Systematic Application of a Cellular Thermal Shift Assay for Inferring Inhibitor Binding Affinity to 17βHSD13 in a Humanized Liver Homogenate

Human loss-of-function variants in 17βHSD13 have been associated with reduced risk and progression of metabolic and alcohol-related liver disease. Although in vitro experiments implicate 17βHSD13 in the processing of steroidal and lipogenic substrates, establishing an unambiguous mechanistic link to disease causation remains challenging, and the absence of robust pharmacodynamic biomarkers complicates clinical dose selection based on tissue target occupancy. To address this, we implemented a medium-throughput cellular thermal shift assay (CETSA) in humanized mouse liver homogenates to quantify the binding of BI-3231, an uncompetitive 17βHSD13 inhibitor, and compared results with functional cell assays. We systematically profiled isothermal dose-response fingerprints (ITDRFCETSA) across 58-70.5 °C to account for temperature-induced relaxation of binding equilibria at elevated temperatures, which produced pronounced right-shifts in the apparent potency and confirmed NAD⁺-dependent binding. Complementary surface plasmon resonance (SPR) measurements across 5-42 °C defined the temperature dependence of NAD⁺ and BI-3231 binding. Extrapolating SPR affinities to the CETSA temperature range showed convergence with ITDRFCETSA fingerprints, supporting the use of the SPR-derived KD for BI-3231 (2.5 nM at 37 °C; pKD = 8.60 ± 0.07, 95% CI) to estimate the occupancy also in liver homogenates. This work provides a generalizable approach to quantify target engagement for CETSA-responsive drug targets, while underscoring that occupancy estimates for uncompetitive inhibitors must incorporate cofactor saturation.