Article | Open

Differentially expressed genes and canonical pathways in the ascending thoracic aortic aneurysm – The Tampere Vascular Study

  • Scientific Reports 7, Article number: 12127 (2017)
  • doi:10.1038/s41598-017-12421-4
  • Download Citation
Received:
Accepted:
Published online:

Abstract

Ascending thoracic aortic aneurysm (ATAA) is a multifactorial disease with a strong inflammatory component. Surgery is often required to prevent aortic rupture and dissection. We performed gene expression analysis (Illumina HumanHT-12 version 3 Expression BeadChip) for 32 samples from ATAA (26 without/6 with dissection), and 28 left internal thoracic arteries (controls) collected in Tampere Vascular study. We compared expression profiles and conducted pathway analysis using Ingenuity Pathway Analysis (IPA) to reveal differences between ATAA and a healthy artery wall. Almost 5000 genes were differentially expressed in ATAA samples compared to controls. The most downregulated gene was homeobox (HOX) A5 (fold change, FC = −25.3) and upregulated cadherin-2 (FC = 12.6). Several other HOX genes were also found downregulated (FCs between -25.3 and -1.5, FDR < 0.05). 43, mostly inflammatory, canonical pathways in ATAA were found to be significantly (p < 0.05, FDR < 0.05) differentially expressed. The results remained essentially the same when the 6 dissected ATAA samples were excluded from the analysis. We show for the first time on genome level that ATAA is an inflammatory process, revealing a more detailed molecular pathway level pathogenesis. We propose HOX genes as potentially important players in maintaining aortic integrity, altered expression of which might be important in the pathobiology of ATAA.

Introduction

Ascending thoracic aortic aneurysms (ATAA) are a group of degenerative, often atherosclerotic, inflammation diseases in which the integrity of the aortic wall weakens due to changes in the extracellular matrix (ECM) and a part of the vessel starts to dilate and bulge out forming an aneurysm.

Degenerative aneurysms of the ascending aorta are the most common type of thoracic aortic aneurysms, with incidence of 6–10 per 100 0001. The most important risk factor for ATAA alongside with genetic disorders is hypertension2. Other important risk factors of ATAA includes smoking, male gender, over 60 years of age1,3 and atherosclerosis. ATAA is caused by weakening changes occurring in the aortic wall, especially in the elastic tunica intima and media4, in which the amount of elastin decreases and elastin becomes fragmented5 thus making the wall more susceptible to bulging out i.e. forming an aneurysm. This type of change predisposing the aortic wall to bulge out is often associated with atherosclerosis6. In addition, changes occur in regard to collagen, especially the type IV collagen related to the vasa vasorum4. Multiple gene expression changes of important inflammatory and ECM-remodeling machinery genes have been demonstrated between healthy and diseased ascending aorta, for example in the expression of ECM remodeling matrix metalloproteases 2 and 9 (MMP-2 and −9)7,8 and a disintegrin and metalloproteinases (ADAMs) 8 and 159. Also, the expression of tissue inhibitors (TIMPs) of these metalloproteases has recently come in to the focus of research in aortic diseases, especially in ATAA research5.

The incidence of aortic dissection is less common (0.5–6 per 100 0001,10,11) and of these only a minority involves the ascending aorta. Ascending aortic dissection is usually caused by a tearing of ATAA as is the case for the patients in our study. ATAA and aortic dissection can also be caused by rare genetic disorders (e.g., Marfan, Ehlers-Danlos (ED) or Loeys-Dietz (LD), or Turner syndrome), a trauma, an inflammatory disease (autoimmune or microbial based) or the cause may remain unclear1.

To our best knowledge this work is the first whole genome wide expression profiling of ATAA. It is important to differentiate the ascending and descending aortic tissues since aorta is not an embryologically uniform structure12, as ascending thoracic and abdominal aortas have different developmental origins12. In previous works, whole genome profiling has revealed an important role of pathways of the immune system and cell adhesion in the pathogenesis of abdominal aortic aneurysms (AAA)13,14. The multifactorial disease of ATAA is regarded as a “silent killer” due to its lack of clinical symptoms before late complications15. Several other conditions, including AAA, have been shown as predisposing factors to ATAA15. Hence it is crucial to research the processes, which are active and affect these, often lethal and requiring aortic surgery to prevent aortic rupture and dissection.

In this study we aim to describe differences in expression of individual genes and pathways between ATAAs and histologically healthy arterial wall obtained from left internal thoracic artery during cardiac bypass surgery from patients with coronary artery disease.

Materials and Methods

Tampere Vascular Study

The samples were collected from patients undergoing cardiac surgery in the Heart Center of Tampere University Hospital with informed consent9. Patient information was collected with questionnaires and from the patient database of the hospital. Of the ATAA samples (N = 32, 23 men, 9 women) six (N = 6) had progressed to dissection. The mean age of our cases was 62.6 years and there were no patients with rare genetic disorders (e.g., Marfan, Ehlers-Danlos (ED) or Loeys-Dietz (LD), or Turner syndrome) or a trauma. The control arteries (N = 28) were collected from atherosclerosis-resistant16 left internal thoracic artery (LITA) from patients undergoing coronary artery bypass surgery. LITA tissue was chosen, as it is a histologically healthy and is ethically accessible. Demographics of the study population are presented in Table 1. The Ethics Committee of Pirkanmaa Hospital District has approved of this study and the clinical investigation followed the principles of Helsinki declaration.

Table 1: Demographics of the study population.

RNA isolation and genome wide expression analysis (GWEA)

The fresh tissue samples obtained from surgery were soaked in RNALater solution (Ambion Inc., Austin, TX, USA) and RNA isolated with Trizol reagent (Invitrogen, Carlsbad, CA, USA) and the RNAEasy Kit (Qiagen, Valencia, CA, USA). The quality as well as concentration of the RNA was evaluated spectrophotometrically (BioPhotometer, Eppendorf, Wesseling-Berzdorf, Germany). Over 23,000 known and candidate genes were analyzed with Illumina HumanHT-12 version 3 Expression BeadChip (Illumina Inc.), following the instructions provided by the manufacturer. (Illumina, San Diego, CA, USA). In brief, 200 ng aliquots of total RNA from each sample were amplified to cDNA using the Ambion’s Illumina RNA Amplification kit according to the instructions (Ambion, Inc., Austin, TX, USA). Samples of cRNA (1500 ng) were hybridized to Illumina’s Expression BeadChip arrays (Illumina). Hybridized biotinylated cRNA was detected with 1 µg/ml Cyanine3-streptavidine (Amersham Biosciences, Piscataway, NJ, USA). BeadChips were scanned with the Illumina BeadArray Reader. The accuracy of this array has been previously tested and in our TVS validation study in which the results of 192 differentially expressed genes were verified by quantitative reverse transcription polymerase chain reaction (qRT-PCR)17. The correlation between expression measurements from GWE and qRT-PCR methods was good (r = 0.87, y = 0.151 + 0.586x), and the Bland-Altman plot showed that the fold changes (FCs) were in agreement with these two methods, although for highly up- or down-regulated transcripts, GWE yielding lower absolute FC values than in qRT-PCR.

Microarray data analysis and pathway analysis

After background subtraction, raw intensity data were exported using the Illumina GenomeStudio software. Raw expression data were imported into R version 3.1.1 (http://www.r-project.org/), log2 transformed and normalized by the locally estimated scatterplot smoothing normalization method implemented in the R/Bioconductor package Lumi (www.bioconductor.org). Locally estimated scatterplot smoothing (LOESS) normalization for the data was selected because it gave the best accuracy in comparison with qRT-PCR data for artery samples17. Data quality control criteria included detection of outlier arrays based on the low number of robustly expressed genes and hierarchical clustering.

In order to estimate FC between groups, we calculated differences between medians (in log2 scale) and then back transformed the log ratios to FCs. To make the interpretation easier, we replaced fold-change values that are <1 by the negative of its inverse. Statistical significance of differences in gene expression was assessed using the nonparametric Wilcoxon signed-rank test and the log-transformed data.

Using the “Core Analysis” function of QIAGEN’s Ingenuity Pathway Analysis (IPA, QIAGEN Redwood City, www.qiagen.com/ingenuity) we analyzed the gene expression data to put it in context with the potential biological functions and known pathological mechanisms. The inclusion criteria for genes selected for the analysis was a FC >  ± 1.5, a p-value < 0.05 and an FDR < 0.05 leading to 4884 genes being selected for further analysis. Gene ontology using GO term analysis with GOrilla online tool18 was also performed.

Results

Differentially expressed genes in ascending thoracic aortic aneurysms in comparison to healthy control arteries

In our ATAA vs. control sample comparison we found a total 4884 significantly differentially expressed genes of which 3115 were up- and 1769 down-regulated (FC >  ± 1.5, p-value < 0.05 and FDR < 0.05). The genes with FC >  ± 5 and FDR < 0.05 are presented in Table 2, and all differentially expressed genes (FC >  ± 1.5 and FDR < 0.05) can be found in Supplementary Table 1. When we excluded the 6 dissected ATAA samples from the analysis, the results remained essentially the same (3228 genes were up- and 1769 genes significantly down regulated).

Table 2: The most up- and down-regulated (FC >  ± 5) genes in ascending thoracic aorta aneurysm samples when compared to left internal thoracic artery.

In genome-wide analysis several homeobox (HOX) genes were amongst the most significantly downregulated in our comparison, especially HOXA5 and HOXC6. A total of 15 different HOX genes were down-regulated (FCs between −25.3 and −1.5) in the samples from ATAA, and one HOX gene (HOXD11) was upregulated with a FC of 1.8 in comparison to controls (Table 2, Supplementary Table 1). HOXA5 was the most down regulated gene in ATAA with a FC of −25.3 in comparison to the healthy control vessel, whilst the most up-regulated gene was cadherin-2 (CDH2; FC = 12.6). Many chemotactic genes were also significantly up-regulated in the ATAA samples as, these included interleukin-6 (IL6), cytokine-like 1 (CYTL1) and secretogranin 2 (SCG2), and several other chemokines that induce chemotaxis of leukocytes to the site of inflammation. Apoptosis- and lipid metabolism-related genes cell death-inducing DFFA-like effector c and a (CIDEC and CIDEA) were also significantly downregulated in our samples.

We also present the results we got when we compared dissected to non-dissected samples in Supplementary Table 2.

Ingenuity Pathway analysis (IPA) and gene ontology (GO term) analysis

In total 43 canonical pathways were significantly (FDR < 0.05) differentially expressed in ATAA samples in comparison to control arteries (Table 3). Many of the pathways are inflammation-related or involved in the alterations in the ECM and cell to cell adhesions. Hepatic Fibrosis/Hepatic Stellate Cell Activation was the pathway with the most significantly different expression in our samples when compared to the control arteries. GO term analysis yielded no statistically significant results in ATAA analysis with a false discovery rate (FDR) < 0.05 and p-value < 0.05. Fifteen GO terms were differentially expressed with a higher FDR-limit of 0.25 (Supplementary Table 3) and these included many GO terms related to leukocyte migration and tissue remodeling. We also performed analysis in which we excluded the dissected ATAA samples, but as the most significant results remained the same, we present the results of both gene expression and IPA only in the supplementary files (Supplementary Table 4). In otherwise similar pathway analysis, the results remained essentially the same when the 6 dissected ATAA samples we excluded from the analysis.

Table 3: The Ingenuity Pathway Analysis results showing the most differentially expressed pathways (FDR < 0.05) in ascending thoracic aorta aneurysm samples when compared to control vessel (LITA).

Discussion

Here we conducted a large-scale genome wide expression analysis of the molecular determinants of ATAA. Our study addresses the important question of the molecular mechanisms of differentially expressed genes in ATAA, and also using novel bioinformatic Ingenuity Pathway Analysis describing the biological processes relating the differentially expressed genes to each other as functional pathways. Much of the research on aortic aneurysms and gene expression has so far focused on aneurysms of the abdominal aorta19. We further confirm many genes previously reported in association with aortic diseases, but also reveal a number of genes previously unreported in association with diseases of the ATAA. Our study offers a novel perspective for the pathobiology ascending thoracic aortic aneurysms, especially regarding the role of HOX genes as most of the HOX genes we found differentially expressed, have not been associated with cardiovascular diseases before. Based on our results it is clear that inflammation remains a crucial part in aneurysm formation and is responsible for many of the structural changes that result in vascular wall weakening and enabling the development of an aneurysm.

Our results are well in accordance with previous results of immune cell response and differential expression of single genes, acquired previously in abdominal aortic aneurysms studies13,14,19. These results further affirm that ATAA is an inflammatory condition20. However, these similarities must be critically considered as ascending aorta and abdominal aorta are of different embryological origin. Previous studies have also demonstrated changes in calcium-signaling14, cell-death21,22 and other inflammatory pathways in abdominal aortic aneurysms, similar to those we here report in ATAA. Interestingly, many of these pathways are similar as those we have previously reported in association with atherosclerosis23. The results we present here suggest ATAAs to have a strong inflammatory component, which agrees with previous studies conducted with both ATAA20 and abdominal aorta aneurysms14. As a novel finding we report the differentially expressed HOX genes in aortic pathological changes, differential expression of which has previously been reported only in few instances in association to cardiovascular pathophysiology24,25,26.

HOX genes are a highly conserved27 subgroup of homeobox-genes, crucial in embryonic development and spatial body patterning. There is a total of 39 HOX genes28, of which 16 were significantly differentially expressed in our data, suggesting these genes have an integral role in the pathogenesis of aortic disease. The only differentially expressed HOXD gene, HOXD11, was upregulated as opposed to 15 different HOXA, -B and –C genes which were downregulated, some of which had several transcripts (namely HOXA9 and HOXC6 had two transcripts). Our results seem to be fairly well in line with previous work of Seo D. et al., who have also shown some of the homeobox genes (HOXA5, -C6, -A4, -D4 and –B2) to be dysregulated in atherosclerosis29. Nonetheless, most studies related to differential expression of HOX genes are of cancer pathogenesis30. There are only a few studies that have shown HOX genes to have a role in cardiovascular diseases29 such as abnormal vasculogenesis24,31, therefore further research in this area is imperative.

When interpreting these results, we need to keep in mind the site-specificity of HOX gene expression, meaning that HOX gene expression is different in different parts of the healthy aorta, in a collinear manner typical for HOX genes28,32,33. The down-regulation of HOXA5 in our samples is in accordance with results from in vivo and murine models presented by Arderiu et al., which show normal HOXA5 expression to reflect a more stable vascular phenotype25. Suppression of HOXA5 has also been shown to be important in angiogenesis34 thus giving merit to our results as angiogenesis, induced by leukocytes35, is active in inflamed tissues such as atherosclerotic and aneurysmatic vessels. We suggest that the downregulation of HOXC6 is related to the apoptotic activity, as this association has been previously shown by Ramachandran S. et al. in human prostate36.

Many chemotactic genes were differentially expressed in the ATAA samples, further demonstrating the inflammatory nature of aortic aneurysms37,38. Amongst the most upregulated were the genes CDH2, CYTL1 and SCG2. We consider the up-regulation of CDH2 to be merely an indicator of mesenchymal stem cells with the potential to differentiate into cardiac myocytes39. The upregulation of secretogranin 2 (SCG2) is a sign of transendothelial migration of leukocytes as shown by Kähler CM. et al.40 in the ascending aorta, and in this manner we believe it contributes to the loss of integrity in the aortic wall, making the aorta susceptible to aneurysm or even progression to dissection. CYTL1, along with other chemokines such IL6 and SPP141, attract leukocytes which in turn infiltrate the aortic wall and cause other inflammatory effects. This results in enhanced vascular permeability and loss of integrity in the aortic wall42, and thereby ensuing in aneurysm formation.

Our results show MMPs 12 and 28 to be differentially expressed with FCs of 2.6 and −2.5 respectively. Many MMPs are associated with cardiovascular diseases, including ATAAs, and particularly ECM-remodeling43,44,45 and as both MMP 12 and 28 are macrophage associated, this further links leukocyte infiltration and function to the changes in the ascending aorta21. MMP28 has been implicated in the up keeping of tissue homeostasis by Balta S. et al.46 and even though it has mostly been investigated in association with various cancers, our results suggest it is related to aortic disease as well, through the same process of dysfunction in tissue homeostasis.

Aortic aneurysms are most commonly located in the abdominal aorta, and due to this, most of the previous similar research has focused on those. The important role of pathways of the immune system and cell adhesion have repeatedly been demonstrated in similar studies of whole genome expression and pathways analysis13,14,19. Several similarities can be recognized between these previous works, many inflammatory genes we report here, such as osteopontin (SPP1) and interleukin-6 (IL6), are introduced in these previous works by Hinterseher I. et al.13,19. It is also noteworthy that the pathways described in these previous works such as Ca2+ -signaling13 and leukocyte transendothelial migration14 are also well in line with our results. Both of the aforementioned and many more similar pathways are also differentially expressed in our samples, but some have slightly different names as the pathway analyses have been conducted with different programs. These similarities in differentially expressed pathways further confirm the role of inflammation to be important in the pathobiology of aortic aneurysms.

The pathways we found differentially expressed suggest that inflammation, leukocyte accumulation and apoptosis are important events occurring in diseased ascending aorta. Pathways we found significantly differentially expressed consist mostly of inflammatory pathways with some hypoxia related signaling as well. We consider the differentially expressed hypoxia induced signaling pathways47, together with the differential expression of HOX genes26, to show that the processes necessary for the formation of neovessels and repair of the vascular wall are active in the aneurysmatic aorta48. Other pathways, that are differentially expressed, are likely to be a result of leukocyte-activity at the aortic site. Pathways such as “Integrin signaling” and “Leukocyte Extravasation Signaling” show that alterations of the arterial wall occur in order to allow leukocytes through, thus further compromising the integrity of the arterial wall49 through complex networks of cytokines and other inflammatory mediators, as has been previously studied in both mice50 and humans51. Our results regarding the p53-signaling in ascending aorta are in line with those of Leeper N. et al.52 relating this to the apoptosis of smooth muscle cells in the aortic wall20. The results we present here support the idea of inflammatory response as the most crucial characteristic of ATAA and show this disease to be an inflammatory condition20,53.

Limitations of the study

The histologically healthy non-atherosclerotic LITA was used as a control vessel. The use of LITA as the control vessel was based on the fact that ethical problems would arise from attempting to collect corresponding healthy aortic samples for gene expression studies during normal surgical operations. Due to LITAs many biologically analogous features we consider it to be the best control readily available16,54,55. We acknowledge that especially the inflammatory pathways are to be expected to differ in LITA56, as has been shown in methodologically similar study by Ferrari G. et al.56, but this also does furthermore also support the crucial role of inflammation in ATAA. However, this means we must take in to account the bias caused by LITA control when interpreting these results, especially in regards to the inflammatory pathways being overestimated. We have also used it in our previous studies in comparison of different arterial beds including descending aorta as a control vessel23. Our LITA samples have been also collected and analyzed with exactly the same method than ATAA and thus are methodologically ideal controls as in our previous TVS works23. To control the effect of ATAA dissection to gene all analyses were performed in two ways with (n = 32) and without dissection (n = 26), with essentially same results. We also emphasize careful interpretation of the results regarding HOX genes, because HOX genes are important in developmental biology and our samples and controls are of different embryological origin.

As many cardiovascular patients have several drugs in use, we must also acknowledge that the use of pharmacologically active agents may affects the gene expression in vascular wall. And as there was a significant difference in statin usage between the cases and controls, this might have an effect on our results. Nonetheless, the differences in medication or in risk factors (i.e. smoking, hypercholesterolemia etc.) are not sufficient to explain the significant differential expression of genes we have presented here and thus we consider our results to be of significance in furthering our understanding the pathobiology of ATAA.

Conclusions

Our findings expand the current understanding of ATAA pathogenesis on the molecular level. Our results show for the first time at the genome level that ATAA is an inflammatory process, and pinpoint its detailed molecular and pathway level pathogenesis. We also propose cautiously HOX genes as possibly crucial players in maintaining aortic integrity, and altered expression of these transcription factors may lead to susceptibility to pathological changes in the ATAA and even lead to a need for aortic surgery. This however might, at least partially, be due to the different embryological origin of the compared vessels. Nonetheless our result well depict important molecular pathways that are significant in ATAA pathogenesis and offer a wide range of possibilities for future research on this serious, yet relatively little studied, disease.

Additional Information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. 1.

    Erbel, R., Aboyans, V. & Boileau, C. et al. 2014 ESC guidelines on the diagnosis and treatment of aortic diseases: Document covering acute and chronic aortic diseases of the thoracic and abdominal aorta of the adult. the task force for the diagnosis and treatment of aortic diseases of the european society of cardiology (ESC). Eur Heart J. 35(41), 2873–2926 (2014).

  2. 2.

    Cozijnsen, L., Braam, R. L. & Waalewijn, R. A. et al. What is new in dilatation of the ascending aorta? review of current literature and practical advice for the cardiologist. Circulation. 123(8), 924–928 (2011).

  3. 3.

    Aggarwal, S., Qamar, A., Sharma, V. & Sharma, A. Abdominal aortic aneurysm: A comprehensive review. Exp Clin Cardiol. 16(1), 11–15 (2011).

  4. 4.

    Tsamis, A., Krawiec, J. T. & Vorp, D. A. Elastin and collagen fibre microstructure of the human aorta in ageing and disease: A review. J R Soc Interface. 10(83), 20121004 (2013).

  5. 5.

    Boyum, J., Fellinger, E. K. & Schmoker, J. D. et al. Matrix metalloproteinase activity in thoracic aortic aneurysms associated with bicuspid and tricuspid aortic valves. J Thorac Cardiovasc Surg. 127(3), 686–691 (2004).

  6. 6.

    Ramanath, V. S., Oh, J. K., Sundt, T. M. 3rd & Eagle, K. A. Acute aortic syndromes and thoracic aortic aneurysm. Mayo Clin Proc. 84(5), 465–481 (2009).

  7. 7.

    Akiyama, M., Ohtani, H., Sato, E., Nagura, H. & Tabayashi, K. Up-regulation of matrix metalloproteinase-2 and membrane-type 1-matrix metalloproteinase were coupled with that of type I procollagen in granulation tissue response after the onset of aortic dissection. Virchows Arch. 448(6), 811–821 (2006).

  8. 8.

    Luttun, A., Lutgens, E. & Manderveld, A. et al. Loss of matrix metalloproteinase-9 or matrix metalloproteinase-12 protects apolipoprotein E-deficient mice against atherosclerotic media destruction but differentially affects plaque growth. Circulation. 109(11), 1408–1414 (2004).

  9. 9.

    Levula, M., Paavonen, T. & Valo, T. et al. A disintegrin and metalloprotease −8 and −15 and susceptibility for ascending aortic dissection. Scand J Clin Lab Invest. 71(6), 515–522 (2011).

  10. 10.

    Howard, D. P., Sideso, E., Handa, A. & Rothwell, P. M. Incidence, risk factors, outcome and projected future burden of acute aortic dissection. Ann Cardiothorac Surg. 3(3), 278–284 (2014).

  11. 11.

    Hagan, P. G., Nienaber, C. A. & Isselbacher, E. M. et al. The international registry of acute aortic dissection (IRAD): New insights into an old disease. JAMA. 283(7), 897–903 (2000).

  12. 12.

    Ruddy, J. M., Jones, J. A. & Ikonomidis, J. S. Pathophysiology of thoracic aortic aneurysm (TAA): Is it not one uniform aorta? role of embryologic origin. Prog Cardiovasc Dis. 56(1), 68–73 (2013).

  13. 13.

    Hinterseher, I., Erdman, R. & Elmore, J. R. et al. Novel pathways in the pathobiology of human abdominal aortic aneurysms. Pathobiology. 80(1), 1–10 (2013).

  14. 14.

    Lenk, G. M. et al. Whole genome expression profiling reveals a significant role for immune function in human abdominal aortic aneurysms. BMC Genomics. 8, 237 (2007).

  15. 15.

    Ziganshin, B. A. & Elefteriades, J. A. Guilt by association: A paradigm for detection of silent aortic disease. Ann Cardiothorac Surg. 5(3), 174–187 (2016).

  16. 16.

    Sajja, L. R. & Mannam, G. Internal thoracic artery: Anatomical and biological characteristics revisited. Asian Cardiovasc Thorac Ann. 23(1), 88–99 (2015).

  17. 17.

    Raitoharju, E., Seppala, I. & Lyytikainen, L. P. et al. A comparison of the accuracy of illumina HumanHT-12v3 expression BeadChip and TaqMan qRT-PCR gene expression results in patient samples from the tampere vascular study. Atherosclerosis. 226(1), 149–152 (2013).

  18. 18.

    Eden, E., Navon, R., Steinfeld, I., Lipson, D. & Yakhini, Z. GOrilla: A tool for discovery and visualization of enriched GO terms in ranked gene lists. BMC Bioinformatics. 10, 48-2105–10-48 (2009).

  19. 19.

    Hinterseher, I., Tromp, G. & Kuivaniemi, H. Genes and abdominal aortic aneurysm. Ann Vasc Surg. 25(3), 388–412 (2011).

  20. 20.

    He, R., Guo, D. C. & Estrera, A. L. et al. Characterization of the inflammatory and apoptotic cells in the aortas of patients with ascending thoracic aortic aneurysms and dissections. J Thorac Cardiovasc Surg. 131(3), 671–678 (2006).

  21. 21.

    Rizas, K. D., Ippagunta, N. & Tilson, M. D. 3rd Immune cells and molecular mediators in the pathogenesis of the abdominal aortic aneurysm. Cardiol Rev. 17(5), 201–210 (2009).

  22. 22.

    Della Corte, A., Quarto, C. & Bancone, C. et al. Spatiotemporal patterns of smooth muscle cell changes in ascending aortic dilatation with bicuspid and tricuspid aortic valve stenosis: Focus on cell-matrix signaling. J Thorac Cardiovasc Surg. 135(1), 8–18 (2008). 18.e1-2.

  23. 23.

    Sulkava, M., Raitoharju, E. & Levula, M. et al. Differentially expressed genes and canonical pathway expression in human atherosclerotic plaques - tampere vascular study. Sci Rep. 7, 41483 (2017).

  24. 24.

    Cantile, M., Schiavo, G., Terracciano, L. & Cillo, C. Homeobox genes in normal and abnormal vasculogenesis. Nutr Metab Cardiovasc Dis. 18(10), 651–658 (2008).

  25. 25.

    Arderiu, G. et al. HoxA5 stabilizes adherens junctions via increased Akt1. Cell Adh Migr. 1(4), 185–195 (2007).

  26. 26.

    Lillvis, J. H., Erdman, R. & Schworer, C. M. et al. Regional expression of HOXA4 along the aorta and its potential role in human abdominal aortic aneurysms. BMC Physiol. 11, 9–6793-11-9 (2011).

  27. 27.

    Santini, S., Boore, J. L. & Meyer, A. Evolutionary conservation of regulatory elements in vertebrate hox gene clusters. Genome Res. 13(6A), 1111–1122 (2003).

  28. 28.

    Lappin, T. R., Grier, D. G., Thompson, A. & Halliday, H. L. HOX genes: Seductive science, mysterious mechanisms. Ulster Med J. 75(1), 23–31 (2006).

  29. 29.

    Seo, D., Wang, T. & Dressman, H. et al. Gene expression phenotypes of atherosclerosis. Arterioscler Thromb Vasc Biol. 24(10), 1922–1927 (2004).

  30. 30.

    Bhatlekar, S., Fields, J. Z. & Boman, B. M. HOX genes and their role in the development of human cancers. J Mol Med (Berl). 92(8), 811–823 (2014).

  31. 31.

    Gorski, D. H. & Walsh, K. Control of vascular cell differentiation by homeobox transcription factors. Trends Cardiovasc Med. 13(6), 213–220 (2003).

  32. 32.

    Pruett, N. D., Visconti, R. P. & Jacobs, D. F. et al. Evidence for hox-specified positional identities in adult vasculature. BMC Dev Biol. 8, 93–213X-8-93 (2008).

  33. 33.

    Soshnikova, N. Hox genes regulation in vertebrates. Dev Dyn. 243(1), 49–58 (2014).

  34. 34.

    Rhoads, K. et al. A role for hox A5 in regulating angiogenesis and vascular patterning. Lymphat Res Biol. 3(4), 240–252 (2005).

  35. 35.

    Lin, E. Y. & Pollard, J. W. Role of infiltrated leucocytes in tumour growth and spread. Br J Cancer. 90(11), 2053–2058 (2004).

  36. 36.

    Ramachandran, S., Liu, P. & Young, A. N. et al. Loss of HOXC6 expression induces apoptosis in prostate cancer cells. Oncogene. 24(1), 188–198 (2005).

  37. 37.

    Motoki, T., Kurobe, H. & Hirata, Y. et al. PPAR-gamma agonist attenuates inflammation in aortic aneurysm patients. Gen Thorac Cardiovasc Surg. 63(10), 565–571 (2015).

  38. 38.

    Tian, L., Liao, M. F., Zhang, L., Lu, Q. S. & Jing, Z. P. A study of the expression and interaction of destrin, cofilin, and LIMK in debakey I type thoracic aortic dissection tissue. Scand J Clin Lab Invest. 70(7), 523–528 (2010).

  39. 39.

    Ishimine, H., Yamakawa, N. & Sasao, M. et al. N-cadherin is a prospective cell surface marker of human mesenchymal stem cells that have high ability for cardiomyocyte differentiation. Biochem Biophys Res Commun. 438(4), 753–759 (2013).

  40. 40.

    Kahler, C. M., Schratzberger, P. & Kaufmann, G. et al. Transendothelial migration of leukocytes and signalling mechanisms in response to the neuropeptide secretoneurin. Regul Pept. 105(1), 35–46 (2002).

  41. 41.

    Lund, S. A., Giachelli, C. M. & Scatena, M. The role of osteopontin in inflammatory processes. J Cell Commun Signal. 3(3-4), 311–322 (2009).

  42. 42.

    Desai, T. R., Leeper, N. J., Hynes, K. L. & Gewertz, B. L. Interleukin-6 causes endothelial barrier dysfunction via the protein kinase C pathway. J Surg Res. 104(2), 118–123 (2002).

  43. 43.

    Cheuk, B. L. & Cheng, S. W. Differential expression of elastin assembly genes in patients with stanford type A aortic dissection using microarray analysis. J Vasc Surg. 53(4), 1071–1078.e2 (2011).

  44. 44.

    Song, Y., Xie, Y. & Liu, F. et al. Expression of matrix metalloproteinase-12 in aortic dissection. BMC Cardiovasc Disord. 13, 34-2261–13-34 (2013).

  45. 45.

    Yasmin, M. E. C. M. & Wallace, S. et al. Matrix metalloproteinase-9 (MMP-9), MMP-2, and serum elastase activity are associated with systolic hypertension and arterial stiffness. Arterioscler Thromb Vasc Biol. 25(2), 372 (2005).

  46. 46.

    Balta, S. et al. Endocan: A novel inflammatory indicator in cardiovascular disease? Atherosclerosis. 243(1), 339–343 (2015).

  47. 47.

    Strauss, E., Waliszewski, K., Oszkinis, G. & Staniszewski, R. Polymorphisms of genes involved in the hypoxia signaling pathway and the development of abdominal aortic aneurysms or large-artery atherosclerosis. J Vasc Surg. 61(5), 1105–13.e3 (2015).

  48. 48.

    Fong, G. H. Potential contributions of intimal and plaque hypoxia to atherosclerosis. Curr Atheroscler Rep. 17(6), 510-015–0510-0 (2015).

  49. 49.

    Carbone, F. & Montecucco, F. Inflammation in arterial diseases. IUBMB Life. 67(1), 18–28 (2015).

  50. 50.

    Xu, J., Ehrman, B., Graham, L. M. & Eagleton, M. J. Interleukin-5 is a potential mediator of angiotensin II-induced aneurysm formation in apolipoprotein E knockout mice. J Surg Res. 178(1), 512–518 (2012).

  51. 51.

    Schonbeck, U., Sukhova, G. K., Gerdes, N. & Libby P.T(H)2 predominant immune responses prevail in human abdominal aortic aneurysm. Am J Pathol. 161(2), 499–506 (2002).

  52. 52.

    Leeper, N. J., Raiesdana, A. & Kojima, Y. et al. Loss of CDKN2B promotes p53-dependent smooth muscle cell apoptosis and aneurysm formation. Arterioscler Thromb Vasc Biol. 33(1), e1–e10 (2013).

  53. 53.

    Maleszewski, J. J. Inflammatory ascending aortic disease: Perspectives from pathology. J Thorac Cardiovasc Surg. 149(2 Suppl), S176–83 (2015).

  54. 54.

    Barry, M. et al. Histologic study of coronary, radial, ulnar, epigastric and internal thoracic arteries: Application to coronary artery bypass grafts. Surg Radiol Anat. 29(4), 297–302 (2007).

  55. 55.

    van Son, J. A., Smedts, F. & de Wilde, P. C. et al. Histological study of the internal mammary artery with emphasis on its suitability as a coronary artery bypass graft. Ann Thorac Surg. 55(1), 106–113 (1993).

  56. 56.

    Ferrari, G., Quackenbush, J. & Strobeck, J. et al. Comparative genome-wide transcriptional analysis of human left and right internal mammary arteries. Genomics. 104(1), 36–44 (2014).

Download references

Acknowledgements

This work was supported by the European Union 7th Framework Programme funding for the AtheroRemo project [201668], the Competitive Research Funding of Tampere University Hospital [grant × 51001 for T.L and 9S054 for E.R], the Academy of Finland: grants 286284 (T.L.), 285902 (E.R.), Finnish Foundation of Cardiovascular Research (T.L.); Finnish Cultural Foundation; Tampere Tuberculosis Foundation (T.L.); Emil Aaltonen Foundation (T.L. and N.O.); Yrjö Jahnsson Foundation (T.L.); and The Clinical Chemistry Research Foundation (M.S).

Author information

Affiliations

  1. Department of Clinical Chemistry, Fimlab Laboratories and Finnish Cardiovascular Research Center Tampere, Faculty of Medicine and Life Sciences, University of Tampere, Tampere, Finland

    • Miska Sulkava
    • , Emma Raitoharju
    • , Mari Levula
    • , Ilkka Seppälä
    • , Leo-Pekka Lyytikäinen
    • , Nina Mononen
    • , Reijo Laaksonen
    • , Niku Oksala
    •  & Terho Lehtimäki
  2. Heart Center, Department of Cardio-Thoracic Surgery, Tampere University Hospital and University of Tampere, Faculty of Medicine and Life Sciences, Tampere, Finland

    • Ari Mennander
    • , Otso Järvinen
    •  & Niku Oksala
  3. Research Unit of Molecular Epidemiology, Helmholtz Zentrum, German Research Center for Environmental Health, Munich, Germany

    • Thomas Illig
    •  & Norman Klopp
  4. Hannover Unified Biobank, Hannover Medical School, Hannover, Germany

    • Thomas Illig
    •  & Norman Klopp
  5. Institute for Human Genetics, Hannover Medical School, Hanover, Germany

    • Thomas Illig
  6. Department of Clinical Physiology, Tampere University Hospital and University of Tampere, Faculty of Medicine and Life Sciences, Tampere, Finland

    • Mika Kähönen

Authors

  1. Search for Miska Sulkava in:

  2. Search for Emma Raitoharju in:

  3. Search for Ari Mennander in:

  4. Search for Mari Levula in:

  5. Search for Ilkka Seppälä in:

  6. Search for Leo-Pekka Lyytikäinen in:

  7. Search for Otso Järvinen in:

  8. Search for Thomas Illig in:

  9. Search for Norman Klopp in:

  10. Search for Nina Mononen in:

  11. Search for Reijo Laaksonen in:

  12. Search for Mika Kähönen in:

  13. Search for Niku Oksala in:

  14. Search for Terho Lehtimäki in:

Contributions

M.S. wrote the manuscript and performed statistical analysis. E.R. acquired data, reviewed/edited the manuscript and acquired funding. A.M. acquired data and reviewed/edited the manuscript. M.L. participated in cohort collection and acquired data. I.S. performed statistical analysis and reviewed the manuscript. L.-P.L. performed statistical analysis. T.I. acquired data and reviewed the manuscript. O.J. and N.K. acquired data. N.M. participated in cohort collection and reviewed/edited the manuscript. R.L. acquired funding. M.K. acquired funding and reviewed the manuscript. N.O. acquired data, participated and organized cohort collection and supervision, and reviewed/edited the manuscript. T.L. handled funding and supervision, participated in cohort collection and reviewed/edited the manuscript.

Competing Interests

The authors declare that they have no competing interests.

Corresponding author

Correspondence to Miska Sulkava.

Electronic supplementary material

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Creative Commons BY

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.