A low level of reactive oxygen species selects for primitive hematopoietic stem cells that may reside in the low-oxygenic niche

Abstract

A low-oxygenic niche in bone marrow limits reactive oxygen species (ROS) production, thus providing long-term protection for hematopoietic stem cells (HSCs) from ROS stress. Although many approaches have been used to enrich HSCs, none has been designed to isolate primitive HSCs located within the low-oxygenic niche due to difficulties of direct physical access. Here we show that an early HSC population that might reside in the niche can be functionally isolated by taking advantage of the relative intracellular ROS activity. Many attributes of primitive HSCs in the low-oxygenic osteoblastic niche, such as quiescence, and calcium receptor, N-cadherin, Notch1, and p21 are higher in the ROS(low) population. Intriguingly, the ROS(low) population has a higher self-renewal potential. In contrast, significant HSC exhaustion in the ROS(high) population was observed following serial transplantation, and expression of activated p38 mitogen-activated protein kinase (MAPK) and mammalian target of rapamycin (mTOR) was higher in this population. Importantly, treatment with an antioxidant, a p38 inhibitor, or rapamycin was able to restore HSC function in the ROS(high) population. Thus, more potent HSCs associated with the low-oxygenic niche can be isolated by selecting for the low level of ROS expression. The ROS-related signaling pathways together with specific characteristics of niche HSCs may serve as targets for beneficial therapies.

Figure 1

Isolation, self-renewal, and differentiation of the ROS^low and ROS^high subsets. (A) Isolation of ROS^low and ROS^high populations. ROS^low or ROS^high Lin⁻CD45⁺AnV⁻ marrow cells were isolated using DCF-DA cell-permeable intracellular ROS indicator. R2 indicates ROS^low; R5, ROS^high. (B) LTC-ICs at 6 weeks (top), CFU-Cs at 10 days (middle and bottom). (C) Competitive repopulation assay with 10 to 2500 cells. A total of 10 mice for each cell dose were used for the transplantation in 2 independent experiments. (D) Multilineage reconstitution potential (left). Myeloid differentiation skewing of the ROS^high population (right). Values represent the means plus or minus SEM. *P < .05; **P < .01.

Figure 2

Association of the ROS^low subsets with properties defining osteoblastic niche–derived HSCs. (A) Flow cytometry histograms for CaR, N-cadherin, Bcrp, TERT, and mTOR. The dotted gray lines are isotype controls. (B) Western analysis for Notch1, p53, and p16. (C) G₀ activity and p21 quantitative PCR. Values represent the means plus or minus SEM. *P < .05. (D) Stem cell surface marker analysis. The variability of CD34⁻LSK cell frequency in 14 independent experiments. The lines represent the means.

Figure 3

Exhaustion of HSCs in serial transplantation of the ROS^high population and the decreased adherence to the specific niche components. Reconstitution capacity in secondary (A) and tertiary (B) transplantation. A total of 5 to 10 mice for each condition were used as recipients. (C) Cell adhesion assay. Values represent the means plus or minus SEM and are significant at *P < .05 and **P < .01.

Figure 4

Mechanisms of HSC activation and exhaustion. (A) Phosphorylated p38 MAPK activity; (B) LTC-IC activity with 100 μM NAC, 10 μM p38 inhibitor, 10 μM JNK inhibitor, 10 μM MEK inhibitor, and 1 μM rapamycin; ROS level (C); and phosphorylated p38 and mTOR levels (D) of viable Lin⁻CD45⁺ marrow cells after in vivo antioxidant treatment. Values represent the means plus or minus SEM. *P < .05; **P < .01.

Figure 5

A schematic summary of LT-HSC activation in the niche. LT-HSC distribution and the characteristics in adult bone marrow niche of normal mice. PP blood indicates peripheral blood.