Vanillic acid attenuates Aβ1-42-induced oxidative stress and cognitive impairment in mice

Increasing evidence demonstrates that β-amyloid (Aβ) elicits oxidative stress, which contributes to the pathogenesis and disease progression of Alzheimer’s disease (AD). The aims of the present study were to determine and explore the antioxidant nature and potential mechanism of vanillic acid (VA) in Aβ1-42-induced oxidative stress and neuroinflammation mediated cognitive impairment in mice. An intracerebroventricular (i.c.v.) injection of Aβ1-42 into the mouse brain triggered increased reactive oxygen species (ROS) levels, neuroinflammation, synaptic deficits, memory impairment, and neurodegeneration. In contrast, the i.p. (intraperitoneal) administration of VA (30 mg/kg, for 3 weeks) after Aβ1-42-injection enhanced glutathione levels (GSH) and abrogated ROS generation accompanied by an induction of the endogenous nuclear factor erythroid 2-related factor 2 (Nrf2) and heme oxygenase 1 (HO-1) via the activation of Akt and glycogen synthase kinase 3β (GSK-3β) in the brain mice. Additionally, VA treatment decreased Aβ1-42-induced neuronal apoptosis and neuroinflammation and improved synaptic and cognitive deficits. Moreover, VA was nontoxic to HT22 cells and increased cell viability after Aβ1-42 exposure. To our knowledge, this study is the first to reveal the neuroprotective effect of VA against Aβ1-42-induced neurotoxicity. Our findings demonstrate that VA could potentially serve as a novel, promising, and accessible neuroprotective agent against progressive neurodegenerative diseases such as AD.


Vanillic acid supplementation attenuated Aβ accumulation and BACE-1 (β-site APP-cleaving enzyme-1) expression. Antioxidants have been reported to inhibit Aβ production
. To determine whether the i.c.v. administration of Aβ  promoted Aβ accumulation in the brain, we performed immunoblot and immunofluorescence analyses. Immunoblot results showed higher levels (1.5-fold) of Aβ in the Aβ 1-42 -treated mice than that in the control mice. Vanillic acid administration results in an increase accumulation of VA in the brain of mice (Suppl. Fig. 1) which attenuated the effects of Aβ 1-42 and significantly reduced (1.4-fold) the levels of Aβ compared to the Aβ 1-42 only mice ( Fig. 2A). Similarly, BACE-1 expression was examined after Aβ 1-42 injection, and the immunoblot results showed that Aβ 1-42 treatment significantly increased (1.4-fold) BACE-1 expression compared to the control mice, and VA treatment significantly reduced (1.2-fold) BACE-1 expression in the Aβ 1-42 -treated mice in comparison to Aβ 1-42 -injected mice without VA treatment ( Fig. 2A).
The immunofluorescence images also supported our western blot results that Aβ 1-42 -treated mice showed increased Aβ immunofluorescence reactivity in the cortex (8.5-fold) and hippocampal DG regions (6.6-fold) compared to the control group. In contrast, VA supplementation in combination with Aβ 1-42 significantly reduced (1.6-and 1.7-fold respectively) Aβ immunofluorescence reactivity in the aforementioned regions of the mice (Fig. 2B).
Vanillic acid abrogates Aβ 1-42 -induced oxidative stress in the mice brain via the Akt/GSK3β/Nrf2 signaling pathway. Oxidative stress is implicated in various neurodegenerative diseases, including AD 28 .
Our results indicated that, VA treatment markedly enhanced (1.2-and 1.3-fold respectively) GSH and GSH/ GSSG levels in the brain homogenates of Aβ 1-42 -treated mice (Fig. 3A,B). To examine the beneficial effects of VA against Aβ 1-42 -induced oxidative stress in the mouse brain, an ROS assay was performed on the brain homogenates of all treated groups. The results showed that Aβ 1-42 induced oxidative stress by significantly increasing (3.8-fold) ROS levels compared to the vehicle-treated mice, whereas VA administration in combination with Aβ 1-42 significantly reduced (1.6-fold) ROS levels in mice (Fig. 3C). Similarly, VA treatment significantly decreased the immunoreactivity of 8Oxo-G (2-fold) and lipid peroxidation (LPO) levels (1.5-fold) in the mice hippocampus against Aβ 1-42 injected group as evident in the Fig. 3D,E respectively.
Previous studies have demonstrated that HO-1 acts as a cellular defense mechanism in protection against ROS attack 29 . The nuclear translocation of Nrf2 and the expression of its target gene products, including HO-1, elicited an antioxidant response that may have therapeutic value for the treatment of AD 30 . Therefore, we assessed the effects of VA on the activation of Nrf2/HO-1 in the brains of Aβ 1-42 -treated mice. The western blot analyses reveal a decreased expression (1.5-fold) of Nrf2 (in the nucleus) and HO-1 (2.6-fold) proteins in Aβ 1-42 -treated mice, whereas VA treatment (30 mg/kg for 3 weeks) significantly increased (1.7-fold) the expression of Nrf2 (in the nucleus) and HO-1 (2.8-fold) in the brains of Aβ 1-42 -treated mice. (Fig. 3F).
Vanillic acid treatment attenuated Aβ 1-42 -induced glial cell activation (microglia and astrocytes) and neuroinflammation in the mouse brain. Several studies have indicated that there is an increase in microgliosis and astrocytosis in old age that can contribute to neurological disorders such as AD 33,34 . Glial fibrillary acidic protein (GFAP) and ionized calcium-binding adaptor molecule 1 (Iba-1) are specific markers for activated astrocytes and microglia, respectively. Therefore, we investigated the protective effect of VA against microglial (Iba-1 reactive cells) and astrocyte (GFAP reactive cells) activation. Immunofluorescence images in cortex and hippocampus revealed a significant increase in the number of Iba-1 and GFAP (7.9-and 6.8-fold respectively) reactive cells in the brains of mice in the Aβ 1-42 -treated group compared to vehicle-treated The double immunofluorescence images of HT22 cells after Aβ 1-42 and VA treatment for 24 h, showing p-NF-kB (green), p-Akt (red), proteins and their respective relative density histograms. DAPI (blue) was used to counterstain the nucleus. These experiments were performed in triplicate. Details are given in the methods section. *Significantly different from vehicle-treated animals; # significantly different from Aβ 1-42 -treated animals. Significance = **P < 0.01, # P < 0.05, ## P < 0.01.
NF-κ B expression has been reported to increase during aging 35 . Similarly Terai et al. reported that NF-κ B has been found in neurons and neurofibrillary tangles in the brain of patient with AD after postmortem 36 . Western blot results revealed that the expression of p-NF-κ B and phospho-IKK-β was increased (1.3-and 1.2-fold respectively) in the brains of Aβ 1-42 -treated mice compared to mice in the vehicle-treated group. Vanillic acid treatment (30 mg/kg for 3 weeks) significantly reduced the expression of p-NF-κ B (1.1-fold) and p-IKK-β (1.4-fold) in Aβ 1-42 -treated mice (Fig. 4C). Similarly, NF-κ B activation could lead to the activation of various pro-inflammatory markers that are implicated in neuronal degeneration 37 . The levels of activated inflammatory markers such as inducible nitric oxide synthase (iNOS) were analyzed in Aβ 1-42 -treated brains via western blot analysis. The results revealed that VA significantly reduced (1.3-fold) Aβ 1-42 -induced neuroinflammation by inhibiting iNOS expression in the mouse brain (Fig. 4C).
Vanillic acid rescued the mouse brain against Aβ 1-42 -induced neuronal apoptosis and neurodegeneration. Pro-apoptotic Bax and anti-apoptotic Bcl-2 are members of the Bcl-2 family and are the main regulators of the apoptotic pathway in mitochondria 38 . Our immunoblot results indicate that Aβ 1-42 caused the Immunoblot analysis of Aβ and BACE-1 protein expression in the mouse brain following Aβ and VA administration. The bands were quantified using Sigma Gel software, and the differences are represented by a histogram. β -Actin was used as a loading control. The density values are expressed in arbitrary units (A.U.) as the mean ± S.E.M. for the indicated protein (n ± 5 mice/group). (B) The immunofluorescence of Aβ was used to evaluate the cortex and hippocampus of experimental mice (n ± 8 mice/ group). Magnification, 10X. *Significantly different from vehicle-treated animals; # significantly different from Aβ 1-42 -treated animals. Significance ± **P < 0.01; ## P < 0.01. Previous reports have determined that Aβ 1-42 plays a critical role in inducing apoptotic neurodegeneration in AD 39 . Aβ -induced neuronal apoptosis is mediated via the apoptotic caspase cascade, which includes caspase-9 and caspase-3 40 . We investigated the levels of caspase-3 in response to Aβ 1-42 and VA treatment via western blot and immunofluorescence analysis. Our results (both western blot and immunofluorescence, Fig. 5A,B) showed higher (1.3-and 3.9-fold respectively) levels of activated caspase-3 in Aβ 1-42 -treated mice compared to the control group. Treatment with VA attenuated the Aβ 1-42 -induced expression of active caspase-3 and significantly decreased (1.3-and 2.2-fold respectively) the level of caspase-3 compared to the mice treated with Aβ 1-42 alone (Fig. 5A,B).
Poly (ADP-ribose) polymerase-1 (PARP-1) is involved in DNA repair, and the overexpression of PARP-1 due to the exposure to different excitotoxic agents induces neurodegeneration 41 . Aβ 1-42 -peptide has been demonstrated to trigger the overactivation of PARP-1 in the adult rat hippocampus 42 . Western blot analysis revealed that PARP-1 cleavage increased (1.5-fold) in the Aβ 1-42 -treated mice, whereas the expression level of cleaved PARP-1

Figure 4. Vanillic acid treatment attenuated the number of Aβ 1-42 -induced activated glial cells (microglia and astrocytes) and reduced neuroinflammation in the mouse brain (n ± 8 mice/group). (A)
Iba-1 immunofluorescence revealed a significant increase in the number of Iba-1 reactive cells in the brains of mice in the Aβ 1-42 -treated group compared to mice in the vehicle-treated group. On the other hand, VA treatment significantly decreased the number of reactive Iba-1 cells in the brains of mice exposed to Aβ 1-42 . (B) Immunofluorescence images revealed a significant increase in the number of GFAP reactive cells in the brains of mice in the Aβ 1-42 -treated group compared to the vehicle-treated group. Vanillic acid treatment significantly decreased the number of reactive GFAP cells in the brains of mice exposed to Aβ 1-42 . (C) Western blot analysis of p-IKKβ , p-NF-κ B and iNOS in the brain of mice. The bands were quantified using Sigma Gel software, and the differences are represented in a histogram. β -Actin was used as a loading control. The density values are expressed in arbitrary units (A.U.) as the mean ± SEM for the indicated proteins (n ± 5 mice/ group). *Significantly different from vehicle-treated mice; # significantly different from Aβ 1-42 -treated mice. Significance = **P < 0.01, ## P < 0.01. was significantly reduced (1.3-fold) in VA treated mice (Fig. 5A). Furthermore, the FJB results indicate that Aβ 1-42 treatment significantly increased (3.8-fold) the number of dead neurons in contrast to untreated mice group (Fig. 5C). While treatment of VA significantly reduced (2.3-fold) the number of FJB positive neurons in mice hippocampus, demonstrating a protective effect of VA against Aβ 1-42 -induced toxicity (Fig. 5C).

Vanillic acid treatment alleviated Aβ 1-42 -induced synaptotoxicity.
To analyze the protective effect of VA against Aβ 1-42 , we assessed pre-and post-synaptic protein markers. Immunoblot results showed a lower level (1.6-and 2.8-fold respectively) of the presynaptic vesicle membrane protein synaptophysin (SYP) and the post-synapse density (PSD95) protein in Aβ 1-42 -treated mice compared to the control group. Vanillic acid administration reversed the synaptotoxic effects of Aβ 1-42 and significantly increased the expression of the SYP (1.2-fold) and PSD95 (2.1-fold) proteins in comparison to Aβ 1-42 treated mice (Fig. 6A). Additionally, immunofluorescence results showed that Aβ 1-42 -treatment decreased (4.3-fold) the immunofluorescence reactivity of PSD95 in cortex and hippocampus compared to the control mice. Vanillic acid treatment reversed the effects of Aβ 1-42 and significantly increased (4.1-fold) the immunofluorescence reactivity of PSD95 in the cortex and the hippocampal CA1 and DG regions (Fig. 6B).

Vanillic acid treatment ameliorates Aβ 1-42 -induced memory impairment. Performance in the
Morris water maze has been shown to be a reliable and noninvasive test to determine cognitive changes in the AD mouse model 43 . To assess whether VA could counteract Aβ 1-42 -induced memory impairment, we administered the MWM and Y-maze tests. We first recorded the learning ability of the mice (n = 13 mice/group) in the MWM test. We observed that Aβ 1-42 -treated mice showed an increased latency to reach the platform, and the mice that had received VA treatment (30 mg/kg, i.p., 3 weeks) showed a decreased escape latency (Fig. 6C). Twenty-four hours after the 5 day training session, we removed the platform and allowed the mice to swim freely. We observed that Aβ 1-42 -treated mice spent less time in the target quadrant and exhibited fewer platform crossings, revealing that Aβ 1-42 caused memory impairment. Vanillic acid treatment improved Aβ 1-42 -induced memory impairment by significantly increasing (1.5-and 2.1-fold respectively) the time spent in the target quadrant and the number of platform crossings (Fig. 6D,E). Following the MWM analysis, we evaluated the spontaneous alteration behavior percentage (%) of mice (n ± 13 mice/group), observing the average total number of arm entries and successive triplets using a Y-maze test. Spontaneous alteration behavior, indicating spatial working memory, is a form of short-term memory. After the single injection of Aβ 1-42 , the % of spontaneous alternation behavior was lower (1.8-fold) in Aβ 1-42 -treated mice compared to the control mice, suggesting that Aβ 1-42 was responsible for the decline in cognition. Treatment with VA significantly increased (1.4-fold) the spontaneous alternation behavior % in Aβ 1-42 -treated mice compared to mice that had received Aβ 1-42 alone (Fig. 6F), indicating that VA treatment ameliorated Aβ 1-42 -induced memory dysfunction in Aβ 1-42 -treated mice.

Discussion
Oxidative stress has been shown to contribute to AD neuropathology 44 . Increased levels of oxidative stress markers were found in neurons surrounding amyloid deposits in transgenic mouse models of the disease 45 , and the experimental induction of oxidative stress leads to Aβ accumulation in primary neurons 46 . ROS can oxidize proteins, lipids, and DNA, and increased levels of specific oxidative markers and redox metals were found in the brains of AD patients 47 . Senile plaque formation in specific regions of the brain induces neuroinflammation and free radical induction that contribute to the destruction of brain areas such as the amygdala, hippocampus, and cortex 48 . Reducing oxidative damage in the brain can be considered a promising strategy for therapeutic intervention in AD 49 . The antioxidative activity of vanillic acid is well known from in vitro experiments 50,51 .
The present study is the first to provide evidence that VA administration (30 mg/kg, i.p., 3 weeks) attenuates Aβ 1-42 -induced ROS, memory impairment, synaptic deficits, neuroinflammation and neurodegeneration in a mouse Aβ 1-42 model. So far, the possible effects of VA have not been studied in an AD model that exhibits amyloidosis and oxidative stress. In the present study, we found that VA significantly ameliorates cognitive deficits accompanied by increased levels of GSH in brain tissues and increased Nrf2/HO-1 expression in Aβ 1-42 -treated mice. Vanillic acid exerts beneficial therapeutic effects via positively regulating Akt/GSK-3β /Nrf2 signaling pathways. Furthermore, we also determined that VA is beneficial against Aβ 1-42 -induced neurotoxicity in the neuronal HT22 cell line in vitro.
The administration of Aβ in mice triggered oxidative stress by increasing ROS level, memory dysfunction, synaptic disorganization (a key feature of early phase AD), neuroinflammation and potentially neuronal degeneration. In the advancement and evaluation of therapeutic strategies for AD pathology, the i.c.v. Aβ 1-42 -infusion model is a useful complement to transgenic mouse models 52 , although the mechanisms that underlie many features of AD, including synaptotoxicity, the hyperphosphorylation of tau, apoptosis, and neurodegeneration are still not clearly known. In AD, Aβ production and aggregation in the human brain has been associated with neuronal dysfunction and memory disorders 53 . BACE-1 is the primary initiating enzyme, and its activity is the rate-limiting step in APP processing and Aβ production 54 . The elevated expression of activated BACE-1 has been examined in the brain during late-onset sporadic AD, which is associated with neuronal loss and spatial memory impairment in 5XFAD APP/PS1 mice 55 . Aβ levels and Aβ 1-42 -induced BACE-1 expression in the Aβ 1-42 -treated mice were alleviated with VA treatment (Fig. 2A,B).
Previous studies involving in vivo and in vitro experiments showed that Aβ increases oxidative damage 56 . Our data in the present study showed that Aβ 1-42 -treated mice exhibited a significant increase in oxidative stress compared to WT mice. Our observations are consistent with other prior reports 57 . Vanillic acid exhibited protective effects due to its free radical scavenging, antioxidant and anti-inflammatory effects 58 . Interestingly, in this work, we found that the enhancement of oxidative stress in the brain of Aβ 1-42 -treated mice was significantly attenuated by VA treatment, which reduced ROS induction and prevented the depletion of endogenous reduced glutathione (GSH) levels, suggesting that antioxidant activity might play some role in the beneficial effects of vanillic acid in Aβ 1-42 -treated mice. Increased levels of reduced glutathione revealed that there was less radical formation, and consequently less oxidized glutathione was formed. This finding clearly revealed the antioxidant effects of VA.
A growing body of literature suggests that the activation of Nrf2 provides neuroprotection in AD 59 . The Nrf2 antioxidant pathway was impaired in transgenic AD mice concomitantly with an increased brain Aβ burden 60 . Previous work by Choudhry showed a 50% reduction in Nrf2 levels in transgenic AD mice 61 . The induction of the Nrf2 pathway by small-molecule compounds protects against neuronal oxidative stress and toxicity induced by Aβ in vitro 62 . A decrease in Nrf2 protein expression was observed in the brains of Aβ 1-42 -treated mice, and VA treatment significantly increased Nrf2 protein expression. Although the nuclear translocation of Nrf2 is responsible for the induction of HO-1 expression, it is uncertain whether the VA-induced enhancement in HO-1 expression contributes to the improved cognitive functions in Aβ 1-42 mice. HO-1 is thought to be highly associated with AD pathology and is expressed in the hippocampus of patients with AD 63 . The upregulation of HO-1 has therapeutic potential for antioxidant function in AD 64 . Although HO-1 expression is known to correlate with oxidative stress, it is uncertain whether increased HO-1 levels are associated with the improvement of cognitive functioning resulting from antioxidant treatment. We observed that Aβ 1-42 -treated mice exhibited decreased HO-1 expression. Treatment with VA increased the expression of HO-1 in the brains of Aβ 1-42 -treated mice.
GSK-3β negatively regulates Nrf2 by controlling its subcellular distribution 65 . Prolonged oxidative stress, as in cases of AD, causes the inactivation of Akt, the activation of GSK-3β and the translocation of Nrf2 from the nucleus to the cytosol, thus limiting the antioxidant response of cells 31,32 . In agreement with previous studies on AD patients and AD mouse models, the current study shows that GSK-3β expression was decreased in the brains of Aβ 1-42 -treated mice, and VA treatment increased the expression of Akt and GSK-3β . Although the causal relationship remains unclear, it is conceivable that improved spatial learning and memory could be attributed to the VA -induced activation of the Akt pathway.
Maqbool et al. demonstrated that activated IKK-β /NF-κ B-induced neuroinflammation promotes neurodegeneration during aging. In this study, we investigated whether Aβ 1-42 increased the expression of IKK-β /NF-κ B, which can trigger other inflammatory mediators 66 . Vanillic acid treatment reversed the Aβ 1-42 -induced elevated expression of IKK-β /NF-κ B and reduced the activity of IKKβ /NF-κ B. Studies have reported that activated NF-κ B increased the expression of other inflammatory mediators such as iNOS2, which might be involved in the induction of memory impairment. Previous studies have described the anti-inflammatory properties of phenolic acids, VA and protocatechuic acids 67 .
We also found that the Aβ 1-42 -induced expression of these inflammatory markers might occur through the activation of the IKK-β /NF-κ B pathway or astrocytic and microglial activation. Treatment with VA decreased the expression of these inflammatory markers, preventing neuroinflammation and memory impairment in the Aβ 1-42 -treated mice.
Numerous mechanisms have been associated with Aβ induced apoptosis and neurodegeneration in both in vivo and in vitro models of AD 39,68 . Our results also revealed that Aβ 1-42 activates caspases. Activated caspase-3 cleaves PARP-1, leading to apoptosis and neurodegeneration 69 . The overactivation of PARP-1 is involved in NAD+ depletion, leading to neuronal cell death 70 . Vanillic acid exerted protective effects on lipids, Bax, Bcl-2 and myocardial infarct size in isoproterenol-induced rats 58 . Our results also showed that VA decreased the expression of activated caspases, which prevented PARP-1 cleavage and reduced the expression of Bax, indicating that VA prevents Aβ 1-42 -induced apoptotic neuronal cell death in Aβ 1-42 -treated mice.
Landmark studies have described Aβ -induced synaptic loss and disorganization in an animal model of AD 71 . However, the underlying mechanism of synaptic loss and disorders is still unknown. The expression of the presynaptic marker synaptophysin was decreased in the brains of patients with AD and in an Aβ animal model of AD 72 . Our results revealed that Aβ 1-42 -treated mice exhibited significantly lower synaptophysin levels in the brain. Moreover, the decreased level of the postsynaptic protein marker PSD95 was also observed in the Aβ model of AD 71 . It has been reported that PSD95 and SNAP-23 regulate AMPARs 73 . Thus, decreased levels of synaptophysin and PSD95 are associated with memory dysfunction; in Aβ 1-42 -treated mice, the levels of synaptophysin and PSD95 were enhanced after VA administration, suggesting that the prevention of synaptic disorganization through various pre-and postsynaptic (LTP)-related protein markers improve memory function.
In the current study, our results showed a significant reduction in memory function as evidenced by MWM and Y-Maze test performance. We also observed that VA treatment (30 mg/kg, i.p., 3 weeks) improved memory, as shown by the reduction in escape latency, the increased amount of time spent in the target quadrant and the number of platform crossings during the probe test. In the Y-maze, we observed a lower percentage of spontaneous alternation behavior, which is related to the function of the hippocampus 74 . Vanillic acid treatment ameliorated the effects of Aβ 1-42 on spontaneous alternation behavior and reduced the degree of spatial memory impairment. The observed improvement in memory function associated with VA treatment demonstrates its neuroprotective effect against Aβ 1-42 -induced memory impairment.

Conclusion
In summary, our results demonstrated that the reversal of cognitive deficits by VA treatment in Aβ 1-42 -treated mice might result from the antioxidant activity of VA; vanillic acid treatment was associated with an increased expression of HO-1, which is mediated by the activation of Akt/GSK-3β /Nrf2 signaling. This unique mechanism explains, at least partially, the potent antioxidant capacity of VA, which might allow VA to succeed in treating AD where other 'regular' antioxidants have failed. In addition, our results do not exclude the possible involvement of any other mechanisms in the inhibition of oxidative stress by VA. Therefore, VA could be a potential candidate for further preclinical studies aimed at the treatment of cognitive impairment and dementia.

Materials and Methods
Chemicals. Vanillic

Oxidative stress (ROS) detection in vitro.
The ROS assay was conducted in HT22 cells as described previously 75 . The cells were cultured in 96-well plates. After 24 h of incubation at 37 °C in a humidified atmosphere of 5% CO 2 , the cells were treated with fresh medium containing Animals. Male wild-type C57BL/6 N mice (25-30 g, 8 wks old, n ± 13 mice/group) were purchased from Samtako Bio (Osan, South Korea). The mice were acclimatized for 1 week in the university animal house under a 12-h/12-h light/dark cycle at 23 °C with 60 ± 10% humidity and were provided with food and water ad libitum. All of the methods and experimental procedures were conducted according to the approved (Approval ID: 125) guidelines and regulations by the animal ethics committee (IACUC) of the Division of Applied Life Sciences, Department of Biology at Gyeongsang National University, South Korea. Drug treatment protocol. Human Aβ 1-42 peptide was prepared as a stock solution at a concentration of 1 mg/mL in sterile saline solution, followed by aggregation via incubation at 37 °C for 4 days. The aggregated Aβ 1-42 peptide or vehicle (0.9% NaCl, 3 μ L/ 5 min/mouse) was stereotaxically administered into the ventricle (i.c.v.) using a Hamilton microsyringe (− 0.2 mm anterioposterior (AP), 1 mm mediolateral (ML), and − 2.4 mm dorsoventral (DV) to Bregma) under anesthesia in combination with 0.05 mL/100 g body weight Rompun (Xylazine) and 0.1 mL/100 g body weight Zolitil (Ketamine). The rest of the protocol was the same as previously reported 76 .

Morris water maze (MWM) test.
Behavior was assessed using a MWM and a Y-maze test (n ± 13 mice/ group). The experimental apparatus consisted of a circular water tank (100 cm in diameter, 40 cm in height), containing water (23 ± 1 °C) at a depth of 15.5 cm, which was made opaque by adding white ink. A transparent escape platform (10 cm in diameter, 20 cm in height) was hidden 1 cm below the water surface and was placed at the midpoint of one quadrant. Each mouse received training for five consecutive days using a single hidden platform in one quadrant with three quadrants of rotational starting. Latency to escape from the water maze (finding the hidden escape platform) was calculated for each trial. Twenty-four hours after the 5th day, the probe test was performed for the evaluation of memory consolidation. The probe test was carried out by removing the platform and allowing each mouse to swim freely for 60 s. The time that mice spent in the target quadrant and the number of times the mouse crossed over the platform location (where the platform was located during hidden platform training) was measured. Time spent in the target quadrant was considered to represent the degree of memory consolidation. All data were recorded using video-tracking software (SMART, Panlab Harvard Apparatus; Bioscience Company, Holliston, MA, USA).
Y-maze test. The Y-maze was made from black-painted wood. Each arm of the maze was 50 cm long, 20 cm high and 10 cm wide at the bottom, and 10 cm wide at the top. Each mouse was placed at the center of the apparatus and was allowed to move freely through the maze for three 8-min sessions. The series of arm entries was visually observed. Spontaneous alteration was defined as the successive entry of the mice into the three arms in overlapping triplet sets. The alteration behavior percentage (%) was calculated as [successive triplet sets (entries into three different arms consecutively)/total number of arm entries-2] × 100. A higher percentage of spontaneous alternation behavior was considered to indicate enhanced cognitive performance.
Protein extraction from mouse brains. The mice were killed and the brains (hippocampus) were immediately removed, and the tissue was frozen on dry ice and stored at − 80 °C. The brains tissues were homogenized in pro-prep TM protein extraction solution according to the provider instructions (iNtRON Biotechnology, Inc., Sungnam, South Korea). The samples were then centrifuged at 13,000 r.p.m. at 4 °C for 25 min. The supernatants were collected and stored at − 80 °C.
Western blot analysis. Western blot analysis was performed according to previously reported methods with minor modifications 77 . Briefly, the animals (n ± 5 mice/group) were sacrificed and the protein concentration was measured (Bio-Rad protein assay kit, Bio-Rad Laboratories, CA, USA). Equal amounts of proteins (25-30 μ g) were subjected to electrophoresis on 4-12% Bolt ™ Mini Gels (Novex; Life Technologies, Kiryat Shmona, Israel).
The membranes were blocked in 5% (w/v) skim milk to reduce the nonspecific binding and were then incubated with the primary antibodies (1:1000 dilution) overnight at 4 °C. After incubating with a horseradish peroxidase-conjugated secondary antibody, the specific immunocomplexes were detected using an ECL detection reagent according to the manufacturer's instructions (Amersham Pharmacia Biotech, Uppsala, Sweden). The X-ray films were scanned, and the optical densities of the bands were analyzed by densitometry using the computer-based Sigma Gel program, version 1.0 (SPSS Inc., Chicago, IL, USA).
ROS assay in vivo. ROS activity was assessed as described previously with some modification 78 . The assay was based on the oxidation of 2′ 7′ -dichlorodihydrofluorescein diacetate (DCFH-DA) to 2′ 7′ dichlorofluorescein (DCF). Brain (hippocampus) homogenates were diluted with ice-cold Lock's buffer at 1:20 to yield the final concentration of 2.5 mg tissue/500 μ L. The reaction mixture of Lock's buffer (1 mL, pH ± 7.4), 0.2 mL of homogenate, and 10 mL of DCFH-DA (5 mM) was incubated at room temperature for 15 min to convert DCFHDA to the fluorescent product DCF. The conversion of DCFH-DA to the DCF was assessed using a spectrofluorimeter with excitation at 484 nm Scientific RepoRts | 7:40753 | DOI: 10.1038/srep40753 and emission at 530 nm. For background fluorescence (conversion of DCFH-DA in the absence of homogenate), we measured parallel blanks. ROS quantification was expressed as pmol DCF formed/mg protein.
GSH and GSH/GSSG assay. The levels of GSH and GSH/GSSG in the brain tissue homogenates of hippocampus region were determined by using the commercially available Glutathione Assay Kit (BioVision's Catalog #K264-100) according to the provided protocol. This BioVision's Glutathione Detection Kit provides a unique, convenient tool for detecting GSH, GSSG, and total glutathione separately. Briefly, in the assay, OPA (o-phthalaldehyde), reacts with GSH (not GSSG), generating fluorescence, so GSH can be specifically quantified. Adding a reducing agent converts GSSG to GSH, so (GSH + GSSG) can be determined. To measure GSSG specifically, a GSH Quencher is added to remove GSH, preventing reaction with OPA (while GSSG is unaffected). Reducing agent is then added to destroy excess quencher and convert GSSG to GSH. Thus, GSSG can be specifically quantified.
Determination of Lipid Peroxidation. Quantification of lipid peroxidation (LPO) is essential to assess the oxidative stress. Free malondialdehyde (MDA), a marker of LPO, was measured in the tissue homogenate of the hippocampus region using lipid peroxidation (MDA) colorimetric/fluorometric assay kit (BioVision, USA, Cat# K739-100) according to the manufacturer's protocol.
HPLC Analysis of vanillic acid Levels in the Brain. Quantitative analysis of vanillic acid was performed by using HPLC (Perkin-Elmer 200 series, Perkin-Elmer Co., Bridgeport, USA). Separation was achieved using Zorbax bonus RP C-18 column (4.6 × 150 mm) 5 lM, Agilent, USA, at room temperature of 25 °C. The mobile phase consisted of (A) water having 0.1% analytical grade acetic acid, and (B) 100% acetonitrile. Elution was carried out in a binary gradient mode in ratio of [A:B 80:20 (5 min), 50:50 (5 mint), 30:70 (6 mint)] having flow rate 1 ml/min. The whole procedure was run for 16 mints, while chromatograms were acquired at 260 nm in UV detector. Stock solution of vanillic acid was prepared in acetonitrile at 1.0 mg/mL and stored at − 20 °C. From this stock solution, 50 μ g/mL working stocks and subsequent working solutions of appropriate concentrations were prepared in ACN. From the working solutions, calibration samples were prepared. After that Calibration curve was constructed using the following concentrations: 6.25, 12.5, 25, and 50 μ g/mL and their respective detected peak area of each concentrations, which were fitted using least squares linear regression correlation analysis. The areas of individual peaks were calculated from extracted LC chromatograms of the given compound. The plotted four point's calibration curves were linear having correlation coefficients higher than 0.999 for the said compound, indicating a good linearity in the proposed investigation range.
In order to determine the levels of vanillic acid in the brain, at 1 h after the last i.p. injection, the brains were rapidly removed, weighed, and washed in ice cold 0.9% NaCl. The brain was minced with scissors and placed in a homogenizer vessel; 5.5 mL acetonitrile was added and tissue was subsequently homogenized. The homogenized samples were transferred to 50 mL conical glass tubes and vortexed for 5 min prior to centrifugation at 2,800 × g for 30 min at 4 °C. The supernatant was placed into a clean tube, filtered (Millipore ® 0.45 μ m) and placed in a sealed vial for HPLC analysis. The injection volume used was 20 μ L for all brain samples. The quantity of vanillic acid was calculated by comparing the peak area ratio from tissue samples of treated animals with those of the corresponding concentration standards of vanillic acid in acetonitrile injected directly into the HPLC system. Tissue collection and sample preparation. For tissue analysis (n ± 8 mice/group), mice were perfused transcardially with 4% ice-cold paraformaldehyde, and the brains were postfixed for 72 h in 4% paraformaldehyde and transferred to 20% sucrose for 72 hr. The brains were frozen in O.C.T. compound (A.O., USA), and 14-μ m coronal sections were cut using a CM 3050 C cryostat (Leica, Germany). The sections were thaw-mounted on probe-on plus charged slides (Fisher, Rockford, IL, USA). Immunofluorescence staining. Immunofluorescence staining was performed according to previously reported methods with minor modifications 77 . Briefly, slides containing either HT22 cells or the brain tissues sections (cortex and hippocampus regions of mice) were washed twice for 10 min each in 0.01 M PBS and incubated for 1 h in blocking solution containing 2% normal bovine serum (Santa Cruz Biotechnology), according to the antibody treatment, and 0.3% Triton X-100 in PBS. After blocking, the slides were incubated overnight at 4 °C with anti-p-NF-KBp65, anti-p-Akt, anti-8-Oxo-dG, anti-Aβ , anti-PSD95, anti-Iba-1 and anti-GFAP antibodies diluted 1:100 in blocking solution. Following this step, the sections were incubated for 2 h with fluorescein isothiocyanate (FITC)-labeled (green) or TRITC labeled (red) secondary antibodies (1:50). The slides were then counterstained with 40,6-diamidino-2-phenylindole (DAPI) for 10 min and mounted with the Prolong Anti-Fade Reagent (Molecular Probes, Eugene, OR, USA). Staining patterns were examined using a confocal laser-scanning microscope (FlouviewFV 1000) and were evaluated by an examiner blind to the treatment groups.
Fluoro-Jade B staining. Fluoro-Jade B staining was performed according to the manufacturer's protocol (Millipore, USA, Cat. #AG310, Lot #2159662) and as previously reported 79 . After air drying the specimens overnight, the slides were immersed for 5 min in a solution containing 1% sodium hydroxide and 80% ethanol. Following this, the slides were immersed in 70% alcohol and distilled water for 2 min each. The specimens were then transferred into a solution of 0.06% potassium permanganate for 10 min, rinsed with distilled water and immersed in a solution of 0.1% acetic acid and 0.01% Fluoro-Jade B for 20 min. These slides were then washed with distilled water and were allowed to dry for 10 min. The glass cover slips were mounted using DPX nonfluorescent mounting medium, and the images were acquired using a confocal laser scanning microscope (FV 1000, Olympus, Japan). Statistical analysis. The western blots were scanned and analyzed via densitometry using the Sigma Gel System (SPSS Inc.). The density values are expressed as the mean ± standard error of the mean (S.E.M.). The Image-J software was used for quantitative immunohistological analysis. A one-way analysis of variance (ANOVA) followed by a two-tailed independent Student's t-test and Tukey's multiple comparison test were used for comparisons among the treatment and control groups. The calculations and graphs were generated with Prism 5 software (GraphPad Software, In., San Diego, CA, USA). P values < 0.05 were considered to be statistically significant: # significantly different from the vehicle treated control group, *significantly different from Aβ 1-42 -treated groups. *P < 0.05, and **P < 0.01; # P < 0.05, and ## P < 0.01.