Acetylcholinesterase in the human erythron. III. Regulation of differentiation.

Acetylcholinesterase (AChE) is present in both primitive and mature erythroid cells, but a role for the enzyme in human hematopoiesis has not been defined. This prospect represented the primary objective of the following study. In clonal culture of normal human bone marrow cells, a "wave" of AChE activity was demonstrated, rising from undetectable levels to a peak (of 1.48 femto-moles per min per cell) at 10 days in the course of progressive erythroid clonogenesis. At concentrations of enzyme inhibitor that clearly reduced AChE activity in a dose-dependent fashion, there was no overall effect on erythropoiesis in vitro, but the clones were generally smaller and significantly more often multi-focal than in control cultures. Furthermore, in the presence of AChE inhibitors, a concentration-dependent increase in the myeloid-erythroid ratios of the culture harvests was observed. Likewise, a clear reduction in hemoglobination was revealed, in cells of 10 day cultures, from a mean hemoglobin concentration of 35.0 pg per cell in controls to 20.1 pg per cell in the presence of the maximal concentration of the inhibitor (10(-6) M eserine). These data point to a role for AChE in the regulation of differentiation in the human erythron.

Lithium and hydrocortisone interactions on cell growth and gene expression in human promyelocytic leukemia (HL60).

Lithium (Li) and hydrocortisone (HC) modify gene expression as well as stimulate human hematopoiesis in vivo and in vitro. The responses of blood-forming tissues, to these natural substances individually, are closely similar. However, clonogenic co-culture experiments, using normal human bone marrow cells, have revealed that the effects of Li and HC are not additive but rather mutually inhibitory. Studies of HL-60 cells in suspension indicate that growth of this human promyelocytic leukemic line is enhanced by HC at 10(-6) M, a "therapeutic" plasma concentration. This phenomenon is abrogated by the co-addition of Li at 3 x 10(-4) M, likewise a "therapeutic" plasma concentration. Neither substrate exerts any influence on the morphological or cytochemical features of differentiation which accompany exposure of HL-60 cells to dimethyl sulphoxide (DMSO), retinoic acid (RA), tetradecanoyl phorbol acetate (TPA) or sodium butyrate. In contrast, the expression of c-fms, a cellular proto-oncogene which codes for the CSF-1 (monocyte-macrophage colony-stimulating factor) surface membrane receptor, is modified by Li in the presence of the differentiation induction agents DMSO and RA which promote the development of mature granulocytes from HL-60 cells. These observations afford further insights into the humoral mechanisms which modulate the expression of genes involved in the regulation of hematopoiesis.

Biochemical assays of hemoglobin in normal human erythroid clones.

Quantitation of small amounts (2-3 micrograms) of hemoglobin (Hb) was achieved by each of three spectrophotometric techniques. An assay employing benzidine base consistently underestimated Hb at this level and gave a wide scatter of results at higher values. The cyanmethemoglobin method was at the limit of its sensitivity and suffers from the disadvantage of having high optical densities in the blank samples. By contrast, pyridine hemochromogen can be detected accurately with 200 ng of Hb (equivalent of 6,000 mature erythrocytes), and the regression line with this technique virtually passes through the origin. Furthermore, the extinction coefficient of pyridine hemochromogen in the Soret band is 13.5-fold greater than that of cyanmethemoglobin. In normal human erythroid clones, generated in vitro from bone marrow cells, mean cell hemoglobin (MCH) values were determined after various intervals of culture. The MCH after 5, 7, 10, and 14 days were 11.8, 15.8, 26.6, and 34.4 pg, respectively, by the pyridine hemochromogen method. Reaction product was also identified in granulocyte-macrophage clones, presumably reflecting the content of other heme proteins such as catalase and cytochromes. Account must be taken of this non-Hb material in computing true MCH values for erythroid cells.

Acetylcholinesterase in the human erythron. I. Cytochemistry.

The successful demonstration and localisation of acetylcholinesterase (AChE), in cells by a cytochemical technique requires maximal expression of enzyme activity, minimal loss of AChE and precise, quantitative generation of reaction product at the actual site of the protein in vivo. These requirements are addressed in a standard technique that has been modified to avoid or optimise fixation and to exhibit enzyme activity under close-to-physiological conditions of osmolality, pH, and temperature. With these refinements and with the use of a variety of substrates and enzyme inhibitors of different specificities, true AChE was demonstrable on the membrane of erythrocytes and in the nucleus and cytoplasm of erythroblasts in bone marrow and of the constituent cells of erythroid clones in vitro. The activity in erythrocytes from umbilical cord blood was less than that in corresponding cells from the peripheral circulation of adults. AChE was observed also in human megakaryocytes and in leucocytes at all levels of differentiation, including the components of granulocyte-macrophage clones. Pseudocholinesterase was detected likewise across the spectrum of erythroid (and leucocyte) ontogeny, suggesting that these enzymes may exercise an important function in hematopoiesis.

Acetylcholinesterase in the human erythron. II. Biochemical assay.

Acetylcholinesterase (AChE) is an integral erythrocyte membrane protein. A role for the enzyme in the developing human erythron is being explored. Assays of AchE by the standard Ellman technique overestimate the amount of enzyme by failing to account for the contribution of hemoglobin to the optical density of the reaction mixture. Furthermore, reliance on substrate selection alone for specificity is unsatisfactory. Incorporation of inhibitors of "true" AchE and of pseudocholinesterase confer greater ability to distinguish one enzyme from the other. In our experience, the inhibitor constant (Kl) for edrophonium, which is highly specific for AChE, is approximately 5 x 10(-5) M against adult human erythrocytes that contain significantly more total cholinesterase activity than do erythrocytes from umbilical cord blood. This consists of both "true" and "pseudo" enzyme, the former predominating and accounting for 0.75-1.65 (mean 1.02, median 0.87) femtomoles of substrate hydrolysed per min per cell in adult blood, with values of 0.15-1.04 (mean 0.71, median 0.73) obtained on cord blood. Moreover, the enzyme activity in neonatal erythrocytes has a rather different inhibitor profile from that of adult cells. AChE was also demonstrated in fresh (ALL) and cultured (K562 and HL60) human leukemic cells, as well as in primitive granulocyte-macrophage and erythroid cells cloned from normal human bone marrow. In the erythroid colonies the enzyme activity was 0-3.76 (mean 1.20, median 0.76) femtomoles per min per cell, apparently the first successful measurement of AChE in such cells.

Interactions of lithium and hydrocortisone in vitro on normal human hematopoiesis.

Lithium salts and glucocorticosteroid hormones stimulate hematopoiesis in vitro and in vivo. In the former circumstance, the impact is demonstrable with concentrations of these substances which reflect physiological and pharmacological conditions. The evident therapeutic benefit of these effects, in patients who experience myelosuppression and pancytopenia with the use of cytotoxic chemotherapy for malignant disease, could be enhanced if the influences of lithium and steroids on blood cell formation were at least additive. However, the observations in this study suggest that these substances can be mutually inhibitory, with respect to hematopoiesis, and do not support the use of lithium and steroids in combination to alleviate iatrogenic reduction in bone marrow function.

Modulation of human erythropoiesis by hydrocortisone in vitro.

The effect of hydrocortisone on the formation of erythroid colonies was studied in vitro in cultures of normal human bone marrow in an agar system. A range of concentrations of hydrocortisone (10(-10) to 10(-3) mol/l) showed significant effects on erythroid burst formation, in terms of the number of colonies, on d 10 and 14 of culture. At subphysiological concentrations (10(-10) to 10(-8) mol/l), no effect was seen, but at both physiological (10(-7) mol/l) and pharmacological (10(-6) and 10(-5) mol/l) concentrations stimulation of erythroid burst formation was noted. At 10(-4) mol/l hydrocortisone inhibited erythroid colony formation and 10(-3) mol/l was uniformly lethal. In the concentration range of 10(-7) to 10(-5) mol/l hydrocortisone also appeared to increase erythroid colony size. Thus hydrocortisone (10(-7) to 10(-5) mol/l) stimulates erythroid colony growth and it is suggested that the hormone may play a role in the physiological regulation of human erythropoiesis.

Lithium and boron in human blood.

A new method for determination of lithium and boron in biologic material has been developed. It uses neutron irradiation and subsequent measurements of helium 3 and helium 4 in a static mass spectrometer. Concentrations of lithium and boron in blood from seven apparently healthy donors were in rather narrow ranges (Li = 2.64 +/- 0.94 ng/gm dry weight, B = 97 +/- 22 ng/gm dry weight), and these concentrations also appeared to be correlated at the 0.05 level of significance.

Putative role for lithium in human hematopoiesis.

Ingestion of lithium salts increases production of neutrophil granulocytes from the bone marrow in human subjects when the concentration of the ion in blood is within the range 5 to 10 X 10(-4) mol/L. Results of preliminary dose-response experiments appeared to indicate that nanomolar levels of lithium stimulated clonal proliferation of granulocyte precursors from normal bone marrow in vitro, suggesting the possibility that this element may contribute to the physiologic regulation of blood cell formation in humans. The present studies confirm that the influence of lithium on hematopoiesis is evident in vitro at concentrations equivalent to that demonstrable in normal blood (2 to 4 X 10(-7) mol/L). Furthermore, such effects are not cell lineage specific, being observed also in clonogenic cultures of erythroid and eosinophil granulocyte progenitor cells, and the phenomenon attributed to lithium is a property shared with rubidium and cesium salts. These findings point to a role for lithium and its elemental relatives in the biophysical mechanisms involved with the control of human blood cell production.

Analysis of ultratrace lithium and boron by neutron activation and mass-spectrometric measurement of 3He and 4He.

A new technique for analysis of lithium and boron at ultratrace concentrations (less than 10(-8)g g-1) is described. The method consists of mass-spectrometric assay of 3He from decay of tritium produced by thermal-neutron reaction on 6Li, and 4He produced by thermal-neutron reaction on 10B. Two neutron-irradiation facilities were used: the McMaster reactor, which is 235U-enriched and light-water moderated; and a graphite-moderated thermal column attached to the 235U-enriched, heavy-water-moderated core at the National Bureau of Standards (NBS) reactor. In the McMaster irradiations, fast neutrons (greater than 0.2 MeV) induce the reactions 14N(n, 3H)12C, 12C(n, alpha)9Be, 16O(n, alpha)13C, and 14N(n, alpha)11B. These reactions become serious sources of error in samples such as human blood which have very low concentrations of lithium and boron, and high concentrations of nitrogen, carbon and oxygen. In the NBS thermal column, fast-neutron reactions are virtually absent, and only corrections for thermal-neutron capture by deuterium, and thermal-neutron (n, alpha) reactions on oxygen, sulfur, chlorine, potassium, and calcium need to be taken into account. Results are presented for various actual samples including human blood and its components, and some standard biological reference materials, to provide a realistic base for other workers to judge the reliability of the method.

Stimulation of human eosinophilopoiesis by hydrocortisone in vitro.

Excess hydrocortisone (HC) evokes neutrophilia and eosinopenia in man. In addition, the hormone enhances human granulopoiesis in vitro at physiological as well as pharmacological concentrations. This study addressed the prospect that HC exerts opposite effects on the precursors of neutrophils and eosinophils, stimulating the former and inhibiting the latter. Experiments conducted on unseparated bone marrow (BM) cells demonstrated an increase in eosinophil clonogenesis in vitro with the addition of HC to the culture system. Secondary cultures, established from such primary harvests, revealed that HC had a direct impact on the clonogenic cells. Furthermore, administration of HC to normal subjects, at a dose which resulted in consistent eosinopenia, prompted an increase in the generation of eosinophil clones from peripheral blood cells ex vivo. Thus the hormone stimulates production of both neutrophils and eosinophils. Previous reports of opposite effects appear to have resulted from deficiencies of growth factors in the cell cultures. The contrasting effects of HC on neutrophil and eosinophil concentrations in the peripheral blood are more likely due to opposite effects on the distribution of these terminally differentiated cells in the circulation and extravascular tissues.

Spurious detection of terminal deoxynucleotidyl transferase in phytohemagglutinin-stimulated lymphocytes.

Expression of terminal transferase (TdT) is believed to be restricted to primitive lymphoid cells; recently, however, indirect immunofluorescent (IF) assays have been used to demonstrate the apparent presence of TdT in phytohemagglutinin (PHA)-stimulated peripheral blood lymphocytes and in various nonlymphoid malignancies. Using an IF assay, we found that a heteroantiserum to TdT reacted with cultured and PHA-stimulated human peripheral blood mononuclear cells, but we were unable to confirm the presence of TdT in these cells using immunoblotting and biochemical assays. We conclude that the IF results are spurious and most likely represent recognition by the heteroantiserum of inducible protein(s) other than TdT.

Hydrocortisone promotes clonal granulopoiesis in cultures of normal human bone marrow--a "nutritional" advantage or true stimulation of growth?

The generation of normal human hemopoietic clones is promoted, in tissue culture, by physiological as well as pharmacological concentrations of hydrocortisone. In part this may reflect a facilitative, nutritional effect of the hormone which may be more evident in particular culture media. However, the findings, in liquid suspension cultures, of an increase in the absolute number of cells and a rise in the mitotic index in the presence of hydrocortisone, point to a real stimulation of cellular proliferation. Measurements of the number of clones early in culture, and the size of clones after longer intervals, indicate respectively that the hormone influences both recruitment of clonogenic cells into cytokinesis and amplification of established clones. The target cell in both processes may be the same, namely a morphologically recognizable entity, in the granulocyte lineage, having limited proliferative potential. Evidence in support of this interpretation includes the high cloning efficiency, small mean clonal size and brief clonal lifespan. These features suggest that members of the myeloblast-promyelocyte-myelocyte hierarchy are likely candidates, but whether the action of hydrocortisone is exerted directly on these cells, or on a more mature accessory population, remains to be determined.

Growth of erythroid colonies in agar cultures of normal human bone marrow.

The use of methylcellulose (MC) gels or plasma clots, for the support of human erythropoiesis in vitro, is associated with several technical disadvantages. Substitution of soft agar offers the prospect of overcoming these difficulties. In comparative studies, normal human bone marrow cells were cultured with erythropoietin (Epo) in agar (0.1%-0.3%) and MC. Concentrations of 0.175% and 0.2% agar proved to be optimal with respect to the combination of cloning efficiency and colony density. Further morphological examination revealed that subcolony formation in erythroid 'bursts' was influenced by gel viscosity. In additional experiments, miniaturising the assay system, to 0.25 ml culture volumes, increased cloning efficiency and reduced Epo utilization. These results confirm and expand earlier observations, and support a preference for the general use of agar in human erythroid cell cultures.

Storage and transportation of samples for analysis of terminal transferase by indirect immunofluorescence.

Expression of the terminal transferase (TdT) antigen, as revealed by indirect immunofluorescence, was retained in samples of human acute lymphoblastic leukemia (ALL) cells, of various phenotypes, during storage at 4 degrees C for almost six months. Similar specimens of ALL cells, of common phenotype, were transported successfully by air without notable loss of TdT. The unsatisfactory outcome of earlier studies may be explained on the basis of methodologic differences alone.

Demonstration of terminal deoxynucleotidyl transferase in single cells by indirect immunofluorescence. II. An examination of specificity.

Advantages in the use of indirect immunofluorescence for the identification of terminal transferase (TdT) in single cells may be offset by lack of specificity, as compared to the biochemical assay of the enzyme, especially in analyses of lymphocyte populations. False positive results were obtained in 15/15 tonsillectomy samples and in 9/27 specimens from children with acute lymphoblastic leukemia in remission, perhaps due to antihuman lymphocyte activity in the rabbit heteroantisera which are used in the indirect immunofluorescence technique. This phenomenon may be more pronounced in 'activated' normal lymphocytes. Such reactions are not due to antibodies directed against adenovirus, papovavirus or EB virus antigens, although these are common constitutents of human tonsillar cells. Additional problems with TdT heteroantisera may result from immunization with non-TdT determinants in calf thymus extracts, as was manifest in human non-lymphoid (KB) cells cultured in fetal bovine serum. These difficulties will be overcome only by production of a monoclonal antibody using human TdT as the antigen.

The intracellular location of terminal transferase does not vary with cell cycle stage.

The biochemical activity of terminal transferase (TdT) in the thymocytes of leukemic AKR mice has no relationship to cell cycle stage, unlike the activity of replicative DNA polymerase which increases during the period of DNA synthesis. Moreover, such assays of DNA polymerase alpha reveal a shift in enzyme activity from cytoplasm to nucleus during S phase. In the present study, the role of TdT in DNA metabolism was explored further by examining the intracellular location of the enzyme during cytokinesis. Single cell suspensions of thymocytes from leukemic AKR mice were partially synchronized by velocity sedimentation in a sucrose gradient at unit gravity and harvested according to cell cycle stage. The content and location of TdT in individual cells was determined by indirect immunofluorescence using a rabbit antiserum to calf thymus TdT as the primary antibody. There was no relationship of fluorescence intensity or of the proportion of TdT-positive cells to cell cycle stage. In all samples examined (n = 6) the enzyme was located almost entirely in the nucleus throughout cytokinesis. These results do not support the hypothesis that the intracellular location of TdT may vary with cell cycle stage and a role for the enzyme in DNA synthesis remains to be defined.

Human granulopoiesis in vitro: an advantage in the use of Iscove's modified Dulbecco's medium.

With the aim of determining whether Iscove's Dulbecco's medium (IMDM) provides a growth advantage in the support of granulopoiesis from cultures of human bone marrow in agar, samples from 20 normal subjects were examined in triplicate after 7, 10 and 14 days in parallel cultures containing IMDM or Dulbecco's medium. From every sample, more granulocyte-macrophage colonies were obtained at each culture interval with IMDM. In particular, the number of colonies with IMDM at 14 days (96 +/- 13 per 2x10(5) bone marrow cells) was almost double that with Dulbecco's medium (50 +/- 10). This increment consisted almost entirely of pure granulocyte colonies (P less than 0.001). No significant change in the proportion of eosinophil colonies was observed. These data indicate that IMDM does provide a growth advantage over Dulbecco's medium in the generation of granulocyte (neutrophil and eosinophil) colonies from agar cultures of normal human bone marrow.

Early relapse of acute lymphoblastic leukemia is not predictable by serial biochemical assays of terminal transferase activity in cells from peripheral blood.

Serial samples of peripheral blood were collected from 37 children with acute lymphoblastic leukemia (ALL) in remission. Activity of terminal transferase (TdT) was assayed by a biochemical technique. False positive results were obtained infrequently (approx. 1%), in contrast to experience with bone marrow analyses. However, early relapse of disease was not predictable in ALL by repeated measurement of TdT in circulating mononuclear cells during remission.