Gradual down-regulation of GM potential in LMPPs but sustained lymphoid potential in vitro and in vivo. (A) Left panel shows clonality of single-cell deposited BM LSKFlt3^hiRag1^GFP−, LSKFlt3^hiRag1^GFPlo, LSKFlt3^hiRag1^GFPint, and LSKFlt3^hiRag1^GFPhi cells cultured under GM conditions (“Methods”) for 8 days. Open bars show cloning frequencies and closed bars show the frequency of high proliferative clones (covering > 50% of well). Mean plus or minus SEM values from 3 experiments. Right panel shows results from morphologic evaluation of MGG-stained cytospin preparations of cells derived from LSKFlt3^hiRag1^GFP−, LSKFlt3^hiRag1^GFPlo, LSKFlt3^hiRag1^GFPint, and LSKFlt3^hiRag1^GFPhi cultures (20 cells each) after 3, 5, and 7 days in culture (“Methods”). Each bar shows relative distribution between wells containing only blast and immature myeloid cells (Bl/Imm), cells of both the granulocyte and monocyte lineages (GM), and cells of only the monocytic (M) lineage, as determined by morphologic evaluation of MGG-stained cytospin slides. Wells with only granulocytes were not detected. Note that although the peak for GM generation varies between the populations, all produce G and M cells. Mean plus or minus SEM values from 2 experiments, each performed with 6 replicate determinations (wells) per population. (B) T- and B-cell potential of single-cell deposited LSKFlt3^hi, LSKFlt3^hiRag1^GFP−, and LSKFlt3^hiRag1^GFPhi BM cells grown for 3 to 4 weeks on OP9-DL1 and OP9 respectively, as evaluated by FACS. T cells were defined as NK1.1⁻Thy1.2^hiCD25^hi and/or NK1.1⁻CD4⁺CD8⁺ and B cells as B220⁺CD19⁺, both negative for the viability stain PI as previously described^10,15  or DAPI. Mean plus or minus SEM values from 3 experiments (n = 16-48 per group and experiment). Right panels show representative FACS profiles of analyzed clone derived from a single cell. To verify the T-cell potential by molecular methods, 7 NK1.1⁻Thy1^hiCD25^hi clones were analyzed for expression of CD3 antigen, epsilon polypeptide (Cd3e), and pre–T-cell antigen receptor alpha (Ptcra). All clones were found to be positive for both T-cell genes, whereas clones grown on OP9 were negative for the T-cell genes but, as expected, positive for hypoxanthine guanine phosphoribosyl transferase (Hprt) and protein tyrosine phosphatase, receptor type C (Ptprc, gene for Cd45) (data not shown). (C) Reconstitution of lethally irradiated recipients that underwent transplantation with 2000 sorted cells of the indicated cell populations sorted from Rag1^GFP mice, in competition with 200 000 BM cells. Results show mean plus or minus the standard deviation (SD) of reconstitution levels (total blood cells) at 3, 6, 9, and 12 weeks after transplantation from 2 experiments (n = 6–7). (D) Lineage analysis of mice that underwent transplantation with LSKFlt3^hiRag1^GFPint/hi cells. Myeloid cell reconstitution was below detection level (less than 0.02%) of total donor-derived cells at all time points after transplantation.