Hiroshi Tsujibo, Department of Microbiology, Osaka University of Pharmaceutical Sciences, 4-20-1 Nasahara, Takatsuki, Osaka 569-1094, Japan. Tel: +81 72 690 1057; fax: +81 72 690 1057; email: email@example.com
Acinetobacter baumannii is a major cause of both community-associated and nosocomial infections worldwide. These infections are difficult to treat because the bacterium rapidly develops resistance to multiple antibiotics. However, little is known about the nature of the innate cellular response to A. baumannii infection. In the present study, we identified the cells infiltrating the lungs of mice with Acinetobacter pneumonia and analyzed their response to infection. Normal mice eradicated the A. baumannii infection within 3 days of inoculation. Neutrophils were rapidly recruited to the lungs, followed by macrophages and NK1.1+ cells. Neutrophil-depleted mice showed acute and severe symptoms, and all of the mice died within 3 days of inoculation. The majority of macrophage-depleted mice responded in a similar manner to the control mice. These results indicate that neutrophils are essential for the elimination of A. baumannii. Half of NK1.1+ cell-depleted mice died within 1 day of inoculation and the number of infiltrating neutrophils was lower than that in control mice up until 3 days post-inoculation. Moreover, the expression levels of keratinocyte chemoattractant protein (KC) decreased in NK1.1+ cell-depleted mice. These results indicate that NK1.1+ cells recruit neutrophils during the early phase of Acinetobacter infection by increasing KC expression.
Acinetobacter baumannii is a ubiquitous Gram-negative bacterium that can survive for prolonged periods in water, soil, and on the skin of healthy humans. During the last decade, A. baumannii has emerged as a major cause of both community-associated and nosocomial infections worldwide (1–3). The urinary tract, intravenous devices, surgical sites, and decubitus are the favored sites of infection. A. baumannii mainly causes pneumonia, particularly in mechanically ventilated patients (4, 5). The mortality rate for ventilator-associated pneumonia caused by A. baumannii has been reported to be <75% (6, 7). However, little is known about the cellular and molecular mechanisms underlying host defenses against respiratory infection by A. baumannii (8–10). Therefore, a deeper understanding of the innate immune system may provide new possibilities for the treatment of nosocomial pneumonia.
The innate immune system is the first line of defense against many bacterial pathogens, including A. baumannii. Bacterial pathogens are recognized by phagocytes, such as macrophages and neutrophils, and are rapidly eliminated from a host suffering from acute infection. CD14 and Toll-like receptor 4 play a key role in the innate sensing of A. baumannii via bacterial lipopolysaccharide (LPS) (9). Recently, van Faassen et al. reported that neutrophils play an important role in host resistance to Acinetobacter pneumonia (11). However, little is known about the innate cellular response and the interactions between these cells in A. baumannii pneumonia. Recent reports suggest that neutrophils engage in cross-talk with other leukocytes during inflammatory responses (12, 13). Immune cells (e.g. macrophages, neutrophils, NK cells, NKT cells, αβT cells, and γδT cells) play an important role in the maintenance of tissue homeostasis in the lungs. Of these, NK cells and NKT cells play a crucial role in the innate immune response to tumors, viruses, and intracellular bacteria, and also have an immunoregulatory effect on other immune cells, such as T cells, B cells, macrophages, and dendritic cells (14–20). Moreover, NK cells modulate neutrophil activation and survival by secreting various cytokines and by direct cell–cell contact (21, 22). However, because most reports are of in vitro studies, little is known about the role and interaction of these cells within infected tissues. The aim of the present study was to identify the cells infiltrating the lungs of mice with Acinetobacter pneumonia and to examine their role in host defense.
MATERIALS AND METHODS
Bacterial strains and media
Acinetobacter baumannii strain A112-II-a was isolated from a patient with chronic nephritis. The bacterium was pre-cultured in LB broth (BD Difco, Franklin Lakes, NJ, USA) at 37°C for 3 hrs. The culture was diluted 1:100 into fresh broth and then shaken at 37°C until the late logarithmic growth phase. To produce agar medium, LB broth was solidified by adding 1.5% (wt/vol) agar (Nacalai Tesque, Kyoto, Japan).
Specific pathogen-free female C57BL/6 mice were purchased from Japan SLC (Shizuoka, Japan). All experimental mice were 8–10 weeks old. The animals were housed under specific pathogen-free conditions in a small level two animal containment facility and given free access to sterile water and certified mouse chow. All experiments were carried out in accordance with the guidelines for the care and use of laboratory animals of Osaka University of Pharmaceutical Sciences.
Respiratory infection with A. baumannii.
Acinetobacter baumannii was grown until the late logarithmic growth phase, centrifuged at 3,500 ×g for 10 min, resuspended and diluted appropriately in PBS, and used immediately. Mice were anesthetized and i.n. inoculated with approximately 107 or 108 CFU A. baumannii in 50 μL PBS. The actual inoculum concentrations were determined by plating 10-fold serial dilutions onto LB ager plates. Clinical signs were monitored and scored as follows: 0, no abnormal clinical signs; 1, ruffled fur and moving slowly; 2, ruffled fur, hunched posture, and moving very slowly; 3, hunched posture, moving very slowly, and squeezed eyes; 4, dead.
Pulmonary lobes were harvested at the indicated time points and fixed in 10% neutral buffered formalin, which was then replaced by a sucrose solution. The lungs were then embedded in OTC (Tissue-Tec; Miles Inc., Elkhart, IN, USA) and frozen at −80°C. The tissue segments were sectioned (6 μm) on a cryostat and stained with hematoxylin and eosin (H & E).
Distribution of A. baumannii
Acinetobacter baumannii-inoculated mice were killed and lungs and spleen were removed. Each tissue was homogenized with PBS in a loose glass homogenizer. Cell suspensions were plated on LB agar plates and cultured at 37°C for 12 hrs.
Anti-M-CSFR (AFS98) was a gift from Dr S. I. Nishikawa (RIKEN, Kobe, Japan) (21). Anti-Gr1 (RB6–8C5) and anti-NK1.1 (PK136) were provided by the Cell Resource Center for Biomedical Research Institute of Development, Aging and Cancer Tohoku University. Anti-CD11b (M1/70), CD45 (30-F11), CD3 (145–2C11) and CD49 (DX5) were purchased from BD Pharmingen (San Jose, CA, USA).
To deplete neutrophils, NK/NKT cells, and macrophages, mice were injected i.p. with 250 μg anti-mouse monoclonal antibodies, RB6–8C5, PK136, and AFS98 (23–25), respectively, on Days 5, 3, and 1 before and Days 1 and 3 post-inoculation with A. baumannii.
Analysis of infiltrating cells
Pulmonary lobes were removed, minced in Hanks’ Balanced Salt Solution (HBSS; Invitrogen, Carlsbad, CA, USA) and incubated with 150 U/mL collagenase (Sigma, St Louis, MO, USA) and 0.1 mg/mL DNase I (Wako Pure Chemicals, Osaka, Japan) for 30 min at 37°C. Spleens were homogenized in PBS using a loose glass homogenizer, centrifuged for 5 min, resuspended in PBS, and passed through nylon mesh (70 μm).
Flow cytometry analysis
Fluorescein isothiocyanate (FITC)-conjugated rat anti-mouse Gr1 mAb (BD Pharmingen), R-PE-conjugated rat anti-mouse CD11b mAb (BD Pharmingen), R-PE-conjugated mouse anti-mouse CD49 mAb (BD Pharmingen), and FITC-conjugated armenian hamster anti-mouse CD3 were used as the primary antibodies. Flow cytometric profiles were analyzed using a FACScan analyzer and CellQuest software (Becton Dickinson, Mountain View, CA, USA).
Mice were anesthetized and inoculated i.n. with approximately 107 CFU of A. baumannii and the lungs harvested on Days 1 and 3 post-infection. Total RNA was isolated from lung tissue using an RNeasy Mini Kit (Qiagen, Tokyo, Japan), and treated with DNaseI (Qiagen). RNA was transcribed to cDNA using M-MLV reverse transcriptase (Promega, Madison, WI, USA) and the cDNA was amplified with AmpliTaq gold (Applied Biosystems, Foster City, CA, USA). The primer pairs used to amplify keratinocyte chemoattractant protein, KC (CXCL1) and hypoxanthine phosphorybosyl transferase (HPRT) were: KC, 5′-TAT CGC CAA TGA GCT GCG C-3′ and 5′-AAG CCA GCG TTC ACC AGA C-3; and HPRT, 5′-CTG TAG ATT TTA TCA GAC TGA AGA G-3′ and 5′-GTC AAG GGC ATA TCC AAC AAC AAA-3′.
Measurement of KC levels
Groups of five PK136 or rIgG-treated C57BL/6 mice were killed 1 and 3 days after i.n. inoculation with 107 CFU A. baumannii. The trachea were exposed through a midline incision and cannulated with a plastic catheter. Lungs were lavaged twice with 400 μL PBS and the lavage fluid centrifuged at 440 ×g for 5 min. The supernatant was collected and stored at −80°C for ELISA. The levels of KC in the BAL fluid were determined using mouse CXCL1/KC Quantikine Kits (R & D Systems, Minneapolis, MN, USA).
The significance of the differences was calculated using one-way analysis of variance. A P value of <0.05 was considered to be significant.
Acinetobacter pneumonia in normal mice
We first examined the host immune responses to Acinetobacter pneumonia. Because A. baumannii was easily eradicated within 3 days by healthy animals, we focused on the innate immune responses and analyzed the physiological mechanisms involved in the exclusion of A. baumannii. First, the effective dose of A. baumannii required for the development of experimental pneumonia in normal C57BL/6 mice was determined. When mice were inoculated with <108 CFU, all the mice survived; however, when a dose of 109 CFU was used, the survival rate was 83% (5/6 mice) after 7 days (data not shown). Therefore, 107 or 108 CFU of A. baumannii was chosen for the pneumonia model. Although all mice inoculated with 107 CFU lost weight up until Day 3 and showed mild clinical signs on Day 1, all recovered completely by Day 4 post-inoculation (Fig. 1A, B). The viable bacterial counts in the lungs and spleens were 105 CFU and 101 CFU, respectively, on Day 1, and no viable bacteria were detected by Day 3 (Fig. 1C). Histological examination of the lungs harvested from mice with pneumonia was undertaken on Days 0, 1, 3, 5, and 7 post-infection (Fig. 2). Many leukocytes had infiltrated the alveoli on Day 1, reaching maximum numbers on Day 3 (Fig. 2A–C). The number of leukocytes decreased on Day 5, and the alveoli had fully recovered by Day 7 (Fig. 2D, E). We next examined the profile of these infiltrating leukocytes using flow cytometry. Mac1+/Gr1high cells, Mac1+/Gr1low/− cells, NK1.1+/CD3− cells, and NK1.1+/CD3+ cells were identified as neutrophils, macrophages, NK cells, and NKT cells, respectively. The number of neutrophils in the alveoli increased up until Day 3 post-inoculation, and then returned to normal levels by Day 5 (Fig. 3A). Macrophages and NK cells also infiltrated the alveoli, reaching maximum levels on Day 3, before returning to normal by Day 7 (Fig. 3B, C). NKT cells were hardly detected in the alveoli, the number of these cells did not show significant change through seven days (Fig. 3D). These results were in agreement with those obtained from the histological analysis (Fig. 2).
Survival rates for pneumonia mice cell depleted of specific cell types
We next assessed the contribution made by neutrophils, macrophages and NK1.1+ cells to the elimination of A. baumannii by depleting each of the cell types using monoclonal antibodies. As described in Materials and Methods, mice were inoculated i.n. with 108 CFU A. baumannii. The survival rate of mice injected with the control Ab was 100%, whereas that of mice injected with anti-Gr1 Ab, anti-NK1.1 Ab, and anti-M-CSFR Ab was 0%, 50%, and 83%, respectively (Fig. 4). These results suggest that neutrophils are essential for the elimination of A. baumannii. They also suggest that NK1.1+ cells play an active protective role in host immune responses against A. baumannii. However, the contribution made by macrophages appears to be very small (Fig. 4). Therefore, we next examined the specific role of neutrophils and NK1.1+ cells in the elimination of A. baumannii.
Lung inflammation in neutrophil- and NK1.1+ cell-depleted mice
To examine the effects of neutrophils on the elimination of A. baumannii, neutrophil-depleted mice were inoculated i.n. with 107 CFU A. baumannii. The viable bacterial count in the lungs of the control mice was 5 × 105 CFU on Day 1, although no bacteria were detected on Day 3 (Fig. 5A). However, in mice injected with anti-Gr1 Ab (neutrophil-depleted), the viable bacterial count was 6 × 107 CFU on Day 1 and 7 × 103 CFU on Day 3. The viable bacterial count in NK1.1+ cell-depleted mice was similar to that in control mice on Day 1, and the count was still 1 × 102 CFU on Day 3 (Fig. 5B). We then examined the profile of leukocytes infiltrating the lungs of cell-depleted mice with pneumonia. Neutrophils were not detected in mice injected with the anti-Gr1 Ab until Day 5 (Fig. 6A). The number of macrophages infiltrating into alveoli was higher than that in control mice up until Day 3, but decreased to similar levels by Day 5 (Fig. 6B). The number of NK cells continued to increase up until Day 7 in both pneumonia and control mice (Fig. 6C). Interestingly, the number of infiltrating neutrophils was less than that in control mice up until Day 3 (Fig. 7A). These results indicate that neutrophils are essential for the elimination of A. baumannii, and that NK1.1+ cells play a role in the migration of neutrophils into the alveoli of Acinetobacter pneumonia mice. The number of infiltrating macrophages was similar to that in the control mice (Fig. 7B). Small numbers of NK cells were observed up until Day 7 in mice injected with the anti-NK1.1 Ab (Fig. 7C).
Chemokine expression in NK1.1+ cell-depleted mice with pneumonia
To elucidate the role played by NK1.1+ cells in the migration of neutrophils, the expression level of chemokines was measured in the lung tissues of anti-NK1.1 Ab-injected mice with pneumonia. RT-PCR was used to detect CXC chemokine mRNAs in lung tissues, as CXC chemokines are chemotactic for neutrophils. As shown in Figure 8A, lung tissues from control mice constantly expressed KC (CXCL1) mRNA, even after Acinetobacter infection; however, the KC levels in mice injected with anti-NK1.1 Ab were lower than those in the control mice on Days 1 and 3.
In addition to KC mRNA levels, the amount of KC protein in the BAL fluid was measured by ELISA (Fig. 8B). There was no significant difference in the level of KC in the BAL fluid between anti-NK1.1 Ab-injected mice and control Ab-injected mice on Day 0. The level of KC in the BAL fluid of the control Ab-injected and anti-NK1.1 Ab-injected mice increased substantially following Acinetobacter challenge, reaching maximum levels in control mice on Day 1, before returning to normal on Day 5. However, KC levels in anti-NK1.1 Ab-injected mice were maximal on Day 3, although they remained lower than those in control mice from Day 1 to Day 5.
Nosocomial infection with A. baumannii pneumonia is an increasing threat because of high mortality rates and antibiotic resistance (6, 26–28). However, little is known about host defense against respiratory infection by this pathogen (9, 11, 29, 30). To investigate the pathology and the responses of immunocompetent cells to A. baumannii, we analyzed the cells infiltrating the lungs of mice with A. baumannii pneumonia and examined their role in the immune response. Normal healthy C57BL/6 mice inoculated i.n. with <108 CFU A. baumannii completely eliminated the pathogen within 3 days, and the inflamed lungs recovered within 7 days (Figs 1, 2). However, large numbers of neutrophils infiltrated the alveoli of mice with Acinetobacter pneumonia (Fig. 3). Increased numbers of macrophages, NK cells, αβT cells, and γδT cells were also observed up until 3 days post-inoculation, decreasing to normal levels thereafter (Fig. 3 and data not shown). Few NKT cells were detected in the alveoli, and the numbers of these cells were constant after A. baumannii infection (Fig. 3D). These results are consistent with earlier observations (11).
Next, we examined the effects of neutrophils on the elimination of A. baumannii using mice depleted of neutrophils by i.p. injection of an anti-Gr1 Ab. Neutrophils play an important role in host defense against bacterial pathogens (31, 32). A. baumannii caused severe pneumonia in mice injected with anti-Gr1 Ab, and the viable bacterial count in the lungs was 100-fold higher than that in control mice (6, and Fig. 5). Furthermore, all of the anti-Gr1 Ab-injected mice died within 3 days of inoculation (Fig. 4). However, 83% of mice injected with the anti-M-CSFR Ab survived (Fig. 4). These results indicate that host innate immune defenses in the respiratory tract of normal mice are mediated by neutrophils rather than by macrophages, which suppress bacterial growth and prevent the development of severe disease. The number of infiltrating NK cells in the lungs of both anti-Gr1 Ab-injected and control mice also increased from Day 1 post-inoculation (Fig. 6C); therefore, we next examined the effect of NK1.1+ cells on the elimination of A. baumannii.
Although NK cells play a key role in the immune response to tumors, viruses, and intracellular bacteria (33–36), little is known about their role in the response to extracellular bacterial infection (37). There are no published reports assessing the contribution of NK cells to the response against A. baumannii pneumonia. The functional role of the NK1.1+ cells was examined by injecting mice with an anti-NK1.1 Ab. As observed for the anti-Gr1 Ab-injected mice, mice injected with anti-NK1.1 Ab showed a reduced ability to eliminate the bacteria, and the overall survival rates were less than those in control mice (Figs 4, 5B). These results indicate that NK1.1+ cells play a crucial role in host defense against respiratory infection by A. baumannii. In anti-NK1.1 Ab-injected mice, the number of infiltrating neutrophils decreased compared with those in control mice up until Day 3 post-inoculation, and the viable bacterial count in the lungs was 100-fold higher than that in control mice by Day 3 (Figs 5B, 7A). Moreover, as shown in Fig. 8, the expression levels of KC in anti-NK1.1 Ab-injected mice were significantly lower than those in control mice. These results suggest that NK1.1+ cells induce the recruitment of neutrophils by increasing the expression of KC during the early phase of Acinetobacter infection. NK1.1 is expressed on NK cells and NKT cells, so anti-NK1.1 Ab treatment depleted NK cells and NKT cells. In this experiment, these results may be caused by NK cells and/or NKT cells. However, it is likely that NK cells rather than NKT cells play an important role in the recruitment of neutrophils during A. baumannii infection, because the numbers of NKT cells were not significantly increased in the lung during infection. NK cells, along with CD8+ T cells, function as key effector cells during Th1-type immune responses, and secrete inflammatory cytokines such as IFN-γ and TNF-α. A recent study shows that A/J mice are much more sensitive to Acinetobacter baumannii infection than C57BL/6 mice, due to delayed neutrophil recruitment during the early phase of infection (38). C57BL/6 and A/J mice exhibit predominantlyTh1-type and Th2-type immune responses, respectively, and Th1-type cytokines, such as IFN-γ and IL-12, induce early neutrophil-mediated host defenses (39, 40). Judging from these reports, the neutrophil recruitment essential for the elimination of A. baumannii may be induced by Th1-type immune responses, and these Th1-type cytokines may be secreted by NK1.1+ cells. NKT cells can make both the Th1-type cytokine IFN-γ and the Th2-type cytokines IL-4 and IL-13. These cells appear to play an important role in allergy, autoimmunity, and tumor control. Moreover, NKT cells play an important protective role in bacterial infection (19, 20). However, Bourgeois et al. reported that NKT cells suppressed neutrophil migration into the lung via Th1-type cytokines IFN-γ and IL-12 (41).It is necessary to clarify whether NK cells or NKT cells are important in the migration of neutrophils.
IL-17A is thought to participate in host defense against various pathogens and induce the production of TNF-α and CXC chemokines in the lung (42–45). In the present study, the expression level of IL-17A increased in lung tissues at 1 day after inoculation of A. baumannii, and up-regulation of IL-17A was delayed by anti-NK1.1 Ab treatment (data not shown). IL-17A and IL-17F may increase the expression level of neutrophil chemotactic factors, including KC (in mouse), MIP-2 (in mouse and humans), and IL-8 (in humans) and may be driven by lung epithelial cells (46). Also, the IL-17A-producing cells in bacterially infected lungs appear to be γδT cells rather than CD4+ Th17 cells (47–49). In the present study, γδT cells were detected in the lungs of mice with Acinetobacter pneumonia, and their numbers rapidly increased up until Day 3 post-inoculation (data not shown). Thus, γδT cells may be involved in neutrophil recruitment and may directly or indirectly interact with NK1.1+ cells. The detailed molecular mechanisms underlying the role of γδT cells on Acinetobacter pneumonia remain to be elucidated.
In conclusion, the results of the present study show that NK1.1+ cells induce neutrophil recruitment by increasing the expression levels of KC during the early phase of Acinetobacter infection. Further understanding of the molecular mechanisms underlying NK1.1+ cell-mediated immune regulation may lead to improved control of A. baumannii infections.
This study was supported in part by a Grant-in-Aid for High Technology Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan. We are grateful to Professor Shin-Ichi Nishikawa for supplying the anti-M-CSF monoclonal antibody, AFS98.
The authors who have taken part in this study declare that they do not have anything to disclose regarding funding or conflict of interest with respect to this manuscript.