Introduction

Top Abstract Introduction Materials and Methods Materials Results Discussion References

p59fyn is a member of the src family of nonreceptor-associated tyrosine-specific protein kinases. The fyn gene encodes the protein products p59fynB and p59fynT, which result from alternative splicing of distinct exons residing in the SH2 and catalytic domains (1). The latter form is restricted in its expression to hematopoietic cells, including peripheral T cells and thymocytes (2). p59fynT has been shown to be associated with the CD3 subunit of the T-cell receptor (TCR) where it participates with lck in signaling (3). In addition, p59fyn has also been associated with the B-cell receptor (BCR) as well as cytokine receptors such as interleukin (IL)-5 and IL-7 (4). The participation or association of fyn described in these processes have been examined primarily in vitro using isolated cell culture systems.

To clarify the role that kinases are believed to play in the signaling of a given receptor, it has been useful to examine the functional consequences of their genetic ablation. Mice homozygous for the p59fynT ablation exhibit no overt phenotype and the development of T lymphocytes proceeds normally (9, 10). These mice do, however, demonstrate a specific defect in thymocyte stimulation through TCR that is not present in peripheral T cells (10). Recently, it has been demonstrated that fyn-deficient mice have a markedly impaired development of natural killer (NK) 1.1⁺ T cells (11, 12). These cells, which were originally defined by their expression of the NK1.1 surface marker, have been shown to produce large amounts of IL-4 upon CD3 crosslinking (13). Furthermore, B-cell development proceeds normally in fynT-deficient mice as does signal transduction through the BCR in mature cells (8).

To better understand the potential importance of p59fynT in response to antigen sensitization and challenge, we compared wild-type mice with those in which the fynT gene had been depleted in the context of a model of allergic pulmonary inflammation. In the absence of p59fynT, we observed an exaggerated response in sensitized mice to aerosolized antigen challenge that was expressed in bronchoalveolar lavage fluid (BALF) by increased numbers of infiltrating eosinophils and T helper (Th) 2 cytokine production. These data suggest the kinase primarily functions as a negative regulator of immune responses that participate in pulmonary allergic disease.

	Materials and Methods

Top Abstract Introduction Materials and Methods Materials Results Discussion References

General

Mice genetically deficient in the p59fynT gene were obtained from Dr. S. Levin (Washington University, St. Louis, MO) as described previously (10). Heterozygote mice were back-crossed to C57Bl/6 mice for eight generations and confirmed 99% for this strain (Charles River, Wilmington, MA). The absence of the kinase in p59fynT-deficient mice (fynKO) was verified by polymerase chain reaction analysis (data not shown). In all experiments, mice homozygous for p59fynT gene ablation (fynKO) were used and compared with age- and sex-matched C57Bl/6 mice. No differences in responses were observed between sexes, hence data from male and female animals were combined for those experiments in which both were used.

Antigen Sensitization and Challenge

Animals were used in accordance with the guidelines set forth by the Pfizer Animal Care and Use Committee. Active sensitization was performed at 6 to 8 wk of age by administration of an intraperitoneal injection of chick egg ovalbumin (100 µg) adsorbed to alum (4.5 mg) (Pierce, Rockford, IL) in a volume of 200 µl on Days 0 and 7. On Days 21 through 24, mice were exposed to aerosols of phosphate-buffered saline (PBS) or ovalbumin (5%) in PBS delivered for 30 min by a DeVilbiss ultrasonic nebulizer (Somerset, PA) into a Plexiglas chamber of dimensions 18 × 43 cm.

Twenty-four or seventy-two hours after the final aerosol challenge, mice were anesthetized intramuscularly with ketamine (450 mg/kg) and xylazine (10 mg/kg). Blood was removed by cardiac puncture for serum immunoglobulin (Ig) analyses and white blood cell (WBC) counts that were analyzed using an automated cell counter (model no. H1, Bayer, Tarrytown, NY). Bronchoalveolar lavage (BAL) was performed by instilling 3 × 1 ml 0.1% bovine serum albumin (BSA) in PBS via tracheal cannula; fluid recovery was  80% of instilled volume. Total BAL cell counts were made using a Cell-Dyn 3500 system (Abbott, Chicago, IL). Differential cell counts were made on 150 µl BALF that had been centrifuged for 2 min at 500 rpm in Cytospin centrifuge (Shandon Instruments, Sewickly, PA), and slides were stained with Diff-Quik (Dade Behring, Newark, DE).

Cytokine Assays

Spleens were isolated from wild-type and fynKO mice that had been sensitized to ovalbumin 3 to 5 wk earlier as described previously. Splenocytes were suspended in RPMI containing 10% fetal calf serum, 2 mM glutamine, 0.55 µM -mercaptoethanol, 0.1 mM nonessential amino acids, and 100 U/ml penicillin/streptomycin (GIBCO-BRL, Gaithersburg, MD) and seeded in 96-well culture plates at a density of 1.5 × 10⁶ cells/ml in duplicate. Cells were incubated for 24 h at 37°C in the presence of buffer, ovalbumin (1 to 1,000 µg/ml), or concanavalin A (ConA) (5 µg/ml).

Murine IL-4 (R&D Systems, Minneapolis, MN), IL-5 (Biosource International, Camarillo, CA), interferon (IFN)-, and IL-2 (R&D Systems) were quantitated by enzyme-linked immunosorbent assay in mouse splenocyte culture supernatants or BALF after 24 h of incubation, or postaerosol exposure, respectively.

Immunoglobulin Assays

Blood was removed by cardiac puncture and total serum IgE (Pharmingen, San Diego, CA) and total IgG (Pierce) were measured in naïve or sensitized mice at both 24 and 72 h after PBS or ovalbumin aerosol challenge.

Assessment of IL-5 Receptor Function In Vivo

Mice were injected intravenously with vehicle (0.1% BSA/PBS) or murine IL-5 (R&D Systems) 1 µg per animal. Two hours after injection, animals were anesthetized with ketamine and xylazine as described previously and bled by cardiac puncture. Total and differential peripheral WBC counts were analyzed as described previously.

Dermal Eosinophil Peroxidase Activity

Mice were briefly anesthetized by inhalation with methoxyflurane. Anesthesia was determined by the failure of the animal to respond to a toe pinch or injection. The dorsal surface was shaved and intradermal injections of vehicle (0.1% BSA/PBS) or murine eotaxin (PeproTech Inc., Rocky Hill, NJ) made in a volume of 20 µl. Animals were killed 4 h after injection and 8-mm biopsies were made at the site of injection. Eosinophil peroxidase (EPO) activity in biopsies was measured by the method of Strath and coworkers (14). Briefly, biopsies were placed in 0.5% hexadecyltrimethylammonium bromide in 50 mM KPO₄ buffer and homogenized. Samples were then frozen twice to lyse cells, centrifuged for 30 min at 2,100 × g, and the supernatant assayed for EPO. Aliquots of 50 µl standard and samples were added to 96-well plates followed by 120 µl of O-phenylenediamine reagent. After 5 min, the reaction was stopped upon addition of 2 M H₂SO₄. Samples were analyzed spectrophotometrically at 490 nm.

	Materials

Top Abstract Introduction Materials and Methods Materials Results Discussion References

All reagents were obtained from Sigma Chemical Company (St. Louis, MO) unless otherwise noted.

Statistics

Unless otherwise specified, results are expressed as mean ± standard error of the mean. Student's t test was used for comparison between two groups, such as the difference between IL-5 levels produced in response to ConA stimulation between wild-type and fynKO splenocytes (SigmaSTAT, Jandel Scientific, San Rafael, CA). One-way analysis of variance (ANOVA) was used to compare the effects of more than two groups, such as peripheral blood eosinophils in PBS-challenged and ovalbumin-challenged wild-type versus fynKO mice. In the absence of equal variances, one-way ANOVA was performed on ranks with Dunn's method used for multiple comparison procedures. Significance was assigned at P < 0.05.

	Results

Top Abstract Introduction Materials and Methods Materials Results Discussion References

Antigen-Induced Inflammatory Cell Influx

Upon gross histologic examination, lungs from sensitized wild-type and fynKO mice exposed to saline aerosol were not different nor was there any evidence in either group of pulmonary inflammation upon microscopic analysis (data not shown). The macrophage was the predominant cell in BALF derived from fynKO (100%) and wild-type (98.7%) mice exposed to PBS aerosol (Figures 1A and 1B). Total WBC numbers in BALF derived from ovalbumin- sensitized wild-type mice exposed to aerosolized antigen for 4 d were approximately 8-fold greater than those receiving PBS aerosol at both 24 and 72 h post-exposure (Figures 1A and 1B). Total WBC numbers after PBS aerosol challenge were not different in BALF from wild-type versus fynKO animals. Antigen challenge increased BALF WBC numbers in fynKO animals 27 times relative to those from wild-type mice under the same conditions at both time points examined. The predominant infiltrating cell after antigen challenge was the eosinophil, which rose from 0 to 55% and 0 to 70% of total BAL WBC in wild-type and fynKO mice, respectively. Total numbers of all BALF WBC, including eosinophils, derived from ovalbumin-sensitized and challenged fynKO mice were significantly greater than those from wild-type mice (Figures 1A and 1B).

View larger version (20K): [in this window] [in a new window]   View larger version (21K): [in this window] [in a new window]   Figure 1.   Inflammatory cell influx into allergic mouse BALF at 24 (A) or 72 h (B) postaerosol challenge. Data are mean ± SEM from two separate experiments with a total of 21 to 26 animals per group per time point. WT-PBS = sensitized wild-type mice exposed to PBS aerosol for 30 min for 4 d; WT-OVA = sensitized wild-type mice exposed to ovalbumin aerosol; KO-PBS = sensitized fynKO mice exposed to PBS aerosol; KO-OVA = sensitized fynKO mice exposed to ovalbumin aerosol. a = P < 0.05 versus wild-type PBS aerosol exposed; b = P < 0.05 versus all groups.

Sensitized wild-type mice demonstrated a moderate (average, 2.18-fold) increase in peripheral blood eosinophils 24 and 72 h after antigen challenge relative to sensitized mice exposed to saline aerosol (Table 1). Eosinophil numbers in the peripheral blood of fynKO mice were slightly higher (average, 2.70-fold) upon antigen challenge at both time points relative to their PBS aerosol controls.

Eosinophil Responsiveness to Nonallergic Stimuli

To determine whether the exaggerated pulmonary eosinophilia observed in sensitized fynKO mice in response to antigen challenge was a consequence of their increased sensitivity to chemotactic agents, mice were injected intradermally with murine eotaxin and EPO activity was measured as an indication of infiltrating eosinophils (13). There was no significant difference between wild-type and fynKO mice in eotaxin-induced eosinophil recruitment at either low (0.1 µg) or high (1.0 µg) doses of the chemokine 4 h after injection (Figure 2).

View larger version (22K): [in this window] [in a new window]   Figure 2.   Eotaxin-induced dermal EPO activity. Animals received intradermal injections of murine eotaxin (0.1 and 1 µg) and biopsies were made 4 h later. Data are mean ± SEM above background EPO (WT = 1.14 ± 0.22 and fynKO = 0.77 ± 0.19 nmol/site) from 10 animals per treatment group.

Because response to IL-5 in B lymphocytes has been reportedly attenuated in fynKO animals (8), the ability of the cytokine to affect eosinophil recruitment from bone marrow into peripheral blood was compared between these and wild-type mice. IL-5 (1 µg, intravenous) increased the percent of peripheral blood eosinophils in naïve wild-type mice from 0.67 ± 0.04 to 0.99 ± 0.12%. Baseline levels of circulating eosinophils were similar in naïve fynKO mouse blood (0.51 ± 0.11%) and increased to a similar degree compared with wild-type mice after IL-5 treatment (0.96 ± 0.11%).

Antigen-Induced BALF Cytokine Production

Ovalbumin aerosol exposure significantly increased BALF IL-4 and IL-5 levels in sensitized wild-type mice relative to saline aerosol exposure (Figure 3). Whereas IL-4 and IL-5 levels in BALF derived from fynKO mice exposed to PBS were not significantly different from the corresponding wild-type mice, levels of these cytokines were significantly increased in fynKO BALF after antigen challenge (Figure 3). IFN- was undetectable in all BAL samples (data not shown).

View larger version (22K): [in this window] [in a new window]   Figure 3.   IL-4 and IL-5 levels in BALF 24 h postantigen challenge. Mice sensitized to ovalbumin were exposed to either PBS aerosol (PBS) or ovalbumin aerosol (OVA) for 4 d. Data are mean ± SEM from n = 10 animals per treatment group. a = significant difference from PBS control at P < 0.05. b = significant difference from all groups at P < 0.05.

Ovalbumin versus Mitogen-Induced Cytokine Production in Splenocytes

Cytokine production during 24 h of incubation with ovalbumin (0.1 to 1 mg/ml) or ConA (5 µg/ml) was compared in splenocytes derived from sensitized wild-type versus fynKO mice. Ovalbumin produced a dose-dependent increase in IL-5 levels that were significantly higher in fynKO splenocytes (Figure 4). IFN- levels were also increased by antigen in a dose-dependent manner but were not significantly different between the groups (Figure 4). IL-4 and IL-2 production in response to ovalbumin was low and variable in both groups although there was a trend toward a reduction in fynKO splenocytes (data not shown). ConA (5 µg/ml) stimulated the production of IL-5, IL-4, IL-2, and IFN- from cultured splenocytes (Figure 5). IL-4 and IL-2 levels after mitogen stimulation were reduced by 58% and 15% in fynKO splenocytes relative to wild-type, respectively. ConA-induced increases in splenocyte IL-5 and IFN- levels were not significantly different between the groups.

View larger version (18K): [in this window] [in a new window]   Figure 4.   IL-5 and IFN- levels in supernatants derived from sensitized mouse splenocytes cultured in the presence of various concentrations of ovalbumin (0.1 to 1 mg/ml) for 24 h in vitro. Data are mean ± SEM from two separate experiments combined for n = 10 spleens per treatment group. *Significant difference from wild-type levels at P < 0.05.

View larger version (48K): [in this window] [in a new window]   Figure 5.   Cytokine levels in supernatants derived from sensitized mouse splenocytes cultured in the presence of ConA 5 µg/ml for 24 h in vitro. Data are mean ± SEM of data from fynKO mice expressed relative to wild-type control in two separate experiments combined for n = 10 spleens per treatment group. *Significant difference from wild-type levels at P < 0.05.

Serum Immunoglobulin Levels

Total serum IgE levels in wild-type mice increased approximately 4-fold from naive (nonsensitized) mice 3 wk after sensitization and did not increase further after repeat antigen challenge (Figure 6). Total IgE levels in naive fynKO mouse serum tended to be lower than those in corresponding wild-type mice and were < 50% of those in corresponding wild-type mice after sensitization and PBS aerosol challenge. However, IgE levels in fynKO mice were no different from those in wild-type mice after chronic antigen challenge. Total serum IgG levels were not significantly affected by antigen challenge or p59fyn gene ablation (data not shown).

View larger version (25K): [in this window] [in a new window]   Figure 6.   Serum IgE levels. Serum was analyzed from nonsensitized (naïve) mice or sensitized mice exposed to PBS aerosol or ovalbumin aerosol for 4 d. Data are mean ± SEM from two separate experiments combined for n = 12 to 20 animals per treatment group. a = significant difference from naïve; b = significant difference from PBS aerosol treatment (P < 0.05).

	Discussion

Top Abstract Introduction Materials and Methods Materials Results Discussion References

Although p59fyn is a nonreceptor tyrosine kinase that has been reported to participate in the signaling of a number of receptors associated with immune cell function, its importance during the development of allergic responses has not been previously examined. Our data demonstrate p59fynT gene ablation markedly increased the airway inflammatory response to aerosol antigen challenge in ovalbumin-sensitized mice. The most striking evidence of this was an exaggerated eosinophilia that occurred in BALF both 24 and 72 h after antigen challenge. Because the recruitment of fynKO eosinophils to a nonallergic stimulus, such as eotaxin, was no different from wild-type, it is unlikely that the difference lies specifically in these cells. The marked increase in pulmonary eosinophilia observed in fynKO mice likely reflects elevated BAL levels of IL-4 and IL-5. IL-4 can contribute to selective eosinophil recruitment by increasing endothelial vascular cell adhesion molecule-1 expression and stimulating cell-specific chemotactic factors such as eotaxin (15). IL-5 is known to be critical to the production of pulmonary eosinophilia because it is important in the differentiation, survival, and activation of these cells (15). The increased pulmonary eosinophilia, coupled with the elevations in BAL cytokines, i.e., IL-4 and IL-5, is consistent with a greater shift to Th2 responses in the absence of p59fynT.

CD4⁺ T lymphocytes, particularly of the Th2 type, contribute to pulmonary inflammation in murine models of allergic disease (16, 17). In our studies, the production of the Th2 cytokine IL-5 was increased in response to ovalbumin not only in BALF but also in splenocytes derived from fynKO mice relative to those from wild-type mice, whereas levels of the Th1 cytokine IFN- were not different between the groups. These data suggest a shift in favor of the Th2 cytokine that is consistent with other studies in which IL-5 levels have been shown to increase in the absence of reduction of IFN- (18, 19). In addition, splenocytes from fynKO mice exhibited a small, but significant, attenuation of mitogen-stimulated IL-2 production relative to wild-type. Because mitogenic lectins, such as ConA can interact with the TCR (20), these data are consistent with reports by Stein and colleagues (9) that describe attenuated IL-2 production in response to anti-CD3 stimulation in splenic T cells derived from fynKO mice. Upregulation of fyn has been correlated with IL-2 production in Th1 clones, whereas production of IL-4 in Th2 cells was associated with reduced fyn activity (21). Consistent with these data are the findings of Gimsa and associates (22) demonstrating a shift in expression of Th2 relative to Th1 cytokines in response to CD4 stimulation in the presence of a pyrazolopyrimidine inhibitor of src kinases, including fyn. Finally, recent studies suggest activation of lck, fyn, and ZAP-70 is required for Th0 responses but that the nature of the ensuing response is dependent on a balance between the three (23). The differentiation of Th0 cells into Th2 is associated with an increase in lck and an absence of detectable fyn and ZAP-70 activities. Hence, the absence of fyn in vitro and in vivo appears to predispose the T lymphocyte to the Th2 phenotype, which leads to increased allergic airway inflammation in the latter.

Not all effects of antigen sensitization and challenge observed in fynKO mice were entirely consistent with a shift in T lymphocytes to Th2 phenotype. In particular, prechallenge serum IgE and splenocyte IL-4 levels in fynKO mice were reduced relative to wild-type animals. It is unlikely that these effects result from perturbations in B-cell development or signaling because p59fyn-deficient mice have not demonstrated abnormalities in B-cell lymphopoiesis or B-cell receptor signaling (8). Rather, it has recently been demonstrated that fyn is required for the burst of IL-4 and IL-13 synthesis that occurs rapidly as a result of NK T-cell activation (12). Mice deficient in p59fyn exhibit a tenfold reduction of thymic NK T cells and a greater than fivefold decrease in the spleen and liver. When NK1.1⁺ cells were depleted before antigen priming in an allergic mouse model, systemic IgE and IgG1a levels were reduced, which is consistent with our own results in the fynKO mouse before antigen challenge (24). In fact, the absence of CD4+ NK1.1⁺ T cells has been proposed to be responsible for the defects in IL-4 and IgE production, which is characteristic of SJL mice (25). The increase in BALF IL-4 levels in fynKO mice and the normalization of serum IgE levels relative to wild-type control after chronic antigen challenge suggest compensatory cell sources for the cytokine and immunoglobulin production (26).

A critical role for the kinase in B cell IL-5 receptor signaling was proposed by Appleby and coworkers (8) who demonstrated a complete blockade of the effects of the cytokine in p59fynT-depleted mice when administered as a comitogenic stimulus with dextran sulfate and CD40 ligand. We used the ability of IL-5 to recruit eosinophils from the bone marrow into the peripheral blood as a method to evaluate IL-5 receptor function in vivo (27). The magnitude of this response was similar in wild-type and fynKO mice and is inconsistent with a major role for p59fynT in the signaling of this receptor in all immune cells. Furthermore, the fact that fynKO mice exhibit exaggerated eosinophilia in response to antigen challenge is inconsistent with defective IL-5 signaling in eosinophils because genetic elimination of the cytokine significantly attenuates antigen-induced influx of these cells in allergic mouse models (28).

p59fyn has also been associated with CD23, which is the low affinity Fc receptor for IgE (FcRII) expressed on a number of immune cell types, including mature B cells, follicular dendritic cells, T cells, and eosinophils (29). The primary function of CD23, whose expression is markedly upregulated in allergic disorders, includes IgE-mediated priming and activation of cells expressing the receptor (30). Interestingly, as observed with fynKO mice, sensitized mice genetically deficient in CD23 exhibited increased airway eosinophilia in response to aerosol antigen challenge relative to their wild-type control (31).

In conclusion, these experiments describe a previously unknown function of p59fyn as a negative regulator of pulmonary inflammation in an allergic mouse model. The myriad of receptors and cell types with which this src family kinase has been associated makes definitive identification of the specific cause difficult. However, the exacerbation of pulmonary eosinophilia and BALF IL-4 and IL-5 suggests a predominance of Th2 cells that can be explained by altered TCR signaling, which occurs in the absence of p59fyn. These results are potentially complicated by the effects of NK1.1⁺ T cell depletion in these animals that may have contributed to the reductions in IL-4 and, hence, IgE levels in fynKO animals after antigen sensitization. Certainly effects on other receptors relevant to allergic disease, such as CD23, cannot be discounted.