To establish a mouse model of intraretinal proliferative vitreoretinopathy by intravitreal injection of low-concentration alkali solution.

66 healthy SPF C57BL/6J mice, aged 6~8 weeks, weighing 18~20 g, of both sexes were used. The right eye of mice was used as the experimental eyes, and the left eye was used as the blank control. 0.3 M, 0.5 M, 0.75 M and 1.0 M NaOH were injected into the vitreous cavity of the experimental eyes, respectively. After injection for 7, 14, and 28 days, eyeballs were enucleated and stained with hematoxylin and eosin staining to observe the changes of retinal morphology and structure. Image J software was used to measure the thickness of full-thickness retina, inner nuclear layer and outer nuclear layer. The glial fibrillary acidic protein (GFAP) and Vimentin were used as markers of Müller cells for immunofluorescence staining to observe the glial response and glial scar of the retina. Image J software was used to quantify the mean fluorescence intensity. The thickness of each retinal layer and the mean fluorescence intensity were described as (±s) and compared by paired t-test. One-way ANOVA was used to compare the differences between multiple groups, and Tukey′s test was used for multiple comparison analysis.

After modeling for 28 days, the thickness of the full-thickness retina, inner nuclear layer and outer nuclear layer in the central retina of eyes injuried by 0.5 M NaOH were (102.40±13.93)μm, (25.08±7.78)μm, and (29.74±1.50)μm, respectively, which were lower than that of the control eyes. There were significantly differeces in the thickness of the full-thickness retina and outer nuclear layer between two eyes (t=8.56, 2.09; P<0.05). The thickness of full-thickness retina, inner nuclear layer and outer nuclear layer in the mid-peripheral retina of eyes injuried by 0.5 M NaOH were (123.60±16.37)μm, (13.05±2.80)μm, and (34.00±1.91)μm, respectively, which were significantly lower than those of eyes in the control group (t=5.01, 11.49, 7.923; P<0.05). The thickness of the full-thickness retina, inner nuclear layer and outer nuclear layer in the peripheral retina of eyes injuried by 0.5 M NaOH were (108.50±24.45)μm, (12.80±4.66)μm, and (34.45±2.90)μm, respectively. The difference of the inner retinal layer thickness between two eyes was statistically significant (t=4.76, P<0.05). After modeling for 7, 14, and 28 days, the average fluorescence intensities of GFAP were (75.43±2.968), (72.16±3.630), and (51.55±5.526), which were higher than those of the control eyes, and the differences were statistically significant (q=12.83, 11.99, 6.699; P<0.05). After modeling for 28 days, the average fluorescence intensity of GFAP was lower than those at 7 and 14 days, and the difference was statistically significant (q=6.131, 5.292; P<0.05). After modeling for 28 days, the average fluorescence intensity of Vimentin labeling Müller cells gliosis and glial scar were stronger than those of GFAP with the statistically significant differences (t=4.563, 15.72; P<0.05).

Intravitreal injection of 0.5 M NaOH in mice could successfully produce a stable model of intraretinal glioses. GFAP staining is a well-suited marker for the early reactive gliosis, while Vimentin staining is of great benifit to evaluate the gilal responses and glial scars in the later stage of injuries in the mice.