Regulation of cell growth by recombinant oncostatin M.

Oncostatin M is a novel growth regulator originally isolated from differentiated human histiocytic lymphoma cells and activated T-lymphocytes based on its ability to inhibit the growth of A375 melanoma cells. We report here that oncostatin M is a widely acting regulator which alters the growth and/or morphology of cells derived from a variety of cancer cell types. At picomolar concentrations, recombinant oncostatin M inhibited the growth of 13/24 tumor cell lines. Six out of 7 lung cancer cell lines were inhibited by oncostatin M, but none of 6 colon cancer cell lines were affected. Oncostatin M also stimulated the growth of some normal cells (3/6), indicating that it, like many growth regulators, is bifunctional. Oncostatin M receptors appear necessary but not sufficient for a growth response to oncostatin M, since none of the cell lines lacking receptor responded to oncostatin M, whereas many but not all cell lines with receptor responded to oncostatin M. Receptor size (Mr congruent to 150,000) was similar for cells in which growth was inhibited, stimulated, or unaffected by oncostatin M.

Identification of Ki (Ku, p70/p80) autoantigens and analysis of anti-Ki autoantibody reactivity.

Anti-Ki (Ku, p70/p80) autoantibodies, named after the prototype patient Kikuta by Tojo et al., occur in approximately 10% of patients with SLE, often in association with anti-Sm autoantibodies. The immunofluorescent staining pattern characteristic of anti-Ki antibodies is diffuse speckled nuclear, although some substrates show nucleolar staining as well. Anti-Ki sera specifically immunoprecipitated two protein antigens, Ki86 (Mr 86,000) and Ki66 (Mr 66,000), from radiolabeled cell extracts. The Ki system was found to be immunologically identical to the Ku system described by Mimori et al. and the p70/p80 system described by Reeves. The Ki primary in vitro translation products were identified and proved similar in size to the cellular antigens. The Ki antigens were purified from human spleen by immunoaffinity chromatography followed by SDS-PAGE. The purified Ki antigens proved to be closely related by amino acid composition, and did not appear to be phosphorylated, glycosylated, or associated with RNA. The Ki antigens were found to bind to DNA, in agreement with the observations on the Ku and p70/p80 antigens. They were found to be widely conserved in mammals and were coordinately expressed in all tissues tested. Anti-Ki autoantibodies were purified by antigen-affinity chromatography and were tested by immunoblotting. The antibodies were classified as class I, II, or III, depending on their reactivity with the Ki antigens in immunoblots. Class I antibodies cross-reacted with both Ki antigens, class II antibodies reacted solely with Ki66, and class III antibodies reacted solely with Ki86. These results suggest that at least three different epitopes are present on the Ki autoantigens and that patients differ in their autoantibody response to each epitope.