(553d) Discovery of Antibodies Targeting Multipass Transmembrane Proteins Using a Suspension Cell-Based Evolutionary Platform

Membrane proteins play a critical role in cell sensing and signal transduction and are thus targeted by the majority (>60%) of all approved drugs. Whereas virtually all clinical drugs targeting complex, multipass transmembrane proteins (such as G protein-coupled receptors, GPCRs) are small molecules, protein therapeutics and, in particular, antibody drugs represent a promising class of interventions for membrane protein targeting. Given their specificity, extended serum persistence, and multi-tiered mechanisms of action, there is great interest in developing antibodies against multipass transmembrane proteins for basic research and pharmaceutical design. Unfortunately, major limitations exist in developing antibodies that recognize multipass transmembrane proteins, largely due to difficulties with recombinant expression of the target proteins. Whole-cell screening platforms overcome these challenges by selecting against cells presenting the target protein in its natural membrane-embedded state. We recently developed a fully suspension cell-based whole-cell screening platform known as “biofloating” that interfaces yeast cells expressing antibody fragments with mammalian cells expressing target membrane proteins (Fig. 1). We demonstrated that this platform shows accelerated binding kinetics and superior sensitivity compared to adherent cell-based platforms, allowing for accurate quantification of protein-protein interactions, as well as directed evolution-mediated enrichment of membrane protein-binding clones within antibody libraries.

Building on the biofloating platform, we developed an optimized workflow for the selection and screening of antibodies targeting membrane proteins from naïve libraries. We also established the first reported stoichiometric binding metric to enable real-time tracking of enrichment progress and inform decisions regarding selection scheme and individual clone screening. We implemented our selection workflow against the single-pass transmembrane protein programmed death-ligand 1 (PD-L1) and discovered three unique target-specific antibody fragments. We further tested our platform against complex, multipass membrane proteins in the GPCR family, and discovered single-domain antibody (nanobody) binders against 4 different proteins: (1) C-X-C motif chemokine receptor 2 (CXCR2); (2) glucagon-like peptide 1 receptor (GLP1R); (3) glucagon receptor (GCGR); and (4) C-X-C motif chemokine receptor 4 (CXCR4). Interestingly, a discovered nanobody against CXCR2 was found to be non-competitive with the IL-8 ligand and previously reported antibodies targeting the N-terminus of the GPCR, suggesting that our evolved molecule binds a unique epitope on the receptor. Finally, we used the biofloating platform to discover antibodies against a ligand-gated ion channel that is aberrantly expressed in leukemia cells. Overall, these studies highlight the robustness and versatility of our novel approach to the discovery of antibodies targeting both single-pass and multipass transmembrane proteins for a broad range of basic and translational research applications.