INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Protection within the airways against infection involves a complex interaction between components of both the innate and adaptive immune systems. For example, neutrophils are central to the control of infection within the lumen of the bronchus. Neutrophil recruitment, activation, and retention within the bronchus are controlled by both T-cell-dependent (1, 2) and autocrine mechanisms, with the latter involving the secretion of proinflammatory cytokines such as tumor necrosis factor (TNF)-, interleukin (IL)-1, and IL-8 by activated neutrophils (3). The improved efficiency of bacterial clearance that comes from the activation of mucosa-targeted T cells is seen in a reduction in sputum purulence of subjects with chronic bronchitis after oral immunization with killed bacteria (4, 5).

A characteristic of intrabronchial neutrophils is the production of lysozyme secreted into the bronchial lumen (6). The level of lysozyme in respiratory tract secretions of subjects with chronic bronchitis has been inversely related to proneness to recurrent episodes of acute bronchitis (7). The mechanism of protection mediated by lysozyme involves aggregation of bacteria, with reduced colonization of mucosal surfaces (8, 9). Thus, control of lysozyme secretion is likely to be a key factor influencing the pattern of respiratory tract infections in patients with damaged airway defenses. Episodes of acute bronchitis have been particularly linked to intercurrent virus infection, with influenza virus infection attracting particular concern in the elderly population. The present study was planned to examine the influence of influenza virus of lysozyme secretion by sputum neutrophils obtained from subjects with bronchiectasis, to determine whether the epidemiologically defined link between viral and bacterial infection of the respiratory tract can in part be understood in terms of an acute reduction in neutrophil function induced by influenza virus infection.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Six subjects with bronchiectasis were studied. The diagnosis of bronchiectasis was confirmed by computed tomographic examination of the lungs. Each subject produced sputum daily. Assessment of the subjects was chronologically remote from episodes of acute infection, and was done when no antibiotics or corticosteroids were being taken. The study was approved by the Ethics Committees of the Hunter Area Health Service and the University of Newcastle.

Sputum Collection

Sputum was collected by expectoration into a 50-ml sterile container, and on some occasions venous blood was drawn at the same time into a tube containing trisodium citrate. All specimens were processed immediately after collection.

Neutrophils

Neutrophils were isolated from sputum through modification of a previously described procedure (3). In brief, sputum was collected in a sterile container and transferred into a 50-ml Falcon centrifuge tube (Becton Dickinson, Mountain View, CA) to which 25 ml of cold, sterile Hank's balanced salt solution (HBSS) was added. The mixture was shaken vigorously, and after addition of an equal volume of HBSS the suspension was filtered through a 10-ml syringe packed with glass wool. The cells were collected by centrifugation, washed twice with HBSS, and then resuspended in 10 ml HBSS. Following this, the cells were layered onto a Percoll gradient (40% to 50% stock solution in saline) and centrifuged for 20 min at 400 × g to remove macrophages, epithelial cells, and lymphocytes. Neutrophils were recovered from the interface of the 50% and 40% phases. Cells were counted, and their viability was assessed with the trypan blue dye exclusion test. The purity of the neutrophil suspension was greater than 98% as assessed by May- Grunwald-Giemsa staining.

Blood neutrophils were isolated from citrated blood by sedimentation in 3% gelatin/0.9% saline, followed by centrifugation through a Ficoll/Hypaque density gradient. Erythrocytes were removed by lysis in 0.15 M NH₄Cl and 0.1 M Tris-HCl, pH 7.2. The neutrophils were recovered, washed in phosphate-buffered saline (PBS), and resuspended in HBSS. The purity of neutrophils was between 98% and 100%.

Preparation of Influenza Virus

All viruses were obtained from Professor Greg Tannock of the Department of Applied Biology and Biotechnology, Royal Medical Institute of Technology, Victoria, Australia, and were grown in fertile hens' eggs for 2 d at 33 to 34° C. The following strains were used: A/ Texas/36/91 (H1N1), A/Qld/6/72 (H3N2), A/Ann Arbor (AA)/6/60 (H2N2), A/Beijing/352/92 (H3N2), B/Panama/45/90, and B/USSR/3/ 87. Viruses were harvested from allantoic fluid and purified on a sucrose gradient (10). Virus stocks were dialyzed against PBS and stored at 70° C until used. The potency of each stock was determined with an infectivity plaque assay, using Madin-Darby Canine Kidney (MDCK) cells, and titers of 10⁸ to 10⁹ pfu/ml were obtained.

Lysozyme Release

Neutrophils (5 × 10⁶/ml) were incubated with 0.1 ml of virus or HBSS for 30 min at 37° C. After various times, the cells were removed by centrifugation and the supernatant was assayed for lysozyme concentration. In some experiments, intracellular lysozyme content was measured in clarified cell lysate obtained by lysing neutrophils with a 0.05% Triton-100 solution followed by centrifugation. Lysozyme was measured with a double-antibody sandwich-type enzyme-linked immunosorbent assay (ELISA) as previously described (11). Briefly, ELISA plates (Immunoplate I; Nunc, Copenhagen, Denmark) were coated overnight at 4° C with rabbit antihuman lysozyme in sodium bicarbonate buffer (pH 9.6) (DAKO, Glostrup, Denmark). After washing of the plates with PBS containing 0.05% (vol/vol) Tween, various concentrations of standard purified human lysozyme (Kallestad, CA) or test supernatants diluted in PBS/Tween/3% bovine serum albumin (BSA) were added to each well. The plates were incubated for 90 min at room temperature (RT), after which they were washed, and horseradish peroxidase-conjugated antihuman lysozyme antibody (Binding Site Ltd, UK) diluted in PBS/Tween/3% BSA was added to each well and incubated for 30 min at RT. After washing, enzyme substrate (tetramethylbenzidine dihydrochloride) (Sigma-Aldrich Ltd., Australia) was added to each well for 20 min. The reaction was stopped with sulfuric acid, and the absorbance was read on an ELISA plate reader (Bio-Rad Laboratories, Richmond, CA). The lysozyme content in micrograms per milliliter of supernatant samples was read from a standard curve. The ELISA assay was antigen-specific, since human lysozyme was undetectable in culture supernatants spiked with purified virus, virus in allantoic fluid, or chicken lysozyme (data not shown).

Myeloperoxidase Production

Myeloperoxidase (MPO) in culture supernatants of neutrophils was measured in triplicate, using commercial ELISA kits (R&D Systems, Minneapolis, MN).

Bactericidal Activity

One-milliliter aliquots of neutrophil suspension (5 × 10⁶/ml) were preincubated with 0.1 ml virus or HBSS for 30 min at 37° C, after which preopsonized Pseudomonas aeruginosa serotype II, phage type 21/44/109/119X/1214 (Central Public Health Laboratory, London, UK) was added at a ratio of 25:1 and the preparation was incubated with gentle mixing on a roller for 30 or 60 min at 37° C. At each time point, several dilutions of the incubation mixture were plated on chocalate agar. After incubation for 3 d, the percentage of bacteria killed was calculated (Table 5). The percentage of bacteria killed varied from 10% to 50% in the neutrophil-bacteria suspension, but no bacterial growth was detected in the cell lysate prepared from neutrophils that were washed three times in HBSS before being homogenized in 0.01% BSA (wt/vol in HBSS) with an Ultra-Turrax T25 homogenizer fitted with a probe (Staufen, Germany).

Statistical Analysis

For individual subjects, data are given as the mean + SEM. Median values and ranges are shown for the subject group. Statistical significance was evaluated with the Mann-Whitney U test. All calculations were performed with a statistical software program (PRISM Version 2; GraphPad, San Diego, CA).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Lysozyme Secretion by Sputum Neutrophils Infected with Influenza Virus

Table 1 shows the amounts of lysozyme secreted by sputum neutrophils obtained from six subjects from bronchiectasis. The results showed that cells infected with influenza virus A/ Qld secreted significantly less lysozyme than did uninfected controls (p < 0.05). Table 2 shows the time course of lysozyme secretion compared with that of MPO in sputum neutrophil cultures infected with influenza virus A/Qld/6/72. Although lysozyme secretion was inhibited by influenza virus, there was no significant reduction in MPO levels in culture supernatants of uninfected neutrophils, irrespective of the time of incubation or the dose of virus (data not shown).

Effects of Virus Dose and Virus Hemagglutin (HA) Subtypes on Lysozyme Secretion

Sputum neutrophils were infected with graded doses of A/ Qld/6/72 virus and then incubated for 30 min at 37° C, after which secreted lysozyme was measured in supernatant. As shown in Table 3, lysozyme secretion by sputum neutrophils was inhibited by the virus in a dose-dependent manner. In addition, inhibition of lysozyme was not restricted to a particular influenza A virus strain, as shown in Table 4, but influenza B viruses (B/USSR and B/Panama) had little or no effect.

Effect of Influenza Virus Infection on Intracellular and Extracellular Lysozyme Concentrations

To determine whether there was a relationship between extracellular and intracellular lysozyme concentrations, we infected blood and sputum neutrophils from six individual subjects with influenza virus and measured the kinetics of the amounts of lysozyme released or retained in the cells. As shown in Figure 1, secretion of lysozyme by sputum neutrophils was inhibited as early as 15 min and for up to 120 min after infection. Although marked inhibition of lysozyme secretion by blood neutrophils infected with virus was observed at 15 min and for up to 60 min after infection, sputum neutrophils contained more lysozyme than did blood neutrophils (p < 0.05). In both cases, however, levels of lysozyme were unaffected by virus infection, irrespective of incubation time.

View larger version (52K): [in this window] [in a new window]   Figure 1.   Comparison of cellular and secreted lysozyme from blood and sputum neutrophils infected with influenza virus. Blood and sputum neutrophils were infected with 105 pfu/ml of influenza virus A/Beijing/352/92 (H3N2). Extracellular (secreted) and intracellular (retained) lysozyme were measured in triplicate in culture supernatants and cell lysates at various times shown. Results shown are mean ± SEM from six individual subjects. Uninfected controls are shown by open bar; infected subjects are shown by hatched bar. *p < 0.05; **p < 0.001 compared with values from uninfected cultures.

Effect of Influenza Virus Infection on Bactericidal Activity of Sputum Neutrophils

Sputum neutrophils were infected with various strains of influenza viruses and then challenged with live P. aeruginosa bacteria. As shown in Table 5, the ability of neutrophils infected with influenza virus strains A/Qld (p < 0.05) and A/Texas (p < 0.001) to kill bacteria was decreased after 30 and 60 min. In contrast, a decrease in killing activity was observed only after 60 min for cells infected with influenza virus strain A/AA (p < 0.05), and minimal to no killing was observed for cells infected with strains B/USSR or B/Panama (data not shown for the latter virus).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Infection with influenza A virus of neutrophils obtained from the sputum of subjects with bronchiectasis induced a rapid downregulation of function, characterized by a reduction in lysozyme secretion and in bactericidal activity. The cells remained functionally intact, with no change in cell loss of MPO. Lysozyme secretion and bactericidal function were similarly reduced by different influenza A virus strains, but an influenza B virus strain had no effect on neutrophil function.

In human airway secretions, lysozyme is derived from various sources, including the submucosal tracheal glands, surface epithelial cells, and pulmonary alveolar macrophages (AM) (12). Although AM provide the primary phagocytic response to inhaled microbes in the normal bronchoalveolar system, the neutrophil is the dominant phagocytic cell in subjects with damaged airways chronically colonized by bacteria, representing more than 98% of the cellular content of sputum from subjects with chronic lung disease (3, 13). We have found that about two thirds of the total sputum lysozyme in subjects with chronic bronchitis and bronchiectasis was contained in the cell pellet of centrifuged sputum (unpublished observations). Neutrophil-derived lysozyme plays a major role in the intracellular destruction of ingested bacteria, through the development of phagolysosomes with primary (or azurophilic) and secondary granules (14, 15), and in the extracellular control of colonization of the mucosa (16), through exocytosis of secondary granules (17, 18). By contrast, MPO is the most abundant enzyme in the primary granule; it is responsible for oxygen-dependent catalytic reactions between hydrogen peroxidase and chlorides to form hypochlorous acid, and is thus largely intracellular (14, 15). These different transport characteristics are consistent with the relative difference in lysozyme and MPO concentrations in culture supernatants, as found in the present study. Furthermore, the lack of effect of virus infection on release of MPO as found in the study reflects the maintenance of cell integrity. Both T-cell-dependent recruitment and activation of neutrophils (14, 19), and the consequent autocrine mechanisms that maintain activated neutrophils within the bronchial lumen (3, 20), contribute to the significant functional differences noted between blood and sputum neutrophils. The higher levels of lysozyme stores within the sputum neutrophil, and the greater and more sustained release of lysozyme in cultures of sputum as compared with blood neutrophils, reflect this altered phenotype. Synthesis of granule contents is considered to occur within the bone marrow before the neutrophil is released into the bloodstream (16). It is unclear whether the increase in sputum neutrophil lysozyme stores reflect reactivated lysozyme synthesis or uptake from the environment. In the lower respiratory tract, both AM and neutrophils are important in host defense against bacterial infection, though they differ from blood neutrophils by continuously synthesizing lysozyme (12, 17). Macrophages secrete lysozyme in amounts greatly exceeding its intracellular stores, whereas blood neutrophils retain the enzyme in lysosomes, releasing it during degranulation (17). Study of lysozyme synthesis in sputum neutrophils may provide an explanation for the increase in intracellular stores of this enzyme.

The mechanism by which influenza A virus inhibits neutrophil function is not entirely clear. The virus enters the cell after binding to the hemagglutinin receptor on the cell surface (21, 22). Recent studies with blood neutrophils show that exposure to influenza A virus inhibits bactericidal activity directed against Staphylococcus aureus, as a consequence of defective formation of phagolysosomes involving primary granules (23). The inhibition of release of lysozyme into culture medium induced by infection with influenza A virus suggests impaired exocytosis, also reflecting defective membrane fusion. Such a mechanism would be consistent with the retention of cell structure (with no leakage from the primary granule) in infected cells, and the time course of 2 h for diminished lysozyme secretion by noninfected cells observed in our study would reflect depletion of a small intracellular compartment (associated with secondary granules), with a decrease of only 10% of the total intracellular lysozyme content. The fairly constant lysozyme secretion rate of infected cells over the 2-h period would then reflect a longer maintenance of this minor pool, owing to the membrane fusion defect induced by infection. An alternate mechanism for the disturbed transport dynamics of lysozyme would be a loss of inhibition caused either by the neutrophil becoming refractory to the effect of virus (24) or by the virus eluting from the cell or becoming damaged.

The difference between the effect of influenza A and influenza B virus on neutrophil lysozyme release and bactericidal activity is consistent with epidemiologic evidence of greater morbidity and mortality following infection with strains of influenza A virus (25). Morbidity and mortality usually involve secondary bacterial infection, which would occur in subjects with lung disease and in the elderly, whose airways are often colonized by bacteria (26, 27). The sudden reduction in the capacity to control bacterial colonization that would follow influenza A virus infection of sputum neutrophils could be critical in the pathogenesis of clinical airways infection with bacteria, and in the subsequent development of pneumonia. Additional mechanisms promoting bacterial infection after influenza virus infection include a specific direct effect on nontypeable Hemophilus influenzae, as well as nonspecific damage to airways defense mechanisms (28). Subjects with chronic lung disease and a low frequency of recurrent acute infectious episodes have significantly lower levels of lysozyme in their saliva than do matched subjects with a high rate of acute episodes (7, 29).

The present study thus provides the first direct evidence of an acute and significant impact of influenza A virus on the protective function of the intrabronchial neutrophil. It identifies a mechanism for the morbidity and mortality caused by bacterial infection of the bronchoalveolar space following infection with influenza A virus.