Identification of the Hub Gene LDB3 in Stanford Type A Aortic Dissection Based on Comprehensive Bioinformatics Analysis

Figures

• 

  FIGURE 1. Flow chart of our study.

• 

  FIGURE 2. Weighted gene coexpression network analysis of the GSE153434 dataset and Venn diagram to obtain two crucial genes. (A) A soft threshold power value of 8 was used, which is the optimum option. (B) Demonstration of scale‐free network validation with a soft threshold of 8. (C) By clustering genes with strong correlations into the same module, different modules were generated. Different modules are displayed in different colors. (D) Network heatmap showing branching of overall genes associated with modules in a hierarchical clustering dendrogram. (E) Analysis of the correlation between each module and TAAD. (F) The black module was significantly positively correlated with TAAD (correlation coefficient = 0.94, p < 0.001). (G) Venn diagram showing the intersection of DEGs obtained after RRA analysis, the WGCNA hub genes, and the TAAD‐related genes in GeneCards, finally yielding TIMP1 and LDB3 as the crucial genes.

• 

  FIGURE 3. Single‐cell sequencing analysis in the GSE213740 dataset. (A) Cell sample quality was assured by the analysis of three metrics: RNA count, gene count, and mitochondrial gene percentage. (B) The 2000 most highly variable genes are shown in red, with the top 10 highly variable genes marked. (C) Dimensionality reduction and clustering analysis were performed on the cells in the dataset, and the clustering of the TAAD and control groups is displayed in the ‘tSNE’ diagram. (D) Annotation of the clustered cells, including T cells, smooth muscle cells, fibroblasts, macrophages, mesenchymal cells, endothelial cells, monocytes, plasmocytes, giant cells, and B cells. (E) Expression of TIMP1 and LDB3 in cell clusters.

• 

  FIGURE 4. Pseudochronological analysis and pathway enrichment analysis of smooth muscle cells in the GSE213740 dataset. (A) Smooth muscle cells in the dataset were extracted and reclustered into nine clusters. (B) Expression analysis of LDB3 in smooth muscle cells from the GSE213740 dataset. (C) Pseudotime analysis of smooth muscle cells. The three images in the illustration show the three different differentiation states based on the aggregation of nine cell clusters and the differences in the time series of cell differentiation. (D) GO enrichment analysis of significant DEGs between cell clusters 1 and 5 (with considerable LDB3 expression) and the remaining seven cell clusters (with insignificant LDB3 expression). (E) KEGG enrichment analysis based on significantly differentially expressed genes between cell clusters 1 and 5 and the remaining seven cell clusters.

• 

  FIGURE 5. LDB3 SNPs analysis in the UK Biobank. Univariate (19 SNPs) and multivariate (rs34346901 and rs117443987) associations between SNPs and aortic aneurysm and dissection. Six SNPs were significantly related, and the relative ratios of each genotype for aortic aneurysm and dissection are shown in the pie charts.

• 

  FIGURE 6. Validation of LDB3 in human samples and upregulation of LDB3 by Ang II stimulation in HA‐VSMC. (A) Histology staining of TAAD and normal tissue. HE, Masson's trichrome staining, and EVG staining were used to reveal the differences in cellularity, collagen fibers, and elastic fibers in ascending aorta sections of TAAD and normal tissues. (B) Tissue collagen and elastin quantification (n = 10). (C) Immunofluorescence staining showed that the expression of LDB3 was significantly decreased in TAAD vascular smooth muscle cells. (D, E) Long and short isoforms of LDB3 were expressed in TAAD aortic wall samples, and Western blot verification revealed that both long and short LDB3 expressions were significantly downregulated in TAAD compared to normal samples (n = 10). (F, G) Ang II induced HA‐VSMC significant bands of LDB3 and quantification analysis. (H, I) Immunofluorescence staining of LDB3 and phalloidin in HA‐VSMC after Ang II treatment and quantification analysis (n = 4). *p < 0.05, **p < 0.01, ****p < 0.0001 in comparisons of two groups as indicated or compared with the corresponding control.

• 

  FIGURE 7. Validation of LDB3 expression in Ang II‐induced aortic dissection aneurysm models. (A) Representative photographs showing macroscopic features of dissection aneurysms induced by Ang II. (B, C) Representative ultrasound images of aortic dimension in two groups and quantification analysis (n = 5). (D, E) HE, Masson's trichrome staining, and EVG staining and elastin degradation score analysis (n = 5). (F, G) Immunofluorescence staining of LDB3 in Ang II‐induced aortic dissection aneurysm and its quantification analysis (n = 5). *p < 0.05, **p < 0.01, ****p < 0.0001 in comparisons of two groups as indicated or compared to the corresponding control.