Interrogating neuronal connectivity through serial two-photon tomography of the whole mouse brain ex-vivo
File(s)
Author(s)
Moore, Gerald
Type
Thesis
Abstract
Ex-vivo, serial two-photon (STP) tomography was used to image the entire mouse brain to interrogate neuronal connectivity across local and distal brain volumes. Large terabyte-sized data sets presented many analytical challenges. To this end, the following pipelines were created and validated; stitching of tile image data, registration of data with common coordinate framework, and a deep learning cell detection algorithm that was comparable to human performance. These pipelines were developed to facilitate two scientific studies. First, the unknown connectivity of local interneurons of the dorsal lateral geniculate nucleus (LGd) of Sox14cre/+ mice were mapped using rabies virus tracing, revealing previously unreported input from the GABAergic reticular nucleus (RT), suggesting a double inhibition mechanism of modulation onto local thalamocortical relay cells. In-vitro patch-clamp recordings confirmed functional connectivity. Thalamic interneurons in the Sox14gfp/+ mouse had high prevalence in first order (FO) sensory nuclei, challenging the notion that interneurons in the mouse thalamus are constrained to visual thalamus only. In-vitro patch-clamp recordings showed functional interneuron-interneuron connectivity as suggested by interneuron morphological analysis. Second, STP tomography and magnetic resonance imaging (MRI) were explored for correlating age-related cognitive changes with anatomical connectivity. Female C57BL/6J mice were subjected to a novel nine-arm radial maze task, and object location and recognition tasks at 4, 8, 12 and 18 months of age. Mice showed remarkable retention of cognitive performance throughout age (n=58). Unique arm choice presented one of many measures that showed performance predictability at 18 months given 4 months performance (R2=0.6166, p<0.0001, n=22). A combination of unique arm sequence length, cumulative arm time and velocity measures showed improved predictability for 18 months performance given 4 months performance (R2=0.6239, p<0.0001, n=22). This level of predictability has been previously seen in Lothian cohort human studies. STP tomography and MRI were used to demonstrate mapping brain connectivity changes with age.
Version
Open Access
Date Issued
2020-09
Date Awarded
2021-05
Copyright Statement
Creative Commons Attribution NonCommercial Licence
License URL
Advisor
Brickley, Stephen
Schultz, Simon
Publisher Department
Life Sciences
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)