Translation of High-Density Lipoprotein Function Into Clinical Practice

HomeCirculationVol. 128, No. 11Translation of High-Density Lipoprotein Function Into Clinical Practice Free AccessResearch ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessResearch ArticlePDF/EPUBTranslation of High-Density Lipoprotein Function Into Clinical PracticeCurrent Prospects and Future Challenges Robert S. Rosenson, MD, H. Bryan BrewerJr, MD, Benjamin Ansell, MD, Philip Barter, MD, PhD, M. John Chapman, PhD, DSc, Jay W. Heinecke, MD, Anatol Kontush, PhD, Alan R. Tall, MD and Nancy R. Webb, PhD Robert S. RosensonRobert S. Rosenson From the Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, NY (R.S.R.); Cardiovascular Research Institute, MedStar Research Institute, Washington Hospital Center, Washington, DC (H.B.B.); Atherosclerosis Research Unit, Division of Cardiology, David Geffen School of Medicine at UCLA, Los Angeles, CA (B.A.); Centre for Vascular Research at the University of New South Wales, Sydney, Australia (P.B.); Dyslipidemia, Atherosclerosis and Inflammation Research Unit 939, National Institute for Health and Medical Research, University of Pierre and Marie Curie - Paris 6, Pitie-Salpetriere Hospital, Paris, France (M.J.C., A.K.) Division of Metabolism, Endocrinology, and Nutrition, University of Washington, Seattle (J.W.H.); Department of Medicine, Columbia University, New York, NY (A.R.T.); and Internal Medicine and Saha Cardiovascular Research Center, University of Kentucky College of Medicine, Lexington (N.R.W.). , H. Bryan BrewerJrH. Bryan BrewerJr From the Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, NY (R.S.R.); Cardiovascular Research Institute, MedStar Research Institute, Washington Hospital Center, Washington, DC (H.B.B.); Atherosclerosis Research Unit, Division of Cardiology, David Geffen School of Medicine at UCLA, Los Angeles, CA (B.A.); Centre for Vascular Research at the University of New South Wales, Sydney, Australia (P.B.); Dyslipidemia, Atherosclerosis and Inflammation Research Unit 939, National Institute for Health and Medical Research, University of Pierre and Marie Curie - Paris 6, Pitie-Salpetriere Hospital, Paris, France (M.J.C., A.K.) Division of Metabolism, Endocrinology, and Nutrition, University of Washington, Seattle (J.W.H.); Department of Medicine, Columbia University, New York, NY (A.R.T.); and Internal Medicine and Saha Cardiovascular Research Center, University of Kentucky College of Medicine, Lexington (N.R.W.). , Benjamin AnsellBenjamin Ansell From the Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, NY (R.S.R.); Cardiovascular Research Institute, MedStar Research Institute, Washington Hospital Center, Washington, DC (H.B.B.); Atherosclerosis Research Unit, Division of Cardiology, David Geffen School of Medicine at UCLA, Los Angeles, CA (B.A.); Centre for Vascular Research at the University of New South Wales, Sydney, Australia (P.B.); Dyslipidemia, Atherosclerosis and Inflammation Research Unit 939, National Institute for Health and Medical Research, University of Pierre and Marie Curie - Paris 6, Pitie-Salpetriere Hospital, Paris, France (M.J.C., A.K.) Division of Metabolism, Endocrinology, and Nutrition, University of Washington, Seattle (J.W.H.); Department of Medicine, Columbia University, New York, NY (A.R.T.); and Internal Medicine and Saha Cardiovascular Research Center, University of Kentucky College of Medicine, Lexington (N.R.W.). , Philip BarterPhilip Barter From the Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, NY (R.S.R.); Cardiovascular Research Institute, MedStar Research Institute, Washington Hospital Center, Washington, DC (H.B.B.); Atherosclerosis Research Unit, Division of Cardiology, David Geffen School of Medicine at UCLA, Los Angeles, CA (B.A.); Centre for Vascular Research at the University of New South Wales, Sydney, Australia (P.B.); Dyslipidemia, Atherosclerosis and Inflammation Research Unit 939, National Institute for Health and Medical Research, University of Pierre and Marie Curie - Paris 6, Pitie-Salpetriere Hospital, Paris, France (M.J.C., A.K.) Division of Metabolism, Endocrinology, and Nutrition, University of Washington, Seattle (J.W.H.); Department of Medicine, Columbia University, New York, NY (A.R.T.); and Internal Medicine and Saha Cardiovascular Research Center, University of Kentucky College of Medicine, Lexington (N.R.W.). , M. John ChapmanM. John Chapman From the Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, NY (R.S.R.); Cardiovascular Research Institute, MedStar Research Institute, Washington Hospital Center, Washington, DC (H.B.B.); Atherosclerosis Research Unit, Division of Cardiology, David Geffen School of Medicine at UCLA, Los Angeles, CA (B.A.); Centre for Vascular Research at the University of New South Wales, Sydney, Australia (P.B.); Dyslipidemia, Atherosclerosis and Inflammation Research Unit 939, National Institute for Health and Medical Research, University of Pierre and Marie Curie - Paris 6, Pitie-Salpetriere Hospital, Paris, France (M.J.C., A.K.) Division of Metabolism, Endocrinology, and Nutrition, University of Washington, Seattle (J.W.H.); Department of Medicine, Columbia University, New York, NY (A.R.T.); and Internal Medicine and Saha Cardiovascular Research Center, University of Kentucky College of Medicine, Lexington (N.R.W.). , Jay W. HeineckeJay W. Heinecke From the Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, NY (R.S.R.); Cardiovascular Research Institute, MedStar Research Institute, Washington Hospital Center, Washington, DC (H.B.B.); Atherosclerosis Research Unit, Division of Cardiology, David Geffen School of Medicine at UCLA, Los Angeles, CA (B.A.); Centre for Vascular Research at the University of New South Wales, Sydney, Australia (P.B.); Dyslipidemia, Atherosclerosis and Inflammation Research Unit 939, National Institute for Health and Medical Research, University of Pierre and Marie Curie - Paris 6, Pitie-Salpetriere Hospital, Paris, France (M.J.C., A.K.) Division of Metabolism, Endocrinology, and Nutrition, University of Washington, Seattle (J.W.H.); Department of Medicine, Columbia University, New York, NY (A.R.T.); and Internal Medicine and Saha Cardiovascular Research Center, University of Kentucky College of Medicine, Lexington (N.R.W.). , Anatol KontushAnatol Kontush From the Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, NY (R.S.R.); Cardiovascular Research Institute, MedStar Research Institute, Washington Hospital Center, Washington, DC (H.B.B.); Atherosclerosis Research Unit, Division of Cardiology, David Geffen School of Medicine at UCLA, Los Angeles, CA (B.A.); Centre for Vascular Research at the University of New South Wales, Sydney, Australia (P.B.); Dyslipidemia, Atherosclerosis and Inflammation Research Unit 939, National Institute for Health and Medical Research, University of Pierre and Marie Curie - Paris 6, Pitie-Salpetriere Hospital, Paris, France (M.J.C., A.K.) Division of Metabolism, Endocrinology, and Nutrition, University of Washington, Seattle (J.W.H.); Department of Medicine, Columbia University, New York, NY (A.R.T.); and Internal Medicine and Saha Cardiovascular Research Center, University of Kentucky College of Medicine, Lexington (N.R.W.). , Alan R. TallAlan R. Tall From the Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, NY (R.S.R.); Cardiovascular Research Institute, MedStar Research Institute, Washington Hospital Center, Washington, DC (H.B.B.); Atherosclerosis Research Unit, Division of Cardiology, David Geffen School of Medicine at UCLA, Los Angeles, CA (B.A.); Centre for Vascular Research at the University of New South Wales, Sydney, Australia (P.B.); Dyslipidemia, Atherosclerosis and Inflammation Research Unit 939, National Institute for Health and Medical Research, University of Pierre and Marie Curie - Paris 6, Pitie-Salpetriere Hospital, Paris, France (M.J.C., A.K.) Division of Metabolism, Endocrinology, and Nutrition, University of Washington, Seattle (J.W.H.); Department of Medicine, Columbia University, New York, NY (A.R.T.); and Internal Medicine and Saha Cardiovascular Research Center, University of Kentucky College of Medicine, Lexington (N.R.W.). and Nancy R. WebbNancy R. Webb From the Cardiovascular Institute, Icahn School of Medicine at Mount Sinai, New York, NY (R.S.R.); Cardiovascular Research Institute, MedStar Research Institute, Washington Hospital Center, Washington, DC (H.B.B.); Atherosclerosis Research Unit, Division of Cardiology, David Geffen School of Medicine at UCLA, Los Angeles, CA (B.A.); Centre for Vascular Research at the University of New South Wales, Sydney, Australia (P.B.); Dyslipidemia, Atherosclerosis and Inflammation Research Unit 939, National Institute for Health and Medical Research, University of Pierre and Marie Curie - Paris 6, Pitie-Salpetriere Hospital, Paris, France (M.J.C., A.K.) Division of Metabolism, Endocrinology, and Nutrition, University of Washington, Seattle (J.W.H.); Department of Medicine, Columbia University, New York, NY (A.R.T.); and Internal Medicine and Saha Cardiovascular Research Center, University of Kentucky College of Medicine, Lexington (N.R.W.). Originally published10 Sep 2013https://doi.org/10.1161/CIRCULATIONAHA.113.000962Circulation. 2013;128:1256–1267IntroductionHigh-density lipoproteins (HDLs) represent a spectrum of particles that vary in their physicochemical and functional properties.1 It has been shown in many population studies that the concentration of HDL cholesterol (HDL-C) is inversely related to the risk of having a cardiovascular disease (CVD) event. In this paradigm, HDL-C has been considered to be a marker of the potentially cardioprotective functions of HDL. However, recent studies have suggested that the simple concentration of HDL-C may not always reflect HDL function, with growing evidence that under some circumstances HDL function may be compromised despite high concentrations of HDL-C.The best known of the potentially antiatherogenic functions of HDLs is their ability to promote cholesterol efflux from cells, including that from macrophages in the arterial wall.2 Cellular cholesterol efflux is achieved by several mechanisms. One involves the interaction of phospholipid-depleted and cholesterol-deficient apolipoprotein (apo) A-I complexes (discoidal, pre–β-migrating particles [very small HDL] with the ATP-binding cassette transporter A1 (ABCA1) in a process that results in the formation of a heterogeneous population of nascent HDL particles that are discoidal in shape and contain apoA-I, phospholipids, and free cholesterol. A proportion of the free cholesterol is subsequently esterified by lecithin:cholesterol acyltransferase (LCAT); this enzyme generates a core of cholesteryl esters in a process that converts HDL particles from discoidal, very small, pre-β1-migrating particles into spherical, α-migrating particles (small HDL]).1 The interaction of spherical HDL particles with other active cellular transporters such as ABCG1 and passive diffusion of cellular cholesterol further increase the cholesterol load of HDL. However, it is often unappreciated that peripheral cholesterol efflux contributes <5% of the cholesterol content of HDL.2 Thus, HDL-C is an inadequate surrogate measure for the most heralded of HDL functions.Various HDL subpopulations differ in other antiatherogenic functions that extend beyond macrophage cholesterol efflux. Small, protein-enriched, cholesterol-depleted HDL particles possess antioxidant, anti-inflammatory, cytoprotective, antithrombotic, anti-infective, and endotoxin-neutralizing activities.1,3 Structure-function analyses suggest that the simple measurement of HDL-C may not always be reflective of HDL functionality.The challenge is to develop laboratory assays that quantify the various HDL functions that may improve CVD risk assessment and augment the evaluation of HDL-modifying therapies. Efforts to develop reproducible, cost-effective, validated assays that measure the potentially protective functions of HDL are now recognized as a major challenge for the cardiovascular field. Currently, there is no consensus concerning the HDL functions that should be targeted, nor are there standardized assays to measure HDL function as a tool to improve either CVD risk assessment or the assessment of therapeutic interventions (Figure 1). Another challenge is to validate measurements of HDL particles to be able to standardize assays of function with HDL quantification.Download figureDownload PowerPointFigure 1. Validation of assays of high-density lipoprotein (HDL) functionality. The development of functional assays for HDL requires validation in studies of atherosclerosis and atherosclerotic cardiovascular events. In this model, we propose that candidate biomarkers of HDL functionality will be evaluated from specimens obtained in trials of HDL-altering therapies. After an association is established, these functional biomarkers will be prospectively evaluated in studies of atherosclerosis progression and clinical atherosclerotic cardiovascular events.In this article, we review currently available measures of HDL function, explore the potential contribution of functional assays to understanding the mechanisms of atherosclerotic CVD, and describe the involvement of the proteome and lipidome in HDL structure-function relationships. To improve the understanding of HDL functionality, we propose a framework for future investigations addressing the validation and clinical application of HDL functional assays that may have a role as surrogates of CVD (Figure 1).Measures of HDL Subclasses and Compositional Determinants of HDL FunctionalityConventionally, HDL concentration is reported in terms of the cholesterol concentration measured within the ultracentrifugally defined density range of 1.063 to 1.21 g/L.1 Further divisions within this density range have given rise to specific terminology for HDL subclasses. Pre-beta HDL particles distribute over a range of hydrated density from approximately 1.21 to 1.25 g/L. Other analytic methods have also been used to describe HDL subclasses based on electrophoretic mobility and apolipoprotein composition. There is also good evidence that the concentration of HDL, HDL particle number (HDL-P), provides clinically useful information that is distinct from HDL-C. Two methods, nuclear magnetic resonance (NMR) spectroscopy4 and ion mobility (IM),5 have been used to quantify HDL-P (details below). However, these methods give different estimates of HDL-P concentration and size.4,5 In future studies, it will be critical for investigators to validate the quantification of HDL-P by NMR and IM.Recently, a uniform nomenclature for HDL subclasses has been proposed that is based on physicochemical properties.1 However, all methods used to assess HDL subclasses have their limitations in that they measure only static concentrations with no assessment of the dynamic processes regulating either HDL subclass concentrations or their potential relationship to atherosclerosis.There is evidence that quantification of lipoprotein particle concentration may be superior to the simple measures of lipoprotein-cholesterol as an indicator of CVD risk assessment. Low-density lipoprotein (LDL) particle number can be measured directly by NMR. The particle concentration of the combined LDL, intermediate-density lipoprotein, and very-low-density lipoprotein fractions can be determined from the plasma concentration of apoB because lipoprotein particles in each of these fractions contain a single molecule of apoB. In individuals with low HDL-C levels, CVD risk is often associated with high LDL particle numbers or its surrogate measure, apoB.6–8 In contrast to LDL, HDL particles contain 2 to 5 molecules of apoA-I.9 As a consequence, the concentration of apoA-I cannot be used to quantify HDL-P. At present, NMR and IM are the only available methods for ascertaining HDL-P.1 In some recent studies, HDL-P concentration has emerged as a predictor of CVD risk that may be superior to that of HDL-C in both population studies10,11 and randomized, clinical trials of lipid-modifying therapies.8,12,13 In the Multi-Ethnic Study of Atherosclerosis (MESA), low HDL-P predicted higher risk of elevated carotid intima-medial thickness regardless of whether the baseline HDL-C level was high (≥55 mg/dL) or low (<42 mg/dL).11 In the Justification for the Use of Statins in Prevention: An Intervention Trial Evaluating Rosuvastatin (JUPITER), HDL-P was a better marker of residual risk in statin-treated patients than chemically measured HDL-C, apoA-I, or average HDL size.13Two separate nested studies of the Veterans Affairs High-Density Lipoprotein Intervention Trial (VA-HIT) trial used 2 different analytical methods for quantifying HDL subclasses to investigate the importance of HDL subclass distribution in the prediction of CVD events among coronary heart disease patients with low HDL-C levels (<40 mg/dL).8,14 In a case-control study, NMR-determined small HDL particle subclass levels measured at baseline and on trial were predictors of coronary heart disease events (odds ratio, 0.71; 95% confidence interval, 0.60–0.84; P<0.0001 for baseline measures; and odds ratio, 0.67; 95% confidence interval, 0.57–0.79; P<0.0001 for on-trial measures), whereas the risk associated with medium HDL particle concentration was weaker (odds ratio, 0.82; 95% confidence interval, 0.70–0.96; P<0.02 for baseline measures; and odds ratio, 0.82; 95% confidence interval, 0.69–0.97; P<0.02 for on-trial measures) and for large HDL particles nonsignificant in multivariate models that included major coronary heart disease risk factors, plasma lipids, and NMR-measured lipoprotein subclasses.8 In gemfibrozil-treated patients, analysis by 2-dimensional gradient gel electrophoresis indicated that elevated levels of pre–β-HDL (very small HDL particles) and low levels of α1 HDL (very-large HDL) and α2 HDL (large HDL) were associated with increased risk of CVD events in multivariate models that adjusted for nonlipid and lipid risk factors.14 These conflicting findings may have resulted from differences in the study design (case-control versus cohort) and variables included in the statistical models. Other possible differences in the associations between very small HDL particles and coronary heart disease risk in VA-HIT may result from the analytical methods in which 2-dimensional gradient gel electrophoresis accurately quantifies pre–β-HDL, whereas it has not been reported whether this HDL subclass is detected by NMR. Consistent with the increased risk associated with high levels of pre–β-HDL in VA-HIT, high pre–β-1 HDL levels predict increased risk of myocardial infarction.15 These data suggest that impaired maturation of HDL particles increases CVD risk.From these studies, we conclude that the inclusion of a measure of atherogenic lipoproteins (LDL-P or apoB) in multivariate models is crucial in the assessment of HDL-associated risk resulting from the inverse correlation between the concentration of atherogenic lipoproteins and HDL-C and large HDL subclasses6 and that determination of HDL-P and individual concentrations of HDL subclasses should be considered in any clinical study that investigates HDL functionality.HDL FunctionalityThe ability of HDL to promote efflux of cholesterol from macrophages in the artery wall is the best known of the potentially cardioprotective functions of HDL.2 However, HDL particles have additional properties with the potential to protect against vascular disease, some of which are related and others are unrelated to cholesterol transport and homeostasis (Table 1). This section discusses the major functional roles of HDL and the available clinical measures currently used to evaluate these functions in clinical studies.Table 1. Major Anti-Atherosclerotic Functional Roles of HDL With Available Clinical Measures•Macrophage cholesterol efflux•Anti-oxidative effects•Anti-inflammatory effects•Endothelial function•Glucose homeostasisCholesterol EffluxReverse cholesterol transport is a term used to describe the efflux of excess cellular cholesterol from peripheral tissues and its return to the liver for excretion in the bile and ultimately the feces. It is believed to be a critical mechanism by which HDL exerts a protective effect on the development of atherosclerosis; equally, it is a critical component of the system that maintains cholesterol homeostasis. In this paradigm, cholesterol is effluxed from arterial macrophages to extracellular HDL-based acceptors through the action of active transporters and passive diffusion.2 After efflux to HDL, cholesterol may be esterified in the plasma by LCAT, and it is ultimately transported from HDL to the liver, either directly via the scavenger receptor BI (SR-BI) or after transfer to apoB-containing lipoproteins by the cholesteryl ester transfer protein for ultimate disposition in the feces. However, isotope kinetic studies and mass measurements of cholesterol and bile acids suggest that effective macrophage cholesterol efflux may be atheroprotective even when biliary and fecal sterol excretion is not increased.2 Thus, revision of earlier models of reverse cholesterol transport was recently proposed to more accurately describe the critical steps required for effective HDL-mediated atheroprotection via promotion of macrophage cholesterol efflux.2Macrophage cholesterol efflux capacity is influenced by the physicochemical properties of HDL and the interaction of these HDL subclasses with cellular transporters.2 As indicated in the previous section, the ABCA1 transporter interacts with cholesterol-deficient and phospholipid-depleted apoA-I complexes, whereas ABCG1 and SR-BI interact with spherical HDL particles of various sizes.2Alterations in HDL protein and lipid composition may alter cholesterol efflux. Loss of apoA-I secondary to its replacement by serum amyloid A, as occurs under proinflammatory conditions of arthritis, uremia, or psoriasis, reduces HDL-mediated cholesterol efflux. Equally, glycation and oxidation of HDL proteins may adversely affect cholesterol efflux and other antiatherogenic functions of HDL.HDL surface lipid composition and interaction between lipid molecules have been shown to affect cholesterol efflux. Enrichment of HDL particles in triglycerides or depletion of phospholipid may render them deficient in their capacity to efflux cellular cholesterol via ABCA116 and SR-BI.18,19 Impaired cholesterol efflux from macrophages can also result from the accumulation in HDL of oxidized sterols, including 7-ketocholesterol.19,20 Qualitatively, the surface rigidity of HDL particles, which is partly regulated by the relative proportions of sphingomyelin and free cholesterol in the surface lipid monolayer of HDL,21,22 influences the capacity of HDL particles to serve as an acceptor of cholesterol.23,24 HDL enrichment in sphingomyelin enhances cholesterol efflux via direct interaction between sphingomyelin and cholesterol molecules, thereby counteracting the effects of diminished fluidity.17,25 In macrophage-like human THP-1 cells, cellular cholesterol efflux capacity correlated with percent weight of phosphatidylcholine and inversely correlated with percent weight of sphingomyelin.24 In addition, sphingomyelin inhibits LCAT activity and impairs maturation of HDL particles.26Clinically, ex vivo assays have been used to assess the capacity of individual patient serum and HDL specimens to remove cholesterol from cultured cholesterol-loaded macrophages. At present, in vivo quantification of macrophage reverse cholesterol transport can be determined only in animal models; however, studies are underway to develop methods for use in humans. The J774 mouse macrophage cell line has been used extensively for ex vivo cholesterol efflux studies. Typically, the cells are lipid loaded and treated with cAMP or a liver X receptor agonist to increase the magnitude of cholesterol efflux, particularly via the ABCA1 pathway.27 Human macrophage THP-1 cells provide an alternative to the J774 model, which features a slightly different expression pattern of major proteins that are involved in cholesterol efflux and may be more relevant for human atherosclerosis.28 SR-BI–mediated cholesterol efflux may be determined with the Fu5AH hepatoma cell line.29 From studies of macrophage cholesterol efflux with the J774 cell line, serum specimens having similar HDL-C or apoA-I levels can exhibit significant differences in fractional efflux.30 A comparison of the contribution of efflux pathways of high- and low-efficiency HDL demonstrated that the increased cholesterol efflux observed in the higher-efficiency sera was attributed largely to greater efflux via the ABCA1 pathway.30A recent study reported that the capacity of individual patient serum to stimulate cholesterol efflux from J774 macrophages has a strong inverse association with angiographically quantified coronary artery disease (CAD) that is independent of HDL-C or apoA-I levels.31 The efficiency of cholesterol efflux in CAD patients was most strongly associated with HDL-C; however, it accounted for only 26% of the reported variation. The role of cholesterol efflux in apoB-depleted serum as a predictor of cardiovascular risk remains controversial.32 A more recent study confirmed the finding that increased cholesterol efflux activity in apoB-depleted serum is associated with reduced risk of prevalent CAD. Unexpectedly, however, higher cholesterol efflux activity was also associated with an increase in prospective (3 years) risk of myocardial infarction, stroke, and death. Regardless of the cellular model, elevated experimental between-assay variability (coefficients of variation close to 10%32) has been reported compared with <4% for analytic measurements.33 Such variability results from the very nature of the cell culture approach used, requires normalization on the basis of the efflux capacity of a serum pool run with each assay, and has been an impediment to the development of these assays for use in clinical practice.Future use of the measurement of HDL-mediated efflux in CVD risk assessment will require integration of advanced vascular imaging in human studies that quantify the volume and composition of atherosclerotic plaques and atherosclerotic CVD events (Figure 1).2 Guidance in the development of new HDL-targeted therapies for humans will require screening of large numbers of serum specimens, and the use of radiolabeled cholesterol for large-scale screening is generally not practical. Thus, fluorescent dipyrromethene boron difluoride cholesterol34 may serve as a substitute for the labeled cholesterol, which would allow the development of a fluorescence-based high-throughput efflux assay.Endothelial FunctionHDL particles have direct effects on endothelial function that are considered antiatherosclerotic and antithrombotic. Specifically, HDL particles isolated from healthy individuals induce the expression of endothelial nitric oxide (NO) synthase (eNOS) and synthesis of NO by endothelial cells, inhibit adhesion molecule expression,35 promote endothelial cell migration contributing to endothelial repair,36 and attenuate tissue factor expression.35ApoA-I appears to be critically involved in the effects of HDL on the endothelium.37 The protective effects of HDL on eNOS production and endothelial repair may be partly dependent on processes that involve SR-BI.38 In a process dependent on apoA-I–dependent binding to SR-BI in endothelial cells, SR-BI initiates a signaling cascade that involves PDZK1-dependent activation of the Src family kinases PI3K and AKt, which phosphorylate eNOS at Ser177, increasing enzyme activity.39,40 Akt-activating phosphorylation (AKt-Ser473) and eNOS-activating phosphorylation (eNOS-Ser1177) diminish LOX-1 activation and inhibit protein kinase Cβ-II activation of AKt and eNOS phosphorylation events. Reduced protection from endothelial apoptosis in CAD and acute coronary syndrome patients was associated with lower HDL content and higher apoC-III content.41 Furthermore, accumulation of symmetrical dimethylarginine in HDL from patients with chronic kidney disease renders HDL dysfunctional and results in the activation of Toll-like receptor-2.42Among the lipid components, S1P, a minor HDL lipid, can serve as a ligand for the family of G protein–coupled S1P receptors that are present on endothelial cells and smooth muscle cells.21 ApoM, a lipocalin that resides primarily on HDL,1 induces endothelial S1P1 receptor internalization, endothelial cell migration, and formation of endothelial adherent junctions.44 HDL-associated S1P may stimulate eNOS through activation of the lysophospholipid receptor S1P3.44 Other HDL-associated sphingolipids such as sphingosylphosphorylcholine and lysosulfatide may also enhance endothelial cell migration and survival and the cytoprotective effects of HDL.45,46 In contrast, elevated content of triglycerides and oxidized lipids47 in HDL can exert deleterious effects on endothelial function, as observed in patients with type 2 diabetes mellitus.Endothelial oxidant stress is another important determinant of endothelium-dependent vasorelaxation.48 HDL isolated from healthy individuals carries active PON1, which inhibits the formation of oxidized lipids and lipoproteins such as malonyldialdephyde.49 In contrast, HDL isolated from CAD patients has a loss of PON1 activity and inhibition of eNOS phosphorylation cascades.There is evidence that reconstituted HDL enhances endothelial function in vivo in humans with normal cholesterol levels50 and in individuals with low HDL-C