Previous studies suggested that the persistence in eukaryotic cells of a Bordetella bronchiseptica mutant carrying an insertion in the locus encoding the response regulator RisAS is impaired. This suggested that ris-dependent products are required for the intracellular survival of bacteria. In this study we demonstrate that ris-regulated products play a role in B. bronchiseptica resistance against both phagosomal acidification and reactive oxygen intermediates.
Bordetella bronchiseptica is a pathogen of the respiratory tract which can cause disease in a broad range of warm-blooded animals . Several in vitro studies demonstrated that Bordetella spp. are able to survive within dendritic, epithelial or macrophage-like cells [2–9]. The persistence in a cellular reservoir may allow bacteria to escape from host clearance mechanisms, favoring a chronic course of the infection.
In Bordetella the expression of many virulence genes is tightly regulated at the transcriptional level by the bvg locus (Bordetella virulence gene) in response to environmental signals [10,11]. The bvg locus encodes two proteins, BvgS and BvgA, which are the environmental sensor and transcriptional activator, respectively, of a two-component regulatory system. In contrast to other Bordetella spp., in which intracellular survival seems to be dependent on bvg-activated products, both wild-type B. bronchiseptica and bvg-negative mutants are able to invade and survive within eukaryotic cells [2,6–9]. Furthermore, bvg-negative B. bronchiseptica mutants may have a selective advantage for long term survival within macrophages , suggesting that bvg-independent or bvg-repressed products may also be required for intracellular survival in macrophages.
Previous studies from our group led to the identification of a novel two-component regulatory system, RisAS, which is present in all Bordetella spp. and is required for bacterial production of acid phosphatase. A B. bronchiseptica mutant carrying a mini-transposon insertion in risAS exhibited an impaired colonization of the mouse respiratory tract . To further characterize the phenotype of the ris mutant and to elucidate the underlying events leading to the observed phenotype, infection studies were performed using mouse peritoneal macrophages.
2Materials and methods
2.1Bacterial strains and media
B. bronchiseptica strain BB7865 and its ris derivative  were grown on Bordet–Gengou (BG) agar (Difco Laboratories, Detroit, MI, USA) supplemented with 1% glycerol and 10% (v/v) defibrinated horse blood, as previously described [6,7].
2.2Tissue culture methods and invasion studies
Macrophages were collected from the peritoneal cavity of 4–6-week-old female BALB/c mice as previously described  and resuspended in Dulbecco's modified Eagle medium containing 25 mM HEPES, 1% fetal calf serum, 2 mM glutamine and 1.5 g l−1 Na2CO3 (Gibco, Karlsruhe, Germany). Macrophages (1×106 cells per well) were seeded on 24-well Nunclon Delta tissue culture plates (Inter Med Nunc, Roskilde, Denmark) and incubated 4 h in an atmosphere containing 5% CO2 at 37°C. Then, they were infected for 1 h as previously described [6,7]. Extracellular bacteria were removed by washing with phosphate-buffered saline (PBS, pH 7.4) and tissue culture medium supplemented with gentamicin (50 μg ml−1) was added to eliminate extracellular bacteria. After washing the monolayer (time 0) and at the different intervals cells were lysed with 0.025% Triton X-100 in PBS and the number of viable intracellular bacteria was determined by plating the lysates onto Brain Heart Infusion agar (Difco Laboratories).
2.3In vivo colonization studies
Approximately 106 bacteria were resuspended in 1% casamino acids in PBS and administered intranasally to 4-week-old female BALB/c mice. Groups of animals (n=3) were sacrificed at different time intervals after challenge, lungs were homogenized using a Polytron PT 1200 (Kinematica, Switzerland) and the number of viable microorganisms was determined by plating appropriate dilutions onto BG agar supplemented with cephalexin (50 μg ml−1).
The statistical significance of the results has been evaluated by the Student's t test using the Statgraphic plus for windows 2.0 software (Statistical Graphic Corp.), differences were considered significant at P≤0.05.
3Results and discussion
3.1Evaluation of bacterial survival within peritoneal macrophages
To evaluate the role played by the products encoded by the ris locus in B. bronchiseptica survival within non-activated professional phagocytes, mouse peritoneal macrophages were infected with wild-type bacteria and the corresponding ris derivative. Subsequently, the number of viable intracellular bacteria was determined at different time intervals post infection, as described in Section 2. The obtained results showed that wild-type B. bronchiseptica can survive better than the ris mutant within professional phagocytes (Fig. 1).
Then, 4-week-old female BALB/c mice were infected intranasally with BB7865 or BB7865ris. Two hours and 5 days after challenge three animals were sacrificed per group, their lungs were homogenized and the number of viable bacteria was determined by plating dilutions onto BG agar with cephalexin (50 μg ml−1). No significant differences were observed between the wild-type strain and its ris derivative in their capacity to establish an infection in mice. However, while the number of wild-type bacteria was increased following challenge, the ris mutant was unable to grow and bacterial clearance was observed shortly after challenge (Fig. 2). The results obtained in these in vitro and in vivo studies indicate that B. bronchiseptica is able to persist within professional phagocytes for only a limited time period and that bacterial clearance is accelerated at the very early stages of the infection process for the ris derivative. This confirmed our hypothesis that ris-regulated genes might be involved in the process of bacterial persistence within host cells.
3.2Characterization of the underlying mechanisms involved in bacterial survival
One of the mechanisms which enables bacterial survival within eukaryotic cells is the neutralization of phagosomal acidification [14–17]. The remarkable difference in survival between the tested strains, and the chronology of bacterial death suggested that intracellular acidification may be an important killing mechanism and that ris-regulated products are somehow affecting this process. To investigate this hypothesis infection studies were performed, in which ammonium chloride was added to inhibit phagosomal acidification. Peritoneal macrophages were infected for 1 h as described above, but after washing to eliminate extracellular bacteria (time 0), macrophage monolayers were incubated for 2 h with ammonium chloride (25 mM). Ammonium chloride was then removed, and cells were further incubated for 2 h with normal infection medium. The impairment of endosomal acidification resulted in a significant increment in B. bronchiseptica survival, being a similar number of viable bacteria recovered from cells infected with either the wild-type or the ris mutant (Fig. 3). This suggests that phagosomal acidification is one of the major bacterial killing mechanisms in this experimental setting, and that ris-dependent products might inhibit this process.
One of the mechanisms by which eukaryotic cells can decrease the phagosomal pH exploits a cellular proton ATPase. To assess the possibility that ris-regulated products can affect this system, peritoneal macrophages were pre-treated during 15 min with the vacuolar proton pump inhibitor bafilomycin (100 nM) and then infected with the B. bronchiseptica strains. Although an increment in intracellular survival was observed for both B. bronchiseptica strains, there was still a remarkable difference in number of viable bacteria recovered from cells infected with the wild-type versus the ris mutant (Fig. 4). Based on these results we can conclude that the ATPase-mediated acidification of the phagosome is only partially responsible for bacterial killing. Therefore, additional mechanism(s) might also be involved in this process.
The production of reactive oxygen intermediates (ROI) is also involved in the bactericidal activity of macrophages. Thus, we examined the role of these agents in bacterial killing by performing macrophage infections in the presence of superoxide dismutase (SOD, 150 U ml−1). Elimination of ROI by SOD resulted in the recovery of a higher number of viable bacteria, in comparison to experiments performed under normal conditions (Fig. 5). These results suggested that the production of oxidative agents plays also a role in anti-Bordetella activity of peritoneal macrophages. On the other hand, bacterial expression of ris-regulated products seems to be involved in the inactivation of ROI.
The innate immune system, particular professional scavenging cells, plays a central role in the initial clearance of bacterial infections . The results obtained here demonstrate that ris-regulated products are required for microbial survival within macrophages, being the impairment of phagosomal acidification and the inactivation of ROI among the underlying effector mechanisms. The acid phosphate produced by B. bronchiseptica, which is up-regulated by ris[12,19], may be a crucial bacterial product for the impairment of endosomal acidification. However, more than one factor and/or regulator seem to be required for the overall B. bronchiseptica survival. The presence of multiple effector mechanisms to fulfil a singular task is a common theme in many biological systems. This functional redundancy ensures the preservation of critical processes, such as the capacity of B. bronchiseptica to persist for long periods in a safe intracellular niche.
This work was in part supported by a Roman Herzog-Fellowship (The Alexander von Humboldt Foundation) awarded to K.Z.