Revealing the nervous system requirements of Alzheimer disease risk genes in Drosophila

Most Alzheimers disease (AD) susceptibility genes have poorly understood roles in the central nervous system (CNS). To address this gap, we systematically characterized 100 conserved candidate AD risk genes using a cross-species strategy in the fruit fly, Drosophila melanogaster. Genes were prioritized based primarily on human functional genomic evidence. We generated custom, loss-of-function alleles for each of the conserved fly orthologs. Most of the genes (80%) are expressed in the adult brain, including 24 neuron- and 13 glia-specific expression patterns. Overall, we identify 50 candidate AD risk gene homologs with requirements for CNS structure or function, including 18 whose loss of function causes neurodegeneration (e.g., Snx6/SNX32 and ClC-a/CLCN1), 35 required for neurophysiology (e.g., Arr1/ARRB2, stai/STMN4), and 8 with diminished CNS resilience following a thermal or mechanical stress (e.g., cindr/CD2AP, Amph/BIN1). In a parallel screen, we found 28 AD risk gene homologs (e.g, Ets98B/SPI1, Yod1/YOD1) that modify the neurotoxicity of either amyloid-{beta} peptide or tau protein, which aggregate to form AD pathology. To translate our findings back to human AD, we developed and deployed oligogenic risk scores based on gene clusters with shared nervous system phenotypes in flies, pinpointing functional pathways that differentially drive AD risk. Our results--available online via the Alzheimers Locus Integrative Cross-species Explorer (alice.nrihub.org)--reveal novel nervous system requirements for dozens of AD risk genes and may enable dissection of causal heterogeneity in AD.