A Laing distal myopathy-associated proline substitution in the β-myosin rod perturbs myosin cross-bridging activity

Abstract

Proline substitutions within the coiled-coil rod region of the β-myosin gene (MYH7) are the predominant mutations causing Laing distal myopathy (MPD1), an autosomal dominant disorder characterized by progressive weakness of distal/proximal muscles. We report that the MDP1 mutation R1500P, studied in what we believe to be the first mouse model for the disease, adversely affected myosin motor activity despite being in the structural rod domain that directs thick filament assembly. Contractility experiments carried out on isolated mutant muscles, myofibrils, and myofibers identified muscle fatigue and weakness phenotypes, an increased rate of actin-myosin detachment, and a conformational shift of the myosin heads toward the more reactive disordered relaxed (DRX) state, causing hypercontractility and greater ATP consumption. Similarly, molecular analysis of muscle biopsies from patients with MPD1 revealed a significant increase in sarcomeric DRX content, as observed in a subset of myosin motor domain mutations causing hypertrophic cardiomyopathy. Finally, oral administration of MYK-581, a small molecule that decreases the population of heads in the DRX configuration, significantly improved the limited running capacity of the R1500P-transgenic mice and corrected the increased DRX state of the myofibrils from patients. These studies provide evidence of the molecular pathogenesis of proline rod mutations and lay the groundwork for the therapeutic advancement of myosin modulators.

Keywords: Molecular pathology; Mouse models; Muscle; Muscle biology.

Figure 1. Transgenic mice expressing the myosin mutant R1500P have reduced muscle strength.

(A) Pie chart shows the relative percentage of the different myosin isoforms expressed in TA muscles of NTG mice, transgenic mice expressing WT MYH7 (βWT), and mice expressing mutant MYH7 (R1500P), quantified by myosin unique peptide proteomics. Six- to 7-month-old male mice were studied (n = 3 mice/group). (B) Average ratio of TA muscle/body weight (left) and TA muscle weight (right); Six- to 10-month-old male mice were studied (n = 20 mice/group). (C) Representative SDH activity of TA muscle cross-sections isolated from 8-month-old male βWT (left) and R1500P (right) animals. Scale bar: 200 μm. (D) Quantification of SDH-positive fibers (left), and fiber CSA, reported in μm² divided by 100 (right). Representative images of sections (fields) used in the CSA quantification are shown in E and Supplemental Figure 2. (E) Myofibrils from TA muscles from 6-month-old male βWT (left) and R1500P (right) mice were prepared and immunostained with the Myc antibody. Samples were imaged on a confocal spinning-disk microscope (Nikon Ti-E; ×100 silicon immersion objective). Scale bar: 5 μm. (F) Four-limb hanging (left) and grip strength (right) tests. The hanging time for each animal was calculated after 3 trials. Eight-month-old male mice were used (n = 6 mice/group). Measurements of forelimb grip strength were carried out with a computerized grip strength meter. Seven- to 8-month-old male mice (n = 6 mice/group). All data are expressed as the mean ± SEM. *P < 0.05 and **P < 0.01, by 2-tailed, unpaired Student’s t test with Welch’s correction.

Figure 2. Expression of the myosin mutant R1500P alters skeletal muscle performance.

EDL muscles isolated from NTG, βWT, and R1500P mice were subjected to an ex vivo contractility assay. (A) Comparison of specific twitch force and tetanic force. Six-month-old male mice (n = 4/mouse group) were used. Data are expressed as the mean ± SEM. *P < 0.05 and **P < 0.01, by 1-way ANOVA with Bonferroni’s multiple-comparison post hoc test. (B) Influence of intermittent tetanic stimulation on fatigue. The tetanic stimulation protocol was 100 Hz for 500 ms, once every second for 60 seconds (representative plot of 3 independent experiments). Six-month-old male mice were used.

Figure 3. Myofibrils containing the myosin mutant R1500P show faster cross-bridge detachment under isometric conditions.

(A) Representative trace of linear and exponential phase relaxation of myofibrils isolated from TA muscles. (B) Rate constant (k _REL,LIN); (C) duration of the early linear phase of relaxation (t _REL,LIN); and (D) rate constant of the final exponential phase of force decline (k _REL,EXP) measured from βWT and R1500P myofibrils. The activity of 4–6 myofibrils/muscle was averaged. Six-month-old male mice were used (βWT, n = 3; R1500P, n = 5). Data are expressed as the mean ± SEM. *P < 0.05 and **P < 0.01, by 2-tailed, unpaired Student’s t test with Welch’s correction. See Supplemental Figure 6 for individual βWT and R1500P myofibril measurements.

Figure 4. The R1500P mutation changes myosin enzymatic activity as well as the thick filament structure.

Percentage of myosin molecules in the SRX or DRX state measured by Mant-ATP chase assay. (A) Analysis of purified muscle fibers from 6-month-old male βWT and R1500P mice. Control mice, n = 10; βWT mice, n = 11; R1500P mice, n = 12. (B) Analysis carried of biopsy specimens from patients with distal myopathy caused by the MYH7 missense mutations L1467P (35-year-old female) and R1560P (52-year-old male). Controls correspond to the mouse TA and human vastus lateralis muscles lacking or having type 1 fibers, respectively. Measurements were recorded in the absence or presence of MYK-581 (0.3 μM and 1 μM). Control mice, n = 17; L1467P mice, n = 4; R1560P mice, n = 6. See Supplemental Figure 7 for relative DRX scatter plots. Data are expressed as the mean ± SEM. *P < 0.05 and **P < 0.01, by 2-way ANOVA (genotype × drug) with Bonferroni’s multiple-comparison post hoc test. (C) MM reflections recorded from x-ray diffraction experiments carried out on βWT and R1500P purified fibers that were immersed in relaxing buffer, mounted in a specific chamber, and examined. All the MM1 and MM2 integrated intensities were normalized to the sixth actin layer line intensity.

Figure 5. The compromised running performance of the transgenic R1500P mice is partially rescued by oral treatment with the myosin modulator MYK-581.

Mouse locomotion analysis was recorded over 15 days of voluntary wheel running. Actograms of distance and activity are shown. Four-month-old male mice were acclimated to the running wheel cages for 4 days before data collection was commenced. Kilometers run (distance) and hours spent on the wheel (activity) are plotted as a function of the total experimental time measured in days. Dosed groups received MYK-581 (10 mg/kg) via chow starting on the first day of acclimatization. βWT mice, n = 7; βWT dosed mice, R1500P mice, and R1500P dosed mice, n = 10/group.

Figure 6. MYK-581 significantly ameliorates the running performance of transgenic R1500P mice.

(A) Daily average distance (Km) covered by each group of 4-month-old male mice over 15 days is shown on the left; statistical analysis of the cumulative distance as well as the comparison between day 1 and day 15 is shown on the right of the panel. (B and C) Group activity (h) and average (Avg) speed (Km/h). βWT mice, n = 7; βWT dosed mice, R1500P mice, and R1500P dosed mice, n = 10/group. All data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001, by 2-way ANOVA (genotype × drug) with Bonferroni’s multiple-comparison post hoc test.

Figure 7. Markov model of mouse activity shows hidden running patterns.

(A) Running wheel activity of 2 representative 4-month-old βWT and R1500P male mice recorded during the circadian α phase. (B) Diagram of the discrete-time Markov chain used for modeling mouse running bouts and rests. The model transitions between 1 running and 3 resting states. The condition p_ss < p_mm < p_ll guarantees that leaving the short, medium, and long rest states is increasingly difficult. (C) PMFs associated with the Markov model, giving the probability of the duration of every single run and rest. PMFs are shown with a linear (Lin) and a logarithmic (log) y axis. (D) PMFs associated with the Markov model describing the duration of a single resting bout. PMFs are shown with a linear and a logarithmic y axis. The last panel on the right reports the first 1,500 seconds of the log plot. Time windows of interest are labeled i–iii. (E) Probability density plot reporting the velocity distributions of the groups as determined experimentally. βWT mice, n = 7; βWT dosed mice, R1500P mice, and R1500P dosed mice, n = 10/group.