Human skeletal myopathy myosin mutations disrupt myosin head sequestration

Figures

• 

  Figure 1. Myosinopathies and location of mutations. (A) A schematic showing the overall composition of striated myosin. The molecule is formed by 2 heavy chains that dimerize to form a coiled coil tail composed of subfragment-2 (S-2) and light-meromyosin (LMM). The 2 heavy chains diverge to form the neck of lever, to which light chains bind, and the 2 motor domains, which bind actin and nucleotide. (B) A schematic showing an end on view of the heptad repeat of 2 interacting α-helices. Residues in “a” and “d” positions form the hydrophobic seam. (C) The frequency of mutations in MYH7 for hypertrophic cardiomyopathy (HCM) (gray dotted line) and for skeletal myopathies (myosinopathy, green line), across the amino acid sequence of MYH7. (D) The sequence in which mutations that cause myosinopathies are most frequent. Positions of mutations (commonly mutation to proline or a single amino acid deletion) are indicated by colored arrows. Mutations in residues mutated in HCM are indicated by an asterisk. Underlined residues indicate the position of mutated residues studied here. The heptad repeat is shown underneath the sequence, with “d” positions highlighted in magenta.

• 

  Figure 2. Secondary structure of LMM constructs. (A and B) The mean circular dichroism (CD) spectra at 10°C for each of the MYH7 (A) and MYH2 (B) LMM constructs. Data are shown as mean ± SD. MRE, mean residue ellipticity. Mutations to proline are shown in shades of red, deletion mutations in shades of gray, and other mutations in shades of blue. (C) The percentage of helicity for each MYH7 and MYH2 LMM construct calculated from the 222 nm MRE values from a minimum of 3 separate experiments. Individual values together with the mean ± SD are shown. The normalized MRE measured at 222 nm from 10°C to 70°C for MYH7 (D) and MYH2 (E) constructs for a minimum of 3 experiments. These data were used to calculate Tm (temperature at which half the protein is melted) for each construct plotted in F. Individual data points for each measurement, together with the mean values ± SD, are shown. Significant differences compared with WT are indicated; the 1-way ANOVA with Dunnett’s test post hoc correction was used with *P < 0.05; ***P < 0.001; ****P < 0.0001. Heptad positions for the mutated residues in MYH7: A1440 and A1636 ‘b’; E1610 and K1729 ‘c’; L1492 ‘e’; E1507 and R1845 ‘f’; E1508 ‘g’ and in MYH2:…

• 

  Figure 3. Molecular dynamics simulations of mutant coiled coil models. (A) The distance between the helices (Dcom) for MYH7 and MYH2 mutants as a function of residue position compared with WT (open circles). (B) Heptad length, the distance between Cα atoms in residues at positions i and i + 7, averaged across both chains of the coiled coil. (C) Interheptad angle, the angle between the lines linking Cα atoms in residues at positions (i – 7), (i), and (i, i + 7), averaged across both chains of the coiled coil. Filled circles, results for the mutant; open circles, results for the WT. Arrows and star indicate the position of the mutated residue in each of the plots. Three independent simulations were performed. Data are shown as mean ± SD for each data point between simulations.

• 

  Figure 4. Effects of mutations on aggregation of GST-LMM into minifilaments in vitro. (A) Examples for GST-filaments, negatively stained and imaged using electron microscopy. The left column shows GST-LMM filaments that are effectively homozygous, as the filaments were generated from a pure population of each type of construct. No images are shown for either of the MYH7 mutations to proline (L1492P or A1630P), as filaments did not form under these conditions. The right column shows heterozygous GST-LMM filaments, in which the mutant LMM was mixed 50:50 with WT LMM. Scale bar: 100 nm. (B) Filament lengths for homozygous populations and the widths. (C) Filament lengths for heterozygous populations and the widths. The mean ± SD is shown for each set of measurements. Significant differences are as shown. The 1-way ANOVA with Dunnett’s test post hoc correction was used with **P < 0.01; ***P < 0.001; ****P < 0.0001.

• 

  Figure 5. Analysis of incorporation of mutant isoforms of eGFP–MHC into muscle sarcomeres in cultured skeletal muscle myotubes formed by C2C12 cells. (A) Images of eGFP-MHC+ myofibrils within cultured skeletal muscle myotubes for WT MYH7 and each of the mutants. (B) An example image of a single eGFP-MYH7+ myofibril together with its associated line profile. Positions of the minimum intensity value (at the M-line) and ends of the sarcomere (at the Z-disc) are indicated. (C). Mean fluorescence intensity profiles for eGFP-MHC organization across a single sarcomere (from Z-disc to Z-disc) for WT and each mutant. SD bars are omitted for clarity. The mean plot profile was calculated from a minimum of 7 myofibrils taken from different myotubes from 3 biological replicates (mean, SD, and n values are in Supporting Data Values). The MYH2 L1871P mutant did not express well enough to quantify sarcomere incorporation. (D) The peak-to-peak distance across the M-line for each of the GFP-MHC constructs, together with the mean ± SD values. Significant differences were calculated using a 1-way ANOVA with Dunnett’s test post hoc correction. **P < 0.01 and ****P < 0.0001.

• 

  Figure 6. Mant-ATP chase experiments to estimate DRX/SRX ratios. Muscle fibers were isolated from controls (CTL) or from patients with MYH7 (MYH7) or with MYH2 (MYH2) mutations. The proportion of myosin molecules in the disordered-relaxed (P1) and super-relaxed states (P2), as well as their respective ATP turnover lifetimes (T1 and T2), are presented. Data are separated according to the myosin heavy chain expression of individual fibers: either β/slow (graphs A, B, E, and F) or type IIA (graphs C, D, G, and H). For the β/slow isoform, n = 73 were from CTL and n = 214 were from patients. For the type IIA isoform, n = 81 were from CTL and n = 88 were from patients. Mean ± SD also appear on the graphs. Significance indicates a difference with control. The 1-way ANOVA with Dunnett’s test post hoc correction was used with *P < 0.05. Typical Mant-ATP chase experimental data showing exponential decays are presented in Supplemental Figure 4.

• 

  Figure 7. Modeling and experimental cellular force production. (A) The 3-component model used for our simulation from ref. 26 with SRX, DRX, and FG (force-generating) states. (B) The result of the simulations. Experimental specific force and Ca2+ sensitivity for muscle fibers isolated from controls (CTL) or from patients with MYH7 (MYH7) or with MYH2 (MYH2) mutations. Data are separated according to the myosin heavy chain expression of individual fibers: either β/slow (graphs C and E) or type IIA (graphs D and F). For the β/slow isoform, n = 72 were from CTL and n = 161 were from patients. For the type IIA isoform, n = 64 were from CTL and n = 92 were from patients. Mean ± SD appear on the graphs. The 1-way ANOVA with Dunnett’s test post hoc correction was used, but no significant differences were seen.