In Vivo divisional Tracking System Reveals Cycling Heterogeneity in Long-Term Hematopoietic Stem Cells.

Abstract 814

Hematopoietic stem cells (HSCs) are defined by their capacity to self-renew and give rise to all mature cells of hemato-lymphoid system for the lifetime of an individual. To ensure this, HSCs are kept at homeostatic levels in adult bone marrow. Steady-state HSC cycling kinetics have been evaluated by in vivo labeling assay using 5-bromo-2-deoxyuridine (BrdU) (Cheshier et. al., PNAS 1999; Kiel et al., Nature 2007), biotin (Nygren et. al., PLoS ONE 2008) and histon 2B-green fluorescent protein (H2B-GFP) transgenic model systems (Wilson et. al., Cell 2008; Foudi et. al., Nat. Biotech. 2008). Based on the latter, it was suggested that one HSC pool turns over faster than another, dormant pool with very limited divisions during a lifetime. However, the fast cycling HSCs did not have long-term multilineage reconstitution capacity in lethally irradiated animals in contrast to dormant HSCs (Wilson et. al., Cell 2008; Foudi et.al., Nat. Biotech. 2008). From these experiments remained unclear, whether the faster cycling HSC loose long-term repopulation potential according to divisional history, or whether they represent progenitors with limited self-renewal potential, sharing a long-term HSC phenotype. Therefore, the dynamics of steady-state long-term HSC homeostasis and blood production remains to be determined.

To address this directly, we set up an in vivo HSC divisional tracking assay. Here we show i.v. transfer of CFSE (carboxyfluorescein diacetate succinimidyl ester) -labeled HSCs into non-conditioned CD45.1/2 congenic F1 recipient mice that allows evaluation of steady-state HSC dynamics as CFSE distributes equally to daughter cells upon each cellular division. Sorted naïve CD4+CD62L+ T cells were used as non-dividing control cell population to determine the zero division CFSE staining level over time. Upon transfer of Lin-c-kit+Sca-1+ cells (LKS) into sublethally irradiated mice, all donor derived Lin-c-kit+ cells had divided >5 times after 3 weeks. However, transfer of LKS cells into non-irradiated mice revealed non-divided LKS cells in recipient bone marrow over 20 weeks. FACS analysis with HSC or progenitor specific marker expression showed that most of 0-2 time-divided and few of >5x divided LKS cells maintained a long-term HSC phenotype (CD150+, c-mpl+, CD34-). In order to test HSC potential, non- or >5x divided cells were sorted based on divisional history from primary recipients at different time points after transplantation, and competitively transplanted into lethally irradiated secondary recipients. At 3 weeks post primary transfer, single non-divided LKS cell was able to multi-lineage repopulate recipients, while 50 of >5x divided LKS cells showed no engraftment. Interestingly, both non- and >5x divided LKS cells at 7 or 12-14 weeks after primary transfer had long-term multilineage repopulating potential. Limiting dilution transplantation experiments demonstrated that HSC with long-term multilineage capacity (LT-HSC) were maintained at constant numbers that fit the numbers of free bone marrow niche space, with non-divided LT-HSC decreasing and >5x divided LT-HSC increasing with a constant division rate. We next tested the effects of hemato-immunological challenge on HSC cycling dynamics. Upon i.p. LPS injection into mice, previously transplanted with CFSE-labeled LKS, almost all LT-HSCs entered cell cycle within one week after challenge.

These findings directly demonstrate that some LT-HSCs are quiescent for up to one fifth of the life-time of a mouse, while other LT-HSCs divide more actively, thus proving asynchronous LT-HSC division and contribution to hematopoiesis in steady-state. In addition, the results demonstrate that quiescent LT-HSCs are driven into division in response to naturally-occurring hematopoietic challenges, such as systemic bacterial infection. The CFSE-tracking model established here now allows to directly test the role of intrinsic versus environmental cues on cycling-dynamics of HSCs as well as leukemia initiating cells in steady-state and upon challenge on multiple genetic and different species background.