Diversity of Gut Microbiota and Metabolites in Benign Prostatic Hyperplasia with Different Prostate Volumes

Take Home Message

This study examines the diversity of gut microbiota and metabolites in benign prostatic hyperplasia via 16S rDNA sequencing and untargeted metabolomics, elucidating the underlying mechanisms of the gut-prostate axis.

Abstract

Background and objective

The gut microbiota, influenced by age and sex hormones, may correlate with the development and progression of benign prostatic hyperplasia (BPH). This study aims to characterize gut microbiota and metabolite profiles in BPH patients with varying prostate volumes.

Methods

Fecal samples from BPH patients were analyzed using 16S rDNA sequencing and untargeted metabolomics. Microbial and metabolic differences were assessed via the Linear discriminant analysis Effect Size, KEGG pathway enrichment, and a mediation analysis.

Key findings and limitations

We identified 26 differential amplicon sequence variants (ASVs) and 70 metabolites, with 18 microbes correlating significantly with clinical BPH indicators. The key pathways included unsaturated fatty acid and steroid hormone biosynthesis. Akkermansia (ASV549) may affect prostate volume through the regulation of intestinal amino acid metabolism and may negatively affect prostate-specific antigen levels by inhibiting heat shock protein (HSP) 90 (luminespib). Limitations include sample size and unmeasured confounders.

Conclusions and clinical implications

Gut microbiota and metabolite diversity are associated with prostate volume; further studies are warranted to elucidate the potential interventions via microbiome modulation or metabolic targeting for BPH management.

Patient summary

In this study, we identified the potential associations between gut and both prostate volume and benign prostatic hyperplasia symptoms. These findings suggest that dietary interventions or fecal microbiota transplantation may represent potential strategies for modulating prostate health in the future.

1. Introduction

In the human microbiome, the gut microbiota is one of the largest microbial communities, consisting of up to 1014 different microbial species, with the dominant phyla being Firmicutes and Bacteroidetes [1]. The composition of the gut microbiota is closely related to host health, and its variation can be influenced by multiple factors, such as diet [2], hormones [3], and age [4]. Meanwhile, the gut microbiota plays an important role in various aspects of host health, including nutrition metabolism, growth and development, immune regulation, and general health [5]. Dysbiosis of the gut microbiota is associated with multisystem diseases, such as intestinal diseases, metabolic syndrome, autism spectrum disorders, and Alzheimer’s disease [6], [7]. Therefore, identification of factors that affect the stability of the gut microbiota is of great significance for preventing and treating microbiota-related diseases.

Benign prostatic hyperplasia (BPH) is one of the most common diseases among elderly men, with a prevalence of about 42% at age 50 yr and approximately 80% at age 80 yr [8]. BPH usually presents with lower urinary tract symptoms (LUTS), including frequency, urgency, nocturia, hesitancy, and reduced urinary volume [9]. It is characterized by a long course of disease and a heavy disease burden, significantly affecting patients' quality of life and consuming considerable medical resources [10]. The severity of symptoms in BPH is associated with disease progression and prostate volume (PV), showing certain heterogeneity [11]. For patients with significant LUTS but small PV or only mild hyperplasia (ie, in whom the degree of prostate enlargement does not match the degree of obstruction), this condition is referred to as small-PV BPH [12]. PV can help predict disease progression and the risk of complications, and guide treatment choices [13]. Although BPH can be treated effectively, it is difficult to cure completely, especially for patients with small PV, for whom surgical outcomes are unclear and bladder neck contracture is common [14]. Therefore, exploring new treatment methods for BPH from other perspectives is necessary. Previous studies have found a close relationship between BPH and gut microbiota [15], [16], [17]; however, the correlation between PV and gut microbiota remains unclear. Therefore, we collected fecal samples and related clinical indicators from BPH patients with different PVs to study the effect of gut microbiota composition and metabolomics on PV at the 16S rDNA sequencing and untargeted metabolomic levels.

2. Patients and methods

2.1. Participants and clinical study design

This study conforms to the principles outlined in the Declaration of Helsinki and its amendments. The study was approved by the local ethics committee of Shanghai General Hospital, Shanghai Jiao Tong University, China (ethics approval no. 2025KS258). The protocol was fully explained to the volunteers who gave their written informed consent prior to participation. The study population comprised men clinically diagnosed with BPH and referred to the Department of Urology, Shanghai General Hospital, for further evaluation and treatment. The inclusion criteria for participating in this study were clinically diagnosed BPH and age over 40 yr.

The exclusion criteria included renal insufficiency, previous gastrointestinal surgery, gastrointestinal disease (hepatic insufficiency or intestinal inflammation) or removal of gall bladder, as well as intake of antibiotics or nutraceuticals (at least 4 wk before the study). Fifty-five men were finally enrolled for the study, including 34 with PV of >50 ml and 21 with PV of <50 ml. The age range was 53–87 yr, with an average age of 70.7 ± 7.8 yr. The average PV was 87.8 ± 54.8 ml, the average International Prostate Symptom Score (IPSS) was 17.3 ± 7.6, and the average postvoid residual (PVR) urine volume was 83.7 ± 150.3 ml.

2.2. Collection of clinical indicators

Basic information and fecal samples were collected from patients treated at Shanghai General Hospital, China, from August 2024 to December 2024. Fecal collection was performed before treatment to avoid potential confounders. LUTS severity was assessed using the IPSS and the quality of life (QoL) scores. Fasting morning blood and urine samples were collected to measure laboratory indicators such as sex hormones, prostate-specific antigen (PSA), urinalysis, and urine culture. PVR was collected by ultrasound.

2.3. Bacterial DNA extraction

Genomic DNA was extracted from fecal samples using the QIAamp PowerFecal Pro Kit (51804; QIAGEN, Hilden, German).

2.4. Use of 16S rDNA V3-V4 sequencing

A 16S rDNA V3-V4 region library was constructed using TransStart astPfu DNA polymerase (TransGen Biotech, Beijing, China) and primers 338F (5′-ACTCCTACGGGAGGCAGCAG-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′). The library was subjected to quality control using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) or Qsep100 Bioanalyzer (BIOptic, New Taipei City, Taiwan, China), and quantified using Qubit 2.0 (Thermo Fisher Scientific, Waltham, MA, USA). Qualified libraries were sequenced using the Illumina MiSeq platform (PE300 strategy, with an average of 50 000 tags per sample and sequencing quality Q30 over 80%; Illumina, San Diego, CA, USA).

2.5. Untargeted metabolomic analysis

Fecal samples were processed using homogenization and cell lysis, and the supernatant was used for an analysis. The samples were separated using an Agilent 1290 Infinity LC ultra-high-performance liquid chromatography system (Agilent Technologies) with a hydrophilic interaction liquid chromatography column and analyzed by a Triple TOF 6600 mass spectrometer (AB SCIEX, Boston, MA, USA). KEGG pathway annotation, clustering, enrichment analysis, trend clustering, and correlation analysis of the metabolites were performed.

2.6. Measurement of PV

The PV was calculated via ultrasound or magnetic resonance imaging defining the anteroposterior (AP), craniocaudal (CC), and laterolateral (LL) diameters through the ellipsoid formula of AP × CC × LL × 0.523. Patients with PV of >50 ml were defined as the large-volume BPH group; patients with PV of <50 ml were defined as the small-volume BPH group.

2.7. Statistical methods

A statistical analysis was performed using SPSS 24.0 software. Data are expressed as mean ± standard deviation ($\overline{X}$ ± s), and group comparisons were conducted using t test, Kruskal-Wallis test, and Fisher’s exact test. A Pearson correlation analysis was used for a correlation analysis. A p value of <0.05 was considered statistically significant. A Benjamini-Hochberg analysis was used for multiple comparison correction.

3. Results

3.1. Participant characteristics

Men participating in this study were on average 70.7 yr old and had a body mass index (BMI) of 23.9 kg/m². There were no significant differences between the two groups in terms of mean age, BMI, PVR, IPSS, QoL score, and six sex hormone levels. However, PV in the large-PV group was significantly larger than that in the small-PV group (122.3 ± 40.8 vs 32.1 ± 9.2 ml, p < 0.001). Additionally, PSA levels were significantly higher in the large-PV group (15.57 ± 14.14 vs 2.14 ± 1.94 ng/ml, p < 0.001), while the free PSA/PSA ratio was significantly lower (18.90 ± 8.52% vs 25.68 ± 13.29%, p = 0.025; Table 1).

Table 1.

Comparison of participant characteristics between groups ($\overline{\mathbf{X}}$ ± s)

Variable	Large-PV BPH (n = 34)	Small-PV BPH (n = 21)	p value
Age (yr)	70.1 ± 7.8	71.8 ± 7.9	0.444
BMI (kg/m2)	24.1 ± 2.9	23.4 ± 1.7	0.322
PV (ml)	122.3 ± 40.8	32.1 ± 9.2	<0.001***
PVR (ml)	72.3 ± 156.6	102.3 ± 141.2	0.477
IPSS	15.0 ± 6.7	16.2 ± 8.7	0.564
QoL	3.7 ± 2.4	3.6 ± 2.0	0.891
PSA (ng/ml)	15.57 ± 14.14	2.14 ± 1.94	<0.001***
fPSA/PSA (%)	18.90 ± 8.52	25.68 ± 13.29	0.025 *
FSH (IU/l)	12.52 ± 5.24	17.09 ± 12.78	0.069
LH (IU/l)	8.05 ± 3.95	7.75 ± 5.78	0.818
E2 (pg/ml)	26.12 ± 9.64	28.10 ± 8.37	0.441
PRO (μg/l)	0.52 ± 0.50	0.51 ± 0.33	0.977
TE (μg/l)	3.83 ± 2.07	3.78 ± 1.29	0.936
PRL (ng/ml)	15.14 ± 7.03	13.67 ± 4.45	0.396

BMI = body mass index; BPH = benign prostatic hyperplasia; E2 = estradiol; fPSA = free PSA; FSH = follicle-stimulating hormone; IPSS = International Prostate Symptom Score; LH = luteinizing hormone; PRL = prolactin; PRO = progesterone; PSA = prostate-specific antigen; PV = prostate volume; PVR = postvoid residual volume; QoL = quality of life; TE = testosterone.

^*p < 0.05.

^***p < 0.001.

3.2. Alpha and beta diversity may not affect PV

Alpha diversity is used to assess the diversity and composition of the microbial community within a sample. The results showed no significant statistical differences in the evenness index (p = 0.42; Fig. 1A), observed index (p = 0.75; Fig. 1B), and Shannon index (p = 0.48; Fig. 1C) between the large- and small-PV groups.

Fig. 1.

Alpha- and beta diversity may not affect PV. Box plots of the (A) evenness index (p = 0.42), (B) observed index (p = 0.75), and (C) Shannon index (p = 0.48) show the alpha diversity of the gut microbiota in the large- and small-PV groups, reflecting the richness and species diversity of each sample. Differences were assessed using the Kruskal-Wallis test. (D) PCoA of the gut microbiota of the large- and small-PV groups based on the Bray-Curtis distance. (E) Box plot displaying the beta diversity of the fecal microbiota between the large- and small-PV groups (p = 0.20). Bars represent the mean and standard error of the mean. PCoA = principal coordinate analysis; PV = prostate volume.

Beta diversity was analyzed using the Bray-Curtis distance method and visualized by principal coordinate analyses and boxplots. The greater the distance between the samples, the better the distinction of differences and diversity between them (Fig. 1D). The results showed no significant difference in beta diversity between the two groups (p = 0.20; Fig. 1E).

3.3. Effect of gut microbiota composition on PV

At the phylum level, the top three dominant bacteria in both groups were Firmicutes, Bacteroidetes, and Proteobacteria (Fig. 2A). At the genus level, the top three dominant bacteria in both groups were Bacteroides, Shigella, and Blautia (Fig. 2B). The gut microbiota composition of different PV groups showed distinct structural and abundance differences at both phylum and genus levels. At the phylum level, the abundance of Proteobacteria, Firmicutes, and Bacteroidota was higher in the large-PV group than in the small-PV group, while the abundance of Actinobacteria was lower (Fig. 2C). At the genus level, the abundance of Anaerostipes (p = 0.018), Prevotella (p = 0.036), and Gemella (p = 0.049) was significantly increased in the large-PV group, while the abundance of Turicibacter (p = 0.027) was significantly decreased (Fig. 2D). The study by Takezawa et al [16] found that the Firmicutes/Bacteroidetes ratio was significantly higher in patients with prostate enlargement than in those without. However, in our study, there was no significant difference in the Firmicutes/Bacteroidetes ratio between different PV groups (Fig. 2E).

Fig. 2.

Differences in gut microbiota composition between different PV groups. (A) Analysis of community structure at the phylum level. (B) Analysis of community structure at the genus level. The composition of the gut microbiota at the (C) phylum and (D) genius levels shows distinct communities. Different colors represent different bacterial taxa, with the length of the colored bars indicating the relative abundance of the taxa. (E) The Firmicutes/Bacteroidetes ratio shows no significant differences between the groups. PV = prostate volume.

3.4. Differential gut microbiota is associated with BPH-related clinical indicators

The Linear discriminant analysis Effect Size (LEfSe) identified 26 different amplicon sequence variants (ASVs) between the two groups. The relative abundance of ASVs corresponding to Akkermansia (ASV26; p = 0.047), Bacteroides (ASV39; p = 0.042), Blautia (ASV17; p = 0.027), and Anaerostipes (ASV44; p = 0.020) was significantly higher in the large-PV BPH group, whereas the relative abundance of ASVs corresponding to Bifidobacterium (ASV243; p = 0.034) and Lactobacillus (ASV95; p = 0.044) was higher in the small-PV group (Fig. 3, Fig. 3).

Fig. 3.

Alteration of gut microbiota composition influences multiple clinical indicators of BPH patients. (A) LDA plot: at the ASV level, microbial communities differ significantly between BPH groups with different PVs (LDA score >2). The colors of the histogram represent each group, with the length of the bars indicating the LDA score. A higher LDA score indicates a more significant difference in this feature between the groups. (B) Phylogenetic tree: microbial composition at various taxonomic levels. Different colors represent each group. Nodes in different colors indicate microorganisms that play a significant role in the corresponding group. From the innermost to the outermost circle, each ring represents taxonomic levels such as phylum, class, order, family, and genus. (C) LDA plot: Microbial communities differ significantly between BPH groups with different PVs (LDA score >4). (D) Correlation heatmap: a Pearson correlation analysis between differentially abundant ASVs and clinical indicators. The color scale represents the correlation coefficient (r value), with values closer to 1 indicating a stronger correlation. ASV = amplicon sequence variant; BMI = body mass index; BPH = benign prostatic hyperplasia; E2 = estradiol; IPSS = International Prostate Symptom Score; LDA = linear discriminant analysis; LH = luteinizing hormone; PRL = prolactin; PRO = progesterone; PSA = prostate-specific antigen; PVR = postvoid residual; QoL = quality of life; TE = testosterone. * p < 0.05. ** p < 0.01.

The LEfSe analysis further identified 17 differential bacteria at genius or species level between the two groups. In the large-PV group, the relative abundance of Family_XIII_AD301 (p = 0.002), Actinomyces (p = 0.018), Sphingomonas (p = 0.038), Gemella (p = 0.002), Prevotella (p = 0.023), and Bacteroides (p = 0.029) was significantly higher than that in the small-PV group. Conversely, the small-PV group showed a higher relative abundance of Clostridium (p = 0.012), Parasutterella (p = 0.037), Oscillospiraceae (p = 0.004), and Enterorhabdus (p = 0.012) than the large-PV group (Fig. 3C).

Eighteen differential ASVs showed significant correlations with clinical indicators. PVR was positively correlated with Phascolarctobacterium (ASV77; r = 0.530, p = 0.002), but negatively correlated with Streptococcus (ASV323; r = –0.438, p = 0.012). IPSS was negatively correlated with Clostridium (ASV405; r = –0.327, p = 0.040). QoL was positively correlated with Parasutterella (ASV366; r = 0.349, p = 0.028), Bacteroides (ASV87; r = 0.442, p = 0.014), Ruminococcus (ASV179; r = 0.447, p = 0.004), Akkermansia (ASV26; r = 0.378, p = 0.016), and Phascolarctobacterium (ASV77; r = 0.420, p = 0.007). PSA was negatively correlated with Parasutterella (ASV366; r = –0.322, p = 0.049), Bifidobacterium (ASV243; r = –0.367, p = 0.006), and Akkermansia (ASV549; r = –0.271, p = 0.045). Estradiol was positively correlated with Dialister (ASV10; r = 0.315, p = 0.025) but negatively correlated with Akkermansia (ASV26; r = –0.360, p = 0.010). Testosterone was positively correlated with Eubacterium (ASV242; r = 0.326, p = 0.020) and Blautia (ASV17; r = 0.322, p = 0.021; Fig. 3D).

3.5. Effect of gut metabolites and metabolic pathways on PV

The untargeted metabolomics analysis was conducted to identify the metabolites associated with PV in BPH patients. The analysis revealed significant differences in the metabolite profiles between the large- and small-PV groups. A principal component analysis showed considerable dispersion in the sample distribution, indicating intragroup heterogeneity and relatively weak intergroup differences (Fig. 4A). Differential metabolites were identified using the Wilcoxon test, revealing 70 distinct gut microbiota–associated metabolites between BPH patients with different PVs. In the large-PV group, the abundance of steroid metabolites (dehydroepiandrosterone and testosterone enanthate), lipid metabolites (phosphocholine), and amino acid metabolites (adenylosuccinate and l-valine) was significantly higher. Conversely, organic acid metabolites (2-methoxycinnamic acid), flavonoid metabolites (gambogic acid), and furan metabolites (marrubiin) were found to be lower in the large-PV group (Fig. 4B). A KEGG clustering analysis suggested that the metabolic differences between the two groups were predominantly related to the biosynthetic pathways of unsaturated fatty acids and steroid hormones (Fig. 4C).

Fig. 4.

PV may lead to alterations in gut metabolites and metabolic pathways of BPH patients. (A) PCA: samples from the two groups are relatively dispersed, indicating heterogeneity within the metabolite profiles and weak intergroup differences. Blue color represents the large-PV group, while green represents the small-PV group. (B) Volcano plot of differential metabolites: differential metabolites were selected using the Wilcoxon test. Green color indicates metabolites upregulated in the large-PV group, yellow represents metabolites upregulated in the small-PV group, and gray represents metabolites with no significant difference. (C) KEGG pathway enrichment bubble plot: the x-axis represents the enrichment factor (rich factor), which is the number of differential metabolites annotated to the pathway divided by the total number of metabolites identified in that pathway. A larger value indicates a higher proportion of differential metabolites in the pathway. The size of the bubbles corresponds to the number of differential metabolites annotated to the pathway. BPH = benign prostatic hyperplasia; PC = principal component; PCA = principal component analysis; PV = prostate volume.

3.6. Association between differential ASVs and key metabolites

The differences in ASVs and key metabolites between BPH patients with different PVs were also correlated, with 18 differential ASVs showing a significant association with key metabolites. Compared with BPH patients with small PV, patients with large PV exhibited a higher abundance of Akkermansia (ASV26) and Streptococcus (ASV323) in the gut microbiota, which were positively correlated with steroid metabolites (testosterone enanthate and 3-dehydroepiandrosterone sulfate). Anaerostipes (ASV44) was positively correlated with amino acid metabolites (n-heptanoyl-l-homoserine lactone) and nucleotide metabolites (3,5,6-trichloro-2-pyridinol), but negatively correlated with steroid hormones (19-norandrostenedione). Eubacterium (ASV329) was negatively correlated with flavonoid metabolites (bidwillon A). In contrast, BPH patients with small PV showed higher abundance of UCG-002 (ASV102), Alistipes (ASV295), and Bifidobacterium (ASV243) in their gut microbiota, which were negatively correlated with steroid metabolites (3-dehydroepiandrosterone sulfate), aromatic metabolites (luminespib), and unsaturated fatty acid metabolites (arachidonic acid; Fig. 5A).

Fig. 5.

Correlation between gut microbiota, gut metabolites, and clinical features of BPH. (A) Correlation heatmap: Pearson correlation analysis between differential ASVs and key metabolites across groups. The colors represent the magnitude of the correlation coefficient (r), with r values closer to 1 indicating a stronger correlation. (B) Correlation heatmap: Pearson correlation analysis between key metabolites and clinical indicators across groups. (C) Sankey diagram: a mediation analysis of differential ASVs, key metabolites, and clinical indicators demonstrates that key metabolites mediate the effect of differential ASVs on clinical indicators. ASV = amplicon sequence variant; BMI = body mass index; BPH = benign prostatic hyperplasia; E2 = estradiol; FSH = follicle-stimulating hormone; fPSA = free PSA; LH = luteinizing hormone; PRL = prolactin; PRO = progesterone; PSA = prostate-specific antigen; PVR = postvoid residual; QoL = quality of life; TE = testosterone; tPSA = total PSA. * p < 0.05. ** p < 0.01.*** p<0.001.

3.7. Gut metabolites are associated with clinical indicators

Among the differential metabolites of the gut microbiota between groups, 26 key metabolites were significantly correlated with clinical indicators. PVR was positively correlated with amino acridine metabolites (acridine orange; r = 0.473, p = 0.047). QoL was negatively correlated with steroid metabolites (19-hydroxyandrost-4-ene-3,17-dione; r = –0.503, p = 0.014). Estradiol was positively correlated with long-chain fatty acid metabolites (13-tetradecynoic acid; r = 0.452, p = 0.006), but negatively correlated with unsaturated fatty acids (levofloxacin; r = –0.433, p = 0.023). Testosterone was positively correlated with protease inhibitors (leupeptin; r = 0.361, p = 0.047). PSA was positively correlated with amino acid metabolites (amfenac; r = 0.400, p = 0.037), nucleotide metabolites (adenylosuccinate; r = 0.412, p = 0.027), and lipid metabolites (lpc 16:0; r = 0.407, p = 0.030), but negatively correlated with leupeptin (r = –0.452, p = 0.008), tryptophan metabolites (L-tryptophanamide; r = –0.477, p = 0.003), and long-chain fatty acid metabolites (13-tetradecynoic acid; r = –0.450, p = 0.006; Fig. 5B).

3.8. Gut microbiota affect BPH-related clinical features through metabolites

A mediation analysis of the microbiota–metabolite-clinical indicator relationships revealed that the correlation between the microbiota and clinical indicators may be mediated by 16 key metabolites. Notably, in BPH patients with small PV, the relatively high abundance of Akkermansia (ASV549) might negatively correlate with PSA levels by inhibiting heat shock protein (HSP) 90 inhibitors (luminespib; p = 0.034). The relatively high abundance of Bifidobacterium (ASV243) in BPH patients with small PV might negatively correlate with PSA levels through the inhibition of the major member of the sialic acid family: N-acetylneuraminate (p = 0.004). The relatively high abundance of Parasutterella (ASV366) in the small-PV group might have a positive indirect effect on prolactin through bilirubin (p = 0.010) and beta-D-fructose 2-phosphate, which was involved in the fructose and mannose system (p = 0.006), while promoting leupeptin (p = 0.048). Moreover, the relatively high abundance of Streptococcus (ASV323) in the large-PV group exerted its positive influence on PVR via 1,1-dimethyl-4-phenylpiperazinium, a selective nicotinic acetylcholine receptor agonist (p = 0.046; Fig. 5C).

4. Discussion

Homeostasis of the gut microbiota is essential for maintaining human health, while gut microbiota dysbiosis may lead to functional abnormalities and increase the risk of diseases [18]. Additionally, gut microbiota is known to undergo changes under pathological conditions [19]. Studies have revealed a strong association between gut microbiota and BPH [15], [16], [17]. In this study, we identified significant differences in gut microbiota composition and metabolomic profiles between patients with large- and small-volume BPH. Furthermore, these differences were found to be potentially associated with patients' clinical symptoms and indicators, providing novel insights into the pathogenesis, progression, and potential therapeutic approaches for BPH.

Our results showed that there were significant differences in gut microbiota composition between BPH patients with different PVs. The large-PV group had a higher abundance of certain genera, such as Akkermansia (ASV26) and Bacteroides (ASV39), while the small-PV group had a higher abundance of Bifidobacterium (ASV243) and Lachnoclostridium (ASV95). These findings are consistent with those of previous studies that have demonstrated the involvement of these bacteria in regulating host immune function, metabolism, and inflammation [20], [21]. Akkermansia (ASV26), for example, has been shown to play a role in regulating the gut barrier function and modulating metabolic pathways that are relevant to prostate health [22]. An analysis of the differential microbiota between groups may contribute to advancing the understanding of the pathogenesis of BPH.

Moreover, a metabolomic analysis revealed distinct differences in the metabolite profiles of BPH patients with different PVs. The large-PV group had a higher concentration of metabolites involved in amino acid metabolism, while the small-PV group showed higher levels of gambogic acid and marrubiin, which were related to oxidative stress and anti-inflammatory responses [23]. Previous studies have demonstrated that unsaturated fatty acid metabolism is closely related to immune function. For example, York et al [24]suggested that unsaturated fatty acid metabolism may participate in the anti-inflammatory mechanisms of interleukin-10. Compared with BPH patients with small PV, patients with large PV had more active immune status. This suggests that BPH patients with large PV may experience immune activation due to differences in gut microbiota metabolic pathways, with greater enrichment of unsaturated fatty acid metabolites. PV is closely related to testosterone levels, and steroid hormones (imbalanced estrogen/testosterone ratios) are considered major risk factors for BPH [25]. We found that the large-PV group had higher levels of steroid metabolites in their gut microbiota, and the steroid hormone biosynthesis pathway exhibited more differential metabolites, suggesting that the gut microbiota may be involved in steroid metabolism, thereby influencing PV during BPH pathogenesis and development. In addition, the correlation between differential metabolites and gut microbiota composition indicates a complex interaction between the microbiota and metabolic pathways in the development of BPH. For example, higher levels of amino acids in the large-volume group could be related to the increased abundance of Bacteroides (ASV39), which are known to have beneficial effects on host metabolism [26]. On the contrary, the increased oxidative stress and inflammatory metabolites in the small-volume group could be related to the higher abundance of Bifidobacterium (ASV243) and Lachnoclostridium (ASV95), which may have a protective effect against oxidative damage and inflammation [27], [28].

The Pearson correlation analysis revealed significant associations between the differential microbiota, key metabolites, and clinical indicators in BPH groups with different PVs. The mediation analysis indicated that the correlation between the gut microbiota and clinical indicators might be mediated by microbiota-derived metabolites. PSA is strongly correlated with PV and can help predict the clinical progression of BPH as well as the development of prostate cancer (PCa) [29]. In our study, we found significant correlations between several differential microbiota species and PSA levels in BPH patients with different PVs. Specifically, the relative abundance of Akkermansia (ASV549) was negatively correlated with PSA levels in the small-PV group. Previous studies have suggested a potential link between Akkermansia and PCa. For instance, Luo et al [21] demonstrated that extracellular vesicles from Akkermansia induce antitumor immunity against PCa through modulation of CD8⁺ T cells and macrophages. Evidence suggests that Akkermansia modulates lactic acid metabolism and induces the production of gut-derived serotonin [30]. Based on our results, we hypothesize that Akkermansia potentially contributes to the regulation of intestinal amino acid metabolism, which may affect PV through inflammatory pathways. Through the mediation analysis, we found that Akkermansia (ASV549) may negatively correlate with PSA levels by inhibiting HSP90 inhibitors, such as luminespib. Several studies have shown that HSP90 inhibitors can significantly impair androgen receptor function and exhibit antitumor activity in PCa [31], [32]. Additionally, the relatively high abundance of Bifidobacterium (ASV243) in small-PV BPH patients may be associated with a negative correlation with PSA levels by inhibiting N-acetylneuraminate, a key member of the sialic acid family. Previous research has shown a close relationship between serum sialic acid levels and PCa, with a significant decrease in the core fucose and an increase in the α2,3-sialic acid percentage of PSA in high-risk PCa patients [33]. Therefore, a study of the gut microbiota in BPH patients may help predict PSA levels, symptom severity, and disease progression, although the underlying mechanisms need further investigation.

5. Conclusions

Our study shows that gut microbiota and metabolite diversity are associated with PV. Gut bacteria may impact PV via microbial metabolites, ultimately influencing the clinical indicators of BPH patients. Thus, further studies are warranted to elucidate the potential interventions via microbiome modulation or metabolic targeting for BPH management.



Author contributions: Chenyi Jiang had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.



Study concept and design: Liu, Chen, Wang, Xia, Jiang.

Acquisition of data: Liu, Chen, Zhang, Qin, Jiang.

Analysis and interpretation of data: Liu, Chen, Li, Zhu.

Drafting of the manuscript: Liu, Zhu, Jiang.

Critical revision of the manuscript for important intellectual content: Xia, Jiang.

Statistical analysis: Liu, Wang, Xu, Cui, Zhu.

Obtaining funding: Wang, Xia, Jiang.

Administrative, technical, or material support: Jiang.

Supervision: Han, Jing.

Other: None.



Financial disclosures: Chenyi Jiang certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.



Funding/Support and role of the sponsor: This research was funded by National Natural Science Foundation of China (no. 81930018 and no. 82300870), and Science and Technology Commission of Jiading District (no. JDKW-2024-0022). The sponsor played a role in the collection, management, and analysis of the data.



Acknowledgments: The authors would like to thank all the families from the study for providing their time and samples, and the staff at the Shanghai General Hospital, Shanghai Jiao Tong University, for their support in sample collection and clinical data management.



Ethics statement: This study was approved by the local ethics committee of Shanghai General Hospital, Shanghai Jiao Tong University, China (ethics approval no. 2025KS258).



Data sharing statement: The DNA sequence data reported in this paper have been deposited in the Genome Sequence Archive (Genomics, Proteomics & Bioinformatics 2021) in National Genomics Data Center (Nucleic Acids Research 2022), China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (GSA: CRA027322), which are publicly accessible at https://ngdc.cncb.ac.cn/gsa/browse/CRA027322. The metabolomics data reported in this paper have been deposited in the OMIX, China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (https://ngdc.cncb.ac.cn/omix: accession no. OMIX011427).

Associate Editor: Véronique Phé