Investigating the contribution of EBV infection and the roles of miR-BARTs in the development of NPC using humanized mouse models

Abstract

Undifferentiated nasopharyngeal carcinoma (NPC) is a type of head-and-neck cancer, which is nearly 100% associated with Epstein-Barr virus (EBV) infection and heavily infiltrated with leukocytes. Moreover, expressions of microRNAs encoded from the EBV-BamHI-A rightward transcripts (miR-BARTs) can constitute up to 45% of total microRNAs in NPC cells. All these imply that the development of NPC may involve the interplay of EBV infection and the tumour infiltrating leukocytes (TILs). Yet, investigation into the roles of the viral-host interactions in NPC pathogenesis is greatly hampered by the lack of reliable experimental models that can recapitulate the tumour microenvironment (TME) with heavy TILs in NPC. Extensive efforts to establish humanized mouse models for NPC and to investigate into the roles of miR-BARTs in EBV-asscociated NPC pathogenesis were carried out in this study. In vitro experiments mimicking the in vivo interactions with tumour-associated macrophages (TAMs) and 3-dimensional (3-D) growth preliminarily showed that these factors may play a role in regulating expressions of miR-BARTs in EBV-infected NPC. Paracrine effects of M2-like macrophages were found to positively regulate expressions of miR-BARTs in NPC cells. Expression levels of miR-BARTs were also found to be increased when NPC cells were grown in 3-D spheroids rather than 2-D monolayer. In order to understand the roles of miR-BARTs in NPC pathogenesis, a new EBV-infected NPC cell system consisting a pair of NPC cell lines re-infected with NPC-strain of EBV (M81) or its mutant with the BART clusters deleted, namely NPC43EBV-veM81WT and NPC43EBV-veM81∆BART respectively, was established. On the other hand, humanized mouse model offers a platform to study the interactions between human immune cells and EBV-infected NPC cells. Reconstitution of human immune system in NOD/SCID/IL-2Rγ-/- (NSG) mice was achieved via engraftment of CD34+ hematopoietic stem cells (HSCs). Flow cytometry analysis of peripheral blood from humanized NSG mice showed presence of human immune components, including monocytes, NK cells, B cells, and T cells. NPC cells propagated in humanized NSG mice successfully grew into tumours. EBV-infected NPC cells grew into bigger tumours than EBV-uninfected NPC cells, which may demonstrate a role of EBV in promoting tumorigenicity in NPC. NPC43EBV-veM81WT also grew into bigger tumours than NPC43EBV-veM81∆BART, which may also suggest that miR-BARTs provide growth advantages for NPC in vivo. EBV-infected NPC grown in humanized NSG mice was histologically similar to human EBV-infected NPC biopsies, indicating that humanized NSG mice may provide a suitable platform to model the TME of EBV-infected NPC. Preliminary data showed that EBV+ve NPC was more active in proliferation and infiltrated with higher proportion of immune cells including regulatory T lymphocytes, when compared to the EBV-ve counterpart. Presence of human TILs in NPC also appeared to have a positive regulatory effects on expressions of miR-BARTs in EBV-infected NPC.