We recently demonstrated the feasibility of ABCG2 inhibition, when coupled with a cytotoxin, as a strategy to improve acute myeloid leukemia therapy using a murine Abcg2 knockout model (1). A high throughput phenotypic screen (based upon inhibition of ABCG2 transport) with a goal of re‐purposing known therapeutic compounds, was used to identify new ABCG2 inhibitors. Among them was a kinase inhibitor (KI) that had not been previously reported to interact with ABCG2. KIs have been reported as ABCG2 inhibitors, possibly by interacting with the transporter, but also by affecting its phosphorylation state and/or subcellular localization. We systematically interrogated this new ABCG2 inhibitor and compared to known ABCG2 kinase inhibitors.Despite numerous reports of inhibition of ABCG2 by KI, it is unknown if a relationship exists between binding to ABCG2 and transport inhibition. Furthermore, it is not known if ABCG2 inhibition is strictly related to a compound's physiochemical properties (e.g., LogP) or a specific pharmacophore or the binding pocket on ABCG2. To determine if a relationship exists between KI binding to ABCG2 and transport inhibition, we employed the membrane protein cellular thermal shift assay (CETSA). CETSA is based on the principle that ligand binding induces target protein thermal stability, leading to increased temperature required for protein unfolding. This method enables the determination of in cellulo ligand binding to the native protein. The ABCG2 melting temperature (Tm) was first established from the sigmoidal thermal denaturation curve. We then used Ko143, a well known ABCG2 inhibitor, and determined that Ko143 dramatically elevated the Tm of ABCG2. We then screened over a dozen known KIs (including our newly identified KI) for their ability to inhibit the transport by ABCG2 of the specific substrate, the fluorescent dye pheophorbide a. In parallel, the EC50 of KI binding to ABCG2 was determined by CETSA. Among many of the KI, a strong relationship was observed between the EC50s for KI binding to ABCG2 and the IC50s for their transport inhibition. Future studies will determine if the binding to ABCG2 depends upon either ATPase activity or ATP‐binding. Furthermore, from these findings, and the recently published ABCG2 structure, a substrate binding model was developed. Inexplicably some KI inhibited ABCG2, but displayed either a weak or no thermal shift, indicating little or no binding to ABCG2. Because some kinases reportedly affect ABCG2 subcellular localization, we are determining if these KIs affect the plasma membrane localization of ABCG2 by using a combination of confocal microscopy and surface biotinylation.In conclusion, the CETSA assay was used, for the first time, to determine the relationship between binding and inhibition of ABCG2 transport for many ABCG2 inhibitors (represented by KIs here). This advance allows the classification of inhibitors into those that bind vs those that inhibit by a mechanism that is independent of binding to ABCG2.This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.
Read full abstract