Small Extracellular Vesicle External Surface Adiponectin-Mediated Adipocytes/Cardiomyocytes Communication in Diabetic Ischemic Heart Failure

Abstract

Background:

Mortality from acute myocardial infarction (MI) has declined significantly in the past decade for nondiabetic patients. However, both morbidity and mobility of ischemic heart failure (IHF) persistently escalate in the diabetic population via incompletely understood mechanisms. Recent studies demonstrated that small extracellular vesicles (sEV) released from nondiabetic and diabetic adipocytes exert opposite effects on acute myocardial ischemia and reperfusion (MI/R) injury. However, whether and how adipocyte sEV may protect against post-MI remodeling and IHF, and more importantly, whether and how diabetes may impair this protective effect, remains unknown.

Methods and Results:

sEV were isolated from epididymal fat pads of nondiabetic animals and intramyocardially injected in nondiabetic or diabetic hearts subjected to MI (90 minutes of MI/4 weeks of reperfusion). sEV treatment significantly attenuated post-MI cardiac remodeling and improved cardiac function in nondiabetic mice. However, the protection was not observed in diabetic hearts. In adult cardiomyocytes isolated from nondiabetic hearts, sEV rapidly (15 min) activated cell salvage kinases (ERK, AMPK, and ACC) and suppressed oxidative stress-induced cell death, suggesting sEV external surface molecules are responsible for the observed cytoprotection. The Exo-Flow (a technology detecting sEV external surface molecules) demonstrated that adiponectin (APN) is enriched on the sEV external surface. The sEV from APN knockout mice or APN neutralization antibody pre-treated sEV failed to protect the heart against IHF. Moreover, the cardioprotective effects of sEV were abolished in AdipoR1-deficient mice (the primary receptor for APN signaling in the heart) or in mice overexpressing GRK2 (a kinase that phosphorylates and inactivates AdipoR1). Finally, diabetes significantly increased cardiac GRK2 expression and AdipoR1 phosphorylation, which prevented sEV from exerting its beneficial effects. Restoring AdipoR1 function by knock-in a mutated phosphorylation-resistant AdipoR1 (AdipoR1^S205A) via AAV9-mediated gene delivery rescued adipocyte sEV cardioprotection in diabetic mice.

Conclusions:

Our study reveals that APN is enriched on adipocyte-derived sEV external surface and biologically active, playing a critical role in adipocyte-cardiomyocyte communication. Diabetes disrupts this communication by enhancing GRK2-mediated AdipoR1 phosphorylation, impairing sEV signaling, and exacerbating IHF. These findings provide new insights into the pathophysiology and therapy of IHF in diabetes.

Introduction

Cardiovascular disease is the leading cause of morbidity and mortality in patients with type 2 diabetes, a disease affecting >23 million people in the United States. Obesity, hyperglycemia, and hyperlipidemia are the most common metabolic disorders in diabetes and are established cardiovascular risk factors. However, recent large-scale clinical trials failed to demonstrate cardiovascular mortality benefit from strict glycemic control in diabetic patients^1–4. Novel strategies capable of protecting diabetic cardiomyocytes from exacerbated post-myocardial infarction (post-MI) remodeling and ischemic heart failure (IHF) are urgently needed.

Research in the past decade has increased understanding of adipocytes’ role in health and disease^5–7. The adipocytes are traditionally considered an inert storage depot for nutrients. However, recent studies demonstrated that adipose tissue is the largest endocrine organ, producing a wide range of hormones and cytokines regulating remote organ functions⁸. Normal (lean) adipocytes are beneficial in maintaining systemic metabolic homeostasis, whereas dysfunctional (obese) adipocytes prominently affect the development/progression of type 2 diabetes and diabetic cardiovascular complications^9–12. Understanding how normal adipocytes exert their cardioprotective effect and how diabetes adversely impacts adipocyte-cardiomyocyte communication will help identify novel effective therapies against MI injury in diabetic individuals.

Small extracellular vesicles (sEV) are released by all cells and have critical roles in homeostatic processes and intercellular communication¹³. Recent studies demonstrate that adipocytes are a significant source of circulation sEV and play a critical role in systemic metabolic hemostasis^14,15. Healthy adipose tissue-derived sEV positively regulate remote organ functions, enhancing systemic insulin sensitivity¹⁶ and promoting pancreatic insulin secretion¹⁷. In contrast, dysfunctional adipocyte sEV promote remote organ pathologic remodeling, including systemic insulin resistance^18–21, cancer cell metastasis²², and atherosclerosis¹⁹. Our recent study demonstrated for the first time that sEV from nondiabetic adipocytes significantly attenuates acute myocardial reperfusion (MI/R) injury. In contrast, diabetic adipocytes-generated sEV are vehicles carrying cytotoxic molecules from diabetic adipocytes to cardiomyocytes, significantly exacerbating MI/R injury²³. However, several important questions remain unanswered. First, the cardioprotective molecules carried by adipocyte sEV remain unidentified. Second, whether adipocyte sEV may attenuate chronic post-MI remodeling and IHF remains unknown. Third and most importantly, whether and how diabetes may alter cardiomyocyte properties and impair protective signaling remains unclear. Clarifying the protective role of adipocyte sEV in adipocyte-cardiomyocyte communication and its alterations by diabetes is the foundation for developing specific interventions that block sEV-mediated, diabetes-exacerbated post-MI remodeling and IHF.

Through a combination of in vitro molecular mechanistic analysis and in vivo concept demonstration, we provided the first evidence that adiponectin (APN) is enriched on the external surface of adipocyte sEV and confers cardioprotection against chronic post-MI remodeling and IHF through the activation of cardiomyocyte AdipoR1. Diabetes disrupts this communication by enhancing GRK2-mediated AdipoR1 phosphorylation, impairing sEV signaling, and exacerbating IHF.

METHODS

Detailed methods for adipocyte sEV isolation, cell isolation and culture, plasmid constructions and transfections, adeno-associated virus 9 vector production and infection, cell viability and apoptosis, Western and coimmunoprecipitation, echocardiography and strain analysis, and immunofluorescent cellular and Masson trichrome staining are provided in Supplemental Material. The data that support the findings of this study are available from the corresponding author on reasonable request.

Animal Study Protocol

All animal experiments were performed in adherence to the National Institutes of Health Guidelines on the Use of Laboratory Animals and were approved by the Thomas Jefferson University Committee on Animal Care. Wild-type (WT), APN knockout (APN-KO), and AdipoR1 knockout (AdipoR1-KO) mice were used in the study. The selection of typical photos for cellular and animal investigations, as well as representative Western blot images, was conducted to align closely with the mean values of the measured parameters. To generate a cardiomyocyte-specific AdipoR1 mutation mouse line, a previously reported method for administering AAV to neonatal mice²⁴ was utilized. In brief, a total of 1×10¹¹ viral genome particles per mouse of AAV9-cTNT-eGFP, AAV9-cTNT-AdipoR1^WT, or AAV9-cTNT-AdipoR1^S205A were injected subcutaneously into the nape of 5- to 7-day-old AdipoR1-KO neonatal mice. Stable cardiac-specific protein expression was evidenced by cardiac-specific eGFP expression 24 weeks after AAV9-cTNT-eGFP injection (Figure S1).

Adult (8 weeks-old) WT C57BL/6J mice, AdipoR1-KO mice (no AdipoR1 was detected, Figure S2) re-expressing AdipoR1WT (AdipoR1KO^AdipoR1WT), or AdipoR1-KO mice re-expressing AdipoR1S205A (AdipoR1KO^AdipoR1S205A) mice were fed a high-fat diet (HFD, 60% kcal fat, 20% kcal protein, 20% kcal carbohydrate, Catalogue No. D12492; Research Diets) for 12 weeks to establish a HFD-induced type 2 diabetic model²³. Age-matched C57BL/6J mice on a normal diet (ND) serve as control. To induce MI/R injury, mice were anesthetized with 2% isoflurane. The heart was temporarily exteriorized via a left thoracic incision. A 6-0 silk suture slipknot was tied around the left anterior descending coronary artery as previously described²⁵. 90 minutes after MI, the slipknot was released, allowing myocardial reperfusion for 3 hours (apoptosis assay), 24 hours (infarct size assay), and 4 weeks (post-MI remodeling and cardiac function assays). Adipocyte sEV (diluted in 20μl PBS, 2×10⁸ /mice) or PBS (vehicle) were intramyocardially injected into the left ventricle at three distinct points distal of the planned coronary ligation site 48 hours before MI/R.

Statistical Analysis

All numerical information is presented as the mean±SEM. Using the Shapiro-Wilk test, the normality of the data was determined. Using the unpaired t-test, comparisons were made between the two groups. Three or more groups were compared using 1- or 2-way ANOVA followed by the Tukey test for multiple comparisons. For the repeated measurements on the same mice in the echo analysis, mixed-effects Two-way ANOVA was used to account for the correlation of observations. GraphPad Prism 9.4.1 was used to conduct all statistical analyses. P values < 0.05 were considered statistically significant.

RESULTS

Adipocyte sEV attenuated post-MI remodeling and IHF in nondiabetic but not in diabetic mice:

We previously demonstrated that nondiabetic adipocyte-derived sEV attenuate, whereas diabetic adipocyte-derived sEV exacerbates reperfusion injury in nondiabetic animals. To determine whether nondiabetic adipocyte sEV may exert a sustained cardioprotective effect against post-MI remodeling and IHF, and more importantly, whether adipocyte sEV might be an effective therapeutic intervention against post-MI remodeling in diabetic animals, 2×10⁸/mice adipocyte sEV or PBS was injected into the left anterior ventricle wall of ND or HFD mice. Similar to what we previously observed in the reperfusion model, administration of adipocyte sEV significantly attenuated chronic post-MI remodeling and IHF in nondiabetic mice (Figures 1A-D). Unfortunately, intramyocardial injection of adipocyte sEV failed to attenuate post-MI remodeling in diabetic mice (Figures 1E-H). These results indicated that adipocyte sEV is an effective therapeutic intervention against chronic post-MI remodeling in nondiabetic mice but ineffective in diabetic mice.

Figure 1. Adipocyte-derived sEV protected against MI/R injury and IHF in non-diabetic mice, but failed to do so in diabetic mice.

A) ND mice were subjected to 90-minute MI and reperfusion, with/without intramyocardial injection of ADp-sEV 48h before MI. 24 hours after MI/R, the hearts were collected, and Evans blue-triphenyltetrazolium chloride double stain was performed to determine the infarct size. 4 weeks after MI/R, the hearts were collected, and Masson’s trichrome staining was performed to determine the collagen volume fraction. Scale bar=1mm. B) Quantification of infarct size and fibrotic area. For the results of infarct size, n=8 in “ND mice + IR” group, n=10 in “ND mice + WT ADp-sEV + IR” group. For the results of fibrotic area, n=6/group, unpaired t test, *p<0.05, ***p<0.001 vs “ND mice + IR” group, respectively. C) 3D regional wall velocity diagrams (B-mode tracing) of LV endomyocardial strain showing contraction (orange) and relaxation (blue) of 3 consecutive cardiac cycles. D) Quantitative analysis of EF% and longitudinal strain measured across the LV endocardium, for the “ND mice” sham group, n=8/group. For “ND mice + IR” and “ND mice + WT ADp-sEV + IR” groups, n=10/group, mixed-effects Two-way ANOVA, ***p<0.001 vs “ND mice + IR” group. E) After MI/R induced IHF in HFD mice, Evans blue-triphenyltetrazolium chloride staining determined myocardial infarct size 24 hours after reperfusion, Masson’s trichrome staining determined collagen volume fraction 4 weeks after reperfusion, with/without intramyocardially injection of ADp-sEV 48h before MI. Scale bar=1mm. F) Quantification of infarct size and fibrotic area. For the results of infarct size, n=8 in “HFD mice + IR” group, n=10 in “HFD mice + WT ADp-sEV + IR” group. For the results of fibrotic area, n=6/group, unpaired t-test, ns indicates not significant, vs “HFD mice + IR” group. G) LV endomyocardial was determined based on representative images generated from speckle-tracking analysis in the long-axis B-mode and radial segmental synchronicity. H) Global strain and strain rate measured in the radial and longitudinal axes across the LV endocardium. For the “HFD mice” sham group, n=8/group. For “HFD mice + IR” and “HFD mice + WT ADp-sEV + IR” groups, n=10/group, mixed-effects Two-way ANOVA, ***p<0.001, ns indicates not significant, vs “HFD mice + IR” group. ND: normal diet; HFD: high-fat diet; IR: myocardial ischemia/reperfusion; WT ADp-sEV: Small extracellular vesicle (sEV) isolated form the primary adipocytes (ADp) of epididymal fat tissue of wild type mice.

Adipocyte sEV rapidly activated injury salvage kinases and attenuated cardiomyocyte oxidative injury:

Post-MI remodeling involves complex cellular mechanisms. To determine whether adipocyte sEV exerted a direct protective effect against cardiomyocyte injury, we isolated primary adult mouse cardiomyocytes (PAMC) and performed in vitro experiments. First, PAMCs were treated with adipocyte sEV, and the time-dependent activation (5 to 30 min) of multiple injury salvage kinases (ISKs), including ERK, AMPK, and ACC, was determined. Interestingly, adipocyte sEV rapidly (as early as 15 min) and significantly (P<0.01) activate these ISKs, as evidenced by increased phosphorylation forms of ISKs without altering their total protein levels (Figures 2A/B). To determine whether sEV activation of ISKs may be translated into a sustained cellular protective action, PAMCs were treated with H₂O₂ in the presence and absence of adipocyte sEV. Cellular injury was determined by MTT and LDH. Treatment with adipocyte sEV significantly attenuated oxidative cell injury (Figures 2C/D). sEV regulate recipient cell functions through two distinctive mechanisms: receptor-mediated signaling (rapid action in minutes) and endocytosis-dependent intracellular cargo delivery (slow but sustained action)^26–28. Collectively, these results suggest that cytoprotective molecules are present on the external surface of adipocyte sEV, protecting cardiomyocytes from oxidative injury through receptor-mediated signaling.

Figure 2. APN is located on the surface of adipocyte sEV and activates injury salvage kinases in cardiomyocytes.

A) PAMC was isolated from ND mice and incubated with adipocyte sEV, adipocyte sEV pre-treated with an APN neutralization antibody, or APN-KO adipocyte sEV for 15 min. Effect of cell salvage kinase activation in cardiomyocytes induced by adipocyte sEV, adipocyte sEV pre-treated with an APN neutralization antibody or APN-KO adipocyte sEV. B) Quantification of the Western blot results normalized to GAPDH. n=5/group, one-way ANOVA, ***p < 0.001 vs “WT ADp-sEV” group, respectively. PAMC was isolated from ND mice and incubated with WT ADp-sEV for 24h, followed by H₂O₂ treatment for 2h. C) PAMC viability was determined by MTT assay, D) PAMC injuries were tested by LDH assay. n=5/group, one-way ANOVA, *p<0.05, **p<0.01, ***p < 0.001 vs “PAMC+H₂O₂” group, respectively. E-F) The sEV-Flow experiment was performed to determine if APN is located on the surface of ADp-sEV. G) Quantification of the Exo-Flow results. n=6, unpaired t test, ***p < 0.001 vs Con group.

Adiponectin (APN) is enriched on the adipocyte sEV external surface and is responsible for sEV cardioprotection:

Adiponectin is an adipocyte-derived cytoprotective molecule. There was no significant difference in APN expression levels between adipocytes and ADp-sEVs (Figure S3). Additionally, all three forms of APN (high, medium, and low molecular weights) were detected in ADp-sEV (Figure S4). Having demonstrated that adipocyte sEV rapidly activate ISKs and attenuates cardiomyocyte injury, we reasoned that adiponectin might be the adipocyte sEV external surface molecule responsible for their cardioprotective action. Several experiments were performed to test this novel hypothesis. First, we performed an in vitro experiment utilizing Exo-Flow (FACS), a recently developed technique that specifically detects sEV surface molecules (SBI, CSFLOWBASICA-1). The results showed that adipocyte sEV carried high levels of APN on their external surface (Figures 2E-G). Second, adipocyte sEV were pre-treated with an APN neutralization antibody before their addition to PAMC. As summarized in Figures 2A and B, adipocyte sEV pre-treated with an APN neutralization antibody (APN-Nu-Adipocyte-sEV) lost their ability to activate ISKs. Third, adipocyte sEV were isolated from APN-KO mice (in which APN expression was not detected, Figure S5), and their effect on ISK activation was determined. Similar to that observed with APN neutralization antibody pre-treated sEV, adipocyte sEV from APN-KO mice (APN-KO-Adipocyte-sEV) failed to activate ISK (Figures 2A/B). Finally, in vivo experiments were performed to determine whether adipocyte sEV cardioprotection is mediated by their external surface APN. Adipocyte sEV were isolated from WT mice or APN-KO mice. WT mice-derived sEV were pre-treated with either nonspecific or APN neutralization antibodies. Three groups of adipocytes sEV (Adipocyte-sEV, APN-Nu-Adipocyte-sEV, APN-KO-Adipocyte-sEV) were administered as described above. The quality control of sEV was performed via NTA traces, Western blots anti-sEV specific markers, and TEM (Figure S6). Their protective effect on post-MI remodeling was determined. Consistent with in vitro observations, administration of adipocyte sEV significantly attenuated post-MI remodeling. However, neither APN-Nu-Adipocyte-sEV nor APN-KO-Adipocyte-sEV significantly protect against post-MI remodeling and IHF. Specifically, administration of APN-Nu-Adipocyte-sEV or APN-KO-Adipocyte-sEV had no significant effects on left ventricular ejection fraction (Figures 3A-C), left ventricular wall velocity (Figure 3D), the radial/longitudinal strain and their strain rates (Figures 3E/F), infarct size (Figures 4A/B), apoptotic death (Figures 4C/D/E/F), and cardiac fibrosis (Figures 4G/H). Collectively, these in vitro and in vivo experimental results demonstrated that adipocyte sEV surface-enriched APN is the molecule responsible for their cardioprotective property.

Figure 3. WT ADp-sEV protected MI/R injury and following IHF, but WT ADp-sEV pre-treated with APN neutralization antibody or APN-KO ADp-sEV lost the cardioprotective effect.

WT mice were subjected to 90 min MI and 4 weeks of reperfusion, 48h after intramyocardial injection with WT ADp-sEV, WT ADp-sEV pre-treated with APN neutralization antibody, or APN-KO ADp-sEV. Echocardiography was performed per week after MI/R to determine the cardiac function. A) In M-mode, echocardiography was traced based on typical images. B-C) Quantitative analysis of cardiac function was performed by echocardiography after MI/R surgery and intramyocardial injection. n=10/group, mixed-effects Two-way ANOVA, ns indicates not significant, vs “WT mice + IR” group. D) In B-mode tracing, contraction (orange) and relaxation (blue) of left ventricular endomyocardial strain were analyzed based on 3D regional wall velocity diagrams. E-F) Quantitative analysis of strain and strain rate in the radial and longitudinal axes in long-axis B-mode. n=10/group, mixed-effects Two-way ANOVA, ns indicates not significant, vs “WT mice + IR” group.

Figure 4. APN on the surface of WT ADp-sEV reduced heart injury and remodeling following MI/R.

WT ADp-sEV, WT ADp-sEV pre-treated with APN neutralization antibody or APN-KO ADp-sEV was intramyocardially injected into heart. 48h later, the mice received MI/R surgery. A) Evans blue-triphenyltetrazolium chloride double stain showed the Infarct size of the heart 24 hours after reperfusion. B) Quantification of infarct size. For “WT mice + WT ADp-sEV + IR” group, n=8/group. For “WT mice + WT ADp-sEV + APN ab + IR” and “WT mice + APN^-/- ADp-sEV + IR” groups, n=10/group, one-way ANOVA, *p<0.05, ns indicates not significant, vs “WT mice + WT ADp-sEV + APN ab + IR” group. C) 3 hours after reperfusion, the heart was collected and TUNEL staining was performed based on the freezing microtome section of the heart. Scale bar=200μm. D) Quantification of TUNEL positive signal. n=6/group, one-way ANOVA, **p<0.01, ns indicates not significant, vs “WT mice + WT ADp-sEV + APN ab + IR” group. E) PAMCs were isolated from WT mice and incubated with WT ADp-sEV for 48 hours. After that, PAMCs were treated with H₂O₂ for 2h, followed by Western blot anti-Cleaved-Caspase 3 or anti-Caspase 3 to determine the apoptosis of PAMCs. F) Quantitation of Western blot results based on E). n=5/group, one-way ANOVA, *p<0.05, ***P<0.001, vs “H₂O₂” group, respectively. G) Representative images of Masson’s trichrome staining of the coronal plane. Scale bar=1mm (upper panel), 100μm (lower panel). H) Quantification of fibrotic area. n=6/group, one-way ANOVA, ***p<0.001, ns indicates not significant, vs “WT mice + WT ADp-sEV + APN ab + IR” group.

Cardiomyocyte AdipoR1 mediates adipocyte sEV cardioprotection, a function blocked by GRK2:

Having demonstrated that APN is the critical molecule responsible for adipocyte sEV cardioprotection, we next investigated how diabetes may adversely impact cardiomyocytes and impair their response to adipocyte sEV cytoprotection. AdipoR1 is the dominant APN receptor expressed in cardiomyocytes. Neonatal cardiomyocytes were isolated from WT mice or AdipoR1-KO mice and treated with vehicle or adipocyte sEV. Adipocyte sEV activated ACC, ERK, and APMK in WT neonatal cardiomyocytes but failed to do so in AdipoR1-KO neonatal cardiomyocytes (Figures 5 A/B). Consistent with in vitro experimental results, the intramyocardial injection of adipocyte sEV in AdipoR1-KO mice showed no protective effect against post-MI remodeling and the development of IHF (Figures 5 and 6). These results demonstrate that the adipocyte-expressed Exo surface-localized APN and cardiomyocyte-expressed AdipoR1 are essential for protective communication between adipocytes and cardiomyocytes. In addition to AdipoR1, the role of T-cadherin in mediating the signaling effects of APN presented on the surface of ADp-sEVs was investigated. Knockdown of T-cadherin using shRNA significantly reduced ADp-sEV-induced AMPK phosphorylation, a key event in APN-mediated cardioprotective signaling. These results indicate that T-cadherin partially contributes to the transmission of cardioprotective signals by ADp-sEVs (Figure S7).

Figure 5. AdipoR1 knockout in cardiomyocytes lost the cardioprotective effect induced by APN located on the WT ADp-sEV.

AdipoR1 in cardiomyocytes is responsible for the ADp-sEV-induced cardioprotection. A) NMVMs were isolated from WT mice or AdipoR1 knock out mice and incubated with/without WT ADp-sEV for 15 min. Cell salvage kinases, including ACC, AMPK, and ERK, were determined in WT and AdipoR1-KO cardiomyocytes. B) Quantification of the Western blot results A) normalized to GAPDH. n=5/group, one-way ANOVA, ***p < 0.001 vs “WT Neonatal CM + WT ADp-sEV” group, respectively. 48 hours after intramyocardial injection with PBS or WT ADp-sEV, MI/R surgery was performed and followed by echocardiography every week. C) Representative echocardiographic images acquired from M-mode tracing. D) Quantitative analysis of LVEF to determine the cardiac function. For “R1^-/- mice” sham group, n=8/group. For “R1^-/- mice + IR” and “R1^-/- mice + WT ADp-sEV + IR” groups, n=10/group, mixed-effects Two-way ANOVA, ***p<0.001, ns indicates not significant, vs “R1^-/- mice + IR” group. E) 3 consecutive cardiac cycles of left ventricular endomyocardial strain in 3D regional wall velocity diagrams were gained via B-model tracing after MI/R surgery. F) Quantitative analysis of longitudinal/radial strain and strain rate based on E). For “R1^-/- mice” sham group, n=8/group. For “R1^-/- mice + IR” and “R1^-/- mice + WT ADp-sEV + IR” groups, n=10/group, mixed-effects Two-way ANOVA, ***p<0.001, ns indicates not significant, vs “R1^-/- mice + IR” group. CM: Cardiomyocyte; R1^-/- mice: AdipoR1 knockout mice.

Figure 6. AdipoR1-KO in cardiomyocytes lost the response to the WT ADp-sEV cardioprotective effect, which enhanced heart injury and remodeling following MI/R.

WT ADp-sEV was intramyocardially injected into the heart of AdipoR1-KO mice. After 48h, the mice received MI/R surgery. A) 24h after reperfusion, heart was collected and Evans blue-triphenyltetrazolium chloride double stain was performed to determine the infarct size. B) Quantification of infarct size was analyzed based on A). For “R1^-/- mice + IR” group, n=8/group. For “R1^-/- mice + WT ADp-sEV + IR” group”, n=10/group, unpaired t test, ns indicates not significant, vs “R1^-/- mice + IR” group. C) 3 hours after reperfusion, the heart was removed for the freezing microtome section, followed by TUNEL staining to show the cardiomyocytes apoptosis after MI/R. Scale bar=200μm. D) Quantification of TUNEL-positive signal. n=6/group, unpaired t test, ns indicates not significant, vs “R1^-/- mice + IR” group. E) Representative images of Masson’s trichrome staining of the coronal plane were acquired after 4 weeks’ reperfusion. Scale bar=1mm (upper panel), 100μm (lower panel). F) Quantification of fibrotic area. n=6/group, unpaired t test, ns indicates not significant, vs “R1^-/- mice + IR” group.

We recently reported that GRK2 causes AdipoR1 phosphorylation at Ser²⁰⁵, promoting its endocytosis and blocking APN transmembrane signaling²⁹. AdipoR1-KO neonatal cardiomyocytes were transfected with a plasmid expressing either a WT AdipoR1 (AdipoR1^WT) or mutant AdipoR1 (AdipoR1^S205A) with/without co-transfection with GRK2. Forty-eight hours after transfection, cardiomyocytes were treated with adipocyte sEV. GRK2 over-expression blocked adipocyte sEV activation of ACC, AMPK, and ERK in AdipoR1^WT neonatal cardiomyocytes. However, adipocyte sEV activation of ACC, AMPK, and ERK was not affected by GRK2 in AdipoR1^S205A re-expression cardiomyocytes (Figures 7A-D). Moreover, GRK2 overexpression in AdipoR1^WT cardiomyocytes enhanced oxidative stress-induced cell death, as evidenced by poorer cell viability and higher LDH release (Figures 7E/G, right). However, GRK2 overexpression in AdipoR1^S205A cardiomyocytes did not increase oxidative stress-induced cell death (Figures 7E-H, Figure S8). Similar to the H₂O₂-induced oxidative stress model, WT ADp-sEV protected against the increased cell death and injury of adult mouse cardiomyocytes caused by the simulated ischemia/reoxygenation (SI/R) model. In the neonatal cardiomyocytes isolated from AdipoR1^KO mice and reconstituted with AdipoR1^WT, WT ADp-sEV significantly attenuated simulated SI/R injury; however, this protective effect was abolished upon co-overexpression of GRK2. Notably, phosphorylation-resistant mutant AdipoR1^S205A overexpression preserved the cardioprotective effect of ADp-sEV even in the presence of GRK2 overexpression (Figure S9).

Figure 7. AdipoR1S205 phosphorylation induced by Grk2 in cardiomyocytes is responsible for the resistance of WT ADp-sEV cardioprotection.

A) NMVMs were isolated from AdipoR1 knock-out mice and transfected with plasmids overexpressing AdipoR1 with/without plasmids overexpressing Grk2. 48 hours later, WT ADp-sEV was added for a 15-minute incubation. A Western blot was performed to determine the activation of ISKs (pACC, pAMPK, and pERK). B) Quantitative analysis of the Western bolt results in A). n=5/group, one-way ANOVA, *p<0.05, **p<0.01, ***p < 0.001 vs “WT Neonatal CM + OE-AdipoR1 + WT ADp-sEV” group, respectively. C) AdipoR1^WT or AdipoR1^S205A was overexpressed in AdipoR1-KO NMVMs, at the same time, Grk2 overexpression plasmid was transfected (Right) or not (Left). 48 hours later, WT ADp-sEV was used to incubate with NMVMs for 15min, followed by Western blot to determine the activation of ISKs (pACC, pAMPK, and pERK). D) Quantitative analysis of the Western bolt results in C). n=5/group, one-way ANOVA, **p<0.01, ***p < 0.001, ns indicates not significant. E-F) AdipoR1-KO NMVMs, overexpressed with AdipoR1^WT or AdipoR1^S205A, were incubated with WT ADp-sEV for 24h and then treated with H₂O₂ for 2h. The MTT assay was performed to test the viability of the NMVMs. G-H) AdipoR1^WT or AdipoR1^S205A overexpressed AdipoR1-KO NMVMs were treated with WT ADp-sEV and H₂O₂ as in E-F), NMVM injuries were tested by LDH assay. n=5/group, one-way ANOVA, *p<0.05, **p<0.01, ***p < 0.001, ns indicates not significant, vs “H₂0₂” group (Left) or “OE-Grk2 + H₂0₂” group (Right), respectively.

Diabetes significantly upregulated GRK2 expression and caused AdipoR1 phosphorylation, and a phosphorylation-resistant AdipoR1 knock-in rescued adipocyte sEV cardioprotection in diabetic animals:

The ischemic upregulation of GRK2 and its critical role in post-MI remodeling are well-recognized. However, whether GRK2 is upregulated in diabetic hearts and contributes to diabetic exacerbation of IHF remains unclear. Two in vivo experiments were conducted to address this critical question. AAV9-cTNT-AdipoR1^WT or AAV9-cTNT-AdipoR1^S205A were injected subcutaneously into the nape of 5- to 7-day-old AdipoR1-KO neonatal mice. At the age of 8 weeks, mice were fed a high-fat diet for 12 weeks. Cardiac GRK2 expression was significantly upregulated compared with nondiabetic hearts, and AdipoR1 phosphorylation was significantly increased in diabetic AdipoR1^WT re-expression hearts without coronary artery ligation, but was resisted in diabetic AdipoR1^S205A re-expression hearts (Figure 8A). In a separate cohort of animals, diabetic mice were randomized to receive vehicle or adipocyte sEV intramyocardial injection and subjected to MI/R, exhibiting enhanced uptake of ADp-sEV (Figure S10). Administration of adipocyte sEV in AdipoR1^WT re-expression mice did not attenuate the diabetic exacerbation of ischemic heart injury. However, the cardioprotective effect of adipocyte sEV was preserved in the AdipoR1^S205A re-expression heart. Specifically, administration of adipocyte sEV in diabetic AdipoR1KO^AdipoR1S205A animals significantly reduced apoptosis (determined 3 hours after reperfusion, Figure 8B), reduced infarct size (determined 24 hours after reperfusion, Figure 8C), improved cardiac function, and reduced interstitial fibrosis (determined 4 weeks after reperfusion, Figures 8D/E/F). These results indicated that re-expression of a phosphorylation-resistant AdipoR1 rescued the cardioprotective effects of adipocyte sEV in the diabetic heart. To further investigate the relationship between ADp-sEVs and cardioprotective signaling, we focused on the AMPK and ERK pathways, given that ACC is a well-established downstream target of AMPK. Primary neonatal mouse cardiomyocytes were transfected with siRNAs targeting either AMPK or ERK, followed by treatment with ADp-sEVs and H₂O₂-induced oxidative stress. Our MTT and LDH results showed that knockdown of either AMPK or ERK significantly reduced the protective effects of ADp-sEVs, with AMPK knockdown producing a more pronounced inhibitory effect (Figure S11).

Figure 8. AdipoR1S205A overexpression in cardiomyocytes resisted AdipoR1 phosphorylation and rescued WT ADp-sEV cardioprotection in HFD mice.

AdipoR1^WT or AdipoR1^S205A was overexpressed in the heart of AdipoR1-KO mice, followed by 12 weeks’ HFD. After that, WT ADp-sEV was intramyocardially injected; 48h later, the mice were subjected to 90-minute MI and reperfusion. A) AAV9-cTNT-AdipoR1^WT or AAV9-cTNT-AdipoR1^S205A were injected into AdipoR1-KO mice to generate AdipoR1^WT or AdipoR1^S205A heart-specific expression mice. After 12 weeks’ HFD, the hearts were harvested, and CoIP was performed to determine the Ser phosphorylation in AdipoR1 protein, and a Western blot was performed to determine the expression of AdipoR1 and Grk2. In HFD mice, Grk2 was overexpressed in CM, which induced phosphorylation of AdipoR1. However, overexpression of the AdipoR1^S205A mutant protein in CM resisted phosphorylation in HF mice. n=5/group, one-way ANOVA, **p<0.01, ***p<0.001, ns indicates not significant. B) 3 hours after MI/R surgery, the heart was removed and frozen-microtome sectioned. Scale bar=200μm. The TUNEL staining was used to show cardiomyocyte apoptosis. n=6/group, unpaired t test, **p<0.01 vs “R1^-/- mice + AAV9-AdipoR1^WT + HFD + WT ADp-sEV + IR” group. C) 24 hours after reperfusion, Evans blue-triphenyltetrazolium chloride double stain was performed in the heart to determine the infarct size, and the quantification of infarct size was analyzed. n=10/group, unpaired t test, *p<0.05 vs “R1^-/- mice + AAV9-AdipoR1^WT + HFD + WT ADp-sEV + IR” group. D) Via B-mode tracing, 3 consecutive cardiac cycles from 3D regional wall velocity diagrams of LV endomyocardial strain showed the contraction (orange) and relaxation (blue) of the heart. And the quantitative analysis of EF% across the LV endocardium. n=10/group, mixed-effects Two-way ANOVA, **p<0.01 vs “R1^-/- mice + AAV9-AdipoR1^WT + HFD + WT ADp-sEV + IR” group. E) Quantitative analysis of longitudinal/radial strain and strain rate. n=10/group, mixed-effects Two-way ANOVA, **p<0.01, ***p<0.001 vs “R1^-/- mice + AAV9-AdipoR1^WT + HFD + WT ADp-sEV + IR” group. F) AdipoR1^S205A overexpression in cardiomyocytes responded to WT ADp-sEV cardioprotection and resisted fibrotic area in HFD mice. Scale bar=1mm. n=6/group, unpaired t test, *p<0.05 vs “R1^-/- mice + AAV9-AdipoR1^WT + HFD + WT ADp-sEV + IR” group. G) Graphic illustration. APN located on the surface of ADp-Exo acts as the critical executor of ADp-CM communication, mediating ADp-Exo cardioprotection. GRK2-induced CM AdipoR1 phosphorylation blocks ADp-Exo protective action, contributing to IHF progression.

Discussion

Cardiovascular complications, particularly ischemic heart disease, are the primary cause of death in patients with obesity and type 2 diabetes. Excessive and dysfunctional visceral adipocytes are the culprits of obesity-induced diabetes, whereas cardiomyocytes are the victims, most significantly contributing to diabetic cardiovascular death. A comprehensive understanding of the molecular mechanisms that mediate communication between adipocytes and cardiomyocytes may lead to the discovery of novel, effective therapies against diabetic cardiac injury. Our current study made several significant observations. First, we demonstrated for the first time that adipocyte sEV carries cardioprotective molecules on their external surface. They activate multiple injury salvage kinases and significantly attenuate ischemic heart injury in nondiabetic but not in diabetic animals. Second, we provided the first direct evidence that sEV external surface APN is biologically active and responsible for adipocyte-cardiomyocyte protective communication. Third, we demonstrated that GRK2 is significantly upregulated in diabetic cardiomyocytes, leading to AdipoR1 phosphorylation, which blocks adipocyte sEV cardioprotection and contributes to the exacerbation of IHF in diabetes (Figure 8G).

The pathologic roles of diabetic adipocyte sEV in the development of cardiovascular complications have been extensively investigated. Research from numerous investigators, including our group, has shown that adipocyte sEV from obesity/diabetes lead to pathological remodeling of distant organs through complex mechanisms^{18,20–22,30}. However, the physiological and protective roles of sEV from nondiabetic adipocytes in regulating distant organs have only started to gain attention in recent years. sEV from nondiabetic adipocytes³¹ or adipose tissue macrophage¹⁶ improves liver and skeletal muscle insulin sensitivity in diabetic mice. Additionally, nondiabetic adipocyte sEV display pancreatic tropism, thereby improving pancreatic cell function¹⁷. We recently reported that diabetes switches adipocyte sEV phenotype from protective against acute reperfusion injury to harmful to reperfused cardiomyocytes²³. However, whether and how sEV from nondiabetic adipocytes may protect the heart from chronic post-MI remodeling and IHF, and whether and how this protective signaling is adversely altered by diabetes have not been previously investigated.

sEV regulates recipient cell functions through two distinctive mechanisms: receptor-mediated signaling and endocytosis-dependent intracellular cargo delivery^26–28. During the past 15 years, most sEV-mediated communications have been focused on the intercellular transfer of sEV cargoes, such as miRNAs, lipids, proteins, and organelles^32,33. However, there is an increasing recognition that proteins located on the sEV external surface play a critical role in mediating sEV communication with remote organs³⁴. sEV derived from malignant osteosarcoma cells carries TGFβ on the external surface (associated with CD63⁺), which binds to TGFβ receptors on mesenchymal stem cells and initiates pathologic signaling³⁵. In contrast, the sEV from human placental-expanded stromal cells promotes angiogenesis, skin regeneration, and immunomodulation³⁶. Removing the sEV external surface proteins abolishes these effects³⁶. Our current study demonstrated that adipocyte sEV rapidly activates multiple injury salvage kinases in nondiabetic cardiomyocytes, significantly attenuating oxidative stress-induced myocardial injury. These results indicate that the novel sEV external surface molecules-mediated intercellular communication is involved in adipocyte-cardiomyocyte protective signaling.

Compared with sEV cargo delivery regulation, sEV external molecule-mediated cell-cell communication has several advantages. Firstly, the sEV external surface proteins activate their receptors on target cells, immediately triggering intracellular signaling. This allows sEV to achieve its regulatory effects more rapidly than through intracellular cargo delivery by sEV. Rapid action is crucial in protecting against acute cellular damage, such as that caused by myocardial ischemia/reperfusion injury. Secondly, sEV modification has been an area of intense research in recent years, with the goal of enhancing drug delivery and therapeutic efficacy. Bioengineering modifications to the molecules on the sEV external surface are likely more practical than modifications to the cargo content within the sEV. Thirdly and more importantly, sEV external surface molecules may determine the cell/tissue selective recognition^28,34,37, which not only ensures cell-selective signaling activation but may also regulate cell-selective cargo delivery (acting either as opsonins to facilitate or as dysopsonins to decrease the cellular uptake of the sEV). In this regard, a recent study demonstrated that the integrin isoforms on the external surface of cancer cell sEV determine organ-specific sEV cargo delivery³⁸.

APN is an adipocyte-specific protein regarded as “adipocyte-derived insulin” with potent cardioprotective effects^39,40. Considering that adipocytes are the primary cells of APN expression, it is unsurprising that adipocyte sEVs contain high levels of APN, which were significantly suppressed by diabetes (Figure S12). However, two unique properties of adipocyte sEV-carrying APN are exciting and important. Firstly, adipocyte sEV contains high levels of APN but deficient levels of leptin⁴¹. Given that leptin is an adipocyte-specific protein (the first identified adipocyte-specific cytokine), the absence of leptin in adipocyte sEV suggests that APN is selectively packaged into adipocyte sEV. We further corroborated this observation in the present study. After culturing primary adipocytes and isolating ADp-sEVs, we collected protein samples from the primary adipocytes, purified the ADp-sEVs, and the ADp-sEV-depleted conditioned medium. Western blot analyses showed that leptin was readily detected in adipocytes and was abundant in the sEV-free conditioned medium, whereas it was extremely low (barely detectable) in the ADp-sEV fraction (Figure S13). These findings support selective cargo packaging and indicate that adipocyte-derived leptin is secreted predominantly as a soluble extracellular protein with minimal loading into ADp-sEVs. Accordingly, leptin carried by ADp-sEVs is unlikely to be a major contributor to the cardiovascular effects observed in our study. This observation further highlights the potential for adipokine-specific sorting into ADp-sEVs, suggesting that distinct adipokines may engage in cardiovascular signaling through different secretion and delivery modes. Secondly, chemical/biochemical experiments indicate that APN exists as an external surface protein in adipocyte-derived sEV⁴¹. A very recent study demonstrated that adipocyte sEV external surface APN improves systemic insulin sensitivity, attenuates IL-1β and TNF-α release, and inhibits liver macrophage infiltration in HFD-induced diabetic mice³¹. Our current study provided more compelling and first evidence that adipocyte sEV external surface APN is responsible for adipocyte remote protection against ischemic cardiomyocyte injury. Firstly, utilizing a recently developed flowcytometry method (Exo-Flow, System Biosciences), which specifically detects sEV external surface-located molecules⁴², we provide direct evidence that APN is present on the adipocyte sEV external surface. Secondly, we demonstrated that adipocyte sEV isolated from APN-KO mice failed to protect the heart against post-MI remodeling and IHF. Thirdly and more interestingly, we showed for the first time that incubation of adipocyte sEV with an APN neutralization antibody blocked their cardioprotective function. Finally, we demonstrated that the AdipoR1-deficient hearts lose their protective response to adipocyte sEV. To determine whether ADp-sEV may also modulate angiogenic signaling, we performed additional in vitro studies using primary cardiac microvascular endothelial cells (PCMECs) isolated from adult mouse ventricles. Hearts were enzymatically digested, and the cell suspension was sequentially filtered. Red blood cells were lysed, and endothelial cells were enriched by anti-CD31 magnetic bead selection, then plated on collagen I-coated dishes in endothelial growth medium. PCMECs were then treated with ADp-sEV or vehicle for 48 hours, followed by Western blot analysis of the pro-angiogenic markers VEGFA and VEGFR2. As shown in Figure S14, ADp-sEV treatment markedly increased VEGFA and VEGFR2 expression in PCMECs, indicating activation of a pro-angiogenic program in vitro that likely contributes, together with the direct cardiomyocyte-protective effects, to the overall cardioprotection observed in vivo. To the best of our knowledge, this is the first study to clarify the molecular mechanism responsible for sEV-mediated protective signaling between adipocytes and cardiomyocytes, as well as cardioprotection.

Compared with soluble APN, sEV external surface APN has several advantages. Firstly, APN association with sEV significantly extends its half-life, thus intensifying its biological activity³¹. Secondly, the APN association with sEV may prevent their pathological modification by circulating molecules, particularly under diabetic conditions⁴³. Thirdly, APN association with sEV may facilitate their crossing of biological barriers and interaction with targeting cells. Finally, the external surface APN of sEV may play a critical role in cell-specific communication between adipocytes and cells expressing high levels of APN receptors (such as cardiomyocytes, hepatocytes, and skeletal muscle cells), promoting the targeted delivery of sEV cargoes.

Two evolutionarily acquired specific APN receptors (AdipoR1 and AdipoR2) have been cloned⁴⁴. They belong to a new family of membrane receptors (PAQR) predicted to contain seven transmembrane domains, similar to GPCRs, but structurally and topologically distinct⁴⁵. APN can also bind to T-cadherin, in an interaction that only tethers APN to the cell surface without transmembrane signaling, as T-cadherin lacks an intracellular domain⁴⁶. Activation of AdipoR1 (primarily expressed in muscular cells) and AdipoR2 (primarily expressed in hepatocytes) increases glucose and free fatty acid utilization, stimulates mitochondrial biogenesis, and inhibits inflammatory response^22–26. We recently demonstrated that AdipoR1, but not AdipoR2, is a potent substrate of GRK2⁴⁷. GRK2 phosphorylates AdipoR1 at S205, promoting its endocytosis and blocking APN cardioprotection²⁹. Although the upregulation of cardiac GRK2 in the ischemic heart and its critical contribution to IHF are well-recognized, the role of GRK2 in diabetic exacerbation of IHF remains unexplored. Our current study demonstrated that HFD-induced type 2 diabetes significantly upregulated cardiac GRK2 and caused AdipoR1 phosphorylation before MI, blocking adipocyte sEV cardioprotection and contributing to diabetic exacerbation of IHF. Notably, the re-expression of a mutant, phosphorylation-resistant AdipoR1 (AdipoR1^S205A) preserves the activation of cardiomyocyte AdipoR1 by adipocyte sEV, protecting diabetic IHF.

A limitation of this study is that ADp-sEV were isolated exclusively from epididymal (visceral) adipose tissue, without direct comparison to sEV derived from other fat depots such as subcutaneous adipose tissue. Our focus on epididymal fat was driven by its central role as the principal visceral depot implicated in diabetic cardiac complications. Future studies will evaluate the efficacy of ADp-sEV administration after MI/reperfusion, using alternative delivery routes and/or animal models that permit post-reperfusion dosing, and will systematically compare sEV from visceral versus subcutaneous and other adipose depots to more fully define depot-specific contributions to diabetic ischemic heart failure. Another limitation of this study is that ADp-sEVs were delivered by pre-MI intramyocardial injection in a preventive paradigm, rather than as post-reperfusion therapy, and thus do not fully recapitulate the clinical scenario in which treatment is initiated after reperfusion. This timing reflects the technical constraints of the Gao/Koch ischemia-reperfusion model, in which intramyocardial injection can be safely performed only during the brief initial thoracotomy. A second thoracotomy at the time of reperfusion would be expected to cause substantial additional trauma, inflammation, and mortality, thereby confounding post-MI remodeling and heart failure readouts. Future studies will evaluate the efficacy of ADp-sEV administered after MI/reperfusion, using alternative delivery routes and/or animal models that permit post-reperfusion dosing without repeat thoracotomy to more closely mirror clinical practice.

In conclusion, our study provides the first evidence that APN on the external surface of adipocyte sEV and AdipoR1 in cardiomyocytes are critical players in sEV-mediated protective communication between adipocytes and cardiomyocytes. The upregulation of GRK2 in diabetes and the subsequent phosphorylation of AdipoR1 disrupt this crucial protective signaling, exacerbating post-MI pathological remodeling and IHF in diabetic animals. Therapeutic administration of sEVs and reactivation of sEV-mediated communication between adipocytes and cardiomyocytes (by inhibiting GRK2-induced phosphorylation of AdipoR1) may represent a promising strategy for treating ischemic heart failure in diabetic patients.

Clinical Perspective.

What Is New?

• This study provides the first evidence that adiponectin enrichment on the external surface of ADp-sEV plays a crucial role in adipocyte-cardiomyocyte communication, activating cardioprotective signaling.

• This study identifies for the first time that AdipoR1 phosphorylation at Ser205 disrupts ADp-sEV transmembrane signaling, contributing to MI/R injury in the diabetic heart.

• This study demonstrates for the first time that blocking AdipoR1 phosphorylation restores ADp-sEV cardioprotection and mitigates ischemic heart failure (IHF) in diabetic animals.

What Are the Clinical Implications?

• By demonstrating that adiponectin enrichment on the surface of ADp-sEV plays a crucial role in cardioprotective signaling, this research could lead to the development of new therapeutic strategies aimed at enhancing cardioprotection, particularly in diabetic patients;

• The finding that blocking AdipoR1 phosphorylation restores ADp-sEV cardioprotection and mitigates ischemic heart failure (IHF) in diabetic animals suggests that similar strategies could be employed in clinical settings to improve outcomes for diabetic patients with heart failure.

This study investigates the role of diabetes in aggravating MI/R injury and the subsequent development of heart failure. For the first time, we identify phosphorylation at the S205 site of AdipoR1 on the cardiomyocyte membrane as a key mechanism contributing to resistance against adiponectin-mediated cardioprotective signaling. Moreover, we demonstrate for the first time that membrane-bound adiponectin on ADp-sEVs effectively transmits cardioprotective signals. Given the excellent biocompatibility, safety, and delivery efficiency of sEVs, the use of ADp-sEVs to supplement adiponectin combined with pharmacological/genetic-modification strategies to inhibit AdipoR1^S205 phosphorylation—may offer a promising therapeutic strategy for mitigating MI/R injury and preventing the progression to heart failure.

Nonstandard Abbreviations and Acronyms:

AdipoR1

Adiponectin Receptor-1

ADp

Adipocyte

APN

Adiponectin

CM

Cardiomyocytes

GPCRs

G-protein-coupled Receptors

Grk2

G-protein-coupled Receptor Kinase 2

HFD

High-Fat diet

IHF

Ischemic Heart failure

Inf

Infarction

ISKs

Injury Salvage Kinases

KO

Knockout

LVEF

Left Ventricular Ejection Fraction

MI/R (I/R)

Myocardial Ischemia and Reperfusion

ND

Normal Diet

NMVM

Neonatal Mouse Ventricular Cardiomyocytes

Nu

Neutralization

PAMC

Primary Adult Mouse Cardiomyocytes

R1^-/- mice

AdipoR1 knockout mice

sEV

Small Extracellular Vesicles

TTC

Triphenyltetrazolium Chloride

TUNEL

Terminal deoxynucleotidyl transferase dUTP nick end labeling

WT

Wild Type