Cobalamin-dependent methionine synthase (MetH) is a large, highly dynamic enzyme that requires extensive domain rearrangements to support catalytic turnover and periodic reactivation. Existing biochemical and structural work has established that MetH cycles between distinct functional states, but a comprehensive view of the conformational changes that bridge turnover and reactivation remained incomplete.
In this study, the authors revisited an existing cryo-electron microscopy dataset to search for structural states that were previously unreported. From a dataset dominated by resting-state particles, they identified and recovered a low-abundance population corresponding to a reactivation-state of MetH. The recovered population was reconstructed to an overall resolution of approximately 4.5–5 Å, enabling interpretation of a fully intact enzyme bound to its physiological cobalamin cofactor.
The recovered structure represents the first intact cryo-EM reconstruction of MetH with physiological cofactor bound reported in the source. Because the minor population was outnumbered by resting-state particles in the original dataset, it had remained hidden until the authors applied updated processing strategies.
Three methodological advances were reported as critical to recovering the minor reactivation-state particles. First, the team used a neural-network particle picker to detect particles that may have been missed or misclassified by earlier, conventional picking approaches. Second, they performed careful classification of particles to segregate heterogeneous conformations and enrich for the minor population representing the reactivation state. Third, they incorporated high-resolution information to determine initial particle orientations during processing, which improved alignment for the small subset of particles and contributed to the quality of the final reconstruction.
Together, these processing choices allowed the authors to overcome the dominance of resting-state particles in the dataset and to extract structural information from a low-abundance ensemble.
The reconstruction of the recovered population reached an overall resolution of roughly 4.5–5 Å, as reported in the source. At this resolution the authors were able to present the first fully intact structure of MetH with its physiological cobalamin cofactor present.
Because the reconstruction comes from a minor particle population that was only revealed through enhanced processing, it provides a complementary view to the more abundant resting-state structures in the original dataset. The recovered conformation is described as a reactivation state, reflecting its relevance to the enzyme’s cycle of turnover and periodic reactivation that depends on cofactor chemistry and domain motions.
The recovery of this minor particle population links several mechanistic elements of MetH function. Specifically, the results illuminate conformational equilibria that connect cofactor loading, catalytic turnover, and the enzyme’s reactivation pathway. Because MetH function requires large-scale domain rearrangements, access to a structure representing the reactivation state with physiological cofactor bound improves understanding of how structural ensembles underpin activity and reactivation.
More broadly, the study highlights that previously recorded cryo-EM datasets can contain biologically meaningful, low-abundance states that remain unrecognized without targeted reanalysis. The authors argue that modern computational picking and classification strategies, together with careful use of high-resolution information in orientation determination, can reveal these hidden states and expand mechanistic insight from existing data.
The work was performed by Haoyue Wang, Maxwell B Watkins, Pasa Suksmith, and Nozomi Ando, with correspondence directed to Nozomi Ando. The manuscript is posted as a preprint on bioRxiv and has not undergone peer review, as noted in the source. The authors declared no competing interests.
Funding reported in the source includes awards from the National Institute of General Medical Sciences and the Simons Foundation. The publication is made available under a CC-BY-NC-ND 4.0 International license, and the DOI assigned to the preprint is https://doi.org/10.64898/2026.07.27.741092.
Because this summary is based solely on the source preprint, detailed experimental parameters, maps, model coordinates, and supplementary analyses were not reported here beyond what appears in the source abstract and metadata.