Abstract
Pediatric low-grade gliomas (PLGGs) are the most common brain tumors in children, with varying degrees of brain invasion. Recent whole-genome sequencing has identified a rare gene fusion involving RAF1, a RAF isoform. Unlike other RAF fusions, RAF1 fusions are resistant to existing RAF inhibitors. Therefore, aside from surgical resection with adjuvant chemotherapy and radiation therapy, there are few targeted therapeutic alternatives for RAF1-fusion-driven PLGGs. Despite the prevalence and challenges this disease presented, our understanding of PLGGs was limited by a lack of genetic models. We ultimately picked Drosophila melanogaster as our model organism due to the conservation of major signaling pathways between flies and humans. Furthermore, this connection between humans and flies, coupled with other technical advantages associated with this model organism, like short generation cycle and its powerful genetic toolbox, makes Drosophila melanogaster an ideal organism to study the genesis and progression of PLGGs. With the help of the GAL4/UAS system, we established four fusion-driven PLGG fly genetic models and found that glial overexpression of QKI-RAF1, a fusion gene in pilocytic astrocytomas, induces an invasion-like phenotype with aberrant glial migration. This migration defect was suppressed by glial overexpression of repulsive guidance signaling receptors Robo2 or PlexA/B, indicating the dysregulation of repulsive guidance signaling pathways. Immunostaining coupled with quantitative analysis revealed that Robo2 expression is downregulated in migrating tumor cells in flies, which is recapitulated in mouse astrocytes overexpressing QKI-RAF1 and PLGG patients with RAF fusions. We further broaden our findings by profiling the tumor transcriptomes, revealing potential downstream effectors, including the G protein-coupled receptor GPR180/CG9304, and inhibition of which suppresses tumor invasion in flies. Taken together, we present the PLGG fly model system, leading to the discovery of Robo2, Plexins, and GPR180/CG9304 as potential therapeutic targets.