INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Bordetella pertussis, the etiologic agent of the severe respiratory disease whooping cough, expresses an array of virulence factors that contribute to its ability to cause disease (13, 14, 34, 46, 48). These factors include cell surface proteins that are thought to be involved in the adherence of the organism to ciliated respiratory epithelial cells, as well as several extracellular toxins that inhibit host defenses and induce damage to host tissues. It is believed that transmission of B. pertussis occurs via aerosol droplets expelled by severe coughing that pass directly from the respiratory tracts of infected individuals to the respiratory tracts of susceptible hosts, who inhale the aerosolized bacteria (20). No animal reservoir for B. pertussis has been identified, and the bacterium appears unable to survive in the environment for prolonged periods of time (17, 35). Once the bacterium is present in the respiratory tract, it adheres to ciliated cells. This interaction is presumably mediated by the adhesins, filamentous hemagglutinin (FHA), pertactin, fimbriae, and possibly tracheal colonization factor (19, 46). The damage to the respiratory epithelium characteristically observed in B. pertussis-infected individuals is presumably caused by tracheal cytotoxin and perhaps other toxins produced by the bacterium. Bacterial toxins adenylate cyclase and pertussis toxin (PT) can inhibit immune system affector cells and may protect the bacteria from clearance later in infection. B. pertussis does not typically disseminate from the respiratory tract or establish a chronic infection, nor is there evidence of significant asymptomatic carriage of the bacterium (17, 21, 24, 26).

The regulated expression of virulence factors in B. pertussis, with the exception of tracheal cytotoxin, is activated at the level of transcription by a single locus, referred to as the bvg locus (originally designated the vir locus) (4, 41, 42, 45, 47). The bvg locus encodes three proteins: BvgA, BvgS, and BvgR (31). BvgS is a 135-kDa transmembrane protein that is thought to be responsible for sensing an environmental signal. Although the relevant environmental signal(s) to which BvgS responds in vivo is unknown, the activity of the bvg locus is repressed when cells are grown in the presence of MgSO₄ or nicotinic acid or when they are grown at reduced temperature in vitro (25). This bvg-mediated change in the patterns of transcription in response to environmental signals is referred to as phenotypic modulation. Under nonmodulating conditions, the autophosphorylation of BvgS at a conserved histidine residue is followed by two intramolecular phosphotransfer reactions ultimately leading to the transfer of the phosphate moiety to a conserved aspartate residue on BvgA (44). Upon phosphorylation by BvgS, BvgA, a 23-kDa cytoplasmic protein, binds to cis-acting sequences in the promoter regions of the bvg-activated genes and activates transcription (7, 8, 22). Transcription of bvgR is activated by bvgA (30). The product of the bvgR gene is responsible for the repression of a class of genes referred to as the bvg-repressed genes (also referred to as vir-repressed genes or vrg's) (31). The nature and role(s) of the bvg-repressed genes are essentially unknown. It is speculated that the proteins encoded by the bvg-repressed genes may be involved in the establishment or persistence of B. pertussis in the host or in the survival of the organism either within a specialized niche in the host or outside of the host.

The presence and continued maintenance of the bvg-activated and bvg-repressed genes suggest that B. pertussis experiences at least two environments during its infectious life cycle. Under one set of conditions the expression of the classically defined virulence factors is required, while under a different set of environmental conditions, the expression of the bvg-repressed genes is advantageous. If, however, our current understanding of the transmission of B. pertussis and the etiology of whooping cough is accurate, there is no obvious period in its life cycle during which the virulence factors would be turned off. It is difficult, therefore, to postulate a role for the regulation of virulence factors in the infectious life cycle of the bacterium. It is possible that the bvg-mediated regulation of gene expression in B. pertussis is an evolutionary remnant. The closely related bacterium Bordetella bronchiseptica causes respiratory disease in a variety of animal species (16). B. bronchiseptica expresses many of the same virulence factors as B. pertussis, and their expression is regulated by the bvg locus (5, 16, 33). B. bronchiseptica also expresses several bvg-repressed factors, although these gene products are not the same as those encoded by the bvg-repressed genes in B. pertussis (2, 11, 15). Unlike B. pertussis, B. bronchiseptica is capable of growth outside the host in a nutrient-poor environment (37). The ability to grow in a nutrient-poor environment appears to be enhanced by the ability to express the bvg-repressed genes (12). Therefore, in B. bronchiseptica, there is an apparent role for bvg-mediated regulation of gene expression. It is possible that the bvg locus evolved within an ancestral Bordetella strain that was a pathogen capable of free living outside of its host. Divergent evolution from this strain could have resulted in the emergence of two strains: B. bronchiseptica, which retained the ability to survive outside of the host and the requirement for regulation by the bvg locus, and B. pertussis, which developed an infectious life cycle that no longer included a phase outside of its host. According to this model, the bvg locus has been retained in B. pertussis only because of its role in activating the expression of the bvg-activated virulence factors, not because of its ability to regulate the transition from one phase to another. If the bvg locus is required to mediate the transition between two environments, the nature of the alternative environment or niche and whether it lies inside or outside of the human host remain to be determined.

Previous studies have demonstrated that the constitutive expression of the bvg-activated genes and the inability to express the bvg-repressed genes do not reduce the ability of B. bronchiseptica to colonize and persist within the respiratory tract of that bacterium's natural host (1, 12, 28). Furthermore, those studies demonstrated that the inappropriate expression of at least one of the bvg-repressed genes in B. bronchiseptica interferes with the bacterium's ability to colonize its host. These results are consistent with a possible role for bvg regulation in B. bronchiseptica: that of mediating the transition from infection of the host to survival outside the host. Although B. bronchiseptica and B. pertussis are closely related, with respect to virulence they are very different. B. pertussis and B. bronchiseptica infect different hosts, and within their respective hosts, they cause different diseases (16). PT, which is one of the major and essential virulence factors in B. pertussis, is not expressed in B. bronchiseptica (18, 36). The modulated states of B. pertussis and B. bronchiseptica appear to be very different. B. pertussis and B. bronchiseptica express different bvg-repressed genes (2, 6, 15, 23, 29). While the bvg-repressed genes of B. bronchiseptica appear to be involved in survival outside of the host, the function(s) of the bvg-repressed genes in B. pertussis is unknown, but presumably they are not involved in long-term survival outside of the host since B. pertussis has not been demonstrated to have that capability. Because of these significant differences, it is not possible to predict from the results of previous virulence studies using B. bronchiseptica what contribution bvg-mediated regulation makes to the virulence of B. pertussis or what role the products of the bvg-repressed genes of B. pertussis play in that bacterium's infectious life cycle.

In this study, the contribution of bvgR-mediated regulation to the virulence of B. pertussis and the importance of the expression of the bvg-repressed genes for virulence were evaluated. We examined the ability of B. pertussis mutants, defective in their ability to regulate the expression of either the bvg-activated or the bvg-repressed genes or both, to cause disease in the mouse aerosol challenge model. Our results indicate that the bvgR-mediated regulation of gene expression contributes to respiratory infection of mice.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Bacterial strains, plasmids, oligonucleotides, and media. The bacterial strains and plasmids which were used in this study are presented in Table 1. Escherichia coli strains were grown on L agar or in L broth supplemented with antibiotics when appropriate (31). B. pertussis strains were grown on Bordet-Gengou (BG) agar (Difco) containing 1% proteose peptone (Difco) and 15% defibrinated sheep blood. Concentrations of antibiotics were as follows: gentamicin sulfate, 10 µg/ml; kanamycin sulfate, 10 µg/ml; nalidixic acid, 50 µg/ml; rifampin, 50 µg/ml; and streptomycin sulfate, 100 µg/ml, unless stated otherwise. Plasmids were transformed into E. coli DH5 (Bethesda Research Laboratories, Bethesda, Md.) and S17 (39).

Strain and plasmid construction. Strain TM1396 bearing a constitutive allele of bvgS was constructed as follows. B. pertussis 18323 (American Type Culture Collection, Manassas, Va.) was mated with E. coli SM10 bearing plasmid pJM503 (32). Exconjugates were selected by growth on BG plates containing colicin and gentamicin. Exconjugates were plated on BG plates supplemented with 50 mM MgSO₄ in the absence of antibiotic selection in order to identify isolates in which plasmid sequences had spontaneously crossed out and the bvgS(Con) allele was retained on the chromosome. Hemolytic colonies were selected and were restreaked in order to insure homogeneity of the colony isolate. A single gentamicin-sensitive isolate that constitutively expressed hemolytic activity was selected and designated TM1396.

View larger version (14K): [in this window] [in a new window]   FIG. 1.   Construction of plasmid pTM126. The construction of plasmid pTM126 bearing an in-frame 606-bp internal deletion of bvgR is diagrammed. B. pertussis sequences are indicated by stipled boxes. Nucleotide numbers given correspond to those of the published sequence of the bvgR locus (30). Details of the construction are presented in Materials and Methods.

Quantitative alkaline phosphatase assays. Alkaline phosphatase assays were performed as described previously (31).

Preparation of B. pertussis protein samples for electrophoresis. B. pertussis 18323, TM1396, TM1423, and TM1424 were grown overnight at 37°C in Stainer-Scholte media either in the presence of 50 mM MgSO₄ or in the absence of MgSO₄. All of the bacterial cultures grew to approximately the same optical density (OD) and were further adjusted to an OD at 600 nm (OD₆₀₀) of 4.5. Bacterial cells were harvested from culture media by centrifugation, and the supernatant fraction and cell pellets were prepared for analysis as previously described (27). Additionally, cell pellet preparations were maintained on ice and sonicated four times for 15 s. When necessary, electrophoresis samples were further diluted in sodium dodecyl sulfate (SDS)-gel loading buffer.

Western blots. Samples were analyzed by electrophoresis with SDS-10% polyacrylamide gels. Electrophoretic transfer of proteins to nitrocellulose membranes was accomplished by a modification of the procedures of Towbin et al. and Burnette (9, 43); a Novex electroblot apparatus was used in accordance with the manufacturer's instructions. After transfer of the proteins, the nitrocellulose membranes were incubated either overnight at 4°C or for 1 h at room temperature in phosphate-buffered saline (PBS) containing 5% (wt/vol) nonfat dry milk. Membranes were incubated overnight at 4°C or for 2 h at room temperature in PBS containing 1% (wt/vol) nonfat dry milk and the appropriate primary antibody. Membranes were washed three times for 30 min in PBS containing 0.05% Tween 20. The blots were incubated for 1 h with either antirabbit or antimouse antibodies conjugated to horseradish peroxidase diluted in PBS containing 0.05% Tween 20. The membranes were washed three times in PBS containing 0.05% Tween 20, and the reacting proteins were visualized with luminol by using the ECL Western blotting analysis system (Amersham International) or SuperSignal chemiluminescent substrate (Pierce) in accordance with the manufacturer's instructions. The chemiluminescent signal was detected by immediate exposure to X-ray film. Primary antibodies used in this study were the following: anti-FHA (rabbit polyclonal antibodies from H. Sato); anti-PT S1 subunit (monoclonal antibody 1B7 from H. Sato); and anti-VraA and anti-VraB (monoclonal antibodies 1G7-8 and 7H1A-5, respectively, from M. Peppler).

Mice. Specific-pathogen-free BALB/cAnNcR mice were obtained at 16 days of age from the Animal Production Program, Division of Cancer Treatment, National Cancer Institute, Frederick, Md. Mice were maintained in microisolators under specific-pathogen-free conditions.

Mouse aerosol challenge. A 21-h culture of each B. pertussis strain grown on BG agar was suspended in sterile PBS at a concentration of approximately 2 × 10⁹ CFU/ml of inoculum. Each strain of challenge inoculum was administered to 17-day-old mice in a separate aerosol challenge for 30 min as described previously (38). Mice were removed from the chamber 1 h after termination of the aerosol challenge, at which time there were no viable B. pertussis cells on the surfaces of the animals or within the chamber that could be cultured. Two mice from each challenge were sacrificed upon removal from the chamber in order to determine the number of viable B. pertussis cells in the lungs. Lungs and tracheas were aseptically removed and were homogenized in 5 (lungs) or 1 ml (tracheas) of sterile PBS. Ten-fold dilutions of homogenates were plated on BG agar plates in order to determine the number of bacteria that could be recovered. Groups of six mice challenged with an inoculum of each bacterial strain were sacrificed 14 days after the challenge, and dilutions of homogenized lungs and tracheas were plated to determine the levels of viable bacteria persisting in the animals. A separate group of 10 mice challenged with an inoculum of each bacterial strain was bled 7 and 14 days after the challenge, and the levels of circulating leukocytes (WBC) were determined. These mice were observed for 21 days to note any deaths. The number of WBC per microliter of blood was determined in a model ZM Coulter Counter.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Analysis of the expression of selected bvg-regulated genes. Strains of B. pertussis which were defective in their ability to regulate the expression of bvg-regulated genes were constructed. In order to confirm that the bvg-activated and bvg-repressed genes were being transcribed and regulated in the manner expected for each strain, transcriptional fusions of the alkaline phosphatase gene of E. coli to the fha, ptx, and vrg6 loci of B. pertussis were crossed onto the chromosomes of strains 18323, TM1396, TM1423, and TM1424. The alkaline phosphatase activities of the resulting strains were assayed by quantitative enzyme assay after growth either in the presence of 50 mM MgSO₄ or in the absence of MgSO₄. The results of this analysis are presented in Table 2. In strain 18323 (wild type), the fha-phoA, ptx-phoA, and vrg6-phoA fusions were all regulated normally. In strain TM1396 [bvgS(Con)], the fha-phoA and ptx-phoA fusions were constitutively expressed and the vrg6-phoA fusion was constitutively repressed. In strain TM1424 (bvgR), the fha-phoA and ptx-phoA fusions were regulated normally and the vrg6-phoA fusion was constitutively expressed. In strain TM1423 [bvgS(Con) bvgR], the fha-phoA, ptx-phoA, and vrg6-phoA fusions were all constitutively expressed. All of the bvg-regulated loci examined in this analysis demonstrated the expected pattern of expression for each of the mutant strains.

View larger version (47K): [in this window] [in a new window]   FIG. 2.   Western blots of bvg-activated and bvg-repressed gene products. Wild-type B. pertussis and the regulation mutants were grown in the presence of 50 mM MgSO4 (+) and in the absence of MgSO4 (). Precipitated supernatants were probed with antibodies that specifically recognize PT and FHA, and cell pellets were resuspended and probed with antibodies that specifically recognize the bvg-repressed surface antigens VraA and VraB.

Analysis of the growth of bvg regulation mutants. The growth rate of each bvg mutant strain was compared with that of the wild-type strain in vitro in order to determine the effect, if any, of the bvg regulation mutations on the growth efficiency. Cells in stationary phase were pelleted and resuspended in Stanier-Scholte media at a starting OD₆₀₀ of 0.1. Their growth was monitored by measuring the increase in turbidity of the culture at 37°C with aeration until stationary phase was reached. The relative growth rates of the mutant strains were indistinguishable from that of the wild-type strain (Fig. 3). Thus, although the expression of bvg-regulated genes was affected in these mutants, the growth efficiency of each strain was not impaired by the introduced mutations.

View larger version (10K): [in this window] [in a new window]   FIG. 3.   Growth curves of bvg mutant strains. B. pertussis 18323 (wild type; open triangle), TM1396 [bvgS(Con); solid triangle], TM1424 (bvgR; open circle), and TM1423 [bvgS(Con) bvgR; solid circle] were inoculated into Stainer-Scholte media at an initial OD600 of 0.1 and grown to stationary phase. Each point shown on the curves is the average of the values from two independent cultures. Error bars represent the standard deviations from the means. Only those error bars that were large enough to be discernible are shown.

Aerosol challenge of mice with bvg mutants of B. pertussis. The contribution of bvg-mediated regulation to the virulence of B. pertussis was determined by measuring the ability of each of the bvg mutants to cause disease in the mouse aerosol challenge model relative to that of the wild-type strain. Wild-type strain 18323 colonizes and proliferates to high titers in the lungs and trachea, induces high levels of leukocytosis, and causes death in 100% of animals by 21 days postinfection. The ability of the wild-type and mutant strains to colonize and proliferate in the upper respiratory tract of the mouse is reflected in the numbers of CFU in the lungs and tracheas of infected animals. All of the animals tested had between 10⁴ and 10⁵ CFU in their lungs 1 h after aerosol challenge regardless of the strain with which they were challenged, demonstrating that all of the test groups were initially infected with approximately equal numbers of bacteria (data not shown). It was clear that all four bacterial strains colonized the lungs and tracheas of infected mice and proliferated to high titers at both sites. Fourteen days after infection, the growth of all of the strains resulted in between 10⁸ and 10⁹ CFU in the lungs of infected mice and between 5 × 10⁵ and 5 × 10⁶ CFU in the tracheas of infected mice (Fig. 4). Although all of the bacterial strains colonized the lungs and tracheas of infected mice and proliferated to high numbers, there was statistically significant variation between experiments as to which strains grew to the highest levels. Therefore, it was not possible to conclude that the proliferative capacity of any of the mutant strains was, or was not, significantly different from that of the wild-type strain.

View larger version (39K): [in this window] [in a new window]   FIG. 4.   Colonization, proliferation, and persistence of wild-type and bvg mutant strains of B. pertussis in the mouse respiratory tract. Mouse aerosol challenges were performed as described in Materials and Methods. The numbers of CFU recovered from the lungs (A) and tracheas (B) of mice infected with the wild-type and bvg mutant strains were determined. The values reported are the averages of two independent experiments. Error bars represent the standard errors of the means. P values are for comparison with the wild type. Those values that are significantly different from the wild-type value are indicated as follows: *, P < 0.01; **, P < 0.001; and ***, P < 0.0001.

View larger version (29K): [in this window] [in a new window]   FIG. 5.   Ability of wild-type and bvg mutant strains of B. pertussis to induce leukocytosis in the mouse. Mouse aerosol challenges were performed as described in Materials and Methods. Leukocytosis values represent the numbers of WBC per microliter of blood from mice infected with the indicated strains on days 7 and 14 postinfection. The values reported are the averages of two independent experiments. Error bars represent the standard errors of the means. P values are for comparison with the wild type. Those values that are significantly different from the wild-type value are indicated as follows: *, P < 0.01; **, P < 0.001; and ***, P < 0.0001.

View larger version (19K): [in this window] [in a new window]   FIG. 6.   Survival of mice infected with wild-type and bvg mutant strains of B. pertussis. Mouse aerosol challenges were performed as described in Materials and Methods. Percent survival is defined as the percentage of mice infected with the indicated strains surviving at the indicated number of days postinfection. The values reported are the averages of two independent experiments. P values are for comparison with the wild type. Those values that are significantly different from the wild-type value are indicated as follows: *, P < 0.01; **, P < 0.001; and ***, P < 0.0001.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Bacterial pathogens produce a wide array of factors that enable them to exploit the environment provided by their host. These factors contribute to the bacterium's ability to colonize and persist within the host, evade or disable host defenses, and ultimately effect transmission to a new host. For many of these pathogens, a major factor in their success is the ability to adapt rapidly to the many environments to which they are exposed during their infectious life cycles. It is not surprising that bacterial pathogens have developed the ability to regulate the expression of their virulence genes since they must have the capacity to adapt to the external environment and to make the transition from the external environment to the environment within the host or from one niche within the host to another. If our current understanding of the transmission of B. pertussis and the etiology of whooping cough is accurate, there is no obvious period during which one would expect expression of the virulence factors to be turned off, and thus it is difficult to postulate a role for the regulation of virulence factors in the infectious life cycle of the bacterium.

Since strain TM1396 [bvgS(Con)] has no discernible defect in its virulence properties relative to the wild-type strain; expression of the bvg-activated genes appears to be sufficient for the bacterium to cause disease in the mouse, and regulation of this expression is not required for virulence. Furthermore, since the bvg-repressed genes are constitutively repressed in strain TM1396 [bvgS(Con)], it is clear that the expression of the bvg-repressed genes is not required for B. pertussis to cause disease in the mouse. If there is a phase during which B. pertussis modulates expression of the bvg-regulated genes inside the host, it must occur late in infection, at a time after the bacteria have established an infection and caused disease.

The higher survival ratio of mice infected with strain TM1424 (bvgR) suggests that the inappropriate expression of one or more of the bvg-repressed genes interferes with the ability of B. pertussis to cause disease. This establishes that, in addition to activating the expression of the bvg-activated genes, another necessary function of the bvg locus is the repression of the bvg-repressed genes at some point in the infectious process. It is intriguing that while the virulence of strain TM1424 (bvgR) is attenuated for mice, the ability of strain TM1423 [bvgS(Con) bvgR) to cause disease is the same as that of wild-type B. pertussis. This suggests that constitutive expression of the bvg-activated genes compensates for the defect that results from the constitutive expression of the bvg-repressed genes. The inappropriate expression of one or more of the bvg-repressed gene products may interfere with the expression, localization, or function of one or more of the bvg-activated gene products, and constitutive expression of the bvg-activated genes may overcome this interference. Alternatively, it is possible that one or more of the bvg-repressed genes provides a target that is exploited by host defenses. Constitutive expression of one or more of the bvg-activated genes may interfere with the expression or localization of the bvg-repressed factor(s) that is targeted by the host. If the inability of strain TM1424 (bvgR) to cause disease in the mouse is a result of the host's ability to more effectively respond to the bacterial infection, strain TM1424 (bvgR) would not be expected to colonize well or persist within the host. This is not the case, which leads us to support the former interpretation. Future studies will address the mechanism by which constitutive expression of the bvg-repressed genes interferes with pathogenesis and how that interference is overcome by constitutive expression of the bvg-activated genes.

We can postulate at least two alternative models to explain the role of bvg-mediated regulation of gene expression in the virulent life cycle of B. pertussis. According to one model, modulation of gene expression by the bvg locus is an evolutionary remnant that does not normally occur during the infectious life cycle of the bacterium because it never, or only transiently, experiences an environment in which expression of the bvg-activated genes is not required. Our observations that the ability to turn off the bvg-activated genes and activate expression of the bvg-repressed genes is not required to cause disease in the mouse supports that interpretation. A second model predicts that B. pertussis experiences at least two environments during its infectious life cycle. If this is the case, our results indicate that in vivo modulation of bvg-mediated gene expression is more likely to occur late in the infectious cycle. The modulated state may represent an as yet unrecognized carrier state or, alternatively, may be a state that facilitates transmission to a new host.

The ability to regulate expression of the bvg-activated genes is not essential to cause disease in the mouse, and expression of the bvg-repressed genes is not required for virulence. The inappropriate expression of the bvg-repressed genes does, however, interfere with the bacterium's ability to cause disease. These observations, although important, are not more supportive of one model over the other. The fact that modulation of the bvg-activated genes and expression of the bvg-repressed genes are not required for colonization, persistence, and the induction of damage to the host does not eliminate the possibility that modulation of gene expression by the bvg locus is required at later stages of the bacterial infectious life cycle. The existence of bvgR and the bvg-repressed genes suggests that modulation does occur and is important for the bacterium's survival. It seems unlikely that the function of the bvg locus and the integrity of the bvg-repressed open reading frames would have been maintained if they did not confer an evolutionary advantage to the bacterium. The further identification and characterization of bvg-repressed genes in B. pertussis should provide insights into the function of these genes and the role of the modulated state in the life cycle of the bacterium. In addition, the development of a model system that allows for the examination of the entire infectious life cycle of B. pertussis will be necessary in order to evaluate the contribution of the regulation by the bvg locus to the ability of B. pertussis to persist and spread within the human population.