The apolipoprotein story

Atherosclerosis Supplements 7 (2006) 23 27 The apolipoprotein story Frank M. Sacks a,b, a Department of Nutrition, Harvard School of Public Health, Boston, MA, USA b Department of Medicine, Harvard Medical School and Brigham & Women s Hospital, Boston, MA, USA Abstract The importance of different lipid and lipoprotein measurements, including LDL particle concentration and levels of apolipoproteins and triglycerides, in the prediction of future cardiac events continues to be debated. In summary, apo B is a strong, independent predictor of initial and recurrent coronary events, even during statin treatment, and recent studies show its predictive superiority over LDL and non- HDL cholesterol. Importantly, determination of apo B levels is unaffected in a non-fasted or hypertriglyceridemic state and is not derived from other measurements. Thus, clear advantages exist for using apo B as a predictor of CHD. Likewise, triglycerides and triglyceride-rich lipoproteins are also strong, independent predictors of coronary events (initial and recurrent) during statin treatment. Triglycerides or especially triglyceride-rich lipoproteins with apo C-III may provide additional information to apo B. Apo C-III not only impairs lipoprotein metabolism but also stimulates directly the vascular inflammatory process. In contrast, evidence from large epidemiological studies is coalescing toward the view that small LDL size is more of a marker of these atherogenic triglyceride-rich lipoproteins than an independent predictor of CHD. Clearly, there remains considerable interest, both in terms of research and clinical practice, in the role of apo B, triglycerides, and specific apo B-containing lipoprotein particles as independent predictors of CHD risk as well as their potential to improve risk prediction and response to lipid-modifying treatment. 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Apolipoproteins; Lipoproteins; Triglycerides; Coronary heart disease; Epidemiology; HMG CoA reductase inhibitors ( statins ); Apolipoprotein B; Apolipoprotein C-III In this brief review, I compare the relationship of the lipid and protein components of plasma lipoproteins with coronary heart disease (CHD). It is well established that plasma lipoprotein concentrations, however, they are measured, strongly predict CHD. Lipid treatment guidelines all use target lipoprotein cholesterol levels as the basis for therapeutic intervention. For example, the National Cholesterol Education Program s third Adult Treatment Panel recommends that, in adults aged 20 years or older, a fasting lipoprotein profile (i.e., total cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, and triglyceride) should be obtained every 5 years [1]. If a person s total or LDL cholesterol is too high or their HDL cholesterol is too low and the presence of other major risk factors or CHD risk equivalents, such as atherosclerotic disease or diabetes, are determined, then they may require intensive Tel.: +1 617 4362 1420; fax: +1 617 432 3101. E-mail address: fsacks@hsph.harvard.edu. lifestyle modifications or initiation of lipid-modifying therapies to reduce the long-term risk of CHD. Furthermore, evidence is building toward the concept that apolipoproteins are linked to atherosclerosis and coronary events more strongly than are plasma lipoproteins. This review will summarize the evidence for apolipoproteins as risk factors for CHD, although I caution the reader that the issue is not fully settled regarding which are the most specific apolipoprotein or lipoprotein lipid risk factors. Lipoproteins are spherical particles that transport variable amounts of cholesterol and triglyceride in the circulation. Lipoprotein particles are divided into several major groups, including LDL and HDL cholesterol, very-lowdensity lipoprotein (VLDL), and chylomicrons, which are rich in recently absorbed dietary triglycerides. Apolipoprotein B (apo B) is a protein found on the surface of LDL, VLDL, and chylomicron particles, and one apo B molecule is found on each particle. For this reason, these lipoproteins are often called apo B-containing lipoproteins. Apolipopro- 1567-5688/$ see front matter 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosissup.2006.05.004

24 F.M. Sacks / Atherosclerosis Supplements 7 (2006) 23 27 tein A-I (apo A-I) is the protein present on all HDL particles, which are sometimes referred to as apo A-I-containing lipoproteins. While a very small amount of apo A-I may be present on apo B-containing lipoproteins, the significance of which is unknown, the converse is not true apo B is never a part of HDL. It is now well known that the LDL cholesterol level is strongly and continuously related across its entire range to atherosclerosis and coronary artery disease. In contrast, high levels of HDL cholesterol have been shown to compensate for some of the harmful effects of LDL cholesterol. Many studies have demonstrated that plasma LDL and HDL cholesterol and plasma total triglycerides are independent predictors of cardiovascular disease [1]. Conventional lipid tests determine the amount of cholesterol or triglyceride carried by all particles within a lipoprotein class or in total plasma. In this way, cholesterol and triglyceride may be viewed as surrogate markers for the lipoprotein particles that carry them. Recently, there has been controversy in the field of atherosclerosis regarding which is more important the cholesterol that the LDL particles carry or the actual LDL particle concentration measured and expressed as apo B. Arguments in favor of cholesterol being the more important factor include the fact that cholesterol ester is the major lipid component responsible for the development of atherosclerosis. Cholesterol is present in foam cells, and it is also present in the extracellular matrix of plaques. Cholesterol esters are susceptible to oxidation, which can potentially make them more atherogenic [2]. There is also a strong case for the particle concentrations of LDL and VLDL being the more important parameters. These parameters are measured by the apo B concentration a measure of the atherogenic particle concentration to which the vessel wall is exposed. Moreover, apo B binds a particle to proteoglycans both on endothelial cells and in the matrix of the plaque. There is a direct relationship between plasma apo B concentrations and how many particles get into a plaque and are retained in a plaque [3]. Macrophages become foam cells by taking up whole particles of native or oxidized apo B lipoproteins by several mechanisms [2,4]. The relative importance of LDL cholesterol and apo B lipoprotein particle concentrations may be influenced by how each component is measured. Whereas apo B is measured directly usually by immunonephelometry using plasma collected in a fed or fasted state [5,6], LDL cholesterol is usually calculated indirectly as the remainder of fasting plasma total cholesterol after HDL cholesterol and VLDL cholesterol are subtracted [7]. The VLDL cholesterol is estimated by dividing the plasma total triglycerides by five, and is accurate only when the triglycerides concentration is <400 mg/dl. In contrast to the limitations of LDL cholesterol measurements, plasma total apo B not only is a good approximation of LDL particle concentration, but also of total atherogenic lipoprotein concentrations. Although in most people, more than 90% of plasma apo B is bound to LDL and the remainder to VLDL, people with high triglyceride levels often have a high VLDL concentration and low LDL concentration (or if they have combined hyperlipidemia, elevated triglycerides and cholesterol). Thus, apo B measurement may be a superior indicator of global atherogenic lipoprotein risk than either LDL cholesterol or total triglyceride concentrations. Several studies have demonstrated that apolipoprotein particle concentrations are related to CHD. The Apolipoproteinrelated Mortality Risk Study (AMORIS) investigated whether apo B and apo A-I are better predictors of risk of fatal myocardial infarction than total cholesterol and LDL cholesterol [8]. It also aimed to establish whether apo B and apo A-I add further information about risk of fatal myocardial infarction to that obtained with total cholesterol, LDL cholesterol, and triglycerides. AMORIS was a large study with 175,553 Swedish men and women enrolled mainly from cholesterol screening programs. The relation between death from acute myocardial infarction and initial values for apo B, apo A-I, and other lipids was examined. In univariate and multivariate analyses adjusted for age, total cholesterol and triglycerides, both apo B and the apo B/apo A-I ratio were strongly and positively related to increased risk of fatal myocardial infarction in men and women. Apo A-I was inversely related to risk of myocardial infarction. In multivariate analysis, apo B was a stronger predictor of risk than LDL cholesterol in both sexes. The study concluded that apo B, apo A-I, and their ratio should be regarded as highly predictive in evaluation of cardiac risk (Fig. 1). In older patients, particularly older women, there is an even stronger prediction. So unlike the conventionally measured cholesterol, which has reduced predictive ability in the elderly, the measurement of apolipoproteins shows a clear advantage in the elderly. The investigators of the AMORIS study acknowledged that LDL cholesterol is the recommended target lipid variable in international guidelines for treatment. However, they concluded that this new information shows the importance of apo B and apo A-I as risk predictors, which may be of great value for an individual. Therefore, the authors suggest Fig. 1. Independent role for apo B and apo A-I in predicting fatal myocardial infarction in Scandinavian men [8]. Reprinted from Lancet, 358, Walldius G et al, High apolipoprotein B, low apolipoprotein A-1, and improvement in the prediction of fatal myocardial infarction (AMORIS), 2026-33, 2001 with permission from Elsevier.

F.M. Sacks / Atherosclerosis Supplements 7 (2006) 23 27 25 Fig. 2. LDL particle concentration predicts cardiovascular disease in American women [9]. that prediction of future cardiac death could be improved by measurement of apo B, apo A-I, and the apo B/apo A-I ratio. Nuclear magnetic resonance spectroscopy measures the plasma concentration of lipids in most lipoproteins, which can be used to estimate particle concentration. In a cohort of healthy, middle-aged women enrolled in the Women s Health Study, the LDL particle concentration, more so than LDL cholesterol, independently predicted coronary events [9] (Fig. 2). This technique also measures the size of the lipoprotein particles; however, LDL particle size was not determined to be a contributor to risk beyond the LDL concentration. Another approach to measuring total atherogenic lipoproteins is to use non-hdl cholesterol. This is the cholesterol contained in VLDL and LDL, and is readily computed by subtracting HDL cholesterol from plasma total cholesterol. Thus, non-hdl cholesterol is essentially the cholesterol analogue to an apo B level as a measurement of atherogenic particle concentration. As one would expect, the concentrations of non-hdl cholesterol and apo B are very highly correlated with one another and highly predictive of coronary heart disease, with correlation coefficients for non-hdl cholesterol and apo B of 0.93 0.94, compared with 0.84 0.85 for LDL cholesterol and apo B [10]. The high correlation between non- HDL cholesterol and apo B remains unchanged after statin therapy. Thus, it is a challenge for multivariate analysis to determine which is better apo B or non-hdl cholesterol. Very recently, Pischon et al. reported in US healthcare professionals that apo B is superior to non-hdl cholesterol [11]. This is illustrated by a stratified analysis that demonstrates the sharp increase in CHD for higher apo B levels at any category of non-hdl cholesterol (Fig. 3). The converse is not true; once the apo B is determined, non-hdl cholesterol does not contribute additional information on risk. Interest has recently focused on the impact of statin treatment on these variables. In the Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS), Fig. 3. Direct comparison of apo B and non-hdl cholesterol to predict coronary heart disease in American men [11]. From Circulation 2005;112:3375 83 with permission. both at baseline and following 1 year of statin therapy, apo B and apo A-I are very strong predictors of atherosclerosis, whereas the lipid parameters are not [12]. In this study, the ratio of apo B/apo A-I is the best predictor of CHD. Even in patients treated with lovastatin, which was very effective in reducing coronary events, apo B/apo A-I ratio remains important (Fig. 4). The Long-term Intervention with Pravastatin in Ischaemic Disease (LIPID) group performed a similar analysis to that done in AFCAPS/TexCAPS and came up with a similar result [13]. At baseline, apo A-I and apo B were strong predictors of CHD. After 1 year of treatment with pravastatin, apo B and apo A-I remain very strong CHD predictors, more so than the lipid parameters. The importance of triglycerides and triglyceride-rich lipoproteins in predicting cardiac events has also been examined. High triglycerides concentrations are associated with several other important risk factors, including obesity, insulin resistance, diabetes, and other dyslipidemias, particularly a profile with low HDL cholesterol levels, elevated apo C-III, and the presence of small, dense LDL particles. This is of particular interest in light of the recent discussions on the metabolic syndrome in which these lipid abnormalities are prevalent. Although much debated, it has been clear for nearly a decade that plasma total triglycerides, the standard clinical measurement, is an independent risk factor for CHD, contributing information on risk beyond that of LDL and HDL cholesterol [14 16]. The Northwick Park Heart Study followed 2508 healthy middle-aged men for 6 years [17]. Not only did it confirm that apo B was superior to LDL cholesterol, but also that apo B and triglycerides were the best combination for predicting CHD in UK men (Fig. 5). The study concluded that there is something additionally predictive about high triglyceride levels that apo B is not capturing, and which may be related to small LDL particle size. Nonetheless, in a cohort of the Physician s Health Study, it has been reported that non-fasting triglyceride lev-

26 F.M. Sacks / Atherosclerosis Supplements 7 (2006) 23 27 Fig. 4. AFCAPS: apo B/Apo A-I is best predictor of CHD [12]. From Circulation 2000;101:477 84 with permission. Fig. 5. Apo B and TG: best combination for predicting CHD in UK men [17]. From Arterioscler Thromb Vasc Biol 2002;22:1918 23 with permission. els strongly predicted first myocardial infarction, regardless of the LDL particle size, and that LDL particle size had no effect beyond that of triglycerides [18] (Fig. 6). Moreover, in a large cohort of patients with CHD, multivariate analysis found that a larger size of LDL was predictive of recurrent Fig. 6. Non-fasting triglycerides but not small LDL size independently predicts first myocardial infarction in US men [18]. events, a relationship eliminated by statin therapy [19]. Considering all epidemiological studies on LDL size, no clear pattern that predicts CHD emerges from multiple variable analysis; in some studies small LDL size was predictive, in others the opposite occurred, and in still others, LDL size did not contribute independently to CHD risk [18]. Thus, we are left with the need to study what it is about triglycerides that affect CHD. Of the triglyceride-rich lipoproteins, those with apo C-III may be most closely linked to abnormal lipoprotein metabolism and CHD. Apo C-III retards lipoprotein metabolism in plasma by inhibiting lipoprotein lipase and blocking the removal of VLDL and LDL by LDL receptors primarily in the liver. Several studies have found that apo C-III in VLDL and LDL is associated with CHD. In one study, apo C-III in VLDL and LDL is associated with progression of coronary stenosis even in those with LDL levels that have been reduced to 82 mg/dl (0.93 mmol/l) on lovastatin 80 mg daily [20]. In the prospective CARE trial, the apo C-III concentration in VLDL and LDL was a strong predictor of coronary events, even in those treated with a statin [21]. In contrast, triglycerides were less important once these more specific measurements of triglyceride-rich lipoproteins were obtained. Finally, in patients with type 2 diabetes, the most potent lipoprotein predictor was the concentration of LDL particles with apo C-III [22]. New adverse properties are being discovered about apo C-III. Apo C-III induces monocytes to adhere to vascular endothelium, an early essential event in atherosclerosis [23]. Apo C-III causes apoptosis of pancreatic beta cells perhaps leading to type 2 diabetes [24]. Thus, it is likely that atherogenic triglyceride-rich lipoproteins may be characterised by the apo C-III content. In conclusion, the importance of different factors, including LDL particle concentration and levels of apolipoproteins and triglycerides, in the prediction of future cardiac events continues to be debated. In summary, apo B is a strong, independent predictor of initial and recurrent coronary events, even during statin treatment, and recent studies show its

F.M. Sacks / Atherosclerosis Supplements 7 (2006) 23 27 27 predictive superiority over LDL and non-hdl cholesterol. Importantly, determination of apo B levels is unaffected in a non-fasted or hypertriglyceridemic state and is not derived from other measurements. Thus, clear advantages exist for using apo B as a predictor of CHD. Likewise, triglycerides and triglyceride-rich lipoproteins are also strong, independent predictors of coronary events (initial and recurrent) during statin treatment. Triglycerides or triglyceride-rich lipoproteins with apo C-III may provide additional information to apo B. Clearly, there remains considerable interest, both in terms of research and clinical practice, in the role of apo B, triglycerides, and specific apo B-containing lipoprotein particles as independent predictors of CHD risk as well as their potential to improve risk prediction and response to lipid-modifying treatment. References [1] Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 2001;285:2486-97.E. [2] Witztum JL, Steinberg D. Role of oxidized LDL in atherogenesis. J Clin Invest 1991;88:1785 9. [3] Smith EB. Transport interactions and retention of plasma proteins in the intima; the barrier function of the internal elastic lamina. Eur Heart J 1990;11(Suppl E):72 81. [4] Kruth HS, Jones HL, Huang W, et al. Macropinocytosis is the endocytic pathway that mediates macrophage foam cell formation with native low density lipoprotein. J Biol Chem 2005;280:2352 60. [5] Marcovina SM, Adolphson JL, Parlavecchia M, Albers JJ. Effects of lyophilization of serum on the measurement of apolipoproteins A-I and B. Clin Chem 1990;36:366 9. [6] Marcovina SM, Albers JJ, Henderson LO, Hannon WH. International Federation of Clinical Chemistry standardization project for measurement of apolipoproteins A-I and B. III. Comparability of apolipoprotein A-I values by use of international reference material. Clin Chem 1993;39:773 81. [7] Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18:449 502. [8] Walldius G, Junger I, Holme I, et al. High apolipoprotein B, low apolipoprotein A-1, and improvement in the prediction of fatal myocardial infarction (AMORIS study). Lancet 2001;358:2026 33. [9] Blake GJ, Otvos JD, Rifai N, Ridker PM. Low-density lipoprotein particle concentration and size as determined by nuclear magnetic resonance spectroscopy as predictors of cardiovascular disease in women. Circulation 2002;106:1930 7. [10] Ballantyne CM, Andrews TC, Hsia JA, et al. Correlation of nonhigh-density lipoprotein cholesterol with apolipoprotein B: effect of 5 hydroxymethylglutaryl coenzyme A reductase inhibitors on nonhigh density lipoprotein cholesterol levels. Am J Cardiol 2001;88: 265 9. [11] Pischon T, Girman CJ, Sacks FM, Rifai N, Stampfer MJ, Rimm EB. Non-HDL cholesterol and apolipoprotein B in the prediction of coronary heart disease in men. Circulation 2005;112:3375 83. [12] Gotto AM, Whitney E, Stein EA, et al. Relation between baseline and on-treatment lipid parameters and first acute major coronary events in the Air Force/Texas Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS). Circulation 2000;101:477 [WGT: Need terminal page number # (see also citation for figure below)]. [13] Simes RJ, Marschner IC, Hunt D, et al., on behalf of the LIPID Study Investigators. Relationship between lipid levels and clinical outcomes in the Long-Term Intervention with Pravastatin in Ischemic Disease (LIPID) trial. To what extent is the reduction in coronary events with pravastatin explained by on-study lipid levels? Circulation 2002;105:1162 9. [14] Hokanson JE, Austin MA. Plasma triglyceride level is a risk factor for cardiovascular disease independent of high-density lipoprotein cholesterol level: a meta-analysis of population-based prospective studies. J Cardiovasc Risk 1996;3:213 9. [15] Stampfer MJ, Krauss RM, Ma J, et al. A prospective study of triglyceride level, low-density lipoprotein particle diameter, and risk of myocardial infarction. JAMA 1996;276:882 8. [16] Miller M. Is hypertriglyceridaemia an independent risk factor for coronary heart disease? The epidemiological evidence. Eur Heart J 1998;19(Suppl H):H18 22. [17] Talmud P, Hawe E, Miller GJ, Humphries SE. Nonfasting apolipoprotein B and triglycerides levels as a useful predictor of coronary heart disease risk in middle-age UK men. Arterioscler Thromb Vasc Biol 2002;22:1918 23. [18] Sacks FM, Campos H. Low-density lipoprotein size and cardiovascular disease: a reappraisal. J Clin Endocrinol Metab 2003;88:4525 32 (Clinical Review 163). [19] Campos H, Moye L, Glasser S, Sacks FM. Low-density lipoprotein size, pravastatin treatment, and coronary events. JAMA 2001;286:1468 74. [20] Alaupovic P, Mack WJ, Knight-Gibson C, et al. The role of triglyceride-rich lipoprotein families in the progression of atherosclerotic lesions as determined by sequential coronary angiography from a controlled clinical trial. Arterioscler Thromb Vasc Biol 1997;17:715 22. [21] Sacks FM, Alaupovic P, Moye LA, et al. Very low density lipoproteins, apolipoproteins B, CIII, and E and risk of recurrent coronary events in the cholesterol and recurrent events (CARE) trial. Circulation 2000;102:1886 92. [22] Lee SJ, Campos H, Moye LA, Sacks FM. LDL containing apolipoprotein CIII is an independent risk factor for coronary events in diabetic patients. Arterioscler Thromb Vasc Biol 2003;23:853 8. [23] Kawakami A, Aikawa M, Libby P, Alcaide P, Luscinskas FW, Sacks FM. Apolipoprotein CIII in apo B lipoproteins enhances the adhesion of human monocytic cells to endothelial cells. Circulation 2006;113:691 700. [24] Juntti-Berggren L, Refai E, Appelskog I, et al. Apolipoprotein CIII promotes Ca 2+ -dependent beta cell death in type 1 diabetes. Proc Natl Acad Sci USA 2004;101:10090 4.