Modulation of Testicular Receptor 4 Activity by ... - Semantic Scholar

Research

Modulation of Testicular Receptor 4 Activity by Mitogen-activated Protein Kinase-mediated Phosphorylation* M. D. Mostaqul Huq‡, Pawan Gupta‡, Nien-Pei Tsai, and Li-Na Wei§ Testicular receptor 4 (TR4) is an orphan member of the nuclear receptor superfamily. Despite the lack of identified ligands, its functional role has been demonstrated both in animals and cell cultures. However, it remains unclear how the biological activity of TR4 is regulated without specific ligands. In this study, we showed that in the absence of specific ligands the activity of TR4 could be modulated by mitogen-activated protein kinase (MAPK)-mediated phosphorylation of its activation function 1 (AF-1) domain. A mass spectrometry-based proteome analysis of TR4 expressed in insect cells revealed three phosphorylation sites in its AF-1 domain, specifically on Ser19, Ser55, and Ser68. Site-directed mutagenesis studies demonstrated the functionality of phosphorylation on Ser19 and Ser68 but not Ser55. We also demonstrated that MAPK-mediated phosphorylation of the AF-1 domain rendered TR4 a repressor, mediated through the preferential recruitment of corepressor RIP140. Dephosphorylation of its AF-1 made TR4 an activator due to its selective recruitment of coactivator, P300/cyclic AMP-responsive element binding protein-binding protein-associated factor (PCAF). The biological effects were validated by using the wild type TR4 and its constitutive negative (dephosphorylated) and constitutive positive (phosphorylated) mutants in the studies of regulation of its natural target gene, apoE. This study uncovered, for the first time, a ligand-independent mechanism underlying the biological activity of TR4 that was mediated by MAPKmediated receptor phosphorylation of AF-1 domain. Molecular & Cellular Proteomics 5:2072–2082, 2006.

Testicular receptor 4 (TR4)1 (also known as TAK1 and Nr2c2) remains an orphan member among the nuclear recepFrom the Department of Pharmacology, University of Minnesota Medical School, Minneapolis, Minnesota 55455 Received, May 16, 2006, and in revised form, July 31, 2006 Published, MCP Papers in Press, August 3, 2006, DOI 10.1074/ mcp.M600180-MCP200 1 The abbreviations used are: TR4, testicular receptor 4; MAPK, mitogen-activated protein kinase; AF, activation function; DR, direct repeat; HA, hemagglutinin; GFP, green fluorescent protein; IDA, information-dependent acquisition; CN, constitutive negative; CP, constitutive positive; FBS, fetal bovine serum; PTM, post-translational modification; TIC, total ion chromatogram; aa, amino acids; LBD, ligand binding domain; DBD, DNA binding domain; PPAR, peroxisome proliferator-activated receptor; NR, nuclear receptor; PCAF, P300/cyclic AMP-responsive element binding protein-binding protein-associated factor.

2072

Molecular & Cellular Proteomics 5.11

tor superfamily (1, 2). Based upon sequence alignment, TR4 is closely related to some other orphan receptors such as retinoid X receptors, chicken ovalbumin upstream promotertranscription factor, and hepatocyte nuclear factor 4 (3). The expression of TR4 was found to be ubiquitous in both embryonic and adult stages (4 – 6). Due to the lack of identified ligands, its physiological roles have been speculated. However, recent gene knock-out studies revealed its potential roles in animal behavior, coordination, and reproduction (7–9). In classical studies, it was shown that TR4 could bind to direct repeat (DR) AGGTCA with variable spacer nucleotides in the promoters of putative target genes (10 –13). These target genes could be activated or repressed. Among the potential target sequences, the DR1 sequence of the apolipoprotein E/C-I/C-II gene cluster was confirmed in the TR4 gene knock-out studies (13). In addition, TR4 was also shown to be an important regulator of other nuclear receptor signaling pathways, such as in pathways involving retinoic acid receptor, retinoid X receptor, peroxisome proliferator-activated receptor (PPAR), vitamin D3 receptor, thyroid hormone receptor, estrogen receptor, and another closely related orphan member, TR2 (14). Together these studies suggested TR4 as an important member of nuclear receptors. Despite the widely demonstrated transcriptional regulatory activity of TR4, it remains unclear how the activity of TR4, as an activator or a repressor, is triggered without the involvement of specific ligands. Our recent proteomics studies of another orphan receptor, TR2, revealed interesting regulatory mechanisms mediated by receptor phosphorylation that could modulate its biological activities (15, 16). We thus set out to determine whether TR4 could also be regulated by protein modification such as phosphorylation. A systematic mass spectrometry-based proteomics analysis of TR4 was initiated by examining modifications of TR4 protein expressed in insect cells, a widely used eukaryotic system for protein expression. In this endeavor, we identified three mitogen-activated protein kinase-mediated phosphorylation sites on the activation function 1 (AF-1) domain of TR4. To validate the biological effects of these modified residues, site-specific mutants were generated, and their biological activities were examined. We also determined the coregulatory mechanisms underlying the activating versus repressive activities of TR4 as well as its mutants by identifying the specific coactivator and corepressor involved. We now report that Ser19, Ser55, and Ser68 at the AF-1 domain of TR4

© 2006 by The American Society for Biochemistry and Molecular Biology, Inc. This paper is available on line at http://www.mcponline.org

Phosphorylation of TR4

can be phosphorylated by MAPK, but only phosphorylation on Ser19 and Ser68 has a biological consequence in terms of the regulation of its target gene expression. We also demonstrate that hyperphosphorylation of its AF-1 domain renders TR4 a repressor, whereas hypophosphorylation of this domain makes TR4 an activator. Its biological activities are mediated through the specific recruitment of coactivator PCAF and corepressor RIP140 by the hypophosphorylated and hyperphosphorylated forms of the protein, respectively. EXPERIMENTAL PROCEDURES

Plasmids—Full-length HA-tagged TR4, GFP-TR4, GST-TR4, GFPTR4⌬LBD, and Gal4TR4-LBD plasmids were constructed as described previously (17). GST-TR4⌬LBD was subcloned from GFPTR4⌬LBD into pGEX-2T expression vector. His-tagged TR4-LBD was subcloned from Gal4-TR4LBD into His-tagged expression vector (pET28, Novagen). Luciferase reporter containing DR1 response element derived from mouse hepatic control region-1 of apoE gene was constructed using synthetic oligonucleotides. The sense (5⬘-CTA GAA TGG CAG AGG TCA TCT AGA ATG GCA GAG GTC A-3⬘) and the antisense (5⬘-GAT CTG ACC TCT GCC ATT CTA GAT GAC CTC TGC CAT TCT AGA TGA CCT CTG CCA T-3⬘) oligonucleotides were annealed, phosphorylated by T4 kinase, and ligated to BamHI/NheI sites into pGL3-promoter vector (Promega). The underlined characters indicate the response element, and the bold character indicates the spacer nucleotide. For the expression of TR4 protein in the baculovirus system, the full-length TR4 cDNA tagged with an HA epitope was inserted into pVL1393 (Invitrogen) at the EcoRI and the XbaI sites. Expression and Purification of TR4 —Sf21 insect cells (1 ⫻ 106) were infected with recombinant baculovirus vector. After 72 h, the cells from 500-ml culture were harvested by centrifugation at 6000 rpm for 10 min at 4 °C, and the pellet was kept frozen at ⫺80 °C. The cell pellet was then resuspended in 30 ml of 10 mM HEPES, pH 7.9, containing 10 mM NaCl and 1.5 mM MgCl2, diluted with an equal volume of HEPES buffer containing 0.1% Nonidet P-40, and incubated on ice for 30 min followed by centrifugation for 5 min at 2000 rpm at 4 °C. The nuclear pellet was resuspended in 60 ml of 20 mM HEPES (pH 7.9) buffer, 420 mM NaCl, 1.5 mM MgCl2, and 25% glycerol and incubated on ice for an additional 30 min prior to centrifugation at 2000 rpm at 4 °C. This nuclear pellet was then resuspended in 40 ml of 20 mM HEPES, pH 7.9, 2 M NaCl, 1.5 mM MgCl2, and 25% glycerol and sonicated on ice until the clear homogeneous solution was obtained. The homogenate was centrifuged at 15,000 rpm for 30 min at 4 °C. The pellet was solubilized in 20 ml of an extraction buffer containing 50 mM sodium phosphate buffer, pH 7.0, 6 M guanidine HCl, and 300 mM NaCl and sonicated to obtain a clear homogenate. The homogenate was then dialyzed against a buffer containing 50 mM phosphate buffer, pH 7.5, 250 mM NaCl, 1 mM DTT, 0.15 mM PMSF, and 1 M urea for 2 h at 4 °C. The protein was then subjected to affinity purification on anti-HA-agarose affinity resins following the manufacturer’s protocol. The purified protein was resolved by 8% SDS-PAGE. Mass Spectrometric Analysis of TR4 —Mass spectral analysis of TR4 protein sample was conducted according to the established procedure described previously (18, 19). Purified HA-tagged TR4 protein from insect cells was resolved by SDS-PAGE. Gel slices containing TR4 were subjected to overnight in-gel tryptic digestion. The samples were analyzed by MALDI-TOF MS (QSTAR XL, Applied Biosystems, Inc., Foster City, CA) using ␣-cyano-4-hydroxycinnamic acid as a matrix in a positive ion reflection mode. For LC-MS, an LC Packings (Dionex, Sunnyvale, CA) Famos autosampler and an LC Packings Switchos pump were used to concentrate and desalt the

sample on an LC Packings C18 nanoprecolumn. The precolumn was connected in line with a capillary column (100-␮m inner diameter, 5 ␮m, 200-Å-pore size C18 particles), and peptides were eluted in a gradient system of ACN and H2O containing 0.1% TFA using an LC Packings Ultimate LC system over 65 min. The LC system was on line with Applied Biosystems, Inc. QSTAR Pulsar quadrupole TOF mass spectrometer, which was equipped with the Protana nanoelectrospray source. As peptides were eluted from the column they were focused into the mass spectrometer. The informationdependent acquisition (IDA) was used to acquire MS/MS. IDA mode was set to measure continuous cycles of full scan TOF MS from 400 to 1200 m/z plus three product ion scans from 50 to 4000 m/z. The data from IDA experiments were searched at MASCOT (www.matrixscience.com) MS/MS data search. The mass tolerance of both precursor ions and the MS/MS fragment ions was set at ⫾0.1 Da, and carbamidomethylcysteine was specified as a static modification. Phosphorylated Ser/Thr/Tyr and oxidized methionines were specified as variable modifications. All MS/MS spectra were manually analyzed to verify sequence assignments. Peaks with a minimum height of 3% relative to the base peak were considered, and a 100 ppm tolerance was used to establish matches with the theoretical b and y ions that were predicted with the help of Bioanalyst software (Applied Biosystems, Inc.). Site-directed Mutagenesis—Site directed mutagenesis on phosphorylated serine residues of full-length HA-TR4, GST-TR4, GFPTR4⌬LBD, and GST-TR4⌬LBD was performed using the QuikChange XL site-directed mutagenesis kit (Stratagene) following the manufacturer’s protocol. Replacement of phosphoserine residues with alanine and with glutamic acid were made by using mutagenic primers. The mutagenic primers were designed such that they were matched to nearest alanine or glutamic acid. The Ser/Thr 3 Ala point/sequential and Ser/Thr3 Glu point/sequential mutations were referred to as the constitutive negative (CN) and the constitutive positive (CP) mutants, respectively. The mutagenic primers used to generate the mutant constructs are: S19A, 5⬘-CTCTGCGGTAGCCGCACCTCAGCGCATTC-3⬘ (sense), and 5⬘-GAATGCGCTGAGGTGCGGCTACCGCAGAG-3⬘ (antisense); S55A, 5⬘-GTTCATCCTAACCGCCCCAGATGGAGCTG-3⬘ (sense) and 5⬘-CAGCTCCATCTGGGGCGGTTAGGATGAAC-3⬘ (antisense); S68A, 5⬘-GTGATCCTGGCTGCTCCGGAAACATCC-3⬘ (sense) and 5⬘-GGATGTTTCCGGAGCAGCCAGGATCAC-3⬘ (antisense); S19E, 5⬘-CTCTGCGGTAGCCGAACCTCAGCGCATTC-3⬘ (sense) and 5⬘-GAATGCGCTGAGGTTCGGCTACCGCAGAG-3⬘ (antisense); S68E, 5⬘-GTGATCCTGGCTGAACCGGAAACATCC-3⬘ (sense) and 5⬘-GGATGTTTCCGGTTCAGCCAGGATCAC-3⬘ (antisense). The underlined characters indicate the genetic code for the mutant amino acid. The positive clones were verified by DNA sequencing. RT-PCR and Western Blot—Total RNA was isolated from H235 cells using a TRIzol姞 kit (Invitrogen), and RT reaction was conducted using SuperscriptTM (Invitrogen) reverse transcriptase enzyme following the manufacturer’s protocol. The specific primers are 5⬘-CTA TGG GGC TGT CAG TTG TG-3⬘ (sense) and 5⬘-CTC CTC CAC TGC TAT CTA TC-3⬘ (antisense) to PCR amplify TR4 cDNA, 5⬘-TGT GGG CCG TGC TGT TGG TCA C-3⬘ (sense) and 5⬘-TGC CTT GTA CAC AGC TAG GCG C-3⬘ (antisense) to amplify cDNA of apoE gene, and actinspecific primers 5⬘-TGGCCTTAGGGTGCAGGG-3⬘ (sense) and 5⬘GTGGGCCGCTCTAGGCACCA-3⬘ (antisense). The expression level of HA-TR4 was detected by anti-HA antibody (Santa Cruz Biotechnology), and TR4-⌬-LBD was detected using anti-TR4 antibody (Santa Cruz Biotechnology) in Western blot analyses. Cell Culture, Transfection, and Reporter Assay—COS-1 cells were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% FBS, and mouse hepatoma H2.35 cells were maintained in Dulbecco’s modified Eagle’s medium containing 4% FBS and 20 nM dexamethasone. Transient transfection in cells was performed using LipofectamineTM 2000 (Invitrogen). In the reporter assay, 0.1 ␮g of


2073


expression plasmids (wild type TR4 full length, TR4-⌬LBD, and their mutants), control plasmids (pCMX/GFP empty vector), DR1-tk-luciferase (0.5 ␮g) reporter, and CMV-lacZ as an internal control (0.05 ␮g) were used in each well of 24-well plates. Thirty-two hours posttransfection cultures were fed a fresh medium containing dextrancharcoal-treated FBS and treated for 8 h with either 3 ␮M MAPK inhibitor (PD98059) or 1 ␮M anisomycin (MAPK activator) (Calbiochem). To monitor the effect of various cofactors on modulation of TR4 biological activity, the wild type TR4 full length and the TR4⌬LBD along with their mutants (0.1 ␮g) were introduced together with PCAF or RIP140 expression vector (0.1 ␮g) into cells. Forty hours posttransfection total cell extracts were collected and tested for luciferase and lacZ activities. The -fold relative luciferase activity was calculated by normalizing relative luciferase unit activity of the experimental groups to the relative luciferase unit activity of the empty vector control group. GST Pulldown Assay—GST and GST fusion proteins were partially purified from bacteria by affinity chromatography using glutathioneagarose beads (Sigma). Preliminary binding studies were done for the various GST-TR4 constructs to determine the amount of bound sample that would yield approximately an equal amount of protein on a Coomassie-stained SDS-polyacrylamide gel. After binding, the beads were washed twice with 20 volumes of PBS and once with a binding buffer (20 mM HEPES (pH 7.5), 100 mM NaCl, 0.5 mM EDTA, 0.1% Triton X-100, 10% glycerol). 35S-Labeled PCAF or RIP140 (2 ␮l) prepared with the TNT kit (Promega) was then added to GST-TR4 samples in 300 ␮l of the binding buffer. The samples were incubated at 4 °C for 90 min followed by three washes with a 20-bead volume of the binding buffer. The beads were collected by centrifugation and suspended in the binding buffer (20 ␮l) and 4⫻ SDS sample buffer (20 ␮l). Samples were divided into two parts, and an equal amount was resolved using SDS-PAGE (10% gel) on two separate gels. One gel was stained with Coomassie Blue, and the second gel was fixed, dried, and exposed to a PhosphorImager screen (GE Healthcare) overnight to detect the bound PCAF and RIP140 proteins. RESULTS

Expression and Purification of TR4 in Sf21 Cells—Many studies including our own have validated the expression of recombinant proteins in insect cells as an efficient way to generate large quantities of low abundance proteins, such as transcription factors, for the study of protein post-translational modification (PTM) (20 –22). Although the stoichiometry appears variable, the sites of modification identified in proteins expressed in insect cells are mostly the same as those found in proteins expressed in mammalian cells (20 –22). We took this approach to express mouse HA-tagged TR4 protein in Sf21 cells followed by purification over anti-HA-agarose beads, allowing the enrichment of purified protein as shown on an 8% SDS-polyacrylamide gel (Fig. 1). Mass Spectral Analysis of TR4 —The proteins were subjected to in-gel trypsin digestion. The tryptic digests were first analyzed on a MALDI-TOF mass spectrometer to identify TR4 and to predict any possible forms of modification according to the mass shift analysis. The MALDI-TOF-MS data were subjected to a MASCOT search at the National Center for Biotechnology Information (NCBI) data bank. The search result confirmed the identification of TR4 protein for 60% sequence coverage (data not shown). The mass shift analysis of the spectral data indicated the presence of several phosphopep-

2074


FIG. 1. Affinity purification of HA-TR4 from Sf21 cells. Affinity purification of HA-TR4 on anti-HA-agarose affinity beads yielded TR4 purified by SDS-PAGE.

tides. However, the MS/MS analysis of those predicted phosphopeptides by MALDI did not yield any quality MS/MS spectrum for assigning the PTM sites accurately. This could be due to improper ionization of phosphopeptides by MALDI. Previously we have reported LC-ESI-MS/MS as a more efficient technique, as compared with MALDI-TOF-MS/MS, for the determination of Ser/Thr phosphorylation on nuclear corepressor RIP140. We identified 12 phosphorylation sites on RIP140 by applying this technique (18, 22). Therefore, we also used a similar LC-ESI-MS/MS analysis for the current TR4 study to identify its phosphorylation sites. The MS/MS data search on MASCOT revealed a 54% sequence coverage of the protein (Table I), covering mostly the N-terminal portion (spanning the AF-1 domain). We also used chymotrypsin digestion for the LC-ESI-MS/MS analysis, but this appeared to yield little improvement. However, an additional 6% coverage was added as shown in bold characters to the total coverage (54%) obtained by tryptic digestion, giving a final coverage of 60% (Table I). Mapping of Phosphorylation Sites on TR4 —To identify the phosphopeptides in the total ion chromatogram (TIC) of the tryptic digests, an 80-amu positive mass shift due to covalent modification of Ser/Thr or Tyr by phosphorylation was considered. Furthermore a 98-amu negative mass shift exerted by dehydroalanine formation from ␤-elimination of H3PO4 from Ser/Thr of the peptide was also taken into account for an indication of phosphorylation. The TIC of the LC-ESI-MS revealed three phosphopeptides from the tryptic digests of TR4 (Table II) located in the AF-1 domain. The MS/MS data of the phosphopeptides were analyzed manually to map the sites of phosphorylation. In addition, the MS/MS spectra of the modified peptides were always aligned to the unmodified version of the same peptide to compare each individual y or b ion for assigning the positions of phosphorylation on a peptide. The modified peptide spanning residues 8 –22 displayed a doubly charged precursor ion at m/z 833.4 (precursor mass, 1664.82 Da) in the TIC. The mass of the precursor ion showed an 80-unit positive mass shift as compared with the unmodified peptide. This suggested that the peptide could be modified by a single phosphorylation site (Table II). The CID of the precursor ion (m/z 833.4) of the peptide yielded a quality


TABLE I PMF by LC-ESI-MS/MS Tryptic digests of TR4 protein were subjected to LC-ESI-MS/MS. The full scan ion chromatogram was recorded in an IDA mode to acquire MS/MS data. The IDA data were searched online at MASCOT (www.matrixscience.com). The sequence coverage is shown below from the result of PMF. See Footnote 1 (fn1) for key.

TABLE II LC-ESI-MS profile of the tryptic phosphopeptides of TR4 The IDA data were searched online at MASCOT (www.matrixscience.com). The MS/MS data were analyzed manually to confirm the sequence of the modified and the unmodified forms of the same peptide identified by the data bank search. The full scan chromatograms were analyzed to assign the charged state, retention time, and intensities of the peptides. Sequences of identified phosphorylation sites were compared to consensus sites of all known protein kinases (23). The underlined amino acids represent the consensus sequence for kinases. The bold character indicates the phosphorylation site. m/z (z), M⫹c, RTd

Sequencea,b

Unmodified

Modified

⌬MSe Da

8

IQIISTDSAVASPQR22

49

QQFILTSPDGAGTGK63

64

VILASPETSSAK75

793.44 529.29 507.26 760.39 601.84

(2), (3), (3), (2), (2),

1584.85, 1584.85, 1518.77, 1518.77, 1201.66,

28.39 28.57 29.10 29.10 23.22

833.41 (2), 1664.82, 29.46

ⴙ80

533.91 (3), 1598.75, 30.64 800.37 (2), 1598.75, 30.46 641.82 (2), 1281.63

ⴙ80 ⴙ80

a

All three could be potentially phosphorylated by MAPK, but Ser19 could also be phosphorylated by Cdc2 protein kinase or CDK2-cyclin A. b MAPK: XX(S/T)P, XP(S/T)XX; Cdc2 protein kinase, CDK2-cyclin A: (S/T)PX(R/K), (K/R)(S/T)P, (S/T)P(K/R) where X can be any amino acid. c Retention time in minutes. d Precursor ion mass in Da. e The 80-unit mass difference between modified and unmodified peptides indicated Ser/Thr phosphorylation.

MS/MS spectrum, which provided almost a total coverage of the amino acid sequence of the peptide mostly contributed by y ions and in part b ions. In the MS/MS spectrum of the modified peptide, the singly charged y3 ion at m/z 400.24 (Fig. 2A, top) was identical to that of the unmodified peptide (Fig. 2A, bottom). However, the y4 ion of the modified peptide

appeared at m/z 469.25 instead of the corresponding y4 ion at m/z 487.30. This net loss of 18-unit mass could be originated either from ␤-elimination of H2O of the unmodified Ser19 or from the ␤-elimination of H3PO4 (loss of 98-unit mass) of the modified version of Ser19. The y5 ion of the modified version appeared at m/z 638.28 instead of m/z 558.30. This 80-unit


2075


2076



positive mass shift clearly indicated that Ser19 was covalently modified by phosphorylation. This also supported the possibility that the net loss of 18-unit mass from y4 ion of the modified peptide was attributed to the dehydroalanine version of the y4 ion generated from the loss of H3PO4 from Ser19. Similar 98-unit mass shifts were also observed at the y5, y6, y8, y9, y10, y11, y12, and y13 ions (Fig. 2A, top) of the modified peptide. They were totally absent in the MS/MS spectrum of the unmodified peptide (Fig. 2A, bottom). Furthermore the y7 ion at m/z 737.33, the y8 ion at m/z 895.39, and the y13 at m/z 1424.67 each displayed an 80-unit positive mass shift relative to the corresponding ions of the unmodified version. This result supported that Ser19 was phosphorylated. We then verified whether the same peptide could produce different species of modified peptides by a single phosphorylation on the other putative Ser/Thr residues at Ser12, Thr13, and Ser15 by individual analysis of each Ser/Thr residue. For this scenario, the MS/MS sequence coverage after the postulated modification sites could not match the peptide (data not shown). Together the data confirmed that only Ser19 was modified by phosphorylation in this peptide. A doubly charged precursor ion at m/z 800.37 and a triply charged precursor ion at m/z 533.91 for the modified peptide spanning aa 43– 69 appeared at 30.64 min in the TIC. The mass of the precursor ion (molecular mass, 1598.75 Da) showed an 80-unit mass shift as compared with the unmodified peptide. This suggested that the peptide was modified by phosphoric acid by a single covalent modification. The CID spectrum of the doubly charged precursor ion (m/z 800.37) showed extensive sequence coverage of the peptide particularly by intense y ions from y1 through y13 ions and b ions from b1 through b9 ions (Fig. 2B, top panel). The CID spectrum of the modified version displayed y ions, in particular from y1 to y8, which could be superimposed to the respective ions of the unmodified peptide (Fig. 2B, bottom). This suggested that the modification was located either at Ser55 or Thr54, respectively, at the y9 and y10 ions. The y9 ion at m/z 771.36 showed a 98-unit negative mass shift, which could be exerted either by ␤-elimination of H3PO4 from the modified peptide or elimination of H2O from the corresponding ion at m/z 789.37 of the unmodified peptide. However, the series of

y ions, from y10 through y13 ions, at m/z 970.39, 1083.47, 1196.59, and 1343.62, each of which displayed a positive 80-unit (y ⫹ P) mass shift relative to those of the unmodified form. Furthermore in the MS/MS spectrum, the y10 and y11 ions of the modified peptide also displayed ion products for ␤-elimination of phosphoric acid (y ⫺ P). These ions were absent in the spectrum of the unmodified peptide, supporting the ␤-elimination of H3PO4 rather than H2O. Together the data supported that the site of phosphorylation was located either at Thr54 or Ser55. Although the ␤-eliminated product of the y9 ion was absent in the unmodified peptide, it did not rule out the possibility of H2O loss from the unmodified version of the y9 ion. Therefore, it was difficult to ascertain whether the PTM site was located at Ser55 or Thr54. We then analyzed the b ion series of the modified peptide. The b ion at m/z 517.28 and the a6 ion at m/z 703.41 corresponded to the respective ions of the unmodified peptide. This analysis eliminated the possibility of Thr54 phosphorylation. The TIC of the modified peptide spanning aa 65–75 displayed a doubly charged precursor ion at m/z 641.82 (precursor mass, 1281.63 Da) instead of the doubly charged precursor ion at m/z 601.84 (precursor mass, 1201.66 Da) of the unmodified version (Table II). The 80-unit mass difference between the two versions of the peptide predicted the modification by phosphorylation. The MS/MS spectrum of the doubly charged precursor ion of the modified peptide showed y ions from y1 through y7 identical to those of the unmodified version (Fig. 2C). However, these y ions, from y8 through y10 ions, which appeared at m/z 886.35, 957.39, and 1070.47, respectively (Fig. 2C, top panel), each clearly showed an 80-unit positive mass shift as compared with the unmodified peptide. Therefore, the phosphorylation site was assigned to Ser68. Kinases Involved in TR4 Phosphorylation—Two sequence isoforms of TR4 (NCBI accession numbers U11688 (aa 1–596) and S75970 (aa 1– 629)) have been documented with 33amino acid (MATNMEGLVQHRVGTQQVAEVPRTQTSWPESPG) differences at the N terminus because of the presence of two potential translation initiation sites within the same reading frame. Our TR4 expression plasmid contained the complete cDNA for translating both variants. However, our mass data covered the sequence of the low molecular weight TR4

Fig. 2. Mapping of phosphorylation sites on TR4 by LC-ESI-MS/MS analysis. A, modified peptide (aa 8 –22, top): precursor m/z 833.41 (z ⫽ 2); and unmodified peptide (bottom): precursor m/z 793.44 (z ⫽ 2) and m/z 529.29 (z ⫽ 3). The y3 ions at m/z 400.24 of the modified and unmodified peptides were identical. The y4 ion at m/z 469.25 of the modified version was shifted by a 98-unit negative mass (y ⫺ P), and the y5 ion at m/z 638.28 was shifted by an 80-unit (y ⫹ P) positive mass. This confirmed Ser19 phosphorylation. B, modified peptide (aa 49 – 63, top): precursor m/z 533.91 (z ⫽ 3) and m/z 800.37 (z ⫽ 2); and unmodified peptide (bottom): precursor m/z 507.26 (z ⫽ 3) and m/z 760.39 (z ⫽ 2). The y8 ions at m/z 702.35 of the modified and unmodified peptides were identical. The y9 ion at m/z 771.36 of the modified version showed a 98-unit negative mass, and the y10 ion at m/z 970.39 shifted by an 80-unit positive mass. Thus, the phosphorylation site was assigned to Ser55. C, modified peptide (aa 64 –75, top): precursor m/z 641.82 (z ⫽ 2); and unmodified peptide (bottom): precursor m/z 601.84 (z ⫽ 2). The y7 ions at m/z 719.37 of the modified peptide corresponded to that of the unmodified peptide. The y8 ion of the modified version appeared at m/z 886.35 instead of m/z 806.37. This 80-unit positive mass shift could be attributed to Ser68 phosphorylation. The y ⫹ 80 (y ⫹ P) and y ⫺ 98 (y ⫺ P) represent y ions due to covalent modification by phosphorylation and loss of phosphate group, respectively. The y* or b* ions represent y or b ions caused by loss of ammonia. The yo and bo ions represent y or b ions caused by loss of H2O. The y⫹ or b⫹ represents the charge status of the y or b ions.


2077


isoform (aa 1–596, Table I). This could be due to the possibility that the low molecular weight TR4 variant was more abundantly expressed in the insect cells. Alternatively the N terminus of the large TR4 variant was not efficiently digested by trypsin for MS analysis. It would be important to explore in what physiological context a particular TR4 variant is preferentially expressed. It would also be equally important to determine the difference between the two variants in terms of their physiological functions. To identify the kinases involved in phosphorylation of its AF-1 domain, we conducted a kinase-specific consensus motif search. The consensus motifs for all known protein kinases (23) involved in the identified phosphorylation sites are shown in Table II. All three phosphorylation sites, at Ser19, Ser55, and Ser68, were potential targets for MAPK-mediated phosphorylation. Only Ser19 could be phosphorylated by other kinases such as Cdc2 protein kinase or Cdk2-cyclinA. The Effects of MAPK on TR4 Activities—Because all phosphorylated residues confirmed by MS were potential sites for MAPK, we then focused on the role of MAPK-mediated phosphorylation in regulating the biological activity of TR4. To determine more comprehensively the biological activities of its AF-1, we used both the TR4 full-length protein and a TR4 construct devoid of LBD, named TR4⌬LBD, where potential complication from ligand binding could be avoided (Fig. 3A). A cell-based reporter assay containing a TR4-responding DR1 element (ATGGCAGAGGTCA) shown by underlined characters was used that was from its natural target in the mouse hepatic control region-1 of apoE/C-I/C-II gene cluster (Fig. 3B) (13). The MAPK activator/inhibitor was first used to determine the effects of MAPK pathway on the biological activity of the full-length TR4 and the TR4⌬LBD. As shown in Fig. 3B, both TR4 (5-fold) and TR4⌬LBD (4-fold) could activate the DR1 reporter, suggesting an LBD-independent activation of TR4. Furthermore MAPK activation by anisomycin led to a nearly complete loss of TR4 full-length protein activity and a significant reduction of TR4⌬LBD activity (AF-1) for the DR1 reporter. On the contrary, the inhibition of MAPK by PD98059 significantly increased the activation of the reporter by both constructs of TR4 (2.4-fold). Together these data revealed a role for MAPK-mediated phosphorylation in modulating the biological activity of TR4 that was independent of the effect of ligands or the LBD. Effects of Site-specific Phosphorylation on TR4 Activity—To unambiguously verify the role of each specific phosphorylated site in the manifestation of TR4 activity, we conducted sitespecific mutagenesis analyses for each identified phosphorylated residue in the context of both TR4 full-length and the TR4⌬LBD proteins (Fig. 4). We generated CN (mimicking dephosphorylation) mutant by replacing phosphorylated Ser with Ala that cannot be phosphorylated by MAPK. Furthermore because one primary effect of protein phosphorylation is to increase site-specific negative charge of the protein, we generated CP (mimicking hyperphosphorylation) mutants of TR4 by replacing phospho-Ser residues, singly or in combi-

2078


FIG. 3. Effect of MAPK on TR4 biological activity. A, graphical structures of TR4 constructs. Two variants of TR4 proteins (aa 1–596 and aa 1– 629) are shown. There is a 33-amino acid addition in the large TR4 variant because of an alternative translational start site. TR4⌬LBD represents a TR4 construct devoid of LBD. B, biological activity of TR4 was monitored using the apoE-DR1-tk-luciferase reporter in the presence of 3 ␮M PD 98059 (MAPK inhibitor) and MAPK activator (anisomycin) as compared with control. Both constructs activated the DR1 reporter, suggesting a ligand-independent activation function of TR4. Inhibition of MAPK was found to potentiate TR4 activation of the reporter, whereas MAPK activation by anisomycin inversely affected the TR4 activity. (* versus **, significantly different (p ⬍ 0.01, Student’s t test)).

nation, with Glu. First we screened all three phosphorylated serine residues individually using the wild type, the CN, and the CP mutants in the context of both the full-length and the truncated proteins with regard to effects on the activation of the reporter. The data showed that the CN mutant of either Ser19 or Ser68 dramatically enhanced TR4 activation of the reporter as compared with the wild type TR4 full length (Fig. 4A, compare Wild and CN mutants). The activity of each individual CN mutant could be further enhanced by MAPK inhibition. However, the mutation on phospho-Ser55 to Ala had no effect on TR4 activity. The CN double mutant of both Ser19 and Ser68 robustly activated the reporter as compared with the wild type TR4 (Fig. 4A). Furthermore the inhibition of MAPK elicited no additional effect on this mutant, confirming that both phosphorylated Ser19 and Ser68 residues were essential for regulating the activity of TR4. The double CP (19CP⫹68CP) mutant repressed the reporter by 3-fold and


FIG. 4. Effects of site-specific phosphorylation on TR4 activities. A, biological activities of TR4 and its mutants assessed in trans-activation assays. (* versus **, not significantly different (p ⬍ 0.05, Student’s t test); # versus ##, significantly different (p ⬍ 0.01, Student’s t test)). B, effects of site-specific phosphorylation on TR4 devoid of the LBD (TR4⌬LBD). (* versus **, not significantly different (p ⬍ 0.05, Student’s t test)). C, effects of TR4 phosphorylation on the endogenous apoE gene expression. The CN mutant elevated the expression of apoE, whereas the CP mutant downregulated the apoE gene expression. IB, immunoblot.

10 –12-fold as compared with the wild type TR4 and the CN (19CN⫹68CN) mutant, respectively. This result further supported the notion that hypophosphorylated TR4 acted as an activator, whereas the hyperphosphorylated TR4 functioned as a repressor, confirming that the activity of TR4 could be modulated by the coordination of its phosphorylation status. To verify the effect of phosphorylation on the ligand-independent activity of wild type TR4, CN and CP mutants were also made in the context of the TR4⌬LBD protein (Fig. 4B). It appeared that the CN mutation at Ser19/Ser68 enhanced the activity of TR4⌬LBD, whereas the CP mutations enhanced the repressive activity of this truncated protein that was devoid of LBD. Taken together, these data confirmed that TR4 activity could be modulated by the phosphorylation on the AF-1 domain (the N-terminal domain) independently of the potential effects of specific ligands. Genetic and biochemical studies demonstrated that TR4

could positively regulate the apoE gene expression (13). To test the effects of phosphorylation on TR4 activity in a physiological context, we conducted gain- and loss-of-function studies by expressing the TR4 wild type and mutants in mouse hepatoma H2.35 cells and monitoring its effect on endogenous target, apoE, gene expression (Fig. 4C). As expected, the CN mutant enhanced the expression of apoE gene as compared with the wild type. On the other hand, the CP mutant down-regulated the expression of apoE. These data were consistent with the findings using the DR1 reporter in terms of the effects of phosphorylation on the biological activity of TR4, verifying the physiological relevance of this finding. It was concluded that hypophosphorylated TR4 positively and hyperphosphorylated TR4 negatively regulated the expression of its target genes such as apoE. The Effect of TR4 Phosphorylation on Its Interaction with Cofactors—To provide a mechanistic insight into the effects


2079


of phosphorylation on the regulatory activities of TR4, the potential coregulators of TR4 were evaluated. Both the wild type and the mutant were examined using the DR1 reporter in the presence of PCAF as a coactivator and RIP140 as a corepressor (Fig. 5A). It appeared that the wild type full-length TR4 activity was positively regulated by coactivator PCAF and negatively regulated by the corepressor RIP140. However, TR4⌬LBD activity was positively affected by PCAF but not affected by RIP140. This suggested that the activation of TR4 AF-1 domain was attributed to, at least in part, PCAF. The CN mutants of both the full-length TR4 and the TR4⌬LBD robustly activated the DR1 reporter in the presence of PCAF. On the other hand, the CP mutant of the wild type TR4 repressed the reporter in the presence of RIP140, suggesting that the hyperphosphorylated TR4 preferentially recruited corepressor RIP140 to exert its repressive activity. Interestingly although the CP mutant of the TR4⌬LBD was also repressive, it was not affected by RIP140. This was consistent with the finding that RIP140 interacted with the LBD but not the AF-1 (see Fig. 5B). This could be due to reduced interaction with coactivator PCAF, or alternatively it could indicate a potentially different repressive mechanism mediated by other unidentified corepressors, which could specifically recognize the hyperphosphorylated AF-1 domain. To determine whether the effect of TR4 phosphorylation on its biological activity was due to the effect on the recruitment of these coregulators, we carried out in vitro protein-protein interaction tests (Fig. 5B, left). The CP mutants were made in the context of GST fusions. To map the interacting domains of TR4 for different coregulators, we first tested the wild type TR4LBD (containing only LBD), the TR4⌬LBD (deleting the LBD), and the full-length proteins. As expected, the full-length TR4 interacted with both PCAF and RIP140. The data also clearly demonstrated that the LBD of TR4 did not interact with PCAF, whereas it strongly interacted with RIP140 (Fig. 5B, left). The TR4⌬LBD was found to interact strongly with PCAF but not with RIP140. These data were consistent with the findings in the cell-based assay, which also revealed the importance of the LBD for the corepressive function of RIP140 but not for the coactivating function of PCAF (Fig. 5A). The interaction tests were then conducted using mutants and the wild type (Fig. 5B, right). As expected, the wild type and the CN mutants of both TR4 and TR4⌬LBD, which should not be phosphorylated, preferentially interacted with PCAF. For the CP mutants of both TR4 and TR4⌬LBD, their interaction with PCAF was dramatically reduced, but the interaction of TR4 full-length CP mutant with RIP140 was enhanced. Together the in vivo and in vitro data supported that the hyperphosphorylated TR4 preferentially recruited corepressor and the hypophosphorylated TR4 recruited coactivators. The effects of phosphorylation on the ability of TR4 to recruit coregulators were in agreement with the effects on its biological activities assessed in both the reporter and the natural target gene systems.

2080


FIG. 5. Effect of TR4 phosphorylation on cofactor recruitment. A, effects of cofactors on modulating TR4 activity for the full-length protein (top) and the truncated TR4⌬LBD (bottom). B, direct protein-protein interactions (in vitro pulldown) of cofactors with TR4. PCAF interacted with AF-1 domain (TR4⌬LBD), whereas RIP140 interacted with LBD (left panel). The CP mutants of both TR4 and TR4⌬LBD interacted less effectively with PCAF, but the CP mutant of TR4 full-length protein preferentially interacted with RIP140 as compared with the wild type and the CN mutant (right panel). The * in the left panel (bottom, lanes 2– 4) and arrows in the right panel (bottom, lanes 2– 4 and lanes 5–7) depict the specific protein input controls. Lane M, molecular mass markers. WT, wild type; TNT, in vitro translated 35S-labeled proteins.


DISCUSSION

The primary sequence of a protein is dictated by the genetic code, and the functional diversity of a protein can be achieved by the lamination of different PTMs. PTM is a hallmark of proteins involved in signal transduction, allowing the proteins to rapidly respond to extracellular signals by triggering a cascade of cellular responses (24). Epigenetic protein modification by phosphorylation and a variety of other PTMs, including acetylation, methylation, and glycosylation, are now known to regulate protein functions and play important roles in multiple cellular processes including DNA repair, protein stability, nuclear translocation, protein-protein interactions, cellular proliferation, differentiation, and apoptosis (25–27). The major challenge is the identification of PTMs occurring on proteins in vivo. It has been difficult, if not impossible, to purify mammalian proteins to homogeneity for analysis (28). In this work, we carried out the identification of phosphorylated sites of TR4 by examining proteins expressed and purified from Sf21 cells. Previous studies with nuclear receptors and some other coregulators showed variation mostly in the stoichiometry but not the sites of PTM on proteins purified from insect cells in most studies (20 –22). Therefore, we predicted that the modified residues on TR4 purified from insect cells could have equal potential for the sites of PTM of this protein in mammalian cells. Importantly these predicted sites were verified using the site-specific mutagenesis approach, which confirmed the biological significance of two specific sites of phosphorylation. We identified three potential phosphorylated sites on TR4 from insect cells using LC-ESI-MS/MS analysis without any enrichment of the phosphopeptides. The phosphorylated sites were Ser19, Ser55, and Ser68, which were located in the AF-1 domain. All these residues were potential for MAPK phosphorylation according to the consensus motif analysis of TR4 for kinase specificity (see Tables I and II). However, some other sites, which could also be potential for MAPK-mediated phosphorylation, were not found according to the MS data as highlighted in Table I. However, we could not confirm the phosphorylation status of two residues located within the uncovered sequence of DBD and LBD. Furthermore a significant portion of the TR4 sequences in the LBD, DBD, and hinge regions was not covered by our PMF data using both trypsin and chymotrypsin for digestion. The theoretical digestion of TR4 showed very few putative digestion sites (Lys/Arg) in the LBD and many sites near the DBD and hinge regions in tryptic digestion. This could generate very large and small peptide fragments from LBD and DBD/hinge regions. These are usually difficult to detect by current MS facilities for PMF. However, the studies of our mutant proteins verified the significance of phosphorylation of the AF-1 domain per se, although it could not rule out potential contributions from the LBD and the DBD/hinge regions. Certain nuclear receptors have been reported to interact

with cofactors through their DBDs independently of their LBDs (29, 30). In this study, we have also shown that TR4 interaction with PCAF does not require LBD. Presumably the PCAF-interacting domain of TR4 is also located at the DBD. With respect to AF-1 function, a synergy has been shown for DBD-mediated cofactor interaction and AF-1 activity for progesterone receptors (31). Our data suggested that phosphorylation of AF-1 domain of TR4 antagonized the synergy between DBD and AF-1 function. Interestingly the transcriptional activity of another orphan receptor, TR2, is positively regulated by protein kinase C-mediated phosphorylation (15, 16), correlated with its enhanced interaction with PACF through the DBD. Because a significant portion of TR4 DBD region was not covered in our mass data, it is unclear whether any protein kinase C-mediated phosphorylation exists at the DBD region of TR4. To this end, it was noted that MAPK-mediated phosphorylation of AF-1 of PPAR-␣ enhanced its transcriptional activity, but MAPK-mediated phosphorylation of AF-1 of PPAR-␥ was shown to suppress its transcriptional activity (32). Thus, there seems to be no general rule for the biological manifestation of AF-1 phosphorylation. Given that AF-1 domain is the most divergent region among nuclear receptors (NRs), without structural information it is probably not a practical exercise to decipher how a single kinase-mediated phosphorylation at the AF-1 domain can differentially regulate the property and function of each receptor. The structural information about AF-1 is urgently needed. Reversible protein phosphorylation is known to control a wide range of biological activities (7–9, 13, 33, 34). Many NRs have been found to be modified by phosphorylation (32). The biological activities of NRs are controlled, in many ways, by kinase signaling pathways. In this study, we described a bidirectional regulation of TR4 function by the MAPK-mediated phosphorylation. We addressed a specific ligand-independent activation of TR4 triggered by phosphorylation of its AF-1 domain. We found that phosphorylated TR4 recruits corepressor RIP140 to exert its repressive function, whereas the dephosphorylated TR4 preferentially recruits coactivator PCAF to activate the genes. Because activating and repressive activities of the full-length TR4 are higher as compared with the truncated TR4 devoid of the LBD (TR4⌬LBD) and AF-1 could modulate RIP140 binding to TR4 via its LBD, a synergy between AF-1 phosphorylation and AF-2 domain may exist to modulate the activity of TR4. This can be verified only if specific ligands for TR4 can be identified in the future. As an orphan nuclear receptor, the physiological function of TR4 has been difficult to assess. Genetic knock-out studies have first indicated the involvement of TR4 in some vital physiological functions (7–9). For examples, TR4 regulates apolipoprotein E/C-I/C-II gene cluster (13), which is very actively involve in progression and regression of atherosclerosis and neuronal regeneration and degeneration processes. TR4 appears to be also essential for normal spermatogenesis, motor coordination, and cerebellum development (7–9). Our demon-


2081


stration of a specific ligand-independent activating mechanism of TR4 mediated by MAPK-triggered phosphorylation as well as the identification of specific sites of phosphorylation provides an enormous opportunity for future studies of the molecular mechanisms mediating the biological activity of TR4 in numerous biological processes even without the presence of ligands.

13.

14.

Acknowledgments—We thank Y. Chen for help in purification of TR4 protein. We also thank the staff of the Mass Spectrometry Consortium for the Life Sciences, University of Minnesota, Department of Biochemistry, Molecular Biology, and Biophysics at St. Paul for recording the mass spectra for the protein samples. * This work was supported by National Institutes of Health Grants DA11190, DA11806, DK54733, DK60521, and K02-DA13926 (to L.-N. W.). 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 U.S.C. Section 1734 solely to indicate this fact. ‡ Both authors contributed equally to this work. § To whom correspondence should be addressed: Dept. of Pharmacology, University of Minnesota Medical School, 6-120 Jackson Hall, 321 Church St. S.E., Minneapolis, MN 55455-0217. Tel.: 612625-9402; Fax: 612-625-8408; E-mail: [email protected]. REFERENCES 1. Beato, M., Herrlich, P., and Schu¨tz, G. (1995) Steroid hormone receptors: many actors in search of a plot. Cell 83, 851– 857 2. Mangelsdorf, D. J., Thummel, C., Beato, M., Herrlich, P., Schu¨tz, G., Umesono, K., Blumberg, B., Kastner, P., Mark, M., Chambon, P., and Evans, R. M. (1995) The nuclear receptor superfamily: the second decade. Cell 83, 835– 839 3. Laudet, V. (1997) Evolution of the nuclear receptor superfamily: early diversification from an ancestral orphan receptor. J. Mol. Endocrinol. 19, 207–226 4. Chang, C., Da Silva, S. L., Ideta, R., Lee, Y., Yeh, S., and Burbach, J. P. (1994) Human and rat TR4 orphan receptors specify a subclass of the steroid receptor superfamily. Proc. Natl. Acad. Sci. U. S. A. 91, 6040 – 6044 5. Hirose, T., Fujimoto, W., Tamaai, T., Kim, K. H., Matsuura, H., and Jetten, A. M. (1994) TAK1: molecular cloning and characterization of a new member of the nuclear receptor superfamily. Mol. Endocrinol. 8, 1667–1680 6. Lee, Y. F., Young, W. J., Burbach, J. P. H., and Chang, C. (1998) Negative feedback control of the retinoid-retinoic acid/retinoid X receptor pathway by the human TR4 orphan receptor, a member of the steroid receptor superfamily. J. Biol. Chem. 273, 13437–13443 7. Chen, Y. T., Collins, L. L., Uno, H., and Chang, C. (2005) Deficits in motor coordination with aberrant cerebellar development in mice lacking testicular orphan nuclear receptor 4. Mol. Cell. Biol. 25, 2722–2732 8. Collins, L. L., Lee, Y. F., Heinlein, C. A., Liu, N. C., Chen, Y. T., Shyr, C. R., Meshul, C. K., Uno, H., Platt, K. A., and Chang, C. (2004) Growth retardation and abnormal maternal behavior in mice lacking testicular orphan nuclear receptor 4. Proc. Natl. Acad. Sci. U. S. A. 101, 15058 –15063 9. Shyr, C. R., Collins, L. L., Mu, X. M., Platt, K. A., and Chang, C. (2002) Spermatogenesis and testis development are normal in mice lacking testicular orphan nuclear receptor 2. Mol. Cell. Biol. 22, 4661– 4666 10. Lee, Y. F., Shyr, C. R., Thin, T. H., Lin, W. J., and Chang, C. (1999) Convergence of two repressors through heterodimer formation of androgen receptor and testicular orphan receptor-4: a unique signaling pathway in the steroid receptor superfamily. Proc. Natl. Acad. Sci. U. S. A. 96, 14724 –14729 11. Lee, Y. F., Pan, H. J., Burbach, J. P., Morkin, E., and Chang, C. (1997) Identification of direct repeat 4 as a positive regulatory element for the human TR4 orphan receptor. A modulator for the thyroid hormone target genes. J. Biol. Chem. 272, 12215–12220 12. Lee, Y. F., Young, W. J., Lin, W. J., Shyr, C. R., and Chang, C. (1999)

2082


15.

16.

17.

18.

19.

20. 21.

22.

23.

24. 25. 26. 27. 28. 29.

30.

31.

32. 33.

34.

Differential regulation of direct repeat 3 vitamin D3 and direct repeat 4 thyroid hormone signaling pathways by the human TR4 orphan receptor. J. Biol. Chem. 274, 16198 –16205 Kim, E., Xie, S., Yeh, S. D., Lee, Y. F., Collins, L. L., Hu, Y. C., Shyr, C. R., Mu, X. M., Liu, N. C., Chen, Y. T., Wang, P. H., and Chang, C. (2003) Disruption of TR4 orphan nuclear receptor reduces the expression of liver apolipoprotein E/C-I/C-II gene cluster. J. Biol. Chem. 278, 46919 – 46926 Lee, Y. F., Lee, H. J., and Chang, C. (2002) Recent advances in the TR2 and TR4 orphan receptors of the nuclear receptor superfamily. J. Steroid Biochem. Mol. Biol. 81, 291–308 Khan, S. A., Park, S. W., Huq, M. D., and Wei, L. N. (2006) Ligandindependent orphan receptor TR2 activation by phosphorylation at the DNA-binding domain. Proteomics 6, 123–130 Khan, S. A., Park, S. W., Huq, M. D., and Wei, L. N. (2005) Protein kinase C-mediated phosphorylation of orphan nuclear receptor TR2: effects on receptor stability and activity. Proteomics 5, 3885–3894 Lee, C. H., Chinpaisal, C., and Wei, L. N. (1998) A novel nuclear receptor heterodimerization pathway mediated by orphan receptors TR2 and TR4. J. Biol. Chem. 273, 25209 –25215 Huq, M. D., Khan, S. A., Park, S. W., and Wei, L. N. (2005) Mapping of phosphorylation sites of nuclear corepressor receptor interacting protein 140 by liquid chromatography-tandem mass spectroscopy. Proteomics 5, 2157–2166 Huq, M. D., and Wei, L. N. (2005) Post-translational modification of nuclear co-repressor receptor-interacting protein 140 by acetylation. Mol. Cell. Proteomics 4, 975–983 Cheng, X., and Hart, G. W. (2000) Glycosylation of the murine estrogen receptor-␣. J. Steroid Biochem. Mol. Biol. 75, 147–148 Wu, R. C., Qin, J., Yi, P., Wong, J., Tsai, S. Y., Tsai, M. J., and O’Malley, B. W. (2004) Selective phosphorylations of the SRC-3/AIB1 coactivator integrate genomic responses to multiple cellular signaling pathways. Mol. Cell 15, 937–949 Gupta, P., Huq, M. D., Khan, S. A., Tsai, N. P., and Wei, L. N. (2005) Regulation of co-repressive activity of and HDAC recruitment to RIP140 by site-specific phosphorylation. Mol. Cell. Proteomics 4, 1776 –1784 Mann, M., Ong, S. E., Gronborg, M., Steen, H., Jensen, O. N., and Pandey, A. (2002) Analysis of protein phosphorylation using mass spectrometry: deciphering the phosphoproteome. Trends Biotechnol. 20, 261–268 Pawson, T., and Scott, J. D. (1997) Signaling through scaffold, anchoring, and adaptor proteins. Science 278, 2075–2080 Pawson, T. (1995) Protein modules and signalling networks. Nature 373, 573–580 Kouzarides, T. (2000) Acetylation: a regulatory modification to rival phosphorylation. EMBO J. 19, 1176 –1179 McBride, A. E., and Silver, P. A. (2001) State of the arg: protein methylation at arginine comes of age. Cell 106, 5– 8 Zhou, H., Watts, J. D., and Aebersold, R. (2001) A systematic approach to the analysis of protein phosphorylation. Nat. Biotechnol. 19, 375–378 Wardell, S. E., Kwok, S. C., Sherman, L., Hodges, R. S., and Edwards D. P. (2005) Regulation of the amino-terminal transcription activation domain of progesterone receptor by a cofactor induced protein folding mechanism. Mol. Cell. Biol. 25, 8792– 8808 Iwasaki, T., Takeshita, A., Miyazaki, W., Chin, W. W., and Koibuchi, N. (2006) The interaction of TR␤1-N terminus with steroid receptor coactivator-1 (SRC-1) serves a full transcriptional activation function of SRC-1. Endocrinology 147, 1452–1457 Wardell, S. E., Boonyaratanakornkit, V., Adelman, J. S., Aronheim, A., and Edwards, D. P. (2002) Jun dimerization protein 2 functions as a progesterone receptor N-terminal domain coactivator. Mol. Cell. Biol. 22, 5451–5466 Rochette-Egly, C. (2003) Nuclear receptors: integration of multiple signalling pathways through phosphorylation. Cell. Signal. 15, 355–366 Koritschoner, N. P., Madruga, J., Knespel, S., Blendinger, G., Anzinger, B., Otto, A., Zenke, M., and Bartunek, P. (2001) The nuclear orphan receptor TR4 promotes proliferation of myeloid progenitor cells. Cell Growth Differ. 12, 563–572 Tanabe, O., Katsuoka, F., Campbell, A. D., Song, W., Yamamoto, M., Tanimoto, K., and Engel, J. D. (2002) An embryonic/fetal beta-type globin gene repressor contains a nuclear receptor TR2/TR4 heterodimer. EMBO J. 21, 3434 –3442