Identification of vaccine candidates by immunization with a complex pool of recombinant VLPs

Abstract

Identifying the regions of a protein that can be readily translated into an effective vaccine remains a major challenge in epitope-focused vaccine design, especially when a protein is poorly characterized. Here, we describe an immune-driven strategy for epitope identification which uses the immune system to reveal regions within a candidate protein capable of eliciting targeted antibody responses. Using IL-17A, a pro-inflammatory cytokine, as a model antigen, we engineered a comprehensive library of MS2 bacteriophage virus-like particles (VLPs) displaying peptides representing every possible linear 10-amino acid peptide spanning the 133-amino acid protein. We then used the library to immunize mice, with the hypothesis that a subset of the IL-17A VLP mixture would elicit anti-IL-17A antibody responses. The IL-17A-reactive antibodies elicited by this complex vaccine were then used to select individual recombinant VLPs from the starting library, which were then tested for their ability to elicit anti-IL-17A antibodies in mice. Selected recombinant VLPs were able to elicit anti-IL-17A antibody responses, and antibody levels could be improved by conjugating synthetic peptides representing the selected epitopes to another bacteriophage platform (Qβ VLPs). Overall, this work describes a new method for antigen-agnostic epitope discovery and vaccine development that does not require pre-existing antibodies or rely on computational approaches to identify potential epitopes.