Thorac cardiovasc Surg 2018; 66(04): 278-286
DOI: 10.1055/s-0036-1583525
Review Article
Georg Thieme Verlag KG Stuttgart · New York

The Value of Circulating Biomarkers in Bicuspid Aortic Valve-Associated Aortopathy

Shiho Naito
Department of Cardiac Surgery, Central Hospital Bad Berka, Bad Berka, Germany
,
Mathias Hillebrand
Department of Cardiology, University Heart Center Hamburg, Hamburg, Germany
,
Alexander Martin Justus Bernhardt
Department of Cardiovascular Surgery, University Heart Center Hamburg, Hamburg, Germany
,
Annika Jagodzinski
Department of Cardiology, University Heart Center Hamburg, Hamburg, Germany
,
Lenard Conradi
Department of Cardiovascular Surgery, University Heart Center Hamburg, Hamburg, Germany
,
Christian Detter
Department of Cardiovascular Surgery, University Heart Center Hamburg, Hamburg, Germany
,
Karsten Sydow
Department of Cardiology, University Heart Center Hamburg, Hamburg, Germany
,
Hermann Reichenspurner
Department of Cardiovascular Surgery, University Heart Center Hamburg, Hamburg, Germany
,
Yskert von Kodolitsch
Department of Cardiology, University Heart Center Hamburg, Hamburg, Germany
,
Evaldas Girdauskas
Department of Cardiovascular Surgery, University Heart Center Hamburg, Hamburg, Germany
› Author Affiliations
Further Information

Address for correspondence

Evaldas Girdauskas, MD, PhD
Department of Cardiovascular Surgery, University Heart Center Hamburg
Martinistraße 52, 20246 Hamburg
Germany   

Publication History

06 January 2016

14 March 2016

Publication Date:
05 May 2016 (eFirst)

 

Abstract

Traditional risk stratification model of bicuspid aortic valve (BAV) aortopathy is based on measurement of maximal cross-sectional aortic diameter, definition of proximal aortic shape, and aortic stiffness/elasticity parameters. However, conventional imaging-based criteria are unable to provide reliable information regarding the risk stratification in BAV aortopathy, especially considering the heterogeneous nature of BAV disease. Given those limitations of conventional imaging, there is a growing clinical interest to use circulating biomarkers in the screening process for thoracic aortic aneurysms as well as in the risk-assessment algorithms. We aimed to systematically review currently available biomarkers, which may be of value to predict the natural evolution of aortopathy in individuals with BAV.


#

Introduction

Bicuspid aortic valve (BAV) is the most common congenital cardiac malformation, with a prevalence estimated between 0.5 and 2%.[1] [2] [3] [4] Although aortic valve dysfunction is the most common complication of BAV disease, dilatation of the proximal aorta, so-called bicuspid aortopathy, is present in approximately one half of the BAV population and might be associated with an increased rate of adverse aortic events (i.e., aortic dissection/rupture).[5] [6]

The current guidelines from both Europe and the United States recommend replacing the ascending aorta in patients who are undergoing concomitant aortic valve surgery if the ascending aortic diameter is > 45 mm.[7] [8] Aggressive recommendations for proximal aortic surgery in BAV patients were mainly extrapolated from surgical guidelines for management of aortopathy in genetic aortic syndromes such as Marfan or Loeys–Dietz.[9] [10] However, several studies documented a low risk for aortic events after isolated aortic valve replacement (AVR) in BAV patients and they therefore questioned the appropriateness of such an aggressive surgical strategy.[11] [12] Moreover, the aortic diameter alone is often not reliable enough to identify patients at risk for rupture or dissection.[13] [14] Additional size-independent predictive tools may help identify BAV patients at increased risk for aortic events ([Table 1]).

Table 1

Potential biomarkers in bicuspid aortopathy

Biomarker

Original publication

MMP-2

Drapisz et al,[34] Wang et al,[45] LeMaire et al,[46] Tzemos et al,[50] Ikonomidis et al,[35] Boyum et al[44]

MMP-8

Ikonomidis et al,[35] Mohamed et al[47]

MMP-9

Boyum et al,[44] Wang et al,[45] Ikonomidis et al[35]

TIMP-1

Ikonomidis et al,[48] Mohamed et al[47]

TIMP-2

Mohamed et al[47]

TIMP-3

Mohamed et al[47]

TIMP-4

Mohamed et al,[47] Ikonomidis et al[48]

miR-1

Ikonomidis et al[35]

miR-21

Ikonomidis et al[35]

miR-29b

Boon and Dimmeler[58]

TGF-β

Hillebrand et al[62]

Alpha 1-antitrypsin

Kilickesmez et al[65]

ADMA

Drapisz et al[34]

S100A12

Hofmann Bowman et al[76]

HMGB-1

Data not available

sRAGE

Branchetti et al[33]

Abbreviations: ADMA, asymmetric dimethylarginine; HMGB-1, high-mobility group box 1; miR, microRNA; MMP, matrix metalloproteinase; sRAGE, soluble for receptor for advanced glycation end product; S100A12, S100 calcium-binding protein A12; TGF-β, transforming growth factor-β; TIMP, tissue inhibitors of metalloproteinase.



#

Predictive Value of Size-Based Parameters/Limitations of Conventional Imaging

Conventional risk stratification model of bicuspid aortopathy is traditionally based on measurement of maximal cross-sectional aortic diameter,[15] definition of proximal aortic shape,[16] and aortic stiffness/elasticity parameters.[17] All these parameters are relatively easy to obtain by means of conventional imaging but are obviously of limited value to predict the risk of adverse aortic events.

Cross-Sectional Aortic Diameter

The current treatment guidelines of bicuspid aortopathy implement cross-sectional aortic diameter as the main criterion for the decision regarding surgery.[7] [8] However, the adherence to size criteria alone would fail to prevent the majority of acute aortic dissections[13] [18] even considering more aggressive treatment guidelines adopted for BAV patients. Furthermore, recent meta-analysis demonstrated the impact of functional BAV phenotype on the risk of late post-AVR aortic dissection, irrespective of aortic size criterion.[19]


#

Proximal Aortic Shapes

The stratification of BAV patients according to the proximal aortic shape was introduced to better accommodate the heterogeneity of BAV-associated aortopathy, separating BAV patients with a possible greater expression of genetically triggered connective tissue disease as compared with BAV patients with predominant contribution of hemodynamic factors.[20] Such scientific efforts resulted in several classification systems of proximal aortic shape.[21] [22] [23] The lack of standardization and prognostic validation of such stratification models are major drawbacks that obviously limit their applicability in the clinical practice.


#

Aortic Stiffness/Elasticity Parameters

Given the limitations of aneurysm size-based criteria, aortic stiffness and elasticity indices were introduced to improve “functional” risk stratification in thoracic aortic aneurysm (TAA) disease.[24] [25] [26] [27] One landmark study revealed markedly abnormal aortic elasticity in 42% BAV patients with normal-sized aorta and no or mild aortic valve impairment.[17] The impaired aortic stiffness did not correlate with aortic dilation and the authors concluded that abnormal load-bearing characteristics of the aortic wall in BAV's can neither be attributed nor identified by simple assessment of aortic size. However, the lack of prospective longitudinal data and validated cutoff values still limit broader clinical application of such functional aortic testing.

The applicability of cross-sectional imaging techniques is further limited by safety concerns, which is particularly true for computed tomography (CT) angiography. Transthoracic echocardiography (TTE) is the most widely used cardiac imaging modality but is limited by intra- and interobserver variabilities.[28] Moreover, proximal aortic measurements by TTE may be less reliable than by CT or cardiac magnetic resonance, especially in the presence of asymmetric aortic root aneurysm.[29] Furthermore, echocardiography has a limited value in visualization of the more distal aorta (i.e., beyond the aortic root) and may therefore miss an aneurysm of the mid-ascending aorta.[30]

Until recently, routine laboratory tests have played a minor role in the risk assessment/diagnosis of aortic diseases (e.g., D-dimer in acute aortic dissection). Novel diagnostic modalities emerged in the last decade, which focused on circulating biomarkers as an initial diagnostic modality.[31] There is a growing clinical interest to use such biomarkers in the screening process for TAAs as well as in the risk-assessment algorithms of aortopathy progression.[32] [33] [34] [35] Efforts are being made to create circulating transcript biomarker panels and aortic tissue biomarker panels for the identification of ascending aortic aneurysms.[36] [37] We aimed to systematically review currently available biomarkers, which may be of value to predict the natural evolution of aortopathy in individuals with BAV ([Fig. 1]).

Zoom Image
Fig. 1 Molecular pathways and potential biomarkers in bicuspid aortopathy.

#
#

Matrix Metalloproteinases

Matrix metalloproteinases (MMPs) are members of family of enzymes being produced by a variety of inflammatory, endothelial, vascular, and periadventitial connective tissue cells.[38] They are involved in the degradation process of extracellular matrix components, including fibronectin, collagen, elastin, and proteoglycans under physiologic and pathologic conditions. Although MMPs have long been considered to be the principal mediators of extracellular matrix (ECM) destruction, notably of fibrillar collagens, they have been shown to function more as processing enzymes, which control numerous cell processes. Moreover, several MMPs (e.g., MMP 12) are key participants in diverse immune, anti-inflammatory processes rather than ECM proteolysis.

There are 23 currently known human tissue MMPs.[38] However, a particular importance for the pathogenesis of vascular pathologies is documented only for MMP-2, MMP-8, and MMP-9.[39] Elevated plasma concentration of MMP-9 has been previously demonstrated in patients with abdominal aortic aneurysm[40] and in a cohort of thoracic aneurysm patients with persisting endoleak after endovascular therapy (thoracic endovascular aneurysm repair, TEVAR).[41] Koullias et al found significantly increased MMP-9 concentrations in aortic tissue from patients undergoing surgery for TAA as compared with the normal aorta by means of a semiquantitative analysis.[42] Similarly, Schmoker et al revealed a significant MMP-2 reduction and MMP-9 elevation in the TAA group as compared with MMPs activity in the normal aorta using an antibody capture technique.[43]

Recent reports demonstrated an increased MMP-9 activity in patients with BAV-associated aortopathy.[44] [45] We summarized currently available data with regard to MMP-9 activity in BAV versus tricuspid aortic valve (TAV) patients and in dilated versus normal BAV aortas (see [Table 2]). LeMaire et al showed a normal MMP-9 expression in ascending aortic aneurysms associated with BAV,[46] whereas Boyum et al demonstrated significantly increased MMP-9 concentrations.[44] Furthermore, Mohamed et al[47] and Ikonomidis et al[35] found no significant difference in MMP-9 activity between BAV and TAV patients, whereas Boyum et al[44] revealed markedly increased MMP-9 and MMP-2 activities in the walls of aneurysms associated with bicuspid as compared with TAVs.

Table 2

Characteristics of biomarkers in bicuspid aortopathy

Biomarker

Expression in BAV vs. TAV[a]

Correlation with aortic diameter in BAV/TAV[b]

MMP-2

BAV > TAV (tissue)[44] [46] [47] [49]

BAV = TAV (tissue/plasma)[35]

Yes (plasma)[34] [45] [50]

MMP-9

BAV > TAV (tissue)[44]

BAV = TAV (tissue)[35] [47]/(plasma)[35]

BAV < TAV (tissue)[46]

No (tissue)[46]/(plasma)[34] [45] [50] [65]

TIMP-1

BAV = TAV (tissue)[35] [44] [46] [47]/(plasma)[35]

BAV < TAV (tissue)[49]

No (plasma)[45] [46] [50]

TIMP-2

BAV = TAV (tissue)[35] [44] [46]/(plasma)[35]

BAV > TAV (tissue)[47]

No (plasma)[45] [46] [50]

TIMP-3

BAV > TAV (tissue)[47]

?

TIMP-4

BAV > TAV (tissue)[47]

No (plasma)[45]

microRNA

Increased in TAV only (tissue)[35]

?

TGF-β

Increased in BAV only (plasma)[62]

?

Alpha 1-antitrypsin

?

Yes (plasma)[65]/(tissue)[64]

ADMA

?

Yes (plasma)[34]

S100A12

BAV = TAV (plasma)[33]

No (plasma)[33]

Yes (tissue)[76]

HMGB-1

BAV = TAV (plasma)[33]

No (plasma)[33]

Yes (tissue)[79]

sRAGE

BAV > TAV (plasma)[33]

No (plasma)[33]

Abbreviations: ADMA, asymmetric dimethylarginine; BAV, bicuspid aortic valve; HMGB-1, high-mobility group box 1; MMP, matrix metalloproteinase; sRAGE, soluble for receptor for advanced glycation end product; S100A12, S100 calcium-binding protein A12; TAV, tricuspid aortic valve; TGF-β, transforming growth factor-β; TIMP, tissue inhibitors of metalloproteinase.


a Comparative biomarker studies which analyzed the expression of circulating/aortic tissue biomarkers in BAV versus TAV cohorts.


b Biomarker studies which analyzed correlation patterns between concentration of circulating/tissue biomarkers and ascending aortic diameters.


Mohamed et al examined aortic tissue from four predefined sites (i.e., concavity and convexity of the tubular ascending aorta as well as distal and proximal aortic sites) in patients with ascending aortic aneurysms.[47] Subsequently, the study cohort was subdivided into different subgroups as a function of age, diameter of aortic aneurysm, gender, and aortic valve disease (i.e., stenosis vs. regurgitation). Using simultaneous multiplex protein detection system (Bio-Plex, Bio-Rad, Hercules, California, United States), the authors showed that MMP-8/MMP-9 concentrations were significantly higher in the convexity than in the concavity of ascending aorta.

Recently, several studies reported an increased MMP-2 activity in TAAs associated with BAV versus TAV.[44] [46] [47] [48] [49] Moreover, strong linear correlation has been demonstrated between MMP-2 concentrations and proximal aortic diameters in patients with BAV disease.[34] [50] In contrast, Ikonomidis et al found no significant difference in MMP-2 concentrations between BAV and TAV patients with TAAs.[35]

In summary, enzyme-linked immunosorbent assays for measuring circulating MMPs appears to be a simple and readily available technique for screening and monitoring individuals with BAV aortopathy. However, published data on MMPs expression in BAV aortopathy remain controversial ([Table 2]). Given the heterogeneity of BAV-associated aortopathy,[51] potential association of MMPs expression with specific BAV phenotypes should be addressed in future prospective studies.


#

Tissue Inhibitors of Metalloproteinases

Tissue inhibitors of metalloproteinases (TIMPs) are important regulators of local MMPs activity. Decreased concentrations of TIMP-1 and TIMP-2 were previously identified in TAAs as compared with controls.[43]

Several studies demonstrated no significant difference in aortic tissue TIMP-1/TIMP-2 expression between patients with BAV- versus TAV-associated aortopathies.[35] [44] [46] Furthermore, Wang et al[45] and Tzemos et al[50] revealed comparable plasma concentrations of TIMP-1/TIMP-2 in BAV patients with and without aortic dilatation. Opposite to these findings, other authors reported a significant elevation of TIMP-2,-3,-4 concentrations in BAV versus TAV patients.[47] Moreover, recent meta-analysis by Rabkin demonstrated a significant reduction of TIMP-1 in BAV patients as compared with TAV patients, while TIMP-2 expression was comparable between both groups.[49]

Different patterns of TIMPs expression have been demonstrated in patients with BAV-associated aortopathy, depending on the site of aortic tissue collection (i.e., concavity vs. convexity)[47] and cusp fusion pattern (i.e., right/left coronary [R/L] vs. right/noncoronary [R/N] cusp fusion).[48] Mohamed et al showed increased TIMP-3 concentrations in various segments of BAV aortas as well as increased TIMP-2/TIMP-4 concentrations in the proximal aorta of BAV patients as compared with TAV patients.[47]

Interestingly, Ikonomidis et al examined ascending aortic aneurysm tissue and blood samples from BAV versus TAV patients and evaluated the relationships between plasma and tissue measurements of MMPs (-1, -2, -3, -8, -9), TIMPs, and microRNAs (miRs).[35] They demonstrated significant differences in tissue and plasma concentrations for the majority of examined biomarkers. Linear relationships between tissue and plasma concentrations were found only for MMP-8 and TIMP-1, -3, and -4. Ikonomidis et al failed to identify one single parameter which would be predictive of aneurysmal disease in TAV or BAV group, while a combination of multiple biomarkers integrated in a stepwise algorithm was highly sensitive and specific for the formation of aneurysm.

Interaction between MMPs and TIMPs

Ikonomidis et al examined ascending aortic aneurysm tissue of BAV patients and calculated MMP/TIMP ratios, global MMP activity, and so-called proteolytic balance as compared with the normal sized aorta.[48] They hypothesized that specific BAV cusp fusion patterns may be associated with distinct combinations of MMPs/TIMPs. All BAV aortas displayed significantly elevated global MMP activity and increased MMP-9 concentrations, whereas MMP-7, MMP-8, TIMP-1, and TIMP-4 were significantly decreased. Moreover, BAV aortas with specific cusp fusion patterns (i.e., R/L, R/N, and left/non-coronary cusp fusion) demonstrated distinct combinations of MMP/TIMP activity. The R/L cusp fusion was associated with an increased proteolytic balance for MMP-1, MMP-9, and MMP-12 as compared with other two cusp fusion patterns. Based on these findings, the authors proposed that R/L fusion type may be accompanied with a more “malignant” form of BAV aortopathy.

In summary, published data on TIMPs expression in BAV aortopathy are controversial ([Table 2]). Such controversial findings may at least in part be explained by marked heterogeneity of BAV-associated aortopathy.[51] Reporting BAV phenotype-specific data would be helpful when evaluating the value of MMPs/TIMPs in BAV aortopathy.


#
#

MicroRNAs

miRs are small noncoding RNA molecules which have been shown to be super ordinate regulators of virtually all biological processes.[52] miRs are endogenously encoded in the genome and do posttranscriptionally inhibit mRNAs by attenuating protein translation and inducing mRNA degradation.[53] miRs were recently reported to be modulators of vascular wall homeostasis and have an impact on vascular smooth muscle cells (VSMC) migration, contraction, differentiation, proliferation, and apoptosis.[52] Recent data indicate that a few miRs (i.e., miR-21, miR-26, miR-29, and miR-143/145) may play a role in the pathogenesis of aortic aneurysm or dissection.[53] Liu et al examined miRs expression in a rat model by exposing abdominal aorta for 20 minutes to saline supplemented with calcium chloride and collagenase and found a set of differentially expressed miRs in abdominal aortic aneurysm tissue after 1 month postexposition.[54] Elia et al examined the miR-143/miR-145 cluster in a murine model of atherosclerosis and reported significantly decreased expression in the miRs knockout mice.[55] They also evaluated biopsies from human ascending aortic aneurysms and found significantly downregulated miR-143/miR-145 expression. Other authors examined miRs expression in aortic tissue of patients with TAAs and dissections.[56] [57] They reported an inverse significant correlation between miRs expression (miRs-1, -21, -29a, and -133a) and TAA diameter.[57]

Ikonomidis et al analyzed miRs (-1, -21, -29a, -133a, -143, and -145) in ascending aortic aneurysm tissue and plasma samples of BAV and TAV patients.[35] The expression of miR-1 and miR-21 in aortic tissue showed significant differences between TAV and BAV groups, but plasma concentrations were similar. Boon and Dimmeler examined specifically miR-29b expression in biopsies of human thoracic aneurysms and found significantly increased concentrations in patients with BAV and TAV valves.[58]

Association between MMP and microRNAs

miRs have been proposed to modulate regulatory processes of specific protein targets within the cardiovascular system, such as MMPs and ECM components.[59] Jones et al examined putative miRs-binding sites in TAA tissue and revealed MMP-2 and MMP-9 as potential targets for miR-29a and miR-133a.[57] The authors found a significant inverse relationship between miR-29a and total MMP-2 expression in TAA tissue samples. The authors concluded that decreased concentrations of aortic tissue miR-29a may represent a potential mechanism by which MMP-2 protein induction occurs, driving aortic remodeling during TAA development.[57]

In summary, the impact of miRs in the development and progression of BAV aortopathy is still insufficiently clarified and requires further evaluation in large-scale prospective studies.


#
#

Transforming Growth Factor-β

Transforming growth factor-β (TGF-β) is a soluble cytokine which impacts vascular remodeling, the impaired signaling of which can alter ECM structure and composition.[60] Excessive TGF-β activation and signaling are currently thought to contribute to the pleiotropic manifestations of Marfan syndrome, including aortic root dilatation and aortic dissection.[61] Matt et al found a significant correlation between serum concentrations of TGF-β and growth rate of the proximal aorta in a murine model of Marfan syndrome.[61] Moreover, TGFBR2 mutations were found responsible for the inherited predisposition to familial TAAs and dissections in 5% of cases.[60]

Hillebrand et al reported elevated circulating concentrations of TGF-β1 in BAV patients as well as in NOTCH1 and FBN1 mutation groups.[62] These authors proposed the use of total serum TGF-β1 as a valuable marker in the wide spectrum of genetic aortic syndromes, including BAV aortopathy.


#

Alpha 1-Antitrypsin

Alpha 1-antitrypsin is the best known circulating serine protease inhibitor which protects human tissues from proteolytic activity of enzymes of inflammatory cells. In the case of alpha 1-antitrypsin deficiency, the most frequent clinical manifestation is lung emphysema.[63]

Schachner et al observed that alpha 1-antitrypsin expression was reduced in the aortic wall of patients presenting with an acute type A aortic dissection.[64] The authors hypothesized that locally decreased alpha 1-antitrypsin activity may potentiate proteolytic tissue damage to the aortic wall, and thereby, weaken the arterial wall which results in either progressive aortic dilatation or aortic dissection. In line with these findings, Kilickesmez et al found that plasma levels of alpha 1-antitrypsin correlated negatively with the dilatation of ascending aorta in patients with BAV disease.[65]

These preliminary studies[64] [65] [66] indicate that plasma concentrations of alpha 1-antitrypsin may be a valuable biomarker for prediction of aortic aneurysm growth and risk for aortic dissection in patients with a BAV disease.


#

Asymmetric Dimethylarginine

Asymmetric dimethylarginine (ADMA) is an endogenous competitive inhibitor of nitric oxide (NO) synthase. Increased ADMA concentrations are associated with reduced NO synthesis resulting in an impaired endothelium-dependent vasodilation and reduced NO metabolite concentrations.[67] Interestingly, ADMA has emerged as an independent predictor of cardiovascular and overall mortality.[68] [69]

Drapisz et al demonstrated that circulating plasma ADMA concentrations were higher in BAV patients as compared with TAV controls.[34] ADMA concentrations in BAV patients correlated significantly with aortic annulus diameter, peak aortic velocity, aortic stiffness index, aortic strain, and MMP-2 values. However, it is unclear whether a single measurement of ADMA concentration may be strong enough to predict the risk for an individual patient. Therefore, a multimarker strategy consisting of circulating ADMA concentrations in combination with MMP-2 concentration may be a valuable approach to identify proximal aortopathy and impaired aortic elastic properties in nonstenotic BAV patients.


#

Receptor for Advanced Glycation End Product, Its Ligands (S100A12 and HMGB-1), and Soluble RAGE

Advanced glycation end products (AGEs) are a heterogeneous group of molecules formed by nonenzymatic glycation and oxidation of proteins or lipids. The receptor for AGEs (RAGE) is a multiligand member of the immunoglobulin superfamily of cell surface molecules which binds also proinflammatory cytokine-like mediators (RAGE ligands) of the S100/calgranulin family (i.e., S100 calcium-binding protein A12 [S100A12] and S100B)[70]; high-mobility group box 1 (HMGB-1)/amphoterin and amyloid-β-peptide and β-sheet fibrils.[71]

The AGE–RAGE axis (i.e., RAGE, its proinflammatory ligands—s100A12 and HMGB-1, and circulating sRAGE) seems to play an important role in the progression of vascular dysfunction and may be useful as potential biomarkers in TAA disease. The members of RAGE family are integrated into the common metabolic pathway, which is initiated by stimulation of RAGE receptors by accumulating AGE products.[72] RAGE signaling triggers the increased production of reactive oxygen species (ROS), at least in part through NADPH oxidase. Moreover, sustained stimulation of RAGE and increased ROS activate key transcription factors, nuclear factor-κB and Egr-1, which induce fundamental inflammatory mechanisms.[72] Increased migration and activation of RAGE-expressing inflammatory cells (i.e., neutrophils, monocytes/macrophages, T cells) result in release of proinflammatory RAGE ligands (S100A12 and HMGB-1).[72] These ligands magnify activation of transcription factors via interaction with RAGE, thereby amplifying oxidative stress and tissue damage. The AGE–RAGE axis may lead to the irreversible tissue injury and cause increased arterial wall stiffness by inducing intracellular oxidative stress responses.[73] sRAGE represents the soluble form of RAGE and may be removed from the cell surface by action of MMPs, thereby producing a circulating form of RAGE.

Schmidt et al postulated the two-hit model for vascular injury mediated by RAGE and its ligands.[74] The first hit is an increased expression of RAGE and its ligands within the vasculature, whereas the second hit is represented by various form of stress (ischemic stress, immune/inflammatory stimuli, physical/hemodynamic stress, or modified lipoproteins) which leads to exaggerated cellular response promoting vascular lesions.

In vivo data indicate that RAGE knockout mice are resistant to aneurysm formation in the genetic model. The inhibition of AGE signaling via targeted gene deletion of RAGE dramatically reduced the incidence of abdominal aortic aneurysms in a murine model. Moreover, the expression of AGE–RAGE axis was found to be significantly upregulated in human aneurysm tissues.[75]

S100 Calcium-Binding Protein A12

S100A12 is a small calcium-binding protein that is a ligand of RAGE and is endogenously expressed in cells intimately linked to vascular disease, that is, granulocytes and myeloid cells.

Hofmann et al found pathologic aortic remodeling with disarray of elastic fibers, increased fibrosis, and increased MMP-2 concentrations which were associated with aneurysmal dilatation of the aorta in a transgenic murine model expressing human S100A12 in VSMCs.[76] Moreover, the authors demonstrated an increased S100A12 expression in VSMCs adjacent to the site of cystic media necrosis in human aortic tissue of TAAs. Such an upregulated S100A12 expression promoted the change of VSMCs phenotype and resulted in an increased interleukin-6 production, MMP2 expression, and enhanced TGF-β signaling pathways.[76]

S100A12 positive cells were identified in the media of ascending aorta from patients with an acute type A dissection and TAA.[77] Moreover, the reduction of S100A12 expression in human aortic VSMCs induced by small hairpin RNA attenuated gene/protein expression of several inflammatory- and apoptosis-regulating factors.[77]

On the contrary, Branchetti et al found no significant correlation between plasma concentrations of S100A12 and proximal aortic diameters in patients undergoing AVR with or without proximal aortic surgery.[33] Moreover, there was no significant difference in circulating S100A12 concentrations between BAV and TAV patients.


#

High-Mobility Group Box 1

HMGB-1 is a RAGE ligand and potent proinflammatory cytokine which is passively released from injured/necrotic cells and actively secreted by stimulated inflammatory cells.[78] Increased HMGB-1 expression has been also linked to the upregulation of MMPs. Kohno et al showed increased HMGB-1 expression in human abdominal aortic aneurysm tissue as compared with normal aortas.[79] They also found a significant correlation between HMGB-1 expression and MMP-2/MMP-9 activity in a murine model of abdominal aortic aneurysm.

In contrast, Branchetti et al found no significant correlation between circulating concentrations of HMGB-1 and ascending aortic diameter in patients with BAV aortopathy.[33] Moreover, plasma values of HMGB-1 were comparable between BAV and TAV patients.


#

Soluble Receptor for Advanced Glycation End Product

The release of sRAGE is related to the shedding of cellular RAGE in response to its activation from ligands. Yamagishi et al reported positive linear correlation between plasma concentrations of AGE and sRAGE in nondiabetic patients.[80]

Branchetti et al analyzed sRAGE values in 135 patients (74 with BAV and 61 with TAV) undergoing AVR surgery due to aortic stenosis/regurgitation and/or ascending aortic replacement.[33] Plasma concentrations of sRAGE were significantly associated with the presence of BAV disease. Within the BAV cohort, patients with ascending aortic dimensions ≥ 45 mm had higher sRAGE values than patients without aortic dilatation (< 45 mm). Moreover, BAV patients aged < 60 years who underwent combined AVR and ascending aortic surgery had significantly higher sRAGE values than those aged > 60 years and without aortic dilatation. Histological examination of ascending aortic tissue revealed a significant association between sRAGE concentrations and dysfunctional aortic wall microstructure (i.e., elastin degradation, proteoglycan deposition). The authors concluded that higher concentrations of sRAGE may be useful as a marker of BAV-associated aortopathy, although no linear correlation between ascending aortic diameter and sRAGE values exist.

In summary, RAGE, its proinflammatory ligands (i.e., s100A12 and HMGB-1), and circulating sRAGE may emerge as novel promising biomarkers in BAV-associated aortopathy and require further validation in future prospective trials.


#
#

Conclusions

Scientific and clinical understanding of BAV-associated aortopathy is rapidly expanding ([Fig. 1]). Given the limited value of aneurysm size-based parameters, clinical practice appears to benefit from readily available and reliable diagnostic tools to predict/monitor BAV-associated aortopathy. In this context, circulating blood-based biomarkers are a novel and promising development which may become an established screening/predictive tool in thoracic aortic aneurysmal disease. Herein, analyzed biomarkers are specifically involved in the maintenance of vascular homeostasis (e.g., proteinases) and have been shown to be prominently expressed in human aortic aneurysm disease. Linear correlation pattern has been consistently demonstrated between concentrations of circulating “vascular” biomarkers and proximal aortic diameters. Furthermore, previous animal studies provide compelling experimental evidence that those markers play a major role in the pathogenesis of aneurysm formation. However, to the best of our knowledge, there are no sensitivity/specificity data for herein presented biomarkers to discriminate BAV patients with and without aortopathy, to predict an increased risk of aortic events (e.g., aortic dissection) and progressive aortic dilatation during follow-up.

No single biomarker has been identified yet to be predictive of aortic aneurysm disease and its progression. Based on the data summarized in this review article, a multimarker strategy seems to be a more promising approach.[35] [36] In particular, the combination of biomarkers consisting of circulating MMP-2/MMP-9 concentrations as combined with miR-29a, ADMA, and sRAGE concentrations seems to be a valuable approach to identify proximal aortopathy in BAV patients. Although MMPs/TIMPs, alpha 1-antitrypsin, and TGF-β are already available on a routine clinical basis, the majority of above-mentioned biomarkers have not yet been validated and might be only used in an experimental setting. Moreover, no prospective large-scale clinical studies are available as evidence to support biomarker-based decision making in BAV aortopathy. Biomarkers may serve as elegant tools to unify the anatomical heterogeneity of BAV aortopathy in a simple serologic risk stratifying model. Moreover, those biomarkers might be of similar value for other aortopathies (e.g., Marfan/TAV/Loeys–Dietz-associated aortic diseases), since they may share similar pathophysiologic mechanisms with BAV aortopathy. The use of circulating biomarkers in combination with a high-definition, noninvasive functional imaging has a great potential to enhance diagnostic accuracy and cost effectiveness of care in BAV-associated aortopathy.


#
#

No conflict of interest has been declared by the author(s).

Acknowledgments

The following coauthors contributed to this study: Shiho Naito—collection of literature/writing of the article; Mathias Hillebrand—TGF-β section; Alexander Bernhardt—formatting of [Fig. 1]/RAGE section; Annika Jagodzinski—MicroRNAs section; Lenard Conradi/Christian Detter—review of the article; Karsten Sydow—ADMA section; Hermann Reichenspurner—final review/re-review of the article; Yskert von Kodolitsch—introduction section/formatting of the article; Evaldas Girdauskas—study design/writing of final document/formatting of article/coordination.


Address for correspondence

Evaldas Girdauskas, MD, PhD
Department of Cardiovascular Surgery, University Heart Center Hamburg
Martinistraße 52, 20246 Hamburg
Germany   


Zoom Image
Fig. 1 Molecular pathways and potential biomarkers in bicuspid aortopathy.