ALBERT

Description: Abstract 1929 Poster Board I-952 Epstein-Barr virus (EBV) can infect not only B cells but also T or NK cells uncommonly and causes lymphoid malignancies, such as extranodal NK/T-cell lymphoma nasal type (ENKL), aggressive NK-cell leukemia, and EBV-positive T/NK-cell lymphoproliferative disease (EBV-T/NK-LPD), which is also known as chronic active EBV infection. However, why and how EBV infects T or NK cells and the mechanism of action responsible for these EBV-induced malignancies have not been elucidated to date. To clarify the molecular mechanism underlying development of EBV-T/NK-LPD, we focused on costimulatory receptor CD137, which is expressed on the surface of activated T cells and plays a pivotal role in their proliferation, survival, and differentiation. We investigated CD137 expression on the surface of EBV-infected T/NK cells (EB-T/NK cells) by flow cytometry. First, three EBV-positive T and NK cell lines, SNT8, SNK6, and SNT16, were obtained for examination. These cell lines had been established from primary lesions of ENKL patients (SNT8 and SNK6) and peripheral blood of an EBV-T/NK-LPD patient (SNT16). CD137 expression was confirmed on the cell surface of these cells, whereas the EBV-negative T and NK cell line, Jurkat and KHYG1 cells, respectively, were negative for CD137. Next, we investigated expression on the surface of EB-T/NK cells derived from EBV-T/NK-LPD patients. EBV-T/NK-LPD was diagnosed according to the following criteria: presence of persistent infectious mononucleosis-like symptoms, elevation of EBV-DNA titer in the peripheral blood (PB), and detection of EBV-infected T or NK cells. To detect the infected cells, we isolated peripheral mononuclear cells and divided them into CD19-, CD4-, CD8-, or CD56-positive fractions using antibody-conjugated magnetic beads. Next, we measured the EBV-DNA titer of each fraction by quantitative RT-PCR. Nine patients (aged 8–41 years; 4 male, 5 female; 4 T and 5 NK cell types) were diagnosed with EBV-T/NK-LPD. Then, we examined surface CD137 expression of the infected cells of each patient. Expression was detected in 7 of 9 patients. Control cells (PB mononuclear cells of a healthy donor, who was negative for EBV-DNA titer in the PB) did not express the molecule. We also examined transcription of CD137 mRNA by RT-PCR assay and detected it in all the 12 EB-T/NK-cell samples described above. From these results we concluded that CD137 expression was induced at the level of both mRNA and protein in EB-T/NK cells. To investigate the molecular mechanism of CD137 overexpression in EBV-T/NK cells, we examined the influence of viral proteins on CD137 expression. EB-T/NK cells express EBV-encoded proteins, including LMP1, LMP2A, LMP2B, and EBNA1 (latency type 2). We cotransfected expression plasmids for these proteins with a luciferase reporter plasmid containing the CD137 gene promoter in Jurkat cells and performed a luciferase assay. LMP1 significantly upregulated the CD137 promoter activity, although the other molecules did not. Furthermore, in a transient expression assay of these viral proteins using Jurkat cells, transcription of endogenous mRNA of CD137 was upregulated only in the LMP1 transfectant. These results indicate that LMP1 may transactivate CD137 transcription and expression in EBV-T/NK cells. Next, we investigated the role of CD137 in developing EBV-T/NK-LPD. We cultured the above-mentioned CD137-expressing EBV-T/NK cells on CHO cells that stably express human CD137L on the cell surface. NF-ĸB activation was detected in CD137-positive EBV-T/NK cells that were cocultured with CD137L-expressing CHO cells. We confirmed that both p50 and p52 translocated to the nucleus, indicating that both canonical and non-canonical pathways for NF-ĸB activation were activated downstream of CD137. Finally, we investigated the role of CD137-mediated NF-ĸB activation in the development of EBV-T/NK-LPD. We cocultured EB-T/NK cells on CHO-wt or CHO-CD137L with VP-16 for 48 h and determined apoptosis by measuring DiCO6 uptake. We noted that stimulation of CD137 significantly suppressed VP-16-induced apoptosis of these cells. Together, these results indicate that EBV-infected T/NK-cells express CD137 on the cell surface, which may be induced by LMP1 and activate the anti-apoptotic intracellular signaling pathway through NF-ĸB activation. This pathway may contribute to immortalization of the infected cells and development of EBV-T/NK-LPD. Disclosures: No relevant conflicts of interest to declare.

Description: Introduction Epstein–Barr virus (EBV) infects B cells and rarely T or NK cells, causing EBV-positive T/NK-cell lymphoproliferative neoplasms (EBV-T/NK-neoplasm), such as extranodal NK/T-cell lymphoma, aggressive NK-cell leukemia, and EBV-positive T- or NK-cell lymphoproliferative disorders (EBV-T/NK-LPDs). EBV-T/NK-LPDs are fatal disorders presenting sustained inflammation, such as infectious mononucleosis-like symptoms, hypersensitivity to mosquito bites, or hydroa vacciniforme-like eruption accompanied by clonal proliferation of EBV-infected T or NK cells. EBV infects B cells making them immortal and resulting in B-cell lymphomas. However, why and how EBV infects T or NK cells and the mechanism of action responsible for the development of these EBV-induced malignancies has not been elucidated yet. Activation-induced cytidine deaminase (AID) is essential for somatic hypermutation and class switch recombination of immunoglobulin genes. Deregulated AID expression acts as a genomic mutator leading to the development of B-cell lymphoma. In addition, EBV infection induces AID expression in B cells. These findings indicate that AID has a role in EBV-induced lymphomagenesis in B cells. However, of interest, AID transgenic mice developed T-cell lymphoma (J. Exp. Med., 197, p.1173, 2003). Recently, Nakamura et al. reported that AID expression was upregulated in the peripheral blood mononuclear cells (PBMCs) of EBV-T/NK-LPDs patients (Eur. J. Dermatol., 21, p.780, 2011). Therefore, we hypothesized that EBV infection induces AID expression in T or NK cells contributing to the tumor development. Objectives We designed this study to investigate AID expression and its contribution to EBV-T/NK-neoplasms development. Materials and Methods EBV-positive T- and NK-cell lines, SNT8, SNK6, SNT13, SNT15, and SNT16, were examined. The EBV-negative T-cell line HPB-ALL, Jurkat and human primary T cells derived from cord blood were used as the negative controls. Clinical samples were obtained from EBV-T/NK-LPDs patients who were diagnosed according to the previously described criteria (Blood, 119, p.673, 2012). To detect and isolate EBV-infected cells, T and NK cells were separated using magnetic beads from PBMCs. AID expression in tissue lesions were examined in clinical samples and xenograft models of EBV-T/NK-LPDs generated by transplantation of PBMCs from the EBV-T/NK-LPDs patients to NOD/Shi-scid, IL-2R γKO mice. For in vitro EBV infection, EBV was prepared from the culture medium of B95-8 cells and added to HPB-ALL or Jurkat cells (Proc. Natl. Acad. Sci., 100, pp.7836-40, 2003). To determine AID-induced mutation of the gene, DNA was extracted from SNT13, SNK6, or EBV-infected Jurkat cells, and c-myc gene was cloned by TA cloning, which was then sequenced. We compared the number of the mutation with that of Jurkat cells. Results We detected AID expression in EBV-positive T- or NK-cell lines by RT-PCR, western blotting, and immunological staining using confocal microscopy, whereas it was not detected in the control. Furthermore, we validated the results in EBV-infected T or NK cells derived from 12 EBV-T/NK-LPDs patients (infected cell types: CD4, 5; CD8, 3; and CD56, 4). Quantitative RT-PCR demonstrated that AID expression was upregulated in EBV-infected T or NK cells compared with the control. Expression was confirmed by immunological staining using confocal microscopy in 5 patients. We also detected AID expression by histopathological staining in EBV-infected cells in lesions of 4 patients and the EBV-T/NK-LPDs xenograft models. Subsequently, we examined the role of EBV for AID expression in T cells. It was demonstrated that AID expression was induced by in vitro EBV infection in the T-cell line HPB-ALL and primary T cells. Moreover, it was demonstrated by the luciferase assay that the viral protein LMP1 upregulated AID promoter activity in HPB-ALL cells. Finally, we found that SNT13, SNK6, and EBV-infected Jurkat cells had increased number of the mutation of c-myc gene compared with Jurkat cells. Conclusions We previously reported that EBV infection of T or NK cells enhances survival of the infected cells by the activation of NF-κB. The results of the present study suggest that EBV contributes to the development of EBV-T/NK-neoplasms through NF-κB-induced cell survival and AID-induced mutagenesis leading to clonal evolution and expansion. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 4953 Chronic active Epstein-Barr virus infection (CAEBV) is a rare and fatal disorder. In Japan and East Asia, most CAEBV patients are the T- or NK-cell-infected type and they have been resistant to the current chemotherapies. L-asparaginase (L-asp) is a reagent that has a well-known effect on extranodal NK-/T-cell lymphoma nasal type (ENKL) even as a single reagent. CAEBV is one of the EBV-positive T- or NK-cell lymphoproliferative diseases (EBV-T/NK-LPD) as is ENKL. In addition, L-asp is not influenced by P-glycoprotein, which is expressed in EBV-infected T-or NK-cells of CAEBV (our unpublished data). In this pilot study, therefore, we investigated its effects on T- and NK-cell type of CAEBV. Adult patients of CAEBV without severe organ dysfunction were enrolled. CAEBV was diagnosed according to the following criteria: persistent infectious mononucleosis-like symptoms, elevation (〉1 X102 copies/μg DNA) in peripheral blood EBV-DNA titer (pEBV-DNA), and the presence of EBV-infected T- or NK-cells. To detect infected cells, we isolated peripheral mononuclear cells and divided them into CD19-, CD4-, CD8-, or CD56-positive fractions using antibody-conjugated magnetic beads. EBV-DNA of each fraction was quantified using a real-time quantitative polymerase chain reaction (PCR) assay. The treatment protocol involved 7 administrations of 6000 U/m2 L-asp every other day. The response was defined by the decrease of pEBV-DNA 1 month after treatment completion. We also performed quantification of asparagine synthetase (AS) in EBV-infected cells of the patients. The results were summarized in the table. Between February 2010 and September 2010, 5 females were enrolled. Two patients were previously treated with the regimen comprising cyclosporine A, prednisolone, and VP16. One patient had EBV-positive cell infiltration of muscles with an elevation of LDH and CK. The mean titer of EBV-DNA in peripheral whole blood was 1.2 × 105 copies/μg DNA. Three patients completed the treatment. pEBV-DNA decreased in 1 patient with a reduction rate of 0.09 with improvement of clinical symptoms, but increased 2.7-fold in another patient and remained almost unchanged in the other. The response rate was 20% (1/5). Of the patients whose treatment was discontinued, one showed progression of the nasal lesion with pEBV increasing 1.5-fold, and the other had a dystonic attack on day 11, and the 2 remaining administrations were stopped although the muscle lesions were improved. Several adverse events (AE) were detected, including liver dysfunction (grade 2 and 3) in 2 patients and neutropenia (grade 3) in 1 patient. One patient had a dystonic attack as described earlier. The brain MRI showed no lesion in the central nervous system that could have caused this attack. Because the patient had been diagnosed with dystonia before she developed CAEBV, the attack was considered not to be directly attributable to L-asp. We examined AS mRNA levels in the EBV-infected cells. mRNA level was low in the patient who achieved pEBV-DNA decrease. As shown in the table, however, we could not find a significant relationship between the effect and AS levels. In general, the effect of L-asp on CAEBV was limited and the rate of AE was high. However, L-asp may have responses in some patients. CAEBV has a diverse phenotype of clinico-pathological findings and associated symptoms. Treatment for CAEBV needs to be planned and evaluated according to the subtype of the disease. Disclosures: No relevant conflicts of interest to declare.

Description: Introduction Epstein-Barr virus (EBV) is a double-stranded DNA virus that infects 〉95% of the human population and is associated with a substantial risk of cancer development. Most infections in children and adolescents are asymptomatic or result in infectious mononucleosis; however, in some patients, EBV is associated with various hematological malignancies including Burkitt lymphoma, diffuse large B-cell lymphoma (DLBCL), and extranodal NK/T-cell lymphoma. EBV infection is also present in a portion of epithelial cell neoplasms such as gastric cancer and nasopharyngeal carcinoma. Despite the large population risk of cancer associated with EBV, it is poorly understood why only a small subset of EBV-infected individuals develop neoplasms, while others do not. Patients and Methods We designed a target enrichment system to capture several EBV strains including the Akata strain, which is responsible for the majority of EBV infections in Japan. We analyzed the genomes of EBV strains in 139 patients with various EBV-associated diseases and 17 EBV-positive cell lines. Next-generation sequencing reads were aligned to the Akata reference genome to analyze nucleotide variations, copy number alterations, and structural variations including sequence insertions in the human genome. The institutional review board of Nagoya University Graduate School of Medicine approved this study. Results We identified a median of 645 single nucleotide variants (SNVs) in the EBV genomes, 78% of which affected coding sequences. SNVs in coding sequences were significantly biased toward synonymous variants, suggesting negative selection pressure. The SNVs detected in noncoding sequences were enriched in two evolutionarily conserved viral noncoding RNAs (EBER1 and EBER2), particularly in the PAX5-binding domain of EBER2. However, most SNVs identified in the EBV genome do not seem to affect the development of neoplasms, as hierarchical clustering of EBV genomes from neoplastic and non-neoplastic diseases based on SNVs revealed no significant association between the EBV strain and disease type. In addition to SNVs, we identified frequent intragenic deletions in the EBV genomes of patients with EBV-positive DLBCL (10/14, 71%), extranodal NK/T-cell lymphoma (10/23, 43%), chronic active EBV infection (27/77, 35%), and other EBV-associated neoplasms (2/7). Such deletions were also identified in several EBV-associated cell lines (6/17), but not in non-neoplastic diseases such as infectious mononucleosis (0/4) and post-transplant lymphoproliferative disorders (0/14), suggesting a unique role of these mutations in the neoplastic proliferation of EBV-infected cells. Frequent deletions were detected in BamHI A rightward transcripts microRNA clusters (31/156), which suppress viral transcription factors (BZLF1 and BRLF1) required for the lytic reactivation of EBV. Deletions also were associated with several genes essential for virus production (20/156). These observed deletions are thought to upregulate lytic cycle-associated genes, some of which benefit neoplasms by inducing genomic instability and immune escape and mitigate cell damage caused by the production of viral particles. In fact, deletion of one essential gene, BALF5, resulted in upregulation of the lytic cycle and promotion of lymphomagenesis in a xenograft model. Discussion Although the essential roles of several latency-associated genes, such as LMP-1 and EBNA-2, in EBV-mediated immortalization and transformation of human lymphocytes have long been discussed, our finding raises the possibility that lytic cycle-associated genes also contribute to lymphomagenesis. This agrees with reports that lytic cycle-associated genes are expressed in Burkitt lymphoma, DLBCL, and chronic active EBV infection, and that BZLF1-deficient lymphoblastoid cells exhibit significantly impaired tumorigenicity in mice. In addition, essential gene deletions lead to the protection of EBV-infected cells from lysis. Further studies are warranted to exploit these findings for the design of novel therapeutics for EBV-associated neoplasms. Disclosures Kiyoi: Sumitomo Dainippon Pharma Co., Ltd.: Research Funding; Novartis Pharma K.K.: Research Funding; Phizer Japan Inc.: Research Funding; Sanofi K.K.: Research Funding; Kyowa Hakko Kirin Co., Ltd.: Research Funding; Celgene Corporation: Research Funding; Eisai Co., Ltd.: Research Funding; Astellas Pharma Inc.: Research Funding; Takeda Pharmaceutical Co., Ltd.: Research Funding; Otsuka Pharmaceutical Co., Ltd.: Research Funding; Chugai Pharmaceutical Co., Ltd.: Research Funding; Nippon Shinyaku Co., Ltd.: Research Funding; FUJIFILM Corporation: Research Funding; Zenyaku Kogyo Co., Ltd.: Research Funding; Bristol-Myers Squibb: Honoraria. Nakamura:Roche/Chugai,: Research Funding; Kyowa-Kirin: Research Funding.

Description: Abstract 1998 Epstein-Barr virus (EBV) can infect not only B cells but also rarely T or NK cells and causes EBV-positive T/NK-cell lymphoproliferative disease (EBV-T/NK-LPD), such as extranodal NK/T-cell lymphoma (ENKL), aggressive NK-cell leukemia, and chronic active EBV infection (CAEBV). However, why and how EBV infects T or NK cells and the mechanism of action responsible for the development of these EBV-induced malignancies have not been elucidated to date. Furthermore, optimal chemotherapy for these, especially for CAEBV, has not been established and the prognosis of EBV-T/NK-LPD remains very poor. NF-kB is a transcription factor that mediates anti-apoptotic molecular signaling and promotes proliferation of cancer cells. NF-kB is known to be constitutively activated in some hematological malignancies; the proteasome inhibitor bortezomib suppresses its activity and is used clinically as an anticancer reagent. To clarify the molecular mechanism underlying the development of EBV-T/NK-LPD, we focused on NF-kB in four EBV-positive T- and NK-cell lines, SNT8, SNT15, SNT16, and SNK6, which had been established from primary lesions of ENKL patients (SNT8 and SNK6) and the peripheral blood (PB) of a CAEBV patient (SNT15, 16), respectively. In these cells, it was demonstrated that p50, p52, RelA, and RelB—the components of NF-kB—existed constitutively in the nucleus. On the other hand, the EBV-negative T-cell lines, Jurkat and Molt4, and an NK-cell line, KHYG1, were negative for nuclear localization of these molecules. Elevated nuclear NF- kB–DNA binding activity was demonstrated by electrophoretic mobility shift assay (EMSA), using oligonucleotides encoding an NF-kB-binding sequence as a probe. Supershift EMSA revealed that NF- kB–DNA binding complexes in these cells involved p50, p52, and RelA. In addition, reporter assay using NF-kB-dependent luciferase reporter gene consisting of an NF-kB-binding site as a promoter demonstrated constitutive activation of NF-kB in SNT8 cells, whereas not in Jurkat cells. From these results, we concluded that NF-kB was constitutively activated through the canonical and noncanonical pathways in these EBV-infected T- or NK-cell (EBV-T/NK cell) lines. Next, we investigated NF-kB activation in EBV-T/NK cells derived from CAEBV patients. CAEBV was diagnosed according to the following criteria: presence of persistent infectious mononucleosis-like symptoms, elevation of EBV-DNA titer in PB, and detection of EBV-infected T or NK cells with clonal proliferation. To detect infected cells, we isolated peripheral mononuclear cells and divided them into CD19-, CD4-, CD8-, or CD56-positive fractions using antibody-conjugated magnetic beads. The EBV DNA of each fraction was quantified using a real-time quantitative polymerase chain reaction assay. Ten patients (aged 8–72 years; 4 male, 6 female; infected cell types CD4:3, CD8:3, γδ:1, and CD56:3) were diagnosed as CAEBV. Then, we examined NF-kB activation in these cells. Similar to the cell lines, immunoblotting and EMSA demonstrated evidence of constitutive activation of NF-kB such as nuclear localization and DNA binding of p50, p52, and RelA in EBV-T/NK cells from CAEBV patients. To investigate the molecular mechanism of NF-kB activation in EBV-T/NK cells, we examined the influence of viral proteins expressed in EBV-T/NK cells, including LMP1, LMP2A, LMP2B, and EBNA1 (latency type 2). We performed luciferase reporter assay using expression vectors for these proteins in Jurkat cells. LMP1, most significantly, and LMP2A, to a lesser extent, upregulated NF-kB-dependent reporter gene expression in Jurkat cells, whereas the other viral proteins did not, suggesting that LMP1 and LMP2A mediated NF-kB activation in T/NK cells. Finally we investigated the contribution of NF-kB to the development of EBV-T/NK-LPD. Bortezomib, in a dose-dependent manner, inhibited translocation of p50, p52, and RelA to the nucleus and reduced the expression of Bcl-XL, which was a transcriptional target of NF-kB in SNT8, SNK6, and CAEBV patients’ cells. In addition, it was also confirmed that bortezomib (0.5nM -5nM) suppressed the survival of and induced apoptosis of EBV-T/NK cells derived from 5 CAEBV patients. These results suggest that constitutive activation of NF-kB contributes to the development of EBV-T/NK-LPD and that the NF-kB signaling pathway can be an attractive molecular target for treatment. Disclosures: No relevant conflicts of interest to declare.

Description: Background and aims Chronic active Epstein-Barr virus infection (CAEBV) is classified into T- or NK-cell neoplasms in the new WHO classification revised in 2017. Allogeneic stem cell transplantation (allo-HSCT) has recently been reported to be an effective treatment for this disorder. Conversely, effects of chemotherapies on CAEBV have not yet been examined in a large number of patients. To clarify clinical features and the current state of chemotherapies for CAEBV under the new definition of the disease, we performed a nationwide survey in Japan. Methods Questionnaires were sent to all educational hospitals certified by the Japanese Society of Hematology and the Japanese Pediatric Society. Subjects were patients newly diagnosed with CAEBV between January 2003 and March 2016. CAEBV was diagnosed according to criteria suggested by the Research group of Measures against Intractable Diseases by Ministry of Health, Labour and Welfare of Japan in 2016: (1) elevated EBV DNA load in peripheral blood (PB) (〉102.5 copies/μg DNA); (2) detection of EBV infection in T or NK cells from the affected tissues or PB; (3) systemic inflammatory symptoms (such as fever, lymphadenopathy, liver dysfunction, progressive skin lesions, vasculitis, and uveitis) persisting for 〉3 months; and (4) exclusion of other possible diagnoses, such as primary EBV infection, autoimmune disease, immunodeficiencies, and lymphomas. Patients who fulfilled (1)-(4) were diagnosed with CAEBV. These criteria were established based on those proposed by Okano et al.(Am J Hematol. 2005;80,p64) and Kimura et al.(Blood. 2012;119,p673) and were compatible with the definition of CAEBV described by WHO in 2017. The disease activity was defined according to the previous reports (Blood. 2012;119,p673 and BMT. 2016;51,p879) as follows: positive for fever, ALT level elevation, vasculitis, progressive skin lesions, or uveitis. Effects of treatments were evaluated as follows: partial response (PR), partial resolution of disease activity; complete response (CR), complete resolution of disease activity with EBV load in PB remaining high (〉102.5 copies/μg DNA, which is the upper limit for healthy people); virological CR (vCR), CR with a significant decrease in the EBV DNA load in PB (