Hyperexpression of the High-Affinity IgE Receptor-β Chain in Chronic Allergic Keratoconjunctivitis

Abstract

purpose. Although the existence of FcεRI-αβγ₂ and FcεRI-αγ₂ receptor subtypes was reported, there has been no direct evidence of these two subtypes of FcεRI in vivo. To investigate the existence of these two subtypes of FcεRI in vivo, the authors evaluated the expression of FcεRI-β in the giant papillae of chronic allergic conjunctivitis and compared the expression level of FcεRI-β with control conjunctivae using the anti–human FcεRI-β antibody.

methods. FcεRI-β expression in giant papillae obtained from patients with atopic keratoconjunctivitis and vernal keratoconjunctivitis in control conjunctivae was evaluated by immunohistochemistry using anti–FcεRI-β, -α, -γ, and anti–human mast cell tryptase, anti–chymase, anti–basophil, and anti–CD1a antibodies.

results. Statistical analyses revealed that the densities of FcεRI-β⁺ cells, FcεRI-α⁺ cells, tryptase⁺ cells, and FcεRI-β⁺/tryptase⁺ cells were significantly increased in giant papillae compared with controls. There were two types of FcεRI (αβγ₂ and αγ₂) on the mast cells of the giant papillae. The ratio of the FcεRI-β⁺ cell number/FcεRI-α⁺ cell number in the giant papillae (0.69 ± 0.08 [mean ± SD]) was significantly higher than that of the controls (0.07 ± 0.16). FcεRI-β/tryptase double immunostaining revealed that 81% ± 13% of tryptase⁺ cells expressed FcεRI-β. FcεRI-β⁺ cells were preferentially localized within and around epithelial tissue. The authors also found that FcεRI-β was expressed by basophils but not by FcεRI-αγ₂–positive Langerhans cells in the giant papillae samples.

conclusions. Preferential FcεRI-β expression observed in the mast cells and basophils of giant papillae suggests important roles of FcεRI-β in the pathophysiology of atopic keratoconjunctivitis and vernal keratoconjunctivitis.

Human high-affinity IgE receptor (FcεRI) exists in two isoform, a tetramer containing the β chain FcεRI-αβγ₂ and a trimer lacking the β chain FcεRI-αγ₂, depending on cell type. ¹ The FcεRI-β gene (MS4A2) has been recognized as an atopy-related gene, initially discovered from a genetic linkage study ² and also from a genetic association study ³ by our group, and the functional roles of FcεRI-β protein have been investigated extensively. For example, the FcεRI-β chain mediates intracellular signaling through the immunoreceptor tyrosine-based activation motif and is phosphorylated in response to antigen cross-linking of the receptor-bound IgE. ¹ The human FcεRI-β chain acts as an amplifier for mast cell activation and cell surface expression of FcεRI. ^{4 5} Although the existence of the FcεRI-αβγ₂ and FcεRI-αγ₂ receptor subtypes has been reported, ⁶ there has been no direct evidence of these two subtypes of FcεRI in vivo at the protein level. Furthermore, the precise pathophysiological roles of the FcεRI-β chain in human atopic diseases remain unclear. Recently, we raised an antibody against human FcεRI-β that was useful for the in situ detection of the FcεRI-β protein. ⁷  

Atopic keratoconjunctivitis (AKC) ^{8 9} and vernal keratoconjunctivititis (VKC) ¹⁰ are the most severe forms of chronic allergic conjunctivitis. Massive infiltration of mast cells occurs, and serum and tear IgE levels are significantly higher than in healthy controls. ^{8 11 12} In addition, AKC and VKC tend to form giant papillae at the upper tarsal conjunctiva. ^{8 10 13} We resected giant papillae for therapeutic purposes ¹⁴ and carried out histopathologic analysis with the resected tissues using our newly generated anti–FcεRI-β–specific antibody. We reported previously that IgE-bearing FcεRI-α⁺ mast cells were increased in the giant papillae of patients with VKC ¹⁵ ; however, the expression of FcεRI-β has yet to be evaluated. We found preferential FcεRI-β expression in the mast cells of giant papillae samples compared with those of the control conjunctivae. 

Materials and Methods

Antibodies

Rabbit antiserum against unique C-terminal sequences of human FcεRI-β (CYSELEDPGEMSPPIDL) was generated and the antiserum was purified on a protein-A column, as previously described. ⁷ Because this antibody was raised against the C-terminal region of human FcεRI-β protein, it did not recognize the truncated form of FcεRI-β protein described previously. ¹⁶ Other antibodies used in this study included Alexa 488–conjugated-goat anti–rabbit-F(ab′)₂ and Alexa 594–conjugated goat anti–mouse IgG-F(ab′)₂ (Invitrogen, Carlsbad, CA), Cy5-conjugated goat anti–mouse IgG1 antibody (Southern Biotechnology, Birmingham, AL), rabbit anti–FcεRI-γ polyclonal antibody (Upstate Biotechnology, Lake Placid, NY), phycoerythrin (PE)-conjugated mouse anti–FcεRI-α monoclonal antibody (clone CRA1; e-bioscience, Tokyo, Japan), mouse anti–chymase monoclonal antibody (clone CC1; LAB Vision, Fremont, CA), mouse anti–CD1a monoclonal antibody (clone O-10; Santa Cruz Biotechnology Inc., Santa Cruz, CA), and mouse anti–mast-cell tryptase monoclonal antibody (clone AA-1; Dako Japan, Kyoto, Japan). Mouse anti–basophil monoclonal antibody (clone BB-1) was generated as previously described. ¹⁷  

AKC/VKC Patient Selection and Giant Papillae Tissue Processing

Giant papillae were resected for therapeutic purposes ¹⁴ from six patients with AKC and four patients with VKC after obtaining informed consent (Table 1) .AKC was defined as bilateral, chronic inflammation of the conjunctiva and lids associated with atopic dermatitis, and VKC was defined as bilateral, chronic inflammation of the conjunctiva associated with predisposition to atopy. ¹⁸ Patients who had atopic dermatitis or corneal stromal neovascularization were excluded from the VKC diagnosis. All patients with AKC or VKC were treated by topical dexamethasone eyedrops for at least 4 weeks before surgery. Some patients were also treated by an oral steroid (20 mg/day prednisolone) or by cyclosporine A eyedrops. Total serum IgE concentration and specific IgE titer against 26 common antigens (including house dust mite and Japanese cedar pollen) were measured by SRL, Inc. (Tokyo, Japan) using the fluorescence enzyme immunoassay method and the enzyme-linked immunosorbent assay (ELISA) method, respectively. Upper bulbar conjunctivae were resected from six patients with conjunctivochalasis and four patients with superior limbic keratoconjunctivitis (SLK) ¹⁹ for therapeutic purposes after informed consent was obtained (Table 2) . Conjunctivochalasis was defined as a redundant conjunctiva typically located between the globe and the lower eyelid, ²⁰ and SLK was defined according to the original clinical descriptions by Theodore. ²¹ Giant papillae were fixed with 4% paraformaldehyde (PFA)-phosphate buffered saline (PBS) for at least 4 hours and were then immersed in a 20% sucrose-PBS solution for 30 minutes, rapidly frozen in OCT compound (Sakura Finetek, Tokyo, Japan), and stored at −80°C. All procedures were approved by the ethics committee of Kyoto Prefectural University of Medicine, and the study was conducted in accordance with the tenets of the Declaration of Helsinki. 

Immunohistochemical Analysis of Giant Papillae

Seven-micrometer cryostat sections were made from the specimens and air dried. Sections were then postfixed with 4% PFA-PBS. After blocking with 1% bovine serum albumin (BSA) in PBS, the slides were reacted with anti–FcεRI-β or with anti–FcεRI-γ polyclonal antibodies for 1 hour and then with Alexa 488-conjugated anti–rabbit IgG antibody for 30 minutes. For double-staining with anti–tryptase, anti–FcεRI-α, anti–basophil, or anti–CD1a monoclonal antibodies, the slides were incubated simultaneously with FcεRI-β or FcεRI-γ polyclonal antibodies and with one of the monoclonal antibodies. In the case of double immunostaining with the anti–chymase antibody, the slides were incubated with the anti–chymase antibody overnight at 4°C, and then the FcεRI-β polyclonal antibody was added and further incubated for 1 hour at room temperature. Alexa 488–conjugated anti–rabbit IgG antibody and Alexa 594 conjugated anti–mouse IgG antibody were mixed and applied simultaneously as second antibodies. For triple immunostaining using the CD1a (class, mouse IgG1), FcεRI-α (class, mouse IgG2a), and FcεRI-β antibodies, the slides were incubated with the CD1a antibody and the FcεRI-β antibody for 1 hour at room temperature. After PBS washes, Alexa 488-anti–rabbit IgG antibody, Cy5-anti–mouse IgG₁ antibody, and PE-anti–FcεRI-α (class, mouse IgG2a) were applied simultaneously for another hour. As negative controls, normal rabbit IgG (Santa Cruz Biotechnology) was used instead of the FcεRI-β polyclonal antibody at the same concentration, or the FcεRI-β IgG antibodies were preabsorbed with a fivefold excess amount of the peptide used for immunization. These slides were then visualized with a confocal laser scanning microscope (FV1000; Olympus Corp., Tokyo, Japan). 

Cell Counting

Numbers of FcεRI-β⁺ and FcεRI-α⁺ cells (membrane staining) and tryptase⁺ cells (intracellular staining) were counted manually using 200× magnification immunostaining images. Positive cells, both in the epithelium and in the substantia propria, were counted per 1-mm unit length of the conjunctival surface as total. We counted three sections per each patient for FcεRI-β immunostaining and two sections for the remaining staining. All immunostained sections were evaluated by an observer who was blinded to the clinical data of the patients. We carried out two independent series of cell staining and counting, and we showed one representative result. 

Results

FcεRI-β Immunostaining of Giant Papillae and Control Conjunctivae

All the tested giant papillae samples (n = 10; Table 1 ) showed focal-positive immunostaining in the epithelium and in the substantia propria with anti–FcεRI-β (Fig. 1A) , and the negative-control slide stained by normal rabbit IgG (Fig. 1B)and by the preabsorbed FcεRI-β antibody (data not shown) did not show any positive staining. In addition to anti–FcεRI-β staining (Fig. 2A) , all the tested giant papillae showed positive staining with anti–FcεRI-α antibodies (Fig. 2B) ; there were FcεRI-α/FcεRI-β double-positive and FcεRI-α single-positive cells (Fig. 2C) . Double-immunohistochemical staining of the nearby section shown in Figure 2Awith anti–FcεRI-γ antibodies (Fig. 2D)and with anti–FcεRI-α antibodies (Fig. 2E)showed that all the FcεRI-α⁺ cells were also FcεRI-γ⁺ (Fig. 2F) . Control upper bulbar conjunctivae from the conjunctivochalasis or SLK patients (n = 10; Table 2 ) showed a few FcεRI-α⁺ cells (Figs. 3A 3C)or a few tryptase-positive mast cells (Figs. 3B 3D)in its substantia propria but a negligible number of FcεRI-β⁺ cells (Fig. 3D ; Table 3 ). 

Immunolocalization and Quantification of FcεRI-β⁺ Cells in Giant Papillae

The giant papillae were also double immunostained with anti–FcεRI-β and the antibodies to typical mast cell proteases (tryptase and chymase). FcεRI-β immunostaining was observed at the cell periphery of the tryptase⁺ cells (Figs. 4C 4G) . Some of the tryptase⁺ cells were FcεRI-β negative (Fig. 4C 4G ; arrow); however, all the chymase⁺ cells were FcεRI-β⁺ (Fig. 4I , asterisks). We quantified the number of FcεRI-β⁺ cells, FcεRI-α⁺ cells, tryptase⁺ cells, and FcεRI-β/tryptase double-positive cells. The numbers of FcεRI-β⁺, FcεRI-α⁺, and tryptase⁺ cells were significantly higher in the giant papillae than in the control conjunctivae (Table 3) . The ratio of the FcεRI-β⁺ cells/FcεRI-α⁺ cells in the giant papillae samples was 0.69 ± 0.08 (mean ± SD). That ratio was significantly higher than the ratio of the control samples (0.07 ± 0.16). Of tryptase⁺ cells, 81% ± 13% expressed FcεRI-β, a rate also significantly higher than that of the control samples. Samples with hypertrophic epithelium showed FcεRI-β⁺ cells predominantly at the epithelial layers (Figs. 4A 4D 4H) . We found that intraepithelial FcεRI-β⁺ cells were dominated by tryptase⁺ and chymase⁻ mast cells (MC_T; Figs. 4G 4I ). 

We also found some FcεRI-β⁺ cells within and around convoluted epithelium ²² and pseudotubules ²³ (Figs. 5A 5B ; asterisks). Double immunostaining with FcεRI-β and tryptase (Figs. 5C 5D)showed cytoplasmic tryptase staining (red) and membranous FcεRI-β staining (green). Double immunostaining with FcεRI-β and chymase (Figs. 5E 5F)showed cytoplasmic chymase staining (red) at some of the FcεRI-β⁺ cells (green). The results of the FcεRI-β⁺ cell quantification are summarized in Table 3 . Statistical analyses revealed that the densities of FcεRI-β⁺ cells, and FcεRI-β⁺-tryptase⁺ mast cells were significantly increased in giant papillae compared with control conjunctivae. Although the average number of FcεRI-β⁺ cells was higher for SLK (1.7 ± 0.7) than for conjunctivochalasis (0.7 ± 1.1) samples, the difference was not statistically significant (P = 0.38, Student’s t-test), and the average number of FcεRI-β⁺ cells in SLK samples was still significantly lower than the number of FcεRI-β⁺ cells in giant papillae (P = 0.00001, Student’s t-test). 

FcεRI-β Expression in the Basophils of Giant Papillae Tissues but Not in Langerhans Cells (LCs)

During immunohistochemical analysis we found some tryptase⁻-FcεRI-β⁺ cells in the giant papillae sections, so we speculated that the FcεRI-β⁺ cells included basophils. Positive basophil staining was observed at the substantia propria of the giant papillae (Fig. 6A) . Double immunostaining using anti–FcεRI-β and anti–basophil antibodies showed FcεRI-β immunostaining (green) at the periphery of the basophil (red) (Fig. 6C) . To examine the FcεRI-receptor subtype in LCs of the giant papillae, immunostaining with CD1a and FcεRI-α/FcεRI-β antibodies was carried out. Double-staining did not reveal any CD1a⁺/FcεRI-β⁺ cells (Figs. 7A 7C 7E) . On the other hand, two CD1a⁺ cells were FcεRI-α⁺ cells (Figs. 7B 7C 7D ; arrows). By triple-immunohistochemical staining, we confirmed the existence of FcεRI-α⁺/CD1a⁺/FcεRI-β⁻ LCs (Fig. 7E , arrow) and of FcεRI-α⁺/FcεRI-β⁺/CD1a⁻ mast cells (Fig. 7E , arrowhead). 

Discussion

To the best of our knowledge, this study is the first to demonstrate the existence of both FcεRI-αβγ₂ and FcεRI-αγ₂ receptor subtypes in the giant papillae of patients with chronic allergic conjunctivitis. FcεRI-β staining (Figs. 1 2)showed a membranous staining pattern at the giant papillae. This staining pattern is consistent with our previous in vitro study using cultured human mast cells. ⁷ FcεRI-α staining of the same slide showed a broader expression at the conjunctiva than that of FcεRI-β (Fig. 2C) , and all the FcεRI-α⁺ cells were FcεRI-γ⁺ (Fig. 2F) . These findings showed the existence of FcεRI-αβγ₂ mast cells and FcεRI-αγ₂ mast cells in the giant papillae. In our previous study, we quantified FcεRI-β protein expression in human mast cells using flow cytometry. ⁷ We found monophasic rather than biphasic FcεRI-β expression pattern in human mast cells, ⁷ and Western blotting showed that the sensitivity of the FcεRI-β antibody is approximately 100 pg in the recombinant FcεRI-β protein (Matsuda A, unpublished data, 2008). These results suggested that there are various levels of FcεRI-β expression in each mast cell, and we identified an area of the mast cell population that expresses the FcεRI-β protein past the threshold as FcεRI-β-positive cells. We could not deny the possibility that FcεRI-αβγ₂ and FcεRI-αγ₂ receptor subtypes could be coexpressed in one mast cell. Nevertheless, we could conclude that there were at least two types of mast cells dominated by FcεRI-αβγ₂ or by FcεRI-αγ₂ receptors. We further examined FcεRI-β expression in the conjunctivae of two nonatopic conjunctival diseases as controls. We selected patients with conjunctivochalasis because relatively large upper bulbar conjunctivae samples could be obtained at the time of surgery. We examined patients with SLK as controls for two reasons: because an increased number of mast cells has been reported in the SLK conjunctivae ²⁴ and to avoid bias from the very low number of the mast cells in the conjunctivochalasis samples. In addition, we excluded allergic conjunctivitis by careful slit-lamp examinations in these cases. The increased number of FcεRI-β⁺ cells and the higher ratio of FcεRI-β⁺/FcεRI-α⁺ cells number in giant papillae samples suggested the preferential expression of FcεRI-β protein in chronic allergic conjunctivitis. 

Double immunohistochemical staining with anti–FcεRI-β and anti–tryptase/anti–chymase antibodies showed that 81% of tryptase⁺ cells were FcεRI-β⁺, and all the chymase⁺ cells were FcεRI-β⁺ (Figs. 4 5 ; Table 3 ). The FcεRI-β⁺ rate among chymase-positive mast cells was higher than it was among among tryptase-positive mast cells; the reason for this is unknown and requires further investigation. During the analysis of double immunostaining with anti–FcεRI-β and anti–tryptase antibodies, we found the preferential localization of FcεRI-β⁺ mast cells within and around hypertrophic or convoluted epithelium and epithelial pseudotubules (Figs. 4 5) . It was reported that antigen challenge in sensitized animals induced mast cell infiltration in esophageal epithelium. ²⁵ Kitaura et al. ²⁶ reported that a combination of antigen and IgE could stimulate mast cell migration by autocrine/paracrine secretion of chemokines. They also showed that Lyn, a signal transduction molecule associated with FcεRI-β immunoreceptor tyrosine-based activation motif, ²⁷ plays an essential role for the antigen-IgE-induced mast cell migration. We considered the possibility that the accumulation of FcεRI-β⁺ mast cells in the hypertrophic epithelium could be a reflection of increased antigen stimulation at the ocular surface of AKC/VKC eyes. Furthermore, Galli et al. ²⁸ reported that mast cells could induce airway epithelial cell proliferation in mouse asthma models in response to antigen stimulation; hence, it is possible that the infiltrating FcεRI-β⁺ mast cells themselves play some roles in epithelial hypertrophy. 

This is the first study to show the existence of FcεRI-αβγ₂ ⁺ basophils (Fig. 6)and FcεRI-αγ₂ ⁺ LC (Fig. 7)at the protein level in giant papillae. Recently, it was reported that basophils play major roles in delayed-phase allergic reaction, independently of T cells and mast cells. ²⁹ We are now investigating the role of FcεRI-β⁺ basophils in the pathophysiology of AKC/VKC. The existence of FcεRI-αγ₂ ⁺ LC is consistent with previous findings in mRNA levels, ³⁰ and FcεRI-αγ₂ ⁺ LCs play important roles for enhanced antigen presentation in atopic disorders. 

Recent studies have also indicated that the role of the FcεRI-β protein in mast cells depends on the amount of IgE-specific antigen, with the low concentration of antigen FcεRI-β chain working as an amplifier ³¹ but with a supraoptimal amount of antigen FcεRI-β chain acting as an inhibitory molecule for degranulation and cytokine expression. ^{31 32} In this study we showed preferential FcεRI-β expression in the giant papillae samples in AKC/VKC, suggesting the involvement of an FcεRI-β–mediated mechanism for amplifying reactions against chronic low concentration of antigens. Additional functional studies will be necessary for investigating the roles of the FcεRI-β chain; this antibody will be a useful tool for in situ analysis. 

 Supported in part by Wellcome Trust Traveling Research Fellowship 066797/Z/02/Z (AM); Grants-in-Aid for Scientific Research Programs 19659454 (SK), 18604009 (AM), and 20591195 (YO) from the Japanese Ministry of Education, Culture, Sports, Science and Technology; and the Japanese National Institute of Biomedical Innovation Program (Project ID05–24).

 Submitted for publication October 29, 2008; revised November 24 and December 13 and 16, 2008; accepted March 30, 2009.

 Disclosure: A. Matsuda, None; Y. Okayama, None; N. Ebihara, None; N. Yokoi, None; J. Hamuro, None; A.F. Walls, None; C. Ra, None; J.M. Hopkin, None; S. Kinoshita, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Clinical Characteristics of Patients with AKC/VKC and FcεRI-β⁺ Cell Numbers of Giant Papillae

Patient	Age (y)	Sex	Total IgE	Specific IgE	Diagnosis	FcεRI-β+ Cells (mean ± SD)	Treatment
1	16	F	509	Positive	VKC	62.5 ± 20.0	Dex, CsA
2	22	M	89	Positive	VKC	53.9 ± 8.3	Dex
3	13	M	2319	Positive	VKC	20.2 ± 8.3	Dex
4	18	M	375	Positive	AKC	49.3 ± 22.0	Dex, CsA
5	17	M	17260	Positive	AKC	56.2 ± 21.2	Dex, CsA, oral steroid
6	21	M	1904	Positive	AKC	34.9 ± 18.0	Dex
7	16	M	3763	Positive	AKC	29.2 ± 17.0	Dex
8	19	M	124	Negative	VKC	49.4 ± 20.0	Dex
9	23	M	20328	Positive	AKC	78.3 ± 25.5	Dex
10	45	F	28	Negative	AKC	48.2 ± 12.0	Dex

 Dex, 0.1% dexamethasone eyedrop; CsA, 0.1% cyclosporine eyedrop.

Table 2.

 

Clinical Characteristics of Control Patients

Patient	Age (y)	Sex	Diagnosis
1	39	F	SLK
2	49	F	SLK
3	70	F	Conjunctivochalasis
4	65	M	Conjunctivochalasis
5	71	F	Conjunctivochalasis
6	74	M	Conjunctivochalasis
7	53	F	Conjunctivochalasis
8	40	M	SLK
9	75	F	Conjunctivochalasis
10	59	F	SLK

Figure 1.

 

Anti–FcεRI-β immunostaining of giant papillae. Immunohistochemical staining was carried out with giant papillae specimens from patients with AKC or VKC using the anti–FcεRI-β antibody (A) and control rabbit IgG (B) at the same concentration. Original magnification, 200×. This is representative data obtained from 1 of 10 patients (patient 5 in Table 1 ).

Figure 2.

 

Two types of FcεRI⁺ (αβγ₂ and αγ₂) cells in giant papillae. Immunohistochemical staining was carried out with giant papillae specimens from patients with AKC or VKC using anti–FcεRI-β (A, green) and anti–FcεRI-α (B, red) antibodies. FcεRI-β was merged with FcεRI-α (C). Arrows indicate FcεRI-α/β double-positive cells (yellow) and arrowheads indicate FcεRI-α single-positive (red) cells. A nearby section from the same patient was immunostained with anti–FcεRI-γ (D, green) and anti–FcεRI-α (E, red) antibodies. FcεRI-γ was merged with FcεRI-α (F). Original magnification, 400×. This is representative data obtained from 1 of 10 patients (patient 1 in Table 1 ).

Figure 3.

 

Expression of the FcεRI-β chain was almost negligible in control conjunctivae. Upper bulbar conjunctivae were obtained from conjunctivochalasis patients (A, B) and SLK patients (C, D). The conjunctivae were double immunostained with the pairs of anti–FcεRI-α (red) and anti–FcεRI-β (A, C, green) or anti–tryptase (red) and FcεRI-β (green) antibodies (B, D). (D) Arrow indicates FcεRI-β^± tryptase⁺ cells. Original magnification, 200×.

Table 3.

 

Comparison of the FcεRI- and Tryptase-Positive Cells between Giant Papillae and Control Samples

	FcεRI-β+ Cells (mean ± SD)	FcεRI-α+ Cells (mean ± SD)	Tryptase+ Cells (mean ± SD)	Ratio of FcεRI-β+ and FcεRI-α+ Cells (mean ± SD)	Ratio of FcεRI-β+ and Tryptase+ Cells (mean ± SD)
Giant papillae	55.9 ± 24.6*	99.0 ± 33.2, †	71.1 ± 24.1, ‡	0.69 ± 0.08, §	0.81 ± 0.13, ∥
Control	1.1 ± 2.3*	19.6 ± 16.9, †	17.6 ± 9.8, ‡	0.07 ± 0.16, §	0.06 ± 0.11, ∥

 Student’s t-test:

*   P = 0.0000007;

†   P = 0.0001;

‡   P = 0.00005;

§   P = 0.00000007;

∥   P = 3 × 10⁻⁹.

Figure 4.

 

Localization of FcεRI-β⁺ cells in hypertrophic epithelium of giant papillae. Giant papillae with hypertrophic epithelium were double immunostained with anti–FcεRI-β (A, green) and anti–tryptase (B, red) antibodies. FcεRI-β was merged with tryptase (C). Adjacent section was also double immunostained with anti–FcεRI-β (D, green) and anti–chymase (E, red) antibodies, FcεRI-β was merged with chymase (F). Arrowheads indicate the boundary line between epithelium and substantia propria. FcεRI-β immunostaining of the hypertrophic epithelial region is shown at higher magnification (H), and merged with tryptase (G) and with chymase (I). Arrows and asterisks indicate FcεRI-β⁻-tryptase⁺ cells (C, G) and FcεRI-β⁺-chymase⁺ cells (I), respectively. Original magnification, 100× (A–F); 400× (G–I). This is representative data obtained from 1 of 10 patients (patient 10 in Table 1 ).

Figure 5.

 

Localization of FcεRI-β⁺ cells in convoluted epthelium and pseudotubules. Giant papillae with convoluted epithelium were immunostained with the anti–FcεRI-β antibody (A, green). Asterisks indicate convoluted epithelium and pseudotubules. The staining around convoluted epithelium (B, C, E) and pseudotubules (D, F) is shown at higher magnification. Some slides were double immunostained with anti–tryptase (C, D, red), and FcεRI-β was merged with them. Arrows indicate FcεRI-β⁺-tryptase⁺ cells. Adjacent sections were double immunostained with anti–chymase (E, F, red), and FcεRI-β was merged with them. Arrows indicate FcεRI-β⁺-chymase⁺ cells. Original magnifications, 100× (A); 200× (B–F). This is representative data obtained from 1 of 10 patients (patient 2 in Table 1 ).

Figure 6.

 

Expression of FcεRI-β in basophils of giant papillae. Giant papillae specimens from a patient with AKC (patient 6 in Table 1 ) were immunostained with the anti–basophil antibody (A; arrows indicate areas of positive staining). The same sections were stained with anti–FcεRI-β antibody and are shown in higher magnification (B). Double immunostaining using anti–FcεRI-β and anti–basophil antibodies is shown in (C). FcεRI-β immunostaining (green) was observed at the periphery of the anti–basophil-positive (red) cells. Original magnifications, 200× (A); 1000× (B, C).

Figure 7.

 

Langerhans cells in giant papillae expressing the FcεRI-αγ₂ subtype. Giant papillae from a patient with AKC (patient 4 in Table 1 ) was stained with the anti–FcεRI-β antibody (A), anti–FcεRI-α antibody (B), and anti–CD1a antibody (C). Double immunostaining with the anti–FcεRI-α antibody and CD1a antibody (D) showed two CD1a/FcεRI-α double-positive cells (D; arrows). Triple immunostaining using CD1a (silver), FcεRI-α (red), and FcεRI-β (green) antibodies showed no existence of FcεRI-β/CD1a double-positive cells. CD1a⁺/FcεRI-α⁺/FcεRI-β⁻ LCs and FcεRI-β⁺/FcεRI-α⁺ mast cells (E, arrows and arrowhead, respectively). Merged image using a differential contrast microscope (F). Original magnification, 400×.

The authors thank Hisako Takeshita for excellent technical assistance and John Bush for reviewing the manuscript.