RUNX1 promotes cardiac fibrosis and exacerbates myocardial infarction by transcriptionally activating PTN expression

Abstract

Background

Myocardial fibrosis is a major pathological change following myocardial infarction (MI). Cardiac fibroblast activation is a key driver in the development of myocardial fibrosis. Runt-related transcription factor 1 (RUNX1) is a transcription factor that promotes tissue fibrosis, but the specific downstream effector through which RUNX1 acts in cardiac fibroblasts after MI has not been defined. This study aims to explore the specific mechanisms by which RUNX1 contributes to cardiac fibrosis following MI.

Methods

The Gene Expression Omnibus (GEO) database was analyzed to explore gene expression changes following MI. Based on its significant upregulation, RUNX1 was selected for investigation of its role in post-MI myocardial fibrosis.

Results

The GEO database analysis revealed that RUNX1 was significantly upregulated in mice with MI. Compared with the control group, the MI group exhibited marked cardiac dysfunction accompanied by significantly increased RUNX1 expression. Fibroblast-specific RUNX1 knockdown alleviated post-MI cardiac fibrosis, while RUNX1 silencing suppressed fibroblast activation and secretory function. Mechanistically, as a transcription factor, RUNX1 binds to the pleiotrophin (PTN) promoter and transcriptionally upregulates PTN, thereby exerting pro-fibrotic effects.

Conclusions

This study identifies PTN as a transcriptional target of RUNX1 in cardiac fibroblasts and shows that fibroblast-targeted RUNX1 knockdown attenuates post-MI fibrosis in a PTN-dependent manner. The RUNX1-PTN axis represents a candidate node for fibroblast-directed intervention in post-MI remodeling.

Highlight box.

Key findings

• Runt-related transcription factor 1 (RUNX1) is significantly upregulated following MI and is associated with cardiac dysfunction.

What is known, and what is new?

• Fibroblast-specific knockdown of RUNX1 attenuates cardiac fibrosis and improves cardiac function following myocardial infarction (MI). RUNX1 promotes cardiac fibroblast activation, migration, and extracellular matrix production. Thus, targeting RUNX1 may represent a potential therapeutic strategy for post-MI cardiac remodeling.

• This study provides a new mechanism that RUNX1 directly binds to the pleiotrophin (PTN) promoter and transcriptionally activates PTN expression. PTN mediates the pro-fibrotic effects of RUNX1, forming a RUNX1/PTN regulatory axis in myocardial fibrosis.

What is the implication, and what should change now?

• RUNX1 promotes post-MI cardiac fibrosis by transcriptionally activating PTN, driving myofibroblast transformation and exacerbating cardiac function. Targeting the RUNX1/PTN axis may represent a novel therapeutic strategy for ameliorating post-MI myocardial fibrosis and heart failure.

Introduction

Myocardial infarction (MI) is a common and life-threatening cardiovascular disease characterized by acute coronary artery occlusion, resulting in insufficient blood supply to the myocardium. This leads to cardiomyocyte necrosis, reparative scar formation, and subsequent cardiac dysfunction, which may ultimately progress to heart failure (1). Early interventions such as percutaneous coronary intervention and coronary artery bypass grafting are widely used clinically; however, they cannot fully reverse the process of post-MI myocardial fibrosis (2). Therefore, investigating the pathogenesis of post-MI myocardial fibrosis is of great clinical significance.

Cardiac fibroblasts play a crucial role in post-MI cardiac remodeling. Following MI, resident cardiac fibroblasts are stimulated by various signaling factors, including angiotensin II, transforming growth factor-β (TGF-β), and endothelin-1, resulting in dynamic phenotypic changes and mature myofibroblast differentiation (3,4). These cells exhibit strong proliferative, contractile, migratory, and collagen-synthesizing capabilities, thereby exerting dual effects in MI. In the early stage, activated fibroblasts migrate to the infarcted region to form scars, helping to preserve structural integrity and prevent acute ventricular rupture. However, while initial reparative fibrosis is beneficial, sustained fibroblast activation results in excessive extracellular matrix (ECM) accumulation, increasing ventricular stiffness, impairing both systolic and diastolic function, and ultimately resulting in heart failure. Previous studies have explored the mechanisms underlying fibroblast activation and proliferation in fibrotic diseases of various other organs (5,6). Thus, targeting anti-fibrotic pathways holds significant potential for clinical translation in the treatment of post-MI cardiac remodeling and heart failure.

Runt-related transcription factor 1 (RUNX1) is a transcription factor first identified in 1991 through cloning of the t(8; 21) chromosomal translocation associated with M2 acute myeloid leukemia (7). Subsequent studies demonstrated that RUNX1 plays essential roles in organismal growth and development. To date, research on RUNX1 has largely focused on hematopoiesis and cancer (8,9); however, recent studies have demonstrated its critical involvement in fibrotic diseases across multiple organs. In renal fibrosis, RUNX1 exacerbates disease progression by enhancing activation of the phosphatidylinositol 3-kinase (PI3K) subunit p110δ, thereby promoting TGF-β-induced partial epithelial-mesenchymal transition (10). In hepatic fibrosis, RUNX1 overexpression leads to more severe fibrotic lesions compared to controls, primarily through activation of the TGF-β/SMAD signaling pathway (11). In pulmonary fibrosis, inhibiting RUNX1 can eliminate the expression of classic myofibroblast markers induced by TFG-β (12). In the heart, RUNX1 is upregulated in various pathological conditions, including diabetic cardiomyopathy, pressure overload-induced heart disease, and acute MI (AMI) (13). However, the precise mechanism by which RUNX1 contributes to post-MI cardiac fibrosis remains unclear.

Pleiotrophin (PTN) is a secreted heparin-binding growth factor involved in diverse biological processes such as cell growth, differentiation, and tumor progression (14). Studies have shown that PTN is upregulated in various neuroinflammatory disorders, and the PTN/RPTPβ/ζ signaling pathway in the connection between obesity, cognitive decline, and the associated brain alterations plays a key role (15). Multiple studies suggest that PTN plays a key role in neonatal cardiac development and heart failure progression, although direct evidence remains limited (16,17). What has not been established is whether PTN is transcriptionally controlled by RUNX1 in cardiac fibroblasts, and whether such an axis operates after MI.

Therefore, we hypothesized that RUNX1 may participate in post-MI myocardial fibrosis by transcriptionally regulating PTN. By investigating the RUNX1/PTN regulatory axis, we aimed to elucidate its molecular mechanism in cardiac fibroblast activation following MI, thereby providing a new theoretical basis for the treatment of post-MI myocardial fibrosis and related drug development. We present this article in accordance with the ARRIVE and MDAR reporting checklists (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1900/rc).

Methods

Animal models

This study was approved by the Animal Ethics Committee of Nanjing First Hospital affiliated to Nanjing Medical University (No. DWSY25094727), in compliance with institutional guidelines for the care and use of animals. Male C57BL/6 mice (6–8 weeks old) were purchased from Jiangsu Huachuang Xinnuo Pharmaceutical Technology Co., Ltd. (Nanjing, China) to establish the MI model. MI was induced by ligation of the left anterior descending coronary artery. In brief, mice were anesthetized with 2% isoflurane inhalation, and a small incision was made between the 3^rd and 4^th intercostal spaces on the left thorax. The left anterior descending artery was ligated with a 6-0 silk suture approximately 2–3 mm below the tip of the left auricle. Control mice underwent thoracotomy alone under the same anesthetic conditions. All surviving mice were housed under specific pathogen-free conditions and data were collected 4 weeks after ligation. At the endpoint, mice were anesthetized with 2% isoflurane and euthanized by cervical dislocation under deep anesthesia, and hearts were harvested for downstream analyses.

Echocardiography

Echocardiographic examinations were performed at the Experimental Animal Center of Nanjing Medical University to assess left ventricular systolic function. Parameters, including left ventricular internal dimension at end-diastole, left ventricular internal diameter at end-systole, left ventricular ejection fraction (LVEF), and left ventricular fractional shortening (LVFS), were measured over at least three consecutive cardiac cycles.

Hematoxylin and eosin (H&E) and Masson’s trichrome staining

After perfusion with normal saline, heart tissues were fixed in 4% paraformaldehyde for 24 hours, embedded in paraffin, and sectioned into 5 µm slices. H&E staining was used to visualize myocardial structure (with nuclei-stained basophilic navy blue, and cytoplasm-stained eosinophilic pink), while Masson’s trichrome staining was performed to assess fibrosis (with collagen-stained ice blue, cytoplasm-stained salmon pink, and chromatin-stained dark purple).

Immunofluorescence and immunohistochemistry

Paraffin sections of heart tissues (fixed in 10% neutral formalin) were dewaxed, rehydrated, subjected to antigen retrieval, and blocked. Primary antibodies [RUNX1: 25315-1-AP, Proteintech, Wuhan, China; α-smooth muscle actin (⍺-SMA): 80008-1-RR, Proteintech; vimentin: GB111308; Servicebio, Wuhan, China; green fluorescent protein (GFP): GB11602; Servicebio] were diluted to 1:200 and incubated at 4 ℃ overnight. For immunofluorescence, sections were incubated with fluorescent secondary antibodies at room temperature for 1 hour, mounted with 4',6-diamidino-2-phenylindole (DAPI)-containing medium, and observed under a fluorescent microscope. For immunohistochemistry, 3,3'-diaminobenzidine (DAB) staining was performed, and brown precipitates were observed under a light microscope.

Fibroblast isolation and culture

Primary cardiac fibroblasts were isolated from 1–3-day-old neonatal mice using collagenase-trypsin digestion. In brief, hearts were minced into 1 mm³ fragments, incubated with 2 mL of 0.25% trypsin (#KGL2101-100, KeyGEN BioTECH, Nanjing, China) and 2 mL of 0.5 g/L type II collagenase (#17101015, Gibco, Shanghai, China) in sterile centrifuge tubes, at 37 ℃ with gentle agitation for 7 minutes. The supernatant was then collected and transferred into Dulbecco’s modified Eagle medium (DMEM)/F12 medium (KeyGEN BioTECH) supplemented with 10% fetal bovine serum (FBS) (#F103, Vazyme Biotech Co., Ltd., Nanjing, China), and the digestion process was repeated until the tissue fragments became translucent. The collected supernatants were centrifuged at 1,200 rpm for 10 minutes, and the cell pellets were resuspended in 10% FBS-DMEM/F12, plated in culture dishes, and allowed to adhere for 2 hours. After differential adhesion, the supernatant was discarded, and fresh medium was added for subsequent experiments. The in vitro fibrosis model was established by treating cells with 10 ng/mL TGF-β1 (RP00671, ABclonal, Wuhan, China) for 48 h.

qRT-PCR

Total RNA was extracted from fibroblasts using the Super FastPure Cell RNA Isolation Kit (#RC102, Vazyme Biotech Co., Ltd.), reverse-transcribed to complementary DNA with the HisyGo RT Red SuperMix for quantitative polymerase chain reaction (qPCR) (#RT101, Vazyme Biotech Co., Ltd.), and quantitative real-time polymerase chain reaction (qRT-PCR) was performed using ChamQ Blue Universal SYBR qPCR Master Mix (#Q312, Vazyme Biotech Co., Ltd.). Transcript levels were normalized to 18s. Primer sequences were as follows: 18s, forward (F), 5'-TGCGGAAGGATCATTAACGGA-3', reverse (R), 5'-AGTAGGAGAGGAGCGAGCGACC-3'; RUNX1, F, 5'-CAAGTGGCGAGATTCAAC-3', R, 5'-TGATGGCTCTATGGTAGGT-3'; collagen type III alpha 1 chain (COL3A1), F, 5'-CGAGTATGGAAGCGAAGG-3', R, 5'-GCAGTGATAGGTGATGTTCT-3'; collagen type I alpha 1 chain (COL1A1), F, 5'-GTTCTCCTGGTGCTGCTGGTC-3', R, 5'-CCATGTGGTCCAACTGGTCCTC-3'; PTN, F, 5'-CAATACCGCCTTGAAGACCAGAAC-3', R, 5'-GGCTTGGGCTTGGTGAGTTTG-3'.

Western blot analysis

Proteins were extracted from mouse heart tissues and neonatal mouse cardiac fibroblasts (NMCFs) using radioimmunoprecipitation assay (RIPA) buffer supplemented with protease inhibitor (phenylmethylsulfonyl fluoride) and phosphatase inhibitors (#P002, MedChemExpress, Suzhou, China). Protein concentration was quantified with a bicinchoninic acid (BCA) assay kit (#KGB2101, KeyGEN BioTECH). Equal amounts of protein were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and transferred to PVDF membranes (#1620177, BIO-RAD, Hercules, CA, USA). The membranes were blocked with 5% skim milk in Tris-buffered saline with Tween-20 (TBST) for 1 hour at room temperature, and incubated with primary antibodies (RUNX1: 25315-1-AP; COL3A1: 22734-1-AP; COL1A1: 14695-1-AP; α-SMA: 80008-1-RR; β-actin: 66009-1-Ig all from Proteintech; PTN: AF06961, AiFang, Wuhan, China.) at 4 ℃ overnight. The membranes were washed with 0.05% TBST (3×10 minutes), and incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (ABclonal) at room temperature for 1 hour. Protein bands were detected with the superFemto enhanced chemiluminescence (ECL) chemiluminescence kit (Vazyme Biotech Co., Ltd.), quantified using ImageJ software, and normalized to β-actin.

Cell transfection

RUNX1-small interfering RNA (siRNA), PTN-siRNA, and RUNX1 overexpression plasmids were purchased from GenePharma Biotechnologies (Shanghai, China). Primary cardiac fibroblasts in logarithmic growth phase were seeded into six-well cell culture plates, and transfection was performed with KeygenMAX 3000 (#KGA9705, KeyGEN BioTECH) at 60–80% confluence, in accordance with the manufacturer’s instructions.

AAV9 transduction

AAV9 vectors [short hairpin RNA Runx1 (shRunx1) or empty vector] were purchased from GeneChem Co., Ltd. (Shanghai, China). Mice were randomly divided into control and experimental groups (n=6 per group). During MI surgery, 200 µL of AAV9 (3×10¹¹ vector genomes per mouse) was administered via tail vein injection. Four weeks after injection, mouse hearts were harvested for Western blot and immunofluorescence analyses to verify transduction efficiency. The AAV-shRunx1 construct contained the Periostin-EGFP-MIR155(MCS)-SV40PolyA sequence.

RNA sequencing and analysis

NMCFs treated with si-NC or si-Runx1 under TGF-β1 in three biological replicates per condition for 24 h had their RNA extracted and analyzed by Ouli Biotechnology Co., Ltd. (Shanghai, China). Libraries were constructed with the VAHTS Universal V6 RNA-seq Library Prep Kit and sequenced on the Illumina NovaSeq 6000 platform in paired-end 150 bp mode, with a target depth of approximately 40 million read pairs per sample. Reads were aligned to the mouse reference genome GRCm39 (Ensembl release 106) using STAR v2.7.10a. Gene-level counts were generated by featureCounts v2.0.2 and normalized using the trimmed mean of M values method in edgeR. Principal component analysis and hierarchical clustering were conducted using R (version 44), after removing genes with read counts <10 in fewer than four samples. Gene Ontology (GO) enrichment analysis was performed using the GOseq R package (adjusted P<0.05), and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was conducted using the ClusterProfiler R package (P<0.05).

Chromatin immunoprecipitation (ChIP) assays

The JASPAR database was used to identify Runx1 binding sites in the potential PTN promoter region. ChIP assays were performed in accordance with the manufacturer’s instructions using the ChIP assay kit (#P2078, Beyotime, Shenzhen, China). Chromatin DNA was extracted from HEK293T cells transfected with the Runx1-Flag plasmid for 72 hours, followed by chromatin cross-linking with 1% formaldehyde. Subsequently, protein-DNA complexes were immunoprecipitated using a Flag antibody or control immunoglobulin G (IgG). After cross-linking, qPCR was conducted to detect the immunoprecipitated chromatin samples. The primer sequences for ChIP-qPCR were as follows: sense sequence (5'-3'): GGAGAGGAAGGTCTCAGGAAGC, and antisense sequence (5'-3'): CGGGTCACATATTTTGTGGCATG.

Luciferase reporter assay

The synthetic promoter regions, comprising only the wild-type and mutant sequences, were cloned into the pGL3 luciferase vector. Following a 48-hour transfection period, luciferase activity was assessed using the dual-luciferase assay kit (Beyotime).

Statistical analysis

All data were presented as the mean ± standard error of the mean (SEM). Normality was assessed using the Shapiro-Wilk test. Comparisons between two groups were performed using an unpaired t-test; otherwise, the Wilcoxon rank-sum test was applied. For comparisons among multiple groups, one-way analysis of variance followed by Tukey’s honestly significant difference post hoc test was conducted. A P value <0.05 was considered statistically significant. All statistical analyses were conducted using GraphPad Prism (version 8.0).

Results

RUNX1 expression increases following MI in mice

To investigate alterations in gene expression in cardiac tissue following MI, we retrieved and analyzed RNA-sequencing data from the Gene Expression Omnibus (GEO) database (GSE236374), comprising infarcted heart tissue samples (n=3) and sham-operated controls (n=3). Following stringent quality control and removal of lowly expressed genes, differential expression analysis was performed based on the following criteria: |log₂fold change| >1 and adjusted P value <0.05. In total, 1,800 significantly upregulated and 291 significantly downregulated genes were identified (Figure 1A). A heatmap was generated to visualize the top 20 upregulated genes (Figure 1B), among which RUNX1 was selected for further investigation due to its marked differential expression and its prior implication in fibroblast activation across organs. KEGG pathway enrichment analysis revealed significant enrichment of ECM deposition and fibrosis pathways (Figure 1C). Subsequently, a mouse model of MI was established, which demonstrated significant upregulation of RUNX1 in the infarcted myocardium compared to the sham-operated controls (Figure 1D).

Figure 1.

RUNX1 expression increases following MI in mice. (A) Volcano plot showing 75 upregulated and 21 downregulated genes in infarcted versus sham hearts (|log₂fold change| >1, P<0.05). (B) Heatmap displaying expression patterns of the top 20 upregulated genes. (C) KEGG enrichment analysis showing significant enrichment of ECM deposition and fibrosis pathways. *, P<0.05. (D) RUNX1 protein expression was significantly upregulated in infarcted myocardium (n=6). ***, P<0.001. Data are presented as mean ± SEM. CPM, counts per million; ECM, extracellular matrix; FC, fold change; FDR, false discovery rate; KEGG, Kyoto Encyclopedia of Genes and Genomes; MI, myocardial infarction; RUNX1, runt-related transcription factor 1; SEM, standard error of the mean.

Knockdown of RUNX1 ameliorates cardiac interstitial fibrosis and improves cardiac function following MI in adult mice

To examine the effects of RUNX1 deficiency on cardiac fibrosis following MI, an AAV9 vector containing the shRunx1 gene with the fibroblast-specific Postn promoter (AAV9-Postn-shRunx1) was delivered via tail vein injection into the myocardial tissue of the adult mice to knockdown RUNX1, with AAV9-Postn-short hairpin RNA negative control (shNC) serving as the control. To assess AAV9 infection efficiency in fibroblasts, cardiac fibroblasts were isolated from the hearts of adult mice 28 days after AAV9-Postn-shRunx1 injection. Knockdown efficiency was confirmed by Western blot analysis (Figure 2A).

Figure 2.

RUNX1 knockdown improved cardiac function and cardiac fibrosis following MI in adult mice. (A) Western blot analysis of RUNX1 expression in cardiac fibroblasts of mice and quantification of the indicated proteins (n=6/group). (B,C) Immunofluorescence staining analysis showing RUNX1, GFP, and vimentin co-staining to confirm knockdown efficiency of the gene transfection system (n=6/group; scale bar =100 µm). (D) Cardiac function was assessed by echocardiography at pre-MI and 28 days after MI (n=6/group). (E) Representative images of H&E and Masson’s trichrome staining of mouse hearts (n=6/group; scale bar =1 mm). (F) Representative immunohistochemical staining of α-SMA and RUNX1 in transverse sections of hearts from the adult mouse MI model at 28 days after MI (n=6/group; scale bar =50 µm). (G) Western blot analysis of α-SMA, COL1A1, and COL3A1 protein levels in fibroblasts. The cells were isolated from the hearts of mice with MI (n=6/group). ***, P<0.001. Data are presented as mean ± SEM. α-SMA, α-smooth muscle actin; COL1A1, collagen type I alpha 1 chain; COL3A1, collagen type III alpha 1 chain; DAPI, 4',6-diamidino-2-phenylindole; GFP, green fluorescent protein; H&E, hematoxylin and eosin; LVEF, left ventricular ejection fraction; LVFS, left ventricular fractional shortening; MI, myocardial infarction; RUNX1, runt-related transcription factor 1; SEM, standard error of the mean; shNC, short hairpin RNA negative control; shRUNX1, short hairpin RNA RUNX1.

Immunofluorescence staining was also performed on myocardial tissues from both the AAV9-Postn-shNC group and the AAV9-Postn-shRunx1 group following MI. The results showed co-localization of GFP and vimentin in the AAV9-injected cardiac tissues (Figure 2B). Further, compared with the AAV9-Postn-shNC group, the AAV9-Postn-shRunx1 group exhibited significantly reduced co-localization of RUNX1 and vimentin (Figure 2C). These results suggest that the gene transfection system can effectively regulate the expression of RUNX1 in mouse fibroblasts.

At 4 weeks after MI, echocardiographic analysis revealed significant reductions in both LVEF and LVFS in the infarcted mice group compared with the sham group. Notably, the mice treated with AAV9-Postn-shRunx1 exhibited a significant improvement in cardiac function (Figure 2D), indicating that targeting RUNX1 effectively alleviated adverse cardiac function following MI.

We further investigated the effects of RUNX1 on myocardial fibrosis following MI. H&E staining revealed that the sham group exhibited smaller cardiomyocyte cross-sectional areas, intact cellular structures, well-organized myofibrils, and an absence of inflammatory cell infiltration. In contrast, the MI group displayed disorganized myocardial fibers, cardiomyocyte hypertrophy, and significant inflammatory infiltration. Notably, myocardial injury was significantly attenuated in the AAV9-Postn-shRunx1-MI mice.

Masson’s trichrome staining was used to evaluate cardiac fibrosis in mice following MI. Compared with the AAV9-Postn-shNC group, the AAV9-Postn-shRunx1-MI group exhibited a significant reduction in myocardial fibrosis (Figure 2E). Subsequently, immunohistochemical staining was performed. Compared with the sham group, the MI mice injected with AAV-shNC exhibited significantly increased expression of α-SMA. In contrast, the mice treated with AAV-shRunx1 showed reduced expression of fibrotic markers compared with the AAV-shNC group (Figure 2F).

Primary cardiac fibroblasts were isolated from the mouse hearts for Western blot analysis. Consistent with Masson’s trichrome results, the mice injected with AAV-shNC showed increased expression of fibrotic markers compared to the sham group. Further, COL3A1, COL1A1, and α-SMA levels were significantly reduced in the AAV-shRUNX1-MI group compared to the AAV-shNC-MI group (Figure 2G). Taken together, our data indicate that RUNX1 acts as a key driver of cardiac interstitial fibrosis following MI.

RUNX1 deficiency inhibits the phenotypic transition of NMCFs

During the process of myocardial fibrosis, cardiac fibroblasts undergo a series of pathological events, including phenotypic transition and collagen secretion. These changes are considered fundamental pathological alterations in the heart following MI (18). To elucidate the role of RUNX1 in cardiac fibroblast activation, primary cardiac fibroblasts were extracted from C57BL/6J lactating mice within 3 days of birth and transfected with siRNA. Cells were stimulated with TGF-β1 for 48 hours to induce activation (19). Runx1 knockdown markedly reduced the messenger RNA (mRNA) levels of Col3A1, Col1A1, α-SMA, and Runx1 itself (Figure 3A), as well as the protein levels of COL3A1, COL1A1, α-SMA, and RUNX1 (Figure 3B), indicating that RUNX1 is required for TGF-β-mediated activation of cardiac fibroblasts. Moreover, wound-healing assays and transwell migration assays revealed that the enhanced migratory capacity of NMCFs under TGF-β was significantly diminished by RUNX1 silencing (Figure 3C,3D). Together, these data provide strong evidence that RUNX1 facilitates multiple aspects of TGF-β-induced fibroblast dysfunction.

Figure 3.

RUNX1 regulates cardiac fibroblast activation. (A) The mRNA levels of Col3A1, Col1A1, α-SMA, and Runx1 were measured by qPCR (n=3/group). (B) Western blot analysis of siRUNX1 transfection efficiency and the protein levels of COL3A1, COL1A1, and α-SMA after RUNX1 knockdown (n=3/group). (C) Representative images of wound-healing assays in fibroblasts transfected with siRunx1. Images of the wounds were captured at 0 and 24 hours. The area between the wound edges was measured and compared between the groups (scale bar =200 µm; n=3/group). (D) Cells were fixed with 4% paraformaldehyde for 30 min, followed by staining with crystal violet at room temperature for 1 h. Representative images of Transwell migration assay evaluating NRCF migration (scale bar =200 μm; n=3/group). *, P<0.05; **, P<0.01; ***, P<0.001. Data are presented as mean ± SEM. α-SMA, α-smooth muscle actin; COL1A1, collagen type I alpha 1 chain; COL3A1, collagen type III alpha 1 chain; mRNA, messenger RNA; NRCF, neonatal rat cardiac fibroblast; qPCR, quantitative polymerase chain reaction; RUNX1, runt-related transcription factor 1; SEM, standard error of the mean; siRUNX1, small interfering RNA RUNX1; TGF-β, transforming growth factor-β.

RNA-sequencing indicates that PTN is involved in ventricular remodeling regulation by RUNX1

To investigate the mechanism by which RUNX1 drives fibroblast activation, RNA-sequencing was performed on primary cardiac fibroblasts after RUNX1 knockdown and TGF-β treatment. The differentially expressed genes were most prominently enriched in pathways related to protein synthesis and secretion (Figure 4A,4B). We then compared these data with the MI transcriptome (GSE236374). PTN was upregulated in the MI database but downregulated upon RUNX1 depletion. A previous study reported that PTN is upregulated in hypertrophic scars (20). Therefore, we speculate that PTN may represent a potential mechanism through which silencing RUNX1 expression can alleviate myocardial fibrosis.

Figure 4.

RNA-sequencing indicates that PTN is involved in ventricular remodeling regulation by RUNX1. (A) Volcano plot of differentially expressed genes in NRCFs after RUNX1 knockdown. (B) KEGG pathway analysis of differentially expressed genes. (C,D) Western blot analysis of PTN protein levels in myocardial tissue from MI mice (n=6; ***, P<0.001). (E,F) Western blot analysis of PTN protein levels in siRunx1-transfected NRCFs (n=3). (G,H) qPCR quantification of Runx1 and Ptn mRNA levels in siRunx1-transfected NRCFs (n=3). (I) Predicted RUNX1-binding sites in the PTN promoter identified using the JASPAR database. (J) ChIP-qPCR confirming RUNX1 occupancy at the Ptn promoter. (K) Dual-luciferase reporter assay demonstrating that RUNX1 positively regulates Ptn transcription. *, P<0.05; **, P<0.01; ***, P<0.001; ns, not significant. Data are presented as mean ± SEM. ChIP, chromatin immunoprecipitation; FC, fold change; IgG, immunoglobulin G; KEGG, Kyoto Encyclopedia of Genes and Genomes; MI, myocardial infarction; mRNA, messenger RNA; Mut, mutant; NRCF, neonatal rat cardiac fibroblast; oe, overexpression; PTN, pleiotrophin; qPCR, quantitative polymerase chain reaction; RUNX1, runt-related transcription factor 1; SEM, standard error of the mean; siRNA, small interfering RNA; siRUNX1, small interfering RNA RUNX1; TGF-β, transforming growth factor-β; WT, wild type.

Compared with the sham hearts, PTN expression was markedly elevated in the MI tissues (Figure 4C,4D). In primary cardiac fibroblasts, RUNX1 silencing significantly reduced PTN mRNA and protein levels (Figure 4E-4H). Using the JASPAR database, we identified a RUNX1 binding site in the Ptn promoter region (Figure 4I). The binding of RUNX1 to the PTN promoter was further confirmed by the CHIP assay (Figure 4J). Dual-luciferase reporter assays further demonstrated that RUNX1 directly transactivated the Ptn promoter (Figure 4K). In conclusion, these results indicate that RUNX1 regulates the expression of PTN.

RUNX1 promotes cardiac fibroblast activation via PTN regulation

To further elucidate whether RUNX1 drives cardiac fibroblast activation via PTN, fibroblasts were co-transfected with a RUNX1 overexpression plasmid and siRNA-PTN. Cells overexpressing RUNX1 alone exhibited marked activation, whereas co-transfection of RUNX1 with PTN-siRNA abolished this effect (Figure 5). These data demonstrate that RUNX1 promotes cardiac fibroblast activation in a PTN-dependent manner.

Figure 5.

RUNX1 promotes cardiac fibroblast activation via PTN regulation. Western blot analysis of COL3A1, COL1A1, α-SMA, RUNX1, and PTN protein levels in NRCFs after RUNX1 overexpression alone or in combination with PTN-siRNA transfection (n=3). *, P<0.05; **, P<0.01; ***, P<0.001; ns, not significant. Data are presented as mean ± SEM. α-SMA, α-smooth muscle actin; COL1A1, collagen type I alpha 1 chain; COL3A1, collagen type III alpha 1 chain; Con, control; NRCF, neonatal rat cardiac fibroblast; oe, overexpression; PTN, pleiotrophin; RUNX1, runt-related transcription factor 1; SEM, standard error of the mean; siRNA, small interfering RNA.

Discussion

AMI is a clinical emergency caused by sudden coronary artery occlusion, leading to symptoms such as chest pain and dyspnea due to acute myocardial ischemia/hypoxia, which can be fatal in severe cases (21). Early revascularization interventions have effectively reduced acute-phase mortality; however, post-MI heart failure remains a major cause of long-term disability and death. Ischemic injury triggers a complex healing process involving cytokine-mediated fibroblast proliferation, migration to the injured area, and myofibroblast differentiation, leading to scar formation. While early reparative fibrosis is essential to prevent ventricular rupture, excessive fibrosis impairs cardiac function and progresses to heart failure (22,23). Therefore, inhibiting cardiac fibroblast activation represents a potential strategy for improving post-MI heart failure outcomes. In the present study, we show that RUNX1 is induced in the infarcted mouse myocardium and acts cell-autonomously in cardiac fibroblasts to drive their activation. We identify PTN as a direct transcriptional target of RUNX1 in fibroblasts, demonstrate that PTN is required for RUNX1-mediated fibroblast activation, and show that fibroblast-restricted RUNX1 knockdown in vivo reduces interstitial fibrosis and preserves systolic function. These data place the RUNX1-PTN axis downstream of MI injury in the fibroblast compartment and complement the emerging picture of cell-type-specific RUNX1 function in the remodeling heart.

Although RUNX1 has been extensively studied in hematologic malignancies, its role in cardiovascular disease remains poorly understood. RUNX1 is essential for early cardiovascular development and is broadly expressed in cardiac mesenchyme and vascular tissues (24). As a transcription factor, RUNX1 exerts its functions through downstream target genes (e.g., protein phosphatase 1), thereby impairing myocardial contractility, and also acts as a pro-fibrotic mediator in multiple organs (25). However, its specific role in post-MI cardiac remodeling is unclear. RUNX1 is induced in both cardiomyocytes and fibroblasts after MI, yet the functional consequences of its loss differ sharply between the two compartments. McCarroll et al. using an inducible cardiomyocyte-specific RUNX1 knockout, reported preserved contractile function that was attributable to improved sarcoplasmic reticulum calcium handling via a protein kinase A/phospholamban/SERCA2a mechanism (26). By contrast, a recent study using Postn-Cre-mediated deletion in activated fibroblasts found that fibroblast-specific Runx1 loss—rather than cardiomyocyte-specific loss—accounts for most of the protective effect on scar size, cardiomyocyte hypertrophy and myofibroblast generation (27). Our results align with the latter model: by delivering AAV9-Postn-shRunx1, we restricted RUNX1 knockdown to fibroblasts and still observed a marked reduction in collagen deposition and α-SMA expression, together with improved ejection fraction and fractional shortening. The simplest reconciliation of these data is that RUNX1 operates through distinct, compartment-specific effector programs—a calcium-handling program in cardiomyocytes and a pro-fibrotic, PTN-dependent program in fibroblasts—and that the dominant pathological contribution in the chronic phase is the fibroblast-autonomous one. Macrophage-specific RUNX1 deletion has also been reported to attenuate inflammation and indirectly restrain fibroblast activation (28), indicating that RUNX1 functions in at least three cardiac compartments. The fact that our fibroblast-targeted intervention phenocopies the global protective effect of RUNX1 inhibition places the fibroblast axis upstream of these indirect effects and supports it as the principal actionable node.

In the heart, RUNX1 has been linked to pressure overload-induced remodeling through epigenetic activation of myofibroblasts (13) and to N6-methyladenosine-dependent cytoskeletal remodeling in cardiac fibrosis (29). Across these studies, however, the direct transcriptional target responsible for RUNX1-driven fibroblast activation in the infarcted myocardium has remained undefined. To clarify the mechanism of RUNX1-mediated myocardial fibrosis, we performed RNA-sequencing on RUNX1-knockdown cardiac fibroblasts and focused on downregulated PTN. PTN is a heparin-binding growth factor involved in mitosis, inflammation, and angiogenesis (30), promoting cell migration via PTN/RPTPζ-mediated cytoskeletal remodeling (31) and proliferation via MAPK/PI3K-AKT pathways (32). PTN also binds anaplastic lymphoma kinase to promote fibroblast proliferation (33) and is upregulated in cirrhotic tissues, mediating cirrhosis-hepatocellular carcinoma progression (34). In the heart, PTN secreted by cardiac fibroblasts in hypertrophic cardiomyopathy promotes autocrine proliferation and macrophage-mediated inflammation via the PTN-syndecan-4 pathway (35). Our study showed that PTN is upregulated in activated fibroblasts and MI myocardium, downregulated after RUNX1 knockdown, and directly transcriptionally regulated by RUNX1 (via promoter binding). Importantly, PTN silencing abrogated RUNX1-induced fibroblast activation. Our use of the Postn promoter to restrict shRunx1 expression to activated fibroblasts demonstrates that the protective effect on cardiac function and fibrosis can be achieved without targeting RUNX1 in other compartments, and mechanistically attributes this effect to the PTN-dependent arm of RUNX1 action. This carries two implications. First, the RUNX1-PTN axis becomes a more selective druggable target than RUNX1 itself: a small molecule or decoy peptide that interrupts RUNX1 binding to the PTN promoter, or a PTN-neutralizing antibody delivered to the infarct region, could in principle reproduce the fibroblast-specific benefit while sparing the calcium-handling and immune functions of RUNX1 in other cells. Second, because PTN is a secreted factor, its levels in the infarct border zone may serve as a pharmacodynamic readout for response to RUNX1-directed therapy, and as a candidate biomarker for fibrotic activity in patients after MI (17).

Limitations

Our mechanistic dissection was performed in NMCFs, which differ from adult activated fibroblasts in proliferation rate and matrifibrocyte potential; the in vivo model captures a single 4-week time point and does not address the early phase of reparative fibrosis; and we did not test whether PTN rescue restores the fibrotic phenotype in Runx1-deficient fibroblasts in vivo. Resolving these questions, and defining the precise chromatin landscape at the PTN promoter in activated versus quiescent fibroblasts, will be required before the RUNX1-PTN axis can be advanced as a clinical target.

Conclusions

RUNX1 is significantly upregulated after MI and promotes myocardial fibrosis by transcriptionally activating PTN and inducing cardiac fibroblast activation. RUNX1 knockdown attenuates post-MI myocardial fibrosis, highlighting RUNX1 as a potential therapeutic target for post-MI ventricular remodeling.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was approved by the Animal Ethics Committee of Nanjing First Hospital affiliated to Nanjing Medical University (No. DWSY25094727), in compliance with institutional guidelines for the care and use of animals.

Data Sharing Statement

Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1900/dss

jtd-18-08-945-dss.pdf ^{(113.2KB, pdf)}

DOI: 10.21037/jtd-2026-1900