Abstract
The stimulation by Flk2-ligand (FL) of blast colony formation by murine bone marrow cells was selectively potentiated by the addition of regulators sharing in common the gp130 signaling receptor–leukemia inhibitory factor (LIF), oncostatin M, interleukin 11, or interleukin 6. Recloning of blast colony cells indicated that the majority were progenitor cells committed exclusively to macrophage formation and responding selectively to proliferative stimulation by macrophage colony-stimulating factor. Reculture of blast colony cells initiated by FL plus LIF in cultures containing granulocyte/macrophage colony-stimulating factor plus tumor necrosis factor α indicated that at least some of the cells were capable of maturation to dendritic cells. The cells forming blast colonies in response to FL plus LIF were unrelated to those forming blast colonies in response to stimulation by stem cell factor and appear to be a distinct subset of mature hematopoietic stem cells.
Hematopoietic stem cells can be separated from other murine bone marrow cells by fluorescence-activated cell sorting as a morphologically uniform population, but the population appears to be heterogeneous based on varying proliferative and differentiative potentials. The least mature stem cells exhibit long-term repopulating capacity in vivo and a capacity for self-renewal (1). The most mature stem cells exhibit no capacity for self-renewal but, when stimulated in vitro by stem cell factor (SCF), can form blast colonies composed of progenitor cells committed to differentiation in a number of lineages (2). This action of SCF is strongly enhanced by the addition of colony-stimulating factors (CSFs) or interleukin 6 (IL-6) (3, 4).
Flk2-ligand (FL), the ligand for the Flt3/Flk2 receptor displayed on immature hematopoietic cells, has a relatively weak proliferative action in cultures of marrow cells to stimulate the formation of blast colonies and small numbers of predominantly granulocytic colonies (5). Combinations of FL with other growth factors exhibit enhanced activity in stimulating the proliferation of stem cells to generate progenitor cells and more differentiated progeny (6, 7). Administration of FL to mice increased the formation of dendritic cells (8). In its actions, FL therefore exhibits some features in common with SCF but does differ in other aspects.
Leukemia inhibitory factor (LIF) is a pleiotropic factor that can suppress self-generation and induce differentiation in cells of the murine myeloid leukemic cell line, M1 (9). When acting alone in semisolid cultures of murine bone marrow cells, LIF has no observable proliferative action on any hematopoietic population (10).
In an analysis of synergistic interactions exhibited by combinations of various growth factors in semisolid cultures of mouse bone marrow cells, it was noted that LIF enhanced colony formation stimulated by FL. The present experiments have analyzed this intriguing example of potentiation of a weak stimulus by addition of a factor with no overt activity and have extended the observations to include other regulators sharing with LIF the common signaling receptor chain, gp130 (11).
MATERIALS AND METHODS
Mice.
Mice used were 6–8 weeks old C57BL/6, DBA/2, CBA, and BALB/c mice of both sexes. In some experiments, day-old C57BL/6 mice were used. Mice were bred and housed under specific pathogen-free conditions.
Growth Factors.
The FL used was purified recombinant murine FL produced as described (5). Purified recombinant murine SCF, LIF, granulocyte/macrophage CSF (GM-CSF), macrophage CSF (M-CSF), and IL-3 were produced in this laboratory as described (10, 12). Recombinant murine IL-6 was a kind gift from R. Simpson (Ludwig Institute for Cancer Research, Melbourne Branch) and murine oncostatin M (OSM) and human IL-11 were purchased from Peprotech (Rocky Hill, NJ) and Genetics Institute (Cambridge, MA), respectively.
Agar Cultures.
Cultures were prepared in 35-mm plastic Petri dishes and contained 1 ml of DMEM with a final concentration of 20% newborn calf serum and 0.3% agar. Stimuli were added to the dishes in volumes of 0.1 ml prior to the addition of the cell suspension in agar medium. After mixing and gelling, the cultures were incubated for 7 days at 37°C in a fully humidified atmosphere of 10% CO2 in air.
Colony formation (clones of 50 or more cells) was scored using an Olympus SZ3 dissection microscope then the cultures were fixed by addition of 1 ml of 2.5% glutaraldehyde. After 4 h, the intact cultures were floated onto glass slides and, after drying, were stained sequentially for acetylcholinesterase (for megakaryocytes), Luxol-Fast-Blue (for eosinophils) then hematoxylin (for other cell types). The cellular content of all colonies in the culture was then determined at ×120 or ×480 magnifications.
The recloning capacity of 7-day colonies was tested by picking off individual colonies using a fine pipette then resuspending each in 8–10 ml of agar medium. Three to five colonies were recloned from each primary culture, choosing the largest present but otherwise selecting only on the basis that the colony was clearly separated from other colonies in the culture. The cell suspensions were then recultured in duplicate cultures containing various stimuli. After a further incubation period of 7 days, the secondary cultures were stained and analyzed for colony and cluster formation.
Dendritic Cell Cultures.
Pools of blast colonies were harvested after 7 days of incubation from cultures of 50,000 C57BL marrow cells stimulated by 500 ng FL plus 10 ng LIF. Parallel pools of macrophage colonies were harvested from cultures stimulated by 10 ng M-CSF.
The pooled colonies were dispersed and cultured for 2 weeks in suspension cultures stimulated by 10 ng/ml M-CSF. The cells were washed then half were cultured in fresh suspension cultures containing 10 ng/ml M-CSF whereas the other half were cultured with 10 ng/ml GM-CSF plus 10 ng/ml tumor necrosis factor α (TNF-α). After 48 h, the cells were harvested, washed, then stained using antibodies to class II major histocompatibility complex antigens and CD11c, and then subjected to flow cytometric analysis, as described (13).
RESULTS
Potentiation of FL-Stimulated Colony Formation by LIF.
In cultures of 50,000 C57BL bone marrow cells, 500 ng FL stimulated the formation of small colonies, the most numerous of which were composed of blast cells. These blast cell colonies differed from the multicentric blast colonies stimulated by SCF (2) in usually being a single loosely dispersed collection of cells (Fig. 1A). In Giemsa-stained preparations, the colony cells had a uniform morphology of blast cells. Characteristically, most blast colonies contained both dividing and dying cells after 7 days of incubation (Fig. 1C). As previously described (5), other colonies developing in FL-stimulated cultures were small numbers of small granulocytic, granulocyte/macrophage or macrophage colonies.
Figure 1.
(A) Blast colony stimulated by 500 ng FL. (B) Enlarged blast colony stimulated by 500 ng FL plus 10 ng LIF. (C) High power magnification of cells in colony stimulated by FL showing the presence both of apoptotic cells (open arrowheads) and mitotic cells (filled arrowheads). (D) A similar view of a colony stimulated by FL plus LIF.
Cultures stimulated by 10 ng LIF developed no clusters or colonies. Combined stimulation with 500 ng FL plus 10 ng LIF resulted in an approximate 2-fold increase in colony numbers due to a selective increase in the number of blast colonies (Table 1). Cell numbers in these blast colonies were increased 5- to 10-fold compared with those stimulated by FL alone (Fig. 1B and Table 1). The blast colony cells stimulated by FL plus LIF appeared healthier than those stimulated by FL alone, although many colonies still contained both dying and mitotic cells (Fig. 1D). There was no consistent increase in the number or size of granulocytic and/or macrophage colonies.
A similar enhancement of blast colony formation was observed in cultures of DBA/2, CBA, or BALB/c bone marrow cells stimulated by the combination of FL plus LIF. In cultures of from 25,000 to 100,000 C57BL bone marrow cells, stimulated by FL plus LIF, the enhanced numbers of blast colonies showed a linear relationship with the numbers of cells cultured, with no evidence that the phenomenon was influenced by cell or colony crowding. No colony or cluster formation was observed in cultures of spleen, thymus, or lymph node cells, stimulated by FL with or without LIF, with or without 5 × 10−5 M 2-mercaptoethanol. In cultures of 25,000 neonatal C57BL bone marrow cells, FL was unable to stimulate significant colony formation and the combination of FL plus LIF stimulated the formation of only a few small blast colonies.
The addition of 10 ng LIF did not increase the size or number of blast colonies stimulated by 100 ng SCF and did not increase granulocyte colony formation stimulated either by SCF or 100 ng IL-6. Previous studies had failed to detect enhancement by LIF of colony formation stimulated by GM-CSF, G-CSF, M-CSF, or IL-3 (10) and these observations were confirmed in the present experiments.
Combination of 500 ng FL with serial dilutions of LIF indicated that significant potentiation of blast colony formation was observable with concentrations of LIF as low as 0.15 ng/ml. Conversely, when 10 ng LIF was combined with increasing concentrations of FL, enhanced blast colony formation was observed with concentrations of 8 ng/ml FL and above.
Recloning of FL/LIF-Induced Blast Colonies.
When cells from 7-day blast colonies initiated by FL plus LIF were recultured in secondary cultures stimulated by FL or LIF, no colonies, or clusters developed. This result eliminated the possibility that FL had induced a novel responsiveness in the cells to proliferative stimulation by LIF. Even the combined stimulus of FL plus LIF produced few colonies in such secondary cultures (Fig. 2) and these were of small size and contained unhealthy cells. This finding was in line with the failure of most colonies in primary cultures stimulated by FL plus LIF to increase in size or remain healthy during a second week of incubation.
Figure 2.
Colony formation by cells from individual blast colonies when recultured in 7-day cultures stimulated by 500 ng FL plus 10 ng LIF; 10 ng IL-3, 10 ng GM-CSF or 10 ng M-CSF. Filled circles are cells from colonies initiated by 500 ng FL plus 10 ng LIF; open circles are cells from colonies initiated by 500 ng FL alone.
Unlike the behavior of SCF-stimulated blast colony cells (3, 4), few colonies developed in secondary cultures of FL plus LIF-initiated blast colony cells when stimulated by IL-3 or GM-CSF (Fig. 2) but, strikingly, these were composed wholly of macrophages. In contrast, when blast colony cells initiated by FL plus LIF were recultured using M-CSF as the secondary stimulus, large numbers of colonies, and clusters developed that were again composed wholly of well differentiated macrophages (Fig. 2).
When cells from conventional M-CSF-elicited macrophage colonies of the same size as the FL plus LIF-elicited blast colonies were recultured with M-CSF, all clones developing were again composed only of macrophages. However, as shown in Fig. 3, cells from macrophage colonies formed few secondary colonies but many macrophage clusters, the inverse of the pattern seen with recultured blast colony cells where most clones were colonies, some containing up to 500 cells. Thus the ratio of clusters to colonies for cells from blast colonies initiated by FL plus LIF was 0.8 ± 0.5 whereas the ratio for macrophage colonies initiated by M-CSF was 64 ± 123.
Figure 3.
Colony and cluster formation in secondary cultures stimulated by 10 ng M-CSF by cells from blast colonies initiated by 500 ng FL plus 10 ng LIF or by cells from macrophage colonies initiated by 10 ng M-CSF. Lines join data from individual colonies.
Action of OSM, IL-11, and IL-6.
The consistent enhancement of FL-initiated blast colony formation by the addition of LIF prompted a parallel study of the possibly similar action of OSM, IL-11, or IL-6, all of which share with LIF a common signaling receptor chain, gp130 (11).
As shown in Table 2, OSM and IL-11, when acting alone, exhibited minimal or no colony-stimulating activity but OSM, IL-11, and IL-6 all enhanced blast colony formation stimulated by FL. At the concentrations used, the action of OSM was similar in magnitude to LIF whereas IL-11 was somewhat weaker. Reculture of these blast colonies produced an identical pattern to that described for recultured colonies initiated by FL plus LIF. In each case, the blast colony cells proliferated preferentially in M-CSF-stimulated secondary cultures and produced large numbers of colonies again composed exclusively of macrophages.
As noted previously (5), combination of IL-6 with FL strongly enhanced colony formation (Table 2). These cultures were more complex than those using FL plus OSM or IL-11 because IL-6 alone is a reasonably strong CSF (14) and the observed enhancement of colony numbers and size included not only blast colonies but also granulocytic and granulocyte/macrophage colonies. In secondary cultures stimulated by M-CSF, most colonies were composed exclusively of macrophages but cells from 11 of 20 recloned colonies, when stimulated either by M-CSF, GM-CSF, or a combination of both factors, also produced in addition smaller numbers of granulocytic and granulocyte/macrophage colonies. No eosinophil or megakaryocyte cells were observed in any of these more complex secondary cultures.
Possible Relationship Between Blast Colony-Forming Cells Responsive to SCF or FL plus LIF.
In five experiments involving the culture of 50,000 C57BL bone marrow cells stimulated by 100 ng SCF, the mean number of blast colonies forming was 11 ± 3, a number intermediate between the numbers of blast colonies forming in response to FL alone or to FL plus LIF (Table 1). On this basis, the possibility was investigated that these might be the same or related populations of clonogenic cells. Twenty resuspended individual day-3 and day-4 blast colonies initiated by SCF were recultured for 7 days in cultures stimulated by FL plus LIF but failed to generate any colonies or clusters, although producing colonies in secondary cultures stimulated by SCF or IL-3. Similarly, cells from 20 individual day-7 blast colonies initiated by FL plus LIF failed to produce colonies or clusters in secondary cultures stimulated by SCF although producing secondary colonies in cultures stimulated by M-CSF.
In cultures of C57BL bone marrow cells stimulated by SCF plus FL plus LIF, the number of blast colonies developing was additive compared with the numbers stimulated by SCF or FL plus LIF alone. More strikingly, the cultures with the combination of three stimuli developed blast colonies of the two characteristic shapes—the multicentric colonies of SCF-stimulated type and the loose globular colonies of dividing and dying cells of FL-stimulated type, with no obvious size enhancement of either type.
The data suggested that the two types of blast colony-forming cells were distinct subsets without evidence of a parent–progeny relationship.
Dendritic Cell Formation by Blast Colony Cells Initiated by FL Plus LIF.
The cells in colonies or clusters forming in secondary cultures of FL plus LIF-initiated blast colony cells stimulated by M-CSF plus GM-CSF were large with an irregular cytoplasmic margin, suggesting that some of the cells could be dendritic cells. This possibility was explored by placing pools of blast colonies in suspension cultures with M-CSF as the proliferative stimulus. In parallel cultures, pools of macrophage colonies initiated by M-CSF were also grown in M-CSF-stimulated suspension cultures. After 2 weeks, the cultured cells were washed, then recultured either with 10 ng/ml M-CSF or 10 ng/ml GM-CSF plus 10 ng/ml TNF-α. After 48 h, the cells were harvested for fluorescence-activated cell sorter (FACS) analysis. At this time, FL plus LIF-derived cells growing in GM-CSF plus TNF-α tended to have formed adherent cell clumps and to be exhibiting the morphology of dendritic cells. FACS analysis indicated that a subset of these cells were highly positive for class II major histocompatibility complex and CD11c, a phenotype characteristic of dendritic cells (Fig. 4). In contrast, M-CSF-initiated macrophage colony cells failed to exhibit elevated expression of these surface markers when cultured with GM-CSF plus TNF-α.
Figure 4.
Flow cytometric analyses of (Left) cell suspensions from cultures derived from pooled macrophage colonies initiated by M-CSF then held either in M-CSF (Upper) or 10 ng/ml GM-CSF plus 10 ng/ml TNF-α (Lower). (Right) Cell suspensions derived from blast colonies initiated by FL plus LIF then held either in M-CSF or GM-CSF plus TNF-α. The boxed areas show the cells with simultaneous CD11c and class II major histocompatibility complex (MHC) expression characteristic of dendritic cells.
The data indicated that at least some blast colony cells initiated by FC plus LIF were capable of generating cells with the morphology and surface markers of dendritic cells.
DISCUSSION
The present experiments have shown that leukemia inhibitory factor, a polyfunctional agent that by itself has no proliferative action on hematopoietic cells (9, 10), reproducibly increased the size and number of blast colonies forming in cultures of mouse bone marrow cells stimulated by FL. These blast colonies differed sharply from previously described stem cell-generated blast colonies stimulated by stem cell factor (3, 4) in not being multicentric in shape and in not containing progenitor cells committed to a variety of differentiation lineages. Blast colonies stimulated by FL alone, or in combination with LIF, were remarkable in containing only macrophage-committed progenitor cells, selectively responsive to stimulation by M-CSF. When the number of macrophage cluster-forming cells also in such colonies was considered, the majority of cells in such colonies appeared to be clonogenic macrophage precursors.
Receptors for LIF have been noted on macrophages (15) but not on progenitor cells (16). The clear enhancing action of LIF suggests either that FL-responsive blast colony-forming cells were a small set of precursors exhibiting LIF receptors or that FL action may have led to the expression of LIF receptors on such cells. However, if the latter, LIF was not able to act alone to support the further proliferation of these cells. FL-stimulated blast colonies contained many dying cells and LIF may merely have improved the survival of such cells rather than acted as a coproliferative stimulus. The poorly sustained growth of blast colonies initiated by FL or FL plus LIF suggested that these stimuli were adequate to initiate cell proliferation but were relatively inactive in supporting sustained cell proliferation.
The uniform macrophage progenitor content of FL plus LIF-stimulated blast colonies raised the possibility that the cells forming these colonies were merely macrophage-committed progenitor cells, essentially similar to those forming macrophage colonies when stimulated by M-CSF. However, M-CSF is not able to stimulate the formation of blast cell colonies. Furthermore, most of the clonogenic cells in blast colonies initiated by FL plus LIF were colony-forming cells, whereas most clonogenic cells in M-CSF-stimulated macrophage colonies merely formed small clusters of macrophages. These differences indicate that the cells forming FL-initiated blast colonies are ancestral to the progenitor cells forming M-CSF-stimulated macrophage colonies.
The ability of FL, on injection into mice, to greatly increase dendritic cell formation (8) raised the possibility that some or all of the “macrophage” progeny of FL-stimulated blast colony cells may in fact have been dendritic cells. Reculture of blast colony cells initiated by FL plus LIF in the presence of GM-CSF plus TNF-α, agents known to be necessary for dendritic cell maturation (17), did lead to the production of cells with the morphology and surface markers of dendritic cells. The FL-responsive blast colony-forming cells may therefore be a subset of stem cells committed to macrophage and dendritic cell formation. Although M-CSF stimulation does not lead to dendritic cell formation, its initial use to stimulate cell proliferation did not appear to suppress the subsequent ability of some of the cells to form dendritic cells.
LIF signals cells through a receptor complex including the gp130 chain shared by the receptors for OSM, IL-11, and IL-6 (11). Parallel studies showed that all three enhanced blast colony formation stimulated by FL, suggesting strongly that signaling through gp130 was necessary to elicit the observed enhancement responses. In the case of cultures stimulated by FL plus IL-6, a more complex outcome was observed because granulocyte and granulocyte/macrophage colony formation was also enhanced. Recloning of the blast colonies from the latter cultures led to the formation of small numbers of granulocytic or granulocyte/macrophage colonies, in addition to the more numerous macrophage colonies. This result may indicate either errors in the selection of “blast” colonies or that IL-6 might have some granulocyte commitment action on FL-responsive blast colony-forming cells.
Previous studies have shown that IL-6 and IL-11 can amplify the proliferation of immature hematopoietic cells when in combination with stem cell factor and other regulators (18, 19). A similar action has been reported for LIF when acting on human cells (20) and LIF, and IL-6 have been reported to amplify the proliferation of purified murine progenitor cells when used in combination with IL-3, GM-CSF, M-CSF, or SCF (21). The present observations have some similarities to this latter study but have identified a subset of FL-responsive precursor cells that is selectively responsive either to survival-enhancement or proliferative stimulation by regulators signaling through the gp130 receptor.
As was noted previously for cells forming blast colonies when stimulated by SCF (3, 4), the blast colony-forming cells responsive to FL stimulation that are enhanced by gp130 signaling, had no capacity for self-generation and appear to be at the maturation border between stem cells and progenitor cells. Blast colony-forming cells responsive to FL plus gp130 signaling were as numerous as cells forming blast colonies in response to stimulation by SCF, but appear to represent a distinct subset of mature hematopoietic stem cells with some capacity, when appropriately stimulated, to form dendritic cell progeny.
Table 1.
Enhancement of FL-stimulated blast colony formation by addition of LIF
| Stimulus | Mean number of colonies
|
Mean number of cells per blast colony | |||
|---|---|---|---|---|---|
| Blast | G | GM | M | ||
| LIF (10 ng/ml) | 0 ± 0 | 0.6 ± 0.4 | 0 ± 0 | 0.2 ± 0.2 | — |
| FL (500 ng/ml) | 7.0 ± 1.0 | 3.6 ± 1.0 | 1.5 ± 0.5 | 4.1 ± 0.7 | 100 |
| FL (500 ng/ml) plus LIF (10 ng/ml) | 19.6 ± 1.6 | 3.4 ± 1.1 | 1.2 ± 0.5 | 4.8 ± 0.7 | 590 |
Mean data ± standard errors from 10 separate experiments in which quadruplicate cultures of 50,000 C57BL bone marrow cells were stimulated for 7 days using 10 ng LIF, 500 ng FL, or a combination of both. Mean cell numbers per blast colony were determined from pools of 20-30 blast colonies. Blast, blast cell; G, granulocytic; GM, granulocyte/macrophage; M, macrophage colonies.
Table 2.
Enhancement of FL-stimulated blast colony formation by addition of OSM, IL-11, or IL-6
| Stimulus | Mean number of colonies
|
Mean cells per blast colony | Mean number of colony-forming cells per blast colony | |||
|---|---|---|---|---|---|---|
| Blast | G | GM | M | |||
| FL | 6 ± 1 | 5 ± 0 | 3 ± 2 | 5 ± 1 | 100 | 10 ± 8 |
| OSM | 0 | 2 ± 1 | 0 | 0 | — | — |
| IL-11 | 0 | 1 ± 0 | 0 | 0 | — | — |
| IL-6 | 0 | 25 ± 2 | 1 ± 0 | 1 ± 1 | — | — |
| FL + OSM | 12 ± 1 | 2 ± 1 | 0 | 6 ± 3 | 1,220 | 226 ± 185 |
| FL + IL-11 | 6 ± 1 | 2 ± 2 | 1 ± 1 | 4 ± 2 | 270 | 54 ± 39 |
| FL + IL-6 | 17 ± 8 | 31 ± 4 | 11 ± 3 | 18 ± 5 | 5,630 | 717 ± 527 |
Primary cultures contained 50,000 C57BL bone marrow cells stimulated by 500 ng FL, 100 ng OSM, 10 ng IL-11 or 100 ng IL-6 alone or in combination. Numbers of blast (blast), granulocytic (G), granulocyte/macrophage (GM), or macrophage (M) colonies were established from quadruplicate-stained cultures (mean values ± SD). Mean cell numbers per blast colony were established from pools of 20-30 blast colonies. In parallel experiments, the numbers of colony-forming cells per blast colony were established from secondary cultures stimulated by 10 ng M-CSF. Mean data from 20 blast colonies ± SD.
Acknowledgments
I am indebted to Sandra Mifsud, Ladina Di Rago, Angela D’Amico, and David Vremec for technical assistance with these experiments and to Dr. Ken Shortman for advice regarding dendritic cells. This work was supported by the Carden Fellowship Fund of the Anti-Cancer Council of Victoria, the National Health and Medical Research Council (Canberra), the AMRAD Corporation (Melbourne), and the National Institutes of Health (Bethesda, MD; Grant CA22556).
ABBREVIATIONS
- FL
Flk2-ligand
- G-CSF
granulocyte colony-stimulating factor
- GM-CSF
granulocyte/macrophage colony-stimulating factor
- IL
interleukin
- LIF
leukemia inhibitory factor
- M-CSF
macrophage colony-stimulating factor
- OSM
oncostatin M
- SCF
stem cell factor
- TNF-α
tumor necrosis factor α
References
- 1.Li C L, Johnson G R. Blood. 1994;84:408–414. [PubMed] [Google Scholar]
- 2.Metcalf D, Nicola N A. Proc Natl Acad Sci USA. 1991;88:6239–6243. doi: 10.1073/pnas.88.14.6239. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Metcalf D. Proc Natl Acad Sci USA. 1991;88:11310–11314. doi: 10.1073/pnas.88.24.11310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Metcalf, D. (1993) Stem Cells 11 (Suppl. 2), 1–11. [DOI] [PubMed]
- 5.Rasko J E J, Metcalf D, Rossner M T, Begley C G, Nicola N A. Leukemia. 1995;9:2058–2066. [PubMed] [Google Scholar]
- 6.Jacobsen S E, Okkenhaug C, Myklebust J, Veiby O P, Lyman S D. J Exp Med. 1995;181:1357–1363. doi: 10.1084/jem.181.4.1357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Piacibello W, Sanavio F, Garetto L, Severino A, Bergandi D, Ferrario J, Fagioli F, Berger M, Aglietta M. Blood. 1997;89:2644–2653. [PubMed] [Google Scholar]
- 8.Maraskovsky E, Brasel K, Teepe M, Roux E R, Lyman S D, Shortman K, McKenna H J. J Exp Med. 1996;184:1953–1962. doi: 10.1084/jem.184.5.1953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Metcalf D. Growth Factors. 1992;7:169–173. doi: 10.3109/08977199209046921. [DOI] [PubMed] [Google Scholar]
- 10.Metcalf D, Hilton D J, Nicola N A. Leukemia. 1988;2:216–221. [PubMed] [Google Scholar]
- 11.Kishimoto T, Akira S, Narazaki M, Taga T. Blood. 1995;86:1243–1254. [PubMed] [Google Scholar]
- 12.Metcalf D, Willson T A, Hilton D J, Di Rago L, Mifsud S. Leukemia. 1995;9:1556–1564. [PubMed] [Google Scholar]
- 13.Vremec D, Lieschke G J, Dunn A R, Robb L, Metcalf D, Shortman K. Eur J Immunol. 1997;27:40–44. doi: 10.1002/eji.1830270107. [DOI] [PubMed] [Google Scholar]
- 14.McKinstry W, Metcalf D, Li C-L. Leukemia. 1994;8:1726–1733. [PubMed] [Google Scholar]
- 15.Hilton D J, Nicola N A, Metcalf D. J Cell Physiol. 1991;146:207–215. doi: 10.1002/jcp.1041460204. [DOI] [PubMed] [Google Scholar]
- 16.McKinstry W J, Li C-L, Rasko J E J, Nicola N A, Johnson G R, Metcalf D. Blood. 1997;89:65–71. [PubMed] [Google Scholar]
- 17.Young J W, Steinman R M. Stem Cells. 1996;14:376–387. doi: 10.1002/stem.140376. [DOI] [PubMed] [Google Scholar]
- 18.Ikebuchi K, Wong G G, Clark S C, Ihle J N, Hirai Y, Ogawa M. Proc Natl Acad Sci USA. 1987;84:9035–9039. doi: 10.1073/pnas.84.24.9035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Neben S, Donaldson D, Sieff C, Mauch P, Bodine D, Ferrara J, Yetz-Adalpe J, Turner K. Exp Hematol. 1994;22:353–359. [PubMed] [Google Scholar]
- 20.Verfaillie C, McGlave P. Blood. 1991;77:263–270. [PubMed] [Google Scholar]
- 21.Keller J R, Gooya J M, Ruscetti F W. Blood. 1996;88:863–869. [PubMed] [Google Scholar]