D3: a gene induced during myeloid cell differentiation of Linlo c-Kit+ Sca-1(+) progenitor cells.

In an effort to characterize molecular events contributing to lineage commitment and terminal differentiation of stem/progenitor cells, we have used differential display reverse transcription polymerase chain reaction (DDRT-PCR) and cell lines blocked at two distinct stages of differentiation. The cell lines used were EML, which is representative of normal multipotential primitive progenitors (Sca-1(+), CD34(+), c-Kit+, Thy-1(+)) able to differentiate into erythroid, myeloid, and B-lymphoid cells in vitro, and MPRO, which is a more committed progenitor cell line, with characteristics of promyelocytes able to differentiate into granulocytes. One clone isolated by this approach was expressed in MPRO but not in EML cells and contained sequence identical to the 3' untranslated region of D3, a gene cloned from activated peritoneal macrophages of unknown function. We have observed a novel pattern of D3 gene expression and found that D3 is induced in EML cells under conditions that promote myeloid cell differentiation (interleukin-3 [IL-3], stem cell factor [SCF], and all-trans-retinoic acid [atRA]) starting at 2 days, corresponding to the appearance of promyelocytes. D3 RNA expression reached a maximum after 5 days, corresponding to the appearance of neutrophilic granulocytes and macrophages, and decreased by day 6 with increased numbers of differentiated neutrophils and macrophages in vitro. Induction of D3 RNA in EML was dependent on IL-3 and was not induced in response to SCF or atRA alone or SCF in combination with 15 other hematopoietic growth factors (HGF) tested. Similarly, D3 was not expressed in the normal bone marrow cell (BMC) counterpart of EML cells, Linlo c-Kit+ Sca-1(+) progenitor cells. D3 RNA expression was induced in these cells when cultured for 7 days in IL-3 plus SCF. A comparison of the expression of D3 RNA in cell lines and normal BMC populations demonstrated that D3 is induced during macrophage and granulocyte differentiation and suggests a potential physiological role for D3 in normal myeloid differentiation.

Engraftment of ex vivo expanded and cycling human cord blood hematopoietic progenitor cells in SCID mice.

The ability of human hematopoietic cells to engraft SCID mice provides a useful model in which to study the efficiency of retroviral gene transfer and expression in primitive stem cells. In this regard, it is necessary to determine whether SCID mice can be engrafted by cycling human hematopoietic progenitor cells. Human cord blood cells from 12 different donors were cultured in vitro for 6 days with interleukin-3 and stem cell factor. Phenotypic analysis indicated that hematopoietic cells were induced to cycle and the number of progenitors was expanded, thus making them targets for retroviral gene transfer. The cells were then transferred to SCID mice. Human hematopoietic progenitor cell engraftment was assessed up to 7 weeks later by growth of human progenitor cells in soft agar. After in vitro culture under conditions used for retroviral gene transfer, human cord blood hematopoietic cells engrafted the bone marrow and spleen of SCID mice. Interestingly, cultured cord blood cells engrafted after intraperitoneal but not after intravenous injection. Furthermore, engraftment of cord blood cells was observed in mice receiving no irradiation before transfer of the human cells, suggesting that competition for space in the marrow is not a limiting factor when these cells have been cultured. Administration of human cytokines after transfer of human cord blood cells to SCID mice was also not required for engraftment. Thus, engraftment of SCID mice with human hematopoietic cells cultured under conditions suitable for gene transfer may provide an in vivo assay for gene transfer to early human hematopoietic progenitor cells.

Raf-1 protein is required for growth factor-induced proliferation of primitive hematopoietic progenitors stimulated with synergistic combinations of cytokines.

Raf-1 is a serine/threonine kinase that has been identified as a component of growth factor-activated signal transduction pathways, and is required for growth factor-induced proliferation of leukemic cell lines and colony formation of hematopoietic progenitors stimulated with single colony-stimulating factors, which promote the growth of committed hematopoietic progenitor cells. However, it is known that the most primitive progenitors in the bone marrow require stimulation with multiple cytokines to promote cell growth. We have determined that c-raf antisense oligonucleotides inhibit the growth of murine lineage-negative progenitors stimulated with two-, three- and four-factor combinations of growth factors, including GM-CSF + interleukin (IL)- 1, IL-3 + steel factor (SLF), IL-3 + IL-11 + SLF and IL-3 + IL-11 + SLF + G-CSF. In addition, c-raf antisense oligonucleotides inhibit the synergistic response of the MO7e human progenitor cell line induced to proliferate with IL-3 + SLF (99%) or GM-CSF + SLF (99%). In contrast, c-raf antisense oligonucleotides only partially inhibited day 14 colony formation of CD34+ human progenitors stimulated with IL-3 + SLF (50%) or GM-CSF + SLF (55%) but completely inhibited day 7 colony formation. However, pulsing CD34+ cells with additional oligonucleotides on day 7 of the colony assay further inhibited day 14 colony formation (70%-80%). Furthermore, a comparison of the effect of c-raf antisense oligonucleotides on the synergistic response of normal human fetal liver cells in [3H]thymidine incorporation assays and colony assays showed strong inhibition in short-term proliferation assays and partial inhibition in 14-day colony assays. Taken together, these results demonstrate that partial inhibition of colony formation of primitive human progenitors stimulated with multiple growth factors is a result of the length (14 days) of the human colony assay and does not represent a differential requirement of primitive progenitors for Raf-1. Thus Raf-1 is required for the proliferation and differentiation of primitive hematopoietic progenitor cells stimulated with synergistic combinations of cytokines.

Direct synergistic effects of leukemia inhibitory factor on hematopoietic progenitor cell growth: comparison with other hematopoietins that use the gp130 receptor subunit.

Because leukemia inhibitory factor (LIF) has little or no effect on murine hematopoietic progenitor cell growth yet enhances hematopoiesis in vivo, we sought to determine whether the effects of LIF were directly or indirectly mediated, or a combination of both. Although LIF alone or in combination with granulocyte-macrophage colony-stimulating factor (GM-CSF) or interleukin-3 (IL-3) has no effect on colony formation of unfractionated bone marrow cells (BMCs), it enhances M-CSF-induced colony formation. In comparison, LIF synergizes with IL-3, GM-CSF, M-CSF, and Steel Factor (SLF) to promote the colony formation of partially purified lineage-negative (Lin-) BM progenitors without altering their differentiation. These effects were directly mediated since identical results were observed in single-cell assays. Comparing the effect of LIF with other members of this subclass of hematopoietins (IL-6, oncostatin M [OSM], and ciliary neurotrophic factor [CNTF]), we found that while LIF and IL-6 equally synergize with M-CSF and SLF to promote the colony formation of Lin- BMCs, OSM, and CNTF have no effect. In agreement with OSMs ability to directly bind gp130, preincubation of BMCs with OSM inhibits progenitor cell growth stimulated by the combination of LIF or IL-6 plus SLF. LIF can also directly enhance the growth of further purified more primitive Lin- c-kit+ progenitor cells in the presence of IL-3, GM-CSF, or SLF. Thus, LIF can directly synergize with growth factors to promote the proliferation of purified hematopoietic progenitors, suggesting that the direct effects of LIF on hematopoietic cell growth can, in part, explain the observed hematopoietic effects in vivo. This is a US government work. There are no restrictions on its use.

Targeted gene transfer to human hematopoietic progenitor cell lines through the c-kit receptor.

In this report, we describe a novel gene therapy approach for hematopoietic stem/progenitor cells using a specific receptor-mediated gene transfection procedure to target c-kit+ cell lines. The vector consists of plasmid DNA containing a luciferase reporter gene that is condensed by electrostatic forces with polylysine (PL) covalently linked to streptavidin (binds biotinylated ligand) and PL covalently linked to adenovirus (AD; to achieve endosomal lysis) with the final addition of biotinylated steel factor (SLF-biotin). Targeted transfection of growth factor-dependent hematopoietic progenitor cell lines that express c-kit showed specific luciferase gene expression over cell lines that did not express c-kit. This effect was dependent on the dose of SLF-biotin and was competed by excess SLF or with monoclonal antibodies that recognize c-kit and block the binding of SLF to its receptor. Maximum transfection efficiency (> 90%) requires a 2-hour incubation period of the vector with the cells, and maximum gene expression occurred 30 hours later. Removal of the endosomalytic agent, AD, from the vector resulted in the loss of gene expression. Vector targeting was versatile and could be changed by the addition of other biotinylated ligands. In principle, this vector should be broadly applicable to deliver genes to hematopoietic stem/progenitor cells in vitro and in vivo.

Engraftment of SCID mice by human bone marrow hematopoietic cells cultured in vitro: an in vivo model for human gene transfer.

Demonstration of the ability of fresh human hematopoietic cells to engraft severe combined immuno-deficient (scid) mice has provided an in vivo assay for expansion and maturation of early human progenitor cells. However, engraftment of cultured hematopoietic cells has been difficult to achieve. We wished to further develop this model as an in vivo assay for efficiency of retroviral gene transfer and expression in the differentiated progeny of adult human bone marrow progenitor cells. Human bone marrow cells were cultured in vitro for six days under conditions suitable for infection by retroviral vectors prior to transfer to irradiated scid mice. Cultured human bone marrow cells introduced by both intravenous (i.v.) and intraperitoneal (i.p.) injection persisted in the bone marrow, spleen and peritoneum of recipient animals up to four weeks after transfer. Following irradiation scid mice receiving cultured human bone marrow cells by either i.v. or i.p. routes demonstrated engraftment of the bone marrow and spleen as determined by the growth of human hematopoietic progenitors in soft agar. By flow cytometric analysis human cells were also detected in the peritoneum of mice receiving cultured human bone marrow cells i.p. These results suggest that the transfer of cultured human bone marrow cells to scid mice with the subsequent engraftment of these cells in the bone marrow, spleen and peritoneum of recipients can routinely occur. This provides an in vivo model for retroviral gene transfer to human cells.

Raf-1 protein is required for growth factor-induced proliferation of hematopoietic cells.

Raf-1 is a 74-kD serine/threonine kinase located in the cell cytoplasm that is activated by phosphorylation in cells stimulated with a variety of mitogens and growth factors, including hematopoietic growth factors. Using c-raf antisense oligonucleotides to block Raf-1 expression, we have established that Raf-1 is required for hematopoietic growth factor-induced proliferation of murine cell lines stimulated by growth factors whose receptors are members of several different structural classes: (a) the hematopoietin receptor family, including interleukin (IL)-2, IL-3, IL-4, granulocyte colony-stimulating factor, granulocyte/macrophage colony-stimulating factor (GM-CSF), and erythropoietin; (b) the tyrosine kinase receptor class, including Steel factor and CSF-1; and (c) IL-6, leukemia inhibitory factor, and oncostatin M, whose receptors include the gp130 receptor subunit. Although results of previous experiments had suggested that IL-4 does not phosphorylate or activate the Raf-1 kinase, c-raf antisense oligonucleotides inhibited IL-4-induced proliferation of both myeloid and T cell lines, and IL-4 activated Raf-1 kinase activity in an IL-4-dependent myeloid cell line. In colony assays, c-raf antisense oligonucleotides completely inhibited colony formation of unseparated normal murine bone marrow cells stimulated with either IL-3 or CSF-1 and partially inhibited cells stimulated with GM-CSF. In addition, c-raf antisense oligonucleotides completely inhibited both IL-3- and GM-CSF-induced colony formation of CD34+ purified human progenitors stimulated with these same growth factors. Thus, Raf-1 is required for growth factor-induced proliferation of leukemic murine progenitor cell lines and normal murine and human bone marrow-derived progenitor cells regardless of the growth factor used to stimulate cell growth.

Distinct and overlapping direct effects of macrophage inflammatory protein-1 alpha and transforming growth factor beta on hematopoietic progenitor/stem cell growth.

Both transforming growth factor beta (TGF beta) and macrophage inflammatory protein 1 alpha (MIP-1 alpha) have been shown to be multifunctional regulators of hematopoiesis that can either inhibit or enhance the growth of hematopoietic progenitor cells (HPC). We report here the spectrum of activities of these two cytokines on different hematopoietic progenitor and stem cell populations, and whether these effects are direct or indirect. MIP-1 alpha enhances interleukin-3 (IL-3)/and granulocyte-macrophage colony-stimulating factor (GM-CSF)/induced colony formation of normal bone marrow progenitor cells (BMC) and lineage-negative (Lin-) progenitors, but has no effect on G-CSF or CSF-1/induced colony formation. Similarly, TGF beta enhances GM-CSF/induced colony formation of normal BMC and Lin- progenitors. In contrast, TGF beta inhibits IL-3/ and CSF-1/induced colony formation of Lin- progenitors. The effects of MIP-1 alpha and TGF beta on the growth of Lin- progenitors were direct and correlate with colony formation in soft agar. Separation of the Lin- cells into Thy-1 and Thy-1lo subsets showed that the growth of Thy-1lo Lin- cells is directly inhibited by MIP-1 alpha and TGF beta regardless of the cytokine used to stimulate growth (IL-3), GM-CSF, or CSF-1). In contrast, two other stem cell populations (0% to 15% Höechst 33342/Rhodamine 123 [Hö/Rh123] and Lin-Sca-1+ cells) were markedly inhibited by TGF beta and unaffected by MIP-1 alpha. Furthermore, MIP-1 alpha has no effect on high proliferative potential colony-forming cells 1 or 2 (HPP-CFC/1 or /2) colony formation in vitro, whereas TGF beta inhibits both HPP-CFC/1 and HPP-CFC/2. Thus, MIP-1 alpha and TGF beta are direct bidirectional regulators of HPC growth, whose effects are dependent on other growth factors present as well as the maturational state of the HPC assayed. The spectrum of their inhibitory and enhancing activities shows overlapping yet distinct effects.

The growth response of Lin-Thy-1+ hematopoietic progenitors to cytokines is determined by the balance between synergy of multiple stimulators and negative cooperation of multiple inhibitors.

The present studies investigated the balance of positive and negative growth signals in direct regulation of hematopoiesis. Interleukin-3 (IL-3) combined with Steel factor (SLF) optimally stimulated proliferation of Lin-Thy-1+ murine bone marrow progenitors in single-cell assays, and that proliferation was inhibited more than 90% by transforming growth factor-beta 1 (TGF-beta 1). Colony-stimulating factor-1 (CSF-1), granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-1, or IL-6 as a third stimulatory growth factor was incapable of counteracting the TGF-beta 1-mediated inhibition of IL-3-plus-SLF-stimulated growth, while G-CSF slightly enhanced the number of TGF-beta 1-resistant clones. As a fourth factor, only IL-1 could partially overcome the TGF-beta 1-induced growth inhibition. While the presence of a cocktail of five additional stimulatory growth factors did not enhanced the frequency of single Lin-Thy-1+ progenitors proliferating in response to IL-3 plus SLF, the number of responding progenitors in the presence of TGF-beta 1 was enhanced nine-fold. Furthermore, tumor necrosis factor-alpha (TNF-alpha) and interferon-gamma (IFN-gamma), but not macrophage inflammatory protein-1 alpha (MIP-1 alpha), cooperated with TGF-beta 1 to reverse the proliferative effects of multiple stimulatory cytokines, resulting in 76% inhibition. Thus, the direct effects of single inhibitory factors on hematopoietic progenitor cell growth can be reversed by multiple stimulatory growth factors, and negative growth factors can directly cooperate to suppress progenitor cell growth stimulated by multiple positive-acting factors.

Interleukin-1 (IL-1) directly and indirectly promotes hematopoietic cell growth through type I IL-1 receptor.

Interleukin-1 (IL-1) has been shown to stimulate hematopoietic progenitor cell growth both in vitro and in vivo. Although IL-1 alone lacks the ability to promote hematopoietic progenitor growth in vitro, it is a potent synergistic factor in combination with other colony-stimulating factors (CSFs). Because it was unknown whether type I (p80), type II (p68), or other IL-1-binding proteins mediated the synergistic effects of IL-1 on purified progenitor cells, we used the difference in immunoreactivity between type I and type II IL-1 receptor (IL-1R) to better assess the role of these receptors in hematopoietic progenitor growth. Therefore, the synergistic effects of IL-1 alpha on IL-3-, CSF-1-, and granulocyte macrophage (GM)-CSF-induced progenitor growth, both in CFU-c and single-cell assays, were determined in the presence of monoclonal antibodies (MoAbs) 35F5 and 4E2 that block the binding of IL-1 alpha to type I and type II IL-1R, respectively. The synergistic effect of IL-1 alpha on IL-3 responsive Lin- and Lin(-)-Thy-1+ progenitors was indirectly mediated and could be inhibited by MoAb 35F5. In contrast, IL-1 alpha directly synergized with CSF-1 and GM-CSF to promote progenitor cell growth. The direct synergistic effect of IL-1 alpha on CSF-1-induced progenitor growth was observed in all progenitor populations examined (Lin-, Lin-Thy-1+, and Lin-Thy-1-) and was inhibited by MoAb 35F5. However, the direct synergistic effect of IL-1 alpha on GM-CSF-responsive progenitors. Lin- and Lin-Thy-1+, was partially inhibited by MoAb 35F5. In contrast, the MoAb antitype II IL-1R (MoAb 4E2) could not inhibit the direct synergistic effects of IL-1 alpha on CSF-1- or GM-CSF-induced progenitor growth. Thus, IL-1 alpha directly and indirectly stimulates the growth and differentiation of purified progenitors through the type I IL-1R but not the type II IL-1R.

Synergy between recombinant human IL-1 alpha (rHuIL-1) and M-CSF (rHuM-CSF) during the recovery of murine hematopoietic activity in myelosuppressed animals: abbreviated versus chronic administration of rHuM-CSF.

The ability of highly purified, recombinant human macrophage colony-stimulating factor (M-CSF) and recombinant human interleukin 1 alpha (IL-1) to rescue hematopoietic activity from the myelosuppressive effects of 5-fluorouracil (5-FU) was investigated in the C57Bl/6 mouse. IL-1 (q24 h x 4) stimulated granulopoietic recovery in the 5-FU-treated animals and reduced the period of severe neutropenia associated with this drug by 7 days. Chronic M-CSF administration (q24 h x 14), on the other hand, resulted in a modest retardation of granulocyte recovery, and, when combined with IL-1, the chronic administration of M-CSF significantly dampened the accelerated recovery of granulopoietic activity observed with IL-1 alone. Consistent with their effects on neutrophil recovery, IL-1 alone markedly enhanced the recovery of the granulocyte erythrocyte macrophage megakaryocyte colony-forming units (CFU-GEMM), macrophage colony-forming units (CFU-M), and erythroid burst-forming units (BFUe) in the marrow, whereas M-CSF failed to demonstrate a significant influence on the restoration of these hematopoietic progenitors (with the exception of delaying the recovery of the BFUe). Unexpectedly, the combination of IL-1 plus M-CSF (q24 h, days 1-4) followed by M-CSF (q24 h, days 5-14) resulted in a more than additive stimulation of progenitor recovery in both the marrow and the spleen that was observed as early as day 3 following 5-FU treatment. Furthermore, in the absence of protracted M-CSF administration on days 5-14, the 4-day rescue with a combination of IL-1 plus M-CSF also resulted in a more than additive effect on the recovery from 5-FU-induced neutropenia. Collectively, these observations demonstrated that IL-1 and M-CSF can interact synergistically to stimulate granulopoietic recovery in the 5-FU-treated animal. However, the data also suggest that the continued administration of M-CSF following the 4-day IL-1 plus M-CSF rescue may interfere with the restoration of neutrophils in the myelosuppressed animal.

Interleukin-1 alpha protects against the toxicity associated with combined radiation and drug therapy.

A dose of total body irradiation sufficient to cause 70% mortality within 30 days (9.5 Gy) and maximally tolerated doses of either adriamycin (10 mg/kg) or cis-dichlorodiammine platinum (8 mg/kg) were administered to C57B1/6 mice and animal survival used as an index of toxicity. Whereas the nature of the toxicity resulting from the radiation alone was hematopoietic, the addition of either drug to the total body irradiation resulted in a pattern of animal death more consistent with that of gastrointestinal toxicity (100% dead within 7 days). However, if the radiation was delivered as a regional abdominal exposure, rather than total body, the gastrointestinal death observed following the combination of total body irradiation drug was not observed. The administration of 2.5 x 10(5) U IL-1 v24 hr prior to total body irradiation demonstrated significant protection against this dose of radiation (90% survival vs 30% survival). Similar protection was also observed when the IL-1 was administered 24 hr prior to the combination of total body irradiation with either drug. While these observations suggested that the IL-1 was protecting against gastrointestinal toxicity, subsequent studies demonstrated that IL-1, in addition to accelerating hematopoietic recovery following radiation insult, was equally effective in advancing the repopulation of the stem cell (CFU-GEMM) and progenitor cell (CFU-M and CFU-GM) compartments following drug treatment. Collectively, the data from these studies demonstrate that the lethal effects resulting from combined total body irradiation + drug treatment contain both a gastrointestinal and a hematopoietic component.

Altered radioprotective properties of interleukin I alpha (IL-1) in non-hematologic tumor-bearing animals.

The radioprotective properties of IL-1 were investigated in the respective murine hosts for the Lewis lung (LLca) and EMT-6 tumors. For these studies, doses of total body irradiation were selected for the C57B1/6 (9.5 Gy) and Balb/c (7.5 Gy) mice that resulted in a 60% mortality over a 30-day interval. When a "priming" dose of 2.5 x 10(5) U IL-1 was administered 24 hr prior to the radiation exposure, animal mortality was markedly reduced (60% vs 5-10%). Under identical experimental conditions, however, the presence of either the LLca or the EMT-6 tumors in their respective host strains was found to compromise the level of radioprotection conferred by this priming dose of IL-1. In Balb/c mice bearing the EMT-6 tumor, a priming dose of IL-1 resulted in only a modest level of radioprotection when compared to non-tumor-bearing control animals (median animal survival increased by 11.5 days). In C57B1/6 mice bearing the LLca tumor, IL-1 failed to demonstrate any evidence of radioprotection. Following a sublethal dose of total body irradiation, the appearance of an accelerated repopulation of the stem cell (8d CFUs and CFU-GEMM) and the myeloid progenitor (CFU-M) compartment in the marrow of the IL-1 primed EMT-6, but not the LLca, tumor-bearing animals was consistent with the hypothesis that the mechanism leading to radioprotection in IL-1 primed rodents involves an accelerated recovery of hematopoietic activity. It was also noted that the presence of the EMT-6 tumor was associated with an increase in the "radiosensitivity" of the Balb/c mouse. Collectively, these data suggest that the use of biological modifiers should be examined under altered physiological conditions prior to attempting to translate them into the clinic.

Tumor-induced altered gastrointestinal steady-states: absence of MHC restriction in the paraneoplastic gastrointestine.

The growth of a number of experimental rodent tumours including the Lewis lung tumour (LLca) progressively compromises the integrity of the host's gastroinestine by inducing cytokinetic alterations in the small bowel resembling those generally defining the intestinal phase of a graft-versus-host reaction (GVHR). To determine whether the induction of this paraneoplastic gastrointestine (PGI) involves, similar to a GVHR, a disparity between the MHC of the donor (LLca tumour) and the recipient (host), PGI development was evaluated in various LLca tumour-bearing murine strains that were either 'syngeneic' [C57BL/6 and BL/10 (H-2b)], 'semisyngeneic' [B6D2F1 (H-2bd) and B6C3F1 (H-2bk)] or 'allogeneic' [C3H/HeJ (H-2k) and DBA/2 (H-2d)] to the H-2b LLca tumour. The temporal appearance and magnitude of a PGI developing in either LLca-syngeneic or semi-syngeneic hosts, but not the allogeneic strains, suggested that the mechanism(s) involved in PGI development like the GVHR, was restricted by the MHC. Subsequent studies using congenic strains [B10.A (H-2k) and B10.D2/nSn (H-2d)], however, demonstrated that the mechanism(s) responsible for the PGI was restricted by the non-MHC loci of the C57BL mouse. These observations were supported by the appearance of a LLca-induced PGI in various B10.A congenic strains carrying mutations at the I-A or I-E/I-J loci of the MHC. Not unlike the intestinal phase of a GVHR, development of the PGI required the participation of enhanced mucosal mast cells which were limited in the WCB6F1 (S1/S1d) but not the (+/+) murine strains. These observations are discussed in light of the postulated premature migration of immature thymocytes that accompany tumour growth and their ability to non-specifically enhance (or suppress) cell mediated immune reactions in the host.