Arsenic Trioxide in APL

Arsenic trioxide-induced apoptosis and differentiation are associated respectively with mitochondrial transmembrane potential collapse and retinoic acid signaling pathways in acute promyelocytic leukemia

Article metrics


Recent studies showed that arsenic trioxide (As2O3) could induce apoptosis and partial differentiation of leukemic promyelocytes. Here, we addressed the possible mechanisms underlying these two different effects. 1.0 μM As2O3-induced apoptosis was associated with condensation of the mitochondrial matrix, disruption of mitochondrial transmembrane potentials (ΔΨm) and activation of caspase-3 in acute promyelocytic leukemia (APL) cells regardless of their sensitivity to all-trans retinoic acid (ATRA). All these effects were inhibited by dithiothreitol (DTT) and enhanced by buthionine sulfoximine (BSO). Furthermore, BSO could also render HL60 and U937 cells, which had the higher cellular catalase activity, sensitive to As2O3-induced apoptosis. Surprisingly, 1.0 μM As2O3 did not induce the ΔΨm collapse and apoptosis, while 0.1 μM As2O3 induced partial differentiation of fresh BM cells from a de novo APL patient. In this study, we also showed that 0.2 mM DTT did not block low-dose As2O3-induced NB4 cell differentiation, and 0.10.5 μM As2O3 did not induce differentiation of ATRA-resistant NB4-derived sublines, which were confirmed by cytomorphology, expression of CD11b, CD33 and CD14 as well as NBT reduction. Another interesting finding was that 0.10.5 μM As2O3 could also induce differentiation-related changes in ATRA-sensitive HL60 cells. However, the differentiation-inducing effect could not be seen in ATRA-resistant HL60 sublines with RARα mutation. Moreover, low-dose As2O3 and ATRA yielded similar gene expression profiles in APL cells. These results encouraged us to hypothesize that As2O3 induces APL cell differentiation through direct or indirect activation of retinoic acid receptor-related signaling pathway(s), while ΔΨm collapse is the common mechanism of As2O3-induced apoptosis.


Arsenic trioxide (As2O3), an effective drug for the treatment of acute promyelocytic leukemia (APL),1234 was shown to exert dose-dependent dual effects in APL cells, ie triggering apoptosis and inducing partial differentiation.5678 Meanwhile, it was also suggested that these effects were associated with As2O3-induced modulation/degradation of PML-RARα protein,589101112 an APL-specific fusion protein resulting from the reciprocal chromosome translocation t(15;17)(q22;q21).1314 However, high-dose (1.02.0 μM) As2O3 could also induce in vitro growth inhibition and/or apoptosis of malignant lymphocytes, myeloma cells, and some solid tumor cell lines such as oesophageal carcinoma and neuroblastoma.15161718192021 Obviously, As2O3-induced apoptosis may involve some common mechanisms. Recently, it was suggested that As2O3-induced collapse of mitochondrial transmembrane potential (ΔΨm) via a direct effect on the mitochondrial permeability transition pore (PT) was a pivotal event in As2O3-induced apoptosis, and sulfhydryl (−SH) groups could be important targets of As2O3 for induction of ΔΨm collapse and apoptosis.1822 However, a recent report showed that GSH reducer dithiothreitol (DTT) effectively enhanced As2O3-induced apoptosis in APL cell line NB4 cells,23 which was inconsistent with the observation in malignant lymphocytes.18 Therefore, it is necessary to evaluate further the role of −SH groups in As2O3-induced apoptosis.

On the other hand, the mechanisms of low-dose (0.10.5 μM) As2O3-induced differentiation, which was observed selectively in APL cells and is possibly the most important mechanism for the remission induction of APL, remain obscure. Therefore, its illustration is important to understand the clinical potential of As2O3 in the treatment of cancers. In the present work, a series of cell biological and biochemical studies were conducted on APL-derived cell lines and HL60 cells with and without the mutation of retinoic acid receptor-α (RARα). Our results urged us to propose that −SH oxidation-related ΔΨm collapse and caspase-3 activation contributes to As2O3-induced apoptosis of APL cells, while As2O3-induced differentiation might be mediated through direct or indirect activation of RARα-related signal pathway(s).

Materials and methods


As2O3, ATRA, dithiothreitol (DTT, a widely used disulfide bond-reducing agent) and buthionine sulfoximine (BSO, a selective inhibitor of γ-glutamylcysteine synthetase) were purchased from Sigma Chemical Company (St Louis, MO, USA). As2O3 was dissolved in small amounts of 1.0 M NaOH, then diluted to 5.0 mM with phosphate-buffered saline (PBS) as stock solutions.

Cell culture, viability and morphology

Fresh bone marrow (BM) cells from two cases of de novo APL patients, with informed consent, were separated by centrifugation on Ficoll's solution. In these two samples, over 85% of cells were leukemic promyelocytes with chromosome translocation t(15;17) and PML-RARα transcripts revealed by cytogenetics and RT-PCR analysis. Both patients were treated with ATRA and achieved clinical remission. Cell lines used in this study included ATRA-sensitive APL cell line (NB4), ATRA-resistant NB4-derived sublines (MR2, R1, R2) and other myeloid leukemia cell lines (HL60 and U937). In addition, HL60 sublines with the dominant-negative mutation of RAR-α (HL60Res), kindly provided by R Gallagher,24 were also used as in vitro models. All cells were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS; Gibco BRL, Grand Island, NY, USA), glutamine and antibiotics (100 IU/ml penicillin and 100 μg/ml streptomycin) in a humidified atmosphere of 95% air/5% CO2 at 37°C. To avoid possible effects of cell density on cell growth and survival, cells were maintained at less than 5 × 105 cells/ml with daily adjusting cell density through adding fresh medium and corresponding concentrations of compounds. Cell viability was estimated by trypan-blue exclusion. Morphology was determined with Wright's staining of cells centrifuged on to slides by cytospin (Shandon, Runcorn, UK; 500 r.p.m., 4 min).

DNA content analysis by flow cytometry

Cells were collected, rinsed and injected into cold 70% ethanol, then kept overnight at 4°C. Subsequently, cells were rinsed with PBS, treated with Tris-HCl buffer (pH 7.4) supplemented with 1% RNase and stained with propidium iodide (PI, 50 μg/ml). Distribution of cells with different DNA contents was determined by flow cytometry (Becton Dickinson, San Jose, CA, USA). Ten thousand cells were analyzed in every sample. All data were collected, stored and analyzed by LYSIS II software (Becton Dickinson).

Gel electrophoresis of genomic DNA

A total of 106 cells were lysed with 2 ml lysis buffer (50 mM Tris-HCl pH 8, 20 mM EDTA pH 8 and 2% SDS) overnight at 37°C. Then, 0.8 ml saturated NaCl solution was added and centrifuged twice to remove proteins. RNase was added to supernatants at a final concentration of 20 μg/ml and incubated at 37°C for 15 min. DNA was precipitated by adding two volumes of 70% ethanol and electrophoresed overnight on a 1.25% agarose gel.

Determination of mitochondrial transmembrane potentials (ΔΨm) on flow cytometry

After washing twice with PBS, about 106 cells were incubated (37°C, 30 min) with 10 μg/ml rhodamine 123 (Rh123), a cationic lipophilic fluorochrome taken up by mitochondria in proportion to the ΔΨm. Then, 50 μg/ml PI, a membrane-impermeable DNA-binding dye, was added to cells. Fluorescent intensities were determined on flow cytometry (Becton Dickinson). Ten thousand cells were analyzed in every sample. All data were collected, stored and analyzed by LYSIS II software (Becton Dickinson).

Transmission electron microscopy (TEM)

Cells were fixed in suspension with 2.5% glutaraldehyde at 4°C overnight, and post-fixed for 1 h on ice with 1% OsO4. Then, cells were dehydrated, infiltrated and embedded according to the usual methods. The sections were stained for 20 min with 2% uranyl acetate, followed by lead citrate and viewed in a Hitachi H-500 electron microscope (Hitachi, Tokyo, Japan).

Determination of cellular caspase-3 activity

Cellular caspase-3 activity was measured according to the Apoalert CPP32 Fluo-kit provided by Clontech (Palo Alto, CA, USA), in which caspase-3 activity was presented with fluorescent intensity (at 505 nm) of 7-amino-4-trifluoromethyl coumarin (AFC) after cleavage from the peptide substrate DEVD-AFC on luminescence spectrometer (Type LS 50B; Perkin-Elmer, Wellesley, MA, USA).

Determination of cellular glutathione (GSH) contents and anti-oxidant enzyme activity

A total of 107 cells were harvested, washed with ice-cold PBS and sonicated in the lysis buffer for 20 min. The homogenates were centrifuged at 10 000 g for 20 min at 4°C and the supernatants were used for the determination of cellular GSH contents and enzyme activity, which was performed with kits provided by the manufacturer (Nanjing Bioengineering Institution, Nanjing, PR China). Briefly, cellular glutathione S-transferase (GST) and glutathione peroxidase (Gpx) activities were determined respectively according to the reduction of GSH contents in the following reaction systems: GSH + 1-chlorin-2,4-dinitrobenzene [C6H3(NO2)2Cl] GST → glutathione dinitrobenzene [C6H3 (NO2)2GS] + HCl, and H2O2 + 2GSH Gpx → oxidized GSH (GSSG) + 2H2O. An activity unit of GST or Gpx was defined as decreasing 1.0 μM GSH in 1 min. GSH contents were measured according to OD420 nm values of a yellow compound that was produced by the reaction between GSH and disulfurate dinitrobenzyl acid. The catalase activity was measured according to its ability to decompose H2O2 into H2O, and an activity unit of catalase was regarded as decomposing 1.0 μM H2O2 in 1 s. A nitrite unit of superoxide dismutase (SOD) activity was defined as 50% of SOD inhibition rate in 1 ml reaction solution, in which SOD specifically inhibited the activity of superoxide (O2) produced by the reaction of xanthine and xanthine oxidase, while O2 oxidized hydroxyamine into nitrite, the latter presenting purple color in the presence of the relative colorant. Cellular protein concentrations were determined by the Bradford method with bovine serum albumin as a standard.

Cell differentiation assays

Cell differentiation antigens CD11b, CD14 and CD33 were determined on flow cytometry (Coulter EPICS XL, South San Francisco, CA, USA). All fluorochrome-labeled monoclonal antibodies including anti-human CD11b/FITC, CD33/PE, CD14/FITC antibody and their corresponding isotypic control (IgG1/FITC, IgG1/PE, IgG2a/FITC) were purchased from Coulter-Immunotech (Paris, France). In addition, NBT reduction and cytochemical reactions were done as previously described.5678

Immunofluorescence analysis for PML/PML-RARα proteins

Cells were centrifuged on to slides as described above and rapidly air-dried. Immunofluorescence staining of the N- terminal region of PML was performed using a high-quality anti-PML N-terminal antiserum,5 kindly provided by Dr Naoe (Branch Hospital, School of Medicine, Nagoya University, Japan).

Reverse transcriptase polymerase chain reaction (RT-PCR)

Expression of some genes regulated by ATRA, including CD52, Bfl-1, RIG-E, preA-PAI-2, transglutaminase (TG), PKC type β I (PKC β I), SUMO-1 and RARβ were measured by RT-PCR in As2O3- and ATRA-treated NB4 cells. In brief, 4 μg total RNAs, isolated from NB4 cells by using Trizol reagents (GIBCO BRL), were reverse transcribed with random primers with Superscript II RT (GIBCO BRL). PCR reactions were performed in 50 μl solutions containing 1 μl cDNA, 1 μM primers, 200 μM dNTPs, 2.5 units of Taq DNA polymerase, and 1.5 mM MgCl2, after denaturation (94°C, 5 min) as follows: 94°C, 1 min, 56°C, 1 min, 72°C 2.5 min, 30 cycles, followed by extension at 72°C for 10 min. All the PCR products were electrophoresed on a 2% agarose gel containing 0.5 μg/ml ethidium bromide in 1× TAE buffer and visualized on a UVP transilluminator. G3PDH was used as an internal control for each sample.


High-dose As2O3-induced ΔΨm collapse and apoptosis were associated with thiols in myeloid leukemic cell lines

As shown in Figure 1, with double staining of PI and Rh123, most untreated NB4 cells presented negative PI (intact plasma membrane) and high Rh123 staining (normal ΔΨm). With treatment of 1.0 μM As2O3 for 1 to 3 days, NB4 cells with negative PI and low Rh123 staining (disrupted ΔΨm) presented a time-dependent increase. These changes appeared in parallel with 1.0 μM As2O3-induced apoptosis in NB4 cells, which was confirmed by the decreased cell viability, typical apoptotic morphology, increased sub-G1 cells and DNA ladder pattern in agarose gel (Figure 1 and data not shown). TEM observation showed that these apoptotic cells presented dense mitochondrial matrix and condensed chromatin and fragmental nuclei (Figure 2a). Furthermore, the activity of caspase-3 was significantly increased 1224 h after the treatment with 1.0 μM, but not with 0.1 μM As2O3 (Figure 2b).

Figure 1

 Effects of As2O3 and DTT or BSO on the ΔΨm (upper panels) and histogram-related nuclear DNA contents (lower panels) of NB4 cells. NB4 cells were treated with 1.0 μM As2O3 in combination with 0.2 mM DTT or 1.0 mM BSO for 48 h. The percentages of PI Rh123low cells and sub-G1 cells are shown in the corresponding panels.

Figure 2

 (a) TEM observation (×10 000) of ultra-microstructures of NB4 cells with and without treatment of 1.0 μM As2O3 for 48 h. N indicates nuclei. (b) The changes in caspase-3 activity of NB4 cells treated with or without 0.1 μM/1.0 μM As2O3 and/or 0.2 mM DTT for 12 to 48 h. The cellular caspase-3 activity was reflected in fluorescence intensity at 505 nm of 7-amino-4-trifluoromethyl coumarin (AFC) after cleavage from the peptide substrate DEVD-AFC. The data are from one of three independent but concordant experiments. (c) The percentages of PIRh123low cells and sub-G1 cells of MR2, U937 and HL60 cells treated with 1.0 μM As2O3 and/or 0.2 mM DTT or 1.0 mM BSO for 48 h.

As seen in As2O3-treated malignant lymphocytes,18 0.2 mM DTT blocked, while 1.0 mM BSO enhanced As2O3-induced ΔΨm collapse, caspase-3 activation and apoptosis in NB4 cells (Figure 1 and Figure 2b). A similar phenomenon also appeared in 1.0 μM As2O3-treated MR2, R1 and R2 cells, which were sensitive to As2O3-induced apoptosis despite their resistance to ATRA (Figure 2c). More interestingly, HL60 and U937 cells did not present ΔΨm collapse and apoptosis under treatment of 1.0 μM As2O3 alone, but these effects could be seen with simultaneous treatment of 1.0 μM As2O3 and 1.0 mM BSO (Figure 2c).

As2O3-insensitive cells showed high level of cellular GSH contents or catalase activity

As mentioned above and previously,58111718 HL-60, U937 and Jurkat cell line, a T lymphocytic lineage from T-ALL, had weak sensitivity to 1.0 μM As2O3-induced apoptosis, which was enhanced by BSO. These results supported the concept that intracellular GSH level could be associated with the sensitivity of cells to As2O3-induced apoptosis, which had been reported by some groups.25 However, only three cell lines were involved in the report. Therefore, we detected cellular GSH contents and the activity of four anti-oxidant enzymes of eight malignant hematopoietic cell lines. The results showed that Gpx, GST and SOD activities showed no significant differences among these cells. However, cells sensitive to As2O3-induced apoptosis had lower GSH level and catalase activity. Contrarily, Jurkat cells had the higher GSH level and HL-60 and U937 cells had significantly higher catalase activities (Figure 3).

Figure 3

 Constitutive levels of cellular GSH content and the anti-oxidant enzyme activities in eight cell lines. In these cell lines, RPMI8226 and Namalwa cells, which are derived respectively from Burkitt's lymphoma and multiple lymphoma, are sensitive to As2O3-induced apoptosis, like NB4, MR2 and R2 cells. The other three cell lines had weaker responses to As2O3-induced apoptosis.58111718

Low-dose As2O3-induced cell differentiation also occurred in ATRA-sensitive HL60, but not in ATRA-resistant NB4-derived sublines and HL60Res

NB4 cells were treated simultaneously with 0.1 μM As2O3 and 0.2 mM DTT for 7 to 14 days. Meanwhile, culture media were changed every day and corresponding concentrations of drugs were supplemented. As reported previously,5 0.1 μM As2O3 induced partial differentiation of NB4 cells, as manifested by morphologic changes, increased expression of differentiation antigen CD11b without the significant changes of NBT reduction and CD33/CD14 expression. However, these differentiation-related alterations were not blocked by 0.2 mM DTT, and they were not present in ATRA-resistant NB4-derived sublines MR2, R1 and R2 with 0.10.5 μM As2O3 treatment for more than 10 days (Figure 4 and data not shown). These findings encouraged us to hypothesize that As2O3-induced differentiation might be associated with retinoic acid (RA) and/or its receptor complex (RARα/RXR) signaling pathways. In order to confirm this hypothesis, we tried to perform the experiment in cell culture medium depleted of ATRA, prepared with charcoal-treated FBS. Unfortunately, cells could not maintain the survival for the sufficient length of time (714 days) in our experimental conditions. We then speculated that if As2O3-induced differentiation was associated with RARα/RXR signaling pathways As2O3 should also be able to induce differentiation of HL60 cells, which are very sensitive to the differentiation-inducing effect of ATRA. Surprisingly, 0.10.5 μM As2O3 could indeed induce HL60 cells to present partial differentiation-related morphological changes (Figure 5a, left panel), and immunophenotypic changes such as elevated CD11b expression (Figure 5b, upper panel). More importantly, unlike low-dose As2O3-induced NB4 cell differentiation, differentiated HL60 cells induced by 0.10.5 μM As2O3 exhibited increased CD14 expression and NBT reduction as well as decreased CD33 expression to different degrees (Figure 5c and e). Then, HL60,Res a cell line with resistance to ATRA-induced differentiation due to RARα mutation, was treated with 0.10.5 μM As2O3 over a relatively long time. The results showed that these concentrations of As2O3 did not induce cell death or partial differentiation-related changes of HL60Res cells including morphology (Figure 5a, right panel), expression of CD11b (Figure 5b, lower panel), CD33 and CD14 (Figure 5d) and NBT reduction (Figure 5e).

Figure 4

 Cell morphology ((a) for NB4 cells and (b) for MR2 cells) and the percentage of CD11b-positive cells (c) of NB4 and MR2 cells undergoing treatment with 0.1 μM As2O3 and/or 0.2 mM DTT for 10 days.

Figure 5

 0.10.5 μM of As2O3 induces partial differentiation in HL60 but not in HL60Res cells, as confirmed by cell morphology (a), expression of CD11b (b) blue and purple lines respectively representing negative control and CD11b) and CD14/CD33 ((c) for HL60, (d) for HL60Res cells) and NBT reduction (e). Treatment time (day, d) and drug doses are shown in the figure. All data were confirmed by three independent experiments.

As2O3 and ATRA at low dose yielded similar gene expression profiles in APL cells

A study in our group showed that a large number of genes could be regulated during the APL cell differentiation induced by pharmacological concentrations of ATRA.26 To further test the hypothesis that mechanisms of As2O3-induced differentiation could be related to RA and/or RARα/RXR signaling pathways, the expression of a group of selected ATRA-responsive genes was measured by RT-PCR in As2O3-treated NB4 cells. Cells treated with 10−8 M and 10−9 M ATRA were used as controls. As shown in Figure 6, both 0.1 μM As2O3 and low concentrations of ATRA yielded similar mRNA expression profiles in most genes tested, including the upregulation of CD52 and Bf1-1, the downregulation of RARβ, and no modulation of RIG-E and Pre A-PAI-2. In addition, synergistic effects were observed in the regulation of PKCβ-1 and SUMO-1 when cells were under the simultaneous treatment of 0.1 μM As2O3 and 10−8M ATRA.

Figure 6

 Effects of As2O3 and/or ATRA on gene expression of NB4 cells. Abbreviations for genes are shown in the Materials and methods.

Effects of DTT on PML/PML-RARα protein degradation induced by high or low concentrations of As2O3

To examine the possible role of −SH groups in As2O3-induced modulation/degradation of PML/PML-RARα proteins, immunofluorescence analysis with anti-PML antibody was performed in NB4 cells treated simultaneously with DTT and As2O3. 0.2 mM DTT treatment alone did not significantly modulate subcellular localization of PML/PML-RARα proteins, but it visibly inhibited the rapid effects of 1.0 μM As2O3 on the degradation of PML/PML-RARα proteins. Also, 0.2 mM DTT treatment over 10 days seemed to have some inhibitory effect, to a lesser extent, on the degradation of PML/PML-RARα proteins induced by 0.1 μM As2O3 (Figure 7).

Figure 7

 Immunofluorescence analysis of subcellular localization of PML/PML-RARα proteins in NB4 cells after treatment with 1.0 μM (a) or 0.1 μM (b) As2O3 and/or 0.2 mM DTT for the indicated times.

1.0 μM As2O3 did not induce apoptosis, while 0.1 μM As2O3 induced cell differentiation in fresh cells from a de novo APL patient

We previously described that dose-dependent dual effects of As2O3 were also seen in fresh APL cells, but the responses of cells from different patients exhibited significant heterogeneity.5 In line with this preliminary observation, the experiment performed here on fresh cells from two de novo APL patients showed quite different dose-responses. Cells from one patient, like NB4 cells, experienced significant ΔΨm collapse and apoptosis under 1.0 μM As2O3, which could be abrogated by 0.2 mM DTT (data not shown). Surprisingly, in fresh BM cells from another de novo APL patient, 1.0 μM As2O3 did not induce ΔΨm collapse and apoptosis (Figure 8a). However, a partial differentiation was induced by 0.1 μM As2O3 which was corroborated by morphology (data not shown) and CD11b expression (Figure 8b).

Figure 8

 Effects of As2O3 on fresh BM cells from a de novo APL patient. (a) Cells were treated with or without 1.0 μM As2O3 for 3 days and the ΔΨm was measured as described above. (b) Cells were treated with 0.1 μM As2O3 for 7 days and CD11b was tested by flow cytometry.


In this work, 1.0 μM As2O3 markedly induced ΔΨm collapse, ultracondensation of the mitochondrial matrix and activation of caspase-3 in ATRA-sensitive and ATRA-resistant APL cell lines, which were coupled with As2O3-induced apoptosis. Furthermore, DTT substantially blocked, while BSO enhanced, As2O3-induced ΔΨm collapse and apoptosis of APL cells. Interestingly, BSO could also render HL60 and U937 cells sensitive to As2O3-induced apoptosis. These observations further supported our previous conclusion that ΔΨm collapse was an important mechanism in As2O3-induced apoptosis, and −SH groups might be an important chemosensor of As2O3 for the induction of ΔΨm collapse and apoptosis.18

Although mitochondria were involved with As2O3-induced apoptosis, 1.0 μM As2O3 did not induce ΔΨm collapse and apoptosis in HL60, U937 cells and Jurkat cells.58111718 It is thus of interest to understand how these cells escape the death pathways in order to explore clinical effectiveness of As2O3 in cancers other than APL. We showed that HL60, U937 and Jurkat cells resistant to As2O3 had either high cellular catalase activity or high GSH level, while their sensitivities to As2O3 were restored due to GSH depletion. These facts suggested that catalase activity was also very important for the sensitivity of cells to As2O3 in addition to intracellular GSH content.25 Of note, a recent report showed that 1.0 μM As2O3 could induce apoptosis of HL60 cells.27 We believe that the most likely reason for the discrepancy between this datum and most reports.58111718 was the variation of cell lines with regard to some critical genetic or biochemical regulation, eg the oxidation and anti-oxidation capacities. Also, As2O3 was widely shown to trigger reorganization and degradation of the PML/PML-RARα proteins, which could be related to the striking increase of SUMO-1-PML conjugates induced by As2O3.28 Here, we showed that DTT inhibited As2O3-induced degradation of PML/PML-RARα proteins, in parallel with the inhibition of As2O3-induced apoptosis. Moreover, trivalent antimonial-provoked apoptosis was also coupled with the degradation of PML-RARα proteins and the reorganization of the PML nuclear bodies (NBs) in APL cells.29 In addition, As2O3 did not induce apoptosis of primary APL blasts with the expression of PLZF-RARα fusion proteins.30 Based on these observations, further consideration that the modulation/ degradation of PML/PML-RAR α influences the sensitivity of APL cells to As2O3-induced apoptosis, is deserved, although As2O3 has a wider spectrum in terms of apoptosis induction.

We previously suggested that low-dose As2O3-induced cell differentiation could be an even more important factor for the remission induction of APL.5 In this study, DTT did not block 0.1 μM As2O3-induced partial differentiation in NB4 cells. Interestingly, fresh cells from one de novo APL patient were resistant to 1.0 μM As2O3-induced apoptosis but sensitive to 0.1 μM As2O3-induced differentiation. Obviously, independent mechanisms were involved in As2O3-induced apoptosis and differentiation, which was also supported by data from an arsenic-resistant NB4 subline (NB4-AsR) which failed to undergo apoptosis, but maintained partial differentiation response to As2O3.7 However, the mechanisms of As2O3-induced differentiation are elusive so far. Previously, we suggested that like ATRA, As2O3-induced differentiation of APL blasts was associated with the degradation of the PML-RARα proteins.5 However, we revealed in this study that As2O3 could also induced partial differentiation of HL60 cells where no dominant negative chimeric receptor is present, and DTT did not block As2O3-induced differentiation but partly inhibited As2O3-induced PML-RARα modulation in NB4 cells. These facts indicated that the modulation/degradation of PML-RARα proteins is not the only mechanism of As2O3-induced differentiation. In fact, the role of PML-RARα in ATRA-induced differentiation is controversial.31

In our study, 0.1 μM As2O3 did not induce differentiation in ATRA-resistant APL cell lines. Although 0.10.5 μM As2O3 could induce differentiation in ATRA-sensitive HL60 cells, this effect disappeared in HL60Res cells with RARα mutation. Finally, 0.1 μM As2O3 could regulate gene expression in a similar pattern to physiological concentrations (10−810−9M) of ATRA, although the similarity of gene regulation between 0.1 μM As2O3 and ATRA could also be associated with modulation/degradation of PML-RARα proteins induced by them. These important findings prompted us to speculate that low-dose As2O3 also provides direct or indirect stimulation to wild-type RARα/RXR signaling pathways, especially in the presence of degradation of PML-RARα proteins, which blocked differentiation and apoptosis of APL cells. Figure 9 represents our working hypothesis for the mechanism of action of As2O3 in remission induction of APL, although some aspects remain to be confirmed at the molecular level: As2O3 induces remission of APL patients via two independent pathways, that is, triggering partial differentiation and inducing apoptosis. High-dose As2O3-induced apoptosis involves mainly the sulfhydryl group-related ΔΨm collapse due to the opening of PT pore, which triggers the release of pre-apoptotic factor from mitochondria to cytoplasma, followed by caspase activation and degradation of specific substrates. On the other hand, low-dose As2O3-induced differentiation might be mediated directly or indirectly by RARα-related signaling pathway(s), in which nuclear receptor coactivators/ corepressors and histone acetylation are involved.

Figure 9

 A hypothesis for mechanisms of As2O3 in the treatment of APL. As2O3 may induce remission in APL patients via inducing partial differentiation and triggering apoptosis. High-dose (12 μM) As2O3 induces APL cell apoptosis through direct or indirect disruption of mitochondrial transmembrane potentials due to opening of permeability transmission (PT) pores. This leads to release of pro-apoptosis factors such as cytochrome C and apoptosis-inducing factor (AIF) from mitochondria to cytoplasma, where they activate caspase and degradate specific substrates. Low-dose (0.10.5 μM) As2O3, in contrast, induces partial differentiation of APL cells through direct or indirect activation of RARα/RXR pathways, in which coactivator (CoAct) and histone acetylase (HAC) are involved. Then, activated RARα/RXR binds to retinoic acid response elements (RARE) and induces the expression of differentiation-related genes. →, stimulation, inhibition.


  1. 1

    Sun HD, Ma L, Hu XC, Zhang TD . Ai-Lin I treated 32 cases of acute promyelocytic leukemia Chin J Integrat of Chinese and Western Med 1992 12: 170–171

  2. 2

    Zhang P, Wang SY, Hu XH . Arsenic trioxide treated 72 cases of acute promyelocytic leukemia Chin J Hematol 1996 17: 58–60

  3. 3

    Shen ZX, Chen GQ, Ni JH, Li XS, Xiong SM, Qin QY, Zhu J, Tang W, Sun GL, Yang KQ, Chen Y, Zhou C, Fang ZW, Wang YT, Ma J, Zhang P, Zhang TD, Chen SJ, Chen Z, Wang ZY . Use of arsenic trioxide (As2O3) in the treatment of acute promyelocytic leukemia (APL): II clinical efficacy and pharmacokinetics in relapsed patients Blood 1997 89: 3354–3360

  4. 4

    Soignet SL, Maslak P, Wang ZG, Jhanwar S, Calleja E, Dardashti LJ, Corso D, deBlasio A, Gabrilove J, Scheinberg DA, Pandolfi PP, Warrell RP Jr . Complete remission after treatment of acute promyelocytic leukemia with arsenic trioxide New Engl J Med 1998 339: 1341–1348

  5. 5

    Chen GQ, Shi XG, Tang W, Xiong SM, Zhu J, Cai X, Han ZG, Ni JH, Shi GY, Jia PM, Liu MM, He KL, Niu C, Ma J, Zhang P, Zhang TD, Paul P, Naoe T, Kitamura K, Miller W, Waxman S, Wang ZY, de The H, Chen SJ, Chen Z . Use of arsenic trioxide (As2O3) in the treatment of acute promyelocytic leukemia (APL): I. As2O3 exerts dose-dependent dual effects on APL cells Blood 1997 89: 3345–3353

  6. 6

    Kitamura K, Yoshida H, Ohno R, Naoe T . Toxic effects of arsenic and other metal ions on APL cells Intl J Hematol 1997 65: 179–185

  7. 7

    Gianni M, Koken MHM, Chelbi-Alix MK, Benoit G, Lanotte M, Chen Z, de The H . Combined arsenic and retinoic acid treatment enhances differentiation and apoptosis in arsenic-resistant NB4 cells Blood 1998 91: 4300–4310

  8. 8

    Chen GQ, Zhu J, Shi XG, Ni JH, Zhong HJ, Si GY, Jin XL, Tang W, Li XS, Xong SM, Shen, ZX, Sun GL, Ma J, Zhang P, Zhang TD, Gazin C, Naoe T, Chen SJ, Wang ZY, Chen Z . In vitro studies on cellular and molecular mechanisms of arsenic trioxide (As2O3) in the treatment of acute promyelocytic leukemia: As2O3 induces NB4 cell apoptosis with down-regulation of Bcl-2 expression and alteration of PML-RARα/PML protein localization Blood 1996 88: 1052–1061

  9. 9

    Zhu J, Koken MH, Quignon F, Chelbi-Alix MK, Degos L, Wang ZY, Chen Z, de The H . Arsenic-inducing PML targeting onto nuclear bodies: implications for the treatment of APL Proc Natl Acad Sci USA 1997 94: 3978–3983

  10. 10

    Andre C, Guillemin MC, Zhu J, Koken MH, Quignon F, Herve L, Chelbi-Alix MK, Dhumeaux D, Wang ZY, Degos L, Chen Z, de The H . The PML and PML/RARα domains: from autoimmuity to molecular oncology and from retinoic acid to arsenic Exp Cell Res 1996 229: 253–260

  11. 11

    Shao W, Fanelli M, Ferrara FF, Riccioni R, Rosenauer A, Davison K, Lamph WW, Waxman S, Pelicci PG, Lo Coco F, Avvisati G, Testa U, Peschle C, Gambacorti-Passerini C, Nervi C, Miller WH Jr . Arsenic trioxide as an inducer of apoptosis and loss of PML/RAR alpha protein in acute promyelocytic leukemia cells J Natl Cancer Inst 1998 90: 124–133

  12. 12

    Look AT . Arsenic and apoptosis in the treatment of acute promyelocytic leukemia J Natl Cancer Inst 1998 90: 86–88

  13. 13

    de The H, Chomienne C, Lanotte D, Degos L, Dejean A . The t(15;17) translocation of acute promyelocytic leukemia fuses the retinoic acid receptor alpha gene to a novel transcribed locus Nature 1990 347: 558–566

  14. 14

    Kakizuka A, Miller WH Jr, Umesono K, Warrel RP Jr, Frankel SR, Murty VV, Dmitrovsky E, Evans RM . Chromosomal translocation t(15;17) in human acute promyelocytic leukemia fuses RAR alpha with a novel putative transcription factor, PML Cell 1991 66: 663–672

  15. 15

    Akao Y, Mizoguchi H, Kojima S, Naoe T, Ohishi N, Yagi K . Arsenic induces apoptosis in B-cell leukaemic cell lines in vitro: activation of caspases and down-regulation of Bcl-2 protein BrJ Haematol 1998 102: 1055–1060

  16. 16

    Zhang W, Ohnishi K, Shigeno K, Fujisawa S, Naito K, Nakamura S, Takeshita K, Takeshita A, Ohno R . The induction of apoptosis and cell cycle arrest by arsenic trioxide in lymphoid neoplasms Leukemia 1998 12: 1383–1391

  17. 17

    Ishitsuka K, Hanada S, Suzuki S, Utsunomiya A, Chyuman Y, Takeuchi S, Takeshita T, Shimotakahara S, Uozumi K, Makino T, Arima T . Arsenic trioxide inhibits growth of human T-cellleukaemia virus type I infected T-cell lines more effectively thanretinoic acids Br J Haematol 1998 103: 721–728

  18. 18

    Zhu XH, Shen YL, Jing YK, Cai X, Jia PM, Huang Y, Tang W, Shi GY, Sun YP, Dai J, Wang ZY, Chen SJ, Zhang TD, Waxman S, Chen Z, Chen GQ . Apoptosis and growth inhibition in malignant lymphocytes after treatment with arsenic trioxide at clinicallyachievable concentrations J Natl Cancer Inst 1999 91: 772–778

  19. 19

    Rousselot P, Labaume S, Marolleau JP, Larghero J, Noguera MH, Brouet JC, Fermand JP . Arsenic trioxide and melarsoprol induce apoptosis in plasma cell lines and in plasma cells from myeloma patients Cancer Res 1999 59: 1041–1048

  20. 20

    Shen ZY, Tan LJ, Cai WJ, Shen J, Chen C, Tang XM, Zheng MH . Arsenic trioxide induces apoptosis of oesophageal carcinoma in vitro Int J Mol Med 1999 4: 33–37

  21. 21

    Akao Y, Nakagawa Y, Akiyama K . Arsenic trioxide induces apoptosis in neuroblastoma cell lines through the activation of caspase 3 in vitro FEBS Lett 1999 455: 59–62

  22. 22

    Larochette N, Decaudin D, Jacotot E, Brenner C, Marzo I, Susin SA, Zamzami N, Xie Z, Reed J, Kroemer G . Arsenite induces apoptosis via a direct effect on the mitochondrial permeability transition pore Exp Cell Res 1999 249: 413–421

  23. 23

    Gurr JR, Bau DT, Liu F, Lynn S, Jan KY . Dithiothreitol enhances arsenic trioxide-induced apoptosis in NB4 cells Mol Pharmacol 1999 56: 102–109

  24. 24

    Li YP, Said F, Gallagher RE . Retinic acid-resistant HL60 cells exclusively contain mutant retinoic acid recepter-alpha Blood 1996 88: 2671–2682

  25. 25

    Dai J, Weinberg RS, Waxman S, Jing YK . Malignant cells can be sensitized to undergo growth inhibition and apoptosis by arsenic through modulation of the glutathione Blood 1999 93: 268–277

  26. 26

    Liu TX, Zhang JW, Zhang RB, Wu JS, Tao J, Tong JH, Yu M, Sun YW, Lanotte M, Waxman S, Zhu L, Chen SJ, Chen Z, Mao M . Identification of the molecular pathways underlying the retinoic acid-induced differentiation of acute promyelocytic leukemia cells Blood 1998 92: 193a (Abstr.)

  27. 27

    Wang ZG, Rivi R, Delva L, Konig A, Scheinberg DA, Gambacorti-Passerini C, Gabrilove JL, Warrell RP Jr, Pandolfi PP . Arsenic trioxide and melarsoprol induce programmed cell death in myeloid leukemia cell lines and function in a PML and PML-RARalpha independent manner Blood 1998 92: 1497–1504

  28. 28

    Muller S, Matunis JM, Dejean A . Conjugation with ubiquitin-related modifier SUMO-1 regulates the partitioning of PML within the nucleus EMBO J 1998 17: 61–68

  29. 29

    Muller S, Miller WH Jr, Dejean A . Trivalent antimonials induce degradation of the PML-RAR oncoprotein and reorganization of the promyelocytic leukemia nuclear bodies in acute promyelocytic leukemia NB4 cells Blood 1998 92: 4308–4316

  30. 30

    Koken MH, Daniel MT, Gianni M, Zelent A, Licht J, Buzyn A, Minard P, Degos L, Varet B, de The H . Retinoic acid, but not arsenic trioxide, degrades the PLZF/RARalpha fusion protein, without inducing terminal differentiation or apoptosis, in a RA-therapy resistant t(11;17)(q23;q21) APL patient Oncogene 1999 18: 1113–1118

  31. 31

    Fanelli M, Minucci S, Gelmetti V, Nervi C, Gambacorti-Passerini C, Pelicci PG . Constitutive degradation of PML/RARalpha through the proteasome pathway mediates retinoic acid resistance Blood 1999 93: 1477–1481

Download references


This work was supported in part by National Key Programme for Basic Research (973) National Natural Science Foundation of China (NNSFC) research grants No. 39970312 and No. 39730270 (GQC), and a NNSFC award for Outstanding Young Scientists (No. 39725011)(GQC), National Ministry of Public Health (GQC), Shanghai Municipal Foundation for Outstanding Young Researcher (GQC), Samuel Waxman Cancer Research Foundation and Clyde Wu Foundation of Shanghai Institute of Hematology.

Author information

Correspondence to Z Chen or G-Q Chen.

Rights and permissions

Reprints and Permissions

About this article


  • arsenic trioxide
  • apoptosis
  • differentiation
  • mitochondrial transmembrane potentials
  • retinoic acid receptor
  • sulfhydryl group

Further reading