doi: 10.1161/hq1201.100220
© 2001 American Heart Association, Inc.
Brief Reviews |
Immune Mechanisms in Atherosclerosis
From the Center for Molecular Medicine and the Department of Medicine at Karolinska Hospital, Karolinska Institute, Stockholm, Sweden.
Correspondence to Göran K. Hansson, Center for Molecular Medicine and Department of Medicine, Karolinska Hospital, Karolinska Institute, SE-17176, Stockholm, Sweden. E-mail Goran.Hansson{at}cmm.ki.se
 |
Abstract |
|---|
 Top Abstract IntroductionImmune Cells in the...Recruitment of Immune Cells...Cellular Immunity of...Immune Specificity in the...Evidence That Immunity Affects...Can We Prevent Atherosclerosis...References |
|---|
Atherosclerosis is an inflammatory disease. Its lesions are filled with immune cells that can orchestrate and effect inflammatory responses. In fact, the first lesions of atherosclerosis consist of macrophages and T cells. Unstable plaques are particularly rich in activated immune cells, suggesting that they may initiate plaque activation. We have seen a rapid increase in the understanding of the mechanisms that govern the recruitment, differentiation, and activation of immune cells in atherosclerosis. Experimental research has identified several candidate antigens, and there are encouraging data suggesting that immune modulation as well as immunization can reduce the progression of the disease. This review provides an overview of our current understanding of the role of immune mechanisms in atherosclerosis.
Key Words: atherosclerosis • pathophysiology • cell biology • cytokines
|
Introduction |
|---|
|
TopAbstractIntroductionImmune Cells in the...Recruitment of Immune Cells...Cellular Immunity of...Immune Specificity in the...Evidence That Immunity Affects...Can We Prevent Atherosclerosis...References |
|---|
The atherosclerotic lesion contains large numbers of immune cells, particularly macrophages and T cells. Furthermore, the disease is associated with systemic immune responses and signs of inflammation. Histopathological and clinical investigations point to inflammatory/immune activation of plaques as a cause of acute coronary syndromes, and seroepidemiological studies have suggested links between atherosclerosis and microbial infections. During recent years, experiments in gene-targeted mice have provided mechanistic evidence in support of the hypothesis that immune mechanisms are involved in atherosclerosis. This review will provide an update on immune mechanisms in atherosclerosis, with particular focus on mechanistic studies.
|
Immune Cells in the Lesions of Atherosclerosis |
|---|
|
TopAbstractIntroductionImmune Cells in the...Recruitment of Immune Cells...Cellular Immunity of...Immune Specificity in the...Evidence That Immunity Affects...Can We Prevent Atherosclerosis...References |
|---|
Adaptive (ie, antigen-specific) immune reactions are initiated when a macrophage (or a dendritic cell of the macrophage lineage) displays a surface complex consisting of an antigenic peptide bound to a major histocompatibility complex (MHC) protein to a neighbor T cell. This can elicit T-cell activation, cytokine secretion, cytotoxicity, antibody production, and many other components of an immune reaction. There is now good evidence that atherosclerosis is associated with immune reactions and that antigen presentation occurs in atherosclerotic lesions. The cellular composition of an atheroma is illustrated in Figure 1. It has been known for many years that monocyte-derived macrophages are present in large numbers in atherosclerotic lesions.1–7 The discovery of the scavenger receptor pathway for uptake of modified lipoproteins8–10 provided an explanation for the finding that most foam cells bear macrophage differentiation markers.
|
The other main immune cell of atherosclerotic lesions, the T cell, remained undetected for a long time. It was observed that many vascular smooth muscle cells (SMCs) in human lesions express human leukocyte antigen (HLA)-DR, a cell surface protein that is induced by the T-cell cytokine interferon-
(IFN-).11 By using T-cell-specific CD3 antibodies, it could be established that T cells are abundant in human plaques.12 A large proportion of them are in an activated state13–16; ie, they express adhesion molecules and other surface molecules, secrete cytokines, and may proliferate. The activation of T cells and macrophages leads to a cascade of cytokines that induce an inflammatory state. With the cDNA cloning and production of recombinant cytokines, it became possible to study their effects in the context of vascular biology. Such studies revealed that vascular endothelial cells and SMCs are important targets for inflammatory cytokines and that they are capable of producing significant amounts of cytokines themselves on stimulation.17,18
Studies during the last decade have identified other types of immune cells in atherosclerotic plaques. Among them, dendritic cells are "professional" antigen-presenting cells that may be particularly important in the initiation of immune responses to plaque antigens.19 Mast cells are immune effector cells with a capacity to modify lipoproteins and digest matrix components; they could be important in plaque activation and rupture.20 All these different cells may be involved in the initiation, progression, and/or development of complications to atherosclerosis. At present, there is evidence for pathogenic roles of monocyte-derived macrophages and T cells.
|
Recruitment of Immune Cells to the Vessel Wall |
|---|
|
TopAbstractIntroductionImmune Cells in the...Recruitment of Immune Cells...Cellular Immunity of...Immune Specificity in the...Evidence That Immunity Affects...Can We Prevent Atherosclerosis...References |
|---|
Leukocyte Adhesion to the Endothelium
During the initiation of atherosclerosis, mononuclear leukocytes, ie, monocytes and T cells, are recruited to the vessel wall across an intact endothelium. This requires activation of the endothelium to express leukocyte adhesion molecules. These cell surface proteins are mainly regulated transcriptionally in a process that involves nuclear factor (NF)-
B,21 a transcription factor that is transactivated when proinflammatory cytokines ligate their receptors on the endothelial surface. Both E-selectin and 2 imunoglobulin-like adhesion molecules, intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1), can be induced in this way.
VCAM-1 can be induced not only on cytokine stimulation but also in response to other proinflammatory macromolecules. Lipopolysaccharides of Gram-negative bacteria activate NF-B by ligating toll-like receptors (TLRs), which transduce NF-B activation through a kinase cascade similar to the one initiated by interleukin-1 (IL-1). In addition, lysophosphatidylcholine and other oxidized phospholipids can induce VCAM-1 expression in endothelial cells.22,23 Interestingly, such phospholipid species are generated during lipoprotein oxidation. Their activity may explain the increased adhesive properties of endothelial cells exposed to oxidized LDL (oxLDL)24 (Figure 2). In vivo studies have shown that hypercholesterolemia rapidly leads not only to lipoprotein deposition and oxidation in the intima but also to VCAM-1 expression on the luminal aortic endothelium.25,26 This provides a direct link between hypercholesterolemia and inflammatory activation of the vessel wall.
|
ICAM-1 is induced in endothelial cells through a cytokine-dependent pathway, but its expression is also promoted by hemodynamic stress. The latter seems to be due to activation of the shear stress response element in its promoter.27 This probably explains the finding of ICAM-1 expression in aortic regions where the endothelium is exposed to variations in shear stress caused by blood flow.27–29 It is likely that the disturbed flow encountered during hypertension can increase ICAM-1 expression. Thus, both hypercholesterolemia and hypertension, 2 major risk factors for atherosclerosis, increase the recruitment of leukocytes to the artery.
Adhesion is a multistep process that starts with leukocyte rolling on the endothelial surface. This is due to selectin ligation, whereas the subsequent firm adhesion depends on interactions between immunoglobulin-like molecules (VCAM-1, ICAM-1, and others) on the endothelium and integrins on the leukocyte surface.30 Very late-activation antigen-4 (VLA-4), a major counterreceptor for VCAM-1, is expressed by monocytes and lymphocytes but not by granulocytes.31 This explains the selective recruitment of mononuclear cells to the arterial intima during early atherosclerosis.32 Immunopharmacological blockade of ICAM-1 and VCAM-1 has been shown to inhibit fatty streak development33 (Table 1). The importance of selectin interactions is illustrated by the finding that mice deficient in E-selectin and P-selectin develop fewer fatty streaks when bred into an atherosclerosis-prone mouse strain, the apoE-knockout (KO) mouse34,35 (Table 1).
|
Chemotactic Transmigration
In the classic inflammatory response, adhesion is followed by transmigration of the leukocytes through the endothelial layer and into the intima. This is governed by chemotactic factors produced in the subendothelial layer. Several studies of hypercholesterolemic rabbits and human samples show that the complement cascade is activated subendothelially during hypercholesterolemia.36,37 This leads to release of small, proteolytic peptide fragments of complement proteins. Such fragments include C5a, which is strongly chemotactic for monocytes and may be important for the recruitment of these cells to the intima.
Several chemotactic cytokines called chemokines are produced by endothelial cells, SMCs, and intimal macrophages during lesion formation. The best-characterized of these chemokines is monocyte chemotactic protein-1 (MCP-1), which can be induced by complement activation or cytokines38,39 and promotes recruitment of monocytes and T cells.40–42 MCP-1 is expressed in significant amounts in all stages of atherosclerosis.40,43 When atherosclerosis-prone mice such as apoE-KO or LDL receptor (LDLR)-KO mice were bred with mice deficient in MCP-1 or its receptors, CCR2, lesion formation was reduced drastically44–46 (Table 1). This demonstrates that chemokine-dependent migration of mononuclear cells into the intima is an important phenomenon in atherogenesis (Figure 2).
|
Cellular Immunity of Atherosclerotic Lesions |
|---|
|
TopAbstractIntroductionImmune Cells in the...Recruitment of Immune Cells...Cellular Immunity of...Immune Specificity in the...Evidence That Immunity Affects...Can We Prevent Atherosclerosis...References |
|---|
Macrophages: Scavenging, Secretory, and Antigen-Presenting Cells
Macrophages are major players in inflammation and innate (ie, antigen-independent) immune responses. These actions largely depend on their capacity to produce free oxygen radicals, proteases, complement factors, and cytokines. Importantly, the macrophage may also initiate adaptive immune responses by presenting foreign antigens to T cells. All these activities may be important in atherogenesis.
The differentiation from monocyte to macrophage is governed by macrophage colony stimulating factor (M-CSF), a cytokine that is produced not only by macrophages but also by vascular and stromal cells. Lack of M-CSF prevents macrophage differentiation, the consequences of which can be observed in many different organs. A striking phenotype of the M-CSF-deficient op/op mouse is the lack of osteoclasts, which prevents bone remodeling and leads to the disease osteopetrosis. The op/op phenotype is also characterized by a lack of macrophages in tissues. When op/op mice were mated with apoE-KO mice, the offspring developed very little atherosclerosis despite high cholesterol levels in the blood.47 This demonstrates that macrophage differentiation is necessary for atherosclerosis.
Macrophage uptake of modified lipoproteins by way of scavenger receptors is tightly regulated by cytokines (Table 2 and Figure 1). IFN-, tumor necrosis factor-
(TNF-), and IL-6 downregulate scavenger receptor-A (SR-A),48–50 IL-4 upregulates CD36,51 and the lectinlike, oxidized LDL receptor (LOX-1) is controlled by TNF- and transforming growth factor-ß (TGF-ß).52 That cytokines regulate scavenger receptors probably reflects important roles for these receptors in the innate host defense. Additional support for this notion is the finding that SR-A-deficient mice are highly susceptible to infections by intracellular bacteria such as Listeria monocytogenes.53 However, SR-A can also internalize antigens, which are routed for presentation to T cells.54,55 By binding foreign antigens and initiating their transfer to antigen-processing as well as degradative compartments, scavenger receptors may be important links between innate and adaptive immunity.
|
Components of oxLDL may not only be degraded or processed for antigen presentation but also may activate the macrophage itself. This was initially demonstrated in studies of peripheral human monocytes and macrophage cell lines56,57 and later confirmed in other cell culture systems. This effect may be mediated by proinflammatory lipids acting on specific receptors. OxLDL contains platelet-activating factor (PAF)-like lipids, which are strongly proinflammatory and activate macrophages as well as endothelial cells.58–61 In addition, oxLDL components might bind to signaling surface receptors such as TLRs or nuclear receptors such as lipid X receptors (LXR). Both of these receptor families are important in the regulation of macrophage function: TLRs mediate macrophage activation in response to bacterial toxins,62 whereas LXRs regulate the ATP binding cassette-1 (ABCA1) transporter and other aspects of cholesterol metabolism in the macrophage.63
T-Cell Types and Their Functions
Important Role for Th1 Cells and Their Cytokines
Most of the T cells in atherosclerotic lesions are CD3+ CD4+ T-cell receptor (TCR) ß+ cells12,15 (Table 3). This implies that they recognize protein antigens presented to them by macrophages after uptake and processing through the endosomal pathway (Figure 3). Such cells represent approximately two thirds of all CD3+ T cells in advanced human lesions and most of the T cells in lesions of apoE-KO mice. Many of the cells exhibit surface markers indicative of a population of memory cells in a state of chronic activation.15 They are largely of the T-helper (Th1) subtype, which secretes IFN-, IL-2, and TNF- and -ß, which cause macrophage activation, vascular activation, and inflammation.64 At least 2 important stimuli for Th1 differentiation are present in the atherosclerotic plaque. The cytokine IL-12, which is produced by many lesion cells, is an important stimulus for Th1 differentiation.65 Osteopontin, also called early T-lymphocyte activation protein-1 (Eta-1), is needed for Th1 responses and promotes IL-12 expression and granuloma formation.66 It is expressed by macrophages, endothelial cells, and SMCs in plaques67,68 and may be important for local immunity as well as for mineralization.
|
|
IFN- is an important immune-activating cytokine that can prime macrophages for activation and induce inflammatory responses such as those observed in delayed-type hypersensitivity and granulomatous lesions (Table 2). It has been detected in human plaques, both on the mRNA and protein levels.13,64 Cell culture studies have shown it to be a powerful growth inhibitor for endothelial cells and SMCs.69,70 In addition, IFN- induces expression of secretory phospholipase A2, which can lead to production of inflammatory lipid mediators such as eicosanoids, lysophosphatidylcholine, and PAF.71
In vivo studies in rats have shown that the proliferative response of SMCs after arterial injury is inhibited by IFN-,72 which also inhibits smooth muscle contractility and collagen synthesis.70,73 It was therefore proposed that IFN--producing T cells could play an important role in plaque destabilization by reducing the fibrous cap.74,75 Histopathological findings of activated T cells at sites of rupture in culprit lesions support this notion.76
Further support for a proatherogenic role came from studies of compound KO mice lacking the IFN- receptor as well as apoE. These mice had an 
60% reduction in atherosclerosis, supporting the notion that IFN- is a strongly proatherogenic cytokine77 (Table 4). The lack of IFN- signaling affected the lipoprotein profile (reduced apoA-IV) as well as the vessel wall (increased collagen), implying that both systemic and local effects could be important for the antiatherogenic action. However, IFN- was recently shown to aggravate transplant vasculopathy by its local action on the vessel wall.78 Even in severe combined-immunodeficient (SCID) mice, IFN- induced vascular pathology in xenografts, indicating that its direct vascular actions are sufficient to promote disease. Finally, recent experiments have shown that administration of this cytokine accelerates atherosclerosis in apoE-KO mice.79 All these data imply that IFN- is a proatherogenic cytokine.
|
Proinflammatory Cytokines: Mediators of Innate and Th1 Responses
TNF- and IL-1 are also present in human lesions.16,80,81 Similarly to IFN-, they affect smooth muscle proliferation, but their effects are more complex and indirect. IL-1 not only stimulates SMC replication by inducing an autocrine platelet-derived growth factor loop but also inhibits proliferation by inducing the growth-inhibitory autacoid, prostaglandin E1.82,83 The net effect on SMC growth therefore depends on the precise conditions in experimental systems and remains controversial in the in vivo situation. TNF- and IL-1 are powerful inducers of local inflammation in blood vessels and elsewhere (Table 2). For instance, they stimulate further activation of macrophages, induce secretion of matrix metalloproteinase-9,84 and promote expression of leukocyte adhesion molecules (see above). IL-1 is also an important costimulatory factor for T-cell activation and TNF-, a proapoptotic cytokine.
The metabolic effects of IL-1 and TNF- are even more striking than those of IFN- (Table 2). They are powerful inhibitors of lipoprotein lipase, which leads to increased systemic levels of VLDL and hypertriglyceridemia.85,86 In addition, they exert important effects on glucose and energy metabolism. High-dose stimulation with TNF- therefore results in cachexia. It is possible that much lower TNF- levels lead to metabolic changes of the kind observed in the metabolic syndrome, such as hypertriglyceridemia, redistribution of adipose tissue, and reduced insulin sensitivity.87–89 Interestingly, this syndrome is considered an important risk factor for atherosclerotic cardiovascular disease.
The proinflammatory cytokine pathway is a cascade that involves activation by TNF- and IL-1 to IL-6 expression90 and also to secretion of pentraxins such as C-reactive protein (CRP).91 This cascade is amplified in each step. For instance, IFN- produced by Th1 cells stimulates macrophages to secrete IL-1. The latter cytokine stimulates SMCs to make copious amounts of IL-6,92 a mediator that is present in atherosclerotic lesions64,93 and that may also be induced by the terminal complement complex C5b-994 (Table 2). CRP and other pentraxins may not only be markers of inflammation but also exert direct effects on leukocyte recruitment and apoptosis in the vessel wall.95,96
Patients with unstable coronary syndromes have elevated levels of IL-6, CRP, and pentraxin-3,97–99 which may be due to increased inflammatory activity in their culprit lesions. Importantly, even a modestly elevated CRP level is a risk factor for coronary heart disease in healthy, middle-aged men.100 This suggests that the modest inflammatory activity of "silent" plaques results in a cascade leading to CRP production and also accelerates disease progression.
Th2, Th3, and More: Are Anti-inflammatory Cytokines Antiatherosclerotic?
Th2 cytokines such as IL-4, IL-5, and IL-10 are less abundant than cytokines of the Th1 type in end-stage human lesions.64 In cell culture systems, IL-4 and IL-10 inhibit inflammatory genes such as inducible nitric oxide (NO) synthase101 and cyclooxygenase 2,102,103 but IL-4 also upregulates the scavenger receptor CD36 (Table 2). In apoE-KO mice, Th2 cytokines can be found in atherosclerotic lesions only in conditions of excessive hypercholesterolemia, with serum cholesterol levels of 40 to 50 mmol/L.104 Th1 and Th2 cytokines are cross-regulatory, IL-10 being an inhibitor of the Th1 pathway and IL-12 a Th2 inhibitor. The low level of Th2 activity in lesions may be at least partly due to local IL-12 secretion.65 There may be a corresponding IL-10 damper of the Th1 response, because IL-10-containing areas of human plaques exhibit reduced apoptosis and expression of inflammatory genes such as cytokine-inducible NO synthase.105 Interestingly, IL-10-/- C57BL/6J mice exhibit increased fatty streak development,106,107 and IL-10-/- x apoE-/- compound-KO mice show increased plaque activation and coronary thrombosis (G. Caligiuri, G.K. Hansson, and A. Nicoletti, unpublished observations, 2001). However, IL-10 appears to be a cytokine secretion-inhibiting cytokine with a broader set of targets than Th1 cells alone, and its effects may not be equivalent to those of Th2 activity.
A third T-helper cell called Th3 was recently described.108 It produces TGF-ß on activation. This cytokine stimulates collagen synthesis and is fibrogenic. Furthermore, it is strongly anti-inflammatory, as illustrated by the fact that TGF-ß-KO mice die of fulminant inflammation by the age of 6 weeks.109 TGF-ß and its receptor are present in the atherosclerotic lesion,64,110 where it may act to stimulate collagen synthesis and cap formation and to dampen inflammation. Several different cell types, including macrophages, SMCs, and Th3 cells, can express TGF-ß, which might be important for plaque stabilization. The SMC growth-promoting action of T cells that release heparin-binding growth factors might add to this effect.111
In conclusion, although Th1 cytokines are abundant in lesions, it seems less likely that atherosclerosis will be considered a strict Th1 disease. It might even be speculated that different phases of this chronic disease are promoted by different effector pathways. Clearly, the development and regulation of Th pathways in atherosclerosis require further studies.
CD8 Cells and 
T Cells: Additional Players in the Plaque Orchestra?
CD8+ T cells are found at varying proportions in human lesions (Table 3). Most of the cytotoxic activity associated with T cells is found in this subpopulation (Tc cells), which is activated by cells that express proteasomally processed peptide fragments of nascent proteins in the context of MHC class I. It was recently shown that expression of a "foreign" antigen on vascular SMCs can lead to cytotoxic attack by CD8+ T cells.112 In an apoE-deficient mouse, this results in significant aggravation of atherosclerosis. Thus, it is possible that CD8+ T cells cause some of the widespread apoptosis that is associated with atherosclerosis.113 However, apoptosis can also be induced by reactive oxygen and nitrogen species, which are generated by macrophages and SMCs on stimulation with proinflammatory cytokines.114 Therefore, activation of CD4+ as well as CD8+ T cells can lead to cell death in lesions. In addition, CD8+ T cells may respond to antigens not only by cytotoxic activity but also by secretion of cytokines in a manner parallel to that observed for Th1 and Th2 cells. Functional Tc1 and Tc2 subsets of CD8+ T cells seem to be important for the tissue response in mycobacterial infections, but it is not known whether they play a role in atherosclerosis.
A third type of T cell has a CD3+ 4- 8- phenotype and expresses TCRs that are rather than ß dimers. Such T cells are important in mucosal immunity and in the recognition of complex lipids as antigens. TCR binds such antigens, which are presented on CD1 rather than MHC proteins. The capacity of T cells to recognize lipids makes them potentially interesting in atherosclerosis. T cells and CD1-expressing cells have been found at varying proportions in arterial inflammatory lesions115 and atherosclerotic lesions,116,117 but their role remains unclear.
Other Immune Cells May Also Participate in the Local Immune Response
The 2 other types of lymphocytes, B cells and natural killer (NK) cells, are less frequent in atherosclerotic lesions than T cells. Relatively few B and NK cells are found in advanced human lesions, but prominent B-cell infiltrates may occur in the adventitia and the periadventitial connective tissue.118 They can develop into pathological perivascular lymphoid aggregates, a condition termed periaortitis.119 In lesions of hypercholesterolemic rabbits and mice, B cells are relatively abundant, and clones of immunoglobulin-producing cells can be found.120,121 In humans as well as animals, IgG accumulation is prominent in lesions.118,122 Some of these IgGs may be specific for antigens present in the tissue. Although a substantial proportion of the IgG may have entered the lesion by filtration, there is also evidence for local synthesis.120 Plasma cells, which may secrete large amounts of IgG, are present in lesions and are the likely source of this locally produced IgG.
A different kind of hematopoietic cell, the mast cell, has been identified in atherosclerotic lesions.20 Mast cells are less frequent than macrophages and T cells but may be important in plaque activation and acute coronary syndromes because they produce several proteases and accumulate at sites of plaque rupture.123–125 Factors released from mast cells may degrade the extracellular matrix126 and could also influence the functions of surrounding cells and modify locally deposited lipoproteins.127 Finally, the dendritic cell, which is a specialized member of the macrophage lineage, has been found in human and experimental atherosclerotic lesions.19,128 This cell performs a key role in immune activation because it is specialized on antigen presentation and appears to be the only cell that can activate naive T cells. It has a high migratory capacity and may "patrol" tissues such as the artery wall in search of antigens, which are endocytosed and transported to regional lymph nodes, where presentation of the antigen to naive and memory T cells, and hence, induction of adaptive immunity, can take place.
|
Immune Specificity in the Atherosclerotic Lesion |
|---|
|
TopAbstractIntroductionImmune Cells in the...Recruitment of Immune Cells...Cellular Immunity of...Immune Specificity in the...Evidence That Immunity Affects...Can We Prevent Atherosclerosis...References |
|---|
Restricted Heterogeneity Suggests Local Immune Activation
The antigen specificity of T cells is encoded in the sequence of their rearranged TCR genes, which determine the conformation of the antigen-binding site in the CDR3 domain of the TCR protein.129 Because the rearrangement process is stochastic, each nascent T cell carries a unique TCR gene and protein. On activation, the stimulated T cell divides to give rise to a clone of cells with identical specificities. The presence in tissue of a population of T cells with an identical TCR is therefore indicative of clonal proliferation and hence, of expansion of these particular T cells due to antigenic stimulation. Such clonal expansions are found in early lesions of apoE-KO mice.130 This finding is suggestive of local clonal proliferation of T cells, which is compatible with activation by local antigens. Interestingly, similar TCRs containing the Vß6 variable domain are frequently expressed by T cells that recognize oxidatively modified LDL, one of the candidate antigens in atherosclerosis (A. Nicoletti, G. Paulsson, and G.K. Hansson, unpublished observations, 2001).
In human lesions, the situation is more complex. A heterogeneous population of TCRs and therefore, of T cells is found in lesions,131 and the enormous allelic variation in MHC molecules, ie, HLA proteins, between individuals makes it difficult to identify patterns that could reflect local expansions of HLA-restricted T cells. Although this may be partly due to technical difficulties, it seems unlikely that the immunological situation in atherosclerosis could be as simple as the one in type 1 diabetes, for example, where one specific HLA-DQ allele permits an autoimmune response that leads to ß-cell destruction and disease. Instead, it is likely that several antigens and antigenic epitopes are involved and can be presented by several different HLA alleles. Nevertheless, autoantigens have also been identified in atherosclerosis.
OxLDL Is a Candidate Autoantigen
Antibodies to oxLDL can be detected in atherosclerotic patients and experimental animals119,132,133 and are present in atherosclerotic lesions.134 These studies identified oxLDL as a candidate antigen in atherosclerosis (Table 5). Further support for this notion came with the identification of cellular immune responses to oxLDL. T cells can be isolated from fresh human plaques, cloned, expanded in culture, and challenged with candidate antigens. By using this approach, oxLDL was shown to be a major autoantigen in the cellular immune response of atherosclerosis.135 One fourth of all CD4+ T cells cloned from human plaques recognized oxLDL in an HLA-DR-dependent manner. OxLDL-specific T cells are present in lymph nodes of apoE-KO mice,136,137 which have strong humoral as well as cellular immune responses to such modified lipoproteins.138,139 In humans, oxLDL induces activation of a subset of peripheral T cells, suggesting that it is a quantitatively important antigen.140
|
The antibody specificity of IgG molecules in lesions can be determined by isolating them and permitting them to react with an antigen, eg, in ELISA or Western blot analyses. With this approach, it was demonstrated that IgGs of atherosclerotic lesions contain antibodies reactive with oxLDL.134 The antigens, ie, LDL carrying various kinds of oxidative modifications, could be demonstrated in the lesions by immunostaining and Western blot. Therefore, oxLDL is an autoantigen that is formed in the lesion, and it elicits a local cellular as well as a humoral immune response.
The immune response to oxLDL plays a pathogenetic role in atherosclerosis because lesion progression can be inhibited by immunization141–143 or induction of neonatal tolerance to oxLDL.137 It seems paradoxical that both tolerization and hyperimmunization can reduce the extent of disease; this may be due to the different effector pathways activated by these two kinds of treatment.
Heat Shock Proteins Are Common Autoantigens in Inflammatory Disease
Heat shock proteins have also been implicated in the pathogenesis of atherosclerosis144 (Table 5). Such proteins are involved in protein folding in the normal cell, are produced in large amounts by injured cells, are released on tissue injury,145 and serve as targets for autoimmune responses in many inflammatory diseases. Interestingly, heat shock proteins are also released by monocytes exposed to oxLDL.146 Antibodies to heat shock protein 60 (hsp60) and its prokaryotic homologue hsp65 are frequently detected in rheumatoid arthritis and other inflammatory conditions.145
Xu and coworkers147 immunized rabbits with hsp65/60 and recorded induction of vascular inflammation, with endothelial activation and mononuclear cell adhesion. The developing lesions also contained T cells, and cell lines derived from such infiltrates exhibited anti-hsp60 reactivity. Anti-hsp60 antibodies occurred in peripheral blood, and immunization with hsp60 was found to increase fatty streak development in hypercholesterolemic rabbits and mice.147,148 In humans, antibodies to hsp 65/60 are elevated in hypertension149 and early atherosclerosis150 and may predict progression of atherosclerotic disease.151 Because heat shock proteins of humans and microbes are structurally and antigenically similar, it is possible that molecular mimicry between immune responses to microbial heat shock proteins and homologues expressed by vascular cells could account for the association between infections and atherosclerosis.152
A third autoantigen, ß2-glycoprotein Ib (ß2-GPI), is present on platelets but may also be expressed by endothelial cells (Table 5). Autoantibodies to ß2-GPI are produced in several inflammatory disorders, including atherosclerosis.153,154 The immune response to ß2-GPI appears to be proatherogenic, because hyperimmunization with ß2-GPI155 or transfer of ß2-GPI-reactive T cells aggravates fatty streak formation in LDLR-/- mice.156 The pathogenetic mechanism by which ß2-GPI acts remains unclear, but it may be related to this protein’s capacity to bind phospholipids.
A Role for ‘Phospholipid Antibodies’ in Atherosclerosis?
Antibodies reacting with phospholipids such as cardiolipin are associated with recurrent thrombosis157 and accelerated atherosclerosis.158 Many of these antibodies recognize phospholipid-binding plasma proteins such as ß-GPI,159 and it could be speculated that ß2-GPI affects atherosclerosis by targeting immune responses to membrane lipids. Interestingly, another set of "phospholipid antibodies" recognizes oxidized phospholipids in oxLDL160 and may represent "natural" antibodies that appear in early life and also bind phospholipids on apoptotic cells and certain bacteria.161 Perhaps pathogenic immune responses are due to "molecular mimicry" between oxLDL and modified lipids on microbes and apoptotic cells.
Microbes as Antigens in Atherosclerosis
In 1988, Saikku et al162 discovered that patients with cardiovascular diseases often have high-titer antibodies to Chlamydia pneumoniae (Table 5). This common pathogen can cause pneumonia, but it has also been associated with a variety of other chronic diseases. Further investigations revealed that C pneumoniae survives intracellularly in macrophages, can be detected in atherosclerotic lesions, and can be isolated and grown from such lesions.163–165 This established that C pneumoniae is involved in atherosclerosis. However, the extent of C pneumoniae infiltrates in lesions remains controversial,166 as does the importance of C pneumoniae infection as an aggravating factor in experimental atherosclerosis.167–170 It therefore remains to be clarified how important C pneumoniae is for disease development and whether the microorganism or the immune response against it can be harmful.
Viruses of the Herpesviridae family have also been implicated in atherosclerosis (Table 5). Both herpes simplex type 1 and cytomegalovirus have been detected in human lesions.171–173 An avian analogue, Marek’s disease virus, aggravates cholesterol-induced atherosclerosis in chickens. Cytomegalovirus has been linked to transplant arteriosclerosis174,175 and may also be associated with "garden-variety" atherosclerosis and restenosis after angioplasty.176,177 Several pathogenic mechanisms have been proposed.178 Cell culture studies have identified a role for this virus as a stimulus for smooth muscle migration and scavenger receptor expression, which may be important in vascular pathology.179 However, it appears that direct action of the virus, rather than immune responses against it, is playing the important pathogenic role.
Systemic Immune Responses and Serodiagnosis
Systemic humoral immune responses are characteristic for atherosclerotic disease in humans and experimental models. Antibodies to oxidatively modified LDL are commonly found in patients; as expected from the incidence of silent disease, they are frequent also in healthy individuals.133,180 Immunodominant epitopes in modified LDL include malondialdehyde-conjugated and 4-hydroxynonenal-conjugated lysine residues.139,181 These modifications are generated by enzymatic or nonenzymatic attack on fatty acid residues, which leads to release of reactive aldehydes that can bind to the polypeptide chain of apoB-100. Other B-cell epitopes on oxLDL include oxidized phospholipids such as phosphorylcholine, which is also present on several microbes and apoptotic cells and is recognized by "natural" antibodies present in most individuals.161
Anti-oxLDL antibody titers are correlated with the progression of advanced atherosclerosis in some studies,133,182 whereas others have been unable to find any such correlation. Interestingly, titers are correlated negatively with early cardiovascular diseases, including borderline hypertension and carotid intima/media thickening.183–185 It appears that no simple, positive correlation occurs between antibodies and disease throughout its natural history. Instead, it may be that bursts of plaque activity boost immune responses, possibly by releasing antigenic material, and give rise to titer increases. Such episodic activity might be interspersed with periods of slow, silent growth of lesions and fading immune activity. It is also possible that antibodies can serve to eliminate modified lipoproteins from the circulation, thus reducing the load of atherogenic lipoproteins in arterial lesions. Antibodies may therefore be viewed both as markers of disease activity and as vehicles for clearance of antigen. The situation with regard to other antigens appears to be similar to that for modified LDL. Anti-hsp65/60 antibodies have been positively correlated with the progression of carotid atherosclerotic disease in an epidemiological study,151 but more data are needed before these assays find their place in clinical diagnostics.
|
Evidence That Immunity Affects Atherosclerosis |
|---|
|
TopAbstractIntroductionImmune Cells in the...Recruitment of Immune Cells...Cellular Immunity of...Immune Specificity in the...Evidence That Immunity Affects...Can We Prevent Atherosclerosis...References |
|---|
Atherosclerosis in Patients With Immune Disorders
It is by now well established that atherosclerosis is accompanied by adaptive immune responses and that the early phase of the disease is dominated by immune cells, particularly macrophages. These facts do not necessarily imply that atherosclerosis can be treated or prevented by interfering with immune mechanisms. To clarify whether that could be possible, it has been important to determine whether the extent of atherosclerosis is different (reduced or increased) in individuals with immune defects.
The situation in immune-deficient patients is complicated and difficult to interpret. Individuals with severe combined immune deficiencies have not survived long enough to develop cardiovascular disease, and most of the congenital immune deficiencies affect too few patients to make epidemiological investigations into cardiovascular disease meaningful. Individuals with selective, congenital defects in the humoral immune responses are relatively frequent, and the clinical syndrome is compatible with life into adult age. These patients do not exhibit any reduced heart disease, suggesting that humoral immunity does not aggravate atherosclerosis.186 HIV patients, who lack CD4+ T cells (and particularly Th1 activity), develop aggressive cardiovascular disease, particularly when modern highly active antiretroviral treatment has been used to prevent rapid development of AIDS.187–189 However, the protease inhibitor treatment may itself cause metabolic syndromes, and it is difficult to determine whether the increased cardiovascular morbidity is due to the disease or its treatment.
Some of the most remarkable data in support of a link between immunity and atherosclerosis come from epidemiological studies of patients with autoimmune disorders. Patients with rheumatoid arthritis have a 2- to 5-fold increase in cardiovascular morbidity and mortality,190 and patients with systemic lupus erythematosus exhibit an even higher increase in cardiovascular disease.191,192 Although some of this morbidity is clearly due to small-vessel vasculitis associated with the underlying autoimmune disease, there is also evidence for atherosclerotic large-vessel disease in many cases. Interestingly, patients with systemic lupus erythematosus share several autoimmune phenomena with atherosclerotic patients.193 Further studies into these aspects are clearly needed.
Mechanistic Insights From Gene-Targeted Mouse Models
To clarify the role for adaptive immunity in atherosclerosis, it was necessary to generate immunodeficient animal models (Table 4). The apoE-KO mouse was crossed with the recombinase-activating gene-1-/- (RAG-1-/-) mouse, which lacks functional T and B cells due to a mutation in the rearrangement machinery. The offspring exhibited a 40% reduction of atherosclerosis, indicating that adaptive immunity accelerates disease.194 However, the impact of the immune defect was undetectable in cases of excessive hypercholesterolemia.194,195 This could be due to either an overwhelming effect of very high cholesterol levels or a qualitative difference in immune activity in this metabolic condition. The finding of a Th1
Th2 switch in severe hypercholesterolemia104 supports the latter conclusion. Interestingly, the atheroprotective effect of estrogens was recently found to be lacking in immunodeficient mice, suggesting that the atheroprotective effect of estrogen depends on estrogen action via immune mechanisms.196 Whether this involves switches in effector functions remains to be determined.
A more recently constructed apoE-/- x scid/scid mouse, which also lacks adaptive immunity, shows an even more striking phenotype, with a 70% reduction of disease in immunodeficient female apoE-/- x scid/scid mice compared with immunocompetent apoE-/- scid/+ littermates197 (Table 4). Neither the RAG nor the SCID immunodeficient mice exhibited any changes in serum cholesterol compared with single-KO apoE-/- mice; this points to important effects on the vascular wall rather than on systemic metabolism for the effect on atherosclerosis.
This conclusion was further supported when immune cell populations were transferred into the apoE-/- x scid/scid mice. Transfer of CD4+ T cells from immunocompetent apoE-/- mice to immunodeficient apoE-/- x scid/scid mice increased atherosclerosis dramatically in the recipient, from 27% to 79% of the level in fully immunocompetent apoE-/- mice.197 Transferred T cells migrated to lesions and induced expression of MHC class II genes, probably through IFN- secretion. These data show that immune activity increases atherosclerosis and identifies CD4+ T cells as the proatherogenic subset.
In another recent study, a T-cell-mediated response to a vascular autoantigen was found to aggravate atherosclerosis in apoE-KO mice.112 These mice were bred with a mouse strain that carries the ß-galactosidase gene under the control of a smooth muscle-specific promotor activated by a tamoxifen-inducible Cre recombinase. When such mice were immunized by injections of dendritic cells presenting ß-galactosidase, CD8+ T cells were activated to specifically recognize this antigen. Subsequent induction of ß-galactosidase expression in SMCs by tamoxifen treatment led to a lytic attack of the SMCs by antigen-specific CD8+ T cells. This resulted in a dramatic increase in atherosclerosis.
Important Roles for Immunoregulatory Cell Surface Molecules
Immune activation depends on interactions between proteins displayed on the surfaces of adjacent cells as well as on soluble cytokines. Such proteins include adhesion molecules, such as ICAM-1, and the more narrowly expressed pairs of costimulatory molecules. The latter include the T-cell protein CD28, which binds 2 alternative counterreceptors, B7.1 (CD80) and B7.2 (CD86). CD28 ligation promotes T-cell activation, whereas ligation of B7 proteins to the alternative T-cell surface receptor, cytotoxic T-lymphocyte antigen-4 (CTLA-4, CD152), induces a negative signal for cell-mediated immune responses. Soluble CTLA-4 proteins have been used to inhibit immune reactions and can prevent chronic rejection of organ transplants.198 However, they do not appear to reduce experimental atherosclerosis.
Another costimulatory molecule, CD40, is expressed by B cells and dendritic cells and can be induced on many different cell types. It ligates the constitutively expressed T-cell protein, CD40 ligand (CD40L, or CD154). This interaction is necessary for T-B cell cooperation in the induction of antibody responses. CD40 is constitutively expressed by endothelial cells and can be induced in both vascular endothelial cells and SMCs by stimulation with proinflammatory cytokines and IFN-.199,200 Similarly, CD40L could be induced on these cells by cytokine stimulation.200 Ligation of CD40 leads to tissue factor expression by endothelial cells and can stimulate proteinase secretion from macrophages and SMCs.201–203 Interestingly, CD40 and CD40L are present on vascular cells and macrophages of atherosclerotic lesions, and their activation may be involved in the progression of atherosclerotic disease. Recent animal experiments support this notion (Table 4). First, anti-CD40L antibody injections reduced fatty streak development in LDLR mice.204 Second, CD40L-/- apoE-/- compound-KO mice exhibited slower induction of lesions than did the CD40L+/+ apoE-/- controls.205 These data suggest that inhibition of the CD40 pathway might be a means of treating atherosclerosis. However, the mechanism of action for inhibitors such as anti-CD40L antibodies remains unclear and could involve not only immunomodulation but also direct effects on vascular cells.
|
Can We Prevent Atherosclerosis by Immunization? |
|---|
|
TopAbstractIntroductionImmune Cells in the...Recruitment of Immune Cells...Cellular Immunity of...Immune Specificity in the...Evidence That Immunity Affects...Can We Prevent Atherosclerosis...References |
|---|
Encouraging Experimental Data for the Immune Strategy
After having cited a wealth of studies that establish adaptive—and innate—immunity as proatherogenic, it may seem totally unrealistic to ask whether immunization, ie, deliberate activation of specific immunity, might protect against atherosclerosis. However, one should consider that several infections in which cellular immunity plays a pathogenic role can be prevented by vaccination. Experimental autoimmune diseases such as experimental autoimmune encephalomyelitis can be prevented by protective immunization. It is now evident that entirely different immune effector responses can be induced against the same antigen, with some important ones elicited by local antigen release in parenchymal tissue and others by subcutaneous or oral immunization.
As discussed, several different antigens have been implicated in the pathogenesis of atherosclerosis. Interestingly, immunization with one of them, oxLDL, can reduce disease in several animal models141–143,206,207 (Table 4), whereas immunization with hsp65/60 or ß2-GPIb aggravates lesion development147,148,155,156 (Table 4). The reason for this apparent discrepancy is not known; however, there are important differences between the antigens. Hsp60 is an intracellular molecule that can be expressed on the surface of stressed cells. OxLDL, in contrast, is an extracellular particle that is present in the interstitial space of the intima and may even circulate in the blood.208 It could be speculated that protective immunization against atherosclerosis works by inducing high-titer antibodies that increase the clearance of oxLDL by way of Fc and C3 receptors. In support of this, the protective effect of oxLDL immunization can be correlated with the titer of T-cell-dependent IgG antibodies,143 and transfer of polyclonal immunoglobulins, which contain anti-oxLDL antibodies, reduces atherosclerosis in apoE-/- mice.209
The detrimental effect of immunization with hsp60 may depend on the fact that this protein can appear on the surface of endothelial cells and/or SMCs. This could lead to antibody binding that is followed by complement activation and lysis of the vascular cells. In addition, peptides derived from hsp60 synthesis might bind to MHC class I molecules, which would permit attack by cytolytic CD8+ T cells. The potential importance of the latter pathway was recently demonstrated in mice in which an autoimmune response to a transgene expressed on SMCs led to a CD8+ T-cell attack on the vessel wall and aggravation of atherosclerosis.112 It is possible that endogenous antigens such as hsp60 could elicit a similar attack once a cellular immune response has developed toward the protein.
Therapeutic Potential of Vaccination and Immunomodulation
Can we use our current knowledge about immune activity in atherosclerosis to prevent or treat the disease? Although no such therapy is used clinically today, there is reason for optimism. The success in protective immunization with oxLDL as an immunogen in experimental models suggests that this could also be a fruitful approach in humans. However, oxLDL is a heterogeneous particle that may be unsuitable for parenteral administration in patients. It may also be extremely difficult to standardize to the extent that is required for vaccines. For these reasons, it will be necessary to identify the most important B- and T-cell epitopes on oxLDL and to test whether either of them can be used as an immunogen with the same success as the intact particle. If this turns out to be the case, clinical trials should be encouraged.
Immunomodulation is also attractive as a possible therapeutic principle. However, broadly acting immunosuppressants such as corticosteroids and cyclosporins have undesirable, direct vascular effects that make them unsuitable for this purpose. Instead, we should try to develop more specific immunomodulators that act on the key mechanisms in the pathogenesis of atherosclerosis. Potential targets are proinflammatory lipids such as PAF and lysophosphatidylcholine, lipoprotein-derived antigens, and signal transduction pathways leading to inflammatory innate and adaptive immune responses.
Interestingly, certain drugs targeting lipid and carbohydrate metabolism may also act as immunomodulators. Statins act as repressors of MHC class II-mediated T-cell activation210 and can improve the outcome of cardiac transplantation.211 Glitazones, which are agonists for peroxisome-proliferator activating receptor-, can also inhibit T-cell activation and cytokine secretion.212 It is tempting to speculate that some of the beneficial effects of these types of compounds are due to their immuomodulatory effects rather than their action on cholesterol or glucose metabolism.
Because the proinflammatory/Th1 pathway appears to play a key role, interference with signal transduction molecules, receptors, or cytokines involved in this cascade could be interesting as a potential therapeutic approach. TNF- antagonists are already in clinical use for rheumatoid arthritis and have been tested clinically for heart failure. It will be important to determine whether they also act against atherosclerosis. Similarly, IFN antagonists and inhibitors of the NF-B and Janus kinase/signal transducer and activator of transcription (Jak/STAT) pathways could turn out to be useful. Experimental studies have already shown beneficial effects of blockers of the costimulatory molecules CD40/CD40L; inhibition of this or similar costimulatory proteins might also be effective against atherosclerosis (see above).
Finally, it would be theoretically attractive to apply an immunomodulatory treatment that promotes anti-inflammatory immunity. This might be accomplished by enhancing the Th2 or TGF-ß/Th3 effector mechanisms, for example, or by the more general inhibition of cell-mediated immunity that can be achieved by treatment with large doses of polyclonal immunoglobulins.209 The latter treatment, if useful clinically, would have to be reserved for selected cases of rapidly accelerating atherosclerosis. However, it may point to a useful principle for immunomodulation that could be used to develop more defined pharmaceuticals.
To summarize, several interesting drugs and immunogens have shown effects against atherosclerosis in experimental systems, and an additional set of molecules is attractive from a theoretical point of view. We should have some interesting years ahead of us.
|
Acknowledgments |
|---|
Our work is supported by the Swedish Heart-Lung Foundation and Science Council (project 6816) and by the Ragnar and Torsten Söderberg foundation. I thank Johan Frostegård, Zhong-qun Yan, and Xinghua Zhou for constructive criticism.
Received May 21, 2001; accepted October 2, 2001.
|
References |
|---|
|
TopAbstractIntroductionImmune Cells in the...Recruitment of Immune Cells...Cellular Immunity of...Immune Specificity in the...Evidence That Immunity Affects...Can We Prevent Atherosclerosis...References |
|---|
- Poole JCF, Florey HW. Changes in the endothelium of the aorta and the behaviour of macrophages in experimental atheroma of rabbits. J Pathol Bacteriol. 1958; 75: 245–252.[Medline] [Order article via Infotrieve]
- Schaffner T, Taylor K, Bartucci EJ, Fischer DK, Beeson JH, Glagov S, Wissler RW. Arterial foam cells with distinctive immunomorphologic and histochemical features of macrophages. Am J Pathol. 1980; 100: 57–80.[Abstract]
- Fowler S, Shio H, Haley NJ. Characterization of lipid-laden aortic cells from cholesterol-fed rabbits, IV: investigation of macrophage-like properties of aortic cell populations. Lab Invest. 1979; 41: 372–378.[Medline] [Order article via Infotrieve]
- Gerrity RG. The role of the monocyte in atherogenesis, I: transition of blood-borne monocytes into foam cells in fatty lesions. Am J Pathol. 1981; 103: 181–190.[Abstract]
- Gerrity RG. The role of the monocyte in atherogenesis, II: migration of foam cells from atherosclerotic lesions. Am J Pathol. 1981; 103: 191–200.[Abstract]
-
Faggiotto A, Ross R. Studies of hypercholesterolemia in the nonhuman primate, I: changes that lead to fatty streak formation. Arteriosclerosis. 1984; 4: 323–340.
[Abstract/Free Full Text] -
Faggiotto A, Ross R. Studies of hypercholesterolemia in the nonhuman primate, II: fatty streak conversion to fibrous plaque. Arteriosclerosis. 1984; 4: 341–356.
[Abstract/Free Full Text] -
Brown MS, Goldstein JL. A receptor-mediated pathway for cholesterol homeostasis. Science. 1986; 232: 34–47.
[Free Full Text] -
Fogelman AM, Shechter I, Seager J, Hokom M, Child JS, Edwards PA. Malondialdehyde alteration of low density lipoproteins leads to cholesteryl ester accumulation in human monocyte-macrophages. Proc Natl Acad Sci U S A. 1980; 77: 2214–2218.
[Abstract/Free Full Text] - Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol: modifications of low-density lipoprotein that increase its atherogenicity. N Engl J Med. 1989; 320: 915–924.[Medline] [Order article via Infotrieve]
- Jonasson L, Holm J, Skalli O, Gabbiani G, Hansson GK. Expression of class II transplantation antigen on vascular smooth muscle cells in human atherosclerosis. J Clin Invest. 1985; 76: 125–131.[Medline] [Order article via Infotrieve]
-
Jonasson L, Holm J, Skalli O, Bondjers G, Hansson GK. Regional accumulations of T cells, macrophages, and smooth muscle cells in the human atherosclerotic plaque. Arteriosclerosis. 1986; 6: 131–138.
[Abstract/Free Full Text] - Hansson GK, Holm J, Jonasson L. Detection of activated T lymphocytes in the human atherosclerotic plaque. Am J Pathol. 1989; 135: 169–175.[Abstract]
- van der Wal AC, Das PK, van de Berg DB, van der Loos CM, Becker AE. Atherosclerotic lesions in humans: in situ immunophenotypic analysis suggesting an immune mediated response. Lab Invest. 1989; 61: 166–170.[Medline] [Order article via Infotrieve]
-
Stemme S, Holm J, Hansson GK. T lymphocytes in human atherosclerotic plaques are memory cells expressing CD45RO and the integrin VLA-1. Arterioscler Thromb. 1992; 12: 206–211.
[Abstract/Free Full Text] - Kishikawa H, Shimokama T, Watanabe T. Localization of T lymphocytes and macrophages expressing IL-1, IL-2 receptor, IL-6 and TNF in human aortic intima: role of cell-mediated immunity in human atherogenesis. Virchows Arch A Pathol Anat Histopathol. 1993; 423: 433–442.[Medline] [Order article via Infotrieve]
- Pober JS, Collins T, Gimbrone MAJr, Cotran RS, Gitlin JD, Fiers W, Clayberger C, Krensky AM, Burakoff SJ, Reiss CS. Lymphocytes recognize human vascular endothelial and dermal fibroblast Ia antigens induced by recombinant immune interferon. Nature. 1983; 305: 726–729.[Medline] [Order article via Infotrieve]
- Warner S, Auger KR, Libby P. Interleukin-1 induces interleukin-1, II: recombinant human interleukin-1 induces interleukin-1 production by adult human vascular endothelial cells. J Immunol. 1987; 139: 1911–1917.[Abstract]
- Bobryshev YV, Lord RSA. S-100 positive cells in human arterial intima and in atherosclerotic lesions. Cardiovasc Res. 1995; 29: 689–696.[Medline] [Order article via Infotrieve]
-
Kaartinen M, Penttilä A, Kovanen PT. Accumulation of activated mast cells in the shoulder region of human coronary atheroma, the predilection site of atheromatous rupture. Circulation. 1994; 90: 1669–1678.
[Abstract/Free Full Text] - Collins T, Read MA, Neish AS, Whitley MZ, Thanos D, Maniatis T. Transcriptional regulation of