Rabbits immunized with Epstein-Barr virus gH/gL or gB recombinant proteins elicit higher serum virus neutralizing activity than gp350

Primary infection with Epstein-Barr virus (EBV) is associated with acute infectious mononucleosis, whereas persistent infection is associated with chronic diseases such as autoimmune diseases and various types of cancer. Indeed, approximately 2% of all new cancer cases occurring annually worldwide are EBV-associated. Currently, there is no licensed EBV prophylactic vaccine. Selection of appropriate viral protein subunits is critical for development of an effective vaccine. Although the major EBV surface glycoprotein gp350/220 (gp350) has been proposed as an important prophylactic vaccine target, attempts to develop a potent vaccine based on gp350 alone have shown limited success in the clinic. We provide data showing that five EBV glycoproteins (gp350, gB, gp42, gH, and gL) involved in viral entry and infection can successfully be incorporated on the surface of EBV-like particles (EBV-LPs). These EBV-LPs, when administered together with aluminum hydroxide and monophosphoryl lipid A as adjuvants to New Zealand white rabbits, elicited EBV glycoprotein-specific antibodies capable of neutralizing viral infection in vitro in both B cells and epithelial cells, better than soluble gp350 ectodomain. Our findings suggest that a pentavalent EBV-LP formulation might be an ideal candidate for development as a safe and immunogenic EBV vaccine.

Herpesvirus entry into host cells is mediated by multiple virally encoded receptor binding and membrane fusion glycoproteins. Despite their importance in host cell tropism and associated disease pathology, the underlying and essential interactions between these viral glycoproteins remain poorly understood. For Epstein–Barr virus (EBV), gHgL/gp42 complexes bind HLA class II to activate membrane fusion with B cells, but gp42 inhibits fusion and entry into epithelial cells. To clarify the mechanism by which gp42 controls the cell specificity of EBV infection, here we determined the structure of gHgL/gp42 complex bound to an anti-gHgL antibody (E1D1). The critical regulator of EBV tropism is the gp42 N-terminal domain, which tethers the HLA-binding domain to gHgL by wrapping around the exterior of three gH domains. Both the gp42 N-terminal domain and E1D1 selectively inhibit epithelial-cell fusion; however, they engage distinct surfaces of gHgL. These observations clarify key determinants of EBV host cell tropism.

Herpesvirus entry into cells requires the coordinated action of multiple virus envelope glycoproteins, including gH, gL, and gB. For EBV, the gp42 protein assembles into complexes with gHgL heterodimers and binds HLA class II to activate gB-mediated membrane fusion with B cells. EBV tropism is dictated by gp42 levels in the virion, as it inhibits entry into epithelial cells while promoting entry into B cells. The gHgL and gB proteins are targets of neutralizing antibodies and potential candidates for subunit vaccine development, but our understanding of their neutralizing epitopes and the mechanisms of inhibition remain relatively unexplored. Here we studied the structures and mechanisms of two anti-gHgL antibodies, CL40 and CL59, that block membrane fusion with both B cells and epithelial cells. We determined the structures of the CL40 and CL59 complexes with gHgL using X-ray crystallography and EM to identify their epitope locations. CL59 binds to the C-terminal domain IV of gH, while CL40 binds to a site occupied by the gp42 receptor binding domain. CL40 binding to gHgL/gp42 complexes is not blocked by gp42 and does not interfere with gp42 binding to HLA class II, indicating that its ability to block membrane fusion with B cells represents a defect in gB activation. These data indicate that anti-gHgL neutralizing antibodies can block gHgL-mediated activation of gB through different surface epitopes and mechanisms.