|Home | About | Journals | Submit | Contact Us | Français|
The CepR-CepI quorum-sensing system has been shown to regulate production of the siderophore ornibactin, extracellular proteases, and N-octanoyl-homoserine-l-lactone (OHL) in Burkholderia cepacia strain K56-2. To examine the effect of cepIR on production of other siderophores, cepR mutants were constructed in strains that produce pyochelin in addition to salicylic acid and ornibactins. Pc715j-R1 (cepR::tp) hyperproduced ornibactin but produced parental levels of pyochelin and salicylic acid, suggesting that CepR is a negative regulator of ornibactin synthesis but not pyochelin or salicylic acid. Pc715j-R1 was also protease deficient and OHL negative. The effects of cepR on ornibactin biosynthetic genes were examined by constructing cepR pvdA-lacZ and cepR pvdD-lacZ mutants and monitoring β-galactosidase activity. There was an increase in expression of pvdA in the cepR mutant compared to the level in its parent strain in both low- and high-iron media during stationary phase. When the outer membrane protein profiles of a cepR mutant and the wild-type strain were compared on sodium dodecyl sulfate-polyacrylamide gel electrophoresis, there did not appear to be any difference in levels of expression of the ornibactin receptor. Experiments with cepI-lacZ and cepR-lacZ transcriptional fusions indicated that cepI was not expressed in the cepR mutant and that cepR acts as a negative regulator of its own expression. By a thin-layer chromatography assay for N-acyl homoserine lactones, OHL and N-hexanoyl-l-homoserine lactone (HHL) were detectable in K56-2 and Pc715j, both wild-type strains. OHL was not detectable and HHL was only weakly detectable in the cepI and cepR mutants. These results suggest that CepR is both a positive and negative transcriptional regulator and that CepR may influence the expression of ornibactin biosynthetic genes in addition to the expression of the cepIR quorum-sensing system.
Burkholderia cepacia is an opportunistic pathogen that primarily infects immunocompromised patients with cystic fibrosis (CF), chronic granulomatous disease, cancer, or burns or patients with indwelling devices (12). B. cepacia infections are a particular cause for concern in CF patients. First, although B. cepacia infections have varied outcomes, a significant number of B. cepacia-infected patients (20 to 35%) experience cepacia syndrome, which often culminates in death (14, 42). Second, there is increasing epidemiological evidence for patient-to-patient transmission (18). Since several epidemics have been shown to be caused by a single lineage, there is a possibility that some strains have an increased capacity for virulence and transmission (41). Last, the inherent multidrug resistance of B. cepacia (27) makes treatment very difficult and eradication almost impossible.
Strains of B. cepacia have recently been divided into a group of five distinct genomovars, referred to as the B. cepacia complex (43). The term genomovar denotes a group of strains with phenotypic similarity but genotypic uniqueness. Of the genomovars (numbered I through V), genomovars II, IV, and V have been renamed Burkholderia multivorans, Burkholderia stabilis, and Burkholderia vietnamiensis, respectively (43, 44). The majority of CF-transmissible or epidemic strains belong to B. cepacia genomovar III (43, 44).
Quorum sensing is a mechanism for regulating virulence factor production that involves the production and detection of signaling molecules (N-acyl homoserine lactones [AHLs]) (for reviews, see references 10, 15, and 28). The luxI gene family encodes an autoinducer synthase responsible for generation of the signaling molecule, which accumulates until a threshold concentration is attained. At this point, luxR-type transcriptional regulators bind the autoinducer and subsequently activate or repress the expression of target virulence genes. We have recently described a quorum-sensing system consisting of the cepI and cepR genes in a B. cepacia genomovar III strain (17). An N-acyl homoserine lactone (HSL) was purified from wild-type but not cepI mutant culture supernatants and identified as N-octanoyl-l-HSL (OHL). The initial characterization of cepI and cepR mutants revealed a lack of protease and OHL production, suggesting that protease and OHL production are positively regulated, and increased production of the siderophore ornibactin, suggesting that ornibactin is negatively regulated (17). A 20-bp lux box-like sequence that partially overlaps the −35 region of the putative cepI promoter was identified, suggesting that CepR binds to the cepI promoter to activate cepI expression (17). The majority of quorum-sensing systems are known to act as positive activators of target genes, although it is now recognized that LuxR-type proteins can act as transcriptional repressors (6, 45, 46). CepR appears to function as both a positive and a negative regulator of B. cepacia genes.
B. cepacia produces at least four types of siderophores: salicylic acid, pyochelin, cepabactin, and ornibactins (21, 22, 35, 38, 40), with salicylic acid and ornibactins being predominant in clinical isolates (3). Ornibactin-mediated iron acquisition has been shown to be important in both chronic and acute models of respiratory infection (36, 37). Pyochelin production was associated with 62% of clinical isolates, and its production correlated with severe pulmonary disease in CF patients (3, 35). Cepabactin is produced by few clinical isolates (11%) (3). Strain K56-2 produces ornibactins, salicylic acid, and negligible amounts of pyochelin (3). K56-2 cepR or cepI mutations resulted in hyperproduction of ornibactin but not salicylic acid.
The objectives of the present study were (i) to determine the role of cepR in the regulation of other B. cepacia siderophores and (ii) to further study the mechanisms of positive and negative regulation of cepI and cepR target genes in B. cepacia. Since the protease gene(s) has not been cloned and characterized, we have examined the role of cepR in the regulation of ornibactin biosynthesis and uptake genes and in the regulation of the cepI and cepR quorum-sensing genes.
The bacterial strains and plasmids used in this study are described in Table Table1.1. For genetic manipulations, Escherichia coli DH5α and B. cepacia K56-2 were grown at 37°C in Luria-Bertani (LB) broth (Life Technologies) or Bacto-Terrific broth (1.2% tryptone, 2.4% yeast extract, 0.4% glycerol, 17 mM KH2PO4, 72 mM K2HPO4) or on 1.5% LB agar plates. The following concentrations of antibiotics were used when necessary: 100 μg of ampicillin, 15 μg of tetracycline, 20 μg of gentamicin, and 1.5 mg of trimethoprim per ml for E. coli and 300 μg of tetracycline and 100 μg of trimethoprim per ml for B. cepacia. For ornibactin production and protease and Chrome Azurol S (CAS) assays, cultures were grown in succinate medium supplemented with ornithine (10 mM) (23) at 37°C. For pyochelin and salicylic acid assays, cultures were grown in deferrated CAA medium (35) at 37°C. For β-galactosidase assays of cepI-lacZ or cepR-lacZ fusions, cultures were grown in tryptic soy broth medium at 37°C. For β-galactosidase assays of pvdA-lacZ or pvdD-lacZ fusions, cultures were grown in TSBD-C medium (low iron) or TSBD-C supplemented with 5 μM FeCl3 (37). All medium components were products of Difco (Detroit, Mich.) unless otherwise stated.
The cepR gene in pSLR101 was insertionally inactivated by the introduction of the trimethoprim cassette from p34E-Tp into an internal PstI site (pSLR101-T). The inactivated cepR fragment was amplified by PCR using the primers pNOT19-1 (5′-GGCATGCGCAAGGCGATTAAGTTGG-3′) and pNOT19-2 (5′-GGCATGCCTTTATGCTTCCGGCTCG-3′), which incorporate SphI linkers, and cloned into the SphI site within the suicide vector pEX18Tc (13). This construct, designated pEXCEPR, was transferred to B. cepacia Pc224c or Pc715j by conjugation using either triparental matings with pRK2013 (7) as the mobilizing plasmid or biparental matings with E. coli SM10 as the donor strain (34). Transconjugants were plated onto Pseudomonas isolation agar plates containing 100 μg of trimethoprim per ml to select for single-crossover events in B. cepacia. Tpr transconjugants were streaked for isolated colonies on LB agar plates containing 100 μg of trimethoprim per ml and 5% sucrose to select for the loss of the vector sequence and screened to confirm tetracycline sensitivity. The insertional inactivation of cepR was confirmed with Southern hybridization analysis.
Molecular biology techniques were performed as generally described by Sambrook et al. (29). Restriction enzymes and oligonucleotide primers were purchased from Life Technologies. T4 DNA ligase was purchased from Promega Corp. (Madison, Wis.). Genomic DNA was isolated as described by Ausubel et al. (1). PCR products were cloned using the pCRR2.1 TOPO cloning system as recommended by the manufacturer (Invitrogen, Carlsbad, Calif.). Plasmids were introduced by electroporation using a Gene Pulser (Bio-Rad, Richmond, Calif.) according to the manufacturer's recommendations into E. coli and into B. cepacia as previously described (4).
Siderophore activity was measured using CAS assays (32). On CAS agar, siderophores remove iron from the CAS dye complex, resulting in a blue-to-orange color change in zones surrounding the colonies. The same dye complex was used to quantitate siderophore activity in culture supernatant fluid by measuring the increase in orange color at an optical density at 630 nm (OD630) as previously described (17). Ornibactins were purified by Sephadex LH-20 chromatography and quantitated using the CAS assay as previously described (3, 17). Pyochelin and salicylic acid were isolated and quantitated by thin-layer chromatography (TLC) as previously described (35, 38).
For protease assays, cultures were grown overnight, normalized to a turbidity of 0.3 at 600 nm, and spotted (3 μl) onto dialyzed brain heart infusion (D-BHI) agar containing 1.5% skim milk (39). The plates were incubated for 2 days at 37°C and examined for clear zones surrounding the colonies.
The cepI reporter strain K56-I2 does not produce autoinducer (OHL) or protease but responds to OHL produced by test strains inoculated at right angles to the reporter. The ability of test strains to produce autoinducers was detected by the restoration of protease production at the junction of the test and reporter strains on D-BHI–skim milk agar.
Agrobacterium tumefaciens A136 (pCF218)(pCF372) was used as a reporter to detect AHLs with 3-oxo-, 3-hydroxy-, and 3-unsubstituted side chains as previously described (33). Plasmid pCF218 contains traR, and pCF372 contains a traI-lacZ fusion (9, 49). AHLs were extracted from 20 ml of culture twice with equal volumes of acidified ethyl acetate (0.1 ml of glacial acetic acid per liter). Ethyl acetate was removed by rotary evaporation. The residue was resuspended in 2 ml of ethyl acetate, dried over N2 gas, and resuspended in 100 μl of acidified ethyl acetate. TLC bioassays were performed as described previously (33) with minor modifications. Samples were spotted onto a 20- by 20-cm C18 reversed-phase TLC plate (Whatman) and chromatographed using methanol-water (60:40, vol/vol) as a solvent. The plates were overlaid with a 20-ml A. tumefaciens A136 culture grown as follows. A 3-ml overnight culture was subcultured at a 1/100 dilution into 30 ml of LB broth and grown to log phase. Cells were pelleted by centrifugation, resuspended in 20 ml of AT medium (10.7g of KH2PO4, 160 mg of MgSO4·7H2O, 10 mg of CaCl2·2H2O, 5 mg of FeSO4·7H2O, 2 mg of MnSO4·7H2O, 2 g of (NH4)2SO4, 0.5% glucose, 1 liter of H20), and incubated for 30 min. This culture was then added to 150 ml of 0.7% AT agar and 60 μg of 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside per ml. TLC plates were incubated for 24 to 48 h at 30°C. Synthetic N-hexanoyl-HSL (HHL) and OHL (Sigma-Aldrich, Oakville, Ontario, Canada) were used as reference standards.
The lacZ-Gmr fragment from pZ1918G was ligated into the PstI site of pSLS200, which contains a 2.3-kb SphI-PstI fragment that encodes the entire cepI gene and the 5′ end of cepR, resulting in a transcriptional cepR-lacZ fusion (pSLR111). The cepI and lac vector promoters are in the orientation opposite to that of the cepR promoter, which is solely responsible for lacZ expression. A 769-bp fragment containing the cepI promoter was PCR amplified using the primers IN-CEPI (5′-GCGGATCCACCAGACGCCCATCTACCTGCTTCG-3′) and EX-CEPI (5′-GCCTGCAGGGCACAACGACGCCTATCATGC-3′), which incorporate BamHI and PstI linkers, respectively. The PCR product was cloned as a PstI-BamHI fragment into pUCP28T (pSLS210). The lacZ-Gmr fragment was ligated into the BamHI site of pSLS210, resulting in the cepI-lacZ transcriptional fusion (pSLS222). The lac vector promoter is in the opposite direction of the cepI promoter, which is solely responsible for lacZ expression.
β-Galactosidase activity was measured as previously described (26). Overnight cultures were subcultured in 50 ml of medium containing 100 μg of trimethoprim per ml to an OD600 of 0.01. One-milliliter aliquots were removed throughout growth and assayed for enzyme activity as previously described (26). β-Galactosidase activity is expressed in Miller units.
Outer membrane proteins were isolated as previously described (33) from cultures grown for 16 to 24 h (stationary phase) after overnight cultures were subcultured in fresh medium to an OD600 of 0.01. Protein samples (10 μg) were electrophoresed by sodium dodecyl sulfate–10% polyacrylamide gel electrophoresis (16).
Previously, the effects of a cepR::Tn5-OT182 insertion were determined in strain K56-2, which produces ornibactin and salicylic acid, negligible amounts of pyochelin, and no cepabactin (17). We attempted to construct cepR mutants by allelic exchange in other strains that produce pyochelin and cepabactin as well as ornibactin and salicylic acid in order to determine if pyochelin or cepabactin is regulated by CepR. CepR::tp mutants were successfully constructed with strains Pc715j and Pc224c, which produce ornibactin, salicylic acid, and pyochelin (3), and designated Pc715j-R1 and Pc224c-R1. It was not possible to obtain double-crossover cepR::tp recombinants using the following strains: ATCC 25416, ATCC 17759, P109, Pc22-20, 34192, and 34930, which also produce cepabactin (3).
The siderophore, protease, and autoinducer phenotypes of Pc715j-R1 were characterized (i) to confirm the role of CepR in autoinducer and protease production and (ii) to determine if pyochelin synthesis is regulated by CepR. CAS activity was measured in culture fluids throughout the growth of Pc715j and Pc715j-R1 (Fig. (Fig.1).1). Siderophore production is growth phase dependent in that siderophore activity is not observed until late log phase and is maximal in stationary phase. The cepR mutant had significantly greater CAS activity than the respective parent strain during stationary phase, indicating that one or more siderophores were being hyperproduced (Fig. (Fig.1).1). Consistent with this observation, Pc715j-R1 produced 50%-larger zones on CAS agar than those of Pc715j (Table (Table2).2). Since CAS activity is a measure of total siderophore activity, the yields of individual siderophores in culture supernatants were also quantitated. Pc715j-R1 produced parental levels of pyochelin and salicylic acid but nearly 100% more ornibactin than Pc715j (Table (Table2).2). These data suggest that CepR is a negative regulator of ornibactin synthesis but not of pyochelin or salicylic acid synthesis.
Interestingly, Pc715j-R1 produced significantly less protease activity than that of the parent strain Pc715j (Table (Table2),2), whereas K56-R2 had no detectable protease in this assay. It is possible that Pc715j produces more than one protease, only one of which is regulated by CepR. Pc715j did not produce detectable autoinducer activity in the bioassay with the cepI reporter K56-12 (Table (Table2),2), indicating that cepR expression is required for OHL production. The phenotypes observed in the cepR mutant Pc224c-R1 were similar to those of Pc715j-R1 (data not shown).
Using an A. tumefaciens traI::lacZ reporter assay, K56-2 and Pc715j produced detectable AHLs that comigrated on TLC plates with synthetic OHL and HHL (Fig. (Fig.2).2). The cepR mutants, K56-R2 and Pc715j-R1, and the cepI mutant, K56-I2, produced very small amounts of HHL, and OHL was only weakly detectable if the plates were incubated for several days. These data indicate that both the cepI and cepR genes are required for OHL synthesis.
Since ornibactin yields are increased in cepR mutants, we hypothesized that ornibactin biosynthetic genes are negatively regulated by CepR at the level of transcription. If our hypothesis is true, the hyperproduction of ornibactin would be a result of the increased expression of ornibactin biosynthetic genes in cepR mutants. Previously we described two Tn5-OT182 mutants, I117 and T10, which contain transposon insertions in ornibactin biosynthetic genes (37). Strain I117 has a Tn5-OT182 insertion in the pvdA gene that encodes l-ornithine N5-oxygenase, which catalyzes the hydroxylation of l-ornithine, forming the hydroxamate ligand in the peptide moiety of ornibactin. T10 contains an insertion in the pvdD gene that shares homology with nonribosomal peptide synthetases (37). The Tn5-OT182 mutants I117 and T10 contain chromosomally borne transcriptional lacZ fusions due to the presence of a promoterless lacZ gene within the transposon (37).
We constructed cepR mutants by allelic exchange in I117 and T10, which resulted in cepR pvdA and cepR pvdD mutants, respectively. β-Galactosidase activity was monitored throughout growth in low-iron medium and iron-supplemented medium (5 μM FeCl3) to examine the effects of cepR on pvdA and pvdD expression. During stationary phase there was a 20% increase in pvdA expression in I117-R21 in low-iron medium and a 62% increase in pvdA expression in iron-supplemented medium compared to levels in I117 (P < 0.02, t test for unpaired observations) (Table (Table3).3). There was a 35% increase in pvdD expression in T10-R10 compared to that in T10 in iron-supplemented medium during stationary phase (P < 0.02, t test for unpaired observations), but there was no significant difference between these strains in low-iron medium. The increase in pvdA expression correlates with the increased levels of CAS activity in the cepR mutants, K56-R2 and Pc715j-R1 (Table (Table2).2). (17).
The orbA gene encoding the 78.5-kDa ornibactin outer membrane receptor protein is located downstream of the pvdA gene (36). To determine if other genes involved in ornibactin-mediated iron acquisition were regulated by cepIR, outer membrane proteins were isolated from K56-I2 (cepI::tp) and K56-R2 and compared by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. There did not appear to be any difference in levels of expression of the ornibactin receptor in the cepR and cepI mutants (data not shown).
To determine the effects of CepR on cepI and cepR expression, transcriptional cepR-lacZ and cepI-lacZ fusions were constructed in pUCP28T and introduced into K56-2 and K56-R2. β-Galactosidase activity was monitored throughout growth (Fig. (Fig.3).3). In K56-2, the levels of expression of both cepR and cepI were low in early log phase but increased rapidly during late log and stationary phase (Fig. (Fig.3).3). This expression pattern is typical of quorum-sensing-regulated genes. Two important observations were made regarding cepIR expression in a cepR mutant background. In K56-R2, cepR expression increased twofold in the absence of functional cepR, indicating that cepR is autoregulated. cepI expression was abolished in the absence of functional cepR, indicating that cepI expression is positively regulated by CepR at the level of transcription (Fig. (Fig.3).3). Therefore, cepR is involved in regulation of both its own and cepI expression, in addition to that of genes involved in ornibactin biosynthesis.
The expression of the cepR-lacZ fusion in K56-2 was examined in cultures grown for 19 h in medium with and without iron supplementation to determine the effect of iron on cepR expression. There was 6,956 ± 1,276, 9,644 ± 472, and 10,178 ± 557 Miller units of β-galactosidase activity (values are means ± standard deviations) in cultures of K56-2 (pSLR211) grown in medium with 0, 5, and 10 μM FeCl3, respectively. The level of cepR expression was significantly higher in cultures supplemented with iron (P < 0.05). This result suggests that iron increases cepR expression, which may lead to higher levels of pvdA and pvdD repression in the parent strain. This result agrees with the observation that there is a greater effect of a cepR mutation on pvdA and pvdD expression in the presence of iron in the culture medium than in the absence of iron.
Analysis of multiple strains and their respective cepR mutants has shown that cepI and cepR comprise a functional quorum-sensing system that consistently regulates protease, ornibactin, OHL, and HHL production. Construction of gene replacement mutants in pyochelin-producing strains made it possible to determine that pyochelin biosynthesis is not regulated by cepIR. The effect of cepIR on siderophore production appears to be specific for ornibactin biosynthesis. Attempts to construct cepR::tp mutants with several strains previously reported to produce cepabactin were unsuccessful. Therefore, it was not possible to determine if cepabactin is regulated by cepIR. Many strains of the B. cepacia complex are not very amenable to genetic manipulations (19). K56-2 was the only previously reported strain to be successfully used for gene replacement studies (17, 19, 37). In this study, we have identified two additional B. cepacia complex strains that may be used for gene replacement studies.
We have determined that the hyperproduction of ornibactins in a cepR mutant is due to the increased expression of pvdA and possibly pvdD. These experiments suggest that cepR directly regulates ornibactin production at the level of transcription of ornibactin biosynthetic genes, although it is possible that this regulation occurs through an intermediate regulator. Since there are so few examples of negative regulation by quorum-sensing transcriptional regulators, it is difficult to understand the significance of a modest level of repression of ornibactin production by CepR. The negative regulation of ornibactin biosynthesis may serve as a fine-tuning mechanism to reduce energy expended synthesizing ornibactin at high cell densities in an environment in which the concentration of ornibactin is probably sufficient for iron acquisition.
Alternatively, cell density may serve as an additional environmental signal to limit the uptake of iron. The accumulation of intracellular iron can be deleterious due to the formation of hydroxyl radicals in the Haber-Weiss and Fenton reactions (2). When present in high concentrations, iron is the predominant environmental signal leading to the Fur-mediated repression of iron acquisition pathways. It is known that other conditions lead to the downregulation of iron acquisition pathways such as oxidative stress mediated by OxyR and SoxRS (48). The negative regulation of ornibactin production by cepR, when cells are growing at high cell densities, regardless of the iron concentration in the medium, may provide an additional level of control of siderophore-mediated uptake.
The roles of cepR in the control of the autoinducer synthase gene cepI and in its own expression were examined. Previously, we had been unable to detect OHL activity in stationary-phase culture supernatants of the cepR mutant K56-R2 (17). In this study using a different reporter assay, we determined that K56-2 and Pc715j produce both OHL and HHL. Production of these two AHLs has also been detected in another strain of B. cepacia using a Chromobacterium violaceum biosensor TLC assay (11). K56-2 and Pc715j cepI and cepR mutants produced little if any OHL and HHL, suggesting that cepI directs the synthesis of both AHLs and that cepI is positively regulated by CepR. This possibility was confirmed by demonstrating that there was no expression of a plasmid-borne cepI-lacZ fusion in a cepR mutant. Transcriptional fusions on high-copy-number plasmids were used for this study; however, Pesci et al. (24) have shown that Pseudomonas aeruginosa lasR expression patterns did not vary when lasR-lacZ fusions were expressed on high- or low-copy-number plasmids. This type of positive feedback control of autoinducer production is thought to provide a mechanism of signal amplification in which AHL signals are upregulated, allowing for more activation of the cognate R protein and subsequent activation of virulence genes. Although commonly employed, the positive control of autoinducer production is not an absolute requirement since several luxI homologs are not controlled by their respective R proteins (8).
Interestingly, the cepR gene was shown to negatively regulate its own expression, but the mechanism of repression is unknown. The repression effect of CepR on cepR expression would possibly be greater if expression was monitored from a single-copy chromosomal cepR-lacZ reporter since the effects of CepR would not be diluted by additional copies of the cepR promoter. Pantoea stewartii subsp. stewartii (formerly Erwinia stewartii) esaR is autoregulated in a similar pattern (45). von Bodman and Farrand proposed that the lux box overlapping the −10 region in the esaR promoter serves as an EsaR binding site that might occlude the RNA polymerase binding site and prevent its own expression (45). In fact, extracts of E. coli expressing esaR retard the mobility of a 61-bp oligonucleotide containing this lux box element (25). Although the significance of autoregulation of LuxR proteins is unknown, it is a possibility that the negative autoregulation of luxR genes may function to turn off the quorum-sensing response.
Using transcriptional lacZ fusions, we have provided evidence that cepR regulates pvdA and cepIR at the transcriptional level. To prove conclusively that CepR directly regulates these genes, further studies are required to purify CepR and examine its ability to bind to promoter regions of CepR-regulated genes.
This study was supported by a grant from the Canadian Cystic Fibrosis Foundation. S.L. is the recipient of a studentship award from the Alberta Heritage Foundation for Medical Research.
We thank M. B. Visser for excellent technical assistance and D. Storey and D. Erickson for advice on the TLC autoinducer bioassays.
University at Buffalo, South Campus
Buffalo, NY 14215