Depression, anxiety, and stress impair sperm quality via dysregulation of the mitochondrial PDK–PDC axis

Abstract

Male infertility, a significant global public health challenge, arises from multifactorial interactions involving genetic, metabolic, and psychological factors. While psychological stress (depression/anxiety/stress, DAS) is increasingly linked to impaired sperm quality, its molecular mechanisms remain unclear. This cross-sectional study investigates mitochondrial metabolic reprogramming via the pyruvate dehydrogenase kinase (PDK)-pyruvate dehydrogenase complex (PDC) axis under psychological stress, exploring multidimensional impacts on semen quality.Approved by three tertiary hospitals in Central China, 557 participants were categorized into DAS and non-DAS groups based on DASS-21 scores, with clinical histories and semen samples collected. Targeted metabolomics (ATP/pyruvate/lactate), lipidomics (free fatty acids), and RT-qPCR (PDK1-4, PDH mRNA) were performed. Results revealed significantly reduced sperm motility in the DAS group (p < 0.01). Mechanistically, DAS upregulated PDK2/PDK4 expression and suppressed PDH expression, driving metabolic reprogramming and mitochondrial dysfunction. This study provides the first evidence that psychological stress impairs sperm motility via PDK-PDC axis dysregulation, offering novel mechanistic insights and potential therapeutic targets for male infertility.

Highlights

• First evidence of psychological stress (depression/anxiety/stress) impairing sperm motility through mitochondrial PDK-PDC axis dysregulation.

• Mechanistic breakthrough: Stress-induced upregulation of PDK2/PDK4 and suppression of PDH drive metabolic reprogramming, leading to ATP depletion and mitochondrial dysfunction.

• Multi-omics validation: Targeted metabolomics and lipidomics reveal reduced TCA cycle intermediates (e.g., citrate, α-ketoglutarate) and elevated free fatty acids (e.g., palmitic acid) in stressed sperm.

• Clinical significance: Identifies PDK-PDC axis as a potential therapeutic target for male infertility linked to psychological stress.

Introduction

Infertility, defined as the inability to achieve pregnancy after 12 months of unprotected intercourse, affects 8%–12% of couples worldwide, with male factors contributing to 40%–50% of cases [1]. Over the past five decades, a concerning decline in semen quality has been observed globally, with sperm concentration dropping by 52.4% and total sperm count by 59.3% in North America, Europe, and Australia [2]. This trend is associated with contemporaneous phenomena including rising obesity rates, increased exposure to environmental pollutants, and elevated psychological stress levels, suggesting underlying etiological mechanisms involving multifactorial interactions [3, 4]. Among these, psychological stress has emerged as a potential modifiable risk factor for male infertility, yet its biological underpinnings remain poorly understood [5, 6].

The burden of mental health disorders has reached epidemic proportions, with an estimated 970 million people suffering from anxiety or depressive disorders in 2022 [7]. Among men seeking fertility treatment, the prevalence of depression and anxiety is markedly elevated, ​with its prevalence being significantly higher than that in the general male population [8, 9]. Previous studies have reported inverse associations between depression and semen parameters, including semen volume, total sperm count, sperm concentration, and motility [10–12], with a mechanism potentially linked to depression-induced oxidative stress that compromises sperm quality [13, 14]. Notably, generalized anxiety disorder (GAD) has been proposed to impair testicular function through chronic stress pathways, ultimately affecting reproductive capacity [15]. These findings provide critical insights into the interplay between psychological factors and reproductive health. However, conflicting evidence suggests minimal or negligible effects of psychological stress on semen quality [16–19]. Such discrepancies may arise from methodological heterogeneity in stress assessment, population selection biases, and geographic variability, underscoring the need for standardized protocols and mechanistic investigations to elucidate causal relationships.

Sperm motility is a pivotal determinant of male fertility. Total motility refers to all motile spermatozoa (including both progressive and non-progressive motility), while progressive motility describes spermatozoa moving actively in straight lines or large circular paths at a minimum speed of 25 μm/s, according to the WHO Laboratory Manual (5th edition).The exceptional motility of spermatozoa necessitates extraordinarily high adenosine triphosphate (ATP) demands for epididymal maturation and post-ejaculatory movement [20]. Mitochondria, serving as the energy hub for sperm motility [21], generate ATP through dual metabolic pathways: glycolysis and oxidative phosphorylation (OXPHOS) [22, 23]. While numerous studies emphasize glycolysis as the primary energy source for mammalian sperm motility [24–26], accumulating evidence highlights the active involvement of fatty acid oxidation (FAO) in sperm energy metabolism and motility regulation [27–29], with FAO potentially surpassing glycolysis in dominance under specific physiological conditions [30]. The dynamic equilibrium of the mitochondrial PDK-PDC axis plays a critical role in maintaining metabolic flexibility. This axis regulates the efficiency of pyruvate conversion to acetyl-CoA, thereby directly modulating the competitive interplay between FAO and glucose metabolism. Aberrant upregulation of PDK phosphorylates and inhibits PDC activity, triggering a substrate preference shift toward FAO [31, 32]. Such metabolic reprogramming disrupts ATP homeostasis, which may constitute a central mechanism underlying reduced sperm motility. However, the molecular pathways through which psychological stress influences sperm quality via this metabolic axis remain uncharacterized, representing a critical knowledge gap at the intersection of psychobiology and reproductive medicine.

This study employed a cross-sectional design to investigate the correlation between semen quality parameters and psychological scale scores. Mechanistically, it pioneers a focused exploration of the mitochondrial PDK-PDC axis in human sperm. Through comprehensive analyses encompassing energy metabolite profiling, free fatty acid quantification, and qRT-PCR-based evaluation of PDK/PDH expression dynamics, we systematically deciphered the molecular pathways by which psychological stress disrupts sperm energy metabolism. These findings partially bridging critical knowledge gaps​ in the molecular mechanisms linking emotional distress to male infertility.

Materials and methods

Study design

In this multicenter cross-sectional study conducted from June 2023 to July 2024, 557 adult male volunteers were consecutively recruited from three clinical departments in Hubei Province, China: the Department of Urology at Hubei Hospital of Traditional Chinese Medicine, the Department of Reproduction at Hubei Maternal and Child Health Hospital, and the Department of Urology at Hubei Hospital of Integrated Traditional and Western Medicine. The study protocol received ethical approval from all participating institutions (Approval Nos.: HBZY2023-C10-01, 2023IEC-094, and 2023-ER-058) and was conducted in compliance with the Declaration of Helsinki. All participants provided written informed consent after receiving detailed instructions about semen sample collection procedures and study objectives, with all data processed anonymously to ensure confidentiality.

Participants were required to complete two standardized instruments. The first instrument was the 21-item Depression, Anxiety and Stress Scale (DASS-21), a validated psychometric tool consisting of three 7-item subscales that measure emotional states over the preceding week using a 4-point Likert response format (0 ="Not applicable at all"to 3 ="Very applicable"). This Chinese-adapted version has demonstrated satisfactory reliability in prior studies [33], with Cronbach's α coefficients ranging from 0.76 to 0.90 [34]. The second questionnaire collected comprehensive data through three domains: sociodemographic characteristics including age, height, body mass index (BMI), marital status, parenthood status, employment status, and economic standing;medical history encompassing psychiatric disorders, hypertension, hyperlipidemia, diabetes mellitus, hyperuricemia, and rheumatoid diseases; and modifiable lifestyle factors such as smoking habits, alcohol consumption patterns, sleep deprivation frequency, occupational overexertion, and prolonged electronic device usage.Each volunteer filled out these two questionnaires and provided a semen sample.

Semen sample collection and analysis

Semen samples were collected through masturbation after 2–7 days of sexual abstinence, adhering to World Health Organization (WHO) guidelines. Immediately after collection into sterile containers, samples were incubated at 37 °C for 30 min to ensure complete liquefaction. Semen analysis was performed using a computer-assisted sperm analysis (CASA) system (SQA-Vision, Hamilton Thorne IVOS II) to evaluate sperm concentration (× 10⁶/mL), motility parameters (total and progressive motility, %), and Morphological characteristics. Morphology assessment followed WHO 5th edition criteria, with Diff-Quik stained smears analyzed by two independent technicians. All procedures were completed within 1 h post-collection, and inter-observer variability was controlled below 5% through standardized protocols.

Targeted energy metabolomics

Sperm purification was performed using PureSperm density gradient centrifugation (40/80%, Nidacon), yielding high-motility sperm (> 90% viability) with > 95% purity. Purified sperm (1 × 10⁶ cells) were lysed in methanol–acetonitrile (1:1) containing 10 mM SUCCINIC ACID-D6 (Sigma-Aldrich, Cat# 488,536) internal standard, followed by ultrasonication water bath (40 kHz, 4 °C, 30 min) and protein precipitation at −20°C. After centrifugation (14,000 g, 20 min, 4 ℃), supernatants were lyophilized using a vacuum concentrator,The dried residues were resuspended in 100 μL of acetonitrile–water (1:1, v/v) prior to injection, and the resulting solution was used for LC–MS analysis. Chromatographic separation was achieved on an Agilent 1290 Infinity II UHPLC system equipped with an ACQUITY UPLC BEH Amide column (1.7 µm, 2.1 mm × 150 mm, Waters) at a column temperature of 35 °C. The mobile phase consisted of two eluents: eluent A was 50 mM ammonium acetate aqueous solution containing 1.2% ammonium hydroxide (pH 9.0), and eluent B was acetonitrile containing 1% acetylacetone. The gradient elution protocol was performed at a flow rate of 300 μL/min with an injection volume of 2 μL, following the time-dependent B-phase gradient: 70% B-phase was maintained from 0 to 1 min; B-phase linearly decreased from 70 to 60% between 1 and 10 min; B-phase further linearly decreased from 60 to 30% between 10 and 12 min; B-phase was held at 30% from 12.1 to 15 min; B-phase rapidly increased from 30% back to 70% between 15 and 15.5 min; finally, 70% B-phase was maintained from 15.5 to 22 min.A 5500 QTRAP mass spectrometer (SCIEX) operated in negative MRM mode with ESI source parameters: 450 °C, ISVF −4500 V.Quality assurance included: 1) Internal standard monitoring (CV < 15%); 2) QC samples every 10 injections (RSD < 15%); 3) Method validation showing linearity (0.01–100 μM, R² > 0.99) and recovery (85–115%). Data processing used Multiquant 3.0.2 with retention time alignment.

Targeted lipidomics-detection of free fatty acids

Sperm purification was performed using 40/80% PureSperm density gradient centrifugation (Nidacon International) with DPBS washes, yielding > 95% pure sperm with > 90% viability. For free fatty acid (FFA) analysis, 1 × 10⁶ purified sperm were homogenized in methanol-MTBE (1:1.3, 0.1% BHT) and centrifuged (12,000 g, 5 min, 4 °C). The lipid phase was derivatized with 15% BF₃-methanol containing 10 μM C17:0 internal standard (Cmass, Cat# CMS-R0542) at 60 °C for 30 min to generate fatty acid methyl esters (FAMEs). Hexane-extracted FAMEs were analyzed via GC–MS (Shimadzu GCMS-QP2020) on a DB-5MS column (30 m × 0.25 mm, 0.25 µm, Agilent Technologies) with an injection volume of 1 μL, using temperature programming: 40 °C (2 min) → 200 °C at 30°C/min → 240 °C at 10°C/min → 285 °C at 5°C/min (3 min). MS detection used electron ionization (70 eV) with ion source/quadrupole temperatures of 230°C/150°C.FFA quantification was performed using GCMS Solution software (Shimadzu) by matching fragment ions (e.g., m/z 74, 87) against the NIST 20 library (similarity > 80%). Absolute concentrations were normalized to C17:0 internal standard (CV < 10%). Metabolic pathway enrichment (KEGG ko04978) and PCA (SIMCA 14.1, UV-scaled) were conducted with significance thresholds of FDR < 0.05 and model validity criteria (R²X > 0.6, Q² > 0.5).Quality assurance included: 1) Derivatization efficiency > 95% (spiked FFA standards); 2) Daily system suitability (Rs > 1.5, theoretical plates > 5000); 3) QC samples (pooled extracts) with RSD < 15%; 4) Method validation (linearity 0.1–100 μg/mL, R² > 0.99; recovery 85–115%; intra-/inter-day precision RSD < 8%/12%); 5) Blank controls (signal < LOQ).

Reverse transcription-quantitative PCR (RT-qPCR)

Sperm purification was performed using 40/80% PureSperm density gradient centrifugation (Nidacon International) with DPBS washes, yielding > 95% pure sperm with > 90% motility. Mitochondria were isolated from 1 × 10⁷ sperm using a commercial kit (Thermo Scientific, #89,874) through differential centrifugation: initial lysis (700 g, 10 min, 4 °C) to remove debris, followed by mitochondrial enrichment (12,000 g, 15 min, 4 °C).Total RNA was extracted using a Servicebio kit (#G3640), with quality verified by A260/A280 ratios (1.8–2.0) and agarose gel electrophoresis. First-strand cDNA synthesis was performed with 1 μg RNA using a Servicebio reverse transcription kit (#G3337) under standard conditions: 42 °C for 20 min, 85 °C for 5 s.Quantitative PCR utilized SYBR Green master mix (Servicebio, #G3326) with gene-specific primers (Table 1, designed by Wuhan Cyvier Biotechnology) in triplicate reactions. Cycling parameters: 95 °C for 30 s, 40 cycles of 95°C/15 s and 60°C/30 s. Relative gene expression was calculated via 2 − ΔΔCT method using GAPDH for normalization.Quality control included: 1) RNA integrity verification; 2) primer specificity confirmation through melt curve analysis; 3) technical triplicates (CV < 5%); 4) NTC controls.

Table 1.

Primer Sequences

Gene	Species	Primer Sequence (5'−3')	Fragment Length (bp)
PDK1	Human	S: TGTGAAGATGAGTGACCGAGGAG	216
		A: GCATCTGTCCCGTAACCCTCTA
PDK2	Human	S: TACCTCAGCCGCATCTCCAT	131
		A: TTGACCACCTCAGAGACGTTG
PDK3	Human	S: GATAATTTACTTAACCGCCCTTCAG	230
		A: GTGCTAATGAAAGGATCAAACCC
PDK4	Human	S: AAGCCCAGATGACCAGAAAGC	210
		A: TGGTTCATCAGCATCCGAGTAGA
PDH	Human	S: CGAATTGGAATCCCAGTCAGAAG	71
		A: AGTTGAGTTGGTGCTGGCATG
GAPDH	Human	S: GGAAGCTTGTCATCAATGGAAATC	168
		A: TGATGACCCTTTTGGCTCCC

Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics 25.0 and RStudio (version 4.3.1), with data visualization conducted in GraphPad Prism 8. The normality of continuous variables was assessed by Shapiro–Wilk tests (α = 0.05). Normally distributed data were analyzed using independent t-tests, while non-normally distributed data were evaluated with Mann–Whitney U tests. Statistical significance was defined as two-tailed p < 0.05.

Results

This study enrolled 557 healthy male volunteers with a mean age of 35.02 ± 8.84 years. Table 2 summarizes sociodemographic characteristics and lifestyle factors. Notably, 30.16% reported smoking, 41.65% alcohol consumption, 61.40% frequent night-time activity, 32.68% occupational overexertion, and 85.82% prolonged electronic device usage.

Table 2.

Descriptive analysis of the questionnaire data from male volunteers. Data are presented as mean ± standard deviation (SD) or percentages

Parameter		N
Age (years)	35.02 ± 8.84	557
Height (m)	1.73 ± 0.05	555
Weight (kg)	72.15 ± 9.59	555
BMI (kg/m2)	23.99 ± 3.20	555
Get married
No	29.8	166
Yes	70.2	391
Have children
No	50.99	284
Yes	49.01	273
Have a regular job
No	12.93	72
Yes	87.07	485
There are other sources of income
No	81.33	453
Yes	18.67	104
Have a history of mental illness
No	97.85	545
Yes	2.15	12
High blood pressure
No	91.74	511
Yes	8.26	46
High blood fat
No	91.35	507
Yes	8.65	48
Diabetes
No	97.48	541
Yes	2.52	14
Hyperuricemia
No	91.53	508
Yes	8.47	47
Internal rheumatism
No	98.92	549
Yes	1.08	6
Smoking status
No	69.84	389
Yes	30.16	168
Drinking status
No	58.35	325
Yes	41.65	232
Stay up late
No	38.6	215
Yes	61.4	342
Overwork
No	67.32	375
Yes	32.68	182
Use Electronics for a long time
No	14.18	79
Yes	85.82	478

Participants were stratified into groups based on DASS-21 thresholds (depression: 0–9; anxiety: 0–7; stress: 0–14), with psychological distress observed in 38.60% (depression), 56.73% (anxiety), and 27.10% (stress) of participants. Severity stratification is detailed in Table 3.

Table 3.

Prevalence of depression, anxiety, and stress among male volunteers. Data are expressed as percentages

	N(%)
DAS level	Depression	Anxiety	Stress
Normal	342 (61.40)	241 (43.27)	406 (72.90)
Mild	99 (17.77)	75 (13.46)	71 (12.74)
Moderate	81 (14.54)	163 (29.26)	48 (8.62)
Severe	26 (4.67)	46 (8.26)	28 (5.02)
Extremely severe	9 (1.62)	32 (5.75)	4 (0.72)

Semen analysis revealed the following parameters (mean ± SD): concentration 57.35 ± 23.95 × 10⁶/mL, total motility 45.94 ± 6.72%, progressive motility 36.95 ± 5.74%, and normal Morphology 7.64 ± 3.46% (Table 4).

Table 4.

Descriptive analysis of the routine semen parameters of male volunteers. Data are presented as mean ± standard deviation (SD)

Semen quality parameters		N
Concentration (× 106/mL)	57.35 ± 23.95	557
Total motility (Progressive and non-progressive, %)	45.94 ± 6.72	557
Progressive motility (%)	36.95 ± 5.74	557
Normal morphological sperm (%)	7.64 ± 3.46	557

The impact of depression, anxiety, and stress on routine semen quality parameters

Bivariate correlations between psychological distress scores (depression, anxiety, stress), age, BMI, and semen parameters were assessed using Pearson's r, with correlation patterns visualized through a heatmap (Fig. 1).

Fig. 1.

Pearson correlation matrix of psychological distress scores and semen parameters. The heatmap illustrates pairwise correlations between depression, anxiety, stress scores (DASS-21 subscales), age, BMI, and semen quality parameters. Color intensity reflects the magnitude of Pearson's correlation coefficient (|r|), with red indicating positive correlations and blue negative correlations (see scale bar). Coefficients range from −1 (perfect inverse) to + 1 (perfect direct relationship)

The analysis revealed that depression, anxiety, and stress exerted significant negative effects on semen quality. Specifically, the DAS group exhibited significantly lower progressive motility (p < 0.001) and total motility (p = 0.002) compared to the Non-DAS group, as detailed in Fig. 2 and Table 5. The DAS group showed no significant differences in sperm concentration or normal morphology compared to the Non-DAS group (Table 5).

Fig. 2.

Comparative analysis of semen parameters between DAS and Non-DAS groups. Bar plots show (A) sperm concentration, (B) progressive motility, (C) total motility, and (D) normal morphology. Data are presented as mean ± SD (n = 557), with significance levels denoted as **** p < 0.001,** p < 0.01,* p < 0.05, ns (not significant)

Table 5.

Comparative analysis of semen parameters between DAS and Non-DAS groups

	M(P25, P75)		p-Value
Parameter	Non-DAS	DAS
	(N = 197)	(N = 360)
Age (years)	34.000(30.000,39.000)	33.000(29.000,38.750)	0.175
Height (m)	1.730(1.700,1.750)	1.730(1.700,1.760)	0.167
Weight (kg)	71.000(65.000,76.000)	71.000(65.375,78.000)	0.531
BMI (kg/m2)	23.700(22.150,25.900)	23.900(22.200,25.575)	0.892
Concentration (× 106/mL)	56.840(38.190,77.615)	56.595(35.625,78.665)	0.707
Progressive motility (%)	37.610(34.760,40.815)	35.695(32.073,40.343)	< 0.001***
Total motility (Progressive and non-progressive, %)	45.810(43.125,50.330)	45.205(40.032,50.020)	0.002**
Normal morphological sperm (%)	8.000(5.000,11.000)	8.000(5.000,10.000)	0.422

^†Data presented as median (IQR); Mann–Whitney U test with *** p < 0.001; **p < 0.01;* p < 0.05

Depression, anxiety, and stress impair glucose metabolism, disrupt TCA Cycle, and reduce electron transport chain efficiency

Targeted metabolomic analysis identified 26 key metabolites involved in glycolysis, the tricarboxylic acid (TCA) cycle, and oxidative phosphorylation. The DAS group exhibited significantly reduced ATP levels (p < 0.001) and downregulation of 22 metabolites compared to controls (Fig. 3A). Critical glycolytic intermediates, including dihydroxyacetone phosphate (DHAP), fructose-1,6-bisphosphate (F1,6BP), fructose-6-phosphate (F6P), and pyruvate, were markedly decreased (Fig. 3B). Concurrently, TCA cycle inhibition was evidenced by reduced citrate, α-ketoglutarate, oxaloacetate, and fumarate levels (all p < 0.05). Furthermore, electron transport chain (ETC) dysfunction was observed through diminished NADH and FMN (flavin mononucleotide) concentrations (p < 0.05).KEGG pathway enrichment analysis of the 22 differentially expressed metabolites (Fig. 3C-D) revealed significant perturbations in core metabolic processes: central carbon metabolism, glucagon signaling, TCA cycle, glycolysis/gluconeogenesis, pyruvate metabolism, and oxidative phosphorylation.

Fig. 3.

Comparative analysis of energy metabolism between DAS and Non-DAS groups. A Schematic of the TCA cycle with quantitative metabolite levels in glycolysis, TCA cycle, and oxidative phosphorylation. B Hierarchical clustering of differential metabolites (red: upregulation; purple: downregulation). C KEGG pathway enrichment map. D Top 10 significantly enriched metabolic pathways. Data expressed as mean ± SD (n = 10 biological replicates per group). ***p < 0.001, **p < 0.01,*p < 0.05

Depression, anxiety, and stress induce lipid metabolic imbalance through free fatty acid accumulation

Targeted lipidomic analysis via GC–MS/MS identified 31 free fatty acids (FFAs). Principal component analysis (PCA) indicated a partial separation of lipid metabolic profiles between the DAS and Non-DAS groups, though the 95% confidence ellipses overlapped (Fig. 4A). Eight FFAs exhibited significant differential expression (P < 0.05): decanoic acid (C10:0), hendecanoic acid (C11:0), linolelaidic acid (trans-C18:2n6), lauric acid (C12:0), myristic acid (C14:0), palmitic acid (C16:0), stearic acid (C18:0), and α-linolenic acid (C18:3n3), all showing upregulation in the DAS group (Fig. 4B-C).KEGG pathway enrichment analysis demonstrated these metabolites were predominantly involved in lipid metabolic pathways, fatty acid biosynthesis and unsaturated fatty acid biosynthesis (Fig. 4D-E).

Fig. 4.

Lipid metabolic alterations in the DAS group (A) PCA score plot showing group separation (DAS vs. Non-DAS). (B) Heatmap of differentially expressed FFAs (red: upregulation; green: downregulation). C Violin plots quantifying FFA levels (***p < 0.001, **p < 0.01,*p < 0.05). (D) KEGG classification showing pathway-to-metabolite ratios. E Pathway impact analysis with differential abundance scores. Data expressed as mean ± SD (n = 4 biological replicates)

RT-qPCR analysis revealed differential expression of pyruvate dehydrogenase kinase (PDK) isoforms 1–4 and pyruvate dehydrogenase complex (PDH) in sperm mitochondria between DAS and Non-DAS groups. PDK2 and PDK4 expressions were significantly upregulated in the DAS group (p < 0.05 and p < 0.01, respectively), whereas PDK1 and PDK3 showed non-significant upward trends(Fig. 5A). Importantly, PDH expression was significantly downregulated in the DAS group (p < 0.01), suggesting potential impairment of pyruvate oxidation capacity (Fig. 5B).

Fig. 5.

Expression profiles of PDK isoforms and PDH between DAS and Non-DAS groups (A) Relative mRNA levels of PDK1, PDK2, PDK3, and PDK4. (B) PDH mRNA expression was significantly reduced in the DAS group. Data expressed as mean ± SD (n = 8 biological replicates). ***p < 0.001, **p < 0.01,*p < 0.05, ns (not significant)

Discussion

The global decline in human semen quality over recent decades [35–37] has prompted urgent investigations into modifiable risk factors. Psychological stress, as a potentially reversible contributor to male infertility, remains mechanistically controversial. By integrating psychometric evaluation (DASS-21), targeted energy metabolomics, lipidomics, and qPCR analysis, this study systematically demonstrates the correlation between depression/anxiety/stress (DAS) severity and impaired sperm motility, while pioneering the identification of the mitochondrial PDK-PDC axis as the molecular hub underlying this association. This breakthrough parallels discoveries in neurodegenerative diseases, where chronic stress-induced metabolic dysregulation similarly compromises neuronal bioenergetics [38], It suggests the existence of conserved stress—response pathways across different tissues, highlighting the universality of stress—related metabolic disruptions in biological systems..

The mitochondrial PDK-PDC axis serves as a central regulatory node in sperm energy metabolism, coordinating glucose and fatty acid oxidation by dynamically balancing the activity of the PDH. This study provides the first experimental evidence that psychological stress hijacks this regulatory system through isoform-specific mechanisms (Fig. 6): overexpression of PDK2 and PDK4 in the DAS group sperm drives PDH suppression, sequestering pyruvate in the cytosol and blocking its conversion to acetyl-CoA. This directly depletes key TCA cycle intermediates, including citrate and α-ketoglutarate. Concurrently, diminished NADH and FMN availability reduces electron transport chain (ETC) efficiency, ultimately leading to decreased ATP synthesis. Termed"dual-engine shutdown"(dual inhibition of glycolysis and TCA cycle), this phenomenon fundamentally differs from the Warburg effect in cancer cells [39].While both cancer cells and stressed sperm undergo metabolic reprogramming, their objectives diverge fundamentally. The Warburg effect fuels cancer proliferation by shunting glycolytic intermediates to biosynthesis despite functional mitochondria [39]. In contrast, DAS-stressed sperm face systemic energy failure due to mitochondrial dysfunction: glycolysis suppression (↓pyruvate/F1,6BP), TCA cycle blockade (↓citrate/α-ketoglutarate), and failed compensatory β-oxidation (Fig. 3–4). This"dual-engine shutdown"arises from sperm's unique bioenergetic constraints—unlike somatic cells, sperm lack biosynthetic demands but require extreme ATP flux for flagellar motility [20], making them uniquely vulnerable to PDK-PDC dysregulation. KEGG pathway enrichment analysis further confirms significant enrichment of differentially expressed metabolites in core energy production pathways (TCA cycle and oxidative phosphorylation), indicating systemic metabolic remodeling under psychological stress, highlighting the complexity of the metabolic response to stress in sperm cells.

Fig. 6.

Overview of metabolic reprogramming in male sperm mitochondria under depression, anxiety, and stress:Compared with those without emotional distress, the expression of PDK2 and PDK4 in male sperm mitochondria was up-regulated under depression, anxiety and stress, which inhibited PDH, inhibited glucose metabolism pathway, blocked TCA cycle and decreased ETC efficiency, fatty acid oxidation is forced up and total ATP production is reduced

Faced with glycolytic collapse, sperm attempt to compensate for the energy deficit through β-oxidation—a process evidenced by significant upregulation of eight free fatty acids (FFAs), including palmitic acid (C16:0) and stearic acid (C18:0), as revealed by lipidomic profiling. However, this pseudo-compensatory response fails to restore ATP depletion and may exacerbate reactive oxygen species (ROS) generation, triggering oxidative stress that damages intracellular organelles and impairs sperm motility and function [40], thereby establishing a vicious cycle of metabolic disturbance → oxidative damage → energy depletion. This maladaptive cascade resembles non-alcoholic fatty liver disease (NAFLD), in which excessive FFAs induce mitochondrial dysfunction through incomplete oxidation [41]. Furthermore, excessive accumulation of long-chain fatty acids (LCFAs) may directly disrupt sperm plasma membrane fluidity through lipotoxicity [42], further compromising motility, as the integrity and fluidity of the plasma membrane are crucial for sperm—egg interaction and the sperm's ability to swim towards the egg.

From an evolutionary perspective, the PDK-PDC axis originally evolved as an adaptive mechanism to combat nutritional deprivation—suppressing glucose oxidation while enhancing lipolysis to sustain energy supply during starvation [43]. However, chronic psychological stress pathologically activates this conserved pathway, trapping sperm in a"metabolic gridlock."Notably, although PDK1 and PDK3 expression trends did not reach statistical significance, their upregulation suggests progressive isoform-specific adaptation to prolonged stress. These molecular compensatory attempts paradoxically exacerbate metabolic dysregulation, mirroring the neurodegenerative consequences of Tau hyperphosphorylation in Alzheimer's disease [44], ultimately leading to irreversible energetic failure in sperm, highlighting the delicate balance of sperm metabolism and the detrimental effects of chronic stress on this balance.

This study presents the first systematic demonstration of the molecular mechanism by which psychological stress impairs sperm motility through mitochondrial PDK-PDC axis-mediated metabolic reprogramming. While prior studies have established significant negative correlations between psychological stress and semen parameters (e.g., sperm concentration and motility) [6, 10, 45–47] their mechanistic focus has largely centered on hormonal imbalances (e.g., testosterone suppression [48, 49] or oxidative stress pathways [50, 51]. Our findings represent a breakthrough by positioning metabolic regulation as the core mechanism, offering novel insights into stress-associated infertility. Through integrated multi-omics analysis, this study delineates the specific pathway by which psychological stress (DAS) drives metabolic reprogramming via the PDK-PDC axis. DAS induces isoform-specific overexpression of mitochondrial PDK2 and PDK4 in sperm, suppressing PDH activity and blocking pyruvate conversion to acetyl-CoA, thereby restricting TCA cycle flux. Although lipidomic profiling detected free fatty acid accumulation (e.g., palmitic acid, C16:0), indicative of compensatory β-oxidation activation, impaired oxidative phosphorylation efficiency (evidenced by significant reductions in NADH and FMN levels) reduced ATP synthesis, ultimately resulting in the loss of sperm progressive motility (Fig. 6). This"dual-engine shutdown"mechanism—distinct from the energy-compensatory Warburg effect in cancer [39]—provides a metabolic explanation for the global epidemiological overlap between semen quality decline and psychological disorders.This metabolic specificity explains the differential vulnerability of semen parameters to psychological stress, prompting further mechanistic exploration into compensatory adaptations.

Although β-oxidation partially compensates for glycolytic failure (as evidenced by FFA accumulation), its inefficiency in sperm mitochondria [52] prompts the exploration of alternative pathways. Drosophila studies indicate that ketone bodies can serve as emergency fuels for neurons to compensate for β-oxidation deficiency [53], suggesting a rationale for exploring their compensatory potential in germ cells. PDK inhibitors (e.g., DCA) significantly improve cellular function in metabolic stress models such as renal injury [54], implying therapeutic potential for sperm energy metabolism disorders. Notably, glucocorticoid receptor (GR) activation directly upregulates PDK4 transcription [55], thereby inhibiting PDH activity and driving metabolic dysfunction—a mechanism linking HPA axis hyperactivity to sperm bioenergetic failure.

Notably, DAS was associated with selective impairment of sperm motility, while concentration and morphology remained unaffected (Table 5), suggesting distinct pathogenic mechanisms. This selectivity may arise because motility primarily depends on mitochondrial bioenergetics—potentially disrupted by PDK-PDC axis dysregulation—whereas concentration relies on spermatogenesis modulated by complex testosterone responses to stress (e.g., chronic suppression versus acute elevation) [56]. Morphogenesis during spermiogenesis involves structural processes that appear less vulnerable to metabolic ATP depletion [57], and compensatory glucocorticoid signaling may upregulate androgen receptors to partially preserve spermatogenesis [58]. Future studies measuring serial hormonal dynamics are needed to validate these hypotheses.

Although our integrated model addresses compensatory mechanisms, several limitations warrant consideration. First, the cohort was exclusively recruited from three hospitals in Hubei Province, China, which may restrict the generalizability of our findings to broader populations. Although we collected basic demographic data, detailed information on participants'geographic origins and long-term residence was not available, precluding subgroup analyses of regional influences.Second, the cross-sectional design inherently limits causal inference between psychological stress and sperm quality impairment. Future longitudinal studies tracking stress dynamics and semen parameters over time, or experimental models (e.g., stress-induced animal studies), are needed to establish causality.Third, while we focused on the mitochondrial PDK-PDC axis, other metabolic pathways (e.g., glycolytic flux, oxidative phosphorylation) may also mediate stress-induced sperm damage and warrant investigation.

Conclusions

This study delineates the molecular mechanism by which depression/anxiety/stress impairs sperm function through mitochondrial PDK-PDC axis-mediated metabolic reprogramming. We confirm that stress-induced PDK2/4 overexpression and PDH suppression establish a"dual metabolic blockade"of glucose and lipid utilization, depleting ATP production and motility. The observed metabolic gridlock differs fundamentally from the Warburg effect, highlighting sperm-specific metabolic vulnerability under stress. Our pioneering integration of evolutionary biology into reproductive medicine reveals that PDK-PDC axis dysregulation represents pathological hijacking of a conserved stress-response pathway. These findings establish a theoretical foundation for developing non-invasive diagnostics (e.g., PDK4 expression levels) and precision therapies (e.g., isoform-specific PDK inhibitors), advancing male infertility management from phenomenological observation to mechanism-targeted regulation.

Abbreviations

PDK

Pyruvate dehydrogenase kinase

PDC

Pyruvate dehydrogenase complex

DASS

Depression anxiety stress stable

PDH

Pyruvate dehydrogenase

ATP

Adenosine triphosphate

BMI

Body mass index

SE

Standard deviation

PR

Progressive motility

NP

Non-progressive motility

TCA

Tricarboxylic acid cycle

NADH

Nicotinamide adenine dinucleotide

FMN

Flavin monoucleotide

KEGG

Kyoto encyclopedia of genes and genomes

Authors’ contributions

WW, WQK., and WZL conceptualized the study 、designed the methodology and drafted the original manuscript;; SJY and JXC performed validation; LQ,CSH.,ZYY. and GMJ conducted formal analysis; CSY carried out the investigation; CJG and XM contributed to review, editing, supervision, and funding acquisition. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China [Grant No. 82104704]; and the Hubei Provincial Administration of Traditional Chinese Medicine [Grant No. ZY2023Z024].

Data availability

The datasets generated during and/or analyzed during the current study are not publicly available but may be made available upon reasonable request to the corresponding author.

Declarations

Ethics approval and consent to participate

This study was conducted in strict accordance with the ethical principles of the World Medical Association Declaration of Helsinki and relevant institutional guidelines. All procedures involving human participants were approved by the Ethics Committees of the three hospitals, with approval numbers HBZY2023-C10-01,2023IEC (094), and (2023) ethical review no. (research 058). And informed consent was obtained from all participants. The privacy rights of the human subjects were upheld throughout the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.