P03-09

Large-scale single nucleus RNA-seq analysis of Lewy body diseases subtypes

Supakorn PONGPAKDEE *, Kosuke HASHIMOTO, Kenji MIZUGUCHI

Osaka University
( * E-mail: u734967b@ecs.osaka-u.ac.jp )

Single cell/nucleus data analysis become advancing to incorporate with biomarker discovery in neurodegenerative diseases, particularly on Lewy body diseases (LBDs) that compose of Parkinson’s disease (PD), Dementia with Lewy body (DLB) and Parkinson’s disease dementia (PDD). While these diseases share a common neuropathological hallmark overlapping in cortical and subcortical brain area by aggregated α-synuclein protein, called Lewy body (LB), and associate with the prevalence of dementia in Alzheimer’s disease (AD), the clinicopathological progressive trajectory of LB between brainstem and olfactory bulb tract initially represents differently among LBDs heterogeneity, suggesting diverse LB pathology distributions. Although several studies of transcriptomic profiling for LBDs have highlighted potential molecular therapeutic targets, the underlying mechanisms in specific cell types to distinguish LBDs remain elusive.
This study aims to identify distinct genes marker for LBDs subtypes using large-scale transcriptomic single nucleus RNA-seq (snRNA-seq) data analysis. Postmortem snRNA-seq data derived from substantia nigra (SN) and midbrain was retrieved from eight public datasets, including unaffected control (CTL), PD, PDD and DLB samples (n=66, 55, 19 and 4, respectively). Two of eight datasets were used as reference to transfer cell-type annotations, and all datasets were integrated into a single matrix using previous probabilistic deep learning model (scVI). Overall, we obtained a single expression matrix containing approximately 800,000 cells with 18,632 genes. These cells were classified into seven major cell types: neuron, oligodendrocyte, oligodendrocyte progenitor cell, astrocyte, dopaminergic neurons, microglia and endothelial cell. The differential expression (DE) and gene ontology (GO) enrichment analysis revealed significant gene sets associated with mitochondrial function, extracellular organization and membrane protein co-translation, between disease conditions. Due to limited data availability, additional collection is necessary to enhance the representation of various subtypes in the dataset.