Anticancer Effect of Icaritin on Prostate Cancer via Regulating Abundance of Akkermansiaceae and Vitamin K2 in Intestinal Fecal

TL;DR

Icaritin helps fight prostate cancer by boosting gut bacteria Akkermansiaceae and vitamin K2, which also improves survival in mice.

Contribution

This study reveals a novel mechanism by which Icaritin suppresses prostate cancer through gut microbiome modulation.

Findings

Icaritin restored Akkermansiaceae and vitamin K2 levels in high-fat diet-fed mice.

Increased vitamin K2 and improved adipokine profiles were linked to longer survival in TRAMP mice.

Gavage with Akkermansiaceae alone also improved survival and adipokine levels.

Abstract

Our study investigated the mechanisms by which Icaritin (ICT) suppresses prostate cancer (PCa) progression via modulation of the gut microbiome. Using C57BL/6 and TRAMP mouse models, we found that a high-fat diet (HFD) significantly reduced in-testinal Akkermansiaceae abundance and vitamin K2 levels relative to control di-et-fed mice. ICT supplementation in HFD-fed TRAMP mice restored Akkermansiaceae pop-ulations and elevated vitamin K2 levels, concurrently decreasing serum leptin and in-creasing adiponectin concentrations. Direct gavage with Akkermansiaceae similarly elevated vitamin K2 levels, improved adipokine profiles, and extended survival. These findings demonstrate that ICT exerts anti-tumor effects by modulating gut Akker-man-siaceae, thereby enhancing vitamin K2 production and regulating adipokine se-cretion, which supports its clinical potential for PCa treatment.

Background: High-fat diet (HFD) induced inflammation and tumorigenesis by altering gut microenvironment and gut microbiome profiles. Vitamin K2 along with Akkermansiaceae, an intestinal symbiote, is involved in anti-tumors including prostate cancer (PCa). Although there have been clinical studies of Icaritin (ICT) for anti-tumor, its detailed mechanism for anti-PCa remains unclear. Here, we explored the effect of ICT on PCa progression that involved altering the mouse gut microbiome. Methods: We used 20 C57BL/6 mice and 85 transgenic adenocarcinomas of mouse prostate (TRAMP) mice as animal models. Then, 16SrRNA pyrophosphate gene sequencing was used to analyze the intestinal fecal microbiome of mice in each group. At the same time, vitamin K2 in fecal samples was determined by High Performance Liquid Chromatography (HPLC), and the content of adipokines leptin and adiponectin in serum was determined by Elisa. Results: It was observed that the relative abundance of Akkermansiaceae (9.14 ± 2.75% (C57BL/6J-N) vs. 5.66 ± 1.35% (C57BL/6J-HFD); 1.21 ± 0.61% (TRAMP-N) vs. 0.67 ± 0.45% (TRAMP-HFD) (p < 0.05) and vitamin K2 (C57BL/6J-HFD was 0.41-fold to C57BL/6J-N, TRAMP-HFD was 0.71-fold to TRAMP-N) in the intestinal fecal of HFD C57BL/6 and TRAMP were significantly lower than those of control mice. Both the ratio of Akkermansiaceae and vitamin K2 in the gut fecal were significantly increased in HFD-fed TRAMP mice supplemented with ICT compared with HFD-fed TRAMP group (p < 0.01). When HFD TRAMP mice were fed with Akkermansiaceae by gavage, the content of vitamin K2 in the intestinal fecal of the mice was increased, the adipokines leptin in the serum were decreased, but the adiponectin was significantly increased, and the overall survival time of the TRAMP mice was significantly prolonged. Our study shows that ICT rescued the reduction in Akkermansiaceae in the gut microbiota of HFD TRAMP mice. ICT suppresses PCa by modulating Akkermansiaceae in the gut, increasing vitamin K2, and regulating adipokines leptin and adiponectin secretion. Conclusions: Our study suggested that ICT might regulate abundance of Akkermansiaceae to increase vitamin K2 to suppress PCa. This provides theoretical basis and data support for the clinical application of ICT in the treatment of PCa.

Figures

Full text

1. Introduction

PCa is one of the most common malignant tumors in male patients [1]. There are 3.3 million new reported PCa each year in China [2]. Although there have been numerous new preclinical treatment strategies for PCa in recent years, such as the use of miRNAs and their derivative products in PCa [3,4], effective and potentially clinically applicable new treatment methods still require further exploration.

Studies have found that HFD may alter biological activities and functions related to inflammation, tumors and other diseases [5,6]. HFD may induce adipose-mediated secretion and production of various growth factors, cytokines, chemokines, and hormone-like molecules to increase PCa morbidity and mortality [7,8]. In addition, researchers have found that the progression of PCa in overweight patients is closely related to the abnormal secretion of adipokines and cytokines [9], highlighting that HFD might be critical for PCa progression. Leptin and adiponectin are the two main components of adipokines. The increase in serum leptin is positively correlated with tumor progression, while the increase in adiponectin inhibited tumor cell proliferation [10], suggesting that HFD induced adipokines releasing. Other studies have shown that lower serum adiponectin levels and decreased expression of adiponectin receptors in tumor tissues of PCa patients which suggests that serum adiponectin level may be a risk factor for PCa [11]. Elevated leptin levels correlated with high risk of diagnosis with high-volume tumor in PCa patients through testosterone and factors related to stature and obesity [12].

Gut microbiota participates in and affects tumor progression and antitumor drug metabolism in various organ systems including PCa [13]. HFD alters the gut microenvironment and gut microbiome to induce the development of inflammation, which may be associated with tumorigenesis [14]. Gut microbes may influence epigenetic modifications by affecting DNA methylation and histone modifications in the immune system [15]. Epigenetics involves various stages of PCa occurrence, development and transfer, among which DNA methylation and histone modification play an important role in the PCa mechanism [16]. Dysregulation of the gut microbiome with reducing beneficial bacteria such as Bacteriodes, Lactobacillus, and Eubacterium is also associated with the inflammatory response and the higher levels of inflammatory factors such as tumor necrosis factor-alpha (TNF-α), interleukins (IL-6, IL-8, and IL-1β) [17]. Chronic prostate inflammation mediated by inflammatory factors such as IL-6 and IL-8 is closely related to the tumorigenesis and progression of PCa [18]. In other studies, it was showed that PCa altered the gut microbiota with loss of protective bacteria such as Akkermansia muciniphila and Lachnospiraceae spp., which could be reversed by multiple ways in a mice model [19,20]. Therefore, further investigation on the gut microbiome profiles and the inflammation-induced microbiome species would provide insights on the biology of tumorigenesis in PCa and the new microbiome-based therapeutics.

Akkermansiaceae is an intestinal symbiote that colonizes the mucosal layer and is regarded as a promising candidate probiotic. Akkermansiaceae has significant value in improving the host’s metabolic function and immune response which is abundantly present in the intestinal microbiota of healthy individuals and has the functions of preventing and treating obesity, type 2 diabetes and other metabolic disorders. The reduction in Akkermansiaceae may be associated with the elevated levels of trimethylamine N-oxide (TMAO) in the peripheral blood and renal tissues of patients with renal injury and in the peripheral blood of patients with atrial fibrillation [21,22]. Supplementation with Akkermansiaceae may be beneficial for inhibiting TMAO production mediated by the intestinal microbiota [23]. Akkermansiaceae affects the progression of various tumors [24]. The components of this bacterium, such as Amuc_2172 and Amuc 1100, affect tumor occurrence and metabolites of which, such as Butyrate, can induce apoptosis of cancer cells. Moreover, Akkermansiaceae can directly stimulate the immune response to inhibit tumor occurrence [24]. Akkermansiaceae is influenced by dietary factors, which is helpful to enhance the therapeutic effect of cancer. Immune-based cancer treatments, such as the clinical response of tumor patients to immune-checkpoint inhibitors (ICIs), are positively correlated with the relative abundance of Akkermansiaceae [24]. In addition, the combined treatment of IL-2 and Akkermansiaceae can achieve a stronger anti-tumor effect. Results of a recent study showed that the oral androgen biosynthesis inhibitor Abiraterone acetate promotes the bacterial biosynthesis of vitamin K2, an inhibitor of androgen-dependent and independent tumor growth, by increasing Akkermansiaceae in the intestinal microbiota, and simultaneously has anti-tumor efficacy against both hormone-dependent and non-hormone-dependent PCa and this anti-PCa effect is highly repeatable [20]. Therefore, Akkermansiaceae combined with vitamin K2 involved in our study to explore potential targets of PCa therapy.

Icariin and Icariin II, the main active components of ICT (Figure 1), which is derived from the traditional Chinese medicine Epimedium, have significant anti-tumor effects on various tumors, such as PCa, kidney cancer, hepatoma, lung cancer and osteosarcoma [25,26,27,28]. In addition, ICT significantly reduces the expression level of androgen receptor (AR) protein in PCa cell line, thereby inhibiting the growth of PCa [29,30]. ICT may serve as a promising cancer therapy targeting tumor cells in PCa [31]. Our previous study applied ICT to normal diet and HFD transgenic TRAMP mice, respectively, and found that the overall survival (OS) of mice treated with ICT was significantly improved, while the release of inflammatory factors was inhibited [31] and the secretion of serum adipokines was regulated [32].

Although our previous studies have used ICT for antitumor therapy and significantly improved OS in TRAMP mice, the underlying mechanism remains unclear. This study intends to investigate the effect of HFD on intestinal flora in mice, and found that Akkermansiaceae and vitamin K2 in the intestine significantly increased, and further explore the regulatory of Akkermansiaceae on the secretion of vitamin K2 along with serum adipokines in the gut of TRAMP mice, so as to elucidate the potential mechanism of ICT inhibiting PCa progression.

2. Materials and Methods

2.1. Chemicals and Reagents

ICT was purchased from Shanghai Win Herb Medical Science Corporation (Shanghai, China). Dimethyl sulfoxide (DMSO) (D8418), Vitamin K2 standard sample (V-031) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Chromatographic grade methanol (M433275) was purechased from ALADDIN-E.COM. DNeasy PowerSoil kit (47016) was supplied by QIAGEN, Inc., (Venlo, The Netherlands). Agencourt AMPure kit (A63882) was purchased from Beckman Coulter, (Milan, Italy). PicoGreen dsDNA detection kit (P7589) was purchased from Invitrogen (Carlsbad, CA, USA). MiSeq Reagent Kit v3 (A1106) was purchased from Shanghai Personal Biotechnology Co., Ltd. (Shanghai, China). Elisa kit of adipokines leptin (MOB00B) and adiponectin (MRP300) was purchased from R&D Sys-tems^®^ (Minneapolis, MN, USA).

2.2. Animals and Diets

TRAMP transgenic PCa mice and non-transgenic C57BL/6 mice were purchased from Jackson Laboratory (Bar Harbor, ME, USA). Among them, TRAMP mice have been screened and identified according to previous genotyping methods [36]. Normal diet consists of 64% calories from carbohydrates, 16% from fats and 20% from proteins. HFD consisted of 40% calories from carbohydrates, 40% from fats and 20% from proteins according to previous study [32]. The diet was provided by Shanghai Proton Mouse Diet Co., Ltd. (Shanghai, China). A total of 20 C57BL/6 mice and 85 TRAMP mice were used for the study after weaning (postnatal day 20). They were randomly divided into normally fed C57BL/6J mice (C57BL/6J-N, n = 10), HFD-fed C57BL/6J mice (C57BL/6J-HFD, n = 10), and normally fed TRAMP mice (TRAMP-N, n = 25), HFD-fed TRAMP mice (TRAMP-HFD, n = 25), HFD-fed + Oral Akkermansiaceae TRAMP mice (TRAMP-HFD + Akkermansiaceae, n = 25), HFD-fed + Oral ICT TRAMP mice (TRAMP-HFD + ICT, n = 10). Among them, TRAMP-N (n = 15), TRAMP-HFD (n = 15), TRAMP-HFD + Akkermansiaceae (n = 15) was used in survival experiment. The timeline of animal experiments has been shown in Figure 2. All animals in each group were housed at room temperature of 22–26 °C and humidity of 50–60%. Animals were approved by the Institutional Animal Care and Use Committee (IACUC) from the Huashan Hospital, Fudan University (ID 22YF1404300).

2.3. ICT Administration

ICT was dissolved in DMSO. The concentration of DMSO is limited to less than 0.1% of the total medium volume which was safe concentration according to previous studies [37,38]. The dose of ICT was set at 30 mg/kg via gavage according to a previous study [31,32]. ICT was administered to each mouse one time/day starting from 7 weeks of age until the completion of the respective experiments (23rd).

2.4. Akkermansiaceae Administration

Akkermansiaceae (ATCC BAA-835) was grown in sterilized brain heart infusion broth (BD Difco) anaerobically in an airtight pot with AnaeroPack (Thermo Fisher Scientific, Waltham, MA, USA) for approximately 48 h at 37 °C to an end concentration of 1 × 10^10^ colony-forming units (cfu)/mL. Then, Akkermansiaceae were washed twice with sterile PBS and re-suspended with sterile PBS to 1 × 10^8^ cfu/0.2 mL. Akkermansiaceae was administered by oral gavage daily at the dose 1 × 10^8^ cfu/0.2 mL suspended in sterile PBS starting from 7 weeks of the age until the completion of the respective experiments (23rd and 67th week) according to a previous study [39].

2.5. Fecal 16S rRNA Gene Pyrosequencing and Sequence Analysis

The TRAMP mice and their original mouse strain mice C57BL/6J feed with norma or HFD diet (n = 10) were reared for 20 weeks, and the fecal were collected for genomic DNA extraction according to the previous method [40]. Briefly, genomic DNA was extracted and the V3-V4 region of the 16S rRNA gene was amplified using the DNeasy PowerSoil kit; then, agarose gel electrophoresis was performed. Agencourt AMPure kit was used to purify Polymerase Chain Reaction (PCR) amplicons, and PicoGreen dsDNA detection kit was used for quantification. Finally, MiSeq Reagent Kit v3 and Illumina MiSeq platform were used to obtain equimolar pools of amplicons and perform paired-end 2 × 300 bp sequencing.

Sequence data were processed using 1.8.0 package [41]. Low-quality reads were removed, and paired-end reads were aligned using Fast Length Adjustment of Short Reads (FLASHs) [42]. Operational taxonomic units (OTUs) were clustered at a 97% cutoff using UCLUST [43], and taxonomic classification was assigned using the Greengenes database. Alpha and beta diversity were calculated to assess microbial richness and community structure. Differentially abundant taxa were identified using LEfSe analysis [44].

2.6. Detection of Vitamin K2

Vitamin K2 levels in the intestinal fecal of TRAMP mice feed with norma or HFD diet and feed HFD companied with Akkermansiaceae supplement (n = 10) was determined by HPLC according to previous studies [45,46].

Briefly, weighed fecal samples were extracted with water–methanol (50:50, v/v), vortexed, sonicated, and centrifuged. The supernatant was acidified, and the upper layer was collected, evaporated to dryness, and reconstituted in methanol. After filtration through a 0.22 μm membrane, samples were analyzed using a water–methanol (15:85, v/v) mobile phase.

A standard curve (0.1–10 μg/mL) was prepared using a vitamin K2 reference standard, and sample concentrations were calculated from peak areas. The detection limit was 0.01 μg/mL.

2.7. Detection of Adipokines

TRAMP-N, TRAMP-HFD, TRAMP-HFD + Akkermansiaceae and TRAMP-HFD + ICT mice (n = 10) were euthanized using CO_2_ asphyxiation after 20 weeks of rearing. Collect 1 mL of blood from the portal vein using a 1 mL sterile syringe and put it into an Ep tube. Subsequently, the blood was centrifuged at 13,000 rpm for 10 min at 4 °C and stored in a −80 °C freezer. The content of adipokines leptin and adiponectin was detected using Elisa kits according to the instructions.

2.8. Molecular Docking

The ligand structure was downloaded from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/, accessed on 15 February 2026). The ligand was optimized using ChemBio software and exported in mol2 format. The receptor structures were obtained from PDB (https://www.rcsb.org/, accessed on 15 February 2026) and preprocessed with PyMol software 3, then saved in pdbqt format. Molecular docking and binding energy prediction were performed using Autodock Vina software. The docking results were visualized and analyzed with PyMol software. Additionally, heatmap analysis of the docking results was conducted using the Weishengxin Visualization Cloud Platform.

2.9. Survival Analysis

The remaining mice in the TRAMP-N, TRAMP-HFD and TRAMP-HFD+ Akkermansiaceae groups (n = 15) were reared until death, and the survival time of each mouse was recorded for survival analysis.

2.10. Statistical Analysis

Data were presented as mean ± 95% confidence interval (CI). Multiple groups were analyzed using Student’s t-test or one-way analysis of variance (ANOVA). All statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA). p values less than 0.05 were considered statistically significant.

3. Results

3.1. Changes in Gut Microbiome of Mice with HFD Intervention

We analyzed the gut microbiome of C57BL/6 and TRAMP mice to determine whether HFD affects the gut microbiome. The gut microbiome in mice treated with HFD was significantly different from that of mice with normal diet. Moreover, the relative abundance of Akkermansiaceae in the HFD C57BL/6J (C57BL/6J-HFD: 5.66 ± 1.35%) and TRAMP mice (TRAMP-HFD: 0.67 ± 0.45%) was significantly lower than that in the normal feed C57BL/6J (C57BL/6J-N: 9.14 ± 2.75%) and TRAMP mice (TRAMP-N: 1.21 ± 0.61%) (p < 0.05, Figure 3A,B). The relative amounts of vitamin K2 in the intestinal contents of mice in each group were consistent with the changes in Akkermansiaceae (Figure 3C). These results suggest that HFD alters the gut microbiome profiles and reduces the abundance of Akkermansiaceae and vitamin K2 levels in PCa mice.

3.2. Effects of ICT on Gut Microbiome in HFD-Fed TRAMP Mice

To assess whether ICT exerts anti-PCa effects by altering the abundance of Akkermansiaceae in the gut, ICT was added to the diet of HFD TRAMP mice (TRAMP-HFD + ICT), the relative abundance of Akkermansiaceae (Figure 4A,C) and vitamin K2 content (Figure 4B) in the gut microbiome of TRAMP-HFD + ICT group were significantly increased compared with the TRAMP-HFD group. In addition, the ratios of Firmicutes and Bacteroidetes in the gut microbiome were reversed, suggesting that ICT may have the effect of promoting the increase in Bacteroidetes.

3.3. Effects of Akkermansiaceae on Gut Microbiome and OS in TRAMP Mice

To understand how ICT affects PCa development by altering Akkermansiaceae in the gut, we added Akkermansiaceae to HFD TRAMP mice (TRAMP-HFD + Akkermansiaceae). The gut microbiome was detected in TRAMP-N, TRAMP-HFD and TRAMP-HFD + Akkermansiaceae groups. While the TRAMP-HFD + Akkermansiaceae group had a significant increase in the relative abundance of Akkermansiaceae (Figure 5A,C), the OS was significantly prolonged (Figure 5B). Similar to Akkermansiaceae, vitamin K2 was elevated in the intestinal contents of Akkermansiaceae-supplemented mice (Figure 5D).

Results of molecular docking showed in Figure 6A,B demonstrated that ICT could bind to protein of leptin or adiponectin (interface-binding energy less to −5.5 kcal/mol) which may lead to protein inhibition or promotion. Elisa results indicated that the adipokine leptin was significantly decreased (Figure 6C), whereas the amount of adiponectin was significantly increased (Figure 6D).

4. Discussion

TRAMP is a model of PCa, encompassing lesions ranging from precancerous to metastatic stages. So far, the TRAMP model is still one of the most essential tools for PCa study, as well as for the development of new strategies to prevent the disease [47]. In the present study, we investigated and validated the effect of HFD intervention on the gut microbiome using C57BL/6 mice and TRAMP mice models, and found significant reductions in Akkermansiaceae and vitamin K2 in the gut microbiome. ICT rescued the reduction in Akkermansiaceae and vitamin K2 in the gut microbiota of HFD TRAMP mice. At the same time, supplementation with Akkermansiaceae significantly prolongs OS in TRAMP mice, antagonized the changes in adipokines leptin and adiponectin induced by HFD, and increased the content of vitamin K2 in the intestinal contents of mice. We speculated that ICT treats PCa by modulating Akkermansiaceae in the gut, increasing vitamin K2, and regulating adipokines leptin and adiponectin secretion.

Eating habits can alter the composition of the gut microbiome. HFD can cause the shift and change in intestinal flora, which is closely related to the progression of obesity [48]. The study found that TRAMP mice on HFD had significantly different abundances and species of 19 bacteria and six fungi in their gut microbiome compared to mice with a regular diet. Microbiota involved in functional metabolism were significantly reduced in the gut microbiome of HFD mice. At the same time, mice with HFD-induced obesity, insulin resistance and other phenotypes were given a normal diet, and the abnormal flora gradually recovered to the level of the control group [49]. Another study also confirmed that HFD-induced dysbiosis can be adjusted to normal levels by changing dietary habits and ingredients [50]. Our results indicated that the relative abundance of Akkermansiaceae was significantly lower in HFD-fed mice than in control-fed mice which may be a beneficial bacteria compared with other bacteria such as Ruminococcus, Streptococcus and Bacteroides species to correlate with tumor development and consistent with previous study [19,20]. The relative amount of vitamin K2 in the intestinal contents of the mice in HFD group was lower than that of the mice in the control group.

ICT can exert anti-tumor effects by inducing apoptosis, regulating cell cycle, anti-angiogenesis, inhibiting tumor metastasis and immune regulation [32,51]. Studies have found that ICT can significantly inhibit the expression of AR protein in LNCaP (PCa cells) and inhibit the proliferation of PCa cells [29], which is consistent with the results of Sun et al. [30]. Another study found that ICT can significantly improve the survival rate of TRAMP mice, and decrease the levels of serum inflammatory cytokines in mice [32]. ICT exerts antitumor effect by down-regulating UBE2C expression and up-regulating miR-381-3p level in human PCa cells [52]. Our recent study found that ICT significantly improves OS in TRAMP mice, suggesting that ICT may serve as a promising cancer preventive agent targeting to tumor cells in obese PCa patients [31]. However, how ICT affects the development of PCa remains unclear, and the underlying mechanisms still require further exploration. This study found that ICT reversed the reduction in Akkermansiaceae and vitamin K2 in the gut microbiome of TRAMP mice induced by HFD. In addition, we also observed a reversal of the ratios of Firmicutes and Bacteroidetes in the gut, and their effects on the occurrence and development of PCa require further study.

Akkermansiaceae is of great value in improving host metabolic function and immune response. In addition, Akkermansiaceae may be valuable in improving cancer treatment. In recent years, several studies have shown that Akkermansiaceae combined with PD-1 plays a synergistic role in cancer treatment and enhances the positive response rate of PD-1 therapy [53]. PD-1/PD-L1 is expressed in a subset of PCa, especially PD-L1 expression is significantly upregulated in non-organ restricted tumors [54]. These studies were consistent with our finding that adding Akkermansiaceae to the diet significantly prolonged OS in HFD TRAMP mice. Combined with our previous research, it was found that the elevation of Akkermansiaceae is a key factor in the treatment of PCa with ICT.

Vitamin K2, which is up-regulated with the increase in Akkermansiaceae, has been shown to inhibit the growth of PCa [8,20]. Vitamin K2 is a potential anticancer agent that targets CRPC in vitro [55] and inhibits androgen-independent and androgen-dependent PCa growth in mice [56]. In a large European prospective study, dietary intake of vitamin K2 was found to be inversely associated with the occurrence of PCa [57]. A recent study found that the use of abiraterone acetate (AA) in patients increased Akkermansiaceae in the gut microbiome and promoted the biosynthesis of vitamin K2, which is beneficial for hormone-dependent and non-independent PCa. Moreover, this anti-PCa effect is highly reproducible and independent of immune involvement [20]. When Akkermansiaceae was administered to TRAMP mice, the content of vitamin K2 was also increased, and the survival of TRAMP mice was significantly prolonged. It indicated that Akkermansiaceae may inhibit PCa progression by affecting the production of vitamin K2.

Leptin is one of the pro-inflammatory factors secreted by adipose tissue [58]. Studies have shown that leptin can participate in tumor proliferation and metabolism by activating various growth factors in PCa to activate the corresponding signaling pathways [59,60]. In addition, the detection of plasma leptin in PCa patients found that it was significantly higher than the normal level, which may contribute to the development of PCa [61]. Adiponectin is also an adipokine secreted by adipose tissue and is involved in various metabolic activities, including cancer. Studies have found an increased risk of PCa at lower adiponectin levels [62]. Bub et al. found that adiponectin can effectively inhibit the proliferation of PCa cells [33]. We analyzed the serum of mice in TRAMP-HFD+ Akkermansiaceae group and found that the adipokine leptin was significantly decreased and the level of adiponectin was significantly increased. Therefore, by interfering with the intestinal flora of HFD TRAMP mice, the secretion of adipokines leptin and adiponectin can be regulated, thereby affecting the development of PCa. Our study revealed that ICT may increase the production of vitamin K2 by increasing Akkermansiaceae in the gut, thereby reducing the adipokines leptin and increasing adiponectin, and improving the OS of HFD TRAMP mice. ICT has been used in clinical research on tumors [34] and exerts anticancer activity along with safety to advanced hepatocellular carcinoma (HCC) patients with poor prognoses, who are unsuitable for conventional therapies. Furthermore, ICT, approved by the State Food and Drug Administration in January 2022 for the treatment of advanced HCC [35], can improve survival, delay progression, and produce clinical benefits in HCC patients, with a favorable safety profile and minimal adverse events [63]. ICT combined with targeted drugs for treating intermediate and advanced malignancies brings hope to patients who cannot or refuse to take chemotherapy due to its less toxic effect on the liver and kidney and bone marrow suppression according to a case report of pancreatic cancer patient treated for a month [64]. However, the study of ICT‘s effect on PCa is limited to animal models. The low bioavailability due to its poor water solubility and membrane permeability would dampen its potential beneficial effects [65]. There is still a long way to go before it can be clinically applied to the treatment of human tumors including PCa.

Although our study found that ICT can exert a good anti-cancer ability by regulating the intestinal flora, there are still some limitations in this study. First, the potential side effects of ICT have not been studied. Second, it is still necessary to further explore whether the changes in the abundance of other intestinal flora such as Bacteroides after ICT treatment have an impact on the occurrence and development of PCa involved animals with more samples or clinical research along with specific molecular mechanism of ICT to modulate Akkermansiaceae and vitamin K2 through transcriptomic or proteomic analyses. Further functional tests are needed to verify the effect of ICT on PCa and provide more reliable evidence for ICT in the treatment of PCa. Lastly, our results are primary. For instance, the lack of pharmacokinetic and safety data cannot fully support the application of ICT in the clinical treatment of PCa, metabolite analysis of vitamin K2 using mass spectrometry as well as correlation of ICT with tumor incidence and stage within survival cohorts of PCa would be involved in our further study.

5. Conclusions

In conclusion, our study showed that the use of ICT rescued the reduction in Akkermansiaceae in the gut microbiota of HFD TRAMP mice. After administering Akkermansiaceae supplementation to HFD TRAMP mice, we found that the OS was significantly prolonged in HFD TRAMP mice. Continuing to explore the potential mechanism, it was found that ICT may increase vitamin K2 in the intestine by increasing Akkermansiaceae, while reducing the serum adipokines leptin and increasing adiponectin. This study provides theoretical basis and data support for the clinical application of ICT in the treatment of PCa. According to some of the limitations of our previous study, potential side effects of ICT, the changes in the abundance of other intestinal flora such as Bacteroides after ICT treatment of PCa including more samples of animals along with transcriptomic or proteomic analyses to explore specific molecular mechanism of ICT to modulate Akkermansiaceae and vitamin K2, more functional tests for ICT on PCa, pharmacokinetic and safety data of ICT for clinical research, metabolite analysis of vitamin K2 using mass spectrometry and correlation of ICT with tumor incidence and stage within survival cohorts of PCa patients would be involved in our further study.