Abstract
Selective markers for human mast cells are of paramount importance for understanding their role in physiological and pathological processes. A mouse monoclonal antibody (MAb) designated 2C7, raised against in vitro-derived human mast cells, was used in immunoenzymatic analysis of sections from a variety of human organs. Double immuno-labeling with 2C7 and tryptase, chymase, Fc∊RIα, and c-kit was performed on cryostat tissue sections from skin, colon, uterus, breast, stomach, bladder, and lung. MAb 2C7 stained greater than 93% of the tryptase+ or chymase+ mast cells in all tissues examined. In addition, the majority of cells stained with the tryptase or chymase also stained for Fc∊RIα. However, there were a significant number of Fc∊RIα+ cells in all tissues studied that were tryptase− and/or chymase−. In contrast, MAb 2C7 in double immunoenzymatic staining co-localized with 93–96% of the Fc∊RIα+ cells in all tissues. Analysis for c-kit expression on the different tissues revealed that the majority of tryptase+ or chymase+ cells in skin, uterus, bladder, and lung stained with c-kit. However, only approximately 70-78% of tryptase+ cells in colon and stomach were c-kit+. These data suggest that MAb 2C7 appears to identify mature mast cells and a population of Fc∊RIα+, chymase−, and tryptase− cells in a variety of human tissues.
Keywords
The development of a method to produce monoclonal antibodies (MAbs) against a defined antigen has had a profound impact in many fields of research and has signaled a shift in the analysis of biological problems (Milstein 2000). Selective markers for human mast cells (MCs) are of paramount importance for the identification of MCs and their progenitors (MCPs), for understanding the ontogeny and function of these cells, and for detailed studies of their role in physiological and pathological processes. The histochemical characterization of human MCs can be assessed by the presence of metachromatically staining secretory granules, surface expression of the high-affinity IgE receptor (Fc∊RI), certain myeloid CD markers, and expression of MC granule proteases. Fc∊RI is expressed on MCs as well as basophils (MacGlashan et al. 1983). However, human MC-committed progenitors (P) often lack Fc∊RI (Rottem et al. 1994). Interestingly, the same observation has been reported using rodent MC-committed progenitors (Rodewald et al. 1996). Kit ligand (KL) has been considered to be a major growth factor for human MCs (Zsebo et al. 1990; Li et al. 1995) and its receptor, c-kit, is expressed in a variety of hemopoietic cells (Lerner et al. 1991; Galli et al. 1994).
Human MCs express various combinations of carboxypeptidase A (CPA), a chymase, cathepsin G, and at least four homologous tryptases (designated I, II/β, III, and α) (Schwartz 1985; Irani et al. 1986,1991; Goldstein et al. 1987; Miller et al. 1989,1990; Reynolds-Gurley et al. 1989; Vanderslice et al. 1990; Schechter et al. 1994). The granule-specific serine proteases tryptase and chymase are presently considered the most selective markers for human MCs. However, a recent study has demonstrated that these neutral proteases are also expressed in metachromatic cells that have features of basophils (e.g., segmented nuclei and expression of the basophil-specific antigen Bsp-1) and other features of MCs (e.g., expression of c-kit and CPA) in the peripheral blood of patients with asthma or adverse drug reactions (Li et al. 1998). Tryptase and chymase are also expressed late in MC differentiation, and antibodies directed against these proteases cannot be used to identify immature MC-committed progenitors.
It has been recognized that mature MCs in the mouse are derived from a committed progenitor (Kitamura et al. 1981; Rodewald et al. 1996). These agranular cells traverse the vascular space and enter the tissues, where they complete their differentiation and maturation processes. Local microenvironmental factors are crucial for MC differentiation in the various tissue sites (Levi-Schaffer et al. 1986; Ghildyal et al. 1993; Gurish et al. 1995). However, because of the lack of a selective marker for MCPs, it has not been possible to identify these cells in humans. Human MCs arise from CD34+/c-kit+ progenitor cells when these cultures are maintained in stem cell factor (SCF) (Kirshenbaum et al. 1991; Rottem et al. 1994). However, this population also gives rise to neutrophils, basophils, and monocytes. In a subsequent study, Kirshenbaum et al. (1999) sorted cells expressing various combinations of CD34, c-kit, and CD13 and cultured them in selective growth factors for 8 weeks. They found that MCs arose from a CD34+/c-kit+/CD13+ progenitor cell. The remaining cells were monocytes. This same subset of MCPs contains some cells with bi-potent macrophage/MC colony-forming potential, as well as some that give rise to monocyte colonies, supporting an evolutionary relationship between the MC and the monocyte/macrophage lineage (Kirshenbaum et al. 1999).
In vitro approaches in recent years have provided investigators an opportunity to study the development of hemopoietic progenitors into mature human MCs. Li and co-workers (1995,1996) have derived relatively mature MCs that express Fc∊RI and the granule proteases tryptase, chymase, and CPA by culturing bone marrow or umbilical cord blood cells in the presence of recombinant human (rh) KL and conditioned medium (CM) derived from the HBM-M cell strain. In the present study we used this population of MCs to generate an antibody that recognizes both mature and immature MCs. Our data indicate that an MAb designated 2C7 recognized in vitro-derived MCs as well as tissue MCs from a variety of human tissues. Moreover, the data also suggest that this MAb identifies a population of Fc∊RI+/chymase− or Fc∊RI+/tryptase− cells.
Materials and Methods
Derivation of MAbs to Cultured Human MCs
Human MCs were cultured from umbilical cord blood (UCB) as previously described (Li et al. 1996). Briefly, UCB cells were obtained at delivery with informed consent and collected in heparinized tubes. After washing with Dulbecco's (D) PBS, mononuclear cells were isolated over Histopaque (specific gravity 1.077) as described (Boyum 1968). Cells were washed and resuspended in 25-cm2 culture flasks at a density of 2 × 106 cells/ml in medium supplemented with 50 ng/ml of rhKL (R&D Systems; Minneapolis, MN) with 50% (v/v) HBM-M-CM. Cultured MCs (1–2 × 106) of ∼80% purity were used to immunize BALB/c mice by three monthly sequential IP injections. Three days after the final injection, 8 × 107 spleen cells were harvested from the immunized mouse and fused with 1 × 107 NS-1 murine myeloma cells as previously described (Krilis et al. 1983). Cells were placed in 96-well microtiter plates and cultured in Dulbecco's modified Eagle's medium (DMEM) containing hypoxanthine, aminopterin, and thymidine (HAT) (Life Technologies; Grand Island, NY). Those hybridomas, which secreted antibody that bound to in vitro-differentiated human MCs and in vivo-differentiated human cutaneous MCs were selected and cloned by limiting dilution into 96 microtiter wells containing normal mouse spleen cells as a feeder layer. An MAb designated 2C7 was cloned and selected for further study. After three limiting dilutions to ensure the clonal nature of the hybridomas, the 2C7 clone was grown in culture and 1 × 106 cells were injected IP to obtain ascites, which was purified by Bioquest (Sydney, Australia). The Ig isotype of the resulting antibody was determined using an Ig subtyping kit (Sigma; St Louis, MO).
Tissues and Cell Preparation
Histologically normal adult human tissues were obtained at surgery with informed consent, with the study approved by the ethics committee of the St. George Hospital. All specimens were from St. George and Sutherland Hospitals (Sydney, Australia). Aliquots of biopsies or surgical specimens (n = 3) from skin, stomach, colon, bladder, breast, uterus, and lung were collected within 1 hr of surgery. Tissues were snap-frozen, cut into small pieces (3–5 mm), and placed in Tissue-Tek OCT embedding compound (Sakura Finetechnical; Tokyo, Japan). Cryostat sections (5-μm) were cut in a Reichert-Jung cryostat (Cambridge Instruments; Nussloch, Germany), mounted on gelatin-coated glass slides, and air-dried for 12 hr at room temperature (RT). One slide was immediately stained with hematoxylin and eosin. The remaining slides were stored at −80C until use.
Mononuclear cells from peripheral blood (PB), UCB, and bone marrow (BM) were obtained with informed consent and collected in heparinized tubes. Cells were diluted 1:1 in DPBS and layered on top of an equal volume of Ficoll. After centrifugation at 400 × g for 30 min at 4C, mononuclear cells from the interface were recovered and washed in DPBS. Slides were prepared from each cell preparation by cytocentrifugation. To assess the expression of 2C7 antigen in vitro, UCB cells were cultured in 50% (v/v) of conditioned medium from HBM-M cells with rhKL at 50 ng/ml to obtain mature MCs. The same cell preparation was cultured in minimum essential medium alpha (α-MEM) supplemented with 10% fetal calf serum (FCS) as a control. Cells assessed for immunostaining for 2C7, tryptase, and Fc∊RI at different time points were cytocentrifuged for staining.
Reagents
Mouse anti-Fc∊RIα IgG1 MAb was kindly provided by Dr. P. Kinet, (Department of Pathology, Harvard Medical School; Boston, MA). Mouse anti-tryptase, alkaline phosphatase-conjugated mouse anti-tryptase, mouse anti-chymase, and biotinylated mouse anti-chymase MAbs were from Chemicon International (Temecula, CA). Normal rabbit serum, rabbit anti-mouse IgG, alkaline phosphatase and anti-alkaline phosphatase complex (APAAP), biotinylated rabbit anti-mouse IgM and peroxidase-conjugated streptavidin were from Dakopatts (Glostrup, Denmark). Anti c-kit rabbit IgG polyclonal antibody was from Collaborative Biomedical Products (Bedford, MA). Sheep anti-rabbit Ig-conjugated horseradish peroxidase (HRP) and sheep anti-mouse Ig conjugated with fluorescein isothiocyanate (FITC) were purchased from Silenus (Hawthorn, Australia). Biotinylated rabbit anti-mouse IgG1 was from Zymed Laboratories (San Francisco, CA). The following antibodies were used in double-staining experiments with 2C7 to assess whether the antigens detected were present on 2C7+ cells: CD1a, CD13, CD14, CD19, CD23, CD34, CD61, and CD68 (Dako; Glostrup, Denmark); CD3, CD4, CD8 (Becton Dickinson; San Jose, CA). Bsp-1, a mouse IgM specific for human basophils, was used as an isotype control for 2C7 (kindly provided by Dr. M. Bodger).
Screening Hybridoma Lines
The initial screening of hybridoma supernatants was carried out using an indirect immunofluorescence approach with anti-mouse Ig-FITC. Cytospun cultured MCs or human skin sections were fixed in acetone for 10 min and incubated with the hybridoma supernatant at RT for 30 min. After washing, the samples were incubated with sheep anti-mouse Ig conjugated with FITC at RT for 30 min, and then were counterstained with Evan's blue. Cells recognized by the hybridoma supernatants were identified by fluorescence.
Double Immunoenzymatic Staining of Peripheral Blood Cells (PBCs) with Anti-leukocyte Antibodies and 2C7
Slides with cytospun cells were air-dried and fixed in acetone for 10 min at RT. After fixation, slides were incubated with 0.3% H2O2 in methanol for 10 min and normal rabbit serum diluted 1:10 with Tris-buffered saline (TBS) for 10 min to reduce nonspecific background staining. The treated slides were incubated with mouse anti-human CD antibodies at RT for 1 hr and then incubated with peroxidase-conjugated rabbit anti-mouse Ig (6 μg/ml) for 1 hr at RT. Slides were then developed with a freshly prepared DAB substrate solution [10 mg of 3,3'-diaminobenzidene tetrahydrochloride in 10 ml of Tris buffer (pH 7.6) containing 0.03% H2O2]. The treated slides were incubated with MAb 2C7 IgM (4.5 μg/ml) or control MAb IgM (4.5 μg/ml), followed by biotinylated rabbit anti-mouse IgM and alkaline phosphatase-conjugated streptavidin (5.0 μg/ml) for 1 hr at RT. The slides were further stained by addition of a freshly prepared alkaline phosphatase substrate containing 0.2 mg/ml of naphthol AS-MX phosphate with 0.1 mg/ml Fast Red TR and levamisole in 0.1 M Tris-HCl (pH 8.2) for 20 min. CD+ cells appeared brown and 2C7+ cells appeared pink, whereas double-staining cells appeared both colors.
Double Immunoenzymatic Staining of Human Tissues with an Anti-Fc∊RIα Antibody and Either Tryptase, Chymase, or 2C7 Antibody
For double immunoenzymatic staining of tissues with Fc∊RIα and either tryptase, chymase or 2C7 MAbs, slides containing tissue sections were fixed in acetone for 10 min at RT and treated with 0.3% H2O2 in methanol and normal rabbit serum for 10 min, and then incubated with anti-Fc∊RIα (1.4 μg/ml) overnight at 4C.
For Fc∊RIα and tryptase double immunostaining, tissue sections were sequentially incubated with biotinylated rabbit anti-mouse IgG (1.0 μg/ml) for 1 hr, with peroxidase-conjugated streptavidin (2.7 μg/ml) for 1 hr, and then with a freshly prepared DAB substrate solution [10 mg 3,3'-diaminobenzidene tetrahydrochloride in 10 ml of Tris buffer (pH 7.6) containing 0.03% H2O2]. The treated slides were then incubated with alkaline phosphatase-conjugated murine anti-tryptase MAb (2 μg/ml) for 3 hr at 37C and further stained by the addition of a freshly prepared alkaline phosphatase substrate included 0.2 mg/ml of naphthol AS-MX phosphate containing 0.1 mg/ml Fast Red TR and levamisole in 0.1 M Tris-HCl (pH 8.2) for 20 min. Fc∊RIα+ cells appeared brown, tryptase+ cells appeared pink, and double staining cells appeared both colors.
For Fc∊RIα and chymase or 2C7 double immunostaining, tissue sections were incubated with rabbit anti-mouse IgG (1:100 dilution) and APAAP complex (1:100 dilution) for 1 hr. Slides were then developed for 20 min in alkaline phosphatase substrate solution. Slides containing Fc∊RIα-immunostained sections were then incubated with either biotinylated murine anti-chymase MAb (0.34 μg/ml) for 1 hr at 37C or with 2C7 MAb IgM (4.5 μg/ml) followed by rabbit anti-mouse IgM conjugated with biotin and peroxidase-conjugated streptavidin and then stained with DAB substrate. Fc∊RIα+ cells appeared pink, chymase+ or 2C7+ cells appeared brown, and double-staining cells appeared both colors. Serial sections fixed in acetone for Fc∊RI antigen or Carnoy's fixative for tryptase or chymase antigen were also performed to confirm the results of double immunochemical staining.
Double Immunoenzymatic Staining of Human Tissues for Tryptase or Chymase and 2C7 or c-kit
For immunohistochemical identification of tryptase+ or chymase+ cells, tissue sections were fixed in Carnoy's fixative for 20 min and then incubated with anti-tryptase (0.5 μg/ml) or anti-chymase (0.5 μg/ml) MAb according to the conditions outlined above for the alkaline phosphatase anti-alkaline phosphatase staining with Fc∊RIα. For double immunohistochemistry carried out on the same slide, slides were then incubated with 2C7 IgM, followed by rabbit anti-mouse IgM conjugated with biotin and peroxidase-conjugated streptavidin or anti c-kit IgG (2.5 μg/ml), followed by anti-rabbit IgG conjugated with HRP. The slides were then treated with a freshly prepared DAB substrate solution. Tryptase+ or chymase+ cells stained pink in this reaction, whereas 2C7+ or c-kit+ cells stained brown. Cells that contained both antigens appeared as both colors. In vitro-derived human MCs from UCB cells and HL-60 cells were used as positive and negative controls, respectively. For the anti-chymase and anti-tryptase antibodies, a further control was performed to confirm the results obtained with the double-immunostaining experiments by reversing the order of staining of the tissue sections for tryptase and chymase before staining with anti Fc∊RIα, 2C7, and c-kit. Serial sections were stained also with either anti-tryptase, chymase, Fc∊RIα, 2C7, or c-kit singly.
Double Immunofluorescent Staining of Skin-and UCB-derived MCs with CD13 and 2C7 Using Confocal Microscopy
Cytospins were double-labeled with antibodies to CD13-FITC (2 μg/ml), tryptase (0.5 μg/ml), and 2C7 (4.5 μg/ ml). Antibodies were diluted in PBS/2% normal goat serum (NGS). Sections of skin were fixed in Carnoy's for 20 min.
Cytospins of cord blood cells (cultured for 21 days in HBM-M conditioned medium and KL) were prepared and fixed in either Carnoy's fixative (for tryptase) or acetone (2C7 and CD13) for 15 min. Slides with cytospins were blocked with a mixture of 10% NGS for 20 min at RT.
The cytospins/sections were incubated with the primary antibody (either 2C7-anti-tryptase or CD13-FITC) for 24 hr at 4C, then rinsed in PBS on a shaking table for 20 min. For CD13-FITC no further labeling was required. However, for 2C7 or tryptase double labeling, cytospins/sections were further incubated with biotinylated anti-mouse Ig secondary antibody (Amersham Pharmacia Biotech, Uppsala, Sweden; 1:100) for 1 hr at RT and rinsed in PBS for 20 min, followed by incubation in FluoroLink Cy3 labeled streptavidin (Zymed; 1:250) for 20 min. After rinsing in PBS for a further 20 min, cytospins/sections were then incubated with secondary antibody for 24 hr at 4C, then rinsed in PBS on a shaking table for 20 min. Cytospins/sections were then labeled with Alexa 488-conjugated anti-mouse IgG (Molecular Probes, Eugene, OR; 1:1000) for 1 hr at RT. After rinsing in PBS for a further 20 min, cytospins were mounted with glycerol and coverslipped.
All specimens double labeled with immunofluorescence were viewed using a Leica (Nussloch, Germany) upright confocal microscope incorporating NTCS Leica software. The confocal microscope consists of an argon-krypton laser with a dual filter that allows both 488-nm and 568-nm wavelengths of light to pass onto the specimens, thus simultaneously exciting both FITC/Alexa 488 and Cy3 labels.
Quantitation of Immunostained Cells
Tissue sections were mounted with glycerol-gelatin solution (Sigma) and then examined by two independent observers using light microscopy (Leica) at X100 magnification under oil. Five to eight adjacent random fields were evaluated for each section. For skin sections, mean cell numbers were calculated for the epidermis, the upper dermis, and the deeper dermis. In the colon and stomach tissues, mean cell numbers were calculated in the mucosal, lamina propria, and submucosal layers. The endometrium and myometrium regions of the uterus and mucosal and muscle layers of the bladder were examined. In the breast and lung tissues, mean cell numbers were calculated in the parenchyma.
Western Blotting and Immunoprecipitation
The 2C7 MAb failed to bind to membrane extracts of in vitro-derived human MCs in Western blotting and in immunoprecipitation experiments (results not shown).
Results
Identification of Hybridomas That Produce Antibodies Recognizing Human MCs
In the initial screening, ∼30% of the wells (102/360 wells) contained proliferating hybridomas. However, only ∼10% of the wells possessed antibodies that bound to human MCs, which were used in the immunization protocol. The 2C7 clone was chosen for further study even though it was an IgM based on its apparent specificity for cultured MCs and human skin MCs. This hybridoma supernatant bound to approximately 1.5% (mean, n = 3) of mononuclear cells in human BM, 1.4 ± 1.3% (mean, n = 5) of PB mononuclear cells, and 0.5% (mean, n = 5) of UCB cells. The 2C7+ cells had single-lobed nuclei and lacked the characteristic MC secretory granules.
Percentage of total cells in cultures of umbilical cord blood with rhKL plus 50% (v/v) HBM-M-CM: ▪ 2C7+; ▴ Fc∊RI+, ▾ tryptase+; control medium • 2C7+.
Double immunostaining of (
Kinetic experiments revealed that the 2C7 antigen was expressed early during the in vitro differentiation of hematopoietic progenitors into MCs. Using purified MAb at 4.5 μg/ml at day 0, 0.5 ±0.1% (mean ± SD, n = 5) of cells in the culture were weakly 2C7+. The percentage and intensity of 2C7+ cells in the culture increased with 59.18 ± 22.54% and 87.88 ± 8.53% (n = 5) of cells strongly positive at days 21 and 35, respectively (Figure 1). In contrast, few tryptase+ cells were detected at day 0, and only 13.0 ± 13.8% (n = 5) of the cells expressed this protease at day 21. As found earlier, Fc∊RI was expressed after these MCs expressed the tryptase(s). When the double-staining technique was used to evaluate the expression of 2C7 and the other MC markers, all Fc∊RIα+ cells were 2C7+, as were all tryptase+ or chymase+ cells. Because the 2C7 antigen was expressed before the tryptase epitope, 25% of the 2C7+ cells did not express tryptase at day 35 of culture. The cellular localization of 2C7 and its co-localization with tryptase and CD13 were examined using a combination of immunofluorescent staining and confocal laser microscopy on cytospins of cord blood-derived MCs. 2C7 appeared predominantly localized intracellularly (Figures 2 and 3). The co-localization of 2C7 and tryptase was further confirmed by confocal microscopy (Figures 3C-3E). To determine the association of 2C7 immunoreactivity with CD13 (a marker for MCPs), we carried out a double immunofluorescent staining procedure with these cells. 2C7 staining was intracellular and co-localized with CD13, which had a membrane staining pattern. Cells that were mononuclear had a relatively large cytoplasm (Figure 3B).
Double Immunoenzymatic Staining of Peripheral Blood Cells (PBCs) with CD Leukocyte Antigens and 2C7
Mononuclear cells isolated from peripheral blood of normal individuals contained 1.4 ± 1.3 2C7+ (n = 5); 73 ± 9.5% of these 2C7+ cells stained with c-kit (n = 3), 21 ± 5.7% with CD34, and 49.3 ± 3.0% (n = 5) with CD13, but were unreactive with antibodies to lymphocyte (CD3, CD4, CD8, CD19, and CD23)-, monocyte (CD14)-, macrophage (CD68)-, basophil (Bsp-1)-, Langerhans cell (CD1a)-, and platelet (CD61)-specific antigens.
Confocal laser scanning microscopic image of immunostained human cord blood-derived MCs (
Double Immunoenzymatic Staining of Human Tissues Using Tryptase, Chymase, and 2C7 Antibodies
Double immunoenzymatic staining was performed on cryostat tissue sections to monitor tryptase, chymase, and 2C7 expression at the protein level. 2C7 immunoreactivity was granular and had a cytoplasmic staining pattern. The intensity of staining in 2C7+ cells varied from weak to strong. In the skin, 2C7+ cells were preferentially located in the dermal layer (Figures 2B and 2D). The majority of tryptase+/2C7+ (Figure 2B) or chymase+/2C7+ cells (Figure 2D) were concentrated in the papillary layer of the dermis, whereas most of the 2C7+ cells, which failed to express tryptase or chymase, were found in the reticular layer of the dermis. There was no reactivity for 2C7 in the epidermis.
In the colon and stomach tissue sections, 2C7+ cells preferentially resided in the mucosa, lamina propria, and submucosa. Only the mucosal layer was present in stomach samples, as these were endoscopic biopsies. The numbers of 2C7+ cells in the submucosal region of the colon were higher than in the mucosa. 2C7+ cells were located in both the endometrial and the myometrial layer of the uterus.
The intensity of 2C7+ staining in the myometrium was stronger than that in the endometrium. 2C7+ cells were mainly located in the mucosal layer of the bladder. In the breast and lung samples, 2C7+ cells were distributed mainly in the parenchyma.
Percentage of positive cells in human tissues stained for (
Almost all tryptase+ and/or chymase+ cells in all tissues expressed the 2C7 epitope. The percent of tryptase+ and/or chymase+ cells that also expressed 2C7 ranged from 93% in the uterus to 96% in the skin (Figure 2B). Results are summarized in Figures 4A and 4B.
Double Immunostaining of Human Tissues with Fc∊RI and Tryptase, Chymase, or 2C7 Antibodies
The majority of the tryptase+ and chymase+ cells in all examined tissues also expressed Fc∊RIα (Figures 5A and 5B). However, a significant difference was observed between double immunostaining with Fc∊RIα and tryptase or 2C7. Approximately 20–29% of the Fc∊RIα+ cells in all tissues studied failed to stain with tryptase (Figure 6A). In contrast, the results obtained with the 2C7 demonstrated that more than 91% of Fc∊RIα+ cells also stained for 2C7 (Figure 6C). In the case of chymase and Fc∊RIα staining, 17-40% of the Fc∊RIα+ cells in skin, breast, stomach, bladder, colon, and uterus failed to stain with chymase, whereas 84% of Fc∊RIα+ cells in lung failed to express chymase (Figure 6B). The latter findings are consistent with those obtained by others, showing that very few MCs in normal lung express chymase (Figure 6B). Figures 2B and 2D show double immunoenzymatic staining for chymase and Fc∊RI in human skin. Cells were identified that expressed Fc∊RI without chymase (open arrow) or Fc∊RIα with chymase (closed arrow).
Percentage of positive cells in human tissues stained for (
Double Immunoenzymatic Staining for c-kit and Tryptase or Chymase Antibodies
Although strong immunoreactivity was demonstrated in all tissues for c-kit, there was a substantial difference in the pattern of staining among tissues. The majority of tryptase+ cells in skin, uterus, breast, bladder, and lung also expressed c-kit (Figure 7A). However, 20–30% of tryptase+ cells in colon and stomach failed to stain with the c-kit antibody (Figures 7A and 2C). A similar pattern of c-kit reactivity was also seen with chymase+ cells in colon, breast, stomach, and bladder (Figure 7B). A significant number of c-kit+ cells in colon, uterus, breast, and stomach failed to stain for tryptase (Figure 7C). The pattern was similar for chymase except for lung, which has few chymase+ cells (Figure 7D).
Double immunoenzymatic staining of 2C7 and c-kit showed that 59.4 ± 9.7% of 2C7+ cells expressed c-kit in the skin and 73 ± 1.8% in the uterus, whereas the numbers were approximately 13–45% in other tissues (data not shown). These data suggested that there is heterogeneity in distribution and staining properties of MCs in human tissues.
Percentage of positive cells in human tissues stained for (
Discussion
Using in vitro-derived human MCs as an immunogen, we have derived an MAb, designated 2C7. 2C7 was initially selected for its lack of reactivity on other hematopoietic cell lines and very low percentage of positive cells observed in peripheral blood, cord blood, and bone marrow. Co-localization of MAb 2C7 and antibodies to CD34+ in normal and c-kit+ cells in asthmatic patients first suggested that 2C7 recognized an antigen that was expressed early during MC development. The percentage of c-kit+ cells in peripheral blood in asthmatic patients is consistent with the discovery of the presence of predominately immature MCs in the circulation of patients with allergic disorders (Li et al. 1998). For quite some time very little was known about the MCP, but in recent years there have been a number of reports on the nature of these progenitors. Not only are they positive for CD34, they also express the receptors for SCF, c-kit, and the low-affinity IgG receptor FcγRII (CD32) (Kirshenbaum et al. 1991; Rottem et al. 1994). Approximately 21% and 49% of 2C7+ cells in the peripheral blood of patients with asthma stained for CD34 and CD13, respectively. These cells might represent an immature human MC that is positive for 2C7, c-kit, CD13, and CD34.
The observation that 2C7+ cells express c-kit (CD117) is in line with the assumption that c-kit is expressed at a continuously high level throughout mastopoiesis (Valent et al. 1992). However, in this study we identified significant numbers of c-kit− MCs in a variety of human tissues. Another interesting observation in this study was that 2C7 recognizes not only mature MCs, but also a subset of CD13+ cells in both PBCs from asthma patients and in UCB cell cultures. Most available data point to the expression of CD13 on immature rather than mature human MCs. These results suggest that, during MC differentiation in vitro, the immature MC retains the 2C7+/c-kit+/ CD34+/CD13+ phenotype.
The 2C7 MAb stained MCs from all tissues in addition to in vitro-derived MCs obtained from human UCB and BM progenitors (Li et al. 1995,1996). Analysis of the time course of 2C7 antigen expression demonstrated that the 2C7 antigen in this culture system appeared earlier than either Fc∊RI or the MC protease tryptase. Double immunostaining confirmed that tryptase+ and chymase+ cells in tissue sections from skin, colon, uterus, breast, stomach, bladder, and lung reacted with the 2C7 MAb. However, a significant number of 2C7+ cells failed to stain for tryptase or chymase. There was no staining with 2C7 MAb in the epidermis, and the 2C7+/tryptase−, or chymase− cells were predominantly distributed in the reticular layer of the dermis, submucosa of the colon, and endometrium of the uterus. Fc∊RI plays a pivotal role in the pathophysiology of allergic and inflammatory disorders and is expressed on a number of cells, including certain dermal dendritic cells (DCs) predominantly located around the dermal microvasculature and related DCs as well as in antigen-presenting cells (APCs) in the lung (Wang et al. 1992; Gounni et al. 1994; Maurer et al. 1994,1996; Humbert et al. 1996; Tunon-De-Lara et al. 1996). The double immunoenzymatic experiments using Fc∊RIα and tryptase, chymase, or 2C7 indicated there was a significant population of Fc∊RIα+/tryptase−, or chymase− cells in all organs studied. Whereas there was Fc∊RIα staining in the epidermis, no staining could be detected for 2C7, tryptase, or chymase. The Fc∊RIα staining in the epidermis is most likely due to Langerhans cells, which are known to express the α- and γ-chains of the Fc∊RI hetrotetrameric complex (Wang et al. 1992). The nature of the Fc∊RIα+/tryptase−, or chymase− cells observed in tissue sections from the various organs is unclear. The 2C7 antibody seems to recognize MCs in the tissues in the same stages of maturation as the MAb AA4 that is specific for rodent MCs (Jamur et al. 2001). Alternatively, some of these cells may be DCs, although there was no 2C7 staining in the epidermis where a rich source of Langerhans cells is present that also express Fc∊RIα. Another possibility is that these cells are degranulated MCs, although there was no detectable extracellular immunoreactivity for tryptase or chymase. There were no monocytes or eosinophils in the tissue preparations when these were examined using hematoxylin and eosin (results not shown).
Percentage of positive cells in human tissues stained for tryptase and c-kit, chymase and c-kit in double immunoenzymatic reactions. (
Human MCs circulate as agranular progenitors that develop distinctive granules after leaving the blood vascular space to reside in the connective tissue-rich stromal compartments of tissues and organs. Recent data suggest that KL induces differentiation of human MCs from their progenitor cells in long-term suspension culture and, in addition, is a specific activator of mature lung, skin, and uterine MCs (Columbo et al. 1992; Sperr et al. 1993). The effect of KL is mediated via the c-kit receptor. C-kit has also been identified in glandular epithelial cells of breast, dermal sweat, and esophageal glands (Lammie et al. 1994).
It has been suggested that c-kit+/tryptase−/Fc∊RIα− cells in nasal mucosal tissue obtained from allergic individuals were a candidate for MCPs (Kawabori et al. 1997). Information on expression of c-kit in normal human tissues is limited. In the present study, c-kit+/ tryptase− or c-kit+/chymase− cells were predominantly distributed in colon, uterus, breast, stomach, and bladder, whereas there were fewer in skin and lung. If we assume that a significant number of the c-kit+/tryptase− or c-kit+/chymase− cells are MCPs, it appears that the skin and lung have low numbers of these cells. C-kit has been shown to be expressed on the cell surface of human tissue MCs obtained by enzymatic digestion from lung, skin, uterus, and gut (Guo et al. 1992; Valent and Bettelheim 1992). However, in the present study we have also demonstrated c-kit− cells that express tryptase and/or chymase in tissue sections from colon, breast, and stomach. It was estimated that only 70–90% of tryptase+ cells in samples from stomach, colon, and breast tissues were positive for c-kit. The reason for this is not clear because c-kit expression is detected early during MC development and the MC serine proteases are normally expressed after c-kit expression. It could well be that there is some factor or factors unique to these tissue sites that may downregulate c-kit expression. However, it also appears that c-kit may not be a good marker in the tissues for MCs because a significant number of the tryptase+ and chymase+ cells in colon, stomach, lung, uterus, and bladder are c-kit−. In addition, there appear to be considerable variations in the amount of c-kit antigen that is expressed by MCs at all anatomic locations, with a trend for larger, more granulated cells to express less antigen. Downregulation of c-kit expression in in vitro-derived human MCs has been induced with IL-4 (Sillaber et al. 1991). Interestingly the addition of IL-4 in this culture system induced an increase in chymase expression, which is considered to be a maturation marker of MCs (Toru et al. 1998).
In conclusion, our results show that a novel MAb, 2C7, binds to MCs from a variety of human tissues. 2C7 identifies a population Fc∊RIα+/tryptase−/chymase− cells. MAb 2C7 can be used to identify immature and mature MCs in peripheral blood of patients with asthma and allergic disorders, suggesting that this antibody is also an invaluable tool for precise determination of MC lineage in these patients. Studies are under way to determine the clinical significance of 2C7+ cells and their quantitation in allergic and inflammatory diseases.
Footnotes
Acknowledgements
Supported by a grant from the National Health and Medical Research Council of Australia and by the Clive and Vera Ramaciotti Foundation.
We thank J. Novak and A. de Nangle for excellent secretarial assistance, Dr P. Kinet for providing the MAb to Fc∊RIα, and Dr M. Carton, Dr A. Ford, Dr I. Simpson, Dr D. Knight, and Dr M. Booth and the staff in theaters at St. George Hospital for providing umbilical cord bloods. We also thank the staff in Department of Pathology for excellent technical assistance.