**1. Introduction**

Epstein–Barr virus (EBV) is a large double-stranded DNA gammaherpesvirus with about 170 kilobases in its genome, encoding over 100 open reading frames (ORFs). EBV accounts for about 1% of all cancer cases worldwide. This complex virus is ubiquitous in the human population, establishing a lifelong latent infection in 90% of people by adulthood [1,2]. The viral strains can be divided into two subgroups, type 1 and type 2, which are broadly similar and designated by differences in their nuclear antigens [3,4]. Primary infection is either asymptomatic, experienced as a non-specific infection, or the cause of infectious mononucleosis, with the latter more likely if exposure occurs during adolescence or later [5]. EBV targets human B cells after being transmitted through the oral epithelium via the saliva of an infected individual, establishing latency and allowing the viral genome to persist.

EBV is linked to the development of several human cancers. It was first identified in a Burkitt lymphoma sample [6], and is now known to be a cause of Hodgkin's lymphoma [7], nasopharyngeal carcinoma [8,9], and gastric carcinoma [10]. EBV infection is also linked to autoimmune disorders, such as multiple sclerosis [11] and systemic lupus erythematosus [12], which are likely tied to EBV-driven immune dysregulation [13]. The cancers associated with EBV are linked to their expression of EBV oncogenes, including Epstein–Barr virus nuclear antigen 1 (EBNA1) and latent membrane proteins 1 and 2 (LMP1 and LMP2) [14]. The latent viral oncoproteins of EBV are important cancer drivers and are implicated in directly contributing to EBV-associated malignancies [15–17].

EBNA1 is important in maintaining the viral genome and is required for EBV latency and associated transformation. LMP1 and LMP2 were discovered to colocalize in the membranes of latently infected lymphocytes [18], and these oncoproteins contribute to cancer progression via diverse signaling pathways [19]. LMP1 interacts with tumor necrosis factor receptor (TNFR)-associated factors (TRAFS) to drive nuclear factor-κB (NF-κB), mitogen-activated protein kinase (MAPK), and phosphatidylinositol 3-kinase (PI3K) pathways [20]. LMP2 mimics the B cell receptor, sending survival signals to B cells without the need for antigen stimulation [21]. The LMP2 gene produces LMP2A and LMP2B, of which LMP2A has an additional 119 amino acids at the N-terminus.

There are no approved vaccines available to prevent initial infection by EBV, and clinical trials of EBV vaccine candidates have had limited success. The target that progressed furthest along in the clinic was a recombinant subunit gp350 prophylactic vaccine adjuvanted with aluminum hydroxide and 3-O-desacyl-4- -monophosphoryl lipid A (AS04) which was tested in a phase 2 trial. The study reported that it statistically decreased the incidence of infectious mononucleosis, but this vaccine did not reduce infections by the virus, despite generating high-titer antibody responses in vaccine recipients [22]. Future vaccines against EBV can further explore the numerous other glycoproteins involved in EBV entry and the latent proteins essential for maintaining the virus [23].

EBV is a viable target for therapeutic approaches to treating cancer. Cellular immune responses are particularly important in targeting malignant cells, and they have been exploited in specific cancer immunotherapies [24,25]. It would be a major advantage for such approaches if they would drive both CD4 T cell responses and induce functional CD8 T cell responses that could clear EBV-infected targets. Prior vaccine approaches particularly lacked potent induction of CD8 cellular immunity.

Newer Synthetic Consensus (SynCon) DNA vaccines, combined with adaptive electroporation (EP), have demonstrated safety, as well as the potent induction of antibodies, T helper responses, and CD8 effector T cells, in multiple clinical trials. Clinical efficacy has been reported in the context of immunotherapy for human papillomavirus (HPV)-driven neoplasia, and clinical regressions with clearance have been described in early studies that use a combination approach involving engineered HPV nuclear gene targets and checkpoint inhibitor therapy with PD-1. Specifically, a therapeutic DNA vaccine targeting HPV E6/E7 antigens from the HPV 16 and 18 strains has shown a positive impact in patients when this vaccine was delivered by Cellectra adaptive EP in a phase 2b trial for the treatment of cervical intraepithelial neoplasia [26]. Importantly, this vaccine induced potent CD8 T cells that infiltrated the tumor and caused the lesions to regress, resulting in both histopathological regression and viral clearance in 40% of treated patients. Similar data has been reported impacting HPV-driven head and neck cancers in a preliminary report [27], where a similar genetically-adjuvanted HPV DNA vaccine has been shown to drive an increase in intratumoral T cell infiltration by CD8 cells, as well as result in complete clinical regression in metastatic head and neck cancer when the vaccine was followed by PD-1 immunotherapy (this outcome was observed in 2/4 patients).

These data support the importance of the synthetic DNA approach for the treatment of virally-driven cancers which rely on viral oncogenes for continued disease. This is the situation for EBV-driven cancer as well. Here we report on studies investigating the generation of a multiantigen immunotherapeutic vaccine for EBV infection. We focused on developing a vaccine cocktail consisting of the episome-maintaining EBNA1 antigen combined with the two important latency-related membrane antigens for EBV, LMP1 and LMP2. We report the immune potency and early impact of the combined immune responses to these constructs.

DNA vaccines have previously reported interesting responses against LMP1 [28] and LMP2 [29] in mouse models. This study furthers this research by exploring the immune responses to a combination of EBV latent proteins using newly designed synthetic DNA-encoded antigens studied in the context of facilitated in vivo local delivery. The results show potent and consistent induction of T cell immunity in targeted mouse models with an impact on antigen-positive tumor growth, suggesting further study of this approach for EBV immunotherapy is important.
