AUTHOR=Bian Rutao , Wang Yakuan , Li Zishuang , Xu Xuegong 

TITLE=Identification of cuproptosis-related biomarkers in dilated cardiomyopathy and potential therapeutic prediction of herbal medicines

JOURNAL=Frontiers in Molecular Biosciences

VOLUME=Volume 10 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/molecular-biosciences/articles/10.3389/fmolb.2023.1154920

DOI=10.3389/fmolb.2023.1154920

ISSN=2296-889X

ABSTRACT=Background: Dilated cardiomyopathy (DCM) is one of the significant causes of heart failure and the mechanisms of metabolic ventricular remodelling due to disturbances in energy metabolism are still poorly understood in cardiac pathology. It is crucial to unravel the biological mechanisms of cuproptosis in DCM for drug development.
Methods: The DCM dataset GSE141910 was downloaded from Gene Expression Omnibus (GEO) and identified expression signatures and immune signatures of cuproptosis genes. LASSO, RF, and SVM-RFE machine learning algorithms were used to identify signature genes and the eXtreme Gradient Boosting model was used to assess diagnostic efficacy. The WGCNA algorithm was used to identify genes specific to different clusters, and Fast Gene Set Enrichment Analysis (FGSEA) was used to capture the biological processes involved in the different clusters. Finally, herbal medicines were predicted from an online database, and binding models for the significant compounds were visualized by molecular docking.
Results: We identified dysregulated cuproptosis genes and activated immune responses between DCM and normal populations. Two signature genes (FDX1, SLC31A1) were identified and performed well in an external validation dataset (AUC = 0.846). Two molecular clusters associated with cuproptosis were further defined in DCM, and immune infiltration analysis showed significant immune heterogeneity between the different clusters. In contrast, cluster 2 showed relatively high levels of immune infiltration, while FGSEA analysis revealed that cluster 2 was mainly involved in the cellular energy metabolism, apoptosis pathway. In addition, 19 and 3 herbal species were predicted based on FDX1 and SLC31A1, and Polygonum cuspidatum Sieb. Zucc. [P. reynoutria Makino; Reynoutria japonica Houtt.] containing 11 natural compounds were predicted simultaneously. Based on the molecular docking model, the natural compounds rutin with FDX1 (-9.3 kcal/mol) and polydatin with SLC31A1 (-5.5 kcal/mol) have high stability.
Conclusion: Our study systematically illustrates the complex relationship between cuproptosis and the pathological features of DCM and identifies two signature genes and two natural compounds. This may enhance our diagnosis of the disease and facilitate the development of clinical treatment strategies for DCM.