High-density lipoprotein: why all the fuss?

Abstract

The current search for new treatments to combat coronary heart disease (CHD) is centred on increasing HDL-cholesterol. The failure of the CETP inhibitor torcetrapib may force a rethink. This perspective briefly reviews the antiatherosclerotic properties of HDL and ways HDL-cholesterol concentration can be raised, but argues – in light of the fact that HDL-cholesterol concentration does not reflect the protective properties of HDL particles – that this approach is flawed and a different approach, targeting know antiatherosclerotic components of HDL, is required.

Introduction

Major clinical trials show that despite aggressive LDL-lowering with statins there is still a significant residual risk of coronary events. This has led to increasing interest in developing therapies to decrease this residual risk, and one of the strategies aims at increasing HDL-C. However, is this approach currently practical or even sensible?

The inverse correlation between plasma HDL-C concentration and risk of coronary heart disease (CHD) has been known for many years, ¹ and several studies have shown that increasing HDL results in a decreased risk of CHD events. ² In the VA-HIT study, intervention with gemfibrozil resulted in a 3% reduction in CHD for every 0.03 mmol/L increase in HDL-C. ³ Thus, it would seem that increasing HDL-C concentration is a sensible move to alleviate excess CHD. However, recent studies have highlighted the complexity of HDL both from the point of view of its metabolism and from the different mechanisms by which HDL protects against CHD. The former has been reviewed many times and will not be dealt with here. ⁴

Antiatherosclerotic properties of HDL

Reverse cholesterol transport

Reverse cholesterol transport is a pathway whereby excess cholesterol is transported from the peripheral tissues by HDL and taken to the liver for excretion in the form of bile acids or reused. ⁵ This pathway is believed to be particularly important in removing cholesterol from macrophages in the artery wall, thereby preventing their development into foam cells, the progenitors of atheroma. ⁵

Antioxidant action

HDL has an antioxidant activity towards lipid-peroxides formed on LDL and within cell membranes. ⁶ Several protein components of HDL contribute to this activity including paraoxonase-1 (PON1), PON3, apolipoprotein A1 (apoA1) and lecithin-cholesterol acyltransferase. However, HDL particles differ in their antioxidant activity depending on the distribution of these antioxidant proteins, which are not equally distributed on all HDL particles (discussed below).

Anti-inflammatory action

Because of the ability of HDL to retard the oxidation of LDL, it can prevent the pro-inflammatory effects of oxidized-LDL (ox-LDL) on the endothelium. Thus, HDL can prevent ox-LDL-induced production of monocyte chemotactic protein-1 (MCP-1), intracellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1). ^7,8 This retards monocyte–endothelium interaction, the first stage in the inflammatory process in atherosclerosis. ⁹ Although the antioxidant proteins of HDL contribute to these anti-inflammatory actions, in certain situations they are not essential as reconstituted HDL (rHDL) micelles can also prevent VCAM-1 and ICAM-1 upregulation by ox-LDL. ⁸

In certain circumstances such as infection-induced inflammation, increased serum amyloid A and haptoglobulin displace certain HDL proteins such as apoA1 and PON1 resulting in a pro-inflammatory HDL particle. ¹⁰

Antiapoptotic action

HDL can prevent ox-LDL-induced apoptosis of endothelial and pancreatic β-cells. The exact mechanism is unclear, but it probably reflects the antioxidant activity of HDL, preventing ox-LDL from activating the intracellular signalling mechanisms. ^11,12

Reversing endothelial dysfunction

ox-LDL is a powerful inducer of endothelial dysfunction causing changes to the vasodilatory properties of the artery wall. ¹³ HDL reverses this effect by reversing the inhibition of nitric oxide synthase among other effects. ¹³ This property of HDL appears to be mediated via the interaction of HDL-associated sphingosine-1-phosphate with specific receptors on the endothelial surface, which is also one of the mechanisms to reduce tumour necrosis factor-α-mediated endothelial ICAM-1 and VCAM-1 expression. ¹⁴

HDL also has little-studied antithrombotic effects including the stimulation of prostacyclin synthesis.

Raising HDL

Life-style factors

There are a large number of life-style changes that bring about moderate increases in HDL-C, including smoking cessation, moderate alcohol consumption and exercise.

Lipid-lowering medication

The most commonly used medications to lower LDL and/or triglycerides also raise HDL. Statins do so by 5–10% (statins also convert pro-inflammatory HDL to anti-inflammatory HDL), fibrates by 10–15% and nicotinic acid (niacin) by 15–35%. The use of niacin is limited because of its widespread side-effects; however, a newly formulated version with less problems is being studied. Obviously, the increases in HDL caused by these medications (which were not designed to increase HDL-C) are insufficient to reduce residual risk, which has led to the investigation of more HDL-specific compounds such as PPARα agonists and liver X receptor agonists. However, at present these appear to be no more effective than fenofibrate or statins in improving lipid profile. ¹⁵

Peptide mimetics

Peptide mimetics are short peptides based on the amino acid sequences of apolipoproteins. Currently, mimetics based on apoA1, apoE and apoJ are being investigated. Generally, orally administered mimetics increase pre-β-HDL, improve HDL-mediated cholesterol efflux (including from macrophages), reduce lipoprotein lipid peroxides, increase PON1 activity, convert pro-inflammatory HDL to anti-inflammatory HDL and reduce atherosclerosis in animal models. ¹⁶ Thus, they appear to act by increasing HDL quality rather than quantity.

Reconstituted HDL

rHDL containing apoA1 or apoA1_Milano infused into experimental animals appears to promote the removal of arterial cholesterol. Studies in humans have shown limited success in reducing/regressing atherosclerosis. ¹⁷ However, more studies are required.

Cholesteryl-ester transfer protein (CETP) inhibitors

Based on observations that certain mutations in the Cholesteryl-ester transfer protein (CETP) gene which led to lower activity, increased plasma HDL concentrations, CETP inhibitors were developed as a therapeutic tool to raise HDL. Many studies have shown that CETP inhibitors substantially increase plasma HDL-C (>40%). ^18–20 However, the first trial which compared the CETP inhibitor torcetrapib (Pfizer, NY, USA) plus atorvastatin with atorvastatin alone, the ILLUMINATE study (Investigation of Lipid Level Management to Understand Its Impact in Atherosclerotic Events) had to be terminated prematurely because of an increase in cardiovascular events and mortality in the torcetrapib group. ¹⁸ The cause(s) of the increased mortality were unknown but were accompanied by an increase in blood pressure and plasma aldosterone concentration. This resulted in a deep depression settling over the lipid community.

An investigation into these effects compared torcetrapib with another CETP inhibitor anacetrapib (Merck, Rahway, NY, USA) using rats, mice, dogs and rhesus monkeys. ²¹ In all the models, torcetrapib but not anacetrapib caused an acute increase in blood pressure and in plasma adrenal steroid concentration. The acute increase in blood pressure was not due to the increase in adrenal steroids but was dependent on the presence of intact adrenals, indicating an adrenal toxicity of torcetrapib not shared by anacetrapib.

Yet, perhaps the most worrying aspect of the ILLUMINATE study was the failure to improve atherosclerosis, despite the massive increase in HDL-C. ^18–20

Is it enough to increase total HDL-C?

There are several reasons for believing that simply increasing HDL may not be the only answer. HDL are an extremely heterogeneous mixture of particles and they are not all equally antiatherosclerotic. ²² For example if HDL is separated by density-gradient ultracentrifugation into HDL_2a+b and HDL_3a,b+c and the particles are investigated for their antioxidant ability, then small dense HDL (HDL_3c) were found to have the highest antioxidant activity and were associated with the largest amount of PON1 activity. Yet, HDL₂ are the most antiatherosclerotic particles. In another study, HDL were separated by immunological methods into particles containing only apoA1 (LpA1) and those containing both apoA1 and apoAII (LpA1, AII). ²³ In this instance, the vast majority of PON1 mass was associated with LpA1, AII (69% vs. 31%), however, PON1 was most active on LpA1 as judged by specific activity (0.116U/μg vs. 0.067 U/μg) illustrating a further level of complexity in that particle composition is also important in determining the antiatherosclerotic properties of HDL particles.

Now is perhaps an apt time to rethink strategies towards HDL from both a pharmaceutical and academic point of view. It would perhaps be more appropriate to target known antiatherosclerotic components of HDL such as apoA1 and PON1 rather than the scattergun approach of raising total HDL-C. In any event, CETP inhibition simply results in larger cholesterol-filled HDL and measuring HDL-C is not a measure of the antiatherosclerotic quality of HDL. ²⁴ Much more research is required to uncover more viable targets to improve the antiatherosclerotic properties of HDL.