当前位置: 首页 > 医学版 > 期刊论文 > 基础医学 > 免疫学杂志 > 2005年 > 第11期 > 正文
编号:11256784
IL-3 Induces Inhibitor of DNA-Binding Protein-1 in Hemopoietic Progenitor Cells and Promotes Myeloid Cell Development1
     Abstract

    Hemopoiesis depends on the expression and regulation of transcription factors, which control the maturation of specific cell lineages. We found that the helix-loop-helix transcription factor inhibitor of DNA-binding protein 1 (Id1) is not expressed in hemopoietic stem cells (HSC), but is increased in more committed myeloid progenitors. Id1 levels decrease during neutrophil differentiation, but remain high in differentiated macrophages. Id1 is expressed at low levels or is absent in developing lymphoid or erythroid cells. Id1 expression can be induced by IL-3 in HSC during myeloid differentiation, but not by growth factors that promote erythroid and B cell development. HSC were transduced with retroviral vectors that express Id1 and were transplanted in vivo to evaluate their developmental potential. Overexpression of Id1 in HSC promotes myeloid but impairs B and erythroid cell development. Enforced expression of Id1 in committed myeloid progenitor cells inhibits granulocyte but not macrophage differentiation. Therefore, Id1 may be part of the mechanism regulating myeloid vs lymphoid/erythroid cell fates, and macrophage vs neutrophil maturation.

    Introduction

    cellular process in which mature blood cells are continuously replenished from a pool of rare pluripotential hemopoietic stem cells (HSC)4 (1, 2, 3). This occurs by the sequential commitment of HSC into multipotential progenitors that give rise to precursor cells with more restricted potential, and finally into terminally differentiated blood cells. Although our understanding of the molecular mechanism(s) that regulate hemopoietic development has greatly increased, questions regarding the mechanism(s) that regulate self renewal and lineage commitment remain to be addressed. It is known that transcription factors are critical for the development of specific cell lineages, and function by regulating the expression of hemopoietic growth factor (HGF) receptors, other transcription factors, and lineage-specific genes (4, 5, 6, 7, 8, 9, 10). Targeted disruption of these genes results in impaired hemopoietic development or defects in the development of specific cell lineages. Furthermore, transcription factors are frequently deregulated in leukemias, resulting in enhanced cell proliferation and inhibition of terminal differentiation (5, 6, 7, 11, 12, 13, 14, 15, 16).

    To identify the molecular pathways that regulate early myeloid development, we compared transcription factor gene expression in multipotential erythroid myeloid lymphoid (EML) cells, which closely resemble normal HSC, and myeloid progenitor (MPRO) cells that resemble more committed MPROs that are roughly 40 h further along in differentiation (17, 18). Among the transcription factors that were differentially expressed by these two cell lines were inhibitor of DNA-binding proteins (Id) 1 and 2. Id proteins form inactive heterodimers with basic (b) helix-loop-helix (HLH) transcription factors, like E2A, rendering them unable to bind DNA and regulate gene expression (19, 20). The Id proteins (Id1-Id4) are a family of HLH transcription factors that regulate the growth and differentiation of muscle, nerve, endothelial, lymphoid, and other cell types (21, 22, 23, 24, 25, 26, 27, 28, 29). Further, Id1 mRNA is expressed in murine and human MPRO cell lines, however, its expression in normal HSC and committed myeloid, lymphoid, and erythroid progenitors, regulation by HGF, and its role in primitive myeloid development is currently not known (30, 31). Therefore, we initiated studies to determine 1) the relative expression levels of Id proteins in normal HSC and committed progenitors, 2) whether Id proteins are induced by HGF in EML cells and normal HSC during myeloid, erythroid, and B cell development, and 3) whether overexpression of Id1 in HSC affects lineage commitment.

    Materials and Methods

    Bone marrow transplantation assays

    Equal numbers of 5FU-Id1 or 5FU-MSCV GFP+ cells were transplanted into irradiated C57BL/6 (CD45.1) mice with support marrow from CD45.1 mice. Hemopoietic reconstitution was determined by analyzing hemopoietic cells from peripheral blood (PB), BMC, spleen and thymus 4–6 mo after transplantation using flow cytometry. Briefly, cells were stained with PE-conjugated Abs including Gr-1, B220, CD4, TER119, IgM, CD19, and TCR, or biotin-conjugated Abs including Mac-1, CD3, CD8, CD71, IgD, TCR. All Abs and Tri-color streptavidin used for lineage determination were from BD Pharmingen.

    Results

    Id genes are differentially expressed EML and MPRO cells

    To identify transcription factors whose expression levels are modulated during the initial stages of myeloid development, we compared mRNA expression in the multipotential EML cell line and the more committed MPRO cell line, using the mouse GEM 2 microarray gene chip (Table I). Of the transcription factors that were differentially expressed by >2-fold, Id1 and Id2 were of particular interest because they are known to regulate the proliferation and differentiation of muscle, neuronal, and lymphoid cells. However, their role in primitive hemopoietic cell development is not understood (21, 28, 30, 31, 35, 36, 37, 38). Low levels of Id1 were detected in EML cells, while higher levels were present in MPRO cells (Table I). We validated the microarray data by verifying the expression of Id1 and Id2 in addition to c-Kit and embigin (as positive controls) by Northern and Western blot analysis. As predicted, c-Kit mRNA was expressed in EML cells, and embigin was not detected (Fig. 1A). Conversely, embigin mRNA was highly expressed in MPRO cells, while c-Kit was not detected, demonstrating that microarray data agrees with the expression of known controls. Furthermore, Id1 mRNA and protein were expressed in MPRO cells, while they were not expressed in EML cells (Fig. 1B). In contrast, Id2 expression was higher in EML cells, and decreased in MPRO cells. These data indicate that expression of Id1 and Id2 may be regulated in EML cells during early myeloid development.

    Id1 expression is induced in EML cells during myeloid, but not B cell or erythroid cell differentiation

    To determine whether Id1 gene expression is regulated in EML cells during the early stages of myeloid cell differentiation, we cultured EML cells in conditions which promote myeloid cell maturation including SCF, IL-3, and atRA (Fig. 1C) (18, 32). IL-3-induced Id1 expression in EML cells after 24 h, which was enhanced by the addition of atRA (Fig. 1C, lanes 4 and 5). IL-3 was essential because atRA alone did not induce Id1 expression. Id1 expression peaked after 3 days and was maintained through 5 days (Fig. 1F). EML cells undergo the initial stages of myeloid cell differentiation (appearance of promyelocytes and more differentiated forms) during the first 5 days in culture in the presence of SCF/IL-3 plus atRA (18, 32). Thus, IL-3 induces Id1 expression in EML cells, which is significantly enhanced by the presence of atRA during myeloid cell differentiation.

    To determine whether Id1 is only induced during myeloid differentiation, we examined Id1 expression in EML cells following induction of B cell and erythroid cell development. To promote B cell differentiation, EML cells were cocultured with S17 stromal cells and IL-7 plus Flt-3L (Fig. 1D) (18). As expected, untreated EML cells did not express Id1 (lane 7), while conditions that promote myeloid development in the same assay (IL-3/atRA) induced Id1 expression (lane 8), and Id1 expression was not affected by the addition of IL-7 alone or IL-7/Flt-3L (lanes 4–6). In contrast to myeloid development, Id1 expression was not induced in EML cells during B cell differentiation in cocultures of EML and S17 cells supplemented with IL-7 or IL-7/Flt-3L (lane 2 and 3), in comparison to control cocultures of EML/S17 alone (lane 1). Thus, Id1 is not induced in EML cells during the initial stages of B cell differentiation.

    To determine whether Id1 is induced during erythroid development, we cultured EML cells with SCF plus erythropoietin (EPO) or SCF/EPO/IL-3 for 3 and 6 days (Fig. 1E) (18). Id1 expression was not detected in EML cells cultured in SCF alone or the combination of SCF and EPO (lanes 1, 2, 4, and 5). However, Id1 expression was observed in the cultures containing a combination of SCF/IL-3/EPO at 3 days, but was decreased after 6 days (lanes 3 and 6, respectively). The increase in Id1 expression at 3 days was expected because IL-3 can induce Id1 expression in EML cells. Thus, Id1 is not induced in EML cells during the initial stages of erythroid cell differentiation.

    Id1 expression is increased in committed MPROs and is induced by IL-3 in normal hemopoietic cells

    Induction of Id1 expression by IL-3 in EML cells suggests that Id proteins may be regulated during normal myeloid development. To examine this, we first determined the expression of Id1 in normal purified hemopoietic cell populations. Id1 protein was expressed at low levels in unfractionated BMC and FACS-purified Gr-1+-granulocytes, while little or no Id1 was detected in purified TER119+-erythroid cells, and B220+-B cells isolated from spleen (Fig. 2A). However, Id expression was greatly increased in progenitor-enriched (lineage-low (Linlow)) BMC (1% of total BMC), the BMC fraction that contains the MPRO-like progenitors. To evaluate whether mature B and T cells expressed Id, we examined lymphoid subsets in the spleen. Id1 protein was weakly expressed in unfractionated spleen cells, purified B220+-B cells, and CD3+-T cells. Thus, Id1 is highly expressed in progenitor-enriched BMC, decreased in purified granulocytes, and is expressed at low levels in erythroid and lymphoid lineages.

    To determine which progenitors within the Linlow BMC progenitor population express Id genes, we purified HSC and their immediate progeny, the common lymphoid progenitor (CLP), common MPRO (CMP), granulocyte/macrophage progenitors (GMP), and megakaryocyte/erythroid progenitors (MEP), and determined whether Id1 and Id2 were expressed by RT-PCR and quantitative PCR (schematic: Fig. 2B). Very low levels of Id1 transcripts were detected in HSC and CLP by RT-PCR (Fig. 2C), and by quantitative PCR in comparison to no transcript conditions (Fig. 2D). Furthermore, when Id1 transcripts were normalized to zero in HSC, Id1 expression in CMP and GMP were significantly increased by 1.5- and 4.9- (p < 0.01) fold, respectively, while Id1 transcripts in MEP were only increased 0.8-fold. Furthermore, Id1 expression in CMP was higher than Id expression in MEP (p < 0.01). In summary, low levels of Id1 transcripts are present in HSC, increase in CMP, and are further increased in GMP, indicating Id1 mRNA expression correlates with increased MPRO cell maturation. In comparison, Id1 transcripts decrease as CMP mature to MEP progenitors and erythroid lineage cells.

    The increased expression of Id1 in CMP and GMP suggests that Id1 is up-regulated during the differentiation and development of HSC to more committed MPROs. To test this, HSC were cultured in SCF, which promotes HSC survival, or SCF plus IL-3, which promotes granulocyte, macrophage, and mast cell differentiation. Id1 and Id2 were induced in HSC by IL-3 and IL3/SCF, while SCF alone showed little effect by RT-PCR after 6 and 3 days (data not shown) (Fig. 2E). When Id1 transcripts were quantitated by PCR and normalized to zero in HSC, SCF alone did not significantly increase Id1 expression in HSC, while Id1 transcripts were increased in HSC treated with IL-3 or SCF/IL-3 x 4-fold and 5.6-fold (p < 0.01), respectively (Fig. 2G). HSC cultured in IL-3/SCF for 3 days maintained a primitive blast cell morphology as demonstrated by Giemsa-stained cytocentrifuge preparations suggesting that Id1 was induced in primitive hemopoietic progenitor cells (Fig. 2F). This effect was unique to myeloid HGF because Id1 transcripts were not significantly increased in CLP by IL-7/Flt-3L (Fig. 2H) or in Linlow BMC by Flt-3L/IL-7 (data not shown). Thus, Id1 is induced in purified HSC by SCF/IL-3, which support myeloid differentiation, while Id1 expression is not increased in CLP during lymphoid development.

    To evaluate whether Id1 expression in purified HSC was regulated by HGF that promote the growth of other cell lineages, we cultured purified progenitors (LinlowKit+ Sca-1–) in SCF/EPO to promote erythropoiesis, in SCF/TPO to generate megakaryocytes, in SCF/M-CSF to promote macrophage development, in SCF/G-CSF to generate granulocytes, and in GM-CSF to promote granulocytes/macrophage growth (differentiation was confirmed by evaluation of morphology and cell surface marker expression, data not shown). Quantitative PCR showed that SCF treatment did not significantly alter Id1 expression levels after 6 days (Fig. 2I), while growth factors that promote granulocyte/macrophage differentiation including IL-3, SCF/IL-3, and GM-CSF showed increased Id1 mRNA levels by 2.2-, 2.0-, and 1.1-fold (p < 0.005), respectively. Id1 expression was not significantly changed in SCF/G-CSF-treated cultures, while cells treated with SCF/M-CSF showed increased Id1 expression suggesting that Id1 expression is decreased in neutrophils but not macrophages. Furthermore, Id1 expression was significantly down-regulated in LinlowKit+ Sca-1– cells by HGF that promote erythroid/megakaryocyte differentiation, including SCF/EPO and SCF/TPO, by 1.0- and 2.7- (p < 0.005) fold, respectively. In summary, Id1 expression is increased in normal stem and progenitor cells treated with HGF that promote myeloid development, but decreased in progenitors treated with HGF that promote erythroid/megakaryocyte development.

    Id1 alters hemopoietic development from EML cells in vitro and HSC in vivo

    We found that Id1 expression decreases during erythroid development, but increases during myeloid development suggesting that high levels of Id1 expression may be permissive for myeloid cell development and suppress erythroid cell maturation. To evaluate this, we established EML clones that overexpress Id1 using retroviral vectors that express Id1 and GFP (EML-Id1), or only express GFP (EML-MSCV), and cultured them in conditions that promote erythroid or myeloid development. The ability of EML-Id1 cells to differentiate into erythroid cells (burst-forming unit-erythroid (BFU-E)) was inhibited by 67% compared with EML-MSCV control cells (Fig. 3A). In addition, the ability of EML-Id1 cells to proliferate in soft agar was significantly enhanced in comparison to EML-MSCV cells. Thus, overexpression of Id1 in EML cells inhibits erythroid development.

    EML cells can also be induced to differentiate into restricted MPROs (GMP-like cells) called EML progenitors (EPRO). We evaluated whether EML cells that overexpress Id1 could be induced to differentiate into EPRO. We were able to establish GM-CSF-dependent EPRO-Id1 cells demonstrating that overexpression of Id1 does not impair the early stages of myeloid cell differentiation.

    To evaluate whether Id1 affects the later stages of myeloid development, we induced neutrophil or macrophage differentiation of EPRO-Id cells, and found that 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced macrophage differentiation (Mac-1+Gr-1– cells) was increased in EPRO-Id cells compared with EPRO-MSCV cells, while GM-CSF plus atRA-induced neutrophil differentiation (Mac-1+Gr-1+ cells) was inhibited (Fig. 3B). In agreement with this, we found that normal neutrophils express low levels of Id1 protein (Fig. 2A), suggesting that Id levels decrease during normal neutrophil maturation. We also found that Id1 and Id2 protein expression decrease in 32D-cl3 cells treated with G-CSF over a 6-day period (>90–95% band and segmented neutrophils; Fig. 3C) (30). Id1 protein levels also decrease in MPRO cells during neutrophil maturation indicating that this effect is not unique to 32D-cl3 cells (Fig. 3D).

    Although overexpression of Id1 did not impair macrophage differentiation, it is not known whether Id1 levels increase or decrease during normal macrophage differentiation. To evaluate Id1 levels during macrophage maturation, we treated normal mouse BMC with M-CSF and determined Id1 expression by Western blot analysis. We found that Id1 protein expression increased during macrophage differentiation, and remained at high levels after 7 days where >95% of the cells were terminally differentiated macrophages by morphology and cell surface marker expression (Fig. 3E). Thus, Id1 overexpression inhibits terminal granulocyte but not macrophage differentiation.

    Collectively, the results of overexpression of Id1 in EML cells suggest that overexpression of Id1 in normal HSC could promote myeloid development at the expense of erythroid or other cell lineages. To examine this effect in vivo, we infected BMC from 5FU-treated mice with retroviral vectors that express Id1 and GFP or GFP alone and transplanted equal numbers of GFP+ cells into lethally irradiated mice with support/competitor marrow. We evaluated myeloid, erythroid, and lymphoid reconstitution after 4–6 mo. For repopulation analyses, we gated on Id1- or MSCV-transduced GFP+ cells and observed a significant increase in Mac-1+Gr-1– macrophages (6.9 ± 0.5 vs 9.6 ± 0.5; * = p < 0.005), Mac-1+Gr-1+ granulocytes (9.5 ± 1.9 vs 19.5 ± 6.65; p < 0.005; Fig. 4B) and Mac-1+Gr-1low immature myeloid cells: (4.3 ± 0.5 vs 7.2 ± 2.1; p < 0.005; data not shown) in the PB. In agreement with these data, increased numbers of macrophages and granulocytes were also observed in the bone marrow of 5FU-Id1 transplanted mice (Fig. 4A). Second, we observed a decrease in B220+-B cells in the PB (37 vs 23%; p < 0.01), and BMC (Fig. 4). This correlated with a decrease in the numbers of immature IgM+IgD– B cells and mature IgM+IgD+ B cells (p < 0.01; Fig. 4A). In addition, we observed a decrease in TER119+ erythroid cells in the PB of 5FU-Id transplanted mice, and these mice were anemic with decreased levels of hemoglobin (Fig. 4B, and complete blood cell count, data not shown).

    T cell reconstitution was evaluated at 16, 20, and 28 days, and 4 mo after transplantation of 5FU-Id1 BMC (Fig. 5, A and B). Thymic cellularity (total GFP+ cells) was reduced by roughly 50% in mice transplanted with 5FU-Id1 BMC after 28 days(Fig. 5A). However, the remaining thymocytes appeared to develop normally as determined by expression of CD4 and CD8 (Fig. 5A). Furthermore, we observed an increase in the percentage of CD3+ T cells in the PB 4 mo after transplantation (24.4 ± 5.7 vs 35.9 ± 6.4; p < 0.005; Fig. 5B). This increase was observed in all populations examined including the single-positive CD8 and CD4, and CD4+CD8+ populations (p < 0.01) (Fig. 5B). The decreased thymic cellularity in 5FU-Id1-transplanted mice is similar to the decrease (90%) in thymic cellularity observed in an lck-Id1 transgenic mouse model (39). However, in contrast to the lck-transgenic mouse model, we observed an increase in the percentage of T cells in the PB after 4 mo. We questioned whether this difference was due to decreased expression of Id1 in repopulating lymphoid cells. Therefore, we compared the levels of GFP expression in T cells and myeloid cells in the PB. We found that the T cells in the PB expressed lower levels of GFP compared with myeloid cells, suggesting that T cells expressing high levels of Id1 do not progress through thymic development (Fig. 5C). Collectively, mice transplanted with 5FU-Id BMC show 1) an increase in immature and mature myeloid cells, 2) an increase in macrophages, 3) decreased B cell and erythroid development, and 4) an increase in the percentage of peripheral T cells.

    Discussion

    Questions remain as to whether cell fate decisions during hemopoietic development are the result of internally driven programs or environmental instructive signals (9, 40). For example, do HGF determine cell fate or are they permissive for survival of certain cell lineages? Collectively, our data support a model where IL-3 may direct HSC and more committed progenitors toward a myeloid cell fate vs a lymphoid or erythroid cell fate by inducing the expression of Id1. In support of this, we found that overexpression of Id1 in a multipotential clonal stem cell line and in HSC impairs erythroid differentiation in vitro and in vivo. Thus, regulation of Id protein levels in CMP could instruct CMP toward a myeloid (GMP) vs erythroid (MEP) cell fate. Furthermore, overexpression of Id1 in HSC impairs B cell development in vivo, which agrees with other studies demonstrating that B cell development is blocked at the pro-B cell stage in transgenic mice that overexpress Id1 during B cell development (41). In comparison, while overexpression of Id1 does not affect the early stages of myeloid development, high levels of Id1 inhibit the final stages of granulocyte but not macrophage differentiation. Thus, in addition to its effects on the early stages of myeloid vs erythroid or lymphoid development, Id1 may also be a part of a mechanism regulating the final choice between macrophage and neutrophil development. Altogether, our data suggest that increased levels of Id1 could promote 1) myeloid vs lymphoid lineage commitment of HSC, 2) myeloid vs erythroid cell development at the CMP stage and 3) macrophage vs a granulocyte cell fate in committed GMPs (Fig. 6).

    A significant effort in stem cell biology is focused on methods to expand HSC with HGF in vitro to improve the quality of bone marrow transplants and provide improved targets for gene therapy (42). However, expansion of HSC in vitro often results in compromised HSC engraftment activity (43). In this regard, and in agreement with our data, inclusion of IL-3 in mixtures of HGF to expand HSC was detrimental to B cell development in vivo (44, 45, 46). For example, the addition of IL-3 abrogated the long-term reconstitution activity of purified HSC cultured in SCF plus IL-11 in vitro (47). Therefore, as an explanation for this effect, we propose that IL-3 might induce Id protein expression in HSC, resulting in impaired HSC engraftment ability and inhibition of lymphoid reconstitution activity. In this regard, while cytokines provide growth and survival signals to cells, HGF can promote cell fate decisions, in part, by affecting the levels of transcriptional regulators in cells. For example, G-CSF drives neutrophil vs macrophage cell differentiation by increasing the expression of C/EBP with respect to PU.1, which drives macrophage development (48).

    In addition to impaired B cell development in mice transplanted with 5FU-Id BMC, we found that there was a decrease in the number of developing thymocytes (28 days after transplantation), and an increase in total PB CD3+, CD4–CD8+, and CD4+CD8– T cells 4 mo after transplantation. In this regard, transgenic mice that overexpress Id1 in the thymus show a T cell deficiency due to apoptosis of developing thymocytes, and develop T cell lymphomas as they age (39, 49). These tumors showed significant heterogeneity in their CD4 and CD8 profiles, including tumors that were double-negative, single CD4 or CD8 positive, or a mixture of cells with different CD4 and CD8 profiles. In addition, some mice only showed enlarged thymus with increased T cell numbers and normal CD4 and CD8 profiles, suggesting that thymocytes in lck-transgenic mice might hyperproliferate before transformation, which is similar to what we observed in 5FU-Id1-transplanted mice.

    Based on previous observations, it is thought that Id proteins negatively regulate B and T cell development, by inhibiting E2A homodimer and E2A/Hela E-box binding heterodimer formation and DNA-binding activity, which are required for B and T cell differentiation, respectively (50, 51). Id1 proteins are also thought to interfere with B cell development by binding to Pax-5 and inhibiting its transcriptional activity (52). Similarly, it has been hypothesized that Id proteins inhibit E47/stem cell leukemia (Tal-1) heterodimer formation and DNA-binding activity and prevent erythroid development (53, 54). In comparison to these observations, we found that B cell development was inhibited and erythroid development was impaired, while T cell numbers were increased.

    Thus far, no defect in myeloid, lymphoid, or erythroid development has been reported in Id1 null mice (55, 56). In addition, while Id2 null mice show defects in NK cell development, Langerhans and splenic dendritic cells and Id3 null show impaired humoral B cell proliferative responses and thymocyte maturation, defects in other hemopoietic lineages have not been reported (57, 58). Id1/Id2 null (R. Benezra, unpublished observation) and Id1/Id3 null mice show severe developmental defects and die in utero at days 14–15 of gestation, suggesting that Id proteins may compensate for each other during development (55).

    In summary, our data suggest that Id gene expression is important in determining cell fate in HSC and their immediate progeny. Furthermore, because Id proteins regulate cell proliferation, future experiments will examine their role in regulating the proliferation of normal and leukemic hemopoietic cells.

    Footnotes

    The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

    1 This project has been funded in whole or in part with Federal funds from the National Cancer Institute, National Institutes of Health, under contract number NO1-CO-12400. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government. The publisher or recipient acknowledges the right of the U.S. Government to retain a nonexclusive, royalty-free license in and to any copyright covering this article.

    2 W.L. and H.W. contributed equally to this work.

    3 Address correspondence and reprint requests to Dr. Jonathan R. Keller, Basic Research Program, SAIC-Frederick, National Cancer Institute-Frederick, Building 560, Room 12-03, Frederick, MD 20702-1201. E-mail address: kellerj{at}ncifcrf.gov

    4 Abbreviations used in this paper: HSC, hemopoietic stem cell; HGF, hemopoietic growth factor; EML, erythroid myeloid lymphoid; MPRO, myeloid progenitor; Id, inhibitor of DNA-binding protein; HLH, helix-loop-helix; BMC, bone marrow cell; HS, horse serum; 5FU, 5-fluorouracil; SCF, stem cell factor; atRA, all trans-retinoic acid; Linlow, lineage low; EPO, erythropoietin; CLP, common lymphoid progenitor; CMP, common MPRO; GMP, granulocyte/macrophage progenitor; MEP, megakaryocyte/erythroid progenitor; EPRO, EML progenitor; TPO, thrombopoietin; TPA, 12-O-tetradecanoylphorbol-13-acetate; MSCV, Moloney sarcoma virus; BFU, burst-forming unit-erythroid.

    Received for publication November 16, 2004. Accepted for publication March 25, 2005.

    References

    Spangrude, G. J., L. Smith, N. Uchida, K. Ikuta, S. Heimfeld, J. Friedman, I. L. Weissman. 1991. Mouse hematopoietic stem cells. Blood 78: 1395-1402.

    opoietic stem cells. Annu. Rev. Cell Dev. Biol. 11: 35-71.

    Kondo, M., A. J. Wagers, M. G. Manz, S. S. Prohaska, D. C. Scherer, G. F. Beilhack, J. A. Shizuru, I. L. Weissman. 2003. Biology of hematopoietic stem cells and progenitors: implications for clinical application. Annu. Rev. Immunol. 21: 759-806.

    Zhang, D. E., S. Hohaus, M. T. Voso, H. M. Chen, L. T. Smith, C. J. Hetherington, D. G. Tenen. 1996. Function of PU. 1 (Spi-1), C/EBP, and AML1 in early myelopoiesis: regulation of multiple myeloid CSF receptor promoters. Curr. Top. Microbiol. Immunol. 211: 137-147.

    Tenen, D. G., R. Hromas, J. D. Licht, D. E. Zhang. 1997. Transcription factors, normal myeloid development, and leukemia. Blood 90: 489-519.

    Ward, A. C., D. M. Loeb, A. A. Soede-Bobok, I. P. Touw, A. D. Friedman. 2000. Regulation of granulopoiesis by transcription factors and cytokine signals. Leukemia 14: 973-990.

    Tenen, D. G.. 2003. Disruption of differentiation in human cancer: AML shows the way. Nat. Rev. Cancer 3: 89-101.(Wilairat Leeanansaksiri2,)