OUP user menu

Reverse cholesterol transport and cholesterol efflux in atherosclerosis

R. Ohashi, H. Mu, X. Wang, Q. Yao, C. Chen
DOI: http://dx.doi.org/10.1093/qjmed/hci136 845-856 First published online: 28 October 2005


Reverse cholesterol transport (RCT) is a pathway by which accumulated cholesterol is transported from the vessel wall to the liver for excretion, thus preventing atherosclerosis. Major constituents of RCT include acceptors such as high-density lipoprotein (HDL) and apolipoprotein A-I (apoA-I), and enzymes such as lecithin:cholesterol acyltransferase (LCAT), phospholipid transfer protein (PLTP), hepatic lipase (HL) and cholesterol ester transfer protein (CETP). A critical part of RCT is cholesterol efflux, in which accumulated cholesterol is removed from macrophages in the subintima of the vessel wall by ATP-binding membrane cassette transporter A1 (ABCA1) or by other mechanisms, including passive diffusion, scavenger receptor B1 (SR-B1), caveolins and sterol 27-hydroxylase, and collected by HDL and apoA-I. Esterified cholesterol in the HDL is then delivered to the liver for excretion. In patients with mutated ABCA1 genes, RCT and cholesterol efflux are impaired and atherosclerosis is increased. In studies with transgenic mice, disruption of ABCA1 genes can induce atherosclerosis. Levels of HDL are inversely correlated with incidences of cardiovascular disease. Supplementation with HDL or apoA-I can reverse atherosclerosis by accelerating RCT and cholesterol efflux. On the other hand, pro-inflammatory factors such as interferon-gamma (IFN-γ), endotoxin, tumour necrosis factor-alpha (TNF-α) and interleukin-1 beta (IL-1β), can be atherogenic by impairing RCT and cholesterol efflux, according to in vitro studies. RCT and cholesterol efflux play a major role in anti-atherogenesis, and modification of these processes may provide new therapeutic approaches to cardiovascular disease. Further research on new modifying factors for RCT and cholesterol efflux is warranted.


Over the last few decades, our understanding of the basic mechanisms involved in atherosclerosis has progressed significantly. The important role of inflammation at all stages of the disease process is now recognized, including triggers, mediators and end-effectors. Many recent reviews have summarized these landmark events of atherogenesis.1–5 Plasma cholesterol levels are an important factor in atherosclerosis. Regulation of cholesterol levels is a complicated process, involving cholesterol uptake, biosynthesis, transport, metabolism, and secretion, and has been well described in several recent reviews.6–9 In this review, we focus mainly on cholesterol transport, and describe recent advances in the understanding of reverse cholesterol transport (RCT) and cholesterol efflux, processes which significantly affect atherogenesis.

RCT is a pathway that transports cholesterol from extrahepatic cells and tissues to the liver and intestine for excretion. By reducing accumulation of cholesterol in the wall of arteries, RCT may prevent development of atherosclerosis. Cholesterol efflux, part of the RCT process, is a major process by which macrophages within the vessel wall secrete cholesterol outside cells. Other important factors include high-density lipoprotein (HDL), a subfraction of human plasma lipoproteins with apolipoprotein A-I (apoA-I) as its principal apolipoprotein. HDL levels can determine the efficiency of RCT and cholesterol efflux.

First, we discuss how clinical studies have demonstrated the role of RCT and cholesterol efflux in cardiovascular disease. We then describe animal models and in vitro studies that shed light on the molecular mechanisms of RCT and cholesterol efflux in the vascular system. These animal and laboratory data may be useful for interpretation of clinical findings, and for future directions of clinical investigations.

Overview of RCT and cholesterol efflux

The sequence of events in RCT is described in Figure 1. ApoA-I is first produced mainly by the liver, and released into the plasma. Circulating apoA-I interacts with serum phospholipids and forms nascent discoidal HDL (ndHDL). Once ndHDL is generated, it triggers cholesterol efflux in the macrophages and fibroblasts in the subendothelial space. Externalized cholesterol is absorbed by ndHDL, and subsequently is esterified by lecithin:cholesterol acyltransferase (LCAT). HDL particles are enriched with cholesteryl ester and become larger, resulting in HDL3 and HDL2. Phospholipid transfer protein (PLTP) is involved in this process: for example, by fusing two HDL3 into one HDL2 molecule. If HDL molecules are enriched with triglyceride, they are processed by the enzyme hepatic lipase (HL) and become smaller and denser. HL can convert the phospholipid-rich HDL2 to HDL3. However, regulation of the balance of HL and PLTP is not clear. Cholesterol ester transfer protein (CETP) facilitates the equimolar exchange of cholesteryl esters from HDL for triglycerides in apoB100-containing lipoproteins. These cholesteryl esters are then delivered back to the liver via low-density-lipoprotein receptor (LDL-R), converted to bile salts, and eliminated through the gastrointestinal tract.

Figure 1.

Reverse cholesterol transport. The reverse cholesterol transport (RCT) pathway delivers free cholesterol from macrophages or other cells to the liver or intestine for excretion. Major constituents of RCT include acceptors such as high-density lipoprotein (HDL) and apolipoprotein A-I (apoA-I), and enzymes such as lecithin:cholesterol acyltransferase (LCAT), phospholipid transfer protein (PLTP), hepatic lipase (HL) and cholesterol ester transfer protein (CETP), which regulate cholesterol transport. Eventually, cholesterols in the HDL are delivered to the liver via scavenger receptor B1 (SR-B-1), converted to bile salts and eliminated through the gastrointestinal tract. Cholesteryl esters (CE) could also be delivered to the liver via the low-density-lipoprotein receptor (LDL-R). ndHDL, nascent discoidal high-density lipoprotein.

As acceptors such as apoA-I and HDL approach macrophages in subintimal space, intracellular cholesterol can be released outside the cells for excretion, a process termed cholesterol efflux of macrophages (Figure 2). In this pathway, ATP-binding membrane cassette transport protein A1 (ABCA1) plays a major role in translocating cholesterol into the extracellular space.10 In addition to ABCA1, four other factors are known to be involved in the pathway. Scavenger receptor B1 (SR-B1) can induce cholesterol efflux by enabling HDL to bind to cells and reorganize lipids within cholesterol-rich domains in the plasma membrane.11,,12 Caveolins are typically associated with caveolae, which are non-clatrin-coated plasma membrane microdomains rich in cholesterol and glycosphingolipids. Caveolins are small proteins (18–24 kDa) that have a hairpin loop conformation, with both the N and C termini exposed to the cytoplasm.13 These proteins have the capacity to bind cholesterol, and can transport cholesterol from the endoplasmic reticulum to the plasma membrane.14,,15 A report showed that over-expression of caveolins enhances cholesterol efflux in hepatic cells without affecting ABCA1 expression, indicating the presence of a caveolin-dependent pathway.15 Sterol 27-hydroxylase (CYP27A1) is also known as a contributor to cholesterol efflux.17 CHOP cells transfected with CYP27A1 showed increased cholesterol efflux. Since ABCA-1 expression was not altered, CYP27A1 could cause cholesterol efflux independent of other factors. In addition to these pathways, cholesterol efflux can also occur via passive diffusion, in which cholesterol is desorbed down to the concentration gradient onto acceptor molecules.18 Thus, RCT and cholesterol efflux constitute an efficient pathway by which excess cholesterol can be removed out of the body. Although extensive studies have recently been performed, RCT is a complicated process and its regulation mechanisms are largely unknown. Several key factors described above are involved in the RCT and cholesterol efflux, but the inter-relationship among these factors is not clear.

Figure 2.

Cholesterol efflux. Cholesterol efflux, a part of RCT, is a pathway transferring intracellular cholesterol from macrophages or other cells to extracellular acceptors such as apolipoprotein A-I (apoA-I) of high-density lipoprotein (HDL). It consists of five independent routes, including ATP-binding membrane cassette transporter A1 (ABCA1), scavenger receptor B1 (SR-B1), caveolin, Cyp27A1 and passive diffusion.

Clinical investigations on RCT and cholesterol efflux

A variety of evidence shows that RCT and cholesterol efflux play a major role in preventing atherosclerosis in humans. In fact, congenital impairment in genes involved in cholesterol efflux may augment atherogenesis in some patients. On the other hand, acceleration of RCT and cholesterol efflux by increasing HDL or apoA-I levels may result in amelioration of atherosclerosis, suggesting a potential therapeutic tool for human atherosclerosis.

HDL levels and atherosclerosis

HDL has various species, identified on the basis of their major apolipoprotein (apo) components (apoA-I or apoA-II), density (HDL2 and HDL3) and electrophoretic mobility (α and pre-β).19 Changes in HDL levels more closely reflect variations in the HDL2 subfraction rather than HDL3.20 Several studies have shown that low levels of HDL2 and HDL3 are associated with increased progression of atherosclerosis and risk of cardiovascular disease.21–25 Since HDL and ApoA-I are major receptors of cholesterol in the cholesterol efflux, increasing HDL levels may increase cholesterol efflux and RCT, contributing to reduced cardiovascular disease risks. Many attempts have been made to enhance HDL levels as anti-atherogenesis therapy.26,,27

Statins are the inhibitors of hydroxymethyl glutaryl coenzyme A (HMG CoA) reductase, the rate-limiting enzyme in cholesterol biosynthesis. Statins could decrease intracellular hepatic cholesterol and up-regulate LDL receptors,28,,29 which remove both LDL and triglyceride-rich lipoproteins (VLDL and IDL) from the circulation, thereby reducing plasma LDL and triglyceride levels. In addition, statins could also raise HDL levels,29,,30 which may be mediated through increase in the synthesis and secretion of apoA-I.31 Clinical studies showed that atorvastatin increased apoA-I levels, and also decreased the CETP-mediated transfer of cholesterol from HDL, as a result of a reduction in apoB-containing lipoproteins (LDL, IDL, VLDL).32 Several intervention studies in both primary and secondary prevention support the benefit of treating patients who have low HDL levels with a statin to decrease cardiovascular risk in primary and secondary prevention settings.33,,34

Fibric acid derivatives (fibrates) are agonists of peroxisome proliferator-activated receptor (PPAR)-α, a nuclear hormone receptor involved in energy and lipid metabolism. Fibrates lower triglyceride levels very effectively, and raise HDL levels moderately.35 Fibrates are more effective in raising HDL levels when triglyceride levels are elevated. Gemfibrozil therapy significantly reduced cardiovascular disease, with a modest increase in HDL levels.36 A recent VA-HIT study demonstrated the benefit of gemfibrozil therapy in patients with low HDL and LDL levels.37,,38

Nicotinic acid, or niacin, is the most effective agent for raising HDL levels, resulting in increases of up to 35%. Niacin also lowers triglyceride levels, reduces LDL levels, and modestly lowers lipoprotein (a) levels.35,39,,40 Several clinical trials have found niacin to be effective, alone or in combination with other drugs, in preventing coronary events, slowing atherosclerotic disease progression, and promoting lesion regression.41

Clinical studies demonstrated that oestrogen replacement therapy in post-menopausal women increased apoA-I concentrations by suppressing the activity of hepatic triglyceride lipase (HTGL) and SR-BI.42 More recent clinical studies showed that oral administration of oestrogen could elevate HDL and inflammation proteins such as C-reactive protein (CRP) and serum amyloid (SAA), which may incorporate into HDL particles.43,,44 SAA incorporation in HDL may inhibit its ability to deliver free cholesterol to the liver for clearance, thereby contributing to higher levels of HDL cholesterol, and also reduce its antioxidant properties.45 These observations may explain the limitations of the cardiovascular benefits of oestrogen replacement therapy, even though HDL levels are increased in these patients.

ApoA-I promotes cholesterol efflux and RCT, transferring cholesterol from peripheral cells to the liver for subsequent elimination. Eriksson et al. infused human proapoA-I (precursor of apo A-I) liposome complexes into patients with familial hypercholesterolemia, and measured the faecal excretion of bile acids and neutral sterols.46 The faecal excretion of cholesterol increased in all subjects, corresponding to the removal of excess cholesterol after infusion. This result implies that infusion of proapoA-I liposomes in humans promotes net cholesterol excretion from the body, providing a mechanism that may be useful in the treatment of atherosclerosis.

ApoA-IMilano is a variant of apoA-I identified in individuals in rural Italy who exhibit very low levels of HDL.47 ApoA-IMilano appears to be atheroprotective by promoting cellular cholesterol efflux.48 In a recent study, infusion of apoA-IMilano reduced the atheroma size in the patients with coronary disease by 4.2%, compared with controls.49 The above reports provide support for the concept that elevating HDL decreases cardiovascular diseases. HDL therapies using infusion of apoA-I or apoA-IMilano might have the potential to be therapeutic tools for the disease.

Gene activities or mutations in RCT and cholesterol efflux

ABCA1 is an integral membrane protein that utilizes ATP as a source of energy for transporting lipids and other metabolites across membranes, where they are removed from cells by apolipoproteins such as apoA-I.50 Tangier disease (TD) is an autosomal recessive disorder of lipid metabolism associated with cardiovascular disease.51 It is also characterized by the absence of plasma HDL, and deposition of cholesteryl esters in the reticulo-endothelial system, with splenomegaly and enlargement of tonsils and lymph nodes. In TD fibroblasts, HDL-mediated cholesterol efflux and intracellular lipid trafficking and turnover are abnormal, and mutations in ABCA1 are believed to be responsible.52,,53 Abnormalities in ABCA-1 can also induce low levels of HDL. Clee et al. investigated the phenotypes of individuals with TD and familial hypoalphalipoproteinaemia.54 They found that patients with mutations in the ABCA1 gene have low levels of HDL and high levels of triglycerides. These data provide direct evidence that impairment of cholesterol efflux and RCT is associated with reduced plasma HDL levels and subsequent cardiovascular disease.

CETP is physically associated with HDL particles and facilitates the transport of cholesteryl ester from HDL to apoB-containing lipoproteins. A decrease of CETP activity may increase HDL-C and decrease VLDL-C and LDL-C. Thus, CETP simultaneously affects the composition and concentration of apoA−/− and apoB-containing lipoproteins. Clinical studies demonstrate a low prevalence of coronary heart disease among subjects with CETP deficiency.55 Genetic polymorphisms causing CETP deficiency are particularly observed in Asian populations.56 Although CETP deficiency might prevent atherogenesis by increasing HDL-cholesterol levels,57 their effect has been controversial. Studies performed in Japan indicate that CETP deficiency might increase cardiovascular disease risks.58 In spite of the confusion about the role of CETP, the use of CETP inhibition in humans has attracted attention as a new strategy. The compound JTT-705, a direct inhibitor of CETP, raises HDL in a dose-dependent fashion.59 However, it remains uncertain whether the observed increase in HDL leads to a reduction in coronary artery disease. Davidson et al. reported on a CETP vaccine that induces auto-antibodies that specifically bind and inhibit endogenous CETP, with the intention of increasing HDL and reducing the development of atherosclerosis.60 They demonstrated that 53% of patients developed the auto-antibodies after two injections of the vaccine. Future research will determine the right dose of the vaccine and its effect on HDL levels.

LCAT is a key enzyme necessary for extracellular cholesterol metabolism.61–63 LCAT may facilitate the uptake of cholesterol from peripheral tissues into HDL particles by maintaining a concentration gradient for the efflux of free cholesterol,64 and may play a major role in RCT. Therefore, if LCAT is impaired, mature HDL generation would presumably be decreased, resulting in augmentation of atherosclerosis. LCAT deficiency syndromes have been reported in humans. Fish-eye disease (FED) is a disorder with a selective defect in HDL- or α-associated LCAT activity, presenting with pronounced corneal opacification and a marked reduction in HDL cholesterol.65–67 However, cardiovascular risk does not seem to be increased in FED, for unknown reasons.68 Various animal studies have attempted to clarify this discrepancy, but the precise mechanism still remains controversial.

Hepatic lipase (HL) plays a major role in lipoprotein metabolism as a lipolytic enzyme that hydrolyses triglycerides and phospholipids in chylomicron remnants, intermediate density lipoprotein (IDL), and HDL. HL can convert the phospholipid-rich HDL2 to HDL3. Patients with HL deficiency present with hypercholesterolaemia or hypertriglyceridaemia and accumulate β-VLDLs, chylomicron remnants, IDLs, triglyceride-rich LDLs, and HDLs.69–71

PLTP facilitates the in vivo transfer of phospholipids from triglyceride-rich lipoproteins to HDL during lipolysis by lipoprotein lipase. These phospholipids are present in surface fragments released from chylomicrons and VLDL during lipolysis, and are important precursors of plasma HDL. The interaction of PLTP with typical HDL results in the release of small, lipid-poor pre-β-HDL particles, and at the same time produces large α-HDL (HDL2-like particles) by a process that involves particle fusion.72,,73 Plasma PLTP activity was elevated in patients with insulin-resistant diabetes and obesity, and it was also correlated with high plasma triglycerides and low HDL-cholesterol.74–76 Elevated PLTP activities in these patients may result from elevated rates of VLDL turnover, and inhibition of VLDL synthesis. The regulation of HL and PLTP levels, and their impact on atherosclerosis, is unclear. Since functions of HL and PLTP may regulate both proatherogenic and antiatherogenic factors, the net effect of these protein-induced alterations in plasma lipoproteins on atherosclerosis is not easily predictable. Further investigations are warranted.

Life-style modification and RCT and cholesterol efflux

Various life-style factors affect the progression of atherogenesis, and recent reports suggest that modulation of RCT and cholesterol efflux can play a major role in that process. Beulens et al. suggest that moderate alcohol intake may reduce cardiovascular risk by increasing RCT and cholesterol efflux.77 They found that ABCA1-dependent cholesterol efflux in macrophages were increased in humans consuming alcohol (vs. water). Higher levels of HDL-C and apoA-I were also seen in the alcohol group. In another report, HDL-C levels were increased in athletes, suggesting that regular exercise might have a protective effect on atherogenesis by stimulating RCT and cholesterol efflux.78 By contrast, smoking can impair remodelling of HDL-C by inhibiting many steps in RCT and cholesterol efflux, increasing risks for cardiovascular disease.79

Animal models of RCT and cholesterol efflux

In support of the human studies, a number of experiments using transgenic animals indicate that disruption of one or more steps in RCT and cholesterol efflux can result in accelerated atherosclerosis, whereas over-expression of pivotal proteins in RCT and cholesterol efflux, such as apoA-I, PLTP, LCAT, and SR-B1, exerts atheroprotective effects.80 For instance, increased accumulation of cholesterol in peripheral tissues is observed in animals with apoA-I or ABCA1 deficiency, while over-expression of those genes may reduce atherogenesis.

ABCA1, which facilitates cellular cholesterol efflux, has generated considerable interest as a potential anti-atherogenic agent. In transgenic mice that over-express ABCA1, increased ABCA1 raised plasma HDL levels, increased cholesterol efflux from macrophages, and reduced diet-induced atherosclerosis in different mouse models.81 For example, transgenic mice strongly expressing ABCA1 showed an anti-atherogenic lipid profile, with elevated levels of HDL-C and apoA-I, and significantly less aortic atherosclerosis.82,,83 In a recent study, increased recruitment of ABCA1-deficient leukocytes (monocytes/macrophages) was seen in the arterial wall of LDL-receptor deficient (LDLr−/−) mice.84 This indicates that leukocyte ABCA1 plays a critical role in the protection against atherosclerosis, and that ABCA1 is a leukocyte factor that controls the recruitment of inflammatory cells.

In ABCA1−/− mouse models, lack of HDL and apoA-I is accompanied by accumulation of lipid-laden macrophages in the lungs.85 Aiello et al. examined whether the complete absence of ABCA1 or selected inactivation in macrophages was accompanied by an increase in atherosclerotic lesion progression in hypercholesterolaemic apolipoprotein-E-deficient (apoE−/−) mice and LDLR−/− mice.86 Interestingly, the absence of ABCA1 led to reduced plasma cholesterol levels in both groups, but the complete absence of ABCA1 did not affect the development of atherosclerotic lesions in either the LDLr−/− or the apoE−/− mice. In contrast, bone marrow transplantation studies demonstrated that the selective inactivation of ABCA1 in macrophages markedly increased atherosclerosis and foam cell accumulation in apoE−/−. It is possible that the complete absence of ABCA1 induced another compensating anti-atherogenic effect. ABCA1 deficiency in macrophages, however, demonstrates the anti-atherogenic properties of ABCA1, independent of plasma lipids and HDL levels. Groen et al.87 studied the effect of ABCA1 on hepatobiliary cholesterol transport in ABCA1 knockout mice. ABCA1 knockout significantly blocked cellular cholesterol efflux to apoA-I, thereby reducing HDL formation (almost completely lacking HDL) in these mice. These data indicate that ABCA1 is critical to cholesterol efflux to HDL. However, there were unexpected data indicating that hepatobiliary cholesterol transport was not changed in ABCA1 knockout mice with very low HDL levels, as compared to wild-type mice. It is possible that compensation may occur in these ABCA1 knockout mice, including increased hepatic cholesterol synthesis or increased VLDL/LDL.88 Thus, this study generates some uncertainties about the role of HDL in RCT in mice. Further studies are required to clarify the underlying mechanisms.

Although CETP is a critical factor in RCT, with the ability to transport cholesteryl ester from HDL, several species, including mice and rats, are naturally deficient in CETP. In transgenic mice with the human CETP gene, HDL levels were decreased, while VLDL and LDL cholesterol and apoB levels were increased.89,,90 In both apoE-knockout mice and LDL-receptor-knockout mice, introduction of the human CETP gene with an atherogenic diet resulted in the formation of spontaneous atherosclerosis.91 Introduction of a simian CETP gene into mice also resulted in advanced atherosclerotic lesion formation, which may be due to a redistribution of cholesterol from HDLs to the VLDL/LDL fraction.92 These studies support the hypothesis that CETP is pro-atherogenic. However, several other studies in mice do not support this hypothesis. Over-expression of human CETP genes in mice, with over-expression of human apoC-III, showed an anti-atherogenic effect.93 This effect was also observed in a hypertriglyceridaemic mouse model produced by streptozotocin-induced diabetes and lipoprotein lipase deficiency.94 The mechanisms underlying these ambiguities are not clear.

Intravenous infusion of apoA-I and transgenic expression of human apoA-I in animal models are both anti-atherogenic, and can stimulate atheromatous plaque resorption.95,,96 Recently, an edible form of apoA-I mimetic peptides composed of D-amino acids (D-4F) was developed, and administration of the peptide decreased atheromatous lesions by 79% in LDLR−/− mice fed a Western diet.97 The authors further demonstrated that D-4F can cause formation of pre-β HDL, which induced increased cholesterol efflux in macrophages in apoE−/− mice.98 The introduction of recombinant apoA-IMilano into apoE-deficient mice and rabbits reduced the lipid content of atheromatous plaques.99,,100 It is likely that the use of apoA-I or mimicking peptides has potential therapeutic applications in patients with atherosclerosis.

In addition to its functions in mediating the transfer of phospholipid between lipoproteins and catalysing HDL conversion reactions, PLTP appears to interact with ABCA1, and mediate the net transfer of lipid from cells to apoA-I and ndHDL particles.101 PLTP may therefore participate in RCT and cholesterol efflux by regenerating ndHDL during HDL conversion, and facilitating the lipidation of ndHDL in the subendothelial space. Despite these observations, when transgenic mice over-expressing human PLTP were fed a high-cholesterol diet, serum HDL decreased, and aortic atherosclerosis increased significantly.102

The roles of LCAT in RCT and cholesterol efflux still remain unclear, given that patients with LCAT deficiency do not have significantly increased cardiovascular disease.68 Various animal studies have attempted to elucidate the mechanisms, but have not provided clear answers. For example, transgenic rabbits over-expressing LCAT have shown increased HDL levels and reduced atherosclerosis,103 whereas mice with enhanced LCAT expression have increased atherosclerosis, despite high HDL levels.104 Oxidative stress is elevated in LCAT−/− mice with enhanced vascular ring superoxide production. Ng et al. cross-bred LCAT−/− mice with apoE−/− mice, and found that LCAT−/−×apoE−/− mice had decreased atherosclerosis compared to LCAT+/+×apoE−/− mice.105 It may be possible that increased oxidative stress was reversed in a hyperlipidaemic background, due to redistribution of paraoxonase, resulting in ameliorated atherogenesis. The above reports suggest that other factors such as modification of oxidative stress may be involved in effects of LCAT.

Rodents under oestrogen therapy have greater concentrations of apoA-I mRNA levels. It is believed that the increase in apoA-I expression by estrogens is modulated by the oestrogen receptor via an indirect effect of the hormone, and in HepG2 cells, estradiol potentiated the synergistic interactions between two transcription factors (hepatic nuclear factors 3a and 4, HNF-3a/HNF-4)106 that bind to the apoA-I promoter, and is believed to up-regulate apoA-I promoter activity.107

In vitro studies on modification of RCT and cholesterol efflux

There should be many factors that modify RCT and cholesterol efflux. For example, some pro-inflammatory factors can impair RCT and cholesterol efflux, whereas others may accelerate the pathway in cell culture models. Understanding and exploration of these factors may reveal underlying mechanisms of RCT and cholesterol efflux, and subsequently provide new therapeutic approaches.

Activated CD4+ T cells present in the atherosclerotic lesion can secrete interferon-gamma (IFN-γ).108 The incubation of foam cells with IFN-γ results in the reduction of HDL3-mediated cholesterol efflux.109 This decrease is not observed in other macrophage-activating factors, such as colony-stimulating factor. These findings suggest that IFN-γ contributes to the progression of an atherosclerotic lesion by altering the pathway of intracellular cholesterol trafficking in macrophage foam cells. Reiss et al. suggest that anti-inflammatory adenosine A2A receptor (A2AR) can minimize atherosclerotic changes.110 They showed that a selective A2AR agonist, CGS-21680, inhibits foam cell formation in human macrophages stimulated with immune-complex and IFN-γ. This implies a possible novel approach to developing agents that prevent atherosclerosis.

In addition to ABCA1, ATP-binding membrane cassette transporter G1 (ABCG1) is also known to mediate cholesterol removal from macrophages to HDL. When endotoxin or cytokines such as tumour necrosis factor-alpha (TNF-α) and interleukin-1 beta (IL-1β) were incubated with macrophages, the mRNA levels of ABCA1 and ABCG1 are decreased.111 The effect is rapid and sustained, and is associated with a reduction in ABCA1 protein levels. These reports support the idea that reduction in ABCA1 and ABCG1 in macrophages during the host response to infection and inflammation may aggravate atherosclerosis.

Elevated levels of triacylglycerol-rich lipoprotein (TGRL), including chylomicron remnants, VLDL, and IDL, are risk factors for coronary artery disease.112 Undegraded TGRL and their remnants are present within human and experimental atherosclerotic lesions.113 A recent study showed that TGRL can inhibit the efflux of cholesterol from macrophages to apoA-I.114 This inhibitory effect of TGRL is seen only in cholesterol-loaded foamy macrophages, and not in naive cells. During extended efflux periods, TGRL effectively blocks the ability of apoA-I to reduce macrophage cholesterol mass. The inhibitory mechanism is unclear, but it seems possible that pre-incubation with TGRL might affect the expression of one of the proteins (ABCA1, SR-B1 and apoE) involved in movement of cholesterol to the cell surface, and affect the intracellular pools of free cholesterol and stored cholesteryl ester available within the cells.

In contrast to the above reports, transforming growth factor-beta (TGF-β) is known to have anti-inflammatory properties, as evidenced by the profound systemic inflammatory response reported for TGF-β−/− mice.115 In atherosclerosis, TGF-β could contribute to plaque stability through the inhibition of metalloproteinase activity and increase in matrix deposition.116 TGF-β-treated macrophages exhibited a significant increase in cholesterol efflux mediated by apoA-I or HDL.117 The increase in apoA-I-mediated efflux is consistent with an increase at both the transcriptional and translational levels in the amount of ABCA1 expression in TGF-β-treated foam cells. TGF-β reverses the IFN-γ mediated inhibition of cholesterol efflux in macrophage-derived foam cells by mitigating the inhibitory effects of IFN-γ on ABCA1 expression. These findings support the notion that TGF-β may have atheroprotective properties by increasing efflux and reducing macrophage foam cell formation.


Numerous studies in humans, animals, and in vitro, are addressing the importance of RCT and cholesterol efflux in atherogenesis. It is very possible that augmentation of RCT and cholesterol efflux could be therapeutically useful. Potential major strategies include accelerating RCT and cholesterol efflux, which can be activated by increasing HDL and apoA-I levels, or by stimulating PLTP or CETP. Cholesterol efflux can be enhanced by facilitating pathways including ABCA1, SR-B1, caveolin and Cyp27A1. Blocking inflammatory factors should also be useful. However, uncertainties remain about the impact of RCT and cholesterol efflux on cardiovascular disease. Further exploration of modifiers of RCT and cholesterol efflux is warranted. Gaining insight into the whole picture of RCT and cholesterol efflux may enable us to develop more effective therapies for atherosclerosis in the future.


This work was supported by National Institutes of Health Grants R01HL61943, R01HL65916, R01HL60135 and R01HL72716 (C. Chen); and R01DE015543 (Q. Yao).


View Abstract