Cyclic Stretch Induces Vascular Smooth Muscle Cells to Secrete Connective Tissue Growth Factor and Promote Endothelial Progenitor Cell Differentiation and Angiogenesis

Abstract

Endothelial progenitor cells (EPCs) play a vital role in endothelial repair following vascular injury by maintaining the integrity of endothelium. As EPCs home to endothelial injury sites, they may communicate with exposed vascular smooth muscle cells (VSMCs), which are subjected to cyclic stretch generated by blood flow. In this study, the synergistic effect of cyclic stretch and communication with neighboring VSMCs on EPC function during vascular repair was investigated. In vivo study revealed that EPCs adhered to the injury site and were contacted to VSMCs in the Sprague-Dawley (SD) rat carotid artery injury model. In vitro, EPCs were cocultured with VSMCs, which were exposed to cyclic stretch at a magnitude of 5% (which mimics physiological stretch) and a constant frequency of 1.25 Hz for 12 h. The results indicated that stretched VSMCs modulated EPC differentiation into mature endothelial cells (ECs) and promoted angiogenesis. Meanwhile, cyclic stretch upregulated the mRNA expression and secretion level of connective tissue growth factor (CTGF) in VSMCs. Recombinant CTGF (r-CTGF) treatment promoted endothelial differentiation of EPCs and angiogenesis, and increased their protein levels of FZD8 and β-catenin. CTGF knockdown in VSMCs inhibited cyclic stretch-induced EPC differentiation into ECs and attenuated EPC tube formation via modulation of the FZD8/β-catenin signaling pathway. FZD8 knockdown repressed endothelial differentiation of EPCs and their angiogenic activity. Wnt signaling inhibitor decreased the endothelial differentiation and angiogenetic ability of EPCs cocultured with stretched VSMCs. Consistently, an in vivo Matrigel plug assay demonstrated that r-CTGF-treated EPCs exhibited enhanced angiogenesis; similarly, stretched VSMCs also induced cocultured EPC differentiation toward ECs. In a rat vascular injury model, r-CTGF improved EPC reendothelialization capacity. The present results indicate that cyclic stretch induces VSMC-derived CTGF secretion, which, in turn, activates FZD8 and β-catenin to promote both differentiation of cocultured EPCs into the EC lineage and angiogenesis, suggesting that CTGF acts as a key intercellular mediator and a potential therapeutic target for vascular repair.

Keywords: angiogenesis; connective tissue growth factor; cyclic stretch; differentiation; endothelial progenitor cells; vascular smooth muscle cells.

Figure 1

EPCs adhered to the injury site 1 h after balloon damaged the intimal. (A) Schematic diagrams of the establishment of rat intimal balloon damage model. (B) HE staining indicated the significant intimal hyperplasia and the thickening blood vessel wall after 28 days compared with control group (n = 4). (C) We harvested the vessels for the immunofluorescence staining right after the surgery. The immunofluorescence staining results showed that vWF (red) was not expressed after the balloon injury, indicating that the intima fell off and the model was successfully constructed. The α-SMA (green) and DAPI (blue) were used to identify VSMCs and nuclei, respectively (n = 4). Scale bar = 50 μm. (D) The adhesion of EPCs in situ 1 h after intimal injury was identified by CD34 (stem cell marker, red) and vWF (EC marker, green) by immunofluorescence staining. The results showed that CD34 and vWF were co-expressed at same place, indicating that EPCs adhered to the damaged endothelial layer. The results showed that CD34 and vWF were colocalized, indicating that EPCs adhered to the inner wall of blood vessels. The staining of vWF in the control group was due to the presence of ECs. Nuclei are stained in blue by DAPI. The white arrows indicate the adhered EPCs at the injury site (n = 4). Scale bar = 50 μm.

Figure 2

Five percent cyclic stretch induces VSMC-derived CTGF secretion and promotes the differentiation of cocultured EPCs into ECs. (A) Schematic diagrams show EPC//VSMC coculture and the cyclic stretch system. (B) EPCs showed a spindle-shaped morphology after 8 days. Staining of FITC-UEA-lectin (green) and Dil-acLDL (red) revealed double-positive cells that were identified as EPCs (a). FACS analysis showed that EPCs were positive for the endothelial cell marker CD31 and hematopoietic stem cell markers CD34 and CD133, and they were negative for the leukocyte marker CD45. Controls (blue area) were overlaid on the histogram of each surface antigen (red areas) tested (b). (C) The mRNA levels of the EC markers CD31, vWF, and KDR showed no differences in EPCs among different cocultured conditions after 6 h (n = 5). Monocultured EPCs under static were used as control groups, shown as the dotted line. (D) The mRNA levels of the EC markers CD31, vWF, and KDR were induced in EPCs cocultured with stretched VSMCs after 12 h (n = 5). (E) EPCs cocultured with stretched VSMCs had an increased tube formation ability after 12 h (n = 5). (F) QPCR results revealed that CTGF mRNA levels were increased at different time points in stretched VSMCs (n = 5). (G) The level of CTGF secretion from stretched VSMCs was significantly elevated at 12 h (n = 5). For quantitative analysis, five fields per plate were photographed, and tube lengths were measured using Image-Pro Plus software. Scale bar = 100 μm. Values are expressed as the mean ± SD. *,^#P < 0.05 compared with the respective control. Statistical analysis was performed with Student's t-test and one-way ANOVA for (C–G).

Figure 3

r-CTGF induced EPC differentiation into ECs under static conditions; stretched VSMCs modulate the differentiation of EPCs via CTGF in vitro. (A) EPCs were treated with different doses of r-CTGF (5, 10, 15, 20, and 30 ng/ml) for 24 h, and the mRNA expression of EC markers CD31, vWF, and KDR were significantly increased at a concentration of 20 ng/ml and 30 ng/ml (n = 5). EPCs treated without r-CTGF were used as control groups. (B) EPCs treated with r-CTGF (20 ng/ml) for 24 h significantly exhibited increased tube formation (n = 5). (C) CTGF siRNA was transfected under 5% cyclic stretch conditions, and the CTGF secretion level was significantly decreased in VSMCs. (D,E) Transfection with CTGF siRNA under 5% cyclic stretch conditions with cocultured EPCs resulted in the inhibition of EPC differentiation toward ECs and decreased their tube formation (n = 5). Scale bar = 100 μm. For quantitative analysis, five fields per plate were photographed, and tube lengths were measured using Image-Pro Plus software. Values are expressed as the mean ± SD. *P < 0.05 compared with the respective control. Statistical analysis was performed with Student's t-test for (A–E).

Figure 4

The potential downstream targets of CTGF that were predicted by IPA are shown. (A) Molecules predicted to interact with CTGF were determined using IPA software. IPA analysis showed that downstream targets are involved in molecular and cellular functions (B) and canonical pathways (C).

Figure 5

CTGF regulates FZD8/β-catenin expression in EPCs. (A) The FZD8 protein level was markedly induced in EPCs treated with r-CTGF (20 ng/ml) for 24 h in comparison with the control (n = 5). (B,C) Both Western blotting and immunofluorescent staining showed upregulated β-catenin protein levels in the nuclei and cytosol of EPCs stimulated with r-CTGF. (D) Transfection with CTGF siRNA in VSMCs under 5% cyclic stretch conditions repressed the FZD8 protein level in cocultured EPCs (n = 5), and (E,F) nuclear and cytosolic β-catenin protein expression was also suppressed in cocultured EPCs. Scale bar = 50 μm. Values are expressed as the mean ± SD. *P < 0.05 compared with the control. Statistical analysis was performed with Student's t-test for (A,B,D,E).

Figure 6

Treatment with FZD8 siRNA and XAV-939 decreased EPC differentiation and tube formation. (A,B) EPCs transfected with FZD8 siRNA for 24 h exhibited significantly decreased tube formation and decreased expression of EC markers (CD31, vWF, and KDR) (n = 5). (C,D) Treatment with XAV-939 (5 μM) decreased tube formation and inhibited EPC differentiation toward ECs (n = 5). Scale bar = 100 μm. For quantitative analysis, five fields per plate were photographed, and tube lengths were measured using Image-Pro Plus software. Values are expressed as the mean ± SD. *P < 0.05 compared with the control. Statistical analysis was performed with Student's t-test for (A–D).

Figure 7

The Matrigel plug assay results shows that r-CTGF or stretched VSMCs promote the differentiation of EPCs in vivo. (A) Transplantation scheme is shown for the in vivo Matrigel plug assay. (B) r-CTGF promoted angiogenesis of EPCs in vivo. Immunofluorescent staining for CD31 revealed increased arterial endothelium formation in the Matrigel with r-CTGF-treated EPCs. In addition, few cells invaded the plug in the Matrigel that was injected r-CTGF in the absence of EPCs. The stem cell marker CD34 staining results showed that protein expression levels in EPCs treated with r-CTGF were significantly reduced compared to the control group. (C) Stretched VSMCs promoted the ability of EPCs to form tubes in vivo. Immunofluorescent staining for CD31 revealed that CD31 expression in EPCs cocultured with stretched VSMCs was significantly increased. EPCs that were cocultured with VSMCs that were exposed to cyclic stretch in Matrigel plug expressed lower level of CD34 compared with monocultured under static conditions (n = 4). Scale bar = 50 μm. Values are expressed as the mean ± SD. *P < 0.05 compared with the control. Statistical analysis was performed with Mann–Whitney t-test for (B,C).

Figure 8

EPC and r-CTGF promoted vascular intimal injury repair. (A) Schematic diagrams of the animal experimental protocol. (B) CM-Dil-labeled EPCs (red) adhered to the vascular lesion in situ. (C) The image of rat left carotid artery. (D) Intimal injury site was incubated with or without EPCs for 25–30 min, and injection of r-CTGF (2 μg/kg/day) promoted vascular repair. Reendothelialization area was observed by Evans blue staining after 7 days of the surgery. The percentage of the white area indicated the degree of reendothelialization. Scale bar = 50 μm. Values are expressed as the mean ± SD. *P < 0.05 compared with the control. Statistical analysis was performed with Mann-Whitney t-test for D.

Figure 9

The scheme here depicts the microenvironmental roles of VSMCs under cyclic stretch in EPC differentiation and angiogenesis during vascular repair. Cyclic stretch promotes the secretion of CTGF from VSMCs, which subsequently activates the FZD8/β-catenin signaling pathway in EPCs, and increased β-catenin nuclear translocation eventually induces the differentiation and angiogenic abilities of EPCs.