Superoxide Production by Transmembrane Electron Transfer – SuperET

1) Use of alphafold software to build models of the NOX2 protein alone or in combination with some of its cytosolic partners

2) Classical Molecular Dynamics simulations to study the structure of NOX proteins inserted in lipid membranes

3) QM+MM approach to compute thermodynamic parameters entering the Marcus theory for electron transfer

4) Production of various forms of NOX2 proteins using molecular biology methods

5) Time-resolved spectrokinetics experiments of electron transfer in NOX2, initiated by pulse radiolysis or photoactivation of a ruthenium complex

Several models of NOX2 proteins, alone or in combination with some of its cytosolic partners have been built with the use of Alphafold software.

The electron transfer chain from the flavin cofactor towards the final heme has been characterized by molecular dynamics simulations in two proteins of the NADPH oxidases family (human NOX5 and spNOX). This has led to the first estimation of the rate of electron transfer between redox cofactors in those systems. On a more general perspective, this work has revealed large compensatory effects between the various constituents of the system, which might be at the origin of the fine regulation of the thermodynamics of electron transfer in large transmembrane complexes. Finally, we have obtained promising results for the calibration of the various redox states of flavins in the AMOEBA forcefield, opening the path towards an improvement in the prediction of redox potential values from molecular simulations.

On the experimental side, several forms of NOX2 proteins have been produced with the use of molecular biology methods. Among them, a chimeric protein has been synthesized that combines NOX2 with some of its cytosolic protein partners. It has been shown that in this last chimeric system it is possible to initiate electron transfer with light in the presence of free ruthenium complexes.

The SuperET project offers many perspectives, both theoretical and experimental. Among them, the most obvious are the following:

- Improvement of the NOX2 models thanks to the use of integrative approaches combining molecular modeling and experimental data to constrain the models.

- Cartography of electron transfer in other members of the NADPH oxidases family, in order to highlight the common characteristics and also the differences between various homologs.

- Consolidation of the production of NOX2 systems, and in particular the chimeric NOX2 protein in order to graft a ruthenium complex close to the flavin cofactor. This would allow us to experimentally measure the kinetics of intramolecular electron transfer in NOX2.

NADPH oxidases are unique enzymes producing superoxide to destroy pathogens. Flavocytochrome Cyt b558 is the core protein which is inserted in the membrane of phagosomes and catalyses the last step of superoxide production. The latter consists in the transfer of two electrons from a reduced flavin encompassed in the reductase domain across the membrane domain to reduce two dioxygen molecules. Despite decades of research, the precise molecular mechanism permitting to couple transmembrane electron transfer (ET) to superoxide production has remained elusive. Kinetic and thermodynamic parameters in particular are still lacking. In 2017, the first crystallographic structure of the NOX5 variant has revealed the relative positions of the various redox moieties: the flavin and the two heme molecules inserted in the transmembrane domain. Putative binding pockets have also been reported for the dioxygen molecules.
Our strategy is to use a combined theoretical and experimental approach to provide estimates/measurements of the kinetic rates of electron transfer among the redox cofactors and to decipher the molecular mechanisms of the ET processes using state-of-the-art molecular simulations. Throughout the project, the mechanistical hypotheses emerging from molecular modeling will be tested with the study of designed mutants both experimentally and with molecular simulations, in order to cross-validate both approaches.
We aspire to perform the first time-resolved measurements of ET within the NOX2 variant. Controlled triggering of the reactions is the key prerequisite to obtain such information and two approaches will be followed in parallel. We will use pulsed water radiolysis in reducing conditions to produce carboxyl anion radicals COO•– that will reduce the flavin to initiate the ETs. An advantage of this method, that will be performed on the ELYSE setup at the ICP, is that it can be used directly on wild-type NOX2 protein systems. On another setup at the I2BC, we will study photoactivatable NOX2 systems using modified NOX2 proteins grafted with Ruthenium complexes in close proximity of the redox cofactors that should reduce the flavin upon excitation by UV-visible light.
As for the theoretical and computational part, we will perform molecular dynamics simulations of NOX2 and NOX5 inserted into a membrane to study its structure and dynamics and to monitor the diffusion of dioxygen and superoxide molecules. Thermodynamic and kinetic parameters of ET steps will be obtained through well-established QM+MM calculations in the framework of the Linear Response Approximation. In our simulations, we will pay specific attention to the treatment of electrostatic interactions that play a major role in ET processes. We will use the multipolar and polarizable forcefield AMOEBA, to which we will add parameters to describe flavin in its various redox states.
Our collaborative consortium brings together experts in atomistic simulations of membrane-bound proteins and of biological ET simulations, biochemists specialized in NOX proteins, radiation chemists, photochemists and spectroscopists with an established experience in elucidation of light-induced electron and proton transfer processes in both natural and modified biological systems. We thereby possess all the skills and know-how required to realize a breakthrough in the understanding of transmembrane ET in NOX.