Excessive intake of animal fat and resultant obesity are major risk factors for prostate cancer. Because the composition of the gut microbiota is known to change with dietary composition and body type, we used prostate-specific Pten knockout mice as a prostate cancer model to investigate whether there is a gut microbiota–mediated connection between animal fat intake and prostate cancer. Oral administration of an antibiotic mixture (Abx) in prostate cancer–bearing mice fed a high-fat diet containing a large proportion of lard drastically altered the composition of the gut microbiota including Rikenellaceae and Clostridiales, inhibited prostate cancer cell proliferation, and reduced prostate Igf1 expression and circulating insulin-like growth factor-1 (IGF1) levels. In prostate cancer tissue, MAPK and PI3K activities, both downstream of the IGF1 receptor, were suppressed by Abx administration. IGF1 directly promoted the proliferation of prostate cancer cell lines DU145 and 22Rv1 in vitro. Abx administration also reduced fecal levels of short-chain fatty acids (SCFA) produced by intestinal bacteria. Supplementation with SCFAs promoted tumor growth by increasing IGF1 levels. In humans, IGF1 was found to be highly expressed in prostate cancer tissue from obese patients. In conclusion, IGF1 production stimulated by SCFAs from gut microbes influences the growth of prostate cancer via activating local prostate MAPK and PI3K signaling, indicating the existence of a gut microbiota-IGF1-prostate axis. Disrupting this axis by modulating the gut microbiota may aid in prostate cancer prevention and treatment.

Significance:

These results suggest that intestinal bacteria, acting through short-chain fatty acids, regulate systemic and local prostate IGF1 in the host, which can promote proliferation of prostate cancer cells.

Prostate cancer is presently the second most common cancer worldwide and the fifth most common cause of cancer death in men (1). In 2018, there were about 1.2 million new cases and about 360,000 deaths from prostate cancer, a trend that is increasing. Therefore, elucidating the mechanisms of prostate cancer progression and the development of new prevention and treatment methods are important goals.

Lifestyles, including the diet, have long been linked to prostate cancer (2). In particular, excessive intake of animal fat and resultant obesity, which are responsible for lifestyle-related diseases, has received significant attention as a clear risk factor for prostate cancer. Multiple epidemiologic and basic studies suggested that fat intake, especially of animal fat, increased the risk of prostate cancer. In our previous study using a prostate-specific Pten knockout mouse model of prostate cancer, we found that a high-fat diet (HFD) containing a lot of lard induced prostate local inflammation and that celecoxib, an anti-inflammatory agent, was able to inhibit prostate cancer growth (3). Other than this inflammatory and immune-mediated mechanism, several mechanisms, including sex hormones or oxidative stress, have been described (4, 5). Thus, multiple mechanisms are likely to be involved in mediating the relationship between HFD-induced obesity and prostate cancer, and there may still be further unknown mechanisms.

Diet and other lifestyle aspects have been shown to alter the metabolites in the gut and the composition of the gut microbiota (6). In recent years, attention has been focused on the effects of gut microbiota on homeostasis in humans, and numerous clinical and basic studies have reported its association with various diseases, such as rheumatoid arthritis, ulcerative colitis, and Alzheimer disease (7–9). In several types of cancer, the gut microbiota has been shown to be involved in carcinogenesis and its progression (10–12). In addition, there have even been reports concerning cachexia and the therapeutic effects or side effects of drug treatment, suggesting that the gut microbiota is deeply involved in various aspects of cancer (13). Many of the reports have focused on organs that are strongly associated with the gut microbiota, such as the intestine and liver; however, reports on organs distant from the gut microbiota, such as the prostate, have been limited. With regard to prostate cancer, increased proportions of certain bacteria have been found in the gut microbiota of patients with prostate cancer (14); however, there is still no basic research describing the effects of gut microbiota on prostate cancer.

We hypothesized that animal fat intake and obesity could be involved in the growth of prostate cancer through its effect on gut microbiota and studied this connection to clarify the relationship between gut microbiota and prostate cancer using a mouse model of prostate cancer.

Animals

Prostate-specific Pten knockout mice [Pb-Cre+; Ptenfl/fl] were used. All mice were weaned at 4 weeks of age and fed control diet (CD: 12.5% energy from fat; MF, Oriental Yeast) or HFD (62.2% energy from fat; HFD-60, Oriental Yeast; Supplementary Table S1). The main source of fat in the HFD was lard (33% of weight), which was largely saturated fat. The HFD was continued until the time of sacrifice. The prostate was extracted with the peripheral fatty tissue removed. All animal procedures were approved by the Osaka University Animal Research Committee and conducted in accordance with relevant regulatory standards.

Cell lines

DU145 was purchased from RIKEN BRC CELL BANK (lot no. 002) in MAY 2009. 22Rv1 was purchased from ATCC (lot no. 58483196) in MAY 2012. Prior to the provision, certification (short tandem repeat analysis) and mycoplasma tests (Hoechst stain) were performed to ensure no Mycoplasma infection or other cell line contamination. Thawed storage cells were used for experiments within 13 weeks. Prior to storage and experimentation, Mycoplasma tests were performed on the cell lines grown in the same culture medium using CycleavePCR Mycoplasma Detection Kit (Takara Bio) and confirmed no infection.

Antibiotics and short-chain fatty acids treatment

For antibiotics experiments, mice were fed with antibiotics in the drinking water from 5 weeks of age. The antibiotics administration consisted of an antibiotic mixture [Abx: 0.2 g/L ampicillin (Wako) + 0.1 g/L vancomycin (Wako) + 0.2 g/L metronidazole (Wako) + 0.2 g/L neomycin (NM; Wako)], 0.035 g/L gentamicin (GM; Wako), or 0.2 g/L NM alone. For short-chain fatty acid (SCFA) supplementation experiments, a mixture of SCFAs [67.5 mmol/L sodium acetate (Wako), 40 mmol/L sodium butyrate (Wako), and 25.9 mmol/L sodium propionate (Nacalai Tesque)] or a sodium matched control was added to the Abx solution (15).

16S rRNA gene sequencing and data processing

Bacterial DNA was extracted from the fecal and tissue samples using a DNeasy Power Soil Kit (Qiagen). 16S rRNA amplicon based metagenomic sequencing was performed using a MiSeq (Illumina). QIIME pipeline (version 1.9.1, RRID: SCR_008249) was used as the bioinformatics environment for processing the raw sequencing data. Linear discriminant analysis effect size (LEfSe, RRID:SCR_014609) of the taxonomic data was performed using the Galaxy web application (https://huttenhower.sph.harvard.edu/galaxy/, RRID:SCR_006281). The reported metagenome data are deposited in DDBJ under the accession number DRA011583, DRA011581, DRA011582, and DRA011628.

Quantitative real-time PCR

A Prime Script RT reagent Kit (Perfect Real Time; Takara Bio) was used to generate cDNA. Quantitative real-time PCR was performed on a Quant Studio 7 Flex System (Applied Biosystems), using a Power Up SYBR Green Master Mix (Applied Biosystems). Gene expression was defined relative to the housekeeping gene, Gapdh, using the delta-delta Ct method. The primers used are indicated in Supplementary Table S2.

IHC analysis

For IHC analysis of nuclear localized antigens, the ratio of positive cells to tumor cells was counted in a 400× field of view. For IHC analysis of cytoplasmic or plasma membrane-localized antigens, the levels of staining were classified into three groups according to the staining intensity of tumor cells in a 100× field of view (staining score, 1 to 3). Following this, the averages of the ratio or score were calculated using five different random fields per sample. The IHC staining using human prostatectomy specimens was approved by the Institutional Review Board of UAB Hospital, and we were allowed to omit the authorization and informed consent (IRB #X140724007). The primary antibodies used are indicated in Supplementary Materials and Methods.

IGF1 measurements

Serum of mice was assayed for insulin-like growth factor-1 (IGF1) using Mouse/Rat IGF-I/IGF1 Duo Set ELISA (R&D Systems) according to the manufacturer's instructions. The absorbance was measured using an iMark Microplate Absorbance Reader (Bio-Rad).

SCFA measurements

SCFA assays of fecal samples were performed by LC/MS-MS. An Acquity UPLC H-Class (Waters) was coupled to a triple quadrupole mass spectrometer, Xevo TQ-S (Waters). The SCFAs were detected in electrospray ionization negative mode. The analytes were monitored in multiple reaction monitoring mode. The identification of SCFAs was carried out by referring to the results of standards derivatized by the same procedure as the samples.

Statistical analysis

Data are expressed as mean ± SE. If the data followed a normal distribution in the Shapiro–Wilk test, comparisons were made with an unpaired t test, and if the data are non-normally distributed, comparisons were made with a Mann–Whitney U test. χ2 test was used to test the qualitative variables in patient characteristics. The Bonferroni correction was used for multiple comparison tests. P < 0.05 was considered significant. All analyses were performed by JMP Pro 14 (SAS Institute, RRID:SCR_008567).

More details can be found in Supplementary Materials and Methods.

Antibiotics inhibit tumor growth in the prostate cancer mouse models

To investigate the effect of intestinal bacteria on prostate cancer, the effects of altering the gut microbiota by changing the diet and administering Abx on tumors in prostate-specific Pten knockout mouse model of prostate cancer were evaluated (Fig. 1A). As in our previous report (3), the HFD-fed prostate cancer mouse models (HFD prostate cancer mice) were obese (Fig. 1B and C) and had significantly higher prostate weights than PCa mouse models fed a control diet (CD prostate cancer mice), at 22 weeks of age (P = 0.012; Fig. 1D). In addition, Abx administration slightly reduced prostate weights in CD prostate cancer mice (CD + Abx prostate cancer mice; P = 0.110; Fig. 1E), but significantly reduced prostate weights in HFD prostate cancer mice (HFD + Abx prostate cancer mice; P = 0.015; Fig. 1E). There was no significant interaction between diet and Abx on prostate weight (Pinteraction = 0.100). Abx administration tended to increase the body weight in CD prostate cancer mice (P = 0.065; Fig. 1F) and significantly increased in HFD prostate cancer mice (P = 0.010; Fig. 1F). Prostate weight per body weight was significantly decreased by Abx administration in both HFD prostate cancer mice (11.4 × 10−3 vs. 7.0 × 10−3 g/g, P = 0.003) and CD prostate cancer mice (8.9 × 10−3 vs. 6.8 × 10−3 g/g, P = 0.022). Macroscopically, compared with the prostate tissues in either CD or CD + Abx prostate cancer mice, the prostate tissue in the HFD prostate cancer mice was larger, and Abx administration of HFD prostate cancer mice resulted in a smaller prostate (Fig. 1G). Hematoxylin and eosin histologic staining of prostate showed cancer was observed in all prostate cancer mice regardless of diet or antibiotic status, and that prostate was not enlarged by adipose or stromal tissue in HFD mice (Fig. 1H). The proliferative potential of prostate cancer tumor cells was evaluated by Ki67 staining (Fig. 1I and J). The ratio of Ki67-positive cells to total tumor cells was not changed by Abx in CD prostate cancer mice (P = 0.800), whereas it was significantly decreased in HFD prostate cancer mice (P < 0.001). In wild-type (WT) mice, Abx administration did not have a significant effect on prostate tissue (Supplementary Fig. S1A). Serum testosterone levels were not significantly altered by Abx administration in both CD and HFD WT mice (P = 0.895 and 0.768, Supplementary Fig. S1B). Oral administration of GM, which is not absorbed from the intestine and therefore selectively targets intestinal bacteria, to HFD prostate cancer mice inhibited prostate growth (P = 0.045; Supplementary Fig. S1C) as did administration of Abx. GM administration did not alter the body weight in HFD prostate cancer mice (P = 0.304; Supplementary Fig. S1D). These results suggested that antibiotics suppress cancer proliferation, particularly in HFD-fed mice. This suppression of prostate cancer was not due to an improvement in obesity, because there was no weight loss with antibiotics in mice.

HFD and antibiotics alter the composition of the gut microbiota in mice

Bacterial DNA from fecal samples from the four different prostate cancer mouse model groups (CD, CD + Abx, HFD, HFD + Abx) were subjected to metagenomic analysis based on 16S rRNA sequences. The groups had different gut microbiota compositions from each other (Fig. 2A; Supplementary Fig. S2A), and a principal coordinate analysis (PCoA) revealed that the HFD + Abx group was notably different from the other three groups (Fig. 2B). Comparing the gut microbiota in HFD and HFD + Abx mice, 49 bacteria were found to differ significantly in their relative abundance (Fig. 2C). In a rarefaction analysis, both phylogenic diversity and Shannon index were used to evaluate the diversity of the gut microbiota in these four groups, and the diversity of the HFD + Abx group was found to be less than that in the HFD group (Supplementary Fig. S2B). No obvious differences in the composition of gut microbiota between the prostate cancer mice and WT mice were found in the PCoA or rarefaction analysis (Supplementary Fig. S2C–S2E). The gut microbiota composition was similar for HFD-fed prostate cancer mice administered the aminoglycoside antibiotics GM and NM (Fig. 2D; Supplementary Fig. S2F and S2G). However, prostate weight was significantly higher in the NM-administered prostate cancer mice (P = 0.045) with no difference in body weight between the groups (P = 0.836; Fig. 2E and F). Prostate weight per body weight was also significantly higher in NM administered prostate cancer mice compared with that of GM-administered prostate cancer mice (11.8 × 10−3 vs. 7.4 × 10−3 g/g, P = 0.031). In the comparison of bacterial abundance between GM- and NM-administered HFD mice, 24 bacteria were found to be altered significantly. Among these, the relative abundances of Rikenellaceae and Clostridiales were increased in the prostate cancer progression group, in common with the comparison between HFD and HFD + Abx mice (Fig. 2C and G). The microbiota in the prostate tissue of HFD prostate cancer mice was not obviously altered by Abx administration (Supplementary Fig. S2H–S2J). Collectively, both HFD and Abx administration affect the composition of intestinal bacteria, and the combination of the two has a particularly strong influence. Moreover, prostate cancer itself has no effect on the composition of the gut microbiota. These results indicate that specific changes in the gut microbiota composition induced by a combination of HFD and Abx could be associated with inhibition of prostate cancer growth.

Abx reduces systemic IGF1 produced by the liver and local IGF1 produced by prostate cancer cells

To investigate the effect of Abx on prostate tissue in HFD prostate cancer mice, we compared the prostate gene expression levels of HFD and HFD + Abx prostate cancer mice using a cDNA microarray analysis. As a result, the expression of 64 coding RNAs was significantly upregulated in HFD prostate cancer mice compared to HFD + Abx prostate cancer mice (Fold change ≥ 1.5; Supplementary Table S3), with Igf1 being one of the upregulated genes (Supplementary Fig. S3A). IGF1 is a growth factor that has functions such as promoting cell proliferation, and its activity is inhibited by binding to IGF binding protein 3, which is produced in the liver (16). We next investigated the expression of IGF1–related genes in the prostate of the different prostate cancer mice as an assessment of local IGF1 activity, using quantitative real-time PCR. The expression of Igf1 was significantly downregulated in HFD + Abx prostate cancer mice compared with HFD prostate cancer mice (P < 0.001; Fig. 3A). As an assessment of systemic IGF1 activity, we investigated the expression levels of IGF1-related genes in the liver, the main organ responsible for the production of circulating IGF1, as well as the serum levels of IGF1. As a result, the expression of Igf1 was significantly downregulated, and Igfbp3 was upregulated in the liver from HFD + Abx prostate cancer mice compared with HFD prostate cancer mice (P = 0.032 and 0.017, respectively; Fig. 3B). The serum IGF1 levels were significantly decreased by Abx administration in HFD prostate cancer mice (P = 0.043; Fig. 3C), and also reduced by Abx administration in HFD WT mice (P = 0.005; Fig. 3D). In CD and CD + Abx prostate cancer mice, there were no significant changes in the expression levels of IGF1–related genes in the prostate and liver (Fig. 3E and F). Abx did not cause significant changes in serum IGF1 levels in CD prostate cancer mice and CD WT mice, (P = 0.962 and 0.078, respectively; Fig. 3G and H). In normal prostate from WT mice, there were no changes caused by Abx in the expression levels of IGF1–related genes (Supplementary Fig. S3B), regardless of whether the mice were fed CD or HFD. An IHC analysis showed that prostate cancer cells and hepatocytes were both positive for IGF1 (Fig. 3I; Supplementary Fig. S3C). These data suggest that Abx leads to the downregulation of IGF1 in the prostate of HFD prostate cancer mice. Furthermore, this reduction is found to occur systemically and is not necessarily restricted to the prostate.

Reduction of IGF1 by Abx suppresses key signaling pathways in prostate cancer and is involved in prostate cancer proliferation in vivo and in vitro

The binding of IGF1 to the IGF1 receptor (IGF1R), has been shown to induce cell proliferation by activating PI3K and MAPK signaling pathways (17, 18). We therefore evaluated activation of the IGF1R, PI3K, and MAPK pathways via IHC analysis in prostate from prostate cancer mice. Activation of the IGF1R was assessed by measuring IGF1R phosphorylation (p-IGF1R). The activity of PI3K was assessed by measuring the phosphorylation of protein kinase B (p-AKT), and the activity of MAPK was assessed by measuring the phosphorylation of extracellular signal-regulated kinases (pERK). Although the loss of phosphatase and tensin homolog deleted from chromosome 10 (PTEN) in the prostate of our mouse model promotes the activity of PI3K–AKT pathway, we investigated the activity of this pathway, which a variation in upstream IGF1 could affect (19). Abx administration significantly reduced p-IGF1R (P < 0.001; Fig. 4A), p-AKT (P < 0.001; Fig. 4B), and pERK (P < 0.001; Fig. 4C) levels in cancer cells from HFD prostate cancer mice (Fig. 4D). In contrast, in CD prostate cancer mice, the levels of p-IGF1R (P = 0.689), p-AKT (P = 0.916), and pERK (P = 0.143) were unchanged following Abx administration (Fig. 4AD). IHC analysis of mouse prostate tumors showed reduced levels of phosphorylated MAPK/ERK kinase (MEK) and S6 ribosomal protein, which are additional proteins, respectively, related to MAPK and PI3K signaling, in HFD + Abx prostate cancer mice (Supplementary Fig. S4A–S4C). There was no effect of Abx on the IGF1 signaling pathway in CD prostate cancer mice, whereas Abx had a clear suppressive effect on IGF1 signaling in HFD prostate cancer mice. Taken together, these results suggest that in our prostate cancer mouse model, the IGF1 reduction induced by HFD + Abx downregulates IGF1R-mediated MAPK and PI3K signaling pathways that play a key role in the prostate cancer proliferation. Using an in vitro cell proliferation assay, we found that the addition of recombinant IGF1 significantly promoted proliferation in DU145 and 22Rv1 cells (Fig. 4E). Furthermore, we confirmed that adding IGF1 to DU145 cells significantly promoted the phosphorylation of IGF1R, ERK, and AKT, in a concentration-dependent manner (Fig. 4F and G). In 22Rv1 cells, IGF1 treatment only promoted the phosphorylation of IGF1R and AKT, while the phosphorylation of ERK was unchanged (Fig. 4F and G). These results suggest that IGF1 directly promotes the prostate cancer proliferation by activating its receptor present on cancer cells, subsequently leading to the activation of downstream pathways, including MAPK and PI3K.

Abx administration decreases SCFA levels in feces, leading to the downregulation of IGF1 and inhibition of cancer proliferation

We investigated the reason why the gut microbiota in HFD + Abx prostate cancer mice reduces IGF1 levels, thereby suppressing prostate cancer growth. SCFAs, a class of major bacterial metabolites, affect the IGF1 production and have been shown to be involved in the growth of the body and bone (20). Majority of the SCFAs produced by intestinal bacteria are acetate, propionate, and butyrate. SCFAs are known to be absorbed by the host and have various effects on the intestinal tract and the whole body (21, 22). The relative abundances of multiple SCFA-producing bacteria, such as Rikenellaceae and Clostridiales, were decreased in the HFD + Abx prostate cancer mice (Fig. 2C; refs. 23, 24). The levels of all fecal SCFAs were markedly decreased in HFD + Abx prostate cancer mice (Fig. 5A). In CD prostate cancer mice, Abx administration significantly reduced butyrate and isobutyrate, but not acetate or propionate (Fig. 5B). To investigate the effects of SCFAs on IGF1 production and prostate cancer proliferation, HFD + Abx mice, which have decreased bacterial-derived SCFAs, were administered a mixture of SCFAs orally. Supplementation of SCFAs to HFD + Abx WT mice significantly increased serum IGF1 levels independent of prostate cancer (P = 0.023; Fig. 5C and D). Supplementation of SCFAs to HFD + Abx prostate cancer mice significantly increased prostate weights at 22weeks of age (P = 0.022; Fig. 5E and F), despite no significant changes in body weight (P = 0.069; Fig. 5G). SCFAs supplementation significantly increased prostate weight per body weight (15.9 × 10−3 vs. 8.2 × 10−3 g/g, P = 0.022) and the ratio of Ki67-positive cells to total tumor cells (P = 0.022; Fig. 5H and I). These results suggest that Abx reduces the production of SCFAs in the gut microbiota of HFD mice, and that this reduction in SCFAs leads to a decrease in IGF1 production and inhibition of prostate cancer proliferation.

Obese patients with prostate cancer have higher expression levels of IGF1 in prostate cancer cells

Obesity or HFD leads to hyperinsulinemia, which in turn increases IGF1 production (25, 26). We performed IHC analysis for detecting IGF1 in prostate sections obtained from nonobese [body mass index (BMI) < 26] and severely obese (BMI ≥ 35) patients with prostate cancer, to investigate whether obesity results in higher local prostate IGF1 production in human. The background characteristics are summarized (Table 1). IGF1 staining of the prostate was mainly positive for cancer cells (Fig. 6A). The IGF1 staining scores in the prostate cancer cells were significantly higher in severely obese patients (P = 0.003; Fig. 6B). These results suggest that gut microbiota and IGF1-mediated mechanism of prostate cancer progression may exist in humans as well as in our prostate cancer mouse model.

In this study, we have shown that HFD mainly consisting of saturated fatty acids induces obesity, changes the gut microbiota composition, and promotes cancer growth in a mouse model of prostate cancer. We have demonstrated that Abx administration inhibits cancer growth. Abx markedly alters the composition of the gut microbiota in HFD-fed mice, leading to the downregulation of prostate and systemic IGF1 levels. Furthermore, the activities of the IGF1R and its downstream signaling pathways, PI3K and MAPK, were suppressed in HFD + Abx prostate cancer mice. The levels of SCFAs were markedly reduced in the feces of HFD + Abx mice, and supplementation of SCFAs increased IGF1 production and promoted prostate cancer growth. These results suggest that the SCFAs produced by intestinal bacteria are involved in the production of host IGF1 and have an effect on the prostate cancer proliferation via IGF1 signaling pathways.

This inhibitory effect was confirmed with the oral ingestion of Abx as well as GM, which is not absorbed systemically through the intestinal tract, suggesting that the inhibition of prostate cancer growth is unlikely to be a direct effect of Abx on the prostate. The resultant obesity might be the main cause of tumor growth induced by the HFD. However, Abx administration further increased the body weight of HFD-fed mice, while the tumor weight was suppressed. This might suggest that the HFD is the actual cause of tumor growth. It has been reported that there is a difference in the microbiota between tumor and nontumor specimens of human prostate tissue, suggesting that local microbiota may be involved in carcinogenesis in the prostate in humans (27). Therefore, the suppression of prostate cancer growth is not likely to be the result of a direct effect on the local prostate microbiota. Instead, all of the data show that the antibiotic-influenced gut microbiota indirectly inhibits the prostate cancer growth, and strongly suggest the existence of gut microbiota-mediated mechanisms to regulate prostate cancer development.

IGF1 is one of the growth factors and associated with intestinal bacteria (20, 28). IGF1 is produced not only in the liver but also in various tissues, including cancer cells, and that IGF1 acts locally through autocrine or paracrine signaling (20, 29). In this regard, prostate cancer tumors have also been shown to produce local IGF1 in both in vitro and in vivo studies (30, 31). In humans, IGF1 is known to increase the risk of prostate cancer. Blood IGF1 levels are correlated with prostate cancer risk and the incidence of prostate cancer is increased in acromegaly patients who have increased growth hormone and IGF1 production (32, 33).

As further evidence for an association between intestinal bacteria and IGF1, systemic IGF1 levels are decreased in germ-free mice, resulting in an inhibition of normal body development (20, 34). We found that Abx administration to the HFD prostate cancer mice suppressed not only prostate but also hepatic Igf1 expression as well as circulating IGF1 levels. The reduction in prostate IGF1 levels is most likely due to decreased production by prostate cancer cells as this reduction was observed only in prostate cancer mice and not in WT mice, and as prostate cancer cells were found to be positive for IGF1 via IHC analysis. Because there was a difference in Igf1 expression in hepatic tissue when comparing HFD and HFD + Abx mice, the decrease in circulating IGF1 levels seen in response to Abx administration could be due to decreased production by the liver. In addition, the expression of Igfbp3 was increased in the liver; this would further serve to reduce the systemic activity of IGF1. Because IGF1 is secreted by prostate cancer cells, the decreased levels of IGF1 in HFD + Abx prostate cancer mice might simply reflect a secondary effect of decreased tumor sizes. However, the finding of the Abx-induced decrease in IGF1 levels in cancer-free hepatic tissue and serum imply that changes in IGF1 levels in HFD prostate cancer mice are a primary driver of tumorigenesis rather than a secondary effect correlated with tumor sizes.

Although it has been reported that HFD is associated with the production of IGF1 and that IGF1 affects prostate cancer growth (25, 35, 36), our study is the first to show that IGF1 production altered by the gut microbiota is important in prostate cancer proliferation. In keeping with this, we found that IGF1 directly promotes the proliferation of prostate cancer cell lines via MAPK and PI3K pathways, in vitro. Both in vitro and in vivo studies have shown that activation of the upstream pathways of PI3K, such as IGF1, further promotes AKT phosphorylation, even in cells with PTEN loss (37, 38). Accordingly, despite enhanced PI3K pathway activation due to the loss of PTEN in our prostate cancer model, it is likely that IGF1 would cause further activation of both the MAPK and PI3K pathways. It should be noted that changes in the IGF1 production induced by the gut microbiota may affect other pathologies, such as other cancer types, because it acts not only locally in the prostate but also systemically. However, the IGF1R is often found to be overexpressed in prostate cancer, and therefore, prostate cancer may be particularly susceptible to changes in IGF1 levels (39). Although IGF1 may be a beneficial hormone for body or bone growth, it may be a harmful factor in terms of prostate cancer growth promotion. We found that a gut microbiota-mediated IGF1 increase resulted in tumor growth in Pten knockout mice. PTEN mutation is one the most common mutations in human prostate cancer; however, this finding should be confirmed in other prostate cancer models.

SCFAs are produced by the fermentation of dietary fiber by intestinal bacteria. The relative abundances of Rikenellaceae and Clostridiales, which were increased in prostate cancer progression groups, are positively correlated with fecal SCFA levels (23, 40). Consistently, our results showed that Abx administration to HFD-fed mice resulted in a substantial decrease in SCFA-producing bacteria and in fecal SCFA levels. In addition, oral supplementation of SCFAs to HFD + Abx mice reversed the antibiotic-induced reduction of IGF1 levels and suppression of prostate cancer growth. Taken together, it is likely that multiple intestinal bacteria such as Rikenellaceae and Clostridiales might be associated with cooperative regulation of prostate cancer growth via the IGF1 pathway due to the production of SCFAs. Further studies of transplanting these bacteria to prostate cancer model mice should be performed to test this hypothesis. SCFAs have anti-inflammatory effects and can suppress intestinal carcinogenesis (22, 41). In addition, butyrate has function as a histone deacetylase inhibitor, which has a inhibitory effect on the proliferation of many types of cancer, including prostate cancer (42). However, we suspect SCFAs are deleterious in prostate cancer due to the increased IGF1 levels. In support of this, low concentrations of butyrate have been show to promote cell proliferation in the colon, although high concentrations of butyrate inhibit histone deacetylase (43, 44). In this study, since much of the SCFAs absorbed from the gut are consumed in the liver (22), butyrate and other types of SCFAs are likely to be at low concentrations in the prostate and would not inhibit prostate cancer growth, resulting in an enhancement of proliferation via the IGF1 signaling pathway.

This study has some limitations. First, we did not evaluate the cancer condition separately for each lobe of the prostate tissue, because it was difficult to distinguish each lobe in some mice with particularly advanced prostate cancer. As the cancer condition is different for each anatomical region in the Pten knockout mouse model, separate evaluation might help to more accurately assess the relationship between prostate cancer and gut microbiota (45). Second, we could not assess the effect of inhibiting IGF1 in the prostate cancer mouse model, because the experimental model takes a long time to establish, and it is difficult to administer IGF1 inhibitors to mice for such a long period of time, because IGF1 also affects the normal growth of the body. Third, we could not clarify the mechanism by which SCFAs elevate IGF1 levels. Unfortunately, the SCFA receptors involved in IGF1 production have not been identified (46). Therefore, it is unclear if SCFA act directly on prostate cancer cells and increase local IGF1 production. However, since our experiments showed a consistent change in IGF1 levels by Abx in the prostate cancer, liver, and serum, SCFAs might directly affect prostate cancer cells. Fourth, the activation of the IGF1 pathway by bacterial SCFAs may be only one of several bacteria-mediated mechanisms for increasing prostate cancer growth. For example, HFD leads to increased gut permeability, also known as leaky gut; therefore, bacterial metabolites and components could leak into the blood and result in a systemic response, such as inflammation and insulin resistance (47). The effects of intestinal bacteria on inflammation have been linked with several types of cancer. In our previous study, we found that proinflammatory cytokines induced prostate cancer growth by HFD (3). There are also numerous other reports showing that obesity is involved in the growth of prostate cancer via inflammatory and immune-mediated mechanisms (48–50). Inflammation-associated prostate cancer growth due to obesity could be mediated by gut microbiota. In this study, a gut microbiota-mediated mechanism on prostate cancer growth was demonstrated in HFD-fed obese mice; however, we also found that Abx might suppress tumor growth even in CD-fed mice. Although the cause of possible Abx-induced suppression of prostate cancer growth in CD mice has not been clarified, other intestinal bacteria-mediated mechanisms of prostate cancer growth on CD mice might exist. Moreover, we did not directly compare CD- and HFD-fed mice for this gut microbiota-mediated mechanism due to the limited number of mice available in this study. Therefore, while the results of our animal experiments certainly implicate that bacterial SCFAs play an important role in prostate cancer growth, it is still unclear whether bacterial production of SCFA is responsible for HFD-induced acceleration of prostate cancer growth. This is because other mechanisms, such as leaky gut, are also possible contributing factors. Finally, this bacterial mechanism has not been validated in humans. The gut microbiota composition in humans and mice varies; therefore, future human studies are needed to determine whether the same mechanism exists in both species.

In conclusion, IGF1 production regulated by SCFAs from gut microbiota affects prostate cancre growth via local MAPK and PI3K signaling pathways, indicating the existence of a “gut microbiota–IGF1–prostate axis”. In the future, it may be possible to reduce the risk or inhibit the progression of prostate cancer by intervention in the gut microbiota. Obese patients may be more likely to benefit from this new treatment. Gut microbiota with no ability to produce SCFAs may reduce the risk of prostate cancer.

K. Takeda reports grants from JSPS and AMED outside the submitted work. N. Nonomura reports grants and personal fees from Takeda Pharmaceutical, AstraZeneca, Nihon Shinyaku, personal fees from Jansen Pharma, Astellas, and Beyer outside the submitted work. No disclosures were reported by the other authors.

M. Matsushita: Data curation, investigation, visualization, methodology, writing–original draft. K. Fujita: Conceptualization, supervision, funding acquisition, writing–review and editing. T. Hayashi: Supervision, funding acquisition, investigation, methodology. H. Kayama: Supervision. D. Motooka: Data curation, investigation. H. Hase: Data curation, investigation. K. Jingushi: Data curation, investigation. G. Yamamichi: Investigation. S. Yumiba: Investigation. E. Tomiyama: Investigation. Y. Koh: Investigation. Y. Hayashi: Investigation. K. Nakano: Investigation. C. Wang: Investigation. Y. Ishizuya: Investigation. T. Kato: Supervision. K. Hatano: Supervision. A. Kawashima: Supervision. T. Ujike: Supervision. M. Uemura: Supervision. R. Imamura: Supervision. M.D.C. Rodriguez Pena: Resources, supervision. J.B. Gordetsky: Resources, supervision. G.J. Netto: Resources, supervision. K. Tsujikawa: Supervision. S. Nakamura: Conceptualization, supervision. K. Takeda: Conceptualization, supervision. N. Nonomura: Conceptualization, supervision.

The authors thank Tak W. Mak (Medical Biophysics and Immunology, University of Toronto; Campbell Family Institute for Breast Cancer Research, Princess Margaret Cancer Centre), Akira Suzuki (Department of Molecular Genetics, Division of Cancer Genetics, Kobe University Graduate School of Medicine), and Toru Nakano (Department of Pathology, Medical School and Graduate School of Frontier Biosciences, Osaka University) for offering the [Pten (fl/fl)] mice, and Aysha Mubeen (Department of Pathology, University of Alabama at Birmingham) for collecting information on prostate cancer patients. This research was partially supported by Platform Project for Supporting Drug Discovery and Life Science Research [Basis for Supporting Innovative Drug Discovery and Life Science Research (BINDS)] from AMED under grant number JP21am0101084 (support number 2675). K. Fujita was supported by research grants of The Japanese Urological Association, Yakult Bio-Science Foundation, Project MEET of Osaka University Graduate School of Medicine. T. Hayashi was supported by JSPS KAKENHI grant number JP18K16693.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Bray
F
,
Ferlay
J
,
Soerjomataram
I
,
Siegel
RL
,
Torre
LA
,
Jemal
A
. 
Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries
.
CA Cancer J Clin
2018
;
68
:
394
424
.
2.
Matsushita
M
,
Fujita
K
,
Nonomura
N
. 
Influence of diet and nutrition on prostate cancer
.
Int J Mol Sci
2020
;
21
:
1447
.
3.
Hayashi
T
,
Fujita
K
,
Nojima
S
,
Hayashi
Y
,
Nakano
K
,
Ishizuya
Y
, et al
High-fat diet-induced inflammation accelerates prostate cancer growth via IL6 signaling
.
Clin Cancer Res
2018
;
24
:
4309
18
.
4.
Chang
SN
,
Han
J
,
Abdelkader
TS
,
Kim
TH
,
Lee
JM
,
Song
J
, et al
High animal fat intake enhances prostate cancer progression and reduces glutathione peroxidase 3 expression in early stages of TRAMP mice
.
Prostate
2014
;
74
:
1266
77
.
5.
Akinsete
JA
,
Ion
G
,
Witte
TR
,
Hardman
WE
. 
Consumption of high ω-3 fatty acid diet suppressed prostate tumorigenesis in C3(1) Tag mice
.
Carcinogenesis
2012
;
33
:
140
8
.
6.
Amabebe
E
,
Robert
FO
,
Agbalalah
T
,
Orubu
ESF
. 
Microbial dysbiosis-induced obesity: role of gut microbiota in homoeostasis of energy metabolism
.
Br J Nutr
2020
;
123
:
1127
37
.
7.
Maeda
Y
,
Kurakawa
T
,
Umemoto
E
,
Motooka
D
,
Ito
Y
,
Gotoh
K
, et al
Dysbiosis contributes to arthritis development via activation of autoreactive T cells in the intestine
.
Arthritis Rheumatol
2016
;
68
:
2646
61
.
8.
Ishikawa
D
,
Sasaki
T
,
Osada
T
,
Kuwahara-Arai
K
,
Haga
K
,
Shibuya
T
, et al
Changes in intestinal microbiota following combination therapy with fecal microbial transplantation and antibiotics for ulcerative colitis
.
Inflamm Bowel Dis
2017
;
23
:
116
25
.
9.
Minter
MR
,
Zhang
C
,
Leone
V
,
Ringus
DL
,
Zhang
X
,
Oyler-Castrillo
P
, et al
Antibiotic-induced perturbations in gut microbial diversity influences neuro-inflammation and amyloidosis in a murine model of Alzheimer's disease
.
Sci Rep
2016
;
6
:
30028
.
10.
Schulz
MD
,
Atay
Ç
,
Heringer
J
,
Romrig
FK
,
Schwitalla
S
,
Aydin
B
, et al
High-fat-diet-mediated dysbiosis promotes intestinal carcinogenesis independently of obesity
.
Nature
2014
;
514
:
508
12
.
11.
Yachida
S
,
Mizutani
S
,
Shiroma
H
,
Shiba
S
,
Nakajima
T
,
Sakamoto
T
, et al
Metagenomic and metabolomic analyses reveal distinct stage-specific phenotypes of the gut microbiota in colorectal cancer
.
Nat Med
2019
;
25
:
968
76
.
12.
Loo
TM
,
Kamachi
F
,
Watanabe
Y
,
Yoshimoto
S
,
Kanda
H
,
Arai
Y
, et al
Gut microbiota promotes obesity-associated liver cancer through pge2-mediated suppression of antitumor immunity
.
Cancer Discov
2017
;
7
:
522
38
.
13.
Dzutsev
A
,
Badger
JH
,
Perez-Chanona
E
,
Roy
S
,
Salcedo
R
,
Smith
CK
, et al
Microbes and cancer
.
Annu Rev Immunol
2017
;
35
:
199
228
.
14.
Liss
MA
,
White
JR
,
Goros
M
,
Gelfond
J
,
Leach
R
,
Johnson-Pais
T
, et al
Metabolic biosynthesis pathways identified from fecal microbiome associated with prostate cancer
.
Eur Urol
2018
;
74
:
575
82
.
15.
Smith
PM
,
Howitt
MR
,
Panikov
N
,
Michaud
M
,
Gallini
CA
,
Bohlooly-Y
M
, et al
The microbial metabolites, short-chain fatty acids, regulate colonic T reg cell homeostasis
.
Science
2013
;
341
:
569
73
.
16.
Kineman
RD
,
del Rio-Moreno
M
,
Sarmento-Cabral
A
. 
40 years of IGF1: Understanding the tissue-specific roles of IGF1/IGF1R in regulating metabolism using the Cre/loxP system
.
J Mol Endocrinol
2018
;
61
:
T187
98
.
17.
Lee
Y
,
Wang
Y
,
James
M
,
Jeong
JH
,
You
M
. 
Inhibition of IGF1R signaling abrogates resistance to afatinib (BIBW2992) in EGFR T790M mutant lung cancer cells
.
Mol Carcinog
2016
;
55
:
991
1001
.
18.
Hakuno
F
,
Takahashi
SI
. 
40 years of IGF1: IGF1 receptor signaling pathways
.
J Mol Endocrinol
2018
;
61
:
T69
86
.
19.
Molinari
F
,
Frattini
M
. 
Functions and regulation of the PTEN gene in colorectal cancer
.
Front Oncol
2014
;
3
:
326
.
20.
Yan
J
,
Herzog
JW
,
Tsang
K
,
Brennan
CA
,
Bower
MA
,
Garrett
WS
, et al
Gut microbiota induce IGF-1 and promote bone formation and growth
.
Proc Natl Acad Sci U S A
2016
;
113
:
E7554
63
.
21.
Cook
SI
,
Sellin
JH
. 
Review article: short chain fatty acids in health and disease
.
Aliment Pharmacol Therm
1998
;
12
:
499
507
.
22.
Morrison
DJ
,
Preston
T
. 
Formation of short chain fatty acids by the gut microbiota and their impact on human metabolism
.
Gut Microbes
2016
;
7
:
189
200
.
23.
Wang
B
,
Kong
Q
,
Li
X
,
Zhao
J
,
Zhang
H
,
Chen
W
, et al
A high-fat diet increases gut microbiota biodiversity and energy expenditure due to nutrient difference
.
Nutrients
2020
;
12
:
3197
.
24.
Chai
LJ
,
Xu
PX
,
Qian
W
,
Zhang
XJ
,
Ma
J
,
Lu
ZM
, et al
Profiling the clostridia with butyrate-producing potential in the mud of Chinese liquor fermentation cellar
.
Int J Food Microbiol
2019
;
297
:
41
50
.
25.
Barnard
RJ
,
Ngo
TH
,
Leung
PS
,
Aronson
WJ
,
Golding
LA
. 
A low-fat diet and/or strenuous exercise alters the IGF axis in vivo and reduces prostate tumor cell growth in vitro
.
Prostate
2003
;
56
:
201
6
.
26.
Calle
EE
,
Kaaks
R
. 
Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms
.
Nat Rev Cancer
2004
;
4
:
579
91
.
27.
Cavarretta
I
,
Ferrarese
R
,
Cazzaniga
W
,
Saita
D
,
Lucianò
R
,
Ceresola
ER
, et al
The microbiome of the prostate tumor microenvironment
.
Eur Urol
2017
;
72
:
625
31
.
28.
Poinsot
P
,
Schwarzer
M
,
Peretti
N
,
Leulier
F
. 
40 years of IGF1: The emerging connections between IGF1, the intestinal microbiome, lactobacillus strains and bone growth
.
J Mol Endocrinol
2018
;
61
:
T103
13
.
29.
Perry
JK
,
Emerald
BS
,
Mertani
HC
,
Lobie
PE
. 
The oncogenic potential of growth hormone
.
Growth Horm IGF Res
2006
;
16
:
277
89
.
30.
Ohlson
N
,
Bergh
A
,
Stattin
P
,
Wikström
P
. 
Castration-induced epithelial cell death in human prostate tissue is related to locally reduced IGF-1 levels
.
Prostate
2007
;
67
:
32
40
.
31.
McKeown
BT
,
McDougall
L
,
Catalli
A
,
Hurta
RAR
. 
Magnolol causes alterations in the cell cycle in androgen insensitive human prostate cancer cells in vitro by affecting expression of key cell cycle regulatory proteins
.
Nutr Cancer
2014
;
66
:
1154
64
.
32.
Watts
EL
,
Goldacre
R
,
Key
TJ
,
Allen
NE
,
Travis
RC
,
Perez-Cornago
A
. 
Hormone-related diseases and prostate cancer: an English national record linkage study
.
Int J Cancer
2020
;
147
:
803
10
.
33.
Roddam
AW
,
Allen
NE
,
Appleby
P
,
Key
TJ
,
Ferrucci
L
,
Carter
HB
, et al
Insulin-like growth factors, their binding proteins, and prostate cancer risk: analysis of individual patient data from 12 prospective studies
.
Ann Intern Med
2008
;
149
:
461
71
.
34.
Schwarzer
M
,
Makki
K
,
Storelli
G
,
Machuca-Gayet
I
,
Srutkova
D
,
Hermanova
P
, et al
Lactobacillus plantarum strain maintains growth of infant mice during chronic undernutrition
.
Science
2016
;
351
:
854
7
.
35.
Narita
S
,
Tsuchiya
N
,
Saito
M
,
Inoue
T
,
Kumazawa
T
,
Yuasa
T
, et al
Candidate genes involved in enhanced growth of human prostate cancer under high fat feeding identified by microarray analysis
.
Prostate
2008
;
68
:
321
35
.
36.
Ngo
TH
,
Barnard
RJ
,
Leung
PS
,
Cohen
P
,
Aronson
WJ
. 
Insulin-like growth factor I (IGF-I) and IGF binding protein-1 modulate prostate cancer cell growth and apoptosis: possible mediators for the effects of diet and exercise on cancer cell survival
.
Endocrinology
2003
;
144
:
2319
24
.
37.
Rodriguez
M
,
Siwko
S
,
Zeng
L
,
Li
J
,
Yi
Z
,
Liu
M
. 
Prostate-specific G-protein-coupled receptor collaborates with loss of PTEN to promote prostate cancer progression
.
Oncogene
2016
;
35
:
1153
62
.
38.
Qin
L
,
Wang
Y
,
Tao
L
,
Wang
Z
. 
AKT down-regulates insulin-like growth factor-1 receptor as a negative feedback
.
J Biochem
2011
;
150
:
151
6
.
39.
Krueckl
SL
,
Sikes
RA
,
Edlund
NM
,
Bell
RH
,
Hurtado-Coll
A
,
Fazli
L
, et al
Increased insulin-like growth factor I receptor expression and signaling are components of androgen-independent progression in a lineage-derived prostate cancer progression model
.
Cancer Res
2004
;
64
:
8620
9
.
40.
Allen
JM
,
Mailing
LJ
,
Niemiro
GM
,
Moore
R
,
Cook
MD
,
White
BA
, et al
Exercise alters gut microbiota composition and function in lean and obese humans
.
Med Sci Sports Exerc
2018
;
50
:
747
57
.
41.
Singh
N
,
Gurav
A
,
Sivaprakasam
S
,
Brady
E
,
Padia
R
,
Shi
H
, et al
Activation of Gpr109a, receptor for niacin and the commensal metabolite butyrate, suppresses colonic inflammation and carcinogenesis
.
Immunity
2014
;
40
:
128
39
.
42.
Kim
J
,
Park
H
,
Ji
YI
,
Wahn
SC
,
Hyung
SK
. 
Sodium butyrate regulates androgen receptor expression and cell cycle arrest in human prostate cancer cells
.
Anticancer Res
2007
;
27
:
3285
92
.
43.
Donohoe
DR
,
Collins
LB
,
Wali
A
,
Bigler
R
,
Sun
W
,
Bultman
SJ
. 
The Warburg effect dictates the mechanism of butyrate-mediated histone acetylation and cell proliferation
.
Mol Cell
2012
;
48
:
612
26
.
44.
Belcheva
A
,
Irrazabal
T
,
Robertson
SJ
,
Streutker
C
,
Maughan
H
,
Rubino
S
, et al
Gut microbial metabolism drives transformation of msh2-deficient colon epithelial cells
.
Cell
2014
;
158
:
288
99
.
45.
Berquin
IM
,
Min
Y
,
Wu
R
,
Wu
J
,
Perry
D
,
Cline
JM
, et al
Modulation of prostate cancer genetic risk by omega-3 and omega-6 fatty acids
.
J Clin Invest
2007
;
117
:
1866
75
.
46.
Yan
J
,
Charles
JF
. 
Gut microbiota and IGF-1
.
Calcif Tissue Int
2018
;
102
:
406
14
.
47.
Garidou
L
,
Pomié
C
,
Klopp
P
,
Waget
A
,
Charpentier
J
,
Aloulou
M
, et al
The gut microbiota regulates intestinal CD4 T cells expressing RORγt and controls metabolic disease
.
Cell Metab
2015
;
22
:
100
12
.
48.
Hatano
K
,
Fujita
K
,
Nonomura
N
. 
Application of anti-inflammatory agents in prostate cancer
.
J Clin Med
2020
;
9
:
2680
.
49.
Hayashi
T
,
Fujita
K
,
Matsushita
M
,
Nonomura
N
. 
Main inflammatory cells and potentials of anti-inflammatory agents in prostate cancer
.
Cancers
2019
;
11
:
1153
.
50.
Fujita
K
,
Hayashi
T
,
Matsushita
M
,
Uemura
M
,
Nonomura
N
. 
Clinical medicine obesity, inflammation, and prostate cancer
.
J Clin Med
2019
;
8
:
201
.