0013-7227/01/$03.00/0 Endocrinology Copyright © 2001 by The Endocrine Society
Vol. 142, No. 4 Printed in U.S.A.
Glucocorticoid Receptor-Interacting Protein-1 and Receptor-Associated Coactivator-3 Differentially Interact with the Vitamin D Receptor (VDR) and Regulate VDRRetinoid X Receptor Transcriptional Cross-Talk* LAURA L. ISSA†, GARY M. LEONG‡, JANELLE B. BARRY, ROBERT L. SUTHERLAND, AND JOHN A. EISMAN Bone and Mineral Research Program (L.L.I., G.M.L., J.B.B., J.A.E.) and Cancer Research Program (R.L.S.), Garvan Institute of Medical Research, St. Vincent’s Hospital, Sydney, New South Wales 2010, Australia ABSTRACT The vitamin D3 receptor (VDR) is a ubiquitously expressed nuclear hormone receptor, and its ligand, calcitriol, has diverse biological effects. The extent to which transcriptional coactivators are involved in modulating tissue-specific functions of the VDR is unclear. Hence, the current studies investigated the role of p160 coactivators in regulating VDR function and interaction with RXR. Two p160 coactivators, glucocorticoid receptor-interacting protein-1 (GRIP1) and receptor-associated coactivator-3 (RAC3), which are expressed in an inverse fashion in cell lines representative of calcitriol target tissues, interacted directly with the VDR, both in vitro and in yeast cells, but only in the presence of calcitriol. Deletional analyses of VDR indicated
T
HE VITAMIN D receptor (VDR) is a member of the superfamily of nuclear hormone receptors (NHRs) that includes receptors for estrogen, testosterone, glucocorticoids, progesterone, thyroid hormone, and retinoids (1). The VDR mediates the genomic actions of the active form of vitamin D3, 1,25-dihydroxyvitamin D3 (or calcitriol). Calcitriol plays a key role in the regulation of calcium homeostasis and the development and maintenance of bone. In addition, calcitriol regulates the growth and differentiation of a diverse number of cell types, including cancer and immune cells (2, 3). The VDR functions as a homodimer or as a heterodimer with the retinoid X receptor (RXR) to bind vitamin D responsive elements (VDREs) in gene promoters. Recent studies have shown that activation of transcription by the VDR involves direct or indirect interaction with the general transcription factors, TFIIA and TFIIB, to recruit basal components to the preinitiation complex (4 – 6). Transcriptional activation by NHRs also requires direct interaction with Received July 14, 2000. Address all correspondence and requests for reprints to: Laura L. Issa, Muscle Development Unit, Children’s Medical Research Institute, Locked Bag 23, Wentworthville, New South Wales 2145, Australia. Email:
[email protected]. * These studies were supported by grants from the National Health and Medical Research Council (NH & MRC) of Australia and the New South Wales Cancer Council. † Recipient of an Australian Postgraduate Award. ‡ Recipient of an NH & MRC Postgraduate Medical Scholarship.
that GRIP1 and RAC3 required an intact VDR activation function (AF-2) domain for efficient interaction as well as additional but distinct regions of the VDR. Coexpression experiments in yeast cells indicated that both GRIP1 and RAC3 coassemble with the VDR to form an active transcriptional complex. They also form ternary complexes with VDR homodimers and VDR:RXR␣ heterodimers. In mammalian cells, GRIP1 augmented VDR activation of the osteocalcin promoter, whereas RAC3 enhanced VDR activation indirectly through RXR. These data suggest different coactivators regulate VDR function via distinct mechanisms and support the hypothesis that the VDR recruits different coactivators depending on specific gene and cellular contexts. (Endocrinology 142: 1606 –1615, 2001)
coactivators that recruit and stabilize the binding of RNA polymerase II to the activated promoter. For example, the TATA-binding protein (TBP)-associated factor, TAFII135, potentiates VDR-, TR-, and RXR-mediated transcription (7). Several other putative transcriptional coactivators for NHRs have been identified. These include SRC-1 (NCoA-1 or p160), GRIP1 (TIF2, NCoA-2 or SRC-2), and RAC3 (pCIP, ACTR, AIB1, or SRC-3; 8 –16). Collectively, these molecules form a group of structurally and functionally related proteins, termed p160 coactivators. These coactivators interact with a range of NHRs in a ligand-dependent manner via the receptors’ activation function (AF-1/AF-2) domains. Members of the p160 coactivator family share a series of LxxLL motifs, the so-called nuclear receptor (NR) boxes, which are required for interaction with NHRs (17–19). Specific receptors exhibit both coactivator and NR box binding preference, which allow for multiple interactions of a few coactivators with a diverse family of NHRs (20, 21). Transcriptional coactivators act as coregulators by integrating multiple signal transduction pathways with NHRs. For instance, SRC-1 interacts with TBP, TFIIB, and the CREB binding proteins, CBP/p300 and p/CAF in addition to NHRs (22, 23). Furthermore, SRC-1, CBP/p300 and RAC3 possess histone acetyltransferase activity, which is thought to modify chromatin structure and enhance access of NHR-cofactor complexes to DNA (15, 24). Whether p160 coactivators participate in forming a transcriptionally active complex with the VDR and precisely how they modulate its cell-type-specific genomic actions is not
1606
VDR-COACTIVATOR INTERACTIONS
known. Coactivators shown to interact with the VDR include GRIP1, RAC3, SUG1, SRC-1, RIP140, and NCoA-62 (11, 12, 14, 25–29). Previous studies investigating VDR interactions with GRIP1 and RAC3 do not address the functional role of these coactivators in VDR regulation of a natural VDRE particularly in the context of RXR, the major heterodimeric partner of VDR, and the ligand for RXR, 9-cis RA. Consequently, the present study investigated whether GRIP1 and RAC3 are bona fide transcriptional coactivators for the VDR in the absence and presence of RXR, and whether they modulate VDR function via similar or distinct mechanisms. In addition to examining ligand-dependent interactions between the VDR, GRIP1, and RAC3 in vitro, and formation of ternary complexes with VDR homodimers and VDR:RXR heterodimers, these interactions were examined using electromobility shift assays and activation of the osteocalcin gene promoter VDRE in a mammalian cell system. Materials and Methods DNA expression constructs All expression vectors were sequenced using manual or dye-terminator automated sequencing. The expression constructs of Saccharomyces cerevisiae gal4-DNA binding domain (gal4DBD), pAS2–1, and the gal4activation domain (gal4AD), pACTII, were obtained from CLONTECH Laboratories, Inc. (Palo Alto, CA). A 1.4-kb hVDR complementary DNA (cDNA) insert was PCR amplified from pPH13-VDR (kindly provided by Dr. P. MacDonald; 30) using Pfu DNA polymerase (Stratagene, La Jolla, CA) and cloned in the BamHI site of pAS2–1 and pACTII. pACTII was modified (pACTII) by replacing the BglII polylinker fragment with the double-stranded oligonucleotide: 5⬘-GATCTGTGAATTCCCGGGGATCCGTCGACCTA-3⬘. The EcoRI/BglII hRXR␣ cDNA fragment from pGST-RXR␣ was cloned into pACTII (31). pGAD424-GRIP1 containing nucleotides 204-4878 of GRIP1 cDNA was kindly provided by Dr. M. Stallcup (11). pGAD-RAC3.1 containing nucleotides 1289 –3698 (amino acids 401-1204) of RAC3 was a gift from Dr. D. Chen (14). The yeast LexA-DNA binding domain expression construct, pEG202 was provided by Dr. R. Brent. hVDR cDNA from pAS2–1-VDR was subcloned into BamHI linearized pEG202. hRXR␣ cDNA from pACTII-RXR␣ was cloned into EcoRI/BamHI-linearized pEG202. GRIP1 cDNA from pGAD424-GRIP1 was subcloned into EcoRI cut pEG202. BglII RAC3 cDNA from pGAD-RAC3.1 was inserted in BamHI linearized pEG202. Yeast constructs expressing truncated hVDR were generated using pPH13-VDR as a template. pAS2–1-VDR⌬1–191 (VDR:LBD) was constructed by PCR amplification of nucleotides 574-1281. pAS2–1VDR⌬365– 427 (VDR⌬AF-2) was generated by PstI digestion followed by ligation of pAS2–1-VDR. pAS2–1-VDR⌬1–280 was constructed by removing the first 954 nucleotides by NcoI digestion and religation of pAS2–1-VDR. pAS2–1-VDR1– 89 was generated by PCR amplification of nucleotides 116 –382 of pPH13-VDR which were subsequently cloned into the BamHI site of pAS2–1. For transient transfections of mammalian cells, hVDR cDNA was PCR amplified from SaoS-2 cellular RNA and cloned into HindIII/ApaI linearized pRC-CMV (Invitrogen, Carlsbad, CA). To clone the EcoRI GRIP1 cDNA fragment in frame with the EcoRI site of pCMV2 (Eastman Kodak Co., Rochester, NY) the vector was first modified by Klenow fill-in of the HindIII site to create pCMV2c. EcoRI/BglII RXR␣ cDNA fragment was subsequently cloned in pCMV2c. pCMV2-RAC3 was generated by PCR amplification of the RAC3 cDNA using pGAD-RAC3.1 as a template with primers containing BglII 5⬘ overhangs. The human osteocalcin gene promoter-luciferase reporter construct, pOSLUC2, containing 1.34 kb of promoter sequence, was created by exchanging the XhoI/Vsp1 fragment (corresponding to the CAT cDNA) of pOSCAT2 (32) with the XhoI/Vsp1 luciferase cDNA fragment of pGL-3basic (Promega Corp., Madison, WI). The Renilla luciferase gene reporter, pRL-TK (Promega Corp.), was used to determine transfection efficiency. VDR and RXR␣ cDNA fragments from pAS2–1-VDR and pACTIIRXR␣ were subcloned into pSG5 (Promega Corp.) and used for in vitro coupled transcription and translation. A modified pSG5 vector,
1607
designated pSG5, was created by replacing the EcoRI/BglII polylinker fragment of pSG5 with a double-stranded-oligonucleotide-(5⬘-AATTACGCTACAACGCCATGGGCGAATTCAGGATCCGCATCGATTAGCTGAATAGTA-3⬘). EcoRI GRIP1 cDNA fragment was inserted in EcoRI linearized pSG5. BglII RAC3 cDNA from pGAD-RAC3.1 was cloned in BamHI linearized pSG5. PCR primer sequences are available on request.
Northern blot analysis Total RNA was isolated from subconfluent cells by denaturation in guanidinium followed by purification using a cesium chloride step gradient. Total RNA (15 g) was subjected to Northern analysis. cDNA probes for hVDR, hRXR␣, mRXR, GRIP1, RAC3, and SRC-1 were labeled with [␣32P]dCTP by random priming using Klenow DNA polymerase (Promega Corp.) and purified through a Nick column (Amersham Pharmacia Biotech, Uppsala, Sweden). Northern blots were prehybridized for 4 h at 50 C in prehybridization buffer (10⫻ Denhardt’s solution; 5⫻ SSC (1⫻ SSC is 0.15 m NaCl, 0.015 m sodium citrate); 50 mm Tris-HCl, pH 7.5; 1% SDS; 0.2⫻ SSPE [1⫻ SSPE is 0.18 m NaCl, 10 mm NaPO4, 1 mm EDTA, pH 7.7), and 180 g/ml denatured salmon sperm DNA]. Blots were hybridized with labeled probes overnight at 50 C in hybridization mixture (1⫻ Denhardt’s solution; 0.8⫻ SSC; 50 mm TrisHCl, pH 7.5; 1% SDS; 0.04⫻ SSPE; 9.6% dextran sulfate; 50% deionized formamide, and 3.6 g denatured salmon sperm DNA). Blots were washed twice in 2⫻ SSC and 0.1% SDS at 63 C for 30 min, followed by two washes in 0.2⫻ SSC and 0.1% SDS at 63 C for 30 min then autoradiographed at ⫺80 C for 4 –24 h.
Coimmunoprecipitation Radiolabeled ([35S]-Methionine; NEN Life Science Products, Boston, MA) VDR, RXR␣, GRIP1, and RAC3 were expressed from pSG5 vectors in vitro using the rabbit reticulocyte lysate coupled transcription-translation kit (TNT; Promega Corp.). Protein lysates (10 l) were incubated for 30 min at 22 C with 30 ng of recombinant hVDR (Affinity BioReagents, Inc., Neshanic Station, NJ), in the presence of 100 nm calcitriol or vehicle. Binding reactions were performed in a binding buffer containing 50 mm HEPES (pH 7.5), 150 mm NaCl, 10% glycerol, 1% Triton X-100, 1.5 mm MgCl2, 1 mm EGTA, 10 mm Na-pyrophosphate, 20 mm NaF, 1 mm phenylmethylsulfonyl fluoride, 200 m Na-orthovanadate, and 10 g/ml of each of aprotinin and leupeptin. Rat anti-VDR antibody (100 ng of 9A7; Affinity BioReagents, Inc.) was added and the mixture incubated 30 min at 4 C. Rabbit antirat IgG2a (100 ng; Zymed Laboratories, Inc., South San Francisco, CA) was added for 45 min at 4 C. The protein-immunocomplexes were immobilized on protein A-Sepharose 4B conjugate (Zymed Laboratories, Inc.) following incubation for 30 min at 4 C and extensive washes in binding buffer. Protein-immunocomplexes were resuspended in 3⫻ Laemmli buffer and resolved by SDSPAGE. The gels were dried and autoradiographed using Kodak Biomax film (Eastman Kodak Co., Rochester, NY).
Yeast two-hybrid assay A two-hybrid mating assay was used to examine interactions between gal4DBD-hVDR and various gal4AD-cofactor fusion proteins. The yeast reporter strain CG-1945 (CLONTECH Laboratories, Inc.) was transformed with 1 g of pAS2–1 or pAS2–1hVDR using an alkali cation method (BIO 101, Inc., La Jolla, CA). Transformants were selected on complete synthetic media (CSM; BIO 101) plates lacking tryptophan. Y187 cells (CLONTECH Laboratories, Inc.) were transformed with 1 g of either pACTII, pACTII-VDR, pACTII-RXR␣, pGAD424-GRIP1, or pGAD10-RAC3.1 and selected on CSM plates lacking leucine. Transformed CG-1945 cells were mated with Y187 cells in microtiter plates containing YEPD medium for 16 h at 30 C. Diploid cells were selected, grown to saturation then used to inoculate (OD600 of 0.1) test cultures. These were grown for 4 h then treated with calcitriol or vehicle and grown for 16 h at 30 C. Yeast cells were harvested in 100 l Breaking Buffer (100 mm Tris pH 8; 20% glycerol; 5 mm phenylmethylsulfonyl fluoride; 2 mm dithiothreitol) and lysed by vortexing with glass beads. Protein concentration was determined by the Bradford method (Bio-Rad Laboratories, Inc., Sydney, New South Wales, Australia). Protein extracts (10 –20 g) were assayed for -galactosidase activity at room
1608
VDR-COACTIVATOR INTERACTIONS
temperature for 1 h using the Galactolight reporter assay kit (Tropix, Bedford, MA) and quantitated with a Berthold AutoLumat LB953 (Bad Wildbad, Germany). For yeast experiments in which cofactor interactions with hVDR dimers were investigated, SFY526 cells (CLONTECH Laboratories, Inc.) were transformed with pAS2–1 or pAS2–1hVDR. SFY526 transformants were mated with the Y187 strains described above. These diploid master strains were subsequently transformed with pEG202, pEG202-hVDR, pEG202-RXR␣, pEG202-GRIP1, and pEG202-RAC3 and selected on CSM plates lacking tryptophan, leucine, and histidine. Test cultures were grown and assayed as for the yeast two-hybrid assay.
Western blot analysis For determination of gal4 fusion protein expression in yeast cells, cells were grown to saturation, harvested, and the protein extracted by a TCA precipitation method. Uniform expression of gal4DBD:VDR across different yeast diploid strains was confirmed by immunoblotting for hVDR. An antibody for the gal4DBD was used to determine expression of truncated gal4DBD:VDR proteins. Protein lysates (20 –30 g) were loaded onto a 10 or 12% SDS-PAGE gel. Gels were electroblotted onto nitrocellulose membrane and blocked with Tris-buffer saline containing 0.1% Triton-X 100 (TBS-T) containing 5% skim milk powder and 1% BSA for 16 h at 4 C. Membranes were exposed to rat anti-VDR (1 g/ml; Affinity BioReagents, Inc.) or mouse anti-gal4DBD antibody (0.5 g/ml; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) in TBS-T ⫹ 1% BSA for 2 h at 22 C, followed by rabbit antirat (1/36,000) or sheep antimouse HRP-conjugated secondary antibody (1/2000; Amersham International, Buckinghamshire, UK) for a further 1–2 h at 22 C. Immunoblotted bands were detected by a chemiluminecent method (ECL; Amersham Pharmacia Biotech).
Electromobility shift assay VDR, RXR, GRIP1, and RAC3 were expressed in vitro using the TNT kit (Promega Corp.). Oligonucleotides corresponding to nucleotides ⫺521 to ⫺450 of the human osteocalcin gene promoter were labeled with [␣32P]dCTP using Klenow DNA polymerase (Promega Corp.), isolated by electrophoresis through a 15% polyacrylamide gel, eluted in TE buffer, pH 8, and purified through glasswool. Receptor lystates (8 l) were incubated with 100 nm calcitriol or/and 1 m 9-cis RA or vehicle in a binding buffer (10 mm Tris HCl/HEPES; 200 mm KCl; 0.5 mm EDTA; 2 mm MgCl2; 1 mm dithiothreitol) for 20 min at 22 C. Labeled probe (45,000 cpm) was added and the reactions incubated a further 20 min at 22 C. Protein-DNA reactions were separated by electrophoresis in TBE buffer (pH 7.4) through a 4% nondenaturing polyacrylamide (29:1) gel containing 1.6% glycerol. Gels were dried and autoradiographed for 4 – 8 h.
Endo • 2001 Vol. 142 • No. 4
and RXR, in a panel of cell lines representative of vitamin D target tissues. VDR was expressed at high levels in mouse NIH3T3 fibroblasts and at low levels in primary cultures of mouse (mOB) and rat (ROS17/2.8) osteoblastic cells and P19 mouse embryonal carcinoma cells (Fig. 1). Although not apparent by Northern blot analysis, low levels of VDR protein expression in CV-1 cells has been confirmed by responsiveness to calcitriol in reporter gene transfections. RXR␣ was ubiquitously expressed but at differential levels; with the highest expression in COS1, mOB, NIH3T3, ROS17/2.8, and ROS24 cells, and the lowest expression in CV-1 and P19 cells. RXR was also widely expressed, with the highest expression in NIH3T3 and P19 cells. Similarly, GRIP1 was highly expressed in NIH3T3, P19, and the bone cell lines. RAC3 messenger RNA (mRNA) expression followed an inverse pattern to that of GRIP1 with the highest expression in the Green Monkey kidney cell lines, COS1 and CV-1, and very low levels in other cell lines. SRC-1 was widely expressed at differential levels, with high expression levels in COS1 and ROS24 cells and low levels in mOB cells. Coactivator interaction domains of the VDR
An in vitro binding assay was used to investigate whether GRIP1 and RAC3 participate in direct contact with the VDR, using RXR␣ as an experimental control. GRIP1 and RAC3 were coimmunoprecipitated with recombinant VDR only in the presence of calcitriol (Fig. 2). Coimmunoprecipitation of RXR␣ with VDR was observed in the absence of calcitriol; however, this was augmented by addition of calcitriol. VDR dimerization and cofactor interactions were reconstituted in the yeast two-hybrid system. gal4AD:RXR␣ inter-
Mammalian cell transient transfections Cells were transfected using the nonliposomal FuGENE reagent (Roche Molecular Biochemicals, Sydney, New South Wales, Australia). Approximately 5 ⫻ 105 CV-1 monkey kidney cells, maintained in DMEM (Life Technologies, Inc., Gaithersburg, MD) supplemented with 10% FCS, were seeded in 10 cm2 dishes. After 4 h in DMEM containing 2% charcoal-stripped serum, cells were transfected with a total of 10 g plasmid DNA containing 2 g of pOSLUC2 or pOSLUC1 and 0.1 g of pRL-TK. Sixteen hours post transfection, the cells were trypsinized and seeded into 12-well plates. Test cultures were treated with appropriate ligand for a further 16 h before being harvested, lysed in 1⫻ lysis buffer (Promega Corp.), and tested for Firefly and Renilla luciferase. Firefly luciferase data were normalized to Renilla luciferase data to correct for transfection efficiency. For transient transfections of P19 mouse embryonal carcinoma cells (maintained in ␣-MEM supplemented with 5% serum), 106 cells were transfected as above.
Results Expression of p160 coactivators in vitamin D-responsive cell lines
Northern blot analysis was performed to investigate the relative expression of p160 coactivators, with respect to VDR
FIG. 1. Expression of p160 coactivators in mammalian cells. Total RNA (15 g) from cultured cell lines was analyzed for VDR, RXR␣, RXR, GRIP1, RAC3, and SRC-1 mRNA by Northern blot hybridization of 32P-labeled cDNA probes. The cell lines investigated include: Green Monkey kidney COS1 and CV-1; FVBN mouse osteoblastic mOB; mouse fibroblastic NIH3T3; mouse embryonal carcinoma P19, and rat osteosarcoma ROS17/2.8 and ROS24 cells. Graphed data show expression levels relative to 18S ribosmal RNA.
VDR-COACTIVATOR INTERACTIONS
1609
FIG. 2. Ligand-dependent interaction between VDR and cofactors in vitro. Coimmunoprecipitation of recombinant hVDR with in vitro translated [35S]-labeled RXR␣ (A), GRIP1 (B), and RAC3 (C). Binding assays were performed either in the absence or presence of hVDR and in the absence or presence of 100 nM calcitriol, as indicated. Interacting proteins were immunoprecipitated using an antibody against hVDR coupled to protein-A coated Sepharose. Positive and negative controls indicate immunoprecipitation of [35S]-labeled-hVDR and luciferase protein, respectively. The input represents 10% of the radiolabeled protein added to the binding reaction. FIG. 4. Expression of VDR mutants. In this schematic representation of the domain structure of the VDR and the truncation mutants (A), the amino acid boundaries of the functional DNA-binding, ligandbinding, and activation function domains (DBD, LBD, and AF-2, respectively) of wild-type VDR are indicated. Western blot analysis (B) using an antibody to gal4DBD detected gal4DBD:VDR and truncation mutants. Numbers above the lanes correspond to the mutants described in (A). *, Specific gal4DBD fusion protein bands; NS, nonspecific bands.
ligand-dependent interaction with the full-length wild-type VDR and with the VDR-LBD mutant but did not interact with either the VDR-⌬AF-2 mutant or the VDR-⌬1–280 (Fig. 5C). FIG. 3. Ligand-dependent interaction between VDR and cofactors in yeast cells. Yeast (CG-1945) cells transformed with gal4 DNA binding domain (gal4DBD) alone or as a fusion protein with VDR were mated with Y187 cells expressing the gal4 activation domain (gal4AD) alone or fused to either VDR or RXR␣ (A), GRIP1 (B), and RAC3 (C). Induction of the -galactosidase reporter gene (expressed as relative light units, RLU) was determined in the absence (p) and presence of 10 nM calcitriol (f). Data are the mean ⫾ SD of three separate yeast cultures and are representative of three to five independent experiments.
acted with gal4DBD:VDR in the absence of ligand (Fig. 3). However, calcitriol increased the response almost 5-fold. The gal4DBD:VDR fusion protein interacted with gal4AD:GRIP1 or gal4AD:RAC3 but only in the presence of calcitriol. When fused to the gal4AD neither VDR, RXR␣, GRIP1, nor RAC3 activated reporter gene expression nonspecifically, i.e. when coexpressed with gal4DBD alone. Regions of the VDR required for interaction with RXR␣, GRIP1, and RAC3 were examined by deletion analysis (Fig. 4A). Comparable levels of expression of these VDR mutants were confirmed by Western blotting with an anti-gal4DBD antibody (Fig. 4B). RXR␣ interacted with full-length VDR, but not with any of the deletion mutants (Fig. 5A). Similarly, GRIP1 displayed ligand-dependent interaction only with the wild-type receptor and failed to interact with either the Nterminal deletion mutant, VDR-LBD, or the AF-2 deletion mutant, VDR-⌬AF-2 (Fig. 5B). In contrast, RAC3 displayed
Coexpression of GRIP1 and RAC3 potentiates VDR transactivation in yeast
Having shown that GRIP1 and RAC3 require different regions of the VDR for interaction, the possibility that both cofactors can interact with the VDR simultaneously was examined. Interaction between gal4DBD:VDR and gal4AD:GRIP1 was partially blocked by coexpression with VDR, and more so by excess RXR (Fig. 6A). Excess GRIP1 did not interfere with this interaction, whereas RAC3 expression potentiated calcitriol-dependent interaction between gal4DBD:VDR and gal4AD:GRIP1 by 2- to 3-fold (Fig. 6A). Interaction between gal4DBD:VDR and gal4AD:RAC3 was largely unaffected by coexpression of VDR, RXR␣ or RAC3. However, coexpression of GRIP1 induced a 2- to 3-fold enhancement of calcitriol-dependent gal4DBD:VDR/gal4AD:RAC3 interaction (Fig. 6B). These results demonstrate that GRIP1 and RAC3 can coassemble with the VDR, possibly by using alternative contact sites. Coactivation of VDR-homodimer and VDR:RXR␣ heterodimer complexes by GRIP1 and RAC3
The VDR regulates transcription by binding VDREs as a homodimer or as a heterodimer with RXR (33, 34). To examine whether p160 coactivators interact with VDR dimers, GRIP1 and RAC3 were coexpressed in yeast two-hybrid strains containing either gal4DBD:VDR/gal4AD:VDR (ho-
1610
VDR-COACTIVATOR INTERACTIONS
Endo • 2001 Vol. 142 • No. 4
FIG. 6. GRIP1 and RAC3 interact with the VDR simultaneously. SFY526/Y187 diploid master strains expressing gal4DBD:VDR and gal4AD:GRIP1 fusion proteins (A) and gal4DBD:VDR and gal4AD: RAC3 fusion proteins (B) were transformed with expression vectors for VDR, RXR␣, GRIP1, or RAC3. Yeast cultures were treated with vehicle (p) or 10 nM calcitriol (f) in triplicate and the -galactosidase activity (mean ⫾ SD) determined. Reproducible data were obtained from three independent experiments.
FIG. 5. Cofactor interaction domains of the VDR. SFY526 yeast cells expressing VDR and mutant VDR fused to the gal4DBD were mated with Y187 cells expressing RXR␣ (A), GRIP1 (B), and RAC3 (C) fused to the gal4AD. Induction of the -galactosidase reporter was tested in the absence (p) and presence of 200 nM calcitriol (f). The values plotted on the graph represent the mean ⫾ SD of three yeast cultures. Similar results were obtained in two additional experiments. Note that the reporter activity of (A) is RLU ⫻ 105, and (B) and (C) is RLU ⫻ 107. Noninteracting strains shown in (B) and (C) are ⬍0.002 RLU on the scales shown.
modimeric strain) or gal4DBD:VDR/gal4AD:RXR (heterodimeric strain). GRIP1 and, to a greater extent, RAC3 potentiated induction of reporter gene expression by dimeric interaction between the gal4DBD:VDR and gal4AD:VDR fusion proteins, in the presence but not the absence of calcitriol (Fig. 7A). Similarly, the coactivators enhanced transactivation by heterodimeric interaction between the gal4DBD:VDR and gal4AD:RXR␣ fusion proteins in a calcitriol-dependent manner (Fig. 7B). Again, the effect of RAC3 was more potent than that of GRIP1. 9-cis RA cotreatment did not affect calcitriolinduced interaction between gal4DBD:VDR and gal4AD:RXR␣ in the presence of GRIP1 or RAC3; hence, the coactivators interact with the heterodimer via ligand-activated VDR.
GRIP1 expression resulted in 4.3 (⫾2.4; n ⫽ 3)-fold increase in transactivation by VDR homodimers and a comparable 3.5 (⫾2.1)-fold induction of VDR:RXR␣ heterodimeric activity, relative to empty vector. RAC3 potentiated transactivation by VDR:RXR␣ heterodimers by 60 (⫾25; n ⫽ 3)-fold compared with 15 (⫾5)-fold for VDR homodimers. Overexpression of VDR or RXR␣ did not compete with the gal4-VDR fusion protein for interaction in the VDR homodimer strain [1.2 (⫾0.2) and 0.9 (⫾0.2), respectively]. However, RXR␣ displaced interaction between gal4DBD:VDR and gal4AD: RXR␣ [0.5 (⫾0.1)], but not between gal4DBD:VDR and gal4AD: VDR [0.9 (⫾0.4)]. The control reactions, in which the cofactors were transformed in a yeast cell background containing gal4DBD with gal4AD, or gal4DBD:VDR with gal4AD, showed low nonspecific activity (Fig. 7, C and D). GRIP1 and RAC3 displaced VDR:RXR heterodimer binding to DNA in vitro
The above data provided evidence that the cofactors retain ability to interact with the VDR when it is dimerized or heterodimerized with RXR. Consequently, ternary interactions between GRIP1 and RAC3 with VDR dimers were examined in the context of a VDRE by an electromobility shift assay. The VDR bound the osteocalcin VDRE probe (⫺521 to ⫺450) as a heterodimer with RXR␣ and in a ligand-dependent manner (Fig. 8A). Neither VDR alone nor RXR␣ alone bound the VDRE probe in the absence or presence of their cognate ligands. Coincubation with GRIP1 displaced the VDR:RXR␣-DNA complex induced by calcitriol, in the absence and presence of 9-cis RA, by almost 50%. These effects
VDR-COACTIVATOR INTERACTIONS
1611
FIG. 7. GRIP1 and RAC3 coactivation of VDR dimeric interactions in yeast cells. VDR:VDR homodimeric (A) and VDR:RXR␣ heterodimeric (B) interactions were investigated in the presence of overexpressed GRIP1 or RAC3. Yeast reporter strains expressing gal4DBD and gal4AD empty vectors (C) and gal4DBD:VDR and gal4AD (D) controls showed minimal nonspecific interactions. Induction of -galactosidase activity was determined in the absence (p) and presence of 10 nM calcitriol (f) alone or in combination with 1 M 9-cis RA (u). Data are the mean ⫾ SD of triplicate cultures and are representative of three independent experiments. Note the activity data in (A) and (B) are RLU ⫻ 105 and in (C) and (D) as RLU ⫻ 104. Also note the broken scales in (A) and (B).
were reproducible both qualitatively and quantitatively in multiple experiments (n ⫽ 2–5). Similarly, addition of RAC3 displaced VDR:RXR␣ heterodimeric binding to DNA. To test whether coactivator displacement of the heterodimer-DNA complex was due to competition for RXR␣, RXR␣ was titrated against a fixed amount of VDR and GRIP1 (Fig. 8B). At a ratio of VDR:RXR␣:GRIP1 of 1:2:4, the intensity of the ligandinduced heterodimeric-DNA complex was restored to the same level as that seen in the absence of GRIP1. At a ratio of VDR:RXR␣:GRIP1 of 1:4:4 the heterodimeric-DNA complex induced by calcitriol and 9-cis RA cotreatment was inhibited by 50%, whereas the calcitriol-induced complex was unaffected. These results suggest that VDR:RXR␣ heterodimeric binding to a VDRE precludes interaction with GRIP1 and RAC3. Coactivation of the osteocalcin gene promoter by GRIP1 and RAC3
GRIP1 and RAC3 have previously been shown to enhance the transcriptional activities of NHRs (18, 20). However, their role in VDR regulation of the osteocalcin VDRE in vivo have not been examined. Coactivator function was investigated by transient transfection assays of the osteocalcin promoter fused to a luciferase reporter. Studies were performed in P19 and CV1 cells because they expressed GRIP1 and RAC3 in an inverse pattern. In P19 cells, which express GRIP1 with low levels of RAC3, overexpression of GRIP1 increased calcitriolmediated transactivation of pOSLUC2 by 3.0 (⫾0.9; n ⫽
FIG. 8. GRIP1 and RAC3 disrupt VDR:RXR binding to the osteocalcin VDRE in vitro. Electromobility shift analysis of VDR:RXR binding to the osteocalcin VDRE in the absence and presence of GRIP1 and RAC3. Protein-DNA reactions were treated with vehicle, 10 nM calcitriol, 1 M 9-cis RA, or both hormones in combination. Graphed data represent the intensity of retarded bands relative to the VDR:RXR heterodimeric complex induced by calcitriol and are representative of experiments repeated two to five times with similar outcomes.
7)-fold, relative to VDR transfected cells (Fig. 9A). Overexpression of GRIP1 with VDR and RXR␣ potentiated calcitriol-mediated transactivation by 3.6 (⫾1.4)-fold compared with the 1.5 (⫾0.7)-fold induction seen in cells transfected with only VDR and RXR␣. GRIP1 also increased the 9-cis RA-induced response by 4.0 (⫾1.5)-fold in the absence of overexpressed RXR␣, and by 37 (⫾7)-fold in the presence of RXR␣, compared with the 17 (⫾2.2)-fold increase seen in the VDR and RXR␣ transfected cells. Cotransfection of GRIP1 with VDR and RXR␣ had an additive effect on the transcriptional response to calcitriol plus 9-cis RA cotreatment, in-
1612
Endo • 2001 Vol. 142 • No. 4
VDR-COACTIVATOR INTERACTIONS
of GRIP1 with VDR also increased the transcriptional response to 9-cis RA treatment by 4.5 (⫾1.1)-fold, and the response to calcitriol plus 9-cis RA cotreatment by 3.8 (⫾1.4)fold, relative to cells transfected with VDR alone. Discussion
FIG. 9. Transactivation of the osteocalcin gene promoter by GRIP1 and RAC3. Coregulation of the human osteocalcin gene promoter (pOSLUC2) by the VDR and cofactors in P19 (A) and CV-1 (B) cells. Cells were transiently transfected with 2 g pOSLUC2, 0.5 g VDR alone or in combination with 0.5 g RXR␣, 5 g GRIP1, or 5 g RAC3 as indicated below the columns. Transfected cells were seeded in triplicate, incubated with 0.1% isopropanol (䡺), 10 nM calcitriol (o), 1 M 9-cis RA (p) or both hormones in combination (f) and assayed for luciferase activity. Data are presented as relative light units (RLU ⫾ SD) and are representative of four to seven reproducible experiments.
ducing 10.4 (⫾3.3)-fold increase in reporter activity, compared with 3.1 (⫾1.3)-fold in VDR and GRIP1 transfected cells and 4.2 (⫾1.0)-fold in VDR and RXR␣ transfected cells. Overexpression of RAC3 had a modest effect on calcitriolinduced VDR transactivation in P19 cells (2.1 ⫾ 0.9; n ⫽ 3; Fig. 9A). However, coexpression of RAC3 with VDR and RXR␣ enhanced 9-cis RA-induced transcriptional response by 89 (⫾28; n ⫽ 4)-fold compared with 17 (⫾2.2)-fold in VDR and RXR␣ transfected cells. Coexpression of RAC3 with VDR and RXR␣ also enhanced induction following calcitriol plus 9-cis RA cotreatment by 23 (⫾10)-fold, compared with 4.2 (⫾1.0)-fold in the VDR and RXR␣ transfected cells. In CV-1 cells, which have abundant endogenous RAC3 but low GRIP1 expression, overexpression of RAC3 did not significantly affect transactivation of pOSLUC2 (not shown). Coexpression of GRIP1 with VDR, in this cell line, resulted in a modest 2.1 (⫾0.6; n ⫽ 5)-fold induction of pOSLUC2 activity by calcitriol (Fig. 9B). This effect was attenuated in cells transfected with GRIP1, VDR, and RXR␣. Coexpression
Activation of gene transcription by NHRs is facilitated by direct interaction with basal transcription factors and transcriptional coactivators that are thought to recruit RNA polymerase II (23, 35). The coactivators, GRIP1, RAC3, and SRC-1 have been observed to interact with the VDR; however, at the time this study commenced, whether the VDR is coexpressed with these p160 coactivators, whether VDR-coactivator complexes are transcriptionally active, and their functional significance in the context of VDR:RXR heterodimeric interactions had not been defined. Hence, the present studies investigated the role of GRIP1 and RAC3 in VDR homodimer and heterodimer formation and in the transactivation of a typical VDRE in the osteocalcin gene promoter. GRIP1 and RAC3 were coexpressed with the VDR in cell lines representative of vitamin d-responsive tissues and participated in direct contact with ligand-activated VDR in vitro. In yeast cells, VDR-coactivator interactions were both ligand and AF-2 domain-dependent. GRIP1 and RAC3 appeared to interact simultaneously with the VDR through distinct regions of the VDR. Interaction studies in yeast cells suggest that the coactivators form ternary complexes with VDR: RXR␣ heterodimers. However, in vitro gel shift studies showed DNA binding by VDR:RXR␣ heterodimers was disrupted by GRIP1 and RAC3 apparently via competition for RXR␣. In mammalian cells, GRIP1 potentiated VDR-mediated transactivation of the osteocalcin VDRE, directly through ligand-activated VDR, and, indirectly through ligand-activated RXR. Hence GRIP1 may function to integrate cooperative signaling between VDR and RXR. In contrast, RAC3 did not potentiate VDR-mediated transactivation directly but facilitated transactivation by VDR:RXR␣ heterodimers through ligand-activated RXR. The presence of three structurally similar cofactors that can serve as coactivators for NHRs, notably GRIP1 and RAC3 as well as SRC-1, suggests that these coactivators may play different roles in different cells (9, 11, 14). Consequently, mRNA expression of these cofactors was examined across a panel of cell lines representative of vitamin D target tissues. This revealed that GRIP1 and RAC3 are expressed differentially and, more importantly, inversely relative to each other. SRC-1, however, is ubiquitously expressed. Possibly, the VDR uses different coactivators to activate transcription depending on the cell line or target promoter. In support of this concept, the bone cell lines, mOB, ROS17/2.8 and ROS24, expressed abundant GRIP1 mRNA with low or undetectable levels of RAC3. Consistent with a possible role in modulating VDR function in bone, GRIP1 potentiated transactivation of the osteocalcin gene promoter, which is expressed only in osteoblasts. The cell lines CV-1 and COS1 are derived from primate kidney, a tissue involved in calcium and phosphate homeostasis and vitamin D activation. These cells had high levels of RAC3 expression but low GRIP1 expression. It remains to be shown whether RAC3 is involved in regulating
VDR-COACTIVATOR INTERACTIONS
vitamin D-responsive genes in kidney-derived cell lines. Coexpression of SRC-1 with RAC3 in the kidney cell lines, and with GRIP1 in the bone cell lines, suggests that coactivators may substitute one another depending on the target promoter, or more than one coactivator can assemble with the VDR at an active promoter. The latter is consistent with data presented here showing GRIP1 and RAC3 interacted with the VDR simultaneously. Thus, different combinations of coactivators could fine-tune VDR function and its cross-talk with other cis-acting transcription factors, such as RXR. The current study confirms, by in vitro immunoprecipitation experiments, that the VDR interacts directly with GRIP1 and RAC3 in a ligand-dependent manner. Hence, a ligandinduced conformational change to the VDR is required to expose coactivator interactive domains (36 –38). Further, ligand-dependency implicates a role for GRIP1 and RAC3 in ligand-mediated gene transactivation via the AF-2 domain. Consistent with this, neither GRIP1 or RAC3 interacted with a VDR AF-2 deletion mutant. Whether different coactivator molecules interact with the same or different sites on the VDR is unknown, hence the ability of GRIP1 and RAC3 to interact with a panel of VDR truncation mutants was investigated. These results show RAC3 interacted with the VDRLBD (⌬1–190) but not the VDR⌬1–280 mutant, and therefore amino acids 191–280 are crucial for generating the appropriate functional conformation necessary for efficient binding. Amino acids 191–280 encompass ␣-helix 3 (225–246) of the ligand binding domain that interacts with the AF-2 domain of ␣-helix 12 following ligand-induced intramolecular folding (27, 36, 37). In contrast, GRIP1 did not interact with the VDR⌬1–190 mutant, requiring both the N-terminal and C-terminal regions of the VDR. Recently, Chen et al. (25) identified specific amino acids within ␣-helices 3, 5, and 12 of the VDR required for interaction with GRIP1 but overlooked the importance of the N terminus. Although further VDR deletion and point mutation analyses are required to identify the amino acids involved in distinguishing GRIP1 from RAC3 binding, these results have led to some novel observations. Firstly, different coactivators bind different sites on the VDR and, secondly, the VDR appears to interact simultaneously with more than one coactivator molecule by using alternative binding sites (Fig. 10A). In support of these conclusions, transactivation by the VDR:GRIP1 complex in yeast cells was markedly enhanced when RAC3 was coexpressed by formation of a ternary complex. Similarly, GRIP1 potentiated VDR:RAC3 interaction. It is unclear precisely how NHRs and coactivators assemble on DNA to form a transcriptionally active, multimeric complex. VDR:VDR homodimers and VDR:RXR heterodimers are transcriptionally functional but differ in their affinity and specificity for target gene VDREs (33). Recently, SRC-1 and TRAM-1 were shown to distinguish between interaction with VDR homodimers vs. interaction with VDR:RXR heterodimer on synthetic VDREs (39). To investigate whether GRIP1 and RAC3 differentially interact with VDR dimers, we reconstituted calcitriol-dependent VDR dimerization in yeast cells then coexpressed GRIP1 and RAC3. Expression of both GRIP1 and RAC3 significantly enhanced the transactivation potential of VDR: VDR homodimers and VDR:RXR␣ heterodimers, suggestive of tripartite interactions. Although GRIP1 did not appear to dif-
1613
FIG. 10. VDR-coactivator interactions. The DNA-binding (DBD), ligand-binding (LBD), and activation function-2 (AF-2) domains of the VDR are indicated in the unoccupied receptor (A). As VDR-coactivator interactions occur only in the presence of calcitriol, ligand binding induces a receptor intramolecular fold exposing sites for coactivator contact. VDR deletion studies indicate the sites for RAC3 interaction differ from those for GRIP1 interaction, consistent with the ability of both coactivators to coassemble with the VDR to form a transcriptionally active complex. GRIP1 interaction with the VDR potentiates calcitriol-induced transactivation of the osteocalcin gene VDRE by VDR:RXR␣ heterodimers and by 9-cis RA activated RXR␣:RXR␣ homodimers (B). With calcitriol plus 9-cis RA costimulation, the two dimeric species could compete for GRIP1 binding such that the transcriptional output would be additive or attenuated, dependent on the relative abundance of receptors and GRIP1 in the cell. RAC3 does not appear to function as a direct coactivator of the VDR in the context of the osteocalcin gene VDRE but rather potentiates transactivation through the RXR␣ partner of VDR:RXR␣ heterodimers and 9-cis RA activated RXR␣:RXR␣ homodimers.
ferentiate between the VDR dimeric forms, RAC3 was a more potent coactivator of VDR:RXR␣ heterodimers. These data are consistent with the VDR deletion experiments that suggest the coactivators require different regions of the VDR for interaction. Possibly, the conformational changes following heterodimerization facilitate stronger binding between VDR and RAC3. Coexpression of RXR disrupted interaction between gal4DBD: VDR and gal4AD:GRIP1 but not between gal4DBD:VDR and gal4AD:RAC3. These observed differences are consistent with independent binding sites on the VDR for these cofactors. The osteocalcin gene promoter, a typical vitamin Dresponsive gene, provides a good model for investigating the interplay between VDR, RXR, and the coactivators. The osteocalcin promoter has both a calcitriol-responsive VDRE
1614
Endo • 2001 Vol. 142 • No. 4
VDR-COACTIVATOR INTERACTIONS
and a 9-cis RA-responsive element (RXRE), which can function synergistically (40). Because GRIP1 has been shown to interact with RXR in yeast cells and in vitro, GRIP1 may coactivate both the VDR heterodimers at the VDRE and RXR homodimers at the RXRE (41). In P19 cells, GRIP1 overexpression with VDR enhanced induction of the osteocalcin gene promoter by calcitriol 2- to 4-fold. When GRIP1 was coexpressed with VDR and RXR, there was an additive effect on the transcriptional response to calcitriol plus 9-cis RA costimulation, relative to either hormone alone. Thus, GRIP1 potentiates both calcitriol- and 9-cis RA-mediated transactivation of the osteocalcin promoter. It is possible that GRIP1 oscillates between interaction with the VDR and with RXR in the presence of both hormones. For instance, in CV-1 cells, which have low endogenous RXR, coexpression of RXR with GRIP1 attenuated GRIP1 coactivation of the VDR, suggesting that VDR:RXR heterodimers and RXR:RXR homodimers were competing for GRIP1. Hence, the VDRE and RXRE, which can function synergistically through allosteric interaction, can also be antagonistic through competition for coactivators, depending on the cell line. Another possibility is that GRIP1 preferentially binds RXR. This is supported by the electromobility shift data where, using a region of the osteocalcin gene promoter containing the VDRE but not the RXRE, GRIP1 displaced RXR from heterodimerization with VDR and disrupted the receptor-DNA complex. Addition of excess RXR␣ restored calcitriol-induced VDR:RXR␣ heterodimerization. However, in the presence of 9-cis RA, GRIP1 competition was further augmented, indicating GRIP1 was preferentially interacting with ligand-activated RXR. Similarly, RAC3 displaced VDR:RXR␣-DNA complexes, which may reflect allosteric inhibition by the RXR AF-2 domain, as has been previously suggested (26). Collectively, these results indicate GRIP1 functions as a coactivator for both the VDR and for RXR. The direction in which the multipartite complex forms is dependent on both the abundance of individual NHRs and coactivators in the cell and the activating hormone (Fig. 10B). Possibly, GRIP1 may function to coordinate dimerization between VDR and RXR in response to hormonal signals. In contrast, RAC3 interacted with the VDR but did not appear to act as a direct coactivator of VDR in the context of the osteocalcin promoter. Although, others have observed that RAC3 potentiates VDR induction of a synthetic consensus VDRE weakly in CV-1 cells, they did not address 9-cis RA-dependent activation (26). RAC3 did not potentiate VDR-mediated transcription in CV-1 cells under our conditions. However, overexpression of RAC3 in P19 cells potentiated 9-cis RA-dependent but not calcitriol-dependent transactivation, and thus RAC3 may regulate the VDR indirectly through RXR (Fig. 10C). In summary, this study examined two functionally and structurally related coactivators in relation to VDR function. GRIP1 and RAC3 played distinct roles in regulating VDR function. They both interacted with the VDR in a ligand- and AF-2-dependent manner but differed in their sites of interaction, their ability to form ternary complexes with VDR homodimers and VDR:RXR heterodimers, and in their effects on the interplay between vitamin D and retinoid signaling pathways. Further, GRIP1 and RAC3 exhibited differential expression across cell lines representative of vitamin D-
responsive tissues. These data lead us to hypothesize a model whereby the cell- and promoter-specific functions of the VDR may be mediated through differential recruitment of coactivators. The interplay between VDR, RXR␣, and the coactivators is complex and dependent on the relative abundance of cofactors and activating hormones. This functional complexity may be the basis for transcriptional switches controlling diverse and complex programs of gene expression in response to hormonal signals. Acknowledgments We thank Dr. P. MacDonald for pH13-VDR, Dr. M. Stallcup for pGAD424-GRIP1, Dr. D. Chen for pGADRAC3.1, and Dr. R. Brent for pEG202. We thank Drs. Colin Watts and Edith Gardiner for invaluable discussions of the manuscript, Liz Buttrose for assistance in tissue culture, Kate Sunn for pRC-CMV-VDR and Ron Enriquez for technical assistance.
References 1. Evans RM 1988 The steroid and thyroid hormone receptor superfamily. Science 240:889 – 895 2. Haussler MR, Haussler CA, Jurutka PW, Thompson PD, Hsieh JC, Remus LS, Selznick SH, Whitfield GK 1997 The vitamin D hormone and its nuclear receptor: molecular actions and disease states. J Endocrinol [Suppl] 154:S57–S73 3. Issa LL, Leong GM, Eisman JA 1998 Molecular mechanism of vitamin D receptor action. Inflammation Res 47:451– 475 4. Blanco JC, Wang IM, Tsai SY, Tsai MJ, O’Malley BW, Jurutka PW, Haussler MR, Ozato K 1995 Transcription factor TFIIB and the vitamin D receptor cooperatively activate ligand-dependent transcription. Proc Natl Acad Sci USA 92:1535–1539 5. Leong GM, Wang KS, Marton MJ, Blanco JCG, Wang IM, Rolfes RJ, Ozato K, Segars JH 1998 Interactions between the retinoid X receptor and transcription factor IIB is ligand-dependent in vivo. J Biol Chem 273:2296 –2305 6. Lemon BD, Fondell JD, Freedman LP 1997 Retinoid X receptor: vitamin D3 receptor heterodimers promote stable preinitiation complex formation and direct 1,25-dihydroxyvitamin D3-dependent cell-free transcription. Mol Cell Biol 17:1923–1937 7. Mengus G, May M, Carre L, Chambon P, Davidson I 1997 Human TAF(II)135 potentiates transcriptional activation by the AF-2s of the retinoic acid, vitamin D3, and thyroid hormone receptors in mammalian cells. Genes Dev 11:1381–1395 8. Onate SA, Boonyaratanakornkit V, Spencer TE, Tsai SY, Tsai MJ, Edwards DP, O’Malley BW 1998 The steroid receptor coactivator-1 contains multiple receptor interacting and activation domains that cooperatively enhance the activation function 1 (AF1) and AF2 domains of steroid receptors. J Biol Chem 273:12101–12108 9. Spencer TE, Jenster G, Burcin MM, Allis CD, Zhou J, Mizzen CA, McKenna NJ, Onate SA, Tsai SY, Tsai MJ, O’Malley BW 1997 Steroid receptor coactivator-1 is a histone acetyltransferase. Nature 389:194 –198 10. Voegel JJ, Heine MJS, Zechel C, Chambon P, Gronemeyer H 1996 TIF2, a 160 kDa transcriptional mediator for the ligand-dependent activation function AF-2 of nuclear receptors. EMBO J 15:3667–3675 11. Hong H, Kohli K, Trivedi A, Johnson DL, Stallcup MR 1996 GRIP1, a novel mouse protein that serves as a transcriptional coactivator in yeast for the hormone binding domains of steroid receptors. Proc Natl Acad Sci USA 93:4948 – 4952 12. Hong H, Kohli K, Garabedian MJ, Stallcup MR 1997 GRIP1, a transcriptional coactivator for the AF-2 transactivation domain of steroid, thyroid, retinoid, and vitamin D receptors. Mol Cell Biol 17:2735–2744 13. Torchia J, Rose DW, Inostroza J, Kamei Y, Westin S, Glass CK, Rosenfeld MG 1997 The transcriptional coactivator p/CIP binds CBP and mediates nuclear-receptor function. Nature 38:677– 684 14. Li H, Gomes PL, Chen JD 1997 RAC3, a steroid/nuclear receptor-associated coactivator that is related to SRC-1 and TIF2. Proc Natl Acad Sci USA 94:8479 – 8484 15. Chen H, Lin RJ, Schiltz RL, Chakravarti D, Nash A, Nagy L, Privalsky ML, Nakatani Y, Evans RM 1997 Nuclear receptor coactivator ACTR is a novel histone acetyltransferase and forms a multimeric activation complex with P/CAF and CBP/p300. Cell 90:569 –580 16. Anzick SL, Kononen J, Walker RL, Azorsa DO, Tanner MM, Guan XY, Sauter G, Kallioniemi OP, Trent JM, Meltzer PS 1997 AIB1, a steroid receptor coactivator amplified in breast and ovarian cancer. Science 277:965–968 17. Heery DM, Kalkhoven E, Hoare S, Parker MG 1997 A signature motif in
VDR-COACTIVATOR INTERACTIONS
18. 19.
20.
21.
22. 23. 24. 25. 26. 27.
28. 29.
30.
transcriptional coactivators mediates binding to nuclear receptors. Nature 387:733–736 Li H, Chen JD 1998 The receptor-associated coactivator-3 activates transcription through CREB-binding protein recruitment and autoregulation. J Biol Chem 273:5948 –5954 McInerney EM, Rose DW, Flynn SE, Westin S, Mullen T-M, Krones A, Inostroza J, Torchia J, Nolte RT, Assa-Munt N, Milburn MV, Glass CK, Rosenfeld MG 1998 Determinants of coactivator LXXLL motif specificity in nuclear receptor transcriptional activation. Genes Dev 12:3357–3368 Ding XF, Anderson CM, Ma H, Hong H, Uht RM, Kushner PJ, Stallcup MR 1998 Nuclear receptor-binding sites of coactivators glucocorticoid receptor interacting protein 1 (GRIP1) and steroid receptor coactivator 1 (SRC-1)multiple motifs with different binding specificities. Mol Endocrinol 12:302–313 Hong H, Darimont BD, Ma H, Yang L, Yamamoto KR, Stallcup MR 1999 An additional region of coactivator GRIP1 required for interaction with the hormone-binding domains of a subset of nuclear receptors. J Biol Chem 274:3496 –3502 Glass CK, Rose DW, Rosenfeld MG 1997 Nuclear receptor coactivators. Curr Opin Cell Biol 9:222–232 McKenna NJ, Lanz RB, O’Malley BW 1999 Nuclear receptor coregulators: cellular and molecular biology. Endocr Rev 20:321–344 Bannister AJ, Kouzarides T 1996 The CBP coactivator is a histone acetyltransferase. Nature 384:641– 643 Chen S, Cui J, Nakamura K, Ribeiro RCJ, West BL, Gardner DG 2000 Coactivator-vitamin D receptor interactions mediate inhibition of the atrial natriuretic peptide promoter. J Biol Chem 275:15039 –15048 Leo C, Li H, Chen DJ 2000 Differential mechanisms of nuclear receptor regulation by receptor-associated coactivator 3. J Biol Chem 275:5976 –5982 Masuyama H, Brownfield CM, St-Arnaud R, MacDonald PN 1997 Evidence for ligand-dependent intramolecular folding of the AF-2 domain in vitamin D receptor-activated transcription and coactivator interaction. Mol Endocrinol 11:1507–1517 Gill RK, Atkins LM, Hollis BW, Bell NH 1998 Mapping the domains of the interaction of the vitamin D receptor and steroid receptor coactivator-1. Mol Endocrinol 12:57– 65 Baudino TA, Kraichely DM, Jefcoat SC, Winchester SK, Partridge NC, MacDonald PNG 1998 Isolation and characterisation of a novel coactivator protein, NCoA-62, involved in vitamin D-mediated transcription. J Biol Chem 273:16434 –16441 Nakajima S, Hsieh J-C, MacDonald PN, Galligan MA, Haussler CA, Whit-
31.
32.
33.
34.
35.
36.
37.
38.
39.
40. 41.
1615
field GK, Haussler MR 1994 The C-terminal region of the vitamin D receptor is essential to form a complex with a receptor auxiliary factor required for high affinity binding to the vitamin D-responsive element. Mol Endocrinol 8:159 –172 Mangelsdorf DJ, Umesono K, Kliewer SA, Borgmeyer U, Ong ES, Evans RM 1991 A direct repeat in the cellular retinol binding protein type II gene confers differential regulation by RXR and RAR. Cell 66:555–561 Morrison NA, Shine J, Fragonas JC, Verkest V, McMenemy ML, Eisman JA 1989 1,25-dihydroxyvitamin D-responsive element and glucocorticoid repression in the human osteocalcin gene. Science 246:1158 –1161 Carlberg C, Bendik I, Wyss A, Meier E, Sturzenbecker LJ, Grippo JF, Hunziker W 1993 Two nuclear signalling pathways for vitamin D. Nature 361:657– 660 Freedman LP, Arce V, Fernandez RP 1994 DNA sequences that act as high affinity targets for the vitamin D3 receptor in the absence of the retinoid X receptor. Mol Endocrinol 8:265–273 Shibata H, Spencer TE, Onate SA, Jenster G, Tsai SY, Tsai MJ, O’Malley BW 1997 Role of coactivators and corepressors in the mechanism of steroid/ thyroid receptor action. Recent Prog Horm Res 52:141–164 Renaud JP, Rochel N, Ruff M, Vivat V, Chambon P, Gronemeyer H, Moras D 1995 Crystal structure of the RAR␥ ligand-binding domain bound to all-trans retinoic acid. Nature 378:681– 689 Wurtz JM, Guillot B, Moras D 1997 3D model of the ligand binding domain of the vitamin D nuclear receptor based on the crystal structure of holo RAR␥. In: Norman AW, Bouillon R, Thomasset M (eds) 10th Workshop on Vitamin D. Chemistry, biology and clinical applications of the steroid hormone. University of California, Riverside, California, pp 165–172 Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179 Takeshita A, Yasunori O, Chin WW 2000 Nuclear receptor coactivators facilitate vitamin D receptor homodimer action on direct repeat hormone response elements. Endocrinology 141:1281–1284 Schrader M, Bendik I, Becker-Andre M, Carlberg C 1993 Interaction between retinoic acid and vitamin D signalling pathways. Nature 269:17830 –17836 Walfish PG, Yoganathan T, Yang YF, Hong H, Butt TR, Stallcup MR 1997 Yeast hormone response element assays detect and characterise GRIP1 coactivator-dependent activation of transcription by thyroid and retinoid nuclear receptors. Proc Natl Acad Sci USA 94:3697–3702
