Purified Primary Culture Of Human Retinal Microvessel Endothelium And Pericytes With Magnetic Beads And Effects Of Inflammation And Hypoxia On Its Expression Of PDGF-B/PDGFR-β And Ang-1/Tie-2

BackgroundDiabetic retinopathy (DRP) is the leading cause of blindness in the working-age population of most developed countries, and the most common retinal vascular disease. In China, there are more than 50 million patients with diabetes now, and would increased by 10% annually. The increasing number of individuals with diabetics worldwide suggests that DRP will continue to be major contributors to vision loss and associated functional impairment for years to come.The exact pathogenesis of DRP still remains unclear. Pericyte loss, microaneurysms, and acellular capillaries are the characteristics for the early diabetic retina. The basic pathologic change of proliferative diabetic retinopathy is the formation of retinal neovascularizaton. Current researches have focused on the pathogenesis of DRP, and would provide the prerequisite of prevention and treatment of blindness from DRP.Common characteristics of pathological neovascularization include abnormal vascular permeability and defective vascular remodeling and maturation, which promote leakage, hemorrhaging and inflammation. Bothretinal microvessel endothelium and pericytes are the key targets involved in the pathologic angiogenesis of DRP, play a vital role in the pathogenesis and development of DRP. In order to investigate the characteristics of retinal microvessel endothelium and pericytes in vitro, many methods had been applied for the primary culture of the cells, but the difficulties of purification and enrichment are still remained. The methods of purification include different selective growth medium, cell strainer, cloning, pericyte weeding and use of growth factors and heparin, and recently, beads coupled with different markers had been used for the purification of some kinds of cells.For DRP, it is speculated that pericytes are the primary affected vascular cells, leading to secondary changes of the endothelium, and to dysregulated angiogenesis. But the cause of pericyte loss during early diabetic retinopahty is unclear. One hypothesis related to the pericytic accumulation of toxic products such as sorbitol or advanced glycation end products (AGEs). Pericytes loss is considered a prerequisite of microaneurysm formation, possibly by local weakening and subsequent outpouching of the capillary wall.It is well known that inflammation and hypoxia play a crucial role in the pathogenesis of DRP, the putative mode of inflammation is that increases adherence of leukocytes to capillary endothelium, which may decreases blood flow and increases hypoxia. Reduced retinal blood flow and accompanying hypoxia may be present even before the early signs of DRP. A large body of evidence now implicates increased TNF-a level in the development of DRP; and for the hypoxia, which is a major stimulus for the retinal neovascularization, had been preliminarily investigated by clinical angiographies and in the mimicked test by C0CI2 in vitro.Retinal capillary coverage with pericytes is crucial for the survival of endothelial cells, particularly under stress conditions such as diabetes. PDGF-B is involved in pericyte recruitment, and brain capillaries of mice with a genetic ablation of PDGF-B show pericyte loss and microaneurysms. PDGFR-p is the receptor of PDGF-B, Ang-1 is the ligand of Tie2, Justbecause the expression of PDGF-B and Tie-2 is endothelium-specific and the expression of PDGFR-P and Ang-1 is pericyte-specific, Ang-1 /Tie-2 and PDGF-B/PDGFR-P play a vital role in the cellular signaling between endothelial cells and pericytes.To date, the expressions of Ang-1/Tie-2 and PDGF-B/PDGFR-P in the human retinal microvessel endothelial cells (HRMECs) and pericytes (HRMPs) when exposed to TNF-a and C0CI2 have not been systemically investigated. In order to deeply understand the pathogenesis of DRP, and to provide new strategy for the prevention and treatment of DRP, the underlying tests were designed. At present study, CD31Dynabeads were used for the purification and enrichment in the primary culture of HRMECs, and the expression of PDGF-B and Tie-2 in HRMECs, the expression of PDGFR-P and Ang-1 in HRMPs were investigated when exposed to different concentration of TNF-a and C0CI2, respectively.Objective1. To develop methods for the culture of microvessel endothelium by CD31 Dynabeads and of microvessel pericytes from human retina In vitro.2. To explore the cellular biological and immunohistochemical characterizations of HRMECs and HRMPs.3. To study the effects of different concentration of TNF-a and C0CI2 on the expressions of Ang-l/Tie-2 and PDGF-B/PDGFR-P of HRMECs and HRMPs.Methods1. Human retinas were cut into pieces using razors and digested by collagenase I, and filtered by series of cell strainers or nylon mesh; HRMECs were isolated and enriched by CD31 Dynabeads, HRMPs were collected from the left suspension. The obtained HRMECs were primarily cultured with EGM in gelatin-coated dishes and HRMPs with 20% fetal bovine serum DMEM; Non-EC were removed manually.2. Morphologic observations and growth patterns of HRMECs and HRMPswere done under the inverted phase contrast microscope; Endothelial cell-specific and pericyte cell-specific properties were assessed by immunohistochemical double staining, and observed under the laser scanning confocal microscope; HRMECs and HRMPs growth curve were examined by MTT assay; Trypan blue viability test was applied for the cell viability (%).3. HRMECs and HRMPs were exposed to different concentration of TNF-a and same dosage of C0CI2 respectively , and the expression of Ang-l/Tie-2 and PDGF-B/PDGFR-J3 were evaluated by immunohistochemical staining evaluation, Western blot analysis and RT-PCR assay.Results1. Sufficient number of HRMECs were successfully harvested by the modified method that CD31 Dynabeads was applied for the purification of the HRMECs, and HRMPs were isolated by the typical method that cells were filtered through series of nylon mesh.2. HRMECs were long and thin at the initial stage in the primary culture, usually spindle-liked, and were polygon, nearly round in shape when growing into sheets; HRMECs appeared to be cobblestone-liked morphology when HRMECs nests became confluent. About 2-4 weeks, HRMECs would form a monolayer on the whole flask floor and obvious contact inhibition would appear, soon many cells will not retain cell anchorage, become round and die if not passaged in time. The margin of HRMPs was the high irregular and their growth patterns was overlapping, without contact inhibition, but retained anchorage-dependent. The time that HRMPs coverd the whole flask floor and form a monolayer would need about 2-3weeks; for the overlapping growth pattern, the HRMPs showed some contractility when became confluent. HRMECs expressed von Willebrand factor and the cell adhesion protein CD31 and CD34, and did not contain smooth muscle a-actin; the HRMECs could stained withCD31 and CD34, vWF and CD31 simultaneously under the laser scanning confocal microscope; The HRMPs expressed a-smooth muscle actin(SMA),actin and reacted with monoclonal antibody 3G5, did not contain von Willebrand factor and keratin; All the negative control group, the first antibody was PBS or serum instead, were not stained with the antibody. The growth curve of HRMECs and HRMPs were nearly S-shaped, HRMECs grew into proliferation after 3-5 days, the growth entered into plate period at about 8- 9 days, then contact inhibition began; HRMPs grew into proliferation after 3-7days, the growth entered into plate period at about 8- 11 days. The cell viability (%) of HRMECs was 87-90% and that of HRMPs was 90-93%; The cell viability (%)of HRMECs were 85-87% and that of HRMPs was 90% after liquid nitrogen cryopreservation.3. By semi-quantity RT-PCR, the expressions of PDGF-B and Tie-2 in HRMECs were up-regulated about 1.7 fold and 1.6 fold at 12h in the routine dosage of TNF-a group (lOOng/ml) respectively (vs. control); the expressions of PDGF-B and Tie-2 were down-regulated about 39% and 48% at 12h in the large dosage of TNF-a group (2450ng/ml) respectively (vs. control); For HRMPs, the expression of PDGFR-p and Ang-ldid not show significant changes in the routine dosage of TNF-a group (vs. control), the expressions of PDGFR-P and Ang-1 were down-regulated about 45% and 51% at 12h in the large dosage of TNF-a group respectively (vs. control); C0CI2 could down-regulated the expressions of PDGF-B and Tie-2 about 60% and 54% respectively (vs. control)in HRMECs, and down-regulated the expressions of PDGFR-P and Ang-1 about 63% and 60% respectively (vs. control) in HRMPs. In the group of large dosage of TNF-a and CoCl2,the results were nearly the same with the group of large dosage of TNF-a.Conclusion1. This study is the first description of successful culture and propagation of...