HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 41, 31722-31732, October 13, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/41/31722    most recent M001678200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by del Mar Gonzalez-Barroso, M. Articles by Cassard-Doulcier, A.-M. Search for Related Content PubMed PubMed Citation Articles by del Mar Gonzalez-Barroso, M. Articles by Cassard-Doulcier, A.-M. Social Bookmarking                      What's this?

Transcriptional Activation of the Human ucp1 Gene in a Rodent Cell Line

SYNERGISM OF RETINOIDS, ISOPROTERENOL, AND THIAZOLIDINEDIONE IS MEDIATED BY A MULTIPARTITE RESPONSE ELEMENT^*

Maria del Mar Gonzalez-Barroso, Claire Pecqueur^§, Chantal Gelly, Daniel Sanchis, Marie-Clotilde Alves-Guerra, Frederic Bouillaud, Daniel Ricquier, and Anne-Marie Cassard-Doulcier^¶

From the Centre de Recherches sur l'Endocrinologie Moléculaire et le Développement, CNRS, 92190 Meudon, France

Received for publication, March 1, 2000, and in revised form, July 18, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Uncoupling protein 1 (UCP1) is uniquely expressed in brown adipocytes and generates heat production by uncoupling respiration from ATP synthesis. The activatory effects of norepinephrine and retinoic acid (RA) on rodent ucp1 gene transcription have been well characterized. These effects are mediated by a 211-base pair (bp) enhancer which is also sufficient to restrict expression to brown adipose tissue. The molecular mechanisms controlling the transcription of the human ucp1 gene are unknown. In order to study the transcriptional regulation of the human gene, we set up chloramphenicol acetyltransferase constructs containing the entire or deleted 5' regions upstream of the transcriptional start site of the gene. These constructs were transiently transfected in a mouse cell line. A 350-bp hormone response region showing a significant homology with the rat ucp1 enhancer and located between the BclI polymorphic site and an AatII site (bp 3820/3470) was detected. This region was sufficient to mediate the stimulation by RA and by combined treatments (RA + isoproterenol (ISO), RA + thiazolidinedione (TZD), or RA + ISO + TZD). The highest stimulation, a 26-fold increase in basal activity, was obtained by RA + ISO + TZD treatment. In contrast to the rodent gene, under our conditions, the effect of ISO and/or TZD is dependent on RA stimulation. Analysis of 105 bp inside the 350-bp element by site-directed mutagenesis and gel retardation experiments demonstrated that a multipartite response element mediates the drug stimulation. This region binds RARs and RXRs nuclear factors, CREB/ATF factors, and also PPAR despite the absence of a consensus peroxisome-proliferator response element. The activation of the human ucp1 gene transcription by certain hormones or drugs, and the identification of the cis-elements involved, will help to identify new compounds activating fat oxidation and energy expenditure in humans.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Uncoupling protein 1 (UCP1)¹ is uniquely expressed in brown adipose tissue (BAT) and is located in the mitochondrial inner membrane of adipocytes. UCP1 stimulates heat production by uncoupling oxidative phosphorylation (1). The amount of UCP1 determines the thermogenic potential of BAT. In rodents it has been demonstrated that this tissue can dissipate energy as heat in response to cold exposure or to excessive ingestion of calories (2). Recent results obtained with transgenic mice after genetic BAT ablation showed that these mice undergo excessive weight gain (3). However, disruption of the mouse ucp1 gene demonstrated that the main role of UCP1 was to maintain body temperature in a cold environment (4).

The main control of UCP1 biosynthesis is situated at the transcriptional level (5, 6). An increase in ucp1 gene transcription and ucp1 mRNA is observed within minutes of cold exposure or exogenous catecholamine administration (7, 8). BAT sympathetic stimulation, via norepinephrine and cAMP, is the primary signal in activation of ucp1 expression in rodents, but other factors, such as thyroid hormones, are critical for a full physiological response (see Ref. 9 for review). It has also been reported that retinoids (10, 11) and thiazolidinediones (TZD) (12, 13) increase rodent ucp1 expression.

Over the last few years, the molecular mechanisms involved in the regulation of rodent ucp1 gene expression have been partially elucidated. A 211-bp enhancer located at positions bp 2494 to bp 2283 upstream of the transcription start site in the rat ucp1 gene has been described (14). Using transgenic mice it has been concluded that the tissue specificity of the ucp1 expression in the rat is only due to the 211-bp enhancer (15). In this enhancer, the presence of a ucp1-gene activating region (UAR) is absolutely required for norepinephrine and RA stimulation, but to achieve RA responsiveness 92-bp containing UAR and forming the 5' moiety of the 211-bp enhancer are necessary (16, 17). Studies with the mouse gene have revealed a similar enhancer between bp 2530 and bp 2310 (18). Two elements essential for mouse enhancer function (CRE-2/BRE-1) have been identified downstream of the rat UAR, in the equivalent 92-bp region (18). By using in vitro analysis, a PPRE element binding PPAR-RXR heterodimers was described (19). This element is located in the equivalent 92-bp region of the mouse gene. It has previously been demonstrated that PPAR plays an important role in processing preadipocyte differentiation (20).

In humans, the functional role of BAT is still under discussion. BAT is detected in newborns but its thermogenic activity decreases rapidly after birth (21). In young humans, BAT seems to be converted into white adipose tissue as has been described for ruminants (22, 23). However, it is possible to detect BAT in certain conditions: in outdoor workers exposed to cold conditions (24) or in several pathological situations such as pheochromocytoma or hibernoma (21, 25, 26). UCP1 and ucp1 mRNA were detected by Western and Northern analysis in the perirenal fat of many patients (27). More recently, ucp1 mRNA was detected by reverse transcriptase-polymerase chain reaction in the adipose deposits of adults (28).

We previously cloned and sequenced the human ucp1 gene (29). A polymorphic BclI site was described in the upstream region of the transcriptional unit at position bp 3826 (30, 31). This polymorphism was associated with percentage fat gain over time, suggesting a role for the ucp1 gene in regulation of fat content in humans.

The aim of our study was to investigate the regulation of the human ucp1 gene expression. In this report, we describe that the transcription of the human ucp1 gene can be strongly activated by hormones and drugs. An enhancer, which is partly homologous to the rat and the mouse ucp1 enhancer, controls these effects. This region is located close to the polymorphic BclI site. Although the human enhancer is in part homologous to rodent enhancers, molecular mechanisms implicated in the stimulation by drugs are different. In particular, the -adrenergic and the TZD stimulations have an effect only in the presence of RA. It is shown that a multipartite element mediates the stimulation of the human ucp1 gene by drugs. These studies will facilitate the search for compounds activating substrate oxidation and thermogenesis in humans.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- DNA restriction, modification enzymes, and oligonucleotides were from Eurogentec (Seraing, Belgium) or New England Biolabs (Ozyme, Montigny-le-Bretonneux, France). ³²P-Labeled nucleotide triphosphates were from Amersham Pharmacia Biotech (Les Ulis, France). Fetal calf serum, LipofectAMINE, and culture products were from Life Technologies, Inc. (Cergy-Pontoise, France). Antibodies against ATF and PPAR were from Santa Cruz (Tebu, Le perray-en-Yvelines, France). The Gene-Editor Mutagenesis kit was from Promega Biotech (Charbonnieres, France), and the QuickChange Site-directed Mutagenesis kit from Stratagene (Ozyme, Saint-Quentin en Yvelines, France). Columns for plasmid purification were from Qiagen (Courtaboeuf, France). Other chemical products were from Sigma Aldrich (Saint-Quentin Fallavier, France). The sequence homologies of enhancers were analyzed using the DNA strider program (version 1.3). Polymerase chain reaction was done in Gene-amp-6000 from PerkinElmer Life Science (Courtaboeuf, France) and sequences were analyzed using the prism cyclic sequencing kits and an ABI 373 DNA sequencer.

Plasmids and Site-directed Mutagenesis-- The human ucp1 6300 bp chloramphenicol acetyltransferase reporter plasmid (6300CAT) was constructed by ligating a BglII-KpnI fragment (bp 6233 to +91) from a 7-kb PstI fragment of the human ucp1 gene in the SalI blunt site of PBLCAT6 (32). Different restriction sites were used to generate several deleted 6300CAT constructs: 3820CAT plasmid (deletion BglII-BclI) and 2790CAT plasmid (deletion BglII-DraIII). After restriction digestion, plasmids were blunt-ended using Klenow polymerase and ligated. Internal deletions were generated from the 6300CAT construct. Three deletions were made after insertion of a BclI site by mutation at the required nucleotide: 6300CAT del 350 plasmid (deletion from bp 3820 to bp 3470), 6300CAT del 90 plasmid (deletion from bp 3820 to bp 3762), 6300CAT del 60 plasmid (deletion from bp 3762 to bp 3705). The 1-kb TK-CAT plasmid was constructed by insertion of the BclI/DraIII fragment of the human region as a blunt fragment in the repaired SalI site of PBLCAT5 (32) in both orientations: 3820/2790TK and 2790/3820TK. Successive deletions were generated using these constructions. After restriction digestion, plasmids were blunt-ended by Klenow polymerase and ligated: 3328/2790TK (digestion by HindIII of the 3820/2790TK), 3328/3820TK (digestion by HindIII of the 2790/3820TK), 3820/3470TK named 350-TK-CAT (digestion by XbaI and AatII of the 3820/2790TK), and 3470/3820TK (digestion by AatII and HindIII of the 2790/3820TK). The Bsu36I linearized 350-TK-CAT plasmid was used to perform constructs by using the Bal-31 exonuclease. After blunt endings, fragments were ligated, transformed, and analyzed. Three plasmids were selected: 350bal 51 (bp 3820/3622 to bp 3527/2790), 350bal1 (bp 3820/3682 to bp 3600/2790), 350bal15 (bp 3820/3716 to bp 3602/2790). We also generated two deleted plasmids from the 350TK-CAT plasmid: 350TK del60 and 350TK del47. They were created by insertion of a BclI site as described previously on the 6300CAT and after deletion from bp 3762 to 3705 and bp 3734 to 3684, respectively. The 105-TK-CAT plasmid was constructed by insertion of a 105-bp double strand oligonucleotide (bp 3743/3636) in the blunt-ended SalI site of PBLCAT5 in the two orientations. Mutations in the 105-TK-CAT construct were inserted by using the Promega Gene-Editor kit, except for the E105TK mutant where the QuickChange Site-directed Mutagenesis Kit was used. Deletions and mutations of all clones were confirmed by DNA sequencing.

Cell Culture, Transfections, and CAT Assays-- An immortalized mouse adipocyte cell line expressing the ucp1 gene, termed 1B8, was used (17). 1B8 cells were grown in standard adipocyte culture medium supplemented with 10% fetal calf serum. After cells reached confluence, fetal calf serum content was reduced to 5% and supplemented with 1 nM T3 and 20 nM insulin until complete differentiation about 1 week later (33). 1B8 cells were transiently transfected in suspension by the calcium phosphate precipitation method as described previously (14, 17). In all transfection experiments, 2 µg of plasmid expressing -galactosidase was included to assess the efficiency of separate transfections. 20 µg of the reporter CAT plasmid was used. After transfection, cells were treated or not with drugs in medium supplemented with 5% charcoal fetal calf serum. After 20 h of treatment, the cells were collected by scraping and -galactosidase and CAT assays were carried out (34). Each transfection was performed a minimum of 3 times, and at least 2 different preparations of each DNA construct were used. The CAT activity was normalized for variation in transfection efficiency using the -galactosidase activity as standard. Drug concentrations for treatments were 10⁶ M for thiazolidinedione BRL 49653, all-trans-RA or 9-cis-RA, isoproterenol, and norepinephrine, and 10³ M for dibutyryl-cAMP.

Nuclear Extracts and Electrophoretic Mobility Shift Assays-- RA receptor (RAR, RAR, RAR, and RXR) and PPAR were overexpressed in COS-1 cells. COS-1 cells were transfected with 0.4 µg of PSG5 expression vector containing RAR, RAR, RAR, RXR, or PPAR cDNAs by the LipofectAMINE method. The total amount of DNA was adjusted to 2 µg in each transfection. After transfection, cells were grown for 20 h in Dulbecco's modified Eagle's medium supplemented with 5% fetal calf serum. Cells were harvested in TEN buffer (40 mM Tris-HCl, pH 7.5, 1 mM EDTA, 150 mM NaCl). After centrifugation, the pellet was resuspended in L buffer (50 mM Tris-HCl, pH 7.9, 500 mM KCl, 20% glycerol, 0.5 mM EDTA, 0.1% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, and 2.5 µg/ml leupeptin). Cells were lysed by three cycles of freeze-thawing. After centrifugation the supernatant was aliquoted and frozen in liquid nitrogen and stored at 80 °C.

Nuclear proteins were isolated from 1B8 cells according to a Shapiro modified preparation (35). For electrophoretic mobility shift assays, oligonucleotides were end-labeled using -dATP and Klenow polymerase enzyme and purified on spin columns. The reaction buffer contained 20 mM Hepes, 150 mM NaCl, 5% glycerol, and 0.1% Nonidet P-40. 1 µg of poly(dI-dC) was included in each reaction mixture as a nonspecific competitor. The final reaction volume was 20 µl. Nuclear extracts were incubated at 22 °C for 20 min with dI-dC in the reaction buffer before the addition of 0.3 ng of a ³²P-labeled double-stranded DNA probe (30000 cpm) and followed by a second 20-min incubation in the presence of the probe. In the competition experiments, a 25-, 50-, or 100-fold molar excess of unlabeled oligonucleotide was included in each binding reaction. When antiserum was used, binding reactions were incubated with antiserum for 20 min at 22 °C prior to the addition of the probe.

Oligonucleotides-- The sequences of double-stranded oligonucleotides were as follows from the 5' to 3' extremities: 43 (agctAAGGGTCAGTTGCCCTTGCTCATACTGACCTATTCTTTACCTC), 43mutB (agctAACCCTCTGTTGCCCTTGCTCATACTGACCTATTCTTTACCTC), 43mutC (agctAAGGGTCAGTTGCCCTTTTTCTTACTGACCTATTCTTTACCTC), 43mutE (agctAAGGGTCAGTGACTAGTGCTCATACTGACCTATTCTTTACCTC), 43mutD (agctAAGGGTCAGTTGCCCTTGCTCATACGACTAGATTCTTTACCTC), 43mutB/C (agctAACCCTCTGTTGCCCTTTTTCTTACTGACCTATTCTTTACCTC), RA/CREB (aaTTGCTACGTCATAAAGGGTCAGTTGCCCTTGCTCATACTG), CREB consensus (aattCCAGGGCTTTGGGAGTGACGCGCGTCTG), DR1 (aattCTGTGACCTCTGACCTAG).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Identification of a 350-bp Enhancer Essential for Drug Stimulation-- A 2-kb upstream region of the human ucp1 gene has been cloned and sequenced (29). However, this region was not able to direct specific expression of ucp1 gene in brown adipose cells.² A 7-kb fragment of the 5' region of the human ucp1 gene was cloned and entirely sequenced. The start site has not yet been well defined due to difficulties in obtaining high quality poly(A)⁺ mRNA from human brown adipose tissue and the first T of the TATA box was defined as the +1 (29). In order to analyze the upstream region, a plasmid containing the entire 5' region (bp 6233 to +91) upstream of the bacterial cat gene was constructed. This plasmid was referred to as 6300CAT construct and was used in transient transfection experiments of mouse 1B8 cells. After 20 h of incubation in the presence or absence of drugs, CAT activity was determined in cell lysates and the values were normalized for differences in transfection efficiency. Treatment with retinoids (all-trans-RA or 9-cis-RA) caused a slight increase in the basal activity (Fig. 1). No stimulation by only -adrenergic agonists as ISO (Fig. 1) or norepinephrine (data not shown), or by cAMP (data not shown), was detected. In addition, thiazolidinedione alone (BRL 49653-TZD) had no effect (Fig. 1). Combined treatments were tested and in those conditions a stimulation was detected by RA and TZD or RA and ISO. No stimulation was noted with a combined TZD and ISO treatment (data not shown). The addition of TZD increased more than 3-fold the stimulation obtained by RA alone whereas the effect of ISO was only less than twice. The combination of the three drugs (RA, ISO, and TZD) induced a very large increase in the basal activity.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Stimulation of human ucp1 gene promoter by retinoids and other drugs. Delineation of a responding sequence between bp 3820 and 2790. Duplicate dishes of 1B8 cells were transiently transfected with 20 µg of the reporter gene and 2 µg of a plasmid encoding -galactosidase under the control of SV40 promoter. The cells were incubated for 20 h in medium supplemented with 5% charcoal fetal calf serum and treated or not with drugs. The plasmids used were: 6300CAT, 3820CAT, and 2790CAT. They have been previously described (see "Experimental Procedures"). Drug concentrations were: 106 M thiazolidinedione BRL 49653 (TZD), 106 M all-trans-RA (RA), 106 M isoproterenol (ISO). Fold stimulation were represented, a value of 1 was given to the reporter plasmid without treatment. The results are representative of at least five separate experiments, with at least two different preparations of plasmid. The mean ± S.E. are given.

These results underlined that RA was necessary to observe a stimulation by ISO and/or TZD. ISO was described previously as an inducer of the rodent gene (7, 8). To verify that under our conditions, there was not any possible defect in the ISO transduction signal, the same experiments were repeated with the rat ucp1 gene. -Adrenergic agonist alone increased the transcriptional activity of the rat transgene 6-fold (data not shown), in agreement with previous results (14, 17). The stimulation of the endogenous mouse ucp1 gene transcription in the presence of ISO, or the other drugs used, was also confirmed by Northern blot analysis (data not shown). These observations were in agreement with a normal ISO transduction signal in 1B8 cells. The ISO response, under the same conditions, was different for the rodent genes compared with the human gene, so the different behavior seemed intrinsic to the human gene. Nevertheless, it should be considered that transient transfections were made in a 1B8 mouse cell line, which could interfere with the normal activity of the human gene. It is possible that the dependence of the human gene to retinoids could be due to different intrinsic level of retinoids in human brown adipose cells compared with 1B8 mouse cells.

To further delineate the regulating regions of the human ucp1 gene, several CAT constructs containing different fragments of the 6300-bp region were performed. The 2790CAT construct did not show any drug stimulation (Fig. 1). In contrast, the 3820CAT construct was able to drive a transcriptional activity stimulated by drugs alike the original plasmid (6300CAT) although the maximal stimulation (RA + ISO + TZD) was lower. Therefore, since the qualitative characteristics of the original promoter were retained in this construct we continued our studies by examining the region comprised between bp 3820 and 2790. This region was cloned in either orientation in front of the tk-heterologous promoter and the cat reporter gene (3820/2790TK and 2790/3820TK in Fig. 2). Since according to previous experiments the presence of RA is a prerequisite for stimulation of the human ucp1 promoter, the effect of ISO and TZD were examined in the context of the RA stimulation. Results presented in the Fig. 2 showed that the bp 3820 to 2790 region was able to trigger stimulation by drugs. It should be noticed that the maximal rate of stimulation obtained with the 3820/2790TK construct in the presence of the three drugs (35-fold increase in Fig. 2) was higher than with the original 6300CAT construct (16-fold increase in Fig. 1). The pattern was slightly different since stimulation by RA alone was markedly enhanced with the 1-kb region constructs. This reinforced the idea that the essential regulatory elements mediating the RA, ISO, and TZD effects were retained within the bp 3820 to 2790 region.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Identification of a 350-bp enhancer region essential for drug stimulation between bp 3820 and 3470. A CAT construct was made with the human ucp1 region (bp 3820 to 2790) in the two orientations upstream of the tk-heterologous promoter and the cat reporter gene. The arrows indicate the orientation of the fragment. Several deletions were achieved in this 1-kb region. Transient transfection and drug stimulation were done in 1B8 cells under the same conditions as described in the legend to Fig. 1. Fold stimulation were represented, a value of 1 was given to the reporter plasmid without treatment. The results are representative of at least three separate experiments with at least two different preparations of plasmid. The mean ± S.E. are given.

Different deletions were made in this 1-kb region (Fig. 2). Deletion from bp 3820 to 3328 region led to the 3328/2790TK construct, which was completely inactive, whereas the 3328/3820TK construct still kept the stimulation by drugs. Moreover, a 350-bp element located between bp 3820 to 3470 could be defined as an enhancer as it was able to drive transcription in the presence of a homologous or a heterologous promoter and to mediate stimulation by drugs. The comparison of the two constructs obtained with the 350-TK-CAT in direct or antisense orientation revealed that according to the orientation, the stimulation by RA alone varied greatly from 3-fold when in antisense orientation (350-TK-CAT antisense, bottom of Fig. 2) to 8-fold when inserted in direct orientation (350-TK-CAT). In this latter case, the synergism between RA and ISO or RA and TZD was masked by this high effect of RA. On the other hand, the synergism between the three drugs was maintained. Finally, the similitude between the stimulation pattern obtained with the original plasmid (6300CAT) and the 350-TK-CAT antisense was striking and confirm that the essential elements mediating the hormonal effects were retained within the 350-bp element, whereas spacial constraints may influence the pattern of stimulation observed with the different constructs containing this 350-bp element (Figs. 1 and 2).

Sequence Homology between Human, Rat, and Mouse Enhancers-- The importance of a 211-bp enhancer has been described in rat and mouse ucp1 gene (14, 18). In humans, the detection of the 350-bp hormone-sensitive region acting as an enhancer allowed the comparison between rat and human enhancers. The sequence homology between the rodent enhancer and the human 350-bp enhancer were 62.5% for rat and 60.1% for mouse, respectively, whereas a 69.7% level of homology was calculated between rat and mouse enhancers. The human enhancer was located between bp 3820 to 3470 compared with bp 2494 to 2283 for the rat gene (14) and to bp 2530 to 2310 for the mouse gene (18). A comparison between the enhancer sequences from the three species is given in Fig. 3. The essential regulatory elements previously described for rat and mouse gene are also indicated in the figure. As will be discussed below in detail, the human enhancer element contains five hexamers referred to as A, B, C, D, and E (Fig. 3), which represent 5 putative binding retinoid response elements (RARE). A, B, and C sites are in a 5' to 3' orientation, whereas D and E sites are in the 3' to 5' orientation. The site A also represents a potential site for the binding of factors of the ATF/CREB family.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Comparison of human, rat, and mouse enhancer sequences. The sequence of the human 350-bp enhancer and corresponding rat and mouse enhancer sequences are given. Lowercase letters in the rat and mouse sequences represent bases which differ from the human sequence. Hyphens were inserted along sequences to improve the alignment between the three sequences. The alignment of these sequences allowed the comparison of essential elements. Essential regulatory elements previously described in the rat (UAR, TRE, up-mid, and dn-90) and in the mouse (CRE2, BRE1, and PPRE) genes are underlined (16-19). The bold letters represent the 92-bp region in the rat sequence and the 105-bp region in the human sequence, as examined in Fig. 4. In the human enhancer sequence, five putative half-sites for binding of nuclear receptors are represented by an arrow and lettered from A to E. One consensus site for the binding of the ATF/CREB family factors is also indicated. A 43-bp fragment is indicated (bp 3730 to 3688), see below for further explanations.

The Essential Regulatory Region Is Contained in a 105-bp Fragment-- With the aim of determining the activity of the regulatory elements in the human gene, shorter internal deletions in the 6300CAT and in the 350TKCAT constructs were made. Deletion of the 350-bp element from the entire 5' construct was sufficient to abolish the drug response (Fig. 4A). These results confirmed that the 350-bp region contained the regulatory elements required for the activity of drugs. A shorter deletion in the 6300CAT plasmid, comprising the hexamer motifs region from bp 3762 to 3705, blunted the stimulation by drugs, whereas the deletion of an upstream region (bp 3820 to 3762) had no effect on the relative CAT activity. Comparable deletions in the 350TKCAT confirmed the importance of bp 3762 to 3705 (350TKdel60, Fig. 4B). The 350TKdel47 and 350bal15 constructs showed a residual CAT activity. On the contrary, deletions in the 3' moiety of the 350-bp element containing the region equivalent to the rat UAR or the mouse BRE1 did not modify the hormonal response (350bal51 and 350bal1). The sequence analysis in the 5' moiety of the 350-bp region (Fig. 3) showed five potential hexamer-binding sequences that match the consensus for binding nuclear hormone receptors such as RARs and RXRs (36). This fragment sequence is given in Fig. 3. Two of these motifs (B and D) were identical to the canonical retinoid receptor binding motif, (A/G)G(G/T)TCA, two of them (A and E) differed in one substitution and the C motif contains two substitutions. The five motifs could be arranged in direct repeat, palindrome, or everted palindrome. Such an arrangement suggests the existence of a putative multipartite RARE. Moreover, a potential CREB site (ACGTCA) was located in the A motif (Fig. 3). These observations pointed out that the region comprising bp 3741 to 3682 and putatively these five hexamer motifs were essential for the ucp1 transcriptional activity.

View larger version (34K): [in this window] [in a new window]   Fig. 4.   Delineation of a 105-bp regulatory element by using internal deletions in the 350-bp enhancer. A, B, C, D, and E, are five putative half-sites for binding of nuclear receptors. PPRE, CRE2, BRE1, UAR, and TREs correspond to regulatory regions described in rodent (see Fig. 3). For plasmid descriptions, see "Experimental Procedures." Experiments were done in 1B8 cells as described in the legend to Fig. 1. Fold stimulation were represented, a value of 1 was given to the reporter plasmid without treatment. RA, treatment with 106 M all-trans-RA; RA+ISO+TZD, combined treatment with 106 M RA, 106 M isoproterenol, and 106 M TZD49653. The results are representative of at least three separate experiments, with at least two different preparations of plasmid. The mean ± S.E. are given. A, deletions in the 6300CAT construct. B, deletions in the 350TKCAT construct. The sequence from bp 3741 to 3636 corresponding to the 105-bp is evidenced.

To further delineate the nucleotides implicated in the binding of nuclear factors, a 105TKCAT plasmid containing the bp 3741 to 3636 region and comprising the essential bp 3741 to 3682 was constructed. The comparison of this 105-bp region between animal species revealed a 70% identity between human and rat, a 73% identity between human and mouse, and an 84% identity between rat and mouse. Transient transfection of 1B8 cells demonstrated that the 105TKCAT construct was able to drive the drug stimulation in both orientations (Fig. 5A). A construct containing two 105-bp elements in the antisense orientation was also obtained, its relative CAT activity was 5-fold higher than the activity of a single 105-bp element (Fig. 5A). However, the 105TKCAT construct presented similar activity than the 350-TK-CAT where RA activation was very high masking the additive effect of ISO or TZD although the effect of the three drugs together was maintained (Fig. 2).

View larger version (25K): [in this window] [in a new window]   Fig. 5.   Analysis of the 105-bp fragment activity and effects of mutations in sites A, B, C, D, and E. Experiments were done in 1B8 cells as described in the legend to Fig. 1. Fold stimulation were represented, a value of 1 was given to the reporter plasmid without treatment. RA, treatment with 106 M all-trans-RA; RA+ISO+TZD, combined treatment of 106 M RA, 106 M isoproterenol, and 106 M TZD49653. The crosses represent the mutated site. The results are representative of at least three separate experiments, with at least two different preparations of plasmid. The mean ± S.E. are given.

Site-directed mutagenesis was carried out to determine the effects of sites A, B, C, D, and E, respectively, on promoter activity and stimulation by drugs (Fig. 5, B and C, mutated nucleotides are listed in Fig. 6A). The activity of the mutated A105TK construct maintained the same RA stimulation than the wild type construct, but the additive effect of the three drugs was lost (Fig. 5B). In our conditions, it was not possible to distinguish if the A motif was involved in the ISO and/or TZD effects but this site is important to mediate the stimulation by three drugs. The activity of the mutated plasmid D105TK was completely abolished and the E105TK showed only a residual RA activity (Fig. 5B). These experiments clearly underlined the importance of the integrity of D and E hexamer motifs for drug stimulation. Such results suggested the implication of these two sites in the RA response which was essential to observe the ISO and TZD effect. In the presence of drugs, the activities of the mutated B105TK and C105TK plasmids were 2-fold higher than that of the wild type (Fig. 5C), thus suggesting that these sites could down-regulate the efficiency of human ucp1 gene transcription. To further explain how this complex response element works, in vitro assays of binding of factors to cis-elements were carried out using EMSA.

View larger version (58K): [in this window] [in a new window]   Fig. 6.   Electrophoretic mobility shift assays using the 43-bp domain as a probe. The 43-bp probe was selected from the 105-bp fragment (bp 3730 to 3688). Nuclear extracts from 1B8 cells (5 µg) were incubated with 0.3 ng of the 32P-labeled 43-bp fragment or mutated oligonucleotides. Protein-bound and free DNAs were separated by gel electrophoresis. A, the sequences of the wild-type probe 43 and the mutated oligonucleotides are given, with mutations in bold type. B, the different oligonucleotides were labeled and used as probes. Complexes formed were named I, II, and III. C, the 43-bp oligonucleotide was used as the radioactive probe and competitions were done in the presence of a 50- or 100-fold molar excess of the non-radioactive oligonucleotides except for the 43mutD, which was used at a 100-fold molar excess, and for DR1, which was used at a 25-, 50-, or 100-fold molar excess. Competition with probes 43, 43mutE, 43mutD, and DR1 came from the same gel migration, whereas competition with 43mutB, 43mutC, and 43mutB/C were derived from another gel, which explains the different migration between complexes I, II, and III. Independent experiments were performed at least 5 times.

Electrophoretic Mobility Shift Assays-- EMSA were done to verify the in vitro ability of the hexamer A (A/CREB motif) to bind retinoid factors or CREB/ATF family factors. Nuclear factors from 1B8 cells bound to a probe containing this element and B, C, and E hexamers (RA/CREB probe, see "Experimental Procedures" for sequence, data not shown). The complex was supershifted by an anti-ATF antibody and a full competition by an oligonucleotide containing a CREB consensus motif was observed (data not shown). Antibodies against RAR and RXR factors did not affect the binding (data not shown). Therefore, the binding to the probe was due to the ability of the A/CREB motif to bind ATF/CREB family factors.

The contribution of each of the other half-sites to factor binding was assayed using synthetic 43-bp oligonucleotides (bp 3730 to 3688) containing mutations in B, C, D, or E half-sites already tested in transfection experiments (Figs. 5 and 6A). An oligonucleotide with mutations in both B and C sites was also synthesized. These oligonucleotides were used as radioactive-labeled probes or as competitors of binding against the wild-type 43-bp probe (referred to as 43). The wild-type oligonucleotide induced three retarded complexes (I, II, and III) in the presence of nuclear factors from 1B8 cells (Fig. 6B). The same complexes (slightly less intense) were obtained with the 43mutB oligonucleotide, indicating that integrity of the B half-site was not required for binding. In contrast, mutation of the D box abolished the binding of factors (Fig. 6B). These data confirmed the results obtained in transfection experiments (Fig. 5). The other mutated oligonucleotides, 43mutC, 43mutE, and 43mutB/C, only induced the shifted I and II complexes, but with a lower affinity (Fig. 6B). A 50-fold molar excess of unlabeled 43-bp oligonucleotide was sufficient to completely prevent the formation of the three complexes with the 43 probe, whereas some binding could still be detected when the same molar excess of mutated oligonucleotide competitors was used (Fig. 6C). In addition, the 43mutD oligonucleotide did not compete at all with the 43 probe (Fig. 6C). Taken together, these results suggest that the D hexamer motif is essential for binding of nuclear factors that leads to the stimulation of transcription. The absence of the E hexamer still allowed a slight binding of nuclear factors in vitro, which could explain the residual activity detected when this hexamer was mutated in the 105-bp DNA fragment in front of the CAT gene (Fig. 5B). A DR1 oligonucleotide described as RA and PPAR consensus response element was used as a competitor of the 43 probe (Fig. 6C, see "Experimental Procedures" for sequence). A 50-fold molar excess of DR1 oligonucleotide efficiently displaced the three complexes (Fig. 6C). These results are in agreement with the suggestion that RA and TZD stimulate the human ucp1 gene transcription by acting via a RA and a PPAR response elements.

Monoclonal antibodies against the nuclear receptors were used to further identify the factors that bind the 43 probe (Fig. 7A). The three complexes were abolished in the presence of an antibody against the different isoforms of RXR, suggesting that the complexes at least contained RXR. An antibody against the RAR isoform essentially decreased the binding of complex I, suggesting it is an RAR/RXR heterodimer, whereas an antibody against the RAR isoform totally removed complex III, suggesting it is an RAR/RXR heterodimer. An antibody against the RAR isoform had no effect on binding (Fig. 7A). The antibody against the PPAR isoform supershifted complex II, indicating the presence of PPAR in this complex.

View larger version (82K): [in this window] [in a new window]   Fig. 7.   RXR, RAR, RAR, and PPAR bind to the 43-bp domain. An additive factor was required. A, 5 µg of nuclear extracts from 1B8 cells were incubated with the 32P-labeled 43-bp fragment. When antibodies were used, they were preincubated with nuclear extracts before adding the probe. Independent experiments were performed at least 7 times. B, nuclear extracts were prepared from COS cells after overexpression of PSG5 vectors expressing RAR, RAR, RXR, or PPAR. In each slot, 0.3 ng of labeled 43 probe was incubated with 2 µg of COS extracts for RAR + RXR and for RAR + RXR or 10 µg for RXR + PPAR or 5 µg for RAR + RAR + RXR + PPAR. COS cells were co-transfected with plasmids as indicated in the figure (see "Experimental Procedures" for details). Independent experiments were performed at least 5 times by using three separate preparation of COS extracts. C, left panel, 0.3 ng of labeled 43 probe was incubated with 2 µg of 1B8 nuclear extracts or/and 5 µg of COS extracts. COS cells were co-transfected with RAR + RAR + RXR + PPAR plasmids. Independent experiments were performed at least twice. Right panel, 0.3 ng of labeled 43 probe was incubated with 2 µg (left lane) or 5 µg (right lane) of 1B8 nuclear extracts. D, 0.3 ng of labeled 43 probe was incubated with 2 µg of 1B8 nuclear extracts and 5 µg of COS extracts. COS cells were co-transfected with RAR + RAR + RXR + PPAR plasmids. When antibodies were used, they were preincubated with nuclear extracts before adding the probe. Independent experiments were performed at least twice. Results in B, C, and D came from the same gel migration.

To determine whether the 43-bp region interacted with RARs, RXRs, and PPAR, EMSA were performed in the presence of RAR, RAR, RXR, and PPAR overexpressed in COS cells (Fig. 7B). The migration of the complexes formed in the presence of the nuclear receptors overexpressed in COS cells was different from that observed with 1B8 cells nuclear factors. There are several possible explanations for the different migration: (i) the post-translational modifications assumed by COS cells compared with differentiated brown adipocytes, (ii) the presence or not of the ligand, (iii) the homodimerization due to the overexpression, and (iv) the absence of necessary cofactors.

The 43 probe formed a complex with RAR and RXR (Fig. 7B, referred as complex A) which was supershifted with antibodies against RXR and RAR but not with anti-RAR antibodies (data not shown). Another complex was also formed with RAR and RXR (Fig. 7B, referred as complex B), whereas no complex was observed when PPAR and RXR were coexpressed (Fig. 7B). When RAR, RAR, and RXR were co-expressed in the absence (data not shown) or presence of PPAR only a large complex was observed. This large complex contains A and B complexes (Fig. 7B), as verified with antibodies (Fig. 7D). Under these conditions, we were not able to detect PPAR binding even though different PPAR expression vectors were tested (data not shown).

The results obtained from combined treatments with TZD suggested a role of PPAR (Fig. 1). Moreover, EMSA with the 43 probe and 1B8 nuclear extracts led to identify a binding of PPAR (complex II, Fig. 7A) despite the absence of a PPRE consensus. However, in our conditions of nuclear receptors overexpression in COS cells, no PPAR binding was detected. Taken together, these results suggested that another factor (or more) could be necessary to observe a PPAR binding. In order to test this hypothesis, a small amount of 1B8 nuclear extract (2 µg) was added to the COS factors prepared after co-expression of RAR, RAR, RXR, and PPAR vectors. When 1B8 nuclear factors were incubated at the concentration of 2 µg with the 43 probe (Fig. 7C, second panel, left lane), the bands corresponding to the three complexes (I, II, and III) were quite faint compared with 5 µg (Fig. 7C, second panel, right lane). When 2 µg of 1B8 nuclear factors were incubated in the presence of nuclear receptors overexpressed, two major bands were observed (Fig. 7C). The lowest band corresponded to the A and B complexes described in Fig. 7B and the other band to a new complex referred as C. Antibodies against RAR, RAR, RXR, and PPAR were tested with the A, B, and C complexes. Results in Fig. 7D demonstrated that the three complexes were supershifted by RXR antibodies. An antibody against RAR disturbed all the complexes, but previous experiments showed that complex A was supershifted by this antibody (data not shown). In our conditions, this antibody always showed a nonspecific interaction with other complexes, more than the other antibodies used (Fig. 7A). The complex B was supershifted by RAR antibody while complex C was supershifted by PPAR antibody (Fig. 7D). RAR antibody disturbed the migration but there was no specific interaction (Fig. 7D).

These results demonstrated that the 43 probe binds RAR, RAR, RXR, and PPAR. However, the overexpression of PPAR and RXR was not sufficient to have a binding. The PPAR binding was detected only when 1B8 nuclear extracts were added, which agree with the requirement of an additional factor for the PPAR binding.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

There is no previous report showing that transcription of the human ucp1 gene can be regulated by hormones or drugs. The aim of this study was to answer this question and in particular to identify cis-elements and transcriptional factors regulating the human ucp1 gene.

The results presented in this report demonstrate that a 6300-bp upstream region is able to mediate strong stimulation of human ucp1 transcription by a mixture of retinoic acid, isoproterenol, and thiazolidinedione. Interestingly, under our conditions, ISO and TZD stimulate gene transcription only in the presence of RA, suggesting a permissive effect of retinoids. The lack of stimulation by ISO alone is the main difference between the human and rodent ucp1 gene.

Deletions made in the human 6300-bp upstream region have shown that a 350-bp enhancer, partially homologous to the rat and mouse enhancers, mediates the stimulation by drugs. A 105-bp element essential for drug-dependent transcriptional activation has been identified by dissecting the 350-bp enhancer. This 105-bp enhancer region contains a multipartite response element where five hexamer motifs (A, B, C, D, and E) of nuclear receptors are present.

Regulation of gene transcription by nuclear hormone receptors is mediated by binding of these receptors to specific DNA sequences located in the regulatory unit of genes. Retinoid effects are mediated by two classes of nuclear receptors, RAR (, , and ) and RXR (, , and ), which belong to the nuclear receptor superfamily, both of which induce transcriptional activation through specific RAREs (36). Three isoforms of peroxisome proliferator-activated receptors (PPAR, -, and -) are also described (37, 38). It is well known that PPAR has a major role in adipocyte differentiation (39, 40). Some fatty acids, eicosanoids, or 15-deoxy-^12,14-prostaglandin have been identified as natural PPAR ligands and thiazolidinediones as synthetic ligands. DNA binding of PPAR requires formation of a heterodimer of PPAR and RXR interacting with a PPRE (for review, see Ref. 38). The multiplicity of half-sites in genes is a common feature of naturally occurring hormone response elements. Furthermore, there is evidence that cooperation between these half-sites contributes to the transactivation process. Tripartite RAREs have been identified in some genes. These elements could be organized in direct and/or everted repeats (41-44). Previous studies with the rat ucp1 gene have shown that an enhancer located at bp 2494 upstream from the startpoint was responsible for tissue specificity and drug stimulation (14, 16, 17). The effect of RA on this enhancer was related to a 92-bp fragment containing an UAR element able to bind retinoid receptor and AP1 family factors. There was no CRE consensus in the 92-bp element and although the UAR was not sufficient to mediate the ISO effect, mutations in this element abolished the ISO stimulation.³ These data suggested a possible synergism of drugs via a common response element. A complex RARE comprising three potential RAREs distributed within the 92-bp element was also defined in the rat ucp1 gene (16). The authors suggested that the enhancer contained a tissue-specific upCRS (cAMP response sequence) necessary for the entire response to norepinephrine (45). In the mouse gene, the essential regulatory regions are CRE2/BRE1 (18) and PPRE element (19). This paper describes a multipartite response element in the human gene different from mouse and rat regulatory regions despite the sequence homology between them.

The results obtained demonstrate that this complex element contains five hexamer motifs (A, B, C, D, and E) located inside the 105-bp region. The use of antibodies against nuclear receptors in EMSA showed that the 43-bp region (containing B, C, D, and E sites) bound RAR, RAR, RXR, and PPAR nuclear receptors and that the A site was able to bind ATF/CREB family factors. In addition, results from transfection and EMSA demonstrated that the D half-site was essential for an efficient transcriptional activity and also to observe any binding of nuclear factors. Results also showed that RA was required to observe a stimulation by ISO or TZD. Taken together these data suggested a hierarchical mechanism requiring the presence of the D site to mediate the RA effect by binding of RAR and RXR. Similar experiments showed that mutation of the E site still kept a residual RA activity and a slight binding of receptors. It could be proposed that D and E half-sites act as a DR9 able to bind retinoic nuclear receptors, even though the D site could interact in vitro with another half-site.

Transfection experiments showed that TZD stimulated the transcription of the human ucp1 gene in the presence of RA, moreover a PPAR binding on the 43-bp region was detected by EMSA despite the absence of a PPRE consensus. Further experiments indicated that the formation of the PPAR·RXR complex required an additional unknown factor to bind to this region. In view of the results, the identification of the half-sites forming a PPRE was not possible.

The two other half-sites within the 43-bp region were B and C, mutation of each site increased the drug stimulation thus suggesting that these sites could down-regulate the efficiency of transcription. However, both sites were important to observe an efficient binding of nuclear receptors. They could not be directly associated to the PPAR binding but they are probably implicated in the stabilization of the complexes.

The A site included in the 105-bp region was implicated in the stimulation by ISO and/or TZD in the presence of RA, as shown from mutagenesis. Experiments done could not distinguish the individual participation of ISO or TZD in the stimulation. Nevertheless, EMSA identified the A motif as a CREB/ATF factor-binding site. Moreover the 43-bp region was able to bind PPAR mediating the TZD effect. Taken together, these observations indicated that the A motif could be involved in the ISO response although this effect is dependent of the RA stimulation.

In conclusion, our findings support the hypothesis that this 105-bp region contains all regulatory elements involved in the drug response of the human ucp1 gene. The molecular analysis revealed a complex region where the binding of factors is achieved according to a hierarchy. Moreover, the presence of the five hexamer motifs is necessary to observe drug response with the correct amplitude. It could be noticed that the rat enhancer was described previously as a complex element within the 92-bp region necessary to mediate the drug response. Despite this comparable organization and the sequence homology, molecular mechanisms involved in the regulation of rat or human gene transcription are different (16, 45). To dissect further the signaling pathways that mediate drug stimulation and especially the permissive effect of retinoids on the human ucp1 gene it is necessary to identify cofactors and transcription factors that directly bind to the enhancer.

Recent results have revealed that transcriptional activation or repression by nuclear hormone receptors can be modulated by other nuclear proteins designated as co-activators and co-repressors (for a recent review, see Ref. 46). Several co-factors interact with retinoid receptors, but few studies have addressed the interaction of these proteins with PPAR (46, 47). Some cofactors were identified to potentiate the PPAR effect (48, 49) as the recently described PGC-1 (50). PGC-1 acts in BAT, through binding to PPAR, to increase the activation of the mouse ucp1 gene (50, 51). Although we were able to confirm an effect of PGC-1 on the rat ucp1 gene by co-transfection of nuclear receptor expression vectors in Chinese hamster ovary cells, no co-activation of the human ucp1 gene by PGC-1 was observed under the same conditions (data not shown). Moreover, in EMSA, addition of PGC-1 overexpressed in COS cells was not sufficient to observe a PPAR binding. It is not excluded that the lack of PGC-1 effect could be due to the absence of the unknown factor essential for the PPAR binding.

Beyond the understanding of the molecular mechanism, we are interested to find natural compounds able to stimulate the transcription of the human ucp1 gene via these regulatory elements. RA is an active metabolite of vitamin A and recent data have shown that dietary vitamin A supplementation in rats could increase the ucp1 mRNA levels (52). -Carotene is a precursor in vitamin A synthesis and a recent report showed that -carotene increased ucp1 mRNA levels in a consistent and dose-dependent manner, probably via the transformation of the -carotene in RA (53). Thus, it is particularly significant that metabolites or precursors of RA naturally occurring in food could directly affect BAT thermogenesis through regulation of ucp1 transcription. UCP1 is a powerful thermogenic protein which activates fatty acid oxidation and energy expenditure. The selection of compounds regulating ucp1 gene expression may be valuable in controlling thermogenesis in babies, and in activating thermogenesis and fat oxidation in obese and type 2 diabetic adult patients.

	ACKNOWLEDGEMENTS

We thank Dr. P. Chambon (IGBMC, Illkirch) for the gift of 4RX-1D12 anti-RXR monoclonal antibody and anti-RAR, RAR, RAR monoclonal antibodies and for the gift of expression plasmids RAR RAR, RAR, and RXR. We thank Dr. G. Schutz for the gift of PBLCAT5, PBLCAT6 plasmids, Dr. A. Dejean for RAR and RXR expression vectors, Dr. J. D. Tugwood for PPAR, and Dr. B. M. Spiegelman for the PGC1 expression vector. We thank Dr. R. Madrid-Gonzalez for help in preparing the E105TK mutant. We also express our gratitude to Dr. David Marsh for critical reading of the manuscript.

	FOOTNOTES

^* This work was supported in part by the Centre National de la Recherche Scientifique (CNRS), the Institut National de la Santé et de la Recherche Médicale (INSERM), the Association pour la Recherche contre le Cancer, the Human Frontier Science Program organization (HFSP), and the Institut de Recherches Servier.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Supported by a grant from the European Union (Marie Curie Mobility Training Fellowship).

^§ Supported by a Ph.D. thesis fellowship from the Association pour la Recherche contre le Cancer.

^¶ To whom correspondence should be addressed: CEREMOD, CNRS, 9 rue Jules Hetzel, 92190 Meudon, France. Tel.: 33-0-1-45-07-57-48; E-mail: doulcier@infobiogen.fr.

Published, JBC Papers in Press, July 31, 2000, DOI 10.1074/jbc.M001678200

² A-M. Cassard-Doulcier, unpublished data.

³ M. Larose, unpublished data.

	ABBREVIATIONS

The abbreviations used are: UCP1, uncoupling protein 1; BAT, brown adipose tissue; ISO, isoproterenol; TZD, thiazolidinedione; RA, retinoic acid; AP1, activator protein 1; CAT, chloramphenicol acetyltransferase; UAR, ucp1-gene activating region; BRE-1, brown fat regulatory element; EMSA, electrophoretic mobility shift assays; RAR, retinoid acid receptor; RXR, retinoid X receptor; PPAR, peroxisome proliferator-activated receptor; RARE, retinoid acid response element; PGC-1, PPAR co-activator-1; PPRE, peroxisome proliferator response element; ATF, activator transcription factor; CRE, cAMP response element; CREB, CRE-binding protein; DR1, direct repeat 1; bp, base pair(s); kb, kilobase pair(s); TK, thymidine kinase.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. Muraoka, A. Fukushima, S. Viengchareun, M. Lombes, F. Kishi, A. Miyauchi, M. Kanematsu, J. Doi, J. Kajimura, R. Nakai, et al. Involvement of SIK2/TORC2 signaling cascade in the regulation of insulin-induced PGC-1{alpha} and UCP-1 gene expression in brown adipocytes Am J Physiol Endocrinol Metab, June 1, 2009; 296(6): E1430 - E1439. [Abstract] [Full Text] [PDF]

				H. Wang, Y. Zhang, E. Yehuda-Shnaidman, A. V. Medvedev, N. Kumar, K. W. Daniel, J. Robidoux, M. P. Czech, D. J. Mangelsdorf, and S. Collins Liver X Receptor {alpha} Is a Transcriptional Repressor of the Uncoupling Protein 1 Gene and the Brown Fat Phenotype Mol. Cell. Biol., April 1, 2008; 28(7): 2187 - 2200. [Abstract] [Full Text] [PDF]

				D. Debevec, M. Christian, D. Morganstein, A. Seth, B. Herzog, M. Parker, and R. White Receptor Interacting Protein 140 Regulates Expression of Uncoupling Protein 1 in Adipocytes through Specific Peroxisome Proliferator Activated Receptor Isoforms and Estrogen-Related Receptor {alpha} Mol. Endocrinol., July 1, 2007; 21(7): 1581 - 1592. [Abstract] [Full Text] [PDF]

				J. Robidoux, W. Cao, H. Quan, K. W. Daniel, F. Moukdar, X. Bai, L. M. Floering, and S. Collins Selective Activation of Mitogen-Activated Protein (MAP) Kinase Kinase 3 and p38{alpha} MAP Kinase Is Essential for Cyclic AMP-Dependent UCP1 Expression in Adipocytes Mol. Cell. Biol., July 1, 2005; 25(13): 5466 - 5479. [Abstract] [Full Text] [PDF]

				G. Leonardsson, J. H. Steel, M. Christian, V. Pocock, S. Milligan, J. Bell, P.-W. So, G. Medina-Gomez, A. Vidal-Puig, R. White, et al. Nuclear receptor corepressor RIP140 regulates fat accumulation PNAS, June 1, 2004; 101(22): 8437 - 8442. [Abstract] [Full Text] [PDF]

				B. CANNON and J. NEDERGAARD Brown Adipose Tissue: Function and Physiological Significance Physiol Rev, January 1, 2004; 84(1): 277 - 359. [Abstract] [Full Text] [PDF]

				C. Tiraby, G. Tavernier, C. Lefort, D. Larrouy, F. Bouillaud, D. Ricquier, and D. Langin Acquirement of Brown Fat Cell Features by Human White Adipocytes J. Biol. Chem., August 29, 2003; 278(35): 33370 - 33376. [Abstract] [Full Text] [PDF]

				J. S. Rim and L. P. Kozak Regulatory Motifs for CREB-binding Protein and Nfe2l2 Transcription Factors in the Upstream Enhancer of the Mitochondrial Uncoupling Protein 1 Gene J. Biol. Chem., September 6, 2002; 277(37): 34589 - 34600. [Abstract] [Full Text] [PDF]

				H. Oberkofler, H. Esterbauer, V. Linnemayr, A. D. Strosberg, F. Krempler, and W. Patsch Peroxisome Proliferator-activated Receptor (PPAR) gamma Coactivator-1 Recruitment Regulates PPAR Subtype Specificity J. Biol. Chem., May 3, 2002; 277(19): 16750 - 16757. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/41/31722    most recent M001678200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by del Mar Gonzalez-Barroso, M. Articles by Cassard-Doulcier, A.-M. Search for Related Content PubMed PubMed Citation Articles by del Mar Gonzalez-Barroso, M. Articles by Cassard-Doulcier, A.-M. Social Bookmarking                      What's this?