Iron biomarkers in men with infertility are associated with sperm DNA hydroxymethylation and cumulative live birth rates: a prospective study.

This prospective study involved 60 infertile couples undergoing intracytoplasmic sperm injection (ICSI) cycles et al. Hadi Laboratory and Medical Center in Beirut, Lebanon, between January 2022 and September 2024. Ethical approval was obtained from Al-Hayat Hospital, Lebanon (ETC-2022–13). The male partners consented to provide a blood sample and agreed to donate for research purposes any leftover semen not used for ICSI. Both partners also consented to the use of their clinical and laboratorial data [ 31 ] (Fig.  1 ). Fig. 1 Study design. Sixty infertile couples were included in this study after applying the exclusion criteria ( a ). A blood sample was collected from each male partner ( b ). A semen sample ( c ) was also obtained. ( d ) Cell-free serum and seminal fluid were used for measuring iron biomarkers. ( e ) A portion of the semen sample was processed by density gradient, and the motile-enriched sperm fraction was subjected to DNA extraction ( f ) and ELISA for quantification of global 5-methylcytosine and 5-hydroxymethylcytosine in the DNA of motile spermatozoa ( g ). ( h ) Another semen fraction was processed using a density gradient to prepare the motile sperm-enriched fraction. ( i ) From each female partner, oocytes were collected, and mature oocytes were selected for ICSI ( j ). ( k , l ) Fertilization and blastulation rates were calculated post-ICSI, along with the cumulative live birth rate ( m ) Study design. Sixty infertile couples were included in this study after applying the exclusion criteria ( a ). A blood sample was collected from each male partner ( b ). A semen sample ( c ) was also obtained. ( d ) Cell-free serum and seminal fluid were used for measuring iron biomarkers. ( e ) A portion of the semen sample was processed by density gradient, and the motile-enriched sperm fraction was subjected to DNA extraction ( f ) and ELISA for quantification of global 5-methylcytosine and 5-hydroxymethylcytosine in the DNA of motile spermatozoa ( g ). ( h ) Another semen fraction was processed using a density gradient to prepare the motile sperm-enriched fraction. ( i ) From each female partner, oocytes were collected, and mature oocytes were selected for ICSI ( j ). ( k , l ) Fertilization and blastulation rates were calculated post-ICSI, along with the cumulative live birth rate ( m ) Cycles with embryo biopsy, frozen gametes, or sperm retrieved via testicular biopsy/epididymal aspiration were excluded. Women with BMI > 30 kg/m 2 , advanced maternal age (> 36 years at ovarian pick-up), uterine malformations, or fewer than three aspirated oocytes were also excluded (Fig.  1 ). In addition, samples with inflammatory cell counts exceeding 1 million/mL were excluded. Semen collection and analysis were conducted according to the World Health Organization 2021 guidelines [ 32 , 33 ] (Supplementary material 1 and 2). The remaining samples were divided into three aliquots (Fig.  1 ). The first two aliquots were processed using density gradient centrifugation with two-layer gradients (80–40 gradient layers, Cook Medical, Bloomington, USA) [ 34 ]. The collected sperm pellet was subsequently centrifuged at 448 g for 5 min using Sperm Medium (Cook Medical, Bloomington, USA). The washed sperm pellet from the first aliquot was used for ICSI. The pellet from the second aliquot was rapidly frozen in liquid nitrogen for DNA methylation analysis. Motile spermatozoa have a greater chance of fertilizing oocytes, which is why they were chosen for this analysis [ 35 ]. The third aliquot was centrifuged at 12,000 g for 5 min, and the cell-free supernatant was stored for further analysis [ 36 ]. The women underwent controlled ovarian stimulation following the GnRH antagonist protocol. Specifically, on the second day of ovarian stimulation, human menopausal gonadotropin (hMG) (Merional, IBSA, Switzerland) was administered (Table  1 ). On the sixth day of stimulation, a daily dose of a Gonadotropin-Releasing Hormone (GnRH) antagonist (Cetrotide, 0.25 mg, Merck Serono, Germany) was introduced. Ovulation was triggered with 5000 IU of highly purified human chorionic gonadotropin (hCG) (Choriomon, IBSA, Switzerland) once three or more leading follicles reached a diameter of 16 mm [ 37 ]. Ultrasound-guided transvaginal aspiration under conscious anesthesia was used for cumulus-oocyte complex (COC) retrieval 34–36 h after ovulation induction [ 38 ]. The collected COCs were incubated in Global Total LP for Fertilization medium, covered with oil (Life Global, Cooper-Surgical, Denmark), at 37 °C with 5% oxygen and 5% carbon dioxide for 2–3 h. Thereafter, oocyte denudation was carried out using hyaluronidase enzyme (Total Global, Life Global, Cooper-Surgical, Denmark). After denudation, the proportion of metaphase II (MII) oocytes at ICSI was determined by calculating the ratio of MII oocytes to the number of COCs retrieved per patient [ 39 ]. Table 1 Population characteristics Parameters All participants (n = 60) Women Age (years); (mean ± SD) 30.93 ± 4.62 BMI (kg/m 2 ); (mean ± SD) 23.74 ± 3.12 Antral follicle count (AFC); (median (IQR)) 12 (9) Anti-Müllerian hormone (AMH) (ng/mL); (median (IQR)) 1.32 (1) PCOS (%) 19.29 Fallopian tube-related factors (%) 12.2 Polypectomy (%) 3.5 Endometriosis (%) 5.2 Number of previous attempts; (median (IQR)) 0 (1) Primary infertility (%) 70 Number of previous miscarriages; (median (IQR)) 0 (1) Smoking status Non-smokers (%) 49.1 (n = 27) Smokers (%) 50.9 (n = 28) • Hookah only (%) 43.6 (n = 24) • Cigarette only (%) 7.3 (n = 4) • Dual users (%) (n = 0) • Hookah frequency: Once/day (%) 7.3 (n = 4) Twice/day (%) 20.0 (n = 11) Three times/day (%) 16.4 (n = 9) • Cigarettes/day:   40 (%) 1.8 (n = 1) Alcohol Consumption Non-drinkers (%) 98.2% (n = 59) Alcohol consumers (%) 1.8% (n = 1) • 1–5 standard drinks/week (%)  98.2% (n = 59) Controlled ovarian stimulation parameters Total dose of hMG (IU); (median (IQR)) 3600 (562.5) Days of stimulation; (median (IQR)) 8 (1) OHSS during the study period, n (%) 0 (0) Men Age (years); (median (IQR)) 37 (6) BMI (Kg/m 2 ); (mean ± SD) 28.17 ± 4.04 Varicocelectomy (%) 9.09 Orchidopexy (%) 0 Inguinal hernia repair (%) 1.7 Smoking status Non-smokers (%) 35.0 (n = 21) Smoker (%) 76.7 (n = 39) • Hookah only (%) 31.7 (n = 19) • Cigarette only (%) 30.0 (n = 18) • Dual users (%) 3.3 (n = 2) • Hookah frequency: Once/day (%) 13.3 (n = 8) Twice/day (%) 15.0 (n = 9) Three times/day (%) 6.7 (n = 4) • Cigarettes/day:   40 (%) 15.0 (n = 9) Alcohol Consumption Non-drinkers (%) 85.0% (n = 51) Alcohol consumers (%) 15.0% (n = 9) • 1–5 standard drinks/week (%)  10.0% (n = 6) • > 10 standard drinks/week (%) 5.0% (n = 3) Semen parameters Volume (ml); (median (IQR)) 2.5 (1) Sperm concentration (× 10 6 /ml); (mean ± SD) 50.32 ± 29.53 Total sperm count (× 10⁶/ejaculate); (median (IQR)) 100 (119.25) Total motility (%); (median (IQR)) 50 (16.5) Progressive motility (%); (mean ± SD) 30.37 ± 13.35 Non-motile spermatozoa (%); (median (IQR)) 50 (10) Global motile sperm DNA 5-mC and 5-hmC Percentage of methylated DNA; (median (IQR)) 0.051 (0.11) Percentage of hydroxymethylated DNA; (mean ± SD) 0.048 ± 0.066 Seminal fluid iron status ferritin (ng/ml); (median (IQR)) 186 (111.4) transferrin (mg/dl); (median (IQR)) 1 (2) UIBC (µg/dl); (median (IQR)) 806.40 (509.7) iron (µg /dl); (mean ± SD) 50.05 ± 19.67 TIBC (µg/dl); (median (IQR)) 850.30 (519.1) Serum iron status of male participants ferritin (ng/ml); (median (IQR)) 106 (77) transferrin (mg/dl); (mean ± SD) 218.95 ± 61.16 UIBC (µg/dl); (mean ± SD) 181.71 ± 67.94 iron (µg /dl); (mean ± SD) 70.48 ± 30.06 TIBC (µg/dl); (mean ± SD) 253.08 ± 79.06 The data are presented as the mean ± standard deviation for continuous variables with a normal distribution, as median (interquartile range) for non-normally distributed variables, and as percentages for categorical variables. BMI = body mass index; % = percentage; PCOS = Polycystic ovary syndrome; hMG = human menopausal gonadotropin; OHSS = ovarian hyperstimulation syndrome; 5-mC = 5-methylcytosine; 5-hmC = 5-hydroxymethylcytosine; ng/ml = nanograms per milliliter; mg/dl = milligrams per deciliter; µg/dl = micrograms per deciliter; UIBC = unsaturated iron-binding capacity; TIBC = Total iron-binding capacity. *A standard drink is equivalent to 14 g of pure alcohol Population characteristics The data are presented as the mean ± standard deviation for continuous variables with a normal distribution, as median (interquartile range) for non-normally distributed variables, and as percentages for categorical variables. BMI = body mass index; % = percentage; PCOS = Polycystic ovary syndrome; hMG = human menopausal gonadotropin; OHSS = ovarian hyperstimulation syndrome; 5-mC = 5-methylcytosine; 5-hmC = 5-hydroxymethylcytosine; ng/ml = nanograms per milliliter; mg/dl = milligrams per deciliter; µg/dl = micrograms per deciliter; UIBC = unsaturated iron-binding capacity; TIBC = Total iron-binding capacity. *A standard drink is equivalent to 14 g of pure alcohol ICSI was performed using the standard technique as routinely practiced in the laboratory. During the ICSI procedure, sperm were loaded into a 10% polyvinylpyrrolidone (PVP) solution (FujiFilm, Irvine Scientific, USA). Each mature oocyte was microinjected with an immobilized spermatozoon in HEPES medium (Life Global, Cooper-Surgical, Denmark) under light oil (Life Global, Cooper-Surgical, Denmark). Microinjected oocytes were cultured in a single Global Total LP medium (Life Global, Cooper-Surgical, Denmark) covered with light oil (Life Global, Cooper-Surgical, Denmark) under controlled conditions (37 °C in an incubator [K-system, G210 InviCell, Germany] with 5% oxygen and 5% carbon dioxide) for 5 days [ 40 ]. Thereafter, fertilization was confirmed by the presence of two polar bodies (2PBs) and two pronuclei (2PN). Cleavage rate is defined as the proportion of fertilized oocytes (2 PB and 2PN on Day 1) that have undergone cellular division by Day 3 (approximately 68 ± 1 h post-microinjection). For this calculation, both good and fair quality embryos on Day 3 were considered. Plus, blastocyst development was assessed as the proportion of 2PN zygotes reaching the blastocyst stage by Day 5, approximately 116 ± 2 h after microinjection [ 39 , 41 ]. Embryos were evaluated on day 5, and a fresh blastocyst transfer was performed under ultrasound guidance. Any additional blastocysts, or all embryos in the case of a 'freeze-only' policy (applied when the progesterone level exceeded 1.5 ng/ml on the day of trigger) [ 42 ], were cryopreserved on day 5 using the vitrification kit protocol (Kitazato Corporation, Japan). A single frozen embryo transfer was later transferred using the warming kit protocol (Kitazato Corporation, Japan). Daily oral and vaginal progesterone (Duphaston, Abbott, Netherlands and Biogest, Germany) were administered for luteal support [ 43 , 44 ]. For frozen embryo transfer (FET) cycles, endometrial preparation was performed using oral oestradiol valerate (Estrofem ® , Novo Nordisk, Denmark) at a dose of 8 mg/day starting from Day 2 of menstruation. Once the endometrial thickness reached ≥ 7 mm, luteal support was initiated using oral dydrogesterone (Duphaston ® , Abbott, Netherlands) administered three times daily, combined with vaginal micronized progesterone (Biogest ® , Besins Healthcare, Germany) at a total dose of 800 mg/day, also administered in three divided doses [ 45 ]. It is worth mentioning that the minimum criterion for embryo transfer or verification is a classification of 1 for size and expansion, corresponding to an early blastocyst (grade 1), according to the Gardner and Schoolcraft grading system [ 46 ]. Embryo transfer cycles were defined as the total number of embryo transfer procedures performed per patient, including both fresh and frozen embryo transfers [ 47 ]. For instance, one ovarian stimulation cycle followed by one fresh and two frozen embryo transfers was counted as three embryo transfer cycles. Biochemical pregnancy was defined as a serum β-hCG level exceeding 20 IU/L in two consecutive blood tests, starting from 13 days after blastocyst transfer [ 48 ]. Implantation rate was calculated as the number of gestational sacs observed per the number of blastocysts transferred. A transvaginal ultrasound was performed 3 weeks after blastocyst transfer to confirm implantation by identifying one or more gestational sacs [ 41 ]. CLBR refers to the occurrence of at least one live-born baby (≥ 24 weeks of gestation) resulting from fresh or subsequent frozen embryo transfer (FET) cycles. A live birth (LB) is defined as any instance where at least one baby is born alive from the transfer of fresh or frozen-thawed embryos. Only the initial delivery is considered in this analysis. All patients were included in the analysis, regardless of whether they underwent a fresh embryo transfer or a freeze-all cycle with subsequent frozen embryo transfer. However, the CLBR was calculated only for those who had at least one embryo transfer [ 43 ]. Each patient was analyzed only once and was monitored for a minimum of two years or until their embryos were utilized [ 43 ]. Experienced phlebotomists collected 4–5 ml of blood from participants through direct venipuncture into Vacutainer ™ tubes containing a gel for serum-clot separation. The samples were then centrifuged at 1,107 g for 15 min to separate the serum from the blood clot [ 49 ]. Serum and seminal fluid iron, UIBC, and transferrin levels were analyzed on the Cobas Integra 400 analyzer (Roche Diagnostics GmbH, Mannheim, Germany) using the Iron Gen.2 kit, UIBC kit, and Tina-quant transferrin ver.2 kit, respectively, following the manufacturer’s instructions [ 36 , 50 – 52 ]. TIBC was determined by adding the iron levels to the UIBC [ 52 ]. Ferritin levels were measured on an automated CLIA platform (Maglumi 2000, Snibe Diagnostic, Shenzhen, China) [ 53 ]. All samples were prepared after centrifugation, with detailed procedures available in the Supplementary material 1 and Supplementary Fig. 1. DNA extraction was performed using the QIAamp DNA Mini Kit (Qiagen, Germany, code 51306) [ 54 , 55 ]. The extracted DNA was visualized through electrophoresis on a 0.8% (w/v) agarose gel (100 mV, 30 min) using a Bio-Rad Laboratories apparatus (Supplementary material 1) [ 56 ]. The Methylated DNA Quantification Kit (Colorimetric) (ab117128; Abcam, USA) and the Hydroxymethylated DNA Quantification Kit (Colorimetric) (ab117130; Abcam, USA) were utilized. The relative quantities of 5-mC and 5-hmC were calculated using the formulas provided by the manufacturer (Supplementary material 1) [ 35 ]. Sample size calculations were conducted using G*Power. For correlation analyses, assuming a rho of 0.32, an alpha of 0.05, and 80% power, the required sample size was 59. Regression models with up to three predictors, using a small effect size of 0.20, an alpha of 0.05, and 80% power, also required a maximum of 59 participants (Supplementary material 1). A total of 60 couples were enrolled, slightly exceeding this minimum to ensure adequate power and to account for potential data loss. SPSS version 26 was used for all analyses, with statistical significance set at a p < 0.05. The normality of numerical variables was tested using the Kolmogorov–Smirnov test. Pearson's correlation (r) was used for normally distributed variables, and Spearman's rho for non-normally distributed data. Following correlation analysis, univariate models were constructed for three outcomes: global motile sperm DNA 5-hmC, global motile sperm DNA 5-mC, and CLBR, with independent factors being serum and seminal iron parameters that showed significant correlation with 5-mC, 5-hmC, or CLBR. Multivariable linear regression models were subsequently calculated based on the statistically significant results of the univariate analysis, further exploring these associations while adjusting for potential confounders.

Demographic and clinical characteristics of study participants are presented in Table  1 (Supplementary material 3). Key embryological and clinical outcomes are summarized in Table  2 . A median of 11 (IQR 11) COCs were collected per participant, with a maturation rate of 76.7% (IQR 35.7). The fertilization rate averaged 73.2 ± 20.4%, cleavage rate was 66.6% (IQR 50), and blastulation rate 49.4 ± 29.0%. Freeze-all cycles occurred in 3.3% of cases. The median number of embryo transfer cycles per patient was 1 (IQR 0). Implantation rates had a median of 0% (IQR 100%). Rates of ectopic pregnancy and miscarriage were 1.6% and 8.3%, respectively. A total of 24 singleton live births were recorded, with a median cumulative live birth rate of 0% (IQR 100%) (Table  2 ). When participants were stratified by normozoospermic status, no significant differences were observed in embryological outcomes, including fertilization and blastulation rates, or in cumulative live birth rates between groups (Supplementary Table 1, Supplementary material 3). Table 2 Embryological and clinical outcome Parameters Values Number of collected COCs; (median (IQR)) 11 (11) Proportion of MII oocytes (%); (median (IQR)) 76.70 (35.71) Fertilization rate (%); (mean ± SD) 73.23 ± 20.36 Cleavage rate (%); (median (IQR)) 66.6 (50) Blastulation rate (%); (mean ± SD) 49.36 ± 28.95 Freeze-all (n); (%) 2; 3.3% Embryo transfer cycles (n); (median (IQR)) 1 (0) Implantation rate (%); (median (IQR)) 0 (100) Ectopic pregnancy (%) 1.6 Miscarriage (%) 8.3 Total number of singleton newborns (n) 24 CLBR (%); (median (IQR)) 0 (100) The data are presented as the mean ± standard deviation for continuous variables with a normal distribution, as median (interquartile range) for non-normally distributed variables, and as percentages for categorical variables. COC = cumulus-oocyte complex, MII = metaphase II, CLBR = cumulative live birth rate Embryological and clinical outcome The data are presented as the mean ± standard deviation for continuous variables with a normal distribution, as median (interquartile range) for non-normally distributed variables, and as percentages for categorical variables. COC = cumulus-oocyte complex, MII = metaphase II, CLBR = cumulative live birth rate Interestingly, serum iron, transferrin, and TIBC were positively correlated with total motility. Additionally, serum iron and TIBC were positively correlated with progressive motility. In parallel, seminal fluid transferrin levels were positively correlated with sperm concentration (p < 0.05) (Supplementary Table 2). No significant correlations were found between parental characteristics or sperm parameters and global DNA methylation marks (5-mC and 5-hmC) or CLBR (Supplementary Figs. 2–3, Tables 3–4). Serum and seminal fluid ferritin levels showed no significant differences, while seminal fluid transferrin and iron levels were significantly lower than serum levels (p < 0.001). Conversely, seminal fluid TIBC and UIBC levels were significantly higher than their serum counterparts (p < 0.001) (Supplementary Fig. 4 a-e). Positive correlations were observed between seminal fluid transferrin and serum transferrin (R = 0.40, p = 0.004), TIBC (R = 0.38, p = 0.007), and UIBC (R = 0.37, p = 0.008) (Supplementary Fig. 4g) (Supplementary material 3). For motile sperm global DNA 5-hmC levels, positive correlations were found with serum iron (R = 0.288, p = 0.037), serum TIBC (R = 0.291, p = 0.035), seminal fluid iron (R = 0.296, p = 0.039), seminal TIBC (R = 0.326, p = 0.027), and seminal fluid UIBC (R = 0.301, p = 0.042). For motile sperm global DNA 5-mC levels, a positive correlation was observed with seminal fluid TIBC (R = 0.323, p = 0.027) (Table  3 ). Table 3 Correlations of serum and seminal iron parameters with 5-mC, 5-hmC, embryological outcomes, and CLBR 5-mC (%) 5-hmC (%) Fertilization rate (%) Blastulation rate (%) CLBR (%) Serum Ferritin (ng/ml) R − 0.05 − 0.09 − 0.26 − 0.10 0.03 p-value 0.70 0.49 0.05 0.57 0.79 Serum iron (µg /dl) R 0.05 0.28 * − 0.18 0.15 0.04 p-value 0.67 0.03 0.17 0.42 0.77 Serum transferrin (mg/dl) R 0.009 0.26 0.07 0.09 − 0.02 p-value 0.94 0.05 0.60 0.63 0.84 Serum TIBC (µg/dl) R 0.04 0.29 * 0.05 0.12 − 0.01 p-value 0.73 0.03 0.68 0.53 0.90 Serum fluid UIBC (ug/dl) R 0.08 0.20 0.12 0.07 − 0.03 p-value 0.53 0.13 0.36 0.69 0.80 Seminal fluid ferritin (ng/ml) R − 0.13 0.01 − 0.33 0.03 − 0.005 p-value 0.50 0.96 0.09 0.92 0.979 Seminal fluid iron (µg/dl) R 0.23 0.29 * − 0.16 0.05 0.30 * p-value 0.09 0.03 0.25 0.79 0.02 Seminal fluid transferrin (mg/dl) R 0.08 0.11 − 0.17 0.12 0.06 p-value 0.54 0.44 0.22 0.52 0.67 Seminal fluid TIBC (µg/dl) R 0.32 * 0.32 * − 0.13 0.002 − 0.11 p-value 0.02 0.02 0.35 0.99 0.43 Seminal fluid UIBC (µg/dl) R 0.27 0.30 * − 0.12 0.02 − 0.17 p-value 0.06 0.04 0.39 0.90 0.24 5-mC (%) R 0.04 − 0.09 − 0.01 p-value 0.74 0.64 0.32 5-hmC (%) R − 0.23 − 0.13 − 0.10 p-value 0.08 0.49 0.46 Correlations between continuous variables were assessed using Spearman and Pearson correlation coefficients, depending on whether the variables were normally or non-normally distributed. Data are presented as R correlation coefficients and p-values. Positive correlations (R > 0) and negative correlations (R < 0) are indicated *Bold values indicate that statistical significance was considered at p < 0.05 Correlations of serum and seminal iron parameters with 5-mC, 5-hmC, embryological outcomes, and CLBR Correlations between continuous variables were assessed using Spearman and Pearson correlation coefficients, depending on whether the variables were normally or non-normally distributed. Data are presented as R correlation coefficients and p-values. Positive correlations (R > 0) and negative correlations (R < 0) are indicated *Bold values indicate that statistical significance was considered at p < 0.05 Seminal fluid iron demonstrated a significant positive correlation with the CLBR (R = 0.309, p = 0.027) ( Table  3 ) . Notably, no significant correlations were noted between 5-hmC and fertilization rate, blastulation rate, or CLBR (p > 0.05). Similarly, no significant correlation was found between 5-mC and fertilization rate, blastulation rate, or CLBR (p > 0.05) (Table  3 ). Correlation analysis revealed no significant associations between parental medical history, semen parameters, and the analyzed outcomes (5-mC, 5-hmC, and CLBR). However, several iron biomarkers showed significant correlations with these outcomes in this primary screening. Therefore, a second step involving univariate linear regression analysis was conducted to quantify the strength of these associations. The univariate linear regression analysis showed that seminal fluid iron levels (B = 0.001, p = 0.039), serum iron levels (B = 0.001, p = 0.037), and serum TIBC levels (B < 0.001, p = 0.035) have a statistically significant influence on 5-hmC levels when considered separately, keeping all other factors constant. For every 1 µg/dl increase in seminal fluid iron levels, 5-hmC levels increased by 0.001%. For every 1 µg/dl increase in serum iron levels, 5-hmC levels increased by 0.001%. Additionally, for every 1 unit increase in serum TIBC levels, 5-hmC levels increased by slightly less than 0.001%. (Table  4 ) (Supplementary material 3). Table 4 Univariate linear regression analysis of iron parameters influence on 5-hmC, 5-mC, and CLBR levels Dependent variable: Global motile sperm DNA 5-hmC t Unstandardized Coefficients Standardized Coefficients Sig 95.0% Confidence Interval for B Collinearity Statistics B Std. Error Beta Lower Bound Upper Bound Tolerance VIF Seminal fluid iron 0.001 0 0.29 2.12 0.03 0 0.002 1 1 Serum iron 0.001 0 0.28 2.14 0.03 0 0.001 1 1 Serum TIBC 0 0 0.29 2.17 0.03 0 0 1 1 Seminal fluid TIBC 6.59E-06 0 0.02 0.15 0.87 0 0 1 1 Seminal fluid UIBC 3.94E-07 0 0.001 0.009 0.99 0 0 1 1 Dependent variable: Global motile sperm DNA 5-mC (Constant) − 1.679 0.378 − 4.44 0 − 2.445 − 0.912 Seminal fluid TIBC 0.001 0 0.307 1.933 0.061 0 0.002 1 1 Dependent variable: CLBR (Constant) 2.683 18.415 0.146 0.885 − 34.323 39.69 Seminal fluid iron 0.638 0.325 0.27 1.962 0.055 − 0.015 1.292 1 1 Bold values indicat that statistical significance was considered at p < 0.05 Univariate linear regression analysis of iron parameters influence on 5-hmC, 5-mC, and CLBR levels Bold values indicat that statistical significance was considered at p < 0.05 The univariate linear regression model revealed a p-value of 0.061 for seminal fluid TIBC levels, indicating that it did not have a statistically significant impact on 5-mC levels (Table  4 ) (Supplementary material 3). Seminal fluid iron concentration showed a trend toward a positive association with CLBR (β = 0.638, 95% CI − 0.015–1.292, p = 0.055), approaching but not reaching conventional statistical significance (p < 0.05) (Table  4 ) (Supplementary material 3). Table 5 presents the stepwise progression of the multivariate regression models. For both 5-hmC and CLBR analyses, each model (e.g., Models 1–3 for 5-hmC and Models 1–7 for CLBR) reflects the sequential removal of variables with the highest p-values, as determined by the backward stepwise regression procedure. The final model in each sequence includes only those predictors that remained statistically significant after this systematic elimination process, thereby representing the most parsimonious and robust set of predictors for the outcome variable. Table 5 Multivariate linear regression analysis of iron parameters effects on 5-hmC levels and CLBR Dependent variable: Global motile sperm DNA 5-hmC Model Unstandardized Coefficients Standardized Coefficients t Sig 95.0% Confidence Interval for B Collinearity Statistics B Std. Error Beta Lower Bound Upper Bound Tolerance VIF 1 (Constant) − 0.06 0.04 − 1.55 0.12 − 0.14 0.019 Seminal fluid iron 0.001 0.001 0.20 1.40 0.16 0 0.002 0.92 1.08 Serum iron 0 0 0.16 0.90 0.37 − 0.001 0.001 0.56 1.76 Serum TIBC 0 0 0.17 0.95 0.34 0 0.001 0.57 1.75 2 (Constant) − 6.50E− 0 0.04 − 1.58 0.12 − 0.14 0.01 Seminal fluid iron 0.001 0 0.22 1.55 0.12 0 0.002 0.94 1.06 Serum TIBC 0 0 0.27 1.95 0.05 0 0.001 0.94 1.06 3 (Constant) − 0.039 0.03 − 1.02 0.31 − 0.11 0.03 Serum TIBC 0 0 0.33 2.35 0.02 0 0.001 1 1 Dependent variable: CLBR 1 (Constant) 2.14 35.88 0.06 0.95 − 70.7 75.06 Seminal fluid iron 1.18 0.42 0.52 2.82 0.008 0.3 2.043 0.66 1.50 Seminal fluid transferrin − 3.50 2.01 − 0.32 − 1.73 0.09 − 7.59 0.59 0.65 1.52 Seminal fluid UIBC − 0.01 0.02 − 0.06 − 0.38 0.70 − 0.07 0.048 0.87 1.14 Serum ferritin 0.07 0.07 0.15 0.96 0.34 − 0.08 0.23 0.91 1.09 Serum iron − 0.20 0.27 − 0.12 − 0.73 0.46 − 0.75 0.35 0.75 1.32 Serum UIBC − 0.02 0.14 − 0.03 − 0.17 0.85 − 0.31 0.26 0.73 1.36 Sperm 5-mC − 13.52 22.16 − 0.09 − 0.61 0.54 − 58.57 31.52 0.85 1.17 Sperm 5-hmC − 17.39 106.82 − 0.02 − 0.16 0.87 − 234.49 199.70 0.79 1.25 2 (Constant) 2.70 35.21 0.07 0.93 − 68.79 74.2 Seminal fluid iron 1.18 0.41 0.51 2.86 0.007 0.34 2.01 0.67 1.48 Seminal fluid transferrin − 3.55 1.96 − 0.32 − 1.81 0.07 − 7.53 0.42 0.67 1.48 Seminal fluid UIBC − 0.01 0.02 − 0.06 − 0.38 0.70 − 0.06 0.04 0.87 1.14 Serum ferritin 0.07 0.07 0.15 0.99 0.32 − 0.07 0.23 0.91 1.09 Serum iron − 0.21 0.26 − 0.13 − 0.81 0.42 − 0.74 0.31 0.80 1.24 Serum UIBC − 0.02 0.14 − 0.03 − 0.2 0.84 − 0.31 0.25 0.74 1.35 Sperm 5-mC − 13.47 21.85 − 0.09 − 0.61 0.54 − 57.83 30.8 0.85 1.16 3 (Constant) − 0.70 30.42 − 0.02 0.98 − 62.40 61.002 Seminal fluid iron 1.18 0.40 0.51 2.90 0.006 0.35 2.007 0.67 1.48 Seminal fluid transferrin − 3.67 1.84 − 0.33 − 1.98 0.05 − 7.42 0.07 0.73 1.35 Seminal fluid UIBC − 0.01 0.02 − 0.06 − 0.40 0.69 − 0.06 0.04 0.87 1.13 Serum ferritin 0.07 0.07 0.15 1.02 0.31 − 0.07 0.22 0.92 1.08 Serum iron − 0.22 0.24 − 0.14 − 0.94 0.35 − 0.72 0.26 0.90 1.10 Sperm 5-mC − 14.21 21.24 − 0.10 − 0.66 0.50 − 57.30 28.86 0.88 1.13 4 (Constant) − 7.81 24.45 − 0.32 0.75 − 57.37 41.73 Seminal fluid iron 1.15 0.39 0.50 2.90 0.006 0.351 1.96 0.68 1.45 Seminal fluid transferrin − 3.69 1.82 − 0.34 − 2.02 0.051 − 7.39 0.009 0.73 1.35 Serum ferritin 0.08 0.07 0.16 1.11 0.27 − 0.06 0.22 0.94 1.05 Serum iron − 0.22 0.23 − 0.14 − 0.92 0.36 − 0.70 0.26 0.91 1.09 Sperm 5-mC − 15.90 20.58 − 0.11 − 0.77 0.44 − 57.62 25.80 0.91 1.09 5 (Constant) − 9.66 24.21 − 0.39 0.69 − 58.67 39.35 Seminal fluid iron 1.07 0.38 0.47 2.81 0.008 0.30 1.85 0.73 1.35 Seminal fluid transferrin − 3.44 1.78 − 0.31 − 1.92 0.06 − 7.06 0.17 0.76 1.31 Serum ferritin 0.08 0.07 0.18 1.23 0.22 − 0.05 0.23 0.96 1.04 Serum iron − 0.20 0.23 − 0.13 − 0.88 0.38 − 0.68 0.27 0.91 1.09 6 (Constant) − 19.09 21.65 − 0.88 0.38 − 62.90 24.71 Seminal fluid iron 1 0.37 0.43 2.69 0.01 0.25 1.75 0.77 1.28 Seminal fluid transferrin − 3.47 1.78 − 0.32 − 1.94 0.05 − 7.08 0.13 0.76 1.31 Serum ferritin 0.08 0.07 0.16 1.12 0.26 − 0.06 0.22 0.97 1.02 7 (Constant) − 8.16 19.4 − 0.42 0.67 − 47.38 31.06 Seminal fluid iron 1.01 0.37 0.44 2.72 0.009 0.26 1.76 0.77 1.28 Seminal fluid transferrin − 3.75 1.77 − 0.34 − 2.11 0.04 − 7.33 − 0.17 0.77 1.28 Bold values indicat that statistical significance was considered at p < 0.05 Multivariate linear regression analysis of iron parameters effects on 5-hmC levels and CLBR Bold values indicat that statistical significance was considered at p < 0.05 Variables included in the multivariate regression model were selected on the basis of their statistically significant associations with 5-hmC levels observed in univariate analyses. A multivariate linear regression analysis was conducted to examine the combined effect of seminal fluid iron levels, serum iron levels, and serum TIBC levels on 5-hmC levels. A backward stepwise procedure was used in the multivariate linear regression model. Initially, all three iron parameters were included, and variables that were not statistically significant in affecting 5-hmC levels were eliminated one by one, starting with the variable having the largest p-value (Table  5 ; Models 1,2, 3). In the final multivariate model (Model 3), serum TIBC demonstrated a statistically significant positive association with sperm DNA 5-hmC levels (β = 0.000, standardized β = 0.33, p = 0.02, 95% CI 0.000–0.001). This indicates that higher serum TIBC levels are associated with increased sperm DNA hydroxymethylation. The standardized coefficient (β = 0.33) suggests a moderate positive relationship, meaning that as serum TIBC increases, global motile sperm DNA 5-hmC levels also increase. Collinearity statistics showed no significant multicollinearity issues, confirming the independent impact of each variable (Table  5 ) (Supplementary material 3). An exploratory approach was employed to identify potential predictors of CLBR among the available iron biomarkers and global sperm DNA methylation and hydroxymethylation levels. A multivariate linear regression analysis was conducted to explore factors influencing CLBR, including various seminal fluid and serum iron parameters and 5-mC/5-hmC levels. Due to multicollinearity, seminal ferritin was excluded from the model. After excluding seminal fluid ferritin due to multicollinearity, a backward stepwise regression was employed to refine the model. Initially, all independent variables were included, and non-statistically significant iron parameters affecting CLBR were systematically removed, starting with the variable having the highest p-value. The final model, referred to as Model 7, identified seminal fluid iron (p = 0.009) and seminal fluid transferrin (p = 0.04) as statistically significant predictors of CLBR (Table  5 ; Models 1,2,3,4,5,6,7). A 1 µg/dl increase in seminal fluid iron was associated with a 1.016% increase in CLBR, while a 1 mg/dl increase in seminal fluid transferrin was associated with a 3.754% decrease in CLBR (Table  5 ) (Supplementary material 3).

In the human body, iron is crucial for oxygen transport (e.g., hemoglobin), temporary oxygen storage (e.g., myoglobin), and energy production (e.g., cytochromes) [ 1 – 3 ]. After its absorption in the intestine, iron binds to plasma transferrin, which distributes it to various sites throughout the body [ 4 ]. Once inside the cell, iron is stored in ferritin [ 5 ]. Clinically, iron status can be evaluated through various serum markers, such as serum iron, ferritin and transferrin. Serum iron represents the concentration of iron circulating in the blood [ 6 ]. Moreover, an abnormal serum ferritin level may indicate a disorder in iron storage, while abnormal transferrin test results may suggest iron-deficiency anemia or iron overload [ 7 ]. Additionally, serum transferrin status can be assessed using several laboratory tests, including unsaturated iron-binding capacity (UIBC) and total iron-binding capacity (TIBC). Under normal conditions, only about one-third of transferrin is saturated with iron, leaving 67% available for binding, known as the unsaturated iron-binding capacity (UIBC). The UIBC test measures the unbound iron-binding sites on transferrin, while the TIBC test reflects the total capacity of the blood to bind iron with transferrin, encompassing both serum iron and UIBC [ 7 – 9 ]. Iron is crucial for spermatogenesis because germ cells undergo multiple mitotic and meiotic divisions, which require active DNA synthesis. Several enzymes essential for DNA synthesis—such as replicative DNA polymerases and primase—depend on iron as a structural or catalytic cofactor to function properly [ 10 , 11 ]. Moreover, iron is indispensable for mitochondriogenesis during spermatogenesis, as it is required for the assembly of iron–containing cofactors, such as iron–sulfur clusters and heme groups, which are essential for the proper formation and function of mitochondrial respiratory complexes [ 12 ]. Maintaining iron homeostasis is essential to prevent the generation of oxidative stress (OS)—an imbalance between reactive oxygen species (ROS) and antioxidants—that can lead to DNA, lipid, and protein damage [ 13 , 14 ]. Excess free iron, often resulting from excessive uptake or malabsorption, promotes ROS formation and contributes to OS [ 15 , 16 ]. Conversely, iron deficiency can lead to iron-deficiency anemia (IDA), which induces hypoxia, elevates oxidant levels, and impairs antioxidant defenses, further disrupting oxidative balance [ 17 – 19 ]. Multiple studies have demonstrated significant associations between iron biomarkers in seminal fluid or serum, sperm parameters, and sperm oxidative DNA damage [ 18 , 20 – 23 ]. Additionally, OS has been shown to reduce the activity of ten-eleven translocation (TET) enzymes [ 24 ]. TET enzymes are expressed in several tissues, including the brain and testicles [ 25 ]. TET activity relies on ferrous iron (Fe 2+ ), α-ketoglutarate, and succinate. TET proteins catalyze the oxidation of 5-methylcytosine (5-mC) to form 5-hydroxymethylcytosine (5-hmC), both of which are DNA modifications that serve as epigenetic marks influencing gene expression [ 26 ]. Of particular concern, iron metabolism during pregnancy has been associated with offspring neurodevelopment and cognitive function, as shown in observational studies. A meta-analysis of 1,286 mother–newborn pairs found that maternal ferritin levels during pregnancy were associated with reduced DNA methylation at specific CpG sites in cord blood, with partial persistence in peripheral blood during childhood [ 27 , 28 ]. While exposure to an unhealthy diet in men may negatively affect offspring health [ 29 ], the specific impact of iron metabolism on changes in sperm DNA methylation and its potential implications for offspring health remains largely unexplored. In men, TET enzymes are expressed from the spermatocyte stage to spermatozoa [ 30 ]. However, the impact of iron status in men on global DNA 5-mC and 5-hmC levels in spermatozoa remains unclear. Therefore, the primary aim of this study was to assess whether there was an association between various iron biomarkers and global DNA 5-mC and 5-hmC levels in spermatozoa. The secondary aim was to evaluate whether these parameters were associated with cumulative live birth rates (CLBR) in infertile patients.

The present study indicated that serum TIBC was positively associated with sperm 5-hmC (%), while seminal iron showed a positive association and transferrin a negative association with CLBR in men of infertile men. These findings suggest that monitoring and managing iron levels may play a crucial role in male fertility. However, it is important to note that our results are observational, reflecting associations rather than direct causations. To establish a causal link, RCTs are needed to investigate the impact of dietary iron on semen quality, potentially contributing to the development of dietary guidelines for men seeking to optimize fertility. Moreover, this study suggested a potential role for iron in DNA 5-hmC profiling in sperm. Future research should use next-generation sequencing technologies to map 5-hmC changes across specific genes in relation to iron biomarkers, providing deeper insights into the influence of iron on sperm genomic integrity. Additionally, longitudinal studies with larger cohorts of infertile patients are recommended to assess the long-term effects of iron status and sperm DNA 5-hmC changes on offspring health.

The primary objective of this study was to investigate the possible associations between different iron biomarkers and global DNA methylation and hydroxymethylation levels in motile-sperm-enriched fractions. Multivariate analysis revealed a statistically significant positive association between serum TIBC and sperm DNA 5-hmC levels (standardized β = 0.33, p = 0.02), indicating that higher TIBC levels are associated with increased DNA hydroxymethylation. This relationship suggests that fluctuations in iron levels, reflected by an increase in TIBC due to elevated transferrin in response to low iron availability [ 57 ], may influence key epigenetic modifications in sperm. Particularly, this observation supports the idea that iron homeostasis could modulate the TET enzyme activity, which are responsible for converting 5-mC to 5-hmC, a key step in active DNA demethylation [ 24 ]. 5-hmC, an epigenetic mark, was initially identified in the T-even bacteriophage. It results from the oxidation of 5-mC, a process catalyzed by TET proteins. TET proteins, whose enzymatic activity depends on the presence of ferrous iron (Fe 2 + ), α-ketoglutarate, and succinate play a crucial role in regulating gene expression in cells [ 25 ]. Of distinct importance, various metabolites resulting from nutrient intake and cellular metabolism can influence the activity of TET enzymes [ 26 ]. For instance, glucose may undergo multiple metabolic pathways, generating byproducts that affect the formation of 5-hmC. In cancer cells, mutations in Krebs cycle enzymes can lead to the production of R-2-hydroxyglutarate, which suppresses TET activity and reduces 5-hmC levels [ 26 ]. Moreover, cellular exposure to oxidative stress has been found to alter both overall and site-specific 5-hmC levels in genes associated with the oxidative stress response [ 58 ]. This can be explained by studies showing that oxidative stress impairs the reduction of ferric iron (Fe 3+ ) to ferrous iron (Fe 2+ ), which deactivates the Fe 2+ and α-KG-dependent TET1-3 enzymes, thereby disrupting the DNA demethylation process [ 24 ]. These mechanisms highlight the susceptibility of 5-hmC levels to metabolic and oxidative changes. Importantly, 5-hmC plays a pivotal role in gene regulation and embryonic development. It is stably retained at specific genomic regions up to the eight-cell stage in human embryos, where it marks active enhancers involved in embryonic genome activation [ 59 , 60 ]. In the context of infertility, decreased expression of TET2 and TET3 enzymes in spermatozoa has been associated with decreased fertilization and pregnancy rates following ICSI [ 30 ]. These findings underscore the critical role of 5-hmC and TET enzymes in epigenetic regulation during embryo development [ 25 ]. Future research should use advanced techniques such as DNA immunoprecipitation and high-throughput sequencing to map epigenetic modifications in sperm in the context of iron levels [ 61 ]. The secondary aim of this study was to determine whether iron biomarkers in serum and seminal fluid were associated with CLBR in infertile patients undergoing ICSI. To address this, we first compared iron biomarker levels between seminal fluid and serum, then examined their relationships with sperm parameters, and finally assessed their impact on CLBR. First, the study found significant differences in transferrin, iron, TIBC, and UIBC levels between seminal fluid and serum. These variations may be related to the selective secretion of testicles and male accessory glands, as well as the unique environment necessary for sperm metabolism and function [ 23 ]. Transferrin in seminal fluid originates mainly from the testes, with additional contributions from the prostate and seminal vesicles [ 62 ]. Notably, the study found a significant positive correlation between serum transferrin and seminal fluid transferrin. This finding suggests that systemic iron regulation is partially reflected in seminal plasma, supporting the notion of coordinated iron homeostasis across body compartments [ 23 , 63 ]. Second, this study identified significant correlations between certain seminal and serum iron biomarkers and sperm concentration and motility. Iron homeostasis plays a crucial role in male fertility, but its balance is delicate. Inadequate iron levels may lead to oxidative stress [ 13 , 14 , 20 ], a well-established contributor to male infertility for over 35 years, has been shown to impact sperm motility, viability, morphology, and DNA integrity. These alterations can result in poor pregnancy outcomes [ 64 , 65 ]. Of particular interest, excess free iron (e.g., due to continued excessive iron uptake or malabsorption issues) serves as a substrate for the formation of reactive oxygen species (ROS), leading to oxidative stress [ 15 , 16 ]. In this context, a series of studies have explored the relationship between elevated seminal fluid iron levels, oxidative stress, impaired semen parameters, and DNA damage highlighting the potential negative impact of iron overload on male fertility [ 14 , 20 , 66 – 68 ]. In contrast, iron deficiency is a prevalent disorder in humans that can lead to iron-deficiency anemia (IDA). IDA creates a hypoxic environment in various organs, including the testes, which can increase oxidant levels and weaken antioxidant defenses. These combined effects, including reduced oxygen availability and impaired oxidative balance, have been associated with decreased sperm count and motility, ultimately leading to male infertility [ 17 – 19 ]. Alongside, several studies have found an association between transferrin levels and sperm quality, suggesting a possible role in male infertility [ 21 – 23 ]. Third, the multivariate analysis identified seminal fluid iron and seminal fluid transferrin as significant predictors of CLBR. The positive impact of seminal fluid iron on CLBR may stem from its essential role in spermatogenesis, mitochondrial function, and DNA integrity [ 5 , 11 ]. This positive effect could also be explained by the unique characteristics of our study population, where subclinical iron deficiency, rather than iron overload, may be more prevalent. In such cases, moderate increases in iron may support sperm function without inducing oxidative damage. In contrast, the negative association of CLBR with seminal fluid transferrin may indicate a compensatory response to lower iron bioavailability [ 57 ], as transferrin levels rise to facilitate iron transport under deficient conditions. This iron deficiency can lead to cellular hypoxia and increased production of reactive oxygen species, resulting in oxidative stress [ 17 – 19 ]. In this context, high transferrin may serve as a marker of an imbalance in the seminal fluid microenvironment, potentially impairing iron-dependent sperm functions and reducing CLBR. From this angle, a systematic review and meta-analysis conducted on 1,690 male partners of women with recurrent spontaneous abortion (RSA) and 1,337 male partners of fertile control women indicated a potential association between semen parameters and the risk of RSA [ 69 ]. It is worth mentioning that the findings from a systematic review and meta-analysis indicate that sperm DNA fragmentation may increase the risk of recurrent pregnancy loss [ 70 ]. Plus, particular attention should be given to oxidative DNA damage in human spermatozoa, as this factor may contribute to a substantial mutation load in offspring [ 71 ]. Given these considerations, we acknowledge the importance of evaluating sperm DNA fragmentation and other uterine factors in future studies. Specifically, future research should explore the impact of serum and seminal fluid iron biomarkers on sperm DNA integrity and how these may influence CLBR outcomes. This could provide a more comprehensive understanding of the factors influencing live birth rates and help refine the conclusions drawn from our study. While our study found no direct association between sperm DNA hydroxymethylation and live birth, its correlation with systemic iron transport raises the question of whether epigenetic modifications are more relevant to offspring health than to immediate reproductive outcomes [ 25 , 72 ]. This idea is supported by studies showing that paternal exposure to an unbalanced diet during the pre-conception period may lead to acquired pathologies that can be passed down to future generations in a non-Mendelian manner [ 73 ]. This inheritance of paternal acquired pathologies may occur due to the sensitivity of paternal gametes to metabolic stressors, which arise from the presence of various receptors for growth factors, hormones, and cytokines in germ cells [ 74 , 75 ]. At the molecular level, a growing body of scientific literature demonstrates that such dietary exposure can modify the sperm epigenome [ 74 , 76 ]. These modifications include differential chemical changes to DNA, alterations in the activity of non-coding RNAs, and changes in post-translational modifications of histones [ 76 ]. These epigenetic changes are hypothesized to play a role in the potential intergenerational or transgenerational transmission of paternal conditions acquired through unbalanced diet [ 74 , 76 ]. While there are several clinical trials and meta-analyses exploring the connection between male fertility and diet, the current evidence regarding the impact of nutrition on semen parameters remains insufficient [ 77 – 80 ]. From this perspective, a study conducted on healthy male university students (n = 209) found a statistically significant inverse association between dietary iron intake, assessed via a validated food frequency questionnaire, and both sperm concentration and the percentage of progressively motile sperm [ 81 ]. Dietary iron exists in two forms: heme (e.g. hemoglobin within red blood cells) and non-heme (e.g. ferric citrate) [ 82 – 84 ]. Animal-based foods, such as meat, poultry, eggs, dairy, and seafood, provide both heme and non-heme iron, while plant-based sources like nuts, beans, legumes, and fortified grains contain only non-heme iron [ 85 ]. Future randomized controlled trials (RCTs) with appropriate sample sizes, conducted on clearly defined participants, and using rigorous inclusion and exclusion criteria, are necessary to test the impact of iron supplementation and iron-rich foods on semen parameters. Given the dual role of iron in oxidative stress and enzymatic activity, these studies should also consider potential benefits and risks based on individual iron status and underlying causes of male infertility. Nonetheless, when assessing infertile patients, it is important to consider nutritional factors to complement medical treatments, tailoring the approach by inquiring about food choices and providing counseling on the benefits of a healthy diet when appropriate [ 80 ]. Additionally, future cohort studies could assess whether paternal unbalanced iron homeostasis at the time of conception affects offspring health and identifies which body systems are impacted. A key strength of this study lies in its prospective design, which minimizes recall bias and enables a more accurate assessment of associations [ 86 ]. The clinical accessibility of serum iron markers supports their potential utility as predictors of CLBR and sperm DNA hydroxymethylation in infertile men—an observation that warrants further investigation. However, the observational nature of the study limits causal inference [ 86 ]. The single-centre setting and absence of direct assessments of sperm DNA damage and oxidative stress constrain the ability to clarify the underlying mechanisms linking systemic iron status to epigenetic alterations and ICSI outcomes. Moreover, interpretation of CLBR should be approached with caution, as several potentially important factors—including serum estradiol and progesterone levels, and embryo source (fresh vs. frozen)—were either unavailable or could not be adequately evaluated due to the limited number of freeze-all cycles in our cohort [ 45 , 87 – 89 ]. Future studies should aim to incorporate these parameters and focus on identifying optimal iron levels in seminal fluid that support reproductive success while minimising oxidative damage.

Additional file 1. Additional file 2. Additional file 3. Additional file 1. Additional file 2. Additional file 3.