Investigation of conformational changes in Endophilin A1/B1 using Small-Angle X-ray Scattering (SAXS) and Cryo-EM

The process of regulating membrane curvature is an important part of cellular function. An increase in curvature can be used to promote vesicle formation, membrane rupture, and more. One protein group which regulates membrane curvature is the Endophilin family, Bin/Amphiphysin/Rvs (BAR) domain proteins which bind to membranes in order to perform essential tasks in the cell. Endophilin A1 (EnA1) bind to the cell membrane and promote endocytosis, which is essential as a method for drug delivery. Endophilin B1 (EnB1) bind to the outer mitochondrial membrane (OMM) as a part of the intrinsic apoptotic pathway, leading to cell death. A lack of EnB1 expression has been detected in some forms of cancers, suggesting EnB1 prevents tumour formation. Despite both proteins sharing a homologous structure, the proteins bind to membranes of different lipid compositions. My thesis set out to investigate the conformational changes between EnA1 and EnB1 in membrane bound form, compared to soluble form. I wanted to detect what conformations play a part in membrane binding and curvature regulation. Is the conformational change different when the lipid composition is changed? Can EnA1 bind to membranes EnB1 bind to? What effect does the H0 domain of Endophilins have on binding affinity? To accomplish this, I analysed data from Small-Angle X-ray Scattering (SAXS) experiments of EnA1 and EnB1 in soluble form to calculate electron density maps. I also analysed EnA1 bound to nanodiscs, small synthetic membranes held together by scaffold proteins, with Cryo-EM to obtain high-resolution electron density maps. The binding to nanodiscs was compared between ones containing phosphatidylinositol-4,5-bisphosphate (PIP2), and ones containing cardiolipin (CL). The volumes were compared with a solved reconstruction of EnB1 already published from my group. While my results were unable to resolve the nanodisc binding enough to distinguish side chains in the protein, it nevertheless revealed a conformational difference in EnA1, depending on the nanodisc it was bound to. It also showed that EnA1 adopt a “side-to-side” conformation when multiple copies bind to the same nanodisc. When EnA1 bind to a PIP2 nanodisc, it replaces the scaffold proteins which hold the nanodisc together. The volumes calculated from SAXS were of very low resolution but seem to suggest a straighter conformation of EnB1 and EnA1 compared to nanodisc bound form. The results suggest that membrane remodelling does not occur whenEnA1 binds to CL membranes. However, membrane remodelling and destabilization do occur when EnA1 bind to PIP2 membranes. As H0 is the interacting domain which differ the most between the two proteins, it seems to have a noticeable effect in binding affinity.