Identification and validation of biomarkers associated with mitochondria and programmed cell death in Barrett's esophagus

Background: Studies have shown that mitochondria and programmed cell death (PCD) are closely related to Barrett's esophagus (BE), but their specific mechanisms in BE are not fully understood. Therefore, the purpose of this study was to find and confirm the biomarkers linked to PCD and mitochondria in BE. Methods: Both the datasets related to BE and the mitochondrial-related genes (MRGs) were obtained from public databases. The literature provided the PCD-related genes (PCDRGs). Firstly, the intersection genes of the differentially expressed genes (DEGs) between control and BE samples, MRGs, and PCDRGs were considered as the candidate genes. After that, biomarkers were identified through 113 machine learning algorithms and expression verification. To evaluate the diagnostic ability of the biomarkers, a nomogram was constructed. Subsequently, functional enrichment, immune infiltration, regulatory networks, drug and disease prediction, and molecular docking analyses were performed. Finally, clinical specimens were taken for immunohistochemical validation. Results: HIP1R, VPS13C, ACSL5, and MAPT were identified as biomarkers. The constructed nomogram model had a good predictive effect. Enrichment analysis showed that the ribosome was commonly enriched by the 4 biomarkers. The largest negative link with memory B cells was discovered to be HIP1R, whereas the strongest positive correlations were observed with gamma delta T cells and natural killer T (NKT) cells for ACSL5. In the regulatory network, FOXC1 co-targeted the four biomarkers. The diseases related to 4 biomarkers included prenatal exposure delayed effects and chemical and drug induced liver injury. Meanwhile, Benzo(a)pyrene (BaP) and cyclosporin co-targeted VPS13C, ACSL5 and MAPT. Additionally, molecular docking experiments showed that the biomarkers had a good ability to connect with drugs. The immunohistochemical results confirmed that the expression levels of HIP1R, VPS13C, and ACSL5 were up-regulated in the esophageal tissue of BE, while the expression level of MAPT was significantly down-regulated. Conclusion: This study identified 4 biomarkers (HIP1R, VPS13C, ACSL5 and MAPT) that provided new insights into potential BE treatment strategies.