Activation of Peroxisome Proliferator-activated Receptor α Increases the Expression and Activity of Microsomal Triglyceride Transfer Protein in the Liver*

Microsomal triglyceride transfer protein (MTP) is rate-limiting in the assembly and secretion of lipoproteins containing apolipoprotein (apo) B. Previously we demonstrated that Wy 14,643 (Wy), a peroxisome proliferator-activated receptor (PPAR) α agonist, increases apoB-100 secretion despite decreased triglyceride synthesis. In this study, we sought to determine whether PPARα activation increases MTP expression and activity. Treatment with Wy increased hepatic MTP expression and activity in rats and mice and increased MTP expression in primary cultures of rat and mouse hepatocytes. Addition of actinomycin D blocked this increase and the MTP promoter (–136 to +67) containing a conserved DR1 element was activated by Wy, showing that PPARα activates transcription of the gene. Wy did not affect MTP expression in the intestine or in cultured hepatocytes from PPARα-null mice. A retinoid X receptor agonist (9-cis-retinoic acid), but not a PPARγ agonist (rosiglitazone), increased MTP mRNA expression in cultured hepatocytes from both wild type and PPARα-null mice. In rat hepatocytes incubated with Wy, MTP mRNA levels increased between 6 and 24 h, and MTP protein expression and apoB-100 secretion increased between 24 and 72 h. In conclusion, PPARα activation stimulates hepatic MTP expression via increased transcription of the Mtp gene. This effect is paralleled by a change in apoB-100 secretion, indicating that the effect of Wy on apoB-100 secretion is mediated by increased expression of MTP.

The peroxisome proliferator-activated receptor (PPAR) 1 ␣ is a nuclear receptor that controls the transcription of genes involved in several lipid metabolism pathways, such as ␤-oxidation and fatty acid uptake and transport, as well as lipoprotein production and clearance (1)(2)(3). PPAR␣ is expressed in tissues with a high degree of fatty acid catabolism, primarily liver, intestine, and skeletal muscle (1,2,4). PPAR␣ heterodimerizes with the retinoid X receptor ␣ (RXR␣), and this complex binds DR1 sequences that constitute PPAR response elements (for reviews, see Refs. 1,2). The endogenous ligands for PPAR␣ are unsaturated fatty acids and eicosanoids, whereas hypolipidemic fibrates such as Wy 14,643 (Wy) are potent synthetic agonists (5). We recently found that PPAR␣ agonists increased apolipoprotein (apo) B-100 secretion 2-fold but did not change apoB-48 secretion (6). The increase occurred despite decreased triglyceride synthesis and unchanged apoB mRNA editing and could be explained by inhibition of the cotranslational degradation of apoB-100.
Microsomal triglyceride transfer protein (MTP) catalyzes the transfer of neutral lipids to apoB and thus has a pivotal role in the assembly of apoB-containing lipoproteins (for reviews, see Refs. 7,8). The 97-kDa MTP protein that confers lipid transfer activity heterodimerizes with protein disulfide isomerase. Mutations in the Mtp gene cause abetalipoproteinemia, which is characterized by the inability to secrete apoB-containing lipoproteins (7,8). MTP influences apoB secretion through its effects on the presecretory degradation of the protein (7, 9 -12), most likely by increasing the cotranslational lipidation of apoB during the first step of VLDL assembly (13). MTP is probably also important for the bulk transfer of lipids during the second step of VLDL assembly (14). Although alterations in MTP activity seem to influence the production of both apoB-48-and apoB-100-containing lipoproteins, inhibition or lack of MTP expression appears to have a greater effect on apoB-100 production (9, 14 -17). Moreover, because the level of MTP expression determines the rate of apoB-100 secretion (10,(17)(18)(19), alterations in MTP levels could change apoB-100 secretion by interfering with its presecretory degradation in the cell.
In this study, we have demonstrated that the PPAR␣ agonist Wy stimulates the expression of MTP via transcriptional activation of the Mtp gene and, in parallel with increased MTP protein expression, increases the secretion of apoB-100. These findings indicated that the effect of PPAR␣ activation on MTP expression and activity can explain the effect of Wy on the secretion of apoB-100.

EXPERIMENTAL PROCEDURES
Animals and Treatment-Homozygous PPAR␣-null mice on pure Sv/129 genetic background and corresponding wild type Sv/129 control mice were kindly provided by Dr. F. J. Gonzalez (National Institutes of Health, Bethesda, MD) and kept on the Sv/129 background (20). C57BL/6 mice and Sprague-Dawley rats were from Harlan (Horst, The Netherlands). The animals were maintained under standardized conditions of temperature (21-22°C) and humidity (40 -60%), with light from 0600 to 1800 h. The animals had free access to water and standard laboratory chow containing (energy %) 12% fat, 62% carbohydrates, and 26% protein, with a total energy content of 12.6 kJ/g (R3; Lactamin AB, Stockholm, Sweden).
Three-month-old male C57BL/6 mice were treated with Wy (30 mol/ kg/day in 0.5% (w/v) methyl cellulose) by gavage once daily for 2 weeks. Age-matched control mice received only vehicle. The mice were anesthetized with isoflurane (Forene; Abbot Scandinavia AB, Sweden), and blood was collected by cardiac puncture. The liver and the first 10 cm of the small intestine were removed, immediately frozen in liquid nitrogen, and stored at Ϫ70°C.
Three-month-old male Sprague-Dawley rats were fed a high fat diet for 5 weeks. During the last 4 weeks, Wy (30 mol/kg/day) or vehicle was administered by gavage once daily. The energy percentage of the high fat diet was 48% fat (mainly saturated), 15% protein, and 37% carbohydrates, with a total energy content of 21.4 kJ/g. Blood and liver were harvested as described above.
The study protocol was approved by the Ethics Committee of Göteborg University. All experiments were conducted in accordance with accepted standards of humane animal care.
Serum Analyses-Plasma triglyceride and cholesterol concentrations were determined with enzymatic colorimetric assays (Roche Applied Science). Serum apoB concentrations were determined with an electroimmunoassay (3,21).
Primary Hepatocyte Cultures-Mouse hepatocytes were obtained by nonrecirculating collagenase perfusion through the portal vein of male C57BL/6 mice or age-and sex-matched male PPAR␣-null and wild type mice on a pure Sv/129 genetic background as described (3). Rat hepatocytes were obtained by perfusion of female Sprague-Dawley rats weighing 200 -300 g (22,23). In brief, the cells were seeded at 120,000 cells/cm 2 in dishes (Falcon, Plymouth, UK) coated with laminin-rich Matrigel (BD Biosciences). The cells were cultured during the first 16 -18 h in Williams' E medium with Glutamax (Invitrogen) supplemented as described (23). The cells were then treated up to 72 h with 1 or 10 M Wy (Chemsyn Science Laboratories, Lenaxa, KS) dissolved in dimethyl sulfoxide (Me 2 SO, 0.15% v/v) in medium supplemented as above plus 1 nM dexamethasone (Sigma) and 3 nM insulin (Actrapid; Novo Nordisk A/S, Denmark). Actinomycin D (5 g/ml) was dissolved in Me 2 SO (0.15% v/v) and added to primary rat hepatocyte cultures during the last 6 h of incubation (23). Primary mouse hepatocytes were also incubated with 1 or 10 M rosiglitazone (Medicinal Chemistry; Astra-Zeneca R&D, Mölndal, Sweden) or 10 M 9-cis-retinoic acid (cRA) (Sigma), each dissolved in dimethyl sulfoxide (0.15% v/v).
MTP Promoter Reporter Gene Assay and McA-RH7777 Cell Culture-A fragment of the mouse MTP promoter (Ϫ136 to ϩ67 bp in reference to human transcription start site (24)) containing a 5Ј-KpnI site and a 3Ј-BglII site was amplified by PCR from mouse liver cDNA using the following primers, forward 5Ј-ACG GTA CCA CTA CAA ACT ATA GCC CAC CTG-3Ј and reverse 5Ј-GCA GAT CTG CTG GCT CCC TCT GCC ACA TCC-3Ј,and TA-cloned into pCR 2.1-TOPO (Invitrogen). A specific mutation in the DR1 sequence (Ϫ52 to Ϫ40 bp in reference to the human transcription start site (24)) of the MTP promoter was generated using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) and the following primers, forward 5Ј-GGA GTT TGG AAT CTG TGC TTT CCC CTA TAG-3Ј and reverse 5Ј-CTA TAG GGG AAA GCA CAG ATT CCA AAC TCC-3Ј (mutated bases in bold), according to Kang et al. (24). Constructs were verified by DNA sequencing and cloned into KpnI and BglII sites of the pGL3basic luciferase reporter vector.
The McA-RH7777 cells were plated in 6-well plates 1 day prior to transfection and were ϳ80% confluent when transfected. Wild type or mutated MTP promoter luciferase reporter gene vectors were transfected using Lipofectamine 2000 (Invitrogen). The day after transfec-tion, cells were trypsinated and replated on 96-well plates (Isoplate TC; Wallac), 50,000 cells/well in 100 l with the indicated concentrations of Wy or Me 2 SO vehicle control. Final Me 2 SO concentration was 0.5% (v/v). Luciferase activity was measured 24 h after the addition of ligand using a Victor 2 luminometer (Wallac) and the Steady-Glo luciferase assay system (Promega).
Estimation of ApoB Secretion-The secretion of apoB-48 and apoB-100 into the medium of primary rat hepatocyte cultures was estimated by labeling the cells with a [ 35 S]methionine-cysteine mix (Amersham Biosciences) for 2 h followed by a 4-h chase in culture medium supplemented with 10 mM methionine as described (6,22,25). Labeled apoB-48 and apoB-100 were isolated by immunoprecipitation and SDS-PAGE (22,25). The bands corresponding to apoB-48 and apoB-100 were cut from the gel, digested, and counted in a ␤-counter (22,25).
cDNA Synthesis and Real-time PCR-Total RNA from frozen liver and cultured hepatocytes was isolated with TriReagent according to the manufacturer's protocol (Sigma). The RNA concentration was determined spectrophotometrically at 260 nm. DNA-free (Ambion, Austin, TX) was used to remove DNA from the RNA preparations. First-strand cDNA was synthesized from 0.4 g of total RNA with TaqMan reverse transcription reagents (Applied Biosystems, Foster City, CA). Specific primers for each gene (Table I) were designed with Primer Express software (Applied Biosystems) and gene sequences from GenBank TM . To avoid amplification of genomic DNA, the primers were positioned to span exon junctions. All primers were synthesized by Applied Biosystems.
Real-time PCR analysis was performed with the ABI Prism 7700 sequence detection system using the SYBR Green labeling system (Applied Biosystems). The reactions were analyzed in triplicate, and the correct sizes of the amplicons were verified by gel electrophoresis. The expression data were normalized to an endogenous control, acidic ribosomal phosphoprotein P0 (36B4). The level of 36B4 was invariable between the groups in this study. The relative expression levels were calculated according to the formula 2 Ϫ⌬Ct , where ⌬Ct is the difference in threshold cycle (Ct) values between the target and the 36B4 endogenous control.
Western Blot-Total protein extracts from frozen livers and cultured hepatocytes were prepared as described (26). For cultured primary hepatocytes, Matrigel was removed by incubation on ice for 60 min with 5 mM EDTA in phosphate-buffered saline (PBS) and washings in PBS. Western blots were performed with a polyclonal anti-MTP-protein disulfide isomerase antibody (27) (kindly provided by Carol Shoulders) as described (26).
MTP Activity-MTP activity was measured with an MTP assay kit (Roar Biomedical, New York, NY) according to the manufacturer's protocol and as described (28). In brief, the tissue was homogenized in ice-cold assay buffer (10 mM Tris, 150 mM NaCl, 2 mM EDTA) supplemented with protease inhibitors (Complete Mini; Roche Diagnostics). Donor vesicles (5 l) containing fluorescent neutral lipid and acceptor vesicles (5 l) were incubated with 10 l of homogenate (containing about 1.5 g/l protein) in a total volume of 200 l of assay buffer at 37°C for 210 -280 min. Transfer activity was stable between 210 and 280 min. The activity was linear with respect to the amount of protein.
To determine the mass of MTP-mediated lipid transfer, the intensity of the fluorescent assay buffer was measured in a fluorescence spectrophotometer (Spectramax Gemini; Molecular Devices, Sunnyvale, CA) at an excitation wavelength of 465 nm and an emission wavelength of 535 nm. Lipid transfer activity was related to protein content by using bovine serum albumin as standard (BCA protein assay reagent kit; Pierce) and expressed as pmol/min/mg protein.
Statistics-Values are expressed as means Ϯ S.E. Comparisons between groups were made by Kruskal-Wallis test and Mann-Whitney U test. p Ͻ0.05 was considered significant.

Effects of Wy on MTP Expression and Activity in Mice and
Rats-In male C57BL/6 mice, Wy increased MTP mRNA ex- pression in the liver, with corresponding increases in hepatic MTP protein and activity levels (Fig. 1). There was no effect on MTP mRNA levels in the intestine (Fig. 1D), indicating a liver-specific effect. Wy also decreased plasma triglyceride and apoB levels, despite the increased hepatic MTP expression and activity, but had no significant effect on cholesterol levels (Table II). Similar effects on MTP mRNA, protein, and activity were seen in Wy-treated rats (Fig. 2). However, in rats, protein disulfide isomerase expression decreased (Fig. 2B), an effect not seen in the mice. Effects of Wy, Rosiglitazone, and 9-cis-Retinoic Acid on MTP Expression in Mouse Hepatocytes-Incubation of primary mouse hepatocytes with Wy for 72 h increased MTP mRNA expression in dose-dependent fashion. However, the PPAR␥ agonist rosiglitazone had no effect on MTP mRNA levels (Fig.  3), even though the cells expressed PPAR␥ mRNA and responded to 10 M rosiglitazone with a 25-fold increase in aP2 mRNA (data not shown). Thus, MTP mRNA expression responds to PPAR␣ activation, but not to PPAR␥ activation.
⌻o exclude the possibility that the effect of Wy was nonspecific, primary hepatocytes from PPAR␣-null and wild type mice were incubated for 3 days with Wy. This treatment had no effect on PPAR␣-null hepatocytes but increased MTP mRNA levels 3-fold in control hepatocytes (Fig. 4). Because PPARs need to heterodimerize with RXR␣ to activate target genes, we also investigated the effect of the RXR ligand cRA on MTP mRNA expression in vitro. Incubation with cRA for 3 days increased MTP mRNA levels to the same degree in hepatocytes from both wild type and PPAR␣-null mice (Fig. 4), indicating that the effect of cRA on MTP mRNA expression is independent of PPAR␣.
Effects of Wy on Transcription of the MTP Gene-Primary rat hepatocytes were incubated with Wy for 24 h, and the transcriptional inhibitor actinomycin D was added during the last 6 h (Fig. 5). Addition of actinomycin D abolished the increase in MTP expression induced by Wy, whereas it had no effect on control cells (Fig. 5). This finding indicated that Wy increases MTP mRNA expression via activation of transcription. We thereafter investigated whether the conserved DR-1 element reported by Kang et al. (24) is activated by Wy in McA-RH7777 cells transfected with either wild type or mutated MTP promoter coupled to the luciferase gene (Fig. 6). We used McA-RH7777 cells because they respond to Wy (10 M) with an increased expression of MTP mRNA (ϩ56%, data not shown). Wild type promoter activity increased dose-dependently as a result of Wy incubation, whereas the cells transfected with the promoter construct with a mutated DR-1 element were unre-

FIG. 2. Effects of Wy on hepatic MTP mRNA expression (A), MTP protein expression (B), and MTP activity (C) in 3-month-old male rats.
Rats were treated with Wy (30 mol/kg/day) or vehicle for 4 weeks. MTP mRNA levels were measured by real-time PCR and MTP protein levels by Western blot. MTP activity was determined with an MTP activity assay. Values are the means Ϯ S.E. of four observations. *, p Ͻ 0.05, Mann-Whitney U test. PDI, protein disulfide isomerase. sponsive to Wy (Fig. 6A). Fig. 6B shows the results of three independent experiments. Wild type promoter activity increased about 2-fold after incubation with 10 M Wy, whereas the mutated promoter was unresponsive. Moreover, the cells transfected with the wild type construct had a higher basal activity than the cells transfected with the mutated promoter construct, indicating an effect of endogenous ligands (Fig. 6B).

Effect of Wy on MTP Expression and ApoB Secretion in Rat
Hepatocytes-Next, we measured MTP expression and secretion of apoB-100 and apoB-48 in primary rat hepatocytes incubated with Wy for 6, 24, and 72 h (Fig. 7). Wy had no effect on apoB-48 secretion. However, MTP mRNA expression increased 2.4-fold between 6 and 24 h, and MTP protein expression and apoB-100 secretion increased 2.8-and 1.8-fold, respectively, between 24 and 72 h. The similar time course of these effects suggests that the increased secretion of apoB-100 is mediated by a Wy-induced increase in MTP protein expression. DISCUSSION This study showed that treatment of mice and rats with the PPAR␣ agonist Wy increased MTP mRNA and protein expression as well as MTP activity in the liver but had no effect on MTP mRNA in the intestine. Wy increased MTP mRNA expression in primary mouse and rat hepatocytes in vitro via increased transcription of the Mtp gene. The transcriptional effect of PPAR␣ activation involved the DR-1 element that has been shown to be activated by RXR␣ (24). Thus, PPAR␣ regulates not only the cotranslational degradation and rate of secretion of apoB-100 but also the expression and activity of MTP. This study also showed that the PPAR␣ agonist-induced stimulation of apoB-100 secretion was preceded by an increase in MTP mRNA and paralleled by a change in MTP protein levels. Because variations in MTP levels are reflected in the cotranslational degradation and secretion of apoB-100 (10,14,17,19), our results indicated that the increase in MTP levels, induced by the PPAR␣ agonist, mediates the effects of this agonist on the secretion of apoB-100. Consistent with our previous results (6), Wy had no effect on apoB-48 secretion.
Despite the increases in MTP expression and apoB-100 secretion, plasma apoB levels decreased in Wy-treated mice. This finding could reflect a decreased triglyceride synthesis, resulting in a shift in secretion from triglyceride-rich VLDL particles to triglyceride-poor apoB-containing lipoproteins (6). It is well known that smaller triglyceride-poor VLDL particles give rise to apoB-containing low density lipoprotein (LDL) particles that turn over more rapidly compared with LDL particles derived from larger triglyceride-rich particles (29). Moreover, PPAR␣ agonists decrease the hepatic expression of apoC-III (30), which can contribute to increased clearance of apoB-containing lipoproteins.
We have previously shown that PPAR␣-deficient hepatocytes have a higher secretion of triglycerides, apoB-48, and apoB-100 than hepatocytes from wild type controls (3). Clearly, that is a metabolic situation that is opposite to what we have observed after Wy incubation of rat hepatocytes where the secretion of apoB-100, but not apoB-48, is increased in the context of decreased triglyceride biosynthesis and secretion (6). We therefore believe that the effects of PPAR␣ deficiency on apoB secretion are primarily because of an increased availability of lipids for VLDL assembly. Moreover, MTP expression has been shown to be unchanged in male PPAR␣-deficient mice (31), showing that changed MTP expression is probably not important for the changed apoB secretion in PPAR␣ deficiency.
In a recent report (32), treatment of low density lipoprotein receptor-deficient mice with ciprofibrate caused a large accumulation of triglyceride-depleted apoB-100 containing lipoproteins associated with an increase in the hepatic secretion of apoB-100 VLDL. Decreased apoB mRNA editing in the liver was suggested to be the cause of these effects of ciprofibrate (32). In this study, we have shown that MTP expression and activity increase by PPAR␣ activation, suggesting another explanation for the increased apoB-100 secretion. Moreover, our previous study (6) shows that PPAR␣ activation can increase apoB-100 secretion without influencing apoB mRNA editing.
We cannot explain why MTP mRNA levels were unaffected by Wy in intestine. Others have also found that Mtp gene expression is differently regulated in the liver and intestine (33,34). We transfected human embryonic kidney (HEK) 293 cells with the MTP promoter construct (Ϫ136 to ϩ67 bp), but they were unresponsive to Wy although they were co-transfected with a PPAR␣ expression vector (data not shown), indicating that the cellular milieu is crucial for the effect of PPAR␣ agonists on the MTP promoter. It has been shown that HEK 293 cells have undetectable activity of the proximal MTP promoter unless an hepatocyte nuclear factor (HNF)-4 expression plasmid were introduced (35), suggesting that expression of ApoB-100 and apoB-48 were isolated from the medium by immunoprecipitation and SDS-PAGE. The apoB values are means Ϯ S.E. of three independent liver perfusions with 3-4 culture dishes in each group. The MTP mRNA and protein levels are representative data from one experiment that was repeated once. *, p Ͻ 0.05, Kruskal-Wallis test followed by comparisons between the control group and the other groups using the Mann-Whitney U test. PDI, protein disulfide isomerase. HNF-4 is important for the effect of other transcription factors on the MTP promoter. Because the expression of HNF-4 is similar in the liver and intestine (36), it is likely that other mechanisms are responsible for the different effects of Wy in intestine and liver.
Recently, RXR␣ and its ligand, cRA, were shown to increase MTP promoter activity in L35 hepatoma cells by stimulating a conserved DR-1 element in the proximal promoter (24). We have now shown that the same DR-1 element in the MTP promoter could be activated by Wy in McA-RH7777 cells, a rat hepatoma cell line. The response was specific because the construct with a mutated DR-1 element did not respond to Wy. We extended the findings of Kang et al. (24) by demonstrating increased MTP mRNA expression after incubation of primary mouse hepatocytes with cRA, indicating the physiologic importance of RXR␣ in the regulation of MTP mRNA levels. Incubation with cRA also increased the expression of MTP mRNA in the absence of PPAR␣. This finding suggested that the effect of cRA is independent of PPAR␣. Therefore, it is likely that a nuclear receptor complex other than PPAR␣⅐RXR␣ is responsible for the effect of cRA on MTP mRNA expression. Recently, it was shown that RXR homodimers selectively bind to functional PPAR response elements and induce transactivation without the presence of PPAR␣ (37). Our findings together with those of Kang et al. (24) suggested that RXR homodimers can activate the PPAR response element in the MTP promoter because the presence of PPAR␣ was not necessary for an effect of cRA.
Activation of PPAR␥ by rosiglitazone reduces MTP protein mass in fructose-fed hamsters, a model of insulin resistance associated with increased VLDL-apoB secretion (38). In our study, rosiglitazone had no direct effect on MTP mRNA expression in primary mouse hepatocytes, indicating that the effect of PPAR␥ activation in fructose-fed hamsters occurs indirectly, presumably through changed insulin sensitivity.
In conclusion, PPAR␣ activation increases hepatic MTP expression through transcriptional activation of the Mtp gene. In addition, RXR␣ had a PPAR␣-independent effect on MTP expression. The PPAR␣ agonist-induced increase in the secretion of apoB-100 was preceded by an increase in MTP mRNA and paralleled by a change in MTP protein levels. This finding indicated that the influence of PPAR␣ agonists on the secretion of apoB-100 is mediated by increased expression of MTP.