FluoEM, virtual labeling of axons in three-dimensional electron microscopy data for long-range connectomics

Area of Science:

• Neuroscience
• Computational Biology
• Microscopy

Background:

• Electron microscopy (EM) struggles to label multiple long-range axonal inputs in dense mammalian brain datasets due to limited color space.
• Identifying axonal origins and targets in complex neural circuits is crucial for understanding brain function.

Purpose of the Study:

• To introduce FluoEM, a novel method for identifying multi-color fluorescently labeled axons within dense EM data.
• To enable virtual color labeling of axons in EM datasets without fiducial markers or label conversion.

Main Methods:

• Correlated tissue imaging and computational matching of neurite reconstructions from fluorescent light microscopy (LM) and EM data.
• Utilizing unique axonal trajectories in dense mammalian neuropil for identification.
• Applying the method to mouse cortical datasets.

Main Results:

• FluoEM successfully identifies fluorescently labeled axons in 3D EM data.
• Identification is reliable in EM datasets as small as 40-50 µm in extent.
• The method was exemplified for tracing long-distance axonal inputs into the mouse cerebral cortex layer 1.

Conclusions:

• FluoEM overcomes EM's limited color label space for identifying multi-color fluorescent axons.
• This computational approach facilitates detailed circuit reconstruction and analysis in dense brain EM data.
• FluoEM is a powerful tool for studying long-range axonal connectivity in the mammalian brain.