Proliferative behavior of hematopoietic stem cells revisited: No evidence for mitotic memory

Abstract

The proliferative activity of adult hematopoietic stem cells (HSCs) is controversially discussed. Inducible fluorescent histone 2B fusion protein (H2B-FP) transgenic mice are important tools for tracking the mitotic history of murine HSCs in label dilution experiments. A recent study proposed that the most primitive HSCs divide only four times, to then enter permanent quiescence. We observed that background fluorescence due to leaky H2B-FP expression, occurring in all H2B-FP transgenes independent of label induction, accumulated with age in primitive HSCs with high repopulation potential. We argue that this background had been misinterpreted as retention of induced label and permanent quiescence. We found cell division-independent half-lives of H2B-FPs to be short, which had led to overestimation of HSC divisional activity. Our data do not support HSC mitotic memory and entry into permanent quiescence after few divisions, but show that primitive HSCs of adult mice continue to cycle rarely.In the present study, we re-evaluated H2B-FP retention. We show that in three different H2B-FP transgenic mouse strains, the most quiescent HSC sub-population inevitably accumulates high background fluorescence in an age-dependent manner. Accordingly, long-term serial repopulation activity was enriched among cells with high background. We estimated divisionindependent degradation of H2B-GFP and H2B-RFP to proceed in HSCs with half-lives of approximately 4-6 and 2 weeks, respectively, and show that neglect of H2B-FP decay leads to overestimation of HSC mitotic activity. Two sequential rounds of H2B-RFP induction and subsequent dilution revealed that HSCs do not abruptly halt mitotic activity upon ageing, arguing against permanent quiescence of HSCs after four divisions as previously hypothesized (Bernitz et al., 2016). Mathematical modeling of H2B-FP systems revealed that background label accumulates over time in the most primitive and quiescent HSCs, providing an explanation for observing seemingly label-retaining HSCs after extended chase periods.We argue that such cells were previously misinterpreted as permanently quiescent HSCs.

Modeling the Dynamics of T-Cell Development in the Thymus

“…Thus, using RAG1 reporter knock-in or transgenic strains, dilution of GFP serves as surrogate for proliferation after termination of TCR gene rearrangement [ 60 , 100 , 101 ]. To overcome the need for normalization to correct for degradation of GFP in this experimental system, the half-life of GFP has been prolonged to weeks or even months by fusing it to histone 2B [ 102 , 103 ]. Such fusions have been used to generate Tcrd-H2B-GFP mice to label T cells [ 104 ].…”

Modelling the Dynamics of T-Cell Development in the Thymus

“…Thus, using RAG1 GFP reporter knock-in or transgenic strains, dilution of GFP serves as surrogate for proliferation after termination of TCR gene rearrangement [60,100,101]. To overcome the need for normalization to correct for degradation of GFP in this experimental system, the half-life of GFP has been prolonged to weeks or even months by fusing it to histone 2B [102,103]. Such fusions have been used to generate Tcrd-H2B-GFP mice to label γδT cells [104].…”