. 2022 Apr 14;13:867918. doi: 10.3389/fimmu.2022.867918

Table 3.

Characteristics of the pre-clinical studies included in the systematic review.

First author, year of publication, and title (glycoprotein[s] targeted)	Animal model	Vaccine	Study features	Measurement of overall antibody response and efficacy	Study outcomes	Relevance to the field
1. Zhang, 2020 (42) A novel vaccine candidate based on chimeric virus-like particle displaying multiple conserved epitope peptides induced neutralizing antibodies against EBV infection (gp350)	Female BALB/c mice (n = 5)	Platform: Virus-like particle (VLP) Antigen and dose: 20µg of hepatitis B virus core antigen (HBc149)-based VLPs expressing three conserved gp350 receptor binding domain peptides. Peptide 1 (P1) (aa 16-29), P2 (aa 142-161) and P3 (aa 282-301) from the gp350 receptor binding domain were fused to HBc149 and expressed in bacteria (BL21). The three peptides were inserted into the multiple insertion region of HBc149 (aa 78-82) in different tandem order combinations, yielding five constructs. The five combinations were: P1-L-P2-L-P3 (3A), where L represents an G4SG4S linker, P1-L-P3-L-P2 (3B), P2-L-P1-L-P3 (3C), P2-L-P3-L-P1 (3D) and P3-L-P2-L-P1 (3E). The five constructs were named 149-3A, 149-3B, 149-3C, 149-3D and 149-3E, respectively Adjuvant: Aluminum hydroxide	Control(s): Wild-type HBc149-based VLP (n = 5) or recombinant gp350-ectodomain₁₂₃ protein (1-425aa) (n = 5). Immunization schedule: Three subcutaneous injections administered at week (0, 2, 4) Sample collection: Week (0, 1, 2, 3, 4, 5, 6, 8, 10 and 14) Virus challenge: NA Study duration: 14 weeks	Antibody assay(s): ELISA (gp350^1-425-His) using sera from immunized mice. In vitro neutralization assay(s): Competitive ELISA (gp350^1-425-His) against anti-gp350 Ab 72A1 using sera from immunized mice; neutralization assay using Akata-EBV-eGFP (from CNE2-EBV cells) in EBV-negative Akata cells, with sera from immunized mice (Week 8 and 10). Neutralization efficacy was reported as % infected cells. Measurement of in vivo infection: NA	Sera from mice immunized with 149-3A and 149-3B showed higher anti-gp350 antibody titers and better competition to neutralizing antibodies (nAbs) 72A1 compared to sera from mice immunized with gp350 ectodomain₁₂₃ or other VLPs. At 5-fold and 10-fold dilution, the sera inhibited more than 50% binding of 72A1-HRP. Sera from mice immunized with 149-3A and 149-3B showed stronger neutralizing efficiency compared to the sera raised against soluble gp350 ectodomain_123, or other VLPs. At 2-fold, 4-fold, and 8-fold dilutions, the sera raised against 149-3A and 149-3B showed over 50% neutralizing efficiency against EBV infection. The corresponding ID₅₀ values are 13.43 and 18.93 for sera from animals immunized by 149-3A and 149-3B, respectively.	This was the first study to incorporate multiple gp350 receptor binding domain epitopes on the surface of chimeric HBc149 VLPs as vaccine candidates. Promising results from this study provide support for a VLP-based vaccine strategy that incorporates multiple known immunogenic and neutralizing epitopes. In addition, the use of different epitope combinations proved very informative, revealing that epitope order in such a platform can have drastic effects on vaccine immunogenicity, which will guide future vaccine design strategies using these types of platforms.
2. Escalante, 2020 (43) A pentavalent Epstein-Barr virus-like particle vaccine elicits high titers of neutralizing antibodies against Epstein-Barr virus infection in immunized rabbits (gp350, gB, gp42, gL, gH in combination)	Female and male New Zealand white rabbits (n = 6)	Platform: VLP Antigen and dose: 50µg of Newcastle disease virus (NDV)-based VLP incorporating EBV gp350^1-864, gB, gp42, gL and gH in a single construct produced in CHO cells Adjuvant: 500μg aluminum hydroxide (alum) mixed with 50µg monophosphoryl lipid A from Salmonella enterica serotype minnesota Re 595 (MPL)	Control(s): TNE buffer (n = 6), 50µg ultraviolet (UV)-inactivated Akata EBV (n = 6), 25µg soluble gp350 ectodomain^4-863 (n = 6). Immunization schedule: Three subcutaneous injections administered at day 0, 28 and 42 Sample collection schedule: Day (-7, 14, 35, 49, 70 and 90) Virus challenge: NA Study duration: 90 days	Antibody assay(s): ELISA (His-tagged g350, gB, gp42, gH/gL from human embryonic kidney cell line 293 (HEK-293 cells) using sera from immunized rabbits. In vitro neutralization assay(s): Neutralization assay using EBV-Akata-eGFP (from AGS cells) in HEK-293 and Raji cells, with purified immunoglobulins (IgGs) from immunized rabbits at a concentration of 1.56-50μg/ml. Neutralization efficacy was reported as % neutralization and IC_50. Measurement of in vivo infection: NA	Rabbits immunized with VLPs elicited high titers of gp350 and gB antibodies. However, low titers of gp42, and gH/gL antibodies were reported. Despite low gp42, gH/gL titers, purified IgGs from immunized rabbits neutralized virus infection in both epithelial and B cells at comparable levels to IgGs from UV-inactivated Akata EBV-immunized rabbits, and better than IgGs from gp350-immunized rabbits.	This was the first study to selectively combine five EBV glycoproteins in a single prophylactic vaccine candidate. Neutralization results are promising and provide support for a multivalent vaccine approach, but whether these results hold true in vivo remains to be tested. In addition, the platform requires optimization to achieve more equal glycoprotein expression to elicit high titers of antibodies against all five selected glycoproteins.
3. Bu, 2019 (44) Immunization with components of the viral fusion apparatus elicits antibodies that neutralize Epstein-Barr virus in B cells and epithelial cells (gH/gL, gp42-gH/gL, gp350-gH/gL, gp350-gp42-gH/gL)	(a) Female Balb/c mice (n = 5)	Platform: Nanoparticle Antigen and dose: 0.5µg of gH/gL or gp42-gH/gL H. pylori ferritin-based nanoparticles produced in Expi293F cells Adjuvant: Sigma Adjuvant System (50%v/v)	Controls: 0.5µg of soluble gH/gL (n = 5) or gp42-gH/gL (n = 5). No negative control reported. Immunization schedule: Three intramuscular injections at week (0, 3, 14) Sample collection: Week (2, 5, 13, 16, 20, 24, 28) Virus challenge: NA Study duration: 28 weeks	Antibody assay(s): Luciferase immunoprecipitation system (LIPS) on HEK-293 cells expressing gp42, and gH/gL using sera from immunized mice. In vitro neutralization assay(s): Neutralization assay using EBV-Akata-eGFP (from Akata BX1 cells) in SVKCR2, AGS and Raji cells, using sera from immunized mice. Neutralization efficacy was reported as IC_50. Measurement of in vivo infection: NA	The gH/gL and gp42-gH/gL ferritin-based nanoparticles elicited antibodies in immunized mice that neutralized B-cell and epithelial cell infection better than antibodies from mice immunized with the soluble version of the proteins. Antibodies elicited by gp42-gH/gL ferritin-based nanoparticles neutralized B-cell infection better than gH/gL ferritin-based nanoparticles, but the addition of gp42 had no effect on epithelial cell neutralization.	This was the first study to report that polyclonal antibodies against the gH/gL and gp42-gH/gL complexes are important components of the EBV neutralizing response in naturally infected individuals, which led to this being the first study to selectively use gp42-gH/gL as a complex for immunization. The results further emphasize the relevance of the three proteins as vaccine targets, as well as the importance of using structurally presented antigens in their native forms over monomeric soluble antigens. Furthermore, results from the gp350 combination studies further provide support for multivalent and multimeric vaccine approaches. However, the ability of the vaccine in eliciting protective nAbs in vivo remains to be tested.
	(b) Female Balb/c mice (n = 5)	Platform: Nanoparticle Antigen and dose: 0.5µg of gp350 gL H. pylori ferritin-based nanoparticle + 0.5µg of gH/gL gL H. pylori ferritin-based nanoparticle, or 0.5µg of gp350 gL H. pylori ferritin-based nanoparticle + 0.5µg of gp42-gH/gL gL H. pylori ferritin-based nanoparticle; produced in Expi293F cells Adjuvant: Sigma Adjuvant System (50%v/v)	Controls: 0.5µg of gp350 (n = 5), gH/gL (n = 5) or gp42-gH/gL (n = 5) ferritin-based nanoparticles. No negative control reported. Immunization schedule: Three intramuscular injections at week (0, 3, 14) Sample collection: Week (2, 5, 13, 16, 20, 24, 28) Virus challenge: NA Study duration: 28 weeks	Antibody assay(s): LIPS assay in HEK-293 cells expressing g350, gp42, and gH/gL using sera from immunized mice. In vitro neutralization assay(s): Neutralization assay using EBV-Akata-eGFP (from Akata BX1 cells) in SVKCR2, AGS and Raji cells, using sera from immunized mice. Neutralization efficacy was reported as IC_50. Measurement of in vivo infection: NA	The gp350-gH/gL and gp350-gp42-gH/gL nanoparticle combinations resulted in antibodies that markedly enhanced neutralization in epithelial cells when compared to antibodies from mice immunized with gp350 ferritin-based nanoparticles alone. However, there was no significant effect observed in B cell neutralization.
	(c) Cynomolgus macaques (Macaca fascicularis); sex not reported (n =5)	Platform: Nanoparticle Antigen and dose: 50µg gH/gL or gp42-gH/gL gL H. pylori ferritin-based nanoparticle produced in Expi293F cells Adjuvant: Sigma Adjuvant System (50%v/v)	Controls: 50µg soluble gH/gL or gp42-gH/gL (n = 5). No negative control reported. Immunization schedule: Three intramuscular injections at week (0, 4, 12) Sample collection: Week (0, 6, 8, 14, and 24) Virus challenge: NA Study duration: 24 weeks	Antibody assay(s): LIPS assay in HEK-293 cells expressing gp42, and gH/gL using sera from immunized macaques; fusion-inhibitory assay testing the fusion of either CHO-K1 cells expressing gB, and gH/gL or CHO-K1 cells expressing gB, gp42, and gH/gL with HEK-293 and Daudi cells, respectively, using sera from immunized macaques. In vitro neutralization assay(s): Neutralization assay using EBV-Akata-eGFP (from Akata BX1 cells) in SVKCR2, AGS and Raji cells, using sera from immunized macaques. Neutralization efficacy was reported as IC_50. Measurement of in vivo infection: NA	As in the first mouse study, the gH/gL and gp42-gH/gL ferritin-based nanoparticles elicited antibodies in immunized macaques that neutralized B-cell and epithelial cell infection better than antibodies from macaques immunized with the soluble version of the proteins. Antibodies elicited by gp42-gH/gL ferritin-based nanoparticles neutralized B-cell infection better than gH/gL ferritin-based nanoparticles, but the addition of gp42 had no effect on epithelial cell neutralization. These results were further confirmed/replicated in fusion-inhibitory assays.
4. Zhao, 2018 (45) Immunization with Fc-based recombinant Epstein–Barr virus gp350 elicits potent neutralizing humoral immune response in a BALB/c mice model (gp350)	(a) BALB/c mice; sex not reported (n = 5)	Platform: Multimeric protein Antigen and dose: 1 or 20µg of soluble dimeric gp350 ectodomain fused with mouse Fc-IgG2a (full-length, gp350-ECD_FL-FC, or truncated, gp350-ECD₁₂₃-FC), produced in Sf9 insect cells Adjuvant: Imject alum	Controls: Soluble monomeric truncated gp350 ectodomain (gp350-ECD₁₂₃-6His) (n = 5). No negative control reported. Immunization schedule: Intraperitoneal injections at week (0, 3) Sample collection: Week (0, 1, 2, 3, 4, 5) Virus challenge: NA Study duration: 5 weeks	Antibody assay(s): ELISA (His-tagged gp350) to determine both overall antibody titers and Ig subtypes, using sera from immunized mice. In vitro neutralization assay(s): Competitive ELISA (His-tagged gp350) against anti-gp350 nAb 72A1 using sera from immunized mice; surface plasmon resonance (SPR) antibody competition assay using 72A1 against sera antibodies from immunized mice; neutralization assay using Akata-EBV-eGFP (from CNE2-EBV cells) in EBV-negative Akata cells, with sera from immunized mice. Neutralization efficacy was reported as % infection rate. Measurement of in vivo infection: NA	Mice immunized with the dimeric proteins in both intraperitoneal and intranasal studies displayed higher levels of anti-gp350 antibody titers than mice immunized with monomeric gp350, as well as IgG1, IgG2a and IgG2b titers. Similar results were obtained in both competition assays and in neutralization assays. In general, antibodies from mice immunized intraperitoneally with the high dose of dimeric proteins performed better in all assays compared to mice immunized intranasally. Overall, the gp350-ECD₁₂₃-Fc vaccine performed best.	This was the first EBV vaccine study to use an Fc domain to multimerize an EBV glycoprotein. A vaccine using this type of platform is clinically relevant as Fc approaches have been validated in previous clinical trials and such a vaccine would thus face less regulatory hurdles to reach the clinic. This study is also one of the few preclinical studies to interrogate the identity of Ig subtypes elicited by an EBV vaccine candidate. Overall, this study also provides support for moving away from the use of monomeric proteins in favor of multimeric proteins or platforms that provide structural protein support.
	(b) BALB/c mice; sex not reported (n = 5)	Platform: Multimeric protein Antigen and dose: 20 µg of soluble dimeric gp350 ectodomain fused with mouse FC-IgG2a (full-length, gp350-ECD_FL-Fc, or truncated gp350-ECD₁₂₃-Fc), produced in Sf9 insect cells Adjuvant: 25µg CpG1826	Controls: Soluble monomeric truncated gp350 ectodomain (gp350-ECD₁₂₃-6His) (n = 5) or PBS (n = 5). Immunization schedule: Intranasal immunization at week (0,2) Sample collection: Week (0, 1, 2, 3, 4, 5) Virus challenge: NA Study duration: 5 weeks
5. Perez, 2017 (46) Novel Epstein-Barr virus-like particles incorporating gH/gL- EBNA1 or gB-LMP2 induce high neutralizing antibody titers and EBV-specific T-cell responses in immunized mice (gH/gL, gB, gp350, gp350-gH/gL, gp350-gB, gB-gH/gL, gp350-gB-gH/gL)	Female BALB/c mice: (n = 5)	Platform: VLP Antigen and dose: 10µg of individual NDV-based VLPs (gB-LMP2, gH/gL-EBNA1, or gp350/220 VLP) or VLP combinations of 10µg each (gp350/220 +gB-LMP2, gp350/220+gH/gL-EBNA1, gB-LMP2+gH/gL-EBNA1, or gp350/220+gB-LMP2+gH/gL-EBNA1), produced in CHO cells Adjuvant: None	Controls: 10µg of UV-inactivated EBV (UV-EBV) (n = 5), or TNE buffer (n = 5). Immunization schedule: Three intraperitoneal injections at day (0, 29, 54) Sample collection: Day (14, 18, 33, 46, 68, 97) Virus challenge: NA Study duration: 97 days	Antibody assay(s): ELISA (lytically-induced AGS-Akata cell lysate), using sera from immunized mice. In vitro neutralization assay(s): Neutralization assay using EBV-Akata-eGFP (from AGS cells) in HEK-293 and Raji cells, with sera from immunized mice. Neutralization efficacy was reported as % neutralization. Measurement of in vivo infection: NA	All vaccine groups, except for negative control, elicited IgGs in immunized mice that recognized cell lysate from lytically induced AGS-Akata cells. However, neutralization experiment data as presented is not interpretable, so it is unclear which immunization group(s) elicited better nAb responses and whether multivalent immunogen groups outperformed single immunogen groups in eliciting nAbs.	The study attempted to explore selective combination of multiple EBV glycoproteins as vaccine immunogens, following the multivalent trend recently started in the previous decade. While the vaccine was no doubt immunogenic, neutralization data as presented is not interpretable, and thus the relevance of this study is unclear.
6. Heeke, 2016 (47) Identification of GLA/SE as an effective adjuvant for the induction of robust humoral and cell-mediated immune responses to EBV-gp350 in mice and rabbits (gp350)	(a) Female BALB/c mice (n = 3)	Platform: Monomeric protein Antigen and dose: 5, 10, or 20µg of adjuvanted soluble gp350 produced in CHO cells. Adjuvant: 5µg of TLR4 agonist glucopyranosyl lipid A (GLA) in 2% stable emulsion (SE), or 100µg Aluminum hydroxide (Alum)	Controls: 5, 10, or 20µg of unadjuvanted soluble gp350 (n = 3), or PBS (n = 3). Immunization schedule: Two intramuscular injections at day (0, 14) Sample collection: Day (28) Virus challenge: NA Study duration: 28 days	Antibody assay(s): ELISA (gp350) to determine anti-gp350 IgG subtypes in sera of immunized mice. In vitro neutralization assay(s): Neutralization assay was performed but strategy or types of cells involved were not provided. Neutralization efficacy was reported as log₂ nAb titers able to reduce 50% infection. Measurement of in vivo infection: NA	Mice immunized with gp350 adjuvanted with GLA/SE performed better at all doses than mice immunized with unadjuvanted gp350 or gp350 adjuvanted with Alum in IgG1 and IgG2a ELISAs and neutralization assays. Minimal to no activity was observed in ELISAs and neutralization assays for mice immunized with unadjuvanted gp350. Gp350 adjuvanted with Alum only displayed meaningful activity in IgG1 ELISAs.	The study reports the first use of GLA/SE as an adjuvant to enhance immune response in animals immunized with gp350 monomeric protein to elicit both humoral and cellular responses, tested in both mice and rabbits (only humoral responses were reported in this table). The results support GLA/SE as a potent humoral and cellular adjuvant that performs better than Alum, but its utility in the clinic is yet to be explored as the product is not yet fully characterized and licensed for clinical use. This study is one of the few preclinical studies to interrogate the identity of Ig subtypes elicited by an EBV vaccine candidate and highlights the contribution of different adjuvants to vaccine immune response.
	(b) Female C57BL/6 mice (n = 7-10)	Platform: Monomeric protein Antigen and dose: 5µg of adjuvanted soluble gp350 produced in CHO cells. Adjuvant: 5µg of GLA/2%SE (n = 10), 5µg of GLA in aqueous formulation (GLA-AF) (n = 8), or 2% SE (n = 7)	Controls: 5µg of unadjuvanted soluble gp350 (n = 3), 5µg of GLA/2%SE (n = 3), 5 µg of GLA-AF(n = 3), 2%SE (n = 3), or PBS (n = 3). Immunization schedule: Two intramuscular injections at day (0, 14) Sample collection: Day (28) Virus challenge: NA Study duration: 28 days	(Same as above)	The results revealed that SE significantly contributes to the generation of nAbs, as adjuvanting with GLA-AF alone did not result in high nAb titers. On the other hand, GLA significantly contributes to the generation of IgG2c responses. The contribution of GLA and SE to IgG1 responses remained unclear.
	(c) Female BALB/c mice (n = 6)	Platform: Monomeric protein Antigen and dose: 10µg of adjuvanted soluble gp350 produced in CHO cells. Adjuvant: 5µg of GLA/2%SE, or 100µg of Alum	Controls: 10µg of unadjuvanted soluble gp350 (n = 6), or PBS (n = 6). Immunization schedule: Three intramuscular injections at day (0, 14, 28) Sample collection: Day (28, 42, 73, 102, 132, 161, 192, 222, 252, 283, 312, 347) Virus challenge: NA Study duration: 347 days	(Same as above)	IgG1, IgG2a, and neutralizing responses were stably maintained up to 347 days in mice immunized with gp350 adjuvanted with GLA/SE. IgG1 and neutralizing response were similarly maintained in mice immunized with gp350 adjuvanted with Alum, although nAb titers were lower than in GLA/SE adjuvanted mice. Low IgG21 activity was observed and maintained in mice immunized with unadjuvanted gp350, but no IgG2a or neutralizing activity was observed.
	(d) Female New Zealand white rabbits (n = 3)	Platform: Monomeric protein Antigen and dose: 50 or 100µg of adjuvanted soluble gp350 produced in CHO cells. Adjuvant: 1 or 2.5µg of GLA/2%SE	Controls: PBS (n = 3). Immunization schedule: Four intramuscular injections at day (0, 21, 42, 63) Sample collection: Day (0, 14, 35, 56, 77) Virus challenge: NA Study duration: 77 days	Antibody assay(s): ELISA (gp350) using sera of immunized animals. In vitro neutralization assay(s): Neutralization assay was performed but strategy or types of cells involved were not provided. Results were reported as log₂ nAb titers able to reduce 50% infection. Measurement of in vivo infection: NA	All dose combinations were shown to be immunogenic in immunized rabbits and resulted in similar levels of overall anti-gp350 IgG titers and neutralizing titers at the end of the study. However, the differences in dosage number and level to the mouse studies makes it difficult to compare both.
7. Cui, 2016 (48) Rabbits immunized with Epstein-Barr virus gH/gL or gB recombinant proteins elicit higher serum virus neutralizing activity than gp350 (gH/gL, gB or gp350)	Male New Zealand white rabbits (n = 5)	Platform: Multimeric protein Antigen and dose: 25µg of tetrameric gp350^1–470, trimeric gH/gL, or trimeric gB, produced in CHO cells. Adjuvant: 6.25μg alum mixed with 50μg 12mer phosphorothioate-modified CpG-ODN optimized for rabbits	Controls: 25µg monomeric gp350 (n = 5) or gH/gL (n = 5). Adjuvant alone is listed as an additional control, but no data is shown for them. Immunization schedule: Three subcutaneous injections at day (0, 21, and 42) Sample collection: Day (0, 10, 31, and 52) Virus challenge: NA Study duration: 52 days	Antibody assay(s): ELISA (gp350, gB, gH/gL) using sera from immunized rabbits. In vitro neutralization: Neutralization assay in Raji and human peripheral blood naïve B cells using B95-8/F-eGFP EBV, with sera from immunized rabbits. Neutralization efficacy was reported as EDI₅₀. Measurement of in vivo infection: NA	All constructs tested elicited glycoprotein-specific IgGs in immunized rabbits, although multimeric versions performed better than their corresponding monomeric versions. Trimeric and monomeric gH/gL, trimeric gB, and tetrameric gp350-induced nAbs that blocked EBV infection in B cells that were >100-fold, 20-fold, 18-fold, and 4-fold higher, respectively, than monomeric gp350. However, both monomeric and trimeric gH/gL, as well as trimeric gB, performed better than either monomeric or tetrameric gp350.	This is the first study to selectively use EBV gH/gL as a vaccine candidate to elicit nAb in immunized animals, and although gB had been tested before (Lockey et al, 2008), this is the first study do produce a trimeric form as a vaccine candidate. The study results support the use of multimeric forms of gH/gL and gB as components of a vaccine to prevent EBV infection of B cells. Although the vaccine candidates are expected to also provide protection against epithelial cell infection due to the role of gH/gL and gB in this process, the ability of the vaccine candidates to do so was not tested, and thus their utility in this context was not clear.
8. Ogembo, 2015 (49) A chimeric EBV gp350/220-based VLP replicates the virion B-cell attachment mechanism and elicits long-lasting neutralizing antibodies in mice (gp350)	BALB/c Mice; sex not reported (n = 5)	Platform: VLP Antigen and dose: 10μg of NDV-based VLP incorporating EBV gp350/220, produced in CHO cells. Adjuvant: None	Controls: 10μg of soluble gp350/220 (n = 5), 10μg of UV-inactivated EBV (n = 5), or TNE buffer (n = 5). Immunization schedule: Five intraperitoneal injections at day (0, 43, 172, 183, 218) Sample collection: Day (0, 14, 28, 43, 56, 70, 84, 154, 186, 197, 228) Virus challenge: NA Study duration: 228 Days	Antibody assay(s): ELISA (gp350/220) to determine anti-gp350 IgG general titers and subtype titers in sera of immunized animals In vitro neutralization assay(s): Neutralization assay in Raji cells using EBV B95-8-eGFP or Akata-eGFP produced in B95-8 or AGS cells, respectively, with sera from immunized mice. Neutralization efficacy was reported as % infected cells. Measurement of in vivo infection: NA	Overall, the gp350/220 VLP was immunogenic, resulting in long-lasting gp350/220-specific IgGs in immunized mice, with a predominant IgG1 response. Sera from immunized mice was able to neutralize EBV infection of Raji cells at various dilutions, at similar levels than sera from mice immunized with soluble gp350/220 protein. However, overall, immunization with UV-EBV resulted in superior responses.	This study provided proof-of-concept for the use of the NDV-LP platform as a potential EBV vaccine platform. The platform proved immunogenic, and the vaccine is produced in CHO cells, which are FDA-approved for clinical production of biologics. However, results support the use of multiple glycoprotein targets rather than gp350/220 alone, given that immunization with UV-EBV resulted in higher titers of nAbs. This study is one of the few to interrogate the identity of Ig subtypes elicited by an EBV vaccine candidate and the first to use inactivated virus as a positive control.
9. Kanekiyo, 2015 (50) Rational design of an Epstein-Barr virus vaccine targeting the receptor-binding Site (gp350)	(a) Female BALB/c mice (n = 5)	Platform: Nanoparticle Antigen and dose: 0.5 or 5µg of gp350 D₁₂₃ ectodomain (1-425aa, D₁₂₃)-ferritin (Helicobacter pylori-bullfrog hybrid)- or gp350 D₁₂₃-encapsulin (Termotoga maritima)-based nanoparticles, purified from FreeStyle 293F or Expi293F cells Adjuvant: 50% (V/V) SAS	Controls: 0.5 or 5µg of soluble gp350 ectodomain, full length, (n = 5), or soluble gp350 D₁₂₃ (n = 5). Mice immunized with irrelevant nanoparticle (n = 5) were additionally used for the challenge study. Immunization schedule: Three intramuscular injections at week (0, 3, 16) Sample collection: At Week (2, 5) and approximately every 5 weeks afterwards until Week 30. Virus challenge: Surrogate gp350-expressing vaccinia virus intranasal challenge (1x10^6 PFU), two months after last immunization. Study duration: ~ 26 weeks	Antibody assay(s): ELISA (g350), biolayer interferometry (gp350), and LIPS on a gp350-Renilla luciferase fusion protein, using sera from immunized mice; SPR antibody competition assay using 72A1 against sera antibodies from immunized mice. In vitro neutralization assay(s): Neutralization assay in Raji cells using EBV B95-8/F-eGFP produced in HEK-293/2089 cells, with sera from immunized mice. Neutralization efficacy was reported as IC_50. Measurement of in vivo infection: gp350-expressing vaccinia infection was measured by weight loss.	Mice immunized with either nanoparticle resulted in high serum gp350-specific IgG titers and long-lasting titers of nAbs that performed better than serum antibodies from mice immunized with soluble gp350 ectodomains (full-length and D₁₂₃), with activity from the latter practically non-existent. In the surrogate challenge study, only immunization with gp350 D₁₂₃-ferritin nanoparticle provided protection against infection.	This study provided proof-of-concept for the use of self-assembling ferritin and encapsulin-based nanoparticles as EBV vaccine platforms, and as such it is the first study to use rational structural nanoparticle design for EBV vaccine development. The results of this study also provide strong support for the use of vaccine platforms that provide conformational protein support, and against the use of monomeric soluble protein. However, it is difficult to discern the relevance of the challenge model used, since vaccinia should be able to infect mice both in the presence or absence of gp350.
	(b) Cynomolgus macaques (Macaca fascicularis); sex not reported (n = 4)	Platform: Nanoparticle Antigen and dose: 25µg of gp350 D₁₂₃-ferritin (Helicobacter pylori-bullfrog hybrid)- or gp350 D₁₂₃-encapsulin (Termotoga maritima)-based nanoparticles, purified from FreeStyle 293F or Expi293F cells Adjuvant: 50% (V/V) SAS	Controls: 50µg soluble gp350 ectodomain, full length (n = 4). No negative control reported. Immunization schedule: Three intramuscular injections at week (0, 4, 12) Sample collection: At week (0, 6, 8, 14) Virus challenge: NA Study duration: ~16 weeks	Antibody assay(s): ELISA (g350) using sera from immunized macaques. In vitro neutralization assay(s): Neutralization assay in Raji cells using EBV B95-8/F-eGFP produced in HEK-293/2089 cells, with sera from immunized macaques. Neutralization efficacy was reported as IC_50. Measurement of in vivo infection: NA	Immunization with the nanoparticles resulted in an increase of nAbs, although these performed similarly to antibodies from macaques immunized with gp350 ectodomain (control). However, these macaques had previous infection with an EBV-homologous lymphocryptovirus and cross-reactive nAbs prior to immunization, so the full relevance of these results is unclear.
10. Servat, 2015 (51) Identification of the critical attribute(s) of EBV gp350 antigen required for elicitation of a neutralizing antibody response in vivo (gp350)	Rabbits; strain and sex not reported (n = 3)	Platform: Monomeric protein Antigen and dose: 50μg native soluble gp350^1-860 or 50μg denatured/alkylated gp350, produced in CHO cells. Adjuvant: 200μl TiterMax (CytRx)	Controls: None reported. Immunization schedule: Three subcutaneous injections at day (0, 28. 63) Sample collection: At (Pre-bleed, day 73) Virus challenge: NA Study duration: 73 days	Antibody assay(s): ELISA (gp350) using sera from immunized animals. In vitro neutralization assay(s): Neutralization assay in Raji cells using Akata GFP-EBV produced in Akata cells, with sera from immunized rabbits. Neutralization efficacy was reported as log₂ sera dilution factor that resulted in 50% neutralization. Measurement of in vivo infection: NA	Immunization with both types of gp350 resulted in the generation of anti-gp350 IgGs that were reactive to both native (glycosylated) and denatured forms of gp350. However, only native gp350 resulted in generation of nAbs. Additionally, although not discussed in this table, native gp350 produced in CHO cells was the only form of gp350 able to bind to 72A1, among native and denatured gp350 produced in CHO cells, and native gp350 produced in E. coli. Likewise, native CHO gp350 competed for CD21 binding against a CD21 antibody, as opposed to denatured gp350.	This was the first study to analyze the conformational requirements of gp350 to generating a nAb response. The study demonstrated that production of native gp350 in mammalian cells results in post-translational modifications, particularly glycosylation, that stabilizes the protein in general as well as the conformation of its main neutralizing epitope, promoting immunogenicity. This highlights the importance of vaccine antigen source, placing mammalian cells as the ideal EBV antigen producers, due to the requirement for post-translational modifications that cannot be met in other types of cells.
11. Tanner, 2015 (52) Peptides designed to spatially depict the Epstein-Barr virus major virion glycoprotein gp350 neutralization epitope elicit antibodies that block virus-neutralizing antibody 72A1 interaction with the native gp350 molecule (gp350)	Female BALB/c mice (n =4)	Platform: Peptide Antigen and dose: 100µg keyhole limpet hemocyanin (KLH)-conjugated gp350 peptide 1 (SKAPESTTTSPTLNTTGFADY), peptide 2 (DDRTLQ L-A-QNPVYLIPETVPYIKWDN), or peptide 3 (GSAKPG NGSYF-A-SVKTEMLGNEID) Adjuvant: SAS	Controls: None reported. Immunization schedule: Two intraperitoneal injections at day (0, 21) Sample collection: At (Pre-bleed, day 35) Virus challenge: NA Study duration: ~35 days	Antibody assay(s): ELISA (gp350 peptide 1, 2 or 3; or gp350-expressing CEM cells) using sera from immunized mice. In vitro neutralization assay(s): Competitive ELISA (gp350-expressing CEM cells) against anti-gp350 Ab 72A1 using sera from immunized mice. Measurement of in vivo infection: NA	Immunization with all peptides resulted in the generation of antibodies specific to each peptide, although only peptide 2 and 3 resulted in antibodies capable of recognizing native gp350. Competition assays indicated that sera from p2- and p3-immunized mice were capable of inhibiting 72A1 recognition of native gp350 by 38% and 23%, respectively, with an additive reduction of 51% when anti-p2 and -p3 sera was pooled, while anti-p1 sera did not result in any competition.	This study was the first to computationally model the interaction of gp350 with the neutralizing anti-gp350 antibody, 72A1. This led to the theoretical identification of putative critical peptides involved in 72A1 binding, which were validated in vitro and then tested in immunization studies. While in vitro characterization studies of the peptides (not discussed in this table) provide strong support for these peptides as important neutralizing epitopes, immunization studies are not fully conclusive since there is no full gp350 ectodomain immunization comparator that could discern whether peptide immunization results in similar or better nAb responses than immunization with full protein. Furthermore, the immunoglobulin heavy and light chain sequences of 72A1 antibody used in this study were later confirmed to be incorrect [see (39, 53)], undermining the conclusion derived from the study.
12. Cui, 2013 (54) A novel tetrameric gp350^1-470 as a potential Epstein-Barr virus vaccine (gp350)	(a) Female BALB/C mice (n = 5)	Platform: Multimeric protein Antigen and dose: 1 or 25µg of tetrameric gp350^1-470 fused to two universal human tetanus toxoid (TT)-specific CD4+ T-cell epitopes (P₂ and P₃₀) produced in CHO cells Adjuvant: 13μg of alum +/− 25μg 30mer stimulatory CpG-ODN	Controls: 1 or 25µg monomeric gp350^1-470 (n = 5) Immunization schedule: Two intraperitoneal injections at day (0, 21) Sample collection: Day (0, 14, 21, 28, 35) Virus challenge: NA Study duration: 35 days	Antibody assay(s): ELISA (gp350) to determine overall anti-gp350 IgG titers, as well as titers of IgG subtypes in sera of immunized mice. In vitro neutralization assay(s): Inhibitory gp350-CD21 binding assay in CD21-expressing human erythroleukemia cells, using fluorescently labeled gp350 and sera from immunized mice (only performed in mice at the 25µg dose adjuvanted with alum alone). Measurement of in vivo infection: NA	In both protein and DNA immunizations, tetrameric gp350 performed significantly better than monomeric gp350 at both doses tested, eliciting higher titers of gp350-specific IgGs than monomeric protein in immunized mice. Likewise, immunization with tetrameric protein resulted in higher levels of nAbs than immunization with monomeric protein.	This is the first EBV vaccine study to use a multimeric protein approach, both in soluble protein and DNA vaccine forms. Results support the use of multimeric protein over monomeric protein, as the tetrameric gp350 vaccines performed better than monomeric gp350 vaccines in all its formats. While the main tetrameric format of the vaccine included two known TT-specific CD4+ T-cell epitopes, immunization with a tetrameric format that omitted the TT epitopes revealed that the epitopes did not have any effect on vaccine humoral immunogenicity; thus suggesting that the enhanced immunogenicity of the tetrameric vaccine as compared to the monomeric vaccine can be attributed to its multimeric form.
	(b) Female BALB/c mice (n = 7)	Platform: DNA vaccine Antigen and dose: 4μg plasmid DNA encoding tetrameric gp350^1-470 fused to TT epitopes P₂ and P₃₀ delivered in gold nanoparticles Adjuvant: None	Controls: 4μg plasmid DNA encoding monomeric gp350^1-470 delivered in gold nanoparticles (n = 7) Immunization schedule: Two epidermal immunizations in the abdominal skin at week (0, 4) Sample collection: Week (0, 4, 6) Virus challenge: NA Study duration: 6 weeks	Antibody assay(s): As above. In vitro neutralization assay(s): NA Measurement of in vivo infection: NA
	(c) Female BALB/C mice (n = 5)	Platform: Multimeric protein Antigen and dose: 14-day daily TT administration (0.25 or 25μg, adjuvanted with alum) followed by 25μg of tetrameric gp350^1-470 fused to TT epitopes P₂ and P₃₀, or by 25μg monomeric gp350^1-470 Adjuvant: 13μg of alum	Controls: TT-unprimed 25μg of tetrameric gp350^1-470 fused to TT epitopes P₂ and P₃₀ (n = 5) or monomeric gp350^1-470 (n = 5). Immunization schedule: Two intraperitoneal injections at day (0, 14) Sample collection: Day (0, 7, 14, 21) Virus challenge: NA Study duration: 21 days	Antibody assay(s): ELISA (gp350) using sera from immunized mice. In vitro neutralization assay(s): NA Measurement of in vivo infection: NA	While TT priming did not affect the immunogenicity of monomeric gp350 at both doses tested, it did result in a significant reduction in anti-gp350 IgGs in mice immunized with tetrameric gp350 at the 25μg TT dose, effectively eliminating the immunogenic advantage of the tetrameric gp350 as compared to the monomeric gp350.
	(d) Female BALB/C mice (n = 5)	Platform: Multimeric protein Antigen and dose: 1 or 25µg of tetrameric gp350^1-470 without TT epitopes Adjuvant: 13μg of alum	Controls: 1 or 25µg monomeric gp350^1-470 (n = 5), or tetrameric gp350^1-470 fused to TT epitopes P₂ and P₃₀ (n = 5). Immunization schedule: Two intraperitoneal injections at day (0, 21) Sample collection: At day (0, 14, 21, 28, 35) Virus challenge: NA Study duration: 35 days	Antibody assay(s): ELISA (gp350) using sera from immunized mice. In vitro neutralization assay(s): Inhibitory gp350-CD21 binding assay in CD21-expressing human erythroleukemia cells, using fluorescently labeled gp350 and sera from immunized mice (only performed in mice at the 25µg dose). Measurement of in vivo infection: NA	Removal of TT epitopes from the tetrameric gp350 did not affect the ability of the vaccine to elicit anti-gp350 IgGs, for both protein and DNA vaccines. Similarly, removal of TT epitopes did not affect the ability of tetrameric gp350 to elicit nAbs.
	(e) Female BALB/c mice (n = 7)	Platform: DNA vaccine Antigen and dose: 4μg plasmid DNA encoding tetrameric gp350^1-470 without TT epitopes delivered in gold nanoparticles Adjuvant: None	Controls: 4μg plasmid DNA encoding tetrameric gp350^1-470 fused to TT epitopes P₂ and P₃₀ (n = 7), or plasmid DNA encoding monomeric gp350^1-470 (n = 7), delivered in gold nanoparticles. Immunization schedule: Two epidermal immunizations in the abdominal skin at week (0, 4) Sample collection: Day (0, ~35, ~50) Virus challenge: NA Study duration: ~50 days	Antibody assay(s): ELISA (gp350) using sera from immunized mice. In vitro neutralization assay(s): NA Measurement of in vivo infection: NA
13. Mok, 2012 (55) Evaluation of measles vaccine virus as a vector to deliver respiratory syncytial virus fusion protein or Epstein-Barr virus glycoprotein gp350 (gp350)	(a) Cotton Rats (Sigmodon spp); sex not reported (n = 4)	Platform: Measles vaccine virus (Edmonston-Zagreb strain) Antigen and dose: 10⁵ PFU of live-attenuated recombinant measles vaccine virus (MV), expressing full-length gp350 ectodomain^1-861 or truncated gp350 ectodomain (gp350tr^1-470 produced in HEp-2 cells. Adjuvant: None	Controls: PBS (n = 3), or 10⁵ PFU of MV not expressing any antigens (rEZ; n = 5). Immunization schedule: Two intramuscular injections at week (0, 4) Sample collection: Week (4, 6) Virus challenge: NA Study duration: ~56 days	Antibody assay(s): ELISA (gp350) using sera from immunized rats. In vitro neutralization assay(s): Neutralization assay in Raji cells using B95-8 GFP-EBV produced in B95-8 cells, with sera from immunized rats. Neutralization efficacy was reported as log₂ sera reciprocal dilution that resulted in 50% neutralization. Measurement of in vivo infection: NA	Immunization of rats with MV expressing either form of gp350 elicited gp350-specific IgGs, albeit at low titers, although MV expressing full-length gp350 ectodomain resulted in higher titers than MV expressing gp350tr. However, no EBV nAbs were detected in the sera of immunized rats in either group. In rhesus macaques, the MV-gp350 vaccine did not yield any detectable anti-gp350 IgGs, or EBV nAbs. While not discussed here, T cell responses elicited by the gp350 vaccines were also analyzed, but in both rats and rhesus macaques the cellular responses were low.	The study indicates that the use of measles vaccine virus as a vector to deliver gp350 is not optimal in eliciting nAbs in either cotton rats or rhesus macaques, despite gp350 antibody being detected in immunized cotton rats at low levels. Since increased immunogenicity was observed in a respiratory syncytial virus MV vaccine also tested in the study, and given the differences in immunogenicity observed between rat and NHP immunizations, the study demonstrates that immunogenicity of foreign proteins expressed by measles virus as a vaccine platform is dependent on the nature of the insert and the animal models used for vaccine evaluation. Importantly, the measles vector was shown not to be ideal as a gp350 vaccine platform.
	(b) NHP, Rhesus macaque (Macaca mulatta); sex not reported (n = 4 measles virus seropositive; n = 4 measles virus naïve)	Platform: Measles vaccine virus (Edmonston-Zagreb strain) Antigen and dose: 10⁵ PFU of MV expressing full-length gp350 ectodomain^1-861 produced in HEp-2 cells. Adjuvant: None	Controls: None reported. Immunization schedule: Two intramuscular injections at week (0, 4) Sample collection: At week (4, 6) Virus challenge: NA Study duration: ~56 days	Antibody assay(s): As above. In vitro neutralization assay(s): As above. Measurement of in vivo infection: NA
14. Ruiss, 2011 (56) A virus-like particle-based Epstein-Barr virus vaccine (multiple antigens)	BALB/c mice; sex not reported (n = 4)	Platform: VLP Antigen and dose: 10μg of exosomal vesicles derived from lytically indued EBV packaging HEK-293 cells (VLPs) that contain various EBV proteins and RNAs, but are devoid of viral DNA Adjuvant: None	Controls: 10μg exosomes from non-EBV-packaging HEK-293 cells (n = 2). Immunization schedule: Two intraperitoneal injections at day (0, 14) Sample collection: Week (0, 6) Virus challenge: NA Study duration: 6 weeks	Antibody assay(s): ELISA, using sera from immunized animals, against cell lysates of HEK-293 cells expressing the following EBV antigens: EBNA1, BFRF3, BMRF1, BKRF4, BVRF1, BDLF3, BZLF2 (gp42), BXLF2 (gH), BNRF1, BALF4 (gB), BZLF1, BLLF1 (gp350); cytomegalovirus pp65 HEK-293 cell lysate was used as a negative control. In vitro neutralization assay(s): Neutralization assay in human primary B cells using GFP-EBV 2089 produced in HEK- 293/2089 cells, with sera from immunized mice. Neutralization efficacy was reported as % infected cells. Measurement of in vivo infection: NA	Immunization of mice with the VLPs elicited IgGs specific to each of the antigens tested in ELISA. Furthermore, immunization with VLPs resulted in generation of EBV nAbs that inhibited infection at a similar rate to 72A1 used at 20-40µg/ml).	The study generated an EBV-based VLP that shares the general morphology of the EBV virion, incorporates various viral proteins and RNAs, and, in theory, does not contain any EBV DNA. The platform proved promising in its immunogenicity both in cellular (not discussed here) and humoral responses. However, the feasibility of producing such a vaccine in a large scale, as recognized by the authors, is a challenge, and its production in human cells could face regulatory hurdles. Thus, an alternative production platform might be required for such an approach to reach the clinic.
15. Lockey, 2008 (57) Epstein-Barr virus vaccine development: a lytic and latent protein cocktail (gp350, gB, gp350-gB)	C57BL/6 mice; sex not reported (n =3 or 6)	Platform: Recombinant vaccinia virus (VV) vector Antigen and dose: 10⁷ PFU recombinant VV produced in MC57G cells expressing individual EBV proteins (n = 3 each): gp350, gp110 (gB), EBNA2, or EBNA2C; or 10⁷ PFU of a combination of all four VVs mixed in equal ratios (n = 6). Adjuvant: None	Controls: VV expressing the hemagglutinin neuraminidase-of human parainfluenza virus type 1 (VV-hPIV-HN) (n = 3), or no injection (n = 1). Immunization schedule: Single intraperitoneal injection at week (0) Sample collection: Week (2, 4, 8) Virus challenge: NA Study duration: 8 weeks	Antibody assay(s): Immunoprecipitation assays against each antigen using sera from immunized mice; immunofluorescence (IFA) assays against B95-8 cells using sera from mice immunized with either VV-gp350, VV-mix, or VV-hPIV-HN. In vitro neutralization assay(s): Neutralization assay (transformation inhibition) in human PBMCs using B95-8 supernatants, with sera from mice immunized with either VV-gp350 or VV-mix. Neutralization efficacy was reported as the highest serum dilution to inhibit transformation in ≥ 3 of 5 test wells, and in ≥ 1 of two independent assays. Measurement of in vivo infection: NA	Both individual antigen immunizations as well as the cocktail immunization successfully generated antibodies against each of the antigens tested, with comparable responses between individual antigens and the cocktail. While immunization of mice with VV-gp350 and VV-mix resulted in nAbs (other groups were not tested), titers were low when compared to those found in human sera from naturally infected individuals, which was used as a positive control.	This study proposes the use of a multivalent antigen cocktail that includes both lytic and latent EBV antigens as a vaccine, delivered as a viral vaccine vector. While the study demonstrated the cocktail approach to be immunogenic, resulting in both cellular (not discussed here) and humoral responses, whether this particular cocktail can prevent or treat infection in vivo remains unclear. The lack of ELISA data for each of the vaccine antigens also complicates assessment of the antibody response, and the low nAb titers generated by the vaccine suggests additional antigens might be needed to mount a more robust nAb response.
16. Wilson, 1999 (58) The major Epstein-Barr virus (EBV) envelope glycoprotein gp340 when incorporated into Iscoms primes cytotoxic T-cell responses directed against EBV lymphoblastoid cell lines. (gp350)	NHP, Cottontop tamarins (Saguinus oedipus oedipus); sex not reported (n =3)	Platform: Nanoparticle Antigen and dose: 30μg of gp340 produced in murine C-127 cells that were incorporated into immune-stimulating complexes (ISCOMs) Adjuvant: ISCOM	Control: Treatment not stated (n = 2) Immunization schedule: Three intramuscular injections at week (0, 3, 6). Sample collection: Week (7) Virus challenge: NA Study duration: ~7 weeks	Antibody assay(s): ELISA (gp340) using sera from immunized tamarins. In vitro neutralization assay(s): Neutralization assay in human fetal cord blood lymphocytes using B95-8 EBV, with sera from immunized tamarins; competitive ELISA (gp340) against anti-gp340 Ab 72A1 using sera from immunized tamarins. Neutralization efficacy was reported as EBV titer, where a reduction of virus titer greater than one log was considered neutralizing. Measurement of in vivo infection: NA	Immunization with the ISCOM-adjuvanted protein resulted in the generation of anti-gp340 IgGs. However, the neutralizing activity of the generated antibodies was very low when compared to sera antibodies from EBV-seropositive humans, attributed to low levels of antibody recognizing the major neutralizing epitope of gp340.	This study was the first to incorporate recombinant gp340 ectodomain produced in cell culture into nanoparticles/ISCOMs, a much more feasible approach than using gp350 isolated from EBV-infected cells, as previous gp340-ISCOM studies. Although cellular responses were very promising (not discussed here), humoral responses were suboptimal, perhaps due to gp340 denaturation occurring during chemical coupling to ISCOMs.
17. Jackman, 1999 (59) Expression of Epstein Barr virus gp350 as a single chain glycoprotein for an EBV subunit vaccine (gp350)	Rabbits; strain and sex not reported (n = 3)	Platform: Monomeric protein Antigen and dose: 50μg (for Freund’s adjuvant) or 100μg (for Alum) of recombinant gp350 non-splicing variant protein produced in CHO cells Adjuvant: Freund’s adjuvant or Alum	Controls: None reported. Immunization schedule: Three injections at day (0, 21, 42); no injection route stated. Sample collection: At (Pre-bleed, day 31, day 52) Virus challenge: NA Study duration: 52 days	Antibody assay(s): ELISA (non-splicing gp350 variant) using sera from immunized rabbits. In vitro neutralization assay(s): Neutralization assay in purified human B-lymphocytes, using unknown strain of EBV, with sera from immunized rabbits. Neutralization efficacy was reported as the reciprocal of the highest serum dilution that resulted in at least 50% inhibition of outgrowth of EBV-infected lymphocytes. Measurement of in vivo infection: NA	Immunization of rabbits with the non-splicing gp350 variant resulted in high titers of anti-gp350 IgGs, with animals receiving the dose in Freund’s adjuvant developing a stronger response that those immunized with the dose in Alum. Both groups developed high titers of EBV nAbs.	This study was the first to produce a recombinant non-splicing gp350 variant free of the gp220 isoform. The generated protein was immunogenic, and led to the second EBV vaccine clinical trial reported, published in 2007 by Moutschen et al. (see Table 2 ).
18. Cox, 1998 (60) Immunization of common marmosets with Epstein-Barr virus (EBV) envelope glycoprotein gp340: effect on viral shedding following EBV challenge (gp350)	(a) NHP challenge model, common marmosets, adults (Callithrix jacchus); sex not reported (n = 5)	Platform: Monomeric protein Antigen and dose: 30µg of recombinant gp340 produced in C-127 cells Adjuvant: Alum (1:3 ratio of alum to protein)	Control(s): Alum only (n = 6) Immunization schedule: Three intramuscular injections at week (-12, -8 -4). Sample collection schedule: Week (-12, 0, 2, 4, 8, 12, 14, 16, 20, 24, 28, 32, 37, 41, 45, 50) Virus challenge: Two oral injections of 3x10⁴ ID₅₀ M81 EBV at week (0, 12). Study duration: 50 weeks	Antibody assay(s): ELISA (gp340), and IFA for VCA/virus lytic antigens (VLA) on EBV-infected P3HR1 cells, using sera from immunized marmosets. In vitro neutralization assay(s): None reported. Measurement of in vivo infection: EBV-specific PCR amplification from whole mouth fluid.	In the first experiment, immunization of adult marmosets with gp340 reduced viral load/shedding in the buccal cavity upon challenge when compared to injection with alum alone, but the immunization did not result in sterile immunity. Similar results were observed in the neonate experiment. However, neonates with seropositive parents, both in immunized and unimmunized groups, displayed indicators of EBV infection prior to challenge, suggesting natural infection occurred before EBV challenge.	As previous marmoset challenge studies, the overall results of this study suggest gp340 alone is not sufficient to induce sterile immunity, although it might be useful in reducing viral loads and potentially EBV disease. The results of the neonate experiment also provide further support for the common marmoset challenge model as an EBV model more closely resembling human EBV infection, where natural infection can occur and be transmitted from parents to children.
	(b) NHP challenge model, common marmosets, neonates (Callithrix jacchus); sex not reported (n = 3 from seronegative parents; n = 5 from seropositive parents)	Platform: Monomeric protein Antigen and dose: 5µg of recombinant gp340 produced in C-127 cells Adjuvant: Alum (1:3 ratio of alum to protein)	Control(s): Alum (n = 3 from seronegative parents, n = 3 from seropositive parents). Immunization schedule: Three intramuscular injections at week (-12, -8 -4). Sample collection schedule: Week (0, 4, 9, 12, 16, 20, 24) Virus challenge: One oral injection of 1.5x10⁴ ID₅₀ M81 EBV at week (0). Study duration: 24 weeks	Antibody assay(s): Indirect IFA for VCA/VLA on EBV-infected P3HR1 cells, using sera from immunized marmosets, but results not shown. In vitro neutralization assay(s): None reported. Measurement of in vivo infection: As above.
19. Mackett, 1996 (61) Immunization of common marmosets with vaccinia virus expressing Epstein-Barr virus (EBV) gp340 and challenge with EBV (gp350)	NHP challenge model, common marmoset (Callithrix jacchus); sex not reported (n =4)	Platform: Vaccinia viral vector (Western Reserve [WR] strain) Antigen and dose: One dose of 5 x 10⁷ PFU and a second dose of 2 x 10⁸ PFU of recombinant vaccinia virus expressing gp340 (vMA1) Adjuvant: None	Control(s): “empty” vaccinia virus (vTK-16; n = 3), PBS (n = 4). Immunization schedule: Two intradermal injections at week (-10, -5) Sample collection: Various timepoints after immunization and before challenge, after challenge, and after immunosuppression, over a period of 2.5 years. Virus challenge: 2 oral injections of 5 x10⁴ ID₅₀ M81 EBV at week (0, 12). Between weeks 36-40 and 83-87, animals were immunosuppressed with a 32-day course of Cyclosporin A. Study duration: 2.5 years	Antibody assay(s): ELISA (gp340), and direct IFA for VCA and early antigens on EBV-infected P3HR1 cells, using sera from immunized marmosets. In vitro neutralization assay(s): None reported. Measurement of in vivo infection: Slot-blot (results not shown) and EBV-specific PCR amplification from whole mouth fluid.	Although the vaccine elicited anti-gp340 IgGs in immunized marmosets prior to EBV challenge, all animals eventually developed anti-gp340 responses following challenge that did not differ across treatment groups. Similarly, VCA responses were observed in immunized marmosets prior to challenge, but after challenge all animals developed similar VCA responses; after first immunosuppression VCA responses increased in all groups, but were unaffected after the second round. Persistent early antigen responses were observed in 1/4 of PBS and 2/3 vTK-16-treated animals, and sporadically in 2/4 PBS and 1/3 vTK-16-treated animals; while only 1/4 vaccine-treated animals developed sporadic early antigen responses, even after immunosuppression. Finally, PCR analyses detected DNA in 42% of samples collected from vaccinated animals, and 67% of samples collected from control animals, after immunosuppression.	The study used antibody responses against early antigen as a surrogate marker for virus replication, and DNA positivity in mouth fluids as a marker for virus burden. The vaccine was able to achieve a reduction of both markers in immunized marmosets when compared to control treatments, even after immunosuppression. This suggests that a gp340 vaccine might be able to reduce viral loads and potentially EBV disease in immunized humans, but that on its own is not sufficient to provide sterile immunity. However, it is important to note that the vaccinia strain used (WR strain), has been shown not to be too immunogenic [see (62)]
20. Finerty, 1994 (63) Immunization of cottontop tamarins and rabbits with a candidate vaccine against the Epstein-Barr virus based on the major viral envelope glycoprotein gp340 and alum (gp350)	(a) NHP challenge model, cottontop tamarins (Saguinus oedipus oedipus); sex not reported (n =5)	Platform: Monomeric protein Antigen and dose: 50μg of recombinant gp340 produced in C-127 cells Adjuvant: Alum (1:3 ratio of alum to protein)	Control(s): PBS with alum (n = 1). Immunization schedule: Four intramuscular injections at approximately day (-111, -81, -51, -21) Sample collection: Various timepoints before and after immunization, and after challenge Virus challenge: Two administrations of B95-8 EBV, one intraperitoneal and one intramuscular, at a dose known to induce lymphoma, at Day (0). Study duration: ~166 days	Antibody assay(s): ELISA (gp340), and IFA for VCA (results not shown), using sera from immunized tamarins. In vitro neutralization assay(s): Neutralization assay (transformation inhibition) in cord blood lymphocytes using unreported EBV strain, with sera from immunized tamarins. Neutralization efficacy was reported as log₁₀ reduction in virus titer. Measurement of in vivo infection: Tumor emergence monitoring (external palpation and measurement of lymph nodes; bodyweight recording).	Immunization of cottontop tamarins with gp340 adjuvanted with alum elicited gp340-specific antibodies only after the 4^th dose, with only a single animal eliciting nAbs. Regardless, 3 out of 5 animals were protected against EBV-induced lymphoma upon EBV challenge.	The results of this study emphasize the importance of the use of different animal models when evaluating vaccine and adjuvant immunogenicity. Alum in tamarins did not result in a robust humoral response, unlike SAF-1 in previous studies cited by the authors, but in rabbits both alum and SAF-1 proved equally immunogenic. Despite the low humoral immunogenicity observed in tamarins, over 50% of them were protected from EBV-induced lymphoma, suggesting cellular immunity had a potential role in controlling cancer development.
	(b) Rabbits; strain and sex not reported (n =2)	Platform: Monomeric protein Antigen and dose: 50μg (high dose) or 5μg (low dose) recombinant gp340 produced in C-127 cells, each administered with one of two adjuvants. Adjuvants: Alum or SAF-1	Control(s): None reported. Immunization schedule: Four intramuscular injections at approximately week (0, 2, 4, 6) Sample collection: At (Pre-bleed, Week 2, 4, 6, 8) Virus challenge: NA Study duration: ~56 days	Antibody assay(s): ELISA (gp340), using sera from immunized rabbits. In vitro neutralization assay(s): As above. Measurement of in vivo infection: NA	All immunized rabbits elicited anti-gp340 antibodies, at higher titers than marmosets, regardless of adjuvant. Similarly, all alum and SF-1-adjuvanted immunized rabbits generated nAbs.
21. Ragot, 1993 (64) Replication-defective recombinant adenovirus expressing the Epstein-Barr virus (EBV) envelope glycoprotein gp340/220 induces protective immunity against EBV-induced lymphomas in the cottontop tamarins (gp350)	NHP challenge model, cottontop tamarins (Saguinus oedipus oedipus); sex not reported (n =4)	Platform: Adenovirus (serotype 5) Antigen and dose: 5x10⁹ PFU, 1x10¹⁰ PFU, and 2 x10¹⁰ PFU replication-defective recombinant adenovirus (serotype 5) expressing gp340/220 produced from HEK-293 cells Adjuvant: None	Controls: Non-recombinant adenovirus (n = 1), unimmunized (n = 1). Immunization schedule: Three intramuscular injections at Week (0, 5, 13) Sample collection: Pre-bleed and then weekly Virus challenge: two administrations of B95-8 EBV, one intraperitoneal and one intramuscular, at a 100% tumorigenic dose, at week (16). Study duration: ~26 weeks	Antibody assay(s): ELISA (gp340/220), using plasma from immunized tamarins. In vitro neutralization assay(s): Neutralization assay in cord blood lymphocytes, using unknown strain of EBV, with plasma from immunized tamarins. Neutralization efficacy was not observed and was therefore not reported. Measurement of in vivo infection: Tumor emergence monitoring (external palpation and measurement of lymph node tumors).	Tamarins immunized with gp340/220 adenovirus elicited gp340/220-specific antibodies, but these antibodies were unable to neutralize infection in vitro. Despite this, all immunized animals were protected against EBV challenge, while control animals developed lymphoma.	The results of this study suggest a recombinant adenovirus approach could be a promising gp340/220 vaccine platform to reduce or prevent EBV disease. However, the fact that the generated antibodies were not neutralizing suggests EBV-induced cancer development was mostly controlled via cellular immunity.
22. Madej, 1992 (65) Purification and characterization of Epstein-Barr virus gp340/220 produced by a bovine papillomavirus virus expression vector system (gp350)	(a) BALB/c mice; sex not reported (n = 5)	Platform: Monomeric protein Antigen and dose: 0.1ml of 200µg/ml of recombinant gp340/220 ectodomain produced in mouse C-127 cells. Adjuvant: Alum	Controls: None reported. Immunization schedule: Four subcutaneous injections at Day (0, 14, 28, 42) Sample collection: Not stated. Virus challenge: NA Study duration: Not stated, but likely ~56 days.	Antibody assay(s): ELISA (gp340/220), using sera from immunized mice. In vitro neutralization assay(s): Neutralization assay (transformation inhibition) in cord blood lymphocytes using unreported EBV strain, with sera from immunized mice. Neutralization efficacy was reported as endpoint titer. Measurement of in vivo infection: NA	Immunized mice adjuvanted with Freund’s adjuvant elicited higher levels of anti- gp340/220 antibodies than immunized mice adjuvanted with alum, and produced nAbs at titers similar to tamarins immunized with gp340 isolated from B-958 cells in previous studies that were protected against EBV-induced lymphoma.	The study shows that it is feasible to produce and purify recombinant gp340/220 ectodomain, which maintains its antigenicity, using bovine papillomavirus virus as an expression system in mammalian cells. The resulting product is immunogenic and could elicit nAbs in immunized mice when given with alum, and its efficacy in vivo was tested in tamarins in a companion publication, Finerty et al., 1992 (see below).
	(b) BALB/c mice; sex not reported (n = 5)	Platform: Monomeric protein Antigen and dose: 0.1ml of 40ug/ml of recombinant gp340/220 ectodomain produced in C-127 cells. Adjuvant: Complete Freund’s adjuvant for primary immunization, incomplete Freund’s adjuvant for boosters	Controls: None reported. Immunization schedule: Four intraperitoneal injections at day (0, 14, 28, 42) Sample collection: Not stated Virus challenge: NA Study duration: Not stated, but likely ~56 days	Antibody assay(s): As above. In vitro neutralization assay(s): As above. Measurement of in vivo infection: NA
23. Finerty, 1992 (66) Protective immunization against Epstein-Barr virus-induced disease in cottontop tamarins using the virus envelope glycoprotein gp340 produced from a bovine papillomavirus expression vector (gp350)	NHP, Cottontop tamarins (Saguinus oedipus oedipus); sex not reported (n = 4)	Platform: Monomeric protein Antigen and dose: 50μg recombinant gp340 ectodomain produced in C-127 cells. Adjuvant: SAF-1	Control: Non-immunized (n = 2). Immunization schedule: Four intramuscular injections at Day (0, 10, 20, 30) Sample collection: At (prebleed, Day 0, 10, 20, 30, 40) Virus challenge: EBV administration, strain and route unstated, at a 100% tumorigenic dose, at Day (40) Study duration: ~95 days	Antibody assay(s): ELISA (gp340), using plasma from immunized tamarins. In vitro neutralization assay(s): Neutralization assay (transformation inhibition) in cord blood lymphocytes using unreported EBV strain, with plasma from immunized tamarins. Neutralization efficacy was reported as log₁₀ reduction in virus titer. Measurement of in vivo infection: Tumor emergence monitoring (external palpation and measurement of enlarged lymph nodes and tumors).	Most immunized tamarins generated nAbs, that had efficacy paralleling the titers of anti-gp340 antibodies (higher anti-gp340 antibody producers had higher neutralizing activity). In turn, animals that generated nAbs were protected against EBV-induced lymphoma.	The study tests the in vivo vaccine efficacy of a recombinant gp340 ectodomain protein extensively characterized in a companion publication, Madej et al., 1992 (see above), for the first time. Administered with SAF-1 adjuvant, the protein induced nAbs and proved protective against EBV-induced lymphoma.
24. Zhang 1991 (67) Mapping of the epitopes of Epstein-Barr virus gp350 using monoclonal antibodies and recombinant proteins expressed in Escherichia coli defines three antigenic determinants. (gp350)	Rabbits; strain and sex not reported (n = not stated)	Platform: Monomeric protein Antigen and dose: 250µg beta-galactosidase fusion gp350 proteins covering various gp350 regions (nucleotides 1782 to 2307, 3576 to 4266, 3069 to 4146, or 2301 to 4287, of the BamHI L fragment of B95-8 EBV) produced in E. coli. Adjuvant: Complete Freund’s adjuvant for primary immunization, incomplete Freund’s adjuvant for booster	Control: Beta-galactosidase protein (n = not stated) Immunization schedule: Two immunizations at Day (0, 30). First immunization was administered both intradermally and subscapularly; second immunization was administered intradermally. Sample collection: At (Prebleed, Week 2, 4, 6) Virus challenge: NA Study duration: ~6 weeks	Antibody assay(s): IFA (B95-8 cells), and dot blot immunoassay (gp350 protein, B95-8 cell lysates), using sera from immunized rabbits. In vitro neutralization assay(s): Neutralization assay (transformation inhibition) in cord blood lymphocytes using B95-5 virus, and neutralization assay in Raji cells using P3HR1 virus, with sera from immunized rabbits. Infection was measured by transformation and EBNA IFA in cord blood lymphocytes, and early antigen IFA in Raji cells. Neutralization efficacy was reported as the presence or absence of transformation, and the presence or absence of positive EBNA1 IFA staining. Measurement of in vivo infection: NA	Although anti-gp350 antibodies were generated by immunization with the various truncated gp350 proteins, no nAbs were elicited.	The study produced and characterized various forms of truncated gp350 protein in bacteria with the purpose of mapping gp350 epitopes. When they tested the immunogenicity of some of these proteins, they found that they elicited no nAb activity, which was not unexpected as none had been recognized in dot blot assays by a known nAb, F29-167-A10. The authors suspect this to be the result of lack of post-translational modifications in the bacterial system, namely glycosylation. While we now know that glycosylation is indeed important for the functionality of gp350 nAb epitopes, none of their tested truncations included the now known major neutralizing epitope.
25. Morgan, 1989 (68) Validation of a first-generation Epstein-Barr virus vaccine preparation suitable for human use (gp350)	NHP, Cottontop tamarins (Saguinus oedipus oedipus); sex not reported (n =4)	Platform: Monomeric protein Antigen and dose: 50μg gp340 isolated from the membrane fraction of B95-8 cells Adjuvant: SAF-1 adjuvant containing 250μg threonyl muramyl dipeptide and 2.5%v/v copolymer	Control: Non-immunized (n = 1). Immunization schedule: 5 subcutaneous injections at day (0, 14, 28, 42, 56) Sample collection: Day (0, 14, 28, 42, 56, and 70) Virus challenge: EBV administration, strain and route unstated, 100% carcinogenic dose (10^5.3 tissue culture transforming units), at day (70). Study duration:120 days	Antibody assay(s): ELISA (gp340) using sera from immunized tamarins. In vitro neutralization assay(s): Neutralization assay (transformation inhibition) in human cord blood lymphocytes using unreported EBV strain with sera from immunized tamarins. Neutralization efficacy was reported as the presence or absence of transformation, and whether 1ml of sample neutralized >10⁵ lymphocyte transforming units of virus. Measurement of in vivo infection: Measurement of emergent tumors; measurement of EBV levels in the blood by in vitro transformation assay (co-culturing PBMCs from infected tamarins with human cord blood lymphocytes).	The vaccine was immunogenic, eliciting high titers of anti-gp340 antibodies, and protected 50% of the animals against virus infection and 100% against EBV-induced lymphoma.	This study used a new purification process to isolate gp340 from B95-8 cell membrane (automated FPLC). While the resulting vaccine was very successful in eliciting nAbs and preventing EBV-induced tumorigenesis in immunized tamarins, the overall results indicate that this formulation is not sufficient to prevent infection in vivo.
26. Emini, 1989 (69) Vero cell-expressed Epstein-Barr virus (EBV) gp350/220 protects marmosets from EBV challenge (gp350)	NHP, Common marmosets (Callithrix jacchus); sex not reported (n =4)	Platform: Monomeric protein Antigen and dose: 100µg recombinant gp350/220 produced in Vero cells delivered in one of two adjuvants Adjuvant: Alum or Freund’s adjuvant (complete for primary immunization, incomplete for boosters)	Controls: Alum (n = 2), or Freund’s adjuvant (n = 2). Immunization schedule: Thre intramuscular injections at Month (0, 1, 2). Sample collection: at Months (0, 1, 2, 3, 4, 5, 6, 7, 8, 9) Virus challenge: One injection of 8 log₁₀ transforming units of B95-8 EBV, route unstated, at Month (3). Study duration: 9 months	Antibody assay(s): ELISA (gp350/220) using sera from immunized marmosets. In vitro neutralization assay(s): Neutralization assay in Loukes cells, using B95-8 EBV, with sera from immunized marmosets (measured by IFA). Neutralization efficacy was reported as the highest serum dilution that yielded a positive neutralization. Measurement of in vivo infection: Measurement of antibody titers (IFA) against early antigen and VCA	All the marmosets immunized with gp350/220 in Freund’s adjuvant elicited high titers of anti-gp350/220 antibodies, compared to very low titers in marmosets immunized with gp350/220 in alum. Only one of the marmosets immunized with gp350/220 in alum had nAbs as tested in vitro, while two from the gp350/220 in alum group had nAbs. However, when it came to in vivo protection, protection was only conferred to 50% of marmosets immunized with gp350/220 in alum, compared to none in the gp350/220 in Freund’s adjuvant group. Thus, the presence of both total anti-gp350/220 antibodies or nAbs did not correlate with in vivo protection.	This study tested gp350/220 produced in mammalian cells as a vaccine in two different adjuvants, alum and Freund’s adjuvant. Neither formulation was able to provide sterile immunity to challenged immunized marmosets, although alum-adjuvanted marmosets performed better than Freund’s adjuvant despite displaying a very limited anti-gp350/220 antibody response. Regardless, no correlation was observed between total anti-gp350/220 antibodies or nAbs and in vivo protection, thus results are inconclusive.
27. Morgan, 1988 (70) Recombinant vaccinia virus expressing Epstein-Barr virus glycoprotein gp340 protects cottontop tamarins against EB virus induced malignant lymphomas (gp350)	NHP, Cottontop tamarins (Saguinus oedipus oedipus); sex not reported (n =4)	Platform: Vaccinia virus (Wyeth strain or WR strain) Antigen and dose: 20µl of 3.9x10⁹ PFU/ml (by scarification) or 2x10⁹ PFU/ml (intradermally) recombinant vaccinia virus expressing gp340, Wyeth strain, or 20µl of 5x10⁹ pfu/ml (by scarification) recombinant vaccinia virus expressing gp340, laboratory (WR) strain Adjuvant: None	Controls: Non-immunized (n = 4). Immunization schedule: For Wyeth strain immunizations, either one intradermal injection at Week (0; n = 2) or two applications by scarification at Week (0, 2; n = 2); for WR strain immunizations, either one application by scarification at Week (0; n = 2) or two applications by scarification at Week (0, 14; n = 2). Sample collection: Every two weeks after primary immunization Virus challenge: EBV administration, strain and route unstated, at 10^5.3 transforming units of EBV (100% tumorigenic dose), six weeks after immunization (unclear whether after primary or secondary immunization). Study duration: ~110 days	Antibody assay(s): ELISA (gp340) and radioimmunoassay (gp340) using sera from immunized tamarins. In vitro neutralization assay(s): Performed with sera from immunized tamarins, but conditions not reported. Neutralization efficacy was not explicitly reported. Measurement of in vivo infection: Tumor emergence monitoring (physical palpation and estimation of tumor size).	None to low gp340-specific antibodies were detected by ELISA and radioimmunoassay in all immunized tamarins. Importantly, no nAbs were detected in tamarins immunized with the Wyeth strain, and only very low nAb levels were detected in tamarins immunized with the WR strain (not shown). When it came to in vivo protection, neither group was fully protected from EBV-driven cancer, but more tamarins were protected in the WR strain group (3/4) than in the Wyeth strain (1/4).	This study explored the use of two different types of vaccinia viruses as gp340 vaccine platforms. The overall anti-gp340 antibody response and nAb responses were low to nonexistent in immunized tamarins, but partial protection was achieved. While there was no correlation between this protection and the observed levels of anti-gp340 antibodies and nAbs, a potential positive correlation was observed with the levels of anti-vaccinia antibodies (not discussed here). Thus, the mechanism behind the observed protection is unclear, although the authors speculate that it might be cell-mediated.
28. Morgan, 1988 (71) Prevention of Epstein-Barr (EB) virus-induced lymphoma in cottontop tamarins by vaccination with the EB virus envelope glycoprotein gp340 incorporated into immune-stimulating complexes (gp350)	NHP, Cottontop tamarins (Saguinus oedipus oedipus); sex not reported (n =4)	Platform: Nanoparticle Antigen and dose: <2 to 5µg of ISCOMS complexed to gp340 produced in B95-8 cells Adjuvant: ISCOM	Controls: Non-immunized (n = 4). Immunization schedule: Two subcutaneous injections of 2-5µg gp350-complexed ISCOMs at Week (0, 2), and 1 subcutaneous injection of <2µg gp340-complexed ISCOMs at Week (4) Virus challenge: EBV administration, strain and route unstated, at 10^5.3 transforming units of EB, at Week (6). Sample collection: Unstated. Study duration: ~102 days	Antibody assay(s): ELISA (gp340) using sera from immunized tamarins. In vitro neutralization assay(s): Neutralization assay (transformation inhibition)in cord blood lymphocytes using unreported EBV strain, with sera from immunized tamarins. Neutralization efficacy was not explicitly reported. Measurement of in vivo infection: Tumor emergence monitoring (external palpation and measurement of lymph node tumors).	Immunization of tamarins with gp340-complexed ISCOMs resulted in the generation of anti-gp340 antibodies as well as in vitro nAb activity (not shown). Importantly, all immunized tamarins were fully protected against EBV challenge.	This was the first study to use ISCOMs as an EBV glycoprotein-delivery system. Results were very encouraging and support the use of ISCOMs as an EBV vaccine platform.
29. Emini, 1988 (72) Antigenic analysis of the Epstein-Barr virus major membrane antigen (gp350/220) expressed in yeast and mammalian cells: implications for the development of a subunit vaccine. (gp350)	New Zealand white rabbits; sex not reported (n = 6)	Platform: Monomeric protein Antigen: 700µg of gp350/220 produced in yeast cells, or 200µg of gp350/220 produced in mammalian cells (either GH3 or Vero cells) Adjuvant: Freund’s adjuvant	Control: 20µg (n = 2) or 100µg (n = 4) of gp350/220 obtained from B95-8 cells. Immunization schedule: Three intramuscular injections at Month (0, 1, 2) Sample collection: Two weeks following last injection Virus challenge: NA Study duration: ~3.5 months	Antibody assay(s): ELISA (gp350/220) using sera from immunized rabbits. In vitro neutralization assay(s): Neutralization assay in Loukes cells, using B95-8 EBV, with sera from immunized rabbits (measured by IFA), with and without complement. Neutralization efficacy was reported as the reciprocal of the highest serum dilutions at which complete neutralization was observed. Measurement of in vivo infection: NA	While all immunized rabbits produced anti-gp350/220 antibodies, only rabbits immunized with gp350/220 produced in mammalian cells generated detectable nAbs. The addition of complement in neutralization assays had varying effects on neutralization depending on the source of gp350/220.	This study was the first to directly compare the properties and immunogenicity of gp350/220 protein produced in yeast versus mammalian cells. Results from the characterization experiments (not discussed here), and most importantly from the immunogenicity experiments, suggest that glycoprotein post-translational modifications originating from the producer cell type play a critical role in antigenic presentation. Results from this study also suggest yeast is not a suitable cell type for EBV glycoprotein expression for vaccine purposes.
30. David, 1988 (73) Efficient purification of Epstein-Barr virus membrane antigen gp340 by fast protein liquid chromatography. (gp350)	BALB/c mice; sex not reported (n = 4)	Platform: Monomeric protein Antigen and dose: 10µg of gp340 isolated from the membrane fraction of B95-8 cells Adjuvant: SAF-1	Control: None reported. Immunization schedule: Five subcutaneous injections at Day (0, 14, 28, 42, and 56) Sample collection: Day (0, 14, 28, 42, 56, 70) Virus challenge: NA Study duration: 70 days	Antibody assay(s): ELISA (gp340) using sera from immunized mice. In vitro neutralization assay(s): Neutralization assay (transformation inhibition) in cord blood lymphocytes using unreported EBV strain, with sera from immunized mice. Neutralization efficacy was not explicitly reported. Measurement of in vivo infection: NA	Immunization of mice with this vaccine resulted in the production of high titers of anti-gp340 antibodies. High sera neutralizing activity was also reported, but results were not shown.	In this study the authors sought to establish a new method (FPLC-based) for purifying gp340 from B95-8 cells for vaccine purposes, due to the fact that gp340 purified using immunoaffinity chromatography using nAb 72A1 did not confer sterilizing protection (Epstein et al., 1986). The results of the study show that gp340 purified in this way and used as a vaccine with SAF-1 adjuvant is immunogenic, but the fact that no results were shown for neutralization assays makes the true immunogenic properties of this vaccine difficult to assess as is.
31. Epstein, 1986 (74) Not all potently neutralizing, vaccine-induced antibodies to Epstein-Barr virus ensure protection of susceptible experimental animals (gp350)	NHP, Cottontop tamarins (Saguinus oedipus oedipus); sex not reported (n = 4)	Platform: Nanoparticle Antigen and dose: 275µg of gp340 isolated from B95-8 cells, delivered in liposomes Adjuvant: 5x10¹⁰ Bordatella pertussis cells (first immunization only)	Control: Non-immunized (n = 2). Immunization schedule: Six intraperitoneal injections at Day (0, 14, 28, 42, 56, and 70) Sample collection: Prebleed and one bleed following each immunization, exact schedule unclear. Virus challenge: A single dose of 10^5.3 transforming units of B95-8 EBV, delivered in an intramuscular and an intraperitoneal injection, at Day (105). Study duration: ~175 days	Antibody assay(s): ELISA (gp340) and IFA, using sera from immunized tamarins. In vitro neutralization assay(s): Neutralization assay (transformation inhibition) in cord blood lymphocytes using unreported EBV strain, with sera from immunized tamarins. Neutralization efficacy was reported on whether 1ml serum neutralized >100,000 lymphocyte transforming units of virus. Measurement of in vivo infection: Tumor emergence monitoring (external palpation and measurement of lymph node tumors).	Despite all immunized tamarins developing high titers of anti-gp340 antibodies and high sera neutralizing activity, no immunized animals were protected from the EBV challenge.	This study utilized immunoaffinity chromatography to isolate gp340 from B95-8 cells to be delivered in a liposome-based EBV vaccine. While a previous liposome formulation incorporating gp340 isolated via a molecular weight-based method fully protected tamarins from EBV challenge, no tamarins in this study were protected against challenge despite displaying high levels of sera neutralizing activity. Several potential reasons were given to explain the failed results, including (i) gp340 must undergo denaturation during SDS-PAGE preparations before immunization; (ii) essential gp340 molecules are not isolated by the 72A1 nAb; and (iii) elution of gp340 from the 72A1 column at pH 11.5 may destroy important antigenic carbohydrate structure. This study highlights the importance of testing the efficacy of EBV vaccine candidates in vivo, and the potential dangers of relying on in vitro efficacy data alone.
32. Mackett, 1985 (75) Recombinant vaccinia virus induces neutralising antibodies in rabbits against Epstein-Barr virus membrane antigen gp340 (gp350)	Rabbits; strain and sex not reported (n = 2)	Platform: Vaccinia virus (WR strain) Antigen and dose: 10⁸ PFU of vaccinia virus expressing gp340 Adjuvant: None	Control: Empty vaccinia (n = 1). Immunization schedule: Two intradermal injections at Week (0, 4) Sample collection: Prebleed and at Week (4,8) Virus challenge: NA Study duration: 3 months	Antibody assay(s): IFA (B95-5 and W91 cells) using sera from immunized rabbits. In vitro neutralization assay(s): Performed with sera from immunized rabbits, but conditions not reported. Neutralization efficacy was reported as the relative strength of neutralization. Measurement of in vivo infection: NA	Immunization of rabbits with gp340-expressing vaccinia elicited high titers of EBV-specific antibodies as well as nAbs.	This was the first study to use a viral vector as an EBV vaccine platform. Although the immunization experiments were done in very few animals, the obtained results supported the use of a vaccinia-based EBV vaccine.
33. Epstein, 1985 (76) Protection of cottontop tamarins against Epstein-Barr virus-induced malignant lymphoma by a prototype subunit vaccine (multiple antigens, gp350)	(a) NHP, Cottontop tamarins (Saguinus oedipus oedipus); sex not reported (n = 2)	Platform: Infected cell membrane Antigen and dose: Purified B95-8 cell membranes containing 2,500 units of gp340. Adjuvant: None	Control(s): Non-immunized (n = 2). Immunization schedule: Eight intraperitoneal injections at Week (0, 2, 4, 6, 8, 10, 12, 14) Sample collection schedule: NA Virus challenge: A single dose of 10^5.3 lymphocyte transforming units of B95-8 EBV (timing NA, route unclear). Study duration: NA	Antibody assay(s): IFA (B95-8 cells) and ELISA (gp340), using sera from immunized tamarins. In vitro neutralization assay(s): Neutralization assay (transformation inhibition) in cord blood lymphocytes using unreported EBV strain, with sera from immunized tamarins. Neutralization efficacy was reported on whether 1ml serum neutralized >10,000, 100,000 or >100,000 lymphocyte transforming units of virus. Measurement of in vivo infection: Tumor emergence monitoring while alive (external palpation and measurement of lymph node tumors); necropsy analysis of tumor lesions in spleen, liver, kidney, gut wall, adrenals, and lymph nodes.	Purified EBV-infected cell membrane, which likely contained multiple glycoproteins and not only gp340, elicited high titers of nAbs that completely protected NHP from lymphoma after 8 immunizations. gp340 purified by size and delivered via liposomes did elicit nAbs, but challenged animals that generated nAbs still developed transient lesions.	The results of this study were regarded as evidence that gp340 might serve as an important EBV vaccine target. Indeed, immunization with gp340 delivered via liposomes resulted in in vitro nAb activity. However, not all immunized animals were fully protected against disease, and 2/3 that were fully protected were immunized with infected cell membrane that likely contained additional glycoproteins. In addition, the immunization regimen (8-17 immunizations) required to achieve the observed level of nAb activity is not a feasible or practical realistic vaccine regimen. Regardless, the small number of animals used in the study is not statistically powered to allow an accurate assessment.
	(b) NHP, Cottontop tamarins (Saguinus oedipus oedipus); sex not reported (n = 4)	Platform: Nanoparticle Antigen and dose: gp340 isolated from B95-8 cell membranes, delivered in liposomes (dose NA) Adjuvant: None	Control(s): None reported. Immunization schedule: NA Sample collection schedule: NA Virus challenge: A single dose of 10^5.3 lymphocyte transforming units of B95-8 EBV (timing NA, route unclear). Study duration: NA
	(c) NHP, Cottontop tamarins (Saguinus oedipus oedipus); sex not reported (n = 2)	Platform: Nanoparticle Antigen and dose: 2,250 units of gp340 isolated from B95-8 cell membranes, delivered in liposomes Adjuvant: None	Control(s): Non-immunized (n = 2) Immunization schedule: Seventeen intraperitoneal injections at Week (0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32). Sample collection schedule: NA Virus challenge: A single dose of 10^5.3 lymphocyte transforming units of B95-8 EBV (timing NA, route unclear). Study duration: NA
34. Morgan, 1984 (77) Comparative immunogenicity studies on Epstein-Barr virus membrane antigen (MA) gp340 with novel adjuvants in mice, rabbits, and cottontop tamarins (gp350)	(a) Female BALB/c mice (n = 4)	Platform: Nanoparticle Antigen and dose: 70 or 15 units of gp340 isolated from B95-8 cells, delivered as (a) liposomes; (b) liposomes incorporating the lipid A fraction from Escherichia coli lipopolysaccharide; or (c) liposomes mixed with Bordatella pertussis cells. Adjuvant: lipid A, or 2x10⁹ killed Bordatella pertussis cells.	Control(s): 70 or 15 units of purified gp340 isolated from B95-8 cells in complete Freund’s adjuvant (n = 4). Immunization schedule: Variable according to treatment. Liposomes alone were delivered intravenously; liposomes with lipid A were delivered either intraperitoneally (at two different doses) or intravenously; liposomes with Bordatella pertussis cells were delivered intraperitoneally; control was delivered intraperitoneally. All groups were boosted twice at monthly intervals, except for liposomes with Bordatella pertussis which were only boosted once. Sample collection schedule: Two weeks after each injection Virus challenge: NA Study duration: ~100 days	Antibody assay(s): Binding assay to radiolabeled gp340; immunoprecipitation/SDS-PAGE analysis of B95-8 and M-ABA cell lysates (results for P3HR-1 and QIMR-WIL cells not shown). In vitro neutralization assay(s): NA Measurement of in vivo infection: NA	In mice, liposome formulations with lipid A were superior to any other formulations in eliciting anti-gp340 antibody titers, and the only treatment to elicit a response with the low gp340 dose. All liposome formulations elicited higher responses than purified gp340 in Freund’s adjuvant. Immune sera successfully immunoprecipitated gp340 from B95-8 and M-ABA cells. Although the data was not shown, reported responses in rabbits to liposomes alone were nonexistent, and only observed in animals that received either liposomes with lipid A, or purified gp340 in Freund’s adjuvant after boosts. Similar to mice, liposomes with lipid A was the only treatment to elicit a response with the low gp340 dose. In tamarins, immunizations with liposomes incorporating lipid A also performed better than liposomes alone, and the generated antibodies were found to be neutralizing. Immune sera immunoprecipitated gp340 from B95-8 cells.	This study explored the role of different factors that could influence immunogenicity of antigens presented in liposomes, including animal species, adjuvants, and route of vaccine delivery. The study found that gp340 delivered as liposomes incorporating lipid A as an adjuvant performed better than any other gp340 treatment independent of the immunization route, a pattern that was conserved across the species tested. This led the authors to establish the gp340 liposome/lipid A formulation as a potential candidate to move forward in future EBV vaccine studies. This was also the first study to test an EBV vaccine candidate in cottontop tamarins, which is used as an EBV challenge model in subsequent studies. While the study emphasizes the importance of testing vaccine candidates in various animal models, unclear details in immunization dose and schedules, and differences in immunization schedules across species, complicate comprehensive analysis.
	(b) Sandy-Lop rabbits; sex not reported (n = unspecified)	Platform: Nanoparticle Antigen and dose: 300-500 units of gp340 isolated from B95-8 cells, delivered as (a) liposomes; or (b) liposomes incorporating lipid A. Adjuvant: lipid A.	Control(s): 300-500 units of purified gp340 isolated from B95-8 cells in complete Freund’s adjuvant. (n = unspecified). Immunization schedule: Variable according to treatment. Liposomes alone and liposomes with lipid A were delivered intravenously; control was delivered subcutaneously. All groups were boosted at 3-4 weekly intervals. Sample collection schedule: Two weeks after each injection. Virus challenge: NA Study duration: NA	Antibody assay(s): Unspecified. In vitro neutralization assay(s): NA Measurement of in vivo infection: NA
	(c) NHP, male and female, Cottontop tamarins (Saguinus oedipus oedipus) (n = 4)	Platform: Nanoparticle Antigen and dose: 1000-2000 antigen units of gp340 isolated from B95-8 cells, delivered as (a) liposomes mixed with Bordatella pertussis cells; or (b) liposomes incorporating lipid A. Adjuvant: 10¹⁰ killed Bordatella pertussis cells, or lipid A	Control(s): None reported. Immunization schedule: Both treatments were delivered intraperitoneally, and groups were boosted every 3-9 weeks. Sample collection schedule: 2-5 weeks after each boost. Virus challenge: NA Study duration: NA	Antibody assay(s): Binding assay to radiolabeled gp340; immunoprecipitation/SDS-PAGE analysis of B95-8 cell lysates. In vitro neutralization assay(s): Neutralization assay (transformation inhibition) in cord blood lymphocytes using unreported EBV strain, with sera from immunized tamarins. Neutralization efficacy was reported as the dilution that neutralized 1000 transforming doses of EBV. Measurement of in vivo infection: NA
35. North, 1982 (78) Purified Epstein-Barr virus Mr340,000 glycoprotein induces potent virus-neutralizing antibodies when incorporated in liposomes (gp350)	Female BALB/c mice (n = 4)	Platform: Monomeric protein and nanoparticle Antigen and dose: 70-80 units of gp340 isolated from B95-8 cells delivered as (a) purified protein mixed with Complete Freund’s Adjuvant; (b) purified protein mixed with Bordatella pertussis organsisms; or (c) liposomes mixed with Bordatella pertussis organsisms. Adjuvant: Freund’s adjuvant, or 2x10⁹ Bordatella pertussis organisms	Controls: None reported. Immunization schedule: Two intraperitoneal injections at Week (0, 6) Sample collection: At Week (4, 10) Virus challenge: NA Study duration: 10 weeks	Antibody assay(s): Radioimmunoassay (gp340); immunoprecipitation analysis of B95-8 and QIMR-WIL cell lysates. In vitro neutralization assay(s): Neutralization assay (transformation inhibition) in cord blood lymphocytes using M-ABA strain EBV, with sera from immunized mice. Neutralization efficacy was reported as the 50% endpoint dilution. Measurement of in vivo infection: NA	Immunization of mice with gp340 elicited high anti-gp340 antibody titers when the antigen was delivered as liposomes with Bordatella pertussis, but no response was observed when delivered as purified protein. Reactive sera specifically immunoprecipitated gp340 from B95-8 and QIMR-WIL cell lysates. Furthermore, pooled sera from mice immunized with the liposomal formulation showed robust neutralizing activity against M-ABA strain EBV.	This study provided the first evidence of preparation of purified gp340 from EBV-infected cell membranes as a subunit vaccine, delivered in a liposomal formulation (nanoparticle). These encouraging results opened the way for various future studies focused on gp340 liposomal formulations as EBV vaccine candidates.
36. Thorley-Lawson, 1979 (79) A virus-free immunogen effective against Epstein-Barr virus (multiple antigens)	Rabbit; sex not reported (n = 1)	Platform: Monomeric protein (multiple) Antigen and dose: 100µg of membrane proteins from P3HR-1 cells Adjuvant: Complete Freund’s adjuvant	Controls: 100µg of membrane proteins from Ramos cells (n = 1) Immunization schedule: Three subcutaneous and intravenous injections over the course of two months. Virus challenge: NA Sample collection: 10 days after last injection Study duration: 8-10 weeks	Antibody assay(s): NA In vitro neutralization assay(s): Neutralization assay (transformation inhibition) in B lymphocytes from adult donors using B95-8 EBV, with sera from immunized rabbits. Neutralization efficacy was reported as the number of wells transformed at different sample dilutions. Measurement of in vivo infection: NA	Immunization with P3HR-1 membrane proteins elicited high titers of nAbs that neutralized infection in adult B lymphocytes, in contrast with immunization from Ramos cell membrane proteins, which did not elicit neutralizing sera.	The study reported the first successful large-scale production of DNA-free immunogens from the plasma membrane of EBV producer cell lines. This protein mix elicited high titers of nAbs, further cementing the fact that membrane antigens are important EBV vaccine targets, and identifying multiple antigens. However, the number of animals used is too small to infer meaningful conclusion.