Anne Grethen

• To render membrane proteins amenable to in vitro functional and structural studies, they need to be extracted from cellular membranes and stabilised using membrane-mimetic systems. Amphiphilic copolymers gain considerable interest, because they are able to coextract membrane proteins and their surrounding lipids from complex cellular membranes to form polymer-bounded nanodiscs. The latter harbour a native-like lipid-bilayer core stabilised by a copolymer rim. Accordingly, these membrane mimics are supposed to provide superior stability to embedded membrane proteins as compared with conventional detergent micelles. Herein, the formation of nanodiscs by the most commonly used styrene/maleic acid (SMA)copolymer, termed SMA(2:1), was elucidated in detail. To this end, the equilibrium solubilisation efficiencies towards model and cellular membranes were quantified and compared with those of the more hydrophobic SMA(3:1) and the more hydrophilic diisobutylene/maleic acid (DIBMA) copolymers. It was shown that, from a thermodynamic viewpoint, SMA(2:1) is the most efficient membrane solubiliser in terms of lipid- and proteinextraction yields. Solvent properties (pH, ionic strength) or membrane characteristics (lateral pressure, charge, or thickness) can affect the polymers’ solubilisation efficiency to a certain extent. In addition, the lipid transfer behaviour of SMA(2:1) nanodiscs was studied. Notwithstanding their high effective negative charge, SMA(2:1) nanodiscs exchange phospholipids more rapidly among each other than vesicles or protein-bounded nanodiscs, thus rendering them highly dynamic nano-assemblies. Two alternative electroneutral polymers, namely SMA(2:1)-SB and DIBMA-SB, were introduced in this thesis. They were generated by polymer backbone modifications of SMA(2:1) and DIBMA, respectively. The derivatised polymers were shown to quantitatively solubilise model and biological membranes and, like DIBMA, only had a mild effect on lipidbilayer integrity. Along these lines, DIBMA-SB preserved membrane-protein complexes of distinct structural classes and extracted them from various cellular membranes. Importantly, the electroneutral polymers were amenable to protein/lipid interaction studies otherwise masked by unspecific interactions of their anionic counterparts with target lipids or proteins. Taken together, the in-depth characterisation of nanodiscs formed by anionic and electroneutral polymers allows for adjusting the nanodisc properties to specifically suit experimental requirements or address membrane-protein research questions.