Molecular mechanisms of suxiao jiuxin pills in ameliorating post-acute myocardial infarction inflammatory response: a combined network pharmacology, Mendelian randomization, and experimental validation study

Figure 1. Identification of differentially expressed genes and construction of chuanXiong-bingPian-active components network. (A) Volcano plot of differentially expressed genes between AMI and control samples in training cohort, showing the distribution of up-regulated and down-regulated genes; (B) Heatmap of differentially expressed genes, showing the expression patterns of 689 DEGs in AMI and control samples; (C) ChuanXiong-BingPian-active components interaction network with 106 nodes and 106 edges, showing 91 compounds for ChuanXiong and 15 compounds for BingPian, with Borneol and Camphor as common active ingredients.

Figure 2. Identification and enrichment analysis of differentially expressed target genes. (A) Venn diagram showing the overlap of 689 DEGs and 969 predicted target genes, resulting in 44 differentially expressed target genes (DE-TGs); (B) Gene Ontology (GO) enrichment analysis of DE-TGs, including 1,253 GO entries; (C) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of DE-TGs, identifying 200 KEGG pathways.

Figure 3. Mendelian randomization analysis identifies genes causally associated with AMI. (A,B) Scatter plots of NAMPT and FOS showing positive significant correlation with AMI, with positive slopes indicating they are risk factors for AMI; (C,D) Forest plots showing significant effects of IVW model for NAMPT and FOS (effect size > 0); (E,F) Funnel plots showing that MR analysis adheres to Mendel's second law of random grouping; (G,H) Leave-one-out sensitivity analysis showing that after gradually eliminating each SNP, the effect of remaining SNPs on outcome variables did not significantly change, indicating the reliability of MR analysis.

Figure 4. Identification of key target genes by machine learning algorithms. (A) Support vector machine-recursive feature elimination (SVM-RFE) algorithm screened out NAMPT, ALDH1A1, FOS, and METRNL; (B,C) Least absolute shrinkage and selection operator (LASSO) algorithm identified NAMPT, SCD5, ABCA1, ALDH1A1, FOS, and METRNL; (D) Overlapping results of two machine learning algorithms revealed four candidate key target genes: NAMPT, ALDH1A1, FOS, and METRNL; (E) Expression levels of four candidate key genes showing significantly higher expression in AMI samples in training cohort. ****p < 0.0001; (F) No significant difference between AMI and control samples in validation cohort for ALDH1A1 and METRNL, while NAMPT and FOS showed significantly elevated expression.*p < 0.05, and ns indicates no significance; (G) Chromosomal distribution of NAMPT and FOS, located on chromosome 7 and 14, respectively.

Figure 5. Gene Set enrichment analysis (GSEA) of key target genes. (A,B) GSEA analysis of NAMPT and FOS.

Figure 6. Immune infiltration analysis. (A) Stacked bar chart displaying the proportion of 22 immune cells in 48 AMI and 48 control samples; (B) Comparative analysis of differential immune cells showing significantly higher proportions of activated Dendritic cells, Eosinophils, activated Mast cells, Monocytes, activated NK cells, resting NK cells, Neutrophils, and T follicular helper cells in AMI samples (P < 0.05), while resting memory CD4T cells and gamma delta T cells showed decreased proportions (P < 0.0001); (C) Correlation analysis among different immune cells showing strongest positive correlation between activated Mast cells and Neutrophils (P < 0.01, cor = 0.57), and strongest negative correlation between gamma delta T cells and Neutrophils (P < 0.01, cor = −0.63); (D) Correlation analysis between differential immune cells and key target genes showing strongest positive correlation between NAMPT and Neutrophils (P < 0.0001, cor = 0.65), and strongest negative correlation between NAMPT and resting memory CD4T cells (P < 0.0001, cor = −0.57). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, and ns indicates no significance.

Figure 7. Disease association, transcription factor regulatory network and molecular docking analysis. (A) Inference scores between NAMPT and FOS and AMI were 85.23 and 277.31, respectively, indicating a strong correlation between these key target genes and AMI, particularly FOS; (B) Transcription factor (TF)-mRNA regulatory network showing 14 TFs regulating key target genes, with NAMPT regulated by six TFs and FOS by 11; (C,D) Molecular docking analysis of NAMPT with Oleic Acid showing binding energy of −5.9 kcal/mol, with LYS-189 residue exhibiting hydrogen bonding interactions with Oleic Acid; (E,F) Molecular docking analysis of FOS with Pentadecanol showing binding energy of −5.4 kcal/mol, with DA B-10 and DG A-12 residues showing hydrogen bonding interactions with Pentadecanol.

Figure 8. Effects of suxiao jiuxin pill on FOS and NAMPT expression. (A) Representative immunofluorescence staining images of FOS protein in myocardial tissues from four groups of rats. DAPI staining shows nuclei (blue), FOS antibody staining shows FOS protein expression (red), and Merge shows merged images. Compared with Sham group, FOS protein expression was significantly increased in myocardial tissue of MI group; compared with MI group, FOS expression was downregulated in MI + SJP group; (B) Representative immunofluorescence staining images of NAMPT protein in myocardial tissues from four groups of rats. DAPI staining shows nuclei (blue), NAMPT antibody staining shows NAMPT protein expression (red), and Merge shows merged images. Compared with the Sham group, NAMPT protein expression in myocardial tissue was significantly decreased in the MI group; compared with the MI group, NAMPT expression was upregulated in the MI + SJP group. *p < 0.05.

Figure 9. Suxiao jiuxin pill improves infarct area and cardiac function in MI. (A,B) Masson staining showing infarct area of cardiac tissue, with MI + SJP group showing significant reduction in infarct area compared to MI group alone; (C–E) Hemodynamic analysis showing decreased left ventricular EF and FS in MI group rats after surgery compared to Sham group; Suxiao Jiuxin Pill improved these indicators compared to MI group alone (P < 0.01); (F–J) ELISA assay showing that IL-1, NE, His and cTNI were significantly reduced in MI + SJP group compared to MI group. *p < 0.05.