sign in sign up
<<
Home , Arsenic trioxide, Fluorouracil

ANTICANCER RESEARCH 30: 1157-1162 (2010)

Arsenic Trioxide Suppresses in 5-FU-resistant Colorectal Cell Line HT29 In Vitro Re-sensitizing Cells to 5-FU

POCHI RAMALINGAM SUBBARAYAN, KELVIN LEE and BACH ARDALAN

Department of Medicine, University of Miami Miller School of Medicine, Miami, FL – 33136, U.S.A.

Abstract. Identification of new agents with antitumor greatly needed. (CRC) is a major disease activity in chemoresistant tumors is urgently needed for the affecting many adult American populations. Fluoropyrimidine treatment of colorectal cancer. trioxide (As2O3), a analog 5- (5-FU) is the mainstay of CRC Food and Drug Administration (FDA) approved drug is treatment. Although 5-FU is no longer used as a single agent, successfully being used to treat acute promyelocytic leukemia its efficacy is enhanced by other drugs. Many of the added (APL). Several clinical trials also suggest its ineffectiveness drugs only work in the presence of 5-FU. on solid tumors. We proposed that may be 5-FU is a analog. It interferes with nucleic acid used as chemosensitizer, especially to 5-fluorouracil (5-FU). synthesis. Under normal physiology, uracil in the form of The effect of arsenic trioxide on cell proliferation of 5-FU- deoxyuridylate (dUMP) is converted to deoxy thymidylate sensitive and -resistant HT29 colorectal cancer cells in vitro (dTMP), by the enzyme thymidylate synthase (TS) (1). was tested by trypan blue dye exclusion assay, 2, 3-bis(2- Fluorinated uracil, viz. 5-FU, is converted to fluorodeo- methoxy-4-nitro-5-sulfophenyl)-S-[(phenylamino) carbonyl]- xyuridine monophosphate (FdUMP). It competes with dUMP 2H-tetrazolium hydroxide (XTT) cell proliferation assay and in tightly binding thymidylate synthase, the enzyme which analysis using flow cytometry. Gene expression was is responsible for the conversion of dUMP to dTMP. This analyzed using real-time polymerase chain reaction (PCR) to the inhibition of DNA synthesis. Upon sequestration and Western blot methods. There was a dose-dependent of free TS by FdUMP, its autoregulation is lost and thus the increase in cell detachment and proliferation in HT29 and cellular TS level increases uncontrollably. Thus in a large HT29FU cells. As a single agent, arsenic trioxide also down- number of patient populations, a higher level of TS is regulated thymidylate synthase (TS) expression without correlated to 5-FU resistance (2, 3). Thus, in general, an affecting the expression of some other genes analyzed in the elevated TS level is linked to treatment failure. above cell lines. Combination of arsenic trioxide and 5-FU If the TS level can be regulated, there is a possibility to increased cytotoxicity. In vitro data show that as a single delay or reduce the incidence of treatment failure. Any agent agent, arsenic trioxide down-regulated TS expression in HT29 that reduces or reverses 5-FU resistance will benefit the cells. Addition of 5-FU to these sensitized cells increased patients. In the recent past, RNA interference was used to cytotoxicity. These findings open up a possibility to use lower the TS level to sensitize 5-FU-resistant cells in vitro arsenic trioxide as a chemosensitizer in combination therapy. (4). However, there are no approved methods to deliver siRNA molecules in vivo. Besides antisense RNA, studies Development of resistance to is a major issue demonstrated that other small molecules also had anti-TS facing cancer researchers and patients. Agents that could activity. Cellular signaling molecule 3-oxododecanoyl lactone delay, reverse, or potentially avoid resistance development are (5) had an effect in down-regulation of TS expression. In the last decade, arsenic trioxide has shown promising results in treatment of many hematopoietic malignancies. In 1997, the Chinese clinical investigators, the first to treat acute Correspondence to: P.R. Subbarayan, Department of Medicine, promyelocytic leukemia (APL) (6, 7) with arsenic trioxide, Division of Hematology and Oncology, University of Miami School reported their study to the worldwide oncology community. of Medicine, 1550 NW 10th Avenue, Fox 431A, Miami, FL – Five years later, Miller et al. illustrated that arsenic trioxide 33136, U.S.A. Tel: +1 3052434887, Fax: +1 3052437722, e-mail: alone induced complete remission (CR) with minimal side- [email protected] effects in APL patients (8). Arsenic trioxide treatment achieved Key Words: Small molecules, gene expression analysis, thymidylate CR in 70 to 90% of newly diagnosed and 65 to >90% in synthase, real-time PCR, chemoresistance. relapsed APL patients. The disease-free survival rate was

0250-7005/2010 $2.00+.40 1157 ANTICANCER RESEARCH 30: 1157-1162 (2010)

Table I. List of primers used.

Primer Primer length Sequence 5’-3’ Product length

TS5’ Primer A 24 mer GGGCAGATCCAACACATCCTCCGC 294 bp TS Rev. Primer 20 mer GCCCAAGTCCCCTTCTTCTC Sp1F2 24 mer AGTTGGTGGCAATAATGGGGGCAA 254 bp Sp1R2 23 mer TTGGCACCCTGTGAAAGTTGTGT p53 (Sense) 23 mer CATTCTGGGACAGCCAAGTCTGT 609 bp p53 (Antisense) 23 mer CTGGGGAGAGGAGCTGGTGTTGT Cyclin B1 F 20 mer AGGCGAAGATCAACATGGCA 272 bp Cyclin B1 R 22 mer TTCTGGCTCAGGTTCTGGCTCT Akt 1 F 21 mer ACAGAGAACTTGTTGAGGGGA 351 bp Akt 1 R 20 mer CATCTGTCACCAGGGGCTTA β-Actin F 25 mer TCACCCACACTGTGCCCATCTACGA 295 bp β-Actin R 25 mer CAGCGGAACCGCTCATTGCCAATGG

These primers span the exon-intron junction so that genomic DNA is not amplified under the given conditions.

significantly prolonged, with minimal side-effects such as skin Cytotoxicity assay. A total of 1×105 HT29, and HT29FU cells rashes, fatigue, nausea/vomiting/diarrhea, musculoskeletal were plated in 24-well plates. After 24 hours, cells were treated pain, leukocytosis, function abnormalities and retinoic with 0.2, 0.4, 0.6, 0.8 and 1.0 μM of arsenic trioxide. At 72 acid syndrome. No cardiomyopathy was reported. In other hours post addition, cells were trypsinized and the total number of cells were counted using hemocytometer. In combination studies, CR was observed in 80 to 90% of APL patients experiments, 5 μM of 5-FU were added 24 hours post arsenic treated with arsenic trioxide (9). Based on these studies, the trioxide treatment and cells were incubated for an additional 48 US FDA approved arsenic trioxide for the treatment of APL. hours. The number of viable cells was determined by trypan blue We tested the ability of arsenic trioxide to lower TS dye exclusion assay. The experiments were carried out in expression in 5-FU-resistant colorectal cancer cells in vitro. duplicates and repeated twice. We used trypan blue exclusion assay to determine its anti- tumorigenic property, real-time PCR and Western blot Immunodetection of proteins. Total cell lysates were prepared by lysing cultured cells in M-PER reagent (Pierce, USA). Protein analyses to analyze the gene expression pattern. In this paper, was quantitated by Lowry’s method using Bio-Rad Protein Assay we present the results of our study on the effect of arsenic (Bio-Rad, USA) with bovine serum albumin (BSA) as the trioxide on cultured human colorectal cancer cell line HT29 standard. A total of 30 μg of solubilized protein were and its 5-FU-resistant derivative cell line HT29FU. electrophoresed on a 10% dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gel. After transferring to nylon Materials and Methods membrane, antibody-positive bands were detected with ECL reagent (Amersham, USA). Bio-Rad pre-stained protein molecular Cell lines and culture conditions. The human colorectal carcinoma weight marker was used as size standards. Horseradish peroxidase cell line HT29 was purchased from American Type Culture Collection (HRP)-conjugated anti-rabbit IgG was purchased from Promega, (ATCC), USA. HT29 cells were maintained in McCoy’s 5A medium USA, HRP-conjugated anti-mouse IgG antibodies were from supplemented with 10% fetal bovine serum (FBS), and 1% penicillin- Santa Cruz Biotechnology, USA; anti-actin was from Sigma, USA; and anti-TS from NeoMarkers, USA. streptomycin (PS) at 37˚C in a 5% CO2 humid atmosphere. All experiments were conducted with exponentially growing cells 24 hours after seeding at 1×105 to 1×106 cells in 24- or 6-well plates. Quantitative real-time PCR (RT-PCR). Total RNA from the cell lines under various treatment conditions was isolated using Tri-Reagent Development of 5-FU resistant HT29 cell line (HT29FU). To mimic (Sigma, USA). Oligonucleotides for cDNA synthesis and real-time in vivo development of 5-FU-chemoresistant colorectal cancer cells PCR were prepared in-house at the University of Miami DNA core and to characterize such resistant cells in vitro, we developed a 5- facility. The list of primers used and their respective sequences are FU-resistant HT29 cell line. The native HT29 cells were exposed to given in Table I. QuantiTect SYBR Green PCR Master Mix was increasing concentrations of 5-FU. This led to a cell line, HT29FU, purchased from Qiagen, USA. M-MLV reverse transcriptase and which was highly resistant to 5-FU. dNTP mix were purchased from Promega, USA. RNase inhibitor was purchased from Applied Biosystems, USA. All other chemicals Cell detachment assay. A total of 1×106 cells were seeded onto 6- were reagent grade. well plates. At 24 hours post seeding, they were treated with arsenic Total RNA (2 μg) was reverse transcribed using M-MLV trioxide at 1, 2 & 5 μM concentration. After 72 hours, the number of reverse transcriptase. Each reaction mix contained 0.5 μg of oligo detached cells was counted by trypan blue dye exclusion assay. The dT16 primer per μg of the total RNA. It was heated to 70˚C for 5 experiments were repeated twice. minutes and immediately chilled on ice. The final reaction

1158 Subbarayan et al: Arsenic Trioxide as Chemosensitizer

Figure 1. Arsenic trioxide induces cell detachment and proliferation arrest in human colon cancer cell lines. A: Cell detachment. 1×106 HT29 cells were cultured in the presence of 0, 1, 2 or 5 μM arsenic trioxide; 72 hours later the total number of detached cells was counted using a hemocytometer. B: Proliferation. 1×105 HT29FU cells were cultured in the presence of 0, 0.2, 0.4, 0.6 and 0.8 μM arsenic trioxide; 72 hours later the total viable cell numbers were determined using trypan blue exclusion assay. Experiments were set up in duplicate and repeated twice.

volume of 25 μl contained 0.5 mM each of dNTP mix, 25 U Results (unit) of RNase inhibitor and 200 U of M-MLV reverse transcriptase. The reaction mixture was incubated at 42˚C for 60 The 5-FU-resistant HT29 derivative cell line (HT29FU) minutes and stored on ice. differed morphologically and physiologically from the First strand synthesis mixture (5 μl) was used for real-time parental sensitive type. A few of the differences are: they quantitative PCR. The reaction mixture contained 200 nM each of overexpressed TS mRNA and consequently the protein. The the gene-specific primers and 25 μl of QuantiTect SYBR Green half maximal inhibitory concentration (IC ) value for 5-FU PCR Master Mix, in a final reaction volume of 50 μl. RT PCR was 50 performed in a 96-well plate on GeneAmp Sequence Detection was in excess of 128 μM. System 5700 (Applied Biosystems, USA). β-Actin gene was used as an internal reference. Each reaction was performed at two Arsenic trioxide has activity against 5-FU-sensitive and - different dilutions of the template. The mean normalized expression resistant colorectal cancer HT29 cells. We initially assessed (MNE) was calculated using the formula (10): the sensitivity of HT29 (5-FU-sensitive) and HT29FU (5-FU- Ctref Ctref Ctref resistant) to different concentration of arsenic trioxide, using (Eref) Well 1 (Eref) Well 2 (Eref) Well 3 ++ cell detachment and proliferation/cell number expansion as Cttarget Cttarget Cttarget (Etarget) (Etarget) (Etarget) measures of sensitivity. We saw a sharp increase in the number Well 1 Well 2 Well 3 of released cells between 1 and 5 μM for HT29 cells (Figure MNE= 1A). Arsenic trioxide does not appear to be cytotoxic to 3 HT29FU, rather at doses above 0.2 μM it is cytostatic. At 0.8 Where E, efficiency of PCR reaction; ref, reference gene (β-actin); μM, arsenic trioxide completely prevents expansion of the target, target gene (TS, p53, Sp1, Akt1, Cyclin B1). input cell numbers over 72 h (compared to a 3-fold expansion

1159 ANTICANCER RESEARCH 30: 1157-1162 (2010)

Figure 2. Arsenic trioxide inhibits TS mRNA and protein expression in a dose-dependent fashion. A: mRNA. HT29FU cells were cultured at the indicated concentrations of arsenic trioxide for 48 h, and then assessed for TS mRNA expression as detailed in the Materials and Methods section. B: Protein expression. HT29 and HT29FU cells were treated with the indicated concentrations of arsenic trioxide for 48 h. Total cell lysates were prepared and protein was quantitated by Lowry’s method. Solubilized protein (30 μg) was electrophoresed on a 10% SDS-PAGE gel. After transferring to nylon membrane antibody positive bands were detected with ECL reagent.

for untreated cells (Figure 1B). Comparison of these cell lines gene transcription (Sp1). As seen in Figure 3 (right panel), 0.6 suggests that it is equally active against 5-FU-sensitive (data μM As2O3 resulted in a three-fold decrease in TS mRNA not shown) and -resistant cells. expression in HT29FU cells. Arsenic treatment also resulted in a ten-fold decrease in p53 mRNA expression, but had little effect Arsenic trioxide specifically inhibits TS and p53 gene on Cyclin B1, Akt and Sp1 expression. Because Sp1 is a positive expression. Having established the cytotoxicity of arsenic regulator of TS gene transcription (11), these findings also trioxide on CRC cells in vitro, we investigated its effects on suggest that the effect of arsenic trioxide on TS mRNA levels is transcription in general and TS in particular. We initially not due to down-regulation of Sp1 expression (although a direct assessed the effect of arsenic trioxide concentration on TS effect on the transcription factor itself cannot be ruled out). expression by RT-PCR. As seen in Figure 2A, suppression of TS gene expression in HT29FU cells is most marked Discussion between arsenic trioxide concentrations of 0.2 and 0.6 μM. This down-regulation in TS mRNA is similarly reflected in Arsenic is an enigmatic compound. On the one hand, it is TS protein levels (Figure 2B). The increased expression of denigrated by the environmentalist and classified as a TS in untreated 5-FU-resistant HT29FU cells compared to carcinogen. From the earliest times, arsenic compounds have sensitive HT29 cells is consistent with previous reports (2). been used as a cure and rejuvenator for many illnesses (8). Following the establishment of 0.6 μM arsenic trioxide as an However, it lost its prime place in medicine during the mid effective dose, we examined the expression of a panel of 20th century. In the late 1990s, arsenic compounds were representative genes involved in other cellular functions (Table revived from relative oblivion by Chinese clinical investigators I). These included DNA damage repair and tumor suppression (7) and subsequently adapted by Western oncologists. Arsenic (p53), cell cycle (Cyclin B1), growth factor signaling (Akt1), and trioxide has proven to be effective against hematopoietic

1160 Subbarayan et al: Arsenic Trioxide as Chemosensitizer

Figure 3. Arsenic trioxide specifically inhibits TS gene expression. HT29FU cells were treated with 0.6 μM arsenic trioxide for 48 h. Total RNA was isolated using Tri-Reagent. cDNA was synthesized using M-MLV and dT16 primer. RT PCR was performed in a 96-well plate on GeneAmp Sequence Detection System 5700 with SYBR Green PCR master mix and appropriate primers. β-Actin gene was used as the internal reference. The mean normalized expression (MNE) was calculated using the formula as given in (10). (9, 12, 13). Its effectiveness against solid tumors had Real-time PCR and Western immunoblot results (Figure been varied. Zhang et al. (14) investigated the cytotoxicity of 2), demonstrated a decrease in TS with arsenic trioxide arsenic trioxide toward many different solid tumor cell lines. treatment. We did not assay TS activity because the They reported that the sensitivity to arsenic trioxide depends correlation between the reduction in mRNA and protein with on tumor origin. The cytotoxic dose ranged from 0.01 μM to 1 a concomitant loss in its enzymatic activity is well μM. Among the analyzed tumor types, gastric cancer cells were established (18). The level of TS mRNA, as monitored here the most sensitive. The serum arsenic concentration reached in by RT-PCR, decreased with arsenic trioxide treatment. TS is APL patients treated with arsenic trioxide was 1.8-2.2 μM (13). cell cycle regulated and shown to accumulate during the S- In our in vitro study, arsenic trioxide exerted a cytostatic effect phase of the cell cycle. It is also reported that arsenic trioxide at 0.6 μM, which is significantly lower than the achievable induces DNA damage, leading to p53 accumulation and serum arsenic concentration. consequently G1 arrest (19). Therefore the down-regulation Arsenic is believed to be a broad-spectrum agent and of TS may be attributed to G1 phase arrest. However, the therefore simultaneously affects many different pathways. Its concurrent decreases in p53 at mRNA and protein level efficacy has been well established in curing APL and its illustrate a different development other than G1 phase arrest. possible mechanism(s) of action is well elucidated. A recent In addition to its key role in the DNA synthesis pathway, TS review summarizes how arsenic and its derivatives influence also down-regulates translation of p53. Hence, a decrease in various signal transduction pathways (8). In this context, our TS should to an increase in p53 protein as seen in other data showing suppression of TS and p53 by arsenic, but not of experiments (20). But we observed the suppression of p53 other genes analyzed here, demonstrate that arsenic exerts and TS following arsenic treatment. Consequently, the differential effects on cellular gene expression. Other possibility of TS down-regulation due to G1 phase arrest investigators found that as a result of degradation of suggests further study is necessary. Arsenic might indeed promyelocytic leukemia (PML) gene product by arsenic, there is play a direct role in hitherto unknown ways. Thus, a most a loss of p53 function in APL. In response to arsenic treatment, interesting topic for further study is to elucidate the p53 was down-regulated at both RNA and protein levels in the mechanism of TS down regulation by arsenic trioxide. colon cancer cell line HT29FU. This is in agreement with Contrary to the success in the case of APL, multiple myeloma previous reports (15). Consequently, we may expect decreased and T-cell lymphoma clinical trials, arsenic trioxide as a single cell death. Consistent with this thought, apparently arsenic agent was not successful in treating solid tumors such as renal trioxide acts more as a cytostatic agent (Figure 1B) than a cell carcinoma (21), CRC (22) metastatic melanoma (23) and cytotoxic agent, at least with reference to the HT29FU cell line. germ cell malignancies (24). Intravenous administration of However, arsenic stimulates other pro-apoptotic pathways, thus arsenic trioxide might deliver the drugs directly to hematopoietic counterbalancing the loss of p53 function (16). In contrast, tumors. In solid tumors, because of the poor vascularity, arsenic arsenic is also reported to up-regulate p53 and p53-regulated may not reach the tumor core. Thus it may not be active on its proteins in human fibroblasts (17). Therefore, it is also possible own. However, when it is combined with an active agent such that cellular origin also determines arsenic action. as 5-FU, the combination could be synergistic.

1161 ANTICANCER RESEARCH 30: 1157-1162 (2010)

The anti-neoplastic potency of arsenic trioxide alone and 10 Muller PY, Janovjak H, Miserez AR and Dobbie Z: Processing its ability to down-regulate TS in 5-FU-resistant cells in vitro of gene expression data generated by quantitative real-time RT- makes a strong case for a to assess the toxicity/ PCR. Biotechniques 32: 1372-1379, 2002. 11 Horie N and Takeishi K: Identification of functional elements in efficacy of this particular combination. Arsenic trioxide is an the promoter region of the human gene for thymidylate synthase approved drug and 5-FU is the mainstay of treatment for and nuclear factors that regulate the expression of the gene. many solid tumors. Hence, starting a clinical trial using a J Biol Chem 272: 18375-18381, 1997. combination of arsenic trioxide and 5-FU may facilitate 12 Barbey JT and Soignet S: Prolongation of the QT interval and development of a new treatment regimen. in patients treated with arsenic trioxide for acute promyelocytic leukemia. Ann Intern Med 135: 842-843, Conflict of interest statement 2001. 13 Bahlis NJ, McCafferty-Grad J, Jordan-McMurry I, Neil J, Reis I, None. Kharfan-Dabaja M, Eckman J, Goodman M, Fernandez HF, Boise LH and Lee KP: Feasibility and correlates of arsenic References trioxide combined with ascorbic acid-mediated depletion of intracellular glutathione for the treatment of relapsed/refractory multiple myeloma. Clin Cancer Res 8: 3658-3668, 2002. 1 Rustum YM: Thymidylate synthase: a critical target in cancer 14 Zhang TC, Cao EH, Li JF, Ma W and Qin JF: Induction of therapy? Front Biosci 9: 2467-2473, 2004. apoptosis and inhibition of human gastric cancer MGC-803 cell 2 Link KH, Kornmann M, Butzer U, Leder G, Sunelaitis E, Pillasch growth by arsenic trioxide. Eur J Cancer 35: 1258-1263, 1999. J, Salonga D, Danenberg KD, Danenberg PV and Beger HG: 15 Pearson M, Carbone R, Sebastiani C, Cioce M, Fagioli M, Saito Thymidylate synthase quantitation and in vitro chemosensitivity S, Higashimoto Y, Appella E, Minucci S, Pandolfi PP and Pelicci testing predicts responses and survival of patients with isolated PG: PML regulates p53 acetylation and premature senescence nonresectable liver tumors receiving hepatic arterial infusion induced by oncogenic Ras. Nature 406: 207-210, 2000. chemotherapy. Cancer 89: 288-296, 2000. 16 Torii S, Egan DA, Evans RA and Reed JC: Human Daxx 3 Salonga D, Danenberg KD, Johnson M, Metzger R, Groshen S, regulates Fas-induced apoptosis from nuclear PML oncogenic Tsao-Wei DD, Lenz HJ, Leichman CG, Leichman L, Diasio RB domains (PODs). EMBO J 18: 6037-6049, 1999. and Danenberg PV: Colorectal tumors responding to 5- 17 Yih LH and Lee TC: induces p53 accumulation through fluorouracil have low gene expression levels of an ATM-dependent pathway in human fibroblasts. Cancer Res dihydropyrimidine dehydrogenase, thymidylate synthase, and 60: 6346-6352, 2000. thymidine phosphorylase. Clin Cancer Res 6: 1322-1327, 2000. 18 Johnston PG, Lenz HJ, Leichman CG, Danenberg KD, Allegra 4 Schmitz JC, Chen TM and Chu E: Small interfering double- CJ, Danenberg PV and Leichman L: Thymidylate synthase gene stranded RNAs as therapeutic molecules to restore and protein expression correlate and are associated with response chemosensitivity to thymidylate synthase inhibitor compounds. to 5-fluorouracil in human colorectal and gastric tumors. Cancer Cancer Res 64: 1431-1435, 2004. Res 55: 1407-1412, 1995. 5 Dolnick R, Wu Q, Angelino NJ, Stephanie LV, Chow KC, Sufrin 19 Liu Q, Hilsenbeck S and Gazitt Y: Arsenic trioxide-induced JR and Dolnick BJ: Enhancement of 5-fluorouracil sensitivity by apoptosis in myeloma cells: p53-dependent G or G /M cell an rTS signaling mimic in H630 colon cancer cells. Cancer 1 2 cycle arrest, activation of caspase-8 or caspase-9, and synergy Research 65: 5917-5924, 2005. with APO2/TRAIL. Blood 101: 4078-4087, 2003. 6 Shen ZX, Chen GQ, Ni JH, Li XS, Xiong SM, Qiu QY, Zhu J, 20 Liu J, Schmitz JC, Lin X, Tai N, Yan W, Farrell M, Bailly M, Tang W, Sun GL, Yang KQ, Chen Y, Zhou L, Fang ZW, Wang YT, Chen T and Chu E: Thymidylate synthase as a translational Ma J, Zhang P, Zhang TD, Chen SJ, Chen Z and Wang ZY: Use of regulator of cellular gene expression. Biochim Biophys Acta arsenic trioxide (As O ) in the treatment of acute promyelocytic 2 3 1587: 174-182, 2002. leukemia (APL): II. Clinical efficacy and in 21 Vuky J, Yu R, Schwartz L and Motzer RJ: Phase II trial of relapsed patients. Blood 89: 3354-3360, 1997. arsenic trioxide in patients with metastatic renal cell carcinoma. 7 Chen GQ, Shi XG, Tang W, Xiong SM, Zhu J, Cai X, Han ZG, Invest New Drugs 20: 327-330, 2002. Ni JH, Shi GY, Jia PM, Liu MM, He KL, Niu C, Ma J, Zhang P, 22 Subbarayan PR, Lima M and Ardalan B: Arsenic trioxide/ ascorbic Zhang TD, Paul P, Naoe T, Kitamura K, Miller W, Waxman S, acid therapy in patients with refractory metastatic colorectal Wang ZY, de The H, Chen SJ and Chen Z: Use of arsenic carcinoma: a clinical experience. Acta Oncol 46: 557-561, 2007. trioxide (As O ) in the treatment of acute promyelocytic 2 3 23 Kim KB, Bedikian AY, Camacho LH, Papadopoulos NE and leukemia (APL): I. As O exerts dose-dependent dual effects on 2 3 McCullough C: A phase II trial of arsenic trioxide in patients APL cells. Blood 89: 3345-3353, 1997. with metastatic melanoma. Cancer 104: 1687-1692, 2005. 8 Miller WH Jr, Schipper HM, Lee JS, Singer J and Waxman S: 24 Beer TM, Tangen CM, Nichols CR, Margolin KA, Dreicer R, Mechanisms of action of arsenic trioxide. Cancer Res 62: 3893- Stephenson WT, Quinn DI, Raghavan D and Crawford ED: 3903, 2002. Southwest Oncology Group phase II study of arsenic trioxide in 9 Soignet SL, Frankel SR, Douer D, Tallman MS, Kantarjian H, patients with refractory germ cell malignancies. Cancer 106: Calleja E, Stone RM, Kalaycio M, Scheinberg DA, Steinherz P, 2624-2629, 2006. Sievers EL, Coutre S, Dahlberg S, Ellison R and Warrell RP Jr: multicenter study of arsenic trioxide in relapsed Received September 11, 2009 acute promyelocytic leukemia. J Clin Oncol 19: 3852-3860, Revised March 12, 2010 2001. Accepted March 19, 2010

1162