Multiplex Cell Type-Specific Nuclei Isolation from Archived Human Brain Tissue

Abstract

Acquiring cell type-specific transcriptomic resolution from archived human brain tissue is key to obtaining a more in-depth understanding of neurodegenerative disease pathogenesis. Fluorescence-activated nuclei sorting (FANS) has emerged as a robust method by which to selectively purify cell type-specific populations of human brain nuclei for downstream transcriptomic interrogation. To date, however, FANS-based purification of human brain cell nuclei has been restricted to neurons, owing to the prominent lack of robust cell type-specific nuclei markers for the other brain cell populations. The present thesis aimed to establish a proof-of-concept for the simultaneous multiplex isolation of discrete cell type-specific nuclei populations from archived human brain tissue that can be used to interrogate transcriptomes of individual brain cell populations both within and between human brain diseases. Immunohistochemical tissue staining was initially used to validate the cell type-specificity of NeuN, PU.1, SOX9, Olig2, ERG and MafF as independent nuclear markers of human neurons, microglia, astrocytes, oligodendrocytes, endothelial cells and pericytes, respectively. In parallel with these marker validations, a novel nuclei isolation protocol was developed and comprehensively optimised to ensure maximal yield, viability and downstream transcriptomic integrity of nuclei isolated from archived middle temporal gyrus specimens. NeuN, PU.1 and MafF were then integrated with the optimised nuclei isolation protocol to demonstrate that human brain nuclei can be multiplex-immunolabelled and simultaneously sorted by FANS into discrete, transcriptomically viable nuclei populations for downstream analysis. This research details and validates a novel multiplex approach that can be used to interrogate the complex cell type-specific transcriptomic landscape of the human brain. By facilitating the simultaneous enrichment of multiple cell type-specific nuclei populations, multiplex FANS also permits the highly-efficient localisation of differential gene expression profiles to lowlyrepresented cell types. The compatibility of this technique with archived human brain tissue will permit the acquisition of cell type-specific transcriptomic resolution from both normal and diseased human brains stored in brain banks the world over.