A quorum sensing regulated small volatile molecule reduces acute virulence and promotes chronic infection phenotypes

Abstract

A significant number of environmental microorganisms can cause serious, even fatal, acute and chronic infections in humans. The severity and outcome of each type of infection depends on the expression of specific bacterial phenotypes controlled by complex regulatory networks that sense and respond to the host environment. Although bacterial signals that contribute to a successful acute infection have been identified in a number of pathogens, the signals that mediate the onset and establishment of chronic infections have yet to be discovered. We identified a volatile, low molecular weight molecule, 2-amino acetophenone (2-AA), produced by the opportunistic human pathogen Pseudomonas aeruginosa that reduces bacterial virulence in vivo in flies and in an acute mouse infection model. 2-AA modulates the activity of the virulence regulator MvfR (multiple virulence factor regulator) via a negative feedback loop and it promotes the emergence of P. aeruginosa phenotypes that likely promote chronic lung infections, including accumulation of lasR mutants, long-term survival at stationary phase, and persistence in a Drosophila infection model. We report for the first time the existence of a quorum sensing (QS) regulated volatile molecule that induces bistability phenotype by stochastically silencing acute virulence functions in P. aeruginosa. We propose that 2-AA mediates changes in a subpopulation of cells that facilitate the exploitation of dynamic host environments and promote gene expression changes that favor chronic infections.

Figure 1. 2-AA synthesis is controlled by MvfR.

(A) LC/MS total ion chromatograms show the relative percentage of the most abundant small molecules in P. aeruginosa (PA14) (upper panel) and isogenic mvfR mutant (lower panel) supernatants after 9 h of growth. The most abundant molecules in PA14 cultures that were absent in mvfR mutant cultures were: DHQ (1); 2-AA (2); HQNO (3); and HHQ+PQS (4+5). The abundant peaks shown in mvfR are negligible compared to the amount of HHQ or PQS produced by PA14 since the percentage value on the Y axis is drawn with respect to the most abundant molecule detected. (B) Production kinetics of 2-AA in PA14 supernatants in LB. PA14 growth is shown as OD_{600 nm} on the secondary Y axis versus time. The chemical structure of 2-AA is shown in the inset. (C) 2-AA is not a MvfR co-inducer. MvfR-dependent pqsA-GFP(ASV) expression in the pqsA::pqsH mutant with or without exogenous molecules (10 µg/ml), reported as relative fluorescence versus time, averaged for six replicates.

Figure 2. 2-AA silences MvfR regulon in a subpopulation of cells and restricts HAQ mediated QS signaling relevant in acute infection.

(A) Quantification of pqsA-GFP(ASV) expression in PA14 cells in response to increasing 2-AA concentration (µg/ml); the error bar represents the standard deviation of the 6 replicates. The difference in fluorescence intensity at the peak was statistically significant at all concentrations with p values <0.005. (B) Expression of pqsA-GFP(ASV) in response to 2-AA (µg/ml). Images of total number of cells in the optical field as assessed by membrane staining (red) and bacterial cells expressing pqsA-GFP(ASV)(green). (C) Quantification of pqsA expressing cells (percentage of green/red) in response to each 2-AA concentration. (D) Production of HHQ in supernatants collected from PA14 cells (OD 2.0) with exogenously added 2-AA at 0–200 µg/ml HHQ levels were quantified in triplicate by LC/MS. The error bars represent the standard deviation from triplicate samples. (E) Levels of pyocyanin in PA14 cultures grown (to OD 3.0) in LB and LB supplemented with varying (0–200 µg/ml) concentrations of 2-AA. Error bars represent standard deviation of the triplicate samples. These experiments were repeated at least three times with similar results. (F) Kinetics of pyoverdine production in the presence of 100 and 200 µg/ml of 2-AA. (-) indicates no exogenously added 2-AA.

Figure 3. Negative regulation of the MvfR regulon by 2-AA is a result of down-regulation of pqsABCDE expression and interference with MvfR activity via inhibition of HHQ biosynthesis.

(A) pqsA promoter response was assessed by measuring pqsA-LacZ expression in the PA14 isogenic pqsA::pqsH double mutant in the presence of HHQ (10 µg/ml) or HHQ together with 2-AA at the indicated concentrations. β-galactosidase activity is given in Miller Units, and plotted against growth assessed at OD_{600 nm}. The experiment was repeated three times with similar results. (B) Levels of HHQ in the supernatant (OD 2.0) of PA14 and mvfR mutant cells constitutively expressing pqsABCD on a plasmid. HHQ levels were quantified in triplicate by LC/MS. The error bars represent the standard deviation of triplicate samples. (C) Growth of PA14 and pqsB::C mutant cells carrying pqsA-SacB after 20 h of inoculation in the presence or absence of sucrose (10%), and/or with exogenously added various concentrations of 2-AA. The pqsA promoter of pqsA-SacB in the pqsB::C mutant was induced by exogenous addition of 10 µg/ml HHQ where indicated. (D) Concentration of pyocyanin in mvfR mutant cells (OD 3.0) over-expressing MvfR or PqsE under a constitutive promoter in the presence or absence of 2-AA (200 µg/ml). Error bars represent standard deviations of triplicate samples. These experiments were repeated at least three times with similar results.

Figure 4. 2-AA reduces P. aeruginosa virulence.

(A) Survival kinetics of flies injected with a PA14 bacterial suspension with (∼1.6 ng/fly) and without 2-AA. Data were averaged from two experiments with n(PA14) = 55 and n(PA14+2-AA) = 56. Significance of difference of survival rate was calculated using the log-rank test of the Kaplan Meier survival estimate (p = 0.0056). (B) Survival rates of mice infected with PA14 versus PA14 + 67 µg of 2-AA/mouse. The data correspond to the averages of two independent experiments, with n(PA14) = 20 and n(PA14+2-AA) = 20. Significance of difference of survival rates was calculated using the Kaplan-Meier method (p = 0.03), with a hazard ratio of 1.8932 (95% CI, 1.0664 to 6.0718). (C and D) Bacterial loads in the local and adjacent muscle and in blood were determined for control and 2-AA treated mice, at 20 h post-burn and infection. The statistical significance of the CFU/mg muscle tissue differences between the control and experimental mice were determined using the Mann-Whitney test for independent samples, with p = 0.43 for the difference in the underlying and adjacent muscle in response to 2-AA, and p = 0.045 for the difference in the blood in response to 2-AA. CFU data are presented as log 10. Black PA14, red PA14+2-AA, green 2-AA only.

Figure 5. 2-AA promotes phenotypes associated with pathogenic bacteria at chronic infection sites.

(A) 2-AA promotes increased accumulation of lasR mutations in PA14 and mutS cells in a concentration-dependent manner. Data are averages of two independent experiments, performed in triplicate. Significance of difference in number of lasR mutants was calculated using Student's t-test, assuming equal variance. Asterisks represent significant differences relative to the untreated control group. For both PA14 and mutS mutant cells, lasR mutant accumulation was significantly increased with 100 µg/ml and 200 µg/ml 2-AA (p's = 0.02 and 0.0001 for PA14, and p's = 0.05 and 0.01 for mutS, respectively). (B) 2-AA promotes prolonged stationary phase growth. Untreated PA14 cells began to lyse by 40 h, whereas PA14 cells treated with 2-AA at all tested concentrations did not. (C) 2-AA promotes persistent infection in flies. Bacterial CFUs per fly were assessed at 7 d post-P. aeruginosa feeding (6–9 flies per time point). Significant differences in CFUs after 7 d were seen for pqsA (pink) versus PA14 (black) or pqsB::C (green); Student's t-test p values are indicated in the figure. The data presented were combined from two independent experiments. (D) A model for 2-AA function. As the volatile (squiggly grey lines) 2-AA accumulates overtime in a microenvironment (i.e. biofilms, wounds or CF lungs) it creates a heterogeneous population of cells that consists of WT and physiologically and genetically different cells. The unaltered WT cells undergo lysis. On the other hand, the subpopulation of cells with silenced MvfR can continue to accumulate mutations that are beneficial for adaption to chronic infection.