Trans geometric isomers of EPA decrease LXRα-induced cellular triacylglycerol via suppression of SREBP-1c and PGC-1β

Dietary mono- or di-trans fatty acids with chain lengths of 18–22 increase the risk of cardiovascular diseases because they increase LDL cholesterol and decrease HDL cholesterol in the plasma. However, the effects of trans isomers of PUFAs on lipid metabolism remain unknown. Dietary PUFAs, especially eicosapentaenoic acid (EPA) in marine oils, improve serum lipid profiles by suppressing liver X receptor α (LXRα) activity in the liver. In this study, we compared the effects of trans geometric isomers of eicosapentaenoic acid (TEPA) on triacylglycerol synthesis induced by a synthetic LXRα agonist (T0901317) with the effects of EPA in HepG2 cells. TEPA significantly decreased the amount of cellular triacylglycerol and the expression of mRNAs encoding fatty acid synthase, stearoyl-CoA desaturase-1, and glycerol-3-phosphate acyltransferase induced by T0901317 compared with EPA. However, there was no significant difference between the suppressive effect of TEPA or EPA on the expression of sterol-regulatory element binding protein-1c (SREBP-1c) induced by T0901317. We found that TEPA, but not EPA, decreased the mRNA expression of peroxisome proliferator-activated receptor γ coactivator 1β (PGC-1β), which is a coactivator of both LXRα and SREBP-1. These results suggest that the hypolipidemic effect of TEPA can be attributed to a decrease not only in SREBP-1 but also in PGC-1β expression.

Abstract Dietary mono-or di-trans fatty acids with chain lengths of 18-22 increase the risk of cardiovascular diseases because they increase LDL cholesterol and decrease HDL cholesterol in the plasma. However, the effects of trans isomers of PUFAs on lipid metabolism remain unknown. Dietary PUFAs, especially eicosapentaenoic acid (EPA) in marine oils, improve serum lipid profiles by suppressing liver X receptor a (LXRa) activity in the liver. In this study, we compared the effects of trans geometric isomers of eicosapentaenoic acid (TEPA) on triacylglycerol synthesis induced by a synthetic LXRa agonist (T0901317) with the effects of EPA in HepG2 cells. TEPA significantly decreased the amount of cellular triacylglycerol and the expression of mRNAs encoding fatty acid synthase, stearoyl-CoA desaturase-1, and glycerol-3-phosphate acyltransferase induced by T0901317 compared with EPA. However, there was no significant difference between the suppressive effect of TEPA or EPA on the expression of sterol-regulatory element binding protein-1c (SREBP-1c) induced by T0901317. We found that TEPA, but not EPA, decreased the mRNA expression of peroxisome proliferator-activated receptor g coactivator 1b (PGC-1b), which is a coactivator of both LXRa and SREBP-1. These results suggest that the hypolipidemic effect of TEPA can be attributed to a decrease not only in SREBP-1 but also in PGC-1b expression.-Zaima, N., T. Sugawara, D. Goto, and T. Hirata. Trans geometric isomers of EPA decrease LXRa-induced cellular triacylglycerol via suppression of SREBP-1c and PGC-1b. J. Lipid Res. 2006. 47: 2712-2717.
Supplementary key words eicosapentaenoic acid & liver X receptor a & sterol-regulatory element binding protein-1c & peroxisome proliferatoractivated receptor g coactivator 1b Most naturally occurring unsaturated fatty acids have only cis double bonds. However, trans fatty acids are found in several foods, including infant formulas, shortenings, vegetable oils and fish oils (1-3), because partial hydrogenation of edible oils converts some cis double bonds to trans double bonds without changing their location.
Dietary trans fatty acids increase the risk of cardiovascular diseases. The intake of mono-or di-trans fatty acids with chain lengths of 18-22, from partially hydrogenated edible oils and fats, significantly increases LDL cholesterol and reduces HDL cholesterol in the plasma (4,5). The trans fatty acid of C18:1 can also stimulate the expression of lipogenic genes in the livers of mice (6). In contrast, there is little information about the effects of trans geometric isomers, without changing the double bond location, of PUFAs on lipid metabolism.
Dietary PUFAs, especially eicosapentaenoic acid (EPA) in marine oils, improve serum lipid profiles (7). The hypolipidemic effect of PUFAs is attributable both to a decrease in lipogenesis and an increase in fatty acid boxidation. It has been well established that liver X receptor a (LXRa) and sterol-regulatory element binding protein-1c (SREBP-1c) play crucial roles in the transcriptional regulation of lipogenic genes in the liver (8,9). Oxysterols, such as 22(R)-hydroxycholesterol, 24(S)-hydroxycholesterol, and 24(S),25-epoxycholesterol, are endogenous LXRa ligands that serve as modulators of LXRa activities. PUFAs compete with LXRa ligands in the activation of the ligand binding domain (LBD) (10,11). LXRa regulates the expression of many lipogenic genes, including SREBP-1c (8), which is synthesized as a 125 kDa precursor protein attached to the endoplasmic reticulum (12). In response to insulin stimulation (13), the membrane-bound precursor is cleaved to a 68 kDa N-terminal fragment that translocates to the nucleus and activates the expression of lipogenic genes such as FAS, stearoyl-coenzyme A desaturase-1 (SCD-1), and glycerol-3-phosphate acyltransferase (GPA). PUFAs suppress SREBP-1c expression by inhibiting LXRa binding to liver X receptor response elements (LXREs), which leads to the decrease in expression of lipogenic genes. A recent study reported that peroxisome proliferator-activated receptor g coactivator 1b (PGC-1b) coactivates the LXRa and SREBP families and increases circulating triacylglycerol and cholesterol in VLDL particles (6). PGC-1b may be a key cofactor linking the dietary intake of trans fatty acids with hyperlipidemia.
Trans geometric isomers of eicosapentaenoic acid (TEPA) have been found as a minor component in vivo (14). Trans isomers of a-linolenic acid are elongated and are desaturated to TEPA in rats fed a diet enriched with those isomers (15). In addition, trans isomerization of EPA may occur in vivo, because trans isomers of arachidonic acid are generated by NO 2 -mediated isomerization (16) and are found in human blood plasma (17). TEPA has different physiological effects on platelet aggregation (18), oxidative stability, and anti-inflammation (19) compared with EPA. However, the effect of TEPA on lipid metabolism remains unknown.
The aim of this study was to clarify the effects of TEPA on lipid metabolism. We compared the effect of TEPA on triacylglycerol synthesis induced by a synthetic LXRa agonist (T0901317) with that of EPA in HepG2 cells. We used a mixture of TEPA prepared from p-toluenesulfinic acid, because there are many possible structures of TEPA.

Preparation of TEPA
The mixture of TEPA was prepared using a chemical catalyst (p-toluenesulfinic acid) and was analyzed by HPLC and GC-MS according to a previously described method (19). The TEPA used in this study consisted of 71% trans isomers, without changing the double bond location, and 29% cis isomers (EPA). The sample contained 0.25% conjugated dienes, with no detectable conjugated trienes, conjugated tetraenes, or conjugated pentaenes.

Lipid extraction and triacylglycerol quantification
HepG2 cells (JCRB 1054; Health Science Research Resources Bank, Osaka, Japan) were plated on six-well plates at 2.0 3 10 5 cells/ml for 24 h in DMEM supplemented with 10% fetal calf serum and antibiotics (penicillin and streptomycin). The cells were then treated with EPA, TEPA, and/or T0901317 (50 nM) in serum-supplemented medium. Fatty acids and T0901317 were dissolved in ethanol. After incubation for 72 h, lipids were extracted from cells with chloroform-methanol (2:1, v/v). Reference control cells were extracted for cellular lipids before incubation (zero time control). Collected supernatants were evaporated gently under an N 2 stream, and triacylglycerol was quantified using a triglyceride E-test kit (Wako Pure Chemical Industries, Osaka, Japan).

Determination of mRNA expression levels by real-time RT-PCR
HepG2 cells were plated on 12-well plates at 2.0 3 10 5 cells/ml in DMEM supplemented with 10% fetal calf serum and antibiotics as detailed above. After 24 h of incubation, each fatty acid was added to HepG2 cells with T0901317 (50 nM) in serum-free medium containing 0.1% BSA. After 24 h of incubation, total RNA was extracted from the cells using TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. RNAs were treated with RNase-free DNase (Invitrogen) to remove contaminating genomic DNA. After inactivating DNase by adding 20 mM EGTA and heating at 658C for 10 min, each RNA was transcribed to cDNA using SuperScript RNase H 2 reverse transcriptase (Invitrogen) with random hexamers at 258C for 10 min and then at 428C for 50 min. The reactions were stopped by incubation at 708C for 15 min, then 0.08 ml of the mixture was added to 4 ml of iQ SYBR Green supermix (Bio-Rad Laboratories, Hercules, CA) and 1.6 ml of gene-specific primers in a final volume of 8 ml. Primers used for the quantification of each gene are listed in Table 1. Primer pairs were selected to yield gene-specific single amplicons based on analyses by melting curves and by agarose gel electrophoresis. Real-time quantitative PCR was performed using a DNA Engine Opticon system (Bio-Rad Laboratories). The thermal cycler parameters were as follows: 3 min at 958C for one cycle, followed by amplification of the cDNA for 40 cycles with melting for 15 s at 958C and with annealing and extension for 30 s at 608C. Values were normalized using 18s rRNA as an endogenous internal standard.

Plasmid constructs used for luciferase assay
The LXRE in the human SREBP-1c promoter was cloned into the pGL3-Promoter vector (Promega, Madison, WI) according to the method of Oberkofler et al. (21). The LXRE construct contained two LXREs in the SREBP-1c promoter (pLXREs-Luc). The cDNA for LXRa (hLXRa) was generated from HepG2 cells by RT-PCR amplification of the entire LXRa coding region (Gen-Bank accession number U22662). A 1,350 bp cDNA fragment was amplified using 59-TGGGTACCAAAGAGATGTCCTTGTGGCT-GG-39 (129 to 150) as the forward primer and 59-TGCTCGAGT-CATTCGTGGACATCCCAG-39 (11,361 to 11,379) as the reverse primer and was ligated into the pcDNA(1) 3.1 vector (Invitrogen). Asp718 and XhoI restriction enzyme sites were introduced into the primer sequences, as shown in boldface. Each construct was verified by sequencing.

Transient transfection and luciferase assay
Before transfection, HepG2 cells and HEK293 cells ( JCRB 9068; Health Science Research Resources Bank) were grown to confluence on 96-well culture plates. Vectors, including pLXREs-Luc (25 ng), hLXRa (10 ng), and an internal Renilla luciferase control vector (40 ng), were cotransfected using Lipofectamine 2000 reagent (2 ml/mg DNA; Invitrogen) in serum-free medium. After overnight transfection, cells were treated with EPA, TEPA, and/or T0901317 for 24 h in serum-free medium containing 0.1% BSA. After incubation, luciferase activities were measured with a luminometer using the Dual-Luciferase reporter assay system according to the manufacturer's instructions (Promega).

Cell fractionation and immunoblotting
HepG2 cells were plated on six-well plates at 2.0 3 10 5 cells/ml for 24 h in DMEM supplemented with 10% fetal calf serum and antibiotics as detailed above. The cells were then treated with EPA, TEPA, and/or T0901317 (50 nM) in serum-free medium containing 0.1% BSA. After incubation for 24 h, membrane fractions and nuclear extracts from cells were prepared by the method of Hannah et al. (22). Protein concentrations were measured according to the Bradford method (23). For immunoblot analysis, given amounts of membrane fractions (25 mg) and nuclear extracts (30 mg) were separated by 7% and 10% SDS-PAGE, respectively, and were then transferred to Clear Blot Membrane-P polyvinylidene difluoride filters (ATTO, Tokyo, Japan). The filters were incubated with rabbit polyclonal anti-SREBP-1 antibody (H-160, 1:400 dilution; Santa Cruz Biotechnology, Santa Cruz, CA). Bound antibodies were visualized with alkaline phosphatase-conjugated anti-rabbit IgG (1:500 dilution; Cell Signaling Technology, Danvers, MA). The intensity of each band was quantified using the Scion Image program (Scion Corp.; available as a free download at http://www.scioncorp.com).

Statistical analysis
Data are reported as means 6 SD. The results were analyzed by one-way ANOVA with Fisher's protected least significant difference using Stat View software (SAS Institute, Cary, NC).

Suppression of triacylglycerol synthesis by TEPA in HepG2 cells
We determined increased triaclyglycerol levels in HepG2 cells to estimate the effect of TEPA on lipogenesis (Fig. 1). The synthetic LXRa agonist, T0901317, significantly increased cellular triaclyglycerol levels (164.6 mg/mg protein) after 72 h of incubation. TEPA at 10 mM significantly suppressed the increase of T0901317-induced triacylglycerol levels more than EPA.

Regulation of lipogenic gene mRNA levels by TEPA
To investigate the mechanism by which TEPA decreased cellular triacylglycerol levels, FAS, SCD-1, and GPA mRNA levels were measured ( Fig. 2A-C). Both EPA and TEPA decreased the expression of FAS and SCD-1 mRNAs in a dose-dependent manner. TEPA significantly decreased the T0901317-induced expression of FAS, SCD-1, and GPA mRNAs more than EPA. No difference in cell viability was observed with TEPA or EPA treatment. Neither EPA nor TEPA had any effect on apolipoprotein A-II mRNA expression as a control gene (Fig. 2D).

Regulation of LXRE in the SREBP-1c promoter by TEPA
We generated a luciferase construct containing the two LXREs that exist in the human SREBP-1c promoter and performed luciferase reporter gene assays in both HepG2 cells and HEK293 cells (Fig. 3), because LXRa and SREBP-1c are key regulators of FAS, SCD-1, and GPA mRNA expression. In the presence of T0901317, pLXREs-Luc activity in HepG2 cells and in HEK293 cells was increased by 160% and 223%, respectively, compared with the vehicle control (Fig. 3A, B). EPA and TEPA suppressed the T0901317 induction of pLXREs-Luc activity (Fig. 3C, D). TEPA suppressed pLXREs-Luc in HepG2 cells and in HEK293 cells maximally at 30 and 20 mM, respectively. There were no significant differences in the suppressive effects of TEPA and EPA.

Regulation of SREBP-1c expression by TEPA
The expression of SREBP-1c mRNA in cells treated with T0901317 was 7.2-fold higher than that in vehicle-treated control cells (Fig. 4A). TEPA significantly downregulated the T0901317 induction of SREBP-1c mRNA in a dosedependent manner, and the suppressive effect of TEPA was equal to that of EPA. To investigate the effect of TEPA on SREBP-1 protein levels, the full-length precursor form in the cell membrane (125 kDa) and the cleaved mature form (68 kDa) in the nuclear extract were estimated by immunoblotting. Because the antibody used cannot distinguish between the SREBP-1c and -1a isoforms, we use the general term SREBP-1 to refer to the results. Representative blots are shown in Fig. 4B. T0901317 increased the levels of both the precursor and the mature forms of SREBP-1. TEPA as well as EPA decreased the T0901317 induction of both the precursor and mature forms of SREBP-1. The decrease of SREBP-1 can be attributed to the suppression of SREBP-1c protein, because neither TEPA nor EPA affected the expression of SREBP-1a mRNA (data not shown).

Downregulation of PGC-1b mRNA level by TEPA
To investigate the mechanism by which TEPA significantly decreased the expression of lipogenic gene mRNAs more than EPA, we examined the effect of TEPA on the  mRNA expression of PGC-1b, the coactivator of LXRa and SREBPs (Fig. 5). We found a significant difference between the effects of TEPA and EPA. TEPA downregulated the level of PGC-1b mRNA at 40 mM by 52% compared with the vehicle control. In contrast, EPA significantly increased the level of PGC-1b mRNA. T0901317 had no effect on PGC-1b mRNA expression.
It has been proposed that the arginine in the LBD of LXRa (R305) interacts with the hydroxyl group of LXRa ligands (24,25). Svensson et al. (25) reported that R305 is the coordinator of the carboxylate group in fatty acids. The interaction between R305 and the carboxylate group of fatty acids may be important for antagonist function. Yoshikawa et al. (11) proposed that the degree of unsaturation of fatty acids is a factor for the inhibitory effect of SREBP-1c expression, but whether they are n-3 or n-6 is irrelevant. The magnitude of inhibition of each PUFA on LXRa LBD activation, using an expression plasmid of the LBD of LXRa fused to the Gal4 DNA binding domain, is as follows: arachidonic acid (C20:4) » docosahexaenoic acid (C22:6) » EPA . linoleic acid (C18:2) . oleic acid (C18:1) .. stearic acid (C18:0) (11). In this study, the suppressive effect of TEPA on T0901317-activated pLXREs-Luc activity was equal to that of EPA. An important factor required for PUFA to antagonize LXRa activation appears to be the number of double bonds and the carbon chain length, but not the double bond configuration. PUFAs having a 20-22 carbon length and more than three double bonds might be suitable antagonists of LXRa.
Unexpectedly, TEPA significantly decreased FAS, SCD-1, and GPA mRNA more than EPA, although there was no significant difference between the suppressive effects of TEPA or EPA on SREBP-1c expression or on pLXREs-Luc activity (Figs. 1A, 2). These results could be attributable in part to the suppression of PGC-1b expression by TEPA  ( Fig. 5), because PGC-1b stimulates hepatic lipid synthesis by coactivating SREBP-1 and LXRa (6). Our data suggest that PGC-1b is important for the full induction of lipogenic genes regulated by SREBP-1 and LXRa; thus, PGC-1b provides a therapeutic target for the metabolic syndrome. Despite the EPA increase of PGC-1b mRNA, EPA decreased the expression of lipogenic genes in this study. This result is caused by the downregulation of SREBP-1c, which is an essential nuclear factor for PGC-1b-mediated transcription of lipogenic genes. The mechanism by which EPA increases PGC-1b mRNA is unclear, but the induction of PGC-1b by EPA is consistent with a previous report (6).
Consequently, trans isomerization of EPA increases the favorable effect of EPA on the expression of lipogenic genes, which leads us to speculate that TEPA decreases serum lipid concentrations more than EPA in vivo. The biological activity of TEPA on lipid metabolism may be distinguished from that of other unfavorable trans fatty acids having a few double bonds with chain lengths of [18][19][20][21][22]. The adverse effect of trans fatty acids on serum lipid profiles may depend on their number of carbon atoms and double bonds. Because the TEPA used in this study was a mixture of trans isomers, it will be important to clarify the structure-activity relationships among a variety of trans isomers in future studies. In addition, insulin is an important inducer of SREBP-1c in vivo. It will be essential to examine the effect of TEPA on insulin-mediated lipogenesis both in vivo and in vitro.