The role of iNKT cells in development of bronchial asthma: a translational approach from animal models to human


Omid Akbari PhD
Division of Immunology
Children's Hospital Boston
Harvard Medical School
Karp Research Laboratories
One Blackfan Circle
10th floor
MA 02115


The iNKT cell represents a unique T-lymphocyte sublineage that has been associated with a broad range of disease processes; including host defense against infectious disease, cancer immunity and allergic autoimmune diseases, such as asthma. Studies in both animal models and human subjects suggest that iNKT cells might significantly affect the course of asthma. This study discusses various aspects of iNKT cell function and how it might lead to an important therapeutic target in asthma (airway hyperreactivity).

Recent increases in asthma prevalence and its economic burden confirm asthma's position as a major public health problem particularly for children (1). The disease is thought to have a Th2-driven immunological basis which, through the elaboration of a variety of mediators, leads to the characteristic increase in mucus production, airway hyperreactivity (AHR) and airway inflammation (dominated by lymphocytes, eosinophils, basophils and mast cells) (2). Activation of mast cells can precipitate acute asthma attacks, but CD4+ Th2 cells secreting interleukin (IL)-4, IL-5 and IL-13 might also play a major role in orchestrating airway function. Th2 cells in particular are believed to cause part of the asthma phenotype, because: (i) IL-4 enhances the isotype switch to IgE, (ii) IL-5 enhances the growth and differentiation of eosinophils and (iii) IL-13 promotes both AHR, a cardinal feature of asthma and increases mucus production in the airways.

Th2-driven immune responses might have a critical role in the development of mouse models of asthma (3, 4), a Th2 response might not be sufficient for the induction of asthma. Th2-biased allergen sensitization can occur without asthma, e.g. only about one-third of individuals with allergic rhinitis develop asthma (5). Therefore, additional essential elements may be required for developing Th2 response in the lungs, distinct from allergen sensitization per se, for asthma to occur. The specific requirements for asthma within the respiratory tract or bone marrow have not been identified, but may involve structural responses of the lower respiratory tract to repeated injury (e.g. airway remodeling; 6, 7), or may involve immune system components that localize to the lower respiratory tract.

There are now substantial data indicating that iNKT cells are abundant in the thymus, spleen, liver and bone marrow, and can also be present in the airways. iNKT cells comprise a population of T cells that express cell surface markers characteristic of both NK cells and conventional T cells. Natural killer (NK) T cells can promote the polarization of conventional CD4 T cells into Th1 and Th2 cells, and NKT cells have been found to play an extremely important role in modulating different types of immune responses.

The expression of specific Th1 and Th2 cytokines by NKT cells is critical to these immunomodulatory roles. Thus, NKT cells are important for a wide variety of immune processes, such as: tumor rejection (8, 9); suppression of antitumour responses (10); immune privilege in the eye (11); tolerance to allografts and xenografts (12–14); fetal–maternal tolerance (15); protection against graft-vs-host disease (16); protection against pathogens such as Cryptococcus (17) and hepatitis B virus (18). Furthermore, NKT cells have been shown to exert strong regulatory activity in autoimmune diseases such as diabetes mellitus (19), experimental autoimmune encephalomyelitis (20) and oxazalone-induced colitis (21). iNKT cells are either CD4+ or double negative (lacking CD4 and CD8). iNKT cells express a limited T cell receptor (TCR) repertoire and express an invariant Vα14Jα281 TCR in mice and a Vα24Jα15 TCR in human beings (22). Through this TCR, NKT cells recognize glycolipid antigens presented by the nonpolymorphic major histocompatibility complex (MHC) class I-like protein CD1d. Most importantly, when activated, NKT cells rapidly produce large quantities of IL-4 and interferon (IFN)-γ. A recent study suggests that common signaling pathways, such as NFAT2, GATA-3 and Stat6 have different roles in iNKT cells than in conventional T cells, by which CD4+ iNKT cells express IL-4 independent of Stat6 and via increased NFAT2 activity (23). This rapid and novel production of cytokines is thought to allow NKT cells to exert strong regulatory activity, and recent studies suggest that iNKT cells exert potent activity in controlling the development of asthma, as discussed below.

NKT cells and airway hyperreactivity in mice

CD1d-deficient and Jα281−/− mouse strains, which both lack Vα14iNKT cells provided us with a convenient animal model to examine the link between the presence of Vα14iNKT cells and the induction of allergen-induced AHR. To answer this question, we first examined CD1d knockout (KO) mice, which lack the restriction element of NKT cells. We compared the development of allergen-induced AHR in CD1d KO mice with that in wild-type BALB/c mice, by sensitizing and challenging these mice with ovalbumin (OVA) to induce airway AHR. Wild-type BALB/c mice, when sensitized and challenged with OVA, as expected, developed severe AHR, as demonstrated by challenge of the mice with increasing concentrations of methacholine, which increases airway resistance. To our surprise, the CD1d KO mice failed to develop AHR when sensitized and challenged in the same way (Fig. 1). These findings were really quite striking to us, and therefore we tried several different methods for inducing AHR, and several different antigens, including bovine serum albumin and Aspergillus antigen. In all cases, while the wild-type BALB/c mice developed severe AHR, the CD1d KO mice had normal airway reactivity (24, 25). These findings have been confirmed by other laboratories (26).

Figure 1.

 CD1d−/− mice (open triangles) do not develop antigen-induced airway hyperreactivity (AHR) when compares with wild type (WT) control (filled triangles). Increasing concentrations of methacholine were used to measure AHR in sensitized mice 1 day after the last challenge with ovalbumin (OVA). Data are the mean ± SEM enhanced pause (Penh) values from five sensitized mice in each group.

Do CD1d KO mice have functioning airways?

The failure of the CD1d KO mice to develop AHR could have been due to a number of possibilities, for example, an intrinsic inability of these mice to develop AHR. To rule this out, we administered recombinant IL-13 (rIL-13), which directly affects bronchial smooth muscle and airway epithelial cells to the CD1d KO mice (27). We found that administration of rIL-13 to the CD1d KO mice induced severe AHR, as it did in wild-type BALB/c mice. These results indicated that the CD1d KO mice had functioning airways.

Failure of Jα281 KO mice to develop AHR

The iNKT cells were required for the development of AHR was confirmed by examining another iNKT cell-deficient strain, Jα281 KO mice. These mice lack the α-chain of the invariant TCR of NKT cells, and therefore lack iNKT cells. The Jα281 KO mice, like CD1d KO mice, when sensitized and challenged with OVA failed to develop AHR. The Jα281 KO mice also had reduced airway eosinophilia, whereas the wild-type BALB/c mice, as expected, developed severe AHR.

To confirm the absence of AHR in iNKT cell-deficient mice, AHR was also assessed by directly measuring airway resistance and its reciprocal, dynamic compliance in anesthetized, tracheostomized, intubated and mechanically ventilated mice. In mice sensitized and challenged with OVA, airway resistance increased in the wild-type BALB/c mice on challenge with methacholine, but did not increase in the CD1d KO mice or in the Jα281 KO mice. Dynamic compliance decreased in the wild-type BALB/c mice on challenge with methacholine, but did not change in the CD1d or Jα281 KO mice. This indicated AHR failed to occur in either the CD1d or Jα281 KO mice.

Adoptive transfer of iNKT cells reconstitutes AHR

To more clearly show that the lack of NKT cells was the specific cause for the failure of the CD1d KO and Jα281 KO mice to develop AHR, we reconstituted the Jα281 KO mice with iNKT cells. iNKT cells were purified with CD1d tetramers loaded with α-galactocylceramide (α-GalCer) from wild-type BALB/c mice and adoptively transferred into the Jα281 KO mice, which were then assessed for allergen-induced AHR. We found that recipient Jα281 KO mice reconstituted with iNKT cells developed severe AHR, whereas the nonreconstituted mice had normal airway reactivity. Furthermore, to determine whether the production of IL-4, IL-13 or IFN-γ by the NKT cells required to reconstitute AHR? We therefore adoptively transferred NKT cells purified from IL-4 KO, IL-13 KO, IL-4/IL-13 double KO, or wild-type mice, and then assessed the recipients for AHR. We found that NKT cells from wild-type BALB/c mice and from IFN-γ KO mice fully reconstituted AHR in the Jα281 KO mice, whereas iNKT cells from IL-4 KO and IL-13 KO mice partially reconstituted AHR (24). However, iNKT cells from IL-4 to IL-13 double KO mice failed to reconstitute AHR in the Jα281 KO mice. These results indicated that the induction of AHR required the production by iNKT cells of both IL-4 and IL-13, but not IFN-γ.

Two independent studies by Hachem et al. and Morishima et al. recently confirmed the significant role of iNKT cells and asthma. They demonstrated that a single systemic (i.v.) or local (i.n.) administration of α-GalCer 1 h before or at the time of the first airway challenge to OVA-sensitized mice abrogates AHR and decreases airway eosinophilia, Th2 cytokine production in bronchoalveolar lavage (BAL), and specific anti-OVA IgE antibodies (28, 29). Both authors suggested that α-GalCer suppresses allergen-induced eosinophilic airway inflammation, possibly by inducing a Th1 response. However, those observations could be due to induction of anergy as single administration of the α-GalCer was shown to induce long-term NKT cell unresponsiveness and anergy in mice, which is reversible by administration of IL-2 (30, 31).

From animal models to humans

Observations in mice cannot accurately predict immunological processes that occur in humans, though it appears that the implications of these studies of iNKT cells in mice might be relevant to our understanding of human asthma. However, in this case the striking similarity between the amino acid sequences of CD1d in mouse and human, we proposed that the observed requirement for iNKT cells in mice might be relevant in humans. Thus, in order to determine whether iNKT cells play an important role in human asthma, we studied the frequency of CD1d-restricted NKT cells in the BAL, lung biopsies and blood of patients with persistent asthma.

Human subjects characteristics

We studied subjects with moderate persistent asthma, six normal healthy subjects and five patients with sarcoidosis, a respiratory inflammatory disease in which large numbers of CD4+ Th1 cells are present in the lungs (32, 33). All asthmatic subjects (n = 14) took albuterol on an as needed basis. Nine took inhaled corticosteroids and long-acting β-agonists in a combination device for at least 6 months prior to bronchoscopy. One asthma subject took inhaled corticosteroids for 6 months. The remaining four asthmatic subjects did not take inhaled corticosteroids for at least 6 months prior to study, in spite of daily symptoms of asthma. No subject took anti-immunoglobulin (Ig)E therapy, and no subject had an exacerbation of asthma within the 3 months prior to the study. The four subjects who had not received inhaled corticosteroids for >3 months had a mean forced expiratory volume (FEV1) of 71% predicted as shown in Table 1. Ten of the asthmatic patients were evaluated for allergy. Seven (four of the six tested in the steroid-treated group and three of four tested in the steroid-untreated group) demonstrated elevated allergen-specific IgE to at least one of a panel of common allergens, and a history of bronchospasm after allergen exposure, indicating that most but not all of the asthmatic patients had allergic asthma. The atopic asthmatic subjects had higher total IgE (250 IU/ml) compared with both nonatopic asthmatic subjects (53 IU/ml) and control subjects (21 IU/ml). Although the steroid-treated asthmatic group had a higher mean IgE (250 IU/ml) compared with the steroid-untreated asthmatic group (118 IU/ml) this difference did not reach statistical significance. All five patients with sarcoidosis had Stage II disease (lymphadenopathy and parenchymal lung findings). All were Caucasian and had noncaseating granulomas on transbronchial biopsy with negative fungal and acid-fast smears and cultures. Their average duration of disease was 6 months. None took corticosteroid or other immunosuppressant agents for sarcoidosis.

Table 1.   Subject characteristics and percentage of/NKT cells in BAL
 Steroid-treated Asthma (n = 10)Steroid-untreated Asthma (n = 4)Sarcoldosis (n = 5)Healthy Controls (n = 6)
  1. Steroid-untreated patients had not received corticosteroids for more than 6 months. Steroid-treated subjects had taken inhaled fluticasone propionate (250 mug twice a day) for at least 6 months. Subjects with sarcoidosis had not taken corticosteroid therapy.

  2. Data are presented as mean ± SD.

  3. BAL, bronchoalveolar lavage; FEV1, forced expiratory volume; FVC, forced vital capacity.

Age (years)44 ± 1336 ± 548 ± 937 ± 18
Disease duration (years)20 ± 1116 ± 150.5 ± 0.1NA
FEV, (% predicted)84 ± 1871 ± 14112 ± 18101 ± 7
FEV,/FVC0.57 ±  0.070.71 ± 0.040.83 ± 0.030.80 ±  0.06
% of BAL CD3+ iNKT+75 ± 1078 ± 61.3 ± 0.140.9 ± 0.32
% of BAL CD4+ iNKT+60 ± 986 ± 120.5 ± 0.270.6 ± 0.17

Bronchoalveolar lavage findings

Bronchoalveolar lavage fluid from these subjects was analysed for the presence of CD4+, CD8+ and iNKT cells. First, we found a generally increased cellularity in the lavage from asthmatic and sarcoidosis patients compared with healthy subjects. There was an increase in the proportion of lymphocytes in sarcoidosis patients. In both the asthmatic and sarcoidosis subjects the majority of lymphocytes were CD4+. We next examined the BAL fluid for the presence of iNKT cells using α-GalCer loaded CD1d tetramers, which specifically bind to the invariant TCR of iNKT cells (34), and with a monoclonal antibody, 6B11 that specifically recognizes the CDR3 region of the Vα24Jα18 TCR of human iNKT cells (35). As shown in Table 1, a large number of cells in the BAL fluid from patients with asthma were iNKT cells. By contrast, virtually no iNKT cells were detected in BAL fluid from either the healthy subjects or the patients with sarcoidosis (36).

As iNKT cells can express the CD4 cell surface marker, and large numbers of CD4+ cells were present in the lungs of patients with asthma, we determined the fraction of the CD4+ T cells in BAL fluid of the asthmatic individuals that were iNKT cells. Surprisingly, we found that the majority of the CD4+ T cells in the BAL fluid of patients with asthma were indeed iNKT cells (36). In the steroid-treated asthmatic subjects, average of 60% and in steroid-untreated average of 86% of the CD4+ cells expressed the invariant TCR Vα24 as determined by tetramer staining, while in normal individuals and in patients with sarcoidosis <1% of the CD4 cells expressed the invariant TCR Vα24 (Table 1). In support of our observation, a recent study by Sen et al., demonstrated the presence of iNKT cells in lungs of patients with asthma and documented that asthmatic Vα24+ iNKT cells selectively express CCR9 (37).

Confocal microscopy findings

We next performed immunofluorescence and confocal laser scanning microscopy of biopsy specimens from patients with asthma. The lower left panel in Fig. 2 demonstrates using immunofluoresence microscopy that many lymphocytes in the lamina propria express the invariant TCR Vα24. In contrast, in sarcoidosis the lymphocytes did express CD4, but not Vα24, and therefore were not iNKT cells (36). Interestingly, the presence of iNKT cells in the lungs of patients with asthma did not appear to be significantly affected by inhaled corticosteroid therapy, because 10 of the 14 patients had been treated with topical inhaled corticosteroids for >6 months prior to bronchoscopy, yet the majority of pulmonary CD3+ cells in these patients expressed the invariant TCR of iNKT cells, similar to that observed in steroid-untreated patients (Table 1). Thus, CD4+ iNKT cells were virtually absent from the lungs of patients with sarcoidosis, but were present in high numbers in asthma, even in patients who take inhaled corticosteroids.

Figure 2.

 Significant numbers of iNKT cells are present in the bronchoalveolar lavage (BAL) fluid of patients with asthma. Cells from the BAL fluid of a patient with asthma were stained with both fluorescein isothiocyanate (FITC)-conjugated monoclonal antibody against iNKT T cell receptor (TCR; 6B11) and phycoerythrin (PE)-conjugated α-galactosylceramide (α-GalCer) loaded CD1d tetramer and analysed for double positive cells, gated on CD3+ cells (upper panel). Laser confocal images of bronchial biopsies from a patient with asthma demonstrate that high number of cells stained with anti-6b11–PE conjugate in red (lower panel).

Functional studies on human NKT cells

We assessed the cytokine production by iNKT cells, using two techniques: intracellular cytokine assays and enzyme-linked immunosorbent assay (ELISA). iNKT cells from the lungs of patients with asthma produced both IL-4 and IL-13, but very little IFN-γ (36). This mirrored our findings in mouse models of asthma. In contrast, iNKT cells in the peripheral blood of all of our subjects (asthmatic, sarcoidosis and normal individuals) produced all three cytokines, suggesting the compartmentalization of one subset of iNKT cells (i.e. those producing Th2 cytokines) in the lungs of patients with bronchial asthma. Thus, the iNKT cell population in the lungs of patients with asthma appeared to be distinct from sarcoidosis patients and normal subjects. Whereas in the peripheral blood of asthmatic patients, normal individuals and sarcoidosis patients only about 40% of the iNKT cells were CD4+ (approximately 50% were CD4/CD8 double negative and 3% were CD8+), in lungs of asthmatic subjects the vast majority (>95%) of the iNKT cells co-expressed CD4+ iNKT cells that were either CD8+ or CD4/CD8 double negative were rare. In patients with sarcoidosis we observed low number of total iNKT cells in peripheral blood when compare to patients with asthma or normal controls. This observation is in support of recent studies demonstrating decreased number of iNKT cells in patients with nonremitting sarcoidosis (38, 39).

Potential glycolipids that can activate NKT cells in lungs

Beside α-GalCer, a series of new glycolipid ligands for iNKT cells have been recently identified, including synthetic and naturally occurring antigens. The naturally occurring antigens isoglobotrihexosylceramide (iGb3; 40) and GSL-1′, a glycosphingolipid antigen from Sphingomonas bacteria that is structurally related to α-GalCer (41, 42), stimulate iNKT cells less strongly than α-GalCer and might be good candidates for immunotherapy (25, 30).

But how about environmental allergens? A study by Bilenki et al., suggest that a common airborne human allergen, ragweed, activate NKT cells and play an important role in the outcome of the allergic responses (43). The authors showed that airway eosinophilic inflammation induced by ragweed and was significantly reduced in iNKT-deficient CD1 KO mice, which is correlated with impaired IL-4 and eotaxin production. Furthermore, in the same study, the CD1 KO mice displayed a significant decrease in serum allergen-specific and total IgE levels in their sera compared with BALB/c mice. In another study by Agea et al., it was demonstrated that plant pollens could play an important role as an environmental antigens to stimulate allergic responses, and it was suggested that pollen lipids may be recognized as antigens by human T cells through a CD1-dependent pathway (44). In this study, T-cell clones specific for pollen's phospholipids possessed functional properties similar to those of regulatory T cells, secreted both Th1 and Th2 type cytokines and displayed helper activity for IgE production. Both studies have introduced natural antigens, which are more physiologically relevant as natural ligands for CD1d-mediated NKT cell activation and can potentially induce NKT cell-dependant allergic diseases and asthma. Figure 3 demonstrating a simplified model of the possible interactions between polarized Th2 responses, NKT cells and regulatory T cells upon entry of allergen into the lungs.

Figure 3.

 A simplified schematic model of the possible interactions between polarized Th2 responses, NKT cells and regulatory T cells (TR).The model shows the cytokines produced by each cell type and how they positively (indicated by a plus sign) and negatively (indicated by a minus sign) regulate each other.

Conclusion and future directions

Our results in both animal models of AHR and human subjects with bronchial asthma strongly suggest that iNKT cells play a prominent role in the pathogenesis of AHR and asthma. Although we have learned much about mouse models of asthma over the past 5 years, progress in the development of effective new compounds for asthma therapy has been comparatively slow. Part of the success of approaches to our understanding of asthma depends on a detailed understanding of how the immune system, in particular Th2 allergic responses, is regulated. More research in this area may guide approaches in primary prevention of allergies because our understanding of this area is becoming very detailed. Recently, the cornerstones of asthma therapy (β2-agonists and inhaled corticosteroids) provide symptomatic relief and some physiological improvements in most patients with asthma; however, there is still no cure for asthma and asthma symptoms return as soon as corticosteroid therapy is withdrawn. Future immunotherapies might focus on control of the immune response to infections and allergens, particularly in patients whose condition is not satisfactorily controlled by currently used approaches. The essential role of the iNKT cells makes them an attractive target for asthma therapy. iNKT cells are an exceptional target for immunotherapy, because of their swift and directed activation of the adaptive immune system. However, much more needs to be learnt about their mode of activation, migration and function before targeted therapy is developed.