Soluble Membrane-Protein Libraries in Native Nanodiscs
Sandro Keller – Timo Mühlhaus – Frederik Sommer/Michael Schroda – Johannes Herrmann – Ekkehard Neuhaus
Integral membrane proteins are notoriously difficult to study in vitro, as they need to be solubilised in a membrane-mimetic environment that retains their native structures and functions. Native nanodiscs are currently receiving great attention because they enable the direct extraction of proteins from cellular membranes while preserving their local lipid-bilayer environment at all times.1 These nanodiscs spontaneously form when membranes are treated with certain amphiphilic polymers such as styrene/maleic acid (SMA) copolymers1–3 or diisobutylene/maleic acid (DIBMA) copolymer.4 Although exciting results have been achieved with a variety of membrane proteins embedded in such nanodiscs, it remains largely unclear to what extent the protein and lipid libraries obtained by polymer solubilisation reflect the composition and dynamics of the parent membrane and, thus, if they can serve as soluble in vitro models of the membrane proteome and lipidome.5 Recent Förster resonance energy transfer and small-angle neutron scattering experiments6 have demonstrated fast content exchange among nanodiscs, implying that dynamic interactions can be studied in nanodisc libraries even though the individual components are dispersed into nanosized particles, where they are amenable to optical-spectroscopic and chromatographic methods.
In this project, we will develop and exploit proteomics and lipidomics approaches to provide quantitative insights into the solubilising capacity of nanodisc-forming polymers and into the resulting soluble membrane-protein libraries.5 In particular, we aim at:
- Defining the subproteome and lipidome in native nanodiscs for membranes from various host organisms and different organelles. To this end, we will compare the polymer-solubilised subproteome and lipidome of E. coli—the workhorse for recombinant membrane-protein production—with those of eukaryotic origin such as S. cerevisiae mitochondria and C. reinhardtii thylakoids, whose membranes are more difficult to solubilise because of their high protein content. Knowledge of the degree of bias in the solubilised protein and lipid populations as compared with the respective parent membranes is essential for all subsequent work packages.
- Comparing SMA and other nanodisc-forming polymers with improved properties with each other and against conventional detergents. As it is unlikely that there exists a “holy grail” for membrane-protein solubilisation, it is imperative to characterise each solubilising agent with respect to solubilisation yield, protein and lipid preferences, etc. For instance, DIBMA differs from the current gold-standard SMA polymers in terms of protein-solubilisation preference but performs better during chromatographic protein purification and downstream in vitro analyses.4
- Quantifying the ”interactome” in nanodisc-based membrane-protein libraries with respect to the dynamics of interparticle transfer of encapsulated proteins and lipids. Native-like protein–protein and protein–lipid interactions are retained in simple models (i.e., polymer-solubilised two-component protein/lipid mixtures), but this favourable property remains to be demonstrated in complex, unpurified nanodisc libraries that contain large parts of the membrane proteome and lipidome. This goal will be accomplished by testing the functional activity of polymer-solubilised membrane-protein complexes.
1 Knowles et al. J. Am. Chem. Soc. 2009, 131, 7484
2 Vargas et al. Nanoscale 2015, 7, 20685
3 Cuevas Arenas et al. Nanoscale 2016, 8, 15016
4 Oluwole et al. Angew. Chem. Int. Ed. 2017, 56, 1919
5 Marty et al. Anal. Bioanal. Chem. 2013, 405, 4009
6 Cuevas Arenas et al. Sci. Rep. 2017, in press