Lipid mixtures are inherently nonrandom as each lipid species differs slightly in its chemical structure. A protein associates not with a lipid but with a membrane comprised of lipids where the chemical activities of each lipid is determined by the composition of the mixture. There can be selectivity in this association because a protein can enhance the underlying tendency of lipids to be heterogeneously distributed. This is dependent on the protein having a preferential association of sufficient magnitude with some of the lipids within the membrane. To measure and model protein-lipid interactions, an understanding of the underlying lipid behavior is necessary to interpret their association constants. Methods to measure protein-lipid interactions are discussed within the context of using these techniques in modeling and a general framework is presented for the use of a signal arising from these interactions. The use of binding partition functions is presented as this allows the modeling of cooperative or independent (noncooperative) interactions of protein with lipids and of proteins with additional ligands as well as lipids. A model is also provided using the binding partition function formalism where protein dimerization, and by extension, oligomerization is enhanced at the membrane compared to in solution.
|Original language||English (US)|
|Number of pages||23|
|Journal||Methods in Enzymology|
|State||Published - 2009|