The multifaceted role of reactive oxygen species in tumorigenesis

Abstract

Redox homeostasis is an essential requirement of the biological systems for performing various normal cellular functions including cellular growth, differentiation, senescence, survival and aging in humans. The changes in the basal levels of reactive oxygen species (ROS) are detrimental to cells and often lead to several disease conditions including cardiovascular, neurological, diabetes and cancer. During the last two decades, substantial research has been done which clearly suggests that ROS are essential for the initiation, progression, angiogenesis as well as metastasis of cancer in several ways. During the last two decades, the potential of dysregulated ROS to enhance tumor formation through the activation of various oncogenic signaling pathways, DNA mutations, immune escape, tumor microenvironment, metastasis, angiogenesis and extension of telomere has been discovered. At present, surgery followed by chemotherapy and/or radiotherapy is the major therapeutic modality for treating patients with either early or advanced stages of cancer. However, the majority of patients relapse or did not respond to initial treatment. One of the reasons for recurrence/relapse is the altered levels of ROS in tumor cells as well as in cancer-initiating stem cells. One of the critical issues is targeting the intracellular/extracellular ROS for significant antitumor response and relapse-free survival. Indeed, a large number of FDA-approved anticancer drugs are efficient to eliminate cancer cells and drug resistance by increasing ROS production. Thus, the modulation of oxidative stress response might represent a potential approach to eradicate cancer in combination with FDA-approved chemotherapies, radiotherapies as well as immunotherapies.

Introduction

Reactive oxygen species (ROS) are characterized as oxygen-carrying molecules having reactive properties which consist of radicals including O₂⁻ (superoxide), HO• (hydroxyl) and non-radicals including H₂O₂ (hydrogen peroxide) [1–4]. These ROS molecules originate from oxygen which is utilized in several metabolic responses in the mitochondria and endoplasmic reticulum (ER) along with peroxisomes [5, 6]. Around 2% of the oxygen is utilized through mitochondria to generate O₂⁻. Therefore, mitochondria are recognized as an utmost source of ROS [3, 6, 7]. The ER provides an oxidizing environment for proper folding of proteins by forming disulfide bonds and increasing ROS levels by oxidation of proteins [8]. Peroxisomes play a dual role: (a) scavenging of ROS through the catalytic degradation of H₂O₂ and (b) generation of ROS via β-oxidation of the fatty acids. ROS can be produced by either enzymatic and/or non-enzymatic mechanisms. The enzymatic mechanism involves NADPH oxidases (NOXs), endothelial nitric oxide synthase (eNOS), xanthine oxidase, arachidonic acid, lipoxygenase, enzymes of cytochrome P450 and cyclooxygenase. Non-enzymatic mechanism of ROS generation is through the mitochondrial respiratory chain [1, 2, 9, 10]. Therefore, coordination of ROS/redox homeostasis is pivotal for regulating the normal biological functions including cell growth, senescence, cell survival and aging. A controlled regulation of ROS inducer, as well as ROS scavenger pathways, is required because low/moderate levels ROS is important for proliferation, differentiation, migration, and survival, whereas excessive ROS levels are harmful [7] (Figs. 1 and 2). Alteration in the H₂O₂ or ROS has a potential effect on cellular functions, because of the fact that signaling pathways and transcription factors (TFs) related to cell division, stem cell differentiation and cellular stress networks are susceptible to the redox environment [11–16]. ROS can easily interact with DNA and other biomolecules. This can lead to DNA damage, incorporation of oncogenic mutations in the normal cells that results in genomic instability and cancer [16–19]. Cancer cells have increased aerobic glycolysis (Warburg effect) which is correlated with augmented ROS/oxidative stress [9]. The increased levels of ROS in cancer cells are because of alterations in key signaling pathways related to cellular metabolism. In the present review, we are focusing on the involvement of ROS as an important regulator of a variety of cellular processes including regulation of cellular homeostasis, various signaling pathways, telomerase, metastasis, angiogenesis, cancer stem cell, immune response and microbiome for the initiation, progression and treatment of human malignancies.

Fig. 1.

Formation and regulation of ROS and its effects on cellular functions. Mitochondria and NADPH oxidases are major sources of O₂⁻, HO•, and H₂O₂ (ROS) formation. Superoxide dismutase (SOD1 or SOD2) can convert O₂⁻ into H₂O₂. H₂O₂ can be converted into H₂O (water) by peroxiredoxin (PRX), glutathione peroxidase (GPX) and catalase (CAT) in mitochondria and cytosol. ROS are generated during normal cellular functioning and homeostasis is maintained by antioxidants expressed by the cells. Low ROS (green) is the basic need to maintain normal cellular proliferation, survival, and differentiation. Moderate to high ROS (tumor favoring ROS; light red) is the signal for the increased cellular proliferation, survival, tumor initiation, immune escape to genomic instability, metastasis, invasion and angiogenesis. Extremely high ROS produced by chemotherapeutic agents (dark red) is dangerous for the cells and leads to cell cycle arrest, apoptosis, senescence and unrepairable DNA damage

Fig. 2.

Maintenance of cellular homeostasis through inducers and scavengers of ROS. ROS can be produced by mitochondria, NADPH oxidases, hypoxia, metabolism, ER stress, cyclooxygenase and oncogenes including HRAS, FLT3-ITD, BCR-ABL, AKT, NF-kB, STAT3 and STAT5. On the other hand, ROS can be eliminated via activation of the dietary antioxidants, glutathione peroxidase, peroxiredoxin, catalase, NRF2, NADPH, SOD and tumor suppressor gnes including BRCA1, BRCA2, TP53, PTEN, FXOP3 and ATM

Regulation of ROS generation

ROS balance is maintained by several enzymes that neutralize toxic oxidants. Superoxide dismutases (SODs) are responsible for the conversion of O₂⁻ into H₂O₂. To avoid cellular damage, catalase (CAT), glutathione peroxidase (GPXs), and peroxiredoxins (PRXs) convert H₂O₂ into water and oxygen [20–22] (Fig. 1). There are six different types of PRXs that are localized in ER, cytosol, peroxisome, and mitochondria and this makes them ideal scavengers for ROS/H₂O₂ [21, 23]. PRXs function is to accept oxidants through active cysteine residue. These oxidized PRXs are then reduced via thioredoxin (TRX), as a result of which TRX gets oxidized and subsequently reduced by TRX reductase [14, 21]. In human cancers, deregulation of TRX metabolism has been found to be involved in drug resistance. Elevated levels of TRX have been noticed in different cancers including colorectal, pancreatic, lung, cervix, liver, and breast [24–29]. Glutathione (GSH) is a well-known antioxidant that functions as a scavenger for free radicals. GSH plays a critical role in multiple cellular processes such as cellular proliferation, division as well as differentiation. GSH is synthesized by glutamate-cysteine ligase (GCL) and GSH synthetase (GSS) [30]. The glutathione antioxidant system comprises GSH, glutathione reductase, GPX and glutathione S-transferases (GST). GSH guards the cells against oxidative stress by minimizing disulfide bond formation to the cysteine residues present on the cytoplasmic proteins. To perform the antioxidant function, GSH has been shown to be oxidized into GSSG. Glutathione peroxidases (GPX) act as a catalyst and accelerate the breakdown of hydroperoxides as well as H₂O₂ [47, 48]. GSH reductase has been shown to reduce GSSG and replenish the pool of GSH via the utilization of NADPH [49] (Fig. 3a). Generation of NADPH inside the cell is mostly controlled by cellular metabolism that includes glucose and glutamine metabolism, pentose phosphate pathway, conversion of pyruvate to malate by malic enzyme and conversion of isocitrate to α-ketoglutarate by isocitrate dehydrogenase (IDH) [1]. Under normal physiological conditions, GSH always occurs in its reduced form inside the cells due to the constitutive activity of glutathione reductase [50]. The reduced form of glutathione plays critical roles to control cellular levels of ROS. Moreover, mitochondrial GSH has been observed to react with ROS and protect from apoptosis. Modification of GSH metabolism has been observed in many tumors [31]. GSH dysregulation has been displayed to be involved in multidrug and radiation resistance. For example, an increase in GSH levels within tumor cells has been correlated with resistance to anthracyclines, platinum-based anticancer drugs, and alkylating agents. Another study showed that overutilization of cysteine for GSH synthesis can mediate tamoxifen resistance against breast cancer cells [32]. GSTs belongs to a class of detoxifying enzymes that accelerate the concurrence of GSH to a number of exogenous and endogenous electrophilic compounds for alimentation of cellular integrity, genomic stability by preventing DNA damage, oxidative stress [33, 34] (Fig. 3b). GSTs showed decreased hydroperoxides and 4-HNE, products of lipid peroxidation, to keep the oxidative stress under control [35]. GSTs have been reported to be robustly expressed in almost all human malignancies to modulate mitogen-activated protein kinase (MAPK) pathways [35]. Also, overexpression of GSTs has been correlated with tumor progression and drug resistance in human cancers [33, 35] (Table 1).

Fig. 3.

Glutathione antioxidant system. a Schematics for the reduction of hydrogen peroxide. Nicotinamide Adenine Dinucleotide Phosphate is essential for the regeneration of GSH via glutathione reductase. Hydrogen peroxide (H₂O₂) is reduced to water (H₂O) via glutathione peroxidase. b Mechanism of the glutathione S transferases (GSTs). Glutathione conjugation with xenobiotic (X) is mainly catalyzed via GST to from glutathione S conjugate

Table 1.

Anticancer drugs or agents that directly or indirectly modulate reactive oxygen species in human malignancies

Name of the drug or agent	Mechanism of the action or their target	Human malignancies	Status	References
Targeting antioxidant system, lipid ROS and ferroptosis
Buthionine sulfoximine (BSO)	Block synthesis of GSH, induce lipid ROS	Ovarian cancer, breast cancer, melanoma, rhabdomyosarcoma	Approved	[51, 54, 55 177, 178, 182, 184]
Sulfasalazine (SSZ)	Inhibition of system Xc-, Induction of ferroptosis	Glioma, pancreatic carcinoma, lung carcinoma	Approved	[52, 54, 55, 194–196]
Artesunate (ART)	Induce ferroptosis through iron metabolism mediated lethal lipid ROS	Pancreatic carcinoma, ovarian cancer, lung carcinoma, head and neck cancer	Phase I/II	[51, 52, 54, 55, 192, 193]
Erastin	Inhibit VDAC2/VDAC3, block GSH synthesis, increase lipid peroxidation and lipid ROS	Fibrosarcoma, lung carcinoma, prostate cancer, osteosarcoma	Phase I/II/III	[51, 52, 54, 55, 184, 197]
Sorafenib	Inhibition of system Xc-, deplete GSH leading to accumulation of lipid ROS	Hepatocellular carcinoma	FDA Approved	[51, 54, 55, 191]
Cisplatin	Suppress GSH and GPX levels	Ovarian cancer, colon cancer	FDA Approved	[54, 200, 249, 250, 253, 254]
RSL-3	Inhibit GPX4 and deplete GSH to induce ROS	Lung carcinoma, colon cancer	Clinical	[52, 54, 55]
ML-162	Inhibit GPX4, enhances ROS production	Colon cancer, melanoma		[51, 54, 55, 202]
ML-210	Inhibit GPX4, increased ROS production	Lung carcinoma, colon cancer		[51, 54, 55, 202]
FIN56	Degrade GPX4 or inhibit the function GPX4	Fibrosarcoma and transformed human fibroblast cells		[51, 53–55, 58, 59]
FINO2	Inhibit GPX4	Fibrosarcoma, renal cell carcinoma		[51, 53, 54, 58]
Lanperisone	Inhibition of system Xc-, enhance ROS production	Lung carcinoma, Kras-mutant mouse embryonic fibroblast	FDA Approved	[201, 202]
Artenimol	Promotes iron metabolism and ROS-mediated ferroptosis	Colon cancer, lung carcinoma		[51, 54, 55, 190]
Salinomycin and Ionomycin	Iron-mediated ROS production	Breast cancer, colon cancer	FDA Approved	[202, 203]
Cotylenin A (CN-A)	Induces ferroptosis by increasing ROS	Pancreatic carcinoma		[54, 208]
N-acetyl-l-cysteine (NAC)	Inhibit ROS production and ferroptosis via oxidative pathway	Fibrosarcoma, colon cancer, breast cancer	FDA Approved	[51, 55, 85, 182, 205, 206, 236]
Vitamin E	Inhibit ferroptosis via suppression of LOX	Knockout Gpx4 murine model		[54, 55, 205, 209]
Ferrostatin	Inhibit ferroptosis via ROS generation from lipid peroxidation	Fibrosarcoma, murine embryonic fibroblast		[54, 56, 204, 205]
Liproxstatin	Inhibit ferroptosis via ROS generation from lipid peroxidation	Murine hippocampal, fibrosarcoma		[54, 56, 204, 205]
EUK-134	SOD mimetic, inhibit H2O2	Lung carcinoma, breast cancer		[1, 54, 180]
NOV-002	Modulate of intracellular GSSG/GSH ratio and increase oxidative stress, glutathione disulfide mimetic	Breast cancer, lung carcinoma	Phase I/II	[1, 7, 37, 180, 181]
DZNep (EZH2 inhibitor)	Silence thioredoxin and increases ROS	Acute myeloid leukemia	Phase I	[186]
All-trans-retinoic acid (ATRA) and arsenic trioxide (ATO)	Inhibit translocation of NRF2, enhance ROS	Leukemia, breast cancer, ovarian cancer	FDA Approved	[187, 188]
AEM1	Repress transcriptional activation of NRF2	Lung carcinoma		[189]
Auranofin	Inhibitor of thioredoxin	Head and neck cancer, ovarian cancer, rhabdomyosarcoma	Phase I/II	[182–185]
Targeting mitochondrial and mitochondrial ROS
Mitoquinone (MitoQ)	Mitochondrial respiratory chain complexes I, III, and IV to enhance ROS production	Renal cell carcinoma, Kras model of pancreatic carcinoma		[85, 157]
MitoTEMPO	Activate SOD2 and inhibit mitochondrial superoxide	Renal cell carcinoma		[85, 157]
Arsenic trioxide (AS2O3)	Enhance ROS production, inhibit the mitochondrial respiratory function	A promyelocytic leukemia, lung carcinoma, myeloma	FDA Approved	[217–220]
Paclitaxel	Increased mitochondria ROS that results in activation of STAT3 signaling	Lung Carcinoma, breast cancer		[117]
Ivosidenib	Specific inhibitors for IDH1/2 mutant and target mROS for the anticancer effect	Acute myeloid Leukemia and Glioblastoma	FDA Approved	[215]
Enasidenib	Specific inhibitors for IDH1/2 mutant and target mROS for the anticancer effect	Acute myeloid Leukemia and Glioblastoma	FDA Approved	[215]
Disulfiram	Inhibit mitochondrial ALDH activity, activate the p38 pathway and ROS	Glioblastoma	FDA Approved	[146, 216]
2-Deoxyglucose	Induce oxidative stress via accumulation of glutathione disulfide and NADP + /NADPH	Pancreatic carcinoma, Prostate cancer, cervical carcinoma	Phase I/II trials	[221–223]
Metformin	Mitochondrial complex I inhibitor, Inhibition of oxygen consumption, activate AMPK signaling	Hepatocellular carcinoma, murine cancer models (B16 for melanoma; MC38 for colon adenocarcinoma)	FDA Approved	[159–161]
Nutraceuticals
Epigallocatechin-3-gallate (EGCG)	Modulation of ROS production, inhibition of NF-κB, regulation of MAPKs	Pancreatic carcinoma, colon cancer, breast cancer, lung carcinoma	Phase I/II	[227, 230]
Phenylethyl isothiocyanate (PEITC)	Deplete GPX and induce ROS	Bladder cancer, renal cell carcinoma, prostate cancer	In clinical trials	[179, 208]
Benzyl isothiocyanate (BITC)	Increase ROS production, activate JNK and p38 pathways	Pancreatic carcinoma, breast cancer, lung carcinoma	Phase I	[231, 234]
Vitamin A	Enhance ROS production	Ovarian cancer		[224]
Vitamin C	Attenuated tumor growth in mutant Kras (G12D)/Apc murine models	Colorectal carcinoma, pancreatic carcinoma	In clinical trials	[225]
Vitamin D	Alteration in the ratio of GSSG and GSH, regulate thioredoxin-interacting protein	Endometrial cancer, breast cancer, colorectal carcinoma	In clinical trials	[226]
Bromelain	Downregulate CoA ligase 4, induce ROS in lipid membranes	KRAS mutant colon cancer		[207]
Pancratistatin	Mitochondrial permeabilization increases ROS	Leukemia, colon cancer	Phase I	[239]
Aminoflavone	Enhance intracellular ROS by degenerating mitochondrial membrane potential	Pancreatic carcinoma, breast cancer, colorectal carcinoma	Phase II	[236]
Curcumin	Enhance intracellular ROS by increasing the potential of mitochondrial membrane	Almost all cancers	Phase II/III	[240, 241]
Nimbolide	Modulation of GSH/GSSG ratio leads to ROS production, inhibit STAT3 pathway			[242, 243]
β-Caryophyllene oxide	Suppress tumor growth and support apoptosis by suppressing ROS-mediated activation of MAPKs	Prostate cancer colon cancer Leukemia, lung carcinoma, multiple myeloma, Prostate cancer		[75, 244]
Chemotherapeutic agents
Doxorubicin, daunorubicin	Block DNA synthesis and topoisomerase II activity; inhibit complex I/II leading to an increase in the production of mitochondrial ROS	Acute myeloid leukemia, acute lymphocytic leukemia, breast cancer, chronic myelogenous leukemia, lymphoma, bladder cancer, Kaposi's sarcoma	FDA Approved	[245, 246]
Salvicine (SAL)	Inhibit topoisomerase II, GSH depletion trigger H2O2 production, DNA double-strand breaks	Gastric carcinoma, leukemia, cervical carcinoma	Phase I/II	[247, 248]
Carboplatin	Maintain very high levels of ROS to induce cell death	Breast cancer, ovarian cancer, lung carcinoma	FDA approved	[249, 250]
Oxaliplatin	Retain DACH by the formation of platinum–DNA adducts, block DNA replication	Colon carcinoma, Ovarian cancer, lung carcinoma	FDA approved	[249, 250]
Temozolomide (TMZ)	Inhibit autophagy, induces cell death via the accumulation of lipid ROS	Glioblastoma stem cells	FDA approved	[190, 197, 198]
PARP inhibitors (Olaparib, niraparib, rucaparib)	Inhibit the activity of PARP enzyme, enhance ROS mediated DNA damage	Breast cancer, ovarian cancer, pancreatic carcinoma, prostate carcinoma, lung carcinoma	FDA approved European Medicines Agency	[253–255]
5-Fluorouracil (5-FU)	Inhibit thymidylate synthetase, block DNA and RNA synthesis, increase ROS	Colorectal carcinoma, breast cancer, pancreatic carcinoma	FDA approved	[251]
Vorinostat	Suppress SLC7A11, enhance ROS lead to DNA damage		FDA approved	[252]

Nuclear factor erythroid 2-related factor 2 (NRF2) is a well-known transcription factor. NRF2 is an important master regulator for maintaining redox balance while enhancing the expression of antioxidant proteins inside the cells [36, 37]. Under normal physiological conditions, NRF2 undergoes proteasomal degradation due to its ability to interact with Kelch-like ECH-associated protein 1(KEAP1), along with Cullin 3 (Cul3) E3 ubiquitin ligase [38, 39]. On the other hand, when there is an increase in the ROS levels during oxidative stress, KEAP1 gets oxidized and obstructs the binding of NRF2 to the KEAP1 degradation complex [40]. This leads to the stabilization of NRF2 in the cytoplasm and its translocation into the nucleus to drive the expression of several genes involved in antioxidants (PRXs, CAT GPXs), redox balance, detoxification, NADPH and GSH synthesis [1, 40–42] (Fig. 4). The constitutive activation of NRF2 has been observed in several human cancers including lung, breast, ovarian, skin, and prostate [41, 43–47]. Moreover, mutations in either KEAP1 or NRF2 and well-established oncogenes (KRAS, Myc) have been found to activate NRF2 [43, 46]. Deregulation in the NRF2–KEAP1 pathway has been reported in drug resistance, genomic instability, resistance to apoptosis, metastasis and metabolic reprogramming in several cancer cells [43, 46–49]. The depletion of NRF2 has displayed decreased tumor growth by enhancing oxidative stress-dependent cell death [40, 46, 47]. Therefore, therapeutic strategies that modulate TRX, PRX, GSH, GPX and NRF2 levels within tumor cells could increase the efficacy of anticancer therapies [50].

Fig. 4.

Role of the NRF2/KEAP1 antioxidant pathway for maintaining cellular homeostasis. Under normal physiological condition, NRF2 interact with KEAP1 to activate Cul3‐dependent ubiquitination and its degradation via the proteasome. Under stress or induced condition, NRF2 dissociates from KEAP1 and translocates into the nucleus. NRF2 forms a heterodimer with sMaf protein as well as to ARE to initiate the transcription of several downstream genes

Role of lipid ROS and ferroptosis in human malignancies

Regulated or programmed cell death is an important process and is required for several key biological processes including development and cellular homeostasis. Programmed cell death can be achieved either via apoptosis or non-apoptotic pathways, including ferroptosis [51–53]. Ferroptosis can be easily distinguished from other types of programmed cell death such as apoptosis, necrosis, and autophagy based on morphology and biochemical reaction [51–54]. Ferroptosis is a different class of cell death that relies on iron metabolism and lipid ROS [51, 52, 55]. Ferroptosis has shown to be initiated either with the depletion of cysteine or loss of glutathione peroxidase 4 (GPX4, an enzyme involved in lipid repair). The loss of GPX4 has been noticed with the accumulation of peroxides in the lipid membrane that leads to aggregation of destructive lipid ROS. The knockout of the Gpx4 gene in the murine model has been observed with increased lethal lipid ROS [56]. Also, the silencing of GPX4 in human cells has been found to induce the accumulation of lipid ROS and ferroptosis cell death [57]. Further, pharmacological inhibitors such as FIN56, FINO2, and RSL3 have reported to either degrade GPX4 or inhibit the function of GPX4 [51, 53–55, 58, 59]. Accumulation of fatal lipid ROS has been noticed with stimulation of polyunsaturated fatty acids (PUFAs) through long-chain fatty acid—CoA ligase 4 (ACSL4) and their addition within the membrane lysophospholipids [60]. However, several reports have proved beyond doubt that the peroxidation of PUFAs is catalyzed by lipoxygenases (LOXs) enzymes [61] (Fig. 5). Moreover, the suppression of system Xc⁻ (erastin or RSL3) linked with indirect repression of GPX4 enzymatic activity [52]. System Xc⁻ belongs to the cystine/glutamate antiporter system, which is associated with the import of extracellular cystine to replace intracellular glutamate [62]. Cysteine (reduced form of cystine) acts as a precursor for the synthesis of glutathione (GSH). GSH functions as a cofactor for GPX4 to catalyze the inhibition of lipid peroxides. The impairment of system Xc⁻ using small molecules displayed an aggregation of lethal lipid peroxides and ROS that led to ferroptosis [57] (Fig. 5).

Fig. 5.

Mechanism and inducer of ferroptosis. Suppression of system Xc⁻/GPX4 activity caused ferroptosis to induce cell death. Elevation of lipid ROS results in the ferroptopsis

Role of ROS in the activated signaling pathways in human malignancies

Human malignancies are one of the major causes of deaths, more than tuberculosis, malaria and acquired immune deficiency syndrome around the world [63]. Cancer is a genetic and metabolic disorder that arises from internal factors (inherited mutations, translocations, abnormal activation of signaling pathways initiated by growth factors and hormones, immune conditions) and external factors (environment, infection, food, alcohol, tobacco, radiation) [63–67]. Both these factors can influence critical genes including proto-oncogenes, tumor suppressor genes, DNA repair, and cell cycle genes through the formation of cellular intermediates such as ROS [68]. The association between ROS and cellular transformation was unveiled by initial studies, where activating RAS mutations and growth factors (epidermal growth factor (EGF), insulin) pathways can enhance the intracellular levels of H₂O₂ to induce tumor growth [69–71]. Now, it is more evident through the laboratory experiments that ROS can lead to carcinogenesis, either by activation of several oncogenic pathways or through oncogenic mutations in the DNA. In this section, we focus on the most relevant signaling pathways such as MAPK/extracellular regulated kinase (ERK)/c-jun N-terminal kinase (JNK) pathway, PI3K/AKT/mTOR pathway, ROS in the NF-κB pathway, signal transducer and activator of transcription (STAT) signaling affected by ROS in cancers.

The MAPK family consisting of ERK1/2, JNK and p38 MAPKs pathways are intracellular signaling pathways required for cellular growth, differentiation and survival. ROS have been shown to oxidize and deactivate MAPK phosphatases, while activating the epidermal growth factor receptor (EGFR) and platelet-derived growth factor receptors (PDGFR) signaling in a ligand-independent fashion through the RAS and ERK pathways [72–78]. Several other studies have demonstrated that H₂O₂ is an important mediator for ligand-independent phosphorylation of receptor tyrosine kinases (RTKs) [79, 80]. For example, metabolism of estrogen in breast carcinoma results in the production of H₂O₂ which in turn activates ERK1/2 to increase cellular proliferation and survival. Mutant HRAS (G12V)-transformed NIH/3T3 fibroblast cells have been shown to generate a huge amount of O₂⁻ via RAC1 [81]. Moreover, ROS can activate HRAS, NRAS, and KRAS oncogenic switch through oxidation of the cysteine residue [82]. Weinberg and colleagues have observed that mitochondrial ROS (mROS) is essential for Kras-mediated tumorigenesis in murine lung carcinoma model via the ERK–MAPK signaling pathway [83]. Mitochondrial transcription factor A (TFAM) is important for the replication of mitochondrial DNA, and depletion of TFAM suppressed the growth of lung tumors in Kras murine models. Moreover, TFAM heterozygous knockout mice have elevated mROS levels and showed increased intestinal tumors in APC Min/+ murine model, suggesting the pivotal role of mROS in carcinogenesis [84]. Similarly, KRAS (G12D, G12V) mutation induces mROS and activates various signaling pathways in the acinar cells for the progression of pancreatic carcinoma [85, 86]. Inhibition of ROS using NAC and MitoQ showed a marked reduction in the initiation and progression of pre-cancerous lesions in Kras-driven murine models of pancreatic cancer [85]. On the contrary, activation of ERK1/2 signaling with exogenous H₂O₂ displayed apoptosis in human pancreatic carcinoma and glioma because of the extremely high level of ROS. Excessive levels of ROS have been found to be positively correlated with senescence, cell cycle arrest, and apoptosis through the ASK1/JNK/p38 signaling cascade [87]. ASK-1 (apoptosis signal-regulated kinase-1) and reduced TRX form a complex that results in the inactivation of ASK-1. During excessive stress, H₂O₂ has been shown to oxidize cysteine residues of TRX, leading to dissociation of ASK-1 for activation of JNK and p38 cascade leading to apoptosis [87–89]. Similarly, glutathione S-transferase P dissociates from JNK to facilitate JNK activation under elevated ROS/H₂O₂ [90] (Fig. 6). Higher levels of H₂O₂/ROS result in prolonged activation of the JNK/p38 that can prevent the proliferation of tumor cells [89–92].

Fig. 6.

ROS activate RAS and PI3/AKT signaling pathways. Growth factor receptor signaling can generate ROS through growth factors, NOXs and mitochondria. ROS can activate RAS/MAPK and PI3K/AKT/mTOR signaling cascade either though inactivation of phosphatases such as PTEN or PTP at cysteine residues or by direct oxidation of kinases. Other mechanisms by which ROS induce cellular signaling are through activation NF-kB signaling

PI3K/PTEN is another important signaling pathway in the tumorigenesis and metastasis where several key intermediates are highly sensitive to redox dysregulation [23, 93]. ROS (O₂⁻ and H₂O₂) can hyperactivate the PI3K/AKT/mTOR pathway through oxidation of the cysteine thiol group of various phosphatases (PTEN, PTP1B, PP2A), resulting in their inactivation [94–97]. Moreover, ROS can indirectly phosphorylate casein kinase II, which promotes degradation of PTEN protein via proteasomes. PTEN is mostly dysregulated in breast, glioblastomas, melanoma, endometrial and prostate cancers because of an increase in ROS (O₂⁻ and H₂O₂) production to favor tumor cell growth and survival [48, 98, 99]. H₂O₂ is generated during the binding of estrogen and growth factors (EGF, PDGF) to their respective receptors (Fig. 6). This has been displayed to activate the PI3K/AKT signaling in breast and ovarian carcinoma [100]. NRF2 protein binds to KEAP1 (E3 ubiquitin ligase) protein to maintain low levels of NRF2 protein in the cytosol under lower concentrations of cellular ROS, whereas high concentrations of ROS lead to oxidation at the cysteine residues of KEAP1, which allows cytosolic NRF2 to translocate into the nucleus to upregulate the expression of antioxidants. Also, activation of PI3K/AKT signaling is essential for the nuclear transportation of NRF2. PI3K inhibitors (LY294002, wortmannin) suppressed NRF2-dependent upregulation of antioxidant genes in neuroblastoma cells [101]. BRCA1 mutant breast tumors are deficient in DNA repair mechanisms and accumulate more ROS, leading to genetic modification. BKM120 treatment impedes estrogen-dependent activation of NRF2-mediated PI3K/AKT signaling, indicating that BRCA1-deficient tumors can be treated by elevating ROS levels [102].

NF-κB is a major TF which plays a critical role in inflammation, cellular proliferation, differentiation, and various immunological responses [103–107]. The NF-κB protein expression has been observed to be triggered via H₂O₂ [108]. For instance, treatment of breast carcinoma cells with IL-1β, TNFα, or sodium arsenite generates H₂O₂ and O₂⁻, which in turn activate NF-κB and enhance cellular growth [109, 110]. Interestingly, knockdown of superoxide dismutase (SOD) showed an increase in the basal ROS levels and NF-κB activity in oral carcinoma. It has been reported that IKK-based NF-κB signaling is activated by increased cellular oxidative stress either by H₂O_2, rotenone-mediated O₂⁻ or by inhibition of the glutathione system. On the other hand, IKK-independent activation of NF-κB occurs through phosphorylation of IκBα at tyrosine residue in response to ROS which releases NF-κB [108] (Fig. 6). ROS have been found to activate NF-κB and NRF2 to support cancer cell survival by increasing the levels of antioxidants to escape cancer cell death in an ROS-dependent fashion [108, 111]. Mutant KRAS generates mROS and activates NF-κB through PKD1, which leads to the formation of precancerous lesions in the pancreas [85].

It has been well established that tumor undergoes metabolic reprogramming due to oxidative phosphorylation (OXPHOS) to control energy requirements. Particularly, tumors addicted to oncogene and drug resistanve have been noticed to rely on ROS/OXPHOS-mediated STAT3 signaling as an alternative mechanism for their survival. Several signaling pathways coincide with STAT3; therefore, translocation of STAT3 to the mitochondria can extend the connection across oncogene-mediated signaling pathways and cancer cell metabolism [112–116]. Radiotherapy treatment has been noticed with markedly lower ROS and elevated protein expression of phospho-STAT3, along with BCL2 in triple-negative breast cancer (TNBC) and radio-resistance. Uncoupling protein 2 (UCP-2) is responsible for reducing ROS levels. UCP2 is highly upregulated to maintain low mROS and resistance to paclitaxel in epithelial lung carcinoma (A549, H460). Paclitaxel resistance was reversed by the silencing of UCP-2 through the STAT3 pathway [117]. Further, niclosamide (STAT3 inhibitor) or STAT3 silencing sensitized the TNBC cells via induction of ROS and inhibition of BCL2 [118]. NOX4 is robustly expressed in NSCLC cells and helps in ROS-dependent IL-6 secretion, which eventually phosphorylates STAT3 (Y705). On the other hand, NOX4 knockdown proved that reduced H₂O₂ inhibited IL-6 dependent STAT3 activity. Also, exogenous IL-6 showed STAT3 activation via NOX4 (Fig. 7). This suggests a positive loop among NOX–ROS–IL-6 and STAT3 [119]. The STAT5 signaling pathway is activated in acute myeloid leukemia (AML) with FLT3/ITD. FLT3/ITD expression in AML has been noticed with increased H₂O₂ in a NOX-dependent manner [120]. FLT3 inhibitor (PKC412) and NOX inhibitors (DPI, VAS2870) have been shown to inhibits ROS production in FLT3/ITD expressing AML cells [121]. STAT5 expression has a positive link with BCR-ABL mutation in chronic myeloid leukemia (CML). STAT5 upregulation has been noticed with high ROS and more BCR-ABL mutation in CML cells. STAT5-induced ROS led to double-strand DNA breaks and witnessed by γH2AX [122].

Fig. 7.

ROS-dependent STAT3 pathway in metastasis and drug resistance. Growth factors, ionizing radiation, mitochondria and NOX4 result in the production of intracellular ROS. ROS activate cancer cells and cancer-associated fibroblast cells to secrete IL-6. IL-6 activates the STAT3 pathway and promotes tumor metastasis, resistance to chemotherapy and radiotherapy, and CSC self-renewal

ROS as an important regulator of telomerase

Human telomerase reverse transcriptase (hTERT) is localized in mitochondria and is important for mitochondrial function [123, 124]. hTERT is critical for respiratory chain function and to maintain low ROS [125–127]. In hepatocellular carcinoma, there is a marked increase in the ROS levels from early to late stage which is positively correlated with increased telomeres length. It has been observed that H₂O₂ extends telomeres by enhancing telomerase activity through AKT signaling in HCC, lung cancer and leukemias. Interestingly, there is a positive association between ROS levels, phosphorylation of AKT, length of telomere and prognosis in human cancers [128]. AKT inhibitors (perifosine, GSK690693, SH-6, and MK‐2206) displayed compromised telomerase activity as well as shortening of telomere length while decreasing ROS levels, viability, H₂O₂-mediated migration and invasion in human malignancies [129, 130]. Now, this is known that mitochondrial TERT can increase intracellular-reduced glutathione to escape ROS-mediated apoptosis [131, 132]. Translocation of hTERT from the nucleus to mitochondria results in multidrug resistance in cancers due to reduced ROS which provides protection to mtDNA. Elevated levels of H₂O₂ have been found to be associated with the shortening of the telomere [133, 134].

ROS is essential for metastasis, angiogenesis and cancer stem cell

Metastasis is the major cause of mortality and only limited number of cells can metastasize to distant organs [135]. Growing pieces of evidence witness the fact that higher levels of ROS are vital to facilitate and sustain the aggressive metastatic phenotype of cancer cells [136]. NOX-dependent ROS/NF-κB pathway accelerates migration and invasion of tumor cells by enhancing TGF-β1, uPA and MMP-9 expression [137]. Mutant TP53 was observed to enhance Nox4-dependent metastasis either through TGF-β1 or independent of TGF-β1 signaling. Treatment of colon carcinoma cells with H₂O₂ stimulated MMP-7 production in an AP1-dependent fashion. Also, ROS can lead to the overexpression of MMP1/2/9 to enhance metastasis. Other reports have displayed that activated integrin-Rac signaling can efficiently generate ROS which results in migration, invasion and epithelial to mesenchymal transition through MMP-3. Matsuno and colleagues found that ROS-activated Nrf2 leads to EMT and metastasis via Notch signaling. ROS can activate TGF-β1 through the TAK1 (TGF-β-activated kinase 1) pathway to metastasize the cancer cells to another organ. NRF2 and ATF4 are involved in antioxidant response by enhancing glutathione synthesis and heme oxygenase 1 to bypass oxidative stress, promoting survival during metastasis by blocking anoikis. Addition of either H₂O₂ or SOD in culture medium displayed EMT phenotype where TWIST1, vimentin and SLUG were upregulated and E-cadherin was downregulated in human malignant mesothelioma and pancreatic carcinoma cells, respectively [138]. It has been noticed that ROS stimulates tumor cells and stromal cells to secrete IL-6, which in turn activates STAT3 signaling and triggers EMT and drug-resistant phenotype by altering the protein expression of E-cadherin, N-cadherin, vimentin, and snail (Fig. 7). ROS activate NF-κB to maintain CSCs and cause resistance to chemotherapy and radiotherapy [119]. Also, the silencing of thioredoxin-like 2 (TXNL2) showed decreased mammosphere formation, metastasis, and tumor growth by inducing ROS levels and suppressing NF-kB activity in breast carcinoma [139].

In normoxia, HIF-1α is degraded due to hydroxylation of PHD2 and recognition through von Hippel–Lindau protein. H₂O₂ has been shown to contribute to metastasis and angiogenesis through the stabilization of HIF and activation of one-carbon metabolism as well as AMPK signaling networks to enhance NADPH production [140]. Hypoxia triggers mROS production which stabilizes HIF-1α subunit by forming a dimer along with HIF-1β to drive the expression of hypoxia-responsive genes to increase angiogenesis in tumor mass [141] (Fig. 8). AKT activation results in the formation of superoxide and H₂O₂, which turn on HIF-1 and induce VEGF expression [142]. Notably, H₂O₂ can promote angiogenesis via the Ang1 and p44/42 MAPK axis. Nox2-generated ROS induces the migration of endothelial cells to tumor mass to promote angiogenesis through several pathways such as PI3K/AKT, Src, and ERK [143]. Importantly, ROS have been noticed to regulate the expression of several TFs and remodeling proteins (p300, VEGF-A, HIF-1α, p53, and MMPs) essential for angiogenesis [144].

Fig. 8.

Mitochondrial ROS in hypoxia and angiogenesis. In oxygen-rich conditions, HIF-1α forms complex with VHL with the help of PHD2. This results in ubiquitination and proteasome-mediated degradation of the complex. On the other hand, mROS can cause the depletion of oxygen levels and inhibition of PHD2 activity resulting in HIF-1 α stabilization, by forming a dimer with HIF-1β. This dimer moves to the nucleus and results in transcriptional activation of VEGF, EPO

Cancer stem cells (CSCs) are correlated with clinical hallmark features such as resistance to therapy, tumor recurrence and metastasis [86, 145, 146]. CD44-positive leukemic stem cells (LSC) have lower ROS because of PKC-θ silencing by NOTCH1 [147]. The frequency of LSC in AML has been correlated with the expression of Gpx3 (ROS scavenger) to keep lower ROS [148]. In breast carcinoma, the Snail-G9a-DNMT1 complex pauses the promoter of E-cadherin and for promoter methylation of fructose-1,6-biphosphatase (FBP1). The silencing of FBP1 cuts down oxygen utilization as well as ROS due to compromised mitochondrial oxidative phosphorylation (OXPHOS). This increases CSC-like properties and tumorigenicity through β-catenin [149, 150]. On the contrary, CSCs are known to have high mROS, which helps them to alter the metabolic reprogramming through fatty acid β-oxidation and MAPK signaling, leading to transcriptional activation of EMT markers in several cancers [151, 152]. Several studies provided evidence that low ROS in CSCs helps them to overcome the effect of chemotherapeutic drugs. These suggested that low levels of ROS are needed to preserve LSC/CSCs.

Role of ROS in the immune response during tumor progression

The tumor microenvironment is composed of myeloid-derived suppressor cells (MDSCs), regulatory T cells (Tregs) and tumor-associated macrophages (TAMs). MDSCs, Tregs cells and TAMs provide an immune-suppressive environment for tumor growth, metastasis, invasion and resistance to chemotherapeutics drugs. CD8⁺ T cells are crucial for anticancer immune response in the tumors. Nonetheless, the tumor microenvironment creates an immunosuppressive environment which eventually results in the suppression of CTL response, leading to cancer progression. High ROS have been noticed as one of the major factors for immunosuppression and inhibition for T cell activation and proliferation, while low ROS can bring the T cell back into action inside the tumor microenvironment. Complexes I and III of the mitochondrial electron transport chain (ETC) are excellent sources of mROS and T cell activation [153–155]. Tumor-infiltrating T cells can be activated by overexpression of PGC1α which is involved in the biogenesis of mitochondria and resumes anticancer activity [156]. ROS scavengers such as MitoQ and MitoTEMPO enhance CD8⁺ tumor-infiltrating lymphocyte activation in kidney tumors by activating SOD2 [157]. T cells expressing chimeric antigen receptor (CAR) and CAT have been shown to be correlated with decreased intracellular oxidative stress and an increased ability of T cells (CAR-CAT) to kill cancer cells [158]. CAR-CAT T cells showed better antitumor response than traditional CAR T cells even under extracellular oxidative stress [158]. Program Death receptor 1 (PD-1) is a negative regulator of the immune system, which is present on the surface of T cells. PD-1 can efficiently bind to either PD-L1 and/or PD-L2, which results in the recruitment of SHP2 and inhibits cytotoxic T-lymphocytes (T-CTLs) to mediate killing of cancer cells. It has been observed that T-CTLs extracted from murine treated with PD-L1 antibody have elevated O₂⁻ and cellular ROS. Further, exposure of these cells to tert-butyl hydroperoxide or a mitochondrial respiratory chain uncoupler showed a synergistic reduction in tumor growth. It has been observed that when HCC xenografts were treated with metformin, oxygen consumption was inhibited in murine tumors, leading to enhanced oxygen supply inside the tumor cells. This results in decreased levels of intratumor hypoxia by suppressing the expression of HIF-1α in HCC xenograft [159]. The combination of metformin with PD-1 blockade markedly enhanced intratumor T cell activation and proliferation, leading to tumor clearance through alleviation of tumor hypoxia [160]. This observation suggests that non-responders to PD-1 antibodies might have high mROS and less hypoxic microenvironment, which results in compromised CTL response. Several studies have observed that elevated ROS or oxidative stress led to immunosuppression inside the tumor microenvironment through Tregs. Furthermore, Tregs hinder the therapeutic ability of the PD-L1 antibody in murine cancer models. Kunisada and colleagues have evaluated that metformin (complex I inhibitor) decreased the number of tumor-infiltrating Tregs by reducing the differentiation ability of the naïve CD4⁺ T cells into Tregs via Foxp3 (the transcriptional regulator for metabolic reprogramming) [161]. Weinberg and group have demonstrated that mitochondrial complex III is needed for inhibiting Treg function [162]. It is clear from the above studies that more research is required to discover the key mechanisms of ROS involved in extracellular and tumor-infiltrating cells in modulating tumor immunity. MDSCs are immunosuppressive cells within the tumor microenvironment (TME). Tumor-induced MDSCs showed a block in T cell proliferation and support colorectal carcinoma cell growth through the production of ROS [163]. Interestingly, catalase (ROS inhibitors) rescued the activity of T cells by suppressing the negative effect of MDSCs [164]. On the contrary, high ROS inhibits T cell responses by suppressing the formation of TCR and MHC antigen complex [165]. TAMs are present within the TME and are important moderators of inflammation and carcinogenesis. ROS are involved in the activation of macrophage signaling. ROS generated from macrophages have been shown to induce Tregs [166]. Another study displayed that ROS promote an invasive phenotype in TAMs extracted from skin cancer (melanoma) through secretion of tumor necrosis factor α [167]. It has been observed that several key mitochondrial genes are highly expressed in TAMs obtained from melanomas, suggesting mROS is the major source of oxidative stress within TAMs. Now, it is very clear that ROS is not only involved in oxidative stress, but also important in immune modulation in human malignancies (Fig. 9).

Fig. 9.

Involvement of ROS in tumor microenvironment and immunosuppression. Myeloid-derived suppressor cells (MDSCs) are generated due to secretion of growth factors (GM-CSF, M-CSF, VEGF) and pro-inflammatory cytokines (IFN-ϒ, IL-1β, IL-4, TNFα by tumor cells. MDSC secrete ROS, nitric oxide (NO) and arginase (ARG) to inactivate T cell and TGFβ, and IL10 to activate regulatory T cells (Tregs). ROS convert M0 macrophages into TAMs and secrete immune-suppressive factors and cytokines to block NK and CTLs. Tumor cells and stromal cells express TGFβ, checkpoint ligands and FasL to cause T cell apoptosis. ROS help tumor cells to overexpress PDL1/2 and CTLA4 to inhibit CTLs. TGFβ stimulates NOXs within the Treg cells to trigger ROS production. Macrophage-induced ROS leads to the accumulation of Treg cells. MDSC produces a large amount of ROS to trigger Tregs and suppress T cells

Importance of ROS in the gut microbiome

It is universally accepted that host microbiota can support tumorigenesis via induction of pro-inflammatory toxins, signaling pathways or escape of antitumor immune functions. Interestingly, several host–microbiota have been associated with the generation of ROS, leading to tumorigenic state [168, 169]. Enterococcus faecalis have been shown to generate extracellular O₂⁻, which is converted to H₂O₂ and can damage DNA in eukaryotes [170]. Bacteroides fragilis generate toxin, which is required for bacterial growth while maintaining polyamine catabolism. This is the major cause of ROS production, DNA damage and tumor initiation in the colon [171]. Several groups have shown that diverse species of bacteria can consume bile acid for their growth and generate ROS as a by-product which induces gastrointestinal cancers and DNA damage [172, 173]. On the contrary, damaged mucosal epithelium utilizes low redox/ROS signaling for repair [174]. It has been observed that host mROS decide diversity in the gut microbiome [175]. High-throughput sequencing of gut microbiota discovered mutations in different genes, leading to change in mitochondrial function and composition of the gut microbiota. Furthermore, modulation of ROS levels displayed higher diversity in the murine gut microbiota [175]. In a recent report, it has been noticed that melanoma patients, who respond well to immunotherapy, have displayed an increase in the diversity of gut microbiota [176]. These studies indicate that modulation of mROS could be used to increase the sensitivity of immunotherapies in cancer patients in clinics.

Role of ROS and ROS scavengers/antioxidants in cancer prevention and treatment

Several chemotherapeutic approaches are designed with the aim of increasing intracellular ROS levels to increase unrepairable damages which result in apoptosis of tumor cells. This is one of the promising approaches which can be easily achieved via chemotherapeutic drugs and radiotherapy depending on the origin of the tumor.

Drugs or agents affecting antioxidant system, lipid ROS and ferroptosis

GCL is an important rate-limiting enzyme in GSH synthesis. GSH metabolism has been displayed to enhance drug resistance by preventing cell death of the tumor cells. Buthionine sulfoximine (BSO) is a well-known inhibitor of de novo GSH synthesis and is clinically used for melanoma, ovarian and breast cancers [177, 178]. Phenylethyl isothiocyanate depletes GPX, and GSH has been reported with anticancer effect in preclinical ovarian cancer murine model [179]. EUK-134 (SOD mimetic) and NOV-002 (glutathione disulfide mimetic) are the antioxidants under clinical development for clinical practice in cancer and other diseases [180]. NOV-002 was injected in patients with HER2-negative breast carcinoma in combination with doxorubicin/cyclophosphamide/docetaxel and demonstrated a favorable antitumor activity with manageable side effect than adjuvant therapy [181]. Auranofin is a well-known inhibitor of thioredoxin and used as an antirheumatic drug in clinics. Importantly, the combination of auranofin with BSO showed enhanced sensitivity in head and neck cancer toward EGFR inhibitors and this effect was reversed in the treatment with NAC [182]. In another study, auranofin treatment of cisplatin-resistant ovarian cancer cells resulted in cytochrome c-mediated cell death via attenuation of TRX reductase [183]. BSO or erastin in combination with auranofin has displayed synergistic anticancer activity in rhabdomyosarcoma by increasing the ubiquitination of proteins [184]. Auranofin inhibited side population, expression of stem cell markers as well as the ability to initiate tumors in lung cancer xenograft model [185]. TRX interacting protein is one of the crucial targets of polycomb-repressive complex 2 and is silenced in AML. DZNep (EZH2 inhibitor) treatment restores the TRX-interacting protein expression, which in turn inhibits thioredoxin and increases ROS, leading the way to apoptosis in AML [186]. These data highlight the importance of thioredoxin metabolism in the survival of cancer cells [183]. Particularly, combination therapy using antioxidants with therapeutic drugs that strongly trigger apoptosis independent of oxidative stress may be effective. Combined treatment of all-trans-retinoic acid (ATRA) and ATO has been reported to prevent the translocation of NRF2 into the nucleus and displayed significant cell death in leukemia and breast cancer cells [187]. ATRA sensitizes the CSCs in ovarian cancer by inhibiting NRF2 and ALDH1 activity [188]. AEM1 showed promising anticancer activity in lung carcinoma by repressing transcriptional activation of NRF2 at ARE site in the nucleus [189]. However, the major challenge for suppressing NRF2 is specificity and toxicity. Sulfasalazine (SSZ), artesunate (ART), erastin, temozolomide (TMZ), sorafenib, BSO, lapatinib, altretamine, ML-162, RSL-3, ML-210, and ATRA are well-known inhibitors for induction of ferroptosism [51, 52, 54, 55, 190]. Sorafenib was initially discovered as an inducer of ferroptosis in hepatocellular cancer cells [191]. Mechanistically, sorafenib depletes GSH along with the accumulation of lipid ROS [191]. ART has been shown to induces ferroptosis in human cancer cells including pancreas, head and neck, and ovarian through iron metabolism-mediated ROS [192, 193]. SSZ induces ferroptosis in glioma cells (GBM), pancreatic carcinoma and lung carcinoma via inhibition of system Xc⁻ [54, 194–196]. Erastin triggered ferroptosis in fibrosarcoma, lung, prostate, and osteosarcoma cells [197]. TMZ in combination with erastin can be a potential therapeutic agent in GBM [197]. TMZ inhibits autophagy in glioblastoma stem cells and induces cell death via the accumulation of lipid ROS [190, 198, 199]. Cisplatin exerts an anticancer effect in HCT116 (colon cancer) and A549 (lung cancer) cells through apoptosis via reduced GSH and GPX [200]. Lanperisone enhances the production of ROS to induce ferroptotic death in K-Ras-mutant mouse embryonic fibroblasts and lung cancer cells in the mouse model [201, 202]. Moreover, salinomycin and ionomycin are clinically approved antibiotics that promote ferroptosis in colon and breast cancer cells through iron metabolism-mediated ROS [202, 203]. Ferrostatin, liproxstatin and zileuton have been reported to suppress erastin and RSL3-induced ferroptosis in fibrosarcoma, murine hippocampal and murine embryonic fibroblasts [54, 56, 204, 205]. Several natural compounds including bromelain, baicalein, artenimol, artemisinin, cotylenin A (CN-A), N-acetyl-l-cysteine (NAC) and vitamins can control cell death via ferroptosis, lipid peroxidation and ROS production [52, 54–56, 190, 206–209].

Drugs or agents affecting mitochondria and mitochondrial ROS

IDH1/2 are mutated in blood cancers and brain tumors and result in the formation of 2-hydroxyglutarate (oncometabolite) [210–213]. In the Idh1 mutant knock-in murine model, there is a decrease in the intracellular ROS, leading to an increase in the NADP(+)/NADPH ratio and expression of Hif1α target gene in brain and hematopoietic cells [214]. The lower levels of ROS have been associated with metabolism and overexpression of BCL2 protein in leukemic stem cells in IDH1/2 mutant AML. Ivosidenib and enasidenib are specific inhibitors for IDH1/2 mutant and target mROS for the anticancer effect. These inhibitors showed promising antileukemic activity in patients with AML in clinical trials and are approved by the FDA for the treatment of elderly AML patients [215]. Disulfiram, an ALDH inhibitor in combination with copper (Cu), has been reported to inhibit cancer stem cells and tumor growth of GBM cells via suppression of mitochondrial ALDH activity and generation of ROS along with the activation of p38 pathway [216]. Disulfiram/Cu specifically eliminates leukemia-initiating cells by silencing of NRF2/NF-kB cascade and elevating ROS-dependent JNK pathway [146]. Arsenic trioxide (AS₂O₃) is one of the most successful FDA-approved therapies for leukemia, lung, and myeloma [217, 218]. AS₂O₃ exposure enhances ROS production and is sensed by PML to enhance nuclear body formation which eventually activates p53 to induces differentiation and cell death of leukemic cells [217, 219]. AS₂O₃ combined with ascorbic acid in phase 1 study and was found to be effective against patients with relapsed/refractory multiple myeloma [220]. Paclitaxel treatment revealed an elevated level of ROS through mitochondria which results in activation of STAT3 and JAK2 through phosphorylation in lung carcinoma cells, leading to BCL-2 mediated programmed cell death [117]. 2DG (2-deoxyglucose; glucose analog) has been shown to impede glucose metabolism that results in the accumulation of GSSG to induce oxidative stress. This was associated with radio-sensitization and marked apoptosis in a variety of cancers including pancreatic, prostate and cervical [221–223].

Nutraceuticals with antioxidant properties

Importantly, the intake of natural antioxidant-rich foods has been recommended as one of the best ways to protect against cancer. Several nutrients (vitamins A, C, and D, epigallocatechin-3-gallate (EGCG), genistein, curcumin, piperine, theanine, and choline) have strong antioxidant properties and have been found to control the expansion of cancer stem cells and tumorigenesis in pancreatic, ovarian, breast, colorectal and brain tumors. Wang and colleagues have performed a meta-analysis in a large cohort to find out the correlation between vitamin A and patients with ovarian cancer [224]. KRAS or BRAF mutations are the most recurrent mutations in colorectal carcinoma. It has been observed that high doses of vitamin C showed selective killing of colorectal cancer cells having either KRAS or BRAF mutations because of increased uptake of the dehydro-ascorbate (DHA, the oxidized form of vitamin C) through GLUT1 [225]. This led to the accumulation of ROS, inhibition of glyceraldehyde 3-phosphate dehydrogenase, energy crisis. Interestingly, vitamin C attenuated tumor growth in mutant Kras (G12D)/Apc murine models [225]. More recently, Grant has observed that vitamin D can lower the risk of colorectal and breast cancer, whereas it was the opposite in prostate cancer [226]. Yang and colleagues have reviewed the role, molecular mechanism and signaling pathways of EGCG in several murine cancer models as well as in human cancers [227]. To date, there is a limited therapeutic option for pancreatic cancer which includes gemcitabine in combination with trichostatin A, EGCG, benzyl isothiocyanate (BITC), and capsaicin [227–233]. The above drugs are known to increase intracellular ROS levels to promote apoptosis. BITC operates through ROS-dependent ERK/JNK/p38MAPK and G₂/M arrest by reducing cyclin B1, Cdc2, and Cdc25C in pancreatic and other cancer [231, 234]. EGCG treatment suppressed the expression of the BCL-2, IAP, BCL-X_L, and cIAP (antiapoptotic) and enhanced the expression of the BAD, FAS, and BAX pro-apoptotic [230]. Sulindac is the FDA-approved drug that enhances intracellular ROS levels in colorectal and lung cancer cells which makes them sensitive to H₂O₂-mediated apoptosis [235]. Aminoflavone induces cell death in breast cancer cells (MCF7, MDA-MB231), but is non-toxic in MCF-10A (non-malignant breast cells). Aminoflavone displayed a marked increase in intracellular ROS and was significantly correlated with the activation of caspase 3-mediated cell death. Further, inhibition of ROS production using NAC reverses the effect of amino flavone [236]. NAC treatment suppressed migration, invasion, and EMT through matrix metallopeptidase 3. Pancratistatin, IOA, thymoquinone, and Triphala induce apoptosis of breast carcinoma cells by enhancing intracellular ROS by increasing the potential of mitochondrial membrane [113, 237–239]. Curcumin is a well-known natural antioxidant that has been used as an anticancer agent in almost all human malignancies. Curcumin at lower concentrations has been correlated with reduced ROS production, while curcumin at higher concentrations displayed increased ROS levels in leukemia and solid tumors [240, 241]. Nimbolide has been found to induce oxidative stress, which caused delay in tumor growth in the transgenic prostate cancer model via STAT3 signaling [242, 243]. β-Caryophyllene oxide has been shown to suppress tumor growth and support apoptosis by suppressing ROS-mediated activation of MAPKs [75, 244].

Chemotherapeutic drugs or cytotoxic agents

Anthracyclines and topoisomerase inhibitors such as doxorubicin, adriamycin, daunorubicin, and epirubicin have been reported with anticancer activity in both solid and blood cancers, because these drugs can block DNA synthesis, topoisomerase II activity and complex I/II leading to increase in the production of mitochondrial ROS [245, 246]. Salvicine (SAL) is a known topoisomerase II poison that has been successful in clinical trials for cancer patients. SAL triggers H₂O₂ production, DNA double-strand breaks which induce G2M arrest and apoptosis in cervical carcinoma, leukemia and gastric carcinoma [247, 248]. Platinum-based drugs including cisplatin, carboplatin, oxaliplatin and other alkylating drugs are known for maintaining very high levels of ROS to induce cell death in several human malignancies [249, 250]. On the other hand, nucleotide analogs, antimetabolites, taxanes, and alkaloids treatments eliminate cancer cells by maintaining low ROS. The 5-fluorouracil (5-FU) is FDA approved for the treatment of patients with various malignancies. 5-FU sensitizes the tumors by producing mROS in a p53-dependent fashion [251]. Vorinostat displayed effective antitumor activity against BRAF and or MEK inhibitors resistant to melanoma in clinical trials. Treatment with vorinostat suppresses SLC7A11 which enhances ROS levels and induces DNA damage and cell death [252]. Under normal conditions, DNA damage is sensed and corrected either by DNA single-strand break repair (SSBR) mechanism or double-strand break (DSB) repair pathways [253]. PARP enzymes are essential for SSBR [253]. It is conceivable that loss of DNA damage repair due to PARP inhibitors can sensitize cancer cells to cisplatin- or carboplatin-induced oxidative stress [254, 255]. Interestingly, PARP inhibitors displayed synergy with cisplatin leading to increase in DNA damage as well as permeabilization of the mitochondrial membrane in lung carcinoma [253, 254]. More research is still required for a deeper and better understanding of clinical-grade ROS scavengers and inducers and will be beneficial for the treatment.

Conclusions

During the last five decades, our knowledge has greatly increased in context with the potential applicability of oxidative stress/ROS in normal physiological functions as well as in human malignancies. As we know, in the current scenario we use several toxic chemicals, preservatives, and plastics to process and preserve packed food items and color in food items, and have harmful practices such as excessive smoking and drinking. These are excellent sources of ROS right from birth and can lead to genomic instability, DNA mutations, activation of growth factor-mediated signaling, change in microbiota, metabolism and compromised immunity which ultimately lead to cancer and other diseases. Currently, with the advancement in novel technologies (DNA sequencing, metabolomics), we are starting to understand that even mutations in oncogenes and tumor suppressor genes induce oxidative stress/ROS. ROS are emerging as one of the key modulators of gut microbiota and tumor microenvironment. In future, modulation of ROS can be utilized to redefine or boost the immune response by releasing the immunosuppressive effect for better efficacy anticancer therapies. Moreover, this is very evident from many reports that ROS are involved in aberrant proliferation, tumorigenesis, angiogenesis, metastasis, and apoptosis through the activation of several signal transduction cascades including MAPK, PI3K, NF-kB, STAT3, HIF-1α, and ferroptosis. Importantly, in 2019, the Nobel Prize has been given for discovering the hypoxia-responsive pathway and how cell responds under varying oxygen levels by altering the transcription of HIF-1α regulated genes [140, 141]. Modulation of H₂O₂ through ROS scavengers in transformed cells has been shown to inhibit tumor growth and angiogenesis by blocking peroxide-dependent HIF-1α. On the other hand, several successful chemotherapeutic agents work by maintaining high ROS. Given the fact that ROS are critical for promoting tumorigenesis, ROS modulator or antioxidant has emerged as an alternative anticancer therapeutic and recently incorporated with chemotherapeutic drugs in clinical trials. Many studies were successful in reducing the tumor burden and provided proof of this concept in patients with late stages [256].

We must be a little careful, knowledgeable and considerable while using ROS modulator because ROS levels are crucial for the alimony of normal cells especially stem cells. ROS may be used as a biomarker for assessing the drug response where the aim of the chemotherapy drugs is to increase the ROS. One can think that ROS not only targets tumor cells, but also activate other cells in the tumor such as immune cells, macrophage, microbiota. This is what is required for a successful antitumor therapy and to overcome the drug resistance.

Abbreviations

5-FU

5-Fluorouracil

ABC

ATP-binding cassette

AML

Adult acute myeloid leukemia

AMPK 5′

AMP-activated protein kinase

APAF1

Apoptosis protease-activating factor 1

ATP

Adenosine triphosphate

ART

Artesunate

ASK-1

Apoptosis signal-regulated kinase 1

BSO

Buthionine sulfoximine

BCL-2

B cell lymphoma 2

CAR

Chimeric antigen receptor

CAT

Catalase

CSCs

Cancer stem cells

EGFR

Epidermal growth factor receptor

EGF

Epidermal growth factor

ER

Endoplasmic reticulum

ERK

Extracellular regulated kinase

ETC

Electron transport chain

EMT

Epithelial–mesenchymal transition

eNOS

Endothelial nitric oxide synthase

GPX

Glutathione peroxidase

GSH

Glutathione

GCL

Glutamate-cysteine ligase

GSS

GSH synthetase

GSSG

GSH disulfide

GPX4

Glutathione peroxidase 4

H₂O₂

Hydrogen peroxide

HCC

Hepatocellular carcinoma

HER2

Human epidermal growth factor receptor 2

HGF

Hepatocyte growth factor

HIF-1

Hypoxia-inducible factor

hTERT

Human telomerase reverse transcriptase

IDH1

Isocitrate dehydrogenase 1

IDH2

Isocitrate dehydrogenase 2

IL-6

Interleukin 6

JNK

C-Jun N-terminal kinase

LDH

Lactate dehydrogenase

LSC

Leukemic stem cells

MAPK

Mitogen-activated protein kinase

MDSC

Myeloid-derived suppressor cell

mROS

Mitochondrial reactive oxygen species

mtDNA

Mitochondrial DNA

NADH

Nicotinamide adenine dinucleotide

NF-κB

Nuclear factor kappa-light-chain-enhancer of activated B cells

NO

Nitrogen oxide

NOS

Nitric oxide synthase

NOX

NADPH oxidase

NRF2

Nuclear factor erythroid 2-related factor 2

O2•

Superoxide

OH•

Hydroxy radical

OXPHOS

Oxidative phosphorylation

PRX

Peroxiredoxins

PDAC

Pancreatic ductal adenocarcinoma

PD-1

Programmed death protein 1

PD-L1

Programmed death ligand 1

PD-L2

Programmed death ligand 2

PDGFR

Platelet-derived growth factor receptors

PDGF

Platelet-derived growth factor receptors

PI3K

Phosphoinositide 3-kinases

PML

Promyelocytic leukemia

PTEN

Phosphatase and tensin homolog

RTK

Receptor tyrosine kinase

RNS

Reactive nitrogen species

ROS

Reactive oxygen species

SAL

Salvicine

SOD

Superoxide dismutase

SSZ

Sulfasalazine

STAT3

Signal transducer and activator of transcription 3

TF

Transcription factor

Treg

Regulatory T cells

TAM

Tumor-associated macrophages

TFAM

Mitochondrial transcription factor A

TMZ

Temozolomide

TNBC

Triple-negative breast cancer

UCP-2

Uncoupling protein 2

Author Contributions

MG conceived the idea and designed the format of the manuscript. MG, AK, and GS wrote the manuscript and presented the concepts in the manuscript. MG and AK created the figures and the tables. MG, AK, and GS revised the manuscript and agreed to the published version of the manuscript.

Compliance with ethical standards

Conflict of interest

All the authors have read the manuscript and have no competing interests.