Transcriptional Activity of Peroxisome Proliferator-activated Receptor (cid:1) Is Modulated by SUMO-1 Modification*

Covalent modification of many transcription factors with SUMO-1 is emerging as a key role of trans-activa-tional regulation. Here, we demonstrate that peroxisome proliferator-activated receptor (PPAR) (cid:1) , which is a li-gand-activated nuclear receptor, is modified by SUMO-1. Sumoylation of PPAR (cid:1) mainly occurs at a lysine residue within the activation function 1 domain. Furthermore, we show that the PIAS family proteins, PIAS1 and PIASx (cid:2) , function as E3 ligases (ubiquitin-protein isopeptide ligase) for PPAR (cid:1) . PPAR (cid:1) interacts directly with PIASx (cid:2) in a ligand-independent manner. Analysis using a PPAR (cid:1) mutant with a disrupted sumoylation site shows that modification of PPAR (cid:1) by SUMO-1 represses its transcriptional activity. Interestingly, PIASx (cid:2) and Ubc9 enhance the transcriptional activity of PPAR (cid:1) independent of PPAR (cid:1) sumoylation. Furthermore, PPAR (cid:1) ligand-in-duced apoptosis in a human hepatoblastoma cell line, HepG2, is significantly enhanced by ectopic production of the sumoylation-mutant PPAR (cid:1) . These results suggest that the PPAR (cid:1) -dependent transactivation pathway seems to be modulated by SUMO-1 modification and may serve as a novel target for apoptosis-induction therapy in cancer cells.

The role of PPAR␥ in adipogenesis has been extensively studied. Many adipocyte-specific genes, such as adipocytokines, contain PPAR␥-responsible elements in their promoter and/or upstream enhancer regions (7)(8)(9)(10). PPAR␥ plays a role as a central transcription factor in cellular differentiation and lipid accumulation during adipogenesis. Recent investigations demonstrate that treatment of a variety of human cancer cell lines with PPAR␥ ligands leads to growth inhibition and apoptosis (2,(11)(12)(13). The use of PPAR␥ ligands in the treatment of cancer is a potentially promising nontoxic and selective chemotherapeutic approach, and consequently, increased understanding of the mechanisms of PPAR␥ in tumor suppression is needed.
Post-translational modifications regulate the function of many proteins. In the case of PPAR␥, transcriptional activity is reduced by mitogen-activated protein kinase-induced phosphorylation of serine residue 112 (14 -16). Knock-in mice expressing PPAR␥ with a Ser 3 Ala mutation at this residue exhibit preserved insulin sensitivity in the setting of diet-induced obesity by changing fat cell size, generation of adiponectin, and increasing the amount of free fatty acid levels in serum (17).
Recently, a number of ubiquitin-like proteins (Ubl) have been identified that are covalently linked to lysine residues in target proteins (18,19). One Ubl, SUMO-1, also known as PIC1, UBL1, sentrin, GMP1, and SMT3, is an 11-kDa protein that is structurally homologous to ubiquitin (20 -22). SUMO-1 modification plays an important role in altering the function of modified proteins, including transcriptional activation, nuclear localization, and increased turnover (23)(24)(25). SUMO-1 is conjugated to proteins through a series of enzymatic steps (26). Initially, the ATP-dependent formation of a thioester bond between SUMO-1 and the E1 enzyme complex (SAE1⅐Uba2) is formed, and SUMO-1 is then transferred to the E2-conjugating enzyme Ubc9. Finally, SUMO-1 is conjugated from Ubc9 directly to a lysine residue of target proteins in vitro. The E3 ligase that conjugates SUMO-1 to target molecules in vitro and in vivo has only recently been identified (27)(28)(29)(30). One group of such E3 ligases, protein inhibitor of activated STAT (PIAS) family proteins, homologous to the yeast Siz family protein, has a conserved RING finger domain that regulates transactivation of many transcription factors including STAT1 (31,32), lymphoid enhancer factor-1 (33), and nuclear receptors (34,35) by conjugating SUMO-1.
To understand the molecular mechanisms of PPAR␥ transcriptional function through post-translational modifications, we explored the possible modification of PPAR␥ by SUMO-1. In this paper we demonstrate that PPAR␥ is a target for SUMO-1 modification, and PIAS proteins function as E3 ligases for SUMO-1 modification. The main sumoylation site of PPAR␥ was mapped to a lysine residue at position 107, located in close proximity to the regulatory Ser-112. Sumoylation at this lysine residue reduced PPAR␥-dependent transcriptional activation * This work was supported by grants-in-aid for Research on Hepatitis and for the second-term comprehensive 10-year strategy for cancer control from the Ministry of Health, Labor, and Welfare, grants-in-aid for scientific research from the Ministry of Education, Culture, Sports, Science, and Technology, grants-in-aid of Research for the Future from the Japanese Society for the Promotion of Science, and by the Program for Promotion of Fundamental Studies in Health Science of the Organization for Pharmaceutical Safety and Research (OPSR) of Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  significantly. However, reporter gene assays suggested that PIAS proteins enhanced the transcriptional activity of PPAR␥ by a mechanism independent of PPAR␥ sumoylation. We also observed that a PPAR␥ sumoylation mutant displayed enhanced ligand-induced apoptosis in a human hepatoblastoma cell line, HepG2, which suggests a possible new target for cancer therapy.

MATERIALS AND METHODS
Cell Culture, Transfection, and Luciferase Reporter Assay-Cells were maintained in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 10% fetal calf serum, 100 units/ml penicillin, and 100 g/ml streptomycin. Transfection was performed using FuGENE 6 (Roche Applied Science) according to the manufacturer's instructions. Luciferase activity was normalized to Renilla luciferase activity derived from co-transfected pRL-CMV-Luc (Promega). All reporter assays were performed in triplicate, and S.E. are denoted by bars in the figures.
Antibodies and Reagents-Rat anti-HA (3F10; Roche Applied Science), mouse and rabbit anti-FLAG (Sigma), and mouse anti-GFP (Clontech) antibodies were purchased commercially. Horseradish peroxidase-linked goat antibodies to rat IgG were from Jackson Immu-noResearch Laboratories. Horseradish peroxidase-linked goat antibodies to mouse or rabbit IgG were from Amersham Biosciences. Rosiglitazone was purchased from Alexis Biochemicals.
Immunoprecipitations-HEK-293T cells (1 ϫ 10 5 per 6-cm-diameter dish) were transfected using FuGENE 6 (Roche Applied Science) according to the manufacturer's instructions. After incubation, cells were lysed in 1 ml of lysis buffer (25 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% Nonidet P-40, 0.1% SDS, 0.5% sodium deoxycholate, 1 mM dithiothreitol, 5 mM EDTA, 10 mM N-ethylmaleimide, 200 M indole-3-acetic acid, and a complete protease inhibitor mixture tablet (Roche Applied Science) for sumoylation analysis or radioimmune precipitation assay buffer (25 mM Tris-HCl, pH 8.0, 125 mM NaCl, 0.1% Nonidet P-40, 1 mM dithiothreitol, 1 mM EDTA, and a complete protease inhibitor mixture tablet) for co-immunoprecipitation analysis. Cell debris was removed by centrifugation for 15 min. Lysates were first cleared with protein G beads for 30 min followed by incubation with antibodies for 1 h at 4°C. Finally, the antibody complexes were captured with protein G beads for 1 h. Beads were washed four times with the same buffer, and immunoprecipitates were eluted and analyzed by Western blot.
Detection and Measurement of Apoptosis and Indirect Immunofluoresence Observation-HepG2 cells were grown to subconfluency on 8-well Lab-Tec Chamber (NUNC) in Dulbecco's modified Eagle's medium with 10% fetal calf serum. Cells were transfected with 500 ng of pcDNA3-FLAG-PPAR␥1 or -PPAR␥1(K/R1) expression vectors. Fortyeight hours after transfection, rosiglitazone was added to culture medium to a final concentration at 1 M for 24 h. Cells were fixed at room temperature with 3.7% formaldehyde for 3 min and then permeabilized with 0.5% Triton X-100 in phosphate-buffered saline for 5 min. After blocking with 3% bovine serum albumin and 0.1% Triton X-100 in phosphate-buffered saline, the cells were incubated with anti-HA antibody for 1 h at 37°C and stained with Alexa Fluor 568 anti-rat secondary antibody (Molecular Probes) for 1 h at room temperature. Apoptotic cells were detected by using the in situ cell death detection kit, Fluorescein, following the manufacturer's instructions (Roche Applied Science). The ratio of apoptotic cells was quantitated by analysis of DNA fragmentation using the cell death detection enzyme-linked immunosorbent assay according to the manufacturer's instructions (Roche Applied Science).

RESULTS
PPAR␥ Is a Substrate for SUMO-1 Modification-SUMO-1 modification of certain transcription factors including nuclear hormone receptors is known to affect transcriptional activity, and consequently, we wished to address whether PPAR␥ was a substrate for SUMO-1 modification. We first examined whether PPAR␥ is modified by SUMO-1 in cells transiently expressing FLAG-PPAR␥2 and HA-SUMO-1. Western blot analysis using anti-FLAG antibody revealed the presence of FLAG-tagged PPAR␥2 in all cells transfected with the plasmid expressing FLAG-PPAR␥2. When HA-SUMO-1 was co-expressed, additional slower migrating bands were detected by the FLAG antibody (Fig. 1A, left panel, lane 3). Moreover, to determine whether these slower migrating bands represent PPAR␥2 conjugated to SUMO-1, the membrane was re-probed with anti-HA antibody, which detects proteins conjugated to HA-SUMO-1. The result showed that the slower migrating forms of PPAR␥2, about 90 and 130 kDa, were indeed sumoylated (Fig. 1A, right panel, lane 6). These data suggest that PPAR␥ is modified by SUMO-1 at least two sites. We next examined whether a specific PPAR␥ ligand, rosiglitazone, affected PPAR␥ sumoylation. As shown in Fig. 1B, SUMO-1conjugated PPAR␥2 was detected in cells co-producing SUMO-1. In lysates prepared from cells treated with rosiglitazone, the amount of SUMO-1-conjugated PPAR␥2 was lower than in mock-treated cells (lane 2 and 4), suggesting that PPAR␥ ligand negatively regulates SUMO-1 conjugation to PPAR␥. Two lysine residues, Lys-107 and Lys-347, in the AF1 and AF2 domains, respectively, of PPAR␥2 conform to the proposed consensus motif KX(D/E) (where is a hydrophobic amino acid residue, X represents any residue, and D or E is an acidic residue) for SUMO-1-conjugating sites (18,20) (Fig. 1C).
To determine whether these lysine residues are targets for sumoylation, mutants with lysine to arginine substitutions, K107R (K/R1) and K347R (K/R2) as shown in Fig. 1C, were generated and analyzed for sumoylation. Two bands migrating slower than the original band were detected with almost the same intensity as cells producing both wild type PPAR␥1, PPAR␥2, and the mutant PPAR␥2(K/R2) (Fig. 1D). In contrast, the slowest band disappeared in cells producing the mutant PPAR␥2(K/R1), and moreover, the amount of the slower migrating band was also reduced in the cells expressing the mutant PPAR␥2(K/R1) (Fig. 1D, lane 2). These results imply that Lys-107 is the major site for sumoylation, and this site may function as the master switch of sumoylation because mutation of this lysine residue greatly impaired sumoylation for PPAR␥. The fact that mutation of lysine residue at 347 did not affect the efficiency of sumoylation suggests the presence of lysine residues other than those not in the consensus motif for SUMO-1 modification in PPAR␥.
PIAS Family Proteins Act as E3 Ligases for PPAR␥ Sumoylation-Recent studies indicated that members of the PIAS family enhanced sumoylation of many proteins including nuclear receptors (25,34,35). Therefore, we investigated whether PIAS family proteins function as E3 ligases for PPAR␥. We generated a mutant of PIASx␤, PIASx␤(C/S), in which the conserved cysteine residue at position 353 within the RING finger domain was changed to serine. This mutant was not able to interact with Ubc9 (data not shown) and completely lacked E3 ligase function (36). SUMO-1 conjugation to PPAR␥2 was analyzed in cells producing either wild-type PIASx␤, PIASx␤(C/S), or PIAS1. Small amounts of SUMO-1-conjugated PPAR␥2 were detected in cells expressing only HA-SUMO-1 ectopically (Fig. 2, lane 2). PPAR␥2 sumoylation was enhanced by exogenous expression of PIAS1 and PIASx␤ but not PIASx␤(C/S) (lanes 3-5). These findings indicate that PIAS family proteins function as E3 ligases for PPAR␥2.
Next, to investigate the association of PPAR␥2 with PIASx␤, we employed a GST pull-down analysis. Full-length PIASx␤ expressed in bacteria as a GST fusion protein was coupled to glutathione S-Sepharose beads, and this complex was incubated with in vitro translated 35 S-labeled PPAR␥2 in buffer with or without rosiglitazone, a specific ligand for PPAR␥. As alone. To analyze the physical interaction of PPAR␥2 with PIASx␤ in cells, a co-immunoprecipitation experiment was conducted using extracts from HEK-293T cells co-expressing HA-PPAR␥2 and FLAG-PIASx␤ treated with or without rosiglitazone. Proteins precipitated with anti-HA antibody were resolved by SDS-PAGE and Western blot using anti-FLAG was conducted. A FLAG-reactive species was detectable in the complex precipitated with the anti-HA antibody. The PIASx␤ in the immunocomplex was increased in cells treated with rosiglitazone, indicating that the binding efficiency of PPAR␥2 and PIASx␤ was significantly enhanced by treatment with this ligand (Fig. 3B, lanes 2 and 3).
SUMO-1 Conjugation to PPAR␥ Represses the Transcriptional Activity of PPAR␥-To evaluate the effect of PPAR␥ sumoylation on its transcriptional function, we analyzed the effects of PPAR␥ sumoylation on expression of the p4xPPRE-Luc reporter gene in which the luciferase gene is driven by a PPAR-responsive promoter. NIH3T3 cells were transfected with various combinations of plasmids expressing wild-type PPAR␥1, wild-type PPAR␥2, PPAR␥1(K/R1), PPAR␥2(K/R2), and mRXR␣, a component of a heterodimeric complex with PPAR␥, together with p4xPPRE-Luc. Cells were then treated with or without rosiglitazone. Additional production of mRXR␣ in cells enhanced reporter activity by PPAR␥. Reporter activity was enhanced by treatment with rosiglitazone, and this was FIG. 4. Mutant PPAR␥ has higher transcriptional activity than wild-type PPAR␥. A, NIH3T3 cells were cotransfected with 25 ng of p4xPPRE-Luc and 100 ng of each plasmid expressing HA-wild-type PPAR␥2 (␥2(WT)), -mutant PPAR␥2 with substitution of Lys-107 to Arg (␥2(K/R1)), -wild-type PPAR␥1 (␥1(WT)), -mutant PPAR␥1 with substitution of Lys-77 to Arg (␥1(K/R1)), mRXR␣, or an empty plasmid (Ϫ). Twenty-four hours after transfection, cells were treated with or without 5 M rosiglitazone. Luciferase activities were then measured 18 h after treatment. Open bars denote no treatment, and closed bars indicate rosiglitazone treatment. The activity in control cells was arbitrarily given a value of 1, and the activities in the other cells were relative to the value of control cells. B, HEK-293T cells were cotransfected with 25 ng of GAL4-luciferase reporter plasmid (pGAL4-Luc) and 100 ng of plasmids expressing fusion proteins of the GAL4 DNA binding domain to PPAR␥1 (␥1(WT)), PPAR␥1(K/R1) (␥1(K/R1)), PPAR␥2 (␥2(WT)), PPAR␥2(K/R1) (␥2(K/R1)), or an empty plasmid (Ϫ). Twenty-four hours after transfection cells were treated with or without 5 M rosiglitazone. Luciferase activities were then measured 18 h after treatment. Open bars denote no treatment, and closed bars indicate rosiglitazone treatment. The activity in control cells was arbitrarily given a value of 1, and the activities in the other cells were relative to the value of control cells.

FIG. 3. Association of PIASx␤ with PPAR␥2 in vitro and in vivo.
A, GST and GST-PIASx␤ fusion proteins were immobilized on glutathione-Sepharose beads and incubated with 35 S-labeled PPAR␥2 translated in vitro in pull-down buffer containing 5 M rosiglitazone as indicated. The pull-down complexes were analyzed by SDS-PAGE followed by an image analyzer. The Input lane represents 20% of total volume of whole cell lysates used for pull-down assay. B, HEK-293T cells were transfected with 2 g of plasmid expressing FLAG-PIASX␤ together with (ϩ) or without (Ϫ) 2 g of plasmid expressing HA-PPAR␥2. Twenty-four hours after transfection cells were treated with (ϩ) or without (Ϫ) 5 M rosiglitazone for 12 h. Cell extracts were then prepared and subjected to immunoprecipitation using anti-HA antibody (IP). The immunoprecipitates were subjected to SDS-PAGE and analyzed by immunoblot (IB) with anti-FLAG antibody (top panel) or anti-HA antibody (second panel). The protein levels of FLAG-PIASx␤ in each cell lysate are indicated (third panel). highest in cell lysates containing PPAR␥2(K/R1) (Fig. 4A). The fact that transcriptional activity of PPAR␥2 was higher than that of PPAR␥1 was in good agreement with a previous report (38). Next, to analyze the direct effect of the SUMO-1 conjugationdependent transcriptional activity of PPAR␥, we utilized PPAR␥ fused with GAL4 to analyze gene expression from pGL2-Luc containing five GAL4 binding sites in the promoter region. GAL4-PPAR␥1(K/R1) and -PPAR␥2(K/R1) showed about 5-fold higher luciferase activities than the activity observed by GAL4-wild-type-PPAR␥1 and GAL4-wild-type-PPAR␥2, respectively, in a ligand-dependent manner (Fig. 4B). Taken together, these data suggest that sumoylation of PPAR␥ represses the transcriptional activity of PPAR␥ itself.
Both Ubc9 and PIASx␤ Enhance PPAR␥-dependent Transactivation-To further investigate the transcriptional role of sumoylated PPAR␥, we examined the effects of Ubc9, an essential factor for sumoylation, on PPAR␥-dependent transcription. GAL4-fused wild-type PPAR␥2 and -PPAR␥2(K/R1) were expressed in HEK-293T cells with increasing amounts of Ubc9 (Fig. 5A). Co-production of Ubc9 enhanced transcription by PPAR␥2 and PPAR␥2(K/R1) in a dose-dependent fashion. We next examined the effects of PIASx␤ and PIASx␤(C/S) on the transcriptional activation of PPAR␥2. Luciferase activities were significantly enhanced by the co-production of PIASx␤. However, co-production of PIASx␤(C/S) only slightly enhanced the activity (Fig. 5B). Similar results were also observed for PPAR␥1 (data not shown). These data suggest that the SUMO-1 conjugation activity of Ubc9 and PIAS positively regulates PPAR␥-mediated transactivation. The observation that the transcriptional activity of PPAR␥2(K/R1) was not only significantly enhanced by co-production of Ubc9 but also by PIASx␤ suggests that Ubc9 and PIAS proteins function as positive regulators for PPAR␥-dependent transcription possibly through SUMO-1 conjugation of a factor(s) other than PPAR␥ involved in transcriptional regulation.
Ligand-induced Apoptosis by PPAR␥ Is Enhanced in Cells Producing PPAR␥(K/R1)-Recent studies have demonstrated that specific ligands for PPAR␥ inhibit cell growth and induce apoptosis in several human cancer cells (2,(11)(12)(13). PPAR␥ activation seems to be important for inducing apoptosis in some cells. However, the molecular mechanisms of PPAR␥-dependent apoptosis, particularly the relationship between the transcriptional activity of PPAR␥ and apoptosis, remain unclear. To investigate the effect of sumoylation on PPAR␥-dependent apoptosis, we compared the apoptotic potential of wild-type PPAR␥1 to that of PPAR␥1(K/R1) in HepG2 cells. Plasmids expressing FLAG-PPAR␥1 or -PPAR␥1(K/R1) were transfected into HepG2 cells, and 48 h after transfection cells were treated with 1 M rosiglitazone for 24 h. PPAR␥ expression in cells and apoptotic cells were detected by immunostaining and terminal dUTP nick-end labeling assay, respectively. Approximately 5% of PPAR␥1-transduced cells became terminal dUTP nick-end label-positive, which stained strongly by the anti-FLAG antibody. In contrast, ϳ40% of PPAR␥1(K/R1)-transduced cells became terminal dUTP nick-end label-positive, and almost all cells expressed high levels of PPAR␥1(K/R1) (Fig. 6A). The numbers of apoptotic cells producing PPAR␥1 or PPAR␥1(K/ R1) were verified by measurement of the accumulation of fragmented nucleosomes. Ligand-induced apoptosis is significantly enhanced when PPAR␥1(K/R1) was produced in cells (Fig. 6B). These results suggest that transcriptional activation of PPAR␥ is involved in enhancing ligand-mediated apoptosis. DISCUSSION In this study we showed that sumoylation of PPAR␥ significantly affected its transcriptional activity. PPAR␥ was predominantly modified by SUMO-1 at Lys-107 within the AF1 domain. Our result suggest that there is a lysine residue(s) in addition to Lys-107 targeted for sumoylation that is likely to lie in a non-consensus SUMO-1 conjugation motif, because mutational analysis of the lysine residues lying in other consensus SUMO-1 conjugation motifs in this protein did not affect SUMO-1 conjugation (Fig. 1D). Because mutation of Lys-107 reduced SUMO-1 conjugation of PPAR␥ severely, Lys-107 is the primary site for modification. Similar observations of the presence of hierarchic lysine residues for SUMO-1 conjugation were reported in other proteins such as promyelocytic leukaemia protein (39), androgen receptor (40), aryl hydrocarbon receptor (41), and DNA topoisomerase I (42). PIAS1 and PIASx␤ acted as E3 ligase factors for SUMO-1 conjugation to PPAR␥ (Fig. 2). We also showed that PIASx␤ associated with PPAR␥2 in vitro and in vivo in a ligand-independent manner, but the association was enhanced by the presence of the ligand in vivo (Fig. 3). Interestingly, ligand treatment led to a reduction in the amount of SUMO-1 conjugated to PPAR␥2 (Fig. 1B). Because of conformational alteration of nuclear receptors, association of co-activator complexes with nuclear receptor seems to be regulated by specific ligands. Thus, it is likely that PPAR␥ sumoylation is suppressed by the association of the co-activator complex with the ligated PPAR␥, in which the sumoylation sites of PPAR␥ may be masked, and/or the E3 ligase activity of PIAS proteins may be blocked.
Using a reporter gene assay, we demonstrated that the transcriptional activity of PPAR␥ was negatively regulated by sumoylation. It has been reported that phosphorylation of Ser-112, adjacent to the sumoylation site, as revealed by this work, on PPAR␥ by mitogen-activated protein kinase significantly inhibited both ligand-independent and ligand-dependent transcriptional activation by PPAR␥ (15). Mutation analysis of the phosphorylation site revealed that this phosphorylation-mediated transcriptional repression was not due to a reduced capacity to make PPAR␥⅐RXR␣ complexes or the impairment of recognition of its DNA binding site (14). An AF-1 domain of PPAR␥ may be negatively regulated by phosphorylation and sumoylation. Alternatively, SUMO-1-conjugated PPAR␥ may recruit the transcriptional repressor complex by providing a novel interaction site. Recently it has been shown that sumoylation of the ETS domain transcription factor, Elk-1, results in the recruitment of histone deacetylase activity to promoters (43). Similarly, SUMO-1-conjugated PPAR␥ may recruit additional cellular factors that repress PPAR␥-dependent transcription.
Our data clearly showed that Ubc9 enhanced the transcriptional activities of both PPAR␥ and PPAR␥(K/R1), possibly by a mechanism independent of SUMO-1 conjugation to PPAR␥. PIASx␤ also enhanced PPAR␥ activity through a RING finger domain-dependent mechanism. Thus, it seems that ectopically produced PIASx␤ regulates PPAR␥-mediated transactivation through not only sumoylation of PPAR␥ itself but also in the conjugation of SUMO-1 to another cellular factor(s) involved in transcriptional regulation. In agreement with our observations, key molecules in the SUMO-1 conjugation system, including SUMO-1, Ubc9, and PIAS, have been shown to modulate the transcriptional activities of p53 (44), androgen receptor (40), aryl hydrocarbon receptor (41), and lymphoid enhancer factor-1 (33) even when these target molecules lacked a major sumoylation site(s) by mutation. Moreover, it has been shown that Ubc9 modulates the transcriptional activity of ETS-1-and TEL-independent of its E2 enzymatic activity (45,46). In view of these reports, a mechanism(s) other than the direct SUMO-1 conjugation to PPAR␥ by Ubc9 and PIASx␤ seem to be important for the regulation of transactivation of PPAR␥. Further studies to clarify the molecular basis of the transcriptional activation of PPAR␥-dependent transcription by Ubc9 and PIAS should provide significant insight.
A role for PPAR␥ in adipogenesis is well characterized. In addition, a novel function of PPAR␥ in tumor pathogenesis has been reported recently, which includes PPAR␥ ligand-dependent growth inhibition and/or apoptosis in a variety of human cancer cells (2,(11)(12)(13). Several studies have demonstrated that induction of apoptosis was accompanied by the up-regulation of several pro-apoptotic genes, Bax and caspase-3 and -9, and down-regulation of the anti-apoptotic gene Bcl-2 (47,48), suggesting that transactivation of PPAR␥ is likely to regulate expression of apoptosis modulators at the transcriptional level and contribute as an important modulator of tumor suppression. In this study we demonstrated that the trans-activation function of PPAR␥ was up-regulated by mutation of Lys-107, the major target for sumoylation in PPAR␥. BecauseHepG2 cells expressing this mutant form of PPAR␥ displayed enhanced PPAR␥ ligand-dependent apoptosis, the increased transactivation function of PPAR␥ seems to play an important role in inducing apoptosis. Sumoylation regulates the transacting function of PAPR␥, which could play a role in the regulation of apoptosis. We suggest here that the sumoylation of PPAR␥ may be a good target for a novel therapeutic agent in cancer cells.