Naringin Protects against Tau Hyperphosphorylation in Aβ25–35-Injured PC12 Cells through Modulation of ER, PI3K/AKT, and GSK-3β Signaling Pathways

Alzheimer's disease (AD) is the most common form of dementia and a significant social and economic burden. Estrogens can exert neuroprotective effects and may contribute to the prevention, attenuation, or even delay in the onset of AD; however, long-term estrogen therapy is associated with harmful side effects. Thus, estrogen alternatives are of interest for countering AD. Naringin, a phytoestrogen, is a key active ingredient in the traditional Chinese medicine Drynaria. Naringin is known to protect against nerve injury induced by amyloid beta-protein (Aβ) 25–35, but the underlying mechanisms of this protection are unclear. To investigate the mechanisms of naringin neuroprotection, we observed the protective effect on Aβ25–35-injured C57BL/6J mice's learning and memory ability and hippocampal neurons. Then, an Aβ25–35 injury model was established with adrenal phaeochromocytoma (PC12) cells. We examined the effect of naringin treatment on Aβ25–35-injured PC12 cells and its relationship with estrogen receptor (ER), phosphatidylinositol 3-kinase/protein kinase B (PI3K/AKT), and glycogen synthase kinase (GSK)-3β signaling pathways. Estradiol (E2) was used as a positive control for neuroprotection. Naringin treatment resulted in improved learning and memory ability, the morphology of hippocampal neurons, increased cell viability, and reduced apoptosis. We next examined the expression of ERβ, p-AKT (Ser473, Thr308), AKT, p-GSK-3β (Ser9), GSK-3β, p-Tau (Thr231, Ser396), and Tau in PC12 cells treated with Aβ25–35 and either naringin or E2, with and without inhibitors of the ER, PI3K/AKT, and GSK-3β pathways. Our results demonstrated that naringin inhibits Aβ25–35-induced Tau hyperphosphorylation by modulating the ER, PI3K/AKT, and GSK-3β signaling pathways. Furthermore, the neuroprotective effects of naringin were comparable to those of E2 in all treatment groups. Thus, our results have furthered our understanding of naringin's neuroprotective mechanisms and indicate that naringin may comprise a viable alternative to estrogen therapy.


Introduction
Alzheimer's disease (AD), a neurodegenerative disorder characterized by progressive cognitive impairment, is the leading cause of dementia [1]. Recent research has shown that estrogen not only regulates the growth and development of the reproductive system but also serves a neuroprotective role in the pathophysiology of AD [2]. Estrogen receptor β (ERβ) is highly expressed in the hippocampus. Binding of estrogen to ERβ activates the phosphatidylinositol 3kinase/protein kinase B (PI3K/AKT) pathway and reduces the toxic effects of amyloid beta-protein (Aβ) on adrenal phaeochromocytoma (PC12) cells [3]. Estrogen also helps repair the central nervous system (CNS), stimulating axon regeneration following subcortical axon injury by modulating the PI3K/AKT pathway [4]. The PI3K/AKT signaling pathway is involved in growth, development, learning, and memory aspects of the CNS, and plays a significant role in the prevention and treatment of AD [5]. Activation of the PI3K/AKT signaling pathway inhibits the activity of glycogen synthase kinase (GSK)-3β, which is closely linked to abnormal Tau protein phosphorylation [6]. As the protein kinase that regulates Tau protein phosphorylation, GSK-3β is a key factor in the occurrence and development of AD.
Deactivation of GSK-3β could potentially retard the development of AD hallmarks namely, Aβ plaques and neurofibrillary tangles (NFTs) [7]. In a dementia mouse model, estrogen deficiency was shown to aggravate cognitive impairment through the ERβ/GSK-3β pathway [8]. Thus, the neuroprotective effects of estrogen in AD are likely related to the ER, PI3K/AKT, and GSK-3β signaling pathways.
Unfortunately, long-term estrogen therapy can cause catastrophic side effects such as breast and uterine cancer. Thus, alternatives to estrogen should be investigated as viable candidates for countering AD.
Phytoestrogens are a class of plant-derived compounds similar to estradiol (E 2 ) in structure and function. Phytoestrogens bind ER and can impart both estrogen-like and anti-estrogen-like effects without the harmful side effects associated with long-term estrogen therapy [9]. Hence, phytoestrogens are considered viable candidates for the prevention and treatment of AD [10]. Previous research conducted by our group showed that Drynaria, a plant used in traditional Chinese medicine, had a significant protective effect on the functionality of the CNS [11]. Naringin, a key active ingredient of Drynaria fortune, is a flavonoid phytoestrogen with antioxidant, anti-inflammatory, and anti-apoptotic activities [12]. The estrogen-like activity of naringin has been confirmed [13], making it a viable therapeutic candidate for AD studies. In a recent study, naringin exhibited Mas receptor-mediated neuroprotective effects against Aβinduced cognitive impairment and mitochondrial toxicity in rats [14]. Another study showed that the nitric oxide signaling pathway is involved in naringin's regulation of intranasal manganese and intraventricular Aβ-induced neurotoxicity [15]. However, studies examining the effect of naringin on the ER, PI3K/AKT, and GSK-3β signaling pathways are lacking.
In the present study, we investigated the protective effects of naringin on Aβ 25-35 -injured C57BL/6J mice and Aβ 25-35 -injured PC12 cells, and compared them to the effects of E 2 . In cell experiments, the ER inhibitor ICI182780, the PI3K inhibitor LY294002, and the GSK-3β inhibitor LiCl were used to study the relationship between naringin's protective effects and the ER, PI3K/AKT, and GSK-3β signaling pathways. We found the neuroprotective effects of naringin were comparable to those of E 2 and involve the inhibition of Aβ 25-35 -induced Tau phosphorylation through modulation of the ER, PI3K/AKT, and GSK-3β signaling pathways.

Animals.
Eight-week-old male C57BL/6J mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). Mice were raised in specific pathogen free (SPF)-grade animal facilities, with a temperature of 20-22°C and relative humidity of 55-65%. The light source was natural, and the light and dark alternated for 12 hours (12L : 12D), with ad libitum access to water and food. In order to make the mice adapt to the environment, they were adaptively fed for 1 week before the experiment. The operations were carried out under general anesthesia to reduce the pain of mice. The study was approved by the Ethical Committee of Heilongjiang University of Chinese Medicine (2020092502) and was conducted according to accepted animal care practices.

Experimental
Treatments. C57BL/6J mice were subjected to the following treatments: 24 mice were randomly assigned into the control group, Aβ 25-35 group, naringin group, and E 2 group, with 6 mice in each group. The Aβ 25-35 group, naringin group, and E 2 group were anesthetized by inhaling isoflurane with portable small animal anesthesia machine (ZS-MV-IV, Beijing Zhongshidichuang Science and Technology Development Co., Ltd., Beijing, China) and fixed on the stereotaxic apparatus (ZS-FD/S, Beijing Zhongshidichuang Science and Technology Development Co., Ltd., Beijing, China). The mouse's scalp was cut open to expose the skull. At the injection point in the hippocampal regions (2.0 mm behind the Bregma, 2.1 mm from the lateral to the left and right, 3.7 mm deep under the skull surface), holes were drilled into the surface with a skull drill. Aβ 25-35 1 μl (10 μg/μl) was subsequently injected on each side with a microinjection needle. Aβ 25-35 1 μl was injected over the course of 5 minutes, and the needle was kept in place for an additional 10 minutes after the injection was completed. The scalp was then sutured closed. Normal saline, naringin 100 mg kg −1 d −1 , or E 2 0.13 mg kg −1 d −1 was administered according to the group classification for 21 days, with an oral administration at the dosage of 10 mL kg −1 d −1 . After the control group was anesthetized in the same way, the same amount of normal saline was injected into the bilateral hippocampus, and the same amount of normal saline was given by oral administration for 21 days (Figure 1).

Novel Object Recognition Test.
To examine learning and memory, we performed the novel object recognition (NOR) test [16]. The test is based on animals' innate preference for a novel object, in which mice remembering the familiar object will devote more time exploring a novel one [17]. Each mouse was placed in a box (without objects) for 5 minutes to adapt to the surroundings (habituation). Two identical small cubes were placed at adjacent corners in the box. Each mouse was positioned with its back to the cubes and placed at the midpoint of the opposite box wall. Then, each mouse was given 5 minutes to explore the new objects (training). One hour after the training period, one of the small cubes was replaced with a completely different novel object (Novel Object Recognition Equipment, Shanghai Yishu Information Technology Co., Ltd., Shanghai, China). Each mouse was placed at the same position in the box, and the time taken by each mouse to explore the novel object (Tn) and to explore the familiar object (Tf) within 5 minutes was recorded (test). Objects and the arena were thoroughly cleaned with 70% ethanol between experiments with individual mice to ensure that their behavior was not guided by odor cues [18]. We used the following indices: Discrimination index = (Tn − Tf)/(Tn + Tf). Recognition index = Tn/(Tn + Tf).

H&E Staining.
After the NOR test, the mice were euthanized per protocol and their brains were removed, the residual blood was washed with normal saline, and the brains were fixed with 4% paraformaldehyde for 48 hours, and stained pathological sections were prepared. The brain tissue fixed in 4% paraformaldehyde was dehydrated and embedded by conventional pathological methods to make wax blocks, and we used a rotary slicer to cut the wax block into brain sections (4.0 μm thick), dewaxed in xylene, and dehydrated with 95% ethanol. Then, the sections were stained with H&E solution and dehydrated with graded alcohol. We then made the sections transparent with xylene and sealed the sections with neutral glue. H&E staining was performed to observe the morphology of the hippocampus.

MTT Cell Viability
Assay. Following treatment, cells were collected into 96-well plates. Twenty microliters of a 5 g/L methyl thiazolyl tetrazolium (MTT) solution (Sigma-Aldrich) were added to each well, and the plates were incubated at 37°C for 4 hours. Following removal of the supernatant, 150 L of dimethylsulfoxide (DMSO) solution was added to each well. Plates were incubated at 37°C for 10 minutes on a shaker, and the absorbance was measured at 570 nm on a microplate reader (MK3; Shanghai Thermoelectric Instrument Co., Ltd., Shanghai, China). The percentage of normalized cell proliferation was calculated as follows: absorbance value of treated cells/absorbance value of untreated cells × 100%. 2.9. qRT-PCR. Treated cells were collected into 6-well plates. Total RNA was extracted, and its quality was detected. CDNA was synthesized and amplified using the Ultra SYBR Mixture kit (Biosharp Life Sciences, Seoul, Korea) on a MX3000p fluorescent quantitative PCR instrument (Agilent Technologies, Santa Clara, CA, USA). The PCR primer sequences were: (1) ERα forward: 5 ′ -ATGGGCACTTC AGGAGACAT-3 ′ and reverse: 5 ′ -AAAGGTGGTTCAGC ATCCAA-3 ′ ; (2) ERβ forward: 5 ′ -TCTGGGTGATTGCG AAGAG-3 ′ and reverse: 5 ′ -TGCCCTTGTTACTGATGTG C-3 ′ ; (3) β-actin forward: 5 ′ -TGTCACCAACTGGGAC GATA-3 ′ and reverse: 5 ′ -GGGGTGTTGAAGGTCTCAAA-3′. The PCR cycling parameters were: 1 cycle of denaturation at 95°C for 10 minutes; followed by 40 cycles of denaturation at 95°C for 30 seconds, annealing at 60°C/57°C for 1 minute, and extension at 74°C for 1 minute. CT values, dissolution curves, and amplification curves were calculated by the PCR instrument. The relative expressions of ERα and ERβ were calculated by the 2 −ΔΔCt method.
2.10. Immunoblot Assays. Treated cells were collected into 6well plates and lysed with radioimmunoprecipitation assay buffer (RIPA) buffer containing 1 mM phenylmethanesulfonyl fluoride (PMSF) for 30 minutes on ice. Following total protein extraction, protein concentration was determined using the bicinchoninic acid assay (BCA) method. Proteins were separated by SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to polyvinylidene fluoride (PVDF) membranes. The PVDF membranes were incubated overnight at 4°C with the following primary antibodies: rabbit anti-ERβ, rabbit anti-p-AKT (Ser473, Thr308), rabbit anti-AKT, rabbit anti-p-GSK-3β (Ser9), rabbit anti-p-Tau (Thr231, Ser396), rabbit anti-Tau (all purchased from Beijing Boosen Biological Technology Co., Ltd., China), and mouse anti-GSK-3β (Wuhan Boster Biological Technology Co., Ltd., Wuhan, China). Membranes were washed and incubated with horseradish peroxidase (HRP)-labeled goat anti-mouse IgG or HRP-labeled goat anti-rabbit IgG (Wuhan Boster Biological Technology Co., Ltd.) at room temperature for 1 hour. Membranes were washed again, the enhanced chemiluminescence (ECL) kit (Beyotime Biotechnology) was used to detect immunoreactive bands. A lane ID gel analysis system was used to analyze the grey values.
2.11. Statistical Analyses. All statistical analyses were performed using SPSS version 21.0 (IBM Corp., Armonk, NY, USA). One-way analysis of variance (ANOVA) was implemented for comparison between groups, least significant difference (LSD) tests were applied to compare means between groups, and SNK was used for post-statistical corrections. Experimental data were expressed as mean ± standard deviation (x ± S), and statistical significance was set at P < 0:05.

Behavioural Neurology
In contrast, in the naringin group, the discrimination index significantly increased, which reflects the ability of learning and memory improvement. Additionally, the amount of time that Aβ 25-35 -injured mice spent exploring the novel object was significantly less than that of the control group, as these mice were less interested in exploring and learning about the novel object. However, the mice supplemented with naringin were observed to have a significant increase in preference for the novel object ( Figure 2).

Effect of Naringin on Morphology of Hippocampal
Neurons of C57BL/6J Mice Injured by Aβ [25][26][27][28][29][30][31][32][33][34][35] . In the control group, the cells in the hippocampus were arranged regularly, the number of neurons was large, the staining was uniform, the nucleolus was clearly visible, and there was no nuclear shrinkage. After the Aβ 25-35 treatment, compared with the control group, the hippocampal cells in the model group were disordered, some neurons were lost, the number of neurons was reduced, the cells were reduced, the nucleolus was not obvious, and nuclear shrinkage was obvious. In contrast, the arrangement of hippocampal neurons in the naringin group was more regular, there were more neurons, the morphology was normal, the nucleolus was clear, and there was no obvious nuclear shrinkage ( Figure 3).
3.5. The Role of the ER Pathway in Naringin Protection. We hypothesized that naringin exerts its protective effects over

Behavioural Neurology
Aβ 25-35 -induced injury by modulating the PI3K/AKT and GSK-3β signaling pathways through activation of the ER pathway. To test this hypothesis, we first monitored the levels of ER subtype mRNAs in treated PC12 cells by qRT-PCR. As we previously found 100 pM E 2 and 1 mM naringin to be the most effective concentrations in PC12 cell viability assays, we used these concentrations for our ER expression analyses. Reverse-transcribed total RNA from PC12 cells treated with E 2 or naringin, alone and in combination with ICI182780, served as the template for quantitative real-time polymerase chain reaction (qRT-PCR) reactions with primers specific to ERα and ERβ. As levels of ERα mRNA in cells were exceptionally low, it was difficult to quantitatively amplify Erα, and we thus instead chose to focus on ERβ. Compared with untreated cells, the level of ERβ mRNA was significantly increased in naringin-treated cells (Figure 6(a)). Exposure of PC12 cells to both ICI182780 and naringin significantly reduced ERβ mRNA levels compared to treatment with naringin alone. Similar results were seen for E 2 treatment in the presence and absence of ICI182780. These results indicate that ERβ mRNA expression can be upregulated by naringin.

Discussion
Alzheimer's disease, the most common degenerative disease of the CNS, is affected by estrogen levels [19]. Senile plaques (SPs) formed by Aβ and NFTs formed by hyperphosphorylated Tau protein are widely accepted neuropathological markers of AD [20]. Studies have shown that compared with Aβ, the density of NFTs formed by hyperphosphorylated Tau is more strongly correlated with cognitive function [21]. Thus, the occurrence and developmental process of Tau protein hyperphosphorylation can be used as a criterion for judging the course of AD [22]. Tau protein mediates the neurotoxic effects of Aβ, and hyperphosphorylated Tau leads to increased Aβ production [23]. Therefore, Tau protein comprises an important therapeutic target in AD drug development.
Tau protein has more than 45 phosphorylation sites. Tau protein molecules with abnormal phosphorylation are separated from microtubules and eventually form NFTs. The affinity between Tau and microtubules is mainly mediated by threonine/serine (Thr/Ser) phosphorylation, with phosphorylation of Thr231 and Ser396 leading to increased NFT formation [24]. Tau protein phosphorylation is regulated by a variety of protein kinases and protein phosphatases. Among them, GSK-3β is the most influential protein kinase in driving AD-like Tau protein hyperphosphorylation. The active form of GSK-3β co-localizes with NFTs in the AD brain and phosphorylates Tau at multiple ADrelated sites, including Ser396 and Thr231 [25]. Furthermore, activation of GSK-3β has been shown to induce Tau hyperphosphorylation and cognitive impairment [26].