Differentiation Of Primitive Hematopoietic Cells Research Articles

β-Catenin Expression in Bone Marrow Stromal Cells Is Required for Normal Proliferation and Differentiation of Primitive Hematopoietic Cells.

Normal function of adult hematopoietic stem cells (HSC) is dependent upon the bone marrow microenvironment, which is comprised of multiple cell types, including fibroblasts, endothelial cells, osteoblasts and stromal progenitors. We hypothesized that canonical Wnt signaling, which is necessary for the development of mesenchymal tissue, regulates the cellular structure and function of the microenvironment. We tested this hypothesis using an in vitro model of bone marrow stroma that is deficient in β-catenin, the critical transactivator of canonical Wnt signals. β-catenin−/ − Dexter stroma cultures were established from whole bone marrow harvested from β-cateninlox/lox Mx-cre+/cre mice treated with 7–10 doses of 300 μg pIpC. PCR analysis of stromal cell DNA showed nearly 100% deletion of β-catenin. Confluent stroma cultures were irradiated and seeded with 4 x 104 lin cells/cm2. We have reported (Nemeth, et al, Blood, 108, 29a) that β-catenin−/ − stroma exhibit decreased ability to support CFU formation and generate osteoblasts. The reduction in CFU-C correlates with a 75% increase in the percentage of lin− progenitors cultured on β-catenin−/ − stroma undergoing apoptosis (23.6 ± 3.4%) compared to wild-type (13.6 ± 1.3% ; n = 3; p < .01). To determine the mechanism by which canonical Wnt signaling regulates microenvironment function, we used a cytokine antibody array to analyze protein levels of 30 different hematopoietic growth factors and adhesion molecules. We observed decreased amounts of the soluble factors bFGF (3.3 ± 0.6-fold) and SCF (2.3 ± 0.3-fold), and the adhesion factor VCAM-1 (2.7 ± 0.3-fold) (n = 3; p < .01) in β-catenin−/ − stroma. The decrease in VCAM-1 corresponded with decreased percentages of VCAM-1+ osteoblasts (26.8 ± 0.9% vs. 43.9 ± 5.7%) and endothelial cells (31.6 ± 5.7% vs. 76.5 ± 10.9%) in β-catenin−/ − stroma compared to wild-type (n = 4, p < .01). From these data, we hypothesized that β-catenin is necessary for maintaining the stromal cells that support HSCs in vivo. We tested this hypothesis by transplanting 2 x 106 wild-type whole bone marrow cells (CD45.1) into lethally-irradiated β-cateninlox/lox Mx-cre+/cre mice and littermate controls. 8 weeks later, transplanted mice were treated with pIpC, resulting in mice with a wild-type hematopoietic system and a β-catenin−/ − microenvironment. Two weeks after the final treatment, we observed a 2.7 ± 0.1-fold reduction in the percentage of long- and short-term HSCs (lin−, c-kitHI, Sca-1HI) in bone marrow from mice with a β-catenin−/ − microenvironment compared to wild-type (n = 4, p < .05). We performed a competitive repopulation assay, transplanting 1 x 106 whole bone marrow cells harvested from primary recipients with a wild-type or β-catenin−/ − microenvironment with 1 x 106 CD45.2 whole bone marrow cells into lethally-irradiated secondary recipient mice. After 16 weeks, there was no difference in mean repopulation by bone marrow cells from the β-catenin−/ − microenvironment (9.2 ± 2.8%) compared to wild-type (10.7 ± 0.6%), indicating that self-renewal was unaffected. However, we did observe a significant 4-fold increase in variability of repopulation by bone marrow cells from the β-catenin−/ − microenvironment (F-test = .01). Since smaller numbers of HSCs will yield greater variability in repopulation than larger numbers, this is consistent with the observation that the β-catenin−/ − microenvironment supports fewer HSCs. From these data, we propose a model in which canonical Wnt signaling in the microenvironment is necessary for hematopoietic proliferation and differentiation.

CBFB-Independent Activities of the Inv(16) Fusion Gene CBFB-MYH11.

Inv(16)(p13;q22), the chromosomal translocation commonly found in patients with AML M4Eo, results in the fusion gene CBFB-MYH11. We have shown previously that mice heterozygous for Cbfb-MYH11 (Cbfb+/MYH11) have severe central nervous system hemorrhage and a complete lack of definitive hematopoiesis. This phenotype is indistinguishable from that seen in mice homozygous for Cbfb null (Cbfb−/−), consistent with the idea that the Cbfb-MYH11 gene product prevents normal Cbfb activity. However, in the peripheral blood of Cbfb+/MYH11 embryos before embryonic day 12.5 (E12.5), we have observed an increase of morphologically immature primitive hematopoietic cells, which has not been described in the Cbfb−/− embryos. This finding raises the possibility that the Cbfb-MYH11 gene product has activities not related to the inhibition of normal Cbfb function. To address this hypothesis, we further characterized differentiation of primitive hematopoietic cells in Cbfb+/+, Cbfb+/MYH11, and Cbfb−/− embryos. By FACS, we found that over 90% of the primitive blood cells in Cbfb+/+ and Cbfb−/− E10.5 embryos showed high expression of the differentiation marker TER-119, and were negative for the progenitor marker, c-kit. In contrast, the primitive blood cells of Cbfb+/MYH11 embryos, had increased expression of c-kit, and decreased expression of TER-119 indicating an increase of immature, progenitor cells. We also saw this delay in differentiation in E11.5 and E12.5 Cbfb+/MYH11 embryos, although the severity decreased with age. In addition, we found increased BrdU incorporation and annexin V staining of primitive blood cells in E12.5 Cbfb+/MYH11 embryos as compared to blood from Cbfb+/+ embryos. The rates of BrdU incorporation and annexin V staining in the blood of Cbfb−/− embryos were indistinguishable from wildtype. Together, these results indicate that the Cbfb-MYH11 gene product causes defects in differentiation, proliferation, and apoptosis in primitive blood that is independent of its ability to inhibit normal Cbfb activity. In order to address potential targets of this Cbfb-independent activity, we performed microarray analysis with mRNA from the peripheral blood of Cbfb+/+ and Cbfb+/MYH11 E12.5 embryos. One gene that showed significant upregulation in the blood from Cbfb+/MYH11 embryos was Csfrβ, the common β dimerization partner of the IL-3, Il-5, and GM-CSF α receptors. By FACS, we confirmed that Csfrβ (also known as CD131) was also upregulated at the protein level. In contrast, expression of Csfrβ in the blood of Cbfb−/− embryos was indistinguishable from wildtype indicating that signaling through Csfrβ and its dimerization partners could play a role in the Cbfb-independent activities of the Cbfb-MYH11 gene product. This effect was not unique to embryonic blood as adult mice that conditionally express Cbfb-MYH11 developed a population of pre-monocytic, Csfrβ+ cells in their peripheral blood as early as one week after induction of the fusion protein. In addition, leukemic cells induced in the Cbfb-MYH11 knockin mice expressed high levels of Csfrβ. These results indicate that expression of Csfrβ is an almost immediate consequence of Cbfb-MYH11 expression and that Csfrβ may be a direct target of the fusion protein. In conclusion, we demonstrated that Cbfb-MYH11 is able to impair hematopoiesis in a Cbfb-independent manner, perhaps through the direct upregulation of Csfrβ and consequent activation of its downstream signaling pathways.

Expression and characterization of the recombinant human FLT-3 ligand extracellular domain in Pichia pastoris

OBJECTIVE The FLT-3 ligand (fms-like tyrosine kinase receptor-3 ligand, FL) is a recently described growth factor affecting early hematopoietic progenitor cells. The FL plays a key role in the growth and differentiation of primitive hematopoietic cells. To yield a high-level of recombinant human FL protein, a recombinant Pichia Pastoris (P. pastoris) strain was constructed.

Survivin Regulates Proliferation of Normal Hematopoietic Stem and Progenitor Cells In Vivo.

The inhibitor of apoptosis protein Survivin is barely detectable in most normal adult tissues but is over-expressed in almost all cancers. Survivin regulates apoptosis, cell division and cell cycle, making anti-Survivin therapy an attractive cancer treatment strategy. We reported that Survivin is expressed and regulated by hematopoietic growth factors in normal human CD34+ cells and that over-expression of wild-type Survivin in bone marrow cells enhances in vitro proliferation and survival of normal hematopoietic progenitor cells, whereas disrupting Survivin reduced their proliferation and survival. These results suggest that Survivin regulates normal hematopoietic progenitor cell function. Although targeted anti-Survivin therapies for cancers demonstrate efficacy without overt toxicity in animal models, the consequences of in vivo Survivin disruption in normal hematopoietic stem and progenitor cells (HSPC) has not been determined. In order to understand the physiological roles of Survivin in normal HSPC function in vivo, we created Cre-ER™/Survivin flox/flox mice, in which the Survivin gene can be excised by Tamoxifen treatment and characterized HSPC growth following Survivin gene deletion. RT-PCR analysis showed that Survivin mRNA is expressed in freshly isolated normal mouse marrow Sca-1+, c-kit+, lin− (SKL) cells and more primitive CD34−SKL cells, which contain long term repopulating hematopoietic stem cells (HSC). Administration of 5mg of Tamoxifen for 6 days (3 days injection, 3 days off, 3 additional days and analyzed 14 days after final injection) in Cre-ER™/Survivin flox/flox mice induced Survivin gene deletion in marrow cells, but had little effect on peripheral blood cell count, marrow cellularity (3.5+/−7.1%, NS) or the proportion or total number of lineage committed cells (Gr-1+, Mac-1+, B220+, CD4+ and/or CD8+) in marrow and in peripheral blood. In contrast, short term Survivin deletion significantly decreased the frequency and the absolute number of undifferentiated linneg cells (37+/−6% reduction), c-kit+, lin− cells (35.2+/−8.4% reduction,), CFU-GM (31+/−9 % reduction), Lin−, IL7Ra−, Sca-1−, c-kit+, CD34+, Fcglow common myeloid progenitor cells (52+/−13% reduction), SKL cells (56.8+/−5.4% reduction) and CD34−SKL cells (60.6+/−5.5% reduction) in bone marrow compared to control mice. The effect of Survivin gene deletion was more dramatic on primitive hematopoietic populations compared to mature cells, which is consistent with down-regulation of Survivin in hematopoietic cells with terminal differentiation. Similarly, treatment of bone marrow cells from Cre-ER™/Survivin flox/flox mice with 1uM of Tamoxifen in vitro significantly reduced the number of CFU-GM, (c-kit+, lin−) KL, SKL and CD34−SKL cells cultured with hematopoietic cytokines and increased apoptosis measured by Annexin-V staining. These results suggest that Survivin is required and regulates normal hematopoietic stem and progenitor function in vivo and that Survivin function may be selectively essential for growth and differentiation of primitive hematopoietic cells. In addition, acute ablation of Survivin may cause adverse toxicity on HSPC that provide long term hematopoiesis in the patients receiving anti-Survivin target therapies.

The Newest Member of the Inhibitors of Apoptosis Family, IAP Like Protein 2 (ILP2) Demonstrates Growth Factor Regulation in CD34+ and Mo7e Cells

Inhibitors of apoptosis proteins are a family of natural cysteine proteases inhibitors that regulate apoptosis through caspase dependent and independent mechanisms. ILP2 is the newest family member with restricted specificity for caspase-9 and neutralization of Smac/Diablo complex. ILP2 expression has been reported in the testis and lymphoblastoid and transformed cell lines. We examined ILP2 mRNA and protein expression before and after incubation with cytokines in CD34+ and MO7e cells and examined if ILP2 expression was associated with cell cycle progression and apoptosis. Western analysis and quantitative-RT-PCR revealed that ILP2 was expressed in MO7e and CD34+ cells and that growth factor starvation reduced ILP2 protein and mRNA. Addition of stem cell factor or granulocyte-macrophage colony-stimulating factor blocked down-regulation of ILP2 protein and enhanced cell survival and cell cycle progression in MO7e cells. ILP2 mRNA was up-regulated by the early acting cytokine SCF but not by the late acting GM-CSF. ILP2 mRNA was down-regulated by the combination GM-CSF + SCF, suggesting ILP2 may be regulated with differentiation. Results suggest that ILP2 may be involved in the growth factor mediated proliferation, survival or differentiation of primitive hematopoietic cells. Supported in part by NIH R25 GM067592-02 and P01 AI56097-02S1.

Wnt5a Inhibits Wnt3a-Mediated HSC Differentiation.

Activation of the canonical Wnt signaling pathway by Wnt3a has been implicated in hematopoietic stem cell (HSC) self-renewal (Reya et al., Nature, 2003). Wnt5a has been observed to inhibit Wnt3a signaling (Topol et al., J Cell Biol, 2004). We hypothesized that Wnt3a and 5a act as antagonists on HSC function. 1 x 106 lineage negative cells (lin−) were cultured for 4 days in the presence of 50 ng/ml SCF and Flt3L (control) plus 100 ng/ml rmWnt3a and/or 500 ng/ml rmWnt5a (all factors added on day 0 and day 2). Control lin− cell numbers expanded more than lin− cells cultured with Wnt3a, 5a, or both (control 8.3 ± 0.3-fold; Wnt3a 6.9 ± 0.2-fold (p < .01); Wnt5a 4.8 ± 0.2-fold (p < .001); Wnt3a and 5a 2.6 ± 0.6-fold (p < .001); n = 3). After 4 days, cells were analyzed for myeloid colony formation. Control cells and cells cultured in Wnt3a had similar numbers of CFU-GM/5000 lin− cells (control 13.1 ± 11.1; Wnt3a 21.8 ± 15.3; p = .21; n = 8), while cells cultured in Wnt5a and Wnt3a and 5a had 2-fold and 5.9-fold more CFU-GM/5000 lin− cells than control (Wnt 5a 26.8 ± 13.3 (p = .04); Wnt3a and 5a 77.9 ± 48.3 (p < .01); n = 8). To analyze repopulating ability, 4 x 105 lin− Ly5.1 cells, cultured under the same conditions, were transplanted with 2 x 106 Ly5.2 bone marrow cells into lethally-irradiated Ly5.2 recipients. 16 weeks after transplant, repopulation by control lin− cells increased 2-fold compared to lin− cells cultured in Wnt3a or Wnt5a (control 7.3 ± 3.8%; Wnt3a 3.37 ± 1.2% (p < .01); Wnt5a 3.6 ± 1.1% (p < .01); n = 9-10). However, lin− cells cultured in Wnt3a and 5a showed normal repopulating activity (n = 10; 8.7 ± 5.3%; p = .52). 1 x 104 HSCs (lin−, c-kitHI, Sca-1HI, IL-7Rα −) were cultured for 6 days with SCF, Flt3L, Wnt3a and 5a (factors added on day 0 and day 3) as described above. Control HSC numbers expanded more than HSCs cultured with Wnt3a, Wnt 5a, or both (control 20.7 ± 10.4-fold; Wnt 3a 7.0 ± 4.1-fold (p = .05); Wnt5a 1.7 ± 1.7-fold (p = .01); Wnt3a and 5a 1.2 ± 1.0-fold (p < .01); n = 4). Similar numbers of control HSCs and HSCs cultured with Wnt3a or 5a were lin+ (control 21.7 ± 0.2%; Wnt 3a 15.4 ± 5.3% (p = .10); Wnt5a 14.4 ± 5.2% (p = .07); n = 3). However, culturing HSCs with Wnt3a and 5a resulted in a 50% decrease in the number of lin+ cells compared to control (12.3 ± 2.0% (p = .001)). Cultured Ly5.1 HSCs were transplanted with Ly5.2 bone marrow cells at a 1:100 ratio. There was no difference in repopulation between control HSCs and HSCs cultured with Wnt3a (control 5.8 ± 6.1%; Wnt3a 3.6 ± 0.4%; p = .43; n = 5). To examine the effects of enforced expression of Wnt ligands in HSCs, 5-FU treated bone marrow was transduced with Wnt3a-IRES-GFP, Wnt5a-IRES-dsRED, or IRES-GFP retroviral vectors. Sorted IRES-GFP+, Wnt3a-GFP+ and Wnt5a-dsRed+ cells (Ly5.1) were transplanted with equal numbers of mock-transduced cells and 3 x 105 Sca-1− bone marrow cells (Ly5.2) into lethally-irradiated Ly5.2 mice. 16 weeks later, recipients of IRES-GFP+ and Wnt5a-dsRed+ cells contained a similar number of engrafted cells expressing the vector (3.4 ± 1.8% GFP+ Ly5.1 and 3.5 ± 0.4% dsRed+ Ly5.1 respectively; n = 8). In contrast, no GFP+ Ly5.1 cells were detected in Wnt3a-GFP+ recipients (n = 8). 33.4 ± 3.7% of bone marrow cells were Ly5.1+ indicating successful engraftment and retroviral DNA was detected by PCR, suggesting that transduction had occurred but that only cells in which the vector was silenced survived. We conclude that activation of the canonical Wnt pathway in HSCs promotes differentiation of primitive hematopoietic cells and that other signals, such as Wnt5a, are required to maintain the balance between HSC differentiation and self-renewal.

Hematopoietic Stem Cells from Fancc −/− Mice Have Lower Growth and Differentiation Potential in Response to Growth Factors

Fanconi anemia (FA) is a complex recessive genetic disease characterized by progressive bone marrow (BM) failure. We have previously shown that stem cells from the FA group C mouse model have lower long-term primary and secondary reconstitution ability, and that bone marrow of Fancc(-/-) mice contained fewer lineage-negative (Lin(-))Thy1.2(low)Sca-1(+)c-kit(+) CD34(+) cells but normal levels of Lin(-)Thy1.2(low)Sca-1(+)c-kit(+)CD34(-) primitive cells. These data suggest that CD34(+) primitive cells have either a lower growth or differentiation potential, or that these cells have greater apoptosis levels. To investigate the role Fancc might have on the growth and differentiation potentials of primitive hematopoietic stem cells, we used a single-cell culture system and monitored cell viability, doubling potential, and apoptosis levels of Fancc(-/-) primitive Lin(-)Thy1.2(-)Sca-1(+) (LTS)-CD34(+) and LTS-CD34(-) stem cells. Results showed that Fancc(-/-) LTS-CD34(-) and LTS-CD34(+) cells had altered growth and apoptosis responses to combinations of stimulatory cytokines, most dramatically in response to a combination of factors that included interleukin-3 (IL-3) and IL-6. In addition, Fancc(-/-) LTS-CD34(-) and LTS-CD34(+) cells showed a lower differentiation potential than Fancc(+/+) cells. These results support a role for Fancc in the growth and differentiation of primitive hematopoietic cells and suggest that an altered response to stimulatory cytokines may contribute to BM aplasia in FA patients.

Leptin Receptor and Leukemia

The receptor for leptin, the gene product of the obese gene, is expressed in hematopoietic stem cells. Leptin stimulates normal myeloid and erythroid development, and is secreted from bone marrow adipocytes, which occupy most of the marrow cavity in humans. Leptin might thus play an important role in the control of the expansion and differentiation of primitive hematopoietic cells through paracrine interaction in the bone marrow microenvironment. Leukemic cells of some patients with acute myeloblastic leukemia, acute lymphoblastic leukemia, and chronic myeloid leukemia (CML) also express the leptin receptor. In cases of CML, higher expression of leptin receptor is observed during blast crisis than in chronic phase. Leptin alone and in combination with other cytokines has stimulative effects on proliferation of leukemia cells as well as anti-apoptotic effects. These findings suggest the possibility that leptin plays roles in the pathophysiology of leukemia.

Extramedullary hematopoiesis in the adult mouse liver is associated with specific hepatic sinusoidal endothelial cells

In previous work, two anatomically distinct-liver sinusoid endothelial cells (LEC): LEC-1 and LEC-2, have been described. We also reported that extramedullary hepatic hematopoiesis occurs only in close contact with LEC-1, suggesting that these cells may provide the microenvironment necessary for the maintenance and growth of hematopoietic cells. In the present work, we studied the capacity of LEC-1 and LEC-2 to maintain in vitro hematopoiesis. LEC-1 and LEC-2 were isolated and cloned from livers of adult mice. Bone marrow cells (BM) enriched with primitive hematopoietic progenitors were isolated from day-2, post-5-FU-treated mice (5-FUBMC). LEC-1 supported the maintenance and differentiation of hematopoietic progenitors for more than 6 weeks in vitro. In contrast, LEC-2 cells poorly supported the proliferation of hematopoietic cells for only two weeks of the co- culture. LEC-1 and 5-FUBMC cocultures showed cobblestone-area formation and the presence of hematopoietic progenitors that are able to form colonies (CFC) in the adhering fraction after six weeks of coculture. LEC-1 co-cultures treated with a cocktail of cytokines (stem cell factor, interleukin [IL]1alpha, IL-3, and Epo) showed that megakaryocyte (CFU-Mk) and erythrocyte progenitors (BFU-e) were present during the entire period of the culture. Granulocyte-macrophage progenitors (CFU-GM) were present only during the first three weeks of the culture. These results suggest that LEC-1, but not LEC-2, provide an appropriate hematopoietic microenvironment for supporting the proliferation and differentiation of primitive hematopoietic cells. This could explain the anatomical restriction of hematopoietic cells for growing in LEC-1 domains during liver extramedullary hematopoiesis. (Hepatology 1997 Jul;26(1):165-75)

Extramedullary hematopoiesis in the adult mouse liver is associated with specific hepatic sinusoidal endothelial cells.

In previous work, two anatomically distinct-liver sinusoid endothelial cells (LEC): LEC-1 and LEC-2, have been described. We also reported that extramedullary hepatic hematopoiesis occurs only in close contact with LEC-1, suggesting that these cells may provide the microenvironment necessary for the maintenance and growth of hematopoietic cells. In the present work, we studied the capacity of LEC-1 and LEC-2 to maintain in vitro hematopoiesis. LEC-1 and LEC-2 were isolated and cloned from livers of adult mice. Bone marrow cells (BM) enriched with primitive hematopoietic progenitors were isolated from day-2, post-5-FU-treated mice (5-FUBMC). LEC-1 supported the maintenance and differentiation of hematopoietic progenitors for more than 6 weeks in vitro. In contrast, LEC-2 cells poorly supported the proliferation of hematopoietic cells for only two weeks of the co-culture. LEC-1 and 5-FUBMC cocultures showed cobblestone-area formation and the presence of hematopoietic progenitors that are able to form colonies (CFC) in the adhering fraction after six weeks of coculture. LEC-1 co-cultures treated with a cocktail of cytokines (stem cell factor, interleukin [IL]1alpha, IL-3, and Epo) showed that megakaryocyte (CFU-Mk) and erythrocyte progenitors (BFU-e) were present during the entire period of the culture. Granulocyte-macrophage progenitors (CFU-GM) were present only during the first three weeks of the culture. These results suggest that LEC-1, but not LEC-2, provide an appropriate hematopoietic microenvironment for supporting the proliferation and differentiation of primitive hematopoietic cells. This could explain the anatomical restriction of hematopoietic cells for growing in LEC-1 domains during liver extramedullary hematopoiesis.

Thec-kit receptor transduces the stem cell factor-triggered growth signal in murine interleukin-3-dependent cell line

The stem cell factor is a glycoprotein hormone which regulates the proliferation and differentiation of primitive hematopoietic cells through its interaction with a tyrosine kinase transmembrane receptor which is encoded by the c-kit proto-oncogene. To examine whether a murine c-kit receptor can be functional in murine interleukin-3 (mIL-3)-dependent hematopoietic cell line, we introduced the murine c-kit cDNA into mIL-3-dependent pro-B cell line Ba/F3. One of the resulting clones, Ba/F3 clone BF-K96, expressed the 140 kDa protein recognized by anti-c-kit monoclonal antibody and the expressed c-kit receptor protein on the cell surface bound to a radiolabeled soluble form of murine stem cell factor (mSCF) with high affinity. BF-K96 clone expressing the c-kit receptor could proliferate in response to mSCF in the absence of mIL-3. The cell clone could also grow in co-culture with mouse 3T3 cells which are endogeneously expressing a membrane-associated type of mSCF on their cell surfaces. These findings demonstrate that the c-kit receptor expressed on mIL-3-dependent hematopoietic cell line Ba/F3 transduce the mSCF-dependent growth signal, indicating that established cell clone will provide a unique cellular system for the study of SCF/c-kit signal transduction mechanism.

Cytokine-dependent long-term culture of highly enriched precursors of hematopoietic progenitor cells from human bone marrow.

Human marrow cells positive for the CD34 antigen but not expressing HLA-DR, CD15, or CD71 antigens were isolated. In a liquid culture system supplemented with 48-hourly additions of recombinant interleukins IL-1 alpha, IL-3, IL-6, or granulocyte/macrophage colony-stimulating factor (GM-CSF), these cells were capable of sustaining in vitro hematopoiesis for up to eight weeks. The establishment of an adherent cell layer was never observed. Cultures containing no exogenous cytokine produced clonogenic cells for only 1 wk. IL-1 alpha and IL-6 were alone able to support hematopoiesis for 2 or 3 wk. Cells maintained with GM-CSF proliferated and contained assayable colony-forming cells for 3 or 4 wk, while maximal cellular expansion and generation of assayable progenitor cells occurred in the presence of IL-3 for 4-5 wk. When IL-3 was combined with IL-1 alpha or IL-6, hematopoiesis was sustained for 8 wks. Basophil numbers were markedly increased in the presence of IL-3. These studies indicate that marrow subpopulations can sustain hematopoiesis in vitro in the presence of repeated additions of cytokines. We conclude that a major function of marrow adherent cells in long-term cultures is that of providing cytokines which promote the proliferation and differentiation of primitive hematopoietic cells.