Isolation of  L I domain mutants mediating firm cell adhesion using a novel flow-based sorting method

Abstract

The inserted (I) domain of αLβ2 integrin (LFA-1) contains the entire binding site of the molecule. It mediates both rolling and firm adhesion of leukocytes at sites of inflammation depending on the activation state of the integrin. The affinity change of the entire integrin can be mimicked by the I domain alone through mutations that affect the conformation of the molecule. High-affinity mutants of the I domain have been discovered previously using both rational design and directed evolution. We have found that binding affinity fails to dictate the behavior of I domain adhesion under shear flow. In order to better understand I domain adhesion, we have developed a novel panning method to separate yeast expressing a library of I domain variants on the surface by adhesion under flow. Using conditions analogous to those experienced by cells interacting with the post-capillary vascular endothelium, we have identified mutations supporting firm adhesion that are not found using typical directed evolution techniques that select for tight binding to soluble ligands. Mutants isolated using this method do not cluster with those found by sorting with soluble ligand. Furthermore, these mutants mediate shear-driven cell rolling dynamics decorrelated from binding affinity, as previously observed for I domains bearing engineered disulfide bridges to stabilize activated conformational states. Characterization of these mutants supports a greater understanding of the structure-function relationship of the αL I domain, and of the relationship between applied force and bioadhesion in a broader context.

Equilibrium and dynamic design principles for binding molecules engineered for reagentless biosensors

“…For dynamic measurement of time-varying analyte levels, the biophysical characteristics of the binding event can significantly impact biosensor accuracy and sensitivity. Given the availability of directed evolution protein engineering methodology to create binding molecules of almost arbitrary affinity and widely varying association and dissociation rates[16,17], these variables are available degrees of freedom for improvement of biosensor performance. Recently Haugh developed a reaction-diffusion model to investigate biosensor signal interpretation in live cell imaging, with an emphasis on capturing intracellular and membrane-localized phenomena[18].…”