Cardiomyocyte Apoptosis Is Associated with Contractile Dysfunction in Stem Cell Model of MYH7 E848G Hypertrophic Cardiomyopathy

Missense mutations in myosin heavy chain 7 (MYH7) are a common cause of hypertrophic cardiomyopathy (HCM), but the molecular mechanisms underlying MYH7-based HCM remain unclear. In this work, we generated cardiomyocytes derived from isogenic human induced pluripotent stem cells to model the heterozygous pathogenic MYH7 missense variant, E848G, which is associated with left ventricular hypertrophy and adult-onset systolic dysfunction. MYH7E848G/+ increased cardiomyocyte size and reduced the maximum twitch forces of engineered heart tissue, consistent with the systolic dysfunction in MYH7E848G/+ HCM patients. Interestingly, MYH7E848G/+ cardiomyocytes more frequently underwent apoptosis that was associated with increased p53 activity relative to controls. However, genetic ablation of TP53 did not rescue cardiomyocyte survival or restore engineered heart tissue twitch force, indicating MYH7E848G/+ cardiomyocyte apoptosis and contractile dysfunction are p53-independent. Overall, our findings suggest that cardiomyocyte apoptosis is associated with the MYH7E848G/+ HCM phenotype in vitro and that future efforts to target p53-independent cell death pathways may be beneficial for the treatment of HCM patients with systolic dysfunction.


Introduction
Hypertrophic cardiomyopathy (HCM), characterized by unexplained left ventricular hypertrophy, affects 1 in 500 individuals in the general population [1,2] While the left ventricular function is generally preserved or hyperdynamic, it is increasingly acknowledged that the primary defect in some cases is impaired contractile function [1][2][3][4][5][6]. Mutations in myosin heavy chain 7 (MYH7), encoding a sarcomeric thick filament protein, are common genetic causes for HCM, accounting for 33% of the cases [7,8]. Recently, mavacamten, a myosin inhibitor, was found to improve symptoms in HCM patients with preserved to hyperdynamic systolic function and left ventricular outflow tract obstruction by reducing contractility; however, mavacamten is contraindicated in patients with reduced ejection fraction as it can further worsen systolic function [9]. Because~10% of HCM patients develop systolic dysfunction, this class of medication is not an option for them [9,10]. In order to develop novel therapies for patients with HCM and systolic dysfunction, a better understanding of the molecular mechanisms that govern this disease is needed.
Cardiomyocyte apoptosis has been observed in various models of cardiac diseases [11][12][13][14][15][16]. The activation of tumor suppressor p53, a major driver of intrinsic apoptosis, has been implicated in the progression of cardiac hypertrophy at both the cellular and tissue level [17][18][19][20][21][22][23][24]. In a human induced pluripotent stem cell-derived cardiomyocyte (hiPSC-CM) model of the MYH7 R403Q/+ HCM associated with hypercontractile function, p53 inhibition partially rescued cardiomyocyte survival but did not normalize the hypercontractile function in cardiac microtissues [17]. Given that the role of p53 in HCM associated with hypocontractile function is unknown, we hypothesize that inhibition of p53 in this setting will improve cardiomyocyte survival and overall contractile function.
In a previous work, patients harboring the heterozygous MYH7 E848G/+ variant presented with adult-onset familial systolic dysfunction and mild ventricular wall thickening [6]. Since that publication, an additional family member presented with significant left ventricular hypertrophy that met criteria for HCM. Given that HCM is now established in this family, the rest of the MYH7 E848G/+ family members exhibiting at least 1.3 cm wall thickening now meet diagnostic criteria for HCM as per the 2020 American College of Cardiology and American Heart Association HCM Guidelines [25]. Thus, MYH7 E848G/+ hiPSC-CM is an ideal model for testing the role of p53 in HCM associated with hypocontractile function. Here, we improve upon the prior viral transgenesis approach using clustered regularly interspaced short palindromic repeat (CRISPR)/Cas9 editing of patient-derived hiPSCs to generate isogenic lines expressing the pathogenic MYH7 E848G variant fused with enhanced green fluorescent protein (EGFP) to better understand the pathophysiology of MYH7 E848G/+ -based HCM associated with hypocontractile function. This model recapitulated the clinical phenotype as we observed increased cardiomyocyte hypertrophy and decreased tissue contractility in both patient-derived and isogenic hiPSC-CMs expressing MYH7 E848G/+ . In cardiomyocytes derived from the hiPSCs, we found that the MYH7 E848G variant increased cytotoxicity, apoptosis markers, and p53 expression, but genetic ablation of TP53 did not restore contractile function or cardiomyocyte survival. Overall, our findings suggest that in HCM patients with systolic dysfunction, cardiomyocyte apoptosis contributes to impaired tissue contractility with p53-independent cell death as a potential mechanism.

Generation of Isogenic βMHC-EGFP-Expressing hiPSC-CMs Using CRISPR/Cas9 Editing
Since the publication of our last study, another family member in the original study presented (patient Id) with clear left ventricular (LV) septal wall thickening on echo (1.9 cm) and severe LV systolic dysfunction (ejection fraction 39%) at age 57 [6] ( Figure 1A). Based on the diagnostic criteria for HCM as recommended, because one family member has clear HCM phenotype, the rest of the family members with at least 1.3 cm wall thickening would now meet diagnostic criteria for HCM (patient Ia) [25]. To study the effects of the MYH7 E848G variant in the context of isogenic gene-edited hiPSCs in vitro, we leveraged previously generated human induced pluripotent stem cells (hiPSCs) derived from an HCM patient (HCM IIb) and a non-variant family member (WT Ib) ( Figure 1A). To generate isogenic hiPSC lines with fluorescent tracking of beta-myosin heavy chain (βMHC), the protein encoded by MYH7, we designed a gene editing strategy to create hiPSC lines expressing βMHC-EGFP fusion proteins ( Figures 1B and S1A). Enrichment with the Pgk-puromycin cassette improved the gene-editing efficiency such that~10% of the colonies screened were correct ( Figure S1B). By knocking MYH7 cDNA in-frame with the sequence of EGFP into the endogenous MYH7 locus of HCM IIb MYH7 E848G/+ -hiPSCs, we enabled direct native control of the expression and tracking of the βMHC-EGFP fusion protein ( Figure 1B). With this approach, we generated four isogenic βMHC-EGFP-expressing hiPSC lines with all combinations of WT and E848G homozygous and heterozygous alleles, with one allele EGFP-tagged: MYH7 WT/WT-EGFP , MYH7 WT/E848G-EGFP , MYH7 E848G/WT-EGFP , and MYH7 E848G/E848G-EGFP ( Figure 1C). Notably, our editing approach yielded successfully edited clones as verified by Sanger sequencing with high efficiency, with cumulatively 10 of 76 picked clones (13.2%) across the four lines ( Figure S1B) correctly gene-edited. Green striated sarcomeres were visible in confocal microscopy in each line upon successful differentiation into hiPSC-CMs ( Figure 1C). Western blot confirmed the presence of two βMHC protein bands of roughly equal intensity, corresponding with untagged and EGFP-tagged βMHC ( Figure S1C), suggesting no preferential expression of one allele over the other. Differentiation of MYH7 WT/WT-EGFP and MYH7 WT/E848G-EGFP hiPSCs generated hiPSC-CMs of roughly 80% purity at differentiation day 25 as measured with EGFP + fraction via flow cytometry ( Figure S1E), with no difference in relative expression of EGFP-tagged βMHC as measured using EGFP + mean fluorescent intensity ( Figure S1F). The establishment of these hiPSC-CM lines enabled various lines of inquiry related to the effects of MYH7 E848G/+ on cardiomyocyte behavior in vitro, while also providing evidence of the utility of our approach for creating multiple edits to study a MYH7 variant.

MYH7 E848G Variant Reduces hiPSC-CM Survival in Monolayer Culture
To expedite the maturation process and, thus, the expression and effects of MYH7 E848G variant, we tested two different cardiomyocyte culture media: one with an RPMI base and low calcium concentration (0.4 mM Ca 2+ ), and one with a DMEM base and more approximately physiological calcium concentration (1.8 mM Ca 2+ ). After 10 days of treatment, EGFP intensity as measured with flow cytometry was significantly increased for MYH7 WT/WT-EGFP hiPSC-CMs cultured in DMEM-based media (93.6 ± 3.2% increase) relative to those in RPMI-based media ( Figure S2A,B); FSC area also increased with DMEMbased media (13.9 ± 8.0% increase) relative to RPMI-based media ( Figure S2C), suggesting increased maturation with the DMEM-based high calcium media. Moving forward, we used this DMEM-based media for all monolayer experiments.
While culturing the MYH7 E848G/+ cardiomyocytes, we noted a significant loss of the mutant cardiomyocytes during prolonged culture (Figure 2A). WT Ib and HCM IIb patient lines and MYH7-EGFP isogenic lines were monolayer-cultured for two weeks in DMEM-based media. Intriguingly, lines without MYH7 E848G had a negligible difference in cTnT + or EGFP + total cell count over time, but those with MYH7 E848G had a significant reduction in cTnT + or EGFP + total cell count ( Figure 2B). To further explore the manner of cell death, MYH7 WT/WT-EGFP and MYH7 WT/E848G-EGFP hiPSC-CMs were stained with terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL). The fraction of TUNEL + nuclei in EGFP + cells after 3 days of DMEM-based media treatment was significantly higher in MYH7 WT/E848G-EGFP (10.2 ± 0.8%) relative to MYH7 WT/WT-EGFP hiPSC-CMs (3.7 ± 0.3%), suggesting apoptosis was the cause of cell death ( Figure 2C,D). Combined, these data indicate MYH7 E848G/+ reduces hiPSC-CM viability when cultured on a stiff tissue culture surface, suggesting that the MYH7 E848G/+ cardiomyocytes may be susceptible to increased afterload.

MYH7 E848G/+ Reduces hiPSC-CM Survival and Increases Cardiomyocyte Size in EHTs
We posited that the reduced monolayer viability of hiPSC-CMs expressing the MYH7 E848G variant may translate to the EHT environment and help explain the impaired hypocontractility phenotype. Thus, we utilized a previously described method for papain-based digestion of EHTs into single cells for analysis with fluorescence-activated cell sorting (FACS) ( Figure 2E) [27]. There was a significant decrease in the EGFP + fraction of the sorted EHT population in MYH7 WT/E848G-EGFP tissues at both 1 week (13.2 ± 6.0%) and 3 weeks (7.6 ± 0.7%) post-cast, relative to the EGFP + fraction in MYH7 WT/WT-EGFP tissues (27.4 ± 0.9%, 20.9 ± 1.6%) as detected with flow cytometry ( Figure 2F,G). Although cardiomyocyte loss was persistent over time in the EHTs, there was differential survival in the 1st week of culture that was not seen between 1 and 3 weeks of culture in EHTs, indicating that once the EHTs had compacted to steady state, the MYH7 E848G expressing cardiomyocytes no longer exhibited increased cytotoxicity compared to the control line. This is in contrast to cardiomyocytes plated on stiff tissue culture plastic, where we found a persistent genotype-dependent decrease in survival over time ( Figure 2B), further suggesting that the genotype-dependent cardiomyocyte apoptosis is, in part, due to increased afterload. We next examined if the cardiomyocytes exhibited a hypertrophic response in the 3D environment. MYH7 WT/E848G-EGFP EHTs yielded EGFP + cardiomyocytes with increased forward scatter area (8.4 ± 1.7% increase) relative to those sorted from MYH7 WT/WT-EGFP EHTs ( Figure 2H), indicating that an E848G-induced hypertrophic response was also present in the 3D environment.

Discussion
In this work, we generated isogenic hiPSCs with MYH7-EGFP fusion expression in the endogenous MYH7 locus, with or without the MYH7 E848G variant. Our editing approach provides a couple advantages. First, the use of patient-derived hiPSCs with MYH7 E848G/+ as the parental cell line ensured that the corrected and variant isogenic lines have the same patient-derived genetic background. Approaches which use previously established wild-type lines as the base cell line do not capture the same genetic background as a patientderived model. Second, this gene-editing strategy leverages an antibiotic enrichment cassette that significantly reduces the number of colonies needed to be screened, thereby permitting the generation of multiple isogenic hiPSC lines with MYH7 variants.
In our isogenic hiPSC-CM model, the MYH7 E848G variant increased cell size, reduced cardiomyocyte survival, and reduced tissue contractility in three-dimensional culture. These findings correlated with reduced survival, cellular hypertrophy, and impaired tissue contractility in our patient-derived non-fluorescent hiPSC-CM lines. The cellular hypertrophy and decrease in cardiomyocyte survival has been reported in an hiPSC-CM model of the MYH7 R403Q/+ HCM associated with hypercontractile function [17]. In that study, p53 activity was elevated, and inhibition with the small molecule pifithrin partially rescued cardiomyocyte survival, but it did not normalize contractile function. We also observed increased p53 activity in our hypocontractile HCM model, and we genetically ablated TP53 to interrogate the role of p53. To our knowledge, this is the first study to fully ablate TP53 expression in the context of HCM-associated cytotoxicity and impaired tissue contractility. We believe genetic ablation provides a definitive answer on the role of p53 in MYH7 E848G/+ HCM associated with systolic dysfunction compared to alternative methods that utilize small molecules or viral transgenesis [17,21]. We have demonstrated that reduced cardiomyocyte survival and tissue hypocontractility are independent of p53 activity in our MYH7 E848G model of HCM with hypocontractile function. This does not rule out p53's role in other HCM-causative MYH7 variants or other sarcomeric variants.
This work represents the first attempt to leverage EHT dissociation [27] to interrogate hiPSC-CM survival, hypertrophy, and expression at a cellular level in the context of an HCM-causative variant with hypocontractile function cultured in a 3D cardiac organoid. Notably, MYH7 E848G increased cytotoxicity and cell size in the three-dimensional context, demonstrating that the variant effects in two-dimensional culture are also present in a more relevant, 3D environment.
In sum, we have shown that the MYH7 E848G HCM-causative variant associated with hypocontractile function yields cardiomyocyte hypertrophy with reduced survival and tissue contractility in a p53-independent manner, suggesting that future efforts to target p53-independent apoptotic mechanisms may be beneficial for the treatment of HCM associated with hypocontractile function.

Casting of Engineered Heart Tissues (EHTs)
EHTs were cast on polydimethylsiloxane (PDMS) microposts as previously described [26,27] Briefly, Sylgard 184 Elastomer Base and Curing Agent (Dow, 1317318, Midland, MI, USA) were mixed at a 10:1 ratio and cured in a custom 3D-printed mold for 18 h at 65 • C, with one flexible post and one glass rod filled stiff post per set of posts, 6 posts per array. Cured post arrays were removed from the mold and trimmed of excess PDMS. A total of 500k Day 25 hiPSC-CMs and 100k human Hs27a stromal cells were mixed with 3 U/mL thrombin from bovine plasma (Sigma, T4648, St. Louis, MO, USA) and 5 mg/mL bovine fibrinogen (Sigma, E8630, St. Louis, MO, USA) in 100 µL EHT media (sterile filtered RPMI, B27 supplement, 5 g/L aminocaproic acid (Sigma, A2 504-256-100G, St. Louis, MO, USA), penicillin/streptomycin). The cell slurry was added into 2% agarose wells between posts in a 24-well plate and incubated for 80 min at 37 • C, 5% CO 2 . Then, 350 µL EHT media was added to the wells, and tissues were incubated for 10 min at 37 • C, 5% CO 2 . Posts were carefully moved to a fresh 24-well plate in 2 mL EHT media, and tissues were cultured on posts for 3 weeks with media change every other day.

Analysis of EHT Contractile Force
Videos with 5 s duration of paced EHTs were analyzed as previously described [26,27].

PCR Amplification and Sequencing
Genomic DNA was isolated from hiPSC subclones using a DNeasy Blood and Tissue Kit (Qiagen, 69506, Hilden, Germany). A MYH7 fragment containing mutation was amplified through PCR using Q5 High-Fidelity DNA Polymerase (New England Biolabs, M0491L, Ipswich, MA, USA) and 500 nM forward and reverse primers (Table 3). PCR products were run on 1% agarose gels and extracted using a Fermentas Gel Extraction Kit (Invitrogen, K0692, Waltham, MA, USA). Sanger sequencing was performed by Eurofins Genomics (Louisville, KY, USA).

Immunocytochemistry
hiPSC-CMs were seeded at 25 k/cm 2 in Matrigel-coated 4-well chamber slides (Millipore, PEZGS0416, Burlington, MA, USA) and cultured for 72 h in DMEM-based cardiomyocyte culture media. Cells were fixed with 4% paraformaldehyde for 5 min at room temperature and permeabilized with 0.2% Triton X-100 (Sigma, X100-100ML, St. Louis, MO, USA) in 1× phosphate buffer saline (PBS) for 5 min at room temperature. Cells were rinsed twice with 1× PBS and incubated for 10 min in the dark at room temperature with 1:2000 Hoechst 33342 (Thermo Fisher, H3570, Waltham, MA, USA). Cells were imaged with 40× objective using a custom Nikon ECLIPSE Ti spinning disk confocal microscope with a Yokogawa W1 spinning disk head (Yokogawa, CSU-W1, Tokyo, Japan), using 405 and 488 nm lasers. Images were captured using NIS Elements AR software (Version 5.02.01, Nikon, Tokyo, Japan).

Apoptosis Antibody Array
Monoculture differentiation day 35 hiPSC-CMs were assessed using the 43-target Human Apoptosis Antibody Array (Abcam, ab134001, Cambridge, UK) according to manufacturer's protocol. Briefly, hiPSC-CMs were lysed with provided lysis buffer and blocked membranes were incubated with 200 µg lysate. Membranes were serially incubated with biotin-conjugated anti-cytokines and streptavidin-HRP. Chemiluminescence images were obtained using the ChemiDoc imaging system as above. Volumetric intensities were analyzed using Bio Rad Image Lab software as above.

Statistics
Statistical comparisons of cell size, forward scatter area, twitch force, cell count, protein expression, and gene expression were performed using one-tailed Student's t-tests with unequal variances and significance criteria p < 0.05.

Institutional Review Board Statement:
The study was conducted in compliance with the requirements stipulated in the U.S. Department of Health and Human Services (DHHS) Protection of Human Subjects regulations at 45 CFR 46, and approved by the Institutional Review Board of the University of Washington (FWA#00006878). Subjects gave informed consent and were enrolled under approved IRB protocols no. 1553 and 1334.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.