ERBB2-modulated ATG4B and autophagic cell death in human ARPE19 during oxidative stress

Abstract

Age-related macular degeneration (AMD) is an ocular disease with retinal degeneration. Retinal pigment epithelium (RPE) degeneration is mainly caused by long-term oxidative stress. Kinase activity could be either protective or detrimental to cells during oxidative stress; however, few reports have described the role of kinases in oxidative stress. In this study, high-throughput screening of kinome siRNA library revealed that erb-b2 receptor tyrosine-protein kinase 2 (ERBB2) knockdown reduced reactive oxygen species (ROS) production in ARPE-19 cells during oxidative stress. Silencing ERBB2 caused an elevation in microtubule associated protein light chain C3-II (MAP1LC3B-II/I) conversion and sequesterone (SQSTM)1 protein level. ERBB2 deprivation largely caused an increase in autophagy-regulating protease (ATG4B) expression, a protease that negatively recycles MAP1LC3-II at the fusion step between the autophagosome and lysosome, suggesting ERBB2 might modulate ATG4B for autophagy induction in oxidative stress-stimulated ARPE-19 cells. ERBB2 knockdown also caused an accumulation of nuclear factor erythroid 2-related factor 2 (NRF2) and enhanced its transcriptional activity. In addition, ERBB2 ablation or treatment with autophagy inhibitors reduced oxidative-induced cytotoxic effects in ARPE-19 cells. Furthermore, ERBB2 silencing had little or no additive effects in ATG5/7-deficient cells. Taken together, our results suggest that ERBB2 may play an important role in modulating autophagic RPE cell death during oxidative stress, and ERBB2 may be a potential target in AMD prevention.

Fig 1. Kinome siRNA screening for cytotoxic effects of ARPE-19 cells during oxidative stress.

(A) Human RPE ARPE-19 cells were treated with non-targeting siRNA for 48 h, followed by treatment with hydrogen peroxide at 62.5, 125, 250, 500, and 1000 μM for 24 h in order to determine cell viability. Cellular ROS production and (B) cell viability were measured with ROS-Glo and Cell-titer Glo, respectively. (C) Cells were treated with kinome siRNA (710 gene) for 48 h followed by treatment with hydrogen peroxide (500 μM) for 24 h in order to measure ROS production in cells. (D) The top 10-ranked hits from kinome siRNA screening were further validated for cellular ROS production in three independent experiments (Three parallel samples were included in each experiment), and the results are shown as mean ± SEM.

Fig 2. Effects of ERBB2 on autophagy in ARPE-19 cells during oxidative stress.

(A) Human RPE ARPE-19 cells were transfected with 5 nM scramble siRNA or siRNA against ERBB2 for 64 h and then treated with hydrogen peroxide (500 μM) for 8 hr. The cells were lysed for western blotting using antibodies against NRF2, SQSTM1, MAP1LC3B, ATG4B, and ACTB. (B) The quantitative results for ratio of MAP1LC3B-II/I and SQSTM1 protein level are shown. (C) The mRNA levels of SQSTM1 in cells as mentioned above were determined by real-time polymerase chain reaction (PCR). The results were analyzed using Prism 5.0 and expressed as mean ± SEM from three independent experiments (Three parallel samples were included in each experiment).

Fig 3. Effects of ERBB2 on ATG4B in ARPE-19 cell during oxidative stress.

Cells were transfected with 5 nM scramble siRNA or siRNA against ERBB2 or ATG4B for 48 h, followed by treatment with hydrogen peroxide (500 μM) for 24 h. The cells were then lysed, and equal amount of proteins were incubated with S-tagged (A) MAP1LC3B and (C) GATE-16 for 2 h. S-tag removal and ATG4B expression were examined by immunoblotting (B and D). The S-tag and ATG4B protein levels were quantitated with image J and expressed as mean ± SEM. (E) The knock-downed cells in the absence or presence of hydrogen peroxide were harvested, and nuclear and cytoplasmic fractions were split. The fractionated proteins were determined by immunoblotting using antibodies against NRF2 and ATG4B. (F) NRF2 transcriptional activity was monitored in cells harboring vector containing NRF2 promoter and luciferase.

Fig 4. Effects of autophagy inhibitors in oxidative stress-induced cell death.

(A) Human RPE ARPE-19 cells were treated with hydrogen peroxide (500 μM) in the absence or presence of autophagy inhibitor CQ (20 μM) or ConA (10 nM) for 24 h. Cell viability was quantified with Cell-titer Glo assay system. (B) The cells were transfected with 5 nM scramble siRNA or siRNA against ERBB2 for 48 h and treated with hydrogen peroxide (500 μM) in the absence or presence of autophagy inhibitors CQ (20 μM) or ConA (10 nM) for 24 h. The results were analyzed with Prism 5 and expressed as mean ± SEM from three independent experiments (Three parallel samples were included in each experiment).

Fig 5. Effects of ERBB2 in autophagy deficient ARPE-19 cells during oxidative stress.

(A) Human RPE ARPE-19 cells were transfected with 5 nM scramble siRNA or siRNA against ERBB2 without or with ULK1, BECN1, ATG5, and ATG7 for 48 h and treated with hydrogen peroxide (500 μM) for 24 h. The cells were lysed for immunoblotting to determine protein level of ERBB2, ATG5, ATG7, BECN1, ULK1, SQSTM1, and MAP1LC3B using ACTB as the internal control. (B) SQSTM1 protein levels and ratio of MAP1LC3B-II/I were quantitated with image J and expressed as mean ± SEM. (C) The knock-downed cells were treated with hydrogen peroxide (500 μM) for 8 h, and cell viability was quantified with Celltiter-Glo assay system. (D) Schematic diagram for the potential role of ERBB2 in autophagic cell death in ARPE-19 cells during oxidative stress.