Forecast of AD-related targets.

To conduct a search relevant to “Alzheimer’s disease,” three databases were utilized: DrugBank database (https://www.drugbank.ca/), GeneCards database (https://www.genecards.org/), and OMIM database (https://omim.org/). These databases were employed to ascertain targets associated with Alzheimer’s disease (AD). After removing the duplicate values from these three databases, the results were prospective AD-related genes.

Filtering active ingredients of PkO.

The active constituents of PkO were achieved from a reference-guided database of traditional Chinese medicine (S1 and S2 Table), utilizing a systems pharmacology-based approach with experimental validation. (TCMSP, http://old.tcmsp-e.com/) [10]. We utilized the SwissADME tool (http://swissadme.ch) to forecast plausible active constituents. The assessment encompassed the permeability of the blood-brain barrier (BBB), evaluation of gastrointestinal (GI) absorption, and conducted drug-likeness (DL) analysis [16]. The ADME screening mentioned the criteria defined by Lipinski’s rules, which include the following: the molecular weight should be less than or equal to 500 g/mol, the Moriguchi octanol-water partition coefficient should be less than or equal to 4.15, the number of nitrogen or oxygen atoms should be less than or equal to 10, and the number of NH or OH groups should be less than or equal to 5. All bioactive compounds fulfilled the requirements of the “Abbott Bioavailability Score (>0.1)” as determined by SwissADME. Typically, active ingredients encompass compounds that exhibit blood-brain barrier permeability, high gastrointestinal (GI) absorption, and conform to Lipinski’s rules during drug-likeness analysis [17,18]. Such compounds, once meeting these screening criteria, are identified as active ingredients possessing therapeutic efficacy.

Forecasting potential targets of active ingredients.

We used SwissTargetPrediction (http://www.swisstargetprediction.ch/) to fetch and forecast the associated targets of compounds in PkO. SwissTargetPrediction, a reputable database, leverages the 2D and 3D structures of established compounds for target prediction [19]. In this study, the target screening criterion was defined as a probability value exceeding zero.

Screening shared AD-related targets and performing visual analysis.

The intersection of the targets of Alzheimer’s disease and active ingredients was identified by Jvenn (https://jvenn.toulouse.inrae.fr/), an interactive Venn diagram viewer [20]. These shared targets are the estimated targets of the active ingredients of PkO for the treatment of AD. Cytoscape 3.7.2 software was utilized to construct, visualize, and analyze the network that connects herbs, active ingredients, and disease-related targets, with the aim of comprehending their intricate relationship.

GO and KEGG pathway enrichment analysis.

We utilized the R package from the “org.Hs.e.g.,db” database to select the estimated targets and used DOSE, clusterProfiler, and pathview package (Bioconductor) to perform GO enrichment analysis from three aspects, Biological Process (BP), Cellular Component (CC), and Molecular Function (MF). The top 10 results of each form were chosen to use bar charts or scatter diagrams for visual analysis. Based on the shared targets, we used DOSE, clusterProfiler, and pathview packages (Bioconductor) to conduct KEGG pathway enrichment analysis to forecast its mechanism of action using. Bar charts or scatter diagrams were conducted for visualization. The analysis was performed with P-value cutoff as 0.05 and Q-value cutoff as 0.05.

Protein-protein interaction data.

We imported the shared targets into the Search Tool for the Retrieval of Interacting Genes (STRING) database (https://string-db.org) and constructed a protein-protein interaction (PPI) network [21]. We chose “Homo sapiens” as the target-oriented species, selected the highest confidence scores (> 0.9) and deleted the isolated vertices. The measurement of the local centrality of a node in the network relied heavily on degree centrality, an essential parameter. In general, a node owning a high degree centrality might be the essential node of the network. In our study, we chose the top thirty targets in the PPI network regarding degree centrality as key targets for further research.

Data screening and visualization.

Based on the outcomes of GO and KEGG pathway enrichment analysis and PPI network, a compound-target-pathway map was constructed using Cytoscape 3.7.2 software to screen targets and pathways for verification in follow-up studies.

Molecular docking validation.

We conducted molecular docking with AutoDock and AutoDock Vina software which is based on molecular modeling techniques to forecast the directional interaction between the pivotal targets and compounds [22]. From Protein Data Bank (PDB, http://www.rcsb.org), we successfully achieved the crystal structure of the several objectives. We gained the structure of components from the PubChem database (https://pubchem.ncbi.nlm.nih.gov). Prior to conducting molecular docking experiments, we employed AutoDock software (version 4.2.6) for removing the ligand, hydrogenation, charge calculation, addition of protein type to macromolecules, and saving the resultant structures in PDBQT format. Following that, molecular docking between macromolecules and ligands was executed using AutoDock Vina, adhering to the predefined parameters. Binding energy of less than “-5” meant a grater combining interaction between macromolecules and small molecules [23]. In final, the PyMOL software (version 2.6.0) was utilized for the visualization of molecular docking.

Reagents and antibodies.

Hancinone (molecular weight 340.4; HPLC grade, > 97%) was purchased from Shanghai Acmec Biochemical Technology Co., Ltd. β-Amyloid (1–42) was purchased from GL Biochem Co., Ltd. 3-(4, 5-dimethyl-2-thiazolyl)-2, 5- diphenyl-2-H-tetrazolium bromide (MTT), and dimethyl sulfoxide (DMSO) were purchased from Sigma-Aldrich. R406 (Syk inhibitor) and Triton X-100, DAPI Staining Solution (C1005) were purchased from Beyotime. 0.25% of trypsin was from Keygen, Ltd. The 5-FITC-(Acp)-β-amyloid (1–42), human oligomer (sequence 5-FITC-(Acp)-DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVVIA; purity 96.17%) was purchased from ChinaPeptides Co., Ltd. RPMI 1640 was from Servicebio and fetal bovine serum (FBS) was purchased from Yuanyi Biotechnology Co., Ltd.

Phospho-Syk (Try525/526) antibody (Cat No: #2710), Syk antibody (Cat No: #13198) and phospho- AKT(Ser473) antibody (Cat No: #4060) were obtained from Cell Signaling Technology (Denver, USA). Pan-AKT antibody (Cat No: A18675), phospho-PI3K antibody (Cat No: AP0427), PI3 Kinase antibody (Cat No: A4992), β-actin antibody (Cat No: AC026) and HRP Goat Anti-Rabbit IgG (H + L) antibody (Cat No: AS014) were obtained from ABclonal Biotech Co., Ltd. TREM2 Polyclonal antibody (Cat No: AF09731) was from AiFang Biological Co., Ltd. mTOR antibody (Cat No: ab134903) was from Abcam. CD68 antibody was from Invitrogen. FITC CD206 (MMR) antibody (Cat No: 141703) was purchased from Biolegend (S3 Table).

The compound Hancinone was prepared by dissolving it in sterile deionized water at a concentration of 2 mM. This solution was then stored at a temperature of -80°C. When needed, the Hancinone solution was diluted to the desired concentrations using culturing medium. Syk inhibitor was dissolved in Phosphate Buffered Saline (PBS) at 40 µM, stored at 4°C, and diluted in culturing medium to desired concentrations immediately before use. We used an alternative approach to simulate Aβ formation because that in vitro experiments could not model complex Aβ production in vivo. Thus, we added sufficient chilled HFIP to the Aβ1–42 powder (GL Biochem) and mixed thoroughly to ensure complete dissolution of the powder. After incubating at room temperature for 60 min, the solution was placed on ice for 10 min before transferred to a centrifuge tube. We leaved the tube uncapped and placed it in a fume hood overnight at room temperature and used a freeze dryer to allow the HFIP to evaporate completely, leaving a clear and transparent film. We dissolved the film completely in DMSO and diluted it with F12 medium to prepare a final concentration of 250 µM Aβ solution (S2 Fig and S1 Report), maintaining sterile operations during the experiment to avoid contamination [24].

Cell culture and stimulation.

Human Microglia Clone 3 (HMC3) cells were purchased from Procell Life Science& Technology Co., Ltd. (Wuhan, China). The growth and maintenance of cells were in RPMI 1640 (Servicebio), supplemented with 20% heat-inactivated fetal bovine serum, 100 U/mL of penicillin, 100 μg/mL of streptomycin and 0.25 μg/mL of Amphotericin B (Bio-Channel). Cells were cultured at the temperature of 37°C and the environment of 5% CO2. Afterwards, cells were plated in 6-well plates until cells reached 80% confluence and then incubated with serum-free medium for another 6 h before Syk inhibitor, β-amyloid (1–42) oligomer, or hancinone supplementation. Cells were then incubated for 24 h in the presence or absence of Syk inhibitor (SI, 100 nmol/L), Aβ oligomer (Aβ, 2.5 µM), or hancinone (HAN, 0.5 µM, 2.5 µM, 10 µM) to estimate the influence of hancinone treatment on HMC3 responsiveness to analyze the impact of hancinone treatment on TREM2/Syk signaling.

Cell viability assay.

The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (MTT) assay is a sensitive survey of the regular metabolic condition of cells [25]. In this experiment, a solution of MTT (5 mg/mL) was added to each sample in a volume of 20 μL. Due to the presence of 180 μL of cell culture medium in each well, the final concentration of MTT is 0.5 mg/mL. MTT is an indirect measurement method that can be influenced by cell type and status, and MTT reagent possesses a certain degree of toxicity, which may have some impact on the experiment. During the experimental process, we refined the accuracy of the experimental results through careful operation. The purpose of this addition was to assess the viability of the cells. The samples were then incubated for 4 h at a temperature of 37°C and in the presence of 5% CO2. Following the incubation period, the culture medium was removed and the formazan products, which had formed because of the reaction with MTT, were dissolved in a volume of 150 μL DMSO each well. This step allowed for the quantification of the formazan products, which in turn provided a measure of cell viability. Each well was measured by light absorbance at 490 nm to calculate the appropriate concentration of Syk inhibitor, β-amyloid (1–42) oligomer, and hancinone for further study, which means that these three drugs did not exert toxic effects on HMC3 cells within the corresponding concentration ranges (S1 Fig).

Western blotting.

To calculate the appropriate concentration of SI and Aβ treated on HMC3 cells, the cells were respectively incubated with: SI group 1–6 (0, 25, 50, 100, 150, and 200 nmol/L SI), Aβ group 1–7 (0, 0.1, 0.25, 0.5, 0.75, 1 and 2.5 μmol/L Aβ). Further, according to the stimulation mentioned above, the cells were respectively incubated with control I (culture medium), model group I (100 nmol/L SI), low-concentration HAN group I (100 nmol/L SI + 0.5 μmol/L HAN), medium-concentration HAN group I (100 nmol/L SI + 2.5 μmol/L HAN), high-concentration HAN group I (100 nmol/L SI + 10 μmol/L HAN), control II (culture medium), model group II (2.5 μmol/L Aβ), low-concentration HAN group II (2.5 μmol/L Aβ + 0.5 μmol/L HAN), medium-concentration HAN group II (2.5 μmol/L Aβ + 2.5 μmol/L HAN) and high-concentration HAN group II (2.5 μmol/L Aβ + 10 μmol/L HAN). Following the administration of medication, the cells underwent two purges using ice-cold PBS (137 mM NaCl, 2.7 M KCl, 10 mM Na2HPO4, and 1.8 mM KH2PO4, pH 7.4). Subsequently, the cells were lysed using SDS sample buffer. Protein-equivalent cell lysates were subjected to SDS-PAGE (Beyotime) and subsequently transferred onto polyvinylidene difluoride (nitrocellulose) membranes. Each primary antibody was diluted using the primary antibody dilution buffer (Beyotime) according to the concentration specified in the corresponding instruction manual and the secondary antibody was diluted using the secondary antibody dilution buffer (Beyotime) at a ratio of 1:2000. Following a blocking step using blocking buffer (NCM Biotech), the membranes were incubated overnight with the suitable primary antibodies at 4°C. Finally, the membranes were exposed to secondary antibodies for 2 h at room temperature. We used the enhanced chemiluminescence system from Nanjing Vazyme Biotech Co., Ltd. to visualize immunoreactive proteins. We used the ratio of the target protein band intensity to the corresponding internal control protein band on the same membrane for subsequent calculations. Imaging and analysis with Image Lab software.

Flow cytometric analysis of phagocytosis and polarization.

Cells were harvested using EDTA-free trypsin solution (Beyotime). Following complete digestion neutralization, the cell suspension was transferred to a pre-chilled centrifuge tube and subjected to centrifugation at 1000 rpm for 5 min (4°C). After careful aspiration of the supernatant, the pellet was resuspended in PBS for gentle washing via pipette mixing, followed by repeated centrifugation under identical parameters. Then, cells were reconstituted in ice-cold FACS Buffer at a concentration of 1–5 × 10⁶ cells/mL for downstream applications. Afterwards, to analyze the phagocytic ability of HMC3 cells to Aβ, cells were stained by the 5-FITC-(Acp)-β-amyloid (1–42), human oligomer (QYAO Biotech) according to the manufacturer’s instructions. After 30-min incubation in the dark at 37°C incubator, cells were washed with 1-ml FACS buffer examined on a FACSCalibur (BD Biosciences) flow cytometer. In the research of polarization, cells were orderly stained by CD68 antibody (Invitrogen) for 30 min and CD206 (MMR) antibody (Biolegend) for 60 min in the dark at 37°C. Cells were then washed with 1-ml FACS buffer and then examined on a FACSCalibur (BD Biosciences) flow cytometer. According to the characteristics of the experimental subject HMC3 cells, corresponding FSC and SSC parameters are set to accurately identify HMC3 cells. Since the cells themselves do not contain stained Aβ oligomers, we could explore the amount of Aβ oligomers phagocytosed by HMC3 cells to a certain extent. We first used single-stain tubes for detection to eliminate interference. At the same time, based on the situation of singly stained cells, omissions or misidentifications are avoided, and further analysis is carried out after setting appropriate gating. The data obtained above were analyzed with FlowJo v10.8.1.