Relation of Increased Prebeta-1 High-Density Lipoprotein Levels to Risk of Coronary Heart Disease

Article Outline

• Abstract
• Methods
• Results
• Discussion
• Supplementary data
• References
• Copyright

Preβ-1 high-density lipoprotein (HDL) plays a key role in reverse cholesterol transport by promoting cholesterol efflux. Our aims were (1) to test previous associations between preβ-1 HDL and coronary heart disease (CHD) and (2) to investigate whether preβ-1 HDL levels also are associated with risk of myocardial infarction (MI). Plasma preβ-1 HDL was measured by an ultrafiltration–isotope dilution technique in 1,255 subjects recruited from the University of California–San Francisco Lipid and Cardiovascular Clinics and collaborating cardiologists. Preβ-1 HDL was significantly and positively associated with CHD and MI even after adjustment for established risk factors. Inclusion of preβ-1 HDL in a multivariable model for CHD led to a modest improvement in reclassification of subjects (net reclassification index 0.15, p = 0.01; integrated discrimination improvement 0.003, p = 0.2). In contrast, incorporation of preβ-1 HDL into a risk model of MI alone significantly improved reclassification of subjects (net reclassification index 0.21, p = 0.008; integrated discrimination improvement 0.01, p = 0.02), suggesting that preβ-1 HDL has more discriminatory power for MI than for CHD in our study population. In conclusion, these results confirm previous associations between preβ-1 HDL and CHD in a large well-characterized clinical cohort. Also, this is the first study in which preβ-1 HDL was identified as a novel and independent predictor of MI above and beyond traditional CHD risk factors.

High-density lipoproteins (HDL) exert atheroprotective activities including promotion of efflux of cholesterol from arterial macrophages mediated by several transporters, chiefly ATP binding cassette transporter A1 (ABCA1).¹, ², ³ Free cholesterol is effluxed to preβ-1 HDL.⁴, ⁵ This 60 kDa particle, a quantitatively minor species of HDL, appears to be the quantum particle of the HDL system.¹ It is formed as nascent apolipoprotein A-I enters plasma⁶ and as a substrate or product in interconversion of HDL species.⁷, ⁸, ⁹, ¹⁰, ¹¹, ¹² After acquiring free cholesterol, it becomes the substrate for lecithin–cholesterol acyl transferase, resulting in esterification of cholesterol. As cholesteryl esters accumulate, they form a core in α-HDL particles, subsuming preβ-1 HDL (Figure 1). The steady-state level of preβ-1 HDL reflects (1) rates of generation by de novo synthesis and from transfer of cholesteryl esters from HDL to acceptor lipoproteins and (2) removal by catabolism and conversion to larger HDL species by the action of lecithin–cholesterol acyl transferase. In vitro addition of preβ-1 HDL promotes efflux of cholesterol from macrophages.¹, ³ Paradoxically, previous studies have described higher levels of preβ-1 HDL in subjects with coronary heart disease (CHD) compared to controls.¹³, ¹⁴, ¹⁵, ¹⁶, ¹⁷ Also, increased preβ-1 HDL have correlated with recurrent events in an HDL intervention trial.¹⁶, ¹⁸ Our aims were to determine whether increased preβ-1 HDL is predictive of CHD beyond traditional risk factors in a large well-characterized cohort and to evaluate whether high levels of preβ-1 HDL are associated with myocardial infarction (MI).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 1. 

  The steady-state level of preβ-1 high-density lipoprotein in plasma reflects opposing rates of formation and removal. It acquires free cholesterol effluxed from the ABCA1 transporter that is then esterified with a free fatty acid derived from lecithin mediated by lecithin–cholesterol acyl transferase (LCAT). Cholesteryl esters are then incorporated into cores of large-diameter high-density lipoprotein particles (α–high-density lipoprotein). Apolipoprotein (apo) A-I is subsumed into those particles. Cholesteryl esters are transferred from α–high-density lipoprotein into cores of acceptor lipoproteins by cholesteryl ester transfer protein (CETP). Most cholesteryl esters are ultimately delivered to the liver in remnant particles and low-density lipoproteins. Cholesteryl esters are also transferred to hepatocytes by the scavenger receptor class B type I (SR-BI) receptor, liberating preβ-1 high-density lipoprotein. Concomitant with transfer of cholesteryl esters, preβ-1 high-density lipoprotein is regenerated. It is also formed de novo during secretion of newly synthesized apolipoprotein A-I from the liver and intestine and participates in catabolism of apolipoprotein A-I in the kidney. IDL = intermediate-density lipoproteins; PLTP = phospholipid transfer protein; VLDL = very low-density lipoproteins.

Back to Article Outline

Methods 

Subjects referred because of demonstrable CHD or for risk-factor assessment and management to the University of California–San Francisco Lipid and Cardiology Clinics and to collaborating cardiology practices were recruited for the study. All were clinically stable at time of recruitment. No samples were collected within 2 weeks after an MI. The study was approved by the committee on human research of University of California–San Francisco and written consent was obtained from all subjects. Because of the potentially confounding effect of lipid-lowering medications, subjects with missing information regarding medication were excluded (n = 1,048), leaving 1,255 available for analysis. Subjects were considered to have CHD if any of the following criteria was fulfilled: coronary disease on angiogram (n = 296), MI (n = 213), coronary revascularization (n = 167), angioplasty (n = 227), or stent placement (n = 116). The 4 principal coronary vessels and their branches were imaged after administration of nitroglycerin. Angiographic disease was graded by a “sum score” (sum of percent luminal intrusion of all lesions). A sum score of 60 was the threshold discriminant for diagnosis of CHD. Mean sum score was 240.6. Diagnosis of MI was based on ≥1 of the following: troponin levels, electrocardiographic evidence, or echocardiographic studies.

Lipoproteins were measured after a 10-hour fast on plasma drawn into buffered K₂–ethylenediaminetetra-acetic acid and chilled immediately to 0°C to stop esterification of cholesterol by lecithin–cholesterol acyl transferase, preventing conversion of preβ-1 HDL to larger HDL species. Lipids were measured on ultracentrifugal fractions¹⁹ in 65% of subjects including all subjects with triglycerides >300 mg/dl. Very low-density lipoproteins were separated at density <1.006 g/L. Low-density lipoproteins (LDL) and HDL were separated using heparan sulfate and magnesium chloride. In the remainder, apolipoprotein B–containing lipoproteins were precipitated with dextran sulfate,²⁰ and LDL cholesterol was calculated by the Friedewald formula.²¹ Subjects with plasma triglycerides >500 mg/dl were excluded to avoid confounding alterations in molecular speciation of HDL. Measurements were standardized by the Standardization Program of the National Center for Disease Control.

Plasma content of preβ-1 HDL was measured by the technique of ultrafiltration–isotope dilution developed in our laboratory.²² We previously showed that preβ-1 HDL can be categorically separated from larger-diameter HDL particles by low-velocity centrifugation in ultrafilters and measured in plasma by application of the isotope dilution principle.²² Ultrafiltered particles show discrete preβ-1 mobility by Western blotting in agarose gel electrophoresis identical to that of preβ-1 HDL from whole plasma. Apparent particle diameters are also identical to those of preβ-1 HDL studied directly from plasma, with a molecular mass circa 60 kDa as confirmed by fast protein liquid chromatography analysis.²² Thus, we believe the ultrafiltration technique measures identical molecular subspecies of HDL that can be observed by Western blotting of 2-dimensional electrophoretograms. These properties of preβ-1 HDL are consistent in subjects with major dyslipidemic phenotypes with the exception of those with triglyceride levels >500 mg/dl where aberrant preβ-1 HDL particles may be observed. Preβ-1 HDL of >99% purity prepared from fresh normal plasma by selected affinity immunosorption²³ and preparative electrophoresis is labeled with a tritium adduct for use as an internal standard. Content of apolipoprotein A-I in the ultrafiltrate is measured by enzyme-linked immunosorbent assay²⁴ for calculation of its specific activity. Counts applied in the sample divided by the specific activity yields the content of preβ-1 HDL in the original sample. Incremental addition of purified preβ-1 HDL to plasma results in linear recovery by this method. Assays are run in duplicate with intra- and interassay coefficients of variation of 6% and 7%, respectively. Samples are stable for extended periods at −80°C. Relations of preβ-1 HDL levels to clinical and biochemical variables using this technique have been reported in normolipidemic subjects.²⁵

Descriptive statistics of demographic and clinical characteristics are presented as mean ± SD or median (median absolute deviation) for continuous variables and as number of subjects (percentage) for categorical variables. Percent preβ-1 HDL level, defined as the percent total plasma apolipoprotein A-I in preβ-1 HDL, was used in all analyses. Percent preβ-1 HDL was categorized into tertiles and distribution of traditional CHD risk factors compared across tertiles. Associations between preβ-1 HDL tertiles and CHD risk factors were examined using analysis of variance for continuous variables and chi-square test of independence for categorical variables. To evaluate the association of preβ-1 HDL with CHD and MI, unadjusted and adjusted odds ratios and their 95% confidence intervals were estimated for preβ-1 HDL in univariable and multivariable logistic regression models. The latter included conventional nonlipid CHD risk factors (age, gender, race, body mass index, smoking status, type 2 diabetes mellitus, and hypertension), lipid risk factors (HDL, LDL, and triglycerides), apolipoprotein A-I, and lipid-altering medication usage. Continuous variables with positively skewed distributions (preβ-1 HDL, body mass index, HDL, and LDL) were log-transformed and standardized. Thus, their odds ratios represent the increased odds of disease prevalence given a 1-SD increment. Age was divided into 5-year intervals such that its odds ratio represents the increased odds of risk for a 5-year increase. Because of the highly skewed nature of triglycerides, levels were dichotomized at 200 mg/dl in regression models. We also examined interactions between preβ-1 HDL and gender. Multivariable models were estimated in subjects who did not report using lipid-lowering medications (n = 1,063) and are presented as supplementary material (available online). To evaluate the contribution of preβ-1 HDL to risk models, likelihood-ratio tests were performed comparing risk models with and without the inclusion of preβ-1 HDL as a predictor. Also, we evaluated improvement in area under receiver operating characteristic curves. Because improvements in area under receiver operating characteristic curves can be insensitive to biomarkers with moderate effect sizes and even established CHD risk factors such as lipids, hypertension, and smoking,²⁶, ²⁷ we evaluated improvement in classification of cases and controls with the addition of preβ-1 HDL in multivariable models with 2 measurements: (1) net reclassification index using predefined risk categories of <6%, 6% to 20% and >20% and (2) integrated discrimination improvement index. The net reclassification index classifies subjects into risk categories defined a priori in a model with and without a biomarker and formally tests whether there is a net gain in reclassification because of the biomarker. A gain in reclassification occurs when cases are reclassified into higher-risk categories and controls are reclassified into lower-risk categories. Conversely, a loss in reclassification occurs when cases are reclassified into lower-risk categories and controls are reclassified into higher-risk categories. These gains and losses are subsequently aggregated to assess statistically whether there is a net gain in reclassification because of a biomarker.²⁶ Because the net reclassification index is dependent on predefined risk categories, we evaluated improvement in classification with the integrated discrimination improvement index, which does not require predefined risk categories, by quantifying improvement in sensitivity and specificity integrated over all possible thresholds.²⁶ Results with a p value <0.05 were regarded as statistically significant. All analyses were performed in R 2.9.0 (MediaFire Woodlands, Texas).

Back to Article Outline

Results 

Characteristics of the 1,255 subjects according to preβ-1 HDL tertiles are presented in Table 1. Onset of clinical CHD occurred at a relatively young age in affected subjects (mean age 60.2 years). Most were of European ancestry, followed by Asians, and the rest were distributed in African-Americans and Hispanics. Few subjects were receiving lipid-lowering medications. Because virtually all samples were drawn at initial visits and many subjects were asymptomatic, lipid-lowering drug therapy had not been initiated by referring physicians. Of the 16% taking lipid-lowering drugs at entry, 64% had been prescribed a statin, 8% niacin, and the remainder a combination of statin with resin or fibrate. It is unlikely that lipid-lowering medications would have a significant impact on the finding of this study because there was no significant difference in levels of preβ-1 HDL in subjects taking versus not taking those medications (mean ± SD, no lipid medication 7.9 ± 6.5 vs lipid medication 7.7 ± 6.4, p = 0.73). Mean ± SD preβ-1 HDL was 7.87 ± 6.45% of plasma apolipoprotein A-I in the entire cohort. Preβ-1 HDL levels were associated with all demographic and clinical characteristics with the exception of LDL cholesterol, diabetes mellitus, and lipid-lowering medication. Subjects with increased preβ-1 HDL were slightly younger and had more adverse clinical characteristics. They had higher body mass indexes and triglyceride levels, lower HDL cholesterol and apolipoprotein A-1 levels, and were more likely to be men, current smokers, and hypertensive. The negative correlation between preβ-1 HDL and HDL cholesterol was more pronounced in subjects without CHD (r = −0.34, p <0.001) than in subjects with CHD (r = −0.16, p <0.001). Preβ-1 HDL differed among races with Asians having significantly lower levels than others. Although menopausal status was not obtained, we examined preβ-1 HDL levels in women >55 years old (who are most likely postmenopausal, n = 212) and women <35 years old (who are most likely premenopausal, n = 132). Preβ-1 HDL levels were similar between these groups (mean ± SD, premenopausal 6.84 ± 6.07 vs postmenopausal 6.92 ± 5.91, p = 0.78).

Table 1. Demographic and clinical characteristics according to tertiles of prebeta-1 high-density lipoprotein

Variable	Total (n = 1,255)	Preβ-1 HDL Tertiles			p Value
		<4.171	4.171–8.838	>8.838
		(n = 418)	(n = 418)	(n = 419)
Age (years)	50.8±16.9	52.2±17.1	51.2±17.0	48.9±16.6	0.005
Men	558(44.4%)	155(37.1%)	201(48.1%)	202(48.2%)	<0.001
Body mass index (kg/m2)	26.6±5.4	25.2±5.0	27.2±5.8	27.3±5.1	<0.001
Caucasian	901(73.4%)	287(70.0%)	299(73.6%)	315(76.6%)	0.02
Asian	192(15.6%)	84(20.5%)	57(14.0%)	51(12.4%)
Other (African-American, Hispanic, other)	134(10.9%)	39(9.5%)	50(12.3%)	45(10.9%)
Hypertension	422(33.7%)	123(29.5%)	138(33.0%)	161(38.4%)	0.02
Smoker (current)	83(6.6%)	16(3.8%)	25(6.0%)	42(10.0%)	0.001
Lipid-lowering medication	193(15.4%)	68(16.3%)	61(14.6%)	64(15.3%)	0.8
Triglycerides (mg/dl)	148(87.5)	109(56.3)	151(77.1)	199(147.8)	<0.001
Low-density lipoprotein cholesterol (mg/dl)	147.9±58.6	146.5±53.3	148.7±56.2	148.5±65.8	0.5
High-density lipoprotein cholesterol (mg/dl)	50.8±18.1	58.4±19.8	48.4±16.8	45.5±14.7	<0.001
Apolipoprotein A-I (mg/ml)	1.17±0.35	1.22±0.35	1.15±0.38	1.14±0.31	<0.001
Type 2 diabetes mellitus	105(8.4%)	32(7.7%)	28(6.7%)	45(10.7%)	0.09
Coronary heart disease	454(36.2%)	126(30.1%)	158(37.8%)	170(40.6%)	0.005
Myocardial infarction	213(17.0%)	56(13.4%)	68(16.3%)	89(21.2%)	0.01

Distributions of demographic and clinical characteristics are shown across tertiles of preβ-1 high-density lipoprotein, defined as percent total plasma apolipoprotein A-I present in preβ-1 high-density lipoprotein. Data are presented as number of subjects (percentage), mean ± SD, or median (median absolute deviation). Number of subjects for each demographic and clinical characteristic varies because of missing data.

Unadjusted and adjusted associations of preβ-1 HDL with CHD and MI are presented in Table 2. preβ-1 HDL levels were significantly and positively associated with increased risk of CHD and MI even after adjustment for conventional risk factors. Likelihood-ratio tests showed that preβ-1 HDL contributed significantly to unadjusted and adjusted models. Estimates for all covariates entered in multivariable risk models are shown in the supplementary material. Results were similar when analysis was limited to the subgroup of subjects who were not taking lipid-lowering medication (see supplementary material).

Table 2. Associations of prebeta-1 high-density lipoprotein tertiles with coronary heart disease and myocardial infarction

	CHD		MI
	OR (95% CI)	p Value	OR (95% CI)	p Value
Unadjusted
Preβ-1 high-density lipoprotein tertiles
<4.171	1.00(reference)	—	1.00(reference)	—
4.171–8.838	1.41(1.06–1.88)	0.02	1.25(0.85–1.84)	0.2
>8.838	1.58(1.19–2.11)	0.002	1.74(1.21–2.51)	0.003
Likelihood-ratio test	21.4	<0.001	18.4	<0.001
Adjusted
Preβ-1 high-density lipoprotein tertiles
<4.171	1.00(reference)	—	1.00(reference)	—
4.171–8.838	1.34(0.92–1.95)	0.1	1.17(0.75–1.83)	0.5
>8.838	1.65(1.11–2.46)	0.01	1.94(1.23–3.05)	0.004
Likelihood-ratio test	6.1	0.047	10.0	0.007

Unadjusted and adjusted odds ratios and their 95% confidence intervals are shown for tertiles of preβ-1 high-density lipoprotein, defined as percent total apolipoprotein A-I present in plasma. Adjusted odds ratios adjust for age, gender, body mass index, race, hypertension, current smoking status, type 2 diabetes mellitus, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, triglycerides, apolipoprotein A-I, and usage of lipid-lowering medication in multivariable logistic regression models. Likelihood-ratio tests were performed comparing nested models with and without inclusion of preβ-1 high-density lipoprotein tertiles in logistic regression models.

CI = confidence interval; OR = odds ratio.

Table 3 presents 3 statistical measurements that were used to evaluate the utility of adding preβ-1 HDL as an additional predictor in conventional risk factor models. Inclusion of preβ-1 HDL in multivariable models led to trivial increments in area under receiver operating characteristic curves for CHD and MI. However, changes in area under receiver operating characteristic curves are known to be insensitive to biomarkers of moderate effect.²⁶, ²⁷ Including preβ-1 HDL in the risk model for CHD resulted in a significant improvement in classification as indicated by the net reclassification index; however, the integrated discrimination improvement index estimate was nonsignificant, suggesting that reclassification improvement was not robust to different definitions of risk categories. In contrast, incorporation of preβ-1 HDL into the risk model for MI led to a significant improvement in the net reclassification index and integrated discrimination improvement index.

Table 3. Effect of adding prebeta-1 high-density lipoprotein tertiles to risk-prediction models

	AUC		NRI		IDI
	Change (95% CI)	p Value	Estimate (95% CI)	p Value	Estimate (95% CI)	p Value
Coronary heart disease	0.001(−0.002to0.004)	0.5	0.15(0.04–0.28)	0.01	0.003(−0.001to0.007)	0.2
Myocardial infarction	0.006(−0.003to0.015)	0.2	0.21(0.05–0.36)	0.008	0.010(0.002–0.018)	0.02

The increment because of inclusion of preβ-1 high-density lipoprotein tertiles in risk models compared to conventional risk factors alone is shown for various calibration statistics. Areas under receiver operating characteristic curves in risk models that exclude preβ-1 high-density lipoprotein are 0.85 for coronary heart disease and 0.78 for myocardial infarction.

AUC = area under receiver operating characteristic curve; IDI = integrated discrimination improvement index; NRI = net reclassification improvement index. Other abbreviation as in Table 2.

Back to Article Outline

Discussion 

This study in a large well-characterized clinical cohort confirms the association between increased levels of preβ-1 HDL and risk of CHD reported previously.¹³, ¹⁴, ¹⁵, ¹⁶, ¹⁷, ¹⁸ Higher levels of preβ-1 HDL remained a significant predictor of CHD even after adjustment for traditional risk factors. This study also reveals a novel association between increased levels of preβ-1 HDL and risk of MI. Higher levels are predictive of MI above and beyond conventional risk factors. The magnitude of effect was comparable to effect sizes observed for factors such as lipids and systolic blood pressure. Inclusion of preβ-1 HDL in traditional risk-factor models significantly improved classification of subjects for MI and modestly improved classification for CHD, suggesting that preβ-1 HDL has more discriminatory power for MI than CHD in our study population. Because progression of coronary atherosclerosis to infarction depends on factors contributing to instability of plaque, this observation may have mechanistic importance.

Consistent with our finding that levels of preβ-1 HDL are higher in subjects with CHD, increased levels were correlated with more adverse characteristics such as higher body mass index, hypertension, and smoking. Preβ-1 HDL was correlated with all lipid risk factors except LDL. It was positively correlated with levels of triglycerides and negatively with levels of HDL cholesterol. Preβ-1 HDL levels were similar between women >55 years old and those <35 years old, suggesting that levels are unaffected by menopause. Few studies to date have examined preβ-1 HDL as a predictor of CHD.¹³, ¹⁴, ¹⁵, ¹⁶, ¹⁷, ¹⁸ These have reported a positive association in men¹⁴, ¹⁶ and women.¹⁵ We did not observe a significant interaction between preβ-1 HDL and gender with respect to any CHD end point, suggesting that the association is similar in men and women, although of subjects with normal values, men have higher levels.²⁵

We observed that the association between preβ-1 HDL and CHD was robust to the adjustment of established CHD risk factors in contrast to previous reports.¹⁴, ¹⁶ These discrepancies may be attributed to differences in sample size, prevalence of CHD in study cohorts, and covariates included in the multivariable model. Because our study included subjects from lipid and cardiology clinics, prevalence of CHD was increased and thus increased the power to detect an association of preβ-1 HDL with CHD.

Measurement of preβ-1 HDL in this study used an entirely different analytic approach (ultrafiltration and isotope dilution²²) from methods in previous studies. However, because its development was based on quantitative analysis of purified preβ-1 HDL added to plasma yielding a linear response, it is a quantitative measurement. Although other HDL particles of preβ mobility exist, their particle mass is much greater and they are excluded by the ultrafilter used. The isolated particle contains 2 copies of apolipoprotein A-I and no other proteins and shows little variation in mass or charge. Its lipid moiety contains approximately 5 molecules of phospholipids, 0.5 to 1.0 molecule of free cholesterol, 1 to 2 molecules of cholesteryl esters, and no triglycerides. Several methods have been applied to assess preβ-1 HDL in 1- and 2-dimensional gel electrophoresis. Because those studies are not based on purified preβ-1 HDL as a standard and involve differences in antibody affinity among HDL species, the reported preβ-1 HDL values cannot be directly compared to those in this study. Size-exclusion chromatography can separate preβ-1 HDL particles that share characteristics of those separated by ultrafiltration including particle mass and percent composition of HDL.²⁸ The physical identity of the preβ-1 HDL species across different methods and concordance between the association of levels of preβ-1 HDL with CHD by ultrafiltration and Western blotting¹³, ¹⁴, ¹⁵, ¹⁶, ¹⁷, ¹⁸ are noteworthy. The ability to make quantitative measurements of individual molecular species of HDL is critical to a mechanistic understanding of the atheroprotective functions of HDL. These include antioxidant and anti-inflammatory activities.²⁹ The best understood is reverse cholesterol transport. At least 4 mechanisms for efflux from macrophages are recognized: the ABCA1 and ABCG1 transporters, the scavenger receptor class B type I receptor, and passive diffusion. Sviridov et al³⁰ recognized that plasma depleted of preβ-1 HDL could still support cholesterol efflux from macrophages. Later, de la Llera-Moya et al¹ studied the relative contributions of the pathways using a selective inhibition protocol. They concluded that the ABCA1 mechanism predominates, with the ABCG1 pathway contributing about 20% and the scavenger receptor class B type I receptor very little.¹ In vitro studies with human macrophages have demonstrated that preβ-1 HDL is the primary acceptor in ABCA1-mediated efflux. In plasma from normal subjects, rate of cholesterol efflux from the ABCA1 transporter has been shown to be positively related to the level of preβ-1 HDL.¹ Esterification of cholesterol by lecithin–cholesterol acyl transferase depletes the level of preβ-1 HDL. Cholesteryl ester transfer protein, phospholipid transfer protein, and scavenger receptor class B type I activities can produce preβ-1 HDL.⁷, ⁸, ⁹, ¹⁰ Our findings of high levels of preβ-1 HDL in subjects with CHD points to some defect in the efflux mechanism or subsequent metabolic processes in reverse cholesterol transport. If efflux by ABCA1 or lecithin–cholesterol acyl transferase activity were impaired, an accumulation of unrepleted acceptor (preβ-1 HDL) would be expected. This is consistent with the observation on 2-dimensional gel analysis that lower levels of large-diameter particles are associated with increased risk of CHD. The finding that the rate of efflux of cholesterol through the ABCA1 pathway in vitro can differ significantly between subjects with identical levels of HDL cholesterol¹ and the association of impaired acceptor properties of HDL with CHD suggest that the acceptor properties of HDL present an additional variable determinant of CHD risk.³ Because measurement of the level of preβ-1 HDL is much easier than determination of kinetics of efflux, it is likely to become a clinically valuable indicator of risk.

The strengths of this study are in its large sample and detailed clinical phenotypes. Nevertheless, several limitations apply. The cohort is not representative of a general population because it was recruited from lipid and cardiology clinics. However, all associations for established risk factors were in the same direction as those reported in general population studies, excepting male gender (see supplementary material). Also, all analyses were cross sectional because the study lacked longitudinal information. Further research regarding the prospective discriminatory and reclassification power of preβ-1 HDL with respect to CHD and MI is warranted to fully evaluate it as a biomarker and to study effects of lifestyle, medications, and other factors. Our analyses focused on preβ-1 HDL and did not include other HDL subpopulations that are highly correlated with each other.

Back to Article Outline

Supplementary data 

Back to Article Outline

References 

1. de la Llera-Moya M, Drazul-Schrader D, Asztalos BF, Cuchel M, Rader DJ, Rothblat GH. The ability to promote efflux via ABCA1 determines the capacity of serum specimens with similar high-density lipoprotein cholesterol to remove cholesterol from macrophages. Arterioscler Thromb Vasc Biol. 2010;30:796–801
   • View In Article
   • CrossRef
2. Tall AR, Yvan-Charvet L, Terasaka N, Pagler T, Wang N. HDL, ABC transporters, and cholesterol efflux: implications for the treatment of atherosclerosis. Cell Metab. 2008;7:365–375
   • View In Article
   • CrossRef
3. Khera AV, Cuchel M, de la Llera-Moya M, Rodriquez A, Barke MF, Jafri K, et al. Cholesterol efflux capacity, high density lipoprotein function and atherosclerosis. N Engl J Med. 2011;364:127–135
   • View In Article
   • CrossRef
4. Kunitake ST, La Sala KJ, Kane JP. Apolipoprotein A-I–containing lipoproteins with pre-beta electrophoretic mobility. J Lipid Res. 1985;26:549–555
   • View In Article
   • MEDLINE
5. Castro GR, Fielding CJ. Early incorporation of cell-derived cholesterol into pre-beta-migrating high-density lipoprotein. Biochemistry. 1988;27:25–29
   • View In Article
   • MEDLINE
   • CrossRef
6. Castle CK, Pape ME, Marotti KR, Melchior GW. Secretion of prebeta-migrating apoA-I by cynomolgus monkey hepatocytes in culture. J Lipid Res. 1991;32:439–447
   • View In Article
   • MEDLINE
7. Jiang X, Francone OL, Bruce C, Milne R, Mar J, Walsh A, et al. Increased prebeta-high density lipoprotein, apolipoprotein AI, and phospholipid in mice expressing the human phospholipid transfer protein and human apolipoprotein AI transgenes. J Clin Invest. 1996;98:2373–2380
   • View In Article
   • MEDLINE
   • CrossRef
8. Lee M, Metso J, Jauhiainen M, Kovanen PT. Degradation of phospholipid transfer protein (PLTP) and PLTP-generated pre-beta-high density lipoprotein by mast cell chymase impairs high affinity efflux of cholesterol from macrophage foam cells. J Biol Chem. 2003;278:13539–13545
   • View In Article
   • MEDLINE
   • CrossRef
9. Francone OL, Royer L, Haghpassand M. Increased prebeta-HDL levels, cholesterol efflux, and LCAT-mediated esterification in mice expressing the human cholesteryl ester transfer protein (CETP) and human apolipoprotein A-I (apoA-I) transgenes. J Lipid Res. 1996;37:1268–1277
   • View In Article
   • MEDLINE
10. Kunitake ST, Mendel CM, Hennessy LK. Interconversion between apolipoprotein A-I–containing lipoproteins of pre-beta and alpha electrophoretic mobilities. J Lipid Res. 1992;33:1807–1816
    • View In Article
    • MEDLINE
11. Barrans A, Collet X, Barbaras R, Jaspard B, Manent J, Vieu C, et al. Hepatic lipase induces the formation of pre-beta 1 high density lipoprotein (HDL) from triacylglycerol-rich HDL2 (A study comparing liver perfusion to in vitro incubation with lipases). J Biol Chem. 1994;269:11572–11577
    • View In Article
    • MEDLINE
12. Wolfrum C, Poy MN, Stoffel M. Apolipoprotein M is required for prebeta-HDL formation and cholesterol efflux to HDL and protects against atherosclerosis. Nat Med. 2005;11:418–422
    • View In Article
    • MEDLINE
    • CrossRef
13. Miida T, Nakamura Y, Inano K, Matsuto T, Yamaguchi T, Tsuda T, et al. Pre beta 1–high-density lipoprotein increases in coronary artery disease. Clin Chem. 1996;42:1992–1995
    • View In Article
    • MEDLINE
14. Asztalos BF, Cupples LA, Demissie S, Horvath KV, Cox CE, Batista MC, 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
    • View In Article
    • CrossRef
15. Lamon-Fava S, Herrington DM, Reboussin DM, Sherman M, Horvath KV, Cupples LA, et al. Plasma levels of HDL subpopulations and remnant lipoproteins predict the extent of angiographically-defined coronary artery disease in postmenopausal women. Arterioscler Thromb Vasc Biol. 2008;28:575–579
    • View In Article
    • CrossRef
16. Asztalos BF, Collins D, Cupples LA, Demissie S, Horvath KV, Bloomfield HE, 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
    • View In Article
    • CrossRef
17. Sethi AA, Sampson M, Warnick R, Muniz N, Vaisman B, Nordestgaard BG, et al. High pre-beta1 HDL concentrations and low lecithin: Cholesterol acyltransferase activities are strong positive risk markers for ischemic heart disease and independent of HDL-cholesterol. Clin Chem. 2010;56:1128–1137
    • View In Article
    • CrossRef
18. 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 Veterans Affairs high-density lipoprotein intervention trial. Metabolism. 2008;57:77–83
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (137 KB)
    • CrossRef
19. Pullinger CR, Hennessy LK, Chatterton JE, Liu W, Love JA, Mendel CM, et al. Familial ligand-defective apolipoprotein B (Identification of a new mutation that decreases LDL receptor binding affinity). J Clin Invest. 1995;95:1225–1234
    • View In Article
    • MEDLINE
    • CrossRef
20. Warnick GR, Benderson J, Albers JJ. Dextran sulfate-Mg2+ precipitation procedure for quantitation of high-density-lipoprotein cholesterol. Clin Chem. 1982;28:1379–1388
    • View In Article
    • MEDLINE
21. 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;18:499–502
    • View In Article
    • MEDLINE
22. O'Connor PM, Naya-Vigne JM, Duchateau PN, Ishida BY, Mazur M, Schoenhaus SA, et al. Measurement of prebeta-1 HDL in human plasma by an ultrafiltration-isotope dilution technique. Anal Biochem. 1997;251:234–240
    • View In Article
    • MEDLINE
    • CrossRef
23. McVicar JP, Kunitake ST, Hamilton RL, Kane JP. Characteristics of human lipoproteins isolated by selected-affinity immunosorption of apolipoprotein A-I. Proc Natl Acad Sci USA. 1984;81:1356–1360
    • View In Article
24. Kunitake ST, O'Connor P, Naya-Vigne J. Heterogeneity of high-density lipoproteins and apolipoprotein A-I as related to quantification of apolipoprotein A-I. Meth Enzymol. 1996;263:260–267
    • View In Article
25. O'Connor PM, Zysow BR, Schoenhaus SA, Ishida BY, Kunitake ST, Naya-Vigne JM, et al. Prebeta-1 HDL in plasma of normolipidemic individuals: Influences of plasma lipoproteins, age, and gender. J Lipid Res. 1998;39:670–678
    • View In Article
    • MEDLINE
26. Pencina MJ, D'Agostino RB, D'Agostino RB, Vasan RS. Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyond. Stat Med. 2008;27:157–172
    • View In Article
    • CrossRef
27. Cook NR. Use and misuse of the receiver operating characteristic curve in risk prediction. Circulation. 2007;115:928–935
    • View In Article
    • CrossRef
28. Nanjee MN, Brinton EA. Very small apolipoprotein A-I–containing particles from human plasma: Isolation and quantification by high-performance size-exclusion chromatography. Clin Chem. 2000;46:207–223
    • View In Article
    • MEDLINE
29. Navab M, Gharavi N, Watson AD. Inflammation and metabolic disorders. Curr Opin Clin Nutr Metab Care. 2008;11:459–464
    • View In Article
    • CrossRef
30. Sviridov D, Miyazaki O, Theodore K, Hoang A, Fukamaki I, Nectel P. Delineation of the role of pre-beta 1-HDL in cholesterol efflux using isolated pre-beta 1 HDL. Arterioscler Thromb Vasc Biol. 2002;22:1482–1488
    • View In Article
    • CrossRef