The role of small, dense low density lipoprotein (LDL): a new look

International Journal of Cardiology 74 (2000) S17–S22 www.elsevier.com / locate / ijcard

The role of small, dense low density lipoprotein (LDL): a new look Chris Packard FRCPath*, Muriel Caslake PhD, James Shepherd FRSE Department of Pathological Biochemistry, Glasgow Royal Infirmary University NHS Trust, 4 th Floor, Queen Elizabeth Building, Alexandra Parade, Glasgow G31 2 ER, UK Accepted 28 June 1999

Abstract Plasma low density lipoprotein (LDL) plays a central role in atherogenesis, and elevated levels of LDL are associated with an increased risk of coronary heart disease (CHD). Studies have now revealed that LDL is structurally heterogeneous, based on its size and density. Patients with combined hyperlipidemia exhibit a lipid profile — the so-called atherogenic lipoprotein phenotype — that is associated with elevated triglyceride levels, low levels of high density lipoprotein and a preponderance of atherogenic, small, dense LDL particles. Such individuals are at an increased risk of CHD events, regardless of their total LDL circulating mass. Evidence suggests that when plasma triglycerides exceed a critical threshold of approximately 133 mg / dl (1.5 mmol / l), this favours the formation of small, dense LDL from larger, less dense species. Lipid-lowering agents that are capable of lowering triglyceride levels below this threshold value will cause a shift to a less dense and, therefore, less atherogenic LDL profile. This effect has been demonstrated for the HMG-CoA reductase inhibitor atorvastatin which, in addition to its ability to markedly decrease the total LDL circulating mass, can also shift the LDL profile towards less dense, larger species. This suggests that atorvastatin may also affect the atherogenic lipoprotein phenotype found in patients with combined hyperlipidemia.  2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Combined hyperlipidemia; Small, dense LDL; Triglycerides; Atorvastatin

1. Introduction Low density lipoprotein (LDL) is the principal cholesterol-carrying lipoprotein in human plasma and, as such, plays a central role in atherogenesis. Elevated levels of LDL are associated with increased risk of coronary heart disease (CHD) [1]. Up until the 1950s, the lipoprotein complexes in the plasma were thought to be homogeneous structures. However, studies by Gofman et al. [2] during the 1950s and 1960s provided the first evidence that LDL may be structurally diverse. Gofman’s group used analytical ultracentrifugation to monitor concentrations of species of differing flotation rates in plasma. LDL was observed to have ‘shoulders’ on a single major *Corresponding author.

peak. Fisher [3] pointed out that the flotation pattern of LDL in hypertriglyceridemics was ‘polydisperse’, with a number of peaks present in the analytical centrifuge profile, whereas normal subjects presented a ‘monodisperse’ pattern. It was the pioneering work of Krauss and Blanche [4] that established LDL structural heterogeneity to be the norm rather than the exception. Using the high resolution technique of non-denaturing gradient gel electrophoresis, these investigators were able to show that many subjects in the population had an LDL ˚ which was smaller when particle size of ,250 A, ˚ compared to the more common LDL size of ¯260 A diameter. Density gradient centrifugation was able to reveal heterogeneity in the LDL density profile. Usually, three maxima are seen in the range 1.019– 1.063 g / ml, and these were termed LDL-I, -II and

0167-5273 / 00 / $ – see front matter  2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 99 )00107-2

S18

C. Packard et al. / International Journal of Cardiology 74 (2000) S17 –S22

-III [5,6]. LDL-I has the lowest density (1.025–1.034 g / ml) while LDL-III, also termed small, dense LDL, has the highest density (1.044–1.060 g / ml). Development of techniques in our laboratory to quantify these subfractions has enabled us to study their distribution in the plasma, their relationship to other lipoproteins and their metabolic origins [6,7].

for instance, have been shown to exhibit a particularly high prevalence of small, dense LDL, to the extent that it has been incorporated into the description of the insulin resistance syndrome [11]. It is likely, however, that the relationship with the type 2 diabetic state is a function of the elevated plasma triglyceride levels that are a characteristic feature of diabetic dyslipidemia [12].

2. Epidemiology of small, dense LDL 4. Metabolic origins of small, dense LDL For convenience, Krauss and Blanche [4] divided the LDL gel electrophoresis profile into two phenotypes, named patterns A and B. In pattern A, large-sized LDL is predominant, whereas in pattern B, there is a greater proportion of small, dense LDL. Pattern B was reported in about 25% of the population but was less frequent in women and younger subjects (,40 years old). Its presence was associated with moderate elevation in plasma triglyceride and low levels of high density lipoprotein (HDL) cholesterol [7,8], a combination termed the atherogenic lipoprotein phenotype. Patients with combined hyperlipidemia exhibit this atherogenic profile and are at an increased risk of CHD events, regardless of their total LDL circulating mass. Plasma triglyceride has been shown to be the major determinant of the appearance of small, dense LDL, regardless of whether the profile was assessed by gel electrophoresis [8], peak particle density [9] or quantitation of subfractions [6]. We observed that a small, dense LDL concentration of .100 mg / dl (2.6 mmol / l) lipoprotein mass was seen infrequently in individuals with plasma triglyceride levels ,115 mg / dl (1.3 mmol / l), while above this level, small, dense LDL increased in proportion to the plasma triglyceride concentration [10] (Fig. 1).

3. Genetic studies Family studies have indicated that pattern B is an inherited trait. Its appearance has been linked to a number of gene loci [4], but there is increasing recognition that no single major gene determines the LDL pattern. Instead, several common genetic variants appear to be important in precipitating the formation of smaller sized LDL. Non-insulin-dependent (type 2) diabetic patients,

The statistical association between the appearance of small, dense LDL and moderately raised plasma triglyceride levels suggests that common metabolic abnormalities that give rise to elevated very low density lipoprotein (VLDL) levels favour the formation of smaller-sized LDL. Our observation that women, at a given plasma triglyceride level, have less small, dense LDL than men indicates the presence of other regulating factors (Fig. 1) [10]. A likely candidate for such a regulatory factor is hepatic lipase. This enzyme is strongly influenced by sex steroid hormones and has been shown to be a determinant of small, dense LDL concentration [7] or pattern B [13]. Recently, we have published a metabolic model for the formation of small, dense LDL (Fig. 2). This incorporates the association with plasma triglyceride and hepatic lipase, and helps to explain the relatively low levels of pattern B in women (probably due to low levels of hepatic lipase) and younger individuals [due to lower plasma triglyceride levels, i.e. ,115 mg / dl (1.3 mmol / l)], compared with older men. Metabolic studies have shown that subjects with plasma triglyceride levels that are in the upper normal range (133–204 mg / dl [1.5–2.3 mmol / l]), or moderately elevated above this, form LDL that is slowly metabolised by the receptor pathway (pool b LDL, Fig. 2). This species of LDL has ample opportunity to be remodelled in the circulation by the action of cholesteryl ester transfer protein-mediated exchange, which transfers cholesteryl ester from LDL [or highdensity lipoprotein (HDL)] to very low density lipoprotein (VLDL) and replaces it with triglyceride from the latter. Triglyceride-enriched LDL is now a good substrate for hepatic lipase and this particle, by the action of hepatic lipase, will lose both core triglyceride and surface phospholipid and, in the

C. Packard et al. / International Journal of Cardiology 74 (2000) S17 –S22

S19

Fig. 1. Scattergrams of LDL subfraction concentrations versus triglyceride (TG) concentrations in normolipemic men and women. The LDL-II concentration in both sexes showed a strong positive relationship below a TG level of ¯115 mg / dl (1.3 mmol / l). Male–female differences were noted at TG levels above this value, with men, but not women, showing a significant negative association. The value of 115 mg / dl (1.3 mmol / l) was the plasma TG level at which the LDL-II concentration peaked in men. The LDL-III concentration in both sexes showed little change at TG,115 mg / dl (1.3 mmol / l). Above this value, men showed a greater rise in LDL-III than did women. Reproduced with permission from Tan CE, Forster L, Caslake MJ et al. Relations between plasma lipids and postheparin plasma lipases and VLDL and LDL subfraction patterns in normolipemic men and women. Arterioscler Thromb Vasc Biol 1995;15(11):1839–1848.

process, be converted to a smaller, denser species of LDL. It appears that hepatic lipase activity has to be in the ‘male’ range in order to convert LDL-II to LDLIII efficiently (Fig. 2). The fact that the total concentration of LDL remains relatively constant in subjects with plasma triglyceride levels in the range 1.0–3.0 mmol / l (88–266 mg / dl) has indicated that when plasma triglycerides exceed a critical ‘threshold’ of approximately 1.5 mmol / l (133 mg / dl), this favours the formation of small, dense LDL from larger, less dense species.

5. Atherogenicity of small, dense LDL Structural studies have revealed aspects of the

composition and properties of small, dense LDL that support the concept of its increased atherogenicity. In a case-controlled study by Austin et al. [8], a pattern B LDL profile was associated with a three-fold increase in risk of CHD, and we have reported a seven-fold increment in risk for small, dense LDL concentrations .100 mg / dl (2.6 mmol / l) [6]. Small, dense LDL has been shown to be more readily oxidised, at least in vitro, than its larger counterparts. Furthermore, because of its reduced size it is likely to penetrate the arterial wall more easily, and we have been able to demonstrate that LDL from pattern B subjects has an enhanced affinity for arterial wall proteoglycan, thus prolonging its residence time in the subendothelial space [14]. All of these features contribute to the enhanced atherogenicity of this lipoprotein species (Table 1).

S20

C. Packard et al. / International Journal of Cardiology 74 (2000) S17 –S22

would also be predicted to give rise to an altered LDL subfraction distribution. Statins lower the three major LDL subspecies in concert [18]. Two mechanisms have been proposed for this. First, statins may enhance removal of LDL-I and -II by receptors (these fractions have high receptor binding ability). Second, statins may reduce the formation of pool b LDL from large, triglyceriderich VLDL (VLDL 1 , Fig. 2), by stimulating clearance of remnants in the VLDL 2 and intermediate density lipoprotein density ranges. Fig. 2. Model integrating LDL structural and metabolic heterogeneity. Kinetic evidence suggests that the two metabolically distinct pools in LDL (a and b) arise from different sources. Pool a is the major source detected by multicompartmental modelling of subjects with low normal plasma triglyceride (TG) levels. It is postulated to arise when apoB is secreted into the Sf 0–60 density range; LDL with the kinetic properties of pool b has been shown to be the product of VLDL 1 delipidation. The two LDL species have substantially differing residence times (RT) in the circulation. LDL-III generation is favoured when plasma TG exceeds 133 mg / dl (1.5 mmol / l) and hepatic lipase (HL) is in the male range (i.e. .15 u / l). VLDL 1 is the principal species that accumulates as plasma TG levels rise in the population as a result of overproduction of the lipoprotein or its defective removal (LpL or elevated apoCIII levels). It is envisaged that above the 133 mg / dl (1.5 mmol / l) threshold, sufficient VLDL 1 is present in both to produce long-lived pool b LDL and to cause TG enrichment of the particle to a level that makes LDL-III formation possible. The action of HL removes lipid from the LDL-II to form LDL-III. Reproduced with permission from Packard CJ, Shepherd J. Lipoprotein heterogeneity and apolipoprotein B metabolism. Arterioscler Thromb Vasc Biol 1997;17(12):3542–3556.

6. Effects of lipid-lowering therapy on the LDL profile Drugs that lower triglyceride levels below the ‘threshold’ value of about 1.5 mmol / l (133 mg / dl) will cause a change in the LDL size profile to larger, less dense and, therefore, less atherogenic species. This has been reported with nicotinic acid [15], fibrates [16,17] and HMG-CoA reductase inhibitors (statins) [18]. Modulation of hepatic lipase activity Table 1 Characteristics of the atherogenic lipoprotein phenotype • Moderate hypertriglyceridaemia — Enhanced thrombic potential — Remnant lipoproteins • Small, dense LDL — Increased ability to penetrate arterial wall — Readily oxidised — Enhanced binding to proteoglycan • Reduced plasma HDL levels

7. Differential effect of statins on LDL subfraction distribution Recent investigations in our laboratory have compared the effects of two statins, atorvastatin and simvastatin, on LDL subfraction distribution. Previous studies have demonstrated both agents to be effective in reducing elevated plasma levels of LDLcholesterol (LDL-C) and triglycerides. However, atorvastatin (10 mg / day) has been shown to provide significantly greater reductions (P,0.05) from baseline in total cholesterol, LDL-C and triglycerides, compared with simvastatin (10 mg / day) [19]. A separate study comparing atorvastatin, simvastatin, pravastatin, lovastatin and fluvastatin showed atorvastatin to have superior lipid-lowering efficacy versus all other agents at milligram-equivalent doses in the range 10–40 mg [20]. Our study compared the effects of atorvastatin and simvastatin, both at doses of 40 mg / day, on the LDL subfraction profile of patients with combined hyperlipidemia. The results confirmed that atorvastatin and simvastatin produce significant reductions from baseline (P,0.05) in lipid parameters, including LDL-C (250 versus 244%) and triglycerides (246 versus 239%). Both statins also produced comparable increases from baseline in HDL-C (12 versus 13%). Both agents also produced an improvement in the LDL size profile, with reductions in all three subfractions. Atorvastatin, however, showed a preferential decrease, compared with simvastatin, in small, dense LDL (64 versus 45%), thereby shifting the overall profile towards less dense LDL subfractions (Fig. 3) [21].

C. Packard et al. / International Journal of Cardiology 74 (2000) S17 –S22

S21

Fig. 3. Mean percent change from baseline in LDL-I, -II and -III in patients with combined hyperlipidemia following treatment with atorvastatin (40 mg / day) or simvastatin (40 mg / day) (all P,0.05 versus baseline). (From Caslake MJ, Forster LF, Stewart G, Roger E, Shepherd J, Packard CJ. Effects of atorvastatin and simvastatin on the LDL subfraction profile in combined hyperlipidemia. Abstract presented at the International Congress on Vascular Disease Prevention; 4–8 May 1998, Glasgow, UK.)

These findings suggest that atorvastatin upregulates LDL receptors, resulting in a reduction of receptoractive LDL-I and -II subfractions. As atorvastatin shifts the LDL subfraction profile to a less atherogenic species, this agent may also, in addition to its ability to markedly decrease the total LDL circulating mass, affect the atherogenic lipoprotein phenotype found in patients with combined hyperlipidemia.

8. Conclusion Small, dense LDL appears to be a particularly atherogenic species of lipoprotein. As a result, individuals with an LDL profile in which the small, dense subfraction predominates are at increased risk of CHD, regardless of their total circulating LDL mass. Studies in our laboratory and by others have

led to the development of metabolic models for the formation of small, dense LDL. These models can be used to reveal potential pathways of intervention to modulate the risk associated with this lipoprotein, and may lead to the development of drugs that can modify the underlying disturbance in the atherogenic lipoprotein phenotype. In this regard, atorvastatin has been shown to alter the LDL subfraction profile to a less dense and, hence, less atherogenic species. Therefore, in addition to its documented clinical benefits in patients with primary hyperlipidemia, atorvastatin may also be useful in the management of patients with combined hyperlipidemia.

Acknowledgements Mrs Nancy Thomson provided excellent secretarial assistance in the preparation of this manuscript.

S22

C. Packard et al. / International Journal of Cardiology 74 (2000) S17 –S22

References [1] Castelli WP, Garrison RJ, Wilson PWF et al. Incidence of coronary heart disease and lipoprotein cholesterol levels. The Framingham Study. J Am Med Assoc 1986;256:2835–8. [2] Gofman JW, Delalla O, Glazier F et al. The serum lipoprotein transport system in health, metabolic disorders, atherosclerosis and coronary artery disease. Plasma 1954;2:413. [3] Fisher WR. Heterogeneity of plasma low density lipoprotein manifestations of the physiologic phenomenon in man. Metabolism 1983;32:283. [4] Krauss RM, Blanche PJ. Detection and quantitation of LDL subfractions. Curr Opin Lipidol 1992;3:377–83. [5] Swinkels DW, Demacker PNM, Hendriks JCM, Van’t Laar A. Low density lipoprotein subfractions and relationships to other risk factors for coronary artery disease in healthy individuals. Arteriosclerosis 1989;9:604–13. [6] Griffin BA, Freeman DJ, Tait GW et al. Role of plasma triglyceride in the regulation of plasma low density lipoprotein (LDL) subfractions: relative contribution of small, dense LDL to coronary heart disease risk. Atherosclerosis 1994;106:241–53. [7] Packard CJ, Shepherd J. Lipoprotein heterogeneity and apolipoprotein B metabolism. Arterioscler Thromb Vasc Biol 1997;17:3542–56. [8] Austin MA, King MC, Vranizan KM, Krauss RM. Atherogenic lipoprotein phenotype. A proposed genetic marker for coronary heart disease risk. Circulation 1990;82:495–506. [9] Campos H, Genest Jr. JJ, Blijlevens E et al. Low density lipoprotein particle size and coronary artery disease. Arterioscler Thromb 1992;12:187–95. [10] Tan CE, Forster L, Caslake MJ et al. Relations between plasma lipids and postheparin plasma lipases and VLDL and LDL subfractions in normolipemic men and women. Arterioscler Thromb Vasc Biol 1995;15:1839–48. [11] Reaven GM, Chen Y-DI, Jeppesen J, Maheux P, Krauss RM. Insulin resistance and hyperinsulinemia in individuals with small, dense, low density lipoprotein particles. J Clin Invest 1993;92:141–6.

[12] Tan KC, Cooper MB, Ling KLE et al. Fasting and postprandial determinants for the occurrence of small, dense LDL species in non-insulin dependent diabetic patients with and without hypertriglyceridemia: the involvement of insulin, insulin precursor species and insulin resistance. Atherosclerosis 1995;113:273–87. [13] Zambon A, Austin MA, Brown BG, Hokanson JE, Brunzell JD. Effect of hepatic lipase on LDL in normal men and those with coronary artery disease. Arterioscler Thromb 1993;13:147–53. [14] Anber V, Griffin BA, McConnell M, Packard CJ, Shepherd J. Influence of plasma lipid and LDL-subfraction profile on the interaction between low density lipoprotein with human arterial wall proteoglycans. Atherosclerosis 1996;124:261–71. [15] Superko HR, Krauss RM. Different effects of nicotinic acid in subjects with different LDL subclass patterns. Atherosclerosis 1992;95:69–76. [16] Gaw A, Packard CJ, Caslake MJ. Effects of ciprofibrate on LDL metabolism in man. Atherosclerosis 1994;108:137–48. [17] Guerin M, Bruckert E, Dolphin PJ, Turpin G, Chapman MJ. Fenofibrate reduces cholesteryl ester transfer from HDL to VLDL and normalises the atherogenic, dense LDL profile in combined hyperlipidemia. Arterioscler Thromb Vasc Biol 1996;16:763–72. [18] Gaw A, Packard CJ, Murray EF. Effects of simvastatin on apoB metabolism and LDL subfraction distribution. Arterioscler Thromb 1993;13:170–89. [19] Dart A, Jerums G, Nicholson G et al. A multicenter, double-blind, one-year study comparing safety and efficacy of atorvastatin versus simvastatin in patients with hypercholesterolemia. Am J Cardiol 1997;80:39–44. [20] Jones P, Kafonek S, Laurora I et al. For the CURVES investigators. Comparative dose efficacy study of atorvastatin versus simvastatin, pravastatin, lovastatin and fluvastatin in patients with hypercholesterolemia (the CURVES study). Am J Cardiol 1998;81:582–7. [21] Caslake MJ, Forster LF, Stewart G, Roger E, Shepherd J, Packard CJ, Effects of atorvastatin and simvastatin on the LDL subfraction profile in combined hyperlipidemia. Abstract presented at the International Congress on Vascular Disease Prevention, 4–8 May 1998, Glasgow, UK.