Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Individuals with cystic fibrosis (CF) are unusually susceptible to bronchopulmonary infection with the Gram-negative bacterium Pseudomonas aeruginosa. The infections, caused by the more virulent, mucoid strains of the bacteria, are chronic and often fatal (1). Although different mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene may account for variation in clinical expression of disease, there is a remarkable heterogeneity among CF patients in the severity of chronic bronchopulmonary P. aeruginosa infection (4, 5). Analyses of the relationship between the CF genotype and phenotype demonstrate that although there is a correlation between genotype and pancreatic function, genotype does not predict the severity and course of pulmonary disease (6). This suggests that genetic factors, distinct from the CFTR gene, regulate the host response to bronchopulmonary infection with P. aeruginosa.

It has been well established that genetic factors regulate the host response to a variety of microbial pathogens, including bacteria, viruses, and parasites (9). Although a number of experimental animals have been used for these studies, murine models have been particularly useful because of the availability of a variety of genetically well- defined, inbred mouse strains. Furthermore, the strategy of comparison of host responses in resistant versus susceptible mouse strains has been used successfully by our laboratory and others to elucidate the immune mechanisms leading to resolution versus disease progression in a number of infectious diseases, including infections peculiar to the lung (10).

A mouse model of chronic bronchopulmonary P. aeruginosa infection was established by intratracheal instillation of bacteria trapped in agar beads into the lungs (15). Recently, this model was used by Morissette and coworkers (16) and by our laboratory (17) to identify inbred mouse strains resistant and susceptible to chronic bronchopulmonary infection with a mucoid strain of P. aeruginosa originally isolated from a CF patient. Experiments performed in resistant and susceptible hosts identified a number of mechanisms that contribute to early and late host responses to bronchopulmonary P. aeruginosa infection. The magnitude of the inflammatory response and the level of tumor necrosis factor (TNF)- in the bronchoalveolar space within the first 3 d after infection were identified as factors important in the early, innate response (16, 18, 19).

Most studies to date of acquired immunity to P. aeruginosa have focused on the role of antibody that has traditionally been thought to be important in host defense against Gram-negative bacteria (20). This is in spite of numerous observations of a positive correlation in CF patients between high serum titers of P. aeruginosa-specific immunoglobulin (Ig)G and both the aggressive course of infection and poor clinical prognosis (21). The role of cell-mediated immunity, in particular the role of T cells, however, has received little attention. It has been observed that peripheral blood lymphocytes from CF patients with advanced lung disease have impaired proliferative responses to Pseudomonas antigens in vitro (24). More recently, comparison of T-cell clones derived from peripheral blood mononuclear cells of CF patients to clones from normal individuals revealed that there were no differences in proliferation in response to mitogen or tetanus toxoid (28). In addition, it was also shown that the helper phenotype of the CF-derived CD4⁺ T-cell clones was not abnormal (28). However, the microbiological status of the CF patients used in this study was not established. Thus, it is unclear if the earlier observations of decreased proliferation to specific antigen were due to an intrinsic defect in T cells of CF patients per se, or if the function of these cells is depressed as a result of chronic Pseudomonas infection. In addition, it is unclear if the responsiveness of peripheral blood lymphocytes correlates with that of the local T cells in the lung interstitium.

Previous studies by our laboratory demonstrated that lung interstitial T cells from BALB/c mice, which were identified as resistant to bronchopulmonary P. aeruginosa infection, had significantly higher in vitro proliferative responses to mitogen and specific antigen than T cells from C57BL/6 (B6) mice, which were found to harbor high numbers of bacteria in the lung at 2 wk after infection (17). Similar to P. aeruginosa-infected CF patients, susceptible B6 mice were found to have high serum levels of Pseudomonas-specific antibodies (17). Here, we further characterized chronic bronchopulmonary P. aeruginosa infection following intratracheal instillation of a mucoid strain of bacteria trapped in agar beads in resistant BALB/c and susceptible B6 mice. We also analyzed the phenotypes of leukocytes in the lungs of these hosts during the course of infection to determine if differences in the composition of cells at the local site of infection contribute to differences in host response to P. aeruginosa.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Mice

Age- and sex-matched B6 and BALB/c mice, 8-10 wk old, purchased from Charles River Laboratories (St. Constant, PQ, Canada) were used in all experiments according to the guidelines and regulations of the Canadian Council on Animal Care. Food and water were provided ad libitum.

P. aeruginosa

P. aeruginosa, strain 508, was a kind gift from Dr. Jacqueline Lagacé (Université de Montréal, Montréal, PQ, Canada). This strain has a mucoid appearance when grown on blood agar and was originally isolated from the sputum of a CF patient at Ste-Justine Hospital, Montréal.

Inoculum

P. aeruginosa trapped in agar beads were prepared according to a modification of previously described methods (15, 17). Briefly, log-phase bacteria grown in 4% proteose peptone (Difco, Detroit, MI) were concentrated 10-fold, and 1 ml was added to 9 ml of 1.5% trypticase soy agar prewarmed to 50°C. This mixture was added to 150 ml heavy mineral oil at 50°C and stirred rapidly with a magnetic stirring bar for 6 min at 22°C, followed by cooling with continuous stirring for 10 min more. The oil-agar mixture was centrifuged at 15,000 × g for 20 min to sediment the beads. The oil was removed and the beads were subsequently washed three times in phosphate-buffered saline (PBS) at 400 × g for 10 min at 22°C. The size of the beads was verified microscopically and only those preparations containing beads predominantly 100 to 150 µm in diameter were used as inoculum. The number of bacteria was estimated by homogenizing the bacteria-bead suspension using a Polytron homogenizer (Brinkmann Instruments Inc., Westbury, NY) and plating 10-fold serial dilutions on trypticase soy agar (BBL; Becton Dickinson & Co., Cockeysville, MD). The plates were incubated overnight at 37°C and the number of colony-forming units (CFU) was counted. The inoculum for infection was prepared by diluting the bead suspension with PBS to 4 × 10⁶ CFU/ml.

Intratracheal Infection

Mice were anesthetized with a combination of ketamine (15 mg/ml) and xylazine (2 mg/ml) administered intramuscularly at a dose of 0.2 ml. Following a transverse cervical incision, the trachea was exposed and intubated with a sterile, flexible 22-g cannula attached to a 1.0-ml syringe. An inoculum of 50 µl, containing approximately 1 to 2 × 10⁵ CFU, was implanted via the cannula into the lung. After inoculation, all incisions were closed by suture. None of the animals developed wound infection and healing occurred in 2 to 3 d.

Lung Cell Suspensions

Mice were killed by CO₂ overdose and exsanguinated by cutting the vena cava. The blood vessels were perfused by infusion via the retro-orbital venous plexus with 7 to 10 ml Mg²⁺- and Ca²⁺-free Hanks' balanced salt solution (HBSS) (GIBCO BRL, Grand Island, NY). The lungs were perfused further by injecting 3 to 5 ml HBSS through the heart. The lungs were excised and washed twice with RPMI 1640 (GIBCO BRL) supplemented with 10% fetal calf serum (FCS) (GIBCO BRL), 20 mM N-(2-hydroxyethyl)piperazine-N'-2-ethanesulfonic acid (GIBCO BRL), and 0.12% gentamycin (Schering-Plough, Point Claire, PQ, Canada). Lung cell suspensions were prepared using a previously described modification (17) of methods described by Holt and colleagues (29). Briefly, the lungs were minced and digested in complete medium containing 2 mg/ml collagenase (Type I; Sigma, St. Louis, MO), 10 U/ml elastase (Boehringer Mannheim, Laval, PQ, Canada), and 100 µg/ ml DNase I (Boehringer Mannheim). The tissues were incubated with stirring for 1 h at 37°C, the cells were dissociated by repeated pipeting, and large tissue fragments were sedimented. The single-cell suspension was washed twice with complete medium, overlayered on Lympholyte-M (Cedar Lane, Hornby, ON, Canada), and centrifuged for 20 min at 1,500 × g at 22°C. Cells at the interface were collected, washed twice, and resuspended in sorting buffer for surface marker staining. The viability of the cells as determined by trypan blue exclusion was > 80%.

Flow Cytometry

The phenotypes of lung cells were determined by single color staining of surface markers. Fluorescein isothiocyanate (FITC)-conjugated monoclonal antibodies (mAbs) against mouse CD3 (clone 145-2C11, hamster IgG), CD4 (clone RM4-5, rat IgG2a), CD8 (clone 53-6.7, rat IgG2a), CD45R/B220 (clone RA3-6B2, rat IgG2a), Mac-1 (clone M1/70, rat IgG2b), F4/80 (clone C1:A3-1, rat IgG2b), Ly-6G/Gr-1 (clone RB6-8C5, rat IgG2b), and CD45 (clone 30F11.1, rat IgG2b) were used. In each experiment, isotype-matched mAbs conjugated to FITC were used as negative controls. Anti-Mac-1 and anti-F4/80 mAbs were purchased from Serotec (Oxford, UK) and all other mAb- FITC conjugates were from PharMingen (San Diego, CA). Aliquots of 2 to 5 × 10⁵ cells were stained with 2 µg mAb for 30 min in sorting buffer consisting of PBS supplemented with 1% bovine serum albumin and 0.1% sodium azide. The cells were washed three times and resuspended in PBS containing 1% FCS and 0.1% sodium azide. All staining steps were carried out at 4°C. Acquisition of the cells was performed immediately after staining using FACScan (Becton-Dickinson, Mountain View, CA), which was calibrated with Calibrite Beads (Becton-Dickinson). Acquisition and analysis of data were executed by CellQuest software (Becton-Dickinson). For each sample, the cell populations with high expression of CD45 were gated for analyses and  10,000 events/sample were examined. Results for each marker are presented as the percentage of CD45⁺ cells, which ranged from 65 to 85% for each experiment. In preliminary experiments, two-color flow cytometry was performed using mAbs against CD3 and either CD4 or CD8. Subsequent experiments were performed using one-color flow cytometry using mAbs against either CD4 or CD8. Results were consistent for the percentages of CD4⁺ and CD8⁺ T cells whether one- or two-color cytometry was used for analysis.

Histology

Lungs from infected animals were instilled with 1 ml of buffered formalin through the trachea. The lobes of the lungs were kept distended by closing the trachea with suture. The entire lung-heart complex was dissected and preserved in formalin before setting up tissue blocks in paraffin. The tissues were cut into 5-µm sections and stained with hematoxylin and eosin.

Statistical Analyses

Data are presented as the means ± SEM of replicate experiments as indicated. Differences between groups of mice for all experiments with the exception of the mortality and clearance data were analyzed by the Student's t test (30). A probability of < 0.05 was considered significant. For the mortality and clearance data, differences between BALB/c and B6 mice were analyzed using ² analysis and a probability of < 0.05 was considered significant (30).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Chronic Bronchopulmonary P. aeruginosa Infection in Resistant BALB/c and Susceptible B6 Mice

To characterize bronchopulmonary P. aeruginosa infection further in resistant BALB/c and susceptible B6 mice, animals were infected by intratracheal inoculation of a mucoid strain of bacteria trapped in agar beads, and the course and outcome of infection were assessed for 28 d. By Day 6, 15% of susceptible B6 mice succumbed to infection in contrast to only 6% among resistant BALB/c mice, which represents a significant difference (Figure 1a). There were no further deaths among mice of either strain through 28 d after infection. In addition, B6 compared with BALB/c mice experienced more severe symptoms as evidenced by significantly greater weight loss during the first 14 d after infection (data not shown). B6 mice were also observed to suffer more severe lacrimation and lethargy. Resistant BALB/c mice, however, appeared to experience a higher and more severe incidence of fever, whereas the incidence and severity of fever were moderate among B6 mice.

View larger version (15K): [in this window] [in a new window]   Figure 1.   Cumulative mortality (a) and bacterial burden in the lungs (b) of P. aeruginosa-infected B6 and BALB/c mice. A total of 46 B6 and 51 BALB/c mice in two to four independent experiments were infected by intratracheal instillation of 1 to 2 × 105 CFU P. aeruginosa trapped in agar beads. The animals were followed daily and killed at the indicated times for determination of the numbers of bacteria in the lung. In a, *2 = 5.28, P < 0.025. In b, data were pooled from two to four independent experiments and are presented as means ± SEM. Total n: Day 7, B6 = 10, BALB/c = 11; Day 14, B6 = 21, BALB/c = 24; Day 21, B6 = 7, BALB/c = 8; and Day 24, B6 = 8, BALB/c = 8. *P = 0.002; **P < 0.001, and ***P = 0.008 for B6 compared with BALB/c mice.

The numbers of bacteria in the lung were assessed by determining CFU at weekly intervals for 28 d following intratracheal inoculation of resistant and susceptible hosts with bacteria trapped in agar beads. Significant differences were apparent between B6 and BALB/c mice in terms of both the P. aeruginosa load in the lung (Figure 1b) and the percentage of animals that completely cleared the bacteria (Table 1). At Day 7 after infection, B6 mice had a significantly greater number of P. aeruginosa in the lung with 1,400 times the mean number of bacteria compared with BALB/c mice. At this time, bacteria were undetectable in 46% of BALB/c mice, whereas none of the B6 mice cleared the infection. Although there were decreases in the bacterial burden in the lungs of both strains by 14 d after infection, the decrease was more dramatic in BALB/c mice. A highly significant difference between the strains in the mean number of bacteria was evident such that B6 mice harbored 2,000 times as many bacteria in the lung. Among B6 mice, only 13% cleared the infection, whereas 77% of BALB/c mice cleared the bacteria from the lung by Day 14. A significant difference in the bacterial burden of B6 versus BALB/c mice was also apparent at 21 d after infection. At this time, B6 mice had 200 times as many P. aeruginosa in the lung as BALB/c mice. Furthermore, 87% of B6 mice still harbored bacteria in the lung at this time, whereas 72% of BALB/c mice had cleared the infection. Although there was no significant difference in the bacterial load between B6 and BALB/c mice at Day 28, it is important to note that overall 67% of all infected BALB/c mice had cleared P. aeruginosa from the lung by Day 28, whereas bacteria were still detectable in almost 80% of B6 mice.

Lung Histopathology of P. aeruginosa-Infected Resistant BALB/c and Susceptible B6 Mice

Histological examination revealed that although bronchial and parenchymal changes were apparent in the lungs of P. aeruginosa-infected mice of both strains, marked differences between BALB/c and B6 hosts were visible in the type of inflammation in and around airways. Compared with the lungs of normal mice, instillation of sterile agar beads into the lungs of either mouse strain resulted in relatively mild, granulomatous-like inflammation localized around small airways where agar beads were lodged (Figures 2a and 2b). By Day 7 after infection, the large- and intermediate-sized airways of susceptible B6 mice were filled with cellular infiltrates, consisting primarily of polymorphonuclear neutrophils (PMN) and few macrophages, which extended into the surrounding parenchyma (Figure 2c). In comparison, inflammation in the airways of resistant BALB/c mice was mostly chronic or granulomatous in nature and consisted primarily of macrophages, epithelial histiocytes, and lymphocytes (Figure 2d). Atelectases and expansion of the interstitium with granuloma formation were observed in both mouse strains. Susceptible B6 mice presented histopathologic changes consistent with acute as well as organizing pneumonia, whereas tissue damage was much less severe in resistant BALB/c mice. By 14 d after infection, acute inflammation was still apparent in the large airways of B6 mice and there were marked reactive changes in the respiratory columnar epithelial cells, consisting mainly of nuclear enlargement and mild hyperplasia (Figure 2e). In contrast, little or no inflammation was apparent at this time in the airways of infected BALB/c mice, although occasional granulomas were present in the interstitium and macrophages were scattered in the alveolar spaces (Figure 2f). Inflammation was resolving in the resistant hosts with minimal signs of long-term sequelae.

View larger version (208K): [in this window] [in a new window]   Figure 2.   Lungs from representative normal (a), sterile agar bead implant (b), Day 7-infected B6 (c) and BALB/c (d), and Day 14- infected B6 (e) and BALB/c ( f ) mice. Mice received sterile agar beads intratracheally 7 d previously or were infected by intratracheal instillation of 1 to 2 × 105 CFU P. aeruginosa trapped in agar beads and killed at Days 7 or 14. In b, arrows indicate sterile agar beads in distal airways surrounded by mild, granulomatous-like inflammation. In c, the arrow indicates airway filled with exudate consisting predominantly of PMN. In d, arrows indicate airways with mild, chronic granulomatous inflammation. In e, the arrow indicates persistence of acute inflammation in airway. In f, the arrow indicates airway with little or no inflammation (hematoxylin and eosin; a, b, e, f, original magnification: ×100 and c, d, ×200). Scale bar = 100 µm.

Phenotype of Lung Leukocytes during Chronic Bronchopulmonary P. aeruginosa Infection in Resistant BALB/c and Susceptible B6 Mice

To determine if differences in leukocyte populations in the lung contribute to differences in the level of host resistance to chronic bronchopulmonary P. aeruginosa infection, flow cytometric analyses were performed on cells prepared from the lungs of resistant BALB/c and susceptible B6 mice at various times after intratracheal infection. First, we compared the numbers of total lung cells in normal, bead implant, and infected BALB/c and B6 mice (Figure 3). There were no significant differences between normal BALB/c versus normal B6 mice or bead implant BALB/c versus bead implant B6 mice. On Day 7 after infection, there was an approximately threefold increase in the number of cells in the lungs of both strains, but there were no significant differences between the strains at this time or at any day through 28 d after infection. We also compared the phenotypes of leukocytes in the lungs of normal animals of both strains or following intratracheal implantation of sterile agar beads. As shown in Table 2, there were no significant differences between normal B6 and BALB/c mice in terms of the major leukocyte compartments in the lung, that is, T cells, B cells, mature F4/ 80+ macrophages, and PMN. There was, however, a significant difference between normal mice of the two strains in the percentages of cells expressing Mac-1, a marker found on macrophages and some T cells as well as PMN following exposure to chemoattractants or cytokines (31- 33). Normal B6 mice were observed to have a significantly higher proportion of Mac-1⁺ cells than lung leukocytes recovered from normal BALB/c mice. Importantly, there were no significant differences between normal mice of either strain and their counterparts receiving bead implants in the percentages of inflammatory cells, that is, macrophages and PMN. Significant differences were apparent between bead implant mice of the two strains in the percentages of CD3⁺ and Mac-1⁺ lung cells.

View larger version (23K): [in this window] [in a new window]   Figure 3.   Numbers of total lung cells in normal, bead implant, and P. aeruginosa-infected B6 and BALB/c mice. Data are means ± SEM. Total n: normal mice, n = 3 per strain; mice implanted with sterile agar beads, n = 6 per strain; or at 7, 14, 21, and 28 d postintratracheal infection with 1 to 2 × 105 CFU P. aeruginosa trapped in agar beads, n = 3 mice per strain per time point.

Laser scatter plots from flow cytometric analyses of lung cells from normal and bead implant mice supported these findings. There were no differences between lung leukocytes from normal B6 and BALB/c mice in forward or side scatter (Figures 4A versus 4B). Implantation of agar beads into the lung did not significantly shift the forward or side scatter of lung leukocytes (Figures 4A versus 4C and 4B versus 4D). However, following intratracheal infection with P. aeruginosa, there were marked shifts in the populations of lung leukocytes. On Day 7 after infection, a population of large, granular inflammatory cells was apparent in the lungs of both B6 and BALB/c mice (Figures 4E and 4F). Moreover, visual analyses of the scatter plots indicated that the inflammatory response was of greater magnitude in B6 than in BALB/c mice. By Day 14 after infection, there were shifts in both mouse strains in the populations of lung leukocytes toward smaller, less granular cells characteristic of lymphocytes (Figures 4G and 4H).

View larger version (51K): [in this window] [in a new window]   Figure 4.   Representative laser light-scatter density plots from flow cytometric analyses of lung leukocytes from normal (A and B), sterile bead implant (C and D), or P. aeruginosa-infected B6 and BALB/c mice (E to H). Mice received sterile agar beads intratracheally 7 d previously or were infected by intratracheal instillation of 1 to 2 × 105 CFU P. aeruginosa trapped in agar beads and were killed at Days 7 (E and F ) or 14 (G and H) after infection. Single-cell suspensions of lung leukocytes were prepared by enzyme digestion. The horizontal axis is forward scatter, an indicator of cell size, and the vertical axis is orthogonal scatter, an indicator of cell granularity and surface irregularity. Arrow indicates area of large, granular inflammatory cells on Day 7 after infection.

Further analyses of the leukocyte populations by flow cytometry demonstrated that there were significant increases in CD3⁺ cells but not B220⁺ cells in both B6 and BALB/c mice as the infection progressed, and there were no significant differences between the strains at any time through Day 21 (Figure 5). In terms of the macrophage markers, Mac-1 and F4/80, there were significant decreases in Mac-1⁺ cells in both strains at Days 14 and 21 after infection, but there were no significant differences between the strains during infection. There were no significant differences in F4/80 expression within the strains or between the strains at any time during infection. Marked differences were, however, apparent in the proportions of PMN in the lungs of B6 and BALB/c mice and between these hosts during infection. Although there were significant increases in PMN in both strains on Days 7 and 14 after infection, the proportions of PMN were markedly and significantly greater in B6 mice than in BALB/c mice at these times.

View larger version (31K): [in this window] [in a new window]   Figure 5.   Proportions of leukocyte populations in the lungs of B6 (a) and BALB/c (b) mice at various times after intratracheal infection with P. aeruginosa. Mice were infected by intratracheal instillation of 1 to 2 × 105 CFU P. aeruginosa trapped in agar beads. Lungs were harvested at the indicated times and single-cell suspensions were prepared by enzyme digestion of lung tissue pooled from two mice per group. The cells were stained with FITC-conjugated mAbs against mouse CD3, B220, Mac-1, F4/80, or Gr-1. Leukocytes were defined by expression of the leukocyte common antigen CD45. Results are presented as a percentage of CD45+ cells, and data are pooled from three to four independent experiments per marker and represent means ± SEM. #Significant difference between infected compared with uninfected mice within a strain. For CD3+ cells, among B6 mice, P = 0.01 (Day 7), P < 0.0001 (Day 14), and P = 0.02 (Day 21); and among BALB/c mice, P = 0.04 (Day 7), P = 0.003 (Day 14), and P = 0.02 (Day 21). For Mac-1+ cells, among B6 mice, P = 0.01 (Day 14) and P = 0.006 (Day 21); and among BALB/c mice, P = 0.007 (Day 14) and P = 0.01 (Day 21). For Gr-1+ cells, among B6 mice, P = 0.004 (Day 7) and P = 0.009 (Day 14); and for BALB/c mice, P = 0.002 (Day 7), P = 0.03 (Day 14), and P = 0.02 (Day 21). *P = 0.004 for B6 compared with BALB/c mice. **P = 0.007 for B6 compared with BALB/c mice.

As described previously, there were significant increases in the total CD3⁺ T-cell population in both susceptible B6 and resistant BALB/c mice during lung P. aeruginosa infection, but there were no significant differences between the strains during infection. However, it is possible that there are differences in the proportions of subpopulations of CD4⁺ and CD8⁺ T cells. To address this issue, the ratio of CD4⁺ to CD8⁺ lung T cells was determined in control and infected B6 and BALB/c mice. There were no significant differences in the CD4/CD8 ratio in the lungs of normal mice versus mice receiving sterile beads of either strain (Table 3). Following intratracheal infection with P. aeruginosa, there was a significant increase in B6 mice in the CD4/CD8 ratio within the first 14 d after infection but not 21 d after infection in comparison with normal B6 mice. Among BALB/c mice, there was a significant increase in the CD4/CD8 ratio in infected mice within the first 14 d of infection as well as a highly significant increase by 21 d in comparison with normal mice. Furthermore, the difference in the ratios of CD4/CD8 T cells later in infection was significant between resistant BALB/c and susceptible B6 mice.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Although numerous studies have addressed the physiologic, genetic, and immunologic bases for the persistence of P. aeruginosa infection in the CF lung, few studies have addressed the basis of the initial vulnerability of the lung to the establishment of a chronic infection (34). Given that the CF genotype does not appear to predict the severity of pulmonary disease, it has been hypothesized that genes outside of the CFTR locus may influence the host response to bronchopulmonary P. aeruginosa infection (6- 8). Efforts to elucidate the mechanism of immunity to the initial encounter with chronic bronchopulmonary P. aeruginosa infection have been hampered by the lack of a suitable experimental model. Instillation via the trachea of bacteria trapped in agar beads has been accomplished in a number of animals, including rats and mice (37). As shown by Cash and colleagues (38), who used rats, the resulting bronchopulmonary infection is chronic, and the serologic and histopathologic changes closely resemble the later stage of the lung infection in CF patients. More recently, Starke and coworkers (15) established chronic bronchopulmonary P. aeruginosa infection in mice using outbred CD-1 animals and obtained similar findings. Using genetically well-defined, inbred mice, we observed that B6 mice had a high bacterial burden in the lungs on Day 14 after infection and concluded that this mouse strain is susceptible to chronic bronchopulmonary P. aeruginosa infection (17). The observation that BALB/c mice did not have any detectable bacteria in the lung at this time, on the other hand, led us to conclude that this host is resistant.

In our study, we confirmed and extended our previous observations by further characterizing the course of chronic bronchopulmonary P. aeruginosa infection in resistant BALB/c and susceptible B6 mice for 28 d after infection. Following intratracheal inoculation of 1 to 2 × 10⁵ CFU of mucoid bacteria trapped in agar beads, B6 mice were found to experience a significantly higher mortality than BALB/c mice. Moreover, physical examination of the mice revealed that the infection was more severe in B6 than in BALB/c mice based on several clinical parameters, including a significantly greater weight loss during the first 14 d after infection. Interestingly, BALB/c mice were observed to experience a higher and more severe incidence of fever than B6 mice. Although we did not quantify body temperatures in our study, this observation may relate to our previous observation that at Day 7 after infection alveolar macrophages, the predominant cell in the bronchoalveolar lavage fluid (BALF) of BALB/c mice at this time, produced significantly higher levels of the pyrogenic cytokine TNF- (39) in response to in vitro stimulation with heat-killed P. aeruginosa than did alveolar macrophages from B6 mice (our unpublished observations).

Consistent with our previous observations, B6 mice were observed to have significantly higher numbers of P. aeruginosa in the lung than did BALB/c mice through 3 wk after infection. In addition, 67% of BALB/c mice cleared the infection from the lung by Day 28. In contrast, only 22% of infected B6 mice cleared the infection by this time. We also observed that differences between B6 and BALB/c mice in the bacterial burden in the lung correlated with histopathologic differences in the type of inflammation and the extent of tissue damage. In accord with previous histologic findings in the lungs of rats and CD-1 mice infected intratracheally with mucoid strains of P. aeruginosa trapped in agar beads, marked bronchial and parenchymal changes were evident in the infected lungs of both resistant BALB/c and susceptible B6 mice (15, 38, 40, 41). As observed in these previous studies, acute inflammation consisting primarily of PMN surrounding microcolonies of P. aeruginosa was apparent in the airways of susceptible B6 mice. In contrast, little or no acute inflammation but rather a predominantly chronic, granulomatous inflammation was visible in the airways of resistant BALB/c mice. Furthermore, by Day 14 after infection, extensive tissue damage was apparent in the lungs of B6 hosts. At this time, little or no inflammation was apparent in the airways of BALB/c mice, granulomas were prominent in the interstitium, and tissue repair and control of infection were evident. Later in infection, at 5 wk, acute inflammation was still prominent in the airways of susceptible B6 mice and fibrosis was evident in the parenchyma of these hosts (data not shown). Again, in contrast to the lungs of B6 mice, low-grade, chronic inflammation characterized by the presence of lymphocytes, activated macrophages, giant plasma cells, and lymphoid follicles was prevalent in the lung tissue of resistant BALB/c mice.

Similar to morphologic changes observed in susceptible B6 mice, studies of pulmonary pathology in CF patients revealed that, in P. aeruginosa-associated lung disease, multiple airways are plugged with bacteria surrounded by a cellular exudate consisting almost entirely of PMN (42- 45). It has recently been appreciated that the acute, predominantly PMN inflammation in the CF lung leads to a vicious cycle of infection and inflammation resulting in severe and persistent Pseudomonas infection with loss of lung function (34). The neutrophilic inflammatory response not only fails to control and eliminate the infection, but products of this response, including chemokines, reactive oxygen intermediates, digestive enzymes, and proteases, appear to contribute to tissue damage as well as to a defect in local, host-immune responses that perpetuates the infection (34). Based on this, there is increasing interest in the use of anti-inflammatory drugs to target specifically the exuberant inflammatory response in the CF lung (34). Because of their recently recognized ability to produce immunoregulatory cytokines such as interleukin (IL)-10 and IL-12, PMN may also play an important role in the CF lung in the afferent arm of the immune response and thereby shape the nature of the subsequent acquired immunity (46).

It is of interest to point out that the observation of chronic, granulomatous inflammation in the airways of P. aeruginosa-infected, resistant BALB/c mice is similar to histopathology observed by Johansen and her colleagues (41) in the lungs of rats undergoing a chronic P. aeruginosa infection and treated with recombinant interferon (rIFN)-. On the basis of their observations, these investigators proposed that a Th1 response occurred in the lungs of IFN-- treated animals, whereas a Th2 response was responsible for the acute inflammation and tissue damage that occurred in the lungs of untreated animals. The balance between these two major Th subsets is known to play a critical role in infectious diseases in terms of protection versus immunopathology because of their functional diversity (50). Whereas Th1 cells secrete IL-2, IFN-, and TNF-, Th2 cells secrete IL-4, IL-5, IL-6, and IL-10. We tentatively reached a similar conclusion of a Th1 response in resistant BALB/c mice versus a Th2 response in susceptible B6 mice (15). This conclusion was based on our observations that BALB/c mice had high delayed-type hypersensitivity (DTH) reactions and low Pseudomonas-specific antibody responses during chronic bronchopulmonary P. aeruginosa infection, whereas B6 mice had low DTH reactions and high serum antibody levels, in particular IgG2b and IgM. High serum titers of P. aeruginosa-specific antibodies have been found to correlate positively with both the aggressive course of infection and poor clinical prognosis in CF patients (21). This observation suggests that not only is the inflammatory response excessive in CF but also that the immune response generated may be inappropriate. Indeed, studies by Matthews and colleagaues (51) showed that hypogammaglobulinemic CF patients had significantly less severe lung disease than age-matched CF patients with normal or elevated IgG levels. Studies are currently under way in our laboratory to determine the cytokine profile in the lungs of resistant BALB/c and susceptible B6 mice during chronic bronchopulmonary P. aeruginosa infection to establish the Th-cell phenotype associated with resistance in this model.

We also performed flow cytometric analyses to determine if differences in the proportions of phagocytes, B cells, or T cells in the lungs of P. aeruginosa-infected mice contribute to differences in the level of host resistance. Consistent with histologic evidence described here and elsewhere (15, 38) that instillation of sterile agar beads into the lung resulted in little or no local inflammation, no significant differences were observed in bead implant mice of either strain compared with normal animals in the numbers of lung cells and the proportions of leukocytes or their forward or side scatter. After P. aeruginosa infection, there were similar increases in the numbers of lung cells in both BALB/c and B6 mice on Day 7, but there were no significant differences between the strains in the numbers of lung cells at this time or any day through 28 d after infection. B6 mice were, however, found to have significantly greater proportions of PMN than did BALB/c mice on Days 7 and 14 after infection, an observation consistent with the histologic findings described previously of acute inflammation in the airways of B6 mice. The finding of a higher proportion of PMN in lungs of B6 mice is also consistent with our previous observations that PMN are the predominant cell type in BALF obtained from these hosts at Day 7 after bronchopulmonary P. aeruginosa infection (unpublished observation).

Despite the concept that antibody is important in protection against mucoid P. aeruginosa infection in the lung, there were no increases in the proportions of B cells in either host (36, 52, 53). We previously demonstrated that lung T cells from BALB/c mice exhibited significantly higher proliferative responses to both specific antigen and concanavalin A than cells from B6 mice through 3 wk after infection (17). In the present study, we observed that there were significant but similar increases in total CD3⁺ cells in the lungs of B6 and BALB/c mice during infection as well as significant increases in the ratios of CD4⁺/CD8⁺ T cells in both strains. However, the timing of the increases in the CD4/CD8 ratios differed between the strains. In B6 mice, a significant increase in the CD4/CD8 ratio compared with control animals occurred within the first 14 d of infection, whereas in BALB/c mice there were significant increases in the CD4/CD8 ratios during the entire course of the infection. Moreover, the difference in the CD4/CD8 ratios between resistant BALB/c compared with susceptible B6 mice was significant after 21 d when the majority of BALB/c but not B6 mice had cleared the infection. These observations suggest that T cells, in particular CD4⁺ T cells, play a protective role in immunity to chronic bronchopulmonary P. aeruginosa infection. More extensive studies, however, are required to address this issue.

Taken together, our findings presented here and elsewhere (17) demonstrate that resistant BALB/c mice, which have a chronic, granulomatous inflammation in the airways early in response to the initial encounter with P. aeruginosa, are able to control bacterial multiplication in the lung. The net result is that the appropriate acquired immune response, possibly a Th1 response, is generated and the bacterial burden in the lung is rapidly and significantly reduced before excessive tissue damage occurs. On the other hand, susceptible B6 mice respond early in infection to the initial encounter with this bacterium in the lung with an acute, predominantly neutrophilic, inflammatory response. We propose that in these hosts, an inappropriate specific immune response develops, resulting in a more severe course of infection with a high bacterial burden for a prolonged period and extensive tissue damage. The use of the model of chronic bronchopulmonary P. aeruginosa infection in resistant BALB/c and susceptible B6 mice should enable us to dissect the cellular events, in particular the roles of PMN, macrophages, and Th cell subsets, underlying these differences.