The geometric determinants of programmed antibody migration and binding on multi-antigen substrates

Abstract

Viruses and bacteria commonly exhibit spatial repetition of surface molecules that directly interface with the host immune system. However the complex interaction of patterned surfaces with multivalent immune molecules such as immunoglobulins and B-cell receptors is poorly understood, and standard characterization typically emphasizes the monovalent affinity. We developed a mechanistic model of multivalent antibody-antigen interactions as well as a pipeline for constructing such models from a minimal dataset of patterned surface plasmon resonance experiments in which antigen pattern geometries are precisely defined using DNA origami nanostructures. We modeled the change in binding enhancement due to multivalence and spatial tolerance,i.e. the strain-dependent interconversion of bound antibodies from monovalently bound to bivalently bound states at varying antigen separation distances. The parameterized model enables mechanistic post hoc characterization of binding behavior in patterned surface plasmon resonance experiments as well as de novo simulation of transient dynamics and equilibrium properties of arbitrary pattern geometries. Simulation on lattices shows that antigen spacing is a spatial control parameter that can be tuned to determine antibody residence time and migration speed. We found that gradients in antigen spacing are predicted to drive persistent, directed antibody migration toward favorable spacing. These results indicate that antigen pattern geometry can influence antibody interactions, a phenomenon that could be significant during the coevolution of pathogens and immunity in processes like pathogen neutralization or affinity maturation.