Eriodictyol ameliorated cognitive dysfunction in APP/PS1 mice through inhibited ferroptosis via vitamin D receptor mediated Nrf2 activation

Lin Li The Third A liated Hospital of Chongqing Medical University Wenjun Li The Third a liated hospital of Chongqing medical university Xiangru Zheng The Third A liated Hospital of Chongqing Medical University Qinglong Liu The third a liated hospital of Chongqing medical university Qian Du The third a liated hospital of Chongqing medical university Yujie Lai The Third A liated Hospital of Chongqing Medical University Songqing Liu (  liusq@hospital.cqmu.edu.cn ) The Third A liated Hospital of Chongqing Medical University https://orcid.org/0000-0002-9269-3961

The Y-maze test was carried out according to a previous protocol [20]. In the spontanenous alternation test, the number of total entries and spontanenous alternation of mice were recorded. The percentage of alternation was calculated using the following formula: [the number of spontanenous alternation/ (total number of arm entries -2)] ×100% [21]. In the novel arm exploration test, the number of times the mice entered the novel arm, the distance traveled, the time spent in the novel arm of mice were recorded.
Congo Red Staining.
The Aβ plaques in the brain of mice were stained by Congo red and photographed using a light microscope (Nikon, Janpan) according to the manufacturer's operating instructions [22].
Iron in the mice's brain was stained using Prussian blue following a previous protocol [23], and take photograph under a uorescence microscopy (Nikon, Janpan) .

Immunohistochemistry (IHE) analysis.
IHE analysis of the brain tissues using anti-VDR and anti-Aβ antibodies, and the steps refers to a previous study [24].
HT-22 cells which were from the Chinese Academy of Sciences (Shanghai, China) were cultured in Dulbecco's modi ed Eagle's medium (DMEM, (Biological Industries, Israel) supplemented with 10% FBS (Biological Industries, Israel) and 1% penicillin/streptomycin (Beyotime Biotechnology, Shanghai). The cells were maintained in a sterile and humidi ed atmosphere at 37 ℃ with 5% CO 2 . Eriodictyol was dissolved in dimethyl sulfoxide (DMSO) and the nal concentration of DMSO in culture medium was ≤ 1‰. The preparation method of Aβ 1−42 oligomers is as follows: Aβ 1−42 was dissolved in DMSO and sonicated for 5 minutes in a cold-water bath, then immediately stored at -80 ℃ as stock. The Aβ 1−42 /DMSO solution was diluted with serum-free DMEM to a nal concentration of 100 µM and stored at 37 ℃ until use [25].
Cell Viability Assay.
The cell viability of HT-22 cells after treated with eriodictyol (0-128 µmol/L) and the role of eriodictyol in Aβ 1−42 oligomers-induced cellular damage were detected using CCK-8 assay according to the manufacturer's instructions, and the absorbance was measured at 450 nm by a SynergyH1 microplate reader (BioTek).
HT-22 cells were collected after treatment for 48 h and xed in 2.5% glutaraldehyde. Then each sample was made as slice according to a previous protocol [26]. The digital images were acquired using a JEM-1400 PLUS transmission electron microscope (TEM).
FITC-phalloidin used to stain cytoskeleton of HT-22 cells. HT-22 cells was plated in 24-well plate and cultured with culture medium containing 10 µM RA to induced neural axons formation for 3 days. Then 20 µM Aβ 1−42 oligomers were added into culture medium after cells pre-treated with eriodictyol in different concentration for 2 h, the vehicle control group only was added culture medium. 48 h later, the cells were xed in 4% (wt/vol) paraformaldehyde for 15 min and incubated with FITC-phalloidin solution and DAPI in turn at 37 ℃. Finally, images were collected under a uorescence microscope at 400× magni cation and the length of axons was measured by ImageJ software [6].
PI assay was implemented according to the manufacturer's instructions. After that, cells were observed and taken pictures by a uorescence microscope at 100× magni cation [27].
The level of iron in HT-22 cells were detected by Calcein-AM following the manufacturer's instructions [28].

ROS Assay.
HT-22 cells after treated were assessed intracellular ROS level using ROS Assay Kit according to the manufacturer's instructions. Observe and take photos under a uorescence microscope at 100× magni cation [30].
Iron concentration in the brain tissues and HT-22 cells were assessed according to the manufacturer's instructions of Iron Assay Kit. MDA Assay.
The malondialdehyde (MDA) content in the brain of mice and HT-22 cells were assessed according to the manufacturer's instructions [22].
Western Blot Analysis.
Total protein of brain tissues or HT-22 cells was extracted. Nuclear and cytosolic proteins of HT-22 cells were extracted according to the manufacturer's instructions of the Nuclear and Cytoplasmic protein Extraction Kit [31]. According a previous protocol, the expression level of proteins was measured by Western blot assay [29,32].

Real-time Quantitative PCR (qPCR).
According to the manufacturer's instructions of the Total RNA Extraction Kit, total RNA of HT-22 cells were extracted. Then, the total RNA was reverted in cDNA by using the All-in-One cDNA Synthesis SuperMix kit. Eventually, qPCR was performed with the 2× SYBR Green qPCR Master Mix to measure the expression of mRNA. The gene-speci c primers were mentioned in Table 1. The expression values were normalized to the expression of β-actin and calculated with 2 -ΔΔCt method [33].  [34]. The single-guide RNA (sgRNA), sequence: 5'-AGTCTGGAAAGCGTCACTTG-3', was cloned into lenti-CRISPRV2 plasmid and co-transfected into HEK-293T with psPAX2, pMD2.G and Lipo8000 in order to generate virus particles. Forty eight hours after being transfected, the virus particles in the cell culture uid were ltered with lter membrane (0.45 µm) and infected into HT22 cell line in the presence of Polybrene (10 µg/mL) at 37 ℃. After 48 h, Puromycin (2 µg/mL) was added to KO cells and NC cells to select stable infected cells. The e ciency of knockout was con rmed by Western blotting [35].
Co-IP analysis was carried out according to the manufacturer's instructions of protein A/G magnetic beads (B23201, Bimake, China, Shanghai) [36,37].
Bioinformatics Analysis.
All microarray data were downloaded from gene expression omnibus (GEO) database (accession no. GSE48350). The raw data were downloaded as MINiML les. The extracted data were normalized and processed by log2 transformation. The microarray data were normalized using the preprocessCore package in R software (version 3.4.1). Differentially expressed genes were identi ed using DESeq2 with the adjusted P value < 0.01. Gene expression correlation was analyzed with Spearman test.
Statistical Analysis.
GraphPad Prism 8.0.1 software was adopted for statistical analyses. All data were analyzed by Student t tests. P < 0.05 was considered statistically signi cant. All data are the means ± SD, and the tests were repeated in at least three independent experiments.
First at all, we carried out Morris water maze experiment to test the spatial learning and memory function of mice. As shown in Fig. 1G, there was no difference in the mean swimming speed of all group mice, which means, eriodictyol treatment did not affect the basic motor ability of APP/PS1 mice. On the fth day of platform trials, the escape lantency of APP/PS1 mice was longer compared with WT mice, while eriodictyol treatment signi cantly reduced the escape lantency of APP/PS1 mice (Fig. 1C). Next, the platform was removed to carry out a probe trails. The typical trajectories of each group mice are shown in Fig. 1H. For APP/PS1 mice, the swimming distance and the time spent in the target quadrant were shorter and the number of platform crossings was fewer than WT mice, however, when APP/PS1 mice was intraperitoneally injected with eriodictyol, the proportion of distance and the time spent in the target quadrant was dramatically lengthened and the number of platform crossings was increased ( Fig. 1D-F). As expected, the results of Morris water maze were suggested that APP/PS1 mice show signi cant spatial learning and memory de cits. However, eriodictyol treatment effectively improved these de cits in APP/PS1 mice.
Subsequently, to further investigate whether eriodictyol had a positive effect on memory function of APP/PS1 mice, Y-maze tests were carried out after completion of Morris water maze tests. Consistent with the results of the Morris water maze tests, the results of Y-maze tests showed that cognitive impairment in APP/PS1 mice was signi cantly ameliorated by eriodictyol treatment. To be speci c, in spontaneous alternation tests, compared with WT mice, the percentage of spontaneous alternation and the number of total entries within 8 minutes were reduced in APP/PS1 mice, while eriodictyol memorably reversed it ( Fig. 2A-B). Then in novel arm exploration tests, we also found that the APP/PS1 mice treated with eriodictyol showed a higher percentage of novel arm entries, more time spent in the novel arm and more number of total entries within 5 minutes compared with the mice in APP/PS1-control group (Fig. 2C-E). The representative track of each group of mouse is demonstrated in Fig. 2F. These results of behavioral tests were indicated that eriodictyol can ameliorate cognitive de cits in aged APP/PS1 mice.
To investigate whether eriodictyol had a bene cial effect on Aβ aggregation and Tau hyperphosphorylation which are in APP/PS1 mice, we performed Congo red staining for amyloid plaques, IHE analysis for Aβ protein and phosphorylated-Tau (p-Tau). The result of Congo red staining was shown in Fig. 3A, we observed that eriodictyol was markedly reduce Aβ aggregation in the brain of APP/PS1 mice. Similarly, IHE assay revealed that there was a marked reduction of Aβ protein in the brain of eriodictyoltreated APP/PS1 mice, and the level of p-Tau also reduced by eriodictyol treatment (Fig. 3B). Then, the results of Western blot assay agreed with above studies, the expression level of Aβ and p-Tau in the cortex and hippocampus were reduced by eriodictyol compared with APP/PS1 control mice, despite there was no remarkable change of Tau ( Fig. 3C-D). These fundings revealed that eriodictyol treatment alleviates Aβ aggregation and Tau hyper-phosphorylation in APP/PS1 mice.
Iron accumulation in brain of mice was test using Prussian blue staining. As expected, more iron deposits which stained blue spots by Prussian blue were observed in brain of APP/PS1 mice, while treatment of eriodictyol could effectively reduce the level of iron in APP/PS1 mice (Fig. 4A). Meanwhile, the level of ferrous iron and total iron in cortex and hippocampus of mice were dectected using Iron Assay Kit. The result was shown in Fig. 4B, which suggested that compared with APP/PS1 mice, eriodictyol treatment signi cantly decreased the level of ferrous iron and total iron in cortex and hippocampus of APP/PS1 mice. Western blot analysis demonstrated that the expression of transferrin receptor (TfRC) and ferritin heavy chain (FTH) were reduced upon eriodictyol treatment in cortex and hippocampus of APP/PS1 mice compared with APP/PS1 control mice. While, Fpn (the only iron export protein) was up-regulated in cortex and hippocampus by eriodictyol treated (Fig. 4C). These data imply that eriodictyol treatment could maintain iron balance in cells through reducing iron intake and increasing iron output.
MDA is a speci c marker of lipid peroxidation [6]. It can be found from Fig. 4D that MDA content in cortex and hippocampus of APP/PS1 mice was signi cantly higher than that of WT mice, while treatment with eriodictyol could reverse this tendency.
Finally, we investigated the expression of GPX4 in brain. Western blot analysis con rmed that the expression level of GPX4 was signi cantly increased upon eriodictyol treatment both in cortex and hippocampus of APP/PS1 mice (Fig. 4E).
Together, these results con rmed that eriodictyol effectively inhibited ferroptosis in brain cells of AD model mice.

cells.
HT-22 cell line was used to carry out relevant experiments in vitro. First at all, we determined that the dosage of Aβ 1−42 oligomer was 20 µM to construct cell model of AD according to relevant literatures [6,21]. As shown in Fig. 5A, eriodictyol had no cytotoxicity to HT-22 cells, when the concentration was less than 32 µM.
In order to explore whether eriodictyol relieves the cytotoxicity of Aβ Next, immuno uorometric assay was used to detect the intracellular ROS content in HT-22 cells. Compared with control group, the uorescence intensity was higher in cells treated with Aβ 1−42 oligomer, while that was lower in cells under co-processing of Aβ 1−42 oligomer and eriodictyol (Fig. 6E). Expectedly, the MDA content in the cells was suppressed with treatment of eriodictyol (Fig. 6F).
What's more, the expression of GPX4, a speci c ferroptosis-related gene, was measured by Western blot analysis and qPCR assay. From these results, the protein and mRNA expression of GPX4 in HT-22 cells were down-regulated by Aβ 1−42 oligomer, while eriodictyol treatment could up-regulate the protein and mRNA expression of GPX4 (Fig. 6G-H).
Additionally, the morphological features of Aβ 1−42 oligomer-treated HT-22 cells were observed under electron microscope, including mitochondria appeared rupture and fragmentation with the increasing of membrane density, which consistent with characteristic of ferroptosis reported in Dixon's study [2].
Eriodictyol up-regulates the expression of VDR and activates Nrf2/HO-1 signaling pathway.
Some researchers discovered that nuclear factor erythroid 2-related factor 2 (Nrf2) is the key regulator of ferroptosis [38] , [39]. Moreover, recent study reported that VDR was involved in ferroptosis, and VDR activation could inhibit ferroptosis [40]. In our study, IHE analysis showed the expression level of VDR was reduced in APP/PS1 mice compared with WT mice. In contrast, the expression of VDR in APP/PS1 mice was improved by eriodictyol treatment (Fig. 7A). The results of Western blot were consistent with IHE analysis (Fig. 7B). Then, the expression level of VDR in HT-22 cells was detected by Western blot and qPCR assay, and the result was consistent with vivo ( Fig. 7. C-D).
We further tested the changes of Nrf2/HO-1 signaling pathway. As expected, the expression of phosphorylation of Nrf2 and HO-1 were up-regulated in cortex and hippocampus of APP/PS1 mice upon eriodictyol treatment compared with APP/PS1-control group, while the expression of total Nrf2 was no clear difference. Consistent with the vivo results, Aβ 1−42 oligomer treatment not only weakened the expression of phosphorylated-Nrf2 (p-Nrf2) and HO-1, but also inhibited Nrf2 enter into nucleus. However, when the HT-22 cells were treated with Aβ 1−42 oligomer and eriodictyol together, the expression level of p-Nrf2 and HO-1 were up-regulated, Nrf2 entering the nucleus also was increased (Fig. 7F-G). These data were suggested that eriodictyol promoted the expression of VDR and activated Nrf2/HO-1 signaling pathway.
VDR is the key mediator in the regulation of Nrf2/HO-1 signaling pathway by eriodictyol.
All the above results indicated that VDR and Nrf2/HO-1 signaling pathway participate in the realization of anti-ferroptosis of eriodictyol, but it is incompletely understood that the relationship of VDR and Nrf2/HO-1 signaling pathway. To further understand whether VDR play an indispensable role in the activation of Nrf2/HO-1 signaling pathway by eriodictyol, we used CRISPR/CAS9 technology to knockout VDR in HT-22 cells (Fig. 8A). Then, the sgVDR HT-22 cell was used for CCK-8 assay with normal group. Eriodictyol relieved the reduction of cell viability induced by Aβ 1−42 oligomer, while VDR knockout reversed the protect effect of eriodictyol (Fig. 8B). In addition, we tested the expression of proteins related to AD and ferroptosis. The results were showed in Fig. 8C, the phosphorylation of Tau, was partly restrained in HT-22 cells by eriodictyol treatment compared with Aβ 1−42 oligomer-treated group, but when the VDR was knocked out, the effect of eriodictyol was attenuated. Meanwhile, the facilitation effect of eriodictyol on GPX4 expression was reversed by VDR knockout. Such results suggested that VDR knockout markedly reversed the positive effects of eriodictyol on the HT-22 cell.
Subsequently, we found that knockout VDR blocked the promoting effect of eriodictyol on Nrf2/HO-1 signaling pathway from the results of Western blot assay (Fig. 8C-D). Speci cally, the expression p-Nrf2 and HO-1 were reduced and nuclear translocation of Nrf2 was weakened in HT-22 cells with sgVDR. In addition, to further measured the interaction between VDR and Nrf2, Co-IP assay was carried out. The results con rmed that VDR could interact with Nrf2, and eriodictyol could improve the protein-protein interaction (Fig. 8E).
Finally, to explore the correlation between Nrf2 and VDR in AD patients, bioinformatics analysis was carried out. Gene difference analysis found that there were signi cant differences in the expression of GPX4 in cortex and hippocampus between normal elderly and AD patients, as shown in Fig. 8F. Whereas, the expression levels of VDR didn't show an obvious difference between the two groups. Next, correlation analysis showed that there was a positive tendency correlation between the expression of Nrf2 and VDR (Fig. 8G). These results indicated that VDR is a necessary factor for activating Nrf2/HO-1 signaling pathway and eriodictyol reduces memory impairment Alzheimer's disease by inhibiting ferroptosis, its mechanism is connected with up-regulation of Nrf2/HO-1 pathway which mediated by VDR.

Discussion
In recent years, the number of AD patients has increased sharply in the world, and Alzheimer's disease has become an important factor affecting the quality of life for the aged [1]. Unfortunately, due to the complex pathogenesis and diverse symptoms of AD, there are no effective drugs to treat or prevent AD until now. Eriodictyol, a natural avonoid compound, has showed multiple biological activities, such as anti-in ammatory, antioxidant, anti-radical and neuroprotective effect [13,14,16,17]. However, the mechanism of eriodictyol in AD remains ambiguous. In this study, we found that eriodictyol could alleviate cognitive impairment in APP/PS1 mice and reduced pathological changes of AD both in vivo and vitro. For mechanism, our study implied that the anti-AD effect of eriodictyol was connected with inhibiting ferroptosis of neural cells, which via activating Nrf2/HO-1 signaling pathway mediated by VDR.
Ferroptosis is a novel form of programmed cell death, which have three main characteristics of ferroptosis, including abnormal accumulation of iron, lipid peroxidation, and decrease of GPX4[2] [41].
Firstly, accumulation of intracellular iron is connected with the transport of extracellular iron and the release of intracellular iron. To be speci c, the expression of transferrin (Tf), TfRC and ferritin (indicated by FTH) were increased, while the expression of Fpn, which could transport iron out of cells, was decreased [2]. Secondly, lipid peroxidation showed the increase of MDA and ROS production [41,42]. Thirdly, GPX4 is the key upstream regulator of ferroptosis [43], and studies found that deletion of GPX4 happens in brain which is suffered ferroptosis, which will damage biological macromolecules such as lipids and proteins [44,45]. Up to this point, more and more evidences support that ferroptosis participates in AD [46,47]. There are many studies found that iron accumulates is preferentially observed in the cortex and hippocampus of AD patients using magnetic resonance imaging (MRI) [48][49][50][51][52][53][54], and iron modulates the cleavage of APP and the hyper-phosphorylation of Tau [55]. When iron is overload, the levels of furin protein are reduced, which up-regulates β-secretase activity and then promotes the amyloidogenesis processing of APP, it causes APP depletion and Aβ deposition [56][57][58]. Meanwhile, APP can also stabilize Fpn on the cell membrane to facilitate the e ux of iron from neurons and APP depletion can lead to iron accumulation in cultured neurons and in mouse models [59][60][61]. Additionally, GPX4 activity is of paramount importance in promoting or enabling neuronal survival [62]. This importance is graphically illustrated by an experiment which indicates ablation of GPX4 leads to the rapid degeneration of motor neurons [63]. Hambright et al. also found that GPX4 knockout markedly induces degeneration of hippocampal neurons and cognitive impairment in mice [4]. What's more, the Gene difference analysis in our study showed that the expression level of GPX4 in cortex and hippocampus of AD patients was signi cantly lower than normal elderly. Therefore, it points out that ferroptosis as a potential target for treatment or prevention of AD. In our study, we also found that the morphologic and biochemical hallmarks of ferroptosis appeared in the brain of APP/PS1 mice and Aβ 1−42 oligomer-treated HT-22 cells while eriodictyol could alleviate these changes ( Fig. 4 and Fig. 6). To be speci c, eriodictyol signi cantly reduced the accumulation of intracellular iron by down-regulated TfRC and FTH, up-regulated Fpn, and decreased the the content of MDA and ROS, as well as activated the expression of GPX4. These researches veri ed that ferroptosis was closely related to the pathologies and dependent of AD and suggested that eriodictyol plays an anti-AD role by suppressing ferroptosis.
There are growing evidences shown that Nrf2 and its related signaling pathways play a necessary regulatory role in the process of ferroptosis [64,65]. On the one hand, Nrf2 can maintain cellular iron homeostasis by regulating the expression and activity of ferritin, Tf, TfRC, divalent metal-iron transporter-1 (DMT1), nuclear receptor coactivator 4 (NCOA4), Fpn, and other related proteins and regulatory factors in the process of iron metabolism [64,66,67]. On the other hand, Nrf2 not only regulates the expression and activity of antioxidant enzymes in antioxidant system, such as superoxide dismutase (SOD), HO-1 [39,66], but also improve the expression of GPX4 by maintain the level of glutathione in cells [38,68].
Recently, researchers found that Nrf2 activation could endow on cells with anti-ferroptosis ability. For instance, Yanan Zhao et al. found that knockdown Nrf2 sensitized cell to ferroptosis by down-regulating the expression of Fpn and HO-1 and then reversed the protection of autophagy inhibition in alcoholinduced HepG2 cell [69]. The importance of Nrf2 in ferroptosis was veri ed again in the research carried out by Shuangwen Li's team [70]. Therefore, activating Nrf2 would seem worthy of consideration to inhibit ferroptosis in cells. Additionally, a number of researches proclaimed that the anti-AD effect of eriodictyol was related with activation of Nrf2/HO-1 signaling pathway. For examples, Lv P et al. found that eriodictyol could protect the RGC-5 cells from high glucose-induced oxidative stress, in ammation, and cell apoptosis by activating the Nrf2/HO-1 signaling pathway [71]. Aother study suggested that eriodictyol could protect neurons from the toxic damage of Aβ 25−35 by up-regulating Nrf2/ARE signaling pathway [72]. In our research, we also observed that eriodictyol inhibited pathology features of AD by activating Nrf2/HO-1 signaling pathway, including promoting the expression of p-Nrf2 nuclear translocation of Nrf2, thereby up-regulating the expression of HO-1. Furthermore, a previous study demonstrated that VDR (a karyophilic protein, which mediates 1,25(OH) 2 D 3 to realize the biological functions [73]) also was associated with ferroptosis, it proposed that VDR activation can reverse ferroptosis-related changes in cisplatin induced AKI by trans-regulation of GPX4 [40]. The correlation between VDR and ferroptosis also was veri ed in our study. The results indicated that eriodictyol induced a signi cant increasing of VDR expression both in vivo and in vitro with a dose-dependent manner (Fig. 7). Then, Gene difference analysis of the expression level of VDR in AD patients and the normal aged didn't apparent difference, maybe it's because the number of samples is too small (Fig. 8F). In 2018, a team reported that the promoter sequence of Nrf2 gene contains the VDR binding sites, and proved that Nrf2 transcriptional activation could be promoted by VDR [74]. Automatically, a question was posed: does eriodictyol activate Nrf2/HO-1 signaling pathway mediated by VDR? Our follow-up studies got the answer to this question. In our researches, we found that VDR knockout in HT-22 cells reversed the protective effect and anti-ferroptosis of eriodictyol. Then, the results of Co-IP assay also proved that VDR interacts with Nrf2. Besides, bioinformatic analysis veri ed that VDR has a positive correlation trend with Nrf2. These founding suggested that VDR play an indispensable role in activation of Nrf2/HO-1 pathway by eriodictyol.
According our research, a hypothesis can be put forward ( Fig. 9): eriodictyol alleviated cognitive injury by preventing neuron cell from ferroptosis, its mechanism may be connected with the activation of Nrf2/HO-1 mediated by VDR. However, there are shortages in this research. Firstly, we cannot explore the the anti-AD effect of oral administration of eriodictyol in APP/PS1 mice. Secondly, it is not veri ed that eriodictyol how up-regulate VDR in this study. These researches will carry out in the follow-up study.

Declarations
Ethics approval and consent to participate All animal experiments were conducted with the approval of Chongqing Medical University Animal Welfare Committee.

Consent for publication
Not applicable.
Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
All nancial and non-nancial competing interests must be declared in this section. mg/kg, and intraperitoneal injection three times a week for 3 months three times a week for 3 months.
Data are presented as means ± SD, n = 6. *P < 0.05, ** P < 0.01, ***P < 0.001, vs. APP/PS1-control group.   Eriodictyol inhibits ferroptosis in brain of APP/PS1 mice. (A) Accumulation of iron in brain of mice was strained by Prussian blue staining. (B) The level of ferrous iron and total iron in cortex and hippocampus of mice were dectected using Iron Assay Kit, respectively. (C) The expression of TfRC, FTH and Fpn in cortex and hippocampus of mice brain was measured by Western blot and β-actin was used as loading control. (D) MDA content in cortex and hippocampus of mice brain was detected using Lipid Peroxidation MDA Assay Kit. (E) The expression level of GPX4 in cortex and hippocampus of mice brain was measured by Western blot. The data are shown as the mean ± SD of three experiments. * P < 0.05, ** P < 0.01, *** P < 0.001 compared with the control group.   cells were observed under electron microscope. The data are shown as the mean ± SD of three experiments. * P < 0.05, ** P < 0.01, *** P < 0.001 compared with the Aβ1-42 oligomer group. # P < 0.05, ## P < 0.01, ### P < 0.001 compared with the control group.

Figure 7
Eriodictyol up-regulates the expression of VDR and activates Nrf2/HO-1 signaling pathway. (A) The expression level of VDR in brain of mice was detected by immunohistochemical analysis. Then (B) the protein expression level of VDR in cortex and hippocampus of mice brain was measured by Western blot. control, respectively. The data are shown as the mean ± SD of three experiments. * P < 0.05, ** P < 0.01, *** P < 0.001 compared with the Aβ1-42 oligomer group. # P < 0.05, ## P < 0.01, ### P < 0.001 compared with the control group. are shown as the mean ± SD of three experiments. * P < 0.05 compared with the control group. ### P < 0.001 compared with Aβ1-42 oligomer group. @ P < 0.05 compared with knockout VDR group.

Figure 9
The mechanism of eriodictyol inhibit AD. Eriodictyol alleviated cognitive injury by preventing neuron cell from ferroptosis, its mechanism may be connected with the activation of Nrf2/HO-1 mediated by VDR.