Molecular Fingerprint for Terminal Abdominal Aortic Aneurysm Disease

Figure 1. Study design. A, Two array analyses were performed in the identification cohort: ruptured (rAAA) vs stable and intermediate (≤55 mm) vs large (>70 mm) size AAA. The Venn diagram above depicts the overlap among the different groups in the identification cohort. The Venn diagrams underneath show the overlap among the identified differentially expressed transcripts of the 2 analyses. An independent validation cohort was used to validate the overlapping candidate genes from the array analyses with quantitative real‐time PCR. B, Volcano plot of differentially expressed transcripts for eAAA vs rAAA array analysis. C, Volcano plot of differentially expressed transcripts for intermediate vs large AAA array analysis (colored dots, |log2 FC| ≥1.5, adjusted P<0.05). AAA indicates abdominal aortic aneurysm; eAAA, elective AAA; iAAA, intermediate AAA; lAAA, large AAA; qRT‐PCR, quantitative real‐time polymerase chain reaction; rAAA, ruptured AAA.

Figure 2. Gene expression levels of 5 genes are associated with AAA rupture and diameter. Normalized gene expression levels (validation cohort) of differentially expressed genes in rAAA, which also showed significant correlation to AAA diameter (terminal progression–associated genes). Gene expression levels were normalized to β‐actin. Data are demonstrated as mean±SEM, and Mann‐Whitney U test was performed to identify differences in expression levels of selected genes between eAAA and rAAA. Spearman rank correlation coefficient (r) was determined to detect associations for gene expression levels and AAA diameter. AAA indicates abdominal aortic aneurysm; eAAA, elective AAA; rAAA, ruptured AAA.

Figure 3. Results of the validation for candidate genes without correlation to AAA diameter. A, ADAMTS9, GFPT2, and STC1 were validated for higher expression levels in rAAA. B, Decreased gene expression for rAAA was also demonstrated for CCL4L1 and GAL3ST4 in the validation cohort, but their gene expression levels showed no significant correlation with diameter (rupture only–associated genes). Differential gene expression levels for SLC39A14 and RAB7B could not be reproduced. Gene expression levels were normalized to β‐actin. Data are illustrated as mean±SEM, and Mann‐Whitney U test was performed to identify differences in expression levels of selected genes between eAAA and rAAA. Spearman rank correlation coefficient (r) was determined to detect associations for gene expression levels and AAA diameter. AAA indicates abdominal aortic aneurysm; eAAA, elective AAA; rAAA, ruptured AAA.

Figure 4. Gene expression signature is unique for AAA disease and originates from mesenchymal cells. A, Identified upregulated candidate genes exclusively associated with AAA, and gene expression is distinct from other dilative vascular disorders (eg, popliteal aneurysm, aortic dissection) as well as atherosclerotic changes. Downregulated genes were expressed at low levels in AAA and showed higher expression levels in other vascular pathologies. B, For almost all upregulated genes (except HILPDA) higher expression levels were detected in mesenchymal cell types (fibroblasts, adipocytes, and smooth muscle cells) than in inflammatory cells. Downregulated candidate genes showed highest expression levels in peripheral blood mononuclear cells (PBMC). Quadruplicate measurements per condition. Data are given as mean±SEM. Kruskal‐Wallis test was performed for multiple comparison analyses. Adjusted P<0.05 was considered significant (*P<0.05; **P<0.01; ***P<0.001). AAA indicates abdominal aortic aneurysm; CFA, common femoral artery; diss. TAA, dissected thoracic aortic aneurysm; EC, endothelial cells; ICA, internal carotid artery; popl., popliteal; SMC, smooth muscle cells; TAA, thoracic aortic…

Figure 5. Results of the immunohistochemistry. Staining for protein products of our candidate genes confirmed presence in the aneurysmatic wall. ANGPTL4, HILPDA, SRPX2, and LOX are localized in pericytes of microvessels in the medial‐adventitial border zone. Staining for SRPX2 is diffuse in the media and also observed in adipocytes of the adventitia. Demonstrated are representative sections from a 55‐mm stable abdominal aortic aneurysm.

Figure 6. Increased neovascularization in rAAA. A, Results of the determination of microvessel density between (B) stable (eAAA) and (C) ruptured AAA (rAAA) samples (n=5/group); *P<0.05. Arrows indicate microvessels in the media‐adventitia border zone of aneurysmatic walls. Data are demonstrated as mean±SEM, and Mann‐Whitney U test was performed to identify differences of the percentage of vasa vasorum. AAA indicates abdominal aortic aneurysm; eAAA, elective AAA; rAAA, ruptured AAA.

Figure 7. HIF‐1α strongly associates with differentially expressed genes in rAAA. Results from the Ingenuity pathway analysis (stable vs ruptured AAAs) for the activated upstream regulator HIF‐1α (activation z‐score 4.329 and an overlap P value of 2.74×10−22) showing the significantly up‐ (red) and downregulated (green) genes and their localization in the cell. Figure ©2000‐2016 by Qiagen (Hilden, Germany). Reproduced with permission.

Figure 8. HIF‐1α expression is detected in cells undergoing EndoMT. Results of HIF‐1α and CD31 staining (CD31 [blue], HIF‐1α [red] double staining): (A) overview (scale bar=500 μm). HIF‐1α staining is particularly prominent in the well‐vascularized areas (B) but minimally expressed in the avascular areas (C). D, Perivascular CD‐31/HIF‐1α double‐positive mononuclear cells (arrows) suggest HIF‐1α expression in cells undergoing EndoMT. This finding is supported by a CD31 (blue) and vimentin (red) double staining: (E) overview (scale bar=500 μm). F, Perivascular (vasa vasorum) CD31/vimentin double‐positive mononuclear cells (arrows) are found in the media‐adventitia border (scale bar=50 μm). G, Again, minimal expression is observed in the avascular areas of the intima‐media zone (scale bar=50 μm). EndoMT indicates endothelial‐to‐mesenchymal transition.