ADAR1 and ADAR2 Expression in the Thoracic Aortic Wall Correlates with Aneurysm Severity and Dissection Risk: Insights into A-to-I RNA Editing Dysregulation in Marfan Syndrome-Derived vSMCs

Figures

• 

  Figure 1. ADAR1 and ADAR2 expression pattern in TAA samples. Representative immunohistochemical images of ADAR1 ((A), upper panel) and ADAR2 ((B), upper panel) staining in the intimal and medial layers of ascending aortic wall sections from patients with BAV-associated TAA, TAV-associated TAA, MFS-associated TAA, compared to a healthy control aorta from an organ donor. Brown immunoperoxidase staining marks ADAR-positive cells; nuclei are counterstained with hematoxylin, scale bars 200 μm. Quantification of ADAR1 ((A), lower panel) protein expression in ECs and vSMCs, scored on a semi-quantitative IHC intensity scale (0–3) and ADAR2 ((B), lower panel) in ECs and vSMCs, expressed as percentage of ADAR2-positive cells per field. Data are presented as mean ± SEM. Statistical comparisons were performed by the Kruskal–Wallis test with Dunn’s post-hoc correction. * p < 0.05; ** p < 0.01; *** p < 0.001. BAV, n = 21; TAV, n = 19; MFS, n = 18; Healthy control, n = 1.

• 

  Figure 2. Correlations between ADAR1 and ADAR2 protein expression and aortic clinical parameters across TAA patient subgroups. Each panel shows Spearman rho coefficients for one protein and cell compartment, (A) ADAR1 or (B) ADAR2 in ECs or vSMCs across four patient subgroups: BAV (n = 21), TAV (n = 19), MFS-TAV (n = 16), and MFS TOT (all MFS patients, n = 18). Color scale ranges from blue (negative correlation) to red (positive correlation). Asterisks (*) indicate statistically significant correlations (p < 0.05, Spearman rank-sum test with listwise deletion). The MFS-BAV (n = 2) subgroup was excluded due to insufficient sample size. Asc. aorta, ascending aorta diameter; Bulb diam., aortic bulb diameter; STJ diam., sinotubular junction diameter; EuroSCORE II, European System for Cardiac Operative Risk Evaluation II.

• 

  Figure 3. Transcriptomic analysis of MFS-derived vSMCs. (A) Volcano plot of differential gene expression between MFS and control vSMCs. The x-axis shows the apeglm-shrunken log2 fold change (MFS vs. Control) and the y-axis shows −log10(adjusted p-value). Red dots indicate significantly upregulated genes (padj < 0.05, log2FC > 0; n = 8), blue dots indicate significantly downregulated genes (padj < 0.05, log2FC < 0; n = 10), and purple dots indicate genes with padj < 0.05 but log2FC shrunk to ~0 by apeglm (n = 12). Grey dots represent non-significant genes. The dashed horizontal line marks the significance threshold (padj = 0.05). Gene labels are shown for all significantly up- and downregulated genes. Gene Set Enrichment Analysis (GSEA) dot plots showing significantly enriched gene sets in MFS vs. control vSMCs in terms of (B) MSigDB Hallmark gene sets (padj < 0.05; n = 27) and (C) Gene Ontology Biological Process (GO:BP) terms (padj < 0.001; n = 19). The x-axis represents the Normalized Enrichment Score (NES); positive NES indicates pathway enrichment in MFS, negative NES indicates enrichment in controls. Dot color reflects statistical significance (padj), with darker red indicating…

• 

  Figure 4. A-to-I RNA editing in MFS-derived vSMCs. (A) Left panel, distribution of significant A-to-I RNA editing sites by genomic region and direction of change in MFS syndrome-derived vSMCs. Significant editing sites (p < 0.05, Wilcoxon rank-sum test) are grouped by direction of change relative to controls: hypoedited (decreased editing in MFS, n = 73) and hyperedited (increased editing in MFS, n = 4). Bar segments indicate the number and proportion of sites in each genomic region: 3′UTR (blue), ncRNA intronic (purple), ncRNA exonic (orange), intronic (green), and exonic (red). Percentages refer to the proportion within each group. Middle panel, heatmap showing the median editing level (%) at 77 significant differentially edited sites (p < 0.05, |ΔMedian| ≥ 10%) in control (CTR) and MFS syndrome (MFS) vSMCs. Colored squares on the right indicate the genomic region of each site (blue: 3′UTR; purple: ncRNA intronic; green: intronic; orange: ncRNA exonic; red: exonic). Right panel, horizontal bar chart showing the ΔMedian editing level (MFS − CTR, %) for each site. Purple bars indicate sites with increased editing in MFS (hyperediting); teal bars indicate sites with decreased editing in MFS…

• 

  Figure 5. Integrated analysis of A-to-I RNA editing, miRNA targetome remodeling, and gene expression changes in MFS-derived vSMCs. (A) Number of miRNA binding sites gained (red) or lost (blue) as a consequence of A-to-I editing at the 47 significant 3′UTR editing sites across 36 target genes. miRNA target site prediction was performed by comparing the unedited (A) and edited (G) sequence variants within a ±50 nt genomic window using miRanda. Genes are ordered by net miRNA change (gained minus lost), from highest net gain to highest net loss. (B) Integrated heatmap displaying three layers of regulation for the 36 target genes, ordered by the magnitude of the editing change (Δ Editing, MFS−CTR, %) from most hypoedited (left) to most hyperedited (right). Top row: Δ Editing (%) between MFS and CTR vSMCs; negative values indicate hypoediting in MFS. Middle row: net miRNA site change (sites gained minus sites lost in the edited sequence); positive values indicate a net gain of miRNA binding sites upon editing. Bottom row: non-shrunken log2 fold change (MFS vs. CTR) from DESeq2 analysis of vSMC RNA-seq data; non-shrunken estimates were used to preserve directional information for genes not…