The activities of integral membrane proteins are often affected by the structures of the lipid molecules that surround them in the membrane. One important parameter is the hydrophobic thickness of the lipid bilayer, defined by the lengths of the lipid fatty acyl chains. Membrane proteins are not rigid entities, and deform to ensure good hydrophobic matching to the surrounding lipid bilayer. The structure of the lipid headgroup region is likely to be important in defining the structures of those parts of a membrane protein that are located in the lipid headgroup region. A number of examples are given where the conformation of the headgroup-embedded region of a membrane protein changes during the reaction cycle of the protein; activities of such proteins might be expected to be particularly sensitive to lipid headgroup structure. Differences in hydrogen bonding potential and hydration between the headgroups of phosphatidycholines and phosphatidylethanolamines could be important factors in determining the effects of these lipids on protein activities, as well as any effects related to the tendency of the phosphatidylethanolamines to form a curved, hexagonal H(II) phase. Effects of lipid structure on protein aggregation and helix-helix interactions are also discussed, as well as the effects of charged lipids on ion concentrations close to the surface of the bilayer. Interpretations of lipid effects in terms of changes in protein volume, lipid free volume, and curvature frustration are also described. Finally, the role of non-annular, or 'co-factor' lipids, tightly bound to membrane proteins, is described.

No abstract

In addition to the annular or boundary lipids that surround the transmembrane surface of the potassium channel KcsA from Streptomyces lividans, x-ray crystallographic studies have detected one anionic lipid molecule bound at each protein-protein interface in the homotetrameric structure, at sites referred to as nonannular sites. The binding constant for phosphatidylglycerol at the nonannular sites has been determined using fluorescence quenching methods with a mutant of KcsA lacking the normal three lipid-exposed Trp residues. Binding is weak, with a binding constant of 0.42 +/- 0.06 in units of mol fraction, implying that the nonannular sites will only be approximately 70% occupied in bilayers of 100% phosphatidylglycerol. However, the nonannular sites show high selectivity for anionic lipids over zwitterionic lipids, and it is suggested that a change in packing at the protein-protein interface leads to a closing of the nonannular binding site in the unbound state. Increasing the anionic lipid content of the membrane leads to a large increase in open channel probability, from approximately 2.5% in the presence of 25 mol % phosphatidylglycerol to approximately 62% in 100 mol % phosphatidylglycerol. The relationship between open channel probability and phosphatidylglycerol content shows cooperativity. The data are consistent with a model in which three or four of the four nonannular sites in the KcsA homotetramer have to be occupied by anionic lipid for the channel to open. The conductance of the open channel increases with increasing concentration of anionic lipid, an effect possibly due to effects of anionic lipid on the concentration of K(+) close to the membrane surface.

Trp fluorescence spectroscopy is a powerful tool to study the structures of membrane proteins and their interactions with the surrounding lipid bilayer. Many membrane proteins contain more than one Trp residue, making analysis of the fluorescence data more complex. The mechanosensitive channels MscL's of Mycobacterium tuberculosis (TbMscL) and Escherichia coli (EcMscL) contain no Trp residues. We have therefore introduced single Trp residues into the transmembrane regions of TbMscL and EcMscL to give the Trp-containing mutants F80W-TbMscL and F93W-EcMscL, respectively, which we show are highly suitable for measurements of lipid binding constants. In vivo cell viability assays in E. coli show that introduction of the Trp residues does not block function of the channels. The Trp-containing mutants have been reconstituted into lipid bilayers by mixing in cholate followed by dilution to re-form membranes. Cross-linking experiments suggest that the proteins retain their pentameric structures in phosphatidylcholines with chain lengths between C14 and C24, phosphatidylserines, and phosphatidic acid. Quenching of Trp fluorescence by brominated phospholipids suggests that the Trp residue in F80W-TbMscL is more exposed to the lipid bilayer than the Trp residue in F93W-EcMscL. Binding constants for phosphatidylcholines change with changing fatty acyl chain length, the strongest interaction for both TbMscL and EcMscL being observed with a chain of length C16, corresponding to a bilayer of hydrophobic thickness ca. 24 A, compared to a hydrophobic thickness for TbMscL of about 26 A estimated from the crystal structure. Lipid binding constants change by only a factor of 1.5 in the chain length range from C12 to C24, much less than expected from theories of hydrophobic mismatch in which the protein is treated as a rigid body. It is concluded that MscL distorts to match changes in bilayer thickness. The binding constants for dioleoylphosphatidylethanolamine for both TbMscL and EcMscL relative to those for dioleoylphosphatidylcholine are close to 1. Quenching experiments suggest a single class of binding sites for phosphatidylserine, phosphatidylglycerol, and cardiolipin on TbMscL; binding constants are greater than those for phosphatidylcholine and decrease with increasing ionic strength, suggesting that charge interactions are important in binding these anionic phospholipids. Quenching experiments suggest two classes of lipid binding sites on TbMscL for phosphatidic acid, binding of phosphatidic acid being much less dependent on ionic strength than binding of phosphatidylserine.

The potassium channel KcsA from Streptomyces lividans has been reconstituted into bilayers of phosphatidylcholines and fluorescence spectroscopy has been used to characterize the response of KcsA to changes in bilayer thickness. The Trp residues in KcsA form two bands, one on each side of the membrane. Trp fluorescence emission spectra and the proportion of the Trp fluorescence intensity quenchable by I(-) hardly vary in the lipid chain length range C10 to C24, suggesting efficient hydrophobic matching between KcsA and the lipid bilayer over this range. Measurements of fluorescence quenching for KcsA reconstituted into mixtures of brominated and nonbrominated phospholipids have been analyzed to give binding constants of lipids for KcsA, relative to that for dioleoylphosphatidylcholine (di(C18:1)PC). Relative lipid binding constants increase by only a factor of three with increasing chain length from C10 to C22 with a decrease from C22 to C24. Strongest binding to di(C22:1)PC corresponds to a state in which the side chains of the lipid-exposed Trp residues are likely to be located within the hydrocarbon core of the lipid bilayer. It is suggested that matching of KcsA to thinner bilayers than di(C24:1)PC is achieved by tilting of the transmembrane alpha-helices in KcsA. Measurements of fluorescence quenching of KcsA in bilayers of brominated phospholipids as a function of phospholipid chain length suggest that in the chain length range C14 to C18 the Trp residues move further away from the center of the lipid bilayer with increasing chain length, which can be partly explained by a decrease in helix tilt angle with increasing bilayer thickness. In the chain length range C18 to C24 it is suggested that the Trp residues become more buried within the hydrocarbon core of the bilayer.