Characterising the enzymology of CP506, a second-generation analogue of the hypoxia activated prodrug PR-104A

Abstract

The absence of oxygen, or hypoxia, is a phenomenon which occurs in cancerous lesions whereby the demand on oxygen exceeds its supply. Hypoxia is a well-validated biomarker which is associated with negative prognostic outcomes, as well a major factor in contributing to treatment resistance. Hypoxia activated prodrugs (HAPs) have been utilized to specifically target the hypoxic fraction in tumours, by activation to an oxygen sensitive nitro-radical intermediate through the action of one-electron reductases. PR-104A was designed as a HAP which was investigated in acute myeloid leukaemia patients. PR-104A failed in these trials due to excessive off-target myelotoxicity. The principal hypothesis behind this was PR-104A activation via an oxygen-insensitive mechanism mediated by AKR1C3, an androgenic enzyme which is expressed strongly in bone marrow progenitor cells. CP506 is a second-generation analogue to PR-104A, designed specifically to circumvent activation by AKR1C3. The extended enzymology of CP506 is poorly defined, despite being a critical factor in determining drug sensitivity This project sought to characterise the enzymology of CP506. A library of F279 V5 expression vectors containing the sequences of 62 genes which encoded either an FMN or FAD (or both) domain in their open reading frame were cloned using Gateway Cloning ® technology was created, 32 of which were synthesized prior. Each plasmid was transfected into the H1299 cell line, a poor wild-type metaboliser of CP506, to create a library of cell lines overexpressing each candidate flavoreductase. Expression of proteins was determined by V5-tag-on-demand ® system. 11 flavoreductases were found to sensitise H1299 to the anti-proliferative effect of CP506 (as determined by an IC50 assay) under anoxic exposure, but only 2 flavoreductases (POR, STEAP4) were found to increase metabolism. Further validation showed sensitisation in only 3 of the 11 (POR, STEAP4, TXNRD1) flavoreductases was statistically significant. CP506 demonstrated a distinct metabolic profile in response to flavoreductase overexpression when compared to PR-104A. Knockdown of siRNA showed POR metabolises CP506 in wildtype HCT-8-Sa and SiHa, but STEAP4 metabolises CP506 in SiHa only. Analysis of flavoreductase expression in clinical tumour samples provides a compelling opportunity for CP506 activity in prostate cancer.