High-fat stimulation induces atrial structural remodeling via the TPM1/P53/SHISA5 Axis

Abstract

Background

Atrial structural remodeling plays a central role in the development and progression of atrial fibrillation (AF) and significantly influences its course. Hyperlipidemia, a potential contributor to AF, affects cardiac function through multiple pathways. This study aimed to investigate the underlying mechanisms by which high lipid levels promote AF progression.

Methods

In vitro cell models were established using palmitic acid (PA) stimulation, and in vivo rat models were generated by feeding a high-fat diet (HFD). Proteomic and transcriptomic sequencing analyses were conducted to identify differentially expressed proteins and genes. Extracellular vesicles (EVs) were isolated and characterized by differential centrifugation. Cell proliferation was assessed using EdU incorporation and flow cytometry, while transmission electron microscopy (TEM) was used to observe autophagy. Protein expression was analyzed by immunoblotting, immunohistochemistry, and immunofluorescence.

Results

High lipid stimulation significantly increased the expression of tropomyosin 1 (TPM1) in cardiomyocytes, which was transferred to cardiac fibroblasts via EVs, activating the P53/SHISA5 signaling axis and inducing endoplasmic reticulum (ER) stress and autophagy, thereby promoting atrial structural remodeling. Activation of P53 and overexpression of SHISA5 in human cardiac fibroblast (HCF) cells reduced ER stress, autophagy, and fibrosis. Furthermore, ER stress and autophagy markers were significantly elevated in the atrial tissues of HFD-fed rats, while SHISA5 overexpression mitigated these effects.

Conclusion

High-fat stimulation may induce atrial fibrosis through the TPM1/P53/SHISA5 axis by modulating the ER stress-autophagy pathway.

Introduction

Atrial fibrillation (AF) is the most prevalent tachyarrhythmia and is often accompanied by clinical symptoms. Persistent AF is strongly associated with stroke, heart failure, cognitive impairment, and other severe complications [1]. According to the Global Burden of Disease (GBD) Study, the global prevalence of AF reached 59.7 million cases in 2019. The incidence of AF is approximately 0.7% in individuals aged 55–65 years and rises to 23.5% in those over 85 years [2, 3]. AF is characterized by high rates of disability and mortality. However, the factors contributing to its occurrence are not yet fully understood [4]. The prevailing view is that atrial remodeling plays a central role in the onset and progression of AF. This process includes electrical, structural, and neural remodeling, with structural remodeling serving as the primary substrate for AF maintenance [5]. Structural remodeling involves alterations in the atrial myocardium and extracellular matrix, characterized by myocardial fibrosis, which is closely associated with AF persistence and the difficulty of restoring sinus rhythm. Moreover, structural remodeling and AF development exhibit a bidirectional causal relationship, with each process exacerbating the other [6, 7]. While other mechanisms, such as chronic ionic remodeling and subcellular structural remodeling, may also contribute to AF, this study primarily focuses on the role of structural remodeling in the context of high-fat stimulation.

Hyperlipidemia, a metabolic disorder characterized by elevated blood lipid levels, is a known risk factor for various cardiac diseases, including atherosclerosis, through multiple pathways [8]. It can activate inflammatory cells, including endothelial cells and monocytes, leading to myocardial dysfunction and lipid peroxidation, which in turn generate excessive free radicals that damage myocardial cells and disrupt myocardial homeostasis [9]. Extracellular vesicles (EVs), including exosomes and microvesicles, play a crucial role in intercellular communication by transferring bioactive molecules such as proteins, miRNAs, and lipids between cells [10]. In cardiovascular diseases, EVs derived from cardiomyocytes, fibroblasts, and adipose tissue have been shown to regulate fibrosis, inflammation, and oxidative stress [11]. For instance, EVs derived from epicardial adipose tissue (eFat) in patients with AF carry proteins such as tropomyosin 1 (TPM1), which may activate cardiac fibroblasts [12]. These findings highlight EVs as key mediators of atrial remodeling.

TPM1 is a member of the tropomyosin superfamily and plays a crucial role in muscle tissue. In muscle cells, tropomyosin primarily regulates muscle contraction in a calcium-dependent manner, whereas in non-muscle cells, it stabilizes the cytoskeleton [13, 14]. TPM1 is also believed to be essential for heart development, as studies have shown that TPM1-knockout mice die between embryonic days 10 and 14 due to disrupted myofibril assembly and the absence of a heartbeat [15, 16]. Mutations in TPM1 can trigger myocardial cell hypertrophy, leading to increased thickness of the left ventricular anterior wall, interventricular septum, and lateral wall—key factors contributing to the onset and progression of hypertrophic cardiomyopathy [17, 18].

The tumor suppressor p53 and its downstream target, SHISA5 (scotin), play central roles in ER stress responses and apoptosis [19]. SHISA5 is localized in the endoplasmic reticulum (ER) membrane and interacts with p53 to modulate caspase-dependent pathways and autophagy [20]. Proper ER function, including correct synthesis and functional folding of proteins, is crucial for cardiac function, particularly in cardiomyocytes. The unfolded protein response (UPR) is a protective mechanism activated in response to ER stress, with its primary function being the identification and clearance of misfolded proteins [21]. ER stress-mediated autophagy mitigates ER stress by clearing damaged proteins from the ER, thereby playing an important role in maintaining normal myocardial cell function [22]. However, both excessive and insufficient ER stress-mediated autophagy can lead to metabolic dysfunction in myocardial cells, ultimately contributing to atrial remodeling and AF development [23]. To investigate the mechanisms by which high lipid conditions promote AF progression, this study hypothesizes that under high lipid stimulation, the secretion of TPM1-enriched extracellular vesicles (EVs-TPM1) from cardiomyocytes activates the TPM1/P53/SHISA5 axis, inducing ER stress-autophagy in myocardial fibroblasts, promoting atrial fibrosis and may be involved in atrial remodeling and the development of AF.

Materials and methods

In vivo experiments

High-fat feeding and Adeno-associated virus (AAV) 9-TCF21-SHISA5 treatment

All animal experiments were approved by the Ethics Committee of The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital (Approval Number: 2024041101). For the rats involved in the experiment, we used the carbon dioxide method to euthanize them after the experiment was completed, aiming to ensure that the animals end the experimental process in a humane manner and reduce unnecessary pain and harm. Male Sprague-Dawley (SD) rats (8 weeks old) weighing 200 g (± 20 g) were purchased from Beijing Huafukang Biotech Co., Ltd. The animals were housed at a temperature-controlled environment of 20 ± 2 °C with a 12-hour light/dark cycle. They were randomly divided into a control group and a high-fat diet (HFD) group. The control group was fed a standard diet for 12 weeks, while the HFD group was fed a high-fat diet providing 40% of energy from fat for 12 weeks. After 12 weeks, recombinant AAV9 vectors carrying the rat SHISA5 sequence (AAV9-TCF21-SHISA5, Dongze, Hanbio Inc, Shanghai, China) were used to overexpress SHISA5. An AAV9 overexpression control (AAV9-TCF21-NC, Dongze, Hanbio Inc, Shanghai, China) served as a negative control. Via the tail vein, 1.6 × 10^12 vg/mL of AAV9-TCF21-SHISA5 or AAV9-TCF21-NC was injected into NC or HFD rats, respectively, at a dosage of 250 µL per 200 g of rat body weight, and tissues were harvested after a 4-week period of action. In this study, 40 SD rats were used, with 10 rats per group (Control, HFD, HFD + AAV9-SHISA5, HFD + AAV9-NC).

Mass spectrometry

Left atrial tissues were collected from rats for proteomics analysis. Atrial myocardium was lysed by ultrasonication in pre-chilled lysis buffer for 10 min. After centrifugation at 14,000 g for 14 min at 4 °C, the supernatant was transferred to a new tube. The proteins were reduced in NH₄HCO₃ buffer with dithiothreitol (DTT) in a 37 °C water bath for 3 h, followed by alkylation with iodoacetamide (IAA) in NH₄HCO₃ in the dark at room temperature for 1 h. The proteins were then digested overnight with sequencing-grade trypsin at 37 °C. After desalting and vacuum drying, the peptides were dissolved in 0.1% formic acid and separated using an Acclaim™ PepMap™ 100 C18 HPLC column (Thermo Scientific, USA). The HPLC system was coupled with an Orbitrap Fusion Lumos Tribrid mass spectrometer for MS1 scans (m/z 350–1800) at a resolution of 30,000 with an AGC target of 4,000,000. Peptide precursors with charge states ≥ 2 were selected for MS2 analysis with an AGC target of 20,000 and a maximum injection time of 80 ms. Up to 20 of the most abundant ions in the Orbitrap full MS spectrum were selected for further MS/MS analysis. The raw MS data were processed using the MaxQuant proteomics platform (version 1.5.2.8) in combination with the Uniprot - Mus musculus database. The search parameters included fixed modification of carbamidomethyl (C) and variable modification of oxidation (M). Trypsin was used as the enzyme, with a maximum of 2 missed cleavages. The peptide mass tolerance was set to 20 ppm, and the fragment mass tolerance was 0.6 Da. Differential expression analysis was performed using a fold change ratio ≥ 1.5 or ≤ 0.67 at a confidence level above 95%. Gene ontology (GO) annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were conducted using the online resources.

HE staining

Tissue samples were fixed in 4% paraformaldehyde for 24 h, followed by routine dehydration, clarification, and paraffin embedding. Paraffin Sect. (5 μm thick) were placed on glass slides, sequentially dewaxed, and rehydrated. The sections were then stained with hematoxylin for 10 min, rinsed with running water, differentiated with 1% hydrochloric acid-ethanol for 10 s, and counterstained with eosin for 1 min. After gradient ethanol dehydration and xylene clarification, the sections were mounted and observed under a light microscope.

Masson staining

Masson’s trichrome staining was performed using a commercial kit (Beyotime) to assess collagen deposition. Paraffin Sect. (5 μm thick) were dewaxed, rehydrated, and stained with Weigert’s iron hematoxylin for 10 min to stain the nuclei. Subsequently, the sections were stained with acidic fuchsin for 5 min, differentiated using 1% phosphoric acid for 1 min, and treated with 1% acetic acid for 1 min. The sections were then treated with phosphomolybdic acid for 5 min and stained with aniline blue for 10 min to visualize collagen fibers. After gradient ethanol dehydration and xylene clarification, the sections were mounted and observed. In Masson’s trichrome staining, nuclei appeared purple-black, collagen fibers appeared blue, and the cytoplasm appeared pink.

Immunohistochemistry

Left atrial tissues were collected from rats, after tissue collection, samples were fixed in 4% paraformaldehyde for 48 h. Paraffin-embedded sections were cut into 5 μm slices. The sections were deparaffinized with a gradient of citrate solution and antigen retrieval was performed. Endogenous peroxidase activity was blocked with a 3% hydrogen peroxide solution. After rinsing with PBS, the sections were incubated with a 5% BSA solution for 30 min. Subsequently, the sections were incubated with the primary antibody at a specific dilution overnight at 4 °C. The next day, the paraffin sections were incubated at room temperature for 30 min, rinsed three times with PBS for 5 min each, and then incubated with the secondary antibody for 50 min at room temperature. After rinsing, DAB was added, and the reaction was terminated under microscopic control of the staining time and rinsed when positivity appeared. Hematoxylin staining was applied for 1 min and followed by hematoxylin differentiation solution for several seconds. After dehydration and sealing, the sections were observed under a microscope and data were collected. The antibodies used are as follows: p-IRE1α (1:200, Cat#, Abclone), GRP78 (1:200, Cat#, Abclone).

In vitro

Cell culture

Human cardiac fibroblasts (HCF) were procured from Zhejiang Meisen Cell Technology Co., Ltd. (Meisen CTCC), and human cardiomyocytes (AC16) were obtained from Wuhan Servicebio. The cell lines were cultured in DMEM (Invitrogen, Waltham, USA) supplemented with 10% fetal bovine serum under a humidified atmosphere of 5% CO2 at 37 °C. Cells were seeded in 6-well plates (5 × 10^5 cells/well) or 12-well plates (5 × 10^4 cells/well). Palmitic acid (PA, Sigma, 100 ng/ml) was used to treat AC16 cells for 48 h, following which cells and extracellular vesicles from the supernatant were collected for further experiments. Human recombinant TPM1 protein (rTPM1, Novus, MBP1-48329, 100 ng/ml) was used to treat HCF cells for 24 h, with or without 4-phenylbutyrate (4-PBA; an ER stress inhibitor, 200 µg/ml 2 h before rTPM1), c16- ceramide (c16-c, a P53 agonist, 10 µM, 6 h before rTPM1), and Bafilomycin A1 (BafA1, an autophagy inhibitor, 200 µM, 2 h before rTPM1). All chemical reagents were purchased from MCE Company, and cells were then collected for subsequent experiments.

Adenovirus (ADV)-SHISA5 overexpression transfection in HCF cells

The ADV-SHISA5 overexpression was designed and packaged by Heyuan Biotech Company, with a viral titer of 10^10. When the cell confluence reached 70%, the viral transfection was performed (MOI = 200). The viral dilution was first added to the culture medium with a volume of 1/2 and incubated. After 2 h, the remaining culture medium was added. After 24 h of culture, the medium was replaced with fresh culture medium, and the cells were collected for further experiments after an additional 48 h of culture.

Western blotting

A lysis buffer mixture was prepared with RIPA lysis buffer, PMSF, and a 50× phosphatase inhibitor (Beyotime, Shanghai, China). This mixture was then added to the samples and lysed on ice before being centrifuged at 4 °C at a speed of 12,000 rpm for 15 min. The supernatant was mixed with a 5× loading buffer (Beyotime, Shanghai, China), denatured in a 95 °C metal bath for 15 min, and then chilled at -20 °C. Proteins were separated using a 10% SDS-PAGE gel, with 15 µg of protein per lane. The proteins were transferred onto a PVDF membrane and then blocked with skim milk or bovine serum albumin for 1 h. The primary antibodies (Table 1) were then incubated with the membrane overnight at 4 °C. The membrane was washed three times with TBST for 10 min each, followed by incubation with horseradish peroxidase-conjugated goat anti-mouse/rabbit IgG (h + L) for 1 h (ZSGB-BIO, China). After another three washes with TBST for 10 min each, the membrane was incubated with an ECL chemiluminescence solution (Yeason) and then exposed.

Table 1 List of all antibodies used in the study

Quantitative Real-Time PCR

Total RNA was extracted from cells treated with recombinant TPM1 protein using the traditional method with Trizol (Thermo Fisher Scientific, Waltham, USA) and reverse transcribed into cDNA using the TransScript^® First-Strand cDNA Synthesis SuperMix (TransGen, China). Quantitative PCR was performed using the QuantStudio™ 7 Flex Real-Time PCR system (Applied Biosystems™) with PerfectStart™ Green qPCR SuperMix (TransGen, China). Data were analyzed using the 2 − ΔCt method with GAPDH as the endogenous control. The gene-specific primers used for qPCR are listed in Table 2.

Table 2 The gene-specific primers used for qPCR

RNA sequencing analysis

RNA isolation and purification were performed according to the manufacturer’s instructions for the TRIzol reagent kit (Thermo Fisher Scientific). RNA concentration and purity were measured using a NanoDrop ND-1000 spectrophotometer, and RNA integrity was assessed using a Bioanalyzer 2100. Only RNA samples with a concentration greater than 50 ng/µL, an RIN value greater than 7.0, and a total quantity greater than 1 µg were selected for subsequent experiments. PolyA mRNA was enriched using Dynabeads Oligo (dT) magnetic beads, and mRNA was fragmented using the NEBNext Magnesium RNA Fragmentation Module at elevated temperatures. Double-stranded cDNA was synthesized using E. coli DNA polymerase I and RNase H, and DNA fragments with adapters were prepared by dUTP treatment. After treatment with UDG enzyme, the library was amplified by PCR to form fragments of approximately 300 bp, which were then sequenced on an Illumina Novaseq 6000 platform. After quality control and filtering of the sequencing data, clean data were obtained. The data were aligned to the reference genome, and gene expression quantification, GSEA, differential gene analysis, and enrichment analysis were performed. The alignment results provided statistics on the number of reads and their regional distribution in the sequencing data. Valid data were categorized based on genomic annotations into exons, introns, and intergenic regions. Using StringTie software, combined with genomic information and de novo assembly methods, the genes and transcripts in the samples were quantified, accurately assessing gene expression levels.

Immunocytochemistry and confocal imaging

After removing the culture medium, the cells on the cover slips were washed three times with PBS. A 4% paraformaldehyde (PFA) solution was added to fix the cells at room temperature for 20 min. After washing, the cells were permeabilized with TritonX-100 (Biyuntian) for 10 min at room temperature, followed by blocking with 5% BSA (in PBS) for 1 h. The cells were then incubated with the primary antibody diluted in the blocking solution at 4 °C overnight. The next day, after washing the cells again, the secondary antibody was applied, and the cells were incubated at room temperature in the dark for 1.5 h. DAPI staining was performed for 10 min (Biyuntian). Finally, an anti-fluorescence quenching agent was used to mount the cover slips, and the cells were observed using a confocal microscope. Images were captured using a Nikon confocal microscope. The antibodies used included LC3B, LAMP1, P53, and SHISA5. The dyes used included Dil cell membrane dye, phalloidin, and PKH67 green fluorescent dye.

Edu assay

The Edu incorporation assay was performed using a kit purchased from Beyotime (Cat#C0075S). After stimulation, an equal volume of Edu reaction solution was added, and the cells were incubated at 37 °C for 2 h. The reaction solution was prepared according to the manufacturer’s instructions and added after washing and fixing the cells. The cells were incubated in the dark for 20 min. After washing with PBS, Hoechst staining was used to label the cell nuclei. Fluorescence microscopy was used for detection with an excitation wavelength of 346 nm and an emission wavelength of 565 nm.

Flow cytometry

Cell cycle analysis was performed using the Cell Cycle and Apoptosis Analysis Kit (Yeason, Shanghai, China) according to the manufacturer’s protocol. HCF cells were plated at a density of 10^5–6 cells per well in 6-well plates. The cells were fixed with 70% ethanol at 4 °C for 2 h and then stained with propidium iodide (PI) staining solution at 37 °C in the dark for 30 min. Flow cytometry analysis was completed within 5 h, and the data were analyzed using Flowjo software (BD Biosciences Inc., New Jersey, USA).

Transmission electron microscopy (TEM)

HCF cells treated with rTPM1 for 24 h were fixed in 2.5% glutaraldehyde at room temperature for 5 min, and cell samples were collected. The cells were centrifuged at 3000 rpm for 2 min, and then the pellet was fixed in electron microscopy fixative solution and kept in the dark for 30 min. A 20 µL sample was dropped onto a carbon-coated copper grid and left for 3–5 min. Phosphotungstic acid 2% was dropped onto the carbon support film copper grid and left for 1–2 min. Excess liquid was wicked away with filter paper and the grid was air-dried. The samples were observed under a transmission electron microscope (Hitachi, HT7800/HT7700) and images were captured for analysis.

Exosome separation

The supernatant from AC16 cells was collected and EVs were isolated using ultracentrifugation. The samples were centrifuged at 4 °C at 500 g for 10 min, 1500 g for 15 min, and 3000 g for 25 min. The supernatant was filtered through a 0.22 μm filter (Steritop; Millipore) to remove dead cells, cellular debris, and large vesicles. The supernatant was then centrifuged at 10,000 g for 60 min to remove dead cells, cellular debris, and large vesicles. The EVs were pelleted by ultracentrifugation at 100,000 g for 120 min at 4 °C, and the isolated EVs were resuspended in prechilled PBS and stored at -80 °C until use. Part of the isolated EVs were labeled with PKH67 (Yeason, Shanghai, China) and given to HCF cell culture.

Nanoparticle tracking analysis (NTA)

The size distribution and concentration of exosomes in the supernatant of AC16 cells with or without PA stimulation were analyzed using nanoparticle tracking analysis (NTA) with a ZETA-VIEW instrument (Particle Metrix). A 5 µL sample of exosomes was diluted in PBS to a final volume of 30 µL. Videos of 60 s were recorded at a frame rate of 30 frames per second, and the average size and concentration of exosomes were determined.

Statistical analysis

All experiments were performed in triplicate. For statistical comparisons between two groups, an unpaired t-test was used, and for multiple group comparisons, one-way ANOVA was applied. These analyses were conducted using GraphPad Prism 9.0 statistical software (San Diego, CA). Data are presented as mean ± Standard Error of the Mean (SEM), and the significance level was set at P < 0.05. In the figures, different asterisks indicate levels of statistical significance: * indicates P < 0.05, **indicates P < 0.01, *** indicates P < 0.001, and **** indicates P < 0.0001.

Results

Identification of TPM1 as a putative target protein through LC-MS/MS and WB analysis

Masson staining showed an increased proportion of blue-stained areas in the atrial tissues of rats in the HFD group, indicating more severe fibrosis (Fig. 1a). LC-MS/MS was then performed to identify differentially expressed proteins (DEPs) between rats on a normal diet and those on a high-fat diet. A total of 198 DEPs were identified between the two groups, including 104 upregulated and 94 downregulated proteins (Fig. 1b). The volcano plot indicated a significant upregulation of TPM1 in the HFD group (Fig. 1c). Notably, a comparison of DEPs between the control and HFD groups revealed that TPM1, an actin-binding protein, was significantly upregulated in the HFD group, suggesting that HFD stimulation may influence TPM1 expression. Thus, we hypothesized that TPM1 is a candidate molecule involved in myocardial fibrosis. A Western blot analysis was conducted to validate the proteomic findings, confirming the TPM1 protein was upregulated in the HFD group, which was consistent with the proteomics results (Fig. 1d).

Fig. 1

Identification of TPM1 as a putative target protein. Masson staining to assess the degree of tissue fibrosis after high-fat diet (A). Analysis of differentially expressed proteins by LC-MS/MS (B). Volcano plot showing up- and down-regulated proteins (C). WB detection of TPM1 protein expression changes after high-fat diet (D). WB detection of TPM1 protein expression abundance differences between AC16 cells and HCF cells (E). Changes in TPM1 protein expression in AC16 cells after stimulation with different concentrations of PA (F)

PA stimulation increases TPM1 protein expression in AC16 cells

Western blot analysis was performed to compare TPM1 protein abundance between HCF cardiomyocytes and AC16 myocardial cells. TPM1 protein expression was higher in HCF cells than in AC16 cells (Fig. 1e). AC16 cells were stimulated with varying concentrations of PA (0–0.5 nM) for 48 h, with TPM1 protein expression peaking at 0.1 nM. However, at higher PA concentrations, TPM1 expression declined (Fig. 1f). Thus, 0.1 nM PA stimulation for 48 h was chosen for subsequent experiments. Confocal fluorescence microscopy was used to assess TPM1 expression in cells after PA stimulation, yielding results consistent with the western blot findings (Fig. 2a).

Fig. 2

PA stimulation increases the secretion of EVs-TPM1 from AC16 cells. Verification of TPM1 changes in AC16 cells after PA stimulation using confocal microscopy (40X), Scale bar = 10 μm (A). WB verification of EVs marker proteins (B). Representative morphological analysis by TEM (C). Particle size distribution detected by NTA (D). Immunofluorescence tracing of exosomes (40X), Scale bar = 10 μm (E)

PA stimulation increases the secretion of EVs-TPM1 from AC16 cells

EVs were extracted using differential ultracentrifugation, and western blot analysis confirmed the presence of the EVs marker heat shock protein 70 (Hsp70) and TSG101, along with an upregulated expression of EVs-TPM1 protein under PA stimulation (Fig. 2b). Additionally, transmission electron microscopy (TEM) revealed vesicles with diameters ranging from 30 to 150 nm, consistent with the characteristics of EVs (Fig. 2c). Nanoparticle tracking analysis (NTA) showed a similar size distribution of exosomes in both the control and PA-stimulated groups (Fig. 2d). To further verify the uptake capacity of HCF cells for EVs, EVs were labeled with PKH67, and scattered green fluorescent signals were observed within the cells, confirming successful internalization (Fig. 2e).

TPM1 stimulation induces proliferation and fibrosis in HCF cells

HCF cells were stimulated with varying concentrations of recombinant TPM1 protein (0–1000 nM) for 24 h, and western blot analysis revealed that fibronectin 1 (Fn1), type I collagen (Col I), matrix metalloproteinase-2 (MMP2), and TGF-β expression peaked at 100 nM stimulation (Fig. 3a). EdU assay demonstrated an increase in the percentage of proliferating HCF cells following rTPM1 stimulation (Fig. 3b). Flow cytometry analysis revealed changes in the cell cycle of HCFs after stimulation with different concentrations of rTPM1. Under 100nM rTPM1, there was an increase in the S phase and a decrease in the G1 phase, indicating cell proliferation. However, at 1000 nM rTPM1, the G1 phase was prolonged while the S phase was shortened compared to the 100 nM condition. (Fig. 3c).

Fig. 3

TPM1 stimulation induces proliferation and fibrosis in HCF cells. WB detection of changes in Fibronectin, Collagen I, MMP2, TGF-β expression in HCF cells after rTPM1 stimulation (A). EdU assay of HCF cell proliferation levels after rTPM1 stimulation (20X), Scale bar = 100 μm (B). Flow cytometry verification of HCF cell proliferation levels after rTPM1 stimulation (C)

RNA-seq analysis identifies SHISA5 as a downstream target of TPM1

Current research on TPM1 remains limited. To further investigate the mechanism by which TPM1 induces fibrosis in HCFs, transcriptome sequencing analysis was performed on HCFs stimulated with rTPM1. RNA-seq analysis identified 985 upregulated and 1185 downregulated transcripts. Among these, SHISA5 was identified as a potential downstream target of TPM1 (Fig. 4a, b). Figure 4d presents significantly enriched gene ontology (GO) terms across biological processes (BP), cellular components (CC), and molecular functions (MF). Five transcribed genes (SHISA5, CAPNS1, NDUFC2-KCTD14, ATG13, and BASP1) were selected and validated via RT-qPCR, which confirmed that their expression patterns were consistent with the transcriptomic findings (Fig. 4c). Based on these results, we hypothesized that in cardiac fibroblasts, the TPM1/P53/SHISA5 signaling axis regulates ER stress, thereby inducing cellular autophagy.

Fig. 4

RNA-seq analysis on HCF cells stimulated with rTPM1. Analysis of differentially expressed genes by Transcriptome sequencing (A,B). qPCR validation of transcriptomics results (C). Gene Ontology (GO) enrichment analysis of differentially expressed proteins (DEPs) in NC and rTPM1-stimulated groups. The Y-axis represents the number of DEP genes (D)

TPM1 induces ER stress and autophagy in HCF cells

The expression of membrane-associated proteins light chain 3-II (LC3-II) and p62, which serve as indicators of autophagosome formation, is commonly used to assess autophagy. Therefore, we investigated ER stress and autophagy in HCF cells following TPM1 protein stimulation. Western blot analysis showed downregulation of p62 expression and upregulation of LC3-II, GRP78, p-IRE1α, and p-JNK in the rTPM1 stimulation group, indicating increased ER stress and autophagy in HCF cells after TPM1 stimulation. The autophagy inhibitor BafA1 and BafA1 + rTPM1 stimulation groups exhibited blocked autophagy, accumulation of autophagosomes, and a subsequent increase in LC3B expression (Fig. 5a). Notably, TEM results supported these findings, revealing blurred cell membranes, pale cytoplasm, mild dilation of the rough ER, widened vesicle spaces, local fuzziness, rupture, and detached ribosomes, indicative of autophagic vacuole formation (Fig. 5b). Lysosome-associated membrane protein 1 (LAMP1) is distributed in autophagy and lysosomal organelles, and colocalization analysis of LAMP1 and LC3B using immunofluorescence demonstrated expression trends consistent with western blot analysis (Fig. 5c). Subsequently, we used 4-PBA to treat HCF cells to confirm that TPM1-induced cellular fibrosis occurs via the ER stress pathway. The results showed that ER stress, autophagy, and fibrosis levels were alleviated in the rTPM1 + 4-PBA group compared with the rTPM1 group (Fig. 5d).

Fig. 5

TPM1 induces ER stress and autophagy in HCF cells. WB detection of P62, P53, SHISA5, LC3-II, p-IRE1α, p-JNK protein expression after rTPM1 and BafA1 stimulation (A). TEM observation of autophagosomes and endoplasmic reticulum structures in HCF cells with or without rTPM1 stimulation (B). Immunofluorescence co-localization analysis of autophagy levels in HCF cells after rTPM1 and BafA1 stimulation, Scale Bar = 10 μm (C). Immunofluorescence observation of P53 and SHISA5 protein expression in HCF cells after rTPM1 stimulation, Scale Bar = 10 μm (D). WB detection of Collagen I, p-IRE1α, GRP78, MMP2, P62, p-JNK, LC3-II proteins after 4-PBA inhibition of ERS (E)

The TPM1/P53/SHISA5 axis mediates ER stress and autophagy

By analyzing cytoplasmic and nuclear proteins in HCF cells, we found that P53 was primarily localized in the cell nucleus, and nuclear P53 expression was significantly downregulated following rTPM1 stimulation (Fig. 6a). To assess the role of P53 activation, HCF cells were treated with C16-ceramide (10 µM, 24 h), a P53 activator, which led to increased P53 and SHISA5 expression, along with reduced ER stress, autophagy, and fibrosis levels (Fig. 6b). To further validate the regulatory role of this signaling axis in HCF cells, we generated an ADV-overexpressing SHISA5 cell model. Following SHISA5 overexpression, ER stress, autophagy, and fibrosis levels were all alleviated (Fig. 6c).

Fig. 6

The TPM1/P53/SHISA5 axis mediates ER stress and autophagy. WB detection of cytoplasmic and nuclear P53 protein expression in HCF cells after rTPM1 stimulation (A). WB detection of P53, SHISA5, Collagen I, MMP2, p-IRE1α, GRP78, p-JNK, P62, LC3-II protein expression in HCF cells after activation of P53 by C16-ceramide stimulation (B). WB detection of Collagen I, p-IRE1α, GRP78, MMP2, P62, p-JNK, SHISA5, LC3-II protein expression after ADV-SHISA5 transfection in HCF cells (C)

In vivo animal experiment validation

To further verify the protective effect of SHISA5 on HFD-induced atrial fibrosis, we established HFD-induced rat models and adeno-associated viral SHISA5 overexpression models. Three weeks after viral construction, we assessed the transfection efficiency of the AAV9 virus construct and confirmed that SHISA5 expression was upregulated in the SHISA5 overexpression group compared to the overexpression control group. In contrast, the HFD model group exhibited downregulated SHISA5 expression relative to the healthy control group, consistent with the results of in vitro experiments. Masson’s staining and hematoxylin and eosin (HE) staining were performed on rat tissue paraffin sections. Increased tissue fibrosis (blue) was observed in the HFD group, while SHISA5 overexpression reduced the proportion of fibrosis. Immunohistochemistry revealed a significant increase in the positive density of p-IRE1α and GRP78 in the HFD group. However, after AAV9-SHISA5 treatment, the density of positive cells significantly decreased (Fig. 7a). Western blot analysis showed that the expression levels of Collagen I, MMP2, p-IRE1α, p-JNK, GRP78, and LC3B-II were elevated in the HFD group and decreased following AAV9-SHISA5 treatment. In contrast, P62, P53, and SHISA5 protein expression levels were reduced in the HFD group (Fig. 7b). Additionally, we used PDGFRα to specifically label cardiac fibroblasts in atrial tissue, and co-expression analysis of PDGFRα with P53 revealed decreased P53 expression in fibrotic tissue in the HFD group (Fig. 7c), (Fig. 8).

Fig. 7

In vivo animal experiment validation. Optical microscopy (20X) observation of rat atrial myocardial tissue, HE staining to observe tissue morphology, Masson staining to observe the degree of tissue fibrosis, IHC to observe the expression proportion of GRP78 and p-IRE1 proteins in the tissue, fluorescence microscopy (20x) observation of AAV virus transfection efficiency of SHISA5 in rat atrial myocardial tissue (A). WB detection of Collagen I, p-IRE1α, GRP78, MMP2, P62, p-JNK, P53, SHISA5, LC3-II expression in atrial tissue (B). Immunofluorescence observation of P53 protein (Red) expression in PDGFRα-marked cardiac fibroblasts (Green) (Yellow) (C)

Fig. 8

Proposed mechanism of the TPM1/P53/SHISA5 axis. Under high-fat stimulation, AC16 cells secrete EVs-TPM1, which act on HCF cells. Within HCF, TPM1 inhibits P53 activation, leading to downregulation of SHISA5 expression in the ER. Reduced SHISA5 triggers ERS and subsequent dysregulation of autophagy, ultimately promoting myocardial fibrosis (By Figdraw)

Discussion

Our study aimed to elucidate the effect of TPM1 protein derived from cardiomyocyte-secreted EVs on ER stress and autophagy in cardiac fibroblasts under high-fat stimulation, as well as its role in atrial structural remodeling. Our experiments demonstrated that TPM1 expression was significantly upregulated in HFD-fed rats. Moreover, PA stimulation of AC16 cardiomyocytes markedly increased TPM1 expression, which was subsequently transferred to cardiac fibroblasts via EVs. The elevated EVs-TPM1 activated the P53/SHISA5 signaling axis, regulated ER stress and autophagy, and induced proliferation and fibrosis in HCF cells. These findings suggest that TPM1 plays a pivotal role in intercellular signaling between cardiomyocytes and cardiac fibroblasts, promoting ER stress and autophagy in HCF cells via an EV-mediated mechanism, thereby contributing to atrial fibrosis and structural remodeling.

Atrial fibrosis is a key factor in the initiation and progression of AF, particularly in the transition from paroxysmal to persistent AF. Emerging evidence suggests that fibrosis may precede AF onset [24]. While previous studies have identified AF triggers at the pulmonary vein orifice, increasing fibrosis may generate additional arrhythmogenic sites independent of the pulmonary veins [25]. Fibrotic remodeling alters myocardial conduction properties, leading to local conduction heterogeneity and conduction blocks, which increase the likelihood of re-entry circuits. Thus, fibrosis is a major contributor to AF pathogenesis [5]. Cardiac fibroblasts, which constitute approximately 24.3% of atrial tissue and reside primarily in the myocardial interstitium, play a fundamental role in maintaining cardiac structure and function while also being central to pathological myocardial fibrosis [26]. Under the influence of various inflammatory mediators, cytokines, and mechanical stress, fibroblasts can differentiate into myofibroblasts, which secrete excessive collagen fibers and contribute to ventricular remodeling. Persistent activation of fibroblasts leads to myocardial interstitial fibrosis [27, 28]. Our previous studies demonstrated that HFD-fed rats exhibited significantly elevated plasma triglyceride, total cholesterol, and low-density lipoprotein (LDL) levels alongside a non-significant decreasing trend in high-density lipoprotein (HDL) levels. Compared to the control group, HFD-fed rats showed a significant reduction in the atrial effective refractory period (AERP), which increased their susceptibility to AF. In this study, we further confirmed that HFD-fed rats developed significant atrial myocardial fibrosis.

Many studies have explored the mechanisms by which an HFD mediates the onset and progression of AF. Previous research has indicated that a fat-rich diet promotes myocardial fibrosis-induced AF by triggering programmed cell death in cardiomyocytes [29]. Zhang et al. found that high-fat stimulation reduces nitric oxide (NO) production through the CRIF1/eNOS/P21 axis, leading to atrial neural remodeling that facilitates AF occurrence [30]. Similarly, Wei et al. demonstrated that a deficiency of MD1, a key regulatory protein, accelerates inflammatory atrial fibrosis and increases the susceptibility of obese mice to AF [31]. In our study, we fed Sprague- Dawley (SD) rats with HFD, and after 12 weeks, the HFD group exhibited significantly elevated levels of ER stress marker proteins—GRP78, p-IRE1α, and p-JNK—as well as autophagy marker protein LC3B-II, compared to the normal diet group. These findings indicate that high-fat feeding promotes ER stress and autophagy in rat atrial tissue. Concurrently, the expression of fibrosis-related marker proteins also increased. HFD is closely associated with ER stress, and ER stress-mediated autophagy serves a regulatory function by mitigating ER stress through the clearance of damaged proteins in the ER, thereby maintaining myocardial cell function. However, excessive or insufficient ER stress-mediated autophagy can result in metabolic dysfunction in myocardial cells, ultimately contributing to atrial remodeling and the progression of AF [22]. Our study elucidated a novel mechanism by which high-fat stimulation promotes fibrosis.

Additionally, EVs play a critical role in intercellular communication during AF development. EVs, which contain DNA, RNA, and proteins, modulate cardiac fibroblasts and either promote or inhibit atrial fibrosis [32]. A study by Olga et al. found that EVs released from the eFat of patients with AF exhibited significantly greater profibrotic effects than those from patients without AF. Comparative proteomic analysis of unpurified eFat-EVs (isolated by ultracentrifugation) from patients with and without AF revealed that TPM1 protein was significantly upregulated in the eFat-EVs of patients with AF [12]. This finding aligns with our proteomic results, which demonstrated a significant upregulation of TPM1 protein in the left atrial tissue of HFD-fed rats compared to the normal-fed group. Our in vitro studies further revealed that under high fatty acid stimulation, TPM1 expression increased predominantly in cardiomyocytes. These cardiomyocytes released EVs containing TPM1, which were subsequently taken up by neighboring cardiac fibroblasts, triggering a cascade of downstream changes that ultimately facilitated AF development. Interestingly, TPM1 levels declined with increasing concentrations of palmitic acid (PA), a phenomenon that may be attributed to PA-induced lipotoxicity, which promotes autophagy and apoptosis, accelerating TPM1 degradation [33,34,35].

Current research on TPM1 proteins primarily focuses on tumors. TPM1 inhibits tumorigenesis by suppressing cell proliferation [36]. Another study confirmed that TPM1 expression is downregulated in many tumors, where it plays a crucial role in inhibiting tumor cell proliferation and reducing apoptosis [37]. Li identified TPM1 as a novel systemic pro-aging factor associated with functional defects in the aging retina. TPM1 triggers the expression of endogenous TPM1 and inflammation in the normal aging retina by phosphorylating protein kinase A (PKA) and regulating the FABP5/NF-κB signaling pathway [38, 39]. Wang reported that TPM1 in EVs derived from young osteocytes promotes osteogenesis by increasing the matrix stiffness of bone marrow mesenchymal stem cells (MSCs) [40]. Jia et al. demonstrated that TPM1 promotes the transformation of rat cardiac fibroblasts into cardiomyocyte-like cells via the PI3K-Akt pathway, improving the efficiency of cardiac fibroblast transformation—a process beneficial for myocardial repair following heart injury [41]. In this study, proteomic analysis of left atrial tissues from HFD-fed rats revealed TPM1 upregulation. Although the absolute log2-fold change appeared modest, this finding aligns with a recent study reporting elevated TPM1 levels in EVs derived from the eFAT of patients with AF compared to non-AF individuals [34]. Additionally, TPM1 has been implicated in hypertrophic cardiomyopathy (HCM), where it promotes myocardial fibrosis and pathological remodeling [42]. These findings suggest that TPM1 plays a critical role in cardiac fibrosis and structural remodeling, making it a compelling target for investigating the mechanisms underlying HFD-induced AF. However, no research has yet explored the role of TPM1 in structural remodeling during AF. In this study, we simulated EVs released by AC16 cells by administering recombinant TPM1 to HCF cells and observed its effects on fibrosis. The results showed that the expression of Fn1, MMP2, Col1, and TGFβ in HCF cells tended to increase following recombinant TPM1 administration. Interestingly, at high concentrations, TPM1 reduced the expression of these fibrotic markers in a dose-dependent manner. This phenomenon may be explained by the saturation of signaling pathways, where excessive TPM1 stimulation reduces the activation of target genes. For instance, a study by Yan et al. demonstrated that excessive TGF-β stimulation in cardiac fibroblasts results in reduced collagen synthesis due to receptor desensitization [43]. Additionally, a negative feedback regulatory mechanism may be involved, wherein physiological TPM1 concentrations upregulate ER stress or autophagy-related proteins, prompting fibroblasts to downregulate fibrosis markers to maintain homeostasis. These findings suggest that TPM1 plays a positive regulatory role in fibrosis.

To further explore the downstream signaling pathways of TPM1, we performed RNA-seq analysis of HCF cells stimulated with rTPM1. Although numerous genes were differentially expressed, we prioritized protein-coding genes with significant GO and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotations supported by existing literature. Among the candidate genes (SHISA5, CAPNS1, NDUFC2-KCTD14, ATG13, and BASP1), we selected SHISA5 due to its well-established role in ER stress, apoptosis, and its interaction with p53. Protein-protein interaction (PPI) network analysis revealed direct interactions between TPM1 and p53, as well as between p53 and SHISA5, supporting the plausibility of the TPM1/P53/SHISA5 signaling axis. Despite the modest relative change in SHISA5 expression, its strong statistical significance (p < 0.001) and consistent downstream effects on ER stress and autophagy markers reinforce its role as a key regulator of ER stress. SHISA5, also known as Scotin, is a human gene and a member of the SHISA family [44]. The protein encoded by this gene is primarily localized in the ER and interacts with p53 to induce apoptosis in a caspase-dependent manner. Bourdon JC et al. identified that p53 selectively binds to a p53-binding site within the promoter region of Scotin, thereby directly activating its transcription. Inhibition of endogenous Scotin expression increases resistance to p53-dependent apoptosis induced by DNA damage, highlighting Scotin’s role in p53-mediated apoptosis [45]. Terrinoni et al. demonstrated that p73 (a transcription factor belonging to the p53 family) promotes apoptosis by inducing Scotin and ER stress, following a pathway analogous to p53-induced apoptosis [46]. Lee et al. found that SHISA5/Scotin inhibits autophagy by preventing the interaction between the ER and autophagosomes. In SHISA5-deficient HeLa cells, contact between the ER exit sites (ERES) and autophagosomes increased, leading to higher autophagy levels. This process was dependent on the activity of class III phosphatidylinositol 3-kinase complex I (PtdIns3K-C1) and the functional assembly of ERES [47]. To enable phagocytes-ER membrane interactions following stimulation, SHISA5 appears to undergo palmitoylation, altering membrane curvature and allowing its detachment, which facilitates membrane fusion and vesicle assembly [48, 49]. During autophagy, the PtdIns3K-C1 complex on the ER membrane is activated, generating PtdIns3P, a crucial membrane component required for autophagosome formation. PtdIns3P recruits PtdIns3P-binding proteins such as WIPI2 and its interacting partner ATG2 to specific ER regions, where these protein complexes extend autophagosome precursors and coordinate with other autophagy-related mechanisms on the ER membrane [50,51,52]. In SHISA5-deficient cells, ER-autophagosome precursor contact is enhanced, leading to increased basal autophagy levels. This suggests that the absence of SHISA5 creates a more favorable intracellular environment for PtdIns3K-C1 activity, thereby promoting autophagosome formation and autophagy. Consequently, when autophagy is activated, SHISA5 may be incorporated into autophagosomes and degraded in lysosomes as part of the routine cellular clearance and recycling process. Based on these findings, we hypothesized that in cardiac fibroblasts, the TPM1/p53/SHISA5 signaling axis regulates ER stress and subsequently induces cellular autophagy. After administering rTPM1 to HCF cells, we observed decreased expression of p53 and SHISA5, along with increased markers of ER stress and autophagy. To further elucidate the regulatory dynamics of this signaling axis, we activated p53 using a specific activator and transfected HCF cells with an SHISA5 overexpression adenovirus and observed downstream changes, confirming the regulatory relationship within the TPM1/p53/SHISA5 axis. Additionally, we injected a SHISA5 overexpression adeno-associated virus into HFD-fed rats, which showed lower levels of ER stress and autophagy, as well as reduced atrial fibrosis compared to HFD-fed control rats.

ER stress plays an important role in inflammation and the stress response induced by high-fat stimulation. Following the induction of ER stress, the UPR acts as a defense mechanism to prevent the accumulation of unfolded or misfolded proteins through two major degradation systems: the proteasome and autophagy [53, 54]. When the burden of unfolded or misfolded proteins in the ER increases, GRP78 inhibits UPR sensor proteins (PERK, IRE1, and ATF6) [55, 56]. Once ER stress is triggered, both the ubiquitin-proteasome system (UPS) and the autophagy-lysosome pathway (ALP) become activated [57]. Previous studies have shown that the link between ER stress and autophagy is mainly mediated by the activation of IRE1-related JNK. During ER stress, UPR signaling can promote autophagy via the IRE1-JNK or PERK pathways. The association between ER stress and autophagy is mainly mediated by the activation of IRE1-related JNK. IRE1, a transmembrane kinase, is activated in response to ER stress and phosphorylates downstream target proteins through its kinase activity [58]. JNK-mediated phosphorylation of Bcl-2 disrupts the Beclin-1–Bcl-2 complex on the ER membrane, thereby increasing free Beclin-1 levels. Bcl-1 interacts with PI3K and autophagy-related proteins involved in autophagosome membrane formation [59]. Recent studies have highlighted the critical role of autophagy in myocardial fibrosis. A key event in myocardial fibrosis is the activation of fibroblasts into myofibroblasts, a process largely dependent on the TGF-β-Smad classical signaling pathway [60, 61]. mTORC1 serves as a crucial negative regulator of autophagy and, in synergy with Smad3, functions as a major transcriptional regulator of amino acid metabolism, promoting collagen synthesis in myofibroblasts [62, 63]. In this study, high-fat stimulation-induced EVs-TPM1 conditions triggered autophagy in cardiac fibroblasts via the IRE1-mediated signaling pathway. After inhibiting ER stress with 4-PBA, autophagy and fibrosis in HCF cells were significantly reduced. In addition, we used BafA1 to block autophagy, as it inhibits the V-ATPase (proton pump) of lysosomes, thereby preventing the fusion of autophagosomes and lysosomes and disrupting autophagic degradation [64, 65]. Confocal microscopy following BafA1 treatment revealed a significant accumulation of autophagosomes, indicating active autophagy in the cells prior to treatment.

However, this study had several limitations. The high mortality risk associated with TPM1 deficiency prevented us from directly interfering with or silencing TPM1 expression. Although this study focused on the atria, AC16 cells were derived from human ventricular cardiomyocytes. Currently, no commercially available human atrial cardiomyocyte cell line has been identified, and obtaining primary human atrial cardiomyocytes remains challenging. While HL-1 cells (mouse atrial cardiomyocytes) provide an alternative, interspecies differences between mice (HL-1 cells) and humans (HCF cells) may introduce confounding factors. Although this choice may not fully replicate atrial-specific responses, our findings offer foundational insights into TPM1-mediated mechanisms. EVs represent only one component within the complex network of paracrine mediators. A broader analysis of paracrine mediators secreted by cardiomyocytes may yield a more comprehensive understanding of atrial fibrosis and structural remodeling associated with HFD. Importantly, this study was conducted exclusively in male rats. This decision was based primarily on previous studies and the additional variability introduced by the estrous cycle in female rats [66, 67]. Future studies should incorporate female models to determine whether the TPM1/P53/SHISA5 axis functions similarly in both sexes.

Conclusions

In this study, we demonstrated that high-fat stimulation increases TPM1 expression in cardiomyocytes, which is subsequently transferred to cardiac fibroblasts via EVs. This process leads to abnormal activation of the ER stress response through the TPM1/P53/SHISA5 pathway, inducing autophagy and promoting myocardial fibrosis. Although atrial fibrosis is a well-recognized substrate for AF maintenance, further electrophysiological studies are needed to directly establish its role in AF initiation. Our study highlights the mechanisms by which TPM1 mediates fibrosis in cardiac fibroblasts and identifies EV-derived TPM1 as a novel contributor to this process. These findings provide mechanistic insights into how HFDs exacerbate atrial fibrosis, potentially predisposing individuals to AF progression and atrial remodeling. This research lays the groundwork for future studies exploring metabolic interventions targeting fibrosis.