New methods of correlated light and electron microscopy provide a multi-molecular overlay for large-volume connectomic data in the brain

Abstract

Mapping neuronal networks underlying behavior has become a central focus in neuroscience. Serial section electron microscopy (ssEM) is an approach that has been used to reveal the fine structure of synaptically connected neuronal networks (“connectomics”). However, this structural data doesn’t provide essential molecular information required to identify particular cell types nor to infer functional properties. Volumetric correlated light and electron microscopy (vCLEM) combines ssEM and volumetric fluorescence microscopy to overlay molecular information onto ssEM datasets. I, in collaboration with my colleagues, developed new multicolor vCLEM approaches using either transgenic mice in which fluorescent proteins are expressed in specific cell types or detergent-free immunofluorescence by small single-chain variable fragment (scFv) and nanobody immuno-probes. My results show both excellent ultrastructure and the superimposition of many fluorescent molecular labels. In one project, I generated a vCLEM dataset of the hypothalamic MPOA region using a transgenic mouse to reveal the morphology and connectivity of a molecularly defined cell type that regulates parenting behavior. In another project, I generated a total of 25 fluorescent scFvs that targeted ten useful markers for brain studies (GFP, GFAP, calbindin, parvalbumin, Kv1.2, VGluT1, PSD-95, neuropeptide Y, somatostatin, and 8-oxo-dG). In addition, I generated five fluorescent nanobodies to target Alzheimer's disease-related molecules (amyloid-β and phosphorylated tau). With these assorted probes, I then investigated several brain regions. First, five different fluorescent scFv probes were imaged in the cerebellar cortex with linear unmixing of confocal image stacks. The same sample was then stained, sectioned, and imaged with ssEM. This approach revealed a poorly described cell type in the cerebellum, different molecular and structural types of mossy fiber terminals, and the subcellular localization of ion channels in a particular class of axons. Second, three nanobody probes were imaged in the hippocampus of an Alzheimer’s disease model mouse followed by ssEM. This dataset revealed a number of ultrastructural abnormalities. It identified intracellular and extracellular localizations of amyloid-β and intracellular phosphorylated tau in novel locations. These approaches hold promise for routine overlays of molecular information directly onto connectomic data of the same tissue samples.