Search

Directed movements on actin filaments within the cell are powered by molecular motors of the myosin superfamily. On actin filaments, myosin motors convert the energy from ATP into force and movement. Myosin motors power such diverse cellular functions as cytokinesis, membrane trafficking, organelle movements, and cellular migration. Myosin generates force and movement via a number of structural changes associated with hydrolysis of ATP, binding to actin, and release of the ATP hydrolysis products while bound to actin. Herein we provide an overview of those structural changes and how they relate to the actin-myosin ATPase cycle. These structural changes are the basis of chemo-mechanical transduction by myosin motors.

After several decades studying different acto-myosin complexes at lower and intermediate resolution - limited by the electron microscope instrumentation available then - recent advances in imaging technology have been crucial for obtaining a number of excellent high-resolution 3D reconstructions from cryo electron microscopy. The resolution level reached now is about 3-4 Å, which allows unambiguous model building of filamentous actin on its own as well as that of actin filaments decorated with strongly bound myosin variants. The interface between actin and the myosin motor domain can now be described in detail, and the function of parts of the interface (such as, e.g., the cardiomyopathy loop) can be understood in a mechanistical way. Most recently, reconstructions of actin filaments decorated with different myosins, which show a strongly bound acto-myosin complex also in the presence of the nucleotide ADP, have become available. The comparison of these structures with the nucleotide-free Rigor state provide the first mechanistic description of force sensing. An open question is still the initial interaction of the motor domain of myosin with the actin filament. Such weakly interacting states have so far not been the subject of microscopical studies, even though high-resolution structures would be needed to shed light on the initial steps of phosphate release and power stroke initiation.

Several small molecule effectors of myosin function that target the motor domains of myosin classes I, II, V, and VI have been identified. Four distinct binding sites in the myosin motor domain have been reported with unique properties and mechanisms of action. This chapter describes the structural basis and activities of known small molecule effectors that allosterically target the myosin motor domain.

Contraction in all muscles must be precisely regulated and requisite control systems must be able to adjust to changes in physiological and myopathic stimuli. In this chapter, we outline the structural evidence for a steric mechanism that governs muscle activity. The mechanism involves calcium and myosin induced changes in the position of tropomyosin along actin-based thin filaments. This process either blocks or uncovers myosin crossbridge binding sites on actin and consequently regulates crossbridge cycling on thin filaments, the sliding of thin and thick filaments and muscle shortening and force production.

Vertebrate nonmuscle cells express multiple tropomyosin isoforms that are sorted to subcellular compartments that have distinct morphological and dynamic properties. The creation of these compartments has a role in controlling cell morphology, cell migration and polarization of cellular components. There is increasing evidence that nonmuscle myosins are regulated by tropomyosin in these compartments via the regulation of actin attachment, ATPase kinetics, or by stabilization of cytoskeletal tracks for myosin-based transport. In this chapter, I review the literature describing the regulation of various myosins by tropomyosins and consider the mechanisms for this regulation.

It has been a great honor and a particular pleasure to participate in this meeting to celebrate the fortieth anniversary of the discovery of troponin by Professor Ebashi, whom I have been privileged to know for many years of my scientific life. I thought therefore it would be appropriate to described briefly what was happening in studies of another aspect of muscle contraction over somewhat the same time period.

Our studies demonstrate that harmonin and myosin VIIa are not localized in the same compartments in the mouse and human retinas, indicating that they do not interact in this organ, contrary to what has been shown in the inner ear. The enrichment of harmonin in the photoreceptor synapses indicates that this protein may form multiple complexes with others to maintain the synaptic structure or to mediate in the release of synaptic vesicles.

Muscles can be studied as complex systems of many interacting proteins and investigated at many different levels of organization. This talk will describe how we modeled the mechanism of Ca activation, the structure of the muscle proteins, and protein complexes (including actin monomers, tropomyosin and troponin complexes, and myosin) to examine two different scientific problems: the mechano-chemical energy transduction mechanism, and the control system of that mechanism. The methods we used--saturation transfer electron paramagnetic resonance, phosphorescence anisotropy, and fluorescence resonance energy transfer--reveal two specific structures: a hinge between the motor and regulatory domains, and a stiff regulatory domain. This indicates that the structure of the myosin head is capable of generating translating conformational changes within the motor domain to the swing of the regulatory domain, and that the regulatory domain is rigid enough to act as a lever arm.

By applying AFM technology, we studied mechanical characteristics of myofibrils of skeletal muscle. The obtained results indicate that (1) the Z-band is the most rigid sarcomere component stabilizing the myofibril structures, (2) various filamentous components are inter-connected in sarcomere with sufficient mechanical strength to support the contractile force, and (3) the molecular structure of the overlap region between actin and myosin filaments is anisotropic. In any case the present studies clearly indicate that the AFM technique is a powerful tool to investigate the mechanical characteristics of sarcomere structure of muscle fiber.

Using electron spin resonance, we have studied dynamic structures of myosin neck domain and troponin C by site-directed spin labeling. We observed two broad but distinct orientations of a spin label attached specifically to a single cysteine (cys156) on the regulatoy light chain (RLC) of myosin in relaxed skeletal muscle fibers. The two probe orientations, separated by a 25 degrees axial rotation, did not change upon muscle activation, but orientational distributions became narrower substantially, indicating that a fraction of myosin heads undergoes a disorder-to-order transition of the myosin light chain domain upon force generation and muscle contraction. These results provide insight into the mechanism how myosin heads move their domains to translocate an actin filament. Site-directed spin-labeling was achieved by cysteine residues of human cardiac troponin C (TnC). Spin dipole-dipole interaction showed that free TnC undergoes a global structural change (extended-to-compact) by Ca2+ or Mg2+. The spectra from the spin labels at N-terminal half domain were broad and almost identical in parallel and perpendicular orientations of fiber, suggesting that the N-terminal of TnC molecule is flexible or disoriented with respect to the filament axis. We also succeeded, for the first time, in fixing the newly-synthesized bifunctional spin label rigidly on TnC molecule in solution (either in +/- Ca2+), giving a promise that we can determine the precise coordinate of the spin principal axis on protein surface.

Actin filament velocities in an in vitro motility assay system were measured both in heavy water (deuterium oxide, D2O) and water (H2O) to examine the effect of D2O on the actomyosin interaction. The dependence of the sliding velocity on pD of the D2O assay solution showed a broad pD optimum of around pD 8.5 which resembled the broad pH optimum (pH 8.5) of the H2O assay solution, but the maximum velocity (4.1 +/- 0.5 microm/sec, n=11) at pD 8.5 in D2O was about 60% of that (7.1 +/- 1.1 microm/sec, n=11) at pH 8.5 in H2O. The Km values of 95 and 80 microM and Vmax values of 3.2 and 5.1 microm/sec for the D2O and H2O assay were obtained by fitting the ATP concentration dependence of the velocity (at pD and pH 7.5) to the Michaelis-Menten equation. The Km value of actin-activated Mg-ATPase activity of myosin subfragment 1(S1) was decreased from 50 microM[actin] in H2O to 33 microM[actin] in D2O without any significant changes in Vmax (9.4 s(-1) in D2O and 9.3 s(-1) in H2O). The rate constants of ADP release from the acto-S1-ADP complex measured by the stopped flow method were 361 +/- 26 s(-1) (n=27) in D2O and 512 +/- 39 s(-1) (n=27) in H2O at 6 degrees C. These results suggest that the decrease in the in vitro actin-myosin sliding velocity in D2O results from a slowing of the release of ADP from the actomyosin-ADP complex and the increase in the affinity of actin for myosin in the presence of ATP in D2O.

Electromechanical coupling by KCl depolarization of bladder preparations elicits an initial phasic and subsequent tonic contraction. Using a smooth-muscle myosin heavy chain (SM-MyHC) knock-out mouse model we could previously demonstrate, that phasic and tonic contraction of intact neonatal bladder preparations could be elicited through the recruitment of SM-MyHC and non-muscle myosin heavy chains (NM-MyHC), respectively. Inhibition of myosin light chain kinase (MLCK) by ML-7 eliminated the phasic contraction of wild-type (+/+), rather than tonic contraction of neonatal bladder strips prepared from both +/+ and homozygous SM-MyHC knock-out (-/-) mice. Pharmacomechanical coupling upon PDBu-induced activation of protein kinase C of neonatal bladder preparations elicited tonic contraction of both +/+ and -/- murine. We suggest that: i) electromechanical coupling activates both SM-MyHC and NM-MyHC systems via a ML-7 sensitive and insensitive pathway, respectively. ii) Pharmacomechanical coupling recruits part of the NM-MyHC system rather than SM-MyHC.

We developed an electromagnet to perform quick changes in load in the motility system consisting of myosin molecules attached to a magnetizable bead and actin filaments The electromagnet was combined with an inverted microscope and load could be quickly changed under optical observation. The magnetic field was generated by high electric current (6V, 0-125A) and the maximum field was 8,000 Oe. The maximum force exerted on a bead was 80pN at 2.5mm distance from a magnet. The change in force was 0.48% at the distance of 5.0mm from the magnet when a bead moved longitudinally for 30microm. The time to change load was about 20ms. The movements of a bead in water were recorded by video when step changes in magnetic field were applied and it was shown that a bead exactly followed the change in force. This apparatus is very much useful to analyze the transient changes in the movement of a bead, if the movement is relatively slow as in the interaction between actin and myosin from molluscan smooth muscle.

Analysis of the myosin-based meridional intensity data in the X-ray diffraction patterns of live frog skeletal muscles was performed to propose a more precise model for a myosin crown periodicity and an axial disposition of two-headed crossbridges along the thick filament in a sarcomere. Modeling studies revealed that the thick filament has a mixed structure of two different periodicities of the myosin crossbridge crown arrangement and that the crown periodicity and the axial disposition of crossbridges are altered when muscle goes from the relaxed state to the contracting state. Factors that primarily affect the meridional intensities were examined.

The atomic force microscope (AFM) is a powerful tool for imaging biological molecules on a substrate, in solution. However, there is no effective time axis with AFM; commercially available AFMs require minutes to capture an image, but many interesting biological processes occur at much higher rate. Hence, what we can observe using the AFM is limited to stationary molecules, or those moving very slowly. We sought to increase markedly the scan speed of the AFM, so that in the future it can be used to study the dynamic behaviour of biomolecules. For this purpose, we have developed various devices optimised for high-speed scanning. Combining these devices has produced an AFM that can capture a 100 x 100 pixel image within 80 ms, thus generating a movie consisting of many successive images of a sample in aqueous solution. This is demonstrated by imaging myosin V molecules moving on mica, in solution.

We have developed a new microscopic technique to measure the force generated on a single actin filament (FA) in the A-band in which the intact lattice structure composed of myosin thick filaments is maintained; we call this newly developed system "Bio-nanomuscle (or an A-band motility assay system)". The A-bands were prepared by selective removal of thin filaments from rabbit skeletal glycerinated myofibrils under optical microscope with the use of gelsolin (a severing and barbed (B)-end capping protein of FA) that was prepared from bovine serum. A polystyrene bead of 1 microm in diameter attached to the B-end of FA (through a gelsolin molecule attached to the surface of the bead) was trapped and manipulated with optical tweezers. The displacement of the bead up to 200 nm (corresponding to the force of approximately 40 pN) was determined by phase-contrast image analysis. At the initial stage of this study, the overlapping length of an FA with the A-band was determined from the fluorescence image of FA labeled with rhodamine-phalloidin (Rh-Ph) and the phase-contrast image, but we later improved the method of determination by moving the sample stage stepwise using the piezo actuator. The average force per overlap was subsequently estimated and the histogram was fitted with two Gaussian distributions. Each peak is supposed to correspond to the force developed by FA interacting outside or inside the A-band, and the peak value of the latter was estimated to be 140 pN/microm. From this value, the average force developed per each cross-bridge (CB; a two-headed myosin molecule) was determined to be 1.3 pN.

Molecular assemblies of actin and myosin for the contractility of smooth muscle are quite different from those of striated muscle. Another striking difference is that vascular smooth muscle has a potential to transform to migratory synthetic cell type. At this point of view, smooth muscle cell has properties similar to those of non-muscle. In fact, myosin Ic, a single headed unconventional myosin, was identified in aorta smooth muscle. During the studies on myosin Ic, we have found another calmodulin related 190kDa protein. This protein binds to calmodulin irrespective on calcium ion and to F-actin in an ATP independent manner. Furthermore, the F-actin binding stoichiometry diminished to half upon the addition of exogenous calmodulin. Partial amino acid sequence indicated a high homology to those of GRD (GTPase Related Domain) of human brain IQGAP1. Western blot analysis using anti-human IQGAP1 antibody also indicated a strong cross-reactivity with the protein. We have tested the protein with respect to the characteristic F-actin gelation by IQGAP1. In the presence of cdc42 and GTPgammaS, 190kDa protein could cause a high viscosity of F-actin. These data indicate a close similarity to human brain IQGAP1. The presence of IQGAP1 in aorta smooth muscle suggests contributions for cellular processes such as actin reorganization during contraction-relaxation cycle, association of cytoskeletal structure to cell membrane, organelle movement.

Unlabelled: We have tested the hypothesis that the transition rate (G) of the cardiac XB from the strong force generating state to the weak state is a linear function V of the sarcomere (VSL); furthermore, we tested whether the ATPase rate of the two isoforms of myosin can be held responsible for the difference between V0 of rat cardiac trabeculae containing V1 isomyosin versus those containing V3 isomyosin., Methods: V1 isomyosin was induced by thyroid hormone treatment of the rats for 2 weeks, V3 isomyosin by PTU treatment for 1 month. Force was measured with a strain gauge in trabeculae from the rat right ventricle in K-H solution ([Ca]o=1.5 mM, 25 degrees C). Sarcomere length (SL) was measured with laser diffraction techniques. Twitch force at constant SL, and the force response to shortening at constant VSL (0-8 microm/s; deltaSL 50-100 nm) were measured at varied time during the twitch., Results: The force response to shortening consisted of a fast initial exponential decline (tau = 2 ms) followed by a slow decrease of F. The instantaneous difference (deltaF) between isometric force (FM) and the declining force depended on shortening duration (deltat), VSL and instantaneous FM: deltaF = G1 x FM x deltat x VSL x (1-VSL/VMAX), where VMAX is the unloaded VSL and G1 was 6.15 +/- 2.12 microm(-1) (mean +/- s.d.; n=6). deltaF/FM was independent of the time onset of shortening. G1 of V1 and V3 trabeculae did not differ. V0 of V1 and V3 trabeculae differed 2-2.5 fold, as did both the ATPase rate and the velocity of actin sliding in a motility assay of the myosin purified from V1 or V3 hearts. The temperature dependence of the ATPase rate (Q10: 4.03 and 4.33, respectively; n.s.) was similar to that of V0 that has previously been reported for predominantly V1 trabeculae. Cross-linking of actin to myosin with the short chain cross linker EDC increased the ATPase rate of the two isomyosins (200-fold and 600-fold respectively) to exactly the same final level and reduced their Q10 by 50%., Conclusion: The linear interrelation between deltaF and VSL is consistent with feedback, whereby XB kinetics depends on VSL. This feedback provides an integrated description of cardiac muscle mechanics and energetics. The results, also, suggests that it is unlikely that the hydrolytic domain of the cross bridge determines V0 and warrant ongoing experiments to investigate the role of the actin binding domain of the XB in cardiac sarcomere kinetics. In order to further investigate the role of the actin binding domain, we have expressed chimeric cardiac myosin, co-assembled with MLC, by mutual substitution of actin binding loop on alpha MHC and beta MHC.

Both the MyBP-C motif between C1 and C2 and the C5 module are important regions for implementing the effect of MyBP-C on myosin and on contractility but in different ways. C5 may determine the folding of MyBP-C and the manner in which MyBP-C interacts with myosin. In spite of its apparent importance this interaction does not appear to be physiologically regulated. Its alteration by mutation however can have a major effect on contractility. On the other hand, the effect of the motif is regulated by phosphorylation and appears to be an important part of a physiological mechanism(s) for modulating contractility. Thick filaments isolated from cardiac muscle exist in one of three different structures (Levine et al 2001). Different degrees of phosphorylation of MyBP-C can produce transitions among the three structures. The combination of the binding results of Flashman et al (2002) with the data of McClellan et al (2001) suggests that the C5 interaction with C8 is critical in maintaining the normal structure of thick filaments and the normal function of the force generators in the filaments. The cardiac-specific sequence in C5 and its normal interaction with another part of the same or a different MyBP-C may be required for the correct 3 dimensional shape of the three MyBP-C molecules at each locus in the C zone and the normal structure of the thick filament. The normal interactions may be necessary to allow transitions in binding and filament structure that are associated with phosphorylation of the MyBP-C motif.

When ATP binds to the active site of myosin heads, Switch II undergoes a large conformational change and the cleft surrounding the bound gamma-phosphate closes. In the closed state, Glu470 in Switch II comes together with Arg247 in Switch I to form a salt-bridge. Here, the functional significance of the two bridging residues was tested by using site-directed mutagenesis. We conclude from such tests that (a) the attractive force between Arg247 and the gamma-phosphate of ATP moves the cleft to close, and (b) during hydrolysis, Glu470 is intimately involved in positioning the lytic water for the attack on the gamma-phosphorus. We also speculate on how the salt-bridge between Arg247 and Glu470 is related to hydrolysis.

The effects of length changes applied to resting, contracting and rigor muscles on the reflection spacings of the X-ray diffraction patterns were summarized. The spacing changes of the actin- and myosin-based meridional reflections as a function of tension relative to an isometric tension of active muscle (P0) were linear and almost identical in the active and rigor states, showing that the extension of both filaments is Hookenian and does not depend upon the states of muscle. In addition to their length changes caused by tension generation, there are small but significant length changes of both filaments due purely to activation of muscle. The actin and myosin filaments are elongated by approximately 0.36% and approximately 0.43%, respectively under the maximum active tension. The results indicate that a large part of the sarcomere compliance of an active muscle is caused by the extensibility of the myofilaments. Inspection of the behavior of the meridional and layer-line reflection spacings reveals that there is a close relationship between the extensibility and helical twisting of the actin filaments under active and passive forces. The extension caused by tension is associated with an unwinding of right-handed helices following the actin monomers in the filament. At the pointed end of the filament could rotate anticlockwise through one fifth the complete turn during contraction.

It has been shown that polyclonal antibody directed against myosin subfragment 2 (S-2) eliminates Ca(2+)-activated isometric force development in glycerinated skeletal muscle fibers while Mg-ATPase activity of the fibers remains unchanged. To further explore possible involvement of myosin S-2 in muscle contraction, we studied the effect of anti-S-2 antibody on rigor linkage formation in the fibers, and found that the antibody inhibited development of rigor force in a dose- and time-dependent manner without changing the relation between muscle fiber stiffness and force. If, however, the antibody was applied after development of rigor force, it had no effect on both the stiffness and force. These results strongly suggest bidirectional functional communication between myosin subfragments 1 and 2 in the fibers.

Scallop subfragment 1 (S1) is an unregulated molecule; it differs from heavy meromyosin (HMM) and myosin in that it has no "off" state, although it contains the full complement of light chains and the triggering calcium binding site. S1 differs from myosin by lacking the head-rod junction and being single-headed. The contribution of the head-rod junction was evaluated by studying single-headed myosin. Isolated single-headed myosins show some regulation; their actin activated ATPase is stimulated about 3-fold by calcium. However, in contrast to HMM and myosin, the calcium dependence of ATPase activation of single-headed myosin is non-cooperative. The single ATP turnover rate of single-headed myosin in the absence of calcium is less than 30 seconds (our experimental resolution) compared to the approximately 5 minute turnover rate of myosin. HMM and myosin exhibit several cooperative features not shown by S1. Calcium binding becomes cooperative in the presence of nucleotide analogues in HMM and myosin, but not in S1. Nucleotide analogues are bound cooperatively by myosin and HMM in the absence of calcium; the introduction of calcium to the system reduces the affinity and abolishes the cooperative binding of nucleotide in the double headed molecules. Conversely, S1 shows normal binding curves for nucleotide analogues both in the presence and absence of calcium. Therefore, there is direct communication between the calcium binding sites and nucleotide binding sites in regulated molecules that is mediated by interaction between the two heads. .

The positions and orientations of the myosin heads in relaxed, active, rigor and S1-labelled fish muscle are being determined by analysis both of electron micrographs and of low-angle X-ray diffraction patterns. The X-ray analysis of resting muscle makes use of the head shape defined from the study of S1 crystals, with variable head configurational parameters being used on each of the three different 3-fold symmetric 14.3 nm-spaced 'crowns' of myosin heads within the 42.9 nm axial repeat of the myosin filaments. Diffraction patterns were stripped using CCP13 fibre diffraction software. Searches and optimisation were carried out using simulated annealing and local refinement procedures to give a 'best fit' relaxed structure with a crystallographic R-factor of about 4%. It had heads oriented all the same way up (i.e. with similar rotations around their own long axes) on the myosin filament, but with a small range of axial tilts. Head configuration in rigor fish muscle is being determined by X-ray diffraction and electron microscopy of normal rigor muscle and of skinned muscle soaked with extrinsic myosin S1. Computed 3-D reconstructions of acto-S1 using X-ray amplitudes and phases from electron microscopy are informative and help to analyse the X-ray diffraction data that extend axially to about 1 nm resolution. An ambiguity is the axial direction of the observed resting myosin head array relative to the known polarity of the actin filaments. One polarity would give little axial displacement (2-3 nm) between opposite ends of the resting and rigor heads, and in this case the heads would need to rotate around their own long axes by about 115 degrees to make a rigor attachment. The other (preferred) filament polarity would provide considerable axial swinging (14-15 nm) between the two states. We are attempting to define the absolute polarity of the resting muscle myosin head array using electron microscopy and image processing either of cryo-sections or of replicas from shadowed, freeze-fractured, rapidly frozen fish muscle fibres.

Time resolved X-ray diffraction experiments in single muscle fibres of the frog at 2.15 microns sarcomere length and 4 degrees C were performed at ID2 (SAXS), ESRF, Grenoble (France) to investigate the structural aspects of cross-bridge action during the development of the isometric tetanic tension (T0). Changes in the low angle myosin-based reflections were measured with 5 ms time resolution by signal averaging data collected with a 10 m camera length and a 2D gas-filled detector. Upon activation the intensity of the first order myosin layer line reflection, I(M1), and the intensity of the second order meridional reflection, I(M2), reduced practically to zero with a half-time which leads the tension rise by 15-20 ms. The complex changes of the intensity of the third order myosin meridional reflection, I(M3), and the increase of its axial spacing from 14.34 nm (at rest) to 14.57 nm (at T0) could be analysed by assuming that they were the result of the combination of the time dependent modulation in intensity of two closely spaced periodicities, one at 14.34 nm, characteristics of the myosin molecule at rest and the other at 14.57 nm, assumed by the myosin as a consequence of the activation and force production. I(14.34) drops monotonically in advance to isometric tension development with a half-time similar to that of I(M1) and I(M2), while I(14.57) rises from zero to a maximum in parallel with tension.

Using a gas environmental (hydration) chamber, in which biological specimens can be kept in wet state, we succeeded in recording images of 'living' muscle thick filaments with gold position markers attached to the myosin heads. The position of individual myosin heads did not change appreciably with time in the absence of ATP, indicating stability of the myosin head mean position. On application of ATP, the position of individual myosin heads was found to move by approximately 20 nm along the filament axis, while no appreciable movement of the filaments was detected. The ATP-induced myosin head movement was not observed in filaments in which ATPase activity of the myosin heads was eliminated. Application of ADP produced no appreciable myosin head movement. These results show that the ATP-induced myosin head movement takes place in the absence of the thin filaments. Since ATP reacts rapidly with the myosin head (M) to form the complex (M.ADP.Pi) having average lifetime of > 10 s, the observed myosin head movement may be mostly associated with reaction, M + ATP-->M.ADP.Pi. This work will open a new research field to study dynamic structural changes of individual biomolecules which are kept in 'living' state in an electron microscope.

To deepen our understanding of the regulatory mechanism of smooth muscle contraction, we examined properties of smooth muscle myosin phosphatase (SMMP) which was purified from chicken gizzard according to the method of Alessi et al. with slight modifications. The SMMP was a heterotrimer of 130, 37 and 20 kDa subunits as reported. Because the enzymatic activity was strongly dependent on the ionic strength, all experiments were carried out at a constant ionic strength of 0.15 M. The Vmax and K(m) of the enzyme toward bovine stomach phosphorylated myosin was 25.2 mumol/mg protein/min and 0.45 microM at 25 degrees C, respectively, which implied that the rate of dephosphorylation by SMMP in smooth muscle cells should be comparable to that of phosphorylation by the myosin light chain kinase in the presence of a saturating Ca2+ concentration. ATP dose-dependently decreased the SMMP activity to one-fifth. The IC50 for ATP was 0.6 microM in the presence of 5 mM MgCl2, whereas it increased to 10 microM in the absence of Mg2+ with the addition of 1 mM EDTA. ADP, AMP, GTP and ITP showed no or only weak effect. The finding that ATP gamma S was ineffective indicates that phosphorylation of the 130 kDa subunit cannot be the underlying mechanism. This modulation was observed with myosin from bovine stomach, but not from chicken gizzard. The difference in conformation of myosin molecules in the presence of MgATP may be critical in its modulatory effect.