Discovery and optimization of N-phenylbenzenesulfonamide analogues as novel ALKBH5 inhibitors for AML therapy.

Acute myeloid leukemia (AML) is an aggressive malignancy with poor prognosis, primarily driven by leukemic stem cells (LSCs). The RNA demethylase ALKBH5 is aberrantly overexpressed in AML and critically involved in leukemogenesis and LSC maintenance. Herein, we designed and synthesized fifty-seven N-phenylbenzenesulfonamide derivatives as potential ALKBH5 inhibitors. Within this series, G4 emerged as a potent ALKBH5 inhibitor that selectively suppressed ALKBH5-high AML cells over ALKBH5-knockout cells and normal human PBMCs. In G4-treated AML cells, cell cycle arrest and apoptosis were accompanied by dose-dependent reductions in MCL-1 and Bcl-xL protein levels. Genetic knockout of ALKBH5 markedly attenuated G4-induced cytotoxicity, supporting the ALKBH5 dependence of G4 activity. CETSA and SPR further supported a direct interaction between G4 and ALKBH5. MD simulations and 2-OG competition assays suggested the involvement of the 2-OG pocket in G4 recognition. Mechanistically, G4 directly inhibited ALKBH5 and elevated global m6A levels. Consistently, G4 dose-dependently downregulated TACC3, a known ALKBH5 target, impaired LSC self-renewal, and showed significant in vivo anti-AML efficacy. Collectively, these findings identify G4 as an ALKBH5 inhibitor that impairs LSCs self-renewal with robust in vivo efficacy, providing a promising lead for further optimization toward AML therapy.