Transcriptional Coactivator PRIP, the Peroxisome Proliferator-activated Receptor γ (PPARγ)-interacting Protein, Is Required for PPARγ-mediated Adipogenesis

Nuclear receptor coactivator PRIP (peroxisome proliferators-activated receptor (PPARγ)-interacting protein) appears to serve as a linker between cAMP response element-binding protein-binding protein (CBP/p300)anchored and PBP (PPARγ-binding protein)-anchored coactivator complexes involved in the transcriptional activity of nuclear receptors. Disruption of PRIP and PBP genes results in embryonic lethality between embryonic day 11.5 and 12.5 (postcoitum), indicating that PRIP and PBP are essential and nonredundant coactivators. Both PRIP and PBP were initially identified as PPARγ coactivators, suggesting a role for these molecules in PPARγ-induced adipogenesis. PBP–/– mouse embryonic fibroblasts fail to exhibit PPARγ-stimulated adipogenesis indicating that PBP is a downstream regulator of PPARγ-mediated adipogenesis. We now show that PRIP–/– mouse embryonic fibroblasts are also refractory to PPARγ-stimulated adipogenesis and fail to express adipogenic marker aP2, a PPARγ-responsive gene. Chromatin immunoprecipitation assays reveal reduced association in PRIP–/– cells of PIMT (PRIP-binding protein) and PBP with aP2 gene promoter, suggesting that PRIP is required for the linking of CBP/p300-anchored cofactor complex with PBP-anchored mediator complex. These data indicate that PRIP, like PBP, is a downstream regulator of PPARγ-mediated adipogenesis and that both these coactivators are required for the successful completion of adipogenic program.

lished the critical role for PBP/TRAP220 in PPAR␥-stimulated adipogenesis in that PBP Ϫ/Ϫ mouse embryonic fibroblasts (MEFs) were found refractory to adipocyte differentiation (27). Since PRIP is also a PPAR␥ coactivator and the disruption of the PRIP gene resulted in lethal phenotype, it appeared necessary to investigate the role of this coactivator in PPAR␥-stimulated adipogenesis. Here we show that PRIP Ϫ/Ϫ MEFs fail to exhibit PPAR␥1-stimulated adipogenesis even in the presence of PBP, suggesting the need for the presence of both PBP and PRIP for PPAR␥-directed adipogenesis.

MATERIALS AND METHODS
Construction of Retrovirus Vector-The retroviral vector expressing PPAR␥1 was constructed by inserting mouse PPAR␥1 cDNA isolated from PCMV-PPAR␥1 into pMSCVneo and the construct was verified by sequencing. Pheonix ecotropic retrovirus packaging cells and PT67 cells (Clontech) were cultured in Dulbecco's modified Eagle's medium with 10% fetal bovine serum. Recombinant retroviruses were produced by transfecting Pheonix cells seeded at a density of 6 ϫ 10 6 cells per 10-cm dish with 15 g of retrovirus vector by LipofectAMINE 2000 reagent (Invitrogen). Twenty-four hours after transfection, the cells were changed to fresh medium and cultured for another 24 h. Virus-containing medium was then filtered through 0.45-m membrane and added to 1 ϫ 10 5 PT67 cells with 5 g/ml polybrene (Sigma). Twenty-four hours after virus infection, the cells were selected with 200 g/ml G418 containing medium for 8 days. Surviving PT67 cells were grown into confluence.
Immortalization of Wild-type and PRIP Ϫ/Ϫ Embryonic Fibroblasts-Wild type and PRIP Ϫ/Ϫ primary MEFs were isolated from E11.5 littermate embryos as described (24). Self-immortalization of the MEFs was achieved by re-plating the cells every 3 days at a density of 2.6 ϫ 10 6 continuously for more than 5 months. PBP Ϫ/Ϫ MEFs were obtained as described previously (22).
Cell Culture, Retrovirus Infection of MEFs, and Induction of Adipogenesis-MEF cell lines were grown in Dulbecco's modified Eagle's medium with 10% fetal bovine serum. To infect MEFs with PPAR␥ retrovirus, 1 ϫ 10 5 cells were grown for 24 h and changed to the medium collected from 10-cm plate of virus producing PT67 cells. Cells were incubated with retrovirus for 24 h and then cultured in fresh medium. Forty-eight hours after infection, cells were selected with 200 g/ml G418 for 8 days. Induction of adipogenesis was carried out as described (5). Briefly, after cells grew into confluence, they were treated with culture medium containing 0.5 mM 3-isobutyl-1-methyl-xanthine (Sigma), 1 M dexamethasone (Sigma), 5 g/ml insulin (Sigma), and 0.5 M rosiglitazone or Me 2 SO for 48 h. Cells were then changed to the medium containing 5 g/ml insulin (Sigma) and 0.5 M rosiglitazone or Me 2 SO. The induction lasted for 8 days with the medium being replaced every 2 days. The fat droplets in cells that exhibited adipogenesis were revealed by Oil Red staining.
Northern Analysis and RT-PCR-Total RNA was extracted from cultured cells using TRIzol (Invitrogen) according to the manufacturer's instructions. For Northern blotting, 20 g of total RNA was used for each sample. To detect PBP mRNA in wild type and PRIP Ϫ/Ϫ MEFs, RT-PCR was carried out with primers 5Ј-TGTATCTGGCTCTC-CAATCC-3Ј and 5Ј-AGTGATGAGTTCATACAGGGG-3Ј (4). To detect PRIP mRNA in wild type and PBP Ϫ/Ϫ MEFs, RT-PCR was performed with primers 5Ј-TTTCATGGTGATGCAGCAGC-3Ј and 5Ј-CAT-CATATTTGGTGGCCCGT-3Ј (15). Total RNA (1 g) was used for each sample for RT-PCR analysis using One-Step RT-PCR kit (Invitrogen).

PPAR␥1-induced Adipogenesis in PRIP
Ϫ/Ϫ MEFs-PPAR␥ plays an essential role in the transcriptional regulation of adipocyte-specific genes during adipogenesis (2,5). Introduction of PPAR␥ into either primary or immortalized MEFs stimulates adipogenesis (5,27). Both PPAR␥ isoforms, PPAR␥1 and PPAR␥2, have been shown to have the similar adipogenic capacity (28). To test the requirement of PRIP in PPAR␥-mediated adipogenesis, immortalized MEF cell lines were established from E11.5 PRIP Ϫ/Ϫ embryos and their wild type littermates. PRIP Ϫ/Ϫ mouse embryos die between E11.5 and E12.5 (postcoitum) due in most part to defects in the development of placenta, heart, liver, nervous system, and retardation of embryonic growth (24 -26). Wild type and PRIP Ϫ/Ϫ MEF cell lines were infected with retroviruses expressing PPAR␥1 and then induced with differentiation medium containing PPAR␥ ligand rosiglitazone to maximize cell differentiation or vehicle (dimethyl sulfoxide) alone. Wild type MEFs expressing PPAR␥1 and treated with rosiglitazone exhibited adipocyte conversion (Fig. 1, A and C). In contrast, retroviral expression of PPAR␥1 failed to induce adipogenesis in PRIP Ϫ/Ϫ MEFs even in the presence of rosiglitazone (Fig. 1, B and D). Consistent with the morphological changes, mRNA and protein levels of adipocyte specific gene aP2 were markedly induced in wild type MEFs exhibiting adipogenesis, whereas aP2 mRNA and protein were not detectable in PRIP Ϫ/Ϫ MEFs that failed to reveal adipogenesis (Fig. 2, A and C). On Northern blotting, PPAR␥1 mRNA level in PRIP Ϫ/Ϫ MEFs were at a higher level than that noted in the wild type MEFs showing the differentiated morphology (Fig. 2A). These results demonstrated that PPAR␥1 was unable to stimulate adipogenesis in the absence of PRIP. Primary PRIP Ϫ/Ϫ MEFs also showed defects in PPAR␥1stimulated adipogenesis (data not shown).

PRIP Is Required for PPAR␥-mediated Adipogenesis 25282
Expression of PBP in PRIP Ϫ/Ϫ Cells-Recently, PPAR␥-coactivator PBP, the anchor protein for TRAP-DRIP-ARC-PRIC complex (13,14), has been shown to be essential for adipogenesis (27). Since PBP Ϫ/Ϫ MEFs have been shown to be defective in PPAR␥2-stimulated adipogenesis, it was important to ascertain the relative levels of PBP mRNA in wild type and PRIP Ϫ/Ϫ MEFs (Fig. 2B). PBP mRNA levels, as assessed by RT-PCR, were essentially similar in wild type and PRIP Ϫ/Ϫ MEFs. There was also no difference in the level of PRIP mRNA expression in PBP Ϫ/Ϫ MEFs (Fig. 2B). These results indicate that the defects in PPAR␥1-induced adipogenesis caused by the absence of PRIP and PBP are independent of each other.
Recruitment of PBP, CBP, and PIMT to aP2 Gene Promoter in PRIP Ϫ/Ϫ MEFs during Adipogenesis-Transcriptional activation by PPAR␥ requires recruitment of nuclear receptor coactivator complexes to remodel chromosome structure and facilitate transcriptional initiation (8,9). To investigate the impact of the absence of PRIP on the ability of PPAR␥ to recruit coactivator complexes, ChIP assay was utilized to examine the association of aP2 gene promoter with coactivators in wild type and PRIP Ϫ/Ϫ MEFs that received adipogenesis treatment (Fig.  3). Using antibodies against PPAR␥, PRIP, CBP, PIMT, or PBP, protein-DNA immune complexes were pulled down from cell lysates, and the DNA fragments in the complexes analyzed by PCR with primers (5Ј-AAATTCAGAAGAAAGTAAACA-CATTATT-3Ј and 5Ј-ATGCCCTGACCATGTGA-3Ј) spanning the PPAR␥-responsive element (AGGTCAAATGTGT) in the aP2 gene promoter region (29). The presence of endogenous PPAR␥ was detected on the aP2 promoter in wild type and PRIP Ϫ/Ϫ cells, and this association was prominent in the presence of ligand. PPAR␥ occupancy was increased by expressing exogenous PPAR␥ and addition of PPAR␥ ligand. Recruitment of CBP, and PBP to the aP2 promoter by PPAR␥, was not visibly altered in PRIP Ϫ/Ϫ MEFs in response to retroviral expression of PPAR␥1, whereas the recruitment of PIMT to the aP2 promoter in response to exogenous PPAR␥1 was less robust in PRIP Ϫ/Ϫ cells (Fig. 3), suggesting that PRIP is needed for PIMT recruitment and for the formation of a large multiprotein transcriptional complex (20,21).
Interaction of Coactivators in PRIP Ϫ/Ϫ Cells-Nuclear receptor coactivators CBP/p300, PBP, PRIP, and PIMT function in concert to promote transcriptional activation (8,9). It appears that PRIP and PRIP-binding protein PIMT serve as important linkers between CBP/p300-and PBP-anchored coactivator complexes (20,21). To investigate the interaction of these coactivators in PRIP Ϫ/Ϫ cells, wild type and PRIP Ϫ/ ϪMEFs were infected with adenovirus expressing His-tagged PIMT (20). PIMT and its putative associated proteins were immunoprecipitated with anti-His antibodies, and the presence of CBP, p300, and PBP in the immunoprecipitate was detected by immunoblotting (Fig. 4). While the interaction of PIMT with PBP or p300 was not altered in PRIP Ϫ/Ϫ cells, binding of PIMT to CBP was weaker in PRIP Ϫ/Ϫ cells compared with wild type cells.

DISCUSSION
In this study, we have examined the role of nuclear receptor coactivator PRIP in the PPAR␥1-directed adipogenesis using wild type and PRIP Ϫ/Ϫ MEFs. Our findings provide evidence that PRIP is required for adipocyte differentiation and for the expression of adipocyte specific aP2 gene. The refractoriness of PRIP Ϫ/Ϫ MEFs to PPAR␥-stimulated adipogenesis is similar to that described for PBP/TRAP220 Ϫ/Ϫ MEFs (27), suggesting that both PRIP and PBP play important roles in mediating the adipocyte differentiating effects of PPAR␥. These two coactivators were isolated using PPAR␥ as bait in the yeast two-hybrid screen and were identified as coactivators for PPAR␥ and other nuclear receptors (13)(14)(15)(16)(17)(18). PBP has emerged as a central piece in large TRAP-DRIP-ARC-PRIC multiprotein cofactor complex CBP is barely detected in PIMT immunoprecipitate derived from PRIP Ϫ/Ϫ cells. (10 -12, 21), whereas PRIP and PRIP-binding protein PIMT have been found recently to serve as linkers between CBP-and PBP-anchored cofactor complexes (18,20). The embryonic lethality observed with the disruption of PBP and PRIP genes and the failure of MEFs derived from the PBP and PRIP null mutants to differentiate into adipocytes under PPAR␥ stimulation suggest that both PBP and PRIP are vital for the successful completion of transcriptional activity of several genes involved in adipogenesis and possibly in the development and differentiation necessary for ontogeny.
The mechanism for the failure of PRIP Ϫ/Ϫ MEFs to undergo PPAR␥1-stimulated adipogenesis may involve inadequate docking or linkage between CBP/p300-anchored coactivators with PBP-anchored TRAP-DRIP-ARC-PRIC multiprotein complex. Since PRIP binds CBP/p300 and TRAP/DRIP130, a component of TRAP complex, the absence of PRIP most likely interferes with linkage of CBP/p300-anchored coactivator complex with PBP-anchored complex, thus curtailing the transcriptional signaling (18,20). Absence of PRIP might further interfere with these protein-protein interactions because PRIP binding protein PIMT directly binds with both CBP/p300 and PBP (20). Interestingly, ChIP assays revealed reductions in PIMT recruitment to the aP2 promoter in response to exogenous PPAR␥1, implying that PRIP is needed to recruit PIMT to the coactivator complex. ChIP assays also revealed reduction in exogenous PPAR␥1 recruitment to aP2 promoter in PRIP Ϫ/Ϫ MEFs (Fig. 3). This may be due to the requirement of PRIP for the stable formation of PPAR␥1-RXR heterodimers on the aP2 promoter or reduced amount of RXR in PRIP Ϫ/Ϫ MEFs in that PRIP Ϫ/Ϫ MEFs were shown to exhibit marked repression of RXR-mediated transcriptional activity (24). Immunoprecipitation and immunoblotting data reveal that in PRIP Ϫ/Ϫ MEFs the binding of CBP to PIMT is reduced, and this may also have functional implications.
It is now well established that PPAR␥ and C/EBP␣ are critical transcription factors in adipogenesis (1,2). Genetic analysis of adipogenesis has revealed that PPAR␥ promotes adipogenesis in C/EBP␣-deficient cells, but the converse is not true in that C/EBP␣ has no ability to promote adipogenesis in the absence of PPAR␥ (2). The studies of Ge and co-workers (27) with PBP Ϫ/Ϫ MEFs, and the observations reported here using PRIP Ϫ/Ϫ MEFs clearly establish the importance of these two coactivators in PPAR␥-stimulated adipogenesis. These two coactivators appear to function as downstream effectors of PPAR␥, and both are required for the successful completion of the adipogenic program. This assumption is based on the observation that PRIP Ϫ/Ϫ MEFs used in this study express PBP (Fig. 2B). Likewise PBP Ϫ/Ϫ MEFs also express PRIP mRNA to the same level as wild type MEFs, and the absence of either one of these coactivators interferes with PPAR␥-stimulated adipo-genesis. In essence, lack of PBP may result in the disruption of TRAP-DRIP-ARC-PRIC complex formation and absence of PRIP may interfere with the linkage and passage of transcriptional signal through the combined CBP/p300-and PBP-anchored complex. Additional studies are needed to assess the role of PRIP in in vivo adipogenesis using PRIP conditional null mice.