Abstract
Doxorubicin is one of the most common antitumor drugs. However, cardiotoxicity's side effect limits its clinical applicability. In the present study, Gene Expression Omnibus (GEO) datasets were applied to reanalyze differentially expressed genes (DEGs) and construct weighted correlation network analysis (WGCNA) modules of doxorubicin-induced cardiotoxicity in wild-type mice. Several other bioinformatics analyses were performed to pick out the hub gene, and then the correlation between the hub gene and immune infiltration was evaluated. In total, 120 DEGs were discovered in a mouse model of doxorubicin-induced cardiotoxicity, and PF-04217903, propranolol, azithromycin, etc. were found to be potential drugs against this pathological condition. Among all the DEGs, 14 were further screened out by WGCNA modules, of which Limd1 was upregulated and finally regarded as the hub gene after being validated in other GEO datasets. Limd1 was upregulated in the peripheral blood mononuclear cell (PBMC) of the rat model, and the area under curve (AUC) of the receiver operating characteristic curve (ROC) in diagnosing cardiotoxicity was 0.847. The GSEA and PPI networks revealed a potential immunocyte regulatory role of Limd1 in cardiotoxicity. The proportion of "dendritic cells activated" in the heart was significantly elevated, while "macrophage M1" and "monocytes" declined after in vivo doxorubicin application. Finally, Limd1 expression was significantly positively correlated with "dendritic cells activation' and negatively correlated with "monocytes" and "macrophages M1'. In summary, our results suggested that limd1 is a valuable biomarker and a potential inflammation regulator in doxorubicin-induced cardiotoxicity.