Coxiella burnetii Interaction with Neutrophils and Macrophages In Vitro and in SCID Mice following Aerosol Infection

Abstract

Coxiella burnetii is an obligate intracellular bacterium that causes acute and chronic Q fever in humans. Human Q fever is mainly transmitted by aerosol infection. However, there is a fundamental gap in the knowledge regarding the mechanisms of pulmonary immunity against C. burnetii infection. This study focused on understanding the interaction between C. burnetii and innate immune cells in vitro and in vivo. Both virulent C. burnetii Nine Mile phase I (NMI) and avirulent Nine Mile phase II (NMII) were able to infect neutrophils, while the infection rates were lower than 29%, suggesting that C. burnetii can infect neutrophils, but infection is limited. Interestingly, C. burnetii inside neutrophils can infect and replicate within macrophages, suggesting that neutrophils cannot kill C. burnetii and C. burnetii may be using infection of neutrophils as an evasive strategy to infect macrophages. To elucidate the mechanisms of the innate immune response to C. burnetii natural infection, SCID mice were exposed to aerosolized C. burnetii. Surprisingly, neutrophil influx into the lungs was delayed until day 7 postinfection in both NMI- and NMII-infected mice. This result suggests that neutrophils may play a unique role in the early immune response against aerosolized C. burnetii. Studying the interaction between C. burnetii and the innate immune system can provide a model system for understanding how the bacteria evade early immune responses to cause infection.

INTRODUCTION

Coxiella burnetii is an obligate intracellular Gram-negative bacterium and the causative agent of the zoonotic disease Q fever. Most commonly, the disease manifests in acute form, causing flu-like symptoms, but occasionally chronic infection occurs, leading to life-threatening endocarditis (1). The organism is mainly spread through aerosol exposure of infected materials, with the infectious dose in humans being as low as 1 to 10 organisms (2). C. burnetii is unique due to its high resistance to environmental stresses and its ability to replicate within a highly acidic parasitophorous vacuole (PV) which is indistinguishable from a secondary lysosome (3).

C. burnetii Nine Mile undergoes a phase variation, when it is passaged within cultured cells, which converts it from virulent Nine Mile phase I (NMI) with a full-length smooth lipopolysaccharide (LPS) to nonvirulent Nine Mile phase II (NMII) with a truncated rough LPS (4). NMI can replicate within tissues and cause disease in immunocompetent mice and is considered a select agent, so it must be handled in a biosafety level 3 (BSL3) facility. NMII does not replicate or cause disease within immunocompetent mice and can be handled in a biosafety level 2 (BSL2) facility. In contrast to their behavior in vivo, NMI and NMII act relatively similarly following uptake by human macrophages in vitro. PVs of both phase variants are moderately acidic (pH, ∼5), and endosomal/lysosomal markers, such as lysosome-associated membrane protein 1 (LAMP-1) and CD63 (LAMP-3), are present on the membrane (5). NMI and NMII were also found to have similar growth rates in human macrophages through day 11 and cause the production of similar amounts of the proinflammatory cytokines tumor necrosis factor alpha (TNF-α) and interleukin-6 (IL-6) (6). Thus, identifying the differences between NMI and NMII in their interaction with host immune cells may provide critical information for understanding the pathogenesis of C. burnetii.

When foreign material, such as bacteria, enters the lower airway through a contaminated aerosol, the first cells that they encounter are primarily alveolar epithelial cells and macrophages. Macrophages engulf the bacteria and release inflammatory cytokines, such as IL-1 β, IL-8, and granulocyte colony-stimulating factor (G-CSF), which cause neutrophils to begin infiltrating the lungs within 24 h postinfection (p.i.) (7). Neutrophils are the first responders to infection, with their main job being to kill the bacteria and release inflammatory cytokines, which promote cell migration toward the site of infection and increased cell proliferation. Previous studies have shown the importance of alveolar neutrophils through the delayed clearance of Streptococcus pneumoniae, Legionella pneumophila, and Klebsiella pneumoniae when neutrophils were selectively depleted (8–10). However, it is unknown how neutrophils respond to natural C. burnetii infection and what role neutrophils play in host defense against C. burnetii infection.

Within neutrophils are granules which contain numerous strong bactericidal enzymes and proteins, such as defensins, NADPH oxidase, and myeloperoxidase (MPO). NADPH oxidase and MPO are involved in the production of highly toxic reactive oxygen species (ROS). Reactive oxygen intermediates, such as H₂O₂ and O₂⁻, are also bactericidal and kill by damaging DNA, oxidizing fatty acids and proteins, and inactivating enzymes (11). In order to survive within a highly bactericidal cell, C. burnetii must have a unique strategy to avoid being degraded. One study suggested that C. burnetii prevents the translocation of proteins needed for production of ROS to the PV (12). A following study found that this is likely due to the release of an acid phosphatase by C. burnetii (13). However, it remains unclear whether C. burnetii can replicate within neutrophils and whether neutrophils can kill C. burnetii.

In addition, neutrophils are extremely important early in infection, and they can be detrimental if they are not taken up by other cells due to the release of cytotoxic enzymes and ROS into the host tissue. Once they are infected or reach a certain age (∼2 days in the tissue), neutrophils undergo a process called apoptosis. Cells undergoing apoptosis have unique characteristics, including DNA fragmentation and blebbing of the membrane. On the outer membrane, they also have specific surface proteins and phospholipids, such as phosphatidylserine (PS), which signal to phagocytic cells that they need to be taken up (14). The main cells in charge of phagocytizing apoptotic cells are macrophages. The uptake of apoptotic neutrophils has been shown to increase the killing of certain intracellular bacteria within macrophages (15). Previous studies have shown that C. burnetii NMII is able to inhibit apoptosis in differentiated THP-1 cells and monkey primary alveolar macrophages (16) or induce apoptosis during an early stage of infection through a caspase-independent pathway in THP-1 cells (17). However, it is unknown whether C. burnetii can inhibit or induce apoptosis in neutrophils.

Due to ease of handling and a lack of aerosol instrumentation, the majority of experimental animal challenges with C. burnetii are performed intraperitoneally. However, the natural route of infection is aerosol, and study of the innate immune responses within the lungs is lacking. This is the first study to use an aerosol model of C. burnetii infection to study the innate immune reactions of neutrophils and macrophages within the lungs.

MATERIALS AND METHODS

Animals.

Specific-pathogen-free 8-week-old female BALB/c and SCID mice were obtained from The Jackson Laboratory (Bar Harbor, ME). All mice were housed in sterile microisolator cages under specific-pathogen-free conditions at the University of Missouri laboratory animal facility. Following aerosol infection, mice were housed in an animal biosafety level 3 (ABSL3) facility at the University of Missouri Laboratory of Infectious Disease Research (MU-LIDR). All research protocols described in this study were approved by the Institutional Biosafety Committee and the Animal Care and Use Committee of the University of Missouri.

C. burnetii isolates.

The C. burnetii NMI clone 7 (RSA439) and NMII clone 4 (RSA393) strains were used for both in vitro and in vivo experiments. The bacteria was propagated in L929 cells and isolated by density centrifugation as previously described (18). NMI was handled under BSL3 conditions in the MU-LIDR, and NMII was handled under BSL2 conditions.

Isolation of neutrophils and macrophages.

Neutrophils and macrophages were collected from the bone marrow of mice as described previously (19). Mice were euthanized, the femurs and tibias were removed, and the ends of the femurs and tibias were cut off, exposing the marrow. A 25-guage needle was placed into the end of the bone and 3 ml of Hanks balanced salt solution (HBSS; HyClone Labs, Logan, UT) was flushed through the bone. Red blood cells were lysed by incubating the marrow with ACK buffer (Lonzana, Walkersville, MD) for 5 min. Mononuclear cells were separated from the neutrophils using discontinuous density centrifugation with 55%, 65%, and 75% Percoll solution (GE Healthcare, Pittsburg, PA). Three milliliters of a 75% solution of Percoll in phosphate-buffered saline (PBS) was layered under 3 ml of a 65% solution, and then 3 ml of a 55% solution was added on top of the 65% solution. The marrow cells, suspended in 3 ml HBSS, were layered on top of the 55% solution, and the tube was centrifuged at 500 × g for 30 min. Neutrophils isolated from between the 65 and 75% layers were counted and processed for further experiments, described later in this report. Mononuclear cells from between the 55 and 65% layers were counted, and the cells were placed into a T75 culture flask containing RPMI, 10% fetal bovine serum (FBS), 20% L929, 100 U/ml penicillin, and 100 μg/ml streptomycin at a concentration of 2 × 10⁶ cells/ml and incubated at 37°C. After 3 days, half of the medium was removed from the macrophages and replaced. Macrophage phagocytosis assays were performed 7 days after isolation and are described later in this report. Trypan blue was used to assess cell viability, which was greater than 98% for each assay.

Neutrophil infection.

A total of 1 × 10⁵ freshly isolated neutrophils were placed per well in a 24-well plate containing RPMI and 10% FBS at 37°C. Either NMI or NMII was added to the neutrophils at a multiplicity of infection (MOI) of 100 for different time points. For immunofluorescence assay (IFA) experiments, neutrophils were placed in a 24-well plate containing collagen-coated coverslips and the neutrophils were allowed to attach to the coverslips during a 15-min incubation at 37°C, followed by infection. The C. burnetii survival assay was performed by cycling infected neutrophils through 3 freeze-thaw cycles, followed by centrifuging at 500 × g for 15 min. Supernatant was added to L929 cells, and the cells were incubated for 1 or 7 days.

Macrophage phagocytosis of neutrophils.

Six days after macrophage collection from bone marrow, ice-cold PBS was added to the macrophages to detach them and they were counted. A total of 1 × 10⁵ cells were placed per well in a 24-well plate on glass coverslips and incubated in RPMI, 10% FBS, and 20% L929 medium for 24 h at 37°C. A total of 1 × 10⁵ neutrophils per well were incubated with NMI or NMII (MOI, 100) in RPMI plus 10% FBS at 37°C for 18 h. The neutrophils were then washed with HBSS to remove extracellular bacteria, resuspended in RPMI plus 10% FBS, and added to the macrophages prepared as described above. Additional wells contained macrophages incubated with medium alone, 1 × 10⁵ polymorphonuclear leukocytes (PMNs) alone, or NMI or NMII (MOI, 100). Wells which were incubated for longer than 24 h were washed at 24 h p.i. and received fresh medium.

Transmission electron microscopy.

A total of 1 × 10⁵ neutrophils were incubated for 18 h at 37°C with NMII (MOI, 100) and then fixed with 2% glutaraldehyde and 2% paraformaldehyde in 100 mM sodium cacodylate buffer at 4°C for 24 h. Samples were then submitted to the University of Missouri Electron Microscope Core, and images were viewed on a JEOL 1400 transmission electron microscope.

IFA.

Cells were fixed with 2% paraformaldehyde for 15 min and permeabilized with cold methanol for 10 min. The cells were blocked in 5% normal goat serum for 1 h. Rabbit anti-NMII/NMI polyclonal antibodies (1:500) were incubated with the cells for 1 h, followed by incubation with goat anti-mouse IgG (1:500; Invitrogen) for 1 h to stain intracellular C. burnetii. Lysosome-associated membrane protein 1 (LAMP-1) antibodies conjugated with phycoerythrin (eBioscience) were added simultaneously with secondary antibodies in selected samples. Host nuclei were stained by DAPI (4′,6-diamidino-2-phenylindole; 1:500; Invitrogen), and slides were examined using a Zeiss LSM 510 META NLO confocal fluorescence microscope. A total of 200 neutrophils and 200 macrophages were counted per slide.

Neutrophil apoptosis.

To determine if C. burnetii can inhibit or induce apoptosis in neutrophils, either NMI or NMII (MOI, 100) was incubated with neutrophils for 4 h or 18 h in RPMI plus 10% FBS at 37°C. Following incubation, the cells were stained with annexin V and propidium iodide (PI) using an annexin-V-fluorescein isothiocyanate staining kit (Roche). Cells were washed 3 times with PBS plus 0.5% bovine serum albumin, fixed with 2% paraformaldehyde, and then washed twice more. Fluorescent cells were detected using a fluorescence-activated cell sorter (FACSCalibur; Becton-Dickson, Palo Alto, CA). To visualize infected cells and apoptotic cells at the same time, the IFA procedure described previously (17) was used to stain C. burnetii, and a terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling (TUNEL) assay kit (Roche) was used to stain the apoptotic cells. Stained neutrophils were examined using a Zeiss LSM 510 META NLO confocal fluorescence microscope. For a positive-apoptosis control, neutrophils were incubated with staturosporin (1 μM) for 4 h.

rtPCR.

Tissue and cells were lysed with 200 μl lysis buffer (1 M Tris, 0.5 M EDTA, 7 mg/ml glucose, 28 mg/ml lysozyme) and 10 μl proteinase K (20 mg/ml) and incubated for 4 h at 60°C, followed by the addition of 21 μl 10% SDS and incubation at room temperature for 1 h. DNA was extracted using a High Pure PCR template preparation kit (Roche Molecular Biomedicals, Indianapolis, IN) and stored at −80°C until use. Real-time PCR (rtPCR) was performed using an Applied Biosystems 7300/7500 real-time PCR system. The recombinant plasmid DNA (the com1 gene ligated into the PET23a vector) was used as standard DNA to quantify com1 gene copy numbers.

Aerosol infection.

Mice were exposed to NMII or NMI in PBS or PBS alone using a liquid sparging aerosolizer (20). This apparatus ensures a uniform dose of bacteria to the lower airways through a nose-only aerosol exposure. A total of 1 × 10⁹ bacteria resuspended in 5 ml of PBS were aerosolized and used to challenge 12 mice, and PBS was aerosolized for 8 mice. The total exposure time was approximately 30 min. Samples collected from bubbling of the aerosol through PBS showed that of the starting concentration of 1 × 10⁹ bacteria, each mouse actually received approximately 1 × 10⁷ C. burnetii bacteria.

BALF collection.

Bronchoalveolar fluid (BALF) was collected as described previously, with modifications (21). Following euthanasia of the mice, the trachea was exposed and a 22-gauge catheter was inserted and secured with suture thread. To obtain BALF, 1 ml of HBSS was slowly infused into the lungs and then withdrawn. This procedure was repeated 3 times. The fluid was centrifuged at 300 × g for 5 min, and the supernatant was collected and frozen at −80°C. Cells were spun onto a glass slide using a cytocentrifuge and stained with Diff-Quick to determine the cell population.

Lung cell collection.

Lung cells were collected as previously described, with some modifications (22). After severing the descending aorta, 3 ml of 0.33 mM EDTA in HBSS was perfused through the right side of the heart to remove the intravascular blood pool. Lungs were removed, minced into small pieces, and incubated with 200 μg/ml collagenase D and 40 μg/ml DNase I in RPMI 1640 medium at 37°C for 1 h. Digested lung tissue was passed through a 70-μm-mesh-size nylon screen to obtain a single-cell suspension, and cells were counted on a hemocytometer. Cells to be used for IFA were incubated on collagen-coated coverslips in RPMI plus 10% FBS for 30 min at 37°C.

Histopathology.

Lungs and spleens were removed from mice at different time points and fixed in 10% formalin for at least 48 h. Tissues were sectioned, embedded in paraffin, and cut to a thickness of 5 μm. They were allowed to adhere to glass slides and stained with hematoxylin-eosin.

Statistical analysis.

Results, expressed as means ± standard deviations, were compared by Student's t test. Differences were considered significant at a P value of ≤0.05.

RESULTS

C. burnetii infection of neutrophils.

To determine if C. burnetii could infect neutrophils, cells were incubated with NMII for 18 h, fixed with 2% paraformaldehyde plus 2% glutaraldehyde, and prepared for transmission electron microscopy. Figure 1A shows NMII organisms within a neutrophil (arrow), confirming infection. Next, early and late time points of infection were examined by incubating neutrophils with either NMII or NMI for 1 and 18 h, followed by staining for IFA. Neutrophil incubation with C. burnetii for 1 and 18 h resulted in a significant difference in infection efficiency between the two strain variants. Incubating cells with NMII for 1 h resulted in 14% infected neutrophils, but incubating cells with NMI resulted in only 9% infected neutrophils (Fig. 1B). The percentage of infected cells increased after 18 h in both NMII- and NMI-infected cells (29% and 17%, respectively). Figure 1C shows a greater number of neutrophils containing NMII organisms than NMI organisms using a confocal fluorescence microscope. This suggests that both C. burnetii phase variants are capable of infecting neutrophils, with NMII being more infective than NMI. However, there is a limit to the infection efficiency, as the highest infection rate after 18 h was only 29%. The colocalization of LAMP-1 and C. burnetii seen in Fig. 1C suggests that the organism is inside the cell rather than attached to the outer membrane. In order to investigate internalization further, IFA was performed without the permeabilization step. The absence of C. burnetii staining in these samples confirms that the pathogen is inside neutrophils (data not shown). In order to determine if C. burnetii was alive inside neutrophils, cells were incubated for 18 h with NMII, washed with HBSS, and lysed through 3 freeze-thaw cycles. The medium was centrifuged to remove cellular debris, and the supernatant was added to L929 cells. At 1 day p.i., 2.5% of the L929 cells were infected (Fig. 2A). At 7 days p.i., 36% of the L929 cells were infected, confirming that C. burnetii was alive inside the neutrophils and able to replicate within cells which have a longer life span.

Fig 1.

C. burnetii NMI and NMII are both able to infect neutrophils. (A) Neutrophils were incubated with NMII for 18 h and then fixed and prepared for transmission electron microscopy as described in Materials and Methods. Two C. burnetii bacteria were identified in the center of the cell (arrow). (B) Neutrophils were incubated with C. burnetii NMI or NMII for 1 h or 18 h. IFA was used to identify bacteria, and 200 cells per sample were counted to detect the number of infected cells (*, P ≤ 0.05). (C) Neutrophils were incubated with C. burnetii NMI or NMII for 18 h and then fixed and stained. C. burnetii is stained green, LAMP-1 is stained red, and nuclei are stained blue.

Fig 2.

C. burnetii is viable within neutrophils but does not replicate. (A) Neutrophils were cultured with NMI or NMII for 18 h, followed by lysis through 2 freeze-thaw cycles. After centrifugation, supernatant was added to L929 cells for 1 and 7 days. IFA was performed to determine the L929 infection rate. (B) Neutrophils were cultured with NMI or NMII for 1 or 18 h. DNA was collected from the samples, and rtPCR was performed to determine the genomic copy number of C. burnetii (*, P ≤ 0.05).

To determine if C. burnetii was able to replicate within neutrophils, cells were incubated with NMI and NMII for 1 and 18 h. When neutrophils were lysed and rtPCR was performed, there was no difference in the amount of C. burnetii detected at 1 h and 18 h, suggesting that the organism does not replicate within neutrophils (Fig. 2B). Contrary to the IFA infection results, rtPCR detected a significantly greater number of NMI organisms than NMII organisms within neutrophils. IFA measures the number of cells containing C. burnetii, regardless of how many C. burnetii organisms are within each cell. rtPCR measures the number of C. burnetii genomic copies per sample, regardless of how many cells contain C. burnetii. The difference in these two measurements could be due to a greater number of NMI bacteria than NMII bacteria entering each neutrophil. Further investigation using more sensitive and advanced microscopy could help determine the cause of this difference.

C. burnetii within neutrophils infects macrophages.

Because the previous experiment showed that neutrophils were able to phagocytize C. burnetii and other studies have found that macrophages are the main cell type to engulf infected neutrophils, our aim for this experiment was to investigate whether C. burnetii within neutrophils could infect macrophages. Neutrophils were cultured with NMII or NMI for 18 h, washed with HBSS to remove extracellular bacteria, and incubated with macrophages. Neutrophils were stained with myeloperoxidase antibodies in order to distinguish them from macrophages, and infected cells were observed using a confocal fluorescence microscope. After 4 h, 40% of macrophages incubated with NMII-cultured neutrophils were infected, while only 11% of macrophages incubated with NMI-cultured neutrophils were infected. This difference was found to be significant (Fig. 3A). Macrophages incubated with NMII-cultured neutrophils still had a significantly higher infection rate at 24 h after incubation than those incubated with NMI (67% versus 42%, respectively), but at 72 h the number of infected cells was similar between NMII- and NMI-cultured cells (86% versus 85%, respectively). The increased amount of macrophages infected with NMII at the early time point could be due to a combination of increased infection rates of neutrophils and the increased cellular infectivity of NMII in vitro. The finding that the infection rate continued to increase over time suggests that C. burnetii is able to replicate within macrophages following uptake of infected neutrophils. Since IFA is an insufficient method to determine replication, rtPCR was used to examine replication of NMI and NMII. The results showed that the genomic copy number of C. burnetii NMII within macrophages incubated with infected neutrophils increased significantly at between 1 and 7 days (Fig. 3B), suggesting that this variant can replicate within macrophages following uptake of infected neutrophils. However, the genomic copy number was similar at between 1 and 7 days for the virulent NMI variant, suggesting that replication was not occurring but the organism was not being killed. This is likely due to the low rate of infection of neutrophils with NMI. The number of NMI-infected neutrophils may have been so low that the macrophages were able to control replication. Figure 3C shows an infected neutrophil (white arrow) attached to the outside of a macrophage (arrowhead) following 1 h of incubation of macrophages with NMII-cultured neutrophils. Following 48 h of incubation, a macrophage containing 2 neutrophils (white arrow) and several C. burnetii bacteria (arrowhead) was observed. In order to determine if macrophages were infected through the release of C. burnetii by neutrophils into the medium or by direct uptake of infected neutrophils, macrophages were incubated with infected neutrophils. Following 4 h of incubation, medium was collected, centrifuged to remove cells, and added to a new set of uninfected macrophages. After 24 h, only 8.6% of the new macrophages were infected, suggesting that in the previous studies, the majority of infected macrophages obtained C. burnetii through direct uptake of infected neutrophils (data not shown). These results suggest that neutrophils may not be able to kill C. burnetii and that C. burnetii may use infection of neutrophils as an evasive strategy to infect macrophages.

Fig 3.

C. burnetii NMI and NMII within neutrophils infect macrophages. (A) Neutrophils were cultured with NMI or NMII for 18 h and washed and incubated with macrophages for 4, 24, and 72 h. Cells were fixed and stained for IFA, and infected cells were counted under a fluorescence microscope. A total of 200 cells were counted per sample (*, P ≤ 0.05). (B) Neutrophils cultured with C. burnetii NMI or NMII were incubated with macrophages for 1 day or 7 days. DNA was collected, and rtPCR was performed to determine the genomic copy number of the bacteria. (C) Neutrophils cultured with NMII were incubated with macrophages for 1 h or 48 h, and cells were fixed and stained for IFA. MPO is stained green, and C. burnetii is stained red. Arrow, infected neutrophil; arrowhead, macrophage.

Neutrophil activation of macrophages increases C. burnetii uptake but not killing.

Within the lungs, macrophages encounter infected neutrophils, which were studied as described in the previous section, as well as extracellular C. burnetii. Neutrophil activation of macrophages has been shown to affect uptake and killing of bacteria by macrophages, and the studies described in this section aimed to determine if this is occurring with C. burnetii. Neutrophils were incubated for 18 h and added to macrophages for 4 h. Macrophages were then washed and incubated with C. burnetii NMII for 24 h. IFA was used to determine the infection rate. At the 24-h time point, macrophages which had been incubated with neutrophils prior to infection had a significantly lower uptake rate than control macrophages (60.4% ± 2% versus 79.3% ± 2%, respectively) (Fig. 4A). rtPCR was performed at 24 h and 7 days p.i. to determine if C. burnetii was able to replicate within macrophages incubated with or without neutrophils. Macrophages incubated with neutrophils and control macrophages contained C. burnetii bacteria which replicated at the same rate (Fig. 4B). These studies suggest that neutrophils may prime macrophages to increase phagocytosis of C. burnetii in early infection. However, this did not increase the ability of macrophages to kill C. burnetii, as replication remained unaffected.

Fig 4.

Apoptotic neutrophils increase macrophage (M0) uptake of C. burnetii but do not affect replication. Neutrophils were cultured for 24 h and then added to macrophages for 4 h. Macrophages cultured with or without neutrophils were washed and infected with C. burnetii NMII. (A) IFA was performed 24 h after addition of C. burnetii (*, P ≤ 0.05). (B) DNA was collected at 24 h and 7 days following addition of C. burnetii to determine the genomic copy number of the organism (*, P ≤ 0.05).

C. burnetii does not affect neutrophil apoptosis.

To measure the ability of C. burnetii to induce or delay apoptosis, neutrophils were incubated with or without NMII and NMI for 4 and 18 h, stained with annexin V, and analyzed by use of a flow cytometer. The result showed a similar number of apoptotic neutrophils between infected and uninfected samples for both time points. This observation occurred for both NMII (Fig. 5A) and NMI (Fig. 5B). The number of apoptotic cells was lower while performing the NMI assay, likely due to day-to-day variation; however, the number of control cells was also lower on the same day. This suggests that C. burnetii does not induce or inhibit apoptosis in neutrophils. In addition, a TUNEL assay was used to determine if apoptotic cells within infected samples contained C. burnetii. The results indicated that less than half of the apoptotic cells contained C. burnetii, suggesting that the bacterium does not play a role in neutrophil apoptosis. As shown in Fig. 4C, a similar number of apoptotic cells (green) were observed between infected and uninfected samples at both time points and only 1 apoptotic cell within the infected culture contained C. burnetii. Staturosporin (1 μM) was used as a control and contained numerous apoptotic neutrophils. These results suggest that C. burnetii does not actively induce or inhibit apoptosis in neutrophils.

Fig 5.

C. burnetii NMI and NMII infection of neutrophils does not induce or inhibit apoptosis. (A) Neutrophils were incubated in RPMI alone or with NMII for 4 h or 18 h, followed by staining with annexin V. The percentage of apoptotic cells was measured on a flow cytometer. (B) PMNs were incubated in RPMI alone or with NMI for 4 h or 18 h, followed by staining with annexin V. The percentage of apoptotic cells was measured on a flow cytometer. (C) Neutrophils were incubated with RPMI, NMII, or staturosporin (Stsp; positive control) for 18 h, followed by fixing and staining using a TUNEL apoptosis kit. The TUNEL assay stains apoptotic cells bright green, NMII was stained red, and nuclei were stained with DAPI. Un, uninfected; In, infected.

C. burnetii within tissues following aerosol challenge.

Our results indicated that C. burnetii could infect neutrophils and macrophages in vitro. To understand how C. burnetii interacts with neutrophils and macrophages in vivo, we established a mouse C. burnetii aerosol infection model. Since C. burnetii NMII cannot cause infection in wild-type BALB/c mice, SCID mice were used to investigate the innate response of neutrophils and macrophages against C. burnetii aerosol infection. Mice were placed in an aerosol chamber specifically designed for nose-only delivery of aerosolized material to the lower airways and challenged with 1 × 10⁷ NMII or NMI bacteria. On day 3 p.i., NMII and NMI were detected by rtPCR in the lungs, but not in the spleen (Fig. 6A). On day 7 p.i., the C. burnetii number in the lungs was increased from that on day 3 and C. burnetii was present in the spleen, indicating a systemic infection at this point. Mice infected with NMI had a stronger systemic response on day 7, as there was an increase in the number of bacteria in the spleen compared to that for NMII-infected mice, which still had the majority of bacteria located in the lungs. Between day 7 and day 14, the NMI and NMII amounts did not increase in the lungs. In contrast, the number of bacteria in the spleen increased significantly from day 7 to day 14 p.i. in both NMI- and NMII-infected mice, indicating that the level of dissemination from the lungs to the spleen was high during this time. Similar to the findings for day 7, the amount of NMI was significantly higher in the spleen than in the lungs on day 14, and for NMII-infected mice, there was a larger amount of bacteria in the lungs than in the spleen on day 14. Over all time points, NMI amounts were higher than NMII amounts in both the lung and spleen. This suggests that NMI is more resistant to in vivo killing and is able to proliferate more efficiently in SCID mice, leading to increased dissemination from the lungs to the spleen compared to that for NMII. Increased splenomegaly at 14 day p.i. further indicated a stronger systemic infection in NMI-infected mice than in NMII-infected mice (Fig. 6B). These results indicate that both NMI and NMII are able to cause systemic infections in SCID mice following aerosol challenge, while NMI has greater dissemination from the lungs to the spleen than NMII. These results suggest that the SCID mouse C. burnetii aerosol infection model may be a useful animal model to understand the innate immune response to natural C. burnetii infection.

Fig 6.

C. burnetii NMI causes a more systemic infection than NMII following aerosol challenge. SCID mice were challenged via aerosol with 1 × 10⁹ C. burnetii NMI or NMII bacteria and sacrificed on days 3, 7, and 14 p.i. (A) DNA was collected from lung and spleen tissue of NMI- or NMII-infected mice, and rtPCR was used to measure the genomic copy number of C. burnetii. (B) Spleen and body weights were collected at each time point. To determine splenomegaly, a ratio of spleen weight to body weight (in grams) was used (*, P ≤ 0.05).

Cell populations within lungs following aerosol challenge.

To determine the cell population within the lungs following aerosol challenge with C. burnetii, bronchoalveolar fluid (BALF) was collected and cells were used for differential counts. Surprisingly, on day 3 p.i., only 10% of the leukocyte population within BALF was neutrophils in NMII-infected mice and 39% was neutrophils in NMI-infected mice (Fig. 7A). An influx of neutrophils, making up 70% of the population of the cells in BALF in NMII-infected mice and 85% in NMI-infected mice, was detected at day 7 p.i. Neutrophils are usually the first cells to respond to an infection and start showing up at the site of infection within 24 h, suggesting that aerosol challenge with C. burnetii delays this innate response. By day 14 p.i., neutrophil populations had decreased to 49% in NMII-infected mice and 37% in NMI-infected mice. The morphology of alveolar macrophages changed from day 3 to day 7, with more inactivated macrophages being found on day 3 p.i. and more activated macrophages with a foamy cytoplasm (arrow) surrounded by neutrophils (arrowhead) being observed on day 7 (Fig. 7B). This change in morphology was observed in both NMI- and NMII-infected mice. Lung tissue samples were collected for histology examination, and results similar to those for the BALF cell population were observed (Fig. 8). In NMI-challenged mice on day 3 p.i., there was a small amount of cell infiltrate, mainly macrophages and neutrophils, within the bronchoalveolar spaces. On day 7 p.i., the airways were full of cell infiltrate, mainly neutrophils and a few macrophages. A small amount of cell infiltrate consisting mainly of macrophages was observed in the airways on day 14 p.i. In NMII-challenged mice, little to no cell infiltrate was observed within the bronchoalveolar spaces on day 3 p.i. On days 7 and 14 p.i., cell infiltrate increased and mainly consisted of neutrophils on day 7 and macrophages on day 14. The amount of infiltrate was less than the amount observed with NMI infection. These results indicate that NMI induces more severe inflammatory cell responses than NMII in SCID mice.

Fig 7.

Neutrophils increase in the BALF on day 7. SCID mice were challenged via aerosol with C. burnetii NMI, C. burnetii NMII, or PBS. BALF samples were collected as described in Materials and Methods on days 3, 7, and 14 p.i. (A) Isolated BALF cells were put onto glass slides using a Cytospin centrifuge, stained with DiffQuick, and counted. A total of 200 cells were counted per slide (*, P ≤ 0.05). (B) The morphology of BALF cells was examined under a bright-field microscope. Unactivated macrophages were present on day 3. On day 7 p.i., activated macrophages with foamy cytoplasm (arrow) and multiple neutrophils (arrowhead) were observed.

Fig 8.

C. burnetii NMI causes more lung inflammation than NMII following aerosol challenge. Lung samples from SCID mice challenged via aerosol with C. burnetii NMI, C. burnetii NMII, or PBS were collected on days 3, 7, and 14 p.i. Samples were fixed, embedded in paraffin, and stained with hematoxylin-eosin.

Lung cell infection following aerosol challenge.

Because rtPCR detected C. burnetii in the lungs and distinct cell populations were observed within the lungs, the next step was to use IFA to detect which lung cells were infected with C. burnetii. Within the alveolar macrophage population, 30 to 40% of the cells were infected at day 3 and day 14 p.i. in both NMI- and NMII-infected mice (Fig. 9A). On day 7, however, the percentage of infected macrophages in mice challenged with NMII increased to 53%, whereas the percentage of infected cells in mice challenged with NMI decreased to 23%. This may be due to the large systemic response observed in NMI-challenged mice, representing migration of infected macrophages from the lungs to the spleen. Within alveolar neutrophils, there was a greater percentage of cells infected with NMI than NMII on day 3 p.i. (11% versus 5%, respectively) (Fig. 9B). On day 7 p.i., the percentage of NMI-infected neutrophils decreased and the percentage of NMII-infected neutrophils increased, again supporting the idea that alveolar cells from NMI-infected mice are trafficking to the spleen by day 7 p.i. In both NMI- and NMII-infected mice, alveolar macrophages had a higher infection rate than alveolar neutrophils over all time points, suggesting that either infected neutrophils are quickly phagocytized by macrophages or C. burnetii uses a unique mechanism to inhibit neutrophil phagocytosis or to be selectively taken up by macrophages.

Fig 9.

Alveolar macrophages were the major cell type harboring C. burnetii NMI and NMII following aerosol challenge. Lung tissue cells were collected from SCID mice challenged via aerosol with C. burnetii NMI, C. burnetii NMII, or PBS, as described in Materials and Methods. Cells were fixed and stained for IFA analysis and counted on a fluorescence microscope. A total of 200 cells of each cell type were counted per slide. (A) The percentage of infected macrophages was counted on days 3, 7, and 14 p.i.; (B) the percentage of infected neutrophils (PMNs) was counted on days 3, 7, and 14 p.i. (*, P ≤ 0.05).

DISCUSSION

Many in vitro studies involving C. burnetii look at the macrophage, dendritic cell, and lymphocyte interaction with the pathogen. Challenge of animals with C. burnetii is mainly performed through intraperitoneal infection and involves sample collection from late time points (14 to 28 days). Because the natural route of infection is aerosol and the early innate immune response, including neutrophil recruitment to the lung and bacterial phagocytosis, is critical for effective clearance of bacteria, this study focused on innate responses to C. burnetii infection both in vitro and in vivo following aerosol challenge.

In general, following entry of bacteria into the airway, a great number of innate immune responses work together to clear the bacteria. Within a few hours of infection, neutrophils begin to exit the blood and enter the lung tissue. However, following aerosol infection with C. burnetii, neutrophils are not present in the BALF until 7 days postinfection, suggesting a delayed response. This response is likely due to a decrease in proinflammatory cytokines or an increase in anti-inflammatory cytokines. BALF samples from infected mice had undetectable levels of the proinflammatory cytokines IL-1β, G-CSF, IL-6, keratinocyte chemoattractant, and TNF-α (data not shown), suggesting that proinflammatory cytokines are not produced at a high concentration following C. burnetii infection. This observation may be explained by the following two possibilities: (i) C. burnetii hides inside resident macrophages and/or epithelial cells without triggering an inflammatory response until the bacteria replicate to high enough numbers to fight the delayed immune response, or (ii) C. burnetii has a lower replication rate with a doubling time of 20 h and is not able to induce a strong inflammatory response in the early stage of infection. Within the lungs, there are 2 types of alveolar macrophages, M1 and M2 macrophages. M1 macrophages produce proinflammatory cytokines, such as IL-1β, TNF, IL-12, and IL-6, promoting Th1 responses. M2 macrophages produce anti-inflammatory cytokines, such as IL-10 and transforming growth factor β1 (TGF-β1), and involve Th2 responses. A 2008 study by Benoit et al. (23) examined macrophage activation following incubation with C. burnetii in vitro and found that the organism induced an M2-type response through increased expression of IL-10 and TGF-β1. Increased expression of these cytokines within the lungs is likely another reason for the low inflammation and neutrophil influx on day 3. A dominant M2 response would also lead to undetectable levels of proinflammatory cytokines within BALF. The increase in neutrophils on day 7 suggests that either anti-inflammatory cytokines decreased or proinflammatory cytokines increased after day 3. The study of Benoit et al. (23) incubated macrophages with C. burnetii for only 4 h before measuring cytokine levels, so long-term cytokine production was not examined. Further investigations looking at the mechanism for this delay in neutrophil influx would be helpful to understand the early innate immune response against C. burnetii natural infection.

Once neutrophils migrate into the lungs and reach the site of infection, they engulf bacteria. Neutrophils contain granules which house numerous strong bactericidal enzymes and proteins, such as defensins, NADPH oxidase, and myeloperoxidase (MPO). Reactive oxygen species (ROS) and intermediates produced inside these granules damage the DNA and oxidize the fatty acids and proteins of ingested bacteria (11). Compared with most bacteria, which are killed within an hour of being ingested by neutrophils, C. burnetii was found inside neutrophils following 1 h and 18 h of incubation. C. burnetii was not able to replicate within neutrophils, likely due to the short half-life of these cells; however, when it was incubated with neutrophils for 18 h, intracellular C. burnetii was able to replicate within L929 cells. This suggests that C. burnetii can evade the harsh neutrophil environment and go on to replicate within longer-living cells. Recent studies suggest that C. burnetii prevents the translocation of proteins needed for the production of ROS to the PV, likely due to the release of an acid phosphatase by C. burnetii (12, 13). When neutrophils were incubated with virulent C. burnetii NMI for 18 h, the infection rate was only 16%, suggesting that C. burnetii can evade phagocytosis by neutrophils. Because avirulent C. burnetii NMII has a truncated LPS and the neutrophil infection rate was 29% after 18 h, LPS could play a partial role in neutrophil phagocytosis evasion. A study by Shannon et al. in 2005 (24) found that NMI did not cause dendritic cell maturation and NMII did cause maturation. They proposed that the full-length LPS acted as a shielding molecule to decrease inflammatory cytokine production and cell maturation. LPS on NMI could have the same effect on neutrophils to decrease phagocytosis. However, the proportion of neutrophils infected with NMII was still only 29% after 18 h, suggesting that there is another mechanism besides LPS causing decreased phagocytosis. The hypothesis of a mechanism to decrease neutrophil phagocytosis is supported in vivo as well. Within the lung, C. burnetii preferentially infected macrophages over neutrophils even on day 7, when neutrophils made up 79% of the lung cell population. This suggests that C. burnetii has a mechanism to either resist neutrophil phagocytosis or increase phagocytosis by macrophages.

Although neutrophils are efficient at clearing bacteria out of the lungs, they can be detrimental if they release their harsh granular contents into the lung tissue, causing damage to the host. One of the main jobs of macrophages at the site of infection is to engulf and destroy infected neutrophils in order to keep the intracellular components from damaging host tissue. Apoptotic neutrophils have on the outer membrane specific surface proteins and phospholipids, such as phosphatidylserine (PS), which signal to macrophages that they need to be taken up (14). Studies have shown that macrophage uptake of apoptotic cells can be either harmful or helpful for the survival of intracellular bacteria. A 2006 study by Tan et al. (15) found that macrophage uptake of apoptotic neutrophils and neutrophil granules helped in killing the intracellular pathogen Mycobacterium tuberculosis. However, a study by Zamboni in 2004 (25) found that apoptotic lymphocytes caused a decrease in NO production by macrophages and subsequently higher numbers of intracellular NMII-infected C. burnetii bacteria. In our study, macrophages incubated with apoptotic neutrophils did have increased uptake of C. burnetii at 24 h p.i. compared to that of macrophages without neutrophils. However, incubation with apoptotic neutrophils did not affect the ability of macrophages to kill C. burnetii or decrease C. burnetii replication. Macrophages were likely activated by neutrophils to increase uptake, but decreased NO production following uptake assisted in C. burnetii replication within the cells. Furthermore, macrophages incubated with neutrophils which had been cultured with either C. burnetii NMI or NMII for 18 h contained viable bacteria up to 7 days following infection. While the NMII variant was able to replicate within macrophages over the 7-day incubation, NMI was not. The reason for this difference is likely due to a combination of neutrophil infection rate variability between the variants and an increased cellular infectivity of NMII in vitro. Although the NMI genomic copy number did not increase at between 1 and 7 days, it also did not decrease, suggesting that C. burnetii was surviving within the macrophage. C. burnetii is likely employing the same survival mechanism used in evading neutrophil killing, such as releasing an acid phosphatase, in order to survive within macrophages which have engulfed an apoptotic infected neutrophil. Another possibility is that NMI escapes the neutrophil in the extracellular fluid and is then taken up by macrophages. However, medium collected 4 h after incubation of infected neutrophils with macrophages caused infection in only 8.6% of new macrophages, suggesting that the majority of infected macrophages in our study obtained C. burnetii through direct uptake of infected neutrophils. This strategy could be a novel survival mechanism which C. burnetii has developed in order to get from a harsh environment, such as the phagosome of a neutrophil, to the preferred environment of a macrophage parasitophorous vacuole. This is likely not an infection mechanism like that used by Leishmania, which uses neutrophils as intermediate hosts; the neutrophils then attract macrophages, which allows Leishmania to enter macrophages silently (26). C. burnetii has a low infection rate in neutrophils and does not cause neutrophils to increase chemotactic proteins like Leishmania does. C. burnetii would likely prefer to avoid neutrophil uptake, but when taken up it can survive long enough to enter a macrophage, its preferred host. Future studies should focus on the detailed mechanism of how Coxiella gets from neutrophils to macrophages. Ideally, if the main goal of C. burnetii is to get from the neutrophil into the macrophage as quickly as possible, the bacteria would likely induce neutrophil apoptosis earlier in order to increase the rate of macrophage uptake. However, no apoptosis increase or decrease was found in this study. A recent paper suggests that C. burnetii replication is needed for induction of apoptosis in THP-1 cells (17). Due to the short half-life of neutrophils, C. burnetii replication likely does not have time to occur in neutrophils. Future studies to examine if depletion of neutrophils significantly affects the ability of C. burnetii to cause disease in mouse aerosol infection models would be helpful to determine the role of neutrophils in host defense against C. burnetii infection.

NMI is a virulent strain of C. burnetii, causing disease in mice and humans, while NMII is a nonvirulent strain, infecting only cultured cells and immune-suppressed mice. Across all time points, NMI-infected mice had more bacteria in both lung and spleen than NMII-infected mice, suggesting a greater ability to replicate within the host and evade the immune system. On day 14 postinfection, NMI-infected animals had more bacteria in the spleen than the lungs in comparison to the numbers of bacteria in NMII-infected animals, which had more bacteria in the lungs than the spleen. This suggests a more systemic infection by NMI, which it causes by either moving extracellularly through the circulatory system or being engulfed by phagocytic cells, which then traffic to the spleen. The increased splenomegaly on day 14 in NMI-infected mice supports the idea of infected phagocytes trafficking from the lung to the spleen. Because the mice were SCID mice with no lymphocytes, the majority of cells causing splenomegaly are likely macrophages and neutrophils, as well as, possibly, dendritic cells and NK cells. A decrease in the percentage of infected alveolar macrophages and neutrophils on day 7 p.i. in NMI-infected mice also supports the idea that infected cells traffic to the spleen. An increase in the number of infected alveolar neutrophils and macrophages on day 7 p.i., a smaller number of bacteria in the spleen on day 14, and a greater number of bacteria in the lung than the spleen across all time points suggest a lower systemic response in NMII-infected mice.

Currently, the most common model for testing C. burnetii vaccine candidates in mice uses the intraperitoneal route infection. In this model, splenomegaly and C. burnetii genomic copy number are indicators of vaccine protectiveness. Because the natural route of infection is aerosol, the next step will be to create a model for aerosol infection. This study, along with a previous study from 2005 (27), found that aerosol challenge causes splenomegaly, increased genomic copy number in the spleen and lungs, and inflammatory histology in the lungs, suggesting a working infection model. A study is under way to test vaccine protectiveness in wild-type mice using aerosol challenge with C. burnetii.