Protein engineering for antiviral therapeutics and vaccines

Abstract

Development of monoclonal antibodies (mAbs) and vaccines eliciting protective immunity is critical for viral therapies. For viruses with high antigenic diversity, like HIV and Influenza, broadly neutralizing antibodies (bNAbs), which neutralize several strains, are critical for controlling viral load and great effort has been expended to engineer immunogens that elicit bNAbs. While bNAbs have been identified for HIV, Influenza, and Dengue (DENV), it has proven difficult to elicit such antibodies with traditional vaccine strategies. The overarching goal of this thesis was to investigate the requirements for bNAb recognition in Dengue and Ebola viruses, with an eye toward the development of novel strategies for immunogen or immunotherapeutic design..;DENV causes 100 million annual infections; of which 250,000 result in severe Dengue due to antibody dependent enhancement of disease (ADE) caused by four co-circulating serotypes (DENV-1-4). In ADE, antibodies arising during primary infection bind but do not neutralize secondary infecting virus, resulting in increased uptake of virus in Fcy-receptor cells, viremia, and pathology. To avoid stimulating cross-reactive, non-neutralizing antibodies with the potential to promote ADE, we developed protein immunogens based on domain III (EDIII) of glycoprotein E, and demonstrate, for the first time, the elicitation of cross-neutralizing sera against DENV-1-3. As a prelude to EDIII immunogen engineering, and to further understand the requirements for bNAb recognition of EDIII, we mapped the functional epitopes of bNAb 4E11 (A/G-strand binder) on EDIII of DENV-1-4 using combinatorial alanine mutagenesis. Four EDIII phage libraries were produced, in which bNAb-contacting residues were allowed to vary between wild type and alanine to determine which residues significantly contribute to the energetics of binding to 4E11. We discovered a conserved, core functional epitope for all serotypes centered on residues: K310, L387, L/1389, and W391. Based on this, previous EDIII epitope characterization, and surface analysis, we designed EDIII-based libraries in which residues of cross-reactive epitopes (AB-loop), serotype-specific epitopes (FG-loop), and other surface exposed residues were allowed to vary among wild type, alanine, and serine to target the immune response to the bNAb epitope. Libraries were sorted for "resurfaced" EDIIIs (rsDllls) that maintained high affinity for bNAb 4E11 and demonstrated no affinity for a panel of DENV mAbs that bind the AB- and FG-loops. Two promising rsD111 candidates were stably folded, and demonstrated high affinity binding only to bNAb 4E11. Mouse immunizations resulted in high serum antibody titers to EDIII for each vaccine group, and excitingly, cross-neutralizing antibody titers for both WT and one rsDlll (rsDlll-Ala 30). Previously, only homotypic responses were observed using WT EDIII constructs.;Ebola virus is the causative agent of Ebola Virus Disease resulting in hemorrhagic fever and case fatality rates of up to 90%. Of the five Ebolavirus species, Zaire Ebola (EBOV) and Sudan (SUDV) account for 95% of Ebola-related deaths. Despite the success of the rVSV-EBOV vaccine, a broad post-exposure therapy is needed due to the sporadic and unpredictable nature of outbreaks. A cocktail of EBOV antibodies (ZMapp(TM)), which bind to the glycoprotein of EBOV (GP1.2), was found to be protective in non-human primates (NHP); however, this cocktail's efficacy is limited to EBOV outbreaks. Previously, no cross- or broadly neutralizing antibodies had been identified for ebolaviruses. As a first, we developed and characterized bispecific antibodies (Bis-mAb) that potently neutralize EBOV and SUDV and provide significant protection in post-exposure mouse models for both EBOV and SUDV. Our Bis-mAbs target the base of GP1.2, a neutralizing epitope for both glycoproteins, via genetic fusion of mAbs KZ52 (EBOV-specific) and F4 (SUDV-specific) and demonstrate similar affinity as the parental mAbs. Excitingly, subsequent studies expanding upon this work demonstrated broad neutralization, protection, and identified a broadly neutralizing epitope on GP1.2.