Erythrocyte membrane antigen expression during friend cell differentiation: Analysis of two non-inducible variants

Erythrocyte membrane antigens have been detected on induced Friend erythroleukemic cells with a rabbit antiserum raised against mouse erythrocyte membranes. The antibody specificities of this antiserum have been quantitatively analyzed using a cellular radioimmunoassay. After absorption with thymocytes, the rabbit anti-erythrocyte membrane serum bound to dimethylsulfoxide (DMSO)-induced Friend erythroleukemic cells and to mouse erythrocytes but not to uninduced Friend cells or thymocytes. Reciprocal inhibition studies demonstrated that, following complete thymocyte absorption, the antiserum detected similar antigenic specificities, termed erythrocyte membrane antigens (EMA), on both mature erythrocytes and induced Friend cells. The expression of these erythrocyte membrane antigens was also induced on Friend cells by other agents, such as ouabain and dimethylacetamide (DMA). In contrast, exogenous hematin, which did not induce hemoglobin synthesis in the Friend cell clones used in this study, also did not induce erythrocyte membrane antigen expression. Two independently derived variant clones which do not produce hemoglobin in reponse to DMSO were analyzed for their ability to produce erythrocyte membrane antigens in response to various inducers of Friend cell differentiation. Clone TG-13 is not inducible by DMSO or hematin but is weakly induced by DMA for both hemoglobin production and erythrocyte membrane antigen expression. Another variant clone, M18, was also analyzed. This clone does not synthesize detectable hemoglobin when grown in either DMSO or hematin alone, but undergoes extensive hemoglobin synthesis when grown in medium containing both DMSO and hematin. M18 does, however, express erythrocyte membrane antigens when grown in DMSO alone: the presence of hematin and DMSO together in the growth medium does not enhance expression of these antigens. Thus M18 appears to be defective for hemoglobin inducibility, and this defect can be overcome by exogenous hematin; however, the expression of erythrocyte membrane antigens is not affected by this block in hemoglobin synthesis. The results with the variant clones are discussed in terms of a program for Friend cell differentiation in which the induction of hemoglobin synthesis and erythrocyte membrane antigen expression are under both co-ordinate and separate controls.     

Increase of 2,3-bisphosphoglycerate synthase/phosphatase during maturation of reticulocytes with high 2,3-bisphosphoglycerate content

In the rabbit and in the rat, which possess erythrocytes with high concentration of 2,3-bisphosphoglycerate, the 2,3-bisphosphoglycerate synthase activity increases more than two fold during reticulocyte maturation. Isolation of the enzymes with 2,3-bisphosphoglycerate synthase activity present in extracts of reticulocytes and mature erytrocytes by ion exchange fast liquid chromatography shows that the increase in the synthase activity is due to the accumulation of the bifunctional enzyme 2,3-bisphosphoglycerate synthase/phosphatase (EC 2.7.5.4/EC 3.1.3.13) which represents more than 80% of the synthase activity of the cell extracts. During reticulocyte maturation phosphoglycerate mutase (EC 5.4.2.1), which makes a small contribution to the 2,3-bisphosphoglycerate synthase activity in the erythroid cells, decreases in the rabbit and remains constant in the rat.     

Acetylcholinesterase, carbonic anhydrase and catalase activity in Friend erythroleukemic cells, non-erythroid mouse cell lines and their somatic hybrids

Friend erythroleukemic cells (clone 745), when compared with transformed mouse fibroblasts, hepatoma, myeloma and teratocarcinoma cells, display high levels of acetylcholinesterase and carbonic anhydrase activity. Dimethylsulfoxide, which enhances hemoglobin production in Friend cells, also increases the activity of both enzymes. Inhibitor studies demonstrate that all the cholinesterase activity present in Friend cells is accounted for by the “true” acetylcholinesterase form of the enzyme. Cultured hepatoma cells have low levels of both acetylcholinesterase and pseudocholinesterase. Hybrids between the Friend cells and either transformed mouse fibroblasts or hepatoma cells not only fail to produce any detectable level of hemoglobin or globin mRNA, but also have no carbonic anhydrase activity and only low levels of acetylcholinesterase activity. Dimethylsulfoxide induces an increase in acetylcholinesterase activity in these hybrids. Catalase, which does not increase during erythropoiesis until the reticulocyte stage, is at roughly the same level in Friend cells and the non-erythroid cells we have examined; dimethylsulfoxide has no effect on the level of catalase activity in any of these cells. The data suggest that the Friend cells represent an intermediate stage of erythroid differentiation. It would appear that dimethylsulfoxide treatment stimulates the cells to differentiate further, along a pathway whose events closely follow normal in vivo erythroid differentiation. The data also support the idea that a set of genes usually expressed together in a particular cell type can be coordinatively affected in hybrids between cells maintaining two different epigenetic states.     

Introduction of Haemoglobin Synthesis in Friend Leukaemia Cells without the Necessity of Mitosis

Friend leukaemia cells (745A) were cultured in a soft agar containing 2% dimethylsulfoxide. In situ observation of the marked individual cells verified that the majority of haemoglobin-synthesized cells had not divided during cultivation. This is a direct proof that mitosis is unnecessary for the induction of erythroid differentiation of Friend leukaemia cells.     

Constitutive synthesis of globins within most of the members of an uninduced, proliferating population of Friend erythroluekemic cells.

The proteins synthesized by Friend erythroleukemic cells (clone 745), before and after they have been induced by treatment with dimethylsulfoxide to undergo overt erythroid maturation, have been analyzed by a number of biochemical methods. None of these techniques has, however, revealed any consistent observable differences (other than quantitative variations in the concentrations of particular proteins) between the “uninduced” and the “induced” cells. Thus, various classes of cellular proteins (whole cell, cytoplasmic, and nuclear) appear to be very similar in the two types of cells when analyzed by two-dimensional gel electrophoresis. Of particular interest is the finding that the strain-specific hemoglobins of $\text{DBA}\text{2}$ mice (the line from which the Friend cells were derived by viral transformation) are being synthesized in both the uninduced and the induced cells but at apparently markedly different rates. A monospecific antibody preparation against purified $\text{DBA}\text{2}$ mouse α- and β-globins was used to quantify the amount of these proteins in Friend cells during the course of dimethylsulfoxide-induced differentiation. Rocket gel immunoelectrophoresis, radial immunodiffusion analysis, and competitive radioimmunoassay revealed that the uninduced Friend cells contained, on the average, about 0.45 pg of hemoglobin per cell and this amount increased about 32-fold (to about 14.4 pg per cell) after 5 days of induction. Furthermore, use of the antibody preparation in indirect immunofluorescence studies revealed the constitutive synthesis of low levels of globins in virtually all of the uninduced, actively dividing, leukemia cells. These studies indicate that the standardly used benzidine staining method for detecting hemoglobins is insensitive for the detection of low levels of these proteins. The results of this study indicate that the cells in this particular leukemic population are, in reality, already “differentiated.” This suggests that the cells may represent an intermediate stage of erythroid maturation whose further progress along the normal in vivo terminal differentiation pathway has been blocked by viral transformation. Perhaps various “inducers,” such as dimethylsulfoxide, may enable these cells to partially overcome this block.     

2,3-Bisphosphoglycerate,fructose, 2,6-bisphosphate and glucose 1,6-bisphosphate during maturation of reticulocytes with low 2,3-bisphosphoglycerate content

In contrast to the species with erythrocytes of high 2,3-bisphosphoglycerate content, in the sheep the concentration of 2,3-bisphosphoglycerate decreases during maturation of reticulocytes. The decrease can be explained by the drop of the phosphofructokinase/pyruvate kinase and 2,3-bisphosphoglycerate synthase/2,3-bisphosphoglycerate phosphatase activity ratios that result from the decline of phosphofructokinase, pyruvate kinase, phosphoglycerate mutase and the bifunctional enzyme 2,3-bisphosphoglycerate synthase/phosphatase. The concentrations of fructose 2,6-bisphosphate and aldohexose 1,6-bisphosphates also decrease during sheep reticulocyte maturation in parallel to the 6-phosphofructo 2-kinase and the glucose 1,6-bisphosphate synthase activities.     

Molecular cloning and nucleotide sequence of murine 2,3-bisphosphoglycerate mutase cDNA

Cloning and sequencing of a murine cDNA with the entire coding region of 2,3-bisphosphoglycerate mutase is reported, as a prerequisite for further expression studies of this erythroid specific enzyme in Friend mouse erythroleukemia cells. A comparison between species of the deduced amino acid sequences of these proteins shows 20 substitutions between mouse and human and 21 between mouse and rabbit: none of these substitutions are in positions assumed to be in the active site. Amino acid alignment with the other related enzymes, the phosphoglycerate mutases, in combination with crystallographic data from yeast phosphoglycerate mutase, gives some insight into the structure/function correlation for this protein family. Amino acid residues which are most likely critical for either 2,3-bisphosphoglycerate mutase or phosphoglycerate mutase function are pointed out. Concerning the phylogenetic analysis, phosphoglycerate mutases B and M from mammalians appear to have diverged with the yeast enzyme from a common ancestor, before the emergence of the 2,3-bisphosphoglycerate mutases.     

Metabolism of glycerate 2,3-P2--XII. Characterization of the 2,3-bisphosphoglycerate synthase-phosphatase and of the hybrid phosphoglycerate mutase/2,3-bisphosphoglycerate synthase-phosphatase from pig brain

2,3-Bisphosphoglycerate synthase-phosphatase and the hybrid phosphoglycerate mutase/2,3-bisphosphoglycerate synthase-phosphatase have been partially purified from pig brain. Their 2,3-bisphosphoglycerate synthase, 2,3-bisphosphoglycerate phosphatase and phosphoglycerate mutase activities are concurrently lost upon heating and treatment with reagents specific for histidyl, arginyl and lysyl residues. The two enzymes differ in their thermal stability and sensitivity to tetrathionate. Substrates and cofactors protect against inactivation, the protective effects varying with the modifying reagent. The synthase activity of both enzymes shows a nonhyperbolic pattern which fits to a second degree polynomial. The Km, Ki and optimum pH values are similar to those of the 2,3-bisphosphoglycerate synthase-phosphatase from erythrocytes and the hybrid enzyme from skeletal muscle. The synthase activity is inhibited by inorganic phosphate and it is stimulated by glycolyate 2-P.     

Functional characterization of the enzymes with 2,3-bisphosphoglycerate phosphatase activity from pig skeletal muscle

In pig skeletal muscle exist four enzymes with 2,3-bisphosphoglycerate phosphatase activity. Two of them (forms I-A and I-C) are multi-functional enzymes which, in addition to the phosphatase activity, possess 2,3-bisphosphoglycerate synthase and phosphoglycerate mutase activities. The other two enzyme forms (II-A and II-B) only show the phosphatase activity. The four enzymes differ in substrate specificity. Form I-C is highly specific for glycerate 2,3-P2; form I-A also hydrolyzes the monophosphoglycerates and forms II-A and II-B are specific for phosphoester bonds adjacent to a C-1 carboxylic group. The enzymes possess similar Km, Kcat and optimum pH value, but they are differently inhibited by the reaction products. They are also differently affected by glycolate-2-P (their main activator) and by other modifiers. Probably form I-A, which corresponds to M-type phosphoglycerate mutase, is the main enzyme implicated in the breakdown of glycerate 2,3-P2 in pig muscle.     

Multifunctional enzyme, bisphosphoglyceromutase/2,3-bisphosphoglycerate phosphatase/phosphoglyceromutase, from human erythrocytes. Evidence for a common active site.

Bisphosphoglyceromutase and 2,3-bisphosphoglycerate phosphatase activities responsible for 2,3-bisphosphoglycerate metabolsim in human red cells are displayed by the same enzyme protein which has phosphoglyceromutase activity [Sasaki, R., et al. (1975) Eur J. Biochem. 50, 581-593]. This enzyme was subjected to chemical modification by trinitrobenzenesulfonate. The three enzyme activities were inactivated by trinitrobenzenesulfonate at the same rate. The sulfhydryl content of the enzyme was unchanged during trinitrophenylation, indicating that derivatization was through the amino group. Trinitrophenylation of about one amino group per mole of the enzyme resulted in complete loss of the three activities. Both 2,3-bisphosphoglycerate and 1,3-bisphosphoglycerate inhibited trinitrophenylation and effectively protected the enzyme from inactivation. Although monophosphoglycerates did not show any protective effect at concentrations which should be adequate based upon their kinetic constants, they were protective at higher concentrations. Inactivation by trinitrophenylation was an apparent first-order reaction. The dissociation constant of the enzyme - 2,3-bisphosphoglycerate complex was determined by analyzing the first-order reaction on the assumption that the protective effect of 2,3-bisphosphoglycerate was due to competition with trinitrobenzenesulfonate. The dissociation constant was in good agreement with kinetic constants of 2,3-bisphosphoglycerate in the enzyme reactions, which indicated that 2,3-bisphosphoglycerate did indeed exert its protective effect through competition with trinitrobenzenesulfonate for an amino group of the enzyme. The protective effect of monophosphoglycerates could be rationalized with kinetic evidence that 2-phosphoglycerate at high concentrations interacts with the 2,3-bisphosphoglycerate binding site. These results indicate that the enzyme exhibits the three enzyme activities at a common active site at which one amino group essential for binding of bisphosphoglycerates is located. Based on the multifunctional properties of this enzyme, a possible mechanism was discussed for regulation of 2,3-bisphosphoglycerate metabolism in human red cells.     

Metabolism of 3-phosphoglyceroyl phosphate in phosphoenolpyruvate-enriched human erythrocytes

The concentration of 3-phosphoglyceroyl phosphate in erythrocytes was increased by more than 100-fold when red cells were incubated with extracellular phosphoenolpyruvate at 37 degrees C. Since these elevated levels were maintained for 60 min, the metabolism of 3-phosphoglyceroyl phosphate and related compounds could be investigated in phosphoenolpyruvate-treated erythrocytes. 2,3-Bisphosphoglycerate synthesis was not affected by intracellular pH when the 3-phosphoglyceroyl phosphate level was constant but did vary with 3-phosphoglyceroyl phosphate concentration. On the other hand, the relationship between the rate of 2,3-bisphosphoglycerate synthesis and 3-phosphoglyceroyl phosphate concentration was not straightforward. At relatively low concentrations of 3-phosphoglyceroyl phosphate, the observed rate of 2,3-bisphosphoglycerate synthesis agreed with a rate calculated from a formula incorporating kinetic parameters of purified 2,3-bisphosphoglycerate synthase (Rose, Z.B. (1973) Arch. Biochem. Biophys. 158, 903-910). However, at high concentrations of 3-phosphoglyceroyl phosphate, the observed rate of 2,3-bisphosphoglycerate synthesis was lower than the calculated value. The concentration of glucose 1,6-bisphosphate did not increase even when 3-phosphoglyceroyl phosphate was elevated to 200 microM. Elevated levels of intracellular 2,3-bisphosphoglycerate did not inhibit glycolytic activity in these erythrocytes. These results suggest that incubation of erythrocytes with phosphoenolpyruvate is a useful technique to investigate the effect of metabolic perturbations at the intermediate stages of glycolysis.     

Induction of erythroid differentiation in Friend leukemia cells by bromodeoxyuridine

Treatment of Friend leukemia cells with BrdU, the thymidine analog which interferes with DMSO induced differentiation in these cells as well as the expression of differentiated character in many other cell systems, is capable of inducing erythroid differentiation. Globin mRNA, as assayed by hybridization to globin cDNA, increases 2.5- to 30-fold after appropriate treatment with BrdU. This effect was observed with several different subclones of three independent Friend tumor cell lines. After BrdU treatment, globin mRNA content may reach up to 10-20% of the levels in DMSO induced cultures. The induction of erythroid differentiation is also apparent when accumulated heme content or the appearance of benzidine positive cells is monitored. One Friend cell line (745) we examined was not induced by BrdU although it incorporated an amount of BrdU into its DNA comparable to that incorporated by the other cell lines. In addition, BrdU did interfere with DMSO induction in this cell line. These results suggest that two different mechanisms may be operative in regulating erythroid differentiation in Friend leukemia cells. While BrdU interferes with the mechanism activated by DMSO treatment, this analog could independently activate an alternative mechanism.     

Fructose 2,6-bisphosphate and glucose 1,6-bisphosphate levels in erythrocytes with high and low 2,3-bisphosphoglycerate content during postnatal development

In rabbit and sheep erythrocytes the concentrations of 2,3-bisphosphoglycerate, fructose 2,6-bisphosphate and glucose 1,6-bisphosphate suffer important changes after birth, which differ in both species. The changes of fructose 2,6-bisphosphate and glucose 1,6-bisphosphate correlate with the changes in the levels of the enzymatic activities involved in their synthesis. The change of 2,3-bisphosphoglycerate levels in rabbit but not in sheep erythrocytes could be explained by the changes of the phosphofructokinase/pyruvate kinase and 2,3-bisphosphoglycerate synthase/2,3-bisphosphoglycerate phosphatase activity ratios.     

Isolation and characterization of cDNA encoding rabbit reticulocyte 2,3-bisphosphoglycerate synthase

We have isolated and sequenced a cDNA clone containing the entire coding region of rabbit reticulocyte 2,3-bisphosphoglycerate (DPG) synthase. The cDNA was verified by translation of the hybridization-selected RNA and by demonstrating identity of the deduced amino acid (aa) sequence to the sequences of CNBr peptides of the purified enzyme. The aa sequence of the enzyme was homologous to the reported sequence of the human enzyme [Haggarty et al., EMBO J. 2 (1983) 1213-1220], especially in the N-terminal half (aa 1-142). Northern blot analysis of rabbit reticulocyte poly(A)+ RNA revealed a single species of mRNA with about 1700 nucleotides.     

Terminal differentiation in cultured Friend erythroleukemia cells.

Two populations of differentiated, hemoglobin-containing cells have been identified in cultures of Friend murine erythroleukemia cells (Friend cells): terminally differentiated benzidine-positive (B+) cells that are no longer capable of proliferation and are arrested in the G1 phase of the cell cycle, and their precursors, traversing B+ cells which undergo two or three cell divisions before reaching their terminally differentiated state. Thus Friend cells in suspension culture retain a limited capacity to synthesize DNA and divide after commitment to erythroid differentiation. We identified terminally differentiated cells using autoradiography after benzidine staining. We also developed a quantitative flow microfluorometric assay to distinguish cells that are terminally differentiated from those cells committed to differentiation but still capable of proliferation.We developed a purification procedure to isolate terminally differentiated Friend cells. Their DNA content was the same as that of the undifferentiated cells in G1 by both the diphenylamine reaction and a fluorescence assay. No loss of DNA was detected during the differentiation of Friend cells. As many as 72% of the total cells in a culture induced with DMSO (88% B+) were differentiated cells arrested in G1. As a control, a DMSO-resistant line derived from 745A neither differentiated nor arrested in G1 after growth in the presence of DMSO. The results of these studies were obtained using several compounds that induce differentiation and three independently isolated clones of 745A. We also observed arrest of differentiated cells in G1 with the two other well characterized, independently derived erythroleukemia cell lines, F4-1 and T3-C1-2.     

New U1 RNA species found in Friend SFFV (spleen focus forming virus)-transformed mouse cells

The U1 RNA species in 10 mouse cell lines were examined by two-dimensional polyacrylamide gel electrophoresis. Seven cell lines that were not infected by Friend spleen focus forming virus gave only one (I) or two (I and II) U1 RNA-containing spots. However, two Friend cell lines (FVTCT and Friend 745a cells) gave three spots (I, II, and III) and another Friend cell line, K-1 cells, gave four spots (I, II, III, and IV). As a result of further separation and fingerprinting analysis of each spot, FVTCT and Friend 745a cells were found to contain U1a-1, U1b-1, -2, and -6 RNAs whereas K-1 cells were found to contain several U1 RNAs, which we call U1a-1 and -2, U1b-4, -5, and -6 RNAs. We determined the sequences of these seven U1 RNAs and found that mouse U1 RNAs had two basic sequences (U1a and -b). The nucleotide sequence of U1a-1 RNA was identical to that of rat U1a RNA, while U1a-2 RNA was one base different from U1a-1 RNA. Relative to U1a-1 RNA all of the U1b RNAs had five base substitutions and one additional base and were under-methylated in the center. U1b-6 RNA contained two base substitutions and one base addition in the 3'-terminal portion of U1b-1 RNA. U1b-2, -4, and -5 RNAs, which were observed only in Friend cells, each had an additional base substitution in the 5'-half of U1b-1 RNA.     

Synthesis of erythrocyte-specific proteins in cultured friend leukemia cells

We have studied synthesis of specific proteins in two permanent ilness of Friend virus-induced erythroleukemia cells (Friend line 745 and Ostertag line FSD-1, both derived from DBA/2 mice). By 96 hr following treatment with 1–2% dimethyl sulfoxide (Me₂SO), up to 25% of the protein being synthesized by both these cultures is hemoglobin. At that time, hemoglobin constitutes up to 10% of the cellular soluble protein. Both lines synthesize heme and globin coordinately, and α and β globin chains in a nearly balanced 1:1 ratio. However, the ratio of β^Major:β^Minor chains synthesized by these induced Friend leukemia (FL) cells is approximately 9 in the FSD-1 line and 1.3 in the Friend Clone 745 line, whereas it is 4 in normal adult DBA/2 mouse erythrocytes. Evidence for the latter conclusion was obtained by electrophoresis of FL hemoglobins on cellulose acetate membranes, and also by chromatographic separation of α, β^Major, and β^Minor globins on carboxymethylcellulose in 8 M urea at 20°C. Carbonic anhydrase activity per mg protein is 3 times higher in induced than in control cultures. 2,3-diphosphoglyceric acid is not found in induced FL cells. Induced and control FL cells agglutinate strongly and equally with Phaseolus vulgaris phytohemagglutinin. The developmental process in these cultured leukemia cells appears to be an aberrant erythropolesis.     

Purification of 2,3-bisphosphoglycerate synthase-phosphatase from pig skeletal muscle

Two enzymes which possess 2,3-bisphosphoglycerate synthase, 2,3-bisphosphoglycerate phosphatase and phosphoglycerate mutase activities have been purified from pig skeletal muscle. One of the enzymes corresponds to type M phosphoglycerate mutase. The other enzyme shows properties similar to those of the 2,3-bisphosphoglycerate synthase-phosphatase present in mammalian erythrocytes. The erythrocyte and the muscle enzyme possess the same molecular (56 000) and subunit (27 000) weights. The synthase, phosphatase and mutase activity ratio is similar in both enzymes, and they are affected by the same inhibitor (glycerate 3-P) and activators (glycolate 2-P, pyrophosphate, sulfite and bisulfite).     

Specific esterase activity induced in mouse Friend erythroleukemia cells by dimethylsulfoxide.

Three cell lines of mouse erythroleukemia transformed by Friend virus (FLC), namely 745, F4-1, and 3BM-78, were grown for six days in the absence or in the presence of 1.5% (v/v) dimethylsulfoxide (DMSO) and compared cytochemically for naphtol-AS D-chloroacetate esterase (E), alkalinephosphatase (AP), myeloperoxidase (MP) and periodic acid Schiff (PAS) reaction activity. In the absence of inducer only 1–2% of slightly E positive cells could be found. E positivity greatly increased in 3BM-78 and F4-1 but poorly in 745 cells, after treatment with DMSO. Unlike E reaction, AP and MP reactions were positive in about 5% 3BM-78 and F4-1 cells without DMSO, but there were no positive cells after DMSO treatment. All three lines were always PAS negative. Hemoglobin synthesis (benzidine staining) was intensively induced by DMSO in all three lines. Morphologically after DMSO treatment, FLC matured displaying characteristics of basophilic megaloblastoid cells. The emergence of specific esterase activity, a marker of granulocytes, in FLC differentiating along the erythroid pathway, suggests that in these leukemia cells the genetic determinants for leukopoietic differentiation are retained and capable of being expressed phenotypically.     

Hybrid forms of phosphoglycerate mutase and 2,3-bisphosphoglycerate synthase-phosphatase

Purified phosphoglycerate mutase from pig skeletal muscle and 2,3-bisphosphoglycerate synthase-phosphatase from pig erythrocytes were hybridized “in vitro”. The hybrid showed a behaviour on electrophoresis and on ion-exchange chromatography similar to that of a naturally occurring enzyme with phosphoglycerate mutase, 2,3-bisphosphoglycerate synthase and 2,3-bisphosphoglycerate phosphatase activities present in pig skeletal and heart muscle. Both the hybrid and the muscle enzyme possess similar activities ratio. From these and previous data it is suggested that the six enzymatic forms with phosphoglycerate mutase, 2,3-bisphosphoglycerate synthase and 2,3-bisphosphoglycerate phosphatase activities detected in mammalian tissues (Carreras et al. 1981, Comp. Biochem. Physiol. 70B, 477–485) result from combination of three subunits (types M, B and E).