Carbohydrates in allergy: from disease to novel immunotherapies.

Respiratory allergic disorders are a global public health problem that are responsible for substantial morbidity and healthcare expenditure. Despite the availability of allergen immunotherapy (AIT), its efficacy is suboptimal and regimens are lengthy, with a significant risk of potentially severe side effects. Studies on the recognition of allergens by immune cells through carbohydrate-lectin interactions, which play a crucial role in immune modulation and pathogenesis of allergy, have paved the way for improvements in AIT. We highlight innovative approaches for more effective and safer AIT, including the use of allergens conjugated to specific carbohydrates that bind to C-type lectins (CLRs) and sialic acid-binding immunoglobulin-type lectins (Siglecs) on immune cells to induce suppressive responses.

Modulating local airway immune responses to treat allergic asthma: lessons from experimental models and human studies.

With asthma affecting over 300 million individuals world-wide and estimated to affect 400 million by 2025, developing effective, long-lasting therapeutics is essential. Allergic asthma, where Th2-type immunity plays a central role, represents 90% of child and 50% of adult asthma cases. Research based largely on animal models of allergic disease have led to the generation of a novel class of drugs, so-called biologicals, that target essential components of Th2-type inflammation. Although highly efficient in subclasses of patients, these biologicals and other existing medication only target the symptomatic stage of asthma and when therapy is ceased, a flare-up of the disease is often observed. Therefore, it is suggested to target earlier stages in the inflammatory cascade underlying allergic airway inflammation and to focus on changing and redirecting the initiation of type 2 inflammatory responses against allergens and certain viral agents. This focus on upstream aspects of innate immunity that drive development of Th2-type immunity is expected to have longer-lasting and disease-modifying effects, and may potentially lead to a cure for asthma. This review highlights the current understanding of the contribution of local innate immune elements in the development and maintenance of inflammatory airway responses and discusses available leads for successful targeting of those pathways for future therapeutics.

Poly(trimethylene carbonate) and poly(D,L-lactic acid) modify human dendritic cell responses to staphylococci but do not affect Th1 and Th2 cell development.

Biomaterial-associated infections (BAIs) are frequent complications in the use of medical devices (biomaterials) correlated with considerable patient discomfort and high treatment costs. The presence of a biomaterial in the host causes derangement of local immune responses increasing susceptibility to infection. Dendritic cells (DCs) have an important role in directing the nature of immune responses by activating and controlling CD4+ T helper (Th) cell responses. To assess the immunomodulatory effect of the combined presence of biomaterials and Staphylococcus aureus (S. aureus) or Staphylococcus epidermidis (S. epidermidis), DC-mediated T cell proliferation and Th1/Th2 cell development were measured using an in vitro human cell system. Poly(trimethylene carbonate) (PTMC) and poly(D,L-lactic acid) (PDLLA) modified the production of the DC pro-inflammatory cytokines TNF-α, IL-6 and IL-23 in response to S. aureus and S. epidermidis. However, this modified cytokine production did not cause differences in Th1/Th2 cell polarisation, showing a Th1 cell predominance. In the absence of staphylococci, neither of the biomaterials induced DC-mediated T cell proliferation or Th1/Th2 cell polarisation. Moreover, either in the absence or presence of the biomaterials, S. aureus was a more potent inducer of DC cytokine secretion, T cell proliferation and Th1 cell development than S. epidermidis. In conclusion, although PTMC and PDLLA modulated DC cytokine responses to staphylococci, this did not alter the resulting Th cell development. This result suggested that, in this human cell model, Th1/Th2 cell responses were mainly determined by the species of bacteria and that PTMC or PDLLA did not detectably influence these responses.

Expression profiling and functional analysis of Toll-like receptors in primary healthy human nasal epithelial cells shows no correlation and a refractory LPS response.

BACKGROUND: Innate immune recognition via Toll-like receptors (TLRs) on barrier cells like epithelial cells has been shown to influence the regulation of local immune responses. Here we determine expression level variations and functionality of TLRs in nasal epithelial cells from healthy donors. METHODS: Expression levels of the different TLRs on primary nasal epithelial cells from healthy donors derived from inferior turbinates was determined by RT-PCR. Functionality of the TLRs was determined by stimulation with the respective ligand and evaluation of released mediators by Luminex ELISA. RESULTS: Primary nasal epithelial cells express different levels of TLR1-6 and TLR9. We were unable to detect mRNA of TLR7, TLR8 and TLR10. Stimulation with Poly(I:C) resulted in a significant increased secretion of IL-4, IL-6, RANTES, IP-10, MIP-1ß, VEGF, FGF, IL-1RA, IL-2R and G-CSF. Stimulation with PGN only resulted in significant increased production of IL-6, VEGF and IL-1RA. Although the expression of TLR4 and co-stimulatory molecules could be confirmed, primary nasal epithelial cells appeared to be unresponsive to stimulation with LPS. Furthermore, we observed huge individual differences in TLR agonist-induced mediator release, which did not correlate with the respective expression of TLRs. CONCLUSION: Our data suggest that nasal epithelium seems to have developed a delicate system of discrimination and recognition of microbial patterns. Hypo-responsiveness to LPS could provide a mechanism to dampen the inflammatory response in the nasal mucosa in order to avoid a chronic inflammatory response. Individual, differential expression of TLRs on epithelial cells and functionality in terms of released mediators might be a crucial factor in explaining why some people develop allergies to common inhaled antigens, and others do not.

Retinoic acid primes human dendritic cells to induce gut-homing, IL-10-producing regulatory T cells.

The vitamin A metabolite all-trans retinoic acid (RA) is an important determinant of intestinal immunity. RA primes dendritic cells (DCs) to express CD103 and produce RA themselves, which induces the gut-homing receptors α4ß7 and CCR9 on T cells and amplifies transforming growth factor (TGF)-ß-mediated development of Foxp3(+) regulatory T (Treg) cells. Here we investigated the effect of RA on human DCs and subsequent development of T cells. We report a novel role of RA in immune regulation by showing that RA-conditioned human DCs did not substantially enhance Foxp3 but induced α4ß7(+) CCR9(+) T cells expressing high levels of interleukin (IL)-10, which were functional suppressive Treg cells. IL-10 production was dependent on DC-derived RA and was maintained when DCs were stimulated with toll-like receptor ligands. Furthermore, the presence of TGF-ß during RA-DC-driven T-cell priming favored the induction of Foxp3(+) Treg cells over IL-10(+) Treg cells. Experiments with naive CD4(+) T cells stimulated by anti-CD3 and anti-CD28 antibodies in the absence of DCs emphasized that RA induces IL-10 in face of inflammatory mediators. The data thus show for the first time that RA induces IL-10-producing Treg cells and postulates a novel mechanism for IL-10 in maintaining tolerance to the intestinal microbiome.

Cholera toxin B suppresses allergic inflammation through induction of secretory IgA.

In healthy individuals, humoral immune responses to allergens consist of serum IgA and IgG4, whereas cellular immune responses are controlled by regulatory T (Treg) cells. In search of new compounds that might prevent the onset of allergies by stimulating this type of immune response, we have focused on the mucosal adjuvant, cholera toxin B (CTB), as it induces the formation of Treg cells and production of IgA. Here, we have found that CTB suppresses the potential of dendritic cells to prime for Th2 responses to inhaled allergen. When we administered CTB to the airways of naïve and allergic mice, it strongly suppressed the salient features of asthma, such as airway eosinophilia, Th2 cytokine synthesis, and bronchial hyperreactivity. This beneficial effect was only transferable to other mice by transfer of B but not of T lymphocytes. CTB caused a transforming growth factor-beta-dependent rise in antigen-specific IgA in the airway luminal secretions, which was necessary for its preventive and curative effect, as all effects of CTB were abrogated in mice lacking the luminal IgA transporting polymeric Ig receptor. Not only do these findings show a novel therapeutic avenue for allergy, they also help to explain the complex relationship between IgA levels and risk of developing allergy in humans.

Interactions between epithelial cells and dendritic cells in airway immune responses: lessons from allergic airway disease.

Micro-organisms constantly invade the human body and may form a threat to our health. Traditionally, concepts of defence mechanisms have included a protective outer layer of epithelia and a vigilant immune system searching for areas where the integrity of the outer layer may be compromised. Instead of considering these elements as two independent mechanisms, we should be treating them as a single integrated system. This review will present and discuss the role of local immune-competent cells and local epithelia in the recognition of potential pathogens and how the interaction between the two components may affect the initiation of the airway immune response. A concept emerges where airway mucosal dendritic cells act as integrators of both immunostimulatory and immunosuppressive signals that act within actively-involved mucosal tissue.

Human liver myeloid dendritic cells maturate in vivo into effector DC with a poor allogeneic T-cell stimulatory capacity.

We hypothesized that the relatively low immunogenicity of liver grafts might be related to a special maturation program of hepatic myeloid dendritic cells (MDC), yielding relatively immature effector MDC with weak allogeneic T-cell stimulatory capacity. To investigate whether maturation of human liver-derived MDC in vivo differs from maturation of MDC at another anatomical location, we compared the immunophenotypes and allogeneic T-cell stimulatory capacity of MDC from hepatic with those from inguinal lymph nodes (LN). MDC were purified by immunomagnetic selection from hepatic LN obtained from multi-organ donors (n = 8) and from inguinal LN of kidney transplant recipients (n = 7). MDC from hepatic LN had a significantly reduced capacity to stimulate allogeneic T-cell proliferation compared to MDC from inguinal LN. However, this was not due to an immaturity, since MDC from hepatic LN had significantly higher expressions of HLA-DR, CD80, and CD86 compared to MDC from inguinal LN. Hepatic MDC maturate in vivo to a mature type of effector MDC with relatively poor allogeneic T-cell stimulatory capacity.

Type I IFNs differentially modulate IL-12p70 production by human dendritic cells depending on the maturation status of the cells and counteract IFN-gamma-mediated signaling.

Type I IFNs (IFNalpha/beta) are approved for the treatment of a variety of diseases, including the autoimmune disease multiple sclerosis (MS). The proinflammatory cytokines IL-12 and IFN-gamma have been proposed to contribute to the pathogenesis of MS. Since dendritic cells (DCs) are recognized as major producers of IL-12p70 and promote the development of IFN-gamma-producing Th1 cells, we investigated the direct effect of IFNalpha/beta on monocyte-derived DCs at different stages of development. We demonstrate that IFNalpha/beta enhance IL-12p70 production by immature DCs but inhibit IL-12p70 production by mature DCs. Importantly, IFNalpha/beta strongly counteracted the IL-12-enhancing effect of IFN-gamma on DCs irrespective of their maturation status. Exposure of DCs to IFNalpha/beta during maturation does not affect their maturation or cytokine profile upon CD40 ligation. The differential modulatory effect of IFNalpha/beta on the IL-12-producing capacity of DCs and their cross-regulatory effect on IFN-gamma may reduce inflammatory processes and therefore be therapeutically effective in MS.

Analysis of the CD4+ T cell responses to house dust mite allergoid.

BACKGROUND: Modified allergen extracts (allergoids) with reduced IgE-binding capacity are successfully used in immunotherapy of atopic allergy. Their reduced T-cell stimulatory capacity is less well studied and is a subject of the present study. METHODS: We compared the ability of native house dust mite extract (Dermatophagoides pteronyssinus; HDM) and the glutaraldehyde-modified allergoid (HDM-GA) to induce the proliferation and cytokine production by fresh PBMC and by DC-stimulated polyclonal Th cells and HDM-specific Th cell clones. RESULTS: Freshly isolated T cells showed a partially reduced responsiveness to HDM-GA, differentially pronounced in different donors. HDM-specific Th cell clones prepared from three donors showed either a complete loss of reactivity to HDM-GA, or completely preserved responsiveness. The frequency of nonreactive clones was donor-dependent (2/3, 3/10 and 1/10). GA modification of HDM did not interfere with the cytokine production profile of HDM-specific T cell clones. CONCLUSIONS: The reduced stimulatory potential of HDM-GA results mainly from a loss of certain Th cell epitopes, rather than impaired allergen uptake and presentation, or induction of suppressive factors. Varying frequencies of allergoid-nonreactive HDM-specific Th cells may result in differential responses of individual patients to immunotherapy.

Prostaglandin E(2) is a selective inducer of interleukin-12 p40 (IL-12p40) production and an inhibitor of bioactive IL-12p70 heterodimer.

Interleukin-12 p70 (IL-12p70) heterodimer, composed of p35 and p40 subunits, is a major Th1-driving cytokine, promoting cell-mediated immunity. In contrast, IL-12p40 homodimer, secreted by APC in the absence of p35 expression, and free p40 monomer do not mediate IL-12 activity but act as IL-12 antagonists. Here it is reported that prostaglandin E(2) (PGE(2)), an inflammatory mediator with a previously known Th2-driving function, dose-dependently enhances the IL-12p40 mRNA expression and the secretion of IL-12p40 protein in human tumor necrosis factor-alpha (TNFalpha)-stimulated immature dendritic cells (DCs). This effect is selective and is not accompanied by the induction of IL-12p35 expression or by secretion of IL-12p70 heterodimer. Inability of TNFalpha/PGE(2) to induce IL-12p70 was not compensated by interferon gamma (IFNgamma), which strongly enhanced the lipopolysaccharide (LPS)-induced IL-12p70 production. In addition to the selective induction of IL-12p40 in TNFalpha-stimulated DCs, PGE(2) inhibited the production of IL-12p70 and IL-12p40 in DCs stimulated with LPS or CD40 ligand. These data suggest an additional level of the Th2-promoting activity of PGE(2), via selective induction of IL-12p40. Selective induction of IL-12p40 and suppression of bioactive IL-12p70 may have negative impact on anticancer vaccination with PGE(2)-matured DCs. (Blood. 2001;97:3466-3469)

IL-4 is a mediator of IL-12p70 induction by human Th2 cells: reversal of polarized Th2 phenotype by dendritic cells.

IL-12 is a key inducer of Th1-associated inflammatory responses, protective against intracellular infections and cancer, but also involved in autoimmune tissue destruction. We report that human Th2 cells interacting with monocyte-derived dendritic cells (DC) effectively induce bioactive IL-12p70 and revert to Th0/Th1 phenotype. In contrast, the interaction with B cells preserves polarized Th2 phenotype. The induction of IL-12p70 in Th2 cell-DC cocultures is prevented by IL-4-neutralizing mAb, indicating that IL-4 acts as a Th2 cell-specific cofactor of IL-12p70 induction. Like IFN-gamma, IL-4 strongly enhances the production of bioactive IL-12p70 heterodimer in CD40 ligand-stimulated DC and macrophages and synergizes with IFN-gamma at low concentrations of both cytokines. However, in contrast to IFN-gamma, IL-4 inhibits the CD40 ligand-induced production of inactive IL-12p40 and the production of either form of IL-12 induced by LPS, which may explain the view of IL-4 as an IL-12 inhibitor. The presently described ability of IL-4 to act as a cofactor of Th cell-mediated IL-12p70 induction may allow Th2 cells to support cell-mediated immunity in chronic inflammatory states, including cancer, autoimmunity, and atopic dermatitis.

Development of Th1-inducing capacity in myeloid dendritic cells requires environmental instruction.

Dendritic cells (DC) are key initiators of primary immune responses. Myeloid DC can secrete IL-12, a potent Th1-driving factor, and are often viewed as Th1-promoting APC. Here we show that neither a Th1- nor a Th2-inducing function is an intrinsic attribute of human myeloid DC, but both depend on environmental instruction. Uncommitted immature DC require exposure to IFN-gamma, at the moment of induction of their maturation or shortly thereafter, to develop the capacity to produce high levels of IL-12p70 upon subsequent contact with naive Th cells. This effect is specific for IFN-gamma and is not shared by other IL-12-inducing factors. Type 1-polarized effector DC, matured in the presence of IFN-gamma, induce Th1 responses, in contrast to type 2-polarized DC matured in the presence of PGE2 that induce Th2 responses. Type 1-polarized effector DC are resistant to further modulation, which may facilitate their potential use in immunotherapy.

Corticosteroids inhibit the production of inflammatory mediators in immature monocyte-derived DC and induce the development of tolerogenic DC3.

Corticosteroids (CS) are potent immunosuppressive agents that are known to affect T cell-mediated inflammation by the inhibition of proliferation and cytokine production, as well as the immunostimulatory function of monocytes and macrophages. Not much is known of the effect of corticosteroids on dendritic cells (DC), the professional T cell stimulatory antigen-presenting cells. We report that the endogenous CS hydrocortisone and the synthetic CS clobetasol-17-propionate strongly inhibited the production of the inflammatory mediators interleukin (IL)-12 p70, tumor necrosis factor alpha (TNF-alpha), and IL-6 by lipopolysaccharide (LPS)-stimulated monocyte-derived immature DC (iDC) in vitro. In contrast, the stimulatory capacity, antigen uptake, and the expression of costimulatory molecules were not affected. In accordance with the decreased production of IL-12 p70, CS-treated iDC induced less production of the inflammatory Th1 cytokine interferon-y and enhanced levels of the Th2 cytokines IL-10 and IL-5 in staphylococcal enterotoxin B-stimulated CD4+ Th cells. Furthermore, CS inhibited the maturation of iDC as assessed by the lack of expression of CD83 as well as by the prevention of the loss of antigen uptake capacities. These type 3 DC (DC3) matured in the presence of CS produce less IL-12 p70 and have a decreased T cell stimulatory capacity. Moreover, uncommitted T cells that encounter the CS-induced DC3 develop into Th2-biased cells, which may additionally decrease the Th1-mediated tissue damage but, on the other hand, Th2 cytokines may promote undesirable elevation of IgE and eosinophilia. These findings indicate that suppression of T cell-mediated inflammation by CS not only relies on direct effects on T cells, but also on various effects on DC, their professional antigen-presenting cells.

The late asthmatic response is associated with baseline allergen-specific proliferative responsiveness of peripheral T lymphocytes in vitro and serum interleukin-5.

BACKGROUND: Increasing insights into the mechanism underlying the allergen-induced late asthmatic response (LAR) have been gained with implication of activated eosinophils and CD4+ T lymphocytes. However, the patient characteristics that indicate the individual capacity to develop a LAR are not well-defined. METHODS: In 22 subjects with mild to moderate house dust mite-allergic asthma, we investigated the relationship between the LAR and two other models of late-phase allergic inflammation, i.e. the allergen-specific proliferative response of peripheral blood T lymphocytes in vitro and the late cutaneous response. Non-specific bronchial responsiveness (PC20histamine), lung function (FEV1), peripheral blood eosinophil count, early phase allergic skin sensitivity, and levels of total and specific immunoglobulin E (IgE) were determined prior to bronchial allergen challenge. Serum levels of interleukin-5 (IL-5) were measured before and at several time points after allergen inhalation. RESULTS: A significant correlation was found between the magnitude of the LAR and the allergen-specific proliferative response of peripheral T lymphocytes (r = 0.44, P = 0.04) but not the late cutaneous response. Stepwise-multiple linear regression of the magnitude of the LAR on the parameters analysed at baseline, resulted in a model combining PC20 histamine, early phase allergic skin sensitivity, and the allergen-specific proliferative response of peripheral T lymphocytes (R2 = 0.84, P<0.001). No contribution of the late cutaneous response to the prediction of the LAR was found. Serum levels of IL-5 increased significantly at 6 h (P = 0.01) and 24 h (P = 0.003) after bronchial allergen challenge and correlated with the allergen-specific proliferative response of peripheral T lymphocytes in vitro (rho = 0.48, P = 0.02). CONCLUSIONS: The findings in this study point to a role of TH2-lymphocyte responses in the development of the allergen-induced LAR. In allergic asthmatic patients, allergen-specific responsiveness of peripheral T-lymphocytes in vitro may serve as a model to determine the individual capacity to develop a LAR after allergen inhalation.

Glucocorticoids inhibit bioactive IL-12p70 production by in vitro-generated human dendritic cells without affecting their T cell stimulatory potential.

Glucocorticoids (GC) are known to affect the immune response at several stages. However, little is known about how GC influence the initiation of the specific immune response at the level of dendritic cells (DC), the highly professional APC for T cells. Therefore, we studied whether GC modulate the cytokine production and T cell stimulatory function of DC. In LPS-stimulated DC, GC strongly reduced the secretion of the Thl-skewing factor IL-12p70 and, to a lesser extent, the production of the proinflammatory cytokines IL-6 and TNF-alpha. Regarding the T cell stimulatory function of DC, GC did not influence the cell surface expression of HLA-DR or the costimulatory molecules CD40 and CD80 and did not influence the ability of DC to take up Ag. Consequently, GC pretreatment of DC indeed did not affect their ability to stimulate CD4+ Th cell proliferation in response to superantigen. However, as a result of their defective production of bioactive IL-12, GC-pretreated DC have a reduced ability to promote the production of IFN-gamma in CD4+ Th lymphocytes, as shown by the observation that IFN-gamma production could be restored by exogenous IL-12. In contrast, GC treatment of DC enhanced the secretion of the antiinflammatory cytokine IL-10 and the type 2 cytokine IL-5 by the T cells. It is concluded that, in addition to their role as potent inhibitors of inflammation via the direct suppression of cytokine production in T cells, GC may further inhibit T cell-mediated inflammation indirectly via the suppression of IL-12 production by DC.

Identification and partial characterization of multiple major allergens in peanut proteins.

BACKGROUND: Peanuts are a major cause of food allergies both in children as in adults which can induce an anaphylactic shock. The identification and characterization of peanut allergens could lead to more insight into the mechanism and contribute to the improvement of diagnostic tests and treatment for peanut allergy. OBJECTIVE: In the present study, the peanut protein-specific immunoglobulin concentrations as well as their recognition of the various peanut proteins or protein subunits was determined in the plasma of peanut-allergic (PA) and non-allergic (NA) individuals. Moreover, two peanut allergens were characterized in more detail to confirm them as the earlier described Ara h1 and Ara h2. METHODS: The presence of Ig-binding sites in peanut proteins was studied by immunoblotting assays whereas the concentrations of peanut-specific Ig was determined by ELISA. RESULTS: Peanut proteins were found to contain multiple binding sites for immunoglobulins. Of these proteins, six were recognized by peanut-specific IgE present in more than 50% of the plasma samples of the PA group. Their molecular weights were approximately 44, 40, 33, 21, 20 and 18 kDa. The last three protein bands were recognized by peanut-specific IgE present in more than 70% of the PA plasma samples and were thought to contain Ara h2. This allergen as well as another protein that was thought to be Ara h1, which was not recognized by the majority of the patients' IgE-containing plasma samples, were isolated and the N terminal amino acid sequence was determined. Peanut protein-specific IgA, IgM, IgG and IgG-subclasses showed a more diverse recognition pattern of peanut protein in the PA group compared to the NA group. No differences were found in the plasma concentrations of peanut protein-specific immunoglobulins of the various classes between the PA and NA group. CONCLUSIONS: From the present study, we conclude that peanuts contain multiple allergens, of which six can be described as major allergens, Ara h2 included. In our population Ara h1 is not a major allergen. The recognition of peanut proteins by immunoglobulins is more diverse in PA individuals compared with NA individuals which, however, is not substantiated in the concentrations of peanut-specific immunoglobulins in plasma, other than IgE.

Food allergen (peanut)-specific TH2 clones generated from the peripheral blood of a patient with peanut allergy.

BACKGROUND: Increasing evidence indicates a prominent role of allergen-specific TH2 cells, with high IL-4 and IL-5 production and low interferon-gamma production, in the regulation of IgE and eosinophil production in allergic disorders. However, most studies have concentrated on T cells reactive with inhalation allergens, whereas little is known about the properties of food allergen-reactive T cells. OBJECTIVES: In this study we therefore characterized peanut-specific T cells, cloned from a patient with severe peanut allergy. METHODS: Peripheral blood mononuclear cells from patients with peanut allergy and nonallergic individuals were stimulated with crude peanut extract (CPE) to compare the proliferative responses and to select a suitable patient for the cloning of CPE-specific T cells. The resultant panel of CPE-reactive T-lymphocyte clones was serologically phenotyped by flow cytometry and analyzed for cytokine secretion by ELISA. RESULTS: The patients' peripheral blood mononuclear cells showed a dose-dependent proliferation response to CPE, which was significantly higher (p < 0.05) than in peripheral blood mononuclear cells of nonallergic donors. The CPE-specific T-lymphocyte clones generated from the selected patient were all CD4+/CD8- T helper cells with a TH2 cytokine profile, secreting high amounts of IL-4 and IL-5, but little or no interferon-gamma. CONCLUSIONS: This study demonstrates that peanut-specific T cells do occur in the peripheral blood of patients with peanut allergy and suggests an increased frequency of these T cells in patients compared with nonallergic control subjects. The CD4+ phenotype and the TH2 cytokine profile of the CPE-specific T-lymphocyte clones suggest a functional role of allergen-specific TH2 cells in the pathophysiology of food allergy, similar to the function of inhalation allergen-specific TH2 cells.