Ischemia‐induced cardiac dysfunction is exacerbated in adiponectin‐knockout mice due to impaired autophagy flux

Abstract

Strategies to enhance autophagy flux have been suggested to improve outcomes in cardiac ischemic models. We explored the role of adiponectin in mediating cardiac autophagy under ischemic conditions induced by permanent coronary artery ligation. We studied the molecular mechanisms underlying adiponectin's cardio‐protective effects in adiponectin knockout (Ad‐KO) compared with wild‐type (WT) mice subjected to ischemia by coronary artery ligation and H9c2 cardiomyocyte cell line exposed to hypoxia. Systemic infusion of a cathepsin‐B activatable near‐infrared probe as a biomarker for autophagy and detection via noninvasive three‐dimensional fluorescence molecular tomography combined with computerized tomography to quantitate temporal changes, indicated increased activity in the myocardium of WT mice after myocardial infarction which was attenuated in Ad‐KO. Seven days of ischemia increased myocardial adiponectin accumulation and elevated ULK1/AMPK phosphorylation and autophagy assessed by Western blotting for LC3 and p62, an outcome not observed in Ad‐KO mice. Cell death, assessed by TUNEL analysis and the ratio of Bcl‐2:Bax, plus cardiac dysfunction, measured using echocardiography with strain analysis, were exacerbated in Ad‐KO mice. Using cellular models, we observed that adiponectin stimulated autophagy flux in isolated primary adult cardiomyocytes and increased basal and hypoxia‐induced autophagy in H9c2 cells. Real‐time temporal analysis of caspase‐3/7 activation and caspase‐3 Western blot indicated that adiponectin suppressed activation by hypoxia. Hypoxia‐induced mitochondrial reactive oxygen species production and cell death were also attenuated by adiponectin. Importantly, the ability of adiponectin to reduce caspase‐3/7 activation and cell death was not observed in autophagy‐deficient cells generated by CRISPR‐mediated deletion of Atg7. Collectively, our data indicate that adiponectin acts in an autophagy‐dependent manner to attenuate cardiomyocyte caspase‐3/7 activation and cell death in response to hypoxia in vitro and ischemia in mice.

Adiponectin plays a pivotal role in protecting against ischemic injury via stimulating cardiac autophagy. In a permanent coronary artery ligation model, adiponectin‐deficient mice exhibited an impaired autophagy response, measured via ULK1/AMPK phosphorylation and altered levels of autophagy markers LC3 (depicted as green dot) and p62 (red dot). In this way, adiponectin provided protective effects by mitigating ischemia/hypoxia‐induced ROS production and cell death. Ad‐KO, adiponectin‐knock out; ROS, reactive oxygen species.

Abbreviations

Ad‐KO

adiponectin knockout

CAL

coronary artery ligation

CHD

coronary heart disease

CT

computed tomography

CVD

cardiovascular disease

FMT

fluorescent molecular tomography

LV

left ventricle

MI

myocardial infarction

WT

wild type C57BL/6J

Study Highlights.

• WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC?

The current knowledge on the topic involves exploring strategies to enhance autophagy flux to improve outcomes in cardiac ischemic models. The study investigates into the role of adiponectin in mediating cardiac autophagy under ischemic conditions induced by permanent coronary artery ligation. Previous studies have demonstrated an inverse correlation of circulating adiponectin levels with the development and progression of cardiovascular diseases, and adiponectin knockout mice show exacerbated cardiac remodeling and dysfunction.

• WHAT QUESTION DID THIS STUDY ADDRESS?

This study aimed to investigate the role of adiponectin in regulating cardiac autophagy during ischemia. Specifically, it addressed how adiponectin influences autophagy flux in cardiomyocytes under ischemic conditions induced by permanent coronary artery ligation. The study also explored the molecular mechanisms underlying adiponectin's cardioprotective effects and its impact on cell death and cardiac dysfunction.

• WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE?

The study contributes to our understanding by demonstrating that adiponectin plays a crucial role in mediating cardiac autophagy. It provides insights into the molecular mechanisms through which adiponectin influences autophagy flux and its impact on reducing caspase‐3/7 activation and cell death during ischemia. The noninvasive fluorescent molecular tomography‐based protocol using a near‐infrared activatable probe adds a novel method for assessing autophagy flux in live mice.

• HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE?

The findings suggest that adiponectin could be a potential therapeutic target for improving cardiac outcomes under ischemic conditions. Understanding the adiponectin autophagy axis and its impact on reducing cell death may lead to the development of interventions that enhance autophagy flux, potentially mitigating the adverse effects of ischemia on the heart. This study could influence future research in clinical pharmacology and translational science, paving the way for novel therapeutic strategies in cardiovascular medicine.

INTRODUCTION

Cardiovascular disease (CVD) stands as the primary cause of mortality in developed regions, with coronary heart disease (CHD) being responsible for the majority of fatalities within this classification. ¹ CHD triggers cardiac remodeling, encompassing changes in both the structure and function of the heart, commonly described as ischemic injury. ² Consequently, there is a significant focus on unraveling the multifaceted mechanisms that drive cardiac remodeling post‐ischemia due to heightened interest in this area. ³ In recent years, substantial endeavors have been dedicated to deciphering the mechanisms behind ischemia or ischemia/reperfusion injury, with a concerted focus on identifying potential therapeutic targets. ⁴ , ⁵

Extensive research has been conducted on the process of autophagy, a mechanism that facilitates the bulk degradation of cellular components, particularly in the context of ischemia, both with and without reperfusion. ⁴ , ⁶ The role of cardiac autophagy is crucial for sustaining heart function and supporting the adaptive response under both basal and stress‐induced conditions. ⁶ , ⁷ Nevertheless, inhibiting autophagy during ischemia has been shown to exacerbate myocardial injury, highlighting the protective role of autophagy in such conditions. ⁸ Overall, maintaining a suitable balance of autophagy emerges as a crucial factor influencing the outcomes of heart failure in the context of ischemia. ⁹ , ¹⁰ Further exploration is required to gain additional insights into the processes through which autophagy activation can exert either cardioprotective or detrimental effects in the heart. ¹¹

Adiponectin, secreted predominantly from the adipose tissue, has been previously reported to have multiple cardioprotective effects, including in ischemic settings. ¹² , ¹³ , ¹⁴ A growing body of clinical and basic research has revealed associations between circulating adiponectin levels and the development or extent of cardiac disease, leading to adiponectin being proposed as a potential therapeutic target to improve outcomes in myocardial ischemia/reperfusion injury. ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ It has also been suggested that myocardial ischemia/reperfusion injury could alter adiponectin levels in the heart. ²⁰ Furthermore, myocardium can crosstalk with adipose tissue and treatment of adipocytes with plasma from mice having ischemia/reperfusion injury resulted in increased adiponectin expression. ²¹

The interaction between adiponectin and autophagy has been acknowledged in previous studies. Adiponectin triggers intracellular AMP kinase (AMPK) signaling, initiating various catabolic processes, including autophagy, and mitigating remodeling induced by myocardial infarction. ¹⁴ , ²⁰ Concurrently, the mechanistic target of rapamycin (mTOR), a pivotal regulator of cell growth and metabolism, plays a crucial role, with adiponectin acting as an inhibitor of mTOR signaling, thereby instigating autophagy processes. ¹⁴ , ²⁰ In addition, exacerbated cardiac dysfunction in mice lacking adiponectin was associated with dampened AMPK‐mTOR‐ULK1 dependent autophagy. ²² Further dissection of the molecular landscape involves Beclin‐1, a pivotal regulator of autophagy initiation. Adiponectin's influence extends to the upregulation of Beclin‐1 expression, facilitating autophagosome formation a pivotal step in the autophagic cascade. Adiponectin's capacity to enhance microtubule‐associated protein 1A/1B‐light chain 3 (LC3) lipidation underscores its induction of the initiation and progression of autophagy. The selective autophagy substrate p62/SQSTM1, degraded upon augmented autophagy flux and enhanced cellular waste clearance, is another marker regulated by adiponectin. ⁴ , ²⁰ , ²² Collectively, we and others have demonstrated that adiponectin is associated with autophagy rates in multiple tissues. ²¹ , ²³ , ²⁴

The objective of this study was to test the hypothesis that adiponectin‐stimulated autophagy was of functional significance in mediating cardioprotection in mice with myocardial ischemia and in cellular cardiomyocyte models subjected to hypoxia. We investigated the role of adiponectin on autophagy and ischemic damage in vivo using an adiponectin knockout (Ad‐KO) mouse model subjected to myocardial infarction. Whether adiponectin mitigates hypoxia‐induced apoptosis through activation of autophagy was determined using autophagy deficient cells generated via deletion of Atg7.

MATERIALS AND METHODS

Animals and echocardiography

All animals were housed in temperature and humidity‐controlled rooms (21 ± 2°C, 35%–40%) with a daily 12 h light/dark cycle in the animal care facility of York University in accordance with the guidelines of the Canadian Council on Animal Care. All study protocols were approved by the Animal Care Committee of York University. In‐house Ad‐KO mice ²⁵ and age‐matched C57BL/6J (wild type [WT]) male mice (Jackson Laboratory) were maintained with ad libitum access to water and regular chow diet until 12 weeks of age when they were randomly separated into surgical groups (n = 5 per group). Myocardial infarction (MI) was induced with coronary artery ligation (CAL) surgery, as previously described. ²⁶ Mice were euthanized 7 days, or in some experiments 1 day, after MI without reperfusion. Echocardiography was performed as previously described ²³ using the Vevo2100 system (Visual Sonics, Canada) equipped with an MS550D transducer. Mice were lightly anesthetized using 2%–3.0% isoflurane mixed with 100% O₂ for the duration of the imaging. M‐mode images of the parasternal short‐axis views at the papillary level were used to calculate cardiac functions. B‐mode movie files of the parasternal short‐axis views were used to perform Speckle‐tracking cardiac strain rate analysis. All parameters were averaged for at least three cardiac cycles for analysis. Primary cardiomyocytes from adult rats were prepared as described by us before. ²⁷

Conditional cardiac depletion of Atg7 in cardiomyocytes was achieved by crossing αMerCreMer mice with Atg7^flox/flox mice; mouse strains are on the C57BL/6 background. All mice bred via this protocol were genotyped and assigned to experimental groups. Primers used for the genotyping of Atg7^flox/flox and αMerCreMer are listed below. DNA was extracted from ear tissue for genotyping. Conditional Atg7 deletion was achieved by single tamoxifen (40 mg/kg) injection at the age of 7 weeks in male mice. Atg7^flox/flox Cre⁺ mice without tamoxifen injection were used as control animals. Animal body weights were monitored before tamoxifen injection (7 weeks) and weekly thereafter. One week before the induction of Atg7 deletion by tamoxifen injection. ²⁸

Measurement of serum adiponectin

After 1 or 7 days MI, blood samples were collected, and serum level of adiponectin was determined by mouse adiponectin ELISA kit (Cat# 32010; ImmunoDiagnostics Limited, Hong Kong).

Tissue immunofluorescence staining

Paraffin embedded heart tissue sections were de‐paraffinized with serial rehydration, and antigen retrieval was performed in sodium citrate buffer (pH 6.0). Heart sections were permeabilized and blocked with blocking buffer (3% BSA, 5% goat serum, Vector Laboratories, S‐1000, 0.3% Triton X‐100) for 120 min and stained with primary antibody solution (Rabbit anti‐LC3B, MBL, #PM036 at a 1:200 dilution or Rabbit anti‐Adiponectin, in‐house polyclonal antibody at a 1:200 dilution or wheat germ agglutinin [WGA], Invitrogen, #W11261 at a 1:1000 dilution) and incubated overnight at 4°C. The slides were kept in secondary antibody (Goat anti‐Rabbit Alexa 488, Goat anti‐Rabbit 546, ThermoFisher, #A‐11008, #A‐11035) at 1:1000 for 1 h at room temperature after three washes. Slides were mounted in DAPI containing medium (Vectashield, #H‐1200) after three washes, and slides were visualized under Zeiss LSM700 confocal microscope at 20X or 40X.

Histology analysis

Paraffin embedded heart sections were subjected to TUNEL (In Situ Cell Death Detection kit, Roche, #11684795910), WGA (Invitrogen, #W11261), or picrosirius red staining (Abcam, #ab150681) according to manufacturer's protocol and visualized using an Olympus IX71 inverted fluorescence microscope (Olympus Canada, Canada) or EVOS2 FL Auto 2 Microscope.

Fluorescent molecular tomography imaging in mice

Two nanomoles of pan‐cathepsin protease sensor (ProSense 680, Perkin Elmer, #NEV10003) were infused into mice via tail vein injection for the assessment of lysosomal activity. Probes were delivered 24 h prior to imaging. The intensity of fluorescence was visualized at 680 nm with fluorescent molecular tomography (FMT) using VisEn FMT 2500 LX Quantitative Tomography System (Perkin Elmer, USA). Mice were anesthetized, placed in a supine position, and the scan region was established to capture the upper half of the mouse (i.e., nose to above the liver) using an in‐plane resolution of 1 × 1 mm². After FMT signals were captured, mice were transferred to micro–computed tomography (CT) imaging (eXplore Ultra, GE Healthcare) while maintained under light anesthesia. FMT and CT images were reconstructed, merged, and a 3D image was constructed using Amide software (Amide's Medical Imaging Data Examiner). Hearts were isolated from mice after FMT scanning and placed on an opaque resin block in the FMT plexiglass mouse holder and imaged at 680 nm. Epifluorescence images were captured for each mouse heart, and analysis of the ex vivo images was performed using the FMT System software (TrueQuant 4.0).

Western blot

The ischemic zone of the left ventricle (LV) heart tissue was snap frozen and ground with mortar and pestle in liquid nitrogen. The powdered tissue was then suspended in RIPA lysis buffer. ²⁹ H9c2 cells were used for Western blot analysis and cells were suspended in lysis buffer. Proteins were separated by reducing SDS‐PAGE and transferred to a PVDF membrane. The following antibodies were used for Western blot: p‐AMPK T172 (1:1000, Cell Signaling, #2535), p‐ULK1 S555 (1:1000, Cell Signaling, #5869), GAPDH (1:2000, ThermoFisher, #MA5‐15738), Adiponectin (1:1000, ImmunoDiagnostics, #12010), ATG5‐12 (1:1000, Novus Biologicals, #NB110‐53818), ATG7 (1:1000, Santa Cruz, #sc‐376212), Beclin1 (1:1000, Cell Signaling, #3738), LC3B (1:1000, Cell Signaling, #2775), p62 (1:1000, Cell Signaling, #5114), BAX (1:1000, Cell Signaling, #2772), cleaved Caspase 3 (1:1000, Cell Signaling, #9661), bcl‐2 (1:1000, Cell Signaling, #3489), and β‐actin (1:2000, ThermoFisher, #PA1‐46296) were used. The quantification of signals was performed by densitometry of scanned autoradiographs with the aid of ImageJ (version 1.4v).

Cell culture

H9c2 rat embryonic cardiac myoblasts (ATCC CRL‐1446) were grown in Dulbecco's Modified Eagle's Medium (DMEM; Gibco, Invitrogen) supplemented with 10% fetal bovine serum (FBS) and 1% (vol/vol) streptomycin/penicillin (Gibco, Invitrogen) at 37°C and 5% CO₂. When cells reached confluence, they were incubated in 1% FBS‐DMEM overnight. Then, medium were switched to FBS‐free DMEM with or without globular adiponectin (Ad) (ImmunoDiagnostics, #42012) at indicated concentrations. Hypoxia was achieved by placing cells in a hypoxic chamber (Onstage Incubator, ThermoFisher, #NX7LIVE001) filled with a pre‐analyzed gas mixture of 1%O₂ 5% CO₂. Normoxia was achieved by placing cells in the Onstage Incubator filled with 20% O₂ and 5% CO₂.

Generation of ATG7‐KO and VC3AI transduced H9c2 cells

CRISPR knock‐out lines were generated as previously described using guide RNAs cloned into pX459. ³⁰ Vectors carrying guides targeting Atg7 were transfected into H9C2 cells using Lipofectamine 3000. Transduced cells were selected using 1 ug/mL puromycin over 5 days, and then expanded in medium without selection. T7 endonuclease assays were carried out using T7 endonuclease I (NEB) according to the manufacturer's recommendations on polymerase chain reaction (PCR) products generated using the following primer pair 5′‐GCTGCTGCAGGTAGGTGTAA and 5′‐GGTGTCCTGTCTGAGACTGC. To generate H9c2 lines stably expressing Venus‐based Caspase‐3 like protease activity indicator (VC3AI; a gift from Dr Binghui Li, Addgene plasmid 78907), lentiviral particles were produced using pCDH‐puro‐CMV‐VC3AI and transduced cells were selected using 1 μg/mL puromycin. ³¹ This genetically encoded biosensor consists of cyclized chimeras containing a caspase‐3 cleavage site as a switch. Upon cleavage by caspase‐3‐like proteases, the non‐fluorescent indicator rapidly becomes fluorescent, and thus detects activation in real‐time.

Flow cytometry

After treatment, cells were harvested with trypsin–EDTA‐0.25% (Gibco, #25200056) and cell pellets were incubated with appropriate staining dye. Monobromobimane (mBBr) (ThermoFisher, #M1378), MitoSox (ThermoFisher, #M36008), annexin V and propidium iodide (PI; ThermoFisher, #V13245) were incubated in 1% FBS‐DMEM at 37°C for 30 min. After incubation, cells were re‐suspended in FACS buffer (2% FBS in PBS) and assessed by flow cytometry (Attune NxT BRV6, ThermoFisher). For oxidative stress stain, cells were stained with 2.5 μM MitoSox Red (ThermoFisher) and 40 μM mBBr (ThermoFisher) diluted in FACS buffer for 10 min at 37°C. Cells were stained in 1:40 dilution of Annexin V (eBioscience) diluted in 1× binding buffer for 10 min at room temperature. Cells were then washed once in binding buffer and resuspended in 190 μL binding buffer. Finally, 10 μL of PI (1.5 μM) was added to the cells and analyzed immediately by flow cytometry. Compensation and analysis were performed in FlowJO software version 10. Compensation was executed by using singly stained cells, which were subjected to the identical staining protocol using either MitoSox Red or mBBr, as used throughout the remainder of the experiment, in comparison to unstained cells serving as the negative control. Preceding each experimental run, Attune NxT autocalibration was conducted utilizing Attune performance tracking beads. Gating procedures were established through the utilization of Fluorescence Minus One (FMO) controls to delineate negative gates for the experiments. The same methodology was applied to Annexin V and PI assessments, encompassing unstained and singly stained cells for compensation gates, followed by FMO controls for gating.

Microscopic imaging

H9c2 cells and adult rat cardiomyocytes, which were isolated as described previously ²⁷ were seeded on glass coverslips. Staining using different dyes including Cyto‐ID, (1:1000, Enzo Life Sciences, #51031), Magic Red (1:260, ImmunoChemistry Technologies, #942), DalGreen (1:1000, Dojindo, #D675) DapRed (1:1000, Dojindo, #D677), cleaved caspase 3/7 (1:500, ThermoFisher, #C10423), and ReadyProbe (ThermoFisher, #R37609) was performed according to manufacturer's protocols and DAPI (HCS NuclearMask Blue Stain, ThermoFisher, #H10325) was added simultaneously. Cells were fixed in 10% formalin (Sigma, #HT501128) after treatment and mounted with ProLong Gold Antifade Mountant (ThermoFisher, #P36930). Slides were observed under Nikon Eclipse Ti2, Zeiss LSM700 confocal microscope at 60×, Evos FL Auto 2 and for continuous measurements Cellnsight CX7 Platform (ThermoFisher, #CX7A1110) was used. Co‐localization analysis was performed with an ImageJ JACOP plug‐in. At least five random fields from one coverslip/well, counting at least 30 cells were average in each independent experiment. The bar chart represents mean ± SEM values from at least three independent experiments.

Statistics

Data are presented as mean ± SEM. Statistical analysis was performed using GraphPad Prism 9. Student's t‐test was used for comparison of two groups and one‐way analysis of variance (ANOVA) or two‐way ANOVA were used for comparison of more than two groups. Any p < 0.05 was considered statistically significant.

RESULTS

Autophagy flux is enhanced following adiponectin treatment in normoxic or hypoxic conditions

We then used H9c2 cardiac myoblasts subjected to hypoxia (1% O₂) as an in vitro model relevant to myocardial ischemia and treated cells with or without adiponectin for up to 24 h. First, we conducted temporal analyses via time‐lapse microscopy using the probe DapRed to detect autophagosomes and autophagolysosomes (Figure 1a), or DalGreen to detect autophagolysosomes (Figure 1b). These studies allowed us to detect an early increase in both autophagosome and autophagolysosome content was observed in hypoxic cells compared to controls, as shown by DapRed fluorescence, yet, after 24 h, these levels were similar (Figure 1a). The effect of adiponectin in cells under normoxic conditions was similar yet diverged after 12 h toward enhanced autophagosome and autophagolysosome levels. Hypoxia elicited a rapid increase in autophagosome and autophagolysosome content which was further elevated in the presence of adiponectin (Figure 1a). A similar pattern was observed when using DalGreen to detect autophagolysosomes (Figure 1b). Based upon the above kinetic observations, autophagy markers were also assessed by Western blotting after 12 h of adiponectin treatment. Adiponectin, in normoxic conditions, increased LC3‐II and decreased p62, indicating more autophagy flux (Figure 1c–e). At the 12 h of hypoxia, both with or without adiponectin, increased LC3‐II and decreased p62 (Figure 1c–e). Hypoxia promoted autophagy flux compared to normal cells as indicated by a reduced green to red ratio which was further enhanced following adiponectin treatment (Figure 1f,g). Lysosomal cathepsin activity, measured using MagicRed, was increased by hypoxia and also after adiponectin treatment in both normoxia and hypoxia conditions compared to control (Figure 1h).

FIGURE 1.

Adiponectin treatment enhanced autophagy flux in hypoxic cells in vitro. (a) DapRed fluorescence measured every hour in H9c2 cells during hypoxia or normoxia incubation with or without Ad (1 μg/mL). (b) DalGreen fluorescence measured every hour in H9c2 cells during hypoxia or normoxia incubation with or without Ad (1 μg/mL) for 24 h. (c) Representative Western blot images of p62, LC3 and GAPDH after hypoxia incubation with Ad for 12 h. (d, e) Densitometry analysis of p62 and LC3B‐II to GAPDH from 5C. (f) Representative confocal microscope images of H9c2 cells stained with DalGreen and DapRed in normoxia or hypoxia conditions with or without Ad for 12 h. (g) Quantification of the green to red ratio from 5F. (h) Representative confocal microscope images of H9c2 cells stained with MagicRed after hypoxia incubation with Ad for 12 h. (i) Quantification of MagicRed fluorescence from 5H. Results are presented as mean ± SEM (n = 3 per group). *p < 0.05, **p < 0.01, ***p < 0.001 versus control. Two‐way ANOVA with Bonferroni correction was run for statistical analysis. Scale bar = 20 μm. Ad, adiponectin; ANOVA, analysis of variance; Con, control; Hypo, hypoxia; Nor, normoxia.

Adiponectin mitigates hypoxia induced apoptosis in H9c2 cells

We measured temporal activation of caspase 3/7 in H9c2 cells using an activatable fluorescent probe in normoxia or hypoxia conditions with or without adiponectin treatment for up to 24 h. As expected, hypoxia increased cleavage of caspase 3/7 compared to normoxia conditions, with the change manifest from 12 h onward (Figure 2a). Adiponectin pretreatment attenuated this effect. Increased cleaved caspase 3 content after 24 h hypoxia was also indicated by Western blot (Figure 2b,c). Again, cotreatment with adiponectin reduced cleavage of caspase 3 (Figure 2b,c). To further confirm our results, H9c2 cells stably expressing the caspase activity reporter VC3AI were used to monitor caspase 3 activity in real time. Cleaved caspase 3 was increased in hypoxic cells and attenuated in the presence of adiponectin (Figure 2d,e). In addition, cell death measured using ReadyProbe Cell Viability kit specify was increased under hypoxia conditions and reduced in adiponectin treated cells (Figure 2f,g). Alterations in oxidative stress under these conditions were assessed, first via examining the percentage of MitoSox positive cells to assess changes in mitochondrial reactive oxygen species (ROS). We observed that hypoxia induced mitochondrial ROS in H9c2 cells was mitigated following adiponectin treatment as reflected by a decrease in the percentage of MitoSox positive cells (Figure S3A,C). Similar findings were observed with mBBr staining which measures cellular glutathione levels. Together, analysis of ROS production indicated that adiponectin treatment alleviated elevated oxidative stress levels in hypoxic conditions (Figure S3B,C). Therefore, under conditions of hypoxia, adiponectin treatment ameliorated apoptosis and reduced mitochondrial ROS and oxidative stress levels.

FIGURE 2.

Adiponectin mitigates apoptosis in H9c2 cells during hypoxia. (a) Caspase 3/7 fluorescence measured for 24 h every 1 h in normoxia or hypoxia conditions with or without Ad treatment. (b) Representative Western blot images of cleaved caspase 3 and β‐actin expression in H9c2 cells after hypoxic incubation with Ad (1 μg/mL) for 24 h. (c) Densitometry analysis of cleaved caspase 3 to β‐actin expression in H9c2 cells from Figure 6b. (d) Representative confocal microscope images of H9c2 cells stably expressing VC3AI showing cleaved caspase 3 (green) after hypoxia incubation with Ad (1 μg/mL) for 48 h. (e) Quantification of cleaved caspase 3 per cells from Figure 6d. (f) Representative confocal microscope images of H9c2 cells stained with ReadyProbe (green) after hypoxia incubation with or without Ad (1 μg/mL) for 24 h. (g) Quantification of cell death from Figure 6f. Results are presented as mean ± SEM (n = 3 per group for Figure 7a–j). **p < 0.01, ***p < 0.001 versus control. Two‐way ANOVA with Bonferroni correction was run for statistical analysis. Scale bar = 20 μm for Figure 6d and 5 μm for Figure 6f. Ad, adiponectin; ANOVA, analysis of variance; Con, control; Hypo, hypoxia; Nor, normoxia.

Adiponectin treatment does not limit apoptosis in autophagy deficient H9c2 cells

To determine whether adiponectin mitigates apoptosis through an autophagy‐dependent mechanism, autophagy‐deficient cells were generated using CRISPR mediated ATG7 deletion. First, to validate autophagy deficiency the cells were treated with rapamycin, an mTOR inhibitor well characterized to stimulate autophagy (Figure S4A–D). Control cells responded to rapamycin treatment and had increased LC3‐II and reduced p62 expression indicating enhanced autophagy flux (Figure S4A–D). In contrast, in ATG7‐KO cells there were no changes upon rapamycin treatment confirming impaired autophagy (Figure 3a–d). Using Western blot for cleaved caspase‐3, cells grown under normoxic conditions showed no significant changes (Figure 3a,b). After hypoxia treatment for 24 h, we observed activation of caspase‐3, and this was reduced by adiponectin in WT but not ATG7‐KO cells (Figure 3a,b). To further investigate this, cell death was measured using fluorescence staining with ReadyProbe Cell Viability kit and similar findings were observed (Figure 3c,d). Finally, flow cytometry using Annexin V, a marker for early apoptosis, and PI, a marker for necrosis or late apoptosis showed that in WT cells, hypoxia increased both early and late apoptosis (Figure 3e,f). Importantly, adiponectin treatment mitigated these increases (Figure 3e,f). In autophagy deficient cells, there was a significant increase in early and late apoptosis under hypoxia conditions, which was not attenuated by adiponectin treatment (Figure 3e,f). Collectively, these results indicate that adiponectin alleviates hypoxia‐induced apoptosis through the stimulation of autophagy.

FIGURE 3.

Hypoxia‐induced apoptosis is alleviated by apoptosis through its effect on autophagy. (a) Representative Western blot image of cleaved caspase 3 and β‐actin in WT or ATG7‐KO H9c2 cells after hypoxia incubation or normoxia incubation with or without Ad for 24 h. (b) Densitometry analysis of cleaved caspase 3 to β‐actin from Figure 7a. (c) Representative confocal microscope images of WT and ATG7‐KO H9c2 cells after hypoxia incubation with Ad treatment for 24 h stained with ReadyProbe (green). (d) Quantification of cell death from 7C. (e) Flow cytometry in WT and ATG7‐KO H9c2 cells after hypoxia or normoxia incubation with or without Ad treatment for 48 h stained with PI and annexin V. (f) Quantification of early and late apoptosis in WT and ATG7‐KO H9c2 cells after hypoxia incubation with Ad for 48 h. Results are presented as mean ± SEM (n = 3 per group for Figure 6a–c and h–j, n = 9 per group for Figure 6f,g). *p < 0.05, **p < 0.01, ***p < 0.001 versus control. Two‐way ANOVA with Bonferroni correction was run for statistical analysis. Scale bar = 5 μm. Ad, adiponectin; ANOVA, analysis of variance; KO, knock out; ns, not significant; PI, propidium iodide.

Adiponectin accumulation and signaling in the infarct zone is increased following 7 days of myocardial infarction in WT but not Ad‐KO mice

To determine the role of adiponectin in cardiac autophagy during ischemia and explore its significance in cardiac remodeling, WT and Ad‐KO mice at 12 weeks of age were subjected to coronary artery ligation surgery without reperfusion for 7 days. Western blot was performed on the ischemic zone of the LV heart tissue and adiponectin was expressed in WT but not Ad‐KO mice as expected, and its expression was increased after 7 days of MI in WT mice (Figure 4a,b). Immunofluorescence staining on heart tissue sections provided further qualitative evidence that adiponectin expression was increased following MI in WT, with no expression observed in Ad‐KO mice (Figure 4c). To assess changes in mRNA expression, the mRNA was isolated from heart tissues after 7 days of MI. No changes were observed in adiponectin mRNA expression after inducing MI in WT mice compared to sham mice, indicating that the changes observed in the protein level are not due to elevated adiponectin gene expression (Figure 4d). Furthermore, enzyme‐linked immunosorbent assay (ELISA) quantification revealed no differences in serum adiponectin levels after 1 day or 7 days of MI compared to sham treated mice (Figure 4e). To observe whether downstream signaling pathways of adiponectin were altered, Western blots were performed on heart tissue homogenates. As expected, the increased adiponectin expression observed in WT mice following 7 days of MI was accompanied by higher phosphorylated ULK1, a protein kinase that is important in initiating autophagy (Figure 4f,g). Basal ULK1 phosphorylation was reduced in Ad‐KO mice and not increased upon MI. Interestingly, AMPK was activated following MI in both WT mice and Ad‐KO mice (Figure 4h) which suggests that AMPK could be activated by other signals in Ad‐KO mice. Therefore, MI increases adiponectin accumulation in the infarct zone, which then activates its downstream targets, including ULK1 and AMPK leading to the initiation of autophagy.

FIGURE 4.

Accumulation of regional adiponectin at infarct zone after ischemia. (a) Representative western blot image of adiponectin and GAPDH in apex heart homogenates after 7 days (7D) MI. (b) Densitometry analysis of adiponectin over GAPDH in 1 day (1D) ischemia hearts. (c) Representative immunofluorescence images showing adiponectin (red) and WGA staining (green) in paraffin embedded heart after CAL. (d) Relative adiponectin mRNA expression over 18srRNA in heart apex after 7D MI. (e) ELISA quantification of adiponectin in serum after 1 day or 7 days MI. (f) Representative Western blot images of p‐AMPK T172, p‐ULK1 S555, and GAPDH in apex heart homogenates after 1D MI. (g–i) Densitometry analysis of adiponectin, p‐AMPK T172 and p‐ULK1 S555 to GAPDH from Figure 1f. Results are presented as mean ± SEM (n = 5 per group for 7 days ischemia, n = 4 per group for 1 day ischemia, biological replicates). *p < 0.05 versus corresponding sham, ^# p < 0.05 versus WT MI. Two‐way ANOVA with Bonferroni correction was performed in Figure 1g–i. Unpaired student t‐test was run for statistical analysis in Figure 1b,d,e. Scale bar = 100 μm. Ad‐KO, adiponectin‐knock out; ANOVA, analysis of variance; CAL, coronary artery ligation; MI, myocardial infarction; ns, not significant; WGA, wheat germ agglutinin; WT, wild type.

Validation of fluorescence molecular tomography detection of cathepsin B (CatB 680) activity as a readout of myocardial autophagy in WT and autophagy‐deficient mice

To validate the use of cathepsin B activity analysis by FMT as a noninvasive indicator of myocardial autophagy flux, we compared control mice and those with cardiomyocyte autophagy deficiency. Inducible cardiomyocyte‐specific ATG7‐deficient or WT mice were treated with rapamycin and starvation for 24 h to induce autophagy. Representative ex vivo FMT images of hearts showed that cathepsin B activity was significantly induced by rapamycin and starvation treatment in WT animals, but not in the autophagy‐deficient group (Figure 5a,b). Whole heart homogenates were analyzed by Western blotting for validated markers of autophagy, p62, LC3‐I and LC3‐II (Figure 5c–g). Elevated LC3‐II levels indicate increased autophagosome content, an observation which could arise due to elevated initiation or suppressed flux (Figure 5c–e). Increased p62 levels in autophagy‐deficient mice indicated that flux was reduced (Figure 5c,f). This correlates with the data obtained using CatB as an indicator of autophagy flux. Finally, inducible cardiomyocyte‐specific ATG7‐deficiency was confirmed by the reduction of total ATG7 protein expression in heart homogenates (Figure 5c,g). Residual levels of ATG7 and the small increase seen in ATG7−/− mice (Figure 5a) could be explained as a result signal coming from other myocardial cell types. Overall, our data indicates that myocardial autophagy activation can be assessed not only by well‐known autophagy markers, p62 and LC3‐II conversion, but also by CatB activity. After we validated the use of CatB activity analysis by FMT, we also used this method to measure the effect of ischemia in WT and Ad‐KO mice. Representative FMT images and four optical sections through the heart, are shown for each group (Figure 5h). The increased fluorescence signal in the hearts of WT but not Ad‐KO mice after MI indicates augmented lysosomal CatB activity. Fluorescent molecular tomography imaging was also performed on hearts ex vivo (Figure 5i) and data again revealed higher lysosomal activity in WT mice compared to Ad‐KO mice (Figure 5i,j). Hence, Ad‐KO mice have disrupted autophagy flux when compared to WT after chronic myocardial ischemia.

FIGURE 5.

Validation of detection of myocardial autophagy flux via cathepsin B activity. Ischemia impairs the autophagy response in Ad‐KO mice. (a) To elevate autophagy flux, Tamoxifen‐induced cardiomyocyte specific ATG7‐deficient and control mice were treated with rapamycin and starved for 24 h. All mice were infused with a cathepsin B 680 probe. Ex vivo FMT 2‐dimensional analysis using isolated hearts is shown. * Indicates versus WT‐control and # indicates versus WT‐“rapamycin and starvation” condition and (b) quantification (n = 3, *^,# p < 0.05). (c–g) Whole heart homogenates were analyzed by Western blotting with antibodies for validated autophagy markers, p62 and LC3. To confirm induction of cardiomyocyte‐specific ATG7‐deletion, ATG7 antibody was used for Western blotting. (c) The representative images shown with their own internal control, GAPDH and (d–g) quantitative analysis in graphs with folding over control. In (d–g), n = 3 and * indicates versus WT‐control. # and & indicates versus WT‐“rapamycin and starvation.” (*^,#,& p < 0.05). (h) Representative live FMT (ProSense 680, pan‐cathepsin protease sensor) in animals after MI. (i) Representative ex vivo FMT scan. (j) Quantification of Figure 2i. Results are presented as mean ± SEM (n = 3 per group for (h–k), biological replicates). *p < 0.05 versus corresponding sham. Two‐way ANOVA with Bonferroni correction was run for statistical analysis. Ad‐KO, adiponectin‐knock out; ANOVA, analysis of variance; FMT, fluorescence molecular tomography; MI, myocardial infarction; WT, wild type.

Myocardial infarction‐induced autophagy flux is attenuated in Ad‐KO mice

Next, we explored whether loss of adiponectin expression disrupts autophagy related processes in mice following MI. LC3B immunofluorescence efficacy was first validated in H9c2 rat embryonic cardiac myoblasts using Torin‐1, an mTOR inhibitor used as a positive control to simulate autophagy. H9c2 cells displayed a red fluorescent punctate signal when treated with Torin‐1 and stained with LC3B or probed with Cyto‐ID, indicating autophagosome formation (Figure S1A). To observe changes in key autophagy‐related proteins, Western blot was performed on infarct region homogenates (Figure 6a–d). Beclin 1, an early marker of autophagy, was increased in both WT and Ad‐KO mice subjected to MI compared to their respective sham controls (Figure 6a,b). LC3B‐II levels were higher in Ad‐KO versus WT mice and decreased after MIinfarction (Figure 6a,c). MI significantly increased p62 levels in Ad‐KO but not WT mice (Figure 6a,d).

FIGURE 6.

Ad‐KO mice have reduced ejection fraction, impaired cardiac function, increased autophagy and apoptosis after ischemia. (a) Representative western blot images of Beclin‐1, LC3B, p62, and GAPDH from apex heart homogenates after MI. (b–d) Densitometry analysis of Beclin‐1, LC3‐II, and p62 to GAPDH from Figure 2a. (e) Representative TUNEL staining images in heart sections from WT and Ad‐KO mice with or without ischemia surgery. (f) Quantification of TUNEL positive nuclei from Figure 3g. (g) Representative western blot images of BAX, Bcl2, Lcn2 (lipocalin 2), and GAPDH from apex hearts after MI. (h, i) Densitometry analysis of Bcl2 over BAX and Lcn2 over GAPDH. (j) Gross heart weight to tibial length ratio in WT and Ad‐KO mice after 7 days MI. (k) Representative M‐mode images of heart after MI. (l) Ejection fraction measured from 3B. (m) Representative images of B‐mode showing changes in strains over three cardiac cycles in radial and longitudinal directions from hearts with or without ischemia surgery. (n, o) Quantification of average strains from six segments in Figure 3d. Results are presented as mean ± SEM (n = 5 per group, biological replicates). *p < 0.05 versus corresponding sham. ^# p < 0.05 versus WT Ischemia. Two‐way ANOVA with Bonferroni correction was run for statistical analysis. Scale bar = 50 μm. Ad‐KO, adiponectin‐knock out; ANOVA, analysis of variance; MI, myocardial infarction; WT, wild type.

Ad‐KO mice had exaggerated infarction‐induced cell death and cardiac dysfunction versus WT mice

We next determined whether loss of adiponectin expression can influence ischemia‐induced apoptosis and cardiac dysfunction. To assess cell death, TUNEL staining was performed on heart sections and the extent of cell death was elevated in Ad‐KO versus WT mice both before and after infarction (Figure 6e,f). Western blot revealed that the infarction‐induced decrease in Bcl2 (anti‐apoptotic) to BAX (pro‐apoptotic) ratio, an index of apoptosis, observed in WT mice was exacerbated in Ad‐KO mice (Figure 6g,h). In addition, the inflammatory and pro‐apoptotic protein lipocalin‐2 (Lcn2), was increased in Ad‐KO mice to a greater extent than in WT mice after myocardial infarction (Figure 6g,i). These data suggest that apoptosis due to prolonged ischemia was exacerbated in Ad‐KO mice. The increased heart weight to tibia length found in Ad‐KO mice with MI compared to WT indicated exaggerated cardiac hypertrophy (Figure 6j). In keeping with this, we also found increased cardiomyocyte hypertrophy (Figure S2C,D). Echocardiography was used to examine cardiac function and myocardial strain in Ad‐KO compared to WT mice. MI caused a small significant reduction in ejection fraction in both WT and Ad‐KO compared to sham mice (Figure 6k). Statistical analysis indicated the reduction in Ad‐KO was significantly greater than in WT mice, suggesting that the presence of adiponectin is important for protecting cardiac function following ischemia (Figure 6k). Speckle tracking echocardiography was used to examine radial and longitudinal strain, with data further confirming that Ad‐KO mice had significant cardiac dysfunction compared to their respective WT controls after MI (Figure 6l,m). Consistent with impaired cardiac function, fibrosis was aggravated in Ad‐KO compared to WT mice (Figure S2A,B). These data confirm that presence of adiponectin reduced the impact of ischemia on cardiac function and cardiac remodeling.

Adiponectin stimulates autophagy in primary adult rat cardiomyocytes and H9c2 cardiac myoblasts

The previous results showed loss of adiponectin expression disrupts induction of autophagy in response to ischemia. We next determined whether adiponectin can directly stimulate autophagy in primary adult cardiomyocytes and H9c2 cardiac myoblasts. We found that the autophagy markers, pULK1, pAMPK, and LC3‐II were all increased following treatment of H9c2 cells with adiponectin for 30 min (Figure 7a–d). To further examine direct effects of adiponectin on autophagy, the probe Cyto‐ID was first validated in H9c2 cells treated with Torin‐1. Quantification of Cyto‐ID in primary cardiomyocytes (Figure 7e) and H9c2 cells (Figure 7e,f) revealed a higher number of autophagosomes per cell following adiponectin treatment. Similarly, lysosomal cathepsin activity measured by Magic Red in H9c2 cells was stimulated by adiponectin treatment (Figure 7e,f). Therefore, adiponectin directly stimulated autophagy flux in cardiomyocytes.

FIGURE 7.

Autophagy is stimulated by adiponectin treatment in vitro in cardiomyocytes and H9c2 cells. (a) Representative Western blot images of p‐ULK1 S555, p‐AMPK T172, LC3B, and β‐actin in H9c2 cells after adiponectin stimulation (1 μg/mL, 30 min). (b–d) Quantification of p‐ULK1 S555, p‐AMPK T172, LC3‐l, and LC‐ll over β‐actin from Figure 4a. (e) Representative confocal microscope images of rat adult cardiomyocytes and H9c2 cells stained with Cyto‐ID and Magic Red after Ad treatment (1 μg/mL) for 30 min. (f) Quantification of Cyto‐ID and Magic Red puncta in 4E for H9c2 cells. Results are presented as mean ± SEM (n = 4 per group for a–d, n = 9 per group for e and f). *p < 0.05 versus control. One‐way ANOVA with Dunnett post‐test was performed for statistical analysis in Figure 4b–d. Unpaired student's t‐test was performed for statistical analysis in Figure 4f. Scale bar = 50 μm. Ad‐KO, adiponectin‐knock out; ANOVA, analysis of variance.

DISCUSSION

Many clinical studies have demonstrated an inverse correlation of circulating adiponectin levels with the development and progression of various forms of cardiovascular diseases. ¹² , ¹³ , ¹⁴ Ad‐KO mice have exacerbated cardiac remodeling and dysfunction in various models of heart failure, which could be prevented by restoring normal adiponectin levels. ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ The mechanisms behind these effects remain to be fully defined. We and others have now demonstrated that adiponectin stimulates autophagy flux in cardiomyocytes. ²¹ , ³² , ³³ The link between autophagy and cell death in models of MI has been previously established ⁴ , ³⁴ and the majority of studies indicate that induction of autophagy is a protective mechanism that reduces the extent of ischemia‐induced acute cell damage and also reduces the extent of postinfarction cardiac remodeling and dysfunction. ⁶ , ³⁵ Nevertheless, it should also be noted that it has been demonstrated that cardiomyocytes can go through various forms of cell death during MI and that excess autophagy may in fact contribute to autophagic cell death. ³⁶ The present study was designed to examine the functional importance of adiponectin‐stimulated myocardial cardiac autophagy in mice during ischemia. Parallel studies were conducted in a cellular model using hypoxia.

Here, we observed that in WT mice, adiponectin accumulates in the damaged myocardium. In keeping with our observation, previous work indicated that the adiponectin level was increased in LV post‐MI in a rat model of heart failure. ³⁷ In that study, there was also a decreased adiponectin levels in plasma, which we did not observe in this study. Furthermore, another study in mice showed that adiponectin accumulates in the heart following ischemic damage and that it principally occurred via leakage from the vascular compartment. ²⁰ Our data supports this notion because we saw elevated adiponectin protein levels in the heart but did not observe any change in mRNA expression. It is also important to point out that the local production by cell types in the myocardium is orders of magnitude smaller than that from adipose tissue entering the circulation, and thus circulating levels do not change significantly even though there is detectable local accumulation in the heart. This localized accumulation of adiponectin may be important in allowing enhanced local signaling leading to downstream consequences, such as stimulation of autophagy. Indeed, the accumulation of adiponectin was associated with significantly increased AMPK and ULK1 (Ser 555) phosphorylation in hearts after 7 days of ischemia, indicating activation of adiponectin targets.

We also developed a noninvasive fluorescent molecular tomography‐based protocol using a near infrared (680 nm) activatable probe to assess lysosomal cathepsin activity as a surrogate for assessing autophagy flux in live mice. The concept for this approach is similar to a novel method which was published recently. ³⁸ , ³⁹ First, to validate the correlation between myocardial Cat‐B 680 signal and autophagy, we began our study using mice with inducible cardiomyocyte autophagy‐deficiency caused by deletion of Atg7. We believe that this approach represents a very useful step forward in allowing researchers to gauge temporal changes in autophagy flux via a noninvasive approach, whereas acknowledging that final conclusions should be supported by established biochemical approaches. This data indicated that hearts of Ad‐KO mice had less lysosomal degradative activity compared with WT mice after ischemia. Ad‐KO mice cannot accumulate adiponectin in damaged myocardium and have a disrupted autophagy flux when compared to WT after chronic myocardial ischemia. Other studies have also indicated that adiponectin can directly stimulate myocardial autophagy and consequently confer cardioprotection. ²¹ , ³² , ³³

Using echocardiography, including strain analysis, we observed that mice lacking adiponectin had an enhanced degree of cardiac dysfunction. Correlated with this was a significant increase in indices of cell death, such as more TUNEL positive cells and increased Bcl2:BAX ratio in Ad‐KO versus WT mice after ischemia. These data were expected because the ability of adiponectin to attenuate myocardial cell death, or adiponectin deletion to exacerbate cell death, is well‐established. ⁴⁰ , ⁴¹ , ⁴² We also discovered in this study that induced ischemia led to increased lipocalin‐2 accumulation in the heart, which was further elevated in Ad‐KO mice. We previously reported that lipocalin‐2, a pro‐inflammatory cytokine mainly produced by neutrophils, causes cell death by enhancing oxidative stress and reducing autophagy flux in cardiomyocytes. ²⁶ , ²⁷ , ⁴³ Hence, it is likely that adiponectin counteracts these effects of lipocalin‐2 in WT mice. This represents one mechanism contributing to the anti‐inflammatory effects of adiponectin and our data suggests that this emerging theme of crosstalk between adiponectin and lipocalin‐2 in models of prolonged ischemia or ischemia–reperfusion should be further explored. ⁴⁴ , ⁴⁵ , ⁴⁶

To further investigate our observations from studies in WT and Ad‐KO mice, we also used the cellular models of primary adult rat cardiomyocytes or H9c2 cells subjected to hypoxia. Previous studies have indicated that adiponectin reduced apoptosis induced by chronic intermittent hypoxia and inhibited hypoxia/reoxygenation‐induced necroptosis and intrinsic apoptosis via AdipoR1‐APPL1‐AMPK signaling in H9c2 cells. ⁴⁷ , ⁴⁸ We found that autophagy flux was enhanced following adiponectin treatment in normoxic or hypoxic conditions. Importantly, to verify the functional significance of adiponectin‐stimulated autophagy, we generated an autophagy‐deficient cell model using CRISPR to delete Atg7. Whereas adiponectin reduced hypoxia‐induced cell death in control cells, this effect was lost in Atg7‐KO cells. A recent study also concluded that adiponectin induced ER‐phagy to attenuate hypoxia‐induced apoptosis in H9C2 cardiomyocytes. ⁴⁷ The same concept has been demonstrated in a model of resuscitation‐induced brain apoptosis, where adiponectin attenuated damage in an autophagy‐dependent manner. ³² Adiponectin was known reduce apoptosis caused by H₂O₂ in chondrocytes via stimulating autophagy. ⁴⁹ Furthermore, adiponectin protected hepatocytes from acetaminophen‐induced cell via promoting autophagy. ⁵⁰ The verification of the functional significance of autophagy in the beneficial anti‐apoptotic effects of adiponectin is an important result from this study.

In conclusion, our study demonstrates that adiponectin is an important mediator of cardiac autophagy. We first evaluated the role of adiponectin in regulating ischemic remodeling using a permanent coronary artery ligation model. We reported that mice lacking adiponectin subjected to MI have an impaired autophagy response, increased cellular apoptosis, and exacerbated cardiac dysfunction. Treatment of H9c2 cells with adiponectin increased autophagy flux and mitigated apoptosis under hypoxia conditions. Finally, lack of adiponectin's beneficial effect in the autophagy deficient cell line engineered by CRISPR‐mediated ATG7 knockout, indicated that adiponectin mitigated hypoxia‐induced apoptosis in an autophagy‐dependent manner. Therefore, we demonstrated the importance of an adiponectin‐autophagy axis in mitigating ischemia or hypoxia induced capase‐3 activation and cell death. We have summarized these findings in a schematic diagram (Figure 8). This study further reinforces the potential of adiponectin as a therapeutic target to improve cardiac outcomes under ischemic conditions.

FIGURE 8.

Adiponectin plays a pivotal role in protecting against ischemic injury via stimulating cardiac autophagy. In a permanent coronary artery ligation model, adiponectin‐deficient mice exhibited an impaired autophagy response, measured via ULK1/AMPK phosphorylation and altered levels of autophagy markers LC3 (depicted as green dot) and p62 (red dot). In this way, adiponectin provided protective effects by mitigating ischemia/hypoxia‐induced ROS production and cell death. Ad‐KO, adiponectin‐knock out; ROS, reactive oxygen species.

AUTHOR CONTRIBUTIONS

H.K.S. wrote the manuscript. H.K.S., J.W.S.J., and G.S. designed the research. J.T., J.W.S.J., E.S., Y.K.C., and A.H.L. performed the research. J.P. and A.A.S. analyzed the data.

CONFLICT OF INTEREST STATEMENT

The authors declared no competing interests for this work.

Supporting information

Figure S1

Figure S2

Figure S3

Figure S4

Sung HK, Tang J, Jahng JWS, et al. Ischemia‐induced cardiac dysfunction is exacerbated in adiponectin‐knockout mice due to impaired autophagy flux. Clin Transl Sci. 2024;17:e13758. doi: 10.1111/cts.13758

DATA AVAILABILITY STATEMENT

Original data for this study are available from the authors upon reasonable request.