Fig 1.
Quorum-sensing circuit in Vibrio cholerae.
Quorum sensing in V. cholerae is controlled by four receptor histidine kinases CqsS, LuxPQ, CqsR and VpsS. At low cell density, these receptors act predominantly as kinases and phosphorylate LuxO through LuxU. Phosphorylated LuxO activates transcription of small RNAs Qrr1-4 which inhibit HapR translation and promote AphA translation, thereby resulting in a low cell density expression profile. At high cell density, when the cognate signals are bound, the kinase activity of these receptors is inhibited. This leads to dephosphorylation of LuxO, preventing Qrr1-4 transcription. Therefore, HapR translation is induced and AphA translation repressed, leading to a high cell density gene expression pattern. Autoinducers CAI-1 (produced by CqsA) and AI-2 (produced by LuxS) have been previously characterized in regulating the kinase activity of CqsS and LuxPQ respectively.
Fig 2.
Effect of QS receptor mutations in V. cholerae infection of the small and large intestine.
Competitive indices (CI) were determined between wild-type ΔlacZ and the indicated V. cholerae mutants in the small intestine (SI) and large intestine (LI) of infant mice 24 hr post-infection (n = 8). Δ3 represents triple receptor mutants with the remaining receptor shown in italics. Each symbol represents the CI in an individual mouse and data is represented with horizontal lines indicating the median with a 95% confidence interval for each competition. ***P < 0.001; ns = no significance (unpaired t test).
Fig 3.
The CqsR CACHE domain is important for signal sensing and signal transduction.
(A) CqsR is predicted to contain all the conserved cytoplasmic domains (phosphoacceptor (PA), histidine kinase (HK), and response regulator (RR) domains) of a hybrid histidine kinase with an N-terminal periplasmic CACHE domain flanked by two transmembrane helices (TM). (B) The predicted structure of the periplasmic CACHE domain of CqsR (right) is similar to that of Mlp37 (left). The residues identified as important for signal sensing and signal transduction are highlighted in cyan in the predicted structure. See main text for further details. (C) Eight periplasmic CqsR residues are involved in signal sensing and signal transduction. Relative light production (lux/OD600) was measured in quadruple QS receptor (Δ4) strains carrying a plasmid producing CqsR with single amino changes as well as a HapR-dependent bioluminescence reporter. Alteration in the CqsR helix domain (R49S), CACHE pocket 1 region (D171V, D198V), CACHE pocket 2 region (L217S, V219E), or the transmembrane proximal regions (A259V, H262Y, L268I) impaired bioluminescence production at high cell-density (OD600 > 1.5). Average values and standard errors from at least three independent replicates are shown.
Fig 4.
Characterization of ethanolamine binding to CqsR.
(A) MST quantification (FNorm; normalized fluorescence) for ethanolamine binding to CqsR was performed by titrating between 200 μM and 0.0061 μM with 20 nM His6-tagged CqsR. Ethanolamine binds to CqsR with a Kd of 0.478 ± 0.076 μM. Binding affinity was calculated from three independent experiments. B) Immunoblots of CqsR using Phos-tag (top) and regular gels (bottom). The singly phosphorylated CqsR runs as a distinct band at a higher molecular weight compared to the unphosphorylated CqsR with a Phos-tag gel. The ratio of phosphorylated to unphosphorylated forms of CqsR is substantially lower in the presence of 10 mM ethanolamine. Differential scanning fluorimetry melt curves of purified C) MBP-CqsR-LBD and D) MBP-CqsRD171V-LBD in the presence of 0 μM (red), 20 μM (blue), 200 μM (green) ethanolamine. A positive shift in Tm of MBP-CqsR-LBD indicates ligand binding, while no shift is observed for MBP-CqsRD171V-LBD in the presence of ethanolamine.
Table 1.
Effects of ethanolamine and its analogs on the melting temp of CqsR-LBD.
Fig 5.
Effects of ethanolamine and its analogs on CqsR quorum-sensing response.
HapR-dependent bioluminescence profiles (lux/OD600) were measured in a Δ3 cqsR+ strain in the presence of 10 mM, 1mM, 0.1 mM A) ethanolamine, B) serinol, C) L-alaninol, D) D-alaninol. Blank indicates LB medium without ethanolamine added. HapR-dependent bioluminescence profiles (lux/OD600) were measured in E) Δ3 cqsRD198V strain and F) Δ3 cqsRD171V strain in the presence of 10 mM, 1mM, 0.1 mM concentrations ethanolamine respectively. Each figure shows a representative profile of each condition with two biological replicates. Each experiment was performed independently at least two times.
Fig 6.
Effect of ethanolamine on CqsR quorum-sensing response inside animal hosts.
CFU counts per A) small intestine homogenate or B) large intestine homogenate collected 8 hr post infection from mice (n = 8) singly infected with a Δ3 cqsR+ or Δ3 cqsRD171V mutant strain. To assess the effect of ethanolamine on strain colonization, mice were gavaged with vehicle (LB) or 10mM ethanolamine (EA) at 2 and 4 hours post infection. C) Competitive indexes (CI) were determined between wild-type ΔlacZ and the indicated V. cholerae mutants in both the small intestine (SI) and large intestine (LI) of infant mice 24 hr post infection (n = 8). The Δ3 cqsR+ data presented in Fig 2 has been included for reference and comparison to Δ3 cqsRD171V. Each symbol represents the CI in an individual mouse and data is represented with horizontal lines indicating the median with a 95% confidence interval for each group. *P < 0.05; **P < 0.01; ***P < 0.001 (unpaired t test). n.s. = not significant.
Fig 7.
Effect of glycerophosphodiesterase deletions on CqsR quorum-sensing response.
HapR-dependent bioluminescence profiles (lux/OD600) were measured in a Δ3 cqsR+ strain and an isogenic strain with deletions in loci vc1554 and vca0136, encoding the two putative glycerophosphodiester phosphodiesterases in LB medium and in the presence of 10 mM ethanolamine. The figure shows a representative profile of each condition with two biological replicates. Each experiment was performed independently at least two times.