Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Because the lung is constantly exposed to a broad variety of environmental toxins and pathogens, it is a main site of infections in immunocompromised individuals, e.g., after transplantation. Alveolar macrophages (AM) play a central role in the defense of the respiratory tract against inhaled pathogens, as these cells are capable of producing various cytotoxic, immunoregulatory molecules or chemoattractive mediators as reactive oxygen intermediates (ROI), tumor necrosis factor (TNF)-, or interleukins (IL) (1). Interstitial lung macrophages (IM) have, up to now, not been similarly characterized as they are not as easily accessible as AM. Since isolation procedures for IM from pulmonary tissues of rodents were described (4), it has been possible to gain insights into their morphologic and functional characteristics. Murine IM were shown to exhibit immunoregulatory and accessory functions (7, 8). So far in rats, mainly phenotypical differences of IM and a reduced microbicidal potential in comparison with AM were described (9), whereas little information on the secretory functions of rat IM is available. Cytotoxic or inflammatory mediators released by IM have a greater effect on the surrounding lung tissue than do secretory products of AM, as IM are directly embedded into the lung tissue. Furthermore, immune reactions in the lung have been shown to be often compartmentalized, so that data obtained from bronchoalveolar lavage (BAL) fluid may reveal little information or might even result in misleading conclusions.

Focusing on the microbicidal potential of lung macrophages, we performed studies to analyze whether this function could be augmented in disease models (bacterial infection in normal or immunosuppressed rats) by treating the animals with interferon (IFN)- intratracheally. This approach could support therapeutical strategies favoring a local immunomodulation, for which the respiratory tract provides optimal conditions.

IFN-, an important immunoregulatory cytokine, is known to enhance the effector functions of macrophages in vitro (12). In addition, systemic application of recombinant murine IFN- resulted in protection of mice against parasitic (13) and bacterial (14) infections in vivo. The respiratory tract is suitable for local immunotherapy because the mediators reach the organ directly without prior dilution or metabolization in the circulation. A local activation of alveolar macrophages was observed in human test subjects (15) and in mice (16) after inhalation of aerosolized IFN-. Likewise, intratracheal instillation of IFN- into rats resulted in an augmentation of cytotoxic functions of AM (17). However, it remained unclear as to what extent IM from the lung tissue were affected and if IFN- application influenced the immunoregulatory functions of pulmonary macrophages.

In this study, we characterized IM in comparison with AM, bone marrow-derived macrophages, and peritoneal macrophages with respect to microbicidal and immunoregulatory functions. Furthermore, we examined the pulmonary and systemic effects of a local cytokine delivery on nonspecific defense mechanisms mediated by organ macrophages. The in vivo relevance of the cytokine treatment was evaluated in normal and immunosuppressed rats infected with a respiratory model infection.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Male Lewis rats weighing between 200 and 250 g were obtained from Charles River (Sulzfeld, Germany). The animals were housed under conventional breeding conditions with two animals per cage and free access to food and water. Analysis for bacterial or viral infection was performed monthly.

Culture Media

RPMI-1640 medium (GIBCO Europe, Karlsruhe, Germany) was supplemented with 2 g NaHCO₃ and 10⁵ U penicillin-streptomycin/liter (Seromed, Munich, Germany). For tissue cell cultures, 10% fetal calf serum (complete medium; GIBCO) was added.

Cell Lines

vP815 mastocytoma cells were maintained as suspension cultures in complete medium. Murine L-929 fibroblasts were grown as adherent cell layers in complete medium. The IL-6-dependent 7TD1 B-cell line was maintained in medium with supernatant of concanavalin A-stimulated spleen cells.

Chemicals and Lymphokines

Recombinant IFN- was kindly provided by P. H. van der Meide (TNO Primate Center, Rijkswijk, The Netherlands). The cytokine was produced in Chinese hamster ovary cells and subsequently purified by monoclonal antibody affinity chromatography (18).

Bacterial lipopolysaccharide (LPS) from Escherichia coli strain 0111:B4 was purchased from Sigma (Munich, Germany). Recombinant human IL-6 was obtained from Genzyme (Boston, MA).

Intratracheal Administration of IFN-

The rats were anesthetized by halothane and put into a vertical position. A blunt cannula was inserted into the trachea via the mouth, and 0.2 ml IFN- diluted in phosphate-buffered saline (PBS) (5 × 10² to 10⁵ U/rat), or an equal volume of PBS alone, was instilled intratracheally (19) followed by 1 ml of air. The animals were held in this position until they recovered from the anesthesia.

Bronchoalveolar Lavage

The animals were killed by injection of pentobarbital. Vascular perfusion was performed via the right ventricle with chilled PBS until the lungs were pale white. Thereafter, the trachea was cannulated and the lungs were flushed eight times with 5 ml ice-cold PBS without calcium and magnesium, containing 0.4 mM ethylenediaminetetraacetic acid (PBS-EDTA), under moderate massage of the lungs. Differential cell counts were performed on cytospin preparations stained according to Pappenheim in order to determine the percentage of macrophages in the lavage fluid. A total of 10⁵ macrophages/well was seeded in a 96-well, flat-bottomed microtiter plate. For IFN- detection in the alveolar space, the organ was lavaged only once and the fluid was separated from cellular material by centrifugation.

Interstitial Lung Macrophages

Macrophages from the lung interstitium were harvested as described by Holt and coworkers (4). Briefly, the lavaged, perfused lungs were minced with a tissue chopper and incubated under moderate agitation for 60 min in complete medium containing collagenase (100 U/ml; Worthington type 1, Bayer Diagnostic, Munich, Germany) and DNAse (50 U/ml; Sigma) at 37°C. The cell suspension was poured through a sterile steel sieve to remove tissue fragments, washed in cold RPMI medium, and layered on a discontinuous Percoll gradient. After centrifugation for 30 min at 300 × g, the 20 to 45% fraction was harvested, and the macrophage percentage was determined by esterase stain of cytospots. The cell suspension was adjusted to 10⁶ esterase positive cells/ml RPMI medium. A total of 100 µl/well was allowed to adhere, and a macrophage monolayer was obtained after vigorously washing the wells with warm culture medium.

Peritoneal Macrophages

Peritoneal macrophages were isolated 4 d after intraperitoneal injection of a sterile 2% (wt/vol) starch solution by lavage of the peritoneal cavity with 20 ml cold PBS-EDTA. Erythocytes were removed by osmotic lysis. After differential cell count by staining according to Pappenheim, the cell suspension was adjusted to 10⁶ macrophages/ml complete medium. A total of 100 µl was plated in each well, and nonadherent cells were removed after a 2-h incubation period.

Spleen Macrophages

A spleen cell suspension in complete medium was separated on a discontinuous Percoll gradient. The interface between 20 and 40% was harvested, analyzed by Pappenheim stain, and a monolayer of 10⁵ macrophages/well was achieved after adherence purification.

In Vitro Activation of Effector Cells

Macrophages (10⁵ cells/well of a 96-well microtiter plate in 100 µl RPMI medium) were incubated with LPS or medium alone for 18 h at 37°C, 85% humidity, and 5% CO₂. Subsequently, the supernatant was removed to be monitored for macrophage secretory products. A total of 100 µl fresh complete medium was added to the cells before the tumoricidal assay was performed.

Detection of IFN- in the Lavage Fluid, Lung Tissue, and Serum

After lavage, the lungs were homogenized in 5 ml PBS with a teflon homogenizer. Blood was taken from the vena cava for serum preparation. IFN- in the body fluids was determined by enzyme-linked immunosorbent assay (Holland Biotechnology, Leiden, The Netherlands).

Bioassay for IFN-

IFN- in the alveolar space was measured by an antiviral protection assay using L-929 fibroblasts infected with encephalomyocarditis virus as decribed previously (20).

Fluorescent Staining and Flow Cytometric Analysis

Macrophages (10⁵ cells) were seeded in polystyrole vials with , and Fc receptors were blocked 5 min with 5% inactivated goat serum to prevent nonspecific binding. The cells were stained 30 min with a 1:100 dilution of the mouse anti-rat major histocompatibility complex class II monoclonal antibody Ox6 (Camon, Wiesbaden, Germany) and incubated for another 30 min with polyclonal fluorescein isothiocyanate-labeled goat antimouse serum (1:40; Dianova, Hamburg, Germany). As control for nonspecific fluorescence, cells were incubated with the second antibody only. After each incubation step at 4°C in the dark, the cells were washed three times with cold PBS. The measurement was performed in a FACScan (Becton Dickinson, Heidelberg, Germany). The percentage of positive cells was calculated as the cells staining above the background staining obtained in the absence of primary antibody.

⁵¹Cr Release Assay against P815 Tumor Cells

Macrophages were cocultured with ⁵¹Cr-labeled P815 tumor target cells. After 18 h, the amount of radioactivity in the supernatant was determined. For spontaneous release, tumor cells were cultured in medium without effector cells. The percentage of specific lysis was calculated as follows:

Measurement of TNF- Activity

TNF- activity in macrophage supernatants was determined in a biologic assay using actinomycin D-treated L-929 cells (21). One unit of TNF- is defined as the reciprocal of the supernatant dilution that would cause lysis of 50% of the L-929 cell layer.

Determination of IL-6 Production

7TD1 cells were incubated with the supernatants (22), and proliferation was detected by 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) incorporation. For evaluation of the IL-6 concentration, a recombinant human IL-6 standard was available.

Formation of Nitrite

Nitrite in the culture supernatants was measured spectrophotometrically using the Griess reaction similar to the procedure described previously (23). Concentrations were determined by using a standard solution of sodium nitrite in RPMI medium.

Lucigenin-Dependent Chemiluminescence

A total of 10⁵ AM in complete medium with 10 mM N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid buffer was incubated 30 min with lucigenin (bis-N-methyl-acridinium nitrate) at a final concentration of 0.24 mM to allow background chemiluminescence to diminish. A total of 10 µl zymosan (12.5 mg/ml) was added and chemiluminescence resulting from subsequent generation of ROI was monitored for 30 min in a six-channel Berthold Biolumat (Berthold, Wildbad, Germany). Software for computerized calculation of integrals was supplied by Berthold.

Immunosuppression

Animals were immunosuppressed daily starting 4 d before macrophage isolation either with cyclosporine A (25 mg/kg bodyweight intraperitoneally) or with a triple drug protocol (cyclosporine A 8 mg/kg + azathioprine 2 mg/kg + prednisolone 0.1 mg/kg intraperitoneally).

Pulmonary Infection with Listeria monocytogenes

Rats were infected intratracheally with 10⁵ CFU L. monocytogenes in 0.2 ml NaCl. A subgroup of animals received daily cyclosporine A treatment (25 mg/kg intraperitoneally) starting 24 h before the infection and continued until termination of the experiment. Bacterial numbers were determined by plating serial dilutions of the inoculum or organ homogenates on trypton-soy agar and counting CFU after 24 h.

Statistical Analysis

The mean and standard deviations were calculated. In vitro experiments were repeated at least three times with five samples per group. In vivo experiments were performed with at least three rats per group and repeated twice. Differences between the treatment groups were analyzed by Wilcoxon's rank test and P < 0.05 was accepted as significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Functional Comparison of AM and IM

Microbicidal mechanisms of pulmonary macrophages. To assess the microbicidal potential of AM and IM, the production of cytotoxic nitric oxides (NO) and reactive oxygen species, the release of TNF-, and cytotoxicity against tumor cells were determined in culture supernatants. Bone marrow macrophages and highly activated peritoneal macrophages are shown for comparison. Figure 1 documents that AM are more potent producers of NO than are IM, peritoneal macrophages (PM), and bone marrow-derived macrophages, especially after in vitro incubation with IFN-+LPS. Regarding the zymosan-induced release of cytotoxic superoxide anions, again AM were more effective producers of the intermediates than were IM (AM: 2.2 + 10⁸ cpm; IM: 8.5 × 10⁶ cpm; P < 0.05).

View larger version (28K): [in this window] [in a new window]   Figure 1.   Production of NO and TNF- by AM and IM compared with data from bone marrow-derived macrophages (BM) and PM. TNF- in culture supernatants was determined after 18 h incubation either in medium alone or in IFN- (100 U/ml) + LPS (100 ng/ml). *Significantly different (P < 0.05) from AM treated with the same stimulus. Data are the mean ± standard deviation (SD) from five experiments.

Immunoregulatory functions of pulmonary macrophages. The two main populations of lung macrophages were compared with respect to production of IL-1, IL-6, and the expression of surface MHC class II. The percentage of MHC class II expression is higher in IM than in AM (AM: 10.5% ± 1%; IM: 36.2% ± 4.5%; P < 0.05). The capacity to secrete IL-1 and IL-6 is also more pronounced in IM than in AM (Figure 2). This is true for spontaneous secretion and for secretion of activated macrophages.

View larger version (17K): [in this window] [in a new window]   Figure 2.   Release of IL-1 and IL-6 by different macrophage populations. Cells were cultured in vitro in medium alone or in medium containing bacterial LPS (100 ng/ml). *Significantly different (P < 0.05) from AM treated with the same stimulus. Data are the mean ± SD from five experiments.

Local Activation of AM by Intratracheal Application of IFN-

Distribution and kinetics of IFN- in lavage fluid, lung tissue, and serum after intratracheal administration. In a first set of experiments, we characterized the distribution of IFN- (5 × 10⁴ U/rat) in different compartments of the experimental animals. The cytokine could be detected in the lavage fluid but not in the serum of the rats (data not shown). There was only a minimal amount of IFN- measured in the lung tissue homogenate by ELISA. Approximately 70% of the cytokine were recovered from the alveolar space 30 min after treatment of the animals. Using the antiviral bioassay, it could be demonstrated that the alveolar IFN- detected by ELISA exhibited biologic activity. The decay curves measured with both test systems showed a similar course. Six hours after cytokine administration, IFN- was not detected by ELISA, nor could any remaining antiviral activity be found in the BAL fluid using the bioassay.

Number, viability, and distribution of cells obtained by BAL after intratracheal cytokine application. The intratracheal instillation had no influence on the cell number or on cell viability, as tested by trypan blue exclusion (data not shown). There was no significant difference noted in the composition of the cell populations isolated from the IFN- group as compared with the control rats. A slight, but not significant, influx of polymorphonuclear neutrophils into the alveolar space in some individual animals was observed after instillation of IFN-. No additional changes were noted with regard to the composition of the cell suspensions after treatment.

Enhancement of Microbicidal Defense Mechanisms of Pulmonary Macrophages (ROI, NO, TNF-)

We subsequently investigated the microbicidal potential of the different macrophage populations harvested after local IFN- administration. Zymosan-induced release of cytotoxic superoxide anion by AM, as detected by lucigenin-mediated chemiluminescence, was enhanced in the IFN- group (PBS group: 1.61 × 10⁸ integral cpm in 30 min; IFN- group: 2.43 × 10⁸ integral cpm in 30 min; representative result with n = 3 in each of three experiments).

As shown in Figure 3, the level of cytotoxic nitrogen intermediates was also progressively enhanced in the supernatant fluids of AM from animals pretreated in vivo with increasing doses of IFN-. The culture medium of phagocytes from animals that had received dosages of 10³ or 10⁴ U IFN- contained 46.4 or 69.8 µM nitrite. The secretory potential of these cells could be further stimulated by adding LPS in increasing concentrations to the culture medium (Figure 3). After in vivo exposure to 5 × 10⁴ U IFN-, maximal spontaneous nitrite release (126.4 µM) was detected in the medium of AM. Addition of LPS to these cultures did not result in any further stimulation, indicating an optimal secretory activity in vivo.

View larger version (19K): [in this window] [in a new window]   Figure 3.   Increasing release of NO by pulmonary macrophages after local treatment of rats with progressing dosages of IFN-. Animals received between 5 × 102 and 5 × 104 U IFN- intratracheally 18 h before isolation of AM and IM. Data from IM are only shown after in vivo treatment with the highest dosage of IFN- used. The macrophages were incubated in vitro for 18 h with LPS (1,000 ng/ml) (closed bars) or medium alone (open bars), and the culture medium was used in a photometric assay for nitrite detection. Statistical analysis showed significant effects (P < 0.01) between AM from IFN--pretreated rats (103 to 5 × 104 U/animal) and control animals. Data are the mean ± SD from five experiments with three animals per group.

The augmentation of NO generation by AM was paralleled by elevated nitrite concentrations in the medium of macrophages isolated from the lung tissue. However, similar to the observations concerning release of IL-6 and tumor cytotoxicity shown subsequently, reactive nitrogene intermediate production by IM was only slightly augmented even after intratracheal instillation of the highest IFN- dose of 5 × 10⁴ U/animal (Figure 3). PM were not affected regarding nitrogen metabolism.

To determine if the IFN- treatment affected the release of TNF-, we assayed supernatants of AM, IM, and PM 18 h after cell isolation for TNF- activity. The phagocytes harvested from control animals exhibited only marginal TNF- production. Intratracheal deposition of IFN- resulted in a significant accumulation of TNF- in the supernatant fluids of AM ex vivo, whereas the secretory activity of the other macrophage populations was not altered (data not shown).

Local Enhancement of Tumoricidal Activity of Pulmonary Macrophages

Macrophages were isolated from animals after intratracheal instillation of PBS or 5 × 10⁴ U IFN-, respectively, and examined with regard to their tumoricidal capacity against the xenogenic tumor target P815. Pulmonary macrophages isolated either from alveolar space or lung tissue of rats from the IFN- group exhibited cytotoxicity against the tumor cells, whereas control macrophages were not capable of lysing the P815 targets. An additional in vitro incubation with LPS resulted also in activation of control alveolar and lung tissue macrophages to lyse P815 tumor targets (Figure 4). IM from IFN--treated rats could be further activated by in vitro incubation with LPS, whereas the cytotoxic potential of AM from these animals was already enhanced. In contrast, the tumoricidal potential of PM and splenic macrophages was not changed after in vivo lymphokine application (data not shown). Thus, local IFN- deposition resulted in a maximum activation of AM in vivo, whereas lung tissue macrophages were only partly influenced.

View larger version (18K): [in this window] [in a new window]   Figure 4.   Induction of cytotoxicity against P815 tumor cells in the lungs after cytokine administration. AM and IM were harvested 18 h after local delivery of 5 × 104 U IFN-. Lysis of P815 tumor cells after an 18-h incubation period with LPS (100 ng/ ml) (open columns) or medium (closed columns). Each column represents the mean of at least three experiments and each group includes four animals. *P < 0.05 as compared with the data from PBS-treated control rats.

Dose-Dependent Increase in MHC Class II Antigen Expression on AM

AM exposed in vivo to IFN- revealed a significant increase of MHC class II expression as evaluated by flow cytometry after staining of the cells with the antirat MHC class II monoclonal antibody Ox6. This effect was dose- dependent using IFN- dosages ranging from 10³ to 10⁵ U/animal (Figure 5). Administration of 5 × 10² U IFN- did not influence the MHC class II expression and resembled macrophages from control animals (10.5% positive stained cells). A maximum of 56.7% AM expressing surface MHC class II molecules above the control level was detected on cells from rats pretreated with 10⁵ U IFN-. The expression of MHC class II determinants on lung tissue macrophages was enhanced by 33.9% after intratracheal instillation of 5 × 10⁴ U IFN-. This enhancement of MHC class II antigen expression was limited to pulmonary macrophages because no changes in the peritoneal cavity or spleen could be demonstrated.

View larger version (16K): [in this window] [in a new window]   Figure 5.   Expression of MHC class II molecules on AM after intratracheal instillation of various concentrations of IFN-. BAL and antibody staining of the macrophages for FACS analysis were performed 18 h later. The expression of MHC class II proteins in the adequate control group was substracted. Each column represents the mean ± SD of at least three experiments with four animals each. Macrophages from all treatment groups were significantly (P < 0.05) different from cells obtained from PBS-treated control animals.

Influence of the Local IFN- Delivery on IL-6 Secretion

IL-6 plays a crucial role in the onset of a specific immune reaction, as it takes part in T-cell activation and enhances the final differentiation and antibody secretion of activated B cells. The secretion of this cytokine by AM was significantly increased in a dose-dependent manner in rats treated with interferon doses between 10³ and 5 × 10⁴ U (Figure 6). AM harvested from animals after in vivo exposure to 5 × 10⁴ U IFN- exhibited a more than 20-fold increase in IL-6 secretion as compared with cells from control animals. IFN- dosages of 10³ or 10⁴ U/animal elevated the amount of IL-6 secreted from AM from 7-fold to 10-fold. Lung tissue macrophages were less stimulated by IFN-, showing a significantly elevated IL-6 release compared with cells from control animals only after exposure to a high dose of interferon (5 × 10⁴ U/rat). PM were not influenced regarding IL-6 production.

View larger version (15K): [in this window] [in a new window]   Figure 6.   Dose-dependent augmentation of IL-6 production by pulmonary macrophages. The IL-6 release of macrophages from PBS-treated animals was considered to be one, and the cytokine production in the IFN- group was calculated as a factor of the PBS control. Data are expressed as mean of three experiments with four animals per group. *P values ranged from < 0.05 to < 0.01 as compared with animals of the PBS-treated group.

Duration of Macrophage Activation

To evaluate the duration of the effects described previously, we instilled 5 × 10⁴ U and killed the animals 3 d later. AM and lung digest macrophages were examined for expression of MHC class II molecules, tumoricidal function, release of IL-6, reactive nitrogen intermediates, and TNF-. No functional changes as compared with macrophages isolated from control rats that had received the same volume of PBS could be detected. Obviously, the organ-specific stimulation by IFN- is a transient effect.

The data on local activation of pulmonary macrophages after intratracheal application of IFN- show that, according to the presence of IFN- in the different compartments, AM display by far the highest degree of activation.

Local Therapy of a Respiratory Model Infection with L. monocytogenes

The distribution of L. monocytogenes in the organs on Days 2, 4, and 6 after intratracheal instillation of a sublethal dose of 10⁵ PFU/animal is shown in Figure 7. Intratracheal application of 5 × 10⁴ U IFN- 1 d before, the same day, and one day after infection significantly reduced the Listeria burden in lung, spleen, and liver.

View larger version (12K): [in this window] [in a new window]   Figure 7.   Local therapy of a respiratory infection with L. monocytogenes by application of IFN-. Animals were treated with 5 × 104 U IFN- intratracheally (open squares) or PBS (open diamonds) and infected intratracheally with 105 CFU L. monocytogenes. Data are the mean ± SD from three experiments with five animals per group.

Protective Effect of Local Macrophage Activation in Immunosuppressed Rats Infected with L. monocytogenes

In immunosuppressed organisms, e.g., after organ transplantation, respiratory infections are a major cause of morbidity and mortality. The commonly used immunosuppressive drug cyclosporine A mainly inhibits specific immune reactions, thus preventing rejection of MHC disparate transplants. By locally activating AM we intended to strengthen the nonspecific defense system of the lung.

Before starting infection experiments, we tested whether the immunosuppressive protocols used had any adversary effects on macrophage functions. After a 4-d intraperitoneal immunosuppresive treatment with cyclosporine A alone or a triple drug protocol, macrophages were isolated and studied in vitro. The immunosuppression did not influence the production of ROI, IL-6, MHC class II expression, tumor cytotoxicity, and NO release by AM, IM, or PM. Decreased cytokine concentrations were detected in the culture medium in the high-dose cyclosporine A group (data not shown) only during in vitro stimulation of AM and PM to release TNF-.

In another set of experiments, effects of a local IFN- administration on the resistance against a respiratory infection with L. monocytogenes in immunosuppressed rats were determined. As can be seen in Figure 8, local activation of the nonspecific defense system resulted in a highly significant reduction of Listeria organisms in the lung, spleen, and liver on Day 2 after infection as compared with PBS-treated rats. Whereas in the control group without interferon treatment, 80% of the animals died between Days 4 and 6 postinfection; in the IFN- group, 70% of the animals survived the infection. In healthy animals, production of IFN- by lymphocytes starts on about Day 4, resulting in macrophage activation and clearance of the infection. In contrast, immunosuppressed animals showed an aggravated course of infection owing to the limited functional capacity of the specific immune system, which could be overcome in our experiments by local activation of pulmonary macrophages.

View larger version (15K): [in this window] [in a new window]   Figure 8.   Local activation of pulmonary macrophages enhances the resistance of immunosuppressed rats against respiratory infections with L. monocytogenes. The animals received either cyclosporine A (25 mg/kg) or a triple drug immunosuppression (cyclosporine A, 8 mg/kg, azathioprine, and prednisolone) before cytokine treatment and infection with 105 CFU bacteria. Control rats were treated with the same volume of PBS (open columns). On Day 2 after infection, the bacterial organ burden was determined by plate counting. Data are the mean ± SD from three experiments with five animals per group.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

AM are the first line of defense against inhaled pathogens and thus exhibit phagocytosis and secretory functions enabling them to keep the alveolar surface sterile. The role of IM residing in the lung tissue is much less well characterized, which is in sharp contrast to their potential beneficial or destructive effects on the surrounding tissue. Owing to the immunologic compartmentalization of the lung, studies focused on analyzing functional alterations of alveolar cells obtained by BAL might result in misleading conclusions concerning immunologic reactions in the entire organ. Lung tissue leukocytes could react differently during infections, exposure to occupational pollutants, lung transplantation, or therapeutic treatment. The data presented here characterize alterations in the activation state of pulmonary macrophages, isolated either from the alveolar space or the lung tissue, after local deposition of recombinant rat IFN-.

For the isolation of IM, enzyme digestion of the lung tissue in collagenase/DNAse followed by Percoll density gradient centrifugation and finally plastic adherence were performed. Control experiments comparing AM parameters before and after the cells underwent the identical procedure excluded effects of the isolation protocol on macrophage functions or phenotype in the assay systems used for our studies, which is in line with results obtained in the mouse system (8). Furthermore, other investigators reported that macrophage populations isolated from rat lungs by mechanical procedures without digesting the lung tissue represent a subpopulation of AM (9, 24). The enzymatic method used for our studies resulted in a highly enriched population of esterase-positive, phagocytosing macrophages with phenotypic and functional properties clearly distinct from AM. Although perfusion of the vascular system and BAL were performed, a low grade contamination of our IM preparation by AM or monocytes cannot be excluded. However, the percentage of AM contaminations in IM preparations obtained by similar methods in the rat and mouse system was below 5% (8), which is not sufficient to explain the functional heterogeneity between the two macrophage populations.

AM obtained by BAL were highly active in releasing microbicidal mediators such as TNF-, NO, interferons, and ROI when compared with IM. In contrast, IM were better able to produce immunoregulatory cytokines such as IL-1 and IL-6. Furthermore IM expressed to a greater degree MHC class II proteins, which are key molecules during the antigen presentation step preceeding the onset of an antigen-specific immune reaction. Studies in the mouse and human system also revealed a reduced accessory activity of AM in comparison with IM (25, 26). Concerning microbicidal functions, rat AM were clearly superior to IM when enzymatic digestion instead of mechanical disruption of the tissue was performed (9). A similar functional distinction between AM and IM could also be demonstrated using murine cells (8), suggesting that our observations describe a common property of pulmonary macrophages. It seems justified to suppose that highly microbicidal AM in the alveoli effectively cooperate with their tissue counterparts, which are specialized in initiating specific immune reactions upon antigen entry into the lung tissue. This functional specialization of the macrophage populations reflects their anatomic position, as the release of ROI, NO, and TNF- by IM would have an injurious effect on the lung tissue. Additionally, the lowered immunoregulatory potential of AM restricts constant overall immune activation in the lung by inhaled occupational particles and pathogens. The high degree of microbicidal activity of AM corresponds to the fact that these cells are located in the first line of defense against respiratory infections.

After intratracheal instillation of IFN- into rats, cytokine detection was limited to the lung, and no activity was present in the serum as determined by ELISA and by a bioassay for the antiviral activity of IFN-. These data are in agreement with observations in human volunteers after inhalation of an IFN- aerosol that resulted in detectable amounts of the protein in the epithelial lining fluid of the lung but not in the serum of the test subjects (15). In this study, an elevated local expression of the IFN--specific IP-10 gene by AM, but not by monocytes, was also demonstrated. Consistent with these findings, functional effects of the local IFN- administration were totally restricted to macrophages isolated from rat lungs. The IFN- deposition into rat lungs was followed by a dose-dependent increase of MHC class II antigens on AM. The expression of MHC class II proteins by interstitial pulmonary macrophages was also augmented, whereas splenic or peritoneal macrophages were not affected.

Recently, it was reported that intratracheal administration of the IFN inducer polyinosinic-polycytidilic acid into rats resulted in an increased phagocytosis and tumor cytotoxicity of AM (27). Studies in mice using an aerosol of IFN- and LPS also revealed an activation of AM to kill certain tumor targets in vitro (16). Moreover, the cytotoxicity of AM against Toxoplasma gondii and P815 tumor cells was enhanced after intratracheal IFN- administration (17). We confirmed these data concerning AM tumor cytotoxicity and could exclude alterations in the cytotoxic potential of PM and splenic macrophages. Furthermore, we could demonstrate that IM were affected as well when using a high IFN- dose for treatment of the rats. Remarkably the IM were not activated to maximum lytic activity in contrast to AM, reflecting a compartmentalization of the effects within the lung.

In all assay systems, IM were less activated than were AM and responded only at high dosages of intratracheal IFN-. We investigated this graduated effect on the two populations of pulmonary macrophages in more detail with regard to the release of IL-6. We could detect a sig- nificant augmentation of IL-6 secretion by cells isolated from IFN--pretreated rats. IM were less sensitive to the cytokine treatment because it required higher dosages of IFN- to achieve a comparable enhancement of IL-6 secretory activity as compared with AM. Obviously, the amount of activating IFN- determines to what extent macrophages from the lung tissue are influenced in relation to AM. Moreover, after treatment of rats with the highest IFN- dose, production of TNF- was only detected in the culture supernatants of AM, whereas IM and PM showed no TNF- activity.

Subsequently, we concentrated on microbicidal defense mechanisms displayed by macrophages. The cytotoxicity of AM was clearly enhanced after the local IFN- administration as shown by an elevated generation of superoxide anions. In macrophages, nitric oxide radicals generated by the nitric oxide synthase are responsible for important aspects of antimicrobial activity. This cytotoxic effector mechanism is stimulated by cytokines such as IFN- and TNF- (28). The release of toxic nitrogen oxide radicals by macrophages facilitates the lysis of tumor targets (29), parasites such as leishmania (30, 31), T. gondii (32), and Schistosoma mansoni (33), or the pathogenic fungus Cryptococcus neoformans (34). The concentration of nitrogen oxides in the culture supernatant of AM from IFN--treated rats significantly exceeded that of control animals. Again, IM were less influenced by the cytokine instillation. However, even when using a low dose treatment protocol in order to modulate mainly the functions of AM, accompanying effects of the IFN- application on the pulmonary interstitium cannot be absolutely excluded.

Whereas our studies mainly focused on the augmentation of the microbicidal activity of AM, an activation of interstitial lung macrophages also could be observed. This activation might reflect both a direct effect of IFN- on these macrophages or a secondary effect due to cytokines secreted by activated AM or epithelial cells, which then induce activation of interstitial lung macrophages. To which extent one or the other pathway is responsible for the activation of interstitial lung macrophages has to be resolved in future studies.

Further, of importance is the fact that other macrophage populations like PM were not affected by the IFN- treatment. Therefore, the activation is just restricted to the lung macrophages. On the basis of these results, future clinical studies should reveal whether these data can be reproduced in patients. This therapeutic approach might be of importance for the treatment of infections of the lung when a systemic immunoactivation should be avoided.

We demonstrated a transient organ-specific enhancement of macrophage functions in the lung after local administration of IFN-. In addition, we could not observe any toxic side effects using IFN- amounts up to 5 × 10⁴ U/rat. Recently, an effective delivery of nebulized IFN- into the lungs of human volunteers was performed, resulting in an AM activation without clinical symtoms (35). This article underlines the feasibility of local IFN administrations for the treatment of pulmonary disease. Data were obtained in animal models suggesting beneficial effects of a local IFN- treatment during infections with Legionella pneumophila (36), in a murine model for metastasis formation in the lung (37), or in murine asthma models (38, 39).

Opportunistic infections of the lung are a major cause of morbidity for immunocompromised patients, e.g., transplant recipients (40, 41). Additionally, infections are known to support the development of rejection episodes after organ transplantation. Immunologic mechanisms contributing to the enhancing effect of rat CMV infection on subacute rejection could be identified in a rat model of lung transplantation (42). As AM are the first line of defense against a broad variety of pathogens, it could be beneficial to support their microbicidal functions in order to prevent bacterial or viral infections in high risk groups.

Having shown that microbicidal activities of AM are enhanced after local IFN- administration, we proved the in vivo relevance of these findings using a respiratory model infection. L. monocytogenes is a facultative intracellular parasite in macrophages and other cells. Immunologic mechanisms mediating host resistance against listerial infections are well established and depend in immunocompetent animals on macrophages activated in vivo by IFN- produced by T lymphocytes or natural killer cells. In immunosuppressed animals, the course of respiratory listerial infections is clearly aggravated as under these conditions the secretory functions of lymphocytes are dramatically reduced. To stimulate AM in the L. monocytogenes (LD80) infection model, the highest dose of the IFN- dose finding study was chosen to prove the efficacy of this approach. The resistance of immunosuppressed rats against local infections with L. monocytogenes was significantly restored by the local IFN- treatment.

In summary, we characterized the distinct functional properties of the two main populations of pulmonary macrophages isolated from the alveoli or the lung interstitium. In accordance with others (8), our data underline the microbicidal capacity of AM. Furthermore, our data demonstrate immunoregulatory functions of IM, indicating that IM are not solely an intermediate maturation stage of AM but contribute actively to inflammatory processes in the lung. Additionally, we analyzed AM and IM after local cytokine treatment, which augmented the local resistance against a respiratory bacterial infection. A local enhancement of nonspecific immune mechanisms, namely functions of pulmonary macrophages avoiding systemic side effects, could be a future option for the treatment of opportunistic infections in immunocompromised hosts.