ABSTRACT A NOVEL HSP90 INHIBITOR TO DISRUPT HSP90/p50CDC37 COMPLEX FOR PANCREATIC CANCER THERAPY by Tao Zhang Chair: Duxin Sun Pancreatic cancer is an aggressive disease with multiple biochemical and genetic alterations. In particularThus, Hsp90 has emerged as a promising target for pancreatic cancer therapeutics because it regulates the stabilization and conformational maturation of various oncogenic proteins simultaneously. Traditional Hsp90 inhibitors inhibit Hsp90 function through the blockage of nucleotide binding to Hsp90. In the present study, disruption of Hsp90/p50Cdc37 complex was revealed and confirmed as a novel and viable mechanism of Hsp90 inhibition. We reported that celastrol, a quinone methide triterpene purified from tripterygium wilfordii Hook F., could disrupt Hsp90/p50Cdc37 complex. A series of in silico, in vitro and in vivo assays were established to characterize the interaction between Hsp90 and p50Cdc37 and to investigate the underlying mechanism of the disruption by celastrol. In a reconstituted protein system, GST pull-down and ELISA demonstrated that p50Cdc37 was bound to ADP-bound/nucleotide-free Hsp90 but not to ATP-bound Hsp90, and that celastrol inhibited Hsp90/p50Cdc37 complex formation whereas the classical Hsp90 inhibitors hadve no effect. Celastrol inhibited Hsp90 ATPase activity without blocking ATP binding. Molecular docking and molecular dynamic simulations showed celastrol blocked the critical interaction between Hsp90 and p50Cdc37. Proteolytic fingerprinting indicated celastrol bound to the Hsp90 C-terminal domainal domain to protect it from trypsin digestion. In pancreatic cancer cells, co-immunoprecipitation confirmed that celastrol inhibited the Hsp90 and p50Cdc37 interaction and induced Hsp90 client protein degradation. Celastrol resulted in pancreatic cancer cell apoptosis in vitro and significantly inhibited human pancreatic cancer xenograft growth in nude mice. Moreover, celastrol effectively suppressed tumor metastasis in a RIP-Tag2 transgenic mouse model with pancreatic islet cell carcinogenesis. These data suggest that celastrol represents a new class of Hsp90 inhibitors that can disrupt theby disrupting Hsp90/p50Cdc37 complex to modulate the chaperone activity. This alternative strategy of Hsp90 inhibition may offer more advantages than classical Hsp90 inhibition and further investigations are warranted to verify its therapeutic potentials.