Background: Pancreatic ductal adenocarcinoma (PDAC) exhibits five-year survival rates below 10% with profound therapeutic resistance mediated by stromal barriers and immunosuppression [1,2]. Recent evidence suggests intratumoral mycobiome alterations may influence disease progression [17,18].
Methods: We performed comprehensive network pharmacology analysis to elucidate potential therapeutic mechanisms of Saccharomyces boulardii (SB) and Clostridium histolyticum collagenase (CHC) in PDAC. Gene sets representing immune activation, metabolic competition, and stromal remodeling were analyzed using protein-protein interaction networks (STRING v11.5), functional enrichment (DAVID v2021), and pathway databases (KEGG, Reactome, WikiPathways, DisGeNET). Structural validation of core targets was conducted using blind molecular docking (CB-Dock2) and coarse-grained molecular dynamics simulations (CABS-flex).
Results: Network analysis identified dense interconnectivity among 20 core immune-metabolic genes. Cytoscape-based topological analysis highlighted TNF, IL6, IFNG, and TLR4 as central hubs (degree >15). Intersection analysis (using a literature-derived CHC stromal remodeling signature) identified IL6, TLR2, MMP9, and IL1B as convergence candidates. Across KEGG, Reactome, and WikiPathways, the curated 20-gene immune signature was over-represented in IL-17, Toll-like receptor, and IL-10 signaling and aligned with the Immune infiltration in pancreatic cancer pathway (WP5285); these findings were obtained using BH-FDR within each database and are contingent on the background gene universe. Co-occurrence analysis across 10,897 tumors revealed coordinated expression of IFNG-TLR4 and IFNG-CD86 gene pairs. Structural validation via blind molecular docking substantiated these associations, revealing high-affinity interactions between β-glucan surrogates and the ATP-binding pocket of AKT1 (Δ G = -7.2 kcal/mol), stabilized by hydrogen bonding with the catalytic residue Lys179. Furthermore, the ligand demonstrated a specific "molecular wedge" mechanism within the KRAS-SOS1 GEF interface (Δ G = -6.9 kcal/mol), suggesting direct steric disruption of oncogenic signaling. Clinical transcriptomic validation across 183 TCGA-PAAD (figure 22) patient samples confirmed high-quality expression (98.6% detection) of the target gene signatures. Kinetic modeling under tumor-specific hypoglycemic conditions (<1 mM glucose) demonstrated a 4.2-fold competitive growth advantage for SB over PDAC cells, validating the metabolic competition mechanism. Coarse-grained molecular dynamics simulations corroborated the structural rigidity of these ligand-receptor assemblies. Docking analysis confirmed high affinities for AKT1 (-7.2 kcal/mol), TLR2 (-7.6 kcal/mol), and TLR4 (-6.9 kcal/mol), pending validation at canonical glycan sites. Additionally, the ligand demonstrated specific interference with the SOS1-KRAS interaction interface (validated across wild-type and mutant structures, -6.9 to -6.2 kcal/mol), suggesting a robust direct blockade of oncogenic signaling. Coarse-grained protein simulations indicated rigid interfaces and preserved assemblies; ligand retention was not directly assessed.
Conclusions: Network analysis and structural modeling support the biological plausibility of SB anti-tumor activity primarily through immune activation and stromal remodeling; metabolic competition is a hypothesis that will be quantitatively verified and, if needed, the vehicle will be reformulated to avoid net tumor fueling. This systems-level framework, validated by thermodynamic stability data, provides a mechanistic rationale for empirical investigation.