Distribution of Langerhans cells and mast cells within the human oral mucosa: new application sites of allergens in sublingual immunotherapy?

Abstract

Background:  Sublingual immunotherapy (SLIT) represents an alternative to subcutaneous immunotherapy. While antigen-presenting cells such as Langerhans cells (LCs) are thought to contribute to the effectiveness of SLIT, mast cells (MCs) most likely account for adverse reactions such as sublingual edema. As little is known about LCs and MCs within the oral cavity, we investigated their distribution in search for mucosal sites with highest LCs and lowest MCs density.

Methods:  Biopsies were taken simultaneously from human vestibulum, bucca, palatum, lingua, sublingua, gingiva, and skin. Immunohistochemistry and flow cytometry were used to detect MCs, LCs and high affinity receptor for IgE (FcεRI) expression of LCs. Mixed lymphocyte reactions were performed to assess their stimulatory capacity.

Results:  Highest density of MCs was detected within the gingiva, while the lowest density of MCs was found within the palatum and lingua. However, sublingual MCs were located within glands, which might explain swelling of sublingual caruncle in some SLIT patients. Highest density of LCs was detected within the vestibular region with lowest density in sublingual region. Highest expression of FcεRI was detected on LCs within the vestibulum. Furthermore LCs from different regions displayed similar stimulatory capacity towards allogeneic T cells.

Conclusions:  In view of our data, different mucosal regions such as the vestibulum might represent alternative SLIT application sites with potent allergen uptake. Our data might serve as a basis for new application strategies for SLIT to enhance efficiency and reduce local adverse reactions.

Abbreviations:

7AAD

7-amino-actinomycin-D

APC

antigen-presenting cells

cpm

counts per minute

FcεRI

high affinity receptor for IgE

FITC

fluorescein-isothiocyanate

HPF

high-power field

LC

Langerhans cells

mAb

monoclonal antibody

MC

mast cell

MHC

major histocompatibility complex

oLC

oral Langerhans cells

oMC

oral mast cells

oMC_t

tryptase-positive/chymase-negative oMC

oMC_tc

tryptase-positive/chymase-positive oMC

PE

phycoerythrin

rFI

relative fluorescence index

SCIT

subcutaneous allergen-specific immunotherapy

SLIT

sublingual allergen-specific immunotherapy

Within recent years, sublingual immunotherapy (SLIT) has been proven to represent a safe and effective alternative to subcutaneous immunotherapy (SCIT) in the treatment of allergic rhinitis (1). While severe systemic adverse reactions such as anaphylaxis have not been observed so far, the most frequently reported side-effects consist of local reactions within the oral mucosa especially in the sublingual region and gastrointestinal tract (2). So far the underlying immunological mechanisms of SLIT are not quite clear. However, while antigen-presenting cells (APC) such as oral Langerhans cells (oLC) are thought to play a major role in the effectiveness of SLIT (3), oral mast cells (oMC) most likely account for adverse reactions such as oral itching and sublingual edema caused by histamine release (4). Langerhans cells (LCs) are outposts of the immune system and are located within the epithelium suprabasal layer of all body surface organs, where they capture and process antigens (5). As part of adaptive immunity, LCs are capable of stimulating immune responses by activating antigen-specific T cells after migration to the regional lymph node (6, 7). However, LCs are not only prone to initiate antigen-specific T-cell immune responses but are also capable of stimulating protective immune reactions, which are characterized by tolerance induction towards allergens (5). We could show recently, that oLCs phenotypically differ from their skin counterpart especially by the expression of lipopolysaccharide receptor/CD14 and the high affinity receptor for IgE (FcεRI) of which the latter might enable oLCs to display more efficient allergen-binding and uptake during SLIT (8, 9). Considering oMC, it has been shown so far that tryptase-positive/chymase-negative oMC (oMC_t) as well as tryptase-positive/chymase-positive oMC (oMC_tc) subpopulations reside within oral mucosal tissue of which the latter predominate and contain histamine (10). As little is known about oLCs and oMCs within the oral cavity, we investigated their distribution in search for alternative mucosal application areas with the highest density of IgE receptor-bearing oLCs for allergen uptake and lowest presence of oMC_tc and likeliness of adverse reactions.

Material and methods

Oral mucosa specimen

Different punch-biopsies (Ø 8 mm) were taken simultaneously from the mucosal tissues of the vestibular, buccal, palatine, lingual, sublingual and gingival regions as well as epidermal skin during human autopsies at the Department of Pathology, University of Bonn (n = 10). Individual’s age ranged from 45 to 65 years. Autopsies of individuals with malignancies of orofacial area and undergoing chemotherapy before decease were excluded. Subjects previously taking corticosteroids and/or immunosuppressants systemically as well as locally were also excluded from the study. All specimens were obtained with the approval of the local ethics committee.

Immunhistochemistry

Serial paraffin wax sections (4 μm) of oral mucosal tissue from the vestibular, buccal, palatine, lingual, sublingual, and gingival regions were stained by monoclonal antibodies (mAb) specific for CD1a (Clone O10; Beckman-Coulter, Krefeld, Germany) to detect epidermal APC and chymase (Clone CC1; AbD Serotec, Dusseldorf, Germany) to detect mast cells (MCs) for 1 h at room temperature after heat-mediated antigen retrieval. LSAB2 kit (DAKO, Hamburg, Germany) was used for secondary labeling with fast red chromogen and Invision kit (DAKO) for secondary labeling with 3,3′-diaminobenzidine (DAB) chromogen. Slides were analyzed by three independent investigators as described in detail elsewhere (11). Briefly, CD1a-positive and chymase-positive cells were counted as cells per 10× magnification in three different high-power fields (HPF) per slide and averaged afterwards. Then results of the three independent investigators were combined and the average score was calculated.

Flow cytometry

For flow cytometry, crude epithelial cell suspension of oral mucosal tissue was prepared by trypsinization in a 0.5% trypsin buffer without Ca²⁺ for 1 h at 37°C as described before (12). Cell number varied between 1 and 4 × 10⁶ cells. CD1a was detected by mAb phycoerythrin (PE)-labeled T6RD1 (IgG1, Beckman-Coulter, Krefeld, Germany) and PE-carbocyanin (Cy) 5-labeled HI149 (IgG1, BD Biosciences, Heidelberg, Germany). allophycocyanin (APC) mAb (IgG1) HB15e detecting CD83 and PE-Cy7 mAb (IgG2a) L243 directed against HLA-DR were purchased from BD Biosciences. PE-labeled mAb DCGM4 (IgG1, Beckman-Coulter) detected LC specific CD207/Langerin. The FcεRI was detected by unlabeled mAb 22E7 (IgG1, generous gift of Dr J. Kochan, Hoffmann La Roche Co., Nutley, NJ, USA) directed against the α-chain of FcεRI but does not interfere with the IgE binding site. MOPC-21 (IgG1, Sigma, Deisenhofen, Germany), IgG2a-PE-Cy7 (BD Biosciences) and IgG1RD1 (Beckman-Coulter) were used as appropriate isotype controls. Fluorescein-isothiocyanate (FITC)-conjugated goat anti-mouse (GaM/FITC) antibody was from Jackson Laboratories (West Grove, PA, USA). Normal mouse serum for blocking purposes was obtained from Dianova (Hamburg, Germany) and 7-amino-actinomycin-D (7AAD) was from Sigma.

An indirect extracellular staining for a number of 0.5–2 × 10⁵ unfixed cells was performed as described elsewhere (13). Finally, the cells were washed twice and analyzed by flow cytometry. Cells were acquired on a FACS-Canto (BD Biosciences, Heidelberg, Germany) as described in detail elsewhere (14) and analyzed by FACSDiva (BD Biosciences), and flowjo (TreeStar Inc., Ashland, OR, U.S.A.) software. For quantitative evaluation, dead cells were excluded by 7AAD staining, CD1a population was gated out manually and either expressed in percentage of positive cells or by the relative fluorescence index (rFI) determined as follows: rFI = [MFI (receptor) − MFI (isotype control)]/MFI (isotype control).

T-cell proliferation assays

Proliferation assays were performed in a total volume of 200 μl in 96-well round bottom plastic culture plates using allogeneic T cells as responder cells. Allogeneic naïve CD4⁺ T cells were isolated from peripheral blood mononuclear cells from healthy donors. Blood was obtained after informed consent from donors according to the approval of the local ethics committee. Briefly, naïve CD4⁺ T cells were enriched using a two-step approach with the help of magnetic microbead-labeled antibodies (Miltenyi Biotec, Bergisch Gladbach, Germany) and the AUTOMACs techniques (Miltenyi) as described in the manufacturer′s instructions. Oral mucosal single cell suspensions from different regions were obtained as described above. Numbers of viable oLC were calculated by CD1a⁺/7AAD⁻ staining. Triplicates of oLC containing 200 viable CD1a⁺ oLC/well were incubated with 100.000 viable allogeneic naïve T cells at 37°C for 3 days. Proliferative response was then measured by addition of 1 μCi ³H-Thymidine incorporation for 12 h. The incorporated radioactivity was measured in counts per minute (cpm) with a Wallac Microbeta Jet 1450 Microplate Scintillation/Luminescence Counter (Long Island Scientific, East Setauket, NY, USA). Results were depicted in cpm.

Statistical analysis

For statistical evaluation of significances, the Wilcoxon (signed-rank) test was performed. The test was performed using the spss 14.0 for Windows software. Results are shown as arithmetic mean ± SD; * = P < 0.05; no indication = not significant unless otherwise indicated.

Results

Mast cells are distributed within all mucosal sites with highest density in the gingiva

Tryptase-positive/chymase-positive oMC are granule-containing, FcεRI-expressing, IgE-bearing immune cells located in all connective tissue and mucosal sites (10). Through allergen binding and crosslinking of FcεRI-bound allergen-specific IgE, MCs release histamine and thereby contribute to allergic reactions (15). As oMC_tc, predominate within the oral mucosa, we investigated their distribution immunohistochemically by chymase-staining. Furthermore, the significantly highest density of oMC_tc could be detected within the gingiva, while the lowest density of oMC_tc was found within the palatum and lingua (Fig. 1). However, in the sublingual region, oMC_tc were located within the lobe and duct of sublingual glands (Fig. 2) in a substantial number of individuals. In line with earlier studies (10), oMC_tc were preferentially located close to the basement membrane in the dermis (Fig. 2).

Figure 1.  Oral mucosal tissues from different anatomical sites (representative slides in 10× magnification, region depicted in right upper corner of each slide) were stained with anti-chymase antibody to detect oMC_tc. OMC_tc were counted as described in Material and methods as positive cells per high power field (HPF) (y-axis), mean values ± SEM of n = 10 independent experiments are given in the bar graphs. Significant higher numbers of oMC_tc were detected within the gingival region followed by vestibulum, bucca, and sublingual region.

Figure 2.  Immunohistochemistry slide shows 10× magnification of sublingual tissue. OMC_tc were stained by anti-chymase antibody. OMC_tc were preferentially located along basal cells within upper dermis (arrows) and sublingual glands and ducts (asterix). Right lower slide shows 40× magnification of sublingual gland and duct.

From these data, we conclude that comparable amounts of oMC_tc are distributed with the highest density in the gingiva within the investigated oral mucosal regions and that the sublingual localization of oMC_tc within the lobe and duct of sublingual glands might explain swelling of sublingual caruncle in some SLIT patients.

Oral Langerhans cells are detected in high numbers in the vestibulum

Antigen-presenting cells, the oLCs are thought to play a major role in the immunological mechanisms involved in SLIT (3). Among the various subsets of dermal and epidermal APC-bearing CD1a⁺ and major histocompatibility complex (MHC) II molecules, the LCs uniquely express CD207/langerin (16). By flow cytometry we could identify oral APC within the investigated oral mucosal tissue as LCs detected by their CD1a, CD207/langerin and MHC II expression. As all oLCs from the investigated mucosal sites expressed comparable amounts of LCs maturation marker CD83, we could conclude that isolated oLC resided in a very similar maturation state in the uninflamed oral mucosa (Fig. 3). As little is known about the distribution of oLC within the oral mucosa, we investigated the oLCs located at the mucosal sites suitable for allergen application such as vestibulum, bucca, hard palatum, lingua, sublingua, and gingiva. Thereby, we could detect significantly the highest distribution of oLCs within the vestibulum, bucca, hard palatum, and lingua (Fig. 4). Interestingly, and in line with previously published data (17), the lowest number of oLCs was found within the sublingual and gingival regions of oral mucosa tissue (Fig. 4). Moreover, the vestibular, buccal, palatine, and lingual regions displayed higher numbers of oLCs in relation to oMCs (Fig. 5). Furthermore, we could detect an accumulation of oLCs paired with an extravasation of T cells along the rete processes. oral Langerhans cells and T cells were co-localized as demonstrated by double-colour staining (Fig. 6).

Figure 3.  Cell suspension was obtained by oral mucosal tissue trypsinization. OLC were detected by CD1a expression. Dead cells were gated out manually by 7AAD staining. Histograms show one representative sample of CD207/langerin (CD207), MHCII (HLA-DR), and CD83 expression. Oral APC were identified as Langerhans cells defined by CD1a, CD207/langerin expression (first column), MHC II expression (second column). After isolation procedures, oLC resided in a comparable maturation state demonstrated by equal CD83 expression (third column). Numbers in the upper right corner of each histogram represent rFI ± SD for CD207,CD83, and MHC II of n = 3 experiments.

Figure 4.  Oral mucosal tissues from different anatomical sites (representative slide shown in 10× magnification, region depicted in right upper corner of each slide). were stained with anti-CD1a antibody for oLC detection. OLC were counted as described in Material and methods as positive cells per high power field (HPF) (y-axis), mean values ± SEM of n = 10 independent experiments are given in the bar graphs. Significantly higher numbers of oLC could be detected within the vestibulum, bucca, palatum, and lingua compared to sublingual and gingival region.

Figure 5.  Ratio of oLC to oMC_tc (box above) was calculated by dividing average number ±SEM of oLC and oMC_tc (n = 10) (numbers depicted on y-axis) of different regions (x-axis). Within the vestibulum, bucca, palatum, and lingua numbers of oLC exceeded oMC_tc numbers.

Figure 6.  Oral mucosal tissue from lingual region was stained for CD1a and CD3. Accumulation of CD1a⁺ oLC along the rete ridges corresponded with CD3⁺ T cell detection within rete processes. Double staining of CD1a (fast red) and CD3 (DAB) confirmed co-localization of oLC and T cells (digital magnification; lower right corner).

In view of this data, the vestibular, buccal, hard palatine, and lingual mucosa appears to be an attractive alternative application site for allergens during SLIT.

Oral Langerhans cells of the vestibulum displayed highest expression of FcεRI

Through allergen capture via FcεRI-bound IgE, the oLCs are equipped for efficient allergen uptake (3, 8). Thus, we investigated intra-individual differences in FcεRI expression on the oLCs from the vestibular, buccal, palatine, sublingual, lingual and gingival regions by flow cytometry. Thereby, we could detect significantly highest FcεRI expression on oLCs from the vestibulum and buccal (Fig. 7). High affinity receptor for IgE expression was comparable on oLCs from the palatine, sublingual, lingual, and gingival regions without any significant difference.

Figure 7.  Cell suspension was obtained by oral mucosal tissue trypsinization. OLC were detected by CD1a expression. Dead cells were gated out manually by 7AAD staining. FcεRI is depicted on y-axis as percent of positive CD1a cells ± SD (n = 6). Histograms show one representative sample of FcεRI (numbers in upper right of each histogram represent % ± SD of (n = 6). The highest FcεRI expression could be detected on oLC from the vestibulum.

Oral Langerhans cells from different regions display comparable stimulatory capacity

As APC-oLCs are able to initiate a primary immune response by stimulating naïve T cells (18), allogeneic naïve T cells were cultured with oLCs from different mucosal sites to investigate their stimulatory capacity. As reported earlier (9), the oLCs displayed a higher stimulatory capacity towards T cells compared with epidermal LCs. Among the oLCs from different regions of oral mucosa, no significant variations could be demonstrated in their capacity to stimulate allogeneic T cells (Table 1).

Table 1.   Cell suspension was obtained by oral mucosal tissue trypsinization

Region	cpm	±SD
oLC were cocultured with allogeneic naïve T cells (n = 6). OLC from different oral mucosal regions displayed comparable stimulatory capacity towards allogeneic T cells. Nevetheless, a significantly higher stimulatory capacity was detected from oral mucosal oLC compared to epidermal LC. SD, standard deviation; cpm, counts per minute. *P < 0.05 as compared to skin LC.
T cells	1282*	512
Vestibulum	11869*	8015
Bucca	9671*	4714
Palatum	11966*	8047
Lingua	12108*	6826
Sublingua	6080*	2405
Gingiva	4858*	1489
Skin	1669	657

Discussion

Within the recent decade, the central question about SLIT focused on its clinical effectiveness. Ever since the meta-analysis by Wilson et al. (1), this question has been partially answered and so other questions came up such as which immunological mechanisms determine whether SLIT is successful or not. It is very likely that APC such as LCs play a crucial role in inducing desired allergen-specific tolerance and protective shift from a Th2 to a modified Th2 immune response, while MCs account for the observed side-effects, such as oral itching or sublingual edema (3, 4). As the oral cavity harbors distinctive types of mucosal tissue, such as masticatory, lining, and lingual mucosa (19) another fundamental but not less important question is – how are immune cells important for SLIT distributed within the oral mucosal tissue. Further on mucosal sites appear to be most promising for allergen application, in respect of cells relevant for allergen uptake such as FcεRI⁺ oLCs on the one hand and cells responsible for adverse reactions such as oMC_tc on the other hand. In this report, we could demonstrate that (i) oMC_tc show similar distribution within the oral mucosal tissue with significantly highest presence within the gingiva; (ii) oLCs are more frequently located within the vestibulum, bucca, palatum, and lingua; and (iii) show the strongest expression of FcεRI in the vestibulum. Furthermore, we could show that oLCs from all investigated regions display comparable stimulatory capacity towards allogeneic T cells. Nevertheless, as oral mucosal cell suspension was used, cells other than oLCs may also in part account for the stimulatory capacity. However, as described previously the oLCs displayed a higher stimulatory capacity than epidermal LCs, which might not only result from their different expression of co-stimulatory molecules and MHC classes I and II expression but also from their higher exposure to antigen as well as allergen. Although SLIT has been shown to be a safe alternative compared to SCIT, various local adverse reactions such as oral itching are often observed and represent one major reason for discontinuation of therapy (4). The herein observed sublingual localization of MCs within the duct and lobe of sublingual glands might explain the caruncle swelling in some SLIT patients so that based on these data, other application sites, such as palatum or vestibulum, which do not contain excessive glands might represent attractive alternative application sites for such patients especially as the quantity of oLCs exceeded that of oMC_tc– reflected by a positive oLC/oMC_tc ratio within these regions. Nevertheless, high numbers of oMC_tc might not only account for the observed side-effects of SLIT but also contribute to allergen presentation as it has been shown that MC activation-induced LCs migration in mice (20). The observed accumulation of oLCs within the rete ridges might point towards a higher migratory activity of oLCs, where they could get in contact with CD3⁺ T cells located in the lamina propria for efficient antigen/allergen presentation. This speculation is further supported by recent publications, which report about the presence of CD4⁺ T cells and CD83⁺ dendritic cells within inflamed lamina propria of gingival tissue suggesting a local antigen presentation site apart from regional lymph nodes, a mechanism which might account for the effectiveness of SLIT (21, 22). In view of the oLCs distribution within different regions of the oral cavity, the fact of postmortem alteration has to be considered. In this context, a previous study could demonstrate that postmortem cell density within the oral mucosa is equivalent to ex vivo density (17). Nevertheless, other APC subsets such as plasmacytoid dendritic cells have been shown to play a crucial role in allergy. However, in a previous study we could show that these cells are virtually absent from the uninflamed oral mucosa (19). It has been shown that the sublingual region has the highest permeability within the oral mucosa (23), which might justify its current preference as allergen-application site. Nevertheless, other oral mucosa sites such as buccal mucosa were found to have a high diffusion rate as well (23, 24), which should be sufficient for allergen uptake as allergens reside up to 20 h within the oral cavity after application (25). At any rate, more studies investigating allergen uptake within the different regions of the oral cavity are necessary to draw a final conclusion on this issue.

In view of our data, different mucosal regions such as the vestibulum or bucca might represent alternative application sites because of high oLCs density and high FcεRI expression on oLC presuming most effective allergen uptake, especially in SLIT patients suffering from sublingual edema and swelling of caruncle for it is anatomically separated from the sublingual region by teeth alignment. Furthermore, our data might serve as a basis for the development of new application forms of SLIT such as tablets or stripes, which assure allergen uptake within a defined and limited oral region to increase the efficacy and safety of SLIT.

Acknowledgment

This work was supported by grants of the Deutsche Forschungsgemeinschaft DFG NO454/4-1, SFB704 TPA4 a BONFOR grants of the University of Bonn and partially by a grant from Stallergenes. N.N. is supported by a Heisenberg-Fellowship of the DFG NO454/3-1.