Triamterene inhibits NRF2 Mut as a molecular glue

Background: NFE2-like bZIP transcription factor 2 (NRF2, encoded by NFE2L2 ) is a master regulator of the oxidative stress response. Under normal conditions, NRF2 is ubiquitinated and degraded via the proteasome following its interaction with Kelch-like ECH-associated protein 1 (KEAP1). Persistent activation of NRF2, commonly caused by gain-of-function NFE2L2 or loss-of-function KEAP1 mutations, is frequently observed in human esophageal squamous cell carcinoma (ESCC) and contributes to tumor progression and therapeutic resistance, underscoring the urgent need for effective therapeutic strategies against NRF2-addicted ESCC. Despite its importance, targeting NRF2 pharmacologically has proven challenging, as it is considered an 'undruggable' target due to its function as a transcription factor lacking classical enzymatic pockets or well-defined ligand-binding sites. In a previous high-throughput NRF2 reporter screen, we identified triamterene (TRM), an FDA-approved potassium-sparing diuretic, as a potential NRF2 inhibitor. The effects of TRM on NRF2 expression were validated by Western blot in two NRF2 Mut ESCC cell lines (KYSE70 and TE14). This study is aimed at understanding its efficacy in vivo and its mechanisms of action. Methods: In vivo efficacy of TRM was evaluated using both a genetically engineered mouse model (GEMM) and a cell line-derived xenograft (CDX) model. Mechanistic studies including biochemical assays (ubiquitination and proximity ligation assay [PLA]) and biophysical analyses (isothermal titration calorimetry [ITC], surface plasmon resonance [SPR], hydrogen-deuterium exchange mass spectrometry [HDX-MS], and molecular docking), were conducted to elucidate the mechanism of action of TRM. Results: TRM significantly suppressed tumor growth and NRF2 signaling in the NRF2 W24C ESCC CDX and the NRF2 D29H GEMM model. Mechanistically, TRM shortened the half-life of NRF2 W24C by promoting KEAP1-dependent ubiquitination and degradation. PLA confirmed enhanced KEAP1-NRF2 W24C interaction both in vitro and in vivo following TRM treatment. ITC demonstrated high-affinity binding between TRM and human recombinant KEAP1 (Kd = 0.975 µM). SPR analysis revealed that TRM enhanced the interaction between the DLG W24C peptide and the Kelch domain, but not between DLG WT and Kelch. HDX-MS identified two TRM-binding sites on KEAP1, including residues 397-417, consistent with molecular docking predictions implicating Arg 415 as a critical residue for TRM binding. Conclusion: TRM selectively inhibits NRF2 Mut ESCC by restoring KEAP1-mediated NRF2 degradation through direct binding to the Kelch domain of KEAP1. These findings support TRM as a promising therapeutic candidate for NRF2-addicted cancers. Ongoing studies aim to further define the structural basis of TRM-KEAP1 interaction and its role in modulating NRF2 stability and signaling.