Abstract

Introduction

Aortic aneurysm is morphologically defined as a focal dilation and structural degradation of the aorta, which can occur at different sites along the aortic tree. Pathologically, formation and progression of an aortic aneurysm is driven by dysregulated cellular and acellular events. While thorough reviews on the cellular mechanisms and current medical management of thoracic aortic aneurysm (TAA)^1,2 and abdominal aortic aneurysm (AAA)^2–4 have been published elsewhere, this review focuses on the role of the aortic extracellular matrix (ECM) and the proteases and their inhibitors that maintain the integrity and homeostasis of the ECM in the pathogenesis of TAA or AAA in humans and animal models.

Aortic wall structure and the ECM

The aorta is the largest conduit artery in the body. Owing to its extraordinary ability to expand and recoil, the aorta also serves as a reservoir that transforms the highly pressured and pulsatile heart output into a flow of moderate fluctuations. The composition and three-dimensional organization of its ECM is critical for optimal physiological functions of the aorta. The aortic wall comprises an intricate arrangement of endothelial cells (ECs), smooth muscle cells (SMCs), fibroblasts (FBs), and ECM proteins in three layers: tunica intima, tunica media, and tunica adventitia (Fig. ​(Fig.1).1). The tunica intima consists of a single layer of ECs that lines the lumen of the blood vessel, and is anchored to the underlying basement membrane, a highly specialized ECM network consisting primarily of laminin, collagen type IV, fibronectin, perlecan, and heparan sulfate proteoglycans⁵. This basement membrane also plays a pivotal role in signaling events that regulate EC migration, invasion, proliferation, and survival⁶. The basement membrane together with the internal elastic lamina (IEL) serves as an interface between the tunica intima and tunica media. IEL is a thin sheet of elastic fiber with a variable number of fenestrations that facilitate diffusion of molecules through the tunica intima to the tunica media⁷. The tunica media consists of concentric elastic lamellae with circumferentially oriented SMCs. This layer is rich in ECM proteins, such as proteoglycans (PGs), glycoproteins, glycosaminoglycans (GAGs), as well as collagen. The intermittent layers of elastic fibers and SMCs in this layer provide the compliance and recoil properties of the aortic wall⁸. The medial layer of the aortic wall is the most variable of the tunics among different arteries. The differences in ECM composition in this layer account for the main variations in the function and properties of the arterial tree. The external elastic lamina (EEL) forms a boundary between the tunica media and tunica adventitia, the outermost layer of the aortic wall that is rich in collagen, FBs, and elastic fiber, and provides the tensile strength of the aortic wall. This collagen-rich tunica also harbors innervations, lymphatics, and the vasa vasorum, which are responsible for the high susceptibility of the adventitial layer to vascular inflammation.

Open in a separate window

Fig. 1

Cross section of the aortic wall and its remodeling during physiological versus pathological remodeling. The tunica media is separated from the tunica intima by the internal elastic lamina, and from the tunica adventitia by the external elastic lamina. Intermittent layers of elastic fiber and smooth muscle cells make up the tunica media, while this region is also rich in proteoglycans and glycoproteins. During physiological remodeling, degradation of ECM proteins by MMPs (and other proteases) is balanced by newly synthesized ECM proteins, while tissue inhibitors of proteases (TIMPs) keep the proteolytic activity of MMPs under check. Excess ECM degradation, or impaired ECM renewal synthesis, along with smooth muscle cell death, lead to adverse ECM remodeling as observed in aortic aneurysm

The ECM is a key constituent of the vascular wall. SMCs and the FBs in the aortic wall synthesize and organize a complex ECM structure that defines the passive mechanical behavior of the large, elastic arteries such as the aorta. Arterial ECM is primarily composed of elastin, collagen, proteoglycans (PGs), and glycoproteins. Elastic fibers and fibrillar collagen, comprising ~50% of the dry weight of larger arteries⁹, are the predominant ECM components in the aortic wall and almost entirely define the mechanical properties of the aorta. Elastic fibers provide the expandability and recoil properties, while fibrillar collagens (predominantly collagen type I and type III) are responsible for the tensile strength of the aortic wall to withstand the high pressure of blood pumped by the heart¹⁰. In addition to their mechanical properties of the aortic wall, elastic fibers and fibrillar collagens can directly interact with various integrins and other proteins to modulate the adhesion, proliferation, and migration of SMCs^11–14. GAGs consist of linear chains of repeating disaccharides and are highly negatively charged, which allow them to sequester water. Hydrated GAGs occupy a large volume of the extracellular space and cushion-compressive forces on cells. Among the four primary classes of GAGs, hyaluronan, chondroitin sulfate, heparan sulfate, and keratan sulfate, the first three are present in the aortic wall^15,16. PGs are proteins with covalently attached GAG chains that provide mechanical support and mediate signaling responses for cells in the aortic wall. Versican, which is rich in chondroitin sulfate, is among the principal proteoglycans present in the aorta¹⁷. Among small leucine-rich proteoglycans (SLRP), biglycan and decorin are reported to promote collagen fibrinogenesis in the aorta^18,19. Thrombospondins, tenascins, and fibronectin are also common glycoproteins present in the aorta^20,21.

Regional heterogeneity of the aortic wall

Morphology, structural composition, and mechanical properties of the aortic wall vary in the ascending and along the descending regions²² to ensure optimal mechanical operation²³. The tunica intima accounts for <1% of the total wall thickness and remains uniform along the aorta, whereas the thickness of tunica adventitia is negligible in the thoracic aorta but increases considerably along the abdominal aorta²⁴. The medial layer exhibits the greatest regional variations and diversitise along the aorta. The thickness of tunica media and the aortic lumen size steadily decrease from the ascending to the descending aorta, such that the wall thickness-to-lumen diameter ratio remains constant throughout the aorta. In the thoracic aorta, elastin occupies a higher proportion of the tunica media compared with that in the abdominal aorta. The amount of elastin decreases with distance from the heart, as less pulse dampening is required in the distal aorta. Meanwhile, the amount of collagen, which provides strength and limits stretch at high pressure, remains approximately constant with distance from the heart²⁵. Therefore, the elastin-to-collagen ratio is lower in the abdominal compared with the thoracic aorta. In addition, there is a linear relationship between the number of lamellar units and the tensional forces within the aortic wall, and the number of lamellar units decreases from the thoracic to the abdominal aorta¹⁰.

Physiological remodeling of the ECM

The ECM composition and integrity are key determinants of the physical characteristics of the aortic wall. Vascular ECM undergoes continuous physiological remodeling, whereby the existing ECM proteins are proteolytically degraded to be replaced by newly synthesized proteins. Several protease families are involved in this process. Matrix metalloproteinases (MMPs) are the most explored ECM-degrading proteinases²⁶, while other metalloproteinases such as ADAMs (a disintegrin and metalloproteinases), ADAM-TSs (ADAMs with thrombospondin motifs), or serine/cysteine proteases, including cathepsins and granzymes, can also contribute to this process.

MMPs are a family of zinc-containing calcium-dependent endopeptidases that are best known for their ECM-degrading function, although they also contribute to a number of other cellular events²⁶. To date, 23 human MMPs have been identified, which are classified into six groups based on their substrate specificities: collagenase, gelatinase, stromelysin, matrilysin, membrane-type MMPs, and others. MMPs are mostly secreted as inactive zymogens and are activated by proteolytic removal of their N-terminal pro-domain. The proteolytic activity of MMPs is tightly regulated in vivo by their endogenous inhibitors, tissue inhibitor of metalloproteinases (TIMPs)^27,28. Other MMP inhibitors, such as α2-macroglobulin, inhibit MMPs in the plasma and liquid phase by irreversibly removing them through endocytosis²⁹. The four members of the TIMP family (TIMP-1, −2, −3, and −4) can collectively inhibit all MMPs and a number of ADAMs and ADAM-TSs³⁰ through a covalent bond between their N terminus and the catalytic domain of the proteinase in a 1:1 stoichiometric ratio. In addition to inhibiting metalloproteinases, TIMPs play a number of MMP-independent functions^31–34. Sustained balance in proteolytic turnover of the ECM proteins is critical in maintaining the structural and functional integrity of the aortic wall. Cathepsins, a family of serine, cysteine, and aspartyl proteases are lysosomal proteases that also possess extracellular functions and can cleave multiple ECM components, including fibronectin, laminin, and elastin, and a number of non-ECM components³⁵. Granzymes are serine proteases that contribute to aortic remodeling and have been the subject of a recent review³⁶.

Aortic aneurysm

Aortic aneurysm is characterized by a localized irreversible dilatation of the aortic lumen by ≥50% of its original diameter. The dilatation is induced by aberrant and adverse remodeling of the aortic wall, including endothelial damage in the intima, SMC loss and ECM degradation in the media, and ECM degradation in the adventitia. Aortic aneurysm affects all three layers of the blood vessel, while any dilation of the aorta that is limited to the outer wall is considered a false aneurysm or pseudo aneurysm³⁷. Aortic aneurysm is asymptomatic, but can progress, gradually leading to aortic dissection or rupture, and as such is associated with critical morbidity and mortality in the aging population and accounts for 1–2% of total death in developed countries³⁸. Although TAA and AAA display similar physical appearances, they each demonstrate distinct etiologies, pathological features, epidemiology, and risk factors.

Abdominal aortic aneurysm (AAA)

AAA represents dilation of the aorta in any portion of the infra-diaphragmatic aorta, including the most common site of AAA formation, the infrarenal aorta. Depending on the degree of dilatation, AAA severity is classified as small (<55mm) or large (≥55mm). In the latter case, the patient will be considered eligible for surgical repair³⁹. AAA is the 12–15^th leading cause of death in the aged population, mostly males, over the age of 65 in many developed countries⁴. Almost 1–2% of all 65-year-old males have AAA, whereas 0.5% of women at 70 years of age are affected with AAA^40,41. The overall mortality due to AAA rupture has been reported to be 65–85%³⁸. However, AAA-related mortalities in both male and female populations are declining in developed countries⁴², with the United States and the United Kingdom showing the fastest decline in AAA mortalities in male (6.7–6.2% per year) and female (3.9–4% per year) populations. This was found to closely correlate with the decline in smoking prevalence in these countries. On the other hand, in many European countries, including Romania and Hungary, AAA mortalities have increased by 1.7–2.7% for males and 1–3.5% for females annually⁴².

AAA is a complex multifactorial disease that is affected by both environmental and genetic factors. AAA incidences depend on age and sex, with a higher prevalence in males over the age of 55 compared with females of the same age range (6:1)⁴³. However, AAA appears more detrimental in females, as they usually experience a higher risk of a small aneurysm rupture and less eligibility for standard surgical repair, because of the more complicated aneurysm morphology^44,45. Other risk factors include smoking, a family history of aneurysm⁴⁶, the presence of other diseases such as coronary heart disease, hyperlipidemia and atherosclerosis, hypertension, as well as acute or chronic infection^4,38. Tobacco smokers have 3–4-fold greater risk of developing AAA than nonsmokers^47,48. The link between hypertension and AAA is paradoxical, since hypertension has been shown to increase the risk of AAA by 30–40%; however, antihypertensive medication further increased the risk by 70–80%⁴⁷. More recently, a randomized placebo-controlled trial reported that antihypertensive medications did not impact AAA growth, despite effective lowering of BP⁴⁹. Interestingly, patients with diabetes mellitus have shown decreased prevalence or progression of AAA⁵⁰, which could be linked to the effects of antidiabetic medications. Hypercholesterolemia, increased body mass index, lack of exercise, and cardiovascular diseases were found to be associative factors for higher incidences of AAAs in Swedish male population⁵¹. Although atherosclerosis shows a close association with AAA, multiple studies have demonstrated the lack of a causal role for atherosclerosis in the initiation of AAA^52–54.

Thoracic aortic aneurysm (TAA)

TAA represents dilation of one or more aortic segments above the diaphragm, including the aortic root, ascending aorta, aortic arch, and descending aorta. Approximately 60% of TAA occurs in the aortic root or the ascending aorta, 10% in the aortic arch, 40% in the descending aorta, and 10% in the thoraco-abdominal aorta⁵⁵. About 13% of patients having aortic aneurysm develop multiple aneurysms, and ~20% of them can have a large TAA along with an AAA⁵⁶. The prevalence of TAA is much less than that of AAA^57,58, but the prognosis of TAA is more dire, and the rate of rupture-associated mortality is 2–3 times higher than that for AAA patients. While AAA is more commonly found in elders, TAA can occur at a young age due to the strong hereditary influence⁵⁹.

Similar to AAA, no causal relationship has been found between atherosclerosis and TAA⁶⁰, and it is relatively less associated with ascending TAA⁵⁵. TAA often results from genetic disorders, and is now referred to as syndromic or familial TAA. In a study of 135 non-syndromic TAA patients without Marfan syndrome (MFS), almost 19% of the patients were found to have a family history of TAA and developed the disease in relatively a younger age⁶¹. Moreover, patients with a family history of TAA were found to have a faster aneurysm growth rate (0.22cm/year) compared to patients with sporadic TAA (0.03cm/year) or patients with MFS (0.10cm/year)⁶¹. The specific genetic mutations associated with TAA are discussed later in this review. TAA has also been linked to inflammation and inflammatory diseases. Syphilis was once considered as one of the most common causes of ascending TAA⁵⁵, and although extremely rare nowadays, recent cases of syphilis aortitis and TAA have been reported⁶². Aortic arteritis, which includes Takayasu’s arteritis and giant-cell arteritis, refers to a group of chronic inflammatory diseases of unknown etiologies, and is associated with a high prevalence of TAA^63,64.

Recent clinical studies with large scale populations have identified fluoroquinolones, a commonly prescribed class of antibiotics as a risk factor for aortic aneurysm^65,66. Fluoroquinolone usage elevated the risk of aortic aneurysm or dissection by 2–3 folds compared with the nontreatment group^67,68, and by 2-fold compared with other antibiotics such as amoxicillin⁶⁹. Fluoroquinolones are believed to affect the synthesis and structural integrity of collagen in the aortic wall⁶⁸, or their chelating properties could affect collagen cross-linking and metalloprotease activities^70,71. In a mouse model of AAA, ciprofloxacin (a fluoroquinolone) treatment significantly reduced lysyl oxidase (LOX) expression and activity, and increased MMP expression and activity⁷². However, the exact biological mechanism of action for this class of antibiotics requires further investigation.

Experimental models of aortic aneurysm

Understanding the cellular and molecular event during the initiation, maturation, and rupture of human aortic aneurysm holds the key to identifying molecule(s) that are critical in the pathogenesis of this disease. However, aneurysmal aorta specimens from patients are only available at the advanced stages of the disease and therefore are not suitable for identifying the initiating factors. As such, a number of experimental models of aortic aneurysm have been developed and utilized on various genetically modified mice to understand the molecular mechanism(s) underlying aortic aneurysm (Table ​(Table1).1). Most of these experimental models share some of the features of human aortic aneurysm, although they all fall short of exactly mimicking the full clinical characteristics of this disease in humans^3,73. Nevertheless, these models have helped provide insight into understanding the disease pathogenesis and can be utilized for the possible discovery of novel therapies for aortic aneurysm.

Table 1

The impact of genetic alterations of ECM-related genes on aortic aneurysm

Mouse strain	Aneurysm model used	Phenotype	Ref.
Mmp2−/−	CaCl2-induced TAA, AAA	Protection against AAA, TAA	83,108
	Ang II infusion	Exacerbated aortic dilation, TAA	83
Apoe−/−/Mmp3−/−	Cholesterol-rich diet	Reduced aneurysm formation	110
Mmp9−/−	Intraluminal elastase-induced AAA	Protection against AAA	84
Mmp12−/−	CaCl2-induced AAA	Protection against AAA	97
Wild-type/Mmp14−/− bone marrow chimera	CaCl2-induced AAA	Reduced formation of AAA	109
Adamts1 transgenic/Apoe−/−	Cholesterol-rich diet and Ang II infusion	No changes observed	115
Adamts4−/−	High-fat diet and Ang II infusion	Reduced aneurysm, dissection, and aortic rupture	98
Adamts5Δcat (Truncated catalytic domain)	Ang II infusion	Increased dilatation of the ascending aorta	114
SMC- or EC-specific Adam17−/−	Adventitial elastase-induced TAA	Protection against TAA	88
SMC-specific Adam17−/−	Ang II+BAPN	Protection against TAA	116
Apoe−/−/Cats−/−	Ang II infusion	Protection against AAA	119
Catk−/−	Intraluminal elastase-induced AAA	Reduced AAA	120
Catl−/−	Intraluminal elastase-induced AAA	Reduced AAA	121
Apoe−/−/Plat−/−	High cholesterol diet	No changes observed	125
Apoe−/−/Plau−/−	High cholesterol diet	Protected against media destruction and aneurysm	125
Plg−/−	CaCl2-induced AAA	Reduced AAA	126
SMC-specific Pai1 overexpression	Aortic xenograft	Prevented AAA formation	127
Apoe−/−/Gzmb−/−	Ang II-induced AAA	Protection against AAA	129
Timp1−/−	Intraluminal elastase-induced AAA	Enhanced aneurysm	131
Timp1−/−	CaCl2-induced TAA	Enhanced aneurysm	132
SMC-specific Timp1 overexpression	Aortic xenograft AAA	Prevented AAA degeneration and rupture	133
Timp2−/−	CaCl2-induced AAA	Protection against AAA	137
Timp3−/−	Ang II infusion	Enhanced AAA	82
Timp3−/−/Mmp2−/−	Ang II infusion	Exacerbated AAA	82
Eln−/−	Spontaneous	Perinatal death	143
Eln+/-	Spontaneous	Supravalvular aortic stenosis	143
Fbln4−/−	Spontaneous	Perinatal death due to abolished elastogenesis	180
SMC-specific Fbln4−/−	Spontaneous	Spontaneous ascending and thoracic aortic aneurysm	181
Fbn1 mutant	Spontaneous	Arteriopathy, aneurysm, and dissection	182
Col1a1 Δ/ Δ (first intron deleted)	Spontaneous	Age-dependent aortic dissection and rupture	145
Col3a1+/−	Spontaneous	Sudden death due to TAA, dissection, and rupture	146
LoxMut/Mut	Spontaneous	Spontaneous death after birth due to aortic rupture	149
Bgn−/−	Spontaneous	Spontaneous aortic dissection and rupture	18
Apoe−/−/Dcn overexpression	Ang II infusion	Reduced AAA and rupture	19
SMC-specific Fbln4−/−/Thbs1−/−	Spontaneous	Attenuated TAA compared with Fbln4−/−	21
Thbs4−/−	Ang II infusion	Attenuated TAA	159

Open in a separate window

Fbln fibulin, Fbn fibrillin, Cat cathepsin, Bgn biglycan, Dcn decorin, Plat total plasminogen activator, Plau urokinase-plasminogen activator, Plg plasminogen, Pai plasminogen activator inhibitor, Gzmb granzyme B, Mmp matrix metalloproteinase, Apoe apolipoprotein E, Adam a disintegrin and metalloproteinase, Adamts ADAMs with thrombospondin motifs, Timp tissue inhibitor of metalloproteinases, Eln elastin, Col collagen, Lox lysyl oxidase, SMC smooth muscle cell, EC endothelial cell, Ang II angiotensin II

ECM-regulatory proteins in aortic aneurysm

Aortic aneurysm involves dramatic changes in the structural components of the vasculature, due to SMC dysfunction, imbalance between proteinases and their inhibitors, and abnormal synthesis, accumulation, or degradation of ECM proteins³. Among different classes of proteases, metalloproteinases (MMPs, ADAMs, and ADAM-TSs)^83,88,98,99, and cathepsins^100,101 are important factors for pathogenesis of aneurysm, because of their involvement in cleavage of different ECM and non-ECM proteins. The availability of genetically modified mouse models has paved the way to understanding the role of the ECM and its remodeling, the importance of protease and their inhibitors, and the contribution of different proteoglycans and glycoproteins in aortic aneurysm.

Matrix metalloproteinases

Multiple MMPs contribute to physiological aortic remodeling and their increased expression are well documented in both human and mouse aortic aneurysms¹⁰². Under physiological conditions, vascular SMCs (VSMC) are the major source of MMPs production in the aorta, which is amplified under inflammatory conditions leading to augmented expression of MMPs¹⁰³. Inflammatory cells, such as neutrophils and macrophages, also contribute to MMP production which further promote ECM degradation and aneurysm formation. Neutrophils can produce MMP8 and MMP9^104,105, whereas macrophages can secrete a larger number of MMPs, including MMP1, MMP3, MMP7, MMP9, and MMP12¹⁰⁶. These MMPs can cleave collagen (MMP1, MMP2, MMP9, MMP13, and MT1-MMP¹⁰⁷), elastin (MMP2, MMP9, and MMP12), proteoglycans and basement membrane proteins, leading to the degradation of aortic ECM and aneurysm formation¹⁰³.

Studies on MMP-deficient mice have provided us with important insight into the role of specific MMPs in different models of aortic aneurysm. Mmp2-deficient mice showed reduced aortic aneurysm formations in CaCl₂-induced TAA and AAA models^83,108. However, in response to Ang II infusion, Mmp2-deficient mice developed severe TAA but not AAA⁸³. The absence of MMP9 in mice attenuated AAA formations⁸⁴; however, the protection was reversed by introducing wild-type macrophages in Mmp9^−/− mice, but not in Mmp2^−/− mice, highlighting the role of macrophage MMP9 and stromal MMP2 in the pathogenesis of AAA¹⁰⁸. Even in the absence of MMP2 or MMP9, MT1-MMP can be a key regulatory factor to induce aortic aneurysm by promoting macrophage-dependent elastolytic activity in CaCl₂-induced AAA¹⁰⁹. Apoe^−/−/Mmp3^−/− mice showed reduced aortic aneurysm formation despite persistence of atherosclerotic plaques compared with Apoe^−/− mice¹¹⁰. Loss of MMP12, also known as macrophage metalloelastase, attenuated aneurysm formation in CaCl₂-induced AAA model⁹⁷.

Although there is significant evidence that MMPs are upregulated in AAA and TAA, and MMP-deficiency often results in protection against aneurysm formation, MMP inhibition may not be an effective approach in treating aortic aneurysm. Consistently, broad-spectrum MMP inhibitors, such as doxycycline, were reported to have promising effects in treating small AAAs¹¹¹. However, recent randomized trials reported no changes in AAA growth or in reducing the need for AAA repair or time to repair in patients receiving doxycycline¹¹².

Disintegrin and metalloproteinases (ADAMs, ADAM-TSs)

A number of ADAM-TSs, including ADAM-TS4, ADAM-TS5, and ADAM-TS7, have been reported to be involved in TAA in humans and in mouse models^98,113,114. Mice expressing truncated ADAM-TS5 (lacking the catalytic domain) showed increased aortic dilation in Ang II-induced TAA compared to wild-type mice¹¹⁴. ADAM-TS4 deficiency reduced aortic dilation, aneurysm formation, dissection and aortic rupture in a mouse model of sporadic aortic aneurysm and dissection (AAD) induced by HFD and Ang II infusion⁹⁸. However, ADAM-TS1 overexpressing Apoe^−/− mice subjected to HFD and Ang II infusion did not show elevated aneurysm response compared to Apoe^−/− mice¹¹⁵.

Compared with ADAM-TSs, the role of ADAMs in aortic aneurysm is less explored. The most studied ADAM, ADAM17, was found to be critically involved in aortic aneurysm development in both humans and mice⁸⁸. SMC- or EC-specific deletion of ADAM17 protected the aorta from elastase-induced TAA formation⁸⁸, whereas ADAM17 deficiency in SMCs provided protection against AAA triggered by Ang II+BAPN treatment¹¹⁶. ADAM10 is an important regulator for vascular remodelling and inhibition of ADAM10 mRNA with miR-103a prevented AAA formation in Apoe^−/− mice following HFD and Ang II treatment¹¹⁷.

Cathepsins

Cathepsins are a group of serine, aspartic, and cysteine proteases, which contribute to vascular remodelling. Several cathepsins, including cathepsins A, D, S, K, and L are reported to be involved or upregulated in humans aortic aneurysmal tissues^100,101,118. Cathepsin K and S are potent elastases that are produced by SMCs and macrophages in the aorta under inflammatory milieus. Deficiency of cathepsin S attenuated Ang II-induced AAA formation in Apoe^−/− mice¹¹⁹. Cathepsin K has been reported to contribute to CD4⁺ T-cell proliferation and SMC apoptosis since its deficiency attenuated the AAA in the intraluminal elastase infusion model in mice¹²⁰. Cathepsin L knockout mice exhibited reduced lesions, in vitro macrophage and T-cell transmigration, and angiogenic responses upon intraluminal elastase-induced AAA development¹²¹.

Other proteases

There are several other proteases that are important in aortic aneurysm formation. Chymase and plasminogen activators are serine proteases that are involved in activation of other proteases, including MMPs. Chymases, secreted mostly by mast cells, activate MMP9 in human AAAs¹²². In addition to converting plasminogen into plasmin, plasminogen activators can directly degrade ECM components, activate collagenases, and degranulate neutrophils to release MMPs and elastases^123,124. Deficiency of urokinase-plasminogen activator, but not tissue-plasminogen activator suppressed aortic aneurysm formation in hypercholesteremic Apoe^−/− mice¹²⁵. Plasminogen deficiency also attenuated CaCl₂-induced AAA by targeting macrophage migration and MMP9 activation¹²⁶. Alternatively, local overexpression of plasminogen activator inhibitor-1 (PAI-1) prevented aneurysm development and rupture in mice¹²⁷. Granzyme B is a proapoptotic serine protease that is abundantly expressed in advanced human atherosclerotic lesions and may contribute to plaque instability¹²⁸. Granzyme B contributes to AAA through an extracellular, perforin-independent mechanism involving ECM cleavage. The absence of Granzyme B decreased AAA formation and increase survival of Apoe^−/− mice upon Ang II infusion¹²⁹.

Tissue inhibitors of metalloproteinase (TIMP)

TIMPs regulate the proteolytic activities in the vascular wall by directly binding to and inhibiting different groups of metalloproteinases, including MMPs, ADAMs, and ADAM-TSs. Although there is a high degree of structural homology among the four members of the TIMP family, they exhibit varying substrate affinities, inhibitory efficiencies, and differential responses to pathological stimuli¹³⁰. TIMP1 can inhibit most members of the MMP family except the membrane-type MMPs (MT-MMPs) and MMP19, and few ADAMs, such as ADAM10¹³⁰. TIMP1 expression is reduced in aneurysmal tissues, and the absence of TIMP1 resulted in severe aneurysm in the intraluminal elastase-perfused AAA¹³¹ and CaCl₂-induced TAA mouse models¹³². Local overexpression of TIMP1 prevented aortic aneurysm degeneration and rupture in a rat model¹³³. TIMP2 which acts as an inhibitor for many MMPs¹³⁴ and ADAM12¹³⁵, also plays an important role in MMP2 activation in coordination with MT1-MMP¹³⁶. TIMP2 deficiency resulted in protection against CaCl₂-induced AAA in mice¹³⁷, perhaps by inhibiting MMP2 activation. TIMP3 has the broadest range of substrates among TIMPs which includes most of the MMPs, many ADAMs (ADAM −10, −12, −17, −19, and −33) and ADAMTSs (ADAMTS-1, −2, −4, and −5)²⁸. Moreover, TIMP3 is distinct from other TIMPs in being ECM-bound¹³⁸. Increased expression of TIMP3 in aneurysmal aortic tissue suggests its protective role against the disease¹³⁹, as the absence of TIMP3 promoted aortic dilation and aneurysm in Ang II-induced (non-hyperlipidemic) AAA⁸². TIMP4 can inhibit a number of MMPs and ADAM28¹⁴⁰. Interestingly, the N-terminal domain of TIMP4, but not the full-length protein, can inhibit ADAM17¹⁴¹. The role of TIMP4 in aortic aneurysm is less explored. Increased TIMP4 expression has been reported in hyperhomocysteinemia-associated aortic aneurysm in humans and mice¹⁴².

ECM proteins that are central to aneurysm formation

Conclusion/future prospects

In conclusion, aortic aneurysm remains a serious health concern as its rupture can cause significant morbidity and mortality. It is well acknowledged that TAA is a distinct disease from AAA and should be treated as such. Research to date suggests that given the regional heterogeneity of the aortic structure and ECM, events that disrupt ECM synthesis or ECM protein assembly could underlie the formation of aneurysm in the thoracic area, whereas factors that trigger enhanced proteolytic degradation of the ECM contribute to AAA formation. This is consistent with the notion that genetic disorders that disrupt collagen or elastin production (or assembly) are associated mostly with TAA and to a smaller extent with AAA. While studies on genetically modified mice have revealed a number of potential molecular and cellular mechanisms that may contribute to TAA or AAA, the lack of availability of an effective pharmacological treatment indicates that we may not yet have the full picture of all the factors that contribute to aortic aneurysm. Identifying the initiating factors in aneurysm formation is key in developing a treatment strategy, since at later stages, to repair the severe damage on the aortic wall if possible, would require effective replenishment of ECM-producing cells and ensuring optimal assembly and organization of the newly synthesized ECM proteins.

Acknowledgements

Authors’ contributions

S.J. and Z.K. planned the outline of the article. S.J. and M.H. prepared the first draft, M.S. performed critical editing of the article. S.J. and Z.K. prepared the tables and the figure, revised, and finalized the article.

Conflict of interest

The authors declare that they have no conflict of interest.

Footnotes

References