A Network Pharmacology-Based Study on the Mechanism of Dibutyl Phthalate of Ocimum basilicum L. against Alzheimer's Disease through the AKT/GSK-3β Pathway

Background Ocimum basilicum L. (OBL) is mainly used to treat neurological diseases in China. The preliminary work of this group showed that OBL improves cognitive impairment in Alzheimer's disease (AD). However, the underlying pharmacological mechanism remains unclear. Methods The components of OBL were compiled by literature search, and their active ingredients were screened by online database. The drug targets of OBL in the treatment of AD were predicted and analyzed using information derived from sources such as the SwissTargetPrediction tool. And through the network visual analysis function of Cytoscape software and protein-protein interaction analysis (PPI), the core targets of OBL treatment of AD are predicted. Furthermore, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) were employed to analyze the related signaling pathways affected by OBL. Moreover, AutoDock software was used to assess the potential binding affinity between the core targets and the active compounds. Subsequently, in vivo experiment was conducted to verify the findings of network pharmacology. Results A total of 35 active compounds and 188 targets of OBL were screened, of which 43 common targets were related to AD. The active compounds of 35 OBLs induced 118 GO and 78 KEGG. The results of PPI and network topology parameter analysis show that targets such as MAPK1, GSK3B, NR3C2, ESR1, and EGFR are known as the core targets for the treatment of AD by OBL and are docked with the active ingredients of OBL. Molecular docking results suggest that diterbutyl phthalate (DBP) may be the main active component of OBL for the treatment of AD. Flow cytometry analysis results showed that apoptosis decreased with increasing DBP dose. In addition, DBP significantly decreased the levels of lactate dehydrogenase (LDH) and reactive oxygen species (ROS) in the supernatant of Aβ25-35-induced injury HT22 cell cultures, and it can be speculated that DBP has the ability to protect the stability of injured neuronal cells and improve the permeability of cell membranes, thus stabilizing the intracellular environment. Mechanistically, DBP may increase the mRNA levels of AKT, GSK-3β, etc. in AD cell models and regulate the phosphorylation of AKT/GSK-3β pathway-related. Conclusions Conclusively, our study suggests that DBP, the main active component of OBL, has potential in the prevention or treatment of AD.


Introduction
Alzheimer's disease (AD) is a neurodegenerative disease characterized by amyloid protein (A) deposition and neurogenic fiber tangles [1]. Data shows that the number of people over 65 years of age with AD is expected to reach 6.7 million in 2025 [2]. Patients with AD present with symptoms such as short-term memory loss, language difficulties, and disorientation [3].There are many different theories on the pathogenesis of AD, such as the A cascade hypothesis, genetic mechanisms, inflammatory mechanisms, mitochondrial dysfunction, neurotransmitter dysregulation, glycogen synthase kinase-3 (GSK-3), and oxidative stress, but the pathogenesis has not been fully elucidated [4].
Research on AD is very limited, and active exploration of effective therapeutic drugs for AD has become an important research direction for therapeutic targets in recent years.
Ocimum basilicum L. (OBL), a genus of basil in the family Labiatae, has pharmacological effects such as free radical scavenging, anti-inflammatory, antitumor, antibacterial, hypolipidemic, and antiatherosclerotic [5]. OBL has been reported to have therapeutic activity against neurological disorders such as depression, anxiety, and sedation, mainly through anti-inflammatory and antioxidant properties [6]. Studies have reported the ameliorative effect of OBL volatile oil on neurodegenerative changes in mice caused by chronic unpredictable mild stress (CUMS), but the therapeutic effect of OBL on AD has not been studied and reported [7]. OBL ameliorates memory and neurological deficits after ischemia-reperfusion-induced brain injury in mice [6]. The composition of OBL is complicated and variable, and further research is needed to screen for active compounds with therapeutic effects, as well as to understand the role and mechanism of the compound in the treatment of AD.
In this study, we investigated the active components, potential targets, and signaling pathways of OBL for the treatment of AD through network pharmacology, which is consistent with the therapeutic principles of Chinese medicine and ethnic medicine for the treatment of complex diseases and then searched for the best match between OBL small molecule compounds and target proteins by complementing the receptor active sites of AD disease targets with spatial structure and minimizing the binding energy. The monomeric component diterbutyl phthalate (DBP) was screened from OBL. In vitro experiments showed the effect of DBP on HT22 cells with Aβ 25-35induced damage. The material basis of the action of OBL and its mechanism of action was investigated to provide some theoretical basis for the treatment of clinical AD at a later stage, see Figure 1.

Collection and Collation of OBL Components.
The OBL chemical components were collected according to the following criteria: (1) regional OBL component that has been publicly reported in the literature within the last 5 years and (2) chemical components that have been quantified in OBL. The collected components were preprocessed and standardized to remove outlier samples and redundant molecular descriptions, and finally, the component names were entered into the PubChem database (https://www .chemicalbook.com/) to retrieve normalized 3D molecular structure descriptors for subsequent data analysis.

Screening of OBL Candidate Components and Their
Related Targets. The OBL-normalized 3D molecular structure descriptors were compiled according to the above criteria and entered into the SIB database (http://www .swissadme.ch/index.php) to derive pharmacokinetic parameters related to the chemical composition, and the biologically active components were selected for further study based on ADME parameters. In SwissADME, the high gastrointestinal absorption is indicated by the OB. The high OB value indicates the important index of pharmacodynamic molecules and drug-like properties. In this paper, high gastrointestinal absorption and bioavailability scores 0.55 were used as criteria for screening. The higher OB value is a key index indicating the potent molecules and drug-like properties. Pharmacodynamic effects can influence the ADME process and therefore lead to changes in drug bioavailability [8].
To estimate the drug similarity of each ingredient, pharmacokinetic parameters were calculated based on the model in the Pipeline Pilot ADMET collection. The obtained ingredients' English names were imported into the TCMIP V2.0 database (http://www.tcmip.cn/TCMIP/index.php) of herbal ingredients for searching to obtain the level of drug-likeness weight [9]. A quantitative index called the quantitative estimate of drug-likeness (QED) was used to assess drug similarity, and the estimated values ranged from 0 to 1. The mean QED values for drug-likeness were 0.49 and 0.67. If QED > 0:67 indicates good drug-likeness, 0:49 ≤ QED ≤ 0:67 indicates if QED > 0:67, it means that the drug forming property is good, 0:49 ≤ QED ≤ 0:67 means that the drug forming property is medium, and QED > 0:67 means that the drug forming property is weak. In this paper, 0:49 ≤ QED ≤ 0:67 was used as the screening criterion for screening. The achieved components were considered as candidate components of OBL, and the candidate components were collated through the SwissTargetPrediction (http://www.swisstargetprediction.ch/) database for their targets of action.
2.4. Drug-Disease Target Association Analysis. The OBL component targets were intersected with AD disease targets, and the intersected targets were imported into the STRING database to obtain the interaction relationship between the targets. With "homo sapiens" selected as the species, "minimum required interaction score" was selected as ≥0.7, and the rest of the default parameters were used. A protein-protein interaction (PPI) association network was constructed, and isolated nodes were deleted to obtain the initial network. The CytoNCA (2.1.6) plug-in was then used to determine the betweenness centrality (BC), closeness centrality (CC), eigenvector centrality (EC), degree centrality (DC), local average connectivity-based method (LAC), network centrality (NC), and other topological attributes as criteria for 2-step screening to simplify the network, screening different target clusters, and the topranked target clusters as key target clusters for OBL treatment of AD.
2.5. Gene Enrichment Analysis. In order to systematically elucidate the role of OBL in the treatment of AD, the intersecting targets of OBL for AD were subjected to GO (Gene Ontology) enrichment analysis and KEGG (Kyoto Encyclopedia of Genes and Genomes) signaling pathway enrichment analysis. The active component-target-pathway network map was constructed by Cytoscape 3.7.1 software (version: 3.7.1, https://cytoscape.org).    [10]. HT22 cells in good growth condition were digested with trypsin, prepared into 5 × 10 4 cells/mL single cell suspension with complete medium, inoculated into 96well plates (100 μL/well, i.e., 5 × 10 3 cells/well), incubated for 24 h at 37°C with 5% CO 2 for wall attachment, the medium was discarded, and 100 μL of Aβ 25-35 at a final concentration of 173.568 μmol/L (IC 50 ) was added separately. L (IC 50 ) of Aβ [25][26][27][28][29][30][31][32][33][34][35] and different concentrations of DBP (0, 1, 5, 10, 20, and 50 μmol) were added, while a blank control group was set up. After 48 h of intervention, the supernatant was collected for LDH assay, while 100 μL of the configured 10% CCK-8 solution was added to each well, and the incubation was continued in the incubator for 1 h. After 1 h, the OD value at 450 nm was measured by enzyme marker. OD value at 450 nm was measured by ELISA after 1 h. The results of CCK-8 and LDH assay were combined to screen the low, medium, and high intervention concentrations of DBP for subsequent experiments.

Pretreatment and
2.7.4. LDH and ROS Detection. HT22 cells in good growth state were taken, cells were digested with trypsin, prepared into 5 × 10 4 cells/mL single cell suspension with complete medium, inoculated into 96-well plates (100 μL/well, i.e., 5 × 10 3 cells/well) and 6-well plates (2 mL/well, i.e., 1 × 10 5 cells/well), incubated at 37°C and 5% CO 2 for 24 h for wall attachment, and discarded medium, and the interventions were performed according to experimental groups, with 3 replicates per group. After the intervention was completed, the medium was discarded and the cells were collected for LDH and ROS detection (fluorescence detection wavelength setting: optimal excitation wavelength 500 nm and optimal emission wavelength 525 nm).
2.7.5. Cell Cycle Detection by Flow Cytometry. HT22 cells in good growth condition were digested with trypsin, prepared into 5 × 10 4 cells/mL single cell suspension with complete medium, inoculated into 25 cm 2 culture flasks, and incubated in a 37°C, saturated humidity, 5% CO 2 cell incubator for 24 h. After the intervention according to 2.6.1 experimental grouping, the cells in each group were digested with trypsin, washed with 5 mL of PBS, and fixed overnight at 4°C. Once, resuspend the cells with 500 μL of precooled PBS, add the cell suspension to 3.5 mL of precooled 80% ethanol, and fix overnight at 4°C. Centrifuge at 2000 rpm for 5 min to precipitate the cells. Carefully aspirate the supernatant to avoid aspirating the cells. Wash 2 times with precooled PBS and discard the clean supernatant. Add 500 μL PI/RNase Staining Buffer to resuspend the cells and pass through a 200 mesh nylon sieve to make a single cell suspension, incubate for 30 min at 4°C, protected from light, detect red fluorescence at an excitation wavelength of 488 nm with a flow cytometer, and detect light scattering. Analysis software was used for cellular DNA content analysis and light scattering.
2.7.6. Flow Cytometry Detection of Apoptotic Cells. HT22 cells in good growth condition were digested with trypsin, prepared into 5 × 10 4 cells/mL single cell suspension with complete medium, inoculated into 25 cm 2 culture flask, and incubated in 37°C, saturated humidity, 5% CO 2 cell culture chamber for 24 h. After the intervention according to the experimental grouping in 2.6.1, the culture fluid in the cell flask was aspirated into the centrifuge tube after the intervention was completed (containing the cells that were washed twice with PBS), and the PBS was collected together into the centrifuge tube. The cells were digested by trypsin, transferred to the centrifuge tube, and centrifuged at 1000 rpm for 5 min, and the supernatant was discarded. Wash the cells twice with precooled PBS and discard the supernatant. Add 500 μL of 1 × Binding Buffer to resuspend the cells and pass through a 200 mesh sieve to make a single cell suspension. Add 5 μL Annexin V-PE and 10 μL 7-AAD to each tube, mix gently, and leave for 5 min at 4°C protected from light. Flow cytometry assay was performed within 30 minutes.     BioMed Research International 2.7.8. Detection of AKT and GSK-3β Gene Expression by qRT-PCR. HT22 cells in good growth condition were digested with trypsin and prepared into 5 × 10 4 cells/mL single cell suspension with complete medium, inoculated into 25 cm 2 culture flasks and incubated in 37°C, saturated humidity, 5% CO 2 cell culture incubator for 24 h. The intervention was carried out according to experimental groups, with 5 replicates in each group. After the intervention was completed, the medium was discarded and 1 mL of TRIzol was added to digest the cells so that TRIzol lay flat on the cell level, and the cell culture flasks were repeatedly shaken until the cells were digested down and loaded into 1.5 mL EP tubes. Follow-up experiments were performed according to the qRT-PCR lab report.
2.7.9. WB Detection of AKT, p-AKT, GSK-3β, and p-GSK-3β Protein Expression. HT22 cells in good growth condition were digested with trypsin, prepared into 5 × 10 4 cells/mL single cell suspension with complete medium, inoculated into 25 cm 2 culture flasks, and incubated in 37°C, saturated humidity, 5% CO 2 cell incubator for 24 h. The intervention was carried out according to experimental groups, with 3 replicates in each group. After the intervention was completed, the culture medium in the flask was discarded,    11 BioMed Research International 3 mL of sterile PBS buffer was added and repeatedly rinsed twice, the PBS buffer was discarded, the cells were digested with trypsin, and the operation method was the same as cell passaging. After centrifugation, the supernatant was discarded and the cell precipitate was left, and the cells were collected after washing with 5 mL of sterile PBS. The subsequent experiments were performed according to the Western Blot experiment report.
2.7.10. Data Processing. All data were expressed as mean ± standard deviation ( x ± SD), and each group's data was statistically analyzed using SPSS 19.0 software. If the data conformed to normal distribution, one-way ANOVA was used for statistics: for Chi-squared, LSD method was used for multiple comparisons; for Chi-squared, Tamhane's method was used for multiple comparisons, and P < 0:05 indicated significant differences; if the data did not conform to normal distribution, the data were first log transformed to normalize the data, and then, the data were statistically analyzed according to the above one-way ANOVA method.        BioMed Research International [5,[11][12][13][14][15][16][17][18][19], and these components were further investigated. Screening parameters of gastrointestinal absorption = high and bioavailability ≥ 0:55 were used to screen 107 ingredients, and then, 35 candidate ingredients were screened using 0:49 ≤ QED ≤ 0:67, as shown in Table 1.

Core
Target Screening for OBL for AD. Canonical SMILES of the 35 candidate components obtained above were imported into the SwissTargetPrediction database to collect their action targets, and 525 action targets were obtained, and 188 action targets were obtained after deleting duplicate targets. The targets were linked to OBL components, and the component-target network map was constructed by Cytoscape software. The network consists of 223 nodes and 437 edges. The edges between the components (red octagonal shape) and the targets (blue circles) represent interactions.
The TTD, CTD, DisGeNET, and DrugBank databases were searched for targets related to AD pathogenesis to obtain 155, 298, 252, and 45 disease targets, respectively, and intersected with 188 OBL targets to obtain 43 shared targets, see Figure 2(a). STRING data screened 43 drug-disease shared targets with the criteria of "Homo sapiens" and " minimum required interaction score " ≥ 0:7. The 43 drugdisease targets were filtered by STRING data with the criteria of "Homo sapiens" and " minimum required interaction scor e " ≥ 0:7 to obtain 37 candidate targets, and the interaction relationship between the targets is shown in Figure 2(b). The network of 37 nodes and 69 edges was simplified as shown in Figure 2(c). The 46 candidate targets were filtered by the CytoNCA plug-in with the median of BC, CC, EC, DC, LAC, NC, and other topological attributes of the initial network nodes ≥ 2 and "value = Default" and "filter =used". and "filter = used by total rank selected to 15% proteins" as filtering criteria to simplify the network and obtain mitogen-activated protein kinase 1 (MAPK1), glycogen synthase-3 beta (GSK3B), mineralocorticoid receptor (NR3C2), estrogen receptor (ESR1), and epidermal growth factor receptor (EGFR). The above five targets were used as core targets for molecular docking with OBL candidate components, see Figure 2(d) and Table 2.

GO and KEGG Enrichment Analysis
Results. The results of GO gene function analysis of 43 common targets of OBL for AD treatment by the DAVID database showed that 289 entries were obtained, and the top 45 results were ranked in accordance with the corrected FDR pairs. G-protein-coupled acetylcholine receptor activity, enzyme binding, G-protein-coupled serotonin receptor activity, RNA polymerase II transcription factor activity, ligand-activated sequence-specific DNA binding, neurotransmitter receptor activity, identical protein binding, alpha1-adrenergic receptor activity, steroid binding, sequence-specific DNA binding, protein serine/threonine/tyrosine kinase activity, zinc ion binding, oxidoreductase activity, acting on single donors with incorporation of molecular oxygen, incorporation of two atoms of oxygen, virus receptor activity, protein kinase binding, and beta-amyloid binding are the first 15 biological functions induced by the 35 OBL components, see Figure 3(a).

BioMed Research International
The first 15 biological processes are adenylate cyclaseinhibiting G-protein-coupled acetylcholine receptor signaling pathway, G-protein-coupled serotonin receptor signaling pathway, signal transduction, positive regulation of vasoconstriction, G-protein-coupled acetylcholine receptor signaling pathway, peptidyl-tyrosine autophosphorylation, aging, Gprotein-coupled receptor signaling pathway, coupled to cyclic nucleotide second messenger, response to xenobiotic stimulus, cellular response to reactive oxygen species, memory, intracellular steroid hormone receptor signaling pathway, cellular response to estradiol stimulus, adenylate cyclase-activating adrenergic receptor signaling pathway, and phospholipase C-activating G-protein-coupled receptor signaling pathway, see Figure 3(c).
Among the above active ingredients, Quercetin and Kaempferol have been reported in the treatment of AD. For example, Quercetin can improve cholinergic function and play a neuroprotective role in AD. The neuroprotective effects of Quercetin have multiple mechanisms, including inhibition of Aβ aggregation [20], inhibition of NFT formation, inhibition of amyloid precursor protein (APP) cleavage enzyme (BACE1) inhibition [21], and acetylcholinesterase (AChE) inhibition [22] to reduce oxidative stress in AD [23]. It plays a role in alleviating Alzheimer's disease in terms of oxidative stress and reactive oxygen species scavenging, as well as improving vascular dysfunction and inhibiting inflammation. In contrast, Kaempferol delays the loss of climbing ability and memory and reduces oxidative stress and acetylcholinesterase activity in AD Drosophila [24].
It was reported [25] that DBP could exacerbate hippocampal tissue damage in AD rats through oxidative stress and upregulate the Bcl-2/Bax/Caspase-3 signaling pathway, 21 BioMed Research International leading to decreased learning memory capacity. DBP exposure aggravates type 2 diabetes by disrupting the insulinmediated PI3K/AKT signaling pathway [26], see Figure 5. DBP epigenetically induces reproductive toxicity via the PTEN/AKT pathway [27]. In this study, molecular docking results showed that DBP has strong binding activity to GSK3B, see Figure 6. Glycogen synthase kinase 3β (GSK-3B) is a key factor of the signal transduction pathway during oxidative stress in AD neurons [28]. Aβ in AD patients has neuronal toxicity and induces oxidative stress in neurons [29]. The downstream direct target gene of phosphatidylinositol (-3) kinase (PI3K) is GSK-3B, and Aβ is able to decrease AKT activity, increase GSK-3β activity, and inhibit AKT/GSK-3-related signaling pathways [30]. It can be seen that if GSK-3β can be effectively inhibited, it can help alleviate the symptoms of AD patients.

22
BioMed Research International while little research has been done on the effects of phthalates on AD. The blood-brain barrier (BBB), resulting in low drug solubility and low bioavailability, has become a bottleneck in the current treatment of AD [35]. DBP has good blood-brain barrier permeability (BBB = 0:56) [36]. Whether DBP exerts its therapeutic effect on AD only by inhibiting AKT/GSK-3β, the next step is to conduct in vitro experiments to verify it, see Figure 5. To further determine the effect of DBP on the activity of HT22 cells with Aβ 25-35 -induced injury, the LDH content in the cell supernatant was measured. The results showed a statistically significant decrease in LDL content with concentration gradient in the DBP 50, 100, and 150 μmol/L groups compared with the model group (P < 0:01). In addition, the cell survival rate was significantly reduced after Aβ [25][26][27][28][29][30][31][32][33][34][35] intervention in mouse hippocampal neuronal cells HT22. Based on the experimental results, 50, 100 μmol/L, and 150 μmol/L were selected as the low, medium, and high concentrations of DBP for subsequent experiments, see Figure 8.
3.6. LDH and ROS Assay Results. As shown in Figure 8(c), compared with the control group, the LDH content in the cell culture fluid of the DBP low-, medium-, and high-dose groups was significantly reduced, and the difference was statistically significant (P < 0:01). Compared with the model group, the LDH content in the cell cultures of the DBP low-, medium-, and high-dose groups was significantly lower than that of the model group, with a statistically significant difference compared with the model group (P < 0:01).
As shown in Figure 8(d), the fluorescence intensity of ROS in the cell culture medium of DBP low-, medium-, and high-dose groups was significantly reduced, and the difference was statistically significant (P < 0:01). Compared with the model group, the fluorescence intensity of ROS in the cell culture fluid of DBP low-, medium-, and high-dose groups was significantly reduced, and the difference was more significantly reduced in the high-dose group, and the difference was statistically significant (P < 0:01). This indicates that DBP can reduce oxidative stress and improve the viability of HT22 cells.
3.7. Cell Cycle Assay. The results are shown in Figure 8(e). Flow cytometric detection of DBP on the cell cycle after Aβ 25-35 -induced injury to HT22 cells showed that, compared with the model group, the Aβ 25-35 -induced injury to HT22AD cell model group showed a gradual decrease in the effect on the G0/G1 phase of the cell cycle with increasing DBP dose, a gradual increase in the effect on the S phase, and almost no effect on the G2/M phase.
3.9. qRT-PCR Assay Results. The results of reverse transcription quantitative PCR of AKT and GSK-3β mRNA expression after different intervention groups are shown in Figure 11. The effect of DBP on AKT and GSK-3β mRNA expression was increased in a concentration-dependent manner. Among them, compared with the model group, the AKT and GSK-3β mRNA expression levels of DBP low-, medium-, and high-dose groups increased with the gradient of DBP concentration, and the difference was not statistically significant.
3.10. WB Test Results. The WB results for detection of target protein AKT, p-AKT, GSK-3β, and p-GSK-3β and internal reference protein β-actin are identified separately in Figure 12. The sizes of the bands in the graphs match the sizes of the proteins. Among them, p-AKT and p-GSK-3β protein expression was significantly increased in a concentration-dependent manner in the DBP low-, medium-, and high-dose groups compared with the model group, and the expression levels were more pronounced in the high dose, with statistically significant differences (P < 0:05).

Discussion
Natural products have complex biological activities; their components are complex and diverse, and the composition of the formula is even more complex. In the field of TCM, natural products are commonly used in disease treatment, and their pathways and modes of action vary after entering the human body. They can act directly on specific targets, produce new products after metabolism, or act indirectly through the regulation of endogenous substances, exerting

24
BioMed Research International multicomponent, multitarget regulation. It is urgent to establish new research strategies and methods that can reflect the overall characteristics of TCM-ethnic medicine. In recent years, the rapid development of network pharmacology in TCM-ethnic medicine research has attracted attention, which integrates three aspects of TCM-ethnic medicine, including components, targets, and related diseases, and constructs a multidimensional network of "components-targetspathways-diseases." The active ingredients and mechanisms of action of TCM-ethnic medicine for diseases are then visualized and analyzed. Visual analysis of the PPI network and its CytoNCA network revealed multiple associations between targets, with higher connectivity values associated with greater potential therapeutic effects. The top-ranked target clusters of MAPK1, EGFR, NR3C2, ESR1, and GSK3B may be key targets. In previous studies, MAPK1 has been associated in previous studies with neurodegeneration, synaptic plasticity, cell survival, and a role in autophagic vesicle formation in AD [37,38]. The MAPK1 gene is thought to be an agedependent transcriptional alteration gene involved in aberrant hyperphosphorylation of tau proteins, leading to aggregated neurogenic fiber tangles [39]. Furthermore, galantamine can treat Alzheimer's disease by attenuating the activation of MAPK1 [40].
EGFR is a transmembrane receptor with tyrosine kinase activity and is an important member of the ErbB family that is involved in regulating brain development, neuronal survival, and functional regulation, among other activities. Many neurodegenerative diseases include AD [41]. High levels of EGFR may improve the metabolism of pathological cerebrospinal fluid biomarkers associated with AD in cognitively normal middle-aged individuals [42]. Many recent studies have shown that EGFR inhibitors enhance autophagy, improve Aβ toxicity, and neuroinflammation [43].
NR3C2 is an important gene involved in the stress response, and its gene product, salt cortico-steroid receptor, is mainly distributed in the hippocampus and amygdala regions involved in the regulation of tension and anxiety and is closely related to tension and anxiety generation and regulation and cognitive function [44]. miR-135b-5p upregulation can reduce neuronal damage and inflammatory response in PSCI by targeting NR3C2, which is useful for poststroke cognitive impairment treatment [45].
Estrogen can cross the blood-brain barrier to act in the brain [46], and the action of estrogen is dependent on at least 2 ESRs (ESR1 and ESR2), potential candidate genes that regulate the development of AD. Variants in the ESR1 gene have been reported to regulate the susceptibility or course of AD. Scacchi et al. may be another gene that promotes interindividual variation in response to treatment with cholinesterase inhibitors (ChEIs) of genes [47].
KEGG enrichment pathway analysis revealed that neuroactive ligand-receptor interaction [48], PI3K-AKT signaling pathway [49], cholinergic synapse [50], regulation of actin cytoskeleton [51], Alzheimer's disease, pathways of neurodegeneration-multiple diseases, MAPK signaling pathway [52], and cGMP-PKG signaling pathway [53] are important pathways related to disease regulation in AD and have been reported in the literature, suggesting that the pathways predicted to be enriched in this study have high confidence. Quercetin mediates activation of the PI3K/AKT/GSK-3β signaling pathway through ER and has a protective effect against Aβ 25-35 -induced damage in PC12 cells [54]. Quercetin protects against okadaic acid-(OA-) induced hippocampal neuronal injury in HT22, a cell line derived from mouse hippocampal neurons, via MAPK and PI3K/AKT/GSK3β signaling pathways [55]. Kaempferol exposure delayed the loss of climbing ability, memory, and reduced oxidative stress and acetylcholinesterase activity in Drosophila AD [24].
The results of the present study showed that DBP was able to reduce the rate of inhibition of HT22 cells with Aβ 25-35 -induced damage, and the results indicated that low doses of DBP had a protective effect on HT22 cells with Aβ 25-35 -induced damage. The present study showed that DBP significantly reduced the LDH and ROS content in the supernatant of Aβ 25-35 -induced injury HT22 cell cultures, and DBP was also able to reduce the apoptosis rate of Aβ 25-35 -induced injury HT22 cells. Thus, it can be speculated from the results of this study that DBP has the ability to protect the stability of injured neuronal cells and improve the permeability of the cell membrane, thus stabilizing the intracellular environment. This effect may be related to the fact that DBP increases the mRNA levels of AKT, GSK-3β, etc. in AD cell models and regulates the phosphorylation of the AKT/GSK-3β pathway.
In summary, OBL has been used to explain the relationship between OBL active ingredients, potential targets, signaling pathways, and the pathogenesis of AD disease at a holistic level through network pharmacology technology and to verify the pharmacodynamic and regulatory mechanisms of OBL main active ingredients through in vitro experimental methods. This paper provides a new idea for the treatment of AD with complex pathogenesis and also lays the foundation for the in-depth study of the synergistic mechanism of OBL.

Data Availability
The data used to support the findings of this study are included within the article.