Abstract 982: Functionally important hotspot interfaces between immune-oncology targets PD-1 and PD-L1 and between Hippo pathway targets YAP2 and tight junction protein ZO-1 are identified using a protein-protein interaction technique optimized with novel dye chemistries

Protein-protein interactions are thought to be the next frontier in drug discovery. However, there are several well-known challenges facing development of protein-protein interaction (PPI) inhibitors that lead to slow development in the field, including difficulty identifying PPIs and difficulty designing small molecule inhibitors of relatively flat, featureless PPIs. We address these difficulties with the development of a novel method of discovering PPI hotspots, called protein painting. This technique relies on non-covalent labeling of solvent- accessible protein surfaces using small molecular dyes optimized for protein binding. The dyes block access to trypsin cleavage sites, allowing for digestion only of undyed interface regions following denaturation of the protein complex. Interface regions can be identified using mass spectrometry and used as target sequences for drug development. Here we introduce new dye chemistries and elucidate their mechanism of protein binding for the first time. This allows for rapid identification of functionally-relevant hotspots without the need to screen many dye chemistries to optimize surface coverage of the protein complex. We applied this method to elucidate functional hotspots for immmuno-oncology targets PD-1 and PD-L1 and Hippo pathway targets YAP2 and tight junction protein ZO-1. To further functionally validate the hotspot regions identified, we focused on the case study of PD-1 and PD-L1. We discovered a hotspot of PD-1 Lys 78 in the protein-protein interface, and rationally designed a series of 8 peptide inhibitors to target this hotspot. The most active peptide YRCMISYGGADYKRITV derived from PD-L1 disrupted the PD-1/PD-L1 complex with an IC50 of 5.07 µM. The predicted binding site of this peptide on PD-1 overlaps the binding site of therapeutic anti-PD-1 antibody pembrolizumab; crystal structures of pembrolizumab and PD-1 show hydrogen bonding between the antibody and our identified hotspot Lys 78. Furthermore, we prepared a cyclized analog peptide CYRAMISYGGADYKRITC by disulfide bond stapling to increase peptide stability and found that this did not significantly reduce inhibitor potency, with an IC50 of 8.02 µM. Taken together, this data suggests a specific region of PD-1 found within the larger PD-1/PD-L1 interface that may serve as a target for development of next generation small molecule PD-1/PD-L1 inhibitors. By focusing drug discovery efforts against only the PPI hotspot regions, we may accelerate drug development against these difficult targets. Citation Format: Amanda Still, Douglass Dey, Rachel Carter, Angela Dailing, Mikell Paige, Lance Liotta, Alessandra Luchini. Functionally important hotspot interfaces between immune-oncology targets PD-1 and PD-L1 and between Hippo pathway targets YAP2 and tight junction protein ZO-1 are identified using a protein-protein interaction technique optimized with novel dye chemistries [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 982.