Integrated Network Pharmacology, Transcriptomics, and Experimental Validation Identify PI3K-AKT and STAT3 as Key Pathways for Pien Tze Huang Against Liver and Colorectal Cancers

Introduction: Pien Tze Huang (PZH), a classic Traditional Chinese Medicine (TCM), has long been used for treating inflammation and cancer, yet its pharmacological mechanisms against liver cancer and Colorectal Cancer (CRC) remain incompletely understood. This study aims to elucidate the potential antitumor targets, signaling pathways, and functional effects of PZH against liver cancer and CRC. Methods: Network pharmacology was applied to identify the active components of PZH from the TCMSP database and predict their potential targets, while disease-associated targets were retrieved from GeneCards. Core targets and pathways were screened via Protein-Protein Interaction (PPI) network analysis and functional enrichment (GO/KEGG) using Metascape. Molecular simulation was performed to assess the binding of key components to core targets. Experimental validation in Huh7, HCT116, and RKO cells included assessments of cell viability, cell cycle, protein expression, and transcriptomic profiling using CCK-8, flow cytometry, Western blot, and RNA sequencing, respectively. results: Identification of targets of PZH against liver cancer and CRC According to the Chinese Pharmacopoeia, PZH is primarily composed of four herbal ingredients including Bovis calculus, Moschus, Panax notoginseng, and Snake gall. We identified the main active components of these herbs using the TCMSP database, supplemented by literature review, which yielded 15 primary active components of PZH (Table 1). Subsequently, the potential protein targets of these components were predicted using the SwissTargetPrediction database, resulting in 355 unique targets after removing duplicates. Meanwhile, 1,833 and 1,488 targets associated with liver cancer and CRC were collected (Fig. 1). The intersection of these target sets via a Venn diagram using Venny 2.1.0 revealed 138 and 102 potential therapeutic targets for PZH against liver cancer and CRC, respectively (Fig. 1). Screening of core targets in the PPI network of PZH against liver cancer and CRC The potential targets for PZH against liver cancer and CRC identified above were imported into the STRING database to construct PPI networks (Fig. 2A for liver cancer; Fig. 2B for CRC). The generated PPI data were subsequently imported into Cytoscape software (version 3.7.2) for topological analysis. Core targets were screened based on a betweenness centrality value of ≥ 100, which yielded 36 and 19 core targets for liver cancer and CRC, respectively. The top 10 core targets ranked by betweenness centrality for each cancer are listed in Table 2 and Table 3. Notably, several prominent nodes were identified in both PPI networks, including AKT1, STAT3, ESR1, EGFR, and MAPK3 (ERK1), which are established key signaling molecules regulating cancer cell proliferation, survival and immune regulation 28, 29. Based on these findings, compound-core target network diagrams were plotted for PZH against liver cancer (Fig. 3A) and CRC (Fig. 3B), in which orange octagons represent the four herbal components of PZH, and amber circles represent the core targets. Collectively, the top-ranked targets by betweenness centrality, such as AKT1, STAT3, and MAPK3, were shared between liver cancer and CRC. Given their roles as critical downstream effectors of EGFR signaling, these results suggest that the therapeutic effects of PZH against both cancers may be mediated through coordinated modulation of the EGFR-associated PI3K-AKT, STAT3, and ERK signaling pathways. GO and KEGG enrichment analysis of core targets To elucidate the biological processes and signaling pathways associated with the candidate core targets, GO and KEGG enrichment analysis were performed using the Metascape database. The GO enrichment analysis for biological processes highlighted that the core targets of PZH were mainly involved in receptor-mediated signaling and cancer-related cellular behaviors. Specifically, in liver cancer, the core targets were enriched in biological processes such as the enzyme-linked receptor protein signaling pathway, cell surface receptor signaling pathway involving protein tyrosine kinase, as well as positive regulation of cell migration and motility (Fig. 4A). In CRC, the enriched biological processes were primarily associated with epidermal growth factor receptor signaling pathway, ErbB signaling pathway, positive regulation of response to external stimulus, and regulation of DNA metabolic process (Fig. 4B). KEGG pathway enrichment analysis further identified that multiple oncogenic and immune-related signaling pathways were significantly associated with the core targets. For liver cancer, enriched pathways included the PI3K-AKT, prolactin, estrogen, and ErbB signaling pathways, along with the PD-L1/PD-1 checkpoint pathway in cancer (Fig. 4C). In CRC, the core targets were predominantly enriched in the VEGF, prolactin, ErbB, C-type lectin receptor, and thyroid hormone signaling pathways (Fig. 4D). These pathways are critically involved in the pathogenesis and progression of both cancers. Thus, these results demonstrate that the therapeutic effects of PZH against liver cancer and CRC are mediated through the modulation of a broad spectrum of biological processes and key signaling pathways, underscoring its characteristic multi-component, multi-target, and multi-pathway mechanism of action. Molecular simulation analysis of interactions between PZH compounds and core targets To further explore the structural basis underlying the interactions between the screened core targets and active components of PZH, molecular simulations were performed. Based on their central roles in the PPI network and established roles in cancer-related signaling, AKT1 and STAT3 were selected as representative core targets for further investigation. Among the active components of PZH, muscone (Fig. 5A, B) and quercetin (Fig. 5C, D) were prioritized for evaluating potential binding interactions with AKT1, while ginsenoside F2 (Fig. 5E, F) and ginsenoside Rh2 (Fig. 5G, H) were selected for assessment with STAT3. Negative total binding energies (≤-7 kJ/mol) indicate favorable binding for all systems . Among these, the binding energies of muscone (-10.3 kJ/mol) and quercetin (-10.0 kJ/mol) with AKT1 are approximately equivalent (Fig. 5A, C). The STAT3-ginsenoside F2 system -9.6 kJ/mol exhibited a higher binding energy compared to the STAT3-ginsenoside Rh2 system -8.7 kJ/mol (Fig. 5E, G). Interactions were evaluated using geometric criteria (hydrophobic distances and hydrogen-bonding analyses) and energetic criteria (binding energies). Overall, the favorable binding energetics of the active components of PZH at the active site were primarily driven by a combination of hydrogen bonding and interactions with hydrophobic residues. Hydrogen bonding at the active site is critical for the binding of PZH’s active components. Specifically, muscone and quercetin form hydrogen bonds with key residues of AKT1, including Glu234, Asp274, Lys276, and Lys297 (Fig. 5B, D). Ginsenoside F2 and ginsenoside Rh2 establish hydrogen bonds with key residues of STAT3, namely Asp334, Lys574, Lys615, and Lys642 (Fig. 5F, H). Notably, all these key interacting amino acids are all charged residues. These computational findings provide structural evidence that active components of PZH can directly engage with core targets such as AKT1 and STAT3, lending further support to the network pharmacology predictions and offering a plausible structural basis for the multi-target mechanism of PZH against liver cancer and CRC. PZH suppresses cancer cell proliferation and induces cell cycle arrest in vitro To experimentally validate the predicted anti-proliferative effects of PZH, we evaluated its impact on cell viability and cell cycle progression. Human hepatocellular carcinoma Huh7 cells and colorectal carcinoma HCT116 cells were treated with PZH, and cell viability was assessed using the CCK-8 assay. As shown in Fig. 6A and 6B, PZH treatment (1 mg/mL) significantly reduced the viability of both Huh7 and HCT116 cells compared with the control group, demonstrating a potent anti-proliferative effect. To investigate whether this growth inhibition was associated with cell cycle regulation, cell cycle distribution was analyzed by flow cytometry following PI staining. The results showed that PZH treatment led to a notable accumulation of cells in the G1 phase in both Huh7 and HCT116 cells (Fig. 6C-F), indicating that PZH suppresses cancer cell proliferation, at least in part, through induction of G1 phase arrest. G1 phase progression is tightly regulated by signaling pathways controlling cyclin-dependent kinase activity, among which PI3K-AKT and STAT3 pathways play pivotal roles 30, 31. Therefore, the observed G1 arrest phenotype aligns with the network pharmacology prediction that PZH may interfere with key oncogenic signaling pathways governing cell cycle progression. PZH suppresses proliferation-associated regulators and PD-L1 in liver cancer and CRC cells To further elucidate the molecular mechanism underlying PZH-induced cell cycle arrest and growth inhibition, we performed RNA-sequencing (RNA-seq) using the human CRC cell line RKO treated with PZH. The heatmap of differentially expressed genes revealed a marked downregulation of multiple proliferation-related genes, including CCND1, PCNA, MYB (Fig. 7A, red box). Notably, transcript levels of the immune checkpoint gene CD274 (also known as PD-L1) were also significantly reduced (Fig. 7A, red box). We further validated these findings at the protein level by W