Skip to main content
NCBI home page
The ISME Journal logoLink to The ISME Journal
. 2012 Jul 5;6(12):2219–2228. doi: 10.1038/ismej.2012.69

Acyl-homoserine lactone-dependent eavesdropping promotes competition in a laboratory co-culture model

Josephine R Chandler 1, Silja Heilmann 2, John E Mittler 1, E Peter Greenberg 1,*
PMCID: PMC3504959  PMID: 22763647

Abstract

Many Proteobacteria use acyl-homoserine lactone (AHL)-mediated quorum sensing to activate the production of antibiotics at high cell density. Extracellular factors like antibiotics can be considered public goods shared by individuals within a group. Quorum-sensing control of antibiotic production may be important for protecting a niche or competing for limited resources in mixed bacterial communities. To begin to investigate the role of quorum sensing in interspecies competition, we developed a dual-species co-culture model using the soil saprophytes Burkholderia thailandensis (Bt) and Chromobacterium violaceum (Cv). These bacteria require quorum sensing to activate the production of antimicrobial factors that inhibit growth of the other species. We demonstrate that quorum-sensing-dependent antimicrobials can provide a competitive advantage to either Bt or Cv by inhibiting growth of the other species in co-culture. Although the quorum-sensing signals differ for each species, we show that the promiscuous signal receptor encoded by Cv can sense signals produced by Bt, and that this ability to eavesdrop on Bt can provide Cv an advantage in certain situations. We use an in silico approach to investigate the effect of eavesdropping in competition, and show conditions where early activation of antibiotic production resulting from eavesdropping can promote competitiveness. Our work supports the idea that quorum sensing is important for interspecies competition and that promiscuous signal receptors allow eavesdropping on competitors in mixed microbial habitats.

Keywords: Burkholderia, Chromobacterium, cell–cell communication, microbial competition, quorum sensing, eavesdropping

Introduction

Quorum sensing affords bacteria the ability to control the expression of specific genes in a cell density-dependent manner (Fuqua et al., 1994, 2001; Bassler, 2002; Waters and Bassler, 2005). Many species of Proteobacteria use small molecules, acylated homoserine lactones (AHLs), as quorum-sensing signals. AHLs are produced by LuxI family synthases, and specifically interact with cytoplasmic LuxR family transcription factors to influence gene expression. AHL specificity is defined by the nature of the acyl side group. AHLs can diffuse through lipid bilayers and thus can move out of and into cells by diffusion. Because of the signal diffusibility, AHLs must reach a critical environmental concentration before they cause changes in gene expression. It is common that the AHL synthase gene is among the genes activated, creating a positive feedback loop that results in increased production of signal (Engebrecht et al., 1983; Seed et al., 1995; Latifi et al., 1996; Duerkop et al., 2009; Stauff and Bassler, 2011). Thus, AHL signaling can coordinate population-wide changes in a cell-density-dependent manner.

Quorum-sensing-regulated genes are predominated by those required for the production of shared ‘public goods', such as secreted or excreted factors. One commonly occurring example is antimicrobials. Quorum-controlled antimicrobials have been described in many saprophytic Proteobacteria including Erwinia carotovora (Bainton et al., 1992), Pseudomonas aeruginosa (Kownatzki et al., 1987; Bainton et al., 1992; Gallagher and Manoil, 2001; Ran et al., 2003; Schuster and Greenberg, 2006), Burkholderia thailandensis (Bt) (Duerkop et al., 2009) and Chromobacterium violaceum (Cv) (Latifi et al., 1995; McClean et al., 1997). Although some groups have proposed that antimicrobial activity of secondary metabolites is a side effect and the primary function of these compounds is as signals (Davies et al., 2006; Yim et al., 2007), the classic view is that they are used for competition with other strains or species in multi-species environments. This classic view suggests that quorum sensing may be important for interspecies competition. Quorum sensing is best understood in the context of virulence, and few studies have addressed its importance in competition (Mazzola et al., 1992; Moons et al., 2005, 2006; An et al., 2006). The advantage of using quorum sensing to control the production of antimicrobials is unknown, but it may allow a population to coordinate delivery of a sudden killing dose that deprives competitors of the ability to adapt during exposure to subinhibitory antimicrobial concentrations (Hibbing et al., 2010, D An and M Parsek, unpublished). Quorum sensing may also defer production of an antimicrobial to minimize the metabolic cost of production.

We are interested in the connection between quorum sensing and production of antibiotics, and specifically whether quorum-sensing-controlled antibiotics are important for interspecies competition. Thus, we developed a dual-bacterial species model with two soil saprophytes, Bt and Cv. Although it is not unlikely that these species coexist in nature, we selected this pair of bacteria because we have a base of knowledge about their quorum-sensing systems, about quorum-sensing control of antibiotic synthesis and because these species exhibit similar laboratory growth characteristics. The Bt genome encodes three LuxR–LuxI pairs. The BtaI1–R1 pair produces and responds to octanoyl-HSL (C8-HSL). Little is known about the genes controlled by this system, but it facilitates clumping under some conditions (Chandler et al., 2009). BtaI3 is a 3-hydroxy-octanoyl-HSL synthase, but little is known about BtaI3–R3 (Chandler et al., 2009). Finally, BtaR2–I2 senses and produces 3-hydroxy-octanoyl-HSL and 3-hydroxy-decanoyl-HSL (Duerkop et al., 2009). The BtaR2–I2 system activates btaI2 and a set of genes responsible for the production of a family of hydrophilic antibiotics, the bactobolins, that have activity against a broad range of bacterial species (Duerkop et al., 2009; Seyedsayamdost et al., 2010; Carr et al., 2011) including Cv (see below). The most potent of these is bactobolin A (Carr et al., 2011).

Cv has a single AHL circuit, the CviR–CviI quorum-sensing system. This circuit activates genes required for the production of a purple pigment called violacein and related compounds that have broad-spectrum antimicrobial activity (McClean et al., 1997). We found that Bt is resistant to purified violacein, but shows sensitivity to other quorum-sensing-dependent factors produced by Cv. The CviI-produced AHL signal is hexanoyl-HSL (C6-HSL), and although CviR is a C6-HSL-responsive transcription factor, it is promiscuous and also responds to a number of different AHL signals (McClean et al., 1997; Swem et al., 2009). This promiscuity may allow Cv to eavesdrop on other AHL-producing species. There are now a number of examples of Proteobacteria with promiscuous LuxR homologs (Pierson et al., 1998; Riedel et al., 2001; Steidle et al., 2001; Venturi et al., 2004; Dulla and Lindow, 2009; Ahlgren et al., 2011; Hosni et al., 2011). It is not known if AHL receptor promiscuity provides any advantage over more signal-specific receptors.

We report here that quorum-sensing-dependent production of antimicrobials can provide a competitive advantage to either Bt or Cv by inhibiting growth of the other species in co-culture. We also present evidence that although Bt and Cv produce different AHLs, the promiscuous signal receptor of Cv can sense Bt signals, and that this ability to eavesdrop on Bt can provide a competitive advantage to Cv. We describe a mathematical model of our dual species system and use this model to show that eavesdropping can promote fitness during competition as long as the population can produce sufficient antibiotic to kill the competitor. Our results support the idea that quorum sensing is important for interspecies competition and that promiscuous signal receptors promote fitness in some situations by enabling eavesdropping on AHLs produced by competitors.

Materials and methods

Bacterial strains and growth

Strains and plasmids are described in the Supplementary Text and Supplementary Table S1. All bacteria were grown in Luria–Bertani (LB) broth containing morpholinepropanesulfonic acid (50 mℳ; pH 7). Bactobolin A was generously supplied by Jon Clardy (Seyedsayamdost et al., 2010) and dissolved in filter-sterilized water. Synthetic C6-HSL and purified violacein were purchased from Sigma-Aldrich (St Louis, MO, USA) and dissolved in acidified ethyl acetate (0.1 ml l−1 glacial acetic acid) or in dimethylformamide, respectively. AHLs were prepared from the Bt bactobolin− strain BD20 by extracting stationary-phase (OD600 8–10) culture fluid with two equal volumes of acidified ethyl acetate and drying to completion under a stream of nitrogen gas. The dried extracts were dissolved in volumes of media equivalent to the volumes from which they were extracted. The extracts did not affect growth of Bt or Cv. Extracts similarly prepared from cultures of an AHL−, bactobolin− double mutant had no effect on the outcome of co-culture experiments. Co-cultures and cultures for AHL preparation were grown at 30 °C. All other growth was at 30 °C for Cv and 37 °C for Bt. Pure cultures and co-cultures containing visibly aggregated cells of Cv were dispersed by homogenization or waterbath sonication before plating for viable counts. Gentamicin was used at 10 μg ml−1 (Cv and Escherichia coli) or 100 μg ml−1 (Bt) and trimethoprim was used at 100 μg ml−1. For selection of Bt and Cv transconjugants, gentamicin was at 10 μg ml−1 and trimethoprim was at 100 μg ml−1.

Antimicrobial susceptibility testing

We determined the minimum inhibitory concentration of bactobolin or violacein using a protocol modified from the 2003 guidelines of the Clinical and Laboratory Standards Institute (CLSI, formerly NCCLS). Inocula were prepared from logarithmic-phase cultures and suspended to 5 × 106 cells in 1 ml morpholinepropanesulfonic acid-buffered LB containing dilutions of antibiotic compounds. The minimum inhibitory concentration was defined as the lowest concentration (μg ml−1) that prevented visible growth of bacteria after 24 h. To assess susceptibility to cell culture fluid, bacteria were similarly suspended in a broth with 10% (Bt) or 75% (Cv) (vol vol−1) filtered fluid from stationary-phase cultures grown for 24 or 16 h, respectively. Culture fluid was filtered through a 0.22-μm pore-size membrane and tested immediately. Fluid from cultures of Cv was diluted into 4 × concentrated LB to a 1 × final LB concentration. Cv and Bt were treated for 24 and 10 h, respectively, before plating for viability. All antimicrobial susceptibility testing was at 30 °C with shaking.

Co-culture experiments

To inoculate co-cultures, pure cultures were grown to mid-logarithmic phase, subcultured to fresh medium at an optical density at 600 nm (OD600) 0.05 and grown an additional 3 h before combining at the appropriate ratios in 10 ml (Figures 1 and 2) or 20 ml (Figures 3 and 4) of medium in 125-ml culture flasks. The initial OD600 of the co-culture was 0.05 (2–4 × 107 cells per ml) for Bt and 0.005 (2–4 × 106 cells per ml) for Cv. Co-cultures were incubated with shaking at 250 r.p.m. Colony-forming units (CFUs) of each species were determined by using differential antibiotic selection on LB agar plates. Bt was selected with gentamicin and Cv was selected with trimethoprim.

Figure 1.

Figure 1

Open in a new tab

B. thailandensis–C. violaceum competition. Initial cell densities were 2–4 × 107 B. thailandensis (Bt) cells per ml and 2–4 × 106 C. violaceum (Cv) cells per ml. The initial and final cell densities of Bt and Cv were determined for each independent experiment by selective plating and colony counts. Each data point represents the log-transformed average of the ratios of the two species from duplicate measurements of an independent co-culture experiment. The lines represent the mean of all of the experiments in each set.

Figure 2.

Figure 2

Open in a new tab

Competition in co-cultures of wild-type C. violaceum (Cv) and wild-type or mutant B. thailandensis (Bt) strains. The dashed line indicates the starting 10:1 ratio of Bt to Cv. The ratio of Bt to Cv after 24 h was determined by selective plating and colony counts. The co-culture results with wild-type Bt are also shown in Figure 1 and the final average CFU of each species is also partially represented in Table 3. Bt AHLs were extracted from culture fluid of a Bt bactobolin mutant (see Materials and methods) and added to culture medium. The solid lines represent means for each group. The vertical bars show the standard error of the mean for each group.

Figure 3.

Figure 3

Open in a new tab

Co-cultures of the C. violaceum (Cv) wild-type Cv017 or the AHL mutant Cv026 and the B. thailandensis (Bt) competition-impaired AHL, bactobolin double mutant JBT125. The dashed line shows the initial ratio of Bt to Cv. After 24 h, the ratio of Bt to Cv was determined by colony counts on selective agar. Co-cultures were grown in 20 ml medium. C6-HSL was added before inoculation where indicated (250 nℳ final concentration). The solid lines represent the means of each group.

Figure 4.

Figure 4

Open in a new tab

C. violaceum (Cv) quorum sensing is activated by B. thailandensis (Bt) AHLs. Quorum-sensing activation is indicated by the Cv quorum-sensing-dependent purple pigment, violacein, in stationary-phase cultures. (a) Cv wild-type (Cv017) and the AHL mutant (Cv026) with or without added Bt AHLs. (b) Co-cultures of the Cv AHL mutant and Bt strains as indicated (AHL mutant JBT112; AHL, bactobolin double mutant JBT125). AHLs were extracted from stationary-phase cultures of Bt BD20, a bactobolin mutant.

Results

Antibiotic sensitivities

As a first step in developing our binary culture model, we needed to test the sensitivity of Cv to bactobolin and the sensitivity of Bt to violacein. Thus, we used purified antibiotics to determine the minimum inhibitory concentrations. The minimum inhibitory concentration of bactobolin A for Cv was 8 μg ml−1, and at concentrations exceeding 8 μg ml−1, Cv was killed during treatment (data not shown). This bactobolin was estimated to be at 5.3 μg ml−1 in pure Bt culture fluid in growth conditions similar to those we use (Seyedsayamdost et al., 2010). Bt produces at least seven other bactobolin compounds (Seyedsayamdost et al., 2010; Carr et al., 2011). To test if Bt-produced bactobolins in cell culture fluid are sufficient to kill Cv, we assessed Cv viability after treatment with filtered fluid from a stationary-phase (OD600 8–10) Bt culture. After treatment with 10% (vol vol−1) culture fluid from a wild-type Bt culture diluted into fresh broth, we were unable to recover viable Cv. After similar treatment with 10% (vol vol−1) culture fluid from a Bt bactobolin-defective mutant (btaK−) or in broth alone, Cv grew to 2–3 × 109 CFU per ml (Table 1). Our results show that stationary-phase Bt cultures produce sufficient bactobolins to kill Cv.

Table 1. Sensitivity of C. violaceum (Cv) strains to B. thailandensis (Bt) culture fluid.

Bt culture fluid testeda Cv (CFU per ml)b
  Wild type AHL−
Wild type <100 <100
AHL− 3 × 109 2 × 109
Bactobolin− 2 × 109 2 × 109
No added culture fluid 2 × 109 1 × 109
Open in a new tab
a

Sensitivity was assessed by growing Cv in the presence of filtered culture fluid from stationary-phase (24 h) Bt cultures as described in the Materials and methods. The Bt AHL (btaI1, I2, I3) mutant JBT125 and the bactobolin (btaK) mutant BD20 were used. The Cv AHL (cviI) mutant Cv026 was used. Experiments were carried out in duplicate and in all cases the ranges did not exceed 10%.

b

Bt cell culture fluid was added to a final concentration of 10% (vol vol−1) in 90% (vol vol−1) in 1 ml Luria–Bertani-morpholinepropanesulfonic acid broth.

Bt was resistant to violacein at the highest concentration tested, 125 μg ml−1 (data not shown), which is in excess of amounts produced by Cv (Tobie, 1935; Strong, 1944). Cv codes for other putative antimicrobial factors, including phenazines and hydrogen cyanide (Brazilian National Genome Project Consortium, 2003). To test whether Cv produces quorum-sensing-dependent antimicrobials with activity against Bt, we incubated Bt with filtered fluid from Cv wild-type or mutant stationary-phase cultures (OD600 4–5). After 10 h, Bt grew modestly to 3 × 108 in the presence of wild-type Cv culture fluid, but grew to 2 × 109 in the presence of fluid from the AHL synthesis mutant (Table 2). This indicates that Cv quorum sensing regulates production of extracellular factors that inhibit growth of Bt, and that this inhibition is not due to violacein alone.

Table 2. Sensitivity of B. thailandensis (Bt) strains to C. violaceum (Cv) culture fluid.

Cv culture fluid testeda Bt (CFU per ml)b
  Wild type AHL− Bactobolin−
Wild type 3 × 108 2 × 108 3 × 108
AHL− 2 × 109 2 × 109 1 × 109
No added culture fluid 7 × 109 8 × 109 8 × 109
Open in a new tab
a

Sensitivity was assessed by growing Bt in the presence of filtered culture fluid from stationary-phase (16 h) Cv cultures as described in the Materials and methods. The Bt AHL (btaI1, I2, I3) mutant JBT125 and the bactobolin (btaK) mutant BD20 were used. The Cv AHL (cviI) mutant Cv026 was used. Experiments were carried out in duplicate and in all cases the ranges did not exceed 10%.

b

Cv cell culture fluid was added to a final concentration of 75% (vol vol−1) in 25% (vol vol−1) concentrated Luria–Bertani-morpholinepropanesulfonic acid broth in 1 ml.

The Bt–Cv co-culture model

In pure culture, the doubling times of all Bt strains were 60 min±5% and Cv strains were 48 min±5% (see Supplementary Table S2), and both species reached densities of about 3 × 109 cells per ml in early stationary phase. Because of the modest growth-rate discrepancy, we used an inoculum of 2–4 × 107 Bt per ml and 2–4 × 106 Cv per ml in our co-culture experiments. Wild-type Bt outcompetes wild-type Cv, increasing in relative abundance by about 100-fold in 24 h (Figure 1). To study the competition further, we enumerated bacteria during logarithmic, early stationary and late stationary growth phases. In logarithmic and early stationary phase, both species reached densities in co-culture that were identical to the densities in pure culture (2–5 × 109 cells per ml). However, the final densities of both species in late stationary phase (24 h) was lower in co-culture than in pure culture (Table 3). The final cell density of Cv decreased over three logs from 5 × 109 cells per ml in early stationary phase to 1 × 106 cells per ml at 24 h. There was no significant decrease in Cv density in pure culture (Table 3). The final density of Bt was 10-fold lower in co-culture than in pure culture (Table 2). Our results are consistent with the hypothesis that both species produce quorum-sensing-controlled antimicrobials during stationary phase that inhibit growth of or kill the other species.

Table 3. Final yields of B. thailandensis (Bt) and C. violaceum (Cv) in pure culture and co-culture.

Strain(s) Final growth yield (CFU per ml) a,b
 
 Bt
 Cv
Pure culture
 Bt wild type 1.4 (±0.7) × 1010  
 Bt AHL− 1.0 (±0.9) × 1010  
 Cv wild type   9.9 (±8.4) × 108
     
Co-culture (with wild-type Cv) c
 Bt wild type 1.3 (±0.8) × 109 1.4 (±2.0) × 106
 Bt AHL− 2.3 (±2.6) × 108 2.1 (±1.2) × 109
 Bt bactobolin− 1.0 (±1.5) × 108 2.7 (±0.5) × 109
Open in a new tab
a

The values are the means of at least three independent experiments with ranges indicated within parantheses. The Bt AHL (btaI1, I2, I3) mutant JBT125 and the bactobolin (btaK) mutant BD20 were used.

b

The growth yield in early stationary phase (9 h) of Bt and Cv in pure and co-culture was 1–3 × 109.

c

Co-culture data from individual experiments are also represented in Figure 1.

Quorum-sensing-controlled bactobolin synthesis promotes Bt competitiveness in binary culture

To test the hypothesis that quorum sensing promotes Bt competitiveness in co-culture, we assessed competition with a Bt AHL mutant and wild-type Cv. We also assessed the competitiveness of a Bt bactobolin mutant. In co-culture conditions where wild-type Bt had a robust competitive advantage, either the Bt AHL or bactobolin mutant were outcompeted by Cv (Figure 2). We could rescue competitiveness of the AHL mutant by supplementing our co-cultures with Bt AHLs that were obtained by ethyl acetate extraction of culture fluid from a stationary-phase (OD600 8–10) Bt bactobolin mutant (Materials and methods). These results demonstrate that quorum sensing and quorum-sensing-dependent bactobolin production are critical for the competitive success of Bt in our co-culture model.

Bactobolin production is controlled by the BtaI2–R2 quorum-sensing system (Duerkop et al., 2009). Next, we assessed the importance of BtaI2–R2 and each of the other two Bt quorum-sensing systems, BtaI1–R1 and BtaI3–R3, to the competitiveness of Bt in our co-culture model. For this, we used Bt strains harboring individual deletions in each of the AHL receptor genes btaR1, btaR2 or btaR3 (Figure 2). Not surprisingly, the btaR2 mutant competed poorly with Cv. Results were similar to those with the bactobolin mutant and the AHL synthesis mutant. The outcome with the btaR3 mutant was identical to wild type, indicating that BtaR3 is not important for competition in our model. The btaR1 mutant showed an intermediate ability to compete with Cv, suggesting that this regulator may be important for the production of bactobolin or production of other factors that enhance competition or bactobolin activity. In support of the former, we found that expression of a bactobolin btaK-lacZ transcriptional fusion is delayed in a btaR1 mutant (data not shown), suggesting that BtaR1 may advance the production of bactobolin. We also tested the competitiveness of strains with individual mutations in each of the AHL synthase genes. All three individual AHL synthase mutants outcompeted Cv with results similar to competitions with wild-type Bt (data not shown). These findings suggest that the AHL synthases have overlapping abilities to induce expression of bactobolin. This is supported by our previous finding that BtaR2 can respond to both 3-hydroxy-octanoyl-HSL and 3-hydroxy-decanoyl-HSL, which are produced by the BtaI3 and BtaI2 synthases, respectively (Duerkop et al., 2009).

Quorum sensing can promote competitiveness of Cv

Our results indicate that Cv also produces quorum-sensing-dependent antimicrobial factors that inhibit growth of Bt (Table 1). Thus, we hypothesized that quorum sensing promotes competitiveness of Cv as it does for Bt. To address this, we compared the competitiveness of the Cv wild-type and AHL mutant strains in co-culture with Bt. We modified our experiment to give wild-type Cv a competitive advantage by using a competition-defective Bt AHL, bactobolin double mutant, and we increased the co-culture volume to 20 ml because we observed that this further improves Cv competitiveness for reasons that are unknown (data not shown). In these conditions, wild-type Cv strongly outcompeted the Bt mutant, whereas the Cv AHL mutant barely outcompeted the Bt mutant (Figure 3). Competitiveness could be restored to the Cv AHL mutant by the addition of C6-HSL (the AHL produced by Cv) (Figure 3). These results show that quorum sensing can promote the competitiveness of Cv. Because violacein does not have any antimicrobial activity against Bt, we note that this is not due to violacein, but must be caused by as-yet undefined quorum-sensing-dependent factors.

Cv can sense and respond to Bt AHLs

The Cv AHL receptor CviR can be activated by a range of AHLs including at least one of the AHLs produced by Bt, C8-HSL (McClean et al., 1997; Swem et al., 2009). We hypothesized that Bt AHLs can activate the Cv quorum-sensing receptor CviR and that this promotes competitiveness of Cv in co-culture with Bt. We first tested whether a pure culture of Cv can sense and respond to Bt AHLs; these AHLs were ethyl acetate extracted and concentrated from stationary-phase (OD600 8–10) culture fluid and added to Cv cultures to match concentrations in the culture from which they were extracted. As a read-out for quorum-sensing activation, we followed the purple pigment violacein. The Cv AHL mutant is not pigmented, but pigmentation can be restored by supplementing the culture medium with Bt AHL extracts (Figure 4a). This result shows that Cv can sense and respond to physiological levels of Bt AHLs.

Next, we tested whether the Cv AHL mutant can respond to Bt AHLs during co-culture growth. Because Cv is killed by Bt-produced bactobolin in co-culture (Table 1), we used the Bt bactobolin mutant BD20 for these experiments (Figure 4b). When in co-culture with a Bt AHL, bactobolin double mutant, the Cv AHL mutant did not turn purple. However, in co-culture with the AHL-producing Bt bactobolin mutant BD20, or with exogenously supplied Bt AHLs, the co-culture turned purple. This finding indicates that the Cv CviR responds to Bt AHLs. We conclude that Bt AHLs are cues that alter the behavior of Cv, although they did not evolve for that purpose (Keller and Surette, 2006). In our experiment, the Cv AHL synthase mutant can eavesdrop on Bt.

Eavesdropping promotes competitiveness of Cv

To determine whether eavesdropping can influence competitiveness of Cv, we enumerated Bt and Cv in co-cultures (Figure 5). As in our previous experiments, we grew the Cv AHL mutant with the Bt bactobolin mutant or an AHL, bactobolin double mutant. The Cv AHL mutant was more competitive with the Bt bactobolin mutant than it was with the double mutant. As a control, we added Bt AHLs to the co-culture with the Bt double mutant and observed that this improved the competitiveness of Cv. These results suggest that eavesdropping on Bt AHLs promotes Cv competitiveness. As an additional control, we tested whether the Cv AHL receptor CviR is required for eavesdropping. To address this, we constructed a Cv AHL synthase, receptor double mutant. We found that CviR is required for the competitive advantage provided to Cv by eavesdropping on Bt AHLs (Figure 5).

Figure 5.

Figure 5

Open in a new tab

Eavesdropping promotes competitiveness of Cv in co-cultures with a B. Bt bactobolin mutant. After 24 h of co-culture, the ratio of Bt to Cv was determined. Co-cultures of the Cv AHL mutant (Cv026), or the Cv AHL synthase, receptor double mutant (Cv026R) and the Bt strains as indicated and described in Figure 4 legend. Co-cultures were grown in 20 ml volumes. The dashed line indicates the initial ratio of Bt to Cv. The solid lines represent the means for each group. AHLs were obtained as described for Figure 4.

An in silico eavesdropping model

Our experimental approach has limitations and with the conditions we used, we could not observe an effect of eavesdropping with wild-type strains (data not shown). However, we suspect there may be conditions where eavesdropping provides an advantage to wild-type Cv. This may be as the population nears the critical density required for quorum-sensing activation. At this density, AHLs produced by a nearby competitor may cause early activation of quorum-sensing-dependent antibiotics and would improve competitiveness of the eavesdropping microbe.

To explore this hypothesis further, we developed a mathematical model of our binary culture system (see Supplementary Text and Supplementary Table S3). The model accounts for two wild-type species that produce antibiotics in response to AHL signals in a well-mixed environment, similar to species Bt and Cv in our experimental system. In silico, the antibiotic produced by each species has equal killing efficiency towards the competing species, but no influence on the producing species. The two species in our in silico model also have identical growth rates, rates of antibiotic and AHL production, and antibiotic-production costs. However, as we observed experimentally, in some conditions one species (which we refer to here as species C) can eavesdrop on the other (species B). In the in silico model, we assume that antibiotic production accelerates once the inducer reaches a critical threshold concentration. However, antibiotic-production rates eventually level off as AHL concentrations exceed the quorum-sensing threshold. We use several different activation thresholds in our analysis.

Our in silico model has a bistable dynamic where one species completely dominates under most conditions. In the absence of eavesdropping, the outcome favors the species that is numerically dominant at the beginning (Figure 6). When we vary the activation thresholds for antibiotic production (by varying KB and KC of B and C, respectively, see Supplementary Text and Supplementary Table S3), there is an optimal value (Koptimal) where one species can dominate the other; if we fix KB at this value, B can dominate C at any value of KC (other than when KC was equal to Koptimal), and the same is true for C if KC is set at Koptimal (see Supplementary Figure S1). For every set of parameter values we explored, we find that Koptimal is greater than zero. Thus, waiting until a population reaches a quorum provides a fitness benefit for antibiotic-producing bacteria.

Figure 6.

Figure 6

Open in a new tab

In silico modeling. Our model accounts for two species with quorum-sensing-controlled antibiotics, similar to our experimental model of Bt and Cv. As in our experimental model, our in silico model accounts for two species (B and C) that produce antibiotics in a density-dependent manner. In our model, species C can eavesdrop on species B (see Supplementary Text). We show relative fitness of each species as a function of the initial ratio (C/B) and the eavesdropping sensitivity (ɛ) of C. The fitness of C relative to B was measured using the log relative fitness measure given in Wu et al. (2006) and is indicated by the color spectrum on the far right. (a) The inducer concentration required for production of antibiotic (activation threshold, KB and KC) is relatively high for both species (0.01, see text). (b) The activation threshold is lower (0.003898) and corresponds to an optimal threshold for each species that gives it an advantage over the other species regardless of the other species' threshold. (c) Both species have an activation threshold lower than the optimal threshold (0.003). (d) The same parameters were used as in (c); however, the antibiotic toxicity is raised 10-fold. This changes the optimal activation threshold to 0.001113, which is below the activation threshold value used (0.003) (see Supplementary Figure S1).

We then investigated eavesdropping in our in silico model when species B and C had identical thresholds above (high), equal to (optimal) and below (low) the optimal threshold. At a relatively high threshold, eavesdropping provided a distinct advantage to C by allowing it to invade B from lower starting frequencies (Figure 6a), supporting our initial hypothesis. However, with an optimal or low threshold, eavesdropping was disadvantageous (Figures 6b and c). We posit that in the latter two cases, the eavesdropping population activates production of antibiotic too early to accumulate a sufficient killing dose and antibiotic production is an ineffective metabolic burden. To test this hypothesis, we kept the same conditions as in Figure 6c and increased the toxicity of the antibiotic of both species. In these conditions, eavesdropping provides an advantage (Figure 6d), supporting our hypothesis. Furthermore, eavesdropping is also advantageous if the antibiotic cost is decreased (Supplementary Figures S2 and S3). However, these changes in toxicity and cost alter the optimal threshold (Supplementary Figure S1B and data not shown), effectively resetting the system so that antibiotic production is induced after the optimal threshold is achieved. Thus, eavesdropping-dependent early production of antibiotics promotes competition in a population that has already reached a sufficient density to produce a killing dose.

Discussion

We have developed a dual-species competition model with two soil saprophytes, Bt and Cv, which both use quorum sensing to control production of antimicrobial factors. We show that both of these species can gain a competitive advantage over the other with success dependent on quorum sensing. The advantage of quorum-sensing control of antimicrobials has also been shown in other laboratory co-culture models (Moons et al., 2005, 2006; An et al., 2006). The previous reports, together with the results reported here, support the idea that quorum-sensing regulation is important in multi-species competition. Our results indicate that competitiveness of Bt relies on the btaI2-R2-controlled antibiotic bactobolin and Cv uses as-yet unidentified quorum-sensing-dependent factors for competition. The bactobolin biosynthetic genes and btaI2-R2 are encoded within a large (120-kb) DNA element that is absent from a close relative, the host-adapted pathogen Burkholderia mallei. That this element is retained in Bt supports the view that btaI2-R2 and bactobolin are important for competition during saprophytic growth.

Why do bacteria use quorum sensing to regulate antibiotic production? Our in silico model provides some possible clues. The results indicate that when antibiotic production is costly, early production slows population growth without effectively killing the competitor. Thus, quorum sensing defers the cost of antibiotic production until a sufficient killing dose can be delivered. We do not include in our model the additional possibility that sublethal concentrations of antibiotics may induce in the competitor an adaptation to higher concentrations of antibiotic. Both of these possibilities can be further explored with our experimental co-culture model. An alternative hypothesis is that deferred production may also protect the producing population against the emergence of non-producing cheaters. Cheaters can exploit public goods producers by utilizing the available goods without incurring the cost of their production. In a recent study by Xavier et al. (2011), delayed production of an exploitable public good, surfactant, protected the producing population against the emergence of cheaters. This strategy maximized growth of the producing population, thereby increasing its ability to compete with cheaters. Quorum-sensing regulation may similarly promote competitiveness with non-producing cheaters.

Our experimental model also showed that cross-species AHL activation of the Cv broad-specificity AHL receptor can promote the competitiveness of Cv (Figure 5). In addition to Cv, there are several other species with broad-specificity AHL receptors and these are also saprophytes: E. carotovora (ExpR2)(Sjoblom et al., 2006); P. aeruginosa (QscR)(Lee et al., 2006); and receptors encoded by two species of Bradyrhizobium (BraR and BjaR)(Ahlgren et al., 2011; Lindemann et al., 2011). ExpR2 and QscR are both orphan receptors without a cognate AHL synthase gene (Cui et al., 2006; Fuqua, 2006; Sjoblom et al., 2006). The potential role of each of these receptors in competition has not been determined. AHL receptor specificity can be easily altered by single amino-acid changes (Collins et al., 2005; Hawkins et al., 2007; Chen et al., 2011; Lintz et al., 2011), suggesting that AHL recognition may be very adaptable in nature. In contrast to these broad-specificity AHL receptors, the receptor of the squid symbiont Vibrio fischeri is quite specific for its cognate AHL (Visick and Ruby, 1999). V. fischeri activates quorum-sensing-dependent functions when it is at high cell densities in its squid host; in this environment it rarely encounters other bacterial species (Visick and McFall-Ngai, 2000). Thus, AHL receptors may evolve broad signal specificity in specific environments where eavesdropping might be of use, although the role of these receptors in inter-species competition and eavesdropping requires further study.

In the conditions of our experimental model, eavesdropping did not provide an observable fitness advantage to wild-type strains during competition. However in another study, AHLs produced by epiphytic bacteria on plant leaves altered the quorum-sensing-regulated virulence phenotype of a wild-type Pseudomonas syringae strain (Dulla and Lindow, 2009), suggesting that wild-type strains can be responsive to AHLs from other species in natural environments. Our co-culture model may provide a limited view of the possible interactions between species in nature, for example, Dulla and Lindow 2009) identified several epiphytic species that produce 10-fold more AHL than their laboratory P. syringae strain. High-level signal producers may play a significant role in cross-species induction.

Our mathematical model allowed a simple assessment of the costs and benefits of eavesdropping between competing wild-type strains. For the model, we made the basic assumption that detection of exogenous AHLs can cause early quorum-sensing-dependent activation of antibiotic genes. We have observed this experimentally in Bt with a transcriptional fusion to the bactobolin biosynthetic gene btaK (data not shown), but it is more difficult to address with Cv because we do not yet know what quorum-controlled genes are involved in competition, and during early logarithmic phase the activity of the antimicrobials is too low for our methods of detection. The in silico model indicates that eavesdropping can promote competition in certain conditions where production of antibiotic occurs relatively late during growth. However, eavesdropping can also be detrimental if the activation threshold is relatively low. We observed similar results in other variations of this model (data not shown). Our results suggest that receptors would evolve broad specificity only in particular circumstances where eavesdropping is beneficial. Our bias is that specificity is the more evolved trait and that highly specific receptors likely arose from receptors with less specificity.

Acknowledgments

This work was funded by a National Institute of Allergy and Infectious Disease (NIAID) award to the Northwest Regional Center of Excellence for Biodefense and Emerging Infectious Diseases (U54AI057141) to EPG. JRC was funded by the NIAID (NRSA 1 F32 AI073027-01A2), SH was funded by the Danish National Research Foundation and JEM was funded by the UW Royalty Research Grant. We thank Drs Ajai Dandekar, Snow Brook Peterson, Ben Kerr and Matt Parsek for helpful discussions regarding this manuscript.

Footnotes

Supplementary Information accompanies the paper on The ISME Journal website (http://www.nature.com/ismej)

Supplementary Material

Supplementary Information
Click here for additional data file. (528KB, doc)

References

  1. Ahlgren NA, Harwood CS, Schaefer AL, Giraud E, Greenberg EP. Aryl-homoserine lactone quorum sensing in stem-nodulating photosynthetic bradyrhizobia. Proc Natl Acad Sci USA. 2011;108:7183–7188. doi: 10.1073/pnas.1103821108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. An D, Danhorn T, Fuqua C, Parsek MR. Quorum sensing and motility mediate interactions between Pseudomonas aeruginosa and Agrobacterium tumefaciens in biofilm cocultures. Proc Natl Acad Sci USA. 2006;103:3828–3833. doi: 10.1073/pnas.0511323103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bainton NJ, Bycroft BW, Chhabra SR, Stead P, Gledhill L, Hill PJ, et al. A general role for the lux autoinducer in bacterial cell signalling: control of antibiotic biosynthesis in Erwinia. Gene. 1992;116:87–91. doi: 10.1016/0378-1119(92)90633-z. [DOI] [PubMed] [Google Scholar]
  4. Bassler BL. Small talk. Cell-to-cell communication in bacteria. Cell. 2002;109:421–424. doi: 10.1016/s0092-8674(02)00749-3. [DOI] [PubMed] [Google Scholar]
  5. Brazilian National Genome Project Consortium The complete genome sequence of Chromobacterium violaceum reveals remarkable and exploitable bacterial adaptability. Proc Natl Acad Sci USA. 2003;100:11660–11665. doi: 10.1073/pnas.1832124100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Carr G, Seyedsayamdost MR, Chandler JR, Greenberg EP, Clardy J. Sources of diversity in bactobolin biosynthesis by Burkholderia thailandensis E264. Org Lett. 2011;13:3048–3051. doi: 10.1021/ol200922s. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chandler JR, Duerkop BA, Hinz A, West TE, Herman JP, Churchill ME, et al. Mutational analysis of Burkholderia thailandensis quorum sensing and self-aggregation. J Bacteriol. 2009;191:5901–5909. doi: 10.1128/JB.00591-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chen G, Swem LR, Swem DL, Stauff DL, O'Loughlin CT, Jeffrey PD, et al. A strategy for antagonizing quorum sensing. Mol Cell. 2011;42:199–209. doi: 10.1016/j.molcel.2011.04.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Collins CH, Arnold FH, Leadbetter JR. Directed evolution of Vibrio fischeri LuxR for increased sensitivity to a broad spectrum of acyl-homoserine lactones. Mol Microbiol. 2005;55:712–723. doi: 10.1111/j.1365-2958.2004.04437.x. [DOI] [PubMed] [Google Scholar]
  10. Cui Y, Chatterjee A, Hasegawa H, Chatterjee AK. Erwinia carotovora subspecies produce duplicate variants of ExpR, LuxR homologs that activate rsmA transcription but differ in their interactions with N-acylhomoserine lactone signals. J Bacteriol. 2006;188:4715–4726. doi: 10.1128/JB.00351-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Davies J, Spiegelman GB, Yim G. The world of subinhibitory antibiotic concentrations. Curr Opin Microbiol. 2006;9:445–453. doi: 10.1016/j.mib.2006.08.006. [DOI] [PubMed] [Google Scholar]
  12. Duerkop BA, Varga J, Chandler JR, Peterson SB, Herman JP, Churchill ME, et al. Quorum-sensing control of antibiotic synthesis in Burkholderia thailandensis. J Bacteriol. 2009;191:3909–3918. doi: 10.1128/JB.00200-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dulla GF, Lindow SE. Acyl-homoserine lactone-mediated cross talk among epiphytic bacteria modulates behavior of Pseudomonas syringae on leaves. ISME J. 2009;3:825–834. doi: 10.1038/ismej.2009.30. [DOI] [PubMed] [Google Scholar]
  14. Engebrecht J, Nealson K, Silverman M. Bacterial bioluminescence: isolation and genetic analysis of functions from Vibrio fischeri. Cell. 1983;32:773–781. doi: 10.1016/0092-8674(83)90063-6. [DOI] [PubMed] [Google Scholar]
  15. Fuqua C. The QscR quorum-sensing regulon of Pseudomonas aeruginosa: an orphan claims its identity. J Bacteriol. 2006;188:3169–3171. doi: 10.1128/JB.188.9.3169-3171.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Fuqua C, Parsek MR, Greenberg EP. Regulation of gene expression by cell-to-cell communication: acyl-homoserine lactone quorum sensing. Annu Rev Genet. 2001;35:439–468. doi: 10.1146/annurev.genet.35.102401.090913. [DOI] [PubMed] [Google Scholar]
  17. Fuqua WC, Winans SC, Greenberg EP. Quorum sensing in bacteria: the LuxR–LuxI family of cell density-responsive transcriptional regulators. J Bacteriol. 1994;176:269–275. doi: 10.1128/jb.176.2.269-275.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gallagher LA, Manoil C. Pseudomonas aeruginosa PAO1 kills Caenorhabditis elegans by cyanide poisoning. J Bacteriol. 2001;183:6207–6214. doi: 10.1128/JB.183.21.6207-6214.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hawkins AC, Arnold FH, Stuermer R, Hauer B, Leadbetter JR. Directed evolution of Vibrio fischeri LuxR for improved response to butanoyl-homoserine lactone. Appl Environ Microbiol. 2007;73:5775–5781. doi: 10.1128/AEM.00060-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hibbing ME, Fuqua C, Parsek MR, Peterson SB. Bacterial competition: surviving and thriving in the microbial jungle. Nat Rev Microbiol. 2010;8:15–25. doi: 10.1038/nrmicro2259. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hosni T, Moretti C, Devescovi G, Suarez-Moreno ZR, Fatmi MB, Guarnaccia C, et al. Sharing of quorum-sensing signals and role of interspecies communities in a bacterial plant disease. ISME J. 2011;5:1857–1870. doi: 10.1038/ismej.2011.65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Keller L, Surette MG. Communication in bacteria: an ecological and evolutionary perspective. Nat Rev Microbiol. 2006;4:249–258. doi: 10.1038/nrmicro1383. [DOI] [PubMed] [Google Scholar]
  23. Kownatzki R, Tummler B, Doring G. Rhamnolipid of Pseudomonas aeruginosa in sputum of cystic fibrosis patients. Lancet. 1987;1:1026–1027. doi: 10.1016/s0140-6736(87)92286-0. [DOI] [PubMed] [Google Scholar]
  24. Latifi A, Foglino M, Tanaka K, Williams P, Lazdunski A. A hierarchical quorum-sensing cascade in Pseudomonas aeruginosa links the transcriptional activators LasR and RhIR (VsmR) to expression of the stationary-phase sigma factor RpoS. Mol Microbiol. 1996;21:1137–1146. doi: 10.1046/j.1365-2958.1996.00063.x. [DOI] [PubMed] [Google Scholar]
  25. Latifi A, Winson MK, Foglino M, Bycroft BW, Stewart GS, Lazdunski A, et al. Multiple homologues of LuxR and LuxI control expression of virulence determinants and secondary metabolites through quorum sensing in Pseudomonas aeruginosa PAO1. Mol Microbiol. 1995;17:333–343. doi: 10.1111/j.1365-2958.1995.mmi_17020333.x. [DOI] [PubMed] [Google Scholar]
  26. Lee JH, Lequette Y, Greenberg EP. Activity of purified QscR, a Pseudomonas aeruginosa orphan quorum-sensing transcription factor. Mol Microbiol. 2006;59:602–609. doi: 10.1111/j.1365-2958.2005.04960.x. [DOI] [PubMed] [Google Scholar]
  27. Lindemann A, Pessi G, Schaefer AL, Mattmann ME, Christensen QH, Kessler A, et al. Isovaleryl-homoserine lactone, an unusual branched-chain quorum-sensing signal from the soybean symbiont Bradyrhizobium japonicum. Proc Natl Acad Sci USA. 2011;108:16765–16770. doi: 10.1073/pnas.1114125108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Lintz MJ, Oinuma K, Wysoczynski CL, Greenberg EP, Churchill ME. Crystal structure of QscR, a Pseudomonas aeruginosa quorum sensing signal receptor. Proc Natl Acad Sci USA. 2011;108:15763–15768. doi: 10.1073/pnas.1112398108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Mazzola M, Cook RJ, Thomashow LS, Weller DM, Pierson LS. Contribution of phenazine antibiotic biosynthesis to the ecological competence of fluorescent pseudomonads in soil habitats. Appl Environ Microbiol. 1992;58:2616–2624. doi: 10.1128/aem.58.8.2616-2624.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. McClean KH, Winson MK, Fish L, Taylor A, Chhabra SR, Camara M, et al. Quorum sensing and Chromobacterium violaceum: exploitation of violacein production and inhibition for the detection of N-acylhomoserine lactones. Microbiology. 1997;143 (Part 12:3703–11. doi: 10.1099/00221287-143-12-3703. [DOI] [PubMed] [Google Scholar]
  31. Moons P, Van Houdt R, Aertsen A, Vanoirbeek K, Engelborghs Y, Michiels CW. Role of quorum sensing and antimicrobial component production by Serratia plymuthica in formation of biofilms, including mixed biofilms with Escherichia coli. Appl Environ Microbiol. 2006;72:7294–7300. doi: 10.1128/AEM.01708-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Moons P, Van Houdt R, Aertsen A, Vanoirbeek K, Michiels CW. Quorum sensing dependent production of antimicrobial component influences establishment of E. coli in dual species biofilms with Serratia plymuthica. Commun Agric Appl Biol Sci. 2005;70:195–198. [PubMed] [Google Scholar]
  33. Pierson EA, Wood DW, Cannon JA, Blachere FM, Pierson LS. Interpopulation signaling via N-acyl-homoserine lactones among bacteria in the wheat rhizosphere. Mol Plant Microbe Interact. 1998;11:1078–1084. [Google Scholar]
  34. Ran H, Hassett DJ, Lau GW. Human targets of Pseudomonas aeruginosa pyocyanin. Proc Natl Acad Sci USA. 2003;100:14315–14320. doi: 10.1073/pnas.2332354100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Riedel K, Hentzer M, Geisenberger O, Huber B, Steidle A, Wu H, et al. N-acylhomoserine-lactone-mediated communication between Pseudomonas aeruginosa and Burkholderia cepacia in mixed biofilms. Microbiology. 2001;147:3249–3262. doi: 10.1099/00221287-147-12-3249. [DOI] [PubMed] [Google Scholar]
  36. Schuster M, Greenberg EP. A network of networks: quorum-sensing gene regulation in Pseudomonas aeruginosa. Int J Med Microbiol. 2006;296:273–81. doi: 10.1016/j.ijmm.2006.01.036. [DOI] [PubMed] [Google Scholar]
  37. Seed PC, Passador L, Iglewski BH. Activation of the Pseudomonas aeruginosa lasI gene by LasR and the Pseudomonas autoinducer PAI: an autoinduction regulatory hierarchy. J Bacteriol. 1995;177:654–659. doi: 10.1128/jb.177.3.654-659.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Seyedsayamdost MR, Chandler JR, Blodgett JA, Lima PS, Duerkop BA, Oinuma K, et al. Quorum-sensing-regulated bactobolin production by Burkholderia thailandensis E264. Org Lett. 2010;12:716–719. doi: 10.1021/ol902751x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sjoblom S, Brader G, Koch G, Palva ET. Cooperation of two distinct ExpR regulators controls quorum sensing specificity and virulence in the plant pathogen Erwinia carotovora. Mol Microbiol. 2006;60:1474–1489. doi: 10.1111/j.1365-2958.2006.05210.x. [DOI] [PubMed] [Google Scholar]
  40. Stauff DL, Bassler BL. Quorum sensing in Chromobacterium violaceum: DNA recognition and gene regulation by the CviR receptor. J Bacteriol. 2011;193:3871–3878. doi: 10.1128/JB.05125-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Steidle A, Sigl K, Schuhegger R, Ihring A, Schmid M, Gantner S, et al. Visualization of N-acylhomoserine lactone-mediated cell–cell communication between bacteria colonizing the tomato rhizosphere. Appl Environ Microbiol. 2001;67:5761–5770. doi: 10.1128/AEM.67.12.5761-5770.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Strong FM. Isolation of violacein. Science. 1944;100:287. doi: 10.1126/science.100.2596.287. [DOI] [PubMed] [Google Scholar]
  43. Swem LR, Swem DL, O'Loughlin CT, Gatmaitan R, Zhao B, Ulrich SM, et al. A quorum-sensing antagonist targets both membrane-bound and cytoplasmic receptors and controls bacterial pathogenicity. Mol Cell. 2009;35:143–153. doi: 10.1016/j.molcel.2009.05.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Tobie WC. The pigment of Bacillus violaceus: I. The production, extraction, and purification of violacein. J Bacteriol. 1935;29:223–227. doi: 10.1128/jb.29.3.223-227.1935. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Venturi V, Friscina A, Bertani I, Devescovi G, Aguilar C. Quorum sensing in the Burkholderia cepacia complex. Res Microbiol. 2004;155:238–244. doi: 10.1016/j.resmic.2004.01.006. [DOI] [PubMed] [Google Scholar]
  46. Visick KL, Ruby EG.1999The emergent properties of quorum sensing: consequences to bacteria of autoinducer signaling in their natural environmentIn: Winans GMDaSC (ed.)Cell–Cell Signaling in Bacteria ASM Press: Washington, DC; 333–352. [Google Scholar]
  47. Visick KL, McFall-Ngai MJ. An exclusive contract: specificity in the Vibrio fischeri–Euprymna scolopes partnership. J Bacteriol. 2000;182:1779–1787. doi: 10.1128/jb.182.7.1779-1787.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Waters CM, Bassler BL. Quorum sensing: cell-to-cell communication in bacteria. Annu Rev Cell Dev Biol. 2005;21:319–346. doi: 10.1146/annurev.cellbio.21.012704.131001. [DOI] [PubMed] [Google Scholar]
  49. Wu H, Huang Y, Dykes C, Liu D, Ma J, Perelson AS, et al. Modeling and estimation of replication fitness of human immunodeficiency virus type 1 in vitro experiments by using a growth competition assay. J Virol. 2006;80:2380–2389. doi: 10.1128/JVI.80.5.2380-2389.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Xavier JB, Kim W, Foster KR. A molecular mechanism that stabilizes cooperative secretions in Pseudomonas aeruginosa. Mol Microbiol. 2011;79:166–179. doi: 10.1111/j.1365-2958.2010.07436.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Yim G, Wang HH, Davies J. Antibiotics as signalling molecules. Philos Trans R Soc Lond Ser B. 2007;362:1195–1200. doi: 10.1098/rstb.2007.2044. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information
Click here for additional data file. (528KB, doc)

Articles from The ISME Journal are provided here courtesy of Oxford University Press

RESOURCES