Astragalin Ameliorates Cognitive Dysfunction Through Inhibiting PI3K/Akt-MTOR-Mediated Autophagic Flow in Hippocampal Neurons of APP/PS1 Mice

Astragalin (AST), a natural small molecule �avonoid, can exert anti-oxidant, anti-inammatory and anti-cancer impacts by regulating autophagy. However, the potential mechanism of the neuroprotective effect of AST on neurological disorders such as Alzheimer’s disease (AD) is still not clear. In the present study, we �rstly screened AST for the treatment of AD from the ingredients of Chinese medicines such as Acori tataninowii Rhizoma, Eucommiae Cortex, Paeoniae Radix Alba through the traditional Chinese medicine systems pharmacology database and analysis platform (TCMSP) database. And then we found that AST could improve the cognitive abilities of APP/PS1 mice by Step-down passive avoidance (SDA) and Morris Water Maze (MWM) Test. Further, we identi�ed that AST diminished Aβ plaques deposition in the brains of APP/PS1 mice and Aβ as well as Aβ42 levels in the serum of APP/PS1 mice. Next, microtubule-associated protein 1 light chain 3B (LC3B), p62, Beclin-1, ATG5, ATG12, LAMP-1 were observed to be co-expressed with NeuN in the hippocampus of APP/PS1 mice by immuno�uorescent multiplex staining, while AST was able to activate autophagy and maintain autophagic �ow in hippocampal neurons of APP/PS1 mice by western blot (WB) analysis. Finally, AST reduced the expressions of p-PI3K, p-Akt, p-mTOR by WB analysis. Taken together, we con�rmed that AST may play key neuroprotective effects on APP/PS1 mice by inhibiting the PI3K/Akt-mTOR signaling pathway to activate autophagy and keep autophagic �ow smooth.


Introduction
Alzheimer's disease (AD) is a neurodegenerative disorder characterized by memory loss and progressive cognitive dysfunction [1,2].According to the latest statistics of AD Research Association, the number of deaths caused by AD has exceeded the total number of patients with breast cancer and prostate cancer, ranking rst in the world [3].In contrast, about 9.83 million people aged ≥ 60 years in China suffer from AD [4].The prevalence and incidence of AD are increasing year by year, and at present, AD has become the most common disease that endangers the health of the elderly.So far, the pathogenesis of AD is not thoroughly understood, and therefore there is still a lack of effective drugs for targeted therapy according to AD in clinical practice.Therefore, nding valid strategies and drugs for prevention and treatment AD is of vital importance.
As a degradation mechanism, autophagy is the process of removing damaged and aging organelles, which is essential for maintaining cell homeostasis in the brain [5].The de cits of autophagy bring about the pathogenesis of numerous neurodegenerative diseases, including AD [6].Mounting evidences suggest that AD is accompanied with autophagy dysfunction, resulting in the abnormal accumulation of Aβ plaques that cannot be degraded in the brain, and the regulation of autophagy can facilitate the clearance of Aβ plaques [7][8][9].Autophagic ux which means the integrate process of autophagy, includes autophagy initiation, autophagosome formation, autolysosome fusion and degradation [10], while abnormalities in any of the above links can cause the excessive aggregation of Aβ plaques [11].
Thus, we evaluated all the primary steps of the autophagic process to estimate the in uence of the autophagic ux on the pathological development of AD.
Astragalin (AST), as a natural small molecule avonoid compound [12], is widely distributed in various medicinal plants and fruits, such as Eucommia cortex, lotus leaf and persimmon [13][14][15].Currently, increasing evidences have veri ed AST can be used as a prescription of traditional Chinese medicine owing to its anti-in ammatory [16], anti-cancer [17], and anti-oxidant [18] effects by regulating autophagy.However, whether AST can ameliorate neurological diseases such as AD by modulating autophagy has not yet been reported in related in vivo studies.Consequently, we conducted this experiment to examine the crucial impact of AST on cognitive function of AD mouse models and the autophagy-related mechanism of eliminating Aβ plaques in the brain, and so as to provide animal experiment and theoretical basis for AST as a potential clinical drug for the treatment of AD.

Animals and treatment
A total of 28 male 8-week-old APP Swe , PSEN1 dE9 transgenic mice and 7 male C57BL/6 (WT-like littermates) mice were obtained from Guangdong Medical Laboratory Animal Center, China (Permit Number: SCXK GUANGDONG 2018-0002).The APP/PS1 transgenic mice were reared in a single cage, and all the animals were housed under a standard condition of an even dark/light cycle at 25 ℃.After adapting to the environment for one week, the APP/PS1 transgenic mice were randomly divided into: the APP/PS1 group, 10 mg/kg AST (APP/PS1+AST 10) group, 20 mg/kg AST (APP/PS1+AST 20) Group, 40 mg/kg AST (APP/PS1+AST 40) group, and the C57BL/6 mice were served as the WT group.AST was freshly prepared in dimethyl sulfoxide (DMSO) to make working solutions at concentrations of 10, 20 and 40 mg/kg [19,20].Subsequently, these mice were administrated orally with AST treatment once daily for one month.The mice of WT group and APP/PS1 group were gavaged with an equal volume of saline con guration of 0.1% DMSO.All experimental protocols in this study were approved by the Ethics Committee of Guangdong Pharmaceutical University and were complied with the National Institutes of Health Guide for the Care and Use of Animals.
Step-down passive avoidance (SDA) test Before and after the administration, all animals were subjected to the SDA test.The SDA test was used to assess the learning and memory of rodents [21,22].The experimental device was composed of an avoidance reaction box (15×15×46 cm 3 ) and a parallel steel rod with energization at the bottom.A circular insulating platform (4.5 cm in a diameter, 4.5 cm in a height) was placed in the middle of the box.
The mice of each group were put into the reaction box to adapt to the environment for 3 min, and then electri ed.The mice were stimulated by the electric current (36 V, 0.5 mA, 5 min) and jumped back to the platform to avoid noxious stimulation.Animals were trained for 3 consecutive days and the experiment was formally started after 24 h.The retention of memory was assessed by measuring the time when the mice rstly stepped down the platform (step down latency) and the number of times the mice jumped off the platform within 5 min (number of errors).The schedule of SDA test is shown in Fig. 1.

Morris Water Maze (MWM) Test
Before and after the treatment of AST, the MWM behavioral experiment was conducted to evaluate the spatial memory and learning and memory abilities of mice [23].The MWM includes a mouse behavior test system, an image automatic monitoring and processing system, and a data processing analysis system.Tests were carried out in a circular tank (120 cm diameter and 40 cm deep) containing opaque water maintained at 25 ± 2°C.Tests include positioning navigation test, space exploration test and visible platform test.All mice were subjected to a 5-day hidden platform test and were regularly trained 4 times a day.Removing the platform on day 6, the mice were placed in the water from the same quadrant and then the time of mice reaching the original platform quadrant within 1min were recorded.On day 7 and 8, the platform was raised to a position 1 cm above the water surface and placed in the other quadrant, and then the swimming track and the times of mice crossing the target platform quadrant within 1 min were recorded.The schedule of MWM test is shown in Fig. 1.

Fluorescent imaging of Aβ plaque in the brain of mice
After completing drug treatment and behavioral studies, the mice were anesthetized and perfused with 4% paraformaldehyde (PFA) through the aorta, and then brain tissues were taken out for xation and washed with owing water.After gradient dehydration, para n embedding and tissue sectioning, the brain sections of mice in each group were selected for routine dewaxing and dehydration.Then 0.01 M citrate buffer (PH: 6.0) was used to antigen retrieval at high-temperature for 20 min.After blocking the binding site of non-speci c antibody, mouse anti-Aβ1-42 (1:300) was incubated overnight at 4°C.Then the sections were washed by PBST and Goat anti-mouse IgG H&L (Alexa Fluor® 594) (1:800) was added and incubated at 37°C for 1 h.After washing by PBST, the sections were dyed in 0.3% thio avin S solution for 15 min, and then washed in 70%, 50% I and 50% II ethanol solutions for 5 min.Next, 50% glycerol was used for sealing [24].Sections were observed by a uorescent microscope.Image J (NIH, Bethesda, MD, USA) software was used to determine the number of plaques.

Determination of Aβ and Aβ42 in the serum of mice by ELISA
To detect Aβ and Aβ42 levels in serum of mice, the blood of mice was collected before sacri cing mice.
The serum was then separated by centrifugation at 1000 rpm for 10 min at 4℃.Subsequently, the levels of Aβ and Aβ42 in serum of mice were measured by mouse Aβ ELISA kit and mouse Aβ42 ELISA kit.At last, the absorbance A under 450 nm wave was measured by microplate reader, and the samples concentration were calculated by Curve Expert 1.4 (Hyams DG, Starkville, MS, USA).

Bioinformatics analysis
The traditional Chinese medicine systems pharmacology database and analysis platform (TCMSP) (https://old.tcmsp-e.com/tcmsp.php)database was searched to obtain the chemical compositions of the therapeutic AD herbs, and then aggregated into a table.The tested gene proteins were imported into the String Gene Interaction Network database (https://string-db.org/)and the species was selected as "Homosapiens".The selected targets were searched in "Multiple proteins" and the PPI network map was drawn.

Statistical analysis
All experiments were repeated at least three times.And all the experiments included a minimum of three mice per group.Graphpad Prism 8.0 (GraphPad Software Inc., California, USA) and SPSS 23.0 software (IBM, Armonk, NY, USA) were used to analyze these data.All of the data were expressed as mean ± standard deviation (SD).Quantitative data meet the normal distribution, and after passing the homogeneity test of variance, one-way ANOVA or two-way ANOVA was used to compare groups followed by Turkey or Student-Newman-Keuls test.For comparison between two groups, statistical signi cance was determined by two-tailed unpaired t-test.And the P < 0.05 was considered to be signi cant.

Results
AST exists in most Chinese herbal medicines for the treatment of AD Firstly, we used the TCMSP database and set the screening condition to the drug-like property (DL) ≥ 0.18, which yielded that among the Chinese herbal medicines used in the treatment of AD, Acori tataninowii Rhizoma [25], Eucommiae Cortex [26], Paeoniae Radix Alba [27], Carthami Flos [28], Forsythiae Fructus [29], Achyranthis Bidentatae Radix [30], Granati Pericarpium [31], and Epimrdii Herba [32] contained AST (Table 1).Learning and cognitive dysfunction in the APP/PS1 mice Firstly, we evaluated the cognitive de cits related to AD by using the SDA test and MWM test, and then selected APP/PS1 mice needed for this experiment (Fig. 2a, b).In the SDA test, it was observed that the APP/PS1 mice showed shorter step-down latency (time to jump off the platform for the rst time) than WT mice (p<0.001,Fig. 2c,).Meanwhile, the number of errors made by APP/PS1 mice were much higher than that of WT mice (p<0.001,Fig. 2d).
The MWM results showed that the escape latency (time to nd the target platform) of APP/PS1 mice was signi cantly longer than WT mice during the hidden-platform period, especially from Day 5 (p<0.001,Fig. 2e).During the space exploration test, APP/PS1 mice took much longer time to visit the target quadrant compared with WT mice (p<0.001,Fig. 2f).After the platform was raised to 1 cm above the water surface, it was observed that the times of APP/PS1 mice crossing the quadrant of the target platform were signi cantly less than WT mice (p<0.001,Fig. 2g, h).Together, the above results indicated that APP/PS1 mice had learning, memory and cognitive dysfunction.
AST treatment ameliorates cognitive impairments of APP/PS1 mice Next, to explore whether AST could improve the cognitive de cits of APP/PS1 mice, we conducted the SDA test and MWM test once again.The results of the SDA test showed that compare with WT mice, APP/PS1 mice had a shorter step-down latency and an increase in the number of errors ( p<0.001, Fig. 3a,  b), while these changes were reversed after treatment with 20, 40 mg/kg AST ( p<0.01, Fig. 3a, b).
In MWM test, during the training period the APP/PS1 mice showed much longer escape compared with WT mice, while AST signi cantly reduced the escape latency of APP/PS1 mice, especially from Day 5 (p<0.001,Fig. 3c).During the trial period, the APP/PS1 mice spent signi cantly much longer time to nd the target quadrant and reached the target platform much less frequently when compared with the WT mice and the mice treated with AST (p<0.01,Fig. 3d, e, f).In summary, the results revealed that AST signi cantly improved the cognitive de cit of APP/PS1 mice.
AST treatment reduces Aβ accumulation of APP/PS1 mice In the following, we evaluated Aβ plaques deposition in the brain of APP/PS1 mice after treatment with different doses of AST drugs.As shown in the Fig. 4a, compared with WT mice, there were a large amount of the depositions of Aβ plaque in the brain of APP/PS1 mice (p<0.001,Fig. 4a, b).However, after treatment with AST, Aβ plaques were considerably reduced in a dose-dependent manner, especially after treatment with 40 mg/kg AST (p<0.001,Fig. 4a, b).
Then we further assessed the impacts of AST on the levels of soluble and insoluble Aβ and Aβ42 in the serum of mice by using ELISA assay.By contrast with WT mice, Aβ and Aβ42 levels in the serum of APP/PS1 mice were more higher (p<0.001,Fig. 4c, d).While after being treated with 20 and 40 mg/kg AST for one month, Aβ and Aβ42 levels in serum of APP/PS1 mice were decreased (p<0.01,Fig. 4c, d).
AST treatment activates autophagy in APP/PS1 mouse hippocampal neurons By considering that AST treatment could attenuate cognitive function and Aβ plaques deposition of mice, we next investigated whether AST-mediated effects could regulate autophagy and autophagic ux.Firstly, we detected the levels of LC3B, p62 and Beclin-1 involved in the initiation of autophagy by immuno uorescent staining and WB analysis.The results of immuno uorescence showed that LC3B positive cells were mainly located in the cytoplasm of mice hippocampal neurons (Fig. 5a), p62 positive cells were mainly located in the in the cytoplasm and nucleus of mice hippocampal neurons (Fig. 5b), and Beclin-1 positive cells were mainly located in the cytoplasm and membrane of mice hippocampal neurons (Fig. 5c).The levels of LC3B and Beclin-1 in APP/PS1 mice hippocampus were lower than those in WT group, while the level of p62 was higher (p<0.01, Figure 5d, e, f).Notably, treated with AST, the levels of LC3B and Beclin-1 increased to varying degrees, while the p62 level decreased in a different degree (p<0.05,Fig. 5d, e, f).Among them, the 40 mg/kg AST treatment was the most obvious (p<0.01, Figure 5d, e, f).Collectively, all abovementioned results showed that AST treatment activated autophagy of hippocampal neurons in APP/PS1 mice.
AST treatment promotes autophagosome formation and autophagolysosome degradation in hippocampal neurons of APP/PS1 mice Next, immuno uorescent staining and WB assays were performed in hippocampal of mice to detect the expressions of ATG5 and ATG12 related proteins involved in autophagosome formation.As shown in the Fig. 6: ATG5 positive cells were mainly located in the cytoplasm of mice hippocampal neurons (Fig. 6a), and ATG12 positive cells were mainly located in the cytoplasm of mice hippocampal neurons (Fig. 6b).The similar phenomenons were found ATG5 and ATG12 in the hippocampus of the APP/PS1 group were obviously lower than those in WT group (p<0.01,Fig. 6d, f, g).Meanwhile, with the AST treatment, the ATG5 and ATG12 were e ciently upregulated in a dose-dependent manner, among which treated with 40 mg/kg AST was especially meaningful (p<0.01,Fig. 6d, f, g).Taken together, the results indicated that AST treatment could promote the formation of autophagosomes in the hippocampal neurons of APP/PS1 mice.
Finally, to further investigate the impact of AST on the degradation of autophagy lysosomes, we also inspected the level of lysosomal membrane protein LAMP-1 by immuno uorescence and WB assays.As indicated in Fig. 6c, LAMP-1 positive cells were mainly located in the cell membrane of mice hippocampal neurons (Fig. 6c).Compared with the WT mice, the level of LAMP-I protein in the hippocampus of APP/PS1 mice was decreased signi cantly (p<0.01,Fig. 6e, h).In contrast, while the treatment with 10, 20 and 40 mg/kg AST enhanced the level of LAMP-I protein, among which 40 mg/kg AST showed the most obvious effect (p<0.001, Fig. 6e, h).Together, the above results demonstrated that AST treatment promoted the degradation of autophagy lysosomes in hippocampal neurons of APP/PS1 mice.
PI3K/Akt-mTOR is involved in the neuroprotective effect of AST on APP/PS1 mice The PI3K/Akt-mTOR pathway plays various important roles in the central nervous system and is closely related to the pathogenesis of AD [33].Therefore, we rstly used STRING for protein interaction network analysis and found that PI3K/Akt-mTOR pathway-related proteins were interacted with related proteins that regulated autophagic ow (Fig. 7a) (Source: 9 items (mouse) -STRING interaction network (stringdb.org)).In order to evaluate whether AST could regulate these proteins to ameliorate the pathological behavior of APP/PS1 mice, we then used WB to detect the proteins expression and phosphorylation forms of PI3K, Akt, and mTOR.As shown in the Fig. 7, p-PI3K, p-Akt and p-mTOR in the hippocampus of APP/PS1 mice were signi cantly increased with a varied degree compared to those in the WT mice (p<0.01,Fig. 7b, c, d, e).Consistent with expectations, treated with AST, the expressions of p-PI3K, p-Akt and p-mTOR in the hippocampus of APP/PS1 mice were distinctly reduced, especially when treated with 40 mg/kg AST (p<0.01,Fig. 7b, c, e), suggesting that the PI3K/Akt-mTOR signaling pathway was inhibited after AST treatment.

Discussion
In the present study, we demonstrated the learning and memory functions were impaired in APP/PS1 mice by SDA and MWM behavioral experiments.Subsequently, we inquired the TCMSP database to ascertain that most of the Chinese medicines used to treat AD contained AST.Based on this discovery, we intraperitoneal injection of AST in APP/PS1 mice for one month noticed that AST treatment productively accelerated the clearance of Aβ plaque deposits in the brains of APP/PS1 mice and further restored their cognitive dysfunction.Importantly, we revealed AST could active autophagy and keep autophagic ow smooth of APP/PS1 mice hippocampal neurons for neuroprotective effect by inhibiting PI3K/Akt-mTOR pathway (Fig. 8).The above ndings indicate AST might be a potential trerapeutic agent for future clinical AD treatment.
AD, as a common progressively neurodegenerative disorder, endangers the memory and cognitive abilities of people around the world [34,35].As the universal animal model of AD, APP/PS1 mice stably display mouse/human APP located in neurons and mutant human PS1 and there are senile plaque deposits in the brain at the age of 6 months, and memory ability decline with age [36].Our data in this study manifested that APP/PS1 mice had severe memory deterioration, which is consistent with several related researches [37][38][39].The brains in AD showing excessive senile plaque depositions are extensively regarded as one of the earliest changes [40].Concurrently, we observed that, dissimilar to WT mice, there were numerous Aβ plaque deposits in the brain of APP/PS1 mice.Interestingly, we found the contents of Aβ and Aβ1-42 which were the most common and toxic [41,42] in the blood of APP/PS1 mice were much higher than those of WT mice.Hence, inhibiting the aggregation of Aβ and reducing the content of Aβ and Aβ1-42 in the serum of mice may be a hopeful treatment strategy for AD.
Currently, traditional Chinese medicines play an indispensable performance in the prevention and treatment of AD in domestic and international studies.Thus, we screened out AST, a potential new natural avonoid isolating from traditional Chinese medicines for AD treatment [43], by consulting a large amount of literatures and operating the TCMSP database.In the central nervous system, AST further exerted a good hypnotic effect by prolonging the convulsion latency and diminishing the convulsion rate of mice [44].Concurrently, AST showed neuroprotective impact on PC12 cells and SH-SY5Y by enhancing cell viability and weakening intracellular oxidative stress [45].Additionally, the experimental evidence of Chung et al. proved that AST could prevent H2O2-induced cell death by inhibiting JNK, p38 and ERK 1/2 pathways in SK-N-SH cell [46].However, studies on the function of AST in neurodegenerative diseases such as AD are comparatively less frequent and solely concentrated on the cell levels in vitro.In the previous investigation of AD in vitro, it has been reported that AST could shorten the intracellular tau protein level and alleviate the cell damage through pharmacological analysis [47].Here, we administered AST to APP/PS1 mice and primarily con rmed that AST was validly in recovering impaired cognitive functions in APP/PS1 mice, including learning and memory and spatial navigation.Subsequently, we were surprised to discover that abundant and widespread Aβ plaques in the brains of APP/PS1 mice were productively abolished, as well as Aβ and Aβ1-42 levels in the serum of mice.Notably, the therapeutic effect of 40 mg/kg AST treatment was the most visible in this process.In line with those, we further discussed the relevant mechanisms of AST on the neuroprotective infection of APP/PS1 mice.
From early to late AD, the autophagy of hippocampal neurons is continuously disrupted and the process of autophagic ow is blocked, bringing about the ine ciency to eliminate the pathogenic Aβ plaques accumulated in the cytoplasm, thus exhibiting high toxicity to cells [48][49][50].Revamping hippocampal neuronal autophagy functional impairment may be of indispensable signi cance in postponing aging and hampering neurodegenerative diseases such as AD [51].In the present research, we revealed NeuN, a labeled neuronal protein, was co-expression with autophagy-related proteins involved in the whole autophagic ow by immuno uorescent multiple staining experiments, which further suggested the presence of autophagy in mouse hippocampal neurons.LC3B presents in autophagosomes in the form of LC3B-and LC3B-, while the level of the lapidated form of LC3B-is a factual measure of autophagic ux relative to the unprocessed LC3B- [52,53].SQSTM1/p62 is a ubiquitin-binding protein that can be e ciently degraded by autophagy, and thus, the level of p62 is another indicator of autophagic ux [54,55].The WB analysis demonstrated that LC3B-II levels in hippocampal neurons of APP/PS1 mice were much lower than those of WT mice, whereas p62 levels showed the opposite trend, which could be the result of blocked autophagic ow pathways.Importantly, the two changes were reversed again with treatment of 20 and 40 mg/kg AST, indicating that autophagy was activated.Owing to autophagic ow is a dynamic process, as a meaningful protein for autophagy initiation, Beclin-1 mediates the positioning of other autophagic proteins, and its increased levels indicate autophagic activation [56].Here, we determined AST predominantly enhanced the decreased Beclin-1 levels coincided with the ability of AST to ensure that the autophagic ow initiation phase was normalized again in the hippocampus of APP/PS1 mice.ATG5 gene exerts an imperatively character in the extension of autophagosomal membrane as well as autophagosome formation [57].Meanwhile, the role of ATG12, an essential gene of the ATG12-ATG5 coupling system, can not be underestimated [58].In this study, ATG5 and ATG12 in the hippocampus of APP/PS1 mice shifted expeditiously from lower to higher after AST treatment, in a sense revealing that AST established the proper functioning of the intermediate link of autophagic ow (autophagosome formation phase).The inability of autophagosomes to fuse with lysosomes to further form autophagic lysosomes, which in turn break down harmful substances, can also lead to serious consequences [59].The reduction of LAMP-1 (a lysosomal membrane-bound protein that works in autophagic lysosome fusion) in our study indicated that autophagic lysosome formation was arrested in the hippocampus of APP/PS1 mice [60-62].Novertheless, AST bolstered the fusion of autophagosomes with lysosomes by extending LAMP-1 level, thereby accelerating the clearance of toxic substances in the brain of APP/PS1 mice.The above results indicated in the APP/PS1 mice hippocampal neurons AST could active autophagy and keep autophagic ow smooth by stimulating the degradation of p62, augmenting the formation of LC3B and Beclin-1, up-regulating ATG5, ATG12 and LAMP-I.
We hypothesized that AST upregulated the level of autophagy and controlled autophagic ow in the hippocampus of APP/PS1 mice by restraining the PI3K/Akt-mTOR pathway.mTOR as known to be the major negative regulator of autophagy, is monitored by PI3K/Akt kinase cascade response, and when it is inhibited, the autophagic activity is enhanced [63,64].Numerous researches have shown that super uous activation of the PI3K/Akt-mTOR signaling pathway made for the inhibition of neuronal autophagy level and the inability to clear intracellular accumulation of Aβ and tau proteins in a timely manner, which to some extent aggravated amyloid plaque production and neuro brillary tangles in the AD brains [33,65].In contrast, inhibition of PI3K/Akt-mTOR pathway was able to scale down Aβ aggregation and improve cognitive function of AD mice [66,67].In our study, equivalently, above mentioned pathway was aberrantly mobilized in the hippocampus of APP/PS1 mice compared to normal mice, whereas AST stimulated autophagy and sustained either process of autophagic ow unimpeded by inhibiting the PI3K/Akt-mTOR pathway.
In conclusion, our study demonstrated that AST could exert the partial neuroprotective effect on APP/PS1 mice by facilitating PI3K/Akt-mTOR-mediated autophagy ux pathway.And our data provides more theoretical support for the application of AST to clinical AD effective drug candidates.However, in the current study we didn't clearly clarify which key link autophagic ux was regulated by AST to exert neuroprotective effects.Therefore, we need to explore it further in the future.Data were presented as mean ± SD, n= 7 in each group, * p < 0.05, ** p < 0.01 or *** p < 0.001 vs WT group, # p < 0.05, ## p < 0.01 or ### p < 0.001 vs APP/PS1 group, one-way and two-way ANOVA Multiple comparisons were used for statistical analysis.

Figures
Figures

Table 1
Part of the components of Chinese herbal medicines for the treatment of AD (L)-alpha-Terpineol, dec-2-enal, copaene, isoliquiritigenin, DFV, Astragalin DeclarationsYuxin Ma and Cuizhu Yang designed the study and wrote the paper.Cuizhu Yang, Runheng Zhang and Shuhan Wang conducted the research.Yinghong Tian, Yaqi Yang and Jing Liu provided technical guidance.