ROS-low HSCs are repopulation-high

In this issue, Jang and Sharkis show that hemopoietic stem cells (HSCs) can be fractionated into 2 major subpopulations based on the cellular content of reactive oxygen species (ROSs). Low ROSs predicted a high repopulating capacity along with multipotency, whereas high ROS content predicted relatively poorer HSC functions.

Attempts at precisely identifying rare HSCs within the hemopoietic heterogeneity have often been frustrating; many markers are shared by mature cells, but more importantly the most rigorous methods that separate HSCs from other cells yield a minuscule fraction that is nevertheless variable in terms of self-renewal.¹  It is of paramount importance, therefore, to redefine the stem state and discover alternative HSC characterization tools (reviewed by Zipori² ). In the Jang and Sharkis study, a novel cell feature, ROS content, is found to discriminate between mouse HSC subtypes. ROS^low HSCs have long-term repopulating capacity upon serial transplantations in irradiated recipients and are also multipotent. Conversely, the ROS^high subpopulation declines upon repeated transplantation and is partially lineage restricted. Importantly, the better functional performance of ROS^low cells corresponded to several properties ascribed to long-term repopulating HSCs, such as quiescence and capacity to interact with niche osteoblasts. By contrast, the stem-cell phenotype CD34⁻LSK was common to the ROS^high and ROS^low populations, highlighting the limitation of cell surface marker usage. The superior repopulating capacity of ROS^low cells suggests that they might be the HSCs found in the reduced oxygen sites within the bone marrow that constitute the stem-cell niches. Indeed, ROS^low cells exhibited a relatively higher adhesion capacity to osteoblastic cells. It is, however, yet to be demonstrated that ROS^low cells do indeed reside within the osteoblastic bone marrow niche in situ.

Stem-cell niches in general are constructed in such a way that the stem cell physically interacts with stromal cells that express antagonists of differentiation. The latter interfere with the activity of differentiation inducing cytokines to allow self-renewal. This was first demonstrated for mammalian HSCs³  and later substantiated by studies on Drosophila gonadal stem cells.⁴  Other components of the stem-cell niche include chemokine instructions and adhesion interactions. It has been further demonstrated that hypoxic areas within the bone marrow recruit hemopoietic progenitor cells.⁵  The present study raises the possibility that it is the abundance of ROSs within the stem cell, rather than in the microenvironment, that contributes to the maintenance of the stem state. These findings raise several questions that should be further experimentally examined: What is the molecular mechanism the controls ROSs in stem cells? Is the formation of ROSs in stem cells instructed by the stromal niche? Are ROS^high cells a differentiated product of ROS^low cells? Pharmacological inhibition of p38 or mTOR, which are found in elevated levels in ROS^high cells, restored their performance in LTC-IC culture assays. The latter are regarded as an in vitro corollary of in vivo repopulation tests. This suggests the need to develop strategies to convert ROS^high cells into ROS^low species, with the aim of improving HSC transplantation and subsequent engraftment in human patients.