HDL Particles, Cell-Cholesterol Efflux, and CHD Risk: The Preβ-1 Paradox

Abstract

Objective

The cell-cholesterol efflux capacity (CEC) of high-density lipoprotein (HDL) is inversely associated with coronary heart disease (CHD) risk. ATP-binding cassette transporter-A1 (ABCA1) plays a crucial role in cholesterol efflux from macrophages to preβ-1-HDL. We tested the hypothesis that CHD patients have functionally abnormal preβ-1-HDL.

Approach and Results

HDL CEC via the ABCA1 and the scavenger receptor class B type-I (SR-BI) pathways, HDL anti-oxidative capacity, apolipoprotein A-I-containing HDL particles, and inflammatory- and oxidative-stress markers were measured in a case-control study of 100 CHD cases and 100 gender-matched controls. There were significant positive correlations between ABCA1-dependent cholesterol efflux and the levels of small lipid-poor preβ-1-particles (R²=0.535) and between SR-BI-dependent cholesterol efflux and the levels of large lipid-rich (α-1+α-2) HDL particles (R²=0.712). Cases had significantly higher (87%) preβ-1 concentrations than controls but the functionality of their preβ-1 particles (preβ-1-concentration-normalized ABCA1-dependent efflux capacity) was significantly lower (−31%). Cases had significantly lower (−12%) mean concentration of large HDL particles, but the functionality of their particles (α-1+α-2-concentration-normalized SR-BI-dependent efflux capacity) was significantly higher (22%) compared to controls. HDL anti-oxidative capacity was significantly lower (−16%) in cases than in controls. There were no significant correlations between either preβ-1 functionality or large-HDL-particle functionality with HDL anti-oxidative capacity or the concentrations of inflammatory- and oxidative-stress markers.

Conclusions

HDL CEC is significantly influenced by both the concentration and the functionality of specific HDL particles participating in cell-cholesterol efflux. CHD patients have higher than normal preβ-1 concentrations with decreased functionality, and lower than normal large-HDL particle concentrations with enhanced functionality.

INTRODUCTION

In the last several decades, cross-sectional and interventional trials have indicated that high-density lipoprotein cholesterol (HDL-C) concentration is inversely associated with coronary heart disease (CHD) risk^(1–3). Although a large spectrum of HDL functions has been described, the underlying mechanisms of how HDL protects against CHD are not fully understood. Moreover, randomized drug-intervention trials have failed to prove that increasing HDL-C level decreases CHD risk^(4–6). It has been shown that apolipoprotein (apo) A–I concentration in individual HDL subpopulations is superior to HDL-C concentration in predicting risk for cardiovascular disease (CVD)^(7–10). Recently several HDL functionality markers, most importantly the cell-cholesterol efflux capacity (CEC) of HDL via the ATP-binding cassette transporter A1 (ABCA1) pathway, have emerged as better CVD-risk markers than HDL-C and/or apoA-I concentration. First, Khera et al. reported strong inverse associations between CEC and both carotid intimae-media thickness and the likelihood of angiographic coronary artery disease, independent of HDL-C level⁽¹¹⁾. Then the Dallas Heart Study investigators reported an inverse association between CEC and incidence of cardiovascular events⁽¹²⁾. The EPIC-Norfolk investigators also reported, in a nested case-control study, a significant inverse association between CEC and incidence of CHD, independent of HDL-C concentration⁽¹³⁾. However, Li et al. reported higher CEC in patients with myocardial infarction compared to those without such events⁽¹⁴⁾. Unfortunately, CEC studies are difficult to compare because different methods have been used to determine HDL CEC: J774 mouse macrophages with ³H-labeled cholesterol were used in Khera’s studies and in the Epic-Norfolk Study^(11,13), J774 cells and BODYPI-labeled cholesterol was used in the Dallas Heart Study⁽¹²⁾, and RAW 264.7 mouse macrophages were used by Li et al.⁽¹⁴⁾.

HDL removes cholesterol from cells by multiple mechanisms. Scavenger receptor class B type I (SR-BI)-dependent CEC occurs primarily with the large lipid-rich α-1 and α-2 HDL particles ^(15,16), which are deficient in CHD subjects compared to controls^(8,9,17). ABCA1-dependent CEC is mediated by the small lipid-poor preβ-1-HDL particles^(15,16,18), whose concentration is known to be higher in CHD patients compared to controls^(17,19–21).

We hypothesized that CHD patients have functionally abnormal preβ-1 particles. We tested this hypothesis by measuring the preβ-1-concentration-normalized efflux capacities of CHD and gender-matched control subjects. In an attempt to elucidate the mechanism of this dysfunction, we also assessed the influence of HDL anti-oxidative capacity and lipid and inflammatory markers on HDL functionality.

Materials and Methods

Study design and population

The data that support the findings of this study are available from the corresponding author upon reasonable request. The study protocol was approved by the Institutional Review Board of Tufts University. The study utilized a case-control design of 200 total subjects with a 1:1 allocation of CHD cases to gender-matched controls. Cases (70 males and 30 females) were randomly selected from the baseline samples of a randomized clinical trial⁽²²⁾. Participants in this trial had stable, established coronary artery disease (including previous myocardial infarction [≥6 months prior], coronary artery bypass surgery [>12months prior] or angioplasty, stable angina, or evidence of CHD on prior imaging), an abnormal exercise tolerance test result, or ischemia by nuclear imaging, deemed by the patients’ clinical cardiologist not to require current intervention. Additional inclusion criteria were age 21–75 years, body mass index (BMI) >25 and <35 in women or <40 in men, a stable regimen of statin dosing, and an estimated creatinine clearance of at least 60 mL/min/1.73m². Participants had at least 1 evaluable segment with plaque on baseline multi-detector computed tomography angiography. Controls (70 males and 30 females) were selected from our blood bank, comprised of a large out-patient sample pool, with no history of cardiovascular, kidney, thyroid, or liver disease or diabetes, no use of lipid-lowering medication, low-density lipoprotein cholesterol (LDL-C) level < 5.18 mmol/L, and triglyceride (TG) level < 3.39 mmol/L.

Laboratory measurements

Plasma and serum samples were collected after an overnight fast. Blood chemistries were determined by standard methods on a COBAS 8000 analyzer at Boston Heart Diagnostics (Framingham, MA): total cholesterol, TG, LDL-C, HDL-C, apoA-I, apoB, and high-sensitivity C-reactive Protein using kits from Roche Diagnostics (Indianapolis, IN), small-dense LDL-C using kits from Denka-Seiken (Tokyo, Japan), and Fibrinogen using kits from Medtest (Cortland Manor, NY). Myeloperoxidase (MPO) was measured on a Dimension clinical chemistry system using kits from Siemens (Newark, DE). ApoA-I concentration in HDL subpopulations were measured in plasma by 2-dimensional native agarose-polyacrylamide gel electrophoresis, immunoblot, and image analysis as previously described⁽¹⁷⁾ and shown in Supplemental Figure I. Serum amyloid A (SAA) was measured by ELISA (Invitrogen, Waltham, MA) at Tufts University. HDL CEC was measured in all subjects at Vascular Strategies (Plymouth Meeting, PA) using methods previously described^(11,15). Briefly, ABCA1-dependent CEC was measured in J774 cells radiolabeled with 2μCi of ³H-cholesterol/mL. ABCA1 was upregulated by 6-hour incubation with 0.3 mM of 8-(4-chlorophenylthio)-cyclic- adenosine monophosphate (cAMP). Medium, containing 2.8% apoB-depleted serum, was then added for 4 hours to cAMP-stimulated and unstimulated cells. Liquid scintillation counting was used to quantify the effluxed radio-labeled cholesterol in the medium. Total (global) efflux was defined as the efflux measured from cAMP-stimulated J774 cells. Non-ABCA1-specific (basal) efflux was measured from unstimulated J774 cells. ABCA1-specific efflux was calculated as the difference between cAMP-stimulated (total) and unstimulated (non-ABCA1) cells. SR-BI-mediated CEC was measured as the fraction of radio-labeled cholesterol released from radio-labeled Fu5AH cells to the media containing 2.8% apoB-free serum. Efflux data were expressed as percent cholesterol efflux/4 hour. In the efflux assays, cells were incubated with serum. When only plasma samples were available, plasma was converted to serum by adding 25 mM CaCl₂ to plasma for clotting, and then the clot was removed by low-speed centrifugation. HDL anti-oxidative (anti-ox) capacity was measured in 64 cases and 64 gender-matched controls also at Vascular Strategies. This test assessed the ability of apoB-depleted serum to inhibit or enhance the oxidation of LDL in the presence of a fluorescent organic substrate dichlorofluorescein (DCF) (2′,7′-Dichlorodihydrofluorescein diacetate, Sigma-Aldrich). DCF was dissolved in methanol at 2.0 mg/mL and was incubated at room temperature in dark for 1 hour with isolated LDL in the presence or absence of apoB-depleted plasma as surrogate for HDL⁽²³⁾. Fluorescence intensity was determined with a spectrofluorometer set at an excitation wavelength of 485 nm and an emission wavelength of 530 nm. HDL anti-ox capacity was expressed in arbitrary units. Values > 1.0 indicated that HDL had anti-ox capacity while values < 1.0 indicated that HDL was pro-oxidative. To correct for inter-assay variability, a pool of human serum was tested in parallel with the test samples in every efflux- and anti-ox capacity assay. For inter-day comparability, measurements were normalized to a standard pool.

Statistical analysis

The study size was based on power calculations with at least 90% power to detect significant differences in CEC, HDL particle concentrations, and HDL-particle-concentration-normalized CEC between cases and controls. Differences in outcomes were assessed using multivariate linear regression. All correlations reported were coefficients of determination (R²). Cases and controls were gender matched, and all analyses were adjusted for age. Information about any other potential confounders such as anthropometrics or smoking status was not available and thus not included in the analyses. All analyses were performed using the R Language for Statistical Computing version 3.2.2.

RESULTS

Major characteristics of the CHD and the control groups are presented in Table 1. Compared to controls, cases had significantly lower concentrations of total cholesterol (−7%), LDL-C (−11%), and HDL-C (−20%) and significantly higher concentrations of small, dense LDL-C (61%) and TG (62%). There was no difference in apoA-I concentration between cases and controls but the apoA-I-containing HDL subpopulation profiles were different. Cases had significantly higher concentrations of the small preβ-1 (87%) and α-3 (22%) particles and significantly lower concentration of the large α-1 particles (−28%). There was no correlation between ABCA1-dependent CEC and HDL-C (R²=0.001) and a weak correlation between ABCA1-dependent CEC and apoA-I concentrations (R²=0.110) (Figure 1). ABCA1-dependent CEC was strongly positively correlated with preβ-1 concentration (R²=0.535) but not with the concentration of any other HDL particle (Figure 1 and Supplemental Figure II). Even though the mean preβ-1 concentration was 87% higher, ABCA1-dependent CEC was only 30% higher in the CHD group compared to the control group. Preβ-1-normalized CEC (a metric of preβ-1 functionality) was calculated for each subject by dividing ABCA1-dependent CEC values with preβ-1 concentrations. Cases had significantly (31%) lower preβ-1 functionality than controls. Figure 2 demonstrates that there was large biological variability in ABCA1-dependent CEC in both cases and controls and a large overlap prior to normalization for preβ-1 concentration (left panel). However, the inter-individual variability in preβ-1 functionality was much less in cases than in controls (right panel). There were positive correlations between TG level and ABCA1-dependent CEC (R²=0.317) and preβ-1 level (R²=0.345), but not preβ-1 functionality (R²=0.01) (Figure 3). TG concentration was associated with the concentrations of small HDL particles, preβ-1 (R²=0.345) and α-3 (R²=0.254) (Supplemental Figure III). Moreover, there was a positive correlation between the concentrations of preβ-1 and α-3 particles (R²=0.329). In Supplemental Table I, we compared the measured parameters by gender. Mean preβ-1 concentration and ABCA1-dependent CEC were significantly higher in cases compared to controls in both genders (p<0.01). Preβ-1 functionality was lower in cases compared to controls (males 0.38 vs. 0.61, p<0.01; females 0.53 vs. 0.57, ns).

Table 1.

Characteristics of the study groups

	Control	CHD	P
Male; Female	70; 30	70; 30
Age (year)	55 ± 16	61 ± 7
BMI (kg/m2)	NA	31.1 ± 3.2
TC (mmol/L)	4.3 ± 0.8	4.0 ± 0.9	0.012
LDL-C (mmol/L)	2.5 ± 0.6	2.2 ± 0.7	1.1×10−8
sdLDL-C (mmol/L)	0.5 ± 0.3	0.8 ± 0.4	0
HDL-C (mmol/L)	1.5 ± 0.3	1.2 ± 0.4	1.6×10−7
TG (mmol/L)	1.0 ± 0.4	1.6 ± 0.8	5.0×10−9
apoB (mg/dL)	81 ± 17	79 ± 22	0.388
apoA-I (mg/dL)	156 ± 23	156 ± 33	0.978
Preβ-1 (mg/dL)	8.0 ± 3.2	15.1 ± 6.9	2.4×10−14
α-1 (mg/dL)	34.0 ± 11.9	24.8 ± 13.7	1.3×10−6
α-2 (mg/dL)	63.7 ± 11.6	61.6 ± 15.1	0.292
α-1+α-2 (mg/dL)	96.4 ± 19.7	86.4 ± 26.7	0.001
α-3 (mg/dL)	21.5 ± 3.9	26.0 ± 4.7	1.3×10−11
α-4 (mg/dL)	16.0 ± 4.2	16.8 ± 3.8	0.025
Global efflux (%)	10.3 ± 2.2	11.4 ± 2.9	0.029
ABCA1-efflux (%)	4.4 ± 1.6	5.9 ± 2.3	1.3×10−5
Preβ-1-normalized ABCA1 efflux*	0.59 ± 0.23	0.43 ± 0.20	7.9×10−7
SR-BI-efflux (%)	3.4 ± 0.7	3.6 ± 0.8	0.270
Large-HDL-normalized SR-BI efflux†	0.035 ± 0.004	0.043 ± 0.007	4.7×10−15
HDL anti-oxidative capacity‡	3.4 ± 0.6	2.9 ± 0.5	2.5×10−7
hsCRP (mg/L)	1.9 ± 2.5	2.5 ± 4.2	0.284
SAA (mg/L)	29.3 ± 19.5	30 ± 20	0.178
Fibrinogen (mg/dL)	329 ± 68	453 ± 133	1.7×10−13
MPO (pMol/L)	303 ± 199	535 ± 437	4.0×10−6

Data are expressed as mean ± SD.

Differences between cases and controls were calculated after adjusting data for age.

^*Preβ-1-normalized ABCA1-efflux (preβ-1-particle functionality) was calculated by dividing ABCA1-dependent cholesterol efflux (%) with preβ-1 concentration (mg/dL).

^†Large-HDL-normalized SR-BI-efflux (large-HDL-particle functionality) was calculated by dividing SR-BI-dependent cholesterol efflux (%) with the combined concentration of α-1+α-2 (mg/dL).

^‡Anti-oxidative capacity was measured in 64 cases and 64 gender-matched controls and is expressed in arbitrary units.

Abbreviations: BMI body mass index, NA not available, TC total cholesterol, LDL-C low-density lipoprotein cholesterol, sd small dense, HDL-C high-density lipoprotein cholesterol, TG triglycerides, apo apolipoprotein, ABCA1 ATP-binding cassette transporter A1, SR-BI Scavenger receptor class B type I, hsCRP high-sensitivity C-reactive protein, SAA serum amyloid A, MPO myeloperoxidase

Figure 1.

Linear regression analysis on ABCA1-dependent cell-cholesterol efflux capacity with HDL-C, apoA-I, and preβ-1-HDL particle levels (n=200).

Figure 2.

Box-plot analysis of unadjusted and adjusted (preβ-1-concentration-normalized) data on cell-cholesterol efflux capacity via the ABCA1 pathway.

Figure 3.

Linear regression analyses on ABCA1-dependent cell-cholesterol efflux capacity, preβ-1, and preβ-1 functionality with TG (n=200).

Compared to controls, cases had significantly lower (−12%) concentrations of large (α-1+α-2) HDL particles but no difference in CEC via the SR-BI pathway. Cases had significantly higher (22%) large-HDL-concentration-normalized SR-BI efflux (a metric of large-HDL-particle functionality). Strong positive correlations were observed between SR-BI-dependent CEC and the levels of HDL-C (R²=0.564), apoA-I (R²=0.685), and large (α-1+α-2) HDL particles (R²=0.712) (Figure 4). There were strong correlations between SR-BI-dependent efflux and the large lipid-rich HDL particles (α-1 R²=0.542, α-2 R²=0.618) but not the smaller lipid-poor HDL particles (Supplemental Figure IV). There was large biological variability in SR-BI-dependent efflux in both the normalized and non-normalized models (Figure 5). Compared to control males, control females had significantly higher SR-BI-dependent CEC due to higher levels of large HDL particles, but there was no gender difference in large-HDL-particle functionality (Supplemental Table I).

Figure 4.

Linear regression analysis on SR-BI-dependent cell-cholesterol efflux capacity with HDL-C, apoA-I, and large-HDL particle concentrations (n=200).

Figure 5.

Box-plot analysis of unadjusted and adjusted (large-HDL-particle-normalized) data on cell-cholesterol efflux via the SR-BI pathway

Data on BMI were available only in cases; therefore, we investigated the relationship between BMI and ABCA1- and SR-BI-dependent CEC only in this group (Supplemental Figure V). There were no significant correlations between BMI and either ABCA1-dependent CEC (R²=0.002) or preβ-1 functionality (R²=0.0003). There was a weak inverse correlation between BMI and SR-BI-dependent CEC (R²=−0.043) and a weak positive correlation between BMI and the functionality of large (α1+α2) HDL particles (R²=0.034).

The mean level of HDL anti-ox capacity was significantly lower (−16%) in CHD cases compared to controls. Cases had higher levels of inflammatory- (C-reactive Protein, SAA, and fibrinogen) and oxidative-stress (MPO) markers than controls, but only fibrinogen and MPO were significantly different. Linear regression analysis indicated weak inverse correlations between preβ-1 functionality and the concentrations of SAA (R²=0.021), fibrinogen (R²=0.046), and MPO (R²=0.031), and weak positive correlations between large-HDL-particle functionality and the concentrations of fibrinogen (R²=0.179) and MPO (R²=0.030) (Supplemental Figure VI). The correlations between HDL anti-ox capacity and the concentrations of SAA (R²=0.002), fibrinogen (R²=0.056), and MPO (R²=−0.013) were weak and inconsistent (Supplemental Figure VII). There was a positive correlation between HDL anti-ox capacity and preβ-1 functionality (R²=0.135) and an inverse correlation between HDL anti-ox capacity and large-HDL-particle functionality (R²=0.117) (Supplemental Figure VIII).

DISCUSSION

ABCA1 plays a crucial role in cholesterol efflux from macrophages, and small lipid-poor preβ-1 particles are the primary acceptors of cholesterol via this pathway^(15,16,18). We have demonstrated that although CHD patients have higher preβ-1 concentration, the functionality of their preβ-1 particles is reduced compared to controls. ABCA1-dependent CEC was strongly correlated with the concentrations of preβ-1 particles, but very weakly with the levels of HDL-C, apoA-I, and other HDL particles (Figure 1 and Supplemental Figure II). We have found that the 87% higher preβ-1 level in CHD patients was associated with only a 30% higher ABCA1-dependent CEC as compared to controls (Table 1). Most importantly, we have shown that preβ-1 functionality was significantly lower (−31%) and less variable in cases as compared to controls indicating that preβ-1 particles were altered in a similar way in cases (Figure 2). We corroborated previous findings^(15,16) that SR-BI-dependent CEC was strongly correlated with the concentrations of HDL-C, apoA-I, and the highly-lipidated large HDL particles (α-1+α-2) (Figure 4). In contrast to preβ-1 particles, the concentration of large-HDL particles was significantly lower in CHD cases compared to controls, but their functionality was significantly higher (Table 1). These findings clearly demonstrate that ABCA1-dependent and SR-BI-dependent CEC are influenced by both the concentration and the functionality of the HDL particles participating in these efflux pathways.

The mechanisms that alter ABCA1-dependent efflux in CHD cases are not well understood. There are a number of possibilities, some of which were addressed by this study and many more which are yet to be investigated. Since the cells used in the efflux analyses are constant, the variability in ABCA1-dependent CEC must relate to variability in the concentration, composition, and physicochemical properties of preβ-1 particles. As the influence of particle concentration on CEC was eliminated by the normalization, the variability in CEC can only be explained by differences in the composition and/or physicochemical properties of preβ-1 particles. Linear regression analysis showed better correlation between preβ-1 concentration and ABCA1-dependent CEC in CHD patients (R²=0.481) as compared to controls (R²=0.312) suggesting that specific alteration(s) generated morphologically more similar, although less functional, preβ-1 particles in cases. Precisely how these particles are altered and why those alterations reduced their functionality requires further exploration.

One initial proposed mechanism was that low preβ-1 functionality in CHD patients was the result of high levels of recycled preβ-1 particles due to elevated TG levels and increased cholesteryl ester transfer protein (CETP) and hepatic lipase activities. Preβ-1 particles are formed in two different ways: 1) de novo preβ-1 particles form from newly synthesized apoA-I; 2) recycled preβ-1 particles are generated during the remodeling of mature large HDL particles. An elevated TG level, which is common in CHD patients, increases CETP and hepatic lipase activities. CETP exchanges cholesterol for TG between TG-rich lipoproteins and large HDL particles^(24,25). Hepatic lipase digests TG in large HDL particles and, as their core shrinks, some surface lipids and apoA-I are shed and are recycled as preβ-1 particles⁽²⁶⁾. At that time, we assumed that recycled preβ-1 particles had altered composition with low functionality. However, there was no association between preβ-1 functionality and TG levels in this study (Figure 3). Moreover, in previous studies, positive associations have been shown between TG levels and preβ-1 functionality in subjects with high TG levels without CHD as well as between TG levels and CEC^(16,27,28). Therefore, we concluded that an increased level of recycled preβ-1 particles, as a consequence of elevated TG level, was not the cause of low preβ-1 functionality in CHD patients.

Another possibility was that inflammatory and oxidative stress damage HDL and impair reverse cholesterol transport^(29–31). It is well documented that MPO oxidizes apoA-I⁽³²⁾. In this study, CHD patients had significantly increased MPO levels compared to controls but the influence of MPO level on preβ-1 functionality was weak (R²=0.031) (Supplemental Figure VI). Moreover, neither in this study nor in an earlier study, have we found significant associations between increased levels of MPO, SAA and fibrinogen and preβ-1 functionality⁽¹⁶⁾, therefore, we excluded high levels of circulating inflammatory- and oxidative-stress markers as significant players in the decreased preβ-1 functionality in CHD patients.

We have also investigated how HDL anti-ox capacity, a key athero-protective function of HDL⁽³³⁾, was associated with HDL CEC. HDL anti-ox capacity was significantly lower (−16%) in cases compared to controls; however, it is worth noting that none of the participants had pro-oxidative HDL (anti-ox capacity <1.0). The relationships between HDL anti-ox capacity and HDL-particle functionality were equivocal. HDL anti-ox capacity was associated positively with preβ-1 functionality (R²=0.135) and inversely with large-HDL-particle functionality (R²=0.117) (Supplemental Figure VIII). Moreover, there were no significant correlations between HDL anti-ox capacity and the concentrations of inflammatory- and oxidative-stress markers (Supplemental Figure VII).

We have investigated the influence of BMI on ABCA1- and the SR-BI-dependent CEC in the CHD group where BMI data were available. We found no correlation between BMI and either ABCA1-dependent CEC or preβ-1 functionality (Supplemental Figure IV). There is no consensus on the influence of BMI on CEC. Attia et al. reported increased ABCA1-dependent CEC in 12 overweight to obese women compared to 9 normal-weight women and a significant positive correlation between BMI and ABCA1-dependent CEC⁽⁴²⁾. In contrast, Talbot et al. reported 9% lower ABCA1-dependent CEC in 52 obese men compared to 25 normal-weight men and a significant inverse correlation between BMI and ABCA1-dependent CEC⁽⁴³⁾. We believe that BMI has no direct effect on ABCA1-dependent CEC, but may have an indirect influence on it by modifying the concentrations of TG and/or HDL particles.

We assume that the physicochemical properties of apoA-I in preβ-1 particles are altered in CHD patients compared to controls and are responsible for the reduced ABCA1-dependent CEC. Preliminary data generated in our laboratory indicate that the lipid composition of preβ-1 is more complex than previously thought and may be different between cases and controls. We also think that the higher functionality of large-HDL particles in cases compared to controls may partially be caused by the elevated TG levels. Generally, increased plasma TG level is associated with increased CETP activity, which in turn decreases cholesteryl ester concentration and increases the phospholipid/cholesterol ratio in large HDL particles. These alterations make large-HDL particles better acceptors of cell cholesterol. However, hepatic lipase remodels the TG-enriched large-HDL particles into smaller HDL particles⁽¹⁶⁾. In our view, high TG level induces the formation of more functional large HDL particles, but simultaneously enhances the degradation of these particles.

Our findings are in line with previous ones showing that drug-induced changes in preβ-1 concentration correlate significantly and positively with changes in ABCA1-dependent CEC^(34–36). As CHD patients have significantly higher preβ-1 levels than controls^{(8,9,19–21)} and there is a strong positive correlation between preβ-1 level and CEC, one would expect to see elevated CEC in cases compared to controls. Our findings are in line with this assumption. However, others reported lower CEC in CHD cases compared to controls^(11–13). One reason for this discrepancy is that the nearly 2-fold higher preβ-1 level in cases may have almost perfectly offset the decreased preβ-1 functionality as seen in this study. Other reasons may be related to differences in the study populations or in lipid-modifying interventions. It has been documented that both statins and niacin decrease preβ-1 concentration in a drug- and dose-specific way^(37,38). However, there is no consensus on the effects of lipid-modifying interventions on CEC. Studies have reported either an increase, a decrease, or no change in ABCA1-dependent CEC after statin or niacin treatment^{(27,28,39–41)}. Moreover, it is possible that lipid-modifying interventions influence the concentration and the functionality of HDL particles independently.

There are several limitations to this study. Since all cases were on statins but none of the controls, we could not dissect the drug effects from the effects of CHD prevalence on HDL CEC. Further limitations include the relatively low number of subjects and the inability to adjust for all potential confounders (e.g. BMI, smoking, and diabetes). Due to these limitations, our results might not be comparable to other populations having different pathological conditions and receiving different medications. A significantly larger and better-characterized population is needed to dissect the effects of CHD from other parameters on HDL CEC.

The strength of this study is that we were able to determine not only the efflux capacity of HDL, but also the functionality of key HDL particles that mediate cell-cholesterol efflux. However, further studies are needed to determine the physicochemical and compositional properties of preβ-1 HDL particles to explain why they are dysfunctional in CHD patients.

Conclusions

CHD patients had higher than normal preβ-1 particle concentrations but those particles had lower than normal CEC reducing their maturation into larger HDL particles. We excluded high plasma levels of TG, MPO, and fibrinogen as significant contributors to the decreased preβ-1 functionality in CHD patients. Cases had lower than normal concentrations of α-1+α-2 particles but those particles had higher CEC compared to controls indicating that the functionality of large-HDL particles was not compromised in CHD patients. Our data clearly support the concept that reverse cholesterol transport (the sum of HDL’s cell-cholesterol efflux and transport capacities from peripheral cells to the bile) is influenced not only by the concentrations of individual HDL particles but also by their functional properties. These properties are altered in CHD cases compared to controls. We strongly believe that measurement of both the concentration and the functionality of HDL particles, involved in cell-cholesterol efflux, is critical for better understanding HDL’s role in CHD and in designing interventions that increase not only the concentration, but also the functionality of HDL.

Highlights.

• Preβ-1 HDL is responsible for about 54% of ABCA1-dependent cell-cholesterol efflux.

• Large HDL particles (α-1+α-2) are responsible for about 72% of SR-BI-dependent cell cholesterol efflux.

• CHD patients have elevated preβ-1 levels (87%) but lower preβ-1 functionality (−31%) compared to controls.

• CHD patients have lower concentrations of large-HDL particles (−12%) but higher large-HDL particle functionality (22%) compared to controls.

• HDL cell-cholesterol efflux capacity is influenced by both the concentrations and the functionality of HDL particles; therefore, to better understand the roles of HDL in health and disease, both HDL properties should be measured.

Abbreviations

anti-ox

Anti-oxidative

apo

Apolipoprotein

ABCA1

ATP-binding cassette transporter A1

BMI

body mass index

CEC

Cell cholesterol efflux capacity

CETP

Cholesteryl ester transfer protein

CHD

Coronary heart disease

HDL

High-density lipoprotein

HDL-C

High-density lipoprotein cholesterol

LDL-C

Low-density lipoprotein cholesterol

MPO

Myeloperoxidase

SR-BI

Scavenger receptor class B type I

SAA

Serum Amyloid A

TG

Triglyceride