Variable epitope library-based vaccines: shooting moving targets.

While the antigenic variability is the major obstacle for developing vaccines against antigenically variable pathogens (AVPs) and cancer, this issue is not addressed adequately in current vaccine efforts. We developed a novel variable epitope library (VEL)-based vaccine strategy using immunogens carrying a mixture of thousands of variants of a single epitope. In this proof-of-concept study, we used an immunodominant HIV-1-derived CD8+ cytotoxic T-lymphocyte (CTL) epitope as a model antigen to construct immunogens in the form of plasmid DNA and recombinant M13 bacteriophages. We generated combinatorial libraries expressing epitope variants with random amino acid substitutions at 2-5 amino acid positions within the epitope. Mice immunized with these immunogens developed epitope-specific CD8+ IFN-gamma+ T-cell responses that recognized more than 50% of heavily mutated variants of wild-type epitope, as demonstrated in T-cell proliferation assays and FACS analysis. Strikingly, these potent and broad epitope-specific immune responses were long lasting: after 12 months of priming, epitope variants were recognized by CD8+ cells and effector memory T cells were induced. In addition, we showed, for the first time, the inhibition of T-cell responses at the molecular level by immune interference: the mice primed with wild-type epitope and 8 or 12 months later immunized with VELs, were not able to recognize variant epitopes efficiently. These data may give a mechanistic explanation for the failure of recent HIV vaccine trials as well as highlight specific hurdles in current molecular vaccine efforts targeting other important antigenically variable pathogens and diseases. These findings suggest that the VEL-based strategy for immunogen construction can be used as a reliable technological platform for the generation of vaccines against AVPs and cancer, and contribute to better understanding complex host-pathogen interactions.