Effects of plasma exosomes on endothelial cells in patients with coronary artery disease
First, we separated exosomes from patients with and without coronary artery disease by differential centrifugation. We observed spherical particles with diameters of 30–150 nm by transmission electron microscopy (TEM) (Fig. 1A-B), which is consistent with the size and morphology of exosomes. For the marker proteins CD9 and CD81 of exosomes, we examined them by nanoflow cytometry (nFCM) (Fig. 1C-D). In addition, we determined the size of exosomes by nanoparticle tracking analysis (NTA) technique (Fig. 1E-F). Interestingly, in electron microscopy of exosome samples from different diseases, we found that more exosomes were released in patients with coronary artery disease compared to those with non-coronary artery disease (Fig. 1A-B). To explore the effect of these exosomes on endothelial cells, we labeled exosomes from patients with and without coronary artery disease with PKH67 and cultured them into human umbilical vein endothelial cells (HUVECs). After 24 hours of continuous labeling, we observed with confocal microscopy that the exosomes had been effectively internalized by HUVECs. But there was no significant change in the degree of internalization of endothelial cells by exosomes in patients with and without coronary artery disease (Fig. 1G-I). It is well known that endothelial cells are the first cells to respond to atherosclerosis. Therefore, to investigate the effect of exosomes on endothelial cells from patients with coronary artery disease, we subsequently tested the proliferative capacity and cell viability of endothelial cells (Fig. 1J-M). The results showed that plasma exosomes from patients with coronary artery disease attenuated the viability and proliferative capacity of endothelial cells.
circ_0001785 Protection in the population
CircRNA is difficult to be hydrolyzed by RNase R due to its unique properties of absence of 5′ and 3′ termini and poly A tail. To determine the structural features of circ_0001785, we first examined its tolerance to RNase R digestion. We treated total RNA extracted from endothelial cells with Rnase R and tested the effect of RNase R treatment with its relative expression level to the host gene ELP3 linear RNA. The results showed that circ_0001785 was significantly tolerant to RNase R, further demonstrating the circular structure of circ_0001785 (Fig. 2A). To verify the reverse splice site of circ_0001785, we performed Sanger sequencing of the cDNA amplified by qRT-PCR for this circRNA Divergent primer. From the sequencing results, we could see that the circRNA contained the back-to-back splice site of the exon of ELP3 gene (Fig. 2B). In addition, we designed Convergent-specific primers for circ_0001785 and GAPDH, and Divergent-specific primers for circ_0001785 and GAPDH to amplify cDNA and gDNA of total RNA reversal of HUVECs, respectively.z Finally, it was found that circ_0001785 in cDNA could be detected in the cDNA, but not in the gDNA of HUVECs (Fig. 2C). We then investigated the effect of Exosome-circ_0001785 on atherosclerosis in patients with coronary artery disease. We first determined the cellular origin of circ_0001785. RNA was extracted from human cardiomyocytes, endothelial cells and monocytes, and then qRT-PCR experiments were performed. We found that circ_0001785 had stable expression in monocytes, and its expression was significantly higher than the other two groups. Therefore, we determined that circ_0001785 was mainly derived from monocytes (i.e., mainly from leukocytes) (Fig. 2E). Next, we extracted leukocytes from patients with coronary artery disease and detected the expression of circ_0001785 by qRT-PCR, and found that the expression level of circ_0001785 in leukocytes from patients with coronary artery disease was significantly lower than that in healthy control patients (Fig. 2F). In contrast, when we examined plaque tissue from patients with lower extremity atherosclerotic plaques, we found significantly more circ_0001785 in plaque tissue than in other tissues next to the plaque (Fig. 2G). Therefore, we suggest that circ_0001785 may be transferred to the site of injury via exosomes during plaque formation or endothelial cell injury. To further determine the relationship between circ_0001785 and patients with clinical coronary artery disease, we divided 20 patients into two groups according to the degree of coronary stenosis (grade III for 50%-75% and grade IV for 75%-100%), and the results showed that patients with severe coronary stenosis had significantly lower circ_0001785 levels than patients with moderate coronary stenosis levels (Fig. 2H).
circ_0001785 induces proliferation of endothelial cells and inhibits apoptosis and migration of endothelial cells
First, we investigated the expression of circ_0001785 after co-culture of exosomes with endothelial cells and found that the expression of circ_0001785 was significantly lower in the CHD patient group (Fig. 3A). Therefore, we suggest that exosome-derived circ_0001785 may play an important role in coronary heart disease. To better mimic the environment in which human coronary heart disease develops, a new model of atherosclerotic endothelial cell injury was established in this study. We co-stimulated endothelial cells with ox-LDL combined with LPS, and confirmed the successful model establishment with functional experiments on endothelial cells, and again analyzed the biological functions of the model. After the model was successfully established, we transfected the overexpressed circ_0001785 lentivirus into the model cells and firstly examined the viability of endothelial cells by CCK-8 assay, and found that the overexpressed circ_0001785 could enhance the viability of endothelial cells (Fig. 3B). Then we examined the degree of endothelial cell proliferation by EdU assay and found that overexpression of circ_0001785 significantly promoted the proliferation ability of endothelial cells (Fig. 3C). Meanwhile, we examined l apoptosis of endothelial cells by Hoechst + PI assay and migration ability of endothelial cells by cell scratch assay, and found that overexpression of circ_0001785 significantly inhibited apoptosis and cell migration (Fig. 3D-E). All these data indicated that the overexpression of circ_0001785 significantly inhibited endothelial cell injury (Fig. 3F).
circ_0001785 exerts its functional through the ceRNA network axis of miR-513a-5p/TGFBR3
To identify the downstream target genes of circ_0001785, we first performed bioinformatics analysis. In a previous study, we identified the targeting pathway of circ_0001785/miR-513a-5p/TGFBR3 (doi:10.3389/fcvm.2023.1070616). To validate the predicted targets, based on the possible binding sites between circ_0001785 and miR-513a-5p, we tested the binding ability of circ_0001785 to miR-513a-5p by a dual luciferase reporter gene assay in a base-complementary manner. We cotransformed miR-513a-5p mimics with a dual luciferase reporter gene plasmid containing circ_0001785 wild-type (Wt) and mutant (Mut) cDNA fragments into endothelial cells and found targeted mutations in the predicted binding site of circ_0001785 to miR-513a-5p. The results showed that overexpression of miR-513a-5p significantly reduced the Wt-type circ_0001785 luciferase activity, but not the Mut-type circ_0001785 luciferase activity (Fig. 4A-B). And the above results also indicated that circ_0001785 could directly bind to miR-513a-5p by base complementation. Similarly, we also verified the binding of miR-513a-5p and TGFBR3 (Fig. 4C-D). To further validate the expression of miR-513a-5p and TGFBR3, we extracted RNA from patients with coronary artery disease and healthy controls and performed qRT-PCR experiments for miR-513a-5p and TGFBR3, respectively, which showed increased expression of miR-513a-5p and significantly decreased expression of TGFBR3 in the blood of patients with coronary artery disease (Fig. 4E -F). Similarly, we examined plaque tissues from patients with lower extremity atherosclerotic plaques and found that the amount of miR-513a-5p was significantly less in plaque tissues than in other tissues next to the plaque, while the amount of TGFBR3 was more in plaque tissues than in other tissues next to the plaque (Fig. 4G-H).
circ_0001785 promotes endothelial cell proliferation and inhibits apoptosis and migration via miR-513a-5p/TGFBR3
Through the above experiments, we have demonstrated that circ_0001785 can inhibit the apoptosis and migration of endothelial cells. And circ_0001785 can indirectly affect the expression of TGFBR3 through miR-513a-5p. To further explore whether circ_0001785 affects endothelial cell proliferation, migration and apoptosis through miR-513a-5p/TGFBR3 pathway. In this study, we cotransfected overexpressed circ_0001785 with miR-513a-5p mimics and miR-513a-5p NC, respectively, and validated them using cell functional assays. The specific groups were control group, oe circ_NCz group, oe circ_0001785 group, oe circ_NC + miR mimic-NC group, oe circNC + miR-513a-5p mimic group, oe circ_0001785 + miR mimic-NC group, oe circ_0001785 + miR-513a-5p mimic group. The results showed that overexpression of miR-513a-5p somewhat attenuated the proliferation promoting effect caused by circ_0001785 overexpression (Fig. 5A-B), as well as the inhibitory effect on apoptosis (Fig. 5C), and endothelial cell migration (Fig. 5D). In contrast, the addition of miR-513a-5p NC plasmid had no significant effect on the function of endothelial cells. We also examined the expression level of TGFBR3 by qRT-PCR and found that the expression level of TGFBR3 was significantly increased in endothelial cells overexpressing circ_0001785, while the addition of miR-513a-5p mimics was able to antagonize this increase in expression, resulting in a decrease in TGFBR3 expression. In contrast, the addition of miR-513a-5p NC plasmid did not significantly change the expression of TGFBR3 in the endothelium (Fig. 5E). The above experiments suggest that circ_0001785 can indirectly regulate the expression of TGFBR3 by inhibiting miR-513a-5p, thus exerting its promotional effect on endothelial cell proliferation and its inhibitory effect on endothelial cell apoptosis and migration.
Construction of an endothelial injury model of atherosclerosis in mice
As with the cellular model, we established a mouse model with both atherosclerotic plaque and inflammatory infiltration. Briefly, we fed mice with a high-fat diet and then performed a uniform left anterior descending branch ligation. However, the aim was not to infarct, but to cause an inflammatory infiltrate around the ligation line at the time of ligation, thus further approximating the nature of human vulnerable plaques. Interestingly, we found that after ligation of the left anterior descending branch in mice, inflammation around the ligature line could spread directly to the aortic sinus, leading to inflammatory infiltration at the aortic sinus plaque, which facilitated our further observation. The specific groups were: control, AMI, ApoE−/−+AMI, ApoE−/−+AMI + PBS, ApoE−/−+AMI + LV-oe circ_0001785, ApoE−/−+AMI + LV-oe circNC. where ApoE−/− mice were given high-fat feeding for 10 weeks before left anterior descending branch ligation (Fig. 6A). ApoE−/−+AMI mice were more prone to endothelial cell detachment and inflammatory infiltration with plaque enlargement compared with the control and AMI groups (Fig. 6B, D, E). Also, mice in the model group had increased atherosclerosis and were prone to increased collagen content and thinning of the fibrous cap (Fig. 6C,F). To observe the consistency of the mouse model with human plaque tissue, we removed plaque tissue from patients with lower extremity atherosclerosis from the clinic and then performed HE and Masson staining, and we found that both inflammatory infiltration and collagen fibers were significantly increased in plaque tissue from patients with lower extremity atherosclerosis, consistent with the lesion characteristics of atherosclerotic vulnerable plaques in the mouse model (Fig. 6G-H).
Overexpression of circ_0001785 leads to reduced endothelial injury and enhanced left ventricular diastolic function in mice
In the present experiment, we induced AMI in mice with ligation of the left anterior descending branch of the coronary artery (LAD) and found inflammatory infiltration at the aortic sinus. Therefore, we investigated the effect of circ_0001785 overexpression on cardiac function after AMI by focusing on echocardiography. Both EF and FS were improved after overexpression of circ_0001785 treatment compared with the PBS group or circ_NC group, suggesting that overexpression of circ_0001785 slowed left ventricular (LV) dysfunction after myocardial infarction (Fig. 7B, E-F). Using Masson staining, we observed a significant reduction in infarct size in circ_0001785 lentivirus-injected mice compared with overexpression of circ_NC and PBS controls (Fig. 7C, G). Since human vulnerable plaques are characterized by intraplaque neovascularization and intraplaque hemorrhage, we tested whether these features are also present in ApoE−/−+AMI mice. We found that neovascularization was observed in the plaques of mice in the ApoE−/−+AMI + PBS group, but no significant neovascularization was observed in mice in the overexpression circ_0001785 group (Fig. 7D). To further verify the role played by circ_0001785 in the mouse model, we verified the expression of circ_0001785 in the blood of healthy and model mice using qRT-PCR assay found that the expression of circ_0001785 in the blood of mice in the model group was significantly reduced..