We have investigated the formation of supported bilayers by coadsorption of dipalmitoyl phosphatidylcholine (DPPC) with the nonionic surfactant β-D-dodecyl maltoside. The adsorption of mixed phospholipid-surfactant micelles on hydrophilic silica surfaces at 25 °C was followed as a function of bulk concentration by neutron reflection. Using chain-deuterated d25-β-D-dodecyl maltoside and d62-DPPC, we demonstrate that it is possible to determine the composition of the bilayers at each stage of a sequential dilution process, which enriches the adsorbed layer in phospholipid and leads to complete elimination of the surfactant. The final supported bilayers have thicknesses of 51 ( 3 Å and are stable to heating to 37 °C once all surfactant has been removed, and the structures agree well with other published data on DPPC supported bilayers. The coadsorption of cholesterol in a DPPC-surfactant mixture was also achieved, and the location and volume fraction of cholesterol in the DPPC bilayer was determined. Cholesterol is located in a 18 ( 1 Å thick layer below the lipid headgroup region and leads to an increased bilayer thickness of 58 ( 2 Å at 26 mol % of cholesterol.

We have investigated the formation of supported model membranes via the adsorption of phospholipid-surfactant mixtures at the Si-water interface by specular neutron reflection. The adsorption of mixed micelles of the nonionic surfactant beta-D-dodecyl maltoside and DOPC or POPC was determined as a function of bulk concentration, and using d25-beta-D-dodecyl maltoside, the composition of DOPC and POPC bilayers was determined. Bilayer thicknesses of 39+/-3 A for DOPC and 41+/-3 A for POPC agree well with data from bulk lamellar phases for both lipids, and the average area per lipid molecule can be varied from 62 to 115 A2 by varying the bulk concentrations used. The amount of surfactant in the bilayer is very sensitive to the bulk volume-to-surface area ratio, but it can be fully eliminated by ensuring a sufficiently large dilution/rinsing volume of the solution.

We have investigated the phospholipase A(2) catalyzed hydrolysis of supported phospholipid bilayers using neutron reflection and ellipsometry. At the hydrophilic silica-water interface, hydrolysis of phosphatidylcholine bilayers by phospholipase A(2) from Naja mossambica mossambica venom is accompanied by destruction of the bilayer at an initial rate, which is comparable for DOPC and DPPC but is doubled for POPC. The extent of bilayer destruction at 25 degrees C decreases from DOPC to POPC and is dramatically reduced for DPPC. Neutron reflectivity measurements indicate that the enzyme penetrates into the bilayers in increasing order for DOPC, POPC, and DPPC, while the amount of enzyme adsorbed at the interface is smallest for DPPC and exhibits a maximum for POPC. Penetration into the hydrophobic chain region in the bilayer is further supported by the fact that the enzyme adsorbs strongly and irreversibly to a hydrophobic monolayer of octadecyltrichlorosilane. These results are rationalized in terms of the properties of the reaction products and the effect of their accumulation in the membrane on the kinetics of enzyme catalysis.

The bending elasticity of the monoglyceride 1-monoolein (MO) in water has been measured in the inverse hexagonal (HII) phase at 37 °C. At this temperature, fully hydrated MO is normally in an inverse bicontinuous cubic phase based on Schwarz's D surface, . The addition of either of the C23 chainlength hydrocarbons, tricosane or 9-cis-tricosene, at mole fractions with respect to MO in excess of 0.05, induces a phase transition into the HII phase. This transition is understood to occur because packing stresses in the hydrophobic regions of the HII phase are reduced to levels where this phase is at a lower free energy than the phase. We have used X-ray diffraction on gravimetrically prepared samples and samples subjected to an osmotic stress to determine the bending energy of MO in this phase. The evidence suggests that 9-cis-tricosene can relieve almost all of the packing stress in the HII phase. In this case, we find that the spontaneous radius of curvature at the pivotal surface, R 0, is −20.0 ± 0.3 Å, and the monolayer bending rigidity, κ, is (1.2 ± 0.1) × 10-20 J. Comparing these energetics with those of the phase indicates that the packing stress for an excess water, HII phase at 37 °C in the absence of 9-cis-tricosene would constitute at least 50% of the total equilibrium free energy. With tricosane, the packing stress cannot be completely relieved because tricosane melts at 47.6 °C in the bulk. The results from geometric measurements on this system suggest that when the packing stress is not fully relieved in the HII phase, the polar/apolar interface is deformed away from being cylindrical. Treating the interface as if it were cylindrical leads to a nonphysical location of the pivotal surface that is in disagreement with all previous measurements, a 35% increase in the magnitude of R 0, and a 4-fold increase in the calculated bending rigidity.