Abstract
The outer membrane (OM) of Gram-negative bacteria such as Escherichia coli (E. coli) creates an impermeable membrane thus requiring active transport proteins for essential nutrient uptake to ensure bacterial survival. This task is executed by TonB-dependent transporters (TBDTs) that utilize the proton motive force (PMF) of the inner membrane (IM) by coupling with the TonB/ExbB/ExbD complex. BtuB is a TonB-dependent transporter responsible for binding and uptake of Cobalamin (Vitamin B12) across the OM, and BtuB is an essential protein for the proper function of human microbiome. Though numerous crystal structures are available for multiple TBDTs, the molecular mechanism of TBDT function is poorly characterized.
Numerous studies have shown that site-directed spin labeling (SDSL) coupled with electron paramagnetic resonance (EPR) spectroscopy is a powerful tool to study structural and functional dynamics of membrane proteins. We have shown that single and double cysteine mutations can be successfully incorporated to BtuB, and spin labeled for continuous wave and pulse EPR spectroscopy. However, labeling cysteine pairs in BtuB in in-vivo proves to be challenging as efficient labeling requires the use of a knockout strain where the bacterial periplasmic redox homeostasis is modified.
In in this thesis we begin by examining a knockout strain, which is deficient in the disulfide bond oxidase dsbA, and demonstrate that is promotes a much more efficient double spin labeling of pairs of cysteines on both periplasmic and extracellular surfaces for BtuB, when the protein is purified and reconstituted into phospholipid bilayers. Surprisingly, double electron-electron resonance (DEER) experiments in the dsbA null strain reveal the presence of intermolecular BtuB-BtuB interactions, which were not previously observed in purified and phospholipid reconstituted preparations. This is likely a result of the improvement in labeling efficiency of BtuB. Second, we demonstrate that this approach is necessary to perform in-vivo double spin labeling of the ferric citrate transporter FecA, and that expression in the dsbA null strain also improves the in-vitro labeling of FecA. The expression of FecA in the dsbA null strain required a two-plasmid transformation that allowed FecA induction with arabinose. As a result of this approach we were able to demonstrated that extracellular loops of FecA execute a different gating behavior in the cell than they do when the protein is purified and reconstituted into a phospholipid bilayer.
In BtuB, a highly conserved N-terminal energy coupling motif termed the Ton box is known to switch from an ordered to a disordered state upon substrate binding, where it extends into the periplasmic space and may interact with the IM protein TonB. This structural change requires the breaking of an ionic interaction between R14 in the core domain and D316 in the BtuB β-barrel. In a second portion of this thesis, we use the dsbA null mutant to show that the permanent disruption of this ionic lock results in a greater extension of the Ton box than is observed with substrate addition alone, and that this behavior can be mimicked by the binding of the C-terminal domain of TonB. This structural change may open a pathway for substrate transport. Finally, it is known that an extracellular substrate binding loop, SB3, moves 2 nm toward the periplasmic surface in-vivo. We show that this transition can also be observed in an isolated OM preparation when either the R14-D316 ionic lock is disrupted or when TonB C-terminal domain is bound to BtuB Ton box. This indicates that structural transitions that are seen in the intact cell can be observed in an isolated preparation provided that components in the native membrane and protein partners critical for TBDT transport are present.
