Distinct developmental patterns of short-term and long-term functioning lymphoid and myeloid precursors defined by competitive limiting dilution analysis in vivo.

Functional abilities of individual marrow precursor cells were defined by competitive limiting dilution without enrichment, tissue culture, or induced marking, manipulations that might affect cell functions. We directly measured long-term repopulating abilities in limiting doses (0.25-1.0 x 10(5)) of genetically marked congenic marrow cells. These were mixed with a standard dose of 4 or 5 x 10(5) competitor marrow cells, which contained a predictable distribution of precursor cells and allowed quantitative assays. Percentages of donor type T and B lymphocytes, granulocytes, platelets, and erythrocytes were measured in recipient blood. Applying the maximum likelihood statistic, concentrations (per 10(5)) of precursors repopulating at least one lineage were: 4.7 and 6.0 after 6 wk, 1.6 and 2.7 after 14 to 15 wk, and 1.2 and 1.9 after 30 to 32 wk; concentrations repopulating at least three lineages were 2.3 and 3.4 after 6 wk, 0.9 and 1.7 after 14 to 15 wk, and 0.9 and 1.3 after 32 wk. Almost all precursors functioning after 14 wk repopulated all lineages. At 6 wk, similar levels of donor cells were produced in recipients of both short- and long-term precursors. However, after 14 to 32 wk, contributions by short-term precursors (about two-thirds of the precursors) dropped to zero, while contributions by long-term precursors (about one-quarter of the precursors) expanded severalfold. The latter permanently repopulated all lineages after 30 to 32 wk, functioning as the most primitive stem cells (PSC) in the immune and myeloid systems. Nearly all the variance in long-term repopulated recipients was explained using the Poisson distribution to calculate donor percentages in a model where each donor and competitor PSC contributed equally.