Arsenic-Based Anticancer-Combined Therapy: Novel Mechanism Inducing Apoptosis of Cancer Cells

Arsenic, known as both a naturally occurring toxic element and a traditionally used drug, has caught a great deal of attention from worldwide people due to its curable anticancer effect in patients with acute promyelocytic leukemia (APL). Among the arsenicals, arsenic trioxide (ATO) has been the most widely used anticancer drug. Since ATO exerts an anticancer effect by mediating apoptosis, numerous studies have made efforts to uncover the molecular mechanisms by which ATO activates and/or mediates the apoptotic signaling pathway in cancer cells. Recent advances in cancer therapeutics have led to a paradigm shift away from the traditional cytotoxic drugs toward the targeting of proteins closely associated with driving the cancer phenotype. Here, we discuss novel current arsenic-based combination therapies to treat cancer in both clinical and experimental settings. We also discuss the novel molecular mechanism underlying apoptosis induced by the combined therapies.


Arsenic in cancer treatment
Arsenic, one of the ancient drugs, is currently used as a therapeutic agent worldwide because of its substantial anticancer activity. Arsenic trioxide (ATO), a trivalent arsenite (As III ), was initially used for the treatment of chronic myelogenous leukemia (CML) in the nineteenth century [1]. ATO readily induces apoptosis by promoting differentiation of acute promyelocytic leukemia (APL) cells and is thereby utilized as a therapeutic drug in the clinical setting [2]. ATO is currently used to treat patients with APL and recurrent/relapsed multiple myeloma (MM) [3]. Numerous studies that examined the tumor suppressive and/or proapoptotic effect of ATO have revealed the molecular mechanism by which ATO exerts an anticancer effect in both solid cancer and hematological malignancies (see below section). Thus, accumulating evidence implicates ATO and/or other arsenicals in clinical use as a promising drug to treat cancer patients. Besides ATO monotherapy, there is a growing body of evidence that ATO may be a favorable drug when combined with not only conventional anticancer therapeutics including radiation and chemotherapy but also recently developed molecular-targeted drugs. In one example, the combined treatment of all-trans-retinoic acid (ATRA) with ATO has been shown to synergistically induce apoptosis of APL cells and clinically shown to be better outcome and less toxicity than the combined treatment of ATRA with chemotherapy in the treatment of patients with APL [4,5]. This chapter summarizes the anticancer effect of ATO-based combination therapies in different types of solid cancer and hematological malignancies. Furthermore, the molecular mechanism by which ATO-based combination therapies exert a proapoptotic effect in cancer cells is discussed.

Novel arsenic-based combinatorial anticancer therapy
Based on the promising effect of arsenic on apoptosis in cancer cells, it has been further utilized as a combinatorial drug with other chemotherapeutic agents and/or molecular-targeted drugs to gain its anticancer effect in various types of cancer. The combination drugs, possible target molecules, molecular basis underlying combination treatment-induced apoptosis, and combination indices (CI) for each cancer are summarized in Figure 1 and Table 1.

Colon cancer
Lee et al. reported that sulindac, a nonsteroidal anti-inflammatory drug (NSAID), enhances ATO-induced apoptosis by inhibiting NF-κB activation mediated through the blocking of phosphorylation and degradation of IκB-alpha in . In addition, Cai et al. reported that combined ATO-PI3K inhibitor LY294002 treatment synergistically suppresses the proliferation of colon cancer cell lines, where ATO decreases Hh pathway transcription factor Gli1 and its downstream gene expression including BCL2 and CCND1 [7].

Prostate cancer
Therapeutics in prostate cancer is based on the progression stage of the cancer, and radiation therapy is widely utilized for treatment. ATO was reported to enhance the radiation sensitivity of androgen-dependent (LNCaP) and androgen-independent (PC-3) human prostate cancer cells by mediating inhibition of the Akt/mTOR signaling pathway both in vitro and in vivo [8]. As shown in colon cancer, it has been reported that ATO inhibits the proliferation of the prostate cancer cell line PC-3 by suppressing the Hh signaling pathway and the tumor suppression effect was further enhanced by a classic Hh pathway inhibitor cyclopamine in vivo [9]. Furthermore, In most cancer cells, ATO generates intracellular ROS, which potentially triggers activation of the apoptotic signaling pathway. A glutathione synthesis inhibitor BSO enhances the effect of ATO-induced ROS generation by depleting GSH [43,60], while a platinum drug CDDP cooperatively enhances ROS generation [12]. ROS-induced ER stress as well as MAPK phosphorylation can occur in mitochondrial dysfunction, which subsequently activates caspase-3/caspase-7 and induces apoptosis. (B) Arsenite methyltransferase (As3MT)-mediated metabolic methylation of ATO decreases methyl donors (S-adenosylmethionine (SAM)) and increases its metabolite S-adenosyl-l-homocysteine (SAH), which may inhibit DNA methyltransferase activity of DNMT. ATO itself downregulates gene expression of DNMTs [117]. Consequently, promoter demethylation of silenced genes, including miR-155, miR-200c, secreted frizzled-related protein-1 (SFRP1), and ERα, upregulates their gene expression, which may exert anticancer activities [22,[118][119][120]. The demethylation effect of ATO and other anticancer therapeutics may cooperatively induce apoptosis in cancer cells. (C) In APL cells, ATO binds to PML, while ATRA binds to RARα. Thus, combined ATO-ATRA treatment synergistically induces proteasomal degradation of PML-RARα oncoprotein, differentiation, and subsequent apoptosis [73]. ATO also can induce proteasomal degradation of oncoproteins including chimeric protein BCR-ABL (generated in CML cells), Tax (HTLV-I-encoded protein), and NPM1 (frequently mutated in AML cells). ATO and combination agents are indicated in the Double Square. MMA, monomethylarsonous acid; DMA, dimethylarsinous acid.

Oral cancer (oral squamous cell carcinoma (OSCC))
OSCC is the most common head and neck neoplasm and is highly associated with poor prognosis, despite advances that have been made in diagnostic and therapeutic strategies such as surgery, chemotherapy, and radiotherapy. ATO was reported as a combinatorial drug with radiotherapy [11] and a platinum-based antineoplastic drug cisplatin (CDDP) [12], both of which ATL -Degradation of HTLV-I transactivator protein (Tax) [109,110] Primary effusion lymphoma -Inhibition of NF-κB activity [111] Combination index value (CI); synergism is indicated by CI < 1. are the most standard therapies for OSCC. Since ATO/CDDP-induced apoptosis was almost completely abrogated by NAC, ROS generation may be closely associated with the tumor suppression effect ( Figure 1A) [12]. Recent publications also implicated the therapeutic application of arsenic in the treatment for OSCC. Wang

Ovarian cancer
The rate of the mortality from ovarian cancer is highest among malignant tumors of the female genital organs. As indicated in other types of cancer, ATO was reported to exert synergistic cytotoxic effects against ovarian cancer cells when it was combined with CDDP, one of the standard chemotherapeutics for ovarian cancer, and/or mTOR inhibitor RAD001 [15,16]. Ong et al. reported that both buthionine sulfoximine and ascorbic acid differentially enhance ATO-mediated cell killing by mediating GSH depletion and the oxidative stress-related pathway, respectively [17].

Cervical cancer
Radiotherapy has been generally used for the treatment of patients with cervical cancer, which is the most common cancer among females worldwide, as a monotherapy and combined therapy with chemotherapeutics, such as platinum-based drugs. The combined ATO-radiation treatment was expectably reported to exert a beneficial antitumoral effect on cervical cancer

Breast cancer
Breast cancer is one of the leading causes of cancer-related deaths among women worldwide. In breast cancer cells, ATO was reported to reduce the expression of DNA methyltransferase-1 (DNMT1) and to induce the expression of estrogen receptor α (ERα), whose expression has been epidemiologically recognized to increase disease-free survival and indicate an overall better prognosis ( Figure 1B)

HCC and bile duct carcinoma (cholangiocarcinoma)
Hepatocellular carcinoma (HCC) is the most common primary liver malignancy, which is the sixth most common type of cancer worldwide. Sorafenib, a known multikinase inhibitor, can extend the survival rate of patients with advanced HCC. ATO was reported to synergize with sorafenib to inhibit the proliferation and promote the apoptosis of HCC cells by diminishing the sorafenib-induced activation of Akt and/or its downstream factors, including glycogen synthase kinase-3β, mTOR, ribosomal protein S6 kinase, and eukaryotic translation initiation factor 4E-binding protein 1 [28]. ATO was also reported to potentiate the anticancer effect of genistein . It has also been reported that combined ATO-sulindac treatment induces synergistic augmentation of cytotoxicity in both human NSCLC cell lines NCI-H157 and NCI-H1299 by mediating ROS-induced MAPK phosphorylation and via c-Jun NH 2 -terminal kinase-dependent Bcl-xL phosphorylation, respectively [40,41]. Indomethacin, a nonselective cyclooxygenase inhibitor (a structural isoform of sulindac), was also shown to enhance the ATO-induced cytotoxic effect in A549 cells by mediating activation of ERK and/or p38 MAPKs [42]. Han et al.
reported that buthionine sulfoximine (BSO), a glutathione synthesis inhibitor, enhances ATOinduced apoptosis in A549 cells, in which the apoptosis was related to the increased level of intracellular ROS (Figure 1A) [43]. In addition, combined ATO-CDDP treatment induced apoptosis and synergistically suppressed the proliferation of human NSCLC cell lines A549 and H460, with CI values 0.5 and 0.6, respectively, where CI <1 defines synergism [44]. Lam et al. indicated that combined treatment with ATO and a selective fibroblast growth factor receptor (FGFR) inhibitor PD173074 cooperatively suppresses tumor proliferation both in vitro and in vivo in the lung squamous cell carcinoma (SCC) cell line SK-MES-1, in which ATO-PD173074 decreased FGFR1, Akt, Src, c-Raf, and Erk, at least in part by mediating proteasomal degradation [45]. Gu et al. reported that ROS-mediated ER stress and mitochondrial dysfunction were involved in the apoptosis induced by resveratrol and ATO in A549 cells [46].

Gastric cancer
In gastric cancer, ATO and ABT-737, a small-molecule drug that inhibits Bcl-2/Bcl-xL antiapoptotic molecules, cooperatively suppressed the proliferation of human gastric cancer cell lines SGC7901 and MGC-803 [47]. Although ABT-737 has low solubility and oral bioavailability, the other Bcl-2 inhibitor ABT-199 has been developed for clinical use for the treatment of hematological malignancies including chronic lymphoid leukemia (CLL) [48].

Pancreatic cancer
In pancreatic cancer, Wang et al. reported that parthenolide (PTL), a sesquiterpene lactone from the medical herb feverfew, enhances apoptosis of human pancreatic cancer cell lines PANC-1 and BxPC-3 by mediating ROS generation and subsequent caspase activation via the mitochondrial pathway [50]. The combined ATO-PTL treatment significantly reduced tumor growth rates of PANC-1 xenografts compared with those treated with either PTL or ATO alone [50]. Another report focused on the limited efficacy of ATO on cytotoxicity in pancreatic ductal adenocarcinoma probably because of the high-cellular ROS scavenging activity. Lang et al. reported that PX-478, a hypoxia-inducible factor-1 inhibitor, robustly strengthens the antigrowth and proapoptosis effect of ATO on Panc-1 and BxPC-3 pancreatic cancer cells in vitro by mediating ROS accumulation [51].

Glioma and glioblastoma (GBM)
In glioma, several reports indicated the effectiveness of arsenic for combination therapy. Kim et al. reported that ATO specifically increases expression of death receptor 5 (DR5), a death receptor of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) in a subset of human glioma cell lines but not in astrocytes [52]. They also showed that combined ATO and TRAIL treatment synergistically reduces the survival of glioma cells, with a CI < 1.0 [52].
ATO-radiation combination treatment enhanced autophagic effects in U118-MG cells through increased mitotic arrest and regulation of PI3K/Akt and ERK1/2 signaling pathways [53]. Similar observations were reported in human fibrosarcoma and osteosarcoma, where the combined treatment arrested their cell cycle at the G 2 /M phase [54,55]. ATO is a ROS inducer, and heme oxygenase-1 (HO-1) is known to be increased in arsenic-treated cells [56]. Liu et al. reported that HO-1 inhibition or Nrf2 knockdown significantly potentiated ATO-induced cytotoxic effects on glioma cells [57]. This result suggests that ATO-induced gene expression including the Nrf2 signaling pathway may be partly involved in the resistance of cancer cells to ATO. Dizaji et al.
reported that combined treatment with ATO and silibinin, a natural polyphenolic flavonoid, synergistically induces apoptosis and inhibits invasiveness in the human GBM cell line U87MG, in which the expression of cathepsin B, uPA, MMP-2, MMP-9, membrane type 1-MMP, survivin, Bcl-2, and CA9 decreased after treatment [58]. Gülden et al. reported that combined treatment with ATO and silibinin reduces the viability of A-172 by mediating intracellular arsenic accumulation [59]. As described in lung cancer, BSO depleted cellular glutathione and acted synergistically with ATO in rat C6 astroglioma cells [60]. Karsy et al. investigated the effect of ATO and ATRA on neurosphere formation of U87MG possessing wild-type (wt) p53 and/ or codon-specific p53 mutant (R273H) [61]. They found that ATO and ATRA treatment could induce apoptosis of both U87-p53 (wt) and U87-p53 (R273H) and could potently suppress CSC properties in vitro [61]. It is well known that c-Myc is also required for the maintenance of CSCs of various cancers including GBM [62]. Yoshimura

Acute promyelocytic leukemia (APL)
Patients with newly diagnosed APL, also known as a AML M3 subtype with APL-specific chromosomal translocation t(15;17) (q22;q21) and PML/RARα chimeric fusion protein, are mostly cured after standard ATRA with chemotherapy, while toxicity and refractoriness to the treatment are observed in some patients. Accumulating evidence shows the superiority of novel combined ATRA and ATO therapy for the treatment of patients with APL in terms of event-free survival, relapse-free survival, and less hematologic toxicity, compared to ATRA with chemotherapy [4,5,[69][70][71][72]. The molecular basis underlying synergistic effects between ATRA and ATO has been biologically demonstrated. ATRA and ATO differentially bind PML/RARα protein, the proteasomal degradation of which readily induces terminal differentiation, and subsequent apoptosis in APL cells ( Figure 1C) [73]. In addition to the effect of ATO on the PML moiety, ATO-induced anticancer activities including inhibition of leukemic progenitor self-replication and antiangiogenic effects might be involved in the antileukemic activity. The combined ATRA with intravenous administration of ATO has been developed to combine ATRA with oral As4S4 administration as a routine treatment option for appropriate patients [74].  [77]. The antileukemic activity of ATO was also enhanced by the combination strategies with granulocytemonocyte colony stimulation factor [78], a noncalcemic vitamin D analog 19-Nor-1,25(OH)2D2 [79], N-(beta-Elemene-13-yl) tryptophan methyl [80], a selective inhibitor of epidermal growth factor receptor (EGFR) gefitinib [81], and high-dose vitamin C (ascorbic acid) [82], all of which enhance ATO-induced differentiation and apoptosis of APL cells.

Acute myeloid leukemia (AML)
AML is a malignant disease of the bone marrow, where juvenile leukocytes are arrested in an early stage of differentiation. It has been reported that AML patients with FLT mutations including FLT3-internal tandem duplication (FLT3-ITD) had significantly shorter overall and disease-free survival [83]. Takahashi et al. reported that combined treatment with ATO and FLT3-specific inhibitor AG1296 synergistically induces apoptosis in FLT3-ITD-positive cells, but not in Flt3 wild-type cells [84]. The combined ATO with ATRA, a novel standard treatment for patients with APL, was shown to exert synergistic cytotoxicity against FLT3-ITD AML cells via co-inhibition of FLT3 signaling pathways [85]. In addition, ATO-ATRA was shown to induce apoptosis of NPM1-mutated AML cells by targeting nucleophosmin (NPM1) oncoprotein, whose mutation possibly represents a therapeutic target because of high frequency in >30% AML [86]. As indicated in lung cancer and glioma, BSO was shown to enhance the ATOinduced anticancer effect by mediating ROS generation in AML cells [87] and other leukemic/lymphoma cells [88], suggesting that combined ATO-BSO treatment would be one of the attractive alternative therapies for cancer treatment. It has also been reported that combined treatment with ATO and dichloroacetate [89], azacytidine [90], rapamycin [91], and aclacinomycin A [92] enhances apoptosis in AML cells. Wang et al. reported that ethacrynic acid and a derivative enhance apoptosis in ATO-treated myeloid leukemia and lymphoma cell lines; this combination treatment generates high levels of ROS, activates JNK, and subsequently decreases the protein level of antiapoptotic molecule Mcl-1 [93].

Multiple myeloma (MM)
ATO, melphalan, and ascorbic acid (AA) combination therapy (MAC) is a therapeutic option for patients with relapsed or refractory MM [94]. Grad et al. initially showed that clinically relevant doses of AA decrease GSH levels and potentiate ATO-mediated cell death of MM cell lines [95]. Current therapeutics for MM, such as a proteasome inhibitor, namely, bortezomib (BOR) and carfilzomib, and immunomodulatory drugs, namely, thalidomide, lenalidomide (LEN), and pomalidomide, have successfully improved the patient survival, though MM remains an incurable disease [96]. In view of the current MM therapeutics, ATO was reported to enhance the anti-myeloma cytotoxicity of BOR [97] and sensitivity of MM cells to lenalidomide (LEN) [98]. Wen et al. showed that the enhanced cytotoxicity of ATO-BOR is associated with augmented STAT3 inhibition, JNK activation, and upregulation of Bim, p21, p27, and p53 as well as downregulation of Bcl-2 [97]. Jian et al. showed that ATO upregulates cereblon, the antimyeloma target of LEN, thus potentiating the sensitivity of MM cells [98]. The anti-myeloma activity of ATO was also enhanced by the combination strategies with a vitamin E analog Trolox [99], a specific MEK inhibitor PD325901 [100], a natural quinoid diterpene cryptotanshinone (also known as STAT3 inhibitor) [101,102], and a phytochemical sulforaphane [103].

Chronic myelogenous leukemia (CML)
Recent therapeutics for CML were developed by targeting the Bcr-Abl fusion protein generated from a Philadelphia (Ph) chromosome with reciprocal translocation of chromosomes 9 and 22. Despite the advances in CML therapeutics including Bcr-Abl tyrosine kinase inhibitors (TKIs), TKI therapy can produce a subpopulation of CML cells with a Bcr-Abl gene mutation that leads to resistance to TKI therapy, which results in a poorer prognosis in 10-15% of patients with CML. Several reports indicated the efficacy of ATO-based combined therapy for CML cells [104][105][106][107][108]. Du et al. reported that combined treatment of ATO with imatinib, which is the first approved TKI, coordinately enhances apoptosis of CML cells by mediating intrinsic (upregulation of BAX), extrinsic (upregulation of TNFR1, CASP8, and CASP10), and ER stressrelated pathways (HSPA5 and DDIT3) [104]. Wang et al. also showed the additive effect of ATOnilotinib, a second-line TKI agent, on the proliferation and differentiation of primary leukemic cells from patients with CML in blast crisis [105]. In addition, combined nilotinib-ATO treatment induces ER stress-mediated apoptosis in imatinib-resistant K562 cells by mediating JNK activation [106]. Li et al. reported that anti-miR-21 oligonucleotide sensitizes K562 cells to ATO and enhances ATO-induced apoptosis probably by mediating upregulation of programmed cell death 4 (PDCD4) [107]. In a CML mice model, combined treatment with ATO and interferon-α (IFN-α) was reported to be superior to imatinib [108] in terms of overall survival of secondary recipients, indicating that ATO-IFN-α may exhaust the activity of CML leukemia-initiating cells.

Other leukemia and lymphoma
As the efficacy was shown in a CML mice model [108], ATO and IFN-α synergized to induce cell cycle arrest and apoptosis in adult T-cell leukemia/lymphoma (ATL)-derived human T-cell lymphotropic virus type I (HTLV-I)-transformed cells [109,110]. El-Sabban et al. reported that combined ATO-IFN-α treatment induces the degradation of Tax, which is the viral transactivator protein that plays a critical role in HTLV-I-induced transformation and apoptosis resistance [110]. Similarly, the enhanced ATO-IFN-α-induced apoptosis was shown in primary effusion lymphoma [111]. Darwiche et al. showed that synergism of ATO-ATRA is especially observed in the HTLV-I-transformed cells expressing RARα protein [112]. In acute lymphoblastic leukemia (ALL), low-dose ATO sensitized glucocorticoid-resistant ALL cells to dexamethasone via an Akt-dependent pathway [113]. Jung et al. and Zhao et al. independently showed the synergistic anticancer effects of ATO with BOR in mantle cell lymphoma, which is an aggressive and highly incurable B-cell non-Hodgkin lymphoma [114,115]. Ding et al. recently reported that combined treatment of ATO with cucurbitacin B, an effective component of the dichloromethane extraction from Trichosanthes kirilowii maxim, synergistically enhances apoptosis by inhibiting STAT3 phosphorylation in Burkitt's lymphoma cell lines both in vitro and in vivo [116].

Summary
In this chapter, we show that arsenical compounds enhance cancer cell apoptosis when combined with other anticancer therapeutics including radiation, chemotherapies, and moleculartargeted drugs. Although a number of reports have shown the anticancer effects of arsenic and have discussed the possible molecular targets of ATO in malignant cells, molecular mechanisms underlying ATO-based synergistic anticancer effects with other anticancer therapeutics remain obscure. In the past decade, next-generation sequencing (NGS) technologies have tremendously improved and have clarified the whole context of genomic alterations in cancer cells, among which phenotypic and functional heterogeneity arises within the same tumor as a consequence of genetic changes, environmental differences, and anticancer therapy [121][122][123]. This indicates that the possibility of targeting single molecules and/or signaling pathways as well as single cellular biological processes may generate a different malignant population of cancer cells, some of which may acquire a certain drug resistance. Therefore, novel therapeutic agents and/or strategies are required to overcome drug resistance and improve both the disease outcome and the quality of life for patients with cancer. Further understanding of the relationship between induction of apoptosis and genetic/epigenetic changes in cancer cells may contribute to improvement in selectivity for cancer treatment. Additional studies are required to understand the synergistic anticancer action regarding ATO-based combination therapeutics to develop a novel combined therapy for cancer.