Toll-Like Receptor 9 Contributes to Defense against Acinetobacter baumannii Infection

Abstract

Acinetobacter baumannii is a common nosocomial pathogen capable of causing severe diseases associated with significant morbidity and mortality in impaired hosts. Pattern recognition receptors, such as the Toll-like receptors (TLRs), play a key role in pathogen detection and function to alert the immune system to infection. Here, we examine the role for TLR9 signaling in response to A. baumannii infection. In a murine model of A. baumannii pneumonia, TLR9^−/− mice exhibit significantly increased bacterial burdens in the lungs, increased extrapulmonary bacterial dissemination, and more severe lung pathology compared with those in wild-type mice. Following systemic A. baumannii infection, TLR9^−/− mice have significantly increased bacterial burdens in the lungs, as well as decreased proinflammatory cytokine and chemokine production. These results demonstrate that TLR9-mediated pathogen detection is important for host defense against the opportunistic pathogen Acinetobacter baumannii.

INTRODUCTION

Acinetobacter baumannii is an opportunistic Gram-negative pathogen that causes a multitude of infections in persons with impaired host defenses (1). A. baumannii is among the most frequent causes of nosocomial infections in critically ill patients worldwide, where it is capable of causing pneumonia, urinary tract, and bloodstream infections and is also a frequent cause of burn wound infections (1–3). Patients with immunosuppression related to solid organ and hematologic transplantation are at increased risk for A. baumannii disease, and these infections are associated with high mortality (4, 5). The public health threat of A. baumannii is amplified by the organism's remarkable ability to acquire resistance to antimicrobial agents. Isolates resistant to all available antibiotics are increasingly encountered in clinical practice, making treatment of these infections difficult if not impossible (6–8). Furthermore, the ability to resist disinfectants and survive on abiotic surfaces allows A. baumannii to persist in clinical settings, where it is capable of causing endemics (9, 10).

The relative paucity of A. baumannii infections among persons without preexisting injury or immune compromise suggests that innate host defenses are sufficient to protect against A. baumannii disease. Despite the apparent importance of host defense, relatively little is known about the immune response to A. baumannii. Neutrophils and macrophages are recruited to the site of A. baumannii infection and are key cellular mediators of the host response, as depletion of these cell types lead to increased mortality and increased bacterial burdens in murine models of A. baumannii pneumonia (11–13) and systemic infection (14). Detection of invading pathogens through recognition of conserved pathogen-associated molecular patterns (PAMPs) by host pattern recognition receptors, such as Toll-like receptors (TLRs) and NOD-like receptors (NLRs), is necessary for a coordinated response to infection (15). TLRs comprise a family of membrane-spanning pattern recognition receptors that respond to a variety of conserved molecular patterns of viral, bacterial, and fungal origins. Upon binding of their cognate PAMP, the adaptor protein MyD88 (or TRIF in the case of TLR3) is recruited by the conserved Toll-interleukin 1 (IL-1) receptor homology domain (TIR) of TLRs, and canonical downstream signaling results in the activation of transcription factors, such as nuclear factor kappa B (NF-κB), with the resultant production of proinflammatory cytokines (16–18). To date, only TLR2, which recognizes bacterial lipoteichoic acid, and TLR4, which recognizes bacterial lipopolysaccharide (LPS), have been studied in the context of A. baumannii infection models.

TLR9 is an intracellular receptor with expression of the active form confined to endolysosomes, where it recognizes unmethylated CpG motifs from bacterial or viral DNA (19–21). Signaling through TLR9 has various effects on disease outcomes in murine models of acute bacterial infections. Mice lacking TLR9 have increased mortality and increased bacterial burdens in Klebsiella pneumoniae, Legionella pneumophila, and Streptococcus pneumoniae models of pneumonia, as well as a Neisseria meningitides model of sepsis (22–26). However, TLR9-mediated pathogen detection is detrimental in a Staphylococcus aureus murine pneumonia model, as well as a cecal ligation and puncture model of polymicrobial sepsis (27, 28). Given the various effects of TLR9 signaling in murine models of acute bacterial infections and the observation that A. baumannii detection in the lungs can occur in a TLR4-independent manner, (29), we sought to determine the role of TLR9 in host defense against A. baumannii. Using murine models of A. baumannii pneumonia and systemic infection, we show that TLR9 contributes to innate defense against A. baumannii disease.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

All experiments were performed using A. baumannii strain ATCC 17978 grown in lysogeny broth or agar at 37°C with shaking.

TLR9 reporter assay.

HEK-Blue hTLR9 and Null1 cells were purchased from Invivogen (San Diego, CA, USA). Cells were cultured according to the manufacturer's recommendations. A. baumannii ATCC 17978 was grown for 3.5 h, washed twice in phosphate-buffered saline (PBS), and resuspended in PBS to the appropriate cell density. Alternatively, A. baumannii ATCC 17978 cells were chemically killed by adding an equal volume of cold ethanol-acetone (1:1) mixture to the culture and incubating for 5 min on ice. Killed bacteria were pelleted, washed once with ethanol-acetone, and then washed and resuspended in PBS. Twenty microliters of the A. baumannii cell suspension was added to a 96-well plate, and 280,000 Null1 cells or 450,000 hTLR9 cells were added to each well to achieve a final volume of 200 μl. Plates were incubated for 20 h at 37°C with 5% CO₂. Alternatively, 1 μg of purified A. baumannii ATCC 17978 genomic DNA or 20 μl of filter-sterilized, conditioned culture supernatant from an overnight culture of A. baumannii ATCC 17978 was added to each well. Following incubation, secreted embryonic alkaline phosphatase (SEAP) activity was measured using the QUANTI-Blue reagent according to the manufacturer's recommendations (Invivogen, San Diego, CA, USA) by measuring the absorbance at 620 nm on a BioTek Synergy 2 microplate reader (BioTek, Winooski, VT, USA).

Gentamicin protection assay.

An amount of 2.5 × 10⁵ HEK cells was plated into each well of a 12-well tissue culture dish in a total volume of 500 μl of Dulbecco modified Eagle medium (DMEM) plus 10% fetal bovine serum (FBS). The following day, A. baumannii ATCC 17978 was subcultured for 3.5 h, washed twice with PBS, and resuspended in PBS to the appropriate cell density. Fifty microliters of the bacterial suspension was added to each well of HEK cells, and the plates were incubated at 37°C under 5% CO₂ for 4 h. Following incubation, the supernatant was removed, each well was washed twice with PBS, and 500 μl of medium containing 100 μg/ml of gentamicin was added. The cells were incubated for 2 h, washed twice with PBS, and suspended in PBS. The cell suspension was serially diluted and plated onto lysogeny agar for bacterial enumeration.

A. baumannii infections.

Wild-type C57BL/6 mice were purchased from Jackson Laboratories. TLR9^−/− mice were provided by Gregory B. Barton at the University of California, Berkeley (19, 30). All animal experiments were approved by the Vanderbilt University Institutional Animal Care and Use Committee. The murine model of A. baumannii pneumonia was performed as previously described (31–36). Briefly, A. baumannii ATCC 17978 bacteria were back diluted 1:1,000 from overnight culture and grown for 3.5 h at 37°C with shaking, washed twice with cold PBS, and resuspended in PBS at an appropriate cell density for infection. Eight-week-old female mice were infected intranasally with 3 × 10⁸ CFU A. baumannii ATCC 17978 bacteria in 30 μl PBS. At 36 h postinfection, mice were euthanized and organs and blood were harvested. For systemic infection, A. baumannii ATCC 17978 was prepared in a similar manner and mice were infected retroorbitally with 5 × 10⁸ CFU A. baumannii ATCC 17978 bacteria in 100 μl PBS. At 96 h postinfection, mice were euthanized and organs and blood were harvested. Lungs, livers, and spleens were homogenized and plated onto lysogeny agar for bacterial enumeration. For histological analyses, lungs were inflated with 1 ml of 10% formalin, fixed, embedded, and stained as described previously (31).

Cytokine quantification.

Interferon gamma (IFN-γ), IL-6, monocyte chemoattractant protein 1 (MCP-1), and tumor necrosis factor alpha (TNF-α) were measured using the BD cytometric bead array mouse inflammation kit according to the manufacturer's recommendations (Becton, Dickinson and Company, San Jose, CA, USA). Briefly, serum and lung homogenates from infected mice were diluted 1:2 to 1:5 in assay diluent, mixed with mouse inflammation capture beads, and incubated for 2 h at room temperature in the dark. Following incubation, samples were washed and resuspended in wash buffer, and flow cytometric data acquired on the 3-laser BD LSRII (Becton, Dickinson and Company, San Jose, CA, USA) at the Vanderbilt Flow Cytometry Shared Resource. Data were analyzed using the FCAP Array analysis software (Soft Flow, Inc., St. Louis Park, MN, USA).

Immune cell recruitment.

Flow cytometric analyses were performed with total erythrocyte-free lung cells isolated at 36 h postinfection from individual mice infected with A. baumannii ATCC 17978 as described above. Lungs were sectioned, digested with collagenase and DNase for 30 min, and passed through a 70-μm cell strainer prior to erythrocyte lysis. Antibodies and reagents for cell surface staining were purchased from BD Pharmingen (San Jose, CA, USA). Analyses were carried out on the 3-laser BD LSRII (Becton, Dickinson and Company, San Jose, CA, USA) at the Vanderbilt Flow Cytometry Shared Resource, and analyses were performed using FlowJo software (TreeStar, Inc.). Neutrophils were gated as Ly6G⁺ CD11b⁺, macrophages as CD11b⁺ F4/80⁺, dendritic cells as CD11c⁺, and NK cells as NK1.1⁺. Absolute cell numbers were determined using AccuCheck counting beads (Invitrogen, Grand Island, NY, USA).

Histology.

Paraffin-embedded mouse tissue sections were stained with hematoxylin and eosin or with antibody directed against type IV collagen by the Vanderbilt University Medical Center Translational Pathology Shared Resource. Lung inflammation was scored by K. L. Boyd, who was blinded to the treatment assignment, according to the following definitions: 0, no pathology; 1, minimal infiltrates of neutrophils in alveolar spaces; 2, low numbers of neutrophils in alveoli; 3, moderate numbers of neutrophils and hemorrhaging in alveoli with occasional lobar involvement, focal necrosis of alveolar walls neutrophils in bronchioles; 4, large numbers of neutrophils, consolidation, and widespread alveolar necrosis.

Statistical analyses.

Two-tailed t tests were used to determine significance (P < 0.05). The standard errors of the means of at least three replicates are shown, and all experiments were performed in triplicate.

RESULTS

A. baumannii activates NF-κB in a TLR9-dependent manner.

To determine whether A. baumannii activates NF-κB in a TLR9-dependent manner, an NF-κB-inducible secreted embryonic alkaline phosphatase (SEAP) reporter system was used in HEK cells stably transfected with the human TLR9 gene (hTLR9) or in HEK cells (Null1) as a control. Dose-dependent NF-κB activation was seen in hTLR9 cells following incubation with A. baumannii at increasing multiplicities of infection. Dose-dependent NF-κB activation was not seen in Null1 cells, indicating that A. baumannii activates NF-κB in a TLR9-dependent manner (Fig. 1A). Gram-negative pathogens are capable of directly injecting effector molecules into host cells through secretion machinery, such as the type IV secretion system (T4SS) (37). Active translocation of effector molecules through a T4SS can result in signaling through host pattern recognition receptors (38). To determine whether A. baumannii-mediated TLR9 activation is dependent upon active bacterial processes, such as type IV secretion, TLR9-dependent NF-κB activation by chemically killed A. baumannii was assessed. Chemically killed A. baumannii activate NF-κB in a TLR9-dependent manner, although to a lesser extent than live bacteria (Fig. 1B), indicating that A. baumannii-mediated TLR9 signaling is largely independent of active bacterial processes. Together, these findings demonstrate that TLR9 detects A. baumannii in vitro.

FIG 1.

A. baumannii activates NF-κB in a TLR9-dependent manner. (A) HEK cells expressing an NF-κB-inducible secreted embryonic alkaline phosphatase (SEAP) reporter system (Null1) and HEK cells expressing human TLR9 and an NF-κB SEAP reporter (hTLR9) were infected with increasing inocula of A. baumannii, and SEAP activity was monitored by measuring the optical density at 620 nm (OD₆₂₀) following 22 h of incubation. (B) Null1 and hTLR9 cells were infected with A. baumannii or chemically killed A. baumannii (MOI of 10,000) and SEAP activity was monitored as described above. (C) Null1 and hTLR9 cells were treated with 1 μg of purified A. baumannii genomic DNA, 20 μl of filter-sterilized A. baumannii culture supernatant, or A. baumannii cells (MOI of 10,000) in the presence or absence of 5 μg/ml of cytochalasin D, and SEAP activity was monitored. (D) Null1 cells were infected with A. baumannii cells (MOI of 10,000). Following a 4-h incubation, cells were washed with PBS and treated with gentamicin, and intracellular bacteria were enumerated. PBS, phosphate-buffered saline; ODN, TLR9-stimulatory CpG DNA (oligodeoxynucleotides); MOI, multiplicity of infection; DNA, 1 μg purified A. baumannii genomic DNA; Sup, filter-sterilized A. baumannii culture supernatant; AB, A. baumannii cells; Cyto, cytochalasin D (5 μg/ml); Gm, gentamicin (100 μg/ml). Mean values of at least three independent experiments are shown, with error bars indicating standard errors of the means. *, P < 0.05.

TRL9 signaling requires intracellular localization of A. baumannii.

TLR9 signaling depends upon intracellular localization of bacterial nucleic acid, and it is unclear how a predominately extracellular pathogen is sensed by TLR9 in nonphagocytic cells. Purified A. baumannii DNA or filter-sterilized A. baumannii culture supernatant, which presumably contains shed nucleic acid, do not result in significant TLR9-dependent NF-κB activation, whereas intact A. baumannii cells do. This activation does not require phagocytosis, as the addition of the actin polymerization inhibitor cytochalasin D does not alter TRL9 detection of A. baumannii (Fig. 1C). To determine whether A. baumannii invades HEK cells during infection, gentamicin was used to kill extracellular bacteria, and intracellular bacteria were enumerated following a 4-h infection. Approximately 10⁴ CFU/ml of A. baumannii bacteria were recovered from the intracellular compartment, indicating that A. baumannii does invade HEK cells and is therefore available for detection by TLR9.

TLR9^−/− mice exhibit increased disease severity in a murine pneumonia model.

To examine the role of TLR9-dependent sensing of A. baumannii in vivo, an A. baumannii pneumonia model in C57BL/6 and TLR9^−/− mice was used. Mice were challenged intranasally with 3 × 10⁸ CFU A. baumannii and humanely euthanized at 36 h postinfection, as previously described (31). This time point was chosen because bacterial burdens and lung pathology peak in severity at approximately 36 h postinfection in this nonlethal pneumonia model (32). Compared to the lung burdens in wild-type mice, TLR9^−/− mice exhibited nearly 10-fold increases in bacterial burdens in the lungs at 36 h postinfection (Fig. 2A). In addition, TLR9^−/− mice had increased inflammatory infiltrates in the lung, increased lung injury, and greater loss of lung architecture visible on hematoxylin-and-eosin-stained lung sections at 36 h postinfection (Fig. 2B and C). Extensive necrosis of alveolar walls was evident in infected TLR9^−/− mice, which included destruction of the alveolar basement membrane, evidenced by immunohistochemical staining for type IV collagen (Fig. 2B). Taken together, these findings indicate that TLR9^−/− mice are impaired in their ability to control A. baumannii pneumonia, resulting in more extensive lung injury.

FIG 2.

TLR9^−/− mice exhibit increased disease severity in a murine pneumonia model. Wild-type C57BL/6 or TLR9^−/− mice were challenged intranasally with 3 × 10⁸ CFU A. baumannii bacteria, and lungs were harvested 36 h postinfection, homogenized, serially diluted, and plated onto LB agar for bacterial enumeration. (A) Bacterial burdens in individual mice are plotted, with the bars indicating the means and standard errors of the means displayed by the error bars. (B) Hematoxylin-and-eosin-stained lung sections from infected wild-type and TLR9^−/− mice at 36 h postinfection are shown at ×100 and ×400 magnification, as well as immunohistochemical staining of type IV collagen at ×600 magnification. (C) Lung inflammation was scored, and mean values are shown, with error bars indicating standard errors of the means (n = 8 in each group). *, P < 0.05.

TLR9^−/− mice have reduced proinflammatory cytokine production but increased inflammatory cell recruitment to the lungs.

TLR signaling results in proinflammatory cytokine and chemokine expression; therefore, we hypothesized that TLR9^−/− mice would exhibit reduced expression of inflammatory cytokines and chemokines. Interleukin 6 (IL-6), monocyte chemoattractant protein 1 (MCP-1), tumor necrosis factor alpha (TNF-α), and interferon gamma (IFN-γ) were measured in lung homogenates from infected wild-type and TLR9^−/− mice. IL-6 and MCP-1 were present in similar amounts in lung homogenates of A. baumannii-infected wild-type and TLR9^−/− mice. However, infected TLR9^−/− mice had reduced expression of TNF-α and IFN-γ in lung homogenates compared to their expression in wild-type mice (Fig. 3A). One potential consequence of differential cytokine and chemokine expression in TLR9^−/− mice is impaired inflammatory cell recruitment to the lungs, which could hamper the ability to control A. baumannii infection. However, A. baumannii-infected TLR9^−/− mice had significantly increased numbers of neutrophils, dendritic cells, and NK cells in the lungs at 36 h postinfection compared to the numbers in the lungs of infected wild-type mice (Fig. 3B). The heightened numbers of inflammatory cells in the lungs of TLR9^−/− mice correlate with the increased bacterial burdens and lung injury but not with proinflammatory cytokine and chemokine production.

FIG 3.

TLR9^−/− mice have increased inflammatory cell recruitment to the lungs but reduced proinflammatory cytokine production. Wild-type C57BL/6 or TLR9^−/− mice were challenged intranasally with 3 × 10⁸ CFU A. baumannii bacteria, and inflammatory cell recruitment was measured by flow cytometric analyses of total erythrocyte-free lung cells isolated from individual mice at 36 h postinfection. (A) Tumor necrosis factor alpha (TNF-α), interleukin 6 (IL-6), interferon gamma (IFN-γ), and monocyte chemoattractant protein 1 (MCP-1) were measured in lung homogenates from infected wild-type and TLR9^−/− mice at 36 h postinfection. (B) Gating of representative flow plots is shown above, with total numbers of neutrophils, macrophages, dendritic cells, and natural killer cells depicted below. DC, dendritic cells; NK, natural killer. Mean values and standard errors of the means are shown. *, P < 0.05.

TLR9^−/− mice have increased extrapulmonary bacterial dissemination and increased systemic cytokine production following pneumonic infection.

The marked lung injury and loss of basement membrane integrity evident in the lungs of A. baumannii-infected TLR9^−/− mice could result in increased extrapulmonary A. baumannii dissemination following pneumonic infection. Consistent with this hypothesis, TLR9^−/− mice infected intranasally with A. baumannii had greater than 10-fold increases in bacterial burdens in the liver and spleen compared to the bacterial burdens in these organs of wild-type mice (Fig. 4A). The increased extrapulmonary dissemination of A. baumannii in TLR9^−/− mice coincides with a significant increase in serum levels of IL-6 without significant differences in serum levels of TNF-α, IFN-γ, and MCP-1 (Fig. 4B). These findings indicate that signaling through TLR9 is necessary to limit extrapulmonary dissemination of A. baumannii and the resultant increase in systemic IL-6 from A. baumannii pneumonia.

FIG 4.

TLR9^−/− mice have increased extrapulmonary bacterial dissemination and increased systemic cytokine production following pneumonic infection. (A) Wild-type C57BL/6 or TLR9^−/− mice were challenged intranasally with 3 × 10⁸ CFU A. baumannii, and livers and spleens were harvested 36 h postinfection, homogenized, serially diluted, and plated onto LB agar for bacterial enumeration. (B) Tumor necrosis factor alpha (TNF-α), interleukin 6 (IL-6), interferon gamma (IFN-γ), and monocyte chemoattractant protein 1 (MCP-1) were measured in serum from infected wild-type and TLR9^−/− mice at 36 h postinfection. Mean values and standard errors of the means are shown. *, P < 0.05.

TLR9 signaling limits colonization of the lung but not the liver or spleen during A. baumannii systemic infection.

To determine whether the importance of TLR9 in control of A. baumannii pneumonia extends to other infection types, the role of TLR9 in a systemic A. baumannii infection model was examined. Mice were challenged retroorbitally with 5 × 10⁸ A. baumannii bacteria, an inoculum that does not result in death, and colonization of the lung, liver, and spleen was assessed 96 h postinfection. The A. baumannii burdens in the liver and spleen were low overall and did not differ between TLR9^−/− and wild-type mice (Fig. 5A). Approximately 15-fold more A. baumannii bacteria were recovered from the lungs of TLR9^−/− mice than from the lungs of wild-type mice (Fig. 5A), indicating that TLR9 signaling limits colonization of the lung during A. baumannii systemic infection. Consistent with the increased bacterial burdens in the lungs of TLR9^−/− mice, the serum levels of TNF-α, IFN-γ, and MCP-1 were significantly decreased in systemically infected TLR9^−/− mice compared to the levels in wild-type mice (Fig. 5B). These findings support a role for TLR9 signaling in host defense against A. baumannii systemic infection.

FIG 5.

TLR9 signaling limits colonization of the lung but not the liver or spleen during A. baumannii systemic infection. (A) Wild-type C57BL/6 or TLR9^−/− mice were challenged retroorbitally with 5 × 10⁸ CFU A. baumannii bacteria, and lungs, livers, and spleens were harvested 36 h postinfection, homogenized, serially diluted, and plated onto LB agar for bacterial enumeration. (B) Tumor necrosis factor alpha (TNF-α), interleukin 6 (IL-6), interferon gamma (IFN-γ), and monocyte chemoattractant protein 1 (MCP-1) were measured in serum from infected wild-type and TLR9^−/− mice at 96 h postinfection. Mean values and standard errors of the means are shown. *, P < 0.05.

DISCUSSION

Much remains to be learned about innate defenses against bacterial pathogens. Furthering this knowledge is of the utmost value when studying pathogens that (i) are relatively incapable of causing disease in healthy persons with intact defenses and (ii) when studying pathogens that have acquired multi- and, in some cases, panresistance to the existing antibiotic armamentarium. Understanding how the healthy host successfully senses and defends against such pathogens may provide insights leading to much-needed therapeutic approaches centered on recapitulating an effective immune response. A. baumannii is one such pathogen that causes severe diseases associated with high mortality in at-risk patients (1–5). In addition, A. baumannii has developed resistance to every antibiotic available to treat these infections (6–8). This has rendered A. baumannii among the most problematic multidrug-resistant nosocomial pathogens, prompting the Infectious Diseases Society of America to list this organism in their Bad Bugs-No Drugs Campaign (39, 40).

Here, we define a role for TLR9 in the host response to A. baumannii in two murine models of infection. TLR9-mediated detection of A. baumannii contributes to control of A. baumannii pneumonia, as mice lacking TLR9 had increased bacterial burdens in the lung, more severe lung injury, and increased extrapulmonary bacterial dissemination. Similarly, TLR9 signaling limits A. baumannii colonization of the lung during systemic infection. These findings add to the existing understanding of host recognition of A. baumannii. In human lung epithelial cells, A. baumannii induces IL-8 expression in a TLR2- and TLR4-dependent manner. A. baumannii-mediated NF-κB activation is dependent upon TLR2 and TLR4 in human monocyte cells (41, 42). The importance of TLR4 was confirmed in a murine model of A. baumannii pneumonia, as mice lacking TLR4 had increased bacterial burdens in the lungs compared to the burdens in wild-type mice 4 and 24 h postinfection. However, using the same pneumonia model, mice lacking TLR2 had reduced bacterial burdens in the lungs compared to the burdens in wild-type mice at 24 h postinfection (43). In contrast, the absence of TLR4 was protective in a lethal murine model of A. baumannii sepsis (44). However, this finding was unique to A. baumannii bacteria with enhanced LPS shedding that induced endotoxemia and heightened systemic inflammation without an increase in bacterial burdens. The survival advantage in TLR4 null mice was not seen when a different A. baumannii strain was used (44). Although TLR2, TLR4, and now, TLR9 have been shown to sense A. baumannii in vitro, TLR4 and TLR9 signaling limits A. baumannii infection in vivo, whereas TLR2 signaling may potentiate infection. Nod1 and Nod2, cytosolic detectors of bacterial peptidoglycan components, have recently been shown to detect A. baumannii in human lung epithelial cells but do not contribute to A. baumannii detection in a macrophage cell line (45). The importance of these receptors in vivo has yet to be elucidated. The role of other pattern recognition receptors, including other TLRs, and NLRs in detection and response to A. baumannii infection remain to be determined.

The finding that TLR9 contributes to host defense in A. baumannii infection is consistent with the role of TLR9 in other bacterial pneumonia models. In K. pneumoniae, L. pneumophila, and S. pneumoniae models of pneumonia, mice lacking TLR9 have increased mortality and increased bacterial burdens in the lungs compared to the mortality and bacterial burdens in wild-type mice (22–25). However, the absence of TLR9 does not affect disease outcomes in murine models of H. influenzae and Brucella abortus pneumonia (46, 47), suggesting that either TLR9 does not detect DNA from these pathogens during the course of infection or TLR signaling is redundant in these models, rendering a single TLR dispensable. Finally, TLR9^−/− mice have reduced mortality and decreased bacterial burdens in an S. aureus pneumonia model, as well as a cecal ligation and puncture model of sepsis (27, 28). These findings reinforce the idea that TLR-mediated pathogen detection is not necessarily advantageous to the host and suggest that pathogen detection via host pattern recognition receptors and coordination of the downstream inflammatory response is highly nuanced and may differ based on the pathogen and infection model studied. Additionally, bacterial pathogens have evolved mechanisms to alter the host immune response to their advantage during infection, and this may contribute to the differential effects of TLR9 signaling in these infection models.

TLR9^−/− mice have increased bacterial burdens and inflammatory cell recruitment to the lungs following pneumonic infection, indicative of more severe pneumonia at 36 h postinfection. However, the levels of TNF-α and IFN-γ are reduced in the lungs of TLR9^−/− mice. Given the more severe infection at this time point, the reduction in TNF-α and IFN-γ may be a consequence of aberrant proinflammatory signaling in the absence of TLR9. In contrast, serum levels of TNF-α and IL-6 are increased in TLR9^−/− mice following pneumonic infection, which is likely a consequence of increased bacterial dissemination and greater extrapulmonary bacterial burdens. The cells responsible for differential proinflammatory cytokine expression in the lungs and serum remain to be elucidated.

The finding that in the systemic infection model, TLR9 signaling limits A. baumannii colonization of the lung but not the liver or spleen suggests that TLR9 has differential functions in these tissues. TLR9 is highly expressed in the pulmonary vascular endothelium, as well as alveolar and bronchial epithelia and lung-resident macrophages and neutrophils (48). The degree of TLR9 expression in nearly all cell types of the lung suggests that this pattern recognition receptor is uniquely important for innate defense in this organ, which is supported by the protective role of TLR9 during systemic infection demonstrated herein.

In conclusion, TLR9-mediated pathogen detection results in reduced bacterial burdens, improved lung pathology, and reduced dissemination in an A. baumannii pneumonia model, as well as reduced bacterial burdens and decreased proinflammatory cytokine production in a systemic model of A. baumannii infection. Improved understanding of the factors contributing to an effective innate immune response may provide avenues for immune modulation as a potential treatment strategy for infections caused by drug-resistant pathogens, including A. baumannii.