Ultraviolet (UV) and hydrogen peroxide activate ceramide-ER stress-AMPK signaling axis to promote retinal pigment epithelium (RPE) cell apoptosis

Abstract

Ultraviolet (UV) radiation and reactive oxygen species (ROS) impair the physiological functions of retinal pigment epithelium (RPE) cells by inducing cell apoptosis, which is the main cause of age-related macular degeneration (AMD). The mechanism by which UV/ROS induces RPE cell death is not fully addressed. Here, we observed the activation of a ceramide-endoplasmic reticulum (ER) stress-AMP activated protein kinase (AMPK) signaling axis in UV and hydrogen peroxide (H2O2)-treated RPE cells. UV and H2O2 induced an early ceramide production, profound ER stress and AMPK activation. Pharmacological inhibitors against ER stress (salubrinal), ceramide production (fumonisin B1) and AMPK activation (compound C) suppressed UV- and H2O2-induced RPE cell apoptosis. Conversely, cell permeable short-chain C6 ceramide and AMPK activator AICAR (5-amino-1-β-D-ribofuranosyl-imidazole-4-carboxamide) mimicked UV and H2O2's effects and promoted RPE cell apoptosis. Together, these results suggest that UV/H2O2 activates the ceramide-ER stress-AMPK signaling axis to promote RPE cell apoptosis.

Figure 1

H₂O₂ activates endoplasmic reticulum (ER) stress, AMP activated protein kinase (AMPK) and mitogen-activated protein kinase (MAPK) signal pathways in cultured retinal pigment epithelium (RPE) cells. APRE-19 cells were either left untreated or treated with H₂O₂ (200 μM) for indicated time points; p-AMPKα (Thr 172) (A), p-ACC (Ser 79) (A), p-JNK (Thr 183/Tyr 185) (B), p-p38 (Thr 180/Tyr 182) (B), p-Erk1/2(Thr 202/Tyr 204) (B), p-PERK (Thr 980) (C), p-eIF2α (Ser 51) (C), p-Akt (Ser 473) (D) and p-S6 (Ser 235/236) (D) were detected by Western blot using specific antibodies. Non-phosphorylated kinases and β-actin were also examined as loading controls (A–D). Blot intensity of phosphorylated kinase was quantified after normalization to non-phosphorylated kinase and was expressed as fold changes vs. control group (0′). Experiments were repeated three times. *p < 0.05 vs. control group (0′).

Figure 2

H₂O₂ induces an early ceramide production, inhibited by fumonisin B1. APRE-19 cells (A,C) or primary mouse RPE cells (D,E) were either left untreated or treated with H₂O₂ (200 μM) or 5-amino-1-β-D-ribofuranosyl-imidazole-4-carboxamide (AICAR) (1 mM) or indicated time points; cellular ceramide production was analyzed, quantified and was expressed as fold changes vs. untreated control group; phospho- and non-phospho- AMPK, ACC and eIF2α were detected as described above (E); (B) APRE-19 cells were pretreated with ceramide de novo synthase inhibitor fumonisin B1 (10 μM) for one hour, followed by H₂O₂ (200 μM) for indicated time points; cellular ceramides production was analyzed, quantified and expressed as fold changes vs. control group (Ctrl). Experiments were repeated three times. *p < 0.05.

Figure 3

H₂O₂-induced AMPK and ER stress is inhibited by salubrinal (Sal), but enhanced by C6 ceramide. APRE-19 cells were pretreated with Sal (10 μM) or C6 ceramide (10 μg/mL) for 1 h, followed by H₂O₂ (200 μM) stimulation for indicated time points; the phosphorylation of AMPKα (Thr 172) and eIF2α (Ser 51) were detected by Western-blots. β-actin, non-phospho- AMPK and eIF2α were also examined (A); Blot intensity of p-AMPK and p-eIF2α was quantified after normalization to non-phospho-kinases and expressed as fold changes vs. control group (Ctrl) (B,C). Experiments were repeated three times. *p < 0.05. (B) (C)

Figure 4

H₂O₂-induced RPE cell apoptosis is suppressed by ceramide-ER stress-AMPK inhibitors. APRE-19 cells were pretreated with Sal (10 μM) for one hour, followed by H₂O₂ (200 μM) for 24 h; cell apoptosis was detected by TUNEL staining (A,B); Cultured APRE-19 cells (RPE cells) were treated as follows: control (Ctrl), H₂O₂ (200 μM), fumonisin B1 (F-B1, 10 μM), compound C (10 μM), fumonisin B1 + H₂O₂, compound C + H₂O₂ (C,D), AICAR (1 mM) and C6-ceramide (10 μg/mL) (E); cell apoptosis was detected by TUNEL staining. Experiments were repeated three times. *p < 0.05.

Figure 5

UV induces ceramide production, ER stress/AMPK activation and RPE cell death. APRE-19 cells were irradiated with indicated dosage of UV; afterwards, cells were further incubated in culture medium for one hour. Phospho- and non-phospho-eIF2α/AMPKα, as well as β-actin were tested (A); APRE-19 cells were irradiated with UV (25 mJ/cm²); afterwards, cell were incubated in culture medium for an additional 15, 30 and 60 min; cellular ceramide level was examined (B, left panel); APRE-19 cells were pretreated with fumonisin B1 (F-B1, 10 μM) (B, right panel), Sal (10 μM) or compound C (10 μM) for 1 h, followed by UV (25 mJ/cm²) radiation. Cells were then cultured in culture medium for an additional 24 h; cell viability was examined by MTT assay (C). Experiments were repeated three times. ** p < 0.05 vs. untreated control group (C, left panel), *p < 0.05 (B,C, right panel).

Figure 6

The proposed signaling pathway of this study. In RPE cells, UV radiation induces ROS production to induce an early ceramide production. Increased ceramide activates ER stress, which serves as an upstream signaling for AMPK activation. AMPK activation appears to be pro-apoptotic in this system. Suppression of this signaling axis by ceramide synthase inhibitor fumonisin B1, ER stress inhibitor salubrinal or by AMPK inhibitor compound C inhibits UV or H₂O₂-induced RPE cell death, while C6 ceramide and AMPK activator AICAR mimicked UV/H₂O₂’s effect. The role of MAPK activation in UV or H₂O₂-induced RPE cell death needs further investigation; also, the mechanism link between these pathways warrants more studies.