The synergistic effect of thrombopoietin in erythropoiesis with erythropoietin and/or IL-3 and myelopoiesis with G-CSF or IL-3 from umbilical cord blood cells of full-term neonates.

The authors sought to determine whether recombinant human thrombopoietin (TPO) acts synergistically with other cytokines on burst-forming unit-erythroid (BFU-E)-derived and colony-forming unit-granulocyte/macrophage (CFU-GM)-derived colony formations from cord blood. Cord blood nonadherent mononuclear cells (MNC) from normal full-term neonates were cultured in a methylcellulose system. When cultured with 5 x 10(4) MNC/mL, erythropoietin (EPO) 2 U/mL, interleukin-3 (IL-3) 50 ng/mL, and/or TPO 400 ng/mL (experiment 1), the addition of TPO to EPO gave rise to more BFU-E-derived colonies (p = .002). The addition of TPO to EPO + IL-3 gave rise to more BFU-E-derived colonies (p = .006) also. TPO synergizes erythropoiesis from cord blood. When cultured with IL-3 50 ng/mL, granulocyte colony-stimulating factor (G-CSF) 25 ng/mL, and/or TPO 400 ng/mL, the addition of TPO to IL-3 gave rise to more CFU-GM-derived colonies (p = .002). The addition of TPO to G-CSF gave rise to more CFU-GM-derived colonies (p = .002) also. TPO synergizes myelopoiesis from cord blood. Thus, TPO has synergistic effects on both erythropoiesis and myelopoiesis from cord blood. In the identical conditions of culture, cord blood had significantly greater BFU-E-derived or CFU-GM-derived colony formation than bone marrow (in a previous report by the authors) did. When cultured under conditions similar to those of experiment 1, but with 1 x 10(4) cord blood MNC/mL and TPO 100 ng/mL (experiment 2), results similar to those in the experiment 1 also revealed that TPO has synergistic effects on erythropoiesis and myelopoiesis from cord blood. In every individual assay, the numbers of BFU-E-derived or CFU-GM-derived colonies in experiment 1 were significantly higher than those in experiment 2.

Role of the microenvironment of the embryonic aorta-gonad-mesonephros region in hematopoiesis.

Although various cytokines, growth factors, and chemokines are known to regulate hematopoiesis, expansion of hematopoietic stem cells (HSCs) in vitro with the use of such agents has proved problematic. Stromal cells are major components of the microenvironment that surrounds hematopoietic cells and are thought to play an important role in hematopoiesis in vivo. Co-culture of HSCs with stromal cells promotes hematopoiesis and self-renewal of HSCs. Definitive hematopoietic cells first appear during mammalian embryonic development in the aorta-gonad-mesonephros (AGM) region, and it is therefore thought that the microenvironment of this region plays an important role in HSC ontogeny. We have adopted two approaches to studying the contribution of the AGM microenvironment to hematopoiesis. In the first approach, we have developed an in vitro culture system for mouse AGM explants. Hematopoiesis is enhanced in such cultures by the presence of the combination of stem cell factor (SCF), basic fibroblast growth factor, leukemia inhibitory factor, and oncostatin M (SFLO culture). However, transplantation assays revealed that HSCs capable of long-term reconstitution of the hematopoietic compartment of irradiated mice (LTR-HSCs) do not expand in AGM-SFLO cultures; rather, these cultures appear to provide a favorable microenvironment for hematogenic angioblasts that are precursors of both endothelial and hematopoietic cells. In our second approach, we have established various stromal cell lines from the mouse AGM region. The AGM-S3 cell line supports human and mouse primitive hematopoietic cells as well as mouse LTR-HSCs. Maintenance of LTR-HSCs is mediated by a mechanism other than SCF signaling through its receptor (c-Kit). These two in vitro approaches should prove useful for further elucidation of the mechanisms that underlie hematopoiesis and HSC self-renewal.

Involvement of protein kinase C-epsilon in signal transduction of thrombopoietin in enhancement of interleukin-3-dependent proliferation of primitive hematopoietic progenitors.

We studied the effect of thrombopoietin (TPO) on interleukin-3 (IL-3)-dependent bone marrow cell colony formation of mice to clarify the role of protein kinase C (PKC) in the signal transduction of TPO for the proliferation of primitive hematopoietic progenitors. TPO alone hardly yielded colonies. However, TPO in combination with IL-3 increased colony numbers synergistically from 2- to 4-fold, compared with those supported by IL-3 alone. Serial observation of colony development showed that TPO may hasten the appearance of colonies by shortening the dormant period (G(0)) of primitive progenitors. Immunocytochemical studies on PKC isoforms in progenitor cells stimulated with TPO have revealed that the expression pattern of PKC-epsilon is changed, but not that of PKC-alpha, -beta, -gamma, -delta, or -zeta. Selective PKC inhibitors, such as calphostin C and GF 109203X, and PKC-epsilon-specific translocation inhibitor peptide abrogated the enhancing effect of TPO on IL-3-dependent colony formation and the changes in the intracellular expression pattern of PKC-epsilon. These data taken together suggest that TPO has a direct effect on primitive progenitors and enhances IL-3-dependent colony formation, at least partly through the activation of PKC-epsilon.

The synergistic effect of thrombopoietin in erythropoiesis and myelopoiesis from human bone marrow cells.

We sought to determine whether recombinant human thrombopoietin (TPO) acts synergistically with recombinant human erythropoietin (EPO) and/or recombinant human interleukin-3 (IL-3) on erythroid burst formation and granulocyte-macrophage colony formation from human bone marrow (BM). BM cells were from 5 adults and 15 children who underwent bone marrow examination because of a clinical suspicion of malignancy; their bone marrows as well as the complete blood counts were normal and were cultured in a methylcellulose system. TPO has a synergistic effect with EPO or EPO + IL-3 on erythropoiesis of human BM, as the addition of TPO to EPO significantly gave rise to more erythroid bursts (p = 0.0001) and the addition of TPO to EPO + IL-3 might give rise to more erythroid bursts (p = 0.05). TPO also has a synergistic effect with recombinant human granulocyte colony-stimulating factor (G-CSF) on myelopoiesis of human BM, since the addition of TPO to G-CSF gave rise to significantly more granulocyte-macrophage colonies (p = 0. 0001). Besides its well-known significant role in megakaryopoiesis, TPO also has synergistic effects on erythropoiesis and myelopoiesis.

Antitumor activity of alpha-galactosylceramide, KRN7000, in mice with the melanoma B16 hepatic metastasis and immunohistological study of tumor infiltrating cells.

Liver metastasis of primary tumors is clinically a major problem. We examined the antitumor activity of KRN7000, an alpha-galactosylceramide, in mice with liver metastasis of the B16 melanoma. KRN7000 significantly inhibited tumor growth in the liver, and its potency was similar to that of interleukin-12. The KRN7000 administration resulted in a high percentage of cured mice, which acquired tumor-specific immunity. To study what kinds of antitumor effector cells participated in killing tumor cells, we then performed immunohistological analysis of tumor-infiltrating cells, and found that KRN7000 induced marked invasion of NK1.1+ cells, CD8+ cells, and F4/80+ cells (macrophages) into B16 tumor nodules. In addition, it appeared that KRN7000-treated, liver-associated macrophages possessed strong lytic activity against tumor cells. These results suggest that NK cells, NK1.1+ T (NKT) cells, cytotoxic T lymphocytes, and macrophages play an important role in killing tumor cells in the liver, and that KRN7000 may be useful for the treatment of cancer liver metastasis.

Effects of thrombopoietin on megakaryocyte colony formation from leukemic cells at diagnosis and from marrow cells after induction chemotherapy for acute leukemias.

We have studied the effects of recombinant human thrombopoietin (TPO, mpl ligand) on the megakaryocyte colony formation from control human bone marrow cells, human leukemia cells at diagnosis, and human bone marrow cells after induction chemotherapy for acute leukemias. In the control human bone marrow cells from four adults and nine children who had localized malignancy and histologically normal-looking marrow. TPO alone effectively stimulated megakaryocyte colony formation, and interleukin-3 (IL-3) synergized this. In 17 patients (13 adults and four children) with acute myeloid leukemia (AML) at diagnosis, TPO stimulated leukemic colony formation in only one patient with FAB M7 subtype. In 11 children with acute lymphoblastic leukemia (ALL) at diagnosis, TPO did not enhance leukemic colony formation. After 17 courses of induction chemotherapy, nine for AML and eight for ALL, TPO stimulated megakaryocyte colony formation to a level of 51%, of that in the control human bone marrow cells. This may suggest that the administration of TPO to patients with M7 subtype warrants caution, whereas it is probably safe to give TPO at any time to patients with ALL. The administration of TPO to patients with acute leukemias after induction chemotherapy could stimulate megakaryocytopoiesis.

Treatment of hepatic metastasis of the colon26 adenocarcinoma with an alpha-galactosylceramide, KRN7000.

Colorectal liver metastasis is clinically a major problem. We examined the antitumor activity of KRN7000, an alpha-galactosylceramide, on mice with liver metastases of adenocarcinoma Colon26 cells. KRN7000 treatment, beginning 1 day after tumor inoculation (day 1), significantly inhibited tumor growth in the liver, and its potency was equal to that of interleukin 12. KRN7000 treatment from day 3 caused regression of established Colon26 nodules. KRN7000 administration resulted in a high percentage of cured mice that acquired tumor-specific immunity. In addition, it appeared that highly activated, liver-associated natural killer cells made the major contribution to the killing of Colon26 cells in the liver. These results suggest that KRN7000 may be useful for the treatment of colorectal liver metastasis.

Serum thrombopoietin levels in patients receiving high-dose chemotherapy with support of purified peripheral blood CD34+ cells.

In a case control study, serum levels of thrombopoietin (TPO) were determined by a sandwich ELISA in 20 patients (median age, 7 years; range, 2-56 years) with various malignancies who received high-dose chemotherapy and a stem cell rescue operation. The patients received two different transplant modalities: (a) 12 patients received purified autologous peripheral blood CD34+ cells; and (b) 8 patients received cells in the CD34(-) fraction, which still contains many CD34+ cells. No significant differences were observed between the two groups with regard to the duration required to achieve an absolute granulocyte count of >0.5 x 10(9)/liter, the duration of dependence on platelet transfusion, or the number of platelet transfusions. In both groups, the serum TPO levels were inversely correlated with the circulating platelet count. Multivariate analysis demonstrated that significant determinants of the serum TPO level included the circulating platelet count (standardized regression coefficient = -0.5179), transplantation with cells in the CD34(-) fraction (0.2414), solid tumor (0.1420), and the age of the patient (-0.1236; r2 = 0.3021; P < 0.0001). These results suggest that the mode of stem cell support (ie., the presence of accessory cells in the inoculum), age, or the type of preceding chemotherapy affects serum TPO levels after transplantation.

Evaluation of a cytokine combination including thrombopoietin for improved transduction of a retroviral gene into G-CSF-mobilized CD34+ human blood cells.

We examined cell culture conditions with various combinations of cytokines including thrombopoietin (TPO) to obtain the most efficient transduction of recombinant retrovirus vectors into G-CSF-mobilized blood CD34+ cells which were obtained from children and purified with an Isolex 50 system (Baxter; Deerfield, IL). Three different 4-day culture conditions for the stimulation of CD34+ cells were compared in terms of a cell-cycle analysis by fluorometry and gene transduction efficiency as determined by resistance to G418 and NeoR polymerase chain reaction (PCR) for individual colony-forming unit-granulocyte/macrophage (CFU-GM) grown in a methylcellulose culture system. The cytokines tested were: A) interleukin (IL)-6 + stem cell factor (SCF); B) IL-3 + IL-6 + SCF, and C) IL-3 + IL-6 + SCF + TPO. Without a cell culture, the percentage of CD34+ cells in the cell cycle (the percentage of cells in phases S and G2/M) was 4.6%. After a four-day culture (n = 5), this value increased with the addition of IL-3 (22%) or IL-3 + TPO (27%, p < 0.05) as compared to that with the baseline cocktail of IL-6 + SCF (15%). The cell number uniformly increased approximately 10-fold in each culture condition. The average efficiency of gene transfer into incubated CD34+ cells with the corresponding combinations of cytokines was, respectively, 57%, 47%, and 30% for G418-screened CFU-GM and 72%, 68%, and 51% for polymerase chain reaction-positive CFU-GM. A statistically significant difference (p < 0.01) was found for G418/CFU-GM with IL-3 + IL-6 + SCF (57%) versus IL-3 + IL-6 + SCF + TPO (30%). Hence, it is likely that the increased cell proliferation produced by the addition of TPO was not necessarily translated into an increased rate of retroviral-mediated gene transduction, possibly because TPO preferentially induced the differentiation of stem cells into mature progenitors in these culture systems.

Thrombopoietin primes human platelet aggregation induced by shear stress and by multiple agonists.

Recombinant thrombopoietin has been reported to stimulate megakaryocytopoiesis and thrombopoiesis and it may be quite useful to treat patients with low platelet counts after chemotherapy. As little is known regarding the possible activation of platelets by thrombopoietin, we examined the effects of thrombopoietin on platelet aggregation induced by shear stress and various agonists in native plasma. Using hirudin as an anticoagulant, thrombopoietin (1 to 100 ng/mL) enhanced platelet aggregation induced by 2 micromol/L adenosine-diphosphate (ADP) in a dose dependent fashion. The enhancement was not affected by treatment of platelets with 1 mmol/L aspirin plus SQ-29548 (a thromboxane antagonist, 1 micromol/L) but was inhibited by a soluble form of the thrombopoietin receptor, suggesting that the enhancement was mediated by the specific receptors and does not require thromboxane production. Epinephrine (1 micromol/L), which does not induce platelet aggregation in hirudin platelet rich plasma (PRP), did so in the presence of thrombopoietin (10 ng/mL). Thrombopoietin (10 ng/mL) also enhanced or primed platelet aggregation induced by collagen (0.5 micron.mL),. thrombin, serotonin, and vasopressin. Thrombopoietin does not induce any rise in cytosolic ionized calcium concentration nor activation of protein kinase C, as estimated by phosphorylation of preckstrin, indicating that the priming effects of thrombopoietin does not require those processes. The ADP- or thrombin-induced rise in cytosolic ionized calcium concentration was not enhanced by thrombopoietin (100 ng/mL). Further, shear (ca. 90 dyn/cm2)-induced platelet aggregation was also potentiated by thrombopoietin. The priming effect on epinephrine-induced platelet aggregation in hirudin PRP was unique to thrombopoietin, with no effects seen using interleukin-6 (IL-6), IL-11, IL-3, erythropoietin, granulocyte-colony stimulating factor, granulocyte macrophage-colony stimulating factor, or c-kit ligand. These data indicate that monitoring of platelet functions may be necessary in the clinical trials of thrombopoietin.

Establishment and characterization of a new human mesothelioma cell line (T-85) from malignant peritoneal mesothelioma with remarkable thrombocytosis.

A mesothelioma cell line, termed T-85, was established from a patient with malignant peritoneal mesothelioma and remarkable thrombocytosis (1.4 x 10(6)/mm3). Electron microscopically, two types of mesothelioma cells have been characterized; the major type of cells with dense-cored granules in the cytoplasm and the minor one with evenly dense granules. Immunologically, the cells showed staining for interleukin-6 (IL-6), cytokeratin, collagen type IV, vimentin, laminin, fibronectin and Factor VIII-related antigen. Quantitation by ELISA revealed a high concentration of IL-6 in T-85 cell culture supernatants. RT-polymerase chain reaction of T-85 cells showed two positive bands of cDNA at 628 and 251 base pairs indicating the constitutive expression of IL-6 and IL-6 receptor mRNA. Moreover, prominent pro-platelet process formation activity in T-85 cell culture supernatants indicated the presence of a thrombopoietic activity due mainly to IL-6 but not the c-Mpl ligand or erythropoietin. However, the fact that 15% of PPF activity remained in the supernatants treated with anti-IL-6 antibody indicated the presence of another thrombopoietic substance. T-85 is so far the first mesothelioma cell line derived from a case with remarkable thrombocytosis.

Thrombopoietin (TPO) induces tyrosine phosphorylation and activation of STAT5 and STAT3.

The growth and differentiation of megakaryocytes are regulated by thrombopoietin (TPO), a recently characterized cytokine which exerts its effects via a member of the hematopoietin receptor superfamily, c-Mpl. Since many cytokines which bind hematopoietin receptors activate the STAT family of transcription factors, we investigated whether STAT proteins were activated by TPO. TPO induced the formation of a DNA-binding complex recognizing a known STAT-binding sequence. STAT5 was a major component of this DNA-binding complex, and STAT5 was tyrosine phosphorylated in response to TPO. Additionally, TPO-induced the tyrosine phosphorylation and DNA-binding activity of STAT3. Together with the recent demonstration of JAK2 activation in response to TPO, the data presented here define a rapid signaling pathway likely to be important in TPO-induced gene regulation.

Thrombopoietin induces tyrosine phosphorylation and activation of the Janus kinase, JAK2.

Thrombopoietin (TPO) is a recently characterized growth and differentiation factor for megakaryocytes and platelets that exerts its effects via the receptor, c-MpI. This receptor is a member of the hematopoietin receptor superfamily and is essential for megakaryocyte maturation; however, the molecular mechanisms of TPO and c-MpI action have not been elucidated. Recently, the Janus kinases have emerged as important elements in signaling via this family of receptors. In this report, we show that, in the M07e megakaryocytic cell line, which expresses c-MpI and proliferates in response to TPO, TPO induces phosphorylation of a number of substrates between 80 and 140 kD. Specifically, we show that stimulation with TPO induces the rapid tyrosine phosphorylation of a 130-kD protein that we identify as the Janus kinase, JAK2. However, no detectable tyrosine phosphorylation of JAK1, JAK3, or TYK2 was observed. TPO also induced activation of JAK2 phosphotransferase activity in vitro. Taken together, these data indicate that JAK2 likely plays a key role in TPO-mediated signal transduction.

Exogenous expression of human granulocyte colony-stimulating factor receptor in a B-lineage acute lymphoblastic leukemia cell line: a possible model for mixed lineage leukemia.

We report evidence that the granulocyte colony-stimulating factor (G-CSF) receptor is present and may be functioning on blast cells from some patients with myeloid surface antigen positive (My+) acute lymphoblastic leukemia (ALL). In the present study, a human G-CSF receptor expression plasmid was transfected into a newly established B-lineage ALL cell line 'Tanoue' and its subclone 'ST' by lipofection to investigate whether expression of the G-CSF receptor and G-CSF stimulation would induce myeloid characteristics on myeloid surface antigen negative (My-) ALL cells. The G-CSF receptor became detectable on the transfected cells (GR-Tanoue and GR-ST), with dissociation constant values of 50-130 pmol/l, and maximal binding sites (Bmax of 77-6100 sites/cell on receptor binding assays. Short term culture with recombinant human G-CSF induced myeloid differentiation (a two to three-fold increase in CD33 and CD15 expression), and a moderate 3H-thymidine uptake (stimulation index, 1.75) only in the GR-ST clone no. 15 which expressed a high number of G-CSF receptors (Bmax, 6100 sites/cell). Our data show that (a) exogenous expression of the G-CSF receptor and G-CSF stimulation can induce myeloid characteristics on ALL cells; and (b) in the G-CSF receptor-expressing cells, there is a correlation between the number of G-CSF receptors and cell responsiveness to G-CSF in either proliferation or differentiation.

Analysis of myeloid characteristics in acute lymphoblastic leukemia.

We examined myeloid characteristics of myeloid-antigen-positive (My+) and -negative (My-) B-precursor acute lymphoblastic leukemia (ALL) blast cells. Immunophenotyping before and after culture, rearrangements of immunoglobulin heavy chain (IgH) and T-cell receptor (TCR) genes, stimulation of DNA synthesis with granulocyte colony-stimulating factor (G-CSF), and G-CSF binding assay were performed. Of My+ ALL blasts, the immunophenotypic staging as B-precursor ALL and rearrangements of IgH and TCR-beta, gamma and delta genes did not differ from findings in My- ALL blasts. Stimulated with G-CSF, cells from one My+ ALL and from one My- ALL patients showed enhancement of DNA synthesis and expression of CD11b and CD13, respectively. G-CSF binding was observed in blasts from 3 My+ ALL patients and one My- ALL child. After culture, blasts from My- ALL children expressed CD13 but showed neither enhanced DNA synthesis with G-CSF nor G-CSF binding. Thus, it would appear that (i) My+ and My- adult ALL blasts are at the same stage of differentiation; (ii) some My+ adult ALL blasts have phenotypic and functional myeloid characteristics; and (iii) induction of CD13 expression in My- ALL after in vitro culture does not correlate with other myeloid characteristics.

Anemia induced in splenectomized mice by administration of rhG-CSF.

Normal and splenectomized mice (SPLXM) were given rhG-CSF for 10 to 128 days and serial observations were made on blood counts for 128 days. After 10 days, mice were killed for histologic studies. All treatment schedules produced, in addition to elevated white blood counts, a macrocytic anemia which only partially responded to large doses of Epo. Stopping rhG-CSF treatment for 2 days resulted in the return of granulocytes, lymphocytes, monocytes, platelets and polychromatophilic erythrocytes to near normal levels, indicating a need for the continued presence of rhG-CSF to maintain peripheral blood increases. Treatment of normal and SPLXM with rhG-CSF induced marked granulocytic hyperplasia of the bone marrow with expansion of the granulocytic marrow into the adjoining muscle as in acute myelocytic leukemia. The hyperplasia is greater in the SPLXM than in the normal mouse where splenic hyperplasia occurs in all cell lines. The rhG-CSF also results in expansion of granulopoiesis into the normally fatty tail bone marrow in SPLXM. The rhG-CSF treatment produced marked increases in the assayable numbers of GM-CFU, G-CFU and M-CFU. The significance and mechanisms of induction of these changes are not clear. It is speculated that treatment with rhG-CSF has multicellular effects, suggesting that it initiates a cascade of molecular reactions that cause the effects observed.

Granulocyte colony-stimulating factor (G-CSF) receptors on acute myeloblastic leukaemia cells and their relationship with the proliferative response to G-CSF in clonogenic assay.

The number and the affinity of granulocyte colony-stimulating factor (G-CSF) receptors expressed by blast cells in acute myeloblastic leukaemia (AML) were determined using radiolabelled recombinant human G-CSF (rhG-CSF). Eighteen of 20 patients demonstrated specific binding, and Scatchard analysis revealed a single class of high affinity (Kd 15-130 pM) G-CSF receptors on the AML blasts. The number of G-CSF receptors varied from 55 to 1,200 per cell (mean 278). In the remaining two patients, specific binding was not observed. The number of G-CSF receptors did not differ significantly between various AML subtypes, but the mean receptor number was the highest on type M2 blasts. A chemical cross-linking study revealed that the G-CSF receptor has an approximate molecular weight of 140,000. Autoradiography showed heterogeneity of the distribution of G-CSF receptors on the AML blasts obtained from a single patient. The number of colonies stimulated by the addition of rhG-CSF varied from 0 to 566 per dish, and blast colony formation was observed in eight of 20 patients. The population mean of G-CSF receptor number expressed by blasts that formed colonies on stimulation with rhG-CSF was significantly higher than that on blasts which did not form colonies. These results suggest that a proliferative response of AML blasts to G-CSF may be predicted when the blasts express a large number of G-CSF receptors. Accordingly, it may be safer to restrict the clinical use of G-CSF to AML patients who have blasts with a low G-CSF receptor expression and no response to G-CSF in blast colony assay.

Is recombinant human granulocyte colony-stimulating factor (G-CSF) orally available in rats?

Oral availability of recombinant human granulocyte colony-stimulating factor (G-CSF) was investigated in rats by measuring the blood total leucocyte (BTL) counts. Oral test G-CSF solution was prepared with 10% HCO-60 (polyoxyethylated, 60 mumol, castor oil derivative), 1% DK ester (sugar ester) or 10% MYS-40 (polyethyleneglycol monostearate), in which the G-CSF concentration was 500 or 250 micrograms/ml. Each test solution was injected into the duodenum of three rats at the G-CSF dose level of 300 or 600 micrograms/kg, and BTL counts were monitored for 48 h. All of the test G-CSF solution raised the BTL levels within 24 h after injection. In particular, the HCO-60 solution increased the BTL levels over 2 times as compared to the predose level at 600 micrograms/kg dose and the effect was apparently dose-dependent. A short-term study suggested that the effect of G-CSF on the BTL level appeared at the fastest at about 5 h after administration of HCO-60 test solution, 300 micrograms/kg. In view of the pattern of BTL dynamics obtained after i.v. injection of HCO-60 solution at 25 and 50 micrograms/kg, the increase of BTL levels observed after oral administration of the HCO-60 solution is considered to be due to the orally supplied G-CSF.