All procedures involving animals were performed in accordance with an approved protocol from the Animal Care and Use Committee of the University of Missouri. The Dmd^mdx-4cv mouse model was used for this study, as a model that approximates human DMD (Yucel et al., 2018). For instance, Dmd ^mdx-4Cv mice have 10 times fewer revertant fibers in skeletal muscle cross-sections, which could impact cardiac muscle cells (Danko et al., 1992; Pigozzo et al., 2013) and led to a slightly more severe and less variable cardiac phenotype in Dmd ^mdx-4Cv compared to the classic Dmd ^mdx mice. Mice were between the ages of 4–12 months. Wild-type (WT) littermate mice were used as the control group.

Relaxing solution for permeabilized cardiac myocytes contained: 1 mM DTT, 100 mM KCl, 10 mM Imidazole, 2.0 mM EGTA, 4.0 mM ATP, 1 mM (free, 5 total) MgCl₂. Minimal Ca²⁺ activating solution (pCa 9.0) for experimental protocol contained: 7.00 mM EGTA, 20 mM Imidazole, 5.42 mM MgCl₂, 72.37 mM KCl, 0.016 mM CaCl₂, 14.50 mM PCr, 4.7 mM ATP. Maximal Ca²⁺ activating solution (pCa 4.5) for experimental protocol contained: 7.00 mM EGTA, 20 mM Imidazole, 5.26 mM MgCl₂, 60.25 mM KCl, 7.01 mM CaCl₂, 14.50 mM PCr, 4.81 mM ATP. A range of Ca²⁺ concentrations for experiments was prepared by varying combinations of minimal and maximal Ca²⁺ solutions. Pre-activating solution contained: 0.5 mM EGTA, 20 mM Imidazole, 5.42 mM MgCl₂, 98.18 mM KCl, 0.016 mM CaCl₂, 14.50 mM PCr, and 4.8 mM ATP. The submaximal calcium concentration was determined by selecting the pCa solution that elicited ∼50% of maximal force obtained in pCa 4.5. For mavacamten experiments, 0.5 µM mavacamten was added to all pCa solutions resulting in a reduction of maximal force by ∼50%, in agreement to previous studies (Awinda et al., 2020; Awinda et al., 2021; Sewanan et al., 2021). For experiments in the presence of 0.5 µM mavacamten, the submaximal calcium concentration was determined by selecting the pCa solution with mavacamten that elicited ∼50% maximal force obtained in pCa 4.5 containing mavacamten.

Mice were weighed and then anaesthetized using isoflurane and their hearts were excised. Following excision, the heart was placed into an ice-cold relaxing solution. The left ventricle was cut into 2–3 mm pieces and then placed into a Waring blender for 5–10 s. The resulting suspension of cells was then centrifuged for 105 s at 165 g. Myocytes were then permeabilized using Triton X-100.

For immunostaining, a drop of permeabilized cardiac myocyte preparations were added to a cover slip, fixed in 4% paraformaldehyde, washed in phosphate buffered saline (PBS), and blocked in 3% bovine serum albumin (BSA) for 1 h. Next the myocyte preparations were incubated with primary antibody to dystrophin (MANDYS1 3B7, Developmental Studies Hybridoma Bank, Iowa City, IA) overnight at 4°C. Coverslips were washed three times in PBS, then incubated in ALEXA 488 anti-mouse secondary antibodies for 1 hour at 37°C. Coverslips were sealed and imaged via confocal microscopy.

For mechanical measurements, permeabilized myocytes were attached between a force transducer and torque motor by placing a myocyte preparation into a stainless steel-troughs (25 gauge) and secured by overlaying the ends of the myocyte with 0.5 mm length of 4–0 monofilament nylon sutures (Ethicon, Inc.), then the sutures were secured into the troughs utilizing loops of 10–0 monofilament (Ethicon, Inc.). Attachment took place using a stereomicroscope (90x zoom). The myocyte preparation apparatus was transferred and mounted onto an inverted microscope (model IX-70, Olympus Instrument Co., Japan). A capacitance-gauge transducer and 10x amplifier (Aurora Scientific, Inc, Aurora ON, Canada) was used for mechanical measurements. Changes in length were introduced using a DC torque motor (model 308, Aurora Scientific, Inc.) by voltage commands from a personal computer using a 16-bit D/A converter (AT-MIO-16E-1, National Instruments Corp., Austin, TX, United States). Length and force signals were digitized at 1 kHz and stored on a personal computer in LabView for Windows (National Instruments Corp.). The length of the sarcomere was measured using a IonOptix SarcLen system (IonOptix, Milton, MA), which uses a fast Fourier transform algorithm on the video image of the myocyte. Once attached, the relaxed cardiomyocyte preparation was adjusted to a SL of ∼2.25 µm and passive tension was assessed by slacking the preparation in pCa 9.0 solution. Force, rate of force, and loaded shortening velocities were made at 15°C ± 1°C at two sarcomere lengths (see Figure 2 for representation of experimental workflow) (Hanft et al., 2021). A final force measurement was made at long sarcomere length in pCa 4.5 solution, and if was found to be below 70% of initial maximal force, data from that myocyte were discarded.

Force redevelopment following a slack-restretch maneuver was fit by a single exponential equation: F = F max 1   –   e − k t r t , where F is force at time t, F_max is maximal force, and k _tr is the rate constant of force development.

Myocyte length traces were fit to a single decaying exponential equation: L = A e − k t + C Where L is cell length at time t, A and C are constants with dimensions of length, and the rate constant of shortening is k. Velocity of shortening at time, t, was determined by the slope of the tangent to the fitted curve at t. Loaded shortening velocities were calculated from the onset of force clamp at t = 0 msec. The Hill equation (Hill, 1938) was used to fit hyperbolic force-velocity curves to relative force-velocity data. P + a   V + b = P 0 + a   b P is the force during shortening at velocity V; P₀ is the maximal isometric force; a is a constant with dimensions of force, while b is a constant with dimensions of velocity. Force-velocity data is normalized to isometric force. Force was multiplied by velocity to obtain power-load curves, and peak normalized power output (PNPO) values were obtained by the multiplication of relative force at optimum power by velocity at optimum power. Curves were fit using commercial software (SigmaPlot) and a custom program written using Qbasic. To calculate cross-sectional area, the cell shape was assumed to be elliptical so the width was multiplied by a factor of 0.7 on mnor axis (Strang et al., 1994).

WT and Dmd^mdx-4cv cardiac myocyte preparations were compared using Student’s t test. Within group sarcomere length dependence of k _tr values, force-velocity and power-load curve parameters were compared using paired t tests. WT and Dmd^mdx-4cv preparations were compared before and after mavacamten treatment using a paired t-test. Between group differences in response to mavacamten were compared using either a Mann-Whitney test (Figure 9B) or One-Way ANOVA (Figure 9C). A p-value below 0.05 was considered significant. Significance was denoted by *. N = number of mice/group and all data shown are biological replicates.

Contractile properties were measured in permeabilized cardiac myocyte-sized preparations from WT and Dmd^mdx-4cv mice. The characteristics of permeabilized cardiac myocyte preparations from both groups are provided in Table 1. Figure 2 illustrates experimental workflow for assessment of contractile properties of permeabilized cardiac myocyte preparations. First, force and rate of force development were measured during maximal Ca²⁺ activation. This was followed by measurements of force, rate of force, and loaded shortening during sub-maximal (∼half-maximal) Ca²⁺ activation at both long (∼2.25 μm) and then short (∼1.95 μm) sarcomere length (SL), followed by a final force measurement during maximal Ca²⁺ activation (Figure 2).

Maximal tension was similar between WT and the Dmd^mdx-4cv permeabilized cardiac myocyte preparations (WT: 46 ± 17 kN × m^-2; Dmd^mdx-4cv 45 ± 23 kN × m^-2) (Figure 3A). Additionally, there was no difference in rates of force development (as indexed by the rate constant, k _tr ) following slack-restretch manoeuvre during maximal Ca²⁺ activations between WT and Dmd^mdx-4cv myocyte preparations (WT: 10.5 ± 1.3 s^-1; Dmd^mdx-4cv: 10.9 ± 0.9 s^-1) (Figure 3B). Sub-maximal force development rates (k_tr) were also similar between WT and Dmd^mdx-4cv preparations at long SL and short SL (WT: Long SL: 6.13 ± 3.53 s^-1, Short SL: 5.94 ± 2.99 s^-1, Dmd^mdx-4cv: Long SL: 4.96 ± 2.27 s^-1, Short SL: 5.02 ± 2.08 s^-1) (Figure 3C).

Figure 4 shows representative force-velocity and power-load relationships (normalized to isometric force) from a permeabilized cardiac myocyte preparation from a WT mouse heart (left panel) and a Dmd^mdx-4cv mouse heart (right panel). Table 2 shows force-velocity and power-load characteristics at long and short sarcomere lengths from permeabilized cardiac myocyte preparations between groups. Interestingly, there was a greater sarcomere length dependence of peak normalized power output in permeabilized cardiac myocyte preparations from Dmd^mdx-4cv mice (WT: ΔPNPO long SL-short SL: 0.0004 ± 0.0076, Dmd^mdx-4cv: ΔPNPO long SL-short SL: 0.032 ± 0.022; p < 0.05) (Figure 4).

Length dependence of absolute power was also determined to look at the effects of dystrophin-deficiency on the product of force and loaded shortening. Figure 5 shows sarcomere length dependence of peak absolute power of cardiac myocyte preparations between groups. Length dependence of peak power (pW) was greater in the Dmd^mdx-4cv preparations compared to the WT preparations. From short SL to long SL power output increased 85.67 ± 40.58 pW in the Dmd^mdx-4cv preparations but only 31.00 ± 19.52 pW in the WT preparations. The sarcomere length dependence of tension was similar between groups (see Figure 8). These results suggest the loss of dystrophin increases sarcomere length dependence of loaded shortening and peak power in Dmd^mdx-4cv cardiac myofilaments.

Since we found greater sarcomere length dependence of loaded shortening and power in dystrophic cardiac myocytes, we next investigated the effects of mavacamten on contractile properties of cardiac myofilaments from DMD mice. Mavacamten has been shown to decrease contractility in myocardial preparations (Green et al., 2016; Awinda et al., 2021), making it a promising drug to treat hypercontractility disorders such as HCM and, it stands to reason, that it may be beneficial for dystrophic cardiomyopathy, which is thought to arise, at least in part, from contraction-induced damage. Mavacamten (0.5 µM) was added to pCa solutions and contractile properties were measured in permeabilized cardiac myocytes from WT and Dmd^mdx-4cv mice. The characteristics of permeabilized cardiac myocyte preparations from both treatment groups are provided in Table 3.

Figure 6 shows that mavacamten significantly decreased maximal tension (by ∼45%) in cardiac myocyte preparations from both WT and Dmd^mdx-4cv mice. This result is consistent with the tension decline by mavacamten previously reported in permeabilized multi-cellular myocardial preparations (Awinda et al., 2021). These results indicate that mavacamten mitigates tension generation in permeabilized cardiomyocytes regardless of dystrophin content.

Maximal Ca²⁺ activated rates of force development (k _tr values) were also significantly decreased by mavacamten in both WT and Dmd^mdx-4cv preparations, highlighting the ability of mavacamten to decrease force kinetics in single permeabilized cardiac myocyte preparations (Figure 7). Intriguingly, mavacamten tended to steepen absolute and normalized sarcomere length-tension relationships in cardiac myocyte preparations from both WT and Dmd^mdx-4cv preparations (Figure 8B), similar to a previous report (Sewanan et al., 2021).

Table 4 shows force-velocity and power-load characteristics at long and short sarcomere lengths from permeabilized cardiac myocyte preparations treated with mavacamten. Figure 9A shows representative absolute force-velocity and power-load relationships from a permeabilized cardiac myocyte preparation from a WT mouse heart (left panel) and a Dmd^mdx-4cv mouse heart (right panel). Interestingly, mavacamten eliminated any difference in sarcomere length dependence of power in cardiac myocyte preparations between WT and Dmd^mdx-4cv groups (Figure 9B). Sarcomere length dependence of absolute peak power (expressed in relative terms) shifted to a higher value in the WT myocyte preparations after mavacamten (Figures 9B,C). In fact, sarcomere length dependence of power with mavacamtem was similar to Dmd^mdx-4cv myocyte preparations without mavacamten (Figure 9C). These results implicate that, while mavacamten decreases overall contractility, it maintains or even augments sarcomere length dependence of contraction in cardiac myofilaments.