**6. Redox alterations in liver cancer**

#### **6.1. Redox and carcinogenesis**

Hepatocellular carcinoma (HCC) is a relevant problem of public health since it is in the sixth place in incidence, and the second in mortality at worldwide level [82]. The main risk factors conducing to HCC are viral hepatitis, steatosis and cirrhosis [83]. HCC could be considered the final stage of chronic liver disease (CLD), characterized by a persistent presence of inflam‐ mation and oxidative stress [84].

One of the best‐characterized effects of redox changes is related to hepatic carcinogenesis [85]. The accumulation of ROS in early stages of hepatic damage produces lipid peroxidation of the cellular membranes; the reaction of ROS with the double bounds of polyunsaturated fatty acids results in the generation of aldehydes as 4‐hydroxynonenal (4‐HNE), acrolein, crotonaldehyde, and malond‐ ialdehyde, all with the capability of forming DNA adducts and genetic instability [86, 87]. 4‐HNE forms exocyclic guanine adducts (4‐HNE‐dG) and consequently induces carcinogenic mutations [85]. These mutations affect loci of oncogenes or tumor suppressors conduce to malignant transfor‐ mation. An example is the demonstration that 4‐HNE induce G‐C to A‐T changes in the codon 243 of the p53 locus, an extensively studied tumor suppressor [88], promoting phenotypic loss [89, 90].

In this context, mechanisms activated by oxidative stress have the capability to influence a variety of proteins whose function impact in cancer, by the ability to oxidize thiol residues in

**Figure 5.** Effect of portacaval anastomosis on conjugated dienes of liver homogenate and subcellular fractions. Sham, false‐operated rats (white bars); shunt, rats with portacaval anastomosis for 6‐8 weeks (black bars). Data are average ± SEM from at least eight independent observations. \* Significant statistical difference by *t*‐student test, *p* < 0.05.

a process known as "redox priming" [91]. Examples of these redox sensors are the factor nf‐κB [92], oncogenes or tumor suppressors as p53 [93] and Src tyrosine kinase [94].

Interestingly, accumulated evidence suggests that antioxidant systems activated in response to oxidative stress, improve the proliferation rate and protect cancer cells from a hostile envi‐ ronment [95, 96]. For example, it was demonstrated that prostaglandin 1 reductase‐1 (Ptgr1), an oxidoreductase involved in the catabolism of eicosanoids and LPO‐derived compound as 4‐HNE, whose expression is regulated by the transcription factor (erythroid‐derived‐2)‐like‐2 (NRF2) [97], is overexpressed in human HCC biopsies and in samples from experimental animals. It was documented that Ptgr‐1 regulates positively the proliferation rate of cells and improves their survival in two models of experimental HCC [97], suggesting that this antioxi‐ dant response plays a protumoral role in HCC.

#### **6.2. Redox and hepatocellular carcinoma**

**6. Redox alterations in liver cancer**

Hepatocellular carcinoma (HCC) is a relevant problem of public health since it is in the sixth place in incidence, and the second in mortality at worldwide level [82]. The main risk factors conducing to HCC are viral hepatitis, steatosis and cirrhosis [83]. HCC could be considered the final stage of chronic liver disease (CLD), characterized by a persistent presence of inflam‐

**Figure 4.** Effect of portacaval anastomosis on thiobarbituric acid reactive substances (TBARS) assay of liver homogenate, subcellular fractions and serum. Sham, false‐operated rats (white bars); shunt, rats with portacaval anastomosis for

Significant statistical

6‐8 weeks (black bars). Data are average ± SEM from at least eight independent observations. \*

One of the best‐characterized effects of redox changes is related to hepatic carcinogenesis [85]. The accumulation of ROS in early stages of hepatic damage produces lipid peroxidation of the cellular membranes; the reaction of ROS with the double bounds of polyunsaturated fatty acids results in the generation of aldehydes as 4‐hydroxynonenal (4‐HNE), acrolein, crotonaldehyde, and malond‐ ialdehyde, all with the capability of forming DNA adducts and genetic instability [86, 87]. 4‐HNE forms exocyclic guanine adducts (4‐HNE‐dG) and consequently induces carcinogenic mutations [85]. These mutations affect loci of oncogenes or tumor suppressors conduce to malignant transfor‐ mation. An example is the demonstration that 4‐HNE induce G‐C to A‐T changes in the codon 243 of the p53 locus, an extensively studied tumor suppressor [88], promoting phenotypic loss [89, 90]. In this context, mechanisms activated by oxidative stress have the capability to influence a variety of proteins whose function impact in cancer, by the ability to oxidize thiol residues in

**6.1. Redox and carcinogenesis**

difference by *t*‐student test, *p* < 0.05.

168 Redox - Principles and Advanced Applications

mation and oxidative stress [84].

A role for redox reactions has been detected during the HCC, from the proliferation of initial cancer cells to the dissemination process [96]. At the beginning of CLD, cells have a high proliferation rate, since the tumor growth demands nutrient supply and metabolic condition‐ ing; in consequence, important architectural changes occur including the formation of new

**Figure 6.** Effect of portacaval anastomosis on thiobarbituric acid reactive substances (TBARS) assay supplemented with FeSO4 50 (μM) of liver homogenate, subcellular fractions and serum. Sham, false‐operated rats (white bars); shunt, rats with portacaval anastomosis for 6‐8 weeks (black bars). Data are average ± SEM from at least eight independent observations. \* Significant statistical difference by *t*‐student test, *p* < 0.05.

blood vessels and the establishment of specific microenvironment within the tumor. These adaptations make the tumor cells adapt to highly variable O2 concentration environment with successive lapses of hypoxia‐reperfusion [98].

From the metabolic point of view, lack of O2 favors the onset of aerobic glycolysis, by the Warburg's effect [99]; hypoxia also induces adaptive responses as the expression of special‐ ized proteins named hypoxia‐inducible factors (HIFs) [100]. HIF‐1 is a transcription factor formed by two subunits (HIF‐1α and HIF‐1β). Although HIF‐1 is constitutively expressed in normoxic conditions, HIF‐α is constantly degraded by prolyl‐hydroxylases; hydroxylated residues serve as docking site for von Hippel‐Lindau tumor suppressor protein that is a con‐ stituent of an E3 ubiquitin ligase complex. The ubiquitinated HIF‐1α suffers proteasomal deg‐ radation [101]. Low oxygen tension and some proinflammatory cytokines stabilize HIF‐1α and allow its nuclear translocation to regulate key genes for the hypoxic response [100]. HIF‐1α mediates the expression of genes that supports tumor growth such as NADH dehydrogenase (ubiquinone) 1α subcomplex, 4‐like 2 (NDUFA4L2), a protein that attenuates the activity of the mitochondrial complex I, reducing the ROS production in low O2 conditions [102].

It was shown that the kinase inhibitor sorafenib, an antineoplasic extensively used in oncol‐ ogy with potent antiangiogenic effects, induces intratumor oxidative stress that favor drug resistance in HCC; the insensibility to the drug requires the action of HIF‐1α regulating the expression of specific genes. For example, Nf‐κB induces the expression of the antiapoptotic protein Bcl‐2 and vascular endothelium growth factor (VEGF) is a potent inductor of angiogen‐ esis [103]. The resultant evasion of apoptosis and genesis of new blood vessels makes evident that protective cellular mechanisms are exploited during the progression of cancerous cells.

A role for ROS has also been described in the invasive process particularly in the epithelium to mesenchymal transition (EMT). EMT is a differentiation process consisting of the change from epithelial to mesenchymal characteristics. Epithelial cells are coupled through specialized structures as adherens junctions and exhibit apicobasal polarity, whereas in the mesenchymal phenotype the cells lose the coupling with neighbors and acquire migratory abilities [104]. EMT is considered the fundamental process in metastasis, being the cytokine TGF‐β an effec‐ tive EMT inductor as well as the growth factors EGF and PDGF [105]. These factors regulate the activity of the transcription factor SNAIL [106]. It has been described that in renal tubular epithelial cells, TGF‐β induces synthesis of H2 O2 and the onset to the EMT [107]. In hepatocar‐ cinoma‐derived cells HepG2, the stimulation with the phorbol‐ester TPA, favors the activation of protein kinase C (PKC), phosphorylation of ERK and accumulation of ROS, resulting in the induction of EMT and cellular migration. These effects were prevented by ROS scavengers, suggesting a key role for these molecules in the cell migration promoted by TPA [108].

Recently, it was shown that ROS regulate EMT and cell migration induction through the activity of eukaryotic translation initiation factor *eIF5A2*; the expression pattern in the HCC‐ derived cell line SUN449 correlated with those modified by the knock down of *eIF5A2*, strongly suggesting that eIF5A2 is an effector of ROS signaling [109].

Hepatocellular tumoral cells show a variety of adaptive mechanism in extreme environmental conditions (hypoxia, low nutrients) to continue their growth and progression. Thus, under‐ standing carcinoma cells biology requires yet exhaustive research and integrative efforts of the available data and an intimate knowledge of redox regulation.
