Neuroprotective Effects of Four Phenylethanoid Glycosides on H2O2-Induced Apoptosis on PC12 Cells via the Nrf2/ARE Pathway

Nuclear factor erythroid 2-related factor 2 (Nrf2) is a key transcription factor against oxidative stress and neurodegenerative disorders. Phenylethanoid glycosides (PhGs; salidroside, acteoside, isoacteoside, and echinacoside) exhibit antioxidant and neuroprotective bioactivities. This study was performed to investigate the neuroprotective effect and molecular mechanism of PhGs. PhGs pretreatment significantly suppressed H2O2-induced cytotoxicity in PC12 cells by triggering the nuclear translocation of Nrf2 and reversing the downregulated protein expression of heme oxygenase 1 (HO-1), NAD(P)H quinone oxidoreductase 1 (NQO1), glutamate cysteine ligase-catalytic subunit (GCLC), and glutamate-cysteine ligase modifier subunit (GCLM). Nrf2 siRNA or HO-1 inhibitor zinc protoporphyrin (ZnPP) reduced the neuroprotective effect. PhGs showed potential interaction with the Nrf2 binding site in Kelch-like ECH-association protein 1 (Keap1). This result may support the hypothesis that PhGs are activators of Nrf2. We demonstrated the potential binding between PhGs and the Keap1-activated Nrf2/ARE pathway, and that PhGs with more glycosides had enhanced effects.


Introduction
Oxidative stress, which is an imbalance of antioxidant homeostasis, induces lipid peroxidation, injury to protein and DNA, cell aging, and cell death. This process likely contributes to several neurodegenerative disorders, such as Alzheimer's disease (AD), Parkinson's disease (PD), and ischemia/reperfusion [1]. Hydrogen peroxide (H 2 O 2 ), which is one of the main reactive oxygen species (ROS), is known to cause lipid peroxidation and DNA damage [2]. Moreover, H 2 O 2 is an endogenous source of hydroxyl free radicals that contributes to the background level of cellular oxidative stress [3,4]. Therefore, therapeutic strategies for preventing oxidative stress-induced apoptosis may have the potential in neurodegenerative diseases treatment.
Nuclear factor erythroid 2-related factor 2 (Nrf2), is a transcription factor that is strongly associated with oxidative stress. Activation of Nrf2 induces the transcription of numerous antioxidant and detoxification genes, including heme oxygenase-1 (HO-1), NAD(P)H quinine oxidoreductase 1 (NQO1), among others [5][6][7]. This process represents a key step in protecting cells from oxidative stress and is emerging as a promising therapeutic target for neuroprotection [8]. Gaia reported that Nrf2 mitigates LRRK2-and α-synuclein-induced neurodegeneration by potently promoting neuronal protein homeostasis in a cell autonomous and time-dependent manner [9].
Phenylethanoid glycosides (PhGs) are characterized by cinnamic acid and hydroxylphenylethyl moieties attached to a β-glucopyranose through ester and glycosidic linkages, respectively. These molecules are members of a group of water-soluble natural products widely distributed in the plant kingdom [10]. In vitro and in vivo studies have shown that these compounds possess antioxidant [11], neuroprotective [12], antibacterial, anti-inflammatory [13], and immunomodulatory [14] bioactivities. Osmanthus fragrans is a common ingredient in several Asian foods and has long been consumed. We previously showed that O. fragrans flower extracts enhanced spatial learning and memory, inhibited oxidative damage, and exhibited neuroprotective activities in a d-galactose-induced aging in an ICR mouse model [15]. Salidroside, acteoside, and isoacteoside are the major PhGs response for the antioxidant activities of O. fragrans flowers extracts [16].
Studies on the neuroprotective effect of PhGs have obtained desirable results. Salidroside significantly reduced cell apoptosis of PC12 cells that was exposed in MPP + [17,18]. Acteoside also alleviated MPP + -induced apoptosis and oxidative stress in PC12 cells [19] and Aβ25-35-induced SH-SY5Y cell injury [20]. Echinacoside was investigated on tumor necrosis factor-α (TNFα)-induced apoptosis in SH-SY5Y cells [21], MPTP-induced dopaminergic toxicity in mice [22], glutamate-injured primary cultures of rat cortical cells [23], and 6-OHDA-induced damage in PC12 cells [24]. The results indicated that PhGs exhibited cytoprotective effect and are potential agents to treat neurodegenerative diseases. Studies have shown the antioxidant properties of PhGs underly many other bioactivities for these compounds [25]. However, few studies have investigated the molecular mechanism of PhGs against oxidative toxicity.
In our study, we selected four typical PhGs as follows: salidroside (phenylethanoid monosaccharides), acteoside (phenylethanoid disaccharides), isoacteoside (phenylethanoid disaccharides), and echinacoside (phenylethanoid trisaccharides). We employed a model of neuronal death using differentiated PC12 cells [26] to investigate the protective effect and molecular mechanism of PhGs on H 2 O 2 -induced PC12 cell model. We demonstrated that PhGs activated the Nrf2/ARE pathway by binding to Kelch-like ECH-associated protein 1 (Keap1). This process upregulated the antioxidant enzymes and increased the resistance of PC12 cells to oxidative stress.

PhGs Alleviated H 2 O 2 -Induced Dysregulation of the Nrf2-ARE Pathways in PC12 Cells
The Nrf2-ARE pathway, as one of the major antioxidant pathways in most cell type, is important in regulating cell growth and cell death. Therefore, we investigated whether the Nrf2-ARE pathway is involved in H 2 O 2 -induced apoptosis by immunofluorescence and Western blot analysis using specific antibodies. The immunofluorescence assay showed that PhGs (0.1 and 10 µg/mL) induced the nuclear translocation of Nrf2 ( Figure 4). After treatment of PhGs, isoacteoside, and echinacoside resulted in the partial recovery of the cell normal morphology. The results from Western blotting and gray density analyses showed that PhGs (0.1 and 10 µg/mL) had no significant influence on the expression of Nrf2 in cytoplasm ( Figure 5A,B) (p > 0.05) but upregulated Nrf2 expression in the nucleus ( Figure 5C,D) (p < 0.01). Then, we studied the role of Nrf2 in H 2 O 2 -induced apoptosis by importing the Nrf2 siRNA. The protein and mRNA expression of Nrf2 was significantly downregulated in all Nrf2 siRNA-treated groups ( Figure 5E-H). PhGs (0.1 and 10 µg/mL) prevented H 2 O 2 -induced cytotoxicity in groups without Nrf2 siRNA treatment (p < 0.01), but such protection was reversed by Nrf2 siRNA. After Nrf2 siRNA treatment, the cell viability decreased significantly even with PhGs treatment ( Figure 5I).

Keap1 Expression and Molecular Docking Analysis
Under physiological conditions, Keap1 acts as a repressor protein of Nrf2 by binding to the Neh2 domain of Nrf2 and targeting Nrf2 to a Cul3-based E3 ubiquitin ligase for ubiquitination and subsequent degradation by the 26S proteasome [28]. Binding capacity to Keap1 of PhGs was evaluated by molecular docking analysis to investigate the mechanism under their antioxidant effect.
The expression of Keap1 protein did not change before and after PhGs treatment ( Figure 7A,B). Molecular docking results showed that the Keap1 protein domain had a relatively active binding pocket ( Figure 7D), and PhGs could bind to Keap1 ( Figure 7E-H). Analytical results showed that salidroside exhibited relatively weaker binding capacity. C score ≥ 4 and Total-score > 6 of echinacoside, isoacteoside, and acteoside indicated their better binding capacity with Keap1 (Table 1).   Crash: The inadequate penetration between the protein and the ligand. Polar: The hydrogen bonding and electrostatic interactions between the protein and the ligand. D-score: The charge and van der Waals interactions between the protein and the ligand. PMF-score: The helmholtz free energy between the protein and the ligand. G-score: The hydrogen bonding, complex (ligand-protein), and internal (ligand-ligand) energies between the protein and the ligand. Chem-score: The hydrogen bonding, metal-ligand interaction, lipophilic contact, and rotational entropy, along with an intercept term between the protein and the ligand. C-score: The concensus score between the protein and the ligand, comprehensive reflecting the D-score, PMF-score, G-score, and Chem-score.

Discussion
We investigated the neuroprotection of PhGs on H 2 O 2 induced-cytotoxicity in PC12 cells. The results show that PhGs pretreatment significantly suppressed H 2 O 2 -induced cytotoxicity, attenuated the intracellular ROS level, improved the level of intracellular antioxidant enzymes, and ultimately reversed H 2 O 2 -induced cytotoxicity in PC12 cells. Moreover, PhGs increased the transcriptional activation of Nrf2, reversed the H 2 O 2 -induced downregulation of the protein expression of HO-1, NQO1, GCLC, and GCLM. In addition, PhGs showed potential interaction with Nrf2 binding site in the Keap1 protein.
In the H 2 O 2 -induced PC12 cell injury, lipid peroxidation, which refers to oxidative degradation of lipid, increased the permeability of membranes, leading to cell damage [29]. MDA formation is widely used as the index of lipid peroxidation [30]. H 2 O 2 enhanced ROS production and exhausted antioxidant defense enzymes, such as SOD, catalase, and GPx. This process leads to oxidative stress [31], which plays a key role in the causation and progression of the majority of neurodegenerative disorders. Consistent with previous studies, we observed an increased level of ROS, reduced intracellular antioxidant enzymes, and enhanced apoptosis of PC12 cells after H 2 O 2 treatment. PhGs pretreatment significantly attenuated H 2 O 2 -induced increase in intracellular ROS, improved intracellular antioxidant enzymes, and ultimately reversed H 2 O 2 -induced cytotoxicity in PC12 cells.
Kuang et al. [32] reported that echinacoside showed significant neuroprotective effect on H 2 O 2 -induced cytotoxicity in PC12 cells through the mitochondrial apoptotic pathway. In this study, we found that echinacoside, salidroside, acteoside, and isoacteoside showed neuroprotective effect by enhancing the antioxidant activity of PC12 cells because they increased the transcriptional activation of Nrf2 and upregulated the downstream protein expression of HO-1, NQO1, GCLC, and GCLM. Numerous studies have clearly demonstrated that activation of Nrf2 target genes, in particular HO-1, in astrocytes and neurons strongly protect against inflammation, oxidative damage, and cell death. The HO-1 system has been reported to be very active in the central nervous system, and its modulation apparently plays a crucial role in the pathogenesis of neurodegenerative disorders [33]. Recent studies also clarified the role of Nrf2 in the progression and risk of PD [9] and Keap1 as an efficient target for the reactivation of Nrf2 in AD [34]. The results support new evidence for Nrf2 as a therapeutic target in neurodegenerative diseases.
Molecular docking analysis showed that PhGs could bind to Keap1, with the following binding capacities: echinacoside > isoacteoside > acteoside > salidroside. Consistent with these results, PhGs pretreatment led to the Nrf2 nuclear translocation, with the following Nrf2 expression in the nucleus: echinacoside > isoacteoside ≈ acteoside > salidroside. We assumed that the number of glycosides affected the possible binding mode of PhGs and Keap1, and the binding mode further caused the release of Nrf2 from Keap1. This process resulted in the activation of Nrf2 and the downstream genes and ultimately protect PC12 cells from H 2 O 2 -induced oxidative stress.
In summary, PhGs with more glycosides showed an enhanced effect on Nrf2 activation. PhGs induced Nrf2 activation by blocking the binding between Nrf2 and Keap1. This process activated the Nrf2/ARE pathway and protected PC12 from H 2 O 2 -induced cytotoxicity.

Cell Culture
Mouse adrenal pheochromocytoma line (PC12 cells) was obtained from the Institute of Biochemistry and Cell Biology, SIBS, (CAS, Shanghai, China). The cells were maintained in RPMI-1640 (Hyclone) containing 10% fetal bovine serum (Hyclone), 100 U/mL penicillin, and 0.1 mg/mL streptomycin at 37 • C with 5% CO 2 . The medium was changed every other day.

Cell Viability Assay
PC12 cells were seeded in 96-well plates at 2 × 10 4 cells/well. After attachment, cells were preincubated with or without inhibitor for 20 min, incubated with or without PhGs for 24 h, and then incubated with H 2 O 2 for another 2 h after the PhGs were removed. After incubation, the cells were treated with 5 mg/mL MTT for 4 h at 37 • C, and the media were carefully removed. The formazan crystals that had formed by surviving cells were dissolved in 150 µL of DMSO to generate a blue color [35], and the absorbance was measured at 570 nm on a plate reader. Controls utilized the same concentration of medium with DMSO alone. Cell viability was normalized as the percentage of control.
The concentrations of PhGs (0.1, 1, 5, and 10 µg/mL) was chosen depending on the cytotoxicity analysis of PhGs and the reported cytoprotective effect of echinacoside [32]. According to the report there was no cytotoxicity effect shown below 10 µg/mL, and echinacoside was reported to show cytoprotective effect in H 2 O 2 -injured cell model.

Apoptosis Assay
Apoptosis was detected with an Annexin V-FITC/PI double staining Kit (Solarbio). PC12 cells were seeded in 6-well plates at 2 × 10 5 cells/well. After attachment, cells were treated with PhGs (0.1 and 10 µg/mL) for 24 h and incubated with H 2 O 2 for another 2 h after the PhGs were removed. After incubation, the cells were washed in cold PBS, centrifuged twice at 1500 rpm for 10 min, and resuspended in 500 µL of binding buffer. FITC-labeled Annexin V (5 µL) and propidium iodide (PI, 5 µL) were then added to cells. Then, the cells were incubated in the dark at room temperature for 20 min according to the manufacturer's instruction. Cell apoptosis was measured using a Gallios™ flow cytometer (Beckman Coulter, Brea, CA, USA). Annexin V-positive and PI-negative cells were scored as early apoptotic cells, while cells double-stained with both Annexin V and PI were considered as late apoptotic cells. Control cells were negative for both stains.

Measurement of Intracellular ROS, MDA Production, and SOD
The intracellular ROS level was determined using a ROS-sensitive fluorescent probe, namely, 2,7-dichlorodihydro fluorescent diacetate (DCFH-DA). PC12 cells were seeded in 6-well plates at 2 × 10 5 cells/well. After attachment, cells were treated with PhGs (0.1 and 10 µg/mL) for 24 h and incubated with H 2 O 2 for another 2 h after the PhGs were removed. After incubation, the cells were washed with PBS thrice and subsequently incubated with 10 µM DCFH-DA. After incubation at 37 • C for 20 min, the cells were washed with PBS and collected by gentle centrifugation. Intracellular ROS was measured using a Gallios™ flow cytometer (Beckman Coulter, Brea, CA, USA).
MDA and SOD were measured by assay kits (Beyotime Biotechnology, Nanjing, Jiangsu, China). All procedures completely complied with the manufacturers' instructions. The contents of MDA and SOD were normalized with the corresponding total protein content.

Nrf2 Nuclear Translocation Immunofluorescence
The PC12 cells were seeded in 6-well glass slides at a density of 2 × 10 5 per well. After attachment, the cells were treated with PhGs (0.1 and 10 µg/mL) for 24 h and incubated with H 2 O 2 for another 2 h after the PhGs were removed. After incubation, the cells were washed in cold PBS and fixed in 4% paraformaldehyde for 15 min at room temperature, followed by membrane permeabilization using 0.5% Triton X-100 in PBS for 5 min. The cells were incubated with anti-Nrf2 Rabbit IgG (Abcam) with 5% FBS overnight at 4 • C. Then, the secondary anti-rabbit antibodies conjugated with FITC were applied to the cells for 20 min at room temperature. The nuclei were stained with 4 ,6-diamidino-2-phenylindole (DAPI) at the final preparation step. The slides were rinsed briefly with PBS, air-dried, and mounted in an anti-fluorescence in fading medium. The slides were visualized under a laser confocal microscope (LMS780, Zeiss, Germany).

Protein Extraction
The PC12 cells were seeded in 100 mm dishes at 1 × 10 7 cells/dish. After attachment, the cells were treated with PhGs (0.1 and 10 µg/mL) for 24 h and then incubated with H 2 O 2 for another 2 h with PhGs removal. After incubation, the cells were washed twice with cold PBS and scraped from the dishes with 1000 µL of PBS. Cell homogenates were centrifuged at 1500 rpm for 10 min. After treatment, cellular proteins were extracted using a Beyotime RIPA cell lysis buffer with 1 mM PMSF according to the manufacturer's instructions. In addition, cytoplasmic and nuclear proteins were isolated as described in the Beyotime nuclear and cytoplasmic extraction kit. Protein concentration of the samples was detected by a Beyotime BCA protein assay kit, and all samples were stored at −80 • C for Western blot analysis.

Quantitative Real-Time PCR
Total RNA was extracted using the RNAiso Plus (Takara, Shiga, Japan), following the manufacturer's instructions. Total RNA samples were reverse transcribed with PrimeScript™RT reagent Kit with gDNA Eraser (Takara, Shiga, Japan) according to the manufacturer's instruction. Quantitative real-time PCR was performed using SYBR ® Premix Ex Taq™ II (Takara, Shiga, Japan) in Applied Biosystems ViiA™ 7 Real-Time PCR System. Relative expression of target genes was normalized to β-Actin and analyzed by 2 −∆∆Ct method. Primer sequences for Nrf2 were as follows: forward, ACAGTGCTCCTATGCGTGAA and reverse, TCTGGGCGGCGACTTTAT. Primer sequences for β-Actin were as follows: forward, GCTGTCCCTGTATGCCTCT; and reverse, TTGATGTCACGCACGATTT.

siRNA Transfection
PC12 cells were cultured in 6-well glass slides at a density of 2 × 10 5 per well (for Western blot analysis) or in 96-well glass slides at a density of 2 × 10 4 per well (for cell viability analysis). Control siRNA and Nrf2 siRNA were transfected into cells using Lipofectamine RNAiMAX (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's instructions. After 24 h of incubation, the cells were treated with or without PhGs and H 2 O 2 for the indicated times and then used in Western blot analysis, qRT-PCR, or cell viability analysis as described above.

Molecular Docking Analysis
Molecular docking analysis was performed to investigate the possible binding mode of PhGs to Keap1. The 3D structure of ligands was obtained from NCBI. The crystalized structure of Keap1 (PDB Code: 4L7B) was prepared using correcting structure issues (such as break bond and miss loop) using SYBYL-X 2.0. In this software simulation, crash represents the inadequate penetration between the protein and the ligand. Polar represents the hydrogen bonding and electrostatic interactions between the protein and the ligand. G-score [36] represents the hydrogen bonding, complex (ligand-protein), and internal (ligand-ligand) energies between the protein and the ligand. PMF-score [37] represents the Helmholtz free energy between the protein and the ligand. D-score [38] represents the charge and van der Waals interactions between the protein and the ligand. Chem-score [39] represents the hydrogen bonding, metal-ligand interaction, lipophilic contact, and rotational entropy, along with an intercept term between the protein and the ligand. C score represents the consensus score between the protein and the ligand, comprehensively reflecting the G, PMF, D, and Chem-scores. Tota-score reflects the binding capacity of ligand to protein. The best modes were generated and evaluated using the Total-score (>6) and C-score (≥4).

Statistical Analysis
Data were analyzed using one-way ANOVA with LSD analyses using SPSS. All results were confirmed from three independent experiments. Data were expressed as means ± SD. Statistically significant differences were considered at p < 0.05.