Chronic prostatitis/chronic pelvic pain syndrome induces metabolomic changes in expressed prostatic secretions and plasma

Abstract

Chronic prostatitis/chronic pelvic pain syndrome (CP/CPPS) is a complex disease that is often accompanied by mental health disorders. However, the potential mechanisms underlying the heterogeneous clinical presentation of CP/CPPS remain uncertain. This study analyzed widely targeted metabolomic data of expressed prostatic secretions (EPS) and plasma to reveal the underlying pathological mechanisms of CP/CPPS. A total of 24 CP/CPPS patients from The Second Nanning People’s Hospital (Nanning, China), and 35 asymptomatic control individuals from First Affiliated Hospital of Guangxi Medical University (Nanning, China) were enrolled. The indicators related to CP/CPPS and psychiatric symptoms were recorded. Differential analysis, coexpression network analysis, and correlation analysis were performed to identify metabolites that were specifically altered in patients and associated with various phenotypes of CP/CPPS. The crucial links between EPS and plasma were further investigated. The metabolomic data of EPS from CP/CPPS patients were significantly different from those from control individuals. Pathway analysis revealed dysregulation of amino acid metabolism, lipid metabolism, and the citrate cycle in EPS. The tryptophan metabolic pathway was found to be the most significantly altered pathway associated with distinct CP/CPPS phenotypes. Moreover, the dysregulation of tryptophan and tyrosine metabolism and elevation of oxidative stress-related metabolites in plasma were found to effectively elucidate the development of depression in CP/CPPS. Overall, metabolomic alterations in the EPS and plasma of patients were primarily associated with oxidative damage, energy metabolism abnormalities, neurological impairment, and immune dysregulation. These alterations may be associated with chronic pain, voiding symptoms, reduced fertility, and depression in CP/CPPS. This study provides a local-global perspective for understanding the pathological mechanisms of CP/CPPS and offers potential diagnostic and therapeutic targets.

INTRODUCTION

Prostatitis is a highly prevalent disease, and 4.5%–9.0% of the male population is diagnosed with prostatitis.1,2 Approximately 90.0% of men with prostatitis symptoms are diagnosed with chronic prostatitis/chronic pelvic pain syndrome (CP/CPPS), known as National Institutes of Health (NIH) category III, due to a lack of evidence of urinary tract infection and a disease duration of more than 3 months.3,4 The clinical presentations of CP/CPPS include chronic pain in the pelvis, lower abdomen, lower back and genitals, obstruction or irritative symptoms of voiding, sexual dysfunction, infertility, and psychological disturbances.4,5,6 The NIH Chronic Prostatitis Symptom Index (NIH-CPSI) has been developed and widely used to quantify the severity of CP/CPPS. This index divides symptoms into three domains: pain, urination, and quality of life.7 However, due to the lack of effective laboratory tests for quantifying the symptoms, the diagnostic criteria for CP/CPPS are still vague. Moreover, the etiology and mechanisms of the heterogeneous clinical presentations of CP/CPPS remain uncertain, so there is no definitive therapeutic approach to mitigate symptoms.8,9 Thus, CP/CPPS significantly impairs quality of life and imposes a substantial psychological burden on patients.10

With the “snow flake hypothesis” and the urinary, psychosocial, organ-specific, infection, neurologic/systemic, and tenderness domains (UPOINT) classification proposed in 2009, CP/CPPS has begun to be considered a systemic multifaceted syndrome, not just a prostate disease.11,12 In particular, psychiatric disorders, including depression and anxiety, caused by CP/CPPS are the most noteworthy systemic changes.13,14 Furthermore, psychological disorders can predict more serious symptoms and lower quality of life in CP/CPPS patients after 12 months.15 These results highlight the importance of screening for psychiatric symptoms in individuals with CP/CPPS. However, thus far, the causal relationship between CP/CPPS and psychological disorders has not been elucidated.16

Metabolomics could serve as a powerful method to discover novel biomarkers and potential mechanisms, as metabolites can represent the downstream output of the genome and proteome.17 Expressed prostatic secretions (EPS) and plasma samples are valuable for investigating both localized and systemic changes in the CP/CPPS. EPS comprises proteins and metabolites that are secreted/released from the prostate into the extracellular environment and reflect the physiological state of the prostate.18,19,20 Although some studies have explored the metabolic content in EPS, only a very few have investigated the metabolomic alterations in direct-EPS of CP/CPPS.21,22 In addition, although numerous studies have provided evidence of alterations in the blood metabolome among individuals with psychiatric disorders,23,24,25 alterations in plasma metabolites among CP/CPPS patients with psychiatric symptoms have yet to be investigated. Such insights might contribute to elucidating the biological mechanisms of these systemic changes.

In the present study, a widely targeted metabolomic platform was used to explore the metabolomic alterations in the EPS and plasma of CP/CPPS patients. We preliminarily investigated the specific metabolites associated with intricate symptoms (pain, urination, and psychiatric symptoms) and sought to uncover potential interactions between EPS and plasma metabolites.

PARTICIPANTS AND METHODS

Study population

From October 2021 to November 2021, 24 patients who were diagnosed with CP/CPPS based on the NIH criteria were recruited from the Department of Urology of The Second Nanning People’s Hospital (Nanning, China), and 35 asymptomatic control individuals (ACs) who underwent premarital reproductive system examinations were recruited from the Center of Reproductive Medicine of First Affiliated Hospital of Guangxi Medical University (Nanning, China). All participants completed the NIH-CPSI questionnaire, the Zung’s Self-Rating Depression Scale (SDS), and the Zung’s Self-Rating Anxiety Scale (SAS) to evaluate the severity of prostatitis symptoms and psychiatric symptoms within the last week. The cut-off values of the SDS and SAS were 53 and 50, respectively.26,27 Furthermore, CP/CPPS patients were evenly divided into high-CPSI (CPSI >20) and low-CPSI (CPSI ≤20) subgroups.

The inclusion criteria for CP/CPPS patients were as follows: (1) aged 20–50 years; (2) had persistent symptoms of CP/CPPS for more than 3 months and had previous negative bacterial cultures; and (3) had a CPSI ≥10 points. The inclusion criteria for ACs were as follows: (1) aged 20–50 years; (2) no prostatitis symptoms within the past 6 months; and (3) underwent a health check-up and were considered asymptomatic and healthy. The exclusion criteria were as follows: (1) received antibiotics within 3 months; (2) had active urinary tract infection, benign prostatic hyperplasia, or other diseases of the genitourinary system potentially affecting the study results; or (3) had a history of pelvic surgery, malignancy, or severe metabolic disorders. This study was approved by the Ethics Committee of First Affiliated Hospital of Guangxi Medical University (Approval No. 2017KY121). All participants provided written informed consent for the collection of clinical information and biological samples.

Sample collection

After prostatic massage, the EPS samples were collected in 1.5-ml sterile centrifuge tubes, placed on ice, and transferred to the laboratory immediately. A small portion of each EPS sample was used for regular laboratory examination. The remaining portion of EPS (at least 100 μl) was centrifuged at 376g at 4°C for 10 min (5424R; Eppendorf, Hamburg, Germany) to remove insoluble solids and then immediately stored at −80°C until analysis. The white blood cell (WBC) count of EPS was evaluated in a high-power mirror field (HP) and divided into 5 levels: occasional or few (0–9 counts per HP), 1+ (10–20 counts per HP), 2+ (21–30 counts per HP), 3+ (31–40 counts per HP), and 4+ (>40 counts per HP). Fasting blood samples were collected in ethylenediaminetetraacetic acid (EDTA) vacutainer tubes. The tubes were immediately placed on ice and centrifuged at 1811g at 4°C for 10 min (5810R; Eppendorf) to obtain the supernatant plasma. Then, the plasma samples were immediately stored at −80°C until analysis.

Widely targeted metabolomic detection

EPS and plasma metabolites were quantified by ultra-performance liquid chromatography (UPLC; ExionLC AD, Sciex, Framingham, MA, USA) and tandem mass spectrometry (MS/MS; QTRAP, Sciex) systems provided by MetWare Biotechnology Co., Ltd. (Wuhan, China).

First, EPS samples were thawed on ice and vortexed for 10 s. Then, 400 μl of 20.0% acetonitrile methanol extractant with an internal standard was added to 100 μl of the EPS. The mixture was mixed and centrifuged at 13 523g at 4°C for 10 min (5424R). The supernatant was concentrated to a dry powder and then redissolved in 100 μl of 70.0% methanol aqueous solution. After ultrasonication in a whirling and ice-water bath, the redissolved mixture was centrifuged again at 13 523g at 4°C for 3 min (5424R). Then, 80 μl of the supernatant was added to the liner of the injection bottle for detection. Similarly, plasma samples were thawed on ice and vortexed for 10 s. Then, 300 μl of 20.0% acetonitrile methanol extractant with an internal standard was added to 50 μl of the plasma sample. The mixture was mixed and centrifuged at 13 523g at 4°C for 10 min (5424R). The supernatant was transferred to a new tube, incubated for 30 min at −20°C, and then centrifuged again at 13 523g at 4°C for 3 min (5424R). Then, 180 μl of the supernatant was added to the liner of the injection bottle for detection. Quality control (QC) samples, mixed from all sample extracts, were added to the detection sequence after every 10 samples to monitor the repeatability of detection. Details of the internal standards used are provided in Supplementary Table 1. The UPLC-MS/MS conditions used here were in accordance with a previously published method.28

Supplementary Table 1.

The specification of the internal standard used in ultra-performance liquid chromatography and tandem mass spectrometry

Compounds	CAS number	Purity	Brand	Item number	Final concentration in the extract
L-2-chlorophenylalanine	103616-89-3	0.98	J and K Scientific (Beijing, Chian)	106151-100 mg	1 µg ml−1
4-Fluoro-L-α-phenylglycine	19883-57-9	0.98	TCI (Shanghai, China)	ZTO-F0862-1 g	1 µg ml−1
[2H5]-Kynurenic acid	350820-13-2	0.98	Isoreag (Shanghai, China)	IR-21812	1 µg ml−1
[2H5]-Phenoxy acetic acid	154492-74-7	0.99	Isoreag (Shanghai, China)	IR-22227-250 mg	1 µg ml−1
Indole-3-butyric-2,2-d2 acid	133-32-4	0.99	C/D/N Isotopes (Pointe-Claire, Canada)	100 mg	1 µg ml−1
LysoPC 19:0	108273-88-7	0.99	Zzstandard (Shanghai, China)	ZA-10515	1 µg ml−1
DL-3-IAL	832-97-3	0.98	Topscience (Shanghai, China)	T5230	1 µg ml−1

IAL: indole-lactic acid; CAS: Chemical Abstracts Service

Metabolomic data processing and analysis

The raw mass spectrometry data were processed using Analyst 1.63 software (Sciex). Based on the self-built targeted standard database by MetWare Biotechnology Co., Ltd., metabolites were identified by retention time, precursor/product ion pair information, and secondary spectrum data. The quantitative analysis of metabolites was accomplished using multiple reaction monitoring analysis via triple quadrupole mass spectrometry. The mass spectrum file was opened with MultiQuant software (Sciex), and the chromatographic peaks were integrated and corrected. The peak area of each chromatographic peak represented the relative content of the corresponding metabolite.

The metabolomic data underwent log transformation and Pareto scaling after probability quotient normalization or sum normalization. For univariate analysis of metabolomic data between any two groups, a two-tailed unpaired t-test was performed. For multivariate analysis, principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA) were performed. The above process was accomplished with MetaboAnalyst 5.0 (https://www.metaboanalyst.ca/). According to a previous widely targeted metabolomic research,29 the differentially expressed metabolites (DEMs) in EPS were identified using the criteria of a fold change >2 or <0.5 and variable importance in projection (VIP) >1. VIP values were obtained from the OPLS-DA results. Due to the unsatisfactory modeling of OPLS-DA in plasma, plasma DEMs were identified using the criteria of a fold change >2 or <0.5 and P<0.05 with Student’s t-test. To examine the correlation between clinical parameters and metabolites and between EPS and plasma, Spearman’s correlation and partial correlation analyses were conducted. The partial correlation analysis of metabolites with all clinical indices excluded the confounding effects of age and body mass index (BMI). Given the frequent coexistence of urinary and pain symptoms in CP/CPPS, as well as the tendency for patients with psychiatric symptoms to exhibit higher CPSI scores, the partial correlation analysis of metabolites with pain and urination scores excluded their mutual influence. Similarly, the partial correlation analysis of metabolites with SAS and SDS scores aimed to further eliminate the confounding effects of CPSI scores.

Enrichment analysis and visualization of metabolic pathways were conducted using Kyoto Encyclopedia of Genes and Genomes (KEGG) database and MetaboAnalyst 5.0 (https://www.metaboanalyst.ca/) based on quantitative enrichment analysis or over representation analysis. Receiver operating characteristic (ROC) curves based on logistic regression were used to assess the diagnostic performance of the panel constructed using specific metabolites.

Weighted gene coexpression network analysis (WGCNA) was subsequently conducted to identify the coexpression modules of the metabolites. Highly correlated metabolites were clustered into modules to reduce the dimensionality of the dataset. The correlation between the module and clinical traits was then calculated. The soft threshold power was computed based on the scaled-free topology model parameters (soft threshold power = 11 for EPS metabolomics, and 13 for plasma metabolomics), and minModuleSize was set to 10 and mergeCutHeight was set to 0.25. All other parameters were set to default values.

The visualizations generated from the aforementioned analyses were created using R (version 4.2.1; https://www.r-project.org/), GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA), and the MetWare Cloud (https://cloud.metware.cn/).

Statistical analyses

Quantitative data were presented as the mean ± standard deviation (s.d.). Quantitative data were analyzed using either parametric tests (Student’s t-test and analysis of variance) or nonparametric tests (Mann-Whitney U test and Kruskal-Wallis test). Qualitative data were analyzed using Fisher’s exact test. P < 0.05 was considered to indicate statistical significance. Statistical analyses were performed using SPSS 22.0 (IBM Corp., Armonk, NY, USA).

RESULTS

Characteristics of the enrolled individuals

Metabolomic analysis of EPS was performed on 23 CP/CPPS patients (except for one patient with insufficient EPS sample volume) and 25 ACs with EPS sample. Additionally, metabolomic analysis of plasma was conducted on 24 CP/CPPS patients and 10 other ACs who provided plasma samples. The characteristics of all participants are provided in Supplementary Table 2. Disease severity, as evaluated by the NIH-CPSI, was significantly different between CP/CPPS patients and ACs, and the subscores for pain, urination, and quality of life were also significantly different. Moreover, compared with ACs, CP/CPPS patients had higher SAS and SDS scores. Among the CP/CPPS patients, 6 patients achieved a cut-off for depression, 3 patients achieved a cut-off for anxiety, and 2 patients of the aforementioned individuals simultaneously experienced both anxiety and depression. There was no significant difference in BMI or WBC count between the CP/CPPS group and the AC group. This result further confirms that the WBC count of EPS is not related to prostatitis symptoms.30 Although age appeared to be different between all the enrolled patients and the 25 ACs, the age distribution was not different between the 23 CP/CPPS patients and 25 ACs who underwent metabolomic analysis of EPS (P = 0.064). This suggests that the two groups are comparable.

Supplementary Table 2.

Characteristics of enrolled chronic prostatitis/chronic pelvic pain syndrome patients and asymptomatic control individuals for expressed prostatic secretions and plasma metabolomics

Characteristic	CP/CPPS	AC for EPS metabolomics	AC for plasma metabolomics
Subject (n)	24	25	10
Age (year)	30.42±8.02	34.68±6.06*	31.30±5.62
BMI (kg m−2)	23.42±4.78	22.48±2.70	22.73±2.07
WBC count in EPS#
Occasional or few	8	14	-
1+	6	8	-
2+	3	3	-
3+	2	0	-
4+	5	0	-
NIH-CPSI
Pain	6.88±5.10	0*	0*
Urination	5.58±2.96	0.28±0.54*	0.30±0.68*
Quality of life	8.33±2.04	0.08±0.40*	0.10±0.32*
Total	20.79±7.67	0.36±0.86*	0.40±0.97*
SAS index	42.76±7.01	30.95±6.06*	30.63±4.80*
SDS index	45.27±12.09	32.35±7.82*	30.75±6.19*

*P<0.05, compared to the CP/CPPS group. ^#WBC count in EPS: occasional or few (0–9 counts per HP), 1+ (10–20 counts per HP), 2+ (21–30 counts per HP), 3+ (31–40 counts per HP), and 4+ (>40 counts per HP). CP/CPPS: chronic prostatitis/chronic pelvic pain syndrome; AC: asymptomatic control individuals; BMI: body mass index; EPS: expressed prostatic secretion; WBC: white blood cell; NIH-CPSI: National Institutes of Health - Chronic Prostatitis Symptom Index; SAS: the Zung’s Self-Rating Anxiety Scale; SDS: the Zung’s Self-Rating Depression Scale; HP: high-power

Metabolomic profiles of EPS differentiate CP/CPPS patients from control individuals

The total ion chromatograms (TICs) and correlation plots of the QC samples demonstrated the excellent repeatability and stability of the metabolomic detection (Supplementary Figure 1 ^{(250.6KB, tif)} ). A total of 959 metabolites were identified in the EPS (Figure 1a). PCA revealed a distinct separation of EPS metabolites between CP/CPPS and AC groups (Figure 1b). Moreover, the OPLS-DA model demonstrated obvious differences in the EPS metabolomic profiles between the two groups. Permutation analysis further validated the OPLS-DA model, which demonstrated strong interpretability and predictability (Figure 1c). As shown in the volcano plot, 78 metabolites were considered as DEMs (Figure 1d and Supplementary Table 3). Furthermore, 17 DEMs upregulated in the CP/CPPS group were enriched in the tryptophan metabolism pathway, pentose phosphate pathway, and thiamine metabolism pathway (Figure 1e), and 61 downregulated DEMs were enriched in the folate biosynthesis pathway, galactose metabolism pathway, steroid hormone biosynthesis pathway, and several pathways related to amino acid metabolism and energy metabolism, such as the citrate cycle (tricarboxylic acid cycle [TCA cycle]), arginine biosynthesis, pyrimidine metabolism, and D-glutamine and D-glutamate metabolism (Figure 1f).

Figure 1.

Widely targeted metabolomic analysis of EPS from CP/CPPS patients and asymptomatic control individuals. (a) The proportion of Class I chemical classifications of the identified metabolites in EPS. (b) Principal component analysis of the identified metabolites in EPS between CP/CPPS patients and AC individuals. (c) Orthogonal partial least squares-discriminant analysis of the identified metabolites in EPS between CP/CPPS patients and ACs. Permutation test was applied to validate the model. Perm R2Y reflects the interpretability, and Perm Q2 reflects the predictability. (d) Volcano plot of DEMs between CP/CPPS patients and ACs. Pathway analysis of (e) upregulated and (f) downregulated DEMs in EPS. EPS: expressed prostatic secretions; CP/CPPS: chronic prostatitis/chronic pelvic pain syndrome; AC: asymptomatic control; DEMs: differentially expressed metabolites; tRNA: transfer ribonucleic acid; PC: principal component; VIP: variable importance for the projection; TCA: tricarboxylic acid.

Supplementary Table 3.

Differentially expressed metabolites of expressed prostatic secretion from the comparison between chronic prostatitis/chronic pelvic pain syndrome patients and asymptomatic control individuals

Compounds	HMDB ID	Pubchem CID	VIP	Fold_Change	Type
1-Methylxanthine	HMDB0010738	80220	1.293	5.149	Up
Riboflavin	HMDB0000244	493570	2.341	3.301	Up
Dl-2-Aminooctanoic acid	HMDB0000991	69522	2.146	2.901	Up
3-Hydroxyhippuric acid	HMDB0006116	450268	1.134	2.278	Up
Gluconolactone	HMDB0000150	7027	2.338	3.178	Up
Isobutyric acid	HMDB0001873	6590	1.289	3.751	Up
4-Hydroxyhippurate	HMDB0013678	151012	1.507	2.151	Up
3-Hydroxypicolinic acid	HMDB0013188	13401	2.234	3.216	Up
Tryptamine	HMDB0000303	1150	2.382	4.402	Up
N-Acetyl-L-Leucine	HMDB0011756	70912	2.328	2.079	Up
Thiamine	HMDB0000235	1130	2.383	3.075	Up
Cyclo (pro-pro)	NA	529063	1.309	2.074	Up
2-Chloroadenosine	NA	8974	3.190	12.963	Up
Cysteinyldopa	METPA1455	96024156	1.968	3.186	Up
Urolithin A	HMDB0013695	5488186	1.112	2.195	Up
4-(aminomethyl) indole	NA	280302	2.475	2.044	Up
3-(4-fluorobenzoyl) propionic acid	NA	101359	2.886	3.703	Up
N-Acetyl-L-Tyrosine	HMDB0000866	68310	2.168	0.289	Down
1,5-Anhydro-D-glucitol	HMDB0002712	64960	3.021	0.407	Down
5-Methyl-2’- deoxycytidine	HMDB0002224	440055	2.148	0.191	Down
Stachyose	HMDB0003553	439531	2.814	0.172	Down
Dihydrocaffeic acid	HMDB0000423	348154	3.117	0.318	Down
Trp-Leu	HMDB0029087	6997510	3.114	0.290	Down
Ile-Ala	HMDB0028900	7009577	3.475	0.098	Down
PI (36:2)	HMDB0009786	138199526	3.504	0.024	Down
Sulfoacetic acid	METPA1121	31257	1.210	0.174	Down
Trp-Tyr	HMDB0029095	88184	3.149	0.196	Down
Gln-Gln	HMDB0028795	7010588	2.473	0.394	Down
Val-Cys	HMDB0029124	10421020	3.195	0.221	Down
Asn-Val	HMDB0028744	7019993	3.191	0.216	Down
Asn-Ala	HMDB0028724	9942455	2.646	0.267	Down
L-Asparagine Anhydrous	HMDB0000168	6267	1.127	0.094	Down
L-Glutamine	HMDB0000641	5961	1.100	0.263	Down
N’- Formylkynurenine	HMDB0001200	910	2.673	0.285	Down
L-Sepiapterin	HMDB0000238	135398579	2.776	0.148	Down
6-O-Methylguanine	NA	65275	2.039	0.454	Down
6-Dimethylaminopurine	HMDB0000473	3134	3.157	0.301	Down
N1, N8-diacetylspermidine	HMDB0041947	389613	1.112	0.370	Down
Cyromazine	HMDB0029862	47866	2.946	0.246	Down
Glu-Ile	HMDB0028822	9813855	2.934	0.023	Down
Ile-Glu	HMDB0028906	7009625	2.186	0.035	Down
Gly-Thr	HMDB0028851	111257	2.333	0.257	Down
N-(2-hydroxyethyl)-3-pyridinecarboxamide	NA	72663	2.665	0.444	Down
Lys-His	HMDB0028953	9857010	3.092	0.107	Down
2,2’- Cyclouridine	NA	806138	2.633	0.266	Down
Ser-Ser	HMDB0029048	7019105	2.812	0.097	Down
Tyr-Ser	HMDB0029114	54564570	2.404	0.283	Down
Caffeic acid	HMDB0001964	1549111	1.097	0.422	Down
Cyclamic acid	HMDB0031340	7533	1.643	0.128	Down
Prostaglandin D1	HMDB0005102	5280936	1.352	0.355	Down
Prostaglandin E1	HMDB0001442	5280723	1.352	0.355	Down
Prostaglandin E2	HMDB0001220	5280360	1.260	0.321	Down
PGF1α	HMDB0002685	5280939	2.066	0.319	Down
Prostaglandin F2α	HMDB0001139	NA	2.134	0.319	Down
Prostaglandin J2	HMDB0002710	5280884	1.884	0.186	Down
7-ketodeoxycholic acid	HMDB0000391	188292	1.010	0.436	Down
3-(pyrazol-1-yl) -L-alanine	NA	151491	1.101	0.273	Down
Prostaglandin B2	HMDB0004236	5280881	1.884	0.186	Down
11β-Prostaglandin E2	HMDB0060041	5283061	1.752	0.330	Down
8-iso Prostaglandin F2α	HMDB0005083	5282263	1.362	0.305	Down
11β-Prostaglandin F2α	HMDB0010199	5280886	1.362	0.305	Down
Isocitric acid	HMDB0000193	1198	1.477	0.392	Down
Hydroxypiperazic acid	NA	129859924	2.563	0.175	Down
3-Methylaspartate	NA	852	1.261	0.348	Down
ent-Prostaglandin F2α	NA	35027306	1.042	0.371	Down
8-iso Prostaglandin E1	HMDB0004686	5283212	1.042	0.371	Down
20-hydroxy Prostaglandin F2α	NA	5283040	1.724	0.300	Down
8-iso Prostaglandin F2β	HMDB0002115	5283216	1.362	0.305	Down
15(R)-prostaglandin E1	NA	5283056	1.106	0.488	Down
3,4,5-Trimethoxycinnamic acid	HMDB0002511	735755	2.398	0.398	Down
Indole	HMDB0000738	798	2.785	0.145	Down
2-Aminomethylpyrimidine	NA	17848325	1.091	0.228	Down
Aldosterone	HMDB0000037	5839	2.390	0.243	Down
Carnitine C7:DC	NA	NA	1.004	0.380	Down
Asp-Tyr	HMDB0028765	152455	1.132	0.306	Down
8-iso Prostaglandin E2	HMDB0005844	5283213	2.672	0.304	Down
15-keto Prostaglandin E1	HMDB0001320	5280710	2.672	0.304	Down
BW-245C	NA	3080928	2.244	0.288	Down

CP/CPPS: chronic prostatitis/chronic pelvic pain syndrome; EPS: expressed prostatic secretion; HMDB: human metabolome database; VIP: variable importance in projection; NA: not available

Construction of an EPS metabolite coexpression network

By utilizing WGCNA, a total of 21 metabolite modules exhibited strong coexpression levels among the EPS metabolites. A total of 7 metabolic modules exhibited significant correlations with the clinical traits of CP/CPPS patients (Figure 2a). The categorization of metabolites within each differential module is presented in Figure 2b. Notably, the pink, light cyan, and cyan modules clearly distinguished CP/CPPS patients from ACs (Figure 2c–2e), and the green, red, turquoise, and midnight blue modules were significantly associated with the CPSI-pain score, CPSI-total score, SAS score, and WBC count, respectively. The negatively correlated cyan module predominantly consisted of small peptides, heterocyclic compounds, and sugars (Figure 2b). Phosphatidylcholine (PC), phosphatidylethanolamine (PE), and sphingomyelin (SM) were primarily observed in the positively correlated pink and light cyan modules. Notably, the green module, which was positively correlated with pain score, predominantly consisted of oxidized lipids.

Figure 2.

The coexpression network of EPS metabolites from CP/CPPS patients and asymptomatic control individuals. (a) The correlation between EPS module eigenmetabolites and different phenotypes of CP/CPPS. (b) The distribution of metabolites in differential modules according to Class II chemical classifications. Eigenmetabolite value analysis of (c) pink, (d) light cyan, and (e) cyan modules between CP/CPPS and control individuals. (f) The river plot of the enriched pathways of metabolites in significantly differential modules. Line thickness represents enrichment significance. ^*P < 0.05, ^***P < 0.001, and ^****P < 0.0001. EPS: expressed prostatic secretions; CP/CPPS: chronic prostatitis/chronic pelvic pain syndrome; AC: asymptomatic control; CPSI: Chronic Prostatitis Symptom Index; WBC: white blood cell; SAS: the Zung’s Self-Rating Anxiety Scale; SDS: the Zung’s Self-Rating Depression Scale.

Pathway analysis revealed that metabolites of the positively correlated pink and light cyan modules were significantly enriched in tryptophan metabolism and several pathways of lipid metabolism. The metabolites of the negatively correlated cyan module were mainly enriched in folate biosynthesis and several pathways of carbohydrate metabolism, which may indicate the failure of the energy supply in the prostate microenvironment of CP/CPPS patients (Figure 2f). Furthermore, the metabolites in other positively correlated modules were also enriched in pathways related to lipid and carbohydrate metabolism and played a part in tyrosine and phenylalanine metabolism (Supplementary Figure 2a ^{(308KB, tif)} ). The turquoise module was the largest negatively correlated module. This module was mainly enriched in amino acid metabolism-related pathways, which was basically consistent with the enriched pathways of the downregulated DEMs (Supplementary Figure 2b ^{(308KB, tif)} ).

Key metabolites and pathways in EPS associated with severity of CP/CPPS

While the DEMs effectively distinguished CP/CPPS patients from ACs, the severity of particular symptoms, such as pain, urination, and psychiatric symptoms, cannot be fully elucidated. Considering WGCNA maximizes the utilization of metabolomic data, we defined the EPS metabolites in the differential modules between CP/CPPS patients and ACs as differential-module metabolites (DMMs) and attempted to further explore the key metabolites in DMMs affecting the severity of particular symptoms. The partial correlation analysis of DMMs with each clinical index was further performed among patients (Figure 3a). Agmatine and (+/-)-high proline exhibited significant positive correlations with the CPSI and pain score (all P < 0.05). Additionally, 3-hydroxyanthranilic acid and 3-guanidinopropionic acid exhibited significantly negative correlations with the CPSI and urination score (P < 0.05). Notably, multiple oxidized lipids were positively associated with the pain score, which is consistent with previous findings. The full list of the EPS phenotype-correlated metabolites among patients is presented in Supplementary Table 4. Moreover, pain-correlated metabolites were predominantly enriched in lipid-related pathways, and urination-correlated metabolites were primarily enriched in pathways associated with amino acid metabolism (Supplementary Figure 2c ^{(308KB, tif)} ). Furthermore, the WBC count was found to be significantly associated with carbohydrate-related pathways. These results suggest that the underlying mechanisms inducing the diverse symptoms of CP/CPPS may be different. However, EPS metabolites correlated with SAS and SDS were insufficient for pathway enrichment analysis.

Figure 3.

Identification of key metabolites and pathways in EPS influencing various symptoms of CP/CPPS. (a) Scatter plots showing the relationships between four clinical indicators and DMMs in patients. The horizontal dashed line is the dividing line with a P value of 0.05, and the vertical dashed line is the dividing line with the correlation coefficient |r| = 0.5. The relative intensity of (b) 5-HT, (c) 5-HIAL, (d) IAA, (e) 3-HAA, and (f) XA among high-CPSI patients (high), low-CPSI patients (low) and AC individuals. (g) AUC values of the 5 specific metabolites of EPS for distinguishing CP/CPPS patients from ACs. (h) AUC values of the 5 specific metabolites of EPS for distinguishing high-CPSI patients from low-CPSI patients. ^*P < 0.05 and ^**P < 0.01. EPS: expressed prostatic secretions; CP/CPPS: chronic prostatitis/chronic pelvic pain syndrome; AC: asymptomatic control; DMMs: differential-module metabolites; CPSI: Chronic Prostatitis Symptom Index; WBC: white blood cell; AUC: area under the curve; CI: confidence interval; 3-HAA: 3-hydroxyanthranilic acid; 5-HT: 5-hydroxytryptamine (serotonin); XA: xanthurenic acid; 5-HIAL: 5-hydroxyindole acetaldehyde; IAA: indole-3-acetic acid.

Supplementary Table 4.

The significant phenotypecorrelated metabolites of expressed prostatic secretion among chronic prostatitis/chronic pelvic pain syndrome patients

Type	Compounds	Correlation	tstatistics	P
CPSIcor	3Hydroxyanthranilic acid	−0.669	−3.919	0.001
CPSIcor	(+/)High Proline	0.614	3.389	0.003
CPSIcor	Pyrrole2Carboxylic acid	−0.597	−3.247	0.004
CPSIcor	Uridine	−0.586	−3.154	0.005
CPSIcor	Carnitine C6:0	−0.556	−2.919	0.009
CPSIcor	Carnitine C7:DC	−0.556	−2.913	0.009
CPSIcor	3Guanidinopropionic acid	−0.532	−2.735	0.013
CPSIcor	Cortisol	−0.522	−2.666	0.015
CPSIcor	Agmatine	0.516	2.624	0.017
CPSIcor	Creatine	−0.498	−2.505	0.022
CPSIcor	ValPheAla	−0.497	−2.499	0.022
CPSIcor	Glucerol?3phosphate	0.486	2.424	0.025
CPSIcor	IleLys	−0.486	−2.423	0.026
CPSIcor	N, N’ Diacetyl Chitobiose	0.451	2.204	0.040
CPSIcor	Galactitol	0.445	2.164	0.043
CPSIcor	ADPribose	0.440	2.134	0.046
CPSIcor	NAcetylbetaalanine	−0.436	−2.109	0.048
CPSIcor	Lactulose	−0.434	−2.101	0.049
Paincor	Biopterin	−0.657	−3.695	0.002
Paincor	Lactulose	−0.650	−3.633	0.002
Paincor	Agmatine	0.588	3.085	0.006
Paincor	(+/)High Proline	0.572	2.961	0.008
Paincor	GlyThr	−0.514	−2.544	0.020
Paincor	3Hydroxyanthranilic acid	−0.481	−2.325	0.032
Paincor	PC (22:6 (4Z,7Z,10Z,13Z,16Z,19Z)/20:1 (11Z))	0.474	2.286	0.035
Paincor	8HDoHE	0.460	2.198	0.041
Paincor	13HDoHE	0.457	2.181	0.043
Paincor	10HDoHE	0.457	2.181	0.043
Paincor	11HDoHE	0.457	2.181	0.043
Paincor	Biliverdin	−0.449	−2.134	0.047
Paincor	14(S)HDHA	0.449	2.133	0.047
Paincor	Carnitine C7:DC	−0.446	−2.112	0.049
Paincor	(±) 17HDHA	0.444	2.102	0.050
Urinationcor	GlyThr	0.618	3.338	0.004
Urinationcor	3Hydroxy2methyl4Hpyran4one	0.617	3.328	0.004
Urinationcor	Creatine	−0.615	−3.309	0.004
Urinationcor	3Hydroxyanthranilic acid	−0.614	−3.299	0.004
Urinationcor	3Guanidinopropionic acid	−0.610	−3.262	0.004
Urinationcor	Sarcosine	−0.603	−3.205	0.005
Urinationcor	Pseudotropine	0.602	3.200	0.005
Urinationcor	Pyrrole2Carboxylic acid	−0.592	−3.115	0.006
Urinationcor	AspIle	0.591	3.106	0.006
Urinationcor	1Acetylindole	−0.579	−3.013	0.007
Urinationcor	2Aminoethanesulfinic acid	−0.578	−3.006	0.008
Urinationcor	Caffeine	0.555	2.833	0.011
Urinationcor	Carnitine C6:0	−0.554	−2.826	0.011
Urinationcor	GlyTrp	−0.553	−2.815	0.011
Urinationcor	1Hydroxylamino2phenylethane	−0.553	−2.813	0.012
Urinationcor	(R)()2phenylglycine	−0.553	−2.813	0.012
Urinationcor	LTryptophanamide	−0.547	−2.773	0.013
Urinationcor	NAcetylbetaalanine	−0.542	−2.737	0.014
Urinationcor	Xanthopterin	0.541	2.732	0.014
Urinationcor	N(1Deoxy1fructosyl) phenylalanine	0.537	2.698	0.015
Urinationcor	1,3Diphenylguanidine	0.524	2.610	0.018
Urinationcor	LTyrosine	0.523	2.606	0.018
Urinationcor	AspLeu	0.516	2.558	0.020
Urinationcor	LLeucine	−0.516	−2.557	0.020
Urinationcor	Uridine	−0.514	−2.543	0.020
Urinationcor	LIsoleucine	−0.514	−2.542	0.020
Urinationcor	5Hydroxy2’ deoxyuridine	−0.505	−2.485	0.023
Urinationcor	DGalacturonic acid	0.505	2.484	0.023
Urinationcor	Hydroxyquinoline	−0.493	−2.407	0.027
Urinationcor	LNorleucine	−0.490	−2.383	0.028
Urinationcor	TrpGlu	−0.486	−2.357	0.030
Urinationcor	Indoleacetaldehyde	−0.485	−2.351	0.030
Urinationcor	2,6pyridinedicarboxylic acid	−0.485	−2.350	0.030
Urinationcor	LAlanine	−0.480	−2.319	0.032
Urinationcor	Benzaldehyde	−0.461	−2.201	0.041
Urinationcor	Theobromine	0.459	2.193	0.042
Urinationcor	1amino3,3diethoxypropane	−0.459	−2.190	0.042
Urinationcor	LGlutamine	−0.453	−2.155	0.045
Urinationcor	NMyristoylglycine	−0.450	−2.135	0.047
Urinationcor	LTryptophan	−0.449	−2.132	0.047
Urinationcor	N, NDimethylglycine	0.447	2.118	0.048
WBCcor	DMaltopentaose	−0.722	−4.550	0.000
WBCcor	2’ Omethyladenosine	−0.662	−3.855	0.001
WBCcor	2’ OMethyluridine	−0.642	−3.653	0.002
WBCcor	LPhenylalanine	−0.588	−3.166	0.005
WBCcor	6’ Sialyllactose	0.579	3.097	0.006
WBCcor	Carnitine C6:0	0.578	3.087	0.006
WBCcor	PE (22:6 (4Z,7Z,10Z,13Z,16Z,19Z)/P16:0)	−0.571	−3.028	0.007
WBCcor	Sedoheptulose	−0.569	−3.013	0.007
WBCcor	NMethy4HydroxyProline	−0.565	−2.984	0.008
WBCcor	LIditol	−0.550	−2.869	0.010
WBCcor	Carnitine C4:0	0.543	2.818	0.011
WBCcor	()alphaTerpineol	−0.542	−2.813	0.011
WBCcor	DMannitol	−0.540	−2.800	0.011
WBCcor	15(R)prostaglandin E1	0.535	2.762	0.012
WBCcor	2’ Omethylguanosine	−0.530	−2.723	0.013
WBCcor	8,15Dihete	−0.526	−2.693	0.014
WBCcor	DTagatose	0.523	2.678	0.015
WBCcor	3Methoxybenzoic acid	−0.520	−2.654	0.016
WBCcor	Carnitine C2:0	0.518	2.642	0.016
WBCcor	Securinine	0.513	2.603	0.017
WBCcor	Carnitine isoC4:0	0.509	2.580	0.018
WBCcor	AspTyr	0.509	2.579	0.018
WBCcor	PE (20:4 (5Z,8Z,11Z,14Z)/P18:0)	−0.505	−2.552	0.019
WBCcor	13oxoODE	−0.503	−2.535	0.020
WBCcor	9oxoODE	−0.503	−2.535	0.020
WBCcor	NicotinamideNOxide	−0.498	−2.506	0.021
WBCcor	Galactitol	−0.498	−2.502	0.022
WBCcor	Carnitine 2methylC4	0.494	2.479	0.023
WBCcor	Carnitine C5:0	0.494	2.479	0.023
WBCcor	(2S,3S)3methylphenylalanine	−0.492	−2.461	0.024
WBCcor	Carnitine C7:DC	0.486	2.427	0.025
WBCcor	DFructose	0.483	2.403	0.027
WBCcor	NAcetyl5Hydroxytryptamine	−0.479	−2.382	0.028
WBCcor	Guanosine 5’ diphosphate	−0.478	−2.373	0.028
WBCcor	Methyl ptertbutylphenylacetate	−0.477	−2.363	0.029
WBCcor	PE (20:4 (5Z,8Z,11Z,14Z)/P18:1 (11Z))	−0.472	−2.335	0.031
WBCcor	2(4Hydroxyphenyl) ethanol	−0.469	−2.312	0.032
WBCcor	Prostaglandin E2	0.464	2.283	0.034
WBCcor	DSorbitol	−0.457	−2.237	0.037
WBCcor	3Hydroxyphenylacetic acid	−0.452	−2.207	0.040
WBCcor	5’ Deoxyadenosine	−0.452	−2.206	0.040
WBCcor	3’ Deoxyadenosine	−0.452	−2.206	0.040
WBCcor	3(Methylthio)1propanol	−0.451	−2.204	0.040
WBCcor	NAcetylornithine	−0.448	−2.184	0.042
WBCcor	13(R)HODE	−0.447	−2.180	0.042
WBCcor	3Methylaspartate	0.445	2.166	0.043
WBCcor	SerVal	0.444	2.158	0.044
WBCcor	Lithocholic acid	−0.443	−2.153	0.044
WBCcor	Allocholic acid	−0.443	−2.153	0.044
WBCcor	NMyristoylglycine	−0.441	−2.139	0.046
WBCcor	LAlanylLLysine	0.439	2.132	0.046
WBCcor	S(5Adenosy)LHomocysteine	−0.439	−2.129	0.047
SAScor	SM (d16:1/24:1 (15Z))	−0.617	−3.323	0.004
SAScor	PyrGlu	−0.595	−3.143	0.006
SAScor	AspIle	−0.539	−2.718	0.014
SAScor	2,4diacetamino2,4,6triphenoxyDmannopyranose	−0.525	−2.617	0.017
SAScor	AsnAla	−0.511	−2.519	0.021
SAScor	IleAsp	−0.502	−2.465	0.024
SAScor	Deoxycytidine	0.465	2.230	0.039
SAScor	GluIle	−0.460	−2.195	0.041
SAScor	2Pyrrolidinone	−0.458	−2.185	0.042
SAScor	LPC (13:0/0:0)	0.447	2.121	0.048
SDScor	Isobutyric acid	0.592	3.115	0.006
SDScor	2,2Dihydroxymethyl1azabicyclo[2.2.2]octan3one	0.494	2.411	0.027
SDScor	GluIle	−0.447	−2.123	0.048
SDScor	NAcetylDGalactosamine	0.444	2.104	0.050
SDScor	Carnitine C7OH	0.444	2.103	0.050

CPSI: chronic prostatitis symptom index; WBC: white blood cell; SAS: the Zung’s SelfRating Anxiety Scale; SDS: the Zung’s SelfRating Depression Scale; Cor: correlation

Importantly, metabolites correlated with those indices were all enriched in tryptophan metabolism (Supplementary Figure 2c ^{(308KB, tif)} ). Moreover, five downstream products of tryptophan exhibited consistent changes in symptom severity across high-CPSI patients, low-CPSI patients, and ACs (Figure 3b–3f). Each of these metabolites exhibited effectiveness in distinguishing CP/CPPS patients from ACs, as well as in distinguishing high-CPSI patients from low-CPSI patients. More importantly, combinations of the five metabolites demonstrated superior diagnostic capability (Figure 3g and 3h). These findings indicate that tryptophan metabolism within the prostate microenvironment plays a crucial role in exacerbating the symptoms of CP/CPPS, and the tryptophan metabolites in EPS are potential indicators for measuring the severity of CP/CPPS.

Key alterations in plasma metabolomics affecting CP/CPPS symptoms

To establish a comprehensive metabolomic profile of CP/CPPS, a total of 746 metabolites were identified in plasma (Figure 4a). The PCA plot displayed a separation of plasma metabolites between CP/CPPS patients and ACs (Figure 4b). However, the permutation analysis of the OPLS-DA model demonstrated limited predictability between the two groups (Figure 4c). A total of 45 plasma metabolites were considered as DEMs (Figure 4d and Supplementary Table 5). The DEMs upregulated in the CP/CPPS group were mostly involved in cysteine and methionine metabolism (Figure 4e), and the downregulated DEMs were mainly involved in arginine and proline metabolism and valine, leucine, and isoleucine metabolism (Figure 4f).

Figure 4.

Widely targeted metabolomic analysis of plasma from CP/CPPS patients and asymptomatic control individuals. (a) The proportion of Class I chemical classifications of the identified metabolites in plasma. (b) Principal component analysis of the identified metabolites in plasma between CP/CPPS patients and AC individuals. (c) Orthogonal partial least squares-discriminant analysis of the identified metabolites in plasma between CP/CPPS patients and ACs. Permutation test was applied to validate the model. Perm R2Y reflects the interpretability, and Perm Q2 reflects the predictability. (d) Volcano plot of DEMs between CP/CPPS patients and ACs. Pathway analysis of (e) upregulated and (f) downregulated DEMs in plasma. CP/CPPS: chronic prostatitis/chronic pelvic pain syndrome; AC: asymptomatic control; DEMs: differentially expressed metabolites; PC: principal component; CoA: coenzyme A.

Supplementary Table 5.

Differentially expressed metabolites of plasma from the comparison between chronic prostatitis/chronic pelvic pain syndrome patients and asymptomatic control individuals

Compounds	HMDB ID	Pubchem CID	P	Fold_Change	Type
12,13-EpOME	HMDB0004702	5356421	0.001	2.187	Up
3-Hydroxy-4-methoxybenzoic acid	HMDB0060003	12575	0.000	2.256	Up
3-Methoxysalicylic acid	HMDB0059713	70140	0.000	2.256	Up
5’- Deoxy-5’- (Methylthio) Adenosine	HMDB0001173	439176	0.000	2.120	Up
5-Methoxysalicylic acid	HMDB0001868	75787	0.000	2.256	Up
6-Dimethylaminopurine	HMDB0000473	3134	0.002	2.136	Up
9,10-EpOME	HMDB0004701	6246154	0.001	2.187	Up
Asn-Val	HMDB0028744	7019993	0.001	2.314	Up
Carnitine C7:1	NA	NA	0.004	2.417	Up
Cytidine 5’- Diphosphocholine	HMDB0001413	13804	0.001	2.273	Up
LPC (14:1/0:0)	HMDB0010380	NA	0.000	2.290	Up
Lys-Lys	HMDB0028956	128837	0.003	2.369	Up
MG (18:1/0:0/0:0)	NA	5283468	0.000	3.072	Up
N-Methy-4-Hydroxy-Proline	NA	11768700	0.047	2.758	Up
(3S,4S,5R)-1,3,4,5,6-Pentahydroxyhexan-2-one	HMDB0003418	92092	0.000	0.336	Down
1,3-Dicyclohexylurea	NA	4277	0.000	0.249	Down
1-Methylxanthine	HMDB0010738	80220	0.039	0.275	Down
2-Mercaptobenzothiazole	HMDB0030524	17395437	0.009	0.457	Down
2’- O-methyladenosine	HMDB0004326	102213	0.022	0.469	Down
2-thiophene acrylic acid	NA	735981	0.005	0.320	Down
3-Methyl-2-Oxobutanoic acid	HMDB0000019	49	0.000	0.221	Down
3-N-Methyl-L-Histidine	HMDB0000479	64969	0.000	0.429	Down
3-pyridine aldoxime	NA	5324568	0.000	0.150	Down
4-Acetamidobutyric acid	HMDB0003681	18189	0.000	0.300	Down
4-toluenesulfonic acid	HMDB0059933	6101	0.000	0.151	Down
Carnitine C18:2	NA	NA	0.000	0.481	Down
Carnitine ph-C14	NA	NA	0.000	0.438	Down
Creatinine	HMDB0000562	588	0.004	0.384	Down
Cyclo (Ala-Pro)	NA	13879951	0.001	0.283	Down
Hypoxanthine	HMDB0000157	790	0.006	0.400	Down
Ile-Arg	HMDB0028901	7021814	0.000	0.033	Down
Maleamic acid	METPA0190	5280451	0.000	0.036	Down
MG-132	NA	462382	0.017	0.488	Down
Monobutyl phthalate	HMDB0013247	8575	0.000	0.486	Down
N-acetyl-D-phenylalanine	METPA0627	101184	0.000	0.122	Down
N-Cinnamylglycine	HMDB0011621	709625	0.030	0.332	Down
N-Methyl-4-aminobutyric acid	NA	70703	0.000	0.051	Down
Phe-Arg	HMDB0028989	150903	0.001	0.223	Down
Pro-Arg	HMDB0029011	151004	0.000	0.236	Down
Riboflavin	HMDB0000244	493570	0.002	0.284	Down
Shinorine	HMDB0039002	101926676	0.001	0.492	Down
S-Sulfo-L-Cysteine	HMDB0000731	115015	0.000	0.198	Down
Taurolithocholic acid	HMDB0000722	23662757	0.009	0.323	Down
tryptophan betaine	HMDB0061115	442106	0.002	0.302	Down
Xanthine	HMDB0000292	1188	0.013	0.237	Down

HMDB: human metabolome database; NA: not available

Moreover, WGCNA illustrated that several plasma modules exhibited a significant association with SAS and SDS scores (Supplementary Figure 3a ^{(596.4KB, tif)} ). These results suggest that plasma metabolites exhibit greater explanatory power for psychiatric symptoms than EPSs. Notably, the light cyan, cyan, salmon, and green-yellow modules, which exhibited a negative association with CP/CPPS and psychiatric symptoms, predominantly comprised phospholipids (Supplementary Figure 3b ^{(596.4KB, tif)} ). Three modules independently exhibited significant associations with psychiatric symptoms. The royal blue and black modules that were positively correlated with SDS were enriched in tyrosine metabolism, phenylalanine metabolism, and tryptophan metabolism pathways, whereas the light cyan module that was negatively correlated with SAS was enriched in several lipid-related metabolism pathways (Supplementary Figure 3c ^{(596.4KB, tif)} ). In particular, the dysregulation of the tryptophan and tyrosine metabolic pathways has been widely reported as a pathogenic mechanism underlying inflammation-induced depression.31,32,33

Key metabolites and pathways in plasma affecting psychiatric symptoms

Considering the significant association between the plasma metabolome and psychiatric symptoms, we conducted further investigations to identify the key plasma metabolites that influence psychiatric symptoms, particularly depression. Partial correlation analyses of plasma metabolites with SAS and SDS scores were performed (Figure 5a). The full list of the plasma phenotype-correlated metabolites is displayed in Supplementary Table 6. In line with previous findings, several metabolites in the tyrosine and tryptophan metabolism pathways, including DL-O-tyrosine, 3,3’,5-triiodo-L-thyronine, L-thyronine, 3-O-methyldopa, kynurenine, and indole-3-acetamide, demonstrated a positive correlation with psychiatric symptoms. However, a noteworthy inverse relationship was observed between psychiatric symptoms and lysophospholipids. The SAS- and SDS-correlated metabolites were both significantly enriched in nitrogen metabolism, D-glutamine and D-glutamate metabolism, and aminoacyl-transfer RNA (tRNA) biosynthesis pathways, which indicates systemic antioxidant stress dysregulation in patients with psychiatric symptoms (Supplementary Figure 3d ^{(596.4KB, tif)} ). Furthermore, the participants were divided into 17 regular CP/CPPS patients without psychiatric symptoms (RCPs), 6 CP/CPPS patients with depression (DCPs) and 10 ACs. We focused on investigating the differential plasma metabolites in the tryptophan and tyrosine metabolic pathways among the 3 groups. Six specific plasma metabolites were markedly upregulated in the DCP group compared to the RCP and AC groups, which suggests that the imbalance in global tyrosine and tryptophan metabolism is linked to the development of depression in the CP/CPPS group (Figure 5b).

Figure 5.

Plasma metabolites associated with psychiatric symptoms in CP/CPPS patients. (a) Scatter plots showing the relationships between mental health-related indicators (SAS and SDS scores) and plasma metabolites among CP/CPPS patients and control individuals. The horizontal dashed line is the dividing line with a P value of 0.05, and the vertical dashed line is the dividing line with a correlation coefficient of |r| = 0.5. (b) The relative intensity of 6 differential plasma metabolites among DCPs, RCPs, and ACs. ^*P < 0.05 and ^**P < 0.01. CP/CPPS: chronic prostatitis/chronic pelvic pain syndrome; AC: asymptomatic control; DCP: CP/CPPS patients with depression; RCP: CP/CPPS patients without psychiatric symptoms; SAS: the Zung’s Self-Rating Anxiety Scale; SDS: the Zung’s Self-Rating Depression Scale; AA: arachidonic acid.

Supplementary Table 6.

The significant the Zung’s Self-Rating Anxiety Scale- and the Zung’s Self-Rating Depression Scale -correlated metabolites of plasma among all participants

Type	Compounds	Cor	t-statistics	P
SAS-cor	L- -Glutamyl-L-glutamic acid	0.569	3.727	0.001
SAS-cor	Acetanilide	0.560	3.641	0.001
SAS-cor	2,6-Di-tert-butyl-4-(hydroxymethyl) phenol	0.550	3.548	0.001
SAS-cor	Glu-Gln	0.544	3.489	0.002
SAS-cor	TriacetonaMine	−0.482	−2.961	0.006
SAS-cor	LPG (18:1/0:0)	−0.479	−2.939	0.006
SAS-cor	11-Beta-hydroxyandrostenedione	0.461	2.796	0.009
SAS-cor	(S)-2-Amino-3-(7-chloro-1H-indol-3-yl) propionic acid	−0.456	−2.761	0.010
SAS-cor	Cis-L-3-hydroxyproline	0.436	2.609	0.014
SAS-cor	Indole-3-acetamide	0.425	2.530	0.017
SAS-cor	2-amino-4-oxovaleric acid	0.423	2.517	0.018
SAS-cor	LPE (14:0/0:0)	−0.423	−2.512	0.018
SAS-cor	Thr-Gln	0.423	2.512	0.018
SAS-cor	Eicosanoyl-EA	0.419	2.485	0.019
SAS-cor	Palatinose	0.416	2.467	0.020
SAS-cor	L-Glutamine	0.407	2.402	0.023
SAS-cor	LPE (15:0/0:0)	−0.402	−2.361	0.025
SAS-cor	4-Acetamidobutyric acid	−0.399	−2.342	0.026
SAS-cor	L-Alanyl-L-Lysine	0.392	2.295	0.029
SAS-cor	5,6-Dimethylbenzimidazole	0.391	2.287	0.030
SAS-cor	alpha-Muricholic acid	−0.390	−2.283	0.030
SAS-cor	Glu-Met	0.385	2.247	0.032
SAS-cor	Met-Glu	0.385	2.247	0.032
SAS-cor	N-Acetyl-L-Histidine	0.380	2.211	0.035
SAS-cor	2,4-Dihydroxy-6-pentylbenzoic acid	0.378	2.196	0.036
SAS-cor	2-Hydroxyphenylacetic acid	0.374	2.170	0.038
SAS-cor	4-Hydroxy-3-methylbenzoic acid	0.374	2.170	0.038
SAS-cor	L-Tryptophanamide	0.372	2.157	0.039
SAS-cor	PC (16:0e/5,6-EET)	−0.371	−2.149	0.040
SAS-cor	Carnitine C4:0	0.367	2.122	0.043
SAS-cor	3-aminobutyric acid	0.366	2.121	0.043
SAS-cor	L-2-Aminobutyric acid	0.366	2.121	0.043
SAS-cor	1-Hydroxylamino-2-phenylethane	0.362	2.089	0.046
SAS-cor	(R)-(-)-2-phenylglycine	0.362	2.089	0.046
SAS-cor	2-Aminomethylpyrimidine	0.360	2.078	0.047
SAS-cor	2’- O-methyladenosine	−0.359	−2.073	0.047
SAS-cor	AA (Arachidonic acid)	0.358	2.064	0.048
SAS-cor	10-Formyl-Thf	0.357	2.061	0.048
SAS-cor	Folic acid	0.357	2.061	0.048
SAS-cor	Phe-Glu	0.356	2.052	0.049
SDS-cor	FFA (20:1)	−0.510	−3.192	0.003
SDS-cor	L-Thyroxine	0.500	3.109	0.004
SDS-cor	Aniline	0.495	3.070	0.005
SDS-cor	2-Hydroxyphenylacetic acid	0.467	2.842	0.008
SDS-cor	4-Hydroxy-3-methylbenzoic acid	0.467	2.842	0.008
SDS-cor	L-Glutamine	0.461	2.799	0.009
SDS-cor	AG-183	−0.446	−2.682	0.012
SDS-cor	Carnitine isoC4:0	0.439	2.634	0.013
SDS-cor	Monobutyl phthalate	−0.430	−2.562	0.016
SDS-cor	Zereno	0.428	2.552	0.016
SDS-cor	Indole-3-Carboxaldehyde	0.426	2.538	0.017
SDS-cor	2-amino-4-oxovaleric acid	0.425	2.525	0.017
SDS-cor	Pantetheine	0.422	2.509	0.018
SDS-cor	Carnitine C4:0	0.421	2.503	0.018
SDS-cor	Acetanilide	0.418	2.480	0.019
SDS-cor	AM-1235	0.412	2.438	0.021
SDS-cor	D-Arabinitol	0.405	2.387	0.024
SDS-cor	TriacetonaMine	−0.404	−2.380	0.024
SDS-cor	Galactitol	0.399	2.345	0.026
SDS-cor	2,4-Dihydroxy-6-pentylbenzoic acid	0.397	2.331	0.027
SDS-cor	Xanthurenic acid	0.396	2.319	0.028
SDS-cor	2,6-Di-tert-butyl-4-(hydroxymethyl) phenol	0.393	2.304	0.029
SDS-cor	N-Acetyl-L-Histidine	0.393	2.301	0.029
SDS-cor	Succinic acid	−0.387	−2.262	0.031
SDS-cor	1-deoxyvaleric acid	0.385	2.244	0.033
SDS-cor	2,6-pyridinedicarboxylic acid	0.384	2.241	0.033
SDS-cor	Methyl salinomycin	−0.381	−2.222	0.034
SDS-cor	LPC (20:2/0:0)	−0.376	−2.186	0.037
SDS-cor	LPC (0:0/20:2)	−0.376	−2.186	0.037
SDS-cor	9,10-dihydroxystearic acid	−0.375	−2.180	0.038
SDS-cor	Terephthalic acid	0.375	2.177	0.038
SDS-cor	Kynurenine	0.371	2.153	0.040
SDS-cor	LPG (18:1/0:0)	−0.371	−2.149	0.040
SDS-cor	L-Carnitine	0.369	2.135	0.041
SDS-cor	DL-O-tyrosine	0.368	2.132	0.042
SDS-cor	3,5-Dichlorosalicylic acid	0.367	2.124	0.042
SDS-cor	3-O-Methyldopa	0.366	2.118	0.043
SDS-cor	LPA (18:1/0:0)	−0.366	−2.118	0.043
SDS-cor	Quinoline-2-carboxylic acid	0.366	2.115	0.043
SDS-cor	3,3’,5-Triiodo-L-Thyronine	0.362	2.093	0.045
SDS-cor	SM (d34:2)	−0.357	−2.060	0.048
SDS-cor	Glycine	0.356	2.053	0.049

SAS: the Zung’s Self-Rating Anxiety Scale; SDS: the Zung’s Self-Rating Depression Scale; Cor: correlation

The use of plasma and EPS to investigate the key mechanisms underlying the aggravation of CP/CPPS

To investigate the associations between EPS and plasma metabolites, we conducted an integrated analysis that combined EPS and plasma metabolomic data obtained from the same CP/CPPS patient. The network plot revealed robust correlations between EPS and plasma metabolites with phenotypic indices, highlighting potential connections between them (Supplementary Figure 4 ^{(230.3KB, tif)} ). The metabolites that strongly correlated with urination symptoms and WBC counts were mainly EPS metabolites, whereas the metabolites that strongly correlated with SAS and SDS were mainly derived from plasma. Notably, pain symptoms exhibited strong associations with multiple plasma metabolites and appeared to have a potential connection with anxiety symptoms. Moreover, kynurenine and 3-hydroxyanthranilic acid, which are involved in tryptophan metabolism, were significantly correlated with several clinical indicators.

Previous findings suggest that local and systemic dysregulations of tryptophan metabolism contribute to the severity of multiple CP/CPPS symptoms. The kynurenine/tryptophan (KYN/TRP) ratio and serotonin/tryptophan (serotonin, is also known as 5-hydroxytryptamine [5-HT]/TRP) ratio have been associated with inflammation and depressive symptoms.31 Interestingly, the 5-HT/TRP ratio in the EPS was positively correlated with the CPSI, and the KYN/TRP ratio in the EPS was negatively correlated with the WBC count (Figure 6a). Moreover, the plasma KYN/TRP ratio and SDS score were also nearly significantly positively correlated. Indole-3-acetamide (IAM) is a downstream product of tryptophan that can be transformed by gut bacteria.34 Notably, the plasma IAM/TRP ratio was also positively correlated with the SAS score. However, there was no correlation between the KYN/TRP ratio and the 5-HT/TRP ratio in EPS. Moreover, the conversion of tryptophan in EPS did not influence the synthesis of kynurenine in plasma. These findings indicate more intricate regulatory crosstalk between systemic and localized dysregulation of tryptophan metabolism in CP/CPPS, especially the potential influence of the gut microbiota (Figure 6b).

Figure 6.

The EPS and plasma metabolites in the tryptophan metabolism pathway influence CP/CPPS symptoms. (a) Spearman’s correlation analysis between the CPSI score and the EPS serotonin/tryptophan (5-HT/TRP) ratio, WBC count and EPS kynurenine/tryptophan (KYN/TRP) ratio, SDS score and plasma KYN/TRP ratio, and SAS score and plasma indole-3-acetamide/tryptophan (IAM/TRP) ratio among all participants. (b) Schematic diagram of the crucial metabolic alterations influencing CP/CPPS symptoms. The color darkness defines the degree of up- or downregulation in EPS and plasma (red/blue and red/green) according to the log₂ (fold change). b were drawn using pictures from Servier Medical Art, which is licensed under a Creative Commons Attribution 3.0 unported license (https://creativecommons.org/licenses/by/3.0/). EPS: expressed prostatic secretions; CP/CPPS: chronic prostatitis/chronic pelvic pain syndrome; AC: asymptomatic control; DCP: CP/CPPS patients with depression; RCP: CP/CPPS patients without psychiatric symptoms; CPSI: chronic prostatitis symptom index; WBC: white blood cell; SAS: the Zung’s Self-Rating Anxiety Scale; SDS: the Zung’s Self-Rating Depression Scale; 5-HT: 5-hydroxytryptamine (serotonin); TRP: tryptophan; KYN: kynurenine; IAM: indole-3-acetamide; IAA: indole-3-acetic acid; NAD: nicotinamide adenine dinucleotide; TCA cycle: tricarboxylic acid cycle (citrate cycle); ROS: reactive oxygen species; ↓: downregulation; ↑: upregulation.

DISCUSSION

In this study, we analyzed widely targeted metabolomics data from EPS and plasma to shed light on the intricate manifestations of CP/CPPS. By performing different analyses, significant alterations in EPS and plasma metabolites were identified between CP/CPPS patients and ACs. Moreover, these analyses revealed specific metabolic pathways and metabolites that are associated with pelvic pain, voiding dysfunction, and mental health disorders associated with CP/CPPS.

Compared to those of ACs, the EPSs of CP/CPPS patients exhibited comprehensive dysregulation of tryptophan metabolism (Figure 6b). In particular, the serotonin (5-HT) pathway was positively associated with symptom exacerbation. 5-HT is detected in the prostate and plays an important role in the regulation of prostate function.35,36,37 Neuroendocrine cells in the human prostate are considered the primary source of 5-HT.38 Importantly, the prostate and urinary bladder smooth muscle contain multiple 5-HT receptors, leading to contraction of the prostate and bladder detrusor muscles.39,40 Moreover, inhibitors of 5-HT receptors decreased prostate contraction in vivo and inhibited human hyperplastic prostate cell growth in vitro.39,41 Neuroendocrine cells can be influenced by inflammation or oxidative damage, which further triggers the secretion of 5-HT. Consequently, inhibitors targeting 5-HT receptors in the prostate and bladder could serve as viable treatment options for alleviating CP/CPPS symptoms. Kynurenine and its downstream metabolites play important roles in inhibiting inflammation and regulating the immune response.42 Although there was a significant association between kynurenine and WBC count, no apparent association was observed between kynurenine and symptom improvement. Moreover, the levels of 3-hydroxyanthranilic acid, a potent anti-neuroinflammatory metabolite, were significantly decreased in CP/CPPS patients.43 Moreover, the concentration of xanthurenic acid, which is typically regarded as the end product of a nonfunctional branch, was significantly increased.44 These results may suggest dysfunction of the kynurenine pathway in the prostate microenvironment of CP/CPPS patients.

In line with the findings in EPS, tryptophan metabolism in plasma also plays a significant role in the progression of psychiatric symptoms. Notably, kynurenine was significantly associated with the SDS score and was elevated in the plasma of DCPs. Numerous previous studies have established that interferon-γ (IFN-γ), tumor necrosis factor-α (TNF-α), and interleukin-1β (IL-1β) trigger the catabolism of tryptophan by activating indoleamine 2,3-dioxygenase (IDO), resulting in the production of immune-active kynurenine.42 Extensive transport of kynurenine across the blood-brain barrier contributes to the development of mental health disorders.31 These results further validate the increased systemic inflammation in CP/CPPS patients. However, we did not observe that the release of 5-HT in EPS could impact systemic tryptophan levels and contribute to the development of depression. This indicates that the mechanisms influencing the homeostasis of tryptophan metabolism in CP/CPPS are likely to be more complex. Moreover, IAM, which was elevated in the plasma of DCPs, is primarily generated through the enzymatic conversion of tryptophan by gut bacteria.34 The levels of the downstream product of IAM, indole-3-acetic acid (IAA), were also increased in the EPS of CP/CPPS. Both IAM and IAA possess anti-inflammatory properties,34,45 which reflect an endogenous regulatory mechanism at the systemic level. As the majority of the downstream metabolites of tryptophan are produced in the gut, exploring the influence of the gut microbiota on the progression of CP/CPPS is crucial. Taken together, the above results affirm the significant role of tryptophan metabolism in regulating CP/CPPS symptoms. Furthermore, local and global tryptophan metabolites were validated as potential biomarkers for assessing the severity of CP/CPPS.

The TCA cycle within the prostate, which serves as a crucial metabolic pathway for generating energy, was significantly downregulated pathway in the EPS of the CP/CPPS patients. The TCA cycle in the luminal epithelial cells within the prostate is truncated, enabling these cells to generate, accumulate, and release high levels of citrate.46 This unique citrate-secretory phenotype of the prostate plays a key role in human reproduction, as it helps maintain the pH of semen, acts as a buffer for calcium ions, and actively participates in coagulation and decoagulation processes.46,47 Furthermore, our findings also confirmed significant downregulation of various amino acid- and carbohydrate-related metabolic pathways. Most of these amino acid pathways, including arginine, glutamine, and nitrogen metabolism pathways, play vital roles in sustaining cellular function, facilitating energy metabolism, and supporting antioxidative stress.48,49,50 More importantly, carbohydrates in semen, especially fructose, provide energy to the sperm, supporting its survival and swimming ability.51 The downregulation of TCA cycle function and amino acid and carbohydrate metabolism in the prostate microenvironment leads to increased reactive oxygen species (ROS), disruptions in energy supply, and reduced citrate production, which further reveals an underlying mechanism of reduced fertility in CP/CPPS.

We also found evidence of elevated systemic inflammation levels in CP/CPPS patients. Multiple amino acid metabolism pathways were dysregulated in the plasma of CP/CPPS patients. In particular, the level of 5’-deoxy-5’-(methylthio) adenosine was increased in the cysteine and methionine metabolism pathway, which might be associated with the systemic inflammatory response and T-cell function.48,52 Furthermore, several plasma metabolites exhibited significant correlations with CPSI scores, and particularly, with pain symptoms in patients. Among the positively correlated metabolites, N-ethylglycine has been shown to modulate glycinergic inhibition in pain signaling,53 and phenylacetyl-L-glutamine has been shown to be elevated in neurodegenerative diseases.54,55 Notably, among the negatively correlated metabolites, pantethine has been demonstrated to possess neuroprotective and anti-inflammatory properties,56 and adenosine 5’-monophosphate acts as a potent agonist of AMP-activated protein kinase (AMPK).57 Numerous AMPK activators have been shown to decrease pain behavior in animal models.58 These findings imply that heightened inflammation, neurological impairment, and disturbances in the pain receptor system could contribute to the exacerbation of CP/CPPS.

More importantly, the elevation in systemic inflammation levels is more pronounced in DCPs. In particular, elevated levels of DL-O-tyrosine and 3-O-methyldopa were observed in the plasma of DCPs. DL-O-tyrosine, an abnormal tyrosine isomer, can be generated during the oxidative process in which hydroxyl radicals oxidize the benzyl ring of phenylalanine. Moreover, an increase in the DL-O-tyrosine concentration has been employed as a biological marker of oxidative stress.59 Moreover, 3-O-methyldopa, a major metabolite of L-dopa, can impede blood-brain barrier penetration of L-dopa and diminish the neuroprotective effects of L-dopa through competitive inhibition.60,61 A decrease in L-dopa functionality may facilitate the occurrence of pathological changes associated with mental health disorders in the DCP group. Additionally, upregulation of L-thyroxine and 3,3,5-triiodo-L-thyronine was observed in the DCP group. According to previous reports, an increase in thyroxine and/or free thyroxine levels, although still within the conventional normal range, has been observed in patients with depression.62 Furthermore, numerous studies have shown that the thyroid axis is inhibited in depressed patients.62

Notably, the levels of multiple types of lipids, including PC, PE, and SM, and several oxidized lipids, were significantly increased in the EPS of CP/CPPS patients. Nevertheless, the plasma levels of these lipids in patients were significantly negatively correlated with CP/CPSS symptoms and psychiatric symptoms, marking a divergence from the observations in EPS. PC, PE, and SM are essential metabolites that serve as constituents of cell membranes and signaling molecules. Increased synthesis of PC, PE, and SM is thought to be linked to antiapoptotic and self-repair processes in inflammatory tissue, particularly in the context of neurogenesis.63,64 Additionally, ceramide, a derivative of SM, may trigger mitochondrial dysfunction, increase ROS levels, and result in cell death.65 Abnormal increases in PC, PE, and SM levels may play a direct role in the pathology of prostatitis, specifically in relation to nerve remodeling and sensitization. However, the systemic downregulation of phospholipids, especially lysophospholipids, has been consistently linked to the onset of depression in various studies.66 This association is related to its influence on neuronal plasticity and neurodevelopment. In summary, our results illustrate that CP/CPPS patients exhibit contrasting trends in local and systemic phospholipid metabolism in response to the inflammatory environment.

Moreover, the oxidation of several products of docosahexaenoic acid (DHA) was positively associated with pain score. Numerous studies have confirmed that oxidized lipids promote inflammatory pain by sensitizing nociceptors and enhancing pain perception in sensory neurons.67 However, the majority of oxidized lipids derived from DHA are typically regarded as having anti-inflammatory properties and reducing pain.68,69 Further research is necessary to determine the potential detrimental or advantageous effects of specific oxylipins on the progression of CP/CPPS.

We acknowledge some limitations in this study. First, this study was primarily limited by its small sample size. Thus, our results should be viewed as a basis for generating hypotheses and need to be validated in larger cohorts. Second, this study did not collect EPS or plasma from the same control source. Samples from the same individual source may provide more comprehensive insights into uncovering the link between EPS and plasma. Finally, further investigation and validation of these potential mechanisms using in vitro and in vivo models are needed.

In conclusion, this study contributes to the limited data in this field by determining EPS/plasma metabolome alterations in CP/CPPS. Our findings suggest important roles for the dysregulation of tryptophan metabolism, lipid metabolism, and the TCA cycle in CP/CPPS, shedding light on the oxidative damage, systemic inflammation, neurological impairment, and immunometabolism involved in the pathogenesis of CP/CPPS and the comorbidity of depression. These findings can serve as a valuable resource for future studies on the local-global relationships involved in CP/CPPS and the potential diagnostic and therapeutic approaches for CP/CPPS.

AUTHOR CONTRIBUTIONS

FXZ, XLS, and ZNM designed the study. FXZ, XC, DCN, and LC performed data analysis and wrote the paper. FXZ, XC, DCN, LC, CSH, ML, and YX carried out sample and data collection and revised the paper. All authors read and approved the final manuscript.

COMPETING INTERESTS

All authors declare no competing interests.

Supplementary Figure 1

Quality control of experimental data. The TIC of EPS QC samples in (a) T3 and (b) Hilic mode. (c) Pearson’s correlation of EPS QC samples; The TIC of plasma QC samples in (d) T3 and (e) Hilic mode. (f) Pearson’s correlation of plasma QC samples. QC: quality control; TIC: total ion chromatogram.

Supplementary Figure 2

River plots of enrichment pathways. (a) Enrichment pathways of EPS metabolites in green, midnight blue, and red module. (b) Enrichment pathways of EPS metabolites in turquoise module. (c) Enrichment pathways of significant phenotype-correlated metabolites (P < 0.05) in EPS. Line thickness represents enrichment significance. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, and ^****P < 0.0001. EPS: expressed prostatic secretions; CPSI: Chronic Prostatitis Symptom Index; WBC: white blood cell; cor: correlation.

Supplementary Figure 3

The coexpression network of plasma metabolites. (a) The correlation between plasma module eigenmetabolites and different phenotypes of CP/CPPS. (b) The distribution of metabolites in differential module according to the class II chemical classifications. (c) Enrichment pathways of metabolites in royal blue, black, and light cyan modules. (d) Enrichment pathways of significant SAS-correlated and SDS-correlated metabolites (P < 0.05) in plasma. Line thickness represents enrichment significance. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, and ^****P < 0.0001. CP/CPPS: chronic prostatitis/chronic pelvic pain syndrome; CPSI: Chronic Prostatitis Symptom Index; SAS: the Zung’s Self-Rating Anxiety Scale; SDS: the Zung’s Self-Rating Depression Scale; WBC: white blood cell; LPA: lysophosphatidic acid; LPC: lysophosphatidylcholine; LPC-O: lysoalkylphosphatidylcholine; LPG: lysophosphatidylglycerol; PC: phosphatidylcholine; PC-O: plasmanyl phosphatidylcholine; PC-P: plasmenyl phosphatidylcholine; PE: phosphatidylethanolamine; PS: phosphatidylserine; SM: sphingomyelin; cor: correlation.

Supplementary Figure 4

Network plot of EPS metabolites, plasma metabolites, and clinical phenotypes among CP/CPPS patients. Only the first-node metabolites with significant correlation were displayed (|r| > 0.55, and P < 0.05). Positive correlations are marked as red lines, while negative correlations are marked as blue lines. EPS: expressed prostatic secretions; CP/CPPS: chronic prostatitis/chronic pelvic pain syndrome; CPSI: Chronic Prostatitis Symptom Index; SAS: the Zung’s Self-Rating Anxiety Scale; SDS: the Zung’s Self-Rating Depression Scale; WBC: white blood cell.