Mathematical Modeling Reveals Quantitative Properties of KEAP1-NRF2 Signaling

In response to oxidative and electrophilic stresses, cells launch an NRF2-mediated transcriptional antioxidant program. The activation of NRF2 depends on a redox sensor, KEAP1, which acts as an E3-ligase adaptor to promote the ubiquitination and degradation of NRF2. While a great deal has been learned about the molecular details of KEAP1, NRF2, and their interactions, the quantitative aspects of signal transfer conveyed by this redox duo are still largely unexplored. In the present study, we examined the signaling properties including response time, half-life, maximal activation, and response steepness (ultrasensitivity) of NRF2, through a suite of mathematical models. The models describe, with increasing complexity, the reversible binding of KEAP1 dimer and NRF2 via the ETGE and DLG motifs, NRF2 production, KEAP1-dependent and independent NRF2 degradation, and perturbations by different classes of NRF2 activators. Our simulations revealed that at the basal condition, NRF2 molecules are largely sequestered by KEAP1, with the KEAP1-NRF2 complex comparably distributed in either an ETGE-bound only (open) state or an ETGE and DLG dual-bound (closed) state, corresponding to the unlatched and latched configurations of the conceptual hinge-latch model. With two-step ETGE binding, the open and closed states operate in cycle mode at the basal condition and transition to equilibrium mode at stressed conditions. Class I-V, electrophilic NRF2 activators, which modify redox-sensing cysteine residues of KEAP1, shift the balance to a closed state that is unable to degrade NRF2 effectively. Total NRF2 has to accumulate to a level that nearly saturates existing KEAP1 to make sufficient free NRF2, therefore introducing a signaling delay. At the juncture of KEAP1 saturation, ultrasensitive NRF2 activation, i.e., a steep rise in the free NRF2 level, can occur through two simultaneous mechanisms, zero-order degradation mediated by DLG binding and protein sequestration (molecular titration) mediated by ETGE binding. These response characteristics of class I-V activators do not require disruption of DLG binding to unlatch the KEAP1-NRF2 complex. In comparison, class VI NRF2 activators, which directly compete with NRF2 for KEAP1 binding, can unlatch or even unhinge the KEAP1-NRF2 complex. This causes a shift to the open state of KEAP1-NRF2 complex and ultimately its complete dissociation, resulting in a fast release of free NRF2 followed by stabilization. Although class VI activators may induce free NRF2 to higher levels, ultrasensitivity is lost due to lower free KEAP1 and thus its NRF2-sequestering effect. Stress-induced NRF2 nuclear accumulation is enhanced when basal nuclear NRF2 turnover constitutes a small load to NRF2 production. Our simulation further demonstrated that optimal abundances of cytosolic and nuclear KEAP1 exist to maximize ultrasensitivity. In summary, by simulating the dual role of KEAP1 in repressing NRF2, i.e., sequestration and promoting degradation, our mathematical modeling provides key novel quantitative insights into the signaling properties of the crucial KEAP1-NRF2 module of the cellular antioxidant response pathway.