Detection and mutation of a luxS-encoded autoinducer in Proteus mirabilis

Rachel Schneider1, C. Virginia Lockatell2,3, David Johnson2,3 and Robert Belas1

Center of Marine Biotechnology, University of Maryland Biotechnology Institute, 701 East Pratt Street, Baltimore, MD 21202, USA1
University of Maryland School of Medicine2 and Research Service3, Veterans Administration Medical Center, Baltimore, MD 21201, USA

Author for correspondence: Robert Belas. Tel: +1 410 234 8876. Fax: +1 410 234 8896. e-mail: belas{at}umbi.umd.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Quorum sensing regulates the expression of virulence factors in a wide variety of pathogenic bacteria. This study has shown that Proteus mirabilis harbours a homologue of luxS, a gene required for the synthesis of the quorum sensing autoinducer 2 (AI-2). AI-2 activity is expressed during and is correlated with the initiation of swarming migration on agar surfaces. The P. mirabilis luxS locus was cloned and a LuxS- strain constructed by allelic-exchange mutagenesis. While lacking AI-2 activity, a null mutation in luxS, however, did not affect swimming or swarming motility, swarmer cell differentiation, or virulence in a mouse model of ascending urinary tract infection.

Keywords: quorum sensing, LuxS, swarming, virulence, urinary tract infection

Abbreviations: AHL, N-acylhomoserine lactone; AI, autoinducer; UTI, urinary tract infection

The GenBank accession number for the sequence reported in this paper is AY044337.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Urinary tract infections (UTIs) are among the most common bacterial infections of any organ system, accounting for an estimated 8 million visits to physicians each year (Mobley & Warren, 1996 ). Of the bacteria responsible for causing uncomplicated cystitis or acute pyelonephritis, Proteus mirabilis represents only a small percentage of cases, but it is much more frequently isolated from patients with complicated urinary tracts, i.e. those with functional or anatomical abnormalities or with chronic instrumentation (Mobley & Chippendale, 1990 ). In these cases, P. mirabilis is a more frequent uropathogen and its infection may be more severe due to recurring infection and the production of urinary stones in the kidney(s). Struvite (magnesium ammonium phosphate) and apatite (calcium phosphate) stones, formed as a result of the breakdown of urea by P. mirabilis urease, are commonly found in the renal pelvis. Stone formation makes P. mirabilis important in complicated UTIs, especially since P. mirabilis infections appear preferentially localized to the kidneys (Johnson et al., 1993b ).

Stickler and his collaborators (McLean et al., 1997 ; Stickler & Hughes, 1999 ; Stickler et al., 1998 ) have demonstrated the importance of P. mirabilis biofilms in the formation of struvite crystals during UTI. When grown on nutrient agar, P. mirabilis also forms a similar biofilm (or bacterial colony) with a unique pattern of development. These colonies frequently develop a concentric bull’s-eye pattern (Fig. 1), underscoring the cyclic nature of this motile behaviour. As shown in Fig. 1, each cycle of biofilm development may be divided into four parts: (i) swarmer cell differentiation, (ii) the lag period prior to active movement, (iii) swarming colony migration, and (iv) consolidation (where the cells stop moving and de-differentiate to swimmer cell morphology). The bull’s-eye pattern is itself mirrored in the activity of a set of proteins that are also expressed coordinately with the cycles of swarming. Included in this group of swarmer-cell-dependent proteins are a set of virulence factors, including flagellin, the ZapA protease, urease and haemolysin (Allison et al., 1992 ). It has been postulated, based in part on the evidence of co-ordinate expression of virulence factors during cellular differentiation, that the swarmer cell and swarming behaviour may be involved in UTI and pathogenesis (Allison et al., 1992 , 1993 ; Chippendale et al., 1994 ).



View larger version (38K):
[in this window]
[in a new window]
 
Fig. 1. Spatial and temporal cycling of P. mirabilis swarming behaviour and colony formation. Moving clockwise from the lower left, vegetative swimmer cells are induced to differentiate into swarmer cell morphology (elongated, hyperflagellated, polynucleoid cells) upon contact with a surface such as L agar. The microcolony of swarmer cells grows through cell division, but swarming motility is not expressed. This lag period is typically about 3 h on L agar at 37 °C. At this point, an unknown event triggers swarmer cell migration, which is manifested as a seemingly spontaneous uniform movement of swarmer cells outwards from the colony periphery. Migration continues for about 3 h and is followed by a short (30–45 min) consolidation phase where the cells dedifferentiate to swimmer cell morphology and cease to swarm. These events are cyclical, such that on agar media, swarming colony biofilm development is observed as a series of ever-expanding concentric rings that are composed of morphologically and physiologically distinct cell types (arrows indicate areas populated by either swimmer or swarmer cells). This cycle leads to the bull’s-eye pattern typical of P. mirabilis colonies.

 
The presence of a lag period prior to swarming behaviour and the co-ordination of the swarming colony biofilm to form bull’s-eye patterns suggests that cell–cell interaction and communication may control these events. Our previous work demonstrated that the lag period prior to P. mirabilis swarming migration is dependent upon the density of cells in the nascent preswarming colony (Belas et al., 1998 ). In this earlier work, mutations in rsbA, a gene encoding a putative membrane sensor protein (RsbA) with homology to Vibrio harveyi LuxQ (Bassler et al., 1994 ), were shown to have a reduced lag period, presumably due to defects in the ability of RsbA to sense the density of cells. These data suggest that RsbA may function in an analogous role as V. harveyi LuxQ to sense the same signal as LuxQ.

Quorum sensing is a common mechanism that acts to control bacterial physiology by regulating gene expression in response to population density. Quorum sensing relies on the accumulation of small extracellular signalling molecules, referred to as autoinducers, to modulate the transcription of target genes and operons. Autoinducer 1 (AI-1) quorum systems, using N-acylhomoserine lactone (AHL) derivatives as the signalling molecules, have been uncovered in many Gram-negative bacteria (Fuqua et al., 1996 ; Swift et al., 1996 ). The genes under the influence of AHL quorum sensing regulation include those encoding virulence factors (Parkins et al., 2001 ; Passador et al., 1993 ; Pirhonen et al., 1993 ; Tang et al., 1996 ; Telford et al., 1998 ), as well as those responsible for biofilm formation (Davies et al., 1998 ; De Kievit et al., 2001 ; Eberl et al., 1996 ). However, despite extensive analysis, AHL AIs do not appear to play a major role in P. mirabilis swarming colony biofilm formation (Belas et al., 1998 ).

Quorum sensing can be achieved through molecules other than AHL. Bassler and co-workers (Bassler, 1999 ; Surette & Bassler, 1998 ; Surette et al., 1999 ) have described an alternative quorum sensing mechanism that uses an autoinducer molecule, referred to as autoinducer 2 or AI-2, that appears to be highly conserved in both Gram-negative and Gram-positive bacteria (Schauder & Bassler, 2001 ). In V. harveyi, the membrane protein LuxQ, which is a homologue of P. mirabilis RsbA, acts as the receptor of AI-2. The gene luxS is crucial for AI-2 activity, and luxS orthologues have been found in many bacteria (Bassler, 1999 ; Surette et al., 1999 ). The molecular identity of AI-2 has been recently reported as a derivative of furanone (Schauder et al., 2001 ).

With the exception of its role in the regulation of V. harveyi luminescence, the function of AI-2 remains enigmatic for most of the bacterial species found to possess luxS orthologues. For example, many pathogenic bacteria possess genes with homology to luxS, leading to speculation that AI-2 may function to regulate aspects of bacterial virulence and pathogenicity. Recent reports have tested this hypothesis using enterohaemorrhagic and enteropathogenic Escherichia coli (EHEC and EPEC) (Sperandio et al., 1999 ), Helicobacter pylori (Forsyth & Cover, 2000 ; Joyce et al., 2000 ), Shigella flexneri (Day & Maurelli, 2001 ) and Porphyromonas gingivalis (Chung et al., 2001 ; Frias et al., 2001 ). The strongest connection between AI-2 activity and pathogenicity has been seen in E. coli O157:H7, where AI-2 controls the expression of type III secretion gene transcription and protein secretion (Sperandio et al., 1999 ). AI-2 activity may also function to control a gene involved in haemin acquisition in P. gingivalis (Chung et al., 2001 ) and could, through this interaction, be involved in P. gingivalis pathogenicity. However, a luxS mutation in S. flexneri had no effect on virulence (Day & Maurelli, 2001 ) and a similar mutation in H. pylori had no effect on the expression of any known virulence factor (Joyce et al., 2000 ). Thus, greater knowledge of the role of AI-2 in the regulation of virulence and biofilm formation is needed to understand the function of AI-2 quorum sensing regulation in bacterial pathogenicity.

We are interested in understanding the genetic mechanisms controlling the development of P. mirabilis biofilms, the swarming colony pattern, and how swarming behaviour enhances the uropathogenicity of these bacteria. We hypothesized that AI-2 quorum sensing may influence both P. mirabilis biofilm and colony pattern formation, as well as virulence. In this report, we show the luxS-dependent synthesis of AI-2 by P. mirabilis and the coordinate expression of AI-2 during swarming migration, and compare the phenotype and virulence of mutants defective in AI-2 production to those of the wild-type.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Strains, plasmids and media.
The strains and plasmids used in this study are listed in Table 1. P. mirabilis BB2000 is wild-type for swimming and swarming behaviours. E. coli and P. mirabilis strains were grown in Luria–Bertani broth (LB) or L agar (1·5%, w/v, agar) at 37 °C. To prevent swarming, P. mirabilis strains were grown on LSW- agar as previously described (Belas et al., 1991a , b ). Swimming behaviour was measured microscopically and macroscopically in Mot agar (Belas et al., 1998 ). Swarming behaviour was assessed using the method of Belas et al. (1998) . All V. harveyi strains were maintained on LM medium (Belas et al., 1984 ). V. harveyi was grown aerobically in Autoinducer Bioassay (AB) medium at 25 °C for the AI-2 bioassay (Greenberg et al., 1979 ). Urease activity was measured by using urea agar medium (Difco) according to the recommendations of the manufacturer. Protease activity was measured as a zone of clearing on Skimmed Milk Agar medium, as described by Walker et al. (1999) . Haemolytic activity was detected as a zone of greenish discoloration surrounding positive colonies on trypticase soy agar with 5% sheep red blood cells (TSA II, Becton Dickinson Microbiology Systems) after an overnight incubation at 37 °C. Antibiotics were used at the following final concentrations: 40 µg kanamycin ml-1, 100 µg ampicillin ml-1, 40 µg chloramphenicol ml-1, and 100 µg rifampicin ml-1.


View this table:
[in this window]
[in a new window]
 
Table 1. Strains and plasmids used

 
Molecular cloning and mutagenesis.
Standard methods were used for the manipulation of DNA (Ausubel et al., 1987 ) unless specified otherwise. The P. mirabilis luxS gene (luxSPM; used to distinguish this gene from the E. coli orthologue, luxSEC) was identified by Southern blot hybridization of genomic DNA cleaved with EcoRI and hybridized to a radioactively labelled DNA fragment obtained by PCR amplification of the E. coli MG1655 (Blattner et al., 1997 ) luxS gene (luxSEC) using oligonucleotides luxSfor (5'-TTGGGATGACGCAACAGCA-3') and luxSrev (5'-GAACGCCGTCAGCAGGAAA-3') as primers. A random EcoRI library of P. mirabilis genomic DNA fragments was then ligated with plasmid pCR2.1 (Invitrogen) and transformed into E. coli DH5{alpha}. The library of recombinants was screened for AI-2 activity using V. harveyi BB170 (described below). A transformant that complemented AI-2 production in DH5{alpha} was identified and the nucleotide sequence of the inserted DNA analysed. This plasmid, pRS104, harboured a 1598 bp insert of P. mirabilis genomic DNA containing luxSPM.

A nonpolar null mutation in luxSPM was constructed by first removing the plasmid-borne EcoRI site in pRS104, resulting in pRS107 (Table 1). A 847 bp SmaI fragment containing the aphA-3 gene from pUC18K (Menard et al., 1993 ) was inserted at the EcoRI site internal to luxSPM by blunt-end ligation to Klenow-treated plasmid. The resulting plasmid was named pRS110. Plasmid pRS110 was digested with SacI and KpnI, and a 2·5 kb fragment containing luxS'::aphA::'luxS was ligated to pGP704 (Miller & Mekalanos, 1988 ) also digested with SacI and KpnI, resulting in pRS112. This plasmid was transformed into E. coli SM10 {lambda}pir and conjugally transferred to P. mirabilis BB2000 by filter mating (Belas et al., 1991a ). Mutation of the chromosomal copy of luxSPM was confirmed by antibiotic resistance spectrum (KmR ApS), PCR amplification of luxSPM locus from the putative luxS strain, and measurement of AI-2 activity using the V. harveyi luminescence assay. Several colonies that lacked AI-2 resulting from the introduction of luxS'::aphA::'luxS were found. A representative was chosen and named RS601.

Nucleotide sequencing and analysis.
Double-stranded DNA was used as a template for nucleotide sequencing by the recommended procedures of the Prism Ready Reaction Dye Deoxy Termination Kit (Applied Biosystems) in conjunction with Taq polymerase and a model 373A DNA sequencer (Applied Biosystems). Nucleotide and deduced amino acid sequences were analysed with Vector NTI V5.0 software (Informax) and the BLAST family of programs (Altschul et al., 1990 , 1997 ; Gish & States, 1993 ; Worley et al., 1995 ). Phylogenetic trees of P. mirabilis LuxS relatedness were constructed by aligning the LuxS deduced amino acid sequences obtained from complete and partial genome databases using CLUSTAL W V1.75 [http://www.es.embnet.org/Doc/phylodendron/clustal-form.html] (Thompson et al., 1994 ) and arranged with the computer program ‘Phylodendron’ [http://iubio.bio.indiana.edu/soft/molbio/java/apps/trees/].

The following nucleotide sequences and accession numbers were used in generating LuxS dendrograms: Escherichia coli MG1655, U00096; Salmonella typhimurium LT, WUGSC_99287; Salmonella paratyphi A, WUGSC_32027; Salmonella typhi CT18, WUGSC_573; Klebsiella pneumoniae, Contig1032; Yersinia pestis strain CO-92, Sanger_632; Vibrio harveyi, AF120098; Vibrio cholerae chromosome I, AE003852; Shewanella putrefaciens, sputre_7833; Neisseria meningitidis serogroup B strain M, AE002098; Neisseria gonorrhoeae, OUACGT_485; Haemophilus influenzae Rd, U32731; Pasteurella multocida PM70, AE004439; Actinobacillus actinomycetemcomitans, OUACGT_714; Campylobacter jejuni, AL111168; Deinococcus radiodurans R1, AE000513; Clostridium perfringens, AB028629; Streptococcus mutans, UOKNOR_1309; Streptococcus pneumoniae, S.pneumoniae_3836; Streptococcus equi, sequi_Contig243; Streptococcus pyogenes strain SF370, AE004092; Enterococcus faecalis unfinished, gef_11360; Clostridium difficile strain 630 (epidemic type X) unfinished, Sanger Centre unassigned; Staphylococcus epidermidis, TIGR_1282; Staphylococcus aureus, Sanger_1280_3; Helicobacter pylori 26695, AE000511; Bacillus subtilis, AL009126; Bacillus anthracis, TIGR_1392; Bacillus halodurans C-125, BA000004; Borrelia burgdorferi, AE000783; Clostridium acetobutylicum, C.aceto_gnl; Porphyromonas gingivalis W83, P.gingivalis_GPG.con; Haemophilus ducreyi strain 35000HP, HTSC_730.

Preparation of cell-free conditioned media for AI-2 assays.
Conditioned media from broth cultures were prepared using a modification of the protocol described by Sperandio et al. (1999) . Overnight cultures grown in LB at 37 °C were diluted 1:100 in fresh LB and cultured in a shaking water bath at 37 °C. When the cultures reached an OD600 of 0·3, the bacteria were again diluted 1:100 in fresh LB and incubated in a shaking water bath at 37 °C until an OD600 of 1·2 was reached (about 12 h incubation). The bacteria were removed by centrifugation (14000 g, 5 min, 20 °C) and the supernatant filter-sterilized (0·2 µm pore-size cellulose acetate filter; Millipore). Aliquots of the conditioned media were kept on ice to be used immediately in AI-2 assays or stored at -20 °C until needed.

Production of AI-2 during P. mirabilis swarming migration and behaviour on L agar medium was assessed as follows. P. mirabilis overnight cultures grown in LB were prepared for swarming behaviour assays as described by Belas et al. (1998) , in which a 5 µl inoculum of about 2·5x106 cells was dispensed on the L agar surface, incubated at 37 °C, and assayed for swarming motility. To determine AI-2 activity during swarming migration and colony formation, a cylinder of agar 1 cm in diameter and 1 cm in height from the outer edge of the developing swarming colony was removed at 1 h intervals using a sterile brass cork borer (Fisher Scientific). Colony-forming units (c.f.u.) were determined in parallel by removing the cells from the surface of the agar core with vigorous mixing, followed by growth of serial dilutions of the suspension on L agar. AI-2 activity was also measured at discrete points within a preformed 5 h old swarming colony that represents one cycle of swarming behaviour. In this case, agar cores were removed from the point of inoculation, a point one-third of the way to the periphery, a point two-thirds of the distance to the periphery, and at the periphery of the colony. Each agar plug with overlying bacteria was frozen at -20 °C for 18 h and then thawed rapidly by addition of 300 µl AB medium. Insoluble agar and bacterial cells were removed by centrifugation and filtration as described above, and the supernatant assayed immediately or stored at -20 °C. This ‘freeze-squeeze’ process resulted in 800–900 µl fluid.

V. harveyi luminescence assay for AI-2.
The presence of AI-2 in the conditioned media was assayed by using the V. harveyi BB170 (luxN::Tn5) reporter strain, which responds only to AI-2 (Surette & Bassler, 1998 ). The luminescence assays were performed as described by Surette & Bassler (1998) , using an EG&G Berthold Microlumat LB 96P luminometer. The data are reported as relative light units (RLUs) or as the fold stimulation of light emission by V. harveyi BB170 compared to the RLU values obtained from the corresponding DH5{alpha} (AI-2-) negative control.

Murine UTI virulence assay.
A modification (Johnson et al., 1987 ) of the well-established mouse model of ascending UTI (Hagberg et al., 1983 ) was used to measure P. mirabilis virulence. Female CBA/J (Harlan Sprague–Dawley) were transurethrally challenged with about 3x107 c.f.u. of bacteria per mouse. After 7 days, the mice were killed, and bacteria recovered from urine, bladder and kidneys were enumerated on LSW- agar plates containing the appropriate antibiotics. The range of detection in this assay is 102–109 c.f.u. (ml urine)-1 or c.f.u. (g tissue)-1 (Li et al., 1999 ).

Statistical methods.
Mean numbers of c.f.u. ml-1 or g-1 from cultures of urine or tissue homogenates were compared by the Mann–Whitney and non-parametric ANOVA tests.

Materials and reagents.
All reagents were of the highest purity available. Components of bacteriological media were purchased from Difco. Restriction endonucleases and DNA-modifying enzymes were obtained from New England Biolabs, Boehringer Mannheim Biochemicals, Promega or Qiagen and were used according to the supplier’s recommendations.


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Identification of AI-2 activity in P. mirabilis
We have previously reported on the density-dependent nature of P. mirabilis swarming, the homology between RsbA and LuxQ, and the lack of identifiable AI-1 activity associated with P. mirabilis swarming behaviour (Belas et al., 1998 ). These led us to speculate that P. mirabilis may possess an AI-2 extracellular signalling molecule that might influence swarming behaviour, biofilm formation and other phenotypes, notably expression of virulence factors, during UTI. To test this hypothesis, the V. harveyi AI-2 reporter strain, BB170 (a gift from B. Bassler; Table 1) lacking a sensor for AI-1 (AHL) but containing an intact sensor for AI-2, was used to detect AI-2 activity in P. mirabilis cell-free conditioned media. Table 2 shows the results from this assay. V. harveyi BB170 was inoculated into AB medium containing 10% cell-free conditioned broth from P. mirabilis swimmer cells and the luminescence measured after 4–5 h incubation. V. harveyi BB7 and E. coli MG1655 served as positive controls for AI-2 production, while E. coli DH5{alpha}, lacking AI-2 activity (Surette et al., 1999 ), was used as a negative control. We found that P. mirabilis swimmer cells produced an extracellular AI-2 activity comparable to that obtained from V. harveyi BB7 (Table 2). We conclude that P. mirabilis swimmer cells possess an AI-2-like activity.


View this table:
[in this window]
[in a new window]
 
Table 2. Induction of luminescence in V. harveyi BB170 by conditioned media

 
AI-2 activity coincides with swarming behaviour
We next measured AI-2 activity during swarmer cell differentiation and migration, as a function of both time post-inoculation on L agar, and position in the developing swarmer colony (at a fixed time of 5 h). An inoculum of swimmer cells was applied to the surface of L agar, and AI-2 was measured during swarmer cell differentiation and swarming migration. Our previous work has shown that P. mirabilis requires an approximately 3 h lag phase after inoculation on L agar before the onset of swarming migration (Belas et al., 1998 ). In such experiments, one full cycle of swarming takes about 5 h at 37 °C. As the swarming biofilm developed, agar plugs containing the bacteria and underlying agar matrix were removed from the leading edge of the swarming colony (Methods), and AI-2 activity was measured using the V. harveyi BB170 AI-2 assay. As is shown in Fig. 2, AI-2 activity rose to a maximum at 3 h after inoculation, then dropped back to near background level. The peak in AI-2 activity at 3 h corresponds to the time at which the cells start swarming migration and suggests that AI-2 may play a role in orchestrating this behaviour.



View larger version (8K):
[in this window]
[in a new window]
 
Fig. 2. Temporal expression of AI-2 activity during swarming behaviour and migration on L agar. An inoculum of swimmer cells was pipetted on the surface of L agar, incubated at 37 °C, and AI-2 activity measured as the cells differentiated and began swarming migration and behaviour. At 1 h intervals, a core containing the cells and underlying agar matrix at the periphery of the bacterial growth was removed, and conditioned medium obtained, as described in Methods. AI-2 activity was assayed using the V. harveyi BB170 sensor strain. AI-2 activity is expressed as the fold induction of relative light units compared to that of an uninduced control, and represents the mean of at least three experiments.

 
AI-2 activity was also assessed at discrete areas of a 5 h swarming colony, i.e. a colony that had completed one cycle of differentiation, migration, dedifferentiation and consolidation (Fig. 3). In this experiment, AI-2 activity was greatest at the site of inoculation and decreased to a low at the periphery of the swarming colony.



View larger version (9K):
[in this window]
[in a new window]
 
Fig. 3. Spatial expression of AI-2 activity during swarming behaviour and migration on L agar. An inoculum of swimmer cells was pipetted on the surface of L agar and incubated at 37 °C for 5 h, a time at which one complete cycle of swarming behaviour has been completed. AI-2 activity was measured by removing cores of cells plus underlying agar (see Methods) at sections of the colony corresponding to the point of inoculation, one-third of the distance to the colony edge and two-thirds of the distance to the colony edge, and at the periphery of the colony. Conditioned cell-free media were obtained from these cores as described in Methods and AI-2 activity was assayed using the V. harveyi BB170 sensor strain. AI-2 activity is expressed as the fold induction of relative light units compared to that of an uninduced control, and represents the mean of at least three experiments.

 
Taken together, the data in Figs 2 and 3 suggest that AI-2 activity is correlated with swarming. Specifically, AI-2 activity is maximal at the site of inoculation and peaks at the same time as swarming migration begins. Other factors, such as cell division and growth, may also contribute to the changes in AI-2 activity observed in these experiments. For example, the sharp increase in AI-2 activity at 3 h may be due to an increase in cells cm-2, i.e. a cell density effect due to quorum sensing. The inoculum contained about 2·5 x 106 c.f.u. cm-2, increasing to approximately 4x108 c.f.u. cm-2 at the 3 h sampling. Thereafter, the cell density remained at 1–2x108 c.f.u. cm-2 as the biofilm of cells moved outwards on the agar surface. The increase in cells cm-2 at 3 h may be responsible for the increase in AI-2 activity; however, the subsequent drop in AI-2 at 4 and 5 h cannot be solely explained by the slight drop in cell density recorded at those times, and may point to other unknown factors, such as the degradation of AI-2, that control P. mirabilis AI-2 expression.

A luxS orthologue is responsible for AI-2 activity in P. mirabilis
The luxS gene is essential for AI-2 production in many bacteria, prompting us to determine whether P. mirabilis genomic DNA contained a luxS orthologue. Southern blots of P. mirabilis genomic DNA separately digested with five restriction endonucleases were hybridized to radioactively labelled pRSluxSEC containing a PCR fragment of E. coli luxS. The analysis of these blots indicated that EcoRI digestion resulted in a 1·6 kb DNA fragment that hybridized to luxSEC. An EcoRI digest of genomic DNA was ligated to pBluescript SK(+) and transformed into E. coli DH5{alpha}, a strain lacking AI-2 activity (Surette et al., 1999 ). The resulting recombinant colonies were screened for AI-2 activity using the V. harveyi BB170 AI-2 assay. Five of the cell-free conditioned media supernatants from the recombinant E. coli produced AI-2 activity equal to or in excess of conditioned media from DH5{alpha} harbouring pRSluxSEC (Table 2). Restriction mapping of the plasmid carried by these bacteria revealed that each of these recombinant E. coli harboured the same foreign DNA, presumably containing luxSPM, and indicating that luxSPM is functional in E. coli. A single representative plasmid, pRS104 (Table 1), was used for further work.

The nucleotide sequence of the EcoRI fragment of P. mirabilis genomic DNA inserted in pRS104 was determined. The analysis of the 1598 bp locus is shown in Fig. 4(a). Three ORFs were found, one of which is highly homologous to other bacterial luxS deduced amino acid sequences (Fig. 4b). The homology of ORF2 to other LuxS sequences is greatest with LuxS of E. coli (67% identity) and Salmonella enterica serovar Typhimurium (65% identity). We refer to this ORF as luxS and use luxSPM to distinguish it from other luxS genes. As shown in Fig. 4(a), two partial ORFs were also discovered on either side of luxSPM. To the left of luxS, as drawn in Fig. 4(a), is an ORF whose deduced amino acid sequence is homologous to E. coli gshA, encoding {gamma}-glutamylcysteine synthetase, and to the right of luxS is an ORF whose deduced amino acid sequence has partial homology to a series of proteins involved in ion or antibiotic transport. The data presented in Fig. 4(a, b) show that the P. mirabilis luxS locus is similar in its arrangement of genes to the luxS locus of E. coli and S. enterica serovar Typhimurium. These data also demonstrate that luxSPM is homologous to the luxS genes of other bacteria (Fig. 4b). The latter point is further strengthened by the ability of luxSPM to complement the luxS defect in DH5{alpha}, thereby producing AI-2 activity in the recombinant strain.



View larger version (19K):
[in this window]
[in a new window]
 
Fig. 4. Organization of the P. mirabilis luxS locus and relatedness to LuxS orthologues of other eubacteria. (a) A 1589 bp EcoRI fragment from the P. mirabilis genome contains, as read from left to right, the 3' end of a gene homologous to E. coli gshA, luxS, and the 3' end of an ORF with partial deduced amino acid homology to antibiotic transport proteins. The direction of the reading frame is indicated by the arrow under each gene. The GenBank accession number for this nucleotide sequence is AY044337. (b) Dendrogram showing the phylogenetic relatedness of P. mirabilis LuxS to the deduced amino acid sequences of other LuxS orthologues obtained from computer databases of completed and partial bacterial genome sequences. Alignment of the deduced amino acid sequences was done using CLUSTAL W and the tree was constructed using the associated phylogenetic tree program, Phylodendron (see Methods for details).

 
Phenotypic analysis of an AI-2 negative P. mirabilis
A nonpolar mutation in luxSPM was constructed by inserting the aphA-3 gene from pUC18K (Menard et al., 1993 ) at an EcoRI site internal to luxS to produce pRS110 (Table 1). Plasmid pRS110 failed to produce AI-2 activity in DH5{alpha} (Table 2). The 2·5 kb DNA fragment containing luxSPM'::aphA::'luxSPM was then ligated into the suicide vector pGP704 (Miller & Mekalanos, 1988 ) for conjugal transfer to P. mirabilis. This plasmid, pRS112, was used to produce a nonpolar mutation of the chromosomal copy of luxSPM through allelic-exchange mutagenesis. The resulting colonies were screened for loss of plasmid-borne ampillicin resistance, acquisition of kanamycin resistance (conferred by aphA-3) and loss of AI-2 activity (measured in cell-free conditioned media). Those strains lacking detectable AI-2 were confirmed to be the result of a double crossover event by a PCR-based analysis of the genomic luxSPM. Six colonies were identified that produced no AI-2 activity, resulting from an allelic exchange of luxS and the mutated gene. One of these, RS601 (AI-2-, luxSPM'::aphA::'luxSPM), was phenotypically characterized.

As anticipated, RS601 did not produce detectable AI-2 activity (Table 2), when assayed using the V. harveyi BB170 sensor. We compared RS601 to the parent strain (BB2000) in swimming motility and chemotaxis behaviour, and swarming motility and behaviour. Swimming speed and chemotaxis were measured microscopically as well as during growth in Mot agar. When compared to the wild-type bacteria, RS601 showed no differences in either swimming speed or chemotaxis behaviour. An analysis of swarmer cell morphology, movement and biofilm formation was conducted. Swarmer cell differentiation, the timing of the beginning of swarming migration and the swarming colony pattern formation of RS601 were also indistinguishable from those of the wild-type. Other phenotypes, such as growth rate and the production of urease, protease and haemolysin, were also unaffected by the luxS mutation. We conclude that the luxS mutation does not lead to an overt change in P. mirabilis motility or biofilm phenotypes as determined by this set of assays.

AI-2 activity does not affect P. mirabilis virulence during UTI
It may be that defects in AI-2 synthesis are not apparent when the bacteria are grown in nutrient-rich conditions, but are manifested under conditions that challenge the survival of the cells. To determine if loss of AI-2 activity affected P. mirabilis virulence, we compared the uropathogenicity of RS601 to that of wild-type P. mirabilis, using a modification (Johnson et al., 1987 ) of the mouse model of ascending UTI (Hagberg et al., 1983 ). This mouse model has been successfully used by others to assess the uropathogenicity of E. coli (Johnson et al., 1993a ) and Providencia stuartii (Johnson et al., 1987 ), and by our group to assess uropathogenicity of P. mirabilis mutants defective in flagellin synthesis (Mobley et al., 1996 ) and strains defective in the production of an extracellular metalloprotease (Walker et al. , 1999 ).

Fig. 5 shows the colonization of mouse urine, bladder and kidneys 7 days after transurethral challenge with either RS601 or wild-type P. mirabilis. Ten mice were used per bacterial strain. Two mice infected with wild-type P. mirabilis died from stone blockage on day 6 and were not used in the analysis. A pairwise comparison of the mean c.f.u. for each tissue (indicated by the horizontal line in each column in Fig. 5) suggests a noticeable increase in the survival of the AI-2-deficient mutant in each tissue. The trend towards increased survival in the luxS strain appears in all samples (urine, bladder and kidney) examined. The P values (>0·05), however, do not reflect a statistical significance to these differences. Thus, a luxS mutant is not attenuated in virulence, suggesting that AI-2 quorum signalling is not required for P. mirabilis UTI.



View larger version (13K):
[in this window]
[in a new window]
 
Fig. 5. Mouse urinary tract colonization of the urine, bladder, and kidneys 7 days after transurethral infection with either an AI-2-negative or wild-type P. mirabilis. The mean c.f.u. per ml urine or per g tissue of 10 mice is indicated by a horizontal bar in each column. The P values, as determined by the Mann–Whitney test, for comparisons of mean numbers of c.f.u. per ml or g of urine or bladder or kidney homogenates for wild-type and LuxS- strain are: urine, P=0·055; bladder, P=0·059; left kidney, P=0·209; and right kidney, P=0·106. The overall P value determined by a non-parametric ANOVA comparing the AI-2-negative strain to wild-type is P=0·061.

 
Concluding remarks
In summary, the results presented show that P. mirabilis luxS is critical for AI-2 activity. The deduced amino acid sequence of LuxS, as well as the organization of the luxS locus, emphasize the homology between luxSPM and luxS genes from other bacteria, especially the Enterobacteriaceae. A luxS mutation in P. mirabilis, however, has no demonstrable effects on growth, the production of urease, protease or haemolysin, swimming motility and behaviour, swarmer cell differentiation, swarming behaviour and biofilm formation, or uropathogenicity. The finding that luxS does not alter either the swarming phenotype or UTI virulence is in agreement with previous observations that swarming behaviour is essential for virulence (Allison et al., 1992 , 1993 ; Chippendale et al., 1994 ). Day & Maurelli (2001) have suggested that the AI-2 quorum sensing is important to those pathogens that are constantly exposed to high levels of AI-2 derived from normal flora. During UTI, P. mirabilis is not likely to encounter high levels of AI-2 activity from other bacteria. However, P. mirabilis is also a member of the intestinal flora of many animals, where high AI-2 activity may well exist. Perhaps, because of this dual mode of existence, P. mirabilis has adapted to use AI-2 quorum sensing regulation while in the intestine, but exploits other regulatory schemes to control its virulence factors in the urinary tract, which is normally free of resident flora. If true, this possibility may adversely impact those drug therapies design to intervene in P. mirabilis UTI through inhibition of AI-2 activity.


   ACKNOWLEDGEMENTS
 
We thank Richard Hebel, University of Maryland School of Medicine, for help with statistical analyses, Bonnie Bassler, Princeton University, for the gift of BB170 and helpful advice, and James Kaper, Vanessa Sperandio, Harry L. T. Mobley and Jim Manos for their comments and suggestions. This work was supported by grants from the National Science Foundation (MCB-0078506) and the National Institutes of Health (DK49720).


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Allison, C., Lai, H. C. & Hughes, C. (1992). Co-ordinate expression of virulence genes during swarm-cell differentiation and population migration of Proteus mirabilis. Mol Microbiol 6, 1583-1591.[Medline]

Allison, C., Lai, H. C., Gygi, D. & Hughes, C. (1993). Cell differentiation of Proteus mirabilis is initiated by glutamine, a specific chemoattractant for swarming cells. Mol Microbiol 8, 53-60.[Medline]

Altschul, S., Gish, W., Miller, W., Myers, E. & Lipman, D. (1990). Basic local alignment search tool. J Mol Biol 215, 403-410.[Medline]

Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W. & Lipman, D. J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25, 3389-3402.[Abstract/Free Full Text]

Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Smith, J. A., Seidman, J. G. & Struhl, K. (1987). Current Protocols in Molecular Biology. New York: Greene Publishing Associates & Wiley.

Bassler, B. L. (1999). How bacteria talk to each other: regulation of gene expression by quorum sensing. Curr Opin Microbiol 2, 582-587.[Medline]

Bassler, B. L., Wright, M. & Silverman, M. R. (1994). Multiple signalling systems controlling expression of luminescence in Vibrio harveyi: sequence and function of genes encoding a second sensory pathway. Mol Microbiol 13, 273-286.[Medline]

Belas, R., Mileham, A., Cohn, D., Hilman, M., Simon, M. & Silverman, M. (1982). Bacterial bioluminescence: isolation and expression of the luciferase genes from Vibrio harveyi. Science 218, 791-793.[Medline]

Belas, R., Simon, M. & Silverman, M. (1984). Transposon mutagenesis of marine Vibro spp. J Bacteriol 158, 890-896.[Medline]

Belas, R., Erskine, D. & Flaherty, D. (1991a). Transposon mutagenesis in Proteus mirabilis. J Bacteriol 173, 6289-6293.[Medline]

Belas, R., Erskine, D. & Flaherty, D. (1991b). Proteus mirabilis mutants defective in swarmer cell differentiation and multicellular behavior. J Bacteriol 173, 6279-6288.[Medline]

Belas, R., Schneider, R. & Melch, M. (1998). Characterization of Proteus mirabilis precocious swarming mutants: identification of rsbA, encoding a regulator of swarming behavior. J Bacteriol 180, 6126-6139.[Abstract/Free Full Text]

Blattner, F. R., Plunkett, G., 3rd, Bloch, C. A. & 14 other authors (1997). The complete genome sequence of Escherichia coli K-12. Science 277, 1453–1474.[Abstract/Free Full Text]

Chippendale, G. R., Warren, J. W., Trifillis, A. L. & Mobley, H. L. (1994). Internalization of Proteus mirabilis by human renal epithelial cells. Infect Immun 62, 3115-3121.[Abstract]

Chung, W. O., Park, Y., Lamont, R. J., McNab, R., Barbieri, B. & Demuth, D. R. (2001). Signaling system in Porphyromonas gingivalis based on a LuxS protein. J Bacteriol 183, 3903-3909.[Abstract/Free Full Text]

Davies, D. G., Parsek, M. R., Pearson, J. P., Iglewski, B. H., Costerton, J. W. & Greenberg, E. P. (1998). The involvement of cell-to-cell signals in the development of a bacterial biofilm. Science 280, 295-298.[Abstract/Free Full Text]

Day, W. A.Jr & Maurelli, A. T. (2001). Shigella flexneri LuxS quorum-sensing system modulates virB expression but is not essential for virulence. Infect Immun 69, 15-23.[Abstract/Free Full Text]

De Kievit, T. R., Gillis, R., Marx, S., Brown, C. & Iglewski, B. H. (2001). Quorum-sensing genes in Pseudomonas aeruginosa biofilms: their role and expression patterns. Appl Environ Microbiol 67, 1865-1873.[Abstract/Free Full Text]

De Lorenzo, M., Herrero, M., Jakubzik, U. & Timmis, K. N. (1990). Mini-Tn5 transposon derivatives for insertion mutagenesis, promoter probing, and chromosomal insertion of cloned DNA in gram-negative eubacteria. J Bacteriol 172, 6568-6572.[Medline]

Eberl, L., Winson, M., Sternberg, C. & 7 other authors (1996). Involvement of N-acyl-L-homoserine lactone autoinducers in controlling the multicellular behaviour of Serratia liquefaciens. Mol Microbiol 20, 127–136.

Forsyth, M. H. & Cover, T. L. (2000). Intercellular communication in Helicobacter pylori: luxS is essential for the production of an extracellular signaling molecule. Infect Immun 68, 3193-3199.[Abstract/Free Full Text]

Frias, J., Olle, E. & Alsina, M. (2001). Periodontal pathogens produce quorum sensing signal molecules. Infect Immun 69, 3431-3434.[Abstract/Free Full Text]

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

Gish, W. & States, D. (1993). Identification of protein coding regions by database similarity search. Nat Genet 3, 266-272.[Medline]

Greenberg, E., Hastings, J. & Ulitzur, S. (1979). Induction of luciferase synthesis in Beneckea harveyi by other marine bacteria. Arch Microbiol 120, 87-91.

Hagberg, L., Engberg, I., Freter, R., Lam, J., Olling, S. & Svanborg Eden, C. (1983). Ascending, unobstructed urinary tract infection in mice caused by pyelonephritogenic Escherichia coli of human origin. Infect Immun 40, 273-283.[Medline]

Johnson, D. E., Lockatell, C. V., Hall-Craigs, M., Mobley, H. L. & Warren, J. W. (1987). Uropathogenicity in rats and mice of Providencia stuartii from long-term catheterized patients. J Urol 138, 632-635.[Medline]

Johnson, D. E., Russell, R. G., Lockatell, C. V., Zulty, J. C. & Warren, J. W. (1993a). Urethral obstruction of 6 hours or less causes bacteriuria, bacteremia, and pyelonephritis in mice challenged with ‘nonuropathogenic’ Escherichia coli. Infect Immun 61, 3422-3428.[Abstract]

Johnson, D. E., Russell, R. G., Lockatell, C. V., Zulty, J. C., Warren, J. W. & Mobley, H. L. (1993b). Contribution of Proteus mirabilis urease to persistence, urolithiasis, and acute pyelonephritis in a mouse model of ascending urinary tract infection. Infect Immun 61, 2748-2754.[Abstract]

Joyce, E. A., Bassler, B. L. & Wright, A. (2000). Evidence for a signaling system in Helicobacter pylori: detection of a luxS-encoded autoinducer. J Bacteriol 182, 3638-3643.[Abstract/Free Full Text]

Li, X., Johnson, D. E. & Mobley, H. L. (1999). Requirement of MrpH for mannose-resistant Proteus-like fimbria-mediated hemagglutination by Proteus mirabilis. Infect Immun 67, 2822-2833.[Abstract/Free Full Text]

McLean, R. J., Whiteley, M., Stickler, D. J. & Fuqua, W. C. (1997). Evidence of autoinducer activity in naturally occurring biofilms. FEMS Microbiol Lett 154, 259-263.[Medline]

Menard, R., Sansonetti, P. J. & Parsot, C. (1993). Nonpolar mutagenesis of the ipa genes defines IpaB, IpaC, and IpaD as effectors of Shigella flexneri entry into epithelial cells. J Bacteriol 175, 5899-5906.[Abstract]

Miller, V. L. & Mekalanos, J. J. (1988). A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires toxR. J Bacteriol 170, 2575-2583.[Medline]

Mobley, H. L. T. & Chippendale, G. R. (1990). Hemagglutinin, urease, and hemolysin production by Proteus mirabilis from clinical sources. J Infect Dis 161, 525-530.[Medline]

Mobley, H. & Warren, J. (1996). Urinary Tract Infections. Washington, DC: American Society for Microbiology.

Mobley, H., Belas, R., Lockatell, V., Chippendale, G., Trifillis, A., Johnson, D. & Warren, J. (1996). Construction of a flagellum-negative mutant of Proteus mirabilis: effect on internalization into human renal epithelial cells and virulence in a mouse model of ascending urinary tract infection. Infect Immun 64, 5332-5340.[Abstract]

Parkins, M. D., Ceri, H. & Storey, D. G. (2001). Pseudomonas aeruginosa GacA, a factor in multihost virulence, is also essential for biofilm formation. Mol Microbiol 40, 1215-1226.[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]

Pirhonen, M., Flego, D., Heikinheimo, R. & Palva, E. T. (1993). A small diffusible signal molecule is responsible for the global control of virulence and exoenzyme production in the plant pathogen Erwinia carotovora. EMBO J 12, 2467-2476.[Abstract]

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

Schauder, S. & Bassler, B. L. (2001). The languages of bacteria. Genes Dev 15, 1468-1480.[Free Full Text]

Schauder, S., Shokat, K., Surette, M. G. & Bassler, B. L. (2001). The LuxS family of bacterial autoinducers: biosynthesis of a novel quorum-sensing signal molecule. Mol Microbiol 41, 463-476.[Medline]

Simon, R., Priefer, U. & Puhler, A. (1982). A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram-negative bacteria. Bio/Technology 1, 784-791.

Sperandio, V., Mellies, J. L., Nguyen, W., Shin, S. & Kaper, J. B. (1999). Quorum sensing controls expression of the type III secretion gene transcription and protein secretion in enterohemorrhagic and enteropathogenic Escherichia coli. Proc Natl Acad Sci U S A 96, 15196-15201.[Abstract/Free Full Text]

Stickler, D. & Hughes, G. (1999). Ability of Proteus mirabilis to swarm over urethral catheters. Eur J Clin Microbiol Infect Dis 18, 206-208.

Stickler, D., Morris, N., Moreno, M. C. & Sabbuba, N. (1998). Studies on the formation of crystalline bacterial biofilms on urethral catheters. Eur J Clin Microbiol Infect Dis 17, 649-652.[Medline]

Surette, M. G. & Bassler, B. L. (1998). Quorum sensing in Escherichia coli and Salmonella typhimurium. Proc Natl Acad Sci USA 95, 7046-7050.[Abstract/Free Full Text]

Surette, M. G., Miller, M. B. & Bassler, B. L. (1999). Quorum sensing in Escherichia coli, Salmonella typhimurium, and Vibrio harveyi: a new family of genes responsible for autoinducer production. Proc Natl Acad Sci USA 96, 1639-1644.[Abstract/Free Full Text]

Swift, S., Throup, J., Williams, P., Salmond, G. & Stewart, G. (1996). Quorum sensing: a population-density component in the determination of bacterial phenotype. Trends Biochem Sci 21, 214-219.[Medline]

Tang, H. B., DiMango, E., Bryan, R., Gambello, M., Iglewski, B. H., Goldberg, J. B. & Prince, A. (1996). Contribution of specific Pseudomonas aeruginosa virulence factors to pathogenesis of pneumonia in a neonatal mouse model of infection. Infect Immun 64, 37-43.[Abstract]

Telford, G., Wheeler, D., Williams, P., Tomkins, P. T., Appleby, P., Sewell, H., Stewart, G. S., Bycroft, B. W. & Pritchard, D. I. (1998). The Pseudomonas aeruginosa quorum-sensing signal molecule N-(3-oxododecanoyl)-L-homoserine lactone has immunomodulatory activity. Infect Immun 66, 36-42.[Abstract/Free Full Text]

Thompson, J. D., Higgins, D. G. & Gibson, T. J. (1994). CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions, specific gap penalties and weight matrix choice. Nucleic Acids Res 22, 4673-4680.[Abstract]

Walker, K., Moghaddame-Jafari, S., Lockatell, C., Johnson, D. & Belas, R. (1999). ZapA, the IgA-degrading metalloprotease of Proteus mirabilis, is a virulence factor expressed specifically in swarmer cells. Mol Microbiol 32, 825-836.[Medline]

Worley, K., Wiese, B. & Smith, R. (1995). BEAUTY: an enhanced BLAST-based search tool that integrates multiple biological information resources into sequence similarity search results. Genome Res 5, 173-184.[Abstract]

Received 17 August 2001; revised 6 November 2001; accepted 12 November 2001.