Calcium has a direct effect on thick filament activation in porcine myocardium

Cardiac muscle contraction is highly regulated to fulfill a wide variety of the body’s specific circulatory needs. Contractile activation had long been regarded as a calcium (Ca²⁺)-dependent, thin filament-based mechanism, where Ca²⁺ binding to the troponin complex allows mobilization of tropomyosin. This structural displacement of tropomyosin exposes myosin binding sites on F-actin to facilitate actomyosin binding (Tobacman, 1996; Gordon et al., 2000). It is now realized, however, that thick filament-based activation is also required during muscle contraction (Irving, 2017), but the mechanisms of this activation are not well understood.

There is strong evidence that in relaxed muscle, a large portion of the myosin heads reside in a quasi-helically ordered state close to the thick filament backbone while the myosin heads during contraction are more associated with actin (Huxley and Brown, 1967). One mechanism of activating resting myosin heads in the thick filament to permit contraction was introduced by Linari et al. (2015), who proposed a strain-dependent thick filament activation model (“mechano-sensing”) derived from experiments on frog skeletal muscle (Linari et al., 2015) and later expanded to rodent myocardium (Reconditi et al., 2017; Brunello et al., 2020). In this model, thin filaments are activated by Ca²⁺ influx, and a small number of “constitutively ON” myosin heads can form crossbridges and generate a small amount of force. In this way, the thick filament acts as a mechanical sensor where the strain generated in the thick filament backbone by this amount of force can release myosin heads from an inactive quasi-helically ordered OFF state to an active disordered ON state thereby regulating the number of myosin heads competent to bind actin and generate force (Linari et al., 2015; Caremani et al., 2019b). In addition to the mechanosensing mechanism, multiple studies point to a titin-based thick filament strain mechanism in myofilament length-dependent activation (Irving et al., 2011; Ait-Mou et al., 2016; Ma et al., 2021; Hessel et al., 2022). It has been shown, however, that thick filament strain cannot be the only trigger for OFF to ON structural transitions that result in the release of myosin heads from the thick filament backbone (Ma et al., 2018b, 2023b; Park-Holohan et al., 2021; Brunello and Fusi, 2023).

We previously showed that myosin heads on the thick filaments undergo OFF to ON, order to disorder, and structural transitions in the thick filaments of permeabilized porcine myocardium in response to Ca²⁺ and in the absence of active force (Ma et al., 2022b). Ca^2+_mediated active contraction was abolished by incubating the myocardium with a small molecule thin filament inhibitor (MYK7660) that allowed Ca²⁺ binding to troponin C but inhibited force by increasing the release rate of Ca²⁺ to prevent downstream movement of tropomyosin (Ma et al., 2022b). The results were interpreted as direct Ca²⁺ activation of the thick filament. An alternative mechanism has subsequently been proposed where binding of Ca²⁺ to troponin initiates a Ca²⁺-dependent “crosstalk” between thin and thick filaments via interfilament connections (Brunello and Fusi, 2023; Caremani et al., 2023; Short, 2023). For example, cardiac myosin binding protein-C (cMyBP-C), whose N terminus can bind to the thin filaments, could act as a conduit for any Ca²⁺-mediated signals to the thick filament. Given that MYK7660 allows the binding of Ca²⁺ to troponin, such mechanisms cannot be excluded from the previous studies. To critically test the “crosstalk” hypothesis, we exchanged native troponin in permeabilized porcine myocardium for troponin containing the cTnC D65A mutation. This mutation inhibits troponin activation caused by Ca²⁺ binding at the N-terminal “trigger” site of cTnC (Gillis et al., 2007; Little et al., 2012). We observed that the changes in X-ray diffraction signatures indicative of OFF to ON transitions in response to increased Ca²⁺ concentration are still present in porcine myocardium with the D65A cTnC. These results are consistent with a direct effect of Ca²⁺ in activating the thick filament rather than an indirect effect due to Ca²⁺-mediated crosstalk between the thick and thin filaments.

Methods for construction and expression of wild type (WT) rat cTnC, cTnI, and cTnT in pET24a vectors were developed previously in our laboratory (Potter, 1982; Köhler et al., 2003). The cTnC (D65A) mutant was constructed from the rat WT cTnC plasmid using a Quikchange site-directed mutagenesis kit from Stratagene and confirmed by DNA sequence analysis. The plasmids for cTnC variants were then transformed into Escherichia coli BL21 cells. Proteins were expressed and purified following the protocols developed as described previously (Dong et al., 1996).

Methods for cTnC labeling and fluorescence measurements were previously described (Wang et al., 2012, 2013). Briefly, purified WT and D65A cTnC were labelled with a fluorescent probe {N-[2-(iodoacetoxy)ethyl]-N-methylamino-7-nitrobenz-2-oxa-1,3-diazole (IANBD, M_w = 406.14, Cat No: I-9; Life technologies)}. at position C84 in the dark overnight at 4°C. IANBD enables the quantification of Ca²⁺–cTnC (K_Ca) binding affinities. The labeling efficiency was obtained by comparing the IANBD fluorophore versus protein concentration ratio. The IANBD fluorophore concentration was computed using the maximal absorbance of labeled protein at a wavelength of 481 nm and dividing by IANBD’s extinction coefficient (21,000 M⁻¹ cm⁻¹). The steady-state fluorescence measurements were recorded by a luminescence spectrometer at 530 nm (LS50B; PerkinElmer) at room temperature with 150 mM KCl, 20 mM MOPS, 3 mM MgCl₂, 2 mM EGTA, and 1 mM DTT (pH 7.0). The emission signal was then recorded during the titration of microliter amounts of Ca²⁺ into 2 ml of WT or D65a cTC (0.3 µM). The concentration of free Ca²⁺ was computed using Maxchelator (Bers et al., 2010). Titration curve data were fitted with the Hill equation.

Reconstitution of the cardiac troponin complex was performed as described previously (Potter, 1982; Korte et al., 2012). Briefly, the cTn subunits cTnI and cTnT were first dialyzed independently against 6.0 M urea, 25 mM TRIS, and 1.0 mM EDTA at pH 8.0. After dialysis, cTnC (WT or D65A mutant) was mixed with both the cTnI and cTnT proteins at a molar ratio of 1:1:1 in the presence of 1.5 mM CaCl₂. The protein solutions were then incubated at room temperature for 30 min. After incubation, solutions were dialyzed through a series of five steps to lower the ionic strength. All solutions contained 20 mM MOPS, 3.0 mM MgCl₂, and 2.0 mM EGTA at pH 7.0. The progress was from 2.0 M urea, 0.75 M KCl to 0 and 0.15 M, respectively, with an additional 1.0 mM DTT in the final solution. All dialyses were done at 4°C in the dark without stirring. Proteins that precipitated during the dialysis with decreasing KCl concentration were removed by centrifugation. The final troponin complex concentration (mg/ml) was determined using a spectrofluorometer (Beckman) by dividing the observed OD value collected at 280 nm by the extinction coefficient (0.45).

Porcine myocardium from three hearts for the X-ray diffraction experiments was permeabilized as described previously (Ma et al., 2022a, 2023a). Briefly, previously frozen muscles were permeabilized in a skinning solution (pCa 8.0 solution [91 mM K⁺-propionate, 3.5 mM MgCl₂, 0.16 mM CaCl₂, 7 mM EGTA, 2.5 mM Na₂ATP, 15 mM creatine phosphate, 20 mM imidazole, and 1% protease inhibitor cocktail at pH 7.0] plus 15 mM 2,3-butanedione monoxime, 1% Triton-X100) overnight at 4°C. The fiber bundles were further dissected into preparations with a length of 5 mm with a diameter of ∼200 µm prior to the attachment of aluminum T-clips to both ends in pCa 8 solution. The preparations were randomly separated into the WT and D65A groups. The number of technical replicates was 20 for WT and 40 for the D65A exchanged myocardium. Preparations were then washed three times in pCa 8.0 solutions containing 1 mg/ml BSA. After washing, preparations were placed in pCa 8.0 solution with 3% dextran on ice.

Passive exchanges of whole WT or D65A cTn complex into dissected porcine strips were done as described previously (Korte et al., 2012). Briefly, t-clipped porcine strips were incubated for 16 h in a 4°C fridge on a mechanical rocker (Beckman) with troponin exchange solution (∼1 mg/ml recombinant WT or D65A cTn, 4 mM ATP, 1 mg/ml DTT, and 1% protease inhibitor cocktail). Exchange solutions were refreshed at 8 h to increase exchange efficiency. Preparations were then washed three times in pCa 8.0 solutions containing 1 mg/ml BSA. After washing, preparations were placed in pCa 8.0 solution with 3% dextran on ice. Samples were used within 12 h after washing. Note that the exchanges of WT and D65A cTn complexes were done under identical conditions so that the results can be compared.

X-ray diffraction experiments were performed at the BioCAT beamline 18ID at the Advanced Photon Source, Argonne National Laboratory (Fischetti et al., 2004). The X-ray beam energy was set to 12 keV (0.1033 nm wavelength) at an incident flux of ∼5 × 10¹² photons per second. The specimen-to-detector distance was ∼3 m. The muscles, bound on each end by aluminum foil T-clips, were attached to a hook on a force transducer (Model 402B; Aurora Scientific Inc.) and a static hook inside a customized chamber. The solution was maintained between 28°C and 30°C with a heat exchanger attached to the bottom of the chamber. The muscles were stretched to a sarcomere length of 2.3 µm as measured by light diffraction with a helium-neon laser (633 nm). X-ray patterns were collected at pCa 8.0, pCa 5.8, and pCa 4.5 in the presence of 3% dextran using a MarCCD 165 detector (Rayonix Inc.) with a 1-s exposure time. The muscle samples were oscillated along their horizontal axes at a velocity of 1–2 mm/s to minimize radiation damage. The irradiated areas are moved vertically after each exposure to avoid overlapping X-ray exposures. The force-pCa data were monitored using a Dynamic Muscle Control system (v5.5) from Aurora Scientific Inc., and absolute forces (mN) without correcting for the cross-sectional area are reported in this manuscript.

The data were analyzed using the MuscleX software package developed at BioCAT (Jiratrakanvong et al., 2024). Three to four patterns were collected under each condition, and spacings and intensities of X-ray reflections extracted from these patterns were averaged. The equatorial reflections were measured by the “Equator” routine in MuscleX to calculate the equatorial intensity ratio, I_1,1/I_1,0, and interfilament lattice spacing, d_1,0, as described previously (Ma et al., 2018a). The intensities and spacings of meridional and layer line reflections were measured using the “Projection Traces” routine in MuscleX after the patterns were quadrant-folded and the diffuse scattering subtracted to improve the signal-to-noise ratio as described previously (Ma et al., 2018a, 2020). There were no significant changes in the radial width of the M3 reflection during activation, thus no correction for width was applied to the measured intensities. To compare the intensities under different conditions, the measured intensities of X-ray reflections were normalized to the sixth-order actin-based layer line intensities as described previously (Ma et al., 2022b). Statistical analyses were performed using GraphPad Prism 10 (GraphPad Software). The results were given as mean ± SEM. Statistical comparisons under each condition were done using nested t tests. Symbols on figures represent: *: P < 0.05, **: P < 0.01, ***: P < 0.001, ****: P < 0.0001.

The steady-state fluorescence measurements performed on purified and IANBD-labeled cTnC proteins provide significant insights into the structural change that occurs to cTnC during Ca²⁺ binding. In Ca²⁺-free environments, the IANBD probe, located at site C84, is buried within the protein structure. Upon the binding of Ca²⁺, activation of cTnC exposes the IANBD probe, thus increasing the fluorescent readout captured by the spectrofluorometer. These results are observed in the WT cTnC experiments, as the fluorescent change increased by ∼65% from baseline (no Ca²⁺) measurements. Interestingly, measurements performed with purified D65A cTnC showed a fluorescent decrease of ∼20% (Fig. 1 A). These measurements would suggest Ca²⁺ may interact in some way with cTnC but not in a manner that promotes the conformational change seen when Ca²⁺ binds in site II and results in exposure of the hydrophobic patch, essential for thin filament activation. These experiments establish that D65A cTnC is not competent to activate the thin filaments in the presence of Ca²⁺.

Inhibition of thin filament activation by D65A cTnC was observed when we exchanged native porcine cTn complex with recombinant rat cTn containing the WT or D65A cTnC mutation. Exchanged tissues eliminated >95% of the active force with D65A cTnC exchange compared to preparations with WT cTnC at both pCa 5.8 (95.6%, 0.316 ± 0.02 mN in WT group versus 0.014 ± 0.01 mN in D65A group, P = 0.02) and pCa 4.5 (96.3%, 1.20 ± 0.13 mN in WT group versus 0.041 ± 0.02 mN in D65A group, P < 0.0001) (Fig. 1 B).

We then examined the structural transitions of the WT or D65A exchanged permeabilized porcine cardiac myocardium tissues in response to different Ca²⁺ concentrations using small-angle X-ray diffraction (Fig. 2 A). At high Ca²⁺ concentration (pCa 4.5), the intensities of all the myosin-based reflections (M3, M6, MLLs in Fig. 2 A) become weaker compared with low Ca²⁺ concentration (pCa 8.0) in tissues exchanged with either WT or D65A cTn. The interfilament lattice spacing, d_1,0, decreased as the Ca²⁺ concentration increased in both WT and D65A groups with no significant difference between the two groups at each Ca²⁺ concentration (Fig. 2 B). The equatorial intensity ratio, I_1,1/I_1,0, is an indicator of the proximity of myosin heads to actin in relaxed muscle and is closely correlated to the number of force-producing crossbridges in activated muscle (Haselgrove and Huxley, 1973; Matsubara, 1980; Ma et al., 2018a, 2020). In the low Ca²⁺ resting state (pCa 8.0), I_1,1/I_1,0 was slightly but significantly lower in D65A group (0.32 ± 0.008) compared with WT group (0.37 ± 0.01, P = 0.04). I_1,1/I_1,0 increased to (0.51 ± 0.02) and (0.64 ± 0.03) pCa 5.8 and pCa 4.5, respectively, from the WT group. In the D65A group, where Ca²⁺ mediated troponin activation is mostly abolished, I_1,1/I_1,0 increased to (0.40 ± 0.012) and (0.45 ± 0.014) pCa 5.8 and pCa 4.5, respectively. The radius of the center of mass of the cross-bridges (R_m), which directly measures the proximity of the helically ordered myosin heads on the surface of the thick filament backbone (Ait-Mou et al., 2016; Ma et al., 2018a), was significantly larger in WT (16.06 ± 0.22) group than D65A group (15.30 ± 0.14, P = 0.0039) at pCa8.

Under resting conditions (pCa 8.0), most of the myosin heads are quasi-helically ordered on the surface of the thick filament, producing the myosin-based layer lines (MLLs). Myosin in resting muscle also has a ∼14.3 nm axial periodicity along the thick filament, which gives rise to the third-order myosin-based meridional reflection (M3). The intensity of the first-order myosin-based layer line (I_MLL1) and the M3 reflection (I_M3) progressively decreased when Ca²⁺ concentrations increased, and there were no significant differences in I_M3 and I_MLL1 between the WT group where active actin-binding crossbridges were present and the D65A group where active contraction was mostly abolished (Fig. 3, A and B). The spacing of the M3 reflection (S_M3) and M6 reflection (S_M6) reports the axial periodicity of quasi-helically arranged myosin heads and structures within the thick filament backbone, respectively (Reconditi, 2006). In the WT group, S_M3 was 14.369 ± 0.003 nm and 14.370 ± 0.003 nm at pCa 8.0 and pCa 5.8, respectively, and it increased to 14.404 ± 0.005 nm at pCa 4.5. In the D65A group, S_M3 decreased from 14.371 ± 0.003 nm at pCa 8.0–14.361 ± 0.003 nm at pCa 5.8 and increased to 14.369 ± 0.003 nm at pCa 4.5 (Fig. 3 C). Increased S_M6 is associated with fewer myosin heads in the quasi-helically ordered state (Ma et al., 2023b) and vice versa (Ma et al., 2021), so changes in S_M6 have been proposed as another indicator of OFF to ON transitions (Linari et al., 2015). S_M6 was significantly larger in WT (7.215 ± 0.001 nm) than in D65A (7.208 ± 0.001, P = 0.02) at pCa 8.0. This difference persists as Ca²⁺ concentration increases (Fig. 3 D). S_M6 increased to 7.230 ± 0.002 nm in WT and 7.218 ± 0.001 in D65A (P = 0.09) at pCa 5.8 and future increased to 7.240 ± 0.001 nm in WT and 7.222 ± 0.001 in D65A (P = 0.09) at pCa 4.5. The larger S_M6 changes in the WT group (0.2% and 0.36% at pCa 5.8 and pCa 4.5, respectively) are expected because it involves both the release of the myosin heads from the OFF state and the increase in strain on the thick filament by active contraction (Reconditi et al., 2019). The increase in S_M6 for the D65A (0.14% and 0.20% at pCa 5.8 and pCa 4.5, respectively), however, is solely from the OFF to ON transition of the thick filament in response to increasing Ca²⁺ (Fig. 3 D).

There is a growing appreciation that Ca²⁺, in addition to regulating the thin filaments through the troponin–tropomyosin system, may also regulate the movement of myosin heads relative to the thick filaments. However, many details regarding the mechanisms involved in thick filament activation are not known. A recent study decoupled the Ca²⁺-mediated force production from Ca²⁺-mediated regulation of the thick filaments by eliminating Ca²⁺-mediated tropomyosin movement via a small-molecule thin filament inhibitor (MYK 7660; Bristol Meyers Squibb). Structural changes in thick filaments in response to Ca²⁺ in the presence of the inhibitor were interpreted as an indication that the thick filaments in porcine myocardium could sense and respond directly to Ca²⁺ through a mechanism that is not yet clear (Ma et al., 2022b). Additionally, Caremani et al. (2023) showed that increasing Ca²⁺ can lead to a loss of ordering of myosin heads, and this loss of ordering is more sensitive to Ca²⁺ than to force in rabbit skeletal muscle. In contrast, our previous studies of porcine cardiac muscle showed that the pCa₅₀ for the loss of ordering, as indicated by loss of layer line intensity, was very similar to that for force development. It has been proposed that the OFF to ON transitions observed in the previous studies (Ma et al., 2022b; Caremani et al., 2023) could be explained by an alternative mechanism where binding of Ca²⁺ to troponin initiates a Ca²⁺ dependent “crosstalk” between the thin filaments and the thick filaments via interfilament connections (Brunello and Fusi, 2023; Caremani et al., 2023; Short, 2023). The aim of this current study was to investigate whether Ca²⁺-mediated thick filament activation is a direct effect of Ca²⁺ on thick filaments or an indirect effect due to Ca²⁺-mediated crosstalk via connecting structures between the thick and thin filaments.

In the D65A cTn group, small amounts of active force (<5% of WT) were detected at both submaximal and maximal Ca²⁺ (pCa 5.8 and pCa 4.5, respectively). This is most likely due to incomplete troponin exchange. We have established in the previous study, where active force was totally abolished, that the Ca²⁺-dependent OFF to ON transition in porcine myocardium is independent of force (Ma et al., 2022b). The OFF to ON structural transitions in porcine myocardium, in the absence of active force, show the exact same response to Ca²⁺ as for active force indicating that the structural transitions can be precisely modulated in a graded manner by Ca²⁺ concentration while the OFF to ON structural transitions in fast skeletal muscle is a quasi-stepwise process (Gong et al., 2022; Zhao et al., 2024). If these graded OFF to ON transitions in porcine myocardium are due to a putative Ca²⁺-troponin initiated crosstalk pathway, the degree of myosin head activation would need to be correlated with the number of troponins with bound Ca²⁺ to result in the graded response we observed in porcine cardiac muscle here. In the current study, the <5% residual force roughly indicates that <5% of native troponins are unexchanged and it seems highly unlikely that a communication pathway involving only <5% of troponins with bound Ca²⁺ could result in 100% of the heads undergoing OFF to ON transitions as observed at maximum Ca²⁺ concentration. We believe, therefore, that the residual (<5%) active force developed in the D65A group as compared with the WT group will not affect the main conclusions drawn in this study. At pCa 8.0, I_1,1/I_1,0, R_m, and S_M6 were significantly lower in the D65A group indicating that myosin heads adopt more inactive OFF states and are more associated with the thick filament backbone in the D65A group. Although we cannot exclude the possibility that exchanging WT cTn and D65A cTn could cause myofilament structural changes that are independent of Ca²⁺, these results could be explained if a small amount of Ca²⁺ binds to a few cTnC N-terminal binding sites at pCa 8.0 in the WT group. Although this should be insufficient to result in active force production, it could potentially be sufficient to maintain a small number of myosin heads in the ON state where they can interact with actin, perhaps aided through electrostatic forces (Ma et al., 2023b). In the D65A group, however, any residual actomyosin interactions should be eliminated as the number of open myosin binding sites on the thin filaments will be greatly reduced resulting in more myosin heads remaining in the OFF state. The increases of I_1,1/I_1,0 and S_M6, as Ca²⁺ concentrations increased, were larger in the WT group than in the D65A group. The increases of I_1,1/I_1,0 seen in the WT group can be explained by both the release of OFF states myosin heads and the binding of myosin heads to actin. The larger S_M6 changes in the control group can be attributed to both the release of OFF states myosin heads and the increase in strain on the thick filament by force-producing crossbridges (Reconditi et al., 2019; Ma et al., 2023b). Here, the majority of the troponin units along the thin filament are deactivated by D65A troponin exchange while the thick filament OFF/ON response was unaffected, as indicated by the indistinguishable decreases of I_MLL1 and I_M3 in both the WT and D65A groups, as Ca²⁺ concentration increases. These results indicate that a direct effect of Ca²⁺ on the myosin in the thick filaments, in the absence of significant numbers of strong force-bearing crossbridges, can account for the majority of the OFF-to-ON transitions of myosin heads consistent with the findings from the previous study (Ma et al., 2022b) and in contradiction to the crosstalk hypothesis. Additionally, the lattice spacing decreases by ∼0.5 nm between pCa 8 and pCa in both WT and D65A exchanged myocardium. A decrease in spacing cannot be explained by a decrease in sarcomere length due to internal shortening. Lattice spacing in skinned muscle depends on a balance of several opposing forces (Millman, 1998) so it is very difficult to devise unique explanations for spacing changes in resting skinned muscle. Without detailed knowledge of how radial forces might change with length in resting muscle, it is premature to attempt to propose a detailed mechanism for these changes in lattice spacing.

It has shown that in skeletal muscle, S_M3 increased >1% during maximum activation (Reconditi, 2006; Caremani et al., 2023), which has been widely interpreted as a structural indicator of myosin heads adopting an ON state (Irving, 2017). Here, we show that S_M3 in porcine myocardium increases significantly (0.21%) at pCa 4.5 in the WT group while in the D65A group, S_M3 shows a significant decrease at pCa 5.8 with no increase at pCa 4.5. These data suggest that S_M3 in porcine myocardium does not increase when myosin heads transition from an OFF to ON state in the absence of significant active force. This implies that the increase in S_M3 during contraction seen in other muscle types is not a reliable indicator of myosin transitioning to the ON state in porcine myocardium. We speculate that when an increase of S_M3 is observed, at least in porcine cardiac muscle, it is the result of a significant fraction of the myosin heads adopting a longer periodicity upon binding to actin (Zhao et al., 2024). This hypothesis is supported by the increase of S_M3 in porcine cardiac muscle only at pCa 4.5 in the WT group where a significant number of force-producing crossbridges will be bound to actin. Increases in the thick filament backbone spacing, S_M6, have also been used as an indicator of structural OFF to ON transitions, but a more detailed interpretation has been difficult. While S_M6 has been shown to be responsive to strain in the thick filaments, albeit in a complex and non-linear way (Ma et al., 2018b), we and other investigators, however, have seen multiple situations where S_M6 spacing can change without necessarily changing the tension on the thick filament (Caremani et al., 2019a, 2023; Ma et al., 2022b, 2023b, 2024a; Jani et al., 2024), indicating that S_M6 can also reflect strain independent structural changes. Without differentiating between these two mechanisms, however, an increase in M6 has consistently been associated with an OFF-to-ON transition in all studies to date and is consistent with the other data presented here.

One question raised from the current and previous study (Ma et al., 2022b) is whether this phenomenon observed under steady state is rapid enough to be meaningful during normal heartbeats in vivo. A preliminary study (Ma et al., 2024b) showed that myosin heads leave their OFF state when filament stress is eliminated by zero load shortening during a normal twitch in intact pig trabeculae. We show here that thick filaments are still sensitive to increasing Ca²⁺ after greatly reducing the numbers of troponin competent to bind Ca²⁺ and participate in any Ca²⁺-dependent interfilament crosstalk pathways. Taken together, it seems that at least some direct Ca²⁺-dependent thick filament activation operates on a beat-by-beat basis in the heart. Another question that needs to be resolved in the future is where are the Ca²⁺ binding or sensing sites? At this time, we cannot pinpoint the Ca²⁺ binding or sensing sites on the thick filament that are responsible for this Ca²⁺-mediated thick filament activation pathway. Our previous study (Ma et al., 2022b) showed that Ca²⁺ can destabilize the SRX states of synthetic myosin filament, indicating that this Ca²⁺-dependent transition is an intrinsic property of the myosin filament. We suggested that the F-hand motif of RLC might be the Ca²⁺ binding or sensing site for this Ca²⁺-mediated thick filament activation pathway. Other sarcomeric proteins such as myosin-binding protein C and titin are also known to bind to Ca²⁺ (Labeit et al., 2003). It is possible that binding of Ca²⁺ to either myosin-binding protein C or titin can disrupt their interactions with myosin as indicated by recent high-resolution cryo-EM C-zone thick filament structures in the presence of mavacamten (Dutta et al., 2023). These hypotheses need to be tested in future studies.

Thick filament dysregulation appears to be the basis of many cardiomyopathies (Spudich, 2019; Nag and Trivedi, 2021), so a better understanding of thick filament regulation mechanisms could help us identify therapeutics necessary to mitigate these diseases. In this study, we provide strong corroboration of our previous demonstrations that Ca²⁺ can directly activate the thick filament. In so doing, Ca²⁺ can coordinate the excitation and contraction of cardiomyocytes. Imbalances in Ca²⁺ homeostasis have been shown to lead to cardiac disorders (Luo and Anderson, 2013). For one example, hypertrophic cardiomyopathy (HCM) can be caused by having more myosin heads in the ON state under resting conditions (Anderson et al., 2018; Spudich, 2019; Nag and Trivedi, 2021), and it has also been shown that intracellular diastolic Ca²⁺ concentration may be higher in HCM cells (Coppini et al., 2018). With our new understanding that Ca²⁺ plays a role in thick filament activation, it is worth considering that elevated diastolic Ca²⁺ concentration may result in more myosin heads in the ON state, further exacerbating the thick filament dysregulation in HCM cells. Our findings suggest that the Ca²⁺-mediated thick filament activation pathway should be considered when investigating the pathological bases of sarcomere-based diseases and developing future drug candidates to treat these diseases.

The datasets generated or analyzed during this study are included in this article. The raw data are available from the corresponding author upon reasonable request.

Henk L. Granzier served as editor.

This project is supported by grants 1R01HL171657 (W. Ma) and 5R01HL157169 (F. Moussavi-Harami) from National Heart Lung and Blood Institute, P30 GM138395 from the National Institute of General Medical Sciences (T.C. Irving) and P30 AR074990 from the National Institute of Arthritis and Musculoskeletal and Skin Diseases (M. Regnier) of the National Institutes of Health (NIH). This research used resources from the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02-06CH11357. The content is solely the authors’ responsibility and does not necessarily reflect the official views of the NIH.

Author contributions: S. Mohran: Conceptualization, Data curation, Formal analysis, Investigation, Writing—original draft, T.S. McMillen: Formal analysis, Investigation, C. Mandrycky: Data curation, Investigation, Writing—review & editing, A.-Y. Tu: Resources, K.B. Kooiker: Investigation, W. Qian: Formal analysis, S. Neys: Investigation, B. Osegueda: Investigation, Visualization, F. Moussavi-Harami: Conceptualization, Funding acquisition, Writing—review & editing, T.C. Irving: Funding acquisition, Project administration, Resources, Supervision, Writing—original draft, Writing—review & editing, M. Regnier: Conceptualization, Data curation, Funding acquisition, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing—review & editing, W. Ma: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing—original draft, Writing—review & editing.

The datasets generated or analyzed during this study are included in this article. The raw data are available from the corresponding author upon reasonable request.