Identification of pathological-related and diagnostic potential circular RNAs in Stanford type A aortic dissection

Figures

• 

  Figure 1. Transcriptome signatures of TAAD aorta were highly related to elevated inflammation, loss of mesenchyme, and ECM reconstruction. (A) Volcano plots of DEGs. The blue and red parts indicate >2-fold decreased and increased expression of the dysregulated mRNAs in TAAD tissues, respectively (P < 0.05). (B) Heatmap represents TAAD DEGs. (C) DEGs enrichment map of enriched terms into a network with edges connecting overlapping gene sets. (D–F) GSEA of HALLMARK (D), KEGG (E), and REACTOME (F) gene sets from the Molecular Signatures Database of the Broad Institute, showing the most significantly enriched gene sets in TAAD and normal control aortic tissues and their normalized enrichment scores (NES).

• 

  Figure 2. Profiles of circRNAs of patients with TAAD and healthy controls. (A) Distribution of the circRNAs on human chromosomes. A circos plot diagram shows the locations of the circRNAs on human chromosomes. The outer and inner layers indicate the distribution and counts of circRNAs, respectively. Red and blue represent the patients and normal controls, respectively. (B) Venn diagram analysis presents the number of unique and common circRNAs found in patients with TAAD and normal controls. (C) Volcano plot displays dysregulated circRNAs between patients with TAAD and normal controls. The horizontal black dashed line represents a p-value of 0.05. The gray points represent circRNAs with no significant difference between TAAD and normal individuals. The blue and red points represent significantly downregulated and upregulated circRNAs in patients with TAAD, respectively. (D) Hierarchical clustering of significant DE circRNAs in patients with TAAD and normal controls. The host genes of DE circRNAs were annotated on the right side. (E, F) The bar plots of circRNA host genes GO annotation (E) and KEGG (F) enrichment results.

• 

  Figure 3. CircNOX4, circUSP3, circAMN1, and circPTGR1 were the key circRNAs of TAAD. (A, B) Risk score model construction for DE circRNAs using lasso regression analysis. (A) The lasso coefficient profiles of DE circRNAs in TAAD. (B) Partial likelihood deviance was plotted against log (lambda). The vertical dotted lines indicate the lambda value with minimum error and with one standard error (SE) of the minimum. (C) ROC curve analysis of the prognostic prediction efficiency of the lasso model. (D) The boxplot of lasso model circRNAs expression. (E) ROC curve analysis of the prognostic prediction efficiency of individual circRNA in the lasso model.

• 

  Figure 4. The key circRNAs of TAAD were functional in inflammation, ECM remodeling, and VSMC dysfunction. (A) CircRNA–miRNA interaction network predicted by circBank. The higher the total score means the greater probability that the predicted miRNA is a real one. (B) Dot plot of representative GO terms of miRNA target genes. The dot size and color represent the gene number and p-value of the corresponding term, respectively. (C) Heatmap of GSEA NES based on the correlation coefficient of circRNAs and gene expression. HALLMARK, KEGG, and REACTOME gene sets were utilized. (D) Heatmap of Spearman's correlation coefficient of the key circRNAs with genes involved in positive regulation of smooth muscle contraction (GO:0045987). The red rectangles indicate the target genes of circRNA-interacted miRNA. *p < 0.05, **p < 0.01, ***p < 0.001 (Spearman's correlation).

• 

  Figure 5. The circRNAs were correlated with the cell composition of TAAD. (A–D) Different cells, including immune cells (A), stromal cells (B), HSC (C), and epithelial cells (D), of TAAD aortic tissue and healthy control. (E) Heatmap of Spearman's correlation of circRNAs expression and different cell scores. (F) Heatmap of Spearman's correlation of circRNAs expression and VSMC score. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (Student's t-test and Spearman's correlation).

• 

  Figure 6. The circRNAs as biomarkers were efficient to diagnose TAAD in peripheral blood serum. (A) The circRNAs relative expression of NC (n = 10), TAAD (n = 8), TBAD (n = 9), and IAH (n = 4) in peripheral blood serum. Student's t-test was utilized to test different disease types compared to NC, respectively. (B) ROC curve analysis of the prognostic prediction efficiency of individual circRNA in TAAD, TBAD, and IAH from NC. (C) ROC curve analysis of the prognostic prediction efficiency of combination circRNAs in TAAD, TBAD, and IAH. (D) ROC curve analysis of the prognostic prediction efficiency of each circRNAs in TBAD and IAH from TAAD. *p < 0.05, **p < 0.01 (Student's t-test).