**4. Looking forward: Considerations for further research in n-3 PUFA and prostate cancer**

#### **4.1. Prostate cancer signaling pathways and n-3 PUFA**

**3. Promising anti-inflammatory natural products in PCa**

222 Using Old Solutions to New Problems - Natural Drug Discovery in the 21st Century

however this effect seems to be linked to epigenetic modifications of DNA [77].

**3.1. PUFA and prostate cancer**

The role of inflammation in prostate cancer etiology stems from studies accessing the rela‐ tionship between intake of anti-inflammatory dietary compounds and prostate cancer risk. Epidemiological evidence associates or significantly correlates consumption of tomato [73], soy and green tea [74, 75] with decreased prostate cancer risk. Furthermore, animal studies confirm that the anti-inflammatory properties of both soy and green tea cause a decrease in prostate cancer [74, 75]. One study involving Transgenic Adenocarcinoma of the Mouse Prostate (TRAMP) mice fed a diet enriched with processed whole tomatoes reported benefits including increased survival, delayed progression from PIN to PCa cancer and a decreased incidence of poorly differentiated PCa cancer [76]. Prostate cancer cell lines treated with phytoestrogens, specifically genistein and daidzein, indicate a decrease in prostate cancer risk,

Interest in dietary fats and disease etiologies has emerged because of their known antiinflammatory properties. Examples of dietary fat which are essential in a variety of mammalian biological processes and impact PCa include omega-3 (n-3) and omega-6 (n-6) PUFA [78, 79]. In published cell line and xenograft studies, n-6 PUFA (linoleic acid and arachidonic acid) typically exert a growth-promoting effect, while n-3 PUFA (EPA and DHA) have growthinhibitory effects [80-85]. *In vitro* and animal studies suggest a trend of opposite effects for cancer development with respect to n-3 and n-6 PUFA. The n-3 PUFA, such as EPA and DHA, indicate a suppression of tumor carcinogenesis, however n-6 PUFA seems to promote tumor development. From an epidemiological perspective, evidence of an association between particular PUFA and PCa is inconsistent, with many studies reporting no association between dietary intake of n-3 or n-6 PUFA and the risk of PCa [86-92], however both pre-clinical *in vivo* and *in vitro* studies clearly indicate that there are biological mechanisms by which omega-3 PUFA can arrest growth in both PCa cells and tumors. Supplementation of n-3 PUFA in animal studies failed to produce an effect on prostate tumor growth or other markers of PCa progres‐ sion and did not significantly reduce tumor growth in a PCa mouse model [93-96]. Treatment of human PCa cells with n-3 PUFA has been shown to consistently inhibit proliferation and /or increase programmed cell death, affect gene expression and deter properties of invasive human PCa in cells supporting the notion that EPA and DHA supplementation could preclude or limit the growth of prostate tumors [80, 85, 94, 97-102]. In general, and as found in a recent case-control study, while proinflammatory n-6 PUFA may present an increase in PCa risk, the anti-inflammatory properties of n-3 PUFA have been noted in their association with decreased risk [103]. Interestingly, this same study reported that mutations in the inflammation and mitogenesis related gene (COX-2), combined with low nutritional consumption of n-3 PUFA, had a higher risk of PCa; this risk was lowered with an increase in the dietary intake of n-3 PUFA [103]. Inflammatory gene expression is usually negatively associated with cancer stage and prognosis [104]. Despite these trends, a recent epidemiology study by Brasky (2011) contradicts the protective role of n-3 PUFA in PCa by reporting a positive association with

Signaling pathways and their associated molecules often have a dual role in events such as homeostasis, tissue repair, and tumorigenesis. For example, the Wnt/β-catenin pathway is critical in maintaining steady-state proliferation and tumorigenesis of tissues [106]. Numerous studies to date indicate that n-3 PUFA treatment can affect cell signaling partially through AKT (protein kinase B), mTOR (mammalian target of rapamycin) and NF-κB (nuclear factor kappa B) associated pathways [94, 95, 99, 107]. Treatment with n-3 PUFA has been indicated in the inhibition of human hepatocellular carcinoma and cholangiocarcinoma cell growth by blocking the Wnt/β-catenin pathway [108, 109], and DHA treatment has been shown to inhibit the production of β-catenin in colon cancer cells [110-112]. However, the effects of n-3 PUFA on Wnt/β-catenin signaling in PCa remain largely unknown. A single study in PCa indicates that fat-1 gene (cloned from *Caenorhabditis elegans*) that encodes for a n-3 PUFA desaturase (which converts n-6 PUFA to n-3 PUFA) expression was shown to reduce phosphorylation of glycogen synthase kinase-3β (GSK-3β), a major element in the Wnt/β-catenin pathway, resulting in subsequent down-regulation of both β-catenin and cyclin D1 thereby inhibiting PCa cell proliferation [113].

#### *4.1.1. An introduction to Wnt/β-catenin signaling*

The Wnt family constitutes 19 highly conserved glycoprotein members in mammals [114]. The most significant molecule implicated in the canonical Wnt cascade is β-catenin, a cytoplasmic protein regulated by a multi-protein destruction complex made up of Axin, adenomatous polyposis coli (APC), GSK-3β and casein kinase 1 (CK1) [115]. In the absence of Wnt signaling, the destruction complex stimulates the phosphorylation of β-catenin by GSK-3β, leading to subsequent ubiquitination and proteasomal degradation [116, 117]. Conversely, in the presence of Wnt signaling, Wnt ligand-frizzled (FZD) binding causes disheveled (DVL) protein dissociation of the β-catenin destruction complex, blocking the phosphorylation of β- catenin, and leading to β-catenin accumulation in the nucleus [116, 117]. Nuclear β-catenin functions as a transcription co-factor of the Tcf/Lef family and leads to the activation of Wnt target genes implicated in cell proliferation, differentiation and apoptosis, including c-myc, cyclin D1, Akt, MMP-7, and AR [114, 116-122].

#### *4.1.2. Using Wnt/β-catenin signaling as a model to determine the pathogenic role of n-3 PUFA in prostate cancer*

Modular in nature, the activity of the Wnt/β-catenin pathway can be modified through several points of intervention. The fundamental event in Wnt/β-catenin signaling occurs in the nucleus and is the result of stabilized β-catenin recruiting Tcf/Lef transcription factors that modulate the expression of oncogenes, such as c-myc and cyclin D1[123-128]. Abnormal expression of β-catenin has been observed in up to 71% of prostate tumor specimens [129-131], is elevated in more than 20% of advanced prostate tumors [132] and is associated with advanced stage PCa [130, 133, 134]. APC alterations alone are considered prognostic with respect to an unfavorable outcome [135] even though it varies with respect to inactivation in PCa reporting somatic loss ranging from 2– 43% [132, 136-138] and promoter hypermethylation in up to 90% of PCa [135, 139, 140]. Mutations in Axin-1 have been identified in 14% of advanced PCa, and several Axin-1 mutations and polymorphisms have been noted in PCa cell lines [138]. DVL-1, is a pathway regulator involved in Axin recruitment and inactivation, is significantly overex‐ pressed in PCa and has also shown a positive correlation with PCa grade [141]. Pathway activators, including WNT-1, -2, -5A, and -6 have been highly overexpressed in primary PCa compared to normal prostate [142-145], and WNT-1 and -2 have been indicated as having a role in invasive PCa [143, 146]. Conversely, but also resulting in stabilized β-catenin, pathway inhibitors are commonly downregulated in PCa. Dickkopf-related protein (DKK)-1 expression is lower in PCa tissue in comparison to normal prostate tissue samples and furthermore is significantly reduced during progression to metastasis [143, 147]. Secreted frizzled-related protein (SFRP)- 1 and DKK-3 show decreased expression in prostate tumor cells [148-151] Downregulation of SFRP-1 was also reported in PCa cell lines [152]. It is interesting to note that SFRP-4 overexpression is associated with decreased proliferation, decreased anchorageindependent growth and decreased invasiveness in a PCa cell line, and additionally predicts a good prognosis in PCa [153, 154]. WIF-1, another Wnt antagonist is downregulated at the mRNA (23%) and protein level (64%) of PCa [155]. Recently, FZD-3 inhibition by soy protein was shown to suppress the growth of PCa cell lines [156]. Similarly, FZD-5 was highly upregulated in the prostate tissue of advanced PCa cancer patients and was downregulated by zoledronic acid in PC3 cells, but not DU145 cells [157]. Appropriate steps to determine the role of n-3 PUFA in prostate cancer may include investigating whether n-3 PUFA treatment will affect subcellular gene expression and localization of β-catenin. Beyond confirmation of the molecular mechanism(s) by which n-3 PUFA inhibits Wnt/β-catenin, subsequent experiments could target degradation of cytoplasmic β-catenin and activation of the Wnt receptor complex. Table 1 summarizes dysregulation of numerous pathway components which may be respon‐ sible for PCa tumor development/progression and as such represents potential targets for further investigation to determine n-3 PUFA/Wnt/β-catenin interactions in PCa.

that fat-1 gene (cloned from *Caenorhabditis elegans*) that encodes for a n-3 PUFA desaturase (which converts n-6 PUFA to n-3 PUFA) expression was shown to reduce phosphorylation of glycogen synthase kinase-3β (GSK-3β), a major element in the Wnt/β-catenin pathway, resulting in subsequent down-regulation of both β-catenin and cyclin D1 thereby inhibiting

The Wnt family constitutes 19 highly conserved glycoprotein members in mammals [114]. The most significant molecule implicated in the canonical Wnt cascade is β-catenin, a cytoplasmic protein regulated by a multi-protein destruction complex made up of Axin, adenomatous polyposis coli (APC), GSK-3β and casein kinase 1 (CK1) [115]. In the absence of Wnt signaling, the destruction complex stimulates the phosphorylation of β-catenin by GSK-3β, leading to subsequent ubiquitination and proteasomal degradation [116, 117]. Conversely, in the presence of Wnt signaling, Wnt ligand-frizzled (FZD) binding causes disheveled (DVL) protein dissociation of the β-catenin destruction complex, blocking the phosphorylation of β- catenin, and leading to β-catenin accumulation in the nucleus [116, 117]. Nuclear β-catenin functions as a transcription co-factor of the Tcf/Lef family and leads to the activation of Wnt target genes implicated in cell proliferation, differentiation and apoptosis, including c-myc, cyclin D1, Akt,

*4.1.2. Using Wnt/β-catenin signaling as a model to determine the pathogenic role of n-3 PUFA in*

Modular in nature, the activity of the Wnt/β-catenin pathway can be modified through several points of intervention. The fundamental event in Wnt/β-catenin signaling occurs in the nucleus and is the result of stabilized β-catenin recruiting Tcf/Lef transcription factors that modulate the expression of oncogenes, such as c-myc and cyclin D1[123-128]. Abnormal expression of β-catenin has been observed in up to 71% of prostate tumor specimens [129-131], is elevated in more than 20% of advanced prostate tumors [132] and is associated with advanced stage PCa [130, 133, 134]. APC alterations alone are considered prognostic with respect to an unfavorable outcome [135] even though it varies with respect to inactivation in PCa reporting somatic loss ranging from 2– 43% [132, 136-138] and promoter hypermethylation in up to 90% of PCa [135, 139, 140]. Mutations in Axin-1 have been identified in 14% of advanced PCa, and several Axin-1 mutations and polymorphisms have been noted in PCa cell lines [138]. DVL-1, is a pathway regulator involved in Axin recruitment and inactivation, is significantly overex‐ pressed in PCa and has also shown a positive correlation with PCa grade [141]. Pathway activators, including WNT-1, -2, -5A, and -6 have been highly overexpressed in primary PCa compared to normal prostate [142-145], and WNT-1 and -2 have been indicated as having a role in invasive PCa [143, 146]. Conversely, but also resulting in stabilized β-catenin, pathway inhibitors are commonly downregulated in PCa. Dickkopf-related protein (DKK)-1 expression is lower in PCa tissue in comparison to normal prostate tissue samples and furthermore is significantly reduced during progression to metastasis [143, 147]. Secreted frizzled-related protein (SFRP)- 1 and DKK-3 show decreased expression in prostate tumor cells [148-151]

PCa cell proliferation [113].

MMP-7, and AR [114, 116-122].

*prostate cancer*

*4.1.1. An introduction to Wnt/β-catenin signaling*

224 Using Old Solutions to New Problems - Natural Drug Discovery in the 21st Century

#### *4.1.3. Clinical significance of targeting the Wnt/β-catenin pathway in PCa with n-3 PUFA*

Despite the noted progress of hormone-based drugs as a therapy, PCa remains as one of the primary causes of cancer deaths worldwide [20, 101]. Thus, new and improved therapeutic strategies to prevent PCa and inhibit its progression are needed. In this regard, n-3 PUFA presents tremendous opportunity as a therapeutic intervention in PCa. There is enough convincing data available to show a positive correlation between Wnt/β-catenin activation and PCa progression. It is also well established now that n-3 FAs supplementation can impact PCa development and progression *in vitro.* However, whether these effects of n-3 PUFA on PCa can occur at least in part through inhibition of Wnt/β−catenin signaling remain largely unknown. Thus, the incorporation of variations as suggested above could start by targeting nuclear interaction and progressing back through the pathway to the point of surface expres‐ sion interactions of the Wnt-FRZ-LRP Receptor Complex. Furthermore, AR expression is positively correlated with an increase in cytoplasmic/nuclear β-catenin levels in epithelial cells of the prostate [175]. β-catenin interacts with AR and acts as a co-activator of AR to increase its transcriptional activity in response to androgen [17]. This leads to activation of AR target genes implicated in PCa progression. Therefore it is possible that n-3 PUFA can inhibit aberrant expression and activation of AR. As β-catenin is considered to be a ligand-dependent coacti‐ vator of AR transcription [17], the effects of n-3 PUFA on AR expression and β-catenin/ AR crosstalk is an area requiring exploration.

A series of experiments to specifically target different components of the Wnt/β-catenin pathway as noted above would be a rational approach to determining the effects of n-3 PUFA in PCa. Goss's recent textbook reviews some innovative strategies that have been utilized to antagonize signaling at various levels of the Wnt/β-catenin signaling [176]. These range from blocking β-catenin-mediated transcription to modulating the catenin destruction complex to attenuating extracellular signaling through the Wnt receptors [176]. Figure 2 illustrates the Wnt/β-catenin pathway and could serve as a model to incorporate multiple targets of interest


**Table 1.** Components of the Wnt/β−catenin Pathway Implicated in PCa: Alterations Of Primary PCa Tissues and PCa Cell Lines Compared to Normal Prostate.

to be explored in n-3 PUFA and PCa. Therefore, research focusing on targeting components of the Wnt/β-catenin pathway or other cellular pathways as a means of inhibiting PCa development and progression with a promising prospect, n-3 PUFA seems reasonable.

**Pathway Component Alteration(s) References**

Wnt-2 Increased Expression [143, 146, 158] Wnt-5A Increased Expression [131, 143, 158]

Wnt-1 Increased Expression [142]

226 Using Old Solutions to New Problems - Natural Drug Discovery in the 21st Century

Wnt-6 Increased Expression [144]

WIF-1 Decreased Expression [155]

FZD-3 Increased Expression [156] FZD-5 Increased Expression [159]

DVL-1 Increased Expression [141]

β-catenin Nucleus/cytoplasm accumulation [129-132, 134, 142,

APC Decreased or Loss of Expression [132, 134-137, 139,

GSK-3β Increased activity [165-170]

Tcf-4 Increased Expression [142, 171]

CyclinD1 Upregulation [123, 124]

c-myc Upregulation [125-128, 172-174]

**Table 1.** Components of the Wnt/β−catenin Pathway Implicated in PCa: Alterations Of Primary PCa Tissues and PCa

Axin-1 Increased Expression [138]

160-163]

140, 164, 165]

DKK-3 Decreased Expression [150-152] DKK-1 Decreased Expression [143, 147]

SFRP-1 Decreased Expression [148, 149] SFRP-4 Decreased Expression [153, 154]

**A. Wnts**

**B. Secreted Wnt Antagonists**

**C. Wnt Receptors/coreceptors**

**E. Destruction Complex Components**

**F. Tcf/Lef Transcription Factor**

Cell Lines Compared to Normal Prostate.

**F. Genes**

**D. Pathway Regulators**

**Figure 2.** Tobin, GA. 2012. Adapted from articles under the Creative Commons Attribution License Copyright © 2011 Chi-Tai Yeh *et al*. [177] and from Lattouf *et al.* (Reprinted by permission from Macmillan Publishers Ltd: Nature Re‐ views Urology [178, 179]. Copyright © 1969). A hypothetical diagram of n-3 PUFA and n-3 PUFA/ androgen receptor induced inhibition of Wnt/β-catenin signaling pathway in PCa (directly and indirectly). When the Wnt signal is absent, a multi-complex destruction unit consisting of CK1/2, GSK3β, APC, and Axin forms the target for ubiquitination and degradation, namely a hyper-phosphorylated β-catenin. In the presence of Wnt, Fzd receptors recruit DVL to the plas‐ ma membrane to inactivate Axin. Overall, binding of Wnt ligand to a Frizzled/LRP receptor complex leads to stabiliza‐ tion of β-catenin. Subsequent interaction with nuclear Tcf/Lef proteins, increases the expression of genes such as cyclin D1 and c-myc. Additionally, as noted above the effects of n-3 PUFA on AR expression and β-catenin/AR crosstalk should be considered.
