High-throughput automated molecular replacement for small-molecule MicroED data

Interest in electron diffraction (ED) for structural characterization of both proteins and small molecules has grown significantly over the last decade. While ab initio phasing methods remain the gold standard for ED data from small-molecule samples, radiation beam damage during data collection and poor crystallinity of the nanocrystalline sample can make this method unfeasible - particularly for challenging molecules that exhibit conformational flexibility. Molecular replacement (MR) is the most commonly used phasing method for protein ED data and can circumnavigate issues related to diminished data quality. However, its application to small molecules has been limited due to the lack of methods for generating optimal trial conformations. Herein, a high-throughput automated molecular replacement workflow has been developed to solve a novel ED structure of corilagin, a macrocyclic gallotannin with pharmaceutical relevance, which could not be solved with ab initio phasing. The method was validated against three similar macrocycles with known structures (paritaprevir-α, paritaprevir-β and grazoprevir) at varying data resolution limits (1.0, 1.2, 1.4, 1.5, 1.6, 1.8 and 2.0 Å). At all these resolutions for all three structures, the developed workflow was successful and produced solutions with R factors and RMSD values within acceptable limits of the ab initio solved structure.