NUP98 dysregulation in myeloid leukemogenesis.

Nucleoporin 98 (NUP98) is a component of the nuclear pore complex that facilitates mRNA export from the nucleus. It is mapped to 11p15.5 and is fused to a number of distinct partners, including nine members of the homeobox family as a consequence of leukemia-associated chromosomal translocations. NUP98-HOXA9 is associated with the t(7;11)(p15;p15) translocation in acute myeloid leukemia (AML), myelodysplastic syndrome, and blastic crisis of chronic myeloid leukemia. Expression of NUP98-HOXA9 in murine bone marrow resulted in a myeloproliferative disease progressing to AML by 7-8 months. Transduction of NUP98 fusion genes into human CD34(+) cells confers a proliferative advantage in long-term cytokine-stimulated and stromal cocultures and in NOD-SCID engrafted mice, associated with a five- to eight-fold increase in hematopoietic stem cells. NUP98-HOXA9 expression inhibited erythroid and myeloid differentiation but enhanced serial progenitor replating. NUP98-HOXA9 upregulated a number of homeobox genes of the A and B cluster as well as MEIS1 and Pim-1, and downmodulated globin genes and C/EBPalpha. The HOXA9 component of the NUP98-HOXA9 fusion protein was protected from cullin-4A-mediated ubiquitination and subsequent proteasome-dependent degradation. In NUP98-HOX-transduced CD34(+) cells and cells from AML patients with t(7;11)(p15;p15) NUP98 was no longer associated with the nuclear pore complex but formed intranuclear aggregation bodies. Analysis of NUP98 allelic expression in AML and myelodysplastic syndrome showed loss of heterozygosity observed in 29% of the former and 8% of the latter. This was associated with poor prognosis.

Long-term bovine hematopoietic engraftment with clone-derived stem cells.

Therapeutic cloning by somatic cell nuclear transfer offers potential for treatment of a wide range of degenerative disease. Nuclear transplantation with neo (r)-marked somatic nuclei from 10-13-year-old cows was used to generate cloned bovine fetuses. Clone fetal liver (FL) hematopoietic stem cells (HSC) were transplanted into two busulfan-treated and one untreated nuclear donor cows. Hematopoiesis was monitored over 13-16 months by in vitro progenitor and HSC assays. Chimerism was demonstrated by PCR in blood, marrow, lymph nodes, and endothelium, peaking at levels of 9-17% in blood granulocytes but at lower levels in lymphocyte subsets (0.1-0.01%). Circulating progenitors showed high levels of chimerism (up to 60% neo (r+)) with persisting fetal features. At sacrifice, the animal that had no pre-transplant myelosupression showed persisting donor cells in blood and lymph nodes, and in marrow 0.25% of progenitor cells and a detectable fraction of stem cells were neo (r+). The fetal HSC showed a 10-fold competition advantage over adult HSC. Cloning generated histocompatible HSC capable of long-term multilineage engraftment in a large animal model.

Tumor necrosis factor alpha enhances the adenoviral transduction of CD34+ hematopoietic progenitor cells.

The purpose of this study was to improve the transduction efficiency of adenoviral vectors (Ad) in human CD34+ hematopoietic progenitor cells. CD34+ cells from cord blood or mobilized peripheral blood were incubated with tumor necrosis factor-alpha (TNF-alpha). After removal of free TNF-alpha, the cells were infected with an Ad encoding green fluorescent protein (GFP). One day later, viable cells were counted and analyzed for GFP and CD34 by flow cytometry. To visualize vectoral trafficking, CD34+ cells were incubated with fluorophore-conjugated Ad. Plating efficiencies of hematopoietic progenitors before and after transduction were evaluated by methylcellulose assays. Pretreatment with TNF-alpha increased the transduction efficiency more than twofold (39.2% versus 15.5%) in a dose-dependent manner and strongly improved the survival of GFP-positive CD34+ cells. Time course experiments showed that TNF-alpha incubation times as short as 10 minutes were still effective. Neutralizing antibodies to TNF receptor II and RGD peptides diminished the TNF-alpha-dependent increase in transduction efficiency. No TNF-alpha-dependent increase in adenoviral receptors (coxsackie-adenovirus receptor, alphavbeta3-integrin) occurred. Analysis of viral binding demonstrated a significantly higher incidence of local concentrations of Ad along the cell surface (caps) in virus-positive cells of the TNF-alpha-treated group. Plating efficiency, especially the formation of granulocyte-macrophage colony forming units, was enhanced by TNF-alpha pretreatment. We conclude that brief incubation with TNF-alpha before addition of the Ad significantly increased the Ad transduction efficiency in CD34+ cells, and improved post-transduction survival of progenitors of the granulocyte-macrophage lineage. This finding correlates with increased Ad capping at the cell surface and suggests an alteration of Ad trafficking.

Expression of constitutively active Notch4 (Int-3) modulates myeloid proliferation and differentiation and promotes expansion of hematopoietic progenitors.

The Notch family of transmembrane receptors has been implicated in the regulation of many developmental processes. In this study, we evaluated the role of Notch4 in immature hematopoietic progenitors by inducing, with retroviral transduction, enforced expression of Int-3, the oncogenic and constitutively active form of mouse Notch4. Int-3-transduced human myeloid leukemia (HL-60) cells demonstrated significantly delayed expression of differentiation markers following retinoic acid and 12-0-tetradecanoylphorbol 13-acetate treatment. Furthermore, HL-60 cells expressing Int-3 displayed a slower growth rate than cells infected with void virus, and accumulation in the G0/G1 phases of cell cycle. Transduction with deletion mutants of Int-3 defined the importance of individual domains of the protein (in particular, the ANK domain and the C-terminal domain) in the inhibition of differentiation and growth arrest of HL-60 cells. When mouse bone marrow enriched for stem cells (5-fluorouracil-resistant, lineage negative) was transduced and cultured for two weeks, the Int-3-transduced population displayed a lower expression of differentiation markers and a three- to five-fold higher frequency of colony-forming cells (CFU-GM/BFU-E) than control cultures. These results strongly support the notion that Notch signaling inhibits differentiation and promotes expansion of hematopoietic stem/progenitor cells.

Mobilization of endothelial and hematopoietic stem and progenitor cells by adenovector-mediated elevation of serum levels of SDF-1, VEGF, and angiopoietin-1.

The chemokine stroma-derived factor-1 (SDF-1) is produced within the bone marrow and mediates chemokinesis and chemotaxis on a variety of cell types that express the CXCR4 receptor. SDF-1-responsive cell types include monocytes and macrophages, B and T lymphocytes, platelets and megakaryocytes, and CD34+ cells, including both hematopoietic progenitors and stem cells. We have used intravenous injection of a replication-incompetent adenovector expressing the SDF-1 gene to elevate serum levels of SDF-1 in Balb/c and SCID mice. Within 3 to 5 days there was a marked leukocytosis, predominantly involving monocytes, and a three-fold increase in platelets. In addition, AdSDF-1 mobilized CFU-GM, CFU-s, and cells with long-term repopulating potential. We have identified a bone marrow-derived, circulating endothelial stem cell characterized by expression of the VEGFR2 (Flk-1/KDR). This cell exhibits a chemotactic and chemokinetic response to SDF-1 and VEGF. We have elevated serum levels of VEGF165 using intravenous adenovector gene delivery and compared this to an adenovector expressing angiopoietin-1 alone or in combination with VEGF. VEGF elevation was associated with rapid mobilization of hematopoietic stem and progenitor cells and a population of Flk-1-positive endothelial progenitors. In contrast angiopoietin induced a delayed mobilization of endothelial and hematopoietic progenitors. The combination of VEGF and angiopoietin produced a more prolonged elevation of these progenitors in the circulation with increased proliferation of capillaries and expansion of sinusoidal spaces in the marrow.

Plasma elevation of stromal cell-derived factor-1 induces mobilization of mature and immature hematopoietic progenitor and stem cells.

The chemokine, stromal cell-derived factor-1 (SDF1), is produced in the bone marrow and has been shown to modulate the homing of stem cells to this site by mediating chemokinesis and chemotaxis. Therefore, it was hypothesized that elevation of SDF1 level in the peripheral circulation would result in mobilization of primitive hematopoietic stem and progenitor cells. SDF1 plasma level was increased by intravenous injection of an adenoviral vector expressing SDF1alpha (AdSDF1) into severe combined immunodeficient mice. This resulted in a 10-fold increase in leukocyte count, a 3-fold increase in platelets, and mobilization of progenitors, including colony-forming units-granulocyte-macrophage to the peripheral circulation. In addition, AdSDF1 induced mobilization of cells with stem cell potential, including colony-forming units in spleen and long-term reconstituting cells. These data demonstrate that overexpression of SDF1 in the peripheral circulation results in the mobilization of hematopoietic cells with repopulating capacity, progenitor cells, and precursor cells. These studies lay the foundation for using SDF1 to induce mobilization of hematopoietic stem and progenitor cells in in vivo studies. (Blood. 2001;97:3354-3360)

Isolation and characterization of a disulfide-linked human stem cell factor dimer. Biochemical, biophysical, and biological comparison to the noncovalently held dimer.

Distinct from the noncovalently linked recombinant human stem call factor (rhSCF) dimer, we report here the isolation and identification of an SDS-nondissociable dimer produced during folding/oxidation of rhSCF. Experimental evidence using various cleavage strategies and analyses shows that the isolated dimer is composed of two rhSCF monomers covalently linked by four disulfide bonds. The cysteines are paired as in the noncovalently associated dimer except that all pairings are intermolecular rather than intramolecular. Other structural models, involving intertwining of intramolecular disulfide loops, are ruled out. The molecule behaves similarly to the noncovalently associated dimer during ion-exchange or gel permeation chromatography. However, the disulfide-linked dimer exhibits increased hydrophobicity in reverse-phase columns and in the native state does not undergo spontaneous dimer dissociation-association as seen for the noncovalent dimer. Spectroscopic analyses indicate that the disulfide-linked and noncovalently associated rhSCF dimers have grossly similar secondary and tertiary structures. In vitro, the disulfide-linked dimer exhibits approximately 3-fold higher biological activity in supporting growth of a hematopoietic cell line and stimulating hematopoietic cell colony formation from enriched human CD34+ cells. The molecule binds to the rhSCF receptor, Kit, with an efficiency only half that of the noncovalently associated dimer. Formation of intermolecular disulfides in the disulfide-linked dimer with retention of biological activity has implications for the three-dimensional structure of noncovalently held dimer and disulfide-linked dimer.

A pilot study of the biologic and therapeutic effects of granulocyte colony-stimulating factor (filgrastim) in patients with acute myelogenous leukemia.

In a carefully monitored pilot study, the in vivo biologic effects of filgrastim were investigated in eight patients with relapsed/refractory acute myelogenous leukemia. Within each patient, filgrastim was administered as a single agent prior to any chemotherapy in escalating doses of 0.12-6.0 micrograms/kg/day as a continuous intravenous infusion. The dose was increased every 14 days until an ANC of > or = 2500/mm3 had been achieved or there was evidence of proliferation of the leukemia. In patients who demonstrated growth of the leukemic clone, cytosine arabinoside was initiated at 200 mg/m2/day for 5 days. Through-out the course of therapy, the effects of filgrastim on maturation and proliferation were assessed by in vitro studies of bone marrow aspirates. Three patients demonstrated a sustained increase in ANC; one achieved a partial remission and remained on therapy for 31 weeks. Two of the three responding patients had hypocellular marrows at the time of initiating filgrastim and demonstrated a low but normal pattern of growth in CFU-GM assay early in the treatment course. This suggested that these two characteristics may define an environment in which filgrastim can induce a growth advantage for the normal residual hematopoietic elements. In this study of selected patients, filgrastim appeared safe.

Bacterial endotoxin regulation of cytokine receptors on murine bone marrow cells: in vivo and in vitro study.

Bacterial endotoxin (LPS) modulation of CSF-1, granulocyte-macrophage (GM)-CSF, G-CSF, IL-1, TNF, and Kit Ligand receptors on murine bone marrow cells (BMC) in vivo and in vitro was investigated. In vivo LPS reduced the binding of hrIL-1 (> 90%), mrTNF (> 62%), mrGM-CSF (> 42%) and hrG-CSF (> 91%) to BMC within 2 h, but elevated IL-1 binding (> 8.4-fold) between 8 to 48 h. In vitro, LPS decreased G-CSF and IL-1 binding after 8 h yet increased TNF binding in a dose-dependent manner, suggesting that in vivo, the LPS up-regulation of IL-1 binding to BMC was indirect. Because in vivo LPS elevated the levels of TNF, IL-6, GM-CSF, and glucocorticoids, we further examined GM-CSF and TNF modulation of cytokine receptors on BMC in vivo. In vivo, TNF decreased the binding of TNF (> 88%), G-CSF (> 89%), and IL-1 (> 73%) within 30 min, but increased IL-1 binding (> 4.8-fold) after 10 h. In contrast, in vitro TNF decreased IL-1 binding after 8 h, implying that in vivo TNF up-regulation of Il-1 binding to BMC was also due to an indirect mechanism. However, GM-CSF increased IL-1 binding to BMC both in vivo and in vitro after 8 h. Further studies showed that in vitro GM-CSF and dexamethasone synergistically increased IL-1 binding to BMC in a time- and dose-dependent manner. This synergistic modulation depended on synthesis of protein and mRNA, and was due to an increase in receptor number rather than an increase in receptor affinity. Because in vivo, LPS and LPS-induced cytokines (IL-1 and TNF) elicited the secretion of glucocorticoid and CSF activities, our results revealed a mechanism for LPS up-modulation of IL-1R on BMC in vivo.

Cytokines and dexamethasone modulation of IL-1 receptors on human neutrophils in vitro.

Modulation of IL-1R on human neutrophils was investigated after in vitro treatment of cells with human recombinant (hr) granulocyte (G)-CSF, hrgranulocyte-macrophage (GM)-CSF, hrCSF-1, hrIL-1 alpha, hrIL-2, hrIL-3, hrIL-4, hrIL-5, hrIL-6, hrIL-7, transforming growth factor-beta 1, or hrTNF-alpha. At 4 degrees C, 125I-IL-1 binding was competed by IL-1 but not by other cytokines tested. At 37 degrees C, GM-CSF, TNF-alpha, and IL-1 decreased 125I-IL-1 binding in a dose-dependent manner. Kinetic studies showed that GM-CSF reduced > 45% IL-1 binding within 15 min but later (8 h) produced a > 2-fold increase. In contrast, TNF decreased > 85% IL-1 binding within 15 min with a recovery of > 80% relative to that of control after 24 h. Scatchard analysis revealed that TNF or GM-CSF down-modulation of IL-1 binding was due to a decrease of IL-1R number. Further studies showed that dexamethasone and GM-CSF (or G-CSF) synergistically increased IL-1 binding after 8 h. This synergistic modulation was a cytokine dose- and time-dependent process, and was due to an increase in IL-1R numbers rather than a change in binding affinity. In addition, human bone marrow neutrophils, cord blood neutrophils, and several human hematopoietic cell lines (HL-60, U-937, and AML-193) responded to dexamethasone and GM-CSF (or G-CSF) with a superadditive increase in IL-1 binding. Because mammalian systems respond to bacterial endotoxins with secretion of TNF, IL-1, glucocorticoids, G-CSF and GM-CSF, our results shed additional light on this highly regulated cytokine network and revealed a novel role for GM-CSF.

Interleukin-1 modulation of cytokine receptors on human neutrophils: in vitro and in vivo studies.

Interleukin-1 (IL-1) modulation of cytokine receptors (human IL-1 receptor [hIL-1R], human granulocyte colony-stimulating factor [hG-CSFR], human granulocyte-macrophage CSF receptor [hGM-CSFR], and human tumor necrosis factor receptor [hTNFR]) on human neutrophils was studied both in vitro and in vivo. In vitro, incubation of neutrophils with IL-1 at 37 degrees C for 0.5 or 8 hours caused a reduction of IL-1 binding in a dose-dependent manner, but did not demonstrably affect binding of the other cytokines tested. In vivo, neutrophils from patients with gastrointestinal malignancies who were participating in a clinical trial of recombinant human IL-1 beta (rhIL-1 beta) demonstrated modulation of cytokine receptors in an IL-1 beta dose- and time-dependent manner. At the two highest dose levels of IL-1 beta (0.068 and 0.1 microgram/kg), reduction (> 40%) of G-CSF binding and elevation (twofold to sixfold) of IL-1 binding to neutrophils was observed after 1 hour and 4 to 8 hours, respectively. In addition, IL-1 beta rapidly elevated G-CSF and glucocorticoid levels in plasma. Patients at the lowest dose level (0.002 microgram/kg) had a less dramatic change in these parameters. Further in vitro studies showed that synthetic glucocorticoids and G-CSF synergistically up-modulated IL-1 binding to neutrophils in a dose- and time-dependent manner. Scatchard analysis of binding data showed that this in vitro synergistic modulation was due to an increase in receptor numbers, rather than an increase in binding affinity. In addition, both human umbilical cord blood and bone marrow neutrophils responded to G-CSF and dexamethasone (Dex) with a superadditive increase in IL-1 binding. Therefore, one of mechanisms for IL-1 up-modulation of IL-1R on human neutrophils in vivo was due to the fact that IL-1 rapidly elevates serum levels of G-CSF and glucocorticoids.

Granulocyte colony-stimulating factor modulation of cytokine receptors on murine bone marrow cells. In vivo and in vitro studies.

Human recombinant granulocyte CSF (G-CSF) modulation of cytokine receptors on murine bone marrow cells (BMC) in vivo and in vitro was investigated. In vivo, G-CSF reduced 125I-G-CSF binding to BMC by greater than 95% within 30 min, with return to base line after 48 h. Human rCSF-1 binding was reduced greater than 85% after 30 min and failed to recover even after 48 h. Murine rTNF-alpha or recombinant granulocyte/macrophage CSF binding was not significantly altered. However, human rIL-1 alpha binding increased greater than 1.5-fold after 3 h, was elevated greater than 5-fold between 6 and 12 h, and declined to base line after 48 h. In vitro, G-CSF induced a greater than 1.5-fold increase in IL-1 binding to BMC after 8 h, suggesting that up-modulation of IL-1 binding in vivo required G-CSF and other influences. Further studies indicated that BMC responded to glucocorticoids and G-CSF with a synergistic increase of IL-1 binding. This synergistic IL-1R modulation was a time- and dose-dependent process and was inhibited by cycloheximide or actinomycin D in a dose-dependent manner. Binding studies further revealed that the synergistic stimulation of IL-1R expression on BMC was probably due to increased receptor number, rather than increased receptor affinity. In addition, this phenomenon was also observed in other hematopoietic cells. Our results demonstrated that G-CSF was capable of stimulating IL-1R expression on BMC both in vivo and in vitro and G-CSF in combination with glucocorticoids synergistically up-modulated IL-1 binding to BMC in vitro. Inasmuch as IL-1 induces the secretion of G-CSF and glucocorticoids in vivo, this synergistic induction may play an important, as yet unknown, role in the inflammatory cascade.

IL-1 modulation of cytokine receptors on bone marrow cells. In vitro and in vivo studies.

We here investigated IL-1 modulation of cytokine receptors on murine bone marrow cells (BMC). In vivo, IL-1 treatment reduced greater than 88% of granulocyte-CSF, greater than 35% of TNF, and greater than 51% of granulocyte/macrophage-CSF binding to BMC after 3 h, and returned to base line after 48 h. However, IL-1 binding to BMC decreased greater than 30% after 30 min, dramatically increased greater than 9-fold between 6 and 10 h, and declined to base line after 48 h. In vitro incubation of BMC with IL-1 did not markedly alter IL-1 and granulocyte-CSF binding, suggesting that modulation of granulocyte-CSF and IL-1 binding to BMC by IL-1 in vivo is due to an indirect mechanism. Further in vitro studies showed that IL-1 binding to BMC was specifically induced by glucocorticoids rather than other steroids, and is a time- and dose-dependent process. IL-1 induced IL-1R up-regulation was suppressed by ketoconazole, cycloheximide, and actinomycin D in a dose-dependent manner. In addition, in vivo dexamethasone imitated the action of IL-1 in stimulating IL-1 binding to BMC and in inducing neutrophilia. Furthermore, IL-1 binding to BMC from sham mice 8 h after IL-1 administration was 2.5 times higher than that observed in adrenalectomized mice. Our results demonstrate, for the first time, that in vivo IL-1-induced increased IL-1 binding to BMC was due to an indirect mechanism, and glucocorticoids stimulated IL-1 binding to BMC in vivo and in vitro. Inasmuch as serum glucocorticoid levels can be elevated by IL-1 in vivo, these results reveal a novel mechanism by which IL-1 modulates its own receptors in vivo.

Modulation of granulocyte colony-stimulating factor receptors on murine peritoneal exudate macrophages by tumor necrosis factor-alpha.

Modulation of granulocyte CSF (G-CSF) receptors on murine peritoneal exudate macrophages (PEM) by various cytokines was investigated. At 4 degrees C, 125I-G-CSF receptor binding on PEM reached a plateau after 6 h and was specifically competed by unlabeled human rG-CSF but not by other cytokines, including human rG-CSF-1, murine recombinant granulocyte-macrophage CSF, murine rIFN-gamma, human rIL-1 beta, and murine rTNF-alpha. 125I-G-CSF bound to PEM has a half-life of 30 min at 37 degrees C. Preincubation of PEM with murine rTNF, murine recombinant granulocyte-macrophage CSF, CSF-1, or G-CSF for 30 min at 37 degrees C resulted in partial reduction of 125I-G-CSF binding capacity, whereas IL-1 or IFN-gamma did not inhibit G-CSF binding. Further studies indicated that reduction of G-CSF binding caused by TNF was a dose- and time-dependent process and did not require FCS. The reduction was transient, and receptor binding was recovered by incubation at 37 degrees C for 8 h. The recovery of G-CSF binding was inhibited in the presence of cycloheximide. In addition, G-CSF binding studies suggested that the TNF-induced decrease in G-CSF binding to PEM was probably due to a reduction in receptor number rather than receptor affinity. Modulation of G-CSFR by TNF was also observed on nonelicited macrophages from various strains of mice. Our results demonstrate a physiologic response of G-CSFR on macrophages that is modulated by TNF. This phenomenon may play an important, as yet unknown, role in the macrophage inflammatory response.

Modulation of colony-stimulating factor-1 receptors on macrophages by tumor necrosis factor.

The effect of murine rTNF-alpha on the binding of human 125I-rCSF-1 to murine thioglycolate-elicited peritoneal exudate macrophages (PEM) was investigated. At 4 degrees C, 125I-CSF-1 binding to PEM was inhibited by preincubation with human rCSF-1, but not by other cytokines. When PEM were incubated with various cytokines at 37 degrees C, murine rTNF-alpha caused greater than 90% decrease in 125I-CSF-1 binding. This decrease was time, temperature and TNF dose dependent, and was not affected by preincubation with cycloheximide. The reduction in CSF-1-binding activity was reversed by prolonged incubation at 37 degrees C even in the presence of TNF. However, PEM preincubated with TNF subsequently washing free of residual TNF resulted in a rapid recovery of CSF-1 binding. This recovery of CSF-1-binding activity required protein synthesis. Binding studies suggested that the decrease in 125I-CSF-1 binding was most likely caused by a reduction in the number of CSF-1 receptors. In addition, preincubation with TNF at 37 degrees C inhibited 125I-CSF-1 binding on mononuclear phagocytes, including the macrophage cell line J774, bone marrow-derived macrophages, and nonelicited macrophages from three different strains of mice. In contrast, 125I-murine rTNF-alpha binding to PEM was not inhibited by preincubation with CSF-1 at 4 degrees C or 37 degrees C. These data suggest that TNF may play a role in the modulation of receptor expression on blood cells, and may point to a role for this pleiotropic cytokine in the regulation of hemopoiesis.

Hematopoietic growth factor receptors.

The formation of the cellular constituents of the blood is regulated by a series of endogenous polypeptides with largely paracrine function. A number of these hematopoietic growth factors (HGF's), which include colony stimulating factors, interleukins, and erythropoietin, have been purified to homogeneity and cloned, which in turn has led to extensive investigations of their biochemical properties and biological effects and functions. The HGF's act on target cells by binding to receptors. The kinetics and, to an even larger extent, dynamics of the factor/receptor associations display several intriguing characteristics, most of which are still poorly understood. Herein, the biochemical characteristics of HGF's receptors as well as the binding properties, post-receptor binding events and receptor modulation resulting from the association of HGF's and their target cells are reviewed.

Purification and characterization of human colony-stimulating factor 1 from human pancreatic carcinoma (MIA PaCa-2) cells.

Colony-stimulating factor 1 (CSF-1) was purified from the serum-free conditioned medium of a human pancreatic carcinoma cell line (MIA PaCa-2) by a combination of conventional chromatography and high-performance liquid chromatography. The purity of human CSF-1 was demonstrated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) with a diffuse single band of Mr 42,000-50,000 and by N-terminal amino acid analysis of glutamate residue. The CSF-1 was stable at 50 degrees C for 30 min. It is sensitive to treatment with trypsin, chymotrypsin, and subtilisin but less sensitive to papain digestion. Treatment of CSF-1 with different glycosidases did not affect the biological activity. Sulfhydryl reagents such as dithiothreitol (DTT), iodoacetic acid, and N-ethylmaleimide did not affect the biological activity at the concentration of 1 mM. However, CSF-1 activity was inhibited totally by the combination of 10 mM DTT and 1 mM SDS. Under denaturing and reducing conditions, CSF-1 appeared on SDS-PAGE as a single protein band of Mr 21,000-25,000 and concurrently lost its activity, indicating that human CSF-1 possibly consists of two similar subunits and that the intact quaternary structure is essential for the biological activity. When treated with neuraminidase and endo-beta-D-N-acetylglucosaminidase D, the molecular weight of CSF-1 was reduced to 36,000-40,000, and to 18,000-20,000 in the presence of mercaptoethanol. Because of the specificity of endo-beta-D-N-acetylglucosaminidase D, it is suggested that the carbohydrate moieties are Asn-linked "complex-type" units.