Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Respiratory syncytial virus-induced bronchiolitis in infants less than 1 yr of age has been reported to be an important risk factor for the development of asthma (1). Although it is not known why some infants go on to develop asthma after bronchiolitis while others fully recover, there is evidence that decreased in vitro interferon (IFN)- production by peripheral blood mononuclear cells obtained from infants during virus-induced bronchiolitis is associated with these developments (4). Morever, the fact that diminished IFN- production can be demonstrated in cord blood mononuclear cells in infants who are at increased risk of developing atopic diseases, including asthma, suggests that the propensity toward cytokine dysregulation is genetically determined and/or developmentally regulated (5). Collectively, these observations suggest that genetic (cytokine dysregulation), environmental (viral respiratory infections), and developmental (in relationship to the development of the immune system and/or the lung) factors may contribute to the inception of childhood asthma.

To investigate these interrelationships, a rat model that parallels the human asthmatic phenotype in several respects has been developed (8). After parainfluenza type 1 (Sendai) virus infection at an early age, high immunoglobulin (Ig) E responder "atopic" Brown Norway (BN) rats, but not low IgE responder "nonatopic" Fischer 344 (F344) rats, develop an asthma-like phenotype characterized by episodic reversible airway obstruction, airway hyperresponsiveness to methacholine, chronic airway inflammation, and airway wall remodeling. The development of this postviral asthma-like phenotype resembles the origins of asthma in children in that genetic (strain differences), environmental (paramyxovirus infection), and developmental (age at time of infection) factors contribute significantly to the observed histopathologic and physiologic outcomes.

In addition, the susceptibility and resistance of the BN and F344 strains, respectively, to the development of postviral chronic airway dysfunction appear to be associated with differences in the host cytokine response to Sendai virus infection: BN weanlings differ from F344 weanlings by exhibiting greater interleukin (IL)-4 and IL-5 expression and less IFN- production during the acute viral illness (11). Further, treatment of BN weanlings with aerosolized IFN- during the acute infection prevents the development of the chronic airway sequelae (10). As with preliminary observations in humans, these results are consistent with the hypothesis that reduced IFN- production during the acute viral illness is an important risk factor in the subsequent development of the asthma-like phenotype.

Given the strain differences in IFN- production during Sendai virus infection, we hypothesized that cells from the immune system of weanling BN rats may have a reduced capacity to secrete IFN- in comparison with cells from F344 weanlings. Natural killer (NK) cells are likely to be an important source of IFN- during the early innate host response to viral infection (12). Innate responses to viral infection may be especially important in naive hosts because there will be a distinct lag period before the adaptive immune response becomes effective. Two other respiratory viruses, rhinovirus and influenza A virus, have been shown to induce IFN- production indirectly, via an innate mechanism, by eliciting the secretion of IFN--inducing cytokines from monocytes and macrophages, respectively (13, 14). IL-12, IL-18, and IFN- are known inducers of IFN- expression, and monocytes and macrophages are among the cell types that produce these cytokines (15). Therefore, to extensively compare IFN- response profiles between the susceptible and resistant rat strains, we evaluated postinfection bronchoalveolar lavage fluid (BALF) and the responses of splenocytes or purified NK cells to a number of stimuli, either alone or in combination, including Sendai virus, IL-12, IL-18, and IFN-. Our results demonstrate the novel finding that in rats susceptible to the development of postviral chronic airway dysfunction, NK cells, which participate in innate inflammatory responses, exhibit reduced IFN- production.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Male pathogen-free inbred BN and F344 rats were purchased from Charles River Breeding Laboratories (Raleigh, NC) as 3-wk-old weanlings, and they were approximately 23 to 25 d of age at the time of the experiments. For viral studies, the animals were housed in an American Association for Accreditation of Laboratory Animal Care-accredited isolation facility. All procedures involving animals were approved by the University of Wisconsin Animal Care and Use Committee.

Virus Inoculation and BALF Studies

Weanling BN and F344 rats were inoculated with aerosolized Sendai virus strain P3193 as previously described (10). At 3, 5, and 7 d after inoculation, virus-infected rats were anesthetized by intraperitoneal injection with urethane (1.5 g/kg body weight; Sigma, St. Louis, MO) and exsanguinated by severing the abdominal aorta. Uninfected control rats were processed in the same manner at equivalent time points. The lungs were lavaged with a total volume of 10 ml of cold phosphate-buffered saline (PBS) (Sigma), representing five individual exchanges of 2 ml each. Cells were removed from the BALF by centrifugation, and each BALF sample was concentrated to a final volume of 1 ml using a centrifugal filter device with a molecular weight cutoff of 5,000 (Millipore, Bedford, MA). The concentrated BALF samples were stored at 70°C until the concentrations of IFN- and IL-12 present were measured using rat IFN-- and rat IL-12-specific enzyme-linked immunosorbent assay (ELISA) kits (Biosource International, Camarillo, CA), respectively. The rat IFN--specific ELISA had a sensitivity of 13 pg/ml, and the rat IL-12-specific ELISA, which recognizes both the p70 heterodimer and the p40 subunit, had a sensitivity of 5 pg/ml.

Antibodies

The following fluorescein isothiocyanate (FITC)-conjugated monoclonal antibodies (mAbs) were purchased from Pharmingen (San Diego, CA): 10/78 (mouse IgG1; antirat NKR-P1A [an NK cell marker]), G4.18 (mouse IgG3; antirat CD3), OX-35 (mouse IgG2a; antirat CD4), OX-8 (mouse IgG1; antirat CD8a), and G155-178 (mouse IgG2a; unknown specificity). Phycoerythrin (PE)-conjugated G4.18 and J606 (mouse IgG3; antifructosan) were also obtained from Pharmingen. The following affinity-purified, carboxy terminus-specific, polyclonal rabbit antibodies were acquired from Santa Cruz Biotechnology (Santa Cruz, CA): antimouse signal transducer and activator of transcription (STAT) 4, antihuman IL-12 receptor 1 chain (IL-12R1), antihuman Tyk2, and antimouse Janus kinase (Jak) 2. These antibodies are cross-reactive with the corresponding rat proteins. A biotin-conjugated antiphosphotyrosine mAb, PY20, and a carboxy terminus-specific antiactin mAb were obtained from Leinco Technologies (St. Louis, MO) and Santa Cruz Biotechnology, respectively. Horseradish peroxidase (HRP)-conjugated avidin, HRP-conjugated or alkaline phosphatase (AP)-conjugated goat antirabbit IgG, AP-conjugated goat antimouse IgG, and normal rabbit IgG were purchased from Southern Biotechnology Associates (Birmingham, AL).

Cells

Spleens were removed from uninfected BN and F344 weanling rats, which had been anesthetized with urethane and killed by thoracotomy. To prepare cell suspensions, the spleens were dispersed in cold PBS. After lysing red blood cells with an ammonium chloride lysis buffer (Sigma), the splenocytes were thoroughly washed with PBS, resuspended in PBS, and counted using a cell counter (model Z1; Coulter, Hialeah, FL). For experiments requiring purified NK cells, NKR-P1A+ CD3 cells were isolated from spleen-cell suspensions (pooled from multiple weanling rats) using a magnetic cell sorting system (Miltenyi Biotec, Auburn, CA) according to the manufacturer's instructions. Splenocytes were incubated with FITC-conjugated anti-CD3 and then with an anti-FITC mAb coupled to super-paramagnetic microbeads (Miltenyi Biotec). Negative selection columns, mounted in a magnetic stand, were used to deplete the CD3+ cells. The resulting CD3 cell fractions, which were shown by flow cytometry to be essentially free of FITC-labeled cells, were incubated with FITC-conjugated anti-NKR-P1A and then with the anti-FITC microbead reagent. Positive selection columns were then used to isolate the NKR-P1A+ CD3 cells. Only NK cell isolations that yielded a purity  90%, as determined by flow cytometry, were used in experiments. The purities of the BN and F344 NK cell preparations were 93 ± 1% and 95 ± 2% (mean ± standard deviation [SD]; n = 10), respectively. Unfractionated splenocytes (5 × 10⁵ cells/well) or purified NK cells (10⁵ cells/well) were incubated as triplicate cultures for 24 h in 96-well round-bottom plates at 37°C in 5% CO₂. The cells were cultured, at a final volume of 200 µl, in complete medium containing RPMI 1640 (Sigma) with 10% fetal bovine serum (Hyclone, Logan, UT), 4 mM L-glutamine, 10 mM N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid, 50 µM 2-mercaptoethanol, 100 U/ml penicillin, and 100 µg/ml streptomycin (Sigma). Unfractionated splenocytes from individual rats were incubated with either Sendai virus strain P3193 (5 × 10⁴ or 5 × 10⁵ plaque-forming units [pfu]/well, which represents a multiplicity of infection of 0.1 or 1, respectively) or 2 ng/ml of recombinant murine IL-12 (R&D Systems, Minneapolis, MN). Dose-response experiments demonstrated that this dose of IL-12 provides maximal stimulation of splenocytes from both strains. Control wells contained splenocytes incubated with either medium or an appropriate concentration of allantoic fluid, the vehicle for the Sendai virus. Purified NK cells were incubated in the presence or absence of recombinant murine IL-12 (0.002 to 20 ng/ml), recombinant rat IL-18 (10 or 50 ng/ml; R&D Systems), or recombinant rat IFN- (1,000 or 2,500 U/ml; specific activity: 2 × 10⁸ U/mg; Biosource International). To test for synergistic effects, NK cells were also incubated in the presence of IL-18 (10 or 50 ng/ml) and either IL-12 (0.02, 0.2, or 2 ng/ml) or IFN- (1,000 or 2,500 U/ml). After 24 h, the supernatant fluids were harvested and stored at 70°C until the concentration of IFN- present was measured using a rat IFN--specific ELISA kit (Biosource International).

Flow Cytometry

Spleen cells were prepared as described earlier but were resuspended in PBS with 0.1% bovine serum albumin and 0.05% sodium azide (Sigma). To determine the percentages of NK cells and T cells in the spleen, 10⁶ cells from individual rats were stained with PE-conjugated anti-CD3 and either FITC-conjugated anti-NKR-P1A, anti-CD4, or anti-CD8a for 15 min at 4°C in a total volume of 100 µl. Cells were also stained with FITC- and PE-conjugated control antibodies (G155-178 and J606, respectively). All antibodies were used at a final concentration of 5 µg/ml. The cells were thoroughly washed with cold buffer and then resuspended in 500 µl of buffer. Two-color analysis was performed using a flow cytometer (FACScan; Becton Dickinson, San Jose, CA). To determine the percentages of positive cells, gates were set, on the basis of staining with control antibodies, using Cell Quest software version 3.1 (Becton Dickinson).

Immunoprecipitations, Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis, and Immunoblotting

Purified splenic NK cells from BN and F344 weanling rats were incubated in the presence or absence of IL-12 (2 ng/ml) for 15 min at 37°C in complete medium. After washing with cold PBS, the cells were solubilized in an ice-cold buffer containing 1% Triton X-100 (Pierce, Rockford, IL); 50 mM Tris-HCl (pH 7.4); 250 mM NaCl; 50 mM NaF; 0.5 mM sodium pyrophosphate; 1 mM sodium orthovanadate; 1 mM phenylmethylsulfonyl fluoride; and 5 µg/ml each of aprotinin, leupeptin, and pepstatin. After vortexing and a 10-min incubation on ice, the insoluble material was removed from the cell lysates by centrifugation at 14,000 × g for 10 min at 4°C. The protein concentrations of the lysates were determined by a Coomassie protein assay (Pierce). Lysate samples, containing equal amounts of total protein, were incubated with 2 µg of anti-STAT4 antibody for 14 to 16 h at 4°C. Protein G agarose beads (Amersham Pharmacia Biotech, Piscataway, NJ) were then added for 1 h at 4°C to isolate the immune complexes. The beads were extensively washed with cold lysis buffer, resuspended in sodium dodecyl sulfate (SDS) sample buffer (Novex, San Diego, CA) with 5% 2-mercaptoethanol, and boiled for 10 min. Proteins were separated by SDS-polyacrylamide gel electrophoresis (PAGE) on 8% Tris-glycine gels (Novex) and transferred to polyvinylidene difluoride membranes (Immobilon-P; Millipore) using a semidry electrophoretic transfer apparatus (Bio-Rad, Hercules, CA). The membranes were blocked with 1% ovalbumin (Sigma) in Tris-buffered saline (150 mM NaCl and 10 mM Tris-HCl, pH 8.0) with 0.05% Tween 20 (TBS-T) for 1 h. Membranes were then incubated with biotin-conjugated PY20 for at least 2 h. All primary and secondary antibodies and reagents were diluted in TBS-T containing 1% ovalbumin. The membranes were washed with TBS-T and then incubated with HRP-conjugated avidin for 1 h. After washing, the immunoblots were developed by enhanced chemiluminescence (ECL) (Amersham Pharmacia Biotech) and exposed to film. These membranes were then reprobed with the immunoprecipitating anti-STAT4 antibody and developed using AP-conjugated goat antirabbit IgG and a nitroblue tetrazolium (NBT) reaction (Life Technologies, Grand Island, NY). In other experiments, lysates were prepared from purified NK cells from BN and F344 weanlings as described earlier, with the exception that the lysis buffer contained 150 mM NaCl. Lysate samples containing 15 µg of total protein were added to an equal volume of 2× SDS sample buffer with 5% 2-mercaptoethanol and boiled for 10 min. The samples were analyzed by SDS-PAGE followed by immunoblotting with either anti-IL-12R1, anti-STAT4, anti-Tyk2, or anti-Jak2 as described above. After washing, the membranes were probed with HRP-conjugated goat antirabbit IgG, developed by ECL, and exposed to film. To show that equal amounts of protein were loaded onto each lane of the gels, the same membranes were then reprobed with an antiactin mAb and developed using AP-conjugated goat antimouse IgG and an NBT reaction. NIH Image 1.62 software was used for densitometric analysis.

Data Analysis

When data met the assumptions for parametric tests, a paired t test, a two-sample t test, or a randomized block (where each block represented an individual experiment day) analysis of variance (ANOVA) were used for planned comparisons. A residual analysis was used to test the adequacy of the ANOVA models. Variables that did not conform to parametric assumptions were tested in an analogous manner using Kruskal-Wallis and Mann-Whitney tests. To conform to parametric assumptions, IFN- ELISA data were log transformed. SYSTAT version 7.0 software (SPSS, Chicago, IL) was used for analyses.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

IFN- Levels in BALF from Sendai Virus-Infected Rats

The levels of IFN- in BALF from Sendai virus-infected BN and F344 weanling rats were determined (Figure 1A). The kinetics of the responses were similar, with both strains having peak levels of IFN- at 5 d after inoculation. However, the levels of IFN- present 5 d after infection were significantly lower in BN rats than in F344 rats (P < 0.01). There was no significant difference between the strains in IFN- levels at 3 or 7 d after inoculation, and no IFN- was detected in BALF from either strain at 10 d after inoculation. No IFN- was detected in BALF from uninfected rats. Therefore, the Sendai virus-induced host response in BN weanling rats, as compared with F344 weanlings, was characterized by a significant reduction in peak levels of IFN- production, as measured in BALF.

View larger version (18K): [in this window] [in a new window]   Figure 1.   IFN- and IL-12 levels in BALF from Sendai virus- infected BN and F344 weanling rats. BN and F344 rats were inoculated as weanlings with aerosolized Sendai virus. BALF was harvested from the lungs of BN (open circles) and F344 (filled circles) rats at the indicated times after inoculation (or from uninfected control rats) and then concentrated to a final volume of 1 ml. (A) IFN- and (B) IL-12 levels were measured by ELISA. Data represent values from individual rats of each strain. Bars indicate medians. BALF samples were obtained after two independent inoculations that yielded one to three rats per time point. BALF from uninfected weanlings of each strain (n = 5) contained no detectable IFN- (< 13 pg/ml). The dashed line represents the lower limit of detection of the IL-12 ELISA (5 pg/ml). *A significant difference (P < 0.01; Mann-Whitney test) between BN and F344 responses. **A significant difference (P < 0.05; Mann-Whitney test) between IL-12 levels at Day 5 after inoculation and those from uninfected rats of the same strain.

IL-12 Levels in BALF from Sendai Virus-Infected Rats

The levels of IL-12 in BALF from Sendai virus-infected BN and F344 weanling rats and from uninfected weanling rats were determined so we could examine whether the strain differences in IFN- production were associated with the differential production of this IFN--inducing cytokine (Figure 1B). There was a significant increase in the levels of IL-12 in BALF from weanlings of both strains at Day 5 after inoculation compared with the levels from uninfected weanlings (P < 0.05). However, the levels of IL-12 in BALF samples at 3 or 7 d after inoculation were not significantly different from those in uninfected rats. Overall, there were no significant differences between the strains in the levels of IL-12 in BALF at 3, 5, or 7 d after inoculation.

Sendai Virus-Induced IFN- Secretion by Splenocytes

To determine whether a strain difference in Sendai virus- induced IFN- levels could also be observed in vitro, splenocytes from either BN or F344 weanling rats were incubated with Sendai virus for 24 h. Sendai virus induced the dose-dependent secretion of IFN- from splenocytes of both strains (Figure 2). However, BN splenocytes produced significantly less IFN- in response to Sendai virus than did F344 splenocytes (P < 0.005). At Sendai virus doses of 5 × 10⁴ and 5 × 10⁵ pfu, the median levels of IFN- secreted by BN splenocytes were 38- and 24-fold lower, respectively, than those secreted by F344 splenocytes. Incubation of splenocytes with either allantoic fluid (the vehicle for the Sendai virus) or medium alone resulted in no detectable IFN- production.

View larger version (17K): [in this window] [in a new window]   Figure 2.   IFN- secretion by splenocytes from weanling BN and F344 rats in response to stimulation with either Sendai virus or IL-12. Splenocytes from BN (open circles) and F344 (filled circles) weanlings were incubated for 24 h in the presence of either Sendai virus or IL-12, and IFN- levels were measured in the supernatant fluids by ELISA. Data represent mean values for triplicate cultures from individual rats of each strain (n = 6). Bars indicate medians. Supernatant fluids from splenocytes incubated with either allantoic fluid or medium contained no detectable IFN- (< 13 pg/ml). *P < 0.005 by Mann-Whitney test (comparison of BN and F344 responses to Sendai virus) or t test (comparison of BN and F344 responses to IL-12).

IL-12-Induced IFN- Secretion by Splenocytes

Because Sendai virus-induced IFN- secretion was dramatically reduced in spleen-cell cultures from BN weanlings, we decided to investigate the possibility that cells from BN rats have a reduced ability to respond to IFN-- inducing cytokines. Accordingly, we examined the responses of splenocytes from both strains to IL-12, an inducer of IFN- production that plays a critical role in both innate and adaptive host responses to infection. Incubation with IL-12 for 24 h induced IFN- secretion in splenocyte cultures from either BN or F344 weanling rats (Figure 2). However, as observed after incubation with Sendai virus, BN splenocytes produced significantly less IFN- in response to IL-12 than did F344 splenocytes (P < 0.005). The median level of IFN- secreted by BN splenocytes was almost 7-fold less than that secreted by F344 splenocytes.

Frequencies of NK Cells and T Lymphocytes in the Spleen

One factor that could contribute to the differences in IFN- production observed between the BN and F344 splenocyte cultures is a strain difference in the relative frequencies of NK (NKR-P1A+ CD3) cells and/or T (CD8+ CD3+ or CD4+ CD3+) lymphocytes. Therefore, the percentages of NKR-P1A+ CD3, CD4+ CD3+, and CD8+ CD3+ cells in spleens from BN and F344 weanling rats were determined by flow cytometry (Table 1). The percentage of NK cells in the spleen was significantly lower in BN weanlings as compared with F344 weanlings (P < 0.05). However, the percentage of CD4+ T cells was significantly higher in BN spleens as compared with F344 spleens (P < 0.05). There was no significant difference between the strains with respect to the frequency of splenic CD8+ T cells.

IL-12-Induced IFN- Secretion by NK Cells

Because splenocytes from weanling BN rats contained a significantly lower percentage of NK cells than did splenocytes from F344 weanlings, it was possible that the reduced levels of IFN- production observed in BN spleen-cell cultures were due mainly to the presence of reduced numbers of NK cells. To determine whether NK cells from weanling BN and F344 rats also differed with respect to their ability to secrete IFN-, NK (NKR-P1A+ CD3) cells were purified from the spleens of these rats by magnetic cell sorting. Splenocytes were first depleted of CD3+ cells, as confirmed by flow cytometry (data not shown). NKR-P1A+ cells were then positively selected from these CD3 cell fractions. Two subpopulations of NK cells were consistently observed in the purified cell isolates, an NKR-P1A+^bright and an NKR-P1A+^dim population (Figure 3). NK cell isolates from BN weanlings contained a significantly lower percentage of NKR-P1A+^bright cells as compared with those from F344 weanlings (27 ± 6% versus 41 ± 7%; n = 10; P < 0.005, Mann-Whitney test), although there was no significant strain difference in the percentage of NKR-P1A+^dim cells. Incubation of NK cells from either BN or F344 weanling rats with IL-12 for 24 h resulted in the dose-dependent secretion of IFN-, with maximal IFN- production occurring at a dose of 2 ng/ml (Figure 4A). However, BN NK cells secreted significantly less IFN- in response to stimulation with IL-12 (2 ng/ml) than did F344 NK cells (P < 0.001), with BN NK cells producing a median level of IFN- that was 4-fold lower than that produced by F344 NK cells (Figure 4B). Incubation of NK cells in the absence of IL-12 resulted in no detectable IFN- production.

View larger version (19K): [in this window] [in a new window]   Figure 3.   Flow cytometry profiles of NK cells purified from the spleens of weanling BN and F344 rats. NK (NKR-P1A+ CD3) cells were isolated from spleens of BN and F344 weanlings by magnetic cell sorting. The intensity of staining of the BN (upper histogram) and F344 (lower histogram) cells with FITC-conjugated anti-NKR-P1A was determined using a flow cytometer. Splenocytes stained with a control FITC-labeled antibody were used to set the gates, and the percentage of positive cells (96%) is indicated. Two subpopulations of NK cells were consistently isolated, an NKR-P1A+bright and an NKR-P1A+dim population. Data shown are from a representative experiment.

View larger version (16K): [in this window] [in a new window]   Figure 4.   Comparison of the ability of NK cells from BN and F344 weanling rats to secrete IFN- in response to stimulation with IL-12. NK (NKR-P1A+ CD3) cells were isolated from spleens of BN and F344 weanlings by magnetic cell sorting. (A) BN (open circles) and F344 (filled circles) NK cells were incubated for 24 h in the presence of the indicated concentrations of IL-12, and IFN- levels were measured in the supernatant fluids by ELISA. Data represent means ± SD of triplicate cultures. Absence of error bars indicates that the SD is too small to be seen. The dashed line represents the lower limit of detection of the ELISA (13 pg/ml). This experiment is representative of three independent experiments. (B) BN (open circles) and F344 (filled circles) NK cells were incubated for 24 h in the presence of IL-12 (2 ng/ml), and IFN- levels were measured in the supernatant fluids by ELISA. Data represent the means of triplicate cultures from 10 independent experiments comparing the responses of BN and F344 NK cells. Bars indicate medians. A Mann-Whitney test was used to compare the BN and F344 responses. Supernatant fluids from NK cells cultured in the absence of IL-12 contained no detectable IFN- (< 13 pg/ml).

IL-12R1, STAT4, Tyk2, and Jak2 Expression in NK cells

Reduced expression of proteins involved in signaling through the IL-12 receptor could possibly account, at least in part, for the reduced ability of BN NK cells to secrete IFN- in response to IL-12 stimulation. Therefore, immunoblot analysis of extracts derived from purified NK cells of BN and F344 weanlings was used to determine the levels of expression of four proteins involved in this signaling pathway. IL-12R1 was expressed at equivalent levels in BN and F344 NK cells (Figure 5A). Similarly, STAT4, Tyk2, and Jak2, components of the Jak/STAT pathway that are involved in IL-12 signaling (19), were expressed at comparable levels in BN and F344 NK cells (Figures 5B-5D). Reprobing of the same immunoblots with an antiactin mAb showed that the lanes had been loaded with equal amounts of protein (Figures 5A-5D).

View larger version (29K): [in this window] [in a new window]   Figure 5.   IL-12R1, STAT4, Tyk2, and Jak2 expression and IL-12-induced STAT4 phosphorylation in NK cells from weanling BN and F344 rats. Extracts (15 µg of protein per lane) of NK (NKR-P1A+ CD3) cells, which had been isolated from spleens of BN (lanes 1 and 3) and F344 (lanes 2 and 4) weanlings by magnetic cell sorting, were analyzed by SDS-PAGE followed by immunoblotting with (A) anti-IL-12R1, (B) anti-STAT4, (C ) anti-Tyk2, or (D) anti-Jak2. The membranes were developed using HRP-conjugated goat antirabbit IgG and ECL. The same membranes were then reprobed with antiactin and developed using AP-conjugated goat antimouse IgG and an NBT reaction (A-D). (E ) NK cells from BN (lanes 1 and 2) and F344 (lanes 3 and 4) weanlings were incubated with either medium (lanes 1 and 3) or 2 ng/ml of IL-12 (lanes 2 and 4) for 15 min at 37°C and then solubilized in a buffer containing 1% Triton X-100. Anti-STAT4 was used to immunoprecipitate STAT4 from lysate samples containing equal amounts of total protein, and the immunoprecipitates were analyzed by SDS-PAGE followed by immunoblotting with a biotinylated antiphosphotyrosine mAb, PY20. The membrane was developed using HRP-conjugated avidin and ECL. IP and STAT4P denote immunoprecipitation and tyrosine phosphorylated STAT4, respectively. (F ) The same membrane was reprobed with the immunoprecipitating anti-STAT4 antibody and developed using AP-conjugated goat antirabbit IgG and an NBT reaction. (A-D) and (E and F ) are representative of two and three independent experiments, respectively.

IL-12-Induced Tyrosine Phosphorylation of STAT4 in NK Cells

The transcription factor STAT4 is rapidly activated by tyrosine phosphorylation after binding of IL-12 to its receptor, leading to the dimerization and subsequent translocation of STAT4 to the nucleus where it binds to specific enhancer elements and induces gene expression (20, 21). Therefore, to determine whether differences could be observed between BN and F344 NK cells with respect to IL-12 signaling, STAT4 was immunoprecipitated from extracts of NK cells that had been incubated for 15 min in the presence or absence of IL-12. The immunoprecipitates were analyzed by SDS-PAGE followed by immunoblotting with an antiphosphotyrosine mAb (Figure 5E). Treatment with IL-12 resulted in the tyrosine phosphorylation of STAT4 in both BN and F344 NK cells (Figure 5E, lanes 2 and 4, respectively). Densitometric analysis showed that the levels of tyrosine phosphorylated STAT4, which were normalized to the levels of STAT4 in the lanes after reprobing of the immunoblot with anti-STAT4 antibody (Figure 5F), were comparable between BN and F344 NK cells. The bands observed in these immunoblots were not present when control immunoprecipitations were performed with normal rabbit IgG (data not shown).

IL-12- and IL-18-Induced IFN- Secretion by NK Cells

To test whether BN NK cells also have reduced responsiveness to other cytokines that induce IFN- secretion, we investigated the response of NK cells from weanling BN and F344 rats to stimulation with IL-18. In addition, we examined the responses of NK cells to a combination of IL-12 and IL-18 because these cytokines have been shown to have synergistic effects in other systems (22). Incubation of NK cells from BN and F344 weanlings for 24 h with IL-18 resulted in the dose-dependent secretion of modest levels of IFN-, which were not significantly different between the strains (Figure 6). The highest dose of IL-18 tested, 50 ng/ml, induced levels of IFN- secretion that were comparable to those observed with 0.02 ng/ml of IL-12, indicating that IL-12 was a more potent inducer of IFN- production in these cells (Figure 6). However, incubation of NK cells from BN or F344 weanlings with a combination of IL-12 and IL-18 at any of the tested doses resulted in a significant synergistic enhancement of the levels of IFN- secreted as compared with the levels that would result if the effects of these cytokines had been merely additive (P < 0.05, ANOVA; Figure 6). These synergistic effects were dose dependent. BN NK cells secreted significantly less IFN- in response to a combination of IL-12 and IL-18 than did F344 NK cells, regardless of the dose tested for either cytokine (P < 0.05; Figure 6). For the different dose combinations of IL-12 and IL-18, there was a mean 2.4-fold decrease in the levels of IFN- produced by BN NK cells as compared with the levels produced by F344 NK cells. Figure 6 shows that BN NK cells secreted significantly less IFN- when cultured with IL-12 at 0.2 and 2 ng/ml than did F344 NK cells (P < 0.05). However, no strain difference was observed when the cells were incubated with IL-12 at 0.02 ng/ml, which differed from the results shown in Figure 4A. This variability may be related to the relatively weak IFN- response induced by this dose of IL-12.

View larger version (18K): [in this window] [in a new window]   Figure 6.   Independent and combined effects of IL-12 and IL-18 on IFN- secretion by NK cells from weanling BN and F344 rats. NK (NKR-P1A+ CD3) cells were isolated from spleens of BN and F344 weanlings by magnetic cell sorting. BN (open circles) and F344 (filled circles) NK cells were incubated for 24 h in the presence of the indicated concentrations of IL-12, IL-18, or both IL-12 and IL-18. IFN- levels were measured in the supernatant fluids by ELISA. Data represent means ± SD of three independent experiments, which were performed in triplicate. The dashed line represents the lower limit of detection of the ELISA (13 pg/ ml). *A significant difference (P < 0.05; ANOVA) between BN and F344 responses.

IFN-- and IL-18-Induced IFN- Secretion by NK Cells

In addition to its direct antiviral properties, IFN- has also been shown to induce IFN- secretion and to synergize with IL-18 (14, 16). Therefore, the response of NK cells from weanling BN and F344 rats to stimulation with IFN- in the presence or absence of IL-18 was also investigated. Incubation of NK cells from either BN or F344 weanlings for 24 h with IFN- resulted in the secretion of relatively modest levels of IFN- (Figure 7). However, BN NK cells secreted significantly less IFN- in response to either dose of IFN- tested than did F344 NK cells (P < 0.05), although these differences were not as marked as those observed with IL-12. Incubation of NK cells from BN or F344 weanlings with a combination of IFN- and IL-18 at any of the doses tested resulted in a significant synergistic increase in the levels of IFN- secretion (P < 0.05, ANOVA; Figure 7). However, the levels of IFN- produced in response to the synergistic effects of IL-12 and IL-18 were still much higher than those produced in response to IFN- and IL-18 (compare Figures 6 and 7). BN NK cells secreted significantly less IFN- in response to the combination of IFN-, at 2,500 U/ml, with IL-18 than did F344 NK cells (P < 0.05) (Figure 7), although this strain difference was less dramatic than that observed with IL-12 and IL-18. When NK cells were incubated with IL-18 and 1,000 U/ml of IFN-, no significant difference was observed between the BN and F344 responses.

View larger version (15K): [in this window] [in a new window]   Figure 7.   Independent and combined effects of IL-18 and IFN- on IFN- secretion by NK cells from weanling BN and F344 rats. NK (NKR-P1A+ CD3) cells were isolated from spleens of BN and F344 weanlings by magnetic cell sorting. BN (open circles) and F344 (filled circles) NK cells were incubated for 24 h in the presence of the indicated concentrations of IFN-, IL-18, or both IFN- and IL-18. IFN- levels were measured in the supernatant fluids by ELISA. Data represent means ± SD of three independent experiments, which were performed in triplicate. The dashed line represents the lower limit of detection of the ELISA (13 pg/ml). *A significant difference (P < 0.05; ANOVA) between BN and F344 responses.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

BN weanling rats, in comparison to F344 weanlings, exhibited significantly reduced peak levels of IFN- in BALF after Sendai virus infection. The in vitro finding that BN splenocytes produced significantly less IFN- in response to Sendai virus than did F344 splenocytes was consistent with this observation. These strain differences could have been due to a variety of factors, including differences in the ability of potential IFN--secreting cells to respond to IFN--inducing cytokines, such as IL-12, IL-18, and IFN-; the frequencies of potential IFN--secreting cells; and the amounts of IFN--inducing cytokines released in response to the virus. Indeed, all of these factors may be involved. In this report, we focused on NK cells because they are likely to be an important early source of IFN- during the innate antiviral host response (25). Innate responses to Sendai virus infection may be especially relevant to this model because the development of the asthma-like phenotype involves the infection of naive hosts, who will experience a distinct lag period before their adaptive immune response becomes effective. Our data show that purified NK cells from BN weanlings, as compared with those from F344 weanlings, had a reduced response to the IFN--inducing cytokines IL-12 and IFN-.

In response to maximal stimulation with IL-12, BN NK cells secreted significantly less IFN- than did F344 NK cells. The dose-response data show that even the addition of excess IL-12 did not reconstitute the response of the BN NK cells to a level comparable to that of F344 NK cells, demonstrating that BN NK cells had a reduced maximal response rather than a reduced sensitivity to IL-12 stimulation. Incubation of NK cells from weanlings of either strain with both IL-12 and IL-18 resulted in a dramatic synergistic enhancement of IFN- secretion, which was consistent with other reports of synergy between these two cytokines (15). However, BN NK cells still secreted significantly less IFN- in response to a combination of IL-12 and IL-18 than did F344 NK cells. In contrast, BN and F344 NK cells responded similarly to IL-18, producing equivalent, but relatively small, amounts of IFN-. Together, these results suggest that the strain differences observed were mainly due to a variation in response to IL-12 rather than to IL-18. Inasmuch as IL-12 is known to play a critical role in the generation of T helper (Th) 1-type immune responses (18, 26), the decreased ability of NK cells from BN weanlings to secrete IFN- in response to this cytokine could lead to a cytokine imbalance and a shift toward a Th2-type response.

NK cells from BN weanlings, when cultured with IFN-, also secreted significantly less IFN- than did NK cells from F344 weanlings, although IFN- induced lower levels of IFN- than did IL-12. No synergistic effects were observed when NK cells were incubated with a combination of IFN- and IL-12 (data not shown). In contrast, the combination of IFN- and IL-18 induced a significant synergistic increase in IFN- secretion by BN and F344 NK cells, which was consistent with results from a report involving human T cells (14). However, the levels of IFN- produced in response to the synergistic effects of IFN- and IL-18 were much lower than those produced in response to IL-12 and IL-18. BN NK cells secreted significantly less IFN- in response to the combination of IL-18 and the higher concentration of IFN- than did F344 NK cells.

The frequency of NK cells in BN spleens was significantly lower than that in F344 spleens. The lower frequency of NK cells in the BN spleen probably contributed, along with the diminished responsiveness of these cells, to the reduced amounts of IFN- produced in BN unfractionated splenocyte cultures after stimulation with Sendai virus or IL-12. One potential problem with these data is that we were not examining cells from the lung, which is the organ of viral insult. It is possible that the frequencies of NK cells in the lung differ from those observed in the spleen. The spleen was selected for these studies because of its convenience and high yield of cells. However, initial studies have shown that the frequency of NK cells in the lungs is also lower in BN weanlings as compared with F344 weanlings (A. Tuffaha, L. Mikus, L. Rosenthal, et al., unpublished observations).

Rhinovirus and influenza A virus have been shown to induce IFN- production by an indirect innate mechanism that involves eliciting the secretion of IFN--inducing cytokines from mononuclear phagocytes (13, 14). Differences between the BN and F344 strains in the amounts of IFN--inducing cytokines released in response to Sendai virus could contribute to the reduced ability of weanling BN rats to produce IFN-. In fact, a preliminary study suggests that BN rats produce less IL-12 in response to Sendai virus than do F344 rats (27). We did not find any significant differences between the BN and F344 strains with respect to IL-12 levels in the BALF at Days 3, 5, or 7 after inoculation. These results reflect the amount of IL-12 that was accessible by lavage and do not rule out the possibility that strain differences could exist locally within microenvironments in the lung. In addition, the ELISA that was used does not distinguish between the p40 subunit and the functional p70 heterodimer. Therefore, it is possible that background levels of p40 could mask differences in the levels of p70. Regardless of whether there are strain differences in BALF IL-12 levels in response to Sendai virus infection, the presence of IL-12 in BALF from BN weanlings suggests that diminished responsiveness of BN NK cells to IL-12 would be an important factor in reduced IFN- production in vivo.

When considering the mechanisms that could underlie reduced IFN- secretion by BN NK cells in response to IL-12, the presence of quantitative differences in the expression of components of the IL-12 receptor signaling pathway would be one possibility. By immunoblot analysis, BN and F344 NK cells expressed comparable levels of IL-12R1, STAT4, Tyk2, and Jak2. Therefore, differential expression of these proteins did not appear to account for the observed strain difference. The levels of expression of the 2 chain of the IL-12 receptor were not determined due to the lack of an antibody of appropriate specificity. Another possible mechanism would be the presence of functional differences between BN and F344 NK cells with respect to IL-12 signaling. Tyrosine phosphorylation of STAT4 is necessary for the IL-12-mediated induction of IFN- expression (20, 21, 28, 29). Stimulation of BN and F344 NK cells with IL-12 resulted in comparable levels of STAT4 tyrosine phosphorylation, suggesting that the reduced capacity of BN NK cells to secrete IFN- in response to IL-12 may involve events downstream of STAT4 tyrosine phosphorylation. It is also possible that the presence of two subpopulations of NK cells, NKR-P1A+^bright and NKR-P1A+^dim cells, may have relevance to the strain differences in IFN- secretion. We consistently found a higher percentage of NKR-P1A+^bright cells in F344 NK cell isolates as compared with BN isolates. This observation could contribute to strain differences in IFN- production if NKR-P1A+^bright cells are more capable of secreting IFN- than are NKR-P1A+^dim cells. The increased frequency of NKR-P1A+^bright cells in F344 NK cell isolates did not appear to be an artifact of the isolation procedure because a similar difference was observed in flow cytometry experiments with unfractionated splenocytes (data not shown). It has been reported that human NK cells, when cultured in the presence of IL-12 or IL-4, develop into two subsets, NK1 and NK2, with cytokine secretion patterns similar to that of Th1 and Th2 cells, respectively (30). We are currently performing additional experiments to further characterize the subpopulations of NK cells present in the weanling BN and F344 rats and to examine IL-12-mediated signaling events in these subpopulations.

These studies have focused on NK cells, but differences between BN and F344 weanlings with respect to the ability of CD8+ and CD4+ T cells to secrete IFN- may also have an important influence on levels of Sendai virus-induced IFN- production in vivo. Preliminary in vitro studies with CD8+ T cells, which had been purified from the spleens of weanling BN and F344 rats, have shown that BN CD8+ T cells, in comparison with F344 CD8+ T cells, have a markedly reduced capacity to secrete IFN- in response to T-cell receptor activation in either the presence or absence of IL-12 (31). Under these same conditions, interestingly, preliminary experiments with purified splenic CD4+ T cells from weanling rats have indicated that BN CD4+ T cells, as compared with F344 CD4+ T cells, have a somewhat increased ability to secrete IFN- (L. Rosenthal, L. Mikus, R. Sorkness, et al., unpublished observations). Therefore, the reduced ability of weanling BN rats to produce IFN- in response to Sendai virus infection may involve specific deficiencies in NK cell and CD8+ T-cell function.

A further intriguing aspect of this rat model is the differences in response to virus infection that develop depending on the age of the animal at the time of inoculation. BN rats are most susceptible to the development of the asthma-like phenotype if infected in early life. If infected closer to adulthood, rats develop transient (2 to 4 wk) alterations in airway physiology (32), but these changes do not develop into the chronic, episodic, reversible, obstructive changes demonstrable in animals that had been infected as weanlings (8, 32). Interestingly, developmental fluctuations have also been noted in humans in terms of IFN- response profiles that may influence the clinical expression of atopy and/or asthma (33). Preliminary results in our laboratory indicate that IFN- production from NK cells may differ not only by strain, but by the age of the rat as well.

In summary, our results demonstrate that NK cells from BN weanlings have a reduced capacity to secrete IFN-. This deficiency may contribute to the diminished peak levels of IFN- in BALF from Sendai virus-infected BN rats. We have previously demonstrated that treatment of BN rats with aerosolized IFN- during the acute viral infection prevents the development of the postviral asthma-like phenotype (10). Consequently, the susceptibility of BN rats to the development of postviral chronic airway dysfunction may be related to their apparent inability to produce protective levels of IFN- during a relatively narrow window in the course of the acute infection. Finally, parallels between observations made in this rat model and similar observations in humans are noteworthy. First, cytokine imbalance (diminished IFN- production), detectable in early life, has been considered to be an important risk factor associated with the subsequent development of atopic diseases and/or asthma in children (5). Second, paramyxovirus (particularly respiratory syncytial virus and parainfluenza virus) infections during infancy are also potential risk factors for the subsequent development of childhood asthma (1, 3). Thus, the inclusion of a genetic (cytokine dysregulation with aberrant IFN- production), an environmental (paramyxovirus infection induces an asthmatic phenotype), and a developmental (weanling rats are most susceptible) component in the rat model reported herein should facilitate further comprehensive analyses of these important relationships and interactions with regard to the inception of childhood asthma.