Purchase this article with an account.
T. Bensaoula, C. Gerard, C. Chatenay-Rivauday, H. Ryckelynck, A. Ottlecz, G.N. Lambrou; Temporal and Spatial Characteristics of Retinal Glial Cell Responses in Oxygen-Induced Retinopathy . Invest. Ophthalmol. Vis. Sci. 2003;44(13):2896.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: To characterize glial cell responses accompanying the development of neovascularization in a rat model of oxygen-induced retinopathy (OIR) Methods: Brown-Norway rats were exposed to hyperoxia (75% oxygen and 25% room air) or room air only from postnatal days (P)7 through P12, followed by room air until P21. Retinas were collected at P12, P13, P14, P15, P17, and P21. Vessels and glial cells, were assessed in cryosections labelled with Griffonia simplicifolia isolectin B4 (GSA) and glial fibrillary acidic protein (GFAP) respectively, and examined with a light microscope equipped for fluorescence. Results: In normoxic retinas, at all time points, GFAP immunoreactivity was present around the inner retinal vessels and in fibrillary structures next to the inner limiting membrane (ILM). This pattern corresponded to astrocytes and end-feet of Müller cells. Vessels and sparse microglial cells in the inner retinal layers, were positive for GSA. In hypoxic retinas obtained from P12 to P17, GFAP immunoreactivity was demonstrated around normal GSA-labeled vessels. However, vessel-depleted retinal patches were observed predominantly in the mid-periphery. These avascular areas were associated to a decrease or an absence of GFAP immunoreactivity along the ILM. Markedly dilated and tortuous non-GFAP-immunoreactive retinal vessels and preretinal neovascular tufts were observed from P14 with a peak at P17; they were mainly located at the edge of avascular zones. New vessel sprouting into the vitreous was demonstrated, where breaches in GFAP immunoreactivity had occurred. At P21, discrete avascular zones with few slender preretinal neovascular tufts were observed. GFAP immunoreactivity was present around the inner retinal vessels and along the ILM. Across all time points, GSA strongly labeled a dense and polymorphous population of microglial cells, which occasionally had migrated to the outer retina. Conclusions: Astrocyte depletion was confirmed in the Brown-Norway rat model of OIR. This depletion was maximal at P17, corresponding to the peak of new vessel growth in our animal model. During the recovery phase, at P21, astrocytes were found to reinvest the vicinity of the inner retinal vessels. These findings suggest that astrocyte degeneration might create conditions for the growth of pre-retinal vessels in hypoxia, and bring additional evidence that astrocyte integrity might play an important role in the prevention of NV. In addition, the presence of numerous activated microglia in hypoxic retinas raises questions about their potential angiogenic function.
This PDF is available to Subscribers Only