Methods in laboratory investigation. Immunoelectron microscopic methods for demonstration of antigens on normal human melanocytes and other epidermal cells.

Different procedures for fixation and processing were evaluated in order to examine the antigenic profile of melanocytes and other epidermal cells for immunoelectron microscopy. For this purpose the monoclonal antibodies anti-HLA-A, B, C, anti-HLA-DR, anti-T6, and the melanoma-associated monoclonal antibody NKI /C-3 were used as markers. Fixation with periodate-lysine-paraformaldehyde yielded better antigenic and ultrastructural preservation than 3% paraformaldehyde or picric acid-paraformaldehyde did. Skin was further processed by five different methods: (a) 15-micron frozen sections; (b) 75-micron agar-embedded, tissue chopper sections; (c) 15-micron polyethylene glycol-embedded sections; (d) epidermal cells in suspension; and (e) epoxy sections (postembedding staining) were prepared for the immunoperoxidase procedure. Antigenicity was best preserved in the cell suspension method and somewhat less, but with a similar staining distribution, with the first three methods. Staining with the polyethylene glycol-embedded sections was only achieved if they were left free-floating in buffer; no staining was observed when the sections were mounted on glass slides and left to dry overnight at 37 degrees C. Epidermal cells remained unreactive in the postembedding method, even after etching. Ultrastructural preservation of the agar-embedded sections and the cells in suspension was superior to the other preembedding methods. Melanocytes mostly showed moderate staining for HLA-A, B, C and slight staining for the antigen that is recognized by NKI /C-3. The latter was further demonstrated on Langerhans cells and indeterminate cells which also expressed HLA-A, B, C, T6, and HLA-DR antigens.

The cell population of cutaneous B-cell lymphomas.

The cellular composition of the dermal infiltrates of eleven patients with a cutaneous B-cell lymphoma (four centroblastic lymphomas, two centroblastic/centrocytic lymphomas and five immunocytomas) was investigated. The distribution of both the neoplastic and the non-neoplastic cells (reactive T cells, macrophages and dendritic reticulum cells) in primary and secondary cutaneous B-cell lymphomas was very similar to that of B-cell lymphomas of the same type in lymph nodes. Reactive T cells and dendritic reticulum cells were only occasionally found in centroblastic lymphoma, but were very numerous in centroblastic/centrocytic lymphoma. The large majority of these T cells in centroblastic/centrocytic lymphoma showed the phenotype of activated T-helper cells (Leu-I+, Leu-3a+, OKT4+, HLA-DR+). In immunocytomas many T cells reactive with Leu-I, Leu-3a, and OKT4 but not with anti-HLA-DR antiserum, and varying numbers of dendritic reticulum cells were found. Since B-cell lymphomas in lymph nodes are the neoplastic counterparts of B-cell reactions which take place after antigenic stimulation in the different lymph node compartments, our results suggest that cutaneous B-cell lymphomas may be the malignant counterparts of similar B-cell reactions in the skin.

In situ localization of lymphocyte subsets in synovial membranes of patients with rheumatoid arthritis with monoclonal antibodies.

Immunological analysis of the mononuclear infiltrates in 12 rheumatoid synovial membranes was performed by means of modified peroxidase antiperoxidase technique using a panel of monoclonal antibodies, directed against T cell differentiation antigens and HLA-DR (Ia-like) antigens. Helper/inducer T lymphocytes (OKT4 +, Leu3a+) were found in large numbers in nodular lymphoplasmocellular infiltrates, whereas the number of cytotoxic/suppressor T lymphocytes (OKT8 +m Leu2a+) was very low, resulting in a high T4/T8 or Leu3a/Leu2a ratio (6á14:1). In diffusely localized lymphoplasmocellular infiltrates this ratio was only slightly increased as compared with the peripheral blood of patients with rheumatoid arthritis (2á4:1). Moreover, most of these T lymphocytes appeared to have Ia-like antigens and seemed to have contact with HLA-DR+, sometimes weakly OKT6+ dendritic nonlymphoid cells. The results showed a constant basic localization pattern of T lymphocyte subsets and suggest that interactions between dendritic nonlymphoid cells, T lymphocyte subsets and B lymphocytes determine the ultimate architecture of the inflammatory infiltrates in the rheumatoid synovial membrane.