We have used fluorescence microscopy, fluorescence photobleaching recovery (FPR), and atomic force microscopy (AFM) to investigate the formation of tethered lipid bilayers on plane aluminum oxide or glass surfaces. The bilayers were assembled with the help of a two-step methodology recently proposed for microporous templates (Proux-Delrouyre et al. J. Am. Chem. Soc. 2001, 123, 8313). The first step consists of the accumulation of intact biotinylated vesicles (PC + DOPE) on a streptavidin sublayer itself immobilized on the substrate. The second step, clearly time separated, is the deliberate triggering of bilayer formation with the help of poly(ethylene glycol) (PEG), a fusion agent of lipidic vesicles. AFM and FPR measurements confirm that the vesicles do not spontaneously fuse during the first step provided that the streptavidin sublayer is present on the substrate. On the contrary, the treatment with PEG provokes the fast formation of a continuous lipid bilayer, as attested at the hundred nanometer scale by the AFM images and at the hundred micrometer scale by the lateral diffusion of a fluorescent probe (D ) 2.2 × 10 -8 cm 2 s -1 for NBD-DMPE at 22 °C). † Part of the Langmuir special issue entitled The Biomolecular Interface.

The long-range diffusion coefficients of isoprenoid quinones in a model of lipid bilayer were determined by a method avoiding fluorescent probe labeling of the molecules. The quinone electron carriers were incorporated in supported dimyristoylphosphatidylcholine layers at physiological molar fractions (<3 mol%). The elaborate bilayer template contained a built-in gold electrode at which the redox molecules solubilized in the bilayer were reduced or oxidized. The lateral diffusion coefficient of a natural quinone like UQ10 or PQ9 was 2.0 +/- 0.4 x 10(-8) cm2 s(-1) at 30 degrees C, two to three times smaller than the diffusion coefficient of a lipid analog in the same artificial bilayer. The lateral mobilities of the oxidized or reduced forms could be determined separately and were found to be identical in the 4-13 pH range. For a series of isoprenoid quinones, UQ2 or PQ2 to UQ10, the diffusion coefficient exhibited a marked dependence on the length of the isoprenoid chain. The data fit very well the quantitative behavior predicted by a continuum fluid model in which the isoprenoid chains are taken as rigid particles moving in the less viscous part of the bilayer and rubbing against the more viscous layers of lipid heads. The present study supports the concept of a homogeneous pool of quinone located in the less viscous region of the bilayer.