A B S T R A C T The characteristics of cyclophosphamideinduced suppression of established cell-mediated immunity were studied in guinea pigs previously sensitized to tuberculin. Cyclophosphamide treatment for 5 days produced a dose-dependent peripheral lymphocytopenia and a disproportionately greater neutropenia which was particularly striking at high doses of 20 mg/kg per day (approximately 200 mg/m' per day). Lymphocytes remaining in the circulation of cyclophosphamidetreated animals showed a dose-dependent reduction of both in vitro proliferative and macrophage migration inhibitory factor responses to tuberculin compared to lymphocyte responses of controls. Proliferative responses to phytohemagglutinin and concanavalin A were not significantly suppressed. Additional studies showed that cyclophosphamide suppressed the proliferative and migration inhibitory factor responses to tuberculin of lymph node and splenic as well as circulating lymphocyte populations. These studies showed that relatively short-term cyclophosphamide administration produced immunosuppression by quantitative as well as qualitative changes in lymphocyte populations. Significant suppression of lymphocyte function, however, was achieved only with doses of cyclophosphamide which also produced a severe neutropenia.

SUMMARYWe studied the immunologic correlates of disease activity and differences among subgroups of patients with idiopathic inflammatory myopathy by analysing phenotypic and activation marker expression on peripheral blood mononuclear cells (PBMC). Compared with controls, myositis patients with clinically active disease (n = 51) had significantly lower proportions of CD8+ cells and higher proportions of PBMC that expressed DR, CD3-DR, CD14-DR, interleukin-2 receptors, and the late T cell activation markers CD26 and TLiSA 1. TLiSA1 expression, a marker for cytotoxic differentiation, correlated significantly with both clinical activity indices and serum levels of muscleassociated enzymes. In serial studies of seven patients, the proportion of PBMC expressing MHC class II antigen and late T cell activation markers decreased as myositis disease activity decreased, independent of type of therapy. Among the clinical subgroups, polymyositis (n = 21) and inclusion body myositis (n = 11) were virtually indistinguishable; dermatomyositis patients (n = 19) showed decreased proportions of CD3+DR+ and TLiSA1 + cells, and increased proportions of CD20+ and CD20+DR+ cells compared with the other two groups. Patients with autoantibodies to histidyltRNA synthetase (Jo-l antigen, n = 11) had significantly lower proportions of CD3+ and CD4+ cells, lower CD4/CD8 ratios, and higher proportions of DR+ cells expressing CD20, compared with patients without anti-Jo-l antibodies. These findings support the concept that activated lymphocytes, especially cells undergoing anamnestic responses and cytotoxic differentiation, are important in the pathogenesis of idiopathic myositis. Moreover, taken together with other studies, these data suggest that groups of patients segregated by clinical or autoantibody status have different mechanisms of systemic immune activation and immunopathology.

A B S T R A C T This study was undertaken to determine the capability of lymphocytes in the bone marrow of normal individuals to mediate nonspecific killer cell functions in assays of phytohemagglutinin (PHA) -induced cellular cytotoxicity, and antibody-dependent cellular cytotoxicity (ADCC) against "1Cr-labeled chicken erythrocyte target cells. Relatively pure mononuclear cell suspensions were obtained from bone marrow aspirates in 30 normal volunteers by sucrose gradient centrifugations and from the peripheral blood of the same individuals by Hypaque-Ficoll density centrifugations. At an effector: target ratio of 10: 1, the PHA-induced cellular cytotoxicity of peripheral blood was 78.8±1.3%, while that of bone marrow was not significantly less at 66+9% (P > 0.1). At low effector: target ratios, the ADCC of bone marrow was negligible, while at higher effector: target ratios (20: 1) bone marrow ADCC was 69±3.7%, which was comparable to that of peripheral blood. The lymphocytes themselves in the mononuclear cell suspensions of both peripheral blood and bone marrow were capable of cytotoxic activity since depletion of monocytes from the suspensions by adherence to rayon wool and G-10 Sephadex columns did not remove the cytotoxic activity. Blocking of the Fc receptor on the effector cells by the addition of aggregated gamma globulin to the cultures suppressed the ADCC but not the PHA-induced cellular cytotoxicity of both peripheral blood and bone marrow, indicating that ADCC is dependent on an Fc receptor on the effector cell in both compartments. These studies demonstrate that the bone marrow of normal humans contains populations of lymphoid cells which have highly efficient killer cell capacities. It is uncertain what portion

No abstract