Volumetric imaging of fast cellular dynamics with deep learning enhanced bioluminescence microscopy

• DOI: 10.1038/s42003-022-04292-x
• Publication Date: 2022-12-03T15:48:34Z
• Authors: Luis Felipe Moral...; Adriana Carolina ...; Gustavo Castro-Ol...; Li-Chun Lin; Malak El-Quessny; Montserrat Porta-...; Jacqueline Severino; Laura Battle Morera; Valeria Venturini; Verena Ruprecht; Diego Ramallo; Pablo Loza-Alvarez; Michael Krieg
• Keywords: Bioluminescence imaging; Dynamics

Abstract Bioluminescence microscopy is an appealing alternative to fluorescence microscopy, because it does not depend on external illumination, and consequently neither produce spurious background autofluorescence, nor perturb intrinsically photosensitive processes in living cells animals. The low photon emission of known luciferases, however, demands long exposure times that are prohibitive for imaging fast biological dynamics. To increase the versatility bioluminescence we present improved low-light microscope combination with deep learning methods image extremely photon-starved samples enabling subsecond exposures timelapse volumetric imaging. We apply our method subcellular dynamics mouse embryonic stem cells, epithelial morphology during zebrafish development, DAF-16 FoxO transcription factor shuttling from cytoplasm nucleus under stress. Finally, concatenate neural networks denoising light-field deconvolution resolve intracellular calcium three dimensions freely moving Caenorhabditis elegans .