Degranulated mast cells may be missed by histochemical and immunohistological techniques, making the histological assessment of activated mast cells difficult.
To use a novel mast cell surface antigen 1 (MASA-1) to detect activated mast cells.
Surgical samples of human intestine were obtained from 15 patients with ulcerative colitis (UC), 14 patients with Crohn's disease (CD) and 11 controls. Frozen sections were cut and MASA-1 was detected and quantified by enzyme immunohistology. Immunohistological double staining with anti-MASA-1 and either anti-CD68 or anti-c-Kit antibody was also performed.
The number of MASA-1-positive cells was significantly higher in the submucosal layer and muscularis propria in patients with CD than in patients with UC or controls. The number of MASA-1-positive cells was significantly higher in the muscularis propria in patients with UC in highly inflamed colon compared with controls and less inflamed colon. Human intestinal MASA-1-positive cells are not from the macrophage lineage and showed heterogeneous expression of c-Kit.
The increased number of mast cells in the submucosa and muscularis propria in patients with CD suggests that the involvement of activated mast cells in transmural inflammatory responses possibly including tissue remodelling.
Transmural chronic inflammation followed by intestinal remodelling is a specific feature of Crohn's disease (CD) that distinguishes it from ulcerative colitis (UC). Several types of inflammatory cells are responsible for this tissue destruction and remodelling.1, 2 Mast cells are involved not only in acute allergic reactions but also in chronic inflammatory diseases producing inflammatory mediators, finally resulting in fibrosis.3 Because of technical and methodological problems, estimates of intestinal mast cell numbers in inflammatory bowel disease (IBD) are variable. Immunohistological detection of mast cells is a recent introduction, replacing toluidine blue staining; however, degranulated mast cells are missed by anti-tryptase immunostaining. In this study, we demonstrate the presence of intestinal mast cells using a novel anti-human mast cell surface antigen 1 (MASA-1) antibody that recognizes a subset of activated mast cells.4 Recombinant mRNA was made from a MASA-1 gene, cloned from a human umbilical vein mast cell cDNA library (NCBI E62944), the functional role of which is yet to be clarified and rabbit antibody against MASA-1 was obtained.4
Surgical specimens of human colon were obtained from 15 patients with UC (7 males, 8 females; mean age 45 years, range 17–71 years), 14 patients with CD (12 males, 2 females; mean age 27 years, range 17–34 years) and 11 controls (7 males, 4 females; mean age 63 years, range 51–82 years). Macroscopic normal colon was obtained from controls whose colon had been resected for carcinoma. Fourteen patients with UC and six patients with CD were taking corticosteroids at the time of surgery.
Surgical specimens from UC, CD and controls were fixed with periodate–lysine–4% paraformaldehyde and 5 μm cryostat sections were cut. Sections were pre-treated with 5% normal goat serum to inhibit non-specific protein binding and endogenous peroxidase activity was inactivated by treatment for 15 min with phosphate-buffered saline (PBS) with 0.05% sodium azide and 0.3% hydrogen peroxide. After washing, sections were reacted with MASA-1 antibody in 1/100 dilution at 4 °C overnight. After washing, sections were reacted with goat anti-rabbit immunoglobulin (IgG) conjugated to peroxidase-labelled dextran polymer (DAKO EnVision, Peroxidase Rabbit Ready-to-use DAKO Corporation, Carpinteria, CA, USA). Peroxidase-substrate reaction was developed with 0.03% 3′, 3-diaminobenzidine tetrahydrochloride containing 0.006% hydrogen peroxide and 0.065% sodium azide Tris-buffered saline. Control staining was performed, omitting first antibodies. Tissue sections were counterstained with haematoxylin and observed under a light microscope. MASA-1-positive cells were counted under 400× magnification for five microscopic fields using an optical grid in the lamina propria, submucosa and muscularis propria layers and expressed as cells/mm2. Conventional haematoxylin and eosin staining was carried out, and histological activity was evaluated according to Matts’ criteria for patients with UC.5
Immunofluorescent double staining
Immunofluorescent double staining was performed to detect MASA-1 and either c-Kit or CD68. Tissue sections were prepared as described above, and non-specific protein binding was blocked after pre-treatment with 5% normal goat serum. Sections were reacted with rabbit anti-human MASA-1 antibody in 1/100 dilution at 4 °C overnight, washed with PBS containing 0.05% Tween-20 (PBS-T) and further reacted with fluorescein-conjugated goat anti-rabbit IgG (Fab2) antibody (CN Pharmaceuticals, Inc., Aurora, OH, USA) in 1/20 dilution, at ambient temperature for 30 min in a dark room after this procedure. Sections were washed with PBS-T, and reacted either with mouse anti-human CD68 (EBM11) monoclonal antibody (DAKO A/S, Glostrup, Denmark) in 1/1600 dilution, or anti-human c-Kit antibody (Neo Markers, Fremont, CA, USA) in 1/40 dilution, at ambient temperature, for 60 min. After washing with PBS-T, sections were reacted with rhodamine-conjugated goat anti-mouse IgG antibody (CHEMICON International, Inc., Temecula, CA, USA) in 1/50 dilution at ambient temperature for 30 min. Sections were mounted in glycerin gel and observed under a fluorescent microscope (OLYMPUS BX50; OLYMPUS, Tokyo, Japan) equipped with a colour camera (Olympus Color Chilled 3CCD Camera, M-3204C).
Data were expressed as mean ± s.d. Differences between groups were analysed using one-way analysis of variance with Bonferroni/Dunn correction. P values less than 0.05 were considered significant.
MASA-1-positive cells were found in the lamina propria (Figure 1a,b), submucosa (Figure 1c) and muscularis propria (Figure 1d). Hardly any MASA-1-positive cells were found in the muscularis propria of controls and patients with UC. MASA-1-positive cells are round, mononuclear cells (Figure 1e), but did not stain with anti-human CD68 monoclonal antibody (Figure 1f,g), demonstrating that MASA-1-positive cells are clearly not from the macrophage lineage. Fluorescent immunostaining revealed that the MASA-1 molecule is expressed on the mast cell surface. Fluorescent double immunostaining revealed heterogeneity of MASA-1-positive and c-Kit-positive cells in human intestine (Figure 1h–j).
The number of MASA-1-positive cells/mm2 in the lamina propria was 44 ± 22, 37 ± 34 and 62 ± 32 in controls, patients with UC, and patients with CD, respectively (Figure 2). There was no significant difference between these three groups in terms of the number of MASA-1-positive lamina propria cells/mm2. The number of MASA-1-positive cells/mm2 in the submucosa was 22 ± 15, 24 ± 18 and 43 ± 24 in controls, patients with UC and patients with CD, respectively (Figure 2). The number of MASA-1-positive cells/mm2 in the submucosa was significantly higher in patients with CD than in controls (P = 0.013) or patients with UC (P = 0.0121). The number of MASA-1-positive cells/mm2 in the muscularis propria was 2 ± 3, 3 ± 6 and 13 ± 10 in controls, patients with UC and patients with CD, respectively (Figure 2). The number of MASA-1-positive cells/mm2 in the muscularis propria was significantly higher in patients with CD than in controls (P = 0.0006) or patients with UC (P = 0.0013).
The number of patients with UC meeting Matts’ criteria for scores of 1, 2, 3, 4 and 5 were 3, 3, 1, 5 and 2 patients, respectively. For statistical analysis, controls, Matts’ 1–3 and Matts’ 4, 5 were analysed. The number of MASA-1-positive cells in the lamina propria and submucosa did not differ among controls, Matts’ 1–3 and Matts’ 4, 5 (Figure 3a,b). However, the number of MASA-1-positive cells in muscularis propria were 2.0 ± 2.6, 0.5 ± 0.8 and 6.8 ± 8.1 in controls, Matts’ 1–3 and Matts’ 4, 5, respectively (controls vs. Matts’ 4, 5, P = 0.0369 and Matts’ 1–3 vs. Matts’ 4, 5, P = .0134). The number of MASA-1-positive cells were significantly higher in Matts’ 4, 5 than controls and Matts’ 1, 3 (Figure 3c). The number of MASA-1-positive cells did not significantly differ between Matts’ 4, 5 group and patients with CD.
Mast cells are bone-marrow-derived cells and are abundantly distributed in connective tissues. Mast cells are implicated in allergic and inflammatory reactions in tissues facing the external environment such as the skin, respiratory tract and gastrointestinal tract. In contrast to the distinct heterogeneity of rodent mast cells defined as mucosal mast cells and connective tissue mast cells,6, 7 heterogeneity of human intestinal mast cells are characterized by their protease content; mast cells containing both tryptase and chymase (MCTC) and mast cells containing tryptase (MCT). Human intestinal mast cells are predominated by MCT (81%) in mucosa and by MCTC (77%) in submucosa.8 Although the functional characteristics of such distinct mast cell subpopulations are not known, mast cells play a unique role in allergic and inflammatory reactions by excreting various mediators such as histamine, prostaglandins, leukotrienes, cytokines, proteases and nitric oxide.9–12
The number of mast cells in the IBD intestine remains a matter of debate13 because mast cells have previously been identified either by histochemical methods (toluidine blue staining) or by antigranule antibody (anti-tryptase and anti-chymase). Because both techniques recognize mast cell granules, degranulated mast cells might be missed. MASA-1 is a surface antigen of the activated mast cell subpopulation and possibly a good marker for those degranulated mast cells, while the precise function of the MASA-1 molecule has yet to be clarified. In our study, the number of activated mast cells detected by anti-MASA-1 antibody was increased in the submucosa and muscularis propria in patients with CD, when compared with controls and patients with UC. This study reports for the first time the location and quantification of activated mast cells in the muscularis propria. An increased number of activated mast cells was found in the muscularis propria as well as the submucosa in patients with CD compared with controls and patients with UC, which suggests that activated mast cells play a significant role in tissue fibrosis,4 a condition that causes intestinal stricture in patients with CD. As control specimens were obtained from the patients with colonic carcinoma, the age distribution was higher in the controls. Dunlop et al.14 reported age-related decrease in the number of rectal mucosal mast cells. However, in the present study, numbers of MASA-1-positive cells did not decline in colonic mucosa and submucosa in controls compared with less inflamed UC colon (Matts’ 1–3).
MASA-1-positive cells in human intestine are round, mononuclear cells and distinctly larger than lymphocytes. Double immunohistological fluorescence showed that MASA-1-positive cells are not derived from the macrophage lineage. Interestingly, heterogeneity of c-Kit expression on mast cells was demonstrated in this study. c-Kit is a proto-oncogene product tyrosine kinase receptor structurally analogous to the receptor for colony stimulating factor. c-Kit is critical to mast cell survival and development and is activated by its ligand, stem cell factor, which is expressed on fibroblasts and endothelial cells.15, 16 Most of the dermal mast cells in chronic skin ulcers are c-Kit positive both in ulcer bed and perilesional skin.17 Expression of c-Kit in mast cells is induced during normal wound healing until day 14,17 although the population of c-Kit-positive mast cells in normal skin varies from all18, 19 to 16%17 of skin mast cells. The present results support the findings of Huttunen et al.17, indicating that there are MASA-1 but not c-Kit-positive mast cells found in human intestine. On the contrary, the MASA-1-negative, c-Kit positive cells found in human intestine may represent an inactive mast cell subpopulation.
Activation of mast cells plays a significant role in fibrogenesis both in experimental models3 and in CD.20 MASA-1 could be a useful marker for both the in situ and ex vivo mast cell activation process when analysing the role of mast cells in fibrogenetic inflammatory diseases such as CD, pulmonary fibrosis, liver cirrhosis and cardiac infarction. Modulating mast cell function may in future allow medical treatment to prevent fibrosis in such situations.
This study was supported by a Grant-in-Aid for Scientific Research 15590676 from the Ministry of Education, Science, Sports and Culture, Japan. This study was approved by the ethical committee of the university.