(416d) Computational Design of Protein Nanopores for Selective Binding to Chemical Warfare Agents

Chemical warfare agents (CWAs) such as Sarin and Mustard gas are dangerous weapons of mass destruction that might be used against combatants, first responders, and civilians in terrorist strikes. Therefore, there exist a significant number of studies exploring the sensing and detection of these CWA molecules. Most currently existing techniques that can sense the presence of CWAs include very expensive and large equipment such as NMR/mass spectroscopy and gas chromatography. Some alternative methods include detection by metal-organic frameworks (MOFs), zeolites, and carbon-based materials. On the other hand, many pore proteins, both in their natural state and after engineering, have been adapted as cheaper and effective alternatives for biosensing, nanopore sequencing, and separations. Protein based nanopores have naturally evolved to selectively transport ions, small molecules, and other macromolecules across cell membranes through specific interactions and gating mechanisms. Several types of pore proteins exist in nature with wide ranges of diameters from sub-nanometer to a few nanometers. Each pore protein has evolved to a precise pore geometry and physicochemical characteristics to perform its specific transport functions. In this study, we explore the possibility of computationally designing the pore interior of the OmpF protein for selective binding to Sarin and Mustard gases. We use iterative computational protein design algorithms - IPRO and RosettaDesign to design library of OmpF variants that offer a selective bioactive pore interior to bind Sarin and Mustard molecules. The designed variants indicate two possible modes of binding enhancement – (1) IPRO designs introduce strong electrostatic attachment with the molecules, while (2) Rosetta designs are enriched in hydrophobic amino acid substitutions at the binding sites. Promising designed pockets show up to 2-fold enrichment in binding scores compared to the wild-type protein. Stability of selected designs are validated by simulating OmpF trimers in a native-like membrane environment using all-atom MD simulations. Top designed OmpF variant is determined to have a binding free energy of ~ -8 kcal/mol to Mustard molecule using alchemical free-energy perturbation calculations. The designed OmpF variants can be embedded in block co polymer membranes and tested for pore-based sensing of CWA molecules by measuring interruptions in current across the membranes caused as a result of the designed selective molecular interactions.