Resveratrol enhances the suppressive effects of arsenic trioxide on primitive leukemic progenitors

Abstract

Efforts to enhance the antileukemic properties of arsenic trioxide are clinically relevant and may lead to the development of new therapeutic approaches for the management of certain hematological malignancies. We provide evidence that concomitant treatment of acute myeloid leukemia (AML) cells or chronic myeloid leukemia (CML) cells with resveratrol potentiates arsenic trioxide-dependent induction of apoptosis. Importantly, clonogenic assays in methylcellulose demonstrate potent suppressive effects of the combination of these agents on primitive leukemic progenitors derived from patients with AML or CML. Taken together, these findings suggest that combinations of arsenic trioxide with resveratrol may provide an approach for targeting of early leukemic precursors and, possibly, leukemia initiating stem cells.

Keywords: :

• arsenic trioxide
• myeloid leukemia
• apoptosis
• autophagy

Introduction

Arsenic trioxide (As₂O₃) exhibits potent antineoplastic activities,¹ including antileukemic effects that have led to the introduction of this agent in the treatment of patients with acute promyelocytic leukemia (APL).² Beyond APL, As₂O₃ has cytotoxic effects on other types of leukemias and solid tumors in preclinical models, but at concentrations that may not be clinically achievable because of toxicities.¹ Thus, approaches to promote generation of its effects at lower concentrations are needed in order to allow its use in non-APL leukemias and other malignancies.¹ Concomitant use of this agent with other drugs that may promote the induction of antineoplastic responses may provide a strategy for the development of combinations for the treatment of other hematological malignancies.

Resveratrol is a phytoalexin substance with previously established antioxidant and antineoplastic properties.³ This agent has been shown to induce apoptosis in a variety of lymphoid and myeloid leukemia cell lines.^4,5 The pro-apoptotic activity of resveratrol in leukemia cells appears to be regulated by the induction of the Fas ligand (FasL), through activation of the ASK1/JNK-dependent signaling pathway.⁶ Other studies in chronic myeloid leukemia (CML) cells have shown that resveratrol sensitizes CML cells to cell cycle arrest and apoptotic cell death when combined with imatinib mesylate.⁷ There has also been work demonstrating that resveratrol induces cell-cycle arrest and decreases the expression of survivin irrespective of p53 status.^8-10 Finally, there is evidence that in acute myeloid leukemia (AML), sensitization occurs through reactive oxygen species-mediated activation of the apoptotic pathway.¹¹ Resveratrol has also been linked to activation of autophagy. Recent studies have shown the ability of resveratrol to trigger an autophagic cell death in CML cells, via AMPK and JNK.¹² Furthermore, resveratrol has been shown to enhance radiation sensitivity in a variety of other malignancies, including melanoma, prostate cancer, and non-small lung cancer.^13-15

In the present study, we sought to determine the effects of resveratrol on arsenic trioxide -induced antileukemic responses. We were particularly interested in examining whether resveratrol can potentiate arsenic trioxide-induced autophagy or apoptosis, as our recent studies have established that in addition to apoptotic cell death, induction of autophagy is a key mechanism for the generation of arsenic-induced antileukemic effects on primary leukemic precursors.^16,17 As expected, both agents were found to induce an autophagic state, but resveratrol does not further enhance induction of autophagy by As₂O₃. On the other hand, resveratrol potently enhances As₂O₃-dependent apoptotic cell death and promotes the suppressive effects of As₂O₃ on primary leukemic progenitors from patients with AML and CML, suggesting a developmental therapeutics approach to target early leukemic precursors and leukemia initiating stem cells.

Results

In initial studies, we examined the induction of autophagy by resveratrol in leukemia cells. Treatment with resveratrol resulted in upregulation of LC3-II in the U937 AML cell line and the K562 CML cell line (Fig. 1A and B). Similarly, and as expected,^16,17 treatment with As₂O₃ resulted in upregulation of expression of LC3II (Fig. 1C). However, there was no synergy between the two agents in inducing expression of this marker of autophagy (Fig. 1C). Treatment of leukemic cells with As₂O₃ resulted in induction of apoptosis, as reflected by annexin V staining (Fig. 2A). Such induction of apoptosis was clearly enhanced by combinations of As₂O₃ with resveratrol (Fig. 2A). Consistent results were also obtained in immunoblotting experiments by probing for the expression of the late apoptosis marker, poly (ADP-ribose) polymerase (PARP) and its cleaved isoform (Fig. 2B). To further define the functional relevance of combined treatment of leukemic cells with As₂O₃ and resveratrol, we examined the effects of these agents on the BCR-ABL-transformed leukemic cell line KT1. Treatment with increasing doses of resveratrol and As₂O₃ resulted in decreased cell viability of KT1 cells with significant effects being recorded even at low resveratrol concentrations at either 48 h or 120 h (Fig. 3A and B).

Figure 1. Induction of autophagy by resveratrol or As₂O₃ in leukemia cells. (A) U937 cells were treated with resveratrol (5 μM) (R5) or (10 μM) (R10), for 24 h, as indicated. Total cell lysates were resolved by SDS-PAGE and immunoblotted with anti-LC3 or anti-GAPDH antibodies, as indicated. (B) K562 cells were treated with the indicated concentrations of resveratrol for the indicated times. Total cell lysates were resolved by SDS-PAGE and immunoblotted with anti-LC3 or anti-GAPDH antibodies, as indicated. (C) K562 cells treated with As₂O₃ at the indicated concentrations for 24 h, in the presence or absence of resveratrol (25 μM), as indicated. Total cell lysates were resolved by SDS-PAGE and immunoblotted with anti-LC3 or anti-GAPDH antibodies, as indicated.

Figure 2. Combination treatment As₂O₃ and resveratrol induces apoptosis in a dose-dependent manner. (A) KT1 cells were treated with the indicated concentrations of As₂O₃ and resveratrol for 48 h and subjected to flow cytometric analysis for annexin V staining. Means ± SE of 2 independent experiments are shown. (B) KT1 cells treated with As₂O₃ and resveratrol for 48 h at the indicated concentrations. Total cell lysates were resolved by SDS-PAGE and immunoblotted with anti-PARP or anti-HSP90 antibodies, as indicated.

Figure 3. Combination treatment of As₂O₃ and resveratrol reduces cell viability in a dose-dependent manner. KT1 cells were incubated in the presence or absence of As₂O₃ (2 μM) and the indicated concentrations of resveratrol for either 48 h (A) or 120 h (B). Cell viability was determined by an MTT assay. Data are expressed as a percentage of viability of control untreated samples. Means ± SE of 2 independent experiments are shown.

In subsequent experiments, we determined whether co-treatment of leukemia cells with As₂O₃ and increasing concentrations of resveratrol amplified the effects of these drugs on primitive leukemic precursors, compared with each drug alone. We exposed the BCR-ABL expressing leukemia cell lines K562 (Fig. 4A) and KT1 (Fig. 4B), and the acute myeloid leukemia cell line U937 (Fig. 4C), to As₂O₃ and, concomitantly, increasing concentrations of resveratrol. Effects on leukemic precursors were assessed by clonogenic assays in methylcellulose. Escalating doses of resveratrol resulted in enhanced antileukemic responses, as assessed by leukemic CFU-L colony formation derived from K562, KT1 or U937 cells (Fig. 4).

Figure 4. Enhanced antileukemic responses by combinations of As₂O₃ and resveratrol. (A–C) Primitive leukemic progenitor colony formation derived from K562 (A), U937 (B), or KT1 (C) cells was assessed in clonogenic assays in methylcellulose, following incubation of cells with As₂O₃ (0.5 μM) and the indicated concentrations of resveratrol. Data are expressed as a percentage of CFU-L of control untreated cells. Means ± SE of 3 (A) and 2 (B and C) independent experiments for each panel are shown.

As these findings suggested,enhanced antileukemic responses by combinations of resveratrol with As₂O₃, we sought to determine whether such effects occur on primary leukemic precursors from patients with CML or AML. As shown in Figure 5A, when the effects of combining As₂O₃ with escalating concentrations of resveratrol were determined on primary leukemic cells from CML patients, we found consistent enhancing effects (Fig. 5A). Similar results were obtained when responses were assessed on leukemic precursors from AML patients (Fig. 5B). Thus, combinations of resveratrol with As₂O₃ result in more potent effects than each agent alone, raising the possibility of combining these agents as a therapeutic approach to target leukemic precursors in myeloid leukemias.

Figure 5. Resveratrol enhances As₂O₃ responses in primary leukemia progenitors. (A) Effects of combinations of As₂O₃ (0.5 μM) and the indicated concentrations of resveratrol on leukemic CFU-GM from CML patients. Data are expressed as a percentage of CFU-GM of control untreated cells. Means ± SE of 6 independent experiments are shown. (B) Effects of combinations of As₂O₃ (0.5 μM) and the indicated concentrations of resveratrol on CFU-L from different AML patients. Data are expressed as a percentage of CFU-L of control untreated cells. Means ± SE of 3 independent experiments are shown.

Discussion

Beyond its clinical activity in APL, As₂O₃ has clinical potential for the treatment of other neoplastic disorders, as shown by extensive work using in vitro and in vivo preclinical models for different malignancies.¹ However, the ability of arsenic compounds to generate such responses is frequently limited in non-APL malignant cells by the emergence of leukemic cell resistance mechanisms and/or the simultaneous activation of negative feedback regulatory pathways that limit induction of arsenic biological responses. Extensive work in the past has provided evidence that such negative feedback regulatory cascades include the p38 MAP kinase^18,19 and its upstream²⁰ and downstream^21,22 effectors, as well as the MEK-ERK pathway and its effector kinase RSK1.^23,24 Other work has also implicated AKT/mTOR signaling pathways as negative feedback regulators activated in response to treatment of leukemia cells with arsenic trioxide^25,26 underscoring the complexity of leukemic cell resistance mechanisms. The simultaneous engagement of multiple feedback loops poses therapeutic challenges and makes selective targeting of leukemic cells difficult and limiting. Thus efforts to develop combinations of arsenic trioxide with other agents that promote its antileukemic effects via unique mechanisms of action are important, especially when the demonstrated capacity of arsenic to eliminate leukemic initiating stem cells under certain circumstances²⁷ is taken into account.

There is emerging evidence that the flavonoid resveratrol exhibits antileukemic effects. Previous studies have demonstrated that resveratrol induces autophagic cell death in BCR-ABL transformed leukemic cells,⁷ while it also causes cell cycle arrest in promyelocytic leukemia cells.¹⁰ We have previously shown that arsenic trioxide is a potent anti-leukemic agent and that, beyond apoptosis, autophagy plays a major role in the generation of arsenic responses in vitro in AML¹⁶ and CML¹⁷ cells. Remarkably, resveratrol has been reported to reduce cardiotoxicity and nephrotoxicity caused by As₂O₃ treatment,²⁸ while in a leukemia setting, there is emerging evidence suggesting that resveratrol can be used in As₂O₃-treated APL patients post-therapy to decrease arsenic-induced side effects such as hepatotoxicity.²⁹ There is also some evidence that resveratrol can enhance the anti-leukemic effects of imatinib mesylate on CML cells.⁷

In the present study, we provide evidence that resveratrol sensitizes AML and CML leukemia cell lines to the effects of As₂O₃ in vitro. We also show that resveratrol enhances the inhibitory effects of As₂O₃ on primary AML or CML primitive leukemic progenitors in vitro. Our studies demonstrate that such effects of resveratrol likely reflect potentiation of the pro-apoptotic properties of arsenic trioxide and do not appear to reflect modulation of arsenic-dependent autophagic cell death. It should be noted that previous studies¹⁶ have firmly established that induction of both apoptosis and autophagy are required for the generation of the suppressive effects of As₂O₃ on primitive leukemic precursors. This was shown by experiments demonstrating that pharmacological inhibitors of autophagy or apoptosis can partially reverse the inhibitory effects of arsenic on leukemic progenitors in clonogenic assays in methylcellulose.¹⁶ Although an enhancing effect of resveratrol on arsenic-dependent induction of an autophagic state could not be demonstrated in the current study, this could be attributed to an interplay between the various cell death pathways occurring simultaneously.³⁰ Future studies should clarify this issue and define specific components of pro-apoptotic or autophagic machineries that are modified and may account for the enhancing effects of resveratrol. Nevertheless, our findings demonstrate the potential of combination approaches to enhance the effects of arsenic on primitive AML and/or CML progenitors and further efforts in that direction are warranted.

Material and Methods

Cells and reagents

The K562, KT1, and U937 human leukemia cell lines were grown in RPMI 1640 medium supplemented with 10% fetal bovine serum and antibiotics. Arsenic trioxide (As₂O₃) and resveratrol were purchased from Sigma. Antibodies against PARP and LC3 were obtained from Cell Signaling Technology, Inc. Antibodies against c-Abl and p62/SQSMT1 were obtained from Santa Cruz Biotechnology Inc., and antibodies against glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were obtained from Millipore.

Cell lysis and immunoblotting

Cells were treated with the designated doses of As₂O₃ for the indicated times and subsequently lysed in the phosphorylation lysis buffer as previously described.^18,19 Immunoblotting using an enhanced chemiluminescence (ECL) method was performed as previously described.^31,32

Cell proliferation assays

Cells were treated with the indicated doses of As₂O₃, in the presence or absence of resveratrol for the indicated times. Cell proliferation assays using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) method were performed as in our previous studies.³³

Evaluation of apoptosis

Apoptosis was evaluated by flow cytometry for annexin V/PI staining as in our previous studies²⁵ using FITC Annexin V Apoptosis Detection Kit from BD PharMingen. For assessment of apoptosis assays, cells were treated with As₂O₃ (1 to 5 μM) with the indicated concentrations of resveratrol for 48 h, prior to flow cytometric analysis.

Clonogenic assays in methylcellulose to assess leukemic progenitor colony formation

Peripheral blood or bone marrows from patients with CML or AML were collected after obtaining consent approved by the Institutional Review Board of Northwestern University. The effects of arsenic trioxide on the growth of leukemic progenitors (CFU-L) were assessed by clonogenic assays in methylcellulose, as in previous studies.^34,35 As₂O₃ was used at final concentrations of 0.5 μM and for resveratrol at concentrations of 10, 25, and 50 μM.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

This work was supported by National Institutes of Health grants CA121192, CA155566, CA77816, and CA174205; and by grant I01CX000916 from the Department of Veterans Affairs. E.B. was supported by NIH training grant T32CA070085.

10.4161/cbt.27824