Abstract
Background: Canine cutaneous histiocytoma (CCH) is an epidermotropic tumour of Langerhans cells, most frequent in young dogs, which undergoes spontaneous regression. Materials and Methods: E-cadherin immunoexpression was analysed in ninety-three CCH, categorized according to Cockerell and Slauson (1976) criteria into four histological groups, representing different stages of tumour regression. Results: All tumours expressed membranous E-cadherin both in keratinocytes, and in tumoural cells and no differences were noted among the histological groups analysed. However, the intensity of the E-cadherin immunolabeling decreased in the presence of lymphoid infiltration and varied significantly among the groups considered (p<0.0001). Conclusion: This study strongly suggests a down-regulation of E-cadherin expression in CCH pathogenesis and progression. The loss of E-cadherin expression might represent an activation/ maturation process of the tumoural cells constituting a switch for CCH regression.
Canine histiocytic proliferative disorders include lesions such as canine cutaneous histiocytoma (CCH), a common and usually solitary, benign neoplasm that appears as a rapidly growing, alopecic, erythematous, dome-shaped nodule, often with ulceration (1-4). In a five-year retrospective study, performed in the Laboratory of Histopathology, University of Trás-os-Montes and Alto Douro, CCH represented 29.4% of all dog skin and soft tissue mesenchymal tumours (5). The disease affects mostly young dogs, between 6 months and 3 years of age; the incidence falls strikingly in older animals (1-4, 6).
Immunophenotype and ultrastructural studies have confirmed that CCH is a proliferation of intraepidermal dendritic antigen-presenting cells, also called Langerhans cells, that express CD1a, CD1b, CD1c, major histocompatibility complex (MHC) class II and CD11c (7-10).
Although the rapid growth and high mitotic index suggest malignancy, CCH is a benign tumour in all aspects: it does not metastasise and spontaneous regression (within a few weeks or months) is the natural course of the disease (11). This regression seems to be associated with an initial infiltration of CD4+ T-cells, followed by an increased expression of T-helper subset 1 (Th1) cytokines and recruitment of antitumour effector cells (12) and appears to be related to a change of CCH cell phenotype, with an increased membranous expression of MHC class II antigen molecules (13).
E-cadherin is a transmembranous glycoprotein, a member of a family of calcium-dependent homophilic cell-cell adhesion molecules that bind to cytoskeleton proteins through catenins (14, 15). Down-regulation of E-cadherin expression often results in tissue disorder, disturbed cellular differentiation, increased invasiveness of tumour cells and ultimately malignancy in animal and human epithelial tumours (16-20).
Immunoexpression of E-cadherin in CCH was described very recently (2008) by Baines (21) and Pires (22). Nevertheless, its implication in histiocytoma regression has not yet been clarified.
The purpose of this study was to evaluate E-cadherin expression in different stages of tumour regression, based on the lymphoid inflammatory infiltrate degree, in order to clarify if there is a link between E-cadherin expression and CCH regression.
Materials and Methods
Tissue processing and tumour classification. Ninety-three cases of CCH and ten normal canine skin samples, obtained from the UTAD Histopathology Laboratory archives were included in this study. Each sample was re-examined by two independent pathologists (IP and AA) in order to confirm the diagnosis, according to the World Health Organization International Histological Classification of Tumours of Domestic Animals criteria (23). Histiocytomas were also grouped according to the degree and pattern of intratumoural lymphocytic infiltrate into four groups, as described elsewhere (24), representing diverse stages of tumour regression (Table I).
Classification of CCH into four histological groups according to the degree and distribution of the lymphocytic infiltration (adapted from (24).
Immunohistochemistry. For immunohistochemical studies, 3-μm sections were cut from each specimen and mounted on silane-coated slides. The detection of E-cadherin was carried out by the streptavidin-biotin-peroxidase complex method, with a commercial detection system (Ultra Vision Detection System; Lab Vision Corporation, Fremont, USA) following the manufacturer's instructions. Antigen retrieval was carried out by micro-wave treatment in a 0.05% detergent solution (Extran; Merck, Frankfurt, Germany) at pH 6, with microwave irradiation (750 W) for 3×5 minutes. A monoclonal mouse anti-E-cadherin antibody (4A2C7; Zymed Laboratory, San Francisco, USA) which in previous studies had shown reactivity with canine species (22, 25) was used as primary antibody. The incubation was performed overnight, at a dilution of 1:150 in phosphate-buffered saline (PBS) solution. Immunoreaction was visualized by incubation with 13,3-diaminobenzidinetetrahydrochloride (DAB) at 0.05% with 0.01% H2O2 as the final substrate for 5 minutes. After a final washing in distilled water, the sections were counterstained with haematoxylin, dehydrated, cleared and mounted. The primary antibody was replaced with PBS for negative controls and positive controls consisted of sections from normal canine skin and from skin adjacent to neoplastic tissue.
Quantification of immunolabelling. Positivity was indicated by the presence of distinct brown labelling. Immunoreactivity was evaluated blind by two pathologists using a semiquantitative method, according to the percentage of immunoreactive cells scored as: 0, negative; 1, <25% positive cells; 2, 25-50% positive cells; 3, >50% positive cells, with a membranous (cell-cell boundaries) pattern of cellular distribution. The cytoplasmatic labelling was also reported. The intensity of reaction was also recorded as negative (-), weak (+), moderate (++) or strong (+++), as compared with the internal control (the epidermis).
Statistical analysis. Associations between histological groups and E-cadherin expression were investigated using the χ2 test, with the SPSS system (version 12.0; SPSS Inc., Chicago, USA). A value of p<0.05 was considered significant.
Results
E-cadherin expression in normal skin. In all cases (n=10), the immunoreactivity for E-cadherin was membranous, diffuse and strong in cell-cell unions in the keratinocytes and in epithelial cells of the adnexal structures.
E-cadherin expression in the skin adjacent to the tumour. The immunoreactivity for E-cadherin in apparently normal skin adjacent to the tumour was evaluated in all the lesions (n=93). The immunolabelling pattern was similar to that described for the normal skin.
Semiquantitative evaluation (score) of E-cadherin expression in CCH. All the lesions analysed expressed E-cadherin in the tumour cell membrane. The majority of the cases (n=85; 91.4%) expressed E-cadherin in more than 50% of the tumoural cells (Figure 1). The percentage of E-cadherin-positive cells was similar among the different histological groups (p=0.16; Table II).
Intensity of E-cadherin expression in CCH. All the lesions expressed E-cadherin in the tumour cells at a lower intensity than in the keratinocytes. The intensity of reaction was not uniform in the tumoural cells, with a decrease of immunoexpression in the areas of the lymphocytic infiltrate. The intensity of reaction was significantly different among the different histological groups considered (p<0.0001).
In group I and group II, the cells of the epidermal surface and the cells of the centre of the lesion showed a moderate (++) reactivity to E-cadherin. However, the intensity decreased with depth, being weak or even absent in the bottom periphery of the tumour, except in one case in which the peripheral inflammatory infiltrate was very scarce.
In tumours of group III, where the lymphocytic infiltrate was more intense and diffuse, the intensity of reaction was predominantly weak (+), both in cells of the surface and the bottom of tumours.
In group IV, the tumour cells showed a weak intensity of reaction, except in one case that presented moderate intensity (Table II). In four cases, besides the membranous reaction, a weak and diffuse cytoplasmatic positivity was observed.
Discussion
The regression phenomena, so typical of this disease, makes CCH an attractive system for analysis of Langerhans cell histiocytosis behaviour and it could be regarded as an unique model to understand the pathogeny of the enigmatic disease of human Langerhans cell histiocytosis (7, 26).
Imunoexpression of E-cadherin in the four histological groups.
In the present study, ninety-three CCH were examined by immunohistochemistry for expression of E-cadherin. Although previous studies showed E-cadherin expression in CCH (21, 22), little is known about its role in tumour progression and regression.
All CCHs studied in this work expressed E-cadherin. However, unlike the results described by Baines et al. (21), in our study the immunolabelling was predominantly membranous both in tumour cells and in keratinocytes. In fact, E-cadherin is a transmembranous protein that constitutes the major adhesion molecule in adherens junctions and mediates the binding of Langerhans cells to keratinocytes (27-29). The membranous location of E-cadherin observed in our study is also supported by ultrastructural studies that revealed adherens junctions between the CCH tumour cells in the dermis, as we reported recently (22). This apparent contradiction between the results of Baines et al. (21) and ours could be related to the use of different primary antibodies that could recognize different domains of E-cadherin: cytoplasmic or extracellular, as described lately in human tumours (30, 31).
In the present study, the lesions were classified into four histological groups, representing diverse stages of tumour regression. No significant differences in the score for E-cadherin expression were found in the different groups. However, when we evaluated the E-cadherin immunolabelling intensity, the differences were significant. A decrease of E-cadherin intensity occurred with an increase of the lymphoid infiltrate. In contrast to epithelial tumours, in which a decrease or lack of E-cadherin expression is associated with invasion and metastasis (19, 32-34), in CCH this phenomena seems to be associated with tumour regression.
The work of Baines et al. (21) on E-cadherin expression in 37 CCHs does not report a variation in the E-cadherin intensity. To our knowledge, this is the first study where this characteristic is described. Interestingly, a similar reduction of E-cadherin occurs in normal Langerhans cells in the inflammatory response. The migration of Langerhans cells from the epidermis to the dermis and lymph nodes, induced by contact allergens and proinflammatory cytokines, is associated with a decrease of E-cadherin messenger ribonucleic acid (mRNA) and its synthesis. This decrease with consequent loss of connection between Langerhans cells and keratinocytes is one of the first events in the skin immune response leading to Langerhans cell activation, facilitating their emigration and the initiation of immune responses against antigens encountered in the epidermis (35, 36). Moreover, it has been speculated that E-cadherin-mediated signalling might suppress Langerhans cell maturation and maintain their immature state in vivo (37, 38). In canines, the histiocytoma tumoural cells seem to mimic normal Langerhans cells in an inflammatory environment that undergo an activation/maturation process. This maturation/activation and the interaction with lymphocytes might result in tumour regression and imply, most probably, many other pathways.
It has been proposed that during CCH regression a shift of MHC class II antigen distribution from the cytoplasm to the tumour cell surface occurs with increased infiltration of CD3+ T-cells. However, this study includes only tumours of groups I and II, the tumours in an early regression process (13).
Canine cutaneous histiocytoma has similarities with some forms of human Langerhans cell histiocytosis. Nevertheless, Langerhans cells histiocytosis has distinct clinical behaviours in dogs and humans. In dogs, regression is the natural course of the process. In humans, the clinical picture of Langerhans cell histiocytosis is greatly variable: a limited number of lesions undergo regression (39-42) and other lesions can disseminate and be potentially fatal (43-45). E-cadherin expression in human Langerhans cell histiocytosis is limited to localized forms. Dissemination of this disease and poor outcome may be related to the loss of functions mediated by E-cadherin (46). However, Kapur et al. (47) found a lack of E-cadherin expression both in cutaneous and disseminated forms of human Langerhans cells histiocytosis, which reflects the need to continue studies on this subject.
Despite the progress in investigation, the aetiology and pathogenesis of CCH remain unknown. However, our results lead us to speculate that a dysregulation of the E-cadherin or even the E-cadherin-β-catenin cascade could be a step in the pathogenesis of CCH, similar to what was proposed in human Langerhans cell histiocytosis. A careful regulation of this cascade is essential in normal Langerhans cell activation (48) and abnormalities in the E-cadherin-β-catenin cascade are a major cause of epithelial neoplastic proliferation (49-52). Nevertheless, further studies involving β-catenin are needed to clarify this possible implication. Additionally, heterotypic adhesive interactions between epithelial cells and intraepithelial lymphocytes in vitro are mediated by E-cadherin and its interaction with αEβ7 intraepithelial lymphocyte integrin. This interaction could have a functional relevance for the control of skin local immune response (46, 53) and perhaps in Langerhans cell pathology.
Predominant membranous moderate immunoreactivity for E-cadherin in a group I canine cutaneous histiocytoma. Bar=30 μm.
Since E-cadherin expression in dendritic cells is limited to the Langerhans cells, the presence of this molecule in CCH tumoural cells might be one indication of a Langerhans cell ontogeny, as suggested by others (7, 54). The expression of this molecule should be considered as a useful diagnostic tool for formalin-fixed tumours.
Furthermore, E-cadherin expression and the existence of adherens junctions between CCH tumoural cells might also suggest a high degree of differentiation and might support the benign behaviour of this tumour.
In conclusion, all the CCHs studied here immunoexpressed E-cadherin and this molecule could be related to tumour regression. This study strongly suggests that pathological Langerhans cells in CCH appear in an immature state in early lesions and undergo a maturation process characterized by a down-regulation of E-cadherin. In the regression lesions, these cells differentiate towards a mature phenotype and might themselves interact with lymphocytes. CCH appears as a dynamic lesion that changes its phenotype and determines its own regression.
Acknowledgements
We thank Mrs Lígia Bento for expert technical assistance.
- Received January 16, 2009.
- Revision received March 17, 2009.
- Accepted April 2, 2009.
- Copyright© 2009 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- A Role for Angiogenesis in Canine Cutaneous Histiocytoma Regression: Insights into an Old Clinical Enigma
- Regression of Canine Cutaneous Histiocytoma: Reduced Proliferation or Increased Apoptosis?
- Immunohistochemical and Immunoelectron Study of Major Histocompatibility Complex Class-II Antigen in Canine Cutaneous Histiocytoma: Its Relation to Tumor Regression
- T-Lymphocytic Infiltrate in Canine Mammary Tumours: Clinic and Prognostic Implications