T cell depletion (TCD) has been achieved using techniques that cause the inactivation, lysis, or physical removal of T cells from the donor marrow. The clinical results of TCD reflect, in part, the degree of TCD achieved and the subsets that are removed. To better evaluate TCD using the monoclonal antibody (mAb) T10B9, we have performed a series of flow cytometry and mAb blocking studies and have developed a new limiting dilution assay (LDA) that allows the detection of T cell subsets that survive treatment. T cell growth was stimulated with PHA, rIL-2, and irradiated feeder PBMC in a total well volume of 20 microliters. Growth was scored by microscopic examination on days 14-16 of incubation. Immunomagnetic beads coated with mAb were added to the growing wells and incubated, then the plates were fixed to a template of samarium cobalt magnets before washing away nonadherent cells. Wells in which > 50 cells bound > or = 2 beads were scored as positive. Flow cytometry indicated that T10B9 recognized all T cells, but complement-mediated lysis spared a significant proportion of the TCR gamma delta + subset. The epitope recognized by T10B9 on TCR gamma delta + cells appears to be differentially expressed compared with TCR alpha beta + T cells based on antibody blocking studies. In contrast to antibodies to CD3 epsilon, T10B9 binds less well to TCR gamma delta + cells, possibly resulting in incomplete complement-mediated lysis of this subset. The relative sparing of TCR gamma delta + cells was found in marrow and peripheral blood. Subset LDA confirmed that the TCR gamma delta + cells detected by flow cytometry were capable of growth and further showed that OKT3 did not spare TCR gamma delta + cells. The subset LDA should prove useful in helping to assess the role of T cell subsets in clinical events post-TCD bone marrow transplantation.