Abstract

Differential expression analysis identified 3242 DEGs in PC. • Multi-database integration, PPI, and WGCNA analyses pinpointed CCNB1 and USF1 as key targets. • Molecular docking verified stable binding of apigenin to CCNB1 (-8.3 kcal/mol) and USF1 (-6.4 kcal/mol). • Normal mode analysis confirmed the dynamic stability of apigenin-CCNB1/USF1 complexes. Background Pancreatic cancer (PC) is characterized by high malignancy and limited therapeutic options. Apigenin, a natural flavonoid compound, has potential anti-tumor activity, yet its molecular targets and anti-PC mechanisms remain unclear. Methods Differentially expressed genes (DEGs) in PC were screened from the GSE43795 dataset. PC-associated and apigenin-targeted genes were retrieved from GeneCards and SwissTargetPrediction, and candidate genes were identified by taking their intersection. The protein-protein interaction (PPI) network was constructed using STRING and Cytoscape to screen core genes, and Gene Ontology (GO)/Kyoto Encyclopedia of Genes and Genomes (KEGG) functional enrichment analyses were conducted with DAVID. Key genes were further screened by weighted gene co-expression network analysis (WGCNA), and molecular docking and normal mode analysis (NMA) were conducted using Autodock Vina and iMODS. In vitro validation was performed in PC cells using qRT-PCR, Western blot, CCK-8, and flow cytometry. Results Multi-database integration yielded 30 overlapping genes as potential apigenin targets in PC. PPI and WGCNA analyses identified Cyclin B1 (CCNB1) and upstream transcription factor 1 (USF1) as core targets. Molecular docking revealed stable binding of apigenin to CCNB1 (-8.3 kcal/mol) and USF1 (-6.4 kcal/mol), with NMA confirming complex stability. In vitro , CCNB1 and USF1 were significantly upregulated in PC cells. Apigenin downregulated both proteins, suppressed cell viability, induced apoptosis, and triggered G2/M phase arrest. Rescue experiments confirmed that overexpression of CCNB1 or USF1 si