Introduction

It is estimated that approximately 30 million people worldwide die from ischemic heart disease each year, and by 2030, this fatal disease is projected to become the second leading cause of death1. Restoring the blood flow of ischemic myocardia in early stage of acute myocardial infarction can effectively limit the infarct size and reduce the risk of death2. However, ischemic myocardium may suffer more severe damage when interrupted blood supply is restored, a phenomenon called myocardial ischemia/reperfusion injury (MI/RI)3,4,5. Various studies have demonstrated that the pathophysiological mechanisms underlying MI/RI involve oxidative stress, endoplasmic reticulum stress, intracellular calcium overload, energy metabolism disorders, autophagy, apoptosis, necrosis, pyroptosis, and ferroptosis6. While numerous strategies have been proposed to mitigate MI/RI in animal and cellular models based on these potential mechanisms, these approaches have proven inadequate in reducing the size of myocardial infarction in clinical settings4,7. Given the complexity of the mechanisms involved in MI/RI, it is crucial to urgently identify novel intervention targets associated with the etiology of MI/RI and develop alternative cardioprotective strategies.

Brown adipose tissue (BAT) plays a critical role in the regulation of cardiovascular disease8. BAT is primarily involved in energy expenditure and the prevention of metabolic disorders through uncoupled respiration and thermogenesis9,10. Recently, studies have shown that BAT can be used as a secretory organ to secrete adipokines to protect or regulate the cardiovascular system11. Fibroblast growth factor 21 (FGF21), interleukin-6 (IL-6), neuregulinin-4 (NRG 4) and vascular endothelial growth factor A (VEGFA) secreted from BAT are involved in thermogenesis and angiogenesis of WAT in an autocratic and/or paracrine manner12,13,14,15,16. Small extracellular vesicles secreted from BAT are also implicated in exercise-induced cardioprotection, particularly in the context of MI/RI17. FGF21, a member of the fibroblast growth factor family, plays a significant role in substance metabolism regulation18. As an important regulator of substance metabolism, FGF21 not only indirectly delays the development of coronary heart disease by improving the risk factors of coronary heart disease such as lipid metabolism disorders, insulin resistance and obesity, but also directly participates in the process of coronary heart disease by becoming a predictive or even therapeutic factor for coronary heart disease19. The benefits of FGF21 have received increasing attention in various aspects of coronary heart disease, including acute myocardial infarction and MI/RI20. Activation of adenosine A2A receptors on BAT promotes the expression of FGF21, which contributes to protection against hypertensive cardiac remodeling21. FGF21 protects myocardial against I/R injury by increasing miR-145 levels and promoting autophagy22. Analogously, FGF21 promotes autophagic flux by upregulating the expression levels of Beclin-1 and Vps34 proteins so as to alleviate hypoxia/reoxygenation (H/R) injury in H9C2 cardiomyocytes23. In addition, FGF21 regulates FGFR-EGR1 pathway to inhibit the inflammation and fibrosis in post-infarcted hearts24.

Dexmedetomidine (DEX), a highly selective α-adrenergic receptor agonist, is widely used in clinical practice. DEX can be used in combination with other narcotic drugs to achieve better analgesia, sedation and adverse stress control during coronary artery bypass surgery25. Studies have shown that DEX could promote cardiomyocyte survival by activating PI3K/AKT, SIRT1/mTOR, MAPK/ERK and AMPK signaling pathways26,27,28,29. Similarly, our previous work demonstrates that DEX protects against MI/RI by inhibiting ferroptosis by enhancing the expression of SLC7A11 and GPX430. A recent study has disclosed that DEX reduces myocardial infarction area and facilitates the recovery of left ventricular function in MI/RI mice by inducing eNOS activation and subsequent NO production31. Surprisingly, treatment with DEX was unable to reduce cell damage in isolated myocardial cells after H/R insult31. However, cardiomyocyte injury induced by H/R stimulation was markedly suppressed when co-culturing them with endothelial cells pre-treated with DEX31, indicating that the protection of DEX against MI/RI requires the interaction of cardiomyocytes with the endothelium. Nevertheless, it remains unknown whether BAT participates in the cardiovascular benefits of DEX through the secretion of FGF21. Therefore, we sought to explore whether DEX alleviated MI/RI in mice through the crosstalk between BAT and hearts. Moreover, the underlying molecular mechanism of action of BAT-secreted FGF21 on cardiomyocyte biology was also explored.

Methods and materials

Animals

Male C57BL/6J mice aged 7 ~ 8 weeks were procured from the SPF Biotechnology Co., ltd (Beijing, China). All mice had free access to standard chow and water under a temperature and humidity room on a 12 h light/dark cycle. All procedures for animal studies were reviewed and approved by the Ethics Committee of Jiangnan University (JN.No20220915c1081231[360]). All methods were performed in accordance with the ARRIVAL guidelines and relevant regulations and animal experiments took place in Jiangnan University. The experiments were complied with the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication, revised 2011), and the ethical standards in the 1964 Declaration of Helsinki. All mice were randomly divided into eight groups: Control (Con) group, ischemia/reperfusion (I/R) group, Dexmedetomidine preconditioning (DEX)group, Dexmedetomidine preconditioning + ischemia/reperfusion (I/R + DEX) group, BAT ablation + Dexmedetomidine preconditioning + ischemia/reperfusion (I/R + DEX + BAT ablation), FGF21 neutralizing antibody + Dexmedetomidine preconditioning + ischemia/reperfusion (I/R + DEX + FGF21-Ab) group, Compound C + Dexmedetomidine preconditioning + ischemia/reperfusion (I/R + DEX + Compound C), SR-18292 + Dexmedetomidine preconditioning + ischemia/reperfusion (I/R + DEX + SR-18292) group. Only sham surgery was performed in the Con group, the DEX group was subjected to daily intraperitoneal injection of Dexmedetomidine (14.28 μg/kg/day, totally for 100 μg) for 7 days before the sham operation or I/R. The I/R + DEX + BAT ablation group received iBAT resection for 48 h, and treated with intraperitoneal injection of Dexmedetomidine (14.28 μg/kg/day) for 7 days before the I/R operation. The I/R + DEX + FGF21-Ab group received FGF21 neutralizing antibody (0.3 mg/kg body weight) and Dexmedetomidine (14.28 μg/kg/day) for 7 days before the I/R operation. The I/R + DEX + Compound C group was intraperitoneally injected with Coumpound C at a dose of 25 mg/kg for 48 h, and treated with intraperitoneal injection of Dexmedetomidine (14.28 μg/kg/day) for 7 days before the I/R operation. I/R + DEX + SR-18292 group received SR-18292 at a dose of 45 mg/kg and Dexmedetomidine (14.28 μg/kg/day) for 7 consecutive days by intraperitoneal injection, and MI/R surgery was performed 3 h after injection on day 732. For euthanasia, animals were anesthetized with 5% isoflurane, anaesthesia was confirmed via tail pinch, and then sacrificed by cervical dislocation before cardiac tissue removal.

Myocardial ischemia/reperfusion injury model

The mice were anesthetized with sevoflurane and the MI/R surgery was established as previously described33. Following left thoracotomy at the fourth intercostal, the left main descending coronary artery (LAD) was ligated using a 6–0 silk suture. Ligation was considered successful when the anterior wall of the left ventricle turned pale. Whereafter, the heart is immediately placed back in the chest followed by closing the chest in layers. After 30 min of ischemia, the slipknot was released to induce reperfusion for 24 h. The control group underwent the same surgical procedure without ligation of the LAD.

Surgical BAT ablation

Mice were anesthetized with 1% anesthetized Sodium (50 mg/kg body weight) and fixed in a prone position. An approximate 1.5 cm incision was cut in the interscapular region, and the iBAT was carefully removed and the incision was sutured loosely. Sham control underwent all the surgical procedures except the surgical ablation of BAT. The mice recovered for 48 h and received the sham or MI/R surgery.

Evaluation of echocardiography

Mice were anesthetized with 1.5% isoflurane after the 24 h reperfusion, and echocardiography was performed using the Vevo high-resolution imaging system (Fujifilm, Japan). A two-dimensional M-mode image was used to evaluate the left ventricular function. The ejection fraction (EF%) and fractional shortening (FS%) parameters were calculated using the computerized algorithms. The results were averaged from the five consecutive cardiac cycles measured from the M-mode images.

Determination of myocardial infarct size

Once the reperfusion reached an end, the heart was removed quickly, rinsed with normal saline and frozen at − 20 °C for 30 min. Then, the heart was equally sliced into four pieces from tip to bottom. The slices were dyed by 2% diphenyltetrazolium chloride (TTC) solution at 37 °C for 15 min without light and the active myocardial was represented as brick red. After fixation with a 4% paraformaldehyde solution for 24 h, the heat sections were captured and the infarct area was calculated using image software (IMAGE J, NIH, USA). The infarct size was defined as the ratio of the infarct area to the risk one.

Hematoxylin–eosin (H&E) staining

The mice were sacrificed after 24 reperfusion and their hearts were taken and fixed with 4% paraformaldehyde. The heart was then encased in paraffin, sectioned into slices (5 μm), and separately stained with standard H&E histology (No. C0105S, Beyotime, China). Morphological characteristics of heart section were observed using an optical microscope with a 40 × objective.

TUNEL detection

Cardiomyocyte apoptosis was detected by TUNEL assay kit (KTA2010, Abokine, China). The apoptosis rate was represented by the percentage of TUNEL positive cardiomyocytes in the total cells of the fields.

Cell culture and cardiomyocyte hypoxia/reoxygenation (H/R) model

Cardiomyocyte hypoxia reoxygenation model was established as follows: cardiomyocytes (HL-1) were cultured in the DMEM without glucose in a humidified atmosphere of 5% CO2, 94% N2 and 1% O2 at 37 °C for 6 h, followed by incubation under normoxia (95% air and 5% CO2) for 24 h in the DMEM with normal glucose (5.5 mM) containing 10% FBS. Brown adipocytes (HIB 1B) were cultured in the DMEM supplemented with 10% FBS.

Cell viability

The cell viability of cardiomyocytes was detected by CCK-8 kits. Cardiomyocytes were seeded at a density of 1.0 × 104 cells/well in a 96-well plate and grown for 24 h. After H/R treatment, CCK-8 solution (10 μL, Biosharp) was added to each well of the plate and the cells were incubated at 37 °C for 2 h. The OD value in each well was detected at 450 nm by a microplate reader (BioTek Instruments, USA). A LDH release assay kit was utilized to detect the LDH from cells in which the absorbance was measured at 490 nm. Cell viability and lactate dehydrogenase release were normalized in the control group.

Enzyme-linked immunosorbent assay (ELISA)

The levels of CK-MB (SBJ-M1037, Sbjbio, China) and mouse FGF21 (E-EL-M0029c, Elabscience, China) in the serum were measured by ELISA, according to the manufacturer’s instructions.

Dihydroethidium (DHE) staining

Intracellular ROS levels were measured by DHE staining. Fixed cardiomyocytes were incubated with DHE probe (10 nM) away from light for 30 min at 37 °C. Then the fluorescence images were photographed by microscope.

Flow cytometry

Cardiomyocytes were seeded at 1.0 × 106 cells/well in a 6-well plate and grown for 24 h, then stimulated by H/R treatment. Cell apoptosis was detected using an Annexin V-FITC/PI Apoptosis Detection Kit (C1062M, Beyotime, China) according to the manufacturer’s instructions. The cells with positive Annexin V-FITC staining and negative PI staining were apoptotic cells. Cell apoptosis rate was analyzed with BD FACSAria III flow cytometry. The data were processed by FlowJo software.

Quantitative real-time PCR

Total RNA in cardiac tissues, brown adipose tissues and cells were harvested using Trizol reagents (15322, CWBIO, China). The equal contents of RNA (1 μg) were reverse transcribed using the cDNA Synthesis SuperMix kit (11141ES60, YEASEN Biotechnology, China). Real-time PCR was performed using the qPCR SYBR Green Master Mix (11202ES08, YEASEN Biotechnology, China). The sequences of primers used are the following: β-actin, Forward-5′GCTACAGCTTCACCACCACAG3′, Reverse-5′GGTCTTTACGGATGTCAACGTC3′; FGF21, Forward-5′AAGACACTGAAGCCCACCTG3′, Reverse-5′ATCAAAGTGAGGCGATCCAT3′; UCP1, Forward-5′GTACCAAGCTGTGCGATGTC3′, Reverse-5′ATTCGTGGTCTCCCAGCATA3′; PPARr, Forward-5′AATCCACGAAGCCTACCTGA′, Reverse-5′AATCGGACCTCTGCCTCTTT3′; PRDM16, Forward-5′GCTCGTGTTTAGCGGTCATTA′, Reverse-5′GCGTCTTCTCGATTCACG3′; CIDEA, Forward-5′CTCGGCTGTCTCAATGTCAA′, Reverse-5′CCGCATAGACCAGGAACTGT3′; HO-1, Forward-5′AAGCCGAGAATGCTGAGTTCA′, Reverse-5′GCCGTGTAGATATGGTACAAGA3′; NQO1, Forward-5′TGGCCGAACACAAGAAGCTG′, Reverse-5′GCTACGAGCACTCTCTCAAACC3′; GCLM, Forward-5′AGGAGCTTCGGGACTGTATCC′, Reverse-5′GGGACATGGTGCATTCCAAAA3′; GCLC, Forward-5′GGACAAACCCCAACCATCC′, Reverse-5′GTTGAACTCAGACATCGTTCCT3′. The relative expression of target genes was calculated by the 2–∆∆Ct method.

Western blot

Proteins of cardiac tissues, brown adipose tissues and cells were extracted by lysis buffer (P0013, Beyotime, China). Proteins (30 μg) were separated by polyacrylamide gel electrophoresis (8%, 10% or 12% SDS-PAGE gel) and transferred to a PVDF membrane (IPVH00010, Millipore, USA). Afterward, the membranes were blocked for 1 h at room temperature 5% non-fat dry milk (5 g skim milk powder in 100 ml TBST). Then the membranes were incubated with primary antibodies such as anti-β-actin (4970T, Cell signaling technology), anti-FGF21 (26272-1-AP, Proteintech), anti-Bcl-2 (BM4985, Boster), anti-Bax (A00183, Boster), anti-cleaved caspase 3 (9661S, Cell Signaling Technology), anti-AMPK (A00994-6, Boster), anti-p-AMPK (AF3423, Affinity), anti-PGC1α (sc-518025, Santa Cruz Biotechnology), anti-Keap1 (10503-2-AP, Proteintech), anti-Nrf2 (16396-1-AP, Proteintech) antibodies at 4 °C, followed by incubation with HRP-conjugated secondary antibodies for 2 h at room temperature. Expose and develop the membranes after treatment with the developing solution; the gray value analysis is performed by Image J, and each experiment was in triplicate.

Co-immunoprecipitation

The cell lysates were immunoprecipitated with the anti-Keap1 antibodies (1 μg) for 2 h at 4 °C and then incubated with protein G PLUS-Agarose (20 μl) at 4 °C on a shaking table overnight. After centrifugation, the supernatant was discarded and the precipitate was resuspended in protein loading buffer (40 μl) and boiled for 10 min. Subsequently, proteins were separated by SDS-PAGE, transferred to PVDF membranes, blocked, incubated with primary antibodies and secondary antibody as described above.

Molecular docking

The 3D structure of dexmedetomidine for molecular docking was built using Discovery Studio. The crystal structure of AMPK protein was then obtained from the RCSB Protein Data Bank database. Molecular docking calculations were executed by the LibDock program in Dock Ligands. In addition, PyMOL was used to complete the molecular binding pattern diagram of AMPK and DEX.

Statistical analysis

All data are calculated as the mean ± standard deviation. Differences between groups were determined by One-way ANOVA followed by Bonferroni’s multiple comparisons test. SPSS version 19 statistical software (IBM Corp.) was used for all statistical analysis. The P value < 0.05 was considered as statistically significant.

Ethics approval and consent to participate

All procedures for animal studies were reviewed and approved by the Ethics Committee of Jiangnan University (JN.No20220915c1081231[360]).

Consent for publication

All the authors are consent for publication this paper.

Results

DEX protects against MI/RI via promoting iBAT-released FGF21

To investigate the cardioprotective effects of DEX against MI/RI, we established a mouse model of MI/RI to assess cardiac function and infarct sizes. Consistent with previous reports34, the DEX-administered group exhibited preserved left ventricular systolic function compared to the vehicle-treated group subjected to MI/RI, as monitored by cardiac ultrasound (Fig. 1A,B). In agreement with the results of echocardiography data, the infarct sizes in the DEX group were remarkably smaller than those in MI/RI mice (Fig. 1C). In support, MI/RI resulted in a substantial increase in serum CK-MB levels and DEX pretreatment reversed it (Fig. 1D). Next, we examined whether DEX treatment affected the release of FGF21 from iBAT, the mRNA and protein expression levels of FGF21 in iBAT, as well as serum FGF21 levels were detected. As shown in Fig. 1E,F, the transcription and protein levels of FGF21 in iBAT of mice were elevated by DEX treatment and MI/RI surgery, while this increase was further in DEX-infused mice subjected to MI/RI. A mouse FGF21 ELISA assay further ascertained that DEX promoted the release of FGF21 following MI/RI (Fig. 1G). These results collectively indicated that DEX protected against MI/RI in mice, with a concomitant increase in iBAT-derived FGF21.

Figure 1
figure 1

DEX protects against myocardial ischemia/reperfusion injury and promotes iBAT-derived FGF21 release. (A) Representative echocardiographic M-mode images were obtained 24 h after myocardial ischemia/reperfusion (MI/R), n = 6/group. (B) Left ventricular ejection fraction (LVEF) and fractional shortening (LVFS), n = 6/group. (C) Representative images of TTC staining 24 h after MI/R. Myocardial infarction area was calculated as a percentage of the myocardial area at risk, n = 6/group. (D) The levels of serum CK-MB, n = 6/group. (E) Relative mRNA levels of FGF21 in iBAT. β-actin was used as an internal control. (F) Relative protein expression of FGF21 in iBAT were measured by western blot analysis. β-actin was used as an internal control. (G) The levels of serum FGF21 in Mice, n = 6/group. * means P < 0.05 vs Control, † means P < 0.05 vs I/R group, the value is expressed as Mean ± SD.

Full size image

Surgical iBAT ablation significantly inhibited the cardioprotective effect of DEX

To investigate the involvement of BAT in the protective effect of DEX on the heart, we performed surgical removal of iBAT in mice. Echocardiography results demonstrated that the beneficial effect of DEX on cardiac function was abolished by iBAT ablation, as evidenced by irregular wall movement (Fig. 2A) and reduced ejection fraction and shortening fraction (Fig. 2B). Consistent with the echocardiography findings, infarct sizes were significantly increased following iBAT ablation, in contrast to the reduced infarct sizes observed in the DEX + I/R group (Fig. 2C). Moreover, serum CK-MB levels were significantly elevated by iBAT ablation (Fig. 2D). Next, we sought to determine if BAT ablation resulted in the antagonistic effect of DEX on cell apoptosis of myocardium. The antiapoptotic effect of DEX was blocked by BAT ablation, as manifested by an increase in cleaved caspase3, Bax protein levels and a reduction in Bcl-2 protein levels in the I/R + DEX + BAT ablation group, as assessed in western blot experiments (Fig. 2E). Similarly, as shown in Fig. 2F, these imaging results of TUNEL staining further revealed that iBAT ablation markedly increased the percentage of cardiomyocytes apoptosis. Moreover, pretreatment with DEX significantly reduced the morphological changes of myocardium induced by I/R, which was disrupted following BAT excisions (Fig. 2G), thereby further supporting the conclusion that surgical iBAT ablation significantly inhibited the cardioprotective effect of DEX. Strikingly, the increase in serum FGF21 levels was significantly compromised in mice when iBAT was surgically removed (Fig. 2H). Collectively, we demonstrated that the release of FGF21 and the cardioprotective effect mediated by DEX were significantly inhibited following surgical iBAT ablation.

Figure 2
figure 2

Suidical iBAT ablation significantly inhibited the cardioprotective effect of DEX. (A) Representative echocardiographic M-mode images were obtained 24 h after myocardial ischemia/reperfusion (MI/R), n = 6/group. (B) Left ventricular ejection fraction (LVEF) and fractional shortening (LVFS), n = 6/group. (C) Representative images of TTC staining 24 h after MI/R. Myocardial infarction area was calculated as a percentage of the myocardial area at risk, n = 6/group. (D) The levels of serum CK-MB, n = 6/group. (E) The protein expression of Bax, Bcl-2, cleaved caspase 3, were determined by western blot analysis. Bar graphs present the quantification analysis of Bax, Bcl-2, cleaved caspase 3 expression, n = 3/group. (F) Representative photographs of cardiomyocyte apoptosis examined by TUNEL assay. Bar graphs present the quantification analysis of cell apoptosis, n = 6/group. Scale bar = 200 μm. (G) Morphological changes in myocardium was determined by HE staining. Scale bar = 100 μm. (H) The levels of serum FGF21 in Mice, n = 6/group.* means P < 0.05 vs Control, † means P < 0.05 vs I/R group, ‡ means P < 0.05 vs. I/R + DEX group, the value is expressed as Mean ± SD.

Full size image

Exogenous administration of FGF21 neutralizing antibody significantly inhibited the cardioprotective effect of DEX

It is well known that FGF21 plays an important role during MI/RI20,21,22,23. In our study, we confirmed that DEX promoted the release of FGF21 from iBAT after MI/RI, suggesting that FGF21 might act as a mediator for the cardioprotective effect of DEX. To investigate this possibility, we administered exogenous FGF21 neutralizing antibodies (FGF21-Ab) to mice via intraperitoneal injection.

As shown in Fig. 3A,B, echocardiography data indicated that the protective effect of DEX on cardiac function was eliminated by exogenous administration of FGF21-Ab, as manifested by irregular wall movement and decreased ejection fraction and shortening fraction. In agreement with these results, we observed a significant increase in the infarct sizes following exogenous administration of FGF21-Ab (Fig. 3C). As expected, FGF21-Ab administration significantly increased the serum CK-MB levels in mice (Fig. 3D). Furthermore, we examined the apoptosis-related proteins levels in left ventricular cardiac tissues by western blot experiments. We found that both cleaved-caspase3 and Bax protein levels in left ventricular cardiac tissues harvested from DEX + I/R + FGF21-Ab group were increased compared to DEX + I/R group (Fig. 3E). In contrast, the Bcl-2 protein levels significantly decreased in left ventricular cardiac tissues harvested from mice following the FGF21-Ab administration (Fig. 3E). In support, the imaging results of TUNEL staining further indicated that in contrast to DEX + I/R group, cardiomyocytes apoptosis was markedly increased by application of FGF21-Ab (Fig. 3F). HE staining showed that DEX could significantly reduce the morphological changes of myocardium after I/R surgery, but the beneficial effect was blocked by exogenous administration of FGF21-Ab (Fig. 3G). When taken together, our findings demonstrate that the FGF21-Ab administration significantly abolished cardioprotective effect of DEX, suggesting that FGF21 directly contributes to the cardioprotective effect of DEX.

Figure 3
figure 3

Exogenous administration of FGF21 neutralizing antibody significantly inhibited the cardioprotective effect of DEX. (A) Representative echocardiographic M-mode images were obtained 24 h after myocardial ischemia/reperfusion (MI/R), n = 6/group. (B) Left ventricular ejection fraction (LVEF) and fractional shortening (LVFS), n = 6/group. (C) Representative images of TTC staining 24 h after MI/R. Myocardial infarction area was calculated as a percentage of the myocardial area at risk, n = 6/group. (D) The levels of serum CK-MB, n = 6/group. (E) The protein expression of Bax, Bcl-2, cleaved caspase 3, were determined by western blot analysis. Bar graphs present the quantification analysis of Bax, Bcl-2, cleaved caspase 3, n = 3/group. (F) Representative photographs of cardiomyocyte apoptosis examined by TUNEL assay. Bar graphs present the quantification analysis of cell apoptosis, n = 6/group. scale bar = 200 μm. (G) Morphological changes in myocardium was determined by HE staining. scale bar = 100 μm.* means P < 0.05 vs Control,† means P < 0.05 vs I/R group,‡ means P < 0.05 vs. I/R + DEX group, the value is expressed as Mean ± SD.

Full size image

Blockade of AMPK/PGC1α signaling pathway attenuates DEX-Mediated FGF21 expression

Next, to further explore whether the AMPK/PGC1α signaling pathway was involved in the DEX-mediated FGF21 expression in iBAT, we measured the protein levels of p-AMPK, AMPK and PGC1α in iBAT harvested from mice after I/R surgery or DEX pretreatment. Pretreatment with DEX before I/R surgery significantly activated the AMPK/PGC1α signaling pathway, as manifested by increased p-AMPK, AMPK and PGC1α protein levels (Fig. 4A). Furthermore, we measured FGF21 protein levels in iBAT harvested from mice post administration of a commonly used AMPK Compound C or PGC1α inhibitor SR-18292, as assessed by western blot experiments. As shown in Fig. 4B, when the AMPK/PGC1α pathway was disrupted by application of Compound C or SR-18292, the increase in FGF21 protein levels induced by DEX pretreatment was significantly compromised in iBAT. Similarly, we observed a significant decrease in serum FGF21 levels in mice treated with Compound C or SR-18292, in contrast to the increased serum FGF21 levels observed in the DEX + I/R group (Fig. 4C). Then, we treated brown adipocytes with different concentrations of DEX. In vitro, we observed a dose-dependent increase in the protein levels of p-AMPK, AMPK, and PGC1α following treatment with different concentrations of DEX (Fig. 4D). The docking of DEX with AMPK at the active site was demonstrated in Fig. 4E, indicating the formation of a stable complex with good binding affinity values. Interestingly, DEX interacted with the 148th aspartic acid of AMPK, resulting in high binding energy values of − 7.0 kcal/mol (Fig. 4E), suggesting that AMPK could be a potential target of DEX for cardioprotection. The transcriptional activation of UCP1, PPARγ, PRDM16, and CIDEA plays an indispensable role in the thermogenesis of brown adipose tissue35,36,37. DEX dose-dependently increased the transcription levels of thermogenic genes UCP1, PPARγ, CIDEA and PRDM16 (Fig. 4F), whereas the application of Compound C or SR-18292 did abolish the increased transcription levels of thermogenic genes resulting from DEX (Fig. 4G). Moreover, we observed a significant reduction in FGF21 mRNA levels in brown adipocytes when the AMPK/PGC1α pathway was blocked (Fig. 4H). Consistent with the results of FGF21 mRNA levels, we observed that the FGF21 protein levels in brown adipocytes were also reduced followed the application of Compound C or SR-18292 (Fig. 4I). These results confirmed that blockade of AMPK/PGC1α signaling pathway attenuated DEX-mediated FGF21 expression.

Figure 4
figure 4

Blockade of AMPK/PGC1a signaling pathway attenuates DEX-mediated FGF21 expression. (A)The protein expression of p-AMPK/AMPK, PGC1α in iBAT, were determined by western blot analysis. Bar graphs present the quantification analysis of p-AMPK/AMPK, PGC1α, n = 3/group. (B) The protein expression of FGF21 in iBAT, were determined by western blot analysis. Bar graphs present the quantification analysis of FGF21, n = 3/group. (C) The levels of serum FGF21 in Mice, n = 6/group. (D) DEX dose-dependently induced the protein expression of p-AMPK/AMPK, PGC1α. Bar graphs present the quantification analysis of p-AMPK/AMPK , PGC1α, n = 3/group. (E) Combination pattern diagram of AMPK and DEX. (F) Relative mRNA levels of thermogenic genes at different concentrations of DEX. (G) Relative mRNA levels of thermogenic genes in brown adipose cells after Compound C/SR-18292 treatment. (H) Relative mRNA levels of FGF21 in brown adipose cells after Compound C/SR-18292 treatment. (I) The protein expression of FGF21 in brown adipose cells after Compound C/SR-18292 treatment. Bar graphs present the quantification analysis of FGF21, n = 3/group.* means P < 0.05 vs Control,† means P < 0.05 vs DEX group, the value is expressed as Mean ± SD.

Full size image

DEX had minimal effect on the H/R-induced injury in HL-1 cells

On the contrary, despite its clear protective effect against I/R injury in vivo, we regretfully discovered that DEX had minimal effect on isolated cardiomyocytes exposed to H/R conditions. Regardless of the concentration of DEX used for pretreatment, it did not rescue cardiomyocytes death induced by H/R, as confirmed by flow cytometry assay and PI staining (Fig. 5A,B,E). Moreover, the application of DEX (0.1 µM), DEX (1 µM), or DEX (10 µM) did not alter cell viability in HL-1 cells exposed to H/R, as evaluated using a CCK8 assay (Fig. 5C). Consistent with the viability experiments, DEX did not reduce LDH release following H/R at different concentrations (Fig. 5D). In addition to this, the increase in cleaved caspase3, Bax protein levels and the reduction in Bcl-2 protein levels resulting from H/R condition were not altered by DEX in different concentrations, as assessed by western blot experiments (Fig. 5F). Collectively, these in vitro findings indicate that DEX has minimal effect on H/R-induced injury in HL-1 cells. Therefore, it is reasonable to conclude that the protective effects of DEX on the mouse heart against I/R injury are not due to a direct effect on cardiomyocytes.

Figure 5
figure 5

DEX had minimal effect on the H/R-induced injury in HL-1 cells. (A) Representative photographs of flow cytometry apoptosis assay. (B) Statistics of Cell apoptosis rate. (C) CCK-8 cell viability assay. (D) lactate dehydrogenase (LDH) release assay.(E) Representative photographs of PI staining. Scale bar = 100 μm. (F) The protein expression of Bax, Bcl-2, cleaved caspase 3, were determined by western blot analysis. Bar graphs present the quantification analysis of Bax, Bcl-2, cleaved caspase 3, n = 3/group.* means P < 0.05 vs Control, the value is expressed as Mean ± SD.

Full size image

FGF21 release from DEX-treated brown adipocytes attenuated H/R-induced injury in HL-1 cells through regulating the Keap1/Nrf2 complex.

To confirm that the brown adipocytes was necessary for DEX to exert its cardioprotection effect in vitro, brown adipocytes were cultured for 2 h in the presence or absence of DEX, before removing the HL-1 cardiomyocytes and subjecting them to H/R condition. As shown in Fig. 6A,B, co-culture of cardiomyocytes with control medium from untreated brown adipocytes (CM-Veh) was not protective, but when brown adipocytes were pre-treated for 2 h with Dex (1 μM) prior to co-culture (CM-DEX), the cell apoptosis of cardiomyocytes was reduced markedly. In contrast to H/R group, co-culture of cardiomyocytes with CM-DEX enhanced cell viability in cardiomyocytes exposed to H/R condition (Fig. 6C). Similar to the cell viability results, LDH release was significantly reduced in the H/R + CM-DEX group (Fig. 6D). Furthermore, co-culturing cardiomyocytes with CM-DEX led to a decrease in cleaved caspase 3 and Bax protein levels, and an increase in Bcl-2 protein levels (Fig. 6E).

Figure 6
figure 6

FGF21 release from DEX-treated brown adipocytes attenuated H/R-induced injury in HL-1 cells through regulating the Keap1/Nrf2 complex. (A) Representative photographs of flow cytometry apoptosis assay. (B) Statistics of Cell apoptosis rate. (C) CCK-8 cell viability assay. (D) Lactate dehydrogenase (LDH) release assay. (E) The protein expression of Bax, Bcl-2, cleaved caspase 3, were determined by western blot analysis. Bar graphs present the quantification analysis of Bax, Bcl-2, cleaved caspase 3, n = 3/group. (F) Representative photographs of DHE staining. Scale bar = 100 μm. (G) The protein expression of Keap1 were determined by western blot. (H) The interaction of Keap1 with Nrf2 was determined by Co-IP. (I) The protein expression of Nrf2 in nucleus and cytoplasm were determined by western blot. (J) Relative mRNA levels of antioxidant genes in cardiomyocytes after CM-DEX treatment.

Full size image

Reactive oxygen species (ROS) are generated massively in cardiomyocytes during reperfusion process38. It has been demonstrated that FGF21 protects the heart from oxidative stress39,40. As such, co-culture of cardiomyocytes with CM-DEX could potentially inhibit the elevated ROS levels generated during reperfusion. To this end, we used a ROS-detecting probe, DHE, to measure the ROS levels by fluorescence microscope. As shown in Fig. 6F, application of CM-DEX significantly restricted ROS elevation in cardiomyocytes exposed to H/R condition. Next, we sought to further explore whether the Keap1/Nrf2 pathway contributes to the inhibited ROS elevation following H/R during co-culture of cardiomyocytes with CM-DEX. We observed that in contrast to H/R group, Keap1 in cardiomyocytes treated with CM-DEX was significantly degraded, which was conducive to the release of Nrf2 (Fig. 6G). In agreement with the results of Keap1 protein levels, we observed that the dissociation of Keap1/Nrf2 complex was aggravated by co-culture of cardiomyocytes with CM-DEX (Fig. 6H). In addition to this, co-culture of cardiomyocytes with CM-DEX resulted in an obvious nuclear translocation of Nrf2, as indicated by a significant increase in Nrf2 protein expression in nuclear proteins but a significant decrease in cytoplasmic proteins (Fig. 6I). Subsequently, Nrf2 upregulated the transcription of several antioxidant genes such as Heme Oxygenase-1 (HO-1), NAD(P)H quinone oxidoreductase (NQO1), glutamate-cysteine ligase modifier (GCLM), and glutamate-cysteine ligase catalytic subunit (GCLC) (Fig. 6J). Collectively, this set of results demonstrated that DEX triggers cardioprotection by modulating the Keap1/Nrf2 signaling pathway, but this pharmacological effect of DEX necessitates its interaction with brown adipocytes.

Discussion

As a highly selective α2 adrenergic receptor (α2AR) agonist, DEX has the potential to effectively improve myocardial ischemia–reperfusion injury (MI/RI) by establishing communication between brown adipose tissue (BAT) and the heart. Preliminary results have shown that DEX pretreatment not only significantly improves cardiac dysfunction associated with myocardial ischemia/reperfusion (MI/R), but also increases mRNA transcription and protein expression levels of FGF21 in the interscapular brown adipose tissue (iBAT) as well as serum FGF21 levels. Building upon the latest research progress both domestically and internationally, along with our preliminary experimental findings, this study aims to investigate whether FGF21 produced by BAT mediates the protective effects of DEX-induced cardiac preconditioning. Mouse models of myocardial ischemia/reperfusion and HL-1 cells oxygen glucose deprivation/ re-oxygenation will be utilized to explore the molecular mechanisms involved.

It is well established that DEX preconditioning represents a protective effect against MI/RI in vivo. Autophagy is central in DEX-mediated cardioprotection in sepsis. DEX prevents septic myocardial dysfunction via activating α7nAChR and PI3K/Akt-mediated autophagy41. Besides, other studies have revealed that DEX preconditioning protects the hearts against I/R injury via alleviating myocardial inflammation and apoptosis42, downregulation of the endoplasmic reticulum stress signaling pathway43, or exerting antioxidant stress through the activation of the Keap1/Nrf2/ARE signal pathway44. In particular, DEX pretreatment prevents maladaptive remodeling from myocardial infarction34.

BAT is densely innervated by the sympathetic nervous system (SNS), and adrenergic stimulation can activate the thermogenic function of BAT and reduce lipids and glucose in the circulatory system45. Accordingly, as a highly selective α2 adrenergic receptor (α2AR) agonist, DEX may protect against MI/RI by establishing organ crosstalk between BAT and the heart. To test this hypothesis and ascertain the essential role of BAT in DEX preconditioning, we performed surgical resection of iBAT. Our results clearly demonstrated that surgical iBAT ablation significantly inhibited the release of FGF21 and the cardioprotective effect mediated by DEX. Some studies have suggested that FGF21 possesses antihypertrophic and cardioprotective properties in models of ischemia, hypertension, and hypertrophy21,46,47. If this holds true, we anticipate that FGF21 acts as a mediator for the cardioprotective effect of DEX. To investigate this hypothesis, mice were administered FGF21-Ab to block the action of FGF21. As expected, the administration of FGF21-Ab significantly abolished the cardioprotective effect of DEX. It is documented that adult humans possess significantly less BAT compared to mice. In mice, BAT is more prominent and plays a crucial role in thermogenesis and metabolic regulation, which can influence the response to various treatments, including Dex. This difference necessitates careful consideration when extrapolating findings from mice to humans. In mice, the protective effects of Dex against I/R damage might involve mechanisms mediated by BAT, such as enhanced thermogenic activity, improved mitochondrial function, and reduced oxidative stress. However, the smaller amount of BAT in adult humans suggests that the protective clinical effects of Dex in humans due to BAT need more evidence, which will be explored in further studies. Comparative studies examining the role of BAT in different species can also provide insights into the potential contributions of BAT-independent mechanisms in Dex-mediated protection against I/R injury.

Distinct from previously described results48,49,50,51, we observed that DEX had minimal effect on the HL-1 cells exposed to H/R condition. This may be attributed to variations in the timing and dosage of DEX pretreatment. Coincidently, a recent study also confirmed that DEX did not affect the cell viability of primary cardiomyocytes exposed to H/R conditions31. In this study, they found that DEX triggered cardiac protection by activating the eNOS/NO signaling pathway in the endothelium31, indicating that the interaction of the hearts with the endothelium is required for DEX-mediated cardioprotection.

The AMPK/PGC1α pathway is a vitally involved in adipogenesis, mitochondrial homeostasis and browning in WAT52,53,54. It was observed that DEX significantly promotes the expression of p-AMPK/AMPK and PGC1α in iBAT harvested from mice post I/R surgery. Blocking the AMPK/PGC1α pathway led to a decrease in the transcription and expression levels of FGF21 in iBAT, as well as the serum levels of FGF21 in mice. To assess the effect of DEX on the AMPK/PGC1α pathway in vitro, brown adipocytes were treated with different concentrations of DEX. It was found that DEX dose-dependently increased the protein expression of p-AMPK/AMPK, PGC1α, and the transcription levels of thermogenic genes UCP1, PPARγ, PRDM16, and CIDEA. Consistent with our in vivo results, blocking the AMPK/PGC1α pathway significantly reduced the transcription and protein expression levels of FGF21 induced by DEX. These results indicate that AMPK/PGC1α signaling-induced FGF21 in brown adipocytes is necessary for the cardioprotective effects of DEX.

Nrf2 plays a central role in oxidative stress and is involved in the cardioprotective effects55. In response to oxidative stress, Keap1 is oxidized at reactive cysteine residues, leading to the inactivation of Keap1 and the stabilization and translocation of Nrf2 into the nucleus56. Consistent with these findings, our results demonstrated that co-culturing cardiomyocytes with CM-DEX significantly promoted the dissociation of the Keap1/Nrf2 complex and the nuclear translocation of Nrf2 in HL-1 cells exposed to the H/R condition. Activation of Nrf2 also led to a significant increase in the transcription of antioxidant genes HO-1, NQO1, GCLM, and GCLC. These experiments clearly indicate that the release of FGF21 from DEX-treated brown adipocytes triggers cardioprotection by modulating the Keap1/Nrf2 signaling pathway. However, there are some limitations in this study. Firstly, due to the widespread distribution of brown adipose tissue (BAT) in mice, completely eliminating the effects of BAT through surgical iBAT ablation is not practical. Additionally, it would be beneficial to collect more serum samples from patients to assess the changes in FGF21 serum levels at different time points following DEX intervention.

Conclusions

In conclusion, our results suggest that DEX activates the AMPK/PGC1α pathway which promotes the secretion of FGF21 from iBAT. As a bridge between BAT and cardiac tissue, FGF21 triggers cardiac protection via modulating Keap1/Nrf2 antioxidant signaling pathway (Fig. 7). Preconditioning DEX thus represents a promising strategy to greatly protect against MI/RI in the clinical settings. Of note, Mimuro et al. reported that Dex deteriorated I/R damage57. This finding contrasts with our results and those of other studies showing protective effects of Dex in similar contexts. This discrepancy highlights the complexity of Dex's effects and suggests that its impact on I/R damage may be context-dependent. In the study of Mimuro et al.57 who reported that Dex deteriorated I/R damage since DEX was administered for 25 min before the initiation of reperfusion in isolated rat hearts. In our study, DEX was administered 30 min before the I/R surgery in vivo. Several aspects can account for this discrepancy, such as differences in Dex dosage and the timing of DEX administration. Mimuro et al. used a different dosage or timing regimen that led to detrimental effects, whereas our study and others used protocols that resulted in protective effects. These results underscore the need for a nuanced understanding of Dex's effects on I/R injury. These findings highlight the importance of optimizing Dex treatment parameters for different clinical scenarios. Future studies should aim to delineate the conditions under which Dex is beneficial versus detrimental. Investigating the dose–response relationship, timing of administration, and specific pathways involved could provide insights into optimizing Dex therapy for I/R injury. One should bear in mind that both Compound C and FGF21 have shown various effects on metabolic pathways in cardiomyocytes. Future studies could explore this interaction in more detail, potentially looking at how Compound C modulates FGF21 expression and activity in cardiomyocytes, both in the presence and absence of Dex.

Figure 7
figure 7

Dexmedetomide can promote the production of FGF21 derived from BAT by activating AMPK/PGC1αpathway, and FGF21 can reduce myocardial ischemia/reperfusion injury by activating Keap1/Nrf2 antioxidant pathway. Solid arrows depict activation, and transverse “T” shape indicates inhibition.

Full size image