Unbiased Sampling and Volume-Sorting of Origin-Specific Terminals Using SBEM Image Stacks

While massive sets of serially imaged electron microscopy stacks are required for large-scale brain circuitry reconstructions, smaller and more easily attainable image stacks can be used for studying the circuitry within the given nuclei and the contributions of different synaptic boutons whose origins can be identified by morphological or morphometric criteria using random sampling strategies. Here we describe an unbiased axon terminal sampling procedure for reconstructing and identifying morphological and morphometric properties of terminal boutons that contribute to the circuitry in a sample volume of the tissue. We also describe a statistical modeling tool for volume-based sorting of terminal boutons using the cutoff criteria obtained with the unbiased terminal sampling (UTS) approach and an independent dataset. The UTS approach, as described, yields a variety of measures, including bouton volume, synaptic zone area, and estimated synapse density, and may allow researchers to obtain a quantification of the synaptic circuitry in a given brain region in a relatively short time and then compare across any given variable such as age, development, or disease progression.