Increased susceptibility to airway responses in CD40-deficient mice

Abstract

The interaction between CD40 and its ligand (CD154) is crucial for IL-12 production and effective humoral immunity such as IgE production. Although the interaction seems to play a crucial role in asthmatic inflammation, previous studies investigating the role of the CD40 and CD154 interaction in experimental animal models of asthma are complicated due to multistep reactions in developing asthma. Here, in order to investigate the role of CD40 in the effector phase in the development of airway responses, we used CD40-deficient mice backcrossed with mice transgenic for an ovalbumin (OVA)-specific TCR (TCRtg). Using intranasal OVA administration followed by aerosol inhalation of OVA, greater airway hyperreactivity and eosinophilia in bronchoalveolar lavage fluid (BALF) were observed in CD40-deficient mice backcrossed with TCRtg mice (CD40^–/–/ TCRtg mice), compared with control littermates (CD40^+/+/ TCRtg mice). CD4⁺ helper T cell subset analysis of lung draining lymph nodes revealed that the Th1 component was significantly decreased in CD40^–/–/ TCRtg mice. Airway hyperreactivity and airway eosinophilia significantly correlated with the predomination of Th2 cells. Cytokine measurements in BALF also showed decreased IL-12 and the predominance of Th2 cells in CD40^–/–/ TCRtg mice. These results suggest that CD40 may play a protective role in developing asthma in the phase after establishing specific memory T cells through the regulation of the balance between Th1 and Th2 cells presumably via induction of IL-12.

INTRODUCTION

Asthma is a complex syndrome with many clinical phenotypes, for example atopic and nonatopic. The major characteristics include bronchial hyperreactivity, and airway inflammation, which involve a number of different cell types such as eosinophils, basophils, mast cells, and, most importantly, Th2 cells. Th2 cells, secreting IL-4 and IL-5, play a central role in initiating and sustaining asthmatic responses by regulating the IgE production and the growth, differentiation, and recruitment of mast cells, basophils, and eosinophils [1–3]. Bronchial-biopsy specimens from patients with atopic and nonatopic asthma contain increased concentrations of mRNA for both IL-4 and IL-5 [4]. Thus, bronchial mucosa in patients with asthma contains an excess of activated Th2 cells, whether atopic or nonatopic. These results indicate that the imbalance between Th1 cells and Th2 cells contributes to the cause and evolution of asthma.

The interaction between CD40 on B cells and its ligand (CD154) on activated CD4⁺ helper T (Th) cells is known to be essential for humoral immune responses as well as cell-mediated immunity [5,6]. In vitro, stimulation of CD40 on B cells with an agonistic anti-CD40 antibody induces a proliferation of B cells and undergoes class switching to secrete IgE under the influence of IL-4 [7–9]. Because IgE synthesis is a hallmark of atopic airway disease, CD40–CD154 interaction may play important role in the development of asthma-associated inflammation. On the other hand, CD40–CD154 interaction is known to induce IL-12 [10,11]. Since IL-12 prevents generation of Th2 cell-mediated responses, it could be speculated that a lack of CD40–CD154 interaction predominates in Th2 cell-mediated diseases. Thus, the role of CD40–CD154 interaction in developing asthma could be bi-directional.

Studies using CD40 knockout (CD40^–/–) or CD154 knockout (CD154^–/–) mice showed complicated results. CD40^–/– mice receiving intraperitoneal sensitization with ovalbumin (OVA) followed by aerosol challenge did not produce OVA-specific IgE, IgG isotypes, or IgA, whereas peripheral blood eosinophilia, the recruitment of airway eosinophils, airway hyperreactivity (AHR) and morphological changes of the airways were similar to those observed in the wild-type (WT) controls [12]. On the other hand, CD154^–/– mice served with a similar sensitization showed significantly reduced airway eosinophilic inflammation [13]. In a recent study by Mehlhop et al.[14], bronchial hyperresponsiveness induced by the repeated inhalation of Aspergillus fumigatus was prevented in CD154^–/– mice but not in CD40^–/– mice. Since the development of asthma is a multistep immunological reaction, the role of CD40–CD154 interaction should be examined separately in each phase. Therefore, in order to investigate the role of CD40–C154 interaction in the effector phase of the development of airway allergic responses, we used the CD40^–/– mice, backcrossed with TCRtg mice that bore a transgenic TCR that recognized OVA_323-339 presented by I-A^d[15]. CD40^–/–/TCRtg mice were administrated OVA with an intranasal droplet, which was considered as a more physiological pathway. The accumulation of inflammatory cells, AHR and Th cell phenotypes were analysed in CD40^–/–/TCRtg mice and control littermates (CD40^+/+/TCRtg mice).

MATERIALS AND METHODS

Transgenic animals

The generation and initial characterization of CD40^–/– mice have been described previously [10,16]. CD40^–/– mice had been backcrossed onto the BALB/c background for at least 10 generations. Mice in which T cells bore a transgenic TCR specific for OVA_323-339 on a BALB/c genetic background were produced as described elsewhere [15]. We crossed both types of mice and produced CD40^–/–/TCRtg mice, genotyped by a standard PCR method. In this study, CD40^–/–/TCRtg mice were compared with control littermates (CD40^+/+/TCRtg mice), which were simultaneously treated.

Induction of airway responses

Mice were exposed to various doses (0·1 mg, 1 mg, 10 mg) of OVA (Sigma, St. Louis, MO, USA) dissolved in 20 µl of PBS or PBS alone with an intranasal droplet. To induce cellular infiltration into alveolar space, mice received an aerosol of 1% (w/v) OVA in PBS in a Plexiglas chamber for 1 h on 3 d after intranasal administration. The aerosol was generated by a nebulizer (TERUMO, Tokyo, Japan).

Measurement of AHR

We examined the AHR 24 h after OVA aerosol challenge following the intranasal OVA or PBS administration. Airway reactivity was assessed by incremental lung resistance (R_L) from anaesthetized mice inhalated with methacholine (Mch). After acceptable anaesthesia was achieved (pentobarbital sodium, 50 mg/kg), the trachea was isolated and cannulated using an 18-gauge needle fixed with a quick-dry glue as a tubing adapter. Mice were placed in a whole body plethysmograph-box (model PLY3114; Buxco Electronics Inc., Sharon, CT, USA) and ventilated with a respiratory ratio of 120/min and tidal volume 8 ml/kg, giving a pleural pressure of about 10 cmH₂O at baseline (Mouse Ventilator Model ‘687’; Harvard apparatus Inc., Holliston, MA, USA). Airway pressure was measured by a pressure transducer and air flow was measured with a transducer (model TRD4510 and model TRD 5100, respectively) connected to preamplifier modules (model Max2270; Buxco Electronics Inc.). On the assumption of a lamina flow in this system, lung mechanics were fit to the equation:

$$\text{P }= {\text{R}}_{\text{L}}\text{vV̇} + {\text{E}}_{\text{L}}\text{V̇} + \text{K}$$

where P is tracheal pressure, V̇ is the flow detected by the pneumotachometer attached to the plethysmograph-box, E_L is the lung elastance, V is the volume obtained by integration of V̇, and K is a constant. R_L was calculated as the change in pressure divided by change in flow (dP/d V̇) at the two time points (either rising or dropping) of 70% tidal volume in the volume curve using BioSystem XA software (model SFT1813; Buxco Electronics Inc.). The average of 5 stable measurements of R_L was adopted as a value. Mch responsiveness was obtained by exposing mice for 30 s to nebulized PBS, followed by sequentially incremental doses (6·25–100 mg/ml) of nebulized Mch and monitoring R_L. Results were expressed for each Mch concentration as the percentage of baseline R_L values after PBS exposure. The concentration which produces a half-maximal response (EC₅₀) of Mch for each animal group was calculated according to the following modified Michaelis-Menten equation using a least square fitting routine by Sigma Plot (SPSS Inc, Chicago, IL, USA):

$${\text{R}}_{\text{L }}= {\text{R}}_{\text{Lmax}}{\text{C}}^{\text{n}}/({\text{C}}^{\text{n}} +{\text{ EC}}_{{\text{50}}^{\text{n}}})$$

where R_L denotes the lung resistance observed with concentration (C), R_Lmax the maximum resistance, n the Hill coefficient.

Analysis of bronchoalveolar lavage fluid (BALF) and lung histology

After the measurement of AHR, BAL was performed to evaluate lung inflammation. Nucleated cells in the BALF were counted using a haemocytometer (Becton Dickinson and Company, Parsippany, NJ, USA), and the rest of the samples were prepared by Cytospin (Shandon, Cheshire, UK) and stained with Diff-Quik (Baxter Healthcare Corp., Miami, FL, USA). At least 300 cells were differentiated by light microscopy based on morphologic criteria. The concentrations of IL-4, IFN-γ and IL-12 p70 in the BALF were determined using ELISA kits (Endogen, Woburn, MA, USA for IL-4 and IFN-γ, and BioSource International Inc, Camarillo, CA, USA for IL-12). ELISAs were performed according to the manufacturer's protocols. Simultaneously, the lungs were excised from the mice, fixed in 10% formalin and embedded in paraffin. Sections were stained with haematoxylin and eosin (H&E). Furthermore, to observe eosinophilic infiltration, sections were stained in a solution of Chromotrope 2R (Sigma, St. Louis, MO, USA), which identify eosinophil granules, as previously described [17].

Analysis of intracellular cytokines

It is hypothesized that upon encountering inhaled antigen, airway dendritic cells migrate to the draining lymph nodes of the lung, and interact with naïve T lymphocytes in the process of initial priming and differentiation into either Th1 or Th2 cells. Hence, T cell subsets were observed in the draining lymph nodes, representing the primary response at the site of airway inflammation. After collecting BALF, draining lymph nodes (peribronchial and parathymic) were isolated and homogenized to make single cell suspensions in DMEM (2 mm l-glutamine [+], Gibco BRL, Grand Island, NY, USA) containing 10% FCS (JRH, Biosciences, Lenaxa, KS, USA). The cells were incubated for 5 h at 37°C with or without stimulation with 10 ng/ml phorbol 12-myristate 13-acetate (PMA) and 1 mm ionomycin (both from Sigma Chemical Co.). CD4⁺ T cells were labelled using APC-anti-CD4 [1 : 100] (PharMingen, San Diego, CA, USA). After washing twice with PBS, the cells were fixed and permeabilized with Cytofix/Cytoperm Kit (PharMingen) according to the manufacturer's instructions. Then, PE-anti-IL-4 [1 : 50] and FITC-anti-IFN-γ[1 : 100] (both from PharMingen) were added to the cell suspensions and incubated for 20 min at 4°C in the dark. After a further wash, each sample was analysed on a FACS scan apparatus using Cell Quest (Becton Dickinson, San Jose, CA, USA). The Th1/Th2 ratio was defined as the value of Th1 cell number divided by Th2 cell number.

Statistical analysis

Data are given as mean ± s.e.m. Data were analysed with statistical analysis software (StatView, Ver 5·0). Comparisons between the two groups were done by Student's t-test for dependent samples. Comparisons among more than three groups were done by one-way anova with Fisher's least-significant-difference test as a posthoc test. Correlations were estimated by least square fitting with simple regression. P < 0·05 was considered significant.

RESULTS

CD40-deficiency deteriorated eosinophilic airway inflammation

Figure 1 shows the BALF cells in CD40^–/–/TCRtg mice and CD40^+/+/TCRtg mice. Both types of animals exhibited increasing total cellular infiltration in response to an increase in the amount of intranasal OVA (Fig. 1a). No significant difference was seen in the increase of total inflammatory cells between CD40^–/–/TCRtg mice and CD40^+/+/TCRtg mice. The number of eosinophils in BALF increased in a dose-dependent manner in both CD40^–/–/TCRtg and CD40^+/+/TCRtg mice. The CD40^–/–/TCRtg mice showed a significantly greater number of eosinophils than CD40^+/+/TCRtg mice in 1 and 10 mg intranasal applications of OVA (Fig. 1b). Without aerosol inhalation of OVA, the numbers of eosinophils after administration of intranasal OVA were not significantly increased in both CD40^–/–/TCRtg (0·05 ± 0·03 × 10⁵/ml n= 3) and CD40^+/+/TCRtg mice (0·02 ± 0·01 × 10⁵/ml, n= 3). There was no marked difference in the increase of lymphocytes between the two experimental groups (Fig. 1c).

Fig. 1.

Dose-responsive effects of intranasal administration with OVA on cellular changes in bronchoalveolar lavage fluid (BALF). Both CD40^+/+/TCRtg mice (□) and CD40^–/–/TCRtg mice (▪) were intranasally administrated with PBS or titrated doses of OVA (0·1, 1, 10 mg) and received an OVA aerosol challenge 3 d later. 24 h later, lavage was performed and airway inflammatory cells were differentiated by morphological criteria. (a) Total inflammatory cell numbers (b) eosinophil numbers, and (c) lymphocyte numbers in BALF were counted. Data represent the mean ± s.e.m for groups of 3–7 mice (CD40^+/+/TCRtg with PBS: n= 4, CD40^–/–/TCRtg with PBS: n= 6, CD40^+/+/TCRtg with 0·1 mg OVA: n= 3, CD40^–/–/TCRtg with 0·1 mg OVA: n= 3, CD40^+/+/TCRtg with 1 mg OVA: n= 5, CD40^–/–/TCRtg with 1 mg OVA: n= 5, CD40^+/+/TCRtg with 10 mg OVA: n= 6, CD40^–/–/TCRtg with 10 mg OVA: n= 7). (a) No marked difference between CD40^+/+/TCRtg mice and CD40^–/–/TCRtg mice (b) *P < 0·02, **P < 0·001. (c) No significant difference between the two groups.

Histological evaluation of the lung/airways obtained from the mice with 10 mg OVA confirmed these findings (Fig. 2). In the sections stained with H&E, at a low magnification to provide an overall impression of the extent of the inflammatory response, no marked difference was seen between both groups. At a high magnification, however, there were intermingling eosinophils in the inflammatory infiltrate of CD40^–/–/TCRtg mice (Fig. 2b). Fur-thermore, eosinophils stained red with Chromotrope 2R were seen accumulating especially around the peribronchial and per-ivascular areas of CD40^–/–/TCRtg mice (Fig. 2d,f), whereas chromotrope-positive cells were hardly identified in the sections obtained from CD40^+/+/TCRtg mice (Fig. 2c,e).

Fig. 2.

Histological evaluation of the lung/airways inflammation. Both types of mice were administrated with 10 mg OVA followed by OVA challenge. Light photomicrographs of paraffin-embedded sections of lung tissues stained with H&E obtained from (a) CD40^+/+/TCRtg mice and (b) CD40^–/–/TCRtg mice at 24 h after OVA aerosolization. Although accumulation of total inflammatory cells in CD40^–/–/TCRtg mice seemed to be equivalent to that in CD40^+/+/TCRtg mice, the eosinophilic nature of the airway inflammation was clearly illustrated in CD40^–/–/TCRtg mice (b). Triangles in (b) indicate eosinophils at the area of inflammation. To clarify eosinophilic infiltration, the sections were stained with Chromotrope 2R (c and e, CD40^+/+/TCRtg mice; d and f, CD40^–/–/TCRtg mice). Eosinophils stained red were seen accumulating in the sections from CD40^–/–/TCRtg mice. Histological evaluation of the lung confirmed the deterioration of eosinophilic airway inflammation in CD40^–/–/TCRtg mice. Scale bars represent 50 µm.

CD40-deficiency induced higher AHR after high dose antigen administration

Figure 3 a shows that an intranasal administration of 10 mg OVA in CD40^–/–/TCRtg mice resulted in the development of a significantly higher AHR than CD40^–/–/TCRtg mice without OVA, CD40^+/+/TCRtg mice with 10 mg OVA, and CD40^+/+/TCRtg without OVA. Figure 3b showed the EC₅₀ values of CD40^+/+/TCRtg and CD40^–/–/TCRtg mice with or without a nasal administration of 10 mg OVA combined with or without further aerosol inhalation of OVA. The EC₅₀ of CD40^–/–/TCRtg mice with 10 mg OVA with further inhalation of OVA were significantly smaller than the other mice (Fig. 3b).

Fig. 3.

Measurement of airway hyperreactivity (AHR) to inhalated methacholine (Mch). (a) Airway reactivity was measured by determining changes in lung resistance (R_L) during exposure to sequentially incremental doses of nebulized Mch and represented as the percentage of baseline R_L after PBS exposure. Data represent the mean ± s.e.m for groups of 4–7 mice (□ CD40^+/+/TCRtg with PBS, n= 4; ○ CD40^–/–/TCRtg with PBS, n= 6; ▪ CD40^+/+/TCRtg with 10 mg OVA, n= 6; • CD40^–/–/TCRtg with 10 mg OVA, n= 7). (a) CD40^–/–/TCRtg mice with 10 mg OVA developed significant AHR. R_L of this group was significantly higher than that of other groups at each concentration higher than 12·5 mg/ml (*P < 0·01). There was no significant difference in the values of R_L with 10 mg OVA CD40^+/+/TCRtg, PBS CD40^+/+/TCRtg, and PBS CD40^–/–/TCRtg for each Mch concentration. (b) 50% effective concentration (EC₅₀) values for Mch among experimental groups. CD40^–/–/TCRtg mice with 10 mg OVA showed significantly smaller EC₅₀ than other animal groups (*P < 0·05).

Th2 predominance was related to AHR and eosinophilia

Since Th2 cells promote airway inflammation in asthma while Th1 cells are proposed to protect against allergic disease by damping the activity of Th2 effector cells, the idea that allergic inflammation in asthma arises from an imbalance between Th1 and Th2 cells is generally accepted [18]. To investigate the contribution of CD40–CD154 interaction on the development of an imbalance between Th1 and Th2 cells, we applied intracellular cytokine staining and flow cytometry analysis to T cells derived from lung draining lymph nodes. Th cell subtyping showed a significant predominance of lymph node Th2 cells in CD40^–/–/TCRtg mice with 10 mg OVA (Th1/Th2 ratio = 0·39 ± 0·13, n= 7) compared with CD40^+/+/TCRtg mice (Th1/Th2 ratio = 2·94 ± 0·45, n = 6) (P < 0·0001) (Fig. 4a).

Fig. 4.

Th2 cell dominance in CD40^–/–/TCRtg mice correlated with asthmatic responses. (a) Representative flow cytometry analysis in CD40^+/+/TCRtg mice (left) and CD40^–/–/TCRtg mice (right). Th1 cells were defined by staining for anti-IFN-γ without staining for anti-IL-4, and Th2 cells were defined by staining for anti-IL-4 without staining for anti-IFN-γ. Correlations between the Th1/Th2 ratio and airway reactivity (b) and between the Th2/Th1 ratio, the reciprocal of the Th1/Th2 ratio, and airway eosinophilia (c) in CD40^+/+/TCRtg mice (○) and CD40^–/–/TCRtg mice (•) administrated with 10 mg OVA. Th1/Th2 ratio was defined as the value of Th1 cell number divided by the Th2 cell number. (b) There was a significant correlation between the Th1/Th2 ratio and EC₅₀ of Mch (r = 0·94, P < 0·0001). (c) There was a significant correlation between the Th1/Th2 ratio and the number of eosinophils in BALF (r = 0·64, P < 0·05).

In order to elucidate the respective roles of an imbalance between Th1 and Th2 cells in airway responses, we examined whether there were any relationships between the value of Th1/Th2 ratio and AHR, and between the value of the Th2/Th1 ratio (the reciprocal of the Th1/Th2 ratio) and airway eosinophilia. We found a strong correlation between the Th1/Th2 ratio and EC₅₀ representing Mch induced AHR (r = 0·94, P < 0·0001) (Fig. 4b). Moreover, there was a significant correlation between the value of the Th1/Th2 ratio and the number of eosinophils in BALF (r = 0·65, P < 0·05) (Fig. 4c). These results suggested that the predominance of Th2 responses in CD40^–/–/TCRtg mice induced the development of eosinophilic AHR.

As for the levels of cytokines in BALF obtained from mice with 10 mg OVA, there was no marked difference in the levels of IL-4 characterizing Th2 activity between CD40^–/–/TCRtg mice and CD40^+/+/TCRtg mice (51·6 ± 13·5 pg/ml and 35·2 ± 9·7 pg/ml, respectively). In contrast, the levels of IFN-γ characterizing Th1 activity in CD40^–/–/TCRtg mice were significantly smaller than those of CD40^+/+/TCRtg mice (271 ± 70 pg/ml versus 706 ± 148 pg/ml, P= 0·008), indicating Th2 cell dominant infiltration of T lymphocytes to the airways in CD40^–/–/TCRtg mice. The IL-12 concentration of BALF in CD40^–/–/TCRtg mice was significantly decreased in CD40^–/–/TCRtg mice compared with CD40^+/+/TCRtg mice (1·1 ± 2·1 pg/ml versus 18·8 ± 6·8 pg/ml, P < 0·0001). Figure 5 summarizes the percentage cytokine concentrations in BALF compared with counter littermates.

Fig. 5.

Concentrations of cytokines in BALF of CD40^–/–/TCRtg mice (n = 6) treated with 10 mg OVA. Values of IL-4, IFN-γ, and IL-12 in CD40^–/–/TCRtg mice are expressed as a percentage of mean values of CD4^+/+/TCRtg mice treated with 10 mg OVA (n = 7, represented by the dashed lines). IFN-γ and IL-12 are significantly smaller in CD40^–/–/TCRtg mice compared with CD40^+/+/TCRtg mice (*P < 0·05 and **P < 0·01, respectively).

DISCUSSION

Our results show that intranasal administration with high doses of OVA followed by aerosol OVA challenge developed greater eosinophilic airway responses in CD40^–/–/TCRtg mice than CD40^+/+/TCRtg mice. CD40–CD154 interaction played a protective role in the asthmatic airway responses in our experimental protocol. To investigate the pure role of CD40–CD154 interaction in the development of the effector phase of antigen-induced airway responses, we recruited TCRtg mice in which CD45RB^low memory cells already expressed the transgenic TCR specific for OVA even without sensitization. Despite the fact that the mice were never exposed to OVA, these cells can be stimulated by OVA to proliferate and perform typical memory functions, such as secreting diverse lymphokines and provide cognitive help to B cells [19]. Thus, in TCRtg mice, CD4⁺ T cells are already primed with OVA and the involvement of CD40–CD154 interaction on CD4⁺ T cells priming was skipped. This suggests that our experiment protocol emphasizes the role of CD40 in asthmatic models, especially in the respondent effector phase after sensitization (sensitization phase).

As an effector phase, we used the one-time intranasal administration of OVA followed by inhalation of aerosol OVA. In this method, despite administration of high dose of OVA, we found the allergic response was weak in the presence of CD40 even with using TCRtg mice. Without a transgene of TCR specific for OVA, this method could not induce any asthmatic responses. Even using this weak allergic challenge, we found that CD40^–/–/TCRtg mice developed asthma like responses such as eosinophilia and high AHR. Thus, CD40–CD154 interaction played a protective role in the asthmatic airway responses in our experimental protocol.

Our result is markedly inconsistent with the result of Lei et al. [13] who reported that OVA-stimulated CD154^–/– mice developed less eosinophilia than the WT mice; pulmonary physiology was not assayed. Moreover, our result is also inconsistent with the result of Hogan et al.[12] in that CD40^–/– mice show a susceptibility to asthmatic hyperreactivity equal to the WT mice. Since we skipped the sensitization phase using TCRtg mice, the difference could be a result of the differential role of CD40–CD154 interaction between the sensitization phase and the effector phase. Taken together with our result, CD40–CD154 interaction may play an important role in asthma development in the sensitization (priming) phase, and may have an inhibitory effect on the development of asthma in the effector phase. In a recent study by Mehlhop et al.[14], bronchial hyperresponsiveness induced by the repeated inhalation of Aspergillus fumigatus was prevented in CD154^–/– mice but not in CD40^–/– mice, suggesting the role of the CD40–CD154 interaction in the sensitization phase is largely CD154-dependent. The final consequence of asthma development in mice with naïve T cells could be decided by the sensitization phase because it is necessary to establish sensitization to develop the effector phase response.

Another possibility for the difference between our result and the others is that it resulted from the different experimental protocol of sensitization and challenge (different pathway, antigen doses and times). Using the same CD154^–/– mice, Lei et al. [13] and Mehlhop et al. [14] showed different results in pulmonary eosinophilia. Mehlhop et al. speculated the difference arises from differences in the allergen used or from the distinct mechanism of antigen sensitization. In Lei's experiment, mice were sensitized by OVA intraperitoneal injection and were subjected to many repeated aerosolizations whereas Mehlhop's experiment used repeated inhalation of Aspergillus fumigatus. Not only the sorts and dose [20] of antigen but also the site of antigen exposure is known to cause a markedly different allergic inflammation [21]. In our study, TCRtg mice were administrated with a droplet of high dose OVA on the nasal mucosa surface and challenged by only one aerosolization. It is possible that the different method of OVA challenge produced a different result regardless of skipping the sensitization phase. Further study using our method is needed to clarify this point.

Although asthma is a complex pulmonary syndrome in which various inflammatory cells are involved, CD4⁺ T cells, in particular the Th2 subset, are believed to promote the allergic responses in the airways of subjects with asthma by releasing factors responsible for inflammatory cell recruitment into the airways. The total numbers of lymphocytes in BALF were increased in both CD40^+/+/TCRtg and CD40^–/–/TCRtg mice after OVA challenge, suggesting that the imbalance between Th1 and Th2 cells played an important role in hyperreactivity of CD40^–/–/TCRtg mice, in the same fashion as noted in patients with bronchial asthma. Therefore, we focused on the Th1/Th2 ratio, the Th1 cell number divided by the Th2 cell number, which is commonly used to evaluate the balance of cellular immunity and humoral immunity in daily practice. Th1/Th2 ratios from the lung draining lymph nodes were smaller in mice deficient in CD40 than in the WT mice. Together with each cytokine level in BALF, the induction of Th1 cells was affected while Th2 cells were maintained, resulting in a decrease in the Th1/Th2 ratio in CD40^–/– mice. This preferential induction of Th2 cells on CD40^–/– mice is consistent with the previous observations that CD40^–/– or CD154^–/– mice mount Th2 responses instead of Th1 responses because of a lack of IL-12 production by antigen presenting cells [10,22]. In our study, actually, the IL-12 level in BALF was significantly depressed in CD40^–/–/TCRtg mice (Fig. 5). Previous studies demonstrated the protective role of IL-12 in Th2 mediated diseases including asthma [23,24]. The treatment of animals with IL-12 inhibits Th2 cytokines synthesis in vitro and in vivo[25]. IL-12 treatment inhibits antigen-induced airway hyperresponsiveness in mice [26]. The mice lacking IL-12 showed higher eosinophil levels in BAL than parallel wild-type control mice [27], consistent with our result. There was a significant correlation between Th1/Th2 ratio and EC₅₀ representing Mch induced AHR (Fig. 4b). Furthermore, we also found a significant correlation between Th1/Th2 ratio and the number of airway eosinophilia (Fig. 4c). These results suggest that a relative imbalance of Th2 responses over Th1 responses deteriorates asthmatic airway responses due to impaired production of IL-12 in CD40^−/−/TCRtg mice.

Since Rackeman's clinical classification of asthma, it has been widely accepted that a subgroup of asthmatic patients are not demonstrably atopic, the so called ‘intrinsic’ variant of the disease [28]. In intrinsic (nonatopic) asthmatics, serum total IgE concentrations are within the normal range and there is no evidence of specific IgE antibodies directed against common allergens. These patients, accounting for a third of total asthmatics, are usually older than their allergic counterparts and have onset of symptoms in later life, often with a more severe clinical course. We preliminarily observed that IgE production was markedly impaired in CD40^–/–/TCRtg mice. Our asthma model using CD40^–/–/TCRtg mice administrated with an intranasal droplet of OVA appears to be similar to nonatopic asthmatic in terms of immunopathological entities such as Th2 cell dominant and eosinophilic AHR without serum IgE elevation. Although further studies are required to elucidate the role of CD40 on asthmatic phenomena, CD40 might be a target in the understanding of nonatopic (intrinsic) asthmatics.