OUP user menu

The role of HDL-cholesterol in preventing atherosclerotic disease

Philip Barter
DOI: http://dx.doi.org/10.1093/eurheartj/sui036 F4-F8 First published online: 30 June 2005


The cholesterol required by peripheral tissues, including vascular cells, is provided both by new synthesis in the cells and by a delivery from low-density lipoproteins (LDLs). When the level of LDLs is high, they accumulate in the artery wall where they are oxidized and taken up by foam cells in a process that leads to the development and progression of atherosclerosis. High-density lipoproteins (HDLs) oppose atherosclerosis directly, by removing cholesterol from foam cells, by inhibiting the oxidation of LDLs, and by limiting the inflammatory processes that underlie atherosclerosis. HDLs also have antithrombotic properties. Thus, HDL-cholesterol interrupts the process of atherogenesis at several key stages.

  • HDL-cholesterol
  • Atherosclerosis
  • Inflammation
  • Cardiovascular risk
  • Dyslipidaemia


Cardiovascular disease remains the world's leading cause of death, with age-standardized annual cardiovascular death rates ranging from about 60/100 000 in Japan to 700–800/100 000 in Russia and the former Soviet Republics.1 Preventing atherosclerosis holds the key to reduce the burden of cardiovascular disease, and a detailed understanding of the pathophysiology of atherosclerotic disease will facilitate the design of innovative therapeutic strategies for the management of dyslipidaemia and the prevention of morbid cardiovascular events.

The relative contributions of individual lipoproteins to overall cardiovascular risk have been intensively studied over the last several decades. The role of LDLs in causing atherosclerosis is well known. It is also well known that HDLs protect against the disease. For example, as long ago as 1976, the Framingham study showed that depressed levels of HDL-cholesterol were significantly and independently associated with an increased risk of coronary death,2 a finding confirmed by further analyses based on longer follow-up.3,4 Further, cohort studies have strengthened the association between low HDL-cholesterol and adverse coronary5 and cerebrovascular6 outcomes. Recent studies have shown that low HDL-cholesterol is common in the insulin-resistant states, such as the metabolic syndrome and type 2 diabetes, and may account for a substantial portion of the excess cardiovascular disease observed in patients with these conditions.7

It is important to understand the complex and multifactorial ways in which HDLs protect the vasculature. This review provides an overview of plasma lipoproteins and atherosclerosis with particular reference to the role of HDLs.

Overview of lipid metabolism

Normal metabolism

The liver secretes very low-density lipoproteins (VLDLs) that are rich in triglyceride and relatively poor in cholesterol (Figure 1). The endothelial enzyme, lipoprotein lipase, converts much of the triglycerides in VLDLs to free fatty acids, which are used by peripheral tissues as an energy source. As a result, the VLDL particle is converted first to an intermediate-density lipoprotein and then to an LDL particle. LDLs (the main cholesterol carrier in blood) are cholesterol-rich and triglyceride-poor when compared with VLDLs. LDLs deliver cholesterol to peripheral tissues, where it contributes to the synthesis and maintenance of cell membranes.

Figure 1 Overview of lipoprotein metabolism with special reference to the role of HDL-cholesterol. FC, free (unesterified) cholesterol; CE, cholesterol ester; TG, triglyceride; FFA, free fatty acids; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; LPL, lipoprotein lipase. Reproduced with permission from Barter PJ, Brewer HB Jr, Chapman MJ et al.11

Most extrahepatic cells (with the exception of those synthesizing steroid hormones) are unable to metabolize cholesterol, which would therefore accumulate if supply exceeded demand. HDLs are the principal means by which excess cholesterol is removed from extrahepatic cells. The major apolipoprotein (apo) of HDLs, apoA-I, is secreted from the liver in a lipid-poor form. Once in plasma, it rapidly acquires lipids to be converted into an HDL particle via several mechanisms. The initial reaction is with a membrane transporter termed the ATP binding cassette transporter A-1 (ABCA1).8 This mediates the transfer of phospholipids and some unesterified cholesterol from the peripheral cell to generate a nascent, disc-shaped HDL particle. This HDL disc then acquires further unesterified cholesterol from other plasma lipoproteins and from cell membranes. ApoA-I-mediated activation of the enzyme, lecithin cholesterol acyltransferase (LCAT), results in the esterification of the free cholesterol to form a spheroidal HDL particle containing a core of cholesterol ester.9 In addition, apoA-I-containing particles may fuse with apo-A-II-containing particles to form spherical HDLs containing both apolipoproteins.

HDL particles dispose of their load of cholesterol either by returning it directly to the liver via the scavenger receptor SR-B1 for excretion in bile or for recycling,10 or by indirectly transferring it to the VLDL/LDL fraction in a process mediated by cholesteryl ester transfer protein (CETP). The cholesterol transferred to other lipoproteins by CETP may then be delivered to the tissues (including the liver) by the LDL-receptor. This CETP-mediated transfer of cholesterol from HDLs to the VLDL/LDL fraction may be pro-atherogenic by delivering cholesterol from the protective HDLs to the pro-atherogenic VLDL/LDL particles. The effects of pharmacologic inhibition of CETP are currently under clinical investigation,11 although treatments based on this mechanism will not be available for several years.

The complexity of these interactions renders HDL particles highly heterogeneous in their shape, size, and surface charge. Most of the HDL particles in the circulation are spherical in shape and have a surface charge producing alpha migration.

Insulin resistance and atherosclerosis

Lipolysis in fat cells is normally effectively suppressed by insulin in the fed state. In insulin-resistant individuals, lipolysis persists despite post-prandial increases in circulating insulin concentrations, leading to an increased release of free fatty acids. A proportion of these free fatty acids is transported to the liver, where they promote an increased synthesis of triglycerides, an increased production of the triglyceride-rich VLDLs, and an increase in the secretion of these particles into the plasma. The increased pool of VLDLs provides a larger pool of acceptors of the cholesterol transferred from both HDLs and LDLs into VLDLs. This transfer is accompanied by a reciprocal transfer of triglyceride into LDLs and HDLs, which therefore become enriched in triglyceride. Hydrolysis of the triglyceride in LDLs and HDLs by hepatic lipase reduces their particle size and generates small, dense LDLs and small, dense HDLs. Small, dense LDLs are especially prone to oxidation and thus more likely to be taken up by macrophages in the artery wall, leading to further progression of the atherosclerotic plaque.

Pathophysiology of atherogenesis

Atherosclerosis is an inflammatory disorder that may be initiated by several factors. One of the most important factor is LDLs. LDLs enter the artery wall from plasma. They may also return to the plasma. However, if the plasma level of LDLs exceeds a threshold, they enter the artery faster than they can be removed and thus accumulate. When they accumulate, they become modified, including being oxidized. The modified LDLs then stimulate endothelial cells to express a protein, monocyte chemotactic protein-1 (MCP-1), that attracts monocytes from the blood into the artery wall. The modified LDLs also promote the differentiation of monocytes into macrophages. Macrophages, in turn, express scavenger receptors to take up the modified LDLs, resulting in the formation of lipid-filled foam cells, the hallmark cells of atherosclerosis. Macrophages express a range of cytokines, including tumour necrosis factor-alpha (TNF-α) and interleukin-1, both of which activate endothelial cells to express the adhesion molecules, E-selectin, VCAM-1, and ICAM-1. These adhesion proteins bind plasma monocytes to the endothelium where they are then attracted into the artery wall by MCP-1. Thus, the entry of LDLs into the artery wall begins a cycle that both commences atherosclerosis but also leads to its progression. Figure 2 shows the role of LDLs in these early stages of atherosclerosis. It also shows the points at which HDLs oppose this process (described subsequently).

Figure 2 Principal steps in early atherogenesis. Stages opposed by HDL-cholesterol are shown in white text on a black background. Reproduced with permission from Barter PJ, Nicholls S, Rye KA et al. Antiinflammatory properties of HDL. Circ Res 2004;95:764–772.

Anti-atherogenic properties of HDL-cholesterol

Removal of excess lipids from the vascular wall by HDL-cholesterol is a key anti-atherogenic mechanism of HDL-cholesterol, as described earlier. However, atherogenesis is more than a straightforward over-supply of cholesterol to vascular cells, and several other important mechanisms are involved. These are shown schematically in Figure 2, and the involvement of HDLs in inhibiting these processes is described as follows.

Inhibition of monocyte adhesion

Endogenous,12 purified,13,14 or reconstituted HDLs,15,16 or HDL-associated lysosphingolipids (sphingosylphosphorylcholine or lysosulfatide),17 have been shown to inhibit the expression of E-selectin or other adhesion molecules by vascular endothelial cells exposed to cytokines. The results of one of these studies,15 in which human vascular endothelial cells were exposed to TNFα in the presence and absence of a reconstituted HDLs, are shown in Figure 3. The TNFα-induced increase in the expression of both E-selectin and VCAM-1 was powerfully inhibited by exposure to HDLs.

Figure 3 Inhibition of TNFα-induced expression of the endothelial adhesion molecules, VCAM-1, and E-selectin by a reconstituted form of HDL-cholesterol. Drawn from data presented by Clay MA, Pyle DH, Rye KA et al.15

This reduced expression of adhesion molecules has also been shown to result in decreased binding of inflammatory cells, which is consistent with functional inhibition of atherosclerosis.16 Further, supportive evidence for these anti-atherogenic mechanisms of HDLs is available from clinical studies, in which increased levels of adhesion molecules correlated with low HDL-cholesterol levels.1820 Interruption of a signalling pathway involving generation of sphingosine-1-phosphate by sphingosine kinase has been proposed as a mechanism for the beneficial effects of HDL-cholesterol on the expression of adhesion molecules.21 Sphingosine-1-phosphate is involved in mediating endothelial activation and adhesion molecule expression in response to certain cytokines.

C-reactive protein is a marker of systemic inflammation but may also play a more direct role in mediating inflammation during atherogenesis. The influence of both native human HDL and a reconstituted apoA-I-containing version on C-reactive protein-induced expression of adhesion molecules was evaluated in cultured human vascular endothelial cells in vitro.22 Physiological concentrations of native HDLs and low concentrations of the reconstituted HDLs effectively abolished the increased expression of E-selectin, ICAM-1, and VCAM-1 (Figure 4). These effects appeared to arise via a novel mechanism distinct from those previously described, by which HDLs inhibit the expression of adhesion molecules by cytokines.

Figure 4 Native human HDL-cholesterol and apoA-I-containing reconstituted HDL-cholesterol inhibit the increase in E-selectin expression induced by C-reactive protein. Per cent inhibitions were calculated from mean levels of expression of E-selectin before and after exposure to HDL-cholesterol and/or C-reactive protein (mean expression after stimulation by C-reactive protein=100%). Drawn from data presented by Wadham C, Albanese N, Roberts J et al.22

Inhibition of LDL-cholesterol oxidation and MCP-1 expression

Oxidized LDLs are a potent inducer of MCP-1 expression within the developing atherosclerotic plaque. HDLs contain an enzyme, paraoxonase, which is believed to confer protection against oxidation of LDL-cholesterol in the artery wall. HDLs with and without paraoxonase were incubated with an endothelial cell line in the presence of LDL-cholesterol.23 The paraoxonase-containing HDLs significantly protected LDL-cholesterol from oxidation and inhibited expression of MCP-1, whereas the paraoxonase-deficient species were without effect on either parameter.

Co-cultures of human arterial endothelial cells with smooth muscle cells provide a useful model for evaluating the effects of HDL-cholesterol on monocyte recruitment. One study, involving incubation of such a co-culture with lipoproteins and human serum, showed that LDL-cholesterol induced a marked increase in the expression of MCP-1, and a seven-fold increase in the rate of migration of monocytes into the sub-endothelial space of the co-culture.24 Addition of purified human HDLs to the LDLs reduced the rate of monocyte infiltration by ∼90%.

Similarly, potentially beneficial effects of HDL-cholesterol have been demonstrated in human subjects. A study in women evaluated the relationships between the plasma lipid profile and expression of the CCR2 receptor, which mediates the binding of monocytes to the MCP-1 receptor.25 In women with low LDL-cholesterol, variations in HDL levels had little effect on CCR2 expression. However, in a subset of patients with high LDL-cholesterol, low HDL-cholesterol was significantly associated with increased CCR2 expression (Figure 5). There was a significant inverse correlation between HDL-cholesterol levels and CCR2 expression in women with high LDL-cholesterol (r=−0.62, P=0.028). Treatment of women who had isolated low cholesterol with oestrogen improved both the LDL-cholesterol/HDL-cholesterol ratio and approximately halved CCR2 expression.

Figure 5 Relationship between levels of HDL-cholesterol and expression of the CCR2 receptor, which recognizes the MCP-1, in women with low and high LDL-cholesterol. *Significantly different from other groups (P<0.05, Mann–Whitney U test). Drawn from data presented by Han KH, Han KO, Green SR et al.25

Antithrombotic properties of HDL-cholesterol

The final event in the evolution of a myocardial infarction or stroke is the generation of an occlusive intra-arterial thrombus. Thus, agents that reduce the coagulability of the blood, either through improved fibrinolysis or through reduced platelet aggregation, may prevent or delay a cardiovascular event. The fibrinolytic capacity of the blood is determined by the balance of activities of tissue plasminogen activator, which normally removes intravascular fibrin and its endogenous inhibitor, plasminogen activator inhibitor-1 (PAI-1). Elevated PAI-1 is often part of the cluster of metabolic risk factors associated with the metabolic syndrome, as is low HDL-cholesterol. It is perhaps not surprising, therefore, that low HDL-cholesterol has been shown to correlate with elevated PAI-1 in humans.26

A further observational study in 60 hypercholesterolaemic men showed that both levels of fibrinogen and an index of platelet aggregability were significantly associated with reduced levels of the anti-atherogenic HDL sub-fraction-2 (HDL2).27 Moreover, platelet aggregability was significantly associated with low HDL2 on multivariate analysis. A study in 132 men without history of cardiovascular disease confirmed the relationship between HDL2-cholesterol concentration and fibrinogen levels.28 The level of HDL2-cholesterol, but not that of total HDL-cholesterol or HDL3-cholesterol, was significantly lower in men with the highest quartile of fibrinogen compared with the other three quartiles. Triglycerides and VLDL-cholesterol were also significantly associated with fibrinogen levels.


Mounting clinical and experimental evidence shows that HDLs exert multiple anti-atherogenic and anti-thrombotic effects that together are consistent with a marked reduction in the risk of a morbid cardiovascular event. Indeed, the epidemiological, clinical, and experimental evidence supporting an anti-atherogenic role for HDL-cholesterol is now overwhelming. These findings support the use of therapeutic strategies to counter the common finding of low HDL-cholesterol in patients with dyslipidaemia.


View Abstract