The CepIR quorum-sensing system contributes to the virulence of Burkholderia cenocepacia respiratory infections

P. A. Sokol1, U. Sajjan2, M. B. Visser1, S. Gingues1, J. Forstner2 and C. Kooi1

1 Department of Microbiology and Infectious Diseases, University of Calgary, Calgary, Alberta, Canada
2 Division of Structural Biology and Biochemistry, Hospital for Sick Children, Toronto, Ontario, Canada

Correspondence
P. A. Sokol
psokol{at}ucalgary.ca


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The cepIR genes encode an N-acyl homoserine lactone (AHL)-dependent quorum-sensing system consisting of an AHL synthase that directs the synthesis of N-octanoyl-L-homoserine lactone (ohl) and n-hexanoyl-L-homoserine lactone and a transcriptional regulator. The virulence of cepIR mutants was examined in two animal models. Rats were infected with agar beads containing Burkholderia cenocepacia K56-2, K56-I2 (cepI : : Tpr) or K56-R2 (cepR : : Tn5-OT182). At 10 days post-infection, the extent of lung histopathological changes was significantly lower in lungs infected with K56-I2 or K56-R2 compared to the parent strain. Intranasal infections were performed in Cftr(-/-) mice and their wild-type siblings. K56-2 was more virulent in both groups of mice. K56-I2 was the least virulent strain and was not invasive in the Cftr(-/-) mice. OHL was readily detected in lung homogenates from Cftr(-/-) mice infected with K56-2 but was only detected at levels slightly above background in a few mice infected with K56-I2. Lung homogenates from mice infected with K56-2 had significantly higher levels of the inflammatory mediators murine macrophage inflammatory protein-2, KC/N51, interleukin-1{beta} and interleukin-6 than those from K56-I2-infected animals. These studies indicate that a functional CepIR quorum-sensing system contributes to the severity of B. cenocepacia infections. A zinc metalloprotease gene (zmpA) was shown to be regulated by CepR and may be one of the factors that accounts for the difference in virulence between the cepI mutant and the parent strain.


Abbreviations: AHL, N-acyl homoserine lactone; CF, cystic fibrosis; cftr, cystic fibrosis transmembrane regulator; IL-1{beta}, interleukin-1{beta}; IL-6, interleukin-6; MIP-2, murine macrophage inflammatory protein-2; KC, KC/N51; OHL, N-octanoyl-homoserine lactone; p.i., post-infection; TNF{alpha}, tumour necrosis factor {alpha}


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Cystic fibrosis (CF) is an autosomal recessive disease caused by mutations in the cystic fibrosis transmembrane regulator (cftr) that encodes a chloride channel. CF is the most common lethal inherited disease in Caucasians. Absence or impaired function of CFTR leads to alterations in the lung environment including altered fluid and ion fluxes in respiratory epithelial cells, excessive mucous production in the airways, reduced mucociliary clearance and increased colonization of the lung by opportunistic pathogens [for reviews, see Rajan & Saiman (2002), Ratjen & Doring (2003), Schwiebert et al. (1998) and Tatterson et al. (2001)].

Although the major pathogen responsible for morbidity and mortality in CF patients is Pseudomonas aeruginosa (Govan & Deretic, 1996; Govan et al., 1996; Mahenthiralingam et al., 2002; Rajan & Saiman, 2002; Tatterson et al., 2001), in the past 20 years Burkholderia cepacia has emerged as an important pulmonary pathogen in this patient population (Mahenthiralingam et al., 2002; Mohr et al., 2001; Speert et al., 2002). Chronic colonization with B. cepacia is of a great concern in the CF community due to patient-to-patient transmissibility and inherent multi-drug resistance that makes eradication of B. cepacia almost impossible. Moreover, colonization with B. cepacia has been correlated with a poor clinical outcome sometimes resulting in death. The complexity of B. cepacia infections in CF patients has also increased due to the determination that the group of organisms that infect CF patients is not a single species but rather a group of at least nine closely related species or genomovars that are now referred to as the B. cepacia complex (Coenye et al., 2001; Vandamme et al., 2002). Burkholderia cenocepacia (Vandamme et al., 2003), formerly B. cepacia genomovar III, is the most common species of the B. cepacia complex that has been reported in CF infections. The majority of strains identified as epidemic or transmissible strains belong to this genomovar (LiPuma et al., 2001; Mahenthiralingam et al., 2002).

Several Gram-negative bacteria employ quorum-sensing-mediated cell-signalling systems that regulate various virulence factors in response to cell density (de Kievit & Iglewski, 2000; Williams et al., 2000). The LuxIR family of quorum-sensing systems is composed of N-acyl homoserine lactone (AHL) synthases and transcriptional regulators that may either activate or repress target genes. The AHLs are freely diffusible and bind to transcriptional regulators when they reach a sufficient threshold concentration generally correlating with cell density. The cepIR quorum-sensing system, originally identified in B. cenocepacia, is widely distributed throughout the B. cepacia complex (Gotschlich et al., 2001; Lewenza et al., 1999; Lutter et al., 2001). CepI is an AHL synthase that directs the synthesis of N-octanoyl-homoserine lactone (OHL) and minor amounts of N-hexanoyl-homoserine lactone. CepR encodes a transcriptional regulator that has both positive and negative regulatory properties. The CepIR quorum-sensing system has been shown to regulate expression of extracellular proteases, swarming motility and biofilm production (Huber et al., 2001; Lewenza et al., 1999, 2002). CepR has been shown to negatively regulate its own expression as well as the biosynthesis of the siderophore ornibactin via the pvdA gene (Lewenza & Sokol, 2001). Riedel et al. (2003) using a proteomics approach demonstrated that a number of proteins are differentially expressed between B. cenocepacia strain H111 and a cepI mutant. N-terminal amino acid sequence analysis identified a few proteins up-regulated in the wild-type strain including a putative superoxide dismutase, a peroxidase and a possible ABC transporter system. In B. cepacia genomovar I, CepR has been shown to positively regulate a secreted polygalacturonase involved in onion-rot pathogenicity and to have a negative effect on the expression of the stationary-phase sigma factor, rpoS (Aguilar et al., 2003).

A B. cepacia genomovar I cepI mutant was less virulent in an onion-rot model as evident by attenuated tissue maceration compared to the wild-type. Complementation of the cepI mutant with cepIR in trans enhanced the virulence above that of the wild-type level indicating a role for the cepIR quorum-sensing mechanism in onion-rot pathogenicity (Aguilar et al., 2003). A functional cepIR system was also shown to be required for efficient killing of the nematode Caenorhabditis elegans by B. cenocepacia H111 (Kothe et al., 2003). Nematode killing appears to involve the production of a diffusible extracellular toxin, which has not yet been identified in the B. cepacia complex.

Several animal models have been developed to investigate the virulence of the B. cepacia complex in respiratory infections. These include agar bead models in rats and mice (Cieri et al., 2002; Sokol et al., 1999, 2000) and agar beads or intranasal infection models in CF mice (Davidson et al., 1995; Sajjan et al., 2001). The chronic agar bead model (Cash et al., 1979) originally developed to mimic P. aeruginosa respiratory infections in CF patients has been used to demonstrate that iron acquisition via the siderophore ornibactin is necessary for persistence of B. cenocepacia chronic infections (Sokol et al., 1999, 2000). Recently, we have used this model to demonstrate that a zinc metalloprotease contributes to the virulence of some strains of B. cenocepacia (Corbett et al., 2003). A modified agar bead model in mice has been used to compare invasiveness of members of the B. cepacia complex during acute infections (Cieri et al., 2002) and to demonstrate that Type III secretion is important for virulence (Tomich et al., 2003). CFTR-deficient (knockout) mice have also been used to study the pathogenesis of lung disease in experimental infections due to B. cepacia. Chronic models of B. cepacia infection have been developed in CFTR-deficient mice by repeated intranasal inoculation or repeated exposure to aerosolized organisms (Davidson et al., 1995; Sajjan et al., 2001). Recently, a CFTR-deficient mouse model has been developed that does not require repeated inoculation of B. cenocepacia. Short-term infections can be established in 10- to 12-week-old mice following a single intranasal inoculation (Sajjan et al., 2002). In the present study, we report that the CepIR quorum-sensing system contributes to the pathogenesis of B. cenocepacia strain K56-2 respiratory infections, using two different animal models, the short-term intranasal colonization mouse model and the chronic agar bead infection model in rats.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Strains, growth conditions and DNA manipulations.
The bacterial strains and plasmids used in this study are listed in Table 1. DNA manipulations were performed using standard techniques as described by Sambrook et al. (1989). B. cenocepacia cultures were routinely grown at 37 °C in Luria–Bertani broth (LB) (Invitrogen Life Technologies) or on LB solidified with 1·5 % agar unless otherwise noted. Medium was supplemented with 200 µg tetracycline ml-1 or 100 µg trimethoprim ml-1 to select for the introduction of plasmids into B. cenocepacia. OHL was extracted from cultures grown in trypticase soy broth (TSB; Difco Laboratories). Cultures were grown overnight in TSBD-C medium (Ohman et al., 1980) for infections using the agar bead model and in TSB medium for mouse infections. Protease activity was monitored using D-BHI-milk medium (Sokol et al., 1979).


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

 
Agrobacterium tumefaciens strains were routinely grown at 30 °C in LB or on LB solidified with 1·5 % agar. Medium was supplemented with 25 µg kanamycin ml-1, 4·5 µg tetracycline ml-1, 50 µg spectinomycin ml-1 or 10 µg rifampicin ml-1 as required. When necessary, liquid medium was supplemented with 3 µg tetracycline ml-1. For preparation of A. tumefaciens electrocompetent cells, cultures were grown in MG/L broth (Chilton et al., 1974) and prepared as described by Cangelosi et al. (1991). Plasmids were introduced into A. tumefaciens as described previously (Cangelosi et al., 1991).

Construction of a traIluxCDABE reporter.
To construct pMV26, primers traI pro2 (5'-TTCTGGTGTTGGTATTGGTC-3') and traIproIn2 (5'-AGTAGTTCGCCAGTTAATAG-3') were used to amplify -143 to +68 of the traI promoter region from pCF372, a traIlacZ fusion (Fuqua & Winans, 1996). The Psa origin of replication was also amplified from pCF372 using primers Psa-PacF (5'-CGCTCATAGGGCTTAATTAACCAACGTTTTA-3') and Psa-PacR (5'-CATACTACAATTAATTAACAGAGCCATC-3'), which incorporate PacI restriction sites. The pSC101 origin of replication in the promoterless luxCDABE vector pCS26-Pac (Bjarnason et al., 2003) was replaced with a 2·41 kb Psa origin PacI fragment to allow for replication in A. tumefaciens, forming pCS26-Psa. A 351 bp BamHI–XhoI pCRtraI fragment containing the traI promoter was then cloned into pCS26-Psa to form pMV26.

Agar bead model.
Groups of 20 male Sprague–Dawley rats (150–170 g; Charles River Canada) were tracheostomized under anaesthesia and inoculated with approximately 1·5x105 c.f.u. of K56-2 or the mutant strains embedded in agar beads as described previously (Cash et al., 1979). On days 3 and 10 post-infection (p.i.), the lungs from five animals per group were removed aseptically and homogenized (Polytron Homogenizer; Brinkman Instruments) in 3 ml of PBS, serially diluted and plated onto trypticase soy agar and incubated at 37 °C to determine the number of bacteria present in the lung. Lung homogenates from K56-R2- or K56-I2-infected animals were also plated onto trypticase soy agar containing tetracycline (300 µg ml-1) or trimethoprim (100 µg ml-1), respectively, to confirm that the mutations were stable throughout the course of the infections. In a representative experiment, approximately 50 colonies from each animal were also tested for protease production on skim milk agar to confirm that the cepI or cepR phenotype was maintained during the infection. Lungs from five additional animals per group were examined for quantitative histopathological changes as described previously (Sokol & Woods, 1984) with the following modifications. The lung sections were scanned using an Epson 1650 scanner, and areas of inflammation were digitized with SCION IMAGE software (http://www.scioncorp.com) and reported as the percentage of lung inflammation.

Intranasal infection model.
Liquid-fed Cftr(-/-) knockout mice (Kent et al., 1996) and their littermate wild-type Cftr(+/+) controls, age 10–12 weeks, were infected with approximately 1x108 c.f.u. of K56-2, K56-I2 or K56-R2 in 40 µl PBS. The inoculum was instilled drop-wise intranasally and allowed to be aspirated into the lungs as described previously (Sajjan et al., 2001). Mice were observed for 3 days. At 72 h p.i., lungs and spleens were harvested, weighed and homogenized as described previously to determine the number of bacteria present (Sajjan et al., 2001). If mice died prior to the 72 h, the lungs and spleens were recovered, homogenized and plated. One millilitre of lung homogenate from each mouse was immediately mixed with complete protease inhibitor cocktail (Roche Diagnostics), centrifuged and the supernatant removed and stored at -70 °C for cytokine assays. Cftr(-/-) mice were also infected with approximately 5x106 c.f.u. of K56-2 or K56-I2. Mice were observed for 5 days, after which lungs and spleens were harvested as above.

Lungs were graded for severity of inflammation by determining the ratio of total wet weight of lungs to body weight. A ratio of less than 0·02 was associated with normal or mild inflammation, 0·021–0·03 indicated moderate inflammation and greater than 0·03 (range 0·031–0·05) indicated severe inflammation. Lungs graded as severe had white patches and were more friable indicating consolidation of the lung upon gross pathology examination. Normal lungs appeared pink with more elasticity. Lungs with small patches of consolidation were classified as mild or moderate. The gross pathology observations were found to correlate with the lung-to-body weight ratios.

Cytokine assays.
Murine tumour necrosis factor {alpha} (TNF{alpha}), interleukin-1{beta} (IL-1{beta}), interleukin-6 (IL-6), interferon {gamma} (IFN-{gamma}) (Biosource International), KC/N51 (KC) and macrophage inflammatory protein-2 (MIP-2) (R&D Systems) were measured by ELISA. Lung homogenates were diluted 1 : 10, 1 : 50 or 1 : 100 in diluent buffer and assayed in duplicate according to the manufacturers' instructions. The limits of detection for the cytokine assays were less than 2·0 pg ml-1 for KC, less than 1·5 pg ml-1 for MIP-2, less than 7 pg ml-1 for IL-1{beta}, less than 1 pg ml-1 for IFN-{gamma} and less than 3 pg ml-1 for IL-6 and TNF{alpha}.

OHL extraction and purification.
AHLs were extracted from 20 h cultures of K56-I2(pSL225), which carries cepI on a high-copy plasmid, with equal volumes of acidified ethyl acetate and dried as described previously (Lewenza et al., 1999). To purify OHL, the dried extract from a 500 ml culture was resuspended in 2 ml of deionized water and subjected to reversed-phase FPLC using a Sephasil Peptide C18 12 µm ST 4·6/250 column fitted to the AKTA Explorer 900 FPLC system (Amersham Pharmacia Biotech). Analysis of the resulting chromatogram was performed using the UNICORN version 3.12.02 program (Amersham Pharmacia Biotech). The AHLs were separated with a 0–100 % acetonitrile gradient at a flow rate of 1·0 ml min-1, and the eluent was analysed using an Agrobacterium TLC assay (Lutter et al., 2001; Shaw et al., 1997) to identify fractions containing OHL and N-hexanoyl-homoserine lactone. The concentration of OHL was estimated by comparison with an OHL standard (Fluka) using the same overlay assay. Fractions containing OHL were pooled and stored at -20 °C.

Detection of OHL in lung tissue.
OHL was extracted from mouse or rat lung homogenates as described with modifications (Erickson et al., 2002). Two-hundred microlitres of mouse lung or 2 ml of rat lung homogenate was extracted three times with dichloromethane (2 : 1, v/v). Samples were centrifuged (2000 r.p.m., 10 min) and the organic layer was removed and pooled. Solvent was removed by evaporation and the residue was resuspended in 10 µl acetonitrile. An overnight culture of A. tumefaciens A136(pCF218)(pMV26) was used for detection of OHL. Assays were performed in 96-well microtitre plates (Costar; Corning). Ten microlitres of extract, 10 µl of the A. tumefaciens A136(pCF218)(pMV26) culture and 80 µl LB were mixed, incubated at 30 °C with shaking and luminescence was measured at selected time intervals using a Wallac Trilux luminescence counter (Perkin-Elmer Life Sciences). Lung homogenates from uninfected animals were used to determine background levels of luminescence. Synthetic OHL (Fluka) was used to prepare a standard curve.

Construction of a zmpA : : lacZ fusion.
The zmpA : : lacZ fusion was constructed by inserting the SmaI fragment of pZ1918G containing a lacZ–Gmr cassette (Schweizer, 1993) into the StuI site of zmpA contained on a 2·6 kb PstI fragment cloned into pUCP28T (Schweizer et al., 1996). This plasmid was designated pSG208. The same zmpA : : lacZ fusion was subsequently cloned into pUCP26 (Schweizer et al., 1996) and designated pSG206. Expression of the fusion was monitored by measuring {beta}-galactosidase activity as described previously (Platt et al., 1972).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Effects of cepIR mutations on virulence in a chronic respiratory infection model
Sprague–Dawley rats were infected with agar beads containing K56-2, K56-I2 or K56-R2 to establish chronic respiratory infections (Cash et al., 1979; Sokol et al., 1999). On days 3 and 10 p.i., lungs were removed and subjected to quantitative bacteriology and histopathological analysis. Although the number of bacteria recovered from the lungs of animals infected with the mutants compared to the parent strain was slightly lower on both day 3 and day 10 (Table 2) the difference was not significant. The numbers of bacteria in the lungs increased by at least 1·5 logs by day 3 and persisted for the 10-day period. These data indicate that mutations in either cepI or cepR do not affect the ability of K56-2 to establish and maintain chronic infections in this model.


View this table:
[in this window]
[in a new window]
 
Table 2. Effect of cepI or cepR mutations on quantitative bacteriology in the agar bead infection model

Results are shown as the mean±SD (x106), and are based on five animals per group for day 3 p.i. and four animals per group for day 10 p.i.

 
Despite the fact that there was no significant difference in the number of bacteria present in the lungs, the extent of lung histopathological changes on day 10 p.i. was significantly lower in rats infected with either K56-I2 or K56-R2 compared to the parent strain (Table 3). There were no qualitative differences in the pathological changes observed (data not shown). These data indicate that the cepIR system contributes to the maximum virulence of strain K56-2 probably by regulating production of extracellular virulence factors that result in increased lung injury.


View this table:
[in this window]
[in a new window]
 
Table 3. Effect of cepI or cepR mutations on histopathology in the agar bead infection model

Results are shown as the mean percentage inflammation±SD.

 
Effects of cepIR mutations on virulence in a mouse intranasal infection model
Cftr(-/-) mice or their Cftr(+/+) littermate controls were infected intranasally with approximately 1x108 c.f.u. of K56-2, K56-I2 or K56-R2. Strain K56-2 was more virulent than K56-I2 and K56-R2 in the wild-type mice since both mice infected with the parent strain died by day 3 p.i. (Table 4). The lungs of these mice when examined for gross pathology changes appeared to have severe inflammation whereas only one mouse infected with K56-I2 and none of the mice infected with K56-R2 had severe lung inflammation. Bacteria were cultured from spleens of both surviving mice infected with K56-2 but from only one of the K56-I2-infected mice.


View this table:
[in this window]
[in a new window]
 
Table 4. Comparison of virulence of B. cenocepacia K56-2, K56-I2 and K56-R2 in CF and wild-type mice infected intranasally with 1x108 c.f.u.

 
K56-2 was also more virulent than K56-I2 in the Cftr(-/-) mice (Table 4). Although none of these mice died during the 3-day period, two of three appeared ill and had severe lung inflammation. All three had bacteria present in the spleen. K56-R2 was also virulent in these mice, with invasion and severe lung inflammation evident in two of three mice. One of the K56-R2-infected mice that died, interestingly, had only mild lung inflammation and no bacteria recovered from the spleen. In contrast to the parent and the cepR mutant, infection with K56-I2 resulted in only mild or moderate lung inflammation in four of five mice, and the infection was not invasive in any of the Cftr(-/-) mice. There were fewer bacteria recovered from the lungs of mice infected with K56-I2 than the other strains.

The virulence of K56-2 and K56-I2 was also compared in Cftr(-/-) mice using a lower inoculum than in the previous experiment. Mice were infected with approximately 5x106 c.f.u. and observed for 5 days, since the mice tolerated this dose better than the higher inoculum. There were insufficient wild-type littermate controls available so they were not included in this experiment. K56-I2 was significantly less virulent than the parent strain as evident by the approximately 3-log difference in bacteria recovered from the lungs and the degree of lung inflammation. The absence of bacteria recovered from the spleen indicated that K56-I2 was not invasive in any of the Cftr(-/-) mice (Table 5).


View this table:
[in this window]
[in a new window]
 
Table 5. Comparison of virulence of B. cenocepacia K56-2 and K56-I2 in CF mice infected intranasally with 5x106 c.f.u.

 
To confirm that differences in virulence in this model between K56-2 and K56-I2 were due to the absence of OHL production, lung homogenates from the infected animals were extracted with dichloromethane and analysed for OHL using a traIluxCDABE reporter system in A. tumefaciens. This reporter system is able to detect femtomolar levels of OHL (M. B. Visser, C. E. Chambers & P. A. Sokol, unpublished observations). OHL was detected in lung homogenates from all three Cftr(-/-) mice infected with 1x108 c.f.u. of K56-2 but was not detectable in the four mice infected with K56-I2 or the one surviving mouse infected with K56-R2. The mean±SD c.p.m. value minus the background for the K56-2-infected animals was 30 714±26 067 (range=5087–57 200) at 8 h, which was the time point with the highest activity in this assay. If all of this activity was due to OHL, the concentration would be approximately 0·0025 pM as determined from a standard curve prepared with synthetic OHL. The c.p.m. values for all the K56-I2-infected lungs were below those obtained with the uninfected control (13 573 c.p.m.) and the vector control strain (19 391 c.p.m.).

In animals infected with 1x106 c.f.u., OHL activity was detected in three of three animals infected with K56-2 with a mean±SD c.p.m. value minus the background of 91 796±69 497 (range=38 532–170 409) at 15 h, which was the time of maximum activity in this assay. Some activity was detected in three of five animals infected with K56-I2, although the values obtained were much lower than in animals infected with K56-2 and ranged from 5344 to 29 102 c.p.m. (mean±SD=9804±12 322). Due to the large standard deviation in these groups of animals the differences were not quite significant. Although the c.p.m. values appear higher in the animals infected with 5x106 than in animals infected with 1x108 c.f.u. the actual amount of OHL detected is lower. The background c.p.m. value for extracts from uninfected lung in this assay was 38 289, therefore the c.p.m. values from the K56-I2-infected lungs were not even twice the background level. The mean c.p.m. values detected are equivalent to less than 0·0012 pmol OHL.

The levels of selected cytokines were also measured in lung homogenates from the surviving mice. There were no significant differences between the amounts of KC, MIP-2, TNF{alpha}, IL-1{beta} or IL-6 in the lungs from Cftr(-/-) mice infected with 1x108 c.f.u. of K56-2 or K56-I2 (data not shown). In the mice receiving the lower inoculum, however, the levels of KC, MIP-2, IL-1{beta} and IL-6 were significantly higher in lungs from animals infected with K56-2 compared to K56-I2 (Table 6). There was no difference in the amount of TNF{alpha}. Interestingly, the levels of IFN-{gamma} were higher in mice infected with K56-I2 than in mice infected with K56-2. It was not possible to compare cytokine levels in Cftr(-/-) and Cftr(+/+) mice infected with K56-2 due to the high mortality rate; however, we were able to compare cytokine levels in these mice infected with 1x108 c.f.u. of K56-I2. The levels of MIP-2, IL-6 and TNF{alpha} were significantly higher in lungs from Cftr(-/-) mice than from Cftr(+/+) mice (Table 7).


View this table:
[in this window]
[in a new window]
 
Table 6. Cytokine levels in lungs of Cftr(-/-) mice infected with 5x106 c.f.u. of B. cenocepacia K56-2 or K56-I2

Results are expressed as the mean log values±SD.

 

View this table:
[in this window]
[in a new window]
 
Table 7. Comparison of cytokine levels in Cftr(-/-) and Cftr(+/+) mice infected with 1x108 c.f.u. of B. cenocepacia K56-I2

Results are expressed as the mean log values±SD.

 
Effect of cepI or cepR mutations on protease expression
A zinc metalloprotease gene (zmpA) which encodes the previously described PSCP protease (McKevitt et al., 1989) has recently been cloned and characterized in our laboratory (Corbett et al., 2003). A K56-2 zmpA mutant was less persistent and caused less lung pathology than the wild-type strain in the rat agar bead infection model. Since the CepIR quorum-sensing system has previously been shown to positively regulate extracellular protease activity, the effects of cepI or cepR mutations on expression of zmpA in K56-2 were examined. Transcriptional fusions were constructed by inserting a promoterless lacZ cassette into the zmpA gene. Expression of the zmpA : : lacZ fusion in K56-I2, K56-R2 and K56-2 was examined over 24 h of growth with and without the addition of OHL (Fig. 1). There was a significant decrease in zmpA : : lacZ expression in the mutants compared to the parent strain, which was restored by the addition of OHL to the K56-I2 cultures. Addition of OHL to K56-I2(pSG206) resulted in an approximately sixfold increase in {beta}-galactosidase activity at 18 h but did not restore expression to parental levels. The K56-R2(pSG208) culture did not respond to the addition of OHL. Interestingly, addition of OHL to K56-2(pSG208) resulted in a twofold increase in expression, suggesting that the production of OHL by the wild-type cepI gene did not saturate the regulatory mechanism governing the expression of zmpA. These experiments indicate that zmpA is regulated by CepIR and that differences in protease expression may be one of the factors that contribute to the differences in virulence between these mutants and the parent strain, although it is likely that a number of virulence factors are involved.



View larger version (24K):
[in this window]
[in a new window]
 
Fig. 1. Expression of zmpA : : lacZ fusions in K56-2, K56-I2 and K56-R2. Overnight cultures were diluted 1 : 200 in 10 ml of fresh medium. Cultures were grown in the presence (solid symbols) or in the absence (open symbols) of 5 nM OHL, and {beta}-galactosidase activity was determined at selected intervals. K56-2(pSG208), triangles; K56-I2(pSG206), diamonds; K56-R2(pSG208), circles; K56-2(pUCP28T), squares.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Using both chronic and acute infection models, we have demonstrated a role for the CepIR quorum-sensing system in the pathogenesis of B. cenocepacia lung disease. Rats infected with K56-I2 or K56-R2 were found to have similar numbers of bacteria recovered from the lungs to K56-2-infected animals, indicating that CepIR is not required for establishment or persistence in this model. A significant decrease in lung histopathology despite similar numbers of bacteria in the lungs could be due to decreased expression of extracellular virulence factors by the cepI and cepR mutant. One factor that has been shown to play a role in the virulence of K56-2 in lung injury in this model is the zinc metalloprotease PSCP (Corbett et al., 2003). In this study, we have determined that the zmpA gene encoding this protease is regulated by cepIR and therefore may be one of the factors that accounts for the difference in virulence between the mutants and the parent strain.

A similar difference in histopathology, quantified by measuring the size of lung abscesses, was reported between P. aeruginosa PAO1 and a rhlI lasI double mutant in the agar bead model (Wu et al., 2001). The decrease in the extent of histopathological changes in lungs infected with the mutant compared to the parent was significant on days 14 and 28 p.i. but not day 7. These data, together with this study, suggest that in both P. aeruginosa and B. cenocepacia infections, quorum-sensing mechanisms have a greater role during the later stages of lung infection in this model. rhlI and lasI mutants have also been shown to be less virulent in an acute pulmonary infection model in neonatal mice (Pearson et al., 2000). In this study both single and double mutants in rhlI or lasI were shown to cause pneumonia and bacteraemia less frequently than the parent strain.

In the agar bead model the bacteria are introduced directly into the lung through the trachea, bypassing the normal route of colonization. The mechanical entrapment in the beads prevents their clearance from the lung. CepR-regulated virulence factors potentially involved in adherence and colonization may not contribute to the difference in virulence observed between the cepR or cepI mutants and the parent strain in this model. Although infections with some strains of the B. cepacia complex in C57/Black 6 mice led to splenic invasion (Cieri et al., 2002), we have not recovered strain K56-2 from the spleens of infected rats (data not shown), suggesting that the infection remains confined to the lungs.

The intranasal infection model may be able to detect differences in virulence due to mutations that affect the ability to colonize mucosal surfaces or invade the bloodstream. K56-2 was lethal in the wild-type mice infected with 1x108 c.f.u., whereas the cepR mutant was less virulent and was not recovered from the spleen indicating that it did not result in an invasive infection. Interestingly, the CF mice tolerated K56-2 better than the wild-type mice. Although the mice infected with K56-2 had greater than 109 c.f.u. in the lung and had bacteria in the spleen, no mice died by day 3 p.i. Mice infected with the cepI mutant had approximately 2 logs less bacteria in the lungs and no bacteria in the spleen. The number of K56-I2-infected mice with severe lung inflammation was also less than in mice infected with K56-2 or K56-R2. Although this experiment suggested that the cepI mutant may be less virulent in CF mice, the degree of virulence due to the high inoculum made it difficult to distinguish differences between the strains. When CF mice were infected with 5x106 c.f.u., there was a marked difference in virulence between K56-2 and K56-I2 as indicated by the lack of invasion and the recovery of approximately 3 logs fewer bacteria from the lungs. K56-R2 was not examined at the lower inoculum due to the limited availability of Cftr(-/-) mice of the appropriate age.

Using a sensitive reporter assay employing a traIluxCDABE fusion that responds to OHL, we were able to detect OHL in lung homogenates of CF mice infected with K56-2 but not K56-I2 in animals inoculated with 1x108 c.f.u. We were able to detect OHL in the mice infected with the lower inoculum but were approaching the limits of sensitivity of the assay in these animals and c.p.m. values were close to background levels. We attempted to detect OHL in lung homogenates from rats infected with approximately 1x105 c.f.u. via the agar bead model but were not able to do so, presumably due to the lower numbers of bacteria present (data not shown). Expression of AHLs by B. cepacia or P. aeruginosa in lung infection models has previously been reported using co-infection experiments with reporter strains containing plasmids with green fluorescent protein (GFP) fusions that respond to AHLs (Riedel et al., 2001; Wu et al., 2001). In these studies, evidence of AHL production was demonstrated by visualization of the GFP-expressing reporter by confocal scanning microscopy. The development of the new lux-based reporter system makes it possible to quantify relative amounts of AHLs with sensitivity in the picomolar to femtomolar range depending on the AHL.

Sajjan et al. (2001) reported previously that infection with strain BC7 resulted in elevated levels of the chemokine KC in bronchoalveolar lavage fluid of CF mice compared to wild-type mice. No differences were observed in levels of TNF{alpha} or MIP-2. In the present study, infection with K56-I2 resulted in significantly higher levels of MIP-2, IL-6 and TNF{alpha} in lung homogenates from CF mice compared to wild-type but no differences in IL-1{beta} or KC. The difference in cytokine profiles observed between the two studies may be due to strain variation, since K56-2 is generally more virulent than BC7 (unpublished observations), or to differences in the age of the mice. In a study comparing P. aeruginosa agar bead infections in CF and wild-type mice, elevated levels of TNF{alpha}, MIP-2 and KC were reported in bronchoalveolar lavage fluid from CF mice compared to the wild-type (van Heeckeren et al., 1997), which also suggests that there is strain variation in the stimulation of inflammatory mediators.

Infection with K56-2 resulted in elevated levels of the cytokines MIP-2, IL-6, IL-1{beta} and KC in the lungs compared to infection with K56-I2. The increase in pro-inflammatory cytokines correlates with the increased inflammation observed in these animals and may be due to either CepIR regulation of virulence factors, such as the zinc metalloprotease, or direct stimulation of the host response by AHLs. N-(3-Oxododecanoyl) homoserine lactone (OdDHL) was shown to induce the production of several inflammatory chemokines and cytokines when injected directly into the skin of mice, including MIP-2, MIP-1{beta}, IL-6 and IL-1{alpha} (Smith et al., 2002). It has also been shown to activate IL-8 production in human lung fibroblasts and epithelial cells (Smith et al., 2001). OdDHL has also been reported to suppress leukocyte proliferation and inhibit lipopolysaccharide-induced secretion of TNF{alpha} and IL-12 (Chhabra et al., 2003; Telford et al., 1998). A comparison of a range of synthetic analogues of OdDHL suggests that AHLs with 11–13 carbon side chains are optimal for immune suppressive activity (Chhabra et al., 2003). Although these studies suggest that long-chain AHLs have more activity than compounds with short side chains (Chhabra et al., 2003; Smith et al., 2002; Telford et al., 1998), the potential for a direct effect of OHL or N-hexanoyl-homoserine lactone on inflammation cannot be ruled out.

This study together with those of Kothe et al. (2003) and Aguilar et al. (2003) indicate that the CepIR quorum-sensing system contributes to virulence in a wide range of hosts including plants, nematodes and murine species. Although the virulence factors regulated by CepR that play a role in virulence in the various models need to be determined, the importance of this cell-signalling mechanism in virulence has been established. The virulence of the cepR and cepI mutants was comparable in the agar bead model and in the wild-type mice; however, the cepR mutant was more virulent than the cepI mutant in the Cftr(-/-) mice. Interestingly, a cepR mutant was also more virulent than the cepI mutant in the C. elegans model (Kothe et al., 2003). CepR functions as both an activator and a repressor of target genes (Lewenza & Sokol, 2001; Riedel et al., 2003) and therefore may be involved in both up-regulation and down-regulation of virulence factors. The importance of different complements of virulence factors in the model systems used may account for the observed differences in the virulence of the cepR mutant. For example, a cepR mutant produces elevated amounts of ornibactin which might contribute to its increased virulence. An additional LuxR homologue has been identified in the B. cenocepacia genome (A. Baldwin, P. A. Sokol, J. Parkhill & E. Mahenthiralingam, unpublished data) and this protein may compensate for the cepR mutation. This possibility is currently being investigated in our laboratory.


   ACKNOWLEDGEMENTS
 
These studies were supported by grants from the Canadian Cystic Fibrosis Foundation and the Special Initiative in Memory of Michael O'Reilly funded by the Canadian Cystic Fibrosis Foundation and the Canadian Institutes of Health Research to P. A. S. S. G. is the recipient of a Natural Sciences and Engineering Research Council of Canada Post Graduate Scholarship. The authors thank Dr D. E. Woods for analysis of histopathology in the agar bead model and C. E. Chambers and C. R. Corbett for helpful discussions.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Aguilar, C., Bertani, I. & Venturi, V. (2003). Quorum-sensing system and stationary-phase sigma factor (rpoS) of the onion pathogen Burkholderia cepacia genomovar I type strain, ATCC 25416. Appl Environ Microbiol 69, 1739–1747.[Abstract/Free Full Text]

Bjarnason, J., Southward, C. M. & Surette, M. G. (2003). Genomic profiling of iron-responsive genes in Salmonella enterica serovar typhimurium by high-throughput screening of a random promoter library. J Bacteriol 185, 4973–4982.[Abstract/Free Full Text]

Cangelosi, G. A., Best, E. A., Martinetti, G. & Nester, E. W. (1991). Genetic analysis of Agrobacterium. Methods Enzymol 204, 384–397.[Medline]

Cash, H. A., Woods, D. E., McCullough, B., Johanson, W. G., Jr & Bass, J. A. (1979). A rat model of chronic respiratory infection with Pseudomonas aeruginosa. Am Rev Respir Dis 119, 453–459.[Medline]

Chhabra, S. R., Harty, C., Hooi, D. S., Daykin, M., Williams, P., Telford, G., Pritchard, D. I. & Bycroft, B. W. (2003). Synthetic analogues of the bacterial signal (quorum sensing) molecule N-(3-oxododecanoyl)-L-homoserine lactone as immune modulators. J Med Chem 46, 97–104.[CrossRef][Medline]

Chilton, M. D., Currier, T. C., Farrand, S. K., Bendich, A. J., Gordon, M. P. & Nester, E. W. (1974). Agrobacterium tumefaciens DNA and PS8 bacteriophage DNA not detected in crown gall tumors. Proc Natl Acad Sci U S A 71, 3672–3676.[Abstract]

Cieri, M. V., Mayer-Hamblett, N., Griffith, A. & Burns, J. L. (2002). Correlation between an in vitro invasion assay and a murine model of Burkholderia cepacia lung infection. Infect Immun 70, 1081–1086.[Abstract/Free Full Text]

Coenye, T., Vandamme, P., Govan, J. R. & LiPuma, J. J. (2001). Taxonomy and identification of the Burkholderia cepacia complex. J Clin Microbiol 39, 3427–3436.[Free Full Text]

Corbett, C. R., Burtnick, M. N., Kooi, C., Woods, D. E. & Sokol, P. A. (2003). An extracellular zinc metalloprotease gene of Burkholderia cepacia. Microbiology 149, 2263–2271.[Abstract/Free Full Text]

Davidson, D. J., Dorin, J. R., McLachlan, G., Ranaldi, V., Lamb, D., Doherty, C., Govan, J. & Porteous, D. J. (1995). Lung disease in the cystic fibrosis mouse exposed to bacterial pathogens. Nat Genet 9, 351–357.[Medline]

de Kievit, T. R. & Iglewski, B. H. (2000). Bacterial quorum sensing in pathogenic relationships. Infect Immun 68, 4839–4849.[Free Full Text]

Erickson, D. L., Endersby, R., Kirkham, A., Stuber, K., Vollman, D. D., Rabin, H. R., Mitchell, I. & Storey, D. G. (2002). Pseudomonas aeruginosa quorum-sensing systems may control virulence factor expression in the lungs of patients with cystic fibrosis. Infect Immun 70, 1783–1790.[Abstract/Free Full Text]

Fuqua, C. & Winans, S. C. (1996). Conserved cis-acting promoter elements are required for density-dependent transcription of Agrobacterium tumefaciens conjugal transfer genes. J Bacteriol 178, 435–440.[Abstract]

Gotschlich, A., Huber, B., Geisenberger, O. & 11 other authors (2001). Synthesis of multiple N-acylhomoserine lactones is wide-spread among the members of the Burkholderia cepacia complex. Syst Appl Microbiol 24, 1–14.[Medline]

Govan, J. R. & Deretic, V. (1996). Microbial pathogenesis in cystic fibrosis: mucoid Pseudomonas aeruginosa and Burkholderia cepacia. Microbiol Rev 60, 539–574.[Medline]

Govan, J. R., Hughes, J. E. & Vandamme, P. (1996). Burkholderia cepacia: medical, taxonomic and ecological issues. J Med Microbiol 45, 395–407.[Abstract]

Huber, B., Riedel, K., Hentzer, M., Heydorn, A., Gotschlich, A., Givskov, M., Molin, S. & Eberl, L. (2001). The cep quorum-sensing system of Burkholderia cepacia H111 controls biofilm formation and swarming motility. Microbiology 147, 2517–2528.[Abstract/Free Full Text]

Kent, G., Oliver, M., Foskett, J. K. & 7 other authors (1996). Phenotypic abnormalities in long-term surviving cystic fibrosis mice. Pediatr Res 40, 233–241.[Abstract]

Kothe, M., Antl, M., Huber, B., Stoecker, K., Ebrecht, D., Steinmetz, I. & Eberl, L. (2003). Killing of Caenorhabditis elegans by Burkholderia cepacia is controlled by the cep quorum-sensing system. Cell Microbiol 5, 343–351.[Medline]

Lewenza, S. & Sokol, P. A. (2001). Regulation of ornibactin synthesis and N-acyl-L-homoserine lactone production by CepR in Burkholderia cepacia. J Bacteriol 183, 2212–2218.[Abstract/Free Full Text]

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

Lewenza, S., Visser, M. B. & Sokol, P. A. (2002). Interspecies communication between Burkholderia cepacia and Pseudomonas aeruginosa. Can J Microbiol 48, 707–716.[CrossRef][Medline]

LiPuma, J. J., Spilker, T., Gill, L. H., Campbell, P. W. 3rd, Liu, L. & Mahenthiralingam, E. (2001). Disproportionate distribution of Burkholderia cepacia complex species and transmissibility markers in cystic fibrosis. Am J Respir Crit Care Med 164, 92–96.[Abstract/Free Full Text]

Lutter, E., Lewenza, S., Dennis, J. J., Visser, M. B. & Sokol, P. A. (2001). Distribution of quorum-sensing genes in the Burkholderia cepacia complex. Infect Immun 69, 4661–4666.[Abstract/Free Full Text]

Mahenthiralingam, E., Coenye, T., Chung, J. W., Speert, D. P., Govan, J. R. W., Taylor, P. & Vandamme, P. (2000). Diagnostically and experimentally useful panel of strains from the Burkholderia cepacia complex. J Clin Microbiol 38, 910–913.[Abstract/Free Full Text]

Mahenthiralingam, E., Baldwin, A. & Vandamme, P. (2002). Burkholderia cepacia complex infection in patients with cystic fibrosis. J Med Microbiol 51, 533–538.[Abstract/Free Full Text]

McKevitt, A. I., Bajaksouzian, S., Klinger, J. D. & Woods, D. E. (1989). Purification and characterization of an extracellular protease from Pseudomonas cepacia. Infect Immun 57, 771–778.[Medline]

Mohr, C. D., Tomich, M. & Herfst, C. A. (2001). Cellular aspects of Burkholderia cepacia infection. Microbes Infect 3, 425–435.[CrossRef][Medline]

Ohman, D. E., Sadoff, J. C. & Iglewski, B. H. (1980). Toxin A-deficient mutants of Pseudomonas aeruginosa PA103: isolation and characterization. Infect Immun 28, 899–908.[Medline]

Pearson, J. P., Feldman, M., Iglewski, B. H. & Prince, A. (2000). Pseudomonas aeruginosa cell-to-cell signaling is required for virulence in a model of acute pulmonary infection. Infect Immun 68, 4331–4334.[Abstract/Free Full Text]

Platt, T., Muller-Hill, B. & Miller, J. H. (1972). Analysis of the lac operon enzymes. In Experiments in Molecular Genetics, pp. 352–355. Edited by J. H. Miller. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Rajan, S. & Saiman, L. (2002). Pulmonary infections in patients with cystic fibrosis. Semin Respir Infect 17, 47–56.[CrossRef][Medline]

Ratjen, F. & Doring, G. (2003). Cystic fibrosis. Lancet 361, 681–689.[CrossRef][Medline]

Riedel, K., Hentzer, M., Geisenberger, O. & 7 other authors (2001). N-Acylhomoserine-lactone-mediated communication between Pseudomonas aeruginosa and Burkholderia cepacia in mixed biofilms. Microbiology 147, 3249–3262.[Abstract/Free Full Text]

Riedel, K., Arevalo-Ferro, C., Reil, G., Gorg, A., Lottspeich, F. & Eberl, L. (2003). Analysis of the quorum-sensing regulon of the opportunistic pathogen Burkholderia cepacia H111 by proteomics. Electrophoresis 24, 740–750.[CrossRef][Medline]

Sajjan, U., Thanassoulis, G., Cherapanov, V. & 9 other authors (2001). Enhanced susceptibility to pulmonary infection with Burkholderia cepacia in Cftr(-/-) mice. Infect Immun 69, 5138–5150.[Abstract/Free Full Text]

Sajjan, U., Kent, G. & Forstner, G. (2002). Early responses of CFTR (-/-) and wild type mice to B. cepacia lung infection. Pediatr Pulmonol 34S, 232, A156.[CrossRef]

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

Schweizer, H. P. (1993). Two plasmids, X1918 and Z1918, for easy recovery of the xylE and lacZ reporter genes. Gene 134, 89–91.[CrossRef][Medline]

Schweizer, H. P., Klassen, T. & Hoang, T. (1996). Improved methods for gene analysis and expression in Pseudomonas spp. In Molecular Biology of Pseudomonads, pp. 229–237. Edited by T. Nakazawa, K. Furukawa, D. Hass & S. Silver. Washington, DC: American Society for Microbiology.

Schwiebert, E. M., Benos, D. J. & Fuller, C. M. (1998). Cystic fibrosis: a multiple exocrinopathy caused by dysfunctions in a multifunctional transport protein. Am J Med 104, 576–590.[CrossRef][Medline]

Shaw, P. D., Ping, G., Daly, S. L., Cha, C., Cronan, J. E., Jr, Rinehart, K. L. & Farrand, S. K. (1997). Detecting and characterizing N-acyl-homoserine lactone signal molecules by thin-layer chromatography. Proc Natl Acad Sci U S A 94, 6036–6041.[Abstract/Free Full Text]

Smith, R. S., Fedyk, E. R., Springer, T. A., Mukaida, N., Iglewski, B. H. & Phipps, R. P. (2001). IL-8 production in human lung fibroblasts and epithelial cells activated by the Pseudomonas autoinducer N-3-oxododecanoyl homoserine lactone is transcriptionally regulated by NF-kappa B and activator protein-2. J Immunol 167, 366–374.[Abstract/Free Full Text]

Smith, R. S., Harris, S. G., Phipps, R. & Iglewski, B. (2002). The Pseudomonas aeruginosa quorum-sensing molecule N-(3-oxododecanoyl)homoserine lactone contributes to virulence and induces inflammation in vivo. J Bacteriol 184, 1132–1139.[Abstract/Free Full Text]

Sokol, P. A. & Woods, D. E. (1984). Relationship of iron and extracellular virulence factors to Pseudomonas aeruginosa lung infections. J Med Microbiol 18, 125–133.[Abstract]

Sokol, P. A., Ohman, D. E. & Iglewski, B. H. (1979). A more sensitive plate assay for detection of protease production by Pseudomonas aeruginosa. J Clin Microbiol 9, 538–540.[Medline]

Sokol, P. A., Darling, P., Woods, D. E., Mahenthiralingam, E. & Kooi, C. (1999). Role of ornibactin biosynthesis in the virulence of Burkholderia cepacia: characterization of pvdA, the gene encoding L-ornithine N(5)-oxygenase. Infect Immun 67, 4443–4455.[Abstract/Free Full Text]

Sokol, P. A., Darling, P., Lewenza, S., Corbett, C. R. & Kooi, C. D. (2000). Identification of a siderophore receptor required for ferric ornibactin uptake in Burkholderia cepacia. Infect Immun 68, 6554–6560.[Abstract/Free Full Text]

Speert, D. P., Henry, D., Vandamme, P., Corey, M. & Mahenthiralingam, E. (2002). Epidemiology of Burkholderia cepacia complex in patients with cystic fibrosis, Canada. Emerg Infect Dis 8, 181–187.[Medline]

Tatterson, L. E., Poschet, J. F., Firoved, A., Skidmore, J. & Deretic, V. (2001). CFTR and Pseudomonas infections in cystic fibrosis. Front Biosci 6, D890–D897.[Medline]

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]

Tomich, M., Griffith, A., Herfst, C. A., Burns, J. L. & Mohr, C. D. (2003). Attenuated virulence of a Burkholderia cepacia type III secretion mutant in a murine model of infection. Infect Immun 71, 1405–1415.[Abstract/Free Full Text]

van Heeckeren, A., Walenga, R., Konstan, M. W., Bonfield, T., Davis, P. B. & Ferkol, T. (1997). Excessive inflammatory response of cystic fibrosis mice to bronchopulmonary infection with Pseudomonas aeruginosa. J Clin Invest 100, 2810–2815.[Abstract/Free Full Text]

Vandamme, P., Henry, D., Coenye, T., Nzula, S., Vancanneyt, M., LiPuma, J. J., Speert, D. P., Govan, J. R. & Mahenthiralingam, E. (2002). Burkholderia anthina sp. nov. and Burkholderia pyrrocinia, two additional Burkholderia cepacia complex bacteria, may confound results of new molecular diagnostic tools. FEMS Immunol Med Microbiol 33, 143–149.[CrossRef][Medline]

Vandamme, P., Holmes, B., Coenye, T., Goris, J., Mahenthiralingam, E., LiPuma, J. J. & Govan, J. R. (2003). Burkholderia cenocepacia sp. nov. – a new twist to an old story. Res Microbiol 154, 91–96.[CrossRef][Medline]

Watson, B., Currier, T. C., Gordon, M. P., Chilton, M. D. & Nester, E. W. (1975). Plasmid required for virulence of Agrobacterium tumefaciens. J Bacteriol 123, 255–264.[Medline]

West, S. E., Schweizer, H. P., Dall, C., Sample, A. K. & Runyen-Janecky, L. J. (1994). Construction of improved EscherichiaPseudomonas shuttle vectors derived from pUC18/19 and sequence of the region required for their replication in Pseudomonas aeruginosa. Gene 148, 81–86.[CrossRef][Medline]

Williams, P., Camara, M., Hardman, A. & 7 other authors (2000). Quorum sensing and the population-dependent control of virulence. Philos Trans R Soc Lond B Biol Sci 355, 667–680.[CrossRef][Medline]

Wu, H., Song, Z., Givskov, M., Doring, G., Worlitzsch, D., Mathee, K., Rygaard, J. & Hoiby, N. (2001). Pseudomonas aeruginosa mutations in lasI and rhlI quorum sensing systems result in milder chronic lung infection. Microbiology 147, 1105–1113.[Abstract/Free Full Text]

Zhu, J., Beaber, J. W., More, M. I., Fuqua, C., Eberhard, A. & Winans, S. C. (1998). Analogs of the autoinducer 3-oxooctanoyl-homoserine lactone strongly inhibit activity of the TraR protein of Agrobacterium tumefaciens. J Bacteriol 180, 5398–5405.[Abstract/Free Full Text]

Received 4 June 2003; revised 28 August 2003; accepted 1 September 2003.