PROJECT SUMMARY/ABSTRACT Epstein-Barr virus is an important cause of AIDS-related lymphomas (ARLs) world-wide, including Hodgkin lymphomas (HLs) that are driven by the EBV latent membrane proteins LMP1 (a CD40 mimic) and LMP2A (a BCR mimic), in conjunction with a supportive tumor microenvironment. Although the EBV protein EBNA2 (which mimics notch signaling) is required for EBV-induced transformation of B cells in vitro, cells that express EBNA2 have the “type III” form of viral latency, which is highly immunogenic. Thus, most EBV-infected human lymphomas (including ARLs in cART-treated patients) do not express EBNA2. However, there is currently no in vivo or in vitro model available to study how EBV infection can cause lymphomas in the absence of EBNA2 expression, and the only EBV+ HL cell line, L591, has switched to type III latency. Increasing evidence suggests that HIV infection collaborates with EBV to induce lymphomas not only by inducing immunosuppression, but by creating a pro-inflammatory microenvironment that promotes tumors with more stringent forms of EBV latency (such as the “type II” form that occurs in HLs) that do not express EBNA2. Our lab has developed a novel cord blood-humanized mouse model which provides a highly supportive CD4 T cell-rich environment that allows certain EBV mutants previously considered “non-transforming” in vitro to induce lymphomas in vivo. Our new exciting preliminary results reveal that a naturally occurring EBNA2-deleted EBV strain (P3HR1) that is completely non-transforming in vitro causes Hodgkin-like lymphomas that express LMP1 and LMP2A in this model. Furthermore, our preliminary results suggest that similar to human HLs, lymphomas induced by EBNA2- deleted EBV are heavily infiltrated with collagen, express the pro-survival collagen-stimulated receptor tyrosine kinase DDR1, and have activated PDGFRA and notch-1 signaling. Thus, we believe we have created the first in vivo model for EBV-induced Hodgkin lymphoma. In this proposal, we will dissect the roles of the EBV proteins LMP1 and LMP2A, as well as the tumor microenvironment, in driving lymphomas induced by EBNA2-deleted EBV in humanized mice. We hypothesize that both LMP1 and LMP2A, as well as CD4 T cell- and collagen- derived signals in the microenvironment, are required for the growth of these lymphomas. In Aim 1, we will compare the phenotypes of wild-type (WT) versus EBNA2-deleted (Akata strain) EBV in humanized mouse models, and identify paracrine and/or autocrine signaling pathways (derived from T cells, B cells and/or collagen) that compensate for loss of EBNA2 expression both in vivo and in L591 HL cells in vitro. In Aim 2, we will define the roles of LMP1 and LMP2A on tumor cell growth, and in regulating the tumor microenvironment, in both the CBH model in vivo and in L591 cells in vitro. In Aim 3, we will determine whether inhibiting notch, PDGFRA and/or DDR1 signaling attenuates the growth of lymphomas induced by WT virus and/or EBNA2-deleted virus in humanized mice or in L591 cells in vitro. The results of these studies should provide key insights into the mechanism(s) by which LMP1 and LMP2A promote AIDS-related Hodgkin lymphomas, and elucidate how the tumor microenvironment cooperates with type II latent EBV infection to induce human lymphomas in vivo when EBNA2 expression is shut off.

PUBLIC HEALTH RELEVANCE Epstein-Barr virus (EBV) is an important cause of AIDS-related lymphomas, especially AIDS-related Hodgkin lymphomas (HLs). EBV-infected HLs have a form of viral latency (type II) that cannot transform B cells in vitro, and there are no in vivo models available to study how the EBV proteins, LMP1 and LMP2A, induce HLs in humans. We have created a new humanized mouse model in which EBV infection collaborates with the tumor microenvironment to cause HL-like lymphomas. In this proposal, we will define mechanisms by which LMP1, LMP2A, and growth promoting signals derived from the tumor microenvironment drive these lymphomas, and determine if blocking these growth signals inhibits tumor growth.