| Escherichia coli swims toward niches ideal for generating maximum energy by monitoring the environment with the aerotaxis sensor, Aer. This receptor has an N-terminal PAS domain, which is a sensor module present in a variety of proteins from all kingdoms of life (including humans). To date, the mechanism of signaling from a PAS sensor domain to a single signaling receiver domain has not been resolved. In Aer, the PAS sensor is separated from the signaling region by a membrane anchor. Most "transmembrane prediction" programs forecast a single membrane span for this membrane anchor, but such a topology is inconsistent our previous studies indicating the PAS and HAMP domains, which flank the membrane region, are in contact with one another.; To determine the general topology of the membrane anchor, I used a cysteine scanning approach. I estimated the proximity of 57 cognate cysteine replacements by measuring the rate and extent of dimer formation after adding an oxidant at 23°C, and at 4°C to limit random collisions. I found that the Aer membrane anchor formed two membrane-spanning segments that flank a central periplasmic loop. To map the membrane boundaries, I used hydrophilic sulfhydryl-reactive agents to measure the surface accessibility of these introduced cysteines. I then expanded the cysteine scanning method to include binary, ternary and quaternary combinations of introduced cysteines. Binary combinations were used to differentiate intra- from inter-dimeric collisions and thereby map the faces of Aer monomers. Ternary and quaternary combinations were used to study larger complexes. Aer formed hexamers at both the periplasmic and cytosolic ends, but this stable structure was orchestrated by the cytosolic signaling domain, because C-terminal truncations missing the signaling region altered the periplasmic neighborhood; here, octamers and decamers were trapped, indicating that the signaling region had limited the degrees of freedom in the periplasmic region. Lastly, strategic cross-linking combinations pairing periplasmic and cytosolic cysteine replacements suggested that different Aer trimers-of-dimers could be trapped if the oxidant copper phenanthroline were added at high concentrations. However, the myriad of complexes formed suggested that these interactions were not specific. |
