S100A9 Aggravates Cardiac Fibrosis by TLR4/PGC‐1α Mediated Macrophage‐Myofibroblast Crosstalk

ABSTRACT

Fibrosis is a critical component of ventricular remodelling after myocardial ischemia, and the activation and expansion of cardiac fibroblasts represent key drivers of this process. Our previous work identified S100A9‐dependent macrophage‐to‐myofibroblast transition (MMT) as a newly recognized source of myofibroblasts in the post‐MIR heart. However, the mechanisms by which S100A9 regulates MMT remain incompletely understood. Here, using multiple genetically engineered mouse models, MIR and MI model, and in vitro multicellular co‐culture systems, we investigated the role of S100A9 in regulating fibroblast‐derived migrasome release during MMT. We found that macrophage‐derived S100A9 promotes mitochondrial dysfunction and migrasome release in cardiac fibroblasts through TLR4/PGC1α signalling following MIR. These fibroblast‐derived migrasomes, in turn, activate integrin/Src signalling in macrophages, triggering MMT and accelerating cardiac fibrosis. Importantly, we delineate how macrophage‐derived S100A9 orchestrates crosstalk between fibroblasts and immune cells and identify migrasome‐mediated activation of MMT as a previously unrecognized mechanism driving post‐MIR fibrotic remodelling. Our findings suggest that targeting S100A9‐induced migrasome release and downstream MMT signalling may represent a promising therapeutic strategy to mitigate pathological cardiac fibrosis.

1. Introduction

Cardiac fibrosis is a fundamental component of post‐infarction cardiac remodelling. Although moderate fibrosis is essential for preserving ventricular structural integrity and preventing cardiac rupture, excessive fibrosis impairs normal systolic and diastolic function (Talman and Ruskoaho 2016). Acute ischemic injury of cardiomyocytes triggers inflammation and cell death, which in turn stimulates proliferation and activation of myofibroblasts, leading to the formation of fibrotic foci (Daseke et al. 2020). These fibrotic foci constitute the pathological basis for the initiation and progression of post‐infarction fibrosis (Ma et al. 2017). Therefore, understanding the mechanisms underlying myofibroblast activation and proliferation is critical for developing strategies to limit maladaptive cardiac remodelling.

Myofibroblasts are the primary effector cells driving cardiac fibrosis. Upon sensing cardiac injury, activated myofibroblasts migrate toward the infarcted region—a process that requires substantial energy and robust metabolic activity. Mitochondria, as the cellular powerhouses, are central to this process (Gibb et al. 2020). Elevated metabolic demand generates excessive reactive oxygen species (ROS) and increases cellular stress, leading to mitochondrial dysfunction (Li et al. 2020). During proliferation and migration, dysfunctional mitochondria are packaged into and extruded via migrasomes (Jiao et al. 2021). Migrasomes are recently described class of extracellular vesicles, 0.5–3 µm in diameter, enclosed by plasma membrane and containing mitochondria and other intracellular components (Ma et al. 2015). They serve as key mediators of intercellular communication. During cell migration, retracting fibres extend from the rear of the cell, and fragments of these fibres are left behind in the extracellular space as migrasomes (Ma et al. 2015). Previous studies have shown that migrasomes released by myofibroblasts can promote fibrosis in lung tissue.

Macrophages are critical regulators of cardiac fibrosis and repair (Yap et al. 2023). They not only secrete pro‐fibrotic factors such as TGF‐β to stimulate myofibroblast activation but can also undergo phenotypic conversion into myofibroblast‐like cells, further promoting fibrosis. Macrophage‐to‐myofibroblast transition (MMT) has been reported in fibrosis of the heart, lung, kidney, and interstitial (Wei et al. 2022, Wang et al. 2017, ZHAO et al. 2024, Shen et al. 2024). The SMAD3/TGF‐β axis has been recognized as a central mechanism controlling MMT, while Src acts as a key regulatory node for SMAD3/TGF‐β signalling (Shen et al. 2024). Migrasomes are enriched in signature proteins such as integrin α5, which is known to regulate Src‐mediated cell migration and invasion in multiple cancers (Bianchi‐Smiraglia et al. 2013). However, its role in macrophage phenotypic conversion remains largely unexplored.

In this study, we hypothesize that after myocardial ischemia‐reperfusion (MIR) or myocardial infarction (MI), infiltrating macrophages release S100A9, which promotes fibroblast migration and activation and induces migrasome release. These fibroblast‐derived migrasomes, in turn, interact with macrophages to drive MMT and exacerbate cardiac fibrosis. Using genetically engineered mouse models of MIR and MI, and in vitro coculture systems, we investigated the mechanisms by which myofibroblasts mediate MMT. Specifically, we examined mitochondrial functional alterations and migrasome formation in S100A9‐stimulated myofibroblasts, highlighting the role of multicellular communication in MMT and identifying potential therapeutic targets for post‐MI cardiac fibrosis.

2. Methods

2.1. Human Experiments

Peripheral blood samples were collected from patients with acute MI within 24 h after coronary reperfusion and from healthy control subjects without a history of myocardial infarction at the Department of Cardiology, The First Affiliated Hospital of University of Science and Technology of China (USTC). Serum or peripheral blood monocytes were isolated for subsequent experiments. All participants provided written informed consent prior to enrolment. All procedures and experiments were approved by the Institutional Bioethics Committee of The First Affiliated Hospital of University of Science and Technology of China (2025KY‐155).

2.2. Animal Experiments

Male C57BL/6 mice (6–8 weeks old) were housed at the Animal Experiment Centre of the First Affiliated Hospital of the University of Science and Technology of China. All experimental procedures were conducted in accordance with protocols approved by the Animal Ethics Committee of the same institution (2025‐N(A)‐066).

2.2.1. Animal Model

Anaesthesia was maintained with mechanical ventilation and continuous inhalation of isoflurane (RWD Life Science, Shenzhen, China). Thoracic hair was removed, and the skin was incised to expose the heart through the third to fourth intercostal space at the left sternal border. For the MI model, the left coronary artery was permanently ligated at a site 1 mm below the left atrial appendage. For the MI/R model, the coronary artery was ligated at the same location for 30 min, after which the ligature was released to allow reperfusion. Subsequently, the ribs, muscle layer, and skin were closed in layers. Sham‐operated mice underwent identical procedures, except that the coronary artery was not ligated.

2.2.2. Animal Experiment 1

S100a9‐gene knockout mice on a C57BL/6 background (S100a9 ^KO) and macrophage‐specific S100a9 knockout mice (S100a9 ^{FL/FL‐Lyz2Cre}, hereafter referred to as S100a9 ^MKO) were generated. MI and MIR models were established in these mice, and animals were euthanized on Day 7 post‐surgery for tissue collection.

2.2.3. Animal Experiment 2

Wild‐type (WT) mice were administered recombinant S100A9 (rS100A9, 0.6 mg/kg, Novoprotein) via tail vein injection (Chalise et al. 2022). Serum S100A9 levels were measured in PBS‐treated controls and at 6, 12, 24, and 48 h post‐injection. Subsequently, following MI or MIR model establishment, mice received Intravenous injection via the tail vein of rS100A9 immediately after surgery, once every 48 h. Animals were euthanized on Day 7 for tissue collection.

2.2.4. Animal Experiment 3

MI and MIR models were established in WT mice. Animals were then administered an S100A9 neutralizing antibody (NTAB, 100 µg/mouse, Sigma, MABF290) or an isotype control via tail vein injection every 2 days (Li et al. 2019). Mice were euthanized on Day 7 for tissue collection.

2.2.5. Animal Experiment 4

Twenty‐four hours prior to MIR, mice were intraperitoneally injected with ATN‐161 (100 mg/kg, every 3 day) (MCE, HY‐13535A) or saline and treated until 24 h before sacrifice (Stoeltzing et al. 2003). Control group received identical solvent without drug following the same protocol.

2.2.6. Animal Experiment 5

MI and MIR models were established in WT mice. Animals were then administered the pharmacological S100A9 inhibitor Tasquinimod (5 mg/kg/day; ABR‐215050, MCE) or saline via intraperitoneal injection at 6 and 24 h post‐surgery (Li et al. 2019). Treatment continued until 24 h before euthanasia.

2.3. Macrophage Fluorescence Tracing

KikGR knock‐in mice were crossed with Lyz2 ^Cre mice to generate KIKGR‐Lyz2 ^Cre offspring, in which macrophages constitutively exhibit green fluorescence. Upon exposure to ultraviolet light (350–410 nm), the green fluorescence is irreversibly converted to red fluorescence. Bone marrow‐derived macrophages (BMDMs) were isolated from these mice, cultured to maturity in vitro, and then infused via the tail vein at 24 h after MIR. On Day 7 post‐MIR, the hearts were collected, fixed, sectioned, and stained to assess colocalization of red fluorescence with α‐SMA in the cardiac tissue.

2.4. Masson Staining

Mouse hearts were collected and fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned. The sections were deparaffinized, rehydrated, and sequentially stained with haematoxylin, Biebrich scarlet–acid fuchsin, and aniline blue (G1006, Servicebio), After washing, the sections were mounted, and images were captured and analysed under a light microscope.

2.5. Echocardiography

Mice were anesthetized with continuous inhalation of isoflurane and secured on the operating platform. Thoracic hair was removed, and ultrasound gel was applied to the probe, which was positioned on the chest to assess cardiac function. Cardiac cycles were recorded using VINNO (VINNO, Suzhou, China), and parameters including left ventricular ejection fraction (LVEF), fractional shortening (FS), and Left Ventricular Posterior Wall Systolic thickness (LVPWs) were calculated.

2.6. Serum S100A9 Measurement

Serum S100A9 levels were measured using the Mouse S100A9 ELISA Kit (E‐EL‐M3049) and Human S100A9 ELISA Kit (E‐EL‐H6197), respectively. Briefly, serum samples were collected, and standard solutions and biotinylated antibody working solution were prepared according to the manufacturer's instructions. Blank, standard, and sample wells were designated on the microplate. After adding the corresponding specimens, the plate was incubated at 37°C for 90 min. The liquid was then discarded, and 100 µL of biotinylated antibody working solution was added to each well, followed by incubation at 37°C for 60 min. After washing, 100 µL of HRP enzyme conjugate working solution was added, followed by 90 µL of substrate solution for colour development. The reaction was terminated, and the optical density (OD) of each well was measured at 450 nm using a microplate reader. Sample concentrations were calculated based on the standard curve generated from the standard concentrations and their corresponding OD values, followed by statistical analysis.

2.7. Immunofluorescence

Mouse heart sections were stained using a tyramide signal amplification (TSA) kit (G1226, Servicebio). Briefly, paraffin‐embedded sections were deparaffinized and rehydrated, followed by antigen retrieval and blocking of endogenous peroxidase. Subsequently, sections were incubated with the primary antibody overnight at 4°C, followed by incubation with fluorescence‐conjugated TSA working solution. After washing away excess TSA solution, sections were sequentially incubated with the second primary antibody and fluorescence‐conjugated TSA working solution. Following additional washes, sections were mounted with an anti‐fading mounting medium containing DAPI.

For cell immunofluorescence, cells grown on coverslips were fixed and permeabilized with Triton X‐100 incubated overnight at 4°C with a mixture of primary antibodies from different species. After washing, the appropriate fluorescent secondary antibody mixtures were applied and incubated at room temperature in the dark for 30 min. Excess secondary antibodies were removed by washing, and the sections were mounted with a DAPI‐containing anti‐fade mounting medium (P0131, Beyotime, China). Images were captured and analysed using a fluorescence microscope. The specific antibodies used in this study included: CD68 (CST, 97778), Col1α1 (CST, 72026), α‐SMA (HUABIO, ET1607‐53), goat anti‐mouse IgG‐488 (Abcam, ab150113), goat anti‐rabbit IgG‐594 (Abcam, ab150080).

2.8. Single‐Nucleus RNA‐Sequencing Data Analysis

Publicly available single‐nucleus RNA‐sequencing (snRNA‐seq) data of murine MI/R injury and sham hearts were obtained from the GEO database (GSE227088). Data processing and downstream analyses were performed using the OmicVerse framework integrated with Scanpy in Python. Low‐quality nuclei were removed according to standard quality‐control criteria, including low gene counts and high mitochondrial transcript proportions. After normalization and log transformation, highly variable genes were identified for downstream analyses. Batch effects among samples were corrected using Harmony. Principal component analysis (PCA), neighbourhood graph construction, Uniform Manifold Approximation and Projection (UMAP), and Leiden clustering were subsequently performed. Cell populations were annotated based on canonical marker genes. Differentially expressed genes between I/R and sham groups were identified using the Wilcoxon rank‐sum test.

2.9. Isolation and Differentiation of Bone Marrow–Derived Macrophages

After euthanasia, the long bones of both hind limbs were collected. Bones were immersed in 75% ethanol for 5 min and then washed in PBS. The bones were cut open, and bone marrow cells were flushed out and filtered through a 70‐µm strainer. The cell suspension was centrifuged at 1500 × g for 5 min, and the pellet was resuspended in red blood cell lysis buffer (R1010, Solarbio). After lysis and subsequent centrifugation, cells were cultured in 10% FBS (KY‐01000S, KangYuan Biology)–supplemented RPMI‐1640 containing M‐CSF (315‐02, PeproTech) at 37°C in a 5% CO₂ incubator for 7 days to obtain bone marrow–derived macrophages (BMDMs) for downstream experiments.

2.10. Isolation and Culture of Primary Mouse Cardiac Fibroblasts

Hearts were collected from 3‐day‐old neonatal mice. The chest was opened at the mid‐sternum to expose the heart, which was then excised, rinsed in sterile PBS to remove blood, and minced into 1 mm³ tissue pieces. The tissue was digested with 0.25% trypsin (CN0004, Sparkjade Biotechnology Co., Ltd.), with repeated pipetting until the solution became turbid. The cell suspension was collected, filtered through a 70‐µm strainer, and digestion was terminated by adding 10% DMEM. Remaining tissue pieces were subjected to repeated digestion until no visible tissue fragments remained. After centrifugation, the cell pellet was resuspended in DMEM supplemented with 10% FBS (KY‐01000S, KangYuan Biology) and incubated at 37°C in a 5% CO₂ incubator for 1 h. The supernatant was then discarded, and fresh DMEM supplemented with 10% FBS was added. The adherent cells at this stage were identified as primary mouse cardiac fibroblasts.

For detection of S100A9 binding effector receptor of fibroblast and downstream signalling, Fibroblasts were stimulated with rS100A9 (1 µg/mL, Novoprotein, C795) in accordance with previous studies (Li et al. 2019). Concurrently, the following inhibitors were applied for 24 h: TLR4 (TAK242, 10 nM, HY‐11109, MCE), RAGE (FPS‐ZM1, 50 nM, HY‐19370, MCE), p38 (PD169316, 10 µM, HY‐10578, MCE), JNK (SP600125, 10 µM, HY‐12041, MCE), ERK (UO126, 10 µM, HY‐12031A, MCE), and p65 (BMS‐345541, 10 µM, HY‐10519, MCE) (Gong et al. 2026). SAR407899 (10 µM, MCE, HY‐15687A) and GW4869 (10 µm, MCE, HY‐19363) were used to inhibit migrasome and exosome formation, respectively (Peng et al. 2025, Essandoh et al. 2015). Cells were then harvested for subsequent analyses.

2.11. Cell Migration Assay

Cardiac fibroblasts were seeded into Transwell chambers (8‐µm pore size). Cells were cultured in serum‐free medium in the upper chambers, while the lower chambers contained 10% serum‐supplemented medium with either PBS or rS100A9 (1 µg/mL, Novoprotein, C795). After 24 h, cells were fixed in 4% PFA and stained with 0.1% crystal violet (G1063, Solarbio) for 30 min. Non‐migrated cells on the upper membrane surface were removed with a cotton swab, and migrated cells were imaged and counted under a microscope.

2.12. Coculture Experiments

Coculture experiments were performed using Transwell systems. To assess the effects of rS100A9‐stimulated fibroblasts on macrophages or cardiomyocytes, macrophages or cardiomyocytes were seeded in 6‐well plates (lower compartment) in 1640 medium with 10% FBS. Upper chambers (3‐µm pore size) received PBS, rS100A9, cardiac fibroblasts alone, or rS100A9‐stimulated cardiac fibroblasts. After 24 h of coculture, macrophages were collected for analysis.

To test whether rS100A9‐stimulated fibroblasts promote macrophage phenotypic transition via migrasome release, rS100A9‐stimulated cardiac fibroblasts were seeded into upper chambers with either 0.4 µm or 3 µm pore sizes. SAR407899, a migrasome release inhibitor, was also co‐administered with rS100A9 to fibroblasts. Macrophages were cocultured for 24 h and subsequently analysed.

2.13. Plasmid and si‐RNA Transfection

PEI (Fushenbio, FSF0001) and plasmids were separately mixed with Opti‐MEM at a 3:1 ratio and incubated for 5 min. The two mixtures were then combined, gently mixed, and incubated for an additional 20 min before being added to the serum‐free cultured cells. After 8 h, the medium was replaced with normal culture medium containing 10% foetal bovine serum (FB15‐15, Clark), and subsequent experiments were performed 48 h later. The sense sequences were as follows: Mouse Tspan4: 5’‐GCAUGAUCCGCUUCUACAATT‐3’. Tlr4: 5’‐ATAGAGGTAGTTCCTAATATT‐3’. Erk1/2:5’‐GGACCAGCUCAACCACAUU‐3’. 5’‐GCUCUUGAAGACACAGCAC‐3’. NfκB:5’‐GCAGAUCAGCCUCCAUCGUTT‐3’. Tnfα:5’‐GUGCCUAUGUCUCAGCCUCUUTT‐3’. Integrinα5:5’‐GCUAUGUCACCAUCCUUAATT‐3’.

2.14. Transmission Electron Microscopy

For transmission electron microscopy (TEM) analysis, cardiac fibroblasts were stimulated with rS100A9, then harvested and fixed in 2.5% glutaraldehyde overnight at 4°C. Subsequently, cells were dehydrated in a graded series of ethanol (30%, 50%, 70%, 90%, and 100%) and embedded in resin. Ultrathin sections (approximately 70–100 nm) were prepared using an ultramicrotome. Sections were stained with uranyl acetate for 10–15 min, followed by lead citrate for 5–10 min. Images were acquired using a transmission electron microscope.

2.15. Isolation and Purification of Migrasomes

Migrasomes were isolated and purified following previously established protocols with minor modifications (7, 20). Briefly, fibroblasts were seeded into 150‐mm culture dishes and harvested by enzymatic dissociation. After removing intact cells by centrifugation at 4000 × g, the supernatant was subjected to centrifugation at 20,000 × g to collect crude migrasome fractions. The pellet was then further purified using a density‐gradient ultracentrifugation step at 150,000 × g with a commercial lysosome isolation kit (Sigma‐Aldrich, LYSISO1), according to the manufacturer's instructions. The final migrasome preparation was resuspended in PBS and validated by immunoblot analysis.

2.16. Measurement of OCR and ECAR

OCR and ECAR were measured using commercial kits (E‐BC‐F070 and E‐BC‐F069, Elabscience). Cardiac fibroblasts were plated in 96‐well plates and treated as indicated. According to the manufacturer's protocol, 100 µL of working solution was added to each well, followed by incubation at 37°C for 30 min. After adding Reagent 2, kinetic measurements were performed at 37°C for 90 min using a microplate reader.

2.17. ATP Assessment

Cellular ATP levels were determined using an Enhanced ATP Assay Kit (S0027, Beyotime). Cardiac fibroblasts cultured in 6‐well plates were treated as indicated. Cells were lysed with 200 µL of lysis buffer, and the lysates were centrifuged at 12,000 × g for 5 min at 4°C. The supernatants were collected for analysis. To measure ATP, assay working solution was first added to each well to eliminate background ATP. Subsequently, 100 µL of assay solution was added, followed by 20 µL of sample or ATP standard. Luminescence was recorded as relative light units (RLU) using a luminometer, and ATP concentrations were calculated based on the standard curve.

2.18. Flow Cytometry

Cells were incubated with probes according to the manufacturer's instructions for the ROS (S0033S, Beyotime) and TMRE assay kits (HY‐D0985A, MCE) to explore the level of ROS and mitochondrial membrane potential (MMP). After removing the culture medium from each group following intervention, cells were incubated with 10 µM DCFH‐DA or 20 µM TMRE working solution for 20 min at 37°C. The staining solution was then removed, and the cells were washed three times with PBS. Subsequently, cells were detached using a non‐enzymatic cell dissociation buffer and harvested for flow cytometric analysis, cells were resuspended in PBS to generate a single‐cell suspension. ROS levels and MMP were measured using the FITC and PE channels of a flow cytometer (Beckman, China), respectively. Data were collected and analysed using appropriate statistical methods.

Macrophages were incubated with primary antibodies against F4/80 (biolegend, 123114) and α‐SMA (HUABIO, ET1607‐53) at 4°C for 30 min, followed by incubation with goat anti‐mouse IgG‐488 (Abcam, ab150113) at room temperature for 1 h. Dual‐positive cells were then analysed using PE/CY7 and FITC channels on a flow cytometer (CytoFLUX0206A, LEGEND TECK), and F4/80⁺α‐SMA⁺ cells were identified as MMT cells.

2.19. WGA Staining

Paraffin‐embedded tissue sections were deparaffinized and rehydrated, then incubated with 10 ng/mL WGA (L4895, Sigma) staining solution at room temperature for 15 min. Excess dye was removed by washing, and sections were mounted with a DAPI‐containing anti‐fade mounting medium. Images were captured and analysed using a fluorescence microscope (MSHOT, China).

For cell monolayers, coverslips were fixed with 4% paraformaldehyde and washed. Subsequent staining procedures were performed as described for tissue sections. Images were captured and analysed using a confocal microscope (Zeiss LSM980).

2.20. MtDNA Detection

Genomic DNA was extracted from tissues and cells according to the manufacturer's instructions (TIANGEN, DP304). Briefly, cells were lysed by sequential addition of 200 µL Buffer GA, 20 µL Proteinase K, and 200 µL Buffer GB, followed by incubation at 70°C for 10 min. After adding 200 µL absolute ethanol, the lysate was transferred to a spin column and centrifuged at 12,000 rpm for 30 s. The column was washed with 500 µL Buffer GD and then twice with 600 µL Wash Buffer PW. After air‐drying, DNA was eluted with 200 µL Buffer TE by centrifugation at 12,000 rpm for 2 min. The eluate was used as the template for mtDNA detection. mtDNA levels were measured by qPCR using the following gene‐specific primers:

Primer sequences for mtDNA:

Forward 5’‐AATCTACCATCCTCCGTGAAACC‐3’;

Reverse: 5’‐TCAGTTTAGCTACCCCCAAGTTTAA‐3’.

Tert control:

Forward 5’‐CTAGCTCATGTGTCAAGACCCTCTT‐3’;

Reverse 5’‐GCCAGCACGTTTCTCTCGTT‐3’.

2.21. Western Blot

Cells or tissues were lysed in RIPA buffer. Lysates were centrifuged at 12,000 × g for 5 min at 4°C, and supernatants were mixed with 5× loading buffer (1:4 ratio) and boiled at 95°C for 5 min. Proteins were resolved by SDS‐PAGE, transferred to PVDF membranes, and blocked with 5% non‐fat milk for 1 h. Membranes were incubated with primary antibodies overnight at 4°C, followed by HRP‐conjugated secondary antibodies for 1 h. Signals were detected using ECL substrate (ECL, SQ201, Epizyme) and imaged. Band intensities were quantified using ImageJ. The specific antibodies used in this study included:

Fibronection (HUABIO, ET1702‐25); Integrin α5 (HUABIO, ET1701‐58); Src (Abways, CY5459); α‐SMA (HUABIO, ET1607‐53); OXPHOS complex (Proteintech, PK30006); PGC1α (Abways, CY6630); TLR4 (CST, 38519); RAGE (Proteintech, 16346‐1‐AP); p‐P38 (Proteintech, 28796‐1‐AP); p‐P65 (CST, 3033); p‐ERK (Proteintech, 28733‐1‐AP); p‐JNK (Proteintech, 80024‐1‐RR); smad3 (Proteintech, 80427‐2‐RR); β‐Tubulin (Proteintech, 10094‐1‐AP); GAPDH (Proteintech, 60004‐1‐Ig).

2.22. Statistical Analysis

Data with a normal distribution are presented as mean ± standard deviation (SD), while non‐normally distributed data are presented as median with interquartile range. Comparisons between two groups of normally distributed data were performed using Student's t‐test, and comparisons among three or more groups were performed using one‐way analysis of variance (ANOVA). For non‐normally distributed data, comparisons between two groups were conducted using the Mann–Whitney U test, and comparisons among three or more groups were conducted using the Kruskal–Wallis H test. Statistical analyses were performed using SPSS software, and graphical representations were prepared with GraphPad Prism.

3. Results

3.1. S100A9 Promotes MMT and Exacerbates the Progression of Cardiac Fibrosis

Based on our previous findings, knockout of S100a9 effectively alleviates acute inflammatory injury following MIR and reduces cardiac fibrosis during the reparative phase (Shen et al. 2024) (Figure 1A–D). Mass cytometry analyses revealed a prominent MMT after MIR (Figure 1E–F). To determine the functional role of S100a9 in this process, we generated global S100a9 knockout mice (S100a9 ^KO) as well as macrophage‐specific S100a9 gene knockout mice (S100a9 ^MKO). Neither genetic manipulation resulted in overt abnormalities or impaired baseline physiological function. Following MIR surgery, both mouse models exhibited markedly reduced cardiac fibrosis compared with their WT counterparts (Figure 1C,D and S1A).

FIGURE 1.

S100A9 promotes macrophage‐to‐myofibroblast transition (MMT) after MI and exacerbates myocardial fibrosis.

(A–B) UMAP plots showing the single‐cell transcriptomic landscape of cardiac macrophages from Sham mice and mice at Day 7 post–MIR. n = 9. (C–D) Masson staining showing cardiac fibrosis at Day 7 after MIR, with corresponding quantitative analysis. n = 5. (E–F) t‐SNE plots showing MMT levels in cardiac macrophages from Sham and Day‐7 MIR mice. n = 9. G. Detection of α‐SMA and endogenously red‐fluorescent macrophages in heart tissue by immunofluorescence. n = 3. (H–I) Immunofluorescence staining of α‐SMA and CD68 in heart tissue, with relative quantification. n = 5. (J–K) WGA staining showing cardiomyocyte cross‐sectional area with normalized quantification. n = 5. (L) Heart weight/body weight ratios and correlative analyses in each group. n = 5. (M–O) Echocardiographic assessment of cardiac function, including LVPWs, LVEF, and FS, with quantitative analysis. n = 5. (P–Q) Western blot analysis of protein expression in mouse heart tissue with normalized quantification. n = 5. ns, not significant; *, p < 0.05; **, p < 0.01.

To track macrophage fate in vivo, KIKGR‐Lyz2 ^Cre reporter mice were generated. BMDMs from these mice were isolated and photoconverted from green to permanent red fluorescence. Following MIR, these labelled macrophages were adoptively transferred into recipient mice. On Day 7 post‐MIR, exogenous red fluorescent macrophages co‐expressing α‐SMA were observed in the cardiac tissue, demonstrating the occurrence of MMT (Figure 1G). Notably, the number of MMT‐derived myofibroblasts in hearts with MIR was significantly decreased in both S100a9 ^KO and S100a9 ^MKO mice (Figure 1H–I). Wheat germ agglutinin (WGA) staining further demonstrated that S100a9 deficiency protected against MIR‐induced cardiomyocytes hypertrophy (Figure 1J–K). Compared to WT mice, S100a9‐gene knockout mice exhibited a significantly reduced heart weight‐to‐body weight ratio following MIR (Figure 1L). Echocardiographic assessment showed that S100a9 deletion—either global or macrophage‐specific—did not alter baseline cardiac function (Figure 1M–O). Although WT mice showed substantial deterioration in cardiac function after MIR, both S100a9‐gene knockout models displayed significant preservation of cardiac performance and left ventricular wall thickness (Figure 1M‐O). Previous studies suggest that migrasomes play an important role in MMT regulation (Peng et al. 2025). We therefore examined migrasome‐related markers and found that MIR markedly increased their expression in the heart. Importantly, S100a9 deficiency significantly suppressed the upregulation of these migrasome‐associated proteins (Figure 1P–Q).

To investigate the role of S100A9 in MMT‐mediated cardiac fibrosis, MI models were established in S100a9‐deficient mice. Both global and macrophage‐specific S100a9 knockout reduced post‐MI cardiac fibrosis and MMT levels (Figure S1B‐E). Compared to MI‐operated WT mice, S100a9 knockout mice showed improved cardiac function and reduced cardiomyocyte hypertrophy (Figure S1F‐I). S100a9 deficiency also downregulated fibrosis markers and migrasome markers in post‐MI hearts (Figure S1J‐K). These findings indicate that S100a9 knockout attenuates MMT and cardiac fibrosis, associated with changes in migrasomes, irrespective of reperfusion.

3.2. Recombinant S100A9 Protein Enhances MMT and Exacerbates Cardiac Fibrosis

To further verify the role of S100A9 in aggravating post‐MIR cardiac fibrosis, we administered recombinant S100A9 protein (rS100A9) to MIR mice. To determine the optimal dosing frequency for rS100A9, we measured serum S100A9 levels at various time points after a single infusion. The levels decreased to their lowest point at 48 h post‐infusion (Figure S2A). Accordingly, a 48 h dosing interval was chosen to sustain serum S100A9 levels and maintain its biological effects. Histological staining showed that rS100A9 supplementation markedly increased fibrotic area in both WT mice and S100a9‐gene knockout mice subjected to MIR (Figures 2A,B and S2B). WGA staining confirmed that S100a9 knockout prevented cardiomyocyte hypertrophy following MIR, whereas rS100A9 infusion completely abolished this benefit and further worsened post‐MIR cardiomyocyte hypertrophy in WT mice (Figure 2C,D). Exogenous rS100A9 reversed the S100a9 knockout‐mediated improvement in cardiac function, leading to exacerbated cardiac dysfunction and reduced left ventricular wall thickness following MIR (Figure 2E–H). Consistently, immunofluorescence analysis and western blot confirmed that administration of rS100A9 further upregulated the number of macrophages undergoing transition toward a myofibroblast phenotype and migrasome‐associated molecules within MIR cardiac tissue (Figure 2I–L).

FIGURE 2.

Recombinant S100A9 protein promotes MMT and exacerbates myocardial fibrosis.

WT and S100a9 ^MKO mice received intravenous rS100A9 after MIR and were sacrificed on Day 7 post‐MIR. (A–B) Masson staining showing cardiac fibrosis in mice, with corresponding quantitative analysis. n = 5. (C–D) WGA staining showing cardiomyocyte cross‐sectional area with normalized quantification. n = 5. (E) Heart weight/body weight ratios and quantitative analyses in each experimental group. n = 5. (F–H) Echocardiographic assessment of cardiac function, including LVPWs, LVEF, and FS, with quantitative analysis. n = 5. (I–J) Immunofluorescence staining of α‐SMA and CD68 in heart tissue, with relative quantification. n = 5. (K–L) Western blot analysis of protein expression in mouse heart tissue with normalized quantification. n = 5. WT mice received intravenous injections of S100A9 NTAB or isotype control after MIR and were sacrificed on Day 7 post‐MIR. (M–N) Masson staining showing cardiac fibrosis in mice, with corresponding quantitative analysis. n = 5. (O–P) WGA staining showing cardiomyocyte cross‐sectional area with normalized quantification. n = 5. (Q) Heart weight/body weight ratios and correlative analyses in each group. n = 5. (R–T) Echocardiographic assessment of cardiac function, including LVPWs, LVEF, and FS, with quantitative analysis. n = 5. (U–V) Immunofluorescence staining of α‐SMA and CD68 in heart tissue, with relative quantification. n = 5. (W–X) Western blot analysis of protein expression in mouse heart tissue with normalized quantification. n = 5. ns, not significant; *, p < 0.05; **, p < 0.01.

In the MI model, exogenous rS100A9 reversed the cardioprotective effects of S100a9 deficiency, leading to increased myocardial fibrosis and cardiomyocyte hypertrophy, as well as reduced cardiac function (Figure S2C–H). Additionally, rS100A9 supplementation promoted MMT and upregulated migrasome‐related molecular expression (Figure S2I–L).

3.3. Neutralization of S100A9 Mitigates MMT and Protects Against Cardiac Fibrosis

To determine whether blockade of S100A9 confers therapeutic benefit after MIR, WT mice were treated with an S100A9‐neutralizing antibody (NTAB) during MIR induction. Isotype control had no significant effect on post‐MIR fibrotic area, cardiomyocyte surface area, or cardiac dysfunction (Figures 2M–T and S3A). In contrast, S100A9 neutralization markedly reduced fibrosis following MIR (Figure 2M,N). Combined WGA staining and echocardiographic assessment demonstrated that S100A9 neutralization preserved cardiac function and alleviated MIR‐induced cardiomyocyte hypertrophy (Figure 2O–T). Immunofluorescence demonstrated that S100A9‐NTAB delivery significantly decreased the number of macrophages undergoing transition toward a myofibroblast phenotype (Figure 2U,V). Immunoblotting analyses confirmed that the NTAB suppressed fibrogenic markers, decreased expression of migrasome‐associated molecules, and significantly reduced the number of macrophages undergoing transition into myofibroblasts (Figure 2W,X). In the MI model, neutralization of S100A9 similarly led to a marked reduction in cardiac fibrosis area and cardiomyocyte surface area following MI. Additionally, S100A9 neutralization attenuated the extent of MMT, protected cardiac function, and downregulated the expression of migrasome‐related markers (Figure S3B–K).

3.4. S100A9 Promotes Myofibroblast Migration, Activation, and Migrasome Release

To investigate whether S100A9 directly modulates fibroblast behaviour, primary cardiac fibroblasts were isolated from mice and stimulated with rS100A9. Crystal violet staining revealed that rS100A9 markedly enhanced fibroblast migration (Figure 3A,B). Both immunofluorescence staining and immunoblotting demonstrated that rS100A9 stimulation activated fibroblasts, as indicated by increased expression of Col1α1 and multiple migrasome‐associated molecules (Figure 3C–G). Migrasomes have been widely recognized as carriers that export damaged mitochondria to the extracellular space. Therefore, we next assessed mitochondrial function in fibroblasts after rS100A9 stimulation. qPCR analysis showed decreased mitochondrial mtDNA levels, while immunoblotting revealed impaired oxidative phosphorylation (OXPHOS) complex I/IV integrity in rS100A9‐treated fibroblasts (Figure 3H–K). Consistent with mitochondrial dysfunction, rS100A9 increased ROS production and reduced MMP (Figure 3L–O). rS100A9 stimulation had no significant effect on ECAR in cardiac fibroblasts but significantly reduced OCR and ATP production levels (Figure 3P–R). To exclude direct mitochondrial toxicity of rS100A9, we performed immunofluorescence co‐staining and found no cytoplasmic colocalization of rS100A9 with mitochondria at various time points after stimulation, indicating that S100A9 does not impair mitochondrial function through direct toxicity (Figure 3S). WGA staining and transmission electron microscopy (TEM) further confirmed that rS100A9 promoted migrasome release from activated myofibroblasts (Figures 3T–U and S4A). Analysis of Single‐nucleus RNA‐sequencing datasets from mouse hearts following MIR revealed that activated fibroblasts highly expressed Tspan4, further corroborating our in vitro findings (Figure S4B–D). Tspan4 is a key gene involved in migrasome formation. Knockdown of Tspan4 significantly reduced rS100A9‐stimulated migrasome release and decreased migrasome marker expression, while having no significant effect on exosome markers (Figure 3V–Y). Collectively, these results demonstrate that S100A9 promotes mitochondrial dysfunction and migrasome release in cardiac fibroblasts.

FIGURE 3.

S100A9 promotes fibroblast migration, activation, and migrasome release.

(A–B) Crystal violet staining showing fibroblast migration, with quantitative analysis. n = 5. (C–E) Co‐immunofluorescence staining of α‐SMA and Col1α1 to assess fibroblast activation, with corresponding fluorescence quantification. n = 5. (F–G) Western blot analysis of Integrin α5 expression in fibroblasts following rS100A9 stimulation, with normalized quantification. n = 5. (H) qPCR measurement of mtDNA levels in cardiac fibroblasts after rS100A9 stimulation. n = 5. (I–K) Western blot analysis of mitochondrial oxidative phosphorylation (OXPHOS) complex proteins, with normalized quantification. n = 5. (L–M) Flow cytometry analysis of ROS levels in cardiac fibroblasts following rS100A9 stimulation, with quantitative assessment. n = 5. (N–O) Flow cytometry analysis of MMP in rS100A9‐treated fibroblasts, with quantitative assessment. n = 5. (P–R) ECAR, OCR, and ATP levels in rS100A9‐stimulated fibroblasts. n = 5. (S) Fluorescence images showing α‐SMA and S100A9 expression in fibroblasts at different time points following rS100A9 stimulation. n = 5. (T–U) WGA staining showing migrasome level and corresponding quantification. n = 5. (V–W) Levels of migrasome, exosome, and apoptotic body markers in Tspan4‐knockdown fibroblasts following rS100A9 stimulation. n = 5. (X–Y) WGA staining showing migrasome levels in Tspan4‐knockdown fibroblasts following rS100A9 stimulation. n = 5. ns, not significant; *, p < 0.05; **, p < 0.01.

3.5. S100a9 Induces Mitochondrial Dysfunction and Migrasome Release in Myofibroblasts Through the TLR4/ERK/PGC1α Pathway

PGC‐1α is a key regulator of mitochondrial biogenesis and function. Stimulation of myofibroblasts with rS100A9 markedly reduced PGC‐1α expression and downstream pathway of S100A9 (Figure 4A,B). In a fibroblast–macrophage coculture system, overexpression of PGC‐1α in myofibroblasts substantially diminished the ability of rS100A9‐stimulated fibroblasts to induce fibrogenic gene expression in macrophages and significantly reduced the number of macrophages undergoing MMT (Figure 4C‐F).

FIGURE 4.

S100A9 induces mitochondrial dysfunction and migrasome release in fibroblasts via the TLR4/PGC‐1α pathway.

(A–B) Western blot analysis of protein expression in mouse cardiac fibroblasts stimulated with rS100A9, with normalized quantification. n = 5. (C–D) Western blot analysis of protein expression in cardiac fibroblasts overexpressing PGC‐1α and stimulated with rS100A9, with normalized quantification. n = 5. (E–F) Flow cytometry analysis showing macrophages co‐expressing α‐SMA and F4/80, with quantitative assessment. n = 5. (G–H) Western blot analysis of PGC‐1α expression in mouse cardiac fibroblasts, with normalized quantification. n = 5. (I–J) Flow cytometry measurement of ROS levels in cardiac fibroblasts after rS100A9 stimulation, with quantitative analysis. n = 5. (K–L) Flow cytometry analysis of MMP in cardiac fibroblasts stimulated with rS100A9 and pharmacological inhibitor, with quantitative assessment. n = 5. (M–O) ECAR, OCR, and ATP levels in rS100A9‐stimulated fibroblasts following treatment with a TLR4 inhibitor or a RAGE inhibitor. n = 5. (P–Q) WGA staining evaluating migrasome release level in cardiac fibroblasts, with quantitative analysis. n = 5. (R–S) Western blot analysis of Tlr4‐gene knock‐down in mouse cardiac fibroblasts treat with rS100A9, with normalized quantification. n = 5. (T–U) WGA staining assessing migrasome release level in cardiac fibroblasts with Tlr4‐gene knock‐down and rS100A9 treatment. n = 5. (V–W) WGA staining assessing migrasome release level in cardiac fibroblasts with Erk1/2‐gene knock‐down and rS100A9 treatment. n = 5. (X–Y) Western blot analysis of Erk1/2‐gene knock‐down in mouse cardiac fibroblasts treat with rS100A9, with normalized quantification. n = 5. ns, not significant; *, p < 0.05; **, p < 0.01.

To determine how S100A9 regulates PGC‐1α signalling, we inhibited candidate S100A9 receptors and assessed mitochondrial function. Inhibition of TLR4 restored PGC‐1α expression in rS100A9‐stimulated myofibroblasts, increased MMP, OCR, and ATP, and reduced ROS levels, without influence on ECAR (Figure 4G–O). WGA staining further demonstrated that TLR4 inhibition markedly suppressed migrasome release (Figure 4P,Q). Given that TLR4 is a well‐established receptor for DAMPs, we stimulated fibroblasts with rS100A9 or LPS. Both rS100A9 and LPS upregulated migrasome‐related molecule expression and promoted migrasome release (Figure S4E–H). However, LPS‐induced migrasome release was markedly lower than that induced by rS100A9 (Figure S4E–H). These results demonstrating that S100A9 is a strong inducer of migrasome release from fibroblasts.

We next examined downstream pathways of TLR4. Pharmacological inhibition of p‐JNK, p‐p38, or p‐p65 had minimal effects on PGC‐1α levels compared with vehicle control. In contrast, inhibition of p‐ERK significantly increased PGC‐1α expression and also blocked S100a9‐induced migrasome release, as confirmed by WGA staining (Figure S4I‐K). To dissect the signalling pathways involved, we knocked down Tlr4 or Erk1/2 in cardiac fibroblasts using specific siRNAs. Tlr4 knockdown reduced rS100A9‐induced upregulation of p‐ERK1/2, PGC1α, and migrasome‐related molecules, as well as migrasome release (Figure 4R–U). Similarly, Erk1/2 knockdown produced comparable effects (Figure 4V–Y). Given that S100A9 primarily signals through TLR4 to regulate inflammation, we assessed inflammatory responses in rS100A9‐stimulated fibroblasts. rS100A9 markedly increased p‐NFκB and TNFα expression (Figure S4L‐M). However, knockdown of NFκB or TNFα did not significantly affect migrasome‐related molecule expression or migrasome release (Figure S4N–Q), indicating that the inflammatory pathway is not directly involved in S100A9‐induced migrasome release.

Collectively, these findings indicate that S100A9 suppresses PGC‐1α expression through the TLR4–p‐ERK signalling axis, leading to mitochondrial dysfunction and subsequent migrasome release in myofibroblasts.

3.6. S100a9 Promotes the Release of Migrasomes From Fibroblasts, Which in Turn Enhances MMT

To determine whether rS100A9‐activated fibroblasts influence macrophage fibrotic‐phenotype, we next established a fibroblast‐macrophage coculture system (Figure 5A). Regardless of whether rS100A9 stimulation was applied or not, the fibrotic‐phenotype of macrophages remained unaffected (Figure 5B–E). In the absence of rS100A9 stimulation, cardiac fibroblasts had minimal impact on macrophage activation or MMT (Figure 5B–E). However, when fibroblasts were stimulated with rS100A9, they robustly promoted the expression of fibrogenic molecules in macrophages (Figure 5B–E). Flow cytometry and Immunoblotting further confirmed that rS100A9‐treated fibroblasts significantly increased the number of macrophages undergoing MMT (Figure 5B‐E). SAR407899 is a migrasome‐formation inhibitor that can effectively block rS100A9‐mediated migrasome release from fibroblasts (Figure 5F‐G). To determine whether migrasomes mediate this fibroblast–macrophage crosstalk, we used Transwell coculture systems with different pore sizes. Fibroblasts were plated in the upper chamber and stimulated with rS100A9 in the presence or absence of the migrasome inhibitor SAR407899 (Figure 5H). In the 3‐µm pore system, rS100A9‐stimulated fibroblasts strongly induced MMT; however, this effect was abolished by SAR407899, as evidenced by reduced fibrogenic marker expression and decreased MMT levels on flow cytometry (Figure 5I–L). Considering the size of migrasomes, we further employed a 0.4‐µm pore Transwell, which effectively blocked migrasome transfer. In this setting, rS100a9‐treated fibroblasts no longer promoted macrophage MMT, confirming that the effect was migrasome‐dependent (Figure 5I–L). To determine whether fibroblast‐derived migrasomes mediate MMT, fibroblasts were transfected with si‐Tspan4 to knock down TSPAN4 expression, stimulated with rS100A9, and then cocultured with macrophages. Under these conditions, Tspan4 knockdown significantly reduced the extent of MMT compared to NC (Figure 5M–P).

FIGURE 5.

S100A9‐induced migrasome release from fibroblasts promotes MMT.

(A) Schematic diagram of the co‐culture system of macrophages and cardiac fibroblasts. n = 5. (B–C) Flow cytometry analysis showing macrophages co‐expressing α‐SMA and F4/80, with quantitative assessment. n = 5. (D–E) Western blot analysis of protein expression in macrophage co‐cultured with cardiac fibroblasts treated with rS100A9 or not, with normalized quantification. n = 5. (F–G) WGA Staining showing migrasome level in cardiac fibroblasts treated with rS100A9 and SAR407899, with corresponding quantitative analysis. n = 5. (H) Schematic diagram of an alternative co‐culture experiment of macrophages and cardiac fibroblasts. n = 5. (I–J) Flow cytometry analysis of macrophages co‐expressing α‐SMA and F4/80 which co‐cultured with fibroblasts in each group, with quantitative assessment. n = 5. (K–L) Western blot analysis of protein expression in cardiac fibroblasts, with normalized quantification. n = 5. (M–N) Western blot analysis of α‐SMA and Fibronection expression in macrophage co‐cultured with cardiac fibroblasts with Tspan4‐gene knock‐down and rS100A9 treatment. n = 5. (O–P) Flow cytometry analysis of macrophages co‐expressing α‐SMA and F4/80 which co‐cultured with cardiac fibroblasts with Tspan4‐gene knock‐down. n = 5. ns, not significant; *, p < 0.05; **, p < 0.01.

Taken together, these findings demonstrate that S100A9 induces mitochondrial dysfunction in fibroblasts, leading to enhanced migrasome release, which subsequently drives macrophage‐to‐myofibroblast transition.

3.7. Fibroblast‐Derived Migrasomes Induce MMT Through Integrin α5/Src Signalling

Integrin α5 is a well‐recognized marker of migrasomes and has been shown to promote tumour invasion through Src activation. Src itself is a critical regulatory node in MMT. To determine whether fibroblast‐derived migrasomes drive MMT via this pathway, migrasomes were isolated and purified from rS100A9‐stimulated fibroblasts and subsequently applied to macrophages in the presence of an integrin α5 inhibitor or Src‐overexpression plasmid. The purity of the isolated migrasomes was assessed by analysing the final samples obtained through crude centrifugation. The results confirmed efficient migrasome enrichment and excluded interference from exosomes and apoptotic bodies (Figure S5A). Immunoblotting revealed that migrasome treatment markedly increased macrophage expression of α‐SMA, Src, and p‐smad3, indicative of MMT activation (Figure S5B‐C). Inhibition of integrin α5 significantly attenuated these effects, whereas addition of a Src‐overexpression plasmid reversed the protective effects of integrin α5 blockade (Figure S5B‐C). Flow cytometry quantification of MMT further supported these results, showing a parallel reduction with integrin α5 inhibition and restoration with Src overexpression (Figure S5D‐E). To exclude the potential involvement of exosomes in fibroblast‐driven MMT, we treated rS100A9‐stimulated fibroblasts with an exosome clearance reagent. Removal of exosomes did not significantly impact migrasome release from fibroblasts (Figure S5F‐I). When these fibroblasts were cocultured with macrophages, the extent of MMT remained unchanged (Figure S5J‐L), indicating that exosomes are not required for this process.

To evaluate the functional relevance of this pathway in vivo, we induced MIR and MI in mice and administered an integrin α5 inhibitor. Compared with vehicle‐treated mice, integrin α5 inhibition markedly reduced cardiac fibrosis area (Figures 6A,B and S6A–C). WGA staining and echocardiography demonstrated that integrin α5 blockade preserved cardiac function and alleviated cardiomyocytes hypertrophy following MIR and MI (Figures 6C–H and S6D‐G). Consistent with the in vitro findings, immunoblotting and immunofluorescence analyses showed that integrin α5 inhibition decreased the expression of migrasome markers and MMT‐related molecules in MIR and MI cardiac tissue and significantly reduced the number of macrophages undergoing MMT (Figure 6I–L and S6H–K). To evaluate the therapeutic potential of integrin α5 blockade, MIR mice were treated with an integrin α5 inhibitor. On Day 28 after surgery, cardiac fibrosis area was significantly reduced, cardiac function was preserved, and the expression levels of fibrosis‐related markers were decreased in treated mice compared to vehicle (Figure 6M–U). Integrin α5 regulates migrasome formation and release via fibronectin (Fn) binding, as confirmed by colocalization of migrasomes with Fn (Figure S6L). Addition of RGD peptide to cocultures of macrophages and rS100A9‐stimulated fibroblasts reduced MMT‐related signals (Figure S6M–N). Knockdown of integrin α5 in BMDMs attenuated fibroblast‐derived migrasome‐induced MMT compared to control (Figure 6V–X).

FIGURE 6.

Fibroblast‐derived migrasomes promote MMT through the integrin α5/Src signaling pathway.

(A–B) Masson staining showing cardiac fibrosis in mice at Day 7 post‐MIR, with corresponding quantitative analysis. n = 5. (C–D) WGA staining showing cardiomyocyte cross‐sectional area with normalized quantification. n = 5. (E) Heart weight/body weight ratios and correlative analyses in each experimental group. n = 5. (F–H) Echocardiographic assessment of cardiac function, including LVPWs, LVEF, and FS, with quantitative analysis. n = 5. (I–J) Immunofluorescence staining of α‐SMA and CD68 in heart tissue, with relative quantification. n = 5. (K–L) Western blot analysis of protein expression in mouse heart tissue with normalized quantification. n = 5. (M) Survival rates of mice following MIR and treatment with integrin α5 inhibitor over a 28‐day period. n = 15. (N–O) Masson staining showing cardiac fibrosis in mice at day 28 post‐MIR, with corresponding quantitative analysis. n = 5. (P) Heart weight/body weight ratios and correlative analyses in each experimental group. n = 5. (Q–S) Echocardiographic assessment of cardiac function, including LVPWs, LVEF, and FS, with quantitative analysis. n = 5. (T–U) Western blot analysis of protein expression in mouse heart tissue with normalized quantification. n = 5. (V–W) MMT signaling expression in macrophages upon migrasome stimulation after integrin α5 knockdown. n = 5. (X) Flow cytometry analysis of macrophages knockdown integrin α5 co‐expressing α‐SMA and F4/80 which co‐cultured with migrasome. n = 5. ns, not significant; *, p < 0.05; **, p < 0.01.

To investigate intercellular crosstalk in cardiomyocyte hypertrophy, cardiomyocytes were cocultured with S100A9, fibroblasts, or S100A9‐stimulated fibroblasts. S100A9 or fibroblasts alone did not affect cardiomyocyte surface area or BNP expression. However, S100A9‐stimulated fibroblasts significantly increased both parameters (Figure S6O‐R).

Together, these results identify integrin α5/Src signalling as a key mechanism by which fibroblast‐derived migrasomes induce MMT and exacerbate post‐MI cardiac fibrosis.

3.8. Pharmacological Inhibition of S100a9 Attenuates MMT and Protects Against Cardiac Fibrosis

To determine whether pharmacological targeting of S100A9 confers therapeutic benefits after MIR, we administered an S100A9 inhibitor to mice at 6 and 24 h following MIR. Analysis of cardiac tissues revealed that post‐surgical S100A9 inhibition significantly reduced the area of cardiac fibrosis and decreased cardiomyocyte hypertrophy, as evidenced by reduced cell surface area (Figure 7A–E). Additionally, treated mice exhibited improved cardiac wall thickness and preserved cardiac function. Consistent with these findings, S100A9 inhibitor treatment also effectively suppressed MMT levels and MMT signalling (Figure 7F–M).

FIGURE 7.

Pharmacological inhibition of S100a9 attenuates MMT and protects against cardiac fibrosis.

(A) Survival rates of mice within 7‐days following MIR, with Tasquinimod administered at different post‐operative time points. n = 13. (B–C) Masson staining showing cardiac fibrosis in mice at Day 7 post‐MIR and treated with tasquinimod, with corresponding quantitative analysis. n = 5. (D–E) WGA staining showing cardiomyocyte cross‐sectional area with normalized quantification. n = 5. (F) Heart weight/body weight ratios and correlative analyses in each experimental group. n = 5. (G–I) Echocardiographic assessment of cardiac function, including LVPWs, LVEF, and FS, with quantitative analysis. n = 5. (J–K) Immunofluorescence staining of α‐SMA and CD68 in heart tissue, with relative quantification. n = 5. (L–M) Western blot analysis of protein expression in mouse heart tissue with normalized quantification. n = 5. (N) Serum S100A9 levels in patients with coronary reperfusion after myocardial infarction compared with those in healthy subjects. n = 8. (O–P) Monocytes isolated from the peripheral blood of the subjects were co‐cultured with S100A9‐stimulated human fibroblasts, and the expression levels of markers related to MMT in monocytes were subsequently assessed. n = 8. ns, not significant; *, p < 0.05; **, p < 0.01.

We collected peripheral blood from patients with myocardial infarction who underwent reperfusion within 24 h. Compared with healthy controls, serum levels of S100A9 were significantly elevated in these patients (Figure 7N). We further isolated peripheral blood monocytes from patients with MIR and co‐cultured them with recombinant human S100A9 (rhS100A9) and/or human fibroblasts, followed by assessment of molecular changes. No significant effects on monocyte molecular expression were observed upon treatment with rhS100A9 alone or co‐culture with fibroblasts (Figure 7O–P). However, when monocytes were co‐cultured with rhS100A9‐stimulated fibroblasts, the expression levels of MMT‐related molecules were markedly increased in monocytes (Figure 7O–P).

4. Discussion

Cardiac fibrosis is a common form of structural remodelling after MI and is essential for maintaining ventricular integrity (Lunde et al. 2024). However, excessive fibrosis predisposes the heart to stiffening, maladaptive remodelling, and heart failure, underscoring the need to delineate the mechanisms driving post‐MI fibrogenesis. MMT has recently emerged as a novel cellular source of myofibroblasts, yet its regulatory signals and pathological relevance in cardiac fibrosis remain poorly understood. In this study, we identify a previously unrecognized mechanism in which macrophage‐derived S100A9 amplifies cardiac fibrosis by activating fibroblast TLR4/p‐ERK/PGC‐1α signalling, thereby inducing fibroblast activation, migration, and migrasome release. These fibroblast‐derived migrasomes are subsequently engulfed by macrophages, where integrin α5/Src signalling facilitates MMT and ultimately accelerates fibrotic remodelling after MI.

Macrophages constitute a major innate immune population that rapidly infiltrates the ischemic myocardium following MI. Building on our previous single‐cell RNA‐sequencing work showing that myeloid macrophages release S100A9 to exacerbate myocardial injury and fibrosis, we generated macrophage‐specific S100a9‐deficient mice and demonstrated that loss of S100a9 significantly attenuated post‐MI fibrosis (Shen et al. 2024). Single‐cell transcriptomics and mass cytometry at Day 7 post‐MI revealed a distinct macrophage subset expressing high levels of α‐SMA, consistent with MMT. Notably, macrophage‐specific deletion of S100a9 markedly reduced α‐SMA⁺ macrophages within infarcted hearts, confirming that S100a9 is a potent upstream driver of MMT and fibrosis progression.

Cardiac fibroblasts represent the primary source of myofibroblasts after myocardial injury. Upon sensing damage‐associated signals, fibroblasts undergo phenotypic conversion into highly synthetic and contractile myofibroblasts, accompanied by profound metabolic remodelling (Humeres and Frangogiannis 2019). Mitochondria are central to these changes; mitochondrial ROS generated through the electron transport chain promote myofibroblast differentiation, while Drp1‐mediated mitochondrial fission drives mtROS accumulation and upregulation of α‐SMA and Col1a1 (Gibb et al. 2020; Wang et al. 2020). In this study, we show that S100A9 robustly activates fibroblasts via TLR4/ERK signalling, triggering increased mitochondrial ROS, reduced membrane potential, and impaired oxidative phosphorylation complexes. Importantly, S100A9 also facilitated the release of migrasomes from activated myofibroblasts.

Migrasomes are newly recognized extracellular vesicular structures derived from retraction fibres and serve as mediators of intercellular communication. They contain damaged mitochondria and misfolded proteins and are increasingly viewed as an important mechanism for maintaining cellular homeostasis (Jiao et al. 2021). Previous work in pulmonary fibrosis has shown that LPS promotes migrasome release from myofibroblasts, which in turn enhances MMT and accelerates fibrogenesis (Peng et al. 2025). Nevertheless, the mechanisms through which migrasomes reprogram macrophages have remained largely unknown. Integrin α5 is a defining structural component of migrasomes and is essential for their generation and function. In the heart, integrin α5 is upregulated in both the infarct and peri‐infarct regions following MI (Nawata et al. 1999). Although macrophage‐derived integrin α5 exerts cardioprotective roles, fibroblast‐derived integrin α5 has been implicated in pathological fibrosis across multiple tissues by promoting fibroblast–immune cell interactions (Li et al. 2023; Lu et al. 2023; Xiao et al. 2024). Importantly, integrin α5 can activate Src to drive profibrotic signalling—an effect that our mechanistic studies now extend to MMT regulation in the post‐MIR heart.

Inhibition of integrin α5 in vivo significantly reduced cardiac fibrosis, improved ventricular remodelling, and preserved cardiac function, providing compelling evidence that targeting fibroblast–macrophage communication through migrasomes offers therapeutic potential. Nevertheless, our study also has limitations. First, validation using human myocardial samples is lacking. Second, large‐scale clinical studies are needed to determine whether integrin α5 inhibition confers beneficial effects in patients with MI and adverse ventricular remodelling.

5. Conclusion

This study demonstrates that infiltrating myeloid macrophages release S100A9 after myocardial ischemia, which activates TLR4/p‐ERK signalling in cardiac fibroblasts to enhance mitochondrial mtROS production and impair oxidative respiratory chain function. These metabolic alterations drive the differentiation of fibroblasts into migratory myofibroblasts and promote the release of migrasomes. Once released into the extracellular space, migrasomes act on macrophages through integrin‐α5/Src signalling to induce their transdifferentiation into myofibroblast‐like cells, thereby exacerbating cardiac fibrosis. Together, our findings identify a previously unrecognized S100A9–migrasome–integrin‐α5 axis that links macrophage activation with fibroblast reprogramming and contributes to post‐ ischemia fibrotic remodelling.

Author Contributions

Yiting Zhao: methodology, data curation, writing – original draft, investigation. Yaqin Zhang: methodology, software, formal analysis, validation. Qiang Yuan: methodology, investigation. XiaoFeng Fang: validation, visualization. Jianlong Sheng: conceptualization, data curation, supervision. Yuhang Deng: conceptualization, investigation, funding acquisition. Shichun Shen: conceptualization, investigation, funding acquisition, writing – review and editing, project administration, resources, visualization.

Conflicts of Interest

The authors report no conflict of interest.

Supporting information

Supporting Information: Figure S1. S100A9 promotes macrophage‐to‐myofibroblast transition (MMT) after MI and exacerbates myocardial fibrosis.

Supporting Information: Figure S2. Recombinant S100A9 protein promotes MMT and exacerbates myocardial fibrosis

Supporting Information: Figure S3. S100A9 neutralizing antibody attenuates MMT and protects against myocardial fibrosis

Supporting Information: Figure S4. S100a9 promotes migrasome release from fibroblasts via the TLR4/ERK1/2 pathway

Supporting Information: Figure S5. Blockade of integrin α5 reduces post‐MI cardiac fibrosis levels.

Supporting Information: Figure S6. Fibroblast‐derived migrasomes promote MMT rather than exosomes.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.