Haem is necessary for a continued increase in ferrochelatase mRNA in murine erythroleukaemia cells during erythroid differentiation

Summary The level of mRNA encoding ferrochelatase (FeC) was examined in two murine erythroleukaemia (MEL) clones, DS and DR , a DMSO-sensitive, and a DMSO-resistant clone, respectively. DS cells undergo erythroid differentiation by DMSO treatment with a marked increase in haem synthesis, while DR cells fail to do so due to the lack of the erythroid-specific δ-aminolaevulinate synthase (ALAS-E). Both DS and DR cells showed an increase in the level of FeC mRNA within 18 h of DMSO treatment. The level of FeC mRNA in DR cells was then decreased, while that in DS cells continued to increase for 72 h. Treatment with haemin significantly increased FeC mRNA in DR cells. When cells were treated with both DMSO and haemin, the level of FeC mRNA in DR cells increased to a level comparable to that in DS cells. These findings suggest that the failure to maintain increased FeC mRNA DR cells after DMSO treatment may be due to a deficiency of haem in these cells.     

The rapid and decremental change in haem oxygenase mRNA during erythroid differentiation of murine erythroleukaemia cells

Changes in mRNA for haem oxygenase (HO), the rate-limiting enzyme for haem catabolism, were examined in murine Friend-virus transformed erythroleukaemia (MEL) cells while they were induced to undergo erythroid cell differentiation by treatment with dimethyl sulfoxide (DMSO). When MEL cells were treated with 1.5% (v/v) DMSO, a rapid decrease in HO mRNA content was observed (less than 12 h) which reached the lowest value at 18 h (18% of the untreated control). HO mRNA levels remained at substantially lower levels (approximately 50%) than those in untreated controls thereafter. A rapid decline in HO mRNA may be involved in the cellular events that determine the onset of erythroid differentiation.     

Differential induction responses of delta-aminolevulinate synthase mRNAs during erythroid differentiation: use of nonradioactive in situ hybridization.

Expression of mRNAs encoding the erythroid-specific delta-aminolevulinate synthase (ALAS-E) and the nonspecific delta-aminolevulinate synthase (ALAS-N) were examined in murine Friend virus-transformed erythroleukemia (MEL) cells using nonradioactive in situ hybridization. Following dimethyl sulfoxide (DMSO) treatment, ALAS-E mRNA increased markedly, while ALAS-N mRNA did not increase in wild-type MEL cells. In contrast, in a DMSO-resistant clone of MEL cells, ALAS-E was not detectable before and after DMSO treatment. These findings suggest that ALAS-E and ALAS-N mRNAs are under separate controls and that the expression of ALAS-E mRNA is a critical event in erythroid differentiation.     

Inhibition of cellular iron uptake by haem in mouse erythroleukaemia cells

Haemin inhibited iron uptake from transferrin (Tf) by mouse erythroleukaemia cells (MELC) induced for differentiation by hexamethylene bisacetamide (HMBA). The rate of 59Fe internalization was decreased, but the rate and the extent of 125I-Tf endocytosis was unaffected by the addition of haemin. Haemin inhibited 59Fe incorporation into haem by a greater proportion than the overall uptake of 59Fe from Tf. The reduction of total cellular 59Fe uptake was more pronounced at 59Fe-Tf concentrations closer to saturation. Exogenous 5-aminolaevulinic acid stimulated 59Fe utilization for haem synthesis in MELC but did not revert the inhibition induced by haemin. Haem synthesis measured by 14C-glycine incorporation into haem was maintained for at least 1 h without an external transferrin iron source and was inhibited by the addition of haemin equally over the whole range of Tf concentrations studied. Desferrioxamine (DFO) stimulated cellular uptake of 59Fe by the uninduced cells and reverted the inhibition of 59Fe transport into HMBA treated cells caused by haemin. Addition of DFO within a short-term incubation had no effect on haem synthesis measured by 14C-glycine incorporation into haem. No evidence for a direct effect of haem on the transferrin cycle or iron release was found. It was concluded that the reduction of iron uptake by haemin treated MELC is secondary to the decrease in iron utilization for haem synthesis.     

Hemin does not cause commitment of murine erythroleukemia (MEL) cells to terminal differentiation

The effect of hemin on the differentiation program of murine erythroleukemia (MEL) cells has been investigated. While hemin treatment does induce increased levels of globin mRNA and hemoglobin, it fails to lead to other biochemical changes associated with MEL cell differentiation induced by DMSO and thioguanine. These include increased levels of the nuclear protein IP25 and of the enzyme cytidine deaminase. Clonal analysis of hemin-treated cells revealed that unlike other inducers, hemin does not cause a reprogramming of MEL cells to a specific limitation of proliferative capacity. These observations suggest that hemin differs from DMSO and thioguanine in that it exerts specific effects on globin expression in MEL cells without triggering commitment to the terminal differentiation program.     

Haem oxygenase augments porcine granulosa cell apoptosis in vitro

Haem oxygenases produce carbon monoxide, which, like nitric oxide, is a gaseous messenger molecule that is one of several important survival factors in ovarian follicles. However, little is known about the expression and possible functions of these enzymes in granulosa cells. The purpose of this study was to investigate the expression and possible role of haem oxygenases in porcine granulosa cells (PGCs). We obtained frozen sections of porcine ovaries and PGCs from ovarian follicles of various sizes by needle aspiration, and examined the expression of haem oxygenase-1 (HO-1; inducible type) and HO-2 (constitutive type) in PGCs by immunohistochemistry, RT-PCR, western blotting and flow cytometry. Both types of haem oxygenase were identified in PGCs throughout follicular development, but HO-1 was expressed primarily in granulosa cells in atretic follicles. We also investigated the effect of haem oxygenases on apoptosis of granulosa cells (flow cytometry to detect subdiploid DNA fluorescence) and on expression of Fas ligand (quantitative analysis of western blotting and flow cytometry). In tightly bound PGCs, the mean proportion of apoptotic cells treated with 1 microM haemin (a haem oxygenase substrate) was approximately 1.7-fold greater than that in untreated controls, and zinc protoporphyrin IX (ZnPP IX; a haem oxygenase inhibitor) completely inhibited the increase in apoptosis induced by haemin in 24-h culture. Conversely, in weakly associated PGCs, the proportion of apoptotic cells was not altered by haemin. The quantity of Fas ligand protein was increased in a dose-dependent manner in tightly bound PGCs treated with haemin compared with controls, and the haemin-induced increase in Fas ligand protein was inhibited by ZnPP IX. Thus we identified inducible HO-1 and constitutive HO-2 in PGCs throughout follicular development, and we conclude that products of reactions catalysed by haem oxygenases are likely to be important autocrine/paracrine factors that regulate apoptosis in PGCs.     

Oxidative stress is involved in hydroxyurea-induced erythroid differentiation

Hydroxyurea (HU), an inhibitor of DNA synthesis, can also induce haemoglobinization in certain erythroid cell lines. In this study, we report that intracellular peroxides levels were increased in HU-treated murine erythroleukaemia (MEL) cells and that l-acetyl-N-cysteine (LNAC), a potent reducing reagent, had a significant inhibitory effect on the HU-mediated induction of beta-globin, delta-aminolaevulinate synthase mRNA expression and haemoglobinization of MEL cells. In contrast, the addition of LNAC to dimethyl sulphoxide (DMSO)-treated MEL cells had a much smaller effect on the number of haemoglobinized cells. These findings suggest that oxidative stress is involved in HU-mediated induction of erythroid differentiation and that HU induces MEL cell differentiation by a mechanism different to that involved in DMSO-mediated differentiation. Our findings also suggest that the induction of MEL cell differentiation by HU does not involve RAS-MAP (mitogen-activated protein) kinase signalling.     

A rosette receptor assay with haem-microbeads. Demonstration of a haem receptor on K562 cells

A rosette assay was developed for the detection of haem receptor-bearing cells. Indicator particles were prepared by covalent binding of haem to acrylic microbeads. The new method was tested on K562 human erythroleukaemia cells, known to take up haem. In tests on several batches, 80-90% of the K562 cells were rosetted with haem-microbeads whereas mature erythrocytes were haem receptor-negative. Rosette formation was inhibited in a dose-dependent manner by micromolar concentrations of free haemin but not by albumin or transferrin. Uncoated microbeads or albumin-coated microbeads did not attach to K562 cells but transferrin-coated microbeads rosetted 50-70% of them. Diferric transferrin inhibited these rosettes, but haemin had no effect.     

Cyclic AMP Maintenance of Rabbit Reticulocyte Haem and Protein Synthesis in the Presence of Ethanol and Benzene

Haemin is necessary for maximal protein synthesis in intact reticulocytes and their cell-free lysate preparations. Under conditions of haemin deficiency a translational protein repressor of initiation (HCR_A) forms from a non-inhibitory form of the same protein (HCR₁) and protein synthesis ceases. It has recently been suggested that this repressor is a protein kinase and that high concentrations of 3′,5’cyclic AMP could substitute for haemin in maintaining reticulocyte lysate cell-free protein synthesis. The present study was undertaken, therefore, to investigate if cAMP had a similar function in intact rabbit reticulocytes. The cells were incubated with eidier epinephrine (10^-6 M), isoproterenol (10^-4 M), theophylline (10^-4 M) or dibutyryl cAMP (10^-4 M). All these compounds elevated the intracellular concentration of cAMP within the first 2 min of incubation, as measured by a specific competitive binding assay. This indicates that rabbit reticulocytes contain both adenyl cyclase and phosphodiesterase systems. The cells were incubated with an inhibitor of haem synthesis (either 0.1 M ethanol or 0.1 M benzene) to form HCR_A prematurely and inhibit protein synthesis. When the cells were incubated simultaneously with any one of these inhibitors and either epinephrine, isoproterenol, theophylline or dibutyryl cAMP, there was no inhibition of haem or protein synthesis and no demonstrable HCR_A. Furthermore, dibutyryl cAMP reversed the inhibition of both haem and protein synthesis induced by ethanol or benzene. Dibutyryl cAMP relieved the inhibition of haem synthesis prior to relieving the inhibition of protein synthesis, which indicates that sufficient haemin must be present before maximal protein synthesis may resume. In contrast, elevation of cyclic AMP levels did not protect against iron deficiency inhibition of either haem or protein synthesis. This indicates that in intact cells cAMP does not directly convert HCR_A to HCR₁ On the basis of these findings, we postulate that in intact reticulocytes (in contrast to cell-free systems) a major role of cAMP is maintenance of maximal haem synthesis, even in the presence of inhibitors. Intracellular haemin concentration appears to be a major control mechanism of protein synthesis via prevention and reversal of HCR_A. This model assures a balance between haem and globin synthesis in erythroid precursors.     

Ethanol Inhibition of Reticulocyte Protein Synthesis: The Role of Haem

Ethanol, in concentrations of 0.05-0.8 M, inhibited intact human and rabbit reticulocyte protein synthesis in the presence of iron-transferrin for endogenous haem synthesis. Associated with this effect there was a conversion of polyribosomes to monoribosomes and a decreased incorporation of radioactive leucine into nascent globin chains. When physiological levels of ethanol (0.05-0.1 M) were used, these effects were prevented by incubation with 50 muM haemin and reversed by removing the alcohol and reincubating with iron-transferrin or haemin. The polyribosomal disaggregation was also prevented by stopping ribosomal movement with 5 mM cycloheximide. Neither ATP nor GSH levels were altered in the presence of ethanol. When non-physiological levels of 0.8 M ethanol were used, haemin did not prevent the inhibition of protein synthesis. Likewise, in the rabbit reticulocyte cell-free lysate system containing haemin inhibition was noted at concentrations greater than 0.05 M ethanol. The polyribosomal disaggregation in reticulocytes incubated with 0.8 M ethanol was associated with decreased dissociation of monoribosomes into subunits. Similarly, when ribosomes were directly suspended cell-free in 0.1 or 0.8 M ethanol there was a decreased percentage of subunits. These results indicate that physiological concentrations of ethanol inhibit initiation of reticulocyte protein synthesis secondary to a block in haem synthesis. When intact cells are exposed to high non-physiological concentrations of ethanol the inhibition is secondary to decreased ribosomal dissociation. The cell-free lysate inhibition is also through this effect on ribosomal dissociation. This study supports the view that alcohol is a direct toxin to developing red cell precursors via its effect on mitochondrial haem synthesis. The physiological role of the decreased dissociation of monoribosomes into subunits is not yet clear.     

The role of haem biosynthetic and degradative enzymes in erythroid colony development: the effect of haemin

S ummary . In vitro culture of murine bone marrow has proved to be a useful system for defining the haem biosynthetic and degradative enzymatic pathways during erythroid colony (CFU-E) growth and development. Previous attempts to elucidate these pathways during erythropoiesis have been limited by the amount of biological material available as well as the sensitivity of specific enzyme assays. These conditions were overcome in the present study. In order to obtain an enrichment of CFU-E, nonadherent mouse bone-marrow cells were cultured in special culture plates using a modified methyl-cellulose medium with erythropoietin. CFU-E yields were increased approximately threefold. Utilizing the sensitive radiochemical assay, direct measurement of ALAS activity was determined in very early developing CFU-E cultures as well as mature cultures. ALAS activity was found to reach a peak after 60 h of culture growth and then started to decline in activity. Cellular synthesis of haem was determined (with ¹⁴C ALA) and minor modifications of standard assays were also made in order to determine spectrophotometrically δ-aminolaevulinic acid dehydratase (ALAD) and haem oxygenase activity in developing CFU-E cultures. These assays were reproducible with as few as four or five culture plates. It was found that ALAD activity rose progressively after 36 h of culture growth and reached a plateau at about 60 h of growth. A continuous increase in ¹⁴C ALA incorporation into haem was seen at later hours of culture growth, suggesting that the induction of other haem enzymes beyond ALAS is still rate limiting in haem synthesis. Conversely, haem oxygenase activity declined up to 60 h of growth and was elevated at later culture periods without a subsequent increase in ALASA and ALAD. The observed increase in haem enzymes brought about by haemin was completely suppressed by addition of cycloheximide to the cultures. These results suggest that     

Imatinib increases cytotoxicity of melphalan and their combination allows an efficient killing of chronic myeloid leukemia cells

BCR/ABL positive cells are known to be resistant to DNA damage induced by chemotherapy while they are sensitive to imatinib (IM), a tyrosine kinase inhibitor (TKI). To evaluate whether this drug can increase the activity of cytotoxic drugs on BCR/ABL positive cells, we measured the toxicity of cytosine arabinoside (ARA-C), hydroxyurea (HU) and melphalan (MEL), after a pretreatment of 24 h with IM on K562 cell line. The highest cytotoxic effect was seen when the TKI was followed by MEL; our results indicate that inhibition of BCR/ABL activity by IM increased the cytotoxicity of MEL by favoring the DNA damage induced by this drug and by shortening the time for DNA repair at the G2/M checkpoint. A stronger activation of some genes involved in both intrinsic and extrinsic apoptotic pathways was also observed with IM/MEL combination compared to IM or MEL alone. The drugs association was further tested in a type of BaF3 cells (TonB.210) where the BCR-ABL expression is inducible by doxycycline; in this model it was confirmed that a reduction of BCR/ABL activity resulted in an increased susceptibility to the cytotoxic effect of MEL. Furthermore, we studied the effect of IM/MEL treatment on the proliferative potential of myeloid progenitors of six CML patients at diagnosis. The analysis of CFU-GM and BFU-E colonies demonstrated that the IM/MEL combination was more effective than IM alone in reducing the overall number of colonies and the number of copies of BCR/ABL. In conclusion, our work shows that inhibition of BCR/ABL activity increases the toxicity of MEL and allows an efficient killing of leukemic cells, suggesting that a clinical development of this approach could have therapeutic advantages for CML patients.     

Relationship of mitochondrial membrane potential to hemoglobin synthesis during Friend cell maturation

Previous studies have shown that exposure to imidazole dissociates hemoglobin synthesis from other aspects of cell maturation in dimethylsulfoxide (DMSO)-treated mouse erythroleukemia (MEL) cells. In the present study, we have found that imidazole causes hyperpolarization of the mitochondrial membrane in MEL cells exposed to DMSO, in contrast to the depolarization observed with DMSO alone. Like the defect in hemoglobin synthesis, membrane hyperpolarization is reversible upon removal of imidazole and incubation of cells with DMSO alone. These correlations suggest that alterations in the electrostatic properties of the mitochondrial membrane, due directly or indirectly to the effects of imidazole, interfere with heme synthesis but not with other aspects of the maturation process in these developing erythroid cells.     

Hypoxia-induced Haem Oxygenase-1 gene expression in neonatal rat cardiac myocytes

Background: Haem oxygenase (HO-1), a heat shock or stress protein, is a rate-limiting enzyme in the conversion of pro-oxidant haem to biliverdin and carbon monoxide (CO). The products of haem catabolism serve regulatory and protective functions. Previous studies have shown that hypoxia induces HO-1 expression in cardiac myocytes. Accordingly, we investigated whether hypoxia-induced HO-1 expression is accompanied by increased CO production in cultured neonatal rat cardiac myocytes, and whether protein kinase C (PKC) is involved in hypoxia induced HO-1 gene expression.

Methods and Results: Expression of HO-1 in hypoxia-treated cells was examined by using northern and western blotting, and immunofluorescent staining. The level of HO-1 mRNA at 24 and 48 h was increased after the onset of hypoxia, with corresponding increase in the HO-1 protein level (6.7- and 8.7-fold at 24 and 48 h of hypoxia, respectively). HO-1 protein was colocalised with sarcomeric -actin in hypoxic myocytes. Hypoxia also significantly increased the production of CO by 2.5- and 8-fold at 24 and 48 h, respectively. Under normoxic conditions, activation of PKC by phorbol-12-myristate-13-acetate (PMA; 100 nmol/L) markedly increased HO-1 gene expression, while inhibition of PKC activity by calphostin C (100 nmol/L) blocked hypoxia-induced HO-1 gene expression in cardiac myocytes.

Conclusions: These results demonstrate that hypoxia markedly induces HO-1 expression and increases the production of CO in cardiac myocytes. This hypoxic response is attributed, at least in part, to activation of PKC. Increased HO-1 expression and resultant CO production may be beneficial with respect to protection of cardiac myocytes under hypoxic conditions.     

On the source of the non-transferrin-bound iron which accumulates in packed red blood cell units during storage

Background

Recent studies have shown large increases in non-transferrin-bound iron (NTBI) and biomarkers of oxidative stress in the extracellular medium of packed red blood cell units during storage. It has been further suggested that transfusion-mediated iron and oxidative load may contribute to transfusion-related morbidity in premature babies. The origin and nature of the NTBI is currently unclear, but the release of iron from oxidatively modified haemoglobin and haem has been suggested. The purpose of this study was to investigate whether this may be the case.

Materials and methods

The concentration of haem in the extracellular fluid of paediatric packed cell units stored from 3 to 35 days was measured using a commercial haem assay. In vitro studies were conducted using haem (haemin; ferriprotoporphyrin IX chloride) to determine whether the NTBI assay was able to react with and measure iron associated with haem in the presence and absence of oxidising agents.

Results

The level of haem in the extracellular fluid of paediatric packed cell units rose gradually from day 3 to day 21, then more rapidly to day 35. Very little NTBI was released from haem in the absence of oxidising agents, but the amount rose in a dose- and time-dependent manner in proportion to the oxidation of haem by incubation with H₂O₂.

Discussion

The results of the study imply that the NTBI measured in previous studies may derive from the oxidatively modified haem that builds up in the extracellular fluid of packed red blood cell units during storage. The potential influence of this on transfusion mediated morbidity is discussed.     

Transfection of mouse erythroleukemia cells with myc sequences changes the rate of induced commitment to differentiate.

We have examined the role of the c-myc protooncogene in chemically induced differentiation of mouse erythroleukemia (MEL) cells by transfecting the cells with recombinant plasmids in which c-myc coding sequences were cloned downstream from the mouse metallothionein I promoter in sense and antisense orientations. We previously showed that treatment of MEL cells with inducers of differentiation leads to a rapid (less than 2 hr) decrease in the level of c-myc mRNA. c-myc mRNA is then transiently restored to pretreatment levels approximately 12-18 hr later. These events occur prior to the detection of cells that are irreversibly committed to erythroid differentiation. MEL cell transfectants containing the plasmid with myc in the sense orientation express a chimeric MT-myc mRNA, which also decreases shortly after addition of inducer. However, these clones reexpress myc RNA more rapidly than the parental line and they also differentiate more rapidly. On the other hand, transfectants containing the plasmid with myc in the antisense orientation exhibited a delay in the reexpression of c-myc mRNA and were found to differentiate more slowly than parental cells. Thus, we find a correlation between the time at which myc RNA is reexpressed following inducer treatment and the rate of entry of cells into the terminal differentiation program.     

Intracellular forms of iron during transferrin iron uptake by mitogen-stimulated human lymphocytes

Transferrin receptors expressed by mitogen stimulated human lymphocytes mediate the uptake of transferrin iron into haem, ferritin and a non-haem, non-ferritin component. In spite of different rates of iron uptake by cells from different individuals, the proportional incorporation of 59Fe into these components was similar, suggesting that there was an obligatory relationship between the different forms of iron in the cells. By 3 h over 60% of the iron taken up was incorporated into ferritin while less than 10% was found in haem. Initially (10 min) non-haem, non-ferritin iron comprised 70% of total iron and this diminished to 30% by 3 h. At 10 min 80% of iron in the non-haem, non-ferritin component was retained by anti-transferrin affinity columns indicating it was transferrin-bound. The proportion retained fell to reach a steady state level of 50% by 60 min. These results indicate that 10-20% of the iron in the cells was not recognized as transferrin, ferritin or haem iron. The finding that iron incorporation into this unidentified pool reached equilibrium while that into haem and ferritin increased suggests the iron may act as a precursor for functional and storage compounds.