The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form

The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. hydrocarbon regions. A molecular understanding of the molecular features that control the diffusion speeds of proteins bound to supported bilayers would enable key molecular information to be extracted from experimental diffusion constants, revealing protein-lipid and protein-bilayer interactions difficult to study by other methods. The present study investigates a range of 11 different peripheral protein constructs comprised by 1 to 3 distinct domains (PH, C1A, C1B, C2, anti-lipid antibody). By combining these constructs with various combinations of target lipids, the study measures 2-D diffusion constants on supported bilayers for 17 different protein-lipid complexes. The resulting experimental diffusion ENOblock (AP-III-a4) constants, together with the known membrane interaction parameters of each complex, are used to analyze the molecular features correlated with diffusional slowing and bilayer friction. The findings show that both 1) individual bound lipids and 2) individual protein domains that penetrate into the hydrocarbon core make additive contributions to the friction against the bilayer, thereby defining the 2-D diffusion constant. An empirical formula is developed that accurately estimates the diffusion constant and bilayer friction of a peripheral protein in terms of its number of bound lipids and its geometry of penetration into the bilayer hydrocarbon core, yielding an excellent global best fit (R2 of 0.97) to the experimental diffusion constants. Finally, the observed additivity of the frictional contributions suggests that further development of current theory describing bilayer dynamics may be needed. The present ENOblock (AP-III-a4) findings provide constraints that will be useful in such theory development. = where is the single lipid diffusion constant and is the total number of PH domains (or, equivalently, the total number of bound PIP3 molecules in the construct) (12). This simple relationship suggests that, to a first approximation, (i) the friction of bound lipids against the bilayer dominates due to 100-fold higher viscosity of bilayer, and (ii) the individual bound lipids make additive ACE contributions to the friction (12). Such simplicity is not predicted by current models of bilayer dynamics, suggesting that the theoretical description of the bilayer is not yet complete (12,16). More generally, in contrast to the PH domain motif, many peripheral membrane binding proteins not only bind specific lipids but also penetrate significantly into the bilayer (18C26). For such proteins both the bound lipid and penetrating protein components may generate friction against the bilayer that would slow diffusion. The present study investigates the relationship between the 2-dimensional ENOblock (AP-III-a4) diffusion constant of an arbitrary peripheral protein, its number of bound lipids, and its degree of protein penetration into the bilayer. The analysis compares the 2-dimensional diffusion constants of 17 different protein-lipid complexes undergoing lateral diffusion on supported bilayers, where the bound lipid stoichiometries and protein-membrane docking geometries are known for each complex. The results indicate that the experimental diffusion constant D slows with increasing contacts between the protein-lipid complex and the bilayer. Strikingly, the findings show that the contributions of individual bound lipids and individual bilayer-contacting domains to the total friction are, to a first approximation, additive. To elucidate the parameter that best defines the contribution of protein-bilayer contacts to friction, a set of candidate parameters is developed and the data reveal the most important frictional contacts are those in the bilayer hydrocarbon core. Looking forward, the study provides a set of constraints that will facilitate future development of dynamic bilayer theories capable of describing the 2D diffusion of peripheral membrane proteins on supported bilayers, and ENOblock (AP-III-a4) perhaps bilayers in general. Materials and Methods Reagents Synthetic phospholipids PC (phosphatidylcholine; 1,2-dioleoyl-(Wolfram Research), IGOR Pro (WaveMetrics) and GraphPad Prism 5 (GraphPad Software, Inc.). Single particle tracking As in our previous studies (11,12), diffusion trajectories of single fluorescent lipid and protein molecules were tracked and quantitated using the Particle Tracker plugin for ImageJ (31), then diffusion data was imported into for further analysis. Briefly, the plugin determines the center position and intensity of each fluorescent particle in each frame, then links the particles in successive frames to form trajectories. Only particles possessing fluorescence intensities within a defined range were included in the analysis, thereby eliminating bright protein aggregates and dim, nonprotein contaminants. Additional exclusions removed immobile particles, rapidly dissociating particles, and overlapping tracks for which particle identity is lost. All exclusions were described and validated previously (11, 12). Determination of diffusion coefficients from single molecule data Each data set was analyzed in three ways to determine the best model (i) a homogeneous, 1-component fit; (ii) a 2-component Rayleigh fit; and (iii) a 3-component Rayleigh fit, as described previously (12). Briefly, trajectory displacement data from had been computed for = 8 structures (0.16 s) and changed into binned possibility distribution histograms. These distributions had been fit to an individual component distribution, or the amount of.