INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The major cause of death in patients with cystic fibrosis (CF) is respiratory insufficiency resulting from extensive bronchiectasis. The basic defect leading to this condition is a dysfunctional CF transmembrane conductance regulator (CFTR) protein at the apical surface of airway epithelial cells (1). CFTR is essential to maintain a normal ionic milieu in airway surface fluid. Abnormal ionic composition of the airway surface fluid may lead to one or more of several defects. Viscous mucous plugs may result from defective serous cell fluid secretion (2, 3). Cationic peptide antibacterial host defenses can be inactivated by high salt concentrations (4, 5). In addition, CFTR dysfunction is associated with an excessive inflammatory response to Pseudomonas aeruginosa gene products (6). Regardless of the mechanism, it is clear that the hallmark of CF lung disease is a chronic infection of the tracheobronchial mucosa in which P. aeruginosa is the most common pathogen. This infection is accompanied by airway neutrophilic inflammation, which is observed within the first few months of life, often before any evidence of airway bacteria (7). The inflammation is so severe that active elastase overwhelms the lung's natural antiprotease defenses and proteolytic activity is detectable in the bronchoalveolar lavage fluid of patients with CF at a very early stage (7).

Human neutrophil elastase (NE) is a serine protease of broad substrate specificity with the potential to initiate and sustain several key pathogenic processes in CF lung disease. In addition to inducing copious mucus discharge from airway submucosal glands, NE destroys connective tissues, thus increasing airway compliance (10, 11). These effects contribute directly to the obstruction of expiratory airflow by decreasing lung elastic recoil and increasing airway resistance. However, not only can NE contribute to airflow obstruction, but it can also enhance airway inflammation and inactivate essential host defense mechanisms. NE increases the transcription of the interleukin-8 (IL-8) gene (12). IL-8 is a potent neutrophil chemoattractant, and it is thought to play a major role in the recruitment of neutrophils to the CF airways (13). Furthermore, NE creates a complement/complement receptor mismatch, rendering neutrophil phagocytosis of P. aeruginosa ineffective (14, 15).

Although abundant alpha₁-protease inhibitor (₁PI) is present in CF blood and airway secretions, it is proteolytically inactivated by an excess of NE at the airway mucosal surface (16). Previous investigators have demonstrated that aerosolized ₁PI inhibits free NE in the airway secretions of patients with CF (19). Inhibition of NE in patients with CF has been associated with a decrease in airway neutrophils and IL-8, suggesting that this may be a beneficial therapeutic approach (20). However, it has also recently been shown that NE inhibition may actually decrease host defenses against gram negative bacteria. Transgenic mice deficient in NE were shown to have an increased susceptibility to sepsis and death after intraperitoneal challenge with gram negative bacteria (21). The effect of NE inhibition on bacterial growth in vivo in a lung chronically infected with P. aeruginosa remains unknown. The present study was therefore designed to test the effect of airway neutrophil elastase inhibition using aerosolized Prolastin, a preparation of ₁PI purified from human blood, on the bacterial load of the lung chronically infected with P. aeruginosa. To do this, we utilized a well characterized rat model of chronic P. aeruginosa lung infection.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In Vitro Rat NE-inhibitory Activity of Prolastin

In vitro neutrophil elastase inhibitory activity of Prolastin was measured as previously described (22). Briefly, rat neutrophils were recovered 18 h after intraperitoneal injection of 16 ml of a 12% (wt/vol) suspension of sodium caseinate. Elastase was extracted from rat neutrophil azurophilic granules as previously described (22). The molarity of active rat neutrophil elastase was determined by titrating the preparation against the specific chromogenic substrate methoxy-succinyl-alanyl-alanyl-prolyl-valyl-p-nitroanilide (MeOSAAPVNA). Using pure human neutrophil elastase (EPC, Pacific, MO), we had previously determined that 20 nM neutrophil elastase yields one unit of activity defined as a _Abs410 of 0.025/min. The neutrophil elastase inhibitory capacity of Prolastin was measured by adding 200 µl of 20 nM rat or human neutrophil elastase in 100 mM HEPES, 0.5 M NaCl, 0.1% Brij-35 at pH 7.5 to 200 µl of 0 to 40 nM Prolastin (Bayer Inc., Etobicoke, ON, Canada) at 23° C for 1 h. At the end of each incubation period, 100 µl 1 mM MeOSAAPVNA was added to the reaction mixture, and residual elastase activity was measured in a spectrophotometer (DU-7; Beckman Instruments, Inc., Fullerton, CA) as the change in absorbance at 410 nm/min.

Chronic Rat Lung Infection

The chronic rat lung model of P. aeruginosa infection as originally described by Cash and colleagues (23) was used to assess the effects of aerosolized Prolastin on the progression of chronic P. aeruginosa lung infection in rats. The end points were the examination of bronchoalveolar lavage fluid taken from animals at death for total and neutrophil cell counts and elastase levels after aerosol treatment with Prolastin. The lungs of animals were also examined for total bacterial counts and histopathologic changes after aerosol treatment with Prolastin.

We utilized a state-of-the-art aerosol delivery chamber in the present studies (Figure 1). The aerosol chamber was housed in a Biosafety Cabinet in a Level 2 containment facility with the airflow isolated for additional safety. The chamber was designed to accept animals ranging in size from 20 to 250 g. The aerosol chamber was equipped with sampling ports for continuously monitoring the size of the particles. The aerosol was removed through HEPA-filtered adjustable vacuum flow, which allowed us to control the flow rates. The aerosol flow rates were calibrated using fluorescent microspheres obtained from Duke Scientific Corporation, and this was done prior to each aerosol treatment.

View larger version (57K): [in this window] [in a new window]   Figure 1.   Aerosol delivery system. The nose-only aerosol chamber was housed in a biosafety cabinet where 20 rats were exposed to aerosol preparations from an Aero-Tech II nebulizer operated at 45 psi. The aerosol was removed through a HEPA filtered vacuum adjusted to a flow rate of 10 L/min. A 10-ml volume of aerosol preparation was dispensed in 25 to 30 min.

Fifty-four animals were utilized in these experiments; 18 for histopathology studies, 18 for quantitative bacterial culture, and 18 for bronchoalveolar lavage fluid neutrophil counts and elastase activity determinations. All animals were inoculated with approximately 10⁶ P. aeruginosa strain PAO in agar beads. Two days after inoculation, rats (treatment group, n = 27) were exposed to aerosol preparations from an Aero-Tech II nebulizer (CIS-US, Bedford, MA). The nebulizer was operated at 45 psi, with a flow rate of 10 L/min and contained 10 ml of the preparation (Prolastin 5 mg/ml) to be aerosolized. The 10-ml volume was dispensed in 25 to 30 min. Animals were treated daily for 7 d; control animals (n = 27) received daily exposure to normal saline. Animals were killed on Days 0 (6 h after Prolastin treatment), 3, and 7. Three animals from treatment and control groups were subjected to bronchoalveolar lavage consisting of 10 ml normal saline at 37° C on Days 0, 3, and 7. Lavage fluids were examined for total cell counts by hemacytometer, neutrophil counts by differential counts after Wright's staining, and total elastase in lavage fluid was measured using the synthetic chromogenic elastase substrate MeOSAAPVNA.

On Days 0, 3, and 7, the left lungs of the treatment and control animals were removed and processed for histopathologic examination. Mounted sections were stained for light microscopy with hematoxylin-eosin. Infiltration of the lung with inflammatory cells and exudate was measured by a point-counting method as previously described (24). The number of points over the entire surface of the inflammatory infiltrate was counted with a Zeiss integrating eyepiece (Zeiss, Oberkochen, Germany) and divided by the total number of points over the entire surface area of the left lower lung lobe to obtain a measure of the percentage infiltration. This procedure was repeated with three left lobe slices from three animals in each group.

Effects of Prolastin on P. aeruginosa Growth

The PAO strain of P. aeruginosa was used to test whether Prolastin was directly bactericidal. Triplicate P. aeruginosa cultures (initial inoculums approximately 4 × 10⁶ CFU/ml) were grown in 10-ml aliquots of trypticase soy broth for 24 h at 37° C in the presence or absence of either 100 µg/ml bovine serum albumin or 100 µg/ml Prolastin. Colony-forming units were determined at time 0, 4, 8, and 24 h after inoculation of the cultures.

Statistical Analysis

All data are expressed as the mean ± standard error of the mean. The data were analyzed using Student's t test, and a p value less than 0.05 was considered significant (25).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

SDS-PAGE and Western Analysis of Prolastin

SDS-PAGE confirmed the presence of low levels of contaminating proteins in the Prolastin preparation. Specifically, very low levels of contaminating immunoglobulins were detected by Western blot; however, these contaminating immunoglobulins did not react with P. aeruginosa whole-cell preparations by Western blot (data not shown).

Rat Neutrophil Elastase Inhibition by Prolastin

Prolastin completely inhibited rat neutrophil elastase activity in vitro at a molar ratio of 1:1 (Figure 2). The inhibition of rat NE by Prolastin was similar to the profile obtained against human NE (p > 0.05 between human and rat NE inhibition by Prolastin).

View larger version (15K): [in this window] [in a new window]   Figure 2.   Human neutrophil elastase inhibitory capacity of Prolastin in vitro as measured against either 20 nM human neutrophil elastase (open circles) or 20 nM rat neutrophil elastase (closed circles). Neutrophil elastase was incubated with 0 to 20 nM Prolastin in vitro at 23° C for 1 h. Residual elastase activity was measured using the chromogenic substrate, MeOSAAPVpNA (n = 2 per data point, one representative experiment of three).

Elastase Activity in BAL of Chronic Rat P. aeruginosa Model

Active neutrophil elastase was detectable in the BAL fluid of all animals during the entire study period. However, 6 h after Prolastin aerolsolization and on Days 3 and 7, elastase activity was clearly lower in the treated groups than in the control groups (results expressed in nM, Day 0 = 75 ± 4 versus 54 ± 3; Day 3 = 66 ± 3 versus 38 ± 7; Day 7 = 77 ± 4 versus 22 ± 3; p < 0.01, all comparisons) (Figure 3).

View larger version (14K): [in this window] [in a new window]   Figure 3.   Neutrophil elastase activity in the bronchoalveolar lavage fluid of control (closed squares) and Prolastin-treated (open squares) rats chronically infected with Pseudomonas aeruginosa was measured using the chromogenic substrate MeOSAAPVpNA (control: n = 9, three per time point; treated group: n = 9, three per time point, *p < 0.01).

Quantitative Histopathology of Lungs

The P. aeruginosa imbedded agar bead instillation of the rat lungs lead to an inflammatory infiltration characterized by neutrophil and mononuclear cell infiltration in the airway lumen and walls, as well as visible epithelial desquamation. The treatment with aerosolized Prolastin lead to a progressive decrease in the amount of inflammatory infiltration expressed as a percentage of lung area (Day 7: control = 65 ± 21% versus ₁PI = 11 ± 4%; n = 3 at each time point; p < 0.01) (Figure 4).

View larger version (14K): [in this window] [in a new window]   Figure 4.   Quantitative histopathology of control (closed squares) and Prolastin-treated (open squares) rats chronically infected with Pseudomonas aeruginosa. The qualitative changes in pathology in the lungs of experimental animals from all groups were not significantly different. The abcissa represents the number of days after Prolastin or saline aerosol treatment, with data for Day 0 obtained 6 h after treatment. Abscissa and numbers of animals are as described in legend of Figure 3 (*p < 0.01).

BAL Total Cell and Neutrophil Counts

The total cell counts were not significantly different between the control and the treated groups at any of the time points (Figure 5A). In contrast, the treated group showed a decrease in neutrophil numbers as early as 6 h after Prolastin treatment (control = 64 ± 5% versus ₁PI = 48 ± 4%), and the suppression of BAL neutrophils was markedly accentuated at Days 3 (56 ± 7% versus 13 ± 4%) and 7 (68 ± 9% versus 2 ± 2%; p < 0.01, all comparisons) (Figure 5B).

View larger version (16K): [in this window] [in a new window]   Figure 5.   Bronchoalveolar lavage fluid (A) total cell counts and (B) BAL neutrophils expressed as a percentage of total BALF cells of control (closed squares) and Prolastin-treated (open squares) rats chronically infected with Pseudomonas aeruginosa. Abscissa and numbers of animals are as described in legend of Figure 3 (*p < 0.01).

Total Lung Bacterial Counts

Quantitative bacterial counts indicated detectable bacteria in all animals on Day 0 without any differences in the treatment group. In contrast, at Day 7 the control group demonstrated a marked rise in the bacterial counts, whereas bacteria were near the lower limit of detection in the animals treated with aerosolized Prolastin (control: 85 ± 21 versus 0.2 ± 0.4 CFU × 10⁵; p < 0.01) (Figure 6).

View larger version (11K): [in this window] [in a new window]   Figure 6.   Total bacterial counts in the lungs obtained from control (closed squares) and Prolastin-treated (open squares) rats chronically infected with Pseudomonas aeruginosa. Abscissa and numbers of animals are as described in legend of Figure 3 (*p < 0.01).

Effect of Prolastin on Bacterial Growth In Vitro

The growth rate of P. aeruginosa was not affected by incubation with as much as 100 µg/ml Prolastin over 24 h (Figure 7) (p > 0.05 at all time points compared with control).

View larger version (12K): [in this window] [in a new window]   Figure 7.   P. aeruginosa growth in the presence of either 100 µg/ml albumin (closed circles) or 100 µg/ml Prolastin (closed circles) in trypticase soy broth over 24 h at 37° C. Each data point represents triplicate measures.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The current study provides evidence indicating that not only does aersolized Prolastin decrease lung inflammation but also accelerates bacterial clearance from the airways during chronic P. aeruginosa lung infection. The effect of aerosolized Prolastin was not due to any intrinsic bactericidal activity since Prolastin alone had no effect on bacterial growth in vitro. These observations have important therapeutic implications since an intervention that can help control lung inflammation and infection without the use of antibiotics may help avoid the inevitable selection of resistant organisms that is associated with repeated use of antibiotics in patients with chronic lung infections such as cystic fibrosis. In addition, the in vivo protection observed with doses administered at 24-h intervals suggests that the biologic half-life of Prolastin at the lung's surface is sufficient to render such therapy feasible without resorting to continuous aerosol delivery.

The chronic rat lung model of P. aeruginosa infection originally described by Cash and colleagues (23) is one of the most widely used animal models to study the pathogenesis of chronic P. aeruginosa lung infections. The model involves the incorporation of P. aeruginosa into agar beads followed by intratracheal deposition into the lungs of rats. A chronic infectious state ensues, and animals have been studied for as long as 1 yr after inoculation and found to be chronically infected. A number of clinical correlates to cystic fibrosis have been demonstrated in the model including: (1) similar, if not identical, lung pathology; (2) development of immune complex disorders; (3) conversion to the mucoid phenotype by P. aeruginosa strains colonizing the airways (26). In addition, this model has served in the past to correctly predict the effects of other drugs in patients with CF such as subinhibitory doses of antibiotics (29). This model is therefore a useful tool for the study of chronic airway infections such as that encountered in patients with CF. In the current study, we clearly observed a marked decrease in the characteristic lung histopathologic changes after daily treatment with aerosolized Prolastin.

Prolastin is a partially purified preparation of ₁PI. The aerosol preparation for control animals did not contain the molecules other than ₁PI that are present in Prolastin (i.e., formulation products and contaminating proteins). It is therefore conceivable that the mechanism by which aerosolized Prolastin accelerates bacterial clearance from the lung may be related to moleules other than ₁PI, particularly contaminating proteins. Our work would suggest that this is not the case. First, Prolastin was unable to kill bacteria in vitro, indicating that it did not have any intrinsic bactericidal activity. Second, although we could detect very low levels of contaminating immunoglobulins by Western blot in the Prolastin preparation, these contaminating immunoglobulins did not react with P. aeruginosa whole-cell preparations by Western blot (data not shown). These results would indicate that enhanced opsonization by contaminating immunoglobulins in the Prolastin preparation is not the mechanism by which the aerosol treatment caused the decrease in the lung bacterial burden.

We submit that the likely mechanisms by which Prolastin prevents inflammation and accelerates bacterial clearance are related to inhibition of the known noxious effects of free elastase in the lung. Because IL-8 is a major neutrophil chemoattractant and its levels correlate tightly with neutrophil numbers in the CF lung, prevention of IL-8 gene transcription by Prolastin-mediated elastase inhibition may play an important role in the marked suppression of lung neutrophils in the treated group. Neutrophil elastase inhibition by aerosolized secretory leukoprotease inhibitor has been shown to markedly decrease not only IL-8 concentrations but also neutrophil numbers in the BAL fluid of patients with CF (20). In the current study, the marked decrease in the airway bacterial load caused by Prolastin is a further mechanism that likely contributed to decrease lung inflammation.

Evidence from previous studies would suggest that the mechanism by which bacteria were more rapidly cleared in the treated group may be related to improved opsonin-dependent phagocytosis. Tosi and colleagues (15) have observed that neutrophil elastase can create a functionally important opsonin receptor mismatch by cleaving both C3bi on opsonized P. aeruginosa, and the complement receptor CR1 on neutrophils. Inhibition of neutrophil elastase activity in patients with CF by aerosolized ₁PI has been shown to partially restore the in vitro capacity of airway BAL neutrophils to phagocytose and kill P. aeruginosa bacteria (19). However, until now it was unknown whether aerosolized ₁PI could decrease the lung bacterial burden in vivo. Recently, it has been shown that mice lacking NE demonstrate increased mortality in the presence of gram negative bacterial sepsis, thus indicating that elastase suppression has the potential to markedly hinder host defenses against gram negative bacteria. The current study shows that neutrophil elastase inhibition with the extracellular inhibitor ₁PI not only is not harmful to host defenses in the lung but actually helps decrease the gram negative bacterial load.

One of the unexpected results of this study is the evidence of increased bacterial clearance from the lungs despite the persistence of detectable levels of active neutrophil elastase in the BAL fluids. Tosi and colleagues (15) have demonstrated that very low levels of active neutrophil elastase are sufficient to cleave C1 and IgG and to induce IL-8 release. This raises the possibility that mechanisms other than elastase inhibition may contribute to the observed increase in bacterial clearance with Prolastin aerosolization. Such potential mechanisms include the inhibition of serine proteases other than elastase, which may be derived from bacteria. However, it is important to note that beneficial effects of neutrophil elastase inhibition have been reported in patients with CF despite the persistence of active elastase in BAL fluids. Marked decreases in BAL IL-8 and neutrophil counts of patients with CF treated with aerosolized SLPI were observed despite the persistence of active neutrophil elastase in nearly all treated patients (20). Our results are consistent with these observations and, importantly, they suggest that it may not be necessary to suppress all of the neutrophil elastase at the CF airway surface to induce a decrease in lung inflammation and bacterial burden. Although the current study did not allow us to define the mechanisms by which inhibition aerosolized Prolastin enhances lung host defenses, these studies clearly have significant implications for the development of novel, nonantibiotic therapeutic strategies to fight the chronic lung infection in cystic fibrosis.

In summary, these data indicate that aerosolization of Prolastin in lungs that are chronically infected with P. aeruginosa markedly decreases lung inflammation and accelerates bacterial clearance. The latter observation is particularly interesting since the current strategy of infection control in patients with CF involves the recurrent use of broad-spectrum antibiotics. This strategy is associated with an increasingly alarming rise in the rate of appearance of antibiotic-resistant organisms. We speculate that aerosol delivery of an effective HNE inhibitor such as ₁PI may prove to be a useful approach to control chronic lung infection and inflammation in cystic fibrosis without providing further pressure to increase antibiotic resistance.