It has previously been reported that rabbit antisera prepared with papain digests of human malignant spleen cell membranes recognize antigens expressed by normal and neoplastic human B lymphocytes (1-3). The antigens recognized by these antisera appear to be similar to, and may be identical with, the p23, 30 antigen identified by Humphreys and colleagues and by Schlossman and colleagues on normal B cells, 15-20% of null cells, and the majority of acute leukemia cells (4,5). A surprising and unexplained observation was the frequent detection of similar antigens on acute and chronic myeloid (granulocytic) leukemia cells. It was not clear why neoplastic myeloid cells should express an antigen found on mature B cells but not mature granulocytes. To answer this question, we examined the effects of similar antisera on normal human myeloid stem cells. These cells, colony-forming unit-culture (CFU-C), were assayed for their ability to form granulecyte and macrophage colonies in agar, in vitro (6, 7) in the presence or absence of antiserum and complement (C). It is now clear that the antigens recognized by such antisera also occur on normal myeloid progenitors.
Materials and MethodsAntisera. As previously described, antisera were raised in rabbits immunized with papain digests of cell membranes from the spleen of patients with a variety of hematologic neoplasms (1).Cells. Heparinized aspirates of bone marrow were obtained from appropriately informed normal adult volunteers, and single cell suspensions were prepared as previously described (8).Heparinized peripheral blood of patients with chronic lymphocytic leukemia (CLL) was used to prepare lymphocyte populations of >98% purity for absorption of antisera. Permanent tissue culture lines of B cells (pooled lines B44, 45, and 47) and T cells (MOLT-4) were also used for absorption (1).Cytotoxicity Testing. Normal bone marrow cells, 2 Ă 106/ml, in 15% heat-inactivated fetal calf serum in McCoy's 5A medium were incubated with 1/10 volume of heated rabbit antiserum or heated normal rabbit serum at room temperature for 30 min. Normal rabbit serum, 1/10 volume, as a source of C was added for an additional 60 min. Cells were then plated in 0.3% agar above a feeder layer of 1 Ă 10 ~ peripheral blood leukocytes for assay of CFU-C as previously described (8). After 10 days incubation the number of colonies with >40 cells were determined. Results were expressed as the percentage inhibition (mean _+ SD) of colony formation in experimental assays relative to appropriate controls. The concentration of fresh rabbit serum used to provide C had no inhibitory effect on human marrow colony formation.
ResultsAntisera to papain-digested spleen membranes inhibited normal marrow