Soutenance de thèse : Structural study of NADPH oxidase complex NOX1 by cryo-electron microscopy

Par Maria Val Pevida (IBS/Groupe Membrane et pathogènes & Groupe Microscopie Electronique et Méthodes)

Membrane proteins (MPs) constitute a major class of pharmaceutical targets due to their central roles in essential physiological processes. Among them, NADPH oxidases (NOX) comprise a family of transmembrane enzymes involved in numerous crucial cellular functions — including host defence, hormone synthesis, vascular tone regulation, and equilibrium — through the regulated production of reactive oxygen species (ROS). NOX1, one of the human NOX isoforms, is predominantly expressed in the intestinal epithelium, where it forms a heterodimeric complex with p22^phox in the membrane. Dysregulation of NOX1 activity, often arising from genetic mutations, has been associated with early-onset colorectal cancer and inflammatory bowel disease (IBD). Elucidating the three-dimensional structures of MPs and their mechanistic properties is essential to accelerate structure-based drug design strategies. Cryo-electron microscopy (cryo-EM) has become a powerful technique for resolving MPs structures at atomic resolutions. However, structural determination of small MPs (<100 kDa) remains challenging due to their low expression levels, low signal-to-noise ratios, dependence on detergent solubilisation, and intrinsic conformational flexibility, all which complicating particle alignment and high-resolution structure reconstruction. To overcome these limitations, antigen-binding fragments (Fabs) are frequently employed as fiducials to increase particle size and stabilise specific conformations by binding to defined epitopes.

In this thesis, a lentiviral-based expression system was developed for the overexpression of the NOX1/p22phox complex in HEK293 T-Rex cells. Initial attempts to determine the structure of wild-type NOX1 were unsuccessful due to low production yield and the small size of the complex, compounded by the lack of a commercially available monoclonal antibody targeting NOX1. To overcome these limitations, co-expression with the EROS chaperone, combined with the engineering of a NOX1 mutant, named NOX12, enabled improved production. This mutant incorporates modifications in extracellular loops to allow specific recognition by the anti-NOX2 Fab7D5. This strategy increased particle size, allowing particle alignment and facilitating structural determination of the complex by cryo-EM. High-resolution structure of the core of NOX12 in complex with p22^phox and Fab7D5 was determined at an overall resolution of 2.8 Å. In addition, the structure of the NOX12/p22^phox in complex with EROS and Fab7D5 was resolved at an overall resolution of 3.2 Å. This structure revealed a previously uncharacterised interaction mode between EROS and the catalytic subunit. Furthermore, the structural framework enabled analysis of disease-associated mutations in NOX1 and provided insights into oxygen access pathways.

This work provides the first structural insights into NOX1 isoform in complex with its partner p22^phox and together with the EROS chaperone, establishing a framework for understanding its stabilisation and maturation. In addition, the successful implementation of protein epitope engineering highlights a broadly applicable strategy for studying other NOX isoforms that lack specific antibodies, in both resting and activated states, thereby enabling cryo-EM characterisation across this family of MPs. Further studies will be required to elucidate the detailed molecular mechanisms underlying NOX1 activation.