Transforming growth factor beta: is it a downregulator of stem cell inhibition by macrophage inflammatory protein 1 alpha?

Transforming growth factor beta 1 (TGF-beta 1) and macrophage inflammatory protein 1 alpha (MIP-1 alpha) have recently been identified as potent inhibitors of hemopoietic stem cell proliferation. From previous studies, these molecules appear to have similar functions in the control of stem cell proliferation. This study was designed to investigate the relationship, if any, between these two negative regulators in an attempt to elucidate possible distinctive roles for each within the hemopoietic system. We report here that both MIP-1 alpha and TGF-beta are capable of inhibiting the same stem cell population (colony-forming unit [CFU]-A/CFU-S) with similar potencies. We further show that TGF-beta potently inhibits MIP-1 alpha gene expression in bone marrow-derived macrophages, the presumed source of MIP-1 alpha in the bone marrow. This inhibition is not specific to MIP-1 alpha in that expression of MIP-1 beta, a related molecule that does not exhibit potent stem cell inhibitory properties, is inhibited in a similar manner. The inhibition of MIP-1 alpha gene expression is also seen as a reduction in MIP-1 alpha protein production, which markedly decreases 24 h after treating RAW 264.7 cells, a murine macrophage cell line, with TGF-beta. These in vitro results suggest that in the presence of active TGF-beta in vivo, and in the absence of upregulators of MIP-1 alpha transcription, very little MIP-1 alpha will be produced. To address how MIP-1 alpha's target cells, the stem cells, would respond to TGF-beta, and the consequently low levels of MIP-1 alpha produced, we analyzed the effect of TGF-beta on MIP-1 alpha receptor levels on FDCP-MIX cells, a murine stem cell line. We show that TGF-beta (100 pM) reversibly downregulates MIP-1 alpha receptor levels on these cells to a maximum of 50-70% after 24 h. This level of downregulation does not change upon increasing the concentration of TGF-beta or the length of exposure of the cells to TGF-beta. Scatchard analysis shows that TGF-beta downregulates MIP-1 alpha receptor numbers with no change in affinity of the remaining receptors. These results suggest that TGF-beta may be capable of interfering with MIP-1 alpha's role as a stem cell inhibitor. Indeed, they suggest that in the presence of active TGF-beta in vivo, MIP-1 alpha is at best a weak contributor to the overall physiological inhibition of stem cells.

Requirement of MIP-1 alpha for an inflammatory response to viral infection.

Macrophage inflammatory protein-1 alpha (MIP-1 alpha) is a chemokine that has pro-inflammatory and stem cell inhibitory activities in vitro. Its biologic role in vivo was examined in mice in which the gene encoding MIP-1 alpha had been disrupted. Homozygous MIP-1 alpha mutant (-/-) mice were resistant to Coxsackievirus-induced myocarditis seen in infected wild-type (+/+) mice. Influenza virus-infected -/- mice had reduced pneumonitis and delayed clearance of the virus compared with infected +/+ mice. The -/- mice had no overt hematopoietic abnormalities. These results demonstrate that MIP-1 alpha is an important mediator of virus-induced inflammation in vivo.

The myeloproliferative sarcoma virus causes transformation or erythroid progenitor cells in vitro.

The myeloproliferative sarcoma virus induces spleen focus formation in vivo and transforms fibroblasts in vitro. We showed in this study that in vitro infection of spleen or bone marrow cells from susceptible mice with the myeloproliferative sarcoma virus leads to the formation of erythroid bursts. Under optimal conditions erythroid bursts formed in the absence of added erythropoietin, but the addition of as little as 0.05 U of erythropoietin per ml to infected cultures resulted in a significant increase in numbers of erythroid bursts and the proportion of hemoglobinized cells. A comparison of the kinetics of burst formation and the size of the induced bursts with those induced with Friend virus suggested that either sarcoma virus such as the myeloproliferative sarcoma virus or the target cells for the two viruses were not the same. Density characterization and heat lability studies indicated that the increased erythroid proliferation in vitro was a virus-induced event, but the possibility that the induced erythroid burst formation is mediated via interaction with a nonerythroid target cell and subsequent release of a soluble factor cannot be ruled out.

CD34-positive cells isolated from cryopreserved peripheral-blood progenitor cells can be expanded ex vivo and used for transplantation with little or no toxicity.

PURPOSE: The objectives of this phase I study were to assess the feasibility of using cryopreserved peripheral-blood progenitor cells (PBPC) for large-scale CD34 selection and subsequent expansion, and the safety of their use for reinfusion following chemoradiotherapy. PATIENTS AND METHODS: For 10 patients with nonmyeloid malignancy, an aliquot from a PBPC harvest was recovered from liquid nitrogen, and CD34 selected using the Isolex system (Baxter Healthcare, Newbury, United Kingdom) and expanded for 8 days ex vivo in a medium free of animal proteins but supplemented with autologous serum, stemcell factor (SCF), interleukin-1 beta (IL-1 beta), IL-3, IL-6, and erythropoietin. RESULTS: The mean increase for cell number was 21-fold, for colony-forming units-granulocyte/macrophage (CFU-GM) 139-fold, and for burst-forming units-erythroid (BFU-E) 114-fold. The expanded cells were reinfused in tandem with unmanipulated material (> or = 25 x 10(4) CFU-GM/kg). The patients did not experience any adverse effects immediately on cell infusion or within 48 hours. The 10 index patients were compared with 10 historical controls for parameters of myelosuppressive morbidity. In this small study, there were no differences in either neutrophil or platelet recovery between the patients who received expanded cells and historical controls. CONCLUSION: These data demonstrate that CD34 cells can successfully be selected from cryopreserved material, expanded ex vivo on a large scale, and safely reinfused following myeloablative conditioning regimens.

Perturbations of multiple hemopoietic lineages in retrovirus-induced murine malignant histiocytosis.

A fatal systemic proliferation of malignant histiocytes resembling human malignant histiocytosis was induced in susceptible mice following infection with the murine retrovirus malignant histiocytosis sarcoma virus (MHSV). It is shown that MHSV additionally caused profound alterations of erythropoiesis, granulocytopoiesis and thrombocytopoiesis, and in the hemopoietic stem cell compartment. In the erythroid lineage, MHSV induced a normocytic peripheral anemia, which was paralleled by an unphysiologic, multifocal clonal expansion of erythroid blasts in the spleen. These cells were not transformed and appeared to have a maturation defect since blood reticulocytes did not increase above control values. Moreover, MHSV exerted cytopathic effects on neutrophilic granulocytes and megakaryocytes, since their numbers transiently decreased in the spleen, and agranulocytosis and thrombocytopenia was observed in the blood. Nonetheless, regeneration was found in both lineages at later stages of the infection, which was accompanied by a terminal granulocytosis. The number of lineage-committed and multipotential colony-forming cells in the CFU-S assay increased transiently, but decreased to very low levels in the final stages of the disease. Thus, the studies demonstrate that the same etiologic agent, MHSV, had different effects on hemopoietic cells, which included malignant transformation, hyperproliferative and cytopathic effects.

Mixed colony formation in vitro by the heterogeneous compartment of multipotential progenitors in human bone marrow.

The physiology of the human haemopoietic primitive progenitor populations can be studied in normal and disease states by clonal in vitro cultures in which the primitive progenitor cells proliferate and differentiate to form mixed colonies. For many applications it is essential that such assays detect a high proportion of primitive progenitor cells. We describe an in vitro assay which detects a high incidence of human CD34+ multipotential progenitor cells. Bone marrow mononuclear cells (MNC) or selected CD34+ cells were plated at low cell concentrations in semisolid agar cultures with synergizing growth factor combinations. The optimum growth factor combination of conditioned medium from Mia PaCa-2 cells (Mia-CM), recombinant granulocyte-macrophage colony-stimulating factor and recombinant stem cell factor (SCF) supported the formation of macroscopic (> or = mm) colonies (97% of which were multilineage), at an average incidence of 250/10(5) MNC. The colony-forming cells (human colony-forming unit, type A) detected, showed a low cycling status (7.3%) and the macroscopic colonies had a high replating efficiency (46%), reflecting their probable primitive nature. This assay should prove invaluable, for studies on the regulation of proliferation of the multipotential compartment and in studies involving the assessment of these cells in transplantation and neoplastic disease.

Expression of stem cell inhibitor (SCI) gene in patients with bone marrow failure.

Stem cell inhibitor (SCI) has been shown to inhibit the proliferation of primitive progenitors. The inhibitor, a product of bone marrow macrophages, activated lymphocytes, and monocytes, is identical to macrophage inflammatory protein (MIP-1 alpha). We report homologous (SCI/hMIP-1 alpha) sequences in freshly isolated lymphocytes, monocytes, and granulocytes and have found that SCI mRNA can be induced in monocytes by lipopolysaccharide (LPS) and interleukins 1, 2, and 6. In contrast, interferon gamma (IFN-gamma) decreases the expression of SCI/hMIP-1 alpha. Although only a low level expression of SCI/hMIP-1 alpha mRNA can be detected in normal human bone marrow nucleated cells (NCBM), very significant increases in the levels of SCI/hMIP-1 alpha RNA transcripts are observed in NCBM from patients with aplastic anemia (AA) and myelodysplastic syndrome (MDS). These data suggest that the expression of SCI/hMIP-1 alpha in bone marrow may reflect dysregulated cytokine production and activation of the immune system that may possibly contribute to disease progression.

Inhibitors of haemopoiesis and their potential clinical relevance.

Current cytotoxic treatment regimens are most frequently dose-limited by the problem of myelotoxicity, and this could theoretically be prevented or reduced by the use of stem-cell inhibitors, since protection of this compartment during treatment could result in a more favourable outcome in terms of bone-marrow regeneration. Several negative stem-cell regulators have been identified, including macrophage inflammatory protein-1 alpha, transforming growth factor-beta, tumour necrosis factor-alpha, tetrapeptide and pentapeptide. All of these molecules have been shown to inhibit the proliferation of normal haemopoietic progenitors in bone marrow, and stem-cell protection from cytotoxic agents both in vitro and in vivo has been demonstrated. The potential use of inhibitors for the purging of tumour cells from stem-cell grafts is suggested by the observation that there is a differential response between normal and leukaemic progenitors to some inhibitors.

Hemopoietic stem cell inhibitor (SCI/MIP-1 alpha) also inhibits clonogenic epidermal keratinocyte proliferation.

The maintenance and regulation of continuously renewing tissues is ultimately controlled at the level of stem-cell proliferation. We have recently identified a reversible inhibitor of hemopoietic stem-cell proliferation (stem-cell inhibitor [SCI]), which is identical to the macrophage inflammatory protein, MIP-1 alpha, a 69-amino-acid heparin-binding cytokine. To test the cell/tissue specificity of the inhibition of proliferation by SCI/MIP-1 alpha, we have investigated its activity on epidermal keratinocytes, the principal cell type of another continuously renewing tissue. Here we show that SCI/MIP-1 alpha inhibits the proliferation of epidermal keratinocytes in vitro and that the MIP-1 alpha mRNA is present in epidermal Langerhans cells but not in keratinocytes. This suggests an important growth regulatory function for SCI/MIP-1 alpha in keratopoiesis, as well as hemopoiesis, and may also indicate a novel role for the epidermal Langerhans cell. As SCI/MIP-1 alpha can inhibit the proliferation of embryologically distinct precursor cells, this raises the possibility that it may also function in a number of other tissues.

The cytostatic effects of alpha-interferon may be mediated by transforming growth factor-beta.

There is some evidence to suggest that transforming growth factor-beta (TGF-beta) mediates the cytostatic effects of the anti-oestrogen tamoxifen. In this study we have demonstrated that alpha-interferon has a significant anti-proliferative effect on the oestrogen receptor-positive human breast cancer cell line ZR-75. There is decreased phenotypic expression of the oestrogen receptors (to about 30% of control values) and increased TGF-beta mRNA. Under the growth conditions used here, ZR-75 cells had approximately 5800 TGF-beta binding sites per cell, with an apparent dissociation constant of 70 pm, and we have shown that the anti-proliferative effects of alpha-interferon can be reduced by 60% by co-treating the cells with a TGF-beta polyclonal antibody. The cytostatic effects of alpha-interferon may therefore be mediated by TGF-beta in this human breast cancer cell line.

Preserved long-term repopulation and differentiation properties of hematopoietic grafts subjected to ex vivo expansion with stem cell factor and interleukin 11.

BACKGROUND: The ex vivo expansion of hematopoietic grafts has been proposed as an efficient procedure for improving the hematological recovery of recipients. The fate of the long-term repopulating cells during the ex vivo manipulation of the graft is, however, a critical issue in ex vivo expansion protocols and ultimately will define the applicability of this new technology in hematopoietic transplants. METHODS: The repopulating ability of mouse hematopoietic samples was determined by means of bone marrow (BM*) competition assays, using congenic strains that express the pan-leukocyte Ly-5.1 and Ly-5.2 antigens. The generation of potential changes in the repopulating properties of human hematopoietic samples subjected to ex vivo expansion was determined by comparing the engraftment of fresh and ex vivo-manipulated CD34+ cord blood cells in irradiated nonobese diabetic/severe-combined immunodeficient (NOD/SCID) mice. RESULTS: Under our optimized conditions of mouse BM incubation (stem cell factor plus interleukin-11, either with or without macrophage inflammatory protein-1alpha or Flt3 ligand), both the short-term and the mid-term repopulating ability of the ex vivo-expanded samples were significantly improved when compared with fresh samples. In the long-term, no changes in the repopulation and differentiation properties of the graft were observed as a result of the ex vivo expansion process. As deduced from the analysis of NOD/SCID mice transplanted with fresh and ex vivo expanded human CD34+ cord blood cells, the in vitro stimulation mediated by SCF/IL-11/FLT3L was capable of preserving the ability of the grafts to repopulate the lympho-hematopoiesis of recipients for at least 3 months. CONCLUSION: These results indicate that under our optimized conditions of ex vivo expansion, the amplification of the hematopoietic progenitors responsible for the short- and mid-term repopulating properties of the graft can take place without compromising the long-term lympho-hematopoietic repopulating properties.

Synergistic interactions allow colony formation in vitro by murine haemopoietic stem cells.

A clonogenic assay for cells that give rise to macroscopic colonies in agar or methyl cellulose cultures using untreated, normal murine bone marrow as a source of stem cells is described. We have characterized the clonogenic cell, which has been designated CFU-A, by comparing its properties with those of multipotential stem cells (assayed as CFU-S) and lineage-restricted progenitor cells (assayed as GM-CFC). The investigations have included assessment of proliferative status and response to CFU-S proliferation regulators, response to 5-fluorouracil, buoyant cell density, radial distribution in the femur and response to ionizing radiation. We conclude that the CFU-A has properties in common with CFU-S that differ from those of GM-CFC. The data are consistent with the CFU-A assay detecting part of the multipotential stem cell population also detected by spleen colony formation.

CD34 positive PBPC expanded ex vivo may not provide durable engraftment following myeloablative chemoradiotherapy regimens.

We have previously demonstrated that CD34+ cells, selected from peripheral blood progenitor cells (PBPC), can be expanded in ex vivo culture and can be infused in tandem with unmanipulated PBPC with little or no toxicity. In this study, four patients (two non-Hodgkin's lymphoma (NHL), two multiple myeloma (MM)) received myeloablative conditioning prior to stem cell rescue using ex vivo expanded cells alone. The two patients with NHL received cyclophosphamide and total body irradiation (CY/TBI) and the two patients with MM, busulphan and melphalan (Bu/M). One case received an inadequate CFU-GM dose, despite expansion, and in one case the expanded cells were contaminated. No definitive conclusions may therefore be drawn concerning engraftment in these two cases. However, the other two cases received high doses of committed progenitors. Following infusion of the expanded material, all four patients failed to show sustained neutrophil engraftment and none showed evidence of platelet engraftment. Back-up, unmanipulated PBPC were therefore infused on days 14, 34, 32 and 28 and subsequently all four cases achieved satisfactory engraftment of both neutrophils and platelets. In conclusion, we feel that, CD34+ cells, expanded ex vivo using the conditions described in this report, may not provide durable engraftment following fully myeloablative conditioning.

Changes in genome composition of the Friend virus complex in erythroleukemia cells during the course of differentiation induced by dimethyl sulfoxide.

The Friend spleen focus-forming virus (SFFV) complex released by Friend virus-transformed erythroid cells has been analyzed with respect to changes in the genome composition that may occur during induction of erythropoiesis with dimethyl sulfoxide. It is shown that: (a) There are three types of virus particles, one with buoyant density 1.20 g/ml, one with density 1.17 g/ml (the density of the cloned lymphatic leukemia virus helper component of the complex), and a major fraction that has a density of 1.14 g/ml. (b) Three RNA subunits-35S, 32S, and 30S-have previously been shown to be detectable in the Friend virus complex. The 1.20-g/ml particles contain only 30S RNA, whilst the 1.14- to 1.17-g/ml particles contain a mixture consisting of predominantly 30S and 32S RNA and about 5-10% 35S RNA. (c) Induction of differentiation results in an increase in the 1.14-g/ml particles and 32S RNA. The amount of 30S RNA does not change. (d) Hybridization of the different genomic viral RNAs with full-length virus cDNA shows that the 30S RNA (of induced and uninduced Friend virus) is more closely related to the 32S RNA of the induced Friend virus than to the 32S RNA of the constitutively released Friend virus. (e) The 30S RNA contains SFFV-specific sequences. (f) A hypothesis is presented in which the induction of the new 32S RNA species is related to the increase of SFFV activity and to a specific function of the SFFV during induction of erythropoiesis.

The absence of globin gene sequences from friend leukaemia virus RNA.

Complementary DNA to mouse globin mRNA was prepared using viral reverse transcriptase. Molecular hybridisation of this cDNA to Friend virus RNA demonstrates the absence of globin gene sequences.