Unraveling the Predictive Value of HGF, RECK, CYP1B1, and ZEB2 for Cancer Stemness and Immune Microenvironment in Pancreatic Adenocarcinoma.

Abstract

Introduction: The progression of pancreatic adenocarcinoma (PAAD) is closely linked to autophagy and microRNA (miRNA) regulation. Therefore, constructing miRNA regulatory networks based on PAAD autophagy-related genes is essential for improving targeted therapy.

Methods: In this study, we downloaded PAAD-related data from The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx), PAAD-related miRNA data from Gene Expression Omnibus (GEO), and mined autophagy from existing literature reports for autophagy-related genes. Autophagy signature scores were assessed employing single-sample gene set enrichment analysis (ssGSEA) and screened for autophagy-associated candidate genes via weighted gene co-expression network analysis (WGCNA) and differential expression analysis. Bioinformatics tools such as CIBERSORT and ESTIMATE were employed to assess the level of immune infiltration. The Encori database and Cytoscape 3.8.0 tools were used to construct the miRNA regulatory network. The PubChem website and AutoDockTools were used for targeted drug prediction and molecular docking of PAAD autophagy-related genes. Cell counting kit-8 (CCK-8), scratch healing assay, transwell assay, etc. were used to investigate the regulatory effect of biomarkers on the PAAD cells.

Results: WGCNA combined differential expression analysis obtained candidate genes related to PAAD autophagy, which were primarily implicated in the pathways of phagocytosis and miRNAs in cancer. Among them, four characteristic genes (HGF, RECK, CYP1B1, and ZEB2) contained in the miRNAs in cancer pathway will be included in the subsequent analyses as biomarkers affecting PAAD autophagy. The expressions of these biomarkers were positively linked to all immune-relevant indicators (Stromal, Immune, and ESTIMATE scores) of PAAD. The regulatory network of miRNAs and four biomarkers in PAAD was constructed, and it was elucidated that hsa-miR-222-3p and hsa-miR-377-3p had targeting relationship to RECK and CYP1B1, respectively. Moreover, results showed that HGF, RECK, and ZEB2 were significantly positively correlated with cancer stem cell (CSC) scores (R > 0.3), and the high expressions of HGF, RECK, and ZEB2 genes all led to the significant enrichment of EPITHELIAL_MESENCHYMAL_TRANSITION and TGF_BETA_SIGNALING. Molecular docking revealed that CYP1B1, HGF, and RECK all stably bound resveratrol. ZEB2 is significantly associated with proliferation, migration, and invasion in the PAAD cells.

Conclusion: The present study elucidated genes associated with autophagy features in PAAD by bioinformatics and constructed corresponding miRNA networks and molecular docking models, and predicted potential target drugs for PAAD, which will guide the development of prognostic therapeutic strategies for PAAD.