[A05*] Systems-guided in vivo identification of causal mechanisms in Alzheimer’s disease from human brain transcriptomic profiles

Author(s): Grant Mangleburg, Baylor College of Medicine; Omar El Fadel, Baylor College of Medicine; Anh Le, Baylor College of Medicine; Ismael Al-Ramahi, Baylor College of Medicine; Ying-Wooi Wan, Baylor College of Medicine; Juan Botas, Baylor College of Medicine; Joshua Shulman, Baylor College of Medicine //

ABSTRACT: Alzheimer’s disease (AD) is the most prevalent cause of dementia. Large RNA-Sequencing data sets have provided detailed molecular profiles of AD, but they do not give insight into the mechanistic underpinnings of disease process or differentiate between causal and downstream changes in gene expression. In this study, we combined bioinformatic prioritization and genetic screens in Drosophila melanogaster to identify human brain transcriptional changes that exacerbate or ameliorate neurotoxicity induced by Tau or amyloid-β (Aβ), the proteins which form the hallmark pathologic aggregates that define AD. First, we used human brain RNA-Seq data to nominate candidate drivers of large, AD-associated coexpression networks based on multiple bioinformatic criteria. We then tested nominated drivers for modification of Tau or Aβ neurotoxicity in vivo using transgenic Drosophila in a locomotor screen. Through our screen, we determined that modulating expression of 144 of 357 (40.3%) nominated human driver genes modified Tau or Aβ neurotoxicity. We additionally found that combining multiple bioinformatic prioritization criteria performed better than individual criteria for identification of genetic modifiers. We identified a module with 51.9% of its prioritized genes modifying Tau/Aβ neurotoxicity in our screen. We used this module with our screen data to identify a subnetwork containing genes involved in synaptic function including GRIN2A, GRIN2B, KCNJ6, GABRB2, and GABRB3. This synaptic subnetwork is downregulated in human AD and contains multiple knockdown suppressors of Tau/Aβ neurotoxicity. We speculate that the subnetwork represents a network-level compensatory downregulation of synaptic genes to ameliorate synaptic toxicity in AD. By integrating human coexpression networks and Drosophila screen data, our results establish causal chains of events connecting Tau/AB with neurodegeneration through network-level gene expression changes.

Source of Funding: NIA F31 AG067677-02