Junb Induced by Constitutive CD30–Extracellular Signal-Regulated

Research Article

JunB Induced by Constitutive CD30–Extracellular Signal-Regulated Kinase 1/2 Mitogen-Activated Protein Kinase Signaling Activates the CD30 Promoter in Anaplastic Large Cell Lymphoma and Reed-Sternberg Cells of Hodgkin Lymphoma

Mariko Watanabe,1 Masataka Sasaki,1 Kinji Itoh,3 Masaaki Higashihara,1 Kazuo Umezawa,2 Marshall E. Kadin,5 Lawrence J. Abraham,6 Toshiki Watanabe,4 and Ryouichi Horie1,4

1Fourth Department of Internal Medicine, School of Medicine, Kitasato University; 2Department of Applied Chemistry, Faculty of Science and Technology, Keio University, Kanagawa; 3Department of Pathology, School of Medicine, Toho University; 4Laboratory of Tumor Cell Biology, Department of Medical Genome Sciences, Graduate School of Frontier Sciences, The University of Tokyo, Tokyo, Japan; 5Department of Pathology, Harvard Medical School and Beth Israel Deaconess Medical Center, Boston, Massachusetts; and 6Biochemistry and Molecular Biology, School of Biomedical and Chemical Sciences, University of Western Australia, Crawley, Western Australia, Australia

Abstract characterization of the promoter region of the CD30 gene revealed High expression of CD30 and JunB is characteristic of tumor that the CD30 promoter is composed of a microsatellite region cells in anaplastic large cell lymphoma (ALCL) and Hodgkin containing CCAT repeats and a core promoter with Sp-1 binding lymphoma (HL). Possible interactions of CD30 and JunB were sites. The Sp-1 site at À43 to À38 within the core promoter is examined in this study. We found that the CD30 promoter in responsible for maximum promoter activity (13, 14). Repeats of tumor cells of both nucleophosmin (NPM)-anaplastic lympho- the CCAT motif within the CD30 microsatellite repress core ma kinase (ALK)–positive and NPM-ALK-negative ALCL and CD30 promoter activity. Thus, it was hypothesized that the CD30 HL is regulated by a constitutively active CD30–extracellular microsatellite may be truncated in CD30-overexpressing cells, signal-regulated kinase (ERK) 1/2 mitogen-activated protein possibly because of microsatellite instability (14, 15). We recently kinase (MAPK). Phosphorylation of ERK1/2 MAPK was characterized the structure and function of the CD30 microsatellite confirmed in nuclei of tumor cells in both ALCL and HL. from H-RS cell lines. The results showed that the CD30 microsatellite of H-RS cell lines are not truncated, and the AP-1 site CD30-ERK1/2 MAPK signals induce JunB expression, which maintains high activity of the CD30 promoter. JunB induction (À377 to À371) located within the microsatellite relieves seems to be largely independent of nuclear factor KB in ALCL suppression by the microsatellite through interaction with JunB and HL. These results show a common mechanism of CD30 (10). This JunB-mediated induction of CD30 promoter seems to be overexpression in ALCL and HL, although the outcome of one mechanism underlying high levels of CD30 promoter activity in CD30 signaling differs between NPM-ALK-positive ALCL and H-RS cells (10). NPM-ALK-negative ALCL, cutaneous ALCL, and HL as we In this report, we studied mechanisms underlying high levels of CD30 and JunB expression in ALCL and H-RS cells. We first recently reported. (Cancer Res 2005; 65(17): 7628-34) examined interactions between overexpressed CD30 and JunB. Next, we investigated whether JunB-mediated induction of CD30 Introduction promoter also operates in ALCL cells. We found that CD30 signals JunB is a member of the activator protein (AP-1) transcription regulate JunB expression dependent on the extracellular signal- family. AP-1 is composed of homo- or heterodimers of the regulated kinase (ERK) 1/2 mitogen-activated protein kinase related Jun (c-Jun, JunB, JunD), Fos (Fos, FosB, Fra1, Fra2) and (MAPK) pathway, which may act to stabilize CD30 induction in ATF, CREB families (1). AP-1 is involved in different biological ALCL and H-RS cells. The results show that tumor cells of ALCL processes including cell differentiation, proliferation, and apo- and Hodgkin lymphoma (HL) share a common mechanism of CD30 ptosis (2, 3). Transcription of AP-1 family members is rapidly and JunB induction. and transiently stimulated by multiple extracellular signals (4). c-Jun and JunB have antagonistic functions in biological processes such as oncogenic transformation and cell prolifera- Materials and Methods tion. However, JunB also regulates distinct target genes in a Cell lines and cell cultures. K562 and HEK293 cell lines were obtained c-Jun-independent manner and exerts specific functions (5). from the Japanese Cancer Research Resources Bank (Tokyo, Japan) and Overexpression of JunB has been reported to be associated with Fujisaki Cell Biology Center (Okayama, Japan). Nucleophosmin (NPM)- neoplastic transformation (2, 6–8). anaplastic lymphoma kinase (ALK)+ ALCL cell lines (SUDHL1, Karpas299, Overexpression of CD30 and JunB is recognized as a common and SR786) and H-RS cell lines (HDLM2, L428, and L540) were purchased from the German Collection of Microorganisms and Cell Cultures feature of Hodgkin-Reed-Sternberg (H-RS) cells and anaplastic (Braunschweig, Germany). Mac1 is a NPM-ALK-negative cutaneous ALCL large cell lymphoma (ALCL) cells (9–12). Recent cloning and cell line (16). Establishment of an HEK293 transformant stably overexpressing CD30 was described previously (17). Nonadherent cell lines were cultured in RPMI 1640 and adherent cells in DMEM with supplementation Requests for reprints: Ryouichi Horie, Fourth Department of Internal Medicine, of recommended concentrations of FCS and antibiotics. School of Medicine, Kitasato University, 1-15-1 Sagamihara, Kanagawa 228-8555, Japan. n n Phone: 81-42-778-8111; Fax: 81-42-778-8441; E-mail: [email protected]. Inhibitors. DHMEQ is a nuclear factor B (NF- B) inhibitor that acts at I2005 American Association for Cancer Research. the level of nuclear translocation of NF-nB (18). Inhibitors for MEK1/2 doi:10.1158/0008-5472.CAN-05-0925 (UO126), c-Jun-NH2-kinase (JNK; SP600125) and p38 MAPK (SB203580),

Cancer Res 2005; 65: (17). September 1, 2005 7628 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 2005 American Association for Cancer Research. CD30 Promoter Induction by CD30-ERK1/2-JunB were purchased from Cell Signaling Technology (Beverly, MA), BioMol JunB expression vector pME-JunB was described previously (10). The (Plymouth Meeting, PA), and Sigma (Tokyo, Japan), respectively. Com- structure of the expression vector for the dominant-negative CD30, CD30D pounds were dissolved with DMSO and used for experiments at specified has been described (17). A c-Jun expression vector was kindly provided by concentrations. Dr. Y. Kasuya (Department of Biochemistry and Molecular Pharmacology, Northern blotting. Northern blot analysis was done essentially as Chiba University Graduate School of Medicine). Expression vectors for described (10). Briefly, polyadenylic acid–selected RNA was size-fractionated Raf-1 and its constitutively active form DRaf-1 were kindly provided by on 1% formalin agarose gel electrophoresis and subsequently blotted onto Dr. S. Hattori (Division of Cellular Proteomics, BML, Institute of Medical Hybond-C extra-nitrocellulose membranes (Amersham Bioscience, Piscat- Science, University of Tokyo). away, NJ). Filters were hybridized in 4Â SSC, 1Â Denhardt’s, 0.5% SDS, Reporter gene assays. Activities of the CD30 promoter were studied by j 6 0.1 mol/L NaPO4 (pH 7.0), 10% dextran Na at 65 C with 1.0 Â 10 cpm/mL transient reporter gene assays. Renilla luciferase expression vector driven by of random prime-labeled probes. After washings to a final stringency of the herpes simplex virus thymidine kinase promoter, pRL-TK (Promega, 0.2Â SSC and 0.1% SDS at 65jC, filters were exposed to XAR-5 films Madison, WI), was cotransfected to standardize the transfection efficiency (Eastman Kodak, Rochester, NY) at À80jC. RT-PCR amplified cDNA in each experiment. For each transfection, 0.5 Ag of reporter construct, fragments of human JunB and human glyceraldehyde-3-phosphate dehydro- 0.1 Ag of pRL-TK, and if indicated, 1 Ag each of the expression vector was genase were used as probes. used. Cells (2 Â 105) were transfected using LipofectAMINE 2000 reagent Immunoblot analysis. Immunoblotting experiments were done as (Invitrogen, Groningen, Netherlands) according to the manufacturer’s described previously (19, 20). Cell lysates were prepared from 1 Â 106 cells instructions. Cells were harvested after 16 hours and luciferase activities in a lysis buffer [10 mmol/L Tris-HCl (pH 7.4), 1% SDS, 1 mmol/L sodium were measured by Dual Luciferase assay kit (Promega). Representative orthovanadate, 0.1 mmol/L sodium molybdate, and 1 mmol/L phenyl- results of triplicate experiments with mean and SDs are shown in the methylsulfonyl fluoride]. Antibodies used were as follows: JunB (C-11), c-Jun figures. (H-79), Raf-1 (E-10), and a-tubulin (TU-02; all from Santa Cruz Bio- Treatment of cells with actinomycin D. To measure the half-life of technology, Santa Cruz, CA), total ERK1/2 MAPK, phosphorylated JunB mRNA, cells were treated with 10 Ag/mL actinomycin D, an inhibitor (p-)ERK1/2 MAPK (Thr202/Tyr204), JNK, p38 MAPK, and p-p38 MAPK of RNA synthesis (Sigma). Cells were incubated for 0, 0.5, 1, 2, and 5 hours, (all from Cell Signaling Technology), and CD30 (Dako, Kyoto, Japan). respectively, after the addition of actinomycin D, then harvested and Immunohistochemistry. Fluorescence immunostaining was done on subjected to Northern blotting. cultured cells and signals were detected using confocal microscopy Electrophoretic mobility shift analysis. Electrophoretic mobility shift (Radiance 2000, Bio-Rad, Richmond, CA) as described (20, 21). Primary analysis (EMSA) was done as described (10). Double-stranded oligonucleo- antibodies used were as follows: JunB (Santa Cruz), p-ERK1/2 MAPK or total tides containing the AP-1 consensus sites (5V-CGCTTGATGAGTCAGCCG- ERK1/2 MAPK (both from Cell Signaling Technology). GAA-3V) were purchased from Promega. The nucleotide sequence of the Immunostaining on paraffin-embedded specimens of primary ALCL and CD30 microsatellite AP-1 site probe is as follows: 5V-CACTCACTGATT- HL samples was done as described (10). Primary antibodies used were as CATTTTACA-3V (nucleotide position À384 to À364). Nuclear extracts follows: p-ERK1/2 MAPK, total ERK1/2 MAPK (both from Cell Signaling were prepared basically according to the method reported by Andrews Technology) or ALK (Immunotech, Marseille, France). and Faller (22). Antibodies used for supershift assays are as follows: DNA constructs. Construction of various CD30 promoter–driven JunB (C-11), c-Fos (H-125), c-Jun (H-79), FosB (H-75), Fra-1 (R-20), luciferase plasmids was described previously (10). An AP-1 site-dependent Fra-2 (L-15), CREB-1 (C-21), ATF-2 (C-19), and JunD (329; all from Santa luciferase vector p(AP-1)7 was purchased from Stratagene (Tokyo, Japan). A Cruz Biotechnology).

Figure 1. Overexpressed CD30 induces JunB in ALCL and H-RS cell lines. A, Northern blot analysis of JunB transcripts in HEK293 transformants stably overexpressing CD30 (293CD30). Two micrograms of polyadenylic acid–selected RNA were subjected to analysis. 293vec, control transformant of HEK293 transfected with vacant vector (a). Expression of JunB protein in HEK293 transformants stably overexpressing CD30. For detection of JunB, immunostaining was done as described using anti-JunB antibody and an FITC-labeled secondary antibody (ref. 20; b). Immunoblot analysis of JunB in HEK293 transformant stably overexpressing CD30. Thirty micrograms of whole cell lysates were subjected to analysis. Arrowhead indicates the position of JunB protein. Left, antibodies used (c). B, induction of JunB promoter activity by signals from overexpressed CD30. Activation of the JunB promoter by transfection of CD30 expression vector with or without a dominant-negative CD30 mutant (CD30D) was examined by reporter gene assay in HEK293 cells. vec, transfection of a vacant pME vector. C, down-regulation of JunB promoter activity by transduction of a dominant-negative CD30 mutant (CD30D) in SUDHL1, Mac1 and L428 cells (a). Down-regulation of JunB expression by transduction of CD30D in Karpas299, SUDHL1, and L428 cells (b). FLAG-tagged CD30D transduced into cells as described (20). After staining of JunB with a Texas Red labeled secondary antibody, samples were washed thrice with PBS (À) and further incubated with FITC-labeled anti-FLAG M2 antibody at 4jC overnight. After washing with PBS (À) thrice, samples were observed by confocal immunofluorescence microscopy. Arrowheads, cells expressing FLAG-tagged CD30D. Antibodies used are indicated above the photos. Left, cells analyzed are indicated.

www.aacrjournals.org 7629 Cancer Res 2005; 65: (17). September 1, 2005

Inhibition of JunB expression by small interfering RNA. Target sequences of small interfering RNA (siRNA) 5V-AATGGAACAGCCCTTC- TACCA-3V for JunB were described previously (10). Twenty picomoles of siRNAs were transduced into 2 Â 105 of Karpas299, SUDHL1, and K562 and 1 Â 105 of HEK 293 cells by LipofectAMINE 2000 (Invitrogen) according to the manufacturer’s instruction. Cells were plated and maintained in appropriate conditions and used in the following experiments. The scrambled oligonucleotide 5V-CATGCTTAAATGGGCCCATGA-3V was served as a control.

Results CD30 signals induce JunB expression in anaplastic large cell lymphoma and Hodgkin-Reed-Sternberg cell lines. Overexpres- sion of CD30 and JunB is a common feature of tumor cells in ALCL and HL cells (9–12). We decided to test the hypothesis that CD30 and JunB regulation are interrelated in these lymphomas. We first used a HEK293 transformant that highly expresses CD30 (293CD30; ref. 17). Northern blot analysis showed significantly increased JunB transcripts in 293CD30 cells compared with that in parental HEK293 cells (Fig. 1A, a). Induction of JunB protein in 293CD30 cells was shown by confocal immunofluorescence microscopy and immunoblot analysis (Fig. 1A, b and c). In reporter gene assays, cotransfection of a CD30 expression vector resulted in activation of the JunB promoter in HEK293 cells, although the magnitude was not great (Fig. 1B). Cotransfection of CD30D inhibited JunB promoter induction by CD30 overexpression (Fig. 1B). These results suggest that signals triggered by overexpressed CD30 can induce JunB expression in 293 cells. To confirm that CD30 signal-dependent induction of JunB Figure 2. JunB is induced independent of NF-nB activation. A, JunB expression actually occurs in ALCL and H-RS cell lines, we tested whether and NF-nB activation in ALCL and H-RS cell lines. Immunoblot analysis of JunB blockade of signaling from overexpressed CD30 abrogates and a-tubulin using 20 Ag of whole cell lysates. Left, positions of JunB and n induction of JunB expression in these cells. We transfected a a-tubulin (top). EMSA of NF- B. Left, position of the shifted band corresponding to NF-nB binding activity (bottom). B, levels of JunB mRNA and protein in CD30D expression vector into ALCL and H-RS cell lines and H-RS cells after inhibition of NF-nB activity by DHMEQ. H-RS cell lines were studied levels of promoter activity and JunB expression. Results of treated with 5 Ag/mL (19.15 Amol/L) of DHMEQ for the indicated times (0, 4, and 8 hours) and cells were harvested for analysis. EMSA for NF-nB reporter gene assay and confocal immunofluorescence microsco- activation (top). Northern blot analysis of JunB and glyceraldehyde-3-phosphate py showed down-regulation of JunB promoter activity and dehydrogenase transcripts. Two micrograms of polyadenylic acid–selected expression in cells transduced with FLAG-tagged CD30D RNA were subjected to analysis (middle). Immunoblot analysis of JunB and a-tubulin. Twenty micrograms of whole cell lysates were subjected to analysis (Fig. 1C, a and b). Collectively, these results provide evidence (bottom). C, analysis of half-life of JunB mRNA. L428 cells were treated with for CD30 signal-mediated induction of JunB expression in ALCL 10 Ag/mL actinomycin D, an inhibitor of RNA synthesis, for indicated periods and and H-RS cells. RNA was extracted for Northern blot analysis. One microgram of polyadenylic acid–selected RNA was subjected to analysis. Induction of JunB expression is not nuclear factor-KB– dependent in anaplastic large cell lymphoma and Hodgkin- Reed-Sternberg cell lines. We next tried to clarify the CD30 and H-RS cells, and suggest the possibility that CD30 signals other signal pathway leading to induction of JunB in ALCL and H-RS than NF-nB activation are involved in the induction of JunB cells. A recent report suggested that NF-nB is involved in the expression in these cells. induction of JunB (11, 23, 24). We showed that high levels of CD30 Constitutive CD30-extracellular signal-regulated kinase 1/2 expression could drive constitutive NF-nB activation in H-RS cells mitogen-activated protein kinase pathway is involved in (17). Therefore, we hypothesized that CD30-mediated JunB induction of JunB. We next tried to elucidate the common induction is NF-nB dependent. To test this hypothesis, we signals that induce JunB depending on CD30 overexpression in compared NF-nB activation and JunB expression in ALCL and ALCL and H-RS cells. MAPKs are evolutionary conserved serine/ H-RS cell lines. Activation of NF-nB and expression of JunB were threonine kinases connecting cell-surface receptors to critical examined by EMSA and by immunoblot analysis, respectively. The targets within cells. Previous reports suggest that CD30 signals can results clearly showed abundant JunB expression in the absence also activate ERK1/2 MAPK, JNK, and p38MAPK pathways (25, 26). of NF-nB activation in ALCL cell lines (Fig. 2A). Abundant JunB We therefore examined the effects of inhibitors of MAPK pathways expression in ALCL and H-RS cell lines was also detected by on JunB expression in ALCL and H-RS cell lines. The results Northern blot analysis (data not shown). Blockade of NF-nB showed suppression of JunB expression by MEK1/2 inhibitor activity by DHMEQ did not affect JunB protein and mRNA UO126. Inhibition of constitutively active phosphorylation of ERK1/ expression. (Fig. 2B). The very short half-life of JunB mRNA 2 MAPK was confirmed by immunoblot using phosphospecific excludes the possibility that the above result is due to a long JunB antibodies for ERK1/2 MAPK in these cells (Fig. 3A). Under the half-life (Fig. 2C). These results indicate that high levels of same experimental conditions, p38 MAPK inhibitor SB203580 JunB expression are independent of NF-nB activation in ALCL (10-50 Amol/L) and JNK inhibitor SP600125 (10-50 Amol/L) did not

Cancer Res 2005; 65: (17). September 1, 2005 7630 www.aacrjournals.org

Figure 3. Constitutive CD30-ERK1/2 MAPK pathway is involved in induction of JunB in ALCL and H-RS cells. A, immunoblot analysis of JunB expression in ALCL and H-RS cell lines treated with MEK1/2 inhibitor UO126. ALCL cell lines (SUDHL1 and Mac1) and H-RS cell line (L540) were treated for 6 hours by UO126. Left, antibodies used for immunoblot analysis. p-ERK, phospho-ERK1/2 MAPK; a Tub, a-tubulin. Top, concentrations of inhibitors used. B, CD30 enhances Raf-1-mediated induction of the JunB promoter in HEK293 cells. Reporter gene analysis of JunB promoter transfected with CD30, Raf-1 or CD30 and Raf-1. A constitutively active form of Raf-1(DRaf-1) was used as a positive control. Bottom, expression of transduced proteins by immunoblot analysis. C, expression of p-ERK1/2 MAPK and total-ERK1/2 MAPK in HEK293 transformants stably overexpressing CD30. For detection, immunostaining was done as described using primary antibody and FITC-labeled secondary antibody (20). H-RS cell line L540 served as a positive control. D, immunohistologic staining of p-ERK1/2 MAPK and total-ERK1/2 MAPK in primary ALCL (ALK+), ALCL (ALKÀ), and HL samples. ALK antibody staining suggests expression of NPM-ALK in ALCL cells. A primary breast cancer sample served as a positive control. Top, antibodies used. p-ERK, phospho-ERK1/2.

affect expression of JunB in ALCL and H-RS cell lines (data not p-ERK1/2 MAPK in tumor cell components, which was largely shown). localized within the nucleus. Total ERK1/2 MAPK was expressed In the ERK1/2 MAPK pathway, activation of cell surface ubiquitously in lymph node cells and was predominantly localized receptors is transduced by the Ras-Raf-1-ERK kinase (MEK)- within the cytoplasm. A primary breast cancer sample served as a ERK1/2 signal cascade and phosphorylates target transcription positive control, p-ERK and total ERK showed the same factors. Therefore, to further investigate the CD30-JunB pathway, expression pattern as those in ALCL and HL samples. Represen- we examined the involvement of Raf-1 in CD30-mediated JunB tative results are shown in Fig. 3D. Collectively, these results promoter activation. Transduction of a constitutively active form of suggest involvement of the CD30-ERK1/2 MAPK pathway in the Raf1 (DRaf-1) activated the JunB promoter and CD30 signals induction of JunB in ALCL and H-RS cells in vivo. enhanced Raf-1-mediated induction of the JunB promoter (Fig. 3B). Strong and constitutive activator protein binding activity in We next directly examined expression of p-ERK1/2 by immuno- anaplastic large cell lymphoma cell lines consists of JunB. We histochemistry in 293CD30 cells. The result shows strong nuclear recently showed that strong and constitutive AP-1 binding activities localized staining of p-ERK1/2 in 293CD30 compared with control consisting of JunB contributed to the high CD30 promoter activity 293vec cells (Fig. 3C). Together, these results suggest that the through the AP-1 site in H-RS cells. Thus, we examined if this ERK1/2 MAPK pathway is involved in the CD30-mediated mechanism also operates in ALCL cells. We first investigated the induction of JunB in ALCL and H-RS cell lines. binding activity for AP-1 in ALCL cell lines. EMSA analysis revealed Because overexpression of CD30 and JunB is reported to be a that ALCL cell lines, like H-RS cell lines, show constitutively strong hallmark of H-RS and ALCL cells in vivo as well, we examined binding activity of AP-1, whereas other cell lines unrelated to H-RS expression of p-ERK1/2 MAPK and total ERK1/2 MAPK in or ALCL cells do not show significant AP-1 binding activity (Fig. 4A, primary samples of ALCL and HL. Five samples of ALCL (ALK+), left). AP-1 binding activity was very weak in K562 and absent in ALCL (ALKÀ) and HL examined showed restricted expression of HEK293. Competition assays using SUDHL1 nuclear extracts and an www.aacrjournals.org 7631 Cancer Res 2005; 65: (17). September 1, 2005

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 2005 American Association for Cancer Research. Cancer Research unlabeled consensus probe clearly inhibited binding activity of JunB activates CD30 promoter through the activator protein AP-1, and an unlabeled oligomer spanning the AP-1 binding sequence site in anaplastic large cell lymphoma cell lines. We next in the CD30 promoter also inhibited binding of AP-1 (Fig. 4A,right). studied the role of JunB in driving CD30 promoter activity in ALCL To clarify the components of the AP-1 binding activity of ALCL cell lines. Introduction of a mutation in the AP-1 site significantly cells, supershift analyses were done using nuclear extracts of three decreased the promoter activities in SUDHL1, Karpas299, and NPM-ALK-positive ALCL cell lines, SUDHL1, Karpas299, SR786, and Mac1 cells, whereas it did not do so in K562 and HEK293 cell lines NPM-ALK-negative cutaneous ALCL line, Mac1, as well as unrelated to ALCL (Fig. 5A). The decreased CD30 promoter activity antibodies for AP-1 family transcription factors. The results caused by the introduction of a mutation in the AP-1 site was showed a complete shift when anti-JunB antibody was used, also observed when we used another ALCL cell line SR786 (data whereas no other antibodies showed significant shifted bands not shown). (Fig. 4B, top). Nuclear extracts of an H-RS cell line L540 showed We next wanted to test whether JunB drives the CD30 promoter the same results, as we recently reported (Fig. 4B, bottom). through binding to the AP-1 site. Transient reporter gene assays Immunoblot analysis revealed constitutive expression of c-Jun and were done using ALCL cell lines and a JunB expression vector. JunB in ALCL and H-RS cell lines. Although c-Jun was expressed c-Jun expression vector served as a control. Luciferase activities ubiquitously, JunB expression was restricted to ALCL and H-RS driven by the wild-type CD30 promoter were activated by cells (Fig. 4C, top and middle). These results suggest that the AP-1 transduction of JunB, whereas those driven by the promoter with binding complex of ALCL cells is localized in JunB, which is the an AP-1 mutation were not. Moreover, transduction of c-Jun did same result as we recently reported in H-RS cell lines (10). not activate the CD30 promoter (Fig. 5B). The relatively small activation observed in SUDHL1 and L428 cells compared with that in K562 cells may be due to higher basal levels of JunB expression in the former (10). Luciferase activities driven by the CD30 promoter were sup- pressed in ALCL cell lines by repression of endogenous JunB with siRNA as we recently reported in H-RS cell lines (Fig. 5C). Modest reduction of CD30 promoter activity by JunB siRNA in K562 cells but not in 293 cells seems to be consistent with the previous observation that K562 cells show weak JunB expression and AP-1 activity compared with control 293 cells (10). Collectively, these results provide evidence that the AP-1 site is functionally responsive to JunB, but not to c-Jun, and JunB expression contributes to high CD30 promoter activities not only in H-RS cell lines but also in ALCL cell lines.

Discussion Although there is accumulating evidence that ALCL and HL are biologically distinct, there are common features such as the high level of CD30 and JunB expression in both ALCL and H-RS cells (9–12). The mechanisms responsible for the constitutive high levels of CD30 and JunB expression seem to be essential to the pathogenesis of ALCL and HL. We reported that overexpression of CD30 could transduce constitutive signals by ligand-independent receptor aggregation (17). Therefore, we wanted to know whether overexpressed CD30 could induce constitutive JunB expression. Because we recently reported that JunB enhances CD30 promoter activity in H-RS cells, we wanted to examine whether this mechanism also operates in ALCL. Here, we present data that supports the existence of an autoregulatory mechanism by which levels of the CD30 promoter activity are controlled by self-activated CD30-ERK1/2-JunB signaling in both ALCL and H-RS cells. Since we found that ligand-independent signaling from overexpressed CD30 triggers constitutive activation of NF-nBin H-RS cells and others reported that multiple NF-nB sites in Figure 4. AP-1 activities in ALCL cells. A, EMSA of AP-1 (left). Cell lines used the promoter regulate expression of JunB (11, 23, 24), we tested are indicated above the lanes. Competition analysis of AP-1 binding activities (right). Nuclear extracts from SUDHL1 cells are used. comp, 50-fold molar the possibility that an autonomous CD30-NF-nB-JunB loop excess of unlabeled consensus AP-1 probe; MS AP-1 comp, unlabeled oligomer maintains CD30 overexpression in ALCL and H-RS cells through containing the AP-1 site (À377 to À371) in the CD30 promoter; , 50- to n 150-fold molar excess of the microsatellite oligomer. B, supershift assays of an NF- B–dependent mechanism. Surprisingly, the results showed AP-1 and CREB/ATF transcription factors. Antibodies used are indicated above that irrespective of the presence or absence of NF-nB activation, the lanes. Left, cell lines used. C, immunoblot analysis of c-Jun and JunB JunB expression is shown both in ALCL and H-RS cell lines expression in ALCL and H-RS cell lines. Twenty micrograms of whole cell lysates n n were subjected to analysis. Left, antibodies used. Cell lines used are indicated (Fig. 2A). Moreover, blockade of NF- B activation by a NF- B above the lanes. inhibitor did not significantly reduce JunB transcripts or CD30

Cancer Res 2005; 65: (17). September 1, 2005 7632 www.aacrjournals.org

in vivo (Fig. 3). Our study further shows that the ERK1/2 MAPK pathway is involved in CD30 promoter regulation in these cells. Recent reports indicate the importance of the ERK1/2 MAPK pathway in the growth of H-RS and ALCL cells and suggest that inhibition of this pathway may have therapeutic value in HL. Our study confirms the importance of this pathway as a possible therapeutic target in HL and extends it to ALCL (27, 28). Other studies showing that CD30 signals activate the ERK1/2 MAPK pathway and that ERK1/2 MAPK activates the JunB promoter support our conclusion that JunB expression is regulated by the CD30-ERK1/2 MAPK pathway (29, 30). Previous reports suggest that CD30 signals can activate ERK1/2 MAPK, JNK, and p38MAPK pathways (25, 26). We reported that NPM-ALK oncoprotein abrogates CD30-mediated activation of NF-nB by direct interaction with TRAF proteins in ALCL cells. JNK and p38MAPK transduce signals depending on TRAF proteins and these pathways also seem to be interrupted by NPM-ALK oncoprotein in ALCL cells (21). These observations support the notion that the CD30-ERK1/2 MAPK-JunB pathway functions similarly in ALCL and H-RS cells. We reported that it is not the truncation of the microsatellite but JunB-mediated relief of suppression by the microsatellite which is a mechanism underlying high levels of CD30 promoter activity in H-RS cells (10). Our observations also revealed the absence of truncation of the microsatellite in the CD30 promoter in ALCL cell lines Karpas299, SUDHL1, and SR786.7 Because the repressive activity of the microsatellite is thought to be mediated by binding of protein(s) to CCAT repeats within the microsatellite, this finding suggests that high levels of CD30 expression in ALCL cells cannot be ascribed only to alleviation of the repressive activity of the microsatellite by reduction of the CCAT repeat number. The results here show that JunB can activate the CD30 promoter through the AP-1 site in ALCL cells. These findings indicate that the same

Figure 5. JunB activates the CD30 promoter through the AP-1 site. A, the AP-1 site activates the CD30 promoter in ALCL cells. Results of luciferase assays using CD30 promoter driven luciferase constructs with or without a mutation in the AP-1 site. Activities were measured by dual luciferase assays using pRL-TK to standardize transfection efficiency. Luciferase activities of cell lines transfected with CD30 promoter construct with an AP-1 mutation are expressed as a percentage of those transfected with CD30 promoter construct without AP-1 mutation. wt, wild-type CD30 promoter construct; mt, CD30 promoter construct with a mutation in the AP-1. Cells used for transfection are indicated below. B, responses of the CD30 promoter through the AP-1 site to transduced JunB but not to c-Jun in ALCL cells. Luciferase constructs driven by the CD30 promoter with or without a mutation in the AP-1 site were cotransfected with a JunB or c-Jun expression plasmid. pRL-TK was cotransfected to standardize transfection efficiency. Luciferase activities are expressed as a percentage of those transfected with an empty vector. C, effects of siRNA-mediated down-regulation of endogenous JunB on the CD30 promoter activity in ALCL cells. pRL-TK was cotransfected to standardize the transfection efficiency. HEK293 cells served as a control. Bottom, immunoblot analysis of JunB and a-tubulin expression. Twenty micrograms of whole cell lysates were subjected to analysis. Antibodies used are indicated on the left. promoter activity in H-RS cell lines (Fig. 2B). Furthermore, it was shown that an atypical H-RS cell line, HDMYZ, does not express CD30 and lacks JunB expression in the presence of NF-nB Figure 6. Schematic view of CD30 signals and CD30 promoter induction by activation (11). Collectively, these results indicate that JunB JunB in ALCL and HL. In H-RS cells, NPM-ALK(À) systemic ALCL and cutaneous ALCL cells (left), overexpressed CD30 activates the TRAF-InB expression is dependent on CD30 signaling but not primarily on kinase (IKK)-InB-NF-nB pathway. In systemic ALCL NPM-ALK(+) cells NF-nB activation. (right), the TRAF-IKK-InB-NF-nB pathway is interrupted by association of Our results present direct evidence for the existence of a NPM-ALK with TRAF; activation of ALK tyrosine kinase supports the growth of systemic ALCL cells. CD30-ERK1/2 MAPK-JunB-AP-1 is a TRAF-independent constitutively active CD30-ERK1/2 MAPK-JunB pathway in ALCL common pathway that stabilizes overexpression of CD30 in H-RS cells, and H-RS cell lines and show that this mechanism operates NPM-ALK(+) and (À) systemic ALCL cells, and cutaneous ALCL cells. www.aacrjournals.org 7633 Cancer Res 2005; 65: (17). September 1, 2005

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 2005 American Association for Cancer Research. Cancer Research regulatory mechanisms of CD30 promoter induction operate in CD30-ERK1/2 MAPK signals induce JunB expression that maintains both ALCL and H-RS cells. high levels of the CD30 promoter in ALCL and H-RS cells. This Our study shows that JunB is the major AP-1 binding autoregulatory mechanism of CD30 expression is common to ALCL component in ALCL cells, and JunB, but not c-Jun, can activate and HL, although the outcome of CD30 signaling between ALCL the CD30 promoter. These results substantiate the regulatory NPM-ALK(+) and other lymphomas—HL, ALCL NPM-ALK(À), and function of JunB in ALCL and H-RS cells. Although JunB has been cutaneous ALCL is different, as we recently reported (Fig. 6). thought to be less active and antagonize c-Jun (2, 31–34), recent However, this amplifying loop does not seem to exist in normal cells reports suggest that JunB also regulates distinct target genes in a because the expression level of CD30 on activated lymphocytes is c-Jun-independent manner (5–8). Thus, in spite of constitutive too weak to transduce self-activating signals and effects of CD30 expression of JunB and c-Jun in ALCL and H-RS cells, JunB seems cross-linking are transient (data not shown). Some negative to be the principal molecule that drives the CD30 promoter feedback systems may prevent unregulated CD30 activation in through the AP-1 site. However, our results do not exclude the nonmalignant lymphocytes. possibility that molecule(s) other than JunB are also involved in In conclusion, although both JunB and c-Jun are constitutively CD30 promoter induction in ALCL and H-RS cells. expressed in ALCL and H-RS cells, JunB activates the CD30 Lymphomas with CD30 overexpression can be divided into four promoter through the AP-1 site. JunB expression is induced by major categories, HL, systemic ALCL NPM-ALK(+), systemic ALCL CD30 signals through an ERK1/2 MAPK pathway and contributes NPM-ALK(À), and cutaneous ALCL NPM-ALK(À). We recently to high levels of constitutive expression of CD30. Thus, ALCL tumor reported that blockade of the constitutive CD30-TRAF-NF-nB cells and H-RS cells share a common mechanism for CD30 pathway by oncoprotein NPM-ALK is responsible for absence of promoter activation. NF-nB activation in systemic ALCL, NPM-ALK(+) as further evidence that this lymphoma represents a unique entity among Acknowledgments CD30-overexpressing lymphomas (21). The results obtained in this Received 3/21/2005; revised 5/28/2005; accepted 6/28/2005. study suggest an autoregulatory mechanism by which levels of CD30 Grant support: Ministry of Education, Culture, Sports, Science and Technology, promoter activity are controlled by CD30-ERK1/2 MAPK signaling. and Japan Society of Promotion of Science grants (to R. Horie and T. Watanabe). NIH Specialized Programs of Research Excellence grant P50-CA-93683-04 (to M. Kadin). 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 7 Unpublished observations. with 18 U.S.C. Section 1734 solely to indicate this fact.

References Hodgkin’s lymphoma-associated CD30 gene. J Interferon that mediates responsiveness to interleukin-6 in a murine Cytokine Res 1998;18:915–20. plasmacytoma. J Biol Chem 1995;270:31129–35. 1. Wagner EF. AP-1-Introductory remarks. Oncogene 2001; 14. Croager EJ, Gout AM, Abraham LJ. Involvement of 25. Harlin H, Podack E, Boothby M, Alegre ML. TCR- 20:2334–5. Sp1 and microsatellite repressor sequences in the independent CD30 signaling selectively induces IL-13 2. Jochum W, Passegue E, Wagner EF. AP-1 in mouse transcriptional control of the human CD30 gene. Am J production via a TNF receptor-associated factor/p38 development and tumorigenesis. Oncogene 2001;20: Pathol 2000;156:1723–31. mitogen-activated protein kinase-dependent mecha- 2401–12. 15. Kadin ME. Regulation of CD30 antigen expression nism. J Immunol 2002;169:2451–9. 3. Karin M, Liu Z, Zandi E. AP-1 function and regulation. and its potential significance for human disease. Am J 26. Matsumoto K, Terakawa M, Miura K, Fukuda S, Curr Opin Cell Biol 1997;9:240–6. Pathol 2000;156:1479–84. Nakajima T, Saito H. Extremely rapid and intense 4. Chang L, Karin M. Mammalian MAP kinase signalling 16. Kadin ME, Cavaille-Coll MW, Gertz R, Massague J, induction of apoptosis in human eosinophils by anti- cascades. Nature 2001;410:37–40. Cheifetz S, George D. Loss of receptors for transforming CD30 antibody treatment in vitro. J Immunol 2004;172: 5. Li B, Tournier C, Davis RJ, Flavell RA. Regulation of growth factor h in human T-cell malignancies. Proc Natl 2186–93. IL-4 expression by the transcription factor JunB during Acad Sci U S A 1994;91:6002–6. 27. Levi E, Wang Z, Petrogiannis-Haliotis T, et al. Distinct T helper cell differentiation. EMBO J 1999;18:420–32. 17. Horie R, Watanabe T, Morishita Y, et al. Ligand- effects of CD30 and Fas signaling in cutaneous 6. Schorpp-Kistner M, Wang ZQ, Angel P, Wagner EF. independent signaling by overexpressed CD30 drives anaplastic lymphomas: a possible mechanism for JunB is essential for mammalian placentation. EMBO J NF-nB activation in Hodgkin-Reed-Sternberg cells. disease progression. J Invest Dermatol 2000;115:1034–40. 1999;18:934–48. Oncogene 2002;21:2493–503. 28. Zheng B, Fiumara P, Li Y, et al. MEK/ERK pathway is 7. Passegue E, Jochum W, Schorpp-Kistner M, Mohle- 18. Ariga A, Namekawa J, Matsumoto N, Inoue J, Umezawa constitutively active in Hodgkin disease: a shared Steinlein U, Wagner EF. Chronic myeloid leukemia with K. Inhibition of tumor necrosis factor-a-induced nuclear signaling pathway among CD30, CD40, and RANK that increased granulocyte progenitors in mice lacking junB translocation and activation of NF-n B by dehydroxyme- regulates cell proliferation and survival. Blood 2003;102: expression in the myeloid lineage. Cell 2001;104:21–32. thylepoxyquinomicin. J Biol Chem 2002;277:24625–30. 1019–27. 8. Robinson CM, Prime SS, Huntley S, et al. Over- 19. Horie R, Gattei V, Ito K, et al. Frequent expression of 29. Shoda T, Fukuda K, Uga H, Mima H, Morikawa H. expression of JunB in undifferentiated malignant rat the variant CD30 in human malignant myeloid and Activation of A-opioid receptor induces expression of oral keratinocytes enhances the malignant phenotype lymphoid neoplasms. Am J Pathol 1999;155:2029–41. c-fos and junB via mitogen-activated protein kinase in vitro without altering cellular differentiation. Int J 20. Horie R, Watanabe T, Ito K, et al. Cytoplasmic cascade. Anesthesiology 2001;95:983–9. Cancer 2001;91:625–30. aggregation of TRAF2 and TRAF5 proteins in the 30. Wendtner CM, Schmitt B, Gruss HJ, et al. CD30 9. Stein H, Foss HD, Durkop H, et al. CD30(+) anaplastic Hodgkin-Reed-Sternberg cells. Am J Pathol 2002;160: ligand signal transduction involves activation of a large cell lymphoma: a review of its histopathologic, 1647–54. tyrosine kinase and of mitogen-activated protein kinase genetic, and clinical features. Blood 2000;96:3681–95. 21. Horie R, Watanabe M, Ishida T, et al. The NPM-ALK in a Hodgkin’s lymphoma cell line. Cancer Res 1995;55: 10. Watanabe M, Ogawa Y, Ito K, et al. AP-1 mediated oncoprotein abrogates CD30 signaling and constitutive 4157–61. relief of repressive activity of the CD30 promoter NF-nB activation in anaplastic large cell lymphoma. 31. Chiu R, Angel P, Karin M. Jun-B differs in its microsatellite in Hodgkin and Reed-Sternberg cells. Cancer Cell 2004;5:353–64. biological properties from, and is a negative regulator Am J Pathol 2003;163:633–41. 22. Andrews NC, Faller DV. A rapid micropreparation of, c-Jun. Cell 1989;59:979–86. 11. Mathas S, Hinz M, Anagnostopoulos I, et al. technique for extraction of DNA-binding proteins from 32. Deng T, Karin M. JunB differs from c-Jun in its DNA- Aberrantly expressed c-Jun and JunB are a hallmark of limiting numbers of mammalian cells. Nucleic Acids Res binding and dimerization domains, and represses c-Jun Hodgkin lymphoma cells, stimulate proliferation and 1991;19:2499. by formation of inactive heterodimers. Genes Dev 1993; synergize with NF-n B. EMBO J 2002;21:4104–13. 23. Frazier-Jessen MR, Thompson CD, Brown R, Rawat 7:479–90. 12. Mao X, Orchard G, Lillington DM, Russell-Jones R, R, Nordan RP, Feldman GM. NF-nB elements contribute 33. Hsu JC, Cressman DE, Taub R. Promoter-specific Young BD, Whittaker SJ. Amplification and overexpres- to junB inducibility by lipopolysaccharide in the murine trans-activation and inhibition mediated by JunB. sion of JUNB is associated with primary cutaneous macrophage cell line RAW264.7. FEBS Lett 2002;513: Cancer Res 1993;53:3789–94. T-cell lymphomas. Blood 2003;101:1513–9. 203–7. 34. Mechta-Grigoriou F, Gerald D, Yaniv M. The mam- 13. Croager EJ, Muir TM, Abraham LJ. Analysis of the 24. Brown RT, Ades IZ, Nordan RP. An acute phase malian Jun proteins: redundancy and specificity. Onco- human and mouse promoter region of the non- response factor/NF-nB site downstream of the junB gene gene 2001;20:2378–89.

Cancer Res 2005; 65: (17). September 1, 2005 7634 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 2005 American Association for Cancer Research. JunB Induced by Constitutive CD30−Extracellular Signal-Regulated Kinase 1/2 Mitogen-Activated Protein Kinase Signaling Activates the CD30 Promoter in Anaplastic Large Cell Lymphoma and Reed-Sternberg Cells of Hodgkin Lymphoma

Mariko Watanabe, Masataka Sasaki, Kinji Itoh, et al.

Cancer Res 2005;65:7628-7634.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/65/17/7628

Cited articles This article cites 34 articles, 14 of which you can access for free at: http://cancerres.aacrjournals.org/content/65/17/7628.full#ref-list-1

Citing articles This article has been cited by 14 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/65/17/7628.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/65/17/7628. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.