Cyclin D2 Gene in Response to IL-2 Stat5 and Sp1 Regulate

Stat5 and Sp1 Regulate Transcription of the Cyclin D2 Gene in Response to IL-2 Anthony Martino, James H. Holmes IV, James D. Lord, James J. Moon and Brad H. Nelson This information is current as of September 27, 2021. J Immunol 2001; 166:1723-1729; ; doi: 10.4049/jimmunol.166.3.1723 http://www.jimmunol.org/content/166/3/1723 Downloaded from

References This article cites 51 articles, 28 of which you can access for free at: http://www.jimmunol.org/content/166/3/1723.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 27, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2001 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Stat5 and Sp1 Regulate Transcription of the Cyclin D2 Gene in Response to IL-21

Anthony Martino,2 James H. Holmes IV, James D. Lord, James J. Moon, and Brad H. Nelson

The IL-2R promotes rapid expansion of activated T cells through signals mediated by the adaptor protein Shc and the transcription factor Stat5. The mechanisms that engage the cell cycle are not well deﬁned. We report on the transcriptional regulation of the cell cycle gene cyclin D2 by the IL-2R. IL-2-responsive induction of a luciferase reporter gene containing 1624 bp of the cyclin D2 promoter/ enhancer was studied in the murine CD8؉ T cell line CTLL2. Reporter gene deletional analysis and EMSAs indicate an IL-2-regulated enhancer element ﬂanks nucleotide ؊1204 and binds a complex of at least three proteins. The enhancer element is bound constitutively by Sp1 and an unknown factor(s) and inducibly by Stat5 in response to IL-2. The Stat5 binding site was essential for IL-2-mediated reporter gene activity, and maximum induction required the adjacent Sp1 binding site. Receptor mutagenesis studies in the pro-B cell line BA/FG (a derivative of the BA/F3 cell line) demonstrated a correlation between Stat5 activity and cyclin D2 mRNA levels when the Downloaded from Stat5 signal was isolated, disrupted, and then rescued. Further, a dominant-negative form of Stat5 lacking the trans-activation domain inhibited induction of cyclin D2 mRNA. We propose that the IL-2R regulates the cyclin D2 gene in part through formation of an enhancer complex containing Stat5 and Sp1. The Journal of Immunology, 2001, 166: 1723–1729.

he cytokine IL-2 contributes to immune responses in part and activates two downstream pathways. The Grb2/Sos complex is

by promoting rapid proliferation of activated T cells. Sev- recruited to Shc (9–11), and Sos activates the Ras-mitogen-acti- http://www.jimmunol.org/ T eral components of the IL-2R complex transduce the pro- vated protein kinase pathway up-regulating genes such as c-fos and liferative signal from the cell surface (1, 2), but the mechanisms c-jun (12, 13). Additionally, the phosphatidylinositol-3 kinase that control expression of cell cycle-regulatory genes remain (PI3K) signaling pathway is activated through recruitment of the poorly defined. In this report we investigate the molecular mech- adaptor protein GAB2 (14–17). While no genes have been linked anisms by which the IL-2R regulates transcription of the cell cycle directly to the PI3K pathway, regulation of cyclin D3 and p27kip1 gene cyclin D2. expression, pRB phosphorylation, and E2F activity is attributed to The high affinity IL-2R complex includes three proteins (3). PI3K activity in response to IL-2 in T cells (18, 19). IL-2R␣ regulates ligand receptor affinity, and IL-2R␤ and the A second proliferative signal activated by the IL-2R involves ␥ ␥ 3 by guest on September 27, 2021 common -chain ( c) initiate intracellular signaling. The tyrosine the transcription factor Stat5 (20, 21). Stat5 is recruited to phos- kinases Janus kinase 1 (Jak1) and Jak3 associate with the mem- phorylated Y510 (7), becomes tyrosine phosphorylated, homo- ␤ ␥ brane proximal regions of IL-2R and c, respectively, and un- and/or heterodimerizes with other Stat molecules, and directly dergo catalytic activation upon ligand-induced heterodimerization translocates to the nucleus (22). Expression of Stat5 mutants that ␤ ␥ of IL-2R and c (4–6). Activation of Jak1 and Jak3 leads to lack trans-activation potential impairs the IL-2-activated prolifer- phosphorylation of at least three tyrosine residues on IL-2R␤ 338 392 510 ative response in lymphocytes (23, 24), and Stat5-deficient mice (Y ,Y , and Y ) (7). Primed by the phosphorylation events, have compromised T cell proliferative responses (25). Growth- the receptor complex generates two major proliferative signals. related target genes of Stat5 in the context of the IL-2R include One signal is mediated by the adaptor protein Shc (7, 8), which c-myc, bcl-x, bcl2 (23), IL-2R␣ (26, 27), and pim-1 (28). Other binds to phosphorylated Y338, undergoes tyrosine phosphorylation, target genes of Stat5 that have been identified downstream of other cytokine receptors include cyclin D1 (29) and p21cip1 (30). The D-type cyclins (D1, D2, and D3) are among the first reg- Virginia Mason Research Center, Seattle, WA 98101; and Department of Immuno- ulatory proteins to appear in G in response to mitogens. Cyclin D2 logy, University of Washington, Seattle, WA 98195 1 and D3 mRNA levels increase in early G in primary T cells stim- Received for publication July 10, 2000. Accepted for publication November 8, 2000. 1 ulated by IL-2 (31) or stimulated by PHA and 12-O-tetradecanoyl- 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 phorbol-13-acetate (32). Cyclin D2 and D3 protein expression is with 18 U.S.C. Section 1734 solely to indicate this fact. up-regulated in primary T cells stimulated by anti-CD3, but not in 1 This work was supported by Grant GM57931 from the National Institutes of Health T cells derived from Stat5-deficient mice (25). Several mecha- and by grants from The M. J. Murdock Charitable Trust and The William Randolph nisms for cyclin regulation that may be relevant to IL-2R signaling Hearst Foundation. A.M. was supported by a National Research Service Award Se- nior Fellowship from the National Cancer Institute. have been identified. In an erythroleukemia cell line, IL-3-activated 2 Address correspondence and reprint requests to Dr. Anthony Martino, Virginia Ma- Stat5 regulates the transcriptional activity of cyclin D1 at a specific son Research Center, Benaroya Research Institute, 1201 9th Avenue, Seattle, WA gene enhancer region (29). In colon carcinoma and breast cancer cell 98101-6907. E-mail address: [email protected] lines, serum-activated PI3K controls mRNA translation of D-type cy- 3 ␥ ␥ Abbreviations used in this paper: c, common -chain; Jak, Janus kinase; PI3K, clins (33). Finally, c-Myc has been implicated in the transcriptional phosphatidylinositol 3-kinase; SOE, splice overlap extension; buffer H, 20 mM HEPES (pH 7.9), 1 mM EDTA, 0.1 mM EGTA, 2 mM magnesium chloride, 1 mM induction of cyclin D2 in Rat1 and NIH-3T3 cells (34). sodium o-vanadate, 20 mM sodium fluoride, 1 mM DTT, 0.1 mM 4-(2-aminoethyl)- We have studied the transcriptional regulation of the cyclin D2 benzenesulfonyl fluoride, and 1 ␮g/ml leupeptin; TTBS, 0.1 M, pH 7.5; Tris base, 0.9% sodium chloride, and 0.05% Tween 20; TAD, trans-activation domain; CREB, gene in T cells. We report evidence of an enhancer element in the cAMP responsive element binding; CBP, CREB binding protein. cyclin D2 promoter/enhancer that binds Stat5, specificity protein

(Sp)1, and an unknown factor(s). Maximal enhancer activity re- and 0.15 M sodium citrate, pH 7.0) and UV cross-linked. Blots were pre- quires both the Stat5 and Sp1 binding sites, suggesting functional hybridized at 43°C in hybridization buffer (1 M sodium phosphate (pH cooperation between these factors. Additionally, receptor mu- 7.1), 2 mM EDTA, 2% BSA, 10% SDS, 50% formamide, and 0.16 mg/ml yeast transfer RNA or herring sperm DNA). To generate nucleic acid tagenesis studies indicate the endogenous cyclin D2 gene is reg- probes, a 1.2-kb EcoRI fragment of the cyclin D2 gene and a 1.2-kb PstI ulated by the Stat5 pathway downstream of IL-2 in the absence of fragment of murine GAPDH cDNA were radiolabeled with [␣-32P]dCTP Shc-mediated signaling. Thus, this work reveals a direct pathway using a random primed labeling kit (Roche, Indianapolis, IN) and were from the IL-2R to a key cell cycle regulatory gene via a mecha- purified with Centri-Sep spin columns (Princeton Separations, Adelphia, NJ). Probes were boiled for 5 min and added to blots in hybridization nism involving the transcription factors Stat5 and Sp1. buffer. After overnight incubation at 43°C, blots were washed two or three times with 2ϫ SSC/0.1% SDS at 55°C before autoradiography. Blots were Materials and Methods stripped between probings with a 2-min immersion in boiling water. Plasmid construction D2-Luc. Part (1624 bp) of the human cyclin D2 promoter/enhancer im- Western blots mediately upstream of the translation start site (provided by Dov Shiffman CTLL2 cells were washed three times in PBS and resuspended at 1 ϫ 106 (35)) was introduced into the luciferase vector PGL-3 (Promega, Madison, cells/ml. Cells were deprived of cytokine for 8 h, then harvested either WI) to generate the plasmid D2-Luc. unstimulated or at serial time points after cytokine stimulation. Cells were ؊1303, ؊1204, ؊444 D2-Luc. Deletions from the 5Ј end of the 1624-bp washed with buffer H (20 mM HEPES (pH 7.9), 1 mM EDTA, 0.1 mM promoter/enhancer were completed using convenient restriction enzyme EGTA, 2 mM magnesium chloride, 1 mM sodium o-vanadate, 20 mM sites for StuI, SmaI, and PvuII to produce 1303-, 1204-, and 444-bp frag- sodium fluoride, 1 mM DTT, 0.1 mM 4-(2-aminoethyl)benzenesulfonyl ments of the cyclin D2 gene in PGL-3 basic, respectively. fluoride, and 1 ␮g/ml leupeptin) and lysed at 108 cells/ml in buffer H plus ؊1303 D2-Luc mutated Sp1. The Sp1 site mutation was created by PCR- 0.2% Nonidet P-40 at 0°C. Nuclei were pelleted by centrifugation, and Downloaded from based site-directed mutagenesis of a 97-bp region between the StuI and protein was extracted with buffer K (buffer H plus 0.42 M sodium chloride SmaI sites of D2-Luc. The 97-bp fragment was introduced into Ϫ1303 and 20% (v/v) glycerol). Nuclear extracts were boiled in sample buffer, D2-Luc between the same restriction sites. separated by SDS-PAGE, and transferred to nitrocellulose. Nitrocellulose ؊1303 D2-Luc mutated Stat5. The Stat5 site mutation was introduced by membranes were blocked with TTBS (0.1 M, pH 7.5; Tris base, 0.9% splice overlap extension (SOE) PCR. The mutated, SOE-amplified region sodium chloride, and 0.05% Tween 20) containing 5% powdered skim was digested with KpnI and AflII and then introduced into Ϫ1303 D2-Luc milk (Carnation, Glendale, CA) and probed with polyclonal Abs to cyclin between the same sites. D2 or cdk2 (Santa Cruz Biotechnology, Santa Cruz, CA). Blots were then http://www.jimmunol.org/ washed with TTBS, probed with peroxidase-conjugated goat anti-rabbit Ab ؊1227 to ؊1168)-Luc. A 60-bp region of the human cyclin D2 gene) (Life Technologies), and washed again with TTBS. Bound Ab was de- from Ϫ1227 to Ϫ1168 was amplified by PCR and cloned into the PGL-3 tected by ECL (Amersham, Arlington Heights, IL). promoter vector between the SacI and XhoI sites. Other plasmids. Mutants of IL-2R␤ and Stat5 have been described previously (23). The DNA sequences of all plasmid regions subjected to re- EMSA striction enzyme- or PCR-based mutagenesis were confirmed by standard methods. Nuclear extracts were prepared as described for Western analysis. Protein concentrations were determined using the Bradford assay (Bio-Rad, Her- Cell culture and transfections cules, CA) and normalized across time points by dilution with buffer K. For EMSA probes Ͻ30 bp long, complementary sense and antisense strands of by guest on September 27, 2021 The murine IL-2-dependent T cell line CTLL-2 was obtained from Amer- DNA oligonucleotides were annealed, radiolabeled by an end-filling T4 ican Type Culture Collection (Manassas, VA) and was maintained in RPMI polymerase reaction, and purified with a MicroSpin G-25 column (Phar- (Life Technologies, Gaithersburg, MD) with 10% FCS, 2 mM L-glutamine, macia, Piscataway, NJ). EMSA probes longer than 30 bp were synthesized 50 U/ml penicillin, 50 ␮g/ml streptomycin, 1 mM sodium pyruvate, and 25 by PCR using Taq and [␣-32P]dCTP for radiolabeling and were purified mM 2-ME. Cells were passaged with 50 U/ml human rIL-2 (Chiron, Em- with a MicroSpin G-25 column. DNA probe (one part at 5000 cpm/␮l) and eryville, CA). In the experiments cells were stimulated with 100 U/ml IL-2. nuclear extract (one part) were mixed with two parts EMSA buffer (50 mM The murine IL-3-dependent pro-B cell line BA/F3 was obtained from potassium chloride, 15 mM HEPES (pH 7.9), 15% glycerol, 1 mM DTT, Immunex (Seattle, WA) and maintained in RPMI with 10% WEHI3-con- and 0.1 mg/ml poly(dI-dC)) and incubated at room temperature for 30 min. ditioned medium as a source of murine IL-3, 10% FCS, 2 mM L-glutamine, In competition reactions, unlabeled probe was added in at least a 10/1 50 U/ml penicillin, and 50 ␮g/ml streptomycin. BA/FG cells were modified molar excess over radiolabeled probe. In supershifting experiments, 2.2 ␮g from BA/F3 by introduction of a chimeric G-CSF receptor/gp130 receptor of Sp1 or Stat5 Ab was added per 10 ␮l of EMSA reaction (Santa Cruz chain (36). BA/FG cells were maintained on 100 ng/ml recombinant human Biotechnology). Reaction mixtures were electrophoresed on a nonreducing G-CSF (Amgen, Thousand Oaks, CA) in the absence of IL-3. Introduction 0.25ϫ Tris-buffered N-acrylamide gel and visualized by autoradiography. of the gp130 cytoplasmic receptor chain allows BA/FG cell growth through The probe sequences (sense strands) used in this study include: Ϫ1227 to SHP-2 and Stat3 rather than Stat5 (37). Ϫ1168, CAC TCG CCC CCT CCC CCT CCC GGG CCA TTT CCT AGA ␥ To generate stable BA/FG transfectants, linearized plasmids were in- AAG CTG CAT CGG TGT GGC CAC GCT; Fc receptor 1 promoter troduced into cells by electroporation, and transfectants were selected for element (FCR), TCG AGT ATT TCC CAG AAA AGG AAC AGC T; and resistance to G418 (Life Technologies) in 96-well plates at limiting dilu- Sp1, ATT CGA TCG GGG CGG GGC GAG C. tion to isolate independent subclones. Receptor expression was assessed by flow cytometry with an Ab to human IL-2R␤ (PharMingen, San Diego, CA). Stat5 expression was assessed by Western blot with Abs to Stat5 and Luciferase reporter gene assays the FLAG epitope tag (Transduction Laboratories (Lexington, KY) and CTLL2 cells were pelleted, resuspended at 1.25 ϫ 107 cells/ml in PBS with Sigma (St. Louis, MO)). Subclones with comparable IL-2R␤ and Stat5 10 mM MgCl , and incubated for 10 min at room temperature with 100 ␮g expression were chosen for further analyses. 2 of circular plasmid DNA. Transfection was conducted by electroporation at Northern blots 250 V and 960 ␮F, and the transfected cells were left at room temperature for an additional 10 min before being incubated overnight in complete BA/FG cells were washed three times in PBS and resuspended at 1 ϫ 106 medium plus IL-2. Cells were washed three times with PBS, split into six cells/ml. Cells were deprived of cytokine for 8 h, then harvested either groups, and incubated4hincomplete medium lacking IL-2. Three groups unstimulated or at serial time points after cytokine stimulation. Cells were were left unstimulated, and three groups were stimulated with IL-2 for 5 h. pelleted by centrifugation and were flash-frozen in a dry ice-ethanol bath. At5h,50␮l of cell culture was mixed with 450 ␮lof1ϫ Promega cell RNA was harvested with the RNA Stat 60 kit (Tel Test, Friendswood, TX); culture lysis reagent on ice. Fifteen microliters of lysate was mixed with 75 denatured for 10 min at 65°C in 20 mM MOPS, 5 mM sodium acetate, 0.5 ␮l of Promega luciferase substrate reagent, and luciferase activity was mM EDTA, 2.4 M formaldehyde, and 50% formamide; and run on a 1.2% measured with a United Technologies Packard Minaxi Tri-Carb 4000 se- agarose gel (containing 20 mM MOPS, 5 mM sodium acetate, 0.5 mM ries liquid scintillation counter (Downers Grove, IL). Means were calcu- EDTA, and 1.1 M formaldehyde) in 20 mM MOPS, 5 mM sodium acetate, lated for the three unstimulated and the three stimulated replicates, and fold and 0.5 mM EDTA at pH 7.0. RNA was passively transferred to Zetabind induction was calculated by dividing the mean for the stimulated cells by membranes (Cuno, Meriden, CT) with 10ϫ SSC (1.5 M sodium chloride the mean for the unstimulated cells. The Journal of Immunology 1725

Results Identification of an IL-2-responsive enhancer region in the cyclin D2 gene Experiments to identify an IL-2-responsive enhancer region in the cyclin D2 gene were performed in the murine CD8ϩ T cell line CTLL2. CTLL2 cells demonstrate robust Ag-independent growth in response to exogenous IL-2. IL-2 induced cyclin D2 mRNA and protein as assessed by Northern and Western analyses (Fig. 1). The short delay in cyclin D2 protein appearance relative to mRNA appearance indicates that cyclin D2 expression in response to IL-2 is largely regulated at the mRNA level. We studied the transcriptional regulation of the cyclin D2 gene in response to IL-2 using a luciferase reporter gene containing 1624 bp of the cyclin D2 promoter/enhancer (referred to as D2- Luc). The 1624-bp fragment represents the region immediately upstream of the translational start site in the cyclin D2 gene. D2- Luc was transiently transfected into CTLL2 cells. Transfected cells were deprived of IL-2 for 4 h and then were either left unstimu- Downloaded from lated or were stimulated with IL-2 for an additional 5 h. All re- FIGURE 2. Deletional analysis of a cyclin D2-luciferase reporter gene porter gene assays used triplicate samples and were repeated (D2-Luc) in CTLL2 cells indicates that IL-2 regulatory sites exist in two Ϫ 10–30 times to assure statistical significance. D2-Luc was induced regions flanking nucleotide 1204. Fold induction was measured for IL-2 stimulated vs unstimulated cells. Shown are the mean and SD obtained 2.7-fold in CTLL2 cells in response to IL-2 (Fig. 2). The level of from multiple experiments, each involving triplicate samples. induction was slightly lower than that observed for the endogenous cyclin D2 gene by Northern analysis, indicating that either additional enhancer sites exist outside of the 1624-bp cyclin D2 gene http://www.jimmunol.org/ probe spanning nucleotides Ϫ1227 to Ϫ1168 showed protein bind- fragment studied here or proper chromatin structure found only in ing changes in response to IL-2 (Fig. 3A). The probe spans nucle- the endogenous gene is required for full induction. otide Ϫ1204 and thus contains a portion of the functionally im- Deletion mutants of D2-Luc were evaluated to identify the IL- portant regions defined by the D2-Luc reporter gene (Fig. 3A). 2-responsive region(s) within the 1624-bp promoter/enhancer. The Before IL-2 stimulation, two bands were clearly observed with the region between Ϫ1624 and Ϫ1303 was dispensable for induction Ϫ1227 to Ϫ1168 probe (bands 1 and 2 at time zero). After stim- by IL-2 (Fig. 2). Deletion of the Ϫ1624 to Ϫ1204 region resulted ulation, a third and a fourth band appeared (bands 3 and 4), and the in a decrease in fold induction to 2.0, and deletion to the Ϫ444 site Ϫ original bands 1 and 2 diminished. Changes in protein-DNA com- diminished fold induction to 1.6. Thus, the regions between 1303 by guest on September 27, 2021 plexes represented by the four bands occurred within 30 min and to Ϫ1204 and Ϫ1204 to Ϫ444 appear to contain important regu- persisted for at least 8 h (data not shown). latory sites for induction of D2-Luc. The region downstream of The TRANSFAC program (38) identified a number of putative Ϫ444 bp contains binding sites for basal transcriptional machinery transcription factor binding sites within the 60-bp region, including (35) and possibly enhancer elements, but was not investigated fur- sites for Sp1 and Stat5. Ab supershifting and cold competition ther in this study. studies confirmed that the transcription factors Sp1 and Stat5 bind Ϫ Ϫ Stat5 and Sp1 binding sites flank the Ϫ1204 enhancer region to the 1227 to 1168 EMSA probe. An unlabeled oligonucleotide encoding a consensus Sp1 site competed with the Ϫ1227 to EMSAs were used to analyze a broad region surrounding nucleo- Ϫ1168 probe to eliminate bands 1 and 3. An unlabeled oligonu- Ϫ tide 1204 for IL-2-inducible binding of proteins to DNA. We cleotide encoding a defined Stat5 binding site (FCR) competed made 10 overlapping 60-bp DNA probes spanning the region from with the Ϫ1227 to Ϫ1168 probe to eliminate bands 3 and 4 (Fig. Ϫ Ϫ 1307 to 848. The DNA probes were mixed with nuclear ex- 3A). Similarly, addition of an anti-Sp1 Ab to the EMSA binding tracts from unstimulated and IL-2-stimulated CTLL2 cells. Several reaction eliminated bands 1 and 3, and anti-Stat5 Abs eliminated probes showed constitutive protein binding, and one probe showed band 4. Thus, band 1 is attributed to Sp1 binding, and band 4 is diminished protein binding with IL-2 stimulation. The EMSA attributed to Stat5 binding. Band 3 required both Sp1 and Stat5 binding. An unidentified protein or protein complex is represented by band 2. Point mutations confirmed the locations of the binding sites for Sp1, Stat5, and the unknown factor(s) within the Ϫ1227 to Ϫ1168 probe. Mutation of Ϫ1217 through Ϫ1214 bp (CTCC to AGAA) abrogated binding of Sp1 and the unknown factor(s), as evidenced by elimination of bands 1, 2, and 3 (Fig. 3B). Substi- tution of the highly conserved AA at Ϫ1192 and Ϫ1191 to CC abrogated Stat5 binding to the Ϫ1227 to Ϫ1168 probe as evidenced by elimination of bands 3 and 4. The relative locations of the Sp1 and Stat5 binding sites are well conserved between the human and rat cyclin D2 genes, further suggesting that this region FIGURE 1. Northern and Western blots showing induction of cyclin D2 may be functionally important (Fig. 3C). mRNA and protein in CTLL2 cells after IL-2 stimulation for the indicated The binding sites for Sp1 and the unknown factor(s) were fur- times. Protein appeared soon after mRNA levels increased, indicating that ther studied using smaller DNA probes. A probe of nucleotides induction is regulated primarily at the mRNA level. GAPDH and cdk2 Ϫ1227 to Ϫ1208 containing the Sp1 site showed two constitutive were used as controls to confirm equal loading across lanes. bands (Fig. 4). The upper band corresponds to Sp1, as determined 1726 CYCLIN D2 GENE REGULATION IN RESPONSE TO IL-2 BY Stat5 AND Sp1

FIGURE 4. EMSA analysis of a DNA probe (Ϫ1227 to Ϫ1208) cen-

tered around the Sp1 binding site. CTLL2 cells were stimulated with IL-2 Downloaded from for 0, 2, and 5 h. Two constitutive bands are seen. The upper band involves Sp1, as demonstrated by adding an unlabeled probe corresponding to a consensus Sp1 site or by adding anti-Sp1 Ab. The lower band corresponds to the unknown factor seen with the longer probe in Fig. 3. Mutation of the Sp1 site (CTCC to AGAA) eliminates both bands. http://www.jimmunol.org/

by supershifting Ab and cold competition EMSA reactions. The same Sp1 site base substitution used previously (Ϫ1217 to Ϫ1214 from CTCC to AGAA) eliminated both bands; therefore, the lower band in the Ϫ1227 to Ϫ1208 probe appears to represent the same factor(s) responsible for band 2 observed with the longer Ϫ1227 to Ϫ1168 probe. Putative binding sites for AP-2, myeloid zinc finger 1 (MZF1), and early growth response 1 (Egr-1) lie within the by guest on September 27, 2021 Ϫ1227 to Ϫ1208 region, but addition of unlabeled oligonucleotides encoding consensus binding sites for each of these three proteins failed to competitively eliminate the lower band. Point mutations throughout the Ϫ1227 to Ϫ1208 region failed to uncouple binding of Sp1 vs the unknown factor(s) (data not shown). EMSA reactions were performed with a 2-fold titration series of DNA probe to achieve limiting concentrations. The upper and lower bands disappeared at the same probe concentration (data not shown). Therefore, Sp1 and the unknown factor(s) bind to the same site with relatively equal affinity. We conclude that Sp1, Stat5, and an unknown factor(s) bind to the Ϫ1227 to Ϫ1168 probe flanking the Ϫ1204 enhancer site. The dependence of EMSA band 3 on the presence of both Sp1 and Stat5 is consistent with the formation of a complex containing FIGURE 3. In vitro characterization of nuclear proteins binding the constitutively bound Sp1 and inducibly bound Stat5. Analysis of Ϫ1227 to Ϫ1168 enhancer region of the cyclin D2 gene in IL-2-stimulated point mutants and smaller probes indicates that Sp1 and Stat5 bind CTLL2 cells. A, EMSA analysis of a 60-bp DNA probe representing the DNA independently of each other. The unknown factor(s) may Ϫ1227 to Ϫ1168 region of the human cyclin D2 gene. Arrows indicate also form an inducible complex with Stat5, which would account four bands of interest. Bands 1 and 2 were present in unstimulated cells and for the reduction in band 2 upon IL-2 stimulation. Alternatively, diminished with IL-2 stimulation. Bands 3 and 4 appeared in response to the unknown factor(s) may be inducibly removed from the DNA, IL-2. Competition EMSA reactions using unlabeled oligonucleotides cor- which would also diminish band 2. responding to consensus Sp1 and Stat5 sites (cold Sp1 and cold FCR) demonstrate that bands 1 and 3 involve Sp1, and bands 3 and 4 involve Stat5. Ab supershifting EMSA reactions (anti-Sp1 and anti-Stat5) confirm Contribution of the Stat5 and Sp1 Sites to transcriptional these conclusions. The dependence of band 3 on the presence of Sp1 and activity Stat5 indicates the formation of a complex of constitutively bound Sp1 and Mutational analysis of the Ϫ1303 D2-Luc reporter gene was used inducibly bound Stat5. An unknown factor(s) accounts for band 2. B, EMSA reactions using mutated versions of the Ϫ1227 to Ϫ1168 DNA to determine the importance of the Stat5 and Sp1 sites to tran- probe (underlined) confirm the binding sites for Stat5, Sp1, and the un- scriptional activity. IL-2-induced reporter gene activity was re- known factor(s). C, Sequence comparison between the human and rat cy- duced to a fold induction of 1.2 after mutation of the Stat5 site (AA clin D2 genes in the Sp1 and Stat5 regions. The Stat5 site is conserved to CC at nucleotides Ϫ1192 and Ϫ1191; Fig. 5A). Inducible re- exactly, while two possible Sp1 sites exist in the rat gene. porter gene activity was reduced by ϳ50% after mutation of the The Journal of Immunology 1727

The Stat5 pathway regulates expression of the endogenous cyclin D2 gene Mutants of IL-2R␤ and Stat5 were introduced into the lymphoid cell line BA/FG to investigate the role of Stat5 in the regulation of the endogenous cyclin D2 gene. BA/FG cells are a derivative of the IL- 3-dependent pro-B cell line BA/F3 (23). They express a chimeric G-CSF/gp130 receptor and so proliferate in response to G-CSF through a Stat5-independent signaling mechanism. Like BA/F3 cells, ␥ BA/FG cells constitutively express the c subunit of the IL-2R and can be made responsive to IL-2 by introduction of IL-2R␤. To test whether Stat5 mediates cyclin D2 induction, we introduced an IL-2R␤ mutant capable of activating Stat5 but not the Shc, Ras/mitogen activated protein (MAP) kinase, or PI3K pathways. The mutant receptor contained a distal portion of the IL-2R ␤-chain with a single Stat5-activating tyrosine residue (Y510)at- tached to a truncated form of IL-2R␤ lacking the Shc binding site at Y338and all other cytoplasmic tyrosines (Fig. 6A, IL-2R␤ ⌬325 ϩ Y510). ⌬325 ϩ Y510 induced cyclin D2 mRNA as effectively as Downloaded from the endogenous IL-3R (Fig. 6B). Stat5 activity was normal as assessed by EMSA (Fig. 6C). Next, we introduced a derivative of IL-2R␤ ⌬325 ϩ Y510 containing a leucine to arginine substitution at residue 511 (⌬325 ϩ Y510 RSL) that impairs the Stat5 binding site (23). ⌬325 ϩ Y510 RSL only weakly induced cyclin D2 mRNA in response to IL-2 (Fig. 6B) consistent with markedly reduced Stat5 activity (Fig. 6C). Finally, we rescued Stat5 activity by http://www.jimmunol.org/ overexpressing a FLAG epitope-tagged version of wild-type Stat5a with ⌬325 ϩ Y510 RSL. Stat5a overexpression overcomes the affinity barrier between Stat5 and ⌬325 ϩ Y510 RSL (23). Cyclin D2 mRNA induction was restored (Fig. 6B) with normal Stat5 activity (Fig. 6C). Cyclin D2 mRNA induction correlated with Stat5 activity when the Stat5 signal was isolated, disrupted, and then restored, indicating that Stat5 regulates the endogenous cyclin D2 gene. FIGURE 5. Contribution of the Sp1 and Stat5 sites near Ϫ1204 to the reporter gene activity. A, Analysis of the Ϫ1303 D2/Luc reporter gene with by guest on September 27, 2021 independent mutations of the Sp1 or Stat5 binding site. While the Stat5 site was essential for transcriptional activity, the adjacent Sp1 site was required The trans-activation domain (TAD) of Stat5 mediates cyclin D2 for maximum transcriptional induction. Ϫ1303, Ϫ1303 D2-Luc; mut Sp1, induction Ϫ1303 D2-Luc with Ϫ1217 to Ϫ1214 mutated from CTCC to AGAA; mut Stat5 ⌬713 is a naturally occurring isoform of Stat5a that lacks the Ϫ Ϫ Ϫ Stat5, 1303 D2-Luc with 1192 and 1191 mutated from AA to CC; TAD, but is capable of receptor-mediated tyrosine phosphoryla- Ϫ Ϫ Ϫ Ϫ 444, 444 D2-Luc. B, The 60-bp enhancer region from 1227 to 1168 tion, nuclear translocation, and DNA binding (39). BA/FG clones was placed upstream of the minimal SV40 promoter in the vector PGL-3 stably coexpressing a FLAG epitope-tagged version of Stat5 ⌬713 promoter (1227–1168 Luc). Luciferase induction was approximately equal ␤ ⌬ ϩ 510 to that of the full promoter/enhancer contained in D2-Luc, indicating that and IL-2R 325 Y were generated to test the requirement the Sp1/Stat5 enhancer region can function independently of other regions of the TAD for Stat5 induction of cyclin D2 mRNA. Stat5 ⌬713 of the cyclin D2 gene. ..mut Sp1, 1227–1168 with mutated Sp1 site as exerts no adverse selective pressure on BA/FG cells maintained on described in Fig. 3B; ..mut Stat5, 1227–1168 with mutated Stat5 site as G-CSF, because proliferative signaling occurs through SHP-2 and described in Fig. 3B. Stat3. Activation of Stat5 ⌬713 by IL-2R␤ ⌬325 ϩ Y510 failed to induce cyclin D2 mRNA (Fig. 6B). The inducible DNA binding activity of Stat5 ⌬713 was high, as confirmed by EMSA with and Sp1 site (CTCC to AGAA at nucleotides Ϫ1217 to Ϫ1214). Fold without a supershifting Ab to the FLAG epitope tag (Fig. 6C). induction measured with the mutated Sp1 site was approximately Thus, Stat5 mediates cyclin D2 induction through its TAD. equal to that measured with the Ϫ1204 D2-Luc reporter gene that lacks this region (refer to Fig. 2). We conclude that Stat5 is essential for IL-2-mediated induction of D2-Luc, and the Sp1 bind- Discussion ing site enhances transcriptional induction. We report on the transcriptional regulation of the cyclin D2 gene by To test whether the Stat5/Sp1 enhancer region is sufficient for the IL-2R in T cells. IL-2-responsive transcriptional activity is depen- gene induction, we cloned the Ϫ1227 to Ϫ1168 region into the dent on the formation of an enhancer complex at Ϫ1220 to Ϫ1191 bp vector PGL-3 promoter immediately upstream of the minimal containing Stat5, Sp1, and an unknown factor(s). Stat5 binds induc- SV40 promoter. Fold induction of (Ϫ1227 to Ϫ1168)-Luc in re- ibly in response to IL-2, and the other factors bind constitutively. In sponse to IL-2 was approximately equal to that observed with D2- a cyclin D2/luciferase reporter gene, the Stat5 site was essential for Luc (Fig. 5B); therefore, the Stat5 and Sp1 enhancer is able to IL-2-induced transcriptional activity, and maximum induction re- function independently of other regions of the cyclin D2 promoter/ quired the adjacent Sp1 binding site. In the lymphoid cell line BA/FG, enhancer. Mutation of the Sp1 or Stat5 site within (Ϫ1227 to we isolated, disrupted, and then rescued the Stat5 signal by introduc- Ϫ1168)-Luc abrogated IL-2 responsiveness, further demonstrating ing mutated versions of the IL-2R␤ subunit and wild-type murine functional cooperation between the two sites. Stat5a. Expression of the endogenous cyclin D2 gene correlated with 1728 CYCLIN D2 GENE REGULATION IN RESPONSE TO IL-2 BY Stat5 AND Sp1

ducibly binds an enhancer element containing constitutively bound Elf-1 (26, 27), and cooperativity between factors is required for maximal transcriptional activity. Similarly, Stat3 interacts directly ␣ with c-Jun to induce transcription of the 2-macroglobulin gene in response to IL-6 (42, 43). Mutants of Stat3 that fail to interact with c-Jun result in decreased gene induction. Finally, Stat5 has been shown to interact with the cofactor Nmi to enhance association with CBP (CREB (cAMP responsive element binding) binding protein)/p300 and promote IL-2-mediated transcription (44). Based on the above results, it has been proposed that Stats interact with other factors to synergistically increase transcription through the formation of enhanceosomes, combinations of proteins that promote gene expression through the recruitment of cofactors and through electrostatic and interfacial surface interactions with the basal transcriptional machinery (45). A proposed model for Sp1 function predicts strong Sp1 binding to consensus sites in promoters of constitutively expressed house- keeping genes and weaker binding and cooperation with signal-

responsive transcriptional activators in mitogen-induced genes Downloaded from (46). Consistent with this model, Sp1 binds to a nonconsensus site on the cyclin D2 gene and depends on adjacent Stat5 binding for transcriptional activity. Cholesterol-induced transcription of the gene encoding the low density lipoprotein receptor has been used as a model system to investigate the role of Sp1 in signal-respon-

sive genes (46, 47). Sp1 acts synergistically with sterol-responsive http://www.jimmunol.org/ element binding protein that recruits the cofactor CBP to the enhancer complex. It is proposed that Sp1 acts as a bridge between sterol-responsive element binding protein/CBP and the TATA binding protein-associated factors to enhance transcription. Simi- larly, on the cyclin D2 promoter Sp1 may link Stat5 and associated cofactors such as Nmi/CBP to the basal transcriptional machinery. Stat factors interact with Sp1 in at least two other cases. In rat Nb2 T cells, IL-2 stimulation induces Sp1 protein accumulation and binding to an Sp1 consensus oligonucleotide (48). Stat3 and by guest on September 27, 2021 Stat5 participation in the Sp1-DNA complex was detected by supershifting Ab EMSA reactions, but the contributions of these factors to transcriptional activity were not evaluated. In a second study induction of the gene encoding intercellular adhesion molecule-1 in response to IFN-␥ was shown to require cooperative FIGURE 6. Regulation of the endogenous cyclin D2 gene by Stat5 in interactions between constitutively bound Sp1 and inducibly BA/FG cells. A, Schematic representations of mutant IL-2R␤ and Stat5 bound Stat1 on an enhancer element (49). In this case Stat1 and isoforms used to isolate the IL-2-responsive Stat5 signaling pathway. B, Sp1 appeared to interact directly, as they could be coimmunopre- Northern blots indicate that the endogenous cyclin D2 gene was induced in cipitated from IFN-␥-stimulated cells. response to IL-2 when the Stat5 signaling pathway was isolated by using Our results establish a direct and immediate link between Stat5 the IL-2R␤ ⌬325 ϩ Y510 receptor mutant. IL-3 was used as a Stat5 sig- activation and cyclin D2 gene transcription, but expression of cy- naling control, and G-CSF was used as a Stat3 signaling control. With clin D2 mRNA in response to IL-2 is a delayed-early event (31) ⌬ ϩ 510 disruption of the Stat5 binding site on 325 Y RSL, cyclin D2 induction (J. J. Moon, unpublished observations). Therefore, it is likely that decreased. Restoring Stat5 activity by overexpressing Stat5a restored cyclin other essential activators or cofactors are synthesized upon IL-2 D2 induction (Y510 RSL ϩ Stat5a). Expression of a Stat5 isoform missing the stimulation that regulate cyclin D2 gene induction. Consistent with TAD (Stat5 ⌬713) blocked cyclin D2 induction, indicating that the TAD is essential. C, EMSA analysis measuring Stat5/DNA activity in each of the the delayed-early kinetics, c-Myc has been implicated in the reg- Northern blot studies from B. A DNA oligonucleotide encoding a defined ulation of cyclin D2 transcription in nonlymphoid cell lines. It has Stat5 binding site (FCR) was used. Cells were stimulated with medium alone been proposed that c-Myc regulates cyclin D2 transcription by (Ϫ), IL-3 (3), or IL-2 (2) for 15 min. In lanes marked Ab, an Ab against the competitively binding to Max and displacing a transcriptionally FLAG epitope on Stat5a and Stat5 ⌬713 was added to the EMSA reactions, repressive Mad-Max complex on an E box at nucleotide Ϫ1594 resulting in supershifting of DNA/protein complexes. (34). This model has not been confirmed using physiological levels of c-Myc, however. In fact, we show that the E box region of the cyclin D2 enhancer is dispensable for reporter gene inducibility Stat5 activity. Furthermore, a version of Stat5a lacking the TAD in- (Ϫ1303 D2-Luc; Fig. 2), and that the Stat5/Sp1 enhancer complex hibited cyclin D2 induction, demonstrating that the trans-activation can act independently of the rest of the cyclin D2 gene (Fig. 5B). function of Stat5 is required. Nevertheless, it remains possible that the E box and the Stat5/Sp1 The functional interaction between Stat5 and Sp1 is consistent enhancer element work in concert to control endogenous cyclin D2 with several examples of mitogen-activated transcriptional regu- mRNA levels with delayed-early kinetics. lation. Stat family members often act with other factors to promote It is not known whether IL-2 has a direct effect on Sp1 activity trans-activation potential (40, 41). In the IL-2R␣ gene, Stat5 in- in T cells. IL-2 does not alter DNA binding (Fig. 4) or Sp1 protein The Journal of Immunology 1729 levels (data not shown) in CTLL2 cells. As discussed above, this 24. Mui, A. L., H. Wakao, T. Kinoshita, T. Kitamura, and A. Miyajima. 1996. Sup- is consistent with several examples of mitogen-induced genes pression of interleukin-3-induced gene expression by a C-terminal truncated Stat5: role of Stat5 in proliferation. EMBO J. 15:2425. where Sp1 contributes to transcription. On the other hand, Sp1 25. Moriggl, R., D. J. Topham, S. Teglund, V. Sexl, C. McKay, D. Wang, phosphorylation is modulated during the cell cycle (50) and in A. Hoffmeyer, J. van Deursen, M. Y. Sangster, K. D. Bunting, et al. 1999. Stat5 is required for IL-2 induced cell cycle progression of peripheral T cells. Immunity response to cytokines such as Neu differentiation factor (51). Our 10:249. results do not exclude the possibility that IL-2 similarly regulates 26. John, S., C. M. Robbins, and W. J. Leonard. 1996. An IL-2 response element in the phosphorylation state of Sp1 within the enhancer complex, the human IL-2 receptor ␣ chain promoter is a composite element that binds Stat5, Elf-1, HMG-1(Y), and GATA family protein. EMBO J. 15:5627. thereby enhancing overall trans-activation potential. 27. Lecine, P., M. Algarte, P. Rameil, C. Beadling, P. Bucher, M. Nabholz, and J. Imbert. 1996. Elf-1 and Stat5 bind to a critical element in a new enhancer of Acknowledgments the human interleukin-2 receptor ␣ gene. Mol. Cell. Biol. 16:6829. 28. Matikainen, S., T. Sareneva, T. Ronni, A. Lehtonen, P. J. Koskinen, and We thank Bryan McIntosh and Meghan Crawford for technical assistance, I. Julkunen. 1999. Interferon-␣ activates multiple STAT proteins and upregulates Dov Shiffman for the cyclin D2 promoter/enhancer, James Ihle for cDNAs proliferation-associated IL-2R␣, c-myc, and pim-1 genes in human T cells. Blood encoding the wild-type and ⌬713 versions of Stat5a, and Charles Sherr for 93:1980. the cyclin D2 cDNA. 29. Matsumara, I., T. Kitamura, H. Wakao, H. Tanaka, K. Hashimoto, C. Albanese, J. Downward, R. G. Pestell, and Y. Kanakura. 1999. Transcriptional regulation of the cyclin D1 promoter by STAT5: its involvement in cytokine dependent growth References of hematopoietic cells. EMBO J. 18:1367. 1. Gesbert, F., M. Delespine-Carmagnat, and J. Bertoglio. 1998. Recent advances in 30. Matsumura, I., J. Ishikawa, K. Nakajima, K. Oritani, Y. Tomiyama, J. Miyagawa, the understanding of interleukine-2 signal transduction. J. Clin. Immunol. 18:307. T. Kato, H. Miyazaki, Y. Matsuzawa, and Y. Kanakura. 1997. Thrombopoietin- 2. Nelson, B. H., and D. M. Willerford. 1998. Biology of the interleukin-2 receptor. induced differentiation of a human megakaryoblastic leukemia cell line, CMK, in- WAF1/Cip1 In Advances in Immunology, Vol. 70. Academic Press, New York, p. 1. volves transcriptional activity of p21 by Stat5. Mol. Cell. Biol. 17:2933. Downloaded from 3. Takeshita, T., K. Ohtami, H. Asao, S. Kumaki, M. Nakamura, and K. Sugamura. 31. Turner, J. M. 1993. IL-2 dependent induction of G1 cyclins in primary T cells is 1992. An associated molecule, p64 with IL-2 receptor ␤ chain: its possible in- not blocked by rapamycin or cyclosporin A. Int. Immunol. 5:1199. volvement in the formation of the functional intermediate IL-2 receptor complex. 32. Ajchenbaum, F., K. Ando, J. A. DeCaprio, and J. D. Griffin. 1993. Independent J. Immunol. 148:2154. regulation of human D-type cyclin gene expression during G1 phase in primary 4. Witthuhn, B. A., O. Silvennoinen. O. Miura, K. S. Lai, C. Cwik, E. T. Liu, and human T lymphocytes. J. Biol. Chem. 268:4113. J. N. Ihle. 1994. Involvement of the Jak-3 Janus kinase in signalling by inter- 33. Muise-Helmericks, R. C., H. Leighton Grimes, A. Bellacosa, S. E. Malstro, leukins 2 and 4 in lymphoid and myeloid cells. Nature 370:153. P. N. Tsichlis, and N. Rosen. 1998. Cyclin D expression is controlled post- 5. Johnston, J. A., M. Kawamura, R. A. Kirken, Y. Chen, T. B. Blake, K. Shibuya, transcriptionally via a phosphatidylinosital 3-kinase/Akt dependent pathway.

J. R. Ortaldo, D. W. McVicar, and J. J. O’Shea. 1994. Phosphorylation and J. Biol. Chem. 273:29864. http://www.jimmunol.org/ activation of the Jak-3 Janus kinase in response to interleukin-2. Nature 370:151. 34. Bouchard, C., K. Thicke, A. Maier, R. Saffrich, J. Hanley-Hyde, W. Ansorge, 6. Higuchi, M., H. Asao, N. Tanaka, K. Oala, T. Takeshita, M. Nakamura, S. Reed, P. Sicinski, J. Bartek, and M. Eilers. 1999. Direct induction of cyclin D2 J. Van Snick, and K. Sugamura. 1996. Dispensability of Jak1 tyrosine kinase for by Myc contributes to cell cycle progression and sequestration of p27. EMBO J. interleukin-2-induced cell growth signaling in a human T cell line. Eur. J. Im- 18:5321. munol. 26:1322. 35. Brooks, A. R., D. Shiffman, C. S. Chan, E. E. Brooks, and P. G. Milner. 1996. 7. Friedmann, M. C., T. S. Migone, S. M. Russell, and W. J. Leonard. 1996. Dif- Functional analysis of the human cyclin D2 and cyclin D3 promoters. J. Biol. ferent interleukin 2 receptor ␤-chain tyrosines couple to at least two signaling Chem. 271:9090. pathways and synergistically mediate interleukin 2-induced proliferation. Proc. 36. Ziegler, S. F., T. A. Bird, K. M. Morella, B. Mosley, D. P. Gearing, and Natl. Acad. Sci. USA 93:2077. H. Baumann. 1993. Distinct regions of the human granulocyte-colony-stimulat- 8. Lord, J. D., B. C. McIntosh, P. D. Greenberg, and B. H. Nelson. 1998. The IL-2 ing factor receptor are required for proliferation and gene induction. Mol. Cell. receptor promotes proliferation, bcl-2 and bcl-x induction, but not cell viability Biol. 13:2384. through the adapter molecule Shc. J. Immunol. 161:4627. 37. Fukada, T., M. Hibi, Y. Yamanaka, M. Takahashi-Tezuka, Y. Fujitani, by guest on September 27, 2021 9. Ravichandran, K. S., V. Igras, S. E. Shoelson, S. W. Fesik, and S. J. Burakoff. T. Yamaguchi, K. Nakajima, and T. Hirano. 1996. Two signals are necessary for 1996. Evidence for a role for the phosphotyrosine binding domain of Shc in cell proliferation induced by a cytokine receptor gp130: involvement of STAT3 interleukin-2 signaling. Proc. Natl. Acad. Sci. USA 93:5275. in anti-apoptosis. Immunity 5:449. 10. Evans, G. E., M. A. Goldsmith, J. A. Johnston, W. Xu, S. R. Weiler, R. Erwin, 38. Heinemeyer, T., X. Chen, H. Karas, A. E. Kel, O. V. Kel, I. Liebich, O. M. Z. Howard, R. T. Abraham, J. J. O’Shea, W. C. Greene, and W. C. Farrar. T. Meinhardt, I. Reuter, F. Schacherer, and E. Wingender. 1999. Expanding the 1995. Analysis of interleukin-2 dependent signal transduction through the Shc/ TRANSFAC database towards an expert system of regulatory molecular mech- Grb2 adaptor pathway. J. Biol. Chem. 270:28858. anisms. Nucleic Acids Res. 27:318. 11. Harmer, S. L., and A. L. DeFranco. 1997. Shc contains two Grb2 binding sites 39. Wang, D., D. Stravdopodis, S. Teglund, J. Kitazawa, and J. N. Ihle. 1996. Nat- needed for efficient formation of complexes with SOS in B lymphocytes. Mol. urally occurring dominant negative variants of Stat5. Mol. Cell. Biol. 16:6141. Cell. Biol. 17:4087. 40. Chatterjee-Kishore, M., F. van den Akker, and G. R. Stark. 2000. Association of 12. Hill, C. S., and R. Treisman. 1995. Transcriptional regulation by extracellular STATs with relatives and friends. Trends Cell Biol. 10:106. signals: mechanisms and specificity. Cell 80:199. 41. Decker, T., and P. Kovarik. 1999. Transcription factor activity of STAT proteins: 13. Schaeffer, H. J., and M. J. Weber. 1999. Mitogen-activated protein kinases: spe- structural requirements and regulation by phosphorylation and interacting pro- cific messages from ubiquitous messengers. Mol. Cell. Biol. 19:2435. teins. CMLS Cell. Mol. Life Sci. 55:1535. 14. Frearson, J. A., and D. R. Alexander. 1997. The role of phosphotyrosine phos- 42. Schaefer, T. S., L. K. Sanders, and D. Nathans. 1995. Cooperative transcriptional phatases in haematopoietic cell signal transduction. BioEssays 19:417. activity of Jun and Stat3␤, a short form of Stat3. Proc. Natl. Acad. Sci., USA 15. Gu. H., J. C. Pratt, S. J. Burakoff, and B. G. Neel. 1998. Cloning of p97/Gab2, 92:9097. the major SHP-2 binding protein in hematopoietic cells, reveals a novel pathway 43. Zhang, X., M. H. Wrzeszczynska, C. M. Horvath, and J. E. Darnell, Jr. 1999. for cytokine-induced gene activation. Mol. Cell 2:729. Interacting regions in Stat3 and c-Jun that participate in cooperative transcrip- 16. Gadina, M., C. Sudarshan, and J. J. O’Shea. 1999. IL-2, but not IL-4 and other tional activation. Mol. Cell. Biol. 19:7138. cytokines, induces phosphorylation of a 98 kDa protein associated with SHP-2, 44. Zhu, M., S. John, M. Berg, and W. J. Leonard. 1999. Functional association of phosphatidylinositol 3-kinase, and Grb2. J. Immunol. 162:2081. Nmi with Stat5 and Stat1 in IL-2 and IFN␥-mediated signaling. Cell 96:121. 17. Gu, H., H. Maeda, J. J. Moon, J. D. Lord, M. Yoakim, B. H. Nelson, and 45. Carey, M. 1998. The enhanceosome and transcriptional synergy. Cell 92:5. B. G. Neel. 2000. New role for Shc in activation of the phosphatidylinositol 46. Naar, A. M., and R. Tjian. 1998. Cofactor requirements for transcriptional acti- 3-kinase/Akt pathway. Mol. Cell. Biol. 20:7109. vation by Sp1. In Cold Spring Harbor Symposia on Quantitative Biology. Vol. 18. Brennan, P., J. W. Babbage, B. M. T. Burgering, B. Groner, K. Reif, and 63. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, p. 189. D. A. Cantrell. 1997. Phosphatidylinositol 3-kinase couples the interleukin-2 re- 47. Naar, A. M., P. Beaurang, K. Robinson, J. D. Oliner, D. Avizonis, J. Zwicker, ceptor to the cell cycle regulator E2F. Immunity 7:679. S. Scheek, J. T. Kadonaga, and R. Tjian. 1998. Chromatin, TAFs, and a novel 19. Brennan, P., J. W. Babbage, G. Thomas, and D. Cantrell. 1999. p70s6K integrates multiprotein co-activator are required for synergistic activation by Sp1 and phosphatidylinositol 3-kinase and rapamycin-regulated signals for E2F regulation SREBP-1␣ in vitro. Genes Dev. 12:3020. in T lymphocytes. Mol. Cell. Biol. 19:4729. 48. Too, C. K. L. 1997. Induction of Sp1 activation by prolactin and interleukin-2 in 20. Gilmour, K. C., R. Pine, and N. C. Reich. 1995. Interleukin 2 activates Stat5 Nb2 T-cells: differential association of Sp1-DNA complexes with Stats. Mol. transcriptional factor (mammary gland factor) and specific gene expression in T Cell. Endocrinol. 129:7. lymphocytes. Proc. Natl. Acad. Sci. USA 92:10772. 49. Look, D. C., M. R. Pelletier, R. M. Tidwell, W. T. Roswit, and M. J. Holtzman. 21. Hou, J., U. Schinder, W. J. Henzel, S. C. Wang, and S. C. McKnight. 1995. 1995. Stat1 depends on transcriptional synergy with Sp1. J. Biol. Chem. 270: Identification and purification of human stat proteins activated in response to 30264. interleukin-2. Immunity 2:321. 50. Black, A. R., D. Jensen, S. Lin, and J. C. Azizkhan. 1999. Growth/cell cycle 22. Schindler, C. 1999. Cytokines and JAK-STAT signaling. Exp. Cell Res. 253:7. regulation of Sp1 phosphorylation. J. Biol. Chem. 274:1207. 23. Lord, J. D., B. C. McIntosh, P. D. Greenberg, B. H. Nelson. 2000. The IL-2 51. Alroy, I., L. Soussan, R. Seger, and Y. Yarden. 1999. Neu differentiation factor receptor promotes lymphocyte proliferation and induction of the c-myc, bcl-2, and stimulates phosphorylation and activation of the Sp1 transcription factor. Mol. bcl-x genes through the transactivation domain of Stat5. J. Immunol. 164:2533. Cell. Biol. 19:1961.