Structure, function and FliM interaction studies of CheY6 in Rhodobacter sphaeroides

Bacteria control their swimming direction by signalling from chemoreceptors via a small protein, CheY, to the rotary flagellar motor. Rhodobacter sphaeroides has a complex chemosensory network with two pathways, including three different CheYs controlling a stop-start motor. Deletions of these cheYs result in a non-chemotactic phenotype. CheY6 is essential for chemotaxis, whereas CheY3 and CheY4 have some functional redundancies. Although CheY6 alone can stop the motor, the presence of either CheY3 or CheY4 is required for a chemotactic phenotype. To date, little is known about how these three CheY proteins interact with the flagellar motor, or the switch mechanism used between the inactive and active states. Structural studies of CheY6 using NMR experiments, highlighted a flexible loop region (residues S109-K118) that is not present in CheY3, CheY4 or CheYs in other bacterial species. This elongated loop region was deleted (CheY6-ÎLoop), and in vivo studies were used to investigate its function. CheY6 ÎLoop is folded, retains the ability to be phosphorylated by CheA3, localises at the cytoplasmic chemoreceptor cluster, but appears unable to stop the flagellar motor. Circular dichroism and NMR data suggest that CheY6-ÎLoop is a folded protein that shows similar peak shifts to wild type CheY6 upon activation. In wild type CheY6, residues in this loop show chemical shift changes upon addition of the phosphoryl mimic BeF3-, suggesting that the loop is involved in the activation mechanism. The switch mechanism of CheY6 was probed using multidimensional NMR studies. The active state was mimicked using BeF3- . Wild type CheY6 was shown to undergo structural changes upon addition of BeF3-. In particular, the β4 α4-loop and residues located near the phosphorylatable D56 show large changes in chemical shift. Superposition of this loop region revealed possible steric clashes with the N-terminus of FliM. S83 shows evidence of involvement in the switch mechanism. Residual dipolar coupling experiments suggest that the published crystal of CheY6 in complex with CheA3, is more like the inactive conformation in solution. CheY has been shown to interact with the motor switch protein FliM in other bacterial species. CheY6-FliM interactions were probed using a combination of in vitro and in vivo studies. Bacterial two hybrid assays and NMR studies suggest that CheY6 cannot interact with monomeric FliM. Single molecule total internal reflection microscopy revealed CheY6 does interact with the motor in vivo. The data in combination suggests a model in which CheY6-P only interacts with FliM when it is part of the switch ring, and structural changes involved in stopping the motor depend on the large conformational changes in the β4-α4-loop.