Reactive oxygen species (ROS), when present at low to medium levels, are known to stimulate cell proliferation by activating crucial signaling pathways like MAPKs and the phosphoinositide 3-kinase (PI3K)–AKT pathway. Therefore, it is not surprising that cancer cells are characterized by sustained ROS production, which fuels their uncontrolled growth. The activation of oncogenic pathways amplifies the generation of intracellular ROS, which, in turn, further activates oncogenes, creating a vicious cycle that accelerates cell proliferation and enhances the aggressiveness of cancer cells, ultimately impacting patient outcomes. The ROS rheostat theory posits that the coordinated action of oncogenic pathways on both ROS production and elimination is crucial in reprogramming various cellular functions that support different stages of tumorigenesis.
Current understanding points to two primary oncogene-mediated mechanisms that influence ROS production through the Etc Electron Transport Chain: (i) increased supply of carbon sources to the TCA cycle, leading to enhanced production of NADH and FADH2, which increases the flow of electrons through the ETC (Mechanism A, Fig. 1a); (ii) destabilization of electron transfer within the ETC, promoting electron leakage at complexes I, II, and III (Mechanism B, Fig. 1b).
RAS
The RAS family, encompassing small GTPases like KRAS, HRAS, and NRAS, plays a critical role in transmitting external signals (e.g., growth factor binding to receptors) that encourage cell proliferation and survival. Mutations at codons 12, 13, or 61 of RAS genes result in the continuous activation of RAS signaling in cancer cells.
The persistent activation of KRAS in human cancers orchestrates significant metabolic reprogramming that affects mitochondria, resulting in ROS generation via Mechanism A (Fig. 1). Oncogenic KRAS signaling boosts glutamine catabolism, which feeds the TCA cycle, thereby increasing mitochondrial ROS production and promoting anchorage-independent growth in colon cancer cells. Notably, this effect appears to be mediated by mitochondrial, rather than cytosolic, ROS, and a functional ETC was essential for KRAS-driven lung tumorigenesis in vivo. Anchorage-independent growth was eliminated in ρ0 cells, lacking mitochondrial DNA, while cybrids with a mutated cytochrome b gene restored O2•– production and anchorage-independent growth. These observations suggest that the oncogenic potential of KRAS is linked to O2•– production from the QO site of complex III within the electron transport chain. Similarly, Liou et al. observed that oncogenic KRAS induces the transformation of pancreatic acinar cells into pancreatic intraepithelial neoplasia through mitochondrial ROS-mediated activation of NF-κB, which drives the transcription of EGFR and its ligands, epidermal growth factor (EGF) and transforming growth factor α (TGFα). Interestingly, the study also demonstrated that MitoQ, a mitochondria-targeted antioxidant, prevents pancreatic cancer development in mice with KRAS mutations, indicating the necessity of mitochondrial ROS generation for KRAS oncogenic activity. Weinberg et al. reported similar antitumor effects using mitochondria-targeted O2•– scavengers MCP and MCTPO. The importance of ROS in KRAS-mediated tumorigenesis is further supported by the fact that mitochondria-targeted drugs like Mito-CP (carboxy proxyl nitroxide) and Mito-Metformin inhibit the proliferation of colon cancer cells.
Son et al. demonstrated that KRAS-driven pancreatic ductal adenocarcinoma relies on glutamine catabolism to produce aspartate, which is channeled into the aspartate transaminase (GOT1)–malic enzyme 1 (ME1) axis, a major source of NADPH. In this context, glutamine deprivation leads to oxidative stress and reduced tumorigenicity, effects that are reversed by glutathione and N-acetylcysteine (NAC). It’s important to note that these findings do not contradict observations by Weinberg et al. and Liou et al. The fact that glutamine is required for maintaining endogenous antioxidant systems does not exclude its role in fueling the TCA cycle, thereby increasing mitochondrial ROS generation from the ETC electron transport chain. Consistent with these observations, oncogenic KRAS promotes tumorigenesis by activating nuclear factor erythroid 2-related factor 2 (NRF2), the master regulator of antioxidant responses. In KRAS-driven pancreatic ductal adenocarcinoma with mutant KRAS, glutamine is crucial for both inducing cancer-promoting ROS production and fueling antioxidant pathways, resulting in an elevated homeostatic ROS set point.
Mutant KRAS (G12V) also translocates to mitochondria and disrupts electron transport, thus promoting ROS production via Mechanism B (Fig. 1). Baracca et al. observed that fibroblasts transformed with KRAS showed reduced oxygen consumption rates when supplied with glutamate–malate as a respiratory substrate, suggesting decreased complex I activity within the electron transport chain. Oxygen consumption was not reduced when succinate was used instead of glutamate–malate, indicating that complexes II, III, and IV activity remained unchanged. The complex I defect resulted in inefficient electron transport, due to the loss of supercomplex assembly, a condition potentially exacerbated by the general effects of ROS on respirasome assembly.
In contrast, Chun et al. observed that disrupting oncogenic KRAS led to decreased expression of genes involved in mitochondrial phospholipid synthesis—ACSL5, PCK2, and AGPAT7. The functional consequence was a reduction in cardiolipin synthesis, a phospholipid that facilitates supercomplex assembly, thus optimizing respiration within the electron transport chain. By promoting cardiolipin synthesis in mitochondria, oncogenic KRAS may enhance the efficiency of electron transport and reduce ROS production by the ETC. However, it remains unclear whether all KRAS-mutated tumors exhibit increased cardiolipin levels in mitochondria. These seemingly paradoxical effects may be explained by the different experimental setups used in studies by Weinberg et al. and Baracca et al., who used healthy cells transfected with mutant KRAS, versus Chun et al., who used KRAS-mutated colon cancer cells (HCT116) where mutated KRAS was removed. In the latter case, the impact of oncogenic KRAS on ETC activity was studied in a heavily mutated genetic background, where oncogenic KRAS acquisition could have a protective role by promoting cardiolipin synthesis and reducing ROS generation driven by other oncogenes. Conversely, introducing mutant KRAS into normal cells allowed for a more direct investigation of mutant KRAS effects on the ETC. Further research is needed to clarify the role of oncogenic KRAS signaling on supercomplex assembly and respiratory efficiency and its potential impact on cancer.
Recent studies suggest that KRAS effects on redox homeostasis are essential for maintaining the cancer phenotype and may represent promising therapeutic targets for KRAS-driven cancers, which remain a significant clinical challenge due to the lack of effective therapies. In this regard, Shaw et al. demonstrated that inducing oxidative stress with the small molecule lanperisone kills mouse embryonic fibroblasts transfected with mutant KRAS and inhibits their growth in vivo. Interestingly, Iskandar et al. observed that hyperactivation of mutant KRAS with the small molecule C1 leads to PI3K–AKT pathway activation, enhancing ROS generation (see below), mitochondrial dysfunction, cell death, and blockade of tumors with mutant KRAS. These effects are mitigated by NAC, indicating that ROS generation through the KRAS–AKT axis is necessary for C1 cytotoxicity, supporting the feasibility of a ROS-based anticancer strategy to target KRAS-driven tumors.
MYC
The MYC family of transcription factors (CMYC, LMYC, and NMYC) regulates cell proliferation and apoptosis by controlling a large number of RNA polymerase I-, II-, and III-dependent genes. MYC amplification is frequently observed in neuroblastoma and breast, ovary, prostate, and uterine cancers, while CMYC-immunoglobulin translocation is a hallmark of Burkitt’s lymphoma.
Similar to RAS, MYC induces significant metabolic reprogramming in cancer cells by stimulating glycolysis, mitochondrial biogenesis, and glutaminolysis. Li et al. demonstrated that inducible MYC expression in the B-cell line P493-6 increases mitochondrial mass and enhances oxygen consumption rate, an indicator of ETC electron transport chain activity. These effects are partially mediated by MYC-induced mitochondrial transcription factor A (TFAM), potentially driving ROS production through increased electron flow via the ETC.
Observations by Wise et al. in glioma cell lines indicated that MYC controls a transcriptional program that promotes glutamine catabolism as a carbon source to fuel the TCA cycle, thereby sustaining ROS production by the ETC. Vafa et al. observed that CMYC overexpression in human fibroblasts increased ROS levels, correlating with DNA damage foci formation; the antioxidant NAC reduced both effects. As MYC promoted cell-cycle entry and proliferation even with DNA damage, these findings suggest a mechanism for MYC-induced genomic instability and selection for p53 loss (a common alteration in CMYC-driven tumors), both of which fuel clonal evolution and tumor progression. The role of ROS production following MYC amplification is supported by the fact that exogenous antioxidants (vitamin C and Tiron) inhibited transformation of MYC-overexpressing NIH/3T3 fibroblasts. Moreover, blunting ROS with mitochondria-targeted vitamin E blocked proliferation and induced cell death in osteogenic sarcoma cells.
In chemotherapy-resistant triple-negative breast cancer, MYC upregulation along with the anti-apoptotic protein MCL1 selects for a stem-cell phenotype dependent on mitochondrial respiration. MCL1 accumulation in the mitochondrial matrix enhances the electron transfer capability of complexes I, II, and IV. The combined action of MYC on mitochondrial mass and MCL1 on the ETC resulted in HIF-1α stabilization, cancer stem cell selection, and chemotherapy resistance. However, similar to KRAS, MYC effects on ROS homeostasis are balanced, as MYC can also upregulate mitochondrial peroxiredoxin 3 to protect cells from ROS in hypoxia. Furthermore, NADPH production through serine and one-carbon metabolism protects hypoxic breast cancer stem cells from oxidative stress. This evidence suggests that ROS triggers the selection of cancer stem cells through upregulated antioxidant defenses, consistent with the lower ROS set point of cancer stem cells.
MYC may also act through Mechanism B by upregulating the expression of several mitochondrial nuclear-encoded proteins, leading to an imbalance between ETC subunits coded by nuclear and mitochondrial genomes, resulting in misassembled respiratory complexes. Interestingly, Herrmann et al. observed a strong correlation between the progression from normal prostate epithelium to invasive prostate carcinoma and imbalanced nuclear-encoded versus mitochondrion-encoded subunits of complex IV within the electron transport chain. This suggests that misassembled respiratory complexes promote tumor progression. MYC also affects ROS production by influencing cancer cell metabolism. In particular, MYC promotes the expression of genes involved in nucleotide synthesis, including DHODH, triggering ROS generation by destabilizing electron flow across the ETC.
MYC overexpression is associated with increased proliferation rates in breast cancer, and MYC amplification in luminal A breast cancer is linked to poor survival and endocrine therapy resistance. Although these studies did not assess ROS involvement in MYC-driven breast cancer aggressiveness, high ROS levels are known to be associated with endocrine therapy resistance. The impact of MYC-driven ROS production on clinical cancer outcomes requires further study. While MYC remains a challenging therapeutic target, the dependence of MYC-driven tumors on NADPH production through serine and one-carbon metabolism suggests that inhibiting these pathways could be an effective anticancer strategy.
PI3K–AKT–mTOR
Consistent with their roles in apoptosis suppression, cell proliferation, metabolism, and anabolic reactions, the interconnected PI3K–AKT and mTOR pathways are hyperactivated in almost 40% of all human cancers.
As the PI3K–AKT–mTOR pathway is central to the Warburg effect induction and autophagy inhibition, cancers with hyperactivation of this pathway accumulate dysfunctional, ROS-producing mitochondria that are not eliminated by autophagy.
Non-essential amino acid metabolism is a key feature of metabolic reprogramming in cancer and highly proliferating cells. Cancer cells metabolize non-essential amino acids to obtain nucleotides and lipids, maintain redox homeostasis, and control epigenetic regulation. Among the 11 non-essential amino acids, glutamine, serine, and proline are crucial in tumorigenesis. Proline synthesis, mediated by Δ1-pyrroline-5-carboxylate (P5C) reductases (PYCRs), and proline degradation through PRODH, form a “proline cycle” between the cytosol and mitochondria. In EGFR-mutated non-small-cell lung cancer, constitutive downstream PI3K pathway activation drives proline synthesis, fueling EGFR-regulated proline oxidation. PRODH activity reduces mitochondrial electron transport chain efficiency, driving ROS production from the QO site of complex III via Mechanism B. These findings suggest proline metabolism’s important role in non-small-cell lung cancer with EGFR mutations.
Although the PI3K–AKT–mTOR pathway is linked to aerobic glycolysis induction, Marchi et al. provided a mechanism by which AKT could mediate ROS generation via Mechanism A. They observed that mitochondria-localized AKT phosphorylates MICU1, a regulatory subunit of the mitochondrial calcium uniporter (MCU), destabilizing the MICU1–MICU2 heterodimer and increasing mitochondrial calcium influx. Decreased MICU1 expression following phosphorylation was associated with increased mitochondrial ROS levels and enhanced in vivo cancer cell growth. Mitochondrial calcium overload could cause increased ROS generation through mitochondrial dysfunction, potentially triggering permeability-transition-pore-mediated cell death. Alternatively, elevated mitochondrial calcium levels could drive ROS production by stimulating TCA cycle and OXPHOS enzymes, accelerating oxygen consumption and ROS generation by the ETC via Mechanism A. Moreover, mitochondrial calcium also increases GPDH activity, producing ROS by direct electron leakage and inducing RET towards complex II and complex I (Mechanism B). The tumor suppressor PTEN, inactivated in several human cancers, controls PI3K–AKT–mTOR pathway activity. Observations by Marchi et al. emphasize PTEN status’s central role in AKT-mediated MICU1 effects. Interestingly, ROS inactivates PTEN through oxidation, favoring AKT activation, which phosphorylates MICU1, leading to mitochondrial calcium uptake, and calcium drives ETC-derived ROS production, establishing a vicious cycle sustaining AKT activation and boosting tumor progression. Consistent with this, scavenging mitochondrial O2•– with MitoTEMPO blunted AKT activation, reversed the Warburg effect, and induced death in melanoma cells. Moreover, Variar et al. observed that the mitochondria-targeted antioxidant Mito-CP enhances apoptotic cell death in a Burkitt’s lymphoma cell line by decreasing AKT activation and HIF-1α stabilization under hypoxia, suggesting the potential to block hypoxic adaptation in cancer cells by reducing ETC electron transport chain-generated ROS.
In a breast cancer cell study, Jin et al. recently demonstrated that PI3K–AKT-mediated inactivation of glycogen synthase kinase-3β (GSK-3β) through phosphorylation induces abnormal activity of complexes I and III, altering electron flow and enhancing ROS production via Mechanism B. The resulting ROS released in the tumor microenvironment impaired NK cell cytotoxic activity by oxidizing a serine residue in initiation factor eIF2B, downregulating NKG2D and its ligands. These results provide evidence for an AKT/ROS-mediated mechanism to inhibit innate immune response in the tumor microenvironment. Interestingly, GSK-3β inactivation also stabilizes CMYC, which can further enhance electron transport chain ROS generation. Whether AKT-mediated GSK-3β inhibition requires ROS-mediated PTEN inactivation remains to be elucidated.
mTOR is a central kinase integrating energy sensing and anabolic pathways, coordinating protein synthesis and cell growth. Novel cellular component synthesis is energy-consuming, so mTOR promotes mitochondrial metabolism by indirectly increasing nuclear-encoded mitochondrial protein levels. For instance, mTOR promotes functional complex formation between transcription factor yin-yang 1 (YY1) and its cofactor peroxisome-proliferator-activated receptor coactivator 1α (PGC-1α), driving expression of many genes encoding mitochondrial proteins, including cytochrome c, resulting in increased mitochondrial respiration and ROS production via Mechanism A. Furthermore, mTOR cooperates with estrogen-related receptor α to promote transcription of genes involved in OXPHOS and the TCA cycle. Additionally, through 4E-BP protein inactivation, mTOR upregulates nuclear-encoded subunits of respiratory complex I and ATP synthase, increasing mitochondrial respiration. Goo et al. observed that in PTEN inactivation, hyperactivated AKT is associated with 4E-BP1 phosphorylation, increased activity of complexes I, III, and IV, and augmented oxygen consumption. These results suggest that mTOR could enhance mitochondrial respiration and thus, ROS production via Mechanism A. PI3K–AKT–mTOR pathway hyperactivation could also lead to imbalances between nuclear-encoded and mitochondrial-encoded respiratory complex subunits, as seen with MYC, resulting in misassembled complexes, supercomplex loss, and increased ROS production via Mechanism B from the electron transport chain.
Given its central role in metabolism coordination, mTOR is at the intersection of anabolic pathways and cancer cell mitogenic signaling. mTORC1 phosphorylates S6K1, which activates carbamoyl-phosphate synthetase 2, aspartate transcarbamoylase, dihydro-orotase (CAD) by phosphorylation on S1859. CAD is a rate-limiting enzyme catalyzing the first three steps of de novo pyrimidine synthesis, generating dihydro-orotate from glutamine, thus fueling DHODH, which produces ROS. Notably, dihydro-orotate oxidation to orotate integrates nucleotide synthesis, necessary for uncontrolled cancer cell proliferation, with ETC electron transport chain ROS production, driving cancer initiation and progression. Besides mTOR, KRAS, MYC, AKT, and other oncogenes also converge on pyrimidine synthesis. Interestingly, Hail et al. observed that DHODH inhibition via terifluonomide decreases mitochondrial ROS levels and has a cytostatic effect in prostate cancer cells. Future studies should investigate whether decreased nucleotide pool or ROS levels account for DHODH inhibitor anticancer activity, and the role of DHODH-produced ROS in cancer development.
PI3K–AKT–mTOR pathway hyperactivation is a poor prognosis marker in several human cancers, such as esophageal squamous cell carcinoma and breast cancer. Yu et al. recently observed that pterostilbene, an antioxidant compound in blueberries, slows mantle cell lymphoma progression by targeting the PI3K–AKT–mTOR pathway. However, pterostilbene also directly affects apoptosis and the cell cycle, complicating result interpretation; the impact of ROS on cancer in PI3K–AKT–mTOR hyperactivation warrants further investigation.
BCR/ABL
The t(9,22) translocation, pathognomonic of chronic myeloid leukemia, creates the Philadelphia chromosome and the chimaeric gene BCR/ABL, coding for a constitutively active tyrosine kinase that transmits mitogenic and anti-apoptotic signals to cancer cells.
The BCR/ABL fusion protein promotes electron transport chain ROS production partly by activating the PI3K–mTOR pathway. Kim et al. observed that glucose metabolism is involved in ROS generation in BCR/ABL-transformed cells. Treatment with 2-deoxyglucose, the BCR/ABL inhibitor imatinib mesylate, or rotenone reduced ROS production. Similar effects were seen with wortmannin and rapamycin, inhibiting PI3K and mTORC1, respectively, indicating tight connections between BCR/ABL, PI3K, mTOR, glucose metabolism, and ETC ROS production via Mechanism A. ROS level reduction following rotenone treatment suggests RET (Mechanism B) involvement in BCR/ABL ROS generation. Direct BCR/ABL action on ETC activity has also been observed. Nieborowska-Skorska et al. observed a sharp decrease in electron flow rates between complexes I and II and II and III, with increased O2•– production via Mechanism B, in BCR/ABL-expressing myeloid precursors. This ROS production was reduced by the mitochondria-targeted antioxidant MitoQ and sustained by complex III, as shown by the rescue effect of complex III inhibitors myxothiazol, stigmatellin, and antimycin A. The small GTPase Rac2 promoted complex III ROS generation, as Rac2 knockout significantly reduced mitochondrial O2•– levels and oxidative stress in BCR/ABL-expressing cells. The authors noted this mechanism underlies ETC ROS production in leukemia cells with various genetic alterations (FLT3–ITD, TEL–ABL1, TEL–JAK2, TEL/PDGFβR, TEL–TRKC, BCR–FGFR1, and mutated JAK2). These observations indicate that different leukemia cell types may promote genomic instability and progression by activating Rac2, which interferes with electron transport from complexes I to III and II to III, ultimately inducing electron leakage from complex III and O2•– generation from the electron transport chain.
BCL-2
The B-cell lymphoma-2 (BCL-2) family includes proteins counteracting intrinsic apoptosis by binding pro-apoptotic proteins along the IMM. Many tumor types, including breast cancer, prostate cancer, B-cell lymphomas, and colorectal adenocarcinomas, exhibit BCL-2 overexpression. Several lines of evidence suggest that BCL-2 upregulation, beyond directly blocking apoptosis, creates a pro-oxidant state promoting cell survival. Chen and Pervaiz showed that BCL-2 overexpression in leukemia cells regulates mitochondrial respiration by affecting ETC electron transport chain activity through direct BCL-2 interaction with complex IV subunits Va and Vb, promoting correct complex assembly in the IMM and upregulating its activity. Based on these findings, they suggested that BCL-2 increases ETC activity and O2•– production, leading to a pro-oxidant environment favoring cell survival. Interestingly, during hypoxia, with BCL-2, subunit Va abundance is higher than Vb, reducing complex IV activity and mitochondrial respiration, maintaining mitochondrial redox state unchanged, preventing oxidative stress that would otherwise favor cell death. This is supported by the observation that decreasing O2•– levels triggers apoptosis in BCL-2-overexpressing cancer cells and suggests that O2•– and H2O2 balance could modulate cancer cell survival through O2•–-mediated apoptosis inhibition. Furthermore, establishing a mild pro-oxidant environment prevents BCL-2 dephosphorylation on S70, improving BCL-2 binding with BAX and cell survival, potentially linked to the electron transport chain.
