Mitochondria-derived vesicles with bioenergetic units from brown adipose tissue attenuate cardiac remodeling post-myocardial infarction

Figures

• 

  Fig. 1. BAT transplantation limits post-MI cardiac remodeling. a Mouse BAT weight proportion over the course of MI (mean ± s.d., n = 4 mice for each time point, repeated two-way ANOVA with Holm-Sidak’s multiple-comparisons test). b Experimental timeline for MI induction, adipose tissue pericardial transplantation, and outcome assessments. Echocardiography was performed at baseline (day -1) and on days 7, 14, and 28. Hearts were harvested on day 28 for Masson’s trichrome and Sirius Red staining. Created in BioRender. Shi, T. (2026) https://BioRender.com/8asl17d.c, d Representative B-mode and M-mode echocardiography of BAT-receiving mice at baseline and days 7, 14, and 28 (c). Ejection fraction and fractional shortening were quantified (d) (mean ± s.d., n = 7 mice per group, repeated two-way ANOVA with Holm-Sidak’s multiple-comparisons test). e, f Representative Masson and Sirius Red staining images of mouse hearts at day 28. Fibrosis percentage was quantified (mean ± s.d., n = 7 mice per group, one-way ANOVA with Bonferroni correction). Scale bar: 2 mm. Source data are provided as a Source Data file.

• 

  Fig. 2. BAT-derived MDVs transfer to cardiac macrophages in infarcted myocardium. a Schematics of pericardial BAT transplantation from ZsGreen-ubiquitous fluorescent reporter donor mice (B6-G/Rfl/+) into wild-type MI recipient mice. Created in BioRender. Shi, T. (2026) https://BioRender.com/755v3rv. b Representative confocal images of left ventricles 4 days after MI in mice with or without BATZsGreen transplantation, stained for sarcomeric α-actinin (red) and nuclei (DAPI, blue). 3D reconstruction of the infarct zone of BATZsGreen-receiving mice. Scale bar: 100 μm. The experiment was independently repeated 3 times with similar results. c, d Representative confocal images of left ventricles in BATZsGreen-receiving mice at sham or MI 4 d, stained for TOMM20 or Cytochrome C or PDH or ATP5A1 (red), and nuclei (DAPI, blue). Co-localization analysis of ZsGreen⁺ BATosomes with mitochondrial markers in the infarct myocardium. Scale bar: 100 μm. The experiment was independently repeated 3 times with similar results. e Schematic illustration of the BAT transplant from BAT mitochondria-labeled mice to wild-type (WT) MI mice. AAV8-MTS-eGFP was injected in situ into interscapular BAT to generate…

• 

  Fig. 3. BAT-derived MDVs mediate cardioprotection via EV release post-MI. a Schematic of MDVs isolation from mouse adipose tissue. Created in BioRender. Shi, T. (2026) https://BioRender.com/xnpoj8f. Purified MDVs were visualized by negative-stain TEM. Scale bar, 100 nm. b Protein quantification of MDVs isolated from WAT or BAT, normalized by tissue weight (mean ± s.d., n = 3 mice per group, one-way ANOVA with Bonferroni correction). Mice have been housed at 22 °C (control) or 4 °C for 14 days before tissue harvest. c Western blots of purified mitochondria and MDVs from mouse adipose tissue. The experiment was independently repeated on 3 mice with similar results. Representative B-mode and M-mode echocardiography of adipose tissue-derived MDV-treated mice at baseline and days 7, 14, and 28 (d). Ejection fraction and fractional shortening were quantified (e) (mean ± s.d., n = 5 mice per group, repeated two-way ANOVA with Holm-Sidak’s multiple-comparisons test). f, g Representative Masson and Sirius Red staining images of MDV-treated mouse hearts at day 28. Fibrosis percentage was quantified (mean±s.d., n = 5 mice per group, one-way ANOVA with Bonferroni correction). Scale bar: 2 mm. h…

• 

  Fig. 4. Heart DAMPs engage a BECN1-VPS35-dependent signaling in BAT to drive TOMM20⁺PDH⁻ MDV biogenesis and release. a Relative mRNA expression of genes associated with BAT thermogenesis and mitochondria dynamics in BAT from sham-operated and MI mice (mean ± s.d., n = 6 mice (Ucp1 and Pparg), 3 mice (other genes) per group, two-tailed unpaired Student’s t test). b Representative images of differentiated brown adipocytes stained by Mitotracker (500 nM) and treated with heart DAMPs (20 μg/mL). Scale bar: 100 μm. Data are representative of 3 independent experiments with similar results. c Flow cytometric histogram of Mitotracker-PE+ MDVs release of NE or CCCP or heart DAMPs-treated brown adipocytes. d Quantitative results of MitoTracker+ vesicles percentage and counts (mean ± s.e.m., n = 3 independent experiments, paired one-way ANOVA with Bonferroni correction). e Venn diagram showing genes associated with vesicle-mediated transport pathways and mitochondrial-associated processes. The overlapping region (n = 38, list in the Supplementary Data 1) highlights genes in both pathways. f Time course of VPS35 and BECN1 protein expression in BAT after MI (mean ± s.d., n = 3 mice per group, one-way…

• 

  Fig. 5. Becn1-knockdown ablates MDV-mediated cardioprotection of BAT-derived MDVs. a Schematic illustration of the BAT transplant from Becn1+/– or WT mice (mitochondria labeled) to WT MI mice. Created in BioRender. Shi, T. (2026) https://BioRender.com/wadta8v. b Protein quantification of MDVs isolated from WT or Becn1+/– BAT, normalized by BAT tissue weight (mean ± s.d., n = 3 mice per group, two-tailed unpaired Student’s t test). c Representative confocal images of BATmtGFP-receiving mice at MI 4 d, stained for GFP (green) and nuclei (DAPI, blue). Scale bar: 100 μm. Mt-EGFP+ cell percentage was quantified (mean ± s.d., n = 4 mice per group, two-tailed unpaired Student’s t test). d Representative B-mode and M-mode echocardiography of WT or Becn1+/– BAT-receiving mice at baseline and days 7, 14, and 28. Ejection fraction and fractional shortening were quantified (mean ± s.d., n = 6 mice per group, repeated two-way ANOVA with Holm-Sidak’s multiple-comparisons test). e Representative Masson and Sirius Red staining images of WT or Becn1+/– BAT-receiving mouse hearts at day 28. Fibrosis percentage was quantified (mean ± s.d., n = 6 mice per group, one-way ANOVA with Bonferroni correction).…

• 

  Fig. 6. Becn1 deficiency alters the respiratory-chain cargo of BAT-derived MDVs. a Quantitative proteomic analysis of MDVs from WT and Becn1+/– BAT. Proteomics analysis of WT and Becn1+/– BAT MDV protein cargos. The top 30 abundant proteins were calculated by the average intensity of all samples (median + least to max, n = 3 mice per group). b Comparative analysis of protein levels of electron-transport-chain Complex I-V key subunits (protein with intensity>20000 was recognized as complex key subunit) in MDVs from WT and Becn1+/– BAT (mean ± s.d., n = 39 subunits for Complex I, 4 proteins for Complex II, 9 subunits for Complex III, 21 subunits for Complex IV, and 11 subunits for Complex V, two-tailed unpaired Student’s t test). c Heatmap of Complex V key subunits in MDVs from WT and Becn1+/– BAT. d Schematic representation of respiratory-chain composition in BAT-derived MDVs from WT versus Becn1+/– mice, summarizing differences in the relative contribution of complexes I-V inferred from the proteomic analysis. Created in BioRender. Shi, T. (2026) https://BioRender.com/djhc1u2. Source data are provided as a Source Data file.

• 

  Fig. 7. Complex V-dependent effects of BAT MDVs on macrophage polarization. a RT-qPCR analysis of inflammatory and reparative cytokine genes in iBMDMs during DAMPs priming (mean ± s.e.m., n = 4 independent experiments, one-way ANOVA with Bonferroni correction). b Radar plot showing relative expression of inflammation-related genes in iBMDMs. c RT-qPCR analysis of reparative cytokine genes in iBMDM co-incubated with WT or Becn1+/– BAT MDVs (mean ± s.e.m., n = 4 independent experiments, one-way ANOVA with Bonferroni correction). d Representative western blots of macrophages treated with DAMPs for 24 h in the presence of PBS, WT-MDVs, or Becn1+/–-MDVs (n = 3 independent experiments). e IL-4 and VEGF concentrations in supernatants from macrophages treated as in (d) (mean±s.e.m., n = 3 independent experiments, one-way ANOVA with Bonferroni correction). f Mitochondria membrane potential analysis of macrophages co-incubated with WT or Becn1+/–-MDVs. (mean ± s.e.m., 6 technical replicates, n = 3 independent experiments with similar results, one-way ANOVA with Bonferroni correction). g Purified WT-MDVs and Becn1+/–-MDVs were incubated with oligomycin. NADH oxidation rate (A340/min) was measured…

• 

  Fig. 8. Macrophage complex V level is positively associated with cardiac functions of patients with HF. Complex V expression score in endomyocardial biopsies from normal control and patients with HF across different cell types (a), in fibroblasts (b), in lymphocytes (c), and in macrophages (d). The scores were compared (mean ± s.d., n = 21 (normal control), n = 29 (patients with HF), two-tailed unpaired Student’s t test). Correlation between cardiac functions (LVEF, left ventricular ejection fraction; LVMI, left ventricular mass index) and complex V expression score in all cell types (e), fibroblasts (f), lymphocytes (g), and macrophages (h) of normal control and patients with HF (Solid lines indicate linear regression curves, and shaded bands represent the 95% confidence intervals, n = 47 for LVEF, 49 for LVMI, two-sided Pearson correlation analysis). Source data are provided as a Source Data file.

• 

  Fig. 9. BAT-derived bioenergetic vesicles promote cardiac repair after MI. MI triggers heart-to-brown adipose tissue signaling that induces the production and release of MDVs enriched in mitochondrial respiratory chain proteins. Through Becn1-dependent recruitment of VPS35 to mitochondria, mitochondrial proteins are packaged and secreted via extracellular vesicle trafficking pathways, then delivered to cardiac macrophages. The bioenergetic cargo enhances oxidative phosphorylation and anti-inflammatory programming, thereby limiting fibrosis and improving cardiac repair. Becn1 deficiency reduces complex V enrichment in these vesicles and abolishes cardioprotection. Created in BioRender. Shi, T. (2026) https://BioRender.com/7lxm9ok.