A new method for network bioinformatics identifies novel drug targets for mucinous ovarian carcinoma

• Published in: NAR genomics and bioinformatics 2024-09, Vol.6 (3), p.lqae096
• Main Authors: Craig, Olivia, Lee, Samuel, Pilcher, Courtney, Saoud, Rita, Abdirahman, Suad, Salazar, Carolina, Williams, Nathan, Ascher, David B, Vary, Robert, Luu, Jennii, Cowley, Karla J, Ramm, Susanne, Li, Mark Xiang, Thio, Niko, Li, Jason, Semple, Tim, Simpson, Kaylene J, Gorringe, Kylie L, Holien, Jessica K
• Format: Article
• Language: English
• ISSN: 2631-9268; 2631-9268
• DOI: 10.1093/nargab/lqae096

Abstract Mucinous ovarian carcinoma (MOC) is a subtype of ovarian cancer that is distinct from all other ovarian cancer subtypes and currently has no targeted therapies. To identify novel therapeutic targets, we developed and applied a new method of differential network analysis comparing MOC to benign mucinous tumours (in the absence of a known normal tissue of origin). This method mapped the protein-protein network in MOC and then utilised structural bioinformatics to prioritise the proteins identified as upregulated in the MOC network for their likelihood of being successfully drugged. Using this protein-protein interaction modelling, we identified the strongest 5 candidates, CDK1, CDC20, PRC1, CCNA2 and TRIP13, as structurally tractable to therapeutic targeting by small molecules. siRNA knockdown of these candidates performed in MOC and control normal fibroblast cell lines identified CDK1, CCNA2, PRC1 and CDC20, as potential drug targets in MOC. Three targets (TRIP13, CDC20, CDK1) were validated using known small molecule inhibitors. Our findings demonstrate the utility of our pipeline for identifying new targets and highlight potential new therapeutic options for MOC patients. Graphical Abstract Graphical Abstract