- Email: [email protected]
Extracellular matrix remodeling in the vascular wall
Pathol Biol 2001 ; 49 : 326-32
2001 Éditions scientifiques et médicales Elsevier SAS. Tous droits réservés S0369-8114(01)00151-1/REV
Revue
Extracellular matrix remodeling in the vascular wall M.P. Jacob 1∗ , C. Badier-Commander 1 , V. Fontaine 1 , Y. Benazzoug 2 , L. Feldman 1 , J.B. Michel 1 1 INSERM U 460, UFR de médecine Xavier Bichat, 16, rue Henri Huchard 75870 Paris cedex 18, France ; 2 laboratoire de la matrice extracellulaire, faculté des sciences biologiques, USTHB, BP n◦ 32, Bab Ezzouar,
16111 Alger, Algérie Summary The extracellular matrix provides a structural framework essential for the functional properties of vessel walls. The three dimensional organization of the extracellular matrix molecules — elastin, collagens, proteoglycans and structural glycoproteins — synthesized during fetal development — is optimal for these functions. Early in life, the vessel wall is subjected to injury: lipid deposition, hypoxia, enzyme secretion and reactive oxygen species production during inflammatory processes, and the extracellular matrix molecules are hydrolyzed by proteases — matrix metalloproteinases, leukocyte elastase, etc. In uninjured arteries and veins, some proteases are constitutively expressed, but through the control of their activation and/or their inhibition by inhibitors, these proteases have a very low activity. During the occurrence of vascular pathologies — atherosclerosis, hypertension, varicosis, restenosis, etc. — the balance between proteases and their inhibitors is temporally destroyed through the induction of matrix metalloproteinase gene expression or the secretion of enzymes by inflammatory cells. Smooth muscle cells, the most numerous cells in vascular walls, have a high ability to respond to injury through their ability to synthesize extracellular matrix molecules and protease inhibitors. However, the three dimensional organization of the newly synthesized extracellular matrix is never functionally optimal. In some other pathologies — aneurysm — the injury overcomes the responsive capacity of smooth muscle cells and the quantity of extracellular matrix decreases. In conclusion, care should be taken to maintain the vascular extracellular matrix reserve and any therapeutic manipulation of the protease/inhibitor balance must be perfectly controlled, because an accumulation of abnormal extracellular matrix may have unforeseen adverse effects. 2001 Éditions scientifiques et médicales Elsevier SAS elastases / elastin / collagens / matrix metalloproteinases / vascular wall
Résumé – Le rôle de la matrice extracellulaire dans le remodelage de la paroi vasculaire. La matrice extracellulaire (MEC) confère aux parois vasculaires une structure de soutien indispensable à leurs propriétés fonctionnelles. Les différentes molécules de la MEC — élastine, collagènes protéoglycannes, glycoprotéines de structure — s’organisent de façon optimale dans l’espace extracellulaire au fur et à mesure de leur synthèse au cours du développement et de la croissance de l’individu. Très tôt, la paroi vasculaire est soumise à différentes agressions et les macromolécules de la MEC sont hydrolysées par des protéases. Dans les parois vasculaires normales, la métalloprotéase matricielle (MMP) 2 est exprimée de manière constitutive mais son activité enzymatique reste très faible car elle reste sous forme proenzyme et des inhibiteurs, les inhibiteurs tissulaires des MMPs, sont présents. Au cours des diverses pathologies vasculaires, l’équilibre protéases/inhibiteurs est temporairement détruit de par l’expression et l’activation des MMPs par les cellules vasculaires elles-mêmes ou la sécrétion d’enzymes par les cellules inflammatoires. Par leur capacité de synthèse des différentes macromolécules de la MEC et des inhibiteurs des MMPs, les cellules musculaires lisses ont une forte capacité à répondre à ces agressions. La structure tridimensionnelle de la MEC néosynthétisée n’est toutefois jamais aussi optimale que celle synthétisée au cours du développement. Dans d’autres pathologies vasculaires — anévrysmes —, l’hydrolyse enzymatique des macromolécules dépasse la
∗ Correspondence and reprints.
E-mail address: [email protected] (M.P. Jacob).
Extracellular matrix remodeling in the vascular wall
327
capacité de réponse des cellules musculaires lisses et la quantité de MEC diminue. Il est donc indispensable de préserver au maximun la MEC des parois vasculaires, l’utilisation d’inhibiteurs de protéases devant être parfaitement contrôlée et dans le temps et dans l’espace pour éviter l’effet secondaire qui peut être lui-même néfaste, l’accumulation de MEC non fonctionnelle. 2001 Éditions scientifiques et médicales Elsevier SAS élastases / élastine / collagènes / métalloprotéinases matricielles / paroi vasculaire
The extracellular matrix (ECM) in the vascular wall provides an essential supporting, structural framework necessary for the structural and functional properties of vessel walls (elasticity, resistance to stretch). Besides the two main structural macromolecules, elastin and collagens, proteoglycans and structural glycoproteins play essential roles in other functional requirements of vessels (hydration, ion filtration, interactions between ECM molecules, growth factor biodisponibility, cell-matrix interactions). These ECM macromolecules are synthesized by the three vascular cell types: intimal endothelial cells, medial smooth muscle cells (SMC) and adventitial fibroblasts. The three dimensional organization of these ECM molecules synthesized during fetal development is optimal for the functions of the vessel wall. But early in life, the vessel wall is subjected to injury (lipid deposition, hypoxia, enzyme secretion and reactive oxygen species production during inflammatory processes, etc.). One of the main lessons of our studies is that smooth muscle cells have an extremely high capacity to respond to injuries via their capacity to synthesize ECM molecules and enzyme inhibitors. However, the three dimensional organization of the newly synthesized ECM is never functionally optimal as is that synthesized during development. This response to injury occurs in several vascular pathologies (atherosclerosis, hypertension, varicosis, etc.). In some other pathologies, the injury overcomes the responsive capacity of smooth muscle cells (aneurysm, etc.). This review summarizes the results of our recent studies on varicosis and hypertension, and of previous studies on aneurysms and restenosis, after a brief description of the ECM molecules of the vascular wall and their degradation by proteinases.
THE ECM IN THE VASCULAR WALL Collagens and elastin are the main protein constituents of vessels. The biomechanical properties of vessels, particularly of the major arteries and veins, are largely dependent on the absolute and relative quantities of these two constituents [1]. Among the 19 different collagen types described until now, collagen types I and III are the major fibrillar collagens detectable in vessels, representing 60% and 30% of vascular collagens, respectively [2 – 4]. The remaining 10% include fibrillar collagen type V, Fibril-Associated Collagens with Interrupted Triple helix (FACIT) types XII
and XIV, microfibrillar collagen type VI, basement membrane collagen type IV and collagen type VIII. The basic unit of fibrillar collagens is the 300 nm triple helix, with each chain comprising about 330 repeats of the amino acid sequence -Gly-X-Y- (where X is often proline and Y is often hydroxyproline). After secretion in the extracellular compartment and removal of N- and C-propeptides, collagen triple helices aggregate in quarter-staggered fibrils. Newly synthesized collagen fibrils are soluble in salt solutions and dilute acid and have no tensile strength. During the formation of intermolecular cross-linking, collagen fibers become increasingly insoluble, more refractory to the action of enzymes and show a progressive increase in tensile strength. The cross-linking process is initiated by the enzyme lysyl oxydase, through the oxidation of specific lysine or hydroxylysine residues in the telopeptide regions. The resulting aldehydes undergo a series of reactions with adjacent reactive residues to give both inter- and intramolecular crosslinks [5]. Elastin is the most abundant protein of the large arteries that are subjected to a large pulsatile pressure generated by cardiac contraction [6 – 8]. However, elastin is also detectable in resistance arteries — mainly in the internal and external elastic lamina — and veins. Elastin represents 90% of the elastic fibers, the other constituents being microfibrillar glycoproteins such as fibrillins and MAGPs (Microfibrillar-Associated GlycoProteins). The precursor of elastin, tropoelastin, is a highly hydrophobic protein (MW, 72 kDa), which is soluble in salt solution like the collagen triple helix. In contrast, elastin is an insoluble protein. This insolubility results from the crosslinking process between lysine residues. Crosslinking of tropoelastin molecules begins with the oxidative deamination of some lysine residues by lysyl oxidase, as previously described for collagen crosslinking. The spontaneous condensation reaction between four lysine/allysine residues leads to the formation of the specific cross-links for elastin, desmosine and isodesmosine. Cross-links resulting from the condensation of two or three lysine/allysine residues are also detectable in elastin [5]. This crosslinking process confers to elastin its function — i.e. elasticity —, essential in large arteries which distend during systole and recoil during diastole. Another property of elastin has been recently demonstrated by the study of patients suffering from supravalvular aortic stenosis and of knock-out mice for elastin: elastin controls, directly or indirectly, the proliferation and phenotype of smooth muscle cells [9 – 11].
328
M.P. Jacob et al.
Several structural glycoproteins are closely associated with elastin in elastic fibers: MAGPs and fibrillins [12 – 14]. Fibrillin 1 has been the focus of important research since mutations in the fibrillin 1 gene were found in patients suffering from Marfan’s syndrome [15, 16]. Fibrillins 1 and 2, 350 kDa glycoproteins, are characterized by the multiple repeats of EGF-like domains, in which the three cystine bridges stabilize the β-turns within the molecule. Several other structural glycoproteins play an essential role in the structure and function of the ECM in the arterial wall, including fibronectin, vitronectin, laminin, entactin/nidogen, tenascin and thrombospondin. These glycoproteins have a multidomain structure, potentially enabling simultaneous interactions between cells and other ECM components [17, 18]. The proteoglycans contained in the vascular wall are the large aggregating proteoglycans aggrecan and versican, the small non-aggregating interstitial proteoglycans biglycan, decorin and fibromodulin and the cell-associated proteoglycans syndecan, fibroglycan and glypican. Proteoglycans are proteins that have one or more attached glycosaminoglycan chains [19, 20]. They contain distinct protein and carbohydrate domain structures, which interact with other ECM molecules. They participate in ECM assembly and confer specific properties to the tissues (hydration, filtration, etc.). Proteoglycans also regulate various cellular activities (proliferation, differentiation, adhesion, migration) and control cytokine biodisponibility and stability (FGF, TGF-β, etc.) [21].
THE PROTEINASES/INHIBITORS INVOLVED IN VASCULAR ECM REMODELING The metabolic turnover of mature collagen and elastic fibers in adult animals is relatively slow [22]. Although only small amounts of these proteins are degraded normally, increased degradation and fragmentation of collagen and elastin fibers are observed in vascular diseases. Collagenolytic and elastinolytic enzymes are found in a number of mammalian cells and tissues including polymorphonuclear neutrophils and monocytes/macrophages. Collagenolytic enzymes are mainly matrix metalloproteinases. Elastinolytic enzymes are found in all the four classes of proteinases (serine, metallo-, aspartic and cysteine proteinases) and react with a broad spectrum of substrates [23, 24]. All elastases studied to date have catalytic activity against protein and peptide substrates other than elastin. Matrix metalloproteinases and other elastolytic enzymes have been described extensively in previous reviews [25 – 29]. We will focus only on the serine and metalloproteinases which participate to vascular ECM remodeling, i.e., proteinases synthesized in vascular cells themselves and those secreted by inflammatory cells (table I).
Table I. Serine and metallo-proteinases active on extracellular matrix macromolecules, detectable in vascular cells and inflammatory cells. Proteinases produced by vascular cells
Proteinases produced by inflammatory cells
Matrix metalloproteinases proMMP-1* (EC) proMMP-2 (EC, SMC, fibroblasts) proMMP-3* proMMP-7 proMMP-9*
Matrix metalloproteinases proMMP-1 (monocytes) proMMP-2 proMMP-7 (monocytes) proMMP-8 (PMN) proMMP-9 (macrophages, PMN) proMMP-12 (macrophages) Serine proteinases Leukocyte elastase (PMN)
Serine proteinases SMC elastase
* enzymes not expressed in basal conditions. EC, endothelial cells; MMP, matrix metalloproteinases; PMN, polymorphonuclear neutrophils; SMC, smooth muscle cells.
The two serine proteinases involved during vascular remodeling are the SMC elastase [30] and the leukocyte elastase which is stored in the azurophil granules of polymorphonuclear neutrophils [31]. One inhibitor of SMC elastase is elafin and leukocyte elastase is inhibited by α1-proteinase inhibitor, α2-macroglobulin and elafin. Matrix metalloproteinases (MMPs) play a fundamental role in the degradation of vascular ECM. MMPs include collagenases (MMP-1, expressed by endothelial cells, MMP8, stored in specific granules of PMN), gelatinases (MMP-2, constitutively expressed by vascular cells; MMP-9, expressed in macrophages and PMN, inducible in vascular cells), ‘elastases’ (MMP-7, expressed at a low level by the vascular wall, MMP-12, synthesized during the differentiation of monocytes into macrophages), stromelysins and membrane-type metalloproteinases (MT-MMPs). The proteolytic activity of each MMP is tightly regulated at three levels: first, gene expression and protein secretion levels; second, activation of the inactive pro-enzyme, and, third, inhibition by the TIMPs (Tissue Inhibitors of Matrix Metalloproteinases) or other inhibitors (α2 -macroglobulin). The activation of secreted pro-MMPs requires the disruption of the Cys-Zn2+ (cysteine switch) interaction and the removal of the propeptide. Pro-MMPs can be activated in vitro by SH-reactive agents, mercurial compounds, SDS or chaotropic agents. In vivo, pro-MMPs are activated by tissue or plasma proteinases — plasmin, thrombin, other MMPs or MTMMPs — and reactive oxygen species. The activation of MT-MMPs occurs intracellularly by furin. According to our experiments [32, 33], the activation of MMPs and their inhibition by TIMPs are the main regulatory mechanisms of MMP activities in the vascular wall. For example, proMMP-2 is constitutively expressed in the normal arterial wall, in which no SMC migration or elastin degradation are detectable. The very
Extracellular matrix remodeling in the vascular wall
329
well controlled balance between active proteinases and inhibitors is perturbed during pathological processes, particularly when PMNs or macrophages are present. However, through their capacity to synthesize enzyme inhibitors, smooth muscle cells have an extremely high capacity to respond to these increased enzyme levels.
PAI amount, were lower in varicose veins and the ratio PAI/uPA + tPA was slightly lower in varicose veins than in control veins [32].
ECM REMODELING IN VARICOSE VEINS
The ECM content (collagen + elastin) of arteries is increased in hypertensive patients and in genetic or experimentally induced hypertension in animals. The increase in collagen and elastin synthesis induced by the increased blood pressure has been carefully studied by Keeley et al. [35 – 37]. The collagen and elastin synthesis returns to basal levels as soon as the blood pressure stops rising. While the absolute aortic contents of elastin and collagen are increased in hypertensive animals, the elastin/collagen ratios are the same in normotensive and hypertensive animals. These increased collagen and elastin contents in hypertensive animals may also result from reduced degradation. The MMP/TIMP balance has been studied in the rat model of hypertension induced by chronic blockade of NO production by Nω-nitro-L-arginine methyl ester [33]. ProMMP-9 was not detected in the aortic extracts of normotensive and hypertensive rats. ProMMP-2 activity, as well as TIMP-2 inhibitory capacity, were not significantly modified in the aortas of hypertensive rats. In contrast, TIMP-1 inhibitory capacity, measured by reverse zymography, and TIMP-1 mRNA levels were significantly increased in hypertensive rats. These data suggest that hypertension is associated with the overexpression of both pro-inflammatory molecules (Interleukin-6, Monocyte Chemoattractant Protein-1 and Macrophage ColonyStimulating Factor) and resistance factors such as TIMPs. TIMP-1, by inhibiting MMP activity, limits inflammatory cell migration into the media and leads to subsequent increase in ECM content. Interestingly, it was recently demonstrated that serum concentration of free TIMP-1 is increased, while serum concentration of MMP-1 is diminished, in patients with essential hypertension. Furthermore, free TIMP-1 concentration was decreased, and free MMP-1 concentration was increased, in the sera of the same patients after one year of treatment with an angiotensin-converting enzyme inhibitor [38]. Taken together, these studies demonstrate that the synthesis of ECM is enhanced and its degradation reduced in the arteries of hypertensive individuals.
Primary varicose veins are functionally characterized by venous back-flow and increased blood stasis in the upright position. Dilatation and tortuosity provide evidence for progressive venous wall remodeling, with disturbance of smooth muscle cell/ECM organization. Affected areas are not uniformly distributed, some areas being hypertrophic, whereas others are atrophic or unaffected. When the proteolytic MMP/TIMP balance, which may participate in the remodeling of the venous wall, was investigated, a higher TIMP-1 level, an equal quantity of TIMP-2, and lower MMP-2 level and activity (based on soluble protein content) were found in varicose veins compared to control veins [32]. Consequently, the protein ratios TIMP-1/MMP-2 and TIMP-2/MMP-2 were 3.6- and 2.1-fold higher in varicose veins compared to control veins, respectively. ProMMP-9 was detected in variable amounts, both in control and varicose veins, but the mean activity was lower in varicose veins. The lower level of proMMP-2, constitutively expressed by smooth muscle cells and endothelial cells, probably reflects a lower production of proMMP-2 by cells of the varicose veins. In contrast, the increase in TIMP-1, detectable in both endothelial cells and smooth muscle cells, indicates a higher expression of the TIMP-1 gene by vascular cells. Indeed, depending on the cell type, TIMP-1 expression can be modulated by various cytokines (Tumor Necrosis Factor-α, Interleukin-1β, Platelet-Derived Growth Factor, Transforming growth factor-β [TGF-β]) and soluble factors (angiotensin II, Phorbol Myristate Acetate, retinoids). All these molecules have not been measured in varicose veins, but an increase in the total TGF-β1 concentration has been reported [34]. In turn, TGF-β1 downregulates MMPs and upregulates TIMP-1. This stimulatory effect of TGF-β1 on TIMP-1 expression is additive to the stimulation of collagen and elastin synthesis by TGF-β. Hypoxia is also a potential candidate as a modulator of TIMP-1 expression. This imbalance between MMP and TIMP production, in favor of an antiproteolytic activity, may explain, at least in part, the accumulation of ECM in varicose veins. As plasmin activates some proMMPs and degrades some ECM molecules, the components of the fibrinolytic system, including tissue-type (tPA) and urokinase-type (uPA) plasminogen activators (PAs) and plasminogen activator inhibitors (PAIs) were also measured in vein extracts. The uPA and tPA activities, as well as the
ECM REMODELING DURING HYPERTENSION
ECM REMODELING DURING RESTENOSIS Percutaneous transluminal coronary angioplasty has become a widely used treatment for occlusive coronary lesions. During angioplasty, the atheromatous plaque is
330
M.P. Jacob et al.
fissured and compressed into the vessel wall using a balloon catheter. However, immediate benefit of angioplasty is hampered by the risk of recurrent stenosis (or restenosis), which occurs in 20–40% of the patients within six months [39]. Restenosis is due to the migration and proliferation of smooth muscle cells from the media to the intima, producing a characteristic lesion of intimal fibro-cellular hyperplasia. Restenosis represents an exaggeration of the normal healing process since intimal hyperplasia is present to some degree in all patients after angioplasty. Using the rat model of balloon injury, the induction of proMMP-9 one day after injury and an increase in proMMP-2 at four to five days after angioplasty were demonstrated [40]. The activity of tissue-type plasminogen activator was also increased at four days after arterial injury [41]. The pig shows the same pattern of increased MMP-9 and MMP-2 activities after arterial injury, with maximal activities at seven days and persistantly elevated activities for up to 21 days [42]. In the normocholesterolemic rabbit, MMP-2 mRNA levels increase after balloon angioplasty but MMP-9 remains undetectable [43, 44]. Only the combination of balloon injury with low flow or an atherogenic diet results in substantial proMMP-9 activity, in addition to the increased MMP-2 activity. The relevance of these increased MMP activities to SMC migration from the media to the intima was confirmed by the demonstration of a 97% decrease in the number of SMC that migrated into the intima 4 days after carotid artery injury in rats receiving an MMP inhibitor [45]. However, an increased rate of SMC replication in these treated rats permitted the growth of the neointima to ‘catch up’, thus contributing to the long-term inefficacity of the MMP inhibitor on neointimal thickening observed after arterial injury. Local delivery of genes encoding for TIMP-1 [46] or TIMP-2 [47] partially inhibits or delays neointimal growth in the same model.
ECM REMODELING IN ANEURYSMS Human abdominal aortic aneurysms are characterized by progressive dilatation that may lead to the rupture of the aortic wall [48]. Mutations in genes encoding type III collagen and fibrillin-1 are responsible for several forms of inheritable aneurysmal-like disease with vascular rupture, such as Ehlers-Danlos type IV and Marfan’s syndrome [11]. However, the great majority of abdominal aneurysms develop as acquired degenerative lesions in the absence of known ECM genetic defects [49]. Associations between aneurysms and aging, atherosclerosis, male gender, as well as smoking and inflammatory reactions have been established. The ECM remodeling in aneurysmal aortas is characterized by the structural disorganization and disappearance of the elastic lamella
in association with an inflammatory, fibrotic adventitia. The inflammatory process is characterized by infiltration of the outer aorta by mononuclear phagocytes and lymphocytes. Another biological phenomenon observed during the occurrence of abdominal aortic aneurysms is the invasion of media by capillaries; hence, proliferative endothelial cells are detectable in aneurysmal aortas. ECM degrading enzymes secreted by both invading and resident cell types are characteristic of all these processes. In fact, increased proteolytic activities have been described in aneurysmal aortas, including serine elastases, MMPs and plasminogen activators, these proteinases being directly or indirectly involved in aortic elastin degradation. The serine elastase involved in aortic aneurysms is neutrophil elastase. The main MMPs detected in aneurysmal aortas have elastolytic activities: MMP-2, constitutively expressed by vascular cells in normal aortas, MMP-9 and MMP-12, expressed in particular by tissue macrophages present in the outer aneurysmal aorta [50, 51]. The quantities of TIMP-1 and TIMP-2 in aneurysmal tissues have been measured at the mRNA and protein levels, and are significantly increased in pathological tissues [50, 52]. In order to evaluate the net proteolytic activity in aneurysmal and occlusive aortas, in situ zymography was used. No difference in the net proteolytic activity was detected between aneurysmal and occlusive aortas and this proteolytic activity was detected in the luminal portion of the specimens [53]. This result is quite surprising since the main gelatinolytic enzyme in aneurysmal aortas, MMP-9, has been reported to colocalize with macrophages and is spatially associated with periadventitial vascularization and lymphocytic infiltration [50]. Finally, the well described decrease in medial SMC density in aneurysmal aortas compared to normal or occlusive aortas may explain the lower production of MMP inhibitors relative to proteinases in aneurysmal aortas [54]. Indeed, the local seeding of SMC, whether they overexpress TIMP-1 or not, prevents elastin depletion, aneurysm formation and rupture in a rat model of aneurysm [55]. The blockade of MMP activation also prevents both early aneurysmal changes and the incidence of induced aneurysms in rat models of aortic and cerebral aneurysms [56, 57].
CONCLUSION In uninjured arteries and veins, some MMPs are constitutively expressed (table I). But through the control of their activation and their inhibition by inhibitors synthesized by the vascular cells themselves (TIMPs), these proteases have a very low activity. During the occurrence of vascular pathologies, this balance between proteases and their inhibitors is temporarily destroyed through the induction of MMP gene
331
Extracellular matrix remodeling in the vascular wall
ACKNOWLEDGEMENTS The authors thank Dr Mary Osborne-Pellegrin for the editing of this manuscript. Our studies were supported by grants from the Institut National de la Santé et de la Recherche Médicale, the Institut de recherches international Servier (varicosis) and Menarini, Florence, Italy (hypertension).
ABBREVIATIONS ECM, extracellular matrix; MMP, matrix metalloproteinase; TIMP, tissue inhibitor of matrix metalloproteinase, tPA, tissue-type plasminogen activator; uPA, urokinase-type plasminogen activator; PAI, Plasminogen activator inhibitor; PMN, polymorphonuclear neutrophils; SMC, smooth muscle cells.
REFERENCES
Figure 1. Extracellular matrix homeostasis in the vascular wall. α1-AT, α1-proteinase inhibitor; α2-M, α2-macroglobulin; EC, endothelial cells; ECM, extracellular matrix; EEL, external elastic lamina; IEL, internal elastic lamina; PGs, proteoglycans; PMN, polymorphonuclear neutrophils; SMC, smooth muscle cells; TIMPs, tissue inhibitors of metalloproteinases.
expression in vascular cells or the secretion of enzymes by inflammatory cells. If the protease levels exceed only for a short time the inhibitor level, this may then be overcompensated and the normal healing and repair process would lead to the deposition of newly synthesized ECM components and the arrest/inhibition of SMC proliferation (figure 1). However, during the long-term evolution of the principal vascular pathologies (atherosclerosis, aneurysm) a minimal imbalance between degradation and synthesis of ECM components could lead either to stenosis or to dilatation of the vessel wall. Furthermore, depending on the extent of the degradation and synthesis of each ECM component, the relative levels of each component, and especially the elastin/collagen ratio, may be modified. In addition, the three-dimensional organization of this newly synthesized ECM, and consequently its functional properties, are never optimal as are those of ECM synthesized during development. In conclusion care should be taken to maintain the vascular ECM reserve and any therapeutic manipulation of the protease/inhibitor balance must be perfectly controlled, because an accumulation of abnormal ECM may have unforeseen adverse effects.
1 Robins SP, Farquharson C. Connective tissue components of the blood vessel wall in health and disease. In: Stehbens WE, Lie JT, editors. Vascular Pathology. London: Chapman & Hall Medical; 1995, p. 89-127. 2 Prockop DJ, Kivirikko KI. Collagens: molecular biology, diseases, and potentials for therapy. Ann Rev Biochem 1995; 64: 403-34. 3 van der Rest M, Garrone R. Collagen family of proteins. FASEB J 1991; 5: 2814-23. 4 Mayne R. Collagenous proteins of blood vessels. Arteriosclerosis 1986; 6: 585-93. 5 Reiser K, Mc Cormick RJ, Rucker RB. Enzymatic and nonenzymatic cross-linking of collagen and elastin. FASEB J 1992; 6: 2439-49. 6 Jacob MP. Élastine : préparation, caractérisation, structure, biosynthèse et catabolisme. CR Soc Biol 1993; 187: 166-80. 7 Rosenbloom J, Abrams WR, Mecham R. Extracellular matrix. 4: The elastic fiber. FASEB J 1993; 7: 1208-18. 8 Debelle L, Tamburro AM. Elastin: molecular description and function. Int J Biochem Cell Biol 1999; 31: 261-72. 9 Li DY, Brooke B, Davis EC, Mecham RP, Sorensen LK, Boak BB, et al. Elastin is an essential determinant of arterial morphogenesis. Nature 1998; 393: 276-80. 10 Li DY, Faury G, Taylor DG, Davis EC, Boyle WA, Mecham RP, et al. Novel arterial pathology in mice and humans hemizygous for elastin. J Clin Invest 1998; 102: 1783-7. 11 Milewicz DM, Urban Z, Boyd C. Genetic disorders of the elastic fiber system. Matrix Biol 2000; 19: 471-80. 12 Handford PA, Downing AK, Reinhardt DP, Sakai LY. Fibrillin: from domain structure to supramolecular assembly. Matrix Biol 2000; 1: 457-70. 13 Sherratt MJ, Wess TJ, Baldock C, Ashworth J, Purslow PP, Shuttleworth CA, et al. Fibrillin-rich microfibrils of the extracellular matrix: ultrastructure and assembly. Micron 2001; 32: 185-200. 14 Gibson MA, Hatzinikolas G, Kumaratilake JS, Sandberg LB, Nicholl JK, Sutherland GR, et al. Further characterization of proteins associated with elastic fiber microfibrils including the molecular cloning of MAGP-2 (MP25). J Biol Chem 1996; 271: 1096-103. 15 Collod G, Boileau C. Fibrillines et fibrillinopathies. M/S 1996; 12: 1077-86. 16 Dietz HC, Mecham RP. Mouse models of genetic diseases resulting from mutations in elastic fiber proteins. Matrix Biol 2000; 1: 481-8.
332
M.P. Jacob et al.
17 Chothia C, Jones EY. The molecular structure of cell adhesion molecules. Ann Rev Biochem 1997; 66: 823-62. 18 Labat-Robert J. Cell-matrix interactions, alterations with aging, involvement in angiogenesis. Pathol Biol 1998; 46: 527-33. 19 Iozzo RV. Matrix proteoglycans: from molecular design to cellular function. Ann Rev Biochem 1998; 67: 609-52. 20 Praillet C, Grimaud JA, Lortat-Jacob H. Les protéoglycannes. I. Molécules aux multiples fonctions . . . futures molécules thérapeutiques ? M/S 1998; 14: 412-20. 21 Praillet C, Lortat-Jacob H, Grimaud JA. Les protéoglycannes. II. Rôles en pathologie. M/S 1998; 14: 421-8. 22 Shapiro SD, Endicott SK, Province MA, Pierce JA, Campbell EJ. Marked longevity of human lung parenchymal elastic fibers deduced from prevalence of D-aspartate and nuclear weapons-related radiocarbon. J Clin Invest 1991; 87: 1828-34. 23 Werb Z, Banda MJ, McKerrow JH, Sandhaus RA. Elastases and elastin degradation. J Invest Dermatol 1982; 79: 154s-9s. 24 Robert L. Les protéases de type élastase. Pathol Biol 1988; 36: 1101-7. 25 Nagase H, Woessner JF Jr. Matrix metalloproteinases. J Biol Chem 1999; 274: 21491-4. 26 Dollery CM, McEwan JR, Henney AM. Matrix metalloproteinases and cardiovascular disease. Circ Res 1995; 77: 863-8. 27 Borden P, Heller RA. Transcriptional control of matrix metalloproteinases and the tissue inhibitors of matrix metalloproteinases. Crit Rev Euk Gene Expr 1997; 7: 159-78. 28 Brew K, Dinakarpandian D, Nagase H. Tissue inhibitors of metalloproteinases: evolution, structure and function. Biochim Biophys Acta 2000; 1477: 267-83. 29 Devy L, Noël A, Cataldo D, Jeannesson P, Foidart JM. Rôle des métalloprotéinases matricielles et des protéases à sérine dans la migration et l’infiltration des cellules hématopoïétiques. Hématologie 1999; 5: 265-72. 30 Rabinovitch M. EVE and beyond, retro and prospective insights. Am J Physiol 1999; 277: L5-12. 31 Bieth JG. Leukocyte elastase. In: Barrett AJ, Rawlings ND, Woessner JF, editors. Handbook of proteolytic enzymes. San Diego: Academic Press; 1998, p. 54-60. 32 Badier-Commander C, Verbeuren T, Lebard C, Michel JB, Jacob MP. Increased TIMP/MMP ratio in varicose veins: a possible explanation for extracellular matrix accumulation. J Pathol 2000; 192: 105-12. 33 Gonzalez W, Fontaine V, Pueyo ME, Laquay N, MessikaZeitoun D, Philippe M, et al. Molecular plasticity of vascular wall during Nω-nitro-L-arginine methyl ester-induced hypertension: modulation of proinflammatory signals. Hypertension 2000; 36: 103-9. 34 Badier-Commander C, Couvelard A, Henin D, Verbeuren T, Michel JB, Jacob MP. Smooth muscle cell modulation and cytokine overproduction in varicose veins. An in situ study. J Pathol 2001; 193: 398-407. 35 Keeley FW, Alatawi A. Response of aortic elastin synthesis and accumulation to developing hypertension and the inhibitory effect of colchicine on this response. Lab Invest 1991; 64: 499-507. 36 Keeley FW, Elmoselhi A, Leenen FH. Effects of antihypertensive drug classes on regression of connective tissue components of hypertension. J Cardiovasc Pharmacol 1991; 17: S64-9. 37 Keeley FW, Elmoselhi A, Leenen FH. Enalapril suppresses normal accumulation of elastin and collagen in cardiovascular tissues of growing rats. Am J Physiol 1992; 262: H1013-21. 38 Laviades C, Varo N, Fernandez J, Mayor G, Gil MJ, Monreal I, et al. Abnormalities of the extracellular degradation of collagen type I in essential hypertension. Circulation 1998; 98: 535-40. 39 Mehran R, Dangas G, Abizaid AS, Mintz GS, Lansky AJ, Satler LF, et al. Angiographic patterns of in-stent restenosis: classification and implications for long-term outcome. Circulation 1999; 100: 1872-8.
40 Bendeck MP, Zempo N, Clowes AW, Galardy RE, Reidy MA. Smooth muscle cell migration and matrix metalloproteinase expression after arterial injury in the rat. Circ Res 1994; 75: 539-45. 41 Clowes AW, Clowes MM, Au YP, Reidy MA, Belin D. Smooth muscle cells express urokinase during mitogenesis and tissue-type plasminogen activator during migration in injured rat carotid artery. Circ Res 1990; 67: 61-7. 42 Southgate KM, Fisher M, Banning AP, Thurston VJ, Baker AH, Fabunmi RP, et al. Upregulation of basement membrane-degrading metalloproteinase secretion after balloon injury of pig carotid arteries. Circ Res 1996; 79: 1177-87. 43 Bassiouny HS, Song RH, Hong XF, Singh A, Kocharyan H, Glagov S. Flow regulation of 72-kD collagenase IV (MMP-2) after experimental arterial injury. Circulation 1998; 98: 157-63. 44 Strauss BH, Chisholm RJ, Keeley FW, Gotlieb AI, Logan RA, Armstrong PW. Extracellular matrix remodeling after balloon angioplasty injury in a rabbit model of restenosis. Circ Res 1994; 75: 650-8. 45 Bendeck MP, Irvin C, Reidy MA. Inhibition of matrix metalloproteinase activity inhibits smooth muscle cell migration but not neointimal thickening after arterial injury. Circ Res 1996; 78: 3843. 46 Cheng L, Mantile G, Pauly R, Nater C, Felici A, Monticone R, et al. Adenovirus-mediated gene transfer of the human tissue inhibitor of metalloproteinase-2 blocks vascular smooth muscle cell invasiveness in vitro and modulates neointimal development in vivo. Circulation 1998; 98: 2195-201. 47 Dollery CM, Humphries SE, McClelland A, Latchman DS, McEwan JR. Expression of tissue inhibitor of matrix metalloproteinases 1 by use of an adenoviral vector inhibits smooth muscle cell migration and reduces neointimal hyperplasia in the rat model of vascular balloon injury. Circulation 1999; 99: 3199-205. 48 Michel JB. Acquired abdominal aortic aneurysm. Nephrol Dial Transpl 1998; 13(Suppl 4): 20-4. 49 Herron GS, Unemori EN, Stoney RJ. Proteinases and inhibitors in aortic aneurysms. In: Weber KT, editor. Wound Healing in cardiovascular disease. Armonk: Futura Publishing Company; 1995, p. 195-209. 50 Thompson RW, Holmes DR, Mertens RA, Liao S, Botney MD, Mecham RP, et al. Production and localization of 92-kilodalton gelatinase in abdominal aortic aneurysms. An elastolytic metalloproteinase expressed by aneurysm-infiltrating macrophages. J Clin Invest 1995; 96: 318-26. 51 Curci JA, Liao S, Huffman MD, Shapiro SD, Thompson RW. Expression and localization of macrophage elastase (matrix metalloproteinase-12) in abdominal aortic aneurysms. J Clin Invest 1998; 102: 1900-10. 52 Tamarina NA, McMillan WD, Shively VP, Pearce WH. Expression of matrix metalloproteinases and their inhibitors in aneurysms and normal aorta. Surgery 1997; 122: 264-72. 53 Knox JB, Sukhova GK, Whittemore AD, Libby P. Evidence for altered balance between matrix metalloproteinases and their inhibitors in human aortic diseases. Circulation 1997; 95: 205-12. 54 Lopez-Candales A, Holmes DR, Liao S, Scott MJ, Wickline SA, Thompson RW. Decreased vascular smooth muscle cell density in medial degeneration of human abdominal aortic aneurysms. Am J Pathol 1997; 150: 993-1007. 55 Allaire E, Forough R, Clowes M, Starcher B, Clowes AW. Local overexpression of TIMP-1 prevents aortic aneurysm degeneration and rupture in a rat model. J Clin Invest 1998; 102: 1413-20. 56 Allaire E, Hasenstab D, Kenagy RD, Starcher B, Clowes MM, Clowes AW. Prevention of aneurysm development and rupture by local overexpression of plasminogen activator inhibitor-1. Circulation 1998; 98: 249-55. 57 Fukuda S, Hashimoto N, Naritomi H, Nagata I, Nozaki K, Kondo S, et al. Prevention of rat cerebral aneurysm formation by inhibition of nitric oxide synthase. Circulation 2000; 101: 2532-8.
2001 Éditions scientifiques et médicales Elsevier SAS. Tous droits réservés S0369-8114(01)00151-1/REV
Revue
Extracellular matrix remodeling in the vascular wall M.P. Jacob 1∗ , C. Badier-Commander 1 , V. Fontaine 1 , Y. Benazzoug 2 , L. Feldman 1 , J.B. Michel 1 1 INSERM U 460, UFR de médecine Xavier Bichat, 16, rue Henri Huchard 75870 Paris cedex 18, France ; 2 laboratoire de la matrice extracellulaire, faculté des sciences biologiques, USTHB, BP n◦ 32, Bab Ezzouar,
16111 Alger, Algérie Summary The extracellular matrix provides a structural framework essential for the functional properties of vessel walls. The three dimensional organization of the extracellular matrix molecules — elastin, collagens, proteoglycans and structural glycoproteins — synthesized during fetal development — is optimal for these functions. Early in life, the vessel wall is subjected to injury: lipid deposition, hypoxia, enzyme secretion and reactive oxygen species production during inflammatory processes, and the extracellular matrix molecules are hydrolyzed by proteases — matrix metalloproteinases, leukocyte elastase, etc. In uninjured arteries and veins, some proteases are constitutively expressed, but through the control of their activation and/or their inhibition by inhibitors, these proteases have a very low activity. During the occurrence of vascular pathologies — atherosclerosis, hypertension, varicosis, restenosis, etc. — the balance between proteases and their inhibitors is temporally destroyed through the induction of matrix metalloproteinase gene expression or the secretion of enzymes by inflammatory cells. Smooth muscle cells, the most numerous cells in vascular walls, have a high ability to respond to injury through their ability to synthesize extracellular matrix molecules and protease inhibitors. However, the three dimensional organization of the newly synthesized extracellular matrix is never functionally optimal. In some other pathologies — aneurysm — the injury overcomes the responsive capacity of smooth muscle cells and the quantity of extracellular matrix decreases. In conclusion, care should be taken to maintain the vascular extracellular matrix reserve and any therapeutic manipulation of the protease/inhibitor balance must be perfectly controlled, because an accumulation of abnormal extracellular matrix may have unforeseen adverse effects. 2001 Éditions scientifiques et médicales Elsevier SAS elastases / elastin / collagens / matrix metalloproteinases / vascular wall
Résumé – Le rôle de la matrice extracellulaire dans le remodelage de la paroi vasculaire. La matrice extracellulaire (MEC) confère aux parois vasculaires une structure de soutien indispensable à leurs propriétés fonctionnelles. Les différentes molécules de la MEC — élastine, collagènes protéoglycannes, glycoprotéines de structure — s’organisent de façon optimale dans l’espace extracellulaire au fur et à mesure de leur synthèse au cours du développement et de la croissance de l’individu. Très tôt, la paroi vasculaire est soumise à différentes agressions et les macromolécules de la MEC sont hydrolysées par des protéases. Dans les parois vasculaires normales, la métalloprotéase matricielle (MMP) 2 est exprimée de manière constitutive mais son activité enzymatique reste très faible car elle reste sous forme proenzyme et des inhibiteurs, les inhibiteurs tissulaires des MMPs, sont présents. Au cours des diverses pathologies vasculaires, l’équilibre protéases/inhibiteurs est temporairement détruit de par l’expression et l’activation des MMPs par les cellules vasculaires elles-mêmes ou la sécrétion d’enzymes par les cellules inflammatoires. Par leur capacité de synthèse des différentes macromolécules de la MEC et des inhibiteurs des MMPs, les cellules musculaires lisses ont une forte capacité à répondre à ces agressions. La structure tridimensionnelle de la MEC néosynthétisée n’est toutefois jamais aussi optimale que celle synthétisée au cours du développement. Dans d’autres pathologies vasculaires — anévrysmes —, l’hydrolyse enzymatique des macromolécules dépasse la
∗ Correspondence and reprints.
E-mail address: [email protected] (M.P. Jacob).
Extracellular matrix remodeling in the vascular wall
327
capacité de réponse des cellules musculaires lisses et la quantité de MEC diminue. Il est donc indispensable de préserver au maximun la MEC des parois vasculaires, l’utilisation d’inhibiteurs de protéases devant être parfaitement contrôlée et dans le temps et dans l’espace pour éviter l’effet secondaire qui peut être lui-même néfaste, l’accumulation de MEC non fonctionnelle. 2001 Éditions scientifiques et médicales Elsevier SAS élastases / élastine / collagènes / métalloprotéinases matricielles / paroi vasculaire
The extracellular matrix (ECM) in the vascular wall provides an essential supporting, structural framework necessary for the structural and functional properties of vessel walls (elasticity, resistance to stretch). Besides the two main structural macromolecules, elastin and collagens, proteoglycans and structural glycoproteins play essential roles in other functional requirements of vessels (hydration, ion filtration, interactions between ECM molecules, growth factor biodisponibility, cell-matrix interactions). These ECM macromolecules are synthesized by the three vascular cell types: intimal endothelial cells, medial smooth muscle cells (SMC) and adventitial fibroblasts. The three dimensional organization of these ECM molecules synthesized during fetal development is optimal for the functions of the vessel wall. But early in life, the vessel wall is subjected to injury (lipid deposition, hypoxia, enzyme secretion and reactive oxygen species production during inflammatory processes, etc.). One of the main lessons of our studies is that smooth muscle cells have an extremely high capacity to respond to injuries via their capacity to synthesize ECM molecules and enzyme inhibitors. However, the three dimensional organization of the newly synthesized ECM is never functionally optimal as is that synthesized during development. This response to injury occurs in several vascular pathologies (atherosclerosis, hypertension, varicosis, etc.). In some other pathologies, the injury overcomes the responsive capacity of smooth muscle cells (aneurysm, etc.). This review summarizes the results of our recent studies on varicosis and hypertension, and of previous studies on aneurysms and restenosis, after a brief description of the ECM molecules of the vascular wall and their degradation by proteinases.
THE ECM IN THE VASCULAR WALL Collagens and elastin are the main protein constituents of vessels. The biomechanical properties of vessels, particularly of the major arteries and veins, are largely dependent on the absolute and relative quantities of these two constituents [1]. Among the 19 different collagen types described until now, collagen types I and III are the major fibrillar collagens detectable in vessels, representing 60% and 30% of vascular collagens, respectively [2 – 4]. The remaining 10% include fibrillar collagen type V, Fibril-Associated Collagens with Interrupted Triple helix (FACIT) types XII
and XIV, microfibrillar collagen type VI, basement membrane collagen type IV and collagen type VIII. The basic unit of fibrillar collagens is the 300 nm triple helix, with each chain comprising about 330 repeats of the amino acid sequence -Gly-X-Y- (where X is often proline and Y is often hydroxyproline). After secretion in the extracellular compartment and removal of N- and C-propeptides, collagen triple helices aggregate in quarter-staggered fibrils. Newly synthesized collagen fibrils are soluble in salt solutions and dilute acid and have no tensile strength. During the formation of intermolecular cross-linking, collagen fibers become increasingly insoluble, more refractory to the action of enzymes and show a progressive increase in tensile strength. The cross-linking process is initiated by the enzyme lysyl oxydase, through the oxidation of specific lysine or hydroxylysine residues in the telopeptide regions. The resulting aldehydes undergo a series of reactions with adjacent reactive residues to give both inter- and intramolecular crosslinks [5]. Elastin is the most abundant protein of the large arteries that are subjected to a large pulsatile pressure generated by cardiac contraction [6 – 8]. However, elastin is also detectable in resistance arteries — mainly in the internal and external elastic lamina — and veins. Elastin represents 90% of the elastic fibers, the other constituents being microfibrillar glycoproteins such as fibrillins and MAGPs (Microfibrillar-Associated GlycoProteins). The precursor of elastin, tropoelastin, is a highly hydrophobic protein (MW, 72 kDa), which is soluble in salt solution like the collagen triple helix. In contrast, elastin is an insoluble protein. This insolubility results from the crosslinking process between lysine residues. Crosslinking of tropoelastin molecules begins with the oxidative deamination of some lysine residues by lysyl oxidase, as previously described for collagen crosslinking. The spontaneous condensation reaction between four lysine/allysine residues leads to the formation of the specific cross-links for elastin, desmosine and isodesmosine. Cross-links resulting from the condensation of two or three lysine/allysine residues are also detectable in elastin [5]. This crosslinking process confers to elastin its function — i.e. elasticity —, essential in large arteries which distend during systole and recoil during diastole. Another property of elastin has been recently demonstrated by the study of patients suffering from supravalvular aortic stenosis and of knock-out mice for elastin: elastin controls, directly or indirectly, the proliferation and phenotype of smooth muscle cells [9 – 11].
328
M.P. Jacob et al.
Several structural glycoproteins are closely associated with elastin in elastic fibers: MAGPs and fibrillins [12 – 14]. Fibrillin 1 has been the focus of important research since mutations in the fibrillin 1 gene were found in patients suffering from Marfan’s syndrome [15, 16]. Fibrillins 1 and 2, 350 kDa glycoproteins, are characterized by the multiple repeats of EGF-like domains, in which the three cystine bridges stabilize the β-turns within the molecule. Several other structural glycoproteins play an essential role in the structure and function of the ECM in the arterial wall, including fibronectin, vitronectin, laminin, entactin/nidogen, tenascin and thrombospondin. These glycoproteins have a multidomain structure, potentially enabling simultaneous interactions between cells and other ECM components [17, 18]. The proteoglycans contained in the vascular wall are the large aggregating proteoglycans aggrecan and versican, the small non-aggregating interstitial proteoglycans biglycan, decorin and fibromodulin and the cell-associated proteoglycans syndecan, fibroglycan and glypican. Proteoglycans are proteins that have one or more attached glycosaminoglycan chains [19, 20]. They contain distinct protein and carbohydrate domain structures, which interact with other ECM molecules. They participate in ECM assembly and confer specific properties to the tissues (hydration, filtration, etc.). Proteoglycans also regulate various cellular activities (proliferation, differentiation, adhesion, migration) and control cytokine biodisponibility and stability (FGF, TGF-β, etc.) [21].
THE PROTEINASES/INHIBITORS INVOLVED IN VASCULAR ECM REMODELING The metabolic turnover of mature collagen and elastic fibers in adult animals is relatively slow [22]. Although only small amounts of these proteins are degraded normally, increased degradation and fragmentation of collagen and elastin fibers are observed in vascular diseases. Collagenolytic and elastinolytic enzymes are found in a number of mammalian cells and tissues including polymorphonuclear neutrophils and monocytes/macrophages. Collagenolytic enzymes are mainly matrix metalloproteinases. Elastinolytic enzymes are found in all the four classes of proteinases (serine, metallo-, aspartic and cysteine proteinases) and react with a broad spectrum of substrates [23, 24]. All elastases studied to date have catalytic activity against protein and peptide substrates other than elastin. Matrix metalloproteinases and other elastolytic enzymes have been described extensively in previous reviews [25 – 29]. We will focus only on the serine and metalloproteinases which participate to vascular ECM remodeling, i.e., proteinases synthesized in vascular cells themselves and those secreted by inflammatory cells (table I).
Table I. Serine and metallo-proteinases active on extracellular matrix macromolecules, detectable in vascular cells and inflammatory cells. Proteinases produced by vascular cells
Proteinases produced by inflammatory cells
Matrix metalloproteinases proMMP-1* (EC) proMMP-2 (EC, SMC, fibroblasts) proMMP-3* proMMP-7 proMMP-9*
Matrix metalloproteinases proMMP-1 (monocytes) proMMP-2 proMMP-7 (monocytes) proMMP-8 (PMN) proMMP-9 (macrophages, PMN) proMMP-12 (macrophages) Serine proteinases Leukocyte elastase (PMN)
Serine proteinases SMC elastase
* enzymes not expressed in basal conditions. EC, endothelial cells; MMP, matrix metalloproteinases; PMN, polymorphonuclear neutrophils; SMC, smooth muscle cells.
The two serine proteinases involved during vascular remodeling are the SMC elastase [30] and the leukocyte elastase which is stored in the azurophil granules of polymorphonuclear neutrophils [31]. One inhibitor of SMC elastase is elafin and leukocyte elastase is inhibited by α1-proteinase inhibitor, α2-macroglobulin and elafin. Matrix metalloproteinases (MMPs) play a fundamental role in the degradation of vascular ECM. MMPs include collagenases (MMP-1, expressed by endothelial cells, MMP8, stored in specific granules of PMN), gelatinases (MMP-2, constitutively expressed by vascular cells; MMP-9, expressed in macrophages and PMN, inducible in vascular cells), ‘elastases’ (MMP-7, expressed at a low level by the vascular wall, MMP-12, synthesized during the differentiation of monocytes into macrophages), stromelysins and membrane-type metalloproteinases (MT-MMPs). The proteolytic activity of each MMP is tightly regulated at three levels: first, gene expression and protein secretion levels; second, activation of the inactive pro-enzyme, and, third, inhibition by the TIMPs (Tissue Inhibitors of Matrix Metalloproteinases) or other inhibitors (α2 -macroglobulin). The activation of secreted pro-MMPs requires the disruption of the Cys-Zn2+ (cysteine switch) interaction and the removal of the propeptide. Pro-MMPs can be activated in vitro by SH-reactive agents, mercurial compounds, SDS or chaotropic agents. In vivo, pro-MMPs are activated by tissue or plasma proteinases — plasmin, thrombin, other MMPs or MTMMPs — and reactive oxygen species. The activation of MT-MMPs occurs intracellularly by furin. According to our experiments [32, 33], the activation of MMPs and their inhibition by TIMPs are the main regulatory mechanisms of MMP activities in the vascular wall. For example, proMMP-2 is constitutively expressed in the normal arterial wall, in which no SMC migration or elastin degradation are detectable. The very
Extracellular matrix remodeling in the vascular wall
329
well controlled balance between active proteinases and inhibitors is perturbed during pathological processes, particularly when PMNs or macrophages are present. However, through their capacity to synthesize enzyme inhibitors, smooth muscle cells have an extremely high capacity to respond to these increased enzyme levels.
PAI amount, were lower in varicose veins and the ratio PAI/uPA + tPA was slightly lower in varicose veins than in control veins [32].
ECM REMODELING IN VARICOSE VEINS
The ECM content (collagen + elastin) of arteries is increased in hypertensive patients and in genetic or experimentally induced hypertension in animals. The increase in collagen and elastin synthesis induced by the increased blood pressure has been carefully studied by Keeley et al. [35 – 37]. The collagen and elastin synthesis returns to basal levels as soon as the blood pressure stops rising. While the absolute aortic contents of elastin and collagen are increased in hypertensive animals, the elastin/collagen ratios are the same in normotensive and hypertensive animals. These increased collagen and elastin contents in hypertensive animals may also result from reduced degradation. The MMP/TIMP balance has been studied in the rat model of hypertension induced by chronic blockade of NO production by Nω-nitro-L-arginine methyl ester [33]. ProMMP-9 was not detected in the aortic extracts of normotensive and hypertensive rats. ProMMP-2 activity, as well as TIMP-2 inhibitory capacity, were not significantly modified in the aortas of hypertensive rats. In contrast, TIMP-1 inhibitory capacity, measured by reverse zymography, and TIMP-1 mRNA levels were significantly increased in hypertensive rats. These data suggest that hypertension is associated with the overexpression of both pro-inflammatory molecules (Interleukin-6, Monocyte Chemoattractant Protein-1 and Macrophage ColonyStimulating Factor) and resistance factors such as TIMPs. TIMP-1, by inhibiting MMP activity, limits inflammatory cell migration into the media and leads to subsequent increase in ECM content. Interestingly, it was recently demonstrated that serum concentration of free TIMP-1 is increased, while serum concentration of MMP-1 is diminished, in patients with essential hypertension. Furthermore, free TIMP-1 concentration was decreased, and free MMP-1 concentration was increased, in the sera of the same patients after one year of treatment with an angiotensin-converting enzyme inhibitor [38]. Taken together, these studies demonstrate that the synthesis of ECM is enhanced and its degradation reduced in the arteries of hypertensive individuals.
Primary varicose veins are functionally characterized by venous back-flow and increased blood stasis in the upright position. Dilatation and tortuosity provide evidence for progressive venous wall remodeling, with disturbance of smooth muscle cell/ECM organization. Affected areas are not uniformly distributed, some areas being hypertrophic, whereas others are atrophic or unaffected. When the proteolytic MMP/TIMP balance, which may participate in the remodeling of the venous wall, was investigated, a higher TIMP-1 level, an equal quantity of TIMP-2, and lower MMP-2 level and activity (based on soluble protein content) were found in varicose veins compared to control veins [32]. Consequently, the protein ratios TIMP-1/MMP-2 and TIMP-2/MMP-2 were 3.6- and 2.1-fold higher in varicose veins compared to control veins, respectively. ProMMP-9 was detected in variable amounts, both in control and varicose veins, but the mean activity was lower in varicose veins. The lower level of proMMP-2, constitutively expressed by smooth muscle cells and endothelial cells, probably reflects a lower production of proMMP-2 by cells of the varicose veins. In contrast, the increase in TIMP-1, detectable in both endothelial cells and smooth muscle cells, indicates a higher expression of the TIMP-1 gene by vascular cells. Indeed, depending on the cell type, TIMP-1 expression can be modulated by various cytokines (Tumor Necrosis Factor-α, Interleukin-1β, Platelet-Derived Growth Factor, Transforming growth factor-β [TGF-β]) and soluble factors (angiotensin II, Phorbol Myristate Acetate, retinoids). All these molecules have not been measured in varicose veins, but an increase in the total TGF-β1 concentration has been reported [34]. In turn, TGF-β1 downregulates MMPs and upregulates TIMP-1. This stimulatory effect of TGF-β1 on TIMP-1 expression is additive to the stimulation of collagen and elastin synthesis by TGF-β. Hypoxia is also a potential candidate as a modulator of TIMP-1 expression. This imbalance between MMP and TIMP production, in favor of an antiproteolytic activity, may explain, at least in part, the accumulation of ECM in varicose veins. As plasmin activates some proMMPs and degrades some ECM molecules, the components of the fibrinolytic system, including tissue-type (tPA) and urokinase-type (uPA) plasminogen activators (PAs) and plasminogen activator inhibitors (PAIs) were also measured in vein extracts. The uPA and tPA activities, as well as the
ECM REMODELING DURING HYPERTENSION
ECM REMODELING DURING RESTENOSIS Percutaneous transluminal coronary angioplasty has become a widely used treatment for occlusive coronary lesions. During angioplasty, the atheromatous plaque is
330
M.P. Jacob et al.
fissured and compressed into the vessel wall using a balloon catheter. However, immediate benefit of angioplasty is hampered by the risk of recurrent stenosis (or restenosis), which occurs in 20–40% of the patients within six months [39]. Restenosis is due to the migration and proliferation of smooth muscle cells from the media to the intima, producing a characteristic lesion of intimal fibro-cellular hyperplasia. Restenosis represents an exaggeration of the normal healing process since intimal hyperplasia is present to some degree in all patients after angioplasty. Using the rat model of balloon injury, the induction of proMMP-9 one day after injury and an increase in proMMP-2 at four to five days after angioplasty were demonstrated [40]. The activity of tissue-type plasminogen activator was also increased at four days after arterial injury [41]. The pig shows the same pattern of increased MMP-9 and MMP-2 activities after arterial injury, with maximal activities at seven days and persistantly elevated activities for up to 21 days [42]. In the normocholesterolemic rabbit, MMP-2 mRNA levels increase after balloon angioplasty but MMP-9 remains undetectable [43, 44]. Only the combination of balloon injury with low flow or an atherogenic diet results in substantial proMMP-9 activity, in addition to the increased MMP-2 activity. The relevance of these increased MMP activities to SMC migration from the media to the intima was confirmed by the demonstration of a 97% decrease in the number of SMC that migrated into the intima 4 days after carotid artery injury in rats receiving an MMP inhibitor [45]. However, an increased rate of SMC replication in these treated rats permitted the growth of the neointima to ‘catch up’, thus contributing to the long-term inefficacity of the MMP inhibitor on neointimal thickening observed after arterial injury. Local delivery of genes encoding for TIMP-1 [46] or TIMP-2 [47] partially inhibits or delays neointimal growth in the same model.
ECM REMODELING IN ANEURYSMS Human abdominal aortic aneurysms are characterized by progressive dilatation that may lead to the rupture of the aortic wall [48]. Mutations in genes encoding type III collagen and fibrillin-1 are responsible for several forms of inheritable aneurysmal-like disease with vascular rupture, such as Ehlers-Danlos type IV and Marfan’s syndrome [11]. However, the great majority of abdominal aneurysms develop as acquired degenerative lesions in the absence of known ECM genetic defects [49]. Associations between aneurysms and aging, atherosclerosis, male gender, as well as smoking and inflammatory reactions have been established. The ECM remodeling in aneurysmal aortas is characterized by the structural disorganization and disappearance of the elastic lamella
in association with an inflammatory, fibrotic adventitia. The inflammatory process is characterized by infiltration of the outer aorta by mononuclear phagocytes and lymphocytes. Another biological phenomenon observed during the occurrence of abdominal aortic aneurysms is the invasion of media by capillaries; hence, proliferative endothelial cells are detectable in aneurysmal aortas. ECM degrading enzymes secreted by both invading and resident cell types are characteristic of all these processes. In fact, increased proteolytic activities have been described in aneurysmal aortas, including serine elastases, MMPs and plasminogen activators, these proteinases being directly or indirectly involved in aortic elastin degradation. The serine elastase involved in aortic aneurysms is neutrophil elastase. The main MMPs detected in aneurysmal aortas have elastolytic activities: MMP-2, constitutively expressed by vascular cells in normal aortas, MMP-9 and MMP-12, expressed in particular by tissue macrophages present in the outer aneurysmal aorta [50, 51]. The quantities of TIMP-1 and TIMP-2 in aneurysmal tissues have been measured at the mRNA and protein levels, and are significantly increased in pathological tissues [50, 52]. In order to evaluate the net proteolytic activity in aneurysmal and occlusive aortas, in situ zymography was used. No difference in the net proteolytic activity was detected between aneurysmal and occlusive aortas and this proteolytic activity was detected in the luminal portion of the specimens [53]. This result is quite surprising since the main gelatinolytic enzyme in aneurysmal aortas, MMP-9, has been reported to colocalize with macrophages and is spatially associated with periadventitial vascularization and lymphocytic infiltration [50]. Finally, the well described decrease in medial SMC density in aneurysmal aortas compared to normal or occlusive aortas may explain the lower production of MMP inhibitors relative to proteinases in aneurysmal aortas [54]. Indeed, the local seeding of SMC, whether they overexpress TIMP-1 or not, prevents elastin depletion, aneurysm formation and rupture in a rat model of aneurysm [55]. The blockade of MMP activation also prevents both early aneurysmal changes and the incidence of induced aneurysms in rat models of aortic and cerebral aneurysms [56, 57].
CONCLUSION In uninjured arteries and veins, some MMPs are constitutively expressed (table I). But through the control of their activation and their inhibition by inhibitors synthesized by the vascular cells themselves (TIMPs), these proteases have a very low activity. During the occurrence of vascular pathologies, this balance between proteases and their inhibitors is temporarily destroyed through the induction of MMP gene
331
Extracellular matrix remodeling in the vascular wall
ACKNOWLEDGEMENTS The authors thank Dr Mary Osborne-Pellegrin for the editing of this manuscript. Our studies were supported by grants from the Institut National de la Santé et de la Recherche Médicale, the Institut de recherches international Servier (varicosis) and Menarini, Florence, Italy (hypertension).
ABBREVIATIONS ECM, extracellular matrix; MMP, matrix metalloproteinase; TIMP, tissue inhibitor of matrix metalloproteinase, tPA, tissue-type plasminogen activator; uPA, urokinase-type plasminogen activator; PAI, Plasminogen activator inhibitor; PMN, polymorphonuclear neutrophils; SMC, smooth muscle cells.
REFERENCES
Figure 1. Extracellular matrix homeostasis in the vascular wall. α1-AT, α1-proteinase inhibitor; α2-M, α2-macroglobulin; EC, endothelial cells; ECM, extracellular matrix; EEL, external elastic lamina; IEL, internal elastic lamina; PGs, proteoglycans; PMN, polymorphonuclear neutrophils; SMC, smooth muscle cells; TIMPs, tissue inhibitors of metalloproteinases.
expression in vascular cells or the secretion of enzymes by inflammatory cells. If the protease levels exceed only for a short time the inhibitor level, this may then be overcompensated and the normal healing and repair process would lead to the deposition of newly synthesized ECM components and the arrest/inhibition of SMC proliferation (figure 1). However, during the long-term evolution of the principal vascular pathologies (atherosclerosis, aneurysm) a minimal imbalance between degradation and synthesis of ECM components could lead either to stenosis or to dilatation of the vessel wall. Furthermore, depending on the extent of the degradation and synthesis of each ECM component, the relative levels of each component, and especially the elastin/collagen ratio, may be modified. In addition, the three-dimensional organization of this newly synthesized ECM, and consequently its functional properties, are never optimal as are those of ECM synthesized during development. In conclusion care should be taken to maintain the vascular ECM reserve and any therapeutic manipulation of the protease/inhibitor balance must be perfectly controlled, because an accumulation of abnormal ECM may have unforeseen adverse effects.
1 Robins SP, Farquharson C. Connective tissue components of the blood vessel wall in health and disease. In: Stehbens WE, Lie JT, editors. Vascular Pathology. London: Chapman & Hall Medical; 1995, p. 89-127. 2 Prockop DJ, Kivirikko KI. Collagens: molecular biology, diseases, and potentials for therapy. Ann Rev Biochem 1995; 64: 403-34. 3 van der Rest M, Garrone R. Collagen family of proteins. FASEB J 1991; 5: 2814-23. 4 Mayne R. Collagenous proteins of blood vessels. Arteriosclerosis 1986; 6: 585-93. 5 Reiser K, Mc Cormick RJ, Rucker RB. Enzymatic and nonenzymatic cross-linking of collagen and elastin. FASEB J 1992; 6: 2439-49. 6 Jacob MP. Élastine : préparation, caractérisation, structure, biosynthèse et catabolisme. CR Soc Biol 1993; 187: 166-80. 7 Rosenbloom J, Abrams WR, Mecham R. Extracellular matrix. 4: The elastic fiber. FASEB J 1993; 7: 1208-18. 8 Debelle L, Tamburro AM. Elastin: molecular description and function. Int J Biochem Cell Biol 1999; 31: 261-72. 9 Li DY, Brooke B, Davis EC, Mecham RP, Sorensen LK, Boak BB, et al. Elastin is an essential determinant of arterial morphogenesis. Nature 1998; 393: 276-80. 10 Li DY, Faury G, Taylor DG, Davis EC, Boyle WA, Mecham RP, et al. Novel arterial pathology in mice and humans hemizygous for elastin. J Clin Invest 1998; 102: 1783-7. 11 Milewicz DM, Urban Z, Boyd C. Genetic disorders of the elastic fiber system. Matrix Biol 2000; 19: 471-80. 12 Handford PA, Downing AK, Reinhardt DP, Sakai LY. Fibrillin: from domain structure to supramolecular assembly. Matrix Biol 2000; 1: 457-70. 13 Sherratt MJ, Wess TJ, Baldock C, Ashworth J, Purslow PP, Shuttleworth CA, et al. Fibrillin-rich microfibrils of the extracellular matrix: ultrastructure and assembly. Micron 2001; 32: 185-200. 14 Gibson MA, Hatzinikolas G, Kumaratilake JS, Sandberg LB, Nicholl JK, Sutherland GR, et al. Further characterization of proteins associated with elastic fiber microfibrils including the molecular cloning of MAGP-2 (MP25). J Biol Chem 1996; 271: 1096-103. 15 Collod G, Boileau C. Fibrillines et fibrillinopathies. M/S 1996; 12: 1077-86. 16 Dietz HC, Mecham RP. Mouse models of genetic diseases resulting from mutations in elastic fiber proteins. Matrix Biol 2000; 1: 481-8.
332
M.P. Jacob et al.
17 Chothia C, Jones EY. The molecular structure of cell adhesion molecules. Ann Rev Biochem 1997; 66: 823-62. 18 Labat-Robert J. Cell-matrix interactions, alterations with aging, involvement in angiogenesis. Pathol Biol 1998; 46: 527-33. 19 Iozzo RV. Matrix proteoglycans: from molecular design to cellular function. Ann Rev Biochem 1998; 67: 609-52. 20 Praillet C, Grimaud JA, Lortat-Jacob H. Les protéoglycannes. I. Molécules aux multiples fonctions . . . futures molécules thérapeutiques ? M/S 1998; 14: 412-20. 21 Praillet C, Lortat-Jacob H, Grimaud JA. Les protéoglycannes. II. Rôles en pathologie. M/S 1998; 14: 421-8. 22 Shapiro SD, Endicott SK, Province MA, Pierce JA, Campbell EJ. Marked longevity of human lung parenchymal elastic fibers deduced from prevalence of D-aspartate and nuclear weapons-related radiocarbon. J Clin Invest 1991; 87: 1828-34. 23 Werb Z, Banda MJ, McKerrow JH, Sandhaus RA. Elastases and elastin degradation. J Invest Dermatol 1982; 79: 154s-9s. 24 Robert L. Les protéases de type élastase. Pathol Biol 1988; 36: 1101-7. 25 Nagase H, Woessner JF Jr. Matrix metalloproteinases. J Biol Chem 1999; 274: 21491-4. 26 Dollery CM, McEwan JR, Henney AM. Matrix metalloproteinases and cardiovascular disease. Circ Res 1995; 77: 863-8. 27 Borden P, Heller RA. Transcriptional control of matrix metalloproteinases and the tissue inhibitors of matrix metalloproteinases. Crit Rev Euk Gene Expr 1997; 7: 159-78. 28 Brew K, Dinakarpandian D, Nagase H. Tissue inhibitors of metalloproteinases: evolution, structure and function. Biochim Biophys Acta 2000; 1477: 267-83. 29 Devy L, Noël A, Cataldo D, Jeannesson P, Foidart JM. Rôle des métalloprotéinases matricielles et des protéases à sérine dans la migration et l’infiltration des cellules hématopoïétiques. Hématologie 1999; 5: 265-72. 30 Rabinovitch M. EVE and beyond, retro and prospective insights. Am J Physiol 1999; 277: L5-12. 31 Bieth JG. Leukocyte elastase. In: Barrett AJ, Rawlings ND, Woessner JF, editors. Handbook of proteolytic enzymes. San Diego: Academic Press; 1998, p. 54-60. 32 Badier-Commander C, Verbeuren T, Lebard C, Michel JB, Jacob MP. Increased TIMP/MMP ratio in varicose veins: a possible explanation for extracellular matrix accumulation. J Pathol 2000; 192: 105-12. 33 Gonzalez W, Fontaine V, Pueyo ME, Laquay N, MessikaZeitoun D, Philippe M, et al. Molecular plasticity of vascular wall during Nω-nitro-L-arginine methyl ester-induced hypertension: modulation of proinflammatory signals. Hypertension 2000; 36: 103-9. 34 Badier-Commander C, Couvelard A, Henin D, Verbeuren T, Michel JB, Jacob MP. Smooth muscle cell modulation and cytokine overproduction in varicose veins. An in situ study. J Pathol 2001; 193: 398-407. 35 Keeley FW, Alatawi A. Response of aortic elastin synthesis and accumulation to developing hypertension and the inhibitory effect of colchicine on this response. Lab Invest 1991; 64: 499-507. 36 Keeley FW, Elmoselhi A, Leenen FH. Effects of antihypertensive drug classes on regression of connective tissue components of hypertension. J Cardiovasc Pharmacol 1991; 17: S64-9. 37 Keeley FW, Elmoselhi A, Leenen FH. Enalapril suppresses normal accumulation of elastin and collagen in cardiovascular tissues of growing rats. Am J Physiol 1992; 262: H1013-21. 38 Laviades C, Varo N, Fernandez J, Mayor G, Gil MJ, Monreal I, et al. Abnormalities of the extracellular degradation of collagen type I in essential hypertension. Circulation 1998; 98: 535-40. 39 Mehran R, Dangas G, Abizaid AS, Mintz GS, Lansky AJ, Satler LF, et al. Angiographic patterns of in-stent restenosis: classification and implications for long-term outcome. Circulation 1999; 100: 1872-8.
40 Bendeck MP, Zempo N, Clowes AW, Galardy RE, Reidy MA. Smooth muscle cell migration and matrix metalloproteinase expression after arterial injury in the rat. Circ Res 1994; 75: 539-45. 41 Clowes AW, Clowes MM, Au YP, Reidy MA, Belin D. Smooth muscle cells express urokinase during mitogenesis and tissue-type plasminogen activator during migration in injured rat carotid artery. Circ Res 1990; 67: 61-7. 42 Southgate KM, Fisher M, Banning AP, Thurston VJ, Baker AH, Fabunmi RP, et al. Upregulation of basement membrane-degrading metalloproteinase secretion after balloon injury of pig carotid arteries. Circ Res 1996; 79: 1177-87. 43 Bassiouny HS, Song RH, Hong XF, Singh A, Kocharyan H, Glagov S. Flow regulation of 72-kD collagenase IV (MMP-2) after experimental arterial injury. Circulation 1998; 98: 157-63. 44 Strauss BH, Chisholm RJ, Keeley FW, Gotlieb AI, Logan RA, Armstrong PW. Extracellular matrix remodeling after balloon angioplasty injury in a rabbit model of restenosis. Circ Res 1994; 75: 650-8. 45 Bendeck MP, Irvin C, Reidy MA. Inhibition of matrix metalloproteinase activity inhibits smooth muscle cell migration but not neointimal thickening after arterial injury. Circ Res 1996; 78: 3843. 46 Cheng L, Mantile G, Pauly R, Nater C, Felici A, Monticone R, et al. Adenovirus-mediated gene transfer of the human tissue inhibitor of metalloproteinase-2 blocks vascular smooth muscle cell invasiveness in vitro and modulates neointimal development in vivo. Circulation 1998; 98: 2195-201. 47 Dollery CM, Humphries SE, McClelland A, Latchman DS, McEwan JR. Expression of tissue inhibitor of matrix metalloproteinases 1 by use of an adenoviral vector inhibits smooth muscle cell migration and reduces neointimal hyperplasia in the rat model of vascular balloon injury. Circulation 1999; 99: 3199-205. 48 Michel JB. Acquired abdominal aortic aneurysm. Nephrol Dial Transpl 1998; 13(Suppl 4): 20-4. 49 Herron GS, Unemori EN, Stoney RJ. Proteinases and inhibitors in aortic aneurysms. In: Weber KT, editor. Wound Healing in cardiovascular disease. Armonk: Futura Publishing Company; 1995, p. 195-209. 50 Thompson RW, Holmes DR, Mertens RA, Liao S, Botney MD, Mecham RP, et al. Production and localization of 92-kilodalton gelatinase in abdominal aortic aneurysms. An elastolytic metalloproteinase expressed by aneurysm-infiltrating macrophages. J Clin Invest 1995; 96: 318-26. 51 Curci JA, Liao S, Huffman MD, Shapiro SD, Thompson RW. Expression and localization of macrophage elastase (matrix metalloproteinase-12) in abdominal aortic aneurysms. J Clin Invest 1998; 102: 1900-10. 52 Tamarina NA, McMillan WD, Shively VP, Pearce WH. Expression of matrix metalloproteinases and their inhibitors in aneurysms and normal aorta. Surgery 1997; 122: 264-72. 53 Knox JB, Sukhova GK, Whittemore AD, Libby P. Evidence for altered balance between matrix metalloproteinases and their inhibitors in human aortic diseases. Circulation 1997; 95: 205-12. 54 Lopez-Candales A, Holmes DR, Liao S, Scott MJ, Wickline SA, Thompson RW. Decreased vascular smooth muscle cell density in medial degeneration of human abdominal aortic aneurysms. Am J Pathol 1997; 150: 993-1007. 55 Allaire E, Forough R, Clowes M, Starcher B, Clowes AW. Local overexpression of TIMP-1 prevents aortic aneurysm degeneration and rupture in a rat model. J Clin Invest 1998; 102: 1413-20. 56 Allaire E, Hasenstab D, Kenagy RD, Starcher B, Clowes MM, Clowes AW. Prevention of aneurysm development and rupture by local overexpression of plasminogen activator inhibitor-1. Circulation 1998; 98: 249-55. 57 Fukuda S, Hashimoto N, Naritomi H, Nagata I, Nozaki K, Kondo S, et al. Prevention of rat cerebral aneurysm formation by inhibition of nitric oxide synthase. Circulation 2000; 101: 2532-8.