Abstract
Little is known of how hematopoietic stem cells (HSCs) differentiate. We have previously suggested that particular myeloid lineages, such as a neutrophil/macrophage (nm) lineage, arise asymmetrically through their first division. In this study, we asked whether lymphoid lineage restriction similarly takes place at the level of HSCs.
CD34−c-Kit+Sca-1+lineage−(KSL) cells and CD34+KSL cells, respectively, highly enriched for long-term and short-term repopulating cells, were isolated from bone marrow (BM) of B6 mice by FACS. These cells were subjected to in vitro colony assay and in vivo repopulation assay with use of Rag-2-deficient BM cells, instead of normal BM cells, as competitor cells to facilitate the detection of lymphoid lineage reconstitution. Single CD34−KSL cells were allowed to divide once in the presence of SCF+TPO or SCF+IL-3 under serum-free culture condition. Resultant paired daughter (PD) cells were subsequently separated by micromanipulation and individually transplanted into lethally irradiated mice. Recipient mice were analyzed 7 weeks after transplantation to detect myeloid (My), B-lymphoid (B), and T-lymphoid (T) lineage repopulation by single test donor cells. When single CD34−KSL cells were injected into lethally irradiated mice, about 20% of the cells were detected as MyBT lineage repopulating cells. Interestingly, about 25% of the cells were detected as MyT lineage repopulating cells without detectable level of B lineage reconstitution. Limiting dilution analysis of CD34+KSL cells estimated the frequency of B lineage repopulating cells as close to 1 in 10 cells and that of T lineage repopulating cells as about 1 in 80 cells. Because most colony forming cells among CD34+KSL cells exclusively gave rise to nm lineage, it was assumed that the differentiation potential of CD34+KSL cells was mostly restricted to nm and B lineages. These data suggest more MyT repopulating cells present in CD34− fraction than in CD34+ fraction and, in contrast, more MyB repopulating cells present in CD34+ fraction than in CD34− fraction among KSL cells. On the other hand, My lineage repopulation was observed in about 60% of the PD cells. B and T, B, or T lineage differentiation potential was also detected in approximately 8%, 8%, or 40% of the PD cells with My lineage repopulating activity when generated in the presence of SCF+TPO and in 2%, 5%, or 2% of the cells when generated in the presence of SCF+IL-3. As compared with freshly isolated CD34−KSL cells, T, but not B lineage differentiation potential was well maintained in PD cells by SCF+TPO whereas either B or T lineage differentiation potential was hardly maintained in PD cells by SCF+IL-3. These data indicate the instructive role of cytokines in the restriction of lymphoid potential in HSCs. MyT lineage repopulating cells appeared to be directly generated from MyBT stem cells via their asymmetric division. Taken together, we conclude that asymmetric division of HSCs results in their lymphoid lineage restriction and that T lineage commitment takes place at the level of HSCs, independent of and prior to B lineage commitment which occurs at later stages of differentiation. We propose a novel differentiation model of HSCs which challenges the CLP and CMP based model.
Author notes
Corresponding author
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal