Lung-Specific Transgenic Expression of KC Enhances Resistance to Klebsiella pneumoniae in Mice¹

Despite the development of new broad spectrum antibiotics, the morbidity and mortality from pneumonia remains substantial (1). The clinical management of patients continues to be challenging, particularly in the face of multidrug-resistant Gram-negative bacteria such as Klebsiella pneumoniae and Pseudomonas aeruginosa (2, 3). Effective host defense against lung bacterial infection is primarily dependent upon the rapid clearance of the organism from the respiratory tract. This clearance is known to be mediated by neutrophils and macrophages (Mφ)³ that are vigorously recruited and/or activated at the site of infection (4). Leukocyte recruitment is an inherently integral component of the host defense cascade and is mediated in part by the production of chemotactic cytokines (5).

Four closely related polypeptide chemotactic cytokine or chemokine families, C-C, C-X-C, C, and C-X-X-X-C chemokines, are now known to exist in humans; most have murine homologues (6, 7, 8, 9). The C-C and C-X-C chemokines, in particular, have been recognized as crucial mediators in several inflammatory disease states (8, 9). More importantly, both families of chemokines have been shown to be critical participants in host defense against infection (10, 11, 12, 13). In Cryptococcus neoformans pneumonia, the neutralization of Mφ inflammatory protein (MIP)-1α and monocyte chemoattractant protein-1 (MCP-1/JE), which are members of the C-C chemokine family, resulted in a significant decrease in Mφ and lymphocyte influx and a consequential increase in fungal burden and mortality (11, 12). In a murine Klebsiella pneumonia model, the C-X-C chemokine MIP-2 has been shown to be an important contributor of neutrophil recruitment and bacterial clearance (13). Specifically, neutralization with anti-MIP-2 Abs significantly attenuated neutrophil influx, resulting in increased bacterial burden and early mortality (13).

KC is a member of the C-X-C chemokine family. Like other members of the family, KC is broadly produced by numerous cells, including fibroblasts, endothelial cells, and peritoneal and alveolar Mφ (14, 15, 16, 17). In vitro, these cells express KC in response to inflammatory signals, such as LPS and the host-derived cytokines IFN-γ and platelet-derived growth factor, suggesting a prominent role for KC in the inflammatory cascade (18, 19). Similar to other members of the C-X-C chemokine family, KC induces neutrophil chemotaxis and weakly stimulates respiratory burst to levels that are comparable with those of its human homologue, growth-regulated gene-α (20). In vivo, KC is expressed in the rat models of acute pulmonary inflammation that are induced by endotoxin, sulfur dioxide-mediated chronic bronchitis, and vanadium exposure (20, 21, 22, 23), while the inhibition of KC has been shown to substantially attenuate the accumulation of neutrophils in the lungs after the intratracheal (i.t.) administration of LPS (20). Furthermore, the intrapulmonary administration of KC induces a dose-dependent neutrophilic influx that is compartmentalized to the lungs (20). Finally, KC and other chemokines have been implicated in extrapulmonary inflammatory responses, including the thrombogenesis that occurs in injured endothelium (24).

Several KC-transgenic mouse models have been developed and have been shown to promote controlled neutrophilic recruitment in selectively targeted compartments (25, 26, 27, 28). The programmed expression of the KC transgene using a thymus-specific or a human keratin promoter resulted in neutrophilic aggregates in the thymic cortical and juxtamedullary regions or epidermis, respectively (25). Moreover, transgenic mice that were generated under control of a myelin basic protein promoter expressed the KC transgene only in the central nervous system, resulting in site-specific polymorphonuclear leukocyte (PMN) sequestration (26). More recently, Lira and colleagues have constitutively expressed KC within the lungs using the transgenic KC gene fused to a Clara cell-specific promoter (27). The increased KC expression resulted in enhanced neutrophil migration within the lungs of transgenic mice (27). The lung overexpression of KC did not result in a substantial activation of recruited neutrophils, as evidenced by the lack of tissue damage histologically. This transgenic model allowed us the unique opportunity to examine the role of KC in lung host defense against Gram-negative bacterial organisms.

In this study, we describe the time-dependent expression of KC protein within the lungs after i.t. K. pneumoniae administration and correlate the changes in KC expression with an influx of neutrophils. In addition, transgenic mice were employed to determine whether overexpression of the KC chemokine could have a beneficial effect on bacterial clearance and survival in the setting of Gram-negative bacterial pneumonia.

Specific pathogen-free transgenic KC mice, CC51^+/+ (2–3-mo-old males and females), on a B6D2 background were generated from a microinjection of fertilized eggs carrying the KC transgene in their genome. The expression of the KC transgene was localized to the lung by incorporating regulating elements from the CC10 gene, which encodes for a 10-kDa protein of unknown function that is produced by nonciliated bronchial epithelial cells (Clara cells) (25, 26, 27). Age- and gender-matched wild-type (wt) B6D2 mice of the same species as the transgenic mice served as control animals. All mice were housed under specific pathogen-free conditions within the animal care facility at the University of Michigan until the day of sacrifice.

We chose to use K. pneumoniae strain 43816, serotype 2 (American Type Culture Collection, Manassas, VA) in our studies, as this strain has been shown to induce an impressive inflammatory response in mice (13, 28, 29). K. pneumoniae was grown in tryptic soy broth (Difco, Detroit, MI) for 18 h at 37°C. The concentration of bacteria in the broth was determined by measuring the amount of absorbance at 600 nm. A standard of absorbancies based on known CFU was used to calculate the inoculum concentration. A dose of 3 × 10³ organisms per transgenic or wt mouse was chosen, since this dose allowed for the development of substantial inflammation by 36 to 48 h without excessive mortality at that timepoint. Animals were anesthetized i.p. with ∼1.8 to 2 mg of pentobarbital per animal. The trachea was exposed, and 30 μl of inoculum or saline was administered via a sterile 26-gauge needle. The skin incision was closed with surgical staples.

Bronchoalveolar lavage was performed to obtain BAL cells. The trachea was exposed and intubated using a 1.7-mm OD polyethylene catheter. BAL was performed by instilling PBS containing 5 mM EDTA in 1-ml aliquots. Approximately 5 ml of lavage fluid was retrieved per mouse. Cytospins were subsequently prepared from BAL cells and stained with Diff-Quick (Baxter, McGaw Park, IL); next, differential counts were determined.

Lung MPO activity (as a measure of neutrophil quantity) was determined using a method that has been described previously (13). Briefly, lungs were homogenized in 2 ml of a solution containing 50 mM potassium phosphate (pH 6.0) with 5% hexadecyltrimethylammonium bromide and 5 mM EDTA. The resultant homogenate was sonicated and centrifuged at 12,000 × g for 15 min. The supernatant was then mixed with assay buffer to a 1:15 ratio and read at 490 nm. MPO units were calculated as the change in absorbance over time.

Mice were sacrificed by carbon dioxide inhalation. The pulmonary vasculature was perfused with 4% paraformaldehyde in PBS via the right ventricle. Lungs were then excised en bloc and inflation-fixed in 4% paraformaldehyde in PBS. The lungs were then embedded in paraffin, and sections were cut and stained with hematoxylin and eosin using standard techniques.

Mice were sacrificed by carbon dioxide inhalation at designated timepoints, and blood was collected by orbital bleeding or direct cardiac puncture. Whole lungs were then harvested for assessment of the various cytokine protein levels. Before lung removal, the pulmonary vasculature was perfused via the right ventricle with 1 ml of PBS containing 5 mM EDTA. After removal, whole lungs were homogenized in 1.5 ml of complete protease inhibitor lysis buffer (Boehringer Mannheim, Indianapolis, IN). Homogenates were incubated on ice for 30 min and then centrifuged at 2500 rpm for 10 min. Supernatants were collected, passed through a 0.45-μ filter (Gelman Sciences, Ann Arbor, MI), and then stored at −20°C for an assessment of cytokine levels.

At the time of sacrifice, plasma was collected, the right ventricle was perfused with 1 ml PBS, and then the lungs were removed aseptically and placed in 3 ml of sterile saline. The tissues were subsequently homogenized with a tissue homogenizer under a vented hood. The lung homogenates were placed on ice, and serial 1/10 dilutions were made. A total of 10 μl of each dilution was plated on soy-based blood agar plates (Difco) and incubated for 18 h at 37°C; colonies were counted after the incubation.

Murine TNF-α, IFN-γ, IL-12, IL-10, MIP-2, MCP-1/JE, and KC were quantitated using a modification of a double ligand method as described previously (13). Briefly, flat-bottom, 96-well microtiter plates (Immuno-Plate I 96-F, Nunc, Fisher Scientific, Itasca, IL) were coated with 50 μl/well of rabbit Ab against the various cytokines (1 μg/ml in 0.6 M NaCl, 0.26 M H₃BO₄, and 0.08 M NaOH, pH 9.6) for 16 h at 4°C and then washed with PBS (pH 7.5) plus 0.05% Tween 20 (wash buffer). Microtiter plate nonspecific binding sites were blocked with 2% BSA in PBS and incubated for 90 min at 37°C. The plates were rinsed four times with wash buffer, and diluted (neat and 1/:10) cell-free supernatants (50 μl) in duplicate were added; next, the plates were incubated for 1 h at 37°C. The plates were washed four times, 50 μl/well of biotinylated rabbit Abs against the specific cytokines was added (3.5 μg/ml in PBS (pH 7.5) 0.05% Tween 20, and 2% FCS), and the plates were incubated for 30 min at 37°C. Next, the plates were washed four times, streptavidin-peroxidase conjugate (Bio-Rad, Richmond, CA) was added, and the plates were incubated for 30 min at 37°C. The plates were washed four additional times, and chromogen substrate (Bio-Rad) was added. We incubated the plates at room temperature to the desired extinction, and the reaction was terminated with 50 μl/well of 3 M H₂SO₄ solution. The plates were read at 490 nm in an ELISA reader. Standards were one-half log dilutions of recombinant murine cytokines from 1 pg/ml to 100 ng/ml. This ELISA method consistently detected murine cytokine concentrations that were >25 pg/ml. The ELISA did not cross-react with IL-1, IL-2, IL-4, or IL-6. In addition, the ELISA did not cross-react with other members of the murine chemokine family, including murine RANTES or epithelial neutrophil-activating protein-78.

Data were analyzed by a Macintosh computer using the Statview II statistical package (Abacus Concepts, Berkeley, CA). The survival data were compared using by χ² analysis. All other data were expressed as the mean ± SEM and compared using a two-tailed Student’s t test. Data were considered statistically significant if p values were <0.05.

In uninfected mice at baseline, there was a significant quantity of KC detected within the lungs of KC-transgenic mice, whereas no KC was detected in the lungs of wt control animals (Fig. 1, p < 0.001). The i.t. administration of K. pneumoniae (3 × 10³ CFU) resulted in a time-dependent production of KC in the lungs of both KC-transgenic and wt mice as determined by a specific ELISA. Maximal lung KC levels peaked at 48 h following bacterial inoculation; the levels of KC were significantly greater in transgenic mice compared with the levels observed in wt controls (p = 0.01). In both KC-transgenic mice and wt control mice, the levels of KC remained elevated at 6 days postinoculation compared with noninfected controls. No increase in KC was observed in the blood of either infected KC-transgenic or wt mice compared with noninfected counterparts (data not shown).

Given the potent neutrophil chemotactic effects of KC, we next sought to determine whether compartmentalized overexpression of KC within the lung would provide protective effects in animals challenged with intrapulmonary K. pneumoniae. For these studies, KC-transgenic and wt mice were administered K. pneumoniae i.t. at 3 × 10³ CFU, which us an approximately LD_80–90 dose in wt mice. At 48 h after K. pneumoniae inoculation, both transgenic and wt animals developed signs of pneumonia, including lethargy and ruffled fur. However, survival was significantly greater in KC-transgenic mice that had been challenged with K. pneumoniae, with 80% long-term survival observed in this group compared with only 10% survival in the wt infected control mice (Fig. 2).

To determine whether the observed improvement in survival in KC-transgenic mice was due to an enhancement in bacterial clearance, transgenic mice and wt control mice were administered K. pneumoniae (3 × 10³ CFU) i.t., and lungs and plasma were harvested at 24 and 48 h postinoculation. At 24 h, wt animals had an approximately eightfold greater number of K. pneumoniae CFU in the lung compared with KC-transgenic animals that had been challenged with K. pneumoniae (data not shown). In addition, 66% of wt animals were bacteremic by 24 h, whereas no mice overexpressing KC were bacteremic at this timepoint after K. pneumoniae administration. By 48 h, a more than fourfold increase in K. pneumoniae CFU in blood was noted in wt control mice compared with transgenic KC mice (p < 0.05). An even more striking 13-fold increase in K. pneumoniae CFU was noted in the lungs (p < 0.005) of wt animals compared with mice overexpressing KC at the latter timepoint (Table I). These results indicate that KC overexpression significantly augmented effective bacterial clearance in the lung and limited the dissemination of the organism to the bloodstream.

To determine whether the improvement in bacterial clearance in transgenic KC mice was due to alterations in the recruitment of inflammatory cells to the airspace, KC-transgenic animals or wt controls were challenged with K. pneumoniae, and BAL was performed at 48 h postinoculation. This timepoint was chosen because a maximum influx of neutrophils in response to the i.t. administration of K. pneumoniae occurs at 48 to 72 h postchallenge (11, 17). In uninfected mice, no differences in the percent or absolute number of total BAL cells, alveolar Mφ, or lymphocytes was noted in KC-transgenic mice compared with control animals. However, a significant increase in the percent neutrophils (12.2 ± 4.1% KC-transgenic vs 0.17 ± 0.17% wt; p < 0.05) and total neutrophils (10.8 ± 3.1 × 10⁴ KC-transgenic vs 0.67 ± 0.67 × 10⁴ wt; p < 0.05) was observed in KC-transgenic mice compared with wt mice. The intrapulmonary administration of K. pneumoniae in transgenic mice resulted in a marked increase in the total number of cells; this increase was largely due to a substantial increase in BAL fluid neutrophils (p < 0.05) (Fig. 3). Challenge with K. pneumoniae did not significantly alter the numbers of BAL fluid Mφ or lymphocytes in either the transgenic or wt mice. To confirm the BAL findings, lungs were harvested at 48 h postinoculation with K. pneumoniae or saline and assayed for lung MPO activity. In animals that were challenged with K. pneumoniae, there was a significant increase in lung MPO activity in the KC-transgenic mice compared with wt controls (data not shown). These results indicate that the augmented host response observed in KC-transgenic mice is associated with an enhanced influx of neutrophils into the airspace.

The transgenic expression of KC using a Clara cell-specific promoter has been shown to result in a predominant expression of KC within the terminal bronchioles, with lesser quantities observed in larger airways, including bronchi (25). Given this anatomic distribution of KC transgene expression, we subsequently wanted to determine whether differences in the distribution of the inflammatory response to Klebsiella challenge were observed in KC-transgenic mice compared with wt control animals. Little to no pulmonary inflammation was observed in either KC or wt mice at 48 h or at 7 or 14 days after the i.t. administration of saline (data not shown). At 48 h following the administration of 3 × 10³ CFU of K. pneumoniae, the lungs of infected wt mice showed focal consolidation with an infiltration of moderate numbers of neutrophils into the airspaces; some airspaces contained large numbers of extracellular K. pneumoniae (Fig. 4,A). In contrast, the lungs of KC mice contained focal areas of dense airspace neutrophil accumulation and no visible extracellular bacteria. In addition, neutrophil aggregates were observed in the terminal airways that were contained and adjacent to areas of alveolar consolidation (Fig. 4 B). By 7 days after Klebsiella administration, a partial resolution of the alveolar inflammatory cell infiltrates was observed in transgenic mice, although persistent neutrophil aggregates within distal airways were also observed (data not shown).

Subsequent experiments were performed to determine whether the compartmentalized overexpression of KC resulted in an altered production of important pro- and antiinflammatory cytokines. Several cytokines have been shown to be crucial to antibacterial host defense in Gram-negative infection, particularly TNF-α, T1 phenotype cytokines (IFN-γ and IL-12), the T2 phenotype cytokine IL-10, and members of the C-X-C and the C-C chemokine families (MIP-2 and MCP-1/JE, respectively). The baseline levels of TNF-α, IFN-γ, IL-12, IL-10, MIP-2, and MCP-1/JE were low or undetectable in the uninfected lungs of KC-transgenic animals, and no differences were seen in these levels compared with wt animals. The i.t. administration of K. pneumoniae resulted in a maximal expression of cytokines at 48 h in both KC-transgenic and wt mice, representing at least a fourfold increase in the protein levels of all cytokines assessed (TNF-α, IFN-γ, IL-12, IL-10, MCP-1/JE, and MIP-2), when compared with the expression seen in lung homogenates prepared from saline-treated control animals. There were no differences in the protein levels of TNF-α, IFN-γ, IL-12, IL-10, or MCP-1/JE in transgenic infected mice compared with wt infected mice (data not shown). However, the expression of MIP-2 in wt mice was delayed compared with the expression of KC; maximal lung MIP-2 levels were noted at 72 h after K. pneumoniae challenge, whereas maximal KC levels were noted at 24 h postchallenge and decreased toward baseline thereafter (Fig. 5). Furthermore, maximal MIP-2 levels tended to be lower or were significantly decreased in KC-transgenic mice at 48 and 72 h after bacterial challenge, respectively, compared with wt infected mice (p = 0.08 and 0.01, respectively). These results suggest that KC does not appear to regulate the expression of the other relevant cytokines involved in host defense against Klebsiella pneumonia. However, the overabundance of KC in transgenic mice was associated with a reduction in the expression of another C-X-C chemokine, MIP-2.

Effective host defense against bacterial pathogens requires the generation of a vigorous inflammatory response involving the recruitment and activation of neutrophils (2). MIP-2, a member of the C-X-C chemokine family has been recognized as a contributor to antibacterial host defense by effecting neutrophil trafficking and activation (13). In this study, we examined the contribution of KC, another member of the C-X-C chemokine family, to neutrophil recruitment and bacterial clearance in a murine model of Klebsiella pneumonia.

In this study, we used transgenic mice in which the KC gene was fused to a Clara cell-specific promoter (26, 27), which resulted in an enhanced constitutive and time-dependent expression of KC within the lungs during the course of Klebsiella pneumonia. This transgenic model allowed us to modulate a single chemokine, KC, in our in vivo bacterial pneumonia model to examine whether the overexpression of KC in these transgenic mice would affect overall antibacterial host defense. The enhancement of KC significantly improved survival when compared with wt control mice and was a result of increased bacterial clearance. The mechanism for enhanced bacterial clearance is likely attributable to significantly augmented neutrophil recruitment, as we observed a greater magnitude of neutrophil influx into the airspaces of KC-transgenic mice after K. pneumoniae challenge compared with wt controls. Given that the predominant site of KC transgene expression in transgenic mice is the terminal bronchioles (26, 27), it appears that the enhanced alveolar PMN influx results from the spillover of KC into the airspace. We also observed neutrophil aggregates in the terminal airways in proximity to the sites of airspace inflammation; this finding was distinctly unique to KC-transgenic animals and probably reflects site-directed PMN migration as a result of KC production from the airway epithelium. We cannot exclude a concomitant effect on neutrophil activation, as KC and other C-X-C chemokines have been shown to activate PMN effector cell activities, including respiratory burst and antimicrobial activity in vitro (8, 9, 20). There is also a modest but significant BAL neutrophilia in KC-transgenic animals at baseline, and it is plausible the these neutrophils facilitate the initial clearance of K. pneumoniae after intrapulmonary challenge. Although alveolar Mφ and other Mφ populations have been shown to play a critical role in the clearance of Gram-negative organisms from the lung (30), there is no evidence to suggest that KC or other C-X-C chemokines have direct stimulatory effects on Mφ antimicrobial activity.

While the enhanced production of KC in KC-transgenic mice provided protective effects in murine Klebsiella pneumonia, we also observed a substantial early induction of KC in wt mice in response to the intrapulmonary delivery of K. pneumoniae. Indeed, the marked beneficial effect of compartmentalized KC overexpression taken together with the fact that KC is maximally expressed early (at 24 h) in the host response suggests that KC may play a more important role in antibacterial host defense than MIP-2, which is expressed in a delayed fashion (maximal at 72 h in wt mice); the neutralization of MIP-2 during the evolution of Klebsiella pneumonia resulted in relatively modest although significant reductions in bacterial clearance and survival (13). However, in vivo neutralization studies have not been possible due to the lack of an appropriate anti-murine KC-neutralizing Ab and of KC-deficient mice.

Unlike C-C chemokines (31, 32), there is no evidence to support a role of C-X-C chemokines in directly regulating the expression of other cytokines. Similarly, the compartmentalized overexpression of KC in infected mice did not alter the expression of the cytokines TNF-α, IFN-γ, IL-12, and IL-10 compared with wt animals that had been challenged with K. pneumoniae, suggesting that KC likely occupies a terminal and distal position in the overall cytokine network involved in the host defense cascade. However, we did observe a significant decrease in the inducible, but not the constitutive, expression of MIP-2 in KC transgenic mice that were administered K. pneumoniae compared with wt animals. This observation raises the possibility that a substantial redundancy in chemokine responses exists, such that the overexpression of a specific chemokine may obviate the need for (or directly inhibit) the production of another family member with similar biologic activities. The attenuated expression of one chemokine by the overexpression of other functionally and structurally related chemokines has not been reported previously.

The compartmentalized augmentation of KC within the lung improved the innate host defense system against K. pneumoniae in our murine model of bacterial pneumonia. Given the emergence of highly resistant bacterial pathogens and the increasing population of immunocompromised hosts (1, 2, 3, 33), the treatment of bacterial infection has and will continue to be quite difficult. Therefore, the immunomodulation of host responses, especially if delivered in a site-directed fashion, may prove to be an attractive adjuvant to conventional antibiotic therapy. Because C-X-C chemokines appear to occupy a distal position in the cytokine network and exhibit more neutrophil chemotactic than activating effects, these favorable characteristics suggest that the augmentation of chemokines such as KC may potentially improve the clinical outcome of patients with severe bacterial pneumonias while limiting the potentially serious adverse effects that are mediated by overzealous leukocyte activation and/or the regulation of other potentially injurious cytokines.