| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biochemsitry, Institute of Medical Biology, Faculty of Medicine, University of Tromsø, Tromsø, Norway
ABSTRACT I. INTRODUCTION II. INITIATION OF ATHEROSCLEROSIS III. EXPERIMENTAL EVIDENCE THAT MONOCYTE RECRUITMENT TO THE INTIMA IS LINKED TO ATHEROGENESIS A. Early Monocyte Recruitment to the Intima B. Role of Adhesion Molecules 1. Selectins and VCAM-1 2. Integrins and ligands 3. Other factors C. Role of Chemoattractants 1. MCP-1 2. MCP-1 receptors 3. Factors affecting the production of MCP-1 and other chemoattractants IV. ROLE OF LOW-DENSITY LIPOPROTEIN A. Early, Low-Grade Events B. Modulation of LDL C. Proinflammatory Reactions Associated With LDL Oxidation 1. Impact of oxLDL V. MONOCYTE DIFFERENTIATION AND THE ROLE OF THE EXTRACELLULAR MATRIX IN ATHEROGENESIS A. Macrophage Formation B. Foam Cell Formation C. Roles of Extracellular Matrix and Metalloproteinases in Atherogenesis VI. MONOCYTES/MACROPHAGES AND THE ROLE OF THEIR ACTIVATION PRODUCTS A. Regulation of Activation B. Autocrine- and Paracrine-Mediated Activation Reactions in Monocytes C. Roles of Cytokines Produced by Monocytes/Macrophages 1. TNF-{alpha} 2. IL-1{beta} 3. IL-4 and IL-13 4. IL-6 5. IL-12 6. IFN-{gamma} 7. IL-10 8. Transforming growth factor-{beta} D. CD14 Polymorphism on the Gene of the CD14 Receptor VII. CELLULAR SIGNALS INVOLVED IN MONOCYTE ACTIVATION A. Role of Phospholipases B. Role of Phospholipases in Atherosclerosis 1. sPLA2 2. cPLA2 C. Pathways for Arachidonic Acid Conversion 1. Cyclooxygenases 2. Prostanoid and thromboxane products 3. Isoprostanes 3. Cytochrome P-450-derived products 4. Lipoxygenases and LO-dependent products D. Role of Lipoxygenases in Atherogenesis 1. The lipoxygenase products LTB4, LTC4, and LTD4 E. Availability of Arachidonic Acid F. PAF 1. Production and mechanism of action 2. Biological and pathophysiological effects G. Eicosanoid Products Involved in the Activation of Monocytes? H. PPAR-{gamma} as Regulator of Monocyte/Macrophage Function VIII. INFECTION, MONOCYTES, AND ATHEROSCLEROSIS A. C-reactive Protein and Its Relation to Atherosclerosis IX. THE PROINFLAMMATORY PLATELET AND ITS ROLE IN THE EARLY PHASE OF ATHEROGENESIS X. CONCLUSIONS
 |
ABSTRACT |
|---|
 Top NextReferences |
|---|
|
I. INTRODUCTION |
|---|
|
TopPreviousNextReferences |
|---|
Recruitment of monocytes and lymphocytes from the peripheral blood to the intima of the vessel wall is a primordial event in atherogenesis (see sect. IIIA), an event that appears to depend on the local presence of high amounts of LDL. As the LDL accumulates, their lipids and proteins undergo oxidation and glycation. Cells in the vessel wall seem to interpret the change as a danger sign, and they call for reinforcement from the body's defense system. These events appear to promote upregulation of adhesion molecules on the endothelial cells (EC), particularly vascular cell adhesion molecule-1 (VCAM-1) and intercellular adhesion molecule-1 (ICAM-1). Thus monocyte and lymphocyte recruitment is initiated, leading to enhanced transmigration of monocytes, upregulated exposure of adhesion molecules on a variety of cells (see sect. IIIB), and chemoattractant production and release (see sect. IIIC). These are all essential elements of the transfer of monocytes to the intima and the concurrent differentiation of these cells into macrophages. Available LDL is a prerequisite for the further conversion of macrophages into lipid-loaded macrophages, major residents of the fatty streak formed just underneath the endothelium of the vessel wall. LDL and its modified forms (oxidized, acylated, etc.; see sect. IV, A and B) are of particular interest, since the modification of LDL is associated with inflammatory reactions (see sect. IVC), amplifying proinflammatory events already initiated due to the adherence and transmigration of monocytes and lymphocytes into the intima.
The emerging notion that chronic infections may unleash atherogenic trigger mechanisms (see sect. VIII) is suggestive of a very important role of monocytes in lesion formation by way of their proficiency in generating proinflammatory products. This warrants focus on the cellular signal transduction network of monocyte activation, including the early phase entailing phospholipase activation and release of arachidonic acid (see sect. VII). Transgenic mice have opened access to a substantial body of information regarding the roles of the various signal-systems of importance in atherosclerosis (see sects. III, A and B, IVC, VB, VIC, and VII, B-D). Much of the work has focused on the enzymes involved in the conversion of arachidonic acid to leukotrienes and hydroxyfatty acids (lipoxygenases; see sect. VII, C and D) and prostaglandins (cycloxygenases; see sect. VIIC), lending support to the notion that lipoxygenases may have a vital role in lesion formation (see sect. VIID), whereas the cyclooxygenases appear less important. According to recent evidence, the thromboxane A2 (TxA2) receptor may have an important function in promoting lesion formation (see sect. VIIG), unrelated to the binding of its major ligand, as demonstrated by the noninterference of acetyl salicylic acid (ASA).
In order that monocytes may promote atherosclerosis, a number of unfavorable factors or conditions may have to merge. Polymorphism of the gene of the monocyte CD14 receptor (see also sect. VID) may be part of the explanation why there are major risk factors unrelated to high cholesterol. This receptor mediates not only the toxic effects of bacteria (lipopolysaccharide; LPS) in chronic infections (e.g., Chlamydia pneumonia), but also the cytokine-like effect of heat shock proteins, which is associated with coronary heart disease (CHD) (see sect. VIII). Platelet activating factor (PAF) (see sect. VIIF) and cytokines, particularly tumor necrosis factor-
(TNF-), interleukin (IL)-1
, IL-6 (see sect. VIC), and growth hormones, promote the inflammatory reactions associated with atherogenesis. Another significant regulator of monocyte/macrophage function is the peroxisome proliferator-activated receptor-
(PPAR), studies on which might shed further light on processes central to atherogenesis (see sect. VIIH).
It has not been the intention in this article to cover all aspects of atherosclerosis. For example, changes in the endothelium during atherogenesis, the role of smooth muscle cells and their interactions with macrophages, T lymphocytes, and fibroblasts and their activation products, prominent in the later phase of atherogenesis, have not been presented. Furthermore, the events leading to plaque rupture and thrombosis have not been included.
|
II. INITIATION OF ATHEROSCLEROSIS |
|---|
|
TopPreviousNextReferences |
|---|
Such recruitment of leukocytes continues for as long as hypercholesterolemic conditions prevail (362), without any apparent impairment of the vascular endothelium, the latter contributing to leukocyte recruitment by upregulated expression of specific leukocyte adhesion molecules, notably VCAM-1 and ICAM-1 (see sect. III). In baboons, monocyte margination or attachment observed over normal vessel areas or over plaques was not associated with morphological evidence of endothelial injury, whether under normal or hypercholesterolemic conditions. Migration of these cells through continuous aortic endothelium occurred primarily at junctional sites, between endothelial cells (464). Furthermore, strong interaction between monocytes and components of the extracellular matrix (ECM), e.g., collagen and proteoglycans, is probably a prerequisite for the formation of lipid-laden macrophages (foam cells) (238).
LDL particles trapped in the artery are prone to advancing oxidation, rendering them recognizable by scavenger receptors present on the surface of macrophages and thus targets for internalization by these cells (186, 201, 255, 346, 370, 499), a process ultimately leading to the formation of foam cells. The extent of LDL modification and hence the vigor of this rampant process may vary greatly (123, 186, 370). LDL modified and taken up by the macrophages activates the nascent foam cells, inducing release of inflammatory mediators such as TNF-, IL-1, IL-8, and macrophage colony stimulating factor (MCSF). This release in turn leads to increased transcription of the LDL-receptor gene and consequently enhanced binding of LDL to the endothelium and smooth muscle cells (194, 503), feeding the process even further.
Part of the proinflammatory effect of modified LDL on endothelial cells indirectly derives from its chemotactic effect on monocytes (105), and furthermore, it can upregulate the endothelial cell genes for MCSF and monocyte chemotactic protein-1 (MCP-1) (478). Either mechanism elicits an expanding number of monocyte-derived macrophages populating the artery wall.
The arterial wall may be subject to immunological injury leading to conditions favoring atherosclerosis development (81). Recent studies have implicated several immune complexes as proatherogenic. The fatty streak type of lesion typical of the earliest stage, frequently occurring in infants and young children (367), is of a purely inflammatory process involving only monocyte-derived macrophages and T lymphocytes (497). Thus the attachment of monocytes and T lymphocytes to the endothelium followed by their migration into the intima is one of the first and most crucial steps in lesion development. Immune complex-bound antigens taken up by phagocytes via Fc receptors are processed and presented to T cells in the context of major histocompatibility complex (MHC) class I or II molecules. Such antigen processing and presentation greatly enhance T-cell stimulation and the subsequent immune responses (1). This scenario is in accordance with the observation that atherosclerotic lesions are rich in immunoglobulins, which can be found in proximity to macrophages even in very early fatty streaks (390). These immunoglobulins comprise specific autoantibodies to neoepitopes arising during oxidative or other types of injury, some of which have been identified in immune complexes incorporating oxidized LDL (583).
The emerging lesson from transgenic mouse models is that monocytes and macrophages partake crucially in the development of atherosclerotic lesions, whereas results pertaining to T and B cells are more ambiguous and model dependent, suggesting that under certain conditions these cell types may also promote lesion formation (430, 491).
The enrichment of white blood cells in lesion-prone areas has been shown to be of primarily mononuclear origin (98, 439, 483, 549, 587). The prevalence of such changes has been affirmed in a normocholesterolemic rabbit model, where adhered white blood cells were detected in the lesion-prone flow divider regions of the large abdominal branches (320). Interestingly, lesion-rich aortic and coronary artery segments had significantly greater numbers of mast cells in the adventitia compared with those with a normal intima (19). In normal aortic segments, higher numbers of mast cells were located in the lesion-prone than in the lesion-deficient regions. These observations seem to suggest that part of the proatherogenic potential latent in a vessel wall derives from mast cells and their released products, such as histamine and tryptase.
The model of atherogenesis initiation outlined above is mainly based on experimental animal models with high rate of developing lesions, i.e., fat-fed and genetically hyperlipidemic animals. One may therefore question the validity of such animal models regarding the uptake of oxidized LDL and the effects exerted by oxidized LDL. In line with this skepticism is the proposal of a concept of lesions arising in preexisting intimal masses (465): "Given the evidence for a concordance of the distribution of lesions in adult humans, it seems likely that some property of the intimal cells accounts for localization of lesions." Perhaps the simplest hypothesis, as put forward by Williams and Tabasas (565), is that cells making up the intimal masses have special properties and contribute to lipid accumulation at focal sites. Whether these intimal cells are part of the intimal cellular infiltration and connective tissue described in the coronary arteries of male children, adolescents and young adults remains to be clarified. Interestingly, black people and women have less than average cellular infiltration and connective tissue in their arteries. Whatever factors that are favoring LDL oxidation and macrophage lipid accumulation and retention, even at normal plasma LDL levels, might explain the emergence of the initial cluster of macrophage foam cells (336).
|
III. EXPERIMENTAL EVIDENCE THAT MONOCYTE RECRUITMENT TO THE INTIMA IS LINKED TO ATHEROGENESIS |
|---|
|
TopPreviousNextReferences |
|---|
An initial event in atherogenesis is endothelial cell activation, probably mediated by atherogenic lipoproteins such as remnants, modified LDL or oxidized LDL, particularly potent forms of which are those modified by oxidation or glycation (in diabetes) or trapped in immune complexes. Such activation of the endothelium causes upregulation of endothelial cell adhesion molecules and selectins, promotion of oxygen radical formation, increased apoptosis, and reduced endothelium-dependent relaxation.
The multistep process of localized accumulation of leukocytes, which is regulated by the expression of specific adhesion molecules, may leave the endothelium unperturbed. Thus focal recruitment and attachment of monocytes, whether over plaques or normal areas, was not associated with evidence of endothelial injury as shown in normo- and hypercholesterolemic baboons (464). In addition to monocyte recruitment, lymphocyte recruitment is one of the earliest detectable cellular responses in the process leading to the formation of atherosclerotic lesions.
In a transgenic mouse model, two types of endothelial cell-derived adhesion molecules, VCAM-1 and ICAM-1, were shown to play an important role at sites where lesions eventually form (231, 362). Thus homozygous apolipoprotein E-deficient (Apo E -/-) mice that develop atherosclerotic lesions much like those found in humans were compared with normal control mice. Whereas in the latter specific VCAM-1 staining was weak and limited to sites of altered blood flow, in the Apo E -/- mice significant amounts of VCAM-1 appeared to be localized across the surface of EC at lesion-prone sites. The expression of VCAM-1 preceded lesion formation and correlated positively with the extent of exposure to plasma cholesterol. ICAM-1 too was expressed across the surface of EC and microvilli at lesion-prone sites, but at a rate insensitive to plasma cholesterol levels. Similarly, in a recent study, recruitment of circulating white blood cells by endothelial adhesion molecules was suggested to be more important during lesion initiation than during the late phase of rapid lesion growth (294). In rabbits, similar upregulation of endothelial adhesion molecules at lesion-prone as well as already atherosclerotic sites has been reported (231).
There is supporting evidence that many of these adhesion molecules are expressed by cells in human and experimental atherosclerotic lesions and that expression of adhesion molecules such as VCAM-1 is temporally related to lesion initiation and progression in animal models (for commentary, see Ref. 443).
The mechanism whereby LDL interacts with the endothelium is not quite understood. In one study it was shown that binding of native LDL (n-LDL) to the LDL receptor triggers a rise in intracellular calcium which acts as a second messenger in inducing VCAM-1 expression in human coronary aortic cells (9).
VCAM-1 expressed before leukocyte accumulation can initiate both monocyte and lymphocyte tethering and rolling (for leukocyte recruitment at the endothelium, see excellent review by Ley, Ref. 297) and can induce firm adhesion in the absence of chemoattractants. ICAM-1 exposed on microvilli at lesion-prone sites enhances firm adhesion of leukocytes that have reached the primary, transient adhesion step, whereas ICAM-1 more strongly expressed across the endothelial surface promotes cellular arrest where cell-cell contact has already been established. Finally, transendothelial leukocyte migration entailing subendothelial homing and formation of early fatty streaks is supported by the universally expressed platelet-endothelial cell adhesion molecule-1 (PECAM-1) and also possibly enhanced by locally expressed VCAM-1 (362).
A more detailed account of the evidence for the functional role of adhesion molecules and chemokines in disease models is given below.
The most comprehensive body of evidence for an important role of adhesion molecules in atherogenesis has emerged from various studies using transgenic mice and a few transgenic rabbit models (286). Thus strong evidence that P-selectin is critically involved in initiation of the process was recently reported based on studies using P-selectin-deficient mice (P -/-) back-crossed onto a C57BL16 background (133). When subjected to a high-fat diet for 20 wk, these mice were significantly less prone to fatty streak formation than were the wild type. Fatty streaks are the first visible sign that foam cells accumulate within the intima, close to the surface of the vessel wall. Furthermore, in the same study, the hypercholesterolemia-prone offspring of P-selectin deficiency (P -/-) animals or a combination of P-and E-selectin deficiency (P/E -/-) animals interbred with mice lacking the LDL receptor (LDLR -/-), were fed an atherogenic diet for 8, 20-22, and 37 wk. At 8 wk, mice with combined P -/- and LDLR -/- had developed significantly smaller fatty streaks than was evident in their half-siblings expressing P-selectin. However, after a more prolonged exposure to the high-fat diet and progression of the lesions to the fibrous plaque stage, the two forms of offspring were no longer discernibly different.
LDLR -/- offspring deficient also in P-selectin as well as E-selectin were even more profoundly protected against lesion development, even into the fibrous plaque stage (205).
Apo E-deficient mice spontaneously develop lesions when fed a normal Chow diet. They have very high cholesterol levels, on the order of that of hypercholesterolemic humans. The protective effect of P-selectin deficiency in Apo E -/- mice appeared to be more pronounced and long lasting than was the case in LDLR -/- mice (132). Even at 4 mo of age the size of fibrous plaque lesions in P-selectin -/-, Apo E -/- mice was one-fourth of that of Apo E -/- mice having wild-type P-selectin. Hartwell and Wagner (205) emphasize the critical role of monocytes in atherogenesis suggested by these findings.
Thus the studies on Apo E deficiency mice indicate that P-selectin/VCAM-1-dependent leukocyte rolling is a mandatory step in the early development of atherosclerotic lesions. Further evidence to this end was obtained ex vivo using an isolated carotid artery preparation from 10- to 12-wk-old Apo E -/- and C57BL16 wild-type mice fed a Western-type diet (21% fat wt/wt) for 4-5 wk (424). Adherence and rolling of cells of the mononuclear U937 line on this surface were significantly impaired when P-selectin or its ligand P-selectin glycoprotein-1 ligand (PSGL-1) were blocked using specific antibodies. It was also shown that rolling velocities increased, corresponding to a weaker adhesion, when 4-integrin or VCAM-1 of the mononuclear cells had been blocked using specific antibodies. This was interpreted as evidence that the interaction between 4-integrin and VCAM-1 has a stabilizing effect on rolling interaction, prolonging the monocyte transit time.
In another study it was shown that endothelin-1 (ET-1), which is a potent vasoconstrictor and postulated to play a role in hypertension (502), ischemia-reperfusion injury (404), and atherosclerosis (148), can directly promote significant leukocyte adherence and rolling in a P-selectin-dependent reaction (453).
In rats subjected to dietary-induced hypercholesterolemia, ICAM-1 expression was upregulated mainly in the lesion-prone areas of aorta during the early stages of atherogenesis. This was associated with a pronounced recruitment of monocytes and T lymphocytes to the intima (552). Prior injections of specific antibodies to ICAM-1/lymphocyte function antigen-1 (LFA-1) significantly reduced monocyte adherence and migration into the intima.
In another study, ApoE-deficient mice (apo E -/-) maintained on an atherogenic Western diet were subjected to arterial injury, and posttrauma effects of the concurrent presence of additional ICAM-1 or P-selectin deficiencies were investigated (325). After 5 wk, a P-selectin deficiency linked 94% inhibition of neointima formation was found, whereas ICAM-1 deficiency gave no discernible protection against injury-induced plaque formation in these mice. It was concluded that absence of P-selectin but not of ICAM-1 reduces the plaque area and that P-selectin is critical for monocyte recruitment to sites of neointima formation after arterial injury.
Very late antigen 4 (VLA-4) is the ligand for VCAM-1 and fibronectin containing segment-1 (CS-1) (86, 138, 189, 555) (Table 1). Evidence has been found favoring the notion that VLA-4 has an important role in regulating leukocyte entry into early (479) as well as advanced (397) lesions. It was suggested that the VLA-4 integrin plays an important role in the initial phase of atherosclerotic lesion formation and lipid accumulation (479). According to the evidence currently available, it seems plausible that P-selectin and its ligand PSGL-1 together with VCAM-1 and its ligand VLA-4 may be the most important adhesion molecules involved in monocyte recruitment to atherosclerotic lesions (228).
|
Von Willebrand factor (vWF) has been suggested to be potentially proatherogenic, due to its important role in platelet functioning (114, 164, 504) and in regulating factor VIII in blood (339, 343). Evidence in support of this notion was recently found in studies using transgenic mice deficient in the LDLR (LDLR -/-) as well as vWF (vWF -/-). When fed a Chow diet, these double deficiency mice had a 50% reduction in leukocyte rolling compared with rats with the LDLR -/- trait only, dropping to 20% when an atherogenic fat diet was given (342). The high reduction in leukocyte rolling when a Chow diet was administered was associated with a 50% size reduction of lesions in the aortic sinus. Furthermore, between the superior mesenteric artery and the renal artery, vWF -/- animals were markedly less liable to lesion formation than were wild-type ones.
A list of adhesion molecules and ligands and their leukocyte and endothelial cell targets is shown in Table 1.
Certain adhesion molecules are currently seen as essential for the recruitment of monocytes to the intima. However, according to several studies it appears that a number of additional factors including several chemoattractants are needed in the orchestration of the process (for reviews, see Refs. 27, 315, 428). The most relevant chemoattractants pertaining to atherogenesis will be discussed in the following sections.
Chemokines or chemotactic cytokines belong to an expanding family of structurally related small protein molecules, allocated to subgroups (C, CC, CXC, CX3C) according to the number of and spacing of cysteine residues in the NH2-terminal region. They are critically involved in directing leukocyte trafficking and activation. MCP-1 emerges as probably having some subordinate role in relation to atherosclerosis. Thus high expression of MCP-1 was found in human atherosclerotic lesions (428). Transgenic mice overexpressing MCP-1 had threefold increased oxidized lipid and revealed increased immunostaining for macrophage cell surface markers unique for the activated state (6). It was concluded that MCP-1 expression mediates enhanced atherogenesis, by increasing macrophage numbers as well as accumulation of oxidized lipid. Corroborative evidence for this notion was found using transgenic C57BL/6 mice deficient in apolipoprotein apo (a). When fed a high-fat diet these mice overexpressed MCP-1 and in a correlating manner macrophages accumulated in their vasculature (431). Correspondingly, when subjecting LDLR-deficient mice to a high-cholesterol diet, the ensuing hypercholesterolemia rapidly triggered MCP-1 expression in resident macrophages. Additional numbers of macrophages expressing MCP-1 accumulated over time, indicating that MCP-1 may initiate as well as amplify monocyte recruitment to the artery wall during early atherogenesis (272). This is consistent with the 50% reduction in lesion formation found in MCP-1-deficient mice relative to wild type (181). Recently, it was shown that MCP-1 induces proliferation and IL-6 production in human smooth muscle cells by differential activation of nuclear factor 
B (NFB) and activator protein-1 (AP-1), which may suggest that some hitherto unrecognized mechanism may be involved in the proatherogenic effect of MCP-1 (542). However, overexpression of MCP-1 at the vessel wall was not sufficient to generate lesion formation in rabbits fed normal Chow, whereas such overexpression under hypercholesterolemic conditions induced infiltration of monocytes/macrophages and subsequent lesion formation in the vessel wall (366). On the other hand, hypercholesterolemia by itself did not cause lesion formation in rabbits with normal MCP-1 despite upregulation of VCAM-1 and ICAM-1. It was concluded that activation of other factors induced by hypercholesterolemia is required.
The function of MCP-1 is totally dependent on its receptors. In monocytes three different receptors, CCR1, CCR2, and CCR5, have been identified, the former two of which were the only ones present in granulocytes (for a review, see Refs. 56, 241, 428). It has been proposed that CCR2 serves as the principal MCP-1 receptor. It is upregulated by cytokines and by LDL (202). Mice deficient in Apo E as well as CCR2 (ApoE -/-,CCR2 -/-) showed a 50% reduction in macrophage recruitment compared with wild type after 5 wk on a Western diet (62, 110). Currently very little is known pertaining to the other receptors for MCP-1 and their relative contribution to monocyte recruitment.
3. Factors affecting the production of MCP-1 and other chemoattractants
Adhesion molecules that tether circulating leukocytes to endothelial cells may also serve as transducers or modulators of incoming signals of cellular activation. P-selectin upregulates the secretion of MCP-1 and TNF- from monocytes stimulated with PAF (560). Furthermore, IL-8 and MCP-1 were induced in monocytes by thrombin-activated platelets exposing P-selectin (559). IL-8 is produced both in monocytes and granulocytes and has been thought to act predominantly on neutrophils. However, recently it was found that it brought on firm adhesion of rolling monocytes onto monolayers expressing E-selectin, in the same manner as did MCP-1 (175), a faculty not shared by related chemokines. The production of IL-8 in leukocytes is upregulated by epinephrine-stimulated platelets (142). The prevailing perception of IL-8 as rather peripherally involved in atherogenesis may have to be corrected in view of these later findings. It could turn out more important for proinflammatory reactions and atherogenesis than anticipated, perhaps via some unexpected role in monocyte recruitment.
Monocyte-derived macrophages in the intima of the vessel wall generate MCP-1 and other chemoattractants important for the transendothelial migration of monocytes to the intima. Lysophosphatidylcholine (lysoPC) formed by the oxidation of LDL particles is a potent chemoattractant (87, 356, 361, 420), potentially enhancing the extravasation process. Moreover, since lysoPC serves to upregulate a series of proinflammatory products in monocytes/macrophages, it could have an impact on proinflammatory reactions taking place in a vessel wall accumulating oxidized LDL (87, 356, 361).
Although it has been proposed that MCP-1 serves as a principal chemokine directing monocyte infiltration (45, 373, 515, 563, 582), it may be anticipated that this role is shared somehow with the several other chemokines expressed in atherosclerotic plaques. Prominent among the latter are the chemoattractants for monocytes, MIP-1, MIP-1, and RANTES (regulated upon activation, normal T-cell expressed) (532). Their common receptor CCR5 is progressively expressed during monocyte differentiation (531). Hence, in contrast to CCR2, the main function of which is in the recruitment of monocytes from the peripheral blood, CCR5 and its ligands appear to have their main role in affecting macrophages already established within the lesions (for a review, see Ref. 428). MIP-1 and -1 colocalize in atherosclerotic plaques (563) and are expressed by T cells in macrophage/foam cell-rich areas of the plaque. This may be signifying chemokine-mediated cross-talk between T cells and macrophages, effectively promoting migration of macrophages more closely to the inflammatory scene.
Macrophages are the richest source of chemokines in atherosclerotic lesions (13, 45, 373, 429, 515, 550, 582). When monocytes are stimulated by the vast array of agonists present at the site of foam cell formation, they too start producing chemokines. The agonists comprise inflammatory cytokines, such as TNF, IL-1, IL-6, IL-2, etc., produced by macrophages and T cells (304, 327), and modified LDL particles (105, 265, 511, 519). Thrombin-activated platelets induce the expression of chemokines MCP-1 and IL-8 in monocytes (559).
|
IV. ROLE OF LOW-DENSITY LIPOPROTEIN |
|---|
|
TopPreviousNextReferences |
|---|
Cholesterol is transported in the circulation by plasma lipoproteins. The principal cholesterol carrier LDL serves as an exogeneous source of cholesterol and other cellular nutrients for hepatic and various extrahepatic tissues, where it is taken up by receptor-mediated endocytosis. Alternatively, LDL may be entrapped extracellularly in arteries, thus being subjected to a milieu conducive to various kinds of enzymatic and chemical modifi-cation. Early stages of arterial lipoprotein modification, marked by the generation of bioactive lipid peroxidative products, may occur without any apparent change in cellular receptor recognition of the particles.
Monocyte-derived macrophages in arteries express cell surface receptors for LDL (LDL-R) as well as scavenger receptors for modified LDL (SR-A, CD36, CD68) (Table 2). Although minimally oxidized LDL particles appear physically indistinguishable from native plasma LDL, they bear a cargo of bioactive molecules. The latter have been shown to cause a 60% increase in arachidonic acid and a 100% increase in 12 (S-HETE) release from EC, associated with induced binding of monocytes to EC (219).
|
Several studies have established that minimally modified LDL (MM-LDL) produces a distinct pattern of EC activation (47, 256, 394). Transcriptional activation of the macrophage-colony stimulating gene by MM-LDL was shown to be mediated by NFB activation (423). This was further confirmed in a study assessing responses to MMLDL by examining the expression of inflammatory genes involved in atherogenesis, including MCP-1 and MCSF, and the oxidative stress gene, heme oxygenase-1 (HO-1) (478). Furthermore, studies on EC isolated from the aorta of inbred mice with different susceptibilities to diet-induced atherosclerosis revealed that EC from the susceptible mouse strain C57BL/6J exhibited dramatic transcriptional induction of the inflammatory genes, whereas EC from the resistant strain showed little or no such induction (478). Altogether this suggests that genetic factors of the vessel wall are of importance in atherogenesis.
Lysosphosphatidic acid (LPA) has been identified as a bioactive compound formed in mildly oxidized LDL and minimally modified LDL. Thus LPA initiated platelet activation and stimulated endothelial stress-fiber and gap formation (482). LPA receptor antagonists prevented platelet and endothelial cell activation by mildly oxidized LDL.
Evidence from human surgical and autopsy material as well as from experimental animal specimens indicates that monocyte and T-lymphocyte adherence is preceded by the deposition of lipids underneath the endothelial cells (14). The flux of LDL into the intima is determined mainly by the LDL concentration in plasma. LDL enters the intima and inner media primarily from the luminal side, passing structures such as the endothelium and the subendothelial basement membrane. Under normal conditions these structures are the major barriers for the entry of any kind of plasma macromolecule into the the inner media (217, 400, 466). The transport of LDL into the intima is probably not dependent on any specific EC receptors. Penetration of intact LDL particles into the arterial wall may occur by vesicular ferrying through the endothelium or by passive sieving through pores in or between endothelial cells. Ongoing monocyte trafficking has been suggested to somehow enhance the transport of LDL from plasma to the intima (for a review, see Ref. 375).
The spherical LDL particle has a cholesteryl ester-rich core and a surface dominated by free cholesterol, phospholipids, and apolipoprotein B100, the normal ligand for the LDL receptor. It is anticipated that under most circumstances LDL circulating in plasma is protected from oxidation by the presence of antioxidants. In contrast, the less protective milieu of the arterial wall renders the particle significantly more vulnerable and subject to oxidation.
Modification of LDL via oxidative processes is believed to be a prerequisite for the development of atherosclerosis. Oxidation of the particle in the arterial wall is thought to be a complex reaction involving several cell types, including monocytes, macrophages, granulocytes, lymphocytes, endothelial cells, and SMC. Typically oxidation may take place in microenvironments that are exhausted in antioxidants such as vitamin E, carotenoids, ubiquinol, etc. (146). The importance of LDL oxidation in lesion formation was recently documented in ApoE-deficient mice consuming red wine for 2 mo (23) with a subsequent 40% reduction in basal LDL oxidation, and a similar decrement in LDL oxidizability and aggregation associated with a 35% reduction in lesion size. The resistance to oxidation was associated with an accumulation of flavonoids in the mouse macrophages whereby their capacity to oxidize LDL was reduced and they took up about 40% less LDL than macrophages from placebo-treated mice.
Polyunsaturated fatty acids present in LDL are oxidatively converted to lipid hydroperoxides, which are subsequently cleaved forming reactive aldehydes. The latter may covalently modify apolipoprotein B100 by forming Schiff's bases with exposed lysine amino groups, thus masking the positive charge and in effect increasing the net negative particle charge, rendering the particles recognizable by macrophage scavenger receptors and subject to unrestrained accumulation by these cells.
Details of the oxidation process have been deduced mainly from the results of numerous in vitro studies. Several different reactive oxygen species including superoxide, hydrogen peroxide, hypochlorus acid, hydroxyl radicals, and peroxynitrite have been implicated in the initiation of lipid oxidation and peroxidation. Recently, it was shown that macrophage NAPDH oxidase, the principal vehicle for generating reactive oxygen species (ROS), does not contribute critically to the development of atherosclerosis (258), a somewhat surprising notion considering the likely exposure to many different ROS of the lipoproteins abundantly accumulating at atherosclerotic sites. The finding is evidence for the presence of alternative systems contributing critically to lipoprotein oxidation. Prime candidates are the lipoxygenase enzymes (described later) catalyzing lipid alkoxyl radical (LOO-) formation from esterified fatty acids. Such lipid peroxidation may bypass the generation of superoxide or any of its derived reactive products.
Although the mechanisms by which lipoproteins are oxidized in vivo are still debated, some of the products present in minimally modified/oxidized LDL (MM-LDL) have been identified, including 1-palmitoyl-2- (5,6)-epoxyisoprostane E2 (PEIPC), 1-palmitoyl-2-glutaroyl-sn-glycero-3-phosphorylcholine (PGPC), and 1-palmitoyl-2-oxovaleroyl-sn-glycero-3-phosphorylcholine (POVPC), principal inflammatory mediators of reactions promoting atherogenesis. These phospholipid oxidation products are prominent in atherosclerotic lesions. Moreover, they have been shown to mediate the activation of monocytes and/or endothelial cells in vitro in the presence of PAF (290, 505). The administration of WEB 2086, a PAF analog and selective PAF receptor antagonist, to C57BL/6J LDLR -/- mice fed a Western diet, reduced the fatty streak formation by 62% (505). Altogether, these studies favor the notion of important roles for PAF and/or PAF-like phospholipid oxidation products in mediating atherosclerotic lesion development.
C. Proinflammatory Reactions Associated With LDL Oxidation
Although granulocytes are generally more prolific than monocytes/macrophages in generating oxidizing products, i.e., oxygen radicals essential in the defense system, it appears that also monocytes/macrophages contribute significantly to LDL oxidation. Furthermore, T-helper lymphocytes, generators of IL-4 and IL-13, have been shown to enhance LDL oxidation mediated by activated monocytes in a 15-lipoxygenase (15-LO)-dependent fashion (155). Studies using 12-LO and 15-LO knock-out mice have affirmed the importance of the lipoxygenase pathway. Thus mice with a combination of ApoE -/- and 12- or 15-LO -/- traits were significantly less prone to lesion formation than were wild-type animals (107). When a specific inhibitor of 15-LO, PD 14167, was administered in a rabbit experimental model, atherosclerotic lesions had reduced monocyte/macrophage numbers, and fibro-foamy and fibrous plaque lesion development was diminished (53).
Once modified (e.g., oxidized or acylated) and taken up by macrophages, LDL activates the nascent foam cell. Modified LDL is chemotactic for monocytes and can up-regulate the expression of genes for macrophage colony stimulating factor (419, 422) and MCP-1 (293) (Fig. 1). Thus oxLDL may help expand the inflammatory response by stimulating the replication of monocyte-derived macrophages and the entry of new monocytes into lesions. Effects attributed to oxLDL pertaining to macrophage function include proinflammatory effects, such as increased proliferation (199, 330, 333, 450) and expression of inflammatory cytokines (222, 408), toxicity (300), increased expression of metalloproteinases (223), inhibited expression of inducible nitric oxide synthase (223), and effects on macrophage lipid metabolism and accumulation (179, 243, 500, 551). Released inflammatory mediators like IL-1, TNF-, and MCSF enhance the binding of LDL to the endothelium and smooth muscle and increase the transcription of the LDLR gene (196, 503), thereby having an impact on the migratory pattern of LDL within the artery. Upon binding of modified LDL to scavenger receptors in vitro, various intracellular events are initiated, including the induction of IL-1 (170, 391, 392) and urokinase (147). Thus, given the generation of sufficient amounts of modified lipids and a progressing inflammation, unless there is some tip of the balance, atherosclerosis development is sustained by the mutual reinforcement of these processes.
|
|
V. MONOCYTE DIFFERENTIATION AND THE ROLE OF THE EXTRACELLULAR MATRIX IN ATHEROGENESIS |
|---|
|
TopPreviousNextReferences |
|---|
When circulating peripheral monocytes migrate from the vascular to the extravascular compartment, a process entailing maturation of the cells to macrophages is concomitantly launched. This differentiation process renders the cells ready for active participation in the inflammatory and immune responses. The process has been shown to depend on many transcription factors, and a novel role for NFB has been suggested due to its accumulation in the cytoplasm of cells differentiated into macrophages (514). Part of the arsenal of the differentiated cells is an acquired acute responsiveness to, e.g., LPS, and enhanced capacity for TNF- secretion.
Comparatively little is known about the functional properties of macrophages in vivo, but on average their overall reactivity appears to be substantially higher than that of circulating monocytes. Thus, although oxidized LDL may fail to induce monocyte activation in whole blood (65), an altogether different situation prevails in the intima, where a whole series of activation products may be anticipated after interaction of macrophages with oxidized LDL.
Macrophages are key players in many aspects of human physiology and disease. A whole spectrum of macrophage subpopulations with distinct and often uncharted functions may be formed, as directed by the microenvironments of the differentiating monocytes. Current knowledge about monocyte differentiation has largely been derived from in vitro studies, many aspects of which may not truly reflect the activation reactions associated with differentiation in vivo. Nevertheless, the upregulation as well as downregulation of certain receptors is well documented. Thus alveolar macrophages showed much lower expression of 4-, 6-, and 2-integrins, CD11a, CD11b, L-selectin, Le (x), and sialyl Le (x) compared with monocytes (416). In one in vitro study it was shown that all adherent monocytes expressed CD14, CD36, and LDLR. In tissue macrophages these antigens were less consistently expressed and defined three cellular subpopulations: CD36+CD14-LDLR- (58 ± 12%), CD36+CD14+LDLR+ (18 ± 5%), and CD36-CD14-LDLR- (remaining cells) (567). Thus CD36 appeared to be present in three-fourths of the macrophages, a significant fraction considering the central role of CD36 as an oxLDL scavenger receptor. Another marker that is also induced during monocyte to macrophage differentiation is the class A macrophage scavenger receptor (SR-A). This protein mediated 80% of the uptake of acetylated LDL by human monocyte-derived macrophages (182). The upregulation of platelet-derived growth factor (PDGF) receptors on human monocyte-derived macrophages has been taken as evidence that PDGF has a role in atherogenesis, regulating the function of macrophages as well as SMC in the vascular wall (232).
A hallmark of the development of atherosclerotic plaques is the prior and concurrent acccumulation in the arterial intima of lipoprotein particles subject to chemical modifications. This is associated with local inflammation in the vessel wall and further recruitment of monocytes from the circulation. By taking up such modified LDL (oxidized or acetylated), monocyte-derived macrophages are turned into fat-loaded macrophages residing in the vessel wall and furthering the local inflammatory response. The mechanisms underlying such foam cell generation has for several years been the focus of intensive research (46, 72, 90, 499, 568) (Fig. 1).
Macrophages are normally protected from the accumulation of toxic cholesterol loads by multiple mechanisms, notably the downregulation of surface LDL receptor molecules in response to replete intracellular cholesterol stores (72). However, oxidized or otherwise chemically modified LDL may be taken up by alternate "scavenger" or "oxidized LDL" receptors that are not similarly downregulated when the cholesterol load is in excess (46, 90, 499, 568), thereby evading regular homeo-static control mechanisms. A series of scavenger receptors have been identified and are listed in Table 2. These comprise several classes of transmembrane receptors, a common characteristic of which is an affinity for negatively charged macromolecules or particles like, e.g., modified LDL.
An important role of the SR-A receptor in atherogenesis was inferred from the first analysis of mice with a combined SR-A and ApoE deficiency. These mice had a 60% reduction in atherosclerotic lesion development compared with wild type (509). However, more recent studies have been less affirmative of such a proatherogenic role for SR-A. Thus a combined SR-A and LDL-R deficiency caused only 20% reduction in atherosclerosis (448). Furthermore, when subjecting apoEJ-Leiden mice (another proatherogenic mouse model) to the added deficiency of SR-A, no decrease in atherosclerosis was observed (118). This controversy has been interpreted as an indication that ApoE has a complex role in pathogenesis, related to the fact that ApoE not only acts in plasma lipoprotein metabolism, but it also has a stimulatory influence on the efflux of cholesterol from macrophages (209, 280). One would anticipate the latter effect to serve as a protective mechanism counteracting foam cell formation.
Recently, more solid support for a central role of the class B scavenger receptor CD36 was provided by the demonstration that crosses of ApoE deficiency mice with CD36 deficiency mice had a 76% reduction in lesion area compared with mice deficient in ApoE only, after 12 wk on a Western diet (150). CD36 has been identified as the receptor facilitating myeloperoxidase-modified (MPO-modified) LDL uptake. MPO-modified LDL has a proven capacity to induce foam cell formation in vitro and is likely to be highly proatherogenic. CD36 deficiency in mice reduced the uptake of MPO-modified LDL by almost 90% and the formation of foam cells by >50% (412). It has to be borne in mind, however, that one cannot automatically infer from this marker role of CD36 in knock-out mouse models, that it has a corresponding role in human pathophysiology. Several other scavenger receptors have also been shown to be important for the uptake of oxidized LDL and the pathogenesis of atherosclerosis (see Table 2). Among these scavengers are macrosialin/CD68, lectinlike oxidized LDL receptor (LOX-1) and SREC (for review, see Ref. 120).
The results emerging from transgenic mouse models should be interpreted with caution, considering such factors as control of genetic background, linkage problems, etc. Furthermore, to better mimic the human situation, more refined heterozygotic models assessing various markers and polymorphisms, alone or in combination, may be called for (for a review, see Ref. 266).
The prominent role of modified LDL in atherogenesis suggests that addressing the undesirable proliferation or action of LDL-related products might have considerable prophylactic or even curative potential. Conceivably the following approaches might prove useful: saturation with relevant antioxidants to prevent LDL oxidation, directly at the level of the particle itself or indirectly at the level of the cellular oxidative machinery, or conversion of already oxidized LDL to a nonatherogenic particle using HDL-associated paroxonase (PON-1) (249).
C. Roles of Extracellular Matrix and Metalloproteinases in Atherogenesis
Monocytes migrating into the subendothelial space probably interact with ECM components, notably collagen type I, a major constituent of the normal arterial wall (495), and the most prominent matrix component of atherosclerotic plaques (496). Strong interaction between monocytes and matrix is proabably a prerequisite for the formation of lipid-laden macrophages, since lipid does not accumulate in monocytes that do not form stable interactions with tissues, even if they are allowed to differentiate into macrophages (558).
The ECM molecules are synthesized by resident cells of vascular tissue, such as endothelial cells, macrophages, and smooth muscle cells. Atherogenic lipoproteins gaining access to the subendothelial space are bound and retained through ionic interactions between positively charged residues on their apolipoproteins (apo B and apoE), and negatively charged sulfate and carboxylic acid residues on the glycosaminoglycan chains of extravascular or cell-associated vascular proteoglycans (for review, see Ref. 84). One of the tenets of the response to retention hypothesis is that prolonged residence time of lipoproteins in the intima leads to lipoprotein modification.
Recently, it was shown that oxidized LDL particles retained by ECM proteoglycans were taken up by macrophages, provided the ECM had been preincubated with lipoprotein lipase before adding ox-LDL (250).
Matrix metalloproteinases (MMP) are a family of enzymes comprising at least 16 zinc-dependent endopeptidases that are catalytically active against ECM components (for review, see Ref. 174). They are essential for cellular migration and tissue remodeling under both physiological and pathophysiological conditions (359). In vitro studies have demonstrated that the expression of MMP-1, MMP-3, and MMP-9 in macrophages and smooth muscle cells is enhanced by several mediators secreted from T lymphocytes and monocytes, e.g., TNF- and IL-1 (165, 282), whilst it is suppressed by the general inhibitors of monocyte activation IL-4, INF-, and IL-10. The activities of matrix-degrading MMP are essential for many of the processes involved in atherosclerotic plaque formation, including infiltration of inflammatory cells as outlined above, SMC migration, and proliferation as well as angiogenesis. Probably the most serious consequences of MMP activities are the unfavorable effects on plaque stability and resistance to rupture, which may lead to unstable angina, myocardial infarction, and stroke (166).
MMP need to be activated, and urokinase (produced by macrophages) and the plasminogen system are known to play central roles in the processes where MMP are important (102). MMP are in fact tightly regulated by molecules controlling their activation and by specific inhibitors known as the tissue inhibitors of metalloproteinases (TIMPs) (149).
|
VI. MONOCYTES/MACROPHAGES AND THE ROLE OF THEIR ACTIVATION PRODUCTS |
|---|
|
TopPreviousNextReferences |
|---|
Monocytes play a central role under several pathophysiological conditions, particularly when the progression of the disease stems from underlying inflammatory reactions, e.g., in diseases like rheumatoid arthritis, atherosclerosis, psoriasis, asthma, and inflammatory bowel disease. The proinflammatory potential of the monocytes can only be unlocked by their activation/differentiation and subsequent secretion of activation products, the main classes of which are summarized in Table 3. More than 100 different biologically active molecules are known to be secreted by monocytes/macrophages.
|
Monocytes are activated only according to what is dictated by their particular environment, notably by its content of agonists including primers. The various monocyte agonists associated with monocyte/macrophage activation during atherogenesis are listed in Table 4 according to their type or the group to which they belong. Most of these substances have so far been reported to have agonist properties in relation to monocytes, i.e., activation is induced/primed by their presence. However, the interpretation of these data has been complicated by the fact that certain poorly charted and hard to avoid low-grade monocyte activation phenomena may be inherent in the isolation and culturing processes per se. Prime suspected culprits are mechanical stress and trace contaminants going undetected or not even looked for in the system, the outcome being already preactivated and/or primed cells, quite commonly not acknowledged as such. Indeed, when monocytes adhere to plastic or to extracellular matrix proteins, signals are delivered that induce expression of immediate-early (IE) response genes (246). While mRNA accumulates in these adherent monocytes, the cytokine expression products including IL-1 and TNF- are not secreted unless there is a "second stimulus" such as bacterial LPS (206).
|
In contrast, in the native environment of whole blood ex vivo, LPS appears to be the sole monocyte activator. Cytokines, growth factors, adhesion molecules, etc., reported as stimulants or agonists in cell culture studies, may in the native environment of whole blood only up-regulate ongoing activation (383). Thus activation effects assigned to particular agonists cannot be seen in isolation from the overall experimental setting, and the relevance for the in vivo situation should be interpreted with caution.
Many of the products generated by activated monocytes/macrophages require similar transcription factors for their synthesis. Thus active nuclear factors NFB and AP-1 are mandatory for the production of cytokines, tissue factor, and growth factors, strongly suggesting that they are crucial for monocyte reactivity (Fig. 2) (122). Evidence to support this hypothesis has been obtained from in vivo studies using a mouse strain subjected to an atherogenic diet. The susceptibility to aortic atherosclerotic lesion formation turned out to be associated with accumulation of lipid peroxidation products, induction of inflammatory genes, and the activation of NFB-like transcription factors (301).
|
Further evidence that NFB-dependent gene transcription has a role in atherosclerosis was obtained by subjecting C57B16 mice to intravenous administration of mildly oxidized LDL. This led to induced expression in the liver of the same set of genes as when the mice were fed an atherogenic diet. An inbred strain of these mice susceptible to fatty streak formation was shown to have particularly profuse expression in liver of genes associated with inflammation, and this trait cosegregated with a propensity for activation of a NFB-like transcription factor and with the level of oxidized lipids in the same organ (302).
Most of the products listed in Table 4 are probably more effective in activating monocyte-derived macrophages in the intima than they are in activating monocytes. In relation to atherogenesis, it has been established that various cytokines, immunostimulatory agents, as well as growth factors are potent activators of macrophages (Table 4). Furthermore, platelet-derived activation products, lipids (e.g., ox-LDL), and certain eicosanoids have similar effects. Recently, it was also shown that monocyte activation is upregulated by fibrinogen, indicating a proinflammatory role of this protein (314). Since physiologically unperturbed macrophages are virtually impossible to obtain in isolation, it remains difficult to infer from such studies which of the agonists listed in Table 4 are playing the most important roles in atherogenesis.
B. Autocrine- and Paracrine-Mediated Activation Reactions in Monocytes
Autocrine activation (effects of a substance secreted by a cell on that cell itself) and paracrine activation (action of substances produced by cells and acting at short range on neighboring cells) are important regulatory processes of monocyte activation. Probably one of the most important autocrine activators is TNF-, produced in monocytes and known to upregulate the synthesis of cytokines (410) and tissue factor in monocytes (137). Furthermore, TNF- may induce the production of IL-1 in monocytes, which in turn may promote a series of inflammatory reactions (273). The autocrine effect of TNF- is mediated through the activation of NF
