Light-sheet microscopy · Grey Matter

Light-sheet microscopy lights a sample from the side with a thin sheet of light and films the fluorescence of that one plane with a camera set at right angles, and on brains made transparent by clearing it records a whole organ, plane after plane, at the scale of single cells.

Light-sheet microscopy. Light-sheet microscopy lights a sample from the side with a thin sheet of light and films the fluorescence of that one plane with a camera set at right angles, and on brains made transparent by clearing it records a whole organ, plane after plane, at the scale of single cells.

The idea goes back to the ultramicroscope of Henry Siedentopf and Richard Zsigmondy in 1903, which lit gold particles in liquid from the side through a narrow slit so that light scattered by them could be seen from above. Jan Huisken and Ernst Stelzer's group at EMBL turned it into SPIM in 2004 to film live embryos, and in 2007 Hans-Ulrich Dodt's group combined it with clearing to image whole fixed mouse brains. Clearing removes the lipids that scatter light and soaks the tissue in a liquid matched to its refractive index, so a sheet can pass through a centimetre of brain. Only the filmed plane is lit, so the method is fast and bleaches little. The open-hardware mesoSPIM, begun in 2015 by Fabian Voigt and Fritjof Helmchen in Zurich and built also at the Wyss Center in Geneva, images a cleared mouse brain of about a cubic centimetre in 7 to 8 minutes.

It fills the gap between MRI and the connectome. It shows single cells, vessels and long projections across a whole organ, while synapses, tens of nanometres across, still need electron microscopy.

It sees only what was labelled. Fluorescent proteins, or antibodies and dyes that must diffuse through the whole sample, decide what appears.

Cleared tissue is fixed tissue. It records where cells and fibres sat when the animal died or the tissue was taken, and nothing of what they were doing.

Live brains work only when they are already transparent. In larval zebrafish it captured more than 80 % of neurons at 0.8 volumes a second through calcium imaging.

Light-sheet microscopy trades life for completeness.

Making a brain transparent lets every labelled cell in it be seen in place, at the cost of the activity that made it a brain.

Questions: Can a light-sheet microscope map the wiring of a whole brain? It maps the long-range wiring and stops short of the synapses. A cleared mouse brain can be imaged whole in minutes at a few micrometres, enough to follow labelled neurons and their projections from one region to another. A synapse is tens of nanometres across, so deciding which cell touches which needs electron microscopy at a few nanometres, the method behind the fly connectome. The two are complementary: light-sheet gives the whole organ at cellular scale quickly, electron microscopy gives every contact in a small volume slowly. Can light-sheet microscopy record a whole living brain at work? Only in animals small and transparent enough to need no clearing. In 2013 a light-sheet microscope recorded the brain of a larval zebrafish carrying a calcium indicator, capturing more than 80 % of its neurons at single-cell resolution, the whole volume about 0.8 times a second. At that rate it follows the slow calcium signal, too coarse to separate single spikes. A mouse or human brain scatters light too much, so in those the method works on cleared, fixed tissue, which keeps the cells and loses their activity.