National Food Biotechnology Centre, University College Cork, Cork, Ireland.
Dueling quorum sensing systems in Pseudomonas aeruginosa control the production of the Pseudomonas quinolone signal (PQS)
Article first published online: 9 JAN 2006
FEMS Microbiology Letters
Volume 230, Issue 1, pages 27–34, January 2004
How to Cite
McGrath, S., Wade, D. S. and Pesci, E. C. (2004), Dueling quorum sensing systems in Pseudomonas aeruginosa control the production of the Pseudomonas quinolone signal (PQS). FEMS Microbiology Letters, 230: 27–34. doi: 10.1016/S0378-1097(03)00849-8
- Issue published online: 9 JAN 2006
- Article first published online: 9 JAN 2006
- Received 27 August 2003, Revised 20 October 2003, Accepted 7 November 2003
- Pseudomonas aeruginosa;
- Quorum sensing;
The opportunistic human pathogen Pseudomonas aeruginosa regulates the production of numerous virulence factors via the action of two separate but coordinated quorum sensing systems, las and rhl. These systems control the transcription of genes in response to population density through the intercellular signals N-(3-oxododecanoyl)-l-homoserine lactone (3-oxo-C12-HSL) and N-(butanoyl)-l-homoserine lactone (C4-HSL). A third P. aeruginosa signal, 2-heptyl-3-hydroxy-4-quinolone [Pseudomonas quinolone signal (PQS)], also plays a significant role in the transcription of multiple P. aeruginosa virulence genes. PQS is intertwined in the P. aeruginosa quorum sensing hierarchy with its production and bioactivity requiring the las and rhl quorum sensing systems, respectively. This report presents a preliminary transcriptional analysis of pqsA, the first gene of the recently discovered PQS biosynthetic gene cluster. We show that pqsA transcription required pqsR, a transcriptional activator protein encoded within the PQS biosynthetic gene cluster. It was also found that the transcription of pqsA and subsequent production of PQS was induced by the las quorum sensing system and repressed by the rhl quorum sensing system. In addition, PQS production was dependent on the ratio of 3-oxo-C12-HSL to C4-HSL, suggesting a regulatory balance between quorum sensing systems. These data are an important early step toward understanding the regulation of PQS synthesis and the role of PQS in P. aeruginosa intercellular signaling.
The Gram-negative bacterium Pseudomonas aeruginosa is an opportunistic pathogen that can cause serious infections in plants, animals and humans. This organism is a major cause of nosocomial infections and is responsible for chronic lung infections that plague most cystic fibrosis patients (see  for review). To augment its pathogenicity, P. aeruginosa carefully controls the production of many virulence factors through a population density-dependent mechanism known as quorum sensing (see  for review). Two distinct but related quorum sensing circuits have been identified in P. aeruginosa. Both of these systems are genetically similar in that they consist of genes encoding transcriptional activator proteins (lasR and rhlR) as well as genes responsible for the production of acylated homoserine lactone signaling molecules (lasI and rhlI) [3–7]. The intercellular signals for the las and rhl quorum sensing systems are N-(3-oxododecanoyl)-l-homoserine lactone (3-oxo-C12-HSL) and N-(butanoyl)-l-homoserine lactone (C4-HSL), respectively [8,9]. Together, these signals have been shown to control hundreds of genes, representing 4–12% of the P. aeruginosa genome [10–12].
In addition to 3-oxo-C12-HSL and C4-HSL, a third intercellular signal is produced by P. aeruginosa. This signal is 2-heptyl-3-hydroxy-4-quinolone and is referred to as the Pseudomonas quinolone signal (PQS) . PQS is required for virulence in a nematode model of infection and it is produced in the lungs of cystic fibrosis patients infected by P. aeruginosa[14–16]. The role of PQS in cell to cell signaling is still undetermined, but it has been shown to be part of the P. aeruginosa quorum sensing hierarchy. PQS production is positively regulated by the las quorum sensing system and PQS bioactivity depends on the presence of RhlR . Interestingly, in addition to increasing LasB elastase production, PQS has also been demonstrated to induce the expression of rhlI, which encodes the C4-HSL synthase . These observations led to the conclusion that PQS acts as a connector signal between the las and rhl quorum sensing systems.
Two recent reports have shown that at least nine genes make up the PQS biosynthetic regulon [16,18]. The discovery of this regulon has provided the opportunity to further our studies on the regulation of PQS synthesis. In this report, we present our preliminary transcriptional analysis which indicated that PQS synthesis is positively controlled by two transcriptional regulators and negatively controlled by a third transcriptional regulator.
2Materials and methods
2.1Bacterial strains, plasmids, and culture conditions
Bacterial strains and plasmids are listed in Table 1. P. aeruginosa strains were maintained at −70°C in 10% skim milk (Becton Dickinson). P. aeruginosa cultures were grown in peptone tryptic soy broth  at 37°C with vigorous shaking. When necessary to maintain plasmids, carbenicillin or tetracycline was added to growth media at a concentration of 200 μg ml−1 or 50 μg ml−1, respectively. Plasmid pLP0996 was constructed as follows. Two primers, 5′-AAAAAGCTTAGGTGTCCTCTTCGGCAGG-3′ and 5′-AAACTGCAGCAGCCGGCTGAGAGTCTGGC-3′, were used to amplify the pqsA promoter region in a polymerase chain reaction (PCR) reaction with Pfu Turbo DNA polymerase (Stratagene). The first primer hybridizes to a DNA region centered 524 bp upstream of the pqsA start codon and contains an added HindIII restriction enzyme recognition site (with three A nucleotides to aid digestion) on the 5′-terminus. The second primer will hybridize to a DNA region centered 89 bp downstream of the pqsA start codon and contains an added PstI restriction enzyme recognition site (with three A nucleotides to aid digestion) on the 5′-terminus. The amplified fragment was digested with HindIII and PstI and ligated into pLP170 that had been digested with the same enzymes. The recombinant plasmid (pLP0996), which contained a pqsA′-lacZ transcriptional fusion, was electroporated into P. aeruginosa strains for conducting expression analysis studies.
|Strain or plasmid||Relevant genotype or phenotype||Reference|
|P. aeruginosa strains|
|MP551||pqsR::TnIS phoA/hah-Tc; derived from strain PAO1|||
|PDO100||rhlI::Tn501-2; derived from strain PAO1|||
|PAO-JP1||lasI::Tet; derived from strain PAO1|||
|PAO-JP2||lasI::Tet, rhlI::Tn501-2; derived from strain PDO100|||
|pLP0996||pqsA′-lacZ transcriptional fusion; bla||This study|
|pMTP58||minimum tiling pathway cosmid 58 which contains pqsA; tetr|||
2.2Primer extension analysis
RNA was purified from P. aeruginosa strain PAO1 (pMTP58) as described previously  and quantified spectrophotometrically. Plasmid pMTP58 was included in order to improve the extension process by increasing the number of copies of pqsA mRNA present. Two primers were used in separate primer extension experiments. These primers were: 5′-AACGGCGGTATCGGGATC-3′, which corresponds to nucleotides +66 to +49 relative to the pqsA start codon, and 5′-GCTGAGAGTCTGGCCCCG-3′, which corresponds to nucleotides +93 to +76 relative to the pqsA start codon. Primers were radioactively labeled using [γ-32P]ATP (NEN Research Products) and T4 polynucleotide kinase (Invitrogen). Primer extension reactions were carried out as described previously  using 20 μg RNA and Superscript II RNase H− Reverse Transcriptase (Invitrogen). Primer extension reactions were electrophoresed on a sequencing gel along with DNA sequencing reactions performed with the same primer that was used for the primer extension reaction. Sequencing reactions were completed using a T7 Sequenase version 2.0 DNA sequencing kit (USB).
2.3mRNA analysis by reverse transcriptase (RT)-PCR
RNA was purified and quantitated from P. aeruginosa strain PAO1 as described above. To remove traces of contaminating DNA, total RNA was treated with amplification grade DNase I (Invitrogen) and subsequently purified with an RNeasy Mini Kit (Qiagen). Approximately 50 ng of DNase-treated RNA was used as a template for RT-PCR performed using the Promega Access RT-PCR system. Primer pairs were specifically designed to span the intergenic regions of the genes in the PQS biosynthetic regulon and are listed in Table 2. First strand cDNA synthesis was carried out at 48°C for 45 min. Second strand synthesis and PCR amplification were carried out in a Stratagene Robocyler. Reactions were heated to 95°C for 2 min and then underwent 30 cycles with the following parameters: 95°C for 1 min, 57–64°C for 1 min (different annealing temperatures were required to optimize each primer pair), and 72°C for 2.5 min. The final cycle was followed by heating at 72°C for 10 min. Positive controls were performed using genomic DNA, and negative controls were performed using RNA that had not undergone retrotranscription. Reaction products were analyzed by agarose gel electrophoresis.
|Primer name||Sequence||Genea||Expected length (bp)b|
2.4β-Galactosidase (β-gal) assays
To assay β-gal from the pqsA′-lacZ fusion, overnight cultures were washed and inoculated to a starting OD660 of 0.08. Cultures were grown for 24 h and β-gal was assayed as described by Miller . All β-gal assays were performed in duplicate and repeated at least three times. Data are presented as the mean±σn−1.
2.5Analysis of PQS production
Freshly plated P. aeruginosa cultures were used to inoculate 10-ml cultures for overnight growth. These cultures were then used to inoculate fresh 10-ml subcultures to an OD660 of 0.05 using washed cells. After 24 h of growth, cultures were extracted once with acidified ethyl acetate as described previously . To acidify ethyl acetate, glacial acetic acid was added to a final concentration of 435 μM. Extracts were dried and resuspended in 1:1 acidified ethyl acetate:acetonitrile. Aliquots of extracts were loaded onto thin layer chromatography (TLC) plates (20×20 cm silica gel 60 F254) (EM Science) and separated as described previously using 17:2:1 methylene chloride:acetonitrile:dioxane as the solvent [13,22]. Prior to use, TLC plates were soaked in 5% w/v KH2PO4 and activated by baking at 100°C for 1 h as described previously . Samples were separated on TLC plates which were then photographed under long-wave UV light. Photographs were analyzed using the ImageQuant program (Molecular Dynamics) in order to determine relative amounts of PQS.
3.1Determination of the pqsA transcriptional start site
The PQS biosynthetic gene cluster is organized as a set of eight genes, the first seven of which are arranged contiguously in the same orientation (see Fig. 1A). The first gene in this group is pqsA (PA0996), which encodes a hydroxybenzoate CoA ligase homolog, which could presumably charge the PQS precursor, anthranilate, with coenzyme A to prepare it for a condensation reaction with a decanoic acid . This gene was shown to be required for the synthesis of PQS  so we set out to learn more about how it is regulated.
To begin our studies, we located the transcriptional start site of pqsA by performing primer extension analysis. This analysis showed that the pqsA transcriptional start site was located 71 bp upstream of the gene's translational start codon (Fig. 1B). This analysis was repeated with a different primer (see Section 2) and the transcriptional start site was mapped to the same nucleotide determined above (data not shown). An inspection of the newly mapped pqsA promoter region revealed two regions of interest. Putative quorum sensing operator sequences were centered at 20.5 and 29.5 bp upstream of the pqsA transcriptional start site (Fig. 1C). The putative operator sequences matched 11 out of 20 nucleotides (for the sequence proximal to the transcriptional start) and 10 out of 20 nucleotides (for the distal sequence) when compared to the lasB OP1 operator sequence . In addition, each sequence matched three out of four of the highly conserved nucleotides found in P. aeruginosa quorum sensing controlled operator sequences (consensus sequences contain CT in positions 3 and 4, and AG in positions 17 and 18 ). The significance of having two putative operators in the −35 region of the pqsA promoter is not known, but they may be indicative of a complex regulation scheme that involves both the las and rhl quorum sensing systems. The pqsA promoter region did not contain a sequence that was similar to a known consensus sequence for a sigma factor recognition site.
3.2Analysis of the pqsA transcript
To learn more about the genetic organization of the PQS regulon, we performed RT-PCR using primer pairs that would indicate whether mRNA continued from one gene to the next. When using mRNA as a template, products were successfully amplified between all genes from pqsA to pqsE (Fig. 2). This indicated that pqsABCDE formed a polycistronic operon, which confirms the proposed operon structure implied by our previous studies on mutation polarity within the pqsABCDE region . RT-PCR did not amplify a product that would span pqsE to phnA (Fig. 2), indicating that the pqsABCDE mRNA terminated after pqsE. We also showed that phnA and phnB were contiguous on a common mRNA (Fig. 2), indicating a phnAB operon. Overall, these studies showed that the linked PQS biosynthetic regulon consists of two polycistronic operons (pqsABCDE and phnAB) and pqsR (formerly mvfR). Other studies have indicated that at least one additional unlinked gene, pqsH (PA2587), is also part of the PQS biosynthetic pathway .
3.3Transcription of pqsABCDE is controlled by quorum sensing and pqsR
Our previous studies showed that PQS production was controlled by the las quorum sensing system . Since pqsA is the first gene in the PQS biosynthetic gene cluster, we thought that it may be controlled by LasR and 3-oxo-C12-HSL. Therefore, we examined the expression of pqsA′-lacZ in a lasI mutant (strain PAO-JP1). As we expected, the loss of the las quorum sensing system in strain PAO-JP1 caused a large decrease in pqsA transcription (Fig. 3). The wild-type strain PAO1 containing pLP0996 produced 9551±2184 Miller units of β-gal activity while strain PAO-JP1 (pLP0996) produced only 1183±317 Miller units of β-gal activity. This indicated that pqsABCDE transcription was controlled by the las quorum sensing system.
We also examined the effect of pqsR on pqsABCDE transcription. This gene is on the opposite strand from pqsA and is located directly downstream of phnB (see Fig. 1A). The pqsR gene encodes a LysR-type regulator which is required for the expression of the phnAB operon . It has also been shown that a pqsR mutant (strain MP551) does not produce PQS [16,18]. Analysis of pqsA′-lacZ expression in strain MP551 showed that pqsR was required for pqsABCDE expression (Fig. 3). This is especially interesting because others have shown that lasR and rhlR have no effect on pqsR expression, and that pqsR has no effect on lasR or rhlR expression . This indicated that PQS production was independently regulated in a positive manner by two separate transcriptional activator proteins, both of which were required for induction of pqsABCDE.
3.4Transcription of pqsABCDE is inhibited by the rhl quorum sensing system
To learn more about the control of PQS production, we tested the effect of the rhl quorum sensing system on pqsABCDE expression. When the pqsA′-lacZ fusion plasmid was placed into an rhlI mutant (strain PDO100), we observed a dramatic increase in β-gal activity (Fig. 3). Strain PDO100 (pLP0996) produced 73 484±7133 Miller units of β-gal activity, which was more than seven-fold higher than the activity seen from strain PAO1 (pLP0996). These intriguing results led us to conclude that C4-HSL may be repressing the induction of pqsABCDE. To further examine this effect, we assayed pqsABCDE expression in the presence of different combinations of 3-oxo-C12-HSL and C4-HSL. Strain PAO-JP2 (pLP0996) was grown with 3-oxo-C12-HSL and/or C4-HSL and β-gal activity from pqsA′-lacZ was measured. As expected, pqsA′-lacZ was not induced in the absence of exogenous cell to cell signals and the addition of 3-oxo-C12-HSL alone caused a large induction of pqsA′-lacZ (Fig. 4). The induction of pqsA′-lacZ by 3-oxo-C12-HSL alone was much higher than that seen from the wild-type strain (see Fig. 3) and indicated again that pqsABCDE repression was lost when C4-HSL was absent. When C4-HSL was added in conjunction with 3-oxo-C12-HSL, the expression of pqsA′-lacZ was negatively affected (Fig. 4). This negative effect became more pronounced with increasing concentrations of C4-HSL. These data indicated that pqsABCDE expression was being positively regulated by the las quorum sensing system and negatively regulated by the rhl quorum sensing system. This suggests that the two quorum sensing systems compete for pqsABCDE regulation and that this regulation is dependent on the ratio of 3-oxo-C12-HSL to C4-HSL.
3.5The C4-HSL signal inhibits PQS production
To extend our studies beyond the level of gene transcription, we examined the effects of C4-HSL on PQS production. Strains PA01, PAO-JP2, and PDO100 were grown for 24 h and PQS was extracted from the cultures. The extracts were separated by TLC and PQS was quantified by densitometric analysis of TLC photographs. As shown previously , we observed that strain PAO-JP2 produced no PQS (Fig. 5). More importantly, we found that strain PDO100 produced an amount of PQS that was more than six-fold higher than that seen from the wild-type strain PAO1 (Fig. 5). This increase in PQS production matched the increase in pqsABCDE transcription that was seen earlier from strain PDO100 (see Fig. 3).
We then analyzed the effect of C4-HSL on the ability of 3-oxo-C12-HSL to stimulate PQS production. In the presence of 3-oxo-C12-HSL, increasing concentrations of C4-HSL had a negative effect on PQS production (Fig. 6). This was similar to the effect seen on pqsABCDE transcription and strengthens our conclusion that PQS production depends on the ratio of the signals from the las and rhl quorum sensing systems.
In this report, we show that the biosynthesis of the intercellular signal PQS is controlled in both a positive and negative manner by multiple transcriptional regulators. To begin these studies we analyzed the transcript produced by pqsA, the first gene in the PQS biosynthetic cluster. Primer extension analysis (Fig. 1B) and RT-PCR experiments (Fig. 2) respectively mapped the pqsA transcriptional start site and showed that pqsA was part of a five-gene operon (pqsABCDE). There were no obvious sigma factor consensus sequences in the pqsA promoter region, but two potential quorum sensing operator sequences overlapped with the −35 region of the pqsA promoter (Fig. 1C). While these sequences were appropriately located to mediate pqsA transcription, their significance is not known. The results from the RT-PCR experiment are in agreement with earlier data which suggested that pqsA was transcriptionally linked to the next four contiguous genes to form the pqsABCDE operon . We also saw that the next gene in the PQS synthetic regulon, phnA, was not transcriptionally linked to pqsE (Fig. 2), which implied that phnA is transcribed from its own promoter. The phnA transcriptional start site has not been identified, but these data support the work of others who showed that the DNA directly upstream of phnA was bound and controlled by PqsR . Finally, our mRNA studies showed that phnA and phnB are transcriptionally linked to form a phnAB operon (Fig. 2). These genes encode the large (phnA) and small (phnB) subunits of anthranilate synthase , which presumably synthesizes the anthranilate required for PQS.
Our initial experiments on pqsABCDE expression determined that both the las quorum sensing system and pqsR are absolutely required for pqsABCDE transcription (Fig. 3). The lack of pqsABCDE expression in a lasI mutant nicely extended our earlier studies that showed PQS production was dependent on the las quorum sensing system . These data also imply that the potential quorum sensing operators located in the −35 region of the pqsA promoter are important. The finding that pqsR was required for pqsABCDE expression was also interesting. Previous studies had shown that pqsR positively controlled the expression of phnA which is required for PQS production [16,25]. Taken in conjunction with our data, it is apparent that pqsR is a master regulator of the PQS biosynthetic gene cluster.
The regulation scheme for pqsABCDE was further complicated after examining the influence of the rhl quorum sensing system. In an rhlI mutant, the expression of pqsA′-lacZ was increased seven-fold (Fig. 3) and PQS production increased six-fold (Fig. 5). This suggested that the rhl quorum sensing system was having repressive effects on pqsABCDE and subsequent PQS production. We then found that the production of PQS was dependent on the ratio of the two quorum sensing signals, 3-oxo-C12-HSL and C4-HSL. C4-HSL actually inhibited the ability of 3-oxo-C12-HSL to activate pqsABCDE expression (Fig. 4) and PQS production (Fig. 6). These data indicated that there was a competition occurring at the pqsABCDE promoter.
The mechanism of competition occurring at the pqsABCDE promoter is not known, but we can speculate as to what might be happening. The simplest explanation is that the RhlR–C4-HSL complex can bind and thereby block the pqsABCDE promoter from being fully activated by PqsR and LasR–3-oxo-C12-HSL. This would explain the presence of two potential quorum sensing operator sequences in the pqsABCDE promoter region. One could be used by LasR–3-oxo-C12-HSL to activate the pqsABCDE promoter, and the other could be used by RhlR–C4-HSL to block the same promoter. A second possible explanation for the repression of the pqsABCDE promoter by C4-HSL is that C4-HSL is competing with 3-oxo-C12-HSL for binding to LasR, which is required for pqsABCDE transcription. The possibility of a competitive inhibition such as this has been seen with 3-oxo-C12-HSL blocking the ability of C4-HSL to activate RhlR . In spite of this, we believe that it is an unlikely scenario because Pearson et al.  indirectly showed that C4-HSL binds to RhlR but not LasR. Finally, C4-HSL may be causing QscR to repress pqsABCDE transcription. QscR is capable of repressing genes activated by either the las or rhl quorum sensing system . This mechanism could also include the two putative operator sequences in the pqsABCDE promoter. However, at this time there is no evidence that QscR responds to any cell to cell signal, including C4-HSL . All three of the theories discussed here can incorporate our data that showed the ratio of C4-HSL to 3-oxo-C12-HSL affected pqsABCDE expression, which does not help us to favor one theory over the other. It should also be noted here that additional theories could also be based on the recently discovered potential for heterodimer formation between the three P. aeruginosa LuxR homologs . While the mechanism behind the observed negative regulation of pqsABCDE is not known, it is obviously complex and requires further study.
Our results imply that at least three transcriptional regulator proteins (LasR, RhlR, and PqsR) could be interacting with the promoter upstream of the pqsABCDE operon. How and why these interactions occur is not known, but as we learn more about the production of the intercellular signal PQS, it becomes apparent that the synthesis of this signal is controlled in a very precise manner. Considering this along with recent studies which suggest that PQS is produced and differentially regulated by P. aeruginosa in the lungs of infected cystic fibrosis patients [14,30], it is interesting to speculate that the timing of PQS production may influence the survival of P. aeruginosa during infections.
This work was supported by research grants from the National Institutes of Allergy and Infectious Disease (Grant R01-AI46682) and The Mary Lynn Richardson Fund. We thank T. de Kievit, J.P. Coleman, and C.S. Pesci for help in manuscript preparation and thoughtful insight.
- Pesci, E.C. and Iglewski, B.H. (2003) Quorum sensing. In: Bacterial Protein Toxins (Burns, D.L., Barbieri, J.T., Iglewski, B.H. and Rappuoli, R., Eds.), pp. 55–65. ASM Press, Washington, DC.
- Miller, J. (1972) Experiments in Molecular Genetics, pp. 352–355. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.