Matrix metalloproteinases: are they antiatherogenic but proaneurysmal?

The matrix metalloproteinases (MMPs) are a family of enzymes (25 identified to date) that have in common the ability to degrade many molecules of the extracellular matrix. MMP activity can be inhibited by the endogenous tissue inhibitors of metalloproteinases (TIMPs 1 through 4), and the net proteolytic activity within a tissue is a function of the balance of MMPs/TIMPs.1 Numerous studies have shown that MMPs and TIMPs are expressed during vascular remodeling in the pathological conditions of atherosclerosis, restenosis, and aneurysm formation.2 Despite this burgeoning knowledge, we are still hampered by an incomplete understanding of the scope and the consequences of MMP/TIMP involvement in the pathogenesis of vascular disease. In atherosclerosis and restenosis, MMPs are produced by the major cell types inhabiting the plaque, vascular smooth muscle cells (SMCs), and leukocytes of the monocyte/macrophage and lymphocytic lineages. MMP-1, -2, -3, -9, -12, and -13 have been detected in plaques, along with the TIMP-1, -2, and -4.2 MMPs produced by SMCs clear a path for migration from media to intima by digesting the extracellular matrix, and SMC migration can be inhibited by administration of nonselective MMP inhibitors3–5 or transfection of the genes for TIMP-1 or TIMP-2 into the injured vessel wall.6,7 Macrophages produce abundant amounts of MMPs, which are used to invade through the endothelium and into the atherosclerotic plaque.8 MMPs are colocalized with macrophages in the core and shoulders of established plaques, areas that are very susceptible to the complications of erosion and rupture.9 MMPs are also expressed by inflammatory cells found in abdominal aortic aneurysms,10 and experimental studies using rat and mouse models point to a causal role for the MMPs in the pathogenesis of aneurysm.11,12 A great deal of effort in vascular biology has centered on the hypothesis that inhibiting MMP activity will reduce plaque volume by inhibiting the migration of SMCs and macrophages into the plaque and prevent the later complications of plaque rupture and aneurysm formation. However, the mechanisms of MMP action in complex models of atherosclerosis are largely unknown. With the advent of transgenic technology, better models of atherosclerosis have been developed, including the cholesterol-fed ApoE-null mouse, which is characterized by elevated circulating lipoproteins, and the development of lipid-rich plaques containing inflammatory macrophages and lymphocytes.13 In an article published in this issue of Circulation Research, Silence et al14 have uncovered dual roles of MMPs in the ApoE-null mouse model of atherosclerosis. Surprisingly, they found that deletion of the TIMP-1 gene resulted in reduction of plaque size in the ApoE-null mouse. TIMP-1 inhibits the activity of many MMPs, including the collagenases, gelatinases, and stromelysins. In the absence of the TIMP-1 gene, there was an increase in the number of macrophages present in aortic intimal lesions. The authors postulate that increased MMP activity (predominantly MMP2), which colocalized with the macrophages, resulted in collagen degradation, thereby reducing plaque size. The reduction in plaque size was evident despite increased lipid accumulation in the lesions of the ApoE-null:TIMP-1–null mice. Unfortunately, the potential for aneurysm formation was substantially elevated in these mice, as evidenced by an increase in the frequency of disruptions in the internal elastic lamina. The results presented here seemingly contradict a central dogma in atherosclerosis—that increased MMP activity leads to the formation of a thicker neointima. However, we must remember that most of the earlier studies with MMP inhibitors used experimental models where SMC migration was the main, if not the only, determinant of intimal lesion formation. By contrast, a growing body of experimental evidence from murine atherosclerosis models supports the postulate that increased macrophage-derived MMP activity may limit plaque progression. For example, plaque size and collagen content were greater in mice with double knockout of the MMP-3 and ApoE genes, compared to ApoE-null littermate controls.15 Consistent with this, plaque size, lipid deposition, and collagen content were reduced in ApoE-null mice that overexpressed MMP-1 in the macrophages.16 Taken together with results from the present study, this suggests that the activity of several plaque MMPs may actually be antiatherogenic. In this light, it is interesting to note that polymorphisms in the human MMP-3 promoter leading to decreased expression of this gene have been correlated with an increased incidence of atherosclerosis.17 Further work will be necessary to determine the full spectrum of antior proatherogenic activities of the many MMPs that are expressed in atherosclerosis. Another caveat is that the previous studies using knockout mice address only plaque progression. By contrast, increasing circulating TIMP-1 levels in the ApoEnull mouse induced the regression of pre-established leThe opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association. From the Department of Laboratory Medicine and Pathobiology, University of Toronto, Ontario, Canada. Correspondence to Dr Michelle P. Bendeck, Dept of Laboratory Medicine and Pathobiology, University of Toronto, Medical Sciences Building, 1 King’s College Cir, Room 6217A, Toronto, Ontario M5S 1A8, Canada. E-mail [email protected] (Circ Res. 2002;90:836-837.) © 2002 American Heart Association, Inc.