Keywords

21.1 Introduction

Reactive oxygen species (ROS), which include superoxide (O2 ), hydrogen peroxide (H2O2), and hydroxyl radicals (HO), are produced by the partial reduction of oxygen and are generated endogenously mainly during mitochondrial oxidative phosphorylation and exogenously predominantly from xenobiotic compounds. ROS levels are controlled through the activity of endogenous antioxidant defense systems such as superoxide dismutase (SOD), catalase, and peroxiredoxin, as well as through exogenous antioxidants such as isoflavones, catechins, carotenes, vitamins, and selenium [1]. Oxidative stress occurs when the cellular antioxidant defense systems are overwhelmed by an increase in ROS levels or a decrease in the antioxidant capacity. Excessive ROS levels lead to damage of macromolecules such as DNA, RNA, proteins, and lipids, which is in part rescued by the DNA repair system and the thioredoxin and glutathione detoxification systems [2]. Damage to DNA can cause genetic aberrations, such as mutations and chromosomal rearrangements, while damage to other molecules can affect epigenetic processes, largely through dysregulation of proteins containing redox-reactive cysteine residues. Oxidation of cysteine produces reactive sulfenic acid (–SOH), which forms disulfide bonds with nearby cysteine residues (–S–S–) or undergoes further oxidation to sulfinic (–SO2H) or sulfonic (–SO3H) acids. With the exception of –SO3H formation, each of these redox modifications can be reversed by reducing systems [3]. These oxidative modifications of cysteines alter the protein structure and function, thereby directly or indirectly affecting a range of events, including intracellular signal transduction and gene expression pathways that modulate various cellular processes (Fig. 21.1) [4].

Fig. 21.1
figure 1

Relationship between treatment resistance and oxidative stress

Oxidative stress not only plays an important role in prostate carcinogenesis and progression of prostate cancer [5,6,7] but also is involved in the resistance of prostate cancer to therapy, especially androgen deprivation therapy (ADT) [1, 7, 8]. ADT, which consists of surgical or pharmacological castration or anti-androgen therapy, has been commonly used for the treatment of advanced or recurrent prostate cancer since 1941 [9]. Although ADT is initially effective for most prostate cancer patients, therapy resistance invariably develops and the disease becomes lethal castration-resistant prostate cancer (CRPC). Increasing evidence suggests the existence of functional cross talk between oxidative stress and CRPC. Here, we summarize our current knowledge in this area.

21.2 Oxidative Stress Induced by ADT in Prostate Cancer

Several experiments in vitro and in vivo have indicated that castration leads to oxidative stress by promoting increased ROS production and decreased ROS-detoxifying enzyme activity [10,11,12,13]. However, there are also several conflicting studies showing that androgens can induce oxidative stress [14, 15]. This discrepancy may be due to differences in the physiological and nonphysiological conditions in the various studies (Fig. 21.2). For example, Ripple et al. demonstrated that oxidative stress was decreased or increased by physiological or excessive androgen levels, respectively, suggesting that stress can be induced nonspecifically under nonphysiological conditions [16]. Several molecular mechanisms may be responsible for castration-induced oxidative stress, as shown by a reduction in the antioxidant molecules thioredoxin 1, peroxiredoxin 5, and SOD2 in rats after castration [11], a reduction in SOD2 in human prostate cancer tissue after ADT [10], and upregulation of pro-oxidant nicotinamide adenine dinucleotide phosphate oxidases (Noxs) in rat prostate after castration [13]. Collectively, these studies suggest that epigenetic alterations in gene expression and protein function lead to a redox imbalance and induction of oxidative stress in prostate cancer; this is supported by the finding of elevated oxidative stress levels in prostate cancer cells and surgically resected prostate cancer tissues [8, 17]. Thus, ADT-induced oxidative stress can lead to wide-ranging genetic and epigenetic alterations in prostate cancer, as described in more depth in the following sections.

Fig. 21.2
figure 2

Dose-response relationship between androgen levels and oxidative stress

21.3 Effects of Oxidative Stress on AR and Non-AR Signaling

21.3.1 Effects of Oxidative Stress on AR Signaling

In CRPC, AR signaling is aberrantly augmented by the low androgen milieu via a number of mechanisms, including AR overexpression, AR activation by co-regulators and intracellular signal transduction pathways, mutation of AR and AR-related proteins, expression of AR splice variants, and de novo androgen synthesis. Over the last decade, ADT-induced oxidative stress has been shown to influence AR signaling in prostate cancer. Sharifi et al. showed that suppression of the antioxidant enzyme SOD2 and increased ROS production activated AR signaling through changes in the expression of genes related to steroid metabolism, nuclear receptor co-regulators, and interleukin-6 receptor [7]. We also independently found that ROS play a crucial role in AR signaling and the development of CRPC [1, 8]. Thus, oxidative stress could contribute to castration resistance through AR reactivation by several mechanisms.

21.3.1.1 AR Overexpression

AR overexpression is thought to be a major cause of CRPC [18]. Indeed, many studies have shown that CRPC progression is associated with increased AR expression [19,20,21,22], which may be attributed to gene amplification, increased transcription and translation, and decreased degradation. Among these, transcriptional upregulation is a particularly important mechanism of increased AR expression. As we summarized previously [1, 18], several transcription factors activated by oxidative stress, including Twist1 [8], YB-1 [23], NF-κB [24], Sp1 [25, 26], Myc [27, 28], CREB [29], and Foxo3a [30], are also known to regulate AR expression, suggesting that ADT-induced oxidative stress may act through these factors to upregulate AR transcription [8].

Many other molecules have also been reported to be involved in regulating AR expression. For example, a pathway linked to 12-hydroxyeicosatetraenoic acid and leukotriene B4 receptor 2 was shown to increase ROS production and upregulate AR expression via the Nox4 pathway [31]. Conversely, treatment with diphenyleneiodonium chloride, an antioxidant that inhibits Nox-mediated ROS production, reduced AR expression via SREBP-1 [32]. The oxidative stress inducers cadmium and zinc chloride increase AR expression in dysplastic prostate glands of rats [33], while the synthetic antimicrobial chemical mequindox induces oxidative stress and AR overexpression in rat testes [34]. Paradoxically, other inducers of oxidative stress, such as a curcumin analog [35] and thymoquinone [36], were reported to suppress AR expression. However, these agents may act through non-redox signaling since the effects were poorly suppressed by the antioxidant N-acetyl cysteine (NAC), an electrophile that supports the production of a major intracellular antioxidant, glutathione.

Collectively, these data suggest that oxidative stress induced by internal and external stimuli induces AR overexpression through stress-induced transcription factors and other pathways.

21.3.1.2 AR Activation by Co-regulators and Intracellular Signal Transduction Pathways

The transcriptional activity of AR is modulated by co-regulators [37], several of which, including peroxiredoxin, Hsp27, and EGR-1, are activated by oxidative stress [1]. We previously showed that cysteine residues in peroxiredoxin are critical for its AR co-regulatory function [1], supporting the possibility that ROS-mediated modification of AR co-regulators affects AR signaling.

In addition, several intracellular signaling pathways play a role in AR transactivation. AR function can be augmented by growth factors and cytokines such as insulin-like growth factor, fibroblast growth factor, epidermal growth factor, and IL-6, as well as key components of their downstream signaling pathways, such as mitogen-activated protein kinase (MAPK), JAK/STAT, protein kinase A, phosphatidylinositol-3-kinase (PI3K)/Akt, and protein kinase C, which may itself be activated by oxidative stress [1]. In fact, we have shown that the ε isoform of protein kinase C increases AR expression through NF-κB signaling and contributes to cellular resistance to castration [38, 39]. Thus, oxidative stress also influences intracellular signaling pathways that interact with transcription factors and co-regulators to modulate AR activity.

21.3.1.3 Mutation of AR-Related Proteins and Generation of AR Splice Variants

Mutations in the AR gene have been shown to change the protein’s ligand-binding affinity, permitting activation by non-cognate steroids and even by anti-androgen agents [6, 40, 41]. Although oxidative stress induces mutations in DNA, it is not yet known whether the AR gene is affected [6]. However, mutations in genes related to AR signaling, including FASN, CYP11B1, HSD17B4 (androgen metabolism), NCOR1, and FOXOA1 (AR cofactors), have been detected in CRPC tissues [42, 43]. Such mutations, probably induced by oxidative stress, may contribute to the development of CRPC through aberrant activation of AR signaling.

Several splice variants of AR exhibit transcriptional activity in the absence of androgen and play a key role in promoting CRPC [44,45,46,47,48]. Although possible, a relationship between expression of the AR splice variants and oxidative stress has not yet been documented. However, we recently reported that the redox-sensitive nuclear factor YB-1 [49] and its upstream kinase RSK [50] regulate the expression of an AR variant [51], supporting a direct link. Based on these intriguing observations, further studies of the effects of oxidative stress on mutation of the AR gene and expression of AR splice variants are warranted.

21.3.1.4 De Novo Androgen Synthesis

De novo synthesis of androgens in the adrenal glands and prostate tumors has been recognized as a potential cause of CRPC [52,53,54], and this was confirmed by clinical trials of abiraterone acetate, an inhibitor of a critical enzyme in androgen biosynthesis, cytochrome P17 (CYP17) [55, 56]. H2O2 regulates androgen synthesis in rat Leydig cells in a biphasic manner, indicating that physiological levels of oxidative stress promote steroidogenesis [57]. Nevertheless, there is no direct evidence at present for the existence of a relationship between oxidative stress and de novo androgen synthesis.

21.3.2 Effects of Oxidative Stress on Non-AR Signaling

In addition to AR signaling, numerous non-AR signaling pathways are activated by oxidative stress and many have been reported to be involved in the development to CRPC through genetic and epigenetic mechanisms. The genotoxic effects of oxidative stress include aberrations such as DNA point mutations and chromosomal rearrangements. In fact, genetic alterations in non-AR signaling molecules, such as PIK3CA, SPOP, RET, RICTOR, and CTNNB1, have been identified in tissues from patients with CRPC [42, 43, 58, 59].

Oxidative stress also causes epigenetic alterations that activate signaling independently of the AR [60]. For example, in prostate cancer, oxidative stress activates PI3K/Akt [61] and MAPK [62] and elevates the transcriptional activity of NF-κB [63], which promotes survival and inhibits apoptosis. However, many components of these pathways are also involved in AR signaling and show elevated activity in CRPC cells and tissues, as is the case for PI3K/Akt [64], MAPK [65], and NF-κB [66]. Additional non-AR-related mechanisms that contribute to the development of CRPC include inflammation, epithelial–mesenchymal transition, and cancer stem-like characteristics of prostate cancer cells [67]. Intriguingly, these phenomena are also affected by oxidative stress, further supporting the multiple mechanisms through which oxidative stress is involved in CRPC development.

21.4 Oxidative Stress and the Development of CRPC

As described above, the mutual link between oxidative stress and AR signaling supports a role for oxidative stress in CRPC development; indeed, there is direct evidence of such a relationship. We chronically exposed LNCaP, an androgen-dependent prostate cancer cell line, to oxidative stress to generate H2O2-resistant sublines, and found that they expressed increased levels of AR mRNA and protein and exhibited a castration-resistant phenotype [8]. Whereas castration-resistant cells normally exhibit elevated antioxidant protein levels [70, 71] and ROS-scavenging activity [72], overexpression of AR in such cells increases oxidative stress, as indicated by higher intracellular ROS levels [68, 69].

A connection between oxidative stress and CRPC is also supported by clinical findings. Compared with prostate specimens from patients who had undergone radical prostatectomy without ADT, prostate cancer tissues obtained from patients post-ADT show increased 4-hydroxy-2-nonenal levels, indicative of elevated oxidative stress [17]. In addition, a genetic polymorphism in the GSTM3 gene, which encodes an antioxidant enzyme glutathione S-transferase, was recently reported to be associated with increased risk of progression of metastatic prostate cancer to CRPC, which was validated in nonmetastatic prostate cancer [69].

Collectively, these experimental and clinical data are consistent with a close link between oxidative stress and progression to CRPC.

21.5 Clinical Implications of Antioxidant Therapy in CRPC

Given the accumulating evidence that oxidative stress contributes to CRPC, it has been speculated that antioxidant therapy could have therapeutic effects in prostate cancer patients receiving ADT.

Various naturally occurring antioxidative compounds, including isoflavones, catechins, carotenes, vitamins, and selenium, have been investigated as possible prophylactic agents for prostate carcinogenesis and as therapeutic agents for prostate cancer [73, 74]. Among these compounds, the carotenoid lycopene was shown to prevent oxidative damage to proteins, lipids, and DNA. In a preclinical study, lycopene suppressed AR activity and had antitumor effects [75]. In clinical studies, lycopene augmented the therapeutic effects of orchiectomy in advanced prostate cancer patients [76]. A phase II study showed that administration of lycopene at 10 mg per day suppressed elevation of prostate-specific antigen (PSA) in 41 men with prostate cancer [75]. In addition, a case report of a CRPC patient described a reduction in serum PSA levels and disease-associated symptoms after intake of saw palmetto and lycopene supplements [77]. Although the number of patients in these clinical studies was small, the findings support the potential use of lycopene combined with castration in the treatment of prostate cancer, including CRPC. The antioxidants vitamin E and α-tocopherol have also been reported to decrease the risk of prostate cancer mortality, suggesting that they may prevent disease progression [78].

In addition to naturally occurring compounds, synthetic antioxidants might also be useful for the treatment of prostate cancer. In the TRAMP mouse model of prostate cancer, NAC administration reduced 8-hydroxy-2′-deoxyguanosine, nitrotyrosine, and 4-hydroxy-2-nonenal levels in the prostate [79]. In addition, we previously showed that NAC reduced AR expression and that NAC plus ADT successfully suppressed tumor growth in a mouse xenograft model of prostate cancer [17]. SOD mimetics have been shown to reduce oxidative stress, reduce the expression of AR and AR splice variants, and have a therapeutic effect in prostate cancer cells [80]. Finally, the anti-angiogenic agent endostatin inhibits CRPC growth by augmenting antioxidant enzyme activity and suppressing ROS levels [81].

An alternative therapeutic strategy to counter oxidative stress in CRPC is inhibition of ROS production. In support of this, the Nox inhibitor diphenyleneiodonium decreases the viability of prostate cancer cells, including LNCaP cells [82] and another Nox inhibitor, apocynin, suppresses prostate cancer cell invasion [83].

These observations highlight several options for antioxidant therapy, including natural and synthetic antioxidants and ROS inhibitors. However, a critical obstacle for the clinical use of antioxidants is their rapid oxidative degradation under physiological conditions, resulting in poor stability and bioavailability. One potential solution to this problem might be to encapsulate the antioxidant compounds in nanoparticles that also act as oxygen radical scavengers. For example, it was recently reported that curcumin-loaded pH-sensitive redox nanoparticles exert excellent antitumor activity in prostate cancer [84].

21.6 Novel Agents for CRPC and Oxidative Stress

Taxanes such as docetaxel and cabazitaxel, AR-targeting agents such as abiraterone acetate and enzalutamide, and the radiopharmaceutical radium-223 all show benefit in prolonging progression-free and overall survival and have been approved globally for use in CRPC [37, 85].

Similar to other cytotoxic anticancer agents, taxanes have been shown to cause oxidative stress in cancer cells [86]. Moreover, many molecules implicated in oxidative stress signaling, including PI3K/Akt [87], MAPK [88], and NF-κB [89], and their downstream effectors such as Twist1 [90] and YB-1 [91, 92], are all involved in the resistance of prostate cancer to taxanes. In addition, the status of TMPRSS2-ERG fusion gene caused by inflammation-induced oxidative stress through DNA breaks [93] was reported to be associated with the therapeutic effect of taxanes [94, 95]. Thus, oxidative stress appears to contribute to taxane resistance in prostate cancer through various mechanisms.

Radiation is known to induce oxidative stress in prostate cancer [96]; however, this is not necessarily beneficial because irradiation-induced oxidative stress can activate pro-survival and anti-apoptotic signaling through molecules such as PI3K/Akt [97], MAPK [98], and NF-κB [99], resulting in resistance to irradiation. Radium-223 is an α particle-emitting isotope [100] and appears to induce oxidative stress in prostate cancer cells. Although the therapeutic effect of this isotope may be affected by oxidative stress-induced signaling, there is currently no direct evidence for this.

Little is known about the interaction between oxidative stress and AR-targeting agents, including abiraterone acetate and enzalutamide. However, oxidative stress levels are increased in enzalutamide-resistant prostate cancer cells established in vitro [69], warranting further investigation. Clinical trials have been initiated for several additional promising agents, including immune checkpoint inhibitors and the poly (ADP-ribose) polymerase inhibitor olaparib. To those agents, biomarkers such as the presence of somatic mutations in DNA repair genes and the number of missense somatic mutations which may be caused by oxidative stress are postulated, and then oxidative stress may commit to the sensitivity to those emerging agents.

21.7 Conclusions and Future Directions

Oxidative stress induced by ADT can activate both AR and non-AR signaling, resulting in the acquisition of castration resistance. Treatment-induced oxidative stress also appears to be involved in the resistance of prostate cancer to therapy. Thus, oxidative stress is a critical resistance factor and a crucial target for prostate cancer treatment. Suppression of oxidative stress signaling by antioxidants or inhibitors of ROS production may thus be a promising strategy to overcome treatment resistance in prostate cancer. However, the relationship between oxidative stress and CRPC is a vast and underexplored area of research, and further investigation is warranted. Such studies will undoubtedly lead to some remarkable discoveries.