Skip main navigation

Fenofibrate Reduces Plasma Cholesteryl Ester Transfer From HDL to VLDL and Normalizes the Atherogenic, Dense LDL Profile in Combined Hyperlipidemia

Originally published, Thrombosis, and Vascular Biology. 1996;16:763–772


    Abstract The effect of fenofibrate on plasma cholesteryl ester transfer protein (CETP) activity in relation to the quantitative and qualitative features of apoB- and apoA-I–containing lipoprotein subspecies was investigated in nine patients presenting with combined hyperlipidemia. Fenofibrate (200 mg/d for 8 weeks) induced significant reductions in plasma cholesterol (−16%; P<.01), triglyceride (−44%; P<.007), VLDL cholesterol (−52%; P=.01), LDL cholesterol (−14%; P<.001), and apoB (−15%; P<.009) levels and increased HDL cholesterol (19%; P=.0001) and apoA-I (12%; P=.003) levels. An exogenous cholesteryl ester transfer (CET) assay revealed a marked decrease (−26%; P<.002) in total plasma CETP-dependent CET activity after fenofibrate treatment. Concomitant with the pronounced reduction in VLDL levels (37%; P<.005), the rate of CET from HDL to VLDL was significantly reduced by 38% (P=.0001), whereas no modification in the rate of cholesteryl ester exchange between HDL and LDL occurred after fenofibrate therapy. Combined hyperlipidemia is characterized by an asymmetrical LDL profile in which small, dense LDL subspecies (LDL-4 and LDL-5, d=1.039 to 1.063 g/mL) predominate. Fenofibrate quantitatively normalized the atherogenic LDL profile by reducing levels of dense LDL subspecies (−21%) and by inducing an elevation (26%; P<.05) in LDL subspecies of intermediate density (LDL-3, d=1.029 to 1.039 g/mL), which possess optimal binding affinity for the cellular LDL receptor. However, no marked qualitative modifications in the chemical composition or size of LDL particles were observed after drug treatment. Interestingly, the HDL cholesterol concentration was increased by fenofibrate therapy, whereas no significant change was detected in total plasma HDL mass. In contrast, the HDL subspecies pattern was modified as the result of an increase in the total mass (11.7%) of HDL2a, HDL3a, and HDL3b (d=1.091 to 1.156 g/mL) at the expense of reductions in the total mass (−23%) of HDL2b (d=1.063 to 1.091 g/mL) and HDL3c (d=1.156 to 1.179 g/mL). Such changes are consistent with a drug-induced reduction in CETP activity. In conclusion, the overall mechanism involved in the fenofibrate-induced modulation of the atherogenic dense LDL profile in combined hyperlipidemia primarily involves reduction in CET from HDL to VLDL together with normalization of the intravascular transformation of VLDL precursors to receptor-active LDLs of intermediate density.

    The atherogenic dyslipidemia CHL is characterized by elevated levels of plasma VLDL-C, TGs, and LDL-C.12 The metabolic abnormalities that underlie this disorder involve increased production rates of VLDL apoB and/or LDL apoB34 ; in contrast, the fractional catabolic rates of LDL are normal.345 In addition, plasma concentrations of HDL-C are reduced,6 while plasma CETP mass and activity are elevated.78 Furthermore, the LDL subspecies profile in CHL is dominated by small, dense LDL particles that are present in plasma at concentrations as much as threefold greater than those typical of normolipidemic subjects.69 Dense LDL particles possess a low binding affinity for the cellular LDL receptor10 and display a prolonged residence time in plasma in vivo.1112 Such dense LDL are therefore exposed to a greater degree to biological modification,61314 which results in their catabolism by atherogenic pathways, such as those represented by the scavenger receptors of monocyte-derived macrophages.13 These observations are consistent with the hypothesis that dense LDL particles are of elevated atherogenicity and contribute significantly to the premature vascular disease typical of CHL.

    Fibric acid and its derivatives are widely used in the therapeutic treatment of the atherogenic dyslipidemias, particularly CHL.1516 These compounds are known to inhibit hepatic VLDL synthesis and secretion,17 to promote the lipolysis of TG-rich lipoproteins through stimulation of lipoprotein lipase activity,1819 and to promote catabolism of LDL particles via the LDL receptor–mediated pathway.12202122 Evidence suggests that fibrates increase excretion of hepatic cholesterol in bile and that endogenous hepatic cholesterol synthesis may be decreased.23

    A second-generation fibric acid derivative, fenofibrate, has hypolipidemic effects on plasma lipid parameters in CHL.24252627 Plasma cholesterol and TG levels are typically reduced by 10% to 30% and 30% to 67%, respectively; equally, VLDL-C and LDL-C are decreased by 30% to 70% and 2% to 29%, respectively, whereas HDL-C is increased by up to 26%. In addition, an augmentation of 3% to 38% in plasma apoA-I levels and a reduction of 10% to 37% in apoB concentrations have been reported.24252627 A major feature of such hypolipidemic action involves modulation of the de novo synthesis, intravascular remodeling, and cellular degradation of plasma lipoproteins.912161720 More specifically, fenofibrate not only reduces absolute circulating concentrations of VLDL but also induces a reduction in the size of VLDL particles of hepatic origin; as a consequence of the intravascular remodeling of such VLDLs, receptor-active LDLs are formed, thereby resulting in an increase in the proportion of plasma LDLs that are catabolized by the nonatherogenic LDL receptor pathway.1220 These latter effects have been demonstrated in vivo both in hypertriglyceridemic and hypercholesterolemic patients.1220 It must be emphasized, however, that the precise mechanisms that underlie the fenofibrate-mediated enhanced formation of LDL receptor–active LDL subspecies are as yet undetermined, although some evidence for drug-induced changes in LDL chemical composition and particle size has been provided.20

    There is a paucity of information on the principal features of CET between donor and acceptor lipoproteins in CHL and on the potential contribution of such transfer to the atherogenic lipoprotein profile characteristic of CHL patients. As it is now established that CETP-facilitated transfer and exchange of neutral lipids between HDL and apoB-containing lipoproteins (VLDL, IDL, and LDL) play a major role in the determination of the apoA-I– and apoB-containing lipoprotein subspecies profile,282930 we hypothesized that the CET to LDL was deficient in CHL patients as a result of the elevated concentrations and high CE acceptor activity of VLDL. We therefore investigated the effect of fenofibrate on CET and exchange between VLDL, LDL, and HDL to assess the potential role of neutral lipid transfer in the fenofibrate-induced formation of receptor-active LDLs in CHL. In parallel, we evaluated the effect of fenofibrate therapy on the quantitative and qualitative features of apoB- and apoA-I–containing lipoprotein subspecies in CHL patients. We observed that fenofibrate significantly reduced CET from HDL to VLDL as a result of reductions in both plasma CETP-dependent CET activity and VLDL concentrations. Moreover, fenofibrate normalized the LDL subspecies profile by inducing a shift from denser LDL toward receptor-active LDL particles of intermediate density. Our data indicate that the modulation of CETP activity is a major factor in the correction of the intravascular metabolism of apoB-containing lipoproteins during drug therapy.


    CHL Patients

    Two major groups of subjects have increased circulating concentrations of LDL, ie, those with familial hypercholesterolemia and those with a combination of hypercholesterolemia and hypertriglyceridemia, the latter disorder being termed CHL. Individuals in the latter group are distinguished by elevated levels of both LDL and VLDL. As there is to date no single genetic or clinical marker allowing identification of CHL patients, our classification is similar to that proposed by Arad et al,1 ie, a lipid profile including an LDL-C level >90th percentile for age and sex with a concomitant plasma TG level ≥150 mg/dL.

    Approval by the human subjects review committee of the hospital and informed consent from all participants were first obtained. CHL patients (n=9) with fasting plasma levels of total cholesterol >200 mg/dL and TGs >150 mg/dL were selected and taken off all lipid-lowering agents 6 weeks before the initial blood sampling. All subjects were nondiabetic. Two of the 9 patients had previously presented with premature coronary heart disease; 5 of the subjects were smokers and 4 were nonsmokers. The clinical characteristics of individual CHL patients are presented in Table 1. Secondary hyperlipidemia was excluded by clinical examination and by the determination of hepatic enzyme activities, the plasma levels of creatinine, glucose, thyroid hormones, and blood proteins, and by evaluation of proteinuria. Four patients had the apoE3/E2 heterozygous phenotype.

    Blood Samples

    Blood samples were obtained by venipuncture after an overnight fast and were placed in sterile EDTA-containing tubes (final concentration, 1 g/L) at the time of inclusion into the study and after 8 weeks of fenofibrate treatment (200 mg/d micronized fenofibrate). Plasma was separated by low-speed centrifugation at 4°C. EDTA (0.1 g/L), sodium azide (0.01%), and gentamicin (0.005%) were added to plasma to chelate metal cations and to inhibit microbial growth.

    ApoE Phenotyping by Isoelectric Focusing

    Plasma samples (10 μL) were delipidated with 2 mL acetone-ethanol (1:1, vol/vol) at −20°C overnight. Precipitated protein was sedimented by centrifugation (15 minutes, 2500 rpm), resuspended in the same volume of diethyl ether, stored at −20°C for 1 hour, and pelleted by centrifugation under the same conditions. After drying, protein pellets were solubilized in 1 mL of 10 mmol/L Tris-HCL buffer, pH 8.6, containing 8 mol/L urea and 10 mmol/L dithiothreitol. Samples were electrophoresed into a gel consisting of 7.5% polyacrylamide, 8 mol/L urea, and 2.8% pharmalyte (pH 4 to 6.5; Pharmacia). Isoelectric focusing was performed at 200 V for 3 hours, after which plasma proteins were electrophoretically blotted onto nitrocellulose sheets for 1 hour at 90 V in a transfer buffer containing 25 mmol/L Tris, 192 mmol/L glycine, and 20% methanol. The nitrocellulose paper was then blocked with 5% nonfat milk in PBS buffer for 30 minutes at room temperature, after which it was probed for 1 hour with a sheep anti-human apoE antibody diluted 1:1000 in PBS buffer containing 0.1% bovine serum albumin. After washing three times in PBS, the blots were subsequently incubated for 15 minutes with a rabbit anti-sheep peroxidase-conjugated IgG diluted 1:15 000 in PBS. Detection was performed by chemiluminescence (ECL, Amersham). Identification of apoE isoforms was determined by reference to an apoE3/E3 standard, the kind gift of Dr K.H. Weisgraber.

    Isolation of Plasma Lipoproteins

    Lipoproteins were isolated from plasma by density gradient ultracentrifugation with a Beckman SW41 Ti rotor at 40 000 rpm for 44 hours at 15°C by a slight modification of the method of Chapman et al.31 Briefly, plasma density was increased to 1.21 g/mL by adding dry solid KBr. A discontinuous density gradient was then constructed as follows: 2 mL of a NaCl/KBr solution (d=1.24 g/mL), 3 mL plasma (d=1.21 g/mL), 2 mL of a solution (d=1.063 g/mL), 2.5 mL of a solution (d=1.019 g/mL), and 2.5 mL of a NaCl solution (d=1.006 g/mL). All density solutions contained sodium azide (0.01%), EDTA (0.01%), and gentamicin (0.005%), pH 7.4. After centrifugation, gradients were collected from the top of the tube in 12 fractions corresponding to VLDL (d<1.017 g/mL), IDL (d=1.018 to 1.019 g/mL), LDL-1 (d=1.019 to 1.023 g/mL), LDL-2 (d=1.023 to 1.0239 g/mL), LDL-3 (d=1.029 to 1.039 g/mL), LDL-4 (d=1.039 to 1.050 g/mL), LDL-5 (d=1.050 to 1.063 g/mL), HDL2b (d=1.063 to 1.091 g/mL), HDL2a (d=1.091 to 1.110 g/mL), HDL3a (d=1.110 to 1.133 g/mL), HDL3b (d=1.133 to 1.156 g/mL), and HDL3c (d=1.156 to 1.179 g/mL). Density limits were taken from a standard curve of density versus volume derived from control gradients containing only salt solutions.31 All lipoprotein fractions were analyzed for their lipid and protein content. Recovery of cholesterol was 98% and between 90% and 95% for all other lipids. Comparison of lipoprotein mass profiles as a function of density, computed from chemical analyses of lipoprotein subfractions isolated from aliquots of the same plasma fractionated in separate tubes during the same ultracentrifugal run, revealed a high degree of reproducibility; indeed, mass profiles were indistinguishable, with coefficients of variation of <5% for the masses of individual subfractions.31

    Lipid and Protein Analyses

    Lipids in plasma or in isolated lipoprotein fractions were quantified enzymatically with Bio-Merieux kits (69280) for TGs and phospholipids and Boehringer Mannheim kits (38240) for total and free cholesterol. CE mass was calculated as (Total Cholesterol−Free Cholesterol)×1.67 and thus represents the sum of the esterified cholesterol and fatty acid moieties. Protein was measured by the bicinchoninic acid assay reagent (Pierce). Lipoprotein mass corresponded to the sum of the mass of the individual components for each lipoprotein fraction. The coefficients of variation (interassay and intra-assay) for our chemical analyses of cholesterol, TGs, phospholipids, and protein were 1.9%, 3.6%, 3.8%, and 4.6%, respectively.

    Estimation of LDL Subspecies Size

    LDL particle size was evaluated by electrophoresis in nondenaturing polyacrylamide gradient gels (2% to 16%; Pharmacia Fine Chemicals) in the GE 2/4 LS gel electrophoresis apparatus.63233 A set of standard proteins with known hydrated diameters (latex beads, 380 Å; thyroglobulin, 170 Å; apoferritin, 122 Å; and catalase, 104Å) was run on each slab as a reference marker. From the migration distances of the different LDL subfractions (LDL-1 through LDL-5) and those of the calibration proteins, it was possible to calculate the Stokes’ diameters of the LDL subspecies with the Stokes-Einstein equation.34 The correlation coefficient for the regression line of the relationship between the logarithm of the diameter of the calibration proteins and their migration distance was typically >.98. The intra-assay coefficient of variation for LDL sizing analyses was <2%.

    Preparation of [3H]CE-Labeled HDL

    Radiolabeled HDLs were obtained from the d>1.063 g/mL fraction of plasma by ultracentrifugation at 100 000 rpm for 3 hours and 30 minutes at 15°C with a Beckman TL100 centrifuge.2835 The d>1.063 g/mL fraction was incubated for a total of 18 hours at 37°C in the presence of 4 μCi [1,2,6,7-3H]cholesterol in an ethanol solution (specific activity, 71 Ci/mmol) to allow endogenous lecithin:cholesterol acyltransferase to esterify the radioactive cholesterol.35 HDL-containing esterified radiolabeled CE was then isolated by adjusting the density to 1.21 g/mL by adding dry solid KBr followed by centrifugation at 100 000 rpm for 5 hours and 30 minutes at 15°C. Radiolabeled-HDL preparations displayed a specific radioactivity that ranged from 10 000 to 15 000 cpm/μg CE; >95% of the total radioactive free cholesterol added was transformed into labeled CE.35

    CETP-Dependent CET Assay

    To estimate the maximal transfer activity in plasma, the CETP-dependent esterified cholesterol assay described by Ahnadi et al36 was employed. Briefly, the density of a 1-mL plasma sample was adjusted to 1.21 g/mL by adding dry solid KBr. After ultracentrifugation for 6 hours at 100 000 rpm at 15°C, plasma lipoproteins were removed, and the bottom fraction was extensively dialyzed in Spectrapor membrane tubing at 4°C against a buffer containing 150 mmol/L NaCl, 10 mmol/L Tris base, 1 mmol/L EDTA, and 1 mmol/L sodium azide, pH 7.4. After dialysis, the volume of the bottom fraction was adjusted to 1 mL by adding buffer and was used as a source of CETP. Labeled HDL CE (200 nmol) was mixed with 800 nmol VLDL/LDL CE in the presence or absence of 400 μL of the CETP-containing bottom fraction (final volume, 1 mL). After 0, 0.5, 1, 2, and 3 hours of incubation at 37°C, a 150-μL sample was withdrawn and immediately chilled on ice. Lipoproteins were then precipitated by the phosphotungstate–magnesium chloride procedure. The VLDL/LDL pellets were dissolved in 100 μL of 0.5 mol/L Na2CO3, and their radioactivity content was determined. The facilitated transfer of CE from HDL to VLDL/LDL was calculated from the difference between the radioactivity transferred in the presence or absence of CETP. The interassay and intra-assay coefficients of variation for the CETP-dependent CET assay were <2.6%.

    Measurement of CET and Exchange Between HDL and ApoB-Containing Lipoproteins

    Radiolabeled HDL (equivalent to 1% of the total HDL CE mass present in 1 mL of a subject’s plasma) and iodoacetamide (to inhibit endogenous lecithin:cholesterol acyltransferase activity; final concentration, 1.5 mmol/L) were added to each subject’s plasma. Aliquots of 1 mL were incubated for 4 hours at 0°C or 37°C. After incubation, lipoproteins were isolated by density gradient ultracentrifugation as described above, and the radioactive content and lipid composition of individual density fractions were determined. Radioactivity was quantified by a Pharmacia Rack Beta 509 for liquid scintillation spectrometry. The recovery of radioactivity was >98% in all experiments. Measurements of CET and exchange between HDL and the apoB-containing lipoproteins were performed in triplicate. Net mass transfer of CE represents the net augmentation or decrease in the CE mass content of individual lipoprotein fractions. Measurement of radioactivity alone is indicative of CET plus exchange. Thus, exchange of CE between lipoprotein particles was calculated as the difference between the mass of labeled CE transferred, which was calculated from the known specific radioactivity of HDL CE and the net mass transfer of CE as determined enzymatically.35 The interassay and intra-assay coefficients of variation for the determination of CET and the exchange between HDL and apoB-containing lipoproteins were <5%.

    Determination of CE Exchange and Transfer Rates Between HDL and ApoB-Containing Lipoproteins

    Radiolabeled HDL (equivalent to 1% of the total HDL CE mass present in 1 mL of a subject’s plasma) and iodoacetamide (to inhibit endogenous lecithin-cholesterol acyltransferase activity; final concentration, 1.5 mmol/L) were added to 3 mL of each subject’s plasma. After 0, 0.5, 1, 2, 3, and 4 hours of incubation at 37°C, aliquots of 0.5 mL were withdrawn and immediately chilled on ice. Plasma lipoproteins were then isolated by sequential ultracentrifugation at 100 000 rpm at 15°C. By this method, we successively isolated the VLDL+IDL (d<1.019 g/mL), LDL (d=1.019 to 1.063 g/mL), and HDL (d=1.063 to 1.21 g/mL) fractions by centrifuging for 1.5, 3.5, and 5.5 hours, respectively. The radioactive content and lipid composition of each individual fraction were determined as described above. The kinetics of CE mass transferred, determined on the basis of radioactivity or by measuring CE mass enzymatically, were linear (r=.99) during the initial 4-hour period of incubation.2835 The rate of CET represents the slope of the individual curves and is expressed in micrograms of CE mass transferred per hour per milliliter of plasma for the rate of net CE mass transfer and in micrograms of labeled CE transferred per hour per milliliter of plasma for the rate of CET+exchange. The difference between these two determinations of CET rates represents the rate of CE exchange between HDL and the apoB-containing lipoproteins. The interassay coefficient of variation for the rates of CET and exchange was <5%.

    Statistical Analysis

    The effects of fenofibrate on plasma lipid levels, on the plasma concentrations and chemical compositions of lipoprotein particles, and on CET between HDL and apoB-containing lipoproteins were determined by comparing these parameters at the time of inclusion into the study with those after 8 weeks of drug therapy with Student’s paired t test.


    Effects of Fenofibrate on Plasma Lipid and Apolipoprotein Parameters

    Plasma lipid and apolipoprotein levels in nine CHL patients before and after 8 weeks of fenofibrate treatment (200 mg/d) are shown in Table 1. At inclusion, all CHL patients displayed plasma lipid levels characteristic of CHL as defined by Arad et al,1 ie, a lipid profile including an LDL-C level greater than the 90th percentile for age and sex with a concomitant plasma TG level ≥150 mg/dL. Drug therapy significantly lowered plasma cholesterol and TG levels, by 16% (P<.02) and 44% (P<.007), respectively; indeed, all nine CHL patients displayed a reduction in their plasma cholesterol and TG levels. However, marked differences in the response to fenofibrate treatment were observed among the subjects (the percent change in cholesterol levels ranged from −3% to −37% and in TG levels, from −17% to −64%). Such variations in individual response to drug therapy almost certainly reflect the underlying genetic heterogeneity in the patient population. After fenofibrate treatment we observed significant reductions in plasma VLDL-C (−52%; P=.01), LDL-C (−14%; P<.001), and apoB (−15%; P<.009), whereas fenofibrate induced increases in plasma HDL-C (19%; P=.0001) and apoA-I (12%; P=.003) concentrations. Fenofibrate therapy for 8 weeks induced no significant changes in lipoprotein(a) levels in this study.

    Effects of Fenofibrate on Plasma Lipoprotein Distribution and Composition

    Plasma lipoprotein particles were separated by density gradient ultracentrifugation to yield multiple LDL and HDL subspecies. The correspondence of density subfractions to lipoprotein subspecies was assessed on the basis of chemical, physical, and immunological analyses.32 The distribution of lipoprotein mass among density gradient subfractions and the relative proportions of LDL and HDL subspecies from the plasma of CHL patients before and after fenofibrate treatment are presented in Table 2. After fenofibrate treatment the mean plasma VLDL and IDL (d=1.018 to 1.019 g/mL) mass concentrations decreased by 37% (P<.005) and 55% (P<.05), respectively, and total LDL mass concentration was reduced by 8% (P=.0276). Before treatment, CHL patients displayed a net asymmetry in the LDL profile in which the denser LDL subfractions, LDL-4 (d=1.039 to 1.050 g/mL) and LDL-5 (d=1.050 to 1.063 g/mL), predominated. After fenofibrate treatment the LDL profile was normalized, and the LDL density peak shifted toward the LDL-3 subfraction (d=1.029 to 1.039 g/mL). Indeed, we observed significant increases in both the mean plasma concentration (26%; P<.05) and the relative proportion (37%; P<.007) of LDL-3 concomitant with a decrease in plasma levels of both LDL-4 (−17%) and LDL-5 (−30%), although these latter changes were not significant (P=.068). In addition, it is noteworthy that absolute concentrations of the light, large LDL subspecies (LDL-1 and LDL-2) were decreased by 16.8% after drug treatment (NS). In contrast, such light subspecies accounted for similar proportions of total LDL before and after fenofibrate (18.8% and 17.0%, respectively). Within the hydrated density range of HDL (d=1.063 to 1.179 g/mL), CHL patients treated by fenofibrate displayed increases in the mass HDL2a (d=1.091 to 1.110 g/mL; 4%), HDL3a (d=1.110 to 1.133 g/mL; 19%), and HDL3b (d=1.133 to 1.156 g/mL; 13%). Fenofibrate therapy also induced significant reductions in plasma HDL2b (d=1.063 to 1.110 g/mL) and HDL3c (d=1.156 to 1.179 g/mL) levels (−22%; P=.0004 and −25%; P<.05, respectively).

    The mean weight chemical compositions of native lipoprotein subspecies (expressed as percent of their free cholesterol, CE, TG, phospholipid, and protein contents) are shown in Table 3. Analysis of the composition of VLDL, IDL, and each LDL subfraction revealed no significant effect of fenofibrate therapy. However, we observed a significant reduction in the relative proportion of TGs in HDL2a (P=.006) and HDL3a (P=.0048) subspecies after fenofibrate treatment, which was equally reflected in an increase in the CE/TG ratio in these subspecies from approximately 3 to 4.5 (Table 3).

    Hydrated particle size was evaluated for each LDL subfraction by electrophoresis in 2% to 16% polyacrylamide gradient gels under nondenaturing conditions. Particle size showed a marked and progressive diminution as density increased (Table 4). Analysis of particle diameters in each LDL subfraction after fenofibrate therapy showed a tendency toward nonsignificant increases in mean particle size in LDL-4 and LDL-5. It is of note that the LDL-1 (d=1.019 to 1.023 g/mL) and LDL-2 (d=1.023 to 1.029 g/mL) subfractions from CHL patients displayed a superior hydrated particle size compared with those of the corresponding subfractions from normolipidemic subjects.6

    Effect of Fenofibrate on Maximal Activity of Plasma CETP

    To determine whether fenofibrate exerted any effect on the maximal plasma activity of CETP, we employed an exogenous assay of CETP activity that accurately reflects plasma CETP mass.736 CETP activity in the plasma of seven CHL patients was assessed before and after fenofibrate treatment with an exogenous system containing donor HDL and VLDL/LDL acceptor particles isolated from normolipidemic plasma. After 3 hours of incubation, the mean transfer of CE radioactivity was significantly reduced in plasma from fenofibrate-treated patients by 26% (26.0±2.5% and 19.2±2.6% before and after treatment, respectively; P=.0014) (Fig 1A). Thus, fenofibrate induced a reduction of CETP-dependent CET activity in CHL patients. Furthermore, fenofibrate therapy resulted in a decrease in the amount of radioactive CE transferred (range, 12% to 43%) in the seven subjects tested (Fig 1B).

    CET and Exchange Between HDL and ApoB-Containing Lipoproteins

    The transfer of radioactive CEs between lipoprotein species is indicative of exchange and/or mass transfer reactions, whereas variations in net mass of CE specifically reflect CET reactions (see “Methods”). Thus, the difference between these two determinations (radioactivity or mass) represents CE exchange reactions. To evaluate the effect of fenofibrate on the reactions of CET and/or the exchange mediated by CETP that occurs between lipoprotein particles, plasma from each CHL patient before and after fenofibrate treatment was incubated with radiolabeled HDL.

    First, the rates of transfer and exchange of CE between HDL and the apoB-containing lipoproteins were determined in plasma from CHL patients before and after fenofibrate treatment by measuring the increase or decrease in both radioactive CE and the CE mass content of the VLDL+IDL, LDL, and HDL fractions isolated by sequential ultracentrifugation at each time point of incubation. The effects of fenofibrate treatment on the rates of transfer and exchange of CEs between HDL and apoB-containing lipoproteins are shown in Table 5. After fenofibrate administration, the rate of transfer of HDL CE to VLDL was significantly reduced (−38%; P=.0001). This finding is consistent with that of Mann et al37 in hypertriglyceridemic patients treated with bezafibrate. However, the rate of CE exchange between HDL and LDL remained unchanged after fenofibrate treatment. It is also important to note that the rate of loss of CE from HDL closely approximates the corresponding rate of gain of CE by the apoB-containing lipoproteins.

    Second, the total transfer of radioactive CE and CE mass transferred to the apoB-containing lipoproteins was subsequently quantified after 4 hours of incubation at 37°C after density gradient ultracentrifugation. Fig 2 represents the distribution of radiolabeled CE, expressed in micrograms of labeled CE transferred per milliliter of plasma (calculated from the known specific radioactivity of HDL CE) and CE mass, expressed in micrograms of CE transferred per milliliter of plasma (determined enzymatically) in CHL patients before and after fenofibrate treatment. Evaluation of CET from HDL to VLDL on the basis of radioactivity or by enzymatic determinations gave similar results, indicating that no exchange of CE occurred between HDL and VLDL during the first 4 hours of incubation at 37°C. In contrast, the transfer of radioactive CET from HDL to LDL occurred to a significant degree, during which the LDL fraction received 20% of total labeled CE transferred from HDL to the apoB-containing lipoproteins. However, no net mass transfer of CE from HDL to LDL was detected. Thus, the acquisition of radioactive CEs by the LDL fraction of CHL patients represents only CE exchange between HDL and LDL in the absence of net mass transfer. Fenofibrate treatment induced a significant reduction (−34%) of CET from HDL to VLDL (93.6±16.7 versus 61.9±11.6 μg CE transferred/mL plasma before and after treatment, respectively; P=.0023). However, drug therapy induced no effect on CE exchange between HDL and LDL.


    The present study has demonstrated for the first time that a second-generation fibrate, fenofibrate (in micronized form), can induce significant reductions in both the elevated total plasma CETP-dependent CET activity and in CET from HDL to VLDL in CHL; such total transfer activity is strongly correlated to plasma CETP mass.36 Moreover, fenofibrate quantitatively normalized the atherogenic LDL subspecies profile characteristic of CHL by specifically reducing the plasma levels of both dense LDL subfractions (LDL-4 and LDL-5, −21%) and proportionately increasing those of the intermediate LDL subspecies (LDL-3, 26%) of high receptor-binding affinity.10 Such changes occurred concomitantly with significant reductions not only in plasma cholesterol (−16%), LDL-C (−14%), and apoB (−15%) but also in plasma TG (−44%) and VLDL-C (−52%) levels. Finally, despite marked increases in HDL-C and apoA-I levels (19% and 12%, respectively), total HDL mass was not modified by fenofibrate; this observation was, however, accounted for by subtle modifications in the HDL subspecies profile, in which the relative proportions of HDL2a, HDL3a, and HDL3b were each significantly increased at the expense of HDL2b and HDL3c.

    Our new methodological approach to the determination of physiological rates of CE exchange and transfer has allowed us to demonstrate that the rate of CE mass transfer from HDL to VLDL is accelerated twofold in CHL patients (24.1±3.9 μg CE transferred·h−1·mL−1 plasma) compared with normolipidemic subjects (12.3±2.8 μg CE transferred·h−1·mL −1 plasma35 ). In contrast, no CE mass transfer from HDL to LDL was detectable in the plasma from CHL patients; radiolabeled CET was observed, however, implying that only reactions of CE exchange occurred between HDL and LDL. The absence of CE mass transfer and the occurrence of CE exchange alone between HDL and LDL thus constitute a characteristic of CHL.38 Indeed, in this respect, CHL patients differ markedly from normolipidemic subjects and from familial heterozygous hypercholesterolemic patients, in whom CE mass transfer from HDL to LDL predominates over that to VLDL.2835 Several mechanisms may underlie the deficiency of CE mass transfer between HDL and LDL in CHL patients. Thus, on the one hand, plasma VLDLs, which qualitatively represent the most active CE acceptors among the apoB-containing particles,35 are present at elevated concentrations (3- to 10-fold) and therefore are at the origin of a diminished CE flux to LDL.38 On the other hand, LDL particles from CHL patients may interact poorly with CETP as a result of their qualitative abnormalities,6 which involve both chemical composition and particle size.38 After fenofibrate therapy the overall CE mass transfer to apoB-containing lipoproteins was significantly reduced (−34%; P=.0023), whereas no effect on CE exchange between HDL and LDL was observed. Concomitant with the decrease in CETP activity, plasma VLDL concentrations were lowered by 37%. Thus, the occurrence of CE exchange between HDL and LDL was not directly related to the predominance of VLDL particles in the plasma of CHL patients. As fenofibrate therapy did not qualitatively modify LDL subspecies and thus did not modulate the potential affinity of LDL particles for CETP, we hypothesize that the abnormal chemical composition and particle structure of LDLs in CHL underlie their low affinity for CETP and that in the presence of elevated levels of VLDL acceptors, reactions of CE exchange occur exclusively between HDL and LDL.

    To determine whether the reduction of plasma CET after fenofibrate therapy might result at least in part from a reduction in plasma CETP mass, we employed an exogenous assay of maximal plasma CETP activity that accurately reflects plasma CETP mass.736 This indirect assay indicated that fenofibrate induced a significant reduction (−26%) in plasma CETP concentration. Information on the expression of the CETP gene gained from CETP-transgenic mice suggests that transcription of the CETP gene is induced by a high-cholesterol diet39 and that this response to cholesterol requires DNA sequences contained in the natural flanking regions of the human CETP gene. Furthermore, the proximal promoter of the gene mediates induction of CETP mRNA by dietary cholesterol, thereby indicating that this region contains a sterol upregulatory element distinct from the known sterol response element.39 In this context, it is relevant that evidence has been presented to indicate that fenofibrate inhibits endogenous cholesterol synthesis before mevalonate, indirectly causing a significant reduction of hydroxymethylglutaryl–coenzyme A reductase activity.40 This drug may also inhibit acyl coenzyme A–cholesterol acyltransferase activity, thereby reducing conversion of free cholesterol to its ester form.40 These observations lead us to suggest that the reduction of cholesterol synthesis by fenofibrate may induce downregulation of human CETP gene expression, resulting in the diminution of plasma CETP concentration.

    Analysis of the LDL subspecies isolated by density gradient ultracentrifugation not only revealed that dense, small particles (LDL-4, d=1.039 to 1.050 g/mL and LDL-5, d=1.050 to 1.063 g/mL) predominated in CHL subjects but also that their plasma levels were elevated up to threefold compared with those of normolipidemic subjects.69 This atherogenic, dense LDL profile (“phenotype B”41 ) was dramatically normalized by fenofibrate therapy as a result of an absolute increase (26%) in levels of intermediate LDL subspecies (LDL-3, d=1.030 to 1.039 g/mL), a modification that occurred simultaneously with diminution in the plasma levels of LDL-4 and LDL-5 (−17% and −30%, respectively). Such normalization was specifically related to the quantitative modifications of LDL subspecies that apparently occurred in the absence of alterations in the composition or size of LDL particles. Similar findings are reported by Caslake et al,12 who failed to observe an effect of fenofibrate on the composition of LDL particles in patients with hypercholesterolemia. It is of interest to note that another fibric acid derivative, ciprofibrate, also normalizes the atherogenic dense LDL profile in CHL patients.9 This effect involves the promotion of LDL removal via the receptor pathway.22 However, in contrast to fenofibrate, ciprofibrate induced normalization of both the composition and size of LDL subspecies.9

    Investigation of neutral lipid transfer in the present studies has afforded new insight into the mechanisms underlying the fenofibrate-induced normalization of the LDL subspecies profile. Thus, transfer rates of CE from HDL to VLDL in CHL patients were elevated twofold over normal; in addition, VLDLs were the major acceptors of CE. In contrast, no net mass transfer of CE to LDL was detected before or after fenofibrate therapy. After treatment, the rate of CET from HDL to VLDL was significantly reduced as a result of both reductions in plasma CETP mass and VLDL concentrations. These data are consistent with the hypothesis that the abnormal hepatic formation and intravascular metabolism of TG-rich VLDLs in CHL are major determinants in the formation of dense LDL subspecies in this atherogenic dyslipidemia. We suggest that several mechanisms underlie the shift in LDL subspecies profile induced by fenofibrate: the normalization of the hepatic formation and secretion of VLDL, of the hepatic catabolism of VLDL remnants, of the intravascular remodeling of VLDL as a result of stimulation of lipoprotein lipase and reduction in CETP mass and CET activity and the reduced production of LDL.12192040 Acting in concert, these complex metabolic actions result in the preferential intravascular formation of intermediate-density LDL of high receptor-binding affinity from hepatic VLDL precursors. Overall, our above postulates are entirely consistent with the significant inverse relationship typically observed between TG levels and mean plasma LDL particle size4243 as well as with studies of the Framingham cohort that show that mean plasma LDL particle size can be modulated by a change in plasma TG concentration.43 Finally, our findings allow us to exclude one potential mechanism whereby dense LDL subspecies might be transformed into intermediate LDL particles in CHL plasma and that would involve progressive CETP-facilitated transfer of CE molecules to the hydrophobic core of dense LDL; such a mechanism as this has been proposed to account for the transformation of dense LDL into lighter subspecies in the guinea pig.44

    Plasma HDL2 levels are some 20% lower in CHL patients than normolipidemic subjects (146.8±21.0 versus 182.6±39.8 mg/dL, respectively [n=9]; P<.0338 ). All nine CHL subjects in the present study displayed significant increases in plasma HDL-C level after fenofibrate treatment. Earlier studies have typically focused on the effects of fenofibrate on plasma HDL-C without regard to total plasma HDL mass. When total plasma HDL concentrations are considered, however, we observed that no significant modification occurred with drug therapy. The effects of fenofibrate on plasma HDL concentration were, however, heterogeneous among the nine patients studied. Thus, three patients displayed a 5% to 34% increase in total HDL concentration after treatment, three showed a reduction of 7% to 18%, and the last three presented similar HDL levels before and after fenofibrate therapy. Fenofibrate induced an increase in HDL2a (4%; d=1.091 to 1.110 g/mL), HDL3a (18%; d=1.110 to 1.133 g/mL), and HDL3b (13%; d=1.133 to 1.156 g/mL) levels concomitant with reductions in HDL2b (−22%; d=1.063 to 1.091 g/mL) and HDL3c (−25%; d=1.156 to 1.176 g/mL). The latter two HDL subfractions quantitatively represented the minor HDL subfractions in CHL patients compared with the HDL2a, HDL3a, and HDL3b subspecies. Thus, HDL2b and HDL3c accounted for 33% and 25% of total plasma HDL mass before and after fenofibrate treatment, respectively. Consequently, the net effect of drug treatment was an increase in plasma HDL-C concentrations in the absence of any modifications in total HDL mass. Similar conclusions were reached by Eisenberg et al45 in hypertriglyceridemic patients, in whom a reduction in plasma HDL2a and an increase in HDL2b levels were found; in addition, treatment of hypertriglyceridemic patients with bezafibrate effected a shift in this HDL pattern as a result of an increase in HDL2a concentration at the expense of HDL2b.45 Thus, the net effect was an elevation in HDL-C.

    During circulation of HDL3 in plasma, CE accumulates in the hydrophobic core of these particles through the action of lecithin:cholesterol acyltransferase; in consequence, HDL3 are transformed into the larger lipoprotein subspecies HDL2a.46 CETP is proposed to mediate the heteroexchange of TGs and CE primarily between HDL2a and the apoB-containing lipoproteins.46 Such neutral lipid exchange results in the formation of HDL2b, which subsequently becomes TG-enriched and deficient in CE. HDL2b is then transformed back to HDL3 by hydrolysis of TGs and phospholipids by the action of hepatic lipase (see Fig 3 in Reference 4646 ). After fenofibrate therapy, the relative proportions of HDL2a, HDL3a, and HDL3b in plasma increased by 20%, 40%, and 29%, respectively, whereas those of HDL2b and HDL3c decreased (−10% and −15%, respectively). In addition, we observed a significant reduction in the relative proportion of TGs in HDL2a and HDL3a. These quantitative and qualitative modifications of HDL subspecies are consistent with the drug-induced reduction of CETP activity observed in CHL plasma and do not appear to reflect an increase in hepatic lipase activity, which is essentially unaffected by fenofibrate.40

    It is of substantial interest to consider the present findings in light of the recent advances in our knowledge of the molecular mechanisms that underlie the lipid-lowering action of the fibrates and of fenofibrate in particular. The fibrates activate a group of transcription factors belonging to the superfamily of nuclear hormone receptors, the so-called peroxisome proliferator–activated receptors.4748 Upon activation, these receptors bind to the response elements of target genes and thus regulate their expression. Several such genes have been identified, including those of apoC-III, apoA-I, apoA-II, apoA-IV, acyl coenzyme A oxidase, and possibly that of lipoprotein lipase.49505152 The transcriptional downregulation of apoC-III and the upregulation of lipoprotein lipase by fenofibrate505152 are of major relevance to our present observations, in which circulating levels of VLDLs and their remnants were markedly reduced in CHL patients. Such actions enhance both the intravascular lipolysis of TG-rich lipoproteins as well as their tissue catabolism via apoE-mediated binding to specific cellular receptors.49 These insights thus shed new light on one of the key mechanisms implicated in the fenofibrate-induced normalization of the atherogenic dense LDL profile in CHL, ie, the correction of the abnormal hepatic and intravascular metabolism of TG-rich VLDL particles and their remnants.

    Selected Abbreviations and Acronyms

    CE=cholesteryl ester
    CET=cholesteryl ester transfer
    CETP=cholesteryl ester transfer protein
    CHL=combined hyperlipidemia
    HDL-C=HDL cholesterol
    LDL-C=LDL cholesterol
    PBS=phosphate-buffered saline
    VLDL-C=VLDL cholesterol

          Figure 1.

    Figure 1. Line graphs show the indirect evaluation of plasma CETP mass in CHL patients (n=7) before (•) and after (○) 8-week treatment with fenofibrate 200 mg/d. The exogenous assay used HDL donor and VLDL/LDL acceptor particles isolated from control plasmas with aliquots of the CHL patient’s plasma serving as the source of CETP (see “Methods”). A, Fenofibrate therapy induced a significant reduction in the mean resulting CET, thereby indicating a reduction of plasma CETP mass concentration after treatment. B, All seven subjects’ plasma showed a reduction of radioactive CET after 3 hours of incubation. Values are mean±SD. *P=.0014, **P=.0001 vs CHL patients before treatment.

          Figure 2.

    Figure 2. Bar graph shows the distribution of radiolabeled CE transferred (solid bars) calculated from the known specific radioactivity of HDL CE and the CE mass transferred (dotted bars) as determined by enzymatic methods after 4 hours of incubation at 37°C in CHL plasmas before and after 8 weeks of fenofibrate (200 mg/d). Labeled CET from HDL to LDL in the absence of CE mass transfer indicates that only reactions of CE exchange occurred between HDL and LDL in CHL patients. Fenofibrate treatment induced a significant reduction of net CE mass transfer from HDL to VLDL. However, drug therapy had no effect on CE exchange between HDL and LDL. Values are mean±SD. *P<.02, **P<.008 vs CHL patients before treatment.

    Table 1. Clinical Characteristics and Plasma Lipid Parameters in CHL Patients: Effect of Fenofibrate Treatment

    PatientAge, ySexBMIApoE IsoformTC, g/LTG, g/LVLDL-C, g/LLDL-C, g/LHDL-C, g/LApoB, g/LApoA-I, g/LLp(a), g/L
    Change, %−16.6%−44.3%−52.1%−14.0%+18.9%−15.3%+12.1%0.0%

    BMI indicates body mass index; TC, total cholesterol; Lp(a), lipoprotein(a); and ND, not determined. Plasma lipid parameters were determined before and after 8 weeks of fenofibrate 200 mg/d in nine CHL patients. VLDL-C, LDL-C, and HDL-C were determined by chemical analysis of VLDL (d<1.017 g/mL), LDL (d=1.019-1.063 g/mL), and HDL (d=1.063-1.179 g/mL) after isolation by density gradient ultracentrifugation (see “Methods”). Probability values were determined by comparing corresponding values for each parameter before and after fenofibrate by Student’s paired t test.

    Table 2. Effect of Fenofibrate Treatment on the Concentrations and Relative Proportions of Plasma Lipoprotein Subspecies in CHL Patients

    Density, g/mLBefore, mg/dLBefore, %After, mg/dLAfter, %
    Total LDL1.019-1.063447.5±100.3410.9±82.61
    Total HDL1.063-1.179322.5±50.2321.9±45.7

    Lipoproteins were isolated by density gradient ultracentrifugation. Plasma lipoprotein mass was determined before and after 8 weeks of fenofibrate 200 mg/d in nine CHL patients. The mass of each chemical component in individual fractions was determined by chemical analysis (see “Methods”), and the total mass of each fraction was then calculated as the sum of the mass of each component. The relative proportion of LDL or HDL subspecies is expressed as a percent of total LDL or HDL mass. Values are mean±SD.


    2P<.009 vs CHL patients before treatment.

    Table 3. Effect of Fenofibrate Treatment on the Chemical Composition of Plasma Lipoprotein Subspecies in CHL Patients

    d, g/mL<1.0171.018-1.0191.019-1.0231.023-1.0291.029-1.0391.039-1.0501.050-1.0631.063-1.0911.091-1.1101.110-1.1331.133-1.1561.156-1.179
    Free cholesterol

    The chemical composition of each lipoprotein fraction was determined before and after 8 weeks of fenofibrate 200 mg/d in nine CHL patients. Analyses were performed as described in “Methods.” Values are mean±SD and are expressed as percent mean weight.

    1P<.006 versus CHL patients before treatment.

    Table 4. Evaluation of LDL Subfraction Particle Size in CHL Patients Before and After Fenofibrate Treatment

    LDL SubspeciesBefore, ÅAfter, ÅP
    LDL-1296±13293±10 (273).5676
    LDL-2282±12281±11 (268).8324
    LDL-3267±11267±9 (265).9483
    LDL-4260±9264±8 (261).2552
    LDL-5260±9264±8 (258).2552

    Particle size was estimated by electrophoresis in nondenaturing polyacrylamide gradient gels (2%-16%; see “Methods”). Particle size was determined before and after an 8-week period of fenofibrate 200 mg/d in nine CHL patients. For comparative purposes, parenthetical values represent particle diameters in a corresponding series of LDL subspecies from normolipidemic men (n=5).6

    Table 5. Effect of Fenofibrate Treatment on the Rates of CE Exchange and Transfer Between HDL and the ApoB-Containing Lipoproteins

    Net mass transfer, μg CE transferred · h−1 · mL−1 plasma+24.1±3.9+14.8±2.8300−21.0±2.1−16.7±3.41
    Transfer+exchange, μg [3H]CE transferred · h−1 · mL−1 plasma24.5±6.516.2±4.327.2±2.28.0±2.730.1±5.120.9±3.22

    Rates of CE exchange and transfer were determined before and after 8 weeks of fenofibrate 200 mg/d to nine CHL patients. Rates of CE exchange and transfer were determined as described in “Methods.” Values are mean±SD. − and + indicate a net loss or net gain, respectively, of CE.



    3P<.0001 vs CHL patients before treatment.

    We are indebted to INSERM, to Laboratoires Fournier, and to the Medical Research Council of Canada (MT-5999 to Dr P.J. Dolphin) for financial support of these studies. Dr M. Guérin was supported by a research fellowship from the Association Claude Bernard.


    Correspondence to Maryse Guérin, Institut National de la Santé et de la Recherche Médicale (INSERM), Unité 321, Pavillon Benjamin Delessert, Hôpital de la Pitié, 83 Boulevard de l’ Hôpital, 75651 Paris cedex 13, France.


    • 1 Arad Y, Ramakrishnan R, Ginsberg HN. Lovastatin therapy reduces low density lipoprotein apo B level in subjects with combined hyperlipidemia by reducing the production of apoB containing lipoproteins: implications for the pathophysiology of apo B production. J Lipid Res.1990; 31:567-582. CrossrefMedlineGoogle Scholar
    • 2 Goldstein JL, Schrott GG, Hazzard WR, Bierman EL, Motulsky AG. Hyperlipidemia in coronary heart disease, II: genetic analysis of lipid levels in 176 families and delineation of a new inherited disorder, combined hyperlipidemia. J Clin Invest.1973; 52:1544-1568. CrossrefMedlineGoogle Scholar
    • 3 Janus ED, Nicoll AM, Turner PR, Magill P, Lewis B. Kinetic bases of the primary hyperlipidaemias: studies of apolipoprotein B turnover in genetically defined subjects. Eur J Clin Invest.1980; 10:161-172. CrossrefMedlineGoogle Scholar
    • 4 Kissebah AH, Alfarsi S, Adamo PW. Integrated regulation of very low density lipoprotein triglyceride and apolipoprotein B kinetics in man: normolipidemic subjects, familial hypertriglyceridemia and familial combined hyperlipidemia. Metabolism.1981; 30:856-868. CrossrefMedlineGoogle Scholar
    • 5 Teng B, Sniderman AD, Soutar AK, Thompson GR. Metabolic basis of hyperapobetalipoproteinemia: turnover of apolipoprotein B in low density lipoprotein and its precursors and subfractions compared with normal and familial hypercholesterolemia. J Clin Invest.1986; 77:663-672. CrossrefMedlineGoogle Scholar
    • 6 Dejager S, Bruckert E, Chapman MJ. Dense low density lipoprotein subspecies with diminished oxidative resistance predominate in combined hyperlipidemia. J Lipid Res.1993; 34:295-308. CrossrefMedlineGoogle Scholar
    • 7 McPherson R, Mann CJ, Tall AR, Hogue M, Martin L, Milne RW, Marcel YL. Plasma concentrations of cholesteryl ester transfer protein in hyperlipoproteinemia: relation to cholesteryl ester protein activity and other lipoprotein variables. Arterioscler Thromb.1991; 11:797-804. CrossrefMedlineGoogle Scholar
    • 8 Tato F, Vega GL, Tall AR, Grundy SM. Relation between cholesteryl ester transfer protein activities and lipoprotein cholesterol in patients with hypercholesterolemia and combined hyperlipidemia. Arterioscler Thromb Vasc Biol.1995; 15:112-120. CrossrefMedlineGoogle Scholar
    • 9 Bruckert E, Dejager S, Chapman MJ. Ciprofibrate therapy normalises the atherogenic low-density lipoprotein subspecies profile in combined hyperlipidemia. Atherosclerosis.1993; 100:91-102. CrossrefMedlineGoogle Scholar
    • 10 Nigon F, Lesnik P, Rouis M, Chapman MJ. Discrete subspecies of human low-density lipoproteins are heterogeneous in their interaction with cellular LDL receptor. J Lipid Res.1991; 32:1741-1754. CrossrefMedlineGoogle Scholar
    • 11 Griffin BA, Freeman DJ, Tait GW, Thomson J, Caslake MJ, Packard CJ, Shepherd J. 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; 106:241-253. CrossrefMedlineGoogle Scholar
    • 12 Caslake MJ, Packard CJ, Gaw A, Murray E, Griffin BA, Vallance BD, Shepherd J. Fenofibrate and LDL metabolic heterogeneity in hypercholesterolemia. Arterioscler Thromb.1993; 13:702-711. CrossrefMedlineGoogle Scholar
    • 13 Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Modifications of low-density lipoprotein that increase its atherogenicity. N Engl J Med.1989; 320:915-924. CrossrefMedlineGoogle Scholar
    • 14 De Graaf J, Hak-Lemmers HLM, Hectors MPC, Demacker PNM, Hendricks JCM, Stalenhoef AFH. Enhanced susceptibility to in vitro oxidation of dense low-density lipoprotein subfraction in healthy subjects. Arterioscler Thromb.1991; 11:298-306. CrossrefMedlineGoogle Scholar
    • 15 O’Connor P, Feely J, Shepherd J. Lipid lowering drugs. BMJ.1990; 300:667-672. CrossrefMedlineGoogle Scholar
    • 16 Eisenberg S. Role of fibric acids in the management of hyperlipidemia. Curr Opin Lipidol.1990; 1:34-38. CrossrefGoogle Scholar
    • 17 Kissebah AH, Adams PW, Harrigan P, Wynn V. The mechanism of action of clofibrate and tetranicotinoylfructose (bradilan) on the kinetics of plasma free fatty acid and triglyceride transport in type IV and type V hypertriglyceridaemia. Eur J Clin Invest.1974; 4:163-174. CrossrefMedlineGoogle Scholar
    • 18 Wolfe BM, Kane JP, Havel RJ, Brewster HP. Mechanism of the hypolipidemic effect of clofibrate on postabsorptive man. J Clin Invest.1973; 52:2146-2159. CrossrefMedlineGoogle Scholar
    • 19 Heller F, Harvengt C. Effects of clofibrate, bezafibrate, fenofibrate and probucol on plasma lipolytic enzymes in normolipaemic subjects. Eur J Clin Pharmacol.1983; 25:57-63. CrossrefMedlineGoogle Scholar
    • 20 Shepherd J, Caslake MJ, Lorimer AR, Vallance BD, Packard CJ. Fenofibrate reduces low-density lipoprotein catabolism in hypertriglyceridemic subjects. Arteriosclerosis.1985; 5:162-168. LinkGoogle Scholar
    • 21 Kleinman Y, Oschry Y, Eisenberg S. Abnormal regulation of LDL receptor activity and abnormal cellular metabolism of hypertriglyceridaemic low density lipoprotein: normalization with bezafibrate therapy. Eur J Clin Invest.1987; 17:538-543. CrossrefMedlineGoogle Scholar
    • 22 Gaw A, Packard CJ, Caslake MJ, Griffin BA, Lindsay GM, Thomson J, Vallance BD, Wosornu D, Shepherd J. Effects of ciprofibrate on LDL metabolism in man. Atherosclerosis.1994; 108:137-148. CrossrefMedlineGoogle Scholar
    • 23 Tikkanen MJ. Fibric acid derivatives. Curr Opin Lipidol.1992; 3:29-33. CrossrefGoogle Scholar
    • 24 Blane GF, Bogaievsky Y, Bonnefous F. Fenofibrate: influence on circulating lipids and side-effects in medium and long-term clinical use. In: Fears R, ed. Pharmacological Control of Hyperlipidaemia. Barcelona, Spain: JR Prous Science Publishers; 1986:187-216. Google Scholar
    • 25 Knopp RH, Brown WV, Dujovne CA, Farquhar JW, Feldman EB, Goldberg AC, Grundy SM, Lasser NL, Mellies MJ, Palmer RH, Samuel P, Schonfeld G, Superko HR. Effects of fenofibrate on plasma lipoproteins in hypercholesterolemia and combined hyperlipidemia. Am J Med. 1987;83(suppl 5B):50-59. Google Scholar
    • 26 Farnier M, Bonnefous F, Debbas N, Irvine A. Comparative efficacy and safety of micronised fenofibrate and simvastatin in patients with primary type IIa or IIb hyperlipidemia. Arch Intern Med.1994; 154:441-449. CrossrefMedlineGoogle Scholar
    • 27 Kornitzer M, Dramaix M, Vandenbroek MD, Everaert L, Gerlinger C. Efficacy and tolerance of 200 mg micronised fenofibrate administered over a 6-month period in hyperlipidaemic patients: an open Belgian multicenter study. Atherosclerosis. 1994;110(suppl):S49-S54. Google Scholar
    • 28 Guérin M, Dolphin PJ, Chapman MJ. Preferential cholesteryl ester acceptors among the LDL subspecies of subjects with familial hypercholesterolemia. Arterioscler Thromb.1994; 14:679-685. CrossrefMedlineGoogle Scholar
    • 29 Guérin M, Dolphin PJ, Talussot C, Gardette J, Berthezene F, Chapman MJ. Pravastatin modulates cholesteryl ester transfer from HDL to apoB-containing lipoproteins and lipoprotein subspecies profile in familial hypercholesterolemia. Arterioscler Thromb Vasc Biol.1995; 15:1359-1368. CrossrefMedlineGoogle Scholar
    • 30 Lagrost L, Gandjini H, Athias A, Guyard-Dangremont V, Lallemant C, Gambert P. Influence of plasma cholesteryl ester transfer activity on the LDL and HDL distribution profiles in normolipidemic subjects. Arterioscler Thromb.1993; 13:815-825. CrossrefMedlineGoogle Scholar
    • 31 Chapman MJ, Goldstein S, Lagrange D, Laplaud PM. A density gradient ultracentrifugation procedure for the isolation of the major lipoprotein classes from human plasma. J Lipid Res.1981; 22:339-358. CrossrefMedlineGoogle Scholar
    • 32 Chapman MJ, Laplaud PM, Luc G, Forgez P, Bruckert E, Goulinet S, Lagrange D. Further resolution of the low density lipoprotein spectrum in normal human plasma: physicochemical characteristics of discrete subspecies separated by density gradient ultracentrifugation. J Lipid Res.1988; 29:442-458. CrossrefMedlineGoogle Scholar
    • 33 Nichols AV, Krauss RM, Musliner TA. Nondenaturing polyacrylamide gradient gel electrophoresis. Methods Enzymol.1986; 128:417-431. CrossrefMedlineGoogle Scholar
    • 34 Anderson DW, Nichols AV, Forte TM, Lindgren FT. Particle distribution of human serum high density lipoproteins. Biochim Biophys Acta.1977; 493:55-68. CrossrefMedlineGoogle Scholar
    • 35 Guérin M, Dolphin PJ, Chapman MJ. A new in vitro method for the simultaneous evaluation of cholesteryl ester exchange and mass transfer between HDL and apo B–containing lipoprotein subspecies: identification of preferential cholesteryl ester acceptors in human plasma. Arterioscler Thromb.1994; 14:199-206. CrossrefMedlineGoogle Scholar
    • 36 Ahnadi CE, Berthezene F, Ponsin G. Simvastatin-induced decrease in the transfer of cholesterol esters from high density lipoproteins to very low and low density lipoproteins in normolipidemic subjects. Atherosclerosis.1993; 99:219-228. CrossrefMedlineGoogle Scholar
    • 37 Mann CJ, Yen FT, Grant AM, Bihain BE. Mechanism of plasma cholesteryl ester transfer in hypertriglyceridemia. J Clin Invest.1991; 88:2059-2066. CrossrefMedlineGoogle Scholar
    • 38 Guérin M, Bruckert E, Dolphin PJ, Chapman MJ. Absence of CETP-mediated cholesteryl ester mass transfer from HDL to LDL particles is a major feature of combined hyperlipidemia. Eur J Clin Invest. In press. Google Scholar
    • 39 Jiang XC, Agellon LB, Walsh A, Breslow JL, Tall A. Dietary cholesterol increases transcription of the human cholesteryl ester transfer protein gene in transgenic mice. J Clin Invest.1992; 90:1290-1295. CrossrefMedlineGoogle Scholar
    • 40 Kloer HU. Structure and biochemical effects of fenofibrate. Am J Med. 1987;83(suppl 5B):3-8. Google Scholar
    • 41 Krauss RM. Heterogeneity of plasma low-density lipoproteins and atherosclerosis risk. Curr Opin Lipidol.1994; 5:339-349.CrossrefMedlineGoogle Scholar
    • 42 Campos H, Blijlevens E, McNamara JR, Ordovas JM, Posner BM, Wilson PWF, Castelli WP, Schaefer EJ. LDL particle size distribution: results from the Framingham offspring study. Arterioscler Thromb.1992; 12:1410-1419. CrossrefMedlineGoogle Scholar
    • 43 McNamara JR, Jenner JL, Li Z, Wilson PWF, Schaefer EJ. Change in LDL particle size is associated with change in plasma triglyceride concentration. Arterioscler Thromb.1992; 12:1284-1290. CrossrefMedlineGoogle Scholar
    • 44 Luc G, Chapman MJ. Guinea pig low density lipoproteins: structural and metabolic heterogeneity. J Lipid Res.1988; 29:1251-1263. CrossrefMedlineGoogle Scholar
    • 45 Eisenberg S, Gavish D, Oschry Y, Fairaw M, Deckelbaum RG. Abnormalities in very low, low, and high density lipoproteins in hypertriglyceridemia: reversal toward normal with bezafibrate treatment. J Clin Invest.1984; 74:470-482. CrossrefMedlineGoogle Scholar
    • 46 Grundy SM, Vega GL. Fibric acids: effects on lipids and lipoprotein metabolism. Am J Med. 1987;83(suppl 5B):9-20. Google Scholar
    • 47 Gottlicher M, Widmark E, Li Q, Gustafsson JA. Fatty acids activate chimera of the clofibric acid-activated receptor and glucocorticoid receptor. Proc Natl Acad Sci U S A.1992; 89:4653-4657. CrossrefMedlineGoogle Scholar
    • 48 Keller H, Dreyer C, Medin J, Mahfoudi A, Ozato K, Wahli W. Fatty acids and retinoids control lipid metabolism through activation of peroxisome proliferator-activated receptor retinoid×receptor heterodimers. Proc Natl Acad Sci U S A.1993; 90:2160-2164. CrossrefMedlineGoogle Scholar
    • 49 Staels B, Vu-Dac N, Kosykh VA, Saladin R, Fruchart JC, Dallongeville J, Auwerx J. Fibrates downregulate apolipoprotein C-III expression independent of induction of peroxisomal acyl coenzyme A oxidase: a potential mechanism for the hypolipemic action of fibrates. J Clin Invest.1995; 95:705-712. CrossrefMedlineGoogle Scholar
    • 50 Berthou L, Saladin R, Yaqoob P, Branellec D, Calder P, Fruchart JC, Denèfle P, Auwerx J, Staels B. Regulation of rat liver apolipoprotein A-I, apolipoprotein A-II and acyl-coenzyme A oxidase gene expression by fibrates and dietary fatty acids. Eur J Biochem.1995; 232:179-187. CrossrefMedlineGoogle Scholar
    • 51 Staels B, Auwerx J. Perturbation of developmental gene expression in rat liver by fibric acid derivatives: lipoprotein lipase and a-fetoprotein as models. Development..1992; 115:1035-1043. CrossrefMedlineGoogle Scholar
    • 52 Schoonjans K, Staels B, Deed S, Auwerx J. Fibrates and fatty acids induce lipoprotein lipase gene expression via peroxisome proliferator–activated receptor. Circulation. 1995;92(suppl I):I-495. Abstract.Google Scholar