Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

Oxidative Stress Gene Expression Profile Correlates with Cancer Patient Poor Prognosis: Identification of Crucial Pathways Might Select Novel Therapeutic Approaches

Oxidative Stress Gene Expression Profile Correlates with Cancer Patient Poor Prognosis:... Hindawi Oxidative Medicine and Cellular Longevity Volume 2017, Article ID 2597581, 18 pages https://doi.org/10.1155/2017/2597581 Review Article Oxidative Stress Gene Expression Profile Correlates with Cancer Patient Poor Prognosis: Identification of Crucial Pathways Might Select Novel Therapeutic Approaches Alessandra Leone, Maria Serena Roca, Chiara Ciardiello, Susan Costantini, and Alfredo Budillon Experimental Pharmacology Unit, Istituto Nazionale Tumori Fondazione G. Pascale-IRCCS, Naples, Italy Correspondence should be addressed to Alessandra Leone; a.leone@istitutotumori.na.it Received 15 March 2017; Accepted 30 May 2017; Published 9 July 2017 Academic Editor: Lars Bräutigam Copyright © 2017 Alessandra Leone et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The role of altered redox status and high reactive oxygen species (ROS) is still controversial in cancer development and progression. Intracellular levels of ROS are elevated in cancer cells suggesting a role in cancer initiation and progression; on the contrary, ROS elevated levels may induce programmed cell death and have been associated with cancer suppression. Thus, it is crucial to consider the double-face of ROS, for novel therapeutic strategies targeting redox regulatory mechanisms. In this review, in order to derive cancer-type specific oxidative stress genes’ profile and their potential prognostic role, we integrated a publicly available oxidative stress gene signature with patient survival data from the Cancer Genome Atlas database. Overall, we found several genes statistically significant associated with poor prognosis in the examined six tumor types. Among them, FoxM1 and thioredoxin reductase1 expression showed the same pattern in four out of six cancers, suggesting their specific critical role in cancer-related oxidative stress adaptation. Our analysis also unveiled an enriched cellular network, highlighting specific pathways, in which many genes are strictly correlated. Finally, we discussed novel findings on the correlation between oxidative stress and cancer stem cells in order to define those pathways to be prioritized in drug development. 1. Introduction lipoxygenase, and cyclooxygenase) or by nonenzymatic reac- tions, such as during the mitochondrial respiratory chain. Reactive oxygen species (ROS) are commonly identified as These considerations highlight the concept that the source of ROS is extremely heterogeneous. Indeed, ROS can be found oxygen reactive molecules associated with a wide variety of in the environment, as pollutants, tobacco smoke, and iron physiologic events [1] as well as cancer, diabetes, obesity, neu- salts, or generated inside the cells through multiple mecha- rodegeneration, and other age-related diseases [2, 3]. A nisms [4]. Within cells, mitochondria, cytosol, single reduction-oxidation (redox) reaction concerns the transfer of electrons (reducing power) from a more reduced (nucleo- membrane-bound organelles (peroxisomes, endosomes, and phagosomes), or exosomes shed from plasma membranes, as philic) to more oxidized (electrophilic) molecules. ROS can well as extracellular fluids, including plasma, are all involved be classified in two groups: (1) free radical ROS containing one or more unpaired electron(s) in their outer molecular in ROS generation [3, 5, 6]. Mitochondria are the main ROS producers, principally because they are the site of the respira- orbitals (i.e., superoxide radicals and hydroxyl radicals); (2) tory chain when electron leakage can react with molecular oxy- nonradical ROS which are chemically reactive and can be con- gen, resulting in the formation of superoxide, which can verted to radical ROS (i.e., hydrogen peroxide), although they subsequently be converted to other ROS molecules. Then, gen- do not have unpaired electron(s). In both cases, ROS can be produced by either enzymatic reactions (i.e., NADPH oxidase, erated ROS either can be detoxified or can leave the organelle through channels such as voltage-dependent anion channels metabolic enzymes such as the cytochrome P450 enzymes, 2 Oxidative Medicine and Cellular Longevity Survival Stress Cell proliferation death adaptation ROS accumulation Oxidative damage Oxidative damage Increased lipid to proteins to DNA peroxidation Altered gene Loss of DNA repair regulation activity Altered signal Mutation transduction Altered cell growth, differentiation, and apoptosis Figure 1: Redox stress activation in physiology. The production of abnormally large amounts of ROS leads to persistent changes in signal transduction and gene expression that, in the last instance, could give to cell death. The steady-state levels of ROS are determined by the rate of ROS production and their clearance by scavenging mechanisms. (VDAC) or aquaporin, or by small vesicles such as exosomes mechanisms, involving several antioxidant ROS scavengers, [3, 5, 7]. However, ROS can also be the product of β-oxidation as glutathione peroxidase (GPx), thioredoxin (Trx), catalase in peroxisomes, of prostaglandin synthesis and detoxification (CAT), superoxide-dismutase (SOD), and the nuclear factor reactions by cytochrome P450, or of NADPH-mediated erythroid 2 (NRF2) pathway [4, 7]. If a further increase in ROS levels occurs, then the cells undergo apoptotic cell death reaction in phagocytes [4, 5]. ROS are biologically important in a variety of physiolog- (Figure 1). Therefore, under physiological conditions, in ical systems, including adaptation to hypoxia, regulation of order to guarantee cellular redox homeostasis, cells regulated autophagy, immunity, differentiation, and longevity. They intracellular ROS levels by applying a tight regulation of ROS regulate many signal transduction pathways by directly generation and of ROS detoxifying pathways. reacting with proteins and by modulating transcription In this review, we first summarized the role of oxidative factors and gene expression [1]. At low levels, ROS promote stress molecules in cancer initiation and progression and the cellular proliferation, differentiation, and migration as well proposed oxidative stress-targeted anticancer approaches. as cellular stress-responsive survival pathways such as Next, in order to derive cancer-type specific oxidative stress nuclear factor-κB (NF-κB), thus inducing proinflammatory gene profiles and their potential prognostic role, we integrated cytokines [4, 8]. Because of ROS’ highly reactive potential a publicly available oxidative stress gene signature [9] with the toward biological molecules, excessive ROS levels can dam- data extracted from the Cancer Genome Atlas (TCGA) data- age cellular components such as DNA, proteins, and lipids. base. Then, we reviewed some of those genes/pathways corre- To counteract these effects, cells activate “ROS adaption” lating with patient’s survival, in order to define potential novel Oxidative Medicine and Cellular Longevity 3 Less evidences are available about the regulation of ROS anticancer therapeutic targets. Finally, we highlighted novel findings on the correlation between oxidative stress and cancer by microenvironment; however, new efforts have been stem cells (CSC). recently focused in this field [5, 12]. In this regard, Chan et al. demonstrated that cancer-associated fibroblast- (CAF-) derived ROS are able to induce the acquisition of an oxidative 2. The Role of Oxidative Stress Molecules in CAF-like state on normal fibroblasts. Then, these oxidatively Cancer Initiation and Progression transformed normal fibroblasts promoted the development A link between ROS and cancer progression dates back to of aggressive tumors via a TGFβ1-mediated Smad3 signaling, suggesting an important relationship between the extracellu- 1981 when increased levels of H O induced by insulin were 2 2, shown to promote tumor cell proliferation. Almost three lar redox state and cancer aggressiveness [25]. decades later, several studies sustained this hypothesis, reporting increased levels of oxidative damage products in 3. Targeting Oxidative Stress as Anticancer clinical tumor specimens and plasma as well as in cancer cell Therapy lines [5]. Based on these evidences, to date, the idea that altered redox balance and deregulated redox signaling are The first approach to prevent or treat cancer, by targeting strongly implicated in any steps of carcinogenesis as well as ROS, was based on the use of antioxidant reagents [11, 15]. in the resistance to treatment, by affecting many, if not all, In one of the first trials, based on supplementation of sele- hallmarks of cancer is widely accepted [10, 11]. Indeed, cur- nium, vitamin E and β-carotene on the diet showed a reduc- rently, the role of ROS in cancer initiation and progression tion of overall mortality and cancer rates [26]. However, a through the modulation of cell proliferation, apoptosis, following trial not only failed to obtain consistent results angiogenesis, and the alteration of the migration/invasion but also indicated that in certain cases, antioxidants can program is well described [7, 12, 13]. For example, ROS rather promote cancer initiation and progression. Concor- may affect proliferation by a ligand-independent transactiva- dantly, two trials of cancer prevention, the CARET on male tion of different receptor tyrosine kinase via ERK activation smokers, treated with vitamin A and/or β-carotene and the and may induce tissue invasion and metastatic dissemination SELECT trial, on older males treated with vitamin E and/or by activation of metalloproteinases. Moreover, the release of selenium, resulted in an increased incidence of lung and vascular endothelial growth factor and angiopoietin induced prostate tumors, respectively [27–29]. Similar contradictory by ROS promote tumor angiogenesis and anoikis [12, 14]. results were shown in the trials using antioxidant treatment Nonetheless, the exact origin of ROS generation during as adjuvant therapy [30]. cancer development and disease progression and how this Based on these results, almost a decade ago, ROS event could be druggable remains still unclear. Increasing inducers were proposed as anticancer strategy, in order to evidences reported a link between ROS activation and the overcome the specific threshold of ROS level beyond which presence of some oncogenes, such as Ras, c-Myc, or Bcr- cancer cells undergo ROS-mediated cell death [4, 5]. The first Abl [2, 15, 16]. Activation of oncogenic signaling might agents used are those improving electrons leak from the contribute to the increase of ROS levels, which in turn by respiratory complexes in the mitochondria, such as the arse- promoting genomic instability could affect both nuclear nic trioxide, or conventional chemotherapeutic drugs such as and mitochondrial DNA. The consequent activation of anti- doxorubicin. Indeed, patients treated with those agents oxidants’ signaling within tumor cells can also promote cancer showed lipid peroxidation in their plasma as well as low progression and metastasis [2, 15–18]. Furthermore, cancer levels of vitamin E, vitamin C, and β-carotene in the blood cells undergo metabolic changes to counteract the oxidative [4]. The mechanism of action of these agents seems to be stress, also contributing to metastatic program [5, 19, 20]. related to their ability to generate ROS directly from the Loss of functional p53 is involved in ROS induction, due mitochondria. Indeed, doxorubicin and arsenic trioxide pen- to p53 “genome guardian” role in sensing and removing etrate in the inner membrane of the mitochondria and oxidative damage to DNA, thus preventing genetic instability induce superoxide radical production by modulating the [5, 21]. Anyhow, unlike oncogenes, the role of tumor sup- electron transport chain. Also 5-fluorouracil increases mito- pressors in the modulation of ROS is more complex, depend- chondrial ROS with a different mechanism, mediated by ing on the specific tumor suppressor itself. For example, p53 [4, 31]. Ionizing radiations represent other important ataxia-telangiectasia mutated (ATM) is a cellular damage ROS inducers, because they are able to promote by them- sensor that by regulating cell cycle and DNA repair preserves selves high level of ROS and also because they might increase genomic integrity. Deficiency of ATM gene, either in patients NADPH oxidase, an important source of ROS [32]. More- or in mice, has been shown to produce elevated ROS levels over, we and others have demonstrated, in different models and a chronic oxidative stress status. Recently, cytoplasmic and in different combination settings, that oxidative injury ATM is described to activate a pathway leading to autophagy played a significant functional role in the antitumor effect through repression of mammalian target of rapamycin com- of histone deacetylase inhibitors (HDACi), a class of epige- plex 1 (mTORC1) in response to elevated ROS levels [22, 23]. netic antitumor compounds currently in clinical practice in Another example regards the loss of PTEN that determines haematological malignancies [7, 13, 33–42]. AKT hyperactivation and inactivation of the forkhead Recently, a new ROS inducer compound, Elesclomol homeobox type O (FoxO) transcription factor and therefore (STA-4783), has been developed and tested, both in enhanced susceptibility to oxidative stress [24]. in vitro and in vivo preclinical studies as well as in clinical 4 Oxidative Medicine and Cellular Longevity cancer, TCGA_COAD for colon cancer, TCGA_HNSCC trials [5, 43]. Interestingly, the result from a phase II trial using Elesclomol in combination with chemotherapy, in for head and neck cancer (HNSCC), GSE31210 for lung can- malignant melanoma patients, showed ROS generation and cer, TCGA_PRAD for prostate cancer, and METABRIC for oxidative damage associated with prolonged progression- breast cancer [52, 53]. PPISURV automatically derives the free survival [44]. Unfortunately, these results were not rep- currently known interactome for a gene of interest and corre- licated in a phase III trial, where Elesclomol treatment was lates expression levels of its interactome, with survival out- suspended due to adverse toxic effects [45]. The reason of come in multiple publicly available clinical expression data this failure could be ascribed, at least in part, to cancer cells’ sets containing microarray expression data set annotated capability to activate ROS adaption mechanisms by increasing with survival data. In details, as reported by Antonov et al. levels of ROS scavengers, especially at advanced stages. This [54], in the case of the option “single gene survival analyses event is particularly efficacious in CSC, as described in the last on a single data set,” the PPISURV program exploits rank paragraph of this review. To counteract the ROS adaptation information from expression data sets that reflect the relative mechanisms, a plausible solution could be the combination mRNA expression level. The samples are grouped with of ROS inducers either with another ROS inducer or with respect to expression rank of the gene in order to correlate compounds that suppress cellular antioxidants, to overcome survival information to the expression level of a gene in a par- the threshold useful to induce cell death, The latest approach ticular data set. The groups are then subdivided in basis to was tested by using an inhibitor of the scavenger SOD2, 2- “low expression” and “high expression” where expression Me, in combination with arsenic trioxide in lymphocytic rank of the gene is less or more than average expression rank across the data set, respectively. This separation of patients leukemia and urothelial carcinoma cells [46, 47]. Similarly, the combination between the inhibitor of the antiapoptotic into “low” and “high” groups in the data set along with sur- protein bcl2 ABT-737 and the ROS inducer, N-(4-hydroxy- vival information is then used to find any statistical differ- phenyl) retinamide, or the combination between an NRF2 ences in survival outcome and to draw Kaplan-Meier plot. inhibitor and a glutathione-depleting agents, showed increas- Hence, PPISURV establishes a correlation of the selected gene with survival and assesses the sign of the effect and if ing therapeutic efficiency compared to single-agent treatment [48, 49]. Based on these data, several clinical trials of combina- the gene deregulation is associated with positive or negative tion treatment between ROS inducers and scavenger inhibi- outcome. tors are ongoing, including a multicenter phase II trial with Notably, a significant number of oxidative stress genes the iron chelator Triapine and gemcitabine in advanced were negatively correlated with survival in solid carcinomas, reinforced the idea that oxidative stress plays a crucial role in non-small-cell lung cancer [5]. cancer cells (Figure 3). Furthermore, going deep to our bioin- formatics analysis, we observed that breast, lung, and 4. Bioinformatics Correlation between HNSCC cancers were those more susceptible to oxidative stress gene expression fluctuations. To explain these data, Oxidative Stress Gene Expression and one hypothesis could be that all these tumors are more vul- Prognosis in Solid Cancer Patients nerable to external insults (i.e., pollutants) that, as mentioned above, are an important source of ROS. Furthermore, we Although the biological role of oxidative stress pathways has been extensively demonstrated, it is still unclear which and speculate that this phenomenon could be also related to the how oxidative stress genes predict bad prognosis and if their high mutational load of those tumors. Indeed, several studies showed that either breast (particularly triple-negative sub- modulation is cancer-type specific. Here, to address this question, we took advantage of Cancer Genome Atlas group) or lung cancer exhibited an elevated mutational load (TCGA) database that, by profiling RNA expression levels which is closely associated to mutations in DNA damage and DNA mutational status for thousands of genes, has gen- repair genes as well as to intrinsic genomic instability erated comprehensive maps of the key genomic changes in [55–57]. Similarly, recently it has been demonstrated that the overall mutational load was higher in old HNSCC patients several types of cancer, enabling correlative analysis of criti- cal cellular pathways involved in each type of cancers [50, that represent a high percentage of all HNSCC cancers, com- 51]. In details, we compared cancer patient overall survival pared to younger patients [58]. On the contrary, pancreatic, (OS) and the mRNA levels of 73 oxidative stress genes, prostate, and even colon (with exception of microsatellite selected from a public available oxidative stress signature instability (MSI) high subgroup) cancers are described as less hypermutated and thus, we speculate, are also less dependent [9], in different solid tumors. Specifically, the signature included peroxidases, which are represented by glutathione to the oxidative stress and genomic instability [59–61]. peroxidases (GPx) and peroxiredoxins (TPx); genes impli- A further detailed analysis of our correlation between cated in ROS metabolism (i.e., DUSP1, FoxM1, and oxidative gene expression signature and OS unveiled that HMOX1); and genes involved in superoxide metabolism, the behavior of modulated genes was different among the cancers examined, with the exception of two genes involved such as superoxide dismutase (SOD). Starting from the selec- tion of the 73 oxidative stress genes, bioinformatics investiga- in ROS metabolism, such as FoxM1 and TXNRD1, found tions were performed as described in Figure 2. as statistically significantly high in poor prognosis patients In details, bioinformatics analysis was made by SynTar- in four out of six of the tumor types analyzed (Figures 3 get online tool (http://www.bioprofiling.de/PPISURV) using and 4). For this reason, those two genes are described below in details in two specific sections of the review. Other five the following public datasets: TCGA_PAAD for pancreatic Oxidative Medicine and Cellular Longevity 5 73 oxidative TGCA data in solid stress genes cancers PPISURV tool Correlation between negative overall survival (OS) and mRNA levels of genes showed by heat map and Kaplan-Meier curves String analyses Enriched cellular networks Figure 2: Bioinformatics analyses. Flow chart reporting step-by-step bioinformatics approach to unveil the most important genes/pathways involved in the correlation between oxidative stress and cancer. genes, DUSP1, EPHX2, NUDT1, RNF7, and SEPP1, demon- damage and cell death [64]. RNF7 (RING finger protein-7) strated a statistically significant modulation in poor progno- acts as a metal chelating protein, a scavenger of ROS at the sis patients in three out of six tumor types (Figure 3 and expense of self-oligomerization. RNF7 was found overex- Suppl. Figure S1 available online at https://doi.org/10.1155/ pressed in several tumor types, especially in lung carcinoma, 2017/2597581). Briefly, DUSP1 is a dual-specificity phospha- and associated with poor prognosis [65]. tase-1, which is recognized as a key player for inactivating SEPP1 is a selenoprotein 1, involved in cellular incorpo- ration of the selenium circulating in the plasma. Moreover, different MAPK isoforms. Recently, a role of DUSP-1 as cen- tral redox-sensitive regulator in monocytes has been demon- SEPP1 has some antioxidant activity, as target of NRF2 fam- strated [62]. EPHX2 is a cytosolic epoxide hydrolase, implied ily. In agreement, Bae et al. showed that some antioxidant in cancer progression and metastasis, in differentially man- genes known also as NRF2 targets, including SEPP1, were ner based on the stages of carcinogenesis. Indeed, Bracalante also transcriptionally modulated by the oncosuppressor et al. demonstrated that in A7 melanotic cells, resembling less BRCA1, thus suggesting that BRCA1 regulates the activity aggressive tumor cells, anti-oxidant genes, including EPHX2, of NRF2 and protects cells against oxidative stress [66]. were upregulated in response to oxidative stress, while they Finally, in order to identify a more relevant oxidative were downregulated in G10 metastatic melanoma cells [63]. stress family in our setting, we performed an additional bio- NUDT1, nudix hydrolase 1, is the most prominent mamma- informatics analysis where, independently from their trend lian enzyme among other enzymes responsible for hydrolyz- of expression associated to poor prognosis, all modulated ing oxidized DNA precursors. NUDT1 is commonly genes were analyzed in the biological database STRING, a upregulated in a wide variety of tumors to avoid incorpora- resource of known and predicted protein-protein interaction. tion of oxidized nucleotides that, in turn, induce DNA As shown in Figure 5, our analysis reveals an enriched 6 Oxidative Medicine and Cellular Longevity Genes Gene levels p value Gene levels p value Gene levels p value Gene levels p value Gene levels p value Gene levels p value Breast cancer Lung Head and neck Pancreas Prostate Colon RNF7 NUDT1 PRDX5 NOX5 OXSR1 HSPA1A PNKP BNIP3 NQO1 SRXN1 TXNRD1 FOXM1 GCLM ATOX1 FTH1 GTF2I MPV17 NCF2 PDLIM1 PRNP SIRT2 APOE GSS LPO SOD1 ALOX12 CCS SOD2 TXN GPX1 GPX5 GPX6 GPX7 HMOX1 KRT1 NOX4 PRDX6 SCARA3 SFTPD GPX2 PRDX2 STK25 GSR MBL2 NOS2 SQSTM1 DHCR24 DUOX1 GCLC MPO MT3 NCF1 TTN CYGB EPX GPX3 MSRA PREX1 SOD3 UCP2 CAT TXNRD2 OXR1 CCL5 DUOX2 DUSP1 TPO GPX4 AOX1 EPHX2 SEPP1 Figure 3: Bioinformatics correlation between oxidative stress gene expressions and poor prognosis in 6 different tumor types. Heat map in which we report in red or in green if the high or low expression of genes was negatively correlated with survival, respectively. Moreover, we evidenced in yellow when the correlation is statistically significant (with p value <0.05). In the first column, we evidenced in magenta, the genes similar modulated among cancers and in blue those oxidative stress family extracted from STRING analysis. Notably TXNRD1 showed a central role in both analyses. Oxidative Medicine and Cellular Longevity 7 TXNRD1 Head and neck Prostate Breast cancer Lung (TCGA_HNSCC) (TCGA_PRAD) (METABRIC) (GSE31210) 1.0 1.0 1.0 1.0 0.8 0.8 0.8 0.8 0.6 0.6 0.6 0.6 0.4 0.4 0.4 0.4 0.2 0.2 0.2 0.2 0.0 0.0 0.0 0.0 0 50 100 150 200 250 300 0 20 400 60 80 100 120 050 100 150 0 50 100 150 200 250 Time in Time in Time in Time in High expression TXNRD1 High expression TXNRD1 High expression TXNRD1 High expression TXNRD1 Low expression TXNRD1 Low expression TXNRD1 Low expression TXNRD1 Low expression TXNRD1 FOXM1 Breast cancer Lung Pancreas Prostate (METABRIC) (GSE31210) (TCGA_PAAD) (TCGA_PRAD) 1.0 1.0 1.0 1.0 0.8 0.8 0.8 0.8 0.6 0.6 0.6 0.6 0.4 0.4 0.4 0.4 0.2 0.2 0.2 0.2 0.0 0.0 0.0 0.0 0 50 100 150 200 250 300 0 20 40 60 80 100 120 020 40 60 80 0 50 100 150 200 250 Time in Time in Time in Time in High expression FOXM1 High expression FOXM1 High expression FOXM1 High expression FOXM1 Low expression FOXM1 Low expression FOXM1 Low expression FOXM1 Low expression FOXM1 Figure 4: TXNRD1 and FoxM1 expression related to patient survival. Kaplan-Meier curves showing the survival in the case of high and low expression of TXNRD1 and FOXM1 in solid cancer patients. PRNP DHCF NUDT1 RNF7 NCF2 STK25 FTH1 APOE UCP2 SCARA3 SIRT2 MT3 PREX Node size Node color Small nodes Colored nodes MSRA FOXM1 Protein of unknown 3D structure Query proteins and first shell of interactors Large nodes White nodes NQO1 Some 3D structure is known or predicted Second shell of interactors TPO Known interactions Predicted interactions Others From curated databases Text mining Gene neighborhood SFTP Experimentally determined Gene fusions Coexpression TTN SRXN1 Gene co-occurrence Protein homology OXR1 MPV17 ALOX12 KRT1 GTF21 SEPP1 BNIP3 Figure 5: STRING analysis of the 58 oxidative stress genes. Association network in STRING analysis shows interactions of glutathione peroxidase, superoxide dismutase, and thioredoxin as principal oxidative stress signaling among six different tumor types. 8 Oxidative Medicine and Cellular Longevity different mechanisms have been proposed. For example, the cellular network, in which many genes, including GPx, SOD, and Trx pathways (the latter including TXNRD1), are induction of EMT by activation of Slug [76], stabilization of strongly correlated, as demonstrated by both experimental Smad3/Smad4 complex, and activation of TGFβ pathway [77] as well as the modulation of extracellular matrix by studies and text mining (see red and green lines, resp.). Sim- ilar analyses were also performed for each tumor type sepa- affecting the levels of uPA, uPAR, MMP-2, MMP-9, and rately, or considering high or low gene expression VEGF have been proposed [78, 79]. Moreover, FoxM1 coop- individually, confirming in almost all tumor types GPx, erates with survivin and nuclear XIAP in the promotion of SOD, and Trx signaling as those predominant (Suppl. chemoresistance [80]. Finally, further studies demonstrated that FoxM1 induces resistance to all the therapeutics tested Figures S2 and S3). Based on these analyses, together with TXNRD1, we in breast cancer (including cisplatin, paclitaxel, and trastuzu- decided to review the correlated pathways enriched in the mab) by several mechanisms: (1) acting on DNA-damage network, in details (see below), analyzing their role in cancer repair pathways, (2) promotion of cell cycle progression, and the possible therapeutic strategies to hit them. (3) inhibition of cell cycle checkpoints, and (4) apoptosis induction [72]. FoxM1 gene is widely described as amplified also in lung 5. FoxM1, a Critical Regulator of Oxidative cancer, regulating cell proliferation by promoting both G1/S Stress during Tumorigenesis and G2/M transition, differentiation, and transformation The highly conserved transcription factor FoxM1 belongs to [81] as well as inhibition of apoptosis [82]. Recently, a direct link between FoxM1-induced ROS and lung cancer progres- the forkhead box transcription factor family, similarly to the best known member of FoxO family. However, different from sion has been proposed by Tahmasbpoura et al. Their study showed elevated rate of lung cell proliferation related to high the members of FoxO family, FoxM1 is expressed only in pro- liferative cells. Indeed, FoxM1 as a target of the cyclinD- FoxM1 expression in patients exposed to sulfur mustard, a CDK4/6 kinases, is reactivated when quiescent cells reenter well known agent able to induce ROS [83]. Beyond the mechanisms described, the molecular basis of in the cell cycle and reach a maximal level in S-phase which is maintained throughout G2 and mitosis [67, 68]. Beyond this FoxM1 dysregulation has been also related to the capability role on proliferation, FoxM1 regulates metastasis, apoptosis, of vitamin D receptor (VDR)/FoxM1 axis to affect cell stem- and DNA damage repair [69–71]. Furthermore, FoxM1 has ness and to induce an invasive and metastatic phenotypes in been shown to prevent oxidative stress-dependent premature pancreatic cancer. Indeed, the authors observed that VDR activation reduced the levels of FoxM1, inducing nuclear senescence. Park et al. showed how ROS themselves are inducers of FoxM1 expression, which in turn is able to stimu- accumulation of β-catenin [84]. late antioxidant genes. The authors proposed the inhibition of In prostate cancer (PCa), only few studies focused on the FoxM1 as a new therapeutic strategy to kill cancer cells selec- role of FoxM1; for instance, FoxM1 and its target CENPF, a tively [71]. In agreement, FoxM1 knocking-down was structural protein of kinetochore, have been both proposed as critical drivers of PCa development and as prognostic reported to sensitize human pluripotent stem cells to oxidative stress, as a consequence of activated-CAT5 downregulation, a markers of poor survival [85]. Concordantly, Lin et al. FoxM1 antioxidant target gene [69]. unveiled different miRNAs regulating FoxM1-CENPF axis A growing body of evidences reported high FoxM1 as fre- taking advantage of miRNA expression profile available in quently related to poor prognosis in multiple cancers, con- Taylor dataset of prostate specimens (normal, localized, and metastatic tissues) [86]. Notably, since CENPF regulates cordantly with our bioinformatic results [50]. To date, several mechanisms have been proposed to explain the activ- several genes important for metastasis, including MMP2, ity of FoxM1 in cancer progression, including the activation MMP9, LOX, CXCR4, and CXCL12, dysregulation of the miRNA-COUP-TFII-FoxM1-CENPF axis can inhibit also of FoxM1 by several oncogenic protein and signalling path- ways, such as c-Myc, Ras, and PI3K/AKT [72]. PCa metastatization [86]. Overall, these considerations identified FoxM1 as a Hereafter, we discussed the role of FoxM1 in the four tumor types where we found statistically significant associa- potential anticancer therapeutic target. Unfortunately, the tion of FoxM1 expression and poor prognosis. druggability of FoxM1 remains a big challange because of The impact of FoxM1 in breast cancer progression is the lack of substrate-binding pockets and hydrophobic sur- widely demonstrated. Indeed, its high level has been corre- faces [72, 87]. Several in vitro studies proposed RNA interfer- ence (RNAi) as a strategy to knockdown FoxM1, either alone lated with large tumor size, lymphovascular invasion, lymph- node metastases, and high stage. Two independent studies or in combination with ROS inducers, in order to provoke carried out on ER+ patients, reported that low FoxM1 ROS-mediated cell death [82]. Some studies reported that proteasome inhibitors, including bortezomib or thiostrepton, expression, compared to high FoxM1 expression, is associ- ated to better survival. Another study proved a positive cor- directly reduce both FoxM1 expression and its transcrip- tional activity with the same efficacy as that obtained by relation between HER2 status and FoxM1 expression in breast cancer tissue compared to normal breast counterpart FoxM1 silencing [82, 88]. This latter approach is very [73–75], suggesting that FoxM1 is a downstream target of promising, considering that bortezomib is already in clinical practice to treat multiple myeloma, and that RNAi treatment, HER2 and could be used as a marker of HER2 overexpres- sion. However, molecular basis underling the described roles so far, is not a reasonable therapeutic approach in patients [50, 82, 88]. Thus, bortezomib treatment has been proposed of FoxM1 in cancer progression still needs to be clarified and Oxidative Medicine and Cellular Longevity 9 transiently increased cytoplasmic Trx1 oxidation by andro- as effective therapeutic strategy in highly expressing FoxM1 solid tumor, also in association with ROS inducers. gen but decreased Trx1 activities with the progression of prostate cancer, despite high levels of Trx1 protein expres- sion in cancer cells [103]. The role of TrxR1 in dysplastic 6. Thioredoxin, Glutathione Peroxidase, and transformation has been pointed out in human breast epithe- lial cells, triggered by chronic oxidative stress [104]. In addi- Superoxide Dismutase Families as Mediators tion, Trx1 has been proposed as serum biomarker for either of Carcinogenesis early diagnosis or prognosis of breast cancer in association with CEA and CA15-3 [105]. In non-small-cell lung cancer, Thioredoxin system, composed of thioredoxin reductase (TrxR), thioredoxin (Trx), and NADPH, senses and responds Trx1 is able to modulate transcription of cyclooxygenase-2 via hypoxia-inducible factor- (HIF-) 1α [106]. It is actually to oxidative stress and modulates the redox status by scaveng- ing ROS and by regulating several redox enzymes and signal- worth to mention that many human cancers have low levels ing proteins. Mammalian genomes encode two main Trx of thioredoxin-binding protein-2 (TBP-2), a Trx regulator which is able to bind Trx, blocking its reducing activity. systems: Trx1 and Trx reductase (TrxR) 1, which together con- stitutes the cytosolic system; Trx2 and TrxR2, which are local- These mechanisms have been identified as druggable: histone ized inmitochondria (a Trx3 isoformhas been also reported, as deacetylase inhibitors (HDACi) have been demonstrated to a testis-specific form, mainly expressed in male germ cells and upregulate TBP-2 in various transformed cells, associated associated to reproductive disorders) [89]. Trx1 reducing with a decrease in Trx levels [102]. Recently, Park and colleagues observed that TrxR2 is a power allows the transfer of two electronsfrom its dithiol motif to an acceptor, then the oxidized disulfide form of the enzyme novel binding protein for ribonucleotide reductase small sub- is recycled to the dithiol form by TrxR1, thereby oxidizing one unit p53R2, which is involved in nuclear and mitochondrial molecule of NADPH. DNA replication and repair, stimulating the enzymatic activ- Interestingly, our analysis revealed that TXNRD1, the ity of TrxR in vitro. Their findings also suggest that p53R2 acts as a positive regulator of TrxR2 activity in the mitochon- gene encoding TrxR1, is upregulated and correlates with bad prognosis in pancreatic, colon, HNSCC, lung, prostate, dria both under normal physiological conditions and during and breast cancers. Trx1 enzyme has been shown to regulate the cellular response to DNA damage [107]. NF-κB, playing opposite roles, depending to its intracellular Although STRING analyses highlighted glutathione localization: overexpression of Trx in cytoplasm reduced peroxidases (GPx) as one of the main family involved in oxidative stress adaptation, we found high heterogeneity in NF-κB activity, blocking the degradation of the NF-κB inhib- itor IκB; in the nucleus, Trx directly reduces the cysteine(s) of the disregulation of GPx family members among the tumor NF-κB allowing the NF-κB-dependent gene expression [90]. types we have investigated (Figure 5). GPx reduces either free hydrogen peroxide to water or lipid hydroperoxides to their Following NF-κB stimuli, such as UVB irradiation and TNFα treatment, Trx quickly translocates from the cytoplasm into corresponding alcohols. So far, eight different isoforms of GPx, 1 to 8, have been identified in humans, carrying differ- the nucleus. Trx1 has also been reported as a secreted protein by normal and neoplastic cells [91], but not via exosomes ent affinities for their substrates and different localizations. [92]. Notably, Trx-increased secretion contributed to high GPx1, found in the cytoplasm of mammalian cells, is mainly able to target the hydrogen peroxide, while GPx4 showed ROS production in cisplatin-resistant lung tumors, both in vitro and in vivo [93]. high affinity for lipid hydroperoxides. GPx2 is an intestinal and extracellular enzyme, while GPx3 is extracellularly Trx1 itself is regulated both by hypoxia and by oxidative stress conditions via binding of NRF2 to an antioxidant secreted [99]. responsive element in the Trx promotor [94]. Moreover, GPx1 allelic loss or polymosphisms have been known for years to contribute to both lung [108] and breast Trx1 complex functions as a molecular switch turning the cellular redox state into kinase signaling. Thus, the system cancers [109]. Interestingly, in HNSCC cancer, almost all the isoforms showed low expression (Figure 3). In agree- is able to regulate DNA synthesis, cell proliferation [95, 96], apoptosis, and transcription. In details, the reduced form of ment, a decrease in GPx activity accompanied by SOD Trxs binds to apoptosis signal-regulating kinase 1 (ASK1) and CAT decrease as well as higher levels of oxidative DNA damage was found in HNSCC patients compared and inhibits its activity to prevent stress- and cytokine- induced apoptosis; when Trx is oxidized, it dissociates from to healthy donors [110]. An increase of both Trx and GSH metabolism is a mech- ASK1 and apoptosis is stimulated [97–100]. The impact of Trx1 intracellular localization on its role may be taken into anism widely implicated in the resistance of cancer cells to account especially in tumors (as colon and prostate) where chemotherapy. Loss of TXNRD1 makes tumors highly sus- ceptible to pharmacological GSH deprivation, and concomi- a low expression of TXNRD1 correlates to poor patient out- come (as described in Figure 3). In fact, although increased tant inhibition of both GSH and TxrR systems was recently Trx1 protein expression has been associated to hypoxic proposed as an anticancer strategy [18, 111]. Recently, regions of certain tumours, tumor grade and chemoresis- Rodman and colleagues demonstrated that depletion of tence, for instance by scavenging ROS species generated by GSH and inhibition of TrxR activity enhanced radiation responses in human breast cancer stem cells by a mechanism various anticancer agents [101, 102], its localization and activity have to be both taken into account. In prostate can- involving thiol-dependent oxidative stress [112]. Further- cer, Shan and colleagues identified constitutive nuclear and more, Scarbrough and colleagues reported that simultaneous 10 Oxidative Medicine and Cellular Longevity as prosurvival mechanisms associated with resistance to GSH/Trx inhibition sensitizes human breast and prostate cancer cells to 2DG + 17AAG-mediated killing [113]. chemotherapy and tumor relapse [133]. Among the most important antioxidant enzymes, it is Few studies reported the behavior of cancer stem cells in also important to highlight the role of SOD. SOD is able to oxidative stress condition, but notably in contrast to their convert the superoxide (O ) radical into either oxygen (O ) normal stem cell counterparts, cancer stem cells are charac- 2 2 or the less reactive hydrogen peroxide (H O ) which can then terized by increased ROS levels, reduced oxidative damage, 2 2 be removed by CAT, GPx, or TPx. Among the three major and thus longer survival [134, 135]. For example, Im and col- families of SOD, those we single out in humans are the cop- leagues showed that significantly higher ROS levels were per and zinc (Cu-Zn) SOD1, whose localization is in cytosol, observed in the supernatant of glioblastoma cells, grown in nucleus, peroxisome, and intermembrane space of the serum-free sphere medium, either in polystyrene-treated tis- sue culture plates or in nonadherent plates. Moreover, it has mitochondria [114], the mitochondrial enzyme manganese SOD2 (MnSOD), and the (Cu-Zn) extracellular SOD3. been also shown that ROS is a critical factor for maintaining SOD enzymes are able to exert a strong antioxidant activity. stemness, regulating the expression of the transcription fac- In a recent study, Elchuri and colleagues observed that mice tor SOX-2 [136]. This can be due to a combination of mech- deficient in CuZn SOD1 (which contributes to the majority anisms that arise in the tumor, such as modulation of (1) of cellular SOD activity [115]) showed a reduced lifespan and multiple antioxidative enzyme systems [137] or (2) redox- increased incidence of neoplastic changes in the liver [116]. sensitive signaling pathways, as NRF2, NF-κB, c-Jun, and Conversely, it has been also observed by several authors that HIFs, leading to the increased expression of antioxidant SOD1 overexpression makes tumor cells resistant to oxidative molecules [5]. stress and chemotherapy [117]. Increased expression and The higher ROS levels in CSC could be associated with activity of MnSOD has been correlated with cancer aggressive- lower basal expression of ROS-scavenging systems, such as ness in several tumors and through different pathways [118]. SODs, CAT, GPx, and TPx, compared to normal stem cells. Recently, dysregulation of MnSOD function has been linked In this regard, Yang et al. published those nonglioma stem cells to an acetylation-mediated impairment [119, 120] which trig- which displayed significantly lower basal GPx1 expression and gers an increase in oxidative stress, leading to AKT activation activity than glioma stem cells and that miR-153/NRF2/GPx1 via oxidative inactivation of PTEN [119]. MnSOD acetylation pathway plays an important role in regulating radiosensitivity (and activity) is regulated by the deacetylase Sirt3, a mito- and stemness of glioma stem cells via ROS [138]. chondrial fidelity protein. Interestinlgly, Zou et al. showed Due to the growing body of studies focused on the differ- ential modulation of redox-sensitive signaling pathways (as that loss of Sirt3 results in endocrine therapy resistance of human luminal B breast cancer [120]. In agreement, we summarized in Figure 6) in CSC subpopulation, compared and others demonstrated that HDAC inhibition increases to cancer cells or normal stem cells, in this paraghraph we MnSOD protein expression in both solid and haematological discuss the relevance of the ROS-related pathways modulated diseases [121, 122]. in CSC phenotype. Overall, similar to FoxM1, the described antioxidant In hypoxic environments, limited amount of oxygen systems represent putative good targets to improve therapeu- leads to metabolic switches in both normal and malignant tical oxidative stress-dependent strategies. In details, several cells by HIFs. Paradoxically, recent studies have shown that recent efforts have focused on the targeting of Trx/TrxR CSC exhibit high HIF activity in normoxic environments system [123–130]. Moreover, increasing evidences on a puta- and that HIF activity is critical in the maintenance of CSC as well as in the differentiation [139]. In agreement, Wang tive key role of HDAC inhibitors in the modulation of these pathways may deserve further investigations. In this regard, et al. found that overexpression of stem cell factor in hepato- our recent study on the effect of HDACi in regulating cellular carcinoma is regulated by hypoxic conditions NRF2/Keap1 pathway is of interest, considering the interplay through a selective HIF2α-dependent mechanism which between this pathway and thioredoxin [7]. promotes metastasis [140]. Several studies showed that HIF factors can enhance CSC population growth by modulating Notch signaling pathway 7. Oxidative Stress and Cancer Stem Cells in glioma [141], Hippo pathway through direct stabilization In the multitude of morphological, functional, and respon- of TAZ in breast cancer [142], Ras-ERK-ELK3 in hepatocel- sive cancer cells, a subset of the so-called “cancer stem cells” lular cancer, hypoxia-NOTCH1-SOX2 in ovarian cancer [143], and IL6-HIF1α in non-small-cell lung cancer [144]. (CSC), carrying peculiar features, was identified almost ten years ago in solid cancers [131]. However, the name CSC is Additionally, Yang et al. established that gastric CSC exhib- not referred to an origin from normal stem counterpart but ited a marked increase in HIF1α expression and increased rather represents a specific population that displays some migration and invasion capabilities compared with the nor- exceptional properties normally attributed to stem cells. Spe- moxic control upon hypoxia treatment. Also HIF-1α was responsible for activating EMT via increased expression of cific features, like hierarchical differentiation, self-renewal, enhanced invasive capacity, metastatic proficiency, and the transcription factor Snail in gastric CSC [145]. NF-κB is also related to hypoxia and HIF1α induction. It tumorigenicity, make CSC critical for tumor initiation and growth [132], while CSC elevated apoptosis resistance, has been shown that inhibition of NFκB signaling promoted drug-efflux pumps, enhanced DNA repair efficiency, detoxi- a significant reduction in the hypoxia-driven expansion of + −/low CD44 fication enzyme expression, and quiescence are all identified CD24 CSC which was due to increased CD24 Oxidative Medicine and Cellular Longevity 11 ROS ROS ROS ROS Oxidative damage to proteins Oxidative damage to DNA Increased lipid peroxidation Redox adaptation in cancer stem cells Stress pathways Stress pathways alterations alterations Antioxidative enzymes activation FoxO family high expression HIF family high expression (i) SODs; (i) critical mediators of the cellular (i) modulate Notch signaling pathway; (ii) catalase; responses tooxidative stress and (ii) induce and stabilize TAZ, a Hippo pathway (iii) GPx; effector molecule; (iv) peroxiredoxin. have been implicated in many of (iii) induced by ELK3 and IL-6 promoting stemness. ROS-regulated processes; (ii) it is strongly related to WNT NRF2 high expression Adaptative mechanisms activation -pathway modulation. (i) mediate drug resistance by modulating ABCG2 trasporter, and anti-apoptotic factors (es. Bcl-2, Bmi1). block of mRNA translation (stress granule (i) Hippo pathway high expression formation); c-Jun high expression (ii) quiescence induction (high p21expression); (i) molecular switch controlling in (iii) cell death inhibition by modulation of anti cellular differentiation and stem (i) activation of JNK/c-Jun/Notch1 signaling upon stress. apoptotic factors (Bcl-2, Bmi1); cell renewal. (iv) DNA repair activation. HIF1𝛼-YB1-G3BP1 high expression p53 loss led to an early expansion of mammary stem/progenitor (i) (i) in stress condition is activated to Drug resistance cells. preserve cellular homeostasis. (i) modulation of ATP-binding cassette ??? NF-𝜅B activated trasporters; (i) promote the hypoxia-driven expansion of CSC (ii) epithelial to mesenchymal transition by population through Aurora-A. molulation of Snail. Figure 6: Redox stress in cancer stem cells. The persistent production of abnormally large amounts of ROS induced the mechanism of redox adaptation that, in turn, is translated in a various alteration in stress signaling. Here, we reported both known and hypothesized modulated pathways. extracellular matrix components and were enriched in can- expression in breast cancer models [146]. Similarly, Aurora A kinase which can activate NF-κB pathway has been found cer stem cells [154]. highly expressed in ovarian CSC [147]. Moreover, Xie et al. found that knockdown of JNK1 NRF2 represents another antioxidant system involved in or JNK2 or treatment with JNK-IN-8, an adenosine the maintenance of quiescence as well as in the determina- triphosphate-competitive irreversible pan-JNK inhibitor, sig- tion of differentiation fate in normal stem cells, as described nificantly reduced cell proliferation, the ALDH1+ and CD44 and reviewed by Ryoo et al. [148]. For example, NRF2- +/CD24- CSC subpopulations, and mammosphere forma- deficient mice showed defective stem cell function. Indeed, tion, indicating that JNK family promotes CSC self-renewal haematopoietic stem cell, derived from those mice, dispayed and maintenance in triple-negative breast cancer [155]. lower levels of prosurvival cytochines and exibited spontane- However, other factors could be implicated in CSC ous apoptosis related to wild-type cells [149]. capability to adapt high level of intracellular ROS and would Recently, several studies showed that high levels of NRF2 be very interesting to better define them as potential therapeu- are related to CSC survival and anticancer drug resistance in tic targets, mostly because many anticancer drugs increase HNSCC, cervical, breast, and ovarian cancers [150–153]. intracellular ROS levels. Notably, it was reported that NRF2 overexpression is related In this regard, the transcription factors FoxO1, FoxO3a, to an induction of ATP-binding cassette trasporters and thus and FoxO4 are critical mediators of the cellular responses drug resistance mechanisms. Other described redox- to oxidative stress and have been implicated in many of signaling pathway implicated in redox regulation in CSC ROS-regulated processes [156]. It is also known that FoxO competes with TCF for the same binding site of β-catenin could be c-Jun and/or p53 and NF-κB and FoxO family. In details, Chiche et al. showed that the loss of p53 in and suppresses β-catenin-TCF signaling toward prolifera- K5ΔNβcat (βcat activated) mice led to an early expansion tion, thus attenuating WNT-mediated signaling activities. of mammary stem/progenitor cells and accelerated the for- Also, FoxO factors reduce mitochondrial output to prevent mation of triple-negative breast cancers. In particular, p53- excess ROS production through inhibition of c-Myc function deficient tumors expressed high levels of integrins and and alter the hypoxia response [157]. Described mechanisms Hypothesized mechanisms 12 Oxidative Medicine and Cellular Longevity redox homeostasis genes, to guarantee the development of Another candidate is the Hippo pathway, which acts as a molecular switch controlling in cellular differentiation and precision medicine-based approaches in selected subgroups stem cell renewal but is also modulated in stress condition of cancer patients. Further mechanistic studies are needed to and is described as highly mutated in cancer. Lehtinen and identify either new compounds or molecules to be reposi- colleagues elegantly demonstrated the activation of Mst1, a tioned, in order to target the described redox pathways. serine/threonine kinase activated in the Hippo cascade, upon oxidative stress induced by exposure to increasing concentra- Conflicts of Interest tions of exogenous H O . This was accompanied by phos- 2 2 phorylation of the transcription factor FoxO3a at S207, No potential conflicts of interest were disclosed. thereby disrupting its association with 14-3-3 binding pro- tein and leading to its nuclear localization and transcriptional Acknowledgments activation of the BH3- only Bcl-2 protein, Bim, which triggered neuronal apoptosis [158]. This study was partially supported by the following One of the first mechanisms modulated upon stress con- Research Grant to Alfredo Budillon: Italian Ministry of dition is messenger RNA translation, likely as a mean to limit Health (RF-2011-02346914). energy demanding protein synthesis, leading to stress gran- ule (SG) formation in cancer cells. Many evidences suggest that altered mRNA translational control is a critical factor References in cancer progression, and in this regard, a new axis has been [1] L. A. Sena and N. S. Chandel, “Physiological roles of mito- described. In details, Somasekharan et al. showed that under chondrial reactive oxygen species,” Molecular Cell, vol. 48, stress condition, a YB1, nuclease-sensitive element-binding no. 2, pp. 158–167, 2012. protein 1, facilitates tumor metastasis through two mecha- [2] J. G. Gill, E. Piskounova, and S. J. Morrison, “Cancer, oxida- nisms: first, it directly binds to HIF1α that drives stress adap- tive stress, and metastasis,” Cold Spring Harbor Symposia on tation and metastatic capacity in vivo; second, YB1 mediates Quantitative Biology, vol. 81, pp. 163–175, 2016. formation of cytosolic SGs through translational activation of [3] M.Tafani,L.Sansone, F.Limanaetal., “The interplay of reactive G3BP1, a SG nucleator [159]. Accumulating evidences sug- oxygen species, hypoxia, inflammation, and sirtuins in cancer gest that SG formation is protective against stress-induced initiation and progression,” Oxidative Medicine and Cellular cell damage and death [160], and few studies suggested SG Longevity, vol. 2016, Article ID 3907147, 18 pages, 2016. implication in cancer biology [161, 162]. [4] C. Gorrini, I. S. Harris, and T. W. Mak, “Modulation of oxidative stress as an anticancer strategy,” Nature Reviews 8. Conclusions and Future Perspectives Drug Discovery, vol. 12, no. 12, pp. 931–947, 2013. [5] D. Trachootham, J. Alexandre, and P. Huang, “Targeting The idea that the oxidative stress modulation has a crucial cancer cells by ROS-mediated mechanisms: a radical thera- role in cancer cells to promote proliferation, adaptation, peutic approach?,” Nature Reviews Drug Discovery, vol. 8, and resistance to therapy is now widely accepted [7, 12, 13]. no. 7, pp. 579–591, 2009. Thus, modulating redox regulatory mechanisms represents [6] D. Hernandez-Garcia, C. D. Wood, S. Castro-Obregón, and an attractive therapeutic strategy. However, to date, the oxi- L. Covarrubias, “Reactive oxygen species: a radical role in dative stress-related therapeutic strategies evaluated in pre- development?,” Free Radical Biology and Medicine, vol. 49, no. 2, pp. 130–143, 2010. clinical and clinical studies did not produce homogenous [7] A. Leone, M. S. Roca, C. Ciardiello et al., “Vorinostat syner- results, due to several variables associated to ROS generation gizes with EGFR inhibitors in NSCLC cells by increasing and redox adaptation mechanisms. ROS via up-regulation of the major mitochondrial porin Furthermore, the identification of tumor-type specific VDAC1 and modulation of the c-Myc-NRF2-KEAP1 oxidative stress gene profiles and how they could predict pathway,” Free Radical Biology and Medicine, vol. 89, prognosis still represent critical challenges. Thanks to the pp. 287–299, 2015. increasing availability of cancer gene expression profile, [8] G. B. Waypa, J. D. Marks, R. Guzy et al., “Hypoxia triggers mutation, epigenetic, and survival data from the TGCA data- subcellular compartmental redox signaling in vascular set, it was possible to use bioinformatics to screen the role of smooth muscle cells,” Circulation Research, vol. 106, no. 3, oxidative stress genes from a publicly available signature in pp. 526–535, 2010. large cohorts of several solid cancer patients [52, 53]. [9] https://www.qiagen.com/de/shop/pcr/primer-sets/rt2-profile The TCGA database provides correlative evidences r-pcr-arrays/?catno=PAHS-065Z#geneglobe. suggesting the involvement of the FoxM1, thioredoxin, [10] D. Hanahan and R. A. Weinberg, “Hallmarks of cancer: the superoxide-dismutase, and glutathione pathways as princi- next generation,” Cell, vol. 144, no. 5, pp. 646–674, 2011. pally and commonly modulated in breast, lung, HNSCC, [11] E. Panieri and M. M. Santoro, “ROS homeostasis and metab- pancreatic, prostate, and colon cancers. The differential olism: a dangerous liason in cancer cells,” Cell Death & expression levels of each gene observed in different settings Disease, vol. 7, no. 6, article e2253, 2016. revealed a precise spatial context where redox alterations [12] T. Fiaschi and P. Chiarugi, “Oxidative stress, tumor microen- may promote genome instability or redox adaptation. vironment, and metabolic reprogramming: a diabolic For this reason, tumors should be classified into subclasses liaison,” International Journal of Cell Biology, vol. 2012, based on different oxidative stress alterations occurring in Article ID 762825, 8 pages, 2012. Oxidative Medicine and Cellular Longevity 13 [28] G. S. Omenn, G. E. Goodman, M. D. Thornquist et al., “Risk [13] F. Bruzzese, B. Pucci, M. R. Milone et al., “Panobinostat synergizes with zoledronic acid in prostate cancer and multi- factors for lung cancer and for intervention effects in CARET, ple myeloma models by increasing ROS and modulating the beta-carotene and retinol efficacy trial,” Journal of the mevalonate and p38-MAPK pathways,” Cell Death & Disease, National Cancer Institute, vol. 88, no. 21, pp. 1550–1559, vol. 4, article e878, 2013. [14] S. F. Rodrigues and D. N. Granger, “Blood cells and endothe- [29] E. A. Klein, I. M. Thompson, C. M. Tangen et al., “Vitamin E and the risk of prostate cancer: the selenium and vitamin E lial barrier function,” Tissue Barriers, vol. 3, no. 1-2, article e978720, 2015. cancer prevention trial (SELECT),” JAMA, vol. 306, no. 14, pp. 1549–1556, 2011. [15] J. Kim, J. Kim, and J. S. Bae, “ROS homeostasis and metabo- [30] C. Jacobs, B. Hutton, T. Ng, R. Shorr, and M. Clemons, “Is lism: a critical liaison for cancer therapy,” Experimental & there a role for oral or intravenous ascorbate (vitamin C) in Molecular Medicine, vol. 48, no. 11, article e269, 2016. treating patients with cancer? A systematic review,” The [16] S. S. Sabharwal and P. T. Schumacker, “Mitochondrial ROS Oncologist, vol. 20, no. 2, pp. 210–223, 2015. in cancer: initiators, amplifiers or an Achilles’ heel?,” Nature [31] D. B. Longley, T. Latif, J. Boyer, W. L. Allen, P. J. Maxwell, Reviews Cancer, vol. 14, no. 11, pp. 709–721, 2014. and P. G. Johnston, “The interaction of thymidylate synthase [17] G. M. DeNicola, F. A. Karreth, T. J. Humpton et al., expression with p53-regulated signaling pathways in tumor “Oncogene-induced Nrf2 transcription promotes ROS detox- cells,” Seminars in Oncology, vol. 30, no. 3, Supplement 6, ification and tumorigenesis,” Nature, vol. 475, no. 7354, pp. 3–9, 2003. pp. 106–109, 2011. [32] T. Yoshida, S. Goto, M. Kawakatsu, Y. Urata, and T. S. Li, [18] I. S. Harris, A. E. Treloar, S. Inoue et al., “Glutathione and “Mitochondrial dysfunction, a probable cause of persistent thioredoxin antioxidant pathways synergize to drive cancer oxidative stress after exposure to ionizing radiation,” Free initiation and progression,” Cancer Cell, vol. 27, no. 2, Radical Research, vol. 46, no. 2, pp. 147–153, 2012. pp. 211–222, 2015. [33] F. Bruzzese, M. Rocco, S. Castelli, E. Di Gennaro, A. Desideri, [19] A. Kuehne, H. Emmert, J. Soehle et al., “Acute activation of and A. Budillon, “Synergistic antitumor effect between oxidative pentose phosphate pathway as first-line response vorinostat and topotecan in small cell lung cancer cells is to oxidative stress in human skin cells,” Molecular Cell, mediated by generation of reactive oxygen species and DNA vol. 59, no. 3, pp. 359–371, 2015. damage-induced apoptosis,” Molecular Cancer Therapeutics, [20] V. P. Sukhatme and B. Chan, “Glycolytic cancer cells lacking vol. 8, no. 11, pp. 3075–3087, 2009. 6-phosphogluconate dehydrogenase metabolize glucose to [34] R. R. Rosato, J. A. Almenara, and S. Grant, “The histone induce senescence,” FEBS Letters, vol. 586, no. 16, deacetylase inhibitor MS-275 promotes differentiation or pp. 2389–2395, 2012. apoptosis in human leukemia cells through a process regu- [21] S. M. Wormann, L. Song, J. Ai et al., “Loss of P53 function lated by generation of reactive oxygen species and induction activates JAK2-STAT3 signaling to promote pancreatic of p21CIP1/WAF1 1,” Cancer Research, vol. 63, no. 13, tumor growth, stroma modification, and gemcitabine pp. 3637–3645, 2003. resistance in mice and is associated with patient survival,” [35] X. Y. Pei, Y. Dai, and S. Grant, “Synergistic induction of Gastroenterology, vol. 151, no. 1, article e12, pp. 180–193, oxidative injury and apoptosis in human multiple myeloma cells by the proteasome inhibitor bortezomib and histone [22] A. Alexander, S. L. Cai, J. Kim et al., “ATM signals to TSC2 in deacetylase inhibitors,” Clinical Cancer Research, vol. 10, the cytoplasm to regulate mTORC1 in response to ROS,” no. 11, pp. 3839–3852, 2004. Proceedings of the National Academy of Sciences of the United [36] C. Yu, B. B. Friday, J. P. Lai et al., “Abrogation of MAPK and States of America, vol. 107, no. 9, pp. 4153–4158, 2010. Akt signaling by AEE788 synergistically potentiates histone [23] A. Alexander, J. Kim, and C. L. Walker, “ATM engages the deacetylase inhibitor-induced apoptosis through reactive TSC2/mTORC1 signaling node to regulate autophagy,” oxygen species generation,” Clinical Cancer Research, Autophagy, vol. 6, no. 5, pp. 672-673, 2010. vol. 13, no. 4, pp. 1140–1148, 2007. [24] N. McCabe, S. M. Walker, and R. D. Kennedy, “When the [37] S. Bhalla, S. Balasubramanian, K. David et al., “PCI-24781 guardian becomes the enemy: targeting ATM in PTEN- induces caspase and reactive oxygen species-dependent apo- deficient cancers,” Molecular and Cellular Oncology, vol. 3, ptosis through NF-kappaB mechanisms and is synergistic no. 1, article e1053595, 2016. with bortezomib in lymphoma cells,” Clinical Cancer Research, vol. 15, no. 10, pp. 3354–3365, 2009. [25] J. S. Chan, M. J. Tan, M. K. Sng et al., “Cancer-associated fibroblasts enact field cancerization by promoting extratu- [38] N. Koshikawa, J. I. Hayashi, A. Nakagawara, and K. Takenaga, moral oxidative stress,” Cell Death & Disease, vol. 8, no. 1, “Reactive oxygen species-generating mitochondrial DNA article e2562, 2017. mutation up-regulates hypoxia-inducible factor-1alpha gene transcription via phosphatidylinositol 3-kinase-Akt/protein [26] Y. L. Qiao, S. M. Dawsey, F. Kamangar et al., “Total and kinase C/histone deacetylase pathway,” The Journal of cancer mortality after supplementation with vitamins and Biological Chemistry, vol. 284, no. 48, pp. 33185–33194, minerals: follow-up of the Linxian general population nutrition intervention trial,” Journal of the National Cancer Institute, vol. 101, no. 7, pp. 507–518, 2009. [39] I. M. Wolf, Z. Fan, M. Rauh et al., “Histone deacetylases inhibition by SAHA/Vorinostat normalizes the glioma [27] G. S. Omenn, G. Goodman, M. Thornquist et al., “Chemo- microenvironment via xCT equilibration,” Scientific Reports, prevention of lung cancer: the beta-carotene and retinol vol. 4, p. 6226, 2014. efficacy trial (CARET) in high-risk smokers and asbestos- exposed workers,” IARC Scientific Publications, vol. 136, [40] K. F. Hui, P. L. Yeung, and A. K. Chiang, “Induction of pp. 67–85, 1996. MAPK- and ROS-dependent autophagy and apoptosis in 14 Oxidative Medicine and Cellular Longevity [55] J. Budczies, M. Bockmayr, C. Denkert et al., “Classical pathol- gastric carcinoma by combination of romidepsin and borte- zomib,” Oncotarget, vol. 7, no. 4, pp. 4454–4467, 2016. ogy and mutational load of breast cancer - integration of two worlds,” Journal of Pathology: Clinical Research, vol. 1, no. 4, [41] P. H. Liao, H. H. Hsu, T. S. Chen et al., “Phosphorylation of pp. 225–238, 2015. cofilin-1 by ERK confers HDAC inhibitor resistance in hepatocellular carcinoma cells via decreased ROS-mediated [56] R. M. Chabanon, M. Pedrero, C. Lefebvre, A. Marabelle, mitochondria injury,” Oncogene, 2016. J. C. Soria, and S. Postel-Vinay, “Mutational landscape and sensitivity to immune checkpoint blockers,” Clinical [42] M. R. Makena, B. Koneru, T. H. Nguyen, M. H. Kang, and Cancer Research, vol. 22, no. 17, pp. 4309–4321, 2016. C. P. Reynolds, “Reactive oxygen species-mediated syner- gism of fenretinide and romidepsin in preclinical models [57] N. B. Jamieson and A. V. Maker, “Gene-expression profiling of T-cell lymphoid malignancies,” Molecular Cancer to predict responsiveness to immunotherapy,” Cancer Gene Therapeutics, 2017. Therapy, 2016. [58] S. Meucci, U. Keilholz, I. Tinhofer, and O. A. Ebner, [43] J. R. Kirshner, S. He, V. Balasubramanyam et al., “Elesclomol induces cancer cell apoptosis through oxidative stress,” “Mutational load and mutational patterns in relation to age in head and neck cancer,” Oncotarget, vol. 7, no. 43, Molecular Cancer Therapeutics, vol. 7, no. 8, pp. 2319–2327, 2008. pp. 69188–69199, 2016. [59] S. J. Murphy, S. N. Hart, J. F. Lima et al., “Genetic alterations [44] S. O'Day, R. Gonzalez, D. Lawson et al., “Phase II, random- ized, controlled, double-blinded trial of weekly elesclomol associated with progression from pancreatic intraepithelial neoplasia to invasive pancreatic tumor,” Gastroenterology, plus paclitaxel versus paclitaxel alone for stage IV metastatic melanoma,” Journal of Clinical Oncology, vol. 27, no. 32, vol. 145, no. 5, pp. 1098–1109, 2013, e1. pp. 5452–5458, 2009. [60] J. P. McCrow, D. C. Petersen, M. Louw et al., “Spectrum of mitochondrial genomic variation and associated clinical pre- [45] S. J. O'Day, A. M. Eggermont, V. Chiarion-Sileni et al., “Final sentation of prostate cancer in South African men,” Prostate, results of phase III SYMMETRY study: randomized, double- vol. 76, no. 4, pp. 349–358, 2016. blind trial of elesclomol plus paclitaxel versus paclitaxel alone as treatment for chemotherapy-naive patients with advanced [61] F. Gelsomino, M. Barbolini, A. Spallanzani, G. Pugliese, and melanoma,” Journal of Clinical Oncology, vol. 31, no. 9, S. Cascinu, “The evolving role of microsatellite instability in pp. 1211–1218, 2013. colorectal cancer: a review,” Cancer Treatment Reviews, vol. 51, pp. 19–26, 2016. [46] S. Takahashi, “Combination therapy with arsenic trioxide for hematological malignancies,” Anti-Cancer Agents in [62] H. S. Kim, S. L. Ullevig, D. Zamora, C. F. Lee, and R. Asmis, Medicinal Chemistry, vol. 10, no. 6, pp. 504–510, 2010. “Redox regulation of MAPK phosphatase 1 controls mono- cyte migration and macrophage recruitment,” Proceedings [47] K. L. Kuo, W. C. Lin, I. L. Ho et al., “2-methoxyestradiol of the National Academy of Sciences of the United States of induces mitotic arrest, apoptosis, and synergistic cytotoxicity America, vol. 109, no. 41, pp. E2803–E2812, 2012. with arsenic trioxide in human urothelial carcinoma cells,” PloS One, vol. 8, no. 8, article e68703, 2013. [63] C. Bracalente, I. L. Ibañez, A. Berenstein et al., “Reprogram- ming human A375 amelanotic melanoma cells by catalase [48] S. Bruno, F. Ghiotto, C. Tenca et al., “N-(4-hydroxyphenyl)- overexpression: upregulation of antioxidant genes correlates retinamide promotes apoptosis of resting and proliferating with regression of melanoma malignancy and with malignant B-cell chronic lymphocytic leukemia cells and potentiates progression when downregulated,” Oncotarget, vol. 7, no. 27, fludarabine and ABT-737 cytotoxicity,” Leukemia, vol. 26, pp. 41154–41171, 2016. no. 10, pp. 2260–2268, 2012. [64] S. Waz, T. Nakamura, K. Hirata et al., “Structural and kinetic [49] H. R. Lee, J. M. Cho, D. H. Shin et al., “Adaptive response to studies of the human nudix hydrolase MTH1 reveal the GSH depletion and resistance to L-buthionine-(S,R)-sulfoxi- mechanism for its broad substrate specificity,” The Journal mine: involvement of Nrf2 activation,” Molecular and of Biological Chemistry, vol. 292, no. 7, pp. 2785–2794, 2017. Cellular Biochemistry, vol. 318, no. 1-2, pp. 23–31, 2008. [65] Y. Sun and H. Li, “Functional characterization of SAG/RBX2/ [50] B. Rotblat, T. G. Grunewald, G. Leprivier, G. Melino, and ROC2/RNF7, an antioxidant protein and an E3 ubiquitin R. A. Knight, “Anti-oxidative stress response genes: bioinfor- ligase,” Protein & Cell, vol. 4, no. 2, pp. 103–116, 2013. matic analysis of their expression and relevance in multiple cancers,” Oncotarget, vol. 4, no. 12, pp. 2577–2590, 2013. [66] I. Bae, S. Fan, Q. Meng et al., “BRCA1 induces antioxidant gene expression and resistance to oxidative stress,” Cancer [51] O. F. Kuzu, M. A. Noory, and G. P. Robertson, “The role of Research, vol. 64, no. 21, pp. 7893–7909, 2004. cholesterol in cancer,” Cancer Research, vol. 76, no. 8, pp. 2063–2070, 2016. [67] S. N. Mowla, E. W. Lam, and P. S. Jat, “Cellular senescence and aging: the role of B-MYB,” Aging Cell, vol. 13, no. 5, [52] I. Amelio, P. O. Tsvetkov, R. A. Knight, A. Lisitsa, G. Melino, pp. 773–779, 2014. and A. V. Antonov, “SynTarget: an online tool to test the syn- ergetic effect of genes on survival outcome in cancer,” Cell [68] M. Katoh, M. Igarashi, H. Fukuda, H. Nakagama, and M. Death and Differentiation, vol. 23, no. 5, p. 912, 2016. Katoh, “Cancer genetics and genomics of human FOX family genes,” Cancer Letters, vol. 328, no. 2, pp. 198–206, 2013. [53] A. V. Antonov, “BioProfiling.de: analytical web portal for high-throughput cell biology,” Nucleic Acids Research, [69] C. T. Kwok, M. H. Leung, J. Qin et al., “The forkhead box vol. 39, pp. W323–W327, 2011. transcription factor FOXM1 is required for the maintenance of cell proliferation and protection against oxidative stress in [54] A. V. Antonov, M. Krestyaninova, R. A. Knight, I. human embryonic stem cells,” Stem Cell Research, vol. 16, Rodchenkov, G. Melino, and N. A. Barlev, “PPISURV: a no. 3, pp. 651–661, 2016. novel bioinformatics tool for uncovering the hidden role of specific genes in cancer survival outcome,” Oncogene, vol. 33, [70] I. Wierstra, “The transcription factor FOXM1 (forkhead box no. 13, pp. 1621–1628, 2014. M1): proliferation-specific expression, transcription factor Oxidative Medicine and Cellular Longevity 15 [84] Z. Li, Y. Gao, D. Xie et al., “Activation of vitamin D function, target genes, mouse models, and normal biological roles,”AdvancesinCancerResearch,vol.118,pp.97–398,2013. receptor signaling downregulates the expression of nuclear FOXM1 protein and suppresses pancreatic cancer cell [71] H. J. Park, J. R. Carr, Z. Wang et al., “FoxM1, a critical stemness,” Clinical Cancer Research, vol. 21, no. 4, regulator of oxidative stress during oncogenesis,” The EMBO pp. 844–853, 2015. Journal, vol. 28, no. 19, pp. 2908–2918, 2009. [85] A. Aytes, A. Mitrofanova, C. Lefebvre et al., “Cross-species [72] R. Saba, A. Alsayed, J. P. Zacny, and A. Z. Dudek, “The role of regulatory network analysis identifies a synergistic interac- forkhead box protein M1 in breast cancer progression and tion between FOXM1 and CENPF that drives prostate cancer resistance to therapy,” International Journal of Breast Cancer, malignancy,” Cancer Cell, vol. 25, no. 5, pp. 638–651, 2014. vol. 2016, Article ID 9768183, 8 pages, 2016. [86] S. C. Lin, C. Y. Kao, H. J. Lee et al., “Dysregulation of miR- [73] H. Ahn, J. Sim, R. Abdul et al., “Increased expression of fork- NAs-COUP-TFII-FOXM1-CENPF axis contributes to the head box M1 is associated with aggressive phenotype and metastasis of prostate cancer,” Nature Communications, poor prognosis in estrogen receptor-positive breast cancer,” vol. 7, article 11418, 2016. Journal of Korean Medical Science, vol. 30, no. 4, pp. 390– 397, 2015. [87] M. Halasi and A. L. Gartel, “Targeting FOXM1 in cancer,” Biochemical Pharmacology, vol. 85, no. 5, pp. 644–652, 2013. [74] A. Bergamaschi, Z. Madak-Erdogan, Y. J. Kim, Y. L. Choi, [88] A. L. Gartel, “Suppression of the oncogenic transcription fac- H. Lu, and B. S. Katzenellenbogen, “The forkhead tran- tor FOXM1 by proteasome inhibitors,” Scientifica, vol. 2014, scription factor FOXM1 promotes endocrine resistance Article ID 596528, 5 pages, 2014. and invasiveness in estrogen receptor-positive breast can- cer by expansion of stem-like cancer cells,” Breast Cancer [89] A. Jimenez, W. Zu, V. Y. Rawe et al., “Spermatocyte/sperma- Research, vol. 16, no. 5, p. 436, 2014. tid-specific thioredoxin-3, a novel Golgi apparatus-associated [75] R. E. Francis, S. S. Myatt, J. Krol et al., “FoxM1 is a down- thioredoxin, is a specific marker of aberrant spermatogene- sis,” The Journal of Biological Chemistry, vol. 279, no. 33, stream target and marker of HER2 overexpression in breast cancer,” International Journal of Oncology, vol. 35, no. 1, pp. 34971–34982, 2004. pp. 57–68, 2009. [90] K. Hirota, M. Murata, Y. Sachi et al., “Distinct roles of thior- edoxin in the cytoplasm and in the nucleus. A two-step [76] C. Yang, H. Chen, G. Tan et al., “FOXM1 promotes the mechanism of redox regulation of transcription factor NF- epithelial to mesenchymal transition by stimulating the kappaB,” The Journal of Biological Chemistry, vol. 274, transcription of slug in human breast cancer,” Cancer Letters, no. 39, pp. 27891–27897, 1999. vol. 340, no. 1, pp. 104–112, 2013. [91] A. Rubartelli, A. Bajetto, G. Allavena, E. Wollman, and R. [77] J. Xue, X. Lin, W. T. Chiu et al., “Sustained activation of Sitia, “Secretion of thioredoxin by normal and neoplastic cells SMAD3/SMAD4 by FOXM1 promotes TGF-beta- through a leaderless secretory pathway,” The Journal of Bio- dependent cancer metastasis,” The Journal of Clinical Investi- logical Chemistry, vol. 267, no. 34, pp. 24161–24164, 1992. gation, vol. 124, no. 2, pp. 564–579, 2014. [92] Y. Manabe, M. Takagi, M. Nakamura-Yamada et al., “Redox [78] A. Ahmad, Z. Wang, D. Kong et al., “FoxM1 down-regulation proteins are constitutively secreted by skeletal muscle,” The leads to inhibition of proliferation, migration and invasion of Journal of Physiological Sciences, vol. 64, no. 6, pp. 401–409, breast cancer cells through the modulation of extra-cellular matrix degrading factors,” Breast Cancer Research and Treatment, vol. 122, no. 2, pp. 337–346, 2010. [93] M. Wangpaichitr, E. J. Sullivan, G. Theodoropoulos et al., “The relationship of thioredoxin-1 and cisplatin resistance: [79] C. T. Karadedou, A. R. Gomes, J. Chen et al., “FOXO3a its impact on ROS and oxidative metabolism in lung cancer represses VEGF expression through FOXM1-dependent cells,” Molecular Cancer Therapeutics, vol. 11, no. 3, and -independent mechanisms in breast cancer,” Oncogene, pp. 604–615, 2012. vol. 31, no. 14, pp. 1845–1858, 2012. [94] Y. C. Kim, H. Masutani, Y. Yamaguchi, K. Itoh, M. Yamamoto, [80] G. N. de Moraes, D. Delbue, K. L. Silva et al., “FOXM1 targets and J. Yodoi, “Hemin-induced activation of the thioredoxin XIAP and survivin to modulate breast cancer survival and gene by Nrf2. A differential regulation of the antioxidant chemoresistance,” Cellular Signalling, vol. 27, no. 12, responsive element by a switch of its binding factors,” The pp. 2496–2505, 2015. Journal of Biological Chemistry, vol. 276, no. 21, pp. 18399– [81] D. K. Yang, C. H. Son, S. K. Lee, P. J. Choi, K. E. Lee, and M. S. 18406, 2001. Roh, “Forkhead box M1 expression in pulmonary squamous [95] J. S. Song, H. H. Cho, B. J. Lee, Y. C. Bae, and J. S. Jung, “Role cell carcinoma: correlation with clinicopathologic features of thioredoxin 1 and thioredoxin 2 on proliferation of human and its prognostic significance,” Human Pathology, vol. 40, adipose tissue-derived mesenchymal stem cells,” Stem Cells no. 4, pp. 464–470, 2009. and Development, vol. 20, no. 9, pp. 1529–1537, 2011. [82] M. Halasi, B. Pandit, M. Wang, V. Nogueira, N. Hay, and A. [96] J. R. Prigge, S. Eriksson, S. V. Iverson et al., “Hepatocyte DNA L. Gartel, “Combination of oxidative stress and FOXM1 replication in growing liver requires either glutathione or a inhibitors induces apoptosis in cancer cells and inhibits xeno- single allele of txnrd1,” Free Radical Biology and Medicine, graft tumor growth,” The American Journal of Pathology, vol. 52, no. 4, pp. 803–810, 2012. vol. 183, no. 1, pp. 257–265, 2013. [97] A. Matsuzawa and H. Ichijo, “Redox control of cell fate by [83] E. Tahmasbpour, M. Ghanei, A. Qazvini, E. Vahedi, and MAP kinase: physiological roles of ASK1-MAP kinase path- Y. Panahi, “Gene expression profile of oxidative stress way in stress signaling,” Biochimica et Biophysica Acta, and antioxidant defense in lung tissue of patients vol. 1780, no. 11, pp. 1325–1336, 2008. exposed to sulfur mustard,” Mutation Research, Genetic Toxicology and Environmental Mutagenesis, vol. 800-801, [98] M. Saitoh, H. Nishitoh, M. Fujii et al., “Mammalian thiore- pp. 12–21, 2016. doxin is a direct inhibitor of apoptosis signal-regulating 16 Oxidative Medicine and Cellular Longevity via oxidative stress,” Free Radical Biology and Medicine, kinase (ASK) 1,” The EMBO Journal, vol. 17, no. 9, pp. 2596– 2606, 1998. vol. 52, no. 2, pp. 436–443, 2012. [99] E.M.Hanschmann,E.M.Hanschmann,J.R.Godoy,C.Berndt, [114] S. Zhang, J. Xue, J. Zheng et al., “The superoxide dismutase 1 C. Hudemann, and C. H. Lillig, “Thioredoxins, glutaredoxins, 3'UTR maintains high expression of the SOD1 gene in cancer andperoxiredoxins—molecularmechanismsandhealthsignif- cells: the involvement of the RNA-binding protein AUF-1,” icance:fromcofactorstoantioxidantstoredoxsignaling,”Anti- Free Radical Biology and Medicine, vol. 85, pp. 33–44, 2015. oxidants&RedoxSignaling,vol.19,no.13,pp.1539–1605,2013. [115] J. D. Crapo, T. Oury, C. Rabouille, J. W. Slot, and L. Y. Chang, [100] J. Lu and A. Holmgren, “Thioredoxin system in cell death “Copper,zinc superoxide dismutase is primarily a cytosolic progression,” Antioxidants & Redox Signaling, vol. 17, protein in human cells,” Proceedings of the National Academy no. 12, pp. 1738–1747, 2012. of Sciences of the United States of America, vol. 89, no. 21, pp. 10405–10409, 1992. [101] A. Mukherjee and S. G. Martin, “The thioredoxin system: a key target in tumour and endothelial cells,” The British Jour- [116] S. Elchuri, T. D. Oberley, W. Qi et al., “CuZnSOD deficiency nal of Radiology, vol. 81, no. 1, pp. S57–S68, 2008. leads to persistent and widespread oxidative damage and hepatocarcinogenesis later in life,” Oncogene, vol. 24, no. 3, [102] P. A. Marks, “Thioredoxin in cancer–role of histone deacety- pp. 367–380, 2005. lase inhibitors,” Seminars in Cancer Biology, vol. 16, no. 6, pp. 436–443, 2006. [117] D. P. Brown, H. Chin-Sinex, B. Nie, M. S. Mendonca, and M. Wang, “Targeting superoxide dismutase 1 to overcome [103] W. Shan, W. Zhong, R. Zhao, and T. D. Oberley, “Thiore- doxin 1 as a subcellular biomarker of redox imbalance in cisplatin resistance in human ovarian cancer,” Cancer Chemo- therapy and Pharmacology, vol. 63, no. 4, pp. 723–730, 2009. human prostate cancer progression,” Free Radical Biology and Medicine, vol. 49, no. 12, pp. 2078–2087, 2010. [118] A. Piecuch, M. Brzozowa-Zasada, B. Dziewit et al., “Immuno- [104] C. Dong, L. Zhang, R. Sun et al., “Role of thioredoxin reduc- histochemical assessment of mitochondrial superoxide dis- tase 1 in dysplastic transformation of human breast epithelial mutase (MnSOD) in colorectal premalignant and malignant cells triggered by chronic oxidative stress,” Scientific Reports, lesions,” Przeglad Gastroenterologiczny, vol. 11, no. 4, vol. 6, article 36860, 2016. pp. 239–246, 2016. [105] B. J. Park, M. K. Cha, and I. H. Kim, “Thioredoxin 1 as a [119] V. Vidimar, D. Gius, D. Chakravarti, S. E. Bulun, J. J. Wei, and J. J. Kim, “Dysfunctional MnSOD leads to redox dysreg- serum marker for breast cancer and its use in combination with CEA or CA15-3 for improving the sensitivity of breast ulation and activation of prosurvival AKT signaling in uter- cancer diagnoses,” BMC Research Notes, vol. 7, p. 7, 2014. ine leiomyomas,” Science Advances, vol. 2, no. 11, article e1601132, 2016. [106] I. Csiki, K. Yanagisawa, N. Haruki et al., “Thioredoxin-1 modulates transcription of cyclooxygenase-2 via hypoxia- [120] X. Zou, C. A. Santa-Maria, J. O'Brien, D. Gius, and Y. Zhu, inducible factor-1alpha in non-small cell lung cancer,” Can- “Manganese superoxide dismutase acetylation and dysregula- cer Research, vol. 66, no. 1, pp. 143–150, 2006. tion, due to loss of SIRT3 activity, promote a luminal B-like breast carcinogenic-permissive phenotype,” Antioxidants & [107] S. J. Park, H. B. Kim, C. Piao et al., “p53R2 regulates thiore- Redox Signaling, vol. 25, no. 6, pp. 326–336, 2016. doxin reductase activity through interaction with TrxR2,” Biochemical and Biophysical Research Communications, [121] L. Bianchi, F. Bruzzese, A. Leone et al., “Proteomic analysis vol. 482, no. 4, pp. 706–712, 2017. identifies differentially expressed proteins after HDAC vorinostat and EGFR inhibitor gefitinib treatments in Hep-2 [108] J. A. Moscow, L. Schmidt, D. T. Ingram, J. Gnarra, B. John- cancer cells,” Proteomics, vol. 11, no. 18, pp. 3725–3742, 2011. son, and K. H. Cowan, “Loss of heterozygosity of the human cytosolic glutathione peroxidase I gene in lung cancer,” [122] R. R. Rosato, S. S. Kolla, S. K. Hock et al., “Histone deace- Carcinogenesis, vol. 15, no. 12, pp. 2769–2773, 1994. tylase inhibitors activate NF-kappaB in human leukemia cells through an ATM/NEMO-related pathway,” The [109] Y. J. Hu and A. M. Diamond, “Role of glutathione peroxidase Journal of Biological Chemistry, vol. 285, no. 13, 1 in breast cancer: loss of heterozygosity and allelic differ- pp. 10064–10077, 2010. ences in the response to selenium,” Cancer Research, vol. 63, no. 12, pp. 3347–3351, 2003. [123] P. V. Raninga, G. Di Trapani, S. Vuckovic, M. Bhatia, and K. F. Tonissen, “Inhibition of thioredoxin 1 leads to apoptosis [110] J. Milonski, H. Zielinska-Blizniewska, J. Olszewski, I. in drug-resistant multiple myeloma,” Oncotarget, vol. 6, no. 17, Majsterek, and M. Mrowicka, “DNA damage and oxidant- pp. 15410–15424, 2015. antioxidant status in blood of patients with head and neck can- cer,” DNA and Cell Biology, vol. 34, no. 3, pp. 213–219, 2015. [124] L. Xie, Z. Luo, Z. Zhao, and T. Chen, “Anticancer and antian- giogenic iron(II) complexes that target thioredoxin reductase [111] P. K. Mandal, M. Schneider, P. Kölle et al., “Loss of thiore- to trigger cancer cell apoptosis,” Journal of Medicinal doxin reductase 1 renders tumors highly susceptible to phar- Chemistry, vol. 60, no. 1, pp. 202–214, 2017. macologic glutathione deprivation,” Cancer Research, vol. 70, no. 22, pp. 9505–9514, 2010. [125] H. Hwang-Bo, J. Cheong, W. J. Kim, Y. H. Yoo, and Y. H. Choi, “Auranofin, an inhibitor of thioredoxin reductase, [112] S. N. Rodman, J. M. Spence, T. J. Ronnfeldt et al., “Enhance- ment of radiation response in breast cancer stem cells by induces apoptosis in hepatocellular carcinoma Hep3B cells by generation of reactive oxygen species,” General Physiology inhibition of thioredoxin- and glutathione-dependent metabolism,” Radiation Research, vol. 186, no. 4, pp. 385– and Biophysics, 2017. 395, 2016. [126] H. L. Ng, X. Ma, E. H. Chew, and W. K. Chui, “Design, [113] P. M. Scarbrough, K. A. Mapuskar, D. M. Mattson, D. Gius, synthesis, and biological evaluation of coupled bioactive W. H. Watson, and D. R. Spitz, “Simultaneous inhibition of scaffolds as potential anticancer agents for dual targeting of glutathione- and thioredoxin-dependent metabolism is dihydrofolate reductase and thioredoxin reductase,” Journal necessary to potentiate 17AAG-induced cancer cell killing of Medicinal Chemistry, vol. 60, no. 5, pp. 1734–1745, 2017. Oxidative Medicine and Cellular Longevity 17 [144] F. Zhang, S. Duan, Y. Tsai et al., “Cisplatin treatment [127] J. Y. Zhu, C. Y. Zhang, J. J. Dai, K. Rahman, and H. Zhang, “Diterpenoids with thioredoxin reductase inhibitory activi- increases stemness through upregulation of hypoxia- ties from Jatropha multifida,” Natural Product Research, inducible factors by interleukin-6 in non-small cell lung pp. 1–6, 2017. cancer,” Cancer Science, vol. 107, no. 6, pp. 746–754, 2016. [128] D. Duan, J. Zhang, J. Yao, Y. Liu, and J. Fang, “Targeting [145] S. W. Yang, Z. G. Zhang, Y. X. Hao et al., “HIF-1alpha thioredoxin reductase by parthenolide contributes to induces the epithelial-mesenchymal transition in gastric can- inducing apoptosis of HeLa cells,” The Journal of Biological cer stem cells through the snail pathway,” Oncotarget, vol. 8, Chemistry, vol. 291, no. 19, pp. 10021–10031, 2016. no. 6, pp. 9535–9545, 2017. [129] J. Zhang, S. Peng, X. Li, R. Liu, X. Han, and J. Fang, “Target- [146] O. Iriondo, M. Rábano, G. Domenici et al., “Distinct breast ing thioredoxin reductase by plumbagin contributes to cancer stem/progenitor cell populations require either inducing apoptosis of HL-60 cells,” Archives of Biochemistry HIF1alpha or loss of PHD3 to expand under hypoxic condi- and Biophysics, vol. 619, pp. 16–26, 2017. tions,” Oncotarget, vol. 6, no. 31, pp. 31721–31739, 2015. [130] A. Schroeder, U. Warnken, D. Röth et al., “Targeting [147] I. Chefetz, J. C. Holmberg, A. B. Alvero, I. Visintin, and G. thioredoxin-1 by dimethyl fumarate induces ripoptosome- Mor, “Inhibition of aurora-a kinase induces cell cycle arrest mediated cell death,” Scientific Reports, vol. 7, article in epithelial ovarian cancer stem cells by affecting NFkB 43168, 2017. pathway,” Cell Cycle, vol. 10, no. 13, pp. 2206–2214, 2011. [131] J. P. Medema, “Cancer stem cells: the challenges ahead,” [148] I. G. Ryoo, S. H. Lee, and M. K. Kwak, “Redox modu- Nature Cell Biology, vol. 15, no. 4, pp. 338–344, 2013. lating NRF2: a potential mediator of cancer stem cell resistance,” Oxidative Medicine and Cellular Longevity, [132] J. E. Visvader and G. J. Lindeman, “Cancer stem cells: current vol. 2016, Article ID 2428153, 14 pages, 2016. status and evolving complexities,” Cell Stem Cell, vol. 10, no. 6, pp. 717–728, 2012. [149] A. A. Merchant, A. Singh, W. Matsui, and S. Biswal, “The redox-sensitive transcription factor Nrf2 regulates murine [133] S. Colak and J. P. Medema, “Cancer stem cells—important players in tumor therapy resistance,” The FEBS Journal, hematopoietic stem cell survival independently of ROS levels,” Blood, vol. 118, no. 25, pp. 6572–6579, 2011. vol. 281, no. 21, pp. 4779–4791, 2014. [134] D. Trachootham, H. Zhang, W. Zhang et al., “Effective elim- [150] B. C. Lu, J. Li, W. F. Yu, G. Z. Zhang, H. M. Wang, and H. M. Ma, “Elevated expression of Nrf2 mediates multidrug resis- ination of fludarabine-resistant CLL cells by PEITC through a redox-mediated mechanism,” Blood, vol. 112, no. 5, tance in CD133+ head and neck squamous cell carcinoma stem cells,” Oncology Letters, vol. 12, no. 6, pp. 4333–4338, pp. 1912–1922, 2008. [135] E. D. Lagadinou, A. Sach, K. Callahan et al., “BCL-2 inhibi- tion targets oxidative phosphorylation and selectively eradi- [151] Y. Jia, J. Chen, H. Zhu, Z. H. Jia, and M. H. Cui, “Aberrantly cates quiescent human leukemia stem cells,” Cell Stem Cell, elevated redox sensing factor Nrf2 promotes cancer stem cell vol. 12, no. 3, pp. 329–341, 2013. survival via enhanced transcriptional regulation of ABCG2 and Bcl-2/Bmi-1 genes,” Oncology Reports, vol. 34, no. 5, [136] C. N. Im, H. H. Yun, H. J. Yoo, M. J. Park, and J. H. Lee, pp. 2296–2304, 2015. “Enhancement of SOX-2 expression and ROS accumulation by culture of A172 glioblastoma cells under non-adherent [152] I. G. Ryoo, B. H. Choi, and M. K. Kwak, “Activation of NRF2 culture conditions,” Oncology Reports, vol. 34, no. 2, by p62 and proteasome reduction in sphere-forming breast pp. 920–928, 2015. carcinoma cells,” Oncotarget, vol. 6, no. 10, pp. 8167–8184, [137] J. Liu and Z. Wang, “Increased oxidative stress as a selective anticancer therapy,” Oxidative Medicine and Cellular [153] T. Mizuno, N. Suzuki, H. Makino et al., “Cancer stem-like Longevity, vol. 2015, Article ID 294303, 12 pages, 2015. cells of ovarian clear cell carcinoma are enriched in the ALDH-high population associated with an accelerated [138] W. Yang, Y. Shen, J. Wei, and F. Liu, “MicroRNA-153/Nrf-2/ scavenging system in reactive oxygen species,” Gynecologic GPx1 pathway regulates radiosensitivity and stemness of gli- Oncology, vol. 137, no. 2, pp. 299–305, 2015. oma stem cells via reactive oxygen species,” Oncotarget, vol. 6, no. 26, pp. 22006–22027, 2015. [154] A. Chiche, M. Moumen, M. Romagnoli et al., “p53 deficiency induces cancer stem cell pool expansion in a mouse model of [139] G. Peng and Y. Liu, “Hypoxia-inducible factors in cancer triple-negative breast tumors,” Oncogene, 2016. stem cells and inflammation,” Trends in Pharmacological Sciences, vol. 36, no. 6, pp. 374–383, 2015. [155] X. Xie, T. S. Kaoud, R. Edupuganti et al., “c-Jun N-terminal kinase promotes stem cell phenotype in triple-negative breast [140] X. Wang, J. Dong, L. Jia et al., “HIF-2-dependent expression cancer through upregulation of Notch1 via activation of of stem cell factor promotes metastasis in hepatocellular c-Jun,” Oncogene, 2016. carcinoma,” Cancer Letters, 2017. [156] P. Storz, “Forkhead homeobox type O transcription factors in [141] Z. Li, S. Bao, Q. Wu et al., “Hypoxia-inducible factors regulate tumorigenic capacity of glioma stem cells,” Cancer Cell, the responses to oxidative stress,” Antioxidants & Redox Signaling, vol. 14, no. 4, pp. 593–605, 2011. vol. 15, no. 6, pp. 501–513, 2009. [142] L. Xiang, D. M. Gilkes, H. Hu et al., “Hypoxia-inducible fac- [157] E. C. Ferber, B. Peck, O. Delpuech, G. P. Bell, P. East, and A. Schulze, “FOXO3a regulates reactive oxygen metabolism by tor 1 mediates TAZ expression and nuclear localization to induce the breast cancer stem cell phenotype,” Oncotarget, inhibiting mitochondrial gene expression,” Cell Death and vol. 5, no. 24, pp. 12509–12527, 2014. Differentiation, vol. 19, no. 6, pp. 968–979, 2012. [143] E. J. Seo, D. K. Kim, I. H. Jang et al., “Hypoxia-NOTCH1- [158] M. K. Lehtinen, Z. Yuan, P. R. Boag et al., “A conserved SOX2 signaling is important for maintaining cancer stem MST-FOXO signaling pathway mediates oxidative-stress cells in ovarian cancer,” Oncotarget, vol. 7, no. 34, responses and extends life span,” Cell, vol. 125, no. 5, pp. 55624–55638, 2016. pp. 987–1001, 2006. 18 Oxidative Medicine and Cellular Longevity [159] S. P. Somasekharan, A. El-Naggar, G. Leprivier et al., “YB-1 regulates stress granule formation and tumor progression by translationally activating G3BP1,” The Journal of Cell Biol- ogy, vol. 208, no. 7, pp. 913–929, 2015. [160] K. Arimoto, H. Fukuda, S. Imajoh-Ohmi, H. Saito, and M. Takekawa, “Formation of stress granules inhibits apo- ptosis by suppressing stress-responsive MAPK pathways,” Nature Cell Biology, vol. 10, no. 11, pp. 1324–1332, 2008. [161] B. J. Moeller, Y. Cao, C. Y. Li, and M. W. Dewhirst, “Radia- tion activates HIF-1 to regulate vascular radiosensitivity in tumors: role of reoxygenation, free radicals, and stress gran- ules,” Cancer Cell, vol. 5, no. 5, pp. 429–441, 2004. [162] M. J. Fournier, C. Gareau, and R. Mazroui, “The chemother- apeutic agent bortezomib induces the formation of stress granules,” Cancer Cell International, vol. 10, p. 12, 2010. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Oxidative Medicine and Cellular Longevity Pubmed Central

Oxidative Stress Gene Expression Profile Correlates with Cancer Patient Poor Prognosis: Identification of Crucial Pathways Might Select Novel Therapeutic Approaches

Oxidative Medicine and Cellular Longevity , Volume 2017 – Jul 9, 2017

Loading next page...
 
/lp/pubmed-central/oxidative-stress-gene-expression-profile-correlates-with-cancer-sIxaLgi0gt

References (166)

Publisher
Pubmed Central
Copyright
Copyright © 2017 Alessandra Leone et al.
ISSN
1942-0900
eISSN
1942-0994
DOI
10.1155/2017/2597581
Publisher site
See Article on Publisher Site

Abstract

Hindawi Oxidative Medicine and Cellular Longevity Volume 2017, Article ID 2597581, 18 pages https://doi.org/10.1155/2017/2597581 Review Article Oxidative Stress Gene Expression Profile Correlates with Cancer Patient Poor Prognosis: Identification of Crucial Pathways Might Select Novel Therapeutic Approaches Alessandra Leone, Maria Serena Roca, Chiara Ciardiello, Susan Costantini, and Alfredo Budillon Experimental Pharmacology Unit, Istituto Nazionale Tumori Fondazione G. Pascale-IRCCS, Naples, Italy Correspondence should be addressed to Alessandra Leone; a.leone@istitutotumori.na.it Received 15 March 2017; Accepted 30 May 2017; Published 9 July 2017 Academic Editor: Lars Bräutigam Copyright © 2017 Alessandra Leone et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The role of altered redox status and high reactive oxygen species (ROS) is still controversial in cancer development and progression. Intracellular levels of ROS are elevated in cancer cells suggesting a role in cancer initiation and progression; on the contrary, ROS elevated levels may induce programmed cell death and have been associated with cancer suppression. Thus, it is crucial to consider the double-face of ROS, for novel therapeutic strategies targeting redox regulatory mechanisms. In this review, in order to derive cancer-type specific oxidative stress genes’ profile and their potential prognostic role, we integrated a publicly available oxidative stress gene signature with patient survival data from the Cancer Genome Atlas database. Overall, we found several genes statistically significant associated with poor prognosis in the examined six tumor types. Among them, FoxM1 and thioredoxin reductase1 expression showed the same pattern in four out of six cancers, suggesting their specific critical role in cancer-related oxidative stress adaptation. Our analysis also unveiled an enriched cellular network, highlighting specific pathways, in which many genes are strictly correlated. Finally, we discussed novel findings on the correlation between oxidative stress and cancer stem cells in order to define those pathways to be prioritized in drug development. 1. Introduction lipoxygenase, and cyclooxygenase) or by nonenzymatic reac- tions, such as during the mitochondrial respiratory chain. Reactive oxygen species (ROS) are commonly identified as These considerations highlight the concept that the source of ROS is extremely heterogeneous. Indeed, ROS can be found oxygen reactive molecules associated with a wide variety of in the environment, as pollutants, tobacco smoke, and iron physiologic events [1] as well as cancer, diabetes, obesity, neu- salts, or generated inside the cells through multiple mecha- rodegeneration, and other age-related diseases [2, 3]. A nisms [4]. Within cells, mitochondria, cytosol, single reduction-oxidation (redox) reaction concerns the transfer of electrons (reducing power) from a more reduced (nucleo- membrane-bound organelles (peroxisomes, endosomes, and phagosomes), or exosomes shed from plasma membranes, as philic) to more oxidized (electrophilic) molecules. ROS can well as extracellular fluids, including plasma, are all involved be classified in two groups: (1) free radical ROS containing one or more unpaired electron(s) in their outer molecular in ROS generation [3, 5, 6]. Mitochondria are the main ROS producers, principally because they are the site of the respira- orbitals (i.e., superoxide radicals and hydroxyl radicals); (2) tory chain when electron leakage can react with molecular oxy- nonradical ROS which are chemically reactive and can be con- gen, resulting in the formation of superoxide, which can verted to radical ROS (i.e., hydrogen peroxide), although they subsequently be converted to other ROS molecules. Then, gen- do not have unpaired electron(s). In both cases, ROS can be produced by either enzymatic reactions (i.e., NADPH oxidase, erated ROS either can be detoxified or can leave the organelle through channels such as voltage-dependent anion channels metabolic enzymes such as the cytochrome P450 enzymes, 2 Oxidative Medicine and Cellular Longevity Survival Stress Cell proliferation death adaptation ROS accumulation Oxidative damage Oxidative damage Increased lipid to proteins to DNA peroxidation Altered gene Loss of DNA repair regulation activity Altered signal Mutation transduction Altered cell growth, differentiation, and apoptosis Figure 1: Redox stress activation in physiology. The production of abnormally large amounts of ROS leads to persistent changes in signal transduction and gene expression that, in the last instance, could give to cell death. The steady-state levels of ROS are determined by the rate of ROS production and their clearance by scavenging mechanisms. (VDAC) or aquaporin, or by small vesicles such as exosomes mechanisms, involving several antioxidant ROS scavengers, [3, 5, 7]. However, ROS can also be the product of β-oxidation as glutathione peroxidase (GPx), thioredoxin (Trx), catalase in peroxisomes, of prostaglandin synthesis and detoxification (CAT), superoxide-dismutase (SOD), and the nuclear factor reactions by cytochrome P450, or of NADPH-mediated erythroid 2 (NRF2) pathway [4, 7]. If a further increase in ROS levels occurs, then the cells undergo apoptotic cell death reaction in phagocytes [4, 5]. ROS are biologically important in a variety of physiolog- (Figure 1). Therefore, under physiological conditions, in ical systems, including adaptation to hypoxia, regulation of order to guarantee cellular redox homeostasis, cells regulated autophagy, immunity, differentiation, and longevity. They intracellular ROS levels by applying a tight regulation of ROS regulate many signal transduction pathways by directly generation and of ROS detoxifying pathways. reacting with proteins and by modulating transcription In this review, we first summarized the role of oxidative factors and gene expression [1]. At low levels, ROS promote stress molecules in cancer initiation and progression and the cellular proliferation, differentiation, and migration as well proposed oxidative stress-targeted anticancer approaches. as cellular stress-responsive survival pathways such as Next, in order to derive cancer-type specific oxidative stress nuclear factor-κB (NF-κB), thus inducing proinflammatory gene profiles and their potential prognostic role, we integrated cytokines [4, 8]. Because of ROS’ highly reactive potential a publicly available oxidative stress gene signature [9] with the toward biological molecules, excessive ROS levels can dam- data extracted from the Cancer Genome Atlas (TCGA) data- age cellular components such as DNA, proteins, and lipids. base. Then, we reviewed some of those genes/pathways corre- To counteract these effects, cells activate “ROS adaption” lating with patient’s survival, in order to define potential novel Oxidative Medicine and Cellular Longevity 3 Less evidences are available about the regulation of ROS anticancer therapeutic targets. Finally, we highlighted novel findings on the correlation between oxidative stress and cancer by microenvironment; however, new efforts have been stem cells (CSC). recently focused in this field [5, 12]. In this regard, Chan et al. demonstrated that cancer-associated fibroblast- (CAF-) derived ROS are able to induce the acquisition of an oxidative 2. The Role of Oxidative Stress Molecules in CAF-like state on normal fibroblasts. Then, these oxidatively Cancer Initiation and Progression transformed normal fibroblasts promoted the development A link between ROS and cancer progression dates back to of aggressive tumors via a TGFβ1-mediated Smad3 signaling, suggesting an important relationship between the extracellu- 1981 when increased levels of H O induced by insulin were 2 2, shown to promote tumor cell proliferation. Almost three lar redox state and cancer aggressiveness [25]. decades later, several studies sustained this hypothesis, reporting increased levels of oxidative damage products in 3. Targeting Oxidative Stress as Anticancer clinical tumor specimens and plasma as well as in cancer cell Therapy lines [5]. Based on these evidences, to date, the idea that altered redox balance and deregulated redox signaling are The first approach to prevent or treat cancer, by targeting strongly implicated in any steps of carcinogenesis as well as ROS, was based on the use of antioxidant reagents [11, 15]. in the resistance to treatment, by affecting many, if not all, In one of the first trials, based on supplementation of sele- hallmarks of cancer is widely accepted [10, 11]. Indeed, cur- nium, vitamin E and β-carotene on the diet showed a reduc- rently, the role of ROS in cancer initiation and progression tion of overall mortality and cancer rates [26]. However, a through the modulation of cell proliferation, apoptosis, following trial not only failed to obtain consistent results angiogenesis, and the alteration of the migration/invasion but also indicated that in certain cases, antioxidants can program is well described [7, 12, 13]. For example, ROS rather promote cancer initiation and progression. Concor- may affect proliferation by a ligand-independent transactiva- dantly, two trials of cancer prevention, the CARET on male tion of different receptor tyrosine kinase via ERK activation smokers, treated with vitamin A and/or β-carotene and the and may induce tissue invasion and metastatic dissemination SELECT trial, on older males treated with vitamin E and/or by activation of metalloproteinases. Moreover, the release of selenium, resulted in an increased incidence of lung and vascular endothelial growth factor and angiopoietin induced prostate tumors, respectively [27–29]. Similar contradictory by ROS promote tumor angiogenesis and anoikis [12, 14]. results were shown in the trials using antioxidant treatment Nonetheless, the exact origin of ROS generation during as adjuvant therapy [30]. cancer development and disease progression and how this Based on these results, almost a decade ago, ROS event could be druggable remains still unclear. Increasing inducers were proposed as anticancer strategy, in order to evidences reported a link between ROS activation and the overcome the specific threshold of ROS level beyond which presence of some oncogenes, such as Ras, c-Myc, or Bcr- cancer cells undergo ROS-mediated cell death [4, 5]. The first Abl [2, 15, 16]. Activation of oncogenic signaling might agents used are those improving electrons leak from the contribute to the increase of ROS levels, which in turn by respiratory complexes in the mitochondria, such as the arse- promoting genomic instability could affect both nuclear nic trioxide, or conventional chemotherapeutic drugs such as and mitochondrial DNA. The consequent activation of anti- doxorubicin. Indeed, patients treated with those agents oxidants’ signaling within tumor cells can also promote cancer showed lipid peroxidation in their plasma as well as low progression and metastasis [2, 15–18]. Furthermore, cancer levels of vitamin E, vitamin C, and β-carotene in the blood cells undergo metabolic changes to counteract the oxidative [4]. The mechanism of action of these agents seems to be stress, also contributing to metastatic program [5, 19, 20]. related to their ability to generate ROS directly from the Loss of functional p53 is involved in ROS induction, due mitochondria. Indeed, doxorubicin and arsenic trioxide pen- to p53 “genome guardian” role in sensing and removing etrate in the inner membrane of the mitochondria and oxidative damage to DNA, thus preventing genetic instability induce superoxide radical production by modulating the [5, 21]. Anyhow, unlike oncogenes, the role of tumor sup- electron transport chain. Also 5-fluorouracil increases mito- pressors in the modulation of ROS is more complex, depend- chondrial ROS with a different mechanism, mediated by ing on the specific tumor suppressor itself. For example, p53 [4, 31]. Ionizing radiations represent other important ataxia-telangiectasia mutated (ATM) is a cellular damage ROS inducers, because they are able to promote by them- sensor that by regulating cell cycle and DNA repair preserves selves high level of ROS and also because they might increase genomic integrity. Deficiency of ATM gene, either in patients NADPH oxidase, an important source of ROS [32]. More- or in mice, has been shown to produce elevated ROS levels over, we and others have demonstrated, in different models and a chronic oxidative stress status. Recently, cytoplasmic and in different combination settings, that oxidative injury ATM is described to activate a pathway leading to autophagy played a significant functional role in the antitumor effect through repression of mammalian target of rapamycin com- of histone deacetylase inhibitors (HDACi), a class of epige- plex 1 (mTORC1) in response to elevated ROS levels [22, 23]. netic antitumor compounds currently in clinical practice in Another example regards the loss of PTEN that determines haematological malignancies [7, 13, 33–42]. AKT hyperactivation and inactivation of the forkhead Recently, a new ROS inducer compound, Elesclomol homeobox type O (FoxO) transcription factor and therefore (STA-4783), has been developed and tested, both in enhanced susceptibility to oxidative stress [24]. in vitro and in vivo preclinical studies as well as in clinical 4 Oxidative Medicine and Cellular Longevity cancer, TCGA_COAD for colon cancer, TCGA_HNSCC trials [5, 43]. Interestingly, the result from a phase II trial using Elesclomol in combination with chemotherapy, in for head and neck cancer (HNSCC), GSE31210 for lung can- malignant melanoma patients, showed ROS generation and cer, TCGA_PRAD for prostate cancer, and METABRIC for oxidative damage associated with prolonged progression- breast cancer [52, 53]. PPISURV automatically derives the free survival [44]. Unfortunately, these results were not rep- currently known interactome for a gene of interest and corre- licated in a phase III trial, where Elesclomol treatment was lates expression levels of its interactome, with survival out- suspended due to adverse toxic effects [45]. The reason of come in multiple publicly available clinical expression data this failure could be ascribed, at least in part, to cancer cells’ sets containing microarray expression data set annotated capability to activate ROS adaption mechanisms by increasing with survival data. In details, as reported by Antonov et al. levels of ROS scavengers, especially at advanced stages. This [54], in the case of the option “single gene survival analyses event is particularly efficacious in CSC, as described in the last on a single data set,” the PPISURV program exploits rank paragraph of this review. To counteract the ROS adaptation information from expression data sets that reflect the relative mechanisms, a plausible solution could be the combination mRNA expression level. The samples are grouped with of ROS inducers either with another ROS inducer or with respect to expression rank of the gene in order to correlate compounds that suppress cellular antioxidants, to overcome survival information to the expression level of a gene in a par- the threshold useful to induce cell death, The latest approach ticular data set. The groups are then subdivided in basis to was tested by using an inhibitor of the scavenger SOD2, 2- “low expression” and “high expression” where expression Me, in combination with arsenic trioxide in lymphocytic rank of the gene is less or more than average expression rank across the data set, respectively. This separation of patients leukemia and urothelial carcinoma cells [46, 47]. Similarly, the combination between the inhibitor of the antiapoptotic into “low” and “high” groups in the data set along with sur- protein bcl2 ABT-737 and the ROS inducer, N-(4-hydroxy- vival information is then used to find any statistical differ- phenyl) retinamide, or the combination between an NRF2 ences in survival outcome and to draw Kaplan-Meier plot. inhibitor and a glutathione-depleting agents, showed increas- Hence, PPISURV establishes a correlation of the selected gene with survival and assesses the sign of the effect and if ing therapeutic efficiency compared to single-agent treatment [48, 49]. Based on these data, several clinical trials of combina- the gene deregulation is associated with positive or negative tion treatment between ROS inducers and scavenger inhibi- outcome. tors are ongoing, including a multicenter phase II trial with Notably, a significant number of oxidative stress genes the iron chelator Triapine and gemcitabine in advanced were negatively correlated with survival in solid carcinomas, reinforced the idea that oxidative stress plays a crucial role in non-small-cell lung cancer [5]. cancer cells (Figure 3). Furthermore, going deep to our bioin- formatics analysis, we observed that breast, lung, and 4. Bioinformatics Correlation between HNSCC cancers were those more susceptible to oxidative stress gene expression fluctuations. To explain these data, Oxidative Stress Gene Expression and one hypothesis could be that all these tumors are more vul- Prognosis in Solid Cancer Patients nerable to external insults (i.e., pollutants) that, as mentioned above, are an important source of ROS. Furthermore, we Although the biological role of oxidative stress pathways has been extensively demonstrated, it is still unclear which and speculate that this phenomenon could be also related to the how oxidative stress genes predict bad prognosis and if their high mutational load of those tumors. Indeed, several studies showed that either breast (particularly triple-negative sub- modulation is cancer-type specific. Here, to address this question, we took advantage of Cancer Genome Atlas group) or lung cancer exhibited an elevated mutational load (TCGA) database that, by profiling RNA expression levels which is closely associated to mutations in DNA damage and DNA mutational status for thousands of genes, has gen- repair genes as well as to intrinsic genomic instability erated comprehensive maps of the key genomic changes in [55–57]. Similarly, recently it has been demonstrated that the overall mutational load was higher in old HNSCC patients several types of cancer, enabling correlative analysis of criti- cal cellular pathways involved in each type of cancers [50, that represent a high percentage of all HNSCC cancers, com- 51]. In details, we compared cancer patient overall survival pared to younger patients [58]. On the contrary, pancreatic, (OS) and the mRNA levels of 73 oxidative stress genes, prostate, and even colon (with exception of microsatellite selected from a public available oxidative stress signature instability (MSI) high subgroup) cancers are described as less hypermutated and thus, we speculate, are also less dependent [9], in different solid tumors. Specifically, the signature included peroxidases, which are represented by glutathione to the oxidative stress and genomic instability [59–61]. peroxidases (GPx) and peroxiredoxins (TPx); genes impli- A further detailed analysis of our correlation between cated in ROS metabolism (i.e., DUSP1, FoxM1, and oxidative gene expression signature and OS unveiled that HMOX1); and genes involved in superoxide metabolism, the behavior of modulated genes was different among the cancers examined, with the exception of two genes involved such as superoxide dismutase (SOD). Starting from the selec- tion of the 73 oxidative stress genes, bioinformatics investiga- in ROS metabolism, such as FoxM1 and TXNRD1, found tions were performed as described in Figure 2. as statistically significantly high in poor prognosis patients In details, bioinformatics analysis was made by SynTar- in four out of six of the tumor types analyzed (Figures 3 get online tool (http://www.bioprofiling.de/PPISURV) using and 4). For this reason, those two genes are described below in details in two specific sections of the review. Other five the following public datasets: TCGA_PAAD for pancreatic Oxidative Medicine and Cellular Longevity 5 73 oxidative TGCA data in solid stress genes cancers PPISURV tool Correlation between negative overall survival (OS) and mRNA levels of genes showed by heat map and Kaplan-Meier curves String analyses Enriched cellular networks Figure 2: Bioinformatics analyses. Flow chart reporting step-by-step bioinformatics approach to unveil the most important genes/pathways involved in the correlation between oxidative stress and cancer. genes, DUSP1, EPHX2, NUDT1, RNF7, and SEPP1, demon- damage and cell death [64]. RNF7 (RING finger protein-7) strated a statistically significant modulation in poor progno- acts as a metal chelating protein, a scavenger of ROS at the sis patients in three out of six tumor types (Figure 3 and expense of self-oligomerization. RNF7 was found overex- Suppl. Figure S1 available online at https://doi.org/10.1155/ pressed in several tumor types, especially in lung carcinoma, 2017/2597581). Briefly, DUSP1 is a dual-specificity phospha- and associated with poor prognosis [65]. tase-1, which is recognized as a key player for inactivating SEPP1 is a selenoprotein 1, involved in cellular incorpo- ration of the selenium circulating in the plasma. Moreover, different MAPK isoforms. Recently, a role of DUSP-1 as cen- tral redox-sensitive regulator in monocytes has been demon- SEPP1 has some antioxidant activity, as target of NRF2 fam- strated [62]. EPHX2 is a cytosolic epoxide hydrolase, implied ily. In agreement, Bae et al. showed that some antioxidant in cancer progression and metastasis, in differentially man- genes known also as NRF2 targets, including SEPP1, were ner based on the stages of carcinogenesis. Indeed, Bracalante also transcriptionally modulated by the oncosuppressor et al. demonstrated that in A7 melanotic cells, resembling less BRCA1, thus suggesting that BRCA1 regulates the activity aggressive tumor cells, anti-oxidant genes, including EPHX2, of NRF2 and protects cells against oxidative stress [66]. were upregulated in response to oxidative stress, while they Finally, in order to identify a more relevant oxidative were downregulated in G10 metastatic melanoma cells [63]. stress family in our setting, we performed an additional bio- NUDT1, nudix hydrolase 1, is the most prominent mamma- informatics analysis where, independently from their trend lian enzyme among other enzymes responsible for hydrolyz- of expression associated to poor prognosis, all modulated ing oxidized DNA precursors. NUDT1 is commonly genes were analyzed in the biological database STRING, a upregulated in a wide variety of tumors to avoid incorpora- resource of known and predicted protein-protein interaction. tion of oxidized nucleotides that, in turn, induce DNA As shown in Figure 5, our analysis reveals an enriched 6 Oxidative Medicine and Cellular Longevity Genes Gene levels p value Gene levels p value Gene levels p value Gene levels p value Gene levels p value Gene levels p value Breast cancer Lung Head and neck Pancreas Prostate Colon RNF7 NUDT1 PRDX5 NOX5 OXSR1 HSPA1A PNKP BNIP3 NQO1 SRXN1 TXNRD1 FOXM1 GCLM ATOX1 FTH1 GTF2I MPV17 NCF2 PDLIM1 PRNP SIRT2 APOE GSS LPO SOD1 ALOX12 CCS SOD2 TXN GPX1 GPX5 GPX6 GPX7 HMOX1 KRT1 NOX4 PRDX6 SCARA3 SFTPD GPX2 PRDX2 STK25 GSR MBL2 NOS2 SQSTM1 DHCR24 DUOX1 GCLC MPO MT3 NCF1 TTN CYGB EPX GPX3 MSRA PREX1 SOD3 UCP2 CAT TXNRD2 OXR1 CCL5 DUOX2 DUSP1 TPO GPX4 AOX1 EPHX2 SEPP1 Figure 3: Bioinformatics correlation between oxidative stress gene expressions and poor prognosis in 6 different tumor types. Heat map in which we report in red or in green if the high or low expression of genes was negatively correlated with survival, respectively. Moreover, we evidenced in yellow when the correlation is statistically significant (with p value <0.05). In the first column, we evidenced in magenta, the genes similar modulated among cancers and in blue those oxidative stress family extracted from STRING analysis. Notably TXNRD1 showed a central role in both analyses. Oxidative Medicine and Cellular Longevity 7 TXNRD1 Head and neck Prostate Breast cancer Lung (TCGA_HNSCC) (TCGA_PRAD) (METABRIC) (GSE31210) 1.0 1.0 1.0 1.0 0.8 0.8 0.8 0.8 0.6 0.6 0.6 0.6 0.4 0.4 0.4 0.4 0.2 0.2 0.2 0.2 0.0 0.0 0.0 0.0 0 50 100 150 200 250 300 0 20 400 60 80 100 120 050 100 150 0 50 100 150 200 250 Time in Time in Time in Time in High expression TXNRD1 High expression TXNRD1 High expression TXNRD1 High expression TXNRD1 Low expression TXNRD1 Low expression TXNRD1 Low expression TXNRD1 Low expression TXNRD1 FOXM1 Breast cancer Lung Pancreas Prostate (METABRIC) (GSE31210) (TCGA_PAAD) (TCGA_PRAD) 1.0 1.0 1.0 1.0 0.8 0.8 0.8 0.8 0.6 0.6 0.6 0.6 0.4 0.4 0.4 0.4 0.2 0.2 0.2 0.2 0.0 0.0 0.0 0.0 0 50 100 150 200 250 300 0 20 40 60 80 100 120 020 40 60 80 0 50 100 150 200 250 Time in Time in Time in Time in High expression FOXM1 High expression FOXM1 High expression FOXM1 High expression FOXM1 Low expression FOXM1 Low expression FOXM1 Low expression FOXM1 Low expression FOXM1 Figure 4: TXNRD1 and FoxM1 expression related to patient survival. Kaplan-Meier curves showing the survival in the case of high and low expression of TXNRD1 and FOXM1 in solid cancer patients. PRNP DHCF NUDT1 RNF7 NCF2 STK25 FTH1 APOE UCP2 SCARA3 SIRT2 MT3 PREX Node size Node color Small nodes Colored nodes MSRA FOXM1 Protein of unknown 3D structure Query proteins and first shell of interactors Large nodes White nodes NQO1 Some 3D structure is known or predicted Second shell of interactors TPO Known interactions Predicted interactions Others From curated databases Text mining Gene neighborhood SFTP Experimentally determined Gene fusions Coexpression TTN SRXN1 Gene co-occurrence Protein homology OXR1 MPV17 ALOX12 KRT1 GTF21 SEPP1 BNIP3 Figure 5: STRING analysis of the 58 oxidative stress genes. Association network in STRING analysis shows interactions of glutathione peroxidase, superoxide dismutase, and thioredoxin as principal oxidative stress signaling among six different tumor types. 8 Oxidative Medicine and Cellular Longevity different mechanisms have been proposed. For example, the cellular network, in which many genes, including GPx, SOD, and Trx pathways (the latter including TXNRD1), are induction of EMT by activation of Slug [76], stabilization of strongly correlated, as demonstrated by both experimental Smad3/Smad4 complex, and activation of TGFβ pathway [77] as well as the modulation of extracellular matrix by studies and text mining (see red and green lines, resp.). Sim- ilar analyses were also performed for each tumor type sepa- affecting the levels of uPA, uPAR, MMP-2, MMP-9, and rately, or considering high or low gene expression VEGF have been proposed [78, 79]. Moreover, FoxM1 coop- individually, confirming in almost all tumor types GPx, erates with survivin and nuclear XIAP in the promotion of SOD, and Trx signaling as those predominant (Suppl. chemoresistance [80]. Finally, further studies demonstrated that FoxM1 induces resistance to all the therapeutics tested Figures S2 and S3). Based on these analyses, together with TXNRD1, we in breast cancer (including cisplatin, paclitaxel, and trastuzu- decided to review the correlated pathways enriched in the mab) by several mechanisms: (1) acting on DNA-damage network, in details (see below), analyzing their role in cancer repair pathways, (2) promotion of cell cycle progression, and the possible therapeutic strategies to hit them. (3) inhibition of cell cycle checkpoints, and (4) apoptosis induction [72]. FoxM1 gene is widely described as amplified also in lung 5. FoxM1, a Critical Regulator of Oxidative cancer, regulating cell proliferation by promoting both G1/S Stress during Tumorigenesis and G2/M transition, differentiation, and transformation The highly conserved transcription factor FoxM1 belongs to [81] as well as inhibition of apoptosis [82]. Recently, a direct link between FoxM1-induced ROS and lung cancer progres- the forkhead box transcription factor family, similarly to the best known member of FoxO family. However, different from sion has been proposed by Tahmasbpoura et al. Their study showed elevated rate of lung cell proliferation related to high the members of FoxO family, FoxM1 is expressed only in pro- liferative cells. Indeed, FoxM1 as a target of the cyclinD- FoxM1 expression in patients exposed to sulfur mustard, a CDK4/6 kinases, is reactivated when quiescent cells reenter well known agent able to induce ROS [83]. Beyond the mechanisms described, the molecular basis of in the cell cycle and reach a maximal level in S-phase which is maintained throughout G2 and mitosis [67, 68]. Beyond this FoxM1 dysregulation has been also related to the capability role on proliferation, FoxM1 regulates metastasis, apoptosis, of vitamin D receptor (VDR)/FoxM1 axis to affect cell stem- and DNA damage repair [69–71]. Furthermore, FoxM1 has ness and to induce an invasive and metastatic phenotypes in been shown to prevent oxidative stress-dependent premature pancreatic cancer. Indeed, the authors observed that VDR activation reduced the levels of FoxM1, inducing nuclear senescence. Park et al. showed how ROS themselves are inducers of FoxM1 expression, which in turn is able to stimu- accumulation of β-catenin [84]. late antioxidant genes. The authors proposed the inhibition of In prostate cancer (PCa), only few studies focused on the FoxM1 as a new therapeutic strategy to kill cancer cells selec- role of FoxM1; for instance, FoxM1 and its target CENPF, a tively [71]. In agreement, FoxM1 knocking-down was structural protein of kinetochore, have been both proposed as critical drivers of PCa development and as prognostic reported to sensitize human pluripotent stem cells to oxidative stress, as a consequence of activated-CAT5 downregulation, a markers of poor survival [85]. Concordantly, Lin et al. FoxM1 antioxidant target gene [69]. unveiled different miRNAs regulating FoxM1-CENPF axis A growing body of evidences reported high FoxM1 as fre- taking advantage of miRNA expression profile available in quently related to poor prognosis in multiple cancers, con- Taylor dataset of prostate specimens (normal, localized, and metastatic tissues) [86]. Notably, since CENPF regulates cordantly with our bioinformatic results [50]. To date, several mechanisms have been proposed to explain the activ- several genes important for metastasis, including MMP2, ity of FoxM1 in cancer progression, including the activation MMP9, LOX, CXCR4, and CXCL12, dysregulation of the miRNA-COUP-TFII-FoxM1-CENPF axis can inhibit also of FoxM1 by several oncogenic protein and signalling path- ways, such as c-Myc, Ras, and PI3K/AKT [72]. PCa metastatization [86]. Overall, these considerations identified FoxM1 as a Hereafter, we discussed the role of FoxM1 in the four tumor types where we found statistically significant associa- potential anticancer therapeutic target. Unfortunately, the tion of FoxM1 expression and poor prognosis. druggability of FoxM1 remains a big challange because of The impact of FoxM1 in breast cancer progression is the lack of substrate-binding pockets and hydrophobic sur- widely demonstrated. Indeed, its high level has been corre- faces [72, 87]. Several in vitro studies proposed RNA interfer- ence (RNAi) as a strategy to knockdown FoxM1, either alone lated with large tumor size, lymphovascular invasion, lymph- node metastases, and high stage. Two independent studies or in combination with ROS inducers, in order to provoke carried out on ER+ patients, reported that low FoxM1 ROS-mediated cell death [82]. Some studies reported that proteasome inhibitors, including bortezomib or thiostrepton, expression, compared to high FoxM1 expression, is associ- ated to better survival. Another study proved a positive cor- directly reduce both FoxM1 expression and its transcrip- tional activity with the same efficacy as that obtained by relation between HER2 status and FoxM1 expression in breast cancer tissue compared to normal breast counterpart FoxM1 silencing [82, 88]. This latter approach is very [73–75], suggesting that FoxM1 is a downstream target of promising, considering that bortezomib is already in clinical practice to treat multiple myeloma, and that RNAi treatment, HER2 and could be used as a marker of HER2 overexpres- sion. However, molecular basis underling the described roles so far, is not a reasonable therapeutic approach in patients [50, 82, 88]. Thus, bortezomib treatment has been proposed of FoxM1 in cancer progression still needs to be clarified and Oxidative Medicine and Cellular Longevity 9 transiently increased cytoplasmic Trx1 oxidation by andro- as effective therapeutic strategy in highly expressing FoxM1 solid tumor, also in association with ROS inducers. gen but decreased Trx1 activities with the progression of prostate cancer, despite high levels of Trx1 protein expres- sion in cancer cells [103]. The role of TrxR1 in dysplastic 6. Thioredoxin, Glutathione Peroxidase, and transformation has been pointed out in human breast epithe- lial cells, triggered by chronic oxidative stress [104]. In addi- Superoxide Dismutase Families as Mediators tion, Trx1 has been proposed as serum biomarker for either of Carcinogenesis early diagnosis or prognosis of breast cancer in association with CEA and CA15-3 [105]. In non-small-cell lung cancer, Thioredoxin system, composed of thioredoxin reductase (TrxR), thioredoxin (Trx), and NADPH, senses and responds Trx1 is able to modulate transcription of cyclooxygenase-2 via hypoxia-inducible factor- (HIF-) 1α [106]. It is actually to oxidative stress and modulates the redox status by scaveng- ing ROS and by regulating several redox enzymes and signal- worth to mention that many human cancers have low levels ing proteins. Mammalian genomes encode two main Trx of thioredoxin-binding protein-2 (TBP-2), a Trx regulator which is able to bind Trx, blocking its reducing activity. systems: Trx1 and Trx reductase (TrxR) 1, which together con- stitutes the cytosolic system; Trx2 and TrxR2, which are local- These mechanisms have been identified as druggable: histone ized inmitochondria (a Trx3 isoformhas been also reported, as deacetylase inhibitors (HDACi) have been demonstrated to a testis-specific form, mainly expressed in male germ cells and upregulate TBP-2 in various transformed cells, associated associated to reproductive disorders) [89]. Trx1 reducing with a decrease in Trx levels [102]. Recently, Park and colleagues observed that TrxR2 is a power allows the transfer of two electronsfrom its dithiol motif to an acceptor, then the oxidized disulfide form of the enzyme novel binding protein for ribonucleotide reductase small sub- is recycled to the dithiol form by TrxR1, thereby oxidizing one unit p53R2, which is involved in nuclear and mitochondrial molecule of NADPH. DNA replication and repair, stimulating the enzymatic activ- Interestingly, our analysis revealed that TXNRD1, the ity of TrxR in vitro. Their findings also suggest that p53R2 acts as a positive regulator of TrxR2 activity in the mitochon- gene encoding TrxR1, is upregulated and correlates with bad prognosis in pancreatic, colon, HNSCC, lung, prostate, dria both under normal physiological conditions and during and breast cancers. Trx1 enzyme has been shown to regulate the cellular response to DNA damage [107]. NF-κB, playing opposite roles, depending to its intracellular Although STRING analyses highlighted glutathione localization: overexpression of Trx in cytoplasm reduced peroxidases (GPx) as one of the main family involved in oxidative stress adaptation, we found high heterogeneity in NF-κB activity, blocking the degradation of the NF-κB inhib- itor IκB; in the nucleus, Trx directly reduces the cysteine(s) of the disregulation of GPx family members among the tumor NF-κB allowing the NF-κB-dependent gene expression [90]. types we have investigated (Figure 5). GPx reduces either free hydrogen peroxide to water or lipid hydroperoxides to their Following NF-κB stimuli, such as UVB irradiation and TNFα treatment, Trx quickly translocates from the cytoplasm into corresponding alcohols. So far, eight different isoforms of GPx, 1 to 8, have been identified in humans, carrying differ- the nucleus. Trx1 has also been reported as a secreted protein by normal and neoplastic cells [91], but not via exosomes ent affinities for their substrates and different localizations. [92]. Notably, Trx-increased secretion contributed to high GPx1, found in the cytoplasm of mammalian cells, is mainly able to target the hydrogen peroxide, while GPx4 showed ROS production in cisplatin-resistant lung tumors, both in vitro and in vivo [93]. high affinity for lipid hydroperoxides. GPx2 is an intestinal and extracellular enzyme, while GPx3 is extracellularly Trx1 itself is regulated both by hypoxia and by oxidative stress conditions via binding of NRF2 to an antioxidant secreted [99]. responsive element in the Trx promotor [94]. Moreover, GPx1 allelic loss or polymosphisms have been known for years to contribute to both lung [108] and breast Trx1 complex functions as a molecular switch turning the cellular redox state into kinase signaling. Thus, the system cancers [109]. Interestingly, in HNSCC cancer, almost all the isoforms showed low expression (Figure 3). In agree- is able to regulate DNA synthesis, cell proliferation [95, 96], apoptosis, and transcription. In details, the reduced form of ment, a decrease in GPx activity accompanied by SOD Trxs binds to apoptosis signal-regulating kinase 1 (ASK1) and CAT decrease as well as higher levels of oxidative DNA damage was found in HNSCC patients compared and inhibits its activity to prevent stress- and cytokine- induced apoptosis; when Trx is oxidized, it dissociates from to healthy donors [110]. An increase of both Trx and GSH metabolism is a mech- ASK1 and apoptosis is stimulated [97–100]. The impact of Trx1 intracellular localization on its role may be taken into anism widely implicated in the resistance of cancer cells to account especially in tumors (as colon and prostate) where chemotherapy. Loss of TXNRD1 makes tumors highly sus- ceptible to pharmacological GSH deprivation, and concomi- a low expression of TXNRD1 correlates to poor patient out- come (as described in Figure 3). In fact, although increased tant inhibition of both GSH and TxrR systems was recently Trx1 protein expression has been associated to hypoxic proposed as an anticancer strategy [18, 111]. Recently, regions of certain tumours, tumor grade and chemoresis- Rodman and colleagues demonstrated that depletion of tence, for instance by scavenging ROS species generated by GSH and inhibition of TrxR activity enhanced radiation responses in human breast cancer stem cells by a mechanism various anticancer agents [101, 102], its localization and activity have to be both taken into account. In prostate can- involving thiol-dependent oxidative stress [112]. Further- cer, Shan and colleagues identified constitutive nuclear and more, Scarbrough and colleagues reported that simultaneous 10 Oxidative Medicine and Cellular Longevity as prosurvival mechanisms associated with resistance to GSH/Trx inhibition sensitizes human breast and prostate cancer cells to 2DG + 17AAG-mediated killing [113]. chemotherapy and tumor relapse [133]. Among the most important antioxidant enzymes, it is Few studies reported the behavior of cancer stem cells in also important to highlight the role of SOD. SOD is able to oxidative stress condition, but notably in contrast to their convert the superoxide (O ) radical into either oxygen (O ) normal stem cell counterparts, cancer stem cells are charac- 2 2 or the less reactive hydrogen peroxide (H O ) which can then terized by increased ROS levels, reduced oxidative damage, 2 2 be removed by CAT, GPx, or TPx. Among the three major and thus longer survival [134, 135]. For example, Im and col- families of SOD, those we single out in humans are the cop- leagues showed that significantly higher ROS levels were per and zinc (Cu-Zn) SOD1, whose localization is in cytosol, observed in the supernatant of glioblastoma cells, grown in nucleus, peroxisome, and intermembrane space of the serum-free sphere medium, either in polystyrene-treated tis- sue culture plates or in nonadherent plates. Moreover, it has mitochondria [114], the mitochondrial enzyme manganese SOD2 (MnSOD), and the (Cu-Zn) extracellular SOD3. been also shown that ROS is a critical factor for maintaining SOD enzymes are able to exert a strong antioxidant activity. stemness, regulating the expression of the transcription fac- In a recent study, Elchuri and colleagues observed that mice tor SOX-2 [136]. This can be due to a combination of mech- deficient in CuZn SOD1 (which contributes to the majority anisms that arise in the tumor, such as modulation of (1) of cellular SOD activity [115]) showed a reduced lifespan and multiple antioxidative enzyme systems [137] or (2) redox- increased incidence of neoplastic changes in the liver [116]. sensitive signaling pathways, as NRF2, NF-κB, c-Jun, and Conversely, it has been also observed by several authors that HIFs, leading to the increased expression of antioxidant SOD1 overexpression makes tumor cells resistant to oxidative molecules [5]. stress and chemotherapy [117]. Increased expression and The higher ROS levels in CSC could be associated with activity of MnSOD has been correlated with cancer aggressive- lower basal expression of ROS-scavenging systems, such as ness in several tumors and through different pathways [118]. SODs, CAT, GPx, and TPx, compared to normal stem cells. Recently, dysregulation of MnSOD function has been linked In this regard, Yang et al. published those nonglioma stem cells to an acetylation-mediated impairment [119, 120] which trig- which displayed significantly lower basal GPx1 expression and gers an increase in oxidative stress, leading to AKT activation activity than glioma stem cells and that miR-153/NRF2/GPx1 via oxidative inactivation of PTEN [119]. MnSOD acetylation pathway plays an important role in regulating radiosensitivity (and activity) is regulated by the deacetylase Sirt3, a mito- and stemness of glioma stem cells via ROS [138]. chondrial fidelity protein. Interestinlgly, Zou et al. showed Due to the growing body of studies focused on the differ- ential modulation of redox-sensitive signaling pathways (as that loss of Sirt3 results in endocrine therapy resistance of human luminal B breast cancer [120]. In agreement, we summarized in Figure 6) in CSC subpopulation, compared and others demonstrated that HDAC inhibition increases to cancer cells or normal stem cells, in this paraghraph we MnSOD protein expression in both solid and haematological discuss the relevance of the ROS-related pathways modulated diseases [121, 122]. in CSC phenotype. Overall, similar to FoxM1, the described antioxidant In hypoxic environments, limited amount of oxygen systems represent putative good targets to improve therapeu- leads to metabolic switches in both normal and malignant tical oxidative stress-dependent strategies. In details, several cells by HIFs. Paradoxically, recent studies have shown that recent efforts have focused on the targeting of Trx/TrxR CSC exhibit high HIF activity in normoxic environments system [123–130]. Moreover, increasing evidences on a puta- and that HIF activity is critical in the maintenance of CSC as well as in the differentiation [139]. In agreement, Wang tive key role of HDAC inhibitors in the modulation of these pathways may deserve further investigations. In this regard, et al. found that overexpression of stem cell factor in hepato- our recent study on the effect of HDACi in regulating cellular carcinoma is regulated by hypoxic conditions NRF2/Keap1 pathway is of interest, considering the interplay through a selective HIF2α-dependent mechanism which between this pathway and thioredoxin [7]. promotes metastasis [140]. Several studies showed that HIF factors can enhance CSC population growth by modulating Notch signaling pathway 7. Oxidative Stress and Cancer Stem Cells in glioma [141], Hippo pathway through direct stabilization In the multitude of morphological, functional, and respon- of TAZ in breast cancer [142], Ras-ERK-ELK3 in hepatocel- sive cancer cells, a subset of the so-called “cancer stem cells” lular cancer, hypoxia-NOTCH1-SOX2 in ovarian cancer [143], and IL6-HIF1α in non-small-cell lung cancer [144]. (CSC), carrying peculiar features, was identified almost ten years ago in solid cancers [131]. However, the name CSC is Additionally, Yang et al. established that gastric CSC exhib- not referred to an origin from normal stem counterpart but ited a marked increase in HIF1α expression and increased rather represents a specific population that displays some migration and invasion capabilities compared with the nor- exceptional properties normally attributed to stem cells. Spe- moxic control upon hypoxia treatment. Also HIF-1α was responsible for activating EMT via increased expression of cific features, like hierarchical differentiation, self-renewal, enhanced invasive capacity, metastatic proficiency, and the transcription factor Snail in gastric CSC [145]. NF-κB is also related to hypoxia and HIF1α induction. It tumorigenicity, make CSC critical for tumor initiation and growth [132], while CSC elevated apoptosis resistance, has been shown that inhibition of NFκB signaling promoted drug-efflux pumps, enhanced DNA repair efficiency, detoxi- a significant reduction in the hypoxia-driven expansion of + −/low CD44 fication enzyme expression, and quiescence are all identified CD24 CSC which was due to increased CD24 Oxidative Medicine and Cellular Longevity 11 ROS ROS ROS ROS Oxidative damage to proteins Oxidative damage to DNA Increased lipid peroxidation Redox adaptation in cancer stem cells Stress pathways Stress pathways alterations alterations Antioxidative enzymes activation FoxO family high expression HIF family high expression (i) SODs; (i) critical mediators of the cellular (i) modulate Notch signaling pathway; (ii) catalase; responses tooxidative stress and (ii) induce and stabilize TAZ, a Hippo pathway (iii) GPx; effector molecule; (iv) peroxiredoxin. have been implicated in many of (iii) induced by ELK3 and IL-6 promoting stemness. ROS-regulated processes; (ii) it is strongly related to WNT NRF2 high expression Adaptative mechanisms activation -pathway modulation. (i) mediate drug resistance by modulating ABCG2 trasporter, and anti-apoptotic factors (es. Bcl-2, Bmi1). block of mRNA translation (stress granule (i) Hippo pathway high expression formation); c-Jun high expression (ii) quiescence induction (high p21expression); (i) molecular switch controlling in (iii) cell death inhibition by modulation of anti cellular differentiation and stem (i) activation of JNK/c-Jun/Notch1 signaling upon stress. apoptotic factors (Bcl-2, Bmi1); cell renewal. (iv) DNA repair activation. HIF1𝛼-YB1-G3BP1 high expression p53 loss led to an early expansion of mammary stem/progenitor (i) (i) in stress condition is activated to Drug resistance cells. preserve cellular homeostasis. (i) modulation of ATP-binding cassette ??? NF-𝜅B activated trasporters; (i) promote the hypoxia-driven expansion of CSC (ii) epithelial to mesenchymal transition by population through Aurora-A. molulation of Snail. Figure 6: Redox stress in cancer stem cells. The persistent production of abnormally large amounts of ROS induced the mechanism of redox adaptation that, in turn, is translated in a various alteration in stress signaling. Here, we reported both known and hypothesized modulated pathways. extracellular matrix components and were enriched in can- expression in breast cancer models [146]. Similarly, Aurora A kinase which can activate NF-κB pathway has been found cer stem cells [154]. highly expressed in ovarian CSC [147]. Moreover, Xie et al. found that knockdown of JNK1 NRF2 represents another antioxidant system involved in or JNK2 or treatment with JNK-IN-8, an adenosine the maintenance of quiescence as well as in the determina- triphosphate-competitive irreversible pan-JNK inhibitor, sig- tion of differentiation fate in normal stem cells, as described nificantly reduced cell proliferation, the ALDH1+ and CD44 and reviewed by Ryoo et al. [148]. For example, NRF2- +/CD24- CSC subpopulations, and mammosphere forma- deficient mice showed defective stem cell function. Indeed, tion, indicating that JNK family promotes CSC self-renewal haematopoietic stem cell, derived from those mice, dispayed and maintenance in triple-negative breast cancer [155]. lower levels of prosurvival cytochines and exibited spontane- However, other factors could be implicated in CSC ous apoptosis related to wild-type cells [149]. capability to adapt high level of intracellular ROS and would Recently, several studies showed that high levels of NRF2 be very interesting to better define them as potential therapeu- are related to CSC survival and anticancer drug resistance in tic targets, mostly because many anticancer drugs increase HNSCC, cervical, breast, and ovarian cancers [150–153]. intracellular ROS levels. Notably, it was reported that NRF2 overexpression is related In this regard, the transcription factors FoxO1, FoxO3a, to an induction of ATP-binding cassette trasporters and thus and FoxO4 are critical mediators of the cellular responses drug resistance mechanisms. Other described redox- to oxidative stress and have been implicated in many of signaling pathway implicated in redox regulation in CSC ROS-regulated processes [156]. It is also known that FoxO competes with TCF for the same binding site of β-catenin could be c-Jun and/or p53 and NF-κB and FoxO family. In details, Chiche et al. showed that the loss of p53 in and suppresses β-catenin-TCF signaling toward prolifera- K5ΔNβcat (βcat activated) mice led to an early expansion tion, thus attenuating WNT-mediated signaling activities. of mammary stem/progenitor cells and accelerated the for- Also, FoxO factors reduce mitochondrial output to prevent mation of triple-negative breast cancers. In particular, p53- excess ROS production through inhibition of c-Myc function deficient tumors expressed high levels of integrins and and alter the hypoxia response [157]. Described mechanisms Hypothesized mechanisms 12 Oxidative Medicine and Cellular Longevity redox homeostasis genes, to guarantee the development of Another candidate is the Hippo pathway, which acts as a molecular switch controlling in cellular differentiation and precision medicine-based approaches in selected subgroups stem cell renewal but is also modulated in stress condition of cancer patients. Further mechanistic studies are needed to and is described as highly mutated in cancer. Lehtinen and identify either new compounds or molecules to be reposi- colleagues elegantly demonstrated the activation of Mst1, a tioned, in order to target the described redox pathways. serine/threonine kinase activated in the Hippo cascade, upon oxidative stress induced by exposure to increasing concentra- Conflicts of Interest tions of exogenous H O . This was accompanied by phos- 2 2 phorylation of the transcription factor FoxO3a at S207, No potential conflicts of interest were disclosed. thereby disrupting its association with 14-3-3 binding pro- tein and leading to its nuclear localization and transcriptional Acknowledgments activation of the BH3- only Bcl-2 protein, Bim, which triggered neuronal apoptosis [158]. This study was partially supported by the following One of the first mechanisms modulated upon stress con- Research Grant to Alfredo Budillon: Italian Ministry of dition is messenger RNA translation, likely as a mean to limit Health (RF-2011-02346914). energy demanding protein synthesis, leading to stress gran- ule (SG) formation in cancer cells. Many evidences suggest that altered mRNA translational control is a critical factor References in cancer progression, and in this regard, a new axis has been [1] L. A. Sena and N. S. Chandel, “Physiological roles of mito- described. In details, Somasekharan et al. showed that under chondrial reactive oxygen species,” Molecular Cell, vol. 48, stress condition, a YB1, nuclease-sensitive element-binding no. 2, pp. 158–167, 2012. protein 1, facilitates tumor metastasis through two mecha- [2] J. G. Gill, E. Piskounova, and S. J. Morrison, “Cancer, oxida- nisms: first, it directly binds to HIF1α that drives stress adap- tive stress, and metastasis,” Cold Spring Harbor Symposia on tation and metastatic capacity in vivo; second, YB1 mediates Quantitative Biology, vol. 81, pp. 163–175, 2016. formation of cytosolic SGs through translational activation of [3] M.Tafani,L.Sansone, F.Limanaetal., “The interplay of reactive G3BP1, a SG nucleator [159]. Accumulating evidences sug- oxygen species, hypoxia, inflammation, and sirtuins in cancer gest that SG formation is protective against stress-induced initiation and progression,” Oxidative Medicine and Cellular cell damage and death [160], and few studies suggested SG Longevity, vol. 2016, Article ID 3907147, 18 pages, 2016. implication in cancer biology [161, 162]. [4] C. Gorrini, I. S. Harris, and T. W. Mak, “Modulation of oxidative stress as an anticancer strategy,” Nature Reviews 8. Conclusions and Future Perspectives Drug Discovery, vol. 12, no. 12, pp. 931–947, 2013. [5] D. Trachootham, J. Alexandre, and P. Huang, “Targeting The idea that the oxidative stress modulation has a crucial cancer cells by ROS-mediated mechanisms: a radical thera- role in cancer cells to promote proliferation, adaptation, peutic approach?,” Nature Reviews Drug Discovery, vol. 8, and resistance to therapy is now widely accepted [7, 12, 13]. no. 7, pp. 579–591, 2009. Thus, modulating redox regulatory mechanisms represents [6] D. Hernandez-Garcia, C. D. Wood, S. Castro-Obregón, and an attractive therapeutic strategy. However, to date, the oxi- L. Covarrubias, “Reactive oxygen species: a radical role in dative stress-related therapeutic strategies evaluated in pre- development?,” Free Radical Biology and Medicine, vol. 49, no. 2, pp. 130–143, 2010. clinical and clinical studies did not produce homogenous [7] A. Leone, M. S. Roca, C. Ciardiello et al., “Vorinostat syner- results, due to several variables associated to ROS generation gizes with EGFR inhibitors in NSCLC cells by increasing and redox adaptation mechanisms. ROS via up-regulation of the major mitochondrial porin Furthermore, the identification of tumor-type specific VDAC1 and modulation of the c-Myc-NRF2-KEAP1 oxidative stress gene profiles and how they could predict pathway,” Free Radical Biology and Medicine, vol. 89, prognosis still represent critical challenges. Thanks to the pp. 287–299, 2015. increasing availability of cancer gene expression profile, [8] G. B. Waypa, J. D. Marks, R. Guzy et al., “Hypoxia triggers mutation, epigenetic, and survival data from the TGCA data- subcellular compartmental redox signaling in vascular set, it was possible to use bioinformatics to screen the role of smooth muscle cells,” Circulation Research, vol. 106, no. 3, oxidative stress genes from a publicly available signature in pp. 526–535, 2010. large cohorts of several solid cancer patients [52, 53]. [9] https://www.qiagen.com/de/shop/pcr/primer-sets/rt2-profile The TCGA database provides correlative evidences r-pcr-arrays/?catno=PAHS-065Z#geneglobe. suggesting the involvement of the FoxM1, thioredoxin, [10] D. Hanahan and R. A. Weinberg, “Hallmarks of cancer: the superoxide-dismutase, and glutathione pathways as princi- next generation,” Cell, vol. 144, no. 5, pp. 646–674, 2011. pally and commonly modulated in breast, lung, HNSCC, [11] E. Panieri and M. M. Santoro, “ROS homeostasis and metab- pancreatic, prostate, and colon cancers. The differential olism: a dangerous liason in cancer cells,” Cell Death & expression levels of each gene observed in different settings Disease, vol. 7, no. 6, article e2253, 2016. revealed a precise spatial context where redox alterations [12] T. Fiaschi and P. Chiarugi, “Oxidative stress, tumor microen- may promote genome instability or redox adaptation. vironment, and metabolic reprogramming: a diabolic For this reason, tumors should be classified into subclasses liaison,” International Journal of Cell Biology, vol. 2012, based on different oxidative stress alterations occurring in Article ID 762825, 8 pages, 2012. Oxidative Medicine and Cellular Longevity 13 [28] G. S. Omenn, G. E. Goodman, M. D. Thornquist et al., “Risk [13] F. Bruzzese, B. Pucci, M. R. Milone et al., “Panobinostat synergizes with zoledronic acid in prostate cancer and multi- factors for lung cancer and for intervention effects in CARET, ple myeloma models by increasing ROS and modulating the beta-carotene and retinol efficacy trial,” Journal of the mevalonate and p38-MAPK pathways,” Cell Death & Disease, National Cancer Institute, vol. 88, no. 21, pp. 1550–1559, vol. 4, article e878, 2013. [14] S. F. Rodrigues and D. N. Granger, “Blood cells and endothe- [29] E. A. Klein, I. M. Thompson, C. M. Tangen et al., “Vitamin E and the risk of prostate cancer: the selenium and vitamin E lial barrier function,” Tissue Barriers, vol. 3, no. 1-2, article e978720, 2015. cancer prevention trial (SELECT),” JAMA, vol. 306, no. 14, pp. 1549–1556, 2011. [15] J. Kim, J. Kim, and J. S. Bae, “ROS homeostasis and metabo- [30] C. Jacobs, B. Hutton, T. Ng, R. Shorr, and M. Clemons, “Is lism: a critical liaison for cancer therapy,” Experimental & there a role for oral or intravenous ascorbate (vitamin C) in Molecular Medicine, vol. 48, no. 11, article e269, 2016. treating patients with cancer? A systematic review,” The [16] S. S. Sabharwal and P. T. Schumacker, “Mitochondrial ROS Oncologist, vol. 20, no. 2, pp. 210–223, 2015. in cancer: initiators, amplifiers or an Achilles’ heel?,” Nature [31] D. B. Longley, T. Latif, J. Boyer, W. L. Allen, P. J. Maxwell, Reviews Cancer, vol. 14, no. 11, pp. 709–721, 2014. and P. G. Johnston, “The interaction of thymidylate synthase [17] G. M. DeNicola, F. A. Karreth, T. J. Humpton et al., expression with p53-regulated signaling pathways in tumor “Oncogene-induced Nrf2 transcription promotes ROS detox- cells,” Seminars in Oncology, vol. 30, no. 3, Supplement 6, ification and tumorigenesis,” Nature, vol. 475, no. 7354, pp. 3–9, 2003. pp. 106–109, 2011. [32] T. Yoshida, S. Goto, M. Kawakatsu, Y. Urata, and T. S. Li, [18] I. S. Harris, A. E. Treloar, S. Inoue et al., “Glutathione and “Mitochondrial dysfunction, a probable cause of persistent thioredoxin antioxidant pathways synergize to drive cancer oxidative stress after exposure to ionizing radiation,” Free initiation and progression,” Cancer Cell, vol. 27, no. 2, Radical Research, vol. 46, no. 2, pp. 147–153, 2012. pp. 211–222, 2015. [33] F. Bruzzese, M. Rocco, S. Castelli, E. Di Gennaro, A. Desideri, [19] A. Kuehne, H. Emmert, J. Soehle et al., “Acute activation of and A. Budillon, “Synergistic antitumor effect between oxidative pentose phosphate pathway as first-line response vorinostat and topotecan in small cell lung cancer cells is to oxidative stress in human skin cells,” Molecular Cell, mediated by generation of reactive oxygen species and DNA vol. 59, no. 3, pp. 359–371, 2015. damage-induced apoptosis,” Molecular Cancer Therapeutics, [20] V. P. Sukhatme and B. Chan, “Glycolytic cancer cells lacking vol. 8, no. 11, pp. 3075–3087, 2009. 6-phosphogluconate dehydrogenase metabolize glucose to [34] R. R. Rosato, J. A. Almenara, and S. Grant, “The histone induce senescence,” FEBS Letters, vol. 586, no. 16, deacetylase inhibitor MS-275 promotes differentiation or pp. 2389–2395, 2012. apoptosis in human leukemia cells through a process regu- [21] S. M. Wormann, L. Song, J. Ai et al., “Loss of P53 function lated by generation of reactive oxygen species and induction activates JAK2-STAT3 signaling to promote pancreatic of p21CIP1/WAF1 1,” Cancer Research, vol. 63, no. 13, tumor growth, stroma modification, and gemcitabine pp. 3637–3645, 2003. resistance in mice and is associated with patient survival,” [35] X. Y. Pei, Y. Dai, and S. Grant, “Synergistic induction of Gastroenterology, vol. 151, no. 1, article e12, pp. 180–193, oxidative injury and apoptosis in human multiple myeloma cells by the proteasome inhibitor bortezomib and histone [22] A. Alexander, S. L. Cai, J. Kim et al., “ATM signals to TSC2 in deacetylase inhibitors,” Clinical Cancer Research, vol. 10, the cytoplasm to regulate mTORC1 in response to ROS,” no. 11, pp. 3839–3852, 2004. Proceedings of the National Academy of Sciences of the United [36] C. Yu, B. B. Friday, J. P. Lai et al., “Abrogation of MAPK and States of America, vol. 107, no. 9, pp. 4153–4158, 2010. Akt signaling by AEE788 synergistically potentiates histone [23] A. Alexander, J. Kim, and C. L. Walker, “ATM engages the deacetylase inhibitor-induced apoptosis through reactive TSC2/mTORC1 signaling node to regulate autophagy,” oxygen species generation,” Clinical Cancer Research, Autophagy, vol. 6, no. 5, pp. 672-673, 2010. vol. 13, no. 4, pp. 1140–1148, 2007. [24] N. McCabe, S. M. Walker, and R. D. Kennedy, “When the [37] S. Bhalla, S. Balasubramanian, K. David et al., “PCI-24781 guardian becomes the enemy: targeting ATM in PTEN- induces caspase and reactive oxygen species-dependent apo- deficient cancers,” Molecular and Cellular Oncology, vol. 3, ptosis through NF-kappaB mechanisms and is synergistic no. 1, article e1053595, 2016. with bortezomib in lymphoma cells,” Clinical Cancer Research, vol. 15, no. 10, pp. 3354–3365, 2009. [25] J. S. Chan, M. J. Tan, M. K. Sng et al., “Cancer-associated fibroblasts enact field cancerization by promoting extratu- [38] N. Koshikawa, J. I. Hayashi, A. Nakagawara, and K. Takenaga, moral oxidative stress,” Cell Death & Disease, vol. 8, no. 1, “Reactive oxygen species-generating mitochondrial DNA article e2562, 2017. mutation up-regulates hypoxia-inducible factor-1alpha gene transcription via phosphatidylinositol 3-kinase-Akt/protein [26] Y. L. Qiao, S. M. Dawsey, F. Kamangar et al., “Total and kinase C/histone deacetylase pathway,” The Journal of cancer mortality after supplementation with vitamins and Biological Chemistry, vol. 284, no. 48, pp. 33185–33194, minerals: follow-up of the Linxian general population nutrition intervention trial,” Journal of the National Cancer Institute, vol. 101, no. 7, pp. 507–518, 2009. [39] I. M. Wolf, Z. Fan, M. Rauh et al., “Histone deacetylases inhibition by SAHA/Vorinostat normalizes the glioma [27] G. S. Omenn, G. Goodman, M. Thornquist et al., “Chemo- microenvironment via xCT equilibration,” Scientific Reports, prevention of lung cancer: the beta-carotene and retinol vol. 4, p. 6226, 2014. efficacy trial (CARET) in high-risk smokers and asbestos- exposed workers,” IARC Scientific Publications, vol. 136, [40] K. F. Hui, P. L. Yeung, and A. K. Chiang, “Induction of pp. 67–85, 1996. MAPK- and ROS-dependent autophagy and apoptosis in 14 Oxidative Medicine and Cellular Longevity [55] J. Budczies, M. Bockmayr, C. Denkert et al., “Classical pathol- gastric carcinoma by combination of romidepsin and borte- zomib,” Oncotarget, vol. 7, no. 4, pp. 4454–4467, 2016. ogy and mutational load of breast cancer - integration of two worlds,” Journal of Pathology: Clinical Research, vol. 1, no. 4, [41] P. H. Liao, H. H. Hsu, T. S. Chen et al., “Phosphorylation of pp. 225–238, 2015. cofilin-1 by ERK confers HDAC inhibitor resistance in hepatocellular carcinoma cells via decreased ROS-mediated [56] R. M. Chabanon, M. Pedrero, C. Lefebvre, A. Marabelle, mitochondria injury,” Oncogene, 2016. J. C. Soria, and S. Postel-Vinay, “Mutational landscape and sensitivity to immune checkpoint blockers,” Clinical [42] M. R. Makena, B. Koneru, T. H. Nguyen, M. H. Kang, and Cancer Research, vol. 22, no. 17, pp. 4309–4321, 2016. C. P. Reynolds, “Reactive oxygen species-mediated syner- gism of fenretinide and romidepsin in preclinical models [57] N. B. Jamieson and A. V. Maker, “Gene-expression profiling of T-cell lymphoid malignancies,” Molecular Cancer to predict responsiveness to immunotherapy,” Cancer Gene Therapeutics, 2017. Therapy, 2016. [58] S. Meucci, U. Keilholz, I. Tinhofer, and O. A. Ebner, [43] J. R. Kirshner, S. He, V. Balasubramanyam et al., “Elesclomol induces cancer cell apoptosis through oxidative stress,” “Mutational load and mutational patterns in relation to age in head and neck cancer,” Oncotarget, vol. 7, no. 43, Molecular Cancer Therapeutics, vol. 7, no. 8, pp. 2319–2327, 2008. pp. 69188–69199, 2016. [59] S. J. Murphy, S. N. Hart, J. F. Lima et al., “Genetic alterations [44] S. O'Day, R. Gonzalez, D. Lawson et al., “Phase II, random- ized, controlled, double-blinded trial of weekly elesclomol associated with progression from pancreatic intraepithelial neoplasia to invasive pancreatic tumor,” Gastroenterology, plus paclitaxel versus paclitaxel alone for stage IV metastatic melanoma,” Journal of Clinical Oncology, vol. 27, no. 32, vol. 145, no. 5, pp. 1098–1109, 2013, e1. pp. 5452–5458, 2009. [60] J. P. McCrow, D. C. Petersen, M. Louw et al., “Spectrum of mitochondrial genomic variation and associated clinical pre- [45] S. J. O'Day, A. M. Eggermont, V. Chiarion-Sileni et al., “Final sentation of prostate cancer in South African men,” Prostate, results of phase III SYMMETRY study: randomized, double- vol. 76, no. 4, pp. 349–358, 2016. blind trial of elesclomol plus paclitaxel versus paclitaxel alone as treatment for chemotherapy-naive patients with advanced [61] F. Gelsomino, M. Barbolini, A. Spallanzani, G. Pugliese, and melanoma,” Journal of Clinical Oncology, vol. 31, no. 9, S. Cascinu, “The evolving role of microsatellite instability in pp. 1211–1218, 2013. colorectal cancer: a review,” Cancer Treatment Reviews, vol. 51, pp. 19–26, 2016. [46] S. Takahashi, “Combination therapy with arsenic trioxide for hematological malignancies,” Anti-Cancer Agents in [62] H. S. Kim, S. L. Ullevig, D. Zamora, C. F. Lee, and R. Asmis, Medicinal Chemistry, vol. 10, no. 6, pp. 504–510, 2010. “Redox regulation of MAPK phosphatase 1 controls mono- cyte migration and macrophage recruitment,” Proceedings [47] K. L. Kuo, W. C. Lin, I. L. Ho et al., “2-methoxyestradiol of the National Academy of Sciences of the United States of induces mitotic arrest, apoptosis, and synergistic cytotoxicity America, vol. 109, no. 41, pp. E2803–E2812, 2012. with arsenic trioxide in human urothelial carcinoma cells,” PloS One, vol. 8, no. 8, article e68703, 2013. [63] C. Bracalente, I. L. Ibañez, A. Berenstein et al., “Reprogram- ming human A375 amelanotic melanoma cells by catalase [48] S. Bruno, F. Ghiotto, C. Tenca et al., “N-(4-hydroxyphenyl)- overexpression: upregulation of antioxidant genes correlates retinamide promotes apoptosis of resting and proliferating with regression of melanoma malignancy and with malignant B-cell chronic lymphocytic leukemia cells and potentiates progression when downregulated,” Oncotarget, vol. 7, no. 27, fludarabine and ABT-737 cytotoxicity,” Leukemia, vol. 26, pp. 41154–41171, 2016. no. 10, pp. 2260–2268, 2012. [64] S. Waz, T. Nakamura, K. Hirata et al., “Structural and kinetic [49] H. R. Lee, J. M. Cho, D. H. Shin et al., “Adaptive response to studies of the human nudix hydrolase MTH1 reveal the GSH depletion and resistance to L-buthionine-(S,R)-sulfoxi- mechanism for its broad substrate specificity,” The Journal mine: involvement of Nrf2 activation,” Molecular and of Biological Chemistry, vol. 292, no. 7, pp. 2785–2794, 2017. Cellular Biochemistry, vol. 318, no. 1-2, pp. 23–31, 2008. [65] Y. Sun and H. Li, “Functional characterization of SAG/RBX2/ [50] B. Rotblat, T. G. Grunewald, G. Leprivier, G. Melino, and ROC2/RNF7, an antioxidant protein and an E3 ubiquitin R. A. Knight, “Anti-oxidative stress response genes: bioinfor- ligase,” Protein & Cell, vol. 4, no. 2, pp. 103–116, 2013. matic analysis of their expression and relevance in multiple cancers,” Oncotarget, vol. 4, no. 12, pp. 2577–2590, 2013. [66] I. Bae, S. Fan, Q. Meng et al., “BRCA1 induces antioxidant gene expression and resistance to oxidative stress,” Cancer [51] O. F. Kuzu, M. A. Noory, and G. P. Robertson, “The role of Research, vol. 64, no. 21, pp. 7893–7909, 2004. cholesterol in cancer,” Cancer Research, vol. 76, no. 8, pp. 2063–2070, 2016. [67] S. N. Mowla, E. W. Lam, and P. S. Jat, “Cellular senescence and aging: the role of B-MYB,” Aging Cell, vol. 13, no. 5, [52] I. Amelio, P. O. Tsvetkov, R. A. Knight, A. Lisitsa, G. Melino, pp. 773–779, 2014. and A. V. Antonov, “SynTarget: an online tool to test the syn- ergetic effect of genes on survival outcome in cancer,” Cell [68] M. Katoh, M. Igarashi, H. Fukuda, H. Nakagama, and M. Death and Differentiation, vol. 23, no. 5, p. 912, 2016. Katoh, “Cancer genetics and genomics of human FOX family genes,” Cancer Letters, vol. 328, no. 2, pp. 198–206, 2013. [53] A. V. Antonov, “BioProfiling.de: analytical web portal for high-throughput cell biology,” Nucleic Acids Research, [69] C. T. Kwok, M. H. Leung, J. Qin et al., “The forkhead box vol. 39, pp. W323–W327, 2011. transcription factor FOXM1 is required for the maintenance of cell proliferation and protection against oxidative stress in [54] A. V. Antonov, M. Krestyaninova, R. A. Knight, I. human embryonic stem cells,” Stem Cell Research, vol. 16, Rodchenkov, G. Melino, and N. A. Barlev, “PPISURV: a no. 3, pp. 651–661, 2016. novel bioinformatics tool for uncovering the hidden role of specific genes in cancer survival outcome,” Oncogene, vol. 33, [70] I. Wierstra, “The transcription factor FOXM1 (forkhead box no. 13, pp. 1621–1628, 2014. M1): proliferation-specific expression, transcription factor Oxidative Medicine and Cellular Longevity 15 [84] Z. Li, Y. Gao, D. Xie et al., “Activation of vitamin D function, target genes, mouse models, and normal biological roles,”AdvancesinCancerResearch,vol.118,pp.97–398,2013. receptor signaling downregulates the expression of nuclear FOXM1 protein and suppresses pancreatic cancer cell [71] H. J. Park, J. R. Carr, Z. Wang et al., “FoxM1, a critical stemness,” Clinical Cancer Research, vol. 21, no. 4, regulator of oxidative stress during oncogenesis,” The EMBO pp. 844–853, 2015. Journal, vol. 28, no. 19, pp. 2908–2918, 2009. [85] A. Aytes, A. Mitrofanova, C. Lefebvre et al., “Cross-species [72] R. Saba, A. Alsayed, J. P. Zacny, and A. Z. Dudek, “The role of regulatory network analysis identifies a synergistic interac- forkhead box protein M1 in breast cancer progression and tion between FOXM1 and CENPF that drives prostate cancer resistance to therapy,” International Journal of Breast Cancer, malignancy,” Cancer Cell, vol. 25, no. 5, pp. 638–651, 2014. vol. 2016, Article ID 9768183, 8 pages, 2016. [86] S. C. Lin, C. Y. Kao, H. J. Lee et al., “Dysregulation of miR- [73] H. Ahn, J. Sim, R. Abdul et al., “Increased expression of fork- NAs-COUP-TFII-FOXM1-CENPF axis contributes to the head box M1 is associated with aggressive phenotype and metastasis of prostate cancer,” Nature Communications, poor prognosis in estrogen receptor-positive breast cancer,” vol. 7, article 11418, 2016. Journal of Korean Medical Science, vol. 30, no. 4, pp. 390– 397, 2015. [87] M. Halasi and A. L. Gartel, “Targeting FOXM1 in cancer,” Biochemical Pharmacology, vol. 85, no. 5, pp. 644–652, 2013. [74] A. Bergamaschi, Z. Madak-Erdogan, Y. J. Kim, Y. L. Choi, [88] A. L. Gartel, “Suppression of the oncogenic transcription fac- H. Lu, and B. S. Katzenellenbogen, “The forkhead tran- tor FOXM1 by proteasome inhibitors,” Scientifica, vol. 2014, scription factor FOXM1 promotes endocrine resistance Article ID 596528, 5 pages, 2014. and invasiveness in estrogen receptor-positive breast can- cer by expansion of stem-like cancer cells,” Breast Cancer [89] A. Jimenez, W. Zu, V. Y. Rawe et al., “Spermatocyte/sperma- Research, vol. 16, no. 5, p. 436, 2014. tid-specific thioredoxin-3, a novel Golgi apparatus-associated [75] R. E. Francis, S. S. Myatt, J. Krol et al., “FoxM1 is a down- thioredoxin, is a specific marker of aberrant spermatogene- sis,” The Journal of Biological Chemistry, vol. 279, no. 33, stream target and marker of HER2 overexpression in breast cancer,” International Journal of Oncology, vol. 35, no. 1, pp. 34971–34982, 2004. pp. 57–68, 2009. [90] K. Hirota, M. Murata, Y. Sachi et al., “Distinct roles of thior- edoxin in the cytoplasm and in the nucleus. A two-step [76] C. Yang, H. Chen, G. Tan et al., “FOXM1 promotes the mechanism of redox regulation of transcription factor NF- epithelial to mesenchymal transition by stimulating the kappaB,” The Journal of Biological Chemistry, vol. 274, transcription of slug in human breast cancer,” Cancer Letters, no. 39, pp. 27891–27897, 1999. vol. 340, no. 1, pp. 104–112, 2013. [91] A. Rubartelli, A. Bajetto, G. Allavena, E. Wollman, and R. [77] J. Xue, X. Lin, W. T. Chiu et al., “Sustained activation of Sitia, “Secretion of thioredoxin by normal and neoplastic cells SMAD3/SMAD4 by FOXM1 promotes TGF-beta- through a leaderless secretory pathway,” The Journal of Bio- dependent cancer metastasis,” The Journal of Clinical Investi- logical Chemistry, vol. 267, no. 34, pp. 24161–24164, 1992. gation, vol. 124, no. 2, pp. 564–579, 2014. [92] Y. Manabe, M. Takagi, M. Nakamura-Yamada et al., “Redox [78] A. Ahmad, Z. Wang, D. Kong et al., “FoxM1 down-regulation proteins are constitutively secreted by skeletal muscle,” The leads to inhibition of proliferation, migration and invasion of Journal of Physiological Sciences, vol. 64, no. 6, pp. 401–409, breast cancer cells through the modulation of extra-cellular matrix degrading factors,” Breast Cancer Research and Treatment, vol. 122, no. 2, pp. 337–346, 2010. [93] M. Wangpaichitr, E. J. Sullivan, G. Theodoropoulos et al., “The relationship of thioredoxin-1 and cisplatin resistance: [79] C. T. Karadedou, A. R. Gomes, J. Chen et al., “FOXO3a its impact on ROS and oxidative metabolism in lung cancer represses VEGF expression through FOXM1-dependent cells,” Molecular Cancer Therapeutics, vol. 11, no. 3, and -independent mechanisms in breast cancer,” Oncogene, pp. 604–615, 2012. vol. 31, no. 14, pp. 1845–1858, 2012. [94] Y. C. Kim, H. Masutani, Y. Yamaguchi, K. Itoh, M. Yamamoto, [80] G. N. de Moraes, D. Delbue, K. L. Silva et al., “FOXM1 targets and J. Yodoi, “Hemin-induced activation of the thioredoxin XIAP and survivin to modulate breast cancer survival and gene by Nrf2. A differential regulation of the antioxidant chemoresistance,” Cellular Signalling, vol. 27, no. 12, responsive element by a switch of its binding factors,” The pp. 2496–2505, 2015. Journal of Biological Chemistry, vol. 276, no. 21, pp. 18399– [81] D. K. Yang, C. H. Son, S. K. Lee, P. J. Choi, K. E. Lee, and M. S. 18406, 2001. Roh, “Forkhead box M1 expression in pulmonary squamous [95] J. S. Song, H. H. Cho, B. J. Lee, Y. C. Bae, and J. S. Jung, “Role cell carcinoma: correlation with clinicopathologic features of thioredoxin 1 and thioredoxin 2 on proliferation of human and its prognostic significance,” Human Pathology, vol. 40, adipose tissue-derived mesenchymal stem cells,” Stem Cells no. 4, pp. 464–470, 2009. and Development, vol. 20, no. 9, pp. 1529–1537, 2011. [82] M. Halasi, B. Pandit, M. Wang, V. Nogueira, N. Hay, and A. [96] J. R. Prigge, S. Eriksson, S. V. Iverson et al., “Hepatocyte DNA L. Gartel, “Combination of oxidative stress and FOXM1 replication in growing liver requires either glutathione or a inhibitors induces apoptosis in cancer cells and inhibits xeno- single allele of txnrd1,” Free Radical Biology and Medicine, graft tumor growth,” The American Journal of Pathology, vol. 52, no. 4, pp. 803–810, 2012. vol. 183, no. 1, pp. 257–265, 2013. [97] A. Matsuzawa and H. Ichijo, “Redox control of cell fate by [83] E. Tahmasbpour, M. Ghanei, A. Qazvini, E. Vahedi, and MAP kinase: physiological roles of ASK1-MAP kinase path- Y. Panahi, “Gene expression profile of oxidative stress way in stress signaling,” Biochimica et Biophysica Acta, and antioxidant defense in lung tissue of patients vol. 1780, no. 11, pp. 1325–1336, 2008. exposed to sulfur mustard,” Mutation Research, Genetic Toxicology and Environmental Mutagenesis, vol. 800-801, [98] M. Saitoh, H. Nishitoh, M. Fujii et al., “Mammalian thiore- pp. 12–21, 2016. doxin is a direct inhibitor of apoptosis signal-regulating 16 Oxidative Medicine and Cellular Longevity via oxidative stress,” Free Radical Biology and Medicine, kinase (ASK) 1,” The EMBO Journal, vol. 17, no. 9, pp. 2596– 2606, 1998. vol. 52, no. 2, pp. 436–443, 2012. [99] E.M.Hanschmann,E.M.Hanschmann,J.R.Godoy,C.Berndt, [114] S. Zhang, J. Xue, J. Zheng et al., “The superoxide dismutase 1 C. Hudemann, and C. H. Lillig, “Thioredoxins, glutaredoxins, 3'UTR maintains high expression of the SOD1 gene in cancer andperoxiredoxins—molecularmechanismsandhealthsignif- cells: the involvement of the RNA-binding protein AUF-1,” icance:fromcofactorstoantioxidantstoredoxsignaling,”Anti- Free Radical Biology and Medicine, vol. 85, pp. 33–44, 2015. oxidants&RedoxSignaling,vol.19,no.13,pp.1539–1605,2013. [115] J. D. Crapo, T. Oury, C. Rabouille, J. W. Slot, and L. Y. Chang, [100] J. Lu and A. Holmgren, “Thioredoxin system in cell death “Copper,zinc superoxide dismutase is primarily a cytosolic progression,” Antioxidants & Redox Signaling, vol. 17, protein in human cells,” Proceedings of the National Academy no. 12, pp. 1738–1747, 2012. of Sciences of the United States of America, vol. 89, no. 21, pp. 10405–10409, 1992. [101] A. Mukherjee and S. G. Martin, “The thioredoxin system: a key target in tumour and endothelial cells,” The British Jour- [116] S. Elchuri, T. D. Oberley, W. Qi et al., “CuZnSOD deficiency nal of Radiology, vol. 81, no. 1, pp. S57–S68, 2008. leads to persistent and widespread oxidative damage and hepatocarcinogenesis later in life,” Oncogene, vol. 24, no. 3, [102] P. A. Marks, “Thioredoxin in cancer–role of histone deacety- pp. 367–380, 2005. lase inhibitors,” Seminars in Cancer Biology, vol. 16, no. 6, pp. 436–443, 2006. [117] D. P. Brown, H. Chin-Sinex, B. Nie, M. S. Mendonca, and M. Wang, “Targeting superoxide dismutase 1 to overcome [103] W. Shan, W. Zhong, R. Zhao, and T. D. Oberley, “Thiore- doxin 1 as a subcellular biomarker of redox imbalance in cisplatin resistance in human ovarian cancer,” Cancer Chemo- therapy and Pharmacology, vol. 63, no. 4, pp. 723–730, 2009. human prostate cancer progression,” Free Radical Biology and Medicine, vol. 49, no. 12, pp. 2078–2087, 2010. [118] A. Piecuch, M. Brzozowa-Zasada, B. Dziewit et al., “Immuno- [104] C. Dong, L. Zhang, R. Sun et al., “Role of thioredoxin reduc- histochemical assessment of mitochondrial superoxide dis- tase 1 in dysplastic transformation of human breast epithelial mutase (MnSOD) in colorectal premalignant and malignant cells triggered by chronic oxidative stress,” Scientific Reports, lesions,” Przeglad Gastroenterologiczny, vol. 11, no. 4, vol. 6, article 36860, 2016. pp. 239–246, 2016. [105] B. J. Park, M. K. Cha, and I. H. Kim, “Thioredoxin 1 as a [119] V. Vidimar, D. Gius, D. Chakravarti, S. E. Bulun, J. J. Wei, and J. J. Kim, “Dysfunctional MnSOD leads to redox dysreg- serum marker for breast cancer and its use in combination with CEA or CA15-3 for improving the sensitivity of breast ulation and activation of prosurvival AKT signaling in uter- cancer diagnoses,” BMC Research Notes, vol. 7, p. 7, 2014. ine leiomyomas,” Science Advances, vol. 2, no. 11, article e1601132, 2016. [106] I. Csiki, K. Yanagisawa, N. Haruki et al., “Thioredoxin-1 modulates transcription of cyclooxygenase-2 via hypoxia- [120] X. Zou, C. A. Santa-Maria, J. O'Brien, D. Gius, and Y. Zhu, inducible factor-1alpha in non-small cell lung cancer,” Can- “Manganese superoxide dismutase acetylation and dysregula- cer Research, vol. 66, no. 1, pp. 143–150, 2006. tion, due to loss of SIRT3 activity, promote a luminal B-like breast carcinogenic-permissive phenotype,” Antioxidants & [107] S. J. Park, H. B. Kim, C. Piao et al., “p53R2 regulates thiore- Redox Signaling, vol. 25, no. 6, pp. 326–336, 2016. doxin reductase activity through interaction with TrxR2,” Biochemical and Biophysical Research Communications, [121] L. Bianchi, F. Bruzzese, A. Leone et al., “Proteomic analysis vol. 482, no. 4, pp. 706–712, 2017. identifies differentially expressed proteins after HDAC vorinostat and EGFR inhibitor gefitinib treatments in Hep-2 [108] J. A. Moscow, L. Schmidt, D. T. Ingram, J. Gnarra, B. John- cancer cells,” Proteomics, vol. 11, no. 18, pp. 3725–3742, 2011. son, and K. H. Cowan, “Loss of heterozygosity of the human cytosolic glutathione peroxidase I gene in lung cancer,” [122] R. R. Rosato, S. S. Kolla, S. K. Hock et al., “Histone deace- Carcinogenesis, vol. 15, no. 12, pp. 2769–2773, 1994. tylase inhibitors activate NF-kappaB in human leukemia cells through an ATM/NEMO-related pathway,” The [109] Y. J. Hu and A. M. Diamond, “Role of glutathione peroxidase Journal of Biological Chemistry, vol. 285, no. 13, 1 in breast cancer: loss of heterozygosity and allelic differ- pp. 10064–10077, 2010. ences in the response to selenium,” Cancer Research, vol. 63, no. 12, pp. 3347–3351, 2003. [123] P. V. Raninga, G. Di Trapani, S. Vuckovic, M. Bhatia, and K. F. Tonissen, “Inhibition of thioredoxin 1 leads to apoptosis [110] J. Milonski, H. Zielinska-Blizniewska, J. Olszewski, I. in drug-resistant multiple myeloma,” Oncotarget, vol. 6, no. 17, Majsterek, and M. Mrowicka, “DNA damage and oxidant- pp. 15410–15424, 2015. antioxidant status in blood of patients with head and neck can- cer,” DNA and Cell Biology, vol. 34, no. 3, pp. 213–219, 2015. [124] L. Xie, Z. Luo, Z. Zhao, and T. Chen, “Anticancer and antian- giogenic iron(II) complexes that target thioredoxin reductase [111] P. K. Mandal, M. Schneider, P. Kölle et al., “Loss of thiore- to trigger cancer cell apoptosis,” Journal of Medicinal doxin reductase 1 renders tumors highly susceptible to phar- Chemistry, vol. 60, no. 1, pp. 202–214, 2017. macologic glutathione deprivation,” Cancer Research, vol. 70, no. 22, pp. 9505–9514, 2010. [125] H. Hwang-Bo, J. Cheong, W. J. Kim, Y. H. Yoo, and Y. H. Choi, “Auranofin, an inhibitor of thioredoxin reductase, [112] S. N. Rodman, J. M. Spence, T. J. Ronnfeldt et al., “Enhance- ment of radiation response in breast cancer stem cells by induces apoptosis in hepatocellular carcinoma Hep3B cells by generation of reactive oxygen species,” General Physiology inhibition of thioredoxin- and glutathione-dependent metabolism,” Radiation Research, vol. 186, no. 4, pp. 385– and Biophysics, 2017. 395, 2016. [126] H. L. Ng, X. Ma, E. H. Chew, and W. K. Chui, “Design, [113] P. M. Scarbrough, K. A. Mapuskar, D. M. Mattson, D. Gius, synthesis, and biological evaluation of coupled bioactive W. H. Watson, and D. R. Spitz, “Simultaneous inhibition of scaffolds as potential anticancer agents for dual targeting of glutathione- and thioredoxin-dependent metabolism is dihydrofolate reductase and thioredoxin reductase,” Journal necessary to potentiate 17AAG-induced cancer cell killing of Medicinal Chemistry, vol. 60, no. 5, pp. 1734–1745, 2017. Oxidative Medicine and Cellular Longevity 17 [144] F. Zhang, S. Duan, Y. Tsai et al., “Cisplatin treatment [127] J. Y. Zhu, C. Y. Zhang, J. J. Dai, K. Rahman, and H. Zhang, “Diterpenoids with thioredoxin reductase inhibitory activi- increases stemness through upregulation of hypoxia- ties from Jatropha multifida,” Natural Product Research, inducible factors by interleukin-6 in non-small cell lung pp. 1–6, 2017. cancer,” Cancer Science, vol. 107, no. 6, pp. 746–754, 2016. [128] D. Duan, J. Zhang, J. Yao, Y. Liu, and J. Fang, “Targeting [145] S. W. Yang, Z. G. Zhang, Y. X. Hao et al., “HIF-1alpha thioredoxin reductase by parthenolide contributes to induces the epithelial-mesenchymal transition in gastric can- inducing apoptosis of HeLa cells,” The Journal of Biological cer stem cells through the snail pathway,” Oncotarget, vol. 8, Chemistry, vol. 291, no. 19, pp. 10021–10031, 2016. no. 6, pp. 9535–9545, 2017. [129] J. Zhang, S. Peng, X. Li, R. Liu, X. Han, and J. Fang, “Target- [146] O. Iriondo, M. Rábano, G. Domenici et al., “Distinct breast ing thioredoxin reductase by plumbagin contributes to cancer stem/progenitor cell populations require either inducing apoptosis of HL-60 cells,” Archives of Biochemistry HIF1alpha or loss of PHD3 to expand under hypoxic condi- and Biophysics, vol. 619, pp. 16–26, 2017. tions,” Oncotarget, vol. 6, no. 31, pp. 31721–31739, 2015. [130] A. Schroeder, U. Warnken, D. Röth et al., “Targeting [147] I. Chefetz, J. C. Holmberg, A. B. Alvero, I. Visintin, and G. thioredoxin-1 by dimethyl fumarate induces ripoptosome- Mor, “Inhibition of aurora-a kinase induces cell cycle arrest mediated cell death,” Scientific Reports, vol. 7, article in epithelial ovarian cancer stem cells by affecting NFkB 43168, 2017. pathway,” Cell Cycle, vol. 10, no. 13, pp. 2206–2214, 2011. [131] J. P. Medema, “Cancer stem cells: the challenges ahead,” [148] I. G. Ryoo, S. H. Lee, and M. K. Kwak, “Redox modu- Nature Cell Biology, vol. 15, no. 4, pp. 338–344, 2013. lating NRF2: a potential mediator of cancer stem cell resistance,” Oxidative Medicine and Cellular Longevity, [132] J. E. Visvader and G. J. Lindeman, “Cancer stem cells: current vol. 2016, Article ID 2428153, 14 pages, 2016. status and evolving complexities,” Cell Stem Cell, vol. 10, no. 6, pp. 717–728, 2012. [149] A. A. Merchant, A. Singh, W. Matsui, and S. Biswal, “The redox-sensitive transcription factor Nrf2 regulates murine [133] S. Colak and J. P. Medema, “Cancer stem cells—important players in tumor therapy resistance,” The FEBS Journal, hematopoietic stem cell survival independently of ROS levels,” Blood, vol. 118, no. 25, pp. 6572–6579, 2011. vol. 281, no. 21, pp. 4779–4791, 2014. [134] D. Trachootham, H. Zhang, W. Zhang et al., “Effective elim- [150] B. C. Lu, J. Li, W. F. Yu, G. Z. Zhang, H. M. Wang, and H. M. Ma, “Elevated expression of Nrf2 mediates multidrug resis- ination of fludarabine-resistant CLL cells by PEITC through a redox-mediated mechanism,” Blood, vol. 112, no. 5, tance in CD133+ head and neck squamous cell carcinoma stem cells,” Oncology Letters, vol. 12, no. 6, pp. 4333–4338, pp. 1912–1922, 2008. [135] E. D. Lagadinou, A. Sach, K. Callahan et al., “BCL-2 inhibi- tion targets oxidative phosphorylation and selectively eradi- [151] Y. Jia, J. Chen, H. Zhu, Z. H. Jia, and M. H. Cui, “Aberrantly cates quiescent human leukemia stem cells,” Cell Stem Cell, elevated redox sensing factor Nrf2 promotes cancer stem cell vol. 12, no. 3, pp. 329–341, 2013. survival via enhanced transcriptional regulation of ABCG2 and Bcl-2/Bmi-1 genes,” Oncology Reports, vol. 34, no. 5, [136] C. N. Im, H. H. Yun, H. J. Yoo, M. J. Park, and J. H. Lee, pp. 2296–2304, 2015. “Enhancement of SOX-2 expression and ROS accumulation by culture of A172 glioblastoma cells under non-adherent [152] I. G. Ryoo, B. H. Choi, and M. K. Kwak, “Activation of NRF2 culture conditions,” Oncology Reports, vol. 34, no. 2, by p62 and proteasome reduction in sphere-forming breast pp. 920–928, 2015. carcinoma cells,” Oncotarget, vol. 6, no. 10, pp. 8167–8184, [137] J. Liu and Z. Wang, “Increased oxidative stress as a selective anticancer therapy,” Oxidative Medicine and Cellular [153] T. Mizuno, N. Suzuki, H. Makino et al., “Cancer stem-like Longevity, vol. 2015, Article ID 294303, 12 pages, 2015. cells of ovarian clear cell carcinoma are enriched in the ALDH-high population associated with an accelerated [138] W. Yang, Y. Shen, J. Wei, and F. Liu, “MicroRNA-153/Nrf-2/ scavenging system in reactive oxygen species,” Gynecologic GPx1 pathway regulates radiosensitivity and stemness of gli- Oncology, vol. 137, no. 2, pp. 299–305, 2015. oma stem cells via reactive oxygen species,” Oncotarget, vol. 6, no. 26, pp. 22006–22027, 2015. [154] A. Chiche, M. Moumen, M. Romagnoli et al., “p53 deficiency induces cancer stem cell pool expansion in a mouse model of [139] G. Peng and Y. Liu, “Hypoxia-inducible factors in cancer triple-negative breast tumors,” Oncogene, 2016. stem cells and inflammation,” Trends in Pharmacological Sciences, vol. 36, no. 6, pp. 374–383, 2015. [155] X. Xie, T. S. Kaoud, R. Edupuganti et al., “c-Jun N-terminal kinase promotes stem cell phenotype in triple-negative breast [140] X. Wang, J. Dong, L. Jia et al., “HIF-2-dependent expression cancer through upregulation of Notch1 via activation of of stem cell factor promotes metastasis in hepatocellular c-Jun,” Oncogene, 2016. carcinoma,” Cancer Letters, 2017. [156] P. Storz, “Forkhead homeobox type O transcription factors in [141] Z. Li, S. Bao, Q. Wu et al., “Hypoxia-inducible factors regulate tumorigenic capacity of glioma stem cells,” Cancer Cell, the responses to oxidative stress,” Antioxidants & Redox Signaling, vol. 14, no. 4, pp. 593–605, 2011. vol. 15, no. 6, pp. 501–513, 2009. [142] L. Xiang, D. M. Gilkes, H. Hu et al., “Hypoxia-inducible fac- [157] E. C. Ferber, B. Peck, O. Delpuech, G. P. Bell, P. East, and A. Schulze, “FOXO3a regulates reactive oxygen metabolism by tor 1 mediates TAZ expression and nuclear localization to induce the breast cancer stem cell phenotype,” Oncotarget, inhibiting mitochondrial gene expression,” Cell Death and vol. 5, no. 24, pp. 12509–12527, 2014. Differentiation, vol. 19, no. 6, pp. 968–979, 2012. [143] E. J. Seo, D. K. Kim, I. H. Jang et al., “Hypoxia-NOTCH1- [158] M. K. Lehtinen, Z. Yuan, P. R. Boag et al., “A conserved SOX2 signaling is important for maintaining cancer stem MST-FOXO signaling pathway mediates oxidative-stress cells in ovarian cancer,” Oncotarget, vol. 7, no. 34, responses and extends life span,” Cell, vol. 125, no. 5, pp. 55624–55638, 2016. pp. 987–1001, 2006. 18 Oxidative Medicine and Cellular Longevity [159] S. P. Somasekharan, A. El-Naggar, G. Leprivier et al., “YB-1 regulates stress granule formation and tumor progression by translationally activating G3BP1,” The Journal of Cell Biol- ogy, vol. 208, no. 7, pp. 913–929, 2015. [160] K. Arimoto, H. Fukuda, S. Imajoh-Ohmi, H. Saito, and M. Takekawa, “Formation of stress granules inhibits apo- ptosis by suppressing stress-responsive MAPK pathways,” Nature Cell Biology, vol. 10, no. 11, pp. 1324–1332, 2008. [161] B. J. Moeller, Y. Cao, C. Y. Li, and M. W. Dewhirst, “Radia- tion activates HIF-1 to regulate vascular radiosensitivity in tumors: role of reoxygenation, free radicals, and stress gran- ules,” Cancer Cell, vol. 5, no. 5, pp. 429–441, 2004. [162] M. J. Fournier, C. Gareau, and R. Mazroui, “The chemother- apeutic agent bortezomib induces the formation of stress granules,” Cancer Cell International, vol. 10, p. 12, 2010.

Journal

Oxidative Medicine and Cellular LongevityPubmed Central

Published: Jul 9, 2017

There are no references for this article.