The combinatorial effect of genetic variants is often assumed to be additive. Although genetic variation can clearly interact non-additively, methods to uncover epistatic relationships remain in their infancy. We develop low-signal signed iterative random forests to elucidate the complex genetic architecture of cardiac hypertrophy. We derive deep learning-based estimates of left ventricular mass from the cardiac MRI scans of 29,661 individuals enrolled in the UK Biobank. We report epistatic genetic variation including variants close to CCDC141, IGF1R, TTN, and TNKS. Several loci not prioritized by univariate genome-wide association analysis are identified. Functional genomic and integrative enrichment analyses reveal a complex gene regulatory network in which genes mapped from these loci share biological processes and myogenic regulatory factors. Through a network analysis of transcriptomic data from 313 explanted human hearts, we show that these interactions are preserved at the level of the cardiac transcriptome. We assess causality of epistatic effects via RNA silencing of gene-gene interactions in human induced pluripotent stem cell-derived cardiomyocytes. Finally, single-cell morphology analysis using a novel high-throughput microfluidic system shows that cardiomyocyte hypertrophy is non-additively modifiable by specific pairwise interactions between CCDC141 and both TTN and IGF1R. Our results expand the scope of genetic regulation of cardiac structure to epistasis.

Introduction: Atrial fibrillation (AF) is a common cardiac arrhythmia. Putative genes associated with AF risk have been identified through large genome wide association studies. How expression of these proteins in the left atria differs by AF state (no history of AF, history of paroxysmal AF, history of persistent AF, persistent AF, long-standing persistent AF) has not been examined. Hypothesis: We hypothesized that proteins associated with AF risk differ by AF state. Methods: Left atrial appendage (LAA) tissue was obtained from 222 Cleveland Clinic patients undergoing cardiac surgery. At surgery, 104 were in sinus rhythm (SR) (24 without a history of AF, 50 had a history of paroxysmal AF, 30 had a history of persistent AF) and 118 patients were in AF rhythm (65 with history of persistent AF, 53 long-standing persistent AF). Proteins (n=2539) were identified by mass spectrometry. Protein levels were modeled for associations with AF state using linear regression and adjusted for sex, age, and 23 inferred surrogate variables. Results: Among the quantified proteins we identified 33 encoded from putative AF risk genes. Of these, CASQ2 was increased (p<0.05) and TTN was decreased in patients with a history of AF in SR compared to patients with no history of AF (Table 1A). There were nine proteins that differed between AF rhythm and sinus rhythm, of which five (SYNPO2L, MYPN, NACA, PKP2, and CALU) were increased (Table 1B) and four (TUBA8, MYOZ1, CASQ2, and CAMK2D) were decreased. In patients diagnosed with persistent AF in AF rhythm versus those with a history of persistent AF but in SR at surgery, four were increased (SYNPO2L, MYPN, NACA, and PKP2) and two (CASQ2 and CAMK2D) were decreased (Table 1C). Expression of proteins encoded by putative AF risk genes was not different in long-standing persistent AF rhythm compared to persistent AF rhythm. Conclusions: In one of the largest proteomic datasets in human LAA, we found that key proteins encoded by genes associated with AF risk are altered in patients in AF rhythm. Expression of these proteins also differs by AF state. These studies provide insight into pathways that may be targeted for AF prevention and highlight the importance of early intervention to prevent AF progression.

Allosteric modulation is a central mechanism for metabolic regulation but has yet to be described for a gut microbiota-host interaction. Phenylacetylglutamine (PAGln), a gut microbiota-derived metabolite, has previously been clinically associated with and mechanistically linked to cardiovascular disease (CVD) and heart failure (HF). Here, using cells expressing β1- versus β2-adrenergic receptors (β1AR and β2AR), PAGln is shown to act as a negative allosteric modulator (NAM) of β2AR, but not β1AR. In functional studies, PAGln is further shown to promote NAM effects in both isolated male mouse cardiomyocytes and failing human heart left ventricle muscle (contracting trabeculae). Finally, using in silico docking studies coupled with site-directed mutagenesis and functional analyses, we identified sites on β2AR (residues E122 and V206) that when mutated still confer responsiveness to canonical β2AR agonists but no longer show PAGln-elicited NAM activity. The present studies reveal the gut microbiota-obligate metabolite PAGln as an endogenous NAM of a host GPCR. Allosteric modulation is crucial in metabolic regulation but unexplored in gut microbehost interactions. Here the authors show gut microbe-derived phenylacetylglutamine acts as a negative allosteric modulator of β2-adrenergic receptors, impacting heart function.

A specific genetic variant associated with atrial fibrillation risk, rs17171731, was identified as a regulatory variant responsible for controlling FAM13B expression. The atrial fibrillation risk allele decreases FAM13B expression, whose knockdown alters the expression of many genes in stem cell-derived cardiomyocytes, including SCN2B, and led to pro-arrhythmogenic changes in the late sodium current and Ca2+ cycling. Fam13b knockout mice had increased P-wave and QT interval duration and were more susceptible to pacing-induced arrhythmias vs control mice. FAM13B expression, its regulation, and downstream effects are potential targets for investigation of patient-specific therapeutics.

BACKGROUND:Genomewide association studies have associated >100 genetic loci with atrial fibrillation (AF), but establishing causal genes contributing to AF remains challenging.OBJECTIVE:The purpose of this study was to determine candidate novel causal genes and mechanistic pathways associated with AF risk loci by incorporating gene expression and coexpression analyses and to provide a resource for functional studies and targeting of AF-associated genes.METHODS:Cis-expression quantitative trait loci were identified for candidate genes near AF risk variants in human left atrial tissues. Coexpression partners were identified for each candidate gene. Weighted gene coexpression network analysis (WGCNA) identified modules and modules with overrepresentation of candidate AF genes. Ingenuity pathway analysis (IPA) was applied to the coexpression partners of each candidate gene. IPA and gene set over representation analysis were applied to each WGCNA module.RESULTS:One hundred sixty-six AF-risk single nucleotide polymorphisms were located in 135 loci. Eighty-one novel genes not previously annotated as putative AF risk genes were identified. IPA identified mitochondrial dysfunction, oxidative stress, epithelial adherens junction signaling, and sirtuin signaling as the most frequent significant pathways. WGCNA characterized 64 modules (candidate AF genes overrepresented in 8), represented by cell injury, death, stress, developmental, metabolic/mitochondrial, transcription/translation, and immune activation/inflammation regulatory pathways.CONCLUSION:Candidate gene coexpression analyses suggest significant roles for cellular stress and remodeling in AF, supporting a dual risk model for AF: Genetic susceptibility to AF may not manifest until later in life, when cellular stressors overwhelm adaptive responses. These analyses also provide a novel resource to guide functional studies on potential causal AF genes.

Glycogen Synthase Kinase 3β (GSK-3β) can modulate myofilament function in vitro . However, it’s in vivo role, mechanism, and translational relevance are unknown, which we investigated here using inducible cardiomyocyte specific GSK-3β KO mice. Compared to tamoxifen-treated GSK-3β fl/fl Cre- mice (Con), skinned myocytes from KO mice had reduced calcium sensitivity at long sarcomere lengths (SL = 2.2 μm), but there were no differences at short SL (~1.9 μm). Thus, myocytes from KO mice did not sensitize to calcium with stretch, a mechanism called length dependent activation (LDA) that underlies the organ-level Frank-Starling law. LDA has been attributed to (1) phosphorylation of myofilament proteins, (2) altered lattice-spacing, and (3) changes to titin’s elastic properties. Using mass spectrometry and small energy x-ray diffraction we ruled out the first two mechanisms, however we did find that GSK-3β KO myocytes had decreased passive tension – indicating the loss of LDA was due to loss of titin as a length sensor. Interestingly, immunofluorescence showed that GSK-3β localized to the z-disc when phosphorylated at Y216 and, via mass spectrometry, phosphorylated primarily z-disc proteins including several sites on the structural protein Ablim-1, which we showed also localizes to the z-disc. These data suggested that GSK-3β’s effect on titin is likely through altering its ability to anchor to the z-disc through targeting of these z-disc proteins. To provide further evidence that GSK-3β is specifically altering LDA, we genetically removed a downstream effecter of LDA, cardiac myosin binding protein-C (cMyBP-C). In vitro treatment with exogenous GSK-3β was able to increase calcium sensitivity at long SLs in both KO and WT mice but had no effect on cMyBPC KO mice or mice lacking the c-terminal domains of cMyBPC that are known to be important for LDA. Lastly, we found that human heart failure patients had less myofilament GSK-3β compared to non-failing patients, and that these same samples had a depressed LDA. This work has identified a novel mechanism by which GSK-3β localizes to the myofilament to modulate LDA and indicates that z-disc localized GSK-3β may be a possible therapeutic target to restore the Frank-Starling mechanism in heart failure patients.

The Sry-related high-mobility-group box (SOX) gene family, with 20 known transcription factors in humans, plays an essential role during development and disease processes. Several SOX proteins (SOX4, 11, and 9) are required for normal heart morphogenesis. SOX9 was shown to contribute to cardiac fibrosis. However, differential expression of other SOXs and their roles in the failing human myocardium have not been explored. Here, we used the whole-transcriptome sequencing (RNA-seq), gene co-expression, and meta-analysis to examine whether any SOX factors might play a role in the failing human myocardium. RNA-seq analysis was performed for cardiac tissue samples from heart failure (HF) patients due to dilated cardiomyopathy (DCM), or hypertrophic cardiomyopathy (HCM) and healthy donors (NF). The RNA levels of 20 SOX genes from RNA-seq data were extracted and compared to the 3 groups. Four SOX genes whose RNA levels were significantly upregulated in DCM or HCM compared to NF. However, only SOX4 and SOX8 proteins were markedly increased in the HF groups. A moderate to strong correlation was observed between the RNA level of SOX4/8 and fibrotic genes among each individual. Gene co-expression network analysis identified genes associated and respond similarly to perturbations with SOX4 in cardiac tissues. Using a meta-analysis combining epigenetics and genome-wide association data, we reported several genomic variants associated with HF phenotype linked to SOX4 or SOX8. In summary, our results implicate that SOX4 and SOX8 have a role in cardiomyopathy, leading to HF in humans. The molecular mechanism associated with them in HF warrants further investigation.

BACKGROUND:Heart failure with preserved ejection fraction (HFpEF) is a growing health problem without effective therapies. Epidemiological studies indicate that diabetes is a strong risk factor for HFpEF, and about 45% of patients with HFpEF are suffering from diabetes, yet the underlying mechanisms remain elusive.METHODS:Using a combination of echocardiography, hemodynamics, RNA-sequencing, molecular biology, in vitro and in vivo approaches, we investigated the roles of SIRT6 (sirtuin 6) in regulation of endothelial fatty acid (FA) transport and HFpEF in diabetes.RESULTS:We first observed that endothelial SIRT6 expression was markedly diminished in cardiac tissues from heart failure patients with diabetes. We then established an experimental mouse model of HFpEF in diabetes induced by a combination of the long-term high-fat diet feeding and a low-dose streptozocin challenge. We also generated a unique humanized SIRT6 transgenic mouse model, in which a single copy of human SIRT6 transgene was engineered at mouse Rosa26 locus and conditionally induced with the Cre-loxP technology. We found that genetically restoring endothelial SIRT6 expression in the diabetic mice ameliorated diastolic dysfunction concurrently with decreased cardiac lipid accumulation. SIRT6 gain- or loss-of-function studies showed that SIRT6 downregulated endothelial FA uptake. Mechanistically, SIRT6 suppressed endothelial expression of PPARγ through SIRT6-dependent deacetylation of histone H3 lysine 9 around PPARγ promoter region; and PPARγ reduction mediated SIRT6-dependent inhibition of endothelial FA uptake. Importantly, oral administration of small molecule SIRT6 activator MDL-800 to diabetic mice mitigated cardiac lipid accumulation and diastolic dysfunction.CONCLUSIONS:The impairment of endothelial SIRT6 expression links diabetes to HFpEF through the alteration of FA transport across the endothelial barrier. Genetic and pharmacological strategies that restored endothelial SIRT6 function in mice with diabetes alleviated experimental HFpEF by limiting FA uptake and improving cardiac metabolism, thus warranting further clinical evaluation.

Introduction Heart failure (HF) is an extremely debilitating condition that currently affects at least 16,500 people in Northern Ireland. Ischemic heart disease and cardiac fibrosis are the primary causes of end-stage HF. Greater understanding of molecular changes associated with this underlying pathophysiology could lead to the identification of novel biomarkers and therapeutic targets for improved diagnosis and management of HF. Methods Unbiased, deep proteomic analysis of individual left ventricular tissue samples from patients with HF (n=30) and patients without HF (NF; n=9) was performed using the diaPASEF workflow on a timsTOF Pro mass spectrometer. Validation of notable protein expression changes were performed by ELISA. Protein expression changes and correlations with clinical data were assessed using appropriate non-parametric testing of log-transformed data. Differentially expressed proteins were identified based on an observed fold change of ≥ 1.5 or ≤-1.5 and q-value ≤ 0.005. Results HF patients included patients with hypertrophic obstructive cardiomyopathy (HOCM; n=12), dilated cardiomyopathy (DCM; n=9) and ischemic cardiomyopathy (ISCM; n=9). One hundred and eighteen proteins were identified as being significantly associated with HF, irrespective of the underlying aetiology. Among these, a collagen sub-type, with a reported role in myocardial development, (referred to as ‘COL-CT’), was identified as being significantly elevated in HF (p=0.012), with greatest increase observed in patients with ISCM. Existing transcriptomic data for these samples corroborated these findings at gene level. Measurement of ‘COL-CT’ is more predictive of HF than combined measurement of the cardiac-specific collagen subtypes I and III (AUC 0.847 vs AUV 0.778). HF-associated changes in ‘COL-CT’ protein expression were validated in silico in an independent LV tissue dataset (n=7 NF v n=20 HF, p<0.001), and further confirmed in-house by ELISA-based analysis of serum samples from an independent patient cohort (n=50 NF v N=54 HF, p= 0.0004). ‘COL-CT’ is enriched in atrial regions of the heart and serum levels were found to be significantly positively correlated with left atrial volume (p<0.0001). Conclusions ‘COL-CT’ has yet to be fully investigated in the context of myocardial disease. Here we report a significant association between ‘COL-CT’ and HF. These observations have been validated in multiple independent clinical cohorts, in various sample types. Circulating levels of ‘COL-CT’ in serum provide evidence that ‘COL-CT’ may have clinical utility as a minimally invasive biomarker for HF. Moreover, strong association with ISCM and a likely role in cardiac fibrosis suggest that ‘COL-CT’ should be further investigated as a therapeutic target for management of HF.

BACKGROUND:Altered kinase localization is gaining appreciation as a mechanism of cardiovascular disease. Previous work suggests GSK-3β (glycogen synthase kinase 3β) localizes to and regulates contractile function of the myofilament. We aimed to discover GSK-3β's in vivo role in regulating myofilament function, the mechanisms involved, and the translational relevance. METHODS:Inducible cardiomyocyte-specific GSK-3β knockout mice and left ventricular myocardium from nonfailing and failing human hearts were studied. RESULTS:Skinned cardiomyocytes from knockout mice failed to exhibit calcium sensitization with stretch indicating a loss of length-dependent activation (LDA), the mechanism underlying the Frank-Starling Law. Titin acts as a length sensor for LDA, and knockout mice had decreased titin stiffness compared with control mice, explaining the lack of LDA. Knockout mice exhibited no changes in titin isoforms, titin phosphorylation, or other thin filament phosphorylation sites known to affect passive tension or LDA. Mass spectrometry identified several z-disc proteins as myofilament phospho-substrates of GSK-3β. Agreeing with the localization of its targets, GSK-3β that is phosphorylated at Y216 binds to the z-disc. We showed pY216 was necessary and sufficient for z-disc binding using adenoviruses for wild-type, Y216F, and Y216E GSK-3β in neonatal rat ventricular cardiomyocytes. One of GSK-3β's z-disc targets, abLIM-1 (actin-binding LIM protein 1), binds to the z-disc domains of titin that are important for maintaining passive tension. Genetic knockdown of abLIM-1 via siRNA in human engineered heart tissues resulted in enhancement of LDA, indicating abLIM-1 may act as a negative regulator that is modulated by GSK-3β. Last, GSK-3β myofilament localization was reduced in left ventricular myocardium from failing human hearts, which correlated with depressed LDA. CONCLUSIONS:We identified a novel mechanism by which GSK-3β localizes to the myofilament to modulate LDA. Importantly, z-disc GSK-3β levels were reduced in patients with heart failure, indicating z-disc localized GSK-3β is a possible therapeutic target to restore the Frank-Starling mechanism in patients with heart failure.

Atrial fibrillation (AF) risk is heritable. High rate electrical activity in AF requires increased energy. Atrial mitochondrial structure and function are altered in AF patients in an effort to generate adequate ATP through oxidative phosphorylation. Genomic studies have identified putative AF risk genes, but the association of AF risk genes with expression of mitochondrial genes is unclear. We tested the hypothesis that putative AF risk genes are co-expressed with mitochondrial genes that play a role in atrial energy production. RNA-seq was performed on left atrial appendage (LAA) tissues obtained from 251 cardiac surgery patients. RNA coexpression profiles were evaluated for 222 putative AF risk genes. Genes encoding proteins that localize to the mitochondria were identified using MitoCarta 2.0. Changes in metabolic pathways were detected using Ingenuity Pathway Analysis (IPA). Our analysis identified 128 AF risk genes that coexpressed with at least one mitochondrial gene. The highest level of mitochondrial gene coexpression was evident with PCCB, in which 30% (253 of 848) of coexpressed genes were mitochondrial. CASQ2 (24%, 104 of 431) and ASAH1 (20%, 37 of 182) also showed high levels of mitochondrial gene coexpression. The IPA Oxidative Phosphorylation Pathway was significantly altered (p<0.05) for 26 AF risk genes, with CASQ2 (9.42E-79), MYH6 (4.73E-78), YWHAE (4.96E-75), and TTN (4.30E-71) the most strongly associated (Table). Additionally, 12 AF risk genes coexpressed with genes encoded by mitochondrial DNA (mtDNA) (Table). ASAH1, CASQ2, MYH6, NACA, NUCKS1, PLN, TTN, and YWHAE coexpressed with all 13 mtDNA encoded components of the electron transport chain. Many AF risk genes show significant coexpression with mitochondrial genes. We propose that genetic risk scores based on these AF risk genes may identify a subset of AF patients that would benefit from AF therapies that enhance metabolic activity.

Introduction: Atrial fibrillation (AF) increases energy demand for contractile and electrical activity. Changes in left atrial (LA) protein expression of AF patients are poorly characterized. Hypothesis: Mitochondrial protein expression in patients with AF is altered in an attempt to meet increased energy demand. Methods: LA appendage tissue was obtained from 198 patients undergoing Maze surgery. At the time of surgery, 80 were in sinus rhythm (SR) (50 paroxysmal AF, 30 persistent AF) and 118 in AF (65 persistent AF, 53 permanent AF). Protein content was assessed by mass spectrometry and 2539 proteins were identified. Results: In AF compared to SR, 257 proteins were differentially expressed (q<0.05); 44 of 62 mitochondrial proteins detected (MitoCarta 3.0) were increased. KEGG pathway analysis revealed Oxidative Phosphorylation was increased (p=1.01E-3) in AF, including 17 subunits of the electron transport chain (A). The Hypertrophic Cardiomyopathy pathway was decreased (p=1.01E-03). Expression of ryanodine receptor, and troponin and tropomyosin subunits decreased, but tropomyosin 4 and myosin heavy chain 9 and 10 increased (B), providing evidence of changes in myofibrillar and calcium regulatory proteins in AF. Among 39 putative AF risk genes detectable at the protein level, 8 were altered in AF (C). The Tricarboxylic Acid Cycle KEGG pathway was decreased (p=2.55E-3) in patients with permanent compared to persistent AF, with no significant changes in other cellular pathways or protein expression of putative AF risk genes. Conclusions: In one of the largest proteomic datasets in human LA to date, we find that expression of proteins in metabolic, myofibrillar, and calcium regulation pathways is altered in patients with AF. Additional metabolic changes were detected with progression to permanent AF. These data identify proteins that are altered in patients with AF providing insight into cellular pathways that may be targeted for AF prevention and therapy.

Background: Long noncoding RNAs (lncRNAs) have been implicated in the pathogenesis of cardiovascular diseases. We aimed to identify novel lncRNAs associated with the early response to ischemia in the heart. Methods and Results: RNA sequencing data gathered from 81 paired left ventricle samples from patients undergoing cardiopulmonary bypass was collected before and after a period of ischemia. Novel lncRNAs were validated with Oxford Nanopore Technologies long-read sequencing. Gene modules associated with an early ischemic response were identified and the subcellular location of selected lncRNAs was determined with RNAscope. A total of 2446 mRNAs, 270 annotated lncRNAs and one novel lncRNA differed in response to ischemia (adjusted p < 0.001, absolute fold change >1.2). The novel lncRNA belonged to a gene module of highly correlated genes that also included 39 annotated lncRNAs. This module associated with ischemia (Pearson correlation coefficient = −0.69, p = 1 × 10−23) and activation of cell death pathways (p < 6 × 10−9). A further nine novel cardiac lncRNAs were identified, of which, one overlapped five cis-eQTL eSNPs for the gene RWD Domain-Containing Sumoylation Enhancer (RWDD3) and was itself correlated with RWDD3 expression (Pearson correlation coefficient −0.2, p = 0.002). Conclusion: We have identified 10 novel lncRNAs, one of which was associated with myocardial ischemia and may have potential as a novel therapeutic target or early marker for myocardial dysfunction.

Myofilament BAG3 expression decreases in male patients with nonischemic DCM but is preserved in female patients with DCM. BAG3 expression in the human heart is tightly linked to HSF-1 expression and nuclear translocation. HSF-1 localizes to the sarcomere Z-disc in the human heart. HSF-1 expression in the myofilament fraction decreases in male patients with DCM and positively correlates with myofilament BAG3.

Introduction The purpose of this study was to gain greater understanding of the pathogenesis of hypertrophic obstructive cardiomyopathy (HOCM), dilated cardiomyopathy (DCM) and ischemic cardiomyopathy (ISCM). These conditions lead to heart failure (HF) and the prognosis for HF differs based on the underlying aetiology. Cardiac tissue represents a challenging sample from the proteomics perspective due to the dominant signal from of a small number of high abundance proteins. Thus, a diaPASEF [1] workflow was applied in order to achieve deep quantitative coverage of cardiac tissue from HOCM (n=12), DCM (n=9), ISCM (n=9) and age/sex matched controls (NF, n=9). RNA-seq analysis has already been performed on these samples. Methods Unbiased, deep proteomic analysis of individual samples was performed using the diaPASEF workflow on a timsTOF Pro mass spectrometer. Analysis of high pH-reversed phase fractionated sample pools was performed in ddaPASEF mode to generate spectral library data. Raw data files were processed through Spectronaut 14 software for spectral library building, protein identification and quantification. Differentially expressed proteins were identified based on an observed fold change of ≥ 1.5 or ≤-1.5 and q-value ≤ 0.005. Pathway analysis was performed using Ingenuity Pathway Analysis (IPA) software. Results and Conclusions/Implications Label-free MS analysis led to over 4,000 protein identifications, with 3,484 proteins commonly identified across all patient samples. Over 1,000 significantly differentially expressed protein candidates were identified for comparisons between NF and DCM, HOCM or ISCM. DCM-specific protein changes were strongly associated with glutamine biosynthesis, HOCM-specific protein changes were strongly associated with LXR/RXR Activation, while ISCM-specific protein changes were most associated with tryptophan degradation pathways. DCM vs NF, ISCM vs NF and HOCM v NF had shared differentially expressed proteins that were also significantly altered at gene level (n=106). Canonical pathway analysis revealed that Choline Degradation and Lysine Degradation pathways were most strongly associated with these candidates. Expression changes for some of the top over- and under-expressed HF candidates were validated in an independent replicate dataset (PXD008934) [2]. This represents one of the largest and deepest proteomic datasets for myocardial tissue reported to date. The dataset, which compliments existing transcriptomic data for these samples, has highlighted a number of significant proteins associated with different underlying aetiologies of HF. Prognosis for HF differs depending on the aetiology from which it arises. Hence, the dataset here will help in further understanding the pathogenesis of the disease, leading towards more personalised treatment. Conflict of Interest N/A