Abstract
Francisella tularensis is a facultative intracellular bacterial pathogen capable of proliferating within host macrophages. The mechanisms that explain the differences in virulence between various strains of the species are not well characterized. In the present study, we show that both attenuated (strain LVS) and virulent (strains FSC200 and SCHU S4) strains of the pathogen replicate at similar rates in resting murine peritoneal exudate cells (PEC). However, when PEC were activated by exposure to gamma interferon (IFN-γ), they killed LVS more rapidly than virulent strains of the pathogen. Addition of NG-monomethyl-l-arginine, an inhibitor of inducible nitric oxide synthase, to IFN-γ-treated PEC, completely inhibited killing of the virulent strains, whereas it only partially blocked the killing of LVS. Similarly, in a cell-free system, SCHU S4 and FSC200 were more resistant to killing by H2O2 and ONOO− than F. tularensis LVS. Catalase encoded by katG is a bacterial factor that can detoxify bactericidal compounds such as H2O2 and ONOO−. To investigate its contribution to the virulence of F. tularensis, katG deletion-containing mutants of SCHU S4 and LVS were generated. Both mutants demonstrated enhanced susceptibility to H2O2 in vitro but replicated as effectively as the parental strains in unstimulated PEC. In mice, LVS-ΔkatG was significantly attenuated compared to LVS whereas SCHU S4-ΔkatG, despite slower replication, killed mice as quickly as SCHU S4. This implies that clinical strains of the pathogen have katG-independent mechanisms to combat the antimicrobial effects exerted by H2O2 and ONOO−, the loss of which could have contributed to the attenuation of LVS.
Francisella tularensis is a gram-negative bacterium and the etiological agent of tularemia in humans and many other mammalian species (21). The species F. tularensis comprises F. tularensis subsp. mediasiatica, holarctica, and tularensis (27). Reports of tularemia in humans elicited by F. novicida and F. tularensis subsp. mediasiatica are rare, whereas cases of tularemia caused by F. tularensis subsp. holarctica regularly occur in many countries and cases caused by F. tularensis subsp. tularensis occur in North America. The latter two, also designated type B and type A, respectively, are extremely infectious, and inhalation or intradermal inoculation of as few as 10 CFU of either is sufficient to cause disease in humans (21). A human live-vaccine strain (LVS) of F. tularensis was derived from a strain of F. tularensis subsp. holarctica in the 1940s in the former Soviet Union (6). It was further developed in the United States during 1950s and thereafter designated F. tularensis LVS. Vaccination of mice or humans with LVS conferred measurable protection from a challenge with virulent type A strains. LVS retains some residual virulence for mice, but it is still considerably less virulent than the wild type A and B strains (7). The aerosol and intradermal 50% lethal dose (LD50) of LVS for mice is approximately 100- and 100,000-fold higher, respectively, than that of clinical strains of the pathogen. The mechanisms underlying the attenuation of LVS are essentially unknown. However, in vivo, LVS is more susceptible than clinical strains to being killed by various host defense mechanisms, including gamma interferon (IFN-γ)-dependent host defenses (5).
The ability to survive or suppress the normally antibacterial activities of host macrophages is considered to be critical for the virulence of F. tularensis (24). In vitro, both LVS and clinical strains of the pathogen can proliferate within resting macrophages. Previously, we and others have shown that activating resting macrophages by exposing them to IFN-γ enhances their ability to kill LVS by a mechanism involving both reactive oxygen and nitrogen species (ROS and RNS, respectively) produced by phagocyte oxidase and inducible nitric oxide synthase (iNOS), respectively (10, 11, 14, 22). The latter generates NO that combines with reactive oxygen to form peroxynitrite (ONOO−) (23), which is a strong oxidant and nitrating agent and a key mediator of the IFN-γ-induced killing of LVS by murine macrophages in vitro (14). However, it remains to be determined whether this mechanism is effective against clinical strains of the pathogen. In this regard, like other intracellular pathogens of professional phagocytes, F. tularensis can produce enzymes that can metabolize and neutralize ROS and RNS (3, 29). Examples of such enzymes are superoxide dismutases (SODs), catalase (KatG), glutathione peroxidase, and alkyl-hydroperoxide reductase. Each of the genomes of SCHU S4, FSC200, and F. tularensis LVS contains genes encoding these enzymes. They have overlapping catalytic activities and, among other things, catalyze reduction and the resulting detoxification of H2O2, organic hydroperoxides, and ONOO− (2, 3, 28). Infection with a FeSOD mutant resulted in a lower mortality than did infection with the wild-type strain F. tularensis LVS (1), but the functions and roles of the other antioxidant enzymes have not been investigated. The present study was conducted to address these issues. In particular, it examined the relative resistance of attenuated and clinical isolates of F. tularensis to killing by ROS and RNS and by resting and IFN-γ-activated macrophages. It also compares the virulence of wild-type strains of the pathogen with that of mutants in which the katG gene has been deleted.
MATERIALS AND METHODS
Bacterial strains used in this study.
F. tularensis LVS was originally obtained from the American Type Culture Collection (ATCC 29684). F. tularensis strains SCHU S4 (F. tularensis subsp. tularensis) and FSC200 (F. tularensis subsp. holarctica) were obtained from the Francisella Strain Collection of the Swedish Defense Research Agency, Umeå, Sweden. The LVS-ΔkatG and SCHU S4-ΔkatG strains were obtained by allelic replacement of the katG gene by a previously described procedure (12) with modifications as described below.
Fragments approximately 1,100 bp upstream and downstream of katG were amplified by PCR. The fragments contained complementary sequences in the 3′ end of the upstream fragment and the 5′ end of the downstream fragment. With these complementary sequences, the fragments were joined during a second round of PCR. The 5′ end of the resulting fragment contained a NotI site, and the 3′ end contained a SalI site. The PCR fragment was ligated into the pGEM-T Easy vector, double digestion with NotI and SalI was performed, and the purified fragment was ligated into the NotI/SalI-digested suicide plasmid pDMK, which is a derivative of pDM4 (20). The former contains a Tn5-derived kanamycin resistance gene and confers resistance against only this antibiotic.
Escherichia coli S17-1 carrying pDMK-ΔkatG grown to log phase was concentrated by centrifugation and mixed with F. tularensis SCHU or LVS that had also been grown to log phase and concentrated by centrifugation. After incubation on Luria agar plates overnight, the bacteria were resuspended in 300 μl of phosphate-buffered saline (PBS) with 10 mM MgSO4 and plated on McLeod plates supplemented with 50 μg/ml polymyxin B and kanamycin at a concentration of 5 μg/ml for LVS and 10 μg/ml for SCHU. The colonies that appeared on the plates were checked by PCR to verify integration of the vector. To select for a second recombination event, the recombinants were plated on McLeod plates supplemented with 5% sucrose. Colonies that were resistant to sucrose but sensitive to kanamycin were checked by PCR to verify the deletion of katG.
Mice.
For experiments conducted at Umeå University, breeding stocks of C57BL/6 mice were obtained from The Jackson Laboratory, Bar Harbor, ME. The mice were bred at the Animal Facility, Swedish Defense Research Agency, Umeå, Sweden, under conventional conditions and given food and water ad libitum. Mice were 8 to 14 weeks old, age and sex matched, and found to be free from specific pathogens. The experiments were approved by the Animal Ethical Committee, Umeå University. Experiments conducted at the National Research Council Canada used C57BL/6 mice purchased from Charles Rivers Laboratories, Quebec. These experiments were approved by the institutional animal care committee.
Exposure of F. tularensis to reactive molecular species in a cell-free system.
3-Morpholinosydnonimine hydrochloride (SIN-1; Molecular Probes, Eugene, OR) generates ONOO−. Under physiological conditions, 1 mM SIN-1 generates 10 μM ONOO−/min. F. tularensis was cultivated in Chamberlain medium (4) to the logarithmic growth phase, diluted to a density of approximately 2 × 106 bacteria/ml in PBS with 20 mM HEPES buffer, and incubated at 37°C. H2O2 or SIN-1 was then added. From samples collected at the indicated time points, 10-fold dilutions were plated for determination of bacterial numbers.
Infection of macrophages.
Peritoneal exudate cells (PEC) were obtained from mice 3 days after intraperitoneal injection of 2 ml of 10% proteose peptone. PEC were washed with Dulbecco modified Eagle medium (GIBCO BRL, Grand Island, NY) and resuspended at a density of 106 cells/ml in culture medium consisting of Dulbecco modified Eagle medium with 10% heat-inactivated fetal calf serum. The suspension was divided into 100-μl volumes in a 96-well tissue culture plate. After incubation for 2 h at 37°C, nonadherent cells were removed and after an additional 24 h, F. tularensis LVS was added to give a multiplicity of infection of 50 bacteria/cell. The actual multiplicity of infection was determined by retrospective plating; thus, there were slight variations between experiments. To some wells, 1 mM NG-monomethyl-l-arginine (N-MMLA; Sigma-Aldrich), an inhibitor of iNOS, and/or 100 U/ml IFN-γ (Peprotech, London, United Kingdom) was added 12 h before bacteria were added. The concentration of 1 mM N-MMLA used has been found to essentially block iNOS activity (9, 18). N-MMLA was also added to the wells after subsequent washing steps. After uptake of bacteria was allowed to occur for 2.0 h, the macrophages were washed to remove extracellular bacteria. Macrophages were reconstituted in culture medium supplemented with 2 μg/ml gentamicin and incubated for the indicated periods of time. Cells were subjected to lysis with 0.1% deoxycholate, and the number of intracellular bacteria was determined by plating 10-fold serial dilutions.
Inoculation and enumeration of bacteria in mice.
Intradermal inocula were injected into the shaved mid-belly in a volume of 50 to 100 μl of saline. At various intervals postchallenge, animals were killed and a 1-cm2 sample of the skin encompassing the inoculation site was swabbed with 70% ethanol, excised, and briefly washed in saline to remove residual bactericidal ethanol. Spleens, livers, and lungs were also collected. Each tissue sample was homogenized, and 10-fold serial dilutions were cultured on agar plates.
Statistical analysis.
For statistical evaluation, one-way analysis of variance and multiple comparisons (post-hoc test with Bonferroni correction [SPSS version 13.0]) were used.
RESULTS
Growth of F. tularensis strains in PEC.
Murine PEC were infected with LVS, SCHU S4, or FSC200, and the number of viable intracellular bacteria was determined over time. The results from one out of four representative experiments are shown in Fig. 1. In all four experiments, the number of intracellular SCHU S4 or FSC200 bacteria had already increased significantly at 6 h (P < 0.05) whereas the number of intracellular F. tularensis LVS bacteria only increased in one out of four experiments. Between 6 and 15 h, the three strains showed similar generation times and the numbers of bacteria increased more than 1.0 log10. After 15 h, the PEC showed signs of cytopathogenicity and therefore the number of intracellular bacteria was not recorded after this time point.
FIG. 1.
Growth and survival of SCHU S4, FSC200, and LVS in PEC after different treatments. (A) Nontreated. (B) Stimulated with IFN-γ (10 μg/ml) 15 h before infection and thereafter. (C) Stimulated with IFN-γ (10 μg/ml) and N-MMLA (1 mM) 15 h before infection and thereafter. The values shown represent the average of triplicates ± the standard deviation. The experiment in panel A was repeated four times and the experiments in panels B and C were repeated three times with similar results.
To test if IFN-γ-activated macrophages killed virulent strains of F. tularensis as efficiently as LVS, PEC cultures pretreated with IFN-γ were infected as described above with SCHU S4, FSC200, or LVS. The viability of all three strains decreased over time, but the virulent strains were killed more slowly (Fig. 1B). At 21 h, IFN-γ-activated macrophage cultures still contained 1 to 2 log10 SCHU S4 or FSC200 bacteria whereas the number of LVS bacteria was below the limit of detection (P < 0.001). Unlike resting PEC, IFN-γ-activated macrophages did not display a cytopathic effect following prolonged infection. Next, macrophages activated with IFN-γ and treated with N-MMLA, a competitive inhibitor of iNOS, were infected with each of the three strains to investigate if killing of the virulent strains is dependent on NO produced by iNOS. Previously, killing of F. tularensis LVS was shown to depend on this mechanism (10, 14, 22). Addition of N-MMLA to the IFN-γ-activated macrophage cultures significantly inhibited the killing of all three strains (Fig. 1B versus C) (P < 0.001). The killing of the virulent strains was essentially blocked in the presence of N-MMLA, and at 21 h, the number of viable bacteria had increased >1 log10. In contrast, the viability of F. tularensis LVS had decreased 1.5 log10 during this time.
Resistance of F. tularensis strains to oxygen and nitrogen radicals.
Oxygen and nitrogen species such as H2O2 and ONOO− are bactericidal molecules that can be generated in IFN-γ-activated macrophages. The lower susceptibility of virulent strains of F. tularensis to killing by IFN-γ-activated macrophages could be due to an increased resistance to such reactive species. This possibility was explored in a cell-free assay by exposing the three test strains to H2O2 or SIN-1, which generates ONOO−. Initially, the strains were exposed to 1 mM H2O2 or 0.8 mM SIN-1, a concentration previously shown to be bactericidal for LVS, and the viability of the bacteria was recorded at 2 and 4 h. Whereas this concentration of H2O2 or SIN-1 did not affect the viability of the virulent strains during the 4-h incubation period (Fig. 2A and B), the viability of F. tularensis LVS gradually decreased during this time.
FIG. 2.
Survival of SCHU S4, FSC200, and LVS grown to the logarithmic phase in Chamberlain medium and thereafter diluted to approximately 106 CFU/ml in PBS and exposed to H2O2 (1 mM) (A), SIN-1 (0.8 mM) that generates ONOO− (B), or the indicated concentration of H2O2 for 2 h (C) or SIN-1 for 4 h (D). The values shown represent the average of triplicates ± the standard deviation. This experiment was performed three times with similar results.
To further assess the resistance of SCHU S4 and FSC200 to H2O2 or ONOO−, these strains and LVS were exposed to high concentrations of H2O2 (2, 4, and 8 mM) for 2 h or SIN-1 for 4 h (0.8, 3.2, and 6.4 mM). As expected, LVS was eradicated by all concentrations of H2O2 (Fig. 2C). The viability of SCHU S4 and FSC200 gradually decreased with increased concentrations of H2O2, and the viability of FSC200 decreased significantly more than that of SCHU S4 at all of the concentrations tested (P < 0.001) (Fig. 2C). At a concentration of 3.2 mM, SIN-1 killed LVS within 4 h whereas this concentration did not affect the viability of SCHU S4 or FSC200 (Fig. 2D). When they were exposed to 6.4 mM SIN-1, the virulent strains' viability decreased approximately 0.5 log10 after a 4-h incubation period.
The role of KatG in protection against oxygen and nitrogen radicals.
The katG gene is present in the genomes of all subspecies of F. tularensis, and the encoded protein is conserved among the SCHU S4, LVS, and the FSC200 strains, as determined from the sequences deposited in GenBank. It shows high homology to genes of other bacterial species known to encode an enzyme with a catalase-peroxidase function. The role of this enzyme for the in vivo and in vitro survival and replication of F. tularensis has not been evaluated. To this end, SCHU S4 and LVS and the respective katG mutants were exposed to H2O2 or SIN-1-generated ONOO−. After a 2-h exposure to 0.1 mM H2O2, the viability of SCHU S4-ΔkatG decreased approximately 1.0 log10 whereas the viability of wild-type SCHU S4 was not affected (P < 0.001) (Fig. 3A). Under the same conditions, the number of LVS-ΔkatG bacteria decreased 5 log10 and the number of LVS bacteria decreased approximately 2.5 log10 (P < 0.001). Thus, although SCHU S4-ΔkatG was more susceptible than the wild-type SCHU S4 strain to killing by 0.1 mM H2O2, it was still more resistant to such killing than wild-type LVS (P < 0.001) (Fig. 3A). After a 2-h exposure to 0.8 mM SIN-1, the viability of SCHU S4 or SCHU S4-ΔkatG was not affected (Fig. 3B) whereas the viability of LVS decreased 2 log10 and the viability of LVS-ΔkatG decreased 3 log10 (P < 0.001 for both LVS versus LVS-ΔkatG and SCHU S4 or SCHU S4-ΔkatG versus LVS). Exposure to 4.6 mM or 6.4 mM SIN-1 (Fig. 3B and data not shown) slightly affected the viability of both SCHU S4 and SCHU S4-ΔkatG (∼0.5 log10 killing) to a similar extent.
FIG. 3.
Survival of SCHU S4, SCHU S4-katG, LVS, and LVS-katG grown to the logarithmic phase in Chamberlain medium, diluted to approximately 106 CFU/ml in PBS, and thereafter exposed to hydrogen peroxide (A) or SIN-1 that generates ONOO− (B) for 2 h. The values shown represent the average of triplicates ± the standard deviation. This experiment was performed three times with similar results.
Role of KatG in F. tularensis survival in macrophages.
On the basis of the aforementioned findings, it is possible that KatG plays a critical role in the intramacrophage survival of SCHU S4 and F. tularensis LVS. To examine this possibility, we determined the abilities of SCHU S4-ΔkatG, F. tularensis LVS-ΔkatG, and their respective parental strains to grow intracellularly in unstimulated or in IFN-γ- or IFN-γ-N-MMLA-treated PEC. In all cases, both SCHU S4-ΔkatG and F. tularensis LVS-ΔkatG survived as well as the parental strains (Fig. 4A and B). Thus, KatG, despite its importance for the detoxification of H2O2, is dispensable for the intramacrophage survival of both attenuated and virulent strains of F. tularensis.
FIG. 4.
Growth and survival of SCHU S4-ΔkatG and SCHU S4 (A) and LVS-ΔkatG and LVS (B) in PEC after different treatments. IFN-γ (10 μg/ml) or IFN-γ and N-MMLA (1 mM) were supplied to the cultures 15 h before infection and thereafter. The values shown represent the average of triplicates ± the standard deviation. This experiment was performed three times with similar results.
Requirement of KatG for in vivo survival.
The importance of KatG for in vivo survival was tested by infecting mice with SCHU S4-ΔkatG, LVS-ΔkatG, or their parental strains. The bacteria were given by the intradermal route, a natural route of infection. LVS and LVS-ΔkatG were administered at doses of 107 and 108 CFU/mouse. All mice challenged with either dose of F. tularensis LVS-ΔkatG survived but showed signs of disease between days 4 and 5. Mice infected with 107 CFU of F. tularensis LVS survived but exhibited signs of severe illness between days 4 and 5. Mice infected with 108 CFU of LVS all died within 5 days (Table 1). After administration of an intradermal inoculum of 8 × 104 CFU of LVS-ΔkatG or LVS, significantly fewer bacteria of the former strain were detected in the skin, spleen, and liver at days 3 and 5 of infection (Table 2).
TABLE 1.
Survival of F. tularensis-infected C57BL/6 mice
| Intradermal inoculum size (no. of CFU) | Strain | % Lethality (avg time [days] to death ± SD) |
|---|---|---|
| 101 | SCHU S4 | 100 (7.6 ± 0.9) |
| SCHU S4-ΔkatG | 100 (7.0 ± 0.8) | |
| 108 | LVS | 100 (5 ± 0) |
| LVS-ΔkatG | 0.00 (0 ± 0) |
TABLE 2.
Growth of F. tularensis strains in C57BL/6 mice after intradermal inoculation
| Strain | Skin
|
Liver
|
Spleen
|
|||
|---|---|---|---|---|---|---|
| Day 3 | Day 5 | Day 3 | Day 5 | Day 3 | Day 5 | |
| LVSa | 6.57 ± 0.06c | 4.70 ± 0.61 | 6.02 ± 0.08 | 5.68 ± 0.59 | 6.07 ± 1.06 | 5.56 ± 0.59 |
| LVS-ΔkatGa | 5.26 ± 0.46d | 3.35 ± 0.45d | 5.09 ± 0.54d | 4.23 ± 0.19d | 4.86 ± 0.48d | 4.77 ± 0.24d |
| SCHU S4b | 6.43 ± 1.44 | 6.86 ± 0.36 | 4.27 ± 1.25 | 7.99 ± 0.82 | 5.32 ± 1.22 | 8.98 ± 0.88 |
| SCHU S4-ΔkatGb | 6.46 ± 0.96 | 5.72 ± 0.84e | 3.89 ± 0.97 | 6.30 ± 0.78e | 5.12 ± 0.81 | 7.43 ± 0.81e |
Inoculum of 4.9 log10 CFU.
Inoculum of 1.2 log10 CFU.
Average no. of CFU per organ ± standard deviation (n = 5).
P < 0.05 compared to LVS.
P < 0.05 compared to SCHU S4.
Mice infected with 10 CFU of either SCHU S4 or SCHU S4-ΔkatG all died between days 6 and 7 postchallenge (Table 1). However, when the bacterial numbers were enumerated in samples from the skin, spleen, and liver, fewer bacteria of the SCHU S4-ΔkatG strain than the SCHU S4 strain were detected in all three organs on day 5 (P < 0.05) but not day 3 of infection (Table 2).
DISCUSSION
F. tularensis is a facultative intracellular pathogen of phagocytes, hepatocytes, and possibly other cell types (8). In murine models of tularemia, type A and B strains of F. tularensis are extremely virulent, with an intradermal LD50 of less than 10 CFU and a time to death of, at most, 9 days. In contrast, the intradermal LD50 of LVS is more than 100,000-fold higher. However, the virulence of LVS approaches that of clinical strains in mice deficient in various host defense mechanisms. In particular, the numbers of F. tularensis LVS bacteria are exacerbated during the early phase of infection in phagocyte oxidase-deficient or neutropenic mice or mice deficient in IFN-γ (15, 25, 26). This suggests that loss of virulence in LVS is at least partly due to enhanced susceptibility to killing by host phagocytes and, by corollary, that type A and type B strains are assumed to be virulent because of their ability to resist such killing mechanisms. The present study was undertaken to examine this hypothesis. Killing of F. tularensis could occur either directly by the phagocytes that ingest it or by uninfected phagocytes marshaled into infectious foci secreting antimicrobial substances therein. Therefore, we examined both cell-free and intramacrophage killing of F. tularensis. In cell-free assays, SCHU S4 (type A) and FSC200 (type B) were very resistant to killing by H2O2 and ONOO− relative to LVS. This result suggests that virulent strains express factors that protect against RNS and ROS, which are absent from or expressed at reduced levels in F. tularensis LVS. Although F. tularensis LVS is able to inhibit activation of the oxidative burst in neutrophils that ingest it (19), the presence of noninfected neutrophils in the infectious foci may still lead to high concentrations of ROS therein and thus an ability to survive the exposure of these toxic molecules may be important in vivo.
Previously, we and others have demonstrated that F. tularensis LVS is killed by IFN-γ-activated murine peritoneal or bone marrow-derived macrophages and that iNOS-derived NO is a critical effector mechanism for this bactericidal activity (10, 14, 22). The present study examined for the first time the ability of this mechanism to kill virulent strains of F. tularensis. The results showed that IFN-γ-activated PEC were more bactericidal than resting cells for both SCHU S4 and FSC200. However, LVS was more susceptible than either clinical strain to this IFN-γ-dependent killing mechanism. Interestingly in this regard, when NO production from iNOS in IFN-γ-activated macrophages was inhibited by N-MMLA, killing of the virulent strains, but not LVS, was essentially inhibited. Thus, killing of the virulent strains was primarily dependent on the activity of iNOS. The mechanism behind both the iNOS-dependent and -independent killing of F. tularensis LVS has previously been demonstrated to rely on ONOO− (14), to which clinical isolates were much more resistant in cell-free assays. This could also help explain why LVS was killed more rapidly by activated macrophages than were the clinical strains. Similarly, another clinical strain of F. tularensis subsp. holarctica was found to be more resistant to ROS (16) and to resist killing by human polymorphonuclear leukocytes better than LVS (17). This result and the results presented in the present study suggest that the higher virulence of clinical strains compared to LVS is at least partly related to increased survival in professional phagocytes because of increased resistance to ROS and RNS.
Like other intracellular bacterial pathogens, F. tularensis is armed with a variety of enzymes that can combat host ROS- and RNS-mediated killing mechanisms. Little is know about their function, but recently a FeSOD mutant of F. tularensis LVS was found to have decreased virulence in mice (1). In the present study, we examined the potential contribution of the enzyme encoded by katG to F. tularensis resistance to ROS- and RNS-mediated killing. Although KatG often is considered to be a catalase, there is abundant evidence that katG-encoded enzymes also have a peroxidase function that could potentially afford protection against ONOO− (28). In this regard, the enhanced resistance of virulent strains not only to ONOO− but also to H2O2 observed in the present study may be partly explained by the recent demonstration that a virulent isolate of F. tularensis secretes much more KatG than LVS into culture medium (13). However, since the SCHU S4-ΔkatG mutant was more resistant than F. tularensis LVS to H2O2 this implies that virulent strains possess mechanisms other than KatG that protect against ROS and that these are not as potent in or are absent from LVS. The F. tularensis genome encodes homologs of several other enzymes known to detoxify ROS and RNS, and it is possible that these, acting alone or in concert with KatG, play a more critical role in this detoxification. Examples of such enzymes include alkyl peroxides, SOD, methionine sulfoxide reductase, and glutathione peroxidase. Future studies will examine the role of these enzymes in the host defense against LVS and virulent strains of the pathogen.
Even though LVS and SCHU S4 in the absence of KatG became more susceptible to killing by ROS in a cell-free system, this did not translate into enhanced killing by resting or IFN-γ-activated PEC but did decrease their virulence for mice. This was evident as early as day 3 of infection, by which time fewer mutant than parental LVS were recovered from the skin, liver, and spleen despite the fact that a sublethal inoculum of either strain was administered. Normally, a sublethal intradermal inoculum of LVS leads to vigorous multiplication in the skin for a day or so before dissemination to the liver and spleen occurs (15). Thereafter, LVS multiplies in all target organs for a few days before host defense mechanisms begin to control and eliminate it. The decreased burden of the katG mutant compared to LVS in the skin indicates enhanced killing therein. The decreased burden in the liver and spleen could be due to decreased dissemination from the skin and/or enhanced killing in the former organs. A more subtle role for KatG was revealed for the fully virulent type A strain SCHU S4. Thus, although a 10-CFU intradermal inoculum of SCHU S4 or SCHU S4-ΔkatG killed mice at the same rate, the latter grew significantly more slowly in the skin, liver, and spleen between days 3 and 5 of infection but not beforehand. However, this was insufficient to prevent the mutant bacteria from subsequently growing to lethal numbers, presumably because an irreversible burden was reached before immune mechanisms capable of controlling the infection were expressed. The fact that KatG was more important for the in vivo survival of LVS than that of SCHU S4 might be related to the type B genetic background of the former per se or the various mutations present in LVS.
In conclusion, clinical strains, in contrast to LVS, express mechanisms that confer high resistance to ROS and RNS. These factors likely contributed to the enhanced survival of the clinical strains in macrophages. KatG was demonstrated to be a more critical virulence factor of F. tularensis LVS than SCHU S4 in mice but was not required to resist the bactericidal effects of macrophages. This suggests that F. tularensis is exposed to ROS and RNS not only in macrophages but also in other cell types or extracellularly in vivo. Moreover, the ability to detoxify ROS may contribute to the survival of F. tularensis in other, more subtle ways, for example, by interfering with host signaling mechanisms in vivo. The factors expressed by the virulent F. tularensis strains that give them high resistance to ROS and RNS will benefit these strains at several stages of the infection.
Acknowledgments
This study was enabled by grant support obtained from the Swedish Medical Research Council, Samverkansnämnden, Norra Sjukvårdsregionen, Umeå, and the Medical Faculty, Umeå University, Umeå, Sweden, and in part by grant AI60689 from the National Institutes of Health, Rockville, MD.
Editor: D. L. Burns
Footnotes
Published ahead of print on 8 January 2007.
REFERENCES
- 1.Bakshi, C. S., M. Malik, K. Regan, J. A. Melendez, D. W. Metzger, V. M. Pavlov, and T. J. Sellati. 2006. Superoxide dismutase B gene (sodB)-deficient mutants of Francisella tularensis demonstrate hypersensitivity to oxidative stress and attenuated virulence. J. Bacteriol. 188:6443-6448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bryk, R., P. Griffin, and C. Nathan. 2000. Peroxynitrite reductase activity of bacterial peroxiredoxins. Nature 407:211-215. [DOI] [PubMed] [Google Scholar]
- 3.Carmel-Harel, O., and G. Storz. 2000. Roles of the glutathione- and thioredoxin-dependent reduction systems in the Escherichia coli and Saccharomyces cerevisiae responses to oxidative stress. Annu. Rev. Microbiol. 54:439-461. [DOI] [PubMed] [Google Scholar]
- 4.Chamberlain, R. E. 1965. Evaluation of live tularemia vaccine prepared in a chemically defined medium. Appl. Microbiol. 13:232-235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chen, W., R. KuoLee, H. Shen, and J. W. Conlan. 2004. Susceptibility of immunodeficient mice to aerosol and systemic infection with virulent strains of Francisella tularensis. Microb. Pathog. 36:311-318. [DOI] [PubMed] [Google Scholar]
- 6.Conlan, J. W. 2004. Vaccines against Francisella tularensis—past, present and future. Expert Rev. Vaccines 3:307-314. [DOI] [PubMed] [Google Scholar]
- 7.Conlan, J. W., W. Chen, H. Shen, A. Webb, and R. KuoLee. 2003. Experimental tularemia in mice challenged by aerosol or intradermally with virulent strains of Francisella tularensis: bacteriologic and histopathologic studies. Microb. Pathog. 34:239-248. [DOI] [PubMed] [Google Scholar]
- 8.Conlan, J. W., and R. J. North. 1992. Early pathogenesis of infection in the liver with the facultative intracellular bacteria Listeria monocytogenes, Francisella tularensis, and Salmonella typhimurium involves lysis of infected hepatocytes by leukocytes. Infect. Immun. 60:5164-5171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.DeGeorge, G. L., D. E. Heck, and J. D. Laskin. 1997. Arginine metabolism in keratinocytes and macrophages during nitric oxide biosynthesis: multiple modes of action of nitric oxide synthase inhibitors. Biochem. Pharmacol. 54:103-112. [DOI] [PubMed] [Google Scholar]
- 10.Fortier, A. H., S. J. Green, T. Polsinelli, T. R. Jones, R. M. Crawford, D. A. Leiby, K. L. Elkins, M. S. Meltzer, and C. A. Nacy. 1994. Life and death of an intracellular pathogen: Francisella tularensis and the macrophage. Immunol. Ser. 60:349-361. [PubMed] [Google Scholar]
- 11.Fortier, A. H., T. Polsinelli, S. J. Green, and C. A. Nacy. 1992. Activation of macrophages for destruction of Francisella tularensis: identification of cytokines, effector cells, and effector molecules. Infect. Immun. 60:817-825. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Golovliov, I., A. Sjostedt, A. Mokrievich, and V. Pavlov. 2003. A method for allelic replacement in Francisella tularensis. FEMS Microbiol. Lett. 222:273-280. [DOI] [PubMed] [Google Scholar]
- 13.Lee, B. Y., M. A. Horwitz, and D. L. Clemens. 2006. Identification, recombinant expression, immunolocalization in macrophages, and T-cell responsiveness of the major extracellular proteins of Francisella tularensis. Infect. Immun. 74:4002-4013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lindgren, H., L. Stenman, A. Tarnvik, and A. Sjöstedt. 2005. The contribution of reactive nitrogen and oxygen species to the killing of Francisella tularensis LVS by murine macrophages. Microbes Infect. 7:467-475. [DOI] [PubMed] [Google Scholar]
- 15.Lindgren, H., S. Stenmark, W. Chen, A. Tarnvik, and A. Sjöstedt. 2004. Distinct roles of reactive nitrogen and oxygen species to control infection with the facultative intracellular bacterium Francisella tularensis. Infect. Immun. 72:7172-7182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Löfgren, S., A. Tärnvik, M. Thore, and J. Carlsson. 1984. A wild and an attenuated strain of Francisella tularensis differ in susceptibility to hypochlorous acid: a possible explanation of their different handling by polymorphonuclear leukocytes. Infect. Immun. 43:730-734. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Löfgren, S., A. Tarnvik, G. D. Bloom, and W. Sjöberg. 1983. Phagocytosis and killing of Francisella tularensis by human polymorphonuclear leukocytes. Infect. Immun. 39:715-720. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Maurer, T. S., and H. L. Fung. 2000. Evaluation of nitric oxide synthase activity and inhibition kinetics by chemiluminescence. Nitric Oxide 4:372-378. [DOI] [PubMed] [Google Scholar]
- 19.McCaffrey, R. L., and L. A. Allen. 2006. Francisella tularensis LVS evades killing by human neutrophils via inhibition of the respiratory burst and phagosome escape. J. Leukoc. Biol. 80:1224-1230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Milton, D. L., R. O'Toole, P. Horstedt, and H. Wolf-Watz. 1996. Flagellin A is essential for the virulence of Vibrio anguillarum. J. Bacteriol. 178:1310-1319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Oyston, P. C., A. Sjöstedt, and R. W. Titball. 2004. Tularaemia: bioterrorism defence renews interest in Francisella tularensis. Nat. Rev. Microbiol. 2:967-978. [DOI] [PubMed] [Google Scholar]
- 22.Polsinelli, T., M. S. Meltzer, and A. H. Fortier. 1994. Nitric oxide-independent killing of Francisella tularensis by IFN-γ-stimulated murine alveolar macrophages. J. Immunol. 153:1238-1245. [PubMed] [Google Scholar]
- 23.Radi, R., G. Peluffo, M. N. Alvarez, M. Naviliat, and A. Cayota. 2001. Unraveling peroxynitrite formation in biological systems. Free Radic. Biol. Med. 30:463-488. [DOI] [PubMed] [Google Scholar]
- 24.Sjöstedt, A. 2006. Intracellular survival mechanisms of Francisella tularensis, a stealth pathogen. Microbes. Infect. 8:561-567. [DOI] [PubMed] [Google Scholar]
- 25.Sjöstedt, A., J. W. Conlan, and R. J. North. 1994. Neutrophils are critical for host defense against primary infection with the facultative intracellular bacterium Francisella tularensis in mice and participate in defense against reinfection. Infect. Immun. 62:2779-2783. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Sjöstedt, A., R. J. North, and J. W. Conlan. 1996. The requirement of tumour necrosis factor-alpha and interferon-gamma for the expression of protective immunity to secondary murine tularaemia depends on the size of the challenge inoculum. Microbiology 142:1369-1374. [DOI] [PubMed] [Google Scholar]
- 27.Sjöstedt, A. 2005. Francisella, p. 200-210. In D. J. Brenner, N. R. Krieg, J. T. Staley, and G. M. Garrity (ed.), The Proteobacteria, part B, Bergey's manual of systematic bacteriology, 2nd ed., vol. 2. Springer, New York, N.Y. [Google Scholar]
- 28.Smulevich, G., C. Jakopitsch, E. Droghetti, and C. Obinger. 2006. Probing the structure and bifunctionality of catalase-peroxidase (KatG). J. Inorg. Biochem. 100:568-585. [DOI] [PubMed] [Google Scholar]
- 29.Wang, G., R. C. Conover, A. A. Olczak, P. Alamuri, M. K. Johnson, and R. J. Maier. 2005. Oxidative stress defense mechanisms to counter iron-promoted DNA damage in Helicobacter pylori. Free Radic. Res. 39:1183-1191. [DOI] [PubMed] [Google Scholar]