Lycium Barbarum Polysaccharide-Based Protection to Combat H2O2-Induced Oxidative Stress via the Nrf2/HO-1 Pathway in ARPE-19 Cells

Background Age-related macular degeneration (AMD) has been closely corelated to visual impairment in the elderly, in particular to the oxidative stress (OxS) and apoptosis of retinal pigment epithelial (RPE) cells. Lycium barbarum polysaccharide (LBP), has been ascertained to promote people’s immune system, as well as to reduce neuronal damage and blood retinal barrier disruption. Nevertheless, the protective function of LBP on AMD has not been investigated. In current study, H 2 O 2 was utilized to stimulate the occurrence of OxS in RPE cells, aiming to investigate the protective function of LBP pretreatment and the underlying principle. Results: The experimental results indicated that LBP pretreatment had a signicant ecacy to reduce oxidative damage, in combination with the increased cell viability and inhibited cell apoptosis. Besides, LBP was ascertained to modulate the expression of apoptotic proteins and to activate the nuclear-related factor 2 (Nrf2) signaling pathway to protect cells. These results demonstrated that LBP could activate the Nrf2/HO-1 pathway, hence protecting ARPE-19 cells from H 2 O 2 -induced cell damage. expression of HO-1 was determined by Western blotting. Protein expression level of nuclear Nrf2 and HO-1 was analyzed by Western blotting. analyzed


Introduction
To our best knowledge, patients with AMD are suffering from progressive visional degeneration due to the affected macular area. It is reported that the worldwide population with AMD is approximately 200 million by 2020, the number will further increase to 300 million by 2040 (1). There are two prominent pathological features of AMD, the rst feature is the formation and accumulation of drusen, and another one is the damage of RPE cells (2). Therefore, protecting RPE from injury is essential in decelerating the pathological development of AMD.
As a critical part of the blood-retinal barrier (BRB), RPE plays essential and irreplaceable roles in supporting the neural retina and visual cycle, by protecting fundus tissue from oxidation (3). As compared with other tissue cells, the RPE cell layer is more easily damaged by reactive oxygen species (ROS) resulting from the markedly high oxygen consumption of the retina (4,5). Moreover, the ROS-induced damage to RPE cells was demonstrated as an irreversible process, which was an early sign in AMD(6).
Hence, therapies against OxS should be e cacious to protect normal RPE and impede the development of AMD.
The antioxidant Nrf2 is of essential importance in the immune defense system. For instance, Nrf2 in the cytoplasm is combined with the inhibitor epichlorohydrin-related protein 1 (Keap1) like Kelch (7) under physiological conditions. When the cell is damaged, Nrf2 leaks out from Keap1 and accumulates in the nucleus, triggering the downstream gene expression, such as heme oxygenase-1 (HO-1)(8). Recent research has revealed that Nrf2 and HO-1 were participated in the origin of AMD(9) by dynamically balancing retinal tissue under stress or trauma (10). Therefore, the therapeutic activation of Nrf2 could be potentially useful for AMD treatments.
LBP has exhibited various crucial biological functions, including immunomodulation, neuroprotection, anti-aging and antioxidative capacities (11). For example, LBP was reported to render a reduced level of ROS and apoptosis in human lens epithelial cells (12). In addition, ischemia-induced retinal damage on diabetic rats (13) was avoided by LBP via activating antioxidant pathway (14). Nevertheless, the protective function of LBP on AMD has not been investigated. Therefore, we aimed to evaluate the inhibitory action of LBP onwards H 2 O 2 -induced OxS and apoptosis in RPE cells, as well as to investigate its effects on the Nrf2/HO-1 pathway. And the AMD model was employed by exposing human retinal epithelial cell lines  to H 2 O 2 , to provide an alternative neoteric strategy for AMD therapy.
Experimental results presented that the cell viability was retained before and after 24 hours of LBP pretreatment, indicating that the tested concentration of LBP was safe for the cells. To evaluate the potential impact of H 2 O 2 , ARPE-19 cells were incubated with 0-1,000 µM H 2 O 2 for 2 hours, and the resulting data of cell toxicity were assessed (Fig. 1B). Notably, as compared to the control group, the cell viability was markedly reduced after the H 2 O 2 treatment. For instance, the cell viability was decreased by 53.6% (P < 0.01) at the concentration of 500 µM H 2 O 2 , and 500 µM was chosen for the following analysis. The antioxidant effect of LBP pretreatment was further evaluated. ARPE-19 cells were rst progressively treated with LBP (0.5, 1 or 2 mg/ml) and 500 µM H 2 O 2 , the nal cell viability was determined via CCK 8 kit. In Fig. 1C, ARPE-19 cell viability was resumed up to 90.33% after pretreatment with 2 mg/ml LBP.
These results suggested that 24 hours of pretreatment with LBP (0.5-2mg/ml) effectively avoided H 2 O 2induced damage in ARPE-19 cells.

LBP ameliorated H O triggered OxS
To explore the protection mechanism of LBP, the indicators of intracellular OxS and the levels of antioxidant enzymes were evaluated. In the current work, DCFH-DA assay was employed to measure ROS levels, as well as to the ability of LBP to scavenge H 2 O 2 -induced ROS. As illustrated in Fig. 2A   the anti-apoptotic effect of LBP at the protein expression level, apoptosis-related proteins including Bax and caspase-3 and the anti-apoptotic protein Bcl-2 were detected by Western blotting. In Fig. 3G, as a comparison with the control, cells exposed to 500 µM H 2 O 2 exhibited higher levels of Bax and caspase-3 but lower level of Bcl-2, which were consistent with the results obtained from ow cytometry. However, the higher level of Bcl-2 but lower levels of Bax and caspase-3 were observed after 24 hours of LBP pretreatment, indicating the dose-dependent capability of LBP on reversing H 2 O 2 -caused apoptosis with a statistically signi cant difference (Fig. 3G). In addition, the Bcl-2/Bax ratio was markedly increased in the LBP pretreatment groups while it was declined evidently in the H 2 O 2 group, demonstrating its effectively protection of H 2 O 2 -induced apoptosis in ARPE-19 cells.

LBP alleviated H 2 O 2 -induced cell damage via the Nrf2/HO-1 pathway
To explore the protective molecular mechanism, the signaling role of Nrf2/HO-1 throughout LBP-based protection was studied by respective investigation of H 2 O 2 -induced oxidative damage and apoptosis.
From Western blotting assays, H 2 O 2 treatment improved the nuclear transcription expression of Nrf2 protein, the main regulators of cellular antioxidant response. In comparison with the H 2 O 2 group, LBP pretreatment also improved the nuclear transcription expression of Nrf2 protein, with a dose-dependent effectiveness (Fig. 4A). Additionally, the downstream gene HO-1 was expressed in a similar trend to that of Nrf2 (Fig. 4B). To further validate the mechanism of LBP on H 2 O 2 -induced ARPE-19 cells, individual LBP treatment showed a negligible in uence on the expression of nuclear Nrf2 and HO-1. Conversely, a statistically signi cant increase was observed in nuclear Nrf2 and HO-1 upon induction of H 2 O 2 (Fig. 4C).
As such, the synergistic combination of LBP and OxS contributed to increasing the nuclear transport of Nrf2 protein.
As for the molecular mechanisms, Nrf2 gene knockout experiments were carried out by the siRNA mixed with lipofectamine 2000. After the intervention of Nrf2 siRNA, the expression of Nrf2 was markedly dropped in ARPE-19 cells (Fig. 4D) and the LBP-mediated expression of HO-1 was almost eliminated (Fig. 4D). Additionally, the intervention of Nrf2-siRNA aggravated the H 2 O 2 -caused cell death, offsetting the protection of LBP on cells (Fig. 4E). In conclusion, LBP could activate the Nrf2/HO-1 pathway, hence protecting ARPE-19 cells from H 2 O 2 -induced cell damage.

Discussion
As an essential part of BRB, RPE cells are of critical importance to maintain the structural integrity of the retina (15,16). And RPE cells, susceptible to the negative effects of OxS, are generally exposed to high levels of ROS (17), resulting from the higher oxygen consumption of the retina. Previous studies have correlated the cumulative ROS-induced damage in RPE cells with the early stage of AMD(18). Therefore, early intervention measurements are of essential importance to prevent OxS-induced damage in RPE cells. The anti-oxidative and anti-apoptotic functions of LBP have so far been addressed on various eye diseases, including retinitis pigmentosa (19), glaucoma (20), retinal ischemia-reperfusion injury (21) and diabetic retinopathy (22). The chemical composition analysis of LBP showed that glycopeptides in LBP could alleviate lipid peroxidation (23)(24)(25). Given the above, our study aims to explore how LBP prevent OxS and apoptosis in ARPE-19 cells and its potential mechanism.
In our experiment, a classic model (26)  According to the previous studies, the activation of the apoptotic triggered by ROS presents an essential in uence throughout AMD pathogenesis (32). In particular, the Bcl-2 family and caspase-3 proteins are two major regulators during cell apoptosis (33,34). The impact of H 2 O 2 exposure resulted in an increase in apoptotic proteins (Bax and caspase-3), but a decrease in the anti-apoptotic protein (Bcl-2). However, 24 hours of LBP pretreatment before H 2 O 2 incubation reversed the previously observed phenomenon, as evidenced by the reduced expression of Bax and caspase-3 protein and the augment of Bcl-2 protein. These results indicated that LBP was capable to lower the apoptosis of ARPE-19 cells, hence preventing the internal oxidative damage.
Furthermore, Nrf2 has heavily participated in the process of cell redox homeostasis, which serves to resist OxS by promoting the expression of antioxidant enzymes (35,36). However, few researches have focused on the relationship between LBP and the Nrf2 pathways in oxidative damage. Once the cells stimulated by OxS, Nrf2 would dissociate from Keap1 and transfer into the nucleus to activate HO-1(37, 38). Our results indicated LBP pretreatment alone haven't increased nuclear translocation of Nrf2, and similarly, the expression of HO-1 was not affected. Cell culture and treatment DMEM/F-12 with FBS (10%), streptomycin (100 mg/ml) and penicillin (100 U/ml) were used to culture the ARPE-19 cells (Procell Life Science & Technology Co. Ltd., certi ed by STR) in a humidi ed incubator (5% CO2, and 37°C). All these treatments were carried out when cells reached approximately 80% con uence.
Cell viability assay ARPE-19 cells were seeded into 96-well plates (1 X 10 4 cells per well) with six replicates for each group. Measuring intracellular ROS DCFH-DA method was utilized in this step. Brie y, ARPE-19 cells (1 X 10 6 cells per well) were cultured with or without different amounts of LBP in six-well plates for 24 hours, before treatment with 500 µM H 2 O 2 .
After that, these cells were cultured in the presence of DCFH-DA (10 mM) for 20 min within a dark environment. After thrice washed using cold PBS, the uorescence intensity of harvested cells was determined using a FACSCalibur ow cytometer (Beckman Coulter, Brea, CA). All experimental results are shown as a percentage relative to that of the control sample.
Measuring levels of MDA, SOD, CAT and GSH-Px activities In 1.5 mL Eppendorf tubes, ARPE-19 cells (1 X 10 6 cells per well) after different treatments were coincubated with 100 µL of RIPA lysis buffer and 10% protease inhibitor for 30 minutes. After lysis and 15 minutes of centrifugation (12000 g and 4°C), the protein in the resulting cell suspension was quanti ed using the BCA kit. Finally, the intracellular activities of MDA, SOD, and levels of CAT and GSH-Px were spectrophotometrically detected using the relevant commercial kits. Notably, SOD, CAT and GPX-Px activities were denoted as units/mg protein, while MDA levels were expressed as nmol/g protein. These experimental results are shown as percentages of the control value.
Quantitation of apoptotic cells Similar to the previously described procedures, the centrifuged ARPE-19 cells were resuspended with 100 µL binding buffer with a cell density at 106 cells/mL. Subsequently, 5 µL of Annexin V-FITC and 5 µL of PI were added and gently mixed into the cell suspension. After 15 minutes of dark incubation at room temperature (RT), FACSCalibur ow cytometry and CellQuest software (BD Biosciences) were utilized for the quantitation of apoptotic cells. Notably, the early, normal, and late apoptotic cells were indicated by annexin V-FITC+/PI-, annexin-/PI-, and annexin+/ PI + cell populations, respectively.

Western blot analysis
Similarly, followed by the detection of protein concentration, SDS-PAGE with protein lysates (30 µg/lane) was transferred onto PVDF membranes (Millipore, Billerica, MA, USA). After 2 hours of blocking treatment (5% skim milk) at RT, the membranes were immersed into the primary antibody at 4°C overnight. After washing, the secondary antibodies were subsequently modi ed at RT for 2 hours. The protein bands on the membranes were observed by incubation with ECL reagent and analyzed via Image Lab Software (BioRad). β-actin and histone H3 were chosen as internal controls.
siRNA Interference ARPE-19 cells (1 X 10 5 cells per well) were rstly transfected with 100 µM siRNA (control or Nrf2, GenePharma, China) using lipofectamine 2000 (Invitrogen) for 12 hours. After 24 hours of pretreatment of LBP and 2 hours of exposure to 500 µM H 2 O 2 , western blot and CCK 8 assay were nally used to evaluate the effect of LBP on protecting ARPE-19 cells and its underlying mechanism.

Statistical analysis
All data in the current study were depicted in the format of means ± SEM. GraphPad Prism software version 8.0, and one-way ANOVA and subsequent Tukey's multiple comparison test were utilized to determine the P value. A value of P < 0.05 was denoted as statistically signi cant and all assays were repeated at least in triplicate.