AmericanChemicalSociety.com

We have developed methods to study interfacial diffusion of small molecule surfactants and proteins at the liquid/liquid interface. These methods included the use of microwell compartments to contain the oil phase in the presence of a contacting aqueous phase. Surfactants and/or proteins were introduced to the oil-water interface in two ways: (1) simple adsorption from the aqueous phase, or (2) Langmuir-Blodgett deposition of layers previously assembled at the air/water interface. The first method permitted the observation of dynamic adsorption events, while the second method provided a quantitative measure of the surface density of adsorbate molecules. For samples prepared using the first method, we used total internal reflection fluorescence microscopy in order to reduce the background level due to the fluorescence of molecules remaining in aqueous solution. For samples prepared by LB deposition, epifluorescence was frequently adequate. For samples prepared using relatively high concentrations of fluorescent probe molecules, fluorescence recovery after photobleaching (FRAP) was used to measure the diffusion rates of phospholipids at the oil/water interface for oil viscosities that varied over more than three orders of magnitude. Surprisingly, we found that the effective interfacial diffusion coefficient did not depend inversely on oil viscosity, but instead decreased as viscosity to the -0.6 power. We speculate that this anomalous dependence on oil viscosity may be due to the presence of a diffusive mechanism other than simple motion within the two-dimensional interfacial layer. For example, adsorbed molecules may undergo brief excursions through the less viscous aqueous phase prior to re-adsorption. We plan to investigate the mechanisms in detail using single molecule trajectories.

We have also used fluorescence microscopy to understand the mobility of proteins at the silicone oil / water interface. Preliminary measurements, have demonstrated the ability to perform FRAP and single-molecule tracking experiments of protein mobility at the interface between an aqueous phase and viscous silicone oil. For example, analysis of FRAP data on FITC-labeled bovine serum albumin (BSA) at the interface between water and a 12,000 cP silicone oil found an effective interfacial diffusion coefficient of 0.02 µm2/s. This is significantly higher than the calculated lower limit for the diffusion coefficient of 0.005 µm2/s, based on the Stokes-Einstein equation assuming a protein molecule immersed in the viscous oil phase. This is an intriguing result that suggests the possibility of complex diffusion mechanisms at the oil/water interface.

We have also performed preliminary experiments demonstrating the ability to follow individual protein molecules at the oil/water interface. The preliminary data suggest that individual BSA molecules exhibit at least two diffusive modes at the oil/water interface. Individual trajectories often exhibited behavior where a given molecule initially executed slow diffusive motion with a diffusion coefficient of 0.005 µm2/s (corresponding to the Stokes-Einstein prediction), and then abruptly shifting to faster diffusion, with a coefficient of 0.025 µm2/s (sometimes the switch was in the opposite direction). This behavior suggests the presence of multiple diffusive modes that may be due to different molecular conformations, or simply different penetration depths into the oil phase. Presumably, the FRAP data represent a weighted average of these modes. These observations illustrate the ability of the single-molecule approach to extract molecular level mechanisms, while FRAP data provide ensemble-average information.