Asthma is a chronic inflammatory disease of the lung with airflow obstruction and bronchospasm, characterized by pulmonary eosinophilia, airway remodeling, increased airway hyperresponsiveness to environmental stimuli, and excessive Th2-type cytokine production. Recent studies indicate that crosstalk between the innate and adaptive immune systems is crucial for this disease. We and others have showed that the Dok (downstream of tyrosine kinases) family adaptors, Dok-1, Dok-2, and Dok-3, play essential roles in negative regulation of a wide variety of signaling pathways in both innate and adaptive immunities. Here, histopathology and bronchoalveolar lavage fluid (BALF) cellularity showed spontaneous pulmonary inflammation in Dok-1−/−Dok-2−/−Dok-3−/− (TKO) mice, but not in Dok-1−/−Dok-2−/− or Dok-3−/− mice, with hallmarks of asthma, including eosinophilia, goblet cell hyperplasia, and subepithelial fibrosis. Consistently, TKO mice, but not the other mutants, showed increased airway hyperresponsiveness to methacholine inhalation. In addition, Th2-type cytokine concentrations in BALF were increased in TKO mice. These findings provide strong evidence that Dok-1, Dok-2, and Dok-3 cooperatively play critical anti-inflammatory roles in lung homeostasis.
The cytoplasmic protein Dok-1 was originally identified as an adaptor in hematopoietic cells and was further showed to be a negative regulator of cell proliferation (Carpino et al. 1997; Yamanashi & Baltimore 1997; Yamanashi et al. 2000; Songyang et al. 2001). Indeed, mice lacking both Dok-1 and its closest homolog, Dok-2, but not those lacking either one alone, often develop a myeloproliferative disorder (Niki et al. 2004; Yasuda et al. 2004). Due to this genetic, functional, mechanical, and structural redundancy, it is believed that Dok-1 and Dok-2 have virtually identical roles (Niki et al. 2004; Yasuda et al. 2004; Mashima et al. 2009). The Dok family has been expanded to seven members, Dok-1 to Dok-7, which share structural similarities characterized by the NH2-terminal Pleckstrin homology and phosphotyrosine binding motifs, followed by the src homology 2–binding motifs in the COOH-terminal moiety (Okada et al. 2006). Among these members, only Dok-1, Dok-2, and Dok-3 are expressed preferentially in hematopoietic cells and comprise a closely related subgroup with regard to primary structure (Mashima et al. 2009). Accumulating evidence indicates in general that these Dok proteins are negative regulators of many activities of hematopoietic cells, including proliferation, survival, and cytokine production. For instance, we previously found that these Dok proteins cooperatively suppress macrophage proliferation both in vitro and in vivo (Mashima et al. 2010). Interestingly, Dok-1−/−Dok-2−/−Dok-3−/− (TKO) mice displayed augmented proliferation of macrophages in the lung (Mashima et al. 2010). However, the lung histopathology was not further examined.
Asthma is a heterogeneous inflammatory disorder of the airways in the lung, associated with chronic airway inflammation and increased airway hyperresponsiveness to environmental stimuli, both of which give rise to recurrent wheezing, coughing, and shortness of breath (Kumar et al. 2002; Galli et al. 2008; Kim et al. 2010). In addition, goblet cell hyperplasia and subepithelial fibrosis increase mucous production and airway wall thickness, together contributing to the pathogenesis of asthma (Bousquet et al. 2000). This disease is induced by many environmental factors, such as allergens, infection, inhaled particles, nonsteroidal anti-inflammatory drugs, cold air, and even exercise (Kim et al. 2010). In general, allergic and nonallergic asthma are induced by the adaptive and innate immune responses, respectively. However, these different pathways to asthma often coexist, and crosstalk between the two immune systems is known to be crucial for the disease, a reason for its heterogeneous nature (Cockcroft & Davis 2006; Galli et al. 2008; Lloyd & Hessel 2010; Moffatt et al. 2010). Therefore, it is important to understand how asthma develops not only in the presence, but also in the absence, of particular allergens.
In addition to its histological signature, asthma is also associated with characteristic molecular and cellular pathology, including augmented production of Th2-type cytokines and eosinophilia. The Th2-type cytokines IL-4, IL-5, and IL-13 initiate and promote pathogenesis (Bochner et al. 1994). For instance, IL-13 induces epithelial cell hyperplasia and mucus production from goblet cells (Grunig et al. 1998). In addition, IL-13 induces epithelial cells to produce eotaxin, a chemokine that stimulates migration of eosinophils, contributing to eosinophilia (Zhu et al. 1999). Also, IL-5 regulates eosinophil responses by promoting eosinophil differentiation, activation, and survival (Webb et al. 2000). IL-4 drives Th2 lymphocyte polarization and induces IgE class switching in B cells, facilitating allergic asthma (Burstein et al. 1991; Perkins et al. 2006). It should be noted that augmented production of the Th2-type cytokines does not necessarily indicate Th2 cell hypertrophy, because other types of cells such as basophils and mast cells can also produce these cytokines (Min et al. 2004; Gessner et al. 2005).
Given that Dok-1, Dok-2, and Dok-3 are negative regulators of hematopoietic cell signaling pathways, including those involved in the adaptive and innate immune systems (Mashima et al. 2009), it is hypothesized that these adaptors might act as negative regulators in pulmonary inflammation. Here, we investigate the lung histopathology of TKO mice, in which we previously reported the above-mentioned abnormal accumulation of macrophages.
Pulmonary inflammation and eosinophilia in TKO mice
To understand the nature of the lung pathology in TKO mice, we first compared lung histology in TKO mice and wild-type (WT) controls, together with Dok-1−/−Dok-2−/− (Dok-1/2 DKO) and Dok-3−/− (Dok-3 KO) mice. All the mice examined in this study were 8–12 weeks of age and had the genetic background of C57BL/6. Lung sections from WT, Dok-1/2 DKO, and Dok-3 KO mice showed no significant infiltration; however, the TKO lung sections had patchy cellular infiltrations (Fig. 1A). These infiltrations include macrophages and eosinophils in the alveoli and airways (Fig. 1B,C). In addition, the alveoli of TKO mice contained long, thin, rectangular, and needlelike crystals (Fig. 1D). Moreover, we found that the infiltrated macrophages had crystalline inclusions (Fig. 1D), a characteristic feature of Th2-type inflammation (Harbord et al. 2002; Rauh et al. 2005). To further examine infiltrated cell types, we investigated cellularity in bronchoalveolar lavage fluid (BALF). Consistent with the histology, WT, Dok-1/2 DKO, and Dok-3 KO mice showed similar BALF cellularity, but TKO mice had significantly increased numbers of macrophages and eosinophils, in addition to neutrophils and lymphocytes (Fig. 2A–E). These results indicate pulmonary inflammation with eosinophilia in the TKO lung, implying an asthma-like airway disorder.
Airway remodeling in TKO mice
Aside from cellular infiltrations, asthma is associated with airway remodeling, which includes goblet cell hyperplasia and subepithelial fibrosis (Lambrecht & Hammad 2012). Although Alcian blue stains acidic mucin, a major component of mucus, we found no positively stained cells in the bronchiolar epithelium of WT, Dok-1/2 DKO, and Dok-3 KO mice (Fig. 3A). By contrast, we found many airway mucus-producing goblet cells (blue stained cells) in the TKO epithelium, indicating goblet cell hyperplasia. In addition, Alcian blue staining of the lung sections showed thickened epithelia in TKO mice, but not the other mutants or WT controls (Fig. 3A). Consequently, the size of the airway lumen was reduced in TKO mice. To examine fibrosis in the lung, we visualized collagen in blue with Masson's trichrome staining. A small amount of wispy blue staining of collagen could be seen in the airway walls of WT, Dok-1/2 DKO, and Dok-3 KO mice. However, strongly enhanced collagen deposition was readily apparent in the subepithelial area and adventitia of the airways of TKO mice (Fig. 3B). These findings together indicate that Dok-1, Dok-2, and Dok-3 act cooperatively to maintain airway homeostasis, and the combined loss of these adaptor proteins induces asthma-like airway remodeling, demonstrating the essential roles of these adaptors.
Increased airway hyperresponsiveness in TKO mice
Airway remodeling in asthma is involved in augmented airway hyperresponsiveness to various stimuli, which is causally associated with the pathogenesis (Cockcroft & Davis 2006). Thus, we assessed lung function as airway resistance in response to methacholine, which is clinically used as a stimulus for diagnosis of the disease. The forced oscillation test using plethysmograph showed normal levels of baseline airway resistance in WT and Dok mutant mice examined in the absence of methacholine (Fig. 4A and data not shown). However, TKO mice but not the other mutants showed significantly increased resistance to methacholine as compared to the WT controls, indicating increased airway hyperresponsiveness in the TKO lung. Calculation of PC200 values, the concentration of methacholine that gives 200% of the baseline resistance value, confirmed this phenotype (Fig. 4B). Again, these findings indicate cooperative roles for the Dok adaptors in homeostasis of lung function.
Augmented production of Th2-type cytokines in the TKO lung
Increased production of Th2-type cytokines is important in the pathogenesis of asthma (Lloyd & Hessel 2010). To understand the molecular basis of asthma-like pulmonary inflammation in TKO mice, which is causally associated with airway remodeling and increased airway hyperresponsiveness, we examined the cytokine profiles in BALF. Measurements using ELISA showed significant increases in the Th2-type cytokine (IL-4, IL-5, and IL-13) concentrations in BALF prepared from TKO mice, as compared to those from the WT controls (Fig. 5A–C). However, levels of the Th1 cytokine IFN-γ were similar in BALF from TKO and WT mice (Fig. 5D). These findings suggest that asthma-like pulmonary inflammation in the TKO lung develops, at least in part, due to aberrant production of Th2-type cytokines.
Dok-1, Dok-2, and Dok-3 adaptor proteins are involved in a wide range of immune signaling pathways, including those downstream of receptors for antigens, immunoglobulins, cytokines, chemokines, cell adhesion molecules, and pathogen-associated molecular patterns (Mashima et al. 2009). Consistently, mice lacking one or two of these proteins showed significant defects in adaptive and/or innate immune responses and immune cell development, including lupus-like renal disease and myeloproliferative disorder observed in Dok-1/2 DKO mice (Niki et al. 2004; Yasuda et al. 2004; Shinohara et al. 2005). These observations generally indicate that Dok-1, Dok-2, and Dok-3 are negative regulators of immune cell activities. As mentioned above, Dok-1 and Dok-2 are the closest homologs within the Dok family and play virtually the same roles. For instance, these Dok adaptors recruit p120 rasGAP, a potent negative regulator of Ras, upon tyrosine phosphorylation and suppress the Ras–Erk signaling pathways (Yamanashi & Baltimore 1997; Di Cristofano et al. 1998; Songyang et al. 2001; Shinohara et al. 2005). Although Dok-3 suppresses Erk and JNK, it does not bind to p120 rasGAP (Cong et al. 1999; Robson et al. 2004; Honma et al. 2006; Ng et al. 2007). Instead, it binds to Grb2 and SHIP-1 to exert negative regulation. Therefore, Dok-1/2 and Dok-3 may play not only redundant but also distinct roles in immune regulation.
Previously, we showed that TKO mice, but not Dok-1/2 DKO and Dok-3 KO mice, on the 129/SvJ and C57BL/6 mixed background often develop aberrant accumulation of macrophages in the lung, implying a cooperative role for these adaptors in pulmonary inflammation (Mashima et al. 2010). To define the nature of the pulmonary pathology and contributions of Dok-1/2 and Dok-3 to the pathogenesis, we investigated the lung histology and airway function of TKO, Dok-1/2 DKO, Dok-3 KO, and WT mice. To minimize any effects from genetic background, all mice were analyzed on the C57BL/6 background in this study.
H&E staining of the lung sections showed infiltration of eosinophils and macrophages in the airway and parenchyma of TKO mice, but not the other mutants (Fig. 1). In addition, BALF analysis confirmed the accumulation of inflammatory cells, particularly eosinophils, specifically in TKO mice (Fig. 2). Pulmonary eosinophilia is a major feature of Th2-type inflammation commonly seen in patients with asthma and in its animal models, although alveolar infiltration of macrophages is not frequently seen in patients with asthma. Mucus hyperproduction is also an important component of Th2-type inflammation and airway remodeling both in patients with asthma and in its mouse models (Shum et al. 2008). Consistently, Alcian blue staining showed the increased numbers of mucin-secreting goblet cells in the airways specifically of TKO mice (Fig. 3A). With regard to airway remodeling, thickened epithelia were also seen in TKO mice, but not in the other mutants (Fig. 3A,B). In addition, Masson's trichrome staining showed subepithelial fibrosis again only in TKO mice (Fig. 3B).
Because goblet cell hyperplasia, which leads to airway mucous hyperproduction and thickening of epithelium, reduces the size of the airway lumen and subepithelial fibrosis leads to altered structure and abnormal mechanical properties, airway remodeling found in TKO mice suggests airway dysfunction. Indeed, TKO mice, but not the other mutants, showed significantly increased airway hyperresponsiveness to methacholine, a characteristic dysfunction in asthmatic patients and their models. These findings together indicate that Dok-1/2 and Dok-3 cooperatively play anti-inflammatory roles in pulmonary homeostasis. The observation that asthma-like airway inflammation in TKO mice develops spontaneously in the absence of overt allergen stimulation suggests involvement of innate immunity. However, adaptive immune responses to antigen(s) that is nonpathogenic in WT mice could also contribute to the pathogenesis in TKO mice, in which a wide range of immune signaling pathways are up-regulated (Mashima et al. 2009).
The mechanisms underlying asthma have been extensively studied, and critical roles for Th2-type cytokines in the lung have been highlighted (Lloyd & Hessel 2010). Therefore, it is likely that aberrant production of Th2-type cytokines in TKO mice facilitates asthma-like airway inflammation. It should be noted that mice lacking the src homology 2 domain–containing inositol 5-phosphatase SHIP-1 showed similar defects to those found in TKO mice (Oh et al. 2007). Namely, SHIP-1 KO mice show (i) pulmonary infiltration of macrophages, lymphocytes, neutrophils, and eosinophils; (ii) goblet cell hyperplasia and subepithelial fibrosis; and (iii) exaggerated production of Th2-type cytokines in BALF. Like Dok-1/2/3 proteins, SHIP-1 acts as a negative regulator in a variety of immunoreceptor-mediated signaling pathways. Moreover, Dok-1/2/3 adaptors each recruit SHIP-1 upon tyrosine phosphorylation, suggesting functional interaction in cellular signaling (Mashima et al. 2009). It is tempting to speculate that common target pathway(s) of these negative regulators could be essential for lung homeostasis.
Although Dok-1/2/3 proteins are preferentially expressed in hematopoietic cells, these adaptors may play roles in nonhematopoietic cells. For instance, Dok-1 expression is reported in airway epithelial cells (Lee et al. 2012), fibroblasts (Zhao et al. 2006), adipocytes (Hosooka et al. 2008), and neuronal cells (Smith et al. 2004). Moreover, Dok-1/2/3 expression was reported in bronchoalveolar stem cells in the lung (Berger et al. 2010). In addition, lentivirus-mediated over-expression of Dok-1 in airway epithelial cells significantly reduced both inflammation and airway remodeling induced by ovalbumin (OVA) exposure (Lee et al. 2012), although naïve Dok-1/2 DKO and Dok-1 KO mice did not show asthma-like disorder (Figs 1-5 and data not shown). To evaluate the contributions of the Dok-1/2/3-deficient hematopoietic cells versus other TKO cells to the disease, we are currently investigating whether bone marrow transplantation from TKO mice to WT mice, or vice versa, could transfer the asthma-like disease.
In conclusion, this study shows that TKO mice develop pulmonary infiltration of inflammatory cells, including eosinophils, and show typical airway remodeling, which is associated with increased airway hyperresponsiveness and augmented production of Th2-type cytokines in BALF, together indicating an asthma-like pulmonary disorder. Because current therapies for asthma have little effect on airway remodeling even though they are beneficial, at least in part, in controlling symptoms and airway inflammation (Doherty et al. 2011), there remains a need for the development of more effective and more comprehensive therapies. Elucidation of molecular mechanisms underlying asthma-like airway inflammation and remodeling in TKO mice may open new possibilities for the treatment of asthma.
Mice were housed on a 12:12-hour light–dark cycle in specific pathogen-free conditions with free access to water and standard mouse chow in the animal facility of the Institute of Medical Science, the University of Tokyo. The experimental protocols have been approved by the institutional animal committee. The C57BL/6 mouse strain was acquired from Japan CLEA. C57BL/6-congenic TKO mice were generated from Dok-1/2 DKO mice (Yasuda et al. 2004) and Dok-3 KO mice on a C57BL/6 genetic background, which had been established by backcrossing at least 9 times. Dok-3 mice on a C57BL/6 and 129/Sv mixed background were kindly provided by Brian Seed (Yang & Seed 2003; Mashima et al. 2010).
Lung tissues were fixed in 10% formalin in neutral PBS. Samples embedded in paraffin were sectioned at 4-μm thickness and stained with hematoxylin and eosin (H&E), Alcian blue, and Masson's trichrome for the evaluation of pulmonary inflammation and airway remodeling (Doherty et al. 2011).
Bronchoalveolar lavage fluid (BALF) analysis
Bronchoalveolar lavage fluid was collected as described elsewhere (Oboki et al. 2010). In brief, mice were intubated with a 22-G blunt needle (NIPRO, Osaka, Japan), and 1 mL of HBSS with 2% FCS was injected through the needle into the lungs. After washing 3 times, BALF was collected and centrifuged at 400 g for 5 min, and the supernatant and the pellet were collected separately. The BAL cells were resuspended in 100 μL of HBSS with 2% FCS, and each cell type was counted with the automated hematology analyzer Sysmex XT-1800i (Sysmex corporation, Kobe, Japan). The separately obtained supernatants were used for the analysis of Th2-type cytokines as described below.
Examination of airway responsiveness
To assess the pulmonary function as resistance of the lung under physiological conditions, an invasive test using plethysmograph was carried out (Nakae et al. 2007; Oboki et al. 2010). First, mice were anesthetized with intraperitoneal injection of ketamine (100 μg/g body weight; Daiichi Sankyo Co., Ltd.; Tokyo, Japan) and xylazine (10 μg/g body weight; Sigma-Aldrich). Then, tracheotomy was carried out using a metal cannula (1 mm inside diameter, 8 mm in length), which was inserted into the proximal trachea of each mouse. Subsequently, mice were connected to a computer-controlled small animal ventilator (ELAN system; Buxco Research Systems, Wilmington, NC, USA), ventilated at a frequency of 150 breadths/minute with a positive end-expiratory pressure of 2 cmH2O. To obtain the baseline values of the resistance of lung (RL), mice were left ventilated, and the data of initial resistance were recorded for 3 min at 2-s intervals. During each run, data were collected using elan rc-4 software (Buxco). Then, an aliquot (10 μL) of methacholine (2-(acetyloxy)-N,N,N-trimethylpropan-1-aminiumtrimethylpropan-1-aminium; Sigma, St. Louis, MO, USA) diluted in sterile saline (Otsuka Pharmaceutical Co., Ltd.) at 0, 6.25, 12.5, 25, or 50 mg/mL was given to each mouse through trachea by an in-line nebulizer (AER-1100; Buxco). The RL value (cmH2O/mL/s) at each methacholine concentration was similarly collected for 3 min at 2-s intervals, averaged, and plotted as a function of methacholine doses. Airway responsiveness was assessed using the provocative concentration of methacholine required to increase the resistance of lung to 200% of baseline value (PC200) (Inoue et al. 2005).
Determination of cytokine production by enzyme-linked immunosorbent assay (ELISA)
The concentrations of cytokines (IL-4, IL-5, IL-13, and IFN-γ) in BALF were determined by ELISA using commercially available kits (eBioscience, San Diego, CA, USA).
Data were assessed by Student's t-test and were expressed as mean ± SEM. The differences between samples with P values of <0.05 were considered as statistically significant.
We thank R. F. Whittier for critically reading the manuscript and thoughtful discussions and Ms M. Zenibayashi for animal care. This work was supported by Grants-in-Aid (R.M., S.N., & Y.Y.), the Program for Improvement of Research Environment for Young Researchers (S.N.), The Special Coordination Funds for Promoting Science and Technology (S.N.) from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and grants from Japan Science and Technology Agency, PRESTO (S.N.).