AKR mice develop a spontaneous thymic leukemia after 8 months of age (1). We have previously shown that thymus and spleen cells from these leukemic animals are capable of suppressing the in vitro generation of an antibody response (2) by normal AKR spleen cells. This suppression requires cell contact (2, 3), has unique genetic restrictions (4), and is usually (but not always) susceptible to x-irradiation and mitomycin treatment (4). The susceptibility to x-irradiation and mitomycin treatment suggests involvement of DNA, and cell separation studies enforced this idea because the cells with the highest endogenous rate of DNA synthesis cosediment with the fraction of cells that carry out suppression (5). The need for cell contact led us to investigate the ability of isolated membranes to suppress normal responses; membranes from leukemic cells are able to suppress the responses of normal cells (unpublished results).In an effort to reconcile the apparent involvement of DNA, the need for cell contact, and the ability of isolated membranes to suppress (unpublished data), we tested one of the several possible explanations-namely, that the leukemic suppressor cell has a form of DNA on its surface that is somehow involved in the suppression. In this paper we show that the suppression of in vitro anti-sheep erythrocyte (SRBC) responses by leukemic cells is susceptible to DNase treatment and the suppressing cells can be removed after being reacted with anti-DNA antibodies and passed through an immunoadsorbant column. Preparation and Use of Cell Affinity Chromatography Columns. Goat anti-rabbit IgG (IgG fraction purified by DEAE-cellulose chromatography) dissolved in 0.1 M borate, pH 8/0.5 M NaCI (coupling buffer) was coupled to CNBractivated Sepharose 6MB (10 mg of antibody per ml of swollen gel) on an end-over-end shaker at room temperature for 2 hr. Any remaining reactive groups on the gel were blocked by incubation of the gel in 1 M ethanolamine at pH 8 for 2 hr. Blocking agent and noncovalently adsorbed protein were removed by washing in 0.1 M acetate, pH 4/0.5 M NaCI followed by coupling buffer.
MATERIALS AND METHODSFor columns 12-ml disposable plastic syringes were fitted with 80-am pore size plastic discs (Bellco Glass) and three-way stopcocks (Pharmaseal, Puerto Rico). Columns were loaded with 8 ml of the Sepharose-antibody conjugate, washed, and stored with coupling buffer containing 0.1% NaN3. Before use, columns were equilibrated with RPM!1 medium containing 5% fetal calf serum and 0.025 M EDTA (RPMI/EDTA) for 30 min at 4VC. Cell suspensions treated with antibody (see antisera treatment) were washed three times in RPMI/EDTA and were suspended at 2-3 X 108 cells per 1.5 ml of RPMI/EDTA. The cells were gently layered onto the column with a sterile pasteur pipette and washed into the column with RPMI/EDTA. Cells were allowed to react on the column for 45 min at 4VC. Nonbound cells (fraction I) were collected in RPMI/EDTA at 0.5 ml/min for the first 15 ml; then an additional 30 ml were collected at a rapid flow rate...