Synthetic lethal screening identifies existing drugs with selective viability effects on Neurofibromatosis type-1 model systems

Abstract

ABSTRACTNeurofibromatosis type 1 (NF1) is a genetic multi-system disorder. Symptoms include near universal benign neurofibromas, as well as malignant tumours, including generally fatal malignant peripheral nerve sheath tumours. There are limited therapies for any NF1-associated tumours; therefore, there is a clear clinical need to discover new drugs that specifically target NF1-deficient tumour cells. Using a Drosophila NF1-KO cell model, we used synthetic lethal screening to identify candidate drug targets for NF1-deficient tumours and performed statistical enrichment analysis to identify further targets. We then assessed the top 72 candidate synthetic lethal partner genes to NF1 using Variable Dose Analysis, resulting in 15 candidate genes that decreased NF1-KO viability by >10% and were novel druggable targets for NF1. Autophagy inhibitors Chloroquine (CQ) and Bafilomycin A1 resulted in a significant reduction in NF1-KO cell viability, which was conserved in a panel of human NF1 mutant cell lines. AZT and Enzalutamide also selectively reduced NF1 mutant cell viability in human cell lines. Furthermore, the effect of CQ was conserved in a Drosophila NF1-mutant in vivo model. This study highlights two key points: 1) The use of Drosophila cells as a model to screen for drugs specifically targeting NF1 mutant cells was highly successful as candidate interactions were conserved across a panel of human NF1 mutant cells and an in vivo fly NF1 mutant model, and 2) NF1-deficient cells have vulnerability to disruption of the autophagy pathway, telomerase activity, and AR activity. These pathways/drugs represent promising targets for the potential treatment of NF1-associated tumours.