Lion’s Mane Mushroom, Hericium erinaceus (Bull.: Fr.) Pers. Suppresses H2O2-Induced Oxidative Damage and LPS-Induced Inflammation in HT22 Hippocampal Neurons and BV2 Microglia

Oxidative stress and inflammation in neuron-glia system are key factors in the pathogenesis of neurodegenerative diseases. As synthetic drugs may cause side effects, natural products have gained recognition for the prevention or management of diseases. In this study, hot water (HE-HWA) and ethanolic (HE-ETH) extracts of the basidiocarps of Hericium erinaceus mushroom were investigated for their neuroprotective and anti-inflammatory activities against hydrogen peroxide (H2O2)-induced neurotoxicity in HT22 mouse hippocampal neurons and lipopolysaccharide (LPS)-induced BV2 microglial activation respectively. HE-ETH showed potent neuroprotective activity by significantly (p < 0.0001) increasing the viability of H2O2-treated neurons. This was accompanied by significant reduction in reactive oxygen species (ROS) (p < 0.05) and improvement of the antioxidant enzyme catalase (CAT) (p < 0.05) and glutathione (GSH) content (p < 0.01). Besides, HE-ETH significantly improved mitochondrial membrane potential (MMP) (p < 0.05) and ATP production (p < 0.0001) while reducing mitochondrial toxicity (p < 0.001), Bcl-2-associated X (Bax) gene expression (p < 0.05) and nuclear apoptosis (p < 0.0001). However, gene expression of Nuclear factor erythroid 2-related factor 2 (Nrf2), heme oxygenase 1 (HO-1) and NAD(P)H quinone dehydrogenase 1 (NQO1) were unaffected (p > 0.05). HE-ETH also significantly (p < 0.0001) reduced nitric oxide (NO) level in LPS-treated BV2 indicating an anti-inflammatory activity in the microglia. These findings demonstrated HE-ETH maybe a potential neuroprotective and anti-inflammatory agent in neuron-glia environment.


Introduction
Oxidative stress is closely related to the pathogenesis of neurodegenerative disorders, including Alzheimer's and Parkinson's diseases [1,2]. The underlying molecular mechanisms include accumulation of reactive oxygen species (ROS) and reactive nitrogen species (RNS) which consequently leads to lipid, protein, and organelle damage, mitochondrial membrane collapse and apoptosis that culminates in neuronal death [3,4]. There is growing evidence that inflammatory reactions and the release of pro-inflammatory cytokines also result in oxidative stress and redox homeostasis disruption associated to mitochondrial dysfunction [5,6]. Excessive generation of ROS and RNS (H 2 O 2 , O 2 − , NO, ONOO − /ONOOH) can easily lead to the neuronal cell injury because these cells possess low oxidative resistance, high metabolic capacity, and non-replicative nature [7].
There is currently no report on the neuroprotective effects of H. erinaceus in H 2 O 2 -induced neurotoxicity. One study carried out in NG108-15 neuroblastoma-glioma cell line using hot water extract of H. erinaceus, was found to be non-beneficial [29]. Similarly, anti-inflammatory effects of H. erinaceus in brain microglia has never been reported. As numerous studies demonstrated the activities in RAW 264.7 murine macrophages, H. erinaceus may have anti-neuro-inflammatory activities. As more studies reported neuro-health promotion by the extracts or compounds, especially erinacines isolated from the mycelia of the mushroom, it is worthy to investigate the fruiting bodies (basidiocarps), the mushroom part where hericenones are found [30]. Hence, the current study focused on the extracts of the basidiocarps of H. erinaceus for neuroprotective activities in H 2 O 2 -induced neurotoxicity of HT22 hippocampal neurons. Both hot water (HE-HWA) and ethanolic (HE-ETH) extracts were also studied for their anti-inflammatory activities in LPS-induced inflammation in BV2 microglia to mimic the inflammation in the brain. Further, the mechanisms of neuroprotection were elucidated via investigation of antioxidant, anti-apoptosis, and mitochondrial functioning.

Preparation of Hot Water and Ethanolic Extracts
Fresh basidiocarps of Hericium erinaceus obtained from Ganofarm Sdn. Bhd., Tanjung Sepat, Malaysia were freeze-dried, blended, and stored at −20 • C prior to the extraction. The hot water extract was prepared as reported previously [31].The freeze-dried powder was dissolved in distilled water at a ratio of 1 : 20 (w/v) and left for 24 h at 27 ± 2 • C at 150 rpm. Then the mixture was double boiled for 30 min, cooled, and filtered. The aqueous extract was then freeze-dried at −50 ± 2 • C for 48 h.
The ethanolic extract was prepared as described previously [30]. The freeze-dried powder was soaked in 80% (v/v) aqueous ethanol at a ratio of 1: 10 (w/v) for three days at room temperature. At one-day interval, the solvent containing extract was decanted, filtered and the residue was re-soaked in 80% (v/v) aqueous ethanol. The extraction and filtration process were repeated for another two times. The solvent containing extract was then pooled and concentrated under vacuum using a rotary evaporator to give the ethanolic extract. Both hot water (HE-HWA) and ethanolic (HE-ETH) extracts were stored at −20 • C prior to assay.

Estimation of Total Phenolic Content (TPC)
TPC was measured as reported previously [32]. Briefly, 50 µL of HE-HWA and HE-ETH or gallic acid standard solution was mixed with 50 µL of 10% FC reagent. Then, 100 µL of 10% (w/v) sodium carbonate solution was added. The mixture was then incubated for one hour in the dark and the absorbance was measured at 750 nm using microplate reader (Sunrise, Tecan, Austria). Results were expressed as mg gallic acid equivalents (GAE) per gram of sample.

2,2-Diphenyl-1-Picrylhydrazyl (DPPH) Free Radical Scavenging Assay
DPPH free radical scavenging activity was measured as described previously [33]. DPPH reagent prepared in absolute ethanol (100 µM; 90 µL) was added to 10 µL of HE-HWA and HE-ETH at final concentration of 1 mg/mL. The mixture was incubated for 30 min in the dark and the absorbance was then measured at 518 nm. Distilled water was used as the control. The percentage inhibition or scavenging effect was calculated using the formula below: DPPH radical scavenging (%) = ((Absorbance of control − Absorbance of sample)/Absorbance of control) × 100

Cell Culture
HT22 mouse hippocampal neurons were obtained from Thermo Fisher Scientific Inc. (Waltham, MA, USA) and BV2 murine microglia were a kind gift from Sharmili Vidyadaran, Universiti Putra Malaysia, Selangor, Malaysia. The cells were maintained in DMEM with 10% FBS and 1% P/S at 37 • C, 5% CO 2 , and 90% humidity. At confluency, they were detached by TrypLE™ Express for assays and sub-culturing.

Nitric Oxide (NO) Level Assay
BV2 cells were seeded on a 96 well plate (5 × 10 4 cells/well) and treated with 1 µg/mL LPS in the presence or absence of HE-HWA or HE-ETH for 24 h. The concentration of nitrite (NO 2 ), a soluble oxidation product of NO, in the culture media was measured using Griess reagent (0.1% N-1-napthylethylenediamine dihydrochloride and 1% sulfanilamide in 5% phosphoric acid). Fifty microliters of supernatant were mixed with an equal volume of the Griess reagent and optical density was measured at 540 nm. The resulting NO level was interpreted by comparing the absorbance with the sodium nitrite standard curve.

Reactive Oxygen Species (ROS) Assay
HT22 cells were plated at a density of 1 × 10 4 cells per well in 96-well plate and incubated overnight at 37 • C in a 5% CO 2 -humidified incubator before treatment. After 6 h of treatment, media was removed and the cells were loaded with 10 µM DCFH-DA in phosphate buffered saline (PBS). The plate was returned to 37 • C incubator and incubated for 20 min. After washing with PBS, images were obtained using fluorescence microscope (Nikon Ti-S eclipse, Melville, NY, USA) and fluorescence intensity was measured at an excitation wavelength of 495 nm and an emission wavelength of 529 nm in a multi-mode reader (Biotek TM Synergy H1 Hybrid Multi-Mode Reader, Winooski, VT, USA).

Tetramethylrhodamine Ethyl Ester (TMRE) Staining of Mitochondrial Membrane Potential (MMP)
HT22 cells were plated at a density 1 × 10 4 cells per well in 96-well plate and incubated overnight at 37 • C in a 5% CO 2 -humidified incubator before treatment. After 6 h of treatment, media was removed, and the cells were loaded with 1 µg/mL of TMRE in PBS. The plate was returned to 37 • C incubator and incubated for 20 min. After washing with PBS, images were obtained using fluorescence microscope and fluorescence intensity was measured at an excitation wavelength of 549 nm and an emission wavelength of 575 nm.

Hoechst 33258 Nuclear Apoptotic Staining
HT22 cells were plated at a density of 1 × 10 4 cells per well in 96-well plate and incubated overnight at 37 • C in a 5% CO 2 -humidified incubator before treatment. After 6 h of treatment, media was removed, and the cells were loaded with 10 µg/mL of Hoechst 33,258 in PBS. The plate was returned to 37 • C incubator and incubated for 20 min. After washing with PBS, images were obtained with fluorescence microscope. Neurons with fragmented or condensed DNA and neurons with normal DNA were counted in three different fields (containing about 200-250 cells each) per well and the data was presented as apoptotic neurons as a percentage of total neurons.

Measurement of Catalase (CAT) and Glutathione (GSH) Level
The activity of antioxidant enzyme CAT and GSH content in HT22 cells were determined using commercially available kits obtained from Cayman Chemical (Ann Arbor, MI, USA). The kit was used in accordance with the manufacturer's instructions and the resulting absorbances were measured using microplate reader.

Measurement of Mitochondrial Toxicity and Adenosine Triphosphate (ATP) Levels
The levels of mitochondrial toxicity and ATP indicating mitochondrial functioning in HT22 cells were measured using Mitochondrial ToxGlo™ Assay kit from Promega™. The kits were used in accordance with the manufacturer's instructions. The resulting fluorescence and luminescence were measured in a multi-mode reader.

Measurement of Caspase 3 Activity
The activity of caspase 3 indicating apoptosis in HT22 cells was measured using Caspase-3 Assay Kit from Abcam (Cambridge, UK). The kits were used in accordance with the manufacturer's instructions. The resulting absorbances were measured using microplate reader.

Quantitative Polymerase Chain Gene (qPCR) Expression Analysis
Total RNA was obtained from the cultured HT22 neurons using FavorPrep TM Total RNA purification mini kit according to the manufacturer's protocol (Favorgen Biotech Corp., Pingtung, Taiwan).

Statistical Analysis
Data were presented as mean of triplicates ± SD. Three independent experiments were carried out as indicated. Statistical analysis was performed with one-way or two-way analysis of variance (ANOVA) with Tukey's multiple comparisons test. Student's t-test was used for TPC and DPPH assays. P values less than 0.05 were considered statistically significant. GraphPad Prism 7.0 software (San Diego, CA, USA) was utilised for the analysis.

Effects of Hot Water (HE-HWA) and Ethanolic (HE-ETH) Extracts on the Viability of BV2 and HT22
Prior to the investigation of anti-neuroinflammatory and neuroprotective activities, the effects of HE-HWA and HE-ETH on the viability of BV2 murine microglia and HT22 neurons from mouse hippocampus was assessed by MTS assay to exclude any possibilities of neurotoxicity and proliferative activities. Figure 2A,B shows that HE-HWA and HE-ETH from 50 µg/mL to 400 µg/mL exerted neither inhibition nor increased the viability of BV2 and HT22 (p > 0.05). These results suggested that both HE-HWA and HE-ETH were neither neurotoxic nor stimulated the growth of BV2 microglia and HT22 neurons. Both extracts at concentrations ranging from 100-400 µg/mL were then investigated for anti-neuroinflammation and neuroprotection activities.

Effects of Hot Water (HE-HWA) and Ethanolic (HE-ETH) Extracts on Lipopolysaccharide (LPS)-Induced NO Production in BV2
As shown in Figure 3, LPS significantly (p < 0.0001) increased the NO production from 5.06 ± 0.23 µM to 83.68 ± 1.04 µM in BV-2 cells. In contrast, co-incubation with HE-HWA and HE-ETH markedly (p < 0.0001) reduced the amount of NO production in dose-dependent manner. Although both HE-HWA and HE-ETH significantly (p < 0.0001) suppressed NO accumulation, Sidak's multiple comparison test revealed that NO reduction ability of HE-ETH was more effective than HE-HWA at 100 µg/mL, 200 µg/mL and 400 µg/mL. The highest NO reduction from 83.68 ± 1.04 µM to 38.73 ± 1.17 µM was accomplished by 400 µg/mL of HE-ETH. Values of &&&& p < 0.0001 between the treatment groups were considered as statistically significant.
The morphological observation of HT22 under light microscope further supported the MTS cell viability data ( Figure 4B). After 24 h of incubation in 250 µM H 2 O 2 , increased extracellular spaces with notable reduction in number of cells were observed. However, HE-ETH at 400 µg/mL managed to restore the viability of HT22 comparable to the control groups.

Effects of Ethanolic Extract (HE-ETH) on the Antioxidant Activities in H 2 O 2 -Treated HT22
The mechanism of neuroprotection by HE-ETH was further investigated by assessing antioxidant activities in H 2 O 2 -treated HT22 neurons. As 400 µg/mL of HE-ETH earlier demonstrated protective ability, and therefore the concentration was selected for further assays. After 6 h, the application of 250 µM H 2 O 2 resulted in significantly (p < 0.05) higher intensity of green fluorescence of DCF from DCFH-DA that strongly indicates ROS accumulation. Co-incubation with 400 µg/mL HE-ETH markedly (p < 0.05) diminished the intensity of green fluorescence from 116.10 ± 4.73% to 97.71 ± 2.66% therefore indicating its ROS-scavenging activity ( Figure 5A,B). Although H 2 O 2 treatment did not lower the CAT activity of HT22, HE-ETH significantly (p < 0.01) increased the activity of the antioxidant enzyme from 11.19 ± 0.44 nmol/min/mL to 15.19 ± 1.07 nmol/min/mL ( Figure 5C). In addition, H 2 O 2 treatment significantly (p < 0.01) depleted GSH content from 100.00 ± 5.26% to 81.42 ± 3.44% which was significantly increased to 109.40 ± 5.83% by HE-ETH ( Figure 5D).

Effects of Ethanolic Extract (HE-ETH) on the Anti-Apoptosis in H 2 O 2 -Treated HT22
Apoptotic features were assessed by nuclear labelling with membrane-permeable reagent Hoechst 33,258. The nuclei of control cells were observed as homogeneously dim-blue with regular contours and rounded shapes. After exposure to H 2 O 2 , majority of the nuclei showed bright-blue fluorescence pattern due to chromatin condensation and nuclear fragmentation indicating typical hallmarks for apoptosis. However, co-incubation with HE-ETH inhibited these characteristics of apoptosis ( Figure 7A). Quantitative analysis of Hoechst 33,258 revealed that treatment with H 2 O 2 resulted in 31.63 ± 4.41% apoptotic cells, significantly (p < 0.0001) higher compared with 1.43 ± 0.82% of apoptotic cells in control group ( Figure 5F). Co-incubation with HE-ETH significantly (p < 0.0001) decreased H 2 O 2 -induced apoptotic cells to 2.74 ± 0.58% ( Figure 7B).  Based on Figure 7C, H 2 O 2 did not alter the anti-apoptotic Bcl-2 and pro-apoptotic Bax gene expressions. However, co-incubation with HE-ETH significantly (p < 0.05) reduced Bax expression. Ratio of Bcl-2/Bax which is known to indicate the apoptotic state, inversely related to the degree of apoptosis [34] was then calculated. HE-ETH increased the Bcl-2/Bax ratio in H 2 O 2 -treated HT22 cells from 1.03 ± 0.34 to 1.77 ± 0.49 although not statistically significant (p > 0.05; p = 0.22). In addition, caspase 3 activity which is a hallmark of apoptosis [35,36] was decreased (p = 0.05) by HE-ETH ( Figure 7E).

Effects of Ethanolic Extract (HE-ETH) on the Transcriptional Expression of Nrf2/NQO1/HO1 Antioxidant Pathway
Analysis by qPCR revealed non-significant (p > 0.05) involvement of HE-ETH in regulation of Nrf2 and NQO1 expression ( Figure 8A,B). On the other hand, H 2 O 2 treatment apparently reduced relative HO-1 expression from 1.02 ± 0.23 to 0.73 ± 0.17 although it was not significant (p > 0.05; p = 0.17). Co-incubation with HE-ETH again resulted in noticeable but not significant (p > 0.05; p = 0.14) increase in HO-1 expression ( Figure 8C).

Discussion
Mycelium and basidiocarps of H. erinaceus are different in term of their nutritional compositions. While H. erinaceus is particularly well-known for two classes of neuroactive compounds, namely hericenones and erinacines, the former can be extracted from the basidiocarps while the latter is available in the mycelium [37]. Undoubtedly, cultured mycelium and the isolated erinacines have gained much more interest in neuro-health promotion. Extensive studies have well documented erinacines as potent NGF-stimulating compounds with remarkable neurite outgrowth activities [13,[38][39][40][41][42][43]. However, we proved in our previous studies that hot water and ethanolic extracts of the mushroom's basidiocarps were also beneficial in promoting neurite outgrowth activities in various immortalized neurons and dissociated cells of brain, spinal cord, and retina [44,45]. In addition to this, we managed to show that the ethanolic extract and the isolated hericenones from the basidiocarps enhanced NGF-mediated neurite outgrowth in PC12 cells via MEK/ERK and PI3K-Akt signaling pathways [30]. Hence, we shifted the focus to clarify whether the mushroom basidiocarps could also be beneficial for neuroprotection.
As basidiocarps are what is available to people in the market, they can easily become a preference in one's diet. In order for society to reap the benefits of consuming the mushroom, it has to be available at its therapeutic potential and the preparation process is within their capability.
In terms of the extraction method, hot water extraction was preferred to yield polysaccharides and other heat stable components [46,47]. Meanwhile, ethanol extraction is generally more favored as it can be used to extract secondary metabolites especially terpenoids including hericenones in H. erinaceus [22]. Despite its significantly higher TPC, the DPPH scavenging activity of HE-HWA in this study was similar to that of HE-ETH. The main issue with screening of TPC and DPPH scavenging activity is always the solubility. It is well-known that TPC is a water-based assay, while DPPH assay is based in ethanol. Hence, HE-HWA which is water soluble could give a higher TPC but lower DPPH scavenging activity due to their solvent preferences [48]. In addition, hot water extraction as the name suggests, could deactivate or degrade the heat-labile antioxidant components hence explain why HE-HWA did not get the better of HE-ETH in DPPH scavenging assay regardless of its higher TPC [49][50][51].
Microglia-mediated neuroinflammation is considered a major source of reactive oxygen species (ROS) and reactive nitrogen species (RNS) in the brain and plays a significant role in the pathogenesis of neurodegenerative diseases [52]. Microglial cells constitute the brain's defence system by inducing phagocytosis and regulating the release of pro-inflammatory cytokines [45,46]. However, chronic and excessive induction of pro-inflammatory mechanisms by activated microglia will lead to overproduction of potentially neurotoxic cytokines, ROS and RNS; including NO and H 2 O 2 [53][54][55][56]. Despite its' role as both neuromodulator and neurotransmitter in the central nervous system (CNS), the overproduction of NO has been implicated in neuroinflammation and subsequent neuronal death in neurodegenerative diseases [57]. In our study, HE-HWA and HE-ETH from the mushroom's basidiocarps were shown to reduce the NO level in LPS-activated BV2 microglia. However, the percentage reduction of NO was more immense by HE-ETH at every concentration tested. A huge number of studies reported anti-inflammatory activities of H. erinaceus by suppressing the release of NO in RAW 264.7 murine macrophage from blood [24][25][26][27][28]. To the best of our knowledge, this was the first report showing NO-suppressing activities by H. erinaceus in BV2 microglia indicating its potential in ameliorating chronic inflammation in the CNS. Nevertheless, the anti-inflammatory activities need to be further clarified by investigating the release of other important pro-inflammatory cytokines such as tumor necrosis factor alpha (TNFα) and interleukins as well as inflammatory mediators including inducible nitric oxide synthase (iNOS), cyclooxygenase (COX), and prostaglandin E2 (PGE2).
Hydrogen peroxide (H 2 O 2 ), a highly reactive ROS, is one of the most important mediators of oxidative stress proven to induce neuronal dysfunction through the activation of mitochondria related apoptotic signals [58,59]. H 2 O 2 has been widely utilised as a neurotoxic challenge paradigm to mimic in vitro oxidative stress in many different cell types including murine hippocampal HT22 cells [60][61][62].
In this study, we demonstrated the ability of HE-ETH to successfully protect HT22 neurons from H 2 O 2 -induced neurotoxicity. Our study further elucidated the mechanisms of neuroprotection by HE-ETH in HT22 through improvement of antioxidant, mitochondrial functioning, and anti-apoptosis. Although H. erinaceus was previously demonstrated to stimulate neuroprotection in various model of induced neurotoxicity, this was the first report showing its ability in H 2 O 2 -induced oxidative neurotoxicity. Other reports utilized different neurotoxic models including glutamate, amyloid beta, and MPTP to assess neuroprotective effects of the mushrooms [19][20][21][22][23]. An attempt to show neuroprotective activity of H. erinaceus against H 2 O 2 in NG108-15 neuroblastoma-glioma cell line was non successful [29]. Instead, a study conducted in human gastric mucosa epithelium cell revealed the ability of an isolated polysaccharide, EP-1 from mycelium of H. erinaceus to prevent H 2 O 2 -induced toxicity death by inhibiting activation of apoptotic cellular signals within mitochondria-dependent apoptosis [46]. This was in accordance with our study, which demonstrated the role of similar mechanisms in the promotion of neuronal viability.
Diverse studies have reported that mushrooms possess bioactive compounds that enhance various antioxidant enzymes in the body including CAT, GSH, and superoxide dismutase (SOD). These enzymes act by catalyzing reactions that neutralize free radicals and ROS in order to protect cells from damage [63,64]. CAT is a primary defence unit against oxidative toxicity, responsible for breakdown of H 2 O 2 to O 2 (47). Numerous mushrooms such as Ganoderma lucidum, Pleurotus cystidiosus, P. ostreatus, P. djamor, and P. eryngii were effective in elevating CAT activity, hence preventing H 2 O 2 -induced toxicity [63,[65][66][67][68][69]. In addition, several mushrooms including Agaricus bisporus, G. lucidum and P. ostreatus demonstrated the GSH-related activities. GSH, often referred to as a master antioxidant compound is crucial to block reactive hydroxyl free radicals and oxygen-centred free radicals hence protecting cells from toxicity [65,[68][69][70][71]. Hence, the significant reduction of ROS in this study could be attributed to the elevated level of CAT activity and GSH content. Antioxidant, mitochondrial health and anti-apoptosis are always interrelated in overcoming oxidative stress-induced toxicity, hence becoming the main targets of neuroprotective therapy [72][73][74][75][76]. Similarly, we found in this study that the increase in antioxidant activities were accompanied by improvement of MMP and reduction of mitochondrial toxicity. This was reflected by the increase in ATP content, an indicator of mitochondrial health, and subsequent nuclear apoptosis and reduction in cellular viability. Just recently, the polysaccharide from H. erinaceus mycelium was found to protect ulcerative colitis-induced rats and Caco-2 cells by counteracting oxidative stress and improving mitochondrial function [47]. A study using the hot water extract reported its significant neuroprotective properties in glutamate-induced toxicity in PC12 cells. The extract was found to prevent nuclear apoptosis by suppression of intracellular ROS accumulation and MMP depolarization [23]. On the other hand, erinacine A-enriched ethanolic extract from H. erinaceus mycelium was shown to exert prominent anti-apoptotic activity via ROS-caspase dependent pathway against glutamate-insult in PC-12 cells [22]. Nevertheless, all those studies above mentioned mycelium as the source for extraction rather than basidiocarps of the mushroom, which were used in our study.
Under normal circumstances, Nrf2 is localized in the cytoplasm and bond with its inhibitor, Kelch-like ECH associated protein 1 (Keap1). Once stimulated with oxidative stress, Nrf2 translocate into the nucleus to binds with the anti-oxidant response elements (AREs) and regulate the expression of antioxidant genes, primarily HO-1 and NQO1 [77,78]. HO-1 is an inducible enzyme that catalyzes the degradation of heme into carbon monoxide (CO), biliverdin, and free iron, which exert anti-oxidative functions via various mechanisms including reduction of ROS production [79][80][81]. Meanwhile, NQO1 functions as quinone reductase by reducing quinones to hydroquinones, hence preventing the one electron reduction of quinones that results in the production of radical species [82,83]. The Nrf2-ARE signaling pathway is actively involved in various CNS diseases including ischemic stroke [84], traumatic brain injury [85], Parkinson's disease (PD) [86], and Alzheimer's disease [87]. In this study, we witnessed non-significant enhancement of Nrf2, NQO1, and HO-1 transcriptional expression by HE-ETH. However, it cannot be concluded that Nrf2-ARE pathway was not involved, because CAT and GSH, which are also by-products of Nrf2/ARE activation [88], were significantly enhanced by HE-ETH. Therefore, further studies need to be conducted to comprehensively elucidate the involvement of Nrf2/ARE pathway by suitable approaches, such as overexpression of Keap1 or transient transfection of Nrf2-small interfering RNA (siRNA) [89]. Since HE-ETH and its constituents were found to induce NGF synthesis and upregulation of MAPK and Akt pathways [30] which were proven elsewhere to be neuroprotective [90][91][92][93], it is worthy to explore whether those pathways are also responsible for the activities of HE-ETH in this study.

Conclusions
HE-ETH compared to HE-HWA exerted potent neuroprotection and higher NO-suppressing anti-inflammatory activity in HT22 hippocampal neurons and BV2 microglia in vitro. The mechanisms of neuroprotection were through the improvement of antioxidant, mitochondrial function and anti-apoptosis. HE-ETH could be potentially explored as a neuroprotective agent in neuron-glia environment. However, further studies are needed to elucidate the mechanistic pathways involved in the neuroprotection and to study its efficacy in vivo.
Author Contributions: N.K. designed and carried out the experiments as well as data analysis. C.W.P., V.S., P.D. and M.N. approved the analyzed data and figures. N.K. wrote the paper with input and correction from C.W.P., V.S., P.D. and M.N. All authors approved the final manuscript.

Conflicts of Interest:
The authors declare no conflict of interest.