The HTLV-I Tax Protein Transcriptionally Modulates OX40 Antigen Expression Rüdiger Pankow, Horst Dürkop, Ute Latza, Hans Krause, Ulrich Kunzendorf, Thomas Pohl and Silvia Bulfone-Paus This information is current as of October 2, 2021. J Immunol 2000; 165:263-270; ; doi: 10.4049/jimmunol.165.1.263 http://www.jimmunol.org/content/165/1/263 Downloaded from References This article cites 48 articles, 23 of which you can access for free at: http://www.jimmunol.org/content/165/1/263.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 October 2, 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 © 2000 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The HTLV-I Tax Protein Transcriptionally Modulates OX40 Antigen Expression1
Ru¨diger Pankow,* Horst Du¨rkop,† Ute Latza,†‡ Hans Krause,§ Ulrich Kunzendorf,¶ Thomas Pohl,* and Silvia Bulfone-Paus2*
OX40 is a member of the TNF receptor family, expressed on activated T cells. It is the only costimulatory T cell molecule known to be specifically up-regulated in human T cell leukemia virus type-I (HTLV-I)-producing cells. In a T cell line, OX40 surface expression was shown to be induced by HTLV-I Tax alone. To understand molecular mechanisms of OX40 gene regulation and modulation by HTLV-I Tax, we have cloned the human OX40 gene and analyzed its 5-flanking region. By reporter gene analysis with progressive 5 deletions from nucleotides ؊1259 to ؊64, we have defined a 157-bp DNA fragment as a minimal promoter for constitutive expression. In addition, we show that in the OX40؉ cell line, Co, Tax is able to further increase OX40 surface expression. Up-regulation of OX40 promoter activity by Tax requires two upstream NF-B sites, which are not active in the Downloaded from constitutive OX40 expression. Their deletion abrogates Tax responsiveness in reporter gene analysis. The site-directed mutagenesis of each NF-B site demonstrates that cooperative NF-B binding is a prerequisite for Tax-directed activity as neither site alone is sufficient for a full Tax responsiveness of the OX40 promoter. Upon Tax expression, both sites bind p65 and c-Rel. These data provide new insight into the direct regulation of OX40 by Tax and add to our understanding of the possible role of the OX40/OX40 .ligand system in the proliferation of HTLV-I؉ T cells. The Journal of Immunology, 2000, 165: 263–270
he T cell costimulatory receptor OX40 and its ligand genes, most of which are associated with cell growth. They include http://www.jimmunol.org/ (OX40L)3 are both members of the superfamilies of TNF growth factors, like GM-CSF (16) or cytokines and their receptors T receptors and ligands, respectively (1–4). Interactions of such as IL-2 (17), IL-15 (18), or IL-2R␣ (CD25) (19–21). This ac- these members with their ligands have been shown to be involved tivation occurs via the interaction of Tax with cellular transcription in cell proliferation, activation, as well as induction of apoptosis. factors, which include the serum response factor (22, 23), members of However, unlike the other known TNF receptor family members, the activating transcription factor/cAMP response element binding OX40 has a restricted cellular expression only on activated lym- protein (24, 25), and NF-B (26). phocytes, predominantly CD4ϩ T cells (3, 5, 6). Furthermore, it NF-B defines a family of transcription factors that result from
has thus far not been associated with a cell death-inducing func- the combination of its protein members, such as p50, p65 (RelA), by guest on October 2, 2021 tion. Rather, signaling through OX40 generates strong costimula- c-Rel, and RelB (27). Inactive NF-B is retained in the cytoplasm, tory effects, which induce T cell proliferation (7, 8), modulate either by complexing to the IB inhibitor or as a precursor mol- cytokine production (9), and influence T cell migration into tissues ecule of the active protein (27). NF-B activation and subsequent (10, 11). Both OX40 and OX40L are constitutively expressed on translocation to the nucleus requires phosphorylation and degra- human T cell leukemia virus type I (HTLV-I)-producing T cell dation of IB or proteolytic cleavage of the precursor molecule. lines (3, 10–12). Tax is one of the signals known to cause liberation of NF-B from HTLV-I infection is associated with the aggressive and lethal adult IB (26, 28). It can also bind NF-B precursors as well as active T cell leukemia (ATL), as well as chronic inflammatory disorders, nuclear NF-B proteins (23, 29–31). such as the tropical spastic paraparesis/HTLV-I-associated myelopa- The OX40L promoter was shown to be activated by the Tax thy and others (13–15). T cell transformation requires the HTLV-I oncoprotein (32) and expression of Tax in the OX40-negative T Tax oncoprotein, which is the activator of viral gene expression. Tax cell line, Jurkat, results in OX40 surface expression (33). These also acts as a transactivator of an increasing number of host cellular findings suggest that auto or paracrine OX40/OX40L interactions on HTLV-I-producing cells may provide necessary costimulatory signals for transformation or survival and proliferation of these *Institute of Immunology, Departments of †Pathology and §Urology, University Hos- pital Benjamin Franklin, Free University Berlin, Berlin, Germany; ‡Berufsgenossen- cells. Thus far, however, little is known about the transcriptional schaftliches Forschungsinstitut fu¨r Arbeitsmedizin, Bochum, Germany; and ¶Depart- regulation of OX40 gene expression and its modulation by the Tax ment of Internal Medicine IV, Friedrich-Alexander-University, Erlangen, Germany protein. Therefore, we have cloned the human OX40 gene, ana- Received for publication December 22, 1999. Accepted for publication April lyzed its promoter region, and defined basal promoter activity. 11, 2000. Moreover, we describe here that Tax further up-regulates OX40 The costs of publication of this article were defrayed in part by the payment of page ϩ charges. This article must therefore be hereby marked advertisement in accordance gene expression in an OX40 cell line via two NF- B elements, with 18 U.S.C. Section 1734 solely to indicate this fact. which recruit at least three members of the NF-B family: c-Rel, 1 This study was supported in part by grants from the Deutsche Forschungsgemein- p65, and p50 in a Tax-dependent manner. schaft (SFB 506 to S.B.P., T.P., and H.D.; SFB 366 to H.D.) and from the Deutsche Krebshilfe (W76-93-Du¨1 to H.D.). Materials and Methods 2 Address correspondence and reprint requests to Dr. Silvia Bulfone-Paus, Institute of Isolation and characterization of human OX40 genomic clones Immunology, University Hospital Benjamin Franklin, Free University Berlin, Hin- denburgdamm 30, D-12200 Berlin, Germany. E-mail: [email protected] The human placental cosmid library pWE15 (Stratagene, Heidelberg, Ger- 3 Abbreviations used in this paper: OX40L, OX40 ligand; ATL, adult T cell leukemia; many) was screened by standard molecular biology techniques using ran- HTLV-I, human T cell leukemia virus type I; Inr, initiator element. dom-primed radio-labeled human OX40 cDNA (3). The cosmid carrying
Copyright © 2000 by The American Association of Immunologists 0022-1767/00/$02.00 264 Tax TRANSACTIVATES OX40 THROUGH NF-B
the OX40 gene was sequenced from both strands of sample DNAs on an of each plasmid DNA. Cell extract preparation and luciferase assay were automated sequencer (Perkin-Elmer/ABI, Weiterstadt, Germany). The se- performed 48 h later according to the Luciferase assay system kit (Pro- quence was searched for possible transcription factor binding sites using mega). Enzyme activity was measured with a Lumat LB9501 (Berthold, the programs FACTOR (HUSAR, EMBL, Heidelberg, Germany) and Mat- Wildbad, Germany). Transfection efficiency was normalized by cotrans- Inspector (34). fection of a reporter plasmid, pCH101 (Pharmacia Biotech, Freiburg, Ger- many), containing the -galactosidase gene under the control of the SV40 Cell lines and cell culture promoter. -galactosidase activity was determined using the Galacto-light Ϫ kit (Tropix, Bedford, MA), following manufacturer’s instructions. For The OX40 lymphoid cell lines L363 and DG75 were obtained from the comparison of promoter activities in different cell lines, luciferase activities Deutsche Sammlung fu¨r Mikroorganismen und Zellkulturen (Braun- ϩ were referred to luciferase control vector, pGL2 control, containing the schweig, Germany). The OX40 but HTLV-I-negative lymphoid cell line SV40 promoter and enhancer. All measurements were conducted in dupli- Co (35) was a generous gift from Dr. D. B. Jones (Southhampton, U.K.). cates from three independent transfections. The OX40ϩ but HTLV-I-negative lymphoid cell line CCRF-CEM and the OX40ϩ and HTLV-I-positive lymphoid cell lines HUT-102 and MT-2 were obtained from the American Type Culture Collection (Manassas, Nuclear extracts and gel mobility shift assay VA). All cell lines were cultured in RPMI 1640, containing 10% FCS, 2 Nuclear extracts were prepared as previously described (37). Five micro- mM L-glutamine, 100 U/ml penicillin, and 100 g/ml streptomycin and grams of extract were incubated with 1 ng of labeled DNA (20,000 cpm) incubated at 37°C and 5% CO2. in a buffer containing 20 mM HEPES (pH 7.9), 60 mM KCl, 0.5 mM dithiotreitol, 4% Ficoll or 1% polyvinylalcohol, 5% glycerol, and 2 g poly RNase protection assay (dI-dC) (Boehringer Mannheim) for 30 min at 4°C. The following oligo- Total RNA from OX40-positive cells was isolated using the RNA-Clean nucleotides with consensus sites underlined and mutations in lower case letters were used as probes: NF-B site 1, 5Ј-ACGGCTGGGAATTTCC isolation kit (AGS, Heidelberg, Germany). A PCR fragment carrying the Ј Ϫ Ϫ Ј sequence from position Ϫ251 to ϩ80 with respect to the transcription start CCACGGGTGGG-3 ( 1182/ 1156); NF- B site 2, 5 -TCCTGGGAG Ј Ϫ Ϫ Ј Downloaded from site was generated using 35 cycles of PCR (94°C, 1 min; 60°C, 2 min; GAGGGGATTTCCAGCCCC-3 ( 1128/ 1102); NF- Bmut, 5 -GATG Ј GATGAGTTGAGGcGACTTTCCCTCTTTACTG-3Ј. Oligonucleotides 72°C, 2 min, and 5 min final extension) with top-strand primer 5 -GCTC Ј Ј CATGGACCTTTTTGCTGAGGG-3Ј (Ϫ260 to Ϫ236) and bottom-strand carrying an Sp-1 site, Sp-1cons, 5 -ATTCGATCGGGGCGGGGCGAGC-3 primer 5Ј-GAGGAGCAGAGCCGCACACGGC-3Ј (ϩ59 to ϩ80) and the were purchased from Santa Cruz (Heidelberg, Germany). The reaction mix- cosmid containing the 5Ј region of the OX40 gene as a template. The tures were electrophoresed through a 4% polyacrylamide gel. In competi- tion experiments a 50-fold excess of unlabeled oligonucleotides, and in resulting fragment was cloned into the pAMP1 vector (Life Technologies, Karlsruhe, Germany). For the generation of radioactively labeled anti- supershift assays 2 g of NF- B p50 (sc-1191X), NF- B p65 (sc-372X), or NF-B c-Rel (sc-70-GX) Abs (Santa Cruz), were added to the binding sense-RNA, the construct was linearized with XbaI and in vitro transcribed http://www.jimmunol.org/ with the T7 RNA polymerase (Riboprobe In Vitro Transcription System; reaction before the addition of the labeled probe. Promega, Heidelberg, Germany). Then, 15 g RNA were used in each RNase protection assay according to manufacturer’s instructions (RNase Results Protection Kit; Boehringer Mannheim, Mannheim, Germany). The result- Analysis of the human OX40 gene ing fragment was resolved on a 6% polyacrylamide gel. To determine the genomic structure of the OX40 Ag, a cosmid Flow cytometry containing the entire OX40 gene was obtained. The complete cod- For FACS analysis, 5 ϫ 105 cells were incubated for 30 min on ice with ing region was found to be within only 2.7 kb and to include 7 the human FITC-conjugated OX40 Ab (Ber-Act35) (3) or with an isotype- exons and 6 introns (DDBJ/EMBL/GenBank Nucleotide Sequence
by guest on October 2, 2021 matched control Ab, mouse IgG1 , FITC-conjugated (Dako, Hamburg, Database accession number AJ277151). To identify the transcrip- Germany) in PBS with 10% FCS. After washing, the cells were analyzed by a FACSort (Becton Dickinson, Heidelberg, Germany). tion initiation site, we performed RNase protection assays with RNA isolated from OX40ϩ HUT-102 cells. We obtained a pro- DNA constructs tected band that mapped the start site to an adenosine 26 nt up- For the PCR-based generation of OX40 promoter deletion constructs, the pa- stream of the ATG initiation codon (Fig. 1A). The specificity of the rental cosmid containing the OX40 5Ј-flanking region was used as a template. transcription start site was confirmed using RNA of a different Top-strand primers were Ϫ1259, 5Ј-GAGGGgtACCAACTTAGGGACC-3Ј OX40ϩ cell line, MT-2 (data not shown). Ϫ Ϫ Ϫ Ј ( 1266 to 1245); 1184, 5 -GCCAGCGGtacCCACGGCTGGGAATT Fig. 1B depicts the 5Ј-flanking region and putative transcription TC-3Ј (Ϫ1195 to Ϫ1168); Ϫ1184 mut1, 5Ј-GCCAGCGGtacCCACGGCT GcGAATTTC-3Ј (Ϫ1195 to Ϫ1168); Ϫ1124 mut2, 5Ј-TTCCTGTGgTaC factor binding sites, based on sequence homology. Although an cGGGAGGAGccGATTTCC-3Ј (Ϫ1137 to Ϫ1108); Ϫ1021, 5Ј-CTGGTacCA initiator element (Inr) matching the consensus sequence Py Py Aϩ1 GTTATTTGAAAGG-3Ј (Ϫ1028 to Ϫ1007); Ϫ944, 5Ј-GGGCAGgTacCA N T/A Py Py (38) overlapped the start site for transcription, the TCCTCTGGGC-3Ј (Ϫ954 to Ϫ933); Ϫ649, 5Ј-ggtaCCACCTCTGCTTTG sequence analysis additionally revealed a TATA-like box at posi- Ј Ϫ Ϫ Ϫ Ј CAACTTC-3 ( 649 to 629); 426, 5 -CGGGAGgtACCGTGCCTG tion Ϫ27 as well as a CCAAT box. Several elements with homol- GCCATGGG-3Ј (Ϫ434 to Ϫ409); Ϫ138, 5Ј-GGTaCCCAGACCCC GCCTTTG-3Ј (Ϫ142 to Ϫ122); Ϫ64, 5Ј-GgTaCCCTCCTCCCCTCTCCC-3Ј ogy to Sp-1 binding sites, such as CCCCTCC (39) or GT box (Ϫ68 to Ϫ48); and the bottom-strand primer ϩ19, 5Ј-CTCGagTCTGCT sequences (40), were found throughout the 5Ј region. Furthermore, GTCGCCAGAG-3Ј (ϩ4toϩ25). Nucleotides that were exchanged to intro- we found two putative E boxes, as well as potential binding sites duce KpnI and XhoI restriction sites or to mutate the OX40 NF- B sites are for AP-1, AP-2, Yin Yang-1, and Icaros-2. Finally, at the very given in lower case letters. The annealing conditions for each PCR standard Ј reaction (as described above) were modified according to the melting temper- 5 -end of the analyzed region two NF- B binding sites were de- atures of the primers used. The site-directed mutagenesis of the NF-B site 2 tected (Fig. 1B). in construct Ϫ1184 mut2 was generated according to the protocol of the QuickChange kit (Stratagene) with the pGL2 basic Ϫ1184 reporter construct Functional analysis of the human OX40 promoter region as the template and a primer pair Ϫ1184 mut2 covering 5Ј-GCTCCTGGGAG GAGccGATTTCCAGCCCC-3Ј (Ϫ1130 to Ϫ1102) on both strands. Promoter First, to test the promoter activity of the 5Ј-flanking sequence of fragments were sequenced throughout the PCR-amplified region. The pSGtax, the OX40 gene, a fragment spanning the region between nucleo- coding for full-length HTLV-I Tax and pSGtaxmutant2, coding for a truncated tides Ϫ1259 to ϩ19 was cloned upstream of the luciferase gene of HTLV-I Tax of amino acids 1–284 (36) were a generous gift from Dr. Kazuo the pGL2 basic vector. This reporter construct was then transiently Sugamura (Sendai, Japan). transfected into OX40Ϫ DG75 and L363 cells, and into Co and Transfections and luciferase assays CCRF-CEM cells, which both constitutively express low levels of Transfections were performed using the DMRIE-C reagent protocol (Life OX40 as determined by FACS analysis (Fig. 2). We compared the Technologies). Briefly, for each transfection 5 ϫ 105 cells were pelleted, ability of the OX40 5Ј fragment to induce luciferase activity in resuspended in Opti-MEM media (Life Technologies) and mixed with 1 g these cell lines. The Journal of Immunology 265
FIGURE 1. A, Determination of the human OX40 transcription initiation site by RNase protection analysis. Total RNA from HUT-102 cells was used
for RNase protection assay, as described in Material and Methods (lane 1); control reactions included yeast tRNA treated with RNase (lane 2) or left Downloaded from untreated (lane 3); HinfI-digested ⌽X174 DNA served as a size marker (lane 4). B, Sequence of the human OX40 5Ј-upstream region. The transcription initiation site is marked as ϩ1. Sequences matching Inr, TATA, and CCAAT elements are boxed. Putative transcription factor binding sites are underlined.
Indeed, luciferase activities in Co and CCRF-CEM cells were activity compared with the Ϫ944 construct, which was the most Ͼ20 times higher than in OX40Ϫ DG75 and L363 cells (Fig. 3A). potent one and therefore set as 100%. The longest construct Ϫ1259 http://www.jimmunol.org/ The OX40 promoter in the OX40ϩ cells was three times less active displayed a Ͼ2.5-fold lower activity than the Ϫ944 construct. De- than the SV40 promoter and enhancer of the pGL2 control vector. leting the region to Ϫ1021 resulted in an increased promoter ac- This showed that the sequence up to Ϫ1259 serves as a promoter tivity being only 1.4-fold lower than the Ϫ944 construct. However, of OX40 gene expression in the OX40ϩ cell lines. as neither Co (see Fig. 5B) nor CCRF-CEM cells (data not shown) To determine cis-acting elements in the 5Ј-flanking region, exhibit nuclear NF-B activity the NF-B sites within this region which regulate OX40 gene expression, we transfected Co and do not play a role in maintaining the constitutive OX40 promoter CCRF-CEM cells with a series of luciferase reporter constructs. expression in these cells. Starting from position Ϫ1259, we progressively deleted the 5Ј- To further define the minimal promoter elements required for
end, to position Ϫ1021, Ϫ944, Ϫ649, Ϫ426, Ϫ138, and Ϫ64. As OX40 expression, other 5Ј deletion constructs were tested. Dele- by guest on October 2, 2021 shown in Fig. 3B, all constructs displayed a reduced luciferase tions from nucleotide Ϫ944 to Ϫ649 and to Ϫ426 resulted in 2- and 2.5-fold lower promoter activities, respectively (Fig. 3B). Re- markably, the promoter activity of the Ϫ138 construct was only 1.4-fold lower than the Ϫ944 construct. Indeed, this sequence lacks the CCAAT element but contains the TATA element and several Sp-1 binding sites, which were found to bind both Sp-1 and Sp-3 in EMSA (data not shown). A further 5Ј truncation to nucle- otide Ϫ64 that deleted the Sp-1 site located at position Ϫ118, diminished the promoter activity 5-fold compared with the Ϫ944 construct, but still maintaining it above background levels. Thus these data demonstrate that the region between Ϫ138 and ϩ19 can serve as a minimal promoter whose activity seems to be dependent on the number of Sp-1 sites present.
Tax modulates OX40 promoter activity To determine whether Tax is able to further up-regulate OX40 Ag expression, we transfected Co cells with Tax expression plasmids: wild-type Tax (pSGtax) or, as a negative control, a truncated Tax (pSGtaxmutant2) lacking amino acids 285–353, which was shown to have lost its transactivation ability (36, 41). FACS analysis dem- onstrated that OX40 expression could be up-regulated only upon wild-type Tax and not upon transfection with the pSGtaxmutant2 plasmid, as shown in Fig. 4A. Thus, functional Tax expression increases OX40 surface expression. To examine the role of the OX40 5Ј-flanking region in Tax- FIGURE 2. OX40 surface expression in Co and CCRF-CEM cells. induced OX40 up-regulation, we cotransfected Co cells with the Cells were stained with FITC-conjugated anti-human OX40 (Ber-Act35) Ϫ1259 OX40 promoter construct together with the pSGtax or with (bold curve) or with an isotype-matched control Ab (mouse IgG1-FITC) the pSGtaxmutant2 plasmids. Coexpression of the truncated Tax (dotted curve) and subjected to flow cytometric analysis. protein (pSGtaxmutant2) did not affect reporter gene expression in 266 Tax TRANSACTIVATES OX40 THROUGH NF-B Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021 FIGURE 3. A, OX40 promoter-directed reporter gene activity. Co and CCRF-CEM (OX40ϩ) and DG75 and L363 (OX40Ϫ) cell lines were transfected with 1 g of the pGL2 basic Ϫ1259 to ϩ19 OX40 construct. After a 48-h incubation, cells were harvested and assayed for luciferase activity. The luciferase activities were measured in duplicates, normalized to -galactosidase activity, and referred to the activity of the pGL2 control vector (set 100,000 random light units). The means of three independent experiments are shown with SDs given on top of each bar. B, 5Ј-deletion analysis of the OX40 promoter. Co (Ⅺ) and CCRF-CEM (f) cells were transfected with a series of pGL2 basic constructs containing 5Ј-terminus promoter deletions, as indicated on the abscissa. Luciferase activities were measured in duplicates and normalized to -galactosidase activity. The mean of three independent experiments is shown as a percentage of the activity of the Ϫ944 construct (100%) with SDs given on top of each bar.
any of the OX40 promoter constructs (Fig. 4B). However, cotrans- NF-B sites could influence Tax modulation, we generated con- fection of the Ϫ1259 construct with pSGtax was associated with a struct Ϫ1184 deleting the region between nucleotides Ϫ1259 and 6-fold increase over control in luciferase activity. Thus Tax ex- Ϫ1184. To this construct we then introduced point mutations ei- pression indeed induced OX40 promoter activity. ther to the NF-B site 1, Ϫ1184 mut1, or to the NF-B site 2, To localize the promoter region responsive to Tax, the 5Ј-end Ϫ1184 mut2. Furthermore, construct Ϫ1124 mut2 lacked the deletion constructs, Ϫ1021, Ϫ944, and Ϫ138, were employed for NF-B site 1 and contained a mutated NF-B site 2. These con- cotransfection. In fact, the deletion of the region between Ϫ1259 structs were transfected into Co cells in the presence or absence of and Ϫ1021, including two Sp-1 sites and NF-B sites 1 and 2, pSGtax cotransfection. In the absence of pSGtax cotransfection, all fully abolished Tax-induced promoter activity (Fig. 4B). In addi- constructs exhibited promoter activities equivalent to the Ϫ1021 tion, Tax expression did not alter the activity of the minimal pro- construct. As expected, point mutations in either NF-B site did moter as evidenced with the Ϫ138 construct. These results indicate not affect the constitutive promoter activity. This is consistent with that the minimal promoter of the OX40 gene is Tax independent our findings in EMSAs where nuclear NF-B activity is absent in and that the upstream region containing Sp-1 and two NF-B sites Co cells (Fig. 5B). Furthermore the point mutations within NF-B accounts for the promoter responsiveness upon Tax. site 2 mutated the overlapping Sp-1 site as well, with no effect on the resulting promoter activity. However, when we cotransfected Two NF- B sites are required for Tax inducibility pSGtax we detected a Ͼ5-fold increase of the activity in the nonmu- We wondered whether the two NF-B sites found within the tated Ϫ1184 construct. The respective point mutations of either Ϫ1259 and Ϫ1021 region might mediate OX40 promoter induc- NF-B site severely reduced the inducibility of luciferase activity by ibility upon Tax. Therefore, a series of reporter constructs carrying Tax to Ͻ2-fold (Fig. 5A). Finally, the deletion of NF-B site 1 and a mutations within either NF-B site was generated (Fig. 5A). To mutation of NF-Bsite2(Ϫ1124 mut2) completely abolished OX40 exclude that any relevant binding to the sequence upstream of both promoter inducibility upon Tax coexpression. Therefore, the NF-B The Journal of Immunology 267
were also detectable with nuclear extracts from untransfected Co cells, they were not considered to be of functional relevance for OX40 promoter induction upon Tax expression. Next, we wished to identify protein components of the NF-B complexes formed by nuclear extracts from Tax-expressing cells. We focused on p65 or c-Rel, because they were both shown to bind Tax and activate transcription (29) and because c-Rel is preferen- tially induced upon HTLV-I infection or in Tax-expressing cells (42, 43). Supershift experiments with extracts from HUT-102 cells were performed using Abs against the two NF-B protein family members, c-Rel and p65 (Fig. 5C). Indeed, NF-B complexes I and II, formed with both NF-B sites, were supershifted by anti-p65 and anti-c-Rel Abs, respec- tively (Fig. 5C). Supershifted bands were also observed upon ad- dition of anti-p50 Abs but did not result in any detectable disap- pearance of either complex I or II (Fig. 5C). However, because the simultaneous addition of anti-p65 and anti-c-Rel Abs did not fully supershift complexes I and II, one is tempted to speculate that p50 is contained within the binding complex. The Sp-1 associated com- plexes detected with NF-B site 2 (Fig. 5B, NF-B site 2) were Downloaded from weakly competed with unlabeled Sp-1 oligonucleotides, as they overlaid the anti-p50 supershift (Fig. 5C, NF-B site 2). Taken together, these results suggest that p65, c-Rel, and, to a lesser FIGURE 4. A, Modulation of OX40 surface expression upon transfec- tion with Tax. Co cells were transiently transfected with the pSGtax vector extent, p50-binding to the NF- B site 1 and site 2 are responsible (bold line), the pSGtaxmutant2 vector (dashed line), or left untreated (dot- for mediating Tax-induced OX40 promoter up-regulation. ted line). After a 48-h incubation, cells were harvested and assayed for http://www.jimmunol.org/ OX40 surface expression by staining with FITC-conjugated human OX40 (Ber-Act35) or an isotype-matched control Ab (mouse IgG1-FITC, not Discussion shown) in flow cytometric analysis. B, Sequence requirements for Tax- Signaling through OX40 provides a potent costimulatory function induced OX40 promoter activity. Co cells were transfected with OX40 for prolonging the life span of T cells. The OX40 Ag is also con- Ј Ⅺ 5 -deletion constructs as indicated, in the absence ( ) or presence of the stitutively expressed upon HTLV-I infection, thus possibly con- f pSGtax vector ( ) or its mutated derivative, the pGStaxmutant2 vector tributing to the survival and proliferation of the infected cells. In (u). Luciferase activities were measured in duplicates and normalized to view of the importance to understand the regulatory mechanisms -galactosidase activity. The mean of three independent experiments is that control OX40 expression, the cloning of the gene and the shown as a percentage of the activity of the Ϫ944 construct (100%) with by guest on October 2, 2021 SDs given on top of each bar. characterization of mechanisms involved in its promoter regulation described here contribute a substantial advance in OX40 biology. We define a 157-bp DNA fragment as a minimal promoter for sites, although not required for a constitutive OX40 expression, play constitutive expression with several Sp-1 binding sites modulating a central role in OX40 promoter regulation upon Tax expression. its activity. Furthermore, we show that Tax can enhance OX40 Furthermore, both sites are required for optimal Tax-induced pro- surface expression and that it can transcriptionally up-regulate the moter modulation. promoter activity. Tax-induced promoter modulation requires a 79-bp sequence located 1.1 kb upstream in the OX40 enhancer Tax induces NF- B binding to its motifs in the OX40 promoter region. The cooperative function of two distinct NF-B sites To determine whether NF-B complexes bind to NF-B site 1 and within this sequence is shown here to be the prerequisite to confer 2 upon Tax transactivation, we performed EMSAs using the two Tax responsiveness by binding nuclear p50, p65, and c-Rel. NF-B sites as synthetic probes. The binding activities of nuclear Examining the OX40 gene promoter, we have determined that extracts from HTLV-Iϩ untransfected HUT-102 and Co cells, tran- transcription could be initiated by an Inr element and a TATA box. siently transfected with the Tax-expressing plasmid pSGtax (Co Inr elements are often found in TATA-less promoters because they tax), were compared with the binding activities of nuclear extracts can replace the TATA box to direct the precise transcription start obtained from untransfected Co cells, which served as controls. (44). However, when Inr and a TATA box are present in the same As shown in Fig. 5B, nuclear extracts from either HUT-102 and core promoter with the TATA box 25–30 bp upstream of the Inr, Co tax cells formed two major NF-B complexes, defined as I and they can cooperate to enhance promoter activity, even if the TATA II, with NF-B site 1 and NF-B site 2. Because both complexes sequence only has a weak TATA box homology (38, 45). Such a were formed only with nuclear extracts from HUT-102 and Co weak TATA box-like element is indeed found at position Ϫ27. tax-transfected but not in the nontransfected extracts, this indicated Our functional studies revealed that the sequence between Ϫ138 that they were Tax induced. Binding specificity was shown by and ϩ19 displays a remarkable promoter activity to function as a complete inhibition of complex formation upon addition of the minimal promoter. Sp-1 binding sites (46) within this region to- unlabeled probe but not by mutated NF-B site oligonucleotides. gether with the TATA box element might control the strength of Additional binding complexes were observed with NF-Bsite2in the minimal promoter. all nuclear extracts (Fig. 5B). These could likely result from Sp-1 Previously, it has been reported that OX40 Ag expression is proteins binding to the NF-B site 2 probe, because an Sp-1 bind- induced by the viral transcriptional activator HTLV-I Tax (33). ing site overlaps the NF-B site on the complementary strand. In Here we show that Tax enhances OX40 expression in an OX40ϩ fact, Sp-1 binding could be specifically competed by Sp-1 consen- cell line. In fact, this suggests that Tax is not only capable to sus oligonucleotides (Fig. 5B). However, as these Sp-1 complexes induce de novo gene expression, but can also modify cellular Ag 268 Tax TRANSACTIVATES OX40 THROUGH NF-B Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021
FIGURE 5. A, Functional role of two NF-B sites for OX40 promoter inducibility by Tax. The pGL2 basic constructs carrying the OX40 Ϫ1184 to ϩ19 fragment with wild-type NF-B sites (Ϫ1184), a mutated NF-Bsite1(Ϫ1184 mut1, GGGAATTTCC 3 GGCAATTTCC), a mutated NF-B site 2 (Ϫ1184 mut2, AGGGATTTCC 3 AGCCATTTCC), or a construct spanning the Ϫ1124 to ϩ19 region and carrying the mutated NF-Bsite2(Ϫ1124 mut2) were transfected into Co cells in the absence or presence of cotransfected pSGtax. Luciferase activities of the NF-B/OX40 promoter constructs in the absence of cotransfected pSGtax vector are shown as a percentage of the activity of the Ϫ944 construct (100%), with the activity of the Ϫ1021 construct included (left diagram). The Tax-dependent increase of luciferase activity of the NF-B/OX40 promoter constructs upon cotransfected pSGtax is shown in the right diagram. Luciferase activities were determined in duplicates and normalized to -galactosidase activity. Shown is the mean of three independent experiments with SDs given on top of each bar. B, Binding of nuclear proteins to the OX40 NF-B sites. 32P-labeled oligonucleotides containing the NF-B site 1 or the NF-B site 2 were assayed for binding by Co, Co tax (Co cells, 48 h transiently transfected with pSGtax), and HUT-102 nuclear extracts. NF-B complexes are indicated as I and II and Sp-1 complexes by arrows. To assay binding specificity, unlabeled duplex oligonucleotides for NF-B site 1 or 2, mutated NF-B (NF-Bmut), or Sp-1 consensus (Sp-1cons) were used as competitors at a 50-fold molar excess. C, Supershift analysis of the NF-B family protein members binding to the OX40 NF-B sites. Supershift experiments were performed by preincubation of HUT-102 nuclear extracts with c-Rel, p65, or p50 Abs, before addition of the 32P-labeled NF-B probes. The respective supershifted bands are indicated by arrows. For a better resolution of the data, complexes binding to the NF-B site 2 probe were analyzed in the presence of a 50-fold molar excess of unlabeled duplex Sp-1 competitor. expression to meet the requirements for viral persistence. On the The NF-B site in the IL-2R␣ promoter, GGGAATCTCC, differs promoter regulatory level, this modified OX40 gene expression from the OX40 NF-B site 1 in one and from site 2 in only two upon Tax relies on the presence of two NF-B sites spanning a positions of the decameric motif. Like OX40, the IL-2R␣ NF-B region of 79 bp in the enhancer region. These sites are sufficient for site has also been shown to bind p65 and c-Rel upon Tax expres- conferring the Tax responsiveness of the whole OX40 5Ј-flanking sion, with c-Rel serving as the major protein responsible for Tax region, examined in Fig. 5A. They are dispensable in the consti- transactivation (19). Interestingly, one NF-B site in the IL-2R␣ tutive expression of the OX40 gene, as untransfected Co cells do promoter is sufficient to mediate Tax-directed transcriptional up- not exhibit any nuclear NF-B binding activity (Fig. 5B). regulation, whereas in the OX40 promoter two cooperatively act- Upon Tax expression, both OX40 NF-B sites bind the same set ing NF-B sites are required (Fig. 5A). However, two NF-B el- of NF-B complexes with equal strength (Fig. 5A). These com- ements, i.e., GGGAAATTCA and GGGAACTTCT, have also plexes consist of mainly p65 and c-Rel (Fig. 5B). Other cellular been described to be required for Tax transactivation in the OX40L promoters have also been shown to be up-regulated by Tax promoter. Upon Tax expression, no binding of c-Rel but rather through NF-B proteins, such as the IL-2R␣ (19) or OX40L (32). of RelB was detected on these sites (32). The respective NF-B The Journal of Immunology 269 motifs in the OX40 and OX40L promoters differ in at least two or leukemia patients mediates cell adhesion to vascular endothelial cells: implication three positions from each other. Thus, the exact NF-B motif in a for the possible involvement of OX40 in leukemic cell infiltration. Blood 89: 2951. given promoter may account for altered binding affinities for 12. Du¨rkop, H., U. Latza, P. Himmelreich, and H. Stein. 1995. Expression of the NF-B proteins, and how NF-B proteins regulate a cellular pro- human OX40 (hOX40) antigen in normal and neoplastic tissues. Br. J. Haematol. moter in response to Tax may depend on both the sequence of the 91:927. 13. Franchini, G. 1995. Molecular mechanisms of human T-cell leukemia/lympho- NF-B motif and the number of sites present. tropic virus type I infection. Blood 86:3619. The fact that OX40 is constitutively expressed on HTLV-I-ex- 14. Yamaguchi, K., and K. Takatsuki. 1993. Adult T cell leukaemia-lymphoma. Bail- pressing cell lines and can be detected in ATL cells could account lieres Clin. Haematol. 6:899. 15. Uchiyama, T. 1996. ATL and HTLV-I: in vivo cell growth of ATL cells. J. Clin. for some of the observations reported for ATL or its related dis- Immunol. 16:305. eases (14, 15). For example, OX40 expressed on leukemic cells 16. Himes, S. R., L. S. Coles, R. Katsikeros, R. K. Lang, and M. F. Shannon. 1993. HTLV-1 tax activation of the GM-CSF and G-CSF promoters requires the inter- isolated from peripheral blood of ATL patients was shown to me- action of NF-B with other transcription factor families. Oncogene 8:3189. diate cell adhesion to OX40L-expressing vascular endothelial cells 17. Ruben, S., H. Poteat, T. H. Tan, K. Kawakami, R. Roeder, W. Haseltine, and (11), suggesting the involvement of OX40 for leukemic cell mi- C. A. Rosen. 1988. Cellular transcription factors and regulation of IL-2 receptor gene expression by HTLV-I tax gene product. Science 241:89. gration into tissues. Studies on normal T cells have demonstrated 18. Azimi, N., K. Brown, R. N. Bamford, Y. Tagaya, U. Siebenlist, and that OX40 is a potent costimulatory molecule, which leads to a T. A. Waldmann. 1998. Human T cell lymphotropic virus type I Tax protein sustained proliferation of differentiated effector T cells (8). In ad- trans- activates interleukin 15 gene transcription through an NF-B site. Proc. Natl. Acad. Sci. USA 95:2452. dition, OX40 has not been shown so far to promote cell death. 19. Crenon, I., C. Beraud, P. Simard, J. Montagne, P. Veschambre, and P. Jalinot. Recently, reported data rather suggest that signaling through OX40 1993. The transcriptionally active factors mediating the effect of the HTLV-I Tax may serve to inhibit activation-induced T cell death (47). This transactivator on the IL-2R␣B enhancer include the product of the c-rel proto-
oncogene. Oncogene 8:867. Downloaded from could be important for supporting the survival of HTLV-I-express- 20. Ballard, D. W., E. Bohnlein, J. W. Lowenthal, Y. Wano, B. R. Franza, and ing T cells as Tax not only guarantees for constitutive OX40 ex- W. C. Greene. 1988. HTLV-I tax induces cellular proteins that activate the B pression, but also of its ligand, OX40L (4, 32). Therefore, auto or element in the IL-2 receptor ␣ gene. Science 241:1652. 21. Leung, K., and G. J. Nabel. 1988. HTLV-1 transactivator induces interleukin-2 paracrine OX40/OX40L interactions might contribute to the sur- receptor expression through an NF-B-like factor. Nature 333:776. vival, proliferation, and clonal expansion of ATL cells and may 22. Fujii, M., H. Tsuchiya, T. Chuhjo, T. Akizawa, and M. Seiki. 1992. Interaction substitute the IL-2/IL-2R pathway when the cell proliferation be- of HTLV-1 Tax1 with p67SRF causes the aberrant induction of cellular imme- diate early genes through CArG boxes. Genes Dev. 6:2066. http://www.jimmunol.org/ comes IL-2 independent (48). Thus, inhibition of OX40 expression 23. Suzuki, T., H. Hirai, J. Fujisawa, T. Fujita, and M. Yoshida. 1993. A trans- is a promising target for therapeutic intervention in ATL. activator Tax of human T-cell leukemia virus type 1 binds to NF-B p50 and serum response factor (SRF) and associates with enhancer DNAs of the NF-B site and CArG box. Oncogene 8:2391. 24. Suzuki, T., J. I. Fujisawa, M. Toita, and M. Yoshida. 1993. The trans-activator Acknowledgments tax of human T-cell leukemia virus type 1 (HTLV-1) interacts with cAMP-re- sponsive element (CRE) binding and CRE modulator proteins that bind to the We thank Gabriela Hein and Edda von der Wall for technical assistance 21-base-pair enhancer of HTLV-1. Proc. Natl. Acad. Sci. USA 90:610. and Dr. Ralf Paus for valuable editorial suggestions. 25. Zhao, L. J., and C. Z. Giam. 1992. Human T-cell lymphotropic virus type I (HTLV-I) transcriptional activator, Tax, enhances CREB binding to HTLV-I 21- base-pair repeats by protein-protein interaction. Proc. Natl. Acad. Sci. USA 89:
7070. by guest on October 2, 2021 References 26. Hirai, H., T. Suzuki, J. Fujisawa, J. Inoue, and M. Yoshida. 1994. Tax protein of human T-cell leukemia virus type I binds to the ankyrin motifs of inhibitory 1. Mallett, S., S. Fossum, and A. N. Barclay. 1990. Characterization of the MRC factor B and induces nuclear translocation of transcription factor NF-B pro- OX40 antigen of activated CD4 positive T lymphocytes—a molecule related to teins for transcriptional activation. Proc. Natl. Acad. Sci. USA 91:3584. nerve growth factor receptor. EMBO J. 9:1063. 27. Siebenlist, U., G. Franzoso, and K. Brown. 1994. Structure, regulation and func- 2. Calderhead, D. M., J. E. Buhlmann, A. J. van den Eertwegh, E. Claassen, tion of NF-B. Annu. Rev. Cell. Biol. 10:405. R. J. Noelle, and H. P. Fell. 1993. Cloning of mouse Ox40: a T cell activation 28. Yin, M. J., L. B. Christerson, Y. Yamamoto, Y. T. Kwak, S. Xu, F. Mercurio, marker that may mediate T-B cell interactions. J. Immunol. 151:5261. M. Barbosa, M. H. Cobb, and R. B. Gaynor. 1998. HTLV-I Tax protein binds to 3. Latza, U., H. Du¨rkop, S. Schnittger, J. Ringeling, F. Eitelbach, M. Hummel, MEKK1 to stimulate IB kinase activity and NF-B activation. Cell 93:875. C. Fonatsch, and H. Stein. 1994. The human OX40 homolog: cDNA structure, 29. Suzuki, T., H. Hirai, and M. Yoshida. 1994. Tax protein of HTLV-1 interacts expression and chromosomal assignment of the ACT35 antigen. Eur. J. Immunol. with the Rel homology domain of NF-B p65 and c-Rel proteins bound to the 24:677. NF-B binding site and activates transcription. Oncogene 9:3099. 4. Miura, S., K. Ohtani, N. Numata, M. Niki, K. Ohbo, Y. Ina, T. Gojobori, 30. Watanabe, M., M. Muramatsu, H. Hirai, T. Suzuki, J. Fujisawa, M. Yoshida, Y. Tanaka, H. Tozawa, M. Nakamura, et al. 1991. Molecular cloning and char- K. Arai, and N. Arai. 1993. HTLV-I encoded Tax in association with NF-B acterization of a novel glycoprotein, gp34, that is specifically induced by the precursor p105 enhances nuclear localization of NF-B p50 and p65 in trans- human T-cell leukemia virus type I transactivator p40tax. Mol. Cell. Biol. 11: fected cells. Oncogene 8:2949. 1313. 31. Mosialos, G. 1997. The role of Rel/NF-B proteins in viral oncogenesis and the 5. Paterson, D. J., W. A. Jefferies, J. R. Green, M. R. Brandon, P. Corthesy, regulation of viral transcription. Semin. Cancer Biol. 8:121. M. Puklavec, and A. F. Williams. 1987. Antigens of activated rat T lymphocytes 32. Ohtani, K., A. Tsujimoto, T. Tsukahara, N. Numata, S. Miura, K. Sugamura, and M. Nakamura. 1998. Molecular mechanisms of promoter regulation of the gp34 including a molecule of 50,000 Mr detected only on CD4-positive T blasts. Mol. Immunol. 24:1281. gene that is trans-activated by an oncoprotein Tax of human T cell leukemia virus 6. al Shamkhani, A., M. L. Birkeland, M. Puklavec, M. H. Brown, W. James, and type I. J. Biol. Chem. 273:14119. A. N. Barclay. 1996. OX40 is differentially expressed on activated rat and mouse 33. Higashimura, N., N. Takasawa, Y. Tanaka, M. Nakamura, and K. Sugamura. T cells and is the sole receptor for the OX40 ligand. Eur. J. Immunol. 26:1695. 1996. Induction of OX40, a receptor of gp34, on T cells by trans-acting tran- scriptional activator, Tax, of human T-cell leukemia virus type I. Jpn. 7. Godfrey, W. R., F. F. Fagnoni, M. A. Harara, D. Buck, and E. G. Engleman. ϩ J. Cancer Res. 87:227. 1994. Identification of a human OX-40 ligand, a costimulator of CD4 T cells 34. Quandt, K., K. Frech, H. Karas, E. Wingender, and T. Werner. 1995. MatInd and with homology to tumor necrosis factor. J. Exp. Med. 180:757. MatInspector—new fast and versatile tools for detection of consensus matches in 8. Gramaglia, I., A. D. Weinberg, M. Lemon, and M. Croft. 1998. Ox-40 ligand: a nucleotide sequence data. Nucleic Acids Res. 23:4878. potent costimulatory molecule for sustaining primary CD4 T cell responses. 35. Jones, D. B., C. S. Scott, D. H. Wright, H. Stein, P. C. Beverley, S. V. Payne, and J. Immunol. 161:6510. D. H. Crawford. 1985. Phenotypic analysis of an established cell line derived 9. Flynn, S., K. M. Toellner, C. Raykundalia, M. Goodall, and P. Lane. 1998. CD4 from a patient with Hodgkin’s disease (HD). Hematol. Oncol. 3:133. T cell cytokine differentiation: the B cell activation molecule, OX40 ligand, in- 36. Fujii, M., H. Tsuchiya, and M. Seiki. 1991. HTLV-1 Tax has distinct but over- structs CD4 T cells to express interleukin 4 and upregulates expression of the lapping domains for transcriptional activation and for enhancer specificity. On- chemokine receptor, Blr-1. J. Exp. Med. 188:297. cogene 6:2349. 10. Imura, A., T. Hori, K. Imada, T. Ishikawa, Y. Tanaka, M. Maeda, S. Imamura, 37. Schreiber, E., P. Matthias, M. M. Mu¨ller, and W. Schaffner. 1989. Rapid detec- and T. Uchiyama. 1996. The human OX40/gp34 system directly mediates adhe- tion of octamer binding proteins with ‘mini-extracts’, prepared from a small sion of activated T cells to vascular endothelial cells. J. Exp. Med. 183:2185. number of cells. Nucleic Acids Res. 17:0305. 11. Imura, A., T. Hori, K. Imada, S. Kawamata, Y. Tanaka, S. Imamura, and 38. Lo, K., and S. T. Smale. 1996. Generality of a functional initiator consensus T. Uchiyama. 1997. OX40 expressed on fresh leukemic cells from adult T-cell sequence. Gene 182:13. 270 Tax TRANSACTIVATES OX40 THROUGH NF-B
39. Minth, C. D., and J. E. Dixon. 1990. Expression of the human neuropeptide Y 44. Smale, S. T. 1997. Transcription initiation from TATA-less promoters within gene. J. Biol. Chem. 265:12933. eukaryotic protein-coding genes. Biochim. Biophys. Acta 1351:73. 40. Kingsley, C., and A. Winoto. 1992. Cloning of GT box-binding proteins: a novel 45. Zenzie-Gregory, B., A. Khachi, I. P. Garraway, and S. T. Smale. 1993. Mecha- Sp1 multigene family regulating T-cell receptor gene expression. Mol. Cell. Biol. nism of initiator-mediated transcription: evidence for a functional interaction be- 12:4251. tween the TATA-binding protein and DNA in the absence of a specific recog- 41. Semmes, O., and K. Jeang. 1992. Mutational analysis of human T-cell leukemia nition sequence. Mol. Cell. Biol. 13:3841. virus type I Tax: regions necessary for function determined with 47 mutant pro- 46. Bulfone-Paus, S., L. A. Dempsey, and N. Maizels. 1995. Host factors LR1 and teins. J. Virol. 66:7183. Sp1 regulate the Fp promoter of Epstein-Barr virus. Proc. Natl. Acad. Sci. USA 42. Li, C. C., F. W. Ruscetti, N. R. Rice, E. Chen, N. S. Yang, J. Mikovits, and 92:8293. D. L. Longo. 1993. Differential expression of Rel family members in human 47. Weinberg, A. D., A. T. Vella, and M. Croft. 1998. OX-40: life beyond the effector T-cell leukemia virus type I-infected cells: transcriptional activation of c-rel by T cell stage. Semin. Immunol. 10:471. Tax protein. J. Virol. 67:4205. 48. Maeda, M., N. Arima, Y. Daitoku, M. Kashihara, H. Okamoto, T. Uchiyama, 43. Lanoix, J., J. Lacoste, N. Pepin, N. Rice, and J. Hiscott. 1994. Overproduction of K. Shirono, M. Matsuoka, T. Hattori, K. Takatsuki, et al. 1987. Evidence for the NFKB2 (lyt-10) and c-Rel: a mechanism for HTLV-I Tax-mediated trans-acti- interleukin-2 dependent expansion of leukemic cells in adult T cell leukemia. vation via the NF-B signalling pathway. Oncogene 9:841. Blood 70:1407. Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021