Pure and mixed erythroid colony formation in vitro stimulated by spleen conditioned medium with no detectable erythropoietin.

Cells from CBA fetal mouse liver formed pure or mixed erythroid colonies in semisolid agarculture after stimulation by medium conditioned by pokeweed mitogen-stimulated mouse spleen cells. In general shape, the erythroid colonies resembled typical 7-day single or multiple (burst) colonies. However one-third to one-half contained, in addition to erythroid cells, macrophages and neutrophils and, less commonly, megakaryocytes or eosinophils. Culture of micro manipulated single colony-forming cells showed these erythroid colonies to be clones. Colony-forming cells declined in frequency with advancing fetal age, but low numbers were detectable in adult bone marrow. Assays of spleen conditioned medium in polycythemic mice failed to detect erythropoietin; the cloning system may detect a fetal type of erythropoietin-independent, erythropoietic cell since few were detected in adult marrow.     

Sex hormones and the regulation of erythroid spleen colonies development of fetal liver origin

The development of erythroid colonies of fetal liver hemopoietic cell origin in adult irradiated polycythemic mice was studied. It was found that orchidectomy sharply reduced the number of erythroid colonies developed in the spleen of these polycythemic male recipients. Estrogen injection to the orchidectomized polycythemic recipient did not further decrease the number of erythroid colonies developed. It is concluded that the development of erythroid colonies of fetal liver origin in polycythemic male recipients is maintained mainly by testicular testosterone. The complete suppression of these colonies in female recipients does not seem to be a result of inhibition by estrogen.     

Elevation of erythroid colony-stimulating activity in the serum of mice with graft-versus-host disease

Serum obtained from (C3H X C57BL) F1 hybrid mice injected with parental (C57BL) lymph node cells after 600 rad irradiation contained a high level of an erythropoietic factor, which enhanced erythroid colony formation by fetal C57BL liver cells in methylcellulose cultures. The serum level of this erythroid colony-stimulating activity increased as the dose of lymph node cells injected increased and as suppression of endogenous spleen colonies (indicative of GVH severity) became more marked. Erythropoietin was also elevated in the sera of these mice. When these animals were transfused with washed packed red cells to induce plethora after sublethal irradiation and injection of lymph node cells, the erythropoietin level decreased, but erythroid colony-stimulating activity remained elevated. On the other hand, in anemic mice treated with phenylhydrazine, this erythropoietic factor was not detected, although the serum level of erythropoietin was elevated. These results suggest that a graft-versus-host (GVH) reaction may induce the production of an erythropoietic factor other than erythropoietin and that anemia itself may not be a stimulus for its in vivo production.     

Stimulation of erythropoiesis in a grafted animal by mouse fetal liver culture media.

Erythropoietin given to polycythemic, irradiated mice receiving a bone marrow transplant restores the number of erythroid colonies produced. In this study mouse fetal liver culture media administered to polycythemic, irradiated mice also restored the number of erythroid colonies and erythropoiesis as measured by 59Fe incorporation. Thus, mouse fetal liver in culture produces a factor which resembles erythropoietin.     

Production of erythropoietin-like activity by a murine erythroleukemia cell line.

A transplantable murine leukemia, primarily induced by a biologically cloned Friend helper virus, was shown to induce polycythemia in recipient ICFW mice. A leukemia cell line (IW.32) was established in vitro from this transplantable leukemia. Sodium butyrate and hemin induced erythroid differentiation in these leukemia cells as has already been shown with other erythroleukemia cells. The supernatant of this cell line was devoid of spleen focus-forming virus activity. However, it induced the incorporation of 59Fe in polycythemic mice and the in vitro differentiation of murine and human cfu-e into erythroid colonies. Therefore, these erythroleukemia cells produced a factor with all the biological properties of erythropoietin. The erythropoietic activity of IW.32 supernatant was higher in vitro [equivalent to 0.5-1 international unit (IU) of erythropoietin per ml] than in vivo (0.15-0.3 IU/ml). This erythropoietin-like activity was stable at 100 degrees C for 3 min, which ruled out the possibility that a virus was responsible for these effects. Preliminary studies demonstrated that the biochemical properties of the IW.32 factor are strongly similar to those of Connaught step 3 erythropoietin, thus supporting the hypothesis that the IW.32 factor is indeed an erythropoietin.     

Identification of distinctive stromal elements in erythroid and neutrophil granuloid spleen colonies: light and electron microscopic study.

Erythroid and granuloid spleen colonies from 1100 R irradiated mice transfused with 5 X 10(4) syngeneic bone marrow cells, fixed in paraformaldehyde-glutaraldehyde followed by osmium tetroxide, were processed for light and electron microscopy. Light microscopic examination of day 7 erythroid colonies revealed distinctive reticular cells indigenous to the erythroid colony; they were not observed outside the confines of the erythroid colony or in granuloid colonies or within the spleens of control irradiated mice that had not been transfused with bone marrow cells. Examinations of sections stained with a newly developed differential stain for thick sections (1 mu) revealed unique staining properties for these reticular cells. Ultrastructural features and the intimate contact of the proerythroblasts and the reticular cells are described. Microscopic examination of day 10 neutrophil granuloid spleen colonies harvested from polycythemic mice showed distinctive stromal cells, different from those of erythroid colonies. These distinctive cell to cell relationships constitute a possible morphological basis for the function of the erythroid and granuloid hemopoietic inductive microenvironments.     

In Vitro Effect of Erythropoietin on the Spleen of The Polycythemic Mouse: II. Radiosensitivity of Stem Cells

1. An in vitro method to observe radiosensitivity of stem cells was developedin the present study. In vivo and in vitro effect of ⁶⁰Co irradiation on the erythropoietin-induced stem cell differentiation into erythroblasts was observed,using a tissue culture method of polycythemic mouse spleen. Response to erythropoietin was demonstrated by an appearance of heme synthesis and erythroblasts in spleen fragments.

2. A significant correlation between the rate of appearance of erythroblastsand heme synthesis of the spleen fragments was observed.

3. After irradiation, marked impairment of both heme synthesis and production of erythroblasts was observed, yielding D₃₇ values in the vicinity of70 r in vivo and 120 r in vitro irradiation, respectively.

4. Marked recovery of erythropoietin-induced heme synthesis in the polycythemic mouse spleen was observed 9 days after 300 r irradiation, with an"overshooting" phenomenon on the 12th day.

Submitted on September 14, 1966 Accepted on December 12, 1966     

Erythroid colony induction without erythropoietin by Friend leukemia virus in vitro.

Erythroid colonies could be produced without the addition of erythropeietin in plasma cultures seeded with bone marrow cells from normal C3Hf/Bi mice by exposure of the cells in vitro to medium from a cell line (IS) that continuously produces Friend leukemia virus in culture. The activity in the culture medium was viral rather than erythropoietin-like, since it was sedimentable by high-speed centrifugation and heat labile. Erythroid colonies did not develop when the bone marrow cells exposed to virus-containing medium were from mice genetically resistant to Friend virus. IS culture medium contained both Friend spleen focus-forming and XC-plaque-forming activities. No erythroid colonies were induced when genetically sensitive cells were exposed to a preparation from which the spleen focus-forming activity had been removed, but which contained XC plaque-forming activity in high concentration. Thus the spleen focus-forming component of Friend virus appeared to be responsible for inducing erythroid colony formation without erythropoietin in vitro. Some erythroid colonies were also found in control cultures to which neither virus nor erythropoietin had been added. Reduction in the concentration of fetal calf serum in the culture medium substantially decreased the number of these colonies but had only a minor effect on the number of virus-induced colonies. The number of erythroid colonies produced after 2 days of culture without erythropoietin or fetal calf serum was approximately proportional to the titer of Friend spleen focus-forming virus to whcih the bone marrow cells had been exposed. This system should prove useful for investigation in vitro of Friend virus--host cell interactions which lead to erythropoietin-independent erythropoiesis.     

Endogenous erythroid colony formation by peripheral blood mononuclear cells from patients with myelofibrosis and polycythemia vera.

Peripheral blood mononuclear cells from patients with polycythemia vera or myelofibrosis with myeloid metaplasia were studied for their erythroid colony growth characteristics in plasma clot cultures. In both diseases, erythroid colonies formed early in culture in the absence of added erythropoietin (endogenous colonies). In no instance did early, endogenous colony formation occur with peripheral blood cells from normals or patients with secondary polycythemia. A normal response to erythropoietin was observed with both control and patients' peripheral blood cells. Spleen mononuclear cells obtained from one patient with myelofibrosis also produced endogenous colonies and showed a response to erythropoietin. This study suggests that culture of peripheral blood mononuclear cells might serve as a useful tool in discriminating polycythemia vera from secondary polycythemia.     

The Cellular Basis for the Defect in Haemopoiesis in Flexed-Tailed Mice

Three assays for erythropoietic progenitor cells have been applied to mice of genotype f/f and to nearly congenic +/+ controls. When f/f mice were tested for their ability to generate transient endogenous erythroid spleen colonies 4-6 days after 800 rads and 10 units of erythropoietin, the numbers of such colonies detected were greatly reduced, although normal numbers of spleen colonies appeared at later times (9-12 days) postirradiation. In contrast, cells capable of erythropoietic colony formation in culture (CFU-E) were present within the normal range in both f/f spleen and marrow and their sensitivity to erythropoietin in culture was the same as that found previously for CFU-E in the marrow and spleen of +/+ mice. Transfusion-induced plethora reduced the number of CFU-E in marrow to a similar extent in both f/f and +/+ mice; likewise, subsequent administration of 10 units of erythropoietin induced a rapid return in the number of marrow CFU-E in both genotypes. In the spleen, CFU-E numbers were approximately three-fold lower in f/f mice in each group. These results support the view that the 5 day assay for transient endogenous spleen colonies detects cells (TE-CFU) that are different from both CFU-E and pluripotent stem cells (CFU-S), although possibly overlapping to some extent with the immediate progenitors of CFU-E. The results also indicate that the generation or maturation of TE-CFU represents a primary site of expression of the f/f defect.     

Effect of Chronic Ethanol Administration on Production of and Response to Erythropoietin in the Mouse

Male adult mice were allowed to drink only a solution of 32% ethyl alcohol for 3 months. Hematocrit and hemoglobin concentration were lower in ethanol-treated than in control mice at the end of the experimental period. Red cell volume was not significantly different between both groups. Plasma volume was higher in experimental than in control mice. Therefore, the anemia found in ethanol-treated mice can be regarded as a dilution anemia. When ethanol-treated and control mice, both made polycythemic by hypertransfusion to suppress their endogenous erythropoietin formation, were injected with doses of erythropoietin in the range of 0.2 to 3.2 IRP units, the derived dose-response curves were markedly different because of a reduced response to the hormone by the treated mice. This finding suggests that the number of "erythropoietin-responsive cells" may be reduced as the result of ethanol, or that their response to the hormone may be delayed or inhibited. Plasma erythropoietin concentration in alcohol- treated mice, as determined in the posthypoxic polycythemic mouse bioassay, was higher than normal in both normoxic and hypoxic conditions, probably as the result of the impaired responsiveness to the hormone mentioned above.     

In vitro assay for erythropoietin: erythroid colony formation in methyl cellulose used for the measurement of erythropoietin in plasma.

Erythroid colony formation in methyl cellulose has been used for the measurement of erythropoietin in plasma. Livers from newborn mice less than 24 hr old were found to provide convenient target cells. Newborn mouse liver contains a substantial number of erythroid colony-forming cells (CFU-e) that have a high sensitivity to erythropoietin, the dose--response curve for erythropoietin reaching a plateau at 50 mU/ml. As little as 0.5 m/ml of the hormone is detectable. Removal of cells that adhered to glass prior to culturing doubled the number of colonies formed in the presence of erythropoietin. Addition of untreated plasmas that showed high erythropoietin titers in the exhypoxic polycythemic mice assay gave variable results. Some of the plasmas stimulated colony formation actively and in a linear fashion. However, the majority of the plasmas were toxic to the cultures. Dialyzing the plasmas for 3 days against distilled water effectively removed the toxicity. Results obtained with the method are in good agreement with the values found using the exhypoxic polycythemic mice assay.     

Erythropoiesis in mice with chemically-induced tumor

The erythropoietic response of Strain A mice with benzo(a)pyrene-induced fibrosarcoma has been studied. The rate of erythropoiesis, expressed in terms of 59Fe incorporation into circulating erythrocytes, was slightly increased in the fibrosarcomatous mice compared to the matched controls. The femoral marrow of the tumor hosts became hypocellular with reduced number of erythroblasts. The spleen, on the other hand, was hypercellular with an increased number of erythroid progenitor cells. Quantitative assessment of erythropoietic activity in the hematopoietic organs by measuring the 6-hr organ uptake of 59Fe revealed erythropoietic suppression in the marrow but enhancement in the spleen following tumor development. While the number of CFU-s in the femoral marrow of the tumor-bearing animals was slightly increased, marked increase in the concentration as well as absolute number of CFU-s was found in the spleen. Bioassay for erythropoietin revealed appreciable increase in the level of plasma erythropoietin in the tumor-bearing animals.     

Induced expression of c-fms in normal hematopoietic cells shows evidence for both conservation and lineage restriction of signal transduction in response to macrophage colony-stimulating factor

Retrovirus-mediated gene transfer was used to obtain expression of the macrophage colony-stimulating factor (MCSF) receptor, c-fms, on hematopoietic lineages that normally do not express this receptor. Cultures of murine fetal liver cells infected with the c-fms retrovirus developed erythroid colonies in cultures stimulated with M-CSF. However, these colonies were fewer and less hemoglobinized than colonies in parallel cultures stimulated by erythropoietin. Culture of isolated clones demonstrated a direct action of M-CSF on erythroid clones. Culture of c-fms retrovirus-infected adult murine bone marrow cells showed megakaryocyte and novel macrophage-megakaryocyte clones when stimulated by M-CSF. Culture of isolated clones again confirmed a direct action of M-CSF on megakaryocyte clones. In contrast, M-CSF stimulation of c-fms-infected granulocytes and granulocyte progenitor cells did not elicit proliferation, enhanced survival, or functional stimulation of granulocytes. These findings provide evidence for both conservation and lineage restriction of signal transduction in normal hematopoietic cells.     

Early fluctuations of BFU-E pool size after transfusion or erythropoietin treatment.

The kinetics of erythroid burst-(BFU-E) and colony-forming units (CFU-E) have been evaluated in marrow and spleen of normal or polycythemic mice, respectively, after transfusion or administration or purified erythropoietin (Ep). Ep injection induces an early but temporary rise of BFU-E number, versus a later but more prolonged expansion of the CFU-E pool. Symmetric patterns are observed after transfusion, i.e., an early but transient depletion of the BFU-E population, versus a later but more persistent decrease of the CFU-E number. It is suggested that the size of the BFU-E compartment is Ep-dependent in the "early" phase after erythroid perturbation. Later on, however, compensatory mechanisms allow the BFU-E pool to "escape" from the early Ep influence, thus allowing its size to return to and stabilize in the near normal range. It is further suggested that the BFU-E may represent an early target cell of Ep stimulus.     

Polycythemia in chronic obstructive pulmonary disease. A study of serum and urine erythropoietin and medullary erythroid progenitors

In order to investigate the mechanism of polycythemia in chronic obstructive pulmonary disease (COPD), serum and urinary levels of erythropoietin and medullary erythroid progenitors were studied in 21 patients; nine were nonpolycythemic (hematocrit, 39 +/- 4 percent; red blood cell [RBC] mass, 28 +/- 5 ml/kg; forced expiratory volume in one second [FEV1], 0.6 +/- 0.1 L), and 12 patients were polycythemic (hematocrit, 52 +/- 7 percent; RBC mass, 46 +/- 7 ml/kg; FEV1, 0.9 +/- 0.3 L). Hypoxia was severe in both groups, with mean arterial oxygen pressure of 47 mm Hg. The following parameters of tissue oxygenation were not significantly different between the two groups: arterial and mixed-venous oxygen saturations; cardiac output; oxygen utilization coefficient; 2, 3-diphosphoglycerate, and carboxyhemoglobin level. The level of erythropoietin was measured by bioassay in vitro. The level was increased in the serum of 85 percent (18) and in the urine of 38 percent (8) of the patients. There was no significant difference between the nonpolycythemic and polycythemic groups. Without exogenous erythropoietin, none of the subjects showed spontaneous colonies of erythroid progenitors. The addition of one unit of erythropoietin induced a similar normal proliferation of erythroid progenitors in both groups. The absence of adaptative polycythemia in the nonpolycythemic group with severe hypoxia was seemingly related neither to a quantitative deficit of erythropoietin nor to a lack of sensitivity of erythroid progenitors to its action.     

Studies on Erythropoiesis. XVII. Some Quantitative Aspects of the Erythropoietic Response to Erythropoietin

The transfusion-induced polycythemic mouse is an excellent subject in whichto study the physiology of erythropoiesis because endogenous erythropoiesisis reduced in a manner that does not appear to impair, to any recognizabledegree, the animal’s ability to respond to an erythropoietic stimulus.

Using a purified glycoprotein, erythropoietin, obtained from the plasma ofanemic sheep, a number of experiments was carried out on polycythemic mice.Morphological evidence of erythropoiesis in the spleen was observed followinga single subcutaneous injection of erythropoietin. A cumulative response tomultiple injections was noted. For a given total dose of erythropoietin, agreater response was produced when this amount was fractioned than whenit was administered as a single dose. Animals made polycythemic by transfusion were maintained in a polycythemic state by daily injections of erythropoietin. Finally, a true erythrocytosis was produced by a prolonged course ofinjections of erythropoietin in normal mice.

Submitted on January 3, 1961 Accepted on February 22, 1961     

Regeneration of erythroid progenitor cells in polycythemic mice treated with cyclophosphamide

Mice with posthypoxic polycythemia treated with a sublethal dose of cyclophosphamide (Cy) were used as a model to investigate, by in vitro methods, the kinetics of regeneration of erythroid committed precursors (ECP) and to study the influence of erythropoietin (Ep) on those precursor cells. The results demonstrated that erythroid burst-forming units (BFU-E), early (d10) and late (d4), and erythroid colony-forming units (CFU-E) recover at different rates after Cy. Early BFU-E recovery was not Ep dependent and closely resembled regeneration of pre-erythropoietin-responsive cells (pre-ERC) found previously using the same experimental model. The absence of spontaneous recovery of mature BFU-E and CFU-E in the bone marrow and spleen of Cy-treated polycythemic mice, which is contrary to the findings in normal mice treated with Cy, indicates the importance of Ep for BFU-E (d4) and CFU-E regeneration. This was confirmed when exogenous Ep was injected. The effect on BFU-E (d4) of exogenous Ep injected into the polycythemic Cy-treated mice at the time when primitive BFU-E have regenerated considerably suggested an influence of Ep on the transition of BFU-E (d10) to BFU-E (d4). The fast regeneration of CFU-E in the spleen of normal mice and after Ep injection in polycythemic Cy-treated mice confirms the well-known and significant role of the spleen in mouse erythropoiesis under stress conditions. It could be suggested that the patterns of BFU-E (d4) and CFU-E recovery as well as Ep responsiveness closely resemble the findings observed earlier for ERC in the same experimental model.     

Fetal hemoglobin analysis of erythroid bursts in patients with chronic myelogenous leukemia (CML)

Early erythroid progenitors (BFUE) form colonies of mature progeny in culture. The development of hemoglobinized red cells within multilineage colonies (CFUGEMM) and erythroid bursts is dependent upon exogenously added erythropoietin and molecules released by hemopoietic subpopulations. Mixed colonies and erythroid bursts were grown from 3 patients with Ph' chronic myelogenous leukemia (CML). It was found that some mixed hemopoietic colonies and erythroid bursts did not require exogenously added erythropoietin. An increase of the plating efficiency of BFUE could be observed when erythropoietin was added. Erythroid bursts grown without added Ep from samples of the patients with chronic myelogenous leukemia have a higher probability to contain HbF than clones grown in the presence of Ep. The data support the view of a phenotypical heterogeneity among clonal descendents of a common ancestor as previously postulated for CML.     

Possible role of tumor necrosis factor-alpha in erythropoietic suppression by endotoxin and granulocyte/macrophage colony-stimulating factor

Injection of bacterial endotoxin or granulocyte/macrophage colony-stimulating factor (GM-CSF) into exhypoxic polycythemic mice simultaneously with erythropoietin (EPO) suppressed erythroid cell formation, as monitored by ⁵⁹Fe incorporation into circulating red blood cells. This effect was dose-dependent and time-dependent. GM-CSF did not inhibit erythroid cell formation directly, as the antibody to the GM-CSF did not neutralize the effect of endotoxin, the inducer of GM-CSF. The suppression of both agents could be partially corrected by prior injection of a monoclonal antibody to tumor necrosis factor α (anti-TNFα). These results indicate that the suppression of EPO-induced erythroid cell formation by endotoxin and GM-CSF was due in part to the production of TNFα. © 1996 Wiley-Liss, Inc.