Untargeted Detection of HIF Stabilizers in Doping Samples: Activity-Based Screening with a Stable In Vitro Bioassay

Hypoxia-inducible factor (HIF) stabilizers are listed in the World Anti-Doping Agency's prohibited list as they can increase aerobic exercise capacity. The rapid pace of emergence of highly structurally diverse HIF stabilizers could pose a risk to conventional structure-based methods in doping control to detect new investigational drugs. Therefore, we developed a strategy that is capable of detecting the presence of any HIF stabilizer, irrespective of its structure, by detecting biological activity. Previously developed cell-based HIF1/2 assays were optimized to a stable format and evaluated for their screening potential toward HIF stabilizers. Improved pharmacological characterization was established by the stable cell-based formats, and broad specificity was demonstrated by pharmacologically characterizing a diverse set of HIF stabilizers (including enarodustat, IOX2, IOX4, MK-8617, JNJ-42041935). The methodological (in solvent) limit of detection of the optimal HIF1 stable bioassay toward detecting the reference compound roxadustat was 100 nM, increasing to 50-100 ng/mL (corresponding to 617-1233 nM in-well) in matching urine samples, owing to strong matrix effects. In a practical context, a urinary limit of detection of 1.15 mu g/mL (95% detection rate) was determined, confirming the matrix-dependent detectability of roxadustat in urine. Pending optimization of a universal sample preparation strategy and/or a methodology to correct for the matrix effects, this untargeted approach may serve as a complementing method in antidoping control, as theoretically, it would be capable of detecting any unknown substance with HIF stabilizing activity.