Differential repositioning of the second transmembrane helices from E. coli Tar and EnvZ upon moving the flanking aromatic residues

Abstract

Aromatic tuning, i.e. repositioning aromatic residues found at the cytoplasmic end of transmembrane (TM) domains within bacterial receptors, has been previously shown to be an efficient way to modulate signal output from the aspartate chemoreceptor (Tar) and the major osmosensor EnvZ of Escherichia coli. In the case of Tar, changes in signal output consistent with the vertical position of the native Trp-Tyr aromatic tandem within TM2 were observed. In contrast, within EnvZ, where a Trp-Leu-Phe aromatic triplet was repositioned, the surface that the triplet resided upon was shown to be the major determinant governing signal output. However, these previous studies failed to determine whether moving the aromatic residues within TM2 of Tar or EnvZ was sufficient to physically reposition the TM helix within a membrane. Recent coarse-grained molecular dynamics (CG-MD) simulations predicted displacement of Tar TM2 upon moving the aromatic residues at the cytoplasmic end of TM2. Here, we have employed a glycosylation-mapping technique to demonstrate that repositioning the Trp-Tyr tandem within Tar TM2 is sufficient to displace the C-terminal boundary of the helix relative to the membrane. In a similar analysis of EnvZ, an abrupt initial displacement of TM2 was observed but no subsequent movement was seen, suggesting that the vertical position of TM2 is not governed by the location of the Trp-Leu-Phe triplet. In summary, our results support recent CG-MD simulations with aromatically tuned Tar segments that demonstrated that the Trp-Tyr tandem is sufficient to displace TM2 within a membrane. Our results also provide another set of experimental data, i.e. the resistance of EnvZ TM2 to being displaced upon aromatic tuning, which could be useful for subsequent refinement of the initial CG-MD simulations. We suggest that differences observed between the behavior of helices are due to the inherently different properties of the residues being repositioned (i.e. Trp or Tyr versus Phe). Finally, we discuss the limitations of these methodologies, how moving flanking aromatic residues might impact steady-state signal output and the potential to employ aromatic tuning in other bacterial membrane-spanning receptors.