Prefusion Stabilized gB for EBV Vaccine Development

Background

Epstein-Barr virus (EBV) is a ubiquitous human herpesvirus that infects the majority of the global population. EBV can cause infectious mononucleosis and contribute to the development of cancers. Despite decades of research, there is no licensed prophylactic vaccine or targeted antiviral therapy. The viral fusion glycoprotein B (gB) mediates membrane fusion and viral entry and represents a principal antigenic target for neutralizing antibodies. However, eliciting robust immune responses against the functionally relevant prefusion conformation of gB remains a critical unmet need in the design of effective EBV vaccines and immunotherapies.

Current vaccine strategies and subunit immunogens often rely on proteins that adopt the postfusion conformation or suffer from conformational heterogeneity, leading to suboptimal presentation of neutralizing epitopes. The transient nature of the prefusion state, extensive glycosylation, and structural metastability hinder production of homogenous, high‐yield antigens. Furthermore, conventional expression and purification methods struggle to preserve prefusion architecture, resulting in low immunogenicity and inconsistent manufacturing batches. These limitations have impeded the development of potent EBV‐specific vaccines capable of eliciting durable and protective antibody responses.

Technology

The researchers have harnessed advanced structural modeling to lock the EBV glycoprotein B (gB) in its prefusion state through engineered disulfide bonds and complementary stabilizing modifications. The prefusion and postfusion ectodomain structures were predicted using ColabFold and AlphaFold Multimer, guided by herpesvirus templates and refined via Disulfide by Design 2 and PyMOL inspection. Eleven strategically placed cysteine pairs were evaluated. These pairs link domains close together in prefusion and far-apart in postfusion with interdomain and interchain disulfide bonds. Additional stabilizing modifications in selected variants include fusion-loop and optimized C-terminal truncations. The variants can be displayed on trimerization scaffolds (e.g., foldon or circular tandem repeat protein nanoparticles). Recombinant proteins expressed in HEK293 cells are purified by Ni-NTA affinity and size-exclusion chromatography, then characterized by SDS–PAGE and negative-stain EM, confirming stable, homogeneous prefusion trimers suitable for vaccine formulation in 4 of the 7 variants evaluated.

This technology is differentiated from previous efforts that relied on single-site mutations or generic scaffolds, by the utilization of an integrated, multi-tool design pipeline that couples high-resolution computational prediction with systematic experimental validation to generate the prefusion stabilized gB variants. The trimerization modules and precise truncation boundaries yield constructs with enhanced thermal and storage stability, rapid manufacturability, and broad applicability to other class III viral fusion proteins, positioning it as a versatile vaccine antigen design paradigm.

Benefits

• Locked prefusion conformation of EBV gB ensures presentation of key neutralizing epitopes 
• Enhanced protein stability via engineered disulfide bonds, proline/fusion‐loop mutations, and trimerization scaffolds 
• Scalable production using HEK293 expression and streamlined Ni-NTA/SEC purification 
• Compatibility with multiple vaccine platforms (protein subunit, mRNA/LNP) for flexible deployment 
• Detailed modeling and design workflow adaptable to other class III viral fusion proteins (e.g., HSV-1, CMV) 
• Potential to elicit potent, durable antibody responses and advance effective EBV prophylactic vaccines

Additional info

National Phase application based on WO 2024/006930

For any questions, email kamya.rajaram@seattlechildrens.org