Beyond KEAP1: The Context-Specific NRF2 Partner Code in Disease and Therapy

Nuclear factor erythroid 2-related factor 2 (NRF2) has traditionally been framed as a Kelch-like ECH-associated protein 1 (KEAP1)-regulated stress-response transcription factor, but three observations now require a broader framework: NRF2 turnover is controlled by parallel E3 ligase systems; transcriptional output can be limited by coactivator assembly despite unchanged NRF2 abundance; and NRF2 activation can be beneficial or harmful depending on disease context, as illustrated by lung cancer models in which NRF2 paradoxically promotes metastasis through BTB and CNC homology 1 (BACH1) stabilization. We synthesize these observations into an NRF2 partner-code framework in which NRF2 acts as a context-dependent transcriptional platform assembled through four partly independent modules: a degradation module (KEAP1; β-transducin repeat-containing protein, β-TrCP; HMG-CoA reductase degradation protein 1/synoviolin 1, Hrd1/SYVN1; WD repeat-containing protein 23/DDB1- and CUL4-associated factor 11, WDR23/DCAF11); a cytoplasmic scaffold module (p62/sequestosome 1, p62/SQSTM1; IQ motif-containing GTPase-activating protein 1, IQGAP1; type I phosphatidylinositol 4-phosphate 5-kinase γ/heat shock protein 27, PIPKIγ-HSP27; peptidyl-prolyl cis-trans isomerase NIMA-interacting 1, PIN1; peptidyl-prolyl isomerase A/cyclophilin A, PPIA); a nuclear coactivator module at Neh4/5 (CREB-binding protein/p300, CBP/p300; receptor-associated coactivator 3/steroid receptor coactivator 3, RAC3/SRC-3; protein arginine methyltransferase 1/coactivator-associated arginine methyltransferase 1, PRMT1/CARM1; Mediator complex subunit 16, MED16); and a DNA/chromatin module at Neh1 (small musculoaponeurotic fibrosarcoma [Maf] proteins, BACH1, and chromodomain helicase DNA-binding protein 6, CHD6). Mapping 22 partners onto the Neh-domain architecture identifies approximately 25 pharmacologically addressable interfaces, stratified into four translational tiers. The framework reframes NRF2 pharmacology around one principle: the most actionable target is often a partner rather than NRF2 itself, with disease context dictating the direction of modulation. We close with five testable hypotheses and a partner-code decision matrix linking disease, biomarker, and candidate target.