Effect of antilymphocyte globulin (ALG) on bone marrow T/non-T cells from aplastic anaemia patients and normal...

The aim of this study was twofold: (a) to test the effect of antilymphocyte globulin (ALG) on bone marrow (BM) T/non-T cells, and (b) to look for a possible differential response of cells from severe aplastic anaemia (SAA) patients and controls. For this purpose bone marrow T/non-T cells from normal individuals (n = 7) or aplastic patients (SAA, n = 13) were kept in liquid culture with or without ALG. Supernatants were then tested for enhancement/suppression on colony forming unit, granulocyte-macrophage (CFU-GM) growth (in the presence of exogenous recombinant granulocyte-macrophage colony stimulating factor (rGM-CSF)), or for their ability to support CFU-GM growth (in the absence of exogenous rGM-CSF). Supernatants from SAA T cells suppressed CFU-GM growth of normal bone marrow cells in 5/12 patients (mean expected growth (EG) 71 +/- 16%), but not after incubation with ALG (mean 110 +/- 29% EG, P = 0.03). No inhibition could be obtained with the supernatants from untreated normal T cells. Significant enhancement was seen with ALG treated versus untreated SAA T cells (142 +/- 28% EG v. 105 +/- 61% EG, P = 0.01) and with ALG treated versus untreated SAA non-T cells (165 +/- 26% EG v. 105 +/- 23% EG, P = 0.01), but not in controls. Supernatants from SAA and control T/non-T cells were capable of promoting colony formation in the absence of rGM-CSF (colony-stimulating activity (CSA) production): 16 +/- 14% for SAA-T cells and 19 +/- 18% EG for non-T cells (100% = 30 ng rGM-CSF/ml). The addition of ALG increased CSA production in T cells to 37 +/- 23% EG (P = 0.04) and in non-T cells to 40 +/- 13% EG (P = 0.04). Similar results could be obtained in controls.

(a) ALG interacts in vitro with bone marrow T and non-T cells from SAA patients, down-regulating the production of negative lymphokines and enhancing the release of haemopoietins; (b) the latter, but not the former effect, can be shown also with cells from normal controls. The two effects are not mutually exclusive, and are likely to provide maximal benefit in vivo.