Segmental flexibility and head-head interaction in scallop myosin. A study using saturation transfer electron paramagnetic resonance spectroscopy

Saturation transfer electron paramagnetic resonance spectroscopy was used to investigate the rotational motion of the head domains of native and desensitized scallop myosin and its proteolytic subfragments. Scallop myosin was spin-labelled with 4-(2-iodoacetamido)-2,2,6,6-tetramethylpiperidinooxyl, which reacted with a heavy chain residue in the subfragment 1 domain. As previously shown for rabbit skeletal muscle myosin (Thomas et al., 1975), the two head domains of native scallop myosin appear to have independent motion (rotational correlation time, pi, = 0.8 X 10(-7) s for subfragment 1; 1.4 X 10(-7) s for myosin). However, removal of a regulatory light chain, to effect desensitization of the actin-activated ATPase, was associated with an increase in pi for myosin to a value of 2.4 X 10(-6) s. The Ca2+ sensitivity and initial correlation time were restored on recombination of the regulatory light chain in the presence of Mg2+. Sedimentation velocity profiles in an analytical ultracentrifuge indicated that the desensitized myosin preparations were largely monomeric and therefore the change in pi appears to reflect an intramolecular event. Addition of EDTA to spin-labelled scallop heavy meromyosin caused an immediate 2.5 to 4-fold increase in pi and a partial desensitization of the ATPase activity. Comparable experiments with subfragment 1 yielded a barely detectable increase in pi (1.5-fold) in the first ten minutes. The restricted rotational motion observed in desensitized myosin and heavy meromyosin could arise by a conformational change in the subfragment 1-subfragment 2 hinge region or by an association of one head with its partner. The latter mechanism, involving the exposed light chain binding site, would also explain the preferential release of one regulatory light chain from scallop myosin, and might account for some other co-operative effects observed in this molecule (Bagshaw, 1980).     

Regulatory light chains in myosins.

Scallop myosin molecules contain two moles of regulatory light chains and two moles of light chains with unknown function. Removal of one of the regulatory light chains by treatment with EDTA is accompanied by the complete loss of the calcium dependence of the actin-activated ATPase activity and by the loss of one of the two calcium binding sites on the intact molecule. Such desensitized preparations recombine with one mole of regulatory light chain and regain calcium regulation and calcium binding. The second regulatory light chain may be selectively obtained from EDTA-treated scallop muscles by treatment with the Ellman reagent (5,5′-dithiobis(2-nitrobenzoic acid)): treatment with this reagent, however, leads to an irreversible loss of ATPase activity. The light chains obtained by treatment with EDTA and then DTNB are identical in composition and function. A different light chain fraction obtained by subsequent treatment with guanidine-HCl does not bind to desensitized or intact myoflbrils and has no effect on ATPase activity.Regulatory light chains which bind to desensitized scallop myofibrils with high affinity and restore calcium control were found in a number of molluscan and vertebrate myosins, including Mercenaria, Spisula, squid, lobster tail, beef heart, chicken gizzard, frog and rabbit. Although these myosins all have a similar subunit structure and contain about two moles of regulatory light chain, only scallop myosin or myofibrils can be desensitized by treatment with EDTA.There appear to be two classes of regulatory light chains. The regulatory light chains of molluscs and of vertebrate smooth muscles restore full calcium binding and also resensitize purified scallop myosin. The regulatory light chains from vertebrate striated, cardiac, and the fast decapod muscles, on the other hand, have no effect on calcium binding and do not resensitize purified scallop myosin unless the myosin is complexed with actin. The latter class of light chains is found in muscles where in vitro functional tests failed to detect myosin-linked regulation.     

Effects of removal and reconstitution of myosin regulatory light chain and troponin C on the Ca(2+)-sensitive ATPase activity of myofibrils from scallop striated muscle

In order to examine the involvement of troponin-linked Ca(2+)-regulation, in addition to well-known myosin-linked Ca(2+)-regulation, in the contraction of molluscan striated muscle, myofibrils from Ezo-giant scallop striated muscle were desensitized to Ca(2+) by removing both myosin regulatory light chain and troponin C by treatment with a strong divalent cation chelator, CDTA. The ATPase level in the desensitized myofibrils was about half the maximum level in intact myofibrils regardless of the Ca(2+)-concentration at 25 and 15 degrees C. In the absence of Ca(2+), the ATPase of the desensitized myofibrils was suppressed by myosin regulatory light chain but not affected by troponin C at either temperature. The ATPase was activated at higher Ca(2+)-concentrations by both myosin regulatory light chain and troponin C, but the activating effects of these two proteins were affected differently by temperature. The activation of ATPase by myosin regulatory light chain was much greater than that by troponin C at 25 degrees C, whereas the activation by troponin C was much greater than that by myosin regulatory light chain at 15 degrees C. The maximum activation was only obtained in the presence of both myosin regulatory light chain and troponin C at these temperatures. These findings strongly suggest that the contraction of scallop striated muscle is regulated through both myosin-linked and troponin-linked Ca(2+)-regulation, and that the troponin-linked Ca(2+)-regulation is more significant at lower temperature.     

扇贝肌钙蛋白及其对肌球蛋白超沉淀的影响

分离了扇贝闭壳肌肌钙蛋白,其分子量为４６（ＩｎＩ）,４０（ＴｎＴ）,和２２（ＴｎＣ）ｋＤ．肌球蛋白Ｂ含有主要的收缩蛋白质与调节蛋白质,在有Ｃａ２＋和ＡＴＰ存在时,它会发生超沉淀作用．经低离子强度溶液反复沉淀处理,即失去Ｃａ２＋－敏感性,成为去敏肌球蛋白Ｂ．在Ｃａ２＋和ＡＴＰ作用下,它仍可发生超沉淀作用,但仅及最大活性的５０％．若加入肌钙蛋白,则反应活性可完全恢复．兔骨骼肌肌钙蛋白可替代扇贝闭壳肌肌钙蛋白．这表明扇贝闭壳肌兼有肌动蛋白相关调节和肌球蛋白相关调节．     

Negative staining of myosin molecules

A reproducible method has been developed for the negative staining of myosin molecules. The dimensions of stained molecules are in close agreement with those obtained by metal shadowing. Sharp bends in the tail, indicative of hinge regions, were observed at two positions 44 nm and 76 nm from the head-tail junction. The tail was often ill-defined at the position of the first (44 nm) bend. The bend positions may be sites of proteolytic cleavage that result in the production of long and short myosin subfragment S2. About half the molecules exhibited bending to various degrees at one or both of these positions, but cases where the tail folded back on itself in a 180 degrees bend were comparatively rare (approximately equal to 10%). However, in the absence of EGTA, a large fraction of the molecules (approximately equal to 80%) exhibited 180 degrees bends. A small region, approximately 20 nm long, at the tip of the tail often appears to be significantly different from the rest. The heads are about 19 nm long and roughly pear-shaped. Although sometimes straight, more often they show a pronounced curvature. Both senses of curvature were observed, but those curved in a clockwise manner were the most common, indicating preferential binding of one side of the head to the carbon substrate. An analysis of the different combinations of head shapes in individual molecules indicates that each head can rotate independently around its long axis. No preferred angle of orientation between the two heads in a molecule, or between either head and the tail could be found. Substructure has been observed within the heads.     

Structure of the cross-striated adductor muscle of the scallop

The structure of the cross-striated adductor muscle of the scallop has been studied by electron microscopy and X-ray diffraction using living relaxed, glycerol-extracted (rigor), fixed and dried muscles. The thick filaments are arranged in a hexagonal lattice whose size varies with sarcomere length so as to maintain a constant lattice volume. In the overlap region there are approximately 12 thin filaments about each thick filament and these are arranged in a partially disordered lattice similar to that found in other invertebrate muscles, giving a thin-to-thick filament ratio in this region of 6:1.The thin filaments, which contain actin and tropomyosin, are about 1 μm long and the actin subunits are arranged on a helix of pitch 2 × 38.5 nm. The thick filaments, which contain myosin and paramyosin, are about 1.76 μm long and have a backbone diameter of about 21 nm. We propose that these filaments have a core of paramyosin about 6 nm in diameter, around which the myosin molecules pack. In living relaxed muscle, the projecting myosin heads are symmetrically arranged. The data are consistent with a six-stranded helix, each strand having a pitch of 290 nm. The projections along the strands each correspond to the heads of one or two myosin molecules and occur at alternating intervals of 13 and 16 nm. In rigor muscle these projections move away from the backbone and attach to the thin filaments.In both living and dried muscle, alternate planes of thick filaments are staggered longitudinally relative to each other by about 7.2 nm. This gives rise to a body-centred orthorhombic lattice with a unit cell twice the volume of the basic filament lattice.     

The "roll and lock" mechanism of force generation in muscle

Muscle force results from the interaction of the globular heads of myosin-II with actin filaments. We studied the structure-function relationship in the myosin motor in contracting muscle fibers by using temperature jumps or length steps combined with time-resolved, low-angle X-ray diffraction. Both perturbations induced simultaneous changes in the active muscle force and in the extent of labeling of the actin helix by stereo-specifically bound myosin heads at a constant total number of attached heads. The generally accepted hypothesis assumes that muscle force is generated solely by tilting of the lever arm, or the light chain domain of the myosin head, about its catalytic domain firmly bound to actin. Data obtained suggest an additional force-generating step: the "roll and lock" transition of catalytic domains of non-stereo-specifically attached heads to a stereo-specifically bound state. A model based on this scheme is described to quantitatively explain the data.     

3D Structure of fish muscle myosin filaments

Myosin filaments isolated from goldfish (Carassius auratus) muscle under relaxing conditions and viewed in negative stain by electron microscopy have been subjected to 3D helical reconstruction to provide details of the myosin head arrangement in relaxed muscle. Previous X-ray diffraction studies of fish muscle (plaice) myosin filaments have suggested that the heads project a long way from the filament surface rather than lying down flat and that heads in a single myosin molecule tend to interact with each other rather than with heads from adjacent molecules. Evidence has also been presented that the head tilt is away from the M-band. Here we seek to confirm these conclusions using a totally independent method. By using 3D helical reconstruction of isolated myosin filaments the known perturbation of the head array in vertebrate muscles was inevitably averaged out. The 3D reconstruction was therefore compared with the X-ray model after it too had been helically averaged. The resulting images showed the same characteristic features: heads projecting out from the filament backbone to high radius and the motor domains at higher radius and further away from the M-band than the light-chain-binding neck domains (lever arms) of the heads.     

A folded (10 S) conformer of myosin from a striated muscle and its implications for regulation of ATPase activity.

Myosin from the striated adductor muscle of the scallop Pecten maximus is shown to fold into a compact 10 S conformer under relaxing conditions, as has been characterized for smooth and non-muscle myosins. The folding transition is accompanied by the trapping of nucleotide at the active site to give a species with a half-life of about an hour at 20 degrees C. Ca2+ binding to the specific, regulatory sites on a myosin head promotes unfolding to the extended 6 S conformer and activates product release by 60-fold. The unfolding transition, however, remains much slower than the contraction-relaxation cycle of scallop striated muscle and could not play a role in the regulation of these events. The dissociation of products from myosin heads in native thick filaments is Ca2(+)-regulated, but under relaxing conditions the nucleotide is released at least an order of magnitude faster than from the 10 S monomeric myosin, at a rate similar to that observed with heavy meromyosin. Thus, there is no evidence for any intermolecular interaction between neighbouring molecules in the filament analogous to the head-neck intramolecular interaction in the 10 S conformer. It is possible that the 10 S myosin state represents an inert form involved in the control of filament assembly during muscle growth and development. Removal of regulatory light chains or labelling the reactive heavy chain thiol of myosin prevents, or at least disfavours, formation of the folded 10 S conformer and allows separation of the modified protein from the native molecules.     

Position of Mercenaria regulatory light-chain Cys50 site on the surface of myosin visualized by electron microscopy

Mercenaria regulatory light-chains, specifically labelled at cysteine 50 with N-iodoacetyl-N'-biotinylhexylenediamine, were rebound to regulatory light-chain denuded scallop myosin, and the hybrid myosin formed was decorated with avidin. These hybrid myosins were visualized by rotary-shadowing electron microscopy. Three distinct images of avidin-decorated hybrid myosin molecules were obtained. These comprise singly decorated molecules, where the avidin is bound symmetrically or asymmetrically with respect to the two heads of myosin, in addition to "figures-of-five", where two myosin molecules associate with a centrally placed avidin molecule. Analysis of these images indicates that the Mercenaria regulatory light-chain Cys50 site is located 15 to 35 A from the head-rod junction when the light-chain is bound in situ to myosin. Implications with respect to head topology and probe studies are discussed.     

Conformation of the myosin motor during force generation in skeletal muscle

Myosin motors drive muscle contraction, cytokinesis and cell locomotion, and members of the myosin superfamily have been implicated in an increasingly diverse range of cell functions. Myosin can displace a bound actin filament several nanometers in a single interaction. Crystallographic studies suggest that this 'working stroke' involves bending of the myosin head between its light chain and catalytic domains. Here we used X-ray fiber diffraction to test the crystallographic model and measure the interdomain bending during force generation in an intact single muscle fiber. The observed bending has two components: an elastic distortion and an active rotation that generates force. The average bend of the force-generating myosin heads in a muscle fiber is intermediate between those in crystal structures with different bound nucleotides, and the C-terminus of the head is displaced by 7 nm along the actin filament axis compared with the in vitro conformation seen in the absence of nucleotide.     

Essential light chain exchange in scallop myosin

The exchange of essential light chains (SH-LCs) of scallop myosin was followed with the aid of scallop SH-LC alkylated with 14C-labeled iodoacetate. More than 70% of the SH-LCs were exchanged in myosin preparations that were desensitized by removal of both regulatory light chains (R-LCs) with ethylenediaminetetraacetic acid (EDTA) treatment. Although desensitized myosin solubilized with 0.6 M NaCl or with 10 mM adenosine 5'-triphosphate (ATP) in the absence of salt equilibrated rapidly with SH-LCs even in the cold, exchange in myosin filaments required elevated temperatures. Equilibration of the SH-LCs in desensitized preparations did not depend on ATP or magnesium ions but was significantly accelerated by actin. The desensitized myosin preparations containing alkylated SH-LCs (approximately 1 mol of thiol alkylated/mol of SH-LC) readily recombined with R-LCs. The preparations regained fully the calcium dependence of the actin-activated magnesium adenosinetriphosphatase (Mg-ATPase), contained R-LCs and SH-LCs in equimolar amounts, and had an ATPase activity similar to that of untreated myosin preparations. R-LCs interfered with the equilibration of the SH-LCs. In intact myosin preparations, the exchange of SH-LCs was slow and was frequently associated with the dissociation of the R-LCs. The blocking action of the R-LC on SH-LC exchange agrees with evidence showing that the two light chain types interact and suggests that parts of the SH-LC may lie between the R-LC and the heavy chain of myosin.     

Calcium-dependent structural changes in scallop heavy meromyosin

The mechanism of calcium regulation of scallop myosin is not understood, although it is known that both myosin heads are required. We have explored possible interactions between the heads of heavy meromyosin (HMM) in the presence and absence of calcium and nucleotides by sedimentation and electron microscope studies. The ATPase activity of the HMM preparation was activated over tenfold by calcium, indicating that the preparation contained mostly regulated molecules. In the presence of ADP or ATP analogs, calcium increased the asymmetry of the HMM molecule as judged by its slower sedimentation velocity compared with that in EGTA. In the absence of nucleotide the asymmetry was high even in EGTA. The shift in sedimentation occurred with a sharp midpoint at a calcium level of about 0.5 microM. Sedimentation of subfragment 1 was not dependent on calcium or on nucleotides. Modeling accounted for the observed sedimentation behavior by assuming that both HMM heads bent toward the tail in the absence of calcium, while in its presence the heads had random positions. The sedimentation pattern showed a single peak at all calcium concentrations, indicating equilibration between the two forms with a t(1/2) less than 70 seconds. Electron micrographs of crosslinked, rotary shadowed specimens indicated that 81 % of HMM molecules in the presence of nucleotide had both heads pointing back towards the tail in the absence of calcium, as compared with 41 % in its presence. This is consistent with the sedimentation data. We conclude that in the "off" state, scallop myosin heads interact with each other, forming a rigid structure with low ATPase activity. When molecules are switched "on" by binding of calcium, communication between the heads is lost, allowing them to flex randomly about the junction with the tail; this could facilitate their interaction with actin in contracting muscle.     

Chimaeric myosin regulatory light chains: sub-domain switching experiments to analyse the function of the N-terminal EF hand.

The regulatory light chains (RLCs) located on the myosin head, regulate the interaction of myosin with actin in response to either Ca2+ or phosphorylation signals. The RLCs belong to a family of calcium binding proteins and are composed of four "EF hand" ancestral calcium binding motifs (numbered I to IV). To determine the role of the first EF hand (EF hand I) in the regulatory process, chimaeric light chains were constructed by protein engineering, by switching this region between smooth muscle and skeletal muscle myosin RLCs. For example, chimaera G(I)S consisted of EF hand I of the smooth muscle (gizzard) RLC and EF hands II to IV of the skeletal muscle RLC, whereas chimaera S(I)G consisted of EF hand I of the skeletal muscle RLC and EF hands II to IV of the smooth muscle RLC. The chimaeric RLCs were expressed in Escherichia coli using the pLcII expression system, and after isolation and purification their regulatory properties were compared with those of wild-type smooth and skeletal muscle myosin RLCs. The chimaeric RLCs bound to the myosin heads in scallop striated muscle myofibrils from which the endogenous RLCs had been removed ("desensitized" myofibrils) with similar affinities to those of the wild-type smooth and skeletal muscle RLCs. Both chimaeric RLCs were able to regulate the actin-activated Mg(2+)-ATPase activity of scallop myosin: G(I)S inhibited the ATPase in the presence and absence of Ca2+, like the wild-type skeletal muscle RLC, while S(I)G inhibited the myosin ATPase in the absence of Ca2+, and this inhibition was relieved on Ca2+ addition, in the same way as the wild-type smooth muscle RLC. Thus the type of regulation that the RLCs confer on the myosin is determined by the source of EF hands II to IV rather than that of EF hand I.     

Rotational dynamics of the regulatory light chain in scallop muscle detected by time-resolved phosphorescence anisotropy.

We have used time-resolved phosphorescence anisotropy (TPA) to study the rotational dynamics of chicken gizzard regulatory light chain (RLC) bound to scallop adductor muscle myofibrils in key physiological states. Native RLC from scallop myofibrils was extracted and replaced completely with gizzard RLC labeled specifically at Cys 108 with erythrosin iodoacetamide (ErIA). The calcium sensitivity of the ATPase activity of the labeled myofibril preparation was quite similar to that of the native sample, indicating that the ErIA-labeled RLC is functionally bound to the myosin head. In rigor (in the absence of ATP, when all the myosin heads are rigidly bound to the thin filament), a slight decay was observed in the first few microseconds, followed by no change in the anisotropy. This indicates small-amplitude restricted motions of the RLC or the entire LC domain of myosin. Addition of calcium to rigor restricts these motions further. Relaxation with ATP (no Ca) causes a large decay in the anisotropy, indicating large-amplitude rotational motion with correlation times of 5-50 micros. Further addition of calcium, to induce contraction, resulted in a decrease in the rate and amplitude of anisotropy decay. In particular, there is clear evidence for a slow rotational motion with a correlation time of approximately 300 micros, which is not present either in rigor or relaxation. This indicates rotational motion that specifically correlates with force generation. The changes in the rotational dynamics of the light-chain domain in rigor, relaxation, and contraction support earlier work based on probes of the catalytic domain that muscle contraction is accompanied by a disorder-to-order transition of the myosin head. However, the motions of the LC domain are different from those of the catalytic domain, which indicates rotation of the two domains relative to each other.     

Refined model of the 10S conformation of smooth muscle myosin by cryo-electron microscopy 3D image reconstruction

The actin-activated ATPase activity of smooth muscle myosin and heavy meromyosin (smHMM) is regulated by phosphorylation of the regulatory light chain (RLC). Complete regulation requires two intact myosin heads because single-headed myosin subfragments are always active. 2D crystalline arrays of the 10S form of intact myosin, which has a dephosphorylated RLC, were produced on a positively charged lipid monolayer and imaged in 3D at 2.0 nm resolution by cryo-electron microscopy of frozen, hydrated specimens. An atomic model of smooth muscle myosin was constructed from the X-ray structures of the smooth muscle myosin motor domain and essential light chain and a homology model of the RLC was produced based on the skeletal muscle S1 structure. The initial model of the 10S myosin, based on the previous reconstruction of smHMM, was subjected to real space refinement to obtain a quantitative fit to the density. The smHMM was likewise refined and both refined models reveal the same asymmetric interaction between the upper 50 kDa domain of the "blocked" head and parts of the catalytic, converter domains and the essential light chain of the "free" head observed previously. This observation suggests that this interaction is not simply due to crystallographic packing but is enforced by elements of the myosin heads. The 10S reconstruction shows additional alpha-helical coiled-coil not seen in the earlier smHMM reconstruction, but the location of one segment of S2 is the same in both.     

Interhead fluorescence energy transfer between probes attached to translationally equivalent sites on the regulatory light chains of scallop myosin

Interhead fluorescence energy transfer studies between probes located at translationally equivalent sites on the two heads of scallop myosin indicates that the distance between such sites is no less than 50 A. Regulatory light chains, possessing either one (Mercenaria, chicken gizzard) or two (Loligo, rabbit skeletal) sulfhydryl groups, were modified either with 1,5-IAEDANS (N'-iodoacetyl-N'-(1-sulfo-5-n-naphthyl)ethylenediamine), as energy transfer donor, or with IAF (5-(iodoacetamido)fluorescein) or DABMI (4-dimethylaminophenylazophenyl-4'-maleimide), as energy transfer acceptor. The sulfhydryl groups on these light chains are located at different positions within the regulatory light-chain primary sequence; this enables one to probe a variety of locations, with respect to regulatory light-chain topology, on each myosin head. These independently modified regulatory light chains were added back to desensitized scallop myosin under a variety of conditions, including biphasic re-addition, the aim being to maximize the number of interhead energy transfer couples present. The efficiency of energy transfer was determined on the same samples by both steady-state and time-decay techniques. Results obtained by these two techniques were in good agreement with each other and indicated that the efficiency of energy transfer did not exceed 20% in any of the hybrids studied. Transfer efficiencies were invariant, irrespective of the presence or absence of MgATP, calcium or actin, either separately or in combination. Results using heavy meromyosin at low ionic strength were identical. It is shown that these results, in conjunction with the results of recent crosslinking studies performed on comparable myosin hybrids, may place certain restrictions on the configurations of the two heads of myosin.     

MgATP binding changes neither the shape nor the flexibility of the heads of single myosin molecules.

The structural mechanism by which myosin heads exert force is unknown. One possibility is that the tight binding of the heads to actin drives them into a force-generating configuration. Another possibility is that the force-generating conformational change is inherent to the myosin heads. In this case the heads would make force by changing their shape according to the species of nucleotide in their active sites, the tight attachment to actin serving only to provide traction. To test this latter possibility, we used negative stain electron microscopy to search for a MgATP-induced shape change in the heads of single myosin molecules. We compared the heads of 10S smooth muscle myosin monomers (wherein MgATP is trapped at the active site) with the MgATP-free heads of 6S monomers. We found that to a resolution of about 2 nm, MgATP binding to the unrestrained myosin head does not drive it to change its shape or its flexibility. This result suggests that the head makes force by virtue of an induced fit to actin.     

Fluorescence intensity and UV absorption changes accompanying dissociation and association of regulatory light chain of scallop adductor myosin

Dissociation and association of regulatory light chains of scallop myosin were found to be accompanied by changes in the fluorescence intensity and in the UV absorption spectrum. The changes in the two optical properties of scallop myosin and the dissociation and association of regulatory light chains were studied as a function of the magnesium and calcium concentrations. The results thus obtained suggested that there are two different types of attachment between regulatory light chains and "desensitized" myosin; one type is a calcium-specific attachment, and the other type of attachment can be mediated by either calcium or magnesium ions. These changes in the optical properties of scallop myosin were distinguishable from those induced by Mg-ATP; for example, with "desensitized" scallop myosin, the former changes were not observed but the latter were.     

Structure of the myosin projections on native thick filaments from vertebrate skeletal muscle

Rabbit psoas muscle filaments, isolated in relaxing buffer from non-glycerinated muscle, have been applied to hydrophilic carbon films and stained with uranyl acetate. Electron micrographs were obtained under low-dose conditions to minimize specimen damage. Surrounding the filament backbone, except in the bare zone, is a fringe of clearly identifiable myosin heads. Frequently, both heads of individual myosin molecules are seen, and sometimes a section of the tail can be seen connecting the heads to the backbone. About half the expected number of heads can be counted, and they are uniformly distributed along the filament. The majority of heads appear curved. The remainder could be curved heads viewed from another aspect. Three times as many heads curve in a clockwise sense than in an anticlockwise sense, suggesting a preferential binding of one side of the head to the carbon film. The two heads of myosin molecules exhibit all the possible combinations of clockwise, anticlockwise and straight heads, and analysis of their relative frequencies suggests that the heads rotate freely and independently. The heads also adopt a wide range of angles of attachment to the tail. The lengths of heads cover a range of 14 to 26 nm, with a peak at 19 nm. The average maximum width is 6.5 nm. Both measurements are in excellent agreement with values for shadowed molecules. Since our data are from heads adsorbed to the film in relaxing conditions and the shadowed molecules were free of nucleotide, gross shape changes are not likely to be produced by nucleotide binding. The length of the link between the heads and the backbone was found to vary between 10 nm and 52 nm, with a broad peak at about 25 nm. Thus, the hinge point detected in the tail of isolated molecules was not usually the point from which the crossbridges swung out from the filament surface. The angle made by the link to the filament axis was between 20 degrees and 80 degrees, with a broad maximum around 45 degrees. These lengths and angles concur with our observation of an average limit of the crossbridges from the filament surface of 30 nm. This is sufficient to enable heads in the myofibril lattice to reach out beyond the nearest thin filament and should allow considerable flexibility for stereospecific binding to actin in active muscle.