ObjectiveTo investigate the changes of glutathione peroxidase 4 (GPX4) in retinal photoreceptor cells, and the related mechanism correlated with retinal photoreceptor cell damage.

MethodsThe posterior segment tissues of 8 age-matched male donors were collected from the Body (Organ) Donation Register and Corneal Receiving Station of Tongji Hospital of Wuhan Red Cross from 2018 to 2021, including 4 non-diabetic donors and 4 diabetic donors.The tissues were divided into diabetes group and control group according to their donors.A total of 14 healthy SPF 8-week-old male C57BL/6 mice were selected and randomly divided into diabetes group and control group by the random number method, with 7 mice in each group.The mice in diabetes group were intraperitoneally injected with streptozotocin at a dose of 50 mg/kg for 5 days, and no intervention was given to mice in control group.Mouse photoreceptor cells 661W were divided into advanced glycation end products (AGEs) group and control group.AGEs group was treated with 100 μg/ml AGEs for 24 hours to simulate diabetic injury, and no intervention was given to control group.The outer segment morphology of retinal photoreceptors in human and mouse retinas was observed by hematoxylin-eosin staining.The expressions of glial fibrillary acidic protein (GFAP), rhodopsin and GPX4 in human and mouse retinas were detected by immunofluorescence staining.The expressions of GFAP, rhodopsin and GPX4 in mouse retina and the expression of GPX4 in 661W cells were determined by Western blot.The activity of 661W cells was detected by cell counting kit-8 (CCK8) method.The concentration of malondialdehyde (MDA) in mouse retina and cells was detected by TBA method.The activity of superoxide dismutase (SOD) in mouse retina and cells was detected by hydroxylamine assay.The use of human tissues was approved by the Ethics Committee of Tongji Hospital, Tongji Medical College of Huazhong University of Science and Technology (No.TJ-C20230301). The animal experiments were conducted with reference to the Standards Guide for the Care and Use of Laboratory Animals of the National Institutes of Health, and the study protocol was approved by the Experimental Animal Ethics Committee of Tongji Hospital, Tongji Medical College of Huazhong University of Science and Technology (No.TJH-2016001).

ResultsHematoxylin-eosin staining showed that retinal photoreceptor outer segments were deformed or broken in diabetic donors and diabetic mice compared with control groups.GFAP fluorescent signal mainly appeared in the inner retina of human and mice, and the stained cells were spindle or polygonal, which was consistent with the shape of glial cells.The retinal GFAP fluorescent signal of diabetic tissue and mouse groups was stronger than that of respective control groups.Rhodopsin was only expressed in the outer segment layer of photoreceptors with clear boundaries, and GPX4 was expressed in the whole retina with strong signal in the outer segment layer of photoreceptors.The fluorescent signals of rhodopsin and GPX4 in diabetic tissue and mouse groups were weaker than those in respective control groups.The relative expressions of GFAP were significantly higher and the relative expressions of rhodopsin and GPX4 were significantly lower in diabetic tissue and mouse groups than in respective control groups (all at P<0.05). The cell viability of AGEs group was significantly lower than that of control group (t=13.490, P<0.001). The relative expression of GPX4 protein in AGEs group was 0.42±0.12, which was significantly lower than 1.00±0.04 in control group (t=9.041, P<0.001). MDA concentration was higher and SOD activity was lower in retinal tissue of diabetic mice and AGEs group than those in respective control groups, and the differences were statistically significant (all at P<0.05).

ConclusionsDiabetes can reduce the GPX4 level in retinal photoreceptor cells and cause the imbalance of oxidation-antioxidant system, which may be the mechanism of the damage to retinal photoreceptor cells caused by diabetes.