STAT3/miR-15a-5p/CX3CL1 Loop Regulates Proliferation and Migration of Vascular Endothelial Cells in Atherosclerosis

Abstract

Endothelial cell proliferation disorder caused by vascular injury seems to be one of the causes of atherosclerosis, which is the pathological basis of coronary heart disease. The role of STAT3 in the regulation of microRNAs and endothelial dysfunction in atherosclerosis is unclear. STAT3 can be activated by cytokine IL-6 and up regulate the expression of CX3CL1. In addition, microRNA-15a-5p (miR-15a-5p) inhibited the transcription of CX3CL1, the proliferation of vascular endothelial cells and the proliferation of STAT3 regulated vascular endothelial cells. STAT3 positively regulates the expression of CX3CL1, and then down-regulates the inhibition of CX3CL1 by over-expression of miR-15a-5p, thus forming an elimination feedback loop to control the proliferation of HUVECs and affect the progression of atherosclerosis. In conclusion, miR-15a-5p may be the therapeutic target of the pathological basis of coronary atherosclerosis.

Keywords: Atherosclerosis; CX3CL1; Endothelial cells; MicroRNA-15a; STAT3.

Figure 1

CX3CL1 induced proliferation and promoted migration in HUVECs. HUVECs after serum starvation (DMEM+0.5% FBS) for 48 hours were transfected with si-control (si-NC) or si-CX3CL1 for 48 hours. 8-week-old male mice were fed a high fat-, high cholesterol diet (40% fat (20%w), 1.25% cholesterol for eight weeks. A: Cell proliferation was measured by counting cell numbers as described in “Experimental Procedures.” n=3. Data are expressed as mean±S.D (**P< 0.01). B and C: Cell proliferation was measured by EdU assay. n=3. Data are expressed as mean±S.D (**P< 0.01). D and E: The effect of overexpressed or knocked down CX3CL1 on HUVECs migration of was detected by Wound-healing assay. (**P< 0.01). The transwell assay was also used to detect the migration-stimulating effects of overexpressed or knocked down CX3CL1 on HUVECs. The number of cells migrated to the lower side of the transwell chambers was counted and photographed in five fields (the upper, the lower, the left, the right, and the middle) of three independent experiments, and the migration capacity was calculated by SPSS statistical soft. (**P< 0.01). F. The effect of atherosclerosis on the expression of VEGFA and CX3CL1 was examined by ELISA assay in ApoE^-/- mice. (**P< 0.01)

Figure 2

CX3CL1 expression activated in HUVECs by STAT3 signaling pathway. A-D: The effect of treated with IL-6 on CX3CL1 expression in HUVECs was detected by western blot analysis and qRT-PCR. Cells were stimulated by IL-6 at various concentrations (0, 10, 20, 40 60 and 80, 100 ng/mL) for 24 hours or at different times (0, 6, 12, 24, and 48h) at a concentration of 80ng/mL. All experiments were repeated three times in duplicate, and the semi-quantitative analysis was calculated by SPSS statistical soft. (**P< 0.01). E and F: The effect of treated with overexpression STAT3 or knocked down STAT3 on the expression of CX3CL1 was examined by western blot analysis and qRT-PCR in HUVECs. All experiments were repeated three times in duplicate, and the semi-quantitative analysis was calculated by SPSS statistical soft. (**P< 0.01)

Figure 3

MiR-15a-5p decreased CX3CL1 and inhibited HUVECs proliferation and migration. HUVECs after serum starvation (DMEM+0.5% FBS) for 48 hours were transfected with miR-15a-5p (mimics) or respective negative control (NC) for 48 hours. A: Cell proliferation was measured by counting cell numbers as described in “Experimental Procedures.” n=3. Data are expressed as mean±S.D (**P<0.01). B and C: Cell proliferation was measured by EdU assay. n=3. Data are expressed as mean±S.D (**P<0.01). D and E: The effect of overexpressed miR-15a-5p (mimics) or respective negative control (NC) on HUVECs migration of was detected by Wound-healing assay. ( *P< 0.05, **P< 0.01). The transwell assay was also used to detect the migration-stimulating effects of overexpressed miR-15a-5p (mimics) or respective negative control (NC) on HUVECs. The number of cells migrated to the lower side of the transwell chambers was counted and photographed in five fields (the upper, the lower, the left, the right, and the middle) of three independent experiments, and the migration capacity was calculated by SPSS statistical soft.(**P< 0.01)

Figure 4

miR-15a-5p inhibits CX3CL1 protein expression by targeting CX3CL1 3'UTR. A: The effect of overexpressed miR-15a-5p (mimics) or respective negative control (NC) on the expression of CX3CL1 was examined by western blot analysis in HUVECs. B: The representative image shows location of CX3CL1 in HUVECs. The above panels (green) show anti-CX3CL1 antibody reactivity to demonstrate gross morphology. The middle panels (blue) show the DAPI staining for nuclei. The below panels show double immunostained for CX3CL1and nuclei. Scale = 50µm. C: Total RNA was isolated and the expression of CX3CL1 was examined by qPCR. GAPDH was used to serve as a loading control. (**P< 0.01). D: MiR-15a-5p target sites were predicted in the 3'UTR of CX3CL1 by TargetScan Release 6.2. Bioinformatics analysis unveiled alone miR-15a-5p binding sites on the CX3CL1 3'UTR. The structure of CX3CL1 3'UTR promoter (CX3CL1-UTR-WT) and mutation the hsa-miR-15a-5p binding sites CX3CL1 3'UTR promoter (CX3CL1-UTR-MUT). E: HUVECs were respectively transfected with CX3CL1-3'UTR-luc and overexpressed miR-15a-5p (mimics) or respective negative control (NC) 24 hours, then luciferase reporter assays was used to test the transcriptional activity of CX3CL1. Values as the Relative luciferase activity (100%). All experiments were repeated three times in duplicate. (**P< 0.01)

Figure 5

The model of MiR-15a-5p-CX3CL1 and STAT3 formed a loop to regulate HUVECs proliferation. A and B: The expression of STAT3 and CX3CL1 expression were analyzed by western blot after 48 hours transfection according to the following groups: pcDNA3.1; pcDNA3.1‐STAT3; pcDNA3.1‐STAT3 + miR‐15a‐5p mimic-NC; pcDNA3.1‐STAT3 + miR‐15a‐5p mimics in HUVECs. GAPDH was a loading control. The data represent mean ± SEM (n = 3). Western blot analysis results in Figure 4B were quantified using the ImageJ software. (**P< 0.01). C: HUVECs were transiently transfected with a STAT3, or a control vector (pcDNA3.1) for 48 hours, than the expression of Cx43 was tested by qPCR assays. (**P< 0.01). D: HUVECs were transfected with the wildtype miR-15a promoter (LUC-miR-15a-5p-WT), or the miR-15a promoter with mutations in GAS site (LUC-miR-15a-5p-MUT), and transfected with STAT3 48 hours or a control vector (pcDNA3.1). Then the luciferase reporter assays was used to test the transactivity of miR-15a. Values as the Relative luciferase activity (fold). All experiments were repeated three times in duplicate. (**P< 0.01). E: HUVECs were transfected with the wild-type CX3CL1 promoter (LUC-CX3CL1-WT), or the CX3CL1 promoter with mutations in GAS site (LUC-CX3CL1-WT), and transfected with STAT3 48 hours or a control vector (pcDNA3.1). Then the luciferase reporter assays was used to test the transactivity of CX3CL1. Values as the Relative luciferase activity (fold). All experiments were repeated three times in duplicate. (**P< 0.01). F: HUVECs were transiently transfected with a STAT3, or a control vector (pcDNA3.1) for 48 hours, and ChIP assays were performed by PCR with primers targeting miR-15a-5p as described in “Materials and Methods”. GAPDH show the fold enrichment of STAT3 AT CX3CL1-GAS-sites and miR-15a-5p-GAS-sites at in-gene region in HUVECs. (**P< 0.01)