Fig. 1. This figure shows that the negative effect of light exposure (1000 lx over a period of 48 h) on RGC-5 cells in culture can be attenuated by the presence of EGCG, dose-dependently (5–20 μM) and also trolox (20 μM), determined by the MTT viability assay (A) or for mitochondrial dehydrogenase activity (B). Light on its own decreases cell viability and mitochondrial dehydrogenase activity by approximately 23% and 20% respectively, relative to cells maintained in the dark (100%). Results are mean values ± S.E.M. where n = 6. **p < 0.01 by one-way ANOVA followed by Tukey multi comparison test.

Fig. 2. This figure shows cultured RGC-5 cells stained for apoptosis using APOPercentage™ (A, B and C) and ROS (D, E and F). A and D are dark maintained cells where no sign of apoptosis or ROS was apparent. In contrast cultures exposed to light at 1000 lx for 48 h show some cells to exhibit a brown staining for membrane alterations (arrow heads in B) and a staining for ROS (arrows in E). When EGCG (10 μM) was present in the medium during the light exposure (C and F) little sign of apoptosis or ROS production was apparent. Scale bars = 40 μm.

Fig. 3. This figure shows experiments conducted on three different cultures of RGC-5 cells. For each culture some cells were kept in the dark and others exposed to light (1000 lx for 48 h) in the absence or presence of 10 μM EGCG or the pan-caspase inhibitor Z-VAD-FMK. A shows western blots of all the experiments were caspase-3 was determined relative to actin. No differences in caspase-3 could be detected between cells in the dark (D), light (L) or light plus EGCG (L + E). Quantification of the western blots by densitometry is shown in B. C shows viability assay data (MTT assays) where it can be seen that the detrimental effect of light exposure (no drug) to the survival of RGC-5 cells is not altered by the presence of Z-VAD-FMK at concentrations of 10 or 20 μM. Results are means ± S.E.M., where n = 3 in each case. NS = non-significant. **p < 0.05.

Fig. 4. This figure shows cultures treated in the same way as described in Fig. 2, except that the serum in the culture media was 1% rather than the normal 10%. Some cells exhibited mild apoptosis (A) and ROS production (D) in the dark but this was significantly enhanced by light exposure (B and E). The light effect was in turn attenuated by the addition of 10 μM EGCG (C and F). Scale bars = 40 μm.

Fig. 5. TUNEL staining for apoptosis for RGC-5 cells exposed to light for 96 rather than the normal 48 h (A), where it can be seen that almost every cell exhibited positive staining for TUNEL. B shows cells maintained in the dark where no evidence for TUNEL staining occurred. C shows cultures exposed to light for 96 h where EGCG (10 μM) was present. In this case only a very few cells showed some positive staining for TUNEL. D shows a positive control for TUNEL where the cells were first exposed to DNase. Scale bars = 40 μm.

Fig. 6. In these studies ERGs from both eyes were recorded. One eye was subsequently given ischemia. Five days later the ERGs from the ischemia and non-ischemic eyes were recorded and the percentage changes in the a- and b-wave amplitudes were compared between the ischemic and control eyes. A shows typical ERG recording while B shows the quantitative analysis from eight different animals in each case. It can be seen that both the a- and b-wave amplitudes of the ERG are significantly reduced by ischemia/reperfusion (results are means ± S.E.M., n = 8, **p < 0.01). Moreover, compared with vehicle-treated animals EGCG attenuated the reduction of the a- and b-wave amplitudes significantly (*p < 0.05).

Fig. 7. This figure shows that ischemia/reperfusion caused a clear change in the thickness of the Thy-1 immunoreactivity (B) in the inner plexiform layer (IPL) and almost a complete loss of the ChAT amacrine cells (E) where their processes show up as a tram line in the IPL. The effect of ischemia on the localisation of retinal Thy-1 and ChAT immunoreactivities is attenuated in animals treated with EGCG (C and F). A shows the broad band of Thy-1 immunoreactivity (large arrows) in the IPL of a control retina. D shows the tram line of ChAT immunoreactivity in the IPL (double arrows) and ChAT positive amacrine cells on either side (single arrows) of a control retina.

Fig. 8. Thy-1 (A), NF-L (B), melanopsin (C), and opsin (D) mRNA levels relative to cyclophilin mRNA in the non-ischemic retina (control) and following ischemia/reperfusion with treatment of vehicle (vehicle) or EGCG. It can be seen that ischemia/reperfusion caused significant changes on the mRNA level in vehicle-treated animals and these changes were all significantly blunted (except for opsin) by treatment with EGCG. Data are means with error bars indicating ± S.E.M.s. n = 6 in each case, *p < 0.05, **p < 0.01.

Fig. 9. This figure shows caspase-3 (A), caspase-8 (B), and GFAP (C) mRNA levels relative to cyclophilin in the non-ischemic retina (control) and following ischemia/reperfusion with treatment of vehicle (vehicle) or EGCG. Ischemia/reperfusion pronouncedly up-regulated the expression of these three mRNA levels in vehicle-treated animals, whereas EGCG treatment significantly attenuated these increases (except for caspase-3). Data are means with error bars indicating ± S.E.M.s. n = 6 in each case, *p < 0.05, **p < 0.01.

Fig. 10. NF-L (A), rhodopsin kinase (B), caspase-3 (C), PARP (D), GFAP (E), and AIF (F) proteins relative to actin in the retina in the non-ischemic ischemia (control) and following ischemia/reperfusion with vehicle treatment (vehicle) or EGCG treatment (EGCG). Ischemia/reperfusion caused a significant change in the amounts of all proteins (except for AIF) and treatment with EGCG significantly counteracted this effect. Data are means with error bars indicating ± S.E.M.s. n = 6 in each case, *p < 0.05, **p < 0.01.

Fig. 11. NF-L (A), and tubulin (B) proteins relative to actin in the non-ischemic optic nerves (control) and following ischemia/reperfusion with vehicle treatment (vehicle) or EGCG treatment (EGCG). Ischemia/reperfusion caused a significant decrease in the amounts of both NF-L and tubulin proteins. EGCG treatment significantly attenuated this reduction. Data are means with error bars indicating ± S.E.M.s. n = 8 in each case, *p < 0.05, **p < 0.01.

A comparison of the influence of light on RGC-5 cell viability (MTT assay) and mitochondrial dehydrogenase activity (WST-1 assay) in culture media contained 10% or 1% serum

p < 0.01, p < 0.05, by unpaired Student's t-test to dark and light, respectively. Results are mean values ± S.E.M. where n = 4.

It can be seen that relative to dark conditions, the influence of light in 1% serum is slightly greater than in 10% serum. Note: EGCG significantly blunted the influence of light in all cases.