Excessive accumulation of epicardial adipose tissue promotes microvascular obstruction formation after myocardial ischemia/reperfusion through modulating macrophages polarization

Abstract

Background: Owing to its unique location and multifaceted metabolic functions, epicardial adipose tissue (EAT) is gradually emerging as a new metabolic target for coronary artery disease risk stratification. Microvascular obstruction (MVO) has been recognized as an independent risk factor for unfavorable prognosis in acute myocardial infarction patients. However, the concrete role of EAT in the pathogenesis of MVO formation in individuals with ST-segment elevation myocardial infarction (STEMI) remains unclear. The objective of the study is to evaluate the correlation between EAT accumulation and MVO formation measured by cardiac magnetic resonance (CMR) in STEMI patients and clarify the underlying mechanisms involved in this relationship.

Methods: Firstly, we utilized CMR technique to explore the association of EAT distribution and quantity with MVO formation in patients with STEMI. Then we utilized a mouse model with EAT depletion to explore how EAT affected MVO formation under the circumstances of myocardial ischemia/reperfusion (I/R) injury. We further investigated the immunomodulatory effect of EAT on macrophages through co-culture experiments. Finally, we searched for new therapeutic strategies targeting EAT to prevent MVO formation.

Results: The increase of left atrioventricular EAT mass index was independently associated with MVO formation. We also found that increased circulating levels of DPP4 and high DPP4 activity seemed to be associated with EAT increase. EAT accumulation acted as a pro-inflammatory mediator boosting the transition of macrophages towards inflammatory phenotype in myocardial I/R injury through secreting inflammatory EVs. Furthermore, our study declared the potential therapeutic effects of GLP-1 receptor agonist and GLP-1/GLP-2 receptor dual agonist for MVO prevention were at least partially ascribed to its impact on EAT modulation.

Conclusions: Our work for the first time demonstrated that excessive accumulation of EAT promoted MVO formation by promoting the polarization state of cardiac macrophages towards an inflammatory phenotype. Furthermore, this study identified a very promising therapeutic strategy, GLP-1/GLP-2 receptor dual agonist, targeting EAT for MVO prevention following myocardial I/R injury.

Keywords: Epicardial adipose tissue; GLP-1/GLP-2 receptor dual agonist; Inflammation; Liraglutide; Macrophage; Microvascular obstruction.

Fig. 1

Representative images of cardiac magnetic resonance of STEMI patients. A Measurements of EAT thickness in the grooved segments at three different locations in the horizontal long-axis plane (left AV groove, right AV groove and anterior IV groove). B Measurements of EAT thickness in the grooved segments at three locations in the short-axis plane (superior IV groove, inferior IV groove and the right ventricular free wall. C Short axis LGE sequence performed at mid left ventricle level showing the presence of MVO

Fig. 2

High EAT load was associated with MVO occurrence in STEMI patients. A Range distribution of MVO occurrence according to EAT and LV EAT volume. Bars represent the percentage of MVO occurrence identified at every quartile. B The association of EAT volume with infarct size, peak TnT and LVEF. C Univariate logistic regression analysis to identify the risk factors associated with the occurrence of MVO in patients with STEMI. D Multivariate logistic regression analysis to identify the risk factors associated with the occurrence of MVO in patients with STEMI. E Predictive accuracy of MVO based on ROC curve analysis. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 3

Association between EAT accumulation and DPP4, GLP-1 and GLP-2. A Correlation between LV EAT mass index and plasma DPP4 concentration in STEMI patients. B Correlation between LV EAT mass index and plasma DPP4 activity in STEMI patients. C Circulating level of total GLP-1 in STEMI patients. D Circulating level of total GLP-2 in STEMI patients. Data are presented as mean ± SD. ***P < 0.001, ****P < 0.0001

Fig. 4

EAT depletion ameliorated microvascular dysfunction and cardiac inflammation in mice. A The experimental protocol for assessing the impact of excessive EAT accumulation on myocardial I/R injured mice. B Quantitative assessment of EF% and FS% using echocardiography 3 days following operation (n = 6). C Representative images of heart slices stained with Evans Blue and TTC from mice 3 days after myocardial I/R induction. The red line represents the area at risk (AAR) and the white dotted line represents the size of the infarct (IS). Scale bar = 5 mm. D Quantitative analysis of the percentage of AAR and percentage of infarct size in hearts in C (n = 5). E Representative thioflavin-S stained hearts isolated from myocardial I/R injured mice 1 day post operation with or without EAT. White dotted line represents the left ventricular area (LV) and red dotted line represents the size of MVO. Scale bar = 5 mm. F quantification of MVO percentage in hearts in E (n = 5). G Cytokine expression of IL-10 in the hearts of mice from different groups 3 days post I/R (n = 7). H Cytokine expression of IL-6 and IL-1β in the hearts of mice from different groups 3 days post I/R (n = 7). I Representative HE staining of I/R injured hearts from EAT-depletion and EAT preservation mice 3 days following operation. Scale bar = 100 μm. J Quantification of inflammatory cell infiltration (%) within the ischemic hearts in I (n = 5). K Immunohistochemical staining of CD68⁺ cells within the ischemic zone 3 days following I/R. Scale bar = 100 μm. L Quantification of CD68⁺ cells from K (n = 5). Data are presented as mean ± SD *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns = not significant

Fig. 5

Excessive EAT accumulation promoted macrophage M1 polarization under myocardial I/R condition through secreting inflammatory EVs. A Representative flow cytometry plots depicting the ratio of total macrophage (CD11b⁺F4/80⁺), M1 phenotype (CD11b⁺F4/80⁺iNOS⁺CD206⁻) and M2 phenotype (CD11b⁺F4/80⁺iNOS⁻CD206⁺) in cardiac tissue of mice 3 days post-operation. B Quantification of flow cytometry data from A (n = 4). C Immunofluorescence staining of iNOS⁺ and Arg1⁺ cells within the ischemic zone respectively in mice 3 days post operation. D Quantification of iNOS⁺ and Arg1⁺ cells from groups defined in C (n = 4). E Experimental scheme of EAT and macrophages co-culture. F Gene expression patterns of M1 markers (iNOS, IL-6, TNFα and IL-1β) and M2 markers (Arg1, CD206, IL-10 and TGFβ) were analyzed in LPS-stimulated macrophages after co-culturing with EAT or not for 48 h (n = 3). G Representative flow cytometry plots indicating the percentages of M1 (iNOS⁺CD206⁻) and M2 (iNOS⁻CD206⁺) phenotype in macrophages. H Quantification of flow cytometry data in G (n = 3). I Experimental scheme of macrophages co-culturing with EAT-conditioned medium with or without EVs. J Representative flow cytometry plots indicating the percentages of M1 (iNOS⁺CD206⁻) and M2 (iNOS⁻CD206⁺) phenotype in different groups described above. K Quantification of flow cytometry data in J (n = 3). Data are presented as mean ± SD *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns = not significant

Fig. 6

GLP-1 receptor agonist and GLP-1/GLP-2 receptor dual agonist attenuated myocardial I/R injury and alleviated cardiac inflammation through modulating EAT inflammatory and metabolic status. A Quantitative assessment of EF% and FS% in mice treated with vehicle, liraglutide and dapiglutide 3 days post-operation (n = 6). B Representative images of heart slices stained with Evans Blue&TTC and quantification of the percentage AAR and percentage infarct 3 days following I/R (n = 5). Scale bar = 5 mm. C Representative thioflavin-S stained heart slices and quantification of MVO percentage in vehicle, liraglutide and dapiglutide treated mice 1 day post operation (n = 5). Scale bar = 5 mm. D Representative images of H.E.stained hearts and quantification of inflammatory cell infiltration (%) 3 days after surgery (n = 5). Scale bar = 100 μm. E Principal components analysis of transcriptional profiling of EAT from vehicle, liraglutide and dapiglutide treated mice 3 days post I/R (n = 3). F Bar plots displaying the top significant KEGG pathways associated with DEGs in dapiglutide versus liraglutide, dapiglutide versus vehicle, and liraglutide versus vehicle groups. G Heatmap displaying expression patterns of top inflammation related DEGs between dapiglutide and vehicle group. H Heatmap displaying expression patterns of top inflammation related DEGs between liraglutide and vehicle group. I A heatmap depicting the expression patterns of top inflammation related DEGs among three groups. J A heatmap depicting the expression patterns of top metabolism related DEGs among three groups. K EF% and FS% of vehicle, liraglutide and dapiglutide treated I/R injured mice with EAT depletion 3 days following surgery (n = 6). L Quantitative assessment of LVIDd and LVIDs measured by echocardiography of vehicle, liraglutide and dapiglutide treated I/R injured mice with EAT depletion 3 days following I/R (n = 6). M Cardiac IL-6 and IL-1β levels of vehicle, liraglutide and dapiglutide treated I/R injured mice with EAT depletion 3 days after operation (n = 6). Data are presented as mean ± SD *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns = not significant