Abstract
Abnormal cardiac development has been observed in individuals with Cornelia de Lange syndrome (CdLS) due to mutations in genes encoding members of the cohesin complex. However, the precise role of cohesin in heart development remains elusive. In this study, we aimed to elucidate the indispensable role of SMC3, a component of the cohesin complex, in cardiac development and its underlying mechanism. Our investigation revealed that CdLS patients with SMC3 mutations have high rates of congenital heart disease (CHD). We utilized heart-specific Smc3-knockout (SMC3-cKO) mice, which exhibit varying degrees of outflow tract (OFT) abnormalities, to further explore this relationship. Additionally, we identified 16 rare SMC3 variants with potential pathogenicity in individuals with isolated CHD. By employing single-nucleus RNA sequencing and chromosome conformation capture high-throughput genome-wide translocation sequencing, we revealed that Smc3 deletion downregulates the expression of key genes, including Ets2, in OFT cardiac muscle cells by specifically decreasing interactions between super-enhancers (SEs) and promoters. Notably, Ets2-SE-null mice also exhibit delayed OFT development in the heart. Our research revealed a novel role for SMC3 in heart development via the regulation of SE-associated genes, suggesting its potential relevance as a CHD-related gene and providing crucial insights into the molecular basis of cardiac development.
Similar content being viewed by others
Congenital heart disease (CHD) is the most common congenital anomaly in newborns, affecting 8-12/1000 live births worldwide and accounting for approximately 40% of prenatal deaths1. CHD arises from abnormal heart development during the embryonic stage and has a strong heritable component. Cardiogenesis is a precise and tightly controlled process involving outflow tract (OFT) formation, the development of four cardiac chambers, and the specialization of cardiac muscle cells2. Many signaling pathways and tissue-specific transcriptional regulators contribute to the regulation of heart development. Dysregulation of the dynamic control of these specific biological processes may lead to CHD. For instance, the Notch pathway plays a crucial role in early OFT formation by arresting proliferation and promoting the differentiation of second heart field progenitors3, and defects in this pathway lead to various CHD phenotypes4.
Super-enhancers (SEs) are a class of regulatory regions comprising clusters of enhancers densely occupied by master regulators and mediators5. SEs exert powerful transcriptional activation effects on genes that are crucial for cell identity6. Previous evidence suggested that SEs play a significant role in the genetic control network of the heart7,8,9. For example, TGF-β-mediated binding of the chromatin reader protein BRD4 to SEs can promote the expression of genes involved in cardiac fibroblast activation8. A conserved SE controls the expression of Nppa and Nppb in heart development and homeostasis. Deletion of this SE results in manifestations similar to those in Nppa-Nppb knockout mice, indicating the physiological importance of this SE in cardiac disease9. Although a study analyzing a large database generated a comprehensive catalog of SEs in 86 human cells and tissue types, including the left ventricle (LV)10, the identification and in-depth analysis of heart-specific SEs remain necessary.
Many SEs are distally related to the genes they control and often regulate their target genes through chromatin loops, which are mediated by cohesin11. Cohesin, a ring-shaped protein complex that acts as a chromatin glue, is loaded onto chromatin by NIPBL to form a chromatin loop through DNA extrusion12, uniting distant SEs with promoters to coordinate gene expression13. Cornelia de Lange syndrome (CdLS, OMIM122470) is a rare genetic disorder resulting from mutations within genes encoding components of cohesin, including NIPBL (65%), HDAC8 (8%), SMC1A (5%) and SMC3 (1–2%)14,15,16. Notably, a considerable proportion of individuals with CdLS also exhibit CHD, which contributes significantly to morbidity and mortality rates17. An analysis of cardiac phenotypes in model animals revealed that NIPBL deficiency contributes to heart defects associated with cohesinopathy. Nipbl-heterozygous mouse models constructed from Cre expressed in diverse cardiac cell lineages of mouse embryos show atrial septal defects (ASDs) at a high frequency due to subtle global transcriptional dysregulation18. Nipbl morphant zebrafish exhibit heart jogging/looping defects in the context of abnormal endoderm development19. However, the detailed mechanism by which cohesin defects lead to CHD remains unclear, particularly in cases involving mutations in other cohesin components.
In this study, we elucidated the fundamental significance of SMC3 in heart development, as it effectively controls the expression of master genes crucial for cardiogenesis by regulating specific spatial chromatin structures containing SEs. Heart-specific Smc3-knockout (SMC3-cKO) mice exhibited varying degrees of OFT malformations, including malformations of the nonseparated aorta and pulmonary artery, transposition of the great arteries (TGA), and membranous ventricular septal defects (VSDs). We also observed 16 rare variants of SMC3 with potential pathogenicity in patients with isolated CHD, with phenotypes similar to those observed in SMC3-cKO mice. These findings shed light on the essential function of SMC3 in heart development and establish an association between SMC3 and isolated CHD.
Materials and methods
RNA scope
An RNA scope assay (Biotend Biotechnology, Shanghai, China) was used to analyze Smc3 and Ets2 expression in mouse hearts. Samples were analyzed using the RNA scope 2.5 HD Reagent-Red kit (Advanced Cell Diagnostics, CA, United States) and the RNA scope H2O2 & Protease Plus Reagents kit (Advanced Cell Diagnostics, CA, United States). The general processes are described below. Sections were deparaffinized using xylene and 100% ethanol and incubated with a hydrogen peroxide solution for 10 min at room temperature. Afterward, sections were incubated in the target retrieval reagent solution for 15 min at 99 °C, and protease was added for digestion. Sections were hybridized with targeted probes against Smc3 (Probe-Mm-Smc3-C1) and Ets2 (Probe-Mm-Ets2-C2). Internal positive controls (against Polr2a, Ppib, and Ubc) and a negative control (against the bacillus gene DapB) were used in addition to staining the same tissue to ensure the specificity of the targeted probes. The sections were subjected to signal amplification using AMP1 for 30 min, AMP2 for 30 min, and AMP3 for 15 min and then incubated with the appropriate fluorescent dye. Finally, HRP blockers were added dropwise, and the sections were counterstained with DAPI for 1 min. The prepared sections were sealed with blocking reagents to avoid fluorescence quenching.
RNA extraction, reverse transcription, and quantitative real-time PCR
Total RNA was extracted from mouse heart tissues and cells using TRIzol reagent (Invitrogen, CA, USA) according to the manufacturer’s instructions. After the evaluation of the quality and concentration, RNA was reverse transcribed to cDNA using a Prime Script RT Reagent Kit (Takara, Shiga, Japan), and quantitative real-time PCR (qPCR) was conducted using SYBR Premix Ex Taq™ (Takara, Shiga, Japan) on a StepOnePlusTM Real-Time PCR System (Thermo Fisher Scientific, MA, USA). The relative expression level of mRNA was normalized to that of GAPDH and calculated using the relative quantification method (2−ΔΔCt). The primers used are shown in Supplementary Table 1.
Construction of heart-specific Smc3-knockout mice
Smc3fl/+ mice (C57BL6/J background) were generated using CRISPR-Cas9 technology at Biomodel Organism Science & Technology Development Co., Ltd., Shanghai, China. The loxp sites flanking exon four were inserted via homologous recombination. Nkx2.5-Cre transgenic mice (C57BL6/J background), which express Cre recombinase in their heart-forming fields in the splanchnic mesoderm at embryonic day (E)7.5, were obtained from GemPharmatech, Nanjing, China. The Smc3fl/+ mice were bred with Nkx2.5-Cre mice to generate mice with a deletion of Smc3 in cardiac progenitors. The genotyping primers used are listed in Supplementary Table 1. The efficiency of Smc3 knockout in the embryonic heart was tested via qPCR. The embryos were dated, and the morning of the vaginal plug was recorded as E0.5. All transgenic mice were backcrossed to wild-type mice (C57BL6/J background) once after each two to three generations of mating. All mouse procedures were approved and monitored by the Research Ethics Committee of the School of Basic Medical Sciences, Fudan University, China. The approval number is 20160520-3.
Observation of cardiac phenotypes in mice
Pregnant mice and neonates were anesthetized by isoflurane inhalation and euthanized by rapid cervical dislocation. Embryos and neonates were collected at specific stages and photographed in phosphate-buffered saline (PBS) using a Leica M205C microscope (Leica, Heilbronn, Germany). The ventricular area and compact layer of ventricular myocardium thickness of the mouse hearts were measured using ImageJ (http://rsb.info.nih.gov/ij). Hematoxylin–eosin (H&E) staining was used to observe the detailed anatomical structure of the mouse hearts and was performed by Biossci Biotechnology Co., Ltd., Shanghai, China.
Clinical subjects and ethics statement
All patients with isolated CHD were recruited from the Children’s Hospital of Fudan University and were diagnosed via echocardiography using published diagnostic criteria. The patients had no other physical abnormalities. A total of 104 healthy Han Chinese individuals recruited from the physical examination department were used as controls. The studies involving human subjects were approved by the Medical Ethics Committee of the Children Hospital of Fudan University, and the approval number is 2016-121.
Sanger sequencing
Whole blood was collected from 56 patients with isolated CHD, including 34 TOF patients and 22 VSD patients, and 104 Han Chinese controls for Sanger sequencing to detect SMC3 variants. Sanger sequencing was subsequently performed at the Medical Laboratory of Nantong ZhongKe, Nantong, China. PCR products were sequenced using an ABI 3730 Genetic Analyzer with the respective forward or reverse primers. Chromatograms were analyzed using Mutation Surveyor DNA Variant Analysis Software (https://softgenetics.com/).
Multiplex PCR-targeting sequencing
The cohort of 1083 patients with isolated CHD20 included 234 tetralogy of Fallot (TOF) patients, 718 VSD patients, 55 pulmonary atresia+VSD patients, 19 double outlet right ventricle + pulmonary stenosis patients and 57 right ventricular hypoplasia patients. Multiplex PCR-targeting sequencing was applied to sequence all exons, splice sites (±20 bp of adjacent intronic sequences) and regulatory regions (5000 bp upstream of the transcription start site, 300 bp downstream of the transcriptional termination site) of SMC3. The primer design covered a theoretical 99.9% of the complete target region. The prepared samples were sequenced on the HiSeq 2500 (Illumina, California, USA) platform. Burrows‒Wheeler Aligner software (version 0.7.17), SAMtools (https://samtools.sourceforge.net/), and Genome Analysis Toolkit (https://gatk.broadinstitute.org/hc/en-us) were used to map sequence reads to hg19 to calculate the read quality and call single nucleotide variants.
Identification of SMC3 variants
Sequence variants were functionally annotated using the ANNOVAR program with annotation databases, including the Reference Sequence (RefSeq, hg19) collection, the pathogenicity prediction databases (the Combined Annotation Dependent Depletion (CADD) (version 1.3), Sorting Intolerant From Tolerant (SIFT) (http://sift-dna.org/sift4g), the polymorphism phenotyping v2 (Polyphen2) (http://genetics.bwh.harvard.edu/pph2/), and reference databases including normal sample sequencing data (the Genome Aggregation Database (GnomAD) (http://gnomad.broadinstitute.org/) and the Exome Aggregation Consortium (ExAC) (http://exac.broadinstitute.org)).
The effect of each screened variant in the promoter and 5’ UTR regions on regulatory DNA elements (DNase hypersensitivity regions, binding sites of transcription factors, and promoter regions) was evaluated via RegulomeDB (https://regulomedb.org/regulome-search/).
For each screened variant at the 3’ UTR, the effect on mRNA–microRNA interactions was predicted using TargetScanHuman 7.0 (http://www.targetscan.org).
Genomnis HSF Pro (https://hsf.genomnis.com/mutation/analysis) was used to evaluate changes in splicing signals and to analyze the effect of variants in introns on consensus splice sites.
The protein sequences of SMC3 from multiple species were obtained from the NCBI database and compared using ClustalW in MEGA11 software (v11.0.13). The evolutionary tree was constructed and tested using the neighbor-joining algorithm and the bootstrap test, respectively. The number of iterations was 1000.
Dual-luciferase reporter assay
The recombinant luciferase plasmids (pGL3-basic-SMC3-wild-type, pGL3-basic-SMC3-498insTGGGG, and pGL3-basic-778_776delATG) were constructed with the corresponding sequences amplified from human genomic DNA. The plasmids were sequenced and shown to be consistent with the sequences in the National Center for Biotechnology Information database. Cells were seeded in 96-well plates (1 × 104 cells per well) and incubated at 37 °C overnight. The pGL3-basic-SMC3-wild-type, pGL3-basic-SMC3-498insTGGGG, and pGL3-basic-778_776delATG plasmids were transfected into cells using Lipofectamine 3000 (Invitrogen, CA, USA). The cells were collected 48 h after transfection. Both firefly and Renilla luciferase activities were measured using a dual-luciferase reporter assay (Promega, Wisconsin, USA). The firefly luciferase activities were normalized to the Renilla luciferase activities.
Single-nucleus RNA sequencing
E12.5 mouse embryos were microdissected in PBS solution using a Leica M205C microscope (Leica, Heilbronn, Germany), and the heart tissues were extracted. Whole heart samples from different embryos, including six Smc3fl/fl embryos, seven Smc3fl/+; Nkx2.5-Cre embryos, and seven Smc3fl/fl; Nkx2.5-Cre embryos, were subjected to mechanical trituration to prepare cell suspensions. Cell counts and viability were determined for the resulting suspension, and the cell concentration was adjusted to the ideal of 300–600 cells/μL. The prepared single-cell suspension was combined with a mixture of gel beads containing barcode information and enzymes and then encapsulated by oil droplets to form a gel bead-in-emulsion (GEM). After the reverse transcription reaction, the GEMs were fragmented to complete the amplification of nuclear cDNA, and the quality-checked amplification products were constructed as sequencing libraries. Briefly, the cDNA was broken into fragments of 200–300 bp, end-repaired, ligated to the P7 adapter, and finally amplified by PCR for introduction into the indices of the samples. Sequencing libraries were created by integrating the screened cDNA fragments, and single-nucleus RNA sequencing (snRNA-seq) was performed on the NovaSeq 6000 (Illumina, California, USA) platform.
Single-nucleus RNA sequencing data analysis
Bioinformatics analysis was performed by CapitalBio Technology (Beijing, China). For each sample, the cleaned data were generated using Cell Ranger (v3.0.2) (https://github.com/10XGenomics/cellranger) and filtered for low-quality reads. The data were aligned to the mouse mm10 reference genome. The 10× Genomics‐derived data (Smc3fl/fl, Smc3fl/+; Nkx2.5-Cre, and Smc3fl/fl; Nkx2.5-Cre) were collected. Downstream analysis using the Seurat (v4.0.5) (https://satijalab.org/seurat) program revealed cells with gene numbers greater than 200, mitochondrial gene ratios less than 25%, and 2000 genes with highly variable expression in 3 or more cells. Principal component analysis was then performed using significantly differentially expressed genes, and uniform manifold approximation and projection (UMAP) was performed to visualize the data. Cells were represented in a two-dimensional UMAP plane, and clusters were identified and annotated according to previously published cardiac cell markers. The significance of differential expression was calculated using the bimod test. The functional enrichment of the differentially expressed genes was then determined by performing Gene Ontology (GO) (http://www.geneontology.org/) analysis. The pseudotime trajectory was determined using the Monocle2 package with the default settings. Specifically, the DDRTree method was used for dimension reduction with max_components set at 2, and the cells were ordered using the orderCells function.
Construction of SMC3-knockdown and SMC3-overexpressing myocardial cell lines
The mouse cardiomyocyte cell line HL-1, human cardiomyocyte cell line AC16 and HEK293T cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum (FBS) (Gibco, Australia) and 1% Pen-Strep antibiotics (Yeasen, Shanghai, China) at 37 °C in 5% CO2.
SMC3 was knocked down using CRISPR-Cas9. The SMC3 CRISPR guide RNA (gRNA) was designed using a web tool developed by the Feng Zhang group at MIT and subcloned between the two BsmBI sites of the lentiCRISPRv2 vector from Feng Zhang’s laboratory. The SMC3 and Smc3 gRNA target sequences were 5’-CGGCCCTTACCGGCCCATGA-3’ and 5’-CATGCTGTTGAAGAAGAAGG-3’, respectively.
SMC3 was tagged at the N- and C-termini with an HA tag and then cloned and inserted into the mammalian expression vector pCDH-CMV-IRSE-Blast (pCDH-SMC3-WT). Two variant plasmids (pCDH-SMC3-T857S and pCDH-SMC3-Y434C) were generated using the KOD-Plus Mutagenesis Kit (Toyobo, Tokyo, Japan). The pCDH-ETS2 plasmid (pCDH-CMV-IRSE-Blast) was obtained from GeneRay (Shanghai, China). All expression plasmids were fully sequenced.
For the construction of stable cell lines, ViaFect™ Transfection Reagent (Promega, Wisconsin, USA) was used to package pseudoviral particles of gRNA in HEK293T cells. The virus-containing supernatant was collected, filtered, and used to infect cells in complete media supplemented with 5 µg/mL polybrene (Sigma‒Aldrich, Massachusetts, USA). Puromycin dihydrochloride and blasticidin (Yeasen, Shanghai, China) were used to screen stably transfected Smc3-knockdown (KD) and Ets2-overexpressing cells, respectively, at 48 hours postinfection.
For the construction of transiently infected cell lines, three plasmids (pCDH-SMC3-WT, pCDH-SMC3-T857S, and pCDH-SMC3-Y434C) diluted to the same concentration (300 ng/μl) were transfected into SMC3-KD AC16 cells using the ViaFect™ Transfection Reagent (Promega, Wisconsin, USA). The transfected cells were harvested and used for subsequent experiments 48 hours later.
Western blotting and antibodies
Cells were harvested and lysed in RIPA buffer (Yeasen, Shanghai, China) containing a protease inhibitor cocktail (Sigma‒Aldrich, Massachusetts, USA). Proteins were quantified using a Bradford Protein Assay Kit (Abcam, MA, USA), and the same amounts of proteins were separated on 10% sodium dodecyl sulfate (SDS)-PAGE gels and transferred to nitrocellulose membranes, which were blocked with 8% milk dissolved in Tris-buffered saline and Tween for 1 h at room temperature. The membranes were then incubated with specific primary antibodies at 4 °C overnight, followed by an incubation with secondary antibodies at room temperature for 1 h. The antibodies used are listed in Supplementary Table 1.
Bulk RNA-seq and analysis
Bulk RNA-seq of mouse embryo hearts and HL-1 cells was performed by LC-Bio Technology (Hangzhou, China). Total RNA was isolated and purified using TRIzol reagent (Invitrogen, CA, USA). The RNA samples that passed quality control were fragmented into small pieces at 94 °C for 5–7 min and reverse-transcribed to generate cDNA. The average insert size for the final cDNA library was 300 ± 50 bp. The FASTQ sequencing data were trimmed via paired-end sequencing. Consecutively, the trimmed reads were mapped to the reference mouse genome (mm10) with HISAT2 software (https://ccb.jhu.edu/software/hisat2). The mRNA expression abundance was analyzed using StringTie software and represented as FPKM. The differentially expressed mRNAs were selected with FC > 1.5 and a parametric test comparing nested linear models (p value < 0.05) using the R package edgeR (https://bioconductor.org/packages/release/bioc/html/edgeR.html). GSEA was performed on all expressed genes that were ranked by log2FC value. The functional enrichment of differentially expressed genes was then assessed by performing GO (http://www.geneontology.org/) analysis.
Chromatin immunoprecipitation assay
HL-1 cells (5–10.0 × 106) were crosslinked with 1% formaldehyde and quenched in a glycine solution (at a final concentration of 0.125 M). Chromatin was prepared as described above and digested with micrococcal nuclease to mainly generate mononucleosomes with a minor fraction of dinucleosomes. The reaction was stopped by the addition of 8 μL of solution containing 0.2 M EDTA and 0.2 M EGTA. The chromatin was incubated overnight at 4 °C with 2 μL of rabbit anti-SMC3 antibody (Abcam, MA, USA), 1 μL of rabbit anti-H3K4me3 antibody (Millipore, Darmstadt, Germany), 1 μL of rabbit anti-H3K27ac antibody (Millipore, Darmstadt, Germany), or 2 μL of control rabbit IgG (CST, MA, USA). Protein A/G magnetic beads (Thermo Fisher Scientific, MA, USA) were added and incubated for an additional 4 h, after which the immunoprecipitates were washed vigorously, and the DNA was purified according to the manufacturer’s instructions for the SimpleChIP® Plus Enzymatic Chromatin IP Kit (CST, MA, USA).
For chromatin immunoprecipitation–sequencing (ChIP-seq), ChIP products were end repaired, dA-tailed and linkers were ligated, and barcoding and Illumina adapter addition were performed by PCR amplification. Libraries were purified using QiaQuick PCR purification reagents (Qiagen, Hilden, Germany), and 0.7× and 0.2× Ampure XP beads (Beckman, California, USA) were used for size selection.
For ChIP‒qPCR, ChIP-enriched DNA samples were quantified by qPCR to determine the binding of SMC3 to the Ets2 promoter. The values are shown as relative enrichment normalized to that of IgG. The primers used for ChIP‒qPCR are shown in Supplementary Table 1.
Identification of SEs
SEs were defined by H3K27ac peaks in the mouse cardiomyocyte cell line HL-1 from ChIP-seq data using the rank ordering of SEs (ROSE) algorithm (https://github.com/stjude/ROSE). The details of this procedure were similar to those in previous research6. SE-associated genes were screened through a conjoint analysis of transcriptomic data from HL-1 cells. The specific steps were as follows: 1. A total of 1544 genes were filtered because they were most closely related to SEs. 2. Using FPKM values, 308 genes in the top 20% of the 1544 genes were identified as SE-associated genes.
3C-high-throughput genome-wide translocation sequencing
A total of 1 × 107 HL-1 cells were subjected to cross-linking by incubating them for 10 min on ice in 10 mL of RPMI-1640 containing 10% FBS and 2% formaldehyde. The reaction was stopped by adding glycine to a concentration of 0.125 M and incubating the mixture for 5 min at room temperature. Pelleted nuclei were incubated for 1 h at 37 °C in 0.5 mL of 1× CutSmart digestion buffer containing 0.3% SDS. The chromatin was then digested by the addition of 400 U of MboI (NEB, MA, USA) overnight at 37 °C, and the reaction was terminated by the addition of 80 μL of 10% SDS. Decrosslinking was performed by adding 30 μL of 10 mg/mL proteinase K and 3 μL of 100 mg/mL RNaseA and incubating the sample for 4 h at 65 °C. A total of 10 μg of 3 C sample was sonicated for 3 min using a Model 550 Sonic Dismembrator (Thermo Fisher Scientific, MA, USA) for 15 s ON and 25 s OFF. As determined by the agarose gel analysis, the chromosomal DNA was reduced to an average size of 300–500 bp. Sonicated DNA was linearly amplified using a biotinylated primer (Supplementary Table 1) that anneals to the selected MboI (NEB, MA, USA) fragment of the Ets2 promoter. The biotin-labeled single-stranded DNA products were enriched with streptavidin C1 beads (Thermo Fisher Scientific, MA, USA) and ligated with HTGTS adaptors containing a 6-nucleotide overhang. The adaptor-ligated products were amplified using a nested primer that anneals upstream of the biotinylated primer and an HTGTS-adaptor-complementary primer (Supplementary Table 1). The products were prepared for sequencing on the Illumina MiSeq platform (Illumina, California, USA) after amplification using the P5-I5 and P7-I7 primers. After quality control, filtered reads were extracted from the sequence file through Cutadapt and Pear (v0.9.6). Paired-end reads containing NestPrimer or AdapterPrimer were obtained, and then the reads were filtered by searching for restriction site sequences. The first MboI enzyme fragment was extracted and mapped to the mouse genome mm10 using Bowtie2 (version 2.3.4.3). The read numbers were counted and normalized to the total mapped reads per sample after self-ligation, relegation, and dumping. For visualization, we converted the final bam file to a bedGraph file. The signal peak was obtained by postcomparison filtering, signal statistics and normalization.
Statistical analysis
All experiments were repeated more than twice, and all data are presented as the means±standard deviations. Statistical tests were performed using GraphPad Prism Software (v7). Two-tailed unpaired Student’s t tests with or without Welch’s correction or Mann‒Whitney U tests were used to analyze the differences between two groups. A p value < 0.05 was regarded as statistically significant.
Results
Cornelia de Lange syndrome patients with SMC3 mutations have high rates of congenital heart disease
Mutations in genes encoding components of the cohesin complex are associated with a wide range of effects on heart development in individuals with CdLS14,21,22. We conducted an extensive literature search and extracted detailed clinical information for 410 CdLS patients documented in 20 papers to gain deeper insights into the origin of cohesinopathy-related heart defects (Supplementary Table 2). None of these patients were from consanguineous families. An analysis of the data revealed that approximately one-quarter of the CdLS patients exhibited congenital heart defects. The occurrence of CHD was as follows: NIPBL mutation carriers (26.73% or 58/217), RAD21 mutation carriers (25.58% or 11/43), SMC1A mutation carriers (24.14% or 14/58), and HDAC8 mutation carriers (16.33% or 8/49). Notably, patients with SMC3 mutations had a significantly greater frequency of CHD (52.4% or 22/42) (Supplementary Table 2). These findings suggest that mutations in SMC3 may have a more pronounced impact on heart development than mutations in other genes encoding cohesin components.
Heart-specific Smc3-knockout mice exhibit defects in outflow tract development
We detected Smc3 expression in mouse heart tissues via the RNA scope assay, a recently developed in situ hybridization technique that allows sensitive and specific localization of gene expression in cells and tissues, to clarify the role of SMC3 in cardiac development. Smc3 was found to be widely expressed across the entire heart at E10.5 and postnatal day (P) 0 (Fig. 1a). Smc3 was expressed at a relatively high level at E9.5 and decreased at E10.5. Afterward, it maintained a relatively low expression level (Fig. 1b). These results indicate that Smc3 expression in the heart is widespread and dynamic in early embryonic stages.
a Location of Smc3 expression in mouse heart at embryonic day (E) 10.5 (top panel) and postnatal day (P) 0 (bottom panel). RNA fluorescence in situ hybridization was performed using the RNA scope platform with probes targeting Smc3. Image scale bars, 0.2 mm. b Temporal expression pattern of Smc3 in mouse hearts from E9.5 to P7 determined using quantitative real-time PCR (qPCR). c Genotype distribution of embryos and newborn mice resulting from Smc3fl/fl and Smc3fl/+; Nkx2.5-Cre crosses. d Representative necropsy images of the ventricular size, length of the outflow tract (OFT), and relative location of the aorta (AO) and pulmonary artery (PA) in Smc3fl/fl, Smc3fl/+; Nkx2.5-Cre, and Smc3fl/fl; Nkx2.5-Cre embryos at E9.5, E10.5, E11.5, E12.5, and E13.5. Image scale bars, 1 mm. e Representative images of hearts from fetal and newborn Smc3fl/fl and Smc3fl/+; Nkx2.5-Cre mice at E14.5, E18.5, and P0. Compared with Smc3fl/fl hearts, Smc3fl/+; Nkx2.5-Cre hearts have different degrees of abnormal arrangement of the large arteries. Image scale bars, 2 mm. f H&E staining of heart sections from Smc3fl/fl and Smc3fl/+; Nkx2.5-Cre mice at E14.5 and P0. A ventricular septal defect (VSD) was observed in 6/18 Smc3fl/+; Nkx2.5-Cre mice at E14.5. In 2/18 Smc3fl/+; Nkx2.5-Cre mice at E14.5, the PA connects to the left ventricle (LV), and the AO connects to the right ventricle (RV). Scale bars for images of E14.5 embryos, 0.5 mm; scale bars for images of P0 mice, 0.25 mm. g H&E staining of heart sections from Smc3fl/fl and Smc3fl/+; Nkx2.5-Cre mice at E18.5 (left panel) and quantification of the RV wall compact zone thickness at E14.5 and E18.5 (right panel). The compact layer of the ventricular myocardium is indicated by the black line. Image scale bars, 0.25 mm. The error bars indicate the means ± standard deviations. *P < 0.05 and **P < 0.01. TGA transposition of the great arteries, RA right atrium, LA left atrium.
Then, we generated SMC3-cKO mice by crossing Smc3-floxed (Smc3fl/+) mice with Nkx2.5-Cre mice (Supplementary Fig. 1a, b) and confirmed efficient knockout at E10.5 using qPCR (Supplementary Fig. 1c). The birth rate of the SMC3-cKO mice did not conform to the Mendelian ratio (Fig. 1c, Supplementary Fig. 1d), indicating embryonic lethality. Moreover, Smc3fl/fl; Nkx2.5-Cre embryos were absent at E14.5 (Fig. 1c, Supplementary Fig. 1d). During heart looping (E9.5 to E10.5), a significant proportion (12/20) of Smc3fl/fl; Nkx2.5-Cre embryos exhibited a short OFT and small right ventricle (RV) compared to Smc3fl/fl embryos (Fig. 1d). During normal heart development, the OFT is separated into the aorta and pulmonary artery from E11.5 to E13.5. However, we did not observe a tendency toward OFT separation in 6 of the 12 Smc3fl/fl; Nkx2.5-Cre embryos at E11.5, and the malformation rate increased to 73.3% (11/15) at E12.5. All surviving Smc3fl/fl; Nkx2.5-Cre embryos presented this type of abnormality, especially at E13.5 (Fig. 1d), which may have led to their death after this time point. Considering the dual roles of cohesin in embryonic development23,24, we propose that the embryonic lethality in Smc3fl/fl; Nkx2.5-Cre embryos may be attributed to defective cell proliferation.
On the other hand, Smc3 heterozygous mutations have been associated with abnormal cardiac development in CdLS patients, possibly resulting from the dysregulated expression of genes involved in heart development24. We analyzed the phenotype of Smc3fl/+; Nkx2.5-Cre embryos to investigate this outcome further. Among the total number of embryos examined (80), only a small proportion (8/80) exhibited OFT developmental defects (Fig. 1d). Surviving Smc3fl/+; Nkx2.5-Cre embryos at E14.5, E18.5, and P0 showed malalignment of the aorta and pulmonary artery, along with small RVs (Fig. 1e). At E14.5, 11% (2/18) of Smc3fl/+; Nkx2.5-Cre embryos exhibited TGA, while 33% (6/18) displayed a membranous VSD (Fig. 1f). Furthermore, at P0, one of ten Smc3fl/+; Nkx2.5-Cre mice had a VSD (Fig. 1f). In addition to structural abnormalities, we noticed a thinner compact layer of the ventricular myocardium in Smc3fl/+; Nkx2.5-Cre embryos at E14.5 and E18.5 (Fig. 1g). Notably, cohesin haploinsufficiency has a mild impact on cell proliferation25. Therefore, the observed malformations in Smc3fl/+; Nkx2.5-Cre mice likely resulted from the dysregulation of genes related to heart development. Notably, due to the early embryonic lethality of SMC3-cKO embryos, we were unable to ascertain the extent of SMC3-dependent developmental progress. Consequently, we cannot rule out structural malformations of the LV as a potential cause for incomplete embryos.
SMC3 variants are identified in patients with isolated congenital heart disease
De novo mutations in SMC3 contribute to approximately 1–2% of CdLS-like phenotypes, with a significant proportion displaying noncanonical CdLS phenotypes14. Upon examination of the CHD type of CdLS individuals26,27,28,29,30,31 with SMC3 mutations, we found that a majority (76.2% or 16 out of 21) exhibited aorta and pulmonary artery dysplasia, consistent with the phenotypes of SMC3-cKO mice (Fig. 2a). We extracted DNA samples for Sanger sequencing from 56 individuals with isolated CHD, including those with TOF and VSD, corresponding to the heart defects observed in SMC3-cKO mice, to understand the relevance of SMC3 variants in CHD patients without a distinct CdLS phenotype. Two candidate pathogenic variants (R221T and T857S) in the coding region were identified in two TOF patients based on the following filtering criteria: 1) excluding synonymous variants; 2) variants with a minor allele frequency (MAF) less than 0.1% according to the public control database; and 3) the variant type was a nonsense, frameshift or missense variant predicted to be deleterious by at least two algorithms (Fig. 2b, c, Supplementary Table 3). Subsequently, through multiplex PCR-targeting sequencing, we searched for SMC3 variants in a larger cohort of 1083 patients with isolated CHD from our previous study20 and found one candidate pathogenic variant (Y434C) in the coding region from a patient with TOF after filtering using the same criteria. We also detected six variants with potential pathogenicity in the noncoding region from patients with VSD or TOF according to a MAF less than 0.1%, and these variants are predicted to affect SMC3 expression (located on regulatory DNA elements, affect the consensus splice site, or influence microRNA binding) (Supplementary Table 3). Furthermore, seven SMC3 candidate pathogenic variants in the coding region were screened from our peers’ whole exome sequencing data related to TOF or VSD without known pathogenic factors (Fig. 2c, Supplementary Table 3). In summary, a total of 16 SMC3 candidate pathogenic variants (ten coding and six noncoding) were detected across 14 patients with TOF or VSD. The ten coding variants, including eight missense variants, one nonsense variant, one frameshift variant, and three other variants (R221T, Y434C, and T857S) were verified by Sanger sequencing (Fig. 2b) due to the availability of DNA samples. We also noticed that the missense variants (L49I and R221T) are located in the ATPase head domain, and one (G534C) is in the hinge domain of SMC3 (Fig. 2f). Then, wild-type or mutated SMC3 was overexpressed in the SMC3-KD cardiomyocyte cell line AC16 to explore the effects of these variants on protein expression. We found that two variants (Y434C and T857S) led to a reduction in SMC3 expression at the protein level (Fig. 2d). Furthermore, the evolutionary tree of eleven species indicated that SMC3 is highly conserved among mammalian species (Fig. 2e). The high conservation of these variants across multiple species illustrated their constant and significant role during species evolution (Fig. 2f). We also performed Sanger sequencing on DNA samples from 104 Han Chinese controls to further verify the rarity of these variants and did not find any of these SMC3 variants through a sequence analysis (Supplementary Fig. 2a).
a Statistical overview of the types of congenital heart disease (CHD) in Cornelia de Lange syndrome (CdLS) patients with SMC3 mutations. The inner layer represents the pathological categorization of CHD types. The outer layer shows CHD types. The digit in brackets indicates the number of CdLS patients. b Sanger sequencing chromatograms of three SMC3 coding variants (R221T, T857S, and Y434C). The SMC3 variant (Y434C) was screened by multiplex PCR-targeting sequencing and verified by Sanger sequencing. The black arrows in the upper panels indicate variant sites. The lower panels show wild-type (WT) sequences. c General information on the ten coding variants. Red markings represent increased pathogenicity. See Supplementary Table 3 for further details. Categorical predictions from each algorithm. ExAC contains high-quality exome sequencing data for 60,706 individuals from different ancestries. The values represent the frequencies of variant occurrence. SIFT, “T”=tolerated, “D”=damaging. The numbers represent the probabilities of similarity between the variant and the actual variant as calculated using the SIFT algorithm, with values less than 0.05 indicating significant differences. PolyPhen2 HDIV, “B”=benign, “P”=possibly damaging, “D”=likely damaging. Numbers represent the probabilities that the variant changes the protein structure and function as calculated using the HumDiv algorithm. CADD, the numbers represent the score of the deleteriousness of the variant based on several factors, including the polymorphism of the allele and the pathogenicity of the variant, with a higher value representing a greater probability that the variant is “deleterious”. A phred score >20 is generally considered “deleterious”. d The effects of variants on SMC3 expression. The level of SMC3 in SMC3-knockdown (KD) AC16 cells overexpressing SMC3-WT was higher than that in cells overexpressing the variant. e Evolutionary tree of SMC3 in eleven species. Each node represents a taxon. The numbers indicate the reliability of the branch, and a larger value indicates greater reliability. The branch lengths represent the genetic variability of the sequence and are calculated based on a distance scale. f Protein domain plot of SMC3 and the site (top panel) and conserved region (bottom panel) of ten coding variants. The color of each arrow pointing to the variant represents the variant type. All identified SMC3 variants are indicated at the corresponding positions. TOF tetralogy of Fallot, PDA patent ductus arteriosus, PS pulmonary stenosis, BAV mitral and aortic valve, AS aortic stenosis, ASD atrial septal defect, PPS peripheral pulmonary stenosis, PAH pulmonary arterial hypertension, NC normal control cells transfected with a blank vector, ExAC exome aggregation consortium, SIFT Sorting Intolerant From Tolerant, Polyphen2 polymorphism phenotyping v2, HDIV HumDiv model, CADD combined annotation-dependent depletion, WES whole-exon sequencing.
The cardiomyocyte cell line AC16 was transfected with luciferase reporter plasmids containing the wild-type SMC3 promoter and two mutant SMC3 promoters (498insTGGGG and -778_-776delATG) to investigate the impact of noncoding variants on SMC3 transcription (Supplementary Fig. 2b). Dual-luciferase reporter assays revealed that the two variants reduced promoter activity, suggesting their potential roles in downregulating SMC3 expression. Additionally, qPCR analysis revealed significantly downregulated SMC3 expression in injured heart tissues from TOF patients compared to normal heart tissues, the cDNAs of which were obtained from the remaining preserved samples from previous research projects32 (Supplementary Fig. 2c). Taken together, these findings suggest an association between SMC3 variants and isolated CHD.
The deletion of SMC3 resulted in altered expression profiles of master genes in cardiac muscle cells
We focused on E12.5, a stage in which SMC3-cKO mice do not exhibit embryonic lethality and OFT separation begins, to obtain insights into the precise cellular and molecular events driven by SMC3 during heart development. Compared with other organs, the heart mainly contains cardiomyocytes, which are large and single cells cannot be easily captured via microfluidics; therefore, we performed snRNA-seq33,34 on cardiac tissues derived from Smc3fl/fl, Smc3fl/+; Nkx2.5-Cre and Smc3fl/fl; Nkx2.5-Cre mouse embryos at E12.5 using the 10× Genomics technique, generating 8923, 10,057, and 10,233 nuclei, respectively (Fig. 3a). All SMC3-cKO embryonic hearts used for sequencing showed defects in OFT development. Normalized snRNA-seq profiles at the Smc3 locus in Smc3fl/fl and SMC3-cKO embryo cardiac tissue further validated the knockout efficiency (Supplementary Fig. 3a).
a Schematic of single-nucleus RNA-seq (snRNA-seq). Uniform manifold approximation and projection (UMAP) representation of single-cell transcriptomes derived from E12.5 Smc3fl/fl, Smc3fl/+; Nkx2.5-Cre, and Smc3fl/fl; Nkx2.5-Cre embryonic cardiac tissue (b) and CMs (d). Dot plots showing the expression of individual marker genes across the cardiac tissue (c) and CM (e) lineages. The color from dark to light represents the average expression from high to low. The size of the dot represents the percentage of cells within that cluster expressing the indicated gene. f Heatmap illustrating scaled expression of CM differentiation master genes in CMs and OFT development master genes in outflow tract like (OFTL)-CMs from E12.5 Smc3fl/fl, Smc3fl/+; Nkx2.5-Cre, and Smc3fl/fl; Nkx2.5-Cre cardiac tissue. The color from dark to light indicates the scaled expression from low to high. g Bar plots displaying representative Gene Ontology (GO) terms in which downregulated genes were enriched in OFTL-CMs from E12.5 Smc3fl/+; Nkx2.5-Cre (top panel) and Smc3fl/fl; Nkx2.5-Cre (bottom panel) cardiac tissues. h Pseudotemporal ordering trajectory map of OFTL-CMs from Smc3fl/fl, Smc3fl/+; Nkx2.5-Cre, and Smc3fl/fl; Nkx2.5-Cre embryonic cardiac tissues. Dark to light colors represent the pseudotime order. Fb_like fibroblast-like cell, Epi epicardial cell, EC endothelial cell, A-CM atrium-cardiac muscle cell, LV-CM left ventricle-cardiac muscle cell, AVC-CM atrioventricular canal-cardiac muscle cell.
The cells were clustered into 27 types based on marker gene expression (Supplementary Fig. 3b), and specific cell types were assigned using the cardiac muscle cell (CM) marker genes Myh6 and Ankrd1, endothelial cell (EC) genes Pecam1 and Flt1, and the epicardial cell (Epi) genes Wt1 and Tbx18 (Fig. 3b, c, Supplementary Fig. 3c, d). Additionally, we identified fibroblast-like cells (Fb_like) and mural cells based on the expression of Postn and Dlk1, respectively (Fig. 3b, c). Immune and blood cells were distinguished by the expression of Dock2 and Slc4a1 (Fig. 3b, c). Our particular focus was on CMs, as disruptions in their differentiation or function can cause arrested OFT remodeling and separation, diminished pump function, arrhythmias and ultimately death35. Compared with CMs from Smc3fl/fl mice, 2129 down-regulated genes were detected in homozygous (Smc3fl/fl; Nkx2.5-Cre) embryonic hearts (Supplementary Table 4) and 2173 downregulated genes were detected in heterozygous (Smc3fl/+; Nkx2.5-Cre) embryonic hearts (Supplementary Table 4). We observed a notable reduction in the expression of critical CM developmental genes, including Tnnt2 and Tnnc1, as well as CM marker genes such as Ankrd1 and Myl1/4/9 (Fig. 3f) in SMC3-cKO embryos. GO analysis revealed the enrichment of these downregulated genes in biological processes involved in CM contraction and differentiation (Supplementary Fig. 3e, f). We also identified different CM subtypes based on previously established markers for embryonic CM heterogeneity36,37, including OFT-like CMs (OFTL-CMs), atrial CMs (A-CMs), CMs of the left ventricle (LV-CMs), and the atrioventricular canal (AVC-CMs) (Fig. 3d, e, Supplementary Fig. 3g, h).
We performed a comprehensive analysis of the differentially expressed genes in OFTL-CMs from Smc3fl/+; Nkx2.5-Cre or Smc3fl/fl; Nkx2.5-Cre embryos to further explore the underlying cause of abnormal OFT development in SMC3-cKO embryos. Our analysis revealed 4577 downregulated genes in Smc3fl/fl; Nkx2.5-Cre embryonic hearts (Supplementary Table 4) (Fig. 3f), while 4628 genes were downregulated in Smc3fl/+; Nkx2.5-Cre embryonic hearts compared with Smc3fl/fl hearts (Supplementary Table 4) (Fig. 3f). GO analysis highlighted enrichment of the downregulated genes in biological processes related to cardiac muscle cell differentiation and heart processes, including OFT morphogenesis (Fig. 3g). Given the similarity in the number and function of differentially expressed genes between Smc3fl/+; Nkx2.5-Cre and Smc3fl/fl; Nkx2.5-Cre embryos at this stage, a plausible hypothesis is that the lethality of homozygous embryos at later stages can be attributed more to the impact of SMC3 on the cell cycle.
As a method to understand whether the underlying differentiation process of OFTL-CMs was affected in each sample, we utilized the R package monocle to construct cell development trajectories. The OFTL-CMs in Smc3fl/fl embryonic hearts were predominantly distributed during early and late differentiation, whereas in SMC3-cKO embryonic hearts, more OFTL-CMs were detected in the transition state (Fig. 3h). These results indicate that SMC3 plays a key role in the development and differentiation of CMs, particularly CMs located in the OFT.
Due to the heterogeneity of tissues and the indispensable role of SMC3 in mitosis, we performed bulk RNA-seq to verify the function of SMC3 in cardiac development using the Smc3 KD mouse cardiac muscle cell line HL-1 (Supplementary Fig. 4a, b). A total of 730 significantly downregulated genes were identified in the KD group compared to the normal control (NC) group (Supplementary Table 5). GO enrichment analysis and gene set enrichment analysis (GSEA) revealed that these genes were enriched in biological processes involved in cardiac morphogenesis and cardiac muscle tissue development (Supplementary Fig. 4c, d). Among the downregulated genes were several established cardiac regulatory genes, including Tbx1, Notch2, Tbx18, Ankrd1, Agt, Mef2a, and Ets2 (Supplementary Fig. 4e). These results are consistent with the findings from the snRNA-seq data.
SMC3 regulates multiple heart development-related genes by controlling the activity of superenhancers
SEs are known to be highly tissue-specific and are often associated with pivotal regulatory genes involved in cell fate determination, embryonic development, and tumor growth38,39,40. Cohesin, with SMC3 as a core component, regulates gene expression by mediating chromatin organization and enhancer activity41. We speculated that SMC3 is involved in heart development by regulating heart-specific SEs. We performed ChIP-seq targeting histone H3 lysine 27 acetylation (H3K27ac) in HL-1 cells to characterize the SE landscape of the heart. Using the ROSE algorithm5, we identified 1544 SEs and 308 SE-associated genes (Fig. 4a) (Supplementary Table 6). Some of these SE-associated genes, including Ets2, Myh9, Id1, Notch2, and Ankrd1, are recognized as master regulatory genes of cardiogenesis in mice (Fig. 4a). GO analysis revealed significant enrichment of these SE-associated genes in biological processes related to anatomical structure development and cell differentiation (Fig. 4b).
a Hockey-stick plots illustrating superenhancers (SEs) based on normalized H3K27ac signals and their ranks in the mouse cardiomyocyte cell line HL-1 using the rank ordering of superenhancers (ROSE) algorithm. Selected SE-associated genes are marked. b Bar plots displaying representative GO terms enriched in 308 SE-associated genes. c Comparison of 4628 downregulated genes in OFTL-CMs from Smc3fl/+; Nkx2.5-Cre embryos, 730 downregulated genes in Smc3-KD HL-1 cells, and 308 SE-associated genes. Four overlapping genes, namely, Ankrd1, Gadd45g, Ets2, and Notch2, were identified. d Violin plots illustrating reduced expression of the four overlapping genes in OFTL-CMs from Smc3fl/+; Nkx2.5-Cre embryos compared to Smc3fl/fl embryos. e Bar graphs showing the expression of the four overlapping genes in the KD group compared to the NC group. f qPCR analysis validated the SMC3 knockdown efficiency in AC16 cells. g Western blot (WB) assays confirmed that SMC3 regulated the protein levels of ETS2, NOTCH2, ANKRD1 and GADD45G in both AC16 (left panel) and HL-1 (right panel) cells. h In situ Hi-C chromatin interaction heatmaps of LV cardiac muscle cells obtained from adult WT mice at the Notch2 locus and SE. H3K4me3 and H3K27ac ChIP-seq tracks of the NC and KD groups at the Notch2 locus and SE. All error bars are the means ± standard deviations. *P < 0.05; **P < 0.01; ***P < 0.001; and ****P < 0.0001. FPKM, fragments per kilobase of exon model per million mapped fragments.
Next, we examined the expression of these SE-associated genes in OFTL-CMs from Smc3fl/+; Nkx2.5-Cre embryos, as well as in Smc3 KD HL-1 cells. Among the 308 SE-associated genes, four genes, Ankrd1, Gadd45g, Ets2, and Notch2, were found to be significantly downregulated under both in vivo and in vitro conditions of Smc3 disruption (Fig. 4c–e). Additionally, we observed that SMC3 downregulation led to reduced expression of these proteins in SMC3 KD HL-1 and AC16 cells (Fig. 4f–g). Importantly, these genes are known to play roles in heart development. For instance, Notch2, a receptor involved in canonical Notch signaling, is crucial for proper OFT development42. Hi-C data from LV cardiac muscle cells in adult mice (GSE96692) also revealed substantial interactions between Notch2 and its SE in a topologically associating domain (TAD) according to our reanalysis (Fig. 4h). ETS2 is a transcription factor belonging to the ETS family known for its critical roles in regulating networks that govern the development of the heart, especially the coronary arteries and myocardial tissue43,44. However, the mechanisms underlying the regulation of its expression remain elusive. These findings suggest that SMC3 plays a critical role in controlling heart development-related genes by regulating the activity of SEs.
A super-enhancer regulates Ets2 expression in the heart
Demonstrating that the Ets2 super-enhancer (Ets2-SE) indeed promotes Ets2 expression during cardiac development is crucial to further confirm the role of SMC3 in regulating Ets2 expression. First, the spatiotemporal expression pattern of Ets2 in mouse hearts from E9.5 to P7 was determined using an RNA scope assay. We found that Ets2 was ubiquitously expressed in the heart (Fig. 5a). A temporal analysis of wild-type mice indicated that the Ets2 mRNA was expressed at its highest level at E9.5 and remained at a relatively low level until a second expression peak at P0 (Fig. 5b), which closely resembles the spatial and temporal specificity of Smc3 expression in the heart during mouse embryonic development. We also observed a region with strong H3K27ac signals at the Ets2-SE in the ChIP-seq data from HL-1 cells (Fig. 5c). This region largely overlaps with the position of the identified Ets2-SE in thymocytes6. Further analysis of ChIP-seq data from human heart tissue (GSM409307, GSM706848, GSM906404, GSM1059445, GSM663427, GSM906396, GSM1013124, GSM432392, GSM772735, and GSM906406) confirmed the presence of an enhancer cluster at the conserved region in the genome (Supplementary Fig. 5a), suggesting that the ETS2-SE is conserved in humans and mice. Furthermore, a robust interaction between this SE and the ETS2 promoter was observed in RV tissues (GSE87112) (Fig. 5d), indicating the regulatory role of ETS2-SE in regulating ETS2 expression. Our existing Ets2-SE-deficient mice, described in our previous publication6, were analyzed to investigate the impact of Ets2-SE on heart development and Ets2 expression. In these mice, the deletion of a 166 kb genomic region (chr16:95745432-95912361, mm10) containing Ets2-SE led to a specific approximately 5-fold reduction in ETS2 expression in the heart, as demonstrated by qPCR and WB assays (Fig. 5e, f). Interestingly, despite its proximity to the Ets2-SE and location in the same TAD, the expression of Psmg1 was not affected by Ets2-SE deletion in the cardiac tissues of mouse embryos at E10.5 (Supplementary Fig. 5b). These results indicate that Ets2-SE specifically regulates the high expression of Ets2 in the heart.
a Localization of Ets2 in the hearts of WT mice at E10.5 (top panel) and P0 (bottom panel). RNA fluorescence in situ hybridization images were obtained using the RNA scope platform with probes targeting Ets2. Image scale bars, 0.2 mm. b Temporal expression pattern of Ets2 in mouse hearts from E9.5 to P7 analyzed using qPCR. c Normalized H3K27ac ChIP-seq profiles at both the Ets2 locus and SE in HL-1 cells. d In situ Hi-C chromatin interaction heatmap of human RV at the ETS2 locus and SE. e, f qPCR and WB results illustrating that Ets2-SE deletion specifically reduces Ets2 expression in the heart at E10.5. All error bars are the means ± standard deviations. **P < 0.01. TAD topologically associating domain.
SMC3 promotes the transcription of Ets2 by enhancing promoter–enhancer interactions
We detected Smc3 and Ets2 expression in the heart tissues of mouse embryos at E10.5 by performing RNA scope and qPCR assays to explore the regulatory effect of SMC3 on Ets2 expression during cardiac development (Fig. 6a, b). The results indicated a notable reduction in the fluorescence intensity of the Ets2 probe across all regions of heart in the Smc3fl/fl; Nkx2.5-Cre embryos (Fig. 6a), which was consistent with the trend detected using qPCR (Fig. 6b). These results indicate that the positive regulatory effect of SMC3 on Ets2 expression is widespread throughout cardiac tissues.
a RNA fluorescence of mouse embryonic hearts revealed reduced expression of Ets2 and Smc3 in Smc3fl/fl; Nkx2.5-Cre mice compared to Smc3fl/fl mice. Image scale bars, 0.1 mm. Low-power image scale bars, 0.25 mm; high-magnification image scale bars, 50 μm. b qPCR results for the Ets2 mRNA extracted from the embryonic hearts of Smc3fl/fl, Smc3fl/+; Nkx2.5-Cre, and Smc3fl/fl; Nkx2.5-Cre mice at E10.5. qPCR analysis of the impact of SMC3 knockdown on ETS2 expression in HL-1 (c) and AC16 cells (d). e In situ Hi-C chromatin interaction heatmaps of LV cardiac muscle cells obtained from adult WT mice at the Ets2 locus and SE. SMC3, RAD21, CTCF, H3K4me3, and H3K27ac ChIP-seq tracks of mouse embryonic stem cells and mouse embryonic hearts for the Ets2 locus and SE. f ChIP‒qPCR revealed reduced occupancy of SMC3 on the Ets2 promoter in KD HL-1 cells compared to NC HL-1 cells. g ChIP-seq and 3C-HTGTS-seq at the Ets2 locus and SE. The 3C-HTGTS bait is located at the Ets2 promoter. h qPCR and WB analyses validated the efficiency of Ets2 overexpression (OE) in Smc3-KD HL-1 cells. i The expression of the genes downregulated in Fig. 3f in four groups of HL-1 cells, namely, NC, KD, overexpression of the blank vector in Smc3-KD HL-1 cells (KD + Ets2-NC), and overexpression of Ets2 in Smc3-KD HL-1 cells (KD + Ets2-OE). The reduced expression of these genes in Smc3-KD cells was reversed by Ets2 overexpression. All error bars are the means ± standard deviations. *p < 0.05, **p < 0.01, ***P < 0.001, and ****P < 0.0001.
We used Smc3 KD HL-1 and AC16 cells to elucidate the mechanism by which SMC3 regulates Ets2 expression (Supplementary Fig. 4a, b, Fig. 4f, g). The downregulation of SMC3 led to a reduction in ETS2 expression at both the mRNA and protein levels in mouse and human cell lines (Figs. 4f, g and 6c, d), consistent with observations in Smc3fl/fl; Nkx2.5-Cre mouse heart tissues. Many studies have provided substantial evidence supporting the concept that the loop extrusion model serves as the foundation for cohesin-mediated control of gene expression45,46,47. According to this model, one or a pair of physically tethered cohesin rings slide in opposite directions along chromatin until they encounter hindrance from CTCF proteins bound to inward-oriented motifs. Then, we analyzed the ChIP-seq data for SMC3 (GSM2111724), RAD21 (GSM591469), CTCF (GSM851286), H3K4me3 (GSM1944082), and H3K27ac (GSM2577068 and GSM851290) in mouse embryonic stem cells and heart tissues. High occupancy of SMC3, RAD21, and CTCF was detected on the promoter region of Ets2 (Fig. 6e), suggesting that cohesin directly regulates Ets2 expression. ChIP‒qPCR confirmed the binding of SMC3 to the Ets2 promoter in HL-1 cells, and SMC3 occupancy was reduced in Smc3-KD HL-1 cells (Fig. 6f). Additionally, we performed ChIP-seq of H3K27ac and H3K4me3 in HL-1 cells to detect the effect of Smc3 downregulation on the SE. We observed reduced H3K27 acetylation in the SE region and decreased H3K4 trimethylation at the Ets2 promoter in Smc3 KD cells (Fig. 6g).
The role of cohesin in facilitating interactions between promoter regions and distant enhancers, thereby activating gene transcription, is widely recognized5,41,48. We conducted an analysis using Hi-C data obtained from LV cardiac muscle cells of adult mice (GSE96692) to explore the involvement of SMC3 in the enhancer–promoter looping of the Ets2 gene. This analysis revealed strong interactions between the Ets2 gene and the SE region. We further explored this relationship in Smc3 KD and NC HL-1 cells utilizing a newly developed technique called 3C-high-throughput genome-wide translocation sequencing (3C-HTGTS-seq) with bait at the Ets2 promoter. 3C-HTGTS is a technique similar to the circularized chromatin conformation capture (4 C) assay, enabling the sensitive detection of regions interacting with the Ets2 promoter6. We observed that the Ets2 promoter strongly interacted with the SE region, and these interactions were significantly reduced in Smc3 KD cells (Fig. 6g). These findings suggest that SMC3 promotes Ets2 transcription by enhancing SE activity and mediating interactions between the Ets2 promoter and the SE region.
We overexpressed Ets2 in Smc3-KD HL-1 cells (Fig. 6h) and assessed the expression levels of the downregulated genes (Fig. 3f) in CMs and OFTL-CMs from SMC3-cKO embryos to verify whether SMC3 regulates the expression of cardiac development-related genes through ETS2. We found that Ets2 overexpression significantly reduced the expression of the majority of genes in Smc3-KD cells (Fig. 6i, Supplementary Fig. 6). In conclusion, these results indicate that SMC3 regulates Ets2 expression by establishing chromatin loops between Ets2-SE and its promoter, thereby influencing the regulation of cardiac development-related genes.
The Ets2 super-enhancer is involved in the regulation of heart development
Previous studies have shown that Ets2 deficiency in mice results in retarded and smaller embryos, abnormal coronary artery development, and thin ventricular walls44,49. We examined the cardiac phenotype of Ets2-SE-deleted mice during embryonic development to validate whether Ets2-SE plays a comparable role in heart development. The analysis of birth rates and gross external morphology in Ets2-SE-deleted mice did not reveal consistent effects on embryo morphology (Supplementary Fig. 7) or embryo mortality (Fig. 7a). However, a detailed examination of Ets2-SE−/− and Ets2-SE+/− embryos or neonates at E18.5 and P0 revealed that membranous VSD occurred in one of ten Ets2-SE−/− embryos at E18.5 and one of seven at P0 (Fig. 7b). At E14.5, a critical period of septal closure, malalignment between the artery and the ventricle was also observed in a portion of (2/15) Ets2-SE−/− mice (Fig. 7c). Specifically, both the aorta and pulmonary artery originated from the RV in Ets2-SE−/− embryo hearts, while the pulmonary artery originated from the RV and the aorta originated from the LV in normal hearts (Fig. 7c). However, this severe cardiac malformation did not result in embryonic lethality, possibly because the deletion of Ets2-SE delayed, rather than completely blocked, OFT remodeling at E14.5. This hypothesis also explains the low penetrance in Ets2-SE−/− neonatal mice. Furthermore, we found a significant patent ventricular septum in the majority (9/15) of Ets2-SE−/− embryos at E14.5. At E12.5, we noted that a significantly smaller RV occurred in the Ets2-SE−/− embryos (Fig. 7d). Additionally, a portion (4/15) of the Ets2-SE+/− embryos also exhibited a significantly patent ventricular septum (Fig. 7c). These observations indicate that the deletion of Ets2-SE led to dysplasia of the OFT.
a Genotype distribution of embryos and newborn mice resulting from Ets2-SE+/− and Ets2-SE+/− crosses. b H&E staining of heart sections from WT, Ets2-SE+/−, and Ets2-SE−/− mice at E18.5 and P0. The black arrow indicates a VSD. The number in the lower right corner represents the frequency of the deformity. Image scale bars, 1 mm. c H&E staining of heart sections at E14.5 showing delayed OFT development in Ets2-SE−/− and Ets2-SE+/− mice. In the WT heart, the PA and AO open into the RV and LV, respectively. In 2/15 of the Ets2-SE−/− embryos, both the PA and AO open into the RV. A patent ventricular septum is observed in 9/15 of Ets2-SE−/− mice and 4/15 of Ets2-SE+/− mice. Image scale bars, 0.5 mm. d Compared to those of WT hearts at E12.5, Ets2-SE−/− hearts exhibited smaller RVs. The size of the LV is indicated by a Cambridge blue outline, while the pink outline surrounds the size of the RV. The outline of the PA is indicated by blue lines, while red lines indicate the outline of the AO. WT, n = 6; Ets2-SE+/−, n = 9; Ets2-SE−/−, n = 7. Low-power image scale bars, 5 mm; high-magnification image scale bars, 1 mm. All error bars are the means ± standard deviations. ***P < 0.001. e Volcano plot illustrating differentially expressed genes between Ets2-SE−/− and WT mice at E12.5. Blue, downregulated genes. Red, upregulated genes. f Bar plots displaying representative GO terms enriched in downregulated genes in the hearts of E12.5 Ets2-SE−/− mice. g Gene set enrichment analysis (GSEA) revealed that the downregulated genes were enriched in biological processes, including homeostasis of the number of cells, and the upregulated genes were enriched in the negative regulation of cell development. h Heatmap illustrating scaled expression of genes associated with heart development in WT and Ets2-SE−/− embryonic hearts at E12.5. The color from dark blue to dark red indicates the scaled expression from low to high. FC fold change, ES enrichment score, NES normalized enrichment score, FDR, false discovery rate.
We conducted a transcriptome comparison between Ets2-SE−/− and wild-type mouse embryos at E12.5 using bulk RNA-seq to explore the genes potentially regulated by ETS2 that could contribute to the observed phenotypes. A total of 2290 upregulated and 2204 downregulated genes were detected in the E12.5 hearts of Ets2-SE−/− mice compared to those of wild-type mice (Fig. 7e) (Supplementary Table 7). Downregulation of Ets2 resulted in the dysregulation of genes involved in heart morphogenesis and cell proliferation (Fig. 7f, g). In addition, ETS2 regulated the expression of critical genes involved in heart development, including Wnt11 and Tbx18 (Fig. 7h). WNT11 is essential for second heart field progenitor development and cardiac chamber development50,51, while TBX18 plays a role in regulating the development of the epicardium and coronary vessels52. In conclusion, these observations reveal that Ets2-SE plays a role in heart development, especially in the OFT, by controlling Ets2 expression within the heart.
Discussion
In this study, we revealed a novel and significant role for SMC3 in establishing promoter–enhancer loops to regulate the expression of SE-associated genes crucial for OFT development. We found that a majority of CdLS patients with SMC3 mutations suffered from CHD, with most exhibiting aorta and pulmonary artery malformations. The formation of the aorta and pulmonary artery results from the correct rotation and separation of the OFT. The various OFT defects observed in SMC3-cKO mice provide a pathological explanation for CHD in CdLS patients. Based on our findings, we speculated that SMC3 might promote the activity of heart-specific SEs to control the expression of master genes essential for OFT development, such as Notch2. NOTCH2 is required for proper OFT development53. Jag1 and Notch2 double heterozygous mice exhibit RV hypoplasia, pulmonary stenosis, VSD, and ASD54. This study provides new insight into the mechanism by which the expression of Notch2 is regulated during OFT development. Furthermore, we highlight the increasing significance of noncoding mutations in individuals with CHD and make a valuable contribution to the recognition of the importance of mutations located in SEs that regulate the expression of master genes in heart development.
During heart development, the fates of various cardiac progenitor cells undergo redefinition to ensure the structural formation and functional stability of the OFT. SEs play a significant role in this process, as they promote robust transcriptional activation of genes within cell fate determination regulatory networks. Our study revealed that several genes, including ETS2, ANKRD1, and GADD45G, are regulated by SMC3 and heart-specific SEs and may be involved in OFT development. ETS2, which is essential for controlling the formation of the anterior–posterior pattern, primitive streak, mesoderm from the ectoderm in the murine embryo55 and normal development of the second heart field56, synergizes with MESP1 to reprogram human dermal fibroblasts into cardiac progenitors43. Ets2 deficiency leads to abnormal coronary artery development44. ANKRD1, which is normally localized to cardiomyocyte sarcomeres, is thought to function as a transcriptional cofactor and negative regulator of myocardial gene expression during heart development. Overexpression of Ankrd1 in cardiomyocytes leads to abnormal rotation of the early OFT57. GADD45G, a member of the GADD45 protein family, is involved in p38 mitogen-activated protein kinase-dependent cell death58. Overexpression of Gadd45g in mice induces cardiac muscle cell apoptosis and myocardial fibrosis59, which contribute to the development of several types of CHD60,61. These genes have previously been linked to various aspects of cardiac development and function, and although the specific roles of these genes in OFT development are unclear, our study offers a potential link that can help shed light on their involvement in this crucial process.
ETS2, located on chromosome 21, has been extensively studied in various biological processes, including cell proliferation and apoptosis62,63. As previously described, ETS2 is highly expressed in patients with heart failure64, and a reduction in ETS2 expression is associated with improved contractile dysfunction in patients with cardiac hypertrophy65. In the present study, we elucidated that high Ets2 expression in the heart is regulated by Ets2-SE and SMC3, providing potential insights for precision therapy of heart failure patients.
The function of isolated SMC3 mainly depends on cohesin. The dual roles of cohesin in heart development should be considered. While cohesin mediates sister chromosome cohesion, which is essential for cell survival and proliferation, its role in regulating gene expression through chromatin organization is equally important23. The phenotype observed in Smc3fl/fl; Nkx2.5-Cre mice may be mainly caused by defects in cohesin function, which leads to cell death after SMC3 deletion. However, the defects in cardiac development experienced by CdLS patients may not result from impaired cell cycling but from abnormal regulation of gene expression because they have a normal SMC3 allele. During early heart development, the same heart defects in Smc3fl/fl; Nkx2.5-Cre embryos were observed in a small proportion of Smc3fl/+; Nkx2.5-Cre embryos. We hypothesize that SMC3 knockout may affect cell proliferation and lead to dysregulated gene expression, which contributes to the observed heart defects. Differentiating between these two functions of cohesin by phenotypic observation alone may not be possible. CdLS is a genetically heterogeneous pleiotropic disorder with multiple structural and functional deficits, including CHD22. A genotype‒phenotype correlation analysis of CdLS revealed that SMC3 missense mutations and in-frame deletions may lead to a milder phenotype, whereas NIPBL truncating pathogenetic mutations result in a more severe phenotype14. Our identification of rare and pathogenic SMC3 variants in individuals with isolated CHD suggests that mild functional variants of genes encoding cohesin members can cause CHD without a classic CdLS phenotype, expanding our understanding of cohesinopathy-related disorders and variants associated with isolated CHD.
In conclusion, our study has shed light on the regulatory role of SMC3 in cardiac development and provides compelling evidence that SMC3 facilitates interactions between promoters and SE loops, thereby regulating the expression of SE-associated genes essential for heart development. These findings contribute to the understanding of the molecular basis of cardiac development and provide valuable insights into the pathogenesis of CHD. Furthermore, our research underscores the elevated risk of CHD in individuals with SMC3 mutations and strongly supports the inclusion of SMC3 mutation testing within the panel of genes considered for CHD screening.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request. All the raw high-throughput sequence data are available in the GEO (GSE242974). All research materials listed in the Methods section are also included in the major resources table (Supplementary Table 1). The ChIP-seq data were downloaded from the GEO database and are displayed in Supplementary Table 1.
References
Boyd, R., McMullen, H., Beqaj, H. & Kalfa, D. Environmental exposures and congenital heart disease. Pediatrics 149, e2021052151 (2022).
Galdos, F. X. et al. Cardiac regeneration: lessons from development. Circ. Res 120, 941–959 (2017).
Rammah, M., Théveniau-Ruissy, M., Sturny, R., Rochais, F. & Kelly, R. G. PPARγ and NOTCH regulate regional identity in the murine cardiac outflow tract. Circ. Res 131, 842–858 (2022).
Chapman, G. et al. Functional genomics and gene-environment interaction highlight the complexity of congenital heart disease caused by Notch pathway variants. Hum. Mol. Genet 29, 566–579 (2020).
Whyte, W. A. et al. Master transcription factors and mediator establish super-enhancers at key cell identity genes. Cell 153, 307–319 (2013).
Feng, D. et al. Chromatin organizer SATB1 controls the cell identity of CD4(+) CD8(+) double-positive thymocytes by regulating the activity of super-enhancers. Nat. Commun. 13, 5554 (2022).
Borck, P. C., Guo, L. W. & Plutzky, J. BET epigenetic reader proteins in cardiovascular transcriptional programs. Circ. Res 126, 1190–1208 (2020).
Dey, A. et al. BRD4 directs hematopoietic stem cell development and modulates macrophage inflammatory responses. Embo j. 38, e100293 (2019).
Man, J. C. K. et al. Genetic dissection of a super enhancer controlling the Nppa-Nppb cluster in the heart. Circ. Res 128, 115–129 (2021).
Hnisz, D. et al. Super-enhancers in the control of cell identity and disease. Cell 155, 934–947 (2013).
Vos, E. S. M. et al. Interplay between CTCF boundaries and a super enhancer controls cohesin extrusion trajectories and gene expression. Mol. Cell 81, 3082–3095.e3086 (2021).
Eagen, K. P. Principles of chromosome architecture revealed by Hi-C. Trends Biochem Sci. 43, 469–478 (2018).
Rao, S. S. P. et al. Cohesin loss eliminates all loop domains. Cell 171, 305–320.e324 (2017).
Sarogni, P., Pallotta, M. M. & Musio, A. Cornelia de Lange syndrome: from molecular diagnosis to therapeutic approach. J. Med Genet 57, 289–295 (2020).
Panarotto, M., Davidson, I. F., Litos, G., Schleiffer, A. & Peters, J. M. Cornelia de Lange syndrome mutations in NIPBL can impair cohesin-mediated DNA loop extrusion. Proc. Natl Acad. Sci. USA 119, e2201029119 (2022).
Singh, V. P. & Gerton, J. L. Cohesin and human disease: lessons from mouse models. Curr. Opin. Cell Biol. 37, 9–17 (2015).
Chatfield, K. C. et al. Congenital heart disease in Cornelia de Lange syndrome: phenotype and genotype analysis. Am. J. Med Genet A 158a, 2499–2505 (2012).
Santos, R. et al. Conditional Creation and Rescue of Nipbl-Deficiency in Mice Reveals Multiple Determinants of Risk for Congenital Heart Defects. PLoS Biol. 14, e2000197 (2016).
Muto, A., Calof, A. L., Lander, A. D. & Schilling, T. F. Multifactorial origins of heart and gut defects in nipbl-deficient zebrafish, a model of Cornelia de Lange Syndrome. PLoS Biol. 9, e1001181 (2011).
Hao, L. et al. WDR62 variants contribute to congenital heart disease by inhibiting cardiomyocyte proliferation. Clin. Transl. Med 12, e941 (2022).
Ayerza Casas, A. et al. Cornelia de Lange syndrome: congenital heart disease in 149 patients. Med Clin. (Barc.) 149, 300–302 (2017).
Kline, A. D. et al. Diagnosis and management of Cornelia de Lange syndrome: first international consensus statement. Nat. Rev. Genet 19, 649–666 (2018).
Ishiguro, K. I. The cohesin complex in mammalian meiosis. Genes Cells 24, 6–30 (2019).
Watrin, E., Kaiser, F. J. & Wendt, K. S. Gene regulation and chromatin organization: relevance of cohesin mutations to human disease. Curr. Opin. Genet Dev. 37, 59–66 (2016).
Pistocchi, A. et al. Cornelia de Lange Syndrome: NIPBL haploinsufficiency downregulates canonical Wnt pathway in zebrafish embryos and patients fibroblasts. Cell Death Dis. 4, e866 (2013).
Gil-Rodríguez, M. C. et al. De novo heterozygous mutations in SMC3 cause a range of Cornelia de Lange syndrome-overlapping phenotypes. Hum. Mutat. 36, 454–462 (2015).
Mannini, L., Cucco, F., Quarantotti, V., Krantz, I. D. & Musio, A. Mutation spectrum and genotype-phenotype correlation in Cornelia de Lange syndrome. Hum. Mutat. 34, 1589–1596 (2013).
Ansari, M. et al. Genetic heterogeneity in Cornelia de Lange syndrome (CdLS) and CdLS-like phenotypes with observed and predicted levels of mosaicism. J. Med Genet 51, 659–668 (2014).
Yuan, B. et al. Global transcriptional disturbances underlie Cornelia de Lange syndrome and related phenotypes. J. Clin. Invest 125, 636–651 (2015).
Liu, C. et al. Analysis of clinical and genetic characteristics in 10 Chinese individuals with Cornelia de Lange syndrome and literature review. Mol. Genet Genom. Med 8, e1471 (2020).
Li, R. et al. A Chinese Case of Cornelia de Lange Syndrome Caused by a Pathogenic Variant in SMC3 and a Literature Review. Front Endocrinol. 12, 604500 (2021).
Ma, J. et al. Hypermethylation-mediated down-regulation of lncRNA TBX5-AS1:2 in Tetralogy of Fallot inhibits cell proliferation by reducing TBX5 expression. J. Cell Mol. Med. 24, 6472–6484 (2020).
Hu, P. et al. Single-nucleus transcriptomic survey of cell diversity and functional maturation in postnatal mammalian hearts. Genes Dev. 32, 1344–1357 (2018).
Wolfien, M. et al. Single-Nucleus Sequencing of an Entire Mammalian Heart: Cell Type Composition and Velocity. Cells 9, 318 (2020).
Liu, X. et al. Single-Cell RNA-Seq of the Developing Cardiac Outflow Tract Reveals Convergent Development of the Vascular Smooth Muscle Cells. Cell Rep. 28, 1346–1361.e1344 (2019).
Hill, M. C. et al. A cellular atlas of Pitx2-dependent cardiac development. Development 146, dev180398 (2019).
Feng, W. et al. Single-cell transcriptomic analysis identifies murine heart molecular features at embryonic and neonatal stages. Nat. Commun. 13, 7960 (2022).
Kitagawa, Y. et al. Guidance of regulatory T cell development by Satb1-dependent super-enhancer establishment. Nat. Immunol. 18, 173–183 (2017).
Gartlgruber, M. et al. Super enhancers define regulatory subtypes and cell identity in neuroblastoma. Nat. Cancer 2, 114–128 (2021).
Sengupta, S. & George, R. E. Super-Enhancer-Driven Transcriptional Dependencies in Cancer. Trends Cancer 3, 269–281 (2017).
Nishiyama, T. Cohesion and cohesin-dependent chromatin organization. Curr. Opin. Cell Biol. 58, 8–14 (2019).
Niessen, K. & Karsan, A. Notch signaling in cardiac development. Circ. Res. 102, 1169–1181 (2008).
Islas, J. F. et al. Transcription factors ETS2 and MESP1 transdifferentiate human dermal fibroblasts into cardiac progenitors. Proc. Natl Acad. Sci. USA 109, 13016–13021 (2012).
Lie-Venema, H. et al. Ets-1 and Ets-2 transcription factors are essential for normal coronary and myocardial development in chicken embryos. Circ. Res. 92, 749–756 (2003).
Kagey, M. H. et al. Mediator and cohesin connect gene expression and chromatin architecture. Nature 467, 430–435 (2010).
Matthews, B. J. & Waxman, D. J. Computational prediction of CTCF/cohesin-based intra-TAD loops that insulate chromatin contacts and gene expression in mouse liver. Elife 7, e34077 (2018).
Brown, J. M. et al. A tissue-specific self-interacting chromatin domain forms independently of enhancer-promoter interactions. Nat. Commun. 9, 3849 (2018).
Cheng, N. et al. STAG2 promotes the myelination transcriptional program in oligodendrocytes. Elife 11, e77848 (2022).
Yamamoto, H. et al. Defective trophoblast function in mice with a targeted mutation of Ets2. Genes Dev. 12, 1315–1326 (1998).
Cohen, E. D., Miller, M. F., Wang, Z., Moon, R. T. & Morrisey, E. E. Wnt5a and Wnt11 are essential for second heart field progenitor development. Development 139, 1931–1940 (2012).
Touma, M. et al. Wnt11 regulates cardiac chamber development and disease during perinatal maturation. JCI Insight 2, e94904 (2017).
Wu, S. P., Dong, X. R., Regan, J. N., Su, C. & Majesky, M. W. Tbx18 regulates development of the epicardium and coronary vessels. Dev. Biol. 383, 307–320 (2013).
Varadkar, P. et al. Notch2 is required for the proliferation of cardiac neural crest-derived smooth muscle cells. Dev. Dyn. 237, 1144–1152 (2008).
McCright, B., Lozier, J. & Gridley, T. A mouse model of Alagille syndrome: Notch2 as a genetic modifier of Jag1 haploinsufficiency. Development 129, 1075–1082 (2002).
Ristevski, S., Tam, P. P., Hertzog, P. J. & Kola, I. Ets2 is expressed during morphogenesis of the somite and limb in the mouse embryo. Mech. Dev. 116, 165–168 (2002).
Bosman, A. et al. Perturbations of heart development and function in cardiomyocytes from human embryonic stem cells with trisomy 21. Stem Cells 33, 1434–1446 (2015).
Piroddi, N. et al. Myocardial overexpression of ANKRD1 causes sinus venosus defects and progressive diastolic dysfunction. Cardiovasc Res. 116, 1458–1472 (2020).
Iwahana, T. et al. Novel myocardial markers GADD45G and NDUFS5 identified by RNA-sequencing predicts left ventricular reverse remodeling in advanced non-ischemic heart failure: a retrospective cohort study. BMC Cardiovasc Disord. 20, 116 (2020).
Lucas, A. et al. Gadd45γ regulates cardiomyocyte death and post-myocardial infarction left ventricular remodelling. Cardiovasc Res. 108, 254–267 (2015).
Rathod, R. H., Powell, A. J. & Geva, T. Myocardial Fibrosis in Congenital Heart Disease. Circ. J. 80, 1300–1307 (2016).
Hu, C., Huang, S., Wu, F. & Ding, H. MicroRNA-219-5p participates in cyanotic congenital heart disease progression by regulating cardiomyocyte apoptosis. Exp. Ther. Med. 21, 36 (2021).
Dwyer, J., Li, H., Xu, D. & Liu, J. P. Transcriptional regulation of telomerase activity: roles of the the Ets transcription factor family. Ann. N. Y Acad. Sci. 1114, 36–47 (2007).
Haridhasapavalan, K. K., Sundaravadivelu, P. K. & Thummer, R. P. Codon Optimization, Cloning, Expression, Purification, and Secondary Structure Determination of Human ETS2 Transcription Factor. Mol. Biotechnol. 62, 485–494 (2020).
Li, J., Su, H., Zhu, Y., Cao, Y. & Ma, X. ETS2 and microRNA-155 regulate the pathogenesis of heart failure through targeting and regulating GPR18 expression. Exp. Ther. Med. 19, 3469–3478 (2020).
Luo, Y. et al. Cooperative Binding of ETS2 and NFAT Links Erk1/2 and Calcineurin Signaling in the Pathogenesis of Cardiac Hypertrophy. Circulation 144, 34–51 (2021).
Acknowledgements
We thank the patients who participated in this study. We thank Professor Yujiang Geno Shi from Fudan University for providing excellent guidance on writing the article and Professor Jikui Wang from Xinxiang Medical University for providing substantial support in diagnosing the cardiac phenotypes of mice. We also thank our peers for providing the WES data from patients with isolated CHD. This work was supported by the National Key Research and Development Program of China (2021YFC2701001 to D.M.), the National Natural Science Foundation of China (No. 82271889 to J.M., 31970836 to B.H., and 32170885 to B.H.), and the National Science Foundation for Young Scientists (No. 81801501 to J.M.).
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Zhang, B., Zhu, Y., Zhang, Z. et al. SMC3 contributes to heart development by regulating super-enhancer associated genes. Exp Mol Med 56, 1826–1842 (2024). https://doi.org/10.1038/s12276-024-01293-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s12276-024-01293-0
