High-Density Lipoproteins Suppress Chemokines and Chemokine Receptors In Vitro and In Vivo
Objective— To investigate whether high-density lipoproteins (HDLs) suppress chemokine (CCL2, CCL5, and CX3CL1) and chemokine receptor (CCR2 and CX3CR1) expression, a mechanism for the atheroprotective properties of HDLs.
Methods and Results— Apolipoprotein (apo) E−/− mice were fed a high-fat diet for 12 weeks. Before being euthanized, the mice received 5 consecutive daily injections of lipid-free apoA-I, 40 mg/kg, or saline (control). The injection of apoA-I reduced CCR2 and CX3CR1 expression in plaques compared with controls (P<0.05). ApoA-I–injected mice had lower plasma CCL2 and CCL5 levels. Hepatic CCL2, CCL5, and CX3CL1 levels were also reduced (P<0.05). In vitro studies found that reconstituted HDL (rHDL) reduced monocyte CCR2 and CX3CR1 expression and inhibited their migration toward CCL2 and CX3CL1 (P<0.05). Preincubation with rHDL reduced CCL2, CCL5, and CX3CL1 expression in monocytes and human coronary artery endothelial cells. The stimulation of CX3CR1 with peroxisome proliferator–activated receptor γ agonist CAY10410 was suppressed by preincubation with rHDL but did not affect the peroxisome proliferator–activated receptor γ antagonist (GW9664)–mediated increase in CCR2. In monocytes and human coronary artery endothelial cells, rHDL reduced the expression of the nuclear p65 subunit, IκB kinase activity, and the phosphorylation of IκBα (P<0.05).
Conclusion— Lipid-free apoA-I and rHDL reduce the expression of chemokines and chemokine receptors in vivo and in vitro via modulation of nuclear factor κB and peroxisome proliferator–activated receptor γ.
National Cholesterol Awareness Month Article
Fisher EA. National Cholesterol Month. Arterioscler Thromb Vasc Biol. 2009;29:1243.
High-density lipoproteins (HDLs) are atheroprotective.1 Human and animal intervention studies have found that infusion of reconstituted HDL (rHDL) or overexpression of apolipoprotein (apo) A-I reduces atherosclerotic plaque size2 and macrophage and lipid content.3–6 The mechanisms for this have been predominantly attributed to reverse cholesterol transport. However, HDLs also have potent anti-inflammatory properties.7 For example, HDLs reduce endothelial expression of adhesion molecules, such as vascular cell adhesion molecule-1 and intercellular adhesion molecule-1. This project seeks to investigate if HDLs also reduce the expression of chemokines and their receptors, which are key mediators of inflammatory processes and atherosclerotic lesion development.
In the early stages of atherosclerosis, chemokines and chemokine receptors are critical for the recruitment of circulating monocytes to the endothelium and assist in their migration into the artery wall.8 The monocytes differentiate into macrophages and express more chemokines, thereby exacerbating the disease process.8
The importance of the chemokine receptors (CCR2 and CX3CR1) and the chemokines (CCL2 [monocyte chemotactic protein-1], CCL5 [regulated upon activation, normal T-cell expressed, and secreted (RANTES)], and CX3CL1 [fractalkine]) in atherosclerosis has been demonstrated in human studies and animal knockout models. For example, CCR2−/− and CX3CR1−/− mice, crossed with apoE−/− mice, develop smaller atherosclerotic plaques9–11; and CX3CR1 has been identified in human lesions.12,13 The chemokines CCL2, CCL5, and CX3CL1 have also been identified in human lesions12,14,15; and mice deficient in CCL2 and CX3CL116,17 have reduced lesion size on an apoE−/− background. Because chemokines and chemokine receptors are critical in the development of atherosclerosis, modulation of their expression by HDL would assist in explaining how HDL reduces plaque size and macrophage content.
Accordingly, we sought to determine if HDL can regulate the expression of chemokine receptors (CCR2 and CX3CR1) and chemokines (CCL2, CCL5, and CX3CL1) in vivo and in vitro. We found find that lipid-free apoA-I significantly reduces CCR2 and CX3CR1 expression in high-fat–fed apoE−/− mice and that rHDL reduces CCR2 and CX3CR1 expression in vitro in cultured human primary monocytes. The reduction in CX3CR1, but not CCR2, expression was mediated via peroxisome proliferator–activated receptor (PPAR) γ. Lipid-free apoA-I also reduced CCL2, CCL5, and CX3CL1 levels in mouse plasma and/or tissues and rHDL reduced all 3 chemokines in vitro in human coronary artery endothelial cells (HCAECs) and monocytes by decreasing the activation of nuclear factor (NF) κB. These studies demonstrate a potential role for chemokines and chemokine receptors in the antiatherosclerotic effects of HDL.
Three-week-old apoE−/− mice were fed a high-fat diet for 12 weeks and then injected daily with lipid-free apoA-I or saline in the final 5 days before being euthanized. Aortic sinus sections were stained for CCR2, CX3CR1, and macrophages. CCR2 and CX3CR1 mRNA levels were determined by RT-PCR. In vitro, the effects of rHDL on CCR2 and CX3CR1 protein expression and mRNA levels were investigated in primary human monocytes, as was the expression of the chemokines CCL2, CCL5, and CX3CL1 in both monocytes and HCAECs. The activation of PPAR-γ and the inhibition of the NF-κB pathway were also investigated.
There were no differences in total or HDL cholesterol concentrations between treatment groups at euthanasia (supplemental Table I; all supplemental materials are available online at http://atvb.ahajournals.org). The plasma concentration of human apoA-I at euthanasia was 30.2±6.5 mg/dL (mean±SEM) in mice that received the intravenous apoA-I injections. This elevation in human apoA-I did not alter endogenous mouse apoA-I levels.
Five daily lipid-free apoA-I, 40 mg/kg, injections reduced CCR2 and CX3CR1 staining in aortic sinus plaques by 38% and 36%, respectively (P<0.05) (Figure 1A and B). CCR2 and CX3CR1 mRNA levels in the aortic arches were decreased by 37% and 42%, respectively (P<0.05) (Figure 1C and D). The aortic sinus plaque macrophage content was not significantly reduced (27%, P=0.06) (Figure 1E), and there were no changes in plaque size (Figure 1F). CCR2 and CX3CR1 staining predominantly colocalized with macrophages within the aortic sinus plaques (supplemental Figure I). Further analysis of the plaques showed no changes in either smooth muscle cell α-actin or lipid content between treatment groups (supplemental Table I).
By using flow cytometry to assess the effect of lipid-free apoA-I injections on circulating mouse monocytes, we found that CD11b+ cells comprised approximately 10% to 14% of the total gated mononuclear cell fraction (supplemental Figure IIA). Analysis of cell populations revealed that 68% to 80% of CD11b+ cells were CCR2+ and 42% to 51% of CD11b+ cells were CX3CR1+. There were also distinct cell populations that were CD11b+CCR2− and CD11b+CX3CR1−. Five daily injections of lipid-free apoA-I had no effect on the expression of either CCR2 or CX3CR1 on CD11b+ cells (supplemental Figure IIB). Of the mice injected with PBS, 72.70±1.17% of the circulating CD11b+ cells were CCR2+ compared with 71.80±1.56% in the apoA-I group. Similarly, 45.60±2.44% of the CD11b+ cells were CX3CR1+ in the PBS group compared with 42.00±2.12% for the apoA-I group. Data presented as mean±SEM.
CCL2, CCL5, and CX3CL1 concentrations were assessed in plasma, homogenized liver samples, and thoracic aortas (Table). Plasma levels of both CCL2 and CCL5 were lower in mice injected with lipid-free apoA-I (21% and 25%, respectively) compared with control mice (P<0.05). There was no difference in CX3CL1 levels between treatment groups.
Liver CCL2 levels were reduced by 38% after apoA-I injection (P<0.05). There were also significant reductions in liver CCL5 (40%) and CX3CL1 (42%) levels in the apoA-I–injected mice compared with control mice (P<0.05 for both) (Table). These reductions in liver chemokine protein levels were supported by reductions in mRNA levels. Mice receiving the lipid-free apoA-I injections had significantly lower CCL2 (37%), CCL5 (22%), and CX3CL1 (24%) liver mRNA levels (P<0.05) (supplemental Table II).
Finally, there was a 40% reduction in CX3CL1 in the aortas of lipid-free apoA-I–treated mice compared with controls (P<0.05) but no changes in aortic CCL2 or CCL5 levels (Table).
Effect of Discoidal rHDL on CCR2 and CX3CR1 Expression in Monocytes
Incubation of monocytes with discoidal rHDL significantly decreased CCR2 (66%) and CX3CR1 (58%) protein levels compared with control cells (P<0.01) (Figure 2A). There were also significant reductions in CCR2 (72%, P<0.05) and CX3CR1 (50%, P<0.05) mRNA levels after incubation with rHDL (Figure 2B). CCL2 and CX3CL1 trigger CCR2- and CX3CR1-directed migration, respectively. By using transwell membranes to assess cell migration (Figure 2C), we found that incubation with rHDL decreased monocyte migration toward both CCL2 (62%, P<0.05) and CX3CL1 (36%, P<0.05).
Effect of Discoidal rHDL on CCL2, CCL5, and CX3CL1 Expression in HCAECs
HCAEC lysates were analyzed for changes in CCL2, CCL5, and CX3CL1 protein levels (Figure 3A). Incubation with tumor necrosis factor (TNF) α increased CCL2 by 3.3-fold (P<0.001). When HCAECs were preincubated with rHDL, CCL2 protein levels in cell lysates decreased by 20% (P<0.05).
Incubation with TNF-α increased the CCL5 protein concentration in HCAEC lysates 20-fold (P<0.001). This cytokine-induced increase in CCL5 was reduced by 41% (P<0.05) in HCAECs preincubated with rHDL. TNF-α increased CX3CL1 protein expression from undetectable levels to 4.8±0.2 ng/mg (mean±SEM) of cell protein (P<0.001). Preincubation with rHDL resulted in a significant 50% reduction in CX3CL1 concentration in HCAEC lysates (P<0.01).
Preincubation with rHDL also reduced secretion of chemokines into the culture media (supplemental Figure IIIA). Incubation with TNF-α significantly increased the medium CCL2, CCL5, and CX3CL1 levels (P<0.001 for all). Preincubation with rHDL decreased the concentration of CCL2, CCL5, and CX3CL1 by 24% (P<0.05), 32% (P<0.05), and 57% (P<0.01), respectively, compared with TNF-α incubation alone.
Consistent with the changes in chemokine proteins, preincubation with rHDL at a final apoA-I concentration of 600 μg/mL reduced mRNA levels of CCL2 (39%, P<0.05), CCL5 (66%, P<0.01), and CX3CL1 (22%, P<0.05) relative to cells that were incubated with TNF-α alone (supplemental Figure IIIB).
Effect of Discoidal rHDL on CCL2, CCL5, and CX3CL1 Expression in Monocytes
Monocyte lysates were analyzed for changes in CCL2, CCL5, and CX3CL1 protein expression (Figure 3B). Incubation with interferon (IFN) γ increased the concentration of CCL2 in monocyte lysates by 22-fold (P<0.001). Preincubation with rHDL decreased lysate CCL2 protein in a concentration-dependant manner (Figure 3B, inset). At a final apoA-I concentration of 600 μg/mL, the CCL2 concentration was reduced by 87% (P<0.001). Incubation with IFN-γ also significantly increased CCL5 and CX3CL1 protein levels in monocyte lysates (Figure 3B). Concentrations of both chemokines were significantly reduced back to baseline levels in monocytes preincubated with rHDL (P<0.05).
Incubation with IFN-γ also increased the concentration of CCL2 protein in the culture media by 57-fold (P<0.001) (supplemental Figure IVA). Preincubation with rHDL mediated a stepwise decrease in CCL2 levels in the culture media that was maximal (98%) after preincubation with the highest rHDL concentration (inset) (P<0.001). Stimulation with IFN-γ also increased the CCL5 concentration in the culture media. This increase was reduced by 37% when the cells were preincubated with rHDL (P<0.05). CX3CL1 protein was not detected in monocyte culture media before or after incubation with IFN-γ.
Consistent with the changes in chemokine protein levels, IFN-γ increased monocyte CCL2, CCL5, and CX3CL1 mRNA levels by 1.5- to 2.5-fold (P<0.01) (supplemental Figure IVB). When cells were preincubated with rHDL, CCL2 mRNA levels decreased in a concentration-dependant manner (inset) (P<0.001). Similarly, preincubation with rHDL, before stimulation with IFN-γ, decreased both CCL5 and CX3CL1 mRNA levels (P<0.05).
Discoidal rHDL Reduces Activation of the NF-κB Pathway
The expression of the chemokines CCL2, CCL5, and CX3CL1 is regulated by changes in the activation of transcription factor NF-κB.18–20 Incubation of monocytes and HCAECs for 24 hours with rHDL significantly reduced IκB kinase activity (as measured by ATP depletion) by 25% in monocytes (P<0.05) and by 22% in HCAECs (P<0.05) compared with control cells (Figure 4A). Preincubation with rHDL also reduced IκBαP relative to IκBαT (IκBαP/IκBαT) in both monocytes (50%) and HCAECs (55%) (Figure 4B) compared with controls (P<0.05 for both). Reductions in the nuclear NF-κB p65 subunit were also observed in monocytes (40%) and HCAECs (66%) incubated with rHDL (P<0.05) (Figure 4C).
Hepatic NF-κB p65 and p50 subunit mRNA levels in apoE−/− mice receiving lipid-free apoA-I injections were reduced by 36% for p65 (P<0.05) and by 46% for p50 (P<0.05) (supplemental Table III).
Regulation of CX3CR1 and CCR2 Expression by rHDL via PPAR-γ
The expression of CX3CR1 and CCR2 is regulated by PPAR-γ.21 To determine if the effects of rHDL on CCR2 and CX3CR1 levels were mediated via PPAR-γ, monocytes were incubated with the PPAR-γ agonist CAY10410. Under these conditions, CX3CR1 protein expression increased by 68% (P<0.05) (Figure 5A). When the cells were preincubated with rHDL, before CAY10410, CX3CR1 protein expression was comparable to baseline levels (P<0.01, compared with CAY10410-only control). Conversely, when monocytes were incubated with the PPAR-γ antagonist, GW9664, CCR2 protein levels increased by 72% (P<0.05) (Figure 5B). Preincubation with rHDL had no effect on the GW9664-induced increase in CCR2 expression.
It is well established that HDL exhibits potent anti-inflammatory and antiatherosclerotic properties.7,22 The objective of this study was to determine if HDL modulates the expression of chemokines (CCL2, CCL5, and CX3CL1) and chemokine receptors (CCR2 and CX3CR1), all of which have critical roles in mediating inflammation, monocyte recruitment, and progression of atherosclerosis. We report the following findings: (1) 5 daily injections of lipid-free apoA-I reduce CCR2 and CX3CR1 expression in the aortic sinus plaques of apoE−/− mice; (2) apoA-I–injected mice have significantly lower plasma CCL2 and CCL5 levels and reduced hepatic CCL2, CCL5, and CX3CL1 relative to control mice; (3) in vitro, incubation of human monocytes with rHDL significantly reduces CCR2 and CX3CR1 expression; (4) preincubation of both HCAECs and monocytes with rHDL significantly reduces CCL2, CCL5, and CX3CL1; (5) rHDL regulates CX3CR1 but not CCR2 expression via PPAR-γ; and (6) rHDL decreases chemokine expression by inhibiting activation of the NF-κB pathway.
One interesting outcome of this study is that the reduction in plaque CCR2 and CX3CR1 expression did not coincide with a reduction in plaque macrophage content. Furthermore, immunocytochemistry revealed that CCR2 and CX3CR1 colocalized with macrophages, suggesting that the reductions in plaque chemokine receptor expression are predominantly by macrophages and that other cell types are likely to play a minor role. These findings indicate that apoA-I may act directly on plaque macrophages to lower chemokine receptor expression in vivo. Consistent with these data, our in vitro findings also demonstrated that rHDL reduced monocyte CCR2 and CX3CR1 expression. However, flow cytometry analysis revealed that the injections of apoA-I did not alter chemokine receptor expression on circulating monocytes. This lack of change may be attributed to the short half-life (approximately 6 hours) of circulating monocytes,23 such that their exposure time to apoA-I is insufficient to decrease chemokine receptor expression. However, it does indicate that the observed reductions in plaque CCR2 and CX3CR1 expression are occurring at the site of the lesion. A reduction in plaque chemokine receptors may be beneficial eg, CX3CR1 anchors macrophages to smooth muscle cells by binding to CX3CL1.24 This interaction between macrophages and smooth muscle cells by CX3CR1/CX3CL1 causes macrophage retention within a plaque and prevents them from emigrating. Our results also indicate that the mechanism for these reductions in chemokine receptor expression may be, at least in part, because of modulation of PPAR-γ expression (Figure 5).
The present study found that lipid-free apoA-I infusions reduce circulating plasma levels of CCL2 and CCL5. However, there was no change in circulating CX3CL1. This is not surprising because CX3CL1 is predominantly a membrane-bound chemokine.8 Consistent with this fact, we observed reductions in CX3CL1 levels in the liver and thoracic aorta.
CCL2 and CCL5 levels were also lower in the liver homogenates of apoE−/− mice. This may explain the reduction in plasma CCL2 and CCL5 because fewer chemokines would have been secreted from the liver into the circulation. The combined plasma and liver data highlight the fact that lipid-free apoA-I can act as a systemic anti-inflammatory agent in addition to its more localized and well-known antiatherosclerotic effects. It may be expected that a reduction in plasma and hepatic chemokines would be atheroprotective because fewer chemokines would be present to participate in monocyte recruitment. However, we did not find a reduction in plaque macrophage content in this particular study. Perhaps with a longer apoA-I intervention period, a reduction in plaque macrophage content would have been observed, as reported elsewhere.5,25 However, our study was designed specifically to investigate changes in chemokines/chemokine receptors. Interestingly, we did not see changes in CCL2 or CCL5 in thoracic aorta homogenates. This is possibly because of the lower levels of expression found at this site. In contrast, our in vitro studies found that rHDL substantially lowered CCL2 and CCL5 in both monocytes and HCAECs. Possibly, therefore, a more sensitive measure of chemokine protein or higher expression levels in thoracic aortas could have detected changes in CCL2 and CCL5.
The decision to use apoA-I–containing discoidal rHDL in the in vitro studies and lipid-free apoA-I in the animal studies was made based on previous observations from our laboratory, which showed that lipid-free apoA-I is rapidly lipidated and then incorporated into the HDL fraction after injection into rabbits.26 This is also likely to be the case when lipid-free apoA-I is injected into mice. In contrast, because lipid-free apoA-I does not become lipidated to a significant extent when it is incubated with monocytes or HCAECs, it was converted into discoidal rHDL before incubation with the cells.
The expressions of CCR2 and CX3CR1 are regulated by PPAR-γ.21,27,28 The PPAR-γ agonist CAY10410 increases CX3CR1 expression in monocytes,21 which was confirmed in this study. However, when monocytes were preincubated with rHDL CX3CR1, protein expression did not increase. Conversely, preincubation of monocytes with rHDL had no effect on the stimulation of CCR2 expression by the PPAR-γ antagonist GW9664. This indicates that PPAR-γ mediates the decrease in CX3CR1 but not CCR2 and suggests that rHDL is modulating the expression of CCR2 via another mechanism. Interestingly, there is some evidence that CCL2 can regulate the expression of CCR2.29 Because rHDL substantially decreases CCL2 expression in monocytes, this may explain the decrease in CCR2 expression.
NF-κB is a heterodimer, of which p65 is reported to be the active subunit that initiates transcription.30 CCL2, CCL5, and CX3CL1 expression is regulated via activation of the NF-κB pathway,18–20 which is also inhibited in HCAECs by HDL.31 Results from the current study indicated that rHDL also decreases several key steps in the NF-κB activation pathway in monocytes. Consistent with this finding, our in vivo studies also found that lipid-free apoA-I–infused mice had lower levels of p65 and p50 NF-κB subunits. Because CCL2, CCL5, and CX3CL1 are all NF-κB target genes, this suggests that the mechanism by which rHDL reduces chemokine expression is via a reduction in activation of the NF-κB pathway.
In conclusion, we have demonstrated that lipid-free apoA-I and rHDL effectively reduce the chemokine receptors (CCR2 and CX3CR1) and the chemokines (CCL2, CCL5, and CX3CL1) in vivo and in vitro. These effects appear to be mediated via modulation of NF-κB and PPAR-γ. Our findings suggest an important role for chemokines and chemokine receptors in the antiatherosclerotic and anti-inflammatory effects of HDL, thereby providing further insight into its mechanism of action.
Sources of Funding
This study was supported by a Heart Foundation Career Development Fellowship (CR07S3331); a grant from the Bushell Foundation (Dr Bursill); and grant 222722 from the National Health and Medical Research Council Programme.
Received on: September 18, 2009; final version accepted on: June 17, 2010.
Nissen SE, Tsunoda T, Tuzcu EM, Schoenhagen P, Cooper CJ, Yasin M, Eaton GM, Lauer MA, Sheldon WS, Grines CL, Halpern S, Crowe T, Blankenship JC, Kerensky R. Effect of recombinant ApoA-I Milano on coronary atherosclerosis in patients with acute coronary syndromes: a randomized controlled trial. JAMA. 2003; 290: 2292–2300.
Tardif JC, Gregoire J, L'Allier PL, Ibrahim R, Lesperance J, Heinonen TM, Kouz S, Berry C, Basser R, Lavoie MA, Guertin MC, Rodes-Cabau J. Effects of reconstituted high-density lipoprotein infusions on coronary atherosclerosis: a randomized controlled trial. JAMA. 2007; 297: 1675–1682.
Shaw JA, Bobik A, Murphy A, Kanellakis P, Blombery P, Mukhamedova N, Woollard K, Lyon S, Sviridov D, Dart AM. Infusion of reconstituted high-density lipoprotein leads to acute changes in human atherosclerotic plaque. Circ Res. 2008; 103: 1084–1091.
Choudhury RP, Rong JX, Trogan E, Elmalem VI, Dansky HM, Breslow JL, Witztum JL, Fallon JT, Fisher EA. High-density lipoproteins retard the progression of atherosclerosis and favorably remodel lesions without suppressing indices of inflammation or oxidation. Arterioscler Thromb Vasc Biol. 2004; 24: 1904–1909.
Nicholls SJ, Cutri B, Worthley SG, Kee P, Rye KA, Bao S, Barter PJ. Impact of short-term administration of high-density lipoproteins and atorvastatin on atherosclerosis in rabbits. Arterioscler Thromb Vasc Biol. 2005; 25: 2416–2421.
Barter PJ, Nicholls SJ, Rye KA, Anantharamaiah GM, Navab M, Fogelman AM. Antiinflammatory properties of HDL. Circ Res. 2004; 95: 764–772.
Combadiere C, Potteaux S, Gao JL, Esposito B, Casanova S, Lee EJ, Debre P, Tedgui A, Murphy PM, Mallat Z. Decreased atherosclerotic lesion formation in CX3CR1/apolipoprotein E double knockout mice. Circulation. 2003; 107: 1009–1016.
Lucas AD, Bursill C, Guzik TJ, Sadowski J, Channon KM, Greaves DR. Smooth muscle cells in human atherosclerotic plaques express the fractalkine receptor CX3CR1 and undergo chemotaxis to the CX3C chemokine fractalkine (CX3CL1). Circulation. 2003; 108: 2498–2504.
Greaves DR, Hakkinen T, Lucas AD, Liddiard K, Jones E, Quinn CM, Senaratne J, Green FR, Tyson K, Boyle J, Shanahan C, Weissberg PL, Gordon S, Yla-Hertualla S. Linked chromosome 16q13 chemokines, macrophage-derived chemokine, fractalkine, and thymus- and activation-regulated chemokine, are expressed in human atherosclerotic lesions. Arterioscler Thromb Vasc Biol. 2001; 21: 923–929.
Saederup N, Chan L, Lira SA, Charo IF. Fractalkine deficiency markedly reduces macrophage accumulation and atherosclerotic lesion formation in CCR2-/- mice: evidence for independent chemokine functions in atherogenesis. Circulation. 2008; 117: 1642–1648.
Bhavsar PK, Sukkar MB, Khorasani N, Lee KY, Chung KF. Glucocorticoid suppression of CX3CL1 (fractalkine) by reduced gene promoter recruitment of NF-kappaB. FASEB J. 2008; 22: 1807–1816.
Barlic J, Zhang Y, Foley JF, Murphy PM. Oxidized lipid-driven chemokine receptor switch, CCR2 to CX3CR1, mediates adhesion of human macrophages to coronary artery smooth muscle cells through a peroxisome proliferator-activated receptor gamma-dependent pathway. Circulation. 2006; 114: 807–819.
Barter PJ. Cardioprotective effects of high-density lipoproteins: the evidence strengthens. Arterioscler Thromb Vasc Biol. 2005; 25: 1305–1306.
Barlic J, Zhang Y, Murphy PM. Atherogenic lipids induce adhesion of human coronary artery smooth muscle cells to macrophages by up-regulating chemokine CX3CL1 on smooth muscle cells in a TNFalpha-NFkappaB-dependent manner. J Biol Chem. 2007; 282: 19167–19176.
Rong JX, Li J, Reis ED, Choudhury RP, Dansky HM, Elmalem VI, Fallon JT, Breslow JL, Fisher EA. Elevating high-density lipoprotein cholesterol in apolipoprotein E-deficient mice remodels advanced atherosclerotic lesions by decreasing macrophage and increasing smooth muscle cell content. Circulation. 2001; 104: 2447–2452.
Kee P, Rye KA, Taylor JL, Barrett PH, Barter PJ. Metabolism of apoA-I as lipid-free protein or as component of discoidal and spherical reconstituted HDLs: studies in wild-type and hepatic lipase transgenic rabbits. Arterioscler Thromb Vasc Biol. 2002; 22: 1912–1917.
Barbier O, Torra IP, Duguay Y, Blanquart C, Fruchart JC, Glineur C, Staels B. Pleiotropic actions of peroxisome proliferator-activated receptors in lipid metabolism and atherosclerosis. Arterioscler Thromb Vasc Biol. 2002; 22: 717–726.
McGrath KC, Li XH, Puranik R, Liong EC, Tan JT, Dy VM, DiBartolo BA, Barter PJ, Rye KA, Heather AK. Role of 3beta-hydroxysteroid-delta 24 reductase in mediating antiinflammatory effects of high-density lipoproteins in endothelial cells. Arterioscler Thromb Vasc Biol. 2009; 29: 877–882.