Brightness analysis in finite geometries: probing protein interactions in cellular, cell-free and aqueous environments

Abstract

Fluorescence fluctuation spectroscopy (FFS) is a powerful technique for quantitatively analyzing protein interactions. Using brightness analysis methods, we are uniquely able to measure the stoichiometry of protein complexes. FFS is particularly valuable because it allows measurements within living cells. This thesis demonstrates that measuring in very small volumes, such as <italic>E. coli cells</italic>, introduces a bias into the measured brightness. We show that this bias is a result of accumulative sample loss, or photodepletion, and that we can account for this effect and recover correct brightness values. Similarly, very thin samples, such as cell cytoplasm, introduce a bias due to the sample being shorter along the vertical axis than the volume of the excitation light. We introduce z-scan FFS and theory to identify and model thin samples and to recover unbiased data. Although measuring in cells is a primary strength of the FFS technique, some studies require the greater degree of experimental control afforded by solution measurements. Thus, we characterize cell-free expression solution for FFS measurements, an environment that offers increased control but permits genetic fluorescent labeling. We take advantage of this system to perform chromophore maturation experiments as a function of temperature on three common fluorescent proteins: EGFP, EYFP and mCherry. Our results prove that EGFP has fast maturation and is a good reporter for fluorescence experiments. Finally, we apply FFS and brightness analysis to the enzyme, APOBEC3G. We reveal that APOBEC3G interactions with RNA and single-stranded DNA are sequence dependent, which has important implications for the mechanism by which APOBEC3G packages itself into HIV-1 viral particles and restricts the virus to prevent infection.