Regulation of Human ApoA-I by Gemfibrozil and Fenofibrate Through Selective Peroxisome Proliferator-Activated Receptor Modulation
http://www.100md.com
动脉硬化血栓血管生物学 2005年第3期
From UR.545INSERM, Département d’Athérosclérose (H.D., P.P., I.P.T., F.P., G.L., J.-C.F., C.F., B.S.), Institut Pasteur Lille and Faculté de Pharmacie, Université de Lille2, France; UR.459INSERM, Biologie Moléculaire des Récepteurs Nucléaires (B.L., P.L.), Faculté de Médecine Lille, France; the Department of Veterinary Science (J.M.P.), Center for Molecular Toxicology, Pennsylvania State University, Hershey, Penn; the Laboratory of Metabolism (F.J.G.), National Cancer Institute, National Institute of Health, Bethesda, Md; GENFIT.SA (R.G., S.H.), Loos, France; and Cardiac Catherization Laboratory & Interventional Cardiology (V.D.), University Health Network, Toronto, Canada.
Correspondence to Hélène Duez or Prof Bart Staels, UR545INSERM, Département d’Athérosclérose, Institut Pasteur Lille, 1 rue Calmette, 59019 Lille, France. E-mail helene.duez@pasteur-lille.fr or bart.staels@pasteur-lille.fr
Abstract
Objective— The objective of this trial was to study the effects of fenofibrate (FF) and gemfibrozil (GF), the most commonly used fibrates, on high-density lipoprotein (HDL) and apolipoprotein (apo) A-I.
Methods and Results— In a head-to-head double-blind clinical trial, both FF and GF decreased triglycerides and increased HDL cholesterol levels to a similar extent, whereas plasma apoA-I only increased after FF but not GF. Results in human (h) apoA-Itransgenic (hA-ITg) peroxisome proliferator-activated receptor (PPAR) –/– mice demonstrated that PPAR mediates the effects of FF and GF on HDL in vivo. Although plasma and hepatic mRNA levels of hapoA-I increased more pronouncedly after FF than GF in hA-ITgPPAR+/+ mice, both fibrates induced acylCoAoxidase mRNA similarly. FF and GF transactivated PPAR with similar activity and affinity on a DR-1 PPAR response element, but maximal activation on the hapoA-I DR-2 PPAR response element was significantly lower for GF than for FF. Moreover, GF induced recruitment of the coactivator DRIP205 on the DR-2 site less efficiently than did FF.
Conclusion— Both GF and FF exert their effects on HDL through PPAR. Whereas FF behaves as a full agonist, GF appears to act as a partial agonist due to a differential recruitment of coactivators to the promoter. These observations provide an explanation for the differences in the activity of these fibrates on apoA-I.
In hyperlipidemic patients and human apolipoprotein A-Itransgenic (hA-ITg) mice, fenofibrate (FF) and gemfibrozil (GF) increase HDL cholesterol, whereas apoA-I levels only increased after FF. Because of differential coactivator recruitment to the promoter, FF and GF, respectively, behave as full and partial PPAR agonists, likely explaining the clinical differences in the activity of these fibrates on apoA-I.
Key Words: apolipoprotein A-I ? high-density lipoprotein ? fibrates ? peroxisome proliferator-activated receptor ? selective PPAR modulator
Introduction
Several clinical trials have indicated a benefit of fibrate treatment in cardiovascular risk reduction and improvement of lipid profiles.1 Angiography trials have shown that administration of fibrates (bezafibrate, gemfibrozil [GF], and fenofibrate [FF]) significantly decreases the progression of coronary narrowing.1 Furthermore, fibrate treatment decreased combined incidence of coronary heart disease events in primary and secondary prevention trials.1 These trials also demonstrated the effectiveness of fibrates in improving dyslipidemia by lowering elevated plasma triglycerides, improving the low-density lipoprotein subfraction distribution, and increasing high-density lipoprotein cholesterol (HDL-C). However, qualitative and quantitative differences appear to exist between fibrates. Qualitative differences in pharmacological modulation of HDL metabolism, and its major apolipoprotein (apo) A-I that confers protection against the development of atherosclerosis in animal models2 and humans,3 may be of clinical importance. In several clinical trials, FF increases plasma HDL-C and apoA-I concentrations.4–8 By contrast, although significantly increasing HDL-C levels, GF seems to have little or no effect on apoA-I levels.9–11 Thus, the 2 most used fibrates (FF and GF), although exerting similar beneficial actions on plasma lipids, may also exert distinct effects by yet unresolved mechanisms.
Even though fibrates are low-affinity ligands, their effects are thought to be mediated principally through peroxisome proliferator-activated receptor (PPAR) . Fibrate-activated PPAR regulates gene expression after heterodimerization with the retinoid X receptor (RXR), subsequent coactivator recruitment, and binding to PPAR response elements (PPREs) which are located in target gene promoters. The PPAR/RXR heterodimer interacts with coactivators that act as bridges with the basal transcription machinery and play crucial roles in the transmission of regulatory signals. Different ligands can induce different receptor conformations in a way that is unique to each ligand, leading to differential coactivator recruitment. This is the case for PPAR agonists. Compared with other glitazones, troglitazone behaves as a partial or full agonist for PPAR depending on the cellular environment and promoter context leading to PPAR selective modulation and differential downstream effects on gene expression.12 This selective PPAR modulator concept may explain differences in biological activity.
In this study the clinical effects of FF and GF were compared on lipid metabolism with a focus on their effects on HDL and apoA-I in a head-to-head trial. Moreover, the molecular mechanisms underlying observed differences were investigated.
Methods
Clinical Study
A multicenter, double-blind, randomized, 24-week study was performed comparing head-to-head the usual doses of micronized FF (200 mg per day) and GF (1200 mg per day). After an 8-week placebo run-in period, 234 patients (82 women, 152 men) aged 18 to 70 years (mean age 51.6 years) with combined hyperlipidemia (total cholesterol [TC]/HDL-C 6.0 mmol/L [males], 5.6 mmol/L [females]; triglycerides between 2.2 and 5.6 mmol/L) were randomized and treated with FF (n=116) or GF (n=118). Both drugs were equally well tolerated. For triglyceride, HDL-C, and apoA-I measurement, please see the online Methods, available at http://atvb.ahajournals.org.
Comparison of percent changes was done by 1-way analysis of variance. The study was approved by the independent ethics committee of each of the 10 participating Canadian centers.
Animals
Hemizygous hA-ITg mice on a PPAR–/–, +/–, and +/+ background were obtained by crossing homozygous human (h) apoA-Itransgenic (hA-ITg) mice13 with PPAR-null mice14,15 and subsequently treated with either FF or GF (0.2 and 0.5% w/w, respectively) for 17 days, a time at which a steady-state response is reached16 (see online Methods). All animal experiments were approved by the Institutional Animal Care and Use Committee.
Lipid, Apolipoprotein, and Lipoprotein Analyses
Plasma TC was determined as described.15 Plasma human (h) apoA-I and mouse apoA-II levels were measured by an immunonephelometric assay using species-specific antibodies.17 Lipoprotein cholesterol distribution in pooled plasma samples was assessed by fast protein liquid.18 hapoA-I was measured in each eluted fraction using the same immunonephelometric assay as above,17 and HDL fractions were defined as apoA-I containing fractions. HDL-C was determined from the cholesterol distribution profiles by calculating the areas under the curves of the apoA-I-containing fractions.
RNA Analysis
RNA was isolated and Northern blot analysis was performed as described19 using hapoA-I and rat acylCoAoxidase (ACO) probes and human ribosomal 28S cDNA as a control probe. hapoA-I, ACO, and mouse apoA-II levels were quantified by reverse transcription followed by real-time polymerase chain reaction, normalized to the 28S internal control, and expressed as fold induction over the untreated control group, set as 1 (see online Methods).
Statistical Analysis
ANOVA was used to compare differences between treatment groups. When overall significance was attained, Scheffe test was used to analyze for significant differences between the experimental groups.
Transient Transfection Assays
The pSG5-mPPAR plasmid was as described.20 The hapoA-II-DR-1 (J6-TK-Luc) and hapoA-I-DR-2 (A3-TK-pGL3)–containing reporter plasmids were obtained by inserting 6 or 3 copies of the hapoA-II PPRE (DR-1)–containing J or the hapoA-I PPRE (A site)21 sites in the pTK-pGL3 plasmid, respectively (see online Methods).
Cos-7 cells were transfected using polyethylenimine before the addition of increasing concentrations of the compound tested or vehicle for 24 hours in 0.2% FCS-DMEM (see online Methods). The pCMV-?-galactosidase expression plasmid was cotransfected as a control for transfection efficiency. Luciferase and ?-galactosidase assays were performed as described.22 EC50 values were estimated using Prism software(GraphPad).
Cofactor Recruitment Assays
The glutathione-S-transferase pull-down protocol was as described.23 DNA-dependent protein-protein interactions were tested as described previously.24 Briefly, hPPAR and 35S-labeled-hRXR were incubated with or without PPAR ligand before a double-stranded oligonucleotide corresponding to the ACO or hapoA-I PPRE and glutathione-S-transferase-DRIP205 were added (see online Methods). Complexes were resolved and quantified with a PhosphorImager (Molecular Dynamics).
Results
FF and GF Regulate ApoA-I Differently in Humans
To compare the effects of FF and GF on plasma lipids and apoA-I, 234 patients with combined hyperlipidemia were treated with FF (n=116) or GF (n=118) in a head-to-head, double-blind, randomized trial. Baseline characteristics of patients are given in Table I (available online at http://atvb.ahajournals.org).
Both fibrates reduced triglycerides (FF, –39±22% versus GF, –41±26%, P>0.5) and increased HDL-C (FF, +16±20% versus GF, +12±21%, P>0.5) to a similar extent (Figure 1). Interestingly, although having similar HDL-C increasing activities, only FF treatment increased apoA-I plasma levels, whereas GF treatment was without effect (FF, +8.6±14% versus GF, +1.5±12%, P<0.001; Figure 1).
Figure 1. Effects of FF or GF treatment on HDL-C, triglyceride, and apoA-I plasma levels. Hyperlipidemic patients were treated with either fenofibrate (FF; 200 mg/d, n=116, black bars) or gemfibrozil (GF; 1200 mg per day, n=118, gray bars) in a direct 24-week head-to-head comparison study. ApoA-I (FF, +8.6±14% vs GF, +1.5±12%), HDL cholesterol (HDL-C; FF, +16±20% vs GF, +12±21%), and triglyceride (TG; FF, –39±22% vs GF, –41±26%) plasma levels are reported as mean percent change from baseline. (1-way analysis of variance, ***P<0.001).
PPAR Mediates the Effects of FF on Plasma HDL-C and hapoA-I Plasma and Hepatic mRNA Levels In Vivo in Mice
To determine the mechanistic basis of this differential effect, we first aimed to establish whether PPAR is required for the in vivo regulation of HDL-C and hapoA-I levels by fibrates. hapoA-I-transgenic (hA-ITg) mice on a PPAR+/+, +/–, and –/– background were treated with FF for 17 days. In PPAR+/+hA-ITg mice, FF treatment resulted in significant 4- and 3.5-fold increases of plasma TC and hapoA-I levels (TC: control [CON], 159±9 versus FF, 618±31 mg/dL, P<0.001; hapoA-I: CON, 411±13 versus FF,1503±117 mg/dL, P<0.001). By contrast, no increase was observed in the PPAR–/–hA-ITg mice (TC: CON, 152±8 versus FF, 108±7 mg/dL, P>0.05; hapoA-I: CON, 315±19 versus FF, 268±8 mg/dL, P>0.05; Figure 2 A and 2B). Strikingly, treatment with FF increased TC (CON, 145±9 versus FF, 328±29 mg/dL, P<0.001) and hapoA-I (CON, 322±20 versus FF, 625±51 mg/dL, P<0.05) to intermediate levels in PPAR+/–hA-ITg mice compared with PPAR+/+ and PPAR–/– mice (Figure 2A and 2B). Consistently, changes in HDL-C distribution profiles in the PPAR+/–hA-ITg mice occurred in an intermediate fashion when compared with the 2 other PPAR genotypes (data not shown). In agreement with the above results, FF treatment induced a marked increase in hepatic hapoA-I mRNA levels in the PPAR+/+hA-ITg mice (CON, 100±15 versus FF, 226±10%, P<0.001), whereas no effect was observed in the PPAR–/–hA-ITg mice (CON, 100±20 versus FF, 121±8%, P>0.05; Figure I, available online at http://atvb.ahajournals.org), and an intermediate effect was observed in the PPAR+/–hA-ITg mice (CON, 100±7 versus FF, 173±25%, P<0.01). Altogether these data indicate that the increase of HDL-C and hapoA-I plasma and hepatic mRNA levels by FF in hA-ITg mice occurs in a PPAR gene dosage-dependent manner.
Figure 2. PPAR is required for the in vivo regulation of plasma cholesterol and hapoA-I by FF in a gene dosage-dependent manner. Plasma TC (A) and hapoA-I (B) levels in PPAR+/+hA-ITg (CON, n=10; FF, n=10), PPAR+/–hA-I Tg (CON, n=7; FF, n=9) and PPAR–/–hA-ITg (CON, n=13; FF, n=13) mice treated with control (CON; white bars) or fenofibrate (FF)-supplemented (black bars) chow. The results are expressed as percentage (mean±SEM) of untreated control mice. Statistically significant differences between groups are indicated by asterisks (Scheffe test, *P<0.05, ***P<0.001).
FF and GF Exert Distinct PPAR-Dependent Effects on HDL-C and hapoA-I Plasma Levels in Mice
The effect of FF and GF on the in vivo regulation of HDL-C and hapoA-I was next assessed in hA-ITgPPAR+/+ and PPAR–/–mice. FF and, albeit to a lower extent, GF significantly increased plasma TC in PPAR+/+hA-ITg mice (CON, 158±7 versus FF, 498±31 mg/dL, P<0.001; CON, 158±7 versus GF, 224±13 mg/dL, P<0.01; Figure 3A), whereas this effect was not observed in the PPAR–/– background (Figure II, available online at http://atvb.ahajournals.org). A large increase in HDL-C was observed in PPAR+/+hA-ITg mice treated with FF and GF, although the effect of GF was again less pronounced compared with FF (1.6-fold versus 3.5-fold, respectively; Figure 3B). No change was observed in the PPAR–/–hA-ITg mice (Figure II). Furthermore, FF and GF treatment led to the appearance of a larger peak overlapping the low-density lipoprotein and HDL size range in a PPAR-dependent manner.
Figure 3. FF but not GF induces plasma hapoA-I levels in a PPAR-dependent manner. A, Plasma TC and hapoA-I levels in PPAR+/+hA-ITg mice (CON, n=6; FF, n=6; GF, n=8) treated with control (CON; white bars), fenofibrate (FF; black bars) or gemfibrozil (GF; gray bars)-supplemented chow. The results are expressed as percentage (mean±SEM) of untreated control mice. Statistically significant differences between groups are indicated by asterisks (Scheffe test, **P<0.01, ***P<0.001). B, Representative gel filtration chromatography distribution profiles of cholesterol in plasma from PPAR+/+hA-ITg mice treated with control () or FF () or GF ()-supplemented chow. "HDL" fractions are defined as apoA-I-containing fractions (CON, 55 to 70; FF, 43 to 62; GF, 50 to 70).
FF increased plasma hapoA-I in PPAR+/+hA-ITg mice (CON, 357±12 versus FF, 886±181 mg/dL, P<0.001; Figure 3A), an effect that was not observed in the PPAR–/– background (CON, 304±20 versus FF, 253±8 mg/dL, P>0.05). In contrast to FF, plasma hapoA-I levels were only slightly modified by GF treatment in the PPAR+/+hA-ITg mice (CON, 357±12 versus GF, 394±12 mg/dL, P>0.05; Figure 3A). Altogether these data demonstrate that both GF and FF induce HDL-C levels in a PPAR-dependent manner. However, in contrast to FF, GF has only a marginal effect on plasma hapoA-I.
FF Induces hapoA-I to a Larger Extent Than GF, Whereas Both Fibrates Increase ACO mRNA Levels Similarly in Mice
In line with the plasma changes, FF significantly increased hapoA-I mRNA levels in PPAR+/+ mice (Figure 4) but was ineffective in the PPAR–/–hA-ITg mice. However, GF treatment was not efficient to induce hepatic hapoA-I mRNA levels in PPAR+/+hA-ITg mice (Figure 4). By contrast, mRNA levels of ACO, a DR-1–driven PPAR target gene,25 were increased by GF treatment to a similar extent as FF (4.7- and 5.3-fold increase, respectively) in PPAR+/+hA-ITg mice, whereas no change was observed with either treatment in the PPAR–/–hA-ITg mice (Figure 4). Hepatic mRNA levels of mouse apoA-II, another important protein constituent of HDL, were also measured. Neither FF nor GF significantly affected mouse (m) apoA-II mRNA levels in PPAR+/+hA-ITg mice (FF, 110±30%; GF, 121±28%; CON, 100±10%). However, both FF and GF treatment increased mapoA-II plasma levels in PPAR+/+hA-ITg mice (FF, 136±20%, P<0.01; GF, 200±65%, P<0.01; CON, 100±19%), whereas no change was observed with either FF or GF in the PPAR–/–hA-ITg mice (FF, 102±23%, P>0.05; GF, 128±18%, P>0.05; CON, 100±10%). Thus, GF appears to behave as a selective PPAR modulator in vivo.
Figure 4. In vivo differential effects of FF and GF on hepatic hapoA-I but not on ACO mRNA levels. Representative Northern blot showing liver hapoA-I and ACO mRNA levels in PPAR+/+hA-ITg and PPAR–/–hA-ITg mice treated with control (CON), or fenofibrate (FF), or gemfibrozil (GF)-supplemented chow. Values reported below the blots are the induction levels of hapoA-I and ACO mRNA in treated vs control PPAR+/+hA-ITg mice quantified by quantitative polymerase chain reaction.
GF and FF Distinctly Induce PPAR Activation and Coactivator Recruitment in a PPRE-Dependent Manner
The PPAR activation profile of GF and FF was determined in a transfection assay in which full-length PPAR was cotransfected with reporter constructs driven by either the hapoA-I DR-2 or a DR-1 PPRE (Figure 5A and 5B). GF and FF increased PPAR activity from the hapoA-I DR-2 site with similar EC50 values (5.8 and 7.8 μmol/L, respectively). However, a submaximal response was observed with GF compared with FF (Figure 5A). By contrast, on the DR-1 PPRE-driven vector, FF and GF activated PPAR with similar EC50 values (12 μmol/L and 11 μmol/L, respectively) and similar maximal response (Figure 5B). These data indicate that GF behaves as a partial agonist compared with FF depending on the type of PPRE. Next, the influence of FF and GF on coactivator recruitment to PPAR/RXR bound to either a DR-1 or the hapoA-I DR-2 site was examined. DRIP205 was chosen as a model system, but similar results were obtained with other coactivators, that is, p300, CBP (CREB binding protein), and glucocorticoid receptor interacting protein (data not shown). Compared with FF, GF led to at least as efficient coactivator recruitment to the DR-1 site but was comparatively less efficient on the hapoA-I DR-2 site (Figure 5C and 5D). Taken together, these findings suggest that GF selectively modulates PPAR transcriptional activity associated with less efficient coactivator recruitment on the hapoA-I DR-2 site as compared with a DR-1 PPRE.
Figure 5. Differential PPAR activation and coactivator recruitment by fenofibric acid (FF) and GF depends on the nature of the PPRE. A and B, Cos cells were transfected with pSG5mPPAR (330 ng) and reporter vectors containing (A) the hapoA-I DR-2-PPRE (A3TkpGL3-hA-I, 1100 ng) or (B) the hapoA-II DR-1-PPRE (J6TkLuc, 110 ng) constructs. Cells were subsequently treated with the indicated concentrations of PPAR ligands. C and D, Interaction of DNA-bound hRXR/hPPAR heterodimers with DRIP205 in the presence of fenofibric acid or gemfibrozil (GF; 1, 10, and 100 μmol/L) on the hapoA-I DR-2 (C) or a classical DR-1 PPRE (D). The amount of hPPAR/hRXR bound to coactivators in the presence of the indicated ligands is expressed relative to that measured in absence of ligand (dimethyl sulfoxide [DMSO]) defined as 1.
Discussion
Fibrates are useful drugs in the treatment of dyslipidemia characterized by elevated triglyceride and reduced HDL levels. Several clinical studies have demonstrated the efficacy of fibrate treatment to improve lipid profiles and in reducing the risk of coronary heart disease events. Even though no comparative studies were available, the 2 most widely used fibrates (FF and GF) appear to exert overlapping but distinct effects. Although both FF and GF increase plasma HDL-C, FF efficiently increases plasma apoA-I,4–7,26 whereas GF seems to have little or no effect on this parameter in humans.9–11,27
In this study, we assessed the effect of FF and GF on plasma lipids and apoA-I levels in hyperlipidemic patients. Both FF and GF induced a quantitatively comparable increase in HDL-C and decrease in triglyceride levels. Even though there was no significant difference in the change from baseline in HDL-C levels between FF and GF, a significant increase in apoA-I levels was seen only in FF-treated patients, whereas GF recipients displayed unchanged apoA-I levels from baseline. Although this is the first head-to-head comparison, our data are in line with results from several other clinical trials showing an increase of both HDL-C and hapoA-I concentrations after FF treatment.4–7 Furthermore, recent turnover studies have shown that FF treatment in humans significantly increases apoA-I plasma levels through enhancing its synthesis rate.28,29 By contrast, in the Veterans Affairs High-Density Lipoprotein Intervention Trial, GF treatment, although significantly increasing HDL-C levels, had no effect on apoA-I.9 Similarly, in other trials GF also failed to increase apoA-I concentrations despite a significant increase in HDL-C levels.10,11,27 Accordingly, the present report clearly establishes in a direct comparison that, although having similar effect on HDL-C levels, FF and GF differentially influence apoA-I levels in humans. By contrast, both FF and GF have been shown to induce apoA-II levels and LpA-I:A-II particles in patients,27,30 further emphasizing a selective increase of apoA-I by FF.
It was next determined whether PPAR mediates fibrate action on HDL-C and hapoA-I in hA-ITgPPAR+/+ and PPAR–/– mice. Humanized transgenic apoA-I mice were studied, because HDL metabolism is regulated in an opposite fashion by fibrates in rodents (decrease) and man (increase) due to sequence divergences in the respective apoA-I promoters.21 FF and, albeit to a lesser extent, GF significantly increased plasma HDL-C levels in PPAR+/+hA-ITg mice, whereas no change was observed in the PPAR–/–hA-ITg mice. In addition, and in line with the human situation,30,31 both FF and GF increased plasma apoA-II levels in mice in a PPAR-dependent manner, although no significant change in hepatic mapoA-II gene expression was observed. The appearance of larger HDL particles after fibrate treatment is likely because of increased expression of phospholipid transfer protein18 and decreased scavenger receptor class B type I protein levels in liver.32 It is noteworthy that quantitative differences exist between the 2 fibrates, because GF was less effective than was FF to increase HDL-C levels and change HDL size distribution.
Consistent with the observations in humans, FF treatment strongly increased hapoA-I plasma and hepatic mRNA levels in a PPAR-dependent manner. Similar results were obtained in ciprofibrate-treated mice (data not shown). Surprisingly, although GF administration to PPAR+/+hA-ITg mice led to increased HDL-C levels, plasma apoA-I remained unchanged. Interestingly, GF treatment did not influence hepatic hapoA-I mRNA levels, whereas mRNA levels of ACO, a classical PPAR target gene,33 were increased to a similar extent by FF and GF. It is important to note that these fibrates were administered at doses and for a time period allowing maximal in vivo effectiveness in rodent.16 Furthermore, the observed differences in apoA-I gene expression after GF and FF were not related to differences in the administered doses (0.2% FF versus 0.5% GF), because FF and GF induced ACO mRNA levels to a similar extent (x5.3 versus x4.7, respectively) indicating equipotent activity in the liver. Altogether, these data establish that PPAR mediates both fibrate effects in our study and suggest that GF behaves as a selective modulator of PPAR in vivo.
The ability of FF and GF to activate PPAR was examined using a reporter vector driven by either a DR-1 or the apoA-I DR-2–type PPREs. FF fully induced PPAR transactivation, whereas GF only elicited a partial transactivation on the hapoA-I DR-2 PPRE. Both GF and FF exhibited similar affinities for PPAR. By contrast, GF and FF were equally active on a DR-1–type PPRE. Detailed analysis has revealed that the binding of PPAR to the hapoA-I promoter occurs through a DR-2–type PPRE within the A site,34 whereas the ACO site is a DR-1 response element.25 Our data show that GF behaves as a partial or full PPAR agonist depending on the type of PPRE, whereas FF fully induces PPAR transactivation activity on both DR-1 and DR-2 sites.
To understand the role of the PPRE structure in the distinct action of FF and GF, the ability of both ligands to induce coactivator recruitment to PPAR on DR-1 versus DR-2 PPREs was compared. Our data indicate that GF is less efficient to induce DRIP205 recruitment to the PPAR/RXR heterodimer bound to the DR-2. By contrast, FF elicited similar recruitment efficiency to both types of complexes. This indicates that GF induces differential coactivator recruitment to the PPAR/RXR-DNA complex depending on the geometry of the PPRE. This is in line with previous reports showing that the nature of the response element plays a crucial role by inducing allosteric conformational changes of the receptor, leading to distinct coactivator recruitment, receptor activity, and gene response.35 Whether DRIP205 is the coactivator explaining the different in vivo responses is impossible to answer. However, it can be concluded from our data that weaker coactivator recruitment to PPAR by GF, because of distinct PPRE architecture, is likely to be the basis of the selective modulation of gene expression by GF in vivo.
Another important observation of this study is that PPAR expression levels determine the response to fibrates. When treated with FF, PPAR+/–hA-ITg mice showed an intermediate response with respect to plasma TC, HDL-C, and hapoA-I plasma and liver mRNA levels. Thus, our data provide genetic evidence that PPAR expression levels determine the response to fibrates. Because we previously reported that human liver PPAR levels vary >2-fold among individuals,36 it is likely that in humans also the response to fibrate treatment will differ between individuals because of differences in PPAR expression levels. In conclusion, this article strongly supports the fact that GF behaves as a selective PPAR modulator in humans. It is interesting to note that GF is chemically different from other fibrates.37 Although differences in pharmacokinetics and metabolism between fibrates may also contribute to distinct clinical phenotypes and even though GF may also have additional effects,38 our data provide a plausible mechanism for the clinically observed differences between FF and GF. Our observations may have potential significant implications with regard to pharmacological effects of fibrates, and future studies may be aimed at identifying selective PPAR modulators to improve therapeutic efficacy.
Acknowledgments
This multicenter clinical study was funded by an unrestricted grant from Laboratoires FournierSA. This work was supported by grants from INSERM, the Région Nord/Pas-de-Calais (Génop?le #01360124), and the Leducq Foundation. I.P.T. was supported by a European Community grant (ERBFMBICT983214). All lipid and safety tests were centrally measured in the Department of Laboratory Medicine (Hamilton Civic Hospitals, Hamilton, Ontario, Canada) under the supervision of Dr Darius Nazir. We thank J.-C. Ansquer, S. Robins, and J.-P. Després for helpful discussion and J. Fremaux, B. Derudas, A. Pilon, M. Bouly, M. Coevoet, and H. DeSafta for excellent technical assistance. We greatly acknowledge the clinical study investigators: Dr Gordon A. Francis (Department of Medicine, Division of Cardiology, University of Alberta, Edmonton, Alberta); Dr Alan L. Edward and Dr Charlotte A. Jones (Department of Medicine, Calgary General Hospital, Calgary, Alberta); Dr Gregory P. Curnew (Lipid Research Clinic, Hamilton Civic Hospitals, Hamilton, Ontario); Dr Lawrence A. Leiter (Department of Endocrinology, St Michael’s Hospital, Toronto, Ontario); Dr Teik Chye Ooi (Department of Medicine, Division of Endocrinology and Metabolism, Ottawa Hospital Riverside Campus, Ottawa, Ontario); Dr Mark H. Sherman (Department of Medicine, Division of Endocrinology and Metabolism, Royal Victoria Hospital, Montréal, Québec); Dr Bruno St-Pierre and Dr Chantal Godin (Department of Medicine, Centre Hospitalier Hotel-Dieu de Sherbrooke, Québec); Dr Patrick T.S. Ma and Dr Neil George Filipchuk (Department of Cardiology, Calgary General Hospital, Calgary, Alberta); and Dr Jorge Federico Bonet and Dr Gordon Neil Hoag (Lipid clinic, Victoria General Hospital, Victoria, British Columbia).
References
Robins SJ. PPAR ligands and clinical trials: cardiovascular risk reduction with fibrates. J Cardiovasc Risk. 2001; 8: 195–201.
Rubin EM, Krauss RM, Spangler EA, Verstuyft JG, Clift SM. Inhibition of early atherogenesis in transgenic mice by human apolipoprotein AI. Nature. 1991; 353: 265–267.
Nissen SE, Tsunoda T, Tuzcu EM, Schoenhagen P, Cooper CJ, Yasin M, Eaton GM, Lauer MA, Sheldon WS, Grines CL, Halpern S, Crowe T, Blankenship JC, Kerensky R. Effect of recombinant ApoA-I Milano on coronary atherosclerosis in patients with acute coronary syndromes: a randomized controlled trial. JAMA. 2003; 290: 2292–2300.
Fruchart JC, Davignon J, Bard JM, Grothe AM, Richard A, Fievet C. Effect of fenofibrate treatment on type III hyperlipoproteinemia. Am J Med. 1987; 83: 71–74.
Mellies MJ, Stein EA, Khoury P, Lamkin G, Glueck CJ. Effects of fenofibrate on lipids, lipoproteins, and apolipoproteins in 33 subjects with primary hypercholesterolemia. Atherosclerosis. 1987; 63: 57–64.
Goldberg AC, Schonfeld G, Feldman EB, Ginsberg HN, Hunninghake DB, Insull W Jr, Knopp RH, Kwiterovich PO, Mellies MJ, Pickering J. Fenofibrate for the treatment of type IV and V hyperlipoproteinemias: a double-blind, placebo-controlled multicenter US study. Clin Ther. 1989; 11: 69–83.
Steinmetz A, Schwartz T, Hehnke U, Kaffarnik H. Multicenter comparison of micronized fenofibrate and simvastatin in patients with primary type IIA or IIB hyperlipoproteinemia. J Cardiovasc Pharmacol. 1996; 27: 563–570.
Poulter N. The impact of micronized fenofibrate on lipid subfractions and on reaching HDL-target levels in 7,098 patients with dyslipidemia. Br J Cardiol. 1999; 6: 682–685.
Robins SJ, Collins D, Wittes JT, Papademetriou V, Deedwania PC, Schaefer EJ, McNamara JR, Kashyap ML, Hershman JM, Wexler LF, Rubins HB. Relation of gemfibrozil treatment and lipid levels with major coronary events: VA-HIT: a randomized controlled trial. JAMA. 2001; 285: 1585–1591.
Durrington PN, Mackness MI, Bhatnagar D, Julier K, Prais H, Arrol S, Morgan J, Wood GN. Effects of two different fibric acid derivatives on lipoproteins, cholesteryl ester transfer, fibrinogen, plasminogen activator inhibitor and paraoxonase activity in type IIb hyperlipoproteinaemia. Atherosclerosis. 1998; 138: 217–225.
Schaefer EJ, Lamon-Fava S, Cole T, Sprecher DL, Cilla DD Jr, Balagtas CC, Rowan JP, Black DM. Effects of regular and extended-release gemfibrozil on plasma lipoproteins and apolipoproteins in hypercholesterolemic patients with decreased HDL cholesterol levels. Atherosclerosis. 1996; 127: 113–122.
Camp HS, Li O, Wise SC, Hong YH, Frankowski CL, Shen X, Vanbogelen R, Leff T. Differential activation of peroxisome proliferator-activated receptor-gamma by troglitazone and rosiglitazone. Diabetes. 2000; 49: 539–547.
Rubin EM, Ishida BY, Clift SM, Krauss RM. Expression of human apolipoprotein A-I in transgenic mice results in reduced plasma levels of murine apolipoprotein A-I and the appearance of two new high density lipoprotein size subclasses. Proc Natl Acad Sci U S A. 1991; 88: 434–438.
Lee SS, Pineau T, Drago J, Lee EJ, Owens JW, Kroetz DL, Fernandez-Salguero PM, Westphal H, Gonzalez FJ. Targeted disruption of the alpha isoform of the peroxisome proliferator-activated receptor gene in mice results in abolishment of the pleiotropic effects of peroxisome proliferators. Mol Cell Biol. 1995; 15: 3012–3022.
Akiyama TE, Nicol CJ, Fievet C, Staels B, Ward JM, Auwerx J, Lee SS, Gonzalez FJ, Peters JM. Peroxisome proliferator-activated receptor- regulates lipid homeostasis, but is not associated with obesity: studies with congenic mouse lines. J Biol Chem. 2001; 276: 39088–39093.
Staels B, van Tol A, Andreu T, Auwerx J. Fibrates influence the expression of genes involved in lipoprotein metabolism in a tissue-selective manner in the rat. Arterioscler Thromb. 1992; 12: 286–294.
Berthou L, Duverger N, Emmanuel F, Langouet S, Auwerx J, Guillouzo A, Fruchart JC, Rubin E, Denefle P, Staels B, Branellec D. Opposite regulation of human versus mouse apolipoprotein A-I by fibrates in human apolipoprotein A-I transgenic mice. J Clin Invest. 1996; 97: 2408–2416.
Bouly M, Masson D, Gross B, Jiang XC, Fievet C, Castro G, Tall AR, Fruchart JC, Staels B, Lagrost L, Luc G. Induction of the phospholipid transfer protein gene accounts for the high density lipoprotein enlargement in mice treated with fenofibrate. J Biol Chem. 2001; 276: 25841–25847.
Staels B, Auwerx J. Perturbation of developmental gene expression in rat liver by fibric acid derivatives: lipoprotein lipase and alpha-fetoprotein as models. Development. 1992; 115: 1035–1043.
Issemann I, Green S. Activation of a member of the steroid hormone receptor superfamily by peroxisome proliferators. Nature. 1990; 347: 645–650.
Vu-Dac N, Chopin-Delannoy S, Gervois P, Bonnelye E, Martin G, Fruchart JC, Laudet V, Staels B. The nuclear receptors peroxisome proliferator-activated receptor and Rev-erbalpha mediate the species-specific regulation of apolipoprotein A-I expression by fibrates. J Biol Chem. 1998; 273: 25713–25720.
Raspe E, Madsen L, Lefebvre AM, Leitersdorf I, Gelman L, Peinado-Onsurbe J, Dallongeville J, Fruchart JC, Berge R, Staels B. Modulation of rat liver apolipoprotein gene expression and serum lipid levels by tetradecylthioacetic acid (TTA) via PPAR activation. J Lipid Res. 1999; 40: 2099–2110.
Lefebvre B, Mouchon A, Formstecher P, Lefebvre P. Distinct modes of interaction of the retinoic acid receptor alpha with natural and synthetic retinoids. Mol Cell Endocrinol. 1998; 139: 161–169.
Mouchon A, Delmotte MH, Formstecher P, Lefebvre P. Allosteric regulation of the discriminative responsiveness of retinoic acid receptor to natural and synthetic ligands by retinoid X receptor and DNA. Mol Cell Biol. 1999; 19: 3073–3085.
Schoonjans K, Staels B, Auwerx J. The peroxisome proliferator activated receptors(PPARS) and their effects on lipid metabolism and adipocyte differentiation. Biochim Biophys Acta. 1996; 1302: 93–109.
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 hyperlipidemic patients: an open Belgian multicenter study. Atherosclerosis. 1994; 110: S49–S54.
Sakai T, Kamanna VS, Kashyap ML. Niacin, but not gemfibrozil, selectively increases LP-AI, a cardioprotective subfraction of HDL, in patients with low HDL cholesterol. Arterioscler Thromb Vasc Biol. 2001; 21: 1783–1789.
Watts GF, Barrett PH, Ji J, Serone AP, Chan DC, Croft KD, Loehrer F, Johnson AG. Differential regulation of lipoprotein kinetics by atorvastatin and fenofibrate in subjects with the metabolic syndrome. Diabetes. 2003; 52: 803–811.
Bilz S, Wagner S, Schmitz M, Bedynek A, Keller U, Demant T. Effects of atorvastatin versus fenofibrate on apoB-100 and apoA-I kinetics in mixed hyperlipidemia. J Lipid Res. 2004; 45: 174–185.
Bard JM, Parra HJ, Luc G, Camare R, Ziegler O, Dachet C, Bruckert E, Douste-Blazy P, Drouin P, Jacotot B. Lipoprotein particle analysis comparing simvastatin and fenofibrate. Atherosclerosis. 1991; 9: S29–S34.
Kahri J, Sane T, van Tol A, Taskinen MR. Effect of gemfibrozil on the regulation of HDL subfractions in hypertriglyceridaemic patients. J Intern Med. 1995; 238: 429–436.
Mardones P, Pilon A, Bouly M, Duran D, Nishimoto T, Arai H, Kozarsky KF, Altayo M, Miquel JF, Luc G, Clavey V, Staels B, Rigotti A. Fibrates down-regulate hepatic scavenger receptor class B type I protein expression in mice. J Biol Chem. 2003; 278: 7884–7890.
Dreyer C, Keller H, Mahfoudi A, Laudet V, Krey G, Wahli W. Positive regulation of the peroxisomal beta-oxidation pathway by fatty acids through activation of peroxisome proliferator-activated receptors (PPAR). Biol Cell. 1993; 77: 67–76.
Klausing KS, Briggs, M. The human apolipoprotein A1 promoter contains a novel PPRE. Circulation. 1995; 92: (suppl I): 295. Abstract.
Hall JM, McDonnell DP, Korach KS. Allosteric regulation of estrogen receptor structure, function, and coactivator recruitment by different estrogen response elements. Mol Endocrinol. 2002; 16: 469–486.
Gervois P, Torra IP, Chinetti G, Grotzinger T, Dubois G, Fruchart JC, Fruchart-Najib J, Leitersdorf E, Staels B. A truncated human peroxisome proliferator-activated receptor alpha splice variant with dominant negative activity. Mol Endocrinol. 1999; 13: 1535–1549.
Todd PA, Ward A. Gemfibrozil. A review of its pharmacodynamic and pharmacokinetic properties, and therapeutic use in dyslipidaemia. Drugs. 1988; 36: 314–339.
Jin FY, Kamanna VS, Chuang MY, Morgan K, Kashyap ML. Gemfibrozil stimulates apolipoprotein A-I synthesis and secretion by stabilization of mRNA transcripts in human hepatoblastoma cell line(HepG2). Arterioscler Thromb Vasc Biol. 1996; 16: 1052–1062.(Hélène Duez; Bruno Lefebv)
Correspondence to Hélène Duez or Prof Bart Staels, UR545INSERM, Département d’Athérosclérose, Institut Pasteur Lille, 1 rue Calmette, 59019 Lille, France. E-mail helene.duez@pasteur-lille.fr or bart.staels@pasteur-lille.fr
Abstract
Objective— The objective of this trial was to study the effects of fenofibrate (FF) and gemfibrozil (GF), the most commonly used fibrates, on high-density lipoprotein (HDL) and apolipoprotein (apo) A-I.
Methods and Results— In a head-to-head double-blind clinical trial, both FF and GF decreased triglycerides and increased HDL cholesterol levels to a similar extent, whereas plasma apoA-I only increased after FF but not GF. Results in human (h) apoA-Itransgenic (hA-ITg) peroxisome proliferator-activated receptor (PPAR) –/– mice demonstrated that PPAR mediates the effects of FF and GF on HDL in vivo. Although plasma and hepatic mRNA levels of hapoA-I increased more pronouncedly after FF than GF in hA-ITgPPAR+/+ mice, both fibrates induced acylCoAoxidase mRNA similarly. FF and GF transactivated PPAR with similar activity and affinity on a DR-1 PPAR response element, but maximal activation on the hapoA-I DR-2 PPAR response element was significantly lower for GF than for FF. Moreover, GF induced recruitment of the coactivator DRIP205 on the DR-2 site less efficiently than did FF.
Conclusion— Both GF and FF exert their effects on HDL through PPAR. Whereas FF behaves as a full agonist, GF appears to act as a partial agonist due to a differential recruitment of coactivators to the promoter. These observations provide an explanation for the differences in the activity of these fibrates on apoA-I.
In hyperlipidemic patients and human apolipoprotein A-Itransgenic (hA-ITg) mice, fenofibrate (FF) and gemfibrozil (GF) increase HDL cholesterol, whereas apoA-I levels only increased after FF. Because of differential coactivator recruitment to the promoter, FF and GF, respectively, behave as full and partial PPAR agonists, likely explaining the clinical differences in the activity of these fibrates on apoA-I.
Key Words: apolipoprotein A-I ? high-density lipoprotein ? fibrates ? peroxisome proliferator-activated receptor ? selective PPAR modulator
Introduction
Several clinical trials have indicated a benefit of fibrate treatment in cardiovascular risk reduction and improvement of lipid profiles.1 Angiography trials have shown that administration of fibrates (bezafibrate, gemfibrozil [GF], and fenofibrate [FF]) significantly decreases the progression of coronary narrowing.1 Furthermore, fibrate treatment decreased combined incidence of coronary heart disease events in primary and secondary prevention trials.1 These trials also demonstrated the effectiveness of fibrates in improving dyslipidemia by lowering elevated plasma triglycerides, improving the low-density lipoprotein subfraction distribution, and increasing high-density lipoprotein cholesterol (HDL-C). However, qualitative and quantitative differences appear to exist between fibrates. Qualitative differences in pharmacological modulation of HDL metabolism, and its major apolipoprotein (apo) A-I that confers protection against the development of atherosclerosis in animal models2 and humans,3 may be of clinical importance. In several clinical trials, FF increases plasma HDL-C and apoA-I concentrations.4–8 By contrast, although significantly increasing HDL-C levels, GF seems to have little or no effect on apoA-I levels.9–11 Thus, the 2 most used fibrates (FF and GF), although exerting similar beneficial actions on plasma lipids, may also exert distinct effects by yet unresolved mechanisms.
Even though fibrates are low-affinity ligands, their effects are thought to be mediated principally through peroxisome proliferator-activated receptor (PPAR) . Fibrate-activated PPAR regulates gene expression after heterodimerization with the retinoid X receptor (RXR), subsequent coactivator recruitment, and binding to PPAR response elements (PPREs) which are located in target gene promoters. The PPAR/RXR heterodimer interacts with coactivators that act as bridges with the basal transcription machinery and play crucial roles in the transmission of regulatory signals. Different ligands can induce different receptor conformations in a way that is unique to each ligand, leading to differential coactivator recruitment. This is the case for PPAR agonists. Compared with other glitazones, troglitazone behaves as a partial or full agonist for PPAR depending on the cellular environment and promoter context leading to PPAR selective modulation and differential downstream effects on gene expression.12 This selective PPAR modulator concept may explain differences in biological activity.
In this study the clinical effects of FF and GF were compared on lipid metabolism with a focus on their effects on HDL and apoA-I in a head-to-head trial. Moreover, the molecular mechanisms underlying observed differences were investigated.
Methods
Clinical Study
A multicenter, double-blind, randomized, 24-week study was performed comparing head-to-head the usual doses of micronized FF (200 mg per day) and GF (1200 mg per day). After an 8-week placebo run-in period, 234 patients (82 women, 152 men) aged 18 to 70 years (mean age 51.6 years) with combined hyperlipidemia (total cholesterol [TC]/HDL-C 6.0 mmol/L [males], 5.6 mmol/L [females]; triglycerides between 2.2 and 5.6 mmol/L) were randomized and treated with FF (n=116) or GF (n=118). Both drugs were equally well tolerated. For triglyceride, HDL-C, and apoA-I measurement, please see the online Methods, available at http://atvb.ahajournals.org.
Comparison of percent changes was done by 1-way analysis of variance. The study was approved by the independent ethics committee of each of the 10 participating Canadian centers.
Animals
Hemizygous hA-ITg mice on a PPAR–/–, +/–, and +/+ background were obtained by crossing homozygous human (h) apoA-Itransgenic (hA-ITg) mice13 with PPAR-null mice14,15 and subsequently treated with either FF or GF (0.2 and 0.5% w/w, respectively) for 17 days, a time at which a steady-state response is reached16 (see online Methods). All animal experiments were approved by the Institutional Animal Care and Use Committee.
Lipid, Apolipoprotein, and Lipoprotein Analyses
Plasma TC was determined as described.15 Plasma human (h) apoA-I and mouse apoA-II levels were measured by an immunonephelometric assay using species-specific antibodies.17 Lipoprotein cholesterol distribution in pooled plasma samples was assessed by fast protein liquid.18 hapoA-I was measured in each eluted fraction using the same immunonephelometric assay as above,17 and HDL fractions were defined as apoA-I containing fractions. HDL-C was determined from the cholesterol distribution profiles by calculating the areas under the curves of the apoA-I-containing fractions.
RNA Analysis
RNA was isolated and Northern blot analysis was performed as described19 using hapoA-I and rat acylCoAoxidase (ACO) probes and human ribosomal 28S cDNA as a control probe. hapoA-I, ACO, and mouse apoA-II levels were quantified by reverse transcription followed by real-time polymerase chain reaction, normalized to the 28S internal control, and expressed as fold induction over the untreated control group, set as 1 (see online Methods).
Statistical Analysis
ANOVA was used to compare differences between treatment groups. When overall significance was attained, Scheffe test was used to analyze for significant differences between the experimental groups.
Transient Transfection Assays
The pSG5-mPPAR plasmid was as described.20 The hapoA-II-DR-1 (J6-TK-Luc) and hapoA-I-DR-2 (A3-TK-pGL3)–containing reporter plasmids were obtained by inserting 6 or 3 copies of the hapoA-II PPRE (DR-1)–containing J or the hapoA-I PPRE (A site)21 sites in the pTK-pGL3 plasmid, respectively (see online Methods).
Cos-7 cells were transfected using polyethylenimine before the addition of increasing concentrations of the compound tested or vehicle for 24 hours in 0.2% FCS-DMEM (see online Methods). The pCMV-?-galactosidase expression plasmid was cotransfected as a control for transfection efficiency. Luciferase and ?-galactosidase assays were performed as described.22 EC50 values were estimated using Prism software(GraphPad).
Cofactor Recruitment Assays
The glutathione-S-transferase pull-down protocol was as described.23 DNA-dependent protein-protein interactions were tested as described previously.24 Briefly, hPPAR and 35S-labeled-hRXR were incubated with or without PPAR ligand before a double-stranded oligonucleotide corresponding to the ACO or hapoA-I PPRE and glutathione-S-transferase-DRIP205 were added (see online Methods). Complexes were resolved and quantified with a PhosphorImager (Molecular Dynamics).
Results
FF and GF Regulate ApoA-I Differently in Humans
To compare the effects of FF and GF on plasma lipids and apoA-I, 234 patients with combined hyperlipidemia were treated with FF (n=116) or GF (n=118) in a head-to-head, double-blind, randomized trial. Baseline characteristics of patients are given in Table I (available online at http://atvb.ahajournals.org).
Both fibrates reduced triglycerides (FF, –39±22% versus GF, –41±26%, P>0.5) and increased HDL-C (FF, +16±20% versus GF, +12±21%, P>0.5) to a similar extent (Figure 1). Interestingly, although having similar HDL-C increasing activities, only FF treatment increased apoA-I plasma levels, whereas GF treatment was without effect (FF, +8.6±14% versus GF, +1.5±12%, P<0.001; Figure 1).
Figure 1. Effects of FF or GF treatment on HDL-C, triglyceride, and apoA-I plasma levels. Hyperlipidemic patients were treated with either fenofibrate (FF; 200 mg/d, n=116, black bars) or gemfibrozil (GF; 1200 mg per day, n=118, gray bars) in a direct 24-week head-to-head comparison study. ApoA-I (FF, +8.6±14% vs GF, +1.5±12%), HDL cholesterol (HDL-C; FF, +16±20% vs GF, +12±21%), and triglyceride (TG; FF, –39±22% vs GF, –41±26%) plasma levels are reported as mean percent change from baseline. (1-way analysis of variance, ***P<0.001).
PPAR Mediates the Effects of FF on Plasma HDL-C and hapoA-I Plasma and Hepatic mRNA Levels In Vivo in Mice
To determine the mechanistic basis of this differential effect, we first aimed to establish whether PPAR is required for the in vivo regulation of HDL-C and hapoA-I levels by fibrates. hapoA-I-transgenic (hA-ITg) mice on a PPAR+/+, +/–, and –/– background were treated with FF for 17 days. In PPAR+/+hA-ITg mice, FF treatment resulted in significant 4- and 3.5-fold increases of plasma TC and hapoA-I levels (TC: control [CON], 159±9 versus FF, 618±31 mg/dL, P<0.001; hapoA-I: CON, 411±13 versus FF,1503±117 mg/dL, P<0.001). By contrast, no increase was observed in the PPAR–/–hA-ITg mice (TC: CON, 152±8 versus FF, 108±7 mg/dL, P>0.05; hapoA-I: CON, 315±19 versus FF, 268±8 mg/dL, P>0.05; Figure 2 A and 2B). Strikingly, treatment with FF increased TC (CON, 145±9 versus FF, 328±29 mg/dL, P<0.001) and hapoA-I (CON, 322±20 versus FF, 625±51 mg/dL, P<0.05) to intermediate levels in PPAR+/–hA-ITg mice compared with PPAR+/+ and PPAR–/– mice (Figure 2A and 2B). Consistently, changes in HDL-C distribution profiles in the PPAR+/–hA-ITg mice occurred in an intermediate fashion when compared with the 2 other PPAR genotypes (data not shown). In agreement with the above results, FF treatment induced a marked increase in hepatic hapoA-I mRNA levels in the PPAR+/+hA-ITg mice (CON, 100±15 versus FF, 226±10%, P<0.001), whereas no effect was observed in the PPAR–/–hA-ITg mice (CON, 100±20 versus FF, 121±8%, P>0.05; Figure I, available online at http://atvb.ahajournals.org), and an intermediate effect was observed in the PPAR+/–hA-ITg mice (CON, 100±7 versus FF, 173±25%, P<0.01). Altogether these data indicate that the increase of HDL-C and hapoA-I plasma and hepatic mRNA levels by FF in hA-ITg mice occurs in a PPAR gene dosage-dependent manner.
Figure 2. PPAR is required for the in vivo regulation of plasma cholesterol and hapoA-I by FF in a gene dosage-dependent manner. Plasma TC (A) and hapoA-I (B) levels in PPAR+/+hA-ITg (CON, n=10; FF, n=10), PPAR+/–hA-I Tg (CON, n=7; FF, n=9) and PPAR–/–hA-ITg (CON, n=13; FF, n=13) mice treated with control (CON; white bars) or fenofibrate (FF)-supplemented (black bars) chow. The results are expressed as percentage (mean±SEM) of untreated control mice. Statistically significant differences between groups are indicated by asterisks (Scheffe test, *P<0.05, ***P<0.001).
FF and GF Exert Distinct PPAR-Dependent Effects on HDL-C and hapoA-I Plasma Levels in Mice
The effect of FF and GF on the in vivo regulation of HDL-C and hapoA-I was next assessed in hA-ITgPPAR+/+ and PPAR–/–mice. FF and, albeit to a lower extent, GF significantly increased plasma TC in PPAR+/+hA-ITg mice (CON, 158±7 versus FF, 498±31 mg/dL, P<0.001; CON, 158±7 versus GF, 224±13 mg/dL, P<0.01; Figure 3A), whereas this effect was not observed in the PPAR–/– background (Figure II, available online at http://atvb.ahajournals.org). A large increase in HDL-C was observed in PPAR+/+hA-ITg mice treated with FF and GF, although the effect of GF was again less pronounced compared with FF (1.6-fold versus 3.5-fold, respectively; Figure 3B). No change was observed in the PPAR–/–hA-ITg mice (Figure II). Furthermore, FF and GF treatment led to the appearance of a larger peak overlapping the low-density lipoprotein and HDL size range in a PPAR-dependent manner.
Figure 3. FF but not GF induces plasma hapoA-I levels in a PPAR-dependent manner. A, Plasma TC and hapoA-I levels in PPAR+/+hA-ITg mice (CON, n=6; FF, n=6; GF, n=8) treated with control (CON; white bars), fenofibrate (FF; black bars) or gemfibrozil (GF; gray bars)-supplemented chow. The results are expressed as percentage (mean±SEM) of untreated control mice. Statistically significant differences between groups are indicated by asterisks (Scheffe test, **P<0.01, ***P<0.001). B, Representative gel filtration chromatography distribution profiles of cholesterol in plasma from PPAR+/+hA-ITg mice treated with control () or FF () or GF ()-supplemented chow. "HDL" fractions are defined as apoA-I-containing fractions (CON, 55 to 70; FF, 43 to 62; GF, 50 to 70).
FF increased plasma hapoA-I in PPAR+/+hA-ITg mice (CON, 357±12 versus FF, 886±181 mg/dL, P<0.001; Figure 3A), an effect that was not observed in the PPAR–/– background (CON, 304±20 versus FF, 253±8 mg/dL, P>0.05). In contrast to FF, plasma hapoA-I levels were only slightly modified by GF treatment in the PPAR+/+hA-ITg mice (CON, 357±12 versus GF, 394±12 mg/dL, P>0.05; Figure 3A). Altogether these data demonstrate that both GF and FF induce HDL-C levels in a PPAR-dependent manner. However, in contrast to FF, GF has only a marginal effect on plasma hapoA-I.
FF Induces hapoA-I to a Larger Extent Than GF, Whereas Both Fibrates Increase ACO mRNA Levels Similarly in Mice
In line with the plasma changes, FF significantly increased hapoA-I mRNA levels in PPAR+/+ mice (Figure 4) but was ineffective in the PPAR–/–hA-ITg mice. However, GF treatment was not efficient to induce hepatic hapoA-I mRNA levels in PPAR+/+hA-ITg mice (Figure 4). By contrast, mRNA levels of ACO, a DR-1–driven PPAR target gene,25 were increased by GF treatment to a similar extent as FF (4.7- and 5.3-fold increase, respectively) in PPAR+/+hA-ITg mice, whereas no change was observed with either treatment in the PPAR–/–hA-ITg mice (Figure 4). Hepatic mRNA levels of mouse apoA-II, another important protein constituent of HDL, were also measured. Neither FF nor GF significantly affected mouse (m) apoA-II mRNA levels in PPAR+/+hA-ITg mice (FF, 110±30%; GF, 121±28%; CON, 100±10%). However, both FF and GF treatment increased mapoA-II plasma levels in PPAR+/+hA-ITg mice (FF, 136±20%, P<0.01; GF, 200±65%, P<0.01; CON, 100±19%), whereas no change was observed with either FF or GF in the PPAR–/–hA-ITg mice (FF, 102±23%, P>0.05; GF, 128±18%, P>0.05; CON, 100±10%). Thus, GF appears to behave as a selective PPAR modulator in vivo.
Figure 4. In vivo differential effects of FF and GF on hepatic hapoA-I but not on ACO mRNA levels. Representative Northern blot showing liver hapoA-I and ACO mRNA levels in PPAR+/+hA-ITg and PPAR–/–hA-ITg mice treated with control (CON), or fenofibrate (FF), or gemfibrozil (GF)-supplemented chow. Values reported below the blots are the induction levels of hapoA-I and ACO mRNA in treated vs control PPAR+/+hA-ITg mice quantified by quantitative polymerase chain reaction.
GF and FF Distinctly Induce PPAR Activation and Coactivator Recruitment in a PPRE-Dependent Manner
The PPAR activation profile of GF and FF was determined in a transfection assay in which full-length PPAR was cotransfected with reporter constructs driven by either the hapoA-I DR-2 or a DR-1 PPRE (Figure 5A and 5B). GF and FF increased PPAR activity from the hapoA-I DR-2 site with similar EC50 values (5.8 and 7.8 μmol/L, respectively). However, a submaximal response was observed with GF compared with FF (Figure 5A). By contrast, on the DR-1 PPRE-driven vector, FF and GF activated PPAR with similar EC50 values (12 μmol/L and 11 μmol/L, respectively) and similar maximal response (Figure 5B). These data indicate that GF behaves as a partial agonist compared with FF depending on the type of PPRE. Next, the influence of FF and GF on coactivator recruitment to PPAR/RXR bound to either a DR-1 or the hapoA-I DR-2 site was examined. DRIP205 was chosen as a model system, but similar results were obtained with other coactivators, that is, p300, CBP (CREB binding protein), and glucocorticoid receptor interacting protein (data not shown). Compared with FF, GF led to at least as efficient coactivator recruitment to the DR-1 site but was comparatively less efficient on the hapoA-I DR-2 site (Figure 5C and 5D). Taken together, these findings suggest that GF selectively modulates PPAR transcriptional activity associated with less efficient coactivator recruitment on the hapoA-I DR-2 site as compared with a DR-1 PPRE.
Figure 5. Differential PPAR activation and coactivator recruitment by fenofibric acid (FF) and GF depends on the nature of the PPRE. A and B, Cos cells were transfected with pSG5mPPAR (330 ng) and reporter vectors containing (A) the hapoA-I DR-2-PPRE (A3TkpGL3-hA-I, 1100 ng) or (B) the hapoA-II DR-1-PPRE (J6TkLuc, 110 ng) constructs. Cells were subsequently treated with the indicated concentrations of PPAR ligands. C and D, Interaction of DNA-bound hRXR/hPPAR heterodimers with DRIP205 in the presence of fenofibric acid or gemfibrozil (GF; 1, 10, and 100 μmol/L) on the hapoA-I DR-2 (C) or a classical DR-1 PPRE (D). The amount of hPPAR/hRXR bound to coactivators in the presence of the indicated ligands is expressed relative to that measured in absence of ligand (dimethyl sulfoxide [DMSO]) defined as 1.
Discussion
Fibrates are useful drugs in the treatment of dyslipidemia characterized by elevated triglyceride and reduced HDL levels. Several clinical studies have demonstrated the efficacy of fibrate treatment to improve lipid profiles and in reducing the risk of coronary heart disease events. Even though no comparative studies were available, the 2 most widely used fibrates (FF and GF) appear to exert overlapping but distinct effects. Although both FF and GF increase plasma HDL-C, FF efficiently increases plasma apoA-I,4–7,26 whereas GF seems to have little or no effect on this parameter in humans.9–11,27
In this study, we assessed the effect of FF and GF on plasma lipids and apoA-I levels in hyperlipidemic patients. Both FF and GF induced a quantitatively comparable increase in HDL-C and decrease in triglyceride levels. Even though there was no significant difference in the change from baseline in HDL-C levels between FF and GF, a significant increase in apoA-I levels was seen only in FF-treated patients, whereas GF recipients displayed unchanged apoA-I levels from baseline. Although this is the first head-to-head comparison, our data are in line with results from several other clinical trials showing an increase of both HDL-C and hapoA-I concentrations after FF treatment.4–7 Furthermore, recent turnover studies have shown that FF treatment in humans significantly increases apoA-I plasma levels through enhancing its synthesis rate.28,29 By contrast, in the Veterans Affairs High-Density Lipoprotein Intervention Trial, GF treatment, although significantly increasing HDL-C levels, had no effect on apoA-I.9 Similarly, in other trials GF also failed to increase apoA-I concentrations despite a significant increase in HDL-C levels.10,11,27 Accordingly, the present report clearly establishes in a direct comparison that, although having similar effect on HDL-C levels, FF and GF differentially influence apoA-I levels in humans. By contrast, both FF and GF have been shown to induce apoA-II levels and LpA-I:A-II particles in patients,27,30 further emphasizing a selective increase of apoA-I by FF.
It was next determined whether PPAR mediates fibrate action on HDL-C and hapoA-I in hA-ITgPPAR+/+ and PPAR–/– mice. Humanized transgenic apoA-I mice were studied, because HDL metabolism is regulated in an opposite fashion by fibrates in rodents (decrease) and man (increase) due to sequence divergences in the respective apoA-I promoters.21 FF and, albeit to a lesser extent, GF significantly increased plasma HDL-C levels in PPAR+/+hA-ITg mice, whereas no change was observed in the PPAR–/–hA-ITg mice. In addition, and in line with the human situation,30,31 both FF and GF increased plasma apoA-II levels in mice in a PPAR-dependent manner, although no significant change in hepatic mapoA-II gene expression was observed. The appearance of larger HDL particles after fibrate treatment is likely because of increased expression of phospholipid transfer protein18 and decreased scavenger receptor class B type I protein levels in liver.32 It is noteworthy that quantitative differences exist between the 2 fibrates, because GF was less effective than was FF to increase HDL-C levels and change HDL size distribution.
Consistent with the observations in humans, FF treatment strongly increased hapoA-I plasma and hepatic mRNA levels in a PPAR-dependent manner. Similar results were obtained in ciprofibrate-treated mice (data not shown). Surprisingly, although GF administration to PPAR+/+hA-ITg mice led to increased HDL-C levels, plasma apoA-I remained unchanged. Interestingly, GF treatment did not influence hepatic hapoA-I mRNA levels, whereas mRNA levels of ACO, a classical PPAR target gene,33 were increased to a similar extent by FF and GF. It is important to note that these fibrates were administered at doses and for a time period allowing maximal in vivo effectiveness in rodent.16 Furthermore, the observed differences in apoA-I gene expression after GF and FF were not related to differences in the administered doses (0.2% FF versus 0.5% GF), because FF and GF induced ACO mRNA levels to a similar extent (x5.3 versus x4.7, respectively) indicating equipotent activity in the liver. Altogether, these data establish that PPAR mediates both fibrate effects in our study and suggest that GF behaves as a selective modulator of PPAR in vivo.
The ability of FF and GF to activate PPAR was examined using a reporter vector driven by either a DR-1 or the apoA-I DR-2–type PPREs. FF fully induced PPAR transactivation, whereas GF only elicited a partial transactivation on the hapoA-I DR-2 PPRE. Both GF and FF exhibited similar affinities for PPAR. By contrast, GF and FF were equally active on a DR-1–type PPRE. Detailed analysis has revealed that the binding of PPAR to the hapoA-I promoter occurs through a DR-2–type PPRE within the A site,34 whereas the ACO site is a DR-1 response element.25 Our data show that GF behaves as a partial or full PPAR agonist depending on the type of PPRE, whereas FF fully induces PPAR transactivation activity on both DR-1 and DR-2 sites.
To understand the role of the PPRE structure in the distinct action of FF and GF, the ability of both ligands to induce coactivator recruitment to PPAR on DR-1 versus DR-2 PPREs was compared. Our data indicate that GF is less efficient to induce DRIP205 recruitment to the PPAR/RXR heterodimer bound to the DR-2. By contrast, FF elicited similar recruitment efficiency to both types of complexes. This indicates that GF induces differential coactivator recruitment to the PPAR/RXR-DNA complex depending on the geometry of the PPRE. This is in line with previous reports showing that the nature of the response element plays a crucial role by inducing allosteric conformational changes of the receptor, leading to distinct coactivator recruitment, receptor activity, and gene response.35 Whether DRIP205 is the coactivator explaining the different in vivo responses is impossible to answer. However, it can be concluded from our data that weaker coactivator recruitment to PPAR by GF, because of distinct PPRE architecture, is likely to be the basis of the selective modulation of gene expression by GF in vivo.
Another important observation of this study is that PPAR expression levels determine the response to fibrates. When treated with FF, PPAR+/–hA-ITg mice showed an intermediate response with respect to plasma TC, HDL-C, and hapoA-I plasma and liver mRNA levels. Thus, our data provide genetic evidence that PPAR expression levels determine the response to fibrates. Because we previously reported that human liver PPAR levels vary >2-fold among individuals,36 it is likely that in humans also the response to fibrate treatment will differ between individuals because of differences in PPAR expression levels. In conclusion, this article strongly supports the fact that GF behaves as a selective PPAR modulator in humans. It is interesting to note that GF is chemically different from other fibrates.37 Although differences in pharmacokinetics and metabolism between fibrates may also contribute to distinct clinical phenotypes and even though GF may also have additional effects,38 our data provide a plausible mechanism for the clinically observed differences between FF and GF. Our observations may have potential significant implications with regard to pharmacological effects of fibrates, and future studies may be aimed at identifying selective PPAR modulators to improve therapeutic efficacy.
Acknowledgments
This multicenter clinical study was funded by an unrestricted grant from Laboratoires FournierSA. This work was supported by grants from INSERM, the Région Nord/Pas-de-Calais (Génop?le #01360124), and the Leducq Foundation. I.P.T. was supported by a European Community grant (ERBFMBICT983214). All lipid and safety tests were centrally measured in the Department of Laboratory Medicine (Hamilton Civic Hospitals, Hamilton, Ontario, Canada) under the supervision of Dr Darius Nazir. We thank J.-C. Ansquer, S. Robins, and J.-P. Després for helpful discussion and J. Fremaux, B. Derudas, A. Pilon, M. Bouly, M. Coevoet, and H. DeSafta for excellent technical assistance. We greatly acknowledge the clinical study investigators: Dr Gordon A. Francis (Department of Medicine, Division of Cardiology, University of Alberta, Edmonton, Alberta); Dr Alan L. Edward and Dr Charlotte A. Jones (Department of Medicine, Calgary General Hospital, Calgary, Alberta); Dr Gregory P. Curnew (Lipid Research Clinic, Hamilton Civic Hospitals, Hamilton, Ontario); Dr Lawrence A. Leiter (Department of Endocrinology, St Michael’s Hospital, Toronto, Ontario); Dr Teik Chye Ooi (Department of Medicine, Division of Endocrinology and Metabolism, Ottawa Hospital Riverside Campus, Ottawa, Ontario); Dr Mark H. Sherman (Department of Medicine, Division of Endocrinology and Metabolism, Royal Victoria Hospital, Montréal, Québec); Dr Bruno St-Pierre and Dr Chantal Godin (Department of Medicine, Centre Hospitalier Hotel-Dieu de Sherbrooke, Québec); Dr Patrick T.S. Ma and Dr Neil George Filipchuk (Department of Cardiology, Calgary General Hospital, Calgary, Alberta); and Dr Jorge Federico Bonet and Dr Gordon Neil Hoag (Lipid clinic, Victoria General Hospital, Victoria, British Columbia).
References
Robins SJ. PPAR ligands and clinical trials: cardiovascular risk reduction with fibrates. J Cardiovasc Risk. 2001; 8: 195–201.
Rubin EM, Krauss RM, Spangler EA, Verstuyft JG, Clift SM. Inhibition of early atherogenesis in transgenic mice by human apolipoprotein AI. Nature. 1991; 353: 265–267.
Nissen SE, Tsunoda T, Tuzcu EM, Schoenhagen P, Cooper CJ, Yasin M, Eaton GM, Lauer MA, Sheldon WS, Grines CL, Halpern S, Crowe T, Blankenship JC, Kerensky R. Effect of recombinant ApoA-I Milano on coronary atherosclerosis in patients with acute coronary syndromes: a randomized controlled trial. JAMA. 2003; 290: 2292–2300.
Fruchart JC, Davignon J, Bard JM, Grothe AM, Richard A, Fievet C. Effect of fenofibrate treatment on type III hyperlipoproteinemia. Am J Med. 1987; 83: 71–74.
Mellies MJ, Stein EA, Khoury P, Lamkin G, Glueck CJ. Effects of fenofibrate on lipids, lipoproteins, and apolipoproteins in 33 subjects with primary hypercholesterolemia. Atherosclerosis. 1987; 63: 57–64.
Goldberg AC, Schonfeld G, Feldman EB, Ginsberg HN, Hunninghake DB, Insull W Jr, Knopp RH, Kwiterovich PO, Mellies MJ, Pickering J. Fenofibrate for the treatment of type IV and V hyperlipoproteinemias: a double-blind, placebo-controlled multicenter US study. Clin Ther. 1989; 11: 69–83.
Steinmetz A, Schwartz T, Hehnke U, Kaffarnik H. Multicenter comparison of micronized fenofibrate and simvastatin in patients with primary type IIA or IIB hyperlipoproteinemia. J Cardiovasc Pharmacol. 1996; 27: 563–570.
Poulter N. The impact of micronized fenofibrate on lipid subfractions and on reaching HDL-target levels in 7,098 patients with dyslipidemia. Br J Cardiol. 1999; 6: 682–685.
Robins SJ, Collins D, Wittes JT, Papademetriou V, Deedwania PC, Schaefer EJ, McNamara JR, Kashyap ML, Hershman JM, Wexler LF, Rubins HB. Relation of gemfibrozil treatment and lipid levels with major coronary events: VA-HIT: a randomized controlled trial. JAMA. 2001; 285: 1585–1591.
Durrington PN, Mackness MI, Bhatnagar D, Julier K, Prais H, Arrol S, Morgan J, Wood GN. Effects of two different fibric acid derivatives on lipoproteins, cholesteryl ester transfer, fibrinogen, plasminogen activator inhibitor and paraoxonase activity in type IIb hyperlipoproteinaemia. Atherosclerosis. 1998; 138: 217–225.
Schaefer EJ, Lamon-Fava S, Cole T, Sprecher DL, Cilla DD Jr, Balagtas CC, Rowan JP, Black DM. Effects of regular and extended-release gemfibrozil on plasma lipoproteins and apolipoproteins in hypercholesterolemic patients with decreased HDL cholesterol levels. Atherosclerosis. 1996; 127: 113–122.
Camp HS, Li O, Wise SC, Hong YH, Frankowski CL, Shen X, Vanbogelen R, Leff T. Differential activation of peroxisome proliferator-activated receptor-gamma by troglitazone and rosiglitazone. Diabetes. 2000; 49: 539–547.
Rubin EM, Ishida BY, Clift SM, Krauss RM. Expression of human apolipoprotein A-I in transgenic mice results in reduced plasma levels of murine apolipoprotein A-I and the appearance of two new high density lipoprotein size subclasses. Proc Natl Acad Sci U S A. 1991; 88: 434–438.
Lee SS, Pineau T, Drago J, Lee EJ, Owens JW, Kroetz DL, Fernandez-Salguero PM, Westphal H, Gonzalez FJ. Targeted disruption of the alpha isoform of the peroxisome proliferator-activated receptor gene in mice results in abolishment of the pleiotropic effects of peroxisome proliferators. Mol Cell Biol. 1995; 15: 3012–3022.
Akiyama TE, Nicol CJ, Fievet C, Staels B, Ward JM, Auwerx J, Lee SS, Gonzalez FJ, Peters JM. Peroxisome proliferator-activated receptor- regulates lipid homeostasis, but is not associated with obesity: studies with congenic mouse lines. J Biol Chem. 2001; 276: 39088–39093.
Staels B, van Tol A, Andreu T, Auwerx J. Fibrates influence the expression of genes involved in lipoprotein metabolism in a tissue-selective manner in the rat. Arterioscler Thromb. 1992; 12: 286–294.
Berthou L, Duverger N, Emmanuel F, Langouet S, Auwerx J, Guillouzo A, Fruchart JC, Rubin E, Denefle P, Staels B, Branellec D. Opposite regulation of human versus mouse apolipoprotein A-I by fibrates in human apolipoprotein A-I transgenic mice. J Clin Invest. 1996; 97: 2408–2416.
Bouly M, Masson D, Gross B, Jiang XC, Fievet C, Castro G, Tall AR, Fruchart JC, Staels B, Lagrost L, Luc G. Induction of the phospholipid transfer protein gene accounts for the high density lipoprotein enlargement in mice treated with fenofibrate. J Biol Chem. 2001; 276: 25841–25847.
Staels B, Auwerx J. Perturbation of developmental gene expression in rat liver by fibric acid derivatives: lipoprotein lipase and alpha-fetoprotein as models. Development. 1992; 115: 1035–1043.
Issemann I, Green S. Activation of a member of the steroid hormone receptor superfamily by peroxisome proliferators. Nature. 1990; 347: 645–650.
Vu-Dac N, Chopin-Delannoy S, Gervois P, Bonnelye E, Martin G, Fruchart JC, Laudet V, Staels B. The nuclear receptors peroxisome proliferator-activated receptor and Rev-erbalpha mediate the species-specific regulation of apolipoprotein A-I expression by fibrates. J Biol Chem. 1998; 273: 25713–25720.
Raspe E, Madsen L, Lefebvre AM, Leitersdorf I, Gelman L, Peinado-Onsurbe J, Dallongeville J, Fruchart JC, Berge R, Staels B. Modulation of rat liver apolipoprotein gene expression and serum lipid levels by tetradecylthioacetic acid (TTA) via PPAR activation. J Lipid Res. 1999; 40: 2099–2110.
Lefebvre B, Mouchon A, Formstecher P, Lefebvre P. Distinct modes of interaction of the retinoic acid receptor alpha with natural and synthetic retinoids. Mol Cell Endocrinol. 1998; 139: 161–169.
Mouchon A, Delmotte MH, Formstecher P, Lefebvre P. Allosteric regulation of the discriminative responsiveness of retinoic acid receptor to natural and synthetic ligands by retinoid X receptor and DNA. Mol Cell Biol. 1999; 19: 3073–3085.
Schoonjans K, Staels B, Auwerx J. The peroxisome proliferator activated receptors(PPARS) and their effects on lipid metabolism and adipocyte differentiation. Biochim Biophys Acta. 1996; 1302: 93–109.
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 hyperlipidemic patients: an open Belgian multicenter study. Atherosclerosis. 1994; 110: S49–S54.
Sakai T, Kamanna VS, Kashyap ML. Niacin, but not gemfibrozil, selectively increases LP-AI, a cardioprotective subfraction of HDL, in patients with low HDL cholesterol. Arterioscler Thromb Vasc Biol. 2001; 21: 1783–1789.
Watts GF, Barrett PH, Ji J, Serone AP, Chan DC, Croft KD, Loehrer F, Johnson AG. Differential regulation of lipoprotein kinetics by atorvastatin and fenofibrate in subjects with the metabolic syndrome. Diabetes. 2003; 52: 803–811.
Bilz S, Wagner S, Schmitz M, Bedynek A, Keller U, Demant T. Effects of atorvastatin versus fenofibrate on apoB-100 and apoA-I kinetics in mixed hyperlipidemia. J Lipid Res. 2004; 45: 174–185.
Bard JM, Parra HJ, Luc G, Camare R, Ziegler O, Dachet C, Bruckert E, Douste-Blazy P, Drouin P, Jacotot B. Lipoprotein particle analysis comparing simvastatin and fenofibrate. Atherosclerosis. 1991; 9: S29–S34.
Kahri J, Sane T, van Tol A, Taskinen MR. Effect of gemfibrozil on the regulation of HDL subfractions in hypertriglyceridaemic patients. J Intern Med. 1995; 238: 429–436.
Mardones P, Pilon A, Bouly M, Duran D, Nishimoto T, Arai H, Kozarsky KF, Altayo M, Miquel JF, Luc G, Clavey V, Staels B, Rigotti A. Fibrates down-regulate hepatic scavenger receptor class B type I protein expression in mice. J Biol Chem. 2003; 278: 7884–7890.
Dreyer C, Keller H, Mahfoudi A, Laudet V, Krey G, Wahli W. Positive regulation of the peroxisomal beta-oxidation pathway by fatty acids through activation of peroxisome proliferator-activated receptors (PPAR). Biol Cell. 1993; 77: 67–76.
Klausing KS, Briggs, M. The human apolipoprotein A1 promoter contains a novel PPRE. Circulation. 1995; 92: (suppl I): 295. Abstract.
Hall JM, McDonnell DP, Korach KS. Allosteric regulation of estrogen receptor structure, function, and coactivator recruitment by different estrogen response elements. Mol Endocrinol. 2002; 16: 469–486.
Gervois P, Torra IP, Chinetti G, Grotzinger T, Dubois G, Fruchart JC, Fruchart-Najib J, Leitersdorf E, Staels B. A truncated human peroxisome proliferator-activated receptor alpha splice variant with dominant negative activity. Mol Endocrinol. 1999; 13: 1535–1549.
Todd PA, Ward A. Gemfibrozil. A review of its pharmacodynamic and pharmacokinetic properties, and therapeutic use in dyslipidaemia. Drugs. 1988; 36: 314–339.
Jin FY, Kamanna VS, Chuang MY, Morgan K, Kashyap ML. Gemfibrozil stimulates apolipoprotein A-I synthesis and secretion by stabilization of mRNA transcripts in human hepatoblastoma cell line(HepG2). Arterioscler Thromb Vasc Biol. 1996; 16: 1052–1062.(Hélène Duez; Bruno Lefebv)