| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Metabolic and Endocrine Diseases (O.v.B., N.H., R.B., E.K.), University Medical Center Utrecht, 3584 EA Utrecht, The Netherlands; Department of Physiological Chemistry (A.B.B., N.J.F.v.d.B.), University Medical Center Utrecht, 3584 CG Utrecht, The Netherlands; and Department of Biochemistry and Molecular Biology (L.G., S.M.), University of Southern Denmark, 5230 Odense M, Denmark
Address all correspondence and requests for reprints to: Eric Kalkhoven, Department of Metabolic and Endocrine Diseases, University Medical Center Utrecht, Room KE.03.139.2, Lundlaan 6, 3584 EA Utrecht, The Netherlands. E-mail: e.kalkhoven{at}umcutrecht.nl.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(PPAR) plays a key role in the regulation of lipid and glucose metabolism in adipocytes, by regulating their differentiation, maintenance, and function. The transcriptional activity of PPAR is dictated by the set of proteins with which this nuclear receptor interacts under specific conditions. Here we identify the HIV-1 Tat-interacting protein 60 (Tip60) as a novel positive regulator of PPAR transcriptional activity. Using tandem mass spectrometry, we found that PPAR and the acetyltransferase Tip60 interact in cells, and through use of chimeric proteins, we established that coactivation by Tip60 critically depends on the N-terminal activation function 1 of PPAR, a domain involved in isotype-specific gene expression and adipogenesis. Chromatin immunoprecipitation experiments showed that the endogenous Tip60 protein is recruited to PPAR target genes in mature 3T3-L1 adipocytes but not in preadipocytes, indicating that Tip60 requires PPAR for its recruitment to PPAR target genes. Importantly, we show that in common with disruption of PPAR function, small interfering RNA-mediated reduction of Tip60 protein impairs differentiation of 3T3-L1 preadipocytes. Taken together, these findings qualify the acetyltransferase Tip60 as a novel adipogenic factor. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, -β/
, and -, all bind to PPAR-responsive elements (PPREs) in the promoter regions of target genes as obligate heterodimers with retinoic acid X receptors (RXRs). However, each receptor is associated with distinct biological effects, which may partly be explained by their tissue-specific expression. An important site of PPAR action is adipose tissue, where it is a key player in the differentiation, maintenance, and function of these cells (5, 6, 7). Its relevance for the differentiation of fibroblast-like mesenchymal stem cells into adipocytes, a process known as adipogenesis, is clearly exemplified by PPAR+/– mice, which lack adipose tissue (8, 9, 10). In addition, in vitro differentiation of fibroblasts into mature adipocytes can be induced by introduction of PPAR, showing that PPAR is not only necessary but also sufficient for adipogenesis (11). Compelling genetic evidence for this view comes from the familial partial lipodystrophy subtype 3 (FPLD3, MIM 604367), an autosomal dominantly inherited disorder, characterized by gradual loss of sc fat from the extremities and an accumulation of excess fat in the intraabdominal regions, which is caused by heterozygous mutations in the PPARG gene (4, 12).
Like other NRs, PPAR consists of distinct functional domains including an N-terminal transactivation domain (AF1), a highly conserved DNA-binding domain (DBD), and a C-terminal ligand-binding domain (LBD) that contains a ligand-dependent transactivation function (AF-2). Furthermore, two isoforms of PPAR exist, named PPAR1 and PPAR2, which differ by the presence of an additional N-terminal 30 amino acids in the PPAR2 isoform. The crystal structure of the PPAR LBD reveals 13 -helices and a small four-stranded β-sheet (13, 14). Upon ligand binding, the AF-2 helix (also referred to as helix 12) is stabilized in an active state, resulting in a charge clamp pocket consisting of a conserved lysine residue in helix 3 and a glutamate in the AF-2 helix (K329 and E499 in PPAR2, respectively) (13, 14). This pocket serves as a binding site for coactivator complexes, including the steroid receptor coactivator 1 (SRC1)/cAMP response element-binding protein (CREB)-binding protein (CBP) and thyroid hormone receptor associated protein (TRAP)/vitamin D receptor-interacting protein (DRIP)/activator-recruited cofactor (ARC) complexes, which interact with the LBD through the LXXLL motifs (in which L is leucine and X is any amino acid) present in these coactivators (15, 16). Interestingly, thiazolidinediones, drugs that ameliorate insulin resistance in humans, are able to initiate such a sequence of events because they are ligands for PPAR (17).
Although numerous studies have focused on the ligand-dependent AF2 function of PPARs, recent studies indicate that the AF1 region also plays an important regulatory role in PPAR signaling. Using chimeric PPAR-PPARβ/ proteins, the AF1 region of PPAR was shown to be essential for adipogenesis (18, 19). Furthermore, gene expression profiling revealed that the AF1 regions of the different PPAR family members are the main determinants of isotype-selective gene expression (19). It seems plausible that, in analogy with the AF2 domain, the activity of the AF1 region is dictated by the set of proteins with which this domain interacts. Despite its emerging importance, only a limited number of proteins have been described so far that interact with and regulate the function of the N-terminal region of PPAR. These AF1-interacting proteins include PPAR coactivator 2 (PGC2)/SCAN domain protein 1 (SDP1) (18, 20) and the transcriptional coactivator with PDZ-binding motif TAZ (21). PGC2 interacts specifically with and increases the activity of PPAR and not PPAR or PPARβ/ (18), although studies on the human homolog SCAN domain protein 1 showed binding to PPAR and -β/ as well (20). Interestingly, overexpression of PGC2 in 3T3-L1 preadipocytes resulted in increased adipogenesis, thereby qualifying this protein as an adipogenic cofactor (18). Because PGC2 lacks intrinsic transactivation potential (18), the molecular mechanism behind coactivation of PPAR by PGC2 remains to be established. TAZ, which is a positive regulator of the osteogenic transcription factor Runx2, inhibits the activity of PPAR, but also in this case, the mechanism is unknown (21). The acetyltransferases CBP and p300 can also interact with the AF1 region of PPAR, but the relative importance of this interaction is unclear because CBP and p300 also bind to the LBD in a ligand-dependent fashion (22). Interestingly, the AF1 region is also subject to phosphorylation at serine residue 112 (23) and SUMOylation at lysine residue 107 (24, 25, 26), and it seems likely that such posttranslational modifications regulate interactions with coregulators and/or vice versa, ultimately controlling the output of the AF1 region.
To identify novel regulators of PPAR activity, we performed a mass spectrometry analysis of PPAR-associated proteins and by this means found the HIV-1 Tat-interacting protein 60 (Tip60). Tip60 is a member of the MYST family of acetyltransferases, named after its founding members monocyte leukemia zinc-finger (MOZ), Ybf2/Sas3, Sas2, and Tip60, which share a highly conserved MYST acetyltransferase domain but display limited homology outside this region (27, 28). Tip60 is part of a large multiprotein complex (29, 30, 31), implicated in many cellular processes such as DNA damage repair, cell cycle control, and apoptosis (32, 33, 34). We show here that Tip60 targets the AF1 region of PPAR, a region of the protein implicated in isotype-selective gene expression and adipogenesis (19). Furthermore, siRNA-mediated knockdown of Tip60 expression resulted in inhibition of the differentiation of 3T3-L1 cells into adipocytes. These findings qualify the MYST acetyltransferase Tip60 as a novel adipogenic factor.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(sc-7196), anti-Tip60 (sc-5725), and anti-Fabp4 (sc-18661) from Santa Cruz Biotechnology (Santa Cruz, CA); anti-FLAG(M2)-HRP (A8592), anti-HA (H9658), antigoat-HRP (A9452), and anti-tropomyosin (T2780) from Sigma-Aldrich (St. Louis, MO); antirabbit-HRP (111035144) and antimouse-HRP (115035146) from Jackson ImmunoResearch Laboratories Inc. (West Grove, PA); and anti-Brd8 (BL 1231) from Bethyl Laboratories (Montgomery, TX). Anti-FLAG(M2) agarose beads (A02220) and Oil-Red-O (O-0625) were purchased from Sigma-Aldrich, crystal violet was from Chroma-Gesellschaft Schmid & Co. (Köngen, Germany). Formaldehyde solution (4%) was purchased from Klinipath (Duiven, The Netherlands), and RNAiMAX was purchased from Invitrogen (Carlsbad, CA). Purified Drososphila core histones were a kind gift from G. Chalkley (Erasmus University Medical Center, Rotterdam, The Netherlands).
Plasmids and small interfering RNA (siRNA) oligonucleotides
All recombinant DNA work was performed according to standard procedures (35). pCDNA-FLAG-MOZ (36), pCDNA-HA-HBO-1 (37), and pCDNA-HA-Tip60 (38) were kind gifts from Drs. D. M. Heery (University of Nottingham, UK), Z. Sun (Stanford University School of Medicine, Stanford, CA), and D. Trouche (Unité Mixte de Recherche 5099, Toulouse, France), respectively. pGEX4T1-PPAR2 deletion constructs (39) were a kind gift from Dr. S. Kato (University of Tokyo, Japan). pSport mMOF was purchased from ImaGenes GmbH (Berlin, Germany) (clone IRAVp968c0560D). Expression vectors pcDNA-hPPAR2/ and pcDNA-hPPAR/ were generated by site-directed mutagenesis of pcDNA-PPAR and -PPAR2 generating NheI restriction sites at codon 147 of PPAR2 and codon 108 of PPAR, and subsequent exchange of the relevant fragments. To generate the Fabp4 promoter reporter construct, the HindIII sites surrounding the 7.9-kb mouse Fabp4 promoter were used to clone the Fabp4 promoter into pBSK and from there subcloned into pGL3-Basic (Promega, Madison, WI) using SmaI and KpnI sites. The thymidine kinase (TK)-Luc reporter was generated by insertion of a BamHI fragment from 5xGal-TK-Luc (40) into a BamHI-BglII-digested pGL3 basic plasmid (Promega). All mutations were generated by QuikChange mutagenesis (Stratagene, La Jolla, CA) and verified by sequencing. All other plasmids have been described before (41). Control (D-001210-01-20) and mTip60 (M-057795-00) SMARTpool siRNA oligonucleotides or individual siRNA duplexes (MQ-057795-00) were purchased from Dharmacon (Lafayette, CO).
Cell culture, transient transfections, and reporter assays
The human osteosarcoma cell line U2OS and the human embryonic kidney 293T cell line (HEK293T) were maintained in DMEM Glutamax (Dulbecco) containing 10% fetal calf serum (Life Technologies, Inc., Rockville, MD), penicillin and streptomycin (both 100 µg/ml; Life Technologies). The murine 3T3-L1 cell line was cultured in the same media but with 10% bovine serum (Life Technologies) and penicillin and streptomycin (both 100 µg/ml; Life Technologies). Reporter assays were performed in 24-well plates with 1 µg 3xPPRE-tk-Luc reporter construct, 2 ng PPAR expression construct, 100 ng Tip60 expression construct (or empty vector), and 2 ng pCMV-Renilla (Promega) as described before (41).
Tandem mass spectrometry and Western blotting
For immunoprecipitation experiments, U2OS cells or 293T cells were grown in 15-cm dishes and transiently transfected with PPAR2 or Tip60 expression vectors (10 µg) using either Fugene6 or PEI transfection reagent. At d 2, cells were put on 2 µM rosiglitazone and harvested at d 3 in 500 µl RIPA buffer (without SDS) per dish. Immunoprecipitation of FLAG-PPAR2 and HA-Tip60 were performed at 4 C for at least 4 h, using anti-FLAG(M2) agarose beads or HA antibodies precoupled to agarose beads, respectively. For tandem mass spectrometry, overexpressed FLAG-PPAR2 and interacting proteins were isolated from 20 dishes (15 cm). Proteolytic digestion of proteins and liquid chromatography-tandem mass spectrometry were performed as described (42). For Western blotting, whole-cell lysates (directly lysed in 2x SDS-PAGE sample buffer) or immunoprecipitated proteins were subjected to SDS-PAGE and transferred to Immobilon membranes (Millipore, Bedford, MA). Enhanced chemiluminescence (Amersham Biosciences, Arlington Heights, IL) was used for detection after incubation with primary and secondary antibodies.
Glutathione S-transferase (GST) pull-down assays and histone acetylation assays
Recombinant Tip60 cDNA in the pCDNA3 expression vector was transcribed and translated in vitro in reticulocyte lysates in the presence of [35S]methionine according to manufacturer’s protocol (TNT T7 Quick Coupled Transcription/Translation Kit; Promega). Rosetta pLysS competent bacteria (Novagen, Madison, WI) were transformed with GST expression plasmids. GST fusion proteins were purified as described earlier (43). 35S-labeled proteins were incubated with GST fusion proteins in NETN buffer [20 mM Tris (pH 8.0), 100 mM NaCl, 1 mM EDTA, 0.5% Nonidet P-40] containing protease inhibitors (Complete; Roche). Samples were subsequently washed and subjected to SDS-PAGE. Signals were enhanced with Amplify (Amersham), gels were fixed and dried, and the 35S-labeled proteins were visualized by fluorography.
Histone acetylation assays were performed exactly as described (40), using immunopurified HA-tagged Tip60 from transiently transfected 293T cells and purified Drosophila core histones.
Chromatin immunoprecipitation (ChIP) assays
ChIP assays were performed as described (44). In short, 3T3-L1 preadipocytes and differentiated 3T3-L1 adipocytes were cross-linked with 1% formaldehyde in PBS and quenched with glycine (final concentration 0.125 M). Cells were washed two times with ice-cold PBS, collected in PBS, and resuspended in lysis buffer (0.1% SDS; 1% Triton X-100; 0.15 M NaCl; 1 mM EDTA; and 20 mM Tris-HCl, pH 8.0). Chromatin was fragmented using the Bioruptor (Diagenode SA, Liège, Belgium), and cellular debris was removed by centrifugation. Chromatin was diluted in lysis buffer and incubated with antibody against PPAR or Tip60 and incubated overnight with protein A or protein G agarose slurry (Amersham) in the presence of BSA (1 µg/µl). Immunoprecipitates were washed twice with buffer 1 [0.1% SDS, 0.1% sodium deoxycholate (NaDOC), 1% Triton X-100, 0.15 M NaCl, 1 mM EDTA, and 20 mM Tris-HCl (pH 8.0)], one time with buffer 2 [0.1% SDS, 0.1% NaDOC, 1% Triton X-100, 0.5 M NaCl, 1 mM EDTA, and 20 mM Tris-HCl (pH 8.0)] and buffer 3 [0.25 M LiCl, 0.5% NaDOC, 0.5% Nonidet P-40, 1 mM EDTA, and 20 mM Tris-HCl (pH 8.0)] and twice in Tris-EDTA. Immunoprecipitated chromatin was eluted with 400 µl 1% SDS and 0.1 M NaHCO for 30 min followed by addition of NaCl to a final concentration of 0.2 M and then de-cross-linked overnight at 65 C. DNA was purified with a phenol/chloroform extraction and precipitated at –20 C with sodium acetate (pH 5.2) and ethanol in the presence of glycogen. The precipitated DNA was dissolved in water and analyzed by quantitative PCR with primers against mouse Fabp4 PPRE (5'-GAGAGCAAATGGAGTTCCCAGA-3'; 5'-TTGGGCTGTGACACTTCCAC-3') and an intergenic region on mouse chromosome 15 as control (5'-TGGTAGCCTCAGGAGCTTGC-3'; 5'-ATCCAAGATGGGACCAAGCTG-3').
Differentiation assays, quantitative RT-PCR, and siRNA transfections
Differentiation assays on mouse 3T3-L1 cells and Oil-Red-O and crystal violet staining were performed as described earlier (41). Three independent samples of total RNA were isolated at different time points and subjected to quantitative PCR analysis (41). The expression of the reference gene Hprt1 was used to calculate the relative expression levels according to Vandesompele et al. (45). The sequences of the primers are as follows: murine Hprt1 sense primer 5'-TCCTCCTCAGACCGCTTTT-3' and antisense primer 5'-CCTGGTTCATCATCGCTAATC-3', murine Fabp4 sense primer 5'-GAAAACGAGATGGTGACAAGC-3' and antisense primer 5'-TTGTGGAAGTCACGCCTTT-3', murine Tip60 (isoform2) sense primer 5'-AGCCTCGGTTTTCCCTCA-3' and antisense primer 5'-TCTGAGCTGTCCTGAGAATCC-3', and murine PPARg (isoform 2) sense primer 5'-CGCTGATGCACTGCCTATGA-3' and antisense primer 5'-AGAGGTCCACAGAGCTGATTCC-3'.
For siRNA experiments, cells were grown to 80% confluency and transfected with siRNA oligonucleotides using RNAiMAX according to the manufacturer’s protocol. siRNA transfections were repeated every 3 d during the differentiation assay. At least three independent siRNA transfections followed by differentiation assays were performed. In parallel, cells were lysed in 2x SDS-PAGE sample buffer and subjected to Western blot analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
22 interacting proteins, ectopically expressed FLAG-tagged PPAR2 was immunopurified from HEK 293T cells and the associated proteins determined by mass spectrometry analysis. One of the peptides we identified corresponded to amino acids TLPIPVQITLR (position 104–114) of the Tip60 protein, a member of the MYST family of histone acetyltransferases (Fig. 1A
). At least three Tip60 isoforms exist in man and mouse as a result of alternative splicing (33). Isoform 1 results from the translation of intron 1, whereas the shorter isoforms 2 and 3 (also referred to as Tip60 and -β, or Tip60a and -b, respectively) are identical apart from the absence of the internal exon 5 in isoform 3. The peptide detected here is encoded by exon 5, indicating that Tip60 isoform 1 and/or 2 was immunopurified. Because mRNA and protein expression of isoform 2 is much higher than isoform 1 in several cell types, including NIH-3T3 mouse fibroblasts (46), all subsequent experiments were performed with Tip60 isoform 2.
|
2 and Tip60 in living cells, HA-tagged Tip60 was overexpressed in U2OS cells in the presence of FLAG-tagged PPAR2 with or without the synthetic ligand rosiglitazone. As shown in Fig. 1B
, Tip60 coimmunoprecipitated with FLAG-tagged PPAR2, both in the absence and presence of rosiglitazone. As controls, immunoprecipitations were performed on cell lysates in which neither of the proteins or only Tip60 was present. No co-immunoprecipitations were observed in these cases (Fig. 1B, lanes 1–4).
To investigate whether Tip60 functions as a transcriptional coactivator for PPAR2, the two proteins were overexpressed in U2OS cells together with a reporter construct containing three copies of the PPRE found in the rat acyl coenzyme A oxidase promoter (47) in the absence or presence of the synthetic ligand rosiglitazone (1 µM). As shown in Fig. 2A, unliganded PPAR2 activated this 3xPPRE-tk-Luc reporter only marginally, but Tip60 further potentiated this activity 4- to 6-fold. The activity of liganded PPAR2 was increased approximately 2- to 3-fold by Tip60 (see also Discussion). Because the interaction between cellular proteins and Tip60 often maps to the 52 amino acids encoded by exon 5 (28), we also performed these assays with isoform 3 of Tip60 and obtained similar results (data not shown). PPAR2, Tip60, or the combination of these proteins failed to activate a reporter that lacks functional PPREs (TK-luc), indicating that coactivation by Tip60 critically depends on DNA-bound PPAR (Fig. 2A).
|
). To investigate whether other MYST acetyltransferases could also up-regulate PPAR activity, the histone acetyltransferase bound to origin-recognition complex subunit 1 (HBO-1), males-absent on the first (MOF), and MOZ proteins were tested under the same conditions. The HBO-1 protein, which has been reported to interact with several other NRs (37, 48), as well as the MOF and MOZ proteins displayed negligible ability to potentiate PPAR2 activity (Fig. 2A). As a control, Western blot analysis was performed to show that all four MYST proteins were efficiently expressed under our experimental conditions (Fig. 2C). Finally, the effect of Tip60 on PPAR-mediated transcription was tested on a less artificial promoter. Reporter assays using a luciferase reporter construct containing the promoter of the murine fatty acid-binding protein 4 (Fabp4) gene (Fabp4, also referred to as A-FABP or aP2), which contains several functional PPREs approximately 5.5 kb upstream of the transcription start site (49, 50). As shown in Fig. 2D, PPAR-mediated activation of this reporter was also potentiated by Tip60. Western blot analysis revealed that Tip60 did not affect the expression of PPAR protein, or vice versa (Fig. 2E), indicating that Tip60 stimulates PPAR activity and not protein expression. Taken together, these findings establish Tip60 as a novel and specific coactivator for PPAR.
Tip60 coactivates PPAR via the AF1 domain
We next determined the specific domains required for the PPAR-Tip60 interaction. For this, bacterially expressed and purified GST or a series of GST-PPAR2 fusion proteins were incubated with in vitro-translated [35S]methionine-labeled Tip60 protein. As a positive control, in vitro-translated RXR, the heterodimeric partner of PPAR2, was used. Consistent with our coimmunoprecipitation experiments (Fig. 1B), Tip60 interacted with full-length PPAR2 independently of the presence of ligand (Fig. 3A). Moreover, the N-terminal 136 amino acids (Fig. 3A, lane 5) were indispensable for this interaction, whereas the C-terminal region, containing the LBD, was not required (Fig. 3A, lane 8). Coomassie staining of the gels showed equal amounts of GST fusion proteins used (data not shown). These data indicate that the interaction between Tip60 and PPAR2 is mediated through the AF1 domain and not the LBD of this NR. To corroborate these findings in living cells, an AF1 deletion mutant (
AF1) was tested in the luciferase reporter assay described above. As observed before, deletion of the AF1 region of PPAR protein resulted in significantly higher activity in the presence of ligand compared with the wild-type protein, probably as a result of loss of repressive functions encoded by the AF1 domain (11). In contrast to the wild-type PPAR2 protein, the activity of this AF1 protein was stimulated only marginally by coexpression of Tip60, suggesting that coactivation of PPAR by Tip60 critically depends on the AF1 region. To substantiate these findings, we made use of the PPAR protein, which displays extensive homology with PPAR in its DBD and LBD but not in its AF1 domain. This NR harbors significant ligand-independent activity that can still be increased by addition of the synthetic PPAR ligand WY14643 (41, 51). As shown in Fig. 3B, Tip60 failed to potentiate either the ligand-independent or ligand-dependent PPAR activity. Next, chimeric PPAR2-PPAR constructs were generated, in which the AF1 domains of these receptors were exchanged. We found that the activity of a PPAR protein with the AF1 domain of PPAR2 (PPAR2/) was stimulated by Tip60, whereas a PPAR2 protein harboring the AF1 domain of PPAR (PPAR/) was insensitive to Tip60 (Fig. 3B). Tip60 was, however, able to stimulate the activity of the PPAR1 isoform (data not shown), which lacks the first 30 amino acids present in PPAR2, indicating that Tip60 specifically requires the AF1 region between amino acids 31 and 136 of PPAR2 for coactivation.
|
2, and not the LBD, is indeed the predominant interaction interface, two additional approaches were taken. First, two LBD mutants of PPAR2 were used in which AF2 activity is affected, while ligand binding is intact (52). Mutation of the conserved lysine in helix 3 (K329A), which is part of the charge clamp, or one of the conserved leucines in helix 12 (L496A) resulted in markedly reduced transcriptional activity, either in the absence or presence of synthetic ligand (Fig. 3C). However, simultaneous overexpression of Tip60 clearly stimulated reporter activity of both mutant proteins; this effect was most pronounced in the presence of ligand (Fig. 3C). Second, we tested whether disruption of the NR box present in the C-terminal region of Tip60 (see Figs. 1A and 2B), which was shown to be required for the interaction between Tip60 and the LBD of the androgen receptor (53), affected the coactivation of PPAR by Tip60. Mutation of the LXXLL motif in Tip60 (amino acids 488–492) to LXXAA had no appreciable effect on the ability of Tip60 to stimulate PPAR2-mediated transcription (Fig. 3D), indicating that indeed interactions between the LBD and the LXXLL motif play a minor role in the interaction between PPAR2 and Tip60. Collectively, these results strongly suggest that the interaction between PPAR2 and Tip60 and coactivation of PPAR2 by Tip60 critically depend on the AF1 region, and not the LBD, of this NR.
The acetyltransferase activity of Tip60 contributes to coactivation of PPAR2
Because Tip60 contains no intrinsic transcriptional activity (54) but can acetylate histones and nonhistone proteins (32, 33), we determined whether this function is necessary for its role as a PPAR coactivator. As described before (55), Tip60 acetylates core histones with a preference for H3 and H4 (Fig. 4A). Mutation of a conserved glycine residue in the MYST domain of Tip60 into glutamic acid (G380E), analogous to the catalytic mutation of the Drosophila MYST family member mof (56), abrogates most of the histone acetyltransferase activity as well as auto-acetylation activity (Fig. 4A). We tested this Tip60 mutant for its ability to activate a PPAR-responsive reporter and found that Tip60-mediated activation of this mutant was decreased, indicating that the acetyltransferase activity of Tip60 plays a role in the coactivation of PPAR2 (Fig. 4B). In addition, mutation of the zinc finger (C263A), which plays a role in the binding of some substrates (57, 58), also resulted in reduced coactivation (Fig. 4B). These findings indicate that coactivation of PPAR2 by Tip60 depends at least in part on the ability of this protein to acetylate histone and/or nonhistone proteins.
|
2 in adipogenesis prompted us to investigate whether Tip60 would also play a role in this process. As a first experiment, the protein expression of Tip60 was studied during differentiation of 3T3-L1 cells into adipocytes. As shown in Fig. 5A, the expression levels of Tip60 decreased from 2 d before the start of the differentiation assay up to d 0, possibly caused by growth inhibition as a result of contact inhibition. Between d 1 and 9 after initiation of differentiation, a progressive increase in Tip60 levels was detected. A similar pattern of protein expression was detected for the bromodomain-containing protein 8 (Brd8) protein (Fig. 5A), which is a stable component of the Tip60 complex (29, 30, 31). Reprobing of the Western blots with an antibody against PPAR showed that this protein was expressed at detectable levels around 3 d after addition of the differentiation medium. Interestingly, quantitative RT-PCR experiments revealed unaltered expression of Tip60 mRNA during adipogenesis, indicating that the increase in Tip60 protein levels was not paralleled by increased mRNA expression (Fig. 5B). As a control, the mRNA expression of Pparg itself and Fabp4, an established PPAR target gene, was determined, and both genes were clearly induced during adipocyte differentiation (Fig. 5B).
|
target genes in intact adipocytes. For this, ChIP experiments were performed in 3T3-L1 preadipocytes and adipocytes. As shown in Fig. 5C, PPAR and Tip60 could readily be detected in the region of functional PPREs in the Fabp4 promoter in mature 3T3-L1 adipocytes (d 6). Little PPAR and Tip60 was detected on this promoter in undifferentiated cells (Fig. 5C; d 0), in which PPAR expression is low (Fig. 5A). Similar results were obtained for a second PPAR target gene, perilipin (data not shown). As a control, ChIP assays were performed on an intergenic region on chromosome 15, to which neither protein bound (Fig. 5C). Because endogenous Tip60 protein could be detected on the Fabp4 promoter in mature adipocytes, which express PPAR, but not in preadipocytes, which lack detectable PPAR expression (Fig. 5A), these data suggest that Tip60 requires the PPAR protein for its recruitment to PPAR target genes.
To address the relevance of Tip60 in adipogenesis, the expression of this protein was reduced using siRNA-mediated knockdown. Two independent siRNA oligonucleotides (siRNA oligonucleotides 1 and 2) reduced the expression of the Tip60 protein, with siRNA oligonucleotide 2 being the most efficient, whereas siRNA oligonucleotide 4 hardly affected Tip60 protein levels when compared with cells treated with nontargeting siRNA oligonucleotides (Fig. 5D). Knockdown of Tip60 expression with siRNA oligonucleotides 1 and 2 resulted in impaired differentiation of 3T3-L1 cells into adipocytes, as illustrated by staining of triglycerides with Oil-Red-O at d 6 of differentiation (Fig. 5E). Similar findings were obtained with an siRNA oligonucleotide pool directed against Tip60 (data not shown). To confirm this reduction in adipocyte differentiation independently, the protein expression of Fabp4 was determined. Although Fabp4 protein was clearly induced at d 6 of differentiation (Fig. 5D, lanes 1 and 2), treatment of cells with siRNA oligonucleotide 2 and to a lesser extent siRNA oligonucleotide 1, blunted this response (Fig. 5D). Treatment of cells with siRNA oligonucleotide 4, which failed to reduce Tip60 expression (Fig. 5D), had no effect on triglyceride accumulation (Fig. 5E) or Fabp4 protein expression (Fig. 5D). Because our experiments indicate that Tip60 contributes to the adipogenic action of PPAR by targeting the N-terminal AF1 domain (Fig. 2) and that the protein levels of this MYST acetyltransferase are rate-limiting during adipogenesis (Fig. 5), we conclude that Tip60 qualifies as a novel adipogenic factor.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is dictated by the set of proteins with which this NR is able to interact. Here we report that the MYST acetyltransferase Tip60 functions as a novel and specific coactivator for PPAR. Our experiments show that, unlike most transcriptional coregulators, Tip60 interacts with the N-terminal AF1 domain of PPAR. Several lines of evidence indicate that this AF1 region, which is poorly conserved between the PPAR, -, and -β/ proteins, plays an essential role in determining the specific gene expression profile elicited by the different isotypes (19, 44, 59). First, gene expression studies in cell lines expressing chimeric PPAR-PPAR proteins have revealed that the AF1 regions are responsible for PPAR-selective gene expression (19). Second, the AF1 regions restrict the transcriptional response, because deletion of these regions resulted in less selective gene expression profiles (e.g. activation of PPAR-specific target genes by PPARAF1) (19). Finally, exchanging the AF1 region of PPAR with the corresponding region of PPAR converts this protein from a nonadipogenic to an adipogenic transcription factor (18, 19). Because Tip60 targets the AF1 domain, this acetyltransferase is, together with PGC2 (18), among the first proteins identified to date that may play a role in determining PPAR isotype-selective gene expression.
Tip60 has previously been implicated in signaling through other NRs, including the androgen receptor, glucocorticoid receptor, estrogen receptor, and retinoic acid receptor (RAR)-related orphan receptor (53, 60, 61, 62, 63). In the case of androgen receptor, glucocorticoid receptor, and estrogen receptor, Tip60 was reported to interact directly with these class I NRs through their LBDs, and the interactions required the presence of the respective ligands as well as an intact LXXLL motif in the Tip60 protein (53). The failure to detect interactions between Tip60 and the LBDs of several class II NRs (thyroid hormone receptor, vitamin D receptor, and RXR), together with the lack of up-regulation of these receptors by Tip60 led to the hypothesis that Tip60 was a coactivator specific for class I NRs (53). Our experiments now indicate that Tip60 can potentiate the activity of the class I NR PPAR but that this occurs through an alternative molecular mechanism. In contrast to the Tip60 class I NR interactions, the Tip60-PPAR interaction can occur in the absence of ligand and is independent of the interaction interface defined by the charge clamp in the LBD and the LXXL motif in the coactivator molecule. Furthermore, using chimeric NRs and in vitro protein-protein interaction assays, we demonstrate that Tip60 targets the N-terminal AF1 region of PPAR. We conclude, therefore, that the same coactivator (or coactivator complexes) can play a role in the activation process of closely related transcription factors through distinct molecular mechanisms. Our data suggest that Tip60 is at least in part responsible for releasing the repressive functions encoded by the AF1 domain of PPAR (11).
Tip60 is part of a large multiprotein complex, which is conserved between yeast and humans (27, 29, 30, 31). Our results indicate that the acetyltransferase activity of Tip60 plays a role in PPAR-mediated transcription, because a catalytic mutant of Tip60 (G380E) displayed reduced ability to potentiate the transcriptional activity of PPAR. Tip60 can acetylate both histones and nonhistone proteins (27, 33), including the AR in its hinge region (60). Whether histones, PPAR itself, or other proteins in the transcription complex are the main acetylation substrates in our experimental setting remains to be established. The Tip60 complex also contains the Brd8 protein, a coactivator for PPAR/RXR, which contains LXXLL motifs (64, 65). The Brd8 protein may stabilize interactions between the Tip60 complex and the PPAR/RXR heterodimers by binding to the AF2 region. Such additional protein-protein interactions may account for the increase in ligand-dependent activity of PPAR by Tip60, whereas we found the protein-protein interaction between PPAR and Tip60 to be independent of ligand. Alternatively, Tip60 may cooperate with other coregulators outside the Tip60 complex, like transcriptional intermediary factor 2 (66), to potentiate PPAR-mediated transcription in a ligand-dependent fashion.
Although lysine modifications are clearly important mechanisms to regulate protein function (67), very little is known about the mechanism that regulates the activity and/or expression of the modifying enzymes, like acetyltransferases. As is shown here, Tip60 protein levels increase during adipogenesis and are clearly not regulated at the mRNA level but are probably controlled posttranslationally via a yet unidentified mechanism. Furthermore, it has been published that Tip60 acetyltransferase activity is regulated during the cell cycle by phosphorylation (46, 68), and it will therefore be of much interest to investigate whether the intrinsic acetyltransferase activity of Tip60 is also regulated in cellular processes like adipogenesis.
In conclusion, besides ligand binding, modulation of AF1 activity through interactions with coregulators and/or posttranslational modifications provides an important extra layer in the regulation of PPAR activity. Given the central role of PPAR in glucose and lipid metabolism, the enzymes involved in regulating PPAR activity can be considered as potential therapeutic targets for the treatment of type 2 diabetes. This underscores the importance of further elucidating the molecular mechanisms that regulate the transcriptional activity of PPAR.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: The authors of this manuscript have nothing to declare.
First Published Online December 20, 2007
1 A.B.B., L.G., and N.H. contributed equally to this work. 
Abbreviations: AF1, N-terminal transactivation domain; Brd8, bromodomain-containing protein 8; CBP, cAMP response element-binding protein-binding protein; ChIP, chromatin immunoprecipitation; DBD, DNA-binding domain; Fabp4, fatty acid-binding protein 4; GST, glutathione S-transferase; HBO, histone acetyltransferase bound to origin-recognition complex subunit 1; LBD, ligand-binding domain; MOF, males-absent on the first; MOZ, monocyte leukemia zinc-finger; NaDOC, sodium deoxycholate; NR, nuclear receptor; PPAR, peroxisome proliferator-activated receptor; PPRE, PPAR-responsive element; RXR, retinoic acid X receptor; siRNA, small interfering RNA; Tip60, Tat-interacting protein 60; TK, thymidine kinase.
Received July 17, 2007.
Accepted for publication December 7, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in energy metabolism and vascular homeostasis. J Clin Invest 116:571–580[CrossRef][Medline]: a dagger in the heart of the metabolic syndrome. J Clin Invest 116:590–597[CrossRef][Medline] and human metabolic disease. J Clin Invest 116:581–589[CrossRef][Medline]. Cell 123:993–999[CrossRef][Medline]: a master regulator of adipogenesis. Mol Biol Cell 6:1343 is required for placental, cardiac, and adipose tissue development. Mol Cell 4:585–595[CrossRef][Medline] mediates high-fat diet-induced adipocyte hypertrophy and insulin resistance. Mol Cell 4:597–609[CrossRef][Medline] is required for the differentiation of adipose tissue in vivo and in vitro. Mol Cell 4:611–617[CrossRef][Medline]2, a lipid-activated transcription factor. Cell 79:1147–1156[CrossRef][Medline]. Int J Obes (Lond) 29(Suppl 1):S31–S35/RXR crystal structure reveals the molecular basis of heterodimerization among nuclear receptors. Mol Cell 5:545–555[CrossRef][Medline]. Nature 395:137–143[CrossRef][Medline] (PPAR). J Biol Chem 270:12953–12956. EMBO J 18:3676–3687[CrossRef][Medline]2 co-activator that binds through its SCAN domain. Biochem J 370:719–727[CrossRef][Medline]2 in a ligand-independent and -dependent manner, respectively. J Biol Chem 274:7681–76882 stability and activity by SUMOylation. Obes Res 12:921–928[Medline] is negatively regulated by SUMO conjugation in the amino-terminal domain. Genes Cells 9:1017–1029 is modulated by SUMO-1 modification. J Biol Chem 279:29551–29557 function through the TAK1/TAB1/NIK cascade. Nat Cell Biol 5:224–230[CrossRef][Medline] function through mutation of a conserved salt bridge (R425C) in familial partial lipodystrophy. Mol Endocrinol 21:1049–10652: tissue-specific regulator of an adipocyte enhancer. Genes Dev 8:1224–1234) function. J Biol Chem 275:3733–3736 coordinates reciprocal signaling in cerebellar development through sonic hedgehog and calcium-dependent pathways. Neuron 40:1119–1131[CrossRef][Medline] directs ordered, cyclical, and combinatorial recruitment of cofactors on a natural target promoter. Cell 115:751–763[CrossRef][Medline]/RXR heterodimers. Mol Endocrinol 13:1695–1703This article has been cited by other articles:
 |
|
 |
|
  A. Bugge, L. Grontved, M. M. Aagaard, R. Borup, and S. Mandrup The PPAR{gamma}2 A/B-Domain Plays a Gene-Specific Role in Transactivation and Cofactor Recruitment Mol. Endocrinol., June 1, 2009; 23(6): 794 - 808. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
5500 bp from transcription start site). Primers corresponding to an intergenic region on mouse chromosome 15 were used as a negative control. Results are shown as percent recovery relative to chromatin input. Results are representative of at least three independent experiments. D, 3T3-L1 cells were transfected with either control or Tip60 siRNA oligonucleotides 2 d before differentiation initiation. siRNA transfection was repeated every 3 d. At d 6 of differentiation, lysates of 3T3-L1 cells treated with control siRNA or Tip60 siRNA oligonucleotides were subjected to Western blot analysis, using antibodies against Tip60, Fabp4, and tropomyosin. Undifferentiated cells (d 0) were used as a control. E, 3T3-L1 cells treated as described for D were fixed at d 0 or 6 of differentiation and stained for triglycerides using Oil-Red-O, and nuclei were stained with crystal violet. Pictures are representative of three independent experiments.