Fig 1.
Model for LuxR-mediated induction of V. fischeri luminescence by C8-HSL and 3OC6-HSL.
The pheromones C8-HSL and 3OC6-HSL interact with LuxR to form complexes with dissociation constants K1 and K2 (and Hill coefficients m and n) respectively. These complexes bind to the lux box (dissociation constants KA and KB respectively) to activate expression of the lux operon and synthesis of LuxA and LuxB, the subunits of the bacterial luciferase. The luminescence is proportional to the concentration of the LuxA-LuxB heterodimer [29].
Table 1.
Bacterial strains and plasmids used in this study.
Fig 2.
The lux operon in V. fischeri and genomic organization of strains engineered for this study.
Panel (A) illustrates the genetic structure of the lux locus in parental wild-type strain ES114 (top) as well as the strains used to assay LuxR activity, wherein the native luxRI is deleted and luxR alleles are placed in an engineered construct between ORF VFA0926 and luxC (hatched box). Panel (B) shows specific sequences of ES114 aligned with those of engineered strains. Red sequences are stop codons for luxR (reverse strand) and luxI. The green ATG represents the start codon for luxC. Further details (e.g. cloning strategy, sequence of the “consensus promoter”, etc.) are provided in Methods.
Fig 3.
Comparing data and fit for ΔainRS mutants of the four LuxR variants.
Each of panels (A)-(D) shows a representative luminescence dataset and fit for one of the ΔainRS strains, where luminescence is measured as a function of C8-HSL and 3OC6-HSL concentration. The luminescence (vertical) axis for data is shown on a logarithmic scale. The lower figure of each group shows the residual on a logarithmic scale, i.e. the ratio data/fit is shown on a logarithmic scale. The set of parameter values obtained in 150 fits of 3 datasets for each LuxR are shown in Fig 6 and summarized in Table 2.
Fig 4.
Effect of ΔainR on the response to C8-HSL.
Each panel shows the luminescence of a LuxR variant in both ainR+ (blue) and ainR- (red) background. Left panels show results in the presence of 3.2 μM 3OC6-HSL and right panels show results in absence of 3OC6-HSL. All data are the average of at least three independent replicates. Error bars indicate standard deviation of the replicates. Luminescence is given in units of detector counts.
Fig 5.
Illustration of the role of the model parameters.
In order to show how the predicted HSL response is shaped by the values of the interaction parameters in the model (see Methods), the predicted response is shown for several different choices of the C8-HSL parameters while the parameters for 3OC6-HSL are held fixed. (Analogous figures are generated if the 3OC6-HSL parameters are varied while the C8-HSL parameters are held fixed). (A) Calculated luminescence signal (vertical scale) versus 3OC6-HSL and C8-HSL concentrations (horizontal scales), for parameter values k1 = 100 nM, k2 = 100 nM, m = 1, n = 1, kA = 10 nM, kB = 10 nM; (B) Calculated response for the same parameter values as in (A), except with k1 increased fivefold to k1 = 500 nM; (C) Response for same parameter values as in (A), except with m decreased to 0.3; (D)-(F) Response for same parameters as in (A), except with kA changed to 3 nM, 30 nM, and 90 nM respectively. In all cases the overall scale parameters (Eq 4) are fixed at a0 = 10 and a1 = 550, which were typical for our datasets and analysis.
Fig 6.
Fit results for the four ΔainR strains, obtained by fitting luminescence data for each strain to the six-parameter model of Fig 1.
For each of four LuxR variants, 150 independent optimizations of the model were performed with respect to three independent luminescence experimental datasets. The histograms below indicate the results obtained for (A)-(C) the C8-HSL interaction parameters and (D)-(F) the 3OC6-HSL interaction parameters. The red line indicates the median result for a given parameter and LuxR, while the yellow box indicates the span of the 25th -75th percentiles for the parameter. The dissociation constants k1, k2, kA, kB are the scaled (relative to [LuxR0]—see Methods) dissociation constants and accordingly have units of nM.
Fig 7.
Correlation between interaction parameters for C8-HSL and 3OC6-HSL.
Correlation between interaction parameters for C8-HSL (horizontal axes) and 3OC6-HSL (vertical axes) is shown for the four LuxR variants. All strains are ΔainR. Each point represents parameter values obtained in one of the 150 fits performed for each LuxR. The color code (blue = LuxRMJ1, green = LuxRES114, red = LuxRA, yellow = LuxRB) indicates the LuxR variant. Panels (A), (B), and (C) show results for complex dissociation (k1, k2), Hill coefficient (m, n), and lux binding (kA, kB) respectively. The black dashed line in each panel corresponds to equality between C8-HSL and 3OC6-HSL parameters; k1 = k2, m = n, or kA = kB. The larger shaded circles highlight (with the same color code) the median value obtained for each LuxR.
Table 2.
Parameter estimates obtained from fit to luminescence (vs. C8-HSL and 3OC6-HSL) data.