**1. Introduction**

Since 1980, the prevalence of obesity has doubled in the world. Obesity is caused by genetic factors, neuroendocrine factors, psychological factors, and environmental factors [1]. In the United States, almost 40% of people suffer from obesity, and the present situation is a "pandemic" of obesity. In Asian countries such as Korea and Japan, the prevalence of obesity is still low, approximately less than 10%, but the prevalence of obesity has increased over the last decade [1,2]. The incidence rate of prostate cancer is also increasing and is now highest in Japan. Overfeeding with a high-fat and/or high-calorie diet and less physical activity result in an energy imbalance and adiposity. Obesity causes insulin resistance, type 2 diabetes, cardiovascular diseases, and several malignancies via systemic inflammation. The resulting medical costs due to obesity are increasing and becoming an important issue worldwide.

Prostate cancer has had high morbidity among elderly men. Many patients with prostate cancer are in the early stage and have good prognosis after several treatments including prostatectomy, radiation therapy, hormonal therapy, and even active surveillance. However, some progressive prostate cancer patients in the late stage with poorly-differentiated cancer cells, or local invasion, or metastatic lesion are more resistant to several treatments including hormonal therapy or chemotherapy, and have poor prognosis. (Figure 1) It is important to elucidate the mechanism of the factors inducing prostate cancer progression.

**Figure 1.** The scheme of different stages and progression of prostate cancer.

Chronic inflammation is the major etiology behind the development of several cancers, such as hepatocellular carcinoma, squamous cell carcinoma in the urinary bladder, colorectal cancer, and gastric cancer. Inflammatory cells migrating to the local area generate reactive oxygen species and reactive nitrogen species that induce mutations of DNA in normal epithelia [3]. Acute or chronic inflammation is a common histological finding in both benign and malignant tissues in prostatectomy specimens [4,5]. The causes of inflammation in the prostate vary among bacteria causing prostatitis and sexually-transmitted disease, hormonal changes of estrogen [6], physical trauma caused by corpora amylacea [7,8], urine reflux to the prostate gland, and environmental factors such as dietary habits [9,10]. Dietary habits cause inflammation of the prostate and can result in carcinogenesis in the early stage [11,12]. Dietary-induced inflammation could last for the entire life, and chronic inflammation can also stimulate the progression of prostate cancer in the late stage. However, the association of immune cells in tumor microenvironments with prostate cancer is still unclear.

In this review, the link between obesity and prostate cancer is discussed based on the recent findings related to inflammation.

### **2. Obesity and Prostate Cancer**

Several studies reported that obesity was associated with the increased risk of several cancers, such as colon, breast, endometrial, kidney, gastric, esophagus, pancreas, liver, and gall bladder [13,14]. Several studies have shown the association of obesity with the risk of prostate cancer. A prospective study of 3673 men in the United States showed that greater body mass index (BMI) was an independent predictor of prostate cancer (relative risk = 1.7 for BMI > 27.8 kg/m<sup>2</sup> compared with <23.6 kg/m2; *p* = 0.1). The percent change in BMI from baseline to age 50 was also positively associated with risk (*p* = 0.01) [15]. Another prospective study in the United States showed that BMI was weakly and positively associated with prostate cancer, and the association of obesity with the risk of clinically-significant prostate cancer strengthened after the exclusion of well-differentiated, localized tumors [16]. However, a prospective study of 36,959 Swedish men showed that the incidence of localized prostate cancer was inversely associated with BMI in middle-to-late adulthood (the rate ratio for 35 kg/m<sup>2</sup> when compared with 22 kg/m<sup>2</sup> was 0.69 (95% confidence interval (CI) 0.52–0.92)), but not in early adulthood. BMI in middle-to-later adulthood was associated with a non-statistically significant increase in the risk of fatal prostate cancer (rate ratio for every five-unit increase: 1.12 (0.88–1.43)) and BMI in early adulthood with a decreased risk of fatal prostate cancer (rate ratio for every five-unit increase: 0.72 (0.51–1.01)) [17]. A prospective study of 141,896 men in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort showed that high BMI at a young age was inversely associated with the overall risk of prostate cancer (relative risk = 0.89, 95% CI 0.80–0.98, BMI ≥ 26 vs. 20–21.9, *p* = 0.01) and with fatal and advanced disease [18]. Obesity at a young age causes the delayed onset of puberty and may result in the lower lifetime exposure of insulin-like growth factor 1 (IGF-I), which may affect the development of prostate cancer later in life [18,19]. A meta-analysis of 12 prospective studies of localized prostate cancer (1,033,009 men, 19,130 cases) and 13 of advanced prostate cancer (1,080,790 men, 7067 cases) showed an inverse linear relationship with BMI for localized prostate cancer (*p* ≤ 0.001, relative risk: 0.94 for every 5-kg/m<sup>2</sup> increase) and a positive linear relationship with BMI for advanced prostate cancer (*p* = 0.001, relative risk: 1.09 for every 5-kg/m<sup>2</sup> increase) [20]. Obesity thus could affect the incidence of the risk of prostate cancer in the early stage in the opposite direction according to the type of prostate cancer. The underlying mechanisms of this inverse association of obesity with localized prostate cancer could be the low testosterone levels in obese men. Obese men have a lower concentration of free testosterone due to a decrease of lutenizing hormone (LH) pulse amplitude and serum LH levels [21]. Plasma total testosterone and free testosterone were positively associated with increased risk of low-grade prostate cancer [22]. However, the association of testosterone, free testosterone, and the free-to-total testosterone ratio with prostate cancer is still controversial [23]. Furthermore, the impact of obesity-induced systemic inflammation on the inverse relationship of localized prostate cancer to BMI is still unknown.

Obesity may also affect the prognosis of prostate cancer in the late stage. An analysis of 4123 men treated by radical prostatectomy showed that higher BMI was associated with biochemical recurrence after radical prostatectomy (hazard ratio (HR) 1.02, 95% CI 1.00–1.02, *p* = 0.008) [24]. A retrospective analysis of 4268 radical prostatectomy patients within the Shared Equal Access Regional Cancer Hospital (SEARCH) database showed that being overweight and obesity were associated with prostate cancer-specific mortality (HR 1.88, *p* = 0.061 and HR 2.05, *p* = 0.039, respectively) [25]. A prospective study of 404,576 men showed a positive linear trend in the prostate cancer death rate with higher BMI (*p* < 0.001) [14]. These epidemiological studies showed obvious evidence of the association of obesity with advance prostate cancer.

### **3. Obesity and Inflammation**

Many studies have shown that obesity causes systemic inflammation through the action of various mechanisms. Adipocytes secrete tumor necrosis factor (TNF)- α in obese mice that causes systemic inflammation [26]. A high-fat diet (HFD) changes the intestinal microbiota and increases the translocation of live Gram-negative bacteria through the intestinal mucosa into the bloodstream and mesenteric adipose tissue, which results in continuous bacteremia [27]. Fatty acids activate toll like receptor 4 (TLR4) signaling in adipocytes and macrophages. Female mice lacking TLR4 show increased obesity, but are partially protected against HFD-induced insulin resistance, possibly due to reduced inflammatory gene expression in the liver and fat [28]. Obesity induces activation of the innate immune system. Adipose depots contain multiple immune cells. Macrophages in adipose tissues are increased in the obese, skewing to the M1-polalized macrophages. These macrophages show a pro-inflammatory phenotype and secrete inflammatory cytokines such as TNFα [29].

It is still unclear how such systemic inflammation affects local inflammation of the prostate (Figure 2). Several chemokines and cytokines secreted from prostate cancer cells may recruit immune cells to the prostate. Which organ are these immune cells activated in? Some immune cells could be "taught" in the intestinal wall [30], but there has been no evidence of the homing of these intestinal immune cells to a distant organ. In bone marrow or regional lymph nodes, the immune cells might be activated by factors related to obesity and subsequently recruited to the prostate. Otherwise, the local immune cells recruited by prostate cancer cells might be activated by the obesity-related factors. The elucidation of these factors related to obesity could lead to the development of new treatments or the prevention of prostate cancer in the early stage.

**Figure 2.** Interaction of immune cells with adipocytes and prostate cancer cells.

### **4. Obesity Promotes Prostate Cancer Growth**

Although the link between obesity and prostate cancer has not been definitively determined, several studies focusing on the cytokines and/or chemokines have been reported. In a mouse xenograft model of the prostate cancer cell line LNCaP, serum monocyte chemoattractant protein-1 (MCP-1) was significantly increased, and tumor growth was promoted in HFD-fed mice [31]. Palmitic acid is one of the saturated free fatty acids abundantly included in HFDs. The addition of palmitic acid induced the expression of macrophage inhibitory cytokine 1 (MIC1) in vitro, and serum levels of MIC1 were increased in the HFD-fed mice xenograft model. Obese patients with prostate cancer were also found to have higher serum levels of MIC1 than those in healthy controls [32]. HFDs also modulate miRNA expression in prostate cancer cells. Prostate cancer cells cultured in the serum of HFD-fed mice showed a marked increase in cell proliferation and the attenuation of miR-130a. miR-130a modulated MET expression in prostate cancer cell lines, and furthermore, cytoplasmic MET in prostate cancer tissues was overexpressed in patients with higher BMI [33]. An HFD also induced increases in leptin, C-C motif ligand (CCL)3, CCL4, CCL5, and C-X-C motif ligand (CXCL)10 in the sera of transgenic adenocarcinoma of mouse prostate (TRAMP) mice. The conditioned medium of sera from HFD-fed TRAMP mice promoted the proliferation, migration, and invasion of DU-145 cells [34]. Obese patients with prostate cancer showed increased expression of epithelial CXCL1, which induces the recruitment of adipose stromal cells from white adipose tissue to the tumor and promotes the tumor's growth [35]. These reports showed that cytokines and chemokines could play important roles in the obesity-associated progression of prostate cancer in the early and late stage. Because TRAMP mice lacking expression of androgen receptor are thought to be models for a very advanced stage with neuroendocrine cancer cells and independent from androgen receptor, the findings using TRAMP mice might be compatible with prostate cancer patients in only the late stage. Moreover, the detailed mechanisms including the tumor microenvironments are still unknown.

### **5. Inflammation in Prostate Cancer**

In the tumor microenvironments, the interactions among cancer cells, immune cells, endothelial cells, and fibroblasts can play important roles. Inflammatory cells consist of innate immune cells and acquired immune cells. Acquired immune cells include B cells and T cells, which act based on antigen recognition. While innate immune cells are the main players in inflammation, innate immune cells and acquired immune cells also orchestrate the inflammation. Innate immune cells including neutrophils, myeloid cells, mast cells, and macrophages are different from acquired immune cells by receptor-mediated activation and their rapid response to invading pathogens and foreign bodies [36]. Macrophages and neutrophils are the most abundant immune cells in the tumor microenvironment [37].

To reveal the relationship between HFD-induced inflammation and tumor progression in the prostate, we used two genetically-engineered prostate cancer mouse models, prostate-specific *Pten* knockout mice (Pb-Cre+; *Pten*(fl/fl)) and *Pten* and *Tp53*-double knockout mice (Pb-Cre+; *Pten*(fl/fl); *Tp53*(fl/fl)) on the C57BL/6 genetic background. The prostate weights and the ratio of Ki67-positive cells to tumor cells, which indicates the proliferative capacity of the tumor, of the mice in the HFD-fed double knockout mouse model were significantly higher than those of the control diet (CD)-fed model mice (*p* = 0.011, *p* = 0.005, respectively) (Figure 3A,B). Total RNA was isolated from prostatic tissues of both the CD-fed mice and HFD-fed double knockout mice, and transcriptome analysis of the two groups was performed using mRNA microarray technology. Gene ontology analysis revealed that many processes related to inflammation and the immune response were ranked in the top 22 processes expressed in the prostate of the HFD-fed double knockout mice (Figure 3C). This finding strongly suggests that local inflammation of the prostate is one of the most important factors for the progression of prostate cancer in obese or HFD-fed mice in the early and late stages. The profiles of the local immune cells in prostate cancer were analyzed in the *Pten* knockout mouse model fed with a CD or HFD. Although the number of B cells, T cells, macrophages, and mast cells and the ratio of CD8/CD4 T cells were not changed by the HFD, the number of myeloid-derived suppressor cells (MDSCs) and the M2/M1 macrophage ratio were significantly increased in the HFD-fed mice compared with the CD-fed mice. The promotion of tumor growth by the HFD was completely cancelled by the administration of celecoxib, a cyclooxygenase 2 (COX-2) inhibitor, which suggests that inflammation plays a central role in tumor progression caused by an HFD. IL-6 expression in prostate tissues was increased in HFD-fed mice, as were the amounts of phosphorylated signal transducer and activator of transcription 3 (STAT3) in prostate cancer cells. Inhibition of the IL-6 pathway resulted in the suppression of tumor growth by an HFD [38]. The HFD and subsequent obesity caused the increased secretion of IL-6 from local macrophages in the prostate tumor via unclear mechanisms. IL-6 might increase the number of local MDSCs and promote the proliferation of prostate cancer cells via signal transducer and activator of transcription 3 (STAT3) pathways. Because transcriptome analysis in double knockout mice resulted in different changes of gene expressions from *Pten* knockout mice after administration of HFD, *Tp53* may have many functions regarding inflammation. In addition, it might result in different findings if model mice on the other genetic background were to be examined.

**Figure 3.** (**A**) Representative gross findings of the prostatic tissues (the black bar indicates 5 mm) (top) and prostate weights (*n* = 6 and 4, respectively) (bottom) of the model mice at 22 weeks of age. (**B**) Representative images of Ki67 staining for the prostatic tissues (top) and the ratio of Ki67-positive cells to tumor cells (*n* = 4 and 3, respectively) (bottom) of the model mice at 22 weeks of age. (**C**) Gene ontology analysis using mRNA microarray technology of the prostatic tissues of the model mice at 22 weeks of age (HFD-fed vs. CD-fed, *n* = 3, respectively; fold change >2.0, *p* < 0.05, biological process). CD, control diet; HFD, high-fat diet. \* *p* < 0.05, \*\* *p* < 0.01.
