Interleaflet and Substrate Coupling in Phospholipid Bilayers

Since the existence of lateral organisation in the cell membrane was first proposed by Erwin London in 1997, much has been discovered about the complex behaviour of lipid bilayers. Whilst some membrane proteins involved in signalling are almost as mobile as lipid molecules, such as the photoreceptor protein rhodopsin, others such as the peripheral glycoprotein fibronectin are virtually static. This has been linked to the existence of phase separated micro-domains, sometimes known as lipid rafts, in model systems. However, there are still many open questions, including the effect of asymmetry and curvature on bilayers. Domains in the two leaflets of a model bilayer always align, or register. Conversely, the plasma membrane is asymmetric in composition, which implies that different phases can exist across the bilayer midplane, known as anti-registration. Hydrophobic mismatch at phase boundaries should favour a fully anti-registered bilayer in model systems, implying an interleaflet coupling force drives registration. In this thesis, hydrophobic mismatch between phases is controlled, with anti-registered domains forming at a mismatch of 8 carbons per leaflet. A coupling free energy of 0.021 kBT/nm2 was determined, in close agreement with the only other experimental study using a different methodology, and refining the values found via simulation. Methods are explored to induce anti-registration with lower mismatch, and to characterise the orientation of the anti-registered states. Arising from this work is a greater understanding of how substrate choice for supported bilayers greatly affects phase behaviour. Glass, used in fluorescence microscopy experiments, and PDMS (Polydimethylsiloxane), used to create flexible and curved bilayer substrates, result in nanoscale domain formation compared to micro-scale domains on atomically flat mica. This difference is investigated and it is found that the hydrodynamic motion of domains is hindered by rougher substrates, having great implications for the study and understanding of supported lipid bilayers