Network pharmacology of Sijunzi decoction in pancreatic cancer

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is one of the most aggressive malignancies of the digestive system. Globally, PDAC ranks 12th in incidence and 6th in mortality. Its major risk factors include advanced age, smoking, obesity, and metabolic diseases. PDAC characterized by late diagnosis, poor response to conventional therapies, and a dismal five-year survival rate. Despite advances in cancer therapeutics, PDAC remains a significant clinical challenge, necessitating the exploration of novel treatment strategies (1-3).

Traditional Chinese medicine (TCM) has garnered increasing attention in oncology due to its multi-target, low-toxicity, and holistic therapeutic effects (4). Sijunzi decoction (SJZD) is a classic Chinese medicine prescription for treating digestive tract tumors. It originated from the Song Dynasty medical book “Taiping Pingfu Pharmacy Prescription”, which has been written for more than 860 years. It can treat symptoms of spleen deficiency such as weakness, abdominal distension, diarrhea and vomiting. Chinese medicine believes that spleen deficiency leads to decreased immune function, which in turn forms digestive tract tumors. SJZD has the effect of tonifying the spleen and removing dampness, and is widely used in the adjuvant treatment of PDAC (5). The prescription is composed of 4 kinds of herbs: Panax ginseng C.A.Mey. (Ren Shen), Atractylodes macrocephala Koidz. (Bai Zhu), Poria cocos (Schw.) Wolf. (Fu Ling), and Glycyrrhiza uralensis Fisch. (or Glycyrrhiza inflata Bat., or Glycyrrhiza glabra L.) (Gan Cao), the ratio of them is 2:2:2:1. Recent pharmacological studies have demonstrated its potential anticancer properties, particularly in regulating the tumor microenvironment (TME), inhibiting tumor cell proliferation (6), and enhancing immune functions (7). The multi-component and multi-target nature of SJZD suggests that it may exert antitumor effects by modulating key cancer-related signaling pathways, such as phosphatidylinositide 3-kinases (PI3K)/protein Kinase B (AKT), NF-κB, and apoptosis. The PI3K/AKT/mechanistic target of rapamycin kinase (mTOR) signaling axis is frequently hyperactivated in PDAC, primarily driven by KRAS mutations and loss of PTEN expression, and plays a critical role in promoting tumor cell proliferation, survival, metabolism, and chemoresistance. Consequently, targeting the PI3K/AKT/mTOR cascade has emerged as a promising therapeutic strategy for this malignancy (8).

In the context of PDAC, several bioactive compounds in SJZD, including quercetin and kaempferol, have been implicated in suppressing tumor cell growth, inducing apoptosis, and reducing invasiveness. It has been reported that quercetin induces ferroptosis in ovarian cancer cells by regulating the HSPB1/Notch1 pathway, while kaempferol inhibits prostate cancer metastasis and tumor angiogenesis through interfering with the LIMK1/Cofilin pathway (9,10). However, the molecular mechanisms underlying SJZD’s effects on PDAC remain insufficiently understood, particularly regarding its role in complex signaling networks.

To address this gap, this study employs network pharmacology, protein-protein interaction (PPI) network analysis, and functional enrichment analyses to systematically elucidate the potential therapeutic targets and mechanisms of SJZD in PDAC. Key gene modules were identified, with a focus on the PI3K/AKT/mTOR signaling pathway and apoptosis-related processes. These findings were further validated by in vivo and in vitro experiments. Additionally, molecular docking studies were conducted to confirm the interactions between SJZD’s active compounds and their target proteins.

This comprehensive investigation not only provides mechanistic insights into the anti-PDAC effects of SJZD but also lays the groundwork for its potential clinical application as an adjuvant therapeutic strategy for PDAC. We present this article in accordance with the ARRIVE and MDAR reporting checklists (available at https://cco.amegroups.com/article/view/10.21037/cco-2025-1-174/rc).

Methods

Screening of active compounds and targets of SJZD in PDAC

To identify the active compounds and their associated targets in the key herbs of SJZD, including Bai Zhu (BS), Fu Ling (FL), Ren Shen (RS), and Gan Cao (GC), we utilized the Traditional Chinese Medicine Systems Pharmacology (TCMSP) database. We applied two key pharmacokinetic criteria—oral bioavailability (OB ≥30%) and drug-likeness (DL ≥0.2)—to select the most promising compounds for further analysis. These parameters are widely used to evaluate the bioavailability and potential therapeutic efficacy of active compounds.

Screening of PDAC-related targets

PDAC-related targets were retrieved by searching the GeneCards (11) and Online Mendelian Inheritance in Man (OMIM) (12) databases with the keyword “Pancreatic ductal adenocarcinoma”. The retrieved genes were merged, and the intersection with the target genes of the four key herbs was identified. These overlapping genes were considered potential PDAC-related targets for subsequent analyses.

Construction of PPI network

The identified potential PDAC-related targets of SJZD were imported into the STRING platform (13) to construct a PPI network. The network was generated with the following parameters: protein type set to “Homo sapiens” and a minimum interaction score of 0.900, which corresponds to high-confidence interactions. The resulting PPI network provided insights into the interactions and relationships between the targets.

Module extraction

To identify key subnetworks and functional modules within the PPI network, we employed SPICi (14), a community detection algorithm, with default parameters. This method allowed for the extraction of significant functional modules that could be further analyzed for their relevance to PDAC.

Network visualization

The PPI network and its functional modules were visualized using Cytoscape v3.9 (15), a widely used tool for the integration and analysis of biological networks. This enabled a clear representation of the interactions among the targets and the identification of potential hub genes.

Functional enrichment analysis

To investigate the biological functions and pathways associated with the identified PDAC-related targets, Gene Ontology (GO) analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, and Hallmark pathway enrichment analysis were performed using the clusterProfiler R package. Significantly enriched terms and pathways were identified based on a P value threshold of <0.05, which indicated statistical significance.

Construction of multi-tissue-target network

The in vivo mechanism of action of SJZD remains incompletely understood, and its therapeutic effects on PDAC may involve multiple organs and tissues. To explore this, we utilized the BioGPS database (16) to retrieve the mRNA expression profiles of each target gene across various organs and tissues. In this study, tissues with expression levels greater than twice the median expression across all tissues were classified as specific high-expression tissues. This network was subsequently visualized using Cytoscape v3.9 to depict the interactions between the identified targets and relevant tissues.

Molecular docking

The chemical structures of 131 active compounds were obtained in MOL format from the TCMSP database and imported into Maestro 11.8 for conversion to the MAE (Maestro) format. The protein structures of the genes of interest were downloaded from the Research Collaboratory for Structural Bioinformatics (RCSB) Protein Data Bank (17). These protein structures were preprocessed using Maestro 11.8, including the addition of hydrogen atoms, assignment of partial charges and protonation states, removal of water molecules, and energy minimization using the OPLS-2005 force field. The original positions of the protein-bound ligands were preserved to define the ligand-binding site, with the original ligands being removed to generate the protein docking file. The compounds were prepared using the LigPrep module in Maestro, applying the OPLS-2005 force field. The ionization states of the compounds were determined at pH 7.0 using the Epik module. Finally, molecular docking was performed in standard precision (SP) mode of Glide, and the results were visualized using PyMol 3.1.

Preparation of SJZD extract

SJZD consists of four key ingredients: ginseng, Atractylodes macrocephala, Poria cocos, and licorice. The formula was obtained from the Chinese Medicine Pharmacy at The First Affiliated Hospital of Zhejiang University School of Medicine (Hangzhou, China). The preparation method involved soaking 15 g of ginseng, 15 g of Atractylodes macrocephala, 12 g of Poria cocos, and 6 g of licorice in 1 L of pure water for 1 hour. The mixture was then simmered for approximately 5 hours, followed by filtration through gauze to obtain 200 mL of liquid extract. The liquid was subsequently frozen at −80 ℃. After freeze-drying, 7 g of dried powder was collected. In in vitro experiments, SJZD powder was dissolved in culture medium at the required concentrations, with medium without SJZD serving as a negative control.

Cell culture

Human pancreatic cancer cell lines MIA-PaCa-2, CFPAC-1, and human pancreatic ductal epithelial (HPDE) cells HPNE were purchased from ATCC (Manassas, USA). Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin at 37 ℃ in a 5% CO₂ incubator.

Immunoblotting

Cells were collected after treatment with different concentrations of SJZD and washed with pre-cooled PBS. Cells were lysed using RIPA containing 1% protease inhibitor and phosphatase inhibitor, and cell lysate concentration was quantified using BCA kit. The proteins were separated by electrophoresis on SDS-PAGE and transferred to PVDF membrane, which was closed with 5% skim milk powder at room temperature for 2 hours, incubated at 4 ℃ for the corresponding primary antibody overnight, and at 4 ℃ for the second antibody for 2 hours. The immunoreactive bands were detected with enhanced chemiluminescence reagent. The following antibodies were used: anti-AKT (4691, Cell Signaling Technology, 1:1,000, Boston, USA), anti-p-AKT (4060, Cell Signaling Technology, 1:1,000), anti-mTOR (2983, Cell Signaling Technology, 1:1,000), anti-p-mTOR (5536, Cell Signaling Technology, 1:1,000), anti-Cleaved-caspase3 (9661, Cell Signaling Technology, 1:1,000), anti-PARP (9542, Cell Signaling Technology, 1:1,000), and anti-GAPDH (ab128915, Abcam, 1:20,000).

Colony formation assay

The cells were inoculated with 1,000 cells per well in a 12-well plate and treated with the corresponding concentration of SJZD. After treatment, the medium was replaced every three days, fixed with 4% paraformaldehyde one week later, stained with 0.5% crystal violet for 30 minutes, washed with PBS for 30 minutes, photographed visible clones with a camera, and calculated the number of clones using Image J software.

Cell Counting Kit-8 (CCK-8) assay

Cells with appropriate density were inoculated into 96-well plates and treated with SJZD at different concentrations and times. After treatment, the supernatant was discarded and 100 µL serum-free medium containing 10% CCK-8 solution was added to each well for 1–2 h. Optical density (OD) values were measured on a Molecule devices spectrophotometer with a wavelength of 450 nm.

Apoptosis assay

Annexin V-FITC/propidium iodide (PI) apoptosis flow cytometry kit was used to determine the apoptosis rate. The cells were collected and washed with pre-cooled phosphate-buffered saline (PBS), then re-suspended in 500 µL binding buffer with 5 µL Annexin V-FITC and 10 µL PI added to each sample and incubated at room temperature for 5 min. Analysis of at least 50,000 cells on cytoflex LX.

Animal experiments

Male Nude mice (6 weeks old) were purchased from Hangzhou Ziyuan Experimental Animal Technology and housed in a specific-pathogen-free (SPF) environment at the Experimental Animal Center, the First Affiliated Hospital of Zhejiang University School of Medicine. Experiments were performed under a project license (No. 2025-091) granted by the Institutional Animal Care and Use Committee of The First Affiliated Hospital of Zhejiang University School of Medicine, in compliance with the institutional guidelines for the care and use of laboratory animals.

For the patient-derived xenograft (PDX) model, pancreatic cancer tissues were resected, minced into 3–4 mm pieces, and stored in pre-cooled PBS. After the mice were anesthetized, a 1-cm incision was made along the right mid-axillary line, and a subcutaneous cavity was created by blunt dissection (avoiding pectoral muscle damage). Tumor fragments were implanted into the cavity and the incision was closed with sutures. After successful establishment of all animal models (approximately 14 days), the mice were randomly divided into two groups and gavaged with SJZD (10 g/kg) once a day, and PBS was used as the control.

Immunohistochemistry

After tissue samples were harvested, they were fixed with 4% paraformaldehyde, dehydrated, and embedded in paraffin before sectioning. Immunohistochemical staining was performed using anti-Ki67, anti-cleaved-caspase3, anti-p-AKT, and anti-p-mTOR antibodies. The staining protocol followed the method described previously. After imaging with a Leica DM 2500 microscope, ImageJ was used to remove the hematoxylin staining and calculate the integrated optical density (IOD) of the images, followed by statistical analysis. The following antibodies were used: anti-Ki67 (12202, Cell Signaling Technology, 1:200) anti-Cleaved-caspase3 (9661, Cell Signaling Technology, 1:200), anti-p-AKT (4060, Cell Signaling Technology, 1:200), anti-p-mTOR (5536, Cell Signaling Technology, 1:200).

Statistical analysis

All statistical analysis in the study was performed using GraphPad prism 9 software, and data were expressed as the mean ± standard deviation (SD) of three experimental results. Differences between groups were analyzed using one-way analysis of variance (ANOVA) or Student’s t-test, and P<0.05 was considered statistically significant.

Results

Construction of drug-active compound-potential target network

A total of 113 active compounds and 260 corresponding targets from SJZD were screened using the TCMSP database, with criteria of OB ≥30% and DL ≥0.2. The active compounds of SJZD are detailed in table online https://cdn.amegroups.cn/static/public/cco-2025-1-174-1.xlsx. Among these compounds, 7 were derived from BS, 15 from FL, 88 from GC, and 22 from RS. Notably, kaempferol was the only common compound identified in both GC and RS. A total of 4,603 target genes related to PDAC were obtained from the GeneCards and OMIM databases (Figure 1A). By intersecting the targets of these active compounds with PDAC-related targets, 284 drug-disease cross-targets were identified as potential candidates for further research (Figure 1B). Figure 1C illustrates the network of active compounds and their associated targets, with quercetin, kaempferol, and 7-Methoxy-2-methyl isoflavone emerging as the compounds with the most interactions with multiple targets, suggesting that SJZD exerts its therapeutic effect on PDAC through multiple compounds acting on different targets.

Figure 1 Active compounds and targets of SJZD. (A) Targets related to pancreatic cancer. (B) Venn diagram showing the number of active targets of each herbal medicine in SJZD and the degree of overlap with disease targets. (C) “Drug-active compound-potential target” network. Blue hexagons represent compounds in SJZD; yellow quadrilaterals represent potential PDAC-related targets; edges represent interactions between compounds and targets. BS, Bai Zhu; FL, Fu Ling; GC, Gan Cao; OMIM, Online Mendelian Inheritance in Man; PDAC, pancreatic ductal adenocarcinoma; RS, Ren Shen; SJZD, Sijunzi decoction.

Construction of protein interaction network

The potential targets of SJZD (see table online https://cdn.amegroups.cn/static/public/cco-2025-1-174-2.xlsx) identified above were imported into the STRING database to construct a PPI network, with the protein type set to Homo sapiens and a high confidence interaction score (≥0.900) for protein interactions (Figure 2A). The resulting network includes numerous oncogenes, indicating a highly complex interaction landscape. To gain deeper insights into the functional implications of this network, we used SPICi to extract significant modules.

Figure 2 Potential targets identified by PPI analysis. (A) The entire PPI network. (B-D) Subnetworks of the PPI network. (E) The core targets of PPI. PPI, protein-protein interaction.

We used SPICi to obtain some specific modules and further explore the functions of these modules. The first module represents the densest network, centered around the tumor suppressor gene TP53, and includes key genes involved in PDAC biology (Figure 2B). For instance, BAD, a pro-apoptotic factor, promotes cell death by inhibiting the anti-apoptotic proteins BCL2 and BCL2L1. Overexpression of the latter two often prevents apoptosis, thus supporting cancer cell survival (18). RAF1 regulates cell proliferation and survival via the MEK/ERK signaling pathway, and its dysregulation is commonly observed in PDAC (19,20). HIF1A, activated under hypoxic conditions, plays a pivotal role in tumor invasiveness and drug resistance (21). Similarly, JUN and FOS form the AP-1 transcription complex, which governs cell proliferation and apoptosis (22). The sustained activation of STAT3 drives tumor growth through various downstream signaling pathways (23,24), while TP53, a well-known tumor suppressor, is often inactivated in PDAC, leading to the loss of cell cycle control (25,26). CCND1 promotes cell proliferation by modulating the cell cycle (27), and both RELA and estrogen receptor 1 (ESR1)/estrogen receptor 2 (ESR2) regulate inflammation and hormonal signaling, both critical in shaping the TME (28,29).

The second gene module includes various cytokines implicated in inflammation and immune responses. Key genes in this module include CXCL8, IL6, TNF, IL1B, CXCL10, IL10, IFNG, IL1A, CXCL2, IL2, ICAM1, CCL2, and IL4 (Figure 2C). These cytokines are crucial for regulating the TME, particularly the infiltration and activation of immune cells. CXCL8 and CCL2 are chemokines that recruit immune cells to the tumor site, enhancing the inflammatory response (30). IL6 and TNF are prominent proinflammatory cytokines involved in the progression and metastasis of PDAC (31). Conversely, IFNG and IL2 promote T-cell activation and anti-tumor immunity, although their efficacy may be diminished in the TME, facilitating tumor immune evasion (32-34). ICAM1 is integral to immune cell adhesion, influencing the invasiveness and metastatic capacity of tumor cells (35). This module highlights the immune regulatory network within the PDAC microenvironment, offering potential therapeutic targets for modulating inflammation and immune escape mechanisms.

The third module is enriched with enzymes involved in drug metabolism and steroid hormone biosynthesis. Key enzymes include CYP1A2, SULT1E1, HSD3B2, CYP1A1, CYP3A4, AKR1C3, HSD3B1, UGT1A1, AKR1C1, and CYP19A1 (Figure 2D). These enzymes play critical roles in the metabolism of drugs and steroid hormones, influencing the response to therapy and the progression of PDAC. CYP1A1 and CYP1A2, for example, metabolize drugs and environmental toxins (36), and their altered expression can impact the therapeutic efficacy of chemotherapeutic agents in PDAC patients (37). CYP3A4 is a major enzyme involved in drug metabolism, and its activity may be modulated in PDAC, affecting drug toxicity and efficacy (38). Enzymes involved in steroid hormone biosynthesis, such as HSD3B1/2 and CYP19A1 (aromatase), contribute to the growth and survival of PDAC cells by regulating steroid hormones (39). Targeting these enzymes may offer new therapeutic avenues for managing PDAC, particularly in overcoming drug resistance and hormonal regulation.

A count of the most highly connected genes within Module 1 (Figure 2E) reveals key oncogenes implicated in various cancer-related biological processes including cell proliferation, apoptosis, invasion, metastasis, and drug resistance. These highly connected genes are predominantly found within Module 1, reinforcing the hypothesis that SJZD exerts its primary anti-tumor effects through this module. Future research will focus on the in-depth study of Module 1’s key genes.

Construction of multi-tissue-target network

To investigate the therapeutic effect of SJZD on PDAC, we first examined the mRNA expression levels of the genes in Module 1 using the BioGPS database. The results revealed that many of the target genes were highly expressed in a variety of immune cells, indicating that the therapeutic effects of SJZD on PDAC likely involve multiple immune processes (Figure 3A).

Figure 3 Functional enrichment analysis results. (A) Tissue-target network. Red arrows indicate tissue locations, and yellow quadrilaterals indicate targets. (B) GO pathway enrichment analysis. (C) KEGG pathway enrichment analysis. (D) Hallmark pathway enrichment analysis. BP, biological process; CC, cellular component; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; MF, molecular function.

Functional enrichment analysis

The integrated GO, KEGG, and Hallmark pathway enrichment analyses revealed that the genes of interest are significantly involved in cellular stress responses, cancer-related pathways, and immune regulation—all of which are critical factors in cancer progression and therapeutic responses. The enriched biological processes, such as cellular response to chemical and oxidative stress and response to radiation, highlight the importance of these genes in cellular damage control. Additionally, pathways associated with neuron death and autophagy suggest potential roles in regulating cell survival and programmed cell death (Figure 3B).

The KEGG analysis identified several important pathways, including endocrine resistance, colorectal cancer, prolactin signaling pathway, and apoptosis, emphasizing potential roles in tumor development, hormonal regulation, and immune modulation (Figure 3C). In the Hallmark analysis, key pathways such as PI3K/AKT/mTOR signaling, apoptosis, and tumor necrosis factor-alpha (TNF-α) signaling via NF-kB were highlighted, underscoring their relevance to cell growth, immune response, and apoptosis. Other pathways, including IL-2/STAT5 signaling and Hypoxia, point to the module’s potential role in immune signaling and low-oxygen conditions commonly found in the TME (Figure 3D).

Together, these findings suggest that apoptosis and PI3K/AKT/mTOR signaling play pivotal roles in mediating the antitumor effects of the genes in Module 1.

Validation of the anti-PDAC effect of SJZD in vitro

To further verify the anti-PDAC effect of SJZD, cell growth was examined following SJZD treatment in two PDAC cell lines (CFPAC-1 and MiaPaCa-2). As shown in Figure 4A,4B, SJZD treatment markedly reduced the cell viability of CFPAC-1 and MiaPaCa-2 cells in both a dose- and time-dependent manner. Consistently, the cell number of tested PDAC cells was found to decreased to SJZD at 10, 30 and 50 g/L over a period of 72 hours (Figure 4C,4D). Moreover, the colony formation assay revealed that SJZD treatment significantly inhibited the proliferation of PDAC cells (Figure 4E,4F). Of note, HPDE cells showed higher tolerance to SJZD in comparison to PDAC cells, as evidenced by an elevated half-maximal inhibitory concentration (IC₅₀) value (Figure 4G,4H). Together, these results indicate that SJZD demonstrates a considerable anti-cancer effect in PDAC cells in vitro.

Figure 4 SJZD inhibits the growth of pancreatic cancer in vitro. (A,B) CCK-8 assay of CFPAC-1 and MiaPaCa-2 cells treated with indicated concentrations of SJZD for 24, 48 and 72 hours. (C,D) Cell counting assay of CFPAC-1 and MiaPaCa-2 cells treated with indicated concentrations of SJZD for 24, 48 and 72 hours. (E,F) Colony formation assay of CFPAC-1 and MiaPaCa-2 cells treated with indicated concentrations of SJZD. Representative images (above) and quantification of colonies (bottom) were shown. **, P<0.01; ***, P<0.001. (G) CCK-8 assay of HPNE cells treated with indicated concentrations of SJZD for 24, 48 and 72 hours. (H) IC50 values of SJZD at different time points for the three cell lines. CCK-8, Cell Counting Kit-8; HPNE, human pancreatic ductal epithelial; IC50, half-maximal inhibitory concentration; SJZD, Sijunzi decoction.

Validation of the functional pathways involved in SJZD

The enrichment analysis results in Figure 3 demonstrate that SJZD may exert anti-tumor effects mainly by regulating the PI3K/AKT/mTOR signaling pathway and apoptosis. To ascertain whether PI3K/AKT/mTOR signaling was modulated by SJZD in PDAC cells, we examined the levels of some classic markers, including AKT, p-AKT, mTOR, and p-mTOR. We observed decreased levels of these proteins upon SJZD treatment (Figure 5A,5B; Figures S1,S2), indicating inhibition of PI3K/AKT/mTOR signaling. Furthermore, we examined the protein expression of cleaved-caspase3 and cleaved-PARP following SJZD treatment and found increased expression in SJZD-treated PDAC cells, indicating an obvious effect of SJZD on apoptosis induction in PDAC cells (Figure 5A, Figures S1,S2). Meanwhile, we evaluated the apoptotic ratio using flow cytometry assays. As shown in Figure 5C,5D, SJZD treatment for 12 hours induced apoptosis of PDAC cells in a dose-dependent manner. Taken together, our data demonstrate that SJZD inhibits PDAC growth by inhibiting PI3K/AKT/mTOR signaling and inducing apoptosis.

Figure 5 The anti-cancer effect of SJZD is mediated by PI3K/AKT/mTOR signaling and apoptosis. (A) Immunoblotting of total and phosphorylated AKT, mTOR and cleaved PARP or caspase 3 in PDAC cells treated with the indicated concentrations of SJZD for 12 hours. (B) Statistical analysis of p-AKT/AKT and p-mTOR/mTOR in PDAC cells. (C) Flow cytometric analysis of apoptosis in PDAC cells treated with the indicated concentrations of SJZD for 12 hours. (D) Statistical analysis of the apoptotic rate in PDAC cells treated with the indicated concentrations of SJZD for 12 hours. *, P<0.05; **, P<0.01; ***, P<0.001. PDAC, pancreatic ductal adenocarcinoma; PI, propidium iodide; SJZD, Sijunzi decoction.

Validation of the anti-PDAC effect and related pathways of SJZD in vivo

Following the experimental outlined in Figure 6A, we evaluated the antitumor activity and related pathways of SJZD using PDX models. As shown in Figure 6B,6C, tumor volume was significantly suppressed in the SJZD-treated group compared to the control group. Similarly, tumor weight in the SJZD-treated group was lower than that of the control group (Figure 6D), demonstrating that SJZD also displayed anti-PDAC effect in vivo. To verify whether SJZD exerted its therapeutic effect through modulation of the PI3K/AKT/mTOR signaling, we performed immunohistochemical analysis of tumor tissues. The results revealed that SJZD treatment significantly inhibited the expression of the proliferation marker Ki67 while markedly upregulating the expression of the apoptosis marker cleaved-caspase3 compared to the control group (Figure 6E,6F). In addition, we further examined the phosphorylation levels of AKT and mTOR. As anticipated, SJZD treatment led to significant downregulation of phosphorylated AKT and mTOR (Figure 6E,6F). Collectively, these data demonstrate that SJZD exhibits remarkable antitumor efficacy in vivo, which can be attributed to the induction of apoptosis through modulation of the PI3K/AKT/mTOR signaling pathway.

Figure 6 SJZD induces pancreatic cancer cell apoptosis by modulating PI3K/AKT/mTOR signaling in vivo. (A) Schematic of the PDX mouse model experimental design. (B) Tumor growth curves showing volume measurements at indicated time points (n=5; *, P<0.05, two-way ANOVA). (C) Representative tumor images at endpoint. (D) Comparison of final tumor weights (*, P<0.05, unpaired t-test). (E) Immunohistochemical analysis of Ki67, cleaved caspase-3, p-AKT and p-mTOR (scale bar =40 µm). (F) Quantification of immunohistochemistry scores (n=5; ***, P<0.001, and ****, P<0.0001, unpaired t-test). Data represent mean ± SEM. ANOVA, analysis of variance; PDX, patient-derived xenograft; SEM, standard error of the mean; SJZD, Sijunzi decoction.

Molecular docking

To verify the results of network pharmacology, we used molecular docking to calculate the binding affinity between the evaluated compounds and the targets. We focused on the targets in Module 1 and 131 compounds for molecular docking. The docking scores (see table online https://cdn.amegroups.cn/static/public/cco-2025-1-174-3.xlsx) are shown in the heat map (Figure 7A). It can be seen that the docking scores of ESR1 and ESR2 with some active compounds are significantly lower than those of other proteins, especially ESR1 with a variety of active compounds. Figure 7B shows the binding details of ESR1 with glyzaglabrin, which exhibits the highest absolute affinity among ESR1 interactions. Figure 7C displays the binding details of ESR2 with Glabrene, which shows the highest absolute affinity among ESR2 interactions. The docking scores are low, which suggests that these two proteins may be potential binding targets for SJZD to exert its effects.

Figure 7 Molecular docking. (A) The heat map of docking score between core targets and active compounds (unit: kcal/mol). (B) The binding modes between ESR1 and the active compounds in SJZD (MOL004907). (C) The binding modes between ESR2 and the active compounds in SJZD (MOL004911). ESR1, estrogen receptor 1; ESR2, estrogen receptor 2; SJZD, Sijunzi decoction.

Discussion

This study explores the mechanisms underlying the therapeutic effects of SJZD on PDAC using network pharmacology, molecular docking, and experimental validation. PDAC is one of the most aggressive malignancies, characterized by poor prognosis and high mortality rates. Its late diagnosis, rapid progression, and resistance to conventional therapies pose substantial challenges in clinical management. Despite advancements in surgical techniques, chemotherapy, and targeted therapies, the 5-year survival rate for PDAC remains dismally low, at less than 10% in advanced stages (40,41). These clinical challenges necessitate innovative therapeutic approaches, and SJZD, rooted in TCM, offers promising potential.

In TCM theory, the spleen (“Pi”) plays a pivotal role in digestion, metabolism, and immune regulation. Weakness of Pi qi and the accumulation of “dampness toxins” are considered key pathogenic mechanisms underlying PDAC (42). SJZD is a classic formula designed to invigorate Pi qi, eliminate dampness, and restore metabolic and immune homeostasis (43).

This study identified and validated critical bioactive compounds within SJZD, including kaempferol, isorhamnetin, and ginsenosides, which exhibit significant anti-tumor effects. For instance, kaempferol has been shown to induce apoptosis in PDAC cells by regulating the Bcl-2 family proteins, inhibiting Akt kinase activity, and disrupting mitochondrial integrity (44). Furthermore, its combination with standard chemotherapy agents like 5-fluorouracil (5-FU) enhances therapeutic efficacy and overcomes drug resistance (45). Similarly, isorhamnetin inhibits PDAC progression by reducing ROS-mediated HIF-1α expression and suppressing β-catenin nuclear translocation, key processes in tumor metastasis (46).

Molecular docking analysis revealed strong binding affinities between SJZD compounds and PDAC-related targets. Notably, functional enrichment and KEGG pathway analyses highlighted the PI3K/Akt/mTOR signaling pathway as a critical therapeutic target of SJZD. This pathway, extensively implicated in PDAC pathogenesis, regulates cell proliferation, apoptosis, angiogenesis, and autophagy. Dysregulation of this signaling cascade contributes to the aggressiveness and therapeutic resistance of PDAC. Our findings demonstrate that SJZD significantly inhibits the phosphorylation of PI3K, Akt, and mTOR in PDAC cells, thereby suppressing tumor growth and promoting apoptosis. Moreover, SJZD enhances autophagic activity, as evidenced by increased LC3B expression and autophagosome formation, which may further contribute to its anti-tumor effects.

Beyond its direct anti-cancer mechanisms, SJZD possesses potent anti-inflammatory properties, mediated by bioactive compounds such as cerevisterol and isotrifoliol, which inhibit inflammatory cytokines like IL-6 and TNF-α and suppress the NF-κB pathway (47). Chronic inflammation is a well-recognized driver of PDAC progression, and SJZD’s ability to mitigate this microenvironmental factor adds an additional therapeutic advantage. Moreover, SJZD regulates gut microbiota composition, which has emerging relevance in PDAC pathogenesis and therapy.

Despite these promising results, our study has limitations. The predictions based on network pharmacology and molecular docking require further in vivo validation to confirm the clinical relevance of SJZD. Additionally, while this study focused on apoptosis and autophagy, other tumor-related processes, such as invasion and metastasis, remain to be explored. The precise roles of individual SJZD components and their synergistic interactions also warrant further investigation.

Conclusions

In conclusion, SJZD demonstrates significant therapeutic potential against pancreatic cancer by targeting the PI3K/Akt/mTOR signaling pathway, modulating apoptosis and autophagy, reducing inflammation, and improving gut microbiota balance. These findings provide a solid foundation for future studies to further investigate the clinical applicability of SJZD as a complementary therapeutic strategy for PDAC.

Acknowledgments

The authors used ChatGPT to assist with language polishing during the preparation of this manuscript. The authors have reviewed and edited the content as needed and take full responsibility for the content of the publication.

Footnote

Reporting Checklist: The authors have completed the ARRIVE and MDAR reporting checklists. Available at https://cco.amegroups.com/article/view/10.21037/cco-2025-1-174/rc

Data Sharing Statement: Available at https://cco.amegroups.com/article/view/10.21037/cco-2025-1-174/dss

Peer Review File: Available at https://cco.amegroups.com/article/view/10.21037/cco-2025-1-174/prf

Funding: This research was supported by the National Natural Science Foundation of China (No. 82504086) and Zhejiang Provincial Natural Science Foundation of China (No. LQN25H310007).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://cco.amegroups.com/article/view/10.21037/cco-2025-1-174/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Experiments were performed under a project license (No. 2025-091) granted by the Institutional Animal Care and Use Committee of The First Affiliated Hospital of Zhejiang University School of Medicine, in compliance with the institutional guidelines for the care and use of laboratory animals.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. McGuigan A, Kelly P, Turkington RC, et al. Pancreatic cancer: A review of clinical diagnosis, epidemiology, treatment and outcomes. World J Gastroenterol 2018;24:4846-61. [Crossref] [PubMed]
2. Wang S, Zheng Y, Yang F, et al. The molecular biology of pancreatic adenocarcinoma: translational challenges and clinical perspectives. Signal Transduct Target Ther 2021;6:249. [Crossref] [PubMed]
3. Mannucci A, Goel A. Advances in pancreatic cancer early diagnosis, prevention, and treatment: The past, the present, and the future. CA Cancer J Clin 2026;76:e70035. [Crossref] [PubMed]
4. Wu J, Tang G, Cheng CS, et al. Traditional Chinese medicine for the treatment of cancers of hepatobiliary system: from clinical evidence to drug discovery. Mol Cancer 2024;23:218. [Crossref] [PubMed]
5. Jia J, Qin Y, Zhang L, et al. Sijunzi decoction-treated rat serum induces apoptosis of side population cells in gastric carcinoma. Exp Ther Med 2018;15:1718-27. [Crossref] [PubMed]
6. Wu B, Xuan ZR. Progress in research on applying Sijunzi Decoction in treating digestive malignant tumor. Chin J Integr Med 2007;13:156-9. [Crossref] [PubMed]
7. Gao B, Peng Y, Peng C, et al. A Comparison of Characterization and Its Actions on Immunocompetent Cells of Polysaccharides from Sijunzi Decoction. Evid Based Complement Alternat Med 2019;2019:9860381. [Crossref] [PubMed]
8. Rozengurt E, Eibl G. Pancreatic cancer: molecular pathogenesis and emerging therapeutic strategies. Signal Transduct Target Ther 2026;11:6. [Crossref] [PubMed]
9. Zhao B, Zhu H, Qian H, et al. Quercetin induces ferroptosis in ovarian cancer through regulating HSPB1/Notch1 pathway. J Ovarian Res 2026;19:75. [Crossref] [PubMed]
10. Li P, Lu L, Sun Y. Kaempferol suppresses prostate cancer metastasis and tumor angiogenesis via disrupting the LIMK1/Cofilin pathway. Biochem Biophys Res Commun 2026;797:153145. [Crossref] [PubMed]
11. Stelzer G, Rosen N, Plaschkes I, et al. The GeneCards Suite: From Gene Data Mining to Disease Genome Sequence Analyses. Curr Protoc Bioinformatics 2016;54:1.30.1-1.30.33.
12. Amberger JS, Bocchini CA, Schiettecatte F, et al. OMIM.org: Online Mendelian Inheritance in Man (OMIM®), an online catalog of human genes and genetic disorders. Nucleic Acids Research 2015;43:D789-D798.
13. Szklarczyk D, Gable AL, Nastou KC, et al. The STRING database in 2021: customizable protein-protein networks, and functional characterization of user-uploaded gene/measurement sets. Nucleic Acids Res 2021;49:D605-12. [Crossref] [PubMed]
14. Jiang P, Singh M. SPICi: a fast clustering algorithm for large biological networks. Bioinformatics 2010;26:1105-11. [Crossref] [PubMed]
15. Shannon P, Markiel A, Ozier O, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 2003;13:2498-504. [Crossref] [PubMed]
16. Wu C, Jin X, Tsueng G, et al. BioGPS: building your own mash-up of gene annotations and expression profiles. Nucleic Acids Res 2016;44:D313-6. [Crossref] [PubMed]
17. Goodsell DS, Zardecki C, Di Costanzo L, et al. RCSB Protein Data Bank: Enabling biomedical research and drug discovery. Protein Sci 2020;29:52-65. [Crossref] [PubMed]
18. Kaloni D, Diepstraten ST, Strasser A, et al. BCL-2 protein family: attractive targets for cancer therapy. Apoptosis 2023;28:20-38. [Crossref] [PubMed]
19. Bahar ME, Kim HJ, Kim DR. Targeting the RAS/RAF/MAPK pathway for cancer therapy: from mechanism to clinical studies. Signal Transduct Target Ther 2023;8:455. [Crossref] [PubMed]
20. McCubrey JA, Steelman LS, Chappell WH, et al. Roles of the Raf/MEK/ERK pathway in cell growth, malignant transformation and drug resistance. Biochim Biophys Acta 2007;1773:1263-84. [Crossref] [PubMed]
21. Denko NC. Hypoxia, HIF1 and glucose metabolism in the solid tumour. Nat Rev Cancer 2008;8:705-13. [Crossref] [PubMed]
22. Nagesh R, Kiran Kumar KM, Naveen Kumar M, et al. Regulation of Jun and Fos AP-1 transcription factors by JNK MAPKs signaling cascade in areca nut extract treated KB cells. Biochem Biophys Rep 2021;27:101090. [Crossref] [PubMed]
23. Gu Y, Mohammad IS, Liu Z. Overview of the STAT-3 signaling pathway in cancer and the development of specific inhibitors. Oncol Lett 2020;19:2585-94. [Crossref] [PubMed]
24. Wang HQ, Man QW, Huo FY, et al. STAT3 pathway in cancers: Past, present, and future. MedComm (2020) 2022;3:e124.
25. Rivlin N, Brosh R, Oren M, et al. Mutations in the p53 Tumor Suppressor Gene: Important Milestones at the Various Steps of Tumorigenesis. Genes Cancer 2011;2:466-74. [Crossref] [PubMed]
26. Konishi M, Kameyama KI, Yamamoto T. Mutations in the tumor suppressor gene p53 in cattle are associated with enzootic bovine leukosis. Vet Microbiol 2021;263:109269. [Crossref] [PubMed]
27. Montalto FI, De Amicis F. Cyclin D1 in Cancer: A Molecular Connection for Cell Cycle Control, Adhesion and Invasion in Tumor and Stroma. Cells 2020;9:2648. [Crossref] [PubMed]
28. Chen L, Deng H, Cui H, et al. Inflammatory responses and inflammation-associated diseases in organs. Oncotarget 2018;9:7204-18. [Crossref] [PubMed]
29. Kovats S. Estrogen receptors regulate innate immune cells and signaling pathways. Cell Immunol 2015;294:63-9. [Crossref] [PubMed]
30. Han ZJ, Li YB, Yang LX, et al. Roles of the CXCL8-CXCR1/2 Axis in the Tumor Microenvironment and Immunotherapy. Molecules 2021;27:137. [Crossref] [PubMed]
31. van Duijneveldt G, Griffin MDW, Putoczki TL. Emerging roles for the IL-6 family of cytokines in pancreatic cancer. Clin Sci (Lond) 2020;134:2091-115. [Crossref] [PubMed]
32. Ding H, Wang G, Yu Z, et al. Role of interferon-gamma (IFN-γ) and IFN-γ receptor 1/2 (IFNγR1/2) in regulation of immunity, infection, and cancer development: IFN-γ-dependent or independent pathway. Biomedicine & Pharmacotherapy 2022;155:113683.
33. Niederlova V, Tsyklauri O, Kovar M, et al. IL-2-driven CD8(+) T cell phenotypes: implications for immunotherapy. Trends Immunol 2023;44:890-901. [Crossref] [PubMed]
34. Im SJ, Lee K, Ha SJ. Harnessing IL-2 for immunotherapy against cancer and chronic infection: a historical perspective and emerging trends. Exp Mol Med 2024;56:1900-8. [Crossref] [PubMed]
35. Qiu Z, Wang Y, Zhang Z, et al. Roles of intercellular cell adhesion molecule-1 (ICAM-1) in colorectal cancer: expression, functions, prognosis, tumorigenesis, polymorphisms and therapeutic implications. Front Oncol 2022;12:1052672. [Crossref] [PubMed]
36. Kapelyukh Y, Henderson CJ, Scheer N, et al. Defining the Contribution of CYP1A1 and CYP1A2 to Drug Metabolism Using Humanized CYP1A1/1A2 and Cyp1a1/Cyp1a2 Knockout Mice. Drug Metab Dispos 2019;47:907-18. [Crossref] [PubMed]
37. Androutsopoulos VP, Tsatsakis AM, Spandidos DA. Cytochrome P450 CYP1A1: wider roles in cancer progression and prevention. BMC Cancer 2009;9:187. [Crossref] [PubMed]
38. Fatunde OA, Brown SA. The Role of CYP450 Drug Metabolism in Precision Cardio-Oncology. Int J Mol Sci 2020;21:604. [Crossref] [PubMed]
39. Lai TC, Li HF, Li YS, et al. Proximal GATA-binding sites are essential for human HSD3B1 gene transcription in the placenta. Sci Rep 2017;7:4271. [Crossref] [PubMed]
40. Bazeed AY, Day CM, Garg S. Pancreatic Cancer: Challenges and Opportunities in Locoregional Therapies. Cancers (Basel) 2022;14:4257. [Crossref] [PubMed]
41. Hu JX, Zhao CF, Chen WB, et al. Pancreatic cancer: A review of epidemiology, trend, and risk factors. World J Gastroenterol 2021;27:4298-321. [Crossref] [PubMed]
42. Lan Q, Fang J, Yu X, et al. Analysis of traditional Chinese medicine syndrome elements and clinical symptoms in prediabetes: A systematic review. Medicine (Baltimore) 2024;103:e36789. [Crossref] [PubMed]
43. Shang L, Wang Y, Li J, et al. Mechanism of Sijunzi Decoction in the treatment of colorectal cancer based on network pharmacology and experimental validation. J Ethnopharmacol 2023;302:115876. [Crossref] [PubMed]
44. Imran M, Salehi B, Sharifi-Rad J, et al. Kaempferol: A Key Emphasis to Its Anticancer Potential. Molecules 2019;24:2277. [Crossref] [PubMed]
45. Azwar S, Seow HF, Abdullah M, et al. Recent Updates on Mechanisms of Resistance to 5-Fluorouracil and Reversal Strategies in Colon Cancer Treatment. Biology (Basel) 2021;10:854. [Crossref] [PubMed]
46. Asgharian P, Tazekand AP, Hosseini K, et al. Potential mechanisms of quercetin in cancer prevention: focus on cellular and molecular targets. Cancer Cell Int 2022;22:257. [Crossref] [PubMed]
47. Alam MB, Chowdhury NS, Sohrab MH, et al. Cerevisterol Alleviates Inflammation via Suppression of MAPK/NF-κB/AP-1 and Activation of the Nrf2/HO-1 Signaling Cascade. Biomolecules 2020;10:199. [Crossref] [PubMed]