Hoxb4-deficient mice undergo normal hematopoietic development but exhibit a mild proliferation defect in hematopoietic stem cells

Enforced expression of Hoxb4 dramatically increases the regeneration of murine hematopoietic stem cells (HSCs) after transplantation and enhances the repopulation ability of human severe combined immunodeficiency (SCID) repopulating cells. Therefore, we asked what physiologic role Hoxb4 has in hematopoiesis. A novel mouse model lacking the entire Hoxb4 gene exhibits significantly reduced cellularity in spleen and bone marrow (BM) and a subtle reduction in red blood cell counts and hemoglobin values. A mild reduction was observed in the numbers of primitive progenitors and stem cells in adult BM and fetal liver, whereas lineage distribution was normal. Although the cell cycle kinetics of primitive progenitors was normal during endogenous hematopoiesis, defects in proliferative responses of BM Lin(-) Sca1(+) c-kit(+) stem and progenitor cells were observed in culture and in vivo after the transplantation of BM and fetal liver HSCs. Quantitative analysis of mRNA from fetal liver revealed that a deficiency of Hoxb4 alone changed the expression levels of several other Hox genes and of genes involved in cell cycle regulation. In summary, the deficiency of Hoxb4 leads to hypocellularity in hematopoietic organs and impaired proliferative capacity. However, Hoxb4 is not required for the generation of HSCs or the maintenance of steady state hematopoiesis.     

Non-side-population hematopoietic stem cells in mouse bone marrow

Most hematopoietic stem cells (HSCs) are assumed to reside in the so-called side population (SP) in adult mouse bone marrow (BM). We report the coexistence of non-SP HSCs that do not significantly differ from SP HSCs in numbers, capacities, and cell-cycle states. When stained with Hoechst 33342 dye, the CD34(-/low) c-Kit(+)Sca-1(+)lineage marker(-) (CD34(-)KSL) cell population, highly enriched in mouse HSCs, was almost equally divided into the SP and the main population (MP) that represents non-SP cells. Competitive repopulation assays with single or 30 SP- or MP-CD34(-)KSL cells found similar degrees of repopulating activity and frequencies of repopulating cells for these populations. Secondary transplantation detected self-renewal capacity in both populations. SP analysis of BM cells from primary recipient mice suggested that the SP and MP phenotypes are interconvertible. Cell-cycle analyses revealed that CD34(-)KSL cells were in a quiescent state and showed uniform cell-cycle kinetics, regardless of whether they were in the SP or MP. Bcrp-1 expression was similarly detected in SP- and MP-CD34(-)KSL cells, suggesting that the SP phenotype is regulated not only by Bcrp-1, but also by other factors. The SP phenotype does not specify all HSCs; its identity with stem cell function thus is unlikely.     

Selection based on CD133 and high aldehyde dehydrogenase activity isolates long-term reconstituting human hematopoietic stem cells

The development of novel cell-based therapies requires understanding of distinct human hematopoietic stem and progenitor cell populations. We recently isolated reconstituting hematopoietic stem cells (HSCs) by lineage depletion and purification based on high aldehyde dehydrogenase activity (ALDH(hi)Lin- cells). Here, we further dissected the ALDH(hi)-Lin- population by selection for CD133, a surface molecule expressed on progenitors from hematopoietic, endothelial, and neural lineages. ALDH(hi)CD133+Lin- cells were primarily CD34+, but also included CD34-CD38-CD133+ cells, a phenotype previously associated with repopulating function. Both ALDH(hi)CD133-Lin- and ALDH(hi)CD133+Lin- cells demonstrated distinct clonogenic progenitor function in vitro, whereas only the ALDH(hi)CD133+Lin- population seeded the murine bone marrow 48 hours after transplantation. Significant human cell repopulation was observed only in NOD/SCID and NOD/SCID beta2M-null mice that received transplants of ALDH(hi)CD133+Lin- cells. Limiting dilution analysis demonstrated a 10-fold increase in the frequency of NOD/SCID repopulating cells compared with CD133+Lin- cells, suggesting that high ALDH activity further purified cells with repopulating function. Transplanted ALDH(hi)CD133+Lin- cells also maintained primitive hematopoietic phenotypes (CD34+CD38-) and demonstrated enhanced repopulating function in recipients of serial, secondary transplants. Cell selection based on ALDH activity and CD133 expression provides a novel purification of HSCs with long-term repopulating function and may be considered an alternative to CD34 cell selection for stem cell therapies.     

Survival and retrovirus infection of murine hematopoietic stem cells in vitro: effects of 5-FU and method of infection.

Gene replacement therapy for diseases of the hematopoietic system requires efficient gene transfer to pluripotent hematopoietic stem cells. We have systematically compared a number of protocols for retrovirus-mediated gene transfer into murine repopulating hematopoietic stem cells. Recipients of infected bone marrow cells were analyzed for the presence of the transduced provirus 4 months after transplantation. Our results show that 5-fluorouracil (5-FU) pretreatment of donor animals was required for efficient gene transfer and that 5-FU-treated bone marrow retained more repopulating activity in culture than untreated bone marrow. A comparison of retrovirus-mediated gene transfer by co-cultivation of bone marrow cells with retrovirus producer cells as opposed to gene transfer by culturing bone marrow cells in retrovirus-containing supernatant revealed that gene transfer by cocultivation was more efficient than supernatant infection. However, the repopulating ability of bone marrow cells cocultured with retrovirus producer cells was reduced compared to bone marrow cells cultured in virus-containing medium.     

Hematopoietic senescence is postponed and hematopoietic stem cell function is enhanced by dietary restriction

OBJECTIVE: The aim of this study was to test dietary restriction (DR) as an intervention to alleviate senescence-associated functional defects in hematopoietic stem cells (HSCs). MATERIALS AND METHODS: BALB/cByJ (BALB) mice were fed ad libitum (AL) or were diet restricted (DR) to 75% of the AL food intake after 1 month of age. Peripheral blood and bone marrow cell compositions were compared in 3- and 25-month-old AL (AL-3, AL-25) mice and in 25-month-old DR (DR-25) mice using fluorescence-activated cell staining. Relative HSC functions in vivo were compared using competitive repopulation, and were also tested in 6-month-old BALB mice to measure the effects of short-term DR. RESULTS: Compared to AL-3, AL-25 blood had significantly lower levels of red blood cells and hemoglobin. AL-25 marrow contained less than half the concentration of Lin(-)CD34(-)Sca1(+)CD117(+) HSCs and showed only half the in vivo functional ability of AL-3 marrow. In vivo, AL-25 HSCs failed to produce the strong correlations over time that demonstrate clonal stability during competitive repopulation. These correlations were shown in AL-3 HSCs. DR for 24 months alleviated hematopoietic deficiencies in the blood, increased concentrations of bone marrow Lin(-)CD34(-)Sca1(+)CD117(+) HSCs and improved HSC functional abilities in DR-25 mice to values far greater than those in normally aged mice. Surprisingly, HSC function in 25-month-old DR mice was better than that in young adults. Degrees of recipient repopulation by HSCs from DR-25 mice also correlated well over time, demonstrating clonal stability. Short-term DR for 5 months also improved HSC function, but to a much smaller degree. CONCLUSIONS: Aged BALB mice show hematopoietic and HSC senescence and clonal succession. Lifelong DR slows hematopoietic senescence and prevents HSC aging.     

Hematopoietic capability of CD34+ cord blood cells: a comparison with CD34+ adult bone marrow cells

The characteristics of hematopoietic progenitor and stem cell (HPC/HSC) populations in mammals vary according to their ontogenic stage. In humans, HPC/HSCs from umbilical cord blood (CB) are increasingly used as an alternative to HPC/HSCs from adult bone marrow (BM) for the treatment of various hematologic disorders. How the hematopoietic activity of progenitor and stem cells in CB differs from that in adult BM remains unclear, however. We compared CD34+ cells, a hematopoietic cell population, in CB with those in adult BM using phenotypic subpopulations analyzed by flow cytometry, the colony-forming activity in methylcellulose clonal cultures, and the repopulating ability of these cells in NOD/Shi-scid (NOD/SCID) mice. Although the proportion of CD34+ cells was higher in adult BM than in CB mononuclear cells, the more immature subpopulations, CD34+ CD33- and CD34+ CD38- cells, were present in higher proportions in CD34+ CB cells. Clonal culture assay showed that more multipotential progenitors were present in CD34+ CB cells. When transplanted into NOD/SCID mice. CD34+ adult BM cells could not reconstitute human hematopoiesis in recipient BM, but CD34+ CB cells achieved a high level of engraftment, indicating that CD34+ CB cells possess a greater repopulating ability. These results demonstrated that human hematopoiesis changes with development from fetus to adult. Furthermore, CD34+ CB cells contained a greater number of primitive hematopoietic cells, including HSCs, than did adult BM, suggesting the usefulness of CD34+ CB cells not only as a graft for therapeutic HSC transplantation but also as a target cell population for ex vivo expansion of transplantable HSCs and for gene transfer in gene therapy.     

Loss of expression of the Hoxa-9 homeobox gene impairs the proliferation and repopulating ability of hematopoietic stem cells

The homeobox gene Hoxa-9 is normally expressed in primitive bone marrow cells, and overexpression of Hoxa-9 markedly expands hematopoietic stem cells, suggesting a function in early hematopoiesis. We present evidence for major functional defects in Hoxa-9-/- hematopoietic stem cells. Hoxa-9-/- marrow cells have normal numbers of immunophenotypic stem cells (Lin(-)c-kit(+)flk-2(-)Sca-1+ [KLFS] cells). However, sublethally irradiated Hoxa-9-/- mice develop persistent pancytopenia, indicating unusual sensitivity to ionizing irradiation. In competitive transplantation assays, Hoxa-9-/- cells showed an 8-fold reduction in multilineage long-term repopulating ability, a defect not seen in marrow cells deficient for the adjacent Hoxa-10 gene. Single-cell cultures of KLFS cells showed a 4-fold reduction in large high-proliferation potential colonies. In liquid cultures, Hoxa-9-deficient Lin(-)Sca-1(+) cells showed slowed proliferation (a 5-fold reduction in cell numbers at day 8) and delayed emergence of committed progenitors (a 5-fold decrease in colony-forming cells). Slowing of proliferation was accompanied by a delay in myeloid maturation, with a decrease in Gr-1hiMac-1hi cells at the end of the culture. Retroviral transduction with a Hoxa-9 expression vector dramatically enhanced the cytokine-driven proliferation and in vivo engraftment of Hoxa-9-/- marrow cells. Hoxa-9 appears to be specifically required for normal hematopoietic stem cell function both in vitro and in vivo.     

Bone marrow collected 14 days after in vivo administration of granulocyte colony-stimulating factor and stem cell factor to mice has 10-fold more repopulating ability than untreated bone marrow

We have examined the repopulating ability of bone marrow and peripheral blood cells collected immediately and at intervals after treatment of donor mice with the combination of granulocyte colony-stimulating factor (G-CSF) and stem cell factor (SCF). Using a competitive repopulation assay we showed that the repopulating ability of peripheral blood cells was highest immediately after cytokine treatment and declined to normal levels within 6 weeks of the termination of treatment with G-CSF and SCF. In contrast the repopulating ability of bone marrow cells was low immediately after cytokine treatment and increased to levels that were 10-fold or more greater than marrow from untreated mice by 14 days after termination of treatment with G-CSF and SCF. This high level of repopulating activity declined to normal levels by 6 weeks after termination of treatment with G-CSF and SCF. The high level of repopulating ability was confirmed by injecting cells from G- CSF- and SCF-treated donors into unconditioned recipients. Peripheral blood cells collected immediately after treatment with G-CSF and SCF engrafted into unconditioned mice sevenfold better than an equivalent number of bone marrow cells from untreated mice. Likewise, bone marrow cells collected 14 days after treatment of the donor animal with G-CSF and SCF engrafted at 10-fold higher levels than an equivalent number of bone marrow cells from untreated mice. We conclude that the treatment of donor mice with G-CSF and SCF causes a transient increase in the repopulating ability of peripheral blood and later of bone marrow. These observations may have applications to clinical hematopoietic stem cell transplantation.     

Hematopoietic stem cells expanded by fibroblast growth factor-1 are excellent targets for retrovirus-mediated gene delivery

OBJECTIVE: For the study of the function of genes of interest in hematopoietic stem cells (HSCs) and for successful gene therapy, it is fundamental to develop a method of efficient gene transfer into HSCs. In mice experiments, efforts have been made to raise the transduction efficiency by modifying the vectors, administrating 5-fluorouracil (5-FU) to donor mice, selecting cytokine cocktails to better sustain the long-term repopulating potential of the stem cells, and so on. The objective of this study is to examine whether the use of fibroblast growth factor-1 (FGF-1)-expanded bone marrow cells provide an improved source for retroviral gene delivery to HSCs. MATERIALS AND METHODS: Unfractionated bone marrow cells from one mouse were cultured in serum-free medium containing FGF-1. Both floating and attached cells were transferred to retronectin precoated dishes and infected with virus supernatant from MP34 cells stably transduced with pMY/GFP retrovirus. After 3-day infection, the green fluorescence protein-positive fraction was sorted and the cells were transplanted to lethally irradiated mice. RESULTS: The experiments illustrated that the number of bone marrow-derived competitive repopulation units (CRUs) was increased from 600 to 9300 per mouse after a 3-week culture period with FGF-1. Following retroviral transduction of the expanded cells, the absolute number of sorted retrovirus-transduced CRUs was 4200. Using these retrovirus-transduced cells in noncompetitive reconstitution assay, we achieved radiation protection and long-term bone marrow reconstitution in 100% of the recipients with average myeloid and lymphoid chimerisms of 70% and 50%, respectively, even if we transplanted 150 recipients with cells derived from a single donor mouse. CONCLUSION: In conclusion, FGF-1-expanded bone marrow cells constitute an excellent source of stem cells that could be used in a range of gene delivery protocols.     

Expansion of hematopoietic stem cells in the developing liver of a mouse embryo

The activity of hematopoietic stem cells in the developing liver of a C57BL/6 mouse embryo was quantified by a competitive repopulation assay. Different doses of fetal liver cells at days 11 to 18 of gestation were transplanted into irradiated mice together with 2 x 10(5) adult bone marrow cells. A long-term repopulation in myeloid-, B-cell, and T-cell lineage by fetal liver cells was evaluated at 20 weeks after transplantation. At day 12 of gestation multilineage repopulating activity was first detected in the liver as 50 repopulating units (RU) per liver. The number of RU per liver increased 10-fold and 33-fold by day 14 and day 16 of gestation, and decreased thereafter, suggesting a single wave of stem cell development in the fetal liver. A limiting dilution analysis revealed that the frequency of competitive repopulating units (CRU) in fetal liver cells at day 12 of gestation was similar to that at day 16 of gestation. Because of an increase of total fetal liver cell number, the absolute number of CRU per liver from days 12 to 16 of gestation increased 38-fold. Hence, the mean activity of stem cells (MAS) that is given by RU per CRU remained constant from days 12 to 16 of gestation. From these data we conclude that hematopoietic stem cells expand in the fetal liver maintaining their level of repopulating potential.     

Hematopoietic stem cell and progenitor defects in Sca-1/Ly-6A-null mice

Despite its wide use as a marker for hematopoietic stem cells (HSCs), the function of stem cell antigen-1 (Sca-1) (also known as lymphocyte activation protein-6A [Ly-6A]) in hematopoiesis remains poorly defined. We have previously established that Sca-1(-/-) T cells develop normally, although they are hyperresponsive to antigen. Here, we report detailed analysis of hematopoiesis in Sca-1-deficient animals. The differentiation potential of Sca-1-null bone marrow was determined from examination of the most mature precursors (culture colony-forming units [CFU-Cs]) to less committed progenitors (spleen CFUs [CFU-Ss]) to long-term repopulating HSCs. Sca-1-null mice are mildly thrombocytopenic with a concomitant decrease in megakaryocytes and their precursors. Bone marrow cells derived from Sca-1(-/-) mice also have decreased multipotential granulocyte, erythroid, macrophage, and megakaryocyte CFU (GEMM-CFU) and CFU-S progenitor activity. Competitive repopulation assays demonstrated that Sca-1(-/-) HSCs are at a competitive disadvantage compared with wild-type HSCs. To further analyze the potential of Sca-1(-/-) HSCs, serial transplantations were performed. While secondary repopulations using wild-type bone marrow completely repopulated Sca-1(-/-) mice, Sca-1(-/-) bone marrow failed to rescue one third of lethally irradiated wild-type mice receiving secondary bone marrow transplants from irradiation-induced anemia and contributed poorly to the surviving transplant recipients. These data strongly suggest that Sca-1 is required for regulating HSC self-renewal and the development of committed progenitor cells, megakaryocytes, and platelets. Thus, our studies conclusively demonstrate that Sca-1, in addition to being a marker of HSCs, regulates the developmental program of HSCs and specific progenitor populations.     

Expansion of hematopoietic stem cell phenotype and activity in Trp53-null mice

OBJECTIVES: To study the effects of transformation-related protein 53 (Trp53) and other genes on hematopoiesis and hematopoietic stem cells (HSCs). METHODS: Frequencies of murine bone marrow cells (BMCs) with the Lin(-)Sca-1(+)c-kit(+)CD34- phenotype were analyzed by flow cytometry, and were increased in mice with germ-line deletion of the Trp53 (Trp53(-/-)) gene but not in 25 other deletions of genes involved in cell cycling, development, cancer, or hematopoiesis. Therefore, Trp53(-/-) and wild-type Trp53(+/+) mice were compared using the following assays: complete blood counts, day-9 colony-forming unit spleen (CFU-S), and competitive repopulation. In the latter assay, donor repopulating ability was analyzed at one, three, and five months, while recipient survival and recipient blood and bone marrow cell composition were analyzed at five months, after transplantation. RESULTS: In comparison to wild-type controls, Trp53(-/-) mice had normal blood and bone marrow cell counts, increased CD11b(+), and decreased CD45R(+) cell proportions in blood and bone marrow, twice as many Lin(-)Sca-1(+)c-kit(+)CD34(-) BMCs, and 37% more day-9 CFU-S. In the competitive repopulation assay, Trp53(-/-) BMCs engrafted lethally irradiated recipients two to four times better than Trp53(+/+) BMCs. The Trp53(-/-) engraftment advantage increased with time in the recipients. Recipients of Trp53(-/-) donors had two to three times more Lin(-)Sca-1(+)c-kit(+)CD34(-) BMCs than recipients of Trp53(+/+) donors at five months after transplantation. However, only 44% of recipients of Trp53(-/-) donors survived five months after trans-plantation, compared with 92% of recipients of Trp53(+/+) donors. CONCLUSION: The Trp53-null allele expands bone marrow Lin(-)Sca-1(+)c-kit(+)CD34(-) cells and the overall activity of HSCs; however, it increases recipient mortality.     

Ex vivo culture with human brain endothelial cells increases the SCID-repopulating capacity of adult human bone marrow

Adult human bone marrow (ABM) is an important source of hematopoietic stem cells for transplantation in the treatment of malignant and nonmalignant diseases. However, in contrast to the recent progress that has been achieved with umbilical cord blood, methods to expand ABM stem cells for therapeutic applications have been disappointing. In this study, we describe a novel culture method that uses human brain endothelial cells (HUBECs) and that supports the quantitative expansion of the most primitive measurable cell within the adult bone marrow compartment, the nonobese diabetic/severe combined immunodeficient (NOD/SCID) repopulating cell (SRC). Coculture of human ABM CD34(+) cells with brain endothelial cells for 7 days supported a 5.4-fold increase in CD34(+) cells, induced more than 95% of the CD34(+)CD38(-) subset to enter cell division, and produced progeny that engrafted NOD/SCID mice at significantly higher rates than fresh ABM CD34(+) cells. Using a limiting dilution analysis, we found the frequency of SRCs within fresh ABM CD34(+) cells to be 1 in 9.9 x 10(5) cells. Following HUBEC culture, the estimated frequency of SRCs increased to 1 in 2.4 x 10(5) cells. All mice that received transplants of HUBEC-cultured cells showed B-lymphoid and myeloid differentiation, indicating that a primitive hematopoietic cell was preserved during culture. Noncontact HUBEC cultures also maintained SRCs at a level comparable to contact HUBEC cultures, suggesting that cell-to-cell contact was not required. These data demonstrate that human brain endothelial cells possess a unique hematopoietic activity that increases the repopulating capacity of adult human bone marrow.     

IGF binding protein 2 supports the survival and cycling of hematopoietic stem cells

The role of IGF binding protein 2 (IGFBP2) in cell growth is intriguing and largely undefined. Previously we identified IGFBP2 as an extrinsic factor that supports ex vivo expansion of hematopoietic stem cells (HSCs). Here we showed that IGFBP2-null mice have fewer HSCs than wild-type mice. While IGFBP2 has little cell-autonomous effect on HSC function, we found decreased in vivo repopulation of HSCs in primary and secondary transplanted IGFBP2-null recipients. Importantly, bone marrow stromal cells that are deficient for IGFBP2 have significantly decreased ability to support the expansion of repopulating HSCs. To investigate the mechanism by which IGFBP2 supports HSC activity, we demonstrated that HSCs in IGFBP2-null mice had decreased survival and cycling, down-regulated expression of antiapoptotic factor Bcl-2, and up-regulated expression of cell cycle inhibitors p21, p16, p19, p57, and PTEN. Moreover, we found that the C-terminus, but not the RGD domain, of extrinsic IGFBP2 was essential for support of HSC activity. Defective signaling of the IGF type I receptor did not rescue the decreased repopulation of HSCs in IGFBP2-null recipients, suggesting that the environmental effect of IGFBP2 on HSCs is independent of IGF-IR mediated signaling. Therefore, as an environmental factor, IGFBP2 supports the survival and cycling of HSCs.     

Hematopoietic stem cells express Tie-2 receptor in the murine fetal liver

Tie-2 receptor tyrosine kinase expressed in endothelial and hematopoietic cells is believed to play a role in both angiogenesis and hematopoiesis during development of the mouse embryo. This article addressed whether Tie-2 is expressed on fetal liver hematopoietic stem cells (HSCs) at day 14 of gestation. With the use of anti-Tie-2 monoclonal antibody, its expression was detected in approximately 7% of an HSC population of Kit-positive, Sca-1-positive, lineage-negative or -low, and AA4.1-positive (KSLA) cells. These Tie-2-positive KSLA (T(+) KSLA) cells represent 0.01% to 0.02% of fetal liver cells. In vitro colony and in vivo competitive repopulation assays were performed for T(+) KSLA cells and Tie-2-negative KSLA (T(-) KSLA) cells. In the presence of stem cell factor, interleukin-3, and erythropoietin, 80% of T(+) KSLA cells formed colonies in vitro, compared with 40% of T(-) KSLA cells. Long-term multilineage repopulating cells were detected in T(+) KSLA cells, but not in T(-) KSLA cells. An in vivo limiting dilution analysis revealed that at least 1 of 8 T(+) KSLA cells were such repopulating cells. The successful secondary transplantation initiated with a limited number of T(+) KSLA cells suggests that these cells have self-renewal potential. In addition, engraftment of T(+) KSLA cells in conditioned newborn mice indicates that these HSCs can be adapted equally by the adult and newborn hematopoietic environments. The data suggest that T(+) KSLA cells represent HSCs in the murine fetal liver. (Blood. 2000;96:3757-3762)