Vasculogenesis in experimental stroke after human cerebral endothelial cell transplantation

Abstract

Background and purpose: Despite the reported functional recovery in transplanted stroke models and patients, the mechanism of action underlying stem cell therapy remains not well understood. Here, we examined the role of stem cell-mediated vascular repair in stroke.

Methods: Adult rats were exposed to transient occlusion of the middle cerebral artery and 3 hours later randomly stereotaxically transplantated with 100K, 200K, or 400K human cerebral endothelial cell 6 viable cells or vehicle. Animals underwent neurological examination and motor test up to day 7 after transplantation then euthanized for immunostaining against neuronal, vascular, and specific human antigens. A parallel in vitro study cocultured rat primary neuronal cells with human cerebral endothelial cell 6 under oxygen-glucose deprivation and treated with vascular endothelial growth factor (VEGF) and anti-VEGF.

Results: Stroke animals that received vehicle infusion displayed typical occlusion of the middle cerebral artery-induced behavioral impairments that were dose-dependently reduced in transplanted stroke animals at days 3 and 7 after transplantation and accompanied by increased expression of host neuronal and vascular markers adjacent to the transplanted cells. Some transplanted cells showed a microvascular phenotype and juxtaposed to the host vasculature. Infarct volume in transplanted stroke animals was significantly smaller than vehicle-infused stroke animals. Moreover, rat neurons cocultured with human cerebral endothelial cell 6 or treated with VEGF exhibited significantly less oxygen-glucose deprivation-induced cell death that was blocked by anti-VEGF treatment.

Conclusions: We found attenuation of behavioral and histological deficits coupled with robust vasculogenesis and neurogenesis in endothelial cell-transplanted stroke animals, suggesting that targeting vascular repair sets in motion a regenerative process in experimental stroke possibly via the VEGF pathway.

Keywords: endothelial cells; neurogenesis; stem cells; stroke.

Figure 1

Immunocytochemical analysis of human cerebral endothelial cells (HEN6), immortalized HEN6, immunocytochemically expressed human-specific mitochondrial marker (Mito), von Willebrand factor (vWF), and CD31. DAPI indicates 4′,6-diamidino-2-phenylindole.

Figure 2

Human cerebral endothelial cells (HEN6) ameliorates stroke-induced behavioral deficits. All animals enrolled in this study displayed no detectable behavioral deficits at baseline, with sham-operated animals (normal) exhibiting normal behaviors throughout the study period. After middle cerebral artery occlusion stroke surgery throughout the 7-day study period, stroke animals that received vehicle infusion displayed significant motor (A) and neurological impairments (B). In contrast, dose-dependent improvements across all times points in both behavioral outcomes were displayed by stroke animals transplanted with HEN6, with the highest dose of 400K most improved. The HEN6 transplanted stroke animals, although significantly improved compared with the vehicle-infused stroke animals, were still significantly impaired compared with the normal group (*P<0.05).

Figure 3

Human cerebral endothelial cells (HEN6) attenuates stroke-induced histological deficits. Transplantation of HEN6 reduced infarct volume (2,3,5-triphenyltetrazolium chloride [TTC]), suppressed reactive gliosis (GFAP), and induced vasculogenesis (collagen IV; A). Graphical rendition of correlations among GFAP (green), collagen IV (red), and infarct volume (line) are presented (B), indicating that with HEN6 reducing the infarct volumes in 400K and 200K transplanted stroke animals, there was a corresponding suppression of reactive gliosis (GFAP) and elevation of collagen IV. Numeric coefficients of correlation are shown (C). Scale bar in TTC-stained brains is 5 mm. Scale bar in GFAP and collagen IV equals 50 μm. GFAP indicates glial fibrillary acidic protein.

Figure 4

Human cerebral endothelial cells (HEN6) induces endogenous and exogenous vasculogenesis and neurogenesis. HEN6 grafts were labeled with human-specific antigen HuNu (A–C; green). The vascular marker collagen IV (A; red) revealed labeling of the exogenous transplanted HEN6 (asterisk in A) and the endogenous vasculature (arrowhead in A). Double positive cells, using the other vascular marker von Willebrand factor (vWF; B; red) colabeled with HuNu, correspond to exogenous vasculature (arrow in B), whereas vWF-positive cells but negative for HuNu represent endogenous vasculature (arrowhead in B). In addition, cells positive for the immature neural marker nestin (C; red) and colabeled with the HuNu-positive transplanted HEN6 indicate exogenous neurogenesis, whereas the nestin-positive cells but negative for HuNu represent endogenous neurogenesis. Abundant nestin-positive cells (D; green) surrounding the transplants suggest that endogenous neurogenesis within the striatum was enhanced by the HEN6 grafts. Scale bar, 50 μm.

Figure 5

Human cerebral endothelial cells (HEN6) reduces oxygen-glucose deprivation (OGD)–induced cell death in cocultured primary neuronal cells (PRNSCs). Representative caspase 3 immunocytochemical (ICC) images of control (no OGD) vs OGD under single culture of PRNCs or cocultured with HEN6 in the routine DMEM culture condition or supplemented with vascular endothelial growth factor (VEGF), anti-VEGF or combination of both (A). In addition, as appropriate coculture control condition, PRNCs were cocultured with fibroblasts (FB; ICC images not shown but comparable with single culture condition). Quantifications of all treatment conditions revealed that OGD produced significant apoptotic cell death (caspase 3; B) and impaired the oxidative metabolism (relative mitochondrial activity; C), which were blocked by VEGF treatment, and such neuroprotective effects were further enhanced by coculture with HEN6 but not with FB. *P<0.05. Scale bar, 50 μm. BM indicates basal medium.