Myosin Heavy Chain Converter Domain Mutations Drive Early-Stage Changes in Extracellular Matrix Dynamics in Hypertrophic Cardiomyopathy

More than 60% of hypertrophic cardiomyopathy (HCM)-causing mutations are found in the gene loci encoding cardiac myosin-associated proteins including myosin heavy chain (MHC) and myosin binding protein C (MyBP-C). Moreover, patients with more than one independent HCM mutation may be at increased risk for more severe disease expression and adverse outcomes. However detailed mechanistic understanding, especially at early stages of disease progression, is limited. To identify early-stage HCM triggers, we generated single (MYH7 c.2167C > T [R723C] with a known pathogenic significance in the MHC converter domain) and double (MYH7 c.2167C > T [R723C]; MYH6 c.2173C > T [R725C] with unknown significance) myosin gene mutations in human induced pluripotent stem cells (hiPSCs) using a base-editing strategy. Cardiomyocytes (CMs) derived from hiPSCs with either single or double mutation exhibited phenotypic characteristics consistent with later-stage HCM including hypertrophy, multinucleation, altered calcium handling, metabolism, and arrhythmia. We then probed mutant CMs at time points prior to the detection of known HCM characteristics. We found MYH7/MYH6 dual mutation dysregulated extracellular matrix (ECM) remodeling, altered integrin expression, and interrupted cell-ECM adhesion by limiting the formation of focal adhesions. These results point to a new phenotypic feature of early-stage HCM and reveal novel therapeutic avenues aimed to delay or prohibit disease onset.

Base editing process using cytidine base editor (CBE) with a target cytidine (c) (red). Briefly, nuclease Cas9 (blue) was directed to the MYH7 c.2167C>T conversion site by the sgRNA (green) complementary to the target strand (gtatcgcatcctgaacccag). A cytidine deaminase converts the target c to uracil (u) which is read by DNA repair enzymes as the correct base due to the nick in the nontarget strand caused by Cas9, resulting in u conversion to thymine (t). (C) Base editing results, with nucleotide ratios (top) and chromatogram (bottom). In the bulk population 74% of the target c were successfully converted to t. Clones were then isolated from this bulk population. Note that we followed the same process to create the MYH6 c.2173C>T conversion and isolated MYH7/MYH6 mutant clone via clonal selection.   that were randomly picked from the bright-field images for each condition. Note that no significant difference between isogenic control 1 and 2 at Day 15 of differentiation, and no significant difference among the MYH7/MYH6 mutant 1, MYH7/MYH6 mutant 2 and MYH7 mutant. Statistically significant differences between any of the isogenic control and any of the mutant at both Day 15 and 45 time points. Data represents mean ± SEM. (N.S.: not significant; ***P < 0.005; Student t test and ANOVA; n = 24 for 6 randomly picked particles with round and intact boundary from 4 bright field images per condition). (D) Quantification of multinucleation. Analyzed from 9 IHC images for each condition. Note that no significant difference between isogenic control 1 and 2         Supplementary Tables   Table S1.

Pathogenic Converter Domain Mutations and Associated hiPSC and Animal Studies
Mutations in MYH7 converter domain [residues 707-774] interpreted as "pathogenic" or "pathogenic/likely pathogenic" in the NCBI ClinVar database, as of April 2021. Myosin conformation changes alter function, energetics and structure (Teekakirikul et al., 2010) 5

G716R
Patient-derived hiPSC-CMs Heterozygous Telomere shortening in hiPSC-CMs (Chang et al., 2018)        Supplementary Video S1. This video shows a side-by-side comparison of morphology and beat profile in MYH7/MYH6 mutant hiPSC-CMs and isogenic control hiPSC-CMs.

Re-plating hiPSC-CMs
To obtain monolayer hiPSC-CMs, a re-plating process was performed. Briefly, 1 mL of 0.25% Trypsin was added to each well and incubated at 37°C for 20 min. Singularized cells were obtained by pipetting with P1000 tips several times. The cell suspension was transferred to a conical tube containing RPMI20 (RPMI1640 with L-glutamine and 20% FBS) with two-fold cell suspension volume. The suspension was centrifuged at 200×g for 5 min to remove supernatant and hiPSC-CMs were seeded at the density of 9 × 10 4 cells/well on 12-well plates in RPMI/B27 medium with 10μM ROCK inhibitor. The re-plated hiPSC-CMs were cultured for 48 h and then fixed with 4% PFA for immunostaining.

Identification of Differentiation Efficiency
Differentiation efficiency was identified by the percentage of cardiomyocytes over total cell number. Cardiomyocytes were recognized via positive cTnT staining and total cell number was based on DAPI positive nuclei in a 10× view under the microscope (ZEISS, Oberkochen, Germany). At least 5 randomly picked 10× images for both mutant hiPSC-CMs and control hiPSC-CMs were utilized for the calculation.

Determination of hiPSC-CM Size
The singularized hiPSC-CMs were obtained via the same trypsinization process for re-plating. The cardiomyocyte size of the singularized hiPSC-CMs was examined by averaging the longest axis of 6 randomly picked particle (with clear boundary and intact shape) over 4 bright field images for each cell line on each time point.
The adherent cell area was measured from re-plated hiPSC-CMs. To determine the cell area, cardiomyocytes and cell boundary were recognized via positive cTnT staining. The cell area was obtained via circling each cell and calculating the area in ImageJ. At least 9 randomly picked 20× images for both mutant and control hiPSC-CMs were utilized for the assessment.

Determination of Multinucleation
Multinucleation was examined on re-plated hiPSC-CMs. To determine multinucleation, cardiomyocytes were recognized via positive cTnT staining and nuclei were recognized via positive DAPI staining. The percentage of multinucleation equals to the percentage of cardiomyocytes that contained more than one nucleus (i.e., the sum of bi-, tri-and multi-nuclei cardiomyocytes) over total cell number (i.e., the sum of single-, bi-, tri-and multi-nuclei cardiomyocytes). At least 9 randomly picked 20× images for both mutant and control hiPSC-CMs were utilized for the calculation.

Determination of Sarcomeric Disorganization
Sarcomeric disorganization was examined on re-plated hiPSC-CMs. Cardiomyocytes and sarcomere structures were recognized via cTnT staining. The annotation of 20× images were removed before performing the blinded classification for highly organized and poorly organized sarcomeres (representative images as shown in Figure 2F). At least 9 randomly picked 20× images for both mutant and control hiPSC-CMs were utilized for the calculation. The percentage of highly organized sarcomere over poorly organized sarcomere was determined as quantitative result of sarcomeric disorganization.

Determination of Cell Coverage
Cell coverage was examined on the undisturbed wells of hiPSC-CMs at Day 15, 30, 45 and 60 time points prior to the re-plating process. All the empty regions in each randomly picked 10× bright field image were measured and summed. Cell coverage percentage was determined by (ATotal -AEmpty)/ATotal × 100% where ATotal and AEmpty represent the total area of each image and the sum of empty areas, respectively.

Calcium Transient Measurements
Movement of Calcium ions was assessed with DMi8 fluorescence microscope (Leica, Wetzlar, Germany) and LAS X software. Cultured hiPSC-CMs were incubated with 5 μM Fluo-4 acetoxymethyl ester (Fluo-4 AM) in culture medium at 37°C for 30 min, and then culture medium was exchanged to Tyrode's salt solution for another 30 min incubation at 37°C. After that, hiPSC-CMs were moved onto the microscope stage and covered with a heating plate to maintain the temperature at 37°C. Fluo-4 AM intensity was recorded at the frame rate of 6.90 Hz with a 30 ms exposure time. The acquired data was processed by ImageJ to obtain a time trace of calcium signal and then analyzed in a custom-written Python script to extract maximal and minimal intensity and corresponding time points for each peak. Peak amplitude was determined by F/F0 where F = <Fmax> -<Fmin> and F0 = <Fmin>. <Fmax> and <Fmin> represent the averaged maximal and minimal intensity for each peak, respectively.

Determination of Relative Cardiomyocyte Size and Integrin Expression via Flow Cytometry
Singularized cell samples were collected and fixed in 4% PFA for 15 min and then washed and stored in PBS at 4°C. At least 0.5 × 10 6 cells per sample were split and assigned to unstained control, secondary antibody only, cTnT only, and cTnT and integrin co-stained samples (integrin beta 1 or alpha 2 or alpha v). Primary antibodies (mouse anti-human cTnT, rabbit anti-human integrin beta 1, rabbit anti-human integrin alpha 2, and rabbit anti-human integrin alpha v) and secondary antibodies (goat anti-mouse 488, and goat anti-rabbit 647) were diluted in 10 μg/mL digitonin 1:100 and 1:500, respectively. Cell samples were permeabilized and immunostained simultaneously (30 min at 4°C for primary antibody, washed with 0.05% (v/v) Tween 20 in PBS twice and then 30 min at 4°C for secondary antibody). Washed with 0.05% (v/v) Tween 20 in PBS twice before resuspended in 200 μL of PBS. Acquire data by Flow Cytometry (BD Accuri C6 Flow Cytometer). To determine cardiomyocyte size and integrin expression level, gated each cell line on each time point with unstained control, adjusted with secondary antibody only control to exclude false positive events and then targeted cTnT+ events as cardiomyocytes for analysis.

GO and KEGG Enrichment Analysis of Differentially Expressed Genes
Gene Ontology (GO) enrichment analysis of differentially expressed genes was implemented by the clusterProfiler R package, in which gene length bias was corrected. GO terms with corrected P-value less than 0.05 were considered significantly enriched by differential expressed genes. We used the clusterProfiler R package to test the statistical enrichment of differential expression genes in KEGG pathways.