Quorum-sensing-dependent regulation of biosynthesis of the polyketide antibiotic mupirocin in Pseudomonas fluorescens NCIMB 10586

A. Kassem El-Sayed1, Joanne Hothersall1 and Christopher M. Thomas1

School of Biosciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK1

Author for correspondence: Christopher M. Thomas. Tel: +44 121 414 5903. Fax: +44 121 414 5925. e-mail: c.m.thomas{at}bham.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Mupirocin (pseudomonic acid) is a polyketide antibiotic, targeting isoleucyl-tRNA synthase, and produced by Pseudomonas fluorescens NCIMB 10586. It is used clinically as a topical treatment for staphylococcal infections, particularly in contexts where there is a problem with methicillin-resistant Staphylococcus aureus (MRSA). In studying the mupirocin biosynthetic cluster the authors identified two putative regulatory genes, mupR and mupI, whose predicted amino acid sequences showed significant identity to proteins involved in quorum-sensing-dependent regulatory systems such as LasR/LuxR (transcriptional activators) and LasI/LuxI (synthases for N-acylhomoserine lactones – AHLs – that activate LasR/LuxR). Inactivation by deletion mutations using a suicide vector strategy confirmed the requirement for both genes in mupirocin biosynthesis. Cross-feeding experiments between bacterial strains as well as solvent extraction showed that, as predicted, wild-type P. fluorescens NCIMB 10586 produces a diffusible substance that overcomes the defect of a mupI mutant. Use of biosensor strains showed that the MupI product can activate the Pseudomonas aeruginosa lasRlasI system and that P. aeruginosa produces one or more compounds that can replace the MupI product. Insertion of a xylE reporter gene into mupA, the first ORF of the mupirocin biosynthetic operon, showed that together mupR/mupI control expression of the operon in such a way that the cluster is switched on late in exponential phase and in stationary phase.

Keywords: N-acylhomoserine lactone, Staphylococcus aureus, reporter genes, genetic manipulation, diffusible activator

Abbreviations: AHL, N-acylhomoserine lactone; C4-HSL, N-butyryl-L-homoserine lactone; C6-HSL, N-hexanoyl-L-homoserine lactone; 3-oxo-C6-HSL, N-(3-oxohexanoyl)-L-homoserine lactone; C10-HSL, N-decanoyl-L-homoserine lactone; 3-OH-C6-HSL, N-(3-hydroxyhexanoyl)-L-homoserine lactone; 3-OH-C8-HSL, N-(3 hydroxyoctanoyl)-L-homoserine lactone; 3-OH-C10-HSL, N-(3 hydroxydecanoyl)-L-homoserine lactone; 3-oxo-C12-HSL, N-(3-oxododecanoyl)-L-homoserine lactone; 3-OH-C14:1-HSL, N-(3-hydroxytetradecenoyl)-L-homoserine lactone

The GenBank accession numbers for the sequences determined in this work are AF318063 (mupA), AF318064 (mupR) and AF318065 (mupI).


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Pseudomonic acid (mupirocin) is a polyketide antibiotic produced by the Gram-negative soil-borne bacterium Pseudomonas fluorescens NCIMB 10586. It has potent antimicrobial activity, which blocks protein synthesis by competitively inhibiting isoleucyl-tRNA synthase (Hughes & Mellows, 1978 , 1980 ). Mupirocin is used as a topical treatment for staphylococcal colonization and infections (Rode et al., 1991 ). It cannot be used systemically due to serum binding as well as rapid degradation, so there is potential to alter these properties by genetic manipulation. We have also been keen to establish how the genes of the mupirocin cluster are regulated since this may provide a way of increasing antibiotic yield.

Many antibiotics are secondary metabolites that accumulate in late exponential and stationary phase. A widespread mechanism that controls this pattern of gene expression is termed quorum sensing (Kaiser & Losick, 1993 ; Fuqua et al., 1994 ; Greenberg, 2000 ). It depends on the ability of bacteria to communicate with each other through constitutive production of a variety of small diffusible signal molecules called autoinducers (Kuo et al., 1994 ). These compounds switch on a response when the bacteria reach a population density, a quorum, at which autoinducer accumulates to a concentration sufficient to induce the target function.

The archetypal example of quorum sensing is the autoinduction of the lux regulon, responsible for bioluminescence of the marine symbiotic bacterium Vibrio fischeri (Kaplan & Greenberg, 1985 ; Kuo et al., 1994 ). This system consists of two divergently transcribed operons (Martin et al., 1989 ). The first operon encodes the transcriptional activator luxR, while the second operon consists of seven ORFs starting with luxI, whose product is involved in production of the N-acylhomoserine lactone (AHL) autoinducer N-(3-oxohexanoyl)-L-homoserine lactone (3-oxo-C6-HSL), followed by six structural genes (luxCDABEG) that are responsible for light production (Cohn et al., 1985 ; Dunn et al., 1973 ; Johnston et al., 1986 ; Meighen, 1988 ).

Quorum-sensing gene regulation is not restricted to control of bioluminescence phenomena, but also regulates many other physiological processes including plasmid conjugative transfer, production of virulence factors and exoenzymes, and antibiotic biosynthesis (Fuqua et al., 1996 ). The best-studied quorum-sensing dependent system in Pseudomonas species is found in Pseudomonas aeruginosa, where two interconnected quorum-sensing regulons have been identified (Greenberg, 2000 ). The first one is the las system, which regulates the expression of a number of virulence and exoenzyme genes such as lasB (elastase; Gambello & Iglewski, 1991 ), aprA (alkaline protease) and exotoxin A ‘exoA (Gambello et al., 1993 ). These genes are positively activated via a complex of the transcriptional activator LasR bound with the autoinducer N-(3-oxododecanoyl)-L-homoserine lactone (3-oxo-C12-HSL), which is produced by LasI. Small amounts of 3-oxo-C12-HSL are required to activate LasR. Once it is activated, it works efficiently to express lasI, producing more 3-oxo-C12-HSL as a positive feedback to activate more LasR, which in turn positively regulates the other genes including lasB, aprA and exoA (Seed et al., 1995 ). An additional layer of control is provided by a recently discovered RsaL, which represses lasI at low cell density, preventing activation of the quorum-sensing cascade in P. aeruginosa during the early stage of growth (De Kievit et al., 1999 ). The second quorum-sensing system consists of RhlR regulatory protein and RhlI as an autoinducer synthase for N-butyryl-L-homoserine lactone (C4-HSL) (Winson et al., 1995 ). This system regulates the rhlAB operon, which directs the biosynthesis of rhamnolipid (Ochsner et al., 1994 ), lasB (Pearson et al., 1997 ) and expression of the stationary-phase sigma factor RpoS (Latifi et al., 1996 ). Furthermore, there is a mutual regulatory link between the lasR/lasI and rhlR/rhlI systems that makes the LasR/LasI combination global quorum-sensing regulators in P. aeruginosa (Latifi et al., 1996 ).

We previously mapped a cluster of genes associated with mupirocin biosynthesis by P. fluorescens NCIMB 10586 to a chromosomal segment more than 60 kb long (Whatling et al., 1995 ). This paper presents the first evidence that these genes are controlled through a quorum-sensing-dependent regulatory system. Two genes within the mup cluster, mupR and mupI, were cloned, sequenced, and identified as possible regulatory loci because their predicted amino acid sequences showed significant identity to LasR/LasI and LuxR/LuxI regulators. Reporter gene fusions and deletion mutations for mupR and mupI established the involvement of these two genes in the mupirocin biosynthesis pathway and confirmed their role in the activation of mup operon transcription. The study also showed that maximum expression of the mup operon occurs in stationary phase.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and plasmids.
The bacterial strains used in this study are listed in Table 1. The plasmids used or constructed during this work are listed in Table 2. Pseudomonas fluorescens NCIMB 10586 was used as the wild-type mupirocin producer. Escherichia coli strains TG1 and NEM259 were used for plasmid transformation and propagation, while S17-1 was used for mobilization of the suicide plasmids to P. fluorescens. Bacillus subtilis was used for bioassay of mupirocin production. Chromobacterium violaceum ATCC 31532 and CV026 (a violacein-negative, mini-Tn5 mutant of ATCC 31532) were used as reporters to assess the presence of exogenous AHLs (McClean et al., 1997 ).


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains used during this work

 

View this table:
[in this window]
[in a new window]
 
Table 2. Plasmids used or constructed during this study

 
Growth and culture conditions.
E. coli strains were grown at 37 °C in L-broth (Kahn et al., 1979 ) and on L-agar (L-broth solidified with 1·5%, w/v, agar) supplemented with appropriate antibiotics to ensure plasmid maintenance. P. fluorescens and its mutants were grown at 30 °C in L-broth, L-agar and mupirocin production medium (MPM) [yeast extract 2·3 g l-1, glucose 1·1 g l-1, Na2HPO4 2·6 g l-1, KH2PO4 2·4 g l-1, and (NH4)2SO4 5·0 g l-1]. For solid medium, agar was added to a concentration of 1·5%. For plasmid maintenance and selection of antibiotic-resistant transformants, the sterile media were supplemented with appropriate antibiotic concentration as follows: 100 µg ampicillin ml-1, 50 µg kanamycin ml-1, 150 and 300 µg penicillin G ml-1 in liquid and solid medium respectively, 600 µg mupirocin ml-1, and 25 µg tetracycline.HCl ml-1.

DNA isolation and manipulation.
Plasmid DNA extractions were performed by the alkaline SDS method of Birnboim & Doly (1979) with slight modifications (Smith & Thomas, 1983 ), as well as with Wizard Plus SV Mini Preps DNA Purification Systems (Promega). Chromosomal DNA was isolated using a genomic DNA purification kit (Nucleon, from Tepnel Life Sciences, Manchester, UK). DNA was digested by appropriate restriction enzymes (MBI Fermentas, Vilnius, Lithuania). DNA was analysed by standard agarose gel electophoresis (Sambrook et al., 1989 ). Extraction of the relevant DNA fragment from agarose gel was performed using the GeneClean kit (Bio101). To convert the sticky-ended DNA resulting from restriction enzyme digestion to blunt-ended, it was treated with DNA polymerase I, large (Klenow) fragment (New England Biolabs). Ligation was performed using T4 DNA ligase (Sambrook et al., 1989 ). Competent E. coli cells were transformed with plasmid DNA using the method of Cohen et al. (1972) .

Conjugative transfer of DNA.
Biparental mating was carried out to mobilize the suicide plasmid derivatives and expression vector pJH1 from E. coli S17-1 to P. fluorescens as follows. A mixture of late-exponential-phase cultures of E. coli containing the relevant plasmid (0·1 ml) and P. fluorescens strain (0·9 ml) were filtered on to a 0·45 µm sterile Millipore filter. The filter was placed on a dry L-agar plate overnight at room temperature. The mating mixture was suspended in 1 ml sterile saline solution, and 0·1 ml aliquots were spread on L-agar plates supplemented with appropriate antibiotics to select for the presence of the plasmid and mupirocin to kill E. coli donor bacteria. To test for the excision of the suicide plasmid from the chromosomal DNA, the cointegrate clone was suspended in 1 ml sterile saline solution and 10 µl aliquots of serial dilutions were plated on L-agar medium supplemented with 5% sucrose.

Construction of mupA::xylE reporter gene fusion.
The 760 bp EcoRI–BamHI fragment containing the 5' end of mupA (Fig. 1) and its putative promoter region was cloned into EcoRI and BamHI sites of pAKE100 suicide plasmid giving pAKE105. The promoterless xylE gene was taken as a BamHI fragment from pSRW41 (Warne, 1986 ) and subcloned into the BamHI site of pAKE105 to be downstream of the putative promoter region, giving plasmid pAKE106.



View larger version (9K):
[in this window]
[in a new window]
 
Fig. 1. Map of the mupA and mupR/mupI segment of the mup cluster. mupA is the first ORF in the mup operon. The 760 bp EcoRI–BamHI fragment in the mupA region possesses a putative promoter sequence for mupA and the mup operon (Fig. 2, A) and was used for the xylE reporter gene fusion strategy. The DNA fragments deleted from mupR and mupI to create the knockout mutants are the 430 bp BamHI–NcoI and 900 bp SphI fragments shown. mupX is the last ORF in the block of ORFs running in the same direction as mupA. Its role in mupirocin biosynthesis is not yet established. mupR is preceded by mupW.

 


View larger version (32K):
[in this window]
[in a new window]
 
Fig. 2. A, Nucleotide sequence analysis of the upstream region of mupA, the first ORF of the mupirocin biosynthesis operon. The putative ribosome-binding site is indicated by a bold line above the sequence. The putative -35 and -10 promoter sequences are underlined and in bold type. There are three inverted repeats identified. Inverted repeats I and III have similarity with lux boxes (Table 3). Inverted repeat II is shorter and has no similarity with the lux box. B, Nucleotide sequence of the upstream region and 5' end of mupI. The putative -10 and -35 promoter sequences are in bold. Two convergent arrows above the 18 bp inverted repeats indicate a possible MupR-binding site. The putative ribosome-binding site is indicated by a bold line above it.

 
Construction of deletions in mupR and mupI.
To make a 430 bp BamHI–NcoI deletion in mupR, a 990 bp PstI–EcoRI fragment that contained mupR (Fig. 1) was subcloned into pUC18, producing pAKE107. This was double digested with BamHI and NcoI followed by Klenow reaction to fill the sticky ends, and then recircularized by ligation, giving pAKE108. The {Delta}mupR region was transferred as an EcoRI–HindIII fragment into pAKE604. The resulting plasmid, pAKE109, was transferred into P. fluorescens by biparental mating. To inactivate mupI a 1706 bp SalI–PstI DNA fragment containing mupI (Fig. 1) and its upstream region was subcloned into pUC18 from which the SphI site in its polylinker had been removed. The resulting clone, pAKE110, was digested with SphI to remove a 900 bp segment that contains the 5' end and upstream region of mupI, giving pAKE111. The deleted mupI gene was transferred as an EcoRI–HindIII fragment into pAKE604, giving pAKE112, which was transferred to P. fluorescens as above.

Construction of a plasmid expressing mupI.
The mupI coding region was amplified by PCR with primers which placed an EcoRI site upstream of the start and a SalI site downstream of the stop codon. The product was inserted into pGEM-TEasy (Promega) and the sequence determined. It was then inserted into pOLE1 (Jagura-Burdzy et al., 1999 ), replacing the incC gene, and thus placing mupI under the control of the tac promoter controlled by the lacIq gene. Finally the tetAR segment conferring inducible tetracycline resistance from pDM6PP (Macartney, 1996 ), was inserted on a PstI fragment to allow easy selection in P. fluorescens, giving pJH1.

Sequence analysis.
DNA sequencing was carried out using the Big Dye Terminator kit (PE-ABI), which is based on the chain-termination method (Sanger et al., 1977 ). The sequence reactions were run on an ABI 377 automated DNA sequencer (Alta Bioscience, Birmingham University). The DNA sequences were analysed and aligned using GCG programs of the Wisconsin package (Devereux et al., 1984 ). The sequences were deposited with GenBank and have been assigned the following accession numbers: mupA, AF318063; mupR, AF318064; mupI, AF318065.

Polymerase chain reaction.
Standard PCR reactions were performed as described by Mullis et al. (1986) in order to test the resulting mupR and mupI deletion mutations as well as to confirm the presence of xylE within the chromosomal DNA of P. fluorescens after integration. The standard reaction involved 3 min at 94 °C followed by 20 cycles of 15 s at 94 °C, 15 s at 51 °C and 15 s at 72 °C, and finally 10 min at 72 °C. The primers used for the PCR reaction were synthesized by Alta Bioscience, Birmingham University. The {Delta}mupR primers were CCTGACTGGTTAGGCTA and AGGCAAATTGTGGCAGC, the {Delta}mupI primers were TTCAACAGCGATGGCTC and CTGCGATAACCAGTCGT, and the xylE primers were GGATCCACGTTGGCGGAAACAAACC and AAGCTTATCAGGTCAGCACGGTCATG.

Bioassay for mupirocin biosynthesis.
A 0·5 ml sample of an overnight culture of B. subtilis was added to 100 ml molten L-agar (<48 °C) and then 10–15 ml of this mixture was overlaid on a plate of mupirocin production medium previously inoculated with P. fluorescens. After 24 h the plate was checked for the presence of clear zones around the P. fluorescens patch.

XylE assay.
To detect the expression of the reporter gene xylE, agar plates with the bacterial colonies or streaks were sprayed with 0·3 mM catechol solution. Only clones which express xylE should turn yellow as a result of oxidation of catechol to 2-hydroxymuconic semialdehyde by catechol 2,3-oxygenase (XylE). To follow expression of the mup genes in a P. fluorescens derivative with xylE inserted into mupA, overnight cultures were diluted 50-fold into mupirocin liquid production medium. A separate flask was inoculated for each sample to be taken. Growth was followed for 120 h at 30 °C with shaking at 250 r.p.m. Three flasks were taken at each time interval for each strain. The optical density was measured at 600 nm, the cultures were harvested by centrifugation, and the pellet resuspended in 0·5 ml of sonication buffer before quantitative measurement of enzyme activity as described by Zukowski et al. (1983) . Protein concentration was determined as described by Gornall et al. (1949) .

Extraction of signalling molecule.
Supernatants (400 ml) from stationary-phase bacterial cultures grown in MPM for 4 d were extracted with dichloromethane (7:3 supernantant/dichloromethane) as described by McClean et al. (1997) . The dichloromethane was removed by rotary evaporation and the extract was resuspended in 2 ml acetonitrile. For bioassays the acetonitrile was evaporated and the residues resuspended in water or medium before use.

Mupirocin biosynthesis inducer reporter plate bioassay.
An aliquot of supernatant extract was spotted onto a streak of AKE5 (mupA::xylE {Delta}mupI) which had been grown for 24 h, and incubated for a further 24 h. The plates were then sprayed with catechol (1%) and examined for the development of a yellow colour through activation of the mupirocin promoter region and XylE expression.

LasRI bioluminescence reporter assay.
Plasmid pSB1075 developed by Winson et al. (1998) encodes a reporter that is preferentially activated by longer-chain AHLs (C10 to C14). This plasmid vector uses a lasRI'::luxCDABE reporter gene fusion which responds to activation of LasR by AHL synthesized from LasI homologues. Induction of bioluminescence in E. coli JM109 carrying pSB1075 was carried out essentially as described by Winson et al. (1998) ; it was detected using an Anthos Lucy1 luminometer and analysed with Stingray Version 1.1 (Dazdaq).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Location and sequence analysis of mupA, mupR and mupI ORFs within the mupirocin biosynthetic cluster
We have cloned and sequenced 84 kb of chromosomal DNA of P. fluorescens carrying genes for the mupirocin biosynthetic pathway (A. K. El-Sayed & C. M. Thomas, unpublished data). The mupirocin gene cluster begins with an ORF that we have named mupA (Fig. 1), whose predicted product has sequence similarity to two proteins involved in light emission: Y4VJ of the terrestrial bacterium Xenorhabdus luminescens (Szittner & Meighen, 1990 ) showed 27% identity, and LuxA of the marine bacterium Vibrio harveyi (Cohn et al., 1985 ; Johnston et al., 1996 ) showed 24·9% identity. Putative -10 and -35 promoter sequences were identified 141 bp upstream of the mupA ORF (Fig. 2).


View this table:
[in this window]
[in a new window]
 
Table 3. Comparison of the possible lux box sequences upstream of the mupA gene with previously identified lux boxes

 

Two divergent ORFs were identified close to the opposite end of the mup cluster (Fig. 1). We tentatively named them mupR and mupI because alignment of the predicted amino acid sequences of their products with the gene bank entries showed significant identity to proteins that are involved in quorum-sensing regulation. mupR encodes a predicted protein of 234 amino acids and molecular mass 26 kDa. MupR showed high identity (41%) to LasR of P. aeruginosa (Gambello & Iglewski, 1991 ) and 31% identity to LuxR of V. fischeri (Devine et al., 1988 ; Gray & Greenberg, 1992 ). The predicted product of mupI has 191 amino acids and a molecular mass of 21 kDa. MupI showed high identity to many proteins involved in producing AHL autoinducers. The best identities were 53·8% to LasI of P. aeruginosa (Passador et al., 1993 ; Pearson et al., 1994 ), 34% to LuxI of V. fischeri ATCC 7744 (Devine et al., 1988 ) and 36·6% to TraI, involved in conjugal transfer of Ti plasmids of Agrobacterium tumefaciens (Hwang et al., 1994 ; Fuqua & Winans, 1994 ).

In the lux system, the LuxR protein appears to be composed of two domains (Choi & Greenberg, 1991 ; Hanzelka & Greenberg, 1995 ). The C-terminal domain contains a putative helix–turn–helix motif, which is thought to bind with the lux promoter target and activate the lux operon. This helix–turn–helix motif is also conserved in MupR and LasR.

A 20 bp inverted repeat termed the lux box located in the luxI promoter region is required for binding the active LuxR to activate the lux operon transcription (Devine et al., 1988 ). The putative promoter region upstream of mupA possesses three inverted repeats (Fig. 2, A): the first and third have significant sequence identity to the lux box of other quorum-sensing systems as shown in Table 3, while the second inverted repeat does not. We found a 20 bp inverted repeat 28 bp upstream of the start codon of mupI gene (Fig. 2, B). This inverted repeat might act as a MupR-binding site but does not show sequence identity to either the lux box DNA sequence or the putative lux box inverted repeats upstream of the mupA promoter.

Integration of the xylE reporter gene downstream of the putative promoter of the mup operon
Integration of xylE downstream of the putative promoter region of the mup operon was achieved using a suicide vector pAKE100 (Fig. 3) It is based on cloning the target chromosomal DNA fragment into a plasmid possessing oriT, which enables its transfer from the donor E. coli to the recipient cell P. fluorescens where it cannot replicate. The plasmid, pAKE106, with xylE in the correct orientation downstream of mupAp was constructed and integrated into chromosomal DNA of P. fluorescens as described in Methods. The cointegrate [P. fluorescens AKE1 (mupA::xylE)] was selected on L-agar with kanamycin and mupirocin. After spraying with catechol, the colonies turned yellow, indicating that the xylE reporter gene is integrated downstream of an active promoter region. PCR was also used to demonstrate the integration of xylE within the chromosomal DNA of P. fluorescens.



View larger version (8K):
[in this window]
[in a new window]
 
Fig. 3. Schematic structure of the two suicide vectors used for manipulation of the mup genes (not strictly to scale). oriT is the transfer origin from RK2/RP4 which allows mobilization from strain S17-1. pMB1 ori is the replicon that allows high copy number maintenance in E. coli. lacZ{alpha} is the coding region that allows blue/white selection in appropriate hosts. mcs, multiple cloning site. KmR and PnR are the regions conferring resistance to kanamycin and penicillin, respectively. sacB is a levansucrase-encoding gene from B. subtilis.

 
{Delta}mupR and {Delta}mupI mutants show decreased mupirocin biosynthesis and transcription of the mupA gene
Deletion mutations of the mupR and mupI ORFs were constructed to determine whether they are involved in mupirocin biosynthesis and its regulation. For this strategy, a second suicide vector, pAKE604 (Fig. 3), was used since it also possesses sacB, which kills Gram-negative bacteria on sucrose-supplemented media due to periplasmic production of levan (Gay et al., 1985 ; Steinmetz et al., 1985 ). This allows one to select the products of a second recombination event, leading to the excision of the suicide plasmid from the chromosome, which in theory should leave a deletion behind at the integration site in 50% of excision events.

A plasmid with a deletion in mupR, pAKE109, was constructed and transferred into P. fluorescens as described in Methods. After purifying and checking integrants they were grown on sucrose agar. PCR was used to check whether DNA extracted from surviving clones carried the desired deletion: the first obtained was designated AKE2. A similar strategy was carried out to inactivate mupI, using a plasmid, pAKE112, constructed and used as described in Methods. Replacement of the wild-type chromosomal segment by the mutant gave a deletion mutant, P. fluorescens AKE3, whose genotype was confirmed by PCR.

Bioassay for mupirocin production was performed using the mupirocin-sensitive bacterium B. subtilis 1604, as described in Methods. Antimicrobial activity was not detected from AKE2 or AKE3, indicating that mupR and mupI are important for biosynthesis of mupirocin. To test this further, pAKE106, carrying the mupA::xylE fusion, was integrated into the first gene of the mup operon within the chromosomal DNA of {Delta}mupR and {Delta}mupI mutants, producing P. fluorescens AKE4 (mupA::xylE {Delta}mupR) and P. fluorescens AKE5 (mupA::xylE {Delta}mupI). These cointegrates did not show any activity for the promoter/xylE fusion when sprayed with catechol and compared to the wild-type P. fluorescens AKE1 construct.

The MupI- phenotype is overcome by a diffusible product of mupI+ bacteria
As described above, we predicted that mupI encodes the ability to produce a diffusible inducer. To test this hypothesis, P. fluorescens NCIMB 10586 and P. fluorescens AKE5 (mupA::xylE {Delta}mupI) were streaked together as two separate lines on mupirocin production agar. After 24 h the plate was sprayed with catechol. As predicted, P. fluorescens AKE5 turned yellow. However, when this plate test was repeated with P. fluorescens AKE3 ({Delta}mupI) in place of NCIMB 10586 there was no restoration of the xylE expression of P. fluorescens AKE5. To confirm that this defect was due to knocking out mupI we amplified the mupI ORF and inserted it into a broad-host-range IncQ expression vector downstream of the tac promoter as described in Methods. We first carried out simple complementation experiments: AKE3(pJH1) grown on MPM agar with 1 mM IPTG (to induce mupI expression) for 24 h showed antimicrobial activity while AKE5(pJH5) turned bright yellow, indicating reactivation of mup operon expression. AKE3(pJH1) also switched on antibiotic synthesis in AKE3 and activated xylE expression in AKE5 when grown alongside them. Finally, pJH1 in E. coli NEM259 directed production of a diffusible activator that could switch on antibiotic synthesis in AKE3 and xylE expression in AKE5. This shows clearly that mupI is involved in producing a diffusible substance released from P. fluorescens that can restore the activity of the mup promoter in the {Delta}mupI mutant.

Although the P. fluorescens AKE2 mutant contains a complete copy of mupI, it did not restore the activity of {Delta}mupI when it was streaked with P. fluorescens AKE5 on the same plate. Therefore, the mupR region is needed to allow mupI to generate the diffusible activator. The same experiment was carried out with the pair P. fluorescens NCIMB 10586 and P. fluorescens AKE4. As expected this did not show any activity for xylE, indicating that the defect in mupR cannot be suppressed by a diffusible product of the wild-type strain.

Time course of xylE expression for P. fluorescens AKE1
To determine the effect of the growth on the regulation of the mupirocin biosynthetic genes, XylE was assayed periodically in cultures of P. fluorescens AKE1. The same experiment was also carried out for a mixture of wild-type P. fluorescens plus P. fluorescens AKE5, to establish the role of the mupI-dependent diffusible product in the mupirocin biosynthesis and at which growth phase and what level it is produced. First, P. fluorescens AKE1 was grown for 120 h to study the effect of the growth phase on the promoter activity (Fig. 4). The XylE activity started low but then decreased further after 5 h, presumably due to decay of existing activity in the absence of new synthesis. The XylE activity started to increase dramatically during entry into stationary phase. It then slightly decreased but increased again followed by a maximum activity at 30 h. By late stationary phase (120 h) the XylE enzyme activity had dropped again.



View larger version (17K):
[in this window]
[in a new window]
 
Fig. 4. Time course of XylE production by P. fluorescens AKE1 (mupA::xylE). The growth phase and XylE activity were followed for 120 h as described in Methods. Bacterial growth was measured as the OD600 ({diamondsuit}). XylE activity ({blacksquare}) was calculated as units, where one unit of catechol 2,3-oxygenase is defined as the amount required to convert 1 mmol substrate to product in 1 min under standard conditions (Zukowski et al., 1983 ). Each time point represents data from three parallel cultures, the error bars showing the standard deviation.

 
The same experiment was carried out for the P. fluorescens AKE5 and P. fluorescens NCIMB 10586 mixture, except that a 250 µl overnight culture inoculum was taken from each strain and added to 25 ml mupirocin production liquid medium. The growth phase and XylE enzyme activity curves had the same patterns as in the previous experiment, but the enzyme activity was lower by a factor of about four (Fig. 5).



View larger version (15K):
[in this window]
[in a new window]
 
Fig. 5. Time course of XylE production by an equal mixture of P. fluorescens AKE5 construct and P. fluorescens wild-type cultures. Growth ({diamondsuit}) and XylE activity ({blacksquare}) were measured as in Fig. 4 and as described in Methods.

 
Characterization of the signalling molecule
To further characterize the diffusible activator we determined whether we could activate available AHL-dependent reporter strains either by growth next to NCIMB 10586 or by addition of concentrated extracts of spent growth medium after NCIMB 10586 cultures had been grown to stationary phase. To determine whether the extraction procedure described in Methods concentrated an active signalling molecule we resuspended the dried extract in sterile distilled water and spotted it onto a streak of AKE5 (mupA::xylE {Delta}mupI) which had been grown for 24 h, and incubated for a further 24 h. When sprayed with catechol (1%) the plates rapidly developed the yellow colour characteristic of XylE activity, which indicates activation of the mupirocin promoter.

The first reporter strains subsequently used were Chromobacterium violaceum ACV026 and ATTC 31532. CV026 is a violacein-negative strain, defective in AHL production, that responds to exogenous AHLs with N-acyl side chains from C4 to C8 in length by the production of the purple pigment violacein (McClean et al., 1997 ). Longer-chain AHLs (C10 to C14) are not only unable to induce violacein production from CV026, but they can also positively interfere with the action of the normal inducer, N-acylhomocysteine thiolactone, in the wild-type strain C. violaceum ATCC 31532. P. fluorescens NCIMB 10586 was streaked beside CV026 and ATCC 31532, as described by Swift et al. (1997) but neither induction nor inhibition of violacein production was detected. Culture supernatant extract from NCIMB 10586 was spotted onto the reporter strains which had been grown for 24 h, and then incubated for a further 24 h, but again no activity was detected. The positive control, P. aeruginosa PAO1161, inhibited violacein production from ATCC 31532 and induced production from CV026.

The second reporter system employed was E. coli carrying a bioluminescence reporter plasmid (pSB1075) developed by Winson et al. (1998) based on the P. aeruginosa las system, which is preferentially activated by longer-chain AHLs (C10 to C14). This plasmid uses a lasRI'::luxCDABE reporter gene fusion which responds to activation of LasR by AHL synthesized from LasI homologues. Bioluminescence induction assays were performed with extracts of NCIMB 10586, AKE3 ({Delta}mupI) and P. aeruginosa. The latter should contain cognate AHL for LasR induction (3-oxo-C12-HSL). Both the NCIMB 10586 extract and the positive control induced bioluminescence from the biosensor (Fig. 6). The extract from AKE3 ({Delta}mupI) induced a negligible level of bioluminescence. This suggests that activation of the LasR reporter by NCIMB 10586 is due to a signal molecule synthesized from MupI that is a longer-chain AHL similar to those synthesized by LasI. Finally therefore we tested extract from P. aeruginosa with the AKE5 (mupA::xylE {Delta}mupI) reporter strain and as predicted found that it would activate the mup promoter, suggesting that the MupR/MupI and the LasR/LasI systems are functionally closely related, as already indicated by their sequence.



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 6. Induction of bioluminescence in E. coli NEM259(pSB1075) by supernatant extracts from P. aeruginosa ({bullet}), P. fluorescens NCIMB 10586 ({blacksquare}) and P. fluorescens {Delta}mupI ({blacktriangleup}). The experiment was done twice, and for each experiment the data points were obtained from duplicate samples. Different extracts from the P. fluorescens strains were used in the two experiments to ensure that the difference between wild-type and mutant was reproducible. The results presented are from one of the experiments. The mean of each pair is plotted and the bars show the variation between the pairs.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In this paper we report that expression of the genes of the mupirocin biosynthetic cluster of P. fluorescens NCIMB 10586 depends on a quorum-sensing regulatory system. Precedents exist in related strains and species. For example, in Pseudomonas aureofaciens 30-84, expression of the phenazine antibiotic operon (phzFABCD) is controlled by PhzR and PhzI, which are homologues of the LuxR and LuxI family quorum-sensing regulators (Pearson et al., 1994 ; Pierson et al., 1995 ; Wood & Pearson, 1996 ). More recently Laue et al. (2000) identified a novel AHL in the biocontrol strain P. fluorescens F113.

The amino acid sequence alignment of MupR and MupI clearly showed similarity to the LasR/LasI and LuxR/LuxI regulatory systems, indicating that MupR and MupI belong to the same family of regulators. Also, the putative consensus helix–turn–helix motif in the C-terminal region of MupR showed strong similarity to LasR and LuxR motifs. This supports the prediction of its function in binding to the DNA promoter region of the mup operon and activation of its transcription. Although we found good candidates for lux boxes upstream of mupA there does not appear to be one upstream of mupI. However, this lack of a lux box is not exceptional. For example, lasI and traM, which are strongly regulated by LasR and TraR respectively (Passador et al., 1993 ; Hwang et al., 1994 ), do not possess a homologous lux box in their promoter regions. Therefore, although the 20 bp inverted repeat upstream of the putative promoter region of mupI is not similar to the lux box sequence, it might work as a target site for MupR binding, although it would not be expected to activate transcription from this position. By contrast with this uncertainty, the homology of the upstream inverted repeat of mupA to the lux box (Table 3) makes this an excellent candidate for binding of the MupR/MupI-product complex to activate mup operon transcription.

The importance of mupR and mupI ORFs within the mupirocin biosynthesis cluster is confirmed by the phenotype of deletion mutants of mupR and mupI, which prevent mupirocin production in a bioassay. The cross-feeding experiment, where the mupirocin producer P. fluorescens restored the activity of {Delta}mupI mutant, was as expected from the prediction, based on amino acid sequence, that MupI produces a diffusible substance. Complementation of the {Delta}mupI mutant by the cloned mupI gene, and also production of a diffusible activator by E. coli carrying the cloned mupI gene, confirmed that the activator really is synthesized by the product of this gene.

The nature of the product was further defined by its having no apparent activity as activator or antagonist of the C. violaceum system, although this result should not be taken completely at its face value. Failure to induce CV026 with another strain of P. fluorescens (F113) was also observed by Laue et al. (2000) . Although NCIMB 10586 may not produce an appropriate AHL to activate or inhibit violacein production, it could also be that the AHL is being produced or extracted below the threshold level for activation of the biosensor; or that additional compounds are also being extracted which inhibit activation. Laue et al. (2000) further purified culture supernatant extracts by HPLC and isolated fractions that were able to activate biosensors. They identified the AHLs as N-(3-hydroxytetradecenoyl)-L-homoserine lactone (3-OH-C14:1-HSL), N-decanoyl-L-homoserine lactone (C10-HSL) and N-hexanoyl-L-homoserine lactone (C6-HSL). We obtained more positive evidence for the nature of the mup activator from its cross-talk with the LasRI system, both activating the LasRI reporter and being substituted by a product of P. aeruginosa. This demonstrates that we have extracted an inducer for expression of the mupirocin cluster from culture supernatant, and suggests that it is similar to signal molecules produced by P. aeruginosa. P. aeruginosa produces two major AHLs, 3-oxo-C12-HSL and C4-HSL, and two minor AHLs, N-(3-oxohexanoyl)-L-homoserine lactone (3-oxo-C6-HSL) and C6-HSL (Pearson et al., 1994 ; Pierson et al., 1995 ; Winson et al., 1995 ). This suggests a longer-chain AHL, which would fit with the results of Laue et al. (2000) described above, and also with another report on a different strain, P. fluorescens 2-79, which produced N-(3-hydroxyhexanoyl)-L-homoserine lactone (3-OH-C6-HSL), N-(3-hydroxyoctanoyl)-L-homoserine lactone (3-OH-C8-HSL) and N-(3-hydroxydecanoyl)-L-homoserine lactone (3-OH-C10-HSL) (Shaw et al., 1997 ). Future work will define the exact nature of the MupI product.

In the lux regulatory family, the LuxR-like proteins not only positively regulate transcription of the biosynthetic genes, they also activate transcription of the luxI homologues (Fuqua et al., 1996 ). The active form of LuxR-like proteins is the result of association with AHL, which binds the N-terminal domain and stops it from binding with and blocking the C-terminal DNA-binding domain. The regulatory process starts with very low expression of luxI, thus producing only a very small amount of AHL, which activates some of LuxR. The LuxR/AHL activates the transcription of luxI, producing more molecules of AHL, which in turn works as a feedback for more LuxR activation (Nealson et al., 1970 ; Fuqua et al., 1994 ). The {Delta}mupR mutant, which contains a complete mupI gene, did not produce sufficient diffusible substance to restore the mupirocin production of the strain with a mupI deletion (P. fluorescens AKE5), indicating that MupR is essential for activating mupI as well as the mup operon. Since mupI alone under the control of the tac promoter in E. coli was able to produce detectable activator, MupR is unlikely to be needed for activator synthesis itself. Therefore the product of mupR is likely to function as an activator of mupI expression, consistent with the role of MupR homologues in other systems. Lack of boosted mupI expression would limit the whole regulatory process and thus expression of the mup operon. The only reservation in this interpretation of the {Delta}mupR mutant is its possible polar effect on mupX (the ORF following mupR; Fig. 1). Such an effect could arise since the mupR defect is not due to a simple in-frame deletion, but also removes the translational start of mupR, leaving a short untranslated section of mRNA between mupW (the ORF preceding mupR; Fig. 1) and mupX. However, the amount of untranslated RNA between the end of mupW and the start of mupX is short, and lacking in potential inverted repeats, so that the chances of transcriptional termination are slight; and in any case polar effects of this sort rarely result in complete loss of expression. Thus, even if mupX is essential for biosynthesis it should not be totally inactivated.

Quorum sensing depends mainly on accumulation of sufficient signal to induce gene expression and hence usually occurs at high cell density (Fuqua et al., 1994 ; Pierson et al., 1994 ). As shown from the XylE reporter gene assay experiments, the mup operon expression reaches the maximum when the growing cells enter stationary phase. Although the XylE assay of the P. fluorescens AKE5 and P. fluorescens wild-type mixture gave an expression profile similar to that of P. fluorescens AKE1, the enzymic XylE activity was approximately fourfold lower. This may be due to the dilution factor since half the volume of the inoculum was taken from P. fluorescens AKE5 followed by another half volume of bacterial cell extract during XylE assay manipulation. We tried to switch on mup expression during exponential phase by supplementing cultures with the extract from stationary-phase cultures, but no enhancement was observed (J.H., unpublished). This may suggest that mup expression is actively repressed during exponential phase.

Many questions remain concerning the role of quorum sensing in regulation of mupirocin biosynthesis in P. fluorescens. Additional studies are being carried out to map the putative promoter region of the mup operon as well as mupR and mupI. Also it is important to find out whether mupR/mupI are the only regulators required for mupirocin biosynthesis or if there is another level of regulation, especially as hierarchical regulation is one of the common features of quorum-sensing systems. The answers to these questions may help both to manipulate expression and to point to the role of mupirocin in the growth and ecology of P. fluorescens.


   ACKNOWLEDGEMENTS
 
This work was in part funded by a project grant from the Biotechnology and Biological Sciences Research Council (Grant 6/C11041). A.K.E.S. was supported by an Egyptian Government Scholarship. Automated DNA sequencing was carried out by AltaBioscience in the School of Biosciences using facilities in part funded by The Wellcome Trust. Mupirocin was a gift from SmithKlineBeecham.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bever, R. A. & Iglewski, B. H. (1988). Molecular characterization and nucleotide sequence of the Pseudomonas aeruginosa elastase structural gene. J Bacteriol 170, 4309-4314.[Medline]

Birnboim, H. C. & Doly, J. (1979). A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res 7, 1513-1523.[Abstract]

Choi, S. H. & Greenberg, E. P. (1991). The C-terminal region of the Vibrio fischeri LuxR contains an inducer-dependent lux gene activating domain. Proc Natl Acad Sci USA 88, 11115-11119.[Abstract]

Cohen, S. N., Chang, A. C. Y. & Hsu, H. (1972). Non-chromosomal antibiotic resistance in bacteria; genetic transformation of Escherichia coli by R factor DNA. Proc Natl Acad Sci USA 69, 2110-2114.[Abstract]

Cohn, D. H., Mileham, A. J., Simon, M. I., Nealson, K. H., Rausch, S. K., Bonam, D. & Baldwin, T. O. (1985). Nucleotide sequence of the luxA gene of Vibrio harveyi and the complete amino acid sequence of the {alpha} subunit of bacterial luciferase. J Biol Chem 260, 6139-6146.[Abstract/Free Full Text]

DeKievit, T., Seed, P. C., Nezezon, J., Passador, L. & Iglewski, B. H. (1999). RsaL, a novel repressor of virulence gene expression in Pseudomonas aeruginosa. J Bacteriol 181, 2175-2184.[Abstract/Free Full Text]

Devine, J. H., Countryman, C. & Baldwin, T. O. (1988). Nucleotide sequence of the luxR and luxI genes and structure of the primary regulatory region of the lux regulon of Vibrio fischeri ATCC7744. Biochemistry 27, 837-842.

Devine, J. H., Shadel, G. S. & Baldwin, T. O. (1989). Identification of the operator of the lux regulon from Vibrio fischeri ATCC7744. Proc Natl Acad Sci USA 86, 5688-5692.[Abstract]

Devereux, J., Haeberli, P. & Smithies, O. (1984). A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res 12, 387-395.[Abstract]

Dunn, D. K., Michaliszyn, G. A., Bogaki, I. B. & Meighen, E. A. (1973). Conversion of aldehyde to acid into the bacterial bioluminescent reaction. Biochemistry 12, 4911-4918.[Medline]

Flavier, A. B., Ganova-Raeva, L. M., Schell, M. A. & Denny, T. P. (1997). Hierarchical autoinduction in Ralstonia solanacearum: control of acyl-homoserine lactone production by a novel autoregulatory system responsive to 3-hydroxypalmitic acid methyl ester. J Bacteriol 179, 7089-7097.[Abstract]

Fuqua, W. C. & Winans, S. C. (1994). A LuxR–LuxI type regulatory system activates Agrobacterium Ti plasmid conjugal transfer in the presence of a plant tumour metabolite. J Bacteriol 176, 2796-2806.[Abstract]

Fuqua, W. C., Winans, S. C. & Greenberg, E. P. (1994). Quorum-sensing in bacteria: the LuxR–LuxI family of cell density-responsive transcriptional regulators. J Bacteriol 176, 269-275.[Medline]

Fuqua, W. C., Winans, S. C. & Greenberg, E. P. (1996). Census and consensus in bacterial ecosystems: the LuxR–LuxI family of quorum-sensing transcriptional regulators. Annu Rev Microbiol 50, 727-751.[Medline]

Gambello, M. J., Kaye, S. & Iglewski, B. H. (1991). Cloning and characterization of the Pseudomonas aeruginosa lasR gene, a transcriptional activator of elastase expression. J Bacteriol 173, 3000-3009.[Medline]

Gambello, M. J., Kaye, S. & Iglewski, B. H. (1993). LasR of Pseudomonas aeruginosa is a transcriptional activator of the alkaline protease gene (apr) and an enhancer of exotoxin A expression. Infect Immum 61, 1180-1184.[Abstract]

Gay, P., Le Coq, D., Steinmetz, M., Berkelman, T. & Hado, C. I. (1985). Positive selection procedure for the entrapment of insertion sequence elements in Gram-negative bacteria. J Bacteriol 164, 918-921.[Medline]

Gibson, T. J. (1984). Studies on the Epstein–Barr virus genome. PhD thesis, Cambridge University, UK.

Gornall, A. G., Bardawill, C. J. & David, M. M. (1949). Determination of serum proteins by means of the biuret reaction. J Biol Chem 177, 751-766.[Free Full Text]

Gray, K. M. & Greenberg, E. P. (1992). Sequence analysis of luxR and luxI, the luminescence regulatory genes from the squid light organ symbiont Vibrio fischeri ES114. Mol Mar Biol Biotechnol 1, 414-419.

Greenberg, E. P. (2000). Acyl-homoserine lactone quorum sensing in bacteria. J Microbiol 38, 117-121.

Hanzelka, B. L. & Greenberg, E. P. (1995). Evidence that the N-terminal region of the Vibrio fischeri LuxR constitutes an autoinducer-binding domain. J Bacteriol 177, 815-817.[Abstract]

Hughes, J. & Mellows, G. (1978). Inhibition of isoleucyl-transfer ribonucleic acid synthase in Escherichia coli by pseudomonic acid. Biochem J 176, 305-318.[Medline]

Hughes, J. & Mellows, G. (1980). Interaction of pseudomonic acid A with Escherichia coli B isoleucyl-tRNA synthase. Biochem J 191, 209-219.[Medline]

Hwang, I., Li, P.-L., Zhang, L., Piper, K. R., Cook, D. M., Tate, M. E. & Farrand, S. K. (1994). TraI, a LuxI homologue, is responsible for production of conjugation factor, the Ti plasmid N-acylhomoserine lactone autoinducer. Proc Natl Acad Sci USA 91, 4639-4643.[Abstract]

Jagura-Burdzy, G., Kostellidou, K., Pole, J., Khare, D., Jones, A. C., Williams, D. R. & Thomas, C. M. (1999). IncC of broad-host-range plasmid RK2 modulates KorB transcriptional repressor activity in vivo and operator binding in vitro. J Bacteriol 181, 2807-2815.[Abstract/Free Full Text]

Johnston, T. C., Thompson, R. B. & Baldwin, T. O. (1986). Nucleotide sequence of luxB gene of Vibrio harveyi and the complete amino acid sequence of the ß subunit of the bacterial luciferase. J Biol Chem 261, 4805-4811.[Abstract/Free Full Text]

Kahn, M., Kolter, R., Thomas, C. M., Figurski, D. H., Meyer, R., Remeut, E. & Helinski, D. R. (1979). Plasmid cloning vehicles derived from plasmids ColE1 and RK2. Methods Enzymol 68, 268-280.[Medline]

Kaiser, D. & Losick, R. (1993). How and why bacteria talk to each other. Cell 73, 873-885.[Medline]

Kaplan, H. B. & Greenberg, E. P. (1985). Diffusion of autoinducer is involved in regulation of Vibrio fischeri luminescence system. J Bacteriol 163, 1210-1214.[Medline]

Kuo, A., Blough, N. V. & Dunlap, P. V. (1994). Multiple N-acyl-L-homoserine lactone autoinducers of luminescence in the marine symbiotic bacterium Vibrio fischeri. J Bacteriol 176, 7558-7565.[Abstract]

Latifi, A., Winson, M. K., Foglino, M., Bycroft, B. W., Stewart, G. S. A. B., Lazdunski, A. & Williams, P. (1995). Multiple homologues of LuxR and LuxI control expression of virulence determinants and secondary metabolites through quorum sensing in Pseudomonas aeruginosa PAO1. Mol Microbiol 17, 333-343.[Medline]

Latifi, A., Foglino, M., Tanaka, K., Williams, P. & Lazdunski, A. (1996). A hierarchical quorum-sensing cascade in Pseudomonas aeruginosa links the transcriptional activators LasR and RhlR (VsmR) to expression of the stationary-phase sigma factor RpoS. Mol Microbiol 21, 1137-1146.[Medline]

Laue, B. E., Jiang, Y., Chhabra, S. R., Jacob, S., Stewart, G. S. A. B., Hardman, A., Downie, J. A., O’Gara, F. & Williams, P. (2000). The biocontrol strain Pseudomonas fluorescens F113 produces the Rhizobium small bacteriocin, N-(3-hydroxy-7-cis-tetradecanoyl)homoserine lactone, via HdtS, a putative novel N-acylhomoserine lactone synthase. Microbiology 146, 2469-2480.[Abstract/Free Full Text]

Lewenza, S., Conway, B., Greenberg, E. P. & Sokol, P. A. (1999). Quorum sensing in Burkholderia cepacia: identification of the LuxRI homologue CepRI. J Bacteriol 181, 748-756.[Abstract/Free Full Text]

Macartney, P. (1996). Molecular and genetic analyses of the central control regions of broad host range bacterial plasmids RK2 and R751. PhD thesis, University of Birmingham.

McClean, K. H., Winson, M. K., Fish, L. & 9 other authors (1997). Quorum sensing and Chromobacterium violaceum: exploitation of violacein production and inhibition for the detection of N-acylhomoserine lactones. Microbiology 143, 3703–3711.[Abstract]

Martin, M., Showalter, R. & Silverman, M. (1989). Identification of the luminescence genes in Vibrio harveyi. J Bacteriol 171, 2406-2414.[Medline]

Meighen, E. A. (1988). Enzymes and genes from the lux operon of bioluminescent bacteria. Annu Rev Microbiol 42, 151-176.

Moir, A., Lafferty, E. & Smith, D. A. (1979). Genetic analysis of spore germination mutants of Bacillus subtilis 168: the correlation of phenotype with map location. J Gen Microbiol 111, 165-168.[Medline]

Mullis, K., Faloona, F., Scharf, S., Saiki, R., Horn, G. & Erlich, H. (1986). Specific enzymic amplification of DNA in vitro: the polymerase chain reaction. Cold Spring Harbor Symp Quant Biol 51, 263–273.[Medline]

Nealson, K. H., Platt, T. & Hastings, J. W. (1970). Cellular control of the synthesis and activity of the bacterial luminescence system. J Bacteriol 104, 313-322.[Medline]

Ochsner, U. S., Koch, A. K. & Reiser, J. (1994). Isolation and characterization of regulatory gene affecting rhamnolipid biosurfactant synthesis in Pseudomonas aeruginosa. J Bacteriol 176, 2044-2054.[Abstract]

Parsek, M. R., Schaefer, A. L. & Greenberg, E. P. (1997). Analysis of random and site-directed mutations in rhlI, a Pseudomonas aeruginosa gene encoding an acylhomoserine lactone synthase. Mol Microbiol 26, 301-310.[Medline]

Passador, L., Cook, J. M., Gambello, M. J., Rust, L. & Iglewski, B. H. (1993). Expression of Pseudomonas aeruginosa virulence genes requires cell-to-cell communication. Science 260, 1127-1130.[Medline]

Pearson, J. P., Gray, K., Passador, L., Tucker, K. D., Eberhard, A., Iglewski, B. H. & Greenberg, E. P. (1994). Structure of the autoinducer required for expression of Pseudomonas aeruginosa virulence genes. Proc Natl Acad Sci USA 91, 197-201.[Abstract]

Pearson, J. P., Pesci, E. C. & Iglewski, B. H. (1997). Roles of Pseudomonas aeruginosa las and rhl quorum-sensing systems in the control of elastase and rhamnolipid biosynthesis genes. J Bacteriol 179, 5756-5767.[Abstract]

Pierson, L. S., Keppenne, V. D. & Wood, D. W. (1994). Phenazine antibiotic biosynthesis in Pseudomonas aureofaciens 30-84 is regulated by PhzR in response to cell density. J Bacteriol 176, 3966-3974.[Abstract]

Pierson, L. S., Gaffney, T., Lam, S. & Gong, F. (1995). Molecular analysis of genes encoding phenazine biosynthesis in the biological control bacterium Pseudomonas aureofaciens 30-84. FEMS Microbiol Lett 134, 299-307.[Medline]

Rode, H., deWet, P. M., Millar, A. J. W. & Cywes, S. (1991). Mupirocin resistance. Lancet 338, 578.

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edition. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Sanger, F., Nicklen, S. & Coulson, A. R. (1977). DNA-sequencing with chain terminating inhibitors. Proc Natl Acad Sci USA 74, 4563-4567.

Schweizer, H. P. & Hoang, T. T. (1995). An improved system for gene replacement and xylE fusion analysis in Pseudomonas aeruginosa. Gene 158, 15-22.[Medline]

Seed, P. C., Passador, L. & Iglewski, B. H. (1995). Activation of Pseudomonas aeruginosa lasI gene by LasR and Pseudomonas autoinducer PAI: an autoinduction regulatory hierarchy. J Bacteriol 177, 654-659.[Abstract]

Shaw, P. D., Ping, G., Daly, S. L., Cha, C., Cronan, J. E., Rinehart, K. L. & Farrand, S. K. (1997). Detecting and characterizing N-acyl-homoserine lactone signal molecules by thin-layer chromatography. Proc Natl Acad Sci USA 94, 6036-6042.[Abstract/Free Full Text]

Simon, R., Priefer, U. & Puhler, A. (1983). A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram-negative bacteria. Biotechnology 1, 784-791.

Smith, C. A. & Thomas, C. M. (1983). Deletion mapping of kil and kor functions in the trfA and trbB regions of broad-host-range plasmid RK2. Mol Gen Genet 190, 245-254.[Medline]

Steinmetz, M., Le Coq, D., Aymerich, S., Gonzy-Treboul, G. & Gay, P. (1985). The DNA sequence of the gene for the secreted Bacillus subtilis enzyme levansucrase and its genetic control sites. Mol Gen Genet 200, 220-228.[Medline]

Swift, S., Karlyshev, A. V., Fish, L., Durant, E. L., Winson, M. K., Chhabra, S. R., Williams, P., MacIntyre, S. & Stewart, G. S. A. B. (1997). Quorum sensing in Aeromonas hydrophila and Aeromonas salmonicida: identification of the LuxRI homologues AhyRI and AsaRI and their cognate N-acylhomoserine lactone signal molecules. J Bacteriol 179, 5271-5281.[Abstract]

Szittner, R. & Meighen, E. (1990). Nucleotide sequence, expression and properties of luciferase coded by lux genes from a terrestrial bacterium. J Biol Chem 265, 16581-16587.[Abstract/Free Full Text]

Thomas, C. M., Hussain, A. A. K. & Smith, C. A. (1982). Maintenance of broad host range plasmid RK2 replicons in Pseudomonas aeruginosa. Nature 298, 674-676.[Medline]

Warne, S. (1986). Studies of gene expression signals suitable for the regulation of foreign proteins in Escherichia coli and Pseudomonas species. PhD thesis, Department of Biological Sciences, University of Birmingham, UK.

Whatling, C. (1993). Molecular analysis of the genes involved in the biosynthesis of pseudomonic acid, a polyketide antibiotic produced by Pseudomonas fluorescens NCIMB10586. PhD thesis, Department of Biological Sciences, University of Birmingham, UK.

Whatling, C. A., Hodgson, J. E., Burnham, M. K., Clarke, N. J., Franklin, F. C. H. & Thomas, C. M. (1995). Identification of a 60 kb region of the chromosome of Pseudomonas fluorescens 10586 required for the biosynthesis of pseudomonic acid (mupirocin). Microbiology 141, 973-982.

Winson, M. K., Camara, M., Latifi, A. & 10 other authors (1995). Multiple N-acyl-L-homserine lactone signal molecules regulate production of virulence determinants and secondary metabolites in Pseudomonas aeruginosa. Proc Natl Acad Sci USA 92, 9427–9431.

Winson, M. K., Swift, S., Fish, L., Throup, J. P., Jorgensen, F., Chhabra, S. R., Bycroft, B. W., Williams, P. & Stewart, G. S. A. B. (1998). Construction and analysis of luxCDABE-based plasmid sensors for investigating N-acyl homoserine lactone-mediated quorum sensing. FEMS Microbiol Lett 163, 185-192.[Medline]

Wood, D. W. & Pierson, L. S. (1996). The phzI gene of Pseudomonas aureofaciens 30-84 is responsible for the production of a diffusible signal required for phenazine antibiotic production. Gene 168, 49-53.[Medline]

Yanisch-Perron, C., Viera, J. & Messing, J. (1985). Improved M13 phage cloning vectors and host strains: nucleotide sequence of M13mp18 and pUC19 vectors. Gene 33, 103-119.[Medline]

Zukowski, M. M., Gaffney, D. F., Speck, D., Kauffman, M., Findelli, A., Wisecup, A. & Lecocq, J. P. (1983). Chromogenic identification of genetic regulatory signals in Bacillus subtilis based on expression of a cloned Pseudomonas gene. Proc Natl Acad Sci USA 80, 1101-1105.[Abstract]

Received 14 November 2000; revised 6 April 2001; accepted 25 April 2001.