Abstract 2192 Characterizing electron-beam induced crystal reorientation, and its impacts on microED of macro- and small molecules

Microcrystal electron diffraction (MicroED) has emerged as a powerful method for atomic structure determination without need for large, pristine crystals, a prerequisite for successful structure determination by X-ray diffraction. However, the accuracy of MicroED structures appears worse to that of structures determined by X-ray crystallography; likewise, X-ray diffraction delivers information about crucial emergent structural properties of molecules, such as a small molecule's chirality or the shape of a substrate in an enzyme's binding pocket, that are less accessible from electron diffraction data. We aim to identify, rigorously characterize, and ultimately account for effects unique and fundamental to electron crystallography which might contribute to this discrepancy in data quality. In particular, we have observed and characterized non-monotonic changes in Bragg reflections produced by the irradiation of thin, stationary, microcrystals with low-flux electron beams, which indicate an unexpected physical reorientation of the crystal lattice induced by electron beam illumination. These diffracted intensity fluctuations, which take on varied character depending on the sensitivity of the crystal to damage by the electron beam, occur concurrently with the anticipated monotonic decay in diffracted signal due to radiolytic damage. Leveraging tools intended for treating serial crystallography data to index diffraction patterns from these stationary crystals, we observe a net rotation of as much as 1-2 degrees in the consensus orientation of certain beam-sensitive crystals in response to the delivery of fewer than 2 e-/Å2 fluence, a regime of dose nearly always encompassed in typical continuous rotation ED experiments. Beam-induced reorientations, if unmodeled during diffraction data processing, necessarily introduce a disagreement between the true orientation of the crystal diffracted in the experiment and the experimental geometry assumed during data reduction and structure determination, increasing potential errors. Importantly, the magnitude of reorientation we observe is consistent with beam-induced motions experienced by single particles in cryoEM, a well-known phenomenon that must be corrected in order to achieve high-quality, high-resolution reconstructions. In accounting for these effects, we anticipate an improvement in agreement between measured electron diffraction intensities and theory, better enabling the extraction of elusive structural information from nanocrystals. This work was supported in part by the NIH award R01AG074954, the BER program of the DOE Office of Science award DE-FC02-02ER63421, and the HHMI Emerging Pathogens Initiative. J.A.R. acknowledges support from the Arnold and Mabel Beckman Foundation, the David and Lucille Packard Foundation, the Pew Charitable Trusts, and the Searle Scholars Program.c