**4. Potential Mechanisms**

Previous studies have proposed several mechanisms in order to explain the possible association between dietary fat and prostate cancer development/progression. Accordingly, growth factor signaling, lipid accumulation, inflammation, and endocrine modulation had been hypothesized to be associated with HFD-induced prostate cancer development and/or progression (Figure 1). Certainly, a more thorough understanding of the possible association between dietary fat and prostate cancer risk requires further inquiry.

**Figure 1.** Scheme of potential mechanisms underlying high-fat diet induced prostate cancer development and/or progression.

### *4.1. Growth Factor Signaling*

Obesity and hyperinsulinemia have been associated with increased amounts of circulating bioactive insulin-like growth factor-I (IGF-I), a growth factor determined to play a pathogenic role in many cancers [81]. Barnard et al. demonstrated that dietary fat reduction combined with a regular exercise intervention in men decreased serum IGF-I and increased serum IGFBP-1 levels, resulting in decreased growth of LNCaP human prostate cancer cells cultured in media containing volunteer serum [14]. The same group showed that LFD-fed mice had significantly slower tumor growth rates, lower levels of serum insulin, tumor IGF-I mRNA expression, and tumor IGFBP-2 immunostaining, and higher levels of serum IGFBP-1, which indicated that IGF-I signaling modulated fat-induced tumor growth in LAPC-4 xenografts [15]. We had previously demonstrated that IGF-I receptor (IGF-IR) mRNA levels were strikingly elevated in HFD-accelerated LNCaP xenografts and that the group having the lowest IGF-IR immunoreactivity tended to have the lowest body mass index in both human normal and prostate cancer epithelia [16]. Kobayashi et al. showed that an LFD reduced the development of prostate cancer in Hi-Myc mouse transgenic model with the suppression of the IGF-AKT pathway, which leads to higher serum IGFBP-1 levels, reduced serum mitogenicity, and lower AKT, GSK3beta, and S6K activities [19].

Several studies have demonstrated that hyperactivation of PI3K-AKT, which is one of the downstream targets for IGF-I signaling, desensitizes tumors to dietary modulations, including calorie restriction and a HFD [23,47]. The PI3K/AKT pathway is naturally inhibited by Pten, which is one of the most frequently lost or mutated tumor suppressor genes in prostate cancer [56]. Partial loss of *PTEN* is observed in 70% of localized prostate cancer, while complete loss thereof is associated with metastatic castration-resistant prostate cancer [56]. *PTEN* inactivation also induces aberrant activation of the PI3K/AKT pathway. As previously mentioned, conditional PTEN knockout produces indolent tumors in mouse prostates. One study that assessed the impact of diet restriction revealed that it does not a ffect a PTEN-null mouse model of prostate cancer but significantly decreases tumor burden in a mouse model of lung cancer lacking constitutive PI3K signaling, which suggests that PI3K signaling is strongly associated with diet-induced cancer progression [23]. Another study involving a GEMM mouse model showed that the loss of both PTEN and the protein tyrosine phosphatase Pypn1, a negative regulator of IGF-IR, enabled the development of a highly invasive prostate tumor, whereas PTEN deficiency alone resulted in tumors that were unresponsive to HFD [47]. Collectively, mechanisms involving PTEN and other related genes may have a higher impact on diet-induced prostate cancer development and progression.

Many other studies have proposed that IGF-I/PI3K/AKT signaling has an impact on diet-induced prostate cancer development and progression [21,28,43]. Therefore, IGF-I/PI3K/AKT signaling has been one of the promising pathways related to HFD-induced prostate cancer development and progression. To determine the impact of treatment, the additive e ffect of IGF-1R inhibition using IGF-IR blockade antibody on 22Rv1 subcutaneous xenografts in SCID mice receiving a HFD (43.3%) or LFD (12.4%) had been investigated [32]. Accordingly, the results showed that the LFD + IGF-1R-Ab group had a significantly smaller mean tumor volume compared to the HF group at day 14 of the intervention. However, no significant di fference in final tumor volumes or final tumor weights had bene observed between the four treatment groups. Therefore, the therapeutic e ffect of IGF-I pathway inhibition remains unknown.

Diet-induced hyperinsulinemia has been shown to accelerate tumor growth in di fferent prostate cancer xenograft models [17,20]. A large prospective survival analysis reported that higher serum C-peptide concentrations, a surrogate of insulin levels, were associated with increased prostate cancer-specific mortality [82]. Insulin and IGF-I are closely related hormones that act on specific tyrosine kinase receptors and elicit the activation of a cascade of intracellular proteins leading to the regulation of gene expression, protein synthesis, cell proliferation or death, and glucose and lipid metabolism. High insulin levels, as well as insulin receptor and IGF-I/IGF-IR axis activation, have been known to be associated with obesity induced cancer progression [83]. Regarding the impact of insulin levels on diet-induced prostate cancer growth, one study involving LAPC-4 xenografts in SCID mice receiving three di fferent diets, NCKD (84% fat), 10% carbohydrate diet (74% fat), or 20% carbohydrate diet (64% fat), proposed that mice receiving a 10% carbohydrate diet had larger tumors than the other groups despite mice receiving a 20% carbohydrate diet having the lowest insulin levels [28]. As such, future studies need to elucidate the relationship between insulin levels and diet-induced prostate cancer carcinogenesis.

In addition, e ffects of di fferent fat sources on the IGF/insulin axis have rarely been discussed and studied. It would be intriguing to know the varying impacts specific fats have on HFD-induced prostate cancer development and progression through growth factor signaling.
