Involvement of Human Herpesvirus-6 Variant B in Classic Hodgkin's Lymphoma Via DR7 Oncoprotein

Home , B virus, Human herpesvirus 6

Published OnlineFirst September 21, 2010; DOI: 10.1158/1078-0432.CCR-10-0470 Published OnlineFirst on September 21, 2010 as 10.1158/1078-0432.CCR-10-0470

Clinical Human Cancer Biology Cancer Research Involvement of Human Herpesvirus-6 Variant B in Classic Hodgkin's Lymphoma via DR7 Oncoprotein

Aurélie Lacroix1, Sophie Collot-Teixeira1,8, Laurent Mardivirin1, Arnaud Jaccard2, Barbara Petit3, Christophe Piguet5, Franck Sturtz6, Pierre-Marie Preux7, Dominique Bordessoule2, and Sylvie Ranger-Rogez1,4

Abstract Purpose: Hodgkin's lymphoma (HL) is associated with the presence of EBV in Reed-Sternberg (RS) cells in ∼40% of cases. Here, we studied the presence of human herpesvirus type 6 (HHV-6) variant B in RS cells of HL patients and correlated results with clinical parameters. We then examined the implication of HHV-6 DR7B protein in cell deregulation. Experimental Design: HHV-6 DR7B protein was produced in a Semliki Forest virus system. Polyclonal antibodies were then generated and used for immunochemical HHV-6 localization in HL biopsies. Binding between DR7B and p53 was studied using a double-hybrid system. Transactivation of NFκB was observed after transient transfection using reporter gene assays. We looked for Id2 factor expression after stable transfection of the BJAB cell line by reverse transcription-PCR and Western blot analysis. Results: HHV-6 was more common in nodular sclerosis subtype HL, and DR7B oncoprotein was detected in RS cells for 73.7% of EBV-negative patients. Colocalization of EBV and HHV-6 was observed in RS cells of doubly infected patients. DR7B protein bound to human p53 protein. p105-p50/p65 mRNA expression and activation of the NFκB complex were increased when DR7B was expressed. Stable expression of DR7B exhibited a strong and uniform expression of Id2. A slightly higher percentage of remission was observed in patients with RS cells testing positive for DR7B than in those testing negative. Conclusions: Collectively, these data provide evidence for the implication of a novel agent, HHV-6, in cases of nodular sclerosis HL. Clin Cancer Res; 16(19); 4711–21. ©2010 AACR.

Hodgkin's lymphoma (HL) is one of the most frequent activity in cultured H/RS cells, this factor being normally lymphomas in the Western world. Four subtypes of classic observed only for limited time intervals after stimulation HL (cHL) have been identified: (a) the nodular sclerosis with diverse inducers. These findings tended to identify (NS) subtype, which is the most frequent; (b) the mixed NFκB as an important component for the understanding cellularity subtype; (c) the lymphocyte-rich subtype; and of HL pathogenesis. (d) the lymphocyte-depleted subtype. cHL (1) is character- On another hand, RS cells were shown to carry clonally ized histologically by the pathognomonic expansion of rearranged IgV genes due to somatic mutations (6). The Hodgkin and Reed-Sternberg (H/RS) cells (2) associated key to this extensive reprogramming was notably attributed with an inflammatory infiltrate consisting of neutrophilic to the B-cell–determining transcription factor E2A. This and eosinophilic granulocytes, histiocytes, T lymphocytes, helix-loop-helix transcription factor plays important roles and plasma cells (3, 4). RS cells are typical binucleated or in promoting cell differentiation and in suppressing cell multinucleated giant CD30+ cells. Interestingly, Bargou growth. E2A is notably inhibited in RS cells by the deregu- et al. (5) have described abundant constitutive nuclear NFκB lated expression of its inhibitor of differentiation and DNA-binding Id2. Indeed, Id2 is strongly and uniformly expressed in RS cells from all cases of cHL (7). Authors' Affiliations: 1Laboratoire de Microbiologie-EA4021, Faculté de Since its initial description, HL has been considered as Pharmacie; 2Service d'Hématologie clinique, 3Service d'Anatomie an infectious disease (8) on the basis of sociogeographic pathologique, and 4Service de Virologie, Centre Hospitalier Universitaire Dupuytren; 5Service de Pédiatrie, Hôpital Mère-Enfant; 6Laboratoire de and epidemiologic data. EBV is a herpesvirus described Biochimie-EA4021 and 7Unité fonctionnelle de Recherche Clinique et as linked to 25% to 40% of HL cases. EBV-associated HL Biostatistiques, Faculté de Médecine, Limoges, France and 8Cardiovascular Division, King's College London, London, United Kingdom is mostly encountered in infants, in elderly people, and in immunocompromised patients. EBV is frequently associated Corresponding Author: Sylvie Ranger-Rogez, Laboratoire de Micro- biologie, Faculté de Pharmacie, 2 rue du Dr Marcland, 87025 Limoges with the mixed cellularity subtype (9). EBV exerts its effects Cédex, France. Phone: 33-5-55-43-58-61; Fax: 33-5-55-05-61-28; principally through its latent membrane protein-1 (LMP1) E-mail: [email protected]. expressed in RS cells and is able to transform rodent fibro- doi: 10.1158/1078-0432.CCR-10-0470 blasts in vitro and to render them tumorigenic in nude ©2010 American Association for Cancer Research. mice (10, 11).

www.aacrjournals.org 4711

Lacroix et al.

was then to analyze the role of the DR7B oncogenic pro- Translational Relevance tein in cell deregulation during this lymphoproliferative disorder. Therefore, we tested the ability of DR7B to bind In this study, we showed the presence of human to human p53, to induce NFκB transactivation, and finally herpesvirus type 6 (HHV-6) and the expression of to overexpress Id2. the viral DR7B oncoprotein in Reed-Sternberg (RS) cells from Hodgkin's lymphoma (HL) patients and Materials and Methods particularly from EBV-negative HL. DR7B influenced proliferation by binding to p53 and by increasing the Case selection κ NF B complex. DR7B expression also induced over- Lymph node biopsies from 48 patients treated for cHL expression of Id2, the inhibitor of differentiation, at Limoges University Hospital (France) from 1998 to and DNA binding of the transcription factor E2A. In 2003 were retrospectively included in this study. These our mind, these findings are of great interest because, patients belong to 68 patients previously found positive for the first time, another virus other than EBV was for HHV-6B DNA in their lymph nodes (≥10 HHV-6 shown to be implicated in HL. Interestingly, in this copies/μg), as quantified by real-time PCR, among 86 pa- study, HHV-6 was found mostly in patients with tients tested (30). For the remaining 20 others, there was nodular sclerosis HL, in contrast to EBV. This study no further sample present. The study was approved by the can be of consequence for clinicians because a slightly Institutional Review Board. HL was initially diagnosed higher percentage of remission was seen in patients based on histologic analyses of lymph node biopsies, harboring DR7B in the RS cells. according to WHO guidelines for neoplastic diseases of lymphoid tissues (35). All patients underwent baseline examinations, and clinical parameters were recorded after diagnosis and during treatment about Ann Arbor stage, Human herpesvirus type 6 (HHV-6) is a ubiquitous relapse, remission, and death in particular. HHV initially isolated from six patients with various Among these 48 biopsies, 38 were negative for EBV de- lymphoproliferative disorders (12, 13). The biological tection in the classic immunohistochemical LMP1 (CS1-4, and immunologic properties and host cell tropism of this Dako) method. These 38 patients had a mean age of 34 principally T lymphotropic virus have led to the identifica- years (range, 14-89 years) and a sex ratio of 0.9; 29 were tion of two different related variants: HHV-6A and HHV-6B hospitalized in the Clinical Haematology Department and (14). Almost all individuals over the age of 2 years are 9 in the Pediatrics Department (≤18 years). All but two infected with HHV-6 (15). After primary infection, the adult patients (one with mixed cellularity cHL and the virus remains latent in the host peripheral blood mono- other with lymphocyte-rich cHL) suffered from NS nuclear cells and vascular endothelial cells (16) until re- subtype cHL. The 10 other biopsies were HHV-6 and activation occurs. Interestingly, HHV-6 association with EBV positive. The mean age of these 10 patients was 58 different cancers (17) and angioimmunoblastic T-cell years (range, 8-81 years) and the sex ratio was 2.3; 9 pa- lymphomas was recently shown (18–20). About HL, sev- tients were hospitalized in the Department of Clinical eral serologic studies have reported significantly higher Haematology and the last one in the Department of Pedi- prevalence of anti–HHV-6 IgG and higher anti–HHV-6 ti- atrics. All except one patient (mixed cellularity) belonged ters in patients compared with healthy controls (21–23). to the NS subtype cHL. HHV-6 DNA sequences have been identified in lymph nodes from patients with HL (24–29), and the B variant Cell lines and virus-infected cell cultures was considerably more prevalent than the A variant (30). The MT4, Molt-3, JJHAN, and BHK-21 cell lines [obtained HHV-6 also shows transforming, transactivating, and on- from the American Type Culture Collection (ATCC)]; cogenic properties. Indeed, a 3.9-kbp SalI-L fragment was the HHV-6– and EBV-negative BJAB B cell line (provided shown to transform human epidermal keratinocytes by I. Joab, U542-INSERM Université Paris 11, Paris, France); RHEK-1 and both primary and established rodent cells and the EBV-positive Raji, P3HR1, and Namalwa cell lines (31, 32). The transforming activity of the SalI-L fragment (provided by J. Icart, Unité de Physiopathologie Cellulaire was localized to the DR7 gene (33), and cells expressing et Moléculaire UPR2163 CNRS, Toulouse, France) were all DR7 protein were tumorigenic when injected into nude grown in RPMI 1640 (Invitrogen) supplemented with 10% mice, whereas cells expressing truncated DR7 protein were FCS (Invitrogen). HEp-2 cell line (ATCC) was grown in not. For the HHV-6 variant A, this oncogenic potential was DMEM (Invitrogen) supplemented with 10% FCS. All cell related notably to the capacity of DR7 to bind and inacti- lines were propagated at 37°C with 5% CO2. HHV-6 variant vate the human tumor suppressor protein p53 (34). A strain U1102 and HHV-6 variant B strain Z29 (provided In this report, we looked for the presence of HHV-6 by H. Agut, Laboratoire de Virologie, CERVI, Groupe variantBandfortheexpressionofDR7proteininRS Hospitalier Pitié-Salpêtrière, Paris, France) and HHV-6 cells from HL patients' lymph nodes, and then we studied variant B strain HST (provided by B.M. Marcille- the respective localization of both oncogenic viruses, Imbert, EA4271, Laboratoire de Virologie, CHU, Nantes, HHV-6 and EBV, in doubly infected patients. Our goal France) were cultured in RPMI 1640 supplemented with

4712 Clin Cancer Res; 16(19) October 1, 2010 Clinical Cancer Research

Involvement of HHV-6B via DR7 Protein in HL

10% FCS, 1 IU/mL of anti-IFN serum (Sigma-Aldrich), and anti-DR7B and monoclonal anti-CD30 antibodies (Serotec). 2.5 μg/mL polybrene (Sigma-Aldrich). They were respec- Alexa Fluor 488–conjugated goat anti-rabbit IgG and Alexa tively propagated in JJHAN, Molt-3, and MT4 T-cell lines. Fluor 594–conjugated goat anti-mouse IgG secondary antibodies (Invitrogen) were used, respectively. DR7 variant B protein production and For the 10 patients positive for HHV-6 and EBV, double im- antibody generation munohistochemical labeling was done with anti-DR7B and The DR7B gene was amplified using primers DR7-BamF anti-LMP1 antibody. Paraffin sections were treated as previ- (5′-ACTTTAAGAGGATCCCGCCACCATGAGGGCG-3′) ously described and then stained with mouse anti-LMP1 an- and DR7-BamR (5′-GCAGCCGGATCCACTAGTAAC- tibody (Dako) followed by Alexa Fluor 594–conjugated GGCCG-3′). The DNA template used (close to the HHV-6 anti-mouse IgG as secondary antibody. The second labeling B HST strain) originated from a patient in this study and wasthendonewithrabbitDR7BantibodyfollowedbyAlexa was selected because its DR7B sequence was representative Fluor 488–conjugated goat anti-rabbit IgG. Observation was of that most frequently found in our patients. A poly-His tag made by confocal fluorescent microscopy. was previously added to this template. A Kozak sequence was included in the forward primer to facilitate expression. Two-hybrid system The amplified product was inserted into the BamHI site Two-hybrid system was carried out using the CheckMate of the modified pSFV1 vector, the central element of the Mammalian Two-Hybrid System (Promega). The cDNA Semliki Forest virus (SFV) system carrying the nonstructural from human p53 (provided by C. Tomasetto, Institute of viral genes. Sequencing confirmed the presence of the insert Genetics and of Molecular and Cellular Biology, Illkirch, in the appropriate orientation. The vector was linearized France), derived from plasmid php53B (ATCC 57255), and transcription was carried out using a strong promoter was inserted into pBIND mammalian vector after the yeast and SP6 RNA polymerase. BHK-21 cells were transfected by GAL4 DNA-binding domain sequence to generate a fusion electroporation (350 V, 750 μF) with the purified RNA protein. The DR7 gene from HHV-6B strain HST or from obtained. RNA directed its own replication in these cells, HHV-6A strain U1102 was cloned (DR7/EcoRI, 5′- resulting in the production of large amounts of the protein TACGTCGAATTCATGAGGGCGGCC-3′;DR7/SalI, 5′- of interest and inhibition of host cell protein synthesis. CCGCCGGTCGACCTAATCCAGGTACGC-3′)intopACT Transfected cells were maintained in growth medium for mammalian vector following the herpes simplex virus 16 hours to allow for DR7B protein production. An anti- VP16 activation domain sequence to generate a fusion pro- His antibody was used to test for the presence of DR7B protein with this sequence. Sequencing checked orientation of tein in cells by immunofluorescence staining and Western each insert. The pG5luc vector contains GAL4-binding sites blotting. A New Zealand female rabbit was inoculated with upstream from a TATA box, which is upstream from the fire- the dialyzed protein fraction purified by high-performance fly luciferase reporter gene. Interaction between proteins liquid chromatography and then diluted with complete p53 and DR7 results in the association of a transcriptional Freund's adjuvant for the first injection or with incomplete activation domain with a DNA-binding domain, which Freund's adjuvant for the second injection 21 days later. may promote the assembly of RNA polymerase II com- Sera were collected from this rabbit before and after immu- plexes at the TATA box and increase transcription of the fire- nization and checked by immunofluorescence staining, fly luciferase reporter gene. The pBIND vector, which first on a viral culture of HHV-6B strain HST and second expresses in addition the Renilla reniformis luciferase under on the HHV-6A strain U1102. It was also checked by West- control of the SV40 promoter, was used to normalize the ern blotting against a protein extract from a viral culture and efficiency of transfection and luciferase gene expression. against DR7B protein used for immunization. Serum was Briefly, HEp-2 cells cultured in six-well plates at a density controlled for specificity by testing on mock cell lines of 106 per well during 24 hours at 37°C were transfected at (BHK-21, MT4, and JJHAN cell lines). It was also tested ∼70% confluence with 3 μg of each plasmid (pBIND, against EBV-positive cell lines (Raji, P3HR1, and Namalwa) pACT, and pG5luc) mixed with 27 μL liposome FuGENE to verify the absence of labeling. 6 Transfection Reagent (Roche) in serum-free Dulbecco's medium. Two milliliters of Dulbecco's medium supple- Immunohistochemistry mented with 5% FCS were added 6 hours after beginning Anti-DR7B antibodies were incubated with Bouin-fixed, transfection. After 24 hours, cells were lysed and the amount paraffin-embedded samples from the 48 patients. After de- of firefly luciferase was quantified using the Dual-Luciferase waxing in toluene, these sections were pretreated with Dako Reporter Assay System (Promega). Values were normalized buffer and endogenous biotin was blocked (Dako). Immuno- to Renilla luciferase activity. For all assays, negative controls histochemistry was then done with the LSAB2 system (Dako). (each plasmid alone or each initial vector alone) were done. Peroxidase staining was carried out with diaminobenzidine Each experiment was done in triplicate. as the substrate, and sections were counterstained with hematoxylin. The same protocol was used for staining with Transient transfection of HEp-2 cells for NFκB anti-gp116/54/64 antibodies (Advanced Biotechnologies). transactivation study For samples testing positive with anti-DR7B antibodies, Fragments corresponding to the promoter region of the double immunohistochemical labeling was done with p105-p50 and p65 human NFκB gene were amplified from

www.aacrjournals.org Clin Cancer Res; 16(19) October 1, 2010 4713

Lacroix et al.

human genomic DNA by PCR using the following primers: p50/ reporter located after multiple copies of the NFκB consensus KpnI, 5′-ACAAAGGTACCGCTGGGTTTTTGTGTGTGTGG-3′, sequence. After endogenous NFκB proteins bind to the κ p50/BglII, 5′-GAGTTAGATCTACCCGGGTGGCGAGTCC-3′; enhancer element, transcription is induced and the reporter p65/KpnI, 5′-GTTTAGGTACCGGATCCATGCAAGGTTCAATG-3′; gene is activated. pTAL-luc vector (Clontech) containing the and p65/HindIII, 5′-TAATAGAAGCTTCATGGCGGGGTCCCG- TATA-like promoter region from the herpesvirus thymidine 3′. After being purified and sequenced, p105-p50 or p65 kinase promoter, but not the NFκB consensus sequence, was amplified fragments were cloned into the pGL3 Enhancer used as a negative control to normalize the effect of the Vector (Promega) upstream from the firefly luciferase report- minimal HSV1 promoter present in the pNFκB-luc vector. er gene to generate p50PROMO-pGL3ev and p65PROMO- DR7-pCI was cotransfected with pNFκB-luc or pTAL-luc pGL3ev plasmids. Sequencing checked orientation of each vector into HEp-2 cells using FuGENE 6. Luciferase activity insert. The DR7B gene of HHV-6 was amplified from the was determined after 24 hours. Negative controls were done HST strain and then cloned directly into the transient and the assay was repeated three times. mammalian expression vector pCI (Promega) to generate DR7-pCI. It was transfected with p50PROMO-pGL3ev or Stable transfection of DR7B gene into Burkitt's p65PROMO-pGL3ev in HEp-2 cells using FuGENE 6. lymphoma B-cell line BJAB pBIND vector was used to normalize transfection efficiency. Amplification of DR7B gene was done from a template Negative controls (each plasmid alone, or pCI or pGL3 ini- originating from a representative patient with the primers tial vectors alone) were done, and the assay was done in DR7/ClaI-S (5′-TGTACGATCGATACATGAGGGCG-3′) triplicate. and DR7/ClaI-AS (5′-CCGGGGATCGATCTACCCGGG- 3′), and the amplified product was cloned into the ClaI Transient transfection of HEp-2 cells for NFκB activity restriction site of the pTRIN76 vector (36). In this plasmid, pNFκB-luc vector (Clontech) is a reporter vector especially elements of the tetracycline regulatable system are in designed for monitoring activation of the NFκB signal trans- opposite directions and the selection gene (NeoR)is duction pathway. It contains the firefly luciferase gene as a downstream from an IRES sequence. Electroporation of

Fig. 1. RS cells labeled with antibody against DR7B. A, RS cells are labeled with the rabbit anti-DR7B primary antibody and horseradish peroxidase– conjugated goat anti-rabbit IgG+A secondary antibody. Magnification, ×400. B, the same labeling reveals DR7B-positive mummified cells. Magnification, ×1,000. C, a RS cell is labeled with the rabbit anti-DR7B primary antibody and Alexa Fluor 488–conjugated goat anti-rabbit IgG secondary antibody. D, the same cell is labeled by mouse anti-CD30 primary antibody and Alexa Fluor 594–conjugated goat anti-mouse IgG secondary antibody. E, double labeling (DR7B/CD30) of the same cell.

4714 Clin Cancer Res; 16(19) October 1, 2010 Clinical Cancer Research

Involvement of HHV-6B via DR7 Protein in HL

Table 1. Characteristics of the 38 HL patients positive for HHV-6B and negative for EBV: general population, DR7B, and gp116/54/64 labeling

Characteristic General DR7B+ in DR7B− in gp complex+ gp complex− DR7B+ in DR7B− in population RS cells RS cells in RS cells in RS cells infiltrating infiltrating (N = 38) (n = 28) (n = 10) (n = 15) (n = 23) cells (n = 17) cells (n = 21)

Mean age ± SD 34.0 ± 17.0 35.6 ± 18.1 29.5 ± 13.1 36.1 ± 19.5 32.6 ± 15.5 25.8 ± 12.7 40.6 ± 17.4 in years (extreme values) (14-89) (14-89) (16-51) (14-89) (15-73) (14-56) (15-89) Sex ratio 0.9 1.0 0.7 0.9 0.9 0.9 0.9 Stages I-II (%) 20/35 (57.1%) 13/25 7/10 4/13 (30.8%) 16/22 (72.7%) 9/16 (56.25%) 11/19 (57.9%) (52.0%) (70.0%) Stages III-IV (%) 15/35 (42.9%) 12/25 3/10 9/13 (69.2%) 6/22 (27.3%) 7/16 (43.75%) 8/19 (52.6%) (48.0%) (30.0%) Remission, OS (%) 27/37 (73.0%) 21/27 6/10 11/15 (73.3%) 16/22 (72.7%) 12/17 (70.6%) 15/20 (75.0%) (77.8%) (60.0%) Median amounts of HHV-6 164 165 147 181 150 116 181 in copies/μg DNA

Abbreviation: OS, overall survival.

107 cells with 10 μg plasmid DNA (pTRIN76-DR7B or ing strategy. These constructs provided standard DNAs pTRIN76 alone) was carried out at a capacitance of 960 μF for calibration of gene copy number. Serial 10-fold dilu- and a voltage of 270 V. After 48 hours of incubation, tions of each recombinant construction were made to clones with stably integrated pTRIN76 or pTRIN76-DR7B prepare stocks containing 1 to 106 copies/5 μL. Real- were selected by culturing in the presence of 400 μg/mL time PCR was done with 1× LightCycler FastStart DNA G418 (Dako). Inducible DR7B expression was revealed Master HybProbe kit (Roche Diagnostics) in a Light- with the same medium supplemented with doxycycline Cycler instrument (Roche). Amplification conditions were (0 and 1 μg/mL). Medium was changed after 24 hours as follows: for DR7B, 4 mmol/L MgCl2,600mmol/L to preserve doxycycline repression. each primer, and 400 nmol/L probe; for Id2, 4 mmol/L MgCl2, 750 mmol/L each primer, and 400 nmol/L probe; Real-time reverse transcription-PCR and for GAPDH, 3 mmol/L MgCl2, 520 mmol/L each DR7B and Id2 expressions were determined by real-time primer, and 100 nmol/L probe. After initial denaturation PCR, and ratios were normalized to glyceraldehyde- at 95°C for 10 minutes, 50 cycles with denaturation at 3-phosphate dehydrogenase (GAPDH) expression. Total 95°C for 10 seconds and annealing at 60°C for 1 minute RNA was extracted from cultured cells by a single-step were carried out. A final cooling step at 40°C for 1 min- guanidium thiocyanate-phenol chloroform method using ute was added. Specificity, sensitivity, and reproducibility Trizol reagent (Life Technologies) according to the manu- criteria were checked to validate the method. facturer's instructions. Any contaminant DNA was elimi- nated by the use of DNase I RNase-free (Roche) before Western blotting reverse transcription (37). After priming with hexanucleo- Transfected cell pellets were suspended in lysis buffer tide mix (Roche), first-strand cDNA synthesis was carried [0.5% NP40, 0.5 mol/L NaCl, 10 mmol/L Tris-HCl out on 200 ng DNase I–treated RNA with Moloney murine (pH 7.5)] and centrifuged. Supernatant was collected, and leukemia virus reverse transcriptase (Invitrogen). PCRs after reading at 595 nm according to the Bradford method, were then done on 100 ng cDNA with the following primers total proteins were separated on 15% SDS-polyacrylamide and Taqman probes: DR7B, 5′-GACCGCTGTTATCGTTCTGG- gels and transferred to nitrocellulose membranes (Sigma- 3′ (forward) and 5′-CCGAATACGTCCAATGTCCT-3′ (re- Aldrich). Membranes were blotted overnight at 4°C verse); Id2, 5′-CCGCACAATGCTCATTACC-3′ (forward) and with various primary antibodies after blocking with 5% 5′-CAGCTCTTCGATACTGACGC-3′ (reverse); GAPDH, 5′- skim milk. Antibody dilutions were 1:2,560, 1:10, and AACAGCCTCAAGATCATCAGC-3′ (forward) and 5′- 1:25,000 for anti-DR7B, anti-Id2 (Invitrogen), or anti- GGTATCGTGGAAGGACTCATG-3′ (reverse); DR7B probe, GAPDH (Ambion), respectively. Blots were then revealed 5′-FAM-ATAACGGCAAGATCTGCCGTCTGG-TAMRA-3′;Id2 with 1:1,000 anti-rabbit IgG-biotin (Sigma) for DR7B probe, 5′-FAM-CCAGAGCCTGGCTGTGGGGA-TAMRA-3′; and Id2 or with 1:1,000 anti-mouse IgG-biotin (Sigma) GAPDH probe, 5′-FAM-TGCACCACCAACTGCTTAGCACC- for GAPDH. After addition of enhanced chemilumines- TAMRA-3′. Each amplified fragment was cloned into cence reagents (Amersham Pharmacia Biotech), blots the pCR2.1 TOPO plasmid (Invitrogen) using a TA clon- were exposed to X-ray film.

www.aacrjournals.org Clin Cancer Res; 16(19) October 1, 2010 4715

Lacroix et al.

Statistical analysis positive for LMP1 (Fig. 2). For the only mixed cellularity Labeling for DR7B or gp116/64/54 in RS cells was com- subtype cHL, DR7B was detected in infiltrating cells and pared between patients as a function of sex, stages, and not in RS cells, which, in contrast, were stained by the remission using the χ2 test. The Student's t and Mann- anti-LMP1 antibody. Despite the low number of patients, Whitney tests were used to compare labeling as a function some trends seem to appear. The six patients who pre- of age and HHV-6 quantification, respectively. Multi- sented DR7B labeling that occurred only in RS cells were variate analysis (stepwise descending logistic regression) all males, whereas the sex ratio was 0.33 for the other was carried out on factors such as HHV-6 quantification, three patients. Ann Arbor stages at diagnosis were notably age, sex, stages, remission, and immunohistochemical higher for these six patients compared with the others (five labeling with antibodies against DR7B and/or gp116/ ofsixversusoneofthreepatientswereatstageIII/IV). 64/54 complex to identify prognostic factors. For NFκB, Finally, long-term remission occurred more frequently p53, and Id2 studies, differences in gene expression or when DR7B was expressed in both infiltrating and RS cells protein activity were analyzed by paired t test. P values (two of three versus three of six patients). of ≤0.05 were considered as statistically significant.

Results

Detection of HHV-6 and its DR7B oncoprotein in RS cells from EBV-negative lymph node sections DR7B protein was detected in the cytoplasm of RS cells (Fig. 1A and B) in 28 of 38 (73.7%) HHV-6–positive and LMP1-negative patients tested. All RS cells staining with anti-DR7B antibody were stained with anti-CD30 antibody (Fig. 1C-E). In 17 of these 28 patients (60.7%), DR7B was detected only in RS cells. All characteristics of patients with DR7B-positive or DR7B-negative RS cells are reported in Table 1. A slightly higher mean age and higher stage at diagnosis were found for patients with RS cells testing positive for DR7B compared with those testing negative. Median values for HHV-6 quantification were similar in patients with and without DR7B in their RS cells. Staining intensity was not correlated with any of the parameters studied. The strongest labeling was observed in the only patient with a lymphocyte-rich subtype. Mummified cells (HL cells that display features of nonclassic apoptosis) were positively labeled for DR7B in 9 of 28 cases (32.1%), this staining being more frequent in stages I and II (77.8%) than in stages III and IV (22.2%). These cells were associated with remission in eight of nine cases (88.9%). In 17 patients (44.7%), DR7B was detected in infiltrating cells (lymphocytes, plasmocytes, polymorphonuclear cells, or histiocytes). These patients were significantly (P = 0.0059) younger (25.8 ± 12.7 years) than those testing negative (40.6 ± 17.4 years). About the HHV-6 gp116/64/54 complex, infiltrating cells tested positive in all patients, whereas RS cells tested positive in only 15 patients (39.5%). RS labeling was significantly more frequent (P = 0.0154) at stages III and IV (60%) than at stages I and II (20%).

Colocalization of DR7B from HHV-6 and LMP1 from EBV oncoproteins in RS cells DR7B was detected in RS cells from all nine patients positive for HHV-6 and EBV with the NS subtype of Fig. 2. A, rabbit anti-DR7B primary antibody was revealed with Alexa cHL. For six of them, DR7B was present only in RS cells, Fluor 488–conjugated goat anti-rabbit IgG secondary antibody. B, mouse anti-LMP1 primary antibody was revealed with Alexa Fluor whereas for the three others, DR7B was also found in in- 594–conjugated goat anti-mouse IgG secondary antibody. C, double filtrating cells. Double immunohistochemical labeling labeling of a RS cell with antibodies against DR7B and LMP1. Images showed that all RS cells stained by anti-DR7B were also were taken with a Nikon Eclipse microscope at ×600 magnification.

4716 Clin Cancer Res; 16(19) October 1, 2010 Clinical Cancer Research

Involvement of HHV-6B via DR7 Protein in HL

Fig. 3. Mean results obtained for two-hybrid system between DR7 (A or B) and p53, expressed in relative luminescence unit (RLU). Each test included the transfection of pG5luc vector except for nontransfected cells. **, P ≤ 0.01.

DR7B binding to p53 protein to 1 μg/mL doxycycline (Fig. 5A), nor in untransfected To test for potential interactions between viral proteins BJAB cells or BJAB cells stably transfected with plasmid DR7(AorB)andhumanp53protein,full-lengthDR7 alone. Similar findings were observed for DR7B protein (A or B) genes fused to the activation domain of VP16 expression by Western blot assay (data not shown). in pACT vector and full-length p53 cDNA fused to the Id2 mRNA was quantified in similar amounts in untrans- GAL4-binding site in pBIND vector were transfected into fected BJAB cells, BJAB cells transfected with control plas- HEp-2 cells at the same time as pG5luc. Overall results mid, and DR7B-transfected cells cultured with 1 μg/mL obtained for luciferase reporter gene expression 24 hours doxycycline. Inversely, in DR7B-transfected cells cultured after transfection of the three plasmids—pBIND, pACT, without doxycycline (overexpression of DR7B), Id2 ex- and pG5luc—are presented in Fig. 3 and revealed signifi- pression was largely increased (∼270-fold). Similar results cant binding (P ≤ 0.01) between DR7B and p53 similar were observed by Western blot, showing higher Id2 protein to DR7A. levels in DR7B stably transfected BJAB cells than in controls, although GAPDH protein levels were similar for each Upregulation of p105-p50 and p65 promoters by condition (Fig. 5B). DR7B viral protein To detect upregulation of p65 and p105-p50 promoters Discussion by DR7B, the expression plasmid DR7-pCI was cotransfected with p105-p50 or p65 promoter constructs. Normal- This report revealed, for the first time, the presence of ized results reported in Fig. 4 showed a statistically HHV-6 variant B and the expression of DR7B protein in significant increase in luciferase reporter gene activity for RS cells from HL patient lymph nodes. In patients dually both p65 (58.2%, P = 0.03) and p105-p50 (43.9%, infected by HHV-6 and EBV, both viruses were localized P = 0.05) promoters when DR7B viral protein was expressed. in the same cells. We provided evidence that DR7B is able to bind to human p53 and induce NFκB transactivation Stimulation of NFκB activity by DR7B viral protein and that it can increase Id2 expression after DR7B stable Because p105-p50 and p65 promoters were regulated by transfection. DR7B viral protein expression, we wondered if these In this preliminary study, conducted on a cohort of proteins could also induce a change in NFκB activity. heterogeneous cHL patients, HHV-6 was detected in RS For this purpose, we cotransfected DR7-pCI with pNFκB- cells in 39.5% of LMP1-negative patients tested by the luc or pTAL-luc vector into HEp-2 cells. As reported in viral anti-gp116/64/54 antibody. To our knowledge, Fig. 4, DR7B viral protein expression induced a significant HHV-6 detection in RS cells has been reported only once increase (P = 0.05) in luciferase reporter gene activation and in a very small number of cases (38). Here, we tested and therefore NFκB activity (90%). a large population in which 44.7% were young adults (18- 40 years) with NS cHL. A much larger number of patients Overexpression of Id2 in BJAB cells stably transfected (73.7%) had RS cells testing positive for DR7B. The CD30 by DR7B phenotype was identified by double labeling of DR7B- DR7B mRNA was expressed in large amounts (∼30 × labeled RS cells for each patient, although RS cells were 107 copies/μg RNA) in BJAB cells stably transfected by recognizable on the basis of their morphology. Only a the DR7B gene cultured in the absence of doxycycline. few patients had RS cells testing positive for DR7B and No expression was detected when these cells were exposed gp116/64/54. Discrepancies in the labeling patterns

www.aacrjournals.org Clin Cancer Res; 16(19) October 1, 2010 4717

Lacroix et al.

Fig. 4. Transactivation and increase in the activity of NFκB when DR7B protein is expressed. A, results obtained for p50 and p65 promoter activation following DR7B (+DR7) viral protein expression. Reactions are compared with similar transfection assays done without viral protein expression (−DR7). The very low amount of luciferase luminescence detected in the different negative controls was used as background. B, comparison of results obtained for NFκB activation with DR7B protein (+DR7) expression or without viral protein expression (−DR7). Columns, mean; bars, SD. *, P ≤ 0.05.

obtained with these antibodies did not result from differ- finding may possibly have some important pathogenic ences in sensitivity. Anti-DR7B antibodies labeled a larger implications. number of RS cells and a smaller number of environmental Interestingly, a large percentage of patients with DR7B- cells than anti-gp116/64/54 antibodies. gp116/64/54 anti- positive RS cells were in remission, although they were bodies detects glycoproteins produced late during the olderageandhadahigherstage—two prognostic factors replication cycle and therefore can be used to monitor classically considered as negative. For these patients, treat- complete viral replication in infected cells. Inversely, ment, whatever it was, seemed to be more effective than for DR7B antibodies label a protein produced early in the viral DR7B-negative patients. Inversely to classic RS cells, mum- cycle. Furthermore, the median amounts of HHV-6 DNA mified cells were largely DR7B positive at early disease were similar, regardless of the labeling pattern. Whereas stages. These observations (older age and higher stage), all patients exhibited gp116/64/54 labeling of at least some not found when gp116/64/54 antibodies were used, led infiltrating cells, DR7B protein was observed in infiltrating us to suppose that DR7B oncoprotein accumulates in RS cells from only 17 patients. These patients were significantly cells at advanced stages of the disease and in older patients, younger than those with no DR7B in infiltrating cells, and as if viral replication stopped at an early step. labeling was not associated with particular features of HL. The accumulation of DR7B in RS cells may be related to HHV-6 DR7B positivity in tumor and bystander cells, or cell proliferation by binding to the human tumor suppres- only in one type, seemed to show different mechanisms sor protein p53, preventing its elimination. In cancer, p53 of pathogenic pathways and different environmental influ- function may be affected in different ways, notably by ences on the tumor cells in different age groups of cHL. This direct interaction with viral proteins, leading to p53

4718 Clin Cancer Res; 16(19) October 1, 2010 Clinical Cancer Research

Involvement of HHV-6B via DR7 Protein in HL

inactivation (39). Human papillomavirus 16 and 18 onco- HBx from hepatitis B virus, Rta from HHV-8, and EBNA-2 proteins E6, SV40 oncoprotein large T antigen, and adeno- and LMP1 from EBV (46, 47) can activate NFκB. All these virus E1B 55-kDa protein have all been shown to interact proteins have transactivating or signal-inducing functions with p53 protein, inhibiting its control of cell cycle arrest on their own. Most of these viruses either can be carcino- after DNA damage. The finding that DR7B was able to genic themselves or are associated with tumor develop- bind to human p53 suggests that this interaction may in- ment. In this study, we showed that DR7 protein from duce or participate in lymphomagenesis, considering the HHV-6B is able to transactivate p105-p50 and p65 NFκB fact that HHV-6 is able to retain p53 within the cytoplasm, subunits in HEp-2 cells and to activate the NFκBcomplex. thus protecting infected cells from apoptosis (40, 41). Therefore, HHV-6B is able to induce or reinforce NFκBcon- This leads to p53 accumulation and upregulation in the stitutive transactivation in H/RS cells. presence of HHV-6 (42). Because DR7B protein was found to be expressed in RS We explored nine NS cHL patients positive for both LMP1 cells of cHL and because the virus was implicated in lym- oncoprotein of EBV and DR7B; double immunohistochem- phomagenesis pathways (p53 and NFκB), it seemed inter- ical labeling showed that all DR7B-positive RS cells were esting to examine the modifications conferred by DR7B also positive for LMP1. Both viruses could interact, either expression in B-cell progenitors of RS cells. It was therefore directly or more probably indirectly particularly through important to construct a model, and a Burkitt's lymphoma p53 (40) and/or NFκB pathway involvement (43). It was cell line (BJAB) negative for EBV and HHV-6 was chosen. previously suggested that HHV-6 can activate EBV rep- The pTRIN76 plasmid allowed us to stably transfect the lication and may contribute to the pathogenesis of EBV- BJAB cell line and to modulate DR7B expression by the associated diseases (44). For the only mixed cellularity HL use of doxycycline concentrations. Id2 factor, an inhibitor patient who had DR7B in infiltrating cells and not in RS cells of E2A transcription factor, has been reported to be in- as opposed to LMP1, the mechanism is certainly different. volved in reprogrammation of the B-cell phenotype of Constitutive activation of NFκB complex seems to be a RS cells and consecutive proliferation of these cells (48). common feature in H/RS cells from HL (5), thus involving In the present study, stable expression of the DR7B protein resistance to cell death and enhancing proliferation. An caused overexpression of Id2. This interesting result can obvious mechanism through which NFκB can promote give rise to different hypotheses about DR7B action on cell tumorigenesis is by inducing expression of antiapoptotic signaling pathway deregulation such as the transforming genes. Other mechanisms also exist: The cyclin D1 growth factor-β signaling pathway (49). In fact, RS cells promoter, for example, was identified as a target of NFκB possess both transforming growth factor-β ligand and re- (45), promoting G1-to-S phase progression in the cell cycle ceptor, so autocrine signaling could contribute to aberrant and resulting in cyclin D1 expression. NFκB can also stim- Id2 expression. Furthermore, Id2 upregulation probably ulate c-Myc promoter activity. Different mechanisms are results from AP-1 deregulation (50). likely to be implicated in NFκB constitutive activation. It In summary, our results revealed the presence and accu- has been shown that expression of a single viral protein mulation of HHV-6B in RS cells of patients suffering from such as Tax from human T-cell lymphotropic virus type 1, HL. The DR7B binding to p53 and the ability of DR7B to

Fig. 5. Increased expression of Id2 when DR7B is expressed. A, Id2 mRNAs are overexpressed in BJAB cells when DR7B mRNAs are produced as quantified by Taqman technology. BJAB cells were stably transfected with DR7B gene and cultured without doxycycline (DR7 0) or in the presence of 1 μg/mL doxycycline for 48 h (DR7 1). A cellular control (BJAB untransfected cells) and a control vector (BJAB cells stably transfected by the pTRIN76 plasmid alone) were shown. The values are expressed relative to that of GAPDH mRNAs. Columns, mean of three separate assays; bars, SD. B, upregulation of Id2 protein expression is observed in DR7B stably transfected BJAB cells cultured without doxycycline (DR7 0) by Western blot. Id2 protein is weakly detected in the diverse controls (BJAB, pTRIN76, and DR7 1). **, P ≤ 0.01.

www.aacrjournals.org Clin Cancer Res; 16(19) October 1, 2010 4719

Lacroix et al.

transactivate NFκB and to overexpress Id2 are different ar- Grant Support guments in favor of an implication of HHV-6 in NS cHL. Finally, in this study, the detection of DR7B in the RS cells This study was supported by grants from the “Ligue contre le Cancer (Comité Haute-Vienne)” and from the National Health Service (Direction of EBV-negative HL patients may be considered as a surro- générale de la Santé, Paris) and promoted by the “Centre Hospitalier gate marker for the identification of patients with a high Universitaire,” Limoges, France. probability of remission. 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 with 18 U.S.C. Section 1734 solely to indicate this fact. Disclosure of Potential Conflicts of Interest Received 02/22/2010; revised 05/28/2010; accepted 07/12/2010; No potential conflicts of interest were disclosed. published OnlineFirst 09/21/2010.

References 1. Fraga M, Forteza J. Diagnosis of Hodgkin's disease: an update on 20. Zhou Y, Attygalle AD, Chuang SS, et al. Angioimmunoblastic T-cell histopathological and immunophenotypical features. Histol Histo- lymphoma: histological progression associates with EBV and HHV6B pathol 2007;22:923–35. viral load. Br J Haematol 2007;138:44–53. 2. Kuppers R. The biology of Hodgkin's lymphoma. Nat Rev Cancer 21. Clark DA, Alexander FE, McKinney PA, et al. The seroepidemiology 2009;9:15–27. of human herpesvirus-6 (HHV-6) from a case-control study of leukae- 3. Weiss LM, Chan JK, MacLennan K, Warnke RA. Pathology of clas- mia and lymphoma. Int J Cancer 1990;45:829–33. sical Hodgkin's disease. In: Mauch P, Armitage J, Diehl V, Hoppe R, 22. Shanavas KR, Kala V, Vasudevan DM, Vijayakumar T, Yadav M. Anti- Weiss L, editors. Hodgkin's disease. Philadelphia: Lippincott HHV-6 antibodies in normal population and in cancer patients in Williams and Wilkins; 1999, p. 101–20. India. J Exp Pathol 1992;6:95–105. 4. Chan WC. The Reed-Sternberg cell in classical Hodgkin's disease. 23. Torelli G, Marasca R, Montorsi M, et al. Human herpesvirus 6 in non- Hematol Oncol 2001;19:1–17. AIDS related Hodgkin's and non-Hodgkin's lymphomas. Leukemia 5. Bargou RC, Emmerich F, Krappmann D, et al. Constitutive nuclear 1992;6 Suppl 3:46–8S. factor-κB-RelA activation is required for proliferation and survival of 24. Torelli G, Marasca R, Luppi M, et al. Human herpesvirus-6 in human Hodgkin's disease tumor cells. J Clin Invest 1997;100:2961–9. lymphomas: identification of specific sequences in Hodgkin's lym- 6. Kuppers R, Rajewsky K, Zhao M, et al. Hodgkin disease: Hodgkin phomas by polymerase chain reaction. Blood 1991;77:2251–8. and Reed-Sternberg cells picked from histological sections show 25. Sumiyoshi Y, Kikuchi M, Ohshima K, Takeshita M, Eizuru Y, clonal immunoglobulin gene rearrangements and appear to be Minamishima Y. Analysis of human herpes virus-6 genomes in lym- derived from B cells at various stages of development. Proc Natl phoid malignancy in Japan. J Clin Pathol 1993;46:1137–8. Acad Sci U S A 1994;91:10962–6. 26. Di Luca D, Dolcetti R, Mirandola P, et al. Human herpesvirus 6: a 7. Renne C, Martin-Subero JI, Eickernjager M, et al. Aberrant expres- survey of presence and variant distribution in normal peripheral sion of ID2, a suppressor of B-cell-specific gene expression, in lymphocytes and lymphoproliferative disorders. J Infect Dis 1994; Hodgkin's lymphoma. Am J Pathol 2006;169:655–64. 170:211–5. 8. Hodgkin T. On some morbid appearances of the absorbent glands 27. Valente G, Secchiero P, Lusso P, et al. Human herpesvirus 6 and and spleen. Med Chir Trans 1832;17:68–114. Epstein-Barr virus in Hodgkin's disease: a controlled study by poly- 9. Rezk SA, Weiss LM. Epstein-Barr virus-associated lymphoprolifera- merase chain reaction and in situ hybridization. Am J Pathol 1996; tive disorders. Hum Pathol 2007;38:1293–304. 149:1501–10. 10. Wang D, Liebowitz D, Kieff E. An EBV membrane protein expressed 28. Collot S, Petit B, Bordessoule D, et al. Real-time PCR for quantifica- in immortalized lymphocytes transforms established rodent cells. tion of human herpesvirus 6 DNA from lymph nodes and saliva. J Clin Cell 1985;43:831–40. Microbiol 2002;40:2445–51. 11. Baichwal VR, Sugden B. Transformation of Balb 3T3 cells by the 29. Ranger-Rogez S, Lacroix A, Denis F, Bordessoule D. HHV-6 and the BNLF-1 gene of Epstein-Barr virus. Oncogene 1988;2:461–7. lympho-hematopoietic system. In: Krueger GR, Ablashi DV, editors. 12. Salahuddin SZ, Ablashi DV, Markham PD, et al. Isolation of a new Human herpesvirus-6: general virology, epidemiology and clinical virus, HBLV, in patients with lymphoproliferative disorders. Science pathology. 2nd ed. Amsterdam: Elsevier; 2006, p. 185–200. 1986;234:596–601. 30. Lacroix A, Jaccard A, Rouzioux C, et al. HHV-6 and EBV DNA 13. Biberfeld P, Kramarsky B, Salahuddin SZ, Gallo RC. Ultrastructural quantitation in lymph nodes of 86 patients with Hodgkin's lymphoma. characterization of a new human B lymphotropic DNA virus (human J Med Virol 2007;79:1349–56. herpesvirus 6) isolated from patients with lymphoproliferative dis- 31. Razzaque A, Williams O, Wang J, Rhim JS. Neoplastic transforma- ease. J Natl Cancer Inst 1987;79:933–41. tion of immortalized human epidermal keratinocytes by two HHV-6 14. Ablashi DV, Salahuddin SZ, Josephs SF, Balachandran N, Krueger GR, DNA clones. Virology 1993;195:113–20. Gallo RC. Human herpesvirus-6 (HHV-6) (short review). In Vivo 1991;5: 32. Thomson BJ, Dewhurst S, Gray D. Structure and heterogeneity of the 193–9. sequences of human herpesvirus 6 strain variants U1102 and Z29 15. Braun DK, Dominguez G, Pellett PE. Human herpesvirus 6. Clin and identification of human telomeric repeat sequences at the Microbiol Rev 1997;10:521–67. genomic termini. J Virol 1994;68:3007–14. 16. Kondo K, Kondo T, Okuno T, Takahashi M, Yamanishi K. Latent 33. Kashanchi F, Thompson J, Sadaie MR, et al. Transcriptional activa- human herpesvirus 6 infection of human monocytes/macrophages. tion of minimal HIV-1 promoter by ORF-1 protein expressed from the J Gen Virol 1991;72:1401–8. SalI-L fragment of human herpesvirus 6. Virology 1994;201:95–106. 17. Ogata M. Human herpesvirus 6 in hematological malignancies. J Clin 34. Kashanchi F, Araujo J, Doniger J, et al. Human herpesvirus 6 (HHV-6) Exp Hematop 2009;49:57–67. ORF-1 transactivating gene exhibits malignant transforming activity 18. Crawford JR, Santi MR, Cornelison R, Sallinen SL, Haapasalo H, and its protein binds to p53. Oncogene 1997;14:359–67. Macdonald TJ. Detection of human herpesvirus-6 in adult central 35. Harris NL, Jaffe ES, Diebold J, et al. The World Health Organization nervous system tumors: predominance of early and late viral anti- classification of neoplastic diseases of the hematopoietic and lym- gens in glial tumors. J Neurooncol 2009;46:37–42. phoid tissues. Report of the Clinical Advisory Committee meeting, Airlie 19. Duncan CG, Leary RJ, Lin JC, et al. Identification of microbial DNA in House, Virginia, November, 1997. Histopathology 2000;36:69–86. human cancer. BMC Med Genomics 2009;2:22. 36. Sturtz FG, Cioffi L, Wittmer S, et al. Tetracycline-regulatable

4720 Clin Cancer Res; 16(19) October 1, 2010 Clinical Cancer Research

Involvement of HHV-6B via DR7 Protein in HL

expression vectors tightly regulate in vitro gene expression of Raab-Traub N. LMP1 signaling and activation of NF-κBinLMP1 secreted proteins. Gene 1998;221:279–85. transgenic mice. Oncogene 2006;25:288–97. 37. Pradeau K, Bordessoule D, Szelag JC, et al. A reverse transcription- 44. Flamand L, Stefanescu I, Ablashi DV, Menezes J. Activation of the nested PCR assay for HHV-6 mRNA early transcript detection after Epstein-Barr virus replicative cycle by human herpesvirus 6. J Virol transplantation. J Virol Methods 2006;134:41–7. 1993;67:6768–77. 38. Krueger GR, Guenther A, Knuefermann R, et al. Human herpesvirus-6 45. Guttridge DC, Albanese C, Reuther JY, Pestell RG, Baldwin AS. (HHV-6) in Hodgkin's disease: cellular expression of viral antigens as NF-κB controls cell growth and differentiation through transcriptional compared to oncogenes met and fes, tumor suppressor gene product regulation of cyclin D1. Mol Cell Biol 1999;19:5785–99. p53, and interleukins 2 and 6. In Vivo 1994;8:501–16. 46. Baeuerle PA, Henkel T. Function and activation of NF-κBinthe 39. Collot-Teixeira S, Bass J, Denis F, Ranger-Rogez S. Human immune system. Annu Rev Immunol 1994;12:141–79. tumor suppressor p53 and DNA viruses. Rev Med Virol 2004; 47. Chien ML, Hammarskjold ML. Epstein-Barr virus latent membrane 14:301–19. protein (LMP1) induces specific NFκB complexes in human 40. Takemoto M, Mori Y, Ueda K, Kondo K, Yamanishi K. Productive T-lymphoid cells. Virus Res 2000;67:17–30. human herpesvirus 6 infection causes aberrant accumulation of 48. Zhao P, Lu Y, Liu L, Zhong M. Aberrant expression of ID2 protein and p53 and prevents apoptosis. J Gen Virol 2004;85:869–79. its correlation with EBV-LMP1 and P16(INK4A) in classical Hodgkin 41. Oster B, Kofod-Olsen E, Bundgaard B, Hollsberg P. Restriction of lymphoma in China. BMC Cancer 2008;8:379. human herpesvirus 6B replication by p53. J Gen Virol 2008;89: 49. Skinnider BF, Mak TW. The role of cytokines in classical Hodgkin 1106–13. lymphoma. Blood 2002;99:4283–97. 42. Oster B, Bundgaard B, Hupp TR, Hollsberg P. Human herpesvirus 6B 50. Mathas S, Hinz M, Anagnostopoulos I, et al. Aberrantly expressed induces phosphorylation of p53 in its regulatory domain by a CK2- c-Jun and JunB are a hallmark of Hodgkin lymphoma cells, and p38-independent pathway. J Gen Virol 2008;89:87–96. stimulate proliferation and synergize with NF-κB. EMBO J 2002;21: 43. Thornburg NJ, Kulwichit W, Edwards RH, Shair KH, Bendt KM, 4104–13.

www.aacrjournals.org Clin Cancer Res; 16(19) October 1, 2010 4721

Involvement of Human Herpesvirus-6 Variant B in Classic Hodgkin's Lymphoma via DR7 Oncoprotein

Aurélie Lacroix, Sophie Collot-Teixeira, Laurent Mardivirin, et al.

Clin Cancer Res Published OnlineFirst September 21, 2010.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-10-0470

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://clincancerres.aacrjournals.org/content/early/2010/09/16/1078-0432.CCR-10-0470. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.