Transcriptional Activation of the SH2D2A Is Dependent on a Cyclic 5 ′ -Monophosphate-Responsive Element in the Proximal SH2D2A This information is current as of September 29, 2021. Ke-Zheng Dai, Finn-Eirik Johansen, Kristin Melkevik Kolltveit, Hans-Christian Aasheim, Zlatko Dembic, Frode Vartdal and Anne Spurkland J Immunol 2004; 172:6144-6151; ; doi: 10.4049/jimmunol.172.10.6144 Downloaded from http://www.jimmunol.org/content/172/10/6144

References This article cites 35 articles, 14 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/172/10/6144.full#ref-list-1

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists by guest on September 29, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

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 © 2004 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Transcriptional Activation of the SH2D2A Gene Is Dependent on a Cyclic Adenosine 5؅-Monophosphate-Responsive Element in the Proximal SH2D2A Promoter1

Ke-Zheng Dai,*§ Finn-Eirik Johansen,† Kristin Melkevik Kolltveit,‡ Hans-Christian Aasheim,¶ Zlatko Dembic,‡ Frode Vartdal,* and Anne Spurkland2*§

The SH2D2A gene, encoding the T cell-specific adapter (TSAd), is rapidly induced in activated T cells. In this study we investigate the regulation of the SH2D2A gene in Jurkat T cells and in primary T cells. Reporter gene assays demonstrated that the proximal 1-kb SH2D2A promoter was constitutively active in Jurkat TAg T cells and, to a lesser extent, in K562 myeloid cells, Reh B cells, and 293T fibroblast cells. The minimal SH2D2A promoter was located between position ؊236 and ؊93 bp from the

first coding ATG, and transcriptional activity in primary T cells depended on a cAMP (CRE) centered around Downloaded from position ؊117. Nuclear extracts from Jurkat TAg cells and activated primary T cells contained binding activity to this CRE, as observed in an EMSA. Consistent with this observation, we found that a cAMP analog was a very potent of SH2D2A mRNA expression in primary T cells as measured by real-time RT-PCR. Furthermore, activation of SH2D2A expression by CD3 stimulation required cAMP-dependent protein kinase activity. Thus, transcriptional regulation of the SH2D2A gene in activated T cells is critically dependent on a CRE in the proximal promoter region. The Journal of Immunology, 2004, 172: 6144–6151. http://www.jimmunol.org/ cell activation is controlled by a variety of different in- regulation of T cell activation are inducible or up-regulated after tracellular signaling , including adapter proteins the initial activation of T cells. One example of the latter is the T that serve as scaffolds for protein-protein interactions dur- CTLA-4 gene, which encodes a membrane receptor competing ing the signal transduction cascade (1). Proper down-regulation of with CD28 for binding to B7. CTLA-4 is induced in T cells after T cell activation is important for the normal functioning of the activation (8) and has been found to play an inhibitory role in T immune system. Failure to turn off activated T cells may lead to cell activation (9). Polymorphism in the CTLA-4 gene influencing lymphoproliferation and autoimmune disease, as seen in patients the expression level of the protein is associated with a number of with mutations in encoding proteins related to regulation of diseases with presumed autoimmune etiology (10). activation-induced apoptosis in T cells (2). The importance of We have recently cloned a novel gene, SH2D2A, that encodes the by guest on September 29, 2021 proper regulation of T cell activation for immune homeostasis is T cell-specific adapter protein (TSAd).3 TSAd inhibits early signal also exemplified by experimental animals in which the expression transduction events when overexpressed in Jurkat T cell lines (11) and of regulatory proteins has been abolished by genetic engineering. is rapidly induced in activated T cells (11, 12) SH2D2A mRNA ex- As an example, mice lacking expression of the adapter protein pression is observed in primary T cells 2 h after stimulation with cas-br-m murine ecotropic retroviral transforming sequence homo- anti-CD3, anti-CD4, or anti-CD8 Abs or after exposure to activated logue (cbl), which is a negative regulator of T cell signaling, de- macrophages in vitro (12). TSAd may have multiple functions in the velop spontaneous autoimmunity (3). cell, as it is observed to translocate to the nucleus in activated Jurkat Proteins involved in signal transduction in T cells may be ubiq- T cells (13). The murine counterpart has been cloned as a binding uitously expressed, such as phosphoinositide 3-kinase, protein ki- partner for Itk and Rlk (14) and MEKK5, a mitogen-activated protein nase C, and phosphoprotein associated with glycosphingolipid-en- kinase kinase (15). We have also found that a GA repeat-length poly- riched microdomains. Other proteins, such as the protein tyrosine morphism in the promoter of the SH2D2A gene is associated with kinases Lck (4) and ZAP-70 (5) and the adapter proteins TCR- multiple sclerosis (16), suggesting that TSAd dysregulation may be interacting molecule (6) and Src kinase-associated phosphoprotein associated with immunological disorders. In support of this idea, it of 55 kDa (7), are preferentially expressed in T cells or cells of was recently observed that mice lacking the murine TSAd gene de- lymphoid origin. Moreover, some of the proteins involved in the velop spontaneous autoimmune disease with age (17). TSAd may therefore represent an inducible modulator of signal transduction in activated T cells. *Institute of Immunology and †Institute of Pathology, National University Hospital; In this study we have examined the regulation of SH2D2A ex- ‡Department of Oral Biology and ¤Institute of Anatomy, University of Oslo; and ¶Department of Immunology, Norwegian Radium Hospital, Oslo, Norway pression in more detail. Luciferase reporter gene assay demonstrated that a 1-kb fragment of the SH2D2A promoter was sufficient for tis- Received for publication May 29, 2003. Accepted for publication February 27, 2004. sue-specific expression. We found that the transcriptional activity of 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 the minimal promoter in primary T cells was dependent on a cAMP with 18 U.S.C. Section 1734 solely to indicate this fact. response element (CRE) located around position Ϫ117 from the first 1 This work was supported by the Norwegian Research Council, the Norwegian Can- ATG. Consistent with this observation, we found that a synthetic cer Society, the Odd Fellow Fund for Multiple Sclerosis, The Anders Jahre Founda- tion, and Medinnova. 2 Address correspondence and reprint requests to Dr. Anne Spurkland, Institute of 3 Abbreviations used in this paper: TSAd, T cell-specific adapter protein; ATF, ac- Anatomy, Faculty of Medicine, University of Oslo, Postboks 1105, Blindern, 0317 tivation factor; 8-CPT-cAMP, 8-(4-chlorophenylthio)-cAMP; CRE, Oslo, Norway. E-mail address: [email protected] cAMP response element; CREB, CRE-binding protein; PKA, protein kinase A.

Copyright © 2004 by The American Association of Immunologists, Inc. 0022-1767/04/$02.00 The Journal of Immunology 6145 cAMP analog was a very potent inducer of SH2D2A mRNA expres- measured (Dual-Luciferase Reporter Assay System; Promega). The pro- sion in primary T cells. moter activity was evaluated as the relative activity between the firefly luciferase activity and the Renilla luciferase activity in each cell lysate. Adherent cells (293T) were transfected by Lipofectamine reagent (In- Materials and Methods vitrogen, Paisley, U.K.) according to the protocol of the manufacturer with Cells 5 ␮g of reporter gene plasmids and 1 ␮g of control plasmid for each 2 ϫ 106 cells. After incubation with transfection solution for 12 h, cells were T cells (Jurkat TAg), B cells (Reh), and myeloid cells (K562) were grown washed and harvested in PBS and subsequently analyzed for luciferase in RPMI 1640 supplemented with 10% FCS, whereas fibroblasts (293T) activity as described above. were grown in RPMI 1640 supplemented with 5% FCS. Fibroblasts were Primary T cells were transfected using the AMAXA nucleofector kept below a cell density of 80% confluence, whereas cells in suspension (AMAXA, Cologne, Germany) according to the manufacturer’s description. ϫ 6 ϫ 6 were kept between 0.3 10 and 1 10 /ml. Cell viability was always Briefly, 5 ϫ 106 cells were transfected with 3 ␮g of each SH2D2A-promoter Ͼ verified to be 95% before transfection. reporter gene construct and 0.2 ␮g of the internal transfection control plasmid. Primary resting T cells for transfection studies and SH2D2A mRNA The cells were then stimulated with anti-CD3/CD28-coated Dynabeads (0.5 expression analysis were positively selected from buffy coats as previously bead/cell; Dynal Biotech) for 16 h before assaying for luciferase activity as described (18) using anti-CD4-coated Dynabeads (Dynal Biotech, Oslo, described above. Norway). Primary T cells for EMSA were isolated from buffy coats by separation over nylon wool columns according to standard procedure. Cells Quantitative RT-PCR of SH2D2A transcripts were stimulated as described in Results using CD3 or CD3/CD28 Dyna- beads (Dynal Biotech), the cAMP analog 8-(4-chlorophenylthio)-cAMP For RT-PCR, total RNA was extracted from T cells using TRIzol reagent (8-CPT-cAMP), or 2 ␮g/ml PHA. (Life Technologies, Merlbebeke, Belgium). Slight modifications were made to the protocol (increased incubation time at all steps, the amount of isopropanol increased from 500 ␮l to 1 ml). cDNA was synthesized in a 20-␮l reaction

Reporter gene constructs Downloaded from buffer containing 20 U/␮l Moloney murine leukemia virus reverse transcrip- A 1-kb fragment of the SH2D2A promoter containing 23 GA repeats was tase, 250 ␮g/ml oligo(dT), 10 ␮M dNTP (2.5 ␮M concentrations of each), 36 subcloned. Briefly, a PCR product was obtained by amplifying the 1-kb U/␮l RNase inhibitor (RNAguard; Pharmacia Biotech), and 5ϫ RT buffer by promoter fragment from the 1-specific clone AH118a10 incubation for1hat37¡C. The enzyme was inactivated for 5 min at 90¡C. The (LL01NC01; HGMP, Cambridge, U.K.) (19) using the following primers: cDNA were stored at Ϫ20¡C until further processing. primer 1, 5Ј-CCAGCCTGGGTGACAGAG-3Ј; and primer 2,5Ј-AACTCCA Quantitation of TSAd transcripts was performed by real-time PCR, TGGGGGCAGCCTC-3Ј to introduce a NcoI site at the first coding ATG. The using the following custom-made primers and probes specific for TSAd blunt-ended PCR product was inserted into the pMOSBlue cloning vector and Zap70, respectively: HuTSAdFW, 5Ј-TGCTACTTGGTGCGGTTCAG; http://www.jimmunol.org/ (Pharmacia Biotech, Uppsala, Sweden) for sequencing. The verified insert huTSAdRV,5Ј-GCAAGTCCGGCTCCTGTAAG;HuTSAd-probe,6AGAGC was then moved into a luciferase reporter plasmid (pGL3-Basic, Promega, GCGGTGACCTTCGTG0; HuZAP-70FW, ACACCCTCAACTCAGATGG Madison, WI). This construct was designated GA23-GL3. An additional ATACA; HuZAP-70RV, TCGGCCGCGGTTTGT; HuZAP-70probe, 6CCC construct containing an additional 2-kb fragment 5Ј of the promoter as well TGAGCCAGCACGCATAACGT0. Both probes were labeled with FAM at as 13 GA repeats at position Ϫ340 was similarly cloned and was desig- the 5Ј end and with TAMRA at the 3Ј end. Primers and probes were designed nated GA13-long-GL3. using Primer Express (PE Applied Biosystems, Foster City, CA) and were Truncations of the promoter construct were performed using the ordered from Sigma-Genosys (Cambridge, U.K.). QuikChange kit (Stratagene, La Jolla, CA) following the instructions of the A standard curve was constructed for TSAd and ZAP-70. Serial dilu- manufacturer. To truncate the promoter, KpnI restriction sites were intro- tions of plasmids containing an insert with either TSAd or ZAP-70 were duced at various positions in the SH2D2A promoter construct GA23-GL3, made containing 1000, 100, 10, and 1 fg/␮l the plasmid. The quantity of and deletions were made by digesting the generated construct with KpnI, TSAd was estimated relative to the quantity of ZAP-70. The choice of by guest on September 29, 2021 followed by religation of the vector fragment. For mutation of the palindrome ZAP-70 as a housekeeping gene was based on our previous studies that at position Ϫ96, an EcoRV site was introduced to disrupt the palindromic showed that the amount of ZAP-70 is relatively stable during the various structure. The palindrome sequence was thus changed from 5Ј-CCTGCCCC phases of T cell activation (16). CGGGGCCAGG-3Ј into 5Ј-CCTGCCCCCGATATCAGG-3Ј. All SH2D2A All PCR were performed using an ABI PRISM 7900 Sequence Detec- promoter constructs generated by PCR were sequenced to ascertain that spu- tion System (PerkinElmer, PE Applied Biosystems). The reaction mixture rious mutations were not included in the analysis. contained the TaqMan Universal Master Mix (PerkinElmer, PE Applied Biosystems) and specific primers/probe set. The thermal cycling conditions Transient transfection assay comprised an initial denaturation step at 95¡C for 10 min and 40 cycles at Cells in suspension (Jurkat TAg, Reh, and K562) were transfected by elec- 95¡C for 15 s and at 60¡C for 1 min. Each PCR run included the four points troporation. Cells at a density of ϳ1 ϫ 106/ml were washed twice with of the standard curves and nontemplate controls. Experiments were per- RPMI 1640 medium without any additives before transfection. Cells (10 ϫ formed in triplicate (or duplicate when indicated) for each data point. 6 10 ) were transfected with 5 ␮g of the reporter plasmid gene (fireflylu- EMSA ciferase) and 1 ␮g of the internal control plasmid, a thymidine kinase promoter-directed Renilla luciferase plasmid (TK-RL2; Promega). Elec- EMSA was performed according to standard methods (20). Nuclear ex- troporation was performed with a BTX electroporator (Genetronics, San tracts of Jurkat T cells were prepared by extraction of isolated nuclei in 0.3 Diego, CA) using a 200 low voltage setting and 70 ms at room temperature M KCl. Radiolabeled probes generated using oligonucleotides (Table I) in a 0.4-cm electroporation cuvette (BTX; Genetronics). Sixteen to 24 h covering the entire minimal promoter (Ϫ246 to Ϫ47) or 45- to 65-bp parts after transfection, 1.5 ϫ 106 cells from each transfection were harvested thereof were generated using PCR amplification and inclusion of and washed twice with PBS (pH 7.4). Transfectants were resuspended in [32P]␣CTP in the PCR. Probes were purified by centrifugation through 50 ␮lof1ϫ Passive Lysis Buffer (Promega), and luciferase activity was Micro Spin S-400 HR columns (Amersham Pharmacia Biotech, Little

Table I. Oligos used for generating electromobility shift probes

Position in SH2D2A Oligo Sequence Promoter from ATG

RT GTGTGCACAC CTTTCCGGA Ϫ63 to Ϫ47 1 AGGGTGAGAG GAGAGATGAT GAGTTGCTGA Ϫ95 to Ϫ66 2 CCTGCCCCCG GGGCCAGGGT GAGAGGAGAG Ϫ110 to Ϫ81 3 TAAGGCGACA TTTCCTGCCC CCGGGGCCAG Ϫ123 to Ϫ94 4 AACTAAGGCT GTGGGTAAGG CGACATTTCC Ϫ138 to Ϫ109 5 ACGTGAGTCC TCGCCAGAAC TAAGGCTGTG Ϫ155 to Ϫ126 6 CTTAAGGCTT TCCATGACGT GAGTCCTCGC Ϫ171 to Ϫ142 240 Kp GAGTGACACA GGTACCCCCC ACCGC Ϫ246 to Ϫ122 6146 TRANSCRIPTIONAL REGULATION OF THE SH2D2A GENE

Chalfont, U.K.). EMSA was performed by incubating 2 ␮g of nuclear extract with 1 fmol of radiolabeled probe in the presence of 1.5 ␮g of dIdC as a nonspecific competitor with or without 100 fmol of cold specific com- petitor or nonspecific competitor derived from the coding part of the SH2D2A cDNA in a 20-␮l reaction buffer containing 20 mM HEPES (pH 7.9), 20% glycerol, 0.2 mM EDTA, 0.1 M KCl, 0.5 mM PMSF, 1 mM DTT, and 1 mM sodium orthovanadate for 20 min at room temperature. Binding reactions were separated on a 4.5% nondenaturing polyacrylamide gel in 1ϫ Tris-boric acid-EDTA buffer with 10% glycerol for 2 h. The gels were dried and exposed for autoradiography. EMSA supershift analysis was performed using a rat polyclonal anti- serum against aa 5Ð45 of CRE-binding protein 1 (CREB-1; Abcam, Cam- bridge, U.K.), a mouse mAb against aa 39Ð217 of activation transcription factor 1 (ATF1) (Santa Cruz Biotechnology, Santa Cruz, CA), a rabbit polyclonal antiserum raised against amino acids 1Ð96 of ATF2 (Santa FIGURE 1. SH2D2A promoter activity in various cell lines. A 1- or Cruz) and a rabbit polyclonal antiserum raised against the C-terminal pep- 3-kb SH2D2A promoter fragment was isolated from the genomic cosmid tide of CREB-2 or ATF4 (Santa Cruz Biotechnology). Nuclear extracts clone AH118a10 and inserted into a firefly luciferase vector to create ␮ ␮ were preincubated with 40 g of antiserum or 2 g of Ab before addition GA23-GL3 and GA13-long-GL3 long, respectively. These constructs were of radiolabeled probe. The binding reaction was separated by electrophore- transfected into cell lines representing fibroblasts (293T), B cells (Reh), T sis as described above. CREB competitor oligo was designed based on a cells (Jurkat Tag), and myeloid cells (K562), respectively. The thymidine previous report (21) as well as on the CREB consensus sequence as found kinase promoter-driven Renilla luciferase construct (TK-RL2) was used as in the TRANSFAC 4.0 (22). (CREB-con: 5Ј-CTCCTTGGCTGACGTCA

GAGAGAG). An additional commercially available CRE gel shift oligo- an internal control. Results are shown as the percentage of firefly luciferase Downloaded from nucleotides (CRE-SC: 5Ј-AGAGATTGCCTGACGTCAGAGAGCTAG) activity normalized to Renilla luciferase activity for each cell type com- was obtained from Santa Cruz Biotechnology. pared with that in Jurkat Tag cells. The results shown are the mean of two independent experiments Ϯ SD. Prediction of the putative transcription factors binding sites in the minimal promoter

The putative transcription factors binding sites were predicted by using Using the MatInspector Professional and Alibaba2 programs, http://www.jimmunol.org/ MatInspector Professional program (Genomatix Software GmbH, Munich, several putative binding sites for transcription factors were iden- Germany; (23) or Alibaba2 (BIOBASE GmbH, Wolfenbu¬ttel, Germany) tified in the Ϫ236 to Ϫ93 fragment (data not shown). Some of which both use transcription factor database TRANSFAC 4.0. The DNA sequence from Ϫ236 to ϩ 1 upstream of the first ATG of the SH2D2A these are ubiquitous transcription factors (Sp1 and OCT-1) (24, 25) promoter containing the minimal promoter was submitted to the program whereas some are implied in expression of genes and retroviruses for analysis. in T cells (NFAT, SEF and CREB) (26Ð28). This region contains also a nearly perfect 18 bp palindromic sequence (CCTGc- Results CCCCGGGGcCAGG), at position Ϫ110 to Ϫ93. As the SH2D2A Ј Tissue specificity and inducibility of the SH2D2A promoter cDNA starts just 3 of the palindrome (12), we speculated that this palindrome could be crucial for transcription initiation of the by guest on September 29, 2021 We previously observed that expression of the SH2D2A gene is SH2D2A gene. To test this hypothesis, SH2D2A promoter con- rapidly induced in T cells after triggering of the CD3, the CD4 or structs where the palindrome had been mutated to a nonpalin- the CD8 receptor (12). To test the tissue specificity of the SH2D2A dromic sequence were analyzed as above. These mutants showed promoter we subcloneda1kbanda3kbfragment of the promoter generally a higher expression of luciferase than the wild type con- extended to the ATG start codon into a luciferase reporter vector. structs (data not shown). Moreover, additional constructs where These constructs contained a 23 GA repeat allele (GA -GL3) and 23 also the putative Sp1 site in the 5Ј end of the palindrome had been a 13 GA repeat allele (GA -long-GL3) respectively, and were 13 removed showed similar expression level as the wild type con- transfected into fibroblast cells (293T), B cells (Reh), myeloid structs (data not shown). Thus, the palindrome is not necessary for cells (K562) and T cells (Jurkat TAg). The results showed that the transcription initiation of the SH2D2A gene. both SH2D2A-promoter constructs promoted luciferase expression in Jurkat T cells and K562 cells, and to a lesser extent in Reh and Identification of nuclear proteins binding to the minimal 293T cells (Fig. 1). The expression of the GA23-GL3 construct was SH2D2A promoter only marginally influenced by stimulation of the transfected Jurkat To identify nuclear proteins binding to the minimal promoter of T cells through the TCR/CD3 complex, as we observed stimulation the SH2D2A gene, we performed electromobility shift assay indexes of 1.4Ð1.6 in repeated experiments (data not shown). (EMSA) with a DNA fragment extending from Ϫ246 to Ϫ47 bp of the SH2D2A gene. We found that nuclear extracts from both un- Identification of promoter sequences influencing the expression stimulated (Fig. 3A) and PMA/ionomycin stimulated Jurkat cells of the SH2D2A gene (data not shown) contained similar amounts of binding activity The SH2D2A gene contains no TATA box or other known regu- toward this fragment. To further characterize the nuclear binding latory elements for transcription initiation (19). To identify the activity, overlapping radiolabeled PCR products (ϳ60 bp each) minimal promoter of the SH2D2A gene, we performed a series of covering the minimal promoter were used as probes in the EMSA ϳ truncations of the GA23-GL3 constructs removing 100 bp from the (Fig. 3B). We found that the overlapping 6Ð3, 5Ð3 and 4Ð1 probes, 5Ј end in each consecutive truncation (Fig. 2A). The constructs were covering bp Ϫ171 to Ϫ66 of the SH2D2A promoter was retarded tested for promoter activity in Jurkat T cells. The data showed that to the same extent by a factor in the nuclear extract of unstimulated there are putative elements between regions Ϫ1010/-820, Jurkat cells (Fig. 3C). Also the 6Ð4, the 2-R and 1-R probes dis- Ϫ630/-500 and Ϫ310/-240, and a putative element in the played retarded bands, which however migrated differently from region encompassed by Ϫ500/-310. The minimal promoter of the that of the probes covering the Ϫ171 to Ϫ66 segment, and was SH2D2A gene was found to be located proximally to position Ϫ236 thus not pursued further in the present study. The 4Ð3 oligonucle- upstream of the first coding ATG (Fig. 2B). We decided to first focus otide representing the minimal shared region between the 6Ð3, 5Ð3 our interest on the minimal promoter of the SH2D2A gene. and 4Ð1 probes DNA (i.e., position Ϫ138 to Ϫ94) still The Journal of Immunology 6147

FIGURE 2. Identification of promoter sequences influencing expression of the SH2D2A gene. A, Schematic diagram of GA23-GL3 and truncated versions generated as described in Materials and Methods. B, Luciferase activity was assessed in lysates of Jurkat TAg cells transiently transfected with Downloaded from the indicated promoter constructs and TK-RL2 as an internal control vector. Results shown are the mean of the normalized firefly luciferase activity Ϯ SD of three independent experiments.

Ϫ contained the binding activity. This oligonucleotide contained sev- at 117 in the SH2D2A promoter is cAMP responsive and im- http://www.jimmunol.org/ eral putative binding sites for transcription factors shown in Fig. portant for transcription of the SH2D2A gene. 3E. To identify more precisely the binding site on probe 4Ð3, we Of note is that the D mutation also affected the basal activity of made several corresponding oligonucleotides with specific point the reporter constructs in primary T cells, suggesting that the CRE- mutations and used these as competitors in EMSA experiments binding factor is present in resting T cells. We therefore examined (Fig. 3E and F). Mutants A, B, C, E and F competed with probe whether nuclear extracts from resting primary T cells display bind- 4Ð3 for binding, suggesting that these mutations do not affect the ing activity to the 240-R probe representing the minimal SH2D2A binding site of this NF. Mutation D, however, did not compete promoter or the 4-3 probe containing the CRE. In freshly isolated with the wild-type probe, indicating that this mutation abolished a primary T cells (isolated through passage through a nylon column), protein binding site. This mutant was located between position nuclear binding activity to the 240-R probe (Ϫ246 to Ϫ47 bp) was by guest on September 29, 2021 Ϫ119 to Ϫ115 and affected a putative cAMP response element observed in EMSA (data not shown). However, in T cells kept in (CRE) site. However, neither a custom made CRE consensus oli- complete medium without stimulation overnight, no binding activity gonucleotide (C-C) (data not shown), nor a commercially available to the 240-R probe and only weak binding activity to the 4-3 probe CRE consensus oligonucleotide (CRE-SC, Fig. 3G) could compete were observed, whereas binding activity to both probes could be in- for binding to probe 4Ð3. duced after2hofstimulation with PHA (Fig. 5, A and B, respec- Several transcription factors are known to promote transcription tively). This result suggests that the NF binding to the CRE in the of genes by binding to CREs (29). We therefore incubated the SH2D2A promoter becomes rapidly activated in T cells that are ma- DNA-binding reaction with antiserum against CRE binding pro- nipulated in vitro. This may explain some of the basal activity ob- tein (CREB1), activating transcription factor 1 (ATF1), ATF2, and served in primary T cells in our in vitro reporter gene experiments. ATF4. However, none of these four Abs displayed inhibition or supershifting of binding activity to probe 4-3 (Fig. 3G). Cyclic AMP analogs induces SH2D2A mRNA expression in primary T cells Role of CRE in SH2D2A expression in primary T cells Having found that the putative CRE in the proximal SH2D2A pro- To test whether the putative CRE at position Ϫ117 was important moter is crucial for transcription of the SH2D2A reporter gene for the expression of SH2D2A in primary T cells, we introduced construct, we examined whether expression of the SH2D2A gene the same mutation as in mutant D into the GA23-GL3 and 240 could be induced by cAMP analogues. Primary resting T cells Kp-GL3 constructs. These two mutants, their corresponding wild- were incubated with increasing concentrations 8-CPT-cAMP, and type reporter genes, and the 110Kp-GL3 reporter gene were indi- the amount of SH2D2A mRNA expression relative to ZAP-70 vidually transfected into primary human T cells that were left un- mRNA expression was evaluated using quantitative PCR. The stimulated or stimulated by CD3 ligation for 16 h. We found that cAMP analog induced a high level of SH2D2A expression at phys- reporter gene activity in CD3-stimulated primary T cells trans- iological concentrations (100 ␮M for 8-CPT) compared with that fected with the two mutated constructs was markedly reduced obtained in T cells stimulated with anti-CD3 magnetic beads (Fig. compared with that in the wild-type constructs (Fig. 4A), indicat- 6A). cAMP mediates much of its effects through activation of protein ing that the CRE is important for transcription of the SH2D2A kinase A (PKA). The synthetic kinase inhibitor H89 will inhibit PKA gene. In an additional series of experiments, stimulation of trans- at low concentrations. To determine whether cAMP-mediated activa- fected primary T cells with a synthetic cAMP analog (8-CPT- tion of SH2D2A mRNA expression is mediated via PKA or another cAMP) induced expression of the GA23-GL3 and the 240 Kp-GL3 kinase that is inhibited by H89, we incubated primary T cells for 20 luciferase reporter constructs, whereas expression of the corre- min in medium containing 10 ␮M H89 before stimulation with 100 sponding mutated constructs was induced to a much lesser extent ␮M 8-CPT. Under these conditions, H89 completely blocked the (Fig. 4B). Taken together these data indicate that the putative CRE cAMP-mediated induction of SH2D2A mRNA expression (Fig. 6B). 6148 TRANSCRIPTIONAL REGULATION OF THE SH2D2A GENE Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 3. Identification of nuclear proteins binding to the minimal SH2D2A promoter. A, EMSA with a radiolabeled PCR fragment, 240-R, covering the minimal SH2D2A promoter (Ϫ246 to Ϫ47). Binding activity was observed in the presence of nuclear extract from resting Jurkat T cells (NE) and was competed by a 100-fold excess of unlabeled PCR fragment, but not by an irrelevant PCR product from the coding part of the SH2D2A gene (irc). B, Schematic representation of the region Ϫ240 to the first coding ATG, showing localization of the primers (f; see also Table I) used to generate radiolabeled PCR fragments for EMSA. C, EMSA with radiolabeled PCR products depicted in B as probes. Binding complexes retarded to a similar extent were seen with the probes 6-3, 5-3, and 4-1. Retarded complexes may also be present with probes 6-4, 2-R, and 1-R. D, EMSA competition experiments with unlabeled PCR product 4-3 as a competitor for the complex binding to probes 6-3, 5-3, 4-1, as well as 4-3. The minimal 4-3 probe competed for binding to the four tested probes, indicating that the nuclear binding activity displayed by these four probes is the same complex. E, The sequence of the minimal oligonu- cleotide 4-3. The introduced mutations AÐF in the minimal oligonucleotide are shown in bold. Possible binding sites for NFAT and the transcription factor Sp1 as well as the putative CRE are indicated directly on the sequence. Their corresponding consensus sites with the mutated bases in the AÐF mutants of the SH2D2A promoter sequence indicated are shown below. F, EMSA competition experiments with radiolabeled PCR probe 4-3 and corresponding unlabeled PCR probes mutated as shown in E as competitors. The PCR probes harboring the D mutation did not compete with the wild-type probe for binding. G, EMSA competition and supershifting experiment with radiolabeled PCR probe 4-3. Commercially available consensus CRE oligonucleotide (CRE-SC or cre in the figure) failed to compete out binding to the 4-3 probe. None of the four Abs against CRE binding transcription factors (Ab 1, anti-CREB1; Ab 2, anti-ATF1; Ab 3, anti-ATF2; Ab 4, anti-ATF4) supershifted or inhibited binding to the 4-3 probe.

When the TCR/CD3 complex is triggered, the level of intracellular mRNA expression (Fig. 6C). Taken together, our data indicate that cAMP increases in the T cell (30). To examine whether induction of the CD3-mediated signal inducing SH2D2A mRNA expression is at mRNA expression of TSAd after stimulation of T cells is dependent least partly mediated by a cAMP-responsive kinase and by NF bind- on cAMP signals, we preincubated primary resting T cells with H89 ing at a putative CRE at Ϫ117 in the SH2D2A promoter. at different concentrations for 45 min before CD3 stimulation. H-89 treatment of primary T cells reduced SH2D2A mRNA expression in- Discussion duction from 35- to 4-fold in anti-CD3-stimulated cells at the lowest In this study we demonstrated that the proximal 1-kb promoter of concentration of H89 used (0.4 ␮M), whereas higher concentrations the SH2D2A gene harbors information for tissue-specific expres- of H89 completely abolished CD3-mediated induction of SH2D2A sion and up-regulation by TCR activation. The minimal promoter The Journal of Immunology 6149 Downloaded from http://www.jimmunol.org/

FIGURE 5. Resting T cells contain only minimal nuclear binding ac- by guest on September 29, 2021 tivity to the CRE. A, EMSA using radiolabeled 240-R probe representing the minimal SH2D2A promoter (Ϫ246 to Ϫ47). Binding activity was ob- served in the presence of nuclear extract from primary T cells activated for 2 h with PHA and was competed by a 100-fold excess of unlabeled PCR FIGURE 4. The D mutation abolishes SH2D2A promoter activity in product or unlabeled 4-3 probe, but not by an unlabeled irrelevant PCR primary T cells. A, Luciferase activity was assessed in lysates of primary product from the coding part of the SH2D2A gene (irc). No binding activity T cells transiently transfected with the indicated promoter constructs and was observed in the presence of nuclear extract from the corresponding TK-RL2 as an internal control vector. Cells were unstimulated or stimu- unstimulated primary T cells. B, EMSA using radiolabeled 4-3 probe con- lated with anti-CD3 Ab coupled to magnetic beads for 16 h. The normal- taining the putative CRE. Binding activity was observed in the presence of ized luciferase activity in each set of experiments was related to the activity nuclear extract from primary T cells activated for 2 h with PHA and was obtained for the unstimulated cells transfected with the GA construct, 23 competed by a 100-fold excess of unlabeled 4-3 probe. Only weak binding which was set at 1. Results shown are the mean of five experiments Ϯ SD. activity was observed in the presence of nuclear extract from the corre- B, The experiment was performed as described above. Cells were unstimu- sponding unstimulated primary T cells. lated or stimulated with 100 ␮M 8-CPT or anti-CD3 Ab on magnetic beads for 16 h. Results shown are the mean of three experiments Ϯ SD.

CRE at Ϫ117 was found in nuclear extracts of Jurkat cells inde- of the SH2D2A gene is located between Ϫ236 and Ϫ93 bp up- pendent of prior stimulation of the cells. Binding activity to the stream of the first coding ATG, and contains a CRE at position CRE at Ϫ117 was also observed in nuclear extracts from activated Ϫ117 that is essential for transcriptional activity of the SH2D2A primary T cells, but not at all or only weakly in nuclear extracts promoter as well as for binding of nuclear factors to the SH2D2A from resting primary T cells, which do not express SH2D2A promoter. Consistent with this, SH2D2A mRNA expression was mRNA. Thus, we favor the idea the transcriptional activity of the found to be induced by synthetic cAMP analogues, and CD3 in- SH2D2A promoter in resting Jurkat cells is due to the partially duced SH2D2A mRNA expression could be abolished by inhibi- activated status of these cells. tion of cAMP responsive kinase activity. We identified the minimal promoter of the SH2D2A gene to be The initial characterization of the SH2D2A promoter was per- between Ϫ236 and Ϫ93 bp 5Ј of the first coding ATG. The formed with the Jurkat TAg T cell line. In this cell line the SH2D2A promoter contains no TATA box or other well-known SH2D2A promoter was transcriptionally active both in nonstimu- regulatory elements necessary for initiation of transcription (19). lated and CD3-stimulated cells. This could be due to lack of the Thus, there must be other elements in this minimal promoter that appropriate regulatory sequences in the promoter fragment studied. direct transcriptional initiation of the gene. However, it could also be explained by the fact that Jurkat T cells In TATA less promoters, an initiator element may compensate are to some extent already activated (31). Binding activity to the for the lacking TATA box to direct transcriptional initiation (32). 6150 TRANSCRIPTIONAL REGULATION OF THE SH2D2A GENE

However, in promoters lacking both TATA and an initiator ele- ment, GC-rich sequences have been shown to provide Sp1 sites that may be sufficient for transcription initiation (33, 34) In such promot- ers, however, there is not a precise start site for transcription (35). A palindromic sequence in the SH2D2A sequence located between po- sitions Ϫ110 and Ϫ93 in the promoter contains two putative over- lapping Sp1 sites. The removal of one or both of these Sp1 sites did not adversely influence the gene expression, suggesting that an as yet unidentified element directs transcriptional initiation of the SH2D2A gene. Moreover, in EMSA experiments, no binding activity to an oligonucleotide containing the palindrome was observed. Thus, we failed to demonstrate participation of this palindrome in transcription initiation of the SH2D2A gene. In EMSA experiments we observed nuclear binding activity to a putative CRE centered on position Ϫ117 upstream of the first ATG. Consistent with the involvement of this CRE, SH2D2A gene expression in primary T cells was rapidly induced by cAMP ana- logues. Moreover, the kinase inhibitor H89 inhibited cAMP as well as

CD3-induced SH2D2A mRNA expression. H89 is a well-known in- Downloaded from hibitor of cAMP-activated kinase (PKA), but also inhibits a number of other kinases (36). Thus, the role of PKA in the regulation of SH2D2A gene expression needs to be further elucidated. Taken together, our present observations fit well with our earlier observations that T cells in the presence of monocytes exposed to

plastic display high levels of SH2D2A mRNA expression (12). http://www.jimmunol.org/ Monocytes exposed to plastic will rapidly adhere, become acti- vated, and secrete mediators such as PGs, which, in turn, may induce cAMP production in the T cells. We were not able to demonstrate the identity of the CRE bind- ing activity in the nuclear extracts from Jurkat T cells. Abs against four different CRE binding transcription factors failed to inhibit or supershift the binding activity in EMSA, nor did two different CRE consensus primers with different flanking sequences compete out the binding activity. The core CRE consensus sequence is the pal- by guest on September 29, 2021 indromic octamer TGACGTCA. The CRE found in the SH2D2A promoter contains only five of these eight bases: aGgCGaCA. Thus, it is possible that the complex binding to the SH2D2A CRE has a different binding specificity from proteins binding to the consensus CRE. Further work to demonstrate the identity of the protein(s) binding to the SH2D2A minimal promoter is therefore clearly needed. In conclusion, we have demonstrated that the tissue-specific ex- pression of the SH2D2A gene is in part determined by the proximal 1-kb SH2D2A promoter. Moreover, we have identified a putative CRE in the minimal promoter of the SH2D2A gene that was crucial for TCR/CD3-mediated activation of SH2D2A-derived reporter genes in primary T cells. An important role of cAMP in the reg- ulation of this gene was further demonstrated by activation of the endogenous gene in primary T cells by cAMP analogues and in- FIGURE 6. The cAMP analogues induce SH2D2A mRNA expression. hibition of cAMP or CD3-induced expression by an inhibitor of A, Primary T cells were incubated with the indicated concentrations of cAMP-activated kinases. Thus, transcriptional regulation of the 8-CPT-cAMP, and SH2D2A (i.e., TSAd) mRNA expression was assessed SH2D2A gene in activated T cells is critically dependent on a CRE using quantitative PCR. Results shown are the level of TSAd mRNA ex- in the proximal promoter region. pression relative to ZAP70 mRNA expression normalized to the level ob- served in resting T cells. Results shown are mean of triplicate determina- Acknowledgments tions from one representative experiment. B, Primary T cells were or were not preincubated in medium containing 10 ␮M H89 for 20 min before The technical assistance of Lise Berven is acknowledged. stimulation with 100 ␮M 8-CPT-cAMP. TSAd mRNA expression was measured and shown as described above. Results shown are the mean of References duplicate determinations from one representative experiment. C, Primary T 1. Leo, A., and B. Schraven. 2001. Adapters in lymphocyte signalling. Curr. Opin. cells were preincubated for 45 min with the indicated concentration of H89 Immunol. 13:307. and then stimulated with anti-CD3 magnetic beads (two beads per cell). 2. Rieux-Laucat, F., F. Le Deist, and A. Fischer. 2003. Autoimmune lymphopro- liferative syndromes: genetic defects of apoptosis pathways. Cell Death. Differ. TSAd mRNA expression was measured and shown as described above. 10:124. Results shown are mean of triplicate determinations from one representa- 3. Bachmaier, K., C. Krawczyk, I. Kozieradzki, Y. Y. Kong, T. Sasaki, tive experiment. A. Oliveira-dos-Santos, S. Mariathasan, D. Bouchard, A. Wakeham, A. Itie, et al. The Journal of Immunology 6151

2000. Negative regulation of lymphocyte activation and autoimmunity by the 20. Carey, M., and S. T. Smale. 2000. Modes of regulating mRNA abundance. In molecular adaptor Cbl-b. Nature 403:211. Transcriptional Regulation in Eukaryotes. Concepts, Strategies and Techniques. 4. Perlmutter, R. M., J. D. Marth, D. B. Lewis, R. Peet, S. F. Ziegler, and M. Carey and S. T. Smale, eds. Cold Spring Harbor Laboratory Press, Cold C. B. Wilson. 1988. Structure and expression of lck transcripts in human lym- Spring Harbor, pp. 65Ð96. phoid cells. J. Cell Biochem. 38:117. 21. Michael, M. D., L. F. Michael, and E. R. Simpson. 1997. A CRE-like sequence 5. Chan, A. C., M. Iwashima, C. W. Turck, and A. Weiss. 1992. ZAP-70: a 70 kd that binds CREB and contributes to cAMP-dependent regulation of the proximal protein-tyrosine kinase that associates with the TCR ␨ chain. Cell 71:649. promoter of the human aromatase P450 (CYP19) gene. Mol. Cell. Endocrinol. 6. Bruyns, E., A. Marie-Cardine, H. Kirchgessner, K. Sagolla, A. Shevchenko, 134:147. M. Mann, F. Autschbach, A. Bensussan, S. Meuer, and B. Schraven. 1998. T cell 22. Heinemeyer, T., E. Wingender, I. Reuter, H. Hermjakob, A. E. Kel, O. V. Kel, receptor (TCR) interacting molecule (TRIM), a novel disulfide-linked dimer as- E. V. Ignatieva, E. A. Ananko, O. A. Podkolodnaya, F. A. Kolpakov, et al. 1998. sociated with the TCR-CD3-␨ complex, recruits intracellular signaling proteins to Databases on transcriptional regulation: TRANSFAC, TRRD and COMPEL. Nu- the plasma membrane. J. Exp. Med. 188:561. cleic Acids Res. 26:362. 7. Marie-Cardine, A., E. Bruyns, C. Eckerskorn, H. Kirchgessner, S. C. Meuer, and 23. Quandt, K., K. Frech, H. Karas, E. Wingender, and T. Werner. 1995. MatInd and B. Schraven. 1997. Molecular cloning of SKAP55, a novel protein that associates MatInspector: new fast and versatile tools for detection of consensus matches in fyn with the protein tyrosine kinase p59 in human T-lymphocytes. J. Biol. Chem. nucleotide sequence data. Nucleic Acids Res. 23:4878. 272:16077. 24. Philipsen, S., and G. Suske. 1999. A tale of three fingers: the family of mam- 8. Freeman, G. J., D. B. Lombard, C. D. Gimmi, S. A. Brod, K. Lee, J. C. Laning, malian Sp/XKLF transcription factors. Nucleic Acids Res. 27:2991. D. A. Hafler, M. E. Dorf, G. S. Gray, and H. Reiser. 1992. CTLA-4 and CD28 25. Hsieh, C. L., R. Sturm, W. Herr, and U. Francke. 1990. The gene for the ubiq- mRNA are coexpressed in most T cells after activation: expression of CTLA-4 uitous octamer-binding protein Oct-1 is on human , region cen- and CD28 mRNA does not correlate with the pattern of lymphokine production. q32, and near Ly-22 and Ltw-4 on mouse chromosome 1. Genomics 6:666. J. Immunol. 149:3795. 26. Serfling, E., F. Berberich-Siebelt, S. Chuvpilo, E. Jankevics, S. Klein-Hessling, 9. Egen, J. G., M. S. Kuhns, and J. P. Allison. 2002. CTLA-4: new insights into its T. Twardzik, and A. Avots. 2000. The role of NF-AT transcription factors in T biological function and use in tumor immunotherapy. Nat. Immunol. 3:611. cell activation and differentiation. Biochim. Biophys. Acta 1498:1. 10. Kristiansen, O. P., Z. M. Larsen, and F. Pociot. 2000. CTLA-4 in autoimmune 27. Amtoft, H. W., A. B. Sorensen, C. Bareil, J. Schmidt, A. Luz, and F. S. Pedersen. diseases: a general susceptibility gene to autoimmunity? Genes Immun. 1:170.

1997. Stability of AML1 (core) site enhancer mutations in T lymphomas induced Downloaded from 11. Sundvold, V., K. M. Torgersen, N. H. Post, F. Marti, P. D. King, J. A. Rottingen, by attenuated SL3Ð3 murine leukemia virus mutants. J. Virol. 71:5080. A. Spurkland, and T. Lea. 2000. Cutting edge: T cell-specific adapter protein 28. Zhang, W., J. W. Nisbet, B. Albrecht, W. Ding, F. Kashanchi, J. T. Bartoe, and inhibits T cell activation by modulating Lck activity. J. Immunol. 165: 2927. M. D. Lairmore. 2001. Human T-lymphotropic virus type 1 p30(II) regulates 12. Spurkland, A., J. E. Brinchmann, G. Markussen, F. Pedeutour, E. Munthe, T. Lea, gene transcription by binding CREB binding protein/p300. J. Virol. 75:9885. F. Vartdal, and H. C. Aasheim. 1998. Molecular cloning of a T cell-specific adapter protein (TSAd) containing an Src homology (SH) 2 domain and putative 29. Hai, T., and M. G. Hartman. 2001. The molecular biology and nomenclature of SH3 and phosphotyrosine binding sites. J. Biol. Chem. 273:4539. the activating transcription factor/cAMP responsive element binding family of 13. Marti, F., N. H. Post, E. Chan, and P. D. King. 2001. A transcription function for transcription factors: activating transcription factor proteins and homeostasis. Gene 273:1.

the T cell-specific adapter (TSAd) protein in T cells: critical role of the TSAd Src http://www.jimmunol.org/ homology 2 domain. J. Exp. Med. 193:1425. 30. Ledbetter, J. A., M. Parsons, P. J. Martin, J. A. Hansen, P. S. Rabinovitch, and 14. Rajagopal, K., C. L. Sommers, D. C. Decker, E. O. Mitchell, U. Korthauer, C. H. June. 1986. Antibody binding to CD5 (Tp67) and Tp44 T cell surface A. I. Sperling, C. A. Kozak, P. E. Love, and J. A. Bluestone. 1999. RIBP, a novel molecules: effects on cyclic nucleotides, cytoplasmic free calcium, and cAMP- Rlk/Txk- and itk-binding adaptor protein that regulates T cell activation. J. Exp. mediated suppression. J. Immunol. 137:3299. Med. 190:1657. 31. Astoul, E., C. Edmunds, D. A. Cantrell, and S. G. Ward. 2001. PI 3-K and T-cell 15. Sun, W., K. Kesavan, B. C. Schaefer, T. P. Garrington, M. Ware, N. L. Johnson, activation: limitations of T-leukemic cell lines as signaling models. Trends Im- E. W. Gelfand, and G. L. Johnson. 2001. MEKK2 associates with the adapter munol. 22:490. protein Lad/RIBP and regulates the MEK5-BMK1/ERK5 pathway. J. Biol. 32. Zenzie-Gregory, B., A. O’Shea-Greenfield, and S. T. Smale. 1992. Similar mech- Chem. 276:5093. anisms for transcription initiation mediated through a TATA box or an initiator 16. Dai, K. Z., H. F. Harbo, E. G. Celius, A. Oturai, P. S. Sorensen, L. P. Ryder, element. J. Biol. Chem. 267:2823. P. Datta, A. Svejgaard, J. Hillert, S. Fredrikson, et al. 2001. The T cell regulator 33. Emami, K. H., T. W. Burke, and S. T. Smale. 1998. Sp1 activation of a TATA-

gene SH2D2A contributes to the genetic susceptibility of multiple sclerosis. less promoter requires a species-specific interaction involving transcription factor by guest on September 29, 2021 Genes Immun. 2:263. IID. Nucleic Acids Res. 26:839. 17. Drappa, J., L. A. Kamen, E. Chan, M. Georgiev, D. Ashany, F. Marti, and 34. Soravia, E., A. Grebe, P. De Luca, K. Helin, T. T. Suh, J. L. Degen, and F. Blasi. P. D. King. 2003. Impaired T cell death and lupus-like autoimmunity in T cell- 1995. A conserved TATA-less proximal promoter drives basal transcription from specific adapter protein-deficient mice. J. Exp. Med. 198:809. the urokinase-type plasminogen activator receptor gene. Blood 86:624. 18. Rasmussen, A. M., E. B. Smeland, B. K. Erikstein, L. Caignault, and S. Funderud. 35. Lin, Y., T. A. Ince, and K. W. Scotto. 2001. Optimization of a versatile in vitro 1992. A new method for detachment of Dynabeads from positively selected B lym- transcription assay for the expression of multiple start site TATA-less promoters. phocytes. J. Immunol. Methods 146:195. Biochemistry 40:12959. 19. Dai, K. Z., G. Vergnaud, A. Ando, H. Inoko, and A. Spurkland. 2000. The 36. Davies, S. P., H. Reddy, M. Caivano, and P. Cohen. 2000. Specificity and mech- SH2D2A gene encoding the T cell specific adapter protein (TSAd) is localized anism of action of some commonly used protein kinase inhibitors. Biochem. J. centromeric to the CD1 gene cluster on chromosome 1. Immunogenetics 51:179. 351:95.