Molecular Bases of the Acute Coronary Syndromes.

HomeCirculationVol. 91, No. 11Molecular Bases of the Acute Coronary Syndromes Free AccessResearch ArticleDownload EPUBAboutView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessResearch ArticleDownload EPUBMolecular Bases of the Acute Coronary Syndromes Peter Libby Peter LibbyPeter Libby From the Vascular Medicine and Atherosclerosis Unit, Cardiovascular Division, Department of Medicine, Brigham and Women’s Hospital, Boston, Mass. Search for more papers by this author Originally published1 Jun 1995https://doi.org/10.1161/01.CIR.91.11.2844Circulation. 1995;91:2844–2850The acute coronary syndromes, including unstable angina and acute myocardial infarction, currently constitute a major preoccupation of clinical cardiology. This century has witnessed a remarkable evolution in our clinical concepts of these syndromes. Herrick1 described the survival of patients with acute coronary thrombosis early in the century. The introduction of the ECG led to major clinical advances in the definition of acute myocardial infarction during the first half of this century and furnished the basis of modern coronary care. In the latter half of this century, the advent of coronary arteriography permitted definition in the living patient of coronary stenoses due to atherosclerosis. The introduction of this diagnostic technique allowed the development of rational treatment modalities such as coronary artery bypass surgery and, subsequently, percutaneous transluminal coronary angioplasty. Until recently, it seemed that we had achieved a firm understanding of the pathophysiology of human coronary artery disease and had devised appropriate modes of therapy for its major manifestations. Yet, recent clinical data suggest that we still have much to learn about the pathophysiology of the acute coronary syndromes. Does the Angiogram Mislead Us About the Propensity of Plaques to Cause Acute Coronary Syndromes? Bypass surgery and angioplasty aim to restore blood flow to sites beyond hemodynamically significant stenoses in the coronary arteries. These revascularization therapies effectively relieve angina pectoris in many cases (although often not permanently). Quite naturally, the availability of these modalities led the cardiology community to focus on high-grade coronary stenosis, visible by angiography, as the critical issue in coronary heart disease. Much of the basis of contemporary cardiology and cardiac surgery rests on the axiom: the greater the stenosis, the greater the risk of a clinical event such as myocardial infarction or unstable angina pectoris. Evolving Concepts of the Pathogenesis of the Acute Coronary Syndromes However, data emerging from clinical and pathological studies over the past decade have occasioned a reassessment of this central dogma of clinical cardiology.2 First, the use of thrombolytic therapy in acute myocardial infarction became widespread in the wake of the GISSI study in 1986.3 Angiographic studies performed after thrombolysis during acute myocardial infarction led to the surprising finding that the atherosclerotic lesions that gave rise to the occlusive thrombus did not cause high-grade stenoses in many cases. Assessment of angiograms obtained before acute myocardial infarctions and those obtained during the infarction corroborated this concept that the lesions most likely to precipitate an infarct-provoking thrombosis often did not appear highly stenotic by angiography.4567Another line of clinical evidence suggested dissociation between the degree of stenosis and coronary events. In the past decade, a number of “regression trials” tested the hypothesis that lipid-lowering regimens would reduce the degree of high-grade coronary stenoses. The angiographic results showed disappointingly minimal effects on established stenotic lesions. Yet, these studies revealed a consistent and resounding decrease in acute clinical coronary events.89Meanwhile, state-of-the-art pathological studies using perfusion fixation of freshly obtained material provided new evidence buttressing the concept that rupture of atherosclerotic plaques precipitates the formation of the occluding thrombus that causes acute myocardial infarction.10 The elegant pathological studies of Davies and colleagues1011 also sought evidence for plaque disruption in hearts from patients dying of noncardiac causes. They documented evidence for plaque disruption in these patients even without overt symptoms of coronary disease or acute myocardial infarction. These results suggested that not all disruptions of atherosclerotic plaques lead to clinically apparent or symptomatic events. Such subclinical episodes of plaque disruption with local thrombin activation and subsequent healing may indeed represent a major pathway for progression of atherosclerotic lesions. Taken together, these new results suggest that while angiographically severe coronary artery disease clearly correlates with the propensity to develop or succumb from acute myocardial infarction, the presence of the severe stenoses may merely serve as a marker for the presence of angiographically modest or even inapparent, non–critically stenotic plaques actually more prone to precipitate acute myocardial infarction. Indeed, pathological studies have shown repeatedly in humans and in experimental animals that over much of its history, growth of an atherosclerotic plaque occurs by outward, abluminal expansion.121314 Hence, most obstructive plaques may pass through a phase that may last many years or even decades of so-called “remodeling” without encroaching on the arterial lumen. Only after the plaque burden approaches half of the luminal area does the plaque usually protrude into the lumen, becoming visible by angiography and capable of impairing flow. It thus appears that high-grade coronary arterial stenoses that can cause angina in the absence of superimposed vasospasm or thrombosis belong to a more advanced or “mature” stage in the life cycle of an atheroma. These recent clinical and pathological studies have focused the attention of those interested in the pathophysiology of the acute coronary syndromes on the plaque itself rather than the lumen as visualized by angiography. Characteristics of “Vulnerable Plaque”: Primacy of the Integrity of the Atheroma’s Fibrous Cap What is it about certain plaques that render them particularly susceptible to disruption and the ensuing acute manifestations such as myocardial infarction or unstable angina? Once again, the pathology of the lesions themselves suggests an answer. Atherosclerotic plaques typically consist of a lipid-rich core in the central portion of the eccentrically thickened intima (Fig 1). The lipid core is bounded on its luminal aspect by a fibrous cap, at its edges by the “shoulder” region, and on its abluminal aspect by the base of the plaque (area of detail, Fig 1). The central, lipid-rich core of the typical lesion contains many lipid-laden macrophage foam cells derived from blood monocytes. Once resident within the arterial wall, these cells imbibe lipid, which accounts for their foamy cytoplasm.15 Such lesional foam cells can produce large amounts of tissue factor, a powerful procoagulant that potently stimulates thrombus formation when in contact with blood.16For this reason, the integrity of the fibrous cap overlying this lipid-rich core fundamentally determines the stability of an atherosclerotic plaque. Rupture-prone plaques tend to have thin, friable fibrous caps.1718 Plaques not liable to precipitate acute myocardial events tend to have thicker fibrous caps that protect the blood compartment in the arterial lumen from potentially disastrous contact with the underlying thrombogenic lipid core (Fig 1). Biomechanical analyses demonstrate maxima of circumferential stress at sites of plaques prone to rupture.1719 Thus, mechanical forces concentrate on the fibrous cap, which must resist these high stresses to avoid rupture and the attendant risk of developing an acute coronary event. This stress-laden fibrous cap is all that stands between the blood and the thrombogenic lipid core of the lesion. Recognition of the primacy of the structure of the fibrous cap in determining the clinical activity of a given atherosclerotic lesion led us to study the cellular and molecular bases of the integrity of this critical region of the plaque. Vascular Smooth Muscle Cell: Guardian of the Integrity of the Plaque’s Fibrous Cap As its name implies, a dense, fibrous extracellular matrix characterizes the plaque’s fibrous cap. This extracellular matrix (formerly called connective tissue) contains several well-characterized macromolecules that account for the strength of the fibrous cap. The principal proteinaceous constituents of this extracellular matrix include the interstitial collagens and elastin. Various classes of proteoglycans also contribute to the extracellular matrix of atheroma but presumably contribute less than collagen or elastin to the rupture-resistance of this matrix. Among these macromolecules, collagen occupies an important position because of its abundance in the fibrous cap and its contribution to the structural integrity of this vulnerable “Achilles’ heel” region of the plaque. Collagen comes in many forms. Principally, the interstitial forms of fibrillar collagen concern us in the context of the plaque’s fibrous cap. Triple helical coils derived from specific procollagen precursors make up the types I and III collagen found in the fibrils of the plaque. Vascular smooth muscle cells can synthesize and assemble these macromolecules and furnish the bulk of both the collagenous and noncollagenous portions of the extracellular matrix of arteries. Evidence for Impaired Collagen Gene Expression at Vulnerable Sites of Human Atheroma For more than a dozen years, our laboratory has studied the role of inflammation in vascular pathobiology. In particular, we have explored the role of cytokines, protein mediators of inflammation and immunity in the pathogenesis of vascular diseases. Data from our laboratory and several others have provided evidence for the presence of various cytokines during different phases of atherogenesis.202122232425 In view of the importance of the collagenous matrix in determining plaque stability, we set out a number of years ago to investigate whether cytokines or growth factors implicated by us and others in atherosclerosis could regulate the synthesis of the interstitial forms of collagen that govern the integrity of the fibrous cap. Using standard techniques of biochemistry and molecular biology, we found that transforming growth factor-β and platelet-derived growth factor increased the mRNA and de novo protein synthesis of the precursors of the interstitial collagens types I and III (Fig 2).26 More notably, we found that a cytokine known as interferon gamma (IFN-γ) markedly decreased the ability of human smooth muscle cells to express the interstitial collagen genes both in the basal, unstimulated state and when exposed to transforming growth factor-β, the most potent stimulus for interstitial collagen gene expression known for these cells (Fig 2). This potent inhibition of interstitial collagen synthesis by human vascular smooth muscle cells did not result from a nonspecific or toxic effect of IFN-γ, since these cells remained viable and this cytokine selectively increased expression of another specific gene (HLA-DR, see below). How might this observation relate to the human atherosclerotic plaque? Among the cells found in human atherosclerotic plaques, only T lymphocytes can elaborate the cytokine IFN-γ. Work from Hansson’s laboratory some years ago provided evidence for IFN-γ production by chronically activated T cells within human atheroma.27 Exciting recent data from van der Wal et al28 in Becker’s laboratory provide further evidence connecting activated T cells and their products to plaque rupture. These investigators painstakingly studied the histology of 20 culprit lesions that provoked fatal acute myocardial infarctions. They found that T cells and macrophages predominated at sites of plaque disruption (frank rupture or superficial erosion of the fibrous cap). These workers further noted that neighboring smooth muscle cells and leukocytes expressed high levels of a transplantation antigen known as HLA-DRα, a finding they took to be an indicator of a state of “activation” of the smooth muscle cells. How could expression of a transplantation antigen by smooth muscle cells possibly pertain to stability of atherosclerotic plaques? We became interested some time ago in the ability of smooth muscle cells to express transplantation antigens in the context of understanding the pathophysiology of accelerated coronary atherosclerosis in transplanted hearts.29 This interest led us to explore the regulation of expression by smooth muscle cells of these molecules, which are important in the immune response. Of a wide variety of cytokines tested, only IFN-γ could induce the expression HLA-DRα in cultures of human vascular smooth muscle cells.30 Therefore, the finding of cells bearing this marker of activation indicates the presence of IFN-γ at the very sites of fatal plaque disruptions in humans. Further observations strongly support this concept. Rekhter and colleagues31 colocalized T lymphocytes with regions of collagen gene expression within human atherosclerotic lesions. Using rigorous morphometric analysis of histological sections of the human atheroma, they found an inverse correlation between the presence of T lymphocytes and interstitial collagen protein and mRNA. Taken together, these results concordantly suggest that chronic immune stimulation within atheroma leads to elaboration of IFN-γ from T cells, inhibiting collagen synthesis in vulnerable regions of the plaque’s fibrous cap. This mechanism provides a molecular explanation for impaired maintenance and repair of the collagenous meshwork in vulnerable plaques, rendering it weak and prone to rupture in the critical region of the plaque. Intact ability to synthesize collagen may sustain the ability of the fibrous cap to resist the concentration of mechanical forces in stable plaques. In addition to inhibiting collagen gene expression by human smooth muscle cells, IFN-γ can inhibit smooth muscle cell proliferation.3233 This cytokine can also contribute to activating the program of cell death, or apoptosis, in human vascular smooth muscle cells (Y-j. Geng and P. Libby, unpublished observations). These findings may help explain the relative paucity of smooth muscle cells in vulnerable regions of human atherosclerotic plaques. Moreover, IFN-γ can activate macrophage functions related to plaque vulnerability, including some of those described below. Curiously, proliferation of smooth muscle cells has dominated our thinking about the pathogenesis of atherosclerosis for decades. Smooth muscle cell growth may indeed contribute importantly to earlier phases of lesion development. Yet the present data, summarized above, suggested that the aspects of the biology of atheroma that actually lead to acute clinical manifestations depend on impaired smooth muscle cell growth and matrix elaboration rather than the contrary. This concept warrants consideration by those embarking on therapeutic quests seeking inhibitors of smooth muscle cell proliferation as treatments for atherosclerosis on the basis of relatively short-term experiments using simple animal models. Inhibition of smooth muscle cell proliferation in human patients might produce the undesired effect of destabilizing vulnerable regions of atherosclerotic plaques by the mechanisms described above. Cells Within Atherosclerotic Plaques Can Inducibly Express Genes Encoding Matrix-Degrading Enzymes In addition to impaired synthesis of collagen, accelerated degradation of collagen and other matrix components could contribute to weakening of the fibrous cap. The macromolecules that form the arterial extracellular matrix generally exhibit considerable metabolic stability. Collagen usually turns over quite slowly in arterial and other tissues. The triple helical structure of fibrillar collagens strongly resists attack by most types of proteolytic enzymes. However, certain enzymes specialize in catabolism of the extracellular matrix. These enzymes doubtless function importantly in normal development, morphogenesis, and wound healing by allowing cells to migrate through the omnipresent extracellular matrix of tissues. They also may contribute to various pathological states such as joint destruction in rheumatoid arthritis or metastasis of malignant cells. In particular, members of a superfamily of such enzymes known as the matrix metalloproteinases merit consideration in this regard. In contrast to the intracellular proteolytic enzymes found in organelles called lysosomes, the matrix metalloproteinases act extracellularly and at physiological pH. The matrix metalloproteinase superfamily includes interstitial collagenase, an enzyme specialized in the initial cleavage of the usually protease-resistant fibrillar collagens that confer strength upon the fibrous cap of the atheroma. Other matrix metalloproteinase family members (the gelatinases) catalyze further breakdown of collagen fragments. Stromelysins can activate other members of the matrix metalloproteinase family and can degrade a broad spectrum of matrix constituents, including the protein backbones of proteoglycan molecules. Stromelysin and one of the gelatinases (gelatinase B, or 92-kD gelatinase) can also break down elastin, an additional structurally important component of the vascular extracellular matrix. Two points merit emphasis in consideration of the potential roles of the matrix metalloproteinases in disruption of atherosclerotic plaques. First, these enzymes require activation from proenzyme precursors to attain enzymatic activity. This type of tight control resembles that found in other critical biological regulatory cascades such as blood coagulation, fibrinolysis, and complement. Also, reminiscent of other protease cascades involved in regulation of key biological processes, ubiquitous inhibitors known as tissue inhibitors of metalloproteinases (TIMPs) hold the activity of these enzymes in check under usual circumstances. In their basal state, human vascular smooth muscle cells express both major isoforms of TIMPs (TIMP 1 and TIMP 2) and one form of gelatinase (gelatinase A, or 72-kD gelatinase).34 Biochemical experiments indicate that this constitutively expressed gelatinase probably exists as a complex with its corresponding inhibitor, TIMP 2, rendering it inactive under normal conditions in vivo. Exposure to inflammatory cytokines such as interleukin-1 or tumor necrosis factor induces smooth muscle cells to express interstitial collagenase, a form of gelatinase not expressed in the basal state (gelatinase B, the form of gelatinase that also exhibits considerable elastolytic activity), and stromelysin.34 Treatment with these cytokines does not alter the expression of TIMPs by these cells. In this manner, cytokines known to localize in atherosclerotic lesions can produce a net increase in the capacity of human smooth muscle cells to degrade constituents of the arterial extracellular matrix. However, as previously noted, regions of the plaque’s fibrous cap particularly prone to disruption contain relatively few smooth muscle cells but abundant macrophages and T cells.3536 For this reason, we tested whether macrophage-derived foam cells can express these matrix-degrading enzymes. Our experimental strategy involved isolation of lipid-laden macrophages from atherosclerotic lesions produced experimentally by diet and balloon injury in the rabbit aorta. Such lesion-derived foam cells expressed stromelysin and interstitial collagenase both in situ and in vitro. In contrast, alveolar macrophages from the same animals, exposed to the same degree of hyperlipidemia, did not display autonomous expression of these matrix metalloproteinases.37 What activates these macrophage foam cells to synthesize these matrix-degrading proteinases? Likely candidates include locally acting cytokines such as IFN-γ, tumor necrosis factor, interleukin-1, or macrophage colony–stimulating factor. Human atherosclerotic plaques can contain each of these candidate stimuli of matrix metalloproteinase expression by lesional foam cells. Evidence for Increased Degradation of the Extracellular Matrix in Vulnerable Regions of Human Atherosclerotic Plaque To extend our in vitro and animal experiments to the human situation, we studied a number of human atherosclerotic plaques and uninvolved arteries. Just as in the culture dish, smooth muscle cells in the nonatherosclerotic arteries express TIMPs 1 and 2 and gelatinase A (as already noted, probably in an inactive complex with its inhibitor TIMP-2). In human atherosclerotic plaques, however, smooth muscle cells, T cells, and macrophages all expressed interstitial collagenase, gelatinase B, and stromelysin.38 Henney and coworkers39 had previously shown evidence for stromelysin mRNA expression by cells within atheroma as well. Moreover, the endothelial cells overlying atheroma, but not normal vessels, contained interstitial collagenase. In addition, the endothelial cells of the plaque’s rich microvasculature expressed this metalloproteinase, which may facilitate new capillary penetration through the dense extracellular matrix of the complicated plaque. However, the mere presence of immunostainable metalloproteinases does not provide assurance that they exist in an active form. The antibodies available do not distinguish the activated forms of these enzymes from their inactive proenzyme or zymogen forms. Moreover, as already noted, the widely distributed tissue inhibitors of metalloproteinases could neutralize any of these enzymes that might become activated in the plaque and prevent their proteolytic action. We tested for the presence of a net excess in matrix-degrading activity by laying frozen sections of human atherosclerotic plaques upon films of suitable substrates and evaluating lysis of these substrates by microscopy. This in situ zymographic technique revealed such activity, particularly in the shoulder region of a series of atherosclerotic plaques.38 These results actually demonstrated net excess of matrix-degrading activity in vulnerable regions of human atherosclerotic plaques, supporting the concept that excessive matrix degradation may contribute to the lability of atheroma (Fig 3).Other Contributors to the Unstable Coronary Syndromes The foregoing discussion has focused on the integrity of the extracellular matrix of the plaque’s fibrotic cap as a key to the understanding of the unstable coronary syndromes. Although these mechanisms probably play a prominent role in the pathophysiology of plaque rupture, other factors may contribute to the unstable coronary syndromes as well. Much recent research has established impaired responsiveness to endothelium-dependent vasodilators in diseased portions of human coronary arteries.40 Such findings suggest a propensity to vasospasm that may contribute to impaired flow in these vessels, particularly at sites of underlying stenoses. Also implicit in the above discussion, the precipitating event in most cases of acute myocardial infarction involves thrombotic occlusion of the vessel, usually at sites of plaque disruption. Indeed, thrombus formation figures prominently in our current concepts of both acute myocardial infarction and unstable angina. In this regard, the molecular bases of the acute coronary syndromes involve a critical intersection of three distinct but interrelated protease–protease-inhibitor cascades: thrombosis, fibrinolysis, and the matrix-degrading proteases (Fig 4). By way of illustration of the interconnections between these cascades, consider that the fibrinolytic enzyme plasmin can cleave interstitial collagenase from its latent zymogen to its active form. Incidentally, this point suggests that administration of plasminogen activators during acute myocardial infarction might actually promote destabilization of atheroma by promoting collagen degradation. As noted above, atherosclerotic lesions usually contain abundant plexi of microvessels.4142 These neovascular channels may themselves be prone to rupture within the plaque, much as the neovessels within the diabetic retina tend to form microaneurysms and hemorrhage. Some episodes of sudden plaque expansion may result from intraplaque hemorrhage due to rupture of the microvessels rather than a disruption of the fibrous cap of the plaque itself. Mechanisms of Stabilization of the Atherosclerotic Plaque: A New Therapeutic Target? The above discussion has summarized elements of a new understanding of the pathophysiology of the acute coronary syndromes based on episodes of plaque disruption rather than gradual progression to complete occlusion of fixed coronary stenoses. As noted above, bypass surgery or transluminal angioplasty provide rational and often effective therapies for these fixed, high-grade stenoses. However, these treatments do not address the nonstenotic but vulnerable plaque. It is of interest in this regard that despite the well-accepted benefit of coronary bypass surgery on anginal symptoms, this treatment aimed at severe stenoses does not prevent myocardial infarction.43 To reduce the risk of acute myocardial infarction, one must stabilize lesions to prevent their disruption, particularly the less stenotic plaques.44How might one achieve such a goal? The results of recent lipid-lowering trials provide a hint. The reductions in clinical events without substantial change in the degree of luminal stenosis could reflect a stabilization of the non–critically stenotic lesions.89 This stabilization might result from reducing the inflammatory stimuli provided by modified lipoproteins that could contribute to activation of lesional foam cells and T lymphocytes as described above (Figs 1 and 3). Accumulating physiological evidence in humans demonstrates that lipid-lowering and/or antioxidant therapy can improve the vasomotor response to endothelium-dependent vasodilators such as acetylcholine.4546 A reduction in inflammation induced by modified lipids might contribute to this salutory effect of pharmacological lipid lowering. The recent demonstration that cholesterol lowering with a 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor decreases cardiac and total mortality under conditions not expected to reduce substantially preexisting high-grade stenoses illustrates the potential of an intervention that could yield plaque stabilization.47 We eagerly await the results of other studies that will include patients with lower cholesterol levels more commonly encountered in our coronary disease populations, and the results of primary prevention studies will also prove quite interesting in this connection. Conclusions Atherosclerosis shares many characteristics of a chronic inflammatory process. Many stimuli may incite this ongoing reaction. As alluded to above, many current data suggest that lipoproteins or their derivatives (eg, oxidized lipoproteins) contribute to smoldering inflammation in atherosclerotic plaques.48 Other potential inciting stimuli may include infectious agents and autoantigens such as heat-shock proteins, among other possible activators of T cells.49 The activated T cell secretes IFN-γ, which may impair collagen synthesis. Macrophages and smooth muscle cells activated by inflammatory mediators such as the cytokines can elaborate enzymes that weaken the connective tissue framework of the plaque’s fibrous cap. Reduction of inflammation should render atherosclerotic plaques more stable. The present observations even provide a potential cellular and molecular mechanism for the marked reduction in acute coronary events observed with lipid-lowering therapies, as discussed above. In any case, the foregoing discussion illustrates how new clinical observations have caused us to reevaluate our traditional concepts of the pathogenesis of the acute coronary syndromes. New insight from clinical and pathological studies reviewed above highlighted the importance of probing the biological basis of the unstable plaque to understand the mechanisms that underlie the acute manifestations of atherosclerosis that occupy a large portion of the efforts of modern cardiologists. The results of such clinical investigations inspired basic studies of the cellular and molecular mechanisms of the acute coronary syndromes such as those described here. The new insights from these “bedside to bench” studies should in turn hasten development of strategies aimed at plaque stabilization. The ultimate goal of this basic research is to return from the bench to the bedside with novel therapies and new understanding to help prevent the development of unstable coronary disease in our patients. Download figureDownload PowerPoint Figure 1. Color diagram showing comparison of the characteristics of “vulnerable” and “stable” plaques. Vulnerable plaques often have a well-preserved lumen, since plaques grow outward initially. The vulnerable plaque typically has a substantial lipid core and a thin fibrous cap separating the thrombogenic macrophages bearing tissue factor from the blood. At sites of lesion disruption, smooth muscle cells (SMCs) are often activated, as detected by their expression of the transplantation antigen HLA-DR (see text). In contrast, the stable plaque has a relatively thick fibrous cap protecting the lipid core from contact with the blood. Clinical data suggest that stable plaques more often show luminal narrowing detectable by angiography than do vulnerable plaques.Download