Cellular distribution of human leucocyte adhesion molecule ICAM-3.

AIMS: To describe the distribution of the recently cloned human leucocyte adhesion molecule ICAM-3 in normal and neoplastic tissues and cell lines. METHODS: A panel of four monoclonal antibodies to ICAM-3 were used to stain cell lines and sections of human lymphoid tissues using the alkaline phosphatase-anti-alkaline phosphatase immunocytochemical method (APAAP). RESULTS: In peripheral blood ICAM-3 was detected on monocytes, granulocytes, and most lymphocytes. In sections of human lymphoid tissue the antigen was also found on most lymphocytes, but many of the proliferating B cells found in the germinal centres of secondary lymphoid follicles were ICAM-3 negative. ICAM-3 was also found on neoplastic white cells (in chronic lymphocytic leukaemia, hairy cell leukaemia, acute and chronic myeloid leukaemia, and multiple myeloma) with the exception of Reed-Sternberg cells in Hodgkin's disease, many of which were negative. ICAM-3 was consistently absent from cells and tissues of non-haemopoietic origin. Endothelium (which expresses ICAM-1) was negative for ICAM-3, with the exception of vessels in some neoplastic lymphoid samples which showed variable staining for ICAM-3. CONCLUSIONS: These findings suggest that ICAM-3 is essentially restricted to the haemopoietic system and is reciprocal in its expression to ICAM-1, in that it is present on resting cells and its level falls as a result of cell activation.

An immunocytochemical study of p53 and bcl-2 protein expression in Hodgkin's disease.

In view of their reported reciprocal effects on apoptosis, the expression of p53 and bcl-2 proteins was studied in 46 cases of Hodgkin's disease by immunocytochemical labeling. We found p53 protein in Reed-Sternberg cells and their mononuclear variants in 16 of the 46 cases (34.7%) of Hodgkin's disease, mainly in a nuclear pattern. This restricted expression on Reed-Sternberg cells and variants supports their neoplastic nature. This overexpression of p53 protein in one third of Hodgkin's disease cases is similar to that seen in many other human malignancies. bcl-2 protein was present in mantle zone B cells and scattered T cells in all cases, and in 17 cases (37.7%) of Hodgkin's disease in Reed-Sternberg cells and their mononuclear variants. Six cases coexpressed both proteins, whereas in 18 cases neither was identified. There is no apparent relationship between p53 and bcl-2 protein expression, and on the basis of the present results there is no reason to suppose that they have any particular complementary effects on the neoplastic transformation in Hodgkin's disease.

CD68 reactivity of non-macrophage derived tumours in cytological specimens.

AIMS: To investigate the presence of the macrophage associated antigen CD68 in non-haematopoietic tumours. METHODS: Cytological specimens from non-macrophage derived tumours were stained using the alkaline phosphatase anti-alkaline phosphatase immunocytochemical method (APAAP) and three monoclonal anti-CD68 antibodies, Y1/82A, EBM11, and KP1. RESULTS: Reactivity of malignant cells with one or more of the antibodies was seen in 11 out of 40 adenocarcinomas and in one of seven poorly differentiated carcinomas; other neoplasms, including 10 cases of squamous carcinoma, three of malignant melanoma, and four of oat cell carcinoma were negative. Monoclonal antibody KP1 gave the strongest staining and reacted with the highest proportion of neoplastic cells. CONCLUSIONS: CD68 is expressed in a proportion of epithelial tumours although the labelling is usually less intense than in macrophages. Anti-CD68 antibodies should therefore be used as part of a panel in the diagnosis of poorly differentiated neoplasms in cytological material.

Immunophenotype of multinucleated and mononuclear cells in giant cell lesions of bone and soft tissue.

AIMS: To compare the antigenic phenotype of giant cells in giant cell lesions of bone and soft tissue with that of osteoclasts and macrophage polykaryons. METHODS: Formalin fixed, paraffin wax embedded sections of 106 giant cell lesions, 19 granulomatous, and 14 osteoclast containing lesions were immuno-histochemically stained for leucocyte common antigen (LCA), CD68, and HLA-DR. RESULTS: Osteoclasts and giant cells of giant cell tumour of bone and giant cell reparative granuloma could be distinguished by their generalised absence of HLA-DR reaction from macrophage polykaryons and giant cells in other giant cell lesions of bone and soft tissue. Staining for LCA, CD68, and HLA-DR was useful in distinguishing reactive histiocytic giant cells and osteoclasts from tumour giant cells. CONCLUSIONS: A panel of macrophage associated antigens should be diagnostically useful in differentiating the histological nature of giant cells in various giant cell lesions of bone and soft tissue.