Distinct mononuclear diploid cardiac subpopulation with minimal cell–cell communications persists in embryonic and adult mammalian heart

Abstract

A small proportion of mononuclear diploid cardiomyocytes (MNDCMs), with regeneration potential, could persist in adult mammalian heart. However, the heterogeneity of MNDCMs and changes during development remains to be illuminated. To this end, 12 645 cardiac cells were generated from embryonic day 17.5 and postnatal days 2 and 8 mice by single-cell RNA sequencing. Three cardiac developmental paths were identified: two switching to cardiomyocytes (CM) maturation with close CM–fibroblast (FB) communications and one maintaining MNDCM status with least CM–FB communications. Proliferative MNDCMs having interactions with macrophages and non-proliferative MNDCMs (non-pMNDCMs) with minimal cell–cell communications were identified in the third path. The non-pMNDCMs possessed distinct properties: the lowest mitochondrial metabolisms, the highest glycolysis, and high expression of Myl4 and Tnni1. Single-nucleus RNA sequencing and immunohistochemical staining further proved that the Myl4⁺Tnni1⁺ MNDCMs persisted in embryonic and adult hearts. These MNDCMs were mapped to the heart by integrating the spatial and single-cell transcriptomic data. In conclusion, a novel non-pMNDCM subpopulation with minimal cell–cell communications was unveiled, highlighting the importance of microenvironment contribution to CM fate during maturation. These findings could improve the understanding of MNDCM heterogeneity and cardiac development, thus providing new clues for approaches to effective cardiac regeneration.

Keywords

mononuclear diploid cardiomyocytes / cell–cell communication / cardiac fibroblast / single-cell RNA sequencing / cardiac regeneration

1 Introduction

Disordered [1] and aging hearts [2] are characterized by a loss of cardiomyocytes (CMs), which may ultimately lead to severe heart dysfunction. Although thought to include only “terminally differentiated” cells for decades [3], the heart has recently been described to sustain a certain proliferative capacity throughout life [4,5]. However, multiple issues remain unclear [6].

Mammalian CMs are proliferative during embryonic development [7]. However, their proliferative capacity gradually ceases during the first week after birth in mouse [8] and over an uncertain time in human [9,10]. In mouse, neonatal (postnatal day 1 (P1)) hearts could fully regenerate and restore heart function after apical resection; thereafter, this capacity disappears from P7 onwards [11]. However, adult amphibian animals [12,13] maintain efficient cardiac regenerative capacity to repair CM loss. Most studies suggest that preexisting CMs contribute to the production of most new CMs [5,14]. Several recent reports describe mononuclear diploid CMs (MNDCMs) as an important cell subpopulation that dominates myocardial regeneration in mammalian [8,15–17], zebrafish [18], and newt [13] hearts. Most mammalian CMs stepwise become binucleated and/or polyploid after birth, leading to their loss of proliferative and regenerative capacity [7].

A small amount of MNDCMs notably persist in adult mammalian heart, and their quantity correlates well with cardiac regenerative capacity after acute myocardial ischemia [16]. Similar findings are observed in other studies [4,5,19–23]. Although these results shed light on cardiac regeneration in hearts with cell loss, efficiently restoring the function of diseased hearts remains arduous due to the lack of effective strategies to activate these MNDCMs. The heterogeneity of these MNDCMs, their interactions with non-CMs (NCMs), and why they could continue the mononuclear diploid status during maturations need further investigation to explore the possible strategy.

Single-cell RNA sequencing (scRNA-seq) approaches are increasingly being used to investigate cardiac development and cardiac diseases [24]. The results have greatly benefitted the understanding of cardiac developmental biology by characterizing cell lineages [25,26], defining human cardiac transcription factor hierarchies [27], and unveiling the function of fibroblasts (FBs) during CM maturation [3]. Moreover, some findings describe transcriptomic differences, suggest cytoskeleton-associated protein 4 as a new target for FB activation in ischemia reperfusion (IR) model [28], and uncover that the active engagement of NCMs could modulate the behavior of CMs [29]. Here, scRNA-seq was performed on single cells from embryonic day 17.5 (E17.5), P2, and P8 murine hearts to investigate the specific cell types that may sustain high proliferative ability during cardiac development on the basis of the transcriptional profiles, characterization of interconnections with other cells, and Gene Ontology (GO) analysis of differentially expressed genes (DEGs). Two novel MNDCM subpopulations with distinct proliferative activity, metabolism property, and communication pattern were identified. Non-proliferative MNDCMs (non-pMNDCMs) displayed low proliferative activity, low mitochondrial metabolisms, low contractility, and minimal interaction with other cells. Proliferative MNDCMs (pMNDCMs) had a special communication with macrophages (MPs) and some FBs through cytokine/inflammatory signals, which may contribute to their unique regenerative potential.

2 Materials and methods

2.1 Animal samples

Mouse experiments were guided by the Institutional Animal Care and Use Committee of Huazhong University of Science and Technology (approval number: S1842) and performed in accordance with the guidelines from Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes. Eight-to-ten-week-old male C57BL/6 mice were purchased from Hubei Experimental Animal Research Center and housed in standard SPF condition (temperature: ~20 °C; humidity: 40%–60%). Two female and one male mice were caged together overnight. On the following morning, females and males were separated, and the embryos were taken as E0.5 if the females were pregnant. Mothers were anaesthetized by an intraperitoneal injection of urethane (1 mg/1 g, ShanPu, China) to harvest the fetus. P2 and P8 neonatal mice were placed on ice and decapitated.

2.2 Preparation of single cells

Offspring were harvested at E17.5, P2, and P8 for further processing to obtain single heart cells. Atria were removed from the hearts without additional valve removal. E17.5 and P2 ventricles were freshly resected from hearts, washed in phosphate buffer saline (PBS), and minced to 1–3 mm-small pieces. The P8 ventricles were transversally sectioned from the heart apex towards the bottom to obtain 0.5–1 mm heart section. Tissue pieces were incubated in 1 mg/mL collagenase type II (1 mg/mL, Worthington Biochemical, Lakewood) solution for 5 min to obtain single cells, and 3–5 rounds of incubation in enzyme solution were processed if necessary. A sterile cell strainer (40 μm) was used to remove big cells and cell clusters. A total of three pregnant female mice with 22 embryos, 12 P2 mice, and 15 P8 mice were used to the above processes to obtain single cells for scRNA-seq.

Single CMs from E16.5, P1, and P8 ventricles were used for immunofluorescence staining. Before the staining was performed, E16.5 and P1 single cells were cultured for 24 h (37 °C and 5% CO₂) in Dulbecco’s modified Eagle’ medium (DMEM, Solarbio, China) supplemented with 20% (v/v) fetal bovine serum (FBS, Gibco, USA), 50 μmol/L β-mercaptoethanol, 1% amino acid, and 1 mmol/L L-glutamine. P8 single cells were maintained for 2 hours in plating medium and cultured for 48 h in maintaining medium. The plating medium was DMEM supplemented with 17% (v/v) M199 (Solarbio, China), 5% (v/v) horse serum (Solarbio, China), and 10% (v/v) FBS. The maintaining medium was composed of (v/v) 19.5% M199, 1% horse serum, and 1% FBS in DMEM. P56 hearts were perfused with 1 mg/mL collagenase type II solution on Langendorff apparatus to obtain single cell suspension for further immunofluorescence staining.

2.3 ScRNA-seq and quality control

The cDNA yielded from single cells with Single Cell Reagent Kit (10X Genomics, Pleasanton, CA, USA) was used to generate the libraries (E17.5: Document CG00052; P2 and P8: Document CG000185). The libraries were then sequenced on the HiSeq Xten platform (Illumina, San Diego, CA, USA).

CellRanger was applied to process the raw data (version 2.1.1 for E17.5 samples and version 3.0.2 for P2 and P8 samples). Thereafter, STAR [30] (version 2.5.1b) was used to map reads to the reference genome (mm10). Finally, the “cellranger-aggr” pipeline was applied to combine individual datasets and normalize read depth across samples to approximately 151 687 reads per cell.

The following criteria were used to screen cells for further analysis: (1) minimally expressed gene number of 200; (2) minimal mitochondrial gene count proportion of 15%; (3) unique molecular identifiers (UMIs) limited to mean of log10 UMI of all cells ±2 standard deviation units and alignment rate of unique read > 89.6%; and (4) maximal mitochondrial read percentage of 86% (58%–86% of total transcripts in CM were mitochondrial gene [24]). Cells with red-blood-cell gene percentage ≥ 50% and all red-blood-cell genes were filtered via downstream analysis to exclude the contamination of these genes (gene symbols containing a prefix of “Hb-”).

2.4 Single-nucleus RNA sequencing (snRNA-seq) and quality control

Single-nucleus RNA-seq libraries were generated using Single Cell 3′ Reagent Kits version 2 (10X Genomics, Pleasanton, CA, USA) in accordance with the manufacturer’s protocol. The libraries were then sequenced on the DNBSEQ-T7HiSeq Xten platform.

CellRanger was applied to process the raw data (version 7.0.1 for w12 samples). Thereafter, STAR [30] (version 2.5.1b) was used to map reads to the reference genome (refdata-gex-mm10-2020-A). Finally, the “cellranger-aggr” pipeline was applied to combine individual data sets and normalize read depth across samples to approximately 46 387 mean reads per cell.

2.5 Single-cell clustering

Cell clusters were detected by the RunMultiCCA function of Seurat to remove batch effect in the single-cell transcriptional data from E17.5, P2, and P8 hearts [31]. The ScaleData function was used to standardize the combined gene expression matrix, and RunPCA and RunUMAP functions were employed to perform dimensionality reduction analyses on the standardized gene expression matrix. Cells from E17.5, P2, and P8 mouse hearts were positioned on a two-dimensional plane. FindNeighbors and FindClusters functions were applied to cluster the reduced dimensionality data. Next, the FindAllMarkers function was applied to find marker genes for each type of cell. Finally, the AddModuleScore function was used to perform module analysis on each type of cell. The same method was used to analyze data from public data.

2.6 Gene Ontology (GO) enrichment analysis

GO enrichment assigns functional annotations to selected gene categories or related genes. The R language clusterProfiler package (clusterProfiler-3.12.0) [32] and human genome annotation information package org.Mm.eg.db (org.Mm.eg.db-3.8.2) were used to extract the GO annotation information of DEGs. The GO enrichment analysis of DEGs and corresponding GO annotation information was performed via hypergeometric test. The GO pathway was defined as enriched if its P value ≤ 0.05. Finally, the GOplot package (GOplot-1.0.2) was applied to draw chord diagrams for DEGs and enriched GO pathways.

2.7 Pseudotime and trajectory analysis

Pseudotime and trajectory analysis was performed on cells from E17.5, P2, and P8 hearts with the OrderCells function in Monocle (version 2.16.0) [33]. The scEpath method [34] was then performed to quantify the entropy (thermal energy) landscape and analyze the developmental potency of single cells. Thereafter, the InferingLineage function was applied to calculate the probability of transformation for each cell type to predict the development pathways of these cells.

2.8 Cell-cycle analyses

The cell-cycle phase of each cell was assigned on the basis of the reported G1/S and G2/M gene panel [35,36] by using the CellCycleScoring function of Seurat.

2.9 Cell-to-cell interaction analysis

The CellChat software package (version 0.0.2) [37] was used to analyze cell–cell interactions between CMs and NCMs. By using CellChatDB, the identifyOverExpressedGenes function was applied to identify highly expressed ligands and receptors in all cell types. Next, the projectData function was applied to project gene expression data to the mouse protein–protein interaction (PPI.mouse) online. Subsequently, the computeCommunProb function was used to calculate the probability value of each interaction and perform permutation tests to infer the interaction between each type of cell. Finally, the NetVisual_aggregate function was applied to draw a network diagram of the interaction between cells. The same method was used to analyze data from Wang et al [3].

2.10 Mapping CM clusters to hearts based on spatial transcriptome analysis

The downloaded raw FASTQ files and histology images were processed sample by sample with the Space Ranger software (version 2.0.0, 10X Genomics), with the CellRanger mm10 reference genome “refdata-gex-mm10-2020-A”. Raw counts were used as the input of Seurat version 4.2.1 for downstream analysis. The gene counts were normalized using SCTransform. The samples with the most clusters identified by FindClusters were maintained for integration analysis with CM clusters identified by scRNA-seq for each condition (sham and post MI days 1/7/14) to preserve the characteristics of myocardial cells. The prediction score was calculated using the functions FindTransferAnchors and TransferData in Seurat to predict the proportion of cardiac muscle cell types from scRNA-seq in each spatial spot.

2.11 Integration and correlation analysis between scRNA-seq and snRNA-seq data sets

Integration analysis was performed using the functions “FindIntegrationAnchors” and “IntegrateData” in Seurat to merge different datasets from scRNA-seq and snRNA-seq. The same method as single-cell clustering was used to cluster the integrated data. For calculation of the Pearson’s correlations between scRNA-seq and snRNA-seq sub-clusters, all genes shared between datasets were used as the features in the function “cor” of R (version 4.0.0).

2.12 Histological and immunohistochemical staining

Single CMs and 6 μm sections from frozen tissues were processed for immunofluorescence staining. The samples were first fixed with 4% paraformaldehyde (Boster, China) and then pretreated with 0.3% Triton X-100 (Sigma, USA) and 3% goat serum (Boster, China) before incubation with primary antibodies overnight. The primary antibodies included mouse monoclonal anti-mouse Myl4 (1:100; Proteintech, USA) and rabbit anti-mouse Tnni1 (1:100, Proteintech, USA). The nuclei were stained with 6-diamidino-2-phenylindole (DAPI; Servicebio, China). Images were captured with a Zeiss FluoView microscope (Zeiss, Germany).

The DNA content and ploidy of nuclei of the Myl4⁺, Myl4⁺Tnni1⁺, Tnni1⁺, and Tnni^–Myl4^– cells were calculated and determined with ImageJ software, as previously described [38].

2.13 Fluorescence in-situ hybridization (FISH)

The probes for Tnni1 and Myl4 were synthesized (Servicebio, China), and 6 μm sections from frozen tissues were fixed with 4% paraformaldehyde, pretreated with proteinase K (20 μg/mL, Servicebio, China), and then incubated with FISH probes in hybridization buffer (Servicebio, China) at 40 °C overnight. The cell nuclei were stained by DAPI. Images were captured with a Zeiss FluoView microscope (Zeiss, Germany). The probe sequences are displayed in Table S1.

2.14 Statistical analysis

R version 4.0.0 and MATLAB version 9.2.0.538062 (R2017a) were employed to all statistical analysis and data visualization. All figures were produced with ggplot2 (version 3.3.1) and CProtMEDIAS (CProtMEDIAS-0.1.0) [39]. Significance levels were defined as following: ns, P > 0.05, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 in violin plot.

2.15 Data availability

The scRNA-seq raw data of mouse CMs were uploaded to the National Genomics Data Center under accession number CRA005078. The snRNA-seq raw data of mouse CMs were uploaded to the National Genomics Data Center under accession number CRA008957. The count matrices for mouse single-cell transcription data from Wang et al. [3] were downloaded from the GEO database under accession numbers GSE123547 and GSE122706. The 10X Genomics Visium platform was used to analyze the spatial transcriptome raw data of E15.5 mouse hearts from Chen et al. [40] and myocardial infarction operation mouse hearts from Yamada et al [41]. The data were downloaded from the GEO database under accession numbers GSE179392 and GSE176092. The count matrices for mouse single-cell transcription data from Cui et al. [42] were downloaded from the GEO database under accession number GSE130699. All the other relevant data are available upon request from the corresponding author.

3 Results

3.1 Single-cell analysis of murine hearts during development from E17.5 to P8

The single cells from E17.5, P2, and P8 murine hearts were acquired for further scRNA-seq analysis to trace the developmental changes of different CM subpopulations, including MNDCMs. The experimental design and analysis workflow are shown in Fig. S1. A total of 12 645 cells, including 4335 cells from E17.5 hearts, 2177 cells from P2 hearts, and 6133 cells from P8 hearts, passed the stringent quality control for further analysis (Table S2). Based on the molecular signatures, five major cell types, CMs, FBs, MPs, endothelial cells (ECs), and smooth muscle cells (SMCs) were identified and visualized with Uniform Manifold Approximation and Projection (UMAP) [43] (Fig.1 and 1B). The expression of key markers confirmed the identified cell populations as CM (Myh6⁺ and Myl2⁺), FB (Col1a⁺ and Dcn⁺), EC (Cdh5⁺ and Pecam1⁺), MP (Adgre1⁺ and Csf1r⁺), and SMC (Acta2⁺ and Myh11⁺), as shown in Fig.1 and 1D. The GO analysis of DEGs described the muscle activities, including muscle system process, muscle contraction, muscle cell and cardiac muscle tissue development in CMs (Fig.1), extracellular matrix organization and tissue remodeling in FBs (Fig. S2A and S2G), immune cell activity in MPs (Fig. S2B and S2H), and angiogenesis and endothelial activity in ECs (Fig. S2C and S2I). Together, these results demonstrated the transcriptional complexity of all the heart cells.

3.2 CM sub-clustered into distinct populations

These cardiac cells displayed certain heterogeneities (Fig.2 and S2D–S2F), and the majority of CMs were from the ventricle, especially the left ventricle (Fig. S3). Eight CM sub-clusters were also identified (Fig.2–2C and S4). The CM0 cluster had the largest number of cells and especially expressed genes related to muscle cell development (Fig.2). Enrichment of striated-muscle-development-related genes indicated their relatively more immature status than other CM cells. This finding was a clue for the CM0 cells to be cardiac-progenitor-like cells. Here, CM0 was marked as cp-like-CM, denoted as CM0_(cp-likeCM). CM1 and CM3 shared similar signatures, Cenpa or Cenpf (Fig.2), which are required for progress through mitosis, chromosome segregation, and cytokinesis [44–49]. Genes highly expressed in CM1 were enriched for functions related to nucleus division and mitotic spindle organization. However, these genes were also observed to be enriched in negative regulation of cell-cycle phase transition, suggesting a multinucleation trend in CM1. Numerous genes related to muscle hypertrophy in CM7 suggested their remarkable cell growth. Myl4 was dominantly expressed in CM4 and CM7 (Fig.2). CM7 also highly expressed other Myl gene subtypes, including Myl7 and Myl1. CM2 expressed abundant mitochondrial genes (Fig.2), and other genes were enriched in cardiac contraction and morphology (Fig.2), indicating a more mature status and reduced proliferative capability [50]. Therefore, CM2 was marked as mature CM (mCM). CM4 exhibited a unique metabolic pattern with a mixture of fatty acid metabolic processes and oxidative phosphorylation (Fig.2). CM5 cells were characterized by their expression of extracellular matrix components, such as Vim, Col1a1, Col1a2, and Col3a1 (Fig.2), and an intimate correlation between CM5 cells and ECs (Fig.2). Hence, they were marked as EC-like CMs.

However, the limited numbers of DEGs for CM3 and CM6 were not sufficient to obtain valuable enrichment of any pathway. Alternatively, the selected populations were compared to one or more other populations to acquire some useful information for enriched pathways (Fig. S5). Obviously, mitosis-related genes were upregulated in CM1/CM3/CM5_(EC-likeCM) compared with CM0_(cp-likeCM) (Fig. S5A and S5B), suggesting that the heart growth due to cell division could increase the heart size. Upregulation of mitosis/cytokinesis-related genes in CM6 were observed when compared with CM1/CM7/CM2_(mCM) (Fig. S5C and S5D) or CM4 (Fig. S5E and S5F). This finding provided a clue for active proliferation in CM6. Compared with CM6 or CM1/CM7/CM2_(mCM), CM4 had a special pattern of metabolism, as previously described in Fig.2; less cytokinesis/mitotic activity; and less cardiac hypertrophy (Fig. S5E–S5H).

3.3 Pseudotemporal trajectory analyses revealing a special MNDCM maintaining path of cardiac differentiation

Next, Monocle [33] and ScEnergy [34] were combined to distinct CM clusters in the context of cardiac differentiation and single cell entropy to investigate the differential trajectory of these CM clusters (Fig.3 and S6). Two different paths were proposed by Monocle, where most CM5_(EC-likeCM) and CM6 were distributed in a path different from the other one (Fig.3–3D and S6A). The lowest energies were detected in CM2_(mCM), suggesting that CMs represent a committed and differentiated cell state and are quiescent [51]. CM5_(EC-likeCM)/CM6 possessed more entrophy than CM2_(mCM) but less than all the other CM clusters. The highest entropy was observed in CM0_(cp-likeCM)/CM3/CM7 (Fig.3 and 3F). Thus, one cluster from CM0_(cp-likeCM)/CM3/CM7 possibly served as the origin of cellular differentiation. On the basis of the above analysis, three distinct developmental paths were established by the system (Fig.3): CM0_(cp-likeCM) was the origin of CM development; CM1/CM7/CM2_(mCM) cells distributed in first path, differentiated into mature functional CMs for contraction and heart pump function (named as contractile CM maturation path, in which CM1 and CM7 as intermediate CMs: interCM); In the second path, CM3 exhibited reserved proliferative potential but later differentiate into two cell fates: CM4 and CM6; CM5 in the third path became specialized to behave like ECs (named as EC-like-CM path).

A notable detail that CM4 cells had relatively higher energies than several CMs clusters, indicating their relative immature status compared with many other CM clusters (Fig.3 and 3F). However, they did not exhibit high mitosis activity as CM6 in the same path (Fig. S5E and S5F). The cell-cycle distribution of CM sub-clusters was well related with the predictions for CM4 and CM6 (Fig.3). Therefore, CM4 and CM6 were named as non-proliferative CM (non-pCM) and proliferative CM (pCM), respectively.

3.4 Non-pCM (CM4) as a novel and unique cardiomyocyte sub-cluster with specific characterization

The major functions of CM4_(non-pCM) were compared to those of other CM subpopulations on the basis of their molecular signatures to explore the possible roles of CM4_(non-pCM) cells (Fig.4 and S7). CM4_(non-pCM) cells uniquely sustained the lowest expression levels of mitochondrial genes (****P < 0.0001, Table S3), the highest glycolysis activity (****P < 0.0001, Table S3), and the most immature cardiac contraction function with regard to calcium handling and structure (Fig.4 and S7A–S7D). By contrast, CM2_(mCM) cells displayed the lowest glycolysis status and the most mature cardiac contraction function. The CM sub-clusters in the first pathway (CM0_(cp-likeCM), CM1_(interCM), CM7_(interCM) and CM2_(mCM)) expressed more calcium handling and structure-related genes than CM4_(non-pCM)/CM6_(pCM).

The special properties of CM4_(non-pCM) fit the previous suspect on MNDCM [16] but they were relatively quiescent in proliferation. They were then marked as non-pMNDCM. To test this view, potential markers of this population were identified by relying on their DEGs. Myl4 was found to be a candidate and also highly expressed in CM7. Hence, Tnni1 was proposed to be an additional marker to distinguish CM4_(non-pMNDCM) from CM7_(interCM) (Fig.4 and 4C).

Then, the snRNA-seq data from P2-11 mouse hearts [42] were utilized to confirm the CM4 existence in mouse immature hearts (Fig. S8A and S8B). On the basis of single-nucleus transcriptomics, the P2-11 CMs were clustered to seven sub-clusters (Fig.4 and S8D). These CMs were integrated with the scRNA-seq data, and 11 CM clusters were identified (Fig.4). The distribution of CM4 was the closest to P2-11 CM3 cells (Fig.4). P2-11 CM3 expressed higher levels of Myl4 and Tnni1 (Fig. S8C) and had the highest correlation to CM4 (Fig.4 and S8E). Furthermore, the anatomic distribution of the scRNA-seq CM clusters was assigned to E15.5 heart in accordance with the spatial and single-cell transcriptomic data [40]. CM4 was mostly distributed in the epicardial and ventricular septum field, and CM5 was located in the vascular area out of the heart (Fig.4, S8F and S8G).

For verification of these markers, immunohistochemical experiment was performed in single CMs acquired from E16.5, P1, and P8 mouse hearts with antibodies specific for these two marker proteins. As shown in Fig. S9, the pink arrow indicated the cells were Myl4⁺Tnni1⁺ and that they were mostly confirmed to be mononuclear diploid by analyzing the DNA content. The CMs positive only for Tnni1 (green arrow) were MNDCM or polyploid. Myl4⁺Tnni1⁺ MNDCMs were also observed in P8 heart cells (Fig.4 upper panel). Some CMs only positive for Myl4 were either binucleated or MNDCM (Fig.4 lower panel). Hence, CM3/CM6/CM4 could be the MNDCM maintaining path. In addition, CM3 was marked as MNDCM progenitor (proMNDCM), and CM6 was marked as pMNDCM.

3.5 Non-pMNDCM (CM4) possessing minimal cell–cell interaction with FBs and MPs

FBs play an essential role in CM maturation during cardiac development, and some other NCMs affect heart function [3]. In the present study, the cell–cell interactions among cardiac cells were quantified by CellChat software package (Fig.5–5C and Tables S4 and S5) [37]. CM4_(non-pMNDCM)/CM6_(pMNDCM) had less interaction with other cell types, and CM4_(non-pMNDCM) exhibited minimal communication, especially the incoming information from other cells (Fig.5 and S10). The highest-scoring interactions between CM clusters and other cell types were mostly observed with FBs (Fig.5, 5E and S10).

The CM clusters in the contractile CM maturation path (CM0_(cp-likeCM)/CM1_(interCM)/CM7_(interCM)/CM2_(mCM)) and EC-like CM path (CM0_(cp-likeCM)/CM5_(EC-likeCM)) shared many incoming signals (Fig. S10), including the ncWNT signal that majorly arouse from FB1 and the others from CM clusters. BMP, a well-known transcriptional factor that modulated the cardiac development and differentiation, was universally detected in these CMs except for CM2_(mCM). The major BMP secreting cells were FB0 and FB1. CM0_(cp-likeCM) exhibited the highest and most significant interaction scores for TWEAK, VISFASTIN, and IGF pathways. CM7_(interCM) had additional incoming signals of SEMA3, VEGF, GRN and GALECTIN pathways, and FB0 were the major cells delivering those signals. These communications between CM0_(cp-likeCM)/CM1_(interCM)/CM7_(interCM)/CM2_(mCM) and FB clusters suggested the important modulation of fibroblast on CM maturation [3]. GAS and PDGF pathways specially interacted with CM5_(EC-likeCM). These signals were majorly secreted by CM clusters.

CM6_(pMNDCM) cells exhibited much fewer incoming signals than these above CM clusters, and more importantly, the closest interaction with MPs (Fig.5, 5E and S10). Chemokine/inflammatory incoming signals, interleukin 16 (IL-16) and tumor necrosis factor (TNF), were intensively detected in CM6_(pMNDCM) and predominantly provided by MPs (Fig.5, 5E, and 5H–5J). IL-16 signals universally originated from all the MP subpopulations and profoundly from MP0 (Fig. S10C and S10D). The C-X-C motif chemokine ligand (CXCL) from the other CM clusters and NCMs specially interacted with CM6_(pMNDCM). Similar chemokine signals, most notably IL-16 and C-C motif chemokine ligand (CCL) from FB7 and FB5, respectively, were also identified in CM6_(pMNDCM) (Fig. S10A and S10B). The PTN pathway was moderately detected in all CM clusters, and the MK pathway was noticeably absent in CM4_(non-pMNDCM) (Fig.5 and 5G).

The communication pattern among different CM clusters with FBs or MPs indicated that the NCMs were closely correlated to different CM developmental fates.

3.6 Non-proliferative MNDCM existence in adult murine heart

Lastly, whether CM4_(non-pMNDCM) persists during heart maturation or experiences developmental changes was determined. Three 12-week-old male mice were used to isolate the heart nuclei for snRNA-seq (Fig. S11A and S11B). Eight W12-CM clusters were defined (Fig.6), and the correlations with the previously identified CM clusters (scRNA-seq CMs) were calculated. As shown in Fig.6, the highest positive correlation with CM4 was W12-CM3, indicating the possible existence of CM4 in adult hearts. By contrast, CM6 did not show a positive correlation to any W12-CM clusters, suggesting their absence in the adult heart. The scRNA-seq CMs were also merged with the snRNA-seq W12-CMs by UMAP analysis, CM4 and W12-CM3 were grouped together (Fig.6). More cells in W12-CM3 shared the high glycolysis level with CM4, and W12-CM2 behaved like CM4 in calcium handling and structure (Fig.6). As expected, immunohistochemical staining detected a few Myl4⁺Tnni1⁺ cells (pink arrow indicated) in P56 heart slices; their nuclei were much smaller than the Tnni1⁺ (red arrow) and Myl4⁺ (green arrow) CMs (Fig.6). FISH also displayed few Myl4⁺Tnni1⁺ CMs lying under the epicardium (Fig. S12). In P56 heart cell suspension, some Myl4⁺Tnni1⁺ CMs were detected as mononuclear, with small round nucleus (Fig. S13). Then, the scRNA-seq CM clusters were assigned to the 9-week-old hearts (sham and MI days 1/7/14) on the basis of the spatial and single-cell transcriptomic data (Fig.6 and S11F). As expected, a few CM4 cells were detected in the 9-week-old heart. After acute myocardial infarction (AMI), CM4 and CM5 were found to accumulate in the infarction zone or edge, thus proposing their participation in heart functional recovery.

The scRNA-seq data from Wang et al. were also acquired and analyzed [3]. In total, 1210 CMs from P1, P7, P14, and P56 were clustered into seven subpopulations (Fig. S14). W-CM5 attracted the author’s interest because Myl4 was highly enriched along with higher Tnni1 expression (Fig. S14B–S14G) and the minimal interactions with FB and other CM sub-clusters (Fig. S15D and S16). However, the correlation analysis did not support this speculation (Fig. S15A–S15C), may be because of the small numbers of CMs in Wang et al.’s dataset (1210 CMs in total, with 47 W-CM5 only).

Taken together, all of these results not only supported the existence of CM4 in adult hearts but also indicated that these MNDCMs may function under some conditions, such as AMI, which highlights the importance of the new identified subpopulations.

3.7 Modeling the cardiomyocyte fate during development

Here, three paths were identified for the distinct CM cellular fates during development of ventricular CMs from E17.5 to P8: first was the contractile CM maturation path involving maturation of functional contracting CMs with abundant expression of calcium handling and contraction-related structural genes (CM0_(cp-likeCM), CM1_(interCM), CM7_(interCM) and CM2_(mCM)); second, which is the most unique path, was MNDCM maintaining path involving CM4_(non-pMNDCM) that was quiescent and CM6_(pMNDCM) that was actively proliferative; and third was EC-like-CM path involving CMs with similarity to ECs (CM5_(EC-likeCM)). Various CM cell fates were paralleled by distinct crosstalk with the NCMs (Fig.7).

ErbB4 and Tnni3k activation may increase cardiac proliferative activity [16]. However, their expression was almost undetectable in silent CM4_(non-pMNDCM) and active CM6_(pMNDCM) (Fig. S17A). ErbB4 and Tnni3k were expressed at a certain level in CMs from the contractile CM maturation path. Further analysis of the data from Wang et al. revealed similar expression levels of these two signals (Fig. S17B), with the exception of W-CM4.

4 Discussion

Recently, researchers have highlighted multiple key factors that relate to cardiac regeneration: low energy requirements and special metabolism pattern [38], unique immune responses [52,53], and mononuclear diploid status [54]. High glycolysis is considered to provide rapid energy for cell proliferation and tumor growth in various tissues and to decrease risk factors associated with myocardial ischemia [55]. It is also closely related to the transition of proliferative immature CMs to non-proliferative mature CMs [26]. Activating glycolysis helps activate cardiomyocyte proliferation [56]. Diploid cardiomyocyte abundance has inverse correlation with standard metabolic rate and temperature [38]. The greater proportion of MNDCMs in adult murine heart is correlated with greater functional recovery following injury [16]. MNDCMs were also visualized in adult mammalian hearts with an immunological technique and identified by snRNA-seq [57]. Their existence was detected in the snRNA-seq data in the present study, and their anatomic distributions were assigned on the basis of spatial transcriptomic analysis. Polyploidization was proposed to act as a hurdle for cardiac proliferation [58]. In the present study, CM4_(non-pMNDCM) displayed the proposed gene expression pattern for proliferative CMs [38,52,53], and Myl4⁺Tnni1⁺ CM4_(non-pMNDCM) were confirmed to be MNDCMs. CM4_(non-pMNDCM) expressed relatively lower levels of mitochondrial and cardiac contraction-related genes and higher levels of glycolysis-related genes than other CM clusters. These characteristics fit the previous characterization of proliferative cells, which contribute to endogenous cardiac regeneration [38,52,53]. Strikingly, CM4_(non-pMNDCM) behaved to the extreme for these properties but appeared quiescent in proliferation compared with CM6_(pMNDCM) cluster, which exhibited the most active cytokinesis among all the CM clusters. CM4_(non-pMNDCM) displayed the additional property of high fatty-acid oxidation and the minimal interaction with other cells. These properties may contribute to their unexpected silent proliferation.

FBs, a crucial NMC type in the heart, contribute to pathophysiological process of heart diseases by secreting factors transforming growth factor β and fibroblast growth factor 2 in hearts with pressure overload [59,60], and exert anti-hypertrophic effects by producing IL-33 [61,62]. They also guide CM maturation [3,63,64] during development and favor embryonic CM proliferation [65]. However, the CM cellular fate remains unclear, whether they lack interaction with FBs. The distinct crosstalk with other cells existed in CM and other sub-clusters, thus providing potential mechanisms for the cell fate of different CM clusters. Single-cardiomyocyte transcriptomics during heart development have been explored by scRNA-seq [3] or snRNA-seq [42] for different purposes. The developmental changes in single-cell transcriptomics during CM maturation have been reported: immature CMs expressed genes enriched in RNA splicing, cell-cycle phase transition, and cardiac muscle development; intermediate CMs expressed abundant genes related to heart contraction and ATP metabolic process; and DEGs of mature CMs were enriched in the regulation of vasculature development, protein maturation, and cell cycle [3,26]. In the present study, similar developmental changes were found in the contractile CM maturation path (CM0_(cp-likeCM), CM1_(interCM), CM7_(interCM) and CM2_(mCM)), in which the majority of CMs were distributed. They maintained the close interaction with FBs and experienced the maturation process to perform cardiac function, so did CM0_(cp-likeCM)/CM5_(EC-likeCM) in another path. Meanwhile, CM4_(non-pMNDCM) lacked crosstalk with FBs, and CM6_(pMNDCM) exhibited obviously less communication with FBs than other CM subpopulations. This finding suggested that FBs play an essential role in determining CM fate during development: CMs having enough communication with FBs could follow the maturation progress, and CMs lacking sufficient communication with FBs may maintain MNDCM status.

Another notable interaction was established between CM6_(pMNDCM) and MPs via chemokine/inflammatory (IL-16, CXCL, and TNF) signals. MP infiltration was proven to be essential for limb regeneration in newts [66] and neonatal CM proliferation through promotion of myocardial angiogenesis [67], indicating the potential of MPs as a mechanism for inducing tissue proliferation. These chemokine/inflammatory signals in CM6_(pMNDCM) may determine the different proliferative capacities of CM4_(non-pMNDCM) and CM6_(pMNDCM). They also suggested that CM6_(pMNDCM) could be possibly converted to CM4_(non-pMNDCM) during development by losing the critical crosstalk with other cells.

The heart nuclei of 12-week-old mouse were isolated for single-nucleus transcriptomic analysis, and CM4 was found to be correlated best to W12-CM3&CM2. These two clusters resembled to CM4 either with high glycolysis or less calcium-handling-related and structural genes. W12-CM2 and CM3 may be the two subpopulations of CM4 in adult hearts. Spatial transcriptomic analysis revealed that CM4 and CM5 increased in the infarct edge and zone, thus providing indirect proof that their functional role in cardiogenesis (CM4) or angiogenesis (CM5). W-CM5 resembled the most to CM4 with minimal crosstalk with NCMs but without significant positive correlation. W-CM6 behaved similarly to pMNDCM, suggested a possible developmental changes or some unknown mechanisms that need to be explored in the future.

Additional analysis of ErbB4 and Tnni3k revealed that they may play minimal roles in non-pMNDCM and pMNDCM cells, although their activation was expected to increase the number of MNDCMs [16]. In particular, the observation that SJL, A, and SWR mouse strains displayed the lowest Tnnik3k expression and significantly different MNDCM frequencies raised questions about their function in MNDCM proliferation [16]. Moreover, high expression of Tnnik3k in CMs of the first path suggested an important role in cardiac regeneration during development, although its ability to directly interfere with MNDCM proliferation in adult CMs remains obscure.

Here, a MNDCM maintaining path during cardiac development was detected. Two MNDCM populations, CM4_(non-pMNDCM) and CM6_(pMNDCM), were also identified. The quantity of MNDCM considerably affected the function recovery after adult heart injury [16], whereas it ignored MNDCM heterogeneity, in which MNDCM may contribute to cardiac regeneration. MNDCM heterogeneity was identified in the present study, and the quality of MNDCMs could serve as another important issue to influence the endogenous cardiac regeneration in adults. Their unique profiles and communication patterns suggested that MNDCM proliferation could be manipulated by activating some chemokine signaling pathways, including IL-16, CXCL, and TNF signaling pathways. However, more efforts are needed to address the species differences to explore the possible strategy for a successful transition from CM4_(non-pMNDCM) to CM6_(pMNDCM) and achieve further knowledge on these MNDCMs by purifying these two MNDCMs separately.

In summary, a special MNDCM maintaining path during CM development was identified, and two MNDCM populations were discovered: one was inactive and the other was actively proliferating. CM4_(non-pMNDCM) had a unique profile with regard to general properties and minimal interactions with other cells. This study provided new knowledge on MNDCMs for enhanced understanding of cardiac regeneration and cardiac development.

References

[1] Murry CE, Reinecke H, Pabon LM. Regeneration gaps. J Am Coll Cardiol 2006; 47(9): 1777–1785

[2] Whelan RS, Kaplinskiy V, Kitsis RN. Cell death in the pathogenesis of heart disease: mechanisms and significance. Annu Rev Physiol 2010; 72(1): 19–44

[3] Wang Y, Yao F, Wang L, Li Z, Ren Z, Li D, Zhang M, Han L, Wang SQ, Zhou B, Wang L. Single-cell analysis of murine fibroblasts identifies neonatal to adult switching that regulates cardiomyocyte maturation. Nat Commun 2020; 11(1): 2585

[4] Bergmann O, Bhardwaj RD, Bernard S, Zdunek S, Barnabé-Heider F, Walsh S, Zupicich J, Alkass K, Buchholz BA, Druid H, Jovinge S, Frisén J. Evidence for cardiomyocyte renewal in humans. Science 2009; 324(5923): 98–102

[5] Senyo SE, Steinhauser ML, Pizzimenti CL, Yang VK, Cai L, Wang M, Wu TD, Guerquin-Kern JL, Lechene CP, Lee RT. Mammalian heart renewal by pre-existing cardiomyocytes. Nature 2013; 493(7432): 433–436

[6] Carvalho AB, de Carvalho AC. Heart regeneration: past, present and future. World J Cardiol 2010; 2(5): 107–111

[7] Laflamme MA, Murry CE. Heart regeneration. Nature 2011; 473(7347): 326–335

[8] Soonpaa MH, Kim KK, Pajak L, Franklin M, Field LJ. Cardiomyocyte DNA synthesis and binucleation during murine development. Am J Physiol 1996; 271(5 Pt 2): H2183–H2189

[9] Mollova M, Bersell K, Walsh S, Savla J, Das LT, Park SY, Silberstein LE, Dos Remedios CG, Graham D, Colan S, Kühn B. Cardiomyocyte proliferation contributes to heart growth in young humans. Proc Natl Acad Sci USA 2013; 110(4): 1446–1451

[10] Ye L, Qiu L, Zhang H, Chen H, Jiang C, Hong H, Liu J. Cardiomyocytes in young infants with congenital heart disease: a three-month window of proliferation. Sci Rep 2016; 6(1): 23188

[11] Porrello ER, Mahmoud AI, Simpson E, Hill JA, Richardson JA, Olson EN, Sadek HA. Transient regenerative potential of the neonatal mouse heart. Science 2011; 331(6020): 1078–1080

[12] Poss KD, Wilson LG, Keating MT. Heart regeneration in zebrafish. Science 2002; 298(5601): 2188–2190

[13] Oberpriller JO, Oberpriller JC. Response of the adult newt ventricle to injury. J Exp Zool 1974; 187(2): 249–259

[14] Ali SR, Hippenmeyer S, Saadat LV, Luo L, Weissman IL, Ardehali R. Existing cardiomyocytes generate cardiomyocytes at a low rate after birth in mice. Proc Natl Acad Sci USA 2014; 111(24): 8850–8855

[15] Li F, Wang X, Capasso JM, Gerdes AM. Rapid transition of cardiac myocytes from hyperplasia to hypertrophy during postnatal development. J Mol Cell Cardiol 1996; 28(8): 1737–1746

[16] Patterson M, Barske L, Van Handel B, Rau CD, Gan P, Sharma A, Parikh S, Denholtz M, Huang Y, Yamaguchi Y, Shen H, Allayee H, Crump JG, Force TI, Lien CL, Makita T, Lusis AJ, Kumar SR, Sucov HM. Frequency of mononuclear diploid cardiomyocytes underlies natural variation in heart regeneration. Nat Genet 2017; 49(9): 1346–1353

[17] Xavier-Vidal R, Mandarim-de-Lacerda CA. Cardiomyocyte proliferation and hypertrophy in the human fetus: quantitative study of the myocyte nuclei. Bull Assoc Anat (Nancy) 1995; 79(246): 27–31

[18] Kikuchi K, Holdway JE, Werdich AA, Anderson RM, Fang Y, Egnaczyk GF, Evans T, Macrae CA, Stainier DY, Poss KD. Primary contribution to zebrafish heart regeneration by gata4⁺ cardiomyocytes. Nature 2010; 464(7288): 601–605

[19] Derks W, Bergmann O. Polyploidy in cardiomyocytes: roadblock to heart regeneration?. Circ Res 2020; 126(4): 552–565

[20] Malliaras K, Zhang Y, Seinfeld J, Galang G, Tseliou E, Cheng K, Sun B, Aminzadeh M, Marbán E. Cardiomyocyte proliferation and progenitor cell recruitment underlie therapeutic regeneration after myocardial infarction in the adult mouse heart. EMBO Mol Med 2013; 5(2): 191–209

[21] Chen X, Wilson RM, Kubo H, Berretta RM, Harris DM, Zhang X, Jaleel N, MacDonnell SM, Bearzi C, Tillmanns J, Trofimova I, Hosoda T, Mosna F, Cribbs L, Leri A, Kajstura J, Anversa P, Houser SR. Adolescent feline heart contains a population of small, proliferative ventricular myocytes with immature physiological properties. Circ Res 2007; 100(4): 536–544

[22] Liao HS, Kang PM, Nagashima H, Yamasaki N, Usheva A, Ding B, Lorell BH, Izumo S. Cardiac-specific overexpression of cyclin-dependent kinase 2 increases smaller mononuclear cardiomyocytes. Circ Res 2001; 88(4): 443–450

[23] Mahmoud AI, Kocabas F, Muralidhar SA, Kimura W, Koura AS, Thet S, Porrello ER, Sadek HA. Meis1 regulates postnatal cardiomyocyte cell cycle arrest. Nature 2013; 497(7448): 249–253

[24] Paik DT, Cho S, Tian L, Chang HY, Wu JC. Single-cell RNA sequencing in cardiovascular development, disease and medicine. Nat Rev Cardiol 2020; 17(8): 457–473

[25] Jia G, Preussner J, Chen X, Guenther S, Yuan X, Yekelchyk M, Kuenne C, Looso M, Zhou Y, Teichmann S, Braun T. Single cell RNA-seq and ATAC-seq analysis of cardiac progenitor cell transition states and lineage settlement. Nat Commun 2018; 9(1): 4877

[26] DeLaughter DM, Bick AG, Wakimoto H, McKean D, Gorham JM, Kathiriya IS, Hinson JT, Homsy J, Gray J, Pu W, Bruneau BG, Seidman JG, Seidman CE. Single-cell resolution of temporal gene expression during heart development. Dev Cell 2016; 39(4): 480–490

[27] Churko JM, Garg P, Treutlein B, Venkatasubramanian M, Wu H, Lee J, Wessells QN, Chen SY, Chen WY, Chetal K, Mantalas G, Neff N, Jabart E, Sharma A, Nolan GP, Salomonis N, Wu JC. Defining human cardiac transcription factor hierarchies using integrated single-cell heterogeneity analysis. Nat Commun 2018; 9(1): 4906

[28] Gladka MM, Molenaar B, de Ruiter H, van der Elst S, Tsui H, Versteeg D, Lacraz GPA, Huibers MMH, van Oudenaarden A, van Rooij E. Single-cell sequencing of the healthy and diseased heart reveals cytoskeleton-associated protein 4 as a new modulator of fibroblasts activation. Circulation 2018; 138(2): 166–180

[29] Wang L, Yu P, Zhou B, Song J, Li Z, Zhang M, Guo G, Wang Y, Chen X, Han L, Hu S. Single-cell reconstruction of the adult human heart during heart failure and recovery reveals the cellular landscape underlying cardiac function. Nat Cell Biol 2020; 22(1): 108–119

[30] Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 2013; 29(1): 15–21

[31] Butler A, Hoffman P, Smibert P, Papalexi E, Satija R. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat Biotechnol 2018; 36(5): 411–420

[32] Yu G, Wang LG, Han Y, He QY. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS 2012; 16(5): 284–287

[33] Trapnell C, Cacchiarelli D, Grimsby J, Pokharel P, Li S, Morse M, Lennon NJ, Livak KJ, Mikkelsen TS, Rinn JL. The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nat Biotechnol 2014; 32(4): 381–386

[34] Jin S, MacLean AL, Peng T, Nie Q. scEpath: energy landscape-based inference of transition probabilities and cellular trajectories from single-cell transcriptomic data. Bioinformatics 2018; 34(12): 2077–2086

[35] Tirosh I, Izar B, Prakadan SM, Wadsworth MH 2nd, Treacy D, Trombetta JJ, Rotem A, Rodman C, Lian C, Murphy G, Fallahi-Sichani M, Dutton-Regester K, Lin JR, Cohen O, Shah P, Lu D, Genshaft AS, Hughes TK, Ziegler CG, Kazer SW, Gaillard A, Kolb KE, Villani AC, Johannessen CM, Andreev AY, Van Allen EM, Bertagnolli M, Sorger PK, Sullivan RJ, Flaherty KT, Frederick DT, Jané-Valbuena J, Yoon CH, Rozenblatt-Rosen O, Shalek AK, Regev A, Garraway LA. Dissecting the multicellular ecosystem of metastatic melanoma by single-cell RNA-seq. Science 2016; 352(6282): 189–196

[36] Guerrero-Juarez CF, Dedhia PH, Jin S, Ruiz-Vega R, Ma D, Liu Y, Yamaga K, Shestova O, Gay DL, Yang Z, Kessenbrock K, Nie Q, Pear WS, Cotsarelis G, Plikus MV. Single-cell analysis reveals fibroblast heterogeneity and myeloid-derived adipocyte progenitors in murine skin wounds. Nat Commun 2019; 10(1): 650

[37] Jin S, Guerrero-Juarez CF, Zhang L, Chang I, Ramos R, Kuan CH, Myung P, Plikus MV, Nie Q. Inference and analysis of cell-cell communication using CellChat. Nat Commun 2021; 12(1): 1088

[38] Hirose K, Payumo AY, Cutie S, Hoang A, Zhang H, Guyot R, Lunn D, Bigley RB, Yu H, Wang J, Smith M, Gillett E, Muroy SE, Schmid T, Wilson E, Field KA, Reeder DM, Maden M, Yartsev MM, Wolfgang MJ, Grützner F, Scanlan TS, Szweda LI, Buffenstein R, Hu G, Flamant F, Olgin JE, Huang GN. Evidence for hormonal control of heart regenerative capacity during endothermy acquisition. Science 2019; 364(6436): 184–188

[39] Zhang Z, Zhu M, Xie Q, Larkin RM, Shi X, Zheng B. CProtMEDIAS: clustering of amino acid sequences encoded by gene families by MErging and DIgitizing Aligned Sequences. Brief Bioinform 2022; 23(4): bbac276

[40] Wu T, Liang Z, Zhang Z, Liu C, Zhang L, Gu Y, Peterson KL, Evans SM, Fu XD, Chen J. PRDM16 is a compact myocardium-enriched transcription factor required to maintain compact myocardial cardiomyocyte identity in left ventricle. Circulation 2022; 145(8): 586–602

[41] Yamada S, Ko T, Hatsuse S, Nomura S, Zhang B, Dai Z, Inoue S, Kubota M, Sawami K, Yamada T, Sassa T, Katagiri M, Fujita K, Katoh M, Ito M, Harada M, Toko H, Takeda N, Morita H, Aburatani H, Komuro I. Spatiotemporal transcriptome analysis reveals critical roles for mechano-sensing genes at the border zone in remodeling after myocardial infarction. Nat Cardiovasc Res 2022; 1: 1072–1083

[42] Cui M, Wang Z, Chen K, Shah AM, Tan W, Duan L, Sanchez-Ortiz E, Li H, Xu L, Liu N, Bassel-Duby R, Olson EN. Dynamic transcriptional responses to injury of regenerative and non-regenerative cardiomyocytes revealed by single-nucleus RNA sequencing. Dev Cell 2020; 53(1): 102–116.e8

[43] Butler A, Hoffman P, Smibert P, Papalexi E, Satija R. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat Biotechnol 2018; 36(5): 411–420

[44] Zeitlin SG, Shelby RD, Sullivan KF. CENP-A is phosphorylated by Aurora B kinase and plays an unexpected role in completion of cytokinesis. J Cell Biol 2001; 155(7): 1147–1157

[45] Kunitoku N, Sasayama T, Marumoto T, Zhang D, Honda S, Kobayashi O, Hatakeyama K, Ushio Y, Saya H, Hirota T. CENP-A phosphorylation by Aurora-A in prophase is required for enrichment of Aurora-B at inner centromeres and for kinetochore function. Dev Cell 2003; 5(6): 853–864

[46] Orthaus S, Biskup C, Hoffmann B, Hoischen C, Ohndorf S, Benndorf K, Diekmann S. Assembly of the inner kinetochore proteins CENP-A and CENP-B in living human cells. ChemBioChem 2008; 9(1): 77–92

[47] Bailey AO, Panchenko T, Sathyan KM, Petkowski JJ, Pai PJ, Bai DL, Russell DH, Macara IG, Shabanowitz J, Hunt DF, Black BE, Foltz DR. Posttranslational modification of CENP-A influences the conformation of centromeric chromatin. Proc Natl Acad Sci USA 2013; 110(29): 11827–11832

[48] Yu Z, Zhou X, Wang W, Deng W, Fang J, Hu H, Wang Z, Li S, Cui L, Shen J, Zhai L, Peng S, Wong J, Dong S, Yuan Z, Ou G, Zhang X, Xu P, Lou J, Yang N, Chen P, Xu RM, Li G. Dynamic phosphorylation of CENP-A at Ser68 orchestrates its cell-cycle-dependent deposition at centromeres. Dev Cell 2015; 32(1): 68–81

[49] Roulland Y, Ouararhni K, Naidenov M, Ramos L, Shuaib M, Syed SH, Lone IN, Boopathi R, Fontaine E, Papai G, Tachiwana H, Gautier T, Skoufias D, Padmanabhan K, Bednar J, Kurumizaka H, Schultz P, Angelov D, Hamiche A, Dimitrov S. The flexible ends of CENP-A nucleosome are required for mitotic fidelity. Mol Cell 2016; 63(4): 674–685

[50] Szibor M, Pöling J, Warnecke H, Kubin T, Braun T. Remodeling and dedifferentiation of adult cardiomyocytes during disease and regeneration. Cell Mol Life Sci 2014; 71(10): 1907–1916

[51] Teschendorff AE, Enver T. Single-cell entropy for accurate estimation of differentiation potency from a cell’s transcriptome. Nat Commun 2017; 8(1): 15599

[52] Ridker PM, Everett BM, Thuren T, MacFadyen JG, Chang WH, Ballantyne C, Fonseca F, Nicolau J, Koenig W, Anker SD, Kastelein JJP, Cornel JH, Pais P, Pella D, Genest J, Cifkova R, Lorenzatti A, Forster T, Kobalava Z, Vida-Simiti L, Flather M, Shimokawa H, Ogawa H, Dellborg M, Rossi PRF, Troquay RPT, Libby P, Glynn RJ; CANTOS Trial Group. Antiinflammatory therapy with canakinumab for atherosclerotic disease. N Engl J Med 2017; 377(12): 1119–1131

[53] Uygur A, Lee RT. Mechanisms of cardiac regeneration. Dev Cell 2016; 36(4): 362–374

[54] Vivien CJ, Hudson JE, Porrello ER. Evolution, comparative biology and ontogeny of vertebrate heart regeneration. NPJ Regen Med 2016; 1(1): 16012

[55] Dodson M, Darley-Usmar V, Zhang J. Cellular metabolic and autophagic pathways: traffic control by redox signaling. Free Radic Biol Med 2013; 63: 207–221

[56] Honkoop H, de Bakker DE, Aharonov A, Kruse F, Shakked A, Nguyen PD, de Heus C, Garric L, Muraro MJ, Shoffner A, Tessadori F, Peterson JC, Noort W, Bertozzi A, Weidinger G, Posthuma G, Grün D, van der Laarse WJ, Klumperman J, Jaspers RT, Poss KD, van Oudenaarden A, Tzahor E, Bakkers J. Single-cell analysis uncovers that metabolic reprogramming by ErbB2 signaling is essential for cardiomyocyte proliferation in the regenerating heart. eLife 2019; 8: e50163

[57] Galow AM, Wolfien M, Müller P, Bartsch M, Brunner RM, Hoeflich A, Wolkenhauer O, David R, Goldammer T. Integrative cluster analysis of whole hearts reveals proliferative cardiomyocytes in adult mice. Cells 2020; 9(5): 1144

[58] Gan P, Patterson M, Sucov HM. Cardiomyocyte polyploidy and implications for heart regeneration. Annu Rev Physiol 2020; 82: 45–61

[59] Pellieux C, Foletti A, Peduto G, Aubert JF, Nussberger J, Beermann F, Brunner HR, Pedrazzini T. Dilated cardiomyopathy and impaired cardiac hypertrophic response to angiotensin II in mice lacking FGF-2. J Clin Invest 2001; 108(12): 1843–1851

[60] Dobaczewski M, Chen W, Frangogiannis NG. Transforming growth factor (TGF)-β signaling in cardiac remodeling. J Mol Cell Cardiol 2011; 51(4): 600–606

[61] Demyanets S, Kaun C, Pentz R, Krychtiuk KA, Rauscher S, Pfaffenberger S, Zuckermann A, Aliabadi A, Gröger M, Maurer G, Huber K, Wojta J. Components of the interleukin-33/ST2 system are differentially expressed and regulated in human cardiac cells and in cells of the cardiac vasculature. J Mol Cell Cardiol 2013; 60: 16–26

[62] Sanada S, Hakuno D, Higgins LJ, Schreiter ER, McKenzie AN, Lee RT. IL-33 and ST2 comprise a critical biomechanically induced and cardioprotective signaling system. J Clin Invest 2007; 117(6): 1538–1549

[63] Xia N, Lu Y, Gu M, Li N, Liu M, Jiao J, Zhu Z, Li J, Li D, Tang T, Lv B, Nie S, Zhang M, Liao M, Liao Y, Yang X, Cheng X. A unique population of regulatory T cells in heart potentiates cardiac protection from myocardial infarction. Circulation 2020; 142(20): 1956–1973

[64] Ambari AM, Setianto B, Santoso A, Radi B, Dwiputra B, Susilowati E, Tulrahmi F, Doevendans PA, Cramer MJ. Angiotensin converting enzyme inhibitors (ACEIs) decrease the progression of cardiac fibrosis in rheumatic heart disease through the inhibition of IL-33/sST2. Front Cardiovasc Med 2020; 7: 115

[65] Ieda M, Tsuchihashi T, Ivey KN, Ross RS, Hong TT, Shaw RM, Srivastava D. Cardiac fibroblasts regulate myocardial proliferation through beta1 integrin signaling. Dev Cell 2009; 16(2): 233–244

[66] Godwin JW, Pinto AR, Rosenthal NA. Macrophages are required for adult salamander limb regeneration. Proc Natl Acad Sci USA 2013; 110(23): 9415–9420

[67] Aurora AB, Porrello ER, Tan W, Mahmoud AI, Hill JA, Bassel-Duby R, Sadek HA, Olson EN. Macrophages are required for neonatal heart regeneration. J Clin Invest 2014; 124(3): 1382–1392

RIGHTS & PERMISSIONS

Higher Education Press