Analogs of imine resveratrol alleviate oxidative stress‐induced neurotoxicity in PC12 cells via activation of Nrf2

Oxidative stress is closely associated with neurodegenerative, cardiovascular and metabolic diseases. Resveratrol and related compounds have shown great potential as antioxidants via either direct scavenging of abundant reactive oxygen species (ROS) or activation of the Kelch‐like ECH‐associated protein 1‐nuclear factor (erythroid‐derived 2)‐like 2‐antioxidant response elements pathway. In the present study, we evaluated imine resveratrol analogs (IRAs) for their neuroprotective effects against ROS in PC12 cells, which are a commonly employed model system for studies of neuronal development and function. We identified that IRA‐3 (4‐[[(4‐hydroxyphenyl)methylene]amino]‐phenol) was more potent than resveratrol at rescuing PC12 cells from H2O2‐induced oxidative damage, exhibiting a recovery percentage of 60.4% at 50 μm. Our findings suggest that the neuroprotective effect of IRA‐3 was achieved via multiple routes, including direct scavenging of ROS, rescue of endogenous antioxidants and activation of the Kelch‐like ECH‐associated protein 1‐nuclear factor (erythroid‐derived 2)‐like 2‐antioxidant response elements pathway. Our results suggest that IRA‐3 may have potential for development into a possible treatment for neurodegenerative diseases.

Oxidative stress is closely associated with neurodegenerative, cardiovascular and metabolic diseases. Resveratrol and related compounds have shown great potential as antioxidants via either direct scavenging of abundant reactive oxygen species (ROS) or activation of the Kelch-like ECHassociated protein 1-nuclear factor (erythroid-derived 2)-like 2-antioxidant response elements pathway. In the present study, we evaluated imine resveratrol analogs (IRAs) for their neuroprotective effects against ROS in PC12 cells, which are a commonly employed model system for studies of neuronal development and function. We identified that IRA-3 (4-[[(4hydroxyphenyl)methylene]amino]-phenol) was more potent than resveratrol at rescuing PC12 cells from H 2 O 2 -induced oxidative damage, exhibiting a recovery percentage of 60.4% at 50 μM. Our findings suggest that the neuroprotective effect of IRA-3 was achieved via multiple routes, including direct scavenging of ROS, rescue of endogenous antioxidants and activation of the Kelch-like ECH-associated protein 1-nuclear factor (erythroidderived 2)-like 2-antioxidant response elements pathway. Our results suggest that IRA-3 may have potential for development into a possible treatment for neurodegenerative diseases.
Oxidative stress is usually caused by the generation and accumulation of reactive oxygen species (ROS), which are produced from both the activation of endogenous metabolic pathways and exogenous stimulation [1]. Many studies have indicated that oxidative stress is closely associated with most neurodegenerative diseases, cardiovascular diseases and metabolic diseases [2][3][4]. At the cellular level, ROS can induce cell apoptosis via peroxidative damage to DNA, lipids and proteins [5]. The elimination of ROS from cells depends on both exogenous antioxidants (e.g. vitamin C and curcumin) and the cellular antioxidative defense system, which includes endogenous antioxidants, such as glutathione (GSH) and cytoprotective proteins. The nuclear factor (erythroid-derived 2)-like 2 (Nrf2)-antioxidant response elements (ARE) pathway is one of the most crucial pathways in regulating antioxidative responses in mammalian cells [6][7][8][9]. In recent years, much effort has been devoted to studying the activation of Nrf2, which occurs mostly through the covalently modification of Kelchlike ECH-associated protein 1 (Keap1), an inhibitor of Nrf2 [10][11][12]. The development of natural product inspired synthetic analogs is included among the effective strategies in searching of Nrf2 activators [13,14].
In previous studies, resveratrol and related natural compounds have shown great potential as antioxidants via either direct scavenging of abundant ROS or activation of the Keap1-Nrf2-ARE pathway [7]. For example, resveratrol has been reported to suppress oxidative stress and the inflammatory response in diethylnitrosamine-initiated rat hepatocarcinogenesis [15]. Pallidol, a naturally occurring resveratrol dimer, was also confirmed to be a bi-functional antioxidant with strong singlet oxygen scavenging ability and moderate Nrf2 inducing activity [16].
A few different series of resveratrol derivatives and analogs have also been designed to improve the bioavailability and enhance the protective effect of resveratrol related compounds, enlightened by the promising results reported for resveratrol [17]. In our previous studies, a series of easily accessible resveratrol analogs, imine resveratrol analogs (IRAs), were designed and synthesized, replacing the C=C double bond of resveratrol with a C=N double bond. Luciferase assays conducted in association with quantitative structure-activity relationship studies helped with the understanding and optimization of several 6-OH IRAs as enhanced Nrf2 activators [18]. In addition, 6-OH IRAs have also been reported to have strong quenching activities against 1,1-diphenyl-2-picrylhydrazyl radical, singlet oxygen ( 1 O 2 ), hydroxyl radical (OH) and super peroxide anion (O À 2 ) [19]. Very recently, 2methoxyl-3,6-dihydroxyl-IRA was demonstrated to ameliorate colitis through the activation of Nrf2 and the inhibition of NACHT, LRR and PYD domainscontaining protein 3 expression both in vivo and in vitro [20]. In the present study, as part of a continuous effort to explore the biological benefits of IRAs, the protective effects of IRAs were investigated in neural cell models. In total, in the rat pheochromocytoma PC12 cell line, which has become a commonly employed model system for studies of neuronal development and function science [21,22], 14 IRAs were screened for their protective effects against H 2 O 2induced oxidative damage. Among them, IRA-3 showed the most significant neuroprotective effect. Further pharmacological studies confirmed that Nrf2 activation is crucial in the protective process.

Synthesis and cytotoxic effects of IRAs
In total, 14 IRAs (Table 1) were synthesized using the same procedure as that described previously [18,19]. The cytotoxicity of each compound at a concentration of 100 μM to PC12 cells was then determined using a CCK-8 assay. As shown in Fig. 1, all IRAs except IRA-1 exhibited low toxicity to PC12 cells, with cell viability exceeding 80%. However, 6-OH substituted IRA-1 was significantly toxic to PC12 cells, in contrast to its effects on other cells. This phenomenon indicated that, although 6-OH IRAs have been optimized as a potent chemopreventors for lung cancer, they might not suitable for neurologic system. It is noteworthy that twelve IRAs exhibited lower cytotoxicity than resveratrol at 100 μM.

Rescue of H 2 O 2 -induced oxidative damage by IRAs
Aiming to investigate the cytoprotective functions of IRAs, the antioxidative effects of IRAs were evaluated in PC12 cell models [23]. H 2 O 2 is a universal oxidant and widely used to generate oxidative stress models, including those involving PC12 cells [24]. In the present study, the suitable concentration of H 2 O 2 was first optimized to establish a stable model. As shown in Fig. 2A Edaravone is an approved brain-protective agent for the treatment of ischemic stroke in Japan, China and other Asian counties [25]. TBHQ is a typical antioxidant, scavenging ROS indirectly [26]. Calcein AM release and LDH leakage content in the culture medium was then determined to confirm the CCK-8 results with IRA-3 ( Fig. 2C-E).

Alleviation of H 2 O 2 -induced apoptosis by IRA-3 treatment
Apoptosis is closely associated with in oxidative stressinduced cell damage. In addition to the PI staining experiment described in Fig. 2E, the population of apoptotic PC12 cells was also determined using fluorescence microscopy after Hoechst 33342 staining [24,27]. As shown in Fig. 3A, most cells in the control group were viable (I), whereas exposure to H 2 O 2 resulted in a significant increase in the number of apoptotic cells, as indicated by blue fluorescence (IV). The nucleus was found to bedensely stained, or fragmented densely stained and condensed in a half-moon shape. Treatment of PC12 cells with 50 μM resveratrol (II) or IRA-3 (III) also induced cell apoptosis to some level. After H 2 O 2 treatment, damaged cells were incubated with resveratrol (V) or IRA-3 (VI), and the percentage of apoptotic cells decreased significantly to various degrees. IRA-3 exerted a stronger antiapoptosis effect than resveratrol. After cell apoptosis, the expression levels of Bcl-2, Bax [28], cytochrome c [29,30], caspase-3, Bim, caspase-9, cleaved PARP [31] and other proteins were varied. The results were further confirmed by the measurement of apoptosisrelated protein expression by an ELISA (Fig. 3B-E).
The changes of the levels of caspase-3 (standard curve equation: y = 0.0028x − 0.0017, r 2 = 0.9993), Bax (standard curve equation: y = 0.0317x + 0.0043, r 2 = 0.9992) and Bcl-2 (standard curve equation: y = 0.0021x + 0.0093, r 2 = 0.9987) activity were observed in the same manner. Exposure to H 2 O 2 led to significant increases in the levels of these indicators, whereas the usage of resveratrol and IRA-3 downregulated the levels of all three apoptosis-related proteins. Notably, IRA-3 showed better effects than resveratrol in all of the ELISA assays and fluorescence intensity determinations, together with lower apoptosis-inducing effect than resveratrol in PC12 cells without pretreatment.
To further confirm the anti-apoptotic effects of IRA-3, annexin V/PI double staining and flow cytometry analysis were performed to quantify early (annexin V + /PI − ) or late apoptotic (annexin V + /PI + ) cells. The representative images for flow cytometry and the summarized data are presented in Fig. 3F. After H 2 O 2 Fig. 1. Cytotoxicity of IRAs to PC12 cells. Cells were incubated with 100 μM of one of the IRAs, resveratrol, edaravone or control (serumcontaining medium with 0.1% dimethylsulfoxide) for 24 h, and then cell viability was determined using a CCK-8 assay. Results are presented as the mean AE SD (n = 4). *P < 0.05; ***P < 0.001 vs. the control group. Data were analyzed using one-way analysis of variance. treatment, the ratio of early apoptosis increased to 13.2% but stayed in relative low level in IRA-3 (9.7%) or resveratrol (7.5%) treated group. Consistently, after exposure to H 2 O 2 , the ratio of late apoptotic cells increased to 51.5%, while treatment with IRA-3 reduced the percentage of late apoptosis to 32.0%, which is more significant than the resveratrol group (47.2%).

Inhibition of ROS accumulation by and antioxidative effect of IRA-3
Previous studies have indicated that antioxidants often achieve their neuroprotective activity by lowering ROS contents in cells. Therefore, we tested the effect of IRA-3 on H 2 O 2 -induced oxidative stress in PC12 cells.
The production and accumulation of intracellular ROS was monitored using DCFH-DA, a fluorescent probe employed for ROS detection. Strong increases in ROS level were observed after exposure to H 2 O 2 (I and IV) (Fig. 4A). The treatment of model PC12 cells with resveratrol or IRA-3 significantly decreased ROS level (V and VI) (Fig. 4A), whereas the treatment compounds had no effects on ROS in normal PC12 cells (II and III) (Fig. 4A). H 2 O 2 treatment can not only directly damage cells but also cause the up-regulation of a series of oxidative stress-related small molecules. The levels of intracellular malondialdehyde (MDA) and NO were determined by corresponding assay kits. As shown in Fig. 4B,C, the levels of both MDA and NO were up-regulated by H 2 O 2 treatment and downregulated by resveratrol and IRA-3 to different extents. In addition, the cellular defensive system was activated by IRA-3, as shown in Fig. 4D,E. The activity of a representative antioxidative enzyme, SOD, was decreased by oxidative stress and rescued by IRA-3. Furthermore, the level of an endogenous antioxidant, GSH, was significantly up-regulated by IRA-3.

Activation of the Keap-Nrf2-ARE pathway by IRA-3
The Keap-Nrf2-ARE pathway is one of the most important pathways in the cellular defensive system for combating oxidative stress. In the present study, the expression of Nrf2-driven antioxidative genes was determined with a RT-PCR. As shown in Fig. 5A, the expression levels of HO-1, NQO-1, TrxR-1, Trx-1, GCLC and GCLM exhibited similar patterns. The incubation of either normal cells or H 2 O 2 -treated cells with IRA-3 resulted in significant increases in the expression of these genes. Furthermore, IRA-3 exhibited enhanced activity compared to resveratrol with respect to up-regulating all six genes. The expression of the respective proteins was then examined by western blotting (Fig. 5B). The expression of antioxidative enzymes was significantly up-regulated after treatment with IRA-3. In particular, the expression of HO-1 and NQO-1 was reduced to approximately 60% of the control under H 2 O 2 damage but was rescued and enhanced by treatment with IRA-3, increasing to over 145% of the control. Of note, several studies have suggested that Nrf2 activation was promoted under ROS stimulation. However, our results showed that pretreatment of PC12 cells with H 2 O 2 slightly decreased the levels of downstream proteins of Nrf2. This phenomenon might be attributed to the different stages of oxidative stress stimulation [32]. In the present study, the analysis was carried out at late oxidative stress state, which was after the cells pretreated with 100 μM H 2 O 2 for 6 h (pretreated with H 2 O 2 ) + 24 h (after same volume of serum-containing medium with 0.1% dimethylsulfoxide was added). In addition, other pathways also possiblly affect the levels of Nrf2 and related proteins. Glycogen synthase kinase-3 beta is an upstream factors of Nrf2 and it has also been demonstrated that H 2 O 2 can induce the activity of glycogen synthase kinase-3 beta and thereby promote the degradation of Nrf2 pathway cullin1/βTRCP [33].

Conclusions
In summary, previously reported IRAs were evaluated for their neuroprotective effects against ROS in PC12 cell models. IRA-3 was selected as the most potent compound, being able to rescue PC12 cells from H 2 O 2 -induced oxidative damage and achieving a recovery percentage of 60.4% at 50 μM. The cytoprotective effect of IRA-3 was much stronger than that of resveratrol or the positive control. The present findings, together with our previous reports, indicate that the neuroprotective effect of IRA-3 is achieved via multiple routes, including the direct scavenging of ROS, the rescue of endogenous antioxidants and the activation of the Keap1-Nrf2-ARE pathway. Our results demonstrate the potential of IRAs as simple but effective and safe neuroprotective drug leads.

Materials and methods
Cell culture and oxidative stress

Preparation of IRAs and cell treatments
Imine resveratrol analogs were synthesized as described previously [18,19] and the characterization data are provided in Appendix S1. In the biological assays, IRAs were dissolved in serum-containing medium with 0.1% dimethylsulfoxide. After treatment with 100 μM H 2 O 2 for 6 h, the cells in treatment groups were then incubated in presence of IRAs, edaravone, t-butylhydroquinone (TBHQ) or resveratrol (25, 50, 100 μM) for 24 h. The cells in H 2 O 2 group were treated in the same way without addition of drugs. For cells in control group, no H 2 O 2 or drug was added. The corresponding assays were then carried out.

Cell viability assay and calculation of cell recovery percentage
A cell counting kit-8 (CCK-8) assay (ZP328-3; Beijing Zoman Biotechnology, Beijing, China) was used to measure cell viability. In brief, 1 × 10 4 cells were seeded into each well of 96-well plates and cultured for 24 h. After appropriate treatments, the cells incubated with 10% CCK-8 serumfree solution, and cells were incubated at 37°C for an additional 3 h. Absorbance was measured at 450 nm using a microplate reader (Molecular Devices, San Jose, CA, USA). Assays were repeated at least three times. The cell recovery percentage is defined as: where V exp is the cell viability of the experimental group, V model is the cell viability of the H 2 O 2 -treated group and V control is the cell viability of the control group.

Calcein AM release assay
The cell viability assay was conducted using a Calcein AM Cell Viability Assay Kit (C2013M; Beyotime Biotechnology, Shanghai, China). After appropriate treatments, cells were incubated with test working fluid at 37°C for 30 min. Absorbance was measured using a microplate reader. In addition, the cells were double-stained with Calcein AM and propidium iodide (PI) (ST511; Beyotime Biotechnology) in the case of 50 μM administration. Cell images were recorded using inverted fluorescence microscopy under the green and red channels.

Assays for antioxidant enzyme activity
The activity of superoxide dismutase (SOD), GSH and MDA content was determined using the corresponding assay kits (S0109, S0052, S0131; Beyotime Biotechnology). PC12 cells (1 × 10 5 cellsÁmL −1 ) were seeded in six-well plates and then 100 μM H 2 O 2 was added, with treatment for 6 h. Next, the cells were cultured with or without 50 μM IRA-3 and resveratrol for 24 h. Then PC12 cells were washed with PBS, and the cells were collected and lysed with cell lysis buffer for western blotting and immunoprecipitation (P0013; Beyotime Biotechnology) treatment. After centrifugation at 14 000 g for 5 min at 4°C, the supernatant fractions were collected. The assays were conducted according to the kit instructions using a Microplate Reader.

Measurement of lactate dehydrogenase (LDH)
The assessment of H 2 O 2 -induced cell injury was performed by detection of LDH released into the medium. For this experiment, PC12 cells were seeded in 96-well plates (1 × 10 4 cells per well) for 24 h and underwent various treatments. The leakage of LDH was evaluated by measuring its activity in the culture medium using the LDH cytotoxicity assay kit in accordance with the manufacturer's instructions (C0017; Beyotime Biotechnology). Absorbance was determined at 490 nm using a microplate reader.

Detection of intracellular NO
Cells (2 × 10 5 cells per well) were seeded into six-well plates and given serum-containing medium for 24 h with diverse treatments. NO levels were determined by reaction with Fig. 3. Effects of IRA-3 on apoptosis of PC12 cells treated with H 2 O 2 and IRA-3. Cells were treated with 0 or 100 μM H 2 O 2 for 6 h, then 50 μM resveratrol or IRA-3 was added. After being incubated for 24 h, apoptotic cells were measured via Hoechst 33342 staining. Scale bars = 50 μm (A). Expression of apoptosis-related proteins (caspase-3, Bcl-2, Bax, Bcl-2/Bax) was determined by ELISA (B-E). Flow cytometry of annexin/PI double staining for H 2 O 2 -treated only, and resveratrol or IRA-3 and H 2 O 2 -treated PC12 cells (F). The results are presented as the mean AE SD (n = 4). **P < 0.01; ****P < 0.0001 vs. the control group.^P < 0.05;^^P < 0.01;^^^P < 0.001 vs. the H 2 O 2 treatment. Data were analyzed using two-way analysis of variance.  Griess reagent in accordance with the manufacturer's instructions (S0021; Beyotime Biotechnology).

Quantitative reverse transcriptase-PCR (RT-PCR)
Total RNA was extracted from PC12 cells using Trizol Reagent (A33252; Invitrogen, Carlsbad, CA, USA). The RNA concentration was measured with 5X All-In-One RT MasterMix (G490; Applied Biological Materials Inc., Richmond, Canada). Extracts were treated with RNase-free DNase to remove any residual genomic DNA. Subsequently, reverse transcription was used to obtain cDNA. Quantitative RT-PCR was carried out on an ABI 7500 Real-time PCR System (Applied Biosystems, Waltham,