Identification of the potential biomarkers associated with circadian rhythms in heart failure

Abstract

Background Heart failure (HF) is a syndrome with multiple clinical symptoms resulting from damage to the heart’s structure and/or function with various pathogenic factors, which has developed as one of the most severe threats to human health. Approximately 13% of genes and about 8% of proteins contained in the heart are rhythmic, which could lead to HF if disrupted. Herein, we aimed to identify the circadian rhythms-related hub genes as potential biomarkers contributing to the identification and treatment of HF. Methods Expression data of ischemic and dilated cardiomyopathy samples with or without HF were collected from the GEO database. First, genes with differential expression in HF and healthy samples were identified, named as differentially expressed genes (DEGs), which were then intersected with circadian rhythms-related genes to identify circadian rhythms-related DEGs. A protein-protein interaction (PPI) network was established to screen hub genes. The performance of the hub genes to identify HF among healthy controls was assessed by referring to the receiver operating characteristic (ROC) curve. Additionally, quantitative real-time polymerase chain reaction (RT-PCR) was run to further validate the hub genes depending on clinical human peripheral blood samples. Results A total of 10,163 DEGs were determined, composed of 4,615 up-regulated genes and 5,548 down-regulated genes in HF patients in comparison to healthy controls. By overlapping the circadian rhythms-related genes in the Circadian Gene DataBase (CGDB), 723 circadian rhythms-related DEGs were obtained, mainly enriched in regulating lipid metabolic process, circadian rhythm and AMPK signaling pathway. Eight hub genes were screened out through the PPI network. The ROC curve indicated the high accuracy of five hub genes with AUC > 0.7, which also showed high accuracy validated by the external validation dataset. Furthermore, according to the results of quantitative RT-PCR, the HF group showed significantly increased relative mRNA expression of CRY2 and BHLHE41 while the decreased ARNTL and NPAS2 in comparison to controls, indicating the four hub genes as potential biomarkers of HF. Conclusion Our study validated that ARNTL, CRY2, BHLHE41 and NPAS2 could serve as potential biomarkers of circadian rhythm in HF. These results may provide a reference for employing novel markers or targets for the diagnosis and treatment of HF.