The chromosomal inversion (16)(p13q22), which is associated with the M4-eosinophilia subtype of human acute myeloid leukemia, causes the fusion of two distinct genes

Summary Recent advances in immunological and molecular technology have prompted proposals to change tumour classification and treatment strategies. Cell surface antigens are now easy to access, and tumour origins and clinical characteristics are now readily identifiable. However, in diffuse large B‐cell lymphoma (DLBCL), one of the heterogeneous forms of haematological malignancy, the clinical significance of tumour surface antigens has not been well documented. We analysed the tumour surface antigens of 50 tumours from newly diagnosed DLBCL patients by flow cytometry in accordance with their clinical characteristics and followed the patients for a median 3·7 years. Statistical analysis showed that CD21 expression was significantly negatively associated with mortality in DLBCL (CD21 negative versus positive; relative risk = 2·36, P < 0·05). As a result of these clinical observations, we generated CD21‐overexpressed (CD21+) lymphoma cell lines after gene transfection and analysed tumour cell growth in vivo in immunocompromised mice. Mice challenged with vector‐only transfectants and parental cells as controls died within 50 d. In contrast, mice injected with CD21+ transfectants exhibited significantly reduced tumour growth and 83% survived long term (versus control groups; P < 0·05). Interestingly, all established CD21+ transfectants (six clones from different bulks) showed homotypic aggregation during in vitro cell culture, and anti‐CD21 antibodies did not block this aggregation. Expression of CD21 is strongly associated with increased survival in DLBCL in vivo. CD21 expression may be indirectly concerned with the expression of additional cell adhesion molecules.

A new human leukemia cell line, designated as ME-1, was established from the peripheral blood leukemia cells of a patient with acute myelomonocytic leukemia with eosinophilia (M4E0). This cell line has the characteristic chromosome abnormality of M4E0, inv(16) (p13q22). When cultured in RPMI 1640 medium containing 10% fetal calf serum, ME-1 cells were monoblastoid, but with the addition of cytokines such as interleukin-3 (IL-3), granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-4, or medium conditioned by phytohemagglutinin-stimulated human peripheral leukocytes (PHA-LCM), the cells exhibited differentiation to macrophage-like cells. PHA-LCM also promoted eosinophilic-lineage differentiation of this cell line, although IL-5 did not do so. To elucidate the mechanism of proliferation and differentiation of ME-1 cells, we studied the effect of a potent inhibitor of protein kinase C, 1-(5-isoquinolinyl-sulfonyl)-2- methylpiperazine (H-7), on colony formation of ME-1 cells. H-7 inhibited colony formation of ME-1 cells by IL-3 or GM-CSF dose dependently, but had little inhibitory effect on colony formation by IL- 4. These results indicate that the proliferation and differentiation of ME-1 cells by IL-3 or GM-CSF were related to the activation of protein kinase C, while those by IL-4 involved other regulatory systems. ME-1 cells should be useful for studying the pathogenesis of M4E0 and the mechanisms of proliferation and differentiation of leukemic and normal progenitors by cytokines.

In an attempt to clarify the mechanisms of cytotoxicity mediated by CD4+ cytotoxic T lymphocytes (CTL), the expression of perforin and membrane-bound lymphotoxin (LT) (tumor necrosis factor-beta) in herpes simplex virus (HSV)-specific CD4+ human cytotoxic and noncytotoxic T- cell clones was examined. Three HSV-specific CD4+ human CTL clones that showed HLA-DR-restricted cytotoxicity and proliferative response were established. The cytotoxicity of these clones in 5-hour 51Cr release assays was found to be mediated by the directional target cell lysis and not by the release of cytotoxic soluble factors, ie, “innocent bystander” killing. Northern blot analysis showed that messenger RNAs for perforin and LT, which were both considered to be important mediators for cytotoxicity of CD8+ CTL and natural killer cells, were abundantly expressed in HSV-specific CD4+ CTL clones. Expression of perforin in the cytoplasm of CD4+ CTL clones was also detected by immunohistochemical staining using a monoclonal antibody against perforin. In addition, LT bound to the cell surface of CD4+ CTL clones was detected by flow cytometry. In contrast, little or no expression of perforin and LT was detected in three HSV-specific CD4+ noncytotoxic T- cell clones. Although the cytotoxicity mediated by lymphokine-activated killer cells was partly inhibited by addition of anti-LT antibody, it did not show any effect on the cytotoxicity of HSV-specific CD4+ CTL clones. In addition, it was found that cytotoxicity mediated by these CD4+ CTL clones was Ca2(+)-dependent. These data thus suggest that perforin and membrane-bound LT are both expressed in HSV-specific CD4+ CTL, although perforin might be the more important mediator in short- term culture.

SummaryPlatelet agonists and RGD-containing peptides can convert platelet membrane glycoprotein (GP) Ilb-IIIa from its resting state to an activated state competent to bind soluble fibrinogen. We examined the effects of two anti-GPIIb-IIIa monoclonal antibodies, PMA1 and PMA5, on fibrinogen binding to agonist- and RGD-activated GPIIb-IIIa. PMA1 abolished aggregation of both agonist- and RGDS peptide-activated fixed platelets, and inhibited the binding of 125I-fibrinogen to these platelets almost completely. PMA5 had the same effects on agonist-activated platelets, but had little effect on the aggregation of RGDS-activated fixed platelets, and inhibited fibrinogen binding to RGDS-activated fixed platelets by only 44%. PMA5 bound to agonist- and RGDS-activated platelets equally. Immunoblot analysis showed that PMA5 bound to intact GPIIIa, but not to a 66 kDa fragment of GPIIIa digested by chymotrypsin. Although PMA5 inhibited platelet adhesion to immobilized fibrinogen by 94%, 44% of the remaining adherent platelets were spread. In contrast, no platelet spreading was observed in the presence of PMA1. These findings indicate that PMA5 is a novel anti-GPIIIa monoclonal antibody with the ability to inhibit fibrinogen binding to agonist- and RGD-activated states of GPIIb-IIIa differentially, and suggest that binding of immobilized fibrinogen to RGD-activated GPIIb-IIIa is necessary for platelet spreading.