Computational Design of α‐Conotoxins to Target Specific Nicotinic Acetylcholine Receptor Subtypes

Abstract

Nicotinic acetylcholine receptors (nAChRs) are drug targets for neurological diseases and disorders, but selective targeting of the large number of nAChR subtypes is challenging. Marine cone snail α‐conotoxins are potent blockers of nAChRs and some have been engineered to achieve subtype selectivity. This engineering effort would benefit from rapid computational methods able to predict mutational energies, but current approaches typically require high‐resolution experimental structures, which are not widely available for α‐conotoxin complexes. Here, we benchmarked five mutational energy prediction methods using crystallographic and mutational data on two acetylcholine binding protein/α‐conotoxin systems. We developed molecular models for six nAChR subtypes in complex with five α‐conotoxins that were studied through 150 substitutions. We determined that the best method was a combination of FoldX and molecular dynamics simulations, resulting in a predictive Matthews Correlation Coefficient (MCC) of 0.68 (85% accuracy). Novel α‐conotoxin mutants designed using this method were successfully validated by experimental assay with improved pharmaceutical properties. This work paves the way for the rapid design of subtype‐specific nAChR ligands and potentially accelerated drug development.