Antioxidant Properties of Cerium Oxide Nanoparticles Prevent Retinal Neovascular Alterations In Vitro and In Vivo

Abstract

In this study, we investigated whether cerium oxide nanoparticles (CeO₂-NPs), a promising antioxidant nanomaterial, may contrast retinal vascular alterations induced by oxidative damage in vitro and in vivo. For the in vivo experiments, the light damage (LD) animal model of Age-Related Macular Degeneration (AMD) was used and the CeO₂-NPs were intravitreally injected. CeO₂-NPs significantly decreased vascular endothelial growth factor (VEGF) protein levels, reduced neovascularization in the deep retinal plexus, and inhibited choroidal sprouting into the photoreceptor layer. The in vitro experiments were performed on human retinal pigment epithelial (ARPE-19) cells challenged with H₂O₂; we demonstrated that CeO₂-NPs reverted H₂O₂-induced oxidative stress-dependent effects on this cell model. We further investigated the RPE-endothelial cells interaction under oxidative stress conditions in the presence or absence of CeO₂-NPs through two experimental paradigms: (i) treatment of human umbilical vein endothelial cells (HUVECs) with conditioned media from ARPE-19 cells, and (ii) coculture of ARPE-19 and HUVECs. In both experimental conditions, CeO₂-NPs were able to revert the detrimental effect of H₂O₂ on angiogenesis in vitro by realigning the level of tubule formation to that of the control. Altogether, our results indicate, for the first time, that CeO₂-NPs can counteract retinal neovascularization and may be a new therapeutic strategy for the treatment of wet AMD.

Keywords: ARPE-19; HUVEC; RPE; VEGF; angiogenesis; cerium oxide nanoparticles; glycative stress; oxidative stress; wet AMD.

Figure 1

Experimental design of the in vivo study. The in vivo experiments were performed on Sprague Dawley albino rats divided in three experimental groups (N = 4/group). (1) CTRL: control group of animals without retinal damage or treatment; (2) light damage (LD): animals subjected to light damage for 24 h and then euthanized seven days thereafter; (3) CeO₂-NPs-LD: animals treated with cerium oxide nanoparticles via intravitreal injection, then subjected to light damage for 24 h and euthanized seven days later.

Figure 2

Representative image of a whole mounted retina. The image was obtained by reconstruction from multiple images acquired with a fluorescence microscope; the retina was stained with bisbenzimide nuclear dye (blue). The black frame highlights the central superior retina, that is, the area selected for the analysis of the retinal vasculature.

Figure 3

Intravitreal injection of CeO₂-NPs prevents neovascularization in the LD animal model. (A) Western blot analysis of VEGFA in CTRL, LD, and CeO₂-NPS-LD groups (n = 4). The graph shown as the mean ± SE (standard error) of VEGFA (folds vs. CTRL) normalized versus α-Tubulin. Statistical analysis: one-way ANOVA followed by Tukey’s test, ** p < 0.01. Entire Western blot bands are shown in Supplementary Figure S2. (B,C) Vessel analysis of the deep retinal plexus. (B) Vessels percentage area and (C) number of tufts of neovascularization. Graphs are shown as mean ± SE (n = 4). Statistical analysis was performed by one-way ANOVA followed by Tukey’s test; * p < 0.05, ** p < 0.01, *** p < 0.001. (D) Representative confocal images of whole mounted retinas stained with Isolectin B4 (green); the white arrows indicate the tufts; scale bar: 50 µm. (E) Representative confocal images of retinal cryosections stained with Isolectin B4 (green) and counterstained with bisbenzimide nuclear dye (blue). The white arrows indicate the vessels sprouting from the choroid into the Outer Nuclear Layer (ONL); scale bar: 50 µm. CTRL: control; LD: rats exposed to high intensity light for 24 h and euthanized seven days thereafter; CeO₂-NPs-LD: rats treated with intravitreal injection of cerium oxide nanoparticles 3 days before light damage and euthanized seven days thereafter.

Figure 4

Antioxidant effects of CeO₂-NPs on ARPE-19 cells treated with H₂O₂. The graphs show the enzyme specific activity of SOD 2 (A), GPx (B), and GST (C), as well as GSH (D) and HO-1 (E) levels. Histograms indicate the means ± SE of three different cultures each of which was tested in triplicate. Statistical analysis was performed by one-way ANOVA followed by Tukey’s test. * p < 0.05, ** p < 0.01 vs. CTRL, and # p < 0.05, ## p<0.01 vs. H₂O₂.

Figure 5

Effect of CeO₂-NPs (0.1 mM) on dicarbonyl stress in ARPE-19 treated with two concentrations of H₂O₂ (250 and 500 µM). (A) Malondialdehyde (MDA) levels; (B) MG-H1 levels; (C) Glo1 specific activity. Histograms indicate the means ±SE of three different cultures, each of which was tested in triplicate. Statistical analysis was performed by one-way ANOVA followed by Tukey’s test * p < 0.05 and ** p < 0.01, *** p < 0.001 vs. control. # p < 0.05 vs. H₂O₂.

Figure 6

Phalloidin staining of ARPE-19 cells. (A) ARPE-19 cells stained with phalloidin (green) and bisbenzimide (blue), which label α-actin and nuclei, respectively. (a) Control cells; (b) cells exposed to 250 µM H₂O₂ and (c) 500 µM H₂O₂; (d) cells treated with 0.1 mM CeO₂-NPs; (e) cells exposed to 250 µM H₂O₂ and treated with 0.1 mM CeO₂-NPs; (f) cells exposed to 500 µM H₂O₂ and treated with CeO₂-NPs 0.1 mM. Scale bar: 100 µm. (B) The graph shows the analysis of the ARPE-19 area with six experimental conditions (control, H₂O₂ 250 µM, 500 µM, CeO₂ 0.1 mM, H₂O₂ 250 µM/CeO₂ 0.1 mM, H₂O₂ 500 µM/CeO₂ 0.1 mM). Data are shown as mean ±SE. One-way ANOVA test followed by Tukey’s test was used to perform statistical analysis (n = 200). ** p < 0.01; *** p < 0.001 versus CTRL; ### p < 0.001 versus H₂O₂.

Figure 7

Effects of H₂O₂ on tubule formation in vitro. (A) Schematic representation of the experimental workflow. (B) Histogram shows the effect of CM from ARPE-19 cells exposed to different treatments (H₂O₂ at 250 and 500 μM; CeO₂-NPs at 0.1 mM, and in combination) on the tubule formation of HUVECs. The number of tubules has been measured by counting the number of junctions/area (branching index). (C) Representative images of tubule formation of HUVECs exposed to CM from ARPE-19 treated with H₂O₂ (250 µM and 500 µM) in the presence or absence of CeO₂-NPs (0.1 mM) (a–f); scale bar: 200 mm. (D) VEGF determination in CM from ARPE-19 cells treated with H₂O₂ (250 µM and 500 µM) in the presence or absence of CeO₂-NPs (0.1 mM). Statistical analysis: one-way ANOVA followed by Tukey’s test (* p < 0.05 vs. CTRL).

Figure 8

Effects of H₂O₂ on a modified co-culture (ARPE-19-HUVECs) tubule formation in vitro. (A) Schematic representation of experimental workflow. (B) Tubule formation in co-cultured HUVECs with ARPE-19 cells treated with H₂O₂ (250, 500, 600, 800, 1000, and 1200 µM) or with H₂O₂ plus CeO₂-NPs (0.1 mM). (C) Representative images of tubule formation in different experimental conditions acquired by a Nikon inverted phase contrast microscope; scale bar: 300 mm. Statistical analysis: one-way ANOVA followed by Tukey’s test; * p < 0.05 vs. CTRL, # p < 0.05 vs. H₂O₂.