Phosphorylation of an actin.tropomyosin.troponin complex from human skeletal muscle.

A human skeletal actin.tropomyosin.troponin complex was phosphorylated in the presence of [gamma-32 P]ATP, Mg2+, adenosine 3':5'-monophosphate (cyclic AMP) and cyclic AMP-dependent protein kinase (protein kinase). Phosphorylation was not observed when the actin complex was incubated in the absence of protein kinase or 1 microM cyclic AMP. In the presence of 10(-7) M Ca2+ and protein kinase 0.1 mole of [32P]phosphate per 196 000 g of protein was incorporated. This was two-fold higher than the [32P]phosphate content of a rabbit skeletal actin complex but two-fold lower than that of a bovine cardiac actin complex. At high Ca2+, 5.10(-5) M, little change in the phosphorylation of a human skeletal actin complex occurred. Phosphoserine and phosphothreonine were identified in the [32P]phosphorylated actin complex. Polyacrylamide gel electrophoresis in sodium dodecyl sulfate showed that 60% of the label was associated with the tropomyosin binding component of troponin. The inhibitory component of troponin contained 16% of the bound [32P]phosphate. Increasing the Ca2+ concentration did not significantly decrease the [32P]phosphate content of the phosphorylated proteins in the actin complex. No change in the distribution of phosphoserine or phosphothreonine was observed. Half maximal calcium activation of the ATPase activity of reconstitute human skeletal actomyosin made with the [32P] phosphorylated human skeletal actin complex was the same as a reconstituted actomyosin made with an actin complex incubated in the absence of protein kinase at low or high Ca2+.     

Glucagon-induced phosphorylation of pyruvate kinase (type L) in rat liver slices.

The effect of glucagon on the phosphorylation of pyruvate kinase in ³²P-labelled slices from rat liver was investigated. Pyruvate kinase was isolated by immunoadsorbent chromatography. The enzyme was partially phosphorylated in the absence of added hormone (0.2 mol of phosphate/mol of enzyme subunit). Upon incubation with 10^?7 M glucagon, the incorporation of [³²P]phosphate was 0.6–0.7 mol/mol of enzyme subunit. Concomitantly, the concentration of intracellular cyclic 3′,5′-AMP increased from 0.3 to 3.2 μM. The phosphorylation inhibited the enzyme activity at low concentrations of phosphoenolpyruvate (60% at 0.5 mM). Almost maximal phosphorylation of the enzyme was reached within 2 min after the addition of glucagon. The concentration of hormone giving half maximal effect on the pyruvate kinase phosphorylation was about 7×10^?9M. The inactivation of the enzyme paralleled the increase in phosphorylation. It is concluded that pyruvate kinase is phosphorylated in the intact liver cell.     

Autophosphorylation of chick brain synaptic polypeptides. Phosphorylation of proteins during incubation of intact synaptosomes with 32PO4

Chick brain synaptosomes incorporated phosphate into proteins when incubated in physiological buffer containing energy sources. Sodium dodecyl sulfate polyacrylamide gel electrophoresis indicated that three synaptosomal polypeptides were significantly phosphorylated after 15 sec incubation while at least fifteen polypeptides were active kinase substrates after 15 min incubation. Labeled synaptosomes were hypotonically lysed and separated by centrifugation into soluble, membrane, and mitochondrial fractions. Every fraction exhibited significant phosphate incorporation. Electrophoresis revealed that each fraction had several unique phosphorylated polypeptides and a distinctive phosphorylation pattern. The same polypeptides appear to be labeled whether MgATP was added to synaptic plasma membranes or synaptic plasma membranes were isolated after synaptosomal autophosphorylation.     

Calyculin A-induced actin phosphorylation and depolymerization in renal epithelial cells

This study reports actin phosphorylation and coincident actin cytoskeleton alterations in renal epithelial cell line, LLC-PK1. Serine phosphorylation of actin was first observed in vitro after the cell lysate was incubated with phosphatase inhibitors and ATP. Both the phosphorylated actin and actin kinase activities were found in the cytoskeletal fraction. Actin phosphorylation was later detected in living LLC-PK1 cells after incubation with the phosphatase inhibitor calyculin A. Calyculin A-induced actin phosphorylation was associated with reorganization of the actin cytoskeleton, including net actin depolymerization, loss of cell-cell junction and stress fiber F-actin filaments, and redistribution of F-actin filaments in the periphery of the rounded cells. Actin phosphorylation was abolished by 3-h ATP depletion but not by the non-specific kinase inhibitor staurosporine. These results demonstrate that renal epithelial cells contain kinase/phosphatase activities and actin can be phosphorylated in LLC-PK1 cells. Actin phosphorylation may play an important role in regulating the organization of the actin cytoskeleton in renal epithelium.     

Phosphorylation of rat kidney pyruvate kinase type L by cyclic 3',5'-AMP-dependent protein kinase.

Pyruvate kinase (ATP:pyruvate 2-O-phosphotransferase, EC 2.7.1.40) type L was partly purified from rat kidney. During the last two purification steps, the incorporation of [32P]phosphate into protein on incubation with [32P]ATP and cyclic 3',5'-AMP-dependent protein kinase was found to parallel the pyruvate kinase activity. After phosphorylation of the enzyme, a major radioactive band with a molecular weight of 57 000 was found on polyacrylamide gel electrophoresis [32P]Phosphorylserine was isolated from the kidney pyruvate kinase. Immunological identity was found between the liver and kidney pyruvate kinases type L. By autoradiography of high-voltage electropherograms after partial acid hydrolysis of the phosphorylated rat liver and kidney pyruvate kinases type L, identical results were obtained. The affinity for phosphoenolpyruvate was found to be decreased by phosphorylation of the enzyme with a change in the apparent Km from 0.15 mM to 0.35 mM. After incubation of the phosphorylated kidney pyruvate kinase with phosphatase the phosphoenolpyruvate saturation curve was found to be identical to that for the unphosphorylated enzyme. Thus, the activity of the rat kidney pyruvate kinase type L is with all probability regulated by a reversible phosphorylation-dephosphorylation reaction, thereby indicating that hormonal regulation of gluconeogenesis via cyclic AMP may be of importance in the renal cortex.     

The binding of smooth muscle heavy meromyosin to actin in the presence of ATP. Effect of phosphorylation

Phosphorylation of the 20,000-dalton light chains of smooth muscle heavy meromyosin (HMM) from turkey gizzards results in a large increase in the actin-activated MgATPase activity over that observed with unphosphorylated HMM. In an attempt to define which step in the kinetic cycle is affected by phosphorylation, we have measured the binding of both unphosphorylated and phosphorylated HMM to actin in the presence of ATP using sedimentation. There was only a 4-fold difference in the actin binding constants of unphosphorylated HMM (5.35 x 10(3) M-1) and fully phosphorylated HMM (2.35 x 10(4) M-1). In contrast, the maximum rate of the actin-activated MgATPase activity (Vmax) of phosphorylated HMM was 25 times greater than that for unphosphorylated HMM. These data rule out a mechanism whereby the unphosphorylated light chain of myosin regulates actin-myosin interaction by directly or indirectly blocking the binding of HMM to actin. This implies that some step in the kinetic cycle other than the binding of HMM to actin must be regulated. We have also measured the rate constant for ATP hydrolysis (the initial phosphate burst) under the same conditions and found that this step was very fast compared to the steady state ATPase rate and was unaffected by phosphorylation. This suggests that the step which is regulated by phosphorylation is either phosphate release or a step preceding phosphate release but following ATP hydrolysis.     

Phosphorylation of an actin · tropomyosin · troponin complex from human skeletal muscle

A human skeletal actin · tropomyosin · troponin complex was phosphorylated in the presence of [γ-³²P]ATP, Mg²⁺, adenosine 3′:5′-monophosphate (cyclic AMP) and cyclic AMP-dependent protein kinase (protein kinase). Phosphorylation was not observed when the actin complex was incubated in the absence of protein kinase or 1 μM cyclic AMP. In the presence of 10⁻⁷ M Ca²⁺ and protein kinase 0.1 mole of [³²P]phosphate per 196 000 g of protein was incorporated. This was two-fold higher than the [³²P]phosphate content of a rabbit skeletal actin complex but two-fold lower than that of a bovine cardiac actin complex. At high Ca²⁺, 5 · 10⁻⁵ M, little change in the phosphorylation of a human skeletal actin complex occurred. Phosphoserine and phosphothreonine were identified in the [³²P]phosphorylated actin complex. Polyacrylamide gel electrophoresis in sodium dodecyl sulfate showed that 60% of the label was associated with the tropomyosin binding component of troponin. The inhibitory component of troponin contained 16% of the bound [³²P]phosphate. Increasing the Ca²⁺ concentration did not significantly decrease the [³²P]phosphate content of the phosphorylated proteins in the actin complex. No change in the distribution of phosphoserine or phosphothreonine was observed. Half maximal calcium activation of the ATPase activity of reconstituted human skeletal actomyosin made with the [³²P]phosphorylated human skeletal actin complex was the same as a reconstituted actomyosin made with an actin complex incubated in the absence of protein kinase at low or high Ca²⁺.     

Protein phosphorylation and the exocytosis-like interaction between isolated adrenal medullary plasma membranes and chromaffin granules

The interaction between isolated adrenal medullary plasma membranes and chromaffin granules has been proposed as a cell-free model for exocytosis. Phosphorylation experiments showed that isolated chromaffin granules as well as isolated plasma membranes contain protein kinases and phosphate accepting membranous proteins. Upon joint incubation however, the chromaffin granule-located proteins are preferentially phosphorylated. β-ν-methylene-ATP, a non-hydrolysable analogue, was able to reduce both the plasma membrane-induced release of the soluble chromaffin granular content and the phosphate incorporation into the protein fraction. The results of these experiments on a cell-free model system fit in the hypothesis originating from work on several types of intact cells that the exocytotic event is linked with protein phosphorylation.     

A single substitution of the insulin receptor kinase inhibits serine autophosphorylation in vitro: evidence for an interaction between the C-terminus and the activation loop

We examined the effects of mutations of tyrosine and serine autophosphorylation sites on the dual specificity of the insulin receptor kinase (IRKD) in vitro using autophosphorylation and substrate phosphorylation and phosphopeptide mapping. For comparable studies, the recombinant kinases were overexpressed in the baculovirus system, purified, and analyzed. The phosphate incorporation into the enzymes was in the range of 3-4.5 mol/mol, and initial velocities of autophosphorylation were reduced up to 2-fold. However, the mutation Y1151F in the activation loop inhibited phosphate incorporation in the C-terminal serine residues 1275 and 1309, due to a 10-fold decrease of the initial velocity of serine autophosphorylation. Although the K(M) and V(MAX) values of this mutant were only slightly altered in substrate phosphorylation reactions using a recombinant C-terminal insulin receptor peptide (K(M): Y1151F, 9.9 +/- 0.4 microM; IRKD, 6.1 +/- 0.2 microM; V(MAX): Y1151F, 72 +/- 4 nmol min(-)(1) mg(-)(1); IRKD, 117 +/- 6 nmol min(-)(1) mg(-)(1)), diminished phosphate incorporation into serine residues of the peptide was observed. In contrast, the phosphorylation of a recombinant IRS-1 fragment, which was shown to be phosphorylated markedly on serine residues by IRKD, was not affected by any kinase mutation. These results underline that IRKD is a kinase with dual specificity. The substrate specificity toward C-terminal serine phosphorylation sites can be modified by a single amino acid substitution in the activation loop, whereas the specificity toward IRS-1 is not affected, suggesting that the C-terminus and the activation loop interact.     

Phosphorylation of a single head of smooth muscle myosin activates the whole molecule

Regulatory light chain (RLC) phosphorylation activates smooth and non-muscle myosin II, but it has not been established if phosphorylation of one head turns on the whole molecule. Baculovirus expression and affinity chromatography were used to isolate heavy meromyosin (HMM) containing one phosphorylated and one dephosphorylated RLC (1-P HMM). Motility and steady-state ATPase assays indicated that 1-P HMM is nearly as active as HMM with two phosphorylated heads (2-P HMM). Single-turnover experiments further showed that both the dephosphorylated and phosphorylated heads of 1-P HMM can be activated by actin. Singly phosphorylated full-length myosin was also an active species with two cycling heads. Our results suggest that phosphorylation of one RLC abolishes the asymmetric inhibited state formed by dephosphorylated myosin [Liu, J., et al. (2003) J. Mol. Biol. 329, 963-972], allowing activation of both the phosphorylated and dephosphorylated heads. These findings help explain how smooth muscles are able to generate high levels of stress with low phosphorylation levels.     

PLC-gamma1 signaling pathway and villin activation are involved in actin cytoskeleton reorganization induced by Na+/Pi cotransport up-regulation

BACKGROUND: The brief incubation of opossum kidney (OK) cells with low P(i) results in Na+/P(i) cotransport up-regulation and in substantial, but transient, cytoskeletal reorganization. In this study, we examined signaling events involved in the depolymerization of microfilaments. RESULTS: Confocal laser scanning microscopy, immunoblot and immunoprecipitation experiments revealed villin co-localization with mainly actin short filaments and monomers, indicating that under the conditions used, villin acted as an actin-severing protein. Further analysis revealed that low concentrations of extracellular phosphate resulted in phospholipase Cgammal (PLC-gammal) translocation to the actin cytoskeleton, without increases in its tyrosine phosphorylation. Additionally, tyrosine phosphorylation of a portion of insoluble villin was increased; whereas, only tyrosine phosphorylated villin associated with PLC-gammal. Although, tyrosine phosphorylation of PLC-gammal was not observed during Na+/P(i) cotransport up-regulation, genistein treatment abolished the enzyme's translocation to the actin cytoskeleton, as well as its association with villin. In addition, villin was found to associate with the 85-KDa subunit (p85) of phosphatidylinositol (PI)-3 kinase, concomitant with PLC-gammal, in the cytoskeletal fraction of Na+/P(i) cotransport up-regulated cells. CONCLUSIONS: Our observations suggest a signaling mechanism linking low ambient P(i) levels to the acute up-regulation of its cotransport with sodium and the depolymerization of the subcortical actin cytoskeleton.     

IQGAP1 promotes neurite outgrowth in a phosphorylation-dependent manner

In eukaryotic cells IQGAP1 binds to and alters the function of several proteins, including actin, E-cadherin, beta-catenin, Cdc42, and Rac1. Yeast IQGAP1 homologues have an important role in cytoskeletal organization, suggesting that modulation of the cytoskeleton is a fundamental role of IQGAP1. Phosphorylation is a common mechanism by which cells regulate protein function. Here we demonstrate that endogenous IQGAP1 is highly phosphorylated in MCF-7 human breast epithelial cells. Moreover, incubation of cells with phorbol 12-myristate 13-acetate (PMA) stimulated phosphate incorporation into IQGAP1. By using mass spectrometry, Ser-1443 was identified as the major site phosphorylated on IQGAP1 in intact cells treated with PMA. Ser-1441 was also phosphorylated but to a lesser extent. In vitro analysis with purified proteins documented that IQGAP1 is a substrate for protein kinase Cepsilon, which catalyzes phosphorylation on Ser-1443. Consistent with these findings, inhibition of cellular protein kinase C via bisindolymaleimide abrogated Ser-1443 phosphorylation in response to PMA. To elucidate the biological sequelae of phosphorylation, Ser-1441 and Ser-1443 were converted either to alanine, to create a nonphosphorylatable construct, or to glutamic acid and aspartic acid, respectively, to generate a phosphomimetic IQGAP1. Although overexpression of wild type IQGAP1 promoted neurite outgrowth in N1E-115 neuroblastoma cells, the nonphosphorylatable IQGAP1 S1441A/S1443A had no effect. In contrast, the S1441E/S1443D mutation markedly enhanced the ability of IQGAP1 to induce neurite outgrowth. Our data disclose that IQGAP1 is phosphorylated at multiple sites in intact cells and that phosphorylation of IQGAP1 will alter its ability to regulate the cytoskeleton of neuronal cells.     

Phosphorylation of ankyrin decreases its affinity for spectrin tetramer

The effects of phosphorylation on the interaction between spectrin and ankyrin were investigated. Spectrin and ankyrin were phosphorylated using purified human erythrocyte membrane and cytosolic (casein kinase A) kinases. These two kinases have similar properties as well as activities toward spectrin and ankyrin. Both kinases catalyzed the incorporation of about 2 mol of phosphate/mol of spectrin and about 7 mol of phosphate/mol of ankyrin. These phosphates were incorporated primarily into seryl and threonyl residues of the proteins. The phosphopeptide maps of ankyrin phosphorylated by the membrane kinase and casein kinase A were identical. Binding studies indicate that ankyrin exhibits different affinities for spectrin dimers (KD = 2.5 +/- 0.9 X 10(-6) M) and tetramers (KD = 2.7 +/- 0.8 X 10(-7) M). These dissociation constants were not appreciably affected by the phosphorylation of spectrin. On the other hand, phosphorylation of ankyrin was found to significantly reduce its affinity for either phosphorylated or unphosphorylated spectrin tetramers (KD = 1.2 +/- 0.1 X 10(-6) M) but not spectrin dimers (KD = 2.5 +/- 0.4 X 10(-6) M). The same results were obtained using either the membrane kinase or casein kinase A as the phosphorylating enzyme. The above observation suggests that ankyrin phosphorylation may provide an important mechanism for the regulation of the erythrocyte membrane cytoskeletal network.     

Tyrosine phosphorylation of villin regulates the organization of the actin cytoskeleton

We have previously shown that tyrosine phosphorylation of the actin-regulatory protein villin is accompanied by the redistribution of phosphorylated villin and a concomitant decrease in the F-actin content of intestinal epithelial cells. The temporal and spatial correlation of these two events suggested that tyrosine phosphorylation of villin may be involved in the rearrangement of the microvillar cytoskeleton. This hypothesis was investigated by analyzing the effects of tyrosine phosphorylation of villin on the kinetics of actin polymerization by reconstituting in vitro the tyrosine phosphorylation of villin and its association with actin. Full-length recombinant human villin was phosphorylated in vitro by expression in the TKX1-competent cells that carry an inducible tyrosine kinase gene. The actin-binding properties of villin were examined using a co-sedimentation assay. Phosphorylation of villin did not change the stoichiometry (1:2) but decreased the binding affinity (4.4 microm for unphosphorylated versus 0.6 microm for phosphorylated) of villin for actin. Using a pyrene-actin-based fluorescence assay, we demonstrated that tyrosine phosphorylation had a negative effect on actin nucleation by villin. In contrast, tyrosine phosphorylation enhanced actin severing by villin. Electron microscopic analysis showed complementary morphological changes. Phosphorylation inhibited the actin bundling and enhanced the actin severing functions of villin. Taken together our data show that tyrosine phosphorylation of villin decreases the amount of villin bound to actin filaments, inhibits the actin-polymerizing properties of villin, and promotes the actin-depolymerizing functions instead. These observations suggest a role for tyrosine phosphorylation in modulating the microvillar cytoskeleton in vivo by villin in response to specific physiological stimuli.     

Multisite phosphorylation of glycogen synthase from rabbit skeletal muscle. Identification of the sites phosphorylated by casein kinase-I

A casein kinase was highly purified from rabbit skeletal muscle whose substrate specificity and enzymatic properties were virtually identical to those of casein kinase-I from rabbit reticulocytes. Prolonged incubation of glycogen synthase with high concentrations of skeletal muscle casein kinase-I and Mg-ATP resulted in the incorporation of greater than 6 mol phosphate/mol subunit and decreased the activity ratio (+/- glucose-6P) from 0.8 to less than 0.02. The sites phosphorylated by casein kinase-I were all located in the N and C-terminal cyanogen bromide peptides, termed CB-1 and CB-2. At an incorporation of 6 mol phosphate/mol subunit, approximately equal to 2 mol/mol was present in CB-1 and approximately equal to 4 mol/mol in CB-2. Within CB-1, casein kinase-I phosphorylated the serines that were 3, 7 and 10 residues from the N-terminus of glycogen synthase, with minor phosphorylation at threonine-5. Within CB-2, approximately equal to 90% of the phosphate incorporated was located between residues 28 and 53, and at least five of the seven serine residues in this region were phosphorylated. The remaining 10% of phosphate incorporated into CB-2 was located between residues 98 and 123, mainly at a serine residue(s). Two of the major sites labelled by casein kinase-I (serine-3 and serine-10 of CB-1) are not phosphorylated by any other protein kinase. This will enable the role of casein kinase-I as a glycogen synthase kinase in vivo to be evaluated.     

Golgi galactosyltransferase contains serine-linked phosphate

In HeLa and HepG2 cells the Golgi complex enzyme galactosyltransferase became phosphorylated following incubation with 32Pi-Analysis on sodium dodecyl sulphate/polyacrylamide gel electrophoresis revealed incorporation of 32P into the mature 54-kDa form. This phosphorylation was independent of protein synthesis. Serine was identified as the sole phosphorylated amino acid; no radioactive phosphate was detected on N-linked oligosaccharide. The phosphate-labelled galactosyltransferase has the same turnover as [35S]methionine-labelled polypeptides (t1/2 = 20 h). Soluble enzyme, released by the cells, contained very little phosphate relative to that which remained cell-associated. Charge heterogeneity arising from phosphorylation contributes in part to the heterodispersed appearance of the enzyme on two-dimensional gels, as the degree of radioactive phosphate differs among the different iso-enzymes.     

Reduced cell adhesion during mitosis by threonine phosphorylation of beta1 integrin

Cell shape and adhesion of cultured mammalian cells change dramatically during mitosis, however, how cell cycle-dependent alterations in cell adhesion are regulated remain to be elucidated. We show here that normal human mammary epithelial (HME) cells which became less adhesive and adopted the rounded morphology during the G(2)/M phase of the cell cycle significantly reduced their dependence on beta1 integrin-mediated adhesion to laminin, by using function blocking antibody to beta1 integrin. In G(2)/M cells, both total and cell surface expressions of beta1 integrin were comparable with those in G(1) cells but it was phosphorylated at threonines 788-789 within its cytoplasmic domain and coimmunoprecipitated Ca(2+)/calmodulin-dependent protein kinase (CaMK) II. The threonine phosphorylated beta1 integrin significantly reduced its intracellular linkage with actin, with no significant reduction in the actin expression. In contrast, beta1 integrin in G(1) cells was not threonine phosphorylated but formed a link with actin and coimmunoprecipitated the core enzyme of the serine/threonine protein phosphatase (PP) 2A. The results suggest that reduced beta1 integrin-mediated cell adhesion of HME cells to the substratum during mitosis may be induced by beta1 integrin phosphorylation at threonines 788-789 and its reduced ability to link with the actin cytoskeleton.     

Sites of in vivo phosphorylation of vesicular stomatitis virus matrix protein.

We mapped the in vivo phosphorylation sites for the matrix (M) protein of the Orsay and San Juan strains of vesicular stomatitis virus, Indiana serotype, using limited proteolysis and phosphoamino acid analysis. M protein was solubilized from 32P-labeled virions by using detergent and high-salt conditions, then treated with either trypsin or Staphylococcus aureus V8 protease, and analyzed by polyacrylamide gel electrophoresis and autoradiography to determine which fragments contained phosphate residues. The M protein fragment extending from amino acid 20 to the carboxy terminus contained approximately 70% of the control 32P label, while the fragment extending from amino acid 35 to the carboxy terminus had only trace amounts of label. These data indicate that the major phosphorylation site was between amino acids 20 and 34 in the Orsay strain M protein. Phosphoamino acid analysis of M protein by thin-layer electrophoresis showed the presence of phosphothreonine and phosphoserine and that phosphothreonine continued to be released after prolonged vapor-phase acid hydrolysis. These data identify Thr-31 as the primary in vivo phosphate acceptor for M protein of the Orsay strain of vesicular stomatitis virus. The San Juan strain M protein has serine at position 32, which may also be an important phosphate acceptor. In addition, phosphorylation at Ser-2, -3, or -17 occurs to a greater extent in the San Juan strain M protein than in the Orsay strain M protein. The subcellular distribution of phosphorylated M protein was investigated to determine a probable intracellular site(s) of phosphorylation. Phosphorylated M protein was associated primarily with cellular membranes, suggesting phosphorylation by a membrane-associated kinase. Virion M protein was phosphorylated to a greater extent than membrane-bound M protein, indicating that M protein phosphorylation occurs at a late stage in virus assembly. Phosphorylation of wild-type and temperature-sensitive mutant M protein was studied in vivo at the nonpermissive temperature. The data show that phosphorylated M protein was detected only in wild-type virus-infected cells and virions, suggesting that association with nucleocapsids may be required for M protein phosphorylation or that misfolding of mutant M protein at the nonpermissive temperature prevents phosphorylation.     

Mechanism of smooth muscle myosin phosphorylation

In vertebrate smooth muscles, phosphorylation of the regulatory light chain appears to be necessary for actin activation of the Mg-ATPase activity and for the in vitro assembly of myosin into filaments. From a correlation between the degree of phosphorylation and enzymatic activity, it was suggested that both myosin heads must be phosphorylated before either head could be activated by actin, and that phosphorylation of filamentous myosin occurred in a negatively cooperative manner (Persechini, A., and Hartshorne, D. J. (1981) Science 213, 1383-1385; Ikebe, M., Ogihara, S., and Tonomura, Y. (1982) J. Biochem. (Tokyo) 91, 1809-1812; Sellers, J. R., Chock, P. B., and Adelstein, R. S. (1983) J. Biol. Chem. 258, 14181-14188). Here we have determined the mechanism of phosphorylation by separating dephosphorylated and phosphorylated myosin species based on their different structural properties in the minifilament buffer system (5 mM citrate, 22 mM Tris). Fully phosphorylated myosin remained assembled as minifilaments in 1 mM Mg-ATP, but dephosphorylated myosin dissociated to a mixture of folded monomers and dimers. Gel filtration was used to separate these two structures. At intermediate levels of phosphorylation, the relative amount of myosin that formed minifilament and dimer and the degree of phosphorylation of the separated species relative to the initial level of phosphorylation was measured. From these data, it was possible to deduce that singly and doubly phosphorylated myosin remained assembled in the presence of nucleotide. Myosin molecules with 0, 1, or 2 heads phosphorylated could also be separated by nondenaturing gel electrophoresis. The amount of myosin which formed each species was quantitated as a function of phosphorylation. Results from the combined approaches are consistent with a model in which light chain kinase randomly phosphorylates myosin, independent of the state of aggregation of the myosin.