An erythropoietic stimulating factor similar to erythropoietin released by macrophages after treatment with silica

Summary An erythropoietic stimulating factor (ESF) can be detected in the supernatant from fetal liver and adult bone marrow and spleen cells when preincubated with the macrophage-specific cytotoxic agent, silica. Stimulation is observed in 12-day fetal liver CFU-E cultures in the absence of added erythropoietin (Ep). The concentration of ESF in the supernatant added to CFU-E cultures is dependent on the preincubated cell dose and the volume added. The stimulating activity is abolished when mice are hypertransfused and increased above normal values when mice are bled. A concentrated silicatreated spleen supernatant was able to stimulate erythropoiesis in the polycythemic mouse bioassay. It is concluded that the ESF is similar, if not identical, to Ep.This work was supported by the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 112/Project A2) and the Volkswagen Foundation.     

Erythropoietin, thrombopoietin and colony stimulating factor in fetal mouse liver culture media.

Fetal mouse livers, days 13 to 19 of gestation, were cultured for 21 days and the culture media tested for erythropoietin (Ep), thrombopoietin (TSF) and colony stimulating factor (CSF). High levels of both Ep and CSF were released into the culture media. However, no detectable TSF was found. Maximum Ep culture activity was detected in day 14 and day 15 fetal liver cultures while maximum CSF was found in the day 16 fetal liver culture. These studies indicate that fetal liver cells in culture are capable of producing and/or releasing both Ep and CSF but not TSF.     

Erythropoietin production in a primary culture of human renal carcinoma cells maintained in nude mice

The present studies report erythropoietin (Ep) production in primary cultures of a human renal carcinoma from a patient with erythrocytosis that has been serially transplanted to BALB/c nude mice. The levels of erythropoietin in the culture media were estimated using the exhypoxic polycythemic mouse assay (EHPCMA), fetal mouse liver erythroid colony- forming technique (FMLC), and a radioimmunoassay (RIA). The spent culture media of the exponentially growing cells contained less than 10 mU/ml of Ep measured by RIA. However, after the cells became confluent, Ep levels (RIA) in the spent media showed a marked increase to approximately 300 mU/ml. Ep levels estimated using the FMLC and EHPCMA were approximately 2/3 and 1/10, respectively, of those measured by RIA. Rabbit antiserum to highly purified human urinary Ep (70,400 U/mg protein) was utilized for immunocytochemical (peroxidase-antiperoxidase method) localization of Ep in the cultured cells. Very few of the cells in exponential growth exhibited Ep-like immunoreactivity, whereas intense Ep-like immunoreactivity was observed in the cytoplasm of the cells maintained in culture for a prolonged period after reaching confluency. The most intense staining was observed in some of the cells forming domes. The domes developed after the cells reached confluency, and their numbers increased with increasing time in confluent culture, in parallel with the increase in Ep levels in the spent media. This primary cell culture system of a renal cell carcinoma maintained in nude mice, which produces immunologically and biologically active Ep, may provide a useful model for studies of the mechanism of Ep production.     

The ultrastructural immunocytochemical demonstration of erythropoietin receptors on developing erythrocytic cells of fetal mouse liver

In spite of the severe restrictions imposed by the lack of high purity erythropoietin (Ep) and Ep specific antiserum wer have attempted to demonstrate presumptive Ep binding sites on the surface of developmental forms of erythrocytic cells of the mouse fetal liver using ultrastructural immunocytochemical methodology. Cells were exposed to exogenous Ep of differing potencies and concentrations and subsequently incubated in an Ep antiserum. The Ep-Ep-antiserum complex was then visualized with a gold-labeled IgG reagent. Limited surface labeling was noted on the developing erythrocytic cells and the evaluation of the mean surface labeling densities of the erythrocytic cell series indicated that the extent of labeling was related to the stage of maturation. Early polychromatophilic erythroblasts proved to be the most heavily labeled of the erythrocytic cells followed by the basophilic erythroblast, proerythroblast and late polychromatophilic erythroblasts respectively; normoblasts, reticulocytes and erythrocytes exhibited, at most, only minimal labeling. In addition, we have been able to demonstrate several primitive cells in the fetal liver which exhibited extensive surface labeling when compared with the developing erythrocytic cells. These cells may represent the erythroid precursors of the fetal liver.     

Developmental regulation of erythropoiesis by hematopoietic growth factors: analysis on populations of BFU-E from bone marrow, peripheral blood, and fetal liver

Fetal hematopoiesis is characterized by expanding erythropoiesis to support a continuously increasing RBC mass. To explore the basis for this anabolic, nonhomeostatic erythropoiesis, the proliferative effect of recombinant hematopoietic growth factors on highly enriched hematopoietic progenitor cells from fetal and adult tissues were compared. Fetal hepatic BFU-E, unlike adult bone marrow (BM) or peripheral blood (PB) BFU-E, were capable of proliferating in response to erythropoietin in the absence of added GM colony-stimulating factor (GM-CSF) or interleukin-3 (IL-3), and erythropoietin (Epo) directly stimulated the expansion of the fetal BFU-E pool in suspension culture. A murine monoclonal antibody (MoAb), Ep 3, was raised against enriched fetal liver progenitor cells, which detected all fetal BFU-E and which reacted with the erythropoietin-responsive, GM-CSF/IL-3-independent fraction of adult BM BFU-E and CFU-E. All adult PB BFU-E were Ep 3- but became Ep 3+ after stimulation with GM-CSF or IL-3. These data indicate that Epo plays a unique role in fetal hepatic erythropoiesis, stimulating proliferation of immature BFU-E in addition to promoting terminal differentiation of later erythroid progenitor cells. In addition, these results demonstrate a MoAb which detects all erythropoietin-responsive progenitor cells and distinguishes the BFU-E compartments in adult BM and PB.     

Investigations into the in Vitro Sensitivity to Erythropoietin of Haematopoietic Tissue from Ex-hypoxic, Erythrocytotic Mice

Summary. If 'sensitivity' is defined as the ratio of haem synthesis in erythropoietin (Ep)-stimulated cultures to that in cultures without added Ep, then haematopoietic tissue of ex-hypoxic mice is less sensitive to Ep in vitro than similar tissue from normal donors. This appears to be due primarily to the loss of erythropoietin responsive cells from the marrow and spleen. In addition, saline extracts of mature erythrocytes inhibited erythropoiesis in normal bone marrow cultures in a manner directly related to the 'haematocrits' used during their preparation. These data support recent suggestions that the preference of erythrocytotic mice over normal animals for Ep assay probably can be attributed to their low baseline level of erythropoiesis rather than an increased sensitivity to Ep.     

Stem cell factor amplifies newborn and sickle erythropoiesis in liquid cultures

A two-phase liquid-culture system was used to substantially amplify and differentiate erythroblasts, starting with mononuclear cells from the blood of normal adults, newborn infants, and patients with sickle cell anemia. After the first 7 days (phase 1), in medium plus fetal bovine serum (FBS) alone, or in combination with stem cell factor (SCF) or conditioned medium (CM), the cell number was unchanged, and the cells all looked like lymphocytes. These cells were then diluted into medium with erythropoietin (Ep) alone, with Ep and either SCF or CM, or in methylcellulose with the same factors (phase 2). After 14 days in liquid phase 2 with SCF and Ep, the cell numbers increased an average of 30-fold in the sickle, 24-fold in the newborn, and 4-fold in the normal adult cultures; almost all the cells were erythroblasts and erythrocytes. SCF in phase 1 increased the number of late progenitors (CFU-E) assayed in methylcellulose, with the largest number in sickle, followed by newborn cultures and then adult cultures. We conclude that erythroid progenitor cells survive for at least 7 days without Ep (but with FBS). Progenitor cells are amplified, particularly with SCF. Later in culture, SCF with Ep increases the final number of differentiated erythroid cells. Both the early and the late effects of SCF are most effective in sickle, followed by newborn cultures and then adult cultures.     

Extraction of erythropoietin from isolated renal glomeruli of hypoxic rats

We investigated whether erythropoietin (Ep) can be extracted from renal glomeruli of hypoxic rats. Hypoxia was induced in a hypobaric chamber at 0.42 atmospheres for usually 6 h. Glomeruli were isolated with a sieving technique from kidneys that were flushed free of blood. Ep activities in glomeruli homogenates were determined in the fetal mouse liver cell assay and in the hypoxia exposed mouse assay. We found in vitro erythropoietic activity of glomerular extracts, which increased about 25-fold during hypoxia. Incubation of the extracts with anti-Ep serum abolished the activity. Its binding characteristics on Wheat germ lectin. Ricinus communis lectin and DEAE-cellulose suggested that glomeruli contained Ep precursor(s) that lacked terminal sugar residues. Such an incomplete carbohydrate portion seems also compatible with the finding that glomerular extracts did not stimulate erythropoiesis in hypoxia exposed mice.     

Erythropoietin changes the globin program of an interleukin 3-dependent multipotential cell line.

B6SUtA is a factor-dependent murine cell line of adult origin displaying the functional properties of a multipotent hematopoietic stem cell. We analyzed the globin programs of B6SUtA cells undergoing erythroid differentiation in both suspension and clonal cultures. In the absence of added erythropoietin, a small number of hemoglobinized cells were present, and these expressed predominantly embryonic globin. Addition of erythropoietin increased the number and maturation of hemoglobinized cells and led to a preferential augmentation of adult globin. Analysis of individual B6SUtA erythroid bursts showed that embryonic and adult globin can be expressed in cells derived from a single progenitor. Furthermore, by studying globin expression in cultured cells from mouse embryos, we found that the globin programs of B6SUtA cells are similar to those of erythroid progenitors at the period of transition from yolk sac to fetal liver erythropoiesis. Since B6SUtA cells are derived from adult bone marrow and they have the capacity to express embryonic globin, we speculate that the globin locus is not irreversibly modified during development and that adult cells at early stages of erythroid differentiation can transiently express ontogenetically primitive globin programs.     

Measurement of human serum erythropoietin by erythroid colony-forming technique using fetal mouse liver cells

A quantitative bioassay for serum erythropoietin in anemic patients was established with erythroid colony-forming technique using methyl cellulose. Fetal mouse liver contains many erythropoietin-dependent erythroid colony-forming cells (CFU-E) with the concentration as well as sensitivity to erythropoietin being the highest at 13–14 days of gestation. Linearity in the erythropoietin response curve for the number of colonies from CFU-E in fetal livers of 13–14 days gestation was obtained for erythropoietin concentrations in the culture medium of 3.9 to 125 mU/ml. When newborn liver cells were used instead, the linearity occurred at 31.2 to 250 mU/ml. Measurement of erythropoietin by the erythroid colony-forming technique for sera from patients with various anemias was compared with that of the assay method using radioiron incorporation into heme in the suspension culture of fetal mouse liver cells. There was a good correlation between the values of both assay methods (r = 0.856, p < 0.001). Also, levels of erythropoietin measured by this assay method showed the expected inverse relationship to the hemoglobin concentration in the sera from the patients.     

Specificity of lactoferrin as an inhibitor of granulocyte-macrophage colony-stimulating activity production from fetal mouse liver cells.

Fetal mouse liver cultures capable of producing both erythropoietin (Ep) and granulocyte-macrophage colony stimulating activity (GM-CSA) were used to study the specificity of lactoferrin as an inhibitor of the production of GM-CSA. Both a granulocyte-derived colony-inhibiting activity (CIA) and lactoferrin inhibited GM-CSA production while having no effect on Ep production. These results demonstrate the specificity of lactoferrin for GM-CSA production.     

Translation of messenger RNA from a renal tumor into a product with the biological properties of erythropoietin

The renal tumor RCC-3-JCK, when transplanted into immunodeficient mice, caused an erythrocytic polycythemia. When grown in culture, the tumor cells secreted a substance into the culture medium that chromatographed by size-exclusion high-performance liquid chromatography similarly to purified human erythropoietin (Ep) and was positive when assayed for Ep by its ability to stimulate erythropoiesis in fetal mouse liver cells (the FMLC assay). The poly(A) + RNA was extracted from the tumor cells and injected into Xenopus oocytes, inducing the appearance of Ep(FMLC) in the oocyte culture medium. Both the tumor cells and oocyte culture media were fractionated by size-exclusion high-performance liquid chromatography, and two fractions with Ep(FMLC) activity were found in the tumor-cell culture medium. Three active fractions were found in the medium from the mRNA-injected oocytes. The largest component from both culture media had the same elution time as a human standard (Ep). The poly(A) + RNA was fractionated by sucrose density-gradient centrifugation and the 8S and 10S fractions were found to induce Ep(FMLC) synthesis when they were injected into the oocytes. We conclude that poly(A) + RNA isolated from the Ep-producing tumor RCC-3-JCK included mRNA for Ep and that the Ep was a translational product of Xenopus oocytes injected with this mRNA.     

Serum erythropoietin measurements using the fetal mouse liver cell culture: the importance of reduction of variation in the specific activity of radioiron--transferrin

Because of varying iron--transferrin concentrations in different serum samples, and varying test serum portions within the culture, variations of the radioiron--transferrin/total iron--transferrin ratio are inevitable, when serum erythropoietin (Ep) concentrations are measured using the fetal mouse liver cell assay. It could be demonstrated that radioiron uptake is directly proportional to the specific activity of radioiron--transferrin, when the latter varies over the range which is inherent to the method. Variations of the ratio of radioiron--transferrin/total iron--transferrin were reduced by preincubating radioiron with human transferrin, and by minimizing the test serum portion of the culture. With this modified in vitro bioassay, serum Ep concentrations of 59 healthy subjects were measured. Mean serum Ep concentration was 136 +/- 66 (s.d.) mU/ml.     

A quantitative bioassay for erythropoietin using mouse fetal liver cells.

The conditions under which mouse fetal liver cells in vitro are most sensitive to erythropoietin have been investigated with the object of establishing a rapid and sensitive bioassay for this hormone. Fetal liver age and incubation times are shown to be relatively unimportant. Of the commercially available media studied, Eagles Minimal Essential Medium plus 5 percent Fetal Calf Serum is superior to other media used to study erythropoiesis in mouse fetal liver cells. The number of cells per 1 ml culture is important since some evidence was obtained for cell cooperation occurring above 10-6 cells per culture. To minimize this possibility 5 x 10-5 cells are used. If (59Fe) ferric citrate is used to assess heme synthesis, no significant advantage is observed by prior binding of the isotope to mouse serum transferrin. Butan-2-one is preferred to cyclohexanone for extracting heme. Using the conditions described in this report the mouse fetal liver cell assay for erythropoietin is capable of detecting erythropoietin concentrations as low as 0.001 unit culture. Preliminary results using normal human sera show that it is sufficiently sensitive to detect erythropoietin over a range of serum concentrations. This allows a full dose-response relationship to the sera to be determined and, by a comparison with an erythropoietin standard, detailed quantitative results can be obtained. Therefore the technique seems to fulfill the need for a rapid and sensitive erythropoietin bioassay for routine clinical use.     

Testosterone stimulation of a rapidly labeled, low-molecular-weight RNA fraction in human hepatic erythroid cells in culture.

The addition of erythropoietin to cell cultures of erythroid cells of human fetal liver resulted in an increased incorporation of thymidine, adenine, and uridine into trichloroacetic acid-insoluble cell fractions and in an increased uptake of adenine and uridine into the cell. Although the effects of testosterone and erythropoietin on heme synthesis in these cells are known to be very similar, there was no effect of testosterone on the total incorporation of radioactive precursors into DNA or RNA. The RNA synthesized after short pulses of radioactive uridine, when analyzed on sucrose gradients containing 1% sodium dodecyl sulfate, consisted of a homogeneous peak sedimenting at 10 plus or minus 2 S, which is quite different from the heterogeneous, high-molecular-weight RNA synthesized under identical conditions in primary cultures of human fetal lung, kidney, or liver parenchymal cells. Addition of testosterone to liver erythroid cells in cultures for 5 hr followed by a 1-hr uridine pulse resulted in a 3-fold increase of RNA species with an average sedimentation coefficient of 14 plus or minus 3 S. The similarity with the sedimentation coefficient of the globin mRNA described in other systems and the high degree of specialization of the erythroid cells suggest that this RNA may be a stable intermediate involved in the synthesis of hemoglobin.     

Effects of dexamethasone on erythroid colony and burst formation from human fetal liver and adult marrow

S ummary . Dexamethasone, a prototypic synthetic glucocorticoid, was added to cultures of human fetal liver and adult marrow cells to assess its effects on erythroid colony and burst formation. Dexamethasone decreased the number of fetal liver erythroid colonies and bursts formed in the presence of erythropoietin, and also decreased the number of cells per colony. The amount and type of haemoglobin produced per cell were unaffected by adding dexamethasone to the cultures. Dexamethasone inhibited the incorporation of ³H-thymidine into DNA in fetal liver cells stimulated with erythropoietin, supporting the hypothesis that dexamethasone inhibits the proliferation but not the differentiation of fetal liver CFU-E and BFU-E. In contrast, addition of dexamethasone to adult bone marrow cultures had a variable effect on erythroid colony and burst formation.     

Erythropoietin production by the fetal liver in an adult environment

To gain insight into the mammalian liver to kidney erythropoietin (Ep) switch, we heterotopically transplanted livers from preswitch, switched, and postswitch fetal and newborn lambs into normal adult sheep. Recipients' serum Ep and circulating reticulocyte levels were serially determined until rejection of the graft and compared with identical samples from sham-operated control adult ewes. Transplantation of preswitch and switched fetal livers caused an impressive rise in recipients' serum Ep activity and provoked a corresponding increase in reticulocytosis. In contrast, Ep activity and reticulocyte counts did not change from preoperative levels in adult ewes transplanted with postswitch livers or in the sham-operated controls. The production of Ep by the preswitch fetal liver in the adult environment was not dependent on the presence or absence of host kidneys and was stimulated by anemic hypoxia. These results suggest that the fetal liver is capable of producing relatively large amounts of Ep activity, and the production of Ep can be maintained in the adult environment in the presence of functional adult kidneys. This argues against suppression of liver Ep production by renal Ep, or some other factor in the postnatal environment, and suggests that the liver to kidney switch of Ep production during ontogeny may represent a genetically determined event.     

Evidence for direct action of human biosynthetic (recombinant) GM-CSF on erythroid progenitors in serum-free culture

The biologic activity of human biosynthetic granulocyte-monocyte colony stimulating factor (GM-CSF) was investigated in serum-free culture of erythroid progenitors derived from adult peripheral blood. The morphology of erythroid bursts and the cloning efficiency of BFU-E under serum-free conditions were similar to those observed in dishes with fetal bovine serum (FBS). For these experiments, progenitor cells were partially purified by Ficoll-Paque density centrifugation, adherence to a plastic surface, and complement-mediated cytotoxicity of Leu-1+ elements. For some studies, blastlike cells were harvested directly from 6-day-old semisolid cultures. In serum-free culture of the light-density cell fraction, biosynthetic erythropoietin (Ep) was sufficient for formation of pure and mixed erythroid colonies whereas GM-CSF was required for granulocyte-monocytic colonies. When adherent and Leu-1+ cells were removed, or when in vitro differentiated blast cells were used as a source of progenitors, neither Ep or GM-CSF alone induced colony formation. In dishes supplemented with both growth factors, erythroid bursts were detected. Although the presence of GM- CSF alone did not induce formation of any colony or clusters, BFU-E were recorded when Ep was added 8 days later, suggesting that BFU-E could be maintained. Terminal maturation of the resulting erythroid bursts was delayed by 8 days. These results provide evidence that GM- CSF acts directly on early erythroid progenitors. Furthermore, they suggest that both Ep and GM-CSF are necessary to start the differentiation process.     

In vitro culture of erythroid colonies from human fetal liver and umbilical cord blood.

Colony formation by erythroid precursors from human fetal liver, umbilical cord blood and adult peripheral blood has been studied in a plasma clot culture system. Fetal liver (FL) was obtained at post-mortem examination from 13-22 week abortuses. After mincing in Hanks' solution, cells in suspension were harvested by Ficoll-Hypaque centrifugation. Mononuclear cells were obtained by centrifugation of umbilical cord blood (CB) and normal adult peripheral blood (PB). All three types of preparations were incubated up to 14 d in 0.1 ml plasma clot cultures containing 0-4 u/ml erythropoietin (Epo) and 10(6) cells/ml. No colonies formed in the absence of Epo. Normal adult PB produced late-appearing colonies; there were no colonies at day 7 and up to 100 colonies/0.1 ml at day 14. CB produced early and late colonies with up to 200 colonies/0.1 ml at day 7 and 125 at day 14. Cells from FL produced many early colonies; over 1500 colonies/0.1 ml were sent at day 7 and there was a subsequent decline in colony count with longer incubation. In cultures of both CB and FL, colonies composed of either mature or immature cells were noted during both early and late stages of incubation suggesting that these cell sources contain a heterogeneous population of erythroid colony progenitors. Measurement of differential beta and gamma globin chain synthesis by erythroid colonies grown from fetal liver and umbilical cord blood gave results similar to those obtained by direct pulse-labelling of the original source of the cultured cells.     

In vitro growth of human fetal CD34+ cells in the presence of various combinations of recombinant cytokines under serum-free culture conditions

Summary. CD34⁺ cells were purified from midtrimester human fetal blood and adult bone marrow samples and seeded in serum-free fibrin-clot cultures in order to evaluate the number and the responsiveness to recombinant cytokines of pluripotent (CFU-GEMM), erythroid (BFU-E), megakaryocyte (BFU-meg and CFU-meg) and granulocyte/macrophage (CFU-GM) haemopoietic progenitor cells.
The number of the different haemopoietic progenitors/1 × 10³ CD34⁺ cells, except CFU-meg, was significantly higher in fetal blood than in adult bone marrow in cultures stimulated by any combination of cytokines including interleukin-3 (IL-3), granulocyte/macrophage colony stimulating factor (GM-CSF) or stem cell factor (SCF) plus erythropoietin (Epo). Nevertheless, whereas adult BFU-E showed a maximal growth in the presence of Epo plus IL-3 or Epo plus SCF, fetal BFU-E showed an optimal growth in the presence of Epo alone, the sensitivity of fetal BFU-E to suboptimal concentrations of Epo being approximately 10–15-fold higher than that of adult BFU-E. Addition of optimal concentrations of IL-3, GM-CSF or SCF, alone or in various combinations, to Epocontaining cultures induced a significant increase in both the number and size of fetal CFU-GEMM, and CFU-GM, and a parallel decrease of fetal BFU-E. Finally, SCF potently syner-gized with IL-3 in increasing the growth of both classes of fetal megakaryocyte progenitors, BFU-meg and CFU-meg.