Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Atopic asthma is a chronic inflammatory disease characterized by increased serum immunoglobulin (Ig)E levels and airway eosinophilia, accompanied by increased airway responsiveness to methacholine (MCh) (1). These pathophysiologic changes, mimicked in murine models of allergic airway inflammation (2, 3) are mediated in part by activated CD4+ T cells. Optimal T-cell activation requires antigen-specific engagement of the T-cell receptor and additional antigen-independent costimulatory interactions, the best-characterized of which is the CD28 pathway (4, 5). CD28 and the homologous molecule cytotoxic T-lymphocyte antigen (CTLA)-4, primarily expressed on T cells, bind to either B7-1 (CD80) or B7-2 (CD86) counterreceptors on antigen-presenting cells (APC). Interruption of this pathway has been shown to prevent the activation of and induce anergy in T cells in vitro (6, 7).

One mechanism by which activated T cells regulate airway hyperresponsiveness (AHR) is by the production of soluble mediators including cytokines. Interleukin (IL)-4 and IL-5, shown to be elevated in bronchial biopsies, bronchoalveolar lavage (BAL) cells, and the blood of allergic asthmatic patients (8), are thought to play a prominent role in generating and perpetuating the asthmatic response. The costimulatory signal delivered to T cells via CD28/ CTLA-4 interactions has a critical role in T-helper (Th) cell differentiation associated with cytokine production, but it is uncertain whether B7-1 and B7-2 differentially influence the subsequent development of a type 1 (Th1) or type 2 (Th2) primary response. In a transfection study of in vitro-stimulated CD4+CD45RA+ T cells, B7-2 induced significantly more IL-4 production than did B7-1, thus potentially directing an immune response toward the Th2 phenotype (11). Additional work in a model of experimental allergic encephalomyelitis (EAE), a model exacerbated by a Th1 immune response, found that anti-B7-1 treatment reduced disease incidence whereas anti-B7-2 increased disease severity (12, 13). In contrast, in a model of autoimmune diabetes (14), also mediated by Th1 cells, anti-B7-1 treatment exacerbated disease while anti-B7-2 treatment abrogated disease development.

We have previously demonstrated that CD28 costimulatory signals play a central role in the pathogenesis of pulmonary inflammation and AHR in a murine model of allergic asthma (4). In this study we found that CTLA4-Ig, a soluble fusion protein that binds both B7-1 and B7-2, inhibits the development of pulmonary disease in mice sensitized and challenged with the allergen ovalbumin (OVA). Further investigation in the same model (15) using antibodies to B7-1 and B7-2, as well as CTLA4-Ig and Y100F, a mutated form of CTLA4-Ig binding only to B7-1, demonstrated that both B7-1 and B7-2 contribute to the development of disease in this model. In vivo blockade of either B7-1 or B7-2 resulted in diminished activation of CD4+ T cells, whereas blockade of both molecules resulted in decreased Th2 cytokine production with a concomitant increase in Th1 cytokine production. In contrast, additional studies in murine models of pulmonary inflammation and AHR have examined the effects of blocking individual B7 molecules and observed that B7-2 costimulation alone appears necessary for development of pulmonary disease (16, 17). Use of soluble mediators to investigate the roles of specific receptor/ligand interactions can be confounded by the complexities of dosing effects, kinetics, and receptor blockade versus active signaling (18). To circumvent these issues, we undertook the present study using mice with germline deletions of the B7-1 and/or B7-2 molecules. Using these mice in a murine model of allergic asthma, we demonstrate that B7-1 and B7-2 have complementary roles in modulating the development of pulmonary allergic inflammation.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Male mice with targeted disruption of the gene encoding B7-1 (B7-1/ mice), B7-2 (B7-2/), or both B7 molecules (B7-1/2/) were generated as previously described

(19, 20). Age-matched 129Sv/Jae male mice were purchased from Jackson Laboratories (Bar Harbor, ME). All animals were maintained according to the guidelines of the Committee on Animals at Harvard Medical School and the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council.

Antigen Sensitization and Challenge

The protocol for induction of pulmonary inflammation via antigen sensitization and aerosol challenge has been described previously (2). Briefly, on Days 0 and 7, mice were immunized via intraperitoneal injection with either chicken OVA (Grade III; Sigma Chemical Co., St. Louis, MO) adsorbed to Al(OH)₃ (Alum) (J.T. Baker Chemical, Phillipsburg, NJ) in phosphate-buffered saline (PBS) (Sigma), or PBS/Alum. On Days 14 to 20, mice underwent aerosolized antigen challenge with either 6% OVA or PBS: mice were placed in a plastic chamber and received OVA/PBS solution via an ultrasonic nebulizer (Model 5000; DeVilbiss, Somerset, PA) for 20 min/d.

Measurement of AHR

As previously described, on Day 21 mice were anesthetized with intraperitoneal injections of pentobarbital sodium (Anthony Products, Arcadia, CA). The trachea was cannulated and connected to a rodent ventilator. An internal jugular vein was cannulated with a Silastic catheter attached to a 0.1-ml microsyringe (Hamilton, Reno, NV) and used to administer MCh (acetyl--methylcholine chloride; Sigma). Pulmonary resistance (RL) and dynamic compliance were determined as previously described. Dose-response curves to MCh were obtained by administering increasing doses of MCh (33 to 1,000 µg/kg).

BAL

The trachea was cannulated and the airways were washed three times with 1.0 ml of PBS supplemented with 0.06 mM sodium ethylenediaminetetraacetic acid (Sigma). The BAL fluid (BALF) was centrifuged (10 min, 4°C, 700 × g) and the supernatant harvested for cytokine analysis. The BAL cells were resuspended in 0.2 ml of cell culture media (RPMI 1640, 8% fetal calf serum, 1% penicillin/streptomycin, and 1% glutamine) and total cell count was determined by trypan blue exclusion using a hemocytometer. Cytospins were prepared (Shandon Scientific, Cheshire, UK), fixed (leukostat fixative solution; Fisher Diagnostics, Pittsburgh, PA), and stained with methylene blue and eosin Y (Leukostat solutions I and II; Fisher Diagnostics). The number of eosinophils in 200 cells was counted based on morphology and staining characteristics. The investigator counting the cells was blinded to the treatment groups.

Lung Histology

The lungs were resected, inflated with ornithine carbamyl transferase (OCT) compound (Miles, Elkhart, IN), immersed in buffered formalin fixative, stained with hematoxylin and eosin (H&E) solution, and examined by light microscopy for histologic changes.

Measurement of Serum Concentrations of IgE

Blood was obtained by cardiac puncture, and serum was obtained by centrifugation at 1,500 × g for 10 min. Serum IgE level was determined with a commercial enzyme-linked immunosorbent assay (ELISA) kit (PharMingen, San Diego, CA).

Cytokine Analysis

BAL supernatant was harvested and analyzed via ELISA (R&D Systems, Minneapolis, MN) for interferon (IFN)-, tumor necrosis factor (TNF)-, IL-4, and IL-5 production as described by the manufacturer.

Statistical Analysis

All results are reported as means ± standard error of the mean (SEM). Data were analyzed using the PRISM (Version 2.01) statistical package. Nonparametric data were analyzed using a Mann-Whitney test, as previously described (4).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Absent Elevation of Serum IgE in B7-1/2/ Mice

129Sv/Jae wild-type and B7-1/2/ mice immunized with OVA were treated with aerosolized OVA. At 24 h after final OVA challenge, the serum concentration of Ig was measured by ELISA. 129Sv/Jae wild-type mice displayed an increase in IgE concentration after allergen treatment (Figure 1A) as we have shown in other strains (BALB/c [4]; and C57Bl/6 [18]). In contrast, the IgE level was below 10 ng/ml in OVA-challenged B7-1/2/ mice, indicating that B7-1/2/ mice displayed no IgE response to OVA-induced sensitization and challenge.

View larger version (13K): [in this window] [in a new window]   Figure 1.   Effect of B7-1/2 deficiency on serum concentration of IgE and airway eosinophilia. (A) The columns represent the concentration of serum IgE in wild-type and B7-1/2/ mice sensitized and challenged with OVA (OVA) or PBS (Control). Results are shown from a single experiment representative of three separate experiments. U.D. = undetectable (< 10 ng/ ml). (B) BALF was obtained from wild-type and B7-1/2/ mice sensitized and challenged with OVA (OVA) or PBS (Control). Each square or triangle represents the absolute number of eosinophils in BALF from a single mouse analyzed. *Denotes values for B7-1/2/ OVA group that differ significantly from wild-type OVA control group (P < 0.05).

Absence of Eosinophilia in BALF of B7-1/2/ Mice

Allergen sensitization and challenge promote a lung inflammatory response characterized by the presence of large numbers of eosinophils in the BALF, lung, and airway tissue. We examined the effect of B7-1/2 deficiency on allergen-induced pulmonary inflammation by quantifying the cellular component of BALF (Figure 1B). In immunized wild-type mice, the number of airway eosinophils was significantly increased after repeated exposure to OVA. In contrast, eosinophils were scarcely detected in BALF of B7-1/2/ mice, demonstrating that B7-1/2/ mice exhibited no eosinophilia in the airway, even after aerosolization with OVA.

Absent Inflammatory Response in Lungs of B7-1/2/ Mice

We also assessed pulmonary inflammation by histologic analysis of lung sections.The usual changes induced by allergen sensitization and challenge in wild-type mice are consistent with pathologic features of human asthma, including cellular peribronchial and perivascular infiltrates (consisting mostly of lymphocytes and eosinophils) and hyperplasia of the mucus-secreting goblet cells lining the bronchial epithelium (Figure 2A), as also found in other strains (BALB/c [4]; and C57Bl/6 [18]). In contrast, congenital absence of either or both B7 molecules resulted in a marked reduction in both cellular infiltrates and inflammatory changes within the bronchial mucosa relative to wild-type controls (Figures 2B-2D). For wild-type control, B7-1/, B7-2/, and B7-1/2/ mice, PBS treatment effected no pathologic change (not shown).

View larger version (123K): [in this window] [in a new window]   Figure 2.   Pulmonary inflammation after antigen challenge in wild-type, B7-1/, B7-2/, and B7-1/2/ mice. Immediately after measurement of AHR, lungs from wild-type, B7-1/, B7-2/, and B7-1/2/ mice sensitized and challenged with OVA were removed from the thoracic cavity, fixed and stained with H&E, and examined by light microscopy. Characteristic inflammatory changes representative of allergen-sensitized mice, including peribronchial and perivascular inflammation consisting of eosinophils, neutrophils, and lymphocytes, are observed in lung sections from wild-type control mice (A). Lungs from B7-1/ (B), B7-2/ (C), and B7-1/2/ mice (D), demonstrating normal bronchial epithelium and a paucity of inflammatory cells. Original magnification: ×100.

Diminished Serum IgE and Eosinophilia in B7-1/ and B7-2/ Mice

To investigate the relative contributions of B7-1 and B7-2 to allergic pulmonary inflammation, we focused on mice lacking either B7-1 or B7-2. Wild-type, B7-1/, and B7-2/ mice, sensitized and challenged with OVA, were examined for serum concentrations of IgE and BAL eosinophilia. Both B7-1/ and B7-2/ mice had reduced levels of serum IgE compared with wild-type controls (Figure 3A), indicating that both phenotypes, each with the loss of a single B7 molecule, had blunted IgE responses to OVA challenge. Quantification of BALF cellularity (Figure 3B) demonstrated significant reduction of the number of BAL eosinophils after OVA challenge in both B7-1/ and B7-2/ mice, indicating that both molecules contribute to the development of airway eosinophilia. Histologic analysis of lung sections revealed that congenital absence of either B7 molecule resulted in a minimal reduction in both cellular infiltrates and inflammatory changes within the bronchial mucosa relative to wild-type controls (Figures 2C and 2D). Our results with B7-deficient mice are consistent with earlier work demonstrating a prominent role for B7-1 in mediating the recruitment of eosinophils to the airway (21) and with work noting that blockade of either B7-1 or B7-2 resulted in significant suppression of BAL eosinophilia (15, 16).

View larger version (17K): [in this window] [in a new window]   Figure 3.   Effect of B7-1 or B7-2 deficiency on serum concentration of IgE and airway eosinophilia. (A) As in Figure 1, the columns represent the concentration of serum IgE in wild-type (WT), B7-1/, and B7-2/ mice sensitized and challenged with OVA. Results are shown from a single experiment representative of three separate experiments. (B) BALF was obtained from wild-type, B7-1/, and B7-2/ mice sensitized and challenged with OVA (OVA) or PBS (Control). Each square or triangle represents the absolute number of eosinophils in BALF from a single mouse analyzed. *Denotes values for treatment groups that differ significantly from wild-type OVA control group (P < 0.05). **Denotes values for treatment groups that differ significantly from wild-type OVA control group (P < 0.01).

Lack of AHR in B7-1/2/ Mice

Previous studies suggested divergence between pathways mediating eosinophilia and a physiologic outcome of inflammation, AHR (22). Thus, the B7-deficient and wild-type mice sensitized and subsequently challenged with aerosolized OVA were analyzed for airway reactivity after intravenous administration of MCh (Figure 4). In wild-type mice, OVA treatment significantly increased the airway response to MCh, as compared with untreated mice. However, the airway response to MCh in OVA-treated B7-1/2/ mice (Figure 4A) was the same as that in nontreated B7-1/2/ and wild-type mice, demonstrating that B7-1/2/ mice did not exhibit AHR after OVA sensitization. The airway response to MCh in OVA-treated B7-1/ (Figure 4B) and B7-2/ (Figure 4C) mice was unchanged from nontreated B7-1/ and B7-2/ mice, respectively.

View larger version (19K): [in this window] [in a new window]   Figure 4.   Effect of B7 deficiency on AHR in OVA-treated mice. RL, expressed as a percent of baseline resistance ± SEM, was measured in living, mechanically ventilated B7-1/2/ (A; OVA, n = 6; control, n = 6), B7-1/ (B; OVA, n = 9; control, n = 6), B7-2/ (C; OVA, n = 16; control, n = 11), and wild-type control mice (OVA, n = 25; control, n = 21), sensitized and challenged with OVA (OVA) or PBS (Control). *Only in wild-type mice, OVA treatment significantly increased AHR compared with that in the untreated group (P < 0.05).

Increase in Th1 Cytokines in B7-1/ and B7-2/ Mice

To investigate the contribution of B7-mediated costimulation on the production of effector cytokines in our model, we measured the levels of two Th1 cytokines (IFN- and TNF-) and two Th2 cytokines (IL-4 and IL-5) in the BALF of OVA-treated B7-1/, B7-2/, and B7-1/2/ mice (Table 1). OVA sensitization and challenge in wild-type mice resulted in decreased production of IFN- and no change in the levels of TNF-, whereas B7-1/ and B7-2/ mice demonstrated increased production of IFN- and TNF- in response to OVA sensitization and challenge (Table 1). The levels of these cytokines were undetectable in B7-1/2/ mice (Table 1). Levels of IL-4 and IL-5 were undetectable in OVA-treated B7-1/, B7-2/, and B7-1/2/ mice (data not shown). These results are consistent with data generated in an EAE model and in a Schistomiasis model, where elevated levels of IFN- and TNF- were observed after antigenic stimulation. Also consistent with our findings, the authors of the EAE study have not been able to detect IL-4 or IL-5 in mice lacking both B7-1 and B7-2.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In this study we demonstrated that pulmonary eosinophilia, AHR, and lung inflammation usually observed in mice sensitized and challenged with allergen are not present in B7-1/2/ mice, suggesting that both B7-1 and B7-2 are required for a normal asthma-like response to OVA in mice. We demonstrated that these same disease parameters are also reduced in B7-2/ mice and reduced to a lesser extent in B7-1/ mice, suggesting that, although B7-2 is quantitatively more important than B7-1 in the induction of this response, B7-1 and B7-2 may have complementary functions in mediating the development of allergic pulmonary inflammation. Further, we observed an increase in Th1 cytokine production concomitant with abrogation of allergen-induced pulmonary responses in mice deficient in either B7-1 or B7-2, suggesting that lack of B7-1 and/or B7-2 costimulation interferes with the generation of a Th2 response.

We have previously demonstrated a role for B7-mediated costimulation in allergic pulmonary inflammation by using a combination of monoclonal antibodies (mAbs) and fusion proteins to block signaling through the individual B7 molecules. The present study represents the first use of mice with germline deletions of the B7 molecules in a model of allergic pulmonary inflammation and AHR. Previous analyses of knockout animals suggested that lifelong deficiency of a cell membrane or secreted molecule can have effects that differ from those caused by acute inhibition of that molecule (26, 27), either because lifelong deficiency causes the development of compensatory mechanisms or because it results in a developmental pattern that exaggerates the effect of the deficiency. Therefore, a dual approach, using both soluble inhibitors and knockout mice, was indicated to adequately assess B7-mediated costimulation in our model. Importantly, the current study corroborates our earlier findings while avoiding the potentially confounding issues of using in vivo antibody administration to study specific receptor/ligand interactions.

Asthma is a chronic disease of the airway with episodic reversible airway obstruction and airway inflammation. The pathophysiologic changes associated with asthma include increased serum IgE level, eosinophilic infiltration of the airway, bronchial mucosal injury, and AHR (28). These changes are thought to be mediated by Th cells with a Th2 cytokine phenotype. In addition to an antigen-specific signal, activation of T lymphocytes requires an antigen-independent costimulatory signal provided by interaction of CD28 with B7 on the APC. Initial characterization of the B7-deficient mice demonstrated that T-cell help in antibody isotype class-switching is critically diminished by the absence of B7-1 and B7-2 (19). Our data are consistent with these observations, as B7-1/2/ mice had undetectable levels of serum IgE after antigen sensitization and challenge. Borriello and associates (19) noted that B7-1 and B7-2 had overlapping roles in Ig class-switching, with B7-2 having a greater role than B7-1. Our data corroborate their findings, as the absence of either B7 molecule blunted serum IgE levels in OVA-sensitized and challenged mice, with the lack of B7-2 more profoundly diminishing serum IgE levels.

Numerous studies have investigated the pathways promoting AHR in asthma (4, 15, 22, 25, 31). Our results demonstrate that both B7-1 and B7-2 signaling participate in the pathogenesis of AHR in this model, as shown both by in vivo blockade (15) and now with germline deletion of the B7 costimulatory molecules. Absence of either or both molecules completely abrogates AHR. Again, although B7-2 appears to play a more prominent role in modulating AHR, the contribution of B7-1 costimulation cannot be ignored. Given that single B7-1 or B7-2 deficiency leads to loss of antigen-induced AHR without similar complete loss of IgE responses, our results support the notion that AHR and pulmonary inflammation can develop by at least two distinct pathways.

Our model of allergic pulmonary inflammation involves intraperitoneal immunization with antigen in the presence of adjuvant, followed by localized pulmonary challenge with aerosolized antigen. One explanation for the phenotypic consequences of absent B7-1 and/or B7-2 derives from the distinct temporal expression patterns of B7-1 and B7-2, with the early expression of B7-2 leading to an important role for B7-2 at the initiation of an immune response. In our model, this response follows intraperitoneal immunization of the mice with OVA in Alum adjuvant. Earlier immunization studies in B7-1/2/ mice (19) suggested that adjuvant-elicited local inflammatory responses can induce B7-1 expression in the B7-2/ mice to provide the necessary costimulation for humoral immune responses after intraperitoneal immunization in the presence of adjuvant. Consequently, whereas absence of B7-2 costimulation at the early stages of our model would tend to have a protective effect on the development of a humoral immune response, compensatory B7-1 upregulation could at least partially support the pathogenesis of pulmonary inflammation.

Additionally, different sites of antigen presentation may influence the relative contributions of B7-1 and B7-2 to the development of subsequent immune responses. Dendritic cells (DC) have been shown to play a prominent role as APC in murine lung (35, 36), and recent work suggests that B7-1 on lung DC can provide a constimulatory signal to allogeneic T cells in the absence of B7-2 signaling but that B7-2 functions poorly except when B7-1 is also engaged (37). Therefore, the pathogenesis of pulmonary inflammation in our odel may be more dependent on B7-2 signaling in the early stages of the protocol, with B7-1- mediated costimulation assuming a more prominent role in the later aeroallergen challenge phase.

Additional studies have suggested that B7-1 and B7-2 can have complementary roles in mediating costimulation of T cells. In vitro study of human peripheral blood T cells demonstrated that both B7-1 and B7-2 transfectants efficiently costimulated anti-CD3 mAb-induced proliferation and the secretion of IL-2 and IFN- (38). Investigation in an in vivo model of type 2 mucosal immune response after oral infection of mice with nematode Heligmosomoides polygyrus has shown that either B7-1 or B7-2 ligand interactions can provide the required costimulatory signals that lead to T-cell effector function (39, 40). In a model of antigen-pulsed DC, B7-1 and B7-2 were demonstrated to be partially interchangeable ligands, inasmuch as either molecule alone was sufficient to initiate a primary humoral response in vivo (41). In vivo study of mice infected with Nippostrongylus brasiliensis suggested that either B7-1 or B7-2 could provide a sufficient costimulatory signal for induction of eosinophilia (42).

Considerable interest has been generated in discerning the relative contribution of B7-1 and B7-2 signaling to the pathogenesis of many autoimmune and infectious diseases, inasmuch as B7 costimulation has been shown to affect the production of both Th1 and Th2 cytokines, depending on the system studied. Together, our data support complementary roles for B7-1 and B7-2 in allergic pulmonary inflammation, with B7-2 costimulation playing a quantitatively more prominent role. Using a therapeutic strategy that exploits the differences in B7-1 and B7-2 costimulation could prove useful in modulating the pathogenesis of allergic pulmonary inflammation and AHR.