The BATMAN-TCM database (Liu et al., 2016) was used to identify the effective components and targets of SJZD through screening using BaiZhu, FuLing, GanCao, and RenShen as the search keywords. The criteria for screening had a correlation score of more than 20 and a p-value of less than 0.05, and the target names were standardized via the Uniprot database. Any duplicate targets were removed and merged into the target database.

After searching for “Parkinson’s Disease” on both the DisGeNet database (https://www.disgenet.org/) (Zhang et al., 2018) and GeneCards database (http://www.genecards.org/) (Guo et al., 2020), we created a database of disease targets related to PD by eliminating any duplicates and compiling the relevant genes.

By intersecting the SJZD and PD target databases, we obtained the intersection target of “SJZD-PD” which was then visualized using the VENNY diagram (https://bioinfogp.cnb.csic.es/tools/venny). Subsequently, the intersection targets, active components, and drug names were imported into Cytoscape 3.7.1 software to construct and visualize the “SJZD-Active Compounds-Intersection Targets” network. Finally, the core drug targets were identified based on the Degree value.

The target obtained from “SJZD-PD” was imported into the String database (https://www.string-db.org/) (Szklarczyk et al., 2015). A PPI network was obtained by limiting the study species to “Homo sapiens” and setting the minimum threshold to “medium confidence (0.400)” (Wang et al., 2017), and core targets were screened by calculating the Degree value. The results were then visually optimized using Cytoscape 3.7.1.

Utilizing the Bioconductor database and the RX64 4.0.0 software, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were conducted on core target genes. This analysis focused on three categories: biological process (BP), cell composition (CC), and molecular function (MF) (Ye et al., 2021). We used a free online platform (http://www.bioinformatics.com.cn) to visualize the results.

Using the PubChem database (https://pubchem.ncbi.nlm.nih.gov/), the core compounds in the network diagram of “SJZD-Active Compounds-Intersection Targets” were obtained in SDF file format. OpenBabel software was used to convert the SDF file into mol2 format, and AutoDock Tool software (Trott and Olson, 2010) was used to perform hydrogenation, ligand detection, and other operations before saving it in pdbqt format. The core protein structure was obtained from the PDB database (http://www.rcsb.org/), and Pymol software (Zhu and Hou, 2020) was used to remove its ligands and water molecules. AutoDock Vina software was then used to link the protein structure and compound molecules for molecular docking, and the results were visualized and analyzed using Discover Studio software 4.5.

Molecular dynamics simulations were conducted on the top three complexes with the highest molecular docking scores using the Amber18 software package. The protein was assigned the ff14SB force field parameter, while the small molecule ligand was assigned the gaff universal force field parameter and had its AM1-BCC atomic charge calculated using the ANTECHAMBER module. The complexes were loaded into the TLAP module, with hydrogen atoms and antagonistic ions added to neutralize charges. The TIP3P explicit water model was selected, and periodic boundary conditions were set. The simulation workflow included four steps: energy minimization, heating, equilibration, and production kinetics simulation. After energy minimization of the heavy atoms of the proteins (and small molecules) in 10,000 steps, constraints were relaxed, and an additional 10,000 steps of energy optimization were conducted on the entire system. The system was then slowly heated to 300 K throughout 50 ps, followed by equilibration at 50 ps under the NPT ensemble. Finally, a molecular dynamics simulation of 200 ns was conducted under the NPT ensemble, with a time step of 2 fs. Trajectory data was saved every 40 ps, and correlation analysis was performed using the CPPTRAJ module. The binding free energy of the ligands and proteins was calculated using the MMPBSA.py module.

Human SH⁃SY5Y neuroblastoma cells were purchased from Wuhan Punosai Life Technology Co., LTD, and cultured in DMEM/F12 medium (AW-M006, Abiowell, Hunan, China) containing 15% fetal bovine serum (FBS, AWC0227a, Abiowell, Hunan, China) and 1% penicillin-streptomycin (AWH0529a, Abiowell, Hunan, China). The cells were maintained in a constant temperature incubator with 5% CO₂ at 37°C. When the cells reached a confluence of 80%–90%, they were digested with trypsin. This digestion process was subsequently halted using a complete medium. Following centrifugation, the cells were suspended and transferred to a new culture bottle. Cellular experiments were conducted on an ultra-clean workbench (YT-CJ-2NB, YATAIKELONG, Beijing, China). During the cell culture process, all containers were sterilized in a high-temperature autoclave, and ultraviolet sterilization was used for the ultra-clean workbench.

Logarithmic SH-SY5Y cells were inoculated on 96-well plates (703001, NEST, Jiangsu, China) with a density of 5 × 10³ cells/well. After the cells adhered to the walls of the plates, they were treated with different concentrations of 6-OHDA (H4381, Sigma-Aldrich, St. Louis, MO, United States) (0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100 µM) and incubated at 37°C in 5% CO₂ for 24 h. Thereafter, 70 µM was determined as the optimal concentration for further analysis. Under the same incubation conditions, SH-SY5Y cells treated with 70 µM 6-OHDA were exposed to different concentrations of SJZD (25, 50, 75, 100, 125, and 150 μg/mL), which was purchased from Hunan Provincial Hospital of Integrated Traditional Chinese and Western Medicine. SH-SY5Y cells and 6-OHDA-treated SH-SY5Y cells were incubated with 10 μL Cell Counting Kit-8 (CCK-8) buffer (Dojindo, Japan) for 4 h to determine half maximal inhibitory concentration (IC₅₀) (Supplementary Figure S1) and optimal SJZD concentration, respectively. The absorbance values for each hole were measured at a wavelength of 450 nm using a multifunctional enzyme label analyzer (Huisong, Shenzhen, China).

To study the effect of different doses of SJZD, the cells were divided into the SJZD-L group (75 μg/mL), the SJZD-M group (100 μg/mL), the SJZD-H group (125 μg/mL), and the control group. RIPA lysate was added into the cells for protein extraction and centrifuged at 12000 rpm for 15 min at 4°C. The protein concentration was determined using the BCA method, with absorbance measured at a wavelength of 562 nm using an enzymoscope, followed by constructing a standard curve. The protein was transferred onto the nitrate cellulose membrane and blocked with 5% skim milk powder at room temperature for 1 h. The membranes were incubated with Bax (1:2000, 50599-2-Ig, Proteintech, United States), Bcl-2 (1:500, 26593-1-AP, Proteintech, United States), Caspase-3 (1:500, 19677-1-AP, Proteintech, United States), p-PI3K (1:500, Bs-5570R, Proteintech, United States), and p-AKT (1:500, 66444-1-Ig, Proteintech, United States) at 4°C overnight. The film was then washed three times with TBST (Biosharp, Anhui, China), and incubated with a secondary antibody (1:5000, AWH0529a, Abiowell, Hunan, China) at room temperature for 1 h. Finally, the washed film was developed with ECL luminescent reagent (ThermoFisher Scientific, Shanghai, China) away from light. ImageJ software (NIH, Bethesda, MD) was used to analyze the gray values of the target protein bands. β-actin (1:500, Proteintech, United States) was used as the loading control.

SPSS 26.0 software was used for statistical analysis, and the data were expressed as mean ± standard deviation x ¯ ± s . For comparison of data among multiple groups, One-way ANOVA was used when the variance was homogeneous, and LSD method was used for multiple comparisons after the event. Dunnett’s T3 method is used when variance is uneven. The statistical significance level was set as p < 0.05.

We obtained 5 active compounds from BaiZhu, 18 from FuLing, 57 from GanCao, and 112 from RenShen. After eliminating duplicates, a total of 188 active compounds were identified. Table 1 shows the effective components of each TCM. We found 82 targets for BaiZhu, 654 for FuLing, 1004 for GanCao, and 4813 for RenShen. After eliminating duplicates, the total number of drug targets was 1568.

Through Genecard database (score ≥10) screening, 1853 targets were identified, with a further 497 DisGeNet (score ≥0.05) targets identified, resulting in a total of 2069 disease target genes after duplicates were removed. Utilizing PERL software, 451 intersection targets of “SJZD-PD” were identified and visualized through VENNY plots (Figure 2).

By uploading active compounds and common targets into Cytoscape 3.7.1, a “SJZD-Active Compounds-Intersection Targets” network comprising 635 nodes and 2341 edges (Figure 3) was created. The network provides insight into the complex effects of active compounds in SJZD on PD. Ranked from highest to lowest, Adenosine Triphosphate, Tridecanoic Acid, Hexadecanoic Acid, Pentadecanoic Acid, and Adenosine were the top five Degree values (Table 2). These are thought to be the primary active components of SJZD for treating PD.

The PPI network consisting of 451 intersection targets included 448 nodes and 9753 edges, with an average node degree of 43.5 and an average local clustering coefficient of 0.483 (Figure 4). The target is represented by a circular node, and the larger the node, the higher the degree value, and the more important the target. Using R X64 4.0.0, a bar chart of the top 30 target genes was generated (Figure 5). In order of Degree, the top 5 were AKT1 (Degree = 224), INS (Degree = 216), TNF (Degree = 208), IL-6 (Degree = 202), and TP53 (Degree = 198). Therefore, these 5 targets can be considered the most important core targets of SJZD in the treatment of PD, warranting further study.

We conducted a GO enrichment analysis on 451 common targets to further investigate the mechanism of SJZD in treating PD. A total of 4211 statistically significant entries were identified in three categories, including 3726 entries for BP, 200 for CC, and 285 for MF. The analysis results are presented as bubble charts, with the redder colors indicating lower adjusted p values and higher enrichment of the GO entries (Figures 6A–C). Furthermore, the top 10 entries with the highest gene counts in each category are presented in Figure 6D and Table 3. These entries were mainly associated with various biological processes such as response to xenobiotic stimuli, response to oxygen levels, response to reduced oxygen content, reactive oxygen species metabolism, and response to hypoxia; cellular components such as neuronal cell bodies, membrane rafts, membrane microdomains, ion channel complexes, and cation channel complexes; and molecular functions such as gated ion channel activity, voltage-gated ion channel activity, voltage-gated channel activity, signal receptor activator activity, and amyloid protein-β binding.

A total of 451 common targets between SJZD and PD were evaluated by KEGG pathway enrichment analysis (p < 0.05 as the significant level). Targets were highly enriched in 213 pathways (Supplementary Table S1), such as those associated with fluid shear stress and atherosclerosis, lipid and atherosclerosis, chemical carcinogenesis receptor activation, cyclic nucleotides (cAMP) signaling pathway, mitogen-activated protein kinase (MAPK) signaling pathway, AGE-RAGE signaling pathway in diabetes complications and other disease pathways. Gene counts revealed the top 20 highly enriched pathways (Figures 7A,B), suggesting that SJZD is essential for the multi-targeting and treatment of PD. To gain insight into the relationships between compounds with Degree ≥50 (Supplementary Table S2), targets, and the top 10 pathways, a “compound-target-pathway” network was created using Cytoscape 3.7.1, containing 123 nodes and 507 edges (Figure 7C).

Molecular docking was performed on the top five target proteins (AKT1, INS, TNF, IL-6, and TP53) and five active components (Tridecanoic Acid, Pentadecanoic Acid, Hexadecanoic Acid, Adenosine Triphosphate, and Adenosine) from the PPI and “SJZD-Active Components-Intersection Targets” network with the highest Degree values. The highest docking score (binding energy of −7.6 kcal·mol⁻¹) was observed between IL-6 and Adenosine. The following highest scores were observed between AKT1 and Adenosine (−7.1 kcal·mol⁻¹), TNF and Adenosine Triphosphate (−7 kcal·mol⁻¹), TNF and Adenosine (−6.4 kcal·mol⁻¹), and AKT1 and Adenosine Triphosphate (−6.4 kcal·mol⁻¹). The results of the docking between active components and target proteins are outlined in Table 4 and illustrated in Figure 8.

Molecular dynamics simulations were conducted on the top three active components with the highest absolute binding scores - compounds (IL-6-Adenosine, AKT1-Adenosine, TNF-Adenosine Triphosphate) - and the results of these simulations were used to calculate the root mean square deviation (RMSD), radius of gyration (Rog), root mean square fluctuation (RMSF), and binding free energy.

The RMSD curve provides insight into the fluctuation of protein conformation (Figure 9A). The RMSD of AKT1-Adenosine protein remained relatively stable after 60 ns, stabilizing around 5Å. Similarly, the RMSD of IL6 Adenosine protein was relatively smooth, with a stable value around 2.5Å after 50 ns. These results suggest that the binding of small molecules to receptor proteins results in a stable complex system that does not cause significant changes in receptor conformation. In contrast, the RMSD of the TNF-Adenosine Triphosphate protein fluctuated significantly, demonstrating that the binding of small molecules to receptor proteins has a significant impact on protein conformation.

The Rog results (Figure 9B) demonstrated the compactness of the protein’s overall structure. The turning radii of AKT1-Adenosine and IL-6-Adenosine remained relatively stable, suggesting that the binding of small molecules to receptors did not significantly affect protein folding. On the other hand, the radius of TNF-Adenosine Triphosphate gyration exhibited some fluctuation, indicating that the binding of small molecules to receptors had a certain degree of influence on protein folding.

The RMSF curve revealed increased flexibility in certain areas of the protein core domain for AKT1-Adenosine, IL6-Adenosine, and Chain A of TNF-Adenosine Triphosphate, as demonstrated in Figure 9C. These regions were characterized by a high number of loop structures, resulting in increased flexibility compared to other sections. AKT1-Adenosine had notably higher flexibility in the 60–100 and 120–150 regions, while IL-6-Adenosine displayed increased flexibility in smaller sections of the protein core domain. Similarly, Chain A of TNF-Adenosine Triphosphate exhibited higher flexibility, particularly in the 50–130 region with a high loop content.

CCK-8 assay demonstrated that SJZD effectively suppressed apoptosis in 6-OHDA-induced SH-SY5Y cells. Except for 150 μg/mL, SJZD groups at various concentrations (25, 50, 75, 100, and 125 μg/mL) demonstrated a notable reduction in cell apoptosis, and the rate of apoptosis was significantly lower than that of the control group (Figure 10A). WB analysis results of different SJZD concentration groups showed that the expression of p⁃PI3K and p⁃AKT in the SJZD-M group and SJZD-H group were increased significantly compared with the control group (all p < 0.01). However, Bax and Caspase-3 decreased significantly (all p < 0.01). The Bcl-2 in the SJZD-H group was increased (p = 0.0159), and the protein expression of cells in each group was shown in Figure 10B, and the comparison of gray value expression of cells in each group was shown in Figure 10C.

This study identified 188 active components for treating PD in SJZD. The four herbs found in SJZD, namely, BaiZhu, FuLing, RenShen, and GanCao, have been proven to have therapeutic effects on PD. Ginsenosides in Ginseng, for instance, have anti-aging, antioxidant, and neuroprotective properties. It has been observed that ginsenosides can counteract the damage to the hippocampus caused by aging in mice, reduce oxidative stress levels, and downregulate the p53∼p21 signaling pathways downstream (Zhang J. J. et al., 2022). Furthermore, ginsenosides have been found to have a protective effect on dopaminergic neurons in the substantia nigra of MPTP-induced PD mice. This effect may be facilitated by γ-Aminobutyric acid neurons, which play a role in regulating the excitatory and inhibitory balance within the direct and indirect pathways associated with PD (Sharma et al., 2022; Dolgacheva et al., 2022). Research has demonstrated that Atractylodes macrocephala extract can significantly improve indicators related to oxidative stress in the brains of rats with cerebral ischemia/reperfusion. This improvement is evidenced by an increase in the activity levels of superoxide dismutase and glutathione, alongside a decrease in malondialdehyde content (Li et al., 2022). Additionally, the polysaccharides in Atractylodes macrocephala have been observed to effectively reduce apoptosis of cortical neurons in hypoxic rats. Poria cocos has also demonstrated a protective effect on the nervous system, as its extract enhances the 5-HT metabolic pathway, regulates the interaction of acetylcholine norepinephrine signal, and maintains a balanced of amino acid neurotransmitters. Additionally, it has been shown to ameliorate neurotransmitter imbalances and circadian rhythm disturbances in rats subjected to unpredictable stress (Zhang D. D. et al., 2022). Furthermore, the flavonoids, glycyrrhetinic acid, and isoliquiritigenin in licorice have been found to possess neuroprotective properties against PD caused by neurotoxins such as MPTP and rotenone (Zulfugarova et al., 2023; Moreira et al., 2023; Verma et al., 2022; Ojha et al., 2016). Moreover, studies have indicated that the flavonoid compound glycyrrhizin in licorice can promote axonal elongation of hippocampal neurons, induce the directional differentiation of neural stem cells into cholinergic neurons, and regulate ERK and AKT/GSK-3β pathways to repair nerve cell damage (Moratilla-Rivera et al., 2023; Sidiropoulou et al., 2023; Teng et al., 2014). Notably, as a TCM formula, SJZD affects multiple systems and has been documented to alleviate chronic gastritis (Wang et al., 2020) and exhibit anti-gastric cancer (Ding et al., 2022) and colorectal cancer (Shang et al., 2023).

This study also revealed that Adenosine Triphosphate, Tridecanoic Acid, Hexadecanoic Acid, Pentadecanoic Acid, and Adenosine in SJZD are the essential active components in managing PD. Adenosine Triphosphate, produced by mitochondria, has been associated with energy expenditure and calcium overload, leading to cell apoptosis, ferroptosis, and, ultimately, neuronal death in neurodegenerative diseases (Xu et al., 2022). Hexadecanoic Acid, a polyunsaturated fatty acid, has been shown to possess antioxidant and antithrombotic properties (Ganesan et al., 2022). Moreover, palmitic acid has been demonstrated to have anti-apoptotic effects by regulating the expression of B cell lymphoma 2 (Bcl-2) and activating the MAPK pathway (Kaewin et al., 2022; Wang T. et al., 2022). Adenosine is an important neuromodulator in the central nervous system involved in many brain diseases such as PD. It can influence dopaminergic signaling in the brain by interacting with its receptors to regulate cognition, pain, sleep and wakefulness, anxiety and depression, motor function, and other important physiological processes (Wang et al., 2023). Furthermore, reports show that it acts in the brain primarily through inhibitory A₁ receptor (A₁R) and promotive A_2A receptor (A_2AR) (Fredholm et al., 2005). It has been evidenced that A_2AR antagonists can reduce behavioral and neurochemical features of PD (Morelli et al., 2009). Therefore, the adenosine system is regarded as the primary focus of research for developing new treatments for neurological diseases.

A total of 1,568 drug targets were identified for the treatment of PD in SJZD through network pharmacology analysis. Among these, five core targets were highlighted: AKT1, INS, TNF-α, IL-6, and TP53. It has been suggested that the onset of PD may be linked to the activity of AKT1, which is subject to oxidative modification that causes its dephosphorylation and subsequent neuronal degeneration in the substantia nigra of the brain (Ahmad et al., 2014). Kim et al. (2020) found that the activation of the AKT1-CREB pathway prevents neuronal cell death and motor and cognitive impairment in PD. Insulin resistance has been identified as closely related to the occurrence and development of PD, and abnormal insulin signaling pathways may lead to nervous system inflammation and mitochondrial dysfunction (Chung et al., 2022). Additionally, insulin can improve cognitive dysfunction in patients with PD and diabetes, likely due to its ability to reduce inflammation and enhance mitochondrial energy metabolism (Carvalho and Moreira, 2023). PD is characterized by the progressive loss of dopamine (DA) neurons in the substantia nigra pars compacta (SNpc) (Balestrino and Schapira, 2020), and evidence suggests that neuroinflammation plays an important role in the pathological features and symptoms of PD (Zhang et al., 2023). It has been noted that the serum of PD patients contains elevated levels of TNF-α and IL-6 compared to that of healthy individuals (Shamsdin et al., 2022; Li et al., 2023a; Moradi et al., 2023; Qu et al., 2023). TNF-α has been shown to significantly regulate modifications of the nigrostriatal pathway and to facilitate the apoptosis of dopaminergic neurons, particularly in chronic inflammation (Amin et al., 2022). Furthermore, TNF-α can activate Caspase-mediated cell apoptosis, thus exacerbating the progression of the disease (Coles et al., 2018). Therefore, targeting TNF-α allow for the optimization of its dual neurotoxic and neuroprotective effects in PD. Notably, a recent study (Green et al., 2019) found that IL-6 was positively associated with motor scores in male patients, while higher levels of IL-6 were associated with poorer cognitive performance in female patients. Another study (Fu et al., 2023) also reported a correlation between IL-6 levels and non-motor symptoms and cognitive dysfunction in PD patients, suggesting that IL-6 may be involved in the pathophysiological processes of non-motor symptoms in PD patients. Moreover, TP53/P53 is a transcription factor that regulates cell growth, proliferation, and repair. In response to stress, TP53 may act on signaling pathways such as the Bcl-2 family protein phase or MAPK, thereby playing a role in modulating cell apoptosis and inflammatory response (Dai et al., 2016; Su et al., 2022).

KEGG pathway enrichment analysis revealed that SJZD treatment of PD primarily targets the Dopaminergic synapse, Lipid and atherosclerosis, cAMP signaling pathway, glycosylated protein products (AGEs) and their ligand (RAGE) pathway in diabetic complications, MAPK signaling pathway, and other signaling pathways. Research has suggested that the lack of dopaminergic synapses in the ventral tegmental area of PD patients may lead to depressive symptoms, while synaptic plasticity damage in the striatum may reduce dopamine levels and trigger the onset of the disease (Cantello et al., 1989). Furthermore, inflammation of peripheral organs has been linked to the exacerbation of the progression of PD (Zhang et al., 2023). cAMP is of great importance in the central nervous system cells, playing a role in neuron growth, development, cognition, and the regulation of dendritic and axonal growth (Zhou et al., 2019). Wu et al. (2018) found that by inhibiting miRNA-200a, the expression of dopamine receptor D2 in the striatum is increased, and apoptosis in the striatum of PD rats is suppressed via the cAMP/PKA pathway, thus improving the symptoms of the disease. Another study has also demonstrated that AGEs-RAGE pathways can contribute to the development of PD, with their mechanism of action associated with inflammatory response (Naz et al., 2023). MAPK signaling pathways are also implicated in the initiation and progression of cell apoptosis. ERK1/2, the most studied member of the MAPK family, is thought to induce cell apoptosis by causing the release of cytochrome C, decreasing the anti-apoptotic protein Bcl-2, increasing the pro-apoptotic protein Bax, and activating the apoptotic gene Caspase-9, leading to a cascade reaction and ultimately activating Caspase-3 (Rivo et al., 2007).

The molecular docking results of this study suggest that IL-6 and Adenosine have the strongest binding affinity, with a binding energy of −7.6 kcal·mol⁻¹. This is followed by AKT1 and Adenosine (−7.1 kcal·mol⁻¹), TNF-Adenosine Triphosphate (−7 kcal·mol⁻¹), TNF-Adenosine (−6.4 kcal·mol⁻¹), and AKT1-Adenosine Triphosphate (−6.4 kcal·mol⁻¹). These findings indicate that the core active components can form a stable ligand-receptor complex system with the target proteins, thus playing a significant role in treating PD. Furthermore, the molecular dynamics simulation results show that the RMSD and Rog of AKT1-Adenosine and IL-6-Adenosine are relatively gentle, implying that the formation of the complex will not cause significant changes in receptor conformation and protein folding. On the other hand, the RMSD and Rog fluctuations of TNF-Adenosine Triphosphate are substantial, indicating that the formation of the complex has an impact on protein conformation and protein folding. In addition, it is worth mentioning that a recent study (Li et al., 2024) found that elevated levels of the pro-inflammatory factor IL-6 can activate the PI3K-AKT pathway.

Studies have shown that the PI3K/AKT pathway is mainly involved in regulating various biological functions such as cell proliferation, cell differentiation, and apoptosis (Thorpe et al., 2015). Of consequence, apoptosis and oxidative stress have been reported to play an important role in the pathogenesis of PD, and inhibiting apoptosis can effectively alleviate PD symptoms (Trist et al., 2019). Recently, research has revealed the significance of activating the PI3K/AKT pathway and its neuroprotective value in PD (Yao et al., 2022; Li et al., 2023b). The PI3K/AKT cascade, which acts on a large number of proteins, is considered a key player in PD models, as research indicates its ability to significantly improve neuronal survival, promote neurogenesis, and inhibit neurotoxin-induced apoptosis (Zheng M. et al., 2021; Khezri and Ghasemnejad-Berenji, 2022; Zheng X. et al., 2021; Wang Q. et al., 2022). Therefore, apoptosis, the most widely explored mechanism of PD neuron loss, was selected for study in this experiment. The results showed that after SJZD intervention, the expressions of PI3K/AKT signaling pathway proteins (p-PI3K, p-AKT) and anti-apoptotic protein (Bcl-2) both increased. Furthermore, it was observed that the expression of these proteins increased in accordance with the concentration of SJZD administered. The higher the expression of the related protein, the stronger the inhibitory impact of SJZD on apoptosis.

This study employed a combination of network pharmacology, molecular docking, and molecular dynamics simulation techniques to construct an “SJZD-Active Components-Intersection Targets” network and PPI network to identify the core active components and action targets of SJZD and to investigate its potential mechanism of action in treating PD. It should be noted that the study has some limitations. First, bioactive compounds and target data were obtained through literature and databases to facilitate predictions, with the reliability and accuracy of these predictions being contingent upon the quality of the data. Future studies can address this issue through liquid chromatography/mass spectrometry (LC/MS) technique, both for the analysis of the active compounds of SJZD, and for the study of metabolomics and pharmacokinetic. Secondly, due to the limitations of network pharmacology, the mechanism of SJZD treatment of PD has only been studied through data mining, and there is a lack of PD animal models and in vivo analyses. Thus, clinical trials and animal experiments are still needed to verify the efficacy and mechanism of SJZD in the treatment of PD. Moreover, the binding sites of SJZD with key targets will become the focus of our subsequent studies.