|Home | About | Journals | Submit | Contact Us | Français|
Expression of uncoupling protein 1 (Ucp1) mRNA is elevated in differentiated adipocytes derived from brown or white adipose tissue devoid of the nuclear receptor corepressor receptor interacting protein 140 (RIP140). Increased expression is mediated in part by the recruitment of peroxisome proliferator activated receptors α and γ, together with estrogen-related receptor α, which functions through a novel binding site on the Ucp1 enhancer. This demonstrates that regulation of Ucp1 expression in the absence of RIP140 involves derepression of at least three different nuclear receptors. The ability to increase expression of Ucp1 by β-adrenergic signaling is independent of RIP140, as shown by the action of the β3-adrenergic agonist CL 316,243 to stimulate expression in both brown and white adipocytes in the presence and absence of the corepressor. Therefore, the expression of this metabolic uncoupling protein in adipose cells is regulated by inhibition as well as activation of distinct signaling pathways.
The maintenance of energy homeostasis requires the regulated expression of gene networks that control metabolic functions in response to changing environmental conditions. A number of studies have demonstrated a fundamental role for nuclear receptors (NRs) and their ligands in the regulation of transcription of metabolic genes (1, 2). These include the peroxisome proliferator activated receptors (PPAR) (3), thyroid hormone receptors (4), the retinoid receptors RAR/RXR (5, 6), and the bile acid and oxysterol receptors FXR and LXR (7), and recent studies have identified an important role for the orphan NR estrogen-related receptor α (ERRα) (6, 8). The ability of NRs to activate gene transcription depends on the recruitment of cofactors, of which peroxisome proliferator-activated receptor-γ coactivator 1α (PGC-1α) and PGC-1β, initially identified as key coregulators of PPARγ, seem to be crucial for the activation of many metabolic genes (9, 10).
In addition to activation of gene expression, many of these same metabolic genes and gene networks are subject to repression by NRs. We found that mice devoid of the NR corepressor receptor interacting protein 140 (RIP140) are lean with reduced trigyceride stores in white adipose tissue (WAT) accompanied by an increase in expression of genes involved in several metabolic pathways that include mitochondrial biogenesis, fatty acid oxidation, and oxidative phosphorylation (11–13). One of these genes, uncoupling protein 1 (Ucp1) is of particular interest because it is normally expressed in brown adipose tissue (BAT) during adaptive thermogenesis and not in WAT and as such has been described as an important factor that defines the function of the adipose cell in terms of its ability to utilize or store energy. Interestingly, in the absence of RIP140, Ucp1 expression is elevated in adipocytes derived from WAT (14).
In BAT, Ucp1 mRNA increases rapidly after cold exposure in response to β3-adrenergic signaling. A 220-bp enhancer region has been identified in the 5′-flanking sequence of the Ucp1 gene that is conserved between species and contains a number of cis-acting elements that regulate tissue-specific expression and hormonal regulation (15–18). Activation of the β3-adrenergic receptor (β3AR) triggers the p38 MAPK signaling pathway, which results in phosphorylation of transcriptional regulators such as cAMP-responsive element binding protein (CREB) and activating transcription factor 2 (ATF2) that act directly at the Ucp1 promoter (19–23). Two CREB response elements have also been identified in the Ucp1 enhancer, the most proximal of which has been shown to be important for norepinephrine-dependent stimulation of Ucp1 gene transcription (17). Activation by β3AR may also be indirect by regulating the expression and activity of transcriptional coregulators, in particular PGC-1α (19). The 220-bp Ucp1 enhancer is also targeted by a number of NRs, including PPARs, RAR, RXR, and thyroid hormone receptors that bind to well characterized response elements (15, 20, 24–28). The PPAR response element (PPRE), which is conserved among species, has been shown to be essential for Ucp1 enhancer activity (15, 22), at least in cell culture models. PPARα and PPARγ activate Ucp1 transcription in BAT (15, 29), whereas an activated form of PPARδ stimulates expression in mice (30).
NR cofactors have also been shown to be controlling factors in adipogenesis and the function of the mature adipocyte. The relative expression levels of the p160 family members steroid receptor coactivator-1 and transcriptional intermediary factor-2 regulate the development of WAT and BAT (31), and the potential to store triglycerides in adipocytes is reduced in the absence of transcriptional intermediary factor-2 (31, 32). PGC-1α is a key transcriptional coactivator and metabolic regulator in BAT (9, 33, 34). Activation or expression of PGC-1α promotes adaptive thermogenesis in BAT, stimulates mitochondrial biogenesis, and increases oxidative metabolism in several cell types (9, 35–37). Overexpression of PGC-1α, together with treatment with ligands for PPARγ, can promote the occurrence of brown fat features in human white preadipocytes (38), demonstrating the importance of the regulation and integration of hormonal and ligand induced signaling pathways in adipocyte function.
Differential regulation of PPARγ target genes by corepressors can also selectively control transcription in adipocytes (39). In common with PGC-1α, RIP140 has been shown to interact in an agonist-dependent manner with most NRs, including all three PPAR isoforms (40–42). RIP140 null cells, when differentiated into adipocytes in vitro, also show elevated energy expenditure and express high levels of Ucp1, implying an intrinsic role of RIP140 in regulating Ucp1 expression (12, 14). In agreement with these observations, RIP140 has been demonstrated to be recruited directly to the Ucp1 enhancer element, indicating a primary role for RIP140 in the repression of Ucp1 gene transcription (14), but the identification of which receptors may be targets for RIP140 has yet to be determined.
The aims of this study are to investigate the contribution of β3AR and NR signaling to the upregulated expression of Ucp1 in RIP140 null adipocytes. By analyzing these pathways in primary cultures and cell lines, we have been able to identify specific NRs that are responsible for the activation of Ucp1 expression and that are subject to repression in white adipocytes.
Primary cultures of BAT and WAT from both wild-type and RIP140 null mice were treated with a hormonal cocktail in vitro to induce adipogenesis. Mature adipocytes were evident as judged morphologically by the accumulation of cytosolic fat droplets (Fig. 1A). This is in agreement with previous reports that adipogenesis is not dependent on the expression of RIP140 (11). Consistent with these morphological changes, the adipocyte marker gene fatty acid-binding protein aP2 was expressed in a similar manner after differentiation in both wild-type and RIP140 null cells (Fig. 1B). Ucp1 mRNA levels were quantitated and found to be higher in both BAT and WAT primary cultures after differentiation, but there was a marked additional increase in expression in the absence of RIP140. Thus, basal Ucp1 expression was increased approximately 50-fold in differentiated RIP140 null brown adipocytes compared with wild-type adipocytes, whereas that in RIP140 null white adipocytes was increased 6-fold (Fig. 1C).
In view of the importance of the β3-adrenergic signaling pathway in the induction of Ucp1 expression in BAT, we investigated the influence of RIP140 on this pathway in the BAT and WAT primary cultures. After adipocyte differentiation, cells were treated with CL 316,243, a specific β3AR agonist. Absolute levels of Ucp1 mRNA were markedly higher in brown adipocytes relative to those in white adipocytes as expected, but importantly CL 316,243 administration increased Ucp1 expression in both wild-type and RIP140 null adipocytes. The ability of CL 316,243 to induce Ucp1 expression in RIP140 null adipocytes, which is particularly evident in brown adipocytes, (Fig. 2A), indicates that β3-adrenergic signaling is maintained in the absence of RIP140. Nevertheless, it was conceivable that the up-regulation of Ucp1 in cells devoid of RIP140, in both the presence and absence of CL 316,243, might result from an increase in the activity of the β3-adrenergic signaling pathway.
We investigated the influence of RIP140 on β3-adrenergic signaling by examining the phosphorylation status of p38 MAPK and ATF2 in differentiated adipocytes before and after treatment with CL 316,243. Protein extracts from wild-type and RIP140 null adipocytes were analyzed by Western blotting. The expression of RIP140 had no effect on the phosphorylation of p38 MAPK in response to CL 316,243 administration in either brown or white adipocytes (Fig. 2B, top panels). Phosphorylated ATF2 levels were also unaffected by RIP140 expression in white adipocytes (Fig. 2B, bottom panels), but total levels of both p38 and ATF2 were slightly elevated in the absence of RIP140. In contrast, in brown adipocytes devoid of RIP140, there was an increase in phosphorylated ATF2, although total ATF2 levels were reduced (Fig. 2B, bottom panels). Given that the phosphorylation of p38 MAPK and ATF2 is unchanged in white adipocytes in the absence of RIP140, we conclude that increased expression of Ucp1 in these cells does not result from up-regulation of β3-adrenergic signaling. In brown adipocytes, however, the increase in phosphorylated ATF2 levels in the absence of RIP140 may contribute to the increased Ucp1 expression. To further investigate the activation of the β3-adrenergic pathway, we examined the expression of the NR cofactor PGC-1α and the β3AR. PGC-1α expression was increased in both wild-type and RIP140 null adipocytes in response to CL 316,243 treatment, consistent with the maintenance of intact β-adrenergic signaling in the absence of RIP140. However, both basal and stimulated levels of PGC-1α were elevated in RIP140 null adipocytes derived from BAT (Fig. 2C), and this may contribute to the higher levels of Ucp1 mRNA in these cells. Expression of β3AR was unaltered in the absence of RIP140 with higher levels in brown adipocytes, as expected (Fig. 2D). Therefore, we conclude from these studies that β3-adrenergic signaling is capable of stimulating Ucp1 gene expression and that this is maintained in the absence of RIP140. However, in brown adipocytes, increased Ucp1 expression may in part result from indirect effects of the repressor acting on the expression of additional transcriptional regulators.
To investigate the function of different PPAR isoforms in the absence of RIP140 repression, the levels of expression of PPARα, PPARγ, and PPARδ were determined in wild-type and RIP140 null adipocytes before and after differentiation. The levels of PPARα, which increased during differentiation, were greater in the absence of RIP140, particularly in brown adipocytes, in which its expression increased approximately 20-fold (Fig. 3). The expression of PPARγ, which also increased during adipogenesis, was unaffected by RIP140 expression. PPARδ expression was unchanged during adipocyte differentiation and unaffected by RIP140 expression. Therefore, because PPARα is a key activator of Ucp1 expression (29), the elevated basal levels of Ucp1 in the absence of RIP140 may result from the increased expression of this isoform.
In view of these changes in expression, we exploited a pharmacological approach together with chromatin immunoprecipitation assays to determine which PPAR isoforms may be responsible for the increased Ucp1 gene expression. To perform these experiments, we used an RIP140 null adipocyte cell line (RIPKO-1) that expresses all three PPAR isoforms and expresses Ucp1 after adipocyte differentiation and an immortalized line of wild-type preadipocytes derived from the Immortomouse mouse line (43). RIPKO-1 cells were induced to undergo adipocyte differentiation for 7 d, treated with specific agonists for PPARα, PPARδ, or PPARγ for an additional 3 d, and then levels of Ucp1 mRNA were determined. Addition of specific agonists to PPARα (GW7647) and PPARγ (roziglitazone) induced Ucp1 expression between 4- and 5-fold relative to no agonist treatment (Fig. 4A), but a PPARδ-specific agonist (GW1516) had no effect. No increase in expression was observed on treatment with T3 alone (data not shown). Additional experiments were performed in the presence of a pan antagonist that blocks the actions of both PPARα and PPARγ (GW496471) to minimize the contribution of potential endogenous ligands. A concentration of 10−6 m GW496471 was sufficient to prevent the increase in Ucp1 expression observed between d 7 and 10 (Fig. 4). In RIP140 null cells, the reduction in Ucp1 expression observed in the presence of the pan antagonist was reversed by treatment with both PPARα- and PPARγ-specific ligands (Fig. 4A), confirming that both PPARα and PPARγ are capable of inducing Ucp1 expression in RIPKO-1 cells. To further address the role of RIP140 in the regulation of Ucp1 expression by these receptors and to minimize the affects of altered PPARα and PPARγ expression in the null cell line, differentiated wild-type adipocytes were treated with adenoviral vectors expressing short hairpin (sh) RNA to RIP140 to reduce RIP140 expression (supplemental Fig. 1, published as supplemental data on The Endocrine Society's Journals Online web site at http://mend.endojournals.org). A decrease in the level of RIP140 in wild-type adipocytes results in a significant agonist dependent up-regulation of Ucp1 expression relative to an sh random control sequence (Fig. 4B), whereas the levels of expression of both PPARα and PPARγ are relatively unaltered (supplemental Fig. 2, published as supplemental data on The Endocrine Society's Journals Online web site at http://mend.endojournals.org). Therefore, a transient decline in RIP140 results in enhanced ligand-dependent activation of both PPARα and PPARγ and increased expression of the Ucp1 target gene.
The up-regulation of Ucp1 gene expression by both PPARα and PPARγ in RIP140 null cells is supported by chromatin immunoprecipitation assays. In agreement with the studies in primary white and brown adipocytes both PPARα and PPARγ mRNA levels were induced in a similar manner on differentiation of RIPKO-1 cells (supplemental Fig. 3A, published as supplemental data on The Endocrine Society's Journals Online web site at http://mend.endojournals.org), and Western blot analysis showed an induction of PPARα and PPARγ protein (supplemental Fig. 3B). Cross-linked protein-DNA extracts prepared from undifferentiated and differentiated RIPKO-1 adipocytes were immunoprecipitated with antibodies specific for PPARα and PPARγ and analyzed by PCR using primers flanking a PPRE in the perilipin gene promoter, which has been shown to bind both of these receptor isoforms (44), the Ucp1 enhancer element and a negative control region 15 kb upstream of the Ucp1 gene. In undifferentiated RIPKO-1 cells, no enrichment of the perilipin response element or the Ucp1 enhancer element sequences were observed with antibodies specific for PPARα and PPARγ (Fig. 5A). Binding of both PPAR isoforms was detected on the perilipin PPRE and the Ucp1 enhancer element in differentiated RIPKO-1 cells but not the negative control region (Fig. 5B).
During the course of analyzing the recruitment of PPARs to the Ucp1 enhancer, we noted potential binding sites for the orphan NR ERRα (supplemental Fig. 4, published as supplemental data on The Endocrine Society's Journals Online web site at http://mend.endojournals.org). In addition to PPARα and PPARγ, we therefore analyzed the expression of ERRα in RIPKO-1 adipocytes and found that mRNA and protein levels were increased during adipocyte differentiation (Fig. 6, A and B). In addition, the loss of RIP140 expression results in only a small 1.4-fold increase in ERRα expression in primary adipocytes (supplemental Fig. 5, published as supplemental data on The Endocrine Society's Journals Online web site at http://mend.endojournals.org). Chromatin immunoprecipitation assays using a specific ERRα antibody demonstrated that the Ucp1 enhancer element but not a control region 15 kb upstream of the mouse Ucp1 gene was precipitated from differentiated RIPKO-1 adipocytes, indicating a direct association of ERRα with the Ucp1 enhancer (Fig. 6C). To investigate a functional role for ERRα on the Ucp1 enhancer, a luciferase reporter gene under the control of the 220-bp enhancer element was stably incorporated into RIPKO-1 adipocytes. The RIP140 null reporter cell line was treated for 10 d with standard differentiation conditions, and, after adipogenesis, the cells were maintained in the presence of increasing concentrations of the ERRα antagonist XCT790 for 48 h. This resulted in a dose-dependent decrease in expression of the reporter gene, indicating a functional role for ERRα on the enhancer element (Fig. 6D). To further characterize the combined effect of ERRα and PPARs, RIPKO-1 adipocytes were differentiated as described previously and treated with both the PPARα/PPARγ antagonist as well as the ERRα antagonist. Antagonist treatment of the differentiated adipocytes resulted in reduction of Ucp1 expression to control levels, indicating that, in these cells lacking RIP140, the elevated expression of Ucp1 is mediated by the combined activity of at least three different NRs (Fig. 6E).
To locate the specific regions required for binding of ERRα to the Ucp1 enhancer, Cos-7 cells were transiently transfected with a luciferase reporter gene under the control of either the wild-type 4-kb Ucp1 promoter or a construct with the 220-bp enhancer region deleted (4kbΔEnh) (Fig. 7A). Although expression of Ucp1 is restricted to a limited number of cell types and the enhancer is complex in nature and regulated by a number of adipocyte-specific factors, coexpression of ERRα and PGC-1α in Cos-7 cells resulted in an 80-fold induction from the wild-type Ucp1 promoter, and this was significantly reduced by deletion of the enhancer element. There was a small increase in promoter activity after expression of ERRα or PGC-1α alone, but this was also reduced in the absence of the enhancer. We next tested whether RIP140 was able to repress the transcriptional activity of exogenously expressed ERRα/PGC-1α. Cos-7 cells were transiently transfected with a luciferase reporter gene containing 220-bp enhancer element linked to the thymidine kinase promoter from −105 to +50. As with the full-length Ucp1 promoter, the activity of this reporter was not induced in the presence of ERRα or PGC-1α alone. Cotransfection of ERRα and PGC-1α together induced the activity of the reporter gene by 8-fold, but this was abrogated by coexpression of RIP140 in a dose-dependent manner (Fig. 7B). In addition, expression of an RIP140-specific small interference (si) RNA increased the ERRα/PGC-1α effect on the Ucp1 reporter gene by more than 2-fold. Therefore, increased expression of RIP140 results in repression, whereas reduction in the levels of RIP140 results in an increase in the transcriptional activation by ERRα/PGC-1α from the Ucp1 enhancer.
Mutational analysis and EMSA were used to identify the role of putative ERR response element (ERRE) binding sites in the Ucp1 enhancer. Potential ERR half-sites were noted at position −2396 and at −2550, overlapping the PPRE/DR1 (direct repeat 1) (supplemental Fig. 3), but only mutation of the downstream site abolished DNA binding activity in EMSA (data not shown). To determine whether this ERRE at −2396 was required for Ucp1 promoter activation by ERRα/PGC-1, Cos-7 cells were transiently transfected with a luciferase reporter gene containing a wild-type enhancer element (UCP1–220bp) or an enhancer element with a mutation in the ERRE site (UCP1–220bp-ERREmut). Expression of ERRα in the presence of either PGC-1α or PGC-1β induced the activity of the wild-type Ucp1 enhancer reporter by 6- to 7-fold, and this induction was abrogated by mutation of the ERRE (Fig. 7D), demonstrating that a single ERRE in the Ucp1 enhancer is sufficient for this activation.
RIP140 plays an essential role in energy homeostasis by repressing metabolic genes that are involved in energy expenditure. We focused on UCP1 expression to investigate the influence of RIP140 on β3AR and NR signaling pathways in adipocytes. Hormonal control of Ucp1 expression in BAT during adaptive thermogenesis has been well characterized (17, 19, 24, 26, 27), and recent studies have shown that this is a key target gene of RIP140 in adipose cells (11, 14). Analysis of primary cultures indicates that, although the expression levels of Ucp1 are markedly higher in brown adipocytes compared with white adipocytes, the absence of RIP140 leads to a marked increase in expression in both types. Thus, we conclude that RIP140 plays an intrinsic role as a corepressor in both brown and white adipose cells.
We previously demonstrated that the ability of RIP140 to repress Ucp1 transcription is dependent on its recruitment to the upstream enhancer element (12). The enhancer, however, contains a number of binding sites for NRs and members of the ATF/CREB family, but their potential contribution to Ucp1 up-regulation in the absence of RIP140 was not determined. By using a combination of specific ligands and chromatin immunoprecipitation experiments, we propose that the absence of RIP140 leads to the recruitment of PPARα, PPARγ, and ERRα to the UCP1 enhancer to allow activation of Ucp1 gene transcription. Although an activated version of PPARδ in WAT results in mice with a similar phenotype to that of RIP140 null mice, including the enhanced expression of Ucp1 (30), the inability of a PPARδ ligand to stimulate Ucp1 expression in RIPKO-1 cells suggests that this isoform may not be primarily responsible for the increased levels observed in RIP140 null adipocytes.
In vivo, Ucp1 expression is regulated by β-adrenergic signaling under the control of the sympathetic nervous system in response to environmental stimuli (19, 24, 45). This is accomplished by activation of the p38 MAPK pathways, which leads to activation of transcription factors such as ATF2 and CREB and increased expression and phosphorylation of the NR cofactor PGC-1α. We found that RIP140 null adipocytes isolated from both BAT and WAT retain the capacity to respond to β-adrenergic stimulation as shown by treatment with a specific β3-adrenergic agonist, which results in activation of p38 MAPK and ATF2 and an increase in Ucp1 mRNA. Nevertheless, there is increased PGC-1α expression and phosphorylated ATF2 levels observed in brown adipocytes devoid of RIP140, suggesting that the corepressor may play a role in suppressing the β3-adrenergic signaling pathway. In contrast, given that the absence of RIP140 in white adipocytes does not affect β3-adrenergic signaling, this pathway does not seem to be involved in the up-regulation of Ucp1 observed in white adipocytes.
ERRα-mediated gene regulation plays important roles in the control of energy balance by regulating fatty acid oxidation (8, 46). We have shown that full transcriptional activation of Ucp1 promoter by ERRα/PGC-1α requires the Ucp1 enhancer element, and, using site-directed mutagenesis, we identified an ERRα binding site in this element that is required for Ucp1 induction by ERRα/PGC1α. The Ucp1 enhancer has been found previously to be a target for repression by RIP140 (14). In this study, we demonstrated that RIP140 is able to repress ERRα/PGC1α transcriptional activity, suggesting that both ERRα and RIP140 are involved in Ucp1 transcriptional regulation. The finding that ERRα, PPARα, and PPARγ are each recruited directly to this element in differentiated adipocytes together with the previous observation that RIP140 is present on the enhancer element in cells in which Ucp1 mRNA expression is abrogated implies that, in mature adipocytes, RIP140 may regulate Ucp1 transcription by targeting one or more of these receptors. In addition, the demonstration that the increase in Ucp1 expression after differentiation can be almost completely prevented by combined PPARα/PPARγ and ERRα antagonist treatment indicates a key role for all of these NRs in the regulation of Ucp1 expression in RIP140 null adipocytes in the absence of β3-adrenergic signaling.
The ability of RIP140 to modulate the action of ERRα may account in part for the increased UCP1 expression found in WAT in ERRα null mice (6) in which the potential for repression by cofactors such as RIP140 would be reduced. Recent studies, however, in ERRα null mice have also demonstrated that this NR is not required for the regulation of Ucp1 expression in the process of adaptive thermogenesis (47). The data described in this study extend these observations by the identification of an ERRE and the finding that the basal expression of a target gene such as Ucp1 may be regulated by repression of different signaling pathways.
In summary, uncoupling proteins such as UCP1 provide an important mechanism for energy dissipation by facilitating the process of thermogenesis as well as a means for maintaining redox balance and reducing the generation of reactive oxygen species. A number of studies have demonstrated that Ucp1 expression is regulated by both NRs and alterations in the levels of intracellular cAMP, with coregulators such as PGC-1α acting as key factors required for the integration of different signaling pathways. This study identifies an additional mechanism for regulating Ucp1 expression and describes a role for RIP140 in determining the basal level of Ucp1 transcription, which in differentiated adipocytes may be mediated by at least three different NRs, ERRα, PPARα, and PPARγ. Therefore, alterations in the level of expression of RIP140 or in the recruitment of RIP140 to these specific receptors may provide a mechanism to control processes that determine energy balance in adipose cells.
The generation of RIP140 null mice has been described previously (48). Mice used in this study were backcrossed six generations to a C57BL/6J background, maintained under standard conditions with controlled light and temperature, and fed a chow diet ad libitum. All experiments were performed according to United Kingdom Home Office guidelines and ethical approval.
Primary brown and white cultures were prepared from inter-scapular and inguinal fat depots, respectively, as described previously (14). Brown preadipocytes were induced to differentiate in DMEM-F12 medium supplemented with 33 μm biotin, 17 μm calcium pantothenate, 10% fetal bovine serum, 100 U/ml penicillin,100 μg/ml streptomycin, and 0.25 μg/ml amphotericin B (Invitrogen, Paisley, UK) supplemented with 170 nm insulin, 1 nm T3, 250 nm dexamethasone, 500 μm isobutylmethylxanthine, and 0.125 mm indomethacin for 2 d (Invitrogen). After induction, the cells were fed every 2 d with maintenance medium supplemented with 170 nm insulin and 1 nm T3. White preadipocytes were differentiated as described previously (49). Differentiated white and brown adipocytes were pretreated with 0.1 μm propranolol (5 min), followed by 10 μm CL 316,243 as indicated. Differentiation of RIPKO-1 cells was performed as described previously (49) in the presence of 2.5 μm roziglitazone, unless indicated. In experiments using the PPARα/PPARγ antagonist GW496471, preliminary transient transfection experiments with a PPAR-responsive reporter gene were used to determine the concentrations of PPARα and PPARγ agonists necessary to overcome the inhibitory effects of the pan PPARα/PPARγ antagonist. In the presence of the PPARα/PPARγ antagonist at 10−6 m, treatment with PPARγ agonist at 10−5 m or a PPARα agonist at 10−6 m was sufficient to restore activation (data not shown). The reporter cell line expressing luciferase under the control of the Ucp1 220-bp enhancer was generated by insertion of the reporter and enhancer into the pLenti6 vector (Invitrogen) from which the cytomegalovirus promoter was removed, and the construct was stably introduced by lentiviral infection of RIPKO-1 cells as described previously (14). The immortalized adipocyte cell line was derived from the H-2Kb-tsA58 transgenic mouse using the stromal vascular fraction of WAT and cultured according to published procedures. The sequences used for depletion of RIP140 using adenoviral vectors are as follows: siRIP, GATCCCCAGAAGATCAAGATACCTCATTCAAGAGATGAGGTATCTTGATCTTCTTTTTTA; siRandom (siRAND),GATCCCCGACGTTAGCAATCGAGCTCTTCAAGAGAGAGCTCGATTGCTAACGTCTTTTTA.
In studies using adenoviral infection, immortalized white preadipocytes were differentiated using a standard cocktail for 48 h as described previously and maintained in standard media containing insulin alone for 7 d. Viral infection was performed in serum-free conditions for 6 h, and cells were then maintained in standard conditions with insulin alone for an additional 48 h before the addition of specific ligands.
Total RNA was isolated from cell lines, primary cultures, and tissue using TRIzol (Invitrogen) according to the instructions of the manufacturer. cDNA was prepared as described previously (11). RIP140, L19, and Ucp1 gene expression levels were determined using specific primers and TaqMan probes. Expression levels of all other genes were determined with SYBR green reagent by using specific primers. Expression levels for all genes were correlated to that for the L19 ribosomal coding gene. Primer sequences may be obtained on request.
Cells were incubated in 1% formaldehyde in DMEM for 15min at 37 C. Cross-linked cells were lysed, sonicated, and immunoprecipitated with protein A/G PLUS-agarose (SC-2003; Santa Cruz Biotechnology, Santa Cruz, CA) according to the instructions of the manufacturer using rabbit-polyclonal antimouse PPARα (SC-9000; Santa Cruz Biotechnology), PPARγ (SC-7196; Santa Cruz Biotechnology), ERRα (LS-A5402; Lifespan, Seattle, WA) antibody or control normal rabbit IgG (SC-2027; Santa Cruz Biotechnology). DNA fragments were purified with a QIAquick PCR purification kit (Qiagen, Valencia, CA) and used as templates in PCRs. The primers used for the Ucp1 enhancer were 5′-AGCTTGCTGTCACTCCTCTACA-3′ and 5′-TGAGGAAAGGGTTGACCTTG-3′, and those for the upstream control region were 5′-GCTTGGGTCCACCTAGAATCAC-3′ and 5′-CCTCCAGGTCAAACTGATCTAGACA-3′. Primers for the perilipin promoter were 5′-GAGTGGTCAAGACCTCTGCTC-3′ and 5′-GCTCTGCTGACAAACCGGTC-3′. Sequences of all additional oligonucleotides used are available on request.
Whole-cell lysates were prepared by rinsing cells with cold PBS, followed by addition of 2× Laemmli sample buffer. Equal amounts of proteins were separated on 10% SDS-polyacrylamide gel and transferred to polyvinylidene difluoride membrane (Hybond P; GE Healthcare, Little Chalfont, UK). The levels of phosphorylated (9211S) and total (9212) p38 MAPK and phosphorylated (9225S) and total (9222) ATF2 were detected with a 1:1000 dilution of each specific antisera (New England Biolabs, Beverly, MA). Rabbit polyclonal antimouse PPARα (SC-9000) and PPARγ (SC-7196) antibodies were diluted 1:1000. Immunoreactive bands were visualized using secondary peroxidase-conjugated antibody and enhanced chemiluminescence.
Ucp1 promoter reporter constructs were generated as described previously (14). Cos-7 cells were transfected in 96-well plates with 20 ng reporter gene and 5 ng pRL-cytomegalovirus, in the absence or presence of ERRα (20 ng), PGC-1α (10 ng), PGC-1β (10 ng), pSUPER-scrambled (20 ng), or pSUPER-siRIP140 (20 ng) using Fugene6 (Roche, Indianapolis, IN), according to the instructions of the manufacturer. Cells were harvested for luciferase assay 48 h after transfection, and renilla luciferase activity was used to correct for differences in transfection efficiencies.
Disclosure Statement: The authors have nothing to disclose.