Heterozygous MYH7 R403Q mutation impairs left atrial mitochondrial function in a Yucatan mini-pig model of genetic non-obstructive hypertrophic cardiomyopathy

Abstract

Hypertrophic cardiomyopathy (HCM) can be caused by a MYH7 R403Q gene mutation, which drives pathological cardiac remodeling and may ultimately lead to heart failure. Here we sought to examine the effects of this mutation on cardiac mitochondrial function in a Yucatan mini-pig model of genetic HCM. Activity of key mitochondrial enzymes, citrate synthase and β-HAD, were significantly reduced in the left atria of HCM animals compared to the control group. However, left atrial mitochondrial respiration was significantly greater in HCM pigs vs controls in the following states: basal (42%, p=0.001), state 2 (47%, p=0.02) and uncoupled (p=0.003), potentiating a compensatory mechanism. Surprisingly, left ventricular mitochondrial respiration and mitochondrial enzymatic activity did not differ between the HCM model vs healthy control pigs. However, proteomic profiling revealed parallel mitochondrial dysfunction and impairment to energy metabolism processes in both chambers, such as inhibited fatty acid metabolism and mitogenesis in the left atria and increased mitochondrial dysfunction and concentration of fatty acids in the left ventricle. Collectively, the MYH7 R403Q mutation may contribute to HCM through chamber-specific mechanisms that promote mitochondrial dysfunction and impaired energy homeostasis. Further, these findings demonstrate the utility of this preclinical large animal model for identifying novel mechanisms underlying genetic heart failure with translational impact for individuals affected with HCM.

NEW & NOTEWORTHY

Changes in mitochondrial function has been proposed in the etiology of hypertrophic cardiomyopathy (HCM). In this report, we examine mitochondrial function and activity in response to a MYH7 R403Q gene mutation that causes HCM. Our findings show chamber-dependent mitochondrial dysfunction and decreased enzymatic activity, which affect key cellular processes such as ATP production and metabolism. These findings highlight chamber-specific metabolic dysfunction that may contribute to the development of HCM.

Graphical Abstract

INTRODUCTION

It is estimated that by 2030, heart failure (HF) will affect more than eight million individuals (1). Familial hypertrophic cardiomyopathy (HCM) is a genetic cardiovascular condition affecting 1 in 500 individuals that may progress to HF and increases the risk of sudden cardiac death (2). Mutations to myosin heavy chain 7 (MYH7), a gene which encodes the β- myosin heavy chain subunit (β-MHC) primarily in the heart but is also present in slow-twitch skeletal muscle fibers, contributes to approximately 1/3 of all familial HCM cases (3). The phenotype of HCM is heterogenous across clinical populations, and one potential contributor to its development is mitochondrial dysfunction (4). Indeed, comorbidities and risk factors associated with cardiovascular disease such as obesity, physical inactivity, type 2 diabetes, and hypertension can contribute to the progression of HCM (5).

The R403Q mutation in the MYH7 gene was one of the first mutations found to cause HCM (6). This mutation promotes HCM through concentric remodeling of the heart, resulting in cardiac hypertrophy, fibrosis, diastolic dysfunction and hypercontractility at rest (7). Smaller animal models exist for studying the R403Q mutation and have been successful in promoting the HCM phenotype seen in humans (8). However, both humans and swine predominately express the β-MHC isoform in the ventricles (9), whereas rodents primarily express the α-MHC (MYH6) isoform. Thus, small animal models may not be the most appropriate model to translate findings for clinical populations (10).

In this study, a Yucatan mini-pig model with a heterozygous MYH7 R403Q mutation was utilized (11, 12) to investigate the ramifications of this genetic alteration on cardiac muscle metabolism in a preclinical large animal setting of experimental HCM, including the left atria (LA) and the left ventricle (LV) (13). Mitochondria within cardiac myocytes are very abundant and heavily involved in cellular functions, producing more than 95% of the ATP (14). Optimal functioning mitochondria in cardiomyocytes is critical to maintain efficient energy production via the ability to utilize metabolic substrates from various molecular sources both aerobic and anaerobically. Although altered cardiac metabolism and mitochondrial dysfunction in the LV and septum has been shown in patients with HCM (15, 16), to our knowledge there have been no studies conducted comparing atrial versus ventricular metabolism. Given the translational relevancy of large animal models due to their functional and anatomical similarities to humans (17), the current study investigated cardiac mitochondrial function in a swine model of HCM with heterozygous MYH7 R403Q mutation to assess chamber-specific differences between the atria and ventricle. Specific measurements of targeted metabolic indicators were complemented with global proteomics analysis to identify differentially expressed proteins and pathways altered between chambers.

MATERIALS AND METHODS

All protocols were submitted and approved by the Animal Care and Use Committee at the University of Missouri-Columbia.

Animal model

Eight-month-old male Yucatan swine were used across all experiments (n=11). Yucatan mini-swine with a heterozygous MYH7 R403Q mutation (n=6; 39.7 ± 7.7 kg) swine served as the experimental model in the current study and referred to as MYH7-HCM. Wildtype Yucatan mini-swine served as controls (n=5; 28.7 ± 3.0 kg). All animals consumed a diet with 12% kcal protein, 84% kcal carbohydrate, and 4% kcal fat (Lonestar Laboratory Swine, Sioux Center, IA).

Tissue collection

Animals were euthanized with 4mg/kg Telazol and 2mg/kg Xylazine and maintained on 5% Isoflurane until the heart was excised. Left atria (LA) and left ventricle (LV) was excised in 1–2 cm pieces and flash frozen in liquid nitrogen. An additional 500mg-1g of each tissue were used to isolate mitochondria using previously described methods (18).

Citrate synthase activity

Citrate synthase activity, the rate limiting enzyme in the TCA, was assessed in the whole tissue and mitochondria in both the LA and LV using previously described methods by our group (19).

β-HAD activity

β-Hydroxyacyl-CoA dehydrogenase (β-HAD), the rate limiting step in fatty acid beta-oxidation, activity was assessed in whole heart tissue and isolated mitochondria in the LA and LV using methods previously described by our group (20, 21).

Cardiac muscle mitochondrial respiration

Mitochondria from LA and LV were isolated as previously reported (18) and placed in mitochondria isolation buffer. Mitochondrial respiration was evaluated by high-resolution respirometry (Oroboros Oxygraph-2k; Oroboros Instruments, Innsbruck, Austria). Basal respiration was assessed by loading approximately 65μg of LV and LA mitochondria without any additional substrates. State 2 respiration was stimulated by the addition of malate (2mM) and glutamate (5mM), state 3 complex I stimulated by titrated ADP (25–125uM), state 3 complex I and II with succinate (10mM), and maximally uncoupled respiration was assessed by the addition of carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP; 0.25uM). Mitochondrial respiration was calculated by O₂ flux per mass (pmol/s*mg) and normalized to protein content obtained using Pierce™ BCA assay kit (Thermo Scientific, Waltham, MA).

Western blotting

Whole tissue homogenates of LA and LV were prepared as previously described by our group (19, 21) and protein content was assessed in triplicate via Pierce™ BCA assay kit (Thermo Scientific, Waltham, MA). Briefly, 20 μg of protein lysate was loaded in a criterion gel (Bio-Rad, Hercules, CA) to further separate and transfer the proteins onto a PVDF membrane. Membranes were blocked and washed appropriately, then imaged and normalized to amido black staining. Protein bands were quantified using the ChemiDoc XRS+ System (Bio-Rad, Hercules, CA). Each membrane was incubated overnight at a 1:1000 dilution with the primary antibodies: oxidative phosphorylation (OXPHOS) complexes (Abcam, cat no. ab110413), lactate dehydrogenase (LDH; Abcam, cat no. ab231903), and pyruvate dehydrogenase (PDH; Invitrogen, cat no. PA5–31519). Following washing, membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (Cell Signaling, Danvers, MA).

Generation of homogenates for proteomic analysis

Tissue homogenates were processed in accordance with suggestions provided by Somalogic, Inc (22). 40 – 60 mg of tissue was excised from flash-frozen cardiac LA or the apex of the LV from either wild-type or R403Q animals (N=3 each group). The excised tissue was placed into a 2 mL screw-top vial and were homogenized by bead-based tissue disruption (Mini-BeadBeater 96, BioSpec Products, Inc., Bartlesville, OK) for 2 cycles of 3 min disruption at 2400 rpm in a pre-chilled tube rack) into 1 mL of T-PER Reagent containing 0.1 % v/v HALT protease inhibitor cocktail (ThermoFisher Scientific). The supernatant was clarified by centrifugation (10,000 xg, 10 min, 4 C) and the protein concentration measured by Pierce 660 protein assay reagent (ThermoFisher Scientific). The samples were diluted to 0.2 mg/mL total protein, per Somalogic’s recommendations, and aliquots flash-frozen on liquid nitrogen for SOMAscan analysis.

SOMAscan assay

Cardiac tissue homogenates, prepared at Edgewise Therapeutics, were sent to Somalogic, Inc., for analysis using their SOMAscan assay (Version 4.1). Proteins were identified using somamers (modified apatamers) and quantified with a microarray-style chip to return raw fluorescence values. Those values were normalized and standardized by comparing calibration samples against species-specific standards to generate relative fluorescence units (RFU) for downstream analysis (23). SOMAscan results were processed through Ingenuity Pathway Analysis at the University of Missouri (IPA; Qiagen) to identify top differentially expressed proteins associated with energy metabolism and mitochondrial function. Pathways with z-scores ≤ −2 and ≥ 2 were considered significant.

Statistical and Data analysis

GraphPad Prism version 10.1.2 (GraphPad Software, La Jolla, CA) was used to perform all statistical analyses. Statistical significance was determined using an α level of p < 0.05. Unpaired two-tailed student’s t-tests were used to detect differences between swine with HCM compared to control swine across each outcome measure. Data is expressed as mean ± standard error (SE), unless noted otherwise.

Data delivered from Somalogic were analyzed using Python (version 3.9 – 3.12 extended with pandas, numpy, scikit-learn, scipy, lxml, and pytorch modules and PyCharm (version 2024.1.1) interactive development environment for writing, debugging, and running custom software. Data from SOMAscan assay runs (ADAT files) contained assay information, quality control (QC) metrics, sample information, reference standards, and RFU values for all 7289 somamers targeting human proteins. ADAT files were imported into Python and analyzed for normalization and standardization notes, which indicated no problems with systematic error or bias. RFU values are heteroscedastic and exhibit a log-normal distribution; therefore, RFUs were log₂ transformed for downstream statistical analyses. To identify differentially expressed proteins, hypothesis testing was performed using t-tests (p < 0.05) to compare distributions of log₂ values for individual proteins. Equal variances and a two-sided alternative hypothesis were assumed.

RESULTS

Citrate synthase and β-HAD Activity

In whole LA tissue, citrate synthase activity was significantly reduced by 20.7% in MYH7-HCM compared to control swine (p=0.02), with no changes in LA isolated mitochondria (p=0.43). Citrate synthase activity in both whole (p=0.21) and mitochondrial LV (p=0.11) was not significantly different between groups (Figure 1A).

Figure 1. Mitochondrial enzymatic activity is reduced in the left atria, with a compensatory increase in mitochondrial respiration of animals with MYH7 R403Q mutation.

Citrate synthase activity (A) and β-Hydroxyacyl-CoA dehydrogenase (β-HAD) activity (B) measured in whole tissue and isolated mitochondria in left atria and left ventricle. Both enzymatic activities decreased in the left atria whole tissue. Mitochondrial respiration assessed with high resolution respirometry on isolated mitochondria from left atria (C) and left ventricle (D). Basal state: no substrates added; state 2: addition of malate and glutamate; state 3 complex I: addition of ADP; state 3 complex I and II: addition of succinate; uncoupled: addition of FCCP. ***p<0.001, **p<0.001, *p<0.05. LA: left atria; LV: left ventricle.

β-HAD enzymatic activity was significantly reduced in the whole tissue of LA of the MYH7-HCM animals compared to controls (p=0.04), without significant changes found in whole LV (p=0.65). Within the isolated mitochondria, β-HAD activity was not significantly altered in either LA (p=0.84) or LV (p=0.54) (Figure 1B).

Mitochondrial respiration

In the MYH7-HCM animals, LA mitochondrial respiration was significantly increased in the basal (p=0.001), state 2 (p=0.02) and maximal uncoupled respiration states (p=0.003), compared to control (Figure 1C). State 3 complex I + II in animals with HCM was increased by 31% and approached significance (p=0.05). There were no significant changes to mitochondrial respiration in the LV under any studied conditions (Figure 1D).

Protein expression

OXPHOS complexes

In the LA, protein expression of complex II (p=0.03) and III (p=0.01) were significantly reduced in MYH7-HCM compared to Yucatan control (Figure 2A). Complex IV trended lower (p=0.10), displaying a 30% decrease in the MYH7-HCM animals, with no significant changes in complex I (p=0.12) or complex V (p=0.20). In LV whole tissue, there were no significant alterations in all five complexes (p>0.05; Figure 2B). However, it is worth noting that a 30% decrease in Complex III expression trended lower in MYH7-HCM animals (p=0.11; Figure 2B).

Figure 2. MYH7 R403Q mutation decreases mitochondrial protein levels and increases the metabolic marker lactate dehydrogenase in the left atria.

Protein expression of oxidative phosphorylation (OXPHOS)(A-B), lactate dehydrogenase (LDH) (C-D), and pyruvate dehydrogenase (PDH) (E-F) assessed in left atria and left ventricle whole tissue. Protein content corrected with amido black staining is graphed. Representative western blot images depicted in far right. *p<0.05.

PDH and LDH

LDH expression in the LA was significantly increased in the MYH7-HCM compared to Yucatan control (p=0.04; Figure 2C), with no significant changes in the LV (p=0.77; Figure 2D). PDH expression was not significantly altered in either the LA (p=0.34; Figure 2E) or the LV (p=0.68; Figure 2F).

SOMAscan and Ingenuity Pathway Analysis

SOMAscan identified 7,289 total proteins in whole LA (Figure 3) and LV (Figure 4) tissue that were filtered to identify differentially expressed proteins between groups as shown in corresponding volcano plots. In the LA, 383 downregulated and 236 upregulated proteins in the LA were identified as significant (Figure 3A), while in the LV 450 downregulated and 551 upregulated proteins were detected (Figure 4A). Differentially expressed proteins from both the LV and LA were inputted into IPA to discover differentiated protein networks and canonical pathways.

Figure 3. Proteomic analysis reveals networks supporting inhibition of fatty acid and glucose metabolism and mitogenesis in the left atria of swine with MYH7 R403Q mutation.

The number of downregulated (green) and upregulated (red) differentially expressed proteins present in the LA of MYH7-HCM swine versus Yucatan control (n=3/group) are shown in the volcano plot and Venn diagram (A). DEPs present in the sample predict significantly altered pathways related to energy metabolism and mitochondrial function ranked by z-score (B). Upstream regulators predicted to inhibit the fatty acid metabolism pathway (C). Clusters of proteins predicted to inhibit uptake of D-glucose and quantity of carbohydrate while activating diabetes mellitus (D), with a separate network inhibiting mitogenesis (E). z-score ≤ −2 and ≥ 2 represents significance.

Figure 4. Proteomic analysis identifies numerous networks associated with metabolic impairment including decreased oxidative phosphorylation and increased fatty acid concentration in the left ventricle of swine with MYH7 R403Q mutation.

The number of downregulated (green) and upregulated (red) differentially expressed proteins present in the LV of MYH7-HCM swine versus Yucatan control (n=3/group) are shown in the volcano plot and Venn diagram (A). DEPs present in the sample predict significantly altered pathways in swine with MYH7 R403Q mutation related to energy metabolism and mitochondrial function ranked by z-score (B). Predicted proteins predicting activation of concentration of fatty acids and inhibition of energy homeostasis and quantity of carbohydrate (C). z-score ≤ −2 and ≥ 2 represents significance.

In the LA, IPA analysis revealed significant downregulation of pathways related to metabolism including PIP3-activated AKT signaling (p=1.55E⁻⁹, z-score = −3.900), extra-nuclear estrogen signaling (p=8.91E⁻⁶, z-score = −3.162), nitric oxide signaling in the cardiovascular system (p=0.0001, z-score = −3.162), integration of energy metabolism (p=0.039, z-score = −2.449), and FOXO-mediated transcription pathways (p=0.0009, z-score = −2.000) (Figure 3B). Network analysis of the LA revealed significant alterations in proteins involved in metabolism and mitochondrial biogenesis (Figure 3C-E). IPA identified numerous proteins that collectively contribute to a predicted inhibition of fatty acid metabolism (p=3.34E⁻⁸, z-score −1.590) (Figure 3C). The downregulated proteins contributing to the reduction in fatty acid metabolism include fatty acid binding proteins (FABP4; log₂FC = −1.12 and FABP5; log₂FC = −0.92) and adiponectin (ADIPOQ; log₂FC = −0.15), which play roles in fatty acid uptake, storage (24), and oxidation (25). A collection of proteins predicted the inhibition of the uptake of glucose and quantity of carbohydrate while activating diabetes mellitus (Figure 3D). An additional cluster of proteins were predicted to inhibit mitogenesis (Figure 3E). Two isoforms of vascular endothelial growth factor (VEGF) were among the downregulated proteins involved in the inhibition of mitogenesis, which aside from their role in angiogenesis (26) have been shown to stimulate genes involved in mitochondrial biogenesis (27).

Despite targeted molecular assessments in Figures 1 and 2 failing to reveal metabolic dysfunction in the LV, IPA analysis interestingly revealed significant impairment of protein networks related to metabolism including upregulation of the mitochondrial dysfunction pathway (p=0.005, z-score = 2.985) and significant downregulation of pathways involved in mitochondrial function and energy metabolism including mitochondrial protein import (p=0.0006, z-score = −3.000), branched-chain amino acid catabolism (p=7.59E⁻⁶, z-score = −2.646), valine degradation I (p=6.76E⁻⁵, z-score = −2.449), oxidative phosphorylation (p=0.22, z-score = −2.449), mitochondrial protein degradation (p=0.001,z-score = −2.111), mitochondrial fatty acid beta-oxidation (p=0.04, z-score = −2.000), complex I biogenesis (p=0.25, z-score = −2.000), and fatty acid β-oxidation I pathways (p=0.04, z-score = −2.000) (Figure 4B). IPA also revealed a cluster of proteins associated with inhibition of PPARGC1A (encodes for PGC1α; z-score = −4.003, p=1.16E⁻⁸), a master regulator of mitochondrial biogenesis (28). This protein cluster was predicted to be associated with an increase in concentration of fatty acids and a decrease in both energy homeostasis and quantity of carbohydrate (Figure 4C). For example, the downregulation of mitochondrial transcription factor A (TFAM; log₂FC = −0.73) and citrate synthase (CS; log₂FC = −0.55) were two proteins that contributed to the increase in fatty acids concentration. TFAM is known to regulate mitochondrial DNA and maintain the electron transport chain (29), thus, a decrease in the protein could result in poor oxidative phosphorylation leading to increases in fatty acid concentration. Furthermore, as a key enzyme in the Krebs cycle and marker for mitochondrial content and function (30), the reduction of CS protein expression would suggest impaired mitochondrial function and reduced content, leading to an increase in metabolic dysfunction (Figure 4B) and additional increase in fatty acid concentration (Figure 4C).

DISCUSSION

In this study, we examine atrial versus ventricular differences for the first time in a swine model of HCM caused by heterozygous MYH7 R403Q mutation. Energetic impairment is a critical pathological feature of not only HCM, but HF in general. The heart is often described as a metabolic omnivore, with fuel metabolic flexibility key to maintaining energy production using various substrates based on availability i.e., fasted-fed states. Thus, shifts in cardiac metabolism that limit substrate utilization are a hallmark feature of HF, and specifically HCM (31). Under normal conditions, the bulk of ATP production in the heart comes from fatty acid metabolism (~85%) (32), which decreases in both systolic and diastolic HF (33). Studies with relevance to HCM have shown impairment to fatty acid, amino acid, ketone, glucose, and high energy phosphate metabolism in humans with HF with preserved ejection fraction (HFpEF) (34) and swine with compensated LV concentric hypertrophy (35, 36), further highlighting the concept of substrate inflexibility in HF.

Our results provide novel evidence of cardiac chamber-dependent functional and proteomic changes related to mitochondria dysfunction and impaired energy metabolism in an experimental setting of large animal HCM. To date, comprehensive studies of mitochondrial respiratory function have exclusively been conducted in mice (4), cats (37), and the LV (38) or septum from patients undergoing septal myectomy (15, 16) while failing to make atrial-ventricular comparisons. Interestingly, more recent human studies did not identify correlations linking mutation-specific effects to metabolic impairment (15, 16). In this study, the MYH7 R403Q mutation resulted in significant reductions in mitochondrial content and fatty acid oxidation in the LA, verified by global proteomics analysis using SOMAScan showing protein clustering associated with decreased mitogenesis (i.e., cell division) and fatty acid metabolism. Given the β-MHC isoform and by extension, the R403Q mutation, is predominantly expressed in the ventricle of large animals (39), it is perhaps somewhat counterintuitive to see the bulk of metabolic impairment in the atria which largely expresses the α-MHC (MYH6) isoform. However, in this animal model β-MHC levels in the atria increase from <10% overall to ~25–30% (40), implying significant shifts in atrial MYH isoform could play a role in the mitochondrial dysfunction observed. We also observed proteomic changes associated with reduced mitochondrial oxidative phosphorylation complexes and fatty acid oxidation in the LV, although changes in these protein networks preceded alterations at the functional level.

The observed decrease in atrial citrate synthase activity in the MYH7-HCM group indicates lower mitochondrial content and mass within the LA, potentially impairing oxidative capacity (41). This is in line with findings from healthy pigs showing lower basal activity of both citrate synthase and β-HAD activity in the LA compared to LV (42). These data were complemented by decreased protein expression of OXPHOS complexes II and III and protein networks associated with the downregulation of mitogenesis observed in the LA proteome of animals with a heterozygous MYH7 R403Q mutation. This downregulation in cell division indicates an impairment in the LA to make new cells, and subsequently could impact the mitochondria as the protein network included proteins within the VEGF family, which have previously been shown to stimulate mitochondrial biogenesis (27). Decreased β-HAD activity and inhibited fatty acid metabolism pathways in the proteome further suggests an impairment in the capacity of the LA to perform fatty acid oxidation in MYH7-HCM animals. Inhibition of fatty acid metabolism protein networks were driven by downregulated proteins including fatty acid binding proteins and adiponectin, which are heavily involved in fatty acid uptake, transport (24), and oxidation (25). Reductions in citrate synthase have previously been reported in a feline model of spontaneously occurring HCM (37) and in human patients with obstructive HCM (15) and end-stage HF (43), supporting impaired capacity for oxidative metabolism. However, one study found no differences in citrate synthase protein expression in patients with HCM (44). These previous studies assessed citrate synthase only in the LV, adding novelty to our work discovering significant reductions of citrate synthase activity in the LA of swine with the MYH7 R403Q mutation. Patients with HCM also showed additional metabolic impairments in the septum such as reduction of genes involved in fatty acid oxidation (including β-HAD) and decreased high-energy phosphate molecules (i.e., ADP, ATP, PCr) (15). In contrast to our findings, no differences occurred in right atrial citrate synthase and β-HAD activity in patients with atrial fibrillation, compared to individuals with normal sinus rhythm (45), indicating alterations in mitochondrial enzymatic activity differ across cardiac pathologies. Our results support previous observations of metabolic insufficiency in HCM using an approach that extends previous studies via assessment of the atria and ventricle, providing novel insight into potential chamber-dependent mitochondrial and metabolic pathogenesis in HCM.

Increased LDH protein observed in the atria of MYH7-HCM animals, considered in parallel with decreased citrate synthase and β-HAD activity, suggests a possible shift from primarily aerobic to anaerobic metabolism in the LA and support our interpretation that atrial mitochondria are becoming less efficient at producing ATP through OXPHOS. Pyruvate dehydrogenase and LDH are pertinent enzymes involved in anaerobic metabolism, catalyzing the irreversible conversion of pyruvate to acetyl-coenzyme A (46) and the reversible conversion of pyruvate to lactate (47), respectively. Considered together, the increase in atrial LDH protein observed in this study provides further evidence that oxidative capacity is impaired in the LA of swine with genetic HCM.

Interestingly, despite reductions in OXPHOS complexes and mitochondrial content, LA mitochondrial respiration was increased in the MYH7-HCM group. Reductions observed in mitochondrial enzymes, concomitant with increases in certain mitochondrial respiration states, suggests a compensatory mechanism exists in the LA at the functional level. Given decreases in enzymes involved in mitochondrial content and fatty acid metabolism were observed, persisting increases in respiration may be indicative that LA mitochondria are attempting to keep up with energy demand but are doing so with less mitochondria.

In contrast to the LA, functional data from the LV suggests that mitochondrial function remains normal. However, significant alterations to protein networks involved in mitochondrial function and fatty acid metabolism were observed in MYH7-HCM animals. Pathway analysis of the LV shows downregulation of several mitochondrial energetic pathways including oxidative phosphorylation, complex I biogenesis, and mitochondrial fatty acid β-oxidation, indicating that dysfunction of these pathways may be a precursor for reductions in functional mitochondrial outcomes. Moreover, IPA predicts the inhibition of mitochondrial proteins (i.e., TFAM, citrate synthase), leading to the activation of pathways that increase the concentration of fatty acids and concurrent inhibition of energy homeostasis and quantity of carbohydrates. Although these pathway changes in the LV do not align with our functional outcomes, they suggest shifts in the preferred source of energy substrate and impairment of mitochondrial fatty acid β-oxidation observed in the atria are certainly possible in the ventricle with further disease development. This shift represents diminished metabolic flexibility and is a well-known feature of pathological hypertrophy (31). Patients with obesity and HF may present with a greater accumulation of lipids (48), and our results indicate mitochondrial dysfunction resulting from impaired lipid metabolism could contribute to the etiology of HCM. Although mitochondria are more abundant in the LV compared to the LA, high resolution respirometry of human cardiac tissue indicate no differences in the bioenergetic profiles between chambers (49).

The current work is not without limitations. First, we included only male swine and findings may not be generalizable to females. While females are typically diagnosed later in life than males (50), the age of diagnosis between sexes did not differ in patients with MYH7 variants, suggesting lesser influence of sex in driving pathological outcomes of this specific mutation (51). Additionally, mitochondria were isolated from the whole LA and LV tissue, thus we cannot distinguish whether the mitochondria were distinctly from cardiomyocytes. Future studies should conduct cell-type specific mitochondria isolation for more precise assessment.

CONCLUSIONS

In summary, we examined mitochondrial function and global proteomic changes in both the LA and LV in a Yucatan mini-pig model of HCM caused by heterozygous MYH7 R403Q mutation. Our data indicate shifts in preferred substrate utilization use and energy production systems likely contribute to developing HCM and identify specific molecular mechanisms and protein networks that may support a basis for metabolic liability. While previous proteomic findings in humans have revealed dysregulated metabolic, inflammatory, and extracellular matrix signaling pathways (44, 52, 53), these studies were limited to examination of only the LV and septum. Expanding upon this existing foundation, our data demonstrate novel chamber-specific differences in mitochondrial dysfunction and protein networks using a preclinical swine model of HCM with high translational significance given β-MHC mutations in the heart contribute to approximately 1/3 of all familial HCM cases (3). Further, these data identify new molecular targets underlying the pathology of HCM and highlight their potential for mitochondria-targeted treatment for patients.

Data availability

Data supporting this study are available upon request.