During force-generating steps of the muscle crossbridge cycle, the tip of the myosin motor, specifically loop-4, contacts the tropomyosin cable of actin filaments. In the current study, we determined the corresponding effect of myosin loop-4 on the regulatory positioning of tropomyosin on actin. To accomplish this, we compared high-resolution cryo-EM structures of myosin S1-decorated thin filaments containing either wild-type or a loop-4 mutant construct, where the seven-residue portion of myosin loop-4 that contacts tropomyosin was replaced by glycine residues, thus removing polar side chains from residues 366–372. Cryo-EM analysis of fully decorated actin-tropomyosin filaments with wild-type and mutant S1, yielded 3.4–3.6 Å resolution reconstructions, with even higher definition at the actin-myosin interface. Loop-4 densities both in wild-type and mutant S1 were clearly identified, and side chains were resolved in the wild-type structure. Aside from loop-4, actin and myosin structural domains were indistinguishable from each other when filaments were decorated with either mutant or wild-type S1. In marked contrast, the position of tropomyosin on actin in the two reconstructions differed by 3 to 4 Å. In maps of filaments containing the mutant, tropomyosin was located closer to the myosin-head and thus moved in the direction of the C-state conformation adopted by myosin-free thin filaments. Complementary interaction energy measurements showed that tropomyosin in the mutant thin filaments sits on actin in a local energy minimum, whereas tropomyosin is positioned by wild-type S1 in an energetically unfavorable location. We propose that the high potential energy associated with tropomyosin positioning in wild-type filaments favors an effective transition to B- and C-states following release of myosin from the thin filaments during relaxation.

The motor protein myosin projects from muscle thick filaments and binds to the actin of thin filaments. Actin-binding itself catalyzes the hydrolysis of ATP by myosin, thus coupling ATP turnover to force generation and ultimately powering thick filament movement along thin filaments during the muscle crossbridge cycle. In striated muscles, the binding of myosin to actin is regulated by troponin and tropomyosin on the thin filaments, thereby controlling myosin-mediated force generation (reviewed in Gordon et al., 2000; Powers et al., 2021).

Tropomyosin is a 40-nm-long coiled coil–protein that stretches over seven neighboring actin subunits on the surface of muscle thin filaments. By polymerizing on actin in a head-to-tail fashion, it forms a continuous cable that follows the thin filament’s long-pitch actin helix (Lorenz et al., 1995; Perry, 2003; Brown and Cohen, 2005; Hitchcock-DeGregori, 2008; Holmes and Lehman, 2008; Li et al., 2011; Orzechowski et al., 2014a; Lehman, 2016). The repositioning of the tropomyosin cable on actin in different regulatory states ultimately controls muscle contraction (Gordon et al., 2000; Lehman, 2016; Powers et al., 2021). This regulatory relocalization of tropomyosin on actin is determined by electrostatic linkages and the dual effects of troponin and myosin-association with actin (Lorenz et al., 1995; Orzechowski et al., 2014b; Orzechowski et al., 2014c; Pavadai et al., 2020). Tropomyosin assumes one of the three average azimuthal locations on the actin of thin filaments, which interfere with or facilitate myosin motor head binding and crossbridge cycling (Vibert et al., 1997; Poole et al., 2006; Lehman, 2016; Yamada et al., 2020; Pavadai et al., 2020; Doran et al., 2020; Doran et al., 2022; Risi et al., 2017; Risi et al., 2021). These are termed the low-Ca2+ troponin-mediated blocked or relaxed B-state, the default Ca2+-induced C-state, and the myosin-induced fully active M-state. However, the positions represented are not likely to be singularly defined discrete steps in a regulatory pathway and are better described as the integrated result of a dynamic equilibrium biased to one or another conformation dependent on troponin- and myosin-binding (McKillop and Geeves, 1993; Geeves, 2012; Lehman, 2017). In the absence of Ca2+, troponin functions to draw tropomyosin away from its energetically favored C-state position to the B-state, thereby blocking myosin-binding and resulting in muscle relaxation (Potter and Gergely, 1974; Lehman et al., 1994; Yamada et al., 2020; Lehman et al., 2021). In contrast, elevated Ca2+ binding to troponin favors the C-state and is followed by myosin interactions that drive tropomyosin to its M-state position, thus reversing tropomyosin’s inhibition and promoting contraction (Vibert et al., 1997; Behrmann et al., 2012; Doran et al., 2020; Doran et al., 2022). These transitions are all allosterically coupled to each other and are cooperative (Geeves, 2012). Relatively low energy barriers separate B- and C-state positions of tropomyosin on actin, whereas the M-state tropomyosin position is energetically unfavorable (Lehman et al., 2000; Kiani et al., 2019; Baldo et al., 2021). Hence, the regulatory effects of troponin and myosin are to move tropomyosin away from its default C-state position and transiently trap it in respective B- and M-states, like the action of a pick pulling on and then releasing a vibrating guitar string (Maytum et al., 2008).

The structural basis for steric inhibition of myosin binding to actin by tropomyosin is well understood. Here at low Ca2+, tropomyosin is constrained in its inhibitory B-state position over myosin-binding sites on actin by a C-terminal extended domain of troponin subunit I (TnI) that simultaneously binds to tropomyosin and F-actin (Potter and Gergely, 1974; Yamada et al., 2020; Lehman et al., 2021). Numerous studies indicate that this structural constraint obstructs actin-myosin interaction and as mentioned results in muscle relaxation (Gordon et al., 2000; Lehman, 2016; Powers et al., 2021). The reverse case, namely, the structural basis for myosin-induced movement of tropomyosin to the fully active M-state is more poorly understood. We and others have shown that the only part of myosin in close contact with tropomyosin in the M-state is a short string of amino acids on the tip of myosin loop-4, which is located at the end of the motor head (Behrmann et al., 2012; Doran et al., 2020; Doran et al., 2022; Risi et al., 2021) and are conserved in all muscle myosin-2 isoforms (Doran et al., 2020). Although the polar side chains projecting from loop-4 residues (R369 and Q368 in β-cardiac myosin) come into close proximity with tropomyosin, it remains uncertain if they drive the tropomyosin to the M-state during activation and/or facilitate return of the coiled coil in the reverse direction during relaxation. In the current study we examine these possibilities by comparing cryo-EM structures of thin filaments decorated with wild-type human myosin S1 (WT-S1) and those containing an engineered myosin S1 in which R369, Q368, and five additional conserved residues that surround them on myosin loop-4 (L366, K367 and E370, E371, Q372) were mutated to glycine (7G-S1). In addition, corresponding in silico work was carried out to determine the interaction energetics associated with the M-state positions. In the 7G-S1 mutant, substitution of glycine for polar residues leads to deficient M-state movement of tropomyosin following S1 binding on actin, accompanied by a local perturbation of both actin-tropomyosin and actin-myosin energetics. These studies suggest that the presence of extended side chains on myosin loop-4 is required for WT-S1 to position tropomyosin optimally in the M-state.

Thin filament sample preparation

Porcine cardiac F-actin, with 100% sequence identity to human cardiac α-actin, and human cardiac α-tropomyosin (Tpm1.1), containing an N-terminal Ala-Ser extension, were prepared as previously described (Orzechowski et al., 2014b). Wild-type recombinant β-cardiac myosin S1 (residues 1–842) was expressed in C2C12 mammalian cells as previously reported (Touma et al., 2022) following high titer infection of cells by recombinant adenovirus (Vector Biolabs). The construct contained standard C-terminal Avi and FLAG tags for FLAG affinity chromatography purification in order to maximize our yield and purity for cryo-EM studies (Swenson et al., 2017; Tang et al., 2019; Tang et al., 2021; Touma et al., 2022). Mutant human β-cardiac myosin S1 in which amino acids 366–372 were replaced by glycine residues was expressed by the same method. In each case, the purified S1 contained mouse skeletal muscle regulatory and essential light chains (Swenson et al., 2017). The S1 containing the murine light chains has been well characterized in previous enzymatic and in vitro motility studies and the properties are similar to those of native cardiac S1 (Swenson et al., 2017; Tang et al., 2019; Tang et al., 2021; Touma et al., 2022). In addition, we note in the Results and discussion section that we did not observe structural differences between the expressed S1 and native cardiac S1 anywhere outside of loop-4.

Thin filament samples were prepared for cryo-EM as previously (Doran et al., 2020; Doran et al., 2022) by first mixing F-actin and tropomyosin to final concentrations of 3 μM actin and 9 μM tropomyosin in 50 mM sodium acetate, 3 mM MgCl2, 1 mM dithiothreitol, and 10 mM MOPS buffer at pH 7.0 at 25°C. Prior to applying 1.5 μl samples to freshly glow-discharged Quantifoil R1.2/1.3 400 mesh Gold grids (Electron Microscopy Sciences), the surfactant octyl β-D-glucopyranoside (Sigma-Aldrich) and bacitracin were added to the actin-tropomyosin mixture to respective final concentrations of 0.3 and 0.1%. Excess tropomyosin was used relative to actin to ensure filament saturation and the surfactants added to promote filament spreading and limit filament adsorption to the air−water interface (Noble et al., 2018; Doran et al., 2020). The grid sample was manually blotted for 1 s at 10°C and 100% humidity, and then a 1.5 μl drop of 9.5 μM myosin-S1 subfragment including 0.3% octyl β-D-glucopyranoside was applied to the blotted grid sample. The sample was blotted again for 6 s and plunge-frozen in liquid ethane using a Vitrobot Mark III System (FEI/Thermo Fisher Scientific) as previously described (Doran et al., 2020; Doran et al., 2022).

Cryo-EM data collection

Samples were examined with a Titan Krios transmission electron microscope (Thermo Fisher Scientific) at 300 kV at the Purdue Cryo-EM Facility. Micrographs were recorded as exposure series movies at a magnification of 81,000× (pixel size, 0.5394 Å) in which 40 frames were sequentially exposed for 0.8 s each and collected on a Gatan K3 Summit direct electron detector (Gatan/AMETEK) using a 20 eV energy filter together with Leginon data acquisition software (Suloway et al., 2005) at a dose rate of 17.2 e2/s for a total exposure of 3.2 s (54 e2). 3,909 movies were collected during data acquisition for the previously documented wild-type actin–tropomyosin-S1 complex (Doran et al., 2022), and 4,171 movies were additionally collected in the current work for the actin-tropomyosin complex containing mutant S1.

Helical reconstruction

The RELION 4.0 suite of programs (Scheres, 2012; Scheres, 2016; He and Scheres, 2017) was used for helical reconstruction of cryo-EM images of mutant S1-decorated thin filaments (Fig. S1). The protocol has been previously described in full for filaments containing wild-type S1 (Doran et al, 2020, 2022; Fig. S1 A). In brief, after motion correction with MotionCor2 (Zheng et al., 2017) and then contrast transfer function (CTF) determination with CtfFind 4.1.13 (Rohou and Grigorieff, 2015), ∼30,000 filament stretches were manually chosen and divided into 47-nm-long overlapping segments with an inter-segment offset of 27.5 Å. Reference-free 2-D classification was then performed to generate initial templates for subsequent automatic selection of filament segments (autopicking) and to facilitate removing damaged or distorted segments. After autopicking and additional 2-D classification, 3-D classification was carried out to yield six 3-D classes using a featureless cylinder as a reference. The highest resolution 3-D class, containing 187,885 segments of the mutant S1-decorated filaments, was selected and used for further 3-D auto-refine processing. The corresponding data set previously analyzed for the wild-type filaments was comparable in size (176,178 segments). The first refinement produced a 4.3 Å resolution map, which improved to 3.6 Å after further CTF refinement and Bayesian polishing. Final helical parameters for the filament reconstruction converged on an actin-actin inter-subunit twist of 166.8° and a protomer rise of 27.9 Å (Table S1).

Model building

An atomic model of wild-type S1-decorated actin-tropomyosin (PDB ID 8EFI) was used as an initial reference to build a corresponding structure of the thin filament containing mutant S1. The wild-type model was first fit into the cryo-EM reconstruction generated above of the mutant complex using the Fit in Map tool in UCSF-Chimera (Pettersen et al., 2004). Myosin residues 366–372 were then manually mutated to glycine and subsequently subjected to iterative rounds of manual rebuilding and refinement in the programs Coot and Phenix (Emsley et al., 2010; Liebschner et al., 2019). Each step in the corresponding Phenix real-space-refinement included conjugate gradient minimization, rigid body fitting, and a local grid search. Secondary structure and Ramachandran constraints as well as ideal bond and angle restraints (target RMSD of 0.005 Å and 0.5°) were applied to maintain standard geometry. The refinement procedure minimized the clash score and produced satisfactory Ramachandran and other fiduciary values (Table S1 and Fig. S1). Those portions of the model that fell outside the reconstruction were not considered further.

Interaction energy determination and molecular dynamics simulations

The models for interaction energy analysis consisted of one tropomyosin dimer, six myosin S1, and six actin monomers aligned along one long-pitch helix of actin fit into the cryo-EM maps of wild-type and mutant S1-decorated thin filaments. Tropomyosin models were first generated to match the actin helical parameters from cryo-EM using previously published programs (Lorenz et al., 1995), with an internal rotation close to that of the refined low calcium structure. This was then rigid body fit into the wild-type and mutant maps in Chimera (Pettersen et al., 2004), minimized and side chain interactions optimized using Molecular Dynamics Flexible Fitting in NAMD 2.13 (Phillips et al, 2005, 2020; Trabuco et al, 2008, 2009). A similar procedure was used to add side chains to tropomyosin in the cryo-EM structure of actin and tropomyosin (PDB ID 5JLF; EMDB ID 8162; von der Ecken et al., 2016) for comparison to the S1-decorated filaments. The interaction energy between tropomyosin and actin/myosin was determined using the NAMDEnergy plugin implemented in VMD (Humphrey et al., 1996). For tropomyosin/actin interactions, only the actin monomers in common between the S1-decorated and S1-free structures were used.

Further molecular dynamics simulations of tropomyosin on actin in the open state, after removal of the S1, started from the mutant and wild-type S1-decorated structures from the molecular dynamics flexible fitting above. First, the mutant S1 structure was rotated and then combined with the wild-type structure to create a full filament with the mutant S1 on one long pitch helix and wild-type on the opposite side. This structure was then centered at the origin and extended along the z-axis using the measured actin helical symmetry to a total of 30 actin monomers and eight tropomyosin dimers. This model was then solvated in water with a box size of 140 × 140 × 780.92 Å in VMD. At this point, the four actin monomers and four tropomyosin dimers on the ends of the system were removed, since these were only placeholders for the solvation step. Sodium chloride and magnesium chloride were then added to the system to final concentrations of 0.15 M and 3 mM, respectively. Since the system is aligned to the z-axis and the length of the z-axis in the solvent box matches the expected actin helical rise, the periodic boundary conditions applied during the simulation will enforce actin and tropomyosin intermolecular contacts at the ends to create an infinite filament. The system was then minimized in the CHARMM 36 force field (Huang et al., 2017) using a non-bonded cutoff of 14 Å with switching in NAMD 2.13 to relieve poor contacts in the initial structure, first with only waters moving, and then the protein atoms moving with slowly released harmonic constraints to the starting coordinates. The system was then heated to 310K at constant volume and the system allowed to equilibrate over 0.72 ns with slow release of harmonic constraints keeping temperature constant using the Langevin temperature bath. Once the system was fully heated and equilibrated, the Langevin piston was used to keep the pressure of the system close to 1 bar for the production runs for a total of 100 ns. The position of tropomyosin over the simulation was monitored by the position of the superhelical axis as measured by the program Twister (Strelkov and Burkhard, 2002).

Online supplemental material

Fig. S1 summarizes the workflow to generate the actin-tropomyosin reconstructions reported and demonstrates the quality of the data presented. Table S1 shows cryo-EM data collection and refinement statistics for F-actin–tropomyosin decorated with human cardiac S1, in which loop-4 residues 366–372 were mutated to glycine (7G-S1) compared to refinement statistics for F-actin–tropomyosin decorated with control wild-type human cardiac S1 (WT-S1).

A major goal of our ongoing studies is to define regulatory states of human cardiac thin filaments by reconstructing cryo-EM images containing human proteins. Once elucidated, cryo-EM structures can be used to determine the interaction energetics of the components that participate in cardiac muscle contraction and relaxation. In the current work, we concentrated on the role played by myosin loop-4 in configuring tropomyosin on actin in the regulatory M-state. Here, we compared thin filament interactions and corresponding tropomyosin positioning induced by side chains extending from loop-4 of wild-type myosin S1 (WT-S1) to those brought about by an engineered S1 lacking these side chains (7G-S1).

S1 and tropomyosin binding to F-actin

Our samples for cryo-EM analysis consisted of porcine cardiac α-actin isolated from tissue (porcine and human α-actin share 100% sequence identity), human cardiac α-tropomyosin (Tpm1.1) expressed in Escherichia coli, and wild-type or mutant human β-cardiac myosin S1 expressed in C2C12 mouse cells. Filament complexes were assembled under nucleotide-free (rigor) conditions. Electron micrographs of F-actin–tropomyosin filaments decorated with either the wild-type or mutant S1 showed distinctive “arrowhead” formation and uniform labeling of the S1 constructs on actin (Fig. 1). SDS-PAGE following cosedimentation of F-actin−tropomyosin containing either bound wild-type S1 or our 7G glycine-substituted loop-4 mutant S1 indicated typical 7:1 mol:mol stoichiometric levels of tropomyosin on actin in such samples (not shown).

3-D reconstruction of wild-type and 7G-mutant S1 bound F-actin–tropomyosin

Reconstructions were generated from the cryo-EM images recorded on a Titan Krios cryo-EM. The RELION suite of image-processing programs was used to produce high-resolution helical reconstruction maps for the wild-type (3.4 Å; EMDB ID EMD28083) and for the mutant (3.6 Å; EMDB ID EMD28270) complexes as done previously (Doran et al., 2020; Doran et al., 2022). Local resolution varies across the reconstructed structures in these and earlier reconstructions of actomyosin complexes containing tropomyosin (Behrmann et al., 2012; Doran et al., 2020; Risi et al., 2021), with the highest resolution region occurring closest to the central filament axis and at the actin-S1 interface, currently reaching a value of 3.2 Å (Fig. 2). In the present case, density is evident for actin side chains (Fig. S1, B and C), as well as for those on the surface of myosin interacting with actin, including loop-4 side chains of WT-S1 approaching actin and tropomyosin, as communicated previously (Doran et al., 2020; Risi et al., 2021). In addition, secondary structure is observed on the remainder of the myosin head in maps of both the wild-type and mutant myosin. The 7G-S1 substituted glycine residues on loop-4 understandably lack side chain density. The backbone structure of the glycine residues themselves is unresolved, and thus the 7G-S1 loop-4 extension is featureless in corresponding reconstructions (Figs. 2 and 3). Aside from loop-4 itself, reconstructions of the wild-type and the 7G mutant human myosin head structures and their corresponding interactions with actin are indistinguishable from each other at the resolution achieved nor are they distinguishable from structures of cardiac myosin that we and others solved previously for the nucleotide–free cardiac actomyosin complex.

Fitting atomic structures, including those for actin (PDB ID 5JLH), tropomyosin (PDB ID 7UTI), and myosin (PDB ID 6X5Z) to the reconstruction of the wild-type actomyosin complex, illustrates loop-4 extending over the region of the actin surface between subdomains 1 and 3 and pointing toward tropomyosin (Fig. 2, D and E; PDB ID 8EFI). As previously observed, the polar side chains of loop-4 residues R369 and Q368 approach tropomyosin closely in the fitting (Fig. 3). In contrast, even though all loop-4 side chains extending toward tropomyosin have been eliminated in the mutant, direct contact between the α-carbon backbone of the mutant myosin motor domain and tropomyosin is not evident, possibly because the tip of the loop is more flexible (PDB ID 8ENC). In fact, apart from myosin loop-4, other loop-like extensions (i.e., the helix-loop-helix motif, cardiomyopathy loop, surface loops 2, 3, and the activation loop) in the 7G mutant myosin appear unperturbed and show residue-to-residue contacts in common with WT-S1 and those found in previously represented reconstructions. This suggests that disorder at the tip of loop-4 does not affect other actin-binding interactions under nucleotide-free conditions used to mimic the M-state of the actomyosin complex. Thus, superposition of the reconstructions shows no obvious differences in the S1 motor heads distal to the actin-S1 interface in the rigor configuration.

Myosin loop-4 positions tropomyosin in the M-state

The tropomyosin coiled coil is well-resolved in the reconstructions of our samples containing either wild-type or mutant 7G-S1, in each case lying over inner actin subdomains 3 and 4. However, while the actin and S1 in the wild-type and mutant filaments are directly superposable, the tropomyosin densities in these reconstructions are not (Fig. 2, C and G–I). The azimuthal position of the human tropomyosin localizes on wild-type S1-decorated filaments as previously observed (Behrmann et al., 2012; Doran et al., 2020), with tropomyosin separated from the tips of actin-bound S1 by the ∼7-Å-long arginine and neighboring glutamine side chains. In contrast, the position of tropomyosin on actin in the 7G-S1 decorated filaments differs and apparently is no longer fixed axially or azimuthally along actin by the glycine residues that replaced R369 and Q368 (Fig. 2). Here, tropomyosin lies 3–4 Å closer to the mutant 7G-S1 lacking these outwardly projecting side chains. Therefore, the tropomyosin coiled coil now localizes marginally nearer to the C-state position which it normally occupies in the absence of myosin-binding. The aberrant M-state location of tropomyosin is not a result of averaging bias associated with RELION processing since the same distinct tropomyosin position was observed when individual reconstructions were made from four independent subsets of data and from images collected from different EM grids. In each case, the maps showed tropomyosin located in the unique loop-4-dependent position displayed by the respective full data sets.

Actin-tropomyosin interaction energetics is affected differently by WT-S1 and 7G-S1–binding

Tropomyosin-actin energetics

On average, the backbone of the tropomyosin cable lies 8–10 Å from the actin surface (Lorenz et al., 1995), i.e., at a distance that precludes significant nonpolar interactions between the two thin filament components. It follows therefore that tropomyosin position and its binding to actin is likely to be dominated by electrostatic interactions, which, as has been extensively reported, still operate at such distances (Lorenz et al., 1995; Brown and Cohen, 2005; Holmes and Lehman, 2008). Electrostatic coupling between respective oppositely charged amino acid side chains is particularly evident when tropomyosin occupies its “ground-state” C-location on actin, namely when its location is not further restricted either by troponin in the absence of Ca2+ or by myosin in the presence Ca2+ (Li et al., 2011; Orzechowski et al., 2014a; Rynkiewicz et al., 2015). The current enhanced resolution described above as well as that obtained from recent cryo-EM derived models of thin filament components have now provided renewed incentive to understand the energetics of tropomyosin-actin interactions with greater precision.

Interaction energy measurements based on our cryo-EM modeling of tropomyosin’s M-state position confirm that actin-tropomyosin interaction weakens after tropomyosin is displaced from favorable B- and C-states positions by the strong binding of wild-type myosin to actin (Table 1). While salt bridges between loop-4 of wild-type myosin residues R369 and Q368 and tropomyosin contribute to the M-state interaction marginally, these links are insufficient to compensate for the loss of B- and C-state linkages or likely to account for meaningful actin-tropomyosin electrostatic contacts as tropomyosin traverses actin to its M-state position. In fact, our modeling and quantitation of interaction energies confirm that the M-state position of tropomyosin does not result from an abundance of newly formed salt bridges between coiled-coil tropomyosin and actin or myosin (Table 1). It is more likely that tropomyosin in the M-state remains bound to actin only because it is part of a continuous polymeric cable which is trapped topologically in a channel formed on one side by loop-4 at the tip of myosin and on its opposite side by a helix (D222-A231) at the edge of the actin inner domain.

The “uphill” movement of tropomyosin to the M-state position apparently is offset by the strong binding of myosin to sites on actin otherwise occupied by tropomyosin in its C-state. It follows that once myosin detaches from actin, for example during relaxation, tropomyosin will move “downhill” to its default lower energy B- or C-state positions. Presumably, a large energetic difference between M-state and C-state or B-state positioned tropomyosin will create a significant mechanistic drive for tropomyosin to move to ground state configurations close to its energy minimum on actin, thus promoting relaxation.

Loop-4 side chains affect tropomyosin-actin energetics

As mentioned, the average positions of tropomyosin on actin noted in cryo-EM maps of WT-S1 and 7G-S1 thin filament complexes differ from each other by ∼3.6 Å. This seemingly minor configurational change produces a considerable difference in M-state energetics when the two conformations are compared (Table 1). In contrast to energetically unfavorable actin-tropomyosin interaction for the wild-type M-state complex, the interaction measured between M-state actin and tropomyosin in the presence of the mutant 7G-S1 association is more favorable. Thus, the diminished potential energy for the mutant 7G-S1 positioned tropomyosin may be responsible for a reduced gradient driving tropomyosin to return effectively to the B- or C-states after release of myosin from thin filaments during relaxation. Transient loss of corresponding S1-tropomyosin interaction caused by extraction of loop-4 side chains from the 7G-S1 may also be important.

Molecular dynamics (MD) simulations highlight loop-4 function

MD simulations were carried out with tropomyosin localized on actin in either the well-characterized M-state position associated with wild-type proteins or in the position associated with the mutant 7G-S1, but in each case free of constraints imposed by myosin, as if following the response to myosin head release from actin and the tendency for tropomyosin to move to more favorable positions. As in previous simulations by Kiani et al. (2019) and confirmed here, during MD tropomyosin moves away from its stereotypical M-state position on myosin-free actin and approaches its C-state position (Fig. 4). In marked contrast, MD initiated with tropomyosin positioned in the aberrant 7G-location shows the coiled coil remains largely in place on actin despite beginning at an azimuthal position closer to the C-state. If anything, here tropomyosin often moves in the opposite direction and away from the C-state (Fig. 4). This unexpected behavior is not observed for tropomyosin moving from its standard M-state starting position. These results suggest that in the wild-type actomyosin complex, myosin loop-4 specifically localizes tropomyosin outside of a broad energy basin (Orzechowski et al., 2014b; Orzechowski et al., 2014c; Rynkiewicz et al., 2015) and that the positioning may be optimized to render an unobstructed path to transition from the M-state to the C-state. In contrast, 7G-S1 localized tropomyosin is already positioned in the energy basin and not ideally positioned to snap back to C-state positions.

The downhill repositioning of tropomyosin away from the stereotypical M-state toward the C-state does not appear to be completely uniform along the length of the coiled coil during MD and appears to be greatest for centrally located pseudo-repeats 3, 4, and 5 of tropomyosin. It may be significant that pseudo-repeat 4 contains non-canonical Asp137 that can provide coiled-coil tropomyosin with enhanced local torsional flexibility to facilitate such movement (Lehman et al., 2020). In turn, such extra flexibility may increase the probability of tropomyosin contacting the troponin I inhibitory and C-terminal domains of Ca2+-free troponin ultimately dragging the tropomyosin cable to the B-state location during muscle relaxation.

Conclusions

All muscle myosin motors studied to date contain conserved residues and associated long arginine side chains on the tip of loop-4 (Doran et al., 2020), which we propose are needed to position tropomyosin in its average M-state position on actin. In the current work, we used cryo-EM and image reconstruction to determine the structural interactions between loop-4 of wild-type cardiac S1 and tropomyosin on actin filaments and likewise between mutant S1, lacking loop-4 side chains, and these thin filament proteins. We studied human proteins in our work to replicate thin filament conformations occurring within the human ventricle as faithfully as is currently possible. Comparison of wild-type and mutant structures provided an opportunity to identify the role played by polar side chains on the tip of loop-4 to configure tropomyosin on actin filaments and then to determine the functional importance of any related tropomyosin positional specificity. Using the Maytum et al. (2008) analogy, we argue that loop-4 and its side chains act as the pick, plucking the M-state tropomyosin guitar string, and that the 7G mutation studied degrades the pick. Cardiac myosin binding protein-C in turn may act as a secondary pick to further modulate the process (Risi et al., 2018).

The strong binding of myosin on actin during the crossbridge cycle displaces tropomyosin from its default C-state position and localizes the coiled coil to an energetically unfavorable site on actin (Orzechowski et al., 2014b; Orzechowski et al., 2014c; Rynkiewicz et al., 2015; Kiani et al., 2019; Baldo et al., 2021; Doran et al., 2022). We posit that M-state tropomyosin therefore is poised to revert to its default energy minimum once myosin releases from actin. In the current study, we found that a mutant S1 lacking loop-4 side chains, like its wild-type counterpart, binds to actin filaments and also displaces tropomyosin from its C-state position. However, in the latter case, tropomyosin does not occupy a well-suited position needed for normal regulatory transitions. We suggest that, absent loop-4 side chains projecting towards the tropomyosin cable, tropomyosin vibrates about a metastable position with insufficient downhill energetics to move appropriately to the C-state. This deficiency precludes effective muscle relaxation. We note that much of the myosin motor head, including myosin loop-4, is a target for point mutations leading to cardiomyopathies (Dellefave et al, 2009; Parker and Peckham, 2020; Trujillo et al, 2022), an observation reinforcing our conclusion that loop-4 structural specificity is vital for cardiac regulatory function.

Structural models generated by this work have been deposited in the Protein Data Bank (PDB ID 8ENC) and Electron Microscopy Data Bank (EMDB ID EMD28270). The corresponding wild-type complex has PDB and EMDB accession codes 8EFI and EMD28083, respectively.

Henk L. Granzier served as editor.

We would like to thank Thomas Klose and Frank Vago for the collection of the data at Purdue. Computational work was carried out in house and using resources provided by the Massachusetts Green High Performance Computing Center.

These studies were supported by National Institutes of Health (NIH) grants R01HL036153 (to W. Lehman) and R01HL127699 (to C.M. Yengo). M.H. Doran was supported by NIH Training Program Grant T32HL007969 (to Katya Ravid) and by the Boston University Division of Graduate Medical Sciences institutional funds. Preliminary screening of electron microscope samples was carried out in house and supported by NIH grant S10RR25434 (to E. Bullitt). Final cryo-EM data collections were done at the Purdue Cryo-EM Facility supported by the NIH Common Fund Transformative High-Resolution Cryoelectron Microscopy Program (U24GM129541).

The authors declare no competing financial interests.

Author contributions: W. Lehman thought of the general approach taken and along with M.J. Rynkiewicz and M.H. Doran developed the principal concepts presented. Under the direction of C.M. Yengo, D. Rasicci and S.M.L. Bodt generated plasmids and expressed and purified the myosin S1 constructs studied. J.R. Moore helped to formulate the 7G-S1 sequence expressed. The Moore laboratory prepared the actin and tropomyosin studied. M.H. Doran carried out the cryo-EM and 3-D reconstruction with supervision from E. Bullitt. M.J. Rynkiewicz and E. Pavadai performed the molecular dynamics simulations and calculated interaction energies. All authors participated in interpreting the data. M.H. Doran, W. Lehman, and M.J.Rynkiewicz prepared figures and tables. W. Lehman wrote the manuscript.

Baldo
,
A.P.
,
J.C.
Tardiff
, and
S.D.
Schwartz
.
2021
.
A proposed mechanism for the initial myosin binding event on the cardiac thin filament: A metadynamics study
.
J. Phys. Chem. Lett.
12
:
3509
3513
.
Behrmann
,
E.
,
M.
Müller
,
P.A.
Penczek
,
H.G.
Mannherz
,
D.J.
Manstein
, and
S.
Raunser
.
2012
.
Structure of the rigor actin-tropomyosin-myosin complex
.
Cell
.
150
:
327
338
.
Brown
,
J.H.
, and
C.
Cohen
.
2005
.
Regulation of muscle contraction by tropomyosin and troponin: How structure illuminates function
.
Adv. Protein Chem.
71
:
121
159
.
Dellefave
,
L.M.
,
P.
Pytel
,
S.
Mewborn
,
B.
Mora
,
D.L.
Guris
,
S.
Fedson
,
D.
Waggoner
,
I.
Moskowitz
, and
E.M.
McNally
.
2009
.
Sarcomere mutations in cardiomyopathy with left ventricular hypertrabeculation
.
Circ. Cardiovasc. Genet.
2
:
442
449
.
Doran
,
M.H.
,
E.
Pavadai
,
M.J.
Rynkiewicz
,
J.
Walklate
,
E.
Bullitt
,
J.R.
Moore
,
M.
Regnier
,
M.A.
Geeves
, and
W.
Lehman
.
2020
.
Cryo-EM and molecular docking shows myosin loop 4 contacts actin and tropomyosin on thin filaments
.
Biophys. J.
119
:
821
830
.
Doran
,
M.H.
,
D.
Rasicci
,
S.M.L.
Bodt
,
E.
Bullitt
,
J.R.
Moore
,
C.M.
Yengo
, and
W.
Lehman
.
2022
.
The structure and function of the human actomyosin complex
.
Biophys. J.
121
:
105a
106a
.
Emsley
,
P.
,
B.
Lohkamp
,
W.G.
Scott
, and
K.
Cowtan
.
2010
.
Features and development of coot
.
Acta Crystallogr. D Biol. Crystallogr.
66
:
486
501
.
Geeves
,
M.A.
2012
.
Thin filament regulation
. In
Comprehensive Biophysics
.
Vol. 4
,
E.H.
Egelman
,
Y.E.
Goldman
,
E.M.
Ostap
, editors.
Oxford: Academic Press
,
San Diego
.
251
267
.
Gordon
,
A.M.
,
E.
Homsher
, and
M.
Regnier
.
2000
.
Regulation of contraction in striated muscle
.
Physiol. Rev.
80
:
853
924
.
He
,
S.
, and
S.H.W.
Scheres
.
2017
.
Helical reconstruction in RELION
.
J. Struct. Biol.
198
:
163
176
.
Hitchcock-DeGregori
,
S.E.
2008
.
Tropomyosin: Function follows structure
.
Adv. Exp. Med. Biol.
644
:
60
72
.
Holmes
,
K.C.
, and
W.
Lehman
.
2008
.
Gestalt-binding of tropomyosin to actin filaments
.
J. Muscle Res. Cell Motil.
29
:
213
219
.
Huang
,
J.
,
S.
Rauscher
,
G.
Nawrocki
,
T.
Ran
,
M.
Feig
,
B.L.
de Groot
,
H.
Grubmüller
, and
A.D.
MacKerell
Jr
.
2017
.
CHARMM36m: An improved force field for folded and intrinsically disordered proteins
.
Nat. Methods
.
14
:
71
73
.
Humphrey
,
W.
,
A.
Dalke
, and
K.
Schulten
.
1996
.
VMD: visual molecular dynamics
.
J. Mol. Graph.
14
:
33
38
.
Kiani
,
F.A.
,
W.
Lehman
,
S.
Fischer
, and
M.J.
Rynkiewicz
.
2019
.
Spontaneous transitions of actin-bound tropomyosin toward blocked and closed states
.
J. Gen. Physiol.
151
:
4
8
.
Lehman
,
W.
2016
.
Thin filament structure and the steric blocking model
. In
Comprehensive Physiology
.
John Wiley & Sons, Inc.
,
Hoboken, NJ
.
1043
1069
.
Lehman
,
W.
2017
.
Switching muscles on and off in steps: The McKillop-Geeves three-state model of muscle regulation
.
Biophys. J.
112
:
2459
2466
.
Lehman
,
W.
,
E.
Pavadai
, and
M.J.
Rynkiewicz
.
2021
.
C-terminal troponin-I residues trap tropomyosin in the muscle thin filament blocked-state
.
Biochem. Biophys. Res. Commun.
551
:
27
32
.
Lehman
,
W.
,
V.
Hatch
,
V.
Korman
,
M.
Rosol
,
L.
Thomas
,
R.
Maytum
,
M.A.
Geeves
,
J.E.
Van Eyk
,
L.S.
Tobacman
, and
R.
Craig
.
2000
.
Tropomyosin and actin isoforms modulate the localization of tropomyosin strands on actin filaments
.
J. Mol. Biol.
302
:
593
606
.
Lehman
,
W.
,
M.J.
Rynkiewicz
, and
J.R.
Moore
.
2020
.
A new twist on tropomyosin binding to actin filaments: Perspectives on thin filament function, assembly and biomechanics
.
J. Muscle Res. Cell Motil.
41
:
23
38
.
Lehman
,
W.
,
R.
Craig
, and
P.
Vibert
.
1994
.
Ca2+-induced tropomyosin movement in Limulus thin filaments revealed by three-dimensional reconstruction
.
Nature
.
368
:
65
67
.
Li
,
X.E.
,
L.S.
Tobacman
,
J.Y.
Mun
,
R.
Craig
,
S.
Fischer
, and
W.
Lehman
.
2011
.
Tropomyosin position on F-actin revealed by EM reconstruction and computational chemistry
.
Biophys. J.
100
:
1005
1013
.
Liebschner
,
D.
,
P.V.
Afonine
,
M.L.
Baker
,
G.
Bunkóczi
,
V.B.
Chen
,
T.I.
Croll
,
B.
Hintze
,
L.W.
Hung
,
S.
Jain
,
A.J.
McCoy
, et al
.
2019
.
Macromolecular structure determination using x-rays, neutrons and electrons: Recent developments in Phenix
.
Acta Crystallogr. D Struct. Biol.
75
:
861
877
.
Lorenz
,
M.
,
K.J.V.
Poole
,
D.
Popp
,
G.
Rosenbaum
, and
K.C.
Holmes
.
1995
.
An atomic model of the unregulated thin filament obtained by x-ray fiber diffraction on oriented actin-tropomyosin gels
.
J. Mol. Biol.
246
:
108
119
.
Maytum
,
R.
,
V.
Hatch
,
M.
Konrad
,
W.
Lehman
, and
M.A.
Geeves
.
2008
.
Ultra short yeast tropomyosins show novel myosin regulation
.
J. Biol. Chem.
283
:
1902
1910
.
McKillop
,
D.F.
, and
M.A.
Geeves
.
1993
.
Regulation of the interaction between actin and myosin subfragment 1: Evidence for three states of the thin filament
.
Biophys. J.
65
:
693
701
.
Noble
,
A.J.
,
H.
Wei
,
V.P.
Dandey
,
Z.
Zhang
,
Y.Z.
Tan
,
C.S.
Potter
, and
B.
Carragher
.
2018
.
Reducing effects of particle adsorption to the air-water interface in cryo-EM
.
Nat. Methods
.
15
:
793
795
.
Orzechowski
,
M.
,
X.E.
Li
,
S.
Fischer
, and
W.
Lehman
.
2014a
.
An atomic model of the tropomyosin cable on F-actin
.
Biophys. J.
107
:
694
699
.
Orzechowski
,
M.
,
S.
Fischer
,
J.R.
Moore
,
W.
Lehman
, and
G.P.
Farman
.
2014b
.
Energy landscapes reveal the myopathic effects of tropomyosin mutations
.
Arch. Biochem. Biophys.
564
:
89
99
.
Orzechowski
,
M.
,
J.R.
Moore
,
S.
Fischer
, and
W.
Lehman
.
2014c
.
Tropomyosin movement on F-actin during muscle activation explained by energy landscapes
.
Arch. Biochem. Biophys.
545
:
63
68
.
Parker
,
F.
, and
M.
Peckham
.
2020
.
Disease mutations in striated muscle myosins
.
Biophys. Rev.
12
:
887
894
.
Pavadai
,
E.
,
W.
Lehman
, and
M.J.
Rynkiewicz
.
2020
.
Protein-protein docking reveals dynamic interactions of tropomyosin on actin filaments
.
Biophys. J.
119
:
75
86
.
Perry
,
S.V.
2003
.
What is the role of tropomyosin in the regulation of muscle contraction?
J. Muscle Res. Cell Motil.
24
:
593
596
.
Pettersen
,
E.F.
,
T.D.
Goddard
,
C.C.
Huang
,
G.S.
Couch
,
D.M.
Greenblatt
,
E.C.
Meng
, and
T.E.
Ferrin
.
2004
.
UCSF Chimera–a visualization system for exploratory research and analysis
.
J. Comput. Chem.
25
:
1605
1612
.
Phillips
,
J.C.
,
R.
Braun
,
W.
Wang
,
J.
Gumbart
,
E.
Tajkhorshid
,
E.
Villa
,
C.
Chipot
,
R.D.
Skeel
,
L.
Kalé
, and
K.
Schulten
.
2005
.
Scalable molecular dynamics with NAMD
.
J. Comput. Chem.
26
:
1781
1802
.
Phillips
,
J.C.
,
D.J.
Hardy
,
J.D.C.
Maia
,
J.E.
Stone
,
J.V.
Ribeiro
,
R.C.
Bernardi
,
R.
Buch
,
G.
Fiorin
,
J.
Hénin
,
W.
Jiang
, et al
.
2020
.
Scalable molecular dynamics on CPU and GPU architectures with NAMD
.
J. Chem. Phys.
153
:
044130
.
Poole
,
K.J.V.
,
M.
Lorenz
,
G.
Evans
,
G.
Rosenbaum
,
A.
Pirani
,
R.
Craig
,
L.S.
Tobacman
,
W.
Lehman
, and
K.C.
Holmes
.
2006
.
A comparison of muscle thin filament models obtained from electron microscopy reconstructions and low-angle x-ray fibre diagrams from non-overlap muscle
.
J. Struct. Biol.
155
:
273
284
.
Potter
,
J.D.
, and
J.
Gergely
.
1974
.
Troponin, tropomyosin, and actin interactions in the Ca2+ regulation of muscle contraction
.
Biochemistry
.
13
:
2697
2703
.
Powers
,
J.D.
,
S.A.
Malingen
,
M.
Regnier
, and
T.L.
Daniel
.
2021
.
The sliding filament theory since Andrew Huxley: Multiscale and multidisciplinary muscle research
.
Annu. Rev. Biophys.
50
:
373
400
.
Risi
,
C.
,
B.
Belknap
,
E.
Forgacs-Lonart
,
S.P.
Harris
,
G.F.
Schröder
,
H.D.
White
, and
V.E.
Galkin
.
2018
.
N-terminal domains of cardiac myosin binding protein C cooperatively activate the thin filament
.
Structure
.
26
:
1604
1611.e4
.
Risi
,
C.
,
J.
Eisner
,
B.
Belknap
,
D.H.
Heeley
,
H.D.
White
,
G.F.
Schröder
, and
V.E.
Galkin
.
2017
.
Ca2+-induced movement of tropomyosin on native cardiac thin filaments revealed by cryoelectron microscopy
.
Proc. Natl. Acad. Sci. USA
.
114
:
6782
6787
.
Risi
,
C.
,
L.U.
Schäfer
,
B.
Belknap
,
I.
Pepper
,
H.D.
White
,
G.F.
Schröder
, and
V.E.
Galkin
.
2021
.
High-resolution cryo-EM structure of the cardiac actomyosin complex
.
Structure
.
29
:
50
60.e4
.
Rohou
,
A.
, and
N.
Grigorieff
.
2015
.
CTFFIND4: Fast and accurate defocus estimation from electron micrographs
.
J. Struct. Biol.
192
:
216
221
.
Rynkiewicz
,
M.J.
,
V.
Schott
,
M.
Orzechowski
,
W.
Lehman
, and
S.
Fischer
.
2015
.
Electrostatic interaction map reveals a new binding position for tropomyosin on F-actin
.
J. Muscle Res. Cell Motil.
36
:
525
533
.
Scheres
,
S.H.
2012
.
RELION: Implementation of a bayesian approach to cryo-EM structure determination
.
J. Struct. Biol.
180
:
519
530
.
Scheres
,
S.H.
2016
.
Processing of structurally heterogeneous cryo-EM data in RELION
.
Meth. Enzym.
579
:
125
157
.
Strelkov
,
S.V.
, and
P.
Burkhard
.
2002
.
Analysis of α-helical coiled coils with the program TWISTER reveals a structural mechanism for stutter compensation
.
J. Struct. Biol.
137
:
54
64
.
Suloway
,
C.
,
J.
Pulokas
,
D.
Fellmann
,
A.
Cheng
,
F.
Guerra
,
J.
Quispe
,
S.
Stagg
,
C.S.
Potter
, and
B.
Carragher
.
2005
.
Automated molecular microscopy: The new Leginon system
.
J. Struct. Biol.
151
:
41
60
.
Swenson
,
A.M.
,
W.
Tang
,
C.A.
Blair
,
C.M.
Fetrow
,
W.C.
Unrath
,
M.J.
Previs
,
K.S.
Campbell
, and
C.M.
Yengo
.
2017
.
Omecamtiv Mecarbil enhances the duty ratio of human β-cardiac myosin resulting in increased calcium sensitivity and slowed force development in cardiac muscle
.
J. Biol. Chem.
292
:
3768
3778
.
Tang
,
W.
,
J.
Ge
,
W.C.
Unrath
,
R.
Desetty
, and
C.M.
Yengo
.
2021
.
Cardiomyopathy mutations impact the actin-activated power stroke of human cardiac myosin
.
Biophys. J.
120
:
2222
2236
.
Tang
,
W.
,
W.C.
Unrath
,
R.
Desetty
, and
C.M.
Yengo
.
2019
.
Dilated cardiomyopathy mutation in the converter domain of human cardiac myosin alters motor activity and response to omecamtiv mecarbil
.
J. Biol. Chem.
294
:
17314
17325
.
Touma
,
A.M.
,
W.
Tang
,
D.V.
Rasicci
,
D.
Vang
,
A.
Rai
,
S.B.
Previs
,
D.M.
Warshaw
,
C.M.
Yengo
, and
S.
Sivaramakrishnan
.
2022
.
Nanosurfer assay dissects β-cardiac myosin and cardiac myosin-binding protein C interactions
.
Biophys. J.
121
:
2449
2460
.
Trabuco
,
L.G.
,
E.
Villa
,
K.
Mitra
,
J.
Frank
, and
K.
Schulten
.
2008
.
Flexible fitting atomic structures into electron microscopy maps using molecular dynamics
.
Structure
.
16
:
673
683
.
Trabuco
,
L.G.
,
E.
Villa
,
E.
Schreiner
,
C.B.
Harrison
, and
K.
Schulten
.
2009
.
Molecular dynamics flexible fitting: A practical guide to combine cryo-electron microscopy and x-ray crystallography
.
Methods
.
49
:
174
180
.
Trujillo
,
A.S.
,
K.H.
Hsu
,
M.C.
Viswanathan
,
A.
Cammarato
, and
S.I.
Bernstein
.
2022
.
The R369 myosin residue within loop 4 is critical for actin binding and muscle function in Drosophila
.
Int. J. Mol. Sci.
23
:
2533
.
Vibert
,
P.
,
R.
Craig
, and
W.
Lehman
.
1997
.
Steric-model for activation of muscle thin filaments
.
J. Mol. Biol.
266
:
8
14
.
von der Ecken
,
J.
,
S.M.
Heissler
,
S.
Pathan-Chhatbar
,
D.J.
Manstein
, and
S.
Raunser
.
2016
.
Cryo-EM structure of a human cytoplasmic actomyosin complex at near-atomic resolution
.
Nature
.
534
:
724
728
.
Yamada
,
Y.
,
K.
Namba
, and
T.
Fujii
.
2020
.
Cardiac muscle thin filament structures reveal calcium regulatory mechanism
.
Nat. Commun.
11
:
153
.
Zheng
,
S.Q.
,
E.
Palovcak
,
J.-P.
Armache
,
K.A.
Verba
,
Y.
Cheng
, and
D.A.
Agard
.
2017
.
MotionCor2: Anisotropic correction of beam-induced motion for improved cryo-electron microscopy
.
Nat. Methods
.
14
:
331
332
.

Author notes

*

M.H. Doran and M.J. Rynkiewicz contributed equally to this paper.

M.H. Doran’s present address is Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, MA.

This article is distributed under the terms of an Attribution–Noncommercial–Share Alike–No Mirror Sites license for the first six months after the publication date (see http://www.rupress.org/terms/). After six months it is available under a Creative Commons License (Attribution–Noncommercial–Share Alike 4.0 International license, as described at https://creativecommons.org/licenses/by-nc-sa/4.0/).