The mechanism mediating the activation of acetyl-coenzyme A carboxylase-alpha gene transcription by the liver X receptor agonist T0-901317.

In birds and mammals, agonists of the liver X receptor (LXR) increase the expression of enzymes that make up the fatty acid synthesis pathway. Here, we investigate the mechanism by which the synthetic LXR agonist, T0-901317, increases the transcription of the acetyl-coenzyme A carboxylase-alpha (ACC alpha) gene in chick embryo hepatocyte cultures. Transfection analyses demonstrate that activation of ACC alpha transcription by T0-901317 is mediated by a cis-acting regulatory unit (-101 to -71 bp) that is composed of a liver X receptor response element (LXRE) and a sterol-regulatory element (SRE). The SRE enhances the ability of the LXRE to activate ACC alpha transcription in the presence of T0-901317. Treating hepatocytes with T0-901317 increases the concentration of mature sterol-regulatory element binding protein-1 (SREBP-1) in the nucleus and the acetylation of histone H3 and histone H4 at the ACC alpha LXR response unit. These results indicate that T0-901317 increases hepatic ACC alpha transcription by directly activating LXR*retinoid X receptor (RXR) heterodimers and by increasing the activity of an accessory transcription factor (SREBP-1) that enhances ligand induced-LXR*RXR activity.

Abstract In birds and mammals, agonists of the liver X receptor (LXR) increase the expression of enzymes that make up the fatty acid synthesis pathway. Here, we investigate the mechanism by which the synthetic LXR agonist, T0-901317, increases the transcription of the acetyl-coenzyme A carboxylase-a (ACCa) gene in chick embryo hepatocyte cultures. Transfection analyses demonstrate that activation of ACCa transcription by T0-901317 is mediated by a cis-acting regulatory unit (2101 to 271 bp) that is composed of a liver X receptor response element (LXRE) and a sterol-regulatory element (SRE). The SRE enhances the ability of the LXRE to activate ACCa transcription in the presence of T0-901317. Treating hepatocytes with T0-901317 increases the concentration of mature sterol-regulatory element binding protein-1 (SREBP-1) in the nucleus and the acetylation of histone H3 and histone H4 at the ACCa LXR response unit. These results indicate that T0-901317 increases hepatic ACCa transcription by directly activating LXR&retinoid X receptor (RXR) heterodimers and by increasing the activity of an accessory transcription factor (SREBP-1) that enhances ligand induced-LXR&RXR activity.-Talukdar, S., and F. B. Hillgartner. The mechanism mediating the activation of acetyl-coenzyme A carboxylase-a gene transcription by the liver X receptor agonist T0-901317. The first committed step of the fatty acid synthesis pathway is the ATP-dependent carboxylation of acetyl-CoA to form malonyl-CoA. This reaction, catalyzed by acetylcoenzyme A carboxylase-a (ACCa), constitutes a key control point in the synthesis of long-chain fatty acids from carbohydrate (1,2). Malonyl-CoA serves as a donor of C 2 units for the synthesis of palmitate catalyzed by fatty acid synthase. Malonyl-CoA is also a substrate of specific elongases involved in the chain elongation of fatty acids to very-long-chain fatty acids (3). The essential role of ACCa in lipid biosynthesis has been confirmed by studies demonstrating that knockout of the ACCa gene disrupts embryonic development before day 7.5 (4).
In lipogenic tissues of birds and mammals, transcription of the ACCa gene is regulated by nutritional and hormonal factors. For example, ACCa transcription is low in livers of starved chicks, and feeding a high-carbohydrate, low-fat diet stimulates an 11-fold increase in ACCa transcription (5). Diet-induced changes in ACCa transcription are mimicked in primary cultures of chick embryo hepatocytes by manipulating the concentrations of hormones and nutrients in the culture medium (6). Incubating chick embryo hepatocytes with the active form of the thyroid hormone 3,59,3-triiodothyronine (T3) stimulates a 5-to 7-fold increase in ACCa transcription. The mechanism by which T3 increases ACCa transcription involves multiple processes. First, T3 interacts with the nuclear 3,59, 3-triiodothyronine receptor (TR) bound to a 3,59,3triiodothyronine response element (T3RE) on the more downstream promoter (promoter 2) of the ACCa gene (7). This T3RE (2101 to 286 bp) is composed of two hexameric half-sites arranged as direct repeats with 4 bp separating the half-sites (DR-4 element). Second, T3 treatment increases the binding of TR&retinoid X receptor (RXR) heterodimers to the ACCa T3RE. The mechanism for this effect has not been defined. Third, T3 treatment increases the binding of sterol-regulatory element binding protein-1 (SREBP-1) to a sterol-regulatory element (SRE) (280 to 271 bp) located immediately downstream of the ACCa T3RE (8). SREBP-1 directly interacts with TR&RXR heterodimers and enhances the ability of this complex to activate ACCa transcription in the presence of T3 (9).
In our studies analyzing the regulation of ACCa transcription by T3, we observed that the ACCa T3RE not only bound protein complexes containing TR&RXR hetero-dimers but also bound protein complexes containing liver X receptor (LXR)&RXR heterodimers (7). LXRs are nuclear hormone receptors that are bound and activated by naturally occurring oxysterols (10,11). Two isoforms of LXR, designated LXRa and LXRb, have been identified in birds and mammals. LXRb is expressed in a wide variety of tissues, whereas LXRa is selectively expressed in liver, adipose tissue, intestine, and macrophages (12,13). LXRs play a key role in regulating cholesterol excretion by mediating the stimulatory effects of oxysterols on the transcription of genes involved in reverse cholesterol transport and bile acid synthesis. For example, naturally occurring oxysterols and synthetic, nonsteroidal LXR agonists activate the transcription of a battery of genes involved in cholesterol efflux (ABCA1, ABCG1, ABCG5, and ABCG8), cholesterol clearance (cholesteryl ester transfer protein and apolipoprotein E), and cholesterol catabolism (cholesterol 7a-hydroxylase) (14)(15)(16). For each of these genes, regulation of transcription by LXR agonists is conferred by a liver X receptor response element (LXRE) that binds LXR&RXR heterodimers. Because oxysterols are produced in proportion to cellular cholesterol content, LXRs have been proposed to function as sensors in a feedforward pathway that stimulates reverse cholesterol transport and cholesterol excretion in response to high cholesterol levels in the diet. Consistent with this proposal, mice lacking the LXRa and/or LXRb gene exhibit diminished cholesterol excretion and increased cholesterol levels in the blood and liver when fed a high-cholesterol diet (17,18).
The ability of LXR agonists to activate genes involved in cholesterol excretion has led to an evaluation of the atheroprotective properties of these compounds in murine models of atherosclerosis. Oral administration of the synthetic LXR ligand/agonist, T0-901317, to mice lacking the LDL receptor or apolipoprotein E causes an increase in blood HDL levels and reverses the formation of atherosclerotic lesions in the aorta (19,20). LXR agonists also improve insulin sensitivity in murine models of type 2 diabetes (21,22). These exciting findings are tempered by the observation that treatment with T0-901317 also causes hypertriglyceridemia and the development of a fatty liver in rodents and chickens (23)(24)(25)(26). These undesired effects of T0-901317 are caused by alterations in the expression of enzymes that make up the fatty acid synthesis pathway. For example, T0-901317 treatment increases the hepatic expression of the mRNAs encoding ACCa, FAS, ATP-citrate lyase (ATP-CL), and stearoyl-coenzyme A desaturase-1 (SCD1) (25)(26)(27). The aim of this study was to determine the mechanism by which T0-901317 regulates the expression of ACCa in avian liver. We show that T0-901317 acts directly on the liver to increase the expression of ACCa and that the extent of this effect is modulated by the presence of insulin and T3. We further show that T0-901317 increases ACCa transcription by activating LXR&RXR heterodimers bound to the ACCa LXRE/T3RE and that SREBP-1 interacts with LXR&RXR to enhance the stimulatory effects of T0-901317 on ACCa transcription.

EXPERIMENTAL PROCEDURES
Cell culture and analytical assays Hepatocytes were isolated from livers of 19 day old chick embryos as described previously (28). Cells were incubated in serum-free Waymouth's medium MD752/1 containing penicillin (60 mg/ml) and streptomycin (100 mg/ml) on untreated Petri dishes at 408C in a humidified atmosphere of 5% CO 2 and 95% air. Hormone and other additions were as described in the figure legends. The triacylglycerol concentration of the culture medium was measured using an enzymatic kit (Sigma).

Plasmids
Reporter plasmids are named by designating the 59 and 39 ends of the ACCa DNA fragment relative to the transcription start site of promoter 2. A series of 59 deletions and 39 deletions of ACCa promoter 2 in the context of p[ACC22054/1274] chloramphenicol acetyltransferase (CAT) have been described (7). An ACCa promoter construct containing a mutation of the SRE between 279 and 272 bp in the context of p[ACC2108/ 1274]CAT has been described (9). p[ACC2108/266]TKCAT, p[ACC284/266]TKCAT, and pTKCAT constructs, containing mutations in the 2108 to 266 bp ACCa fragment, have been described (9).

Transient transfection
Chick embryo hepatocytes were transfected as described by Zhang, Yin, and Hillgartner (7). Briefly, cells were isolated and incubated on 60 mm Petri dishes. At 6 h of incubation, the medium was replaced with one containing 10 mg of lipofectin (Invitrogen), 1.5 mg of p[ACC22054/1274]CAT or an equimolar amount of another reporter plasmid, and pBluescript KS(1) to bring the total amount of transfected DNA to 1.5 mg per plate. At 18 h of incubation, the transfection medium was replaced with fresh medium with or without T0-901317 (6 mM). At 66 h of incubation, chick embryo hepatocytes were harvested and cell extracts were prepared as described (38). CAT activity (39) and protein (40) were assayed by the indicated methods.

Western blot analysis
Nuclear extracts were prepared from chick embryo hepatocytes as described (9). The proteins of the nuclear extract were subjected to electrophoresis on 10% SDS-polyacrylamide gels and then transferred to polyvinylidene difluoride membranes (Millipore) using an electroblotting apparatus (Bio-Rad Laboratories, Hercules, CA). Immunoblot analysis was carried out using a mouse monoclonal antibody against SREBP-1 (IgG-2A4; American Type Culture Collection, Manassas, VA). Antibody/protein complexes on blots were detected using enhanced chemiluminescence (Amersham Biosciences). Chemiluminescence on the blots was visualized using a FluorChem 8000 imager (Alpha Innotech Corp.), and signals for the mature form of SREBP-1 were quantified using FluorChem V200 software.

Histone acetylation
The extent of histone acetylation on ACCa promoter 2 was measured using a chromatin immunoprecipitation (ChIP) assay. The procedure for this assay was as described by Yin et al. (41). ChIP assays were carried out with antibodies against acetylhistone H3 (06-599) and acetyl-histone H4 (06-866; Upstate Biotechnology). Precipitated DNA was analyzed by PCR using Taq DNA polymerase (New England Biolabs) and primers specific for the ACCa and SCD1 promoters. The cycling parameters were as follows: 1 cycle of 958C for 4 min; 30 cycles of 958C for 1 min, 618C for 1 min, and 728C for 1 min; and 1 cycle of 728C for 10 min. The forward primer of the ACCa gene was 59-TCC-CCTCCGTCAGCAGCCAATGGG-39; the reverse primer was 59-ATCCCCGGTCCCGCCCTCGGCTCC-39. The forward primer of the SCD1 gene was 59-AGCGAACAGCAGATTGCGGCAG-39; the reverse primer was 59-TCTCGGCGTGCCAGAAGGGAGGT-39. Amplified products were subjected to electrophoresis on 2% agarose gels and visualized by ethidium bromide staining.

Statistical methods
Data were subjected to ANOVA, and statistical comparisons were made with Dunnett's test or Student's t-test. Statistical significance was defined as P , 0.05.

LXR activation increases the expression of ACCa in primary cultures of chick embryo hepatocytes
Oral administration of T0-901317 to chickens and rats causes a 2-to 3-fold increase in hepatic ACCa mRNA levels (25,26). To investigate whether this phenomenon was attributable to a direct effect of T0-901317 in the liver, we determined whether T0-901317 regulated the expression of ACCa in primary cultures of chick embryo hepatocytes. Incubating hepatocytes with T0-901317 in the absence of other hormones for 24 h caused a 2.4-fold increase in the abundance of ACCa mRNA (Fig. 1). Treatment with the RXR agonist 9-cis retinoic acid had no effect on ACCa mRNA abundance in the absence or presence of T0-901317. Thus, LXR agonists but not RXR agonists regulate ACCa expression in hepatocyte cultures.
In chick embryo hepatocytes, insulin has no effect by itself but amplifies the increase in ACCa transcription caused by T3 (6). This observation prompted us to investigate whether insulin modifies the effects of T0-901317 on ACCa expression. Incubating hepatocytes with T0-901317 stimulated a greater increase in ACCa mRNA abundance in the presence of insulin (3.5-fold) than in the absence of insulin (2.4-fold) (Fig. 1). Thus, as observed for T3 regulation of ACCa, insulin enhances the stimulatory effects of T0-901317 on ACCa expression.
In previous work, we showed that both TR and LXR bind the ACCa T3RE as heterodimers with RXR (7). This observation raised the possibility that a common cis-acting regulatory sequence is involved in mediating the effects of T0-901317 and T3 on ACCa transcription. As a first step in investigating this possibility, we determined the effects of T0-901317 on the expression of ACCa in the presence of T3. Incubating hepatocytes with T3 and insulin caused a 4.6-fold increase in the abundance ACCa mRNA abundance (Fig. 1). Addition of T0-901317 in the presence of T3 and insulin caused a further increase in ACCa mRNA abundance, although the magnitude of this effect (39%) was substantially smaller than the effect of T0-901317 on ACCa expression in the presence of insulin alone (350%). Addition of T0-901317 in the presence of T3, insulin, and 9-cis retinoic acid had no effect on ACCa mRNA abundance. The nonadditive effects of T0-901317 and T3 on ACCa mRNA abundance support the proposal that a common cis-acting sequence(s) mediates the actions of T0-901317 and T3 on ACCa transcription.
We also investigated the effects of T0-901317 on the expression of other lipogenic enzymes. In hepatocytes incubated in the absence and presence of insulin, addition of T0-901317 to the culture medium increased the abundance of mRNAs encoding FAS, SCD1, and ATP-CL (Fig. 1). T0-901317-induced expression of FAS, SCD1, and ATP-CL was higher in the presence of insulin than in the absence of insulin. In hepatocytes incubated with T3 and insulin, addition of T0-901317 caused a small increase (32%) in the abundance of FAS mRNA but had no effect on the abundance of SCD1 mRNA or ATP-CL mRNA. Incubating hepatocytes with 9-cis retinoic acid in the absence or presence of T0-901317 had no effect on FAS, SCD1, and ATP-CL mRNA levels. Thus, regulation of the expression of FAS, SCD1, and ATP-CL by agonists of LXR and RXR is similar to that of ACCa.
In contrast to ACCa, FAS, SCD1, and ATP-CL, the ability of T0-901317 to increase the expression of malic enzyme was substantially lower than that of T3. Treatment with T0-901317 and insulin caused a 2.9-fold increase in malic enzyme mRNA abundance, whereas treatment with T3 and insulin caused a 35-fold increase in malic enzyme mRNA abundance (Fig. 1). These observations are consistent with previous work demonstrating that the major T3RE mediating T3 regulation of malic enzyme transcription lacks the ability to bind LXR&RXR heterodimers (7).
To determine whether the T0-901317-induced increase in lipogenic enzyme expression was associated with an increase in triacylglycerol production, the triacylglycerol concentration in the culture medium was monitored in hepatocytes treated with or without T0-901317. The triacylglycerol concentration of the culture medium increased progressively during a 48 h incubation period (Fig. 2). The extent of the increase in triacylglycerol concentration was higher in cells treated with T0-901317 relative to cells not receiving T0-901317. An increase in hepatic lipogenic enzyme expression and triacylglycerol secretion likely contributes to the hypertriglyceridemia observed in animals treated with T0-901317 (23,24,26).

Identification of a LXR response unit that mediates the effects of T0-901317 on ACCa transcription
Previous studies have shown that T3 regulation of ACCa transcription is mediated by a 23 bp region (2101 to 271 bp) in promoter 2 of the ACCa gene (7). This region contains a DR-4 element (2101 to 286 bp) that binds heterodimers composed of TR&RXR and LXR&RXR and a SRE (282 to 271 bp) that binds SREBP-1. To determine the role of these sequences and other sequences in the ACCa gene in mediating the stimulatory effects of T0- Fig. 1. The liver X receptor (LXR) agonist T0-901317 increases the expression of acetyl-coenzyme A carboxylase-a (ACCa) and other lipogenic enzymes in primary cultures of chick embryo hepatocytes. Hepatocytes were isolated as described in Experimental Procedures and incubated in serum-free Waymouth's medium. At 18 h of incubation, the medium was changed to one of the same composition supplemented with or without T0-901317 (6 mM) in the absence or presence of insulin (50 nM), insulin and 9-cis retinoic acid (9-Cis RA; 100 nM), insulin and 3,59,3-triiodothyronine (T3; 1.5 mM), or insulin, T3, and 9cis retinoic acid. After 28 h of treatment, total RNA was isolated and the abundance of mRNA encoding ACCa, FAS, stearoyl-coenzyme A desaturase-1 (SCD1), ATP-citrate lyase (ATP-CL), malic enzyme (ME), ABCA1, and LXRa was measured by Northern analysis. Levels of mRNA in cells treated without T0-901317, 9-cis retinoic acid, insulin, and T3 were set at 1, and the other values were adjusted proportionally. Values are means 6 SEM of four experiments. Hybridization signals from a representative experiment are shown for each mRNA. Ribosomal subunits (27S and 18S) stained with ethidium bromide are shown as controls for RNA loading. a Mean is significantly (P , 0.05) different from that of cells treated without T0-901317, 9-cis retinoic acid, insulin, and T3; b mean is significantly (P , 0.05) different from that of cells treated with insulin; c mean is significantly (P , 0.05) different from that of cells treated with insulin and T3. 901317 on ACCa transcription, transient transfection experiments were performed using reporter constructs containing portions of ACCa promoter 2 linked to the CAT gene. In chick embryo hepatocytes transfected with a reporter construct containing 2,054 bp of 59-flanking DNA, T0-901317 caused a 2.9-fold increase in promoter activity (Fig. 3). 59 deletion of ACCa sequences to 2391, 2136, and 2108 bp had no effect on T0-901317 responsiveness. Deletion of ACCa sequences containing the DR-4 element (2108 to 284 bp) abolished the stimulatory effect of T0-901317 on ACCa transcription. Further deletion to 241 bp had no effect on T0-901317 responsiveness. Mutation of the SRE (280 to 271 bp) in the context of 108 bp of 59-flanking DNA caused a 49% reduction in T0-901317 responsiveness. These results indicate that the DR-4 element (LXRE) is required for T0-901317 regulation of ACCa transcription and that the SRE enhances the ability of the LXRE to activate ACCa transcription in the presence of T0-901317.
To determine whether the functional interaction between the ACCa LXRE and SRE required the presence of additional cis-acting sequences, hepatocytes were transfected with constructs containing fragments of the ACCa gene linked to the minimal promoter of the herpes simplex virus thymidine kinase (TK) gene. The TK promoter alone was unresponsive to T0-901317 (Fig. 4). When a DNA fragment containing both the ACCa LXRE and the ACCa SRE (2108 to 266 bp) was linked to the TK promoter, treatment with T0-901317 caused a 5.5-fold increase in promoter activity. Mutation of the ACCa SRE in the context of the ACCa 2108 to 266 bp fragment caused a 49% decrease in T0-901317 responsiveness. When a DNA frag-ment containing the ACCa SRE alone (284 to 266 bp) was appended to the TK promoter, T0-901317 treatment had no effect on promoter activity. These data demonstrate that the ACCa SRE can function alone to enhance T0-901317 regulation conferred by the ACCa LXRE. Thus, a region of the ACCa gene containing a LXRE and a SRE is responsible for mediating the effects of T0-901317 on ACCa promoter 2 activity. We refer to this region as the ACCa liver X receptor response unit (LXRU).

LXR activation increases the abundance of mature SREBP-1 in chick embryo hepatocytes
In our studies analyzing the regulation of the ACCa gene by T3, we showed that T3 treatment increased the abundance of the mature, active form of SREBP-1 in chick embryo hepatocytes and that this effect contributed to the activation of ACCa transcription by T3 (8). This observation prompted us to ask whether T0-901317 regulated SREBP-1 levels in chick embryo hepatocyte cultures. The time course of the effects of T0-901317 on the abundance of mature SREBP-1 protein, SREBP-1 mRNA, and ACCa mRNA was determined in hepatocytes cultured in the presence of insulin. Treatment with T0-901317 for 6 h caused a 1.5-fold increase in mature SREBP-1 concentration (Fig. 5). A larger increase in mature SREBP-1 concentration (2.2-fold) was observed after 24 h of treatment with T0-901317. In contrast to the time course for mature SREBP-1, the T0-901317-induced stimulation of ACCa mRNA levels occurred at a later time point (between 24 and 48 h of treatment). This observation is consistent with the proposal that alterations in SREBP-1 levels play a role in mediating the regulation of ACCa transcription by T0-901317. Treatment with T0-901317 increased the abundance of SREBP-1 mRNA, and this effect was maximal (1.9-to 2.1-fold) at or before 2 h of incubation. Thus, T0-901317-induced changes in mature SREBP-1 concentration appear to be mediated by a pretranslational mechanism.

Effect of LXR activation on histone acetylation of the ACCa gene
The T0-901317-induced increase in ACCa transcription may be mediated in whole or in part by alterations in histone acetylation. Previous studies have shown that the activation of LXR&RXR heterodimers by LXR ligands/ agonists triggers the recruitment of coactivator complexes containing histone acetyltransferase activity (42,43). Increased histone acetylation causes a chromatin decondensation that enhances the accessibility of the basal transcriptional machinery and other transcription factors to the target promoter. To investigate the role of histone acetylation in mediating the activation of ACCa transcription by T0-901317, ChIP experiments were performed in chick embryo hepatocytes incubated in the absence or presence of T0-901317. Hepatocytes were treated with 1% formaldehyde to cross-link DNA to associated proteins. Protein-DNA complexes were immunoprecipitated with an antibody against acetylated histone H3 or an antibody against acetylated histone H4. Immunoprecipitated DNA was analyzed by PCR using primers that flanked the ACCa LXRU. In hepatocytes incubated in the absence of T0-901317, acetylation of histone H3 and histone H4 was detected at the ACCa LXRU. Addition of T0-901317 to the culture medium stimulated a rapid increase (<2 h) in the acetylation of histone H3 and H4 (Fig. 6, Table 1). Histone acetylation remained high for 6 h of T0-901317 treatment and then declined between 6 and 24 h of T0-901317 treatment. These results support the proposal that an increase in histone acetylation plays a role in mediating the effects of T0-901317 on ACCa transcription.
We also used ChIP to assess the effects of T0-901317 treatment on histone acetylation in an uncharacterized region of the SCD1 promoter using a primer set that amplified SCD1 sequences between 2369 and 2127 bp. In contrast to the data for the ACCa gene, treatment with T0-901317 had no effect on histone acetylation of this region of the SCD1 gene (Fig. 6). This observation indicates that the effects of T0-901317 on histone acetylation are sequence-specific.

DISCUSSION
In previous work analyzing the effects of T3 on ACCa transcription in avian hepatocytes, we identified a T3RE that conferred T3 regulation on ACCa promoter 2 (7). Interestingly, this T3RE not only bound protein complexes containing TR&RXR heterodimers but also bound protein complexes containing LXR&RXR heterodimers. In this report, we provide functional evidence that LXR&RXR heterodimers regulate ACCa transcription. A specific ligand/agonist of LXR (T0-901317) activates ACCa transcription, and this effect is mediated by the LXRE/T3RE in ACCa promoter 2. We also demonstrate that SREBP-1 is an accessory factor that enhances the ability of LXR&RXR to increase ACCa transcription in the presence of T0-901317 and that LXR activation by T0-901317 increases the concentration of the mature, active form of SREBP-1 in chick embryo hepatocytes.
SREBP-1 also enhances the ability of TR&RXR to activate ACCa transcription in the presence of T3 (9). In analyzing the mechanism mediating this effect, we showed that SREBP-1 interacted physically with TR and that binding of T3 to TR enhanced this interaction. We postulated that SREBP-1&SREBP-1 homodimers formed a tetrameric complex with TR&RXR heterodimers and that tetrameric complex formation enhanced the recruitment of coactivators to ACCa promoter 2. We also showed in these studies that LXR&RXR heterodimers do not interact physically with SREBP-1. Thus, in contrast to the mechanism by which SREBP-1 enhances TR&RXR activity, direct interactions between SREBP-1 and LXR&RXR do not play a role in mediating the stimulatory effects of SREBP-1 on LXR&RXR activity.
The mechanism by which LXR agonists regulate transcription has been analyzed for other lipogenic genes. In human hepatoma cells, T0-901317 activation of FAS transcription is mediated by a single LXRE in the FAS promoter (52). In primary rat hepatocyte cultures, the T0-901317-induced increase in SREBP-1c transcription is mediated by two LXREs in the SREBP-1c promoter (53,54). Both the FAS promoter and the SREBP-1c promoter contain one or more SREs that are located z110-540 bp downstream of the LXRE(s). In cells that express physiological levels of LXR, these SREs enhance basal transcription but have little or no effect on the regulation of transcription by T0-901317. In contrast to these findings, the SRE in ACCa promoter 2 enhances the T0-901317 regulation of transcription but has no effect on basal transcription in chick embryo hepatocytes (Figs. 3, 4). These gene-specific differences in SRE activity may be attributable to variations in the proximity of the SRE relative to the LXRE and other cis-acting regulatory sequences. For example, the close association of the SRE with the LXRE/T3RE in ACCa promoter 2 may facilitate interactions between LXR&RXR and SREBP-1 and the ability of SREBP-1 to enhance T0-901317 responsiveness, whereas the wide separation of the SREs and LXREs in the FAS and SREBP-1c promoters may impede interactions between LXR&RXR and SREBP-1 and the ability of SREBP-1 to enhance T0-901317 responsiveness. Previous studies have shown that the SREs in the FAS and SREBP-1c promoters are closely linked to a binding site for nuclear factor-Y (NF-Y) and that SREBP-1 activity is dependent on interactions between SREBP-1 and NF-Y (53)(54)(55)(56). These interactions enhance the ability of SREBP-1 to stimulate basal transcription. In contrast, the SRE in ACCa promoter 2 is not closely associated with binding sites for NF-Y. The lack of interaction of SREBP-1 with NF-Y on ACCa promoter 2 may explain why the ACCa SRE is not effective in modulating basal transcription.
Promoter 2 of the human ACCa gene and the rat ACCa gene contains two closely spaced SREs that mediate the effects of sterol depletion on ACCa transcription (57,58). In contrast to promoter 2 of the chicken ACCa gene, a sequence resembling a DR-4 element is not present in the region flanking the SREs in the human and rat ACCa promoters. Thus, the role of SREBP-1 in mediating the stimulatory effects of T0-901317 on ACCa transcription may vary depending on the class of animals.
Another finding of this study is that LXR plays a permissive role in mediating the actions of insulin on ACCa transcription in chick embryo hepatocytes. Insulin stimulates ACCa expression in the presence of T0-901317 but has no effect on ACCa expression in the absence of T0-901317 (Fig. 1). In rat hepatocytes, insulin enhances the ability of T0-901317 to increase mature SREBP-1 concentration as a result of a stimulatory effect of insulin on the processing of precursor SREBP-1 to mature SREBP-1 (59). We have confirmed this finding in chick embryo hepatocytes (data not shown). Because SREBP-1 enhances the LXR activation of ACCa transcription in chick embryo hepatocytes, we postulate that alterations in mature SREBP-1 abundance mediate the stimulatory effects of insulin on ACCa transcription. In support of this hypothesis, insulin does not increase the expression of ABCA1 in the presence of T0-901317 (Fig. 1). Previous work has shown that the ABCA1 gene is not a target of SREBP-1 (60).
Previous studies performed in rat hepatocytes indicate that insulin induces ACCa expression by increasing the activity of LXR. For example, Tobin et al. (61) have shown that insulin stimulates a 10-fold increase in the expression of LXRa mRNA and that ablation of the LXRa gene abolishes the stimulatory effect of insulin on ACCa expression. Others have shown that insulin increases the transcription of SREBP-1c and that this effect is mediated by two LXR&RXR binding sites in the SREBP-1c promoter (53,54). SREBP-1c is a key factor mediating the effects of insulin on ACCa transcription in rat hepatocytes (27,62). These findings contrast with our data indicating that LXR plays a permissive role in mediating the effects of insulin on ACCa transcription in chick embryo hepatocytes. The reason for the differences between birds and rodents in the mechanism by which insulin regulates ACCa transcription is not clear. They may reflect subtle classdependent differences in the role of insulin in the control of lipogenesis and/or other metabolic processes in liver.
The ACCa LXRE/T3RE enhances ACCa transcription in both the absence and presence of T0-901317 and T3, with a greater activation observed in the presence of T0-901317 and T3 (Figs. 3, 4) (7). We previously hypothesized that the enhancer activity of the ACCa LXRE/ T3RE in the absence of T0-901317 and T3 was mediated by LXR&RXR heterodimers, as the primary protein complexes that bind the ACCa LXRE/T3RE in the absence of T0-901317 and T3 contain LXR&RXR heterodimers (7). Recent studies have shown that the unliganded form of LXR&RXR represses gene transcription as a result of its ability to interact with corepressor proteins and that the addition of LXR ligand increases gene transcription by  1) in chick embryo hepatocyte cultures. Chick embryo hepatocytes were isolated and incubated in serum-free Waymouth's medium containing insulin. At 18 h of incubation, the medium was changed to one of the same composition. T0-901317 was added at this time. After 2, 6, 12, 24, and 48 h of T0-901317 treatment, cellular extracts or total RNA were prepared as described in Experimental Procedures. A: The abundance of mature SREBP-1 in nuclear extracts was measured by Western analyses. The abundance of SREBP-1 mRNA and ACCa mRNA was measured by Northern analysis. These data are from a representative experiment. B: Signals for mature SREBP-1 protein from Western analyses and SREBP-1 mRNA and ACCa mRNA from Northern analyses were quantitated. Levels of mature SREBP-1 protein, SREBP-1 mRNA, and ACCa mRNA in hepatocytes treated with T0-901317 for 0 h were set at 1. Values are means 6 SEM of four experiments. * The mean is significantly (P , 0.05) different from that of cells incubated with T0-901317 for 0 h or without T0-901317 for 24 or 48 h. causing the release of corepressors and the recruitment of coactivators to LXR&RXR (63). In view of these observations, we further hypothesize that LXR&RXR complexes associated with the ACCa LXRE/T3RE in the absence of T0-901317 and T3 are bound by endogenous LXR and/or RXR ligands. Several lines of evidence support this proposal. First, treatment of chick embryo hepatocytes with naturally occurring agonists of LXR [22-(R)-hydroxy cholesterol] and RXR (9-cis retinoic acid) has little or no effect on ACCa mRNA abundance ( Fig. 1 and data not  shown). Second, the ACCa LXRE/T3RE is not associated with the corepressor protein, nuclear receptor corepressor, in the absence of T0-901317 and T3 (41). Last, the ACCa LXRE/T3RE is associated with acetylated histone H3 and acetylated histone H4 in the absence of T0-901317 and T3 (Fig. 6). The ability of T0-901317 to increase the transcriptional activity and histone acetylation of the ACCa LXRE/T3RE is likely attributable to the fact that this synthetic agonist is more effective than endogenous LXR agonists in stimulating the recruitment of histone acetyltransferase-containing coactivators to LXR&RXR (42).
In summary, we show that T0-901317 activates ACCa transcription by increasing the activity of LXR&RXR and SREBP-1 and the acetylation of histone H3 and histone H4 on ACCa promoter 2. The identification of small molecules that selectively inhibit one or more of these processes represents a potential strategy to enhance the utility of T0-901317 in the treatment of atherosclerosis and type 2 diabetes. The feasibility of this approach is supported by recent studies demonstrating that suppression of ACCa expression reverses hepatic steatosis and insulin resistance in rats fed a high-fat diet (64).  6. Treatment of chick embryo hepatocyte cultures with T0-901317 causes a transient increase in histone acetylation at the ACCa liver X receptor response unit (LXRU). Chick embryo hepatocytes were isolated and incubated in serum-free Waymouth's medium containing insulin. At 18 h of incubation, the medium was changed to one of the same composition. T0-901317 was added at this time. After 2, 6, 24, and 48 h of T0-901317 treatment, the association of acetylated histone H3 (Ac-H3) and acetylated histone H4 (Ac-H4) with ACCa and SCD1 genomic sequences was measured. Chromatin immunoprecipitation assays were performed as described in Experimental Procedures. Immunoprecipitates were analyzed by PCR using primers that flanked the LXRU of ACCa promoter 2 and an uncharacterized region of the SCD1 promoter. The regions of the ACCa and SCD1 genes that were amplified by PCR are indicated at top. Chromatin samples that were processed in parallel without the application of primary antibody served as controls (No Ab). The input lanes show the results of PCR using chromatin samples taken before the immunoprecipitation step. Results are representative of three independent experiments.  Signals from chromatin immunoprecipitation assays using the indicated antibodies were quantitated. The data are expressed as fold differences relative to samples from hepatocytes treated with T0-901317 for 0 h. Values are means 6 SEM of three experiments. Results from a representative experiment are shown in Fig. 6.
a Significantly different (P , 0.05) from cells treated with T0-901317 for 0 h.