American Journal of Respiratory Cell and Molecular Biology

Generalized underrepresentation of IFN-γ has been implicated in the development of allergic asthma. However, the role of local IFN-γ in the lung during the development of this disease has not been completely elucidated. We studied the influence of local pulmonary IFN-γ expression on the development of allergen-induced lung inflammation. To restrict our analysis to IFN-γ expression in the lung and to exclude influences of systemic IFN-γ production, we generated a transgenic mouse line with a targeted deletion of the IFN-γ gene and constitutive, lung-specific IFN-γ expression (Clara cell 10 [CC10]–IFN-γ–tg–IFN-γ–KO mice), and compared allergen-induced airway inflammation in these mice with that of wild-type and IFN-γ– KO mice on the C57BL/6 background. Cytokine quantification in lungs of mice with allergic airway inflammation revealed that pulmonary IFN-γ expression increased expression of IL-5 and IL-13. Consistent with this observation, eosinophilia in bronchoalveolar lavage of CC10–IFN-γ–tg–IFN-γ–KO mice was profoundly increased, indicating that this critical component of asthma is enhanced by local IFN-γ expression. In contrast, airway hyperresponsiveness and anti-ovalbumin-IgE serum levels were reduced by local IFN-γ expression. Together, our results demonstrate pleiotropic action of constitutive IFN-γ expression in the lung, and question the therapeutic value of IFN-γ in allergic asthma. Local expression of IFN-γ in the lung increases markers of allergic airway inflammation, but decreases airway hyperresponsiveness in a murine model of allergic-asthma

The incidence of bronchial asthma has increased dramatically over the past decades (1). This chronic inflammatory disease is characterized by hyperresponsiveness of the airways toward unspecific bronchoconstrictors, variable airflow obstruction, and profound infiltration of the airways by eosinophils, leading to their accumulation in bronchoalveolar lavage (BAL). Although the fundamental mechanisms underlying the development of the asthmatic state remain elusive, it is well appreciated that a dysregulated T helper (Th) cell type 2 response is critically involved (2). Experimental animal models have been established and proven useful for the analysis of the mechanisms that lead to allergic lung inflammation. In these models, asthma-like reactions are induced by systemic sensitization, and subsequent intranasal challenge of mice with a surrogate protein antigen (3, 4). These experiments reveal a critical involvement of the Th2 cytokines IL-4, IL-5, IL-13, and of Th2-associated IgE antibodies in development of allergen-induced responses (5). Although both IL-5 and IL-4 are of major importance for the development of inflammation, they play different roles. IL-5 recruits and activates eosinophils, and IL-4 promotes the Ig class switch toward IgE in plasma cells (57). The IL-4–related cytokine IL-13 has been specifically implicated in the development of airway hyperresponsiveness (AHR) (8). However, the exact mechanisms by which these cytokines induce asthma-associated airway alterations and AHR remain elusive.

Multiple causes have been described as being responsible for the development of asthma; in particular pulmonary infections, repeated allergen exposure, and genetic predisposition. At least 12 polymorphic genes have been linked to asthma development, including genes that control IgE, cytokine and chemokine production, and tissue remodeling. In the last two decades, incidences of asthma have nearly doubled in industrialized countries. Because the genetic composition of the population remained unchanged, improved hygiene was suggested as a major reason for this increase (9). Infections or exposure to components of microorganisms early in life could shape the immune system in a way that it is less prone to the development of allergies and asthma later in life (912). By enhancing Th1 immune reactions (e.g., through IFN-γ), these environmental factors could inhibit development of proallergic Th2 immune responses. Infections with or exposure to Mycobacterium tuberculosis have been implicated in the prevention of asthma development (13), and skin test reactivity to M. tuberculosis antigens is significantly lower in individuals with asthma as compared with healthy control subjects (14).

Studies in the mouse model support this concept (15, 16) and implicate a role of IFN-γ produced by mycobacteria-specific Th1 cells in reduction of allergic airway inflammation (17). A protective effect of IFN-γ is also observed when mice are treated with IFN-γ before or during induction of allergic lung inflammation (1822). As a consequence, IFN-γ treatment of patients with asthma is considered as a therapeutic approach. However, the mechanisms by which IFN-γ reduces asthmatic responses are unclear, and further research is needed to define the in vivo role of IFN-γ during development of this disease. Furthermore, there are also contradictory reports on the effects of IFN-γ or IFN-γ–inducing cytokines on the development of asthma-like reactions in mouse models (2326).

In the present study, the role of local IFN-γ expression for the development of allergen-induced lung inflammation was investigated. To restrict our analyses to IFN-γ expression in the lung and to exclude influences of systemic IFN-γ production, we generated a transgenic mouse line with constitutive, lung-specific, low IFN-γ expression and a targeted deletion of the IFN-γ gene (Clara cell 10 [CC10]–IFN-γ–tg–IFN-γ–KO mice). Using these mutant mice, we analyzed the inflammatory response after allergic sensitization and subsequent airway challenge and compared the response with that observed in IFN-γ–KO and wild-type control mice on a C57BL/6 background.


CC10–IFN-γ transgenic, C57BL/6, and IFN-γ–KO mice were bred in our facility at the Federal Institute for Health Protection of Consumers and Veterinary Medicine (BGVV) in Berlin. Mice were kept under specific pathogen–free conditions in filter bonnet cages with food and water ad libitum. Experiments were conducted according to German animal protection laws.

Generation of CC10–IFN-γ–tg–IFN-γ–KO Mice

Transgenic mice were generated, in which IFN-γ is solely, but constitutively, expressed in the lung. A construct was generated containing the CC10 promoter, the IFN-γ cDNA, a β-globin enhancer, and a bovine growth hormone polyadenylation sequence. After removal of prokaryotic vector sequences, the CC10-IFN-γ−pA construct was injected into IFN-γ–KO embryos (C57BL/6 background), yielding three transgenic founder animals, each giving rise to a different mouse line. The different lines were analyzed for specificity and abundance of IFN-γ mRNA expression, and the line with the most lung-specific (not the strongest) IFN-γ mRNA expression was further bred with IFN-γ–KO mice and used for all experiments. No significant differences were detected between heterozygous and homozygous transgenic animals.

Ovalbumin Sensitization and Challenge

Mice were sensitized and challenged with chicken ovalbumin (OVA) grade VII (Sigma-Aldrich, Taufkirchen, Germany) as previously described (27), with some modifications. Briefly, 6- to 8-wk-old CC10–IFN-γ–tg–IFN-γ–KO, C57BL/6, and IFN-γ–KO animals were immunized by intraperitoneal injection of 20 μg OVA emulsified in 100 μl Alum (Serva, Heidelberg, Germany) on Days 0 and 14. Where indicated, mice were challenged three times intranasally with 100 μg OVA in 25 μl endotoxin-free PBS or with PBS alone 14 d after the last sensitization on Days 28, 29, and 30. Mice were sacrificed and analyzed on Day 31, 24 h after the last challenge.

Bronchoalveolar Lavage and Cytospin

To collect BAL fluid (BALF), mice were killed by cervical dislocation, and the trachea was dissected free from surrounding soft tissue. A 0.6-mm canula was inserted into the trachea and held in place using a clamp. BALF was collected by perfusing the lung three times with 1 ml of PBS and gently aspirating the fluid back. The three washes were pooled, centrifuged, and cell numbers were determined. The cell-free BAL was stored at −80°C for further analysis. Cells were spun onto glass slides using a Cytospin (Shandon, Waltham, MA). Cells on the slide were stained with hematoxylin and eosin, and differential counts were performed microscopically using standard histologic criteria.

Histologic Analysis

Organs were embedded in tissue-tek (Shandon), snap frozen in liquid nitrogen, and subsequently kept at −80°C. Using a Cryostat (Shandon), 5-μm sections were cut, air dried, fixed with acetone for 10 min, and stained with hematoxylin and eosin. For each organ, multiple stained sections were evaluated microscopically.

mRNA Quantification

Total RNA was extracted from tissues using TRIzol reagent (Invitrogen, Carlsbad, CA) as recommended by the manufacturer. For Northern blotting, 10 μg RNA from each organ were separated by electrophoresis in a 1% agarose gel containing 0.4 M formaldehyde. The gel was denatured using 0.5 M ammonium acetate (Sigma), and RNA was blotted onto Hybond-N nylon membrane (Amersham Biosciences, Piscataway, NJ) by standard procedures. Blots were probed with randomly primed [α-32P]dCTP-labeled IFN-γ cDNA or murine β-actin cDNA. Hybridization was performed overnight at 65°C in a solution containing 0.5 M NaPO4 (pH 7.2), 7% SDS, 1 mM EDTA, 1% BSA, and 100 μg/ml salmon sperm DNA. Blots were washed at 65°C with 2× SSC, 0.5% SDS (twice for 30 min), and subsequently with 0.2× SSC, 0.5% SDS for 30 min. Autoradiography was done using Hyperfilm-MP films (Amersham Biosciences). Stripping of IFN-γ cDNA-probed membranes was done as recommended by the manufacturer. Equal loading of RNA was verified by reprobing the stripped blots with β-actin cDNA. The cDNA hybridization probes for IFN-γ and β-actin were prepared by RT-PCR from RNA isolated from C57BL/6 spleens using primers β-actin–fw/β-actin–rev and IFN-γ–68-fw/IFN-γ–425-rev.

For semiquantitative real-time RT-PCR, RNA from at least five similarly treated animals per groups was pooled and treated with DNase I (Invitrogen) to eliminate genomic DNA contamination. DNase I digested RNA (5 μg) were used for reverse transcription using 200 ng random hexamers as primers for Superscript II (Invitrogen) according to the manufacturer's recommendation. All real-time PCRs were run for 45 cycles with 15 s at 94°C and 60 s at 60°C in the ABI Prism 7700 Sequence Detection System (Applied Biosystems, Foster City, CA) using ABI PRISM optical 96-well plates (Applied Biosystems). When possible, primers were designed to span large introns and to produce product sizes between 100 and 200 bp (Table 1). Reaction mixtures were set up in 30 μl final volume using 15 pmol of each primer and 15 μl 2× SYBR-Green PCR Master mix (Applied Biosystems). Quantification was performed at least twice with RNA pools derived from independent experiments and in triplicates for each cDNA and primer pair. Data analysis was performed using the ABI Prism 7000 SDS software (Applied Biosystems) and Excel (Microsoft, Redmond, WA). Threshold cycles (Ct) were determined for each sample, and fold differences relative to the expression level in one of the analyzed cDNA samples was calculated for each cDNA sample and primer pair (fold difference = 2−ΔCt). Resulting fold differences for cytokines and eotaxin expression levels were corrected for different amounts of cDNA by multiplication with the average fold difference of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and β-actin expression within the same sample. DNase I digested, not reverse transcribed RNA, was used as template in separate reactions to control for genomic contamination.




Definition of abbreviations: GAPDH, glyceraldehyde-3-phosphate dehydrogenase; IIGP, IFN-inducible GTPase; IP-10, IFN-γ–inducible protein-10.

Quantification of OVA-Specific IgE Levels

OVA-specific IgE was measured by ELISA using standard procedures. Immuno-Maxisorp ELISA plates (Nunc, Wiesbaden, Germany) were coated with OVA, washed four times, and blocked with PBS containing 1% BSA. After washing, PBS-diluted samples were added to the wells and left overnight. The next day, plates were washed and the detection antibody anti-mouse IgE-horseradish peroxidase clone LO-ME3 (Biosource, Camarillo, CA) was added. After washing, phosphatase substrate (Sigma) in diethanolamine buffer was added, and the reaction was allowed to proceed in the dark before being stopped by the addition of EDTA (pH 8.0). Plates were read by a Spectramax ELISA reader (Molecular Devices, Sunnyvale, CA) and analyzed with Softmaxpro software. Control wells were left uncoated, filled with PBS only, or not treated with anti-mouse IgE.

Measurement of AHR

At 24 h after the last challenge, mouse lungs were prepared, ventilated, and perfused as previously described (27). Briefly, anesthetized mice were tracheotomized and ventilated. After sternotomy and cannulation of the left atrium and pulmonary artery, lungs were perfused with 37°C sterile Krebs-Henseleit hydroxyethylamylopectine buffer (1 ml · min−1) (Serag-Wiessner, Naila, Germany) in a nonrecirculating system, and left atrial pressure was adjusted at +2.2 cm H2O. Pulmonary arterial and left atrial pressure were monitored continuously. Upon isolation, lungs were ventilated by negative pressure (−4.5 to −9.0 cm H2O, 90 breaths · min−1) in a closed chamber. Hyperinflation (−24 cm H2O) was performed at 4-min intervals. The chamber pressure was continuously measured by a differential pressure transducer, and airflow velocity was monitored by means of a pneumotachograph connected to a second differential pressure transducer. All data were amplified and processed with Pulmodyn software (Hugo Sachs Elektronik-Harvard Apparatus, March-Hugstetten, Germany), and the data for airway resistance were analyzed as previously described (28). All hardware and software was purchased from HSE Harvard Apparatus (March-Hugstetten, Germany).

Each experiment consisted of an initial period of baseline measurement and three treatments, at 12-min intervals, with increasing doses of the unspecific bronchoconstrictor methacholine. From the data gathered during the baseline measurement, airway resistance and compliance were calculated for each analyzed lung and, to allow the comparison between lungs, the values were designated as 100% resistance or compliance of the respective lung. For each dose of methacholine, resistance and compliance were then determined in relation to the baseline level.

FACS Analysis and Intracellular Cytokine Staining

Lung lymphocytes were obtained by homogenizing the lung using an iron mash sieve and lysing red blood cells. After two washes in complete RPMI medium, cell suspensions were filtered through a 70-μm nylon sieve. To further purify these cells, a 40% /70% Percoll gradient was performed for 30 min at 600 × g. Intracellular cytokine staining was performed as previously described (29). Briefly, 2 × 106 cells were cultured for 5 h and were either stimulated with 3 μg anti-CD3 monoclonal antibody (mAb) and 5 μg anti-CD28 mAb, or left untreated. During the final 4 h of culture, 10 μg/ml of brefeldin A (Sigma) were added. After incubation, cells were washed, blocked with rat serum and anti-CD16/CD32 mAb, and stained with FITC-conjugated anti-CD4 mAb. Cells were then fixed at room temperature with PBS, 4% paraformaldehyde (Sigma), permeabilized with PBS containing 0.1% BSA, 0.5% saponin (Sigma), and incubated in this buffer with rat serum and anti-CD16/CD32 mAb. To stain intracellular cytokines, phycoerythrin (PE)-conjugated anti–IL-10 mAb and allophycocyanin (APC)-conjugated anti–IL-5 mAb or anti–IFN-γ mAb, or APC- and PE-conjugated isotype control mAb were added. To determine surface-expressed proteins, 2 × 106 cells were incubated in 200 μl PBS with fluorochrome-conjugated mAb for 20 min on ice, washed in PBS, and resuspended in 300 μl PBS containing 0.1% BSA. Stained cells were counted with a FACSCalibur (Becton Dickinson PharMingen, San Diego, CA), and the software FCS-Express (De Novo Software, Thornhill, ON, Canada) and Cell Quest (Becton Dickinson PharMingen) were used to analyze the data.

Statistical Analysis

In real-time RT-PCR experiments, standard deviation was calculated by addition of root mean square errors of GAPDH, β-actin, and cytokine measurements. Statistical significance of results was determined with the unpaired t test and one-way ANOVA, followed by a Tukey post hoc test included in the GraphPad Prism program (GraphPad Software, San Diego, CA).

Generation of Transgenic Mice

Transgenic IFN-γ KO mice that constitutively express IFN-γ under the lung epithelial cell-specific CC10 promoter (CC10–IFN-γ–tg–IFN-γ–KO mice) were generated. In these mice, IFN-γ mRNA expression is detected in the lung but not in spleen, liver, intestine, or kidney. Due to the IFN-γ KO background, the transgenically expressed IFN-γ in the lung is the only detectable IFN-γ mRNA in these mice (Figure 1A). Compared with lung of naive wild-type (C57BL/6) mice, IFN-γ mRNA expression was ∼ 15-fold higher in lungs of CC10–IFN-γ–tg–IFN-γ–KO mice (Figure 1B). Despite the relatively strong expression in comparison to naive C57BL/6 (wild-type) mice, pulmonary IFN-γ expression in CC10–IFN-γ–tg–IFN-γ–KO mice was far lower than that observed in M. tuberculosis–infected C57BL/6 animals (data not shown). In fact, IFN-γ protein expression was below 30 pg/ml, and was not detectable in BALF or serum. However, IFN-γ–induced genes, like IFN-γ–inducible protein-10 (IP-10) and IFN-inducible GTPase (IIGP), were upregulated in lung, demonstrating that IFN-γ mRNA produced from the transgene was translated into active protein. Histologic evaluation of naive CC10–IFN-γ–tg–IFN-γ–KO lungs and analysis of the cellular content of lung lavage fluid did not reveal pathologic aberrations compared with nontransgenic littermates, indicating that, at low IFN-γ levels, normal lung development was not affected (data not shown).

Effect of Pulmonary IFN-γ on OVA-Specific IgE Responses after Sensitization and Challenge

To determine whether systemic antigen sensitization could be accomplished in CC10–IFN-γ–tg–IFN-γ–KO mice, transgene+ and nontransgenic littermates were immunized intraperitoneally with OVA plus alum or with alum alone on Days 0 and 14. On Day 31, sera were collected and OVA-specific IgE was quantified. In the absence of antigen exposure, OVA-specific IgE was not detectable. In contrast, OVA-specific IgE was readily detected in sensitized mice. Interestingly, levels of OVA-specific IgE were comparable only between sensitized transgene+ and nontransgenic littermates (or IFN-γ–KO mice). Moreover, serum IgE levels were comparable only between sensitized CC10–IFN-γ–tg–IFN-γ–KO and C57BL/6 mice (Figure 2A), indicating that systemic sensitization was not affected by lung-specific IFN-γ expression.

To assess the effects of locally produced IFN-γ on OVA-specific IgE during an asthma-like response, groups of OVA-sensitized CC10–IFN-γ–tg–IFN-γ–KO, C57BL/6, and IFN-γ–KO animals were subjected to 3 consecutive intranasal OVA challenges on Days 28, 29, and 30. Twenty-four hours after the last challenge, OVA-specific IgE concentrations in serum and lung homogenates were quantified (Figure 2). Although C57BL/6 and IFN-γ–KO (nontransgenic littermates) animals demonstrated comparable IgE levels, OVA-specific IgE was reduced in CC10–IFN-γ–tg–IFN-γ–KO mice as compared with other groups, indicating that lung-specific IFN-γ expression downmodulated antigen-specific IgE secretion in sensitized and challenged mice.

Effect of Pulmonary IFN-γ on BAL Eosinophilia

Cellular composition of BALF from naive, sensitized, and OVA- or saline-challenged animals was determined 24 h after the last challenge. Cells in BALF from each animal were spun onto glass slides, stained, and counted microscopically using standard histologic criteria. In naive, sensitized, and saline-challenged animals, macrophages made up ∼ 99% of all cells in BALF (data not shown). Eosinophils and neutrophils were virtually absent, with a proportion of 0–2% and 0–1%, respectively. No significant differences were detected between CC10–IFN-γ–tg–IFN-γ–KO mice and non-transgenic littermates or C57BL/6 controls (data not shown).

A profound increase in absolute cell numbers recovered from BAL was observed in all mice that had been OVA sensitized and challenged compared with naive, only sensitized, or saline-challenged animals (0.68 × 105 ± 0.03 × 105 cells versus 5.2 × 105 ± 0.45 × 105 cells). Among all mice, CC10–IFN-γ–tg–IFN-γ–KO mice showed the strongest increase. In all OVA-challenged groups, numbers of macrophages increased only 2-fold (0.61 × 105 ± 0.06 × 105 cells versus 1.1 × 105 ± 0.16 × 105 cells), whereas numbers of eosinophils and neutrophils within BAL increased more than 100-fold (0.73 × 103 ± 0.31 × 103 cells versus 365.9 × 103 ± 121.8 × 103 cells), demonstrating that CC10–IFN-γ–tg–IFN-γ–KO animals were able to mount an allergic inflammatory response. CC10–IFN-γ–tg–IFN-γ–KO mice showed a higher proportion and absolute number of eosinophils as compared with C57BL/6 mice (P = 0.07) and IFN-γ–KO mice (P = 0.015) (Figure 3). The proportion and number of neutrophils in BAL of sensitized and challenged mice did not differ significantly between the three groups (Figure 3). These data demonstrate that lung-specific IFN-γ expression in an IFN-γ–KO animal increases BAL eosinophilia and airway inflammation after aeroallergen challenge.

Effect of Pulmonary IFN-γ on Th2 Cytokine Levels

To investigate the underlying mechanisms that lead to increased eosinophilia and lower IgE levels in CC10–IFN-γ–tg–IFN-γ–KO animals, mRNA expression levels of Th2 cytokines and other cytokines involved in inflammation were measured. RNA was extracted from lungs of naive mice or from mice 24 h after the last challenge. RNA from animals of the same group was pooled and reverse transcribed to cDNA. With these cDNA, semiquantitative real-time PCR was performed using primers for β-actin and GAPDH as controls for equal amounts of cDNA and intron-spanning primer pairs, specific for cytokine cDNA. Cytokine mRNA levels were calculated relative to the expression levels of the analyzed cytokine in lungs of naive C57BL/6 mice (Figure 4).

IL-4, one of the major cytokines of Th2 immune responses, was strongly upregulated in lungs of all sensitized and challenged mice (Figure 4A). Sensitized and challenged CC10–IFN-γ–tg–IFN-γ–KO animals showed the lowest levels of IL-4 mRNA, whereas similarly treated IFN-γ–KO animals exhibited the highest levels. In contrast to IL-4 levels, but in accordance with the stronger BAL eosinophilia, IL-5 levels were highest in CC10–IFN-γ–tg–IFN-γ–KO animals, followed by those in IFN-γ–KO animals (Figure 4B). Of all sensitized and challenged groups, C57BL/6 mice displayed the lowest levels of IL-5 mRNA. Eotaxin expression in lungs of challenged mice corresponded to that of IL-5 mRNA (Figure 4C).

IL-13, the presumed central mediator of murine allergic airway inflammation, was virtually undetectable in lungs of naive animals, but highly upregulated in sensitized and challenged animals (Figure 4D). Despite these high levels, the expression patterns in the three groups of mice were almost comparable to those measured for IL-5. Sensitized and challenged CC10–IFN-γ–tg–IFN-γ–KO animals had almost twice as much IL-13 mRNA in their lungs as similarly treated IFN-γ–KO animals, and threefold more than C57BL/6 animals.

IL-10, which downregulates both Th1- and Th2-driven inflammatory processes, was upregulated in lungs of all sensitized and challenged animals (Figure 4E); however, there was no difference in IL-10 expression between the three groups of mice. In contrast to the other cytokines analyzed, slightly elevated levels of IL-10 were observed in lungs of naive CC10–IFN-γ–tg–IFN-γ–KO and IFN-γ–KO animals, possibly reflecting compensatory effects of unregulated IFN-γ expression and the total absence of IFN-γ.

To determine whether IFN-γ levels in lungs of C57BL/6 mice changed during allergic airway inflammation, real-time RT-PCR was performed. Surprisingly, IFN-γ mRNA levels were upregulated in lungs of challenged C57BL/6 mice as compared with untreated naive and sensitized controls (Figure 4F). As expected, IFN-γ levels did not differ between lungs of challenged and naive control CC10–IFN-γ–tg–IFN-γ–KO mice, because IFN-γ is constitutively produced in these transgenic mice.

Increased IL-10 mRNA in naive transgene+ and transgene IFN-γ–KO animals corresponded with increased numbers and proportions of CD4+ IL-10+ T cells in the lung, as determined by intracellular cytokine staining (data not shown). In contrast, differential IL-5 expression between the three groups was not reflected by increased numbers or proportions of IL-5+ CD4+ T cells (data not shown).

These results suggest that lung-specific IFN-γ expression enhances IL-5 and IL-13 responses during an asthma-like reaction. However, enhanced IL-5 expression is not correlated with enhanced numbers of IL-5+ CD4+ T cells. Moreover, upregulation of IFN-γ in C57BL/6 mice after challenge suggests involvement of this cytokine in the development of allergic airway inflammation in wild-type animals.

Effects of Pulmonary IFN-γ on AHR

To determine the effects of exclusive lung-specific expression of IFN-γ on airway responsiveness, measurement of AHR was performed in perfused lungs using increasing doses of intravascular methacholine. Airway resistance and dynamic compliance were monitored continuously, and the methacholine-induced changes were determined relative to baseline levels. In the first experiment, we studied AHR induced by OVA sensitization and challenge. Lungs of naive, sensitized, and saline- or OVA-challenged animals were successively subjected to 1 × 10−6 M, 5 × 10−6 M, and 1 × 10−5 M methacholine. Although naive and saline-challenged lungs of all groups displayed minute reactions to 1 × 10−6 M methacholine (airway resistance 105 ± 5%), OVA-sensitized and -challenged lungs reacted with massive bronchoconstriction (airway resistance > 250%). To compare airway responsiveness between OVA-sensitized and -challenged C57BL/6, CC10–IFN-γ–tg–IFN-γ–KO and IFN-γ–KO mice, lungs of five animals in each group were subjected to increasing doses of methacholine. Surprisingly, airway responses to all analyzed methacholine concentrations were considerably weaker in CC10–IFN-γ–tg–IFN-γ–KO mice than in C57BL/6 and IFN-γ–KO mice, indicating a lower AHR in these animals (Figure 5). The less pronounced AHR in CC10–IFN-γ–tg–IFN-γ–KO mice indicates a protective effect of IFN-γ on asthma-like airway responses in these animals.

The role of Th1-type responses in asthma is of great interest, as there are several proposed therapy schemes aimed at enhancing Th1 inflammatory responses to reduce allergic airway inflammation. Specifically, underrepresentation of IFN-γ has been implicated in the development of asthma bronchiale (17, 30). However, the role of Th1-type responses during the development of this allergic disease is controversial, as conflicting data suggest that Th1-type responses can both augment and attenuate allergic inflammation (3135).

We studied the influence of local IFN-γ on the development of allergen-induced airway inflammation. To focus our analyses on effects of pulmonary IFN-γ, and to exclude influences of systemic IFN-γ, we generated a transgenic mouse line with constitutive, lung-specific IFN-γ expression and a targeted deletion of the IFN-γ gene (CC10–IFN-γ–tg–IFN-γ–KO mice). In these animals, IFN-γ is exclusively expressed under the control of the CC10 promoter, which restricts expression to the lung (36). Consequently, IFN-γ expression cannot be regulated, allowing confined analysis of local IFN-γ influences on asthma development.

Pulmonary IFN-γ expression in CC10–IFN-γ–tg–IFN-γ–KO animals was relatively low compared with local IFN-γ abundance in M. tuberculosis–infected C57BL/6 animals (unpublished observation), but high compared with naive C57BL/6 animals. In contrast to previously generated transgenic animals with inducible pulmonary IFN-γ expression (36), our animals displayed normal lung development without tissue inflammation or increased cellular infiltration, indicating that IFN-γ levels were sufficiently low to not interfere with normal lung development.

To determine in vivo functions of pulmonary IFN-γ expression during allergen-induced acute lung inflammation, we compared the inflammatory response of CC10–IFN-γ–tg–IFN-γ–KO mice to that of IFN-γ–KO mice and C57BL/6 mice using a short-term model of allergic sensitization and challenge.

In accordance with previous studies (26, 37), we detected increased IFN-γ levels in sensitized and challenged mice compared with naive C57BL/6 mice. The role of these low amounts of IFN-γ remains unclear, as we did not detect significant differences between sensitized and challenged IFN-γ–KO and C57BL/6 mice in OVA-specific IgE, AHR, and infiltration of eosinophils into the lung. Similarities in the reaction of C57BL/6 and IFN-γ–KO mice are not surprising, as a major part of the inflammatory response is considered to be Th2 cytokine–mediated in this model of acute asthma. However, our results contrast with previous reports of IFN-γ–KO or IFN-γR–KO mice, which developed a stronger anti-OVA IgE response (18), or increased BAL eosinophilia (22). These studies used Th2-predisposed BALB/c mice, 129SvJ mice, or different sensitization and challenge protocols. In BALB/c mice, IFN-γ is probably a more important opponent of the Th2 response, thereby reducing effects of allergic sensitization and challenge and promoting stronger inflammatory responses in IFN-γ–KO or IFN-γR–KO mice on BALB/c background. Thus, differences in mouse strains and experimental protocols used for sensitization, challenge, and analysis could be responsible for the discrepancy between our observations and those of other workers.

Comparison of sensitized and challenged CC10–IFN-γ–tg–IFN-γ–KO mice to similarly treated IFN-γ–KO or C57BL/6 mice revealed that lung-specific IFN-γ expression affected BAL eosinophilia and AHR differentially. Although OVA-induced BAL eosinophilia was considerably higher, AHR to the unspecific bronchoconstrictor methacholine was significantly lower in CC10–IFN-γ–tg–IFN-γ–KO mice compared with control C57BL/6 and IFN-γ–KO mice. These diverse effects of constitutive lung-specific IFN-γ expression are also reflected in differential cytokine, eotaxin, and IgE levels in lungs of OVA-sensitized and -challenged CC10–IFN-γ–tg–IFN-γ–KO mice compared with control animals. In line with augmented eosinophilia, eotaxin and IL-5 mRNA levels were increased, whereas OVA-specific IgE and, to a lesser degree, IL-4 mRNA were decreased as compared with equally treated C57BL/6 and IFN-γ–KO mice.

The divergent effects of lung-specific IFN-γ expression partly contrast with previous findings, which describe decreased BAL eosinophilia accompanied by lower AHR after IFN-γ treatment (20, 22, 31, 38). The reasons for these contrasting findings regarding BAL eosinophilia remain unclear. However, apart from mouse-strain differences, some of the studies applied systemic IFN-γ treatment during the phase of sensitization or airway challenge, whereas, in our model, IFN-γ expression was constitutive and restricted to the lung. Different effects of systemic versus local treatment were described previously. Hofstra and colleagues demonstrated that parenteral IFN-γ treatment of C57BL/6 mice downregulated OVA-specific IgE, BAL eosinophilia, and AHR, whereas aerosolized IFN-γ treatment only suppressed AHR (32).

It is not clear how local IFN-γ enhanced BAL eosinophilia in our model, but this effect has been recognized previously in studies analyzing the role of a Th1-inducing respiratory tract infection (33) and allergen-specific Th1 cells (34). Hansen and colleagues found that adoptive transfer of allergen-specific Th1 cells into naive recipients increased airway inflammation after sensitization and challenge (34). In agreement with this observation, IFN-γ produced during an influenza infection leads to augmented inflammation with increased eosinophilia in a postinfection acute asthma model. Dahl and colleagues proposed that this effect is mediated by a modification of local dendritic cells, which could also happen in our transgenic animals (33).

Lung-specific IFN-γ expression lowered AHR in our model of acute asthma. However, treatment with anti–IFN-γ antibodies during the challenge period abolished AHR induction in a model of chronic asthma, indicating the opposite role for IFN-γ in development of AHR in such models (23, 35). This difference is not surprising: models of chronic asthma seem to involve distinctly different mechanisms of AHR development as compared with models of acute asthma, given that mice impaired in their Th2 cytokine signaling or deficient for IL-4 and IL-13 can only develop AHR in models of chronic as opposed to models of acute asthma (reviewed in Ref. 39).

Consistent with the idea that AHR in models of acute asthma is Th2 cytokine dependent, the lower IL-4 mRNA levels in lungs of OVA-sensitized and -challenged CC10–IFN-γ–tg–IFN-γ–KO mice may contribute to the lower degree of AHR in these mice, in spite of increased IL-5 levels, as previously described (40). Corry and colleagues showed that anti–IL-4 mAb treatment reduces AHR without affecting airway eosinophilia, whereas anti–IL-5 mAb treatment did not alter AHR but reduced airway eosinophilia (40).

Because IL-4 strongly promotes class switching in plasma cells, it is possible that lower levels of IL-4 are responsible for the decreased anti-OVA IgE levels in sera and lung homogenates of the transgenic mice (41). IgE binds to high-affinity Fc-ϵI receptors on granulocytes and mast cells and promotes discharge of proinflammatory mediators, such as histamine and leukotrienes, from these cells, and subsequently the induction of AHR (42).

A plausible scenario for the reduced IL-4 and IgE expression in CC10–IFN-γ–tg–IFN-γ–KO mice is that the abundance of IFN-γ in the lungs of CC10–IFN-γ–tg–IFN-γ–KO mice was sufficiently high to reach the draining lymph nodes, where T cell differentiation and IgE synthesis take place after challenge (43). It is feasible that lung-specific IFN-γ could have reduced the induction and proliferation of Th2 cells. However, IL-13 mRNA levels were increased in allergen-sensitized and -challenged CC10–IFN-γ–tg–IFN-γ–KO animals compared with equally treated C57BL/6 and IFN-γ−KO control mice. The abundance of IL-13 in lungs of challenged CC10–IFN-γ–tg–IFN-γ–KO mice was unexpected, as IL-13 has been shown to induce AHR (4446) and CC10–IFN-γ–tg–IFN-γ–KO mice displayed lower AHR than did control animals. Possibly, IFN-γ itself may lead to lower AHR by inducing nitric oxide production, which could counteract contraction of airway smooth muscle (47) independently of IL-13–mediated AHR. Note that AHR in challenged CC10–IFN-γ–tg–IFN-γ–KO mice was not abolished. These mice were still more sensitive to low doses of methacholine than were naive control animals, indicating that AHR was still induced, possibly by IL-13.

Further studies are needed to clarify the exact proinflammatory and antiinflammatory activities of lung-specific IFN-γ in the development of allergic asthma. It remains unclear how AHR develops and how IFN-γ in the lung influences AHR. It also remains to be elucidated how pulmonary IFN-γ expression can lead to increased eotaxin, IL-5, and IL-13 levels. One possible mediator of IFN-γ effects is IP-10 (CXCL10), which was upregulated in lungs of CC10–IFN-γ–tg–IFN-γ–KO mice. IP-10 is strongly induced by IFN-γ (48), and the results of studies dissecting the role of IP-10 in allergic asthma are consistent with our observations (25, 37). Treatment with exogenous IP-10 at the time of allergen challenge increased eosinophilia and decreased AHR for 24 h after the last challenge (25). In this line, Medoff and colleagues showed that IP-10 induces eosinophilia in transgenic mice, although accompanied by increased AHR in their model (37).

In sum, IFN-γ appears to express both proinflammatory and antiinflammatory activities, depending on the local concentration and timing of expression or application in the development of allergic asthma. Our results demonstrate pleiotropic activities of constitutive lung-specific IFN-γ expression in a model of acute allergic asthma. Local IFN-γ differentially modulates allergen-induced eosinophilic inflammation, promotes expression of a Th2 cytokine subset, and reduces AHR. These results indicate that IFN-γ contributes to the pathogenesis of asthma instead of counterbalancing a detrimental Th2 response, as proposed by the “hygiene hypothesis.” Therefore, our study questions the therapeutic value of local IFN-γ in allergic asthma.

The authors thank Mischo Kursar, Hans-Willi Mittrücker, Uwe Klemm, Manuela Primke, and the staff of the animal facility at the Max-Planck-Institute for Infection Biology for their help in facilitating these experiments.

1. Bach JF. The effect of infections on susceptibility to autoimmune and allergic diseases. N Engl J Med 2002;347:911–920.
2. Romagnani S. Immunologic influences on allergy and the TH1/TH2 balance. J Allergy Clin Immunol 2004;113:395–400.
3. Wills-Karp M. Murine models of asthma in understanding immune dysregulation in human asthma. Immunopharmacology 2000;48:263–268.
4. Leong KP, Huston DP. Understanding the pathogenesis of allergic asthma using mouse models. Ann Allergy Asthma Immunol 2001;87:96–109; quiz 110.
5. Kips JC. Cytokines in asthma. Eur Respir J Suppl 2001;34:24s–33s.
6. Foster PS, Hogan SP, Ramsay AJ, Matthaei KI, Young IG. Interleukin 5 deficiency abolishes eosinophilia, airways hyperreactivity, and lung damage in a mouse asthma model. J Exp Med 1996;183:195–201.
7. Finkelman FD, Katona IM, Urban JF Jr, Holmes J, Ohara J, Tung AS, Sample JV, Paul WE. IL-4 is required to generate and sustain in vivo IgE responses. J Immunol 1988;141:2335–2341.
8. Wills-Karp M, Luyimbazi J, Xu X, Schofield B, Neben TY, Karp CL, Donaldson DD. Interleukin-13: central mediator of allergic asthma. Science 1998;282:2258–2261.
9. Romagnani S. The increased prevalence of allergy and the hygiene hypothesis: missing immune deviation, reduced immune suppression, or both? Immunology 2004;112:352–363.
10. McKeever TM, Lewis SA, Smith C, Collins J, Heatlie H, Frischer M, Hubbard R. Early exposure to infections and antibiotics and the incidence of allergic disease: a birth cohort study with the West Midlands General Practice Research Database. J Allergy Clin Immunol 2002;109:43–50.
11. Riedler J, Braun-Fahrlander C, Eder W, Schreuer M, Waser M, Maisch S, Carr D, Schierl R, Nowak D, von Mutius E. Exposure to farming in early life and development of asthma and allergy: a cross-sectional survey. Lancet 2001;358:1129–1133.
12. Braun-Fahrlander C, Riedler J, Herz U, Eder W, Waser M, Grize L, Maisch S, Carr D, Gerlach F, Bufe A, et al. Environmental exposure to endotoxin and its relation to asthma in school-age children. N Engl J Med 2002;347:869–877.
13. von Hertzen L, Klaukka T, Mattila H, Haahtela T. Mycobacterium tuberculosis infection and the subsequent development of asthma and allergic conditions. J Allergy Clin Immunol 1999;104:1211–1214.
14. Shirakawa T, Enomoto T, Shimazu S, Hopkin JM. The inverse association between tuberculin responses and atopic disorder. Science 1997;275:77–79.
15. Major T, Wohlleben G, Reibetanz B, Erb KJ. Application of heat killed Mycobacterium bovis-BCG into the lung inhibits the development of allergen-induced Th2 responses. Vaccine 2002;20:1532–1540.
16. Herz U, Gerhold K, Gruber C, Braun A, Wahn U, Renz H, Paul K. BCG infection suppresses allergic sensitization and development of increased airway reactivity in an animal model. J Allergy Clin Immunol 1998;102:867–874.
17. Erb KJ, Holloway JW, Sobeck A, Moll H, Le Gros G. Infection of mice with Mycobacterium bovis-Bacillus Calmette-Guerin (BCG) suppresses allergen-induced airway eosinophilia. J Exp Med 1998;187:561–569.
18. Coyle AJ, Tsuyuki S, Bertrand C, Huang S, Aguet M, Alkan SS, Anderson GP. Mice lacking the IFN-γ receptor have impaired ability to resolve a lung eosinophilic inflammatory response associated with a prolonged capacity of T cells to exhibit a Th2 cytokine profile. J Immunol 1996;156:2680–2685.
19. Flaishon L, Topilski I, Shoseyov D, Hershkoviz R, Fireman E, Levo Y, Marmor S, Shachar I. Cutting edge: anti-inflammatory properties of low levels of IFN-γ. J Immunol 2002;168:3707–3711.
20. Li XM, Chopra RK, Chou TY, Schofield BH, Wills-Karp M, Huang SK. Mucosal IFN-γ gene transfer inhibits pulmonary allergic responses in mice. J Immunol 1996;157:3216–3219.
21. Rais M, Wild JS, Choudhury BK, Alam R, Stafford S, Dharajiya N, Sur S. Interleukin-12 inhibits eosinophil differentiation from bone marrow stem cells in an interferon-γ–dependent manner in a mouse model of asthma. Clin Exp Allergy 2002;32:627–632.
22. Yoshida M, Leigh R, Matsumoto K, Wattie J, Ellis R, O'Byrne PM, Inman MD. Effect of interferon-γ on allergic airway responses in interferon-γ–deficient mice. Am J Respir Crit Care Med 2002;166:451–456.
23. Hessel EM, Van Oosterhout AJ, Van Ark I, Van Esch B, Hofman G, Van Loveren H, Savelkoul HF, Nijkamp FP. Development of airway hyperresponsiveness is dependent on interferon-γ and independent of eosinophil infiltration. Am J Respir Cell Mol Biol 1997;16:325–334.
24. Kumano K, Nakao A, Nakajima H, Hayashi F, Kurimoto M, Okamura H, Saito Y, Iwamoto I. Interleukin-18 enhances antigen-induced eosinophil recruitment into the mouse airways. Am J Respir Crit Care Med 1999;160:873–878.
25. Thomas MS, Kunkel SL, Lukacs NW. Differential role of IFN-γ–inducible protein 10 kDa in a cockroach antigen–induced model of allergic airway hyperreactivity: systemic versus local effects. J Immunol 2002;169:7045–7053.
26. Wild JS, Sigounas A, Sur N, Siddiqui MS, Alam R, Kurimoto M, Sur S. IFN-γ–inducing factor (IL-18) increases allergic sensitization, serum IgE, Th2 cytokines, and airway eosinophilia in a mouse model of allergic asthma. J Immunol 2000;164:2701–2710.
27. von Bethmann AN, Brasch F, Nusing R, Vogt K, Volk HD, Muller KM, Wendel A, Uhlig S. Hyperventilation induces release of cytokines from perfused mouse lung. Am J Respir Crit Care Med 1998;157:263–272.
28. Held HD, Uhlig S. Basal lung mechanics and airway and pulmonary vascular responsiveness in different inbred mouse strains. J Appl Physiol 2000;88:2192–2198.
29. Kursar M, Bonhagen K, Kohler A, Kamradt T, Kaufmann SH, Mittrucker HW. Organ-specific CD4+ T cell response during Listeria monocytogenes infection. J Immunol 2002;168:6382–6387.
30. Walzl G, Humphreys IR, Marshall BG, Edwards L, Openshaw PJ, Shaw RJ, Hussell T. Prior exposure to live Mycobacterium bovis BCG decreases Cryptococcus neoformans–induced lung eosinophilia in a γ interferon–dependent manner. Infect Immun 2003;71:3384–3391.
31. Lack G, Bradley KL, Hamelmann E, Renz H, Loader J, Leung DY, Larsen G, Gelfand EW. Nebulized IFN-γ inhibits the development of secondary allergic responses in mice. J Immunol 1996;157:1432–1439.
32. Hofstra CL, Van Ark I, Hofman G, Nijkamp FP, Jardieu PM, Van Oosterhout AJ. Differential effects of endogenous and exogenous interferon-γ on immunoglobulin E, cellular infiltration, and airway responsiveness in a murine model of allergic asthma. Am J Respir Cell Mol Biol 1998;19:826–835.
33. Dahl ME, Dabbagh K, Liggitt D, Kim S, Lewis DB. Viral-induced T helper type 1 responses enhance allergic disease by effects on lung dendritic cells. Nat Immunol 2004;5:337–343.
34. Hansen G, Berry G, DeKruyff RH, Umetsu DT. Allergen-specific Th1 cells fail to counterbalance Th2 cell-induced airway hyperreactivity but cause severe airway inflammation. J Clin Invest 1999;103:175–183.
35. Kumar RK, Herbert C, Webb DC, Li L, Foster PS. Effects of anticytokine therapy in a mouse model of chronic asthma. Am J Respir Crit Care Med 2004;170:1043–1048.
36. Wang Z, Zheng T, Zhu Z, Homer RJ, Riese RJ, Chapman HA Jr, Shapiro SD, Elias JA. Interferon γ induction of pulmonary emphysema in the adult murine lung. J Exp Med 2000;192:1587–1600.
37. Medoff BD, Sauty A, Tager AM, Maclean JA, Smith RN, Mathew A, Dufour JH, Luster AD. IFN-γ–inducible protein 10 (CXCL10) contributes to airway hyperreactivity and airway inflammation in a mouse model of asthma. J Immunol 2002;168:5278–5286.
38. Iwamoto I, Nakajima H, Endo H, Yoshida S. Interferon γ regulates antigen-induced eosinophil recruitment into the mouse airways by inhibiting the infiltration of CD4+ T cells. J Exp Med 1993;177:573–576.
39. Kumar RK, Foster PS. Modeling allergic asthma in mice: pitfalls and opportunities. Am J Respir Cell Mol Biol 2002;27:267–272.
40. Corry DB, Folkesson HG, Warnock ML, Erle DJ, Matthay MA, Wiener-Kronish JP, Locksley RM. Interleukin 4, but not interleukin 5 or eosinophils, is required in a murine model of acute airway hyperreactivity. J Exp Med 1996;183:109–117.
41. Zhou CY, Crocker IC, Koenig G, Romero FA, Townley RG. Anti–interleukin-4 inhibits immunoglobulin E production in a murine model of atopic asthma. J Asthma 1997;34:195–201.
42. Holgate ST, Hardy C, Robinson C, Agius RM, Howarth PH. The mast cell as a primary effector cell in the pathogenesis of asthma. J Allergy Clin Immunol 1986;77:274–282.
43. Chvatchko Y, Kosco-Vilbois MH, Herren S, Lefort J, Bonnefoy JY. Germinal center formation and local immunoglobulin E (IgE) production in the lung after an airway antigenic challenge. J Exp Med 1996;184:2353–2360.
44. Kuperman DA, Huang X, Koth LL, Chang GH, Dolganov GM, Zhu Z, Elias JA, Sheppard D, Erle DJ. Direct effects of interleukin-13 on epithelial cells cause airway hyperreactivity and mucus overproduction in asthma. Nat Med 2002;8:885–889.
45. Walter DM, McIntire JJ, Berry G, McKenzie AN, Donaldson DD, DeKruyff RH, Umetsu DT. Critical role for IL-13 in the development of allergen-induced airway hyperreactivity. J Immunol 2001;167:4668–4675.
46. Zhu Z, Homer RJ, Wang Z, Chen Q, Geba GP, Wang J, Zhang Y, Elias JA. Pulmonary expression of interleukin-13 causes inflammation, mucus hypersecretion, subepithelial fibrosis, physiologic abnormalities, and eotaxin production. J Clin Invest 1999;103:779–788.
47. Barnes PJ, Liew FY. Nitric oxide and asthmatic inflammation. Immunol Today 1995;16:128–130.
48. Ohmori Y, Hamilton TA. Cooperative interaction between interferon (IFN) stimulus response element and κ B sequence motifs controls IFN γ– and lipopolysaccharide-stimulated transcription from the murine IP-10 promoter. J Biol Chem 1993;268:6677–6688.
Correspondence and requests for reprints should be addressed to Markus Koch, Department of Immunology, Max-Planck-Institute for Infection Biology, Campus Charité Mitte, Schumannstrasse 21/22, 10117 Berlin, Germany. E-mail:


No related items
American Journal of Respiratory Cell and Molecular Biology

Click to see any corrections or updates and to confirm this is the authentic version of record