key: cord-102608-1ilforzm authors: Litviňuková, Monika; Talavera-López, Carlos; Maatz, Henrike; Reichart, Daniel; Worth, Catherine L.; Lindberg, Eric L.; Kanda, Masatoshi; Polanski, Krzysztof; Fasouli, Eirini S.; Samari, Sara; Roberts, Kenny; Tuck, Liz; Heinig, Matthias; DeLaughter, Daniel M.; McDonough, Barbara; Wakimoto, Hiroko; Gorham, Joshua M.; Nadelmann, Emily R.; Mahbubani, Krishnaa T.; Saeb-Parsy, Kourosh; Patone, Giannino; Boyle, Joseph J.; Zhang, Hongbo; Zhang, Hao; Viveiros, Anissa; Oudit, Gavin Y.; Bayraktar, Omer; Seidman, J. G.; Seidman, Christine; Noseda, Michela; Hübner, Norbert; Teichmann, Sarah A. title: Cells and gene expression programs in the adult human heart date: 2020-04-10 journal: bioRxiv DOI: 10.1101/2020.04.03.024075 sha: doc_id: 102608 cord_uid: 1ilforzm Cardiovascular disease is the leading cause of death worldwide. Advanced insights into disease mechanisms and strategies to improve therapeutic opportunities require deeper understanding of the molecular processes of the normal heart. Knowledge of the full repertoire of cardiac cells and their gene expression profiles is a fundamental first step in this endeavor. Here, using large-scale single cell and nuclei transcriptomic profiling together with state-of-the-art analytical techniques, we characterise the adult human heart cellular landscape covering six anatomical cardiac regions (left and right atria and ventricles, apex and interventricular septum). Our results highlight the cellular heterogeneity of cardiomyocytes, pericytes and fibroblasts, revealing distinct subsets in the atria and ventricles indicative of diverse developmental origins and specialized properties. Further we define the complexity of the cardiac vascular network which includes clusters of arterial, capillary, venous, lymphatic endothelial cells and an atrial-enriched population. By comparing cardiac cells to skeletal muscle and kidney, we identify cardiac tissue resident macrophage subsets with transcriptional signatures indicative of both inflammatory and reparative phenotypes. Further, inference of cell-cell interactions highlight a macrophage-fibroblast-cardiomyocyte network that differs between atria and ventricles, and compared to skeletal muscle. We expect this reference human cardiac cell atlas to advance mechanistic studies of heart homeostasis and disease. The heart is a complex organ, composed of four morphologically and functionally distinct chambers ( Figure 1A ), that perpetually pumps blood throughout our lives. Deoxygenated blood enters the right atrium and is propelled into the low pressure vascular beds of the lungs by the right ventricle. Oxygenated pulmonary blood enters the left atrium and then to the left ventricle, which propels blood across the body at systemic vascular pressures. Right and left chambers are separated by the atrial and interventricular septa and unidirectional flow is established by the atrio-ventricular valves (tricuspid and mitral) and ventricular-arterial valves (pulmonary and aortic). The heart contains an intrinsic electrophysiologic system, composed of the sinoatrial node in the right atria where depolarization begins and spreads to the atrioventricular node located at the top of the interventricular septum. This electrical impulse is then rapidly propagated by Purkinje fibers to the apex where contraction begins. Orchestration of the anatomical and functional complexity of the heart requires highly organized and heterogeneous cell populations that enable continuous contraction and relaxation across different pressures, strains, and biophysical stimuli in each chamber. The importance of these variables is reflected in the differences in wall thickness and mass (left, 116±20g; right 84±14g) of adult ventricular chambers 1 . Specialized properties of cells that enable adaptation to different biophysical stimuli in each chamber are established early in development. The heart is derived from multipotent progenitor cells residing within two heart fields. Cells of the first heart field primarily populate the left ventricle and second heart field-derived cells populate the right ventricle; both heart fields contribute to atrial cells. The distinct gene regulatory networks operant in these heart fields likely establish and prime the patterns of gene expression observed in adult cardiac cells which are further impacted by the establishment of postnatal circulation 2 . The cellular composition of the adult human heart, their anatomical specificities, molecular signatures, intercellular networks and spatial relationships between the various cardiac cells remain largely unknown. Single-cell and single-nuclei transcriptomics (scRNA-Seq, snRNA-seq) and multiplex smFISH imaging now enable us to address these issues at unprecedented resolution 3 . These technologies illuminate the coordinated communication of cells within their microenvironments that in the heart enable electromechanical connectivity, biophysical interactions, and autocrine/paracrine signaling required for tissue homeostasis but are perturbed in disease. 3 While previous studies using a combination of conventional bulk genomics and microscopy have hinted at the cellular complexity of the myocardium, limitations of these techniques have allowed definition of very few distinct cell populations 4 . Bulk RNA-seq analysis is unable to assign gene expression to defined cell subpopulations, light microscopy fails to define features beyond morphology of cell subpopulations and immunostainings are limited to the analysis of few markers at once. Moreover, the large size of cardiomyocytes (length/width:~100/25µM) limits the unbiased capture of single cells requiring analyses of single nuclei transcriptomics to ensure a comprehensive approach. Here, we present a broad transcriptomic census of multiple regions of the adult human heart. We profiled RNA expression of both single cells and nuclei, capturing them from six distinct cardiac anatomical regions. We also analysed the spatial distribution of selected cell populations using multiplex smFISH imaging with RNAscope probes. Our anatomically defined resolved adult human heart cell census provides a reference framework for studies directed towards understanding the cellular and molecular drivers that enable functional plasticity in response to varying physiological conditions in the normal heart, and will inform the heart's responses to disease. Overview of the cellular landscape of the adult human heart Samples were obtained from six cardiac regions including the free wall of each chamber (left/right ventricle, left/right atrium), denoted as LV, RV, LA, RA, and from the LV apex (AX) and interventricular septum (SP). To capture the heterogeneity of cardiac cell populations, samples were collected as transmural tissue segments that span the three cardiac layers (epicardium, myocardium and endocardium; Figure 1A ) from 14 normal hearts (seven females, seven males) of North American and British organ donors (ages 40-75 years; Figure 1B and Supplementary Table A1 ). We isolated single cells and single nuclei, as the large sizes of cardiomyocyte (CM) are not captured by the 10X Genomics Chromium platform. Fresh tissues were mechanically and enzymatically processed to dissociate single cells, and subsequently cardiac immune cells were enriched from the cell fraction using CD45+ magnetic selection. Single nuclei were isolated from frozen heart tissues and purified by fluorescent activated cell sorting. The transcriptome of single cells and nuclei were profiled using the 10X Genomics Single Cell Gene Expression Solution ( Figure 1A ). After processing, all data from nuclei, cells and CD45+ enriched cells were batch-aligned using a generative deep variational autoencoder 5 , prior to unsupervised clustering. We found differences in the distribution of cell types across donors, even within the same region, and correlations between different cell types at the same site ( Supplementary Table A3 ). For example, in LV, AX, SP and RV tissues, the proportions of vCM and FB were negatively correlated ( p -value=2.0e-4), while there was a positive correlation among the proportions of PC, SMC, and NC cells ( p -value<1e-4). We suggest that these data reflect random sampling that included vessels with EC, PC, SMC, and concurrently fewer CM. However, the observed correlation between NC, SMC and PC implies a potential functional organization. The cell distributions were generally similar in tissues of male and female hearts, but left ventricular regions (AX, SP, LV) from female donors had significantly higher mean percentages of CM (46.7±12%) compared to male donors (33.9±10%; p -value=0.008). This is unexpected given the average smaller heart mass of women and may reflect the small donor pool. If replicated, these data may explain higher cardiac stroke volume in women 6 and lower rates of cardiovascular disease. Table B1 ), including HCN1 , MYH6 , MYL4 and NPPA . In addition, we identified a higher aCM expression of ALDH1A2 (9-fold increased), the catalytic enzyme required for synthesis of retinoic acid, that may reflect aCM derived from the second heart field 9 . aCM also showed higher levels of ROR2 (22-fold increased), which participates in cardiomyocyte differentiation via Wnt-signaling, as well as SYNPR (29-fold increased), a synaptic vesicle membrane protein with functions in TRP-channel mechanosensing by atrial volume receptors 10 , 11 . In contrast, vCM expressed 79 genes with 3-fold higher expression than aCM ( Supplementary Table B1 ). Genes with significantly enriched expression included 7 prototypic sarcomere protein genes MYH7 , MYL2 , and transcription factors ( IRX3, IRX5, IRX6, MASP1, HEY2 ). vCM also had 20-fold higher expression of PRDM16 than aCM, which harbors damaging variants associated with LV non-compaction 12 . Similarly highly expressed were PCDH7 ( Supplementary Figure B1A ,B ), a molecule with strong calcium-dependent adhesive properties and SMYD2 , which promotes the formation of protein stabilizing complexes in the Z-disc and I-band of sarcomeres 13 , 14 . Expression of these genes is likely to promote tissue integrity under the conditions of high ventricular pressure and strain. Clustering of vCM data identified five subpopulations ( Figure 2A We also identified two subpopulations (vCM3 and vCM4) across all ventricular regions. The transcriptional profile of vCM3 was remarkably similar to a prominent RA subpopulation (aCM3, discussed below), and suggestive that these are derived from the second heart field 18 . These cells had higher levels of transcripts associated with retinoic-acid responsive smooth muscle cell genes, including MYH9 , CNN1 19, 20 , and NEXN . vCM3 also expressed stress-response genes including ANKRD1 21 , FHL1 22 , DUSP27 23 , XIRP1 and XIRP2. The XIRP proteins interact with cardiac ion channel proteins Nav1.5 and Kv1.5 within intercalated 8 discs, and have been implicated in lethal cardiac arrhythmias prevalent in cardiomyopathies 26, 24 . vCM4 contributed 6-10% to vCM populations, and expressed nuclear-encoded mitochondrial genes ( NDUFB11 , NDUFA4 , COX7C , and COX5B ; Figure 2E ) suggestive of a high energetic state. Indeed, Gene Ontology analyses of vCM4 transcripts identified significant terms of "ATP metabolic process" and "oxidative phosphorylation" ( Supplementary Figure B1C ). These CM also had high levels of CRYAB encoding a heat shock protein with cytoprotective roles and antioxidant responses by CM 25 . With c oncomitant high expression of genes encoding sarcomere components ( Figure 2E ) and PLN ( Supplementary Table B3 ) , we deduced that these vCM are outfitted to perform higher workload than other vCM. Unlike scRNA-Seq analysis that identified prominent RV expression of PLN in embryonic mouse hearts 26 , vCM4 were similar in both ventricles. A small proportion (~1%) of cells comprised vCM5 and expressed high levels of DLC1 and EBF2 27 . These molecules participate in regulating brown adipocyte differentiation and may be involved in cardiac pacemaker activity. In addition, vCM5 nuclei had higher levels of transcripts also expressed in neural lineages ( SOX5, EBF1, and KCNAB1 ) . Notably, mice with deleted EBF1 have profound hypoplasia of the ventricular conduction system 28 . Further confirmation and investigation of this subpopulation is required, given the small number of cells and shared marker genes with other cell types. We identified six subpopulations of aCM, indicating considerable heterogeneity ( Figure 2B ) particularly between the right and left chambers. Notably, HAMP , a master regulator of iron homeostasis, was significantly enriched in over 50% of RA CM compared to 3% LA CM ( Supplementary Table B4 ) 29 , consistent with prior studies of RA tissues 8 and confirmed by smFISH ( Figure 2G ). HAMP has unknown roles in cardiac biology, but Hamp-null mice have electron transport chain deficits and lethal cardiomyopathy 30 . The RA enrichment of HAMP may imply energetic differences between right and left aCM. ligand for ROBO receptors in the heart 31 , ALDH1A2 9 and BRINP3, involved in retinoic acid signaling, and GRXCR2, a molecule that supports cilia involved in mechanosensing 32 . As noted above, aCM3 shared a remarkably similar transcriptional profile with vCM3 including the smooth muscle cell gene CNN1 , which we confirmed by smFISH with RNAscope probes ( Figure 2G, Supplementary Figure B1A ). The transcriptional profiles of aCM2, aCM3 and vCM3 likely indicate their derivation from the second heart field that forms the right heart chambers and associated vascular structures 18 . We captured a small subpopulation of mesothelial cells expressing MSLN , WT1 and BNC1 , but not EC, FB or mural lineage genes, indicating that these are likely epicardial cells 45 , 46,47 . Table C2 ). This is in line with in vitro models predicting that while arterial SMC are more contractile, the venous seem more dedifferentiated and potentially proliferative 52 . EC_art and SMC_art ( Figure 3C, D ) . We also predicted differences between the Figure C1 ). Transdifferentiation of stromal populations has been described but remains controversial in the literature 54 55 56 . We Immune cells are known to play key roles in both cardiac homeostasis, as well as inflammation, repair and remodeling; however, they were unrepresented in our single nuclei dataset. Therefore, we used the pan-immune marker CD45 to enrich for this cell population. We defined cardiac-resident versus circulating cells by calculating the enrichment in To evaluate cell -cell interactions of the immune cells in cardiac homeostasis and remodelling we used cellphonedb 66 . Thus we predicted putative cross-talk among myeloid and lymphoid cells, CM, and FB ( Figure 4C and Supplementary Table E3 ) 64 . This analysis predicted that DC and MØ_trTMSB4X+ interacted with FB3 in a distinct manner in atria and ventricles, as shown in Figure 4D . The FB3 subpopulation signaled to aCM and vCM in turn, forming a cellular circuitry which may be relevant for healthy cardiac homeostasis. 18 The heart is innervated by both sympathetic and parasympathetic components of the autonomic nervous system, which reside in ganglionated plexuses on the epicardial surface and contribute to regulation of heart rhythm 67 . Neural cells (NC) innervate the sinoatrial and atrioventricular nodes, from which an activating wave-front propagates throughout myocardial tissue via nerve bundles and Purkinje fibers. We identified 3485 NC from cardiac tissues, predominantly (60%) from left chambers, presumably due to multiple left ventricular sites. All NC expressed transcripts ( NRXN1, NRXN3, KCNMB4 ) typically found in the central nervous and cardiac conduction system including sodium and potassium channels 68 ( Supplementary Table F1 ) . LGI4 , which is required for glia development and axon myelination 69 . NC2 was predominantly derived from atrial tissues and expressed GRIA1 , encoding glutamate receptor 1 which has altered expression levels in ischemic heart disease 70 . Additionally, GINS3 is enriched in these cells, which is known to participate in the regulation of cardiac repolarization 71, 72 . NC3 and NC4 subpopulations shared a broadly similar transcriptional profile including LGR5 , a G-protein-coupled receptor involved in Wnt -signaling that promotes CM differentiation 73 and demarcates a subset of CM in the outflow tract 74 ; this region often becomes arrhythmogenic in heart disease 75 . These cells also expressed genes associated with coronary artery disease, PPP2R2B 76 , LSAMP 77 and LPL an endothelial enzyme involved transporting lipoproteins into the heart 78 . Additional smaller subpopulations (NC5-10) shared marker genes with other cell types and required further analyses to define their identities. We CellPhoneDB.org predicted different interactions among arterial ECs and skeletal or heart muscle cells ( Figure 5A ). While NOTCH interactions were shared by both, EC-PC/SMC interactions in the heart were predicted to involve JAG1/2 NOTCH3 and NOTCH1 JAG1 but not in skeletal muscle. Additionally, the EFNA1-EPHA3 receptor-ligand complex pair was consistently present in the heart but not detected in skeletal muscle. These differences may reflect distinct microvasculatures in the two tissues due to different oxygen requirements. We also compared the immune compartment of heart, skeletal muscle and adult kidney 79 We suggest that these tissue residents populations have developed transcriptional circuits tailored to the heart that differ from other tissues ( Figure 5B , 5C and Supplementary Table G4 ). Myeloid cells in the skeletal muscle are known to interact with satellite cells to promote myogenesis 80 . As we did not find a cardiac cell population analogous to skeletal muscle satellite cells in the heart, we considered if the heart had other potential repair mechanisms. To identify these we compared predicted interactions ( Supplementary Table G5 ) among immune, fibroblast and myocyte populations in heart and skeletal muscle. These analyses showed that cardiac FB3 secrete FN1 and TNC which is bound by CM-expressed integrins. In skeletal muscle, predicted cell-cell interactions between PRG4+ fibroblasts and skeletal muscle myocytes involved COL1A2 , COL6A2 and a10b1 integrins ( Figure 5D ). Skeletal muscle fibroblasts and monocytes appear to interact via the CXCR4_CXCL12 chemokine, while cardiomyocytes have a distinct interaction with macrophages as described above in Figure 4D . Altogether, these results imply interactive mechanisms are driven by different transcriptional circuits in heart and skeletal muscle. The circuit between CXCR4_CXCL12, which has been described to promote repair after myocardial infarction 81 appears to be primed by myeloid populations that initiate fibrotic repair. 22 FB in atrial and ventricular tissues exhibited different transcriptional profiles and subpopulations, suggesting distinct functions. We found that ECM-producing and -organising fibroblasts, while present in atria and ventricles, differed in their mode of action, as exemplified by regional-specific expression of different collagen and ECM remodelling factor genes. Together with differences in the frequency of other cell types, we suggest that region-specific FB heterogeneity is critical to support CM across varying biophysical stimuli. 86 . Related corona viral infections (SARS and MERS) also had cardiac involvement. Here, we found that expression of the viral receptor ACE2 is higher in pericytes than in CM, as previously reported 87 , but also that neither pericytes nor CM express the protease that prime viral entry, TMPRSS2 . Instead CM and pericytes express CTSB and CTSL , which may also promote viral entry. ACE2 expression correlated with AGTR1 (angiotensin receptor-1) which was highest in pericytes, and consistent with its role of renin-angiotensin-aldosterone system (RAAS) signalling in cardiac hemodynamics. ACE2 cleaves the vasoconstrictor angiotensin II that binds AGTR. ACE2 -null mice have reduced cardiac contractility, myocardial ischemia and hypoxia 88 , 89 suggesting a profound role of ACE2 in the regulation of cardiovascular hemodynamics. We expect that our results will allow further insights into other cardiac disease processes. Going forward, our results will furthermore be of value for deconvolution of existing bulk transcriptomic data, for transfer learning in analyses of other cardiac regions (valves, papillary muscle, conduction system), and to enable interpretation of the cellular responses to human heart disease. All of our data can be explored at www.heartcellatlas.org . This publication is part of the Human Cell Atlas -www.humancellatlas.org/publications . We Normal values of left ventricular mass and cardiac chamber volumes assessed by 320-detector computed tomography angiography in the Copenhagen General Population Study The deployment of cell lineages that form the mammalian heart Single Cell Gene Expression to Understand the Dynamic Architecture of the Heart scGen predicts single-cell perturbation responses Women Have Higher Left Ventricular Ejection Fractions Than Men Independent of Differences in Left Ventricular Volume Differential gene expressions in atrial and ventricular myocytes: insights into the road of applying embryonic stem cell-derived cardiomyocytes for future therapies Regional Differences in mRNA and lncRNA Expression Profiles in Non-Failing Human Atria and Ventricles Human Pluripotent Stem Cell-Derived Atrial and Ventricular Cardiomyocytes Develop from Distinct 29 Distinctive Roles of Canonical and Noncanonical Wnt Signaling in Human Embryonic Cardiomyocyte Development Expression of transient receptor potential channels TRPC1 and TRPV4 in venoatrial endocardium of the rat heart Cardiomyopathy due to PRDM16 mutation: First description of a fetal presentation, with possible modifier genes Non-clustered protocadherin Smyd2 controls cytoplasmic lysine methylation of Hsp90 and myofilament organization Conserved expression of the PRELI domain containing 2 gene (Prelid2) during mid-later-gestation mouse embryogenesis T-cadherin is critical for adiponectin-mediated cardioprotection in mice CDH13 promoter SNPs with pleiotropic effect on cardiometabolic parameters represent methylation QTLs Single-Cell RNA-Seq of the Developing Cardiac Outflow Tract Reveals Convergent Development of the Vascular Smooth Muscle Cells Calponin isoforms CNN1, CNN2 and CNN3: Regulators for actin cytoskeleton functions in smooth muscle and non-muscle cells Smooth muscle dysfunction in resistance arteries of the staggerer 30 The enigmatic role of the ankyrin repeat domain 1 gene in heart development and disease Fhl-1, a new key protein in pulmonary hypertension H19 lncRNA Promotes Skeletal Muscle Insulin Sensitivity in Part by Targeting AMPK Critical Roles of Xirp Proteins in Cardiac Conduction and Their Rare Variants Identified in Sudden Unexplained Nocturnal Death Syndrome and Brugada Syndrome in Chinese Han Population Alpha B-crystallin induction in skeletal muscle cells under redox imbalance is mediated by a JNK-dependent regulatory mechanism. Free Radic Single-cell analysis of cardiogenesis reveals basis for organ-level developmental defects EBF2 determines and maintains brown adipocyte identity AHA Late-Breaking Basic Science Abstracts New Insights into the Hepcidin-Ferroportin Axis and Iron Homeostasis in iPSC-Derived Cardiomyocytes from Friedreich's Ataxia Patient An essential cell-autonomous role for hepcidin in cardiac iron homeostasis Expression of Slit and Robo genes in the developing mouse heart GRXCR2 Regulates Taperin Localization Critical for Stereocilia Morphology and Hearing Single-cell reconstruction of the adult human heart during heart failure and recovery reveals the cellular landscape underlying cardiac function Liprin-mediated large signaling complex organization revealed by the liprin-α/CASK and liprin-α/liprin-β complex structures SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor Single-Cell Transcriptome Atlas of Murine Endothelial Cells Signaling pathways in the specification of arteries and veins Single-Cell Transcriptome Analyses Reveal Endothelial Cell Heterogeneity in Tumors and Changes following Antiangiogenic Treatment Resolving the fibrotic niche of human liver cirrhosis at single-cell level A high endothelial venule-expressing promiscuous chemokine receptor DARC can bind inflammatory, but not lymphoid, chemokines and is dispensable for lymphocyte homing under physiological conditions Hippo Signaling Plays an Essential Role in Cell State Transitions during Cardiac Fibroblast Development Role of secreted modular calcium-binding protein 1 (SMOC1) in transforming growth factor β signalling and angiogenesis Genetic Fate Mapping Defines the Vascular Potential of Endocardial Cells in the Adult Heart A selective marker of normal and neoplastic mesothelial cells in serous effusions Adult mouse epicardium modulates myocardial injury by secreting paracrine factors BNC1 regulates cell heterogeneity in human pluripotent stem cell-derived epicardium The NG2 Proteoglycan in Pericyte Biology A molecular atlas of cell types and zonation in the brain vasculature Anatomically and Functionally Distinct Lung Mesenchymal Populations Marked by Lgr5 and Lgr6 Regulator of G-Protein Signaling 5 Prevents Smooth Muscle Cell Proliferation and Attenuates Neointima Formation In vitro differences between venous and arterial-derived smooth muscle cells: potential modulatory role of decorin It Takes Two: Endothelial-Perivascular Cell Cross-Talk in Vascular Development and Disease Endothelial cells are progenitors of cardiac pericytes and vascular smooth muscle cells Pericytes, an overlooked player in vascular pathobiology Pericytes of Multiple Organs Do Not Behave as Mesenchymal Stem Cells In Vivo Calderone Angelino & Nattel Stanley. Differential Behaviors of Atrial Versus Ventricular Fibroblasts Differences in atrial versus ventricular remodeling in dogs with ventricular tachypacing-induced congestive heart failure Delineating an oncostatin M-activated STAT3 signaling pathway that coordinates the expression of genes involved in cell cycle regulation and extracellular matrix deposition of MCF-7 cells Transforming growth factor β1 (TGFβ1) regulates CD44V6 expression and activity through extracellular signal-regulated kinase (ERK)-induced EGR1 in pulmonary fibrogenic fibroblasts Macrophage hypoxia signaling regulates cardiac fibrosis via Oncostatin M Self-renewing resident cardiac macrophages limit adverse remodeling following myocardial infarction Progranulin attenuates liver fibrosis by downregulating the inflammatory response Reappraising the role of inflammation in heart failure Novel Role of Lck in Leptin-Induced Inflammation and Implications for Renal Aging CellPhoneDB: inferring cell-cell communication from combined expression of multi-subunit ligand-receptor complexes The autonomic nervous system and cardiac arrhythmias: current concepts and emerging therapies Development of the Cardiac Conduction System LGI proteins in the nervous system Molecular remodeling of ion channels, exchangers and pumps in atrial and ventricular myocytes in ischemic cardiomyopathy Common variants at ten loci influence QT interval duration in the QTGEN Study Drug-sensitized zebrafish screen identifies multiple genes, including GINS3, as regulators of myocardial repolarization Downregulation of LGR5 Expression Inhibits Cardiomyocyte Differentiation and Potentiates Endothelial Differentiation from Human Pluripotent Stem Cells Population and Single-Cell Analysis of Human Cardiogenesis Reveals Unique LGR5 Ventricular Progenitors in Embryonic Outflow Tract Outflow tract ventricular arrhythmias: An update Fine mapping of a linkage peak with integration of lipid traits identifies novel coronary artery disease genes on chromosome 5 Polymorphisms of the tumor suppressor gene LSAMP are associated with left main coronary artery disease Role of lipoprotein lipase in lipid metabolism Spatiotemporal immune zonation of the human kidney Macrophages fine tune satellite cell fate in dystrophic skeletal muscle of mdx mice The CXCL12/CXCR4 chemokine ligand/receptor axis in cardiovascular disease MAGMA: generalized gene-set analysis of GWAS data Genetic mapping of cell type specificity for complex traits Revisiting Cardiac Cellular Composition Single-cell reconstruction of the early maternal-fetal interface in humans Clinical features of patients infected with 2019 novel coronavirus in Wuhan The ACE2 expression in human heart 36 indicates new potential mechanism of heart injury among patients infected with SARS-CoV-2 Male-female differences in fertility and blood pressure in ACE-deficient mice Angiotensin-converting enzyme 2 is an essential regulator of heart function