STRUCTURAL AND FUNCTIONAL STUDIES OF SALMONELLA TYPHIMURIUM EFFECTORS GTGE AND SPVB

Abstract

Salmonella is a genus of Gram-negative bacteria, which is a major cause of foodborne disease. To create a safe intracellular environment for the pathogen within the host cells, Salmonella secretes a large number of effectors into the host cytosol. My research concentrates on a structural and functional characterization of two Salmonella effectors, GtgE and SpvB. GtgE is a cysteine protease that specifically cleaves closely related GTPases, Rab29, Rab32 and Rab38. The full-length GtgE and several truncated constructs were cloned, expressed and purified. Extensive crystallization trials were performed with these constructs. In addition, several crystallization rescue strategies have been employed, including introducing entropy-reducing surface mutations, chemical modification to the protein surface, and co-crystallization of inactive GtgE variants with substrate peptide. Despite these extensive efforts, no crystals were obtained. However, new information about GtgE was discovered such as low protease activity in vitro against GST-Rab32 and identification of the N-terminal ~30 residues of GtgE as required for the full function. SpvB is a mono-ADP-ribosyltransferase that modifies G-actin. SpvB is composed of two structural domains. The C-terminal domain of SpvB (SpvB-C) possesses the mono-ADP-ribosyltransferase activity and its structure has been determined. This study provides the first structural determination of the N-terminal domain of SpvB (SpvB-N) at 2.4 Å resolution. This domain is made primarily of β-strands and shows similarity to the N-terminal segment of YenB, an ABC toxin component. A long groove on the protein surface suggests that it functions as a recognition domain. The hypothesis that SpvB-N guides the localization of SpvB and targets the protein to actin was tested, but a co-immunoprecipitation assay excluded strong interaction between SpvB-N and actin. Moreover, SpvB-N expressed as a GFP fusion localized to the nucleus while SpvB and SpvB-C localized to the cytosol. I have shown further that SpvB-C was sufficient to disrupt the actin cytoskeleton and induce cell apoptosis. The similarity of SpvB-N to YenB prompted us to investigate its cytotoxicity but only a marginal effect on the host cells was noted. Experiments to study the function of the SpvB-N were able to exclude some possibilities and narrowed down the spectrum of potential functions.