|Home | About | Journals | Submit | Contact Us | Français|
The liver X receptors (LXRs) are nuclear receptors which play important roles in the regulation of lipid metabolism. In this study, we demonstrate that RIP140 is a cofactor for LXR in liver. Analysis of RIP140 null mice and hepatocytes depleted of RIP140 indicate that the cofactor is essential for the ability of LXR to activate the expression of a set of genes required for lipogenesis. Furthermore we demonstrate that RIP140 is required for the ability of LXR to repress the expression of the PEPCK gene in Fao cells and mice. Thus, we conclude that the function of RIP140 as a cofactor for LXR in liver varies according to the target genes and metabolic process, serving as a coactivator in lipogenesis but as a corepressor in gluconeogenesis.
Liver X receptors (LXR) are members of the nuclear receptor superfamily of transcription factors that regulate the expression of a number of genes involved in lipid, cholesterol, and glucose metabolism in hepatocytes and other cell types (1-5). LXRs serve as intracellular sensors of cholesterol, and oxidized derivatives of cholesterol (oxysterols) have been identified as endogenous ligands (6, 7). This signalling pathway is important for the control of three important processes in hepatocytes, namely the synthesis of steroid hormones, bile acids and cholesterol (6).
In liver, excess cholesterol is converted into bile acids and exported from the cell while, at the same time, cholesterol biosynthesis and uptake of lipoprotein cholesterol is reduced. Functional response elements for LXRα and β have been identified in the promoters of several genes that encode rate-limiting enzymes, transporters and regulators of these processes (7-11). Insights into the function of LXR in cholesterol metabolism were provided by the generation of mice devoid of LXRα (LXRα−/−) (12). For example, transcription of the gene encoding cholesterol 7α-hydroxylase (Cyp7a), the rate-limiting enzyme in bile acid synthesis, is impaired in LXRα−/− mice leading to cholesterol accumulation and ultimately defective liver function.
LXR also plays a key role in the regulation of hepatic lipid metabolism by activating lipogenesis. This is achieved by increasing the expression of sterol regulatory element-binding protein (SREBP)-1c that controls the expression of fatty acid synthase (FAS) and other key genes involved in fatty acid biosynthesis (13-15). Basal LXR activity is essential for the expression of SREBP1c in hepatocytes underlining the importance of LXR for hepatic lipogenesis (16). In addition, LXR regulates the expression of the FAS and other lipogenic genes directly (17) while fatty acids were identified as positive regulators of LXRα gene expression in cultured hepatocytes (18). These observations suggest an important cross-talk between fatty acid- and cholesterol-mediated regulation of lipid metabolism.
LXRs are also regulators of glucose metabolism. LXR agonists have been shown to improve glucose tolerance and insulin sensitivity in diabetic animals by increasing GLUT4 expression and glucose uptake in adipocytes and by suppressing gluconeogenesis, in particular the genes encoding rate-determining enzymes, such as phosphoenolpyruvate carboxykinase (PEPCK) and glucose-6-phosphatase (G6Pase) (19-21).
A number of transcriptional cofactors have been found to play a crucial role in the integration of metabolic processes, including PGC-1α, PGC-1β, SRC1 and TIF2 (22, 23). In addition, corepressors can regulate networks of metabolic genes. For example, RIP140 promotes the storage of lipids in adipose tissue by inhibiting the expression of genes involved in mitochondrial biogenesis, fatty acid oxidation and oxidative phosphorylation (24-26). RIP140 interacts with a number of nuclear receptors, including PPARs, ERRs and LXR that regulate metabolic pathways (27-29). In this study we address the role of RIP140 in LXR regulated hepatic lipid and glucose metabolism. Experiments in WT and RIP140 null mice fed different diets and cultured cells depleted of RIP140 reveal that RIP140 is required for LXR function in two different ways: the induction of lipogenesis and the repression of the PEPCK gene. Therefore we conclude that RIP140 is involved in both positive and negative effects of LXR in transcriptional regulation in hepatocytes.
To study whether RIP140 plays a regulatory role in hepatic lipid metabolism we monitored the levels of triglycerides in wild-type (WT) and RIP140 null mice fed chow or western diet for 4 weeks. Hepatic and plasma triglyceride levels were similar in both WT and RIP140 null mice fed a chow diet however, the expected increase in hepatic triglycerides after challenge with a western diet was less marked in mice devoid of RIP140 (Figure 1A). Plasma triglyceride levels are also slightly less in RIP140 null mice after feeding with western diet (Figure 1A). Given the importance of LXR in hepatic lipogenesis (13) we investigated the possibility that RIP140 might function as a cofactor for LXR in the regulation of this process. Mice were fed a synthetic LXR agonist (T0901317) for 3 days. Analysis of hepatic lipogenic gene expression indicated that while LXR mRNA levels were essentially unchanged after T0901317 administration (data not shown) there was a significant increase in SREBP1c, steroyl-CoA desaturase (SCD-1), FAS and acetyl-CoA carboxylase (ACC-1) expression in WT but not RIP140 null mice (Figure 1B). Next we measured the amount of hepatic triglycerides after LXR activation. As expected the treatment with T0901317 resulted in lipid accumulation in the liver of WT mice. In contrast, mice devoid of RIP140 have much less hepatic triglycerides after the same treatment (Figure 1C). These results suggest that the failure of RIP140 null mice to increase lipogenesis in resonse to LXR activation contributes to the reduced accumulation of triglycerides in the liver of these animals.
We isolated hepatocytes from wild-type and RIP140 null mice to test if the observed effects of RIP140 are autonomous to hepatocytes. First we measured the total lipid content in both cell types after stimulation with the LXR agonist. Treatment of wild-type hepatocytes with T0901317 resulted in an increased lipid accumulation however hepatocytes derived from RIP140 null mice lacked this increase (Figure 2A). Next we measured the expression of the lipogenic genes FAS, SCD-1, and SREBP1c in cells treated with either T0901317 or vehicle (DMSO). As expected this treatment resulted in a increased expression of these genes in wild-type cells. In contrast the expression of these genes remains unchanged in the cells derived from RIP140 null mice (Figure 2B). These findings suggest that the effect of RIP140 on LXR is autonomous for hepatocytes.
To investigate the function of RIP140 in lipogenesis in more detail we used human HuH7 hepatoma cells. Cells were infected with an adenovirus expression siRNA for RIP140 or a control virus, respectively. Treatment with the siRIP140 virus resulted in an effective knock-down of RIP140 protein as shown by Western blotting (Figure 3A). The expression of LXR, SREBP1c, SCD-1, and FAS was markedly increased in response to treatment with T0901317 in control samples, however, this was completely abrogated when RIP140 was depleted by siRNA (Figure 3B). Thus, it appears that RIP140 functions as a positive cofactor for LXR and is required to stimulate expression of the lipogenic programme.
Nuclear extracts were prepared from HuH7 cells, which had been treated with 1μM T0901317 or vehicle alone for 2 hours. Coimmunoprecipitation experiments were performed using conjugated RIP140 antibody or a control antibody. RIP140 and LXR are expressed in HuH7 cells and LXR protein was detected by Western blot in the precipitates with RIP140 antibody but not the control antibody. Consistent with previous in vitro observations the LXR agonist had very little effect on the interaction of RIP140 and LXR (28)(Figure 4A). Next we performed chromatin immunoprecipitation (ChIP) assays to test if RIP140 binds to known target genes of LXR involved in lipogenesis in these cells. Chromatin fragments for FAS, SREBP1c and LXRα gene promoters were amplified by PCR using primers for reported functional LXREs in these genes (11, 15, 17). Our experiments indicate that both LXR and RIP140 bind directly to the FAS, SREBP1c and LXRα gene promoter in the vicinity of the LXRE but not to a distal region of the FAS gene which does not contain a LXRE and was used as a control for specific binding in this assay (Figure 4B). Consistent with the LXR-RIP140 interaction studies there is little effect of the LXR agonist on the promoter occupation in these cells. These results suggest that RIP140 is directly involved in the transcriptional regulation of these genes in HuH7 cells.
Finally, we investigated the ability of RIP140 to stimulate transcription from the FAS promoter in transiently transfected HuH7 cells (Figure 4C). The activity of a FAS luciferase reporter gene was increased approximately 3-fold in the presence of T0901317 and was further increased up to 10-fold with increasing expression of RIP140. Importantly the ability of RIP140 to potentiate FAS promoter activity is dependent on a functional LXRE in the promoter (Figure 4C). Thus we conclude that RIP140 is required for the induction of FAS expression by LXR.
Recently a role for LXR in the regulation of hepatic glucose metabolism has emerged (19-21). We therefore investigated the possibility that RIP140 is also important for this aspect of LXR function in liver. In particular LXR has been shown to inhibit the expression of two key gluconeogenic enzymes, phosphoenolpyruvate carboxykinase (PEPCK) and glucose-6-phosphatase (G6Pase) (20). We therefore determined the effect of T0901317 administration on the expression of these genes in RIP140 null mice and WT controls. The expression of PEPCK and G6Pase was found to be down-regulated in WT animals following administration of the LXR agonist, as expected (Figure 5A). However, the repression of PEPCK, but not G6Pase, by T0901317 was abrogated in mice devoid of RIP140 (Figure 5A). Next we measured the effect of LXR activation in hepatocytes isolated from wild-type or RIP140 null mice. Glucose production (Figure 5B) and PEPCK gene expression (Figure 5C) was reduced in wild-type but not in RIP140 null cells. We next investigated the expression of PEPCK and G6Pase in Fao cells, which were previously utilized to study their transcriptional regulation (30). We were able to confirm that the expression of both genes was suppressed by LXR ligand and that the depletion of RIP140 by means of siRNA results in a loss of PEPCK but not G6Pase repression by T0901317 (Figure 5D). These results suggest that the repression of PEPCK gene transcription by LXR is dependent on RIP140.
We next investigated the ability of RIP140 and LXRα to interact in Fao cells by coimmunoprecititation experiments. As shown in Figure 6A RIP140 and LXRα interact but the LXR agonist had very little effect on the interaction. Next we performed ChIP assays to determine whether RIP140 is recruited to the PEPCK gene in Fao cells. We found that LXR and RIP140 bind in the vicinity of the putative LXRE in the PEPCK gene promoter but not to a distal region of the PEPCK gene which does not contain a LXRE and was used as a control for specific binding in this assay (Figure 6B). In contrast, we could not detect LXR or RIP140 binding to the proximal promoter region of the G6Pase gene which has been shown to bind nuclear receptors such as HNF4 (31). HNF4 binds to both genes as shown before and was used as positive control for the ChIP assay (Figure 6B). These findings demonstrate that RIP140 is directly involved in the repression of PEPCK by LXR and suggest that different mechanisms are involved in the repression of G6Pase and PEPCK expression by LXR in these cells.
An inspection of the PEPCK gene promoter sequence revealed a putative LXRE, which is very similar to known response elements for LXR (3). This sequence is relatively well conserved between species (Figure 7A). To test if LXR can bind to this element in the PEPCK promoter we incubated biotin-labelled PEPCK probe with RXRα and LXRα protein. Gel mobility shift analysis showed that this PEPCK sequence indeed binds LXR/RXR heterodimers (Figure 7B). Competition assays using unlabeled oligonucleotides confirmed that the LXR/RXR binding was specific for the LXRE sequence in the PEPCK gene promoter (Figure 7B).
Previous studies indicate that RIP140 plays a crucial role in fat accumulation (24) by suppressing the expression of genes involved in oxidative metabolism, uncoupling and mitochondrial biogenesis in adipose tissue (25, 26). The role of RIP140 in other tissues is much less clear although the broad expression pattern suggests additional functions in transcriptional regulation (24). Gene expression profiling and the analysis of biochemical parameters in WT and RIP140 null mice led us to identify genes involved in hepatic lipogenesis as targets of RIP140. In this paper, we report that RIP140 is required for the ability of LXR to stimulate the expression of genes involved in lipogenesis. These observations prompted us to investigate whether RIP140 is also involved in other functions of LXR in liver. As a result we have identified the PEPCK gene, a rate-controlling gene in the process of gluconeogenesis as a direct target for LXR and RIP140. In contrast to the stimulation of lipogenesis RIP140 acts as a corepressor for the inhibition of PEPCK by LXR (20). Thus, we propose that the function of RIP140 as a cofactor for LXR in hepatocytes varies according to the target genes and metabolic process.
Impaired accumulation of triglycerides in RIP140 null mice is accompanied by a failure of LXR to stimulate expression of the lipogenic genes SREBP1c, ACC-1, SCD-1 and FAS. That RIP140 functions as a coactivator for LXR is supported by the failure of the LXR agonist T0901317 to stimulate the expression of these genes in cultured HuH7 cells devoid of the cofactor. Further evidence is provided by our observation that RIP140 potentiates the ability of LXR to stimulate transcription from the FAS promoter in transfected cells and by the binding of LXR and RIP140 to the LXRE in chromatin immunoprecipitation assays. The reduction of hepatic lipogenesis might contribute to the lean phenotype of the RIP140−/− mice similar to the LXR−/− mice (24, 32). LXRs are also important for the regulation of lipid metabolism in muscle and white adipose (32, 33). Future work will address a possible role of RIP140 in the function of LXR in these tissues. This will involve the treatment of isolated adipocytes with LXR agonists and the generation of tissue-specific knock-out mice for RIP140.
The agonist for LXR is required for lipogenic gene expression, both in vivo and in vitro, however its effect on the binding of LXR and RIP140 to the FAS promoter is minimal. This is consistent with our observation that endogenous LXR and RIP140 interact in a ligand independent manner and with previous findings characterising the in vitro interaction (28). In fact, RIP140 seems to bind to LXR in an atypical way involving regions of the RIP140 protein other than the NR-boxes (34). We presume that ligand binding to LXR serves another function, possibly the recruitment of other cofactors. Recent studies identified PGC-1β as a coactivator for SREBP in hepatocytes (35). It is thus conceivable that RIP140 and PGC-1β co-stimulate the expression of key genes for hepatic lipogenesis.
Although RIP140 does not seem to be involved in glucose homeostasis during fasting (data not shown) it is important for the repression of PEPCK expression by dietary cholesterol. Indeed, the repression of PEPCK by LXR depends on RIP140, both in vivo and in cultured cells. The function of RIP140 as a corepressor for LXR is also supported by chromatin immunoprecipitation assays showing the binding of both LXR and RIP140 to the PEPCK promoter. Interestingly the down-regulation of another key gluconeogenic enzyme, G6Pase, is independent of RIP140. Thus, we conclude that the ability of LXR to suppress the expression of PEPCK and G6Pase is achieved by distinct mechanisms. Indeed the indirect regulation of G6Pase gene expression by LXR has been observed (36). These findings may have implications for the development of anti-diabetic drugs. Partial agonists of LXR (37), which bypass the requirement for RIP140 should still repress the expression of G6Pase and hepatic glucose output but would not stimulate lipogenesis (38, 39). A role for LXR in the regulation of glucose metabolism is suggested by a recent report that LXR acts as a glucose sensor (40).
The function of RIP140 as a corepressor has been shown to involve the recruitment of repressive enzyme complexes such as histone deacetylases and CtBP (41, 42). At present we can only speculate about the underlying molecular mechanisms for the activation of transcription. Recently, however, Cavailles and coworkers have demonstrated that RIP140 is also capable of activating transcription from ERR target genes possibly by sequestration of repressive enzymes (43). On the other hand, it is conceivable that RIP140 is capable of binding either repressive or activating enzyme complexes, like PGC-1 (44). Thus, RIP140 bound to LXR in the liver may activate transcription from lipogenic genes and repress PEPCK gene transcription to regulate metabolic pathways and lipid homeostasis.
Generation of RIP140 knockout mice (RIP140 null) has been described (45). Mice (age matched and 3-4 months old a time of analysis) were housed under standard conditions and fed ad libitum with chow diet (SDS) or a western diet (D12079B, Research Diets) for 4 weeks. The LXR agonist T0901317 was administered by gavage feeding of 50 mg/kg/day for 3 days. Mice were sacrificed 24 h after the last treatment and tissue samples collected. All procedures were performed in accordance with the guidelines for animal care and use of the United Kingdom Home Office.
Triglyceride content in saponified, neutralized liver extracts, plasma and cell extracts was measured using the Triglyceride (GPO Trinder) Reagent (Sigma). Glucose production assay was performed as described (46).
FAS luciferase reporter constructs and RIP140 expression plasmid have been described (17, 25). LXRα expression plasmid was a gift from M. Needham, Astra Zeneca, UK. The adenoviral vector expressing siRNA for RIP140 was generated as described previously (47) and targets the sequence 5'-AGAAGATCAAGATACCTCA-3' of human, rat and mouse RIP140.
Hepatocytes were isolated by collagenase perfusion of liver from wild-type and RIP140 null mice as described (48). HuH7 and Fao cells were grown in Dulbecco's modified Eagle's medium containing 10% fetal calf serum. Cells were transfected using the FuGENE 6 reagent (Roche). Luciferase reporter activities were determined ca. 18 h after transfections using the Dual-Glo Luciferase Assay System (Promega). For adenovirus infection, cells were cultured in 6-well format and infected with the siRIP140 or a control virus (expressing GFP) at an estimated multiplicity of infection of 20 (HuH7 cells) or 50 (Fao cells) for 36 h and then treated with 1 μM T0901317 or vehicle alone for ca. 12 h and harvested for RNA isolation.
RXRα and LXRα protein (ProteinOne) and biotin labeled rat PEPCK probe (5'-CAAGAGGCGTCCCGGCCAGCCCTGTCCTTGACCCCCACCTGA-3') were incubated at RT for 20min. The competition reactions were performed by adding 10-50fold excess unlabeled double-stranded apoE ME enhancer (5'-GATCGCTGCCAGGGTCACTGGCGGTCAAAGGCAG -3'), mouse PEPCK (5'-CAAGAGGCGTCTCGGCTAGGCCTGCCCTTGACCCCCACCTGA-3'), rat PEPCK or mutated LXRE probe 5'-CAAGAGGCGCGGAGGACTGTCCTCCGACCAACCCCCACCTGA-3') oligonucleotide to the reaction mixture. The reactions were electrophoresed on a 6% precasted Tris-borate-EDTA gel at 100 V for 1 h in a 100 mM Tris-borate-EDTA buffer and transferred to a nylon membrane (Invitrogen). The biotin-labeled DNA was detected with LightShift chemiluminescent electrophoretic mobility shift assay kit (Pierce).
Antibodies against RIP140 used in this study were 6D7 (mouse monoclonal against AA301-478 of human RIP140) and a rabbit polyclonal antiserum provided by Dr. H. Chen, UCDCC, Sacramento), Western blot analysis was performed using antibodies against RIP140 (6D7), LXR (sc-1000) and CtBP (sc-5963, Santa Cruz Biotechnology).
For coimmunoprecipitations 5 mg nuclear extracts were precleared with 50 μl A/G plus agarose (Santa Cruz Biotechnology). Precleared extracts were incubated with 20 μl anti-RIP140 conjugated agarose beads (PIERCE) for 4 h with rotation and washed three times with wash buffer (50 mM TRIS-HCl (pH 8.0), 300 mM NaCl, 10% glycerol 0.1% NP40). Bound proteins were eluted with elution buffer (0.1 M glycine pH 3.2) for 10 min at room temperature. Chromatin immunoprecipitations were performed with antibodies against RIP140 (Chen), LXRα/β (sc-1000), HNF4α (sc-8987) or IgG (sc-2027, Santa Cruz Biotechnology), as described (49) using gene specific primers (table S1).
Total RNA was isolated using the TRIzol reagent (Invitrogen). RNA transcripts were quantified by real-time PCR using the Opticon 2 (MJ Research) and the SYBR Green JumpStart Taq ReadyMix system (Sigma) using gene specific primer (table S1).
Data are expressed as mean ± SEM. The statistical significance of differences between samples was determined by a two-tailed Student t test.
We thank Drs. T. Osborne, P. Tontonoz, and D. Granner for sharing their reagents and A. Soutar for a critical review of the manuscript.
This work was supported by the Wellcome Trust (061930 to B.H. and A.S.) and the BBSRC (BB/C5O4327/1 to M.H).