p-Coumaric acid protects against D-galactose induced neurotoxicity by attenuating neuroinflammation and apoptosis in mice brain

D-galactose (D-gal) induced senescence in rodents is a widely used model for assessment of molecules affecting brain ageing. Chronic administration of D-gal causes neuroinflammation leading to cognitive deficit and memory impairment which represent Alzheimer’s dementia. In present study, we investigated the neuroprotective effects of the natural phenol, p-Coumaric acid (PCA) and its underlying mechanism in the chronic D-gal treated mice. Subcutaneous administration of D-gal (150 mg/kg) to Swiss albino mice for 42 consecutive days resulted in cognitive impairment as observed in Morris water maize (MWM) and Y maze test, which was ameliorated by concurrent treatment with PCA (80 mg/kg, and 100 mg/kg, p.o.). Importantly, PCA treatment attenuated the D-gal induced oxidative stress and significantly inhibited acetylcholinesterase (AChE) activity in mice brain. Furthermore, PCA treatment significantly lowered levels of inflammatory marker nuclear factor kappa B (NFκB) and reduced levels of proapoptotic enzyme caspase3. We also observed that PCA treatment exhibited β-secretase enzyme (BACE1) inhibitory effect. However, our results revealed that PCA treatment failed to decrease the level of advanced glycation end products both in vitro and in vivo. Taken together, current study demonstrated the significant neuroprotective effect of PCA against D-gal induced oxidative stress, neuroinflammation, cognitive impairment and apoptosis.


Introduction
Brain aging is proved to be a prominent and critical factor contributing to pathophysiology of Alzheimer's disease (AD). Population suffering from dementia accounts to almost 9.9 million per year, globally (Shwe et al. 2018). The deterioration of memory in dementia and AD is accompanied with structural changes in brain such as deposition of Aβ plaques, neurofibrillary tangles and damaged cellular proteins, nucleic acids or lipids and functional alterations such as reduced synaptic plasticity, neurogenesis and stimulation of various oxidative and inflammatory pathways (Wu et al. 2008).
In AD, as the age progresses some of the factors contributing predominantly to oxidative stress and neurodegeneration include mitochondrial abnormalities, Aβ plaques, hyperphosphorylated tau deposition, hyperactivation and abnormal function of astrocytes, microglia and excessive generation of advanced glycation end products (AGEs) (Su et al. 2008). Imbalance between production and inactivation of reactive oxygen species (ROS) in brain culminates into neuronal damage leading to neurodegeneration and impairment of vital brain functions including learning and memory (Xian et al. 2014).
Considerable research has highlighted the crucial role of excessive level of free radicals and reactive carbonyl compounds mainly methylglyoxal, glyoxal, triosephosphates in causing protein crosslinking. These carbonyl compounds react nonenzymatically with the proteins, more specifically, N-terminal amino group and side chains of lysine and arginine resulting in formation of amadori products which subsequently undergo oxidation and dehydration resulting in formation of AGEs and advanced lipoxidation end products (ALEs) (Srikanth et al. 2011). Elevated levels of AGEs contribute to increasing the risk of development of AD (Ko et al. 2010).
D-gal is a reducing sugar, which readily gets metabolized to glucose. However, at higher concentrations, it triggers the formation of ROS and leads to the formation of AGEs, that do not get metabolised but expedites the aging process in brain.
AGEs cause oxidative stress and vice versa (Sato et al. 2006). Interaction of AGEs with receptors for advanced glycation end products (RAGE), has many crucial consequences such as activation of NFκB pathway which culminates into enhanced synthesis of ROS and inflammatory mediators like TNF-α, IL-1, IL-6 via activated microglia and astrocytes (Lu et al. 2010a;Srikanth et al. 2011). Further, activation of NFκB pathway leads to neuronal apoptosis (Mastrocola et al. 2005). Increased ROS enhances amyloid precursor protein (APP) expression (Ko et al. 2015). APP thus formed is normally subjected to successive cleavage by enzymes β-secretase (BACE1) and γ secretase resulting in Aβ generation. BACE1 is the first enzyme which acts on rate limiting step in Aβ production. This supports the finding that AGEs could accelerate Aβ aggregation and accumulation which led to generation of Aβ plaques (Vitek et al. 1994). Taken together, chronic D-gal induced senescence and neuronal dysfunction serves as a promising model for neurodegenerative disorders (Shwe et al. 2018).
In spite of an alarming increase in population suffering from AD worldwide, there is lack of effective therapeutic agents. The clinically available drugs offer only symptomatic relief. Thus, emphasizing an urgent need for search of effective disease modifying or therapeutic intervention.
A plethora of studies performed previously has revealed neuroprotective potential of PCA in embolic cerebral ischemia (Guven et al. 2015). PCA was found to attenuate oxidative stress induced by cisplatin in rat brain (Ekinci-Akdemir et al. 2017). PCA enhanced long term potentiation and facilitated recovery of scopolamine induced memory impairment (Kim et al. 2017). Further, PCA was also found to protect against type 2 diabetes induced neurodegeneration in rats (Abdel-Moneim et al. 2017).
However, the effect of PCA against D-gal induced oxidative stress and neurocognitive impairments remains to be explored. The current study aims to investigate the potential protective effect of PCA in D-gal induced neurochemical and memory alterations and understand the underlying mechanisms.

Animals
Male Swiss albino mice (10 weeks old weighing 30-32 g) were procured from National Institute of Biosciences (NIBS), Pune. 6 mice per cage were housed in opaque polypropylene cages and kept under standard laboratory condition i.e. 22 ± 2 °C temperature, 60-70% humidity, maintenance of 12 h natural day and night cycle with food and water ad libitum. Prior to the testing, the mice were allowed to get habituated to the testing rooms for 1 week.

Ethics approval
The study protocol was approved by the Institutional Animal Ethics Committee (IAEC) of Institute of Chemical Technology, Mumbai, India with protocol number: ICT/ IAEC/2018/P21. All experimental procedures were performed as per the guidelines of the Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA), India.

Drugs, reagents and treatment schedule
PCA (≥ 98% HPLC), Rivastigmine (RVS), D-gal and methylglyoxal were purchased from Sigma Aldrich Co. (St. Louis, MO). L-Reduced glutathione (GSH), Trisbuffer Thiobarbituric acid (TBA) and 5, 5'-dithiobis (2-nitrobenzoic acid) (DTNB) were procured from Himedia Laboratories, Mumbai, India. Aminoguanidine HCl (AG) was purchased from TCI Ltd., bovine serum albumin (BSA) was purchased from S.D. Fine chemicals, nitroblue tetrazolium and sodium azide were purchased from Loba chemie pvt ltd. Mouse ELISA kits for NFκB, BACE1, Caspace3 and AGEs were purchased from Kine-sisDX, USA. All other chemicals used were of analytical grade. PCA was suspended in Tween 80 for oral administration. All other drugs were dissolved in 0.9% saline. At the end of study, the animals were sacrificed by carbon dioxide asphyxiation.
The dose of PCA was determined based on earlier studies (Sakamula & Thong-asa 2018). The total 40 mice were divided into five groups, each group having 8 mice: The work plan of the study is illustrated in Fig. 1.

In vitro BSA glycation study
In vitro protein glycation was carried out as described previously (Peng et al. 2008). For AGEs formation 200 µl of BSA (35 mg/ml) containing sodium azide (0.1 g/ml) was incubated with 400 µl of glucose (175 mg/ml) at 37 °C for 21 days, in absence or presence of PCA solution at different concentrations (10-200 µg/ml). AG (4 mg/ml) was used as a standard. AGEs, pentosidine and fructosamine compounds were assessed.

In vitro glycation assay with BSA-methylglyoxal
BSA-methylglyoxal assay was performed as described previously (Lunceford & Gugliucci 2005). 200 µl of BSA was incubated with 400 µl of methylglyoxal (0.4 mg/ml) at 37 °C for 14 days in absence or presence of different concentrations of PCA (10-200 µl/ml). Blanks containing BSA-glucose but no test sample were kept at -80 °C, until the measurement. AG (4 mg/ml) was used as a standard.

Determination of fructosamine (amadori) content
The fructosamine content was measured by nitroblue tetrazolium (NBT) reduction assay (Islam et al. 2014). The reduction of NBT was recorded by increase in absorbance at 530 nm from 5 to 15 min at 37 °C and expressed as % inhibition.

Morris water maze (MWM) test
The spatial memory and learning in mice were assessed using MWM test, performed according to method described % Inhibition = (absorbance of control − absorbance of sample) absorbance of control x 100 by Kumar et al. (Kumar et al. 2010), with slight modifications. A plastic pool having 122 cm diameter and 50 cm height and divided into 4 quadrants was filled with water (22 ± 2 °C) up to 30 cm and rendered opaque by dissolving powdered milk. In one quadrant, a white platform was placed 1 cm below water level. Three bright-coloured visual clues were fixed around the pool. Position of escape platform and visual cues were kept unchanged during entire experiment. The MWM test was divided into two phases as follows:

Maze acquisition phase
On day 20, mice underwent 4 consecutive training sessions, with 15 min intertrial interval. Each time, mouse was carefully put in one of the four quadrants and allowed to explore the platform for 90 s. In case the mouse could not locate the platform during this time, it was gently guided to the platform. The mouse was allowed to remain on the platform for 60 s. The latency period was recorded as initial acquisition latency (IAL).

Maze retention phase
After 24 h of maze acquisition phase, the escape latency was again evaluated and first retention latency (FRL) was recorded. Similarly, second retention latency was calculated on day 40 of treatment and recorded as second retention latency (SRL).

Probe test
The memory consolidation in mice was assessed by probe test performed 24 h after recording of SRL. In probe test the platform was removed and mice were allowed to swim freely for 60 s. Time spent by individual mice in the target quadrant was recorded.

Y maze test
Y-maze test is an excellent tool for assessing exploratory behaviour, spatial recognition and working memory. In Y maze test, apparatus (8 cm × 30 cm × 15 cm) with each of the arms at 120° apart was used. The test was conducted on day 22 and day 42, one hour after the dosing and included two trials. During the first trial, the mouse was allowed to explore only two arms (arm A and arm B) for 5 min keeping the third one (arm C) closed. Arm C was considered a novel arm. Post 30 min interval, the second trial was conducted, allowing each mouse to explore all 3 arms for a period of 5 min. Percentage %SAB and the number of entries in the novel arm were analysed for individual mouse. Consecutive entries in three different arms were counted as one alteration. Following formula was used to calculate percentage spontaneous alterations (Ghods-Sharifi et al. 2008):

Brain tissue collection
Animals sacrificed by CO 2 asphyxia were perfused with phosphate buffer solution (PBS). Isolated brains were immediately transferred in ice-cold isotonic saline and stored at -80ºC.

Biochemical assessment
Brain tissue was homogenised in 0.1 M PBS (pH 7.4). The homogenate (10% w/v) was subjected to cold centrifuge (10 000 g for 15 min). The supernatant thus obtained was used for further estimations. The protein estimation was done by the method prescribed by Lowry et. al. 1951(Lowry et al. 1951).

Evaluation of AChE
As described by Ellman et al. (Ellman et al. 1961), to 50 μl of the brain tissue homogenate, 3 ml of PBS (pH 8), 100 μl of acetyl thiocholine iodide (0.75 nM), and 100 μl of 10 mM Ellman's reagent were added. Absorbance was measured spectrophotometrically at an interval of 30 s for 5 min at 412 nm.

Estimation of reduced glutathione
As described by Smith et al. (Smith et al. 1988), each 3 ml reaction mixture (2.9 ml of DTNB in PBS and 0.1 ml of supernatant) was incubated at 37 °C for 15 min. Absorbance was measured spectrophotometrically at 412 nm. The level of GSH/mg of protein was estimated.

%SAB = Number of alterations (Number of total entries − 2)
x 100

Estimation of lipid peroxidation
The extent of lipid peroxidation was estimated by measuring MDA level as described by Draper et al. (Draper et al. 1993). Mixture of 0.5 ml Tris-HCl buffer and 0.1 ml of the brain homogenate, was incubated for 2 h. 1 ml of 10% w/v TCA solution was then added to above mixture and it was subjected to cold centrifuge (1000 g) for 10 min. The supernatant was collected and to it, an equal volume of TBA (0.67% w/v) was added. The mixture was then subjected to heating in a boiling water bath for 15 min. The mixture was cooled in an ice bath and to it, 1 ml distilled water was added. The level of MDA was estimated by measuring absorbance spectrophotometrically at 532 nm.

Estimation of superoxide dismutase
As per method by Nandi and Chatterjee (Nandi & Chatterjee 1988), to 2.8 ml of PBS, 0.1 ml supernatant and 0.1 ml pyrogallol solution (2.6 mM in 10 mM HCl) was added. Absorbance was measured spectrophotometrically at 325 nm. The level of SOD units/mg of protein was estimated.

Immunoassays for biomarkers
NFκB, AGEs, BACE1 and Caspase3 were estimated using ELISA kits. The manufacturer's protocol was strictly followed and readings were taken using ELISA plate reader (BIORAD).

Statistics
GraphPad Prism Version 5, San Diego, CA was used for statistical analysis. Data are represented as mean ± SEM. Escape latency in MWM and % SAB in Y maze were analysed using repeated measure two-way ANOVA followed by post hoc Bonferroni's test. In other tests, one-way ANOVA followed by post hoc Tukey's multiple comparison tests was used to calculate statistical significance.

PCA does not show in vitro anti-glycating effect
PCA at various concentrations (0.1 mg/ml to 2 mg/ml) did not show significant inhibition of the formation of methylglyoxal compounds compared to the BSA control since the fluorescence was not observed to be decreased significantly. AG was effective in inhibiting methylglyoxal formation at concentration of 1 mg/ml and showed significant activity (P < 0.001). AGEs formation in the BSA-glucose system was measured by auto-fluorescence. For both, AGEs and pentosidine compounds, aminoguanidine hindered the BSA glycation and decreased the formation of fluorescent glycation products as evidenced by significantly lower relative fluorescence unit values as compared to negative control (P < 0.001 was observed for both AGEs and pentosidine compounds). PCA at concentrations (0.1 mg/ml to 2 mg/ ml), failed to significantly inhibit the formation of fluorescent glycation products.
PCA (0.1 mg/ml to 2 mg/ml) did not exhibit significant inhibition of formation of fructosamine (amadori) compounds as compared to the BSA control. AG at a concentration of 1 mg/ml significantly inhibited the formation of amadori compounds.
The results of in vitro anti-glycating activity thus revealed that PCA does not inhibit formation of AGEs or reduce the formation of amadori products. Also, the results of methylglyoxal assay suggested that PCA failed to neutralise the carbonyl compounds formed (results not shown).

MWM test
To investigate effect of PCA on memory we evaluated escape latency by performing MWM test on day 20, 21 and 40 respectively followed by probe test on day 41.
As shown in Fig. 2a, for D-gal group, observed IAL (p < 0.01), FRL (p < 0.001) and SRL (p < 0.001) were significantly higher than vehicle treated group. PCA and RVS treated groups showed progressive decline in escape latency over the treatment period. Treatment with PCA (80 mg/kg) exhibited significantly decreased FRL (p < 0.01) and SRL (p < 0.001) and PCA (100 mg/kg) exhibited significantly decreased IAL (p < 0.05), FRL (p < 0.001) and SRL (p < 0.001) as compared to D-gal group, The results of PCA were comparable with RVS (p > 0.05).
To assess memory consolidation, a probe test was performed 24 h after the MWM test. During the 60 s probe trial, time spent by mouse in each quadrant was recorded. As depicted in Fig. 2b we observed that the time spent in the target quadrant area was significantly higher in PCA (100 mg/kg) (p < 0.05) and RVS (p < 0.05) as compared to D-gal group, while D-gal group spent less time in target quadrant (p < 0.01) as compared to vehicle. The results of probe test for PCA (80 mg/kg) were not significant as compared to D-gal group.

Y maze test
Y maze test performed on day 22 and day 42, evaluated % SAB which signifies spatial and working memory along with learning, whereas the number of entries in the novel arm reflects locomotor activity and exploratory behaviour. % SAB was significantly lower for the D-gal group on day 22 (p < 0.05) and day 42 (p < 0.001) as compared to the vehicle control. As shown in Fig. 2c, % SAB was significantly increased in PCA treatment groups. On day 22, % SAB for PCA (100 mg/kg) but not PCA (80 mg/kg), was significantly higher (p < 0.001) as compared to D-gal group. On day 42, % SAB was significantly higher (p < 0.001) for both doses of PCA. RVS group exhibited significantly increased % SAB on day 22 and day 42 of evaluation (p < 0.001 on both days).
As shown in Fig. 2d, the number of entries in novel arm of Y maze were significantly reduced in D-gal treated group as observed on day 22 and day 42 (p < 0.01 and p < 0.05, respectively). However, on day 22 and day 42, PCA (100 mg/ kg) (p < 0.05 and p < 0.01) and RVS (p < 0.05 and p < 0.01), treated groups exhibited significant improvement in frequency of novel arm entries as compared to D-gal group. However, PCA (80 mg/kg) did not significantly increase number of arm entries.

PCA decreases AChE activity in mice brain
As depicted in Fig. 3a AChE activity was significantly higher in D-gal group (p < 0.001), compared to vehicle control. In RVS group, significant inhibition of AChE was observed (p < 0.001). Treatment with PCA (80 mg/kg & 100 mg/kg) significantly attenuated enhanced AChE level (p < 0.01 & p < 0.001 respectively). However, AChE inhibition exhibited by PCA was lower than RVS treated group.

PCA ameliorates D-gal Induced oxidative stress
As shown in Fig. 3b, c, and d we observed that D-gal treatment significantly decreased SOD and GSH levels (p < 0.001 and p < 0.001, respectively) and elevated brain MDA (p < 0.001) compared to vehicle control. Treatment with PCA (80 mg/kg and 100 mg/kg) significantly elevated SOD level (p < 0.05 and p < 0.01, respectively) and GSH (p < 0.01 and p < 0.001, respectively) as compared to the D-gal group. RVS significantly increased both SOD and GSH levels as compared D-gal group (p < 0.01 and p < 0.001) respectively. Further PCA (100 mg/kg) (p < 0.01) and RVS (p < 0.01) significantly attenuated D-gal induced increased MDA levels but PCA (80 mg/ kg) did not significantly lower MDA level.

Effect of PCA on AGEs levels in mice brain
Estimation of brain AGEs level using ELISA revealed that, D-gal treatment resulted in significantly increased AGEs levels in mice brain compared to vehicle control (p < 0.001). We observed that treatment with PCA (80 mg/kg & 100 mg/ kg) did not significantly reduced the AGEs levels as compared to D-gal treated group as shown in Fig. 4a. For RVS however, significant AGEs inhibitory effect was observed (p < 0.01).

PCA reduces levels of NFκB in mice brain
The estimation of NFκB in brain homogenates was done using ELISA kits. As shown in Fig. 4b, we found that chronic treatment with D-gal markedly increased NFκB levels in brain (p < 0.001). PCA (100 mg/kg) significantly reduced NFκB levels in mice brain (p < 0.05) which was comparable with the results obtained in RVS treated group (p < 0.001). However, PCA (80 mg/kg) did not exhibit significant reduction in brain NFκB level.

PCA reduces levels of BACE1 in mice brain
Estimation of BACE1 levels in mice brain homogenates by ELISA revealed increased BACE1 levels in D-gal treated group (p < 0.001). As shown in Fig. 4c, PCA (80 mg/kg and 100 mg/kg) significantly reduced BACE1 levels in mice brain (p < 0.05 at both dose levels). In RVS treated group also, significant reduction in BACE1 levels was observed as compared to D-gal group (p < 0.05).

PCA prevents D-gal induced neuronal apoptosis
Quantitative evaluation Caspase3 in mice brain homogenates was performed using ELISA kit. As shown in Fig. 4d, Cas-pase3 levels were significantly higher after D-gal treatment Time spent in the target quadrant for MWM test and c % SAB and d number of entries in the novel arm for Y maze test. Data expressed as mean ± SEM (n = 6). For Escape latency and % SAB, evaluation was performed by repeated measure two-way ANOVA followed by post hoc Bonferroni multiple comparison test. [ # p < 0.05, ### p < 0.01, ### p < 0.001 vs Vehicle group; *p < 0.05, **p < 0.01, ***p < 0.001 vs D-gal group]. All other results were evaluated by one-way ANOVA followed by post hoc Tukey's test [[ # p < 0.05, ### p < 0.01, ### p < 0.001 vs. vehicle group and *p < 0.05 and **p < 0.01 vs. D-gal group] which was attenuated by treatment with PCA (100 mg/kg, p < 0.05) and RVS (p < 0.05). However, we observed that there was no significant reduction in caspase3 level in mice brain from PCA (80 mg/kg) group.

Discussion
Memory decline is a characteristic of aging and also serves as an indicator of neurodegenerative changes in brain. D-gal crosses the blood brain barrier and is taken up by brain cells via glucose transport type 1 (GLUT-1) (Shwe et al. 2018). Over accumulation of D-gal induces senescence, by causing oxidative and glycative stress, neuroinflammation and neuronal apoptosis resembling the old age related cognitive impairment and behavioural dysfunction which is suggested to recapitulate symptoms of AD (Wei et al. 2005).
Consistent with the reported data (Li et al. 2015;Wang et al. 2020), we observed that the learning ability and memory function in D-gal group was significantly impaired, which was ameliorated by treatment with PCA at (80 mg/ kg and 100 mg/kg) dose levels. Further, findings of probe test confirmed the memory consolidation at both dose levels of PCA. Moreover, in Y maze test increased % SAB in PCA treated groups at both dose levels indicated significantly improved short term memory and cognition compared to D-gal group. Thus, prolonged treatment with PCA protected against D-gal induced memory impairment.
Memory decline in AD is associated with cholinergic deficiency. Enhanced levels of brain AChE can cause cholinergic deficiency as depicted in AD. D-gal increases brain AChE activity (Zhong et al. 2009) and causes impairment of brain cholinergic function (Lu et al. 2010b). Consequently, we observed increased AChE level in D-gal treated group which was significantly ameliorated by PCA at both (80 mg/kg and 100 mg/kg) dose levels. These results were comparable with RVS, indicating that protective effect of PCA against D-gal induced neurodegeneration and memory impairment could be at least partially mediated through its AChE inhibitory activity. Fig. 3 Effect of PCA on a AChE activity b brain SOD level, c brain GSH level and d brain MDA level. Data expressed as mean ± SEM (n = 6) and evaluated by one-way ANOVA followed by post hoc Tukey's test. [ ### p < 0.001 vs. vehicle group, *p < 0.05, **p < 0.01 and ***p < 0.001 vs. D-gal group] On overdose, D-gal is converted to aldose and hydrogen peroxide via the action of galactose oxidase, leading to the formation of a superoxide anions and oxygen-derived free radicals (Lu et al. 2010b), which ultimately results in diminished levels of SOD and GSH and elevated MDA level in D-gal group. Chronic treatment with PCA at both dose levels (80 mg/kg and 100 mg/kg), significantly attenuated the D-gal induced oxidative stress by increasing SOD and GSH level and decreasing MDA level in mice brain which suggested the neuroprotective effect of PCA through antioxidant mechanism. Thus, our observations support the previous findings by Ekinci et al., who demonstrated ability of PCA to attenuate oxidative stress induced by cisplatin (Ekinci-Akdemir et al. 2017) and the work by Cosma et. al., showing attenuation of low density lipoprotein cholesterol oxidation by PCA (Cosma et al. 2000).
These findings demonstrate protective effect of PCA in D-gal induced behavioural and neurochemical alterations. Chronic overdose of D-gal leads to formation and accumulation of AGEs. Elevated AGEs levels in body are implicated in aggravating the oxidative stress and triggering the production of proinflammatory cytokines, which culminates in impaired synaptic plasticity and neuronal death. The oxidative stress and AGEs initiate a vicious cycle leading to neurodegeneration and cognitive impairment.
Investigation of in vitro antiglycation effect of PCA in glucose-BSA system and methylglyoxal scavenging assay revealed that, PCA does not exhibit significant AGEs inhibitory or methylglyoxal scavenging activity in vitro. Our results are consistent with the reported literature (Chen et al. 2016). Further, estimation of brain AGEs levels using ELISA revealed no significant decrease in AGEs levels after PCA treatment, thus supporting our findings of in vitro study. This also suggests that, PCA may have shown neuroprotective effect by other than AGEs inhibitory mechanisms, thus we further explored the effect of PCA treatment on inflammatory and apoptotic pathways.
Several studies have revealed that interaction between AGEs and RAGE leads to activation of NFκB pathway, which plays a significant role in cell proliferation, apoptosis, and inflammatory response (Zhong et al. 2020). Consistent with these findings, we observed elevated levels of NFκB in D-gal treated mice brain. Fig. 4 Effect of PCA on a brain AGEs level, b brain NFκB level c BACE1 level and d brain Caspase3 level. Data expressed as mean ± SEM (n = 6) and evaluated by one-way ANOVA followed by post hoc Tukey's test. [ ## p < 0.01, ### p < 0.001 vs. vehicle group, *p < 0.05, **p < 0.01 and ***p < 0.001 vs. D-gal group] Decreased NFκB levels observed in PCA treated mice suggests that protective effect of PCA against D-gal induced neuroinflammation might result through inhibition of nuclear translocation of NFκB or its expression.
Excessive formation of AGEs promotes Aβ formation by enhancing expression of APP and upregulates BACE1 and presenilin 1 (Ko et al. 2015;Lubitz et al. 2016). BACE1 is essential for cleavage of APP which is an endogenous precursor of Aβ. Our investigation revealed that chronic D-gal treatment led to upregulation of BACE1 levels in mice brain. Significant reduction in BACE1 level after PCA treatment suggests that PCA may protect against Aβ generation through its potential BACE1 inhibitory effect. These findings are consistent with the study reporting BACE1 inhibitory effect of PCA isolated from Corni fructus (Yoon et al. 2014). However, further detailed investigation is required to confirm the BACE1 inhibitory effect of PCA.
Neuronal damage due to D-gal induced oxidative stress can be reflected from the levels of proapoptotic marker caspase3 in mice brain. In our observations, we found that PCA treatment reduced the brain caspase3 level at 100 mg/ kg dose, indicating its potential antiapoptotic effect which may be attributed to inhibition of NFκB and BACE1. Our findings revealed that RVS treatment also ameliorated D-gal induced apoptosis which is in agreement with the previous findings reported in literature (Gupta et al. 2021). These results demonstrate protective effect of PCA against D-gal induced neurotoxicity via attenuating D-gal induced neuroinflammation and neuronal apoptosis.

Conclusion
Based on our findings we conclude that neuroprotective effect of PCA in D-gal induced neurotoxicity and memory impairment may be at least partly mediated through: i) Protecting activity of antioxidant enzymes in brain, ii) Decreasing brain AChE activity, iii) Inhibiting BACE1 activity in mice brain and iv) Modulating inflammatory and proapoptotic markers in mice brain. These findings provide a new insight suggesting p-Coumaric acid as a potential agent for the treatment of age-related brain disorder. However, further research is needed to elucidate the exact molecular basis for its use in the management of AD.
Authors' contributions Prof. Archana Juvekar: Conception of the work, supervised the research, critical revision of the article.
Pratibha Daroi: Conception and design of the work, performed experiments, data collection, analysis, Interpretation and original draft preparation.
Shrikant Dhage: Performed analysis, data collection and interpretation, critical revision of the article. All authors reviewed the results and approved the final version of the manuscript.

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethics approval The study protocol was approved by the Institutional Animal Ethics Committee (IAEC) of Institute of Chemical Technology, Mumbai, India with protocol number: ICT/IAEC/2018/P21. All experimental procedures were performed as per the guidelines of the Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA), India.

Conflicts of interest/Competing interests
The authors declare there are no conflict of interest/ no competing interests.