Tadashi Nagai, MD, PhD, Division of Haematology, Department of Medicine, Jichi Medical School, 3311–1 Yakushiji, Minamikawachi-machi, Kawachi-gun, Tochigi, 329–0498, Japan. E-mail: firstname.lastname@example.org
Summary. 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 β-globin, δ-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.
Hydroxyurea (HU) is a ribonucleotide reductase inhibitor that exhibits an antitumour effect by inhibiting DNA synthesis. HU is also used for the clinical treatment of patients with sickle cell anaemia, because it induces the synthesis of fetal haemoglobin (HbF), which then inhibits the polymerization of HbS (Rodgers, 1997). Previous studies showed that HU increased fetal haemoglobin synthesis in immature cultured erythroid cells (Fibach et al, 1993) and globin mRNA in the human erythroleukaemia cell lines, K562 and HEL (Xu & Zimmer, 1998; Zhang et al, 2001). However, the mechanism of HU-mediated induction of erythroid differentiation has not been elucidated. Recently, it was shown that treatment of K562 cells with HU significantly altered the levels of phosphorylation of extracellular signal-regulated kinase 1 (Erk1) and p38, which belong to the family of RAS-mitogen-activated protein kinases (MAPKs) (Park et al, 2001). MAPKs are stress-activated protein kinases that are activated through the phosphorylation of threonine and tyrosine residues, and are known to be involved in the regulation of various cellular processes such as cell survival, proliferation and differentiation.
The proper redox state, which is controlled by the balance between reactive oxygen intermediates (ROI) and antioxidants, is known to be critically important for various cell functions. For example, ROI regulate the function of gene products, and are important regulators of intracellular signal transduction pathways and certain transcription factors (Nakamura et al, 1997). Oxidative stress might, therefore, be involved in cellular responses, including erythroid cell differentiation. In fact, it has been suggested that oxidative stress may be involved in chemically induced differentiation of K562 cells (Chenais et al, 2000), and murine erythroleukaemia (MEL) cells (Comelli et al, 1994).
In the present study, HU was shown to induce erythroid differentiation of MEL cells and that oxidative stress may be involved in the HU-mediated induction of differentiation, but in a RAS-MAPK-signalling-independent manner.
Materials and methods
Cell culture. MEL cells, clone DS19, were grown in Roswell Park Memorial Institute (RPMI)-1640 medium supplemented with 10% fetal bovine serum, and cultures were split every 3 d. For the induction of erythroid differentiation, cells were first seeded at a density of 5 × 104 cells/ml and incubated for 16 h to ascertain logarithmic cell growth. Then cells were treated with a chemical or hyperoxia (50% O2). HU (100 mmol/l) or dimethyl sulphoxide (DMSO) 2%[v/v] was used as the chemical for the induction of erythroid differentiation. Incubation was continued in the presence of the chemical or hyperoxia, for various periods of up to 96 h, and then the number of dianisidine-positive cells was determined as a percentage of the total number of cells.
Determination of intracellular peroxides in DS19 cells. Cells were incubated with 2′,7′-dichlorofluorescein diacetate (Molecular Probes, Eugene, OR, USA) as the fluorogenic substrate for 30 min. The level of intracellular peroxides was determined by flow cytometry as green fluorescence generated by the oxidation of the substrate.
Northern blot analysis. Total RNA from untreated and treated DS19 cells was isolated by the acid guanidium thiocyanate phenol chloroform method (Chomczynski & Sacchi, 1987). Twenty micrograms of RNA were loaded onto a 1·2% agarose–formaldehyde gel, electrophoresed and transferred to a Zeta-probe filter (Bio-Rad, Richmond, CA, USA). A mouse β-globin cDNA (pCR1), a rat erythroid-specific δ-aminolaevulinate synthase (ALAS-E) cDNA (pREAL1), a rat δ-aminolaevulinate dehydratase (ALAD) cDNA (pALAD1) and a rat porphobilinogen deaminase (PBGD) cDNA (p44SB1) were used for these analyses (Fujita et al, 1991). Human ribosomal DNA was used as an internal control. Hybridization and washing of the filters were performed as described previously (Nagai et al, 1998).
Western blot analysis. Nuclear extracts were prepared from 1 × 107 cells, according to a previously described method (Lassar et al, 1991). Ten micrograms of nuclear extracts were separated electrophoretically using a 10% polyacrylamide gel. Immunoblotting and detection by enhanced chemiluminescence (ECL) were performed as illustrated previously (Nagai et al, 1997). Haemoglobin was quantified using a rabbit anti-mouse haemoglobin antiserum (CAPPEL/Organon Teknica, Durham, NC, USA). A goat anti-rabbit immunoglobulin (Ig)G, which had been coupled with horseradish peroxidase, was used as a secondary antibody in ECL assay. Anti-p44/42, anti-phospho p44/42, anti-p38 and anti-phospho p38 rabbit polyclonal antibodies were purchased from Cell Signalling Technology, Beverly, MA, USA. Mouse antiglyceraldehyde-3-phosphate dehydrogenase (GAPDH) monoclonal antibody was purchased from Chemicon International, Temecula, CA, USA, and used as an internal control.
HU increased β-globin and ALAS-E mRNA levels in DS19 cells
We first examined the effect of HU on the haemoglobinization of DS19 cells using dianisidine staining, and on mRNA levels of β-globin and ALAS-E by Northern blot analysis. HU treatment increased the percentage of dianisidine-positive cells from 0% to 45%, at 96 h, although the percentage of HU-induced dianisidine-positive cells was less than that induced by DMSO (c. 70%), a potent inducer of MEL cell differentiation (Fig 1A). Both β-globin and ALAS-E mRNA levels were upregulated by HU as well as by DMSO treatment (Fig 1B). HU is known to induce erythroid cell differentiation of human erythroleukaemia cells such as K562 and HEL (Xu & Zimmer, 1998; Zhang et al, 2001). Our findings suggested that DS19 cells undergo erythroid differentiation by HU treatment, similarly to that seen in other human erythroleukaemia cell lines (Xu & Zimmer, 1998; Zhang et al, 2001).
L-Acetyl-N-cysteine-inhibited HU-induced differentiation of DS19 cells
It has been shown that HU incurs oxidative damage on haemoglobin in human red blood cells, suggesting that HU induces the production of intracellular ROI (Iyamu et al, 2001). In fact, the level of intracellular peroxides of HU-treated DS19 cells was significantly higher than that of untreated cells (Fig 2). As ROI influence many important biological functions, we hypothesized that oxidative stress resulting from HU treatment might be associated with the induction of erythroid differentiation. To examine this possibility, we treated DS19 cells with l-acetyl-N-cysteine (LNAC), a potent reducing reagent. While LNAC treatment alone had no effect on the percentage of dianisidine-positive cells (Fig 3A) and mRNA levels of β-globin and ALAS-E (data not shown), combined treatment of DS19 cells with HU and LNAC resulted in a significant reduction of the HU-mediated increase in the percentage of dianisidine-positive cells, β-globin and ALAS-E mRNA levels (Fig 3A and B). In contrast, LNAC only slightly inhibited the DMSO-mediated induction of erythroid differentiation (Fig 3A).
Hyperoxia induces erythroid differentiation of DS19 cells
We next carried out an experiment to determine whether oxidative stress mediated by hyperoxia might induce erythroid differentiation of DS19 cells. Treatment of DS19 cells with hyperoxia (50% O2) resulted in an increase in the percentage of dianisidine-positive cells from 0% to 16%, at 72 h. Consistent with these findings, mRNA and protein levels of β-globin, and mRNA levels of haem synthetic pathway enzymes, i.e. ALAS-E, the rate-limiting enzyme of the haem synthetic pathway, ALAD and PBGD, were all increased 72 h after hyperoxia treatment (Fig 4 and Table I). Hyperoxia treatment increased intracellular peroxides to the same level as that induced by HU treatment (data not shown). These results indicate that erythroid differentiation of DS19 cells could also be induced by another oxidative stress, hyperoxia. These results suggest that oxidative stress is involved both in the HU- and hyperoxia-induced erythroid differentiation, while it plays a minor role in DMSO-mediated erythroid differentiation.
Table I. Changes in mRNA levels of the haem synthetic pathway enzymes following hyperoxic treatment.
Fold increase of hyperoxia over normoxia
DS19 cells were incubated for 72 h, under 21% and 50% O2 concentrations, for normoxic and hyperoxic treatment respectively. Northern blot analysis was performed as described in Materials and methods. mRNA concentrations were quantified using LKB Ultrascan XL enhanced laser densitometer (Pharmacia Piscataway, NJ, USA). The values are shown as means ± SD (n = 3).
× 3·00 ± 0·22
× 1·78 ± 0·10
× 2·51 ± 0·33
HU-induced differentiation of DS19 cells is independent of MAP kinase activation
MAPKs play an important role in the regulation of cell survival, proliferation and differentiation. Previous studies showed that phosphorylation of Erk1 was decreased, while that of p38 was increased, after HU treatment of K562 cells. To determine whether a change in MAP kinase activity might be associated with HU-induced MEL cell differentiation, Western blot analysis of Erk1 and p38 proteins was performed using anti-phospho p44/42 and anti-phospho p38 antibodies. The results showed that the levels of phosphorylated p44/42 and phosphorylated p38 were similar between untreated control and HU-treated cells (Fig 5). LNAC had also no effect on the phoshorylation of p44/42 and p38 both in untreated and in HU-treated DS19 cells (Fig 5). Consistent with these results, the addition of PD98059 or U0126, which are MAP kinase signalling inhibitors, produced no change in the percentage of dianisidine-positive cells in HU-treated cells (data not shown). Therefore, we concluded that HU induced the erythroid differentiation of MEL cells by a mechanism independent of RAS-MAP kinase signalling.
HU was previously proved to induce differentiation of K562 and HEL cells into haemoglobin producing cells. However, most human leukaemic cell lines, including K562, already exhibit substantial amounts of β-globin mRNA before the addition of an inducer to the culture medium (Nagai et al, 1997). In contrast, the levels of mRNAs of β-globin and ALAS-E are low in undifferentiated MEL cells, but are markedly increased in differentiated cells following DMSO treatment (Fujita et al, 1991). We, therefore, decided to use DS19, which is one of the subclones of MEL cells, for the examination of the mechanism of HU-mediated induction of erythroid differentiation.
Our results showed that a potent reducing reagent, LNAC, inhibited HU-induced erythroid differentiation (Fig 3), and that treatment of DS19 cells with hyperoxia increased gene expression of β-globin and haem synthetic pathway enzymes (Fig 4, Table I). These results suggest that oxidative stress is involved in HU-induced erythroid differentiation in DS19 cells. Although it is yet unclear how oxidative stress induces upregulation of erythroid-specific genes, it is possible that ROI may upregulate functions of certain erythroid transcription factors or stimulate other intracellular signalling pathways. It should also be noted that oxidative stress has been demonstrated to influence the activities of various signalling pathways and transcription factors such as nuclear factor (NF)-κB (Schreck et al, 1991; Nakamura et al, 1997).
As oxidative stress plays an important role in the action of HU, increases in the levels of intracellular antioxidants, such as glutathione (GSH), might counteract the HU action. In fact, it has been shown that an increased level of GSH increases resistance to anticancer drugs such as cisplatin, doxorubicin, cytosine arabinoside and daunorubicin, which produce intracellular ROI (Iida et al, 1999; Takemura et al, 2001). Although it is not clear whether GSH is involved in drug resistance to HU, it is possible that an increased concentration of intracellular antioxidant levels may protect cells from HU-induced oxidative damage.
In conclusion, we have demonstrated that oxidative stress plays a critical role in the erythroid differentiation of MEL cells induced by HU. These results also suggest that it may be possible to find or design a new reagent with an oxidant property similar to HU that can induce erythroid differentiation, and thus may be useful in the treatment of disorders of erythroid maturation and/or haemoglobin synthesis.
This work was supported in part by grants-in-aid from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and USPHS grant DK-32890. We are grateful to Drs Hiroyoshi Fujita and Yasuo Ohshima for their helpful discussions. We also thank Ms M. Nakamura for her technical assistance, Ms T. Hirabayashi for reviewing the manuscript and Ms E. Yamakawa for the preparation of the manuscript.