Polymorphisms of pro-inflammatory genes and prostate cancer risk: a pharmacogenomic approach

Abstract

In this paper, we consider the role of the genetics of inflammation in the pathophysiology of prostate cancer (PCa). This paper is not an extensive review of the literature, rather it is an expert opinion based on data from authors’ laboratories on age-related diseases and inflammation. The aim is the detection of a risk profile that potentially allows both the early identification of individuals at risk for disease and the possible discovery of potential targets for medication. In fact, a major goal of clinical research is to improve early detection of age-related diseases, cancer included, by developing tools to move diagnosis backward in disease temporal course, i.e., before the clinical manifestation of the malady, where treatment might play a decisive role in preventing or significantly retarding the manifestation of the disease. The better understanding of the function and the regulation of inflammatory pathway in PCa may help to know the mechanisms of its formation and progression, as well as to identify new targets for the refinement of new treatment such as the pharmacogenomics approach.

Cancer, aging and inflammation

Cancer is a hyperproliferative disorder that involves cellular transformation, dysregulation of apoptosis, uncontrolled cellular proliferation, angiogenesis, invasion, and metastasis. It is generally recognized as an age-related disease; the median age for cancer diagnosis in industrialised countries is nearly 70 years of age and is expected to increase [1]. This may be due to different reasons, such as an increased duration of carcinogenesis or the susceptibility of aging cells to environmental carcinogens [2, 3]. However, incidence and mortality rates of most human cancers consistently increase up to age 90, but they plateau and decline thereafter, suggesting that centenarians are protected from developing cancer. A low-grade systemic inflammation characterizes aging and this pro-inflammatory status may underlie biological mechanisms responsible for age-related inflammatory diseases, including cancer [4]. Another mechanism responsible might be the reduced immune function, i.e., immunosenescence in the elderly [5]. The importance of the immune system in preventing tumor formation, immunosurveillance, has been repeatedly shown in animal models and is supported by epidemiological evidence, such as increased frequency of certain cancer types in immunosuppressed individuals [6, 7].

Clinical and epidemiologic studies have suggested a strong association between chronic infection, inflammation, and cancer: it is known that alcohol abuse can lead to liver and pancretic cancer, smoking to airway inflammation and lung carcinoma, and infectious disease such as hepatitis C virus or papilloma (HPV) infections to liver, cervical, and prostate cancer [4, 8–10]. In addition, many other inflammatory triggers are associated with other malignancies, including microbial infection, such as Helicobacter pylori infection, associated with gastric cancer [11] or autoimmune responses associated with inflammatory bowel disease and colorectal cancer [12].

Epidemiological studies not only show that chronic inflammation predisposes individuals to certain cancers but also that nonsteroidal anti-inflammatory drugs (NSAIDs) seem to prevent several cancers [13]. Accordingly, most pre-neoplastic and neoplastic tissues show signs of inflammation as demonstrated by the presence of immune cells into neoplastic tissue together with inflammatory signalling molecules, such as cytokines and chemokines, which are responsible for changes in tissue structure, i.e., remodelling, and the formation of new blood vessels [4].

Inflammation seems to play a role in carcinogenesis by causing cellular and genomic damage, promoting cellular turnover, and creating a tissue microenvironment that can enhance cell replication, angiogenesis and tissue repair [14]. Actually, tumors are thought to develop from chronic inflammation, where uncontrolled cell proliferation occurs in a milieu rich in pro-inflammatory cytokines, inflammatory mediators, and growth factors normally involved in chronic and unresolved inflammation [4, 15]. In normal tissues, anti-inflammatory cytokines are synchronically upregulated after the pro-inflammatory cytokines are produced, leading to inflammation resolution. In chronic inflammation, this does not occur, resulting in a continuous production of reactive oxygen species, leading to oxidative DNA damage and reduced DNA repair, and increased proliferation [4, 15].

Cancer-related inflammation has been recently described as extrinsic, driven by inflammatory conditions that increase the risk of cancer, or intrinsic, due, for example, to the activation of oncogenes that induce a transcriptional pattern similar to that which occurs during inflammation [16]. Accordingly, it has been demonstrated that an early genetic event that is necessary and sufficient for the development of a human tumor directly promotes the buildup of an inflammatory microenvironment, although the role of the various components of inflammatory microenvironment in tumor progression need to be specified [17].

A further line of evidence involving inflammation in the pathogenesis of cancer is represented by the relationship between sex hormones and inflammation, since it has been shown that the capability of sex hormones to cause tumor promotion in breast and prostatic cancer could be related to inflammation (see below).

It is known that there are two main prostate pathologies affecting men during their lifetime: benign prostatic hyperthophy (BHP) and prostate cancer (PCa) [18]. PCa is currently the commonest non-skin neoplasm and the second leading cause of cancer death in men in the USA, with 218,890 new cases and 27,050 deaths from this disease in the year 2007 [19]. In Europe, human PCa has exhibited a steady increase of incidence in recent years, although there is now some reduction in the mortality rates [20].

Today, it is widely accepted that inflammation has a role in PCa [21–24], and a list of genes discussed in this study is listed in Table 1. The “risk factor” lesion in PCa, the regenerative epithelium in response to environmental insults, often precedes the development of prostatic intraepithelial neoplasia (PIN) and early carcinoma. Proliferative inflammatory atrophy (PIA), the proliferative glandular epithelium with the morphological appearance of simple atrophy, occurs in association with inflammation and is thought to be a possible precursor of PCa too [25–27]. PIA lesions seem to arise as a consequence of regenerative proliferation of prostate epithelial cells in response to inflammatory injury. PIA has been proposed to be considered a precursor of high-grade PIN and PCa. Furthermore, chronic and/or acute glandular inflammation is indeed observed in many radical prostatectomy specimens. High-grade PIN is often observed in proximity to PIA, and morphologic transitions between high-grade PIN and PIA frequently occur within the same acinus/duct [28–34].

Table 1.

Inflammatory molecules discussed in this paper

Genes	Biological effect
CCL5	A chemotactic for T cells, eosinophils, and basophils that plays an active role in recruiting leukocytes into inflammatory sites
CCR5	A β-chemokine receptor involved in the migration of monocytes, NK cells, and some T cells to the inflammation site
COXs	Enzymes involved in the conversion of arachidonic acid to inflammatory mediators, prostaglandins. The isoform COX-2 is upregulated in a variety of malignancies, including prostate cancer, throughout the tumorigenic process
LO	Enzymes involved in the conversion of arachidonic acid to inflammatory mediators, leukotrienes. LTs exert profound biological effects on the development and progression of human cancers
IL-1	IL-1α and IL-1β might be considered the prototypic pro-inflammatory cytokines of the IL-1 family
IL-1Ra	Competes for receptor binding with IL-1α and IL-1β, blocking their role in immune activation
IL-8	Regulates angiogenesis and tumor growth
IL-10	Anti-inflammatory function, with pleiotropic effects in immunoregulation and inflammation. It downregulates the expression of type-1 cytokines, MHC class II antigens, and costimulatory molecules on macrophages
IL-18	Produced by macrophages and other cells, belongs to the IL-1 superfamily. It is a multifunctional cytokine that induces interferon-gamma secretion and plays an important role in antitumor immunity mediated by type-1 positive regulation loop
TGF-β1	Polypeptide member of the transforming growth factor beta superfamily of cytokines. It is a multi-functional modulator of cellular proliferation, differentiation, and production and degradation of extracellular matrix
TIRAP	Intracitoplasmatic protein that mediates downstream signalling of TLR2 and TLR4
TLRs	Receptors able to detect microbial conserved components and trigger protective host responses

For the references, see text

Sexually transmitted diseases (STDs) are hypothesized to play a role in the development of PCa, perhaps due to inflammation-induced oncogenesis. Epidemiological studies reported an increased risk for PCa among men with a history of clinical or symptomatic prostatitis [21, 31, 32, 34]. Frequent sexual encounters with prostitutes and unprotected sexual intercourses were shown to be associated with increased risk of STDs, such as syphilis and gonorrhoea, also associated with an increased risk for PCa [35, 36]. A meta-analysis showed an increased PCa risk directly associated with the number of sexual partners and history of sexually transmitted infection [37]. Another meta-analysis [38] demonstrated significantly increased risk for PCa in subjects affected by gonorrhea and HPV infection. Other indicators of promiscuous sexual behavior and of exposure to STDs, including number of sexual partners, age at first intercourse, and frequency of sex, have each been reported to be associated with prostate cancer risk in previous studies [37, 39].

Estrogens, aromatase, eicosanoids and inflammation

Today, evidence is accumulating in support of an important role for estrogens in human PCa [40], although mechanisms underpinning estrogen implication in prostate malignancies still remain largely unspecified. Clear-cut evidence from both experimental animal models and in vitro studies suggested that androgens and estrogens may be important in PCa. Long-term administration of testosterone and estradiol produces a high incidence of rat prostate adenocarcinomas [41].

As previously discussed, chronic inflammation has been associated with development and/or progression of many human cancers. Estrogens are commonly regarded as pro-inflammatory hormones and have also been implicated in explaining the large prevalence of autoimmune diseases in women [42]. In the Noble rats, it has been shown that elevated estrogen levels in the presence of testosterone cause an early (4 weeks) prostate-specific inflammatory response and a later development of prostate carcinomas (after nearly 50 weeks) [43]. It is likely that estrogen-induced early inflammatory events are a prerequisite for the onset of PCa [44].

Chronic exposure of Wistar rats to estradiol and dihydrotestosterone propionate determined an early upregulation of interleukin(IL)-1β, IL-6, and inducible nitric oxide synthase, later accompanied by an increase of IL-4 and IL-5 expression; this pattern of upregulated pro-and ant-inflammatory genes occurred irrespective of the presence of inflammatory cells and resembled a type-2 helper response [45].

In the hypogonadal mice, an animal model system completely deficient in pituitary gonadotropins and, hence, in sex steroids, exposure to estrogen resulted in an inflammatory and proliferative response involving both prostatic epithelium and stroma [46]. Furthermore, estrogens neonatally administered to rodents induce an imprinted state, called developmental estrogenization, that results in the development of lobe-specific inflammation, hyperplasia and dysplasia, and/or PIN [47].

In steroid target tissues, including prostate, the ultimate effects of steroid hormones depends on the uptake from circulation and local metabolism. A divergent expression and/or activity of key steroid enzymes (including dehydrogenases, hydroxylases, sulfotransferases, sulfatases, and aromatase) may lead to a differential accumulation of steroid derivatives with distinct biological activity. However, studies assessing sex steroid metabolism in vivo in either normal, hyperplastic, or malignant human prostate are surprisingly rare [40].

There is consistent evidence that aromatase is expressed in hyperplastic or malignant human prostate. In this respect, the aromatase enzyme may provide an important source of local estrogen production through an alternative pathway, withdrawing androgen from metabolic conversion to either 5α-reduced bioactive metabolites, such as dihydrotestosterone, or 5β products of degradation [48].

Aberrant expression of aromatase is believed to contribute to the development and progression of breast cancer [49]. Unfortunately, little is known about the potential role of aromatase in prostate carcinogenesis and/or tumor progression. The aromatase overexpressing mouse mantains a locally elevated estrogen production and an extensive prostate inflammation later in life [50]. On the other hand, the aromatase knockout (ArKO) mouse, which is deficient in estrogens because of a non-functional aromatase enzyme, develops prostatic hyperplasia after a lifelong exposure to elevated androgens, but no malignant changes could be detected in the prostate at any time. However, transient neonatal treatment of the ArKO mouse with the synthetic estrogen diethylstilbestrol results in both prostate epithelial dysplasia and inflammation in adult life [51].

In view of the key role of aromatase in the interplay between inflammation and estrogen implicated in the pathophysiology of PCa, different polymorphisms of aromatase gene have been recently analyzed to evaluate their capacity in modifying individual PCa susceptibility. Literature, based on population case–control studies, proposed some polymorphisms as possible markers of susceptibility for PCa [52–61]. In addition, it has been suggested that also PCa prognosis seems to be influenced by some genetic variants of aromatase gene, although conflicting results have been obtained [55, 62].

Eicosanoids, such as prostaglandins (PGs) and leukotrienes (LTs), have been implicated in the pathogenesis of a variety of human diseases [63, 64], including cancer, and are now believed to play a role in tumor promotion, progression, and metastatic disease (see below). The enzymes involved in the conversion of arachidonic acid to PGs and LTs are Cyclooxygenases (COXs) and Lipoxygenase (LO). The two isoforms of COXs, COX-1 and COX-2, are almost identical in structure, but have important differences in substrate and inhibitor selectivity and in their intracellular locations The constitutive COX-1 is present in many tissues and synthesizes PGs involved in maintaining normal tissue homeostasis. The second isoform, COX-2, is responsible for PGs produced in sites of inflammation and is induced by growth factors, cytokines, and various carcinogens [65–67].

COX-2 is upregulated in a variety of malignancies including prostate cancer, throughout the tumorigenic process from early hyperplasia to metastatic disease [9, 68–70]. In particular, its overexpression has been described in cases with evolution of PIA and PIN [71]. In vitro COX-2 overexpression inhibits apoptosis and induces tumor angiogenesis [72]. Some studies have a shown that prolonged aspirin ingestion reduces the incidence of PCa. This effect might result, at least in part, from COX-2 inhibition [73, 74]. However, it is possible that the anti-cancer activity of these compounds might also reflect COX-independent effects, since high concentrations of NSAIDs suppress the growth of cells in culture that do not express COX-2 [75].

The exposure of cultured human breast fibroblasts to PGE2 and proinflammatory cytokines, such as IL-6 or tumor necrosis factor (TNF)-α, produces a marked increase of aromatase activity [76, 77]. In addition, previous experimental evidence obtained in breast cancer patients indicates a direct association of COX-1 and COX-2 with aromatase expression [78]. On the other hand, in estrogen receptor-positive human breast cancer cells, estradiol, combined with TNF-α or IL1-β, rapidly and robustly upregulates PGE synthase (PTGES), a key enzyme in the production of PGE2. This synergistic upregulation of PTGES may lead to an increase in the formation of PGE2 that may in turn enhance further aromatase expression and local estrogen production [79]. This may represent a vicious circle by which aromatase and PGE2 reciprocally induce their own expression and/or activity and sustain local inflammation, ultimately favoring tumor development.

Single nucleotide polymorphisms (SNPs) in the COX-2 gene may influence its function and/or its expression and may also modify the protective effect of NSAIDs [80], affecting an individual’s risk of developing cancer. Few studies have investigated the association between COX-2 polymorphisms and cancer. The association between haplotypes in the COX-2 promoter region and risk of PCa has been studied in four different ethnic groups; great population diversity of the haplotype frequency and divergent risk associations across different haplotypes have been found [81]. Another case–control study examining the role of five SNPs in a large Swedish cohort showed that only two SNPs were significantly associated with the risk of PCa [82]. In addition, the same group, 1 year later, demonstrated that frequent consumption of fatty fish appears to reduce the risk for PCa, and this association is modified by genetic variation in the COX-2 gene [83].

Mounting evidence suggests that LO-catalyzed products, LTs, and hydroxyeicosatetraenoic acids (HETEs) also exert profound biological effects on the development and progression of human cancers. Significant increase of LO metabolites in patients with lung, breast, colon, and skin carcinoma has been observed and increased 12-LO mRNA and 12(S)-HETE levels have been positively correlated with the metastatic potential of colon and prostate carcinoma [84, 85]. Increased levels of 5-HETE were shown to be able to stimulate lung and prostate tumor cell growth, while depletion in their levels has been revealed to result in massive apoptosis in these cells [85, 86]. It was also shown that 5-LO was overexpressed in patients with PCa, suggesting that selective 5-LO inhibitors may be useful for prevention or therapy in patients with PCa [87]. Newly developed selective LO inhibitors have been shown to inhibit the growth of various carcinoma cells and afford protection against carcinogen-induced lung adenomas and rat mammary gland tumors [85].

We have examined the role of Cox-2 −765 G/C and of 5-LO −1708 G/A SNPs in PCa, in the Sicilian population. Our preliminary results show a significant increased frequency of pro-inflammatory SNPs in PCa patients, when compared to age-related controls and centenarians (Table 2). In fact, as a further control group, we have utilized centenarians, since they are a typical human model of disease-free subjects, cancer included [88]. Incidentally, this confirms our previous data suggesting that pro-inflammatory alleles have an opposite role in longevity and age-related diseases, including cancer [4, 14]. To the best of our knowledge, no other genetic study has analyzed the functional variants of the 5-LO gene promoter in PCa.

Table 2.

Studies on the association of the pro-inflammatory polymorphisms of COX-2 and 5-LO gene between patients with prostate carcinoma, age-related controls and centenarians

Gene polymorphisms	Pro-inflammatory SNPs	Significance (P value with Yates’correction)
COX-2 −765 G/C	−765G	0.05 (PCa vs. controls) 0.018 (PCa vs. centenarians)
5-LO −1708 G/A	−1708A	0.01 (PCa vs.controls) 0.0005 (PCa vs. centenarians)

Toll-like receptors (Tlrs): a possible link between infection, inflammation, and prostate cancer

TLRs represent a family of pattern-recognition receptors able to detect microbial conserved components and trigger protective host responses. To date, 13 mammalian TLR members have been identified. Each TLR recognizes distinct pathogen-associated molecular patterns (PAMPs) derived from various microorganisms, including bacteria, viruses, protozoa, and fungi [89, 90]. TLRs, in particular TLR4, are also able to mediate responses against host molecules, such as ox-LDL, Aβ amyloid peptide, heat shock proteins, and those produced in response to tissue injury [91–93]. Recognition of their ligands leads to a series of signalling events, the first step of which is represented by the activation of IL-1 receptor (IL-1R) family followed by the activation of the transcription factor NF-KB with the transcription of pro-inflammatory genes [94]. Recently, the molecular mechanisms that connect infection, inflammation, and cancer have been reviewed, and it has been hypothesized that activation of NF-KB is a crucial mediator of inflammation-induced tumor growth and progression, as well as an important modulator of tumor surveillance and rejection [95–101].

In the PCa milieu, TLRs seem to constitute a possible link between infections, chronic inflammation, and tumor development. In particular, it has been proposed that activation of the innate immune response induced through TLR-mediated recognition of infectious agents or endogenous molecules, such as molecules produced by cell and DNA damage, can promote the development of PCa within an inflammatory environment [44, 98, 102, 103]. Inflammatory mediators released by NF-KB activated by TLRs pathway may contribute to cancer development and progression, such as PCa (Fig. 1).

Fig. 1.

Pathogens (sexually transmitted organisms, such as Neisseria gonorrhoea, Chlamydia trachomatis, Trichomonas vaginalis and Treponema pallidum, and non-sexually transmitted ones such as Propionibacterium acnes, Escherichia coli, and viruses such as papillomavirus, human herpes simplex type-2, Cytomegalovirus) or the endogenous molecules (molecules produced by cell and DNA damage), determine the activation of TLR-NF-kB pathway, on inflammatory and tumor cells, and the release of different pro-inflammatory mediators. The inflammatory microenvironment provides signals inducing tumor initiation and promotion, tumor growth, anti-apoptosis effects, angiogenesis, and metastasis (MMPs, metalloproteases; VEGF, vascular endothelial growth factor)

Taking into account the key role of TLRs in PCa development and immune tumor evasion, sequence variations of the genes codifying these molecules have been analyzed. To date variants in TLR4 gene and the TLR1-6-10 gene cluster have been linked to PCa risk.

The TLR4 gene maps on the chromosome 9 in the 9q32-q33 region and encodes the typical receptor able to recognise products of Gram-negative bacteria, such as the lipopolysaccharide and endogenous molecules. These ligands determine a differential activation in the expression of several pro-inflammatory genes as well as the regulation of growth/apoptosis [90], contributing to various inflammatory diseases, including cancer. These observations have been confirmed by three population-based PCa case–control studies, suggesting a role of TLR4 genetic [104–106] (Table 3). In light of recent literature data on the role of TLR4 polymorphisms in PCa, we examined the role of two TLR4 SNPs, +896A/G and +1196C/T, in PCa within the Sicilian population. This preliminary study shows a significant increase of +896A pro-inflammatory allele frequency in PCa patients, when compared to age-related controls and centenarians (Table 3). In fact, as a further control group, we utilized centenarians, since they are typical disease-free human models (see above).

Table 3.

Studies reporting the association of TLR4 gene with PCa risk

Gene	Polymorphisms	Risk association	Population	References
TLR4	11381G/C in the 3′-untranslated region (rs11536889)	Yes	Sweden	[104]
TLR4	a,bTLR4_1893,TLR4_2032,TLR4_2437, TLR4_7764,TLR4_11912,TLR4_16649, TLR4_17050, and TLR4_17923 11381G/C in the 3′-untranslated region	Negativeb No	USA	[105]
TLR4	5′ UTR polymorphism (rs10759932)	Yes	USA	[106]
TLR4	c+896A/G (Asp299Glyrs4986790) and b+1196C/T (Thr399Ile rs4986791)	Yesc Negativeb	Sicily	Unpublished data

^aInnate immunity names (http://innateimmunity.net) of SNPs

^bThese polymorphisms determine a lower risk of PCa

^cThe +896A proinflammatory allele is significantly increased in PCa patients with respect to centenarians (P = 0.0008, by χ² test with Yates’s correction) and age-related controls (P = 0.001 by χ² test with Yates’s correction)

The TLR1-6-10 cluster maps on the chromosome 4 in the 4p14 region and encodes molecules with a high degree of homology in their amino acid sequences. TLR6 and TLR1 recognize diacylated lipoprotein and triacylated lipoprotein, respectively. No specific ligand has been identified for TLR-10. Each TLR1 and TLR6 molecules form heterodimers with TLR2 that recognize several PAMPs [89, 90]. The association of some sequence variants of TLR1, TLR6, and TLR10 gene cluster with PCa [107–109] have been demonstrated by three recent studies (Table 4).

Table 4.

Studies on the association of TLR1-6-10 gene cluster with PCa risk

Gene	Polymorphisms	Risk association	Population	References
TLR6	−1401 A/G (rs5743795)	Yes	Sweden	[107]
	−673 C/T (rs5743806)	Yes
TLR1	−7202 A/G (rs5743551)	Yes
	−6399 C/T (rs5743556)	Yes
	−833 C/T (rs5743604)	Yes
TLR10	−3260 C/T (IIPGA-TLR10 −995)a	Yes
	−1692 C/T (IIPGA-TLR10 −1962)	Yes
	−260 A/G (rs4274855)	Yes
	720 A/C (IIPGA-TLR10 −4983)a	Yes
	1104 A/C (IIPGA-TLR10 −5367)	Yes
	2322 A/G (rs4129009)	Yes
TLR1			US population	[108]
TLR6
TLR10	19 common (>5%) haplotype-tagging SNPs	No
TLR1	N248S (rs4833095), S26L (rs5743596), rs5743595 and rs5743551	Negativeb	US population	[109]
TLR10	I369L (rs11096955) and N241H (rs11096957)	Negativeb

^aInnate immunity names (http://innateimmunity.net) of SNPs

^bThese SNPs seem to reduce the PCa risk

Taken together, these studies indicate that TLR-mediated inflammation from bacterial and viral infection or other endogenous or exogenous agents can influence the development of cancer, such as PCa.

Cytokine network

Several studies actually showed an increased level of pro-inflammatory cytokines, both in prostatic tissue and fluids in prostatitis. Recent reports indicate that both pro- and anti-inflammatory cytokine levels in prostatic fluid may be useful not only for assessment of prostatic inflammation, but also for early cancer detection and prognosis [23, 24].

Polymorphisms of cytokine genes can influence inflammation and immune response [14]. Cytokine gene polymorphisms associated with susceptibility to prostate cancer are observed in a large number of case–control studies, twin studies and segregation analyses (see below). It is widely hypothesized that the interactions of cytokine network genes, either additively or epistatically, determine the individual risk for PCa, as well as for BPH, also described as an immune-mediated inflammatory disease [110].

Transforming growth factor (TGF)-β1

TGF-β1 is a powerful and multi-functional modulator of cellular proliferation, differentiation, and the production and degradation of extracellular matrix [111]. Recent advances in research on the TGF-β1 signalling pathway have revealed that it is regulated by several mechanisms, and many oncogenic and anti-oncogenic proteins play important roles in tumor development under the control of the TGF-β1 [112]. Changes in the ability of the cells to react to stimulations of this cytokine have been associated with progression and transformation of the malignancy [113, 114]. Among the reported polymorphisms in the gene, two point mutations in codons 10 and 25 of the signal peptide sequence have been linked to variations in serum levels of the protein [111, 115]. The first one (Leu10Pro) is a transition of thymine to cytosine and the second one (Arg25Pro) is a transformation of guanine to cytosine. In both cases, the resultant amino acid is a proline. TGF-β1 is secreted as a latent complex and needs to be activated in order to bind to its receptors and release its signal. Hence, the presence of modified sequences in the signal peptide could be the factor responsible for the refractory behavior of TGF-β1 signalling, a characteristic of tumor cells in advanced stages of malignant progression. The amino acid substitution with distinct polarities, observed in the Leu10Pro SNP, drastically affects the three-dimensional conformation of the protein and its hydrophobicity, breaking the characteristic alpha-helix structure of the region and, consequently, losing its ability to localize the active TGF-β1 into a cell [116]. This would hinder an efficient transduction of the anti-proliferative and pro-apoptotic signals of the TGF-β1, favoring the clonal expansion of refractory tumoral cells. As reported in a recent paper, the 10Pro positive genotypes conferred a risk 1.8 times higher for the PCa event relative to the normal population [117].

IL-1

IL-1α and IL-1β might be considered the prototypic pro-inflammatory cytokines of the IL-1 family. The IL-1R antagonist (IL-1Ra) competes with IL-1α and IL-1β [118, 119]. IL-1α is produced by prostatic epithelial cells and induces fibroblast growth factor 7 (FGF-7) expression in prostatic stromal cells. FGF-7, in turn, induces epithelial proliferation and further increases IL-1α expression, thus ultimately leading to increased tissue mass in the prostatic transition zone, which is critical in the pathogenesis of BPH [120]. Senescent prostatic epithelial cells are reported as the source of IL-1α [121]. Thus, secreted IL-1α might be one of the major factors responsible for age-related growth and proliferation of prostatic epithelial cells observed in BPH. In a recent study, it was found that IL-1α is expressed in BPH, but not in PCa. IL-1Ra, in contrast, is expressed in both BPH and PCa, but its expression is progressively reduced with the tumor grade [122]. A recent study on the possible association between the polymorphisms of IL-1Ra demonstrated a significant association between the genotype distribution of IL-1Ra and the risk of BPH [123]. Regarding PCa, both significant and nonsignificant association were found with IL-1Ra alleles [124, 125].

IL-18

IL-18 is a multifunctional cytokine that induces interferon-gamma secretion and plays an important role in antitumor immunity mediated by type-1 positive regulation loop. In a study enrolling 265 patients with PCa and 280 age-matched male controls, statistically significant differences in the genotype and allele distribution of the -137 G/C SNP of the IL-18 gene among cases and controls were observed [126]. The IL-18 −137C positive genotypes were associated with a significantly increased risk of PCa. Although the functional significance of the aforementioned SNP is not yet fully established, it has been suggested that SNPs at positions −137 are associated with considerable changes in IL-18 expression. Upon stimulation, a low promoter activity was observed for C allele at position −137. So a reduced stimulation of type-1 response might be involved in PCa susceptibility [126–128].

IL-8 and IL-10

IL-8 and IL-10 are the most studied cytokines in PCa. IL-8 has known functions in the regulation of angiogenesis and tumor growth. A recent report showed that neutralizing antibodies to IL-8 inhibited angiogenesis in a human PCa cell line/murin model and reduced tumorigenicity in vivo, implicating IL-8 as an important modulator of PCa growth [129]. IL-10 may play a role in PCa through its ability to inhibit both the production of pro-inflammatory cytokines, such as IL-1, IL-6, and IL-8, and to inhibit angiogenesis, as shown by experimental studies [130, 131]. A preliminary investigation showed an association with the risk for PCa for both the IL-8 SNP −251 TT (low producer of IL-8) and the IL-10 SNP −1082 AA (low producer of IL-10), but subsequent investigations did not fully replicate these findings [132–137]. Heterogeneity in the patients and controls enrolled across the individual studies and small sample size might be the main causes of inconsistency of the results. To overcome these limitations and to better define the role of these cytokine polymorphisms in the disease, we performed meta-analyses of published case–control studies on IL-10 −819 and −1082, and IL-8 −251 SNPs. The results of the random effects model meta-analyses for the correlation between the examined SNPs and the risk for PCa are shown in Table 5. The retrieved studies are arranged according to the year of publication [132–137]. The low producer genotypes of IL-10 (−1082 AA and −819 CC) and IL-8 (−251 TT) were compared with the other two aggregated genotypes. Overall, no association was detected between the three examined SNPs and PCa risk. As expected, a significant heterogeneity among the results of individual studies was reported. This is possibly related to heterogeneity of both under study and control populations as well as to differences in the sample size between the studies. The first published and the one reporting significant associations was a small sample, and controls were unrepresentative of the cases, since they were composed of men and women\drawn from an organ donor bank [134]. In contrast, the other included studies were all age-matched case–control studies enrolling only men. The sample size ranged from 484 to 1,245, larger than the sample size of the fist study (number of cases = 247), as well as the number of controls.

Table 5.

Random effects meta-analyses on pooled genotype raw data for the correlation between IL-10 −819, IL-10 −1082 and IL-8 −251 SNPs and PCa risk

Meta-analysis	Studies (year of publication)	Model	OR (95% CI), P value	Heterogeneity P value	Cases versus controls (numbers of independent samples)
IL-10 −819	Michaud (2006) [135]	CC vs. CT + TT	1.05 (0.84–1.30), P = 0.67	0.04	1,246 vs. 1,742
	Eder (2007) [132]				545 vs. 547
	Faupel-Badger (2008) [133]				507 vs. 384
	Zabaleta (2008) [137]				557 vs. 547
					Total 2,298 vs. 2,693
IL-10 −1082	McCarron (2002) [134]	AA vs. AG + GG	1.07 (0.88–1.30), P = 0.49	0.07	247 vs. 223
	Michaud (2006) [135]				1,245 vs. 1,743
	Eder (2007) [132]				545 vs. 547
	Faupel-Badger (2008) [133]				509 vs. 382
	Zabaleta (2008) [137]				557 vs. 547
					Total 2,546 vs. 2,915
IL-8 −251	McCarron (2002) [134]	TT vs. AT + AA	1.02 (0.72–1.46), P = 0.75	0.02	238 vs. 235
	Yang (2006) [136]				541 vs. 448
	Michaud (2006) [135]				484 vs. 613
					Total 1,263 vs. 1,296

However, it must be noted that the lack of consistency of single SNP analysis in PCa susceptibility may be due to the relatively minor effect that a single SNP may have on the disease. It is more probable that a combination of SNPs in haplotypes or an SNP–SNP interaction may modify the risk for developing a malignancy. In one of the studies included in our meta-analysis [137], it was shown that on analyzing the interactions of nine functionally characterized SNPs of three cytokine genes (IL1-β −511 CT, IL-1β −31 TC, IL-1β +3954 CT, IL-10 −1082 AG, IL-10 −819 CT, IL-10 −592 CA, TNF-α −857 CT, TNF-α −308 GA and TNF-α −238 GA), a single SNP did not modify, unless marginally when adjusted for age, family, smoking, and BPH, the risk of developing PCa, but it was noted that the risk was greatly modified by SNP–SNP interaction.

Chemokines

Many types of cancer cells, including PCa cells, express chemokines and their receptors [138]. Recent studies have shown the production of the CC-chemokine CCL5 (RANTES), a potent chemotactic factor for inflammatory cells and its receptor (CCR5) in different human PCa cell lines [138, 139]. Possible functions of the CCL5 and CCR5 axis in PCa progression are the induction of PCa cells proliferation and invasion [139]. It has been recently demonstrated that the inflammatory chemokine CCL5 may function as an autocrine factor that binds to its CCR5 receptor, expressed on PCa cell surface, and activates cellular responses involved in cancer progression [138]. We have recently evaluated whether CCR5Δ32 deletion of CCR5 gene might be associated with PCa susceptibility. As a control group, we utilized centenarians, since they are a typical human disease-free model (see above). Our preliminary results suggest that CCR5Δ32 anti-inflammatory variant is negatively associated with PCa development [140].

Prostate cancer, inflammation and aging

PCa is the most common non-cutaneous malignant neoplasm in men in Western populations. Its incidence is rapidly increasing in men over 50 years of age [141]. The correlation of PCa incidence with aging suggests that the disease burden associated with PCa will dramatically increase over the next several decades. Recent suggestions emphasize the crucial role of several genetic factors, including inflammatory mediator genetic polymorphisms and environmental factors, such as infectious agents and dietary carcinogens, and hormonal imbalances in the development of PCa [21, 44, 141]. Chronic inflammation seems to be a pathogenic factor common to all these conditions. As discussed, inflammation may be a direct carcinogen by damaging DNA. In addition, it creates a tissue microenvironment rich in cytokines, chemokines, and growth factors that can enhance cell replication, angiogenesis, and tissue repair [4, 14, 16, 21, 44].

PCa is a heterogeneous disease with multiple loci contributing to its susceptibility. It has also been evidenced that more important in terms of inherited susceptibility for PCa are common SNPs in a number of low penetrance alleles of several genes, the so-called genetic modifier alleles. The list of these variants is long, but the major pathways currently under examination include those involved in DNA repair, carcinogen metabolism, and inflammation pathways (for an extensive list see [21, 44]).

We reviewed evidences involving inflammation in the development of human PCa. Data seem to reinforce the largely shared feeling that many modifier genes underlie the genetic susceptibility to prostate cancer. It is widely hypothesized that the interactions of these genes, either additively or epistatically, determine the individual risk for PCa. In addition, the above data summarized clearly indicate that new tools and strategies are necessary to identify unequivocal genetic markers in PCa. A winning strategy might be linked to systematic genotype haplotype-tagging SNPs in a large number of genes among thousands of subjects, and the availability of the mathematical analysis methods and the computing power required for model high-order interactions.

A data mining method, multifactor dimensionality reduction, as a way to reduce the dimensionality of multilocus informations, was recently applied, in order to improve the identification of polymorphism combinations associated with disease risk [142]. A four-SNP interaction, one SNP each from IL-10 and IL-1RN and two from TLR signalling pathway, was identified to classify and predict which individuals were affected by PCa based on any combination of two, three, or four variants from all the genotyped variants [142]. From our point of view, the most relevant finding presented in this paper is that among the four implicated SNPs of four genes, only the SNP in TIRAP, an intracitoplasmatic protein that mediates downstream signalling of TLR2 and TLR4 [143, 144], had a significantly different allele frequency between cases and controls when analyzed with formal statistical methods, whereas no significant differences in the allele frequencies between cases and controls were observed for the other three SNPs. In other words, only by applying an advanced method of data analysis, crucial information regarding polymorphisms involved in PCa susceptibility could be obtained and possibly used as therapeutic target. In addition, the interaction identified among TLR5, IL-1RN, TIRAP, and IL-10 in PCa seems biologically plausible. TLR5 and IL-1R recognize and bind bacteria, viruses, and other ligands. IL-1RN is a protein that binds to IL-1R receptors and inhibits the binding of IL-1α and IL-1β. The engagement of ligands on these receptors initiates a series of downstream signalling cascades, including adaptor proteins such as TIRAP. The union of adaptor molecules with receptors leads to the activation of IL-1R-associated kinase, IRAK, and results in the production of various pro- or anti-inflammatory cytokines [142].

Another contribution to the full understanding of the genetic component of PCa susceptibility might be achieved from studies on longevity. The majority of cases of cancer and in particular prostate cancer occur in patients over the age of 65 [1, 141]. However, a recent investigation on autopsy records revealed that Japanese centenarians are characterized by a lower than expected incidence of cancer, together with a decline of metastatic rate, and a decrease in mortality due to cancer [145]. Furthermore, a study of 507 autopsies of Italian individuals revealed that the prevalence of cancer was 35, 20, and 16% among people aged 75–90 years, 95–99 years, and over 99 years, respectively. The prevalence of metastases was 63% in people aged 75–90 years, 32% in patients aged 95–98 years, and 29% in centenarians [146]. As already stated, centenarians are exceptional individuals who have escaped major common age-related diseases, cancer included [88].

As previously reported, we have recently demonstrated that CCR5Δ32 deletion polymorphism of CCR5 gene as well as the anti-inflammatory alleles of TLR4, COX-2 and 5-LO alleles, might be a component of the protective genetic background against PCa as their frequencies are highly significantly different from that observed in a group of centenarians. In fact, data obtained in Italian centenarians, particularly in males, suggest that a pro-inflammatory genotype is unfavorable to reach extreme longevity in good health and likely favors the onset of age-related inflammatory diseases. Actually, the data coming from the longevous male population shows that the genetic polymorphisms responsible for a low inflammatory response are significantly increased in centenarians when compared to young controls (for a list of pro-inflammatory and anti-inflammatory alleles significantly different between centenarians and controls see [14, 64, 88, 93, 147–149]). Hence, it can be hypothesized that a systematic analysis of pro- or anti-inflammatory allele frequencies in centenarians, in PCa patients, and age-matched controls in different populations might allow us to obtain strong convincing evidence of the genetic profile characterizing susceptibility and/or resistance to PCA.

On the other hand, relationships between cancer and aging are outstandingly complex, therefore most of cancer-predisposing factors may substantially influence the probability to achieve old age. In fact, most of tissue changes, which are classified as hallmarks of cancer may be also hallmark of aging [150, 151]. As an example, the apoptosis-deficient Pro/Pro genotype p53 gene polymorphism increases both the probability to achieve the extreme limit of human longevity and the chance of dying from cancer [152]. For prostate cancer, we can consider also genes coding detoxification enzymes, such as the cytochrome P450 1B1 (CYP1B1) gene. CYP1B1 plays a role in the process of activation of many carcinogens, like polycyclic aromatic hydrocarbons that are deposited in adipose tissues. In this view, it is of interest that a post hoc comparison does not allow us to find differences of CYP1B1 polymorphism frequencies between cancer patient and elderly control groups, whereas statistical differences were observed with the age-related group [151].

Conclusions: a pharmacogenomic approach

Early detection of PCa has proved difficult, and current detection methods are inadequate. Prostate-specific antigen (PSA) testing is a significant advance for early diagnosis of patients with PCa. PSA is produced almost exclusively in the prostate, and abnormalities of this organ are frequently associated with increased serum concentrations. Because of PSA’s lack of specificity for PCa, however, many patients undergo unnecessary biopsies or treatments for benign or latent tumors, respectively [153]. Thus, other diagnostic approaches, such as those described here, are required to augment or replace screening with PSA.

In any case, in age-related diseases, the search for parameters capable of identifying the pro-inflammatory profile of individuals is highly welcome and represents a priority for research, owing to its implication for diagnosis, prevention, and therapy [154, 155]. The translation of pharmacogenomics into clinical practice will allow bold steps to be taken toward personalized medicine.

The research of drugs that selectively interfere with the sequence of events triggered by the genetic mechanism(s) underlying this inflammatory age-related disease may be a good example of how genotyping is incorporated into clinical drug therapy in order to bridge the gap between pharmacogenetic research and clinical application. Taking into account the immune-inflammatory genes,we have reviewed in the present paper, we can formulate the following working hypothesis for the therapeutic treatment of subjects with severe risk factors for PCa before the clinical appearance of the disease. In particular, concerning the COX-2 and 5-LO genes, the presence of high responder alleles suggests the possibility of preventive treatment with specific inhibitors of eicosanoids or their enzymes. For people who do not respond or handle NSAIDs therapy [156], other more sophisticated possibilities include: (1) the occurrence of a high risk genetic profile linked to the presence of high responder alleles of pro-inflammatory cytokines or of low responder alleles of anti-inflammatory cytokines might suggest the treatment with biologics as monoclonal antibodies directed versus the pro-inflammatory cytokines. To decrease the level of systemic inflammation in PCa patients might be a suitable chemopreventive treatment. In addition, it may be relevant that in an animal models of rheumatoid arthritis, where cytokines play an important role, promising results have been obtained utilizing suppressors of cytokine signalling [157]. In addition, ongoing studies are aimed at examining the effectiveness of chemokine receptor antagonists as novel strategies in cancer treatment [158]. Their application as combined therapeutic agents in PCa may be an important area of future study, as demonstrated by our data on CCR5Δ32 deletion. (2) Subjects, carriers of high responder TLR SNPs, might be selected for clinical trial with agonists to block TLR-NF-KB signalling pathway required for induction of inflammation or the release of pro-inflammatory mediators. This blockade might, indeed, reduce the risk of tumor development [159].

In conclusion, the better knowledge of the function and the regulation of inflammatory pathway in cancer may help to understand the mechanisms of PCa formation and progression, as well as to identify new targets for the refinement of new treatment such as the pharmacogenomics approach.