Effects of Caveolin-1-ERK1/2 pathway on endothelial cells and smooth muscle cells under shear stress

Abstract

Hemodynamic forces have an important role in venous intimal hyperplasia, which is the main cause of arteriovenous fistula dysfunction. Endothelial cells (ECs) constantly exposed to the shear stress of blood flow, converted the mechanical stimuli into intracellular signals, and interacted with the underlying vascular smooth muscle cells (VSMCs). Caveolin-1 is one of the important mechanoreceptors on cytomembrane, which is related to vascular abnormalities. Extracellular signal-regulated kinase1/2 (ERK1/2) pathway is involved in the process of VSMCs proliferation and migration. In the present study, we explore the effects of Caveolin-1-ERK1/2 pathway and uremia toxins on the endothelial cells and VSMCs following shear stress application. Different shear stress was simulated with a ECs/VSMCs cocultured parallel-plate flow chamber system. Low shear stress and oscillating shear stress up-regulated the expression of fibroblast growth factor-4, platelet-derived growth factor-BB, vascular endothelial growth factor-A, ERK1/2 phosphorylation in endothelial cells, and proliferation and migration of VSMCs but down-regulated the Caveolin-1 expression in endothelial cells. Uremia toxin induces the proliferation and migration of VSMCs but not in a Caveolin-1-dependent manner in the static environment. Low shear stress-induced proliferation and migration of VSMCs is inhibited by Caveolin-1 overexpression and ERK1/2 suppression. Shear stress-regulated VSMC proliferation and migration is an endothelial cells-dependent process. Low shear stress and oscillating shear stress exert atherosclerotic influences on endothelial cells and VSMCs. Low shear stress modulated proliferation and migration of VSMCs through Caveolin-1-ERK1/2 pathway, which suggested that Caveolin-1 and ERK1/2 can be used as a new therapeutic target for the treatment of arteriovenous fistula dysfunction.

Impact statement: Venous intimal hyperplasia is the leading cause of arteriovenous fistula (AVF) dysfunction. This article reports that shear stress-regulated vascular smooth muscle cells (VSMCs) proliferation and migration is an endothelial cell (EC)-dependent process. Low shear stress (LSS) and oscillating shear stress (OSS) exert atherosclerotic influences on the ECs and VSMCs. LSS-induced proliferation and migration of VSMCs is inhibited by Caveolin-1 overexpression and extracellular signal-regulated kinase1/2 (ERK1/2) suppression, which suggested that Caveolin-1 and ERK1/2 can be used as a new therapeutic target for the treatment of AVF dysfunction.

Keywords: Arteriovenous fistula; caveolin-1; coculture; endothelial cells; extracellular signal-regulated kinase1/2; fibroblast growth factor; platelet-derived growth factor; shear stress; uremia toxins; vascular endothelial growth factor; vascular smooth muscle cells.

Figure 1.

(a) The components of the three-dimensional models of ECs/VSMCs coculture parallel plate flow chamber system. (b) The model of ECs/VSMCs coculture system. ECs and VSMCs are on the two sides of a porous polyethylene terephthalate membrane. (c). The model of parallel plate flow chamber system. The EC side of the coculture was applied to the designated shear stress, whereas the opposite VSMCs side was maintained under static conditions. (A color version of this figure is available in the online journal.)

Figure 2.

(a) The image shows 40× magnified hematoxylin and eosin-stained venous segments near the distal end of anastomosis (red arrow points to the inner wall and the green arrow points to the outer wall). The inner wall has evident neointimal hyperplasia. (b) The graph shows the intima-media thickening of the venous segments at the inner wall and outer wall near the distal end of anastomosis. **P < 0.001 indicates statistically significant. (c) The 100×magnified Caveolin-1 stained pictures of venous segments near the distal end of anastomosis (red arrow points to the inner wall and the green arrow points to the outer wall). The positive immunostaining for Caveolin-1 was more apparent in the outer wall. (d) Western blotting to detect phosphorylation of ERK1/2 between the inner wall and outer wall near the distal end of anastomosis. ERK1/2 phosphorylation was up-regulated in the inner wall. **P < 0.001 indicates statistically significant. (A color version of this figure is available in the online journal.)

Figure 3.

(a) The image shows a streamline diagram of fluid at 25 mL/min flow rates (4 dyn/cm²). (b) The image shows a streamline diagram of fluid at 75 mL/min flow rates (12 dyn/cm²). (c) The image shows a streamline diagram of fluid at 125 mL/min flow rates (20 dyn/cm²). (d) The changing curves of velocity and wall shear stress with time under unsteady flow conditions. The sine functions with a frequency of 1 Hz and amplitudes of 4 dyn/cm². (A color version of this figure is available in the online journal.)

Figure 4.

Western blot shows expression level of protein in ECs treated with different shear stress for 12 h. In ECs, application of LSS and OSS suppressed expression of Caveolin-1 (a), up-regulated the expressions of PDGF-BB, VEGF-A, FGF-4 (a) and increased the phosphorylation of ERK1/2 (b), compared with the static group. The proliferation (c) and migration (d and e) of cocultured VSMCs were significantly up-regulated following ECs treated with LSS and OSS, compared to the static group. NSS treatment decreased expression of VEGF-A in ECs (a). HSS treatment increased expression of Caveolin-1, PDGF-BB, FGF-4 and decreased expression of VEGF-A in ECs (a), but had no significant effects on cocultured VSMCs (c, d and e). Values were expressed as mean ± SD for each condition from five independent experiments (*P < 0.05 vs. static; **P < 0.001 vs. static). (A color version of this figure is available in the online journal.)

Figure 5.

In ECs, LSS treatment suppressed the expression of Caveolin-1 and increased the expression of VEGF-A, PDGF-BB, FGF-4, and phospho-ERK1/2 in a time-dependent fashion (a, b). The migration and proliferation of cocultured VSMCs increased gradually as time increased following the application of LSS to the ECs (c, d, e), reaching a peak level at 24 h. Compared to 0 h group, statistically significant differences were found at 6 h, 12 h, and 24 h. Values were expressed as mean ± SD for each condition from four independent experiments. (**P < 0.001 vs. 0 h). (A color version of this figure is available in the online journal.)

Figure 6.

The uremia toxin increased the expression of VEGF-A, PDGF-BB, FGF-4 and phospho-ERK1/2 (a, b) in ECs and up-regulated proliferation (c) and migration (d, e) of VSMCs, compared to normal serum static group. However, uremia toxin had no notable effects on expression of Caveolin-1 in ECs (a), compared to normal serum static group. Under uremic environment, LSS and OSS showed similar effects on ECs and cocultured VSMCs as in normal growth culture medium (a–e). NSS up-regulated the phosphorylation of ERK1/2, decreased VEGF-A, PDGF-BB in ECs and increased migration of VSMCs (a, b d, e). HSS increased the expression of Caveolin-1, phospho-ERK1/2, decreased the expression of VEGF-A, PDGF-BB, FGF-4 in ECs and down-regulated the migration of cocultured VSMCs (a, b d, e). However, both NSS and HSS had no effect on proliferation (c) in VSMCs. Values were expressed as mean ± SD for each condition from four independent experiments (#P < 0.05 compared to the normal serum static group; *P < 0.05 compared to the uremic static group; **P < 0.001 compared to the uremic static group). (A color version of this figure is available in the online journal.)

Figure 7.

The effect of Caveolin-1-ERK pathway on the LSS-induced migration and proliferation of cocultured VSMCs could be reduced by LV-Cav-1 and PD98059. In ECs, the expressions of PDGF-BB, VEGF-A, FGF-4, and phospho-ERK1/2 were down-regulated by pretreatment with LV-Cav-1 and PD98059 (a and b). Overexpressed Caveolin-1 suppressed the bioactivity of ERK1/2 in ECs also markedly repressed proliferation (c) and migration (d and e) of cocultured VSMCs. Incubation of both LV-Cav-1 and PD98059 in ECs resulted in the most significant difference. Compared to LV/Ctrl ECs, the phosphorylation of ERK1/2 was repressed in LV/Cav-1-infected ECs following the application of LSS (b). Incubation of PD98059 in ECs decreased the production of PDGF-BB, VEGF-A, FGF-4 in ECs (a), and suppressed the migration and proliferation of VSMCs (c, d and e) by LSS stimulation, but the expression of Caveolin-1 had no specific change (a). Values were expressed as mean ± SD for each condition from four independent experiments (**P < 0.001 vs. LV/Ctrl). (A color version of this figure is available in the online journal.)