Electrostatic potential is the dominant force in antigenic selection of naïve T-cells and B-cells for activation and maturation.

Peptide antigenicity can be predicted from sequence using a simple method invented by Hopp and Woods in the early 1980's. Since then, a much clearer understanding of T-cell/B-cell signaling and maturation calls for a new understanding of the amino acid determinants of antigenicity. We show that short peptides with more charged side chains generate significantly higher titers of peptide specific antibodies in co-immunized mice. Peptide docking simulations using linearized Poisson-Boltzmann calculations of electrostatic potential show that immunoglobulins distinguish the cognate peptide sequence from randomly selected sequences at "arms length" (10-20 Å) with >70 % of alternative sequences having higher energy at this distance, consistent with the weak specificity observed for naive T-cell and B-cell receptor interactions with MHC-bound antigen. Orders of magnitude lower complexity of the state space of charged surfaces as compared to the state space of surface shapes suggests a dominant role of electrostatics in selecting naive immune cells from the population of circulating cell lines. We propose a two-stage antigen recognition process, first electrostatic and then shape-based, that explains the dominant contribution of charged residues to peptide immunogenicity.