Diabetic retinopathy (DR) is the leading cause of blindness in working-age people. Pericyte loss is one of the pathologic cellular events in DR, which weakens the retinal microvessels. Damage to the microvascular networks is irreversible and permanent; thus further progression of DR is inevitable. In this study, we hypothesize that multipotent perivascular progenitor cells derived from human embryonic stem cells (hESC-PVPCs) improve the damaged retinal vasculature in the streptozotocin-induced diabetic rodent models. We describe a highly efficient and feasible protocol to derive such cells with a natural selection method without cell-sorting processes. As a cellular model of pericytes, hESC-PVPCs exhibited marker expressions such as CD140B, CD146, NG2, and functional characteristics of pericytes. Following a single intravitreal injection into diabetic Brown Norway rats, we demonstrate that the cells localized alongside typical perivascular regions of the retinal vasculature and stabilized the blood-retinal barrier breakdown. Findings in this study highlight a therapeutic potential of hESC-PVPCs in DR by mimicking the role of pericytes in vascular stabilization. Significance: This study provides a simple and feasible method to generate perivascular progenitor cells from human embryonic stem cells. These cells share functional characteristics with pericytes, which are irreversibly lost at the onset of diabetic retinopathy. Animal studies demonstrated that replenishing the damaged pericytes with perivascular progenitor cells could restore retinal vascular integrity and prevent fluid leakage. This provides promising and compelling evidence that perivascular progenitor cells can be used as a novel therapeutic agent to treat diabetic retinopathy patients.

Patients with vascular dementia, caused by cerebral ischemia, experience long-term cognitive impairment due to the lack of effective treatment. The mechanisms of and treatments for vascular dementia have been investigated in various animal models; however, the insufficient information on gene expression changes that define pathological conditions hampers progress. To investigate the underlying mechanism of and facilitate treatment development for vascular dementia, we established a mouse model of chronic cerebral hypoperfusion, including bilateral carotid artery stenosis, by using microcoils, and elucidated the molecular pathway underlying vascular dementia development. Rho-associated protein kinase (ROCK) 1/2, which regulates cellular structure, and inflammatory cytokines (IL-1 and IL-6) were upregulated in the vascular dementia model. However, expression of claudin-5, which maintains the blood–brain barrier, and MAP2 as a nerve cell-specific factor, was decreased in the hippocampal region of the vascular dementia model. Thus, we revealed that ROCK pathway activation loosens the tight junction of the blood–brain barrier and increases the influx of inflammatory cytokines into the hippocampal region, leading to neuronal death and causing cognitive and emotional dysfunction. Our vascular dementia model allows effective study of the vascular dementia mechanism. Moreover, the ROCK pathway may be a target for vascular dementia treatment development in the future.

We describe the isolation of endothelial cells from the center region of the attached embryoid body (EB) by a two-step enzyme treatment. The isolated cells from the center and outgrowth region of the EB were characterized separately. As human embryonic stem (hES) cells differentiated in EB, they lost expression of the undifferentiated marker Oct-4, whereas expression levels of endothelial-specific markers were increased. Using RT-PCR, fluorescence-activated cell sorting (FACS) analysis, and immunofluorescence, we have shown that various endothelial cell markers, including platelet/endothelial cell adhesion molecule (PECAM), von Willebrand factor (vWF), Flk-1, and Tie2, were expressed on the attached EB. Compared with the outgrowth region of EB, the center region had a higher population of cells with these endothelial cell markers. Once isolated by the twostep enzyme treatment, cells from the center region continued to differentiate into endothelial cell lineage while expression level of endothelial cell markers in cells from the outgrowth region decreased in subcultures. This study has demonstrated that the isolation of EB by a two-step enzyme treatment is a useful technique to obtain endothelial cell marker-positive cells with high yield. Furthermore, a similar approach can be taken to identify the location and distribution of specific cell types in EB and thereby allow us to isolate and expand specific cell types.