E47 regulates hematopoietic stem cell proliferation and energetics but not myeloid lineage restriction

Abstract

The immune system is replenished by self-renewing hematopoietic stem cells (HSCs) that produce multipotent progenitors (MPPs) with little renewal capacity. E-proteins, the widely expressed basic helix-loop-helix transcription factors, contribute to HSC and MPP activity, but their specific functions remain undefined. Using quantitative in vivo and in vitro approaches, we show that E47 is dispensable for the short-term myeloid differentiation of HSCs but regulates their long-term capabilities. E47-deficient progenitors show competent myeloid production in short-term assays in vitro and in vivo. However, long-term myeloid and lymphoid differentiation is compromised because of a progressive loss of HSC self-renewal that is associated with diminished p21 expression and hyperproliferation. The activity of E47 is shown to be cell-intrinsic. Moreover, E47-deficient HSCs and MPPs have altered expression of genes associated with cellular energy metabolism, and the size of the MPP pool but not downstream lymphoid precursors in bone marrow or thymus is rescued in vivo by antioxidant. Together, these observations suggest a role for E47 in the tight control of HSC proliferation and energy metabolism, and demonstrate that E47 is not required for short-term myeloid differentiation.

Figure 1

E47-deficient HSCs show efficient myeloid differentiation under pathogen-free conditions and after stimulation with LPS. (A-B) Flk2⁻ LSKs from E47 HET or KO mice were cultured at 300 cells per well in 96-well plates for 72 hours in the presence or absence of 10 μg/mL LPS. Cells were harvested, counted, and stained with lineage specific antibodies. The number and frequency of CD11b⁺ myeloid cells were measured (n = 9 wells). **P < .05. ns indicates not significant. The data are representative of 2 independent experiments. (C) The expression of CD14 on flk2⁻ LSKs and flk2⁺ LSKs from WT and KO littermates was examined by flow cytometric analysis. The data are representative of 2 independent experiments. (D) flk2⁻ LSKs from E47 WT or KO mice were cultured in MethoCult at 1, 2, and 5 cells per well in 96-well plates for 7 days in the presence or absence of 10 μg/mL LPS. Wells with colony-forming unit-granulocyte, macrophage colonies were scored positive. The frequency of colony-forming cells was calculated according to Poisson statistics (n = 48 wells per group). The data are representative of 2 independent experiments.

Figure 2

E47-deficient bone marrow cells display compromised long-term competitive repopulation activity in vivo. (A) CD45.2 WT and E47 KO bone marrow cells (2 × 10⁵) mixed with an equal number of CD45.1 competitor cells were adoptively transferred into lethally irradiated CD45.1 recipient mice. The proportion of test donor-derived (CD45.2⁺) Gr-1⁺cells was monitored every 4 weeks in the peripheral blood of the recipient mice. Then, the recipients were killed at 16 weeks after transplantation, and lineage repopulation in the spleen (Gr-1⁺, CD11c⁺) or bone marrow (flk2⁻ LSK) was measured (n = 5 mice). P < .05. The data are representative of 2 independent experiments. (B) Competitive lymphoid repopulation activity of WT and E47 KO bone marrow cells was measured at 1:1 test donor/competitor ratio. The percentage of test donor-derived (CD45.2⁺) lymphoid cells in the peripheral blood of the recipient mice at 16 weeks after transplantation is shown (n = 8 mice). PB indicates peripheral blood; SPL, spleen; and BM, bone marrow.

Figure 3

E47-deficient HSCs display poor self-renewal efficiency in vivo. Serial transplantation was performed to examine the long-term self-renewal efficiency of HSCs from WT and E47 KO mice. The bone marrow cells from WT and E47 KO littermates were first adoptively transferred into primary recipients for 16 weeks and then serially transferred into secondary recipients. Blood reconstitution was measured monthly. Recipients were killed at 16 weeks after transplantation, and lineage reconstitution in the spleen was examined. Myeloid or lymphoid reconstitution efficiency was indicated by the number of donor-derived CD45.2⁺ Gr-1⁺ cells (A) or CD3⁺ cells (B), respectively (n = 6 to 8 mice per group). **P < .05. ns indicates not significant.

Figure 4

E47 null HSCs show normal homing, niche engraftment, and apoptosis. (A) CD45.2 WT or E47 KO bone marrow cells (2 × 10⁶) were injected into lethally irradiated CD45.1 recipient mice. The number of donor-derived flk2⁻ LSKs was examined at 2 weeks (n = 3 mice) and at 16 weeks (n = 8 mice) after transplantation. (B) Lethally irradiated CD45.1 recipient mice reconstituted with E47 KO or WT CD45.2 donor bone marrow cells were killed at 12 weeks after transplantation. The bone marrow cells from these recipients were stained with antibodies to resolve donor-derived CD45.2 flk2⁻ LSKs and then labeled with annexin V and DAPI for apoptosis analysis. (Left panels) Representative flow cytometry profiles used to generate the bar graph on the right (n = 3 mice). **P < .05. ns indicates not significant.

Figure 5

E47 null HSCs displayed hyperproliferation under steady state and after transplantation stress. (A) Flk2⁻ LSKs from WT and E47 KO littermates were sorted by flow cytometry, and the expression of p21 and β-actin was examined by quantitative reverse-transcription PCR. The data are presented as KO/WT ratios for each transcript (n = 4 independent sorts). **P < .05. (B) Bone marrow cells from WT and E47 KO littermates were stained with cell surface antibodies to resolve HSC-enriched flk2⁻ LSKs, and the number of flk2⁻ LSKs was counted (n = 11 mice). (C) Surface-stained flk2⁻ LSKs from E47 WT or KO mice were fixed and then stained with antibodies to the Ki67 proliferation antigen and DAPI for cell-cycle analysis. (Left panels) Representative flow cytometric profiles used to generate the bar graph on the right (n = 3 or 4 mice). **P < .05. (D) Lethally irradiated mice reconstituted with E47 KO or WT CD45.2 donor bone marrow cells were killed at 3 weeks after transplantation. A total of 100 μg BrdU was injected into recipient mice at a 12-hour interval for 24 hours before death. The donor-derived CD45.2 flk2⁻ LSKs were fixed and stained with antibodies to BrdU, Ki67, or DAPI for proliferation and cell-cycle analysis (n = 4 mice). **P < .05. ns indicates not significant. (E) Bone marrow from E47 KO (CD45.2) and WT (CD45.1/2) mice was cotransferred into lethally irradiated CD45.1 hosts, and cell-cycle status was examined as in panel D at 2 weeks after transplantation. (n = 6 mice). P < .05.

Figure 6

Quantitative analysis of long-term HSC defects in E47-deficient mice. (A) Long-term culture-initiating cell assay was performed with double-sorted flk2⁻ LSKs from WT and E47 KO littermates to determine the frequency of long-term colony-forming cells. Plotted is the percentage of wells that did not give rise to colonies after a single plating (left panel) or serial replating (right panel) at the indicated input cell numbers. The frequency of long-term colony-forming cells was calculated according to Poisson statistics. (B) Limit dilution doses (0.67 × 10⁵, 0.22 × 10⁵, and 0.073 × 10⁵) of CD45.2 WT and E47 KO bone marrow cells mixed with a constant number (2 × 10⁵) of CD45.1 competitor cells were adoptively transferred into lethally irradiated CD45.1⁺ recipient mice. The graph depicts the percentage of CD45.1 recipient mice that had less than 1% of donor CD45.2⁺ Gr-1⁺ cells at 16 weeks after adoptive transfer of the indicated donor cell numbers. The frequency of functional HSCs was calculated using Poisson statistics. Eight recipient mice were used at each cell dose per genotype. **P < .05.

Figure 7

Antioxidant restoration of E47-deficient MPPs. (A) Quantitative PCR analysis of redox-associated genes in purified HSCs (flk2⁻CD150⁺CD48⁻ LSKs) or MPPs (flk2⁺ LSKs) from E47 KO mice compared with E47 WT mice. Data are mean ± SD of triplicate wells, representative of 2 or 3 independent sorts. *P < .05. (B) Mice treated with 1 mg/mL N-acetyl cysteine in the drinking water were killed after 14 days, and lymphoid organs stained to detect bone marrow multipotent progenitors (flk2⁻ and flk2⁺ LSKs), common lymphoid progenitors (CLPs: lin⁻IL7R⁺AA4.1⁺Sca-l^lo), and early thymic progenitors precursors (ETPs: lin⁻CD44⁺CD25⁺c-kit⁺IL-7R⁻). (C) Quantitation of hematopoietic subsets in B (n = 5 mice/group). *P < .05. ns indicates not significant.