FL118, acting as a ‘molecular glue degrader’, binds to dephosphorylates and degrades the oncoprotein DDX5 (p68) to control c‐Myc, survivin and mutant Kras against colorectal and pancreatic cancer with high efficacy

BACKGROUND: Pancreatic ductal adenocarcinoma (PDAC), a difficult‐to‐treat cancer, is expected to become the second‐largest cause of cancer‐related deaths by 2030, while colorectal cancer (CRC) is the third most common cancer and the third leading cause of cancer deaths. Currently, there is no effective treatment for PDAC patients. The development of novel agents to effectively treat these cancers remains an unmet clinical need. FL118, a novel anticancer small molecule, exhibits high efficacy against cancers; however, the direct biochemical target of FL118 is unknown. METHODS: FL118 affinity purification, mass spectrometry, Nanosep centrifugal device and isothermal titration calorimetry were used for identifying and confirming FL118 binding to DDX5/p68 and its binding affinity. Immunoprecipitation (IP), western blots, real‐time reverse transcription PCR, gene silencing, overexpression (OE) and knockout (KO) were used for analysing gene/protein function and expression. Chromatin IP was used for analysing protein‐DNA interactions. The 3‐[4,5‐dimethylthiazol‐2‐yl]‐2,5‐diphenyltetrazolium bromid assay and human PDAC/CRC cell/tumour models were used for determining PDAC/CRC cell/tumour in vitro and in vivo growth. RESULTS: We discovered that FL118 strongly binds to dephosphorylates and degrades the DDX5 oncoprotein via the proteasome degradation pathway without decreasing DDX5 mRNA. Silencing and OE of DDX5 indicated that DDX5 is a master regulator for controlling the expression of multiple oncogenic proteins, including survivin, Mcl‐1, XIAP, cIAP2, c‐Myc and mutant Kras. Genetic manipulation of DDX5 in PDAC cells affects tumour growth. PDAC cells with DDX5 KO are resistant to FL118 treatment. Our human tumour animal model studies further indicated that FL118 exhibits high efficacy to eliminate human PDAC and CRC tumours that have a high expression of DDX5, while FL118 exhibits less effectiveness in PDAC and CRC tumours with low DDX5 expression. CONCLUSION: DDX5 is a bona fide FL118 direct target and can act as a biomarker for predicting PDAC and CRC tumour sensitivity to FL118. This would greatly impact FL118 precision medicine for patients with advanced PDAC or advanced CRC in the clinic. FL118 may act as a ‘molecular glue degrader’ to directly glue DDX5 and ubiquitination regulators together to degrade DDX5.