Published December 5, 2012, doi:10.4049/jimmunol.1201657 The Journal of Immunology

Nuclear Translocation of MEK1 Triggers a Complex T Cell Response through the Silencing Mediator of Retinoid and Thyroid Hormone Receptor

Lei Guo,* Chaoyu Chen,* Qiaoling Liang,* Mohammad Zunayet Karim,* Magdalena M. Gorska,*,† and Rafeul Alam*,†

MEK1 phosphorylates ERK1/2 and regulates T cell generation, differentiation, and function. MEK1 has recently been shown to translocate to the nucleus. Its nuclear function is largely unknown. By studying CD4 T cells, we demonstrate that a low level of MEK1 is present in the nucleus of CD4 T cells under basal conditions. T cell activation further increases the nuclear translocation of MEK1. MEK1 interacts with the nuclear receptor corepressor silencing mediator of retinoid and thyroid hormone receptor (SMRT). MEK1 reduces the nuclear level of SMRT in an activation-dependent manner. MEK1 is recruited to the of c-Fos upon TCR stimulation. Conversely, SMRT is bound to the c-Fos promoter under basal conditions and is removed upon TCR stimulation. We examined the role of SMRT in regulation of T cell function. Small interfering RNA-mediated knockdown of SMRT results in a biphasic effect on cytokine production. The production of the cytokines IL-2, IL-4, IL-10, and IFN-g increases in the early phase (8 h) and then decreases in the late phase (48 h). The late-phase decrease is associated with inhibition of T cell proliferation. The late-phase inhibition of T cell activation is, in part, mediated by IL-10 that is produced in the early phase and, in part, by b-catenin signaling. Thus, we have identified a novel nuclear function of MEK1. MEK1 triggers a complex pattern of early T cell activation, followed by a late inhibition through its interaction with SMRT. This biphasic dual effect most likely reflects a homeostatic regulation of T cell function by MEK1. The Journal of Immunology, 2013, 190: 000–000.

itogen-activated play an essential role motif leads to the nuclear translocation of MEK1 (12, 13). MEK1 in many fundamental cellular functions, including cell also has an N-terminal nuclear export signal (ALQKKLEE- M proliferation, differentiation, survival, locomotion, and LELDE, residues 32–44). The presence of the nuclear localization secretion (1). ERK1 and ERK2 represent a major subfamily of motif and an export signal allows MEK1 to shuttle between the MAPKs. They are activated through unique threonine–tyrosine nucleus and cytosol (14). The exact function of MEK1 in the phosphorylation. MEK1 and MEK2 specifically phosphorylate nucleus is unclear. MEK1, but not MEK2, was reported to cause

by guest on October 1, 2021. Copyright 2012 Pageant Media Ltd. the TEY motif of ERK1 and ERK2. MEK1 knockout is em- nuclear translocation of ERK2 (15). In addition to activating bryonic lethal (2, 3). Pharmacological inhibitors of MEK1/2 po- ERK1 and ERK2, MEK1 phosphorylates STAT5 (16) and MyoD tently inhibit ERK1/2 activation. This approach allowed extensive (17). The phosphorylation of these factors, especially characterization of the role of the MEK–ERK1/2 pathway in cellular MyoD, is likely to occur in the nucleus. MEK1 also interacts with function. The MEK–ERK1/2 signaling pathway plays an important the nuclear receptor peroxisome proliferator-activated receptor role in different stages of thymic differentiation of CD4 and CD8 (PPAR)-g and the nuclear corepressor, silencing mediator of ret- T cells (4–6). It is also important for mature T cell activation (7) inoid and thyroid hormone receptor (SMRT, also known as nuclear and differentiation (8). receptor corepressor [NCoR]2), and triggers their nuclear export MEK1 has previously been localized to the cytosol (9) and late (14, 18, 19). MEK1-mediated phosphorylation of SMRT prevents

http://classic.jimmunol.org endosome (10). Recent studies have identified a novel and non- its interaction with the nuclear receptors. The interaction with canonical nuclear localization motif (11). Phosphorylation of this SMRT was studied in an overexpression model with fusion pro- teins (16, 17). The direct interaction of endogenous MEK1 and SMRT in primary cells remains unknown. *Division of Allergy and Immunology, Department of Medicine, National Jewish SMRT is a NCoR-related transcriptional corepressor (18, 20– Health, Denver, CO 80206; and †School of Medicine, University of Colorado Denver, 22) and a component of a multimolecular repressor complex that Denver, CO 80045

Downloaded from includes mSin3, TBL1, TBLR1, GPS2, and histone deacetylase Received for publication June 15, 2012. Accepted for publication November 5, 2012. (HDAC)3 (23). The presence of HDACs in the complex prevents This work was supported by National Institutes of Health Grants RO1 AI091614, gene transcription. The SMRT targets two major groups of mol- AI68088, PPG HL36577, and N01 HHSN272200700048C. ecules in the nucleus. The first group includes the nuclear recep- Address correspondence and reprint requests to Dr. Rafeul Alam, Division of Allergy and Immunology, Department of Medicine, National Jewish Health, 1400 Jackson tors retinoic acid receptor (RAR), retinoid X receptor (RXR), liver Street, Denver, CO 80206. E-mail address: [email protected] X receptor (LXR), receptor, and thyroid hormone Abbreviations used in this article: ChIP, ; HDAC, receptors (21, 22, 24, 25). The second group represents the tran- histone deacetylase; LXR, liver X receptor; NCOR, nuclear receptor corepressor; scription factors, as follows: AP1, NF-kB, SRF, MEF2C, FOXP1, PPAR, peroxisome proliferator-activated receptor; RAR, retinoic acid receptor; RXR, retinoid X receptor; shRNA, short hairpin RNA; siNT, nontargeting siRNA ETO1/2, and Ets family members (26–28). SMRT represses the pool; siRNA, short interfering RNA; siSMRT, short interfering SMRT; SMRT, si- histone 3 K27 methylase JMJD3, which derepresses many poly- lencing mediator of retinoid and thyroid hormone receptor; TPT, Threonine-Proline- comb group-silenced genes (29). SMRT knockout is embryonic Threonine. lethal due to malformation of heart and palate (27). The function Copyright Ó 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00 of SMRT in T cells is unknown.

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1201657 2 NUCLEAR MEK1 AND SMRT IN T CELL FUNCTION

In this manuscript, we examined nuclear translocation of MEK1 tracellular staining, cells were further permeabilized in ice-cold 90% and its consequences following activation of human CD4 T cells. methanol for 30 min before staining with Abs. Primary and fluorescence- We specifically examined the interaction of MEK1 with SMRT and conjugated secondary Abs were diluted in 0.1% BSA dissolved in PBS at desired concentration, according to the supplier’s recommendation, and the effect of SMRT inhibition on T cell function. We show that incubated for 30 min. Cells were incubated with primary and secondary Ab MEK1 interacts with SMRT in the nucleus. Both MEK1 and SMRT plus 10% goat serum for 30 min at room temperature. Fluorescence in- bind to the c-Fos promoter and regulate its transcription. SMRT tensity was detected by CYAN (Beckman Coulter, Fullerton, CA), unless knockdown results in an early-phase stimulation, followed by a late- otherwise stated. Data were analyzed by the software FlowJo (Tree Star, Ashland, OR). phase inhibition of T cell activation. IL-10 and b-catenin signaling, induced in the early stimulation phase, play an important role in the Immunofluorescence staining late-phase negative feedback inhibition of T cell activation. Immunofluorescence staining was performed, as described previously (34, 35). CD4 T cells were fixed in 2% paraformaldehyde, and then Materials and Methods placed and dried on slides. Slides were blocked for 1 h in 5% serum, Human subjects 0.03% Triton X-100, and then incubated at 4˚C overnight with primary Ab diluted at the supplier’s recommendation or isotype control Ab. Alexa The protocol for human blood draw and T cell signaling studies was ap- Fluor fluorochrome-conjugated secondary Ab was used at 1 mg/ml. Cells proved by the Institutional Review Board of National Jewish Health were analyzed by a Nikon TE 2000 inverted microscope using a 3100 (Denver, CO). Blood was drawn from healthy subjects upon written consent. objective. The microscope is equipped with a z-motor (Prior Scientific, Blood was anticoagulated with EDTA. In some experiments, buffy coat (red Rockland, MA), excitation and emission filter wheels, and a CoolSnap HQ cell-depleted leukocyte pack) was obtained from the blood bank donor camera (Roper Scientific-Photometrics, Tuscon, AZ). The data acquisition through the Bonfils Blood Center. and analysis were done by Metamorph, version 7.1 (Molecular Devices, Reagents Downingtown, PA). Images of experimental groups to be compared were acquired using the same software setting (e.g., same exposure time). The The mouse monoclonal anti-human CD3 (clone OKT3) and mouse pixel level of the images was raised to an arbitrary level (a process called monoclonal anti-human CD28 (clone CD28.2) were obtained from eBio- thresholding) for quantitative analyses to eliminate the input from the science (San Diego, CA). A concentration of 1–5 mg/ml of each Ab was background fluorescence. The same threshold was applied to all images in used for stimulation. Anti-human IL-2, IL-4, IL-10, and IFN-g ELISA kits comparison studies. Thresholded areas within the selected regions were were obtained from BD Biosciences. The dilution of each Ab was per- used for integrated fluorescence intensity measurement. The integrated formed according to manufacturer’s instructions. Rabbit monoclonal anti– intensity is defined as a sum of intensities of all selected pixels. p-ERK1/2, anti-pp38, and anti-MEK1 (clone 47E6, catalog 9126) Abs were obtained from Cell Signaling Technology (Danvers, MA), and a di- Real-time PCR lution of 1:1000 was used for each Ab, as instructed by each Ab data sheet Real-time PCR was done as described previously (35). RNA purified from for Western blotting purpose. For immunostaining purpose, the same anti- TRIzol (Invitrogen) lysates was reverse transcribed using ImProm-II Re- MEK1 Ab was used with a dilution of 1:50. Fluorescently labeled sec- verse Transcription System (Promega), according to the manufacturer’s ondary Abs were purchased from Molecular Probes (Invitrogen). PE- and protocol. Absolute SYBR Green ROX Mix (Thermo Fisher Scientific) was FITC-labeled mouse anti-FOXP3 mAbs were purchased from BD used for DNA amplification. Real-time PCR was performed with the ABI Biosciences (San Jose, CA). Rabbit anti-SMRT Ab was purchased from Prism 7000 Sequence Detection System (Applied Biosystems). Affinity Bioreagents/Thermo Scientific/Pierce Antibodies (Rockford, IL). For Western blotting purpose, a concentration of 4 mg/ml was used. All Chromatin immunoprecipitation assay other Abs were purchased from Santa Cruz Biotechnology (Santa Cruz,

by guest on October 1, 2021. Copyright 2012 Pageant Media Ltd. CA), and a final concentration of 0.5–2 mg/ml was used for Western The chromatin immunoprecipitation (ChIP) assay was performed using the 6 blotting purpose. The PPAR-g inhibitor T0070907 and the b-catenin/ ChIP-IT Express Enzymatic kit from Active Motif. We used 5–10 3 10 PPAR-g inhibitor FH535 are from EMD/Millipore USA. CD4 T cells per ChIP assay, as described (36). We stimulated cells with PMA (50 ng/ml) and ionomycin (1 mM) for 15 min and then cross-linked CD4+ T cell isolation and transfection with formaldehyde for 10 min. Cells were treated immediately with gly- cine to stop cross-linking, washed three times, and then frozen at 280˚C. Human CD4 T cells were purified according to the method described The thawed cells were subjected to the enzymatic shearing step, as per the previously (30). Briefly, PBMCs were prepared by Ficoll-Hypaque (His- kit instruction. Immunoprecipitation was performed with ChIP-grade topaque) gradient centrifugation (Sigma-Aldrich, St. Louis, MO). Un- + + mouse anti-SMRT (Thermo Fisher Scientific), anti-MEK1 (Santa Cruz touched CD4 T cells were isolated by CD4 T cell isolation kit-II Biotechnology), anti-phosphohistone H3 Abs (Millipore, Upstate, NY), (Miltenyi Biotec, Auburn, CA), according to the manufacturer’s instruc- + isotype control IgGs (Jackson ImmunoResearch Laboratories) (2 mg each), http://classic.jimmunol.org tion. CD4 T cells were transfected with 30–300 nM short interfering RNA and protein A/G magnetic beads overnight at 4˚C. The next day, the beads (siRNA) against SMRT (siGENOME SMARTpool; Dharmacon siRNA were washed three times with the provided buffer. DNA was eluted after Technologies) or nontargeting control siRNA (siGENOME Non-Targeting reversal of cross-linking and used as a template for subsequent PCR. We Pool 2; Dharmacon siRNA Technologies). Electroporation reagents were designed five sequential primer sets spanning the proximal c-Fos promoter purchased from Lonza (Allendale, NJ). Electroporation was performed from +3 to 2889 bp position and used them in the PCR. The primer sets according to the Amaxa protocol. After electroporation, cells were rested were numbered as follows: 1, 2889 to 2665; 2, 2680 to 2444; 3, 2463 overnight before setting them up for experiments. Short hairpin RNAs to 2210; 4, 2348 to 2141; and 5, 2155 to +3. (shRNA) for human SMRT and luciferase (control) were cloned into the Downloaded from retroviral vector Banshee-GFP, as described previously (30). The SMRT- Data analysis and luciferase-targeting shRNA sequences were as follows: 59-GTTCA- CAACACAGGCATGAACTC-39 and 59-AACGTACGCGGAATACTTC- Each experiment was done 3–10 times using blood lymphocytes from 39, respectively. Production of retroviruses, T cell infection, and sorting different donors. The statistical significance of difference between two was done as described (31, 32). samples was calculated by paired t test. Results are shown as mean 6 SEM. The p value ,0.05 was considered statistically significant. CD4+ T cell activation and proliferation

+ CD4 T cells transfected with siRNA were stimulated with plate-bound Results anti-CD3 and anti-CD28 for various times. Cell culture supernatant was harvested for ELISA. For proliferation, cells were labeled with CFSE Nuclear translocation of MEK1 (Invitrogen) at a concentration of 500 nM, according to manufacturer’s MEK1 and MEK2 are two threonine–tyrosine protein kinases that instruction, and then stimulated for 72 h. [3H]Thymidine uptake assay was performed, as described previously (31, 33). play a crucial role in T cell activation by activating ERK1/2. MEK1, but not MEK2, has a noncanonical nuclear localization motif, Flow cytometry Threonine-Proline-Threonine (TPT) (residues 386–388) (11). Flow cytometry was performed, as described previously (33). Cells were Phosphorylation of the threonine residues in this motif causes fixed with 2% paraformaldehyde for 10 min at room temperature. For in- its translocation to the nucleus. MEK1 shuttles between the The Journal of Immunology 3

nucleus and cytosol at a low level under basal conditions. The consistently detected by Abs obtained from three different ven- nuclear translocation increases upon stimulation (12). We exam- dors, Affinity Bioreagents/Thermo Fisher Scientific (Waltham, ined the expression of MEK1 and MEK2 in human CD4 T cells MA), Santa Cruz Biotechnology (Santa Cruz, CA), and Abcam using an anti-MEK1 and an anti-MEK1/2 Ab. The anti-MEK1 Ab (Cambridge, MA). Its mRNA expression was reduced by SMRT detected a major band with an approximate molecular mass of siRNA (Fig. 3D). For these reasons, we will present data on the 45 kDa (Fig. 1A). There was a faint band below the 45-kDa species. 170-kDa doublet. Next, we studied the effect of TCR stimulation The anti-MEK1/2 Ab did not reveal any additional band. Based on SMRT mRNA expression by real-time PCR. Anti-CD3/CD28 upon these findings, we conclude that the 45 kDa is the major MEK1 stimulation decreased the expression of mRNA for SMRT in the species in human CD4 T cells. The expression of MEK2 in this cell first 4 h and then increased its expression at 24 h (Fig. 2C). Im- type is negligible. Next, we examined the nuclear translocation of munofluorescence staining showed a predominant nuclear locali- MEK1 in resting and activated CD4 T cells by Western blotting zation of SMRT under basal conditions (Fig. 4). Anti-CD3/CD28 and immunofluorescence staining. Resting CD4 T cells from healthy stimulation for 1 h reduced the nuclear level and increased the human subjects had a low level of MEK1 in the nucleus (Fig. 1B– cytosolic level of SMRT, suggesting a nuclear export (Fig. 4). D). Stimulation of CD4 T cells with anti-CD3 and anti-CD28 Abs Thus, TCR stimulation resultedinanimmediatereductionin increased the overall expression of MEK1 and its nuclear trans- nuclear SMRT through two distinct mechanisms, inhibition at the location. Treatment of cells with leptomycin B, a known inhibitor transcriptional level and nuclear export at the protein level. The of nuclear export, further increased the nuclear level of MEK1. latter most likely occurs due to the action of MEK1 as its inhi- bition with U0126 resulted in increased nuclear SMRT, as shown MEK1 physically interacts with and regulates the by Western blotting (Fig. 2D) and immunofluorescence staining transcriptional corepressor SMRT (Fig. 4). We examined this issue further by overexpressing MEK1 Next, we examined the nuclear function of MEK1. MEK1 has using a bicistronic GFP-expressing retroviral vector. Overex- previously been shown to phosphorylate SMRT and prevents its pression of MEK1 increased its level in the nucleus (Fig. 2E). This interaction with the nuclear receptor–thyroid hormone receptor was associated with a reduction in the nuclear level of SMRT. and the PLZF (19). The foregoing studies were performed in overexpression models, so their physiological impli- MEK1 and SMRT bind to the promoter of c-Fos and cations remain unknown. We examined physical interaction of reciprocally regulate its expression MEK1 with SMRT using nuclear extract from anti-CD3/CD28 MEK1 along with ERK1/2 has been shown to bind the promoter of Ab-stimulated CD4 T cells. MEK1 coprecipitated with SMRT c-Fos and insulin in a ChIP assay (38). The exact role of MEK1 at from the nuclear extract (Fig. 2A). SMRT exists in multiple mo- the promoter site is unknown, but could include phosphorylation lecular mass forms due to alternative splicing (19, 21, 22). We of transcriptional activators and gene repressors, and the removal examined the presence of various molecular mass forms of SMRT of the latter from the nucleus. We examined the binding of MEK1 in CD4 T cells by Western blotting using a polyclonal Ab that has and SMRT to the c-Fos promoter because of its importance in IL-2 been used in other laboratories (28, 37). We observed a prominent production. We performed ChIP studies using five different primer high molecular mass form ∼170 kDa and a low molecular mass sets spanning the promoter site from +3 to 2889 bp position. We ∼

by guest on October 1, 2021. Copyright 2012 Pageant Media Ltd. form 120 kDa (Fig. 2B) in the nucleus and only the high mo- detected the presence of SMRT at the distal site (primer set 1) of lecular mass form in the cytosol. There were two faint bands just the c-Fos promoter in nonstimulated CD4 T cells and its removal below and above the 170-kDa molecular mass form. In most after stimulation of cells with PMA and ionomycin (Fig. 3A, 3B). experiments, we detected a doublet of the 170-kDa form and its MEK1 was present at a low level under basal conditions and was slightly higher molecular mass form. The 170-kDa doublet was recruited further upon stimulation (Fig. 3B). As a positive control, http://classic.jimmunol.org Downloaded from

FIGURE 1. (A) Expression of MEK1 and MEK2 in human CD4 T cells. Purified human CD4 T cells from a healthy subject were Western blotted using an anti-MEK1 and an anti-MEK1/2 Ab (n = 3). Nuclear expression of MEK1. (B) Western blotting of nuclear extract. Purified human CD4 T cells from a healthy subject were incubated in medium alone or with anti-CD3/CD28 Abs for 1 h and then were processed for nuclear and cytosolic extracts and Western blotted for MEK1. Nuclear and cytosolic extracts were reprobed for JunD and tubulin, respectively (n = 4). (C and D) Immunofluorescence staining. Purified CD4 T cells were incubated in medium alone or with anti-CD3/CD28 Abs 6 leptomycin B (LMB; 10 ng/ml) for 1 h and then stained for MEK1 (green). Nuclei were stained blue with DAPI. Original magnification 3100. All images were thresholded at the same upper and lower limits of the pixel count for comparison across the experimental groups. Integrated fluorescence intensity of nuclear (blue colored region) MEK1 from 25 cells per experiment and from four separate experiments was analyzed statistically (C). *p , 0.05, paired t test. 4 NUCLEAR MEK1 AND SMRT IN T CELL FUNCTION

FIGURE 2. (A) Coprecipitation of MEK1 with SMRT. We prepared nuclear extract from anti-CD3/CD28–stimulated CD4 T cells, immunoprecipitated with an anti-SMRT or control IgG Ab and then Western blotted for MEK1. Because of disparate molecular masses of MEK1 (45 kDa) and SMRT (170 kDa), the membrane was not reprobed for SMRT. Instead we show equal loading of IgH and IgL chains. The immunoprecipitate was separately Western blotted for SMRT (bottom plan)(n = 3). (B) Identification of isoforms of SMRT in CD4 T cells. Nuclear and cytosolic extracts from CD4 T cells were Western blotted for SMRT (n =6).(C) Early inhibition and late induction of SMRT by CD3/CD28 stimulation. CD4 T cells were stimulated with anti-CD3/CD28 Abs for the indicated periods of time and then used to isolate RNA and assay for mRNA for SMRT and b-actin by real-time PCR. Data are presented as normalized ratio of SMRT to b-actin (n =3,*p , 0.05, paired t test). (D) MEK inhibition increases the level of nuclear SMRT. CD4 T cells were stimulated with anti-CD3/CD28 Abs for 1 h in the presence or absence of the MEK inhibitor U0126 (10 mM) and then Western blotted for SMRT using the nuclear and cytosolic extracts. The membrane with the nuclear extract was reprobed for the nuclear repressor ZEB1 (n = 3). The pixel density is shown above the bands. (E) Effect of overexpression of MEK1 on SMRT expression. CD4 T cells were infected with a GFP-expressing bicistronic retrovirus containing cDNA for MEK1 (MEK1) or the empty retrovirus (RV). Cells were sorted for GFP and then Western blotted for SMRT and MEK1 (n = 3). A 50-kDa nonspecific band in the MEK1 blot shows equal protein loading.

we used an anti–phospho-H3 Ab. As anticipated, phospho-H3 was Its dissociation from the c-Fos promoter upon T cell stimulation absent under basal conditions, but became detectable at the distal suggests that SMRT regulates the expression of c-Fos. To test this

by guest on October 1, 2021. Copyright 2012 Pageant Media Ltd. promoter after stimulation. possibility, we developed two approaches to knockdown SMRT in http://classic.jimmunol.org Downloaded from

FIGURE 3. (A and B) ChIP studies with c-Fos promoter. CD4 T cells were stimulated (stim) with and without (basal) PMA (50 ng/ml)/ionomycin (1 mM). DNA was cross-linked to using formaldehyde and sheared by enzymatic processing using the ChIP-IT Express Enzymatic kit. ChIP grade rabbit anti-MEK1, anti-SMRT, and anti–phospho-H3 Abs, and control rabbit IgG were used to immunoprecipitate the bound chromatin. Following reversal of cross-linking and digestion of the proteins, the DNA was used for PCR. Five sets of primers (1–5, in which 5 is proximal to the start codon site) were designed from the proximal promoter sequence of c-Fos and used in PCR. An image of the electrophoresed amplicons is shown in (A). The input DNA was electrophoresed at 1:50 dilution. The density of the bands generated with the 1 primer set was analyzed and presented as enrichment over the input. Results from three separate experiments are shown in Fig. 5B. (C) Expression of SMRT mRNA following siRNA- and shRNA-mediated knockdown. Purified CD4 T cells were either transfected with the Smartpool siRNA or infected with the shRNA retrovirus for SMRT. After resting overnight, cells were stimulated with anti-CD3/CD28 Abs for 30 min. Purified RNA was used for real-time PCR (n = 3). (D and E) CD4 T cells were transfected with siSMRT or control siNT and stimulated as above for 0.5, 2, and 4 h. An aliquot of cells that were stimulated for 0.5 h was Western blotted for SMRT, c-Fos, MEK1, and b-actin (control) (D). The other aliquots were used to isolate RNA and real-time PCR for c-Fos (E) and c-Jun (F). n =3,*p , 0.05, paired t test. The Journal of Immunology 5

(siSMRT) and shRNA for SMRT, but not the nontargeting siNT and shRNA, reduced the expression of SMRT by .65% (Fig. 3C, 3D). The siSMRT-mediated knockdown was associated with an increase in c-Fos protein (Fig. 3D) and mRNA (Fig. 3E) within 30 min. SMRT knockdown did not affect the expression of mRNA for c-Jun (Fig. 3F). SMRT knockdown results in an early increase, followed by a late decline in cytokine production Next, we studied the effect of SMRT knockdown on T cell func- tion. We measured cytokine production by ELISA following stim- ulation of CD4 T cells with anti-CD3/CD28 stimulation. Because we observed a biphasic (an early-phase inhibition and a late- FIGURE 4. CD3/CD28 stimulation increases nuclear export of SMRT. phase upregulation) effect of TCR stimulation on SMRT expres- Isolated CD4 T cells were incubated with medium and anti-CD3/CD28 (2 sion (Fig. 2C), in preliminary studies we examined T cell cytokine mg/ml each) Abs 6 U0126 (10 mM) for 1 h and then stained for SMRT. production at multiple time points following anti-CD3/CD28 3 Nuclei were counterstained with DAPI. Original magnification 100. All stimulation. We observed an initial rise and a late fall in IL-2 images were thresholded at the same upper and lower limits of the pixel production in SMRT knockdown CD4 T cells (Fig. 5A). In sub- count for comparison across the experimental groups. Representative cells sequent studies, we studied cytokine production at an early (8 h) from one of four separate experiments are shown. and a late (48 h) time point. We confirmed the biphasic profile of IL-2 production in a larger study group. We also observed a sim- the cells. We used a smartpool siRNA for SMRT and a control ilar early rise and a late fall in the production of IL-4, IL-10, and nontargeting siRNA pool (siNT) and transfected cells with these IFN-g in SMRT knockdown T cells as compared with the control siRNA using the Amaxa protocol. We also constructed bicistronic cells (Fig. 5B, 5C). SMRT knockdown was associated with an GFP-expressing retroviral vectors encoding shRNA against lu- inhibition of T cell proliferation, as demonstrated by reduced ciferase (negative control) and SMRT. Short interfering SMRT CFSE dilution (Fig. 5D, 5E). The effect on proliferation most by guest on October 1, 2021. Copyright 2012 Pageant Media Ltd.

FIGURE 5. (A–C) Effect of SMRT knockdown on cytokine production. Purified CD4 T cells were trans- fected with siSMRT and siNT and then cultured on anti- CD3/CD28 Ab-coated plates for the indicated periods of time. Supernatant collected at different time points was assayed for cytokines by ELISA. IL-2 production from three donors is shown in (A). (B and C) Shown are the production profiles of IL-2, IL-4, IL-10, and IFN-g http://classic.jimmunol.org at 8 and 48 h. The numbers at the end of the line graphs represent p values for difference (n = 6–8 donors per group). (D and E) SMRT knockdown inhibits T cell proliferation. CD4 T cells, transfected as per (A), were labeled with CFSE, cultured on an anti-CD3/CD28– coated plate, and then analyzed by flow cytometry (n = 3). (F and G) IL-2 prevents late decline in cytokine Downloaded from production and restores proliferation. CD4 T cells, transfected as per (A), were cultured on an anti-CD3/ CD28–coated plate with and without IL-2 (10 ng/ml) for 48 h. (F) Supernatant was assayed for cytokines by ELISA (n = 3). (G)[3H]Thymidine uptake was mea- sured at 48 h. *p , 0.04, paired t test, n =3. 6 NUCLEAR MEK1 AND SMRT IN T CELL FUNCTION

likely results from the inhibition of IL-2 production that com- duction changed to 153 and 56% of the baseline, respectively, in mences by ∼24 h. The late decline in IL-4 and IFN-g is secondary control T cells (Fig. 7F). The result suggests that b-catenin and to the dramatic reduction in the level of IL-2. Supplementation PPAR-g inhibit and augment, respectively, IL-2 production by with IL-2 reversed the inhibitory effect of SMRT knockdown on human CD4 T cells. SMRT knockdown T cells without the IL-4 and IFN-g production (Fig. 5F). The supplementation also inhibitors had an 89% inhibition of IL-2 production. This inhibi- reversed the inhibition of T cell proliferation with siSMRT (Fig. tion was reduced to 84 and 34% of the baseline in the presence 5G). To understand the phenotype of SMRT knockdown T cells, of T0070907 and FH535, respectively. Thus, FH535 increased we studied activation of three signaling pathways that are in- IL-2 production in SMRT knockdown cells by nearly 7-fold. The volved in T cell activation, ERK1/2, p38, and p65 NF-kB. We also results suggest that SMRT antagonizes b-catenin signaling, which examined the expression of JunD, a negative regulator of T cell is lost in SMRT knockdown T cells. As a consequence, b-catenin– activation (37). There was no difference in ERK1/2 activation mediated inhibition dominates in the late phase of T cell acti- after TCR stimulation of SMRT knockdown T cells (Fig. 6A). The vation (Fig. 8). phosphorylation of p38 was mildly elevated. There was no dif- ference in the nuclear expression level of p65 NF-kB and JunD Discussion (Fig. 6B). We recognize that the p65 NF-kB is just one of many In this work, we have shown tht MEK1 translocates to the nucleus members of the NF-kB family, and our result does not rule out the following TCR stimulation and interacts with the nuclear core- involvement of other NF-kB members in SMRT knockdown cells. pressor SMRT in CD4 T cells. MEK1 is bound to the c-Fos pro- IL-10 establishes a negative feedback loop to inhibit T cell moter at a low level under basal conditions and is further recruited activation following T cell activation. In contrast, SMRT is bound to the c-Fos Because SMRT knockdown CD4 T cells produced high quantities promoter under basal condition and is removed following stimu- of IL-10 at the early time point, we asked whether this early rise in lation. These reciprocal changes in MEK1 and SMRT binding to IL-10 mediated the late inhibition of cytokine production. To this the promoter site result in increased c-Fos expression and cytokine goal, we infected CD4 T cells with the control shRNA- or SMRT production in the early stage of T cell activation. The role of SMRT shRNA-encoding bicistronic retrovirus. Infected cells were sorted in T cell activation is far more complex than would be expected for expression of GFP, rested overnight in the medium alone, and from a simple corepressor function. In the early stage of T cell then cultured on an anti-CD3/CD28–coated plate with a neutral- activation, SMRT functions as a corepressor and its nuclear re- izing anti–IL-10 or an isotype control Ab. The anti–IL-10 Ab duction plays a permissive role (Fig. 8). In the late phase, SMRT partially reversed the inhibitory effect of SMRT knockdown on functions as an activator of T cells. SMRT antagonizes two in- the late-phase IL-2 production (Fig. 7A). We also checked FOXP3 hibitory signals that arise in the late phase, IL-10 and b-catenin. expression in SMRT knockdown cells. There was no difference in By antagonizing these two signals, SMRT promotes sustained IL- FoxP3 expression following SMRT knockdown (Fig. 7B). Next, 2 production and T cell proliferation. we examined whether IL-10 exerted its inhibitory effect on CD4 The translocation of MEK1 to the nucleus has previously been T cells through the corepressor SMRT. We cultured CD4 T cells demonstrated in other cells (10). MEK1 is translocated to the with anti-CD4/CD28 Abs for 48 h in the presence and absence of nucleus in an importin 7-dependent manner. Two distinct mech- by guest on October 1, 2021. Copyright 2012 Pageant Media Ltd. IL-10 and then examined the expression of SMRT by immuno- anisms have been identified for nuclear translocation (11). fluorescence staining and Western blotting. Anti-CD3/CD28 Translocation following a mitogenic stimulation requires MEK1 stimulation significantly upregulated SMRT expression (Fig. 7C– activity. MEK1 also translocates to the nucleus at a low E). This expression was further amplified by the addition of IL-10. level under basal conditions. This is largely independent of MEK1 Cells cultured with IL-10 alone showed a negligible effect. kinase activity. It requires phosphorylation of a unique TPT (residues 386–388) motif for translocation. ERK2 has been shown b -catenin signaling contributes to the late-phase inhibition of to phosphorylate the TPT motif. We have shown that MEK1 di- T cell IL-2 production rectly binds to the c-Fos promoter in T cells. This is in agreement SMRT antagonizes gene transcription induced by b-catenin (39, with a previous report in which human pancreatic b cells were http://classic.jimmunol.org 40) and PPAR-g (41). The latter molecules are known to regulate studied (38). We have also shown that the recruitment of MEK1 to T cell function (42, 43). We applied two pharmacologic inhibitors the promoter increases upon TCR stimulation. MEK1 has been to explore their involvement in inhibition of IL-2 production in shown to phosphorylate SMRT, which could cause its dissociation SMRT knockdown T cells. T0070907 is a specific inhibitor of from HDAC3 and removal of this repressor complex from the PPAR-g (44). FH535 inhibits both b-catenin and PPAR-g (45). promoter. SMRT knockdown and control CD4 T cells were stimulated with The N-terminal nuclear export signal allows MEK1 to exit the Downloaded from anti-CD3/CD28 Abs for 48 h in the presence or absence of the nucleus. MEK1 has previously been shown to cause nuclear export inhibitors. In the presence of FH535 and T0070907, IL-2 pro- of SMRT (19) and PPAR-g (14). We have demonstrated a direct

FIGURE 6. Effect of SMRT knockdown on cytosolic and nuclear signaling. (A) CD4 T cells were transfected with siSMRT and siNT Smartpool siRNA; stimulated with and without (medium) anti-CD3/CD28 Abs for 10 min; and then Western blotted for p-p38 and pERK1/2. The p-p38 membrane was reprobed for p38, and the pERK1/2 membrane was reprobed for b-actin. (n = 4). (B) CD4 T cells were transfected with siSMRT and siNT siRNA and then stimulated with anti-CD3/CD28 Abs for 8 and 48 h. Nuclear extract was isolated and Western blotted for p65 NF-kB and JunD. The mem- brane was reprobed for HDAC3 (n = 4). The Journal of Immunology 7

FIGURE 7. (A) Anti–IL-10 Ab partially prevents late decline in cytokine production. CD4 T cells were transfected with siSMRT and siNT siRNA and then stimulated with anti-CD3/CD28 Abs for 48 h in the presence of a neutralizing anti–IL-10 Ab or IgG control (10 mg/ml each). IL-2 in the supernatant was measured by ELISA. n =3,p = 0.04, paired t test. (B) SMRT knockdown does not affect FOXP3 expression. CD4 T cells were transfected with siSMRT and siNT siRNA and then stimulated with anti-CD3/CD28 Abs for 48 h. Cells were then fixed, stained with an anti- FoxP3 Ab, and analyzed by flow cytometry (n = 3). (C–E) Effect of IL-10 on SMRT expression. CD4 T cells were cultured with anti-CD3/CD28 Abs 6 IL-10. The expression of SMRT was examined by immunofluorescence staining (C) and Western blotting (E, n = 3) at 48 h. Immunofluorescence staining (C) and statistical analyses of the morphometric data (D) are shown. Original magnification 3100. *p , 0.01, paired t test, n = 4. Note that all images were thresholded at the same upper and lower limits of the pixel count for cross-comparison. (F) Isolated CD4 T cells were transfected with siNT or siSMRT, and then cultured in anti-CD3/CD28 Ab-coated plates for48hinthepresenceofthePPAR-g inhibitor T0070907 (T70907), b-catenin/PPAR-g inhibitor FH535, or the vehicle (DMSO). IL-2 in the culture supernatant was assayed by ELISA. One of three representative experiments done in triplicates is shown. Results were normalized to IL-2 production in DMSO- and siNT-treated cells. *p , 0.05 compared with siNT and DMSO-treated cells (paired t test). by guest on October 1, 2021. Copyright 2012 Pageant Media Ltd.

interaction of endogenous MEK1 with SMRT in CD4 T cells in SMRT has been shown to antagonize the transcriptional activity this study. Inhibition of MEK1 increases the nuclear content of of b-catenin (39, 40). b-catenin is important for T cell develop- SMRT, suggesting that MEK1 regulates gene repression by con- ment and function. b-catenin positively regulates thymopoiesis trolling the nuclear level of the corepressor SMRT. MEK1 is one (50) and Th2 cell differentiation in the mouse model (51, 52). http://classic.jimmunol.org of a number of molecules that regulate the nuclear level of SMRT. However, it inhibits differentiation of CD4 Th17 cells (53) and The others include MEK kinase 1 (19), 14-3-3, and PIN1 (46). CD8 effector cells (54). b-catenin induces T cell anergy in mature T cell stimulation results in activation and engagement of many T cells (55). In human T cells, b-catenin signaling inhibits IL-2 of these molecules (47–49). The individual contribution of these and IFN-g production and blocks upregulation of the IL-2R (56). molecules in regulating nuclear export of SMRT following T cell As a result, T cell proliferation is blocked. b-catenin impairs activation will require further studies. maturation of CD45RA naive T cells into CD45RO memory cells.

Downloaded from Our results are in agreement with the foregoing studies. b-catenin modestly, but significantly inhibited IL-2 production in human CD4 T cells (Fig. 7F). This inhibition was remarkably pronounced in SMRT knockdown T cells, as uncovered in our inhibitor studies with FH535. The results suggest that SMRT antagonizes b-catenin signaling in T cells. Although FH535 inhibits both b-catenin and PPAR-g, we think PPAR-g is unlikely to play any role in our model. The PPAR-g–specific inhibitor T0070907 actually inhib- ited IL-2 production. IL-2 production in SMRT knockdown T cells was unchanged in the presence of T0070907. Previously, SMRT was studied in Jurkat T cells (57). Through the use of reporter genes, it was shown that CD3/CD28 ligation FIGURE 8. A diagrammatic presentation of the effect of nuclear MEK increased the association of SMRT with nuclear RAR, RXR, and on SMRT and its early- and late-phase consequences on cytokine pro- thyroid hormone receptor. This association resulted in silencing duction and proliferation. the target gene expression. The increase in SMRT–nuclear re- 8 NUCLEAR MEK1 AND SMRT IN T CELL FUNCTION

ceptor association was protein kinase C-u dependent. RAR, RXR, 10. Wunderlich, W., I. Fialka, D. Teis, A. Alpi, A. Pfeifer, R. G. Parton, F. Lottspeich, and L. A. Huber. 2001. A novel 14-kilodalton protein interacts and LXR are negative regulators of T cells, although their effect with the mitogen-activated protein kinase scaffold mp1 on a late endosomal/ on a specific Th cell subtype may vary. A conducive T cell acti- lysosomal compartment. J. Cell Biol. 152: 765–776. vation signaling requires silencing these receptors through SMRT. 11. Chuderland, D., A. Konson, and R. Seger. 2008. Identification and character- ization of a general nuclear translocation signal in signaling proteins. Mol. Cell The late effect of SMRT knockdown in our study is in agreement 31: 850–861. with the foregoing study. We speculate that in the absence of 12. Jaaro, H., H. Rubinfeld, T. Hanoch, and R. Seger. 1997. Nuclear translocation of SMRT, gene transcription through RAR, RXR, and LXR is un- mitogen-activated protein kinase kinase (MEK1) in response to mitogenic stimulation. Proc. Natl. Acad. Sci. USA 94: 3742–3747. inhibited, which also contributes to the inhibition of cytokine 13. Wortzel, I., and R. Seger. 2011. The ERK cascade: distinct functions within production and T cell proliferation. various subcellular organelles. Genes Cancer 2: 195-209. 14. Burgermeister, E., D. Chuderland, T. Hanoch, M. Meyer, M. Liscovitch, and In addition to NFs, SMRT targets certain transcription factors R. Seger. 2007. Interaction with MEK causes nuclear export and downregulation and regulates the expression of their target genes. In macrophages, of peroxisome proliferator-activated receptor gamma. Mol. Cell. Biol. 27: 803– SMRT regulates the expression of 87 LPS-inducible and LXR- 817. 15. Skarpen, E., L. I. Flinder, C. M. Rosseland, S. Orstavik, L. Wierod, M. P. regulated genes (25). Along with NCoR, it coregulates another Oksvold, B. S. Skalhegg, and H. S. Huitfeldt. 2008. MEK1 and MEK2 regulate 80 genes. We have observed SMRT recruitment to the c-Fos distinct functions by sorting ERK2 to different intracellular compartments. promoter. NF-kB has been reported to be a target of SMRT FASEB J. 22: 466-476. 16. Maki, K., and K. Ikuta. 2008. MEK1/2 induces STAT5-mediated germline (26). We have not observed any difference in the nuclear level of transcription of the TCRgamma locus in response to IL-7R signaling. J. p65 NF-kB in SMRT knockdown T cells. The results suggest that Immunol. 181: 494–502. 17. Jo, C., S. J. Cho, and S. A. Jo. 2011. Mitogen-activated protein kinase kinase 1 the nature of the gene targeted by SMRT varies depending upon (MEK1) stabilizes MyoD through direct phosphorylation at tyrosine 156 during the type of the cell. Our studies have revealed a biphasic effect of myogenic differentiation. J. Biol. Chem. 286: 18903–18913. SMRT knockdown in human T cells. We believe that the early- 18. Chen, J. D., and R. M. Evans. 1995. A transcriptional co-repressor that interacts with nuclear hormone receptors. Nature 377: 454–457. phase rise in cytokine production results from the absence of 19. Hong, S. H., and M. L. Privalsky. 2000. The SMRT corepressor is regulated by corepressor activity of SMRT for c-Fos and possibly other SMRT- a MEK-1 kinase pathway: inhibition of corepressor function is associated with regulated genes. SMRT interacts with many histone deacetylases SMRT phosphorylation and nuclear export. Mol. Cell. Biol. 20: 6612–6625. 20. Perissi, V., K. Jepsen, C. K. Glass, and M. G. Rosenfeld. 2010. Deconstructing belonging to the class I and class II HDACs (58). Absence of repression: evolving models of co-repressor action. Nat. Rev. Genet. 11: 109– SMRT could lead to the dissociation of HDACs from the gene 123. 21. Goodson, M. L., B. A. Jonas, and M. L. Privalsky. 2005. Alternative mRNA promoters and initiation of gene transcription. The disruption of splicing of SMRT creates functional diversity by generating corepressor iso- HDAC11 has been shown to induce IL-10 production (59). forms with different affinities for different nuclear receptors. J. Biol. Chem. 280: SMRT is a broad-spectrum corepressor. It represses not only 7493–7503. 22. Chen, J. D., K. Umesono, and R. M. Evans. 1996. SMRT isoforms mediate inhibitory, but also stimulatory genes (25). Based upon this tenet, repression and anti-repression of nuclear receptor heterodimers. Proc. Natl. we have anticipated a complex function of SMRT in T cells. Acad. Sci. USA 93: 7567–7571. Accordingly, we have observed early-phase repressive and late- 23. Perissi, V., C. Scafoglio, J. Zhang, K. A. Ohgi, D. W. Rose, C. K. Glass, and M. G. Rosenfeld. 2008. TBL1 and TBLR1 phosphorylation on regulated gene phase stimulatory effects of SMRT in CD4 T cells. Our results promoters overcomes dual CtBP and NCoR/SMRT transcriptional repression suggest that a complete deficiency would be detrimental to CD4 checkpoints. Mol. Cell 29: 755–766. 24. Hu, X., S. Li, J. Wu, C. Xia, and D. S. Lala. 2003. Liver X receptors interact with T cells. However, a low-level upregulation and downregulation of to regulate gene expression. Mol. Endocrinol. 17: 1019–1026.

by guest on October 1, 2021. Copyright 2012 Pageant Media Ltd. SMRT would most likely affect a wide spectrum of T cell func- 25. Ghisletti, S., W. Huang, K. Jepsen, C. Benner, G. Hardiman, M. G. Rosenfeld, tions. and C. K. Glass. 2009. Cooperative NCoR/SMRT interactions establish a core- pressor-based strategy for integration of inflammatory and anti-inflammatory signaling pathways. Genes Dev. 23: 681–693. 26. Lee, S. K., J. H. Kim, Y. C. Lee, J. Cheong, and J. W. Lee. 2000. Silencing Disclosures mediator of retinoic acid and thyroid hormone receptors, as a novel transcrip- The authors have no financial conflicts of interest. tional corepressor molecule of activating protein-1, nuclear factor-kappaB, and serum response factor. J. Biol. Chem. 275: 12470–12474. 27. Jepsen, K., A. S. Gleiberman, C. Shi, D. I. Simon, and M. G. Rosenfeld. 2008. Cooperative regulation in development by SMRT and FOXP1. Genes Dev. 22: References 740–745. 1. Roux, P. P., and J. Blenis. 2004. ERK and p38 MAPK-activated protein kinases: 28. Ogawa, S., J. Lozach, K. Jepsen, D. Sawka-Verhelle, V. Perissi, R. Sasik,

http://classic.jimmunol.org a family of protein kinases with diverse biological functions. Microbiol. Mol. D. W. Rose, R. S. Johnson, M. G. Rosenfeld, and C. K. Glass. 2004. A nuclear Biol. Rev. 68: 320–344. receptor corepressor transcriptional checkpoint controlling activator protein 1- 2. Giroux, S., M. Tremblay, D. Bernard, J. F. Cardin-Girard, S. Aubry, L. Larouche, dependent gene networks required for macrophage activation. Proc. Natl. Acad. S. Rousseau, J. Huot, J. Landry, L. Jeannotte, and J. Charron. 1999. Embryonic Sci. USA 101: 14461–14466. death of Mek1-deficient mice reveals a role for this kinase in angiogenesis in the 29. Jepsen, K., D. Solum, T. Zhou, R. J. McEvilly, H. J. Kim, C. K. Glass, labyrinthine region of the placenta. Curr. Biol. 9: 369-372. O. Hermanson, and M. G. Rosenfeld. 2007. SMRT-mediated repression of an 3. Scholl, F. A., P. A. Dumesic, D. I. Barragan, K. Harada, V. Bissonauth, H3K27 demethylase in progression from neural stem cell to neuron. Nature 450: J. Charron, and P. A. Khavari. 2007. Mek1/2 MAPK kinases are essential for 415–419. mammalian development, homeostasis, and Raf-induced hyperplasia. Dev. Cell 30. Herna´ndez-Hoyos, G., M. K. Anderson, C. Wang, E. V. Rothenberg, and Downloaded from 12: 615–629. J. Alberola-Ila. 2003. GATA-3 expression is controlled by TCR signals and 4. Page`s, G., S. Gue´rin, D. Grall, F. Bonino, A. Smith, F. Anjuere, P. Auberger, and regulates CD4/CD8 differentiation. Immunity 19: 83–94. J. Pouysse´gur. 1999. Defective thymocyte maturation in p44 MAP kinase (Erk 1) 31. Gorska, M. M., S. J. Stafford, O. Cen, S. Sur, and R. Alam. 2004. Unc119, knockout mice. Science 286: 1374–1377. a novel activator of Lck/Fyn, is essential for T cell activation. J. Exp. Med. 199: 5. Fischer, A. M., C. D. Katayama, G. Page`s, J. Pouysse´gur, and S. M. Hedrick. 369–379. 2005. The role of erk1 and erk2 in multiple stages of T cell development. Im- 32. Naviaux, R. K., E. Costanzi, M. Haas, and I. M. Verma. 1996. The pCL vector munity 23: 431–443. system: rapid production of helper-free, high-titer, recombinant retroviruses. J. 6. D’Souza, W. N., C. F. Chang, A. M. Fischer, M. Li, and S. M. Hedrick. 2008. Virol. 70: 5701–5705. The Erk2 MAPK regulates CD8 T cell proliferation and survival. J. Immunol. 33. Liang, Q., L. Guo, S. Gogate, Z. Karim, A. Hanifi, D. Y. Leung, M. M. Gorska, 181: 7617–7629. and R. Alam. 2010. IL-2 and IL-4 stimulate MEK1 expression and contribute to 7. Jorritsma, P. J., J. L. Brogdon, and K. Bottomly. 2003. Role of TCR-induced T cell resistance against suppression by TGF-beta and IL-10 in asthma. J. extracellular signal-regulated kinase activation in the regulation of early IL-4 Immunol. 185: 5704–5713. expression in naive CD4+ T cells. J. Immunol. 170: 2427–2434. 34. Gorska, M. M., Q. Liang, Z. Karim, and R. Alam. 2009. Uncoordinated 119 8. Goplen, N., Z. Karim, L. Guo, Y. Zhuang, H. Huang, M. M. Gorska, E. Gelfand, protein controls trafficking of Lck via the Rab11 endosome and is critical for G. Pages, J. Pouyssegur, and R. Alam. 2012. ERK1 is important for Th2 dif- immunological synapse formation. J. Immunol. 183: 1675–1684. ferentiation and development of experimental asthma. FASEB J. 26: 1934-1945. 35. Liu, W., K. Tundwal, Q. Liang, N. Goplen, S. Rozario, N. Quayum, M. Gorska, 9. Schipper, H. M., A. Agarwal-Mawal, and H. K. Paudel. 1999. The topography S. Wenzel, S. Balzar, and R. Alam. 2010. Establishment of extracellular signal- and subcellular distribution of mitogen-activated protein kinase kinase1 (MEK1) regulated kinase 1/2 bistability and sustained activation through Sprouty 2 and its in adult rat brain and differentiating PC12 cells. Neuroscience 93: 585–595. relevance for epithelial function. Mol. Cell. Biol. 30: 1783–1799. The Journal of Immunology 9

36. Yu, M., M. Wan, J. Zhang, J. Wu, R. Khatri, and T. Chi. 2008. Nucleoprotein Itch-induced ubiquitination with MEKK1-JNK signaling in Th2 tolerance and structure of the CD4 locus: implications for the mechanisms underlying CD4 airway inflammation. J. Clin. Invest. 116: 1117–1126. regulation during T cell development. Proc. Natl. Acad. Sci. USA 105: 3873– 48. Di Bartolo, V., B. Montagne, M. Salek, B. Jungwirth, F. Carrette, J. Fourtane, 3878. N. Sol-Foulon, F. Michel, O. Schwartz, W. D. Lehmann, and O. Acuto. 2007. A 37. Meixner, A., F. Karreth, L. Kenner, and E. F. Wagner. 2004. JunD regulates novel pathway down-modulating T cell activation involves HPK-1-dependent lymphocyte proliferation and T helper cell cytokine expression. EMBO J. 23: recruitment of 14-3-3 proteins on SLP-76. J. Exp. Med. 204: 681–691. 1325–1335. 49. Esnault, S., Z. J. Shen, E. Whitesel, and J. S. Malter. 2006. The peptidyl-prolyl 38. Lawrence, M. C., K. McGlynn, C. Shao, L. Duan, B. Naziruddin, M. F. Levy, isomerase Pin1 regulates granulocyte-macrophage colony-stimulating factor and M. H. Cobb. 2008. Chromatin-bound mitogen-activated protein kinases mRNA stability in T lymphocytes. J. Immunol. 177: 6999–7006. transmit dynamic signals in transcription complexes in beta-cells. Proc. Natl. 50. Ioannidis, V., F. Beermann, H. Clevers, and W. Held. 2001. The beta-catenin– Acad. Sci. USA 105: 13315–13320. TCF-1 pathway ensures CD4(+)CD8(+) thymocyte survival. Nat. Immunol. 2: 39. Song, L. N., and E. P. Gelmann. 2008. Silencing mediator for retinoid and 691–697. thyroid hormone receptor and nuclear receptor corepressor attenuate transcrip- 51. Yu, Q., A. Sharma, S. Y. Oh, H. G. Moon, M. Z. Hossain, T. M. Salay, tional activation by the beta-catenin-TCF4 complex. J. Biol. Chem. 283: 25988- K. E. Leeds, H. Du, B. Wu, M. L. Waterman, et al. 2009. T cell factor 1 initiates 25999. the T helper type 2 fate by inducing the transcription factor GATA-3 and 40. Penda´s-Franco, N., J. M. Garcı´a, C. Pen˜a, N. Valle, H. G. Pa´lmer, M. Heina¨- repressing interferon-gamma. Nat. Immunol. 10: 992–999. niemi, C. Carlberg, B. Jime´nez, F. Bonilla, A. Mun˜oz, and J. M. Gonza´lez- 52. Notani, D., K. P. Gottimukkala, R. S. Jayani, A. S. Limaye, M. V. Damle, Sancho. 2008. DICKKOPF-4 is induced by TCF/beta-catenin and upregulated S. Mehta, P. K. Purbey, J. Joseph, and S. Galande. 2010. Global regulator SATB1 in human colon cancer, promotes tumour cell invasion and angiogenesis and is recruits beta-catenin and regulates T(H)2 differentiation in Wnt-dependent repressed by 1alpha,25-dihydroxyvitamin D3. Oncogene 27: 4467-4477. manner. PLoS Biol. 8: e1000296. 41. Yu, C., K. Markan, K. A. Temple, D. Deplewski, M. J. Brady, and R. N. Cohen. 53. Yu, Q., A. Sharma, A. Ghosh, and J. M. Sen. 2011. T cell factor-1 negatively 2005. The nuclear receptor corepressors NCoR and SMRT decrease peroxisome regulates expression of IL-17 family of cytokines and protects mice from ex- proliferator-activated receptor gamma transcriptional activity and repress 3T3- perimental autoimmune encephalomyelitis. J. Immunol. 186: 3946–3952. L1 adipogenesis. J. Biol. Chem. 280: 13600-13605. 54. Gattinoni, L., X. S. Zhong, D. C. Palmer, Y. Ji, C. S. Hinrichs, Z. Yu, 42. Xue, H. H., and D. M. Zhao. 2012. Regulation of mature T cell responses by the C. Wrzesinski, A. Boni, L. Cassard, L. M. Garvin, et al. 2009. Wnt signaling . Ann. N. Y. Acad. Sci. 1247: 16–33. arrests effector T cell differentiation and generates CD8+ memory stem cells. 43. Glass, C. K., and K. Saijo. 2010. Nuclear receptor transrepression pathways that Nat. Med. 15: 808–813. regulate inflammation in macrophages and T cells. Nat. Rev. Immunol. 10: 365– 55. Ding, Y., S. Shen, A. C. Lino, M. A. Curotto de Lafaille, and J. J. Lafaille. 2008. 376. Beta-catenin stabilization extends regulatory T cell survival and induces anergy 44. Lee, G., F. Elwood, J. McNally, J. Weiszmann, M. Lindstrom, K. Amaral, in nonregulatory T cells. Nat. Med. 14: 162–169. M. Nakamura, S. Miao, P. Cao, R. M. Learned, et al. 2002. T0070907, a selective 56. Muralidharan, S., P. J. Hanley, E. Liu, R. Chakraborty, C. Bollard, E. Shpall, ligand for peroxisome proliferator-activated receptor gamma, functions as an C. Rooney, B. Savoldo, J. Rodgers, and G. Dotti. 2011. Activation of Wnt sig- antagonist of biochemical and cellular activities. J. Biol. Chem. 277: 19649- naling arrests effector differentiation in human peripheral and cord blood- 19657. derived T lymphocytes. J. Immunol. 187: 5221–5232. 45. Handeli, S., and J. A. Simon. 2008. A small-molecule inhibitor of Tcf/beta- 57. Ishaq, M., G. DeGray, and V. Natarajan. 2003. Protein kinase C theta modulates catenin signaling down-regulates PPARgamma and PPARdelta activities. Mol. nuclear receptor-corepressor interaction during T cell activation. J. Biol. Chem. Cancer Ther. 7: 521-529. 278: 39296–39302. 46. Rajendran, P., B. Delage, W. M. Dashwood, T. W. Yu, B. Wuth, D. E. Williams, 58. Kao, H. Y., M. Downes, P. Ordentlich, and R. M. Evans. 2000. Isolation of E. Ho, and R. H. Dashwood. 2011. Histone deacetylase turnover and recovery in a novel histone deacetylase reveals that class I and class II deacetylases promote sulforaphane-treated colon cancer cells: competing actions of 14-3-3 and Pin1 in SMRT-mediated repression. Genes Dev. 14: 55–66. HDAC3/SMRT corepressor complex dissociation/reassembly. Mol. Cancer 10: 59. Villagra, A., F. Cheng, H. W. Wang, I. Suarez, M. Glozak, M. Maurin, 68. D. Nguyen, K. L. Wright, P. W. Atadja, K. Bhalla, et al. 2009. The histone 47. Venuprasad, K., C. Elly, M. Gao, S. Salek-Ardakani, Y. Harada, J. L. Luo, deacetylase HDAC11 regulates the expression of interleukin 10 and immune C. Yang, M. Croft, K. Inoue, M. Karin, and Y. C. Liu. 2006. Convergence of tolerance. Nat. Immunol. 10: 92–100. by guest on October 1, 2021. Copyright 2012 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from