Identification of monocyte-associated genes as predictive biomarkers of heart failure after acute myocardial infarction

Abstract

Background: Acute myocardial infarction (AMI) is a major contributor of heart failure (HF). Peripheral blood mononuclear cells (PBMCs), mainly monocytes, are the essential initiators of AMI-induced HF. The powerful biomarkers for early identification of AMI patients at risk of HF remain elusive. We aimed to identify monocyte-related critical genes as predictive biomarkers for post-AMI HF.

Methods: We performed weighted gene co-expression network analysis (WGCNA) on transcriptomics of PBMCs from AMI patients who developed HF or did not. Functional enrichment analysis of genes in significant modules was performed via Metascape. Then we obtained the single-cell RNA-sequencing data of recruited monocytes/macrophages from AMI and control mice using the Scanpy and screened 381 differentially expressed genes (DEGs) between the two groups. We validated the expression changes of the 25 genes in cardiac macrophages from AMI mice based on bulk RNA-sequencing data and PBMCs data mentioned above.

Results: In our study, the results of WGCNA showed that two modules containing 827 hub genes were most significantly associated with post-AMI HF, which mainly participated in cell migration, inflammation, immunity, and apoptosis. There were 25 common genes between DEGs and hub genes, showing close relationship with inflammation and collagen metabolism. CUX1, CTSD and ADD3 exhibited consistent changes in three independent studies. Receiver operating characteristic curve analysis showed that each of the three genes had excellent performance in recognizing post-AMI HF patients.

Conclusion: Our findings provided a set of three monocyte-related biomarkers for the early prediction of HF development after AMI as well as potential therapeutic targets of post-AMI HF.

Keywords: Acute myocardial infarction; Biomarker; Heart failure; Monocyte; Systems biology.

Fig. 1

Study flow chart. GEO Gene Expression Omnibus, AMI acute myocardial infarction, HF Heart Failure, PBMCs peripheral blood mononuclear cells, WGCNA Weighted Gene Co-expression Network Analysis, GO Gene Ontology, KEGG Kyoto Encyclopedia of Genes and Genomes, DEGs differentially expressed genes, ROC receiver operating characteristic

Fig. 2

Identification of significant modules highly correlated with post-AMI HF. a Heatmap of the correlation between module eigengenes and post-AMI HF. Each row corresponds to a module eigengenes, column to post-AMI HF or non-HF. The correlation coefficient (cor) and p value are shown in each cell. The modules with |cor|> 0.5 and p < 0.05 were considered as significantly associated with post-AMI HF. Red, positive correlation; blue, negative correlation. b Scatterplots of genes in turquoise, blue and midnightblue modules using the GS and MM measures. Genes in turquoise and blue modules had a high significance for post-AMI HF and high module membership

Fig. 3

GO enrichment and KEGG pathway analysis of genes in significant modules through Metascape. a, b Bar chart of GO biological process enrichment of genes in turquoise and blue modules. Top 20 GO terms were shown. The color of the bar represents − log₁₀ transformation of p value. Gene Ontology: GO; KEGG: Kyoto Encyclopedia of Genes and Genomes. (C-D) Dot plot of KEGG pathway analysis of genes in turquoise and blue modules. Top 20 pathways were shown. Dot size represents the percentage of genes in each pathway. The color scale reflects represents p value

Fig. 4

Identifying recruited monocytes/monocytes-derived macrophages cluster (CCR2+ monocytes/macrophages) in the infarcted cardiac tissue of mice. a UMAP visualization of cell clusters present in pooled control and post-AMI samples. CCR2+ monocytes/macrophages were identified as cluster 4. b UMAP plot depicting all cell types in control (blue) and MI samples (yellow). c Dot plot of marker genes for each macrophage cluster. Dot size represents the percentage of cells expressing the marker gene in each cell cluster. The color scale reflects the gene expression level from low to high. d Heatmap of top 20 differently expressed genes (10 upregulated and 10 downregulated genes) in CCR2+ monocytes/macrophages between AMI and control mice. The color scale indicates the gene expression level (blue: low; red: high). UMAP: Uniform Manifold Approximation and Projection

Fig. 5

GO biological process enrichment of common genes between DEGs and hub genes. a Overlap between DEGs and hub genes of turquoise and blue modules. A total of 25 common genes were identified. b The GO biological process enrichment of 25 common genes through Metascape. DEGs differentially expressed genes, GO Gene Ontology

Fig. 6

Validation of common genes expression in cardiac macrophages from AMI mice and PBMC from AMI patients. a The gene expression changes in cardiac macrophages over the AMI time course, before (day 0) and on days 1, 3 and 7 post AMI, detected by RNA-seq. Seven out of 25 common genes exhibited consistent changes with that in the recruited monocytes/macrophages from infarcted myocardium were shown. *p < 0.05, **p < 0.01 compared with group AMI-d0. b The expression levels of 7 common genes in PBMC from AMI patients at admission measured by microarray. *p < 0.05; **p < 0.01 compared with group post-AMI non-HF; n.s.: no significance. post-AMI non-HF group: n = 8, post-AMI HF group: n = 9. Data are presented as mean ± SEM

Fig. 7

Receiver operator characteristic (ROC) curves analysis of potential biomarkers. The areas under the ROC curves (AUCs) were given for single gene (CUX1, CTSD and ADD3) and combinations (CTSD/CUX1, CTSD/ADD3) to recognize AMI patients who developed HF during a 6-month followed-up. post-AMI non-HF group: n = 8, post-AMI HF group: n = 9. ROC curves were constructed using the log₂ transformed expression data of GSE59867