- Email: [email protected]
High-density lipoprotein particles, coronary heart disease, and niacin
Journal of Clinical Lipidology (2010) 4, 405–410
NLA Symposium on High Density Lipoproteins
High-density lipoprotein particles, coronary heart disease, and niacin Bela F. Asztalos, PhD* Lipid Metabolism Laboratory, Tufts University, 711 Washington Street, Boston, MA 02111, USA KEYWORDS: High density lipoprotein; HDL particles; HDL-C; CHD; Niacin; HDL raising
Abstract: The use of statins in patients with high risk for cardiovascular disease (CVD) has resulted in a 30–40% decrease in clinical events in the last couple of decades. However, despite of a marked reduction (up to 60%) in LDL-C, about 30% of patients continue to have CVD events. This high residual risk in statin-treated patients initiated the search for new ways to reduce CVD risk. HDL is the next logical target. Epidemiological and cross-sectional studies identified low HDL-C level as an independent risk for CVD.1,2 Based on the Framingham Heart Study data, HDL-C ,35 mg/dl was established an independent risk factor and HDL-C >60 mg/dl as protective.3 Presently the cut point is ,40 mg/dl for men and ,50 mg/dl for women. Ó 2010 National Lipid Association. All rights reserved.
In clinical trials, the use of statins in patients with high risk for cardiovascular disease (CVD) has resulted in a 25% to 40% decrease in major clinical events. However, despite a marked reduction (up to 60%) in LDL-C, approximately 50% (or more) of patients continue to have CVD events. This high residual risk in statin-treated patients initiated the search for new ways to reduce CVD risk. High-density lipoprotein (HDL) is the next logical target. The authors of epidemiological and cross-sectional studies identified low HDL-C level as an independent risk factor for CVD.1,2 On the basis of the Framingham Heart Study data, HDL-C less than 35 mg/dL was established as an independent risk factor and HDL-C greater than 60 mg/dl was defined as protective3 and considered to reduce the global risk estimate. Presently the cut point for defining HDL-C as low is less than 40 mg/dL for men and less than 50 mg/dL for women.
* Corresponding author. E-mail address: [email protected] Submitted August 3, 2010. Accepted for publication August 7, 2010.
Whether an HDL-C level greater than 60 mg/dL can be regarded as protective in all patients has recently been debated. Some investigators have provided evidence for the existence of dysfunctional HDL in patients with vascular disease; however, the term ‘‘dysfunctional HDL’’ has not been well and consistently defined. Dysfunctional HDL is generally thought of as not protective against CVD even if HDL-C is present in high levels. It is assumed that dysfunctional HDL has altered composition and altered functions. Studies in mice have demonstrated that decreased scavenger receptor B1 (SR-B1) reduces clearance of HDL-C and results in greater plasma concentrations. HDL particles become larger with more cholesterol content, but the cholesterol transport into the liver and into the bile is decreased; cholesterol transport from peripheral tissues is impeded and the resulting fails to protect from developing arteriosclerosis. Low HDL-C levels can be inherited dyslipidemia expressed as increased concentrations of the triglyceride-rich lipoproteins, very low density lipoproteins (VLDL), and chylomicrons. This form of dyslipidemia is responsible for 15% or more of premature CHD cases. The specific disorders
1933-2874/$ - see front matter Ó 2010 National Lipid Association. All rights reserved. doi:10.1016/j.jacl.2010.08.012
406
Figure 1 Reduction of CVD events (%) in patients treated with an array of drugs versus placebo treatment.
of familial combined hyperlipidemia (FCHL)4 have been reported to be responsible for 14% of premature CHD cases, and fhmya be responsible for approximately 4% of premature CHD cases. Factors associated with sedentary lifestyle, obesity, smoking, and poor control of diabetes mellitus contribute to both increased triglycerides and low HDL-C. In a population of 1439 patients pooled from four clinical trials of statin therapy, LDL-C was reduced from a mean of 124 to 88 mg/dL and HDL-C increased from a mean of 42.5 to 45.1 mg/dL. These statin-induced changes in both LDL-C and HDL-C were found to be significant and independent correlates of atheroma regression assessed with intravascular ultrasound.5 In another meta analysis in which researchers used data from several drug intervention trials, reduction in 1-year event rates was the greatest in subjects with the greatest reduction in LDL-C and the greatest increase in HDL-C (Fig. 1).6 HDL is the smallest and densest of all plasma lipoproteins. It consists of a number of distinct particles that vary in size, shape, density, surface charge, composition, and physiological functions. HDL has several potentially antiatherogenic properties. The best known of these is the ability of HDL to remove cholesterol from peripheral cells, including macrophages in the artery wall and transporting cholesterol to the liver for biliary excretion. Other HDL functions include inhibition of LDL oxidation, improvement in endothelial function, inhibition of the binding of monocytes to the endothelium, promotion of endothelial repair, stabilization of nitric oxide synthesis, as well as antithrombotic and anti-inflammatory properties. HDL is also a key player in innate immunity. At least 49 proteins have been identified in HDL by mass spectrometry, some related to lipid metabolism, others to complement regulation or to acute-phase response, and a few related to proteinase inhibitors.7 Several cell surface-bound and soluble plasma proteins influence the charge, size, shape, and composition of HDL. These factors include ATP binding cassette transporter A1 (ABCA1), ABCG1, lecithin:cholesterol acyltransferase (LCAT), cholesteryl ester transfer protein (CETP), phospholipid transfer protein, lipoprotein lipase, hepatic lipase (HL), endothelial lipase, secretory phospholipase A2, and SR-B1. The various HDL particles in human plasma are classified by the use of many different parameters dependent on the
Journal of Clinical Lipidology, Vol 4, No 5, October 2010 separation method used. In this article, we discuss only the apoA-I2containing HDL particles, separated by twodimensional (2D), nondenaturing gel electrophoresis, detected by immunoblotting for apoA-I, quantified by image-analysis (Fig. 2). With this method, 12 distinct apoA-I–containing HDL subclasses are separated by electrophoretic mobility in the first dimension (pre-b-, a-, and pre-a-mobility particles) and by size in the second dimension (5.4211.7 nm).8 In addition to apoA-I, these HDL particles contain other lipoproteins and plasma proteins and a variety of lipids depending on their size (Fig. 3). Ten of the 12 HDL subpopulations contain no apoA-II (LpA-I particles) whereas two of them, a-2 and a-3, contain both apoA-I and apoA-II (LPA-I:A-II particles). By using the nondenaturing 2D method, we have mapped the sequence and mechanisms of many of the steps in HDL metabolism/remodeling in humans by collecting samples from patients with mutations in genes influencing HDL metabolism/remodeling. We have collected plasma samples from subjects carrying one (heterozygous) and two (homozygous) defective alleles encoding: apoA-I,9 ABCA1,10 LCAT,11 CETP,12 HL, and lipoprotein lipase. In addition, we have performed cell-cholesterol efflux assays via the ABCA1, ABCG1, and SR-B1 pathways.13 On the basis of the results from these studies, we have built a working hypothesis of HDL metabolism/remodeling illustrated in Figure 4. ApoA-I is synthesized in the liver and in the intestine and secreted as a lipid-free protein. Lipid-free apoA-I picks up some phospholipids and forms preb-1 particles (precursor HDL particles) in the circulation. Preb-1 HDL is a homodimer apoA-I particle. Preb-1 particles interact with ABCA1 on the surface of liver and peripheral cells and pick up more phospholipids and cell-derived cholesterol (unesterified cholesterol) and are transformed into the smallest a-mobility particle, a-4. LCAT can act on a-4 particles, transforming free cholesterol into cholesteryl ester, acquiring other neutral lipids and proteins, and growing in size. The two largest HDL particles are a-1 and a-2. These are the major HDL subclasses in control subjects, but their concentrations in CHD patients are low (Fig. 2). These large HDL particles can deliver cholesterol directly to the bile via the SR-B1 transporter expressed on the service of hepatocytes. Alternatively, cholesterol can be transferred from these large HDL particles to TG-rich lipoproteins by CETP. A high concentration of TG-rich lipoproteins— substrate for CETP—results in cholesterol removal from a-1 HDL particles in exchange for TG. TG-enriched a-1 HDL substrate for progressive lipolysis by HL are converted into smaller particles. For this reason, subjects with increased TG levels often have reduced HDL-C and small HDL particles. Patients with CHD often have decreased concentrations of HDL-C and fewer large a-1 HDL but increased concentrations of small a-3 and preb-1 particles (Fig. 3). In the Framingham Offspring Study, we found that in men, with and without CHD, for every 1 mg/dL decrease in a-1 HDL level, there was a 26% increase for CHD risk.14
Asztalos
HDL particles, coronary heart disease, and niacin
407
Figure 2 Apo-A-I-containing HDL subpopulations of a control and a CHD patient. Panel (a): apoA-I-containing HDL subpopulations of a normolipidemic male subject, separated by 2d electrophoresis. The inserted rectangle shows separation in the 1st dimension on agarose gel by charge into preb-, a-, and prea-mobility subfractions. In the 2nd dimension, HDL particles were further separated by size on concave-gradient polyacrylamide gels under non-denaturing conditions. Molecular weight/size standard proteins can be seen on the left side of panel (a). The asterisk represents the position of endogenous albumin marking the a-front. Panel (b): apoA-I-containing HDL subclasses of a male CHD patient. Panel (c): schematic representation of HDL particles containing only apoA-I (LpA-I) light grey, and HDL particles containing both apoA-I and apoA-II (LpA-I:A-II) dark grey.
We also found that HDL particles were significantly better predictors for CHD events than HDL-C itself. Earlier it was shown that every 1 mg/dL increase in HDL-C levels was associated with approximately a 2% decrease in CVD risk. In the Veterans Affairs HDL Intervention Trial (VA-HIT), low levels of a-1 and a-2 HDL and high level of preb-1 HDL particles predicted recurrence of CVD events versus no recurrence in men who were selected with low HDL-C level and established CVD.15 A low a-1 HDL level was the most significant predictor for recurrence of CVD events.
Ways to increase HDL-C levels
Figure 3 Working theory of the physicochemical structure and size of the major apoA-I-containing HDL subpopulations. Light yellow cylinders represent apoA-I and darker yellow cylinders represent apoA-II molecules. Blue, green, and purple spheres represent phospholipid (PL), free cholesterol (FC), and TG respectively. The chemical composition of preb-2 has not been investigated yet; therefore it is labeled as solid black.
Currently, there are three major drug classes—fibrates, statins, and nicotinic acid—that are used to treat dyslipidemia or increase HDL-C level. Fibrates increase HDL-C level by approximately 10%. This is variable with no increase in some studies to more than 20% in those with very high triglyceride concentrations. Despite increasing HDL-C, fibrates slightly decrease the concentration of large a-1 HDL and slightly increase the concentration of small a-3 HDL. As shown previously, concentrations of the large HDL particles are inversely associated while those of the small HDL particles which are positively associated with CVD. Therefore, the effects of fibrates on HDL subclasses may be unfavorable.16 Considering this result and other uncertainties about fibrates, it is not understood how this drug class protects against CVD. There is the possibility that fibrates protect from CVD not through increasing HDL levels but rather through other mechanisms. Statins increase HDL-C level up to 10% and increase large a-1 HDL particles up to 36%. Moreover, statins decrease preb-1 concentrations up to 40%. The efficacy of
408
Journal of Clinical Lipidology, Vol 4, No 5, October 2010
Figure 4 Illustration of HDL metabolism based on research in patients with rare inborn errors of HDL metabolism. Summary of findings: 1) apoA-I is not absolutely necessary for the formation of lipidated HDL particles; 2) preb-1, a very small discoidal precursor HDL particle, interacts with ABCA1 to promote cellular cholesterol and phospholipid efflux and in doing so is converted into small, LpA-I a-4 HDL particles; 4) a-4 is acted upon by LCAT which converts its free cholesterol to cholesteryl ester. Then a-4 serves as an acceptor for apoA-II (secreted in already lipidated form by the liver), eventually forming the smaller LpA-I:A-II HDL subclass (a-3); 5) the small LpA-I a-4 particles can pass through a number of other steps in order to form large spherical a-1 HDL (also LpA-I), which can effectively deliver cholesterol to the liver. Data collected from subjects with defects in the genes encoding CETP, LPL, and HL indicated that HDL maturation is not a simple linear size-increase process. These steps are explained in more detail in “High Density Lipoproteins, Dyslipidemia, and Coronary Heart Disease”, Springer 2010, edited by EJ Schaefer.
the different statins regarding beneficial HDL particle alteration in decreasing order is rosuvastatin, atorvastatin, simvastatin, pravastatin, and lovastatin.17,18 Statin-mediated changes in HDL are associated with decreased CETP activity but not with changes in apoA-I kinetics. Niacin increases the concentration of HDL-C up to 25% and that of the large a-1 HDL particles up to two-fold while decreasing the concentration of the small preb-1 particles. The effects of niacin on HDL are similar to those of statins but produce greater and more consistent changes. Niacinmediated changes in HDL are also associated with increases in apoA-I production and ABCA1 expression in the liver (unpublished data). Niacin’s effects on serum cholesterol were published first by Rudolph Altschul19 in 1955. Niacin has been used clinically to lower cholesterol for more than five decades, but we have no full explanation of its mechanism of action. We need more data on how niacin works in humans and how it protects against CVD. The authors of placebo-controlled studies show that niacin increases HDL-C, decreases LDL-C and TG levels, as well as lipoprotein(a) (Lp(a)) levels, in a dose-responsive pattern. Elevated Lp(a) is associated with increased risk, and niacin is the only commercially available lipid lowering medication that decreases Lp(a) concentrations.
We have studied the effects of niacin in combination with simvastatin in the HDL-Atherosclerosis Treatment Study (HATS) population. Patients received niacin and simvastatin, niacin, simvastatin and antioxidant vitamins, antioxidant vitamins alone, or placebo. Coronary artery atherosclerosis was assessed quantitative angiography, and lipoproteins were analyzed in samples both on and off treatment.20 A significant association between change in coronary artery stenosis and change in a-1 HDL concentrations was observed in the entire cohort including all treatment groups. In the first tertile of a-1 HDL concentrations, the mean particle level decreased 15% and stenosis increased 2.1%. This group of subjects was primarily composed of those taking antioxidants and placebo. In the third tertile with the highest mean a-1 HDL particle level, having increased 157%, there was no significant change in stenosis suggesting stabilization of the disease. This group consisted primarily of niacin and simvastatin. Many patients experienced regression of the disease. This was the first study to show that a known drug-combination can cause reductions in plaque volume.21 Increasing HDL particles appeared to be confirmed in humans by a study using the direct infusion of A-I Milano.22 In conclusion, niacin treatment in patients with combined dyslipidemia significantly decreases TG and LDL-C
Asztalos
HDL particles, coronary heart disease, and niacin
concentrations and significantly increases HDL-C and apoA-I levels. Moreover, with niacin-simvastatin combination therapy, the level of the large protective a-1 HDL particles, believed to be linked to regression of coronary stenosis, is doubled. We hypothesize that the increase of apoA-I mass during niacin treatment is caused by a significant increase in the production of apoA-I. This is one of the differences between niacin and statins, because statins decrease the clearance while niacin increases the production of HDL particles. We have found that niacin has no effect on apoA-II metabolism, and the percent change in apoA-I production rate is positively associated with the percent change of HDL-C concentrations. Recently, we developed a monkey model to study the mechanisms of the protective effects of niacin in vivo. Monkeys fed an atherogenic diet (high in saturated-fat and cholesterol) developed a very similar lipoprotein profile as humans on a similar diet, with significant increases in TG, apoB, and cholesterol in apoB-containing particles. Similar to humans, HDL-C concentration significantly increased with no significant increase in apoA-I concentration. This usually signifies an increase in the concentration of the large lipid-rich HDL particles. In contrast, a significant increase in the concentration of small HDL particles and a significant decrease in the concentration of the largest HDL particles were observed. This change in the HDL particle subpopulation profile indicates a less favorable risk status, even though HDL-C is increased. We hypothesize that an atherogenic diet increases plasma triglyceride and TG-rich lipoprotein concentrations, which in turn increases CETP activity and, very likely, hepatic lipase activity on HDL (see Fig. 4). Augmented CETP activity increases TG concentrations in the large HDL particles. TG-enriched HDL particles provide good substrate for HL-dependent lipolysis. The increased CETP and hepatic lipase activities promote large HDL particle (mainly a-1) becoming smaller preb-1 and a-4 HDL particles. This change was documented by 2D electrophoresis in this model, the concentrations of small HDL particles increase while the concentrations of large HDL particles decrease. We have also investigated the effects of niacin on HDL in subjects with metabolic syndrome or atherogenic dyslipidemia, manifested by high TG and low HDL-C concentrations. Niacin significantly decreases TG, LDL-C, and apoB levels, increasing HDL-C and apoA-I levels in both monkeys and humans. The increase in HDL-C and apoA-I is more pronounced than that observed with statin treatment. The reason is that stains act only to reduce the HDL fractional catabolic rate (by decreasing TG and CETP activity), whereas niacin acts in a dual fashion. In monkeys, niacin increases HDL formation by increasing apoA-I production and s HDL catabolism, possibly by a mechanism similar to statins. Niacin increases ABCA1 expression in the liver (Asztalos, Lamon-Fava, unpublished data, 2010). As a result, we believe HDL metabolism is not only enhanced by more apoA-I production and more ABCA1 expression but also by reduced clearance of HDL particles. In our opinion, this is why niacin is able to increase HDL-C level
409 and the large athero-protective a-1 HDL particle concentration more than any other drug. In contrast to the present paradigm, niacin significantly enhances the functional clearance of apoB100 in TG-rich lipoproteins. Niacin markedly increases ABCA1 and moderately increases apoA-1 gene expression in the liver, resulting in the formation of very large HDL particles. Niacin also has beneficial effects on fat by markedly increasing levels of anti-inflammatory adiponectin levels. It is unlikely that these effects are through the niacin receptor, since the liver lacks this receptor. Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol (ARBITER-2) was the first randomized clinical trial designed to test the hypothesis that the increase of HDL via the addition of niacin to statins could significantly decrease the growth of coronary atherosclerotic lesions more than further LDL reduction with ezetimibe added to the statin therapy.23 However, the trial was prematurely stopped without the investigators ever concluding whether the HDL increase or the LDL decrease was most strongly associated with the beneficial effects of this drug. The data indicate that niacin increased HDL-C significantly more but decreased levels of LDL-C less than ezetimibe while decreasing TG more than ezetimibe. There was a very small but statistically significant reduction in the percent stenosis with the niacin combination. However, because the lack of definitive answers from ARBITER 6, the question of benefit from increasing HDL remains.
References 1. Castelli WP, Anderson K, Wilson PW, et al. Lipids and risk for coronary heart disease: the Framingham Study. Ann Epidemiol. 1992;2:23–28. 2. Gordon DJ, Rifkind BM. High-density lipoprotein—the clinical implications of recent studies. N Engl J Med. 1989;321:1311–1316. 3. Wilson PW, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation. 1998;97:1837–1847. 4. Genest JJ Jr., Martin-Munley SS, McNamara JR, et al. Familial lipoprotein disorders in patients with premature coronary artery disease. Circulation. 1992;85:2025–2033. 5. Nicholls SJ, Tuzcu EM, Sipahi I, et al. Statins, high-density lipoprotein cholesterol, and regression of coronary atherosclerosis. JAMA. 2007;297:499–508. 6. Brown BG, Stukovsky KH, Zhao XQ. Simultaneous low-density lipoprotein-C lowering and high-density lipoprotein-C elevation for optimum cardiovascular disease prevention with various drug classes, and their combinations: a meta-analysis of 23 randomized lipid trials. Curr Opin Lipidol. 2006;17:631–636. 7. Vaisar T, Pennathur S, Green PS, et al. Shotgun proteomics implicates protease inhibition and complement activation in the antiinflammatory properties of HDL. J Clin Invest. 2007;117:746–756. 8. Asztalos BF, Sloop CH, Wong L, Roheim PS. Two-dimensional electrophoresis of plasma lipoproteins: recognition of new apo A-I-containing subpopulations. Biochim Biophys Acta. 1993;1169:291–300. 9. Santos RD, Schaefer EJ, Asztalos BF, et al. Characterization of high density lipoprotein particles in familial apolipoprotein A-I deficiency. J Lipid Res. 2008;49:349–357. 10. Asztalos BF, Brousseau ME, McNamara JR, Horvath KV, Roheim PS, Schaefer EJ. Subpopulations of high density lipoproteins in
410
11.
12.
13.
14.
15.
16.
homozygous and heterozygous Tangier disease. Atherosclerosis. 2001; 156:217–225. Asztalos BF, Schaefer EJ, Horvath KV, et al. Role of LCAT in HDL remodeling: investigation of LCAT deficiency states. J Lipid Res. 2007;48:592–599. Asztalos BF, Horvath KV, Kajinami K, et al. Apolipoprotein composition of HDL in cholesteryl ester transfer protein deficiency. J Lipid Res. 2004;45:448–455. Asztalos BF, de la Llera-Moya M, Dallal GE, Horvath KV, Schaefer EJ, Rothblat GH. Differential effects of HDL subpopulations on cellular ABCA1- and SR-BI2mediated cholesterol efflux. J Lipid Res. 2005;46:2246–2253. Asztalos BF, Cupples LA, Demissie S, et al. High-density lipoprotein subpopulation profile and coronary heart disease prevalence in male participants of the Framingham Offspring Study. Arterioscler Thromb Vasc Biol. 2004;24:2181–2187. Asztalos BF, Collins D, Cupples LA, et al. Value of high-density lipoprotein (HDL) subpopulations in predicting recurrent cardiovascular events in the Veterans Affairs HDL Intervention Trial. Arterioscler Thromb Vasc Biol. 2005;25:2185–2191. Asztalos BF, Collins D, Horvath KV, Bloomfield HE, Robins SJ, Schaefer EJ. Relation of gemfibrozil treatment and high-density lipoprotein subpopulation profile with cardiovascular events in the
Journal of Clinical Lipidology, Vol 4, No 5, October 2010
17.
18.
19. 20.
21.
22.
23.
Veterans Affairs High-Density Lipoprotein Intervention Trial. Metabolism. 2008;57:77–83. Asztalos BF, Horvath KV, McNamara JR, Roheim PS, Rubinstein JJ, Schaefer EJ. Comparing the effects of five different statins on the HDL subpopulation profiles of coronary heart disease patients. Atherosclerosis. 2002;164:361–369. Asztalos BF, Le Maulf F, Dallal GE, et al. Comparison of the effects of high doses of rosuvastatin versus atorvastatin on the subpopulations of high-density lipoproteins. Am J Cardiol. 2007;99:681–685. Altschul R, Hoffer A, Stephen JD. Influence of nicotinic acid on serum cholesterol in man. Arch Biochem. 1955;54:558–569. Brown BG, Zhao XQ, Chait A, et al. Simvastatin and niacin, antioxidant vitamins, or the combination for the prevention of coronary disease. N Engl J Med. 2001;345:1583–1592. Asztalos BF, Batista M, Horvath KV, et al. Change in alpha1 HDL concentration predicts progression in coronary artery stenosis. Arterioscler Thromb Vasc Biol. 2003;23:847–852. Nissen SE, Tsunoda T, Tuzcu EM, et al. Effect of recombinant ApoA-I Milano on coronary atherosclerosis in patients with acute coronary syndromes: a randomized controlled trial. JAMA. 2003;290:2292–2300. Taylor AJ, Villines TC, Stanek EJ, et al. Extended-release niacin or ezetimibe and carotid intima-media thickness. N Engl J Med. 2009; 361:2113–2122.
NLA Symposium on High Density Lipoproteins
High-density lipoprotein particles, coronary heart disease, and niacin Bela F. Asztalos, PhD* Lipid Metabolism Laboratory, Tufts University, 711 Washington Street, Boston, MA 02111, USA KEYWORDS: High density lipoprotein; HDL particles; HDL-C; CHD; Niacin; HDL raising
Abstract: The use of statins in patients with high risk for cardiovascular disease (CVD) has resulted in a 30–40% decrease in clinical events in the last couple of decades. However, despite of a marked reduction (up to 60%) in LDL-C, about 30% of patients continue to have CVD events. This high residual risk in statin-treated patients initiated the search for new ways to reduce CVD risk. HDL is the next logical target. Epidemiological and cross-sectional studies identified low HDL-C level as an independent risk for CVD.1,2 Based on the Framingham Heart Study data, HDL-C ,35 mg/dl was established an independent risk factor and HDL-C >60 mg/dl as protective.3 Presently the cut point is ,40 mg/dl for men and ,50 mg/dl for women. Ó 2010 National Lipid Association. All rights reserved.
In clinical trials, the use of statins in patients with high risk for cardiovascular disease (CVD) has resulted in a 25% to 40% decrease in major clinical events. However, despite a marked reduction (up to 60%) in LDL-C, approximately 50% (or more) of patients continue to have CVD events. This high residual risk in statin-treated patients initiated the search for new ways to reduce CVD risk. High-density lipoprotein (HDL) is the next logical target. The authors of epidemiological and cross-sectional studies identified low HDL-C level as an independent risk factor for CVD.1,2 On the basis of the Framingham Heart Study data, HDL-C less than 35 mg/dL was established as an independent risk factor and HDL-C greater than 60 mg/dl was defined as protective3 and considered to reduce the global risk estimate. Presently the cut point for defining HDL-C as low is less than 40 mg/dL for men and less than 50 mg/dL for women.
* Corresponding author. E-mail address: [email protected] Submitted August 3, 2010. Accepted for publication August 7, 2010.
Whether an HDL-C level greater than 60 mg/dL can be regarded as protective in all patients has recently been debated. Some investigators have provided evidence for the existence of dysfunctional HDL in patients with vascular disease; however, the term ‘‘dysfunctional HDL’’ has not been well and consistently defined. Dysfunctional HDL is generally thought of as not protective against CVD even if HDL-C is present in high levels. It is assumed that dysfunctional HDL has altered composition and altered functions. Studies in mice have demonstrated that decreased scavenger receptor B1 (SR-B1) reduces clearance of HDL-C and results in greater plasma concentrations. HDL particles become larger with more cholesterol content, but the cholesterol transport into the liver and into the bile is decreased; cholesterol transport from peripheral tissues is impeded and the resulting fails to protect from developing arteriosclerosis. Low HDL-C levels can be inherited dyslipidemia expressed as increased concentrations of the triglyceride-rich lipoproteins, very low density lipoproteins (VLDL), and chylomicrons. This form of dyslipidemia is responsible for 15% or more of premature CHD cases. The specific disorders
1933-2874/$ - see front matter Ó 2010 National Lipid Association. All rights reserved. doi:10.1016/j.jacl.2010.08.012
406
Figure 1 Reduction of CVD events (%) in patients treated with an array of drugs versus placebo treatment.
of familial combined hyperlipidemia (FCHL)4 have been reported to be responsible for 14% of premature CHD cases, and fhmya be responsible for approximately 4% of premature CHD cases. Factors associated with sedentary lifestyle, obesity, smoking, and poor control of diabetes mellitus contribute to both increased triglycerides and low HDL-C. In a population of 1439 patients pooled from four clinical trials of statin therapy, LDL-C was reduced from a mean of 124 to 88 mg/dL and HDL-C increased from a mean of 42.5 to 45.1 mg/dL. These statin-induced changes in both LDL-C and HDL-C were found to be significant and independent correlates of atheroma regression assessed with intravascular ultrasound.5 In another meta analysis in which researchers used data from several drug intervention trials, reduction in 1-year event rates was the greatest in subjects with the greatest reduction in LDL-C and the greatest increase in HDL-C (Fig. 1).6 HDL is the smallest and densest of all plasma lipoproteins. It consists of a number of distinct particles that vary in size, shape, density, surface charge, composition, and physiological functions. HDL has several potentially antiatherogenic properties. The best known of these is the ability of HDL to remove cholesterol from peripheral cells, including macrophages in the artery wall and transporting cholesterol to the liver for biliary excretion. Other HDL functions include inhibition of LDL oxidation, improvement in endothelial function, inhibition of the binding of monocytes to the endothelium, promotion of endothelial repair, stabilization of nitric oxide synthesis, as well as antithrombotic and anti-inflammatory properties. HDL is also a key player in innate immunity. At least 49 proteins have been identified in HDL by mass spectrometry, some related to lipid metabolism, others to complement regulation or to acute-phase response, and a few related to proteinase inhibitors.7 Several cell surface-bound and soluble plasma proteins influence the charge, size, shape, and composition of HDL. These factors include ATP binding cassette transporter A1 (ABCA1), ABCG1, lecithin:cholesterol acyltransferase (LCAT), cholesteryl ester transfer protein (CETP), phospholipid transfer protein, lipoprotein lipase, hepatic lipase (HL), endothelial lipase, secretory phospholipase A2, and SR-B1. The various HDL particles in human plasma are classified by the use of many different parameters dependent on the
Journal of Clinical Lipidology, Vol 4, No 5, October 2010 separation method used. In this article, we discuss only the apoA-I2containing HDL particles, separated by twodimensional (2D), nondenaturing gel electrophoresis, detected by immunoblotting for apoA-I, quantified by image-analysis (Fig. 2). With this method, 12 distinct apoA-I–containing HDL subclasses are separated by electrophoretic mobility in the first dimension (pre-b-, a-, and pre-a-mobility particles) and by size in the second dimension (5.4211.7 nm).8 In addition to apoA-I, these HDL particles contain other lipoproteins and plasma proteins and a variety of lipids depending on their size (Fig. 3). Ten of the 12 HDL subpopulations contain no apoA-II (LpA-I particles) whereas two of them, a-2 and a-3, contain both apoA-I and apoA-II (LPA-I:A-II particles). By using the nondenaturing 2D method, we have mapped the sequence and mechanisms of many of the steps in HDL metabolism/remodeling in humans by collecting samples from patients with mutations in genes influencing HDL metabolism/remodeling. We have collected plasma samples from subjects carrying one (heterozygous) and two (homozygous) defective alleles encoding: apoA-I,9 ABCA1,10 LCAT,11 CETP,12 HL, and lipoprotein lipase. In addition, we have performed cell-cholesterol efflux assays via the ABCA1, ABCG1, and SR-B1 pathways.13 On the basis of the results from these studies, we have built a working hypothesis of HDL metabolism/remodeling illustrated in Figure 4. ApoA-I is synthesized in the liver and in the intestine and secreted as a lipid-free protein. Lipid-free apoA-I picks up some phospholipids and forms preb-1 particles (precursor HDL particles) in the circulation. Preb-1 HDL is a homodimer apoA-I particle. Preb-1 particles interact with ABCA1 on the surface of liver and peripheral cells and pick up more phospholipids and cell-derived cholesterol (unesterified cholesterol) and are transformed into the smallest a-mobility particle, a-4. LCAT can act on a-4 particles, transforming free cholesterol into cholesteryl ester, acquiring other neutral lipids and proteins, and growing in size. The two largest HDL particles are a-1 and a-2. These are the major HDL subclasses in control subjects, but their concentrations in CHD patients are low (Fig. 2). These large HDL particles can deliver cholesterol directly to the bile via the SR-B1 transporter expressed on the service of hepatocytes. Alternatively, cholesterol can be transferred from these large HDL particles to TG-rich lipoproteins by CETP. A high concentration of TG-rich lipoproteins— substrate for CETP—results in cholesterol removal from a-1 HDL particles in exchange for TG. TG-enriched a-1 HDL substrate for progressive lipolysis by HL are converted into smaller particles. For this reason, subjects with increased TG levels often have reduced HDL-C and small HDL particles. Patients with CHD often have decreased concentrations of HDL-C and fewer large a-1 HDL but increased concentrations of small a-3 and preb-1 particles (Fig. 3). In the Framingham Offspring Study, we found that in men, with and without CHD, for every 1 mg/dL decrease in a-1 HDL level, there was a 26% increase for CHD risk.14
Asztalos
HDL particles, coronary heart disease, and niacin
407
Figure 2 Apo-A-I-containing HDL subpopulations of a control and a CHD patient. Panel (a): apoA-I-containing HDL subpopulations of a normolipidemic male subject, separated by 2d electrophoresis. The inserted rectangle shows separation in the 1st dimension on agarose gel by charge into preb-, a-, and prea-mobility subfractions. In the 2nd dimension, HDL particles were further separated by size on concave-gradient polyacrylamide gels under non-denaturing conditions. Molecular weight/size standard proteins can be seen on the left side of panel (a). The asterisk represents the position of endogenous albumin marking the a-front. Panel (b): apoA-I-containing HDL subclasses of a male CHD patient. Panel (c): schematic representation of HDL particles containing only apoA-I (LpA-I) light grey, and HDL particles containing both apoA-I and apoA-II (LpA-I:A-II) dark grey.
We also found that HDL particles were significantly better predictors for CHD events than HDL-C itself. Earlier it was shown that every 1 mg/dL increase in HDL-C levels was associated with approximately a 2% decrease in CVD risk. In the Veterans Affairs HDL Intervention Trial (VA-HIT), low levels of a-1 and a-2 HDL and high level of preb-1 HDL particles predicted recurrence of CVD events versus no recurrence in men who were selected with low HDL-C level and established CVD.15 A low a-1 HDL level was the most significant predictor for recurrence of CVD events.
Ways to increase HDL-C levels
Figure 3 Working theory of the physicochemical structure and size of the major apoA-I-containing HDL subpopulations. Light yellow cylinders represent apoA-I and darker yellow cylinders represent apoA-II molecules. Blue, green, and purple spheres represent phospholipid (PL), free cholesterol (FC), and TG respectively. The chemical composition of preb-2 has not been investigated yet; therefore it is labeled as solid black.
Currently, there are three major drug classes—fibrates, statins, and nicotinic acid—that are used to treat dyslipidemia or increase HDL-C level. Fibrates increase HDL-C level by approximately 10%. This is variable with no increase in some studies to more than 20% in those with very high triglyceride concentrations. Despite increasing HDL-C, fibrates slightly decrease the concentration of large a-1 HDL and slightly increase the concentration of small a-3 HDL. As shown previously, concentrations of the large HDL particles are inversely associated while those of the small HDL particles which are positively associated with CVD. Therefore, the effects of fibrates on HDL subclasses may be unfavorable.16 Considering this result and other uncertainties about fibrates, it is not understood how this drug class protects against CVD. There is the possibility that fibrates protect from CVD not through increasing HDL levels but rather through other mechanisms. Statins increase HDL-C level up to 10% and increase large a-1 HDL particles up to 36%. Moreover, statins decrease preb-1 concentrations up to 40%. The efficacy of
408
Journal of Clinical Lipidology, Vol 4, No 5, October 2010
Figure 4 Illustration of HDL metabolism based on research in patients with rare inborn errors of HDL metabolism. Summary of findings: 1) apoA-I is not absolutely necessary for the formation of lipidated HDL particles; 2) preb-1, a very small discoidal precursor HDL particle, interacts with ABCA1 to promote cellular cholesterol and phospholipid efflux and in doing so is converted into small, LpA-I a-4 HDL particles; 4) a-4 is acted upon by LCAT which converts its free cholesterol to cholesteryl ester. Then a-4 serves as an acceptor for apoA-II (secreted in already lipidated form by the liver), eventually forming the smaller LpA-I:A-II HDL subclass (a-3); 5) the small LpA-I a-4 particles can pass through a number of other steps in order to form large spherical a-1 HDL (also LpA-I), which can effectively deliver cholesterol to the liver. Data collected from subjects with defects in the genes encoding CETP, LPL, and HL indicated that HDL maturation is not a simple linear size-increase process. These steps are explained in more detail in “High Density Lipoproteins, Dyslipidemia, and Coronary Heart Disease”, Springer 2010, edited by EJ Schaefer.
the different statins regarding beneficial HDL particle alteration in decreasing order is rosuvastatin, atorvastatin, simvastatin, pravastatin, and lovastatin.17,18 Statin-mediated changes in HDL are associated with decreased CETP activity but not with changes in apoA-I kinetics. Niacin increases the concentration of HDL-C up to 25% and that of the large a-1 HDL particles up to two-fold while decreasing the concentration of the small preb-1 particles. The effects of niacin on HDL are similar to those of statins but produce greater and more consistent changes. Niacinmediated changes in HDL are also associated with increases in apoA-I production and ABCA1 expression in the liver (unpublished data). Niacin’s effects on serum cholesterol were published first by Rudolph Altschul19 in 1955. Niacin has been used clinically to lower cholesterol for more than five decades, but we have no full explanation of its mechanism of action. We need more data on how niacin works in humans and how it protects against CVD. The authors of placebo-controlled studies show that niacin increases HDL-C, decreases LDL-C and TG levels, as well as lipoprotein(a) (Lp(a)) levels, in a dose-responsive pattern. Elevated Lp(a) is associated with increased risk, and niacin is the only commercially available lipid lowering medication that decreases Lp(a) concentrations.
We have studied the effects of niacin in combination with simvastatin in the HDL-Atherosclerosis Treatment Study (HATS) population. Patients received niacin and simvastatin, niacin, simvastatin and antioxidant vitamins, antioxidant vitamins alone, or placebo. Coronary artery atherosclerosis was assessed quantitative angiography, and lipoproteins were analyzed in samples both on and off treatment.20 A significant association between change in coronary artery stenosis and change in a-1 HDL concentrations was observed in the entire cohort including all treatment groups. In the first tertile of a-1 HDL concentrations, the mean particle level decreased 15% and stenosis increased 2.1%. This group of subjects was primarily composed of those taking antioxidants and placebo. In the third tertile with the highest mean a-1 HDL particle level, having increased 157%, there was no significant change in stenosis suggesting stabilization of the disease. This group consisted primarily of niacin and simvastatin. Many patients experienced regression of the disease. This was the first study to show that a known drug-combination can cause reductions in plaque volume.21 Increasing HDL particles appeared to be confirmed in humans by a study using the direct infusion of A-I Milano.22 In conclusion, niacin treatment in patients with combined dyslipidemia significantly decreases TG and LDL-C
Asztalos
HDL particles, coronary heart disease, and niacin
concentrations and significantly increases HDL-C and apoA-I levels. Moreover, with niacin-simvastatin combination therapy, the level of the large protective a-1 HDL particles, believed to be linked to regression of coronary stenosis, is doubled. We hypothesize that the increase of apoA-I mass during niacin treatment is caused by a significant increase in the production of apoA-I. This is one of the differences between niacin and statins, because statins decrease the clearance while niacin increases the production of HDL particles. We have found that niacin has no effect on apoA-II metabolism, and the percent change in apoA-I production rate is positively associated with the percent change of HDL-C concentrations. Recently, we developed a monkey model to study the mechanisms of the protective effects of niacin in vivo. Monkeys fed an atherogenic diet (high in saturated-fat and cholesterol) developed a very similar lipoprotein profile as humans on a similar diet, with significant increases in TG, apoB, and cholesterol in apoB-containing particles. Similar to humans, HDL-C concentration significantly increased with no significant increase in apoA-I concentration. This usually signifies an increase in the concentration of the large lipid-rich HDL particles. In contrast, a significant increase in the concentration of small HDL particles and a significant decrease in the concentration of the largest HDL particles were observed. This change in the HDL particle subpopulation profile indicates a less favorable risk status, even though HDL-C is increased. We hypothesize that an atherogenic diet increases plasma triglyceride and TG-rich lipoprotein concentrations, which in turn increases CETP activity and, very likely, hepatic lipase activity on HDL (see Fig. 4). Augmented CETP activity increases TG concentrations in the large HDL particles. TG-enriched HDL particles provide good substrate for HL-dependent lipolysis. The increased CETP and hepatic lipase activities promote large HDL particle (mainly a-1) becoming smaller preb-1 and a-4 HDL particles. This change was documented by 2D electrophoresis in this model, the concentrations of small HDL particles increase while the concentrations of large HDL particles decrease. We have also investigated the effects of niacin on HDL in subjects with metabolic syndrome or atherogenic dyslipidemia, manifested by high TG and low HDL-C concentrations. Niacin significantly decreases TG, LDL-C, and apoB levels, increasing HDL-C and apoA-I levels in both monkeys and humans. The increase in HDL-C and apoA-I is more pronounced than that observed with statin treatment. The reason is that stains act only to reduce the HDL fractional catabolic rate (by decreasing TG and CETP activity), whereas niacin acts in a dual fashion. In monkeys, niacin increases HDL formation by increasing apoA-I production and s HDL catabolism, possibly by a mechanism similar to statins. Niacin increases ABCA1 expression in the liver (Asztalos, Lamon-Fava, unpublished data, 2010). As a result, we believe HDL metabolism is not only enhanced by more apoA-I production and more ABCA1 expression but also by reduced clearance of HDL particles. In our opinion, this is why niacin is able to increase HDL-C level
409 and the large athero-protective a-1 HDL particle concentration more than any other drug. In contrast to the present paradigm, niacin significantly enhances the functional clearance of apoB100 in TG-rich lipoproteins. Niacin markedly increases ABCA1 and moderately increases apoA-1 gene expression in the liver, resulting in the formation of very large HDL particles. Niacin also has beneficial effects on fat by markedly increasing levels of anti-inflammatory adiponectin levels. It is unlikely that these effects are through the niacin receptor, since the liver lacks this receptor. Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol (ARBITER-2) was the first randomized clinical trial designed to test the hypothesis that the increase of HDL via the addition of niacin to statins could significantly decrease the growth of coronary atherosclerotic lesions more than further LDL reduction with ezetimibe added to the statin therapy.23 However, the trial was prematurely stopped without the investigators ever concluding whether the HDL increase or the LDL decrease was most strongly associated with the beneficial effects of this drug. The data indicate that niacin increased HDL-C significantly more but decreased levels of LDL-C less than ezetimibe while decreasing TG more than ezetimibe. There was a very small but statistically significant reduction in the percent stenosis with the niacin combination. However, because the lack of definitive answers from ARBITER 6, the question of benefit from increasing HDL remains.
References 1. Castelli WP, Anderson K, Wilson PW, et al. Lipids and risk for coronary heart disease: the Framingham Study. Ann Epidemiol. 1992;2:23–28. 2. Gordon DJ, Rifkind BM. High-density lipoprotein—the clinical implications of recent studies. N Engl J Med. 1989;321:1311–1316. 3. Wilson PW, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation. 1998;97:1837–1847. 4. Genest JJ Jr., Martin-Munley SS, McNamara JR, et al. Familial lipoprotein disorders in patients with premature coronary artery disease. Circulation. 1992;85:2025–2033. 5. Nicholls SJ, Tuzcu EM, Sipahi I, et al. Statins, high-density lipoprotein cholesterol, and regression of coronary atherosclerosis. JAMA. 2007;297:499–508. 6. Brown BG, Stukovsky KH, Zhao XQ. Simultaneous low-density lipoprotein-C lowering and high-density lipoprotein-C elevation for optimum cardiovascular disease prevention with various drug classes, and their combinations: a meta-analysis of 23 randomized lipid trials. Curr Opin Lipidol. 2006;17:631–636. 7. Vaisar T, Pennathur S, Green PS, et al. Shotgun proteomics implicates protease inhibition and complement activation in the antiinflammatory properties of HDL. J Clin Invest. 2007;117:746–756. 8. Asztalos BF, Sloop CH, Wong L, Roheim PS. Two-dimensional electrophoresis of plasma lipoproteins: recognition of new apo A-I-containing subpopulations. Biochim Biophys Acta. 1993;1169:291–300. 9. Santos RD, Schaefer EJ, Asztalos BF, et al. Characterization of high density lipoprotein particles in familial apolipoprotein A-I deficiency. J Lipid Res. 2008;49:349–357. 10. Asztalos BF, Brousseau ME, McNamara JR, Horvath KV, Roheim PS, Schaefer EJ. Subpopulations of high density lipoproteins in
410
11.
12.
13.
14.
15.
16.
homozygous and heterozygous Tangier disease. Atherosclerosis. 2001; 156:217–225. Asztalos BF, Schaefer EJ, Horvath KV, et al. Role of LCAT in HDL remodeling: investigation of LCAT deficiency states. J Lipid Res. 2007;48:592–599. Asztalos BF, Horvath KV, Kajinami K, et al. Apolipoprotein composition of HDL in cholesteryl ester transfer protein deficiency. J Lipid Res. 2004;45:448–455. Asztalos BF, de la Llera-Moya M, Dallal GE, Horvath KV, Schaefer EJ, Rothblat GH. Differential effects of HDL subpopulations on cellular ABCA1- and SR-BI2mediated cholesterol efflux. J Lipid Res. 2005;46:2246–2253. Asztalos BF, Cupples LA, Demissie S, et al. High-density lipoprotein subpopulation profile and coronary heart disease prevalence in male participants of the Framingham Offspring Study. Arterioscler Thromb Vasc Biol. 2004;24:2181–2187. Asztalos BF, Collins D, Cupples LA, et al. Value of high-density lipoprotein (HDL) subpopulations in predicting recurrent cardiovascular events in the Veterans Affairs HDL Intervention Trial. Arterioscler Thromb Vasc Biol. 2005;25:2185–2191. Asztalos BF, Collins D, Horvath KV, Bloomfield HE, Robins SJ, Schaefer EJ. Relation of gemfibrozil treatment and high-density lipoprotein subpopulation profile with cardiovascular events in the
Journal of Clinical Lipidology, Vol 4, No 5, October 2010
17.
18.
19. 20.
21.
22.
23.
Veterans Affairs High-Density Lipoprotein Intervention Trial. Metabolism. 2008;57:77–83. Asztalos BF, Horvath KV, McNamara JR, Roheim PS, Rubinstein JJ, Schaefer EJ. Comparing the effects of five different statins on the HDL subpopulation profiles of coronary heart disease patients. Atherosclerosis. 2002;164:361–369. Asztalos BF, Le Maulf F, Dallal GE, et al. Comparison of the effects of high doses of rosuvastatin versus atorvastatin on the subpopulations of high-density lipoproteins. Am J Cardiol. 2007;99:681–685. Altschul R, Hoffer A, Stephen JD. Influence of nicotinic acid on serum cholesterol in man. Arch Biochem. 1955;54:558–569. Brown BG, Zhao XQ, Chait A, et al. Simvastatin and niacin, antioxidant vitamins, or the combination for the prevention of coronary disease. N Engl J Med. 2001;345:1583–1592. Asztalos BF, Batista M, Horvath KV, et al. Change in alpha1 HDL concentration predicts progression in coronary artery stenosis. Arterioscler Thromb Vasc Biol. 2003;23:847–852. Nissen SE, Tsunoda T, Tuzcu EM, et al. Effect of recombinant ApoA-I Milano on coronary atherosclerosis in patients with acute coronary syndromes: a randomized controlled trial. JAMA. 2003;290:2292–2300. Taylor AJ, Villines TC, Stanek EJ, et al. Extended-release niacin or ezetimibe and carotid intima-media thickness. N Engl J Med. 2009; 361:2113–2122.