Nature Medicine
8, 1218 - 1226 (2002)
doi:10.1038/nm1102-1218
Innate and acquired immunity in atherogenesisChristoph J. Binder, Mi-Kyung Chang, Peter X. Shaw, Yury I. Miller, Karsten Hartvigsen, Asheesh Dewan
& Joseph L. WitztumDepartment of Medicine, University of California San Diego, La Jolla, California, USA
Correspondence should be addressed to Joseph L. Witztum jwitztum@ucsd.eduTraditional risk factors like hypercholesterolemia are important for atherogenesis, but it is now apparent that the immune system also plays an important role. Uncovering the mechanisms by which specific components of the immune system impact atherogenesis will not only provide new insights into the pathogenesis of lesion formation, but could also lead to novel therapeutic approaches that involve immune modulation.Atherosclerosis is a chronic disease that begins in fetal life, slowly progresses during childhood and adolescence, and then accelerates in fits and spurts in adult life to result in plaque erosion or rupture, effecting morbid or fatal clinical events. Autopsy studies have confirmed the presence of advanced and obstructive lesions in young and clinically robust individuals, and intracoronary ultrasound studies have demonstrated the remarkable prevalence of coronary atherosclerosis in 37% of 'healthy' heart donors 20−29 years of age, 60% of those 30−39 years of age and 85% of those over 50 years of age1. Therefore, it is not unexpected that the initial manifestation of cardiovascular disease (CVD) in 50% of patients is either sudden death or myocardial infarction. These data emphasize the need to intervene early in the 'natural history' of atherosclerosis to prevent the development of clinical disease and death.
Hypercholesterolemia seems to be an absolute prerequisite for lesion initiation and progression. If all adults had life-long ideal plasma cholesterol (below 150 mg/dl), there would probably be few if any cases of symptomatic disease. However, there is diversity in expression of disease even in cases of extreme hypercholesterolemia: a 3-year-old male with homozygous familial hypercholesterolemia developed myocardial infarction and died, whereas a 33-year-old female with the same mutation in the gene encoding the low-density-lipoprotein receptor (LDLR) and the same plasma cholesterol concentration (800 mg/dl) died from a non-coronary cause2.
What accounts for these differences in the rate of lesion formation and clinical presentation? For the most part, this is unknown; furthermore, because atherosclerosis is a complex disease whose etiology is multifactorial, it is likely that different factors will dominate in different individuals3. There are many 'traditional' risk factors discovered by classical epidemiology that define much of the risk, such as hypercholesterolemia, smoking, male gender, hypertension, diabetes and age. However, newly defined 'non-traditional' risk factors are emerging as being equally important. Among these, it is now apparent that the immune system, when considered in the broadest context, is one of the dominant factors, aside from cholesterol, modulating atherogenesis4,
5,
6. Here we will review the rationale for immune involvement—the evidence that both adaptive and innate immune mechanisms come into play in response to antigens in the atherosclerotic lesion and that these responses can modulate lesion development. Many of the immune responses involved in atherogenesis most likely evolved as responses to other pathogens that are more fundamentally important for the survival of the host, and provide examples of antigens that show molecular mimicry between epitopes of atherosclerosis-associated antigens and other endogenous and exogenous pathogens. Understanding the mechanisms by which the immune system affects atherogenesis not only will provide new insights into the pathogenesis of lesion formation but also could lead to new therapeutic approaches that involve immune modulation.
Why is the immune system involved?
Many genes are involved in immune function, indicative of the vital survival advantage of such a complex system. Atherosclerosis is manifested well beyond the reproductive period and thus could not have exerted evolutionary pressure for any selected immune response. However, it is now recognized that there are responses to inflammatory components of the atherogenic process that are shared with similar disease processes in infectious as well as other acute and chronic diseases. It is likely that those other diseases have 'selected' certain immune responses that in turn affect atherogenesis for better or worse. The fundamental appreciation that inflammation is an important and possibly even obligatory component of lesion initiation and progression, and also participates in the plaque rupture that mediates thrombotic complications and clinical events, has fundamentally changed the view of the pathogenesis of atherosclerosis7,
8. Immune activation must be viewed in the context of responses to inflammatory components of atherogenesis.
Inflammation is a process in which blood leukocytes leave the vascular space and enter a tissue site in response to a perceived pathogen. In atherosclerotic lesions, the nature of the inciting pathogen is not entirely clear. There are probably several different candidate pathogens or pathogenic processes that can elicit localized inflammatory responses in the artery; one prime set of candidates is minimally oxidized LDL (called minimally modified LDL, or mmLDL) and late forms of oxidized LDL (oxLDL). Once trapped in the artery wall by binding to extracellular proteoglycans, a key event in atherogenesis9,
10, LDL is oxidized by mechanisms not yet understood11 and/or undergoes other types of modifications, such as non-enzymatic glycation, enzymatic degradation, aggregation or combinations of these, all of which result in alterations of 'self'. The consequent generation of a wide spectrum of oxidation-specific (or other modification-specific) neo-epitopes renders the modified LDL immunogenic, and leads to both a cellular and a humoral response6. In addition, the oxidation of LDL generates oxidized lipids that are toxic, pro-inflammatory and ultimately pro-atherogenic12. For example, oxidized cholesterol moieties can promote apoptosis and cell death13. Oxidized phospholipids can induce artery wall cells to secrete chemotactic molecules, such as monocyte chemoattractant protein 1 (MCP-1); activate endothelial cells to express adhesion molecules for monocytes and T cells; and induce expression of growth factors, such as monocyte colony-stimulating factor, that facilitate the phenotypic transformation of monocytes into macrophages and stimulate the proliferation of smooth muscle cells14,
15. Macrophages, a central mediator of cellular innate (and adaptive) immunity, are essential in lesion initiation and progression3,
4. Once activated, they initiate the oxidation of LDL and rapidly take up oxLDL through specific scavenger receptors, leading to foam-cell formation16. This is a key event in disease progression, as mice deficient in scavenger receptor A and CD36 have significantly reduced atherosclerosis17,
18. Activated macrophages also secrete a variety of pro-inflammatory products that affect lesion progression and plaque stability3.
Other potential candidate pathogens include infectious agents19, pathogenic molecules that incite one or more events of the inflammatory cascade (such as lysophospholipids or oxidized lipids), and metabolic events, which can mediate increased production of reactive oxygen species (such as hypercholesterolemia or hypertension, by means of angiotensin II)20. Evidence indicates that the chief risk factors for atherogenesis, such as dyslipidemia and diabetes as well as the insulin resistance associated with obesity, also contribute to inflammatory conditions through complex mechanisms, including enhanced lipid peroxidation, glycoxidation and increased secretion of pro-inflammatory cytokines, among other effects. Evidence that inflammation is important is strongly supported by many studies showing that increased plasma concentrations of markers of inflammation, such as C-reactive protein (CRP), interleukin 6 (IL-6), serum amyloid A, IL-1 receptor (IL-1R) antagonist and soluble adhesion molecules, are independent predictors of coronary events8 (Fig. 1).
 | | |
Evidence of immune system involvement in atherogenesis
Both adaptive and innate immune responses modulate the rate of lesion progression as well as the composition of the lesions (Table 1). Both apolipoprotein E−deficient (Apoe-/-) and Ldlr-/- mice have been individually crossed into a recombination activating gene (Rag)-deficient background (Rag1-/- or Rag2-/-), generating hypercholesterolemic mice lacking both T and B cells. When such mice are fed an atherogenic diet to produce very high plasma cholesterol (above 1,300 mg/dl) for a sufficiently long period of time (more than 16 weeks), they do not show any change in the extent of lesion formation as compared with that of Apoe-/- mice with normal immune function21,
22,
23. However, when these same immunodeficient mice are examined at earlier time points (4−8 weeks) or even at more extended periods of time but in the presence of much lower plasma cholesterol concentrations (about 600−800 mg/dl), the effect of immune deficiency results in a 40−80% decrease in the extent of lesions formed21,
23,
24. In addition, crosses of Apoe-/- mice with severe combined immunodeficiency mice, which also lack B and T cells, generated mice that also had 70% fewer lesions. However, when CD4+ T cells from atherosclerotic Apoe-/- mice were transferred into these mice, they developed atherosclerosis similar to that of immunocompetent controls25. These immune-mediated effects were site specific, as Apoe-/-  Rag2-/- mice showed a 60−80% reduction in lesions at the site where lesions first appear in mice, the aortic root, whereas there were no differences in lesions of the brachiocephalic trunk in male mice, although this was not the case in female mice24.
| | |
Cumulatively, these studies indicate that T and B cells are not obligatory for lesion initiation or progression if there is sufficiently high atherogenic pressure generated by substantial hypercholesterolemia for a sufficient period of time. However, with less atherogenic pressure, these immune responses are influential in modulating the course of atherogenesis. These studies indicate a net pro-atherogenic effect of T- and B-cell function, but even these effects must be qualified by site and possibly even gender effects. Furthermore, there are many examples of specific protective mechanisms of immune activation as well, such as the demonstration that immunization of hypercholesterolemic animals with an immunodominant antigen of lesions, oxLDL, ameliorates the progression of lesion formation despite very high plasma cholesterol26,
27,
28,
29,
30. Infusion of polyspecific immunoglobulin into Apoe-/- mice is also protective31. The important lesson is that the impact of immune function is complex, as might be expected, and atherosclerosis can be both enhanced and inhibited by immune modulation. However, there is no evidence so far that immune mechanisms are primary causes of atherosclerosis.
Innate immunity and atherosclerosis
Innate immunity is characterized by a natural selection of germline-encoded receptors, which focuses on highly conserved motifs in pathogens. It provides the first line of defense for the host and is characterized by fast and blunt responses. It involves several cell types, most importantly macrophages (and dendritic cells), which express a limited repertoire (about 100) of highly conserved pattern-recognition receptors (PRRs), such as scavenger receptors and Toll-like receptors (TLRs). Such PRRs typically recognize a restricted pattern of ligands, called pathogen-associated molecular patterns (PAMPs). In addition to being ligands on pathogens, PAMPs include a diverse array of compounds, including lipopolysaccharides, teichoic acids, aldehyde-derivatized proteins, mannans, bacterial DNAs and denatured DNAs. After receptor ligation, cells either endocytose PAMP-expressing particles or are activated (for example, through nuclear factor- B), which elicits an inflammatory response32,
33.
The recruitment of monocytes is essential for lesion formation (Fig. 2), as hypercholesterolemic mice that are deficient in MCP-1 or in expression of CCR2, its cognate receptor on monocytes, have a greatly reduced incidence of atherosclerosis 34,
35. Similarly, hypercholesterolemic Op/Op (Csf1-/-) mice, which lack monocyte colony-stimulating factor and therefore lack differentiated macrophages in their tissues, show minimal atherosclerosis36. During early events in atherosclerosis, activated endothelial cells express various adhesion molecules, resulting in leukocyte rolling and adhesion37. This activation is triggered most prominently by products of oxLDL. For example, the oxidized phospholipid 1-palmitoyl-2-(5-oxovaleroyl)-sn-glycero-3-phosphorylcholine (POVPC) selectively induces connecting segment-1 (CS-1) fibronectin, whereas other oxidized phospholipids show an induction of vascular cell adhesion molecule 1, which might explain the preferential recruitment of monocytes and T cells because of their interaction with the integrin very late antigen 4 (ref. 14). Cytokines such as tumor necrosis factor- (TNF-) and IL-1 are also involved. Oxidized LDL and locally secreted chemokines, such as MCP-1, effect the chemotactic migration of monocytes and T cells, but not neutrophils, into the subendothelial space38. In turn, the expression of MCP-1 by arterial cells is strongly stimulated by oxidized phospholipids as well as cytokines and activated complement.
| | |
Macrophages (and dendritic cells) are important in adaptive immunity in their capacity to ingest pathogens and present antigens to initiate adaptive immune responses. Macrophages are fundamental innate immune cells through their production of reactive oxygen species, proteases and cytokines as well as their scavenging functions mediated through PRRs. The scavenger receptors CD36 and SR-B1, as well as SR-A, are examples of such PRRs, which internalize oxLDL, leading to foam-cell formation, a rate-limiting step in atherogenesis17,
18. TLRs are expressed in atherosclerotic lesions39 and may also participate in inflammatory signaling. Although TLR4 deficiency does not decrease atherosclerosis in cholesterol-fed Apoe-/- mice40, a TLR4 polymorphism that attenuates receptor signaling is associated with decreased atherosclerosis in humans41. The function of other TLRs has not yet been examined. CD14, the non-transmembrane receptor for lipopolysaccharide, initiates inflammatory responses through interactions with TLRs. A polymorphism in the CD14 promoter, resulting in a significantly higher density of CD14 on monocytes, has been identified as a risk factor for myocardial infarction42. These data indicate a convergence of innate immunity to pathogens and atherosclerosis. A variety of stimuli can activate macrophages to secrete an array of cytokines that tightly regulate scavenger receptor expression and inflammatory responses in general. Components of oxLDL, in turn, can also activate macrophages and affect gene expression, such as expression of genes encoding peroxisome proliferator-activated receptor- , CD36 and ATP-binding cassette transporter A1, which profoundly influence macrophage inflammatory and atherogenic activity (discussed in the accompanying article by Li and Glass43).
Soluble factors of innate defense, such as complement and CRP, which act together to rapidly eliminate microbes, are also involved in atherogenesis. Deposition of C3, C4 and terminal C5b−C9 complement complexes occur in atherosclerotic lesions44. Ldlr-/- mice deficient in C3 develop increased atherosclerosis, indicating an important function for the complement system45. As discussed above, CRP reflects inflammation and is an independent risk factor for CVD. In the context of immunity, it functions as an acute-phase protein to rapidly clear pathogens46. CRP may be involved in atherogenesis, however, as indicated by the observation that it is secreted by intimal macrophages, as discussed below47.
Natural antibodies (generated in the absence of known antigen stimulation), mainly immunoglobulin M (IgM), mediate immediate responses against systemic bacterial and viral infections, preventing their dissemination. Many of these antibodies are produced by a specialized set of innate B lymphocytes, the B1 cells, which typically express a restricted set of germline-encoded antigen receptors with specificity for natural or self ligands and generate a link between the innate and adaptive immune responses. In atherosclerosis, IgM responses to oxLDL have been found in both animal models of atherosclerosis and in humans (discussed below). Immunoglobulins, including IgM, are deposited in atherosclerotic lesions, and some of these have been shown to have specificity for epitopes of oxLDL (ref. 48). CD1d-restricted natural killer T cells, equivalent to B1 cells as innate T lymphocytes, have not yet been reported to be involved in atherosclerosis.
Adaptive immunity and atherosclerosis
In contrast to innate immunity, adaptive immunity is more precise but slower. Specific molecular structures on antigens are recognized by antigen receptors, such as T-cell receptors (TCRs) and B-cell receptors (BCRs), which provide great specificity and affinity by somatic rearrangements in blast cells. In contrast to the limited number of PRRs, TCRs have been estimated to number about 1018 and BCRs about 1014, providing an almost unlimited number of receptors.
Dendritic cells and macrophages can activate T cells. This process involves the presentation of antigen to the antigen-specific TCR and additional co-stimulatory signals, such as the interaction of CD40 ligand (CD40L) with CD40. Most antigens cannot stimulate B cells without assistance from CD4+ T cells, which recognize these peptide−major histocompatibility complex (MHC) complexes on B cells, become activated and provide signals (co-stimulatory molecules and cytokines) that promote somatic hypermutation and immunoglobulin class switching. Most CD4+ cells are cytokine-secreting T-helper (Th) cells and express  -TCR, which interacts with MHC class II molecules. A smaller portion expresses  -TCR; most of these cells interact with the non-polymorphic non-classical MHC molecule CD1, which presents certain antigens (particularly lipids and glycolipids). Based on the cytokines they secrete, Th cells are traditionally divided into Th1 cells, which secrete interferon- (IFN-) and IL-2 and promote cell-mediated immunity, and Th2 cells, which secrete IL-4, IL-5, IL-10 and IL-13 and help B cells produce antibodies. These two subsets of Th cells also cross-regulate each others' activity. CD8+ T cells are mainly cytotoxic killer cells, although they too can secrete cytokines, such as TNF-, lymphotoxin and IFN-.
Some T cell−independent antigens can activate B cells without the help of cognate T-cell function; these responses are dominated by IgM, because class switching is severely limited49. Such thymus-independent antigens are typically represented by closely spaced repeated epitopes expressed at high density50. Oxidized LDL, which expresses multiple copies of oxidation-specific epitopes on a single LDL particle, is probably an example of such an antigen.
In general, antibodies provide protection against exogenous pathogens and endogenous altered self molecules to maintain homeostasis by neutralization and clearance. Antibodies can also induce other components of the immune system, such as complement pathways and effector functions of other immune cells (macrophages, B cells, mast cells and so on). The latter process involves binding to antibody class−corresponding Fc receptors, leading to either negative or positive regulation of immune cell responses, the efficient uptake of immune complexes for the degradation of antigens, and efficient antigen presentation. These adaptive responses cooperate in a highly regulated manner and arm the host for quick and specific responses in subsequent encounters with the antigen.
T cells are a prominent component of both early and late lesions (Fig. 2). Most T cells in lesions bear CD3 and CD4 markers and -TCR. These represent about two-thirds of all CD3+ cells in advanced human lesions and more than 90% of T cells in lesions of Apoe-/- mice5. There is also evidence for the presence of dendritic cells, which could be part of a hypothesized Langerhans cell−like network in the vessel wall51. In addition, moderate numbers of CD8+ T cells are found. Small numbers of natural killer cells are also found in early lesions; however, cytolytic activity mediated by natural killer cells does not affect lesion progression52. In contrast, there are relatively few B cells, although they are found in the adventitia surrounding lesions53,
54.
All lesion cells express MHC class II molecules, indicative of IFN--mediated activation. In turn, MHC class II molecules can interact with the TCRs of CD4+ Th cells5. Unexpectedly, the co-stimulatory factors CD40 and CD40L have been reported to be expressed widely in all cells in the lesion, not just in T cells and B cells55. These interactions are important, as genetic disruption of CD154 (CD40L) in Apoe-/- mice56, as well as treatment of cholesterol-fed Ldlr-/- mice with antibody against CD40L, reduced lesion formation by more than 60% (ref. 57). Furthermore, treatment with antibodies against CD40L inhibited the evolution of already established lesions and induced a stable plaque phenotype58,
59. IFN- induces CD40, and CD40 ligation induces matrix metalloproteinase expression by macrophages, which can cause plaque destabilization.
Th1 cells are dominant in atherosclerotic lesions, especially early lesions, as the cytokines IFN-, IL-2 and TNF- are highly expressed, whereas only low amounts of the Th2 cytokines IL-4, IL-5 and IL-10 can be detected. In addition, IgG2a antibodies against oxLDL epitopes, typical of Th1 help, predominate in plasma during early stages of atherosclerosis in Ldlr-/- and Apoe-/- mice. IL-4 expression was detected in lesions of Apoe-/- mice only at very advanced stages of disease in the presence of excessive hypercholesterolemia. Only at this later stage were IgG1 antibodies, typical of Th2 help, found more prominently5.
In the context of atherogenesis, the Th1 cytokine IFN- is pro-inflammatory and pro-atherogenic. In addition to activating macrophages, it inhibits smooth muscle cell proliferation and collagen synthesis, and thereby promotes plaque destabilization. IL-1 and TNF- have similar functions in promoting inflammatory responses (IL-6 and CRP) and the induction of matrix metalloproteinase 9. IL-12 and IL-18 are also highly expressed in atherosclerotic lesions and can further augment IFN- secretion60,
61. Th1-mediated effects appear to be atherogenic: Apoe-/- mice deficient in the IFN- receptor have significantly decreased lesions and increased collagen content62; daily administration of IFN- promoted atherosclerosis in Apoe-/- mice63; and daily administration of IL-12 or IL-18, both of which promote Th1 effects, also increased atherosclerosis in Apoe-/- mice64,
65. The pro-atherogenic effects of IL-18 are mediated through IFN- (ref. 65). Thus, Th1 cells secrete IFN-, which activates macrophages and leads to their release of IL-12, which in turn augments IFN- secretion by T cells. IFN- also inhibits the production of the Th2 cytokines IL-4 and IL-10.
In contrast, Th2 responses seem to antagonize pro-atherogenic Th1 effects and thereby confer atheroprotection. IL-10 can potently suppress IL-12 and IFN- secretion. In vitro studies have shown that recombinant IL-10 inhibited the oxLDL-induced production of IL-12 by human monocytes, indicating a protective function for IL-10. Indeed, cholesterol-fed Ldlr-/- mice that overexpress IL-10 in T cells under control of the human IL-2 promoter had 50% smaller lesions66 compared to Ldlr-/- mice. These same mice also had decreased circulating IFN--secreting CD4+ cells in peripheral blood and in the spleen, and showed an increased ratio of IgG1 to IgG2a antibodies against malondialdehyde-(MDA)−LDL. Furthermore, treatment of Apoe-/- mice with pentoxifylline, which inhibits Th1 differentiation, decreases lesion formation67. In addition, mildly hypercholesterolemic, genetically modified C57Bl/6 mice demonstrated a protective function for Th2-biased responses in fatty streak formation68. The function of the Th2 cytokine IL-4 might be more complex, as other studies found this cytokine to be pro-atherogenic69. In summary, Th2 responses seem to have a protective effect. T cells and macrophages engage in an interactive 'dialog'60, and the local dominance of one subset of Th cells could well influence the course of lesion progression and stability.
The anti-inflammatory cytokine transforming growth factor- (TGF-) is secreted by macrophages, smooth muscle cells and the subset of Th cells, Th3, that exerts regulatory functions. TGF- could be involved in plaque stabilization, because, in contrast to IFN-, it stimulates collagen synthesis and is fibrogenic. Indeed, inhibition of TGF- signaling by neutralizing antibodies led to larger lesion size with a less stable plaque phenotype70,
71.
Adding to the complexity of immune effects is the recent report that splenectomy of cholesterol-fed Apoe-/- mice led to significantly increased atherosclerosis, indicating that the spleen has anti-atherogenic activity72,
73. Further experiments in this study established that transfer of either purified B cells or T cells from the spleens of atherosclerotic Apoe-/- donors, which presumably were already 'educated' with respect to relevant atherosclerotic antigens, could rescue this effect. The mechanisms for these important findings remain to be elucidated. Similar effects may be found in humans, moreover, as long-term studies of soldiers who underwent splenectomy after trauma showed they had a twofold elevated incidence of coronary artery disease74.
Antigens in atherosclerosis
There is ample evidence of immune activation in the atherosclerotic lesion. Of specific interest is the finding that atherosclerotic lesions of Apoe-/- mice showed preferential expression of certain TCR-variable gene segments, suggesting the oligoclonal expansion of T cells and indicating that a limited number of antigens mediate specific proliferation75. What are the antigens involved? There are many candidates, including exogenous infectious pathogens such as bacteria and viruses, and endogenous proteins such as heat-shock proteins (Hsps) and 2GP-1 (ref. 76), which could lead to true autoimmune responses, modified artery wall proteins and oxLDL, as discussed above. There has been much interest in infectious agents as pathogens in atherogenesis19, such as Chlamydia pneumoniae, as patients with CVD have high titers of antibodies against these agents. In addition, C. pneumoniae, herpes simplex and cytomegalovirus have been detected in human lesions. Infectious agents are not necessary for lesion development in Apoe-/- mice40, although this does not rule out the possibility of a modifying function of infection in disease progression. Infection at a site distant from the lesion, such as gingivitis, could also influence cellular activation in the lesion through systemic events mediated by cytokines or antibodies77. Many Hsps of microbes (such as Hsp65) and humans (such as Hsp60) are highly conserved and show molecular mimicry. Consequently, antibodies against Hsp65 could target arterial cells, such as endothelial cells, which express Hsp60 in large amounts when exposed to stress78. Indeed, antibodies against Hsp65 and Hsp60 correlated with the progression of carotid disease in one study in humans79, and immunization of Ldlr-/- mice with Hsp65 promoted early atherosclerosis80, whereas mucosal administration of antigens led to tolerance and decreased atherogenesis81.
The most extensive data obtained so far support the idea that oxLDL is important as an antigen82. Oxidized LDL is present in the atherosclerotic lesions of all animal models and humans examined16 (Fig. 2). When polyunsaturated fatty acids of LDL phospholipids undergo peroxidation, a variety of highly reactive breakdown products are formed, such as MDA, which in turn can form covalent adducts with the lysine residues of the protein of LDL, apoB. In addition, reactive oxidation products derived from phospholipids, such as POVPC, can also form covalent adducts with apoB, and these adducts retain the intact phosphorylcholine (PC) headgroup6. These 'neo-self' determinants have been called oxidation-specific epitopes and have been shown to trigger a substantial humoral response specific to the given modification. For the measurement of these immune responses, two models of oxLDL are widely used: MDA−LDL, which is generated by the derivatization of LDL with MDA, yielding mainly MDA−lysine epitopes, and CuSO4-oxidized LDL (CuOx−LDL), which has many different oxidation-specific epitopes, although the oxidized PC-containing phospholipid seems to be an immunodominant epitope82,
83,
84.
Circulating IgG and IgM antibodies against both MDA−LDL and CuOx−LDL are present in the plasma of animals and humans and form immune complexes with oxLDL in atherosclerotic lesions48,
85. These antibodies closely correlate with atherosclerosis progression and regression in murine models and correlate with measures of lipid peroxidation6,
86,
87. In humans, many (but not all) studies have shown that plasma titers of antibodies against oxLDL epitopes, particular IgG, correlate with risk factors for CVD, and can even be used to predict the progression of carotid disease as well as myocardial infarction and death82. However, many variables affect such titers in humans, and the clinical utility of such measurements remains to be determined, particularly for the individual patient. Further evidence of the importance of oxLDL as an immunodominant antigen was obtained from the observation that 15% of CD4+ T cells cloned from human carotid atherosclerotic plaques specifically proliferated in response to oxLDL in a MHC class II−restricted way88. Compelling evidence that the immune response to oxLDL is important in modulating lesion formation comes from studies demonstrating that immunization of hypercholesterolemic rabbits and mice with MDA−LDL or CuOx−LDL reduced the progression of lesion formation26,
27,
28,
29,
30.
Molecular mimicry between PC of oxLDL and pathogens
This review began with the thesis that many of the immune responses found in atherosclerosis were in fact directed at specific components of inflammatory responses in general. One example of this was discovered by detailed study of a specific immune response noted in Apoe-/- mice. Because the titers of IgM autoantibodies against oxLDL are so high in these mice, it was possible to clone a panel of IgM monoclonal autoantibodies from their spleens that specifically bound to CuOx−LDL, but not native LDL, and specifically to oxidized phospholipids with the PC headgroup, such as POVPC (refs. 89,90). These monoclonal antibodies, such as the prototypic EO6, bind to intact oxLDL and block its binding and uptake by macrophages in a dose-dependent way, and specifically block the binding of oxLDL to the scavenger receptors CD36 and SR-B1 (refs. 90,
91,
92). In addition, POVPC covalently bound to bovine serum albumin could also block the uptake of oxLDL by these cells. These studies showed that oxidized phospholipids bearing the PC headgroup as ligands on oxLDL mediated uptake by macrophage scavenger receptors. A wide variety of PC-bearing oxidized phospholipids present in oxLDL have this property and are present in lesions93,
94. Furthermore, EO6 binds to apoptotic cells, which are known to be under oxidative stress and would be expected to express common oxidation-specific epitopes on their surface. EO6 blocks apoptotic cell uptake by macrophages, as does POVPC bound to bovine serum albumin95. Thus, PC-oxidized phospholipids represent a previously unrecognized PAMP, which is bound by EO6 as well as by the PRRs CD36 and SR-B1, and possibly others. These data strongly indicate a common innate immune response to apoptotic cells and oxLDL.
Further evidence in support of this hypothesis came from studies in which the DNA sequences of the antigen-binding domains of these IgM antibodies were determined. All of the cloned IgM antibodies against oxLDL (including EO6) were shown to be 100% homologous in their entire VH−VL regions to an IgA natural antibody cloned more than 30 years ago48. This antibody, made by the T15 clone, is directed against PC and confers optimal protection to mice against lethal infection with pathogens, such as Streptococcus pneumoniae96. PC, the headgroup of many phospholipids, is a prominent component of the capsular polysaccharide (C-PS) of many bacteria. T15(or EO6) bound equally well to C-PS and oxLDL, and each of these could compete for the binding to the other. T15 is a natural IgA antibody made by B1 cells that has the T15 idiotype, and even arises in mice grown in completely pathogen-free conditions. Thus, the actual positive selection agent for expansion of the B1 cell clone secreting T15/EO6 antibodies is postulated to be apoptotic cells or oxidized membranes or both; only later in life would pathogens contribute to further selection48. This represents positive selection of a natural antibody by an endogenous 'neo-self' antigen; there is an enhanced expansion of B cells secreting T15/EO6 clonospecific antibodies in Apoe-/- mice because of their atherogenic burden. Thus, PC-containing phospholipids, which are prominent components of mammalian membranes and LDL, are 'cryptic' epitopes in viable cells and native LDL. However, after oxidation, conformational changes occur that expose the PC epitope for recognition by EO6 (ref. 84) and by macrophage scavenger receptors. In contrast, the PC moiety of the pathogen C-PS is constitutively exposed for antibody recognition. Although this has not been tested, it is hypothesized that the pathogen PC also would be recognized by one or more of the same scavenger receptors (Fig. 3).
| | |
CRP is a member of the phylogenetically ancient and highly conserved pentraxin family of proteins that is thought to be important in primary, innate host defenses. It shows a rapid and substantial increase in plasma in response to inflammatory events such as tissue injury and infection. It is known to bind to PC of infectious pathogens such as S. pneumoniae and to mediate their clearance46. CRP binds to apoptotic cells as well, although the identity of the ligand involved has not been determined97. CRP also binds to oxLDL and exclusively to oxidized PC-containing phospholipids, and not to native LDL (ref. 98). In addition, the binding of CRP to apoptotic cells is also mediated by the PC moiety of oxidized phospholipids. Although CRP does not bind to native LDL, it does bind to LDL whose native structure was altered by plating on a microtiter well or by aggregation, indicating that PC exposure for CRP could occur by non-oxidative events as well98 (Fig. 3).
These observations indicate that CRP, the natural antibody EO6 and certain scavenger receptors of macrophages are all PRRs, with the PC moiety as the PAMP to which they bind. Thus, PC exposure generates a potent pathogen, mediating highly conserved and concerted innate responses. With infectious pathogens, these responses are undoubtedly protective, which is the basis for their conservation. In atherogenesis, however, the responses could have complex effects, and further study will be needed to sort out the potential benefits or 'penalties'. An increased titer of EO6 antibodies would be expected to be protective, as these antibodies potently block macrophage uptake of oxLDL. Although scavenger receptor−mediated uptake of oxLDL is itself apparently atherogenic, it could be protective if the concentration of oxLDL generated were minimized by low concentrations of LDL or inhibition of oxidative events. With CRP, the situation is equally complicated: CRP could bind to the PC moiety of oxLDL and block uptake by macrophages. However, CRP could also enhance the uptake of opsonized oxLDL through Fc receptors. In addition, ligand-bound CRP can activate complement pathways in a complex way, and can activate macrophages to release pro-atherogenic cytokines such as IFN- (ref. 46). Because CRP is known to be present in atherosclerotic lesions, and to co-localize with oxidized98 and otherwise modified LDLs, these observations indicate that CRP could be involved in complex ways in modulating atherogenesis. Undoubtedly, this complexity characterizes all immune interactions in the course of atherosclerosis.
Summary
There is now much evidence that both innate and adaptive immune mechanisms are involved in atherogenesis, as might be anticipated for a disease that is a chronic inflammatory process. Immune mechanisms have both protective and adverse effects in animal models. Elucidation of the pathways involved could lead to insights into pathogenic events that could explain in part the diversity in the expression of this disease in individuals apparently equal in regard to risk factors, such as plasma LDL. In turn, new therapeutic options could be developed, such as immunization with oxidation-specific epitopes of oxLDL, or interference with Th1-mediated pathways that lead to secretion of IFN-. Study of the action of the immune system in atherogenesis is in its infancy, and much remains to be learned. Most of the work so far has been done on experimental animals, mainly mice, and the relevance of these observations to human disease remains to be determined. However, lesion formation in humans occurs with much less atherogenic pressure and thus proceeds at a more leisurely pace than in hypercholesterolemic mice. Therefore, there may be more opportunity for immune mechanisms to function. That inflammatory markers are a powerful predictor of CVD in humans, autoantibody titers against epitopes of oxLDL and Hsp are increased in CVD patients, and antigen-specific T cells and immunoglobulins are present in lesions indicates that immune mechanisms are relevant to humans as well. The challenge will be to translate what has been learned already, and what will be learned through future experimental studies, to human populations.
REFERENCES
- Tuzcu, E.M. et al. High prevalence of coronary atherosclerosis in asymptomatic teenagers and young adults: Evidence from intravascular ultrasound. Circulation 103, 2705–2710 (2001). | PubMed | ISI | ChemPort |
- Leitersdorf, E., Tobin, E.J., Davignon, J. & Hobbs, H.H. Common low-density lipoprotein receptor mutations in the French Canadian population. J. Clin. Invest. 85, 1014–1023 (1990). | PubMed | ISI | ChemPort |
- Glass, C.K. & Witztum, J.L. Atherosclerosis: The road ahead. Cell 104, 503–516 (2001). | Article | PubMed | ISI | ChemPort |
- Hansson, G.K., Libby, P., Schonbeck, U. & Yan, Z.Q. Innate and adaptive immunity in the pathogenesis of atherosclerosis. Circ. Res. 91, 281–291 (2002). | Article | PubMed | ISI | ChemPort |
- Hansson, G.K. Immune mechanisms in atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 21, 1876–1890 (2001). | PubMed | ISI | ChemPort |
- Hörkkö, S. et al. Immunological responses to oxidized LDL. Free Radic. Biol. Med. 28, 1771–1779 (2000). | Article | PubMed | ISI | ChemPort |
- Ross, R. Atherosclerosis—an inflammatory disease. N. Engl. J. Med. 340, 115–126 (1999). | Article | PubMed | ISI | ChemPort |
- Libby, P., Ridker, P.M. & Maseri, A. Inflammation and atherosclerosis. Circulation 105, 1135–1143 (2002). | Article | PubMed | ISI | ChemPort |
- Williams, K.J. & Tabas, I. The response-to-retention hypothesis of atherogenesis reinforced. Curr. Opin. Lipidol. 9, 471–474 (1998). | Article | PubMed | ISI | ChemPort |
- Skalen, K. et al. Subendothelial retention of atherogenic lipoproteins in early atherosclerosis. Nature 417, 750–754 (2002). | Article | PubMed | ISI | ChemPort |
- Gaut, J.P. & Heinecke, J.W. Mechanisms for oxidizing low-density lipoprotein. Insights from patterns of oxidation products in the artery wall and from mouse models of atherosclerosis. Trends Cardiovasc. Med. 11, 103–112 (2001). | Article | PubMed | ISI | ChemPort |
- Pratico, D. Lipid peroxidation in mouse models of atherosclerosis. Trends Cardiovasc. Med. 11, 112–116 (2001). | Article | PubMed | ISI | ChemPort |
- Colles, S.M., Maxson, J.M., Carlson, S.G. & Chisolm, G.M. Oxidized LDL-induced injury and apoptosis in atherosclerosis. Potential roles for oxysterols. Trends Cardiovasc. Med. 11, 131–138 (2001). | Article | PubMed | ISI | ChemPort |
- Berliner, J.A., Subbanagounder, G., Leitinger, N., Watson, A.D. & Vora, D. Evidence for a role of phospholipid oxidation products in atherogenesis. Trends Cardiovasc. Med. 11, 142–147 (2001). | Article | PubMed | ISI | ChemPort |
- Marathe, G.K., Prescott, S.M., Zimmerman, G.A. & McIntyre, T.M. Oxidized LDL contains inflammatory PAF-like phospholipids. Trends Cardiovasc. Med. 11, 139–142 (2001). | Article | PubMed | ISI | ChemPort |
- Witztum, J.L. & Steinberg, D. The oxidative modification hypothesis of atherosclerosis: Does it hold for humans? Trends Cardiovasc. Med. 11, 93–102 (2001). | Article | PubMed | ISI | ChemPort |
- Suzuki, H. et al. A role for macrophage scavenger receptors in atherosclerosis and susceptibility to infection. Nature 386, 292–296 (1997). | Article | PubMed | ISI | ChemPort |
- Febbraio, M. et al. Targeted disruption of the class B scavenger receptor CD36 protects against atherosclerotic lesion development in mice. J. Clin. Invest. 105, 1049–1056 (2000). | PubMed | ISI | ChemPort |
- Libby, P., Egan, D. & Skarlatos, S. Roles of infectious agents in atherosclerosis and restenosis: An assessment of the evidence and need for future research. Circulation 96, 4095–4103 (1997). | PubMed | ISI | ChemPort |
- Landmesser, U. & Harrison, D.G. Oxidant stress as a marker for cardiovascular events—Ox marks the spot. Circulation 104, 2638–2640 (2001). | PubMed | ISI | ChemPort |
- Dansky, H.M., Charlton, S.A., Harper, M.M. & Smith, J.D. T and B lymphocytes play a minor role in atherosclerotic plaque formation in the apolipoprotein E-deficient mouse. Proc. Natl. Acad. Sci. USA 94, 4642–4646 (1997). | Article | PubMed | ChemPort |
- Daugherty, A. et al. The effects of total lymphocyte deficiency on the extent of atherosclerosis in apolipoprotein E-/- mice. J. Clin. Invest. 100, 1575–1580 (1997). | PubMed | ISI | ChemPort |
- Song, L., Leung, C. & Schindler, C. Lymphocytes are important in early atherosclerosis. J. Clin. Invest. 108, 251–259 (2001). | Article | PubMed | ISI | ChemPort |
- Reardon, C.A. et al. Effect of immune deficiency on lipoproteins and atherosclerosis in male apolipoprotein E-deficient mice. Arterioscler. Thromb. Vasc. Biol. 21, 1011–1016 (2001). | PubMed | ISI | ChemPort |
- Zhou, X., Nicoletti, A., Elhage, R. & Hansson, G.K. Transfer of CD4+ T cells aggravates atherosclerosis in immunodeficient apolipoprotein E knockout mice. Circulation 102, 2919–2922 (2000). | PubMed | ISI | ChemPort |
- Palinski, W., Miller, E. & Witztum, J.L. Immunization of low density lipoprotein (LDL) receptor-deficient rabbits with homologous malondialdehyde-modified LDL reduces atherogenesis. Proc. Natl. Acad. Sci. USA 92, 821–825 (1995). | PubMed | ChemPort |
- Ameli, S. et al. Effect of immunization with homologous LDL and oxidized LDL on early atherosclerosis in hypercholesterolemic rabbits. Arterioscler. Thromb. Vasc. Biol. 16, 1074–1079 (1996). | PubMed | ISI | ChemPort |
- Freigang, S., Hörkkö, S., Miller, E., Witztum, J.L. & Palinski, W. Immunization of LDL receptor-deficient mice with homologous malondialdehyde-modified and native LDL reduces progression of atherosclerosis by mechanisms other than induction of high titers of antibodies to oxidative neoepitopes. Arterioscler. Thromb. Vasc. Biol. 18, 1972–1982 (1998). | PubMed | ISI | ChemPort |
- George, J. et al. Hyperimmunization of apo-E-deficient mice with homologous malondialdehyde low-density lipoprotein suppresses early atherogenesis. Atherosclerosis 138, 147–152 (1998). | Article | PubMed | ISI | ChemPort |
- Zhou, X., Caligiuri, G., Hamsten, A., Lefvert, A.K. & Hansson, G.K. LDL immunization induces T-cell-dependent antibody formation and protection against atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 21, 108–114 (2001). | PubMed | ISI | ChemPort |
- Nicoletti, A., Kaveri, S., Caligiuri, G., Bariaety, J. & Hansson, G.K. Immunoglobulin treatment reduces atherosclerosis in apo E knockout mice. J. Clin. Invest. 102, 910–918 (1998). | PubMed | ISI | ChemPort |
- Medzhitov, R. & Janeway, C.A., Jr. Decoding the patterns of self and nonself by the innate immune system. Science 296, 298–300 (2002). | Article | PubMed | ISI | ChemPort |
- Medzhitov, R. Toll-like receptors and innate immunity. Nature Rev. Immunol. 1, 135–145 (2001). | Article | PubMed | ChemPort |
- Gosling, J. et al. MCP-1 deficiency re
|
