Docking into the KEAP1 Kelch domain for NRF2 activation

The KEAP1–NRF2 interaction is a master regulator of cellular antioxidant and detoxification pathways. Under basal conditions, KEAP1 (Kelch-like ECH-associated protein 1) binds NRF2 (nuclear factor erythroid 2-related factor 2) and targets it for ubiquitin-mediated degradation. Disrupting this protein-protein interaction with a small-molecule KEAP1 inhibitor stabilizes NRF2, allowing it to translocate to the nucleus and activate hundreds of cytoprotective genes. This mechanism is pursued for chronic kidney disease, neurodegenerative disorders, and inflammatory conditions—but also as a liability in cancer, where constitutive NRF2 activation can confer chemo-resistance.

The Kelch domain binding pocket

KEAP1 is a homodimer; each monomer contains an N-terminal BTB domain responsible for Cul3 E3 ligase recruitment, a central IVR (intervening region), and a C-terminal Kelch domain comprising six kelch repeats that form a six-bladed beta-propeller. NRF2 binds KEAP1 through two motifs: a high-affinity ETGE motif (Kd ~10 nM) and a lower-affinity DLG motif (Kd ~1 μM). Both motifs dock into the central pocket on the bottom face of the Kelch propeller, where they engage a network of arginine and other polar residues lining the pocket walls.

Crystal structures (PDB: 2FLU, 1X2R, and others) reveal that the ETGE motif adopts an extended conformation, with the glutamate side chains (E79, E82 in the human NRF2 Neh2 domain) forming bidentate salt bridges with KEAP1 residues Arg415, Arg483, and Asn414. The threonine and glycine residues fit into narrower sub-pockets. Small-molecule KEAP1 inhibitors mimic this acidic pharmacophore: they typically present two carboxylates or other anionic groups spaced ~7–9 Å apart to engage the same arginine anchors, with a hydrophobic or polar linker occupying the central channel.

Docking strategy for KEAP1 inhibitors

Docking into the Kelch domain is technically a protein-protein interaction (PPI) site—wide, shallow, and predominantly polar—rather than a classic druggable orthosteric pocket. Success requires careful pocket definition and scoring-function tuning:

• Pocket centering: Define the docking box around the ETGE binding site. If working from apo KEAP1 (no peptide bound), use Arg415 and Arg483 as geometric references; these residues are invariant across structures. A 20 × 20 × 12 Å box centered between the two arginines typically captures the relevant space.
• Charge states: All carboxylate-bearing ligands must be docked in their ionized (deprotonated) form at physiological pH. The arginine side chains (Arg415, Arg483, Ser508, Asn414) form the electrostatic floor of the pocket; neutral acids will not score correctly.
• Water-mediated contacts: Crystal structures often show ordered water molecules bridging ligand carboxylates and backbone carbonyls (e.g., Ser363, Gly509). Retaining these crystallographic waters or enabling water-toggle features in your docking protocol can improve pose accuracy.
• Scoring recalibration: Standard scoring functions trained on kinase or GPCR pockets may under-reward salt bridges. If your platform supports custom weighting, increase the electrostatic term; alternatively, post-process poses with a protein-ligand interaction fingerprint (PLIF) filter that requires engagement of at least one arginine anchor.

Liganx Studio allows you to upload a KEAP1 Kelch domain structure (e.g., PDB 4IQK or 2FLU), define the pocket interactively, and dock libraries of carboxylate-containing fragments or clinical candidates. Because the platform runs entirely in your browser, you retain full control of proprietary ligand structures and can iterate rapidly on docking-box placement and charge-state assignments.

Clinical and preclinical KEAP1 inhibitors

The first-generation KEAP1 inhibitor to reach clinical testing was bardoxolone methyl, a semisynthetic triterpenoid. It showed efficacy in phase 2 trials for chronic kidney disease but was halted in phase 3 due to cardiovascular adverse events. Bardoxolone is a covalent modifier of KEAP1 Cys151 (in the IVR) rather than a direct Kelch-domain binder; its mechanism is more complex than simple PPI disruption.

True competitive Kelch-domain inhibitors emerged later. Representative scaffolds include naphthalene-dicarboxylates, tetrahydroisoquinoline cores, and bi-aryl acids. Compound 16 from a 2014 Astex/Evotec series (J. Med. Chem. 2014, 57, 2210) is a well-characterized example: a naphthalene-2,7-dicarboxylic acid derivative with submicromolar biochemical potency. Co-crystal structures confirm the expected bidentate arginine coordination.

More recent disclosures include tetrahydroisoquinoline-based inhibitors optimized for oral bioavailability and CNS penetration, motivated by Parkinson's and Alzheimer's disease models. These molecules typically retain the dual-carboxylate motif but add lipophilic extensions to improve cell permeability—a challenge for highly polar PPI disruptors.

Mutation context: KEAP1 loss-of-function in cancer

Somatic KEAP1 mutations are frequent in non-small cell lung cancer (NSCLC), especially adenocarcinomas with KRAS co-mutations. The majority cluster in the Kelch domain or IVR and are loss-of-function, leading to constitutive NRF2 stabilization. In this context, KEAP1 inhibitors are contraindicated—they would further activate an already hyperactive NRF2 pathway. Conversely, strategies to degrade NRF2 (PROTACs, molecular glue degraders) are under investigation for KEAP1-mutant tumors.

When docking into mutant KEAP1 structures, pay particular attention to residues within 5 Å of the peptide-binding groove. Mutations such as G186C, G333C, R470C, and G423S can directly alter pocket geometry or disrupt arginine salt bridges. If you are modeling resistance or compensatory binding in a KEAP1 mutant background, our mutation docking guide outlines best practices for introducing point substitutions, local minimization, and ensemble docking across rotamer states.

Practical tips for Kelch-domain docking

• Start with high-resolution co-crystal structures (2.0 Å or better) to ensure arginine side-chain conformations are well-defined. PDB entries 4IQK, 3VNG, and 5FNQ are good references.
• If your ligand library contains mixtures of mono- and di-acids, pre-filter or flag di-acids—they dramatically outperform mono-acids in this pocket.
• Check for beta-propeller distortions in apo structures. The Kelch domain can undergo subtle breathing motions; peptide-bound structures are generally more reliable for docking.
• Use RMSD clustering or interaction-fingerprint clustering to collapse pose ensembles. Many docking runs will produce multiple poses with identical arginine contacts but different linker conformations; these are functionally equivalent.
• Post-docking, inspect the electrostatic complementarity visually. A successful pose should show the ligand carboxylates penetrating into the arginine-rich crater, not riding on the surface.

Beyond direct inhibitors: allosteric and covalent alternatives

While competitive Kelch-domain inhibitors have dominated the literature, covalent KEAP1 inhibitors targeting Cys151 (dimethyl fumarate, bardoxolone) or Cys273/288 offer an alternative mechanism. These compounds modify KEAP1 cysteines, triggering a conformational change that releases NRF2 without directly occupying the Kelch pocket. Docking covalent warheads requires specialized protocols (covalent docking, transition-state modeling) not typically needed for reversible PPI disruptors.

A third emerging strategy is targeted NRF2 degradation via heterobifunctional molecules (PROTACs) in KEAP1-mutant settings, effectively reversing the gain-of-function phenotype. These large molecules (MW > 800 Da) are not docked in the traditional sense but can be modeled by docking the NRF2-binding warhead into the Kelch domain and the E3-ligase ligand separately, then validating ternary-complex geometry.

Conclusion

Docking into the KEAP1 Kelch domain requires attention to the unique geometry of a shallow, arginine-lined PPI site. Dual-carboxylate pharmacophores, correct ionization states, and retention of key water molecules are critical for accurate pose prediction. While clinical development of direct KEAP1 inhibitors has been cautious—balancing therapeutic benefit against the liability of NRF2 hyperactivation in cancer—the target remains attractive for diseases of oxidative stress and inflammation. Mutation-aware docking can clarify resistance mechanisms in KEAP1-mutant tumors and guide next-generation inhibitor design for preserved or altered Kelch-domain pockets.

References

• Davies, T. G., Wixted, W. E., Coyle, J. E., et al. (2016). Monoacidic inhibitors of the Kelch-like ECH-associated protein 1: nuclear factor erythroid 2-related factor 2 (KEAP1:NRF2) protein-protein interaction with high cell potency identified by fragment-based discovery. J. Med. Chem. 59(8), 3991–4006. DOI: 10.1021/acs.jmedchem.6b00228
• Padmanabhan, B., Tong, K. I., Ohta, T., et al. (2006). Structural basis for defects of Keap1 activity provoked by its point mutations in lung cancer. Mol. Cell 21(5), 689–700. DOI: 10.1016/j.molcel.2006.01.013
• Tong, K. I., Katoh, Y., Kusunoki, H., et al. (2006). Keap1 recruits Neh2 through binding to ETGE and DLG motifs: characterization of the two-site molecular recognition model. Mol. Cell. Biol. 26(8), 2887–2900. DOI: 10.1128/MCB.26.8.2887-2900.2006
• Marcotte, D., Zeng, W., Hus, J. C., et al. (2013). Small molecules inhibit the interaction of Nrf2 and the Keap1 Kelch domain through a non-covalent mechanism. Bioorg. Med. Chem. 21(14), 4011–4019. DOI: 10.1016/j.bmc.2013.04.019
• Cuadrado, A., Rojo, A. I., Wells, G., et al. (2019). Therapeutic targeting of the NRF2 and KEAP1 partnership in chronic diseases. Nat. Rev. Drug Discov. 18(4), 295–317. DOI: 10.1038/s41573-018-0008-x