Letter to the Editors

Intermediate density lipoprotein cholesterol as the best lipoprotein predictor of atherosclerosis severity in the Watanabe Heritable Hyperlipidemic rabbit

Intermediate-density lipoprotein (IDL) and low-density lipoprotein (LDL) consist of heterogeneous particles whose subpopulations may have different atherogenic characteristics. This study investigated the associations between these subpopulations and other lipids, lipoproteins and atherosclerosis-related markers.

A total of 416 subjects (124 males and 292 females, mean age: 50.8 years) were enrolled in this study. Using polyacrylamide gel electrophoresis, serum lipoproteins were separated according to their specific electrophoretic mobility based on particle size. The IDL particles were separated into three midbands (MID-A to C), and the LDL particles were separated into seven subfractions (LDL1 to 7).

MID-B, MID-C, LDL2 and LDL3 to 6 (as a small LDL fraction) were significantly and positively correlated with very LDL (VLDL), while MID-A and LDL1 were significantly and inversely correlated with VLDL. MID-A and LDL1 were significantly and positively correlated with high-density lipoprotein (HDL). The correlation patterns between MID-A or LDL1 and triglycerides, apolipoprotein A-I, glucose, the insulin resistance index, creatinine and the mean LDL particle size had similar trends to those between HDL and these parameters.

The respective subpopulations of IDL and LDL particles can vary in their ability to predict cardiovascular disease risks. These variations may partially explain why quantitative assessments using LDL-cholesterol concentrations, as typically performed in conventional practice, are not perfect predictors of cardiovascular disease. Further studies are required to determine the clinical relevance of analyzing the IDL and LDL subpopulations.

IDL is considered an atherogenic lipoprotein but little progress has been made on methods to subfractionate this density class. Furthermore, previous work suggests that lipoproteins retained by the arterial wall, of which chondoitin sulphate is the major arterial wall proteoglycan, are potentially atherogenic. The aim of this study was to assess the subfractionation of IDL particles using chondroitin sulphate (CS). Forty healthy subjects were recruited from laboratory staff and/or their partners. Fasted plasma samples were obtained and IDL (1.006 g/ml < d < 1.030 g/ml) was isolated. Approximately 1 mg protein of IDL was allowed to interact with CS. The unbound and bound IDL particles were eluted using a low salt and high salt buffer, respectively. On average 70% of IDL bound to CS ranging from 56 to 92%. Total, unbound and bound IDL particles were analysed for lipid composition and particle size. The unbound IDL particles were larger (32 nm) and triglyceride rich (40% versus 33%, P < 0.01), whereas the bound IDL particles were smaller (26–28 nm) and cholesterol rich (21% versus 14%, P < 0.01). The unbound particles contain at least double the amount of apo C-II and apo C-III per IDL particle compared with the bound IDL particles. There are specific IDL particles that bind to CS in vitro, these being the cholesterol rich IDL particles. It remains to be determined if these cholesterol rich IDL particles are potentially more atherogenic than the triglyceride rich IDL particles.

In order to study aortic atherosclerosis and atherosclerotic response to dietary cholesterol and soybean oil in homozygous LDLR^−/− mice, the 16 weeks old animals were randomized in 4 groups either fed standard diet (no cholesterol added, group I, 12♂ and 12♀), standard diet added 0.5% cholesterol (group II, 12♂ and 12♀), standard diet added 10% soybean oil (group III, 7♂) or standard diet added 0.5% cholesterol and 10% soybean oil (group IV, 7♂) for 14 weeks. At termination, the plasma cholesterol of males was: 9.4 mmol/l ± 0.3 (SD) (group I), 34.4 ± 6.2 (group II), 9.9 ± 0.07 (group III), 32.6 ±0.1 (group IV), and of females 6.9 ± 2.7 (group I) and 31.7 ± 4.4 (group II). No apparent difference in plasma triglyceride levels was observed between the groups of either sexes. Aortic atherosclerosis (ratio intima/media) in males was 0.17 ± 0.09 (SD) (group I), 0.96 ± 0.32 (group II, p<0.05 compared to other male groups), 0.21 ± 0.09 (group III), and 0.59 ± 0.34 (group IV, p<0.05 compared to other male groups), and in females 0.12 ± 0.08 (group I) and 0.83 ± 0.30 (group II, p<0.05 compared to female group I). No sex difference in the ratio intima/media was recorded for LDLR^−/− mice on standard or cholesterol diet. It was concluded that spontaneous hypercholesterolemia and aortic atherosclerosis in LDLR^−/− mice were significantly increased by 0.5 % cholesterol addition to the standard diet but were not affected by 10% soybean oil addition to the standard diet, and that 10% soybean oil addition to the 0.5% cholesterol diet did not affect cholesterol induced hypercholesterolemia but significantly decreased cholesterol induced aortic atherosclerosis.

There is increasing evidence that alterations in metabolism of triglyceride-rich lipoproteins are of importance in the pathogenesis of atherosclerosis and its clinical consequences. Particles with the characteristics of triglyceride-rich lipoprotein remnants have been related to the extent and severity of atherosclerosis in humans and in animal models. These particles can be identified using ultracentrifugal procedures as small, very low-density lipoprotein (VLDL) and intermediate-density lipoprotein (IDL) with Svedberg flotation rates (S_f) of 12–60. Postprandial triglyceride levels also have been related to risk of coronary artery disease, consistent with a pathologic role for remnant lipoproteins. In studies in which measurements of lipoprotein subfractions have been carried out, levels of IDL have been more predictive than low-density lipoprotein (LDL) of atherosclerosis progression as assessed by coronary artery angiography or carotid artery ultrasonography. These findings suggest that a considerable portion of the coronary disease risk attributed to LDL may be accounted for by the IDL particles included in standard LDL measurements. Other metabolic changes associated with increased levels of plasma triglyceride may also adversely affect cardiovascular disease risk. These include reductions in HDL-cholesterol and apoprotein A1, increased levels of small dense LDL particles, redistribution of apoC-III from HDL to apoB-containing lipoproteins, diminished insulin sensitivity, and procoagulant changes, including increased levels of the fibrinolysis inhibitor, plasminogen-activator inhibitor-1 (PAI-1). A predominance of small dense LDL (subclass pattern B) is a discrete marker for this cluster of interrelated abnormalities and is found in 40–50% of patients with coronary artery disease. Therapeutic interventions with favorable effects on components of this dysmetabolic profile appear to be of value in decreasing atherosclerosis risk in a substantial proportion of the population.