Abstract
Numerous gram-negative bacterial pathogens regulate virulence factor expression by using a cell density mechanism termed quorum sensing (QS). An in silico analysis of the Burkholderia mallei ATCC 23344 genome revealed that it encodes at least two luxI and four luxR homologues. Using mass spectrometry, we showed that wild-type B. mallei produces the signaling molecules N-octanoyl-homoserine lactone and N-decanoyl-homoserine lactone. To determine if QS is involved in the virulence of B. mallei, we generated mutations in each putative luxIR homologue and tested the pathogenicities of the derivative strains in aerosol BALB/c mouse and intraperitoneal hamster models. Disruption of the B. mallei QS alleles, especially in RJ16 (bmaII) and RJ17 (bmaI3), which are luxI mutants, significantly reduced virulence, as indicated by the survival of mice who were aerosolized with 104 CFU (10 50% lethal doses [LD50s]). For the B. mallei transcriptional regulator mutants (luxR homologues), mutation of the bmaR5 allele resulted in the most pronounced decrease in virulence, with 100% of the challenged animals surviving a dose of 10 LD50s. Using a Syrian hamster intraperitoneal model of infection, we determined the LD50s for wild-type B. mallei and each QS mutant. An increase in the relative LD50 was found for RJ16 (bmaI1) (>967 CFU), RJ17 (bmaI3) (115 CFU), and RJ20 (bmaR5) (151 CFU) compared to wild-type B. mallei (<13 CFU). These findings demonstrate that B. mallei carries multiple luxIR homologues that either directly or indirectly regulate the biosynthesis of an essential virulence factor(s) that contributes to the pathogenicity of B. mallei in vivo.
Burkholderia mallei, the etiologic agent of glanders, is a gram-negative, oxidase-positive, nonmotile bacillus that is an obligate animal pathogen (4). The natural hosts for B. mallei are horses, donkeys, and mules (solipeds). Until the early 20th century and the development of motorized transportation, glanders was common throughout the world (4). After the implementation of quarantine strategies for imported animals, no naturally occurring human cases of glanders have been reported in the United States since the 1930s. Human glanders is uncommon now, occasionally occurring in individuals such as veterinarians, slaughterhouse workers, and laboratory scientists whose occupation exposes them to infection. In solipeds, two distinctive forms of glanders may arise, either an acute (observed with mules and donkeys) or a chronic (common in horses) form. Symptoms of acute glanders include weight loss, difficulty breathing, and an elevated temperature. In contrast, horses with chronic glanders may exhibit pulmonary and cutaneous (farcy) symptoms. Human acute glanders is characterized by fever and fatigue as well as inflammation of and nodule formation on the face and peripheral limbs (4). Chronic glanders in humans presents with swollen lymph nodes, ulcerating nodules in the alimentary and respiratory tracts, weight loss, and numerous subcutaneous abscesses (4). B. mallei can cause disease in a variety of animals, including mice, hamsters, ferrets, guinea pigs, and monkeys, in addition to solipeds and humans (9, 22).
Many gram-negative bacteria possess sophisticated communication systems that allow microorganisms to detect and respond, in a cell-density-dependent manner, to fluctuating environmental conditions at the transcriptional level. This ability to couple extracellular and intracellular signals, termed quorum sensing (QS), involves the synthesis and accumulation of N-acyl-homoserine lactones (AHLs) (6, 10, 13). AHL biosynthesis is enzymatically mediated by the LuxI family of proteins, which are N-acyl-homoserine lactone synthases (AHSs), and a single AHS may produce multiple AHLs with various acyl chain lengths and chemical modifications (12). Cytosolic LuxR proteins respond to AHLs in a concentration-dependent manner through binding of the membrane-permeative signal molecule (AHL). This AHL-protein interaction facilitates conformational changes and multimerization, which in turn induces or represses target gene expression (11). In animal and plant pathogens, this coordinated gene expression of alleles encoding proteins needed for virulence allows microorganisms to elicit an overwhelming attack before host cells can mount an effective defense (2, 7, 12, 17, 25, 30, 37).
Functional QS networks have recently been identified in Burkholderia cepacia, Burkholderia vietnamiensis, Burkholderia thailandensis, and Burkholderia pseudomallei (1, 3, 15, 20, 32, 34-36). Collectively, these Burkholderia QS networks have been shown to both positively and negatively regulate various cellular processes, including AHL and protease production, siderophore biosynthesis, biofilm formation, lipase and beta-hemolytic activities, swarming and twitching motilities, and substrate utilization (18, 20, 21, 32, 35, 36). Furthermore, disruption of these cell signaling systems has been shown to reduce the pathogenicity of B. cepacia and Burkholderia pseudomallei in murine and hamster models of infection (1, 31, 34, 36).
Considering that no effective vaccine is available against glanders as well as the risk of B. mallei weaponization, investigations focusing on vaccine development against this highly infectious Burkholderia species are essential. The objective of this study was to analyze the functional role between QS and the pathogenicity of B. mallei. Utilizing two animal models of infection, we clearly demonstrate that QS is involved in the pathogenicity of B. mallei.
MATERIALS AND METHODS
Bacterial strains and plasmids.
The bacterial strains and cloning vectors used for this study are described in Table 1. B. mallei was cultured in Luria-Bertani (LB) broth or on plates containing 4% glycerol (LBG) (Sigma, St. Louis, Mo.). Escherichia coli strains containing recombinant clones were grown on LB plates or in broth containing 25 μg of kanamycin (Sigma)/ml and 50 μg of 5-bromo-4-chloro-3-indolyl-β-d-galactoside (Sigma)/ml by using standard procedures (26). For AHL detection, Agrobacterium tumefaciens NTL4 was cultured in AT minimal medium at 30°C (11).
TABLE 1.
Bacterial strains and plasmids used for this study
| Strain or plasmid | Descriptiona | Source |
|---|---|---|
| Strains | ||
| Escherichia coli | ||
| SM10 | Mobilizing strain, RP4 tra genes, Kmr | 29 |
| TOP10 | Used for cloning, gene expression, and blue-white screening | Invitrogen |
| RJ23 | TOP10 containing pRUR6 | This study |
| RJ24 | TOP10 containing pRUR7 | This study |
| Agrobacterium tumefaciens | ||
| NTL4 | Ti plasmidless derivative, nopaline chromosomal background | 11 |
| Burkholderia mallei | ||
| ATCC 23344 | Type strain genome sequence completedb | American Type Culture Collection |
| RJ16 | ATCC 23344 bmaI1::pRUI1, Gmr | This study |
| RJ17 | ATCC 23344 bmaI3::pRUI3, Gmr | This study |
| RJ18 | ATCC 23344 bmaR1::pRUR1, Gmr | This study |
| RJ19 | ATCC 23344 bmaR3::pRUR3, Gmr | This study |
| RJ20 | ATCC 23344 bmaR5::pRUR5, Gmr | This study |
| RJ21 | RJ16 containing pRUR6 | This study |
| RJ22 | RJ17 containing pRUR7 | This study |
| Plasmids | ||
| pGSV3 | Mobilizable suicide vector, Gmr | 5 |
| pCR2.1-TOPO | TA cloning vector, Kmr Apr | Invitrogen |
| pBHR1 | Broad-host-range expression vector, Kmr Cmr | MoBiTec |
| pRUI1 | pGSV3 containing a 369-bp PCR product from the ATCC 23344 bmaI1 gene | This study |
| pRUI3 | pGSV3 containing a 398-bp PCR product from the ATCC 23344 bmaI3 gene | This study |
| pRUR1 | pGSV3 containing a 397-bp PCR product from the ATCC 23344 bmaR1 gene | This study |
| pRUR3 | pGSV3 containing a 402-bp PCR product from the ATCC 23344 bmaR3 gene | This study |
| pRUR5 | pGSV3 containing a 401-bp PCR product from the ATCC 23344 bmaR5 gene | This study |
| pRUR6 | pBHR1 containing the bmaI1 gene | This study |
| pRUR7 | pBHR1 containing the bmaI3 gene | This study |
r, resistance; Km, kanamycin; Gm, gentamicin; Ap, ampicillin; Cm, chloramphenicol. bma represents the QS genes present in B. mallei; bmaI1 to I3 depict AHL synthases (luxI genes); and bmaR1, bmaR3, and bmaR5 indicate transcriptional regulators (luxR alleles).
The sequence is available at http://www.tigr.org/.
Cloning of B. mallei QS genes, mutant construction, gene disruption, and mutant confirmation.
PCR primers for disruption cassettes were made by using the B. mallei ATCC 23344 sequences (The Institute for Genomic Research) that were confirmed in silico to carry putative luxIR genes. Genomic DNA for PCR amplification was purified by using a MasterPure DNA purification kit according to the manufacturer's instructions (Epicentre Technologies, Madison, Wis.). Internal gene fragments were PCR amplified with the primer pairs listed in Table 2 under the following conditions: 1 cycle at 94°C for 5 min; 30 cycles at 94°C for 30 s, 56°C for 30 s, and 72°C for 30 s; and a final 7-min extension at 72°C. Site-specific integration was confirmed by using the cycling conditions described above, with a 5-min extension time, by use of the gene-specific PCR primer sets listed in Table 2. PCR amplification was performed by using an Epicentre FailSafe kit with buffer J (Epicentre Technologies). Reactions were analyzed by standard methods, and the products were subcloned into pCR2.1-TOPO (Invitrogen, Carlsbad, Calif.). Ligation products were transformed into One Shot chemically competent E. coli TOP10 cells and then screened accordingly (26). Mutant construction was performed as previously described (33). For gene expression in E. coli TOP10 cells, the B. mallei luxI genes were PCR amplified as described above, cloned into pCR2.1-TOPO, and chemically transformed into E. coli TOP10. Plasmid purification was performed by using a QIAprep Spin miniprep kit (Qiagen, Valencia, Calif.), and the resulting clones were digested with EcoRI (New England Biolabs, Beverly, Mass.) by standard methods (26). Digestion reactions were separated in a 1% agarose gel, and the bands were excised by use of a QIAquick gel extraction kit (Qiagen). Gel-purified amplicons were ligated into EcoRI-digested pBHR1 by using a Fast-Link DNA ligation kit (Epicentre) and were chemically transformed into E. coli TOP10. Recombinant clones were screened for AHL biosynthesis (cross-streaking) by using the A. tumefaciens NTL4 bioreporter strain.
TABLE 2.
Primers used for this investigation
| Primer target | Primer sequence (5′-3′)c |
|---|---|
| Internal gene primersa | |
| bmaI1 | F CCGCGACGACGACGGGGAAATC |
| R TCGATCCAGCACGCGACGACCAT | |
| bmaI3 | F TCGCGGGCCGATTGAACGAACTGC |
| R GAGCGACGCGGCCACCGTGAGCAC | |
| bmaR1 | F CGGCTTCGAATATTGCTGCTATGG |
| R GAGAAAACGGCTCATCAGCGAGTG | |
| bmaR3 | F AGACGTCGTCTCGCTGCACTATCC |
| R ACCCACGTGAGGCACATCTGTTCG | |
| bmaR4 | F GGCGTTCGACAGATGAAACACGAC |
| R GCTCATCTGGCACGACGACCTCTA | |
| bmaR5 | F CGCGTGCCGTGGCCGCTGTCCA |
| R CCGCGCTCCGGGTCCGCCATCAG | |
| Mutant confirmation primersb | |
| bmaI1 | F GCGCGAAACACGAGTCCCTGTCT |
| R TTTTCCTCGAACGTTGCGGATTGA | |
| bmaI3 | F CGGCGGTCCGGTTAGAGGAGAACG |
| R CGCCTTCGTGTCGCGCAACAGC | |
| bamR1 | F GAAGCGGAACCGTTGATGGAGTGA |
| R AGCGTGAAGCTGCTGGAGAACGAA | |
| bmaR3 | F GCGACACGAAGGCGCGGCGATAC |
| R GTGCGGGGTCGTCGTCGGGAGAAA | |
| bmaR4 | F AAACTGCCTGCACCTACGCTTTTG |
| R CTTGAGCTGGGCGGTTCTATGTTC | |
| bmaR5 | F AAACCGCATAAGCACAATCAATCA |
| R GAGCTTCAGGATCGCGTTCTTCAC |
Primers used for the construction of gene disruption cassettes.
Primers utilized to confirm site-specific integration of the suicide vector.
F, forward primer; R, reverse primer.
MS analysis of culture extracts.
AHLs were extracted from culture supernatants with ethyl acetate (Sigma), which was then evaporated, and the extracts were resuspended in 1 ml of acetonitrile (Sigma) (28). Aliquots (150 μl) were dried under a steady stream of dry nitrogen, reconstituted in 100 μl of 50% acetonitrile, and passed through a 0.45-μm-pore-size nylon filter. Approximately 500 nl of each extract was injected onto a CAPLC capillary liquid chromatograph (Waters Corporation, Milford, Mass.) fitted with an Aquasil C18 high-performance liquid chromatography column (10 cm by 75 μm) (New Objective, Woburn, Mass.) operating at a flow rate of 500 nl/min. A gradient elution was employed starting at 100% buffer A (2% acetonitrile-0.1% formic acid) and ending at 100% buffer B (80% acetonitrile-0.1% formic acid) in 30 min. A voltage of 2.1 kV was applied to the column effluent entering the nanoelectrospray source attached to a Q-TOF-2 mass spectrometer (Micromass, Beverly, Mass.). The source temperature was 125°C, and a cone voltage of 18 V was applied. Argon (10 Pa of nominal pressure) was used as the collision gas, with an energy setting of 15 V. The results obtained by mass spectrometry (MS) (scanning from m/z 160 to 330 in 1.5 s) were acquired by the use of data-directed analysis software (Waters Corporation). Ions meeting selected intensity and charge state criteria were further characterized by MS/MS. Precursor ions yielding a fragmentation ion at m/z 102, representing the lactone ring of AHL signaling molecules, were recorded, and the (M + H)+ values were determined. Fragmentation ions of MS/MS spectra containing an ion at m/z 102 were compared to the fragmentation mass spectra of the corresponding AHL standard when possible. If a precursor ion with an (M + H)+ value that was not equal to any of the AHL standards yielded an MS/MS spectrum containing an ion at m/z 102, the mass spectra were further analyzed for the presence of ions that are characteristic of acyl side chains containing substitutions that lose a water molecule(s) after collision-induced dissociation.
Aerosol challenge, LD50 analysis, and IgG titers.
An inhalational challenge of female BALB/c mice, an analysis of the 50% lethal dose (LD50), and measurements of immunoglobulin G (IgG) titers were performed as previously described (19, 24, 33, 38). Briefly, for aerosol exposures, wild-type B. mallei and each QS mutant were inoculated (100 μl from a 3-ml overnight culture) into 10 ml of LBG broth and cultured with aeration (250 rpm) for 18 h at 37°C. Aerosolization (10 mice for each bacterial strain) was performed by nebulizing the entire 10-ml overnight culture (stationary phase), which delivered approximately 10 LD50s.
RESULTS
Structural analysis and ClustalW nucleotide alignments of B. mallei QS alleles.
Using the B. cepacia CepIR and Pseudomonas aeruginosa LasIR and RhlIR proteins, we searched the B. mallei ATCC 23344 genome (The Institute for Genomic Research) in silico for putative LuxIR homologues. This in silico analysis, which was confirmed with PCR amplification (data not shown), indicated that B. mallei carries at least four luxR and two luxI homologues (Fig. 1). The structural organization of B. mallei luxIR and the surrounding genes is depicted in Fig. 1. The results of Blastx homology searches for each B. mallei QS allele are summarized in Table 3, and ClustalW nucleotide sequence alignments are described below.
FIG. 1.
Genetic organization of B. mallei QS loci. Approximately 6-kb segments that were confirmed in silico to carry putative luxIR homologues were used for structural analysis and ORF prediction. Genes identified in B. mallei ATCC 23344 are represented by bma, and triangles (▴) denote the mutated alleles analyzed in this study. Attempts to create mutations in B. mallei bmaR4 were unsuccessful.
TABLE 3.
Blastx searches using the B. mallei QS proteins
| Proteina | % Identity | % Similarity | Homologous proteinb | Protein IDc |
|---|---|---|---|---|
| BamI1 | 99 | 99 | B. pseudomallei BpsI AHL synthase | AAQ901683 |
| BamI3 | 40 | 54 | Burkholderia multivorans CepI AHL synthase | AF330013_1 |
| BamR1 | 100 | 100 | B. pseudomallei BpsR AHL receptor | AAR88244 |
| BamR3 | 44 | 61 | Burkholderia fungorum hypothetical protein | ZP_00030469.1 |
| BamR4 | 33 | 47 | Pseudomonas putida PpuR transcriptional regulator | AAM75413.1 |
| BamR5 | 51 | 67 | Ralstonia solanacearum transcriptional activator | NP_522339.1 |
Quorum-sensing proteins found in B. mallei.
Corresponding species containing similar LuxIR proteins to those of B. mallei.
GenBank protein accession numbers.
Briefly, the bmaIR1 loci are divergently transcribed and are separated by a GeneMark (http://opal.biology.gatech.edu/GeneMark/gmhmm2_prok.cgi)-predicted 444-bp open reading frame (ORF) with no significant similarity to proteins in the National Center for Biotechnology Information databases (Fig. 1). In contrast, the bmaIR3 alleles are genetically linked but are not disrupted by an intergenic sequence and are transcribed in the same direction (Fig. 1). Interestingly, a Blastx analysis of the 5′ and 3′ (3 kb on each side) regions flanking the bpmR4 and bpmR5 loci failed to recover any putative luxI genes, suggesting that these QS alleles are orphaned for a putative LuxI protein (Fig. 1).
Amino acid alignments (ClustalW) between the B. mallei BmaI1 and BmaI3 proteins, B. cepacia CepI, P. aeruginosa LasI and RhlI, B. pseudomallei BpsI, and the B. pseudomallei DD503 BpmI2 and BpmI3 (34) proteins revealed the presence of the 10 invariant amino acids (between residues 24 and 109) that are commonly found in LuxI proteins (23; also data not shown). Similarly, an alignment of the B. mallei LuxR proteins with P. aeruginosa RhlR and LasR, B. cepacia CepR, B. pseudomallei BpsR, and the B. pseudomallei DD503 BpmR2 to -5 transcriptional regulators identified six of the seven invariant amino acids that are found in LuxR proteins (8; also data not shown).
Detection and characterization of AHLs produced by wild-type B. mallei.
To determine the AHL moieties that are synthesized by wild-type B. mallei and each luxI QS mutant, we performed an MS analysis of crude culture extracts. B. mallei produced both N-octanoyl-homoserine lactone (C8-HSL) and N-decanoyl-homoserine lactone (C10-HSL), which was confirmed by an analysis of synthetic AHL standards (Table 4). In culture supernatants of RJ16, which contains a disruption in the bmaI1 AHS locus, the signaling molecules C8-HSL, C10-HSL, and N-(3-hydroxyoctanoyl)-l-homoserine lactone (3-hydroxy-C8-HSL) were identified (Table 4). In contrast to that for wild-type B. mallei and RJ16 (bmaI1 mutant), the only AHL moiety identified in RJ17 (bmaI3 mutant) culture supernatants was C8-HSL (Table 4). Disruption of the bmaI1 gene had no effect on AHL biosynthesis, and in fact, supernatants of RJ16 contained an additional AHL (3-hydroxy-C8-HSL) that was not detected in wild-type B. mallei (Table 4). Likewise, culture extracts of RJ17 (bmaI3 mutant) contained C8-HSL (Table 4). To address these discrepancies, we cloned the bmaI1 and bmaI3 genes into the broad-host-range expression vector pBHR1, transformed them into E. coli, and monitored the AHL biosynthesis profiles as described in Materials and Methods. RJ23 (expresses the bmaI1 gene) supernatants contained C8-HSL, C10-HSL, and 3-hydroxy-C8-HSL, while extracts from overnight cultures of RJ24 (heterologously expresses the bmaI3 gene) contained C8-HSL, C10-HSL, and N-(3-hydroxydecanoyl)-l-homoserine lactone (3-hydroxy-C10-HSL) (Table 4). All AHLs identified by MS in this investigation produced a fragment ion at m/z 102, which is characteristic of the lactone ring bound to the acyl side chain of AHLs. AHL standards for the hydroxy-substituted signaling molecules identified in this work (3-hydroxy-C8-HSL and 3-hydroxy-C10-HSL) were not analyzed, but the MS profiles matched spectra from a previous study (28). A summary of the relevant fragmentation ions is shown in Table 4.
TABLE 4.
AHL profiles of B. mallei and each luxI QS mutant
| AHL moleculea | Signature peaks (m/z values)b | Bacterial strains expressing moleculec |
|---|---|---|
| C8-HSL | 228, 127, 109, 102 | ATCC 23344, RJ16 (bmaI1), RJ17 (bmaI31), RJ23, RJ24 |
| C10-HSL | 256, 155, 137, 102 | ATCC 23344, RJ16 (bmaI1), RJ23, RJ24 |
| 3-Hydroxy-C8-HSL | 244, 125, 102, 97 | RJ16 (bmaI1), RJ23 |
| 3-Hydroxy-C10-HSL | 272, 153, 135, 102 | RJ24 |
N-Acyl-homoserine lactones synthesized by B. mallei and each QS mutant.
Signature peaks from mass spectrometry analysis of overnight culture extracts.
Wild-type B. mallei is represented by ATCC 23344, and bma depicts the luxI homologues. RJ23 (bmaI1) and RJ24 (bmaI3) are E. coli strains expressing the bmaI1 and bmaI3 genes.
Disruption of the B. mallei QS system reduces virulence in an aerosol BALB/c mouse model.
To analyze the course of acute infection for wild-type B. mallei and each QS mutant, we monitored animal survival after bacterial exposure for 39 days postexposure (p.e.) (Fig. 2). Groups of 10 female BALB/c mice were challenged with 104 CFU (10 LD50s) of B. mallei and each QS mutant. Deaths for the group exposed to wild-type B. mallei began on day 5, and the remaining mice succumbed by 6 days p.e. (Fig. 2). Surprisingly, 100% survival at 39 days p.e. was observed for the experimental groups that were aerosolized with RJ16 (bmaI1 mutant), RJ17 (bmaI3 mutant), and RJ20 (bmaR5 mutant) (Fig. 2). Deaths for mice challenged with RJ18 (bmaR1 mutant) and RJ19 (bmaR3 mutant) began on days 22 and 7, respectively, and continued over the 39-day course of analysis (Fig. 2). Although they were chronically infected (with splenic and hepatic abscesses, animal huddling, and fur ruffling), seven and four animals survived an aerosol challenge with RJ18 (bmaR1 mutant) and RJ19 (bmaR3 mutant), respectively, at 39 days p.e. (Fig. 2), in contrast to mice receiving wild-type B. mallei.
FIG. 2.
Time to death of BALB/c mice who were aerosolized with B. mallei and each QS mutant. The survival patterns of animals who were challenged with wild-type B. mallei ATCC 23344 and the QS mutants after aerosol exposure are shown. A targeted dose of 104 CFU (10 LD50s) was delivered, and animal mortality was monitored for 39 days. Disrupted B. mallei luxIR homologues are denoted by bmaI or bmaR.
LD50 determination, IgG titers, and vaccine efficacy.
To further assess the correlation between QS and the pathogenicity of B. mallei, we employed an acute hamster model of glanders. The relative LD50 for wild-type B. mallei at 4 days p.e. was <13 CFU, whereas individual mutagenesis of the B. mallei QS genes increased the LD50 up to approximately 100-fold (Table 5). Due to the sensitivity of Syrian hamsters to B. mallei, we performed complementation studies of the luxI mutants with this animal model. As expected, a reduction in the LD50 occurred by heterologous expression of bmaI1 and bmaI3 in RJ16 and RJ17, respectively (Table 5).
TABLE 5.
Bacterial LD50s for hamsters and IgG antibody titer determination
| Strain | LD50 (CFU) for Syrian hamstersa | IgG titerb |
|---|---|---|
| Wild-type B. mallei | <13 | |
| RJ16 (bmaI1) | >967 | 400 |
| RJ17 (bmaI3) | 115 | 100 |
| RJ18 (bmaR1) | 17 | 200 |
| RJ19 (bmaR3) | 98 | 400 |
| RJ20 (bmaR5) | 151 | 100 |
| RJ21 | 135 | |
| RJ22 | 51 | |
| Positive control | 12,800 |
For LD50 determination, male Syrian hamsters (five for each dose) were challenged with 101, 102, and 103 CFU of B. mallei and each QS mutant.
The B. mallei positive control was an irradiated culture aliquot; IgG titers were determined at 21 days p.e.
In addition to determining the time to death for aerosolized BALB/c mice and determination of LD50s in hamsters, we determined the IgG titers (in BALB/c mice) against B. mallei and each QS mutant, as previously described (38). Seropositive reactions, expressed as reciprocals of the highest dilutions producing positive results, were obtained for each of the B. mallei QS mutants in addition to the B. mallei positive control. The IgG titers for each strain are reported in Table 5.
For determinations of whether the B. mallei QS mutants provided protection against a challenge with wild-type B. mallei, experimental groups that were initially aerosolized with RJ16 (bmaI1 mutant), RJ17 (bmaI3 mutant), and RJ20 (bmaR5 mutant) were rechallenged at 14 and 27 days p.e. and then aerosolized at 36 days p.e. with 10 LD50s of wild-type B. mallei. Only pre-exposure to RJ17 (bmaI3 mutant) conferred partial protection (3 of 10 animals) to a challenge with B. mallei, with mice surviving 11 days postaerosolization, compared to unimmunized animals exposed to wild-type B. mallei, who died by 6 days p.e. (Fig. 2).
DISCUSSION
This investigation analyzed the role of QS in the pathogenicity of B. mallei in vivo by using aerosol BALB/c mouse and Syrian hamster models. In gram-negative bacteria, QS represents a complex mechanism for gene regulation through the synthesis and recognition of AHL signaling molecules. For human and plant pathogens, AHL-based communication systems allow a microbial community to strategically induce or repress expression of genes, specifically alleles encoding putative virulence factors, in response to environmental stimuli. Several reports have identified functional QS systems in various Burkholderia species and have shown that these bacterial communication networks both positively and negatively regulate numerous extracellular virulence factors in addition to contributing to animal pathogenicity (B. cepacia) (1, 3, 20, 21, 31, 34, 36).
The B. mallei QS system is extremely complex and is comprised of multiple luxIR homologues (Fig. 1). Our genome analysis in silico indicated that the B. pseudomallei and B. mallei QS networks are genetically similar, and as with B. pseudomallei, B. mallei does not encode a putative LuxS system (Fig. 1) (34). While our findings were under review, Valade et al. characterized two of the eight luxIR genes carried by B. pseudomallei (36). The pmlIR QS genes reported by Valade et al. correspond to the B. mallei bmaIR1 QS alleles characterized in this work (36). Interestingly, our in silico and in vitro (PCRs of internal gene amplicons) analyses indicated that B. mallei does not carry two of the luxIR pairs (bpmIR2) that were identified in the B. pseudomallei DD503 genome (34). Although they are preliminary, and considering that B. mallei is a pathoadaptive obligate mammalian pathogen as well as proposed to be a clone of B. pseudomallei (14), these findings suggest that these QS alleles are not required for the in vivo pathogenicity of B. mallei. This hypothesis is further supported by the observation that the B. mallei ATCC 23344 genome is 1.5 Mb smaller than the B. pseudomallei K96243 chromosomes (data not shown). Through the evolution and divergence of B. mallei from B. pseudomallei, we hypothesize that B. mallei has undergone genomic modifications (i.e., insertion sequence-mediated deletions) that have resulted in the loss of the additional luxIR (bpmIR2) pair that is carried by B. pseudomallei and Burkholderia thailandensis (34, 35).
It has been proposed that microbial species carrying multiple luxIR genes obtain these alleles through horizontal gene transfer (16). In some instances, these horizontally acquired segments of nucleic acid deviate in G+C content compared to the recipient host. The relative G+C ratio of the ORFs carrying the putative B. mallei luxIR homologues as well as the flanking genes is consistent with the overall G+C content of the B. mallei genome, suggesting that these QS alleles have been present throughout the evolution of this highly infectious Burkholderia species (data not shown).
Our initial approach for AHL detection and characterization, which had limited success, incorporated thin-layer chromatography overlays with the bioreporter strain A. tumefaciens NTL4. To circumvent these limitations, we performed MS with culture extracts of B. mallei and each luxI mutant. In supernatants from B. mallei, the signaling molecules C8-HSL and C10-HSL were detected (Table 4). As with B. mallei, it was recently shown that B. pseudomallei 008 produces C10-HSL via the PmlI protein (36). Similar to B. mallei, with the exception of N-(3-oxotetradecanoyl)-l-homoserine lactone, B. pseudomallei DD503 synthesizes C8-HSL, 3-hydroxy-C8-HSL, C10-HSL, and 3-hydroxy-C10-HSL (34). Furthermore, it has been shown that culture extracts from B. thailandensis DW503, a closely related Burkholderia species to B. mallei and B. pseudomallei, contain the signaling molecules N-hexanoyl-homoserine lactone, C8-HSL, and C10-HSL (35). Surprisingly, disruption of the B. mallei luxI homologues had a marginal effect on AHL biosynthesis, and in fact, mutagenesis of these genes resulted in the detection of signaling molecules that were not identified in wild-type B. mallei supernatants (Table 4). There are multiple scenarios that may have contributed to these observations: (i) the B. mallei genome may encode an additional LuxI protein(s), (ii) the B. mallei LuxI proteins may interact with multiple acyl-acyl carrier proteins (i.e., QS is involved in cellular metabolism), and (iii) the BmaI1 and BmaI3 proteins may synthesize overlapping signaling molecules. With regard to the hypothesis that QS in B. mallei may be involved in carbon metabolism (i.e., it may affect the biosynthesis of AHL precursors), it has been shown at the transcriptional level that QS in P. aeruginosa both positively and negatively regulates numerous enzymes that are involved in carbon metabolism (27, 39). Likewise, mutagenesis of the B. thailandensis QS system and enzymatic cleavage of the AHLs produced by this closely related Burkholderia species also affect substrate utilization (32, 35). To determine if the BmaI1 and BmaI3 proteins produce overlapping signaling molecules, each B. mallei luxI homologue was heterologously expressed in E. coli, and the AHL profiles were monitored. Table 4 clearly demonstrates that with the exception of 3-hydroxy-C8-HSL (unique to RJ23) and 3-hydroxy-C10-HSL (found only in RJ24 extracts), the BmaI1 and BmaI3 proteins, when expressed in E. coli, produce structurally similar AHLs, which may account for the AHL profiles observed for the B. mallei luxI mutants. However, before any definitive conclusions with regard to AHL biosynthesis can be made, it will be necessary to construct multideletion B. mallei luxI strains. Despite these fluctuations in AHL accumulation following mutagenesis of the B. mallei luxI homologues, definitive phenotypes for the bmaI1 (RJ16) and bmaI3 (RJ17) mutants were confirmed by the use of two independent animal models, indicating that AHL biosynthesis plays an essential role, either directly or indirectly, in the virulence of B. mallei. As proposed for B. pseudomallei DD503, it is possible that the timing of biosynthesis and the concentration of the B. mallei signaling molecules are important for in vivo pathogenicity.
It was recently demonstrated by the use of murine models of infection that QS is involved in the pathogenicity of B. cepacia and B. pseudomallei (1, 31, 34, 36). Mutagenesis of the B. mallei QS alleles caused a significant reduction in animal mortality compared to the mortality of mice aerosolized with wild-type B. mallei (Fig. 2). The most notable decrease in pathogenicity was observed for strains containing disruptions in the bmaI1 (RJ16) and bmaI3 (RJ17) luxI homologues (Fig. 2). Additionally, inactivation of the bmaR5 (RJ20) gene resulted in a drastic reduction in animal mortality compared to that with wild-type B. mallei (Fig. 2). In fact, 100% of the animals (10 for each group) that were exposed to RJ16 (bmaI1 mutant), RJ17 (bmaI3 mutant), and RJ20 (bmaR5 mutant) survived an aerosol challenge of 104 CFU, representing 10 LD50s (Fig. 2). As with the aerosol BALB/c model, a reduction in virulence was also observed for several of the B. mallei QS mutants in Syrian hamsters (Table 5). Similar to the BALB/c aerosolization results, although to a lesser degree, RJ16 (bmaI1 mutant), RJ17 (bmaI3 mutant), RJ19 (bmaR2 mutant), and RJ20 (bmaR5 mutant) demonstrated the largest reductions in pathogenicity compared to wild-type B. mallei (Table 5). Interestingly, 70% of the mice that were exposed to RJ18 (bmaR1 mutant) survived the challenge, whereas in hamsters RJ18 exhibited an LD50 similar to that of wild-type B. mallei (Fig. 2 and Table 5). These findings suggest that the B. mallei QS system may regulate unidentified host-specific virulence factors that are needed for mouse versus hamster pathogenicity.
For B. cepacia and B. pseudomallei, QS has been shown to both negatively and positively regulate the biosynthesis of potential extracellular virulence factors (1, 20, 21, 36). Recently, we found that QS in B. thailandensis DW503, a species that is closely related to B. mallei and B. pseudomallei, both positively and negatively regulates lipase and beta-hemolytic activities, swarming and twitching motilities, and carbon metabolism (32, 35). With the exception of swarming and twitching motilities (B. mallei is nonmotile), B. mallei and each QS mutant were tested for defects in beta-hemolytic, protease, lipase, and phospholipase C activities. It should be noted that B. mallei as well as B. thailandensis are normally nonhemolytic; however, mutagenesis of the B. thailandensis luxIR genes resulted in enhanced β-hemolysis of sheep erythrocytes (35). Surprisingly, and consistent with the case for mutagenesis of the B. pseudomallei DD503 QS network, the phenotypes of the parental and mutant strains of B. mallei were identical (data not shown). These results suggest that QS in B. mallei does not regulate a factor(s) that contributes to beta-hemolytic, lipase, protease, and phospholipase C activities.
For B. mallei, the only definitive virulence determinants that have been shown to be required for pathogenicity are an extracellular capsule and type III secretion (5, 33). A transmission electron microscopy analysis of capsule biosynthesis in each B. mallei QS mutant indicated that capsule production was not affected (data not shown). We are currently using whole-genome DNA microarrays to determine if QS affects the transcription of the B. mallei type III secretion operon. The B. mallei QS system represents one of the most complex intraspecies communication systems identified for obligate mammalian pathogens. These findings for two animal models of infection clearly demonstrate that QS plays an essential role in the in vivo pathogenicity of B. mallei. Further studies, utilizing whole-genome DNA microarrays, will be needed to identify the virulence factor(s) regulated by this intricate bacterial cell signaling network.
Acknowledgments
We thank Melanie Ulrich, Tim Hoover, William Day, Jeffery Adamovicz, and Katheryn Kenyon for critically reviewing the manuscript; Marilyn England and David Waag for measuring IgG titers; the USAMRIID Aerobiology Division for directing the aerosol challenges; and Lynda Miller, Anthony Bassett, Kathy Kuehl, and Ron Lind for their technical assistance.
This research was sponsored by the Medical Biological Defense Research Program, U.S. Army Medical Research and Material Command (project 02-4-5X-026).
All research was conducted in compliance with the Animal Welfare Act and other federal statutes and regulations relating to animals and experiments involving animals and adhered to principles stated in the Guide for the Care and Use of Laboratory Animals, National Research Council, 1996. The facility where this research was conducted is fully accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care International.
Opinions, interpretations, conclusions, and recommendations are those of the authors and are not necessarily endorsed by the U.S. Army.
Editor: J. B. Bliska
REFERENCES
- 1.Adam Baldwin, P. A. S., J. Parkhill, and E. Mahenthiralingam. 2004. The Burkholderia cepacia epidemic strain marker is part of a novel genomic island encoding both virulence and metabolism-associated genes in Burkholderia cenocepacia. Infect. Immun. 72:1537-1547. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Camara, M., P. Williams, and A. Hardman. 2002. Controlling infection by tuning in and turning down the volume of bacterial small-talk. Lancet Infect. Dis. 2:667-676. [DOI] [PubMed] [Google Scholar]
- 3.Conway, B. A., and E. P. Greenberg. 2002. Quorum-sensing signals and quorum-sensing genes in Burkholderia vietnamiensis. J. Bacteriol. 184:1187-1191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.DeShazer, D., and D. Waag. Glanders: new insights into an old disease. In L. Lindler, F. Lebeda, and G. W. Korch (ed.), Biological weapons defense: infectious diseases and counterbioterrorism, in press. The Humana Press Inc., Totowa, N.J.
- 5.DeShazer, D., D. M. Waag, D. L. Fritz, and D. E. Woods. 2001. Identification of a Burkholderia mallei polysaccharide gene cluster by subtractive hybridization and demonstration that the encoded capsule is an essential virulence determinant. Microb. Pathog. 30:253-269. [DOI] [PubMed] [Google Scholar]
- 6.Donabedian, H. 2003. Quorum sensing and its relevance to infectious diseases. J. Infect. 46:207-214. [DOI] [PubMed] [Google Scholar]
- 7.Dunn, A. K., and J. Handelsman. 2002. Toward an understanding of microbial communities through analysis of communication networks. Antonie Leeuwenhoek 81:565-574. [DOI] [PubMed] [Google Scholar]
- 8.Flavier, A. B., L. M. Ganova-Raeva, M. A. Schell, and T. P. Denny. 1997. Hierarchical autoinduction in Ralstonia solanacearum: control of acyl-homoserine lactone production by a novel autoregulatory system responsive to 3-hydroxypalmitic acid methyl ester. J. Bacteriol. 179:7089-7097. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Fritz, D. L., P. Vogel, D. R. Brown, D. Deshazer, and D. M. Waag. 2000. Mouse model of sublethal and lethal intraperitoneal glanders (Burkholderia mallei). Vet. Pathol. 37:626-636. [DOI] [PubMed] [Google Scholar]
- 10.Fuqua, C., M. R. Parsek, and E. P. Greenberg. 2001. Regulation of gene expression by cell-to-cell communication: acyl-homoserine lactone quorum sensing. Annu. Rev. Genet. 35:439-468. [DOI] [PubMed] [Google Scholar]
- 11.Fuqua, C., and S. C. Winans. 1996. Conserved cis-acting promoter elements are required for density-dependent transcription of Agrobacterium tumefaciens conjugal transfer genes. J. Bacteriol. 178:435-440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Fuqua, C., S. C. Winans, and E. P. Greenberg. 1996. Census and consensus in bacterial ecosystems: the LuxR-LuxI family of quorum-sensing transcriptional regulators. Annu. Rev. Microbiol. 50:727-751. [DOI] [PubMed] [Google Scholar]
- 13.Fuqua, W. C., S. C. Winans, and E. P. Greenberg. 1994. Quorum sensing in bacteria: the LuxR-LuxI family of cell density-responsive transcriptional regulators. J. Bacteriol. 176:269-275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Godoy, D., G. Randle, A. J. Simpson, D. M. Aanensen, T. L. Pitt, R. Kinoshita, and B. G. Spratt. 2003. Multilocus sequence typing and evolutionary relationships among the causative agents of melioidosis and glanders, Burkholderia pseudomallei and Burkholderia mallei. J. Clin. Microbiol. 41:2068-2079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gotschlich, A., B. Huber, O. Geisenberger, A. Togl, A. Steidle, K. Riedel, P. Hill, B. Tummler, P. Vandamme, B. Middleton, M. Camara, P. Williams, A. Hardman, and L. Eberl. 2001. Synthesis of multiple N-acylhomoserine lactones is wide-spread among the members of the Burkholderia cepacia complex. Syst. Appl. Microbiol. 24:1-14. [DOI] [PubMed] [Google Scholar]
- 16.Gray, K. M., and J. R. Garey. 2001. The evolution of bacterial LuxI and LuxR quorum sensing regulators. Microbiology 147:2379-2387. [DOI] [PubMed] [Google Scholar]
- 17.Hardman, A. M., G. S. Stewart, and P. Williams. 1998. Quorum sensing and the cell-cell communication dependent regulation of gene expression in pathogenic and non-pathogenic bacteria. Antonie Leeuwenhoek 74:199-210. [DOI] [PubMed] [Google Scholar]
- 18.Huber, B., K. Riedel, M. Hentzer, A. Heydorn, A. Gotschlich, M. Givskov, S. Molin, and L. Eberl. 2001. The cep quorum-sensing system of Burkholderia cepacia H111 controls biofilm formation and swarming motility. Microbiology 147:2517-2528. [DOI] [PubMed] [Google Scholar]
- 19.Jeddeloh, J. A., D. L. Fritz, D. M. Waag, J. M. Hartings, and G. P. Andrews. 2003. Biodefense-driven murine model of pneumonic melioidosis. Infect. Immun. 71:584-587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Lewenza, S., B. Conway, E. P. Greenberg, and P. A. Sokol. 1999. Quorum sensing in Burkholderia cepacia: identification of the LuxRI homologs CepRI. J. Bacteriol. 181:748-756. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Lewenza, S., and P. A. Sokol. 2001. Regulation of ornibactin biosynthesis and N-acyl-l-homoserine lactone production by CepR in Burkholderia cepacia. J. Bacteriol. 183:2212-2218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Miller, W. R., L. Pannell, L. Cravitz, W. A. Tanner, and T. Rosebury. 1947. Studies on certain biological characteristics of Malleomyces mallei and Malleomyces pseudomallei. J. Bacteriol. 55:115-126. [PMC free article] [PubMed] [Google Scholar]
- 23.Parsek, M. R., A. L. Schaefer, and E. P. Greenberg. 1997. Analysis of random and site-directed mutations in rhII, a Pseudomonas aeruginosa gene encoding an acylhomoserine lactone synthase. Mol. Microbiol. 26:301-310. [DOI] [PubMed] [Google Scholar]
- 24.Roy, C. J., M. Hale, J. M. Hartings, L. Pitt, and S. Duniho. 2003. Impact of inhalation exposure modality and particle size on the respiratory deposition of ricin in BALB/c mice. Inhal. Toxicol. 15:619-638. [DOI] [PubMed] [Google Scholar]
- 25.Rumbaugh, K. P., J. A. Griswold, and A. N. Hamood. 2000. The role of quorum sensing in the in vivo virulence of Pseudomonas aeruginosa. Microbes Infect. 2:1721-1731. [DOI] [PubMed] [Google Scholar]
- 26.Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
- 27.Schuster, M., C. P. Lostroh, T. Ogi, and E. P. Greenberg. 2003. Identification, timing, and signal specificity of Pseudomonas aeruginosa quorum-controlled genes: a transcriptome analysis. J. Bacteriol. 185:2066-2079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Shaw, P. D., G. Ping, S. L. Daly, C. Cha, J. E. Cronan, Jr., K. L. Rinehart, and S. K. Farrand. 1997. Detecting and characterizing N-acyl-homoserine lactone signal molecules by thin-layer chromatography. Proc. Natl. Acad. Sci. USA 94:6036-6041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Simon, R., U. Priefer, and A. Pühler. 1983. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram-negative bacteria. Biotechnology 1:784-791. [Google Scholar]
- 30.Smith, R. S., and B. H. Iglewski. 2003. Pseudomonas aeruginosa quorum-sensing systems and virulence. Curr. Opin. Microbiol. 6:56-60. [DOI] [PubMed] [Google Scholar]
- 31.Sokol, P. A., U. Sajjan, M. B. Visser, S. Gingues, J. Forstner, and C. Kooi. 2003. The CepIR quorum-sensing system contributes to the virulence of Burkholderia cenocepacia respiratory infections. Microbiology 149:3649-3658. [DOI] [PubMed] [Google Scholar]
- 32.Ulrich, R. L. Quorum quenching: Enzymatic disruption of N-acylhomoserine lactone-mediated bacterial communication in Burkholderia thailandensis. Appl. Environ. Microbiol., in press. [DOI] [PMC free article] [PubMed]
- 33.Ulrich, R. L., and D. DeShazer. 2004. Type III secretion: a virulence factor delivery system essential for the pathogenicity of Burkholderia mallei. Infect. Immun. 72:1150-1154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Ulrich, R. L., D. DeShazer, E. E. Brueggemann, H. B. Hines, P. C. Oyston, and J. A. Jeddeloh. The role of quorum sensing in the pathogenicity of Burkholderia pseudomallei. J. Med. Microbiol., in press. [DOI] [PubMed]
- 35.Ulrich, R. L., H. B. Hines, N. Parthasarathy, and J. A. Jeddeloh. 2004. Mutational analysis and biochemical characterization of the Burkholderia thailandensis DW503 quorum-sensing network. J. Bacteriol. 186:4350-4360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Valade, E., F. M. Thibault, Y. P. Gauthier, M. Palencia, M. Y. Popoff, and D. R. Vidal. 2004. The PmlI-PmlR quorum-sensing system in Burkholderia pseudomallei plays a key role in virulence and modulates production of the MprA protease. J. Bacteriol. 186:2288-2294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Von Bodman, S. B., W. D. Bauer, and D. L. Coplin. 2003. Quorum sensing in plant-pathogenic bacteria. Annu. Rev. Phytopathol. 41:455-482. [DOI] [PubMed] [Google Scholar]
- 38.Waag, D., J. Chulay, T. Marrie, M. England, and J. Williams. 1995. Validation of an enzyme immunoassay for serodiagnosis of acute Q fever. Eur. J. Clin. Microbiol. Infect. Dis. 14:421-427. [DOI] [PubMed] [Google Scholar]
- 39.Wagner, V. E., D. Bushnell, L. Passador, A. I. Brooks, and B. H. Iglewski. 2003. Microarray analysis of Pseudomonas aeruginosa quorum-sensing regulons: effects of growth phase and environment. J. Bacteriol. 185:2080-2095. [DOI] [PMC free article] [PubMed] [Google Scholar]