Abstract
Light-sheet microscopy (LSM) has proven a useful tool in neuroscience to image whole brains with high frame rates at cellular resolution, and, in combination with tissue clearing methods, is often employed to reconstruct the cyto-architecture over the intact mouse brain. Inherently to LSM, however, residual opaque objects, always present to some extent even in extremely well optically cleared samples, cause stripe artifacts, which, in the best case, severely affect image homogeneity and, in the worst case, completely obscure features of interest. Here, demonstrating two example applications in intact optically cleared mouse brains, we report how Bessel beams reduce streaking artifacts and produce high-fidelity structural data for the brain-wide morphology of neuronal and vascular networks. We found that a third of the imaged volume of the brain was affected by strong striated image intensity inhomogeneity and, furthermore, a significant amount of information content lost with Gaussian illumination was accessible when interrogated with Bessel beams. In conclusion, Bessel beams produce high-fidelity structural data of improved image homogeneity and might significantly relax demands placed on the automated tools to count, trace or segment fluorescent features of interest.
Significance statement The standard way to measure neuronal structure in light-sheet microscopy applies Gaussian beams, a method prone to artifacts introduced by residual opaque objects present even in extremely well optically cleared samples. These objects cause streaky shadows, which severely affect image homogeneity and, furthermore, obscure features of interest. In measurements using Bessel beam illumination, however, high-fidelity imaging of micro-anatomical detail is restored. Since whole mouse brain data sets now routinely comprise several terabytes, automated tools to count, trace or segment features of interest are needed to extract meaningful insights. This is why high-fidelity structural imaging has the potential to relax computational demands on the algorithms used to turn data into knowledge.
Footnotes
The authors declare no competing financial interests.
This project received funding from the European Union’s H2020 research and innovation programme under grant agreements No. 720270 (Human Brain Project) and 654148 (Laserlab-Europe), and from the EU programme H2020 EXCELLENT SCIENCE - European Research Council (ERC) under grant agreement ID n.692943 (BrainBIT). The project has also been supported by the Italian Ministry for Education, University, and Research in the framework of the Flagship Project NanoMAX and of Eurobioimaging Italian Nodes (ESFRI research infrastructure), and by “Ente Cassa di Risparmio di Firenze” (private foundation).
This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license, which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.
Jump to comment: