Undue elevation of ROS levels commonly occurs during cancer evolution as a result of various antitumor therapeutics and/or endogenous immune response. Overwhelming ROS levels induced cancer cell death through the dysregulation of ROS-sensitive glycolytic enzymes, leading to the catastrophic depression of glycolysis and oxidative phosphorylation (OXPHOS), which are critical for cancer survival and progression.
1. Introduction
Redox homeostasis is essential to maintain the normal structure and functions of cellular components, but oxidative stress frequently occurs in cancer cells as a result of oncogene activation, hypoxia, inflammation, and therapeutics
[1][2][3][4]. Abrupt accumulation of reactive oxygen species (ROS) has detrimental effects on various components of cancer cells, leading to cellular dysfunction or even cell death
[5][6][7][8]. In particular, metabolic enzymes are sensitive to ROS, with the most noted examples being glyceraldehyde-phosphate dehydrogenase (GAPDH) and pyruvate kinase M2 (PKM2) in the glycolytic pathway
[9]. The wide application of 2-flourine-18[(18)F] fluoro-2-deoxy-D-glucose (FDG) positron emission tomography–computed tomography (PET–CT) demonstrated the glycolytic phenotype in most cancers. Therefore, ROS-induced oxidation and inactivation of GAPDH and PKM2 can cause the depression of both aerobic and anaerobic glycolysis, leading to decreased proliferation and/or cell death due to shortages of energy and tricarboxylic acid cycle (TCA)-derived biosynthesis, especially in cancer cells in early stages that are more dependent on glycolysis
[10][11][12]. Despite these reports linking overt ROS damage to metabolic pathways and other cellular components, it is noteworthy that cancer cells also adapt to such overwhelming ROS levels and metabolic impairment
[13][14]. It has been well documented that the oxidative pentose phosphate pathway (PPP) and the synthesis of reduced glutathione (GSH) are enhanced, largely contributing to the production of nicotinamide adenine dinucleotide phosphate (NADPH) and GSH, the most prominent antioxidant molecules
[15][16]. On the other hand, cancer cells tend to recruit carbon flux from lipids and glutamine into nucleotide synthesis through the non-oxidative PPP
[17][18][19] and into TCA-coupled oxidative phosphorylation (OXPHOS) and biosynthesis
[20][21], which meet substrate and energy requirements. This relies on the wide crosslinks in the metabolic pathways of glucose, lipids, and amino acids. This metabolic regulation is known to be driven by a complex network consisting of several metabolic modulators, including nuclear factor erythroid 2-related factor 2 (NRF2), hypoxia-inducible factors (HIFs), forkhead box proteins (FOXOs), nuclear factor kappa-B (NF-κB), and/or RAC-alpha serine/threonine-protein kinases (AKTs). Their activations depend on ROS levels and their specific inducers, suggesting the heterogeneity of metabolic adaptation under different pro-oxidant conditions.
Accordingly, metabolic regulation plays a central role in cancer adaptation to oxidative stress. Mounting evidence has shown that metabolic regulation, involving the activation of different metabolic modulators with oncogenic properties, metabolic reprogramming, and optimized ROS levels, is tightly linked with cell fate decisions in cancer
[22][23][24][25]. A better understanding of how cancer cells orchestrate these metabolic modulators to achieve stress adaptation has potential implications for developing redox- and metabolism-targeting therapeutic strategies.
2. Oxidative Stress in Cancer Cells
Oxidative stress in cancer cells is induced by various endogenous or exogenous pro-oxidant elements, such as hypoxia, inflammation, and numerous therapeutics. Upon these stimuli, overwhelming ROS may be produced through a host of oxidoreductases in several compartments of cells, primarily including mitochondria, cytoplasm, and the endoplasmic reticulum (ER), inter alia
[6][26].
Mitochondria contribute the most to both physiologically and pathologically endogenous ROS
[9][27][28]. Numerous studies have revealed that cancer cells exhibit remarkable plasticity in metabolic phenotypes
[29][30], with the most notable example being glucose catabolism, which impacts endogenous ROS generation
[31][32]. The selective switch to anaerobic glycolysis in rapidly proliferating cancer cells, even under sufficient nutrition and O
2 (Warburg effect), favors the acquisition of a ‘pro-oxidant state’ partly due to the diversion of pyruvate away from the mitochondria
[33][34][35]. Despite the previously held notion that cancer cells selectively rely on the Warburg effect, there is ample evidence to indicate that cancer cells can switch between glycolysis and OXPHOS to cater to impending energy demands
[33][34][36]. To this end, high OXPHOS-coupled aerobic glycolysis is also a potential source of increased ROS levels in cancer cells
[37][38][39]. For example, CEM leukemia and HeLa cervical cancer cells overexpressing BCL-2 were shown to have increased OXPHOS and mitochondrial ROS generation
[40][41][42]. This suggests that both hypo-functional and/or hyper-functional mitochondria are linked to the increased generation of ROS. Intriguingly, various antitumor drugs, as well as radiation, induce significant oxidative distress in cancer cells
[43][44][45][46], due in part to the impairment of mitochondrial function and metabolism
[40][45][47] (
Figure 1). For example, the doxorubicin, bleomycin, or platinum coordination complexe causes mitochondrial DNA damage or prevent DNA synthesis by inducing cellular oxidative distress
[47][48]. Furthermore, genetically unstable clones generated upon irradiation also displayed higher intracellular ROS levels, potentially due to reduced mitochondrial activity and respiration
[49][50]. In addition, cancer cells under hypoxia are associated with an increase in ROS, likely because of the deficiency in O
2 that prevents electron transfer across the mitochondrial complexes, thereby increasing the possibility of electron leakage to generate ROS
[27][51][52]. These findings indicate that the plasticity of mitochondrial function, under the control of various pro-oxidant elements, contributes significantly to ROS regulation in cancer cells.
Figure 1. Oxidative stress and metabolic impairment in cancer cells. Mitochondria contribute the most to both physiology- and pathology-derived ROS, of which a fraction is generated by tricarboxylic acid cycle (TCA) enzymes, while the major portion is produced along the electron transport chain (ETC) due to electron leakage at complexes I, II, and III. NOXs are another major source of ROS production in cancer cells. These ROS-generating enzymes function by transmitting one electron from cytosolic NADPH to O2 to produce a superoxide anion radical (O2•−) that can be transformed to H2O2 immediately by superoxide dismutase (SOD) families. There are seven human NOX homologues, NOX1–5, dual oxidase 1 (DUOX1), and DUOX2, distributed at the cell membrane, mitochondria, ER, and nucleus. They can be activated by a wide variety of ligands, such as tumor necrosis factor (TNF), angiotensin II, platelet-derived growth factor (PDGF), and pro-epidermal growth factor (EGF). The endoplasmic reticulum (ER) serves as a repository wherein nascent proteins are folded and modified, in which disulfide bond formation is essential for the primary structure of proteins and is catalyzed by protein disulfide isomerase (PDI), which can be reduced by ER oxidoreductases, represented by oxidoreductase 1 (ERO1), to generate H2O2 as a byproduct. Furthermore, several glycolytic enzymes are vulnerable to elevated ROS levels and can be inactivated by redox modification, such as glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and pyruvate kinase M2 isoform (PKM2), which collaboratively cause metabolic stress characterized by a deficiency of carbon sources for ATP generation, lactate secretion, and TCA-driven anabolic biosynthesis.
Cytoplasmic ROS-producing enzymes are represented by members of the NADPH oxidase (NOX) family, such as NOX1-4, whereby cancer cells respond to various extra-cellular stimuli
[53][54]. Several studies have shown the role of ROS generation in cancer cells, particularly from the standpoints of oncogene activation, chemoresistance, survival, inflammation, and metastasis. For instance, NOXs can be activated by a wide variety of ligands, such as tumor necrosis factor (TNF), angiotensin II, platelet-derived growth factor (PDGF), and pro-epidermal growth factor (EGF)
[55][56] (
Figure 1). Furthermore, the function of NOX1 in generating superoxide anion radicals (O
2•−) was shown to be critical in RAS-mediated cell transformation
[57]. Similarly, the introduction of the dominant negative mutant (N17) of GTPase RAC1, a subunit of NOX, reduced superoxide levels, thus inhibiting the growth of mutant KRAS-driven cells
[58].
The ER is also a ROS factory where nascent proteins are folded and modified to become functional
[59]. Disulfide bond formation is essential for the primary structure of proteins and is catalyzed by protein disulfide isomerase (PDI). In a normal state, the oxidized PDI can be reduced by ER oxidoreductases, represented by oxidoreductase 1 (ERO1), to generate H
2O
2 as a byproduct
[60], which maintains a high basic level of ROS in ER
[61][62]. Furthermore, ROS production and release from the ER are significantly increased during ER stress
[63][64]. When misfolded proteins accumulate beyond a tolerable threshold within the ER, an unfolded protein response (UPR) is induced to mobilize protein-folding capacity and otherwise to promote cell apoptosis
[60][65]. ER stress is frequently documented in chemotherapies and radiotherapies because DNA and enzyme damage basically produce a large amount of abnormal proteins
[66][67]. In addition, NOX4 and NOX5 were found in the ER and act, in a way, as their cytoplasmic members. However, the restoration of ROS levels with the endogenous antioxidant N-acetylcysteine (NAC) did not significantly prevent cancer cell death
[68], indicating that ER stress-derived ROS seldom causes cancer cell death directly.