The nose is the air-conditioner of the airways. Since normal breathing is through the nose, this is where most airborne particles are filtered. Therefore, the nasal mucosa is the first line of defence against particles in the air ( 1). Allergic rhinitis is characterized by an inflammation of the nasal mucosa induced by allergens. The symptomatology of allergic rhinitis is considered to be the result of the accumulation and activation of infiltrating inflammatory cells, releasing mediators and cytokines ( 1). Our understanding of the mechanisms leading to sensitization in the airway mucosa is still incomplete. The stimulation of the immune system with allergens is a necessary step in the development of atopic disease ( 2, 3). Antigen-presenting cells (APCs) play an important role in the immune response ( 4, 5), enabling T cells to recognize antigens. T-cell antigen receptors on immunocompetent T cells can recognize antigens when they are presented by APCs in the context of a MHC class molecule ( 6). MHC class I molecules are present on the surface of all cells ( 1). They present antigens, which are generated within the cells as a result of viral or bacterial infection or malignant transformation to CD8+ suppressor/cytotoxic T cells. In this way, these “infected cells” can be recognized and eliminated ( 6). The APCs are MHC class II-positive cells which can take up an antigen from the environment, process it through an endolysosomal pathway into peptide fragments ( 7, 8), and present the processed antigen in a groove of the MHC class II molecule to T cells ( 9). Typical APCs are monocytes/macrophages, dendritic cells (DC), and B cells ( 6). However, DCs are the only APCs capable of inducing a primary humoral and cellular immune response ( 10-12). It has been shown that an increasing number of other cells are able to express MHC class II after stimulation. It has also been claimed that some of these cells – such as endothelial cells ( 13, 14); (airway) epithelial cells ( 15-17); and even mast cells ( 18-20), eosinophils ( 21), and neutrophils ( 22) – play an antigen-presenting role. When stimulating naive T cells, other MHC class II positive nonprofessional APCs either stimulate poorly or produce an unusual immune reaction in vivo ( 6). By contrast, these cells are able to present antigen in vitro to activated or memory T cells in a secondary immune response. The role of these nonprofessional APCs in stimulating low-level secondary immune responses in vivo needs further clarification ( 23). DCs are classified according to function and location: blood DCs ( 24); tissue DCs, including Langerhans cells (LCs) of the skin ( 25) and the airways ( 26-28) and digestive tract mucosa ( 29-31); indeterminate cells in the dermis, lamina propria of the mucosa ( 32), and the submucosa ( 27); veiled cells in afferent lymphatics ( 33); interdigitating cells in the regional lymph nodes ( 33) and lymphoid structures around the mucosa ( 34, 35); and interstitial DCs in various organs such as the kidney, lung, and thyroid gland ( 33, 36) (for review, see Ref. 4). After antigen encounter at the periphery, the DCs migrate as veiled cells to the draining lymph nodes ( 33), where they reside in the T-cell-dependent areas to become interdigitating DCs (IDCs) ( 37). During the interaction between an APC and a T cell, both the ligation of the T-cell receptor by an antigenic peptide associated with an MHC molecule and the ligation of the CD28 molecule on the T cell by one of the costimulatory molecules on the APCs are needed ( 38, 39). These costimulatory molecules can be B7-1 (CD80) or B7-2 (CD86) interacting with CD28 ( 6, 39) and members of the tumour necrosis factor receptor family (TNF-R) such as the OX40 ligand and CD40 ( 40, 41). DCs are highly mobile cells which can move several centimetres per hour, and which can serve as an “early warning system” to alert the adaptive immune system to incoming pathogens and allergens ( 32, 42). Recent investigation indicates that DCs are of diverse origin, with at least two types of myeloid precursors and a lymphoid precursor involved in their generation ( 43, 44). Mature DC subtypes, while sharing the capacity to activate T cells, show additional functional specialization. Some DCs are equipped with additional mechanisms for regulating the type of response of the T cells they activate, while others are able to interact with B cells and modify B-cell responses ( 44). The ability to generate large numbers of DCs from haematopoietic precursors in vitro has been a major breakthrough in the understanding of the functions of DCs ( 24, 45-48). So far, few unique DC markers are available to study DCs in tissues, particularly in the mucosa ( 49, 50). These markers would provide more opportunities, not only to analyse cells in complex microenvironments but also to determine directly the interactions and mechanisms at play within the diseased tissues. An important group of DCs in the mucosa are the LCs ( 26, 32, 50, 51). The peripheral APC which is best characterized and studied is the LC. Most of the available knowledge about the general properties of LCs has been obtained from studies of the skin. After antigen binding, LCs move to the dermis and leave it as veiled cells via the lymphatics, finally presenting antigen to T cells as interdigitating cells in the paracortical area of the draining lymph node ( 52-56). LCs are characterized by the expression of CD1 ( 57, 58) and Birbeck granules ( 59). The function of the CD1 molecule has long been undetermined, but recent evidence has established that CD1 molecules comprise a novel lineage of antigen-presenting molecules, distinct MHC class I and class II molecules ( 60, 61). Unlike the MHC molecules, which bind short peptides in their antigen-binding groove for presentation to either CD4+ or CD8+ T cells bearing αβ T-cell receptors, the CD1 molecules appear to accommodate lipid and glycolipid antigens in their hydrophobic cavity for presentation to a wide variety of T cells, including double-negative αβ and γδ T cells and CD8+αβ T cells ( 62). Resting LCs and freshly isolated LCs are well equipped for antigen binding and processing, and for the stimulation of activated T cells ( 63). However, these cells are weak in stimulating resting T cells ( 38, 64). After maturation, LCs lose their typical characteristics (CD1, Birbeck granules, Fc-receptor) and change their structure, phenotype, and functional capacities into those of DCs ( 56, 64-66). In particular, they start to express molecules of the B7 family necessary for stimulating T cells ( 38). These LCs are extremely powerful in stimulating resting T cells, but they are weak in binding and processing the antigen. Therefore, resting LCs are considered to be good APCs which mature into DCs, after which they are powerful sensitizers of T cells ( 67). The maturation of DCs is primarily influenced by GM-CSF ( 48, 68-70). However, it is now apparent that local, i.e., environmental, factors can govern this process. LCs are able to internalize antigens by various mechanisms. Soluble antigens are internalized by macropinocytosis ( 45). The pinocytosis can also occur after binding of the antigen to receptors such as the mannose receptor ( 71, 72). These mechanisms of internalization are upregulated by IL-4 and IL-13 ( 73). The receptor-mediated endocytosis can also take place through the internalization of FcR-bound immune complexes ( 67). In the case of atopy, LCs capture and internalize allergens by surface-bound FcεRI (see below, section on the secondary immune response) ( 74-77). Pathogenic and nonpathogenic antigens continuously bombard the epithelium of the airways. These antigens are mainly removed nonimmunologically by the first defence layer of the mucosa consisting of mucus, ciliated epithelial cells, and glycoproteins/lysosomes. If the antigen passes this defence layer, it encounters specific and nonspecific immunologic defence mechanisms. The nonspecific defence consists, for example, of phagocytizing cells such as neutrophils and macrophages and the complement activation ( 78, 79). The specific defence mechanism (resulting in a specific immunologic reaction in relation to a certain antigen) is formed by the antibodies, mainly secretory IgA and to a lesser extent IgG ( 80), and immunocompetent cells in the nasal mucosa ( 1). Activation of the specific defence mechanisms may lead to inflammation, which can be allergic. The common induction of tolerance, despite the bombardment of antigens, suggests a strictly regulated T-cell system ( 81). Active immunosuppression seems to play an important role in differentiation between trivial and potential pathogenic antigens, and results in the normal situation of tolerance of trivial antigens ( 82, 83). Of particular importance to allergic disease is the recognition of the regulation of helper immune function by two lineages of T helper cells, i.e., Th1 and Th2, by these cytokines ( 84, 85). The Th2 hypothesis of allergy considers atopy to be a Th2-driven hypersensitivity reaction to allergens of complex genetic and environmental origins, in which the Th1 lineage, normally driven by IL-2, TNF, and IFN-γ, is deficient, and in which a predominant Th2 response is seen that is driven by IL-4, IL-13, IL-5, and IL-10. The Th2 response recruits, mobilizes, and activates eosinophils for subsequent mucosal tissue damage, and it also results in IL-4, an essential cofactor for local or generalized IgE production. Thus, although eosinophils are largely responsible for airway symptoms, their function appears to be under T-cell control. Animal trials for the lung in the last decade have shown that the balance between dendritic APCs and macrophages and/or the reaction of the T-cell system to the stimuli given by these cells play an important role in the occurrence of Th1 tolerance or Th2 hypersensitivity ( 86). In normal circumstances, repeated exposure to nonpathogenic antigens induces T-cell-mediated immunologic tolerance ( 87, 88). In the lung, alveolar macrophages play a role in the induction of immunologic tolerance ( 89, 90). The elimination of alveolar macrophages in sensitized rats induces an increased function of antigen-presenting DCs and an increased IgE response to inhalant allergens ( 91, 92). Intratracheally introduced DCs can be followed into the regional lymph nodes, where they are able to induce a primary antigen-specific T-cell response ( 35, 93, 94). There is recent experimental evidence that antigen presentation by airway DCs leads to the preferential development of a Th2 response which can be short-lived (a short boost of IgE production, followed by active suppression), or which can develop into a polarized, long-lived Th2 response associated with atopy ( 95). Factors which could contribute to tolerance instead of allergic reaction, such as the dose of antigen, the type of APC, the presence of cytokines, other microenvironmental costimuli, and concomitant viral infection, are currently under investigation. Antigen dose, and therefore the extent of TCR ligation, has also been shown to direct the development of a Th1 or Th2 phenotype from naive CD4+ T cells ( 6, 96, 97). The role of costimulatory molecules such as the B7 molecule family and CD30 and the CD30 ligand in the determination of the preferential development of Th2 and the relationship to cytokine environment is a fast-growing field of knowledge ( 39, 98, 99). Moreover, monocyte-derived DCs from allergic asthmatic patients have already shown phenotypic differences in the expression of HLA-DR, CD11b, and the high-affinity receptor for IgE from that of healthy controls. The involvement of DCs in the pathogenesis of allergic disease could depend on an increased immunostimulatory capacity of DCs ( 100). One of the factors increasing the immunostimulatory capacity of DCs could be the ability produce IL-12 ( 101). Kinetic studies in rats have shown that T cells are primarily activated in the superficial cervical lymph nodes; this indicates that induction of tolerance primarily takes place in the upper airways ( 102). Not only are airway DCs essential for stimulating naive T cells in a primary immune response to inhaled allergen and for the development of allergic sensitization, but they have also been shown to be essential for presenting inhaled allergen to previously primed Th2 cells, for stimulating memory T cells in a secondary response to inhaled allergen, and for the subsequent development of chronic airway inflammation leading to chronic eosinophilic airway inflammation ( 103). The depletion of DCs in animal airways led to an almost complete suppression of eosinophilic airway inflammation ( 23). LCs and DCs have been shown to be positive for IgE ( 25, 32, 104, 105). While earlier observations suggested that this phenomenon results from the binding of (complexed) IgE to the low-affinity IgE receptor (FcεRII/CD23), recent evidence suggests that LCs and DCs in the skin, in the blood, and in the mucosa can bind monomeric IgE via the high-affinity receptor for IgE (FcεRI) ( 100, 106-109). The FcεRI on APCs targets, as an allergen-focusing molecule, the APCs, resulting in more efficient uptake, processing, and presentation to T cells than allergen-binding to APCs in the conventional manner ( 75, 76, 110, 111). In the subepithelial layer of the nasal mucosa, clusters of (IgE+) LCs and activated (HLA-DR+) T cells are regularly found, suggesting a role for LCs in the secondary response to inhaled allergen and the subsequent development of chronic airway inflammation leading to chronic eosinophilic airway inflammation ( 50). In vivo, FcεRI-IgE-dependent allergen presentation may critically lower the threshold of atopic individuals to mount allergen-specific T-cell responses. This would result in the perpetuation of allergen-specific IgE production (type I reactions) and perhaps even in the occurrence of T- cell-mediated delayed-type hypersensitivity reactions in allergen-exposed tissues ( 77, 112, 113). DCs, macrophages, and B cells are present in the epithelium and lamina propria of the nasal mucosa ( 50, 114, 115). The APC most extensively studied is the DC. Different members of the DC population are observed in the nasal mucosa ( 50). DCs form a network of APCs in the animal ( 116) and human respiratory mucosa. The density of DCs is highest in the epithelial surface of the upper airways and decreases along the small airways of the peripheral lung ( 117). The DC most extensively studied in the nasal mucosa is the LC. These dendritic, MHC class II+, CD1a+ cells are found in the epithelium and the lamina propria ( 26, 51). LCs are characterized by the presence of Birbeck granules in the cytoplasm. Electron microscope studies of the nasal mucosa of allergic patients showed many DCs with electron-lucent cytoplasm and Birbeck granules, with extensions between the adjoining epithelial cells in the basal half of the epithelium. Occasionally, a DC was found in the basal membrane, with extensions in the lamina propria and in the epithelium. The total number of HLA-DR+ DCs in the epithelium of the nasal mucosa is 2–3 times higher than the number of CD1+ LCs ( 50). These cells can be activated macrophages, L25+ DCs, LC precursors, or possibly also LCs which have matured and therefore lost the CD1 marker ( 50). Although CD1+ cells are also found in the lamina propria of the nasal mucosa, most HLA-DR+ DCs are not CD1+ cells. Most CD1+ cells are found in the subepithelial layer. This suggests that many of the other HLA-DR+ DCs are matured LCs ( 50). Macrophages in the airways are capable of effectively eliminating invading antigens/allergens by phagocytosis, but they cannot selectively present allergens to the immune system ( 118). Macrophages increase easily after nonspecific stimulation of the nasal mucosa, as in daily lavage or brushing ( 50, 115). Care should be taken not to attribute this increase, which also occurs in nonallergic controls, to a specific increase due to allergen stimulation. Animal experiments suggest that the balance between macrophages and DCs plays a crucial role in the occurrence of tolerance or hypersensitivity ( 119, 120). Another possible function of monocytes/macrophages, which have recently been found to express FcεRI also, may be to remove IgE from the circulation/tissues and redirect the immune response from naive B cells ( 76). The lineage relationship between DCs and monocytes/macrophages remains to be clarified. The microenvironment could be decisive in terms of which phenotype evolves from the CD34+ precursor ( 121-124). It has been established that IL-4 and CD40 affect differentiation into DCs ( 121). B cells can be found in the epithelium and the lamina propria of the nasal mucosa ( 114). Unlike DCs and macrophages, their numbers do not increase during allergic inflammation ( 125, 126). The role of B cells as APCs in the nasal mucosa is virtually unknown. In blood/other tissues, B cells have been extensively studied. B cells possess unique characteristics that make them efficient APCs. They express HLA-DR and antigen-specific immunoglobulins, allowing them to bind foreign proteins selectively from a large pool of serum proteins. B cells also express costimulatory molecules such as CD80 and CD86. From birth, the nasal mucosa is continuously exposed to potential allergens. In a percentage of “genetically predisposed” children, this contact leads to sensitization of the immune system. After this process, allergen-specific IgE is produced. This will develop into the typical symptoms associated with allergic rhinitis (e.g., rhinorrhoea, nasal congestion, itching, and sneezing). Animal experimental work has shown that repeated intranasal inoculation of ovalbumin leads to tolerance. This contrasts with intratracheal instillation. Moreover, DCs have been shown to be virtually absent neonatally, but their numbers increase rapidly in the following weeks in the rat airway epithelium ( 127). In young children also, LCs are found in the epithelium and lamina propria of the nasal mucosa ( 128). They are mainly seen in close proximity to the basal membrane and in the subepithelial space of the lamina propria. A typical dendritic appearance was seen in a majority of the cells. By contrast with findings in adults, no differences from controls could be detected in the number of LCs in the epithelium and lamina propria of the nasal mucosa of allergic children during the sensitization phase ( 128). However, only the children who already showed signs of sensitization to aeroallergens had IgE on LCs. This IgE was bound on the FcεRI ( 128). In animal experiments, it has been shown that the number of DCs increases during inoculation with viruses, bacteria, and soluble proteins ( 42). This might explain the absence of an increase in LCs in the nasal mucosa of these young children exceeding the increase caused by the allergic rhinitis. However, in man, an increase in DCs after viral or bacterial infection has not yet been proven. Research on adult nasal mucosa showed the antigen-presentation process to be located in the lamina propria, suggesting a local immune response to allergen provocation ( 129). LCs and activated T cells are able to form clusters in the subepithelial layer of the nasal mucosa ( 129). In the case of APC/T-cell interactions in other epithelial structures such as skin and intestinal mucosa, it is more likely that the majority of these interactions take place in regional draining lymphoid structures such as lymph nodes ( 130, 131). The regional draining lymphoid structures of the nasal mucosa are the cervical lymph nodes and Waldeyer's ring, which consists of the palatine and lingual tonsils and the adenoid. The NALT (nasal associated lymphoid tissue) is located in the upper airways of rodents much like the adenoid tissue in man. The adenoid and NALT have functional similarities. Experiments in mice have shown that the NALT consists of immunocompetent cells involved in the defence of the upper airways and is the most important mucosal lymphoid tissue of the airways ( 102). Its anatomical position and cell populations make Waldeyer's ring a prime site for antigen presentation ( 132). Moreover, animal experimental studies have indicated an important role for the NALT, the murine equivalent of Waldeyer's ring, in the process of allergic sensitization ( 102). After antigen installation in the nose of mice, the stained antigen can be detected in the NALT and in the cervical lymph nodes. Uptake in the NALT was faster and continued for a longer period than in the cervical lymph nodes ( 133). Moreover, nasal sensitization has been shown to be much more effective than oral sensitization ( 134, 135). LCs and other HLA-DR+ DCs can be found in the human adenoid ( 136-139). Raised levels of LCs have been found in the epithelium and the interfollicular areas of adenoids of young allergic children compared to nonallergic controls of the same age group ( 34). LCs can be found in the epithelium and lamina propria of the nasal mucosa ( 51, 140, 141). The number of LCs in the epithelium of the nasal mucosa of allergic rhinitis patients is significantly higher than in nonallergic patients with polyposis nasi or normal controls ( 51). In addition, the numbers of LCs and HLA-DR+ cells increase significantly in the epithelium and the lamina propria during symptomatic allergic rhinitis ( 50, 104). The same has recently been found in the bronchial mucosa ( 28, 109) and in bronchial alveolar lavage ( 90). In grass-pollen-allergic patients, LC levels are significantly higher during the grass-pollen season than in the asymptomatic period before and after the season ( 104). The number of LCs increases sevenfold in the epithelium and lamina propria after repeated threshold allergen provocation ( 50). The increase after the first allergen provocation is fast (a threefold increase in the first 30 min after allergen provocation) ( 50). Most of this increase seems to occur by redistribution from the lamina propria ( 127). The LCs in the lamina propria are found mainly in the subepithelial layer. In this layer, clusters of LCs and activated (HLA-DR+) T cells are regularly found ( 50). The number of LCs in the lamina propria increases at a much slower rate than in the epithelium. In the lamina propria, a significant increase was found after 24 h ( 50). After a period of allergen provocation, the LC level remained increased in the total nasal mucosa for a long period. Two weeks after threshold allergen provocation, the number of LCs was still very much higher than before allergen provocation ( 50). The entry of DCs into the mucosa (of tissues) is regulated through the expression of adhesion molecules, which in turn are upregulated by cytokines ( 42). There seems to be a rapid turnover rate of resident DCs in the airway mucosa ( 127). It is not known whether DC precursors can divide after arrival in the airway mucosa. Therefore, it is also unclear whether the long-term increase in LCs after allergen provocation is caused by the division of precursor cells or the entry of new DCs because of the upregulated adhesion molecules and cytokine expression. LCs are able to bind allergens to IgE by the high-affinity receptor for IgE ( 107, 108, 125, 126). This IgE has been shown to be specific ( 108). Double-staining immunomethods with CD1 and anti-IgE-FITC on sections of the nasal mucosa of patients with allergic rhinitis revealed IgE on 20–40% of the CD1+ cells in the epithelium and occasionally on CD1+ cells in the lamina propria ( 104). Although the absolute number of IgE+/CD1+ cells increases during the grass-pollen season, no difference was observed in the percentage of IgE+/CD1+ cells during, before, or after the grass-pollen season. IgE+/CD1+ cells are not present in the nasal mucosa of nonallergic polyposis patients and nonallergic controls. Anti-CD23 antibodies directed against the FcεR on B cells did not bind to LCs in the nasal mucosa ( 32). In the epithelium, numerous CD1+ cells were in close contact with IgE+ cells, presumably mast cells. Indeed, electron microscopy showed LCs close to mast cells ( 26). Even mast-cell granules and eosinophilic granules were observed between the extensions of the LCs. As has been described before, airway DCs are essential for stimulating naive T cells in a primary immune response to inhaled allergen and for the development of allergic sensitization. However, airway DCs have also been shown to be essential for presenting inhaled allergen to previously primed Th2 cells, for stimulating memory T cells in a secondary response to inhaled allergen, and for the subsequent development of chronic airway inflammation leading to chronic eosinophilic airway inflammation ( 103). Depletion of DCs in animal airways led to an almost complete suppression of eosinophilic airway inflammation ( 23). Studies performed in perennial allergic rhinitis and in provocation studies showed that even low dosages of local steroids result in the total disappearance of CD1+ cells during disease and a virtually complete inhibition of the influx of LCs during allergen provocation in the epithelium ( 142-144). Furthermore, a significant reduction in the number of CD1+ cells and HLA-DR+ cells during disease and a significant inhibition of the influx is seen in the lamina propria of the nasal mucosa epithelium ( 142-144). Apart from diminishing the number of DCs in the airways, corticosteroids also inhibit the uptake and/or processing, but not the presentation, of antigen by airway DCs ( 145). Studies in allergic rhinitis have indicated substantial variations in cell sensitivity to corticosteroid therapy ( 144). Some cells, such as APCs (LCs) and eosinophils, are highly sensitive, while others, such as T cells, are only significantly reduced in exaggerated situations. Some cells, such as macrophages and neutrophils, are not influenced at all. No differences were found in any of these studies in the number of monocytes/macrophages. This is in accordance with studies in the lung ( 28, 146). Furthermore, monocyte-derived DCs from patients with mild/moderate atopic asthma using corticosteroid inhalation exhibited decreased accessory potency. These functional differences could not be explained by the changes in the expression of MHC II and/or the costimulatory molecules CD40 and CD86 ( 147). In conclusion, we can say not only that the accumulation of dendritic APCs, particularly LCs, during allergen provocation is all but abolished by nasal steroids, but also that, in perennial disease, even low doses of nasal steroids result in a considerable reduction of these APCs. Abnormal APC activity seems to play an important role in the propensity to develop allergic reactions. APC function could theoretically be influenced in a number of ways. Epidemiologic studies have indicated that an important role is played by infections, particularly by bacteria, in shifting a deteriorating Th2 response back to a Th1 response ( 148). Moreover, the function of APCs may be influenced in a more experimental way. It has been shown that the B7 family, which interacts with CD28/CTLA4, plays an important role, and could be involved in maintaining a balance between Th1 and Th2 ( 149, 150). Blocking of B7-2 by anti-B7-2 Ab results in an inhibition of the generation of Th2 cells and enhancement of the generation ofTh1 cells ( 151, 152). However, these ways of influencing the immune system are studied not only to change allergic reactions but also, conversely, in autoimmune disease, where the prevention of autoimmune disease is studied by the induction of self-antigen-specific Th2 cells, which can prevent autoimmune disease ( 150, 153). Whether it is possible to find the right balance in immune tolerance is something which has to be elucidated. The strong reduction in the number of APCs by local corticosteroid therapy could explain the subsequent reduction of the secondary inflammatory response and symptomatology in allergic disease ( 154). The animal experimental work showing that depletion of DCs in the airways led to an almost complete suppression of eosinophilic airway inflammation indicates that DCs could be a very good target in a more precise treatment of allergic inflammation ( 23). Since most of the allergen is deposited in the nose and reduction of allergic inflammation in the nose leads to the reduction of asthma symptoms, the reduction of APCs in the upper airways might even be important for the reduction of lower airway symptomatology. Treatment with anti-IgE has been shown to be effective in the early and late phases in asthma ( 155). It may very well be that the anti-IgE affects not only the mast cells in the early phase but also the DCs enhancing the late phase. Hyposensitization has been shown to shift the balance of T-cell subsets away from the TH2 type (producing particularly IL-4 and IL-5) in favour of a TH1-type T-cell response (with the preferential production of IFN-γ) ( 156). The mechanisms of this shift are not known, but they could very well be sought in the APC function. Different adjuvants can be used to potentiate the effects of allergen in inducing immune deviation. These include IL-12, immunostimulatory sequences of DNA, and bacterial proteins such as those used in HibVax. ( 156-158). APCs, particularly DCs, seem to play an important role in allergic rhinitis. The nasal mucosa is an important first site of sensitization. Airway DCs lead to the preferential development of a Th2 response. Which mechanisms lead to a persistent deviation of the normal Th1 response to develop a polarized, long-term Th2 response associated with atopy is not known. DCs seem to play a role in the sensitization phase but also in the induction of ongoing allergic reaction. In the sensitization phase, DC might be influenced to induce a Th1 response. The exact consequences of these manipulations are under investigation. Even now, and probably in ways which will be even more elaborate in the near future, the function of DCs in the secondary immune response leading to allergic (eosinophilic) inflammation and symptomatology of allergic disease can be therapeutically influenced. Local corticosteroid treatment is currently the most effective way to reduce DC numbers and function. However, in the near future, DCs could be a very good target in a more precise treatment of allergic inflammation. I thank Bert Lambrecht and Alex Kleinjan for valuable comments and review of the paper, Petra Boon for editorial assistance, and Pete Thomas for correcting the English.