Superior Impact of p18INK4c over p27kip1 on Long-Term Repopulating Ability of Hematopoietic Stem Cells.

The G1-phase is a critical window during the cell cycle in which stem cell self-renewal may be balanced with differentiation and apoptosis. Increasing evidence suggests that the cyclin-dependent kinase inhibitors (CKIs) such as p21Cip1/Waf1, p27kip1, p16INK4A, and p18INK4C (p21, p27, p16 and p18 hereafter) are involved in stem cell self-renewal, as largely demonstrated in murine hematopoietic stem cells (HSCs). For example, we have recently demonstrated a significant increase of HSC self-renewal in the absence of p18 (Yuan et al, Nature Cell Biology 2004). But the actual roles of these CKIs in HSCs appear to be distinct as p21 and p18 have opposite effects (Yu H et al, ASH 2004) whereas p16 has a limited effect (Stepanova et al, Blood 2005) on HSC exhaustion after serial bone marrow transfer. Like p18, however, p27 was recently reported to also inhibit HSC self-renewal due to the fact that the competitive repopulating units (CRUs) were increased in p27−/− mouse bone marrow (Walkley et al, Nature Cell Biology 2005) in contrast to the results in a previous report (Cheng T et al, Nature Medicine 2000). To further gauge the impact of p18 versus p27 on the long-term repopulating ability (LTRA) of HSCs, we have generated different congenic strains (CD45.1 and CD45.2) of p18−/− or p27−/− mice in the C57BL/6 background, allowing us to compare them with the competitive repopulation model in the same genetic background. The direct comparison of LTRA between p18−/− and p27−/− HSCs was assessed with the competitive bone marrow transplantation assay in which equal numbers of p18−/− (CD45.2) and p27−/− cells (CD45.1) were co-transplanted. Interestingly, the p18−/− genotype gradually dominated the p27−/− genotype in multiple hematopoietic lineages and p18−/− HSCs showed 4-5 times more LTRA than p27−/− HSCs 12 months after cBMT. Further self-renewal potential of HSCs was examined with secondary transplantation in which primarily transplanted p18−/− or p27−/− cells were equally mixed with wild-type unmanipulated cells. Notably, while the p18−/− cells continued to outcompete the wild-type cells as we previously observed, the p27−/− cells did not behave so in the secondary recipients. Based on the flow cytometric measurement and bone marrow cellularity, we estimated that transplanted p18−/− HSCs (defined with the CD34−LKS immunophenotype) had undergone a 230-fold expansion, while transplanted p27−/− and wild-type HSCs had only achieved a 6.6- and 2.4-fold expansion in the secondary recipients respectively. We further calculated the yield of bone marrow nucleated cells (BMNCs) per HSC. There were approximately 44 x 103, 20.6 x 103, and 15 x 103 BMNCs generated per CD34−LKS cell in p18−/−, p27−/− and wild-type transplanted recipients respectively. Therefore, the dramatic expansion of p18−/− HSCs in the hosts was not accompanied by decreased function per stem cell. Our current study demonstrates that hematopoietic engraftment in the absence of p18 is more advantageous than that in the absence of p27, perhaps due to a more specific role of p18 on self-renewal of the long-term repopulating HSCs.