Thematic review series : The Immune System and Atherogenesis Cytokine regulation of macrophage functions in atherogenesis

This review will focus on the role of cytokines in the behavior of macrophages, a prominent cell type of atherosclerotic lesions. Once these macrophages have immigrated into the vessel wall, they propagate the development of atherosclerosis by modifying lipoproteins, accumulating intracellular lipids, remodeling the extracellular environment, and promoting local coagulation. The numerous cytokines that have been detected in atherosclerosis, combined with the expression of large numbers of cytokine receptors on macrophages, are consistent with this axis being an important contributor to lesion development. Given the vast literature on cytokine-macrophage interactions, this review will be selective, with an emphasis on the major cytokines that have been detected in atherosclerotic lesions and their effects on properties that are relevant to lesion formation and maturation. There will be an emphasis on the role of cytokines in regulating lipid metabolism by macrophages. We will provide an overview of the major findings in cell culture and then put these in the context of in vivo studies. —Daugherty, A., N. R. Webb, D. L. Rateri, and V. L. King. Cytokine regulation of macrophage functions in atherogenesis. J. Lipid Res. 2005. 46: 1812–1822. Supplementary key words lipoprotein • modification • metabolism • matrix • coagulation As noted in Getz’s overview (1), lesions contain large numbers of cytokines that can be derived from several cell types. These cytokines may affect the function of many cell types in atherogenesis. The effects of cytokines on endothelial and smooth muscle cells are discussed in Raines and Ferri’s contribution to this series (2). The purpose of this review is to focus on the effects of cytokines on macrophages in the evolution of atherosclerotic lesions. This is a vast literature that has necessitated some selectivity in the areas that can be covered. Given the subject area of this journal, we have elected to focus particular attention on the effect of cytokines on lipid metabolism in macrophages. MACROPHAGE FUNCTIONS IN ATHEROSCLEROTIC LESIONS Macrophages are hypothesized to be attracted to the subendothelial space to remove noxious materials deposited at atherosclerosis-prone regions of arteries. The precise chemical identity of the substance that attracts macrophages has not been unequivocally defined, although many candidate molecules are components of modified lipoproteins (3, 4). However, the function of infiltrating cells becomes subverted and leads to their retention within the subendothelial space. In this region, it is proposed that macrophages modify adjacent lipoproteins while also providing major mechanisms of removal for modified materials from the extracellular environment. The combination of lipoprotein modification and uptake leads to macrophages becoming engorged with lipids and resulting in a morphology that is given the descriptive name of “foam cells.” Lipid engorgement causes pronounced cellular hypertrophy, with the cell diameter being 10 times that of the originating monocyte. Probably as a result of the immense size increase, lipid-laden macrophages are chronically entrapped in the subendothelial space. Trapped macrophages can then invoke processes that perpetuate the continual recruitment of monocytes, leading to an expanded lesion volume. In addition, the subendothelial macrophages can influence the behavior of neighboring cell types within atherosclerotic lesions. Abbreviations: apoE, apolipoprotein E; GM-CSF, granulocyte macrophage colony-stimulating factor; IFN, interferon; IL, interleukin; LOX-1, lectin-like oxidized low density lipoprotein receptor-1; LRP, low density lipoprotein receptor-related protein; M-CSF, monocyte colonystimulating factor; MMP, matrix metalloproteinase; SR-A, class A scavenger receptor; SR-BI, scavenger receptor class B type I; TGF, transforming growth factor; TNF, tumor necrosis factor. 1 To whom correspondence should be addressed. e-mail: alan.daugherty@uky.edu Manuscript received 5 May 2005 and in revised form 13 June 2005. Published, JLR Papers in Press, July 1, 2005. DOI 10.1194/jlr.R500009-JLR200 thematic review by gest, on S etem er 7, 2017 w w w .j.org D ow nladed fom

As noted in Getz's overview (1), lesions contain large numbers of cytokines that can be derived from several cell types. These cytokines may affect the function of many cell types in atherogenesis. The effects of cytokines on endothelial and smooth muscle cells are discussed in Raines and Ferri's contribution to this series (2). The purpose of this review is to focus on the effects of cytokines on macrophages in the evolution of atherosclerotic lesions. This is a vast literature that has necessitated some selectivity in the areas that can be covered. Given the subject area of this journal, we have elected to focus particular attention on the effect of cytokines on lipid metabolism in macrophages.

MACROPHAGE FUNCTIONS IN ATHEROSCLEROTIC LESIONS
Macrophages are hypothesized to be attracted to the subendothelial space to remove noxious materials deposited at atherosclerosis-prone regions of arteries. The precise chemical identity of the substance that attracts macrophages has not been unequivocally defined, although many candidate molecules are components of modified lipoproteins (3,4). However, the function of infiltrating cells becomes subverted and leads to their retention within the subendothelial space. In this region, it is proposed that macrophages modify adjacent lipoproteins while also providing major mechanisms of removal for modified materials from the extracellular environment. The combination of lipoprotein modification and uptake leads to macrophages becoming engorged with lipids and resulting in a morphology that is given the descriptive name of "foam cells." Lipid engorgement causes pronounced cellular hypertrophy, with the cell diameter being Ͼ 10 times that of the originating monocyte. Probably as a result of the immense size increase, lipid-laden macrophages are chronically entrapped in the subendothelial space. Trapped macrophages can then invoke processes that perpetuate the continual recruitment of monocytes, leading to an expanded lesion volume. In addition, the subendothelial macrophages can influence the behavior of neighboring cell types within atherosclerotic lesions.
This includes the well-characterized interaction of macrophages and T-lymphocytes (5). During late stages of atherosclerosis development, exposure of macrophage-rich areas of lesions provides a nidus for thrombus attachment that is thought to account for a high proportion of the catastrophic consequences of atherosclerosis. At each stage of lesion development described above, cytokine interactions with macrophages have the potential to be major determinants of the mechanism and magnitude of the response.

EXPERIMENTAL MODES IN ELUCIDATING MACROPHAGE BIOLOGY IN ATHEROSCLEROSIS
The macrophage is probably the most phenotypically diverse cell type in the body (6). From the common origin of monocytes, macrophages take on different characteristics that are presumably determined by the local milieu of the tissue into which they migrate. Although dissimilar between tissues, macrophages within a specific tissue have less heterogeneity. However, this is not the case in atherosclerosis, in which there are considerable variances in the structures of macrophages that may reflect a divergent array of functional abilities.
Many contributions to our knowledge of cytokine effects on macrophage biology in atherosclerosis are derived from studies of cultured cells. The classic system in this area of investigation is primary cultures of macrophages derived from peritoneal lavage of mice (7). Many studies have also used macrophages derived from human monocytes. In addition to primary cells, there are also a wide variety of macrophage-like cell lines. Some of the most commonly used cell lines include the human-derived THP-1 (8) and the mouse lines J774 and P388D1 (9,10). Although these cultured cells have some macrophage characteristics, the extent to which they mimic the cells present in atherosclerosis remains an open question. The assessment of similarities and differences of macrophages in culture versus those within atherosclerotic lesions is hampered by the fluid nature of macrophage phenotypes and the dearth of reagents to specifically define these phenotypes. The combination of this extensive heterogeneity and the imprecision of phenotyping may contribute to the conflicting reports of the effects of cytokines on some properties of macrophage biology that are described in this review.

CYTOKINE REGULATION OF MACROPHAGE-MEDIATED LIPOPROTEIN MODIFICATION
An early event in the development of atherosclerotic lesions is the subendothelial deposition of lipoproteins (11,12). The presence of lipoproteins in the subendothelial space leads to substances that promote both chemotaxis and lipid accumulation in macrophages. Several modifications can promote both of these properties (3,13). In the early phases of lesion formation, lipoprotein modification occurs, presumably mostly mediated by endothelial mechanisms (14). However, once atherogenesis has been initiated, macrophages may become a major cell type responsible for lipoprotein modification within lesions.
There have been several pathways proposed for lipoprotein modification that are regulated by interactions of cytokines with macrophages. Much of the focus of the study of these modifications has been on oxidative mechanisms. Oxidation of LDL by cultured mouse peritoneal macrophages, as defined by the content of thiobarbituric acid-reacting substances, is decreased by incubating cells with interferon (IFN)-␥ (15). One potential mechanism of IFN-␥ -induced reduction of LDL oxidation is an acceleration of extracellular tryptophan degradation (16). Several reports have demonstrated a role of IFN-␥ via a lipoxygenase pathway that is also thought to involve oxidation (17). Lipoxygenase, specifically the 15-isoenzyme, is expressed in lesional macrophages and colocalizes with immunologically defined oxidative epitopes of lipoproteins (18,19). Expression of the enzyme in macrophages has been shown to promote atherosclerosis and lipoprotein oxidation (20)(21)(22), although there are contrary data (23,24). 15-Lipoxygenase is not expressed in cultured human monocytes, but its expression is profoundly stimulated by incubation with interleukin (IL)-4 or IL-13 (25,26). The stimulation of lipoxygenase expression in these cells is ablated by coincubation with IFN-␥ (25). However, incubation of IFN-␥ with mouse peritoneal macrophages, which contain abundant lipoxygenase activity on isolation, does not influence lipoxygenase protein abundance or activity (27). Therefore, the effect of IFN-␥ appears to be via a mechanism that inhibits the synthesis of the lipoxygenase. The contribution of these cytokines to lipoxygenase regulation in vivo is unclear, because mice deficient in IL-4 do not have reduced expression of lipoxygenase (27). This may be attributable to the continued presence of IL-13 (26). However, lipoxygenase expression is unexpectedly increased in total lymphocyte-deficient mice that are assumed to have low circulating concentrations of cytokines. Also, there is no effect of STAT-6 deficiency, which is a common pathway for the effects of both IL-4 and IL-13 (28).
Myeloperoxidase is present in large amounts in monocytes and has been proposed as a major oxidative enzyme in atherosclerosis (29). The role of myeloperoxidase in the development of atherosclerosis has not been defined, although the protein and many oxidative products of its activity have been detected in lesions (29)(30)(31)(32). The cytokine regulation of myeloperoxidase in macrophages has been given limited attention. Granulocyte macrophage colony-stimulating factor (GM-CSF), but not monocyte colony-stimulating factor (M-CSF), increases the expression of myeloperoxidase in macrophages (33).
Lipoproteins can also be modified within lesions by nonoxidative processes. These include the effects of LPL expressed in macrophages of atherosclerotic lesions (34). The absence of LPL in macrophages decreases atherosclerosis (35,36), whereas overexpression in this cell type increases lesion size (37). IFN-␥ decreases the expression of LPL in human monocyte-derived macrophages from early to late stages of culture (38,39) and by transcriptional regulation (40). Macrophage LPL expression is also inhibited by transforming growth factor (TGF)-␤ through a transcriptional effect (41,42), but it is upregulated by tumor necrosis factor (TNF)-␣ (43). Thus, there is substantial evidence that cytokine regulation may be an important contributor to the atherogenic effects of macrophage-expressed LPL. Although LPL can modify specific lipoprotein fractions, it is possible that nonlipolytic properties of the protein are responsible for its effects in atherogenesis (44).

CYTOKINE REGULATION OF LIPOPROTEIN ENTRY INTO MACROPHAGES
One of the most prominent changes in macrophages after entry into the subendothelial space of developing atherosclerotic lesions is the engorgement of these cells with lipid. Intracellular lipid stores are initially formed with simple droplet morphology. With progressive lipid engorgement, there is the formation of intracellular complexes of cholesterol and phospholipid and cholesterol crystals (45,46). These complexes and crystals are frequently encased by an acid phosphatase-positive layer, consistent with entrapment in lysosomes or late endosomes.
It is now recognized that many receptors are present on macrophages that bind a wide range of native and modified lipoproteins. Several major receptor types that recognize native lipoproteins may be regulated by cytokines in macrophages. LDL receptors have a clearly defined role in the cholesterol homeostasis of the whole body. Their role in macrophages has not been explored widely because of the assumption that they are downregulated in lipid-laden lesional macrophages. However, LDL receptor protein is detectable in experimental atherosclerotic lesions (47,48). Furthermore, macrophage LDL receptors influence atherogenesis under conditions of modest hyperlipidemia (48,49). The expression of LDL receptors on macrophages in cultured cells is a function of the origin and differentiation status (50). Macrophage LDL receptors have been demonstrated to be upregulated and downregulated by IFN-␥ and TGF-␤ , respectively (41,51). Expression of LDL receptors may also have implications on the mode of metabolism of the highly atherogenic lipoprotein fraction, ␤ -VLDL, which may have characteristics similar to those of postprandial chylomicron lipoproteins (48,49).
In addition to LDL receptors, ␤ -VLDL is also recognized by VLDL receptors. This receptor type is also expressed on macrophages and is downregulated by IFN-␥ (52). The functional significance of this downregulation can be shown by the inhibition of ␤ -VLDL-induced foam cell formation by IFN-␥ .
Another lipoprotein receptor type that recognizes native lipoproteins is the low density lipoprotein receptorrelated protein (LRP; also designated CD91) (53,54). Although the embryonic lethality of LRP-deficient mice has impeded a full characterization of its function, its proper-ties appear consistent with being the major system for the clearance of chylomicron remnants from the plasma (54). LRP on macrophages has consistently been shown to be downregulated in the presence of IFN-␥ (55-57). Conversely, TGF-␤ upregulates macrophage LRP, whereas M-CSF has no effect (56). These changes in macrophage LRP expression may have implications for lipoprotein accumulation within lesional macrophages. LRP, which is also termed the ␣ 2 macroglobulin receptor, has a wide range of ligands in addition to lipoproteins (54). Many of these ligands are responsible for regulating the extracellular proteolytic environment of macrophages (58).
Although there has been limited work on the cytokine regulation of native lipoprotein receptors, there has been considerable effort to study the cytokine regulation of receptors for modified lipoproteins. As discussed in the preceding section, there are several mechanisms of lipoprotein modification that can be regulated by cytokines. Many of these modifications involve some form of oxidative damage. The original receptor for modified lipoproteins was designated a "scavenger receptor" based on its ability to mediate the endocytosis of acetylated LDL (59).
There are now many proteins that have been designated as scavenger receptors that are broadly classified by gross structural characteristics in an alphabetic system (8).
The initially discovered scavenger receptor is now referred to as class A scavenger receptor (SR-A) (60). This receptor is able to transport acetylated and oxidized forms of LDL into macrophages by a process that is not downregulated by intracellular cholesterol content (61). Several studies have used genetically manipulated mice to define the effect of SR-A on atherosclerotic lesion formation (62). The ability of cytokines to regulate the SR-A receptor has been the subject of many publications, of which some selected examples are summarized in Table 1 . IFN-␥ is the most widely investigated cytokine. Earlier studies demonstrated a downregulation of SR-A by IFN-␥ in human monocyte-derived macrophages (63,64). IFN-␥ has also been reported to inhibit SR-A at the transcriptional level (65). Decreased foam cell formation occurs by the incubation of IFN-␥ with modified lipoprotein and macrophages (64). These effects, combined with studies on smooth muscle proliferation (66,67), suggest that IFN-␥ may reduce the atherogenic process. However, there are also reports of IFN-␥ having other effects on SR-A activity, protein, and transcription that are either neutral or inhibitory (27,68,69). Markedly different results have also been reported for GM-CSF, which has shown both upregulation and downregulation of SR-A activity by this cytokine (27,70). These differing responses may be a reflection of the heterogeneity of macrophages, which may lead to differences in the phenotype of the cultured cells. Consistent with this theme, IFN-␥ increases SR-A expression in the early differentiation phase of blood-borne monocytes or the human cell line THP-1, but it downregulates this receptor in mature macrophages (69). These conflicting findings may also be related to the mode of data representation, because both IFN-␥ and GM-CSF may markedly increase cellular protein content without influencing cell number. Under these conditions, the normalization of data to either cellular protein or cell number can dramatically influence the conclusion of the study.
Lipopolysaccharide downregulates SR-A both in cultured cells and in vivo (71,72), although induction has also been shown in both primary macrophage cells and mouse macrophage-like cell lines (73). Of the many cytokines released after the incubation of macrophages with lipopolysaccharide, the downregulation of SR-A has been attributed to the secretion of TNF-␣ (71). Consistent with this effect, SR-A is downregulated by incubation of cultured macrophages with TNF-␣ (74). Furthermore, deficiency of the TNF-␣ receptor, p55, has been associated with increased SR-A expression in atherosclerotic lesions (75). Some of the inconsistencies in SR-A expression during TNF-␣ incubation may be attributable to temporal considerations. Indeed, SR-A expression in macrophages has been shown to be increased before the subsequent inhibition (76).
In addition to the extensive literature on the effects of IFN-␥ , CSFs, and TNF-␣ on SR-A regulation, there are a smaller number of publications that have studied the ef-fects of other cytokines on this receptor. These are also summarized in Table 1.
Compared with SR-A, there is a relatively modest literature on the role of cytokines in regulating other scavenger receptors. CD36 is a receptor that has a number of diverse properties, including the binding of modified forms of lipoproteins, although its ability to facilitate endocytosis of these particles is not universally accepted (77). CD36 deficiency has been shown to decrease the extent of atherosclerosis, primarily through a macrophage-mediated effect (78,79). The expression of CD36 on macrophages is increased by IL-4 and M-CSF (80,81). The effects of IFN-␥ and TGF-␤ on CD36 expression in macrophages have been inconsistent (81)(82)(83)(84)(85).
The lectin-like oxidized low density lipoprotein receptor-1 (LOX-1) was originally identified in endothelium and subsequently shown to be expressed in macrophages (86,87). Much of the focus of the study of its regulation has been on endothelial cells. In macrophages, variable effects of TGF-␤ 1 on LOX-1 regulation have been reported, although TNF-␣ has consistently been shown to upregulate this receptor (88-90). GM-CSF, granulocyte macrophage colony-stimulating factor; IFN, interferon; IL, interleukin; LOX-1, lectinlike oxidized low density lipoprotein receptor-1; LRP low density lipoprotein receptor-related protein; M-CSF, monocyte colony-stimulating factor; SR-A, class A scavenger receptor; SR-BI, scavenger receptor class B type I; SR-PSOX, scavenger receptor that binds phosphatidylserine and oxidized lipoprotein; TGF, transforming growth factor; TNF, tumor necrosis factor. Scavenger receptor class B type I (SR-BI) was originally identified as a protein that binds oxidized LDL before the realization that it is a physiologically important HDL receptor (91,92). SR-BI is expressed in macrophages and is distinct from the other receptors discussed above in that it can transfer lipid bidirectionally (8). The presence of SR-BI in macrophages has important consequences on the development of atherosclerosis, although the effect is complex, as may be expected for a receptor with this duality of function (93,94). SR-BI expression is variable in macrophages present in atherosclerotic lesions (95). The mechanisms responsible for this variable SR-BI expression have not been determined. In the scant cytokine literature, it has been demonstrated that both IFN-␥ and TNF-␣ downregulate SR-BI (96).
Overall, there are many receptors that facilitate the entry of native and modified lipoproteins into macrophages by transfer of either the whole particle or selective components. Although there has been some consistency in the effects of selected cytokines in regulating some of these receptors on macrophages, there are also many inconsistencies in the literature. Resolution of these conflicting results needs to account for the many variables in these studies, such as macrophage phenotype and temporal effects.

CYTOKINE REGULATION OF INTRACELLULAR LIPID DISTRIBUTION AND REMOVAL
The previous section described processes that predominantly influence the mass of cholesterol transported into the intracellular compartment. The mode in which cholesterol is distributed within macrophages has dramatic effects on its toxicity and availability for efflux. With regard to toxicity, unesterified cholesterol appears to be benign when excess is stored in lysosomes, but it has profound effects on cell function and viability when present in high concentrations in the cell membrane (97,98). Efflux of cholesterol is dependent on several known variables, including the efficiency of cholesteryl ester hydrolysis and transport pathways involved in shuttling cholesterol to the cell surface and its subsequent removal.
In addition to its role in macrophage-mediated lipoprotein modification and uptake, IFN-␥ has also been extensively studied with regard to its effects on macrophage lipid metabolism ( Table 2 ). IFN-␥ is the only cytokine in which a change in intracellular cholesterol storage has been demonstrated, with an increased cholesteryl ester pool and decreased cholesterol efflux (82,99). This increase could be accomplished by increased activity of ACAT-1, which esterifies cholesterol to cholesteryl ester. The activity of this enzyme is increased in macrophages incubated with IFN-␥ (100). M-CSF and TGF-␤ 1 also increase ACAT-1 activity (101,102). The effect of these cytokines on intracellular cholesterol distribution has not been determined.
The removal of cholesterol from the cell requires the hydrolysis of cholesteryl ester, which is achieved by either acid or neutral forms of cholesteryl ester hydrolase. The cytokine regulation of these enzymes is not well studied, with only one report demonstrating the increased activity of both forms of cholesteryl ester hydrolase after incubation of macrophages with M-CSF (102). Another cholesterol removal process is via the activity of cholesterol 27-hydroxylase, which is known to be present in many extrahepatic cells, including macrophages (103). Cholesterol 27-hydroxylase converts cholesterol to more polar products that are readily removed from macrophages. This process is upregulated by IFN-␥ (104, 105).
The secretion of cholesterol from macrophages is influenced by the ability to secrete apolipoprotein E (apoE), which has been demonstrated to promote cholesterol efflux, particularly when endogenously produced (106)(107)(108). IFN-␥ decreases apoE secretion via a posttranslational mechanism (109). Several other cytokines have also been demonstrated to regulate apoE secretion, although these have not been linked to changes in cholesterol efflux (Table 2).
ABCA1 has a well-defined role in cholesterol efflux in many cell types, including macrophages (110). The cytokine regulation of ABCA1 in macrophages is consistent with its role in efflux. For example, decreased abundance after IFN-␥ incubation was associated with decreased HDL binding and cholesterol efflux. This effect was independent of changes in the expression of SR-BI (111,112). Conversely, TGF-␤1 increased ABCA1 expression and cholesterol efflux (41,105,113), although the binding of HDL was decreased by this cytokine (85). Cytokines have many effects on macrophages with regard to intracellular cholesterol metabolism and its consequences on sterol efflux. With few anomalies, the most extensively studied cytokine, IFN-␥, increases intracellular cholesterol storage when cells are not incubated with modified lipoproteins. Further studies to demonstrate whether these changes affect macrophage viability would be of interest (98).

CYTOKINE REGULATION OF THE EXTRACELLULAR MATRIX OF MACROPHAGES
The extracellular matrix of the artery contains many different proteins. The subendothelial region contains type IV collagen, laminin, and fibronectin. The integrity of extracellular matrix proteins in atherosclerotic lesions may be compromised by a large number of enzymes, most of which belong to the matrix metalloproteinease (MMP), cysteine protease, or serine protease families.
Many MMPs have been detected in atherosclerosis, although the dominant MMPs expressed by macrophages that have been implicated in lesion development are MMP-9 and MMP-12. Of these MMPs, only MMP-9 has been shown to alter the atherogenic process with its deficiency, reducing lesion size in several vascular areas (114). Moreover, this study demonstrated that bone marrowderived stem cells were a significant source of the MMP-9 that modulates atherogenesis.

CYTOKINE REGULATION OF COAGULATION IN ATHEROSCLEROSIS
The presence of tissue factor in atherosclerotic lesions is proposed to exert a major effect on the development of atherosclerosis complications at late stages of the disease (126). Tissue factor is expressed in all of the major cell types of atherosclerotic lesions, with a preponderance in macrophages (127,128). Tissue factor expression in cultured macrophages is downregulated by the Th2 cytokines IL-4, IL-10, and IL-13, whereas the Th1 cytokine IFN-␥ has the opposite effect (129,130). Some of the effects on tissue factor expression, particularly those of IFN-␥, are differentiation-specific (131,132). Overall, there is limited information regarding the role of cytokine-macrophage interactions on the effect of coagulation responses in atherosclerotic lesions.

EXTRAPOLATION OF CYTOKINE EFFECTS ON CULTURED CELLS TO THE ATHEROGENIC PROCESS IN VIVO
There have been numerous studies to determine the role of specific cytokines in the development of atherosclerosis. As described above, one cytokine that has been studied extensively in cell culture studies is IFN-␥, which is also one of the more extensively investigated cytokines in in vivo studies of atherogenesis. Studies with cultured cells have demonstrated many effects of IFN-␥ on the intracellular accumulation of lipids in macrophages. These findings lead to the notion that IFN-␥ would retard atherosclerosis, especially by minimizing intracellular lipid accumulation in macrophages. However, there are conflicting results in cultured cells. In contrast, the effects of IFN-␥ on the development of atherosclerosis in mouse models of the disease have been quite consistent, but they have contradicted the original concept of IFN-␥ being antiatherogenic. Thus, deletion of both IFN-␥ and its receptor decreased the size of atherosclerotic lesions in apoE-deficient mice (133,134). Deletion of IFN-␥ also decreased the size of atherosclerotic lesions in LDL receptor-deficient mice (135). Conversely, exogenous administration of IFN-␥ or its upstream regulators, IL-12 or IL-18, increased the extent of atherosclerosis (136)(137)(138). Currently, the only published report suggesting an antiatherogenic role for IFN-␥ comes from a study of irradiated LDL receptor-deficient mice that were repopulated with cells from IFN-␥-deficient mice. Under these conditions, the absence of IFN-␥ in bone marrow-derived stem cells increased the size of lesions in three different vascular regions (139).
Although gene manipulation and exogenous cytokine administration have provided valuable insight into the atherogenic process, these do not localize the effect to a specific action of the cytokine on macrophages. One mode of focusing on cytokine-macrophage interactions is to use mice that are deficient in specific cytokine receptors. Previous studies have used the technique of bone marrow transplantation as a mode of determining a leukocyte-specific role in atherosclerosis (140,141). This approach has been used in the chemokine field, for example, in studies using CCR2 and CXCR-2 (142,143). However, bone marrow transplantation studies with cytokine receptor-deficient mice have not been performed to date. Although such studies provide valuable insight, their interpretation should be tempered by the potential for bone marrow-derived stem cells to differentiate beyond myeloid and lymphoid lineages (144,145). Thus, the definition of an interaction of a macrophage with a specific cytokine as a functionally significant event in atherogenesis will be facilitated by the ability to specifically regulate cytokine receptors in this cell type.

CONCLUSIONS
The macrophage is a pivotal cell type throughout the initiation and maturation of atherosclerotic lesions. The combination of the many cytokines present in atherosclerotic lesions and the abundant cytokine receptors on macrophages is consistent with an important role of cytokinemacrophage interactions in lesion development. However, the abundance of both cytokines and receptors on macrophages also provides some challenges to unequivocally determining the relative importance and mechanism of a specific cytokine. The most obvious road ahead is the use of molecular engineering to enable the ablation and enhancement of cytokine responses in macrophages. With the increasing identification of specific cytokine receptors, coupled with evolving modes of altering gene expression in macrophages (146,147), it is likely that this will spawn many studies to elucidate the role of cytokine interactions with macrophages.
Work in the authors' laboratory is supported by the National Institutes of Health (Grants HL-62846, HL-70239, and HL-71098) and the American Heart Association.