Human cardiomyocytes AC16 (CL-0790, Procell, Wuhan, Hubei, China) were cultivated in Dulbecco’s modified Eagle’s medium (DMEM, #G4512-500ML, Servicebio, Wuhan, Hubei, China) supplemented with 10% fetal bovine serum (FBS, #164210; Pricella, Shanghai, China) under standard conditions (37 °C, 5% CO₂). AC16 cells were subjected to short tandem repeat (STR) analysis and Mycoplasma testing. The cells were passaged using 0.25% trypsin (#BL501B; Biosharp, Hefei, Anhui, China) once they reached 70–80% confluence. For cryopreservation, the cells were resuspended in freezing medium consisting of DMEM, FBS, and dimethyl sulfoxide (DMSO, #HY-Y0320C, MCE, Monmouth Junction, NJ, USA) at a ratio of 5:4:1. The cell suspension was transferred into cryovials and gradually cooled to –80 °C overnight before being stored in liquid nitrogen.

Angiotensin II (Ang II) is a major regulator of cardiovascular disease, causing cardiomyocyte hypertrophy, fibrosis, and oxidative stress [16]. In this study, AC16 human cardiomyocytes were treated with 1 µmol/L Ang II (#HY-13948, MCE, Monmouth Junction, NJ, USA) for 24 hours to establish a cell model that mimics the pathological conditions of HF. The small interfering RNA (siRNA) sequences targeting SFXN2 (Supplementary Material 1), the TRPV1 overexpression plasmid (Supplementary Material 2) and the negative control (OE-NC, pRP[Exp]-CMV$>$ORF_Stuffer) were synthesized by General Biotechnology Co., Ltd. (Hefei, Anhui, China) and Vectorbuilder (Guangzhou, Guangdong, China). For TRPV1 overexpression, AC16 cells were added to 6-well plates to 70% confluency and transfected with the TRPV1 overexpression plasmid using Lipo8000™ transfection reagent (#C0533; Beyotime, Shanghai, China) according to the manufacturer’s instructions. Transfection complexes were prepared by mixing plasmid DNA with Lipo8000™ reagent in serum-free medium and incubating for 10‒15 minutes. For SFXN2 knockdown, cells were transfected with 2.5 µL of siRNA targeting SFXN2 (Si-SFXN2) or a negative control (Si-NC) using the same method. For the cotransfection experiments, the TRPV1 plasmid and Si-SFXN2 (or Si-NC) were co-delivered into the cells after 6 h, and the medium was changed to complete medium supplemented with 10% FBS. The cells were cultured for 24 hours before further analysis.

Eight- to twelve-week-old C57BL/6J male mice (20–25 g) (total = 32) were obtained from Liaoning Changsheng Biotechnology Co., Ltd. (Shenyang, Liaoning, China), which maintains specific pathogen-free certification (SCXK[Liao]2020–000). Every animal experiment was carried out in compliance with the rules and regulations that were authorized by the Animal Ethics Committee’s Institutional Animal Care and Use Committee of The First Affiliated Hospital of Anhui University of Chinese Medicine (AZYFY-2024-1006). All experiments were conducted in accordance with the 3R principles. Research was conducted at the Anhui Provincial Key Laboratory of Meridians and Organs. The animals were maintained at 22 $\pm$ 2 °C under controlled lighting (12 h cycles) with unrestricted access to standard chow and water.

The mice were randomly assigned to four groups: the sham, TAC, TAC+AAV9-cardiac troponin T (CTNT)-control, and TAC+AAV9-CTNT-TRPV1 groups (n = 8 per group). The mice were anesthetized with ketamine (100 mg/kg) and xylazine (10 mg/kg) (6–1712, Bingene, Beijing, China) via intraperitoneal injection. TAC surgery was conducted as previously described [17] to create a mouse model of cardiac hypertrophy and HF. Sham-operated control mice underwent the same surgery but without aortic constriction. AAV9 was used to achieve cardiac-specific overexpression of TRPV1. The AAV9-TRPV1 vector was constructed by inserting mouse TRPV1 complementary DNA (cDNA) under the control of the promoter (Supplementary Material 3). The AAV9-CTNT-control vector (AAV00274Z, Creative-Biogene, Shirley, NY, USA) without any exogenous gene was used as a control. AAV9 vectors (1 $\times$ 10¹¹ viral genome particles per mouse) were administered by tail vein injection. After 2 weeks, the infected mice were subjected to TAC for 28 days. After the study, euthanasia was performed using an overdose of sodium pentobarbital (150 mg/kg, intraperitoneal injection).

Cardiac function was evaluated using a Vevo 2100 system (VisualSonics, Toronto, ON, Canada) with a 30 MHz transducer at week 8 post-surgery. Under 2% isoflurane (#R510-22-10, RWD, Shenzhen, Guangdong, China) anesthesia and temperature monitoring, parasternal short-axis M-mode images were acquired at the papillary muscle level. Left ventricular (LV) functional parameters, including the ejection fraction (EF) and fractional shortening (FS), were analyzed via Vevo Lab software (version 3.1.1, FUJIFILM VisualSonics Inc., Toronto, ON, Canada) by an independent operator unaware of group allocation. All echocardiographic measurements were carried out by an experienced operator who was unaware of the experimental groups.

Following the manufacturer’s instructions, TRIzol reagent (#G3013-100ML; Life Technologies, Carlsbad, CA, USA) was used to separate total RNA from AC16 cells. The RNA concentration and purity were evaluated via an OD1000+ ultramicrospectrophotometer (Nanjing Wuyi Technology Co., Ltd., Nanjing, Jiangsu, China). Samples showing intact RNA bands on 1% agarose gels (ST004L, Beyotime, Shanghai, China) and RNA integrity number (RIN) values exceeding 8 qualified for library construction. The NEBNext Ultra RNA Library Prep Kit for Illumina (#E7530L; New England Biolabs, Ipswich, MA, USA) was used to produce RNA sequencing (RNA-seq) libraries in accordance with the manufacturer’s instructions. Poly-T magnetic beads were used to isolate mRNA for fragmentation. Double-stranded cDNA was generated through reverse transcription with random hexamers followed by second-strand synthesis. After that, the double-stranded cDNA underwent PCR amplification, adapter ligation, end repair, and A-tailing. A Qubit 2.0 fluorometer (Thermo Fisher Scientific, Waltham, MA, USA) was used to quantify the final cDNA libraries, and an Agilent 2100 Bioanalyzer (Agilent Technologies, Inc., Santa Clara, CA, USA) was used for validation. The libraries were sequenced using 150 bp paired-end reads on an Illumina NovaSeq 6000 platform (Illumina, San Diego, CA, USA). Shanghai OE Biotech. Co., Ltd. (Shanghai, China) carried out the RNA sequencing. For every group, three biological replicates were conducted.

The “limma” package (version 3.54.0) in R (version 4.0.3, R Foundation for Statistical Computing, Vienna, Austria) was used to perform differential gene expression analysis between the control and TRPV1 overexpression groups. Genes with $|$log₂-fold change (FC)$|$ $\ge$1 and p 0.05 were considered differentially expressed genes (DEGs). Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed using the Database for Annotation, Visualization, and Integrated Discovery (DAVID; https://david.ncifcrf.gov/). Both the GO and KEGG pathway enrichment results were deemed statistically significant when the p value was less than 0.05.

A total of 1576 Mito-RGs were collected from the Molecular Signature Database (MSigDB, https://www.gsea-msigdb.org/gsea/msigdb/). The overlapping genes between 249 TRPV1-related genes and 1576 Mito-RGs were identified using the Venn online graphical tool (https://bioinformatics.psb.ugent.be/webtools/Venn/).

The sequencing datasets (containing the control group and the TRPV1 overexpression group) were used to assess the expression levels of seven overlapping genes. The raw counts were normalized with DESeq2 (version 1.30.1) in R, and the log₂-transformed normalized expression values were visualized with pheatmap (version 1.0.12). The correlation matrix of the seven genes was calculated and visualized using a corrplot (version 0.84). The Spearman correlation coefficient between TRPV1 and each of the seven Mito-RGs was calculated in R using the Spearman method. Scatter plots with trend lines were generated using ggplot2 (version 3.3.3) to visualize the correlations.

Mitochondria and the cytosol were isolated from cardiac tissue via a Tissue Mitochondria Isolation Kit (#C3606, Beyotime, Shanghai, China) and a Tissue Cytoplasm and Nucleus Separation Kit (#NT-032, Invent, Wilmington, DE, USA) according to the manufacturer’s instructions. Similarly, mitochondria and cytosol from AC16 cells were extracted using the Cell Mitochondria Isolation Kit (#C3601, Beyotime, Shanghai, China) and the Cell Cytoplasm and Nucleus Separation Kit (#BB-36021, Beibokit, Shanghai, China) following the manufacturer’s protocols.

Proteins in cardiac tissue and AC16 cells (mitochondria) were extracted with radioimmunoprecipitation assay (RIPA) buffer (#P0013, Beyotime, Shanghai, China). After the samples were homogenized, they were incubated on ice for 30 minutes and then centrifuged at 12,000 $\times$g for 15 minutes (4 °C). The supernatant was then collected. Protein quantification was performed with a bicinchoninic acid (BCA) kit (P0012, Beyotime, Shanghai, China). The samples (30 µg) were denatured in loading buffer (100 °C, 10 min) and resolved by 10% sodium dodecyl sulfate‒polyacrylamide gel electrophoresis (SDS‒PAGE) (#P0690, Beyotime, Shanghai, China) (stacking: 80 V; separating: 120 V). Proteins were electroblotted onto methanol-activated polyvinylidene fluoride (PVDF) (#FFP24, Beyotime, Shanghai, China) membranes (300 mA, 90 min). After being blocked with 5% milk-Tris-buffered saline with Tween 20 (TBST) (60145ES76, YEASEN, Shanghai, China) (2 h), the membranes were probed with primary antibodies (Table 1) overnight at 4 °C, followed by incubation with HRP-conjugated secondary antibodies (goat anti-mouse IgG (ZB-2305, Zs-BIO, Beijing, China), goat anti-rabbit IgG (ZB-2301), Zs-BIO, Beijing, China) (1:10,000, 1 h). The membrane was exposed to enhanced chemiluminescence detection reagent (KGC4902, Keygene, Nanjing, Jiangsu, China), and gel imaging was performed using gel imaging equipment (Bio-Rad, Hercules, CA, USA) for protein visualization. The grayscale density of the protein bands was quantified using ImageJ software (version 1.8.0; National Institutes of Health, Bethesda, MD, USA). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and voltage-dependent anion channel 1 (VDAC1) were used as cytosolic and mitochondrial markers, respectively.

TRIzol reagent (#G3013-100ML; Life Technologies, Carlsbad, CA, USA) was used to extract total RNA from AC16 cardiomyocytes. An OD1000+ ultramicrospectrophotometer (Nanjing Wuyi Technology Co., Ltd., Nanjing, Jiangsu, China) was used to measure the concentration and purity of the RNA, and 1% agarose gel electrophoresis was used to confirm the integrity of the RNA. With a reverse transcription kit with dsDNAse (#BL699A; Biosharp, Hefei, Anhui, China) and a PTC-200 heat cycler (Bio-Rad, Hercules, CA, USA), first-strand cDNA was generated. The reaction proceeded for 30 min at 37 °C followed by 5 min at 85 °C. qRT‒PCR was carried out with an ABI StepOne Plus Real-Time PCR System with Taq SYBR Green qPCR Premix Universal (#EG20117M, Best-enzymes, Lianyungang, Jiangsu, China). A 20 µL PCR mixture was prepared with SYBR Green premix (10 µL), primers (0.4 µL each, 10 µM, Sangon Biotech, Shanghai, China), cDNA (3 µL), and Diethylpyrocarbonate-treated Water (DEPC-H₂O) (6.2 µL, #D1007; Generay Biotech, Shanghai, China). Initial denaturation at 95 °C for 30 s was followed by 40 cycles of 95 °C for 15 s and 60 °C for 30 s as part of the amplification process. Melting curve analysis was used to establish the specificity of amplification. The expression levels of TRPV1 and mitochondria-related genes were analyzed. The internal reference gene was GAPDH. Table 2 lists the specific sequences of the primers used. The relative expression of genes was calculated using the 2^{-Δ⁢Δ⁢CT} method.

A bicinchoninic acid (BCA) protein assay kit (#BL521A; Biosharp, Hefei, Anhui, China) was used to measure the protein concentration of the homogenate after AC16 cells from various treatment groups were obtained and homogenized in ice-cold phosphate-buffered saline (PBS, #G4207; Servicebio, Wuhan, Hubei, China). The malondialdehyde (MDA), superoxide dismutase (SOD), glutathione (GSH), and ferrous iron (Fe²⁺) levels associated with lipid peroxidation were subsequently measured using an MDA assay kit (#A003-1-2), a SOD kit (#A001-3-1), a reduced GSH assay kit (#A006-2-1), and a ferrous ion colorimetric kit (#E-BC-K773-M; Elabscience, Wuhan, Hubei, China). All unlabeled kits were provided by Nanjing Jiancheng Bioengineering Institute (Nanjing, Jiangsu, China). Every test was carried out in compliance with the manufacturer’s instructions. The absorbance was measured at 532 nm (MDA), 450 nm (SOD), 420 nm (GSH), and 593 nm (Fe²⁺) using a microplate reader (Bio-Rad, Hercules, CA, USA). The quantitative methods were performed according to the instructions provided in the manual.

The intracellular Ca²⁺ concentration of AC16 cells was analyzed using a calcium assay kit (#40776ES50, YEASEN, Shanghai, China) according to the manufacturer’s instructions. In brief, AC16 cells were extracted via centrifugation at 1000 rpm for 5 minutes and then rinsed with PBS. The cells were then incubated with 5 µM Rhod-2 AM (prepared in DMSO with a 2–5 mM stock solution) for 30 minutes at 37 °C in the dark. The cells were washed twice with PBS, followed by centrifugation (1000 rpm, 5 min). Fluorescence images were captured with a Leica DMI3000B inverted fluorescence microscope (Leica, Wetzlar, Hesse, Germany) at 200$\times$ magnification. Calcium-bound Rhod-2 AM fluorescence was detected using an appropriate red fluorescence filter, and the green fluorescent protein (GFP) signal was monitored in the green channel. Quantification was performed using ImageJ software.

Following the manufacturer’s instructions, cell viability was evaluated using the Calcein AM/PI Live/Dead Viability/Cytotoxicity Assay Kit (#C2015M; Beyotime, Shanghai, China). After being washed with PBS, AC16 cells from various treatment groups were treated with the working solution for 30 minutes at 37 °C in the dark. PI (red fluorescence) was used to identify dead cells, whereas Calcein AM (green fluorescence) was used to identify live cells. Fluorescence images were taken at 200$\times$ magnification using an Olympus CKX53 inverted microscope (Olympus, Tokyo, Japan), and at least three random fields of view were taken for each group. Images were acquired using appropriate filter sets for the PI (red) and Calcein AM (green) channels and then merged for analysis. Cell viability was quantified using ImageJ software.

MitoSOX Red (HY-D1055, MCE, Monmouth Junction, NJ, USA) was used to assess the levels of ROS in the mitochondria. AC16 cells were plated, rinsed with PBS, and then incubated with 5 µM MitoSOX Red working solution at 37 °C for 30 minutes in the dark. Nuclei were counterstained with 4^′,6-diamidino-2^′-phenylindole (DAPI, #C1002; Beyotime, Shanghai, China). Fluorescence images were captured at 200$\times$ magnification using a Leica DMI3000B inverted fluorescence microscope. At least three random fields were taken for each group. Red fluorescence indicates mitochondrial superoxide levels, while DAPI staining (blue) was used for nuclear visualization. The ROS level was quantified using ImageJ software.

Apoptotic cells were detected via a one-step TUNEL kit (#ECK-A320; Elabscience, Wuhan, Hubei, China). The samples were fixed in 4% paraformaldehyde (CF189021, Solarbio, Beijing, China) (30 min) and permeabilized with 0.2% Triton X-100 (P0096, Beyotime, Shanghai, China) (5 min) in PBS. After washing, the samples were incubated in the dark (37 °C, 1 h) with a TdT/luciferin-dutP reaction mixture. Nuclear staining was performed with DAPI (25 µg/mL, 10 min). Images were acquired at 100$\times$ magnification, and TUNEL-positive cells (green) were quantified relative to total DAPI signals (blue) using ImageJ software.

Data analysis was performed with R (version 4.0.3, R Foundation for Statistical Computing, Vienna, Austria) and GraphPad Prism (version 8.0.0, GraphPad Software, San Diego, CA, USA). The results represent the mean $\pm$ standard deviation (standard deviation) from three independent experiments. Differences between pairs were evaluated by Student’s t-test, whereas multigroup analyses employed one-way analysis of variance (ANOVA) with Tukey’s post hoc comparisons. Statistical significance was set at p 0.05.

AC16 human cardiomyocytes were treated with 1 µmol/L Ang II to simulate pathological cardiac conditions. TRPV1 mRNA expression was significantly downregulated according to qRT‒PCR (p 0.0001, Fig. 1A). Western blot analysis revealed that TRPV1 protein levels were similarly reduced following Ang II stimulation (p 0.001, Fig. 1B). To investigate the effect of TRPV1, we first overexpressed TRPV1 in AC16 cells (p 0.0001, Fig. 1C,D). Next, the intracellular Ca²⁺ level in AC16 cells was assessed using fluorescent probes. Compared with that in the control group, TRPV1 overexpression significantly reduced the intracellular Ca²⁺ level (p 0.05, Fig. 2A). OE-TRPV1 increased the level of TRPV1 in the HF cell model, which was validated via qRT‒PCR and Western blot analyses (p 0.0001, Fig. 2B,C).

Fig. 1.

Analysis of transient receptor potential cation channel subfamily V member 1 (TRPV1) level in AC16 cells induced by angiotensin II (Ang II) and overexpressing TRPV1. (A) Quantitative real-time polymerase chain reaction (qRT–PCR) detected the TRPV1 mRNA level in AC16 cells induced by Ang II. (B) Western blot analysis of TRPV1 protein levels in AC16 cells after Ang II induction was performed and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used for normalization. (C) qRT–PCR detected the TRPV1 mRNA level of AC16 cells in overexpression negative control (OE-NC) and overexpression TRPV1 (OE-TRPV1) groups. (D) Western blot analysis of TRPV1 protein levels of AC16 cells in OE-NC and OE-TRPV1 groups was performed, and GAPDH was used for normalization. Each experiment was independently repeated three times. Data are presented as mean $\pm$ standard deviation. ***p 0.001, ****p 0.0001.

Fig. 2.

Effects of TRPV1 overexpression on intracellular Ca²⁺ levels and TRPV1 expression in heart failure (HF) cell model. (A) Representative fluorescence microscopy images showing intracellular calcium ion (Ca²⁺) levels (red), green fluorescent protein (GFP) expression (green), and merged images in OE-NC and OE-TRPV1 groups. Magnification: 200$\times$. Scale bar = 50 µm. (B) qRT–PCR was used to detect the TRPV1 mRNA level in HF cell model transfected with OE-TRPV1. (C) Western blot detection of TRPV1 protein level in HF cell model transfected with OE-TRPV1, and GAPDH was used for normalization. Each experiment was independently repeated three times. Data are presented as mean $\pm$ standard deviation. *p 0.05, ****p 0.0001.

To investigate whether TRPV1 affects cell viability under Ang II stimulation, we performed PI/Calcein AM double staining. Compared with the control treatment, Ang II treatment dramatically inhibited cell viability (p 0.001, Supplementary Fig. 1). Overexpression of TRPV1 reversed the Ang II-induced decrease in cell viability (p 0.05, Supplementary Fig. 1). Next, the intracellular ROS levels were examined using fluorescence imaging. A comparison of the Ang II-treated group with the control group revealed a substantial increase in ROS generation (p 0.0001, Supplementary Fig. 2). Notably, overexpression of TRPV1 significantly attenuated the increase in ROS levels induced by Ang II (p 0.001, Supplementary Fig. 2). These results suggest that by preventing oxidative stress, TRPV1 overexpression shields cardiomyocytes from Ang II-induced harm.

According to the screening criteria for DEGs, 195 upregulated genes and 54 downregulated genes were screened from the sequencing data (Supplementary Fig. 3). The results of the biological process (BP), cellular component (CC), and molecular function (MF) enrichment analyses revealed that the upregulated genes were associated primarily with nucleosomes (GO: 0000786), the deoxyribonucleic acid (DNA) packaging complex (GO: 0044815), protein heterodimerization activity (GO: 0046982), DNA binding (GO: 0003677), and nucleosome assembly (GO: 0006334) (Supplementary Fig. 4A). The downregulated genes were related to the extracellular region (GO: 0005576), cytokine activity (GO: 0005125), cell adhesion (GO: 0007155), and the cell surface receptor signaling pathway (GO: 0007166) (Supplementary Fig. 4B). KEGG enrichment pathway analysis revealed that the pathways enriched with the upregulated genes included the apelin signaling pathway, herpes simplex virus 1 infection, and alcoholism (Supplementary Fig. 5A). The downregulated genes were associated with KEGG pathways such as the chemokine signaling pathway, NOD-like receptor signaling pathway, influenza A pathway, and the interleukin-17 (IL-17) signaling pathway (Supplementary Fig. 5B).

To explore the possible molecular mechanisms underlying the relationship between TRPV1 and mitochondrial function, we performed an intersection analysis of TRPV1-related genes and Mito-RGs. Venn diagram analysis revealed that there were 7 overlapping genes between the TRPV1-related genes (249 genes) and Mito-RGs (1576 genes) (Fig. 3A). Among these overlapping genes, heatmap analysis revealed different expression patterns between the groups (Fig. 3B). Further analysis revealed differences in correlations between different genes (Fig. 3C). For example, solute carrier family 25 member 24 (SLC25A24) and G0/G1 switch gene 2 (G0S2) were strongly positively correlated, whereas SLC25A24 was significantly negatively correlated with tubulin polymerization promoting protein (TPPP) and lactate dehydrogenase A like 6B (LDHAL6B) (p 0.01, Fig. 3C). These findings suggest that there is a complex regulatory network between TRPV1 and Mito-RGs, which may contribute to the role of TRPV1 in mitochondrial function.

Fig. 3.

Analysis of TRPV1 and mitochondria-related gene expression patterns. (A) Venn diagram showing the overlap between TRPV1-related genes and mitochondria-related genes. (B) Heatmap visualization of expression patterns for the seven overlapping genes across different experimental groups. Color scale represents normalized expression values, with red indicating upregulation and blue indicating downregulation. (C) Correlation matrix showing the relationships among the seven overlapping genes. The size and color intensity of squares represent correlation strength (–1 to 1), with blue indicating negative correlations and red indicating positive correlations. Statistical significance is denoted by asterisks (*p 0.05, **p 0.01, ***p 0.001). LDHAL6B, lactate dehydrogenase A like 6B.

Further analysis confirmed the correlation between TRPV1 and seven genes (Supplementary Fig. 6A–G). Strong correlations were observed between TRPV1 and both LDHAL6B (p = 0.03, r = 0.86) and SFXN2 (p = 0.04, r = 0.83) (Supplementary Fig. 6C,E). In vitro experiments verified the expression of these genes in AC16 cells treated with Ang II. After TRPV1 overexpression, qRT‒PCR analysis revealed that the expression of schindler disease (SDS), SFXN2, creatine kinase mitochondrial 2 (CKMT2), TPPP, and LDHAL6B was dramatically greater than that in the Ang II+OE-NC group (p 0.001), whereas the G0S2 and SLC25A24 levels were not significantly altered (Supplementary Fig. 6H). Among these genes, LDHAL6B and SFXN2 presented the strongest correlations with TRPV1. The role of SFXN2 in mitochondrial function has been previously established. However, its specific involvement in mitochondrial function during HF remains unclear. Therefore, we selected SFXN2 for further analysis.

To assess the therapeutic potential of TRPV1 in HF, we generated a model of cardiac dysfunction caused by pressure overload using TAC. Cardiac function was assessed using echocardiography. Compared with those in the sham group, the cardiac function of the mice in the TAC-treated group was considerably worse, as evidenced by a reduction in FS and EF (p 0.0001, Supplementary Fig. 7A–C). Notably, compared with TAC+AAV9-CTNT-control, AAV9-mediated cardiac-specific TRPV1 overexpression (TAC+AAV9-CTNT-TRPV1) improved cardiac function, as shown by a moderate increase in EF and FS (p 0.0001, Supplementary Fig. 7A–C). Compared with those in the sham group, TAC dramatically decreased the expression of TRPV1 and SFXN2 while increasing the protein levels of atrial natriuretic peptide (ANP) and beta-myosin heavy chain ($\beta$-MHC) (p 0.0001, Fig. 4A–E). Moreover, compared with those in the TAC+AAV9-CTNT-control group, TRPV1 overexpression significantly increased TRPV1 and SFXN2 protein levels but significantly reduced the expression of the cardiac stress markers ANP and $\beta$-MHC (p 0.05, Fig. 4A–E). These results suggest that TRPV1 overexpression can alleviate pathological cardiac remodeling in HF caused by pressure overload.

Fig. 4.

TRPV1 overexpression regulates the expression of HF-related markers in the TAC-induced HF model. (A) Representative Western blot images showing protein expression of HF-related markers (atrial natriuretic peptide (ANP), beta-myosin heavy chain ($\beta$-MHC), TRPV1, and sideroflexin 2 (SFXN2)) in different experimental groups (sham, transverse aortic constriction (TAC), TAC+adeno-associated virus 9 (AVV9)-cardiac troponin T (CTNT)-control, TAC+AVV9-CTNT-TRPV1), with GAPDH as a loading control. (B–E) Quantitative analysis of protein expression levels of TRPV1 (B), SFXN2 (C), ANP (D), and $\beta$-MHC (E) with GAPDH as a standard. Each experiment was independently repeated three times. Data are presented as mean $\pm$ standard deviation. *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001.

Using the TAC mouse model, we examined how TRPV1 regulates the levels of autophagy-related proteins. Compared with sham surgery, TAC surgery increased the expression level of sequestosome 1 (SQSTM1) in the cytoplasm while drastically lowering the amounts of Parkin and PTEN-induced kinase 1 (PINK1) (p 0.05, Fig. 5A,B). TRPV1 overexpression decreased the amount of SQSTM1 and restored Parkin expression (p 0.0001), but it had no discernible effect on PINK1 expression (Fig. 5A,B). In the mitochondrial fraction, TAC increased Parkin and PINK1 levels while decreasing SQSTM1 levels (p 0.001, Fig. 5A,C). Overexpression of TRPV1 decreased the levels of Parkin and PINK1 (p 0.001, Fig. 5A,C). Autophagy is a lysosome-dependent degradation pathway that promotes ferroptosis by degrading iron storage proteins such as ferritin, leading to increased intracellular iron levels and subsequent lipid peroxidation. Therefore, on the basis of the connection between autophagy and ferroptosis, we examined the expression of glutathione peroxidase 4 (GPX4), a ferroptosis marker. Western blot analysis revealed that TAC significantly reduced GPX4 protein levels (p 0.0001, Supplementary Fig. 8A,B). Compared with the TAC+AAV9-CTNT-control group, TRPV1 overexpression significantly upregulated GPX4 expression, indicating that TRPV1 has a protective effect against TAC-induced ferroptosis (p 0.001, Supplementary Fig. 8A,B). These findings suggest that TRPV1 may respond to TAC-induced cardiac stress by regulating autophagy and ferroptosis pathways.

Fig. 5.

TRPV1 overexpression regulates the expression of mitophagy-related proteins in the cytoplasm and mitochondria in the TAC-induced HF model. (A) Representative Western blot images showing the protein expression of mitophagy markers (Parkin, PTEN-induced kinase 1 (PINK1), and sequestosome 1 (SQSTM1)) in the cytoplasm and mitochondria fractions. GAPDH and voltage-dependent anion channel 1 (VDAC1) were used as loading controls for the cytoplasm and mitochondria fractions, respectively. (B) Quantification of the cytoplasmic protein levels of Parkin, PINK1, and SQSTM1 in the experimental groups, with GAPDH as the standard. (C) Quantification of the mitochondrial protein levels of Parkin, PINK1, and SQSTM1 in the experimental groups, with VDAC1 as the standard. n = 8 per group. Data are presented as mean $\pm$ standard deviation. *p 0.05, ***p 0.001, ****p 0.0001.

Previously, we analyzed and verified that TRPV1 and SFXN2 interact in HF, but the specific mechanism was unclear. Therefore, we continued to analyze this phenomenon through in vitro experiments. To analyze the role of SFXN2, we designed three siRNAs targeting SFXN2. The qRT‒PCR results demonstrated that all three siRNAs effectively reduced SFXN2 expression in cells, with Si-SFXN2#3 (referred to as Si-SFXN2) exhibiting the highest knockdown efficacy (p 0.0001, Supplementary Fig. 9A). Western blot analysis further confirmed that Si-SFXN2 significantly decreased SFXN2 protein levels (p 0.01, Supplementary Fig. 9B,C). First, qRT‒PCR experiments revealed that, compared with OE-NC, TRPV1 overexpression significantly increased SFXN2 mRNA levels, but when TRPV1 was combined with Si-SFXN2, its mRNA level was reduced (p 0.0001, Fig. 6A). These results were also confirmed at the protein level, and SFXN2 knockdown effectively reversed TRPV1-induced SFXN2 upregulation (p 0.0001, Fig. 6B,C). Next, we assessed ferroptosis-related markers. The results revealed that TRPV1 overexpression significantly increased GPX4 protein levels, whereas SFXN2 knockdown partially reversed this effect (p 0.05, Fig. 6D,E). Consistent with these findings, TRPV1 overexpression significantly reduced the MDA and Fe²⁺ levels while increasing the GSH content and SOD activity (p 0.0001, Fig. 6F,I). Importantly, these effects were partially reversed after SFXN2 knockdown (p 0.001, Fig. 6F–I). Taken together, these results suggest that TRPV1 regulates ferroptosis through an SFXN2-dependent pathway.

Fig. 6.

TRPV1 regulates ferroptosis in HF through an SFXN2-dependent mechanism. (A) qRT–PCR was used to detect the changes in the mRNA level of SFXN2 after TRPV1 was overexpressed alone or in combination with down-expressed SFXN2 in Ang II-induced AC16 cardiomyocytes. (B,C) Western blot was used to detect the protein level of SFXN2 after TRPV1 was overexpressed alone or in combination with down-expressed SFXN2 in Ang II-induced AC16 cardiomyocytes. (D,E) Western blot was used to detect the protein level of glutathione peroxidase 4 (GPX4) in Ang II-induced AC16 cardiomyocytes after TRPV1 was overexpressed alone or in combination with down-expressed SFXN2. (F–I) Commercial kits were used to detect the changes in the levels of malondialdehyde (MDA), glutathione (GSH), superoxide dismutase (SOD), and ferrous iron (Fe²⁺) after TRPV1 was overexpressed alone or in combination with down-expressed SFXN2 in Ang II-induced AC16 cardiomyocytes. Each experiment was independently repeated three times. Data are presented as mean $\pm$ standard deviation. *p 0.05, ***p 0.001, ****p 0.0001. Si-NC, small interfering RNA negative control; Si-SFXN2, small interfering RNA targeting SFXN2.

Next, we further analyzed the mechanism of action of SFXN2 and TRPV1 in an HF model. TUNEL staining demonstrated that TRPV1 expression alone resulted in a substantial decrease in apoptosis compared with that in the OE-NC group (p 0.0001, Fig. 7A). When SFXN2 knockdown reversed the inhibitory effect of TRPV1 overexpression on cell apoptosis (p 0.0001, Fig. 7A). Subsequent Western blot experiments were used to measure the levels of autophagy-related proteins in the cytoplasm and mitochondria after the transfection of TRPV1-overexpressing and SFXN2-knockdown vectors. Compared with those in the OE-NC group, TRPV1 overexpression significantly increased Parkin levels in the cytoplasm while reducing SQSTM1 levels (p 0.05, Fig. 7B,C). SFXN2 knockdown reversed these TRPV1-mediated effects, specifically reducing cytoplasmic Parkin and increasing SQSTM1 levels (p 0.01, Fig. 7B,C). In mitochondria, TRPV1 overexpression resulted in a substantial reduction in the protein expression of Parkin and PINK1 but increased SQSTM1 expression compared with that in the OE-NC group (p 0.0001, Fig. 7B,D). Conversely, SFXN2 knockdown significantly reversed these changes, resulting in increased mitochondrial accumulation of Parkin and PINK1, while decreasing SQSTM1 levels (p 0.0001, Fig. 7B,D).

Fig. 7.

TRPV1 inhibits mitochondrial autophagy by regulating PINK1-Parkin-SQSTM1 through upregulating SFXN2. (A) In Ang II-induced AC16 cardiomyocytes, terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) staining was used to detect the apoptosis levels in different experimental groups (OE-NC; OE-TRPV1; OE-TRPV1+small interfering RNA negative control (Si-NC); OE-TRPV1+small interfering RNA targeting SFXN2 (Si-SFXN2)). Left: Representative fluorescence images showing TUNEL-positive cells (green) and nuclear 4^′,6-Diamidino-2^′-phenylindole (DAPI) staining (blue). Right: Quantification of apoptosis rates among experimental groups. Magnification: 100$\times$. Scale bar = 50 µm. (B) Western blot images showing the protein levels of Parkin, PINK1, and SQSTM1 in the cytoplasmic and mitochondrial fractions. GAPDH and VDAC1 were used as loading controls for the cytoplasmic and mitochondrial fractions, respectively. (C) Quantification of cytoplasmic protein levels with GAPDH as the standard. (D) Quantification of mitochondrial protein levels with VDAC1 as the standard. Each experiment was independently repeated three times. Data are presented as mean $\pm$ standard deviation. *p 0.05, **p 0.01, ****p 0.0001.