Abstract
The presence and density of tryptase-positive/chymase-positive mast cells (MCs) (MCTC), chymase-positive/tryptase-negative MCs (MCC), and tryptase-positive/chymase-negative MCs (MCT) in lesional skin from cats with eosinophilic conditions were investigated. Skin biopsy specimens from eight cats with eosinophilic plaque (three cats), eosinophilic granuloma (two cats), and eosinophilic dermatitis (three cats) were studied. Toluidine blue staining and a double-enzyme-immunohistochemical staining technique were performed to determine MC density and MC subtypes, respectively. MC density varied from 170.3 to 503 cells/mm2 (mean value of 314.9 cells/mm2). In the superficial dermis, 5.9% of the MC belonged to the MCT, 12.8% to the MCC, and 81.2% to the MCTC subtype. In the deep dermis, 12.8% belonged to the MCT, 12.8% to the MCC, and 73.8% to the MCTC subtype. It is the first time that MCC have been identified. The double-labeling procedure proved to be a reliable tool for identifying simultaneously the presence of MC subtypes in feline skin.
Cutaneous mast cells (MCs) play an important role in the homeostasis of normal tissues and in the induction of inflammation after stimulation. These cells produce and store numerous mediators that are important in inflammation. Among these, two serine proteases, namely, chymase and tryptase, modulate and direct the inflammatory response. 11 In humans, dogs, and bovines, three MC subtypes have been recognized in healthy skin. 11 , 12 In the cat, tryptase and chymase have been demonstrated in skin MCs. 1 In dogs and humans, the major MC subtype is tryptase-positive/chymase-positive MC (MCTC), followed by tryptase-positive/chymase-negative MC (MCT) and chymase-positive/tryptase-negative MC (MCC). In bovines, only MCT can be detected in the skin. Determination of MC subtypes and numbers has been performed in dogs with skin disease 13 but not in diseased cats. In dogs, a double-staining technique for MC subtypes on paraffin-embedded tissue samples has been described, whereas in cats, only the staining of serial cryosections for one protease at a time has been reported. The aim of this study was to use a chymase-tryptase double-staining technique to determine the percentage distribution of MCTC, MCT, and MCC in lesional skin from cats affected by eosinophilic plaque and eosinophilic granuloma.
Eight European shorthaired cats with eosinophilic conditions were included in the study. Ivermectin had been administered subcutaneously (0.3 mg/kg, Ivomec®, Merck, Italy) and fipronil had been applied to the interscapular skin (Frontline spot on®, Mérial, Italy) 2 weeks before biopsy sampling to exclude parasitic infestation. Fungal cultures, performed with the MacKenzie brush technique on Dermatophyte test medium and Sabouraud culture medium (DIFCO, Detroit, MI), were negative in all animals. None of the cats had received short-acting steroids or antihistamines within the 2 weeks before skin sampling or had received injectable long-acting steroids within the 6 weeks before sampling. Biopsy samples of lesional skin were obtained under general anesthesia with an 8-mm biopsy punch. Samples were fixed in 4% buffered formalin solution (pH 7.4). After embedding in paraffin, 3-μm-thick sections were obtained and stained with hematoxylin and eosin and toluidine blue (TB) following standard procedures. MC density was evaluated on sections stained with TB by using a Quantimet 500 MC analyzer system (Leica, Cambridge, UK) as described previously. 10
The protease content of MCs was examined using an enzyme-histochemical reaction for the detection of chymase activity and an immunohistochemical staining method with a polyclonal antibody for the detection of tryptase. This double-labeling procedure was performed as follows. After deparaffinizing the sections in xylol and acetone for 15 minutes each, the slides were immersed in Tris-buffered saline (TBS), pH 7.4, and subsequently in deionized water. The enzyme cytochemical reaction for the detection of chymase was performed with a commercially available detection kit (Sigma kit 91-C, Sigma Chemical Co., St. Louis, MO) using Naphthol-AS-D-Chloroacetate as substrate. The only modification of the protocol provided by the company was the use of Fast Blue BB Base (Sigma) instead of Fast Red Violet LB Base (Sigma). No counterstaining was performed. The immunohistochemical staining for MC tryptase was performed at room temperature immediately after the enzyme cytochemical staining and a predigestion with 0.1% Protease XIV (Sigma) for 4 minutes at 37 C. Tryptase was detected with a polyclonal rabbit anti–human skin tryptase antibody at a dilution of 1:2,000. 4 After incubation for 30 minutes, a mouse anti–rabbit IgG (1:200, Dianova, Hamburg, Germany) was applied for 30 minutes. This step was followed by incubation with the alkaline phosphatase-anti-alkaline phosphatase (APAAP) complex (1:100, Dianova) for 30 minutes. The incubation with the mouse anti–rabbit IgG and the APAAP complex was performed twice. Between each step, the sections were washed thoroughly in TBS (pH 7.4). The antibody and the APAAP complex were diluted in RPMI buffer (pH 7.4). The alkaline phosphatase substrate was prepared by dissolving 20 mg Naphthol-AS-MX-Phosphat (Dianova) in 2 ml dimethylformamide and then adding 98 ml TBS (pH 8.2). Immediately before this preparation was used, Fast Red TR (100 mg) and levamisole (1 M, 2.408 mg) were added. Incubation with the substrate was performed for 30 minutes with continual agitation. Subsequently, sections were rinsed in tap water and mounted in Kaiser's glycerine gelatine. For each tissue specimen, negative controls were prepared by omitting the incubation with the primary antibody.
The percentage distribution of the MC subtypes, MCTC, MCT, and MCC, was assessed at a high-power field (magnification 400×). Images were captured on the monitor, and blue, red, and mixed-colored MCs were counted in 10 fields from the subepidermal dermis and 10 fields from the deep dermis for each sample. Only nucleated MCs were counted.
Descriptive statistic was calculated. The nonparametric Mann-Whitney
Histologically, three cases were classified as eosinophilic plaque and two as eosinophilic granuloma, and three cases, with only a moderate eosinophilic infiltrate, were diagnosed as eosinophilic dermatitis. These last three cases corresponded clinically to lesions of an eosinophilic plaque or an eosinophilic granuloma. However, the eosinophils were not as diffusely present in the interstitium as in the eosinophilic plaque, and no granuloma formation was seen. MC density assessed on TB-stained sections varied from 170.3 to 503.0 cells/mm2 (mean of 314.9 cells/mm2). MCT showed bright-red intracytoplasmic staining with a diffuse granular pattern; MCC showed azure to light-blue intracytoplasmic staining; and MCTC showed dark-purple to dark-blue intracytoplasmic fine granular content. In the subepidermal dermis, 5.9 ± 3.6% of the MC belonged to the MCT, 12.8 ± 11.5% to the MCC, and 81.2 ± 12.3% to the MCTC subtype. In the deep dermis, 12.8 ± 8.4% were MCT, 12.8 ± 7.4% were MCC, and 73.8 ± 7.1% were MCTC. No statistically significant differences in the distribution of MC subtypes were detected between the subepidermal dermis and the deep dermis. In the eosinophilic granuloma, a perivascular distribution of mostly MCT was observed in the deep dermis.
Three MC subtypes, with different biologic functions, have been identified in normal human and canine skin: MCT, MCC, and MCTC, with the last type being more numerous in the skin than the first and second types, respectively. 7 , 8 , 11 A number of studies support an important role of both MC proteases in inflammation. Tryptase has been demonstrated to play a critical role in the recruitment of both neutrophils and eosinophils. 5 A contribution to allergic inflammation has been demonstrated for chymase. 6 Furthermore, differences in the biochemical properties of the two proteases have been demonstrated. 8
In normal feline skin, MCT and MCC have been identified, but MCTC has not been described due to unavailability of double-labeling techniques in this species. 1
This is the first time that the MCTC subtype is reported in feline skin. The detection of this subtype was achieved by using a double-labeling procedure on paraffin-embedded biopsy specimens, which simultaneously allows the demonstration of the presence of MCT, MCC, and MCTC. This double-labeling technique has already been used on dogs and bovines, and a similar technique was used on human tissue. 4 , 12 It has now been shown that this technique is also a reliable and accurate method for marking MC subtypes in the feline species.
The main MC subtype in both subepidermal and deep dermis is the MCTC. This subtype has also been described to be predominant in canine 12 and human skin. 8 It is difficult to compare our results with those obtained in a previous study performed on normal feline skin 1 because, as already mentioned above, in this study, staining for chymase and tryptase was performed separately on serial sections. However, in this previous study, the MCC/MCT ratio was approximately 90% in cryosections of normal feline skin. Even though it was not possible to assess the true number of MCT, MCC, and MCTC, these data suggested that MCs containing both proteases represent the major subtype in normal feline skin. 1 However, this study could not determine the presence of MCC, which were clearly identified by our study.
This is the first report that documents the distribution of MC subtypes, according to protease content, in lesional skin of cats affected by eosinophilic conditions. As expected, MC numbers were increased when compared with those in the skin of healthy cats, 3 , 10 but the distribution of MC subtypes was not very different from that described in the skin of healthy cats 1 and of healthy 9 and atopic 13 dogs. MC protease content seems therefore not to be influenced by disease and localization in the dermis. However, the finding of the perivascular distribution of mostly MCT in one case of eosinophilic granuloma was very interesting. Preliminary data suggest that activated MC may selectively degranulate different granule contents in a piecemeal fashion, 2 thus changing labeling properties in lesional skin. Different etiopathogenic mechanisms are thought to be implicated in feline eosinophilic conditions, and the role of different MC subtypes in deep perivascular infiltrates (eosinophilic plaque and granuloma) compared with the mostly superficial perivascular infiltrate (eosinophilic dermatitis) deserves further investigation.