharose 4B, scanning densitometry indicated that the 130,000-dalton band comprises 80% and the 105,000-dalton band 8% of the total protein present (Fig. 4, lane F). No other band represents more than about 1% of the total protein. The specific activity of the aminohexyl column fraction is close to 1 pmol/min.mg when assayed using Dictyostelium myosin at a concentration of 0.3 mg/ml. This represents a 14,000-fold increase in specific activity over the initial extract with a yield of close to 7%. Kinase at this stage of the purification retains full activity for several weeks if stored at 0 "C in 60% sucrose, 0.1 M KC1, 10 mM imidazole, 2 mM dithiothreitol, pH 7.0. At any stage of the purification, the myosin heavy chain kinase rapidly loses activity if exposed to low salt (50 mM KC1 or less) in the absence of sucrose.
The 130,000-dalton myosin heavy chain kinase band is phosphorylated when incubated with M$+ and ATP and at the same time its mobility on SDS-polyacrylamide gels is reduced (Fig. 4, lanes G and H). Preliminary results indicate that a maximum of 2 mol of phosphate can be incorporated per mol of the 130,000-dalton band.

Optimal Conditions for Myosin Heavy C h i n Kinase
Actiuity-The myosin heavy chain kinase displays optimal activity between pH 7.0 and 7.5 (Fig. 5A). Specific activities for the kinase at pH 8.0 and above were difficult to determine accurately because of a downward curve in the time courses, suggesting that the kinase is unstable at high pH. In contrast, at pH values below 7.0, the time courses remained linear even though the rate of phosphate incorporation decreased.
The myosin heavy chain kinase has no activity in the absence of divalent cations but is fully active in the presence of 1-2 mM M P (Fig. 5B). In the absence of M$+, neither Ca2+ nor Mn2+ (1 or 5 mM) could support kinase activity. The slow decline in kinase activity as the MgCl, concentration is increased is probably an effect of ionic strength. The kinase activity is strongly inhibited by increasing concentrations of KC1 (Fig. 5C). If the activity at 15 mM KC1 is taken as loo%, the myosin heavy chain kinase is 90% inhibited at 60 mM KC1, and no activity could be detected at 100 mM KCl.
The assays for myosin heavy chain kinase activity therefore must be performed at low ionic strength, under conditions where Dictyostelium myosin is filamentous. We have found the assays to be most reproducible if soluble myosin in 60% sucrose, 0.1 M KC1 (usually 2.0-2.5 mg/ml) is diluted 6-or 7-  Purification of myosin heavy chain kinase on phosphacellulose. Chromatography of the actomyosin supernatant was performed on a 5 X 10-cm phosphocellulose column. Fractions of 12 ml were collected. The eluant was monitored for Am, conductivity, and myosin heavy chain kinase activity. Phosphate incorporation into Dictyostelium myosin was determined by the filter paper method as described under "Experimental Procedures." Fractions of 5 pl were incubated with 50 p1 of assay mixture a t 25 "C for 10 min prior to spotting 45 pl onto the filter paper.

Fraction Number
FIG. 2. Chromatography of the myosin heavy chain kinase on Bio-Gel A-1.6m. Chromatography was performed on a 1.6 X 100-cm Bio-Gel A-1.5m column of the 0-60% ammonium sulfate precipitate of the phosphocellulose pool. An 8-ml volume was loaded onto the column, and fractions of 1.9 ml were collected. 5-pl samples were assayed for kinase activity as described for Fig. 1. The column was calibrated with the following protein molecular weight standards: thyroglobulin (Thy), 670,000; y-globulin (yG), 158,000; ovalbumin (Ou), 45,000; myoglobin (My), 17,000; and vitamin B-12,1,350. Standards were dissolved in 8 ml of column buffer and chromatographed over the Bio-Gel column in an identical manner to the kinase sample. V, was determined from the standard run. fold into the assay mixture. While maintaining a low salt concentration in the assays, this method produces small, evenly dispersed aggregates of myosin so that samples taken from the assay at different times contain a constant amount of myosin. We have attempted to perform assays at higher myosin concentrations by using Dictyostelium myosin which has been dialyzed against a low ionic strength buffer and then concentrated by centrifugation. However, this method produces large aggregates of myosin which tend to stick to glass or plastic, and reproducible amounts of myosin are not recovered from the assays.
The rate of phosphorylation of Dictyostelium myosin by the myosin heavy chain kinase was affected less than 20% by Chromatography of the Bio-Gel A-1.5m kinase pool was performed on a 1.0 X 3.0-cm hydroxylapatite column. The column was eluted with a 40-ml gradient from 50 mM to 0.2 M K 2 P 0 , pH 7.0. The eluant was monitored for Am, ionic strength, kinase activity, and protein using the Bradford assay (33). Samples of 5 pl were taken to assay kinase activity as described for Incorporation of Phosphate into Dktyostelium Myosin-The myosin heavy chain kinase incorporates phosphate only into the heavy chains, not into the light chains of Dictyostelium myosin (Fig. 6). Partial chymotryptic digestion, carried out as described by Peltz et al. (19), demonstrates that all of the phosphate is present in the tail region of the myosin (data not shown). Only phosphothreonine is detected by partial acid hydrolysis (Fig. 6). C, KC1 was added to increase the final KC1 concentration in the assay from 15 to 100 mM. Each assay, 200 pl in volume, was initiated by the addition of 5 pl of hydroxylapatite-purified kinase (0.09 mg/ml); and samples of 45 pl were taken at 30,60,90, and 120 s for the filter paper assay as described under "Experimental Procedures.'' Maximum activity for the kinase was 190 nmol/min.mg. To determine rates near 100% activity, only the first two time points were used as phosphate incorporation approached 1 mol/mol of myosin by 90 s.
When Pi incorporation is measured by the filter paper assay and protein concentration is measured by the Bradford assay (33), a maximum of 2.2 mol of phosphate is calculated to be incorporated per mol of Dictyostelium myosin (Fig. 7). If the Markwell-Lowry assay ( (2.4 mg/ml), and the reaction was initiated by the addition of 25 pl of hydroxylapatite column-purified myosin heavy chain kinase (0.09 mg/ml). Samples of 45 pl were taken at the indicated times and spotted onto Whatman 3MM filter papers which were washed and counted as described under "Experimental Procedures." An additional 5 pl of kinase was added to the assay at 6 min. Samples identical to these but containing the appropriate buffer instead of myosin were used to correct for radioactivity incorporated into the kinase fraction alone; this represented less than 5% of the incorporation obtained in the presence of myosin for each time point. At 2, 7, and 11 min, 45pl samples were taken for protein concentration determinations by the Bradford assay (33). All assays indicated a protein concentration of 0.3 mg/ml. Stoichiometry was calculated using M, of 500,000 for Dictyostelium myosin. myosin concentration, the stoichiometry becomes 1.9 mol of phosphate/mol of myosin. Identical results are obtained when phosphate incorporation into Dictyostelium myosin is measured directly using myosin phosphorylated by the myosin heavy chain kinase in the presence of [y-32P]ATP and then Dictyostelium Myosin Heavy Chain Kinase separated from the free ATP and kinase by chromatography over Sepharose CL-4B.
The Dictyostelium myosin used as a substrate in these assays (isolated as described under "Experimental Procedures") appears to contain little covalently bound phosphate. Measurement of acid-stable phosphate indicated the presence of only 0.5-1.0 mol of phosphate/mol of myosin. We have found that the catalytic subunit of smooth muscle phosphatase I rapidly removes 32P which has been incorporated into the Dictyostelium myosin heavy chain by the myosin heavy chain kinase. When incubated with the smooth muscle phosphatase as described under "Experimental Procedures," fully phosphorylated Dictyostelium myosin loses 75% of its heavy chain phosphate in 5 min and 95% of the phosphate in 30 min. Dictyostelium myosin which had been dephosphorylated by a 60-min incubation with the catalytic subunit of smooth muscle phosphatase I again incorporated 2 mol of phosphate/ mol of myosin when incubated with the myosin heavy chain kinase.
Incorporation of 2 mol of phosphate/mol of Dictyostelium myosin by the myosin heavy chain kinase inhibits the actinactivated ATPase activity of the myosin by about 70-80% at all actin concentrations (Fig. 8). Treatment of the fully phosphorylated myosin with the smooth muscle phosphatase returns the actin-activated ATPase activity of the myosin to its original level (Fig. 8).
Substrate Specificity-As stated above, the insolubility of Dictyostelium myosin limits the concentration of myosin which can be used in the myosin heavy chain kinase assay. Increasing the myosin concentration from 0.3 to 0.6 mg/ml increases the specific activity of the myosin heavy chain kinase by 60%, indicating that the kinase is being assayed as a myosin concentration well below V,,,. Nevertheless, even at low concentrations (less than 1 pM) of Dictyostelium myosin, the kinase was much more active toward the myosin than toward several generally used phosphate-accepting proteins (histone, casein, phosvitin, and smooth muscle myosin light chaiis present in much higher molar amounts) ( Table  11). The Dictyostelium myosin heavy chain kinase also did not FIG. 8. Actin-activated ATPase activity of Dictyostelium myosin before and after phosphorylation. Shown is actin-activated Mg+-ATPase activity of Dictyostelium myosin before (0) and after (0) incorporation of 2.2 mol of phosphate/mol of myosin by the myosin heavy chain kinase. Assays were performed for 20 min at 30 "C with varying concentrations of rabbit skeletal muscle actin as described under "Experimental Procedures." Each assay contained 13 pg of myosin in a 300-pl volume. Ca2'-ATPase activity for the myosin was 1.1 pmol/min.mg before phosphorylation and 1.0 pmol/ min. mg after phosphorylation. Myosin was phosphorylated, separated from the kinase by Sepharose CL-4B chromatography, dialyzed, and concentrated as described under "Experimental Procedures." X represents the activity of phosphorylated myosin following treatment for 30 min with smooth muscle phosphatase.

Substrate specificity of Dictyostelium myosin heavy chain kinase
The initial rate of phosphorylation of the indicated substrates was measured in 10 mM imidazole, pH 7.0, 2 mM MgC12, 1 m M dithiothreitol, 0.5 mM [Y-~ZPIATE' (50 pCi/pmol) with 5 pg/ml hydroxylapatite column-purified kinase and the indicated concentration of substrate.

Substrate
Specific incorporate phosphate into either turkey gizzard smooth muscle myosin or human platelet myosin.

DISCUSSION
Dictyostelium myosin is composed of a pair of 210,000dalton heavy chains and two pairs of light chains of 18,000 and 16,000 daltons. In vivo both the heavy chain and 18,000dalton light chain of Dictyostelium myosin are phosphorylated (14). At present, the effects of light chain phosphorylation on the biological activity of Dictyostelium myosin remain to be elucidated. However, Kuczmarski and Spudich (14), using a partially purified Dictyostelium myosin heavy chain kinase, were able to show that incorporation of phosphate into the Dictyostelium myosin heavy chains inhibited the actin-activated M$"ATPase activity of the myosin and its ability to self-assemble into filaments. The phosphorylation of myosin heavy chains therefore probably plays an important role in regulating contractile activity in Dictyostelium as well as in other lower eukaryotes, such as A. castellunii (15), and perhaps also in most mammalian nonmuscle tissues (23). Hammer et al. (38) have described the purification of a kinase which phosphorylates the heavy chain of the single-headed Acanthamoeba myosin I enzymes. We report here the first purification to near homogeneity of a heavy chain kinase responsible for the phosphorylation of a conventional double-headed myosin.
The myosin heavy chain kinase is present in Dictyostelium in levels high enough so that its activity can readily be assayed in crude extracts or dialyzed extracts of the cells. It can be calculated that if myosin represents close to 1% of the total Dictyostelium protein (39) (20 pmol of myosin/mg of cellular protein), the activity of the myosin heavy chain kinase measured in crude extracts (70 pmol/min. mg) would be sufficient to fully phosphorylate the myosin heavy chains in less than 1 min (assuming a rate of myosin heavy chain phosphorylation in vivo similar to that assayed in vitro).
Using four columns and an ammonium sulfate precipitation step, the Dictyostelium myosin heavy chain kinase has been purified 14,000-fold based on the increase in specific activity.
Throughout the purification procedure, the Dictyostelium kinase activity behaves as a single enzymatic species. The high level of myosin heavy chain kinase activity recovered from the initial phosphocellulose column, as compared to that present in crude Dictyostelium extracts, suggests that this single enzyme can account for the large majority of myosin heavy chain kinase activity in Dictyostelium. Although we cannot rule out the possibility that Dictyostelium contains other myosin heavy chain kinases which are inactive under our assay conditions, the myosin heavy chain kinase we have purified appears to be the kinase which in vivo is responsible for regulating Dictyostelium myosin activity. Incorporation of 2 mol of phosphate/mol of Dictyostelium myosin by the purified kinase significantly inhibits the actin-activated M$+-ATPase of Dictyostelium myosin (Fig. 8). It is interesting to note that fully phosphorylated Dictyostelium myosin still retains a low degree of actin-activated M$+-ATPase activity; however, as has been shown with Acanthamoeba myosin 11, the effects of heavy chain phosphorylation may be highly dependent on the assay conditions used (17).
The most highly purified myosin heavy chain kinase fraction displays a single major band on SDS-polyacrylamide gels with an estimated M , of 130,000. The evidence that this band represents the myosin heavy chain kinase can be summarized as follows. (a) SDS-polyacrylamide gel analysis of the proteins eluting from the hydroxylapatite column or from the Bio-Gel A-1.5m column when the hydroxylapatite pool is rechromatographed indicates that only the intensity of the 130,000dalton-band correlates with myosin heavy chain kinase activity. (b) Quantitation of scanning densitometry demonstrates that during the last step of purification, the 130,000-dalton band increases from 25 to 80% of the total protein present, in good agreement with the observed increase in specific activity of the kinase, whereas the amount of the 105,000-dalton band drops from 18 to 8%. The 105,000-dalton band, the major contaminant present, therefore cannot represent the kinase. ( c ) Scanning densitometry further indicates that no other band represents more than about 1% of the total protein present in the aminohexyl column fraction. The high specific activity of the myosin heavy chain kinase at this stage of purification makes it very unlikely that any of these minor contaminants could represent the myosin heavy chain kinase. The specific activity of the most highly purified myosin heavy chain kinase (1 pmol/min. mg) compares favorably to that of other protein kinases, especially when it is considered that this activity is measured at a myosin concentration (0.6 p~) which is probably well below that required to saturate the enzyme (Table 11). For comparison, the K,,, of the Acanthamoeba myosin I1 heavy chain kinase for a synthetic peptide corresponding to the Acanthumoeba myosin I1 heavy chain phosphorylation sites is 6 p~ (22). Given the insolubility of Dictyostelium myosin under the assay conditions, it will probably not be possible to accurately measure the kinetic parameters of the Dictyostelium myosin heavy chain kinase until the amino acid sequence around the Dictyostelium myosin heavy chain phosphorylation site is determined and synthetic peptides corresponding to this sequence are prepared.
One of the most interesting questions to be answered is how the activity of the Dictyostelium myosin heavy chain kinase is regulated. It has been shown that the distribution of myosin in Dictyostelium can be rapidly and transiently altered in response to external chemotactic signals (40), and Malchow et al. (41) have demonstrated that chemotactic stimulation of Dictyostelium results in a transient dephosphorylation of the myosin heavy chains. Evidence has been presented to indicate that Dictyostelium myosin heavy chain kinase activity present in crude membrane fractions (41) or in partially purified kinase fractions (26) is inhibited by increased Ca2+ levels in the presence of calmodulin. The myosin heavy chain kinase isolated here requires only M$+ and ATP to display high activity, in agreement with the suggestion that second messengers produced by chemotactic signals might act to inhibit what is normally an active myosin heavy chain kinase. How-ever, we have not yet been able to demonstrate an effect on the purified kinase of any of several common second messengers, including Ca2+ and calmodulin.
The Dictyostelium myosin heavy chain kinase electrophoreses as a 130,000-dalton band on SDS-polyacrylamide gels but elutes from gel filtration columns with an apparent M, in excess of 700,000. This could indicate that the native enzyme is a very asymmetric molecule or that it is an aggregate consisting of several 130,000-dalton subunits. The observation that the kinase can be phosphorylated suggests that this could be a potential mechanism for regulating the activity of the enzyme. As with many other protein kinases, the myosin heavy chain kinase may autophosphorylate; however, as the kinase is not completely pure, further studies will be required to confirm this point. At present it is not clear what effect phosphorylation has on the activity of the myosin heavy chain kinase since it is phosphorylated quite rapidly under the same conditions used to assay for incorporation of phosphate into myosin.