|Home | About | Journals | Submit | Contact Us | Français|
Burkholderia pseudomallei is a bacterial pathogen that causes a broad spectrum of clinical symptoms collectively known as melioidosis. Since it can be acquired by inhalation and is difficult to eradicate due to its resistance to a wide group of antibiotics and capacity for latency, work with B. pseudomallei requires a biosafety level 3 (BSL-3) containment facility. The bsa (Burkholderia secretion apparatus)-encoded type III secretion system (TTSS) has been shown to be required for its full virulence in a number of animal models. TTSSs are export devices found in a variety of gram-negative bacteria that translocate bacterial effector proteins across host cell membranes into the cytoplasm of host cells. Although the Bsa TTSS has been shown to play an important role in the ability of B. pseudomallei to survive and replicate in mammalian cells, escape from the endocytic vacuole, and spread from cell to cell, little is known about its effectors mediating these functions. Using bioinformatics, we identified homologs of several known TTSS effectors from other bacteria in the B. pseudomallei genome. In addition, we show that orthologs of these putative effectors exist in the genome of B. thailandensis, a closely related bacterium that is rarely pathogenic to humans. By generating a Bsa TTSS mutant B. thailandensis strain, we also demonstrated that the Bsa TTSS has similar functions in the two species. Therefore, we propose B. thailandensis as a useful BSL-1 model system to study the role of the Bsa TTSS during Burkholderia infection of mammalian cells and animals.
Burkholderia pseudomallei is a gram-negative bacterial pathogen responsible for melioidosis, an infectious disease endemic in the tropics of Southeast Asia and Northern Australia (6, 49). The bacterium can be acquired by inhalation, ingestion, or skin penetration of contaminated soil or ground water, leading to a wide spectrum of clinical outcomes ranging from an asymptomatic state to pneumonia, skin abscess, and highly fatal bacteremia (6, 40). The disease has a remarkable capacity for latency, as it has been shown to develop in human patients up to 62 years after geographic exposure (29). Therapy of melioidosis with antibiotics is usually long and difficult, because the bacterium is intrinsically resistant to a diverse group of antibiotics. Furthermore, relapse of melioidosis is common even after apparently successful treatment (6). For these reasons, B. pseudomallei has been considered a potential bioterrorism threat and is listed as a category B agent by the Centers for Disease Control and Prevention (CDC, 2000).
B. pseudomallei is a facultative intracellular bacterium that has evolved mechanisms to subvert host cellular processes. It has the ability to invade nonphagocytic cells and to survive and replicate in both phagocytic and nonphagocytic cells (16). It can also enter and survive in amoebae, which may serve as its host in soil and aquatic environments (15). Following internalization by mammalian cells, the bacterium is capable of lysing the endocytic membrane and escaping into the cytoplasm of the host cell, which may contribute to its ability to evade killing by macrophages (12, 16). In addition, cytoplasmic B. pseudomallei is able to induce actin tail formation, which enables it to move intracellularly and spread from cell to cell. This also promotes cell fusion, resulting in the formation of multinucleated giant cells (18). The presence of fused cells in tissues of melioidosis patients suggests that they may be important for the pathogenesis of the organism, perhaps by sequestering the bacteria from host immune surveillance and prolonging latency.
The B. pseudomallei genome contains three distinct gene clusters with homology to type III secretion system (TTSS) genes found in various gram-negative bacteria (1, 30). Two of the putative B. pseudomallei TTSSs show significant similarities to the plant pathogen-like TTSS of Ralstonia solanacearum and are speculated to be involved in symbiotic or pathogenic bacterium-plant interaction in rice paddies, a common environment of B. pseudomallei. The third B. pseudomallei TTSS, termed Bsa (Burkholderia secretion apparatus), resembles the Salmonella enterica serovar Typhimurium pathogenicity island 1 (SPI1)-encoded and the Shigella flexneri Ipa/Mxi/Spa TTSSs. B. pseudomallei mutants in the Bsa TTSS have been shown to be attenuated in various inbred mouse and Syrian hamster infection models and have reduced ability to escape from endocytic vesicles, replicate intracellularly, and form actin tails and multinucleated giant cells (41-43, 46). This suggests that the Bsa TTSS is essential for the pathogenesis of B. pseudomallei.
B. thailandensis is a bacterium closely related to B. pseudomallei (3). In fact, when it was first isolated from the environment, it was mistakenly identified as B. pseudomallei due to the many similar characteristics of the two species (50). A notable difference between them is the ability of B. thailandensis to assimilate l-arabinose, in contrast to B. pseudomallei, which lacks the entire arabinose-assimilation operon. In addition, B. thailandensis is rarely pathogenic to humans and its infectious dose is significantly higher in animal models (4, 37). Interestingly, the virulence-associated Bsa TTSS-encoding region is highly conserved between B. pseudomallei and B. thailandensis (19). However, the B. thailandensis Bsa TTSS is negatively regulated by growth in medium containing l-arabinose, which may partly explain the reduced pathogenicity of this species (27). The absence of other bacterial factors, such as a gene cluster involved in the production of capsular polysaccharides, also contributes to the reduced virulence of B. thailandensis (32).
TTSSs are specialized transport machineries activated under specific conditions for the delivery of bacterial virulence proteins, called effectors, into the host cell cytoplasm (28). Once intracellular, these effectors function to alter host cellular processes in order to promote bacterial survival and colonization. Many TTSS effectors are conserved among bacterial species (44). Based on their similarities to known Salmonella and Shigella effector genes and their proximities to the bsa locus, nine putative Burkholderia effector genes have been discovered (39, 41-43, 46). However, only the encoded product of bopE, a homolog of the Salmonella SopE, which is a guanine nucleotide exchange factor for host Rho GTPases, has actually been shown to be transported by the Bsa TTSS (39). Since other effectors may not be linked to the bsa locus, as is the case with many bacterial TTSS effectors, and since the Bsa TTSS appears to have a more complex role in the pathogenesis of B. pseudomallei than can be explained by the putative functions of the nine known homologs, we hypothesized that there are additional Bsa TTSS effectors encoded elsewhere in the B. pseudomallei genome and were interested in identifying them by using bioinformatics (9). In addition, because the B. thailandensis genome also encodes the Bsa TTSS, we wanted to determine whether the effector contents of the two bacteria were also the same. If so, B. thailandensis could be utilized as an attractive model system to facilitate the study of the role of the Bsa TTSS during Burkholderia infection, since, in contrast to the mandated B. pseudomallei working conditions, work with B. thailandensis does not require a biosafety level 3 (BSL-3) containment facility and there is no restriction on the use of antibiotic-resistance markers for its genetic manipulation.
The genomic sequences of B. pseudomallei K96243 and B. thailandensis E264 were those of the published annotations by the Sanger Institute and The Institute for Genomic Research (currently part of the J. Craig Venter Institute), respectively. The protein sequences of known bacterial TTSS effectors from a variety of gram-negative bacteria were compiled from the Entrez Protein database (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=Protein). To identify homologs of these known bacterial TTSS effectors, their protein sequences were compared to those of the translated B. pseudomallei K96243 or B. thailandensis E264 genome by the use of the Basic Local Alignment Search Tool (BLASTP). An E value of 1 × 10−5 or less was considered significant.
B. thailandensis E264 (American Type Culture Collection [ATCC] 700388) was obtained from the ATCC (3). Transfer of plasmids into B. thailandensis was carried out by conjugal mating with Escherichia coli SM10 lambda pir+ (25). The bsaZ gene of B. thailandensis 700388 was inactivated by insertion of an all-frame STOP (TGA CTG AGT AG) after the 20th codon by allelic exchange, yielding AH174. This was accomplished by PCR amplification of bp −939 to 60 of bsaZ and of the all-frame STOP with bp 61 to 1061 of bsaZ and by ligation of these two DNA fragments into pDM4 (26). The resulting plasmid, pAH59, was introduced into E. coli SM10 lambda pir+ by electroporation and then transferred into B. thailandensis 700388 by conjugal mating. Transconjugants were selected using medium containing chloramphenicol (CAM) and polymyxin B, and the all-frame STOP insertion was confirmed by PCR analysis and sequencing. To complete the allelic exchange, the integrated plasmid was forced to recombine out of the chromosome by growth on medium with 10% sucrose. Colonies that were sensitive to CAM were then verified for the insertion by PCR. For complementation, a plasmid expressing bsaZ, pAH60, which was constructed by PCR amplification and insertion of the B. thailandensis 700388 bsaZ gene into pMLS7, was introduced into AH174, yielding AH186 (21). The enhanced green fluorescent protein (eGFP)-expressing B. thailandensis strains AH181 and AH183 were created by introducing pMLS7-eGFP into AH174 and 700388, respectively (21). B. thailandensis AH191 and AH194 were generated by PCR amplification of B. thailandensis 700388 bopE along with an in-frame carboxy-terminal hemagglutinin (HA) tag, insertion of this PCR fragment into pMLS7, and introduction of the resulting plasmid, pAH62, into B. thailandensis 700388 and AH174, respectively. The bacterial strains and plasmids used in this study are also listed in Table Table11.
All bacterial strains were grown on Luria-Bertani (LB) agar plates or in LB broth with aeration at 37°C. When appropriate, CAM, polymyxin B, and trimethoprim (TMP) were each used at 50 μg/ml. For infection of HeLa cells and preparation of secreted proteins, bacterial strains were grown to mid-logarithmic phase by diluting overnight cultures 1:20 and growing them to an optical density of 1.7 to 2 at 600 nm.
HeLa cells were obtained from the ATCC and were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 μg/ml streptomycin at 37°C in humid air with 5% CO2. Bacteria were added to HeLa cells at an approximate multiplicity of infection (MOI) of 10 in antibiotic-free culture medium and centrifuged onto the cells at 800 × g for 5 min, and the infection was allowed to proceed at 37°C in humid air with 5% CO2. After 2 h, the infected cells were washed three times with phosphate-buffered saline (PBS) to remove extracellular bacteria and were either lysed or incubated with culture medium containing 10 μg/ml imipenem (IMP) or 500 μg/ml ceftazidime for the indicated lengths of time.
For quantification of intracellular B. thailandensis 700388, AH174, or AH186, infected HeLa cells were washed three times with PBS and lysed with 0.1% Triton X-100-PBS. Serial dilutions of cell lysates were made in LB broth and were plated onto LB agar with or without TMP, as appropriate. CFU were counted after 24 to 48 h at 37°C.
Vacuolar acidification was blocked by pretreating HeLa cells with 0 to 20 mM ammonium chloride (NH4Cl) in an equal volume of water or with 0 to 100 nM bafilomycin A1 (Sigma-Aldrich) in an equal volume of dimethyl sulfoxide (DMSO). After 1 h, the cells were infected with B. thailandensis 700388, AH174, AH183, or AH186, while NH4Cl or bafilomycin A1 was maintained in the culture medium throughout the infection. At 2 h postinfection, the cells were washed with PBS to eliminate extracellular bacteria and IMP was added to the culture medium. After 18 h, the infected cells were washed again three times with PBS and intracellular bacteria were quantified or the cells were examined by microscopy.
HeLa cells seeded onto 12-mm-diameter glass coverslips were infected with AH181 or AH183, as described above. After 2 h, the infected cells were washed three times with PBS and incubated with culture medium containing IMP for an additional 18 h. The infected cells were then washed three times with PBS and were fixed with 10% formyl saline for 15 min. Cells were washed again and then permeabilized with ice-cold acetone for 20 s. After being washed again with PBS, the cells were blocked with 10% fetal bovine serum in PBS for 1 h. The samples were probed with primary and secondary antibodies or stains for 1 h each time with washes with PBS afterwards. Lysosome-associated membrane protein 1 (LAMP-1) was detected with mouse monoclonal antibody H4A3 (Developmental Studies Hybridoma Bank, University of Iowa) followed by staining with anti-mouse immunoglobulin G-tetramethylrhodamine isothiocyanate (Sigma), while filamentous actin and cell nuclei were visualized by staining with phalloidin labeled with Texas Red (Invitrogen) and 4′,6′-diamidino-2-phenylindole (DAPI; Sigma-Aldrich), respectively. Stained coverslips were mounted onto microscope slides with Fluoromount G (VWR Scientific) and were examined with a Nikon Eclipse TE2000-E microscope using a 40×, 60×, or 100× lens. Images were acquired with a Photometrics CoolSnap HQ camera using MetaMorph imaging software (Universal Imaging).
For preparation of secreted proteins, AH191 and AH194 were grown to mid-logarithmic phase in LB broth at pH 4.5 and the proteins from the supernatants were precipitated with 10% (vol/vol) trichloroacetic acid, as previously described, resuspended in 50 mM Tris (pH 8.3), and concentrated using Microcon filters (Millipore) (10 kDa nominal molecular mass limit) (20). Bacterial pellets were lysed in sodium dodecyl sulfate sample buffer. Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and BopE-HA was detected by immunoblotting using the mouse monoclonal antibody HA.11 (Covance) as the primary antibody and the anti-mouse immunoglobulin G-horseradish peroxidase antibody conjugate as the secondary antibody (Amersham Biosciences).
Specific-pathogen-free C57BL/6 mice were obtained from the Jackson Laboratories (Bar Harbor, ME). All animals were housed in laminar flow cages and were permitted ad libitum access to sterile food and water. Euthanasia was accomplished with intraperitoneal pentobarbital injection followed by exsanguination from a cardiac puncture. The Institutional Animal Care and Use Committee of the University of Washington approved all experimental procedures.
For infection, 20 ml of LB broth was inoculated with a single colony of B. thailandensis 700388 or AH174 at 37°C. After 18 h, the bacteria were washed twice and resuspended in PBS to the desired concentration. Mice were exposed to aerosolized bacteria by use of a snout-only inhalation system (In-Tox Products, Moriarty, NM) (47). Aerosols were generated from a MiniHEART high-flow nebulizer (Westmed, Tucson, AZ) driven at 40 lb/in2. Airflow through the system was maintained for 10 min at 24 liters/min followed by a 5-min purge with air. Bacterial deposition in each experiment was determined from quantitative cultures of lung tissue from four sentinel mice per strain sacrificed immediately after infection. Animals were examined daily for illness or death, and abdominal surface temperatures were measured using a Ranger MX4P digital infrared thermometer (Raytek, Santa Cruz, CA). Ill animals with temperatures of <21.5°C, ruffled fur, eye crusting, hunched posture, and lack of resistance to handling were deemed terminal and were euthanized. At specific time points after infection, the remaining mice were sacrificed, the left lung, median hepatic lobe, and spleen were each homogenized in 1 ml of PBS, and serial dilutions were plated onto LB agar. Colonies were counted after 2 to 4 days of incubation at 37°C in humid air with 5% CO2.
Combined data are reported as means ± standard deviations. P values for inferences between two groups were calculated by log base 10 transformation of the data to correct for inequality of the variances, followed by the Welch two-sample t test. P values for inferences derived from the data from three groups were calculated by log base 10 transformation of the data, followed by analysis of variance and the Tukey posttest. The number one was substituted for values of zero, where appropriate, to allow log transformation. Survival analyses were performed with the log rank test. Testing was undertaken using the statistical language R (R Development Core Team, 2007; R Foundation for Statistical Computing, Vienna, Austria [http://www.R-project.org]) or GraphPad Prism 4.0 (San Diego, CA). A two-sided P value of less than 0.05 was considered statistically significant.
To identify effectors of the Bsa TTSS, the protein sequences of 85 known bacterial TTSS effectors from other bacteria were compiled from the Entrez Protein database and compared to the translated B. pseudomallei genome using BLASTP. Since B. pseudomallei has two additional TTSSs that are plant-like and are not associated with virulence in animal models, only those effectors that are secreted by bacteria pathogenic to animals, and not those pathogenic to plants, were screened. This search yielded 27 B. pseudomallei proteins, including multiple copies for some of the effectors (Table (Table2).2). A similar search was performed for the translated B. thailandensis genome to determine whether these putative effectors of the Bsa TTSS are also present in B. thailandensis. Interestingly, nearly identical orthologs of all but the E. coli EspFU, Cif, and Bordetella bronchiseptica BopN homologs, though some with different copy numbers, were found in the translated genome of B. thailandensis (Table (Table2).2). A further comparison of the protein sequences of the B. pseudomallei putative effectors to those of the B. thailandensis translated genome identified orthologs of two of these missing proteins, indicating that only the E. coli Cif homolog is absent in B. thailandensis (Table (Table1).1). Since the B. thailandensis genome also encodes a nearly identical Bsa TTSS, we hypothesized that the putative effectors in the two bacterial species have similar functions and therefore that B. thailandensis could be utilized as a BSL-1 model system to study the role of the Bsa TTSS during Burkholderia infection.
To determine whether B. thailandensis could serve as a model system to study the function of the Bsa TTSS and its effectors, the role of the Bsa TTSS during B. thailandensis infection was tested. A mutant of the Bsa TTSS was constructed by inactivating the B. thailandensis bsaZ gene, a homolog of Salmonella spaS and Shigella spa40, which encode a structural component of the TTSS. An all-frame STOP was inserted in the bsaZ gene by allelic exchange to ensure that the mutation is nonpolar. This mutant strain was then investigated for its ability to invade and replicate within epithelial cells by quantifying intracellular bacteria 2, 4, and 20 h postinfection. As shown in Fig. Fig.1A,1A, at 2 h postinfection bsaZ mutant B. thailandensis (AH174) invaded HeLa cells as wild-type B. thailandensis (700388), suggesting that the Bsa TTSS is not required for invasion of epithelial cells. However, it had a pronounced replication defect, as reflected by the smaller numbers of intracellular bsaZ mutant B. thailandensis (AH174) bacteria compared to the results seen with the wild-type strain (700388) at 4 and 20 h postinfection (Fig. 1B and C). Since the bsaZ mutant has the same growth rate in LB medium as wild-type B. thailandensis, the replication defect is a result of an intracellular growth defect due to the all-frame STOP insertion (data not shown). In addition, when trans-complemented with a wild-type copy of bsaZ, the mutant strain (AH186) replicated as well as wild-type B. thailandensis (700388) within HeLa cells, indicating that the bsaZ mutation is nonpolar and that bsaZ is required for intracellular replication (Fig. (Fig.1C1C).
To determine whether this replication defect was due to the inability of bsaZ mutant B. thailandensis to escape from the mammalian vacuole, ceftazidime was added to the culture medium after bacterial internalization. Unlike IMP, which at the concentration used cannot penetrate cellular membranes, at high concentrations ceftazidime can cross the mammalian plasma but not the vacuolar membrane. When ceftazidime was added to HeLa cells following infection, the intracellular numbers of wild-type B. thailandensis (700388) declined compared to those from infected cells treated with IMP (Fig. (Fig.1C1C and D). In contrast, the numbers of the bsaZ mutant (AH174) stayed the same regardless of which antibiotic was used. Since ceftazidime does not penetrate the vacuolar membrane and kill vacuolar bacteria, we hypothesized that the majority of wild-type B. thailandensis bacteria were then present in the cytosol and thus were susceptible to killing, whereas the bsaZ mutant was protected in the vacuole. This localization pattern was verified by constructing mutant and wild-type B. thailandensis strains expressing GFP and by visualizing the strains during infection by microscopy. As shown in Fig. 2A and B, respectively, wild-type B. thailandensis (AH183) was dispersed in the cytoplasm and formed host actin tails whereas the bsaZ mutant (AH181) was trapped in the vacuolar compartment and was unable to form actin tails. This indicates that the replication defect of the bsaZ mutant B. thailandensis strain is due to its inability to escape from the endocytic vacuole.
To demonstrate that the mutation in bsaZ abolished the ability of B. thailandensis to secrete effectors, a plasmid expressing HA-tagged B. thailandensis BopE was introduced into bsaZ mutant B. thailandensis and secretion of BopE-HA was examined in vitro. As a positive control, BopE-HA was also expressed in wild-type B. thailandensis, which has an intact Bsa TTSS. As shown in Fig. Fig.3,3, BopE-HA was expressed by both wild-type B. thailandensis (AH191) and the bsaZ mutant strain (AH194) but was secreted only by the former, indicating that the bsaZ mutant is defective in secretion of TTSS effectors. Interestingly, BopE-HA was expressed but not secreted by wild-type B. thailandensis unless the culture medium during growth was acidic, which suggested that the TTSS is not active at neutral pH. Since the Bsa TTSS is not required for invasion of HeLa cells but for escape of B. thailandensis from the mammalian vacuole, it is plausible that it is not fully functional until the bacterium encounters the acidic environment of the vacuolar compartment. To test this hypothesis, acidification of the vacuole in HeLa cells was blocked by either NH4Cl, a weak base known to neutralize the endocytic vacuole in HeLa cells, or bafilomycin A1, a specific inhibitor of the mammalian vacuolar H+ ATPase, and intracellular replication of wild-type B. thailandensis was examined. As shown in Fig. 4A and B, treatment of HeLa cells with increasing concentrations of NH4Cl and bafilomycin A1, respectively, resulted in decreased intracellular numbers of B. thailandensis (700388). In fact, this replication defect was about equal to that seen with the bsaZ mutant strain (AH174) (Fig. (Fig.4C).4C). Furthermore, microscopy of eGFP-expressing wild-type B. thailandensis (AH183) demonstrated that when vacuolar acidification was blocked with bafilomycin A1, the bacteria were unable to escape from the vacuolar compartment, as we previously observed with HeLa cells infected with bsaZ mutant B. thailandensis (AH181) (Fig. (Fig.4D).4D). These results suggest that acidic vacuolar pH is required for secretion of at least one effector, BopE, and for escape of B. thailandensis from the mammalian vacuole.
To determine whether our findings for cultured cells reflected altered bacterial virulence in vivo, we compared infection with bsaZ mutant B. thailandensis to infection with the wild-type strain in a murine pneumonia model (47). A volume of 105 CFU/lung of wild-type or 105 CFU/lung of bsaZ mutant B. thailandensis was deposited in the lungs of C57BL/6 mice, and the survival of the infected mice was assessed. Mice infected with wild-type bacteria (700388) became ill and died within 4 days, while mice that were infected with the bsaZ mutant (AH174) remained healthy for 2 weeks postinfection, when the experiment was ended (Fig. (Fig.5A).5A). To analyze bacterial replication and dissemination, bacterial loads in the lung, liver, and spleen of mice were compared 72 h after infection with the bsaZ mutant or the wild-type strain. Organ cultures revealed that pulmonary replication of the bsaZ mutant (AH174) was contained and that there was little dissemination to the liver and spleen, whereas multiorgan dissemination and exponential replication of the wild-type strain (700388) was observed (Fig. (Fig.5B,5B, C, and D). We also included the trans-complemented bsaZ mutant B. thailandensis (AH186) in our comparison; however, since this strain lost the plasmid carrying bsaZ early during infection, the infected mice survived, like those infected with the mutant strain, and had intermediate bacterial loads in the organs at 72 h postinfection (data not shown).
The discovery of an attenuated phenotype in vivo prompted us to investigate whether infection with the bsaZ mutant confers protection against wild-type B. thailandensis inhalational challenge. A total of 105 CFU/lung of bsaZ mutant bacteria was deposited by aerosol in the lungs of C57BL/6 mice. One month later, these mice as well as naïve controls were challenged with 4 × 104 CFU/lung of aerosolized wild-type B. thailandensis and the survival of the infected mice was assessed. Whereas all naïve mice died within 4 to 5 days after infection with the wild-type strain (700388), all mice previously vaccinated with the bsaZ mutant (AH174) remained well for 1 month, when the experiment was ended (Fig. (Fig.5E).5E). These results demonstrate that the Bsa TTSS is essential for full virulence of B. thailandensis in mice but is not required for generating protective immunity. The similarity of these data to those observed with B. pseudomallei in infections of mice supports the idea that B. thailandensis could serve as a good model system to study the role of the Bsa TTSS during Burkholderia pathogenesis.
B. thailandensis is not considered to be a human-pathogenic member of the Burkholderia genus. Although there have been reports of infected patients in Thailand and recently in the United States manifesting disease symptoms similar to melioidosis, the infectious dose of B. thailandensis, at least in animal models, is significantly higher than that of B. pseudomallei (4, 10, 22, 37). Its low pathogenicity for humans and the findings that it shares the virulence-associated Bsa TTSS and possibly its effectors with B. pseudomallei makes it an attractive and useful model system for the study of the intracellular lifestyle of B. pseudomallei. Although B. thailandensis has been shown to invade A549 cultured human respiratory epithelial cells less efficiently than B. pseudomallei, it can replicate in these cells as well as B. pseudomallei, form actin tails, and induce formation of multinucleated giant cells (12, 17). In fact, the BimA protein of B. thailandensis has been demonstrated to be capable of functionally complementing bimA mutant B. pseudomallei, which is unable to form actin tails and move intracellularly (38). In addition, there are a growing number of studies demonstrating the virulence of B. thailandensis in animal models (8, 48). For example, inhalation of B. thailandensis by BALB/c and C57BL/6 mice is lethal above deposition doses of 104 CFU/lung (47). Furthermore, Wiersinga and colleagues have recently shown that intranasal B. thailandensis infection induces lung pathologies in C56BL/6 mice and inflammatory responses in lung-derived cell lines similar to those seen with B. pseudomallei, suggesting that the two bacterial species have similar pathogenesis mechanisms for inbred mice (48).
Here we show that the Bsa TTSS is essential for the intracellular lifestyle and virulence of B. thailandensis. We demonstrate that a mutant defective in secretion of effectors by the Bsa TTSS invades epithelial cells as well as wild-type B. thailandensis; however, it has a pronounced replication defect due to its inability to escape from the endocytic vacuole. In addition, this mutant strain has greatly reduced virulence in a murine pulmonary infection model compared to wild-type B. thailandensis. These are findings similar to those obtained with B. pseudomallei, with the exception of the lack of an invasion phenotype observed with our mutant. Stevens and colleagues, who reported that the Bsa TTSS of B. pseudomallei is required for invasion of epithelial cells, have shown only a modest invasion defect of bipD::pDM4 B. pseudomallei, which contains an insertion in the homolog of the Salmonella translocon gene sipD (39). However, the authors quantified intracellular bacteria 6 h postinfection, by which time intracellular replication can occur. Therefore, it is possible that the differences observed between the mutant and wild-type strain results were due to the inability of the mutant to escape from the endocytic vacuole and replicate intracellularly and not to its inability to invade cells, as seen for B. thailandensis. In addition, the mutant was not complemented, so it is unknown whether the insertion had additional polar effects on the bacterium.
Since the Bsa TTSS is not required for invasion of epithelial cells but is required for escape of B. thailandensis from the endocytic vacuole, it was reasonable to assume that it is induced intracellularly, inside that compartment. Therefore, we investigated and here demonstrate that the Bsa TTSS requires acidic vacuolar pH for full functionality. This is similar to results seen with the SPI2-encoded TTSS of S. enterica serovar Typhimurium, which is also activated by the low pH of the vacuolar environment and is required for intracellular replication of the bacterium and establishment of systemic disease (5, 7, 23, 31, 36). Acidic conditions have been reported to induce expression of SPI2 genes as well as assembly of the SPI2 TTSS, resulting in translocation of Salmonella effector proteins into the cytoplasm through the vacuolar membrane. In the case of the Bsa TTSS, it is not known how the acidic environment activates it and whether translocation of its effectors also occurs through the vacuolar membrane. This raises the interesting question of whether a Burkholderia effector is directly involved in lysis of the vacuolar membrane from the cytosolic side after translocation by the TTSS or, perhaps, from the vacuolar side after secretion into the vacuolar space by the TTSS. Alternatively, it is also possible that simply, insertion of the TTSS into the vacuolar membrane causes pore formation and therefore release of the bacteria into the cytosolic space, as TTSS-induced contact-dependent hemolysis by several bacteria due to the insertion of translocon proteins into the membrane has been described previously (2, 11, 13, 24). One candidate effector from our bioinformatics screening that may cause damage to the vacuolar membrane is the homolog of Pseudomonas aeruginosa ExoU, which is a patatin-like phospholipase responsible for the cytotoxicity of P. aeruginosa toward mammalian cells (34, 35).
Currently, there is no effective vaccine for the prevention of melioidosis. Iliukhin and colleagues have proposed wild-type B. thailandensis as a potential vaccine candidate. They have shown that more than 50% of guinea pigs vaccinated with B. thailandensis were protected against challenge with a lethal dose of B. pseudomallei (14). However, since wild-type B. thailandensis has been reported to cause disease in humans and can also kill mice, it may not be a practical vaccine for humans, particularly for those with chronic diseases and immunosuppression, who would be the main candidates for such a vaccine. Here we show that bsaZ mutant B. thailandensis is avirulent in mice and elicits the same immune response as wild-type B. thailandensis, as it provides 100% protection against lethal challenge with wild-type B. thailandensis. Therefore, bsaZ mutant B. thailandensis may be a possible vaccine candidate for prevention of melioidosis.
In conclusion, our results describing the phenotypes of bsaZ mutant B. thailandensis in cultured cells and a murine infection model demonstrate that the Bsa TTSS has a role during infection with B. thailandensis similar to that of B. pseudomallei. Therefore, B. thailandensis could be used as a model system to study the function of this TTSS and its translocated effector proteins in mammalian cells. This could accelerate our understanding of B. pseudomallei pathogenesis, as research on this important bacterium is currently hampered in part by strict safety regulations and the need of investigators for BSL-3 facilities. Furthermore, the Bsa TTSS is also present in B. mallei, another category B agent that is responsible for human and animal glanders. Since the B. mallei Bsa TTSS has a function in cultured cells and mice equivalent to those of B. thailandensis and B. pseudomallei, the use of B. thailandensis to elucidate the role of this TTSS during infection may also provide more insight into the mechanism by which B. mallei exploits host cells and causes disease (30, 33, 45).
We thank Loren Kinman and Matthew McKevitt for helping with some of the experiments and Debra Milton and Miguel Valvano for generously providing reagents.
This work was supported by Career Development Awards to A. H. and T.E.W. and other grants to the authors as part of the Northwest Regional Center of Excellence for Biodefense and Emerging Infectious Diseases Research (National Institute of Allergy and Infectious Diseases grant U54 AI057141).
Editor: J. B. Bliska
Published ahead of print on 8 September 2008.