The medium used for semen washing and centrifugation was Equiplus®, an antibiotic-free equine semen extender (Minitüb GmbH, Germany). Curcumin (Sigma-Aldrich, Cat. No. C7727) was prepared as a stock solution by dissolving 1 mg in 1 mL of DMSO. A 1:10 dilution was made to obtain a working concentration of 100 μg/mL, and further 1:2 dilutions were made in PBS to reach the desired concentrations, ensuring that the final DMSO concentration in the sample did not exceed 1%.

For microbiological analyses, two different media were employed depending on the purpose: MacConkey Agar, a selective and differential medium for Gram-negative bacteria, prepared in-house in the university laboratory, and Petrifilm® Aerobic Count plates (3 M, Barcelona, Spain), containing standard methods agar nutrients, a cold water-soluble gelling agent, and a red indicator dye for total aerobic counts.

Epididymal spermatozoa were obtained from the epididymides of 12 stallions collected at the local slaughterhouse (Mercazaragoza, Zaragoza) and transported to the laboratory at room temperature. Spermatozoa were collected using the retrograde epididymal tail washing technique (19). Briefly, the epididymis was dissected, and a 25 G needle was inserted into the vas deferens and connected to a syringe containing 10 mL of Equiplus® medium without antibiotics at room temperature. Manual pressure was applied to collect spermatozoa into a beaker.

Following collection, initial sperm motility was assessed using the ISAS® computer analysis system (Proiser, Spain). Only samples with motility between 30% and 40% were used. This range was selected to standardize baseline sperm quality among samples and to mimic suboptimal but clinically relevant conditions frequently encountered in epididymal recoveries, thereby facilitating the detection of both protective and detrimental effects of curcumin during storage. Sperm samples were then diluted with Equiplus® to a final concentration of 50 × 10⁶ sperm/ml, centrifuged at 1000 × g for 5 min at room temperature, and the supernatant was removed. Each pellet was subsequently resuspended in the four experimental extenders corresponding to the different curcumin concentrations: Control (Equiplus® + 0.5% DMSO), C1 (Equiplus® + 0.5% DMSO + 0.125 mM), C2 (Equiplus® + 0.5% DMSO + 0.25 mM), and C3 (Equiplus® + 0.5% DMSO + 0.5 mM). Finally, the samples were prepared in doses containing a total of 25 × 10⁶ sperm/ml and stored in Eppendorf tubes at 4 °C for up to 96 h.

Before motility assessment, samples were warmed in a thermoblock at 37 °C for 5 min to allow recovery after cold storage. Sperm motility and movement patterns were objectively assessed using a computerized analysis system (ISAS®; Proiser, Valencia, Spain). For each evaluation, a volume of 4 μL was placed on a microscope slide at 37 °C, with three fields per sample, each containing a minimum of 200 spermatozoa. The set parameters were 25 frames/s, where spermatozoa had to be present in at least 15 frames to be counted.

The parameters evaluated were total motile spermatozoa (TM%), curvilinear velocity (VCL, μm/s), straight-line velocity (VSL, μm/s), average path velocity (VAP, μm/s), curvilinear trajectory linearity (LIN; ratio of VSL/VCL, %) and lateral head amplitude (ALH, μm). Spermatozoa with a mean velocity (VAP) ≥ 20 μm/s were considered motile, whereas those with a VAP < 10 μm/s were considered immotile. Patterns of motile quality were also analysed (40).

Plasma membrane integrity (viability) and acrosomal integrity were jointly assessed by double staining with fluorescein isothiocyanate–conjugated peanut agglutinin (PNA-FITC) and propidium iodide (PI) (43). Samples were incubated with FITC-PNA (1 mg/mL) and PI (500 μg/mL) for 5 min at 37 °C and subsequently fixed in 4% paraformaldehyde. A minimum of 200 spermatozoa per sample were evaluated under a fluorescence microscope (Leica DM2500 LED, Spain). Spermatozoa excluding PI were considered viable, while FITC-PNA labeling of the apical head region indicated acrosomal damage (Figure 1).

To evaluate bacterial growth using two complementary approaches, semen samples were analysed by culture on MacConkey agar plates for the selective detection of Gram-negative bacteria, and on Petrifilm® Aerobic Count plates to estimate overall aerobic cultivable bacterial growth under the incubation conditions used.

For each sample, a 1:10 dilution was prepared by mixing 100 μL of semen with 900 μL of sterile saline solution. From this dilution, 1 mL was inoculated onto MacConkey agar plates using the pour plate technique, specifically aimed at detecting Gram-negative aerobic and facultative anaerobic bacteria. In parallel, 1 mL of the same dilution was inoculated onto Petrifilm® Aerobic Count plates (3 M, St. Paul, MN, USA) following the manufacturer’s instructions to assess total aerobic cultivable bacteria. All samples were processed in duplicate for each method.

Plates were incubated aerobically at 37 °C for 24 h. After incubation, colony-forming units (CFU) were counted, and results were expressed as CFU/mL after correction for the dilution factor. Results obtained from both culture systems were used to compare bacterial growth among treatments and storage times.

Strict anaerobic bacteria were not evaluated, as neither MacConkey agar nor Petrifilm® Aerobic Count plates, under aerobic incubation, are suitable for their detection.

Sperm quality, including total motility, kinetic parameters, viability, and acrosome integrity, as well as bacterial growth assessed by traditional culture methods and Petrifilm®, were evaluated in relation to the addition of different concentrations of curcumin to stallion sperm extenders. A total of 12 samples (n = 12) were analyzed after 1 and 96 h of refrigerated storage at 5 °C. The overall experimental design is presented in Figure 2.

Statistical analysis was performed using IBM SPSS Statistics version 22.0. For each storage time (1 and 96 h), quantitative variables were analysed using a univariate general linear model for repeated measures, with curcumin treatment (Control, C1, C2, C3) as a within-subject factor and stallion as the subject (random) factor. Prior to analysis, model residuals were checked for normality and homoscedasticity. Data are presented as mean ± standard error. When a significant main effect of treatment was detected, pairwise comparisons between groups were carried out with Bonferroni adjustment for multiple testing. Different superscript letters within a row indicate statistically significant differences between treatments (p < 0.05).

Figures 3 and 4 present the mean values of sperm quality parameters evaluated after 1 and 96 h of refrigerated storage. Figure 3 shows the sperm quality parameters after 1 h of refrigeration, whereas Figure 4 shows the corresponding values after 96 h of storage. After 1 h of refrigeration, supplementation of the extender with curcumin did not result in significant differences in total sperm motility, plasma membrane integrity (viability), or acrosomal integrity when compared with the control group. Although a slight reduction in total motility was observed in samples supplemented with the highest curcumin concentration (0.5 mM; C3), this decrease was not statistically significant.

After 96 h of refrigerated storage, differences among treatments were observed (Figure 4). Samples supplemented with 0.25 mM (C2) and 0.5 mM (C3) curcumin showed significantly lower percentages of spermatozoa with intact plasma membrane and acrosomal integrity compared with the control and 0.125 mM (C1) groups. Total sperm motility was significantly reduced in the C3 group compared with the control, whereas no significant differences were detected between the control and C1 samples.

The extender supplemented with 0.125 mM curcumin (C1) did not significantly affect sperm quality parameters at either evaluation time compared with the control extender.

Analysis of the individual factors included in the statistical model revealed that, at the 1-h evaluation, significant differences were associated exclusively with the stallion factor (p < 0.001). In contrast, after 96 h of refrigeration, significant effects were detected for stallion, curcumin concentration, and the stallion × curcumin interaction (p < 0.001), indicating that both individual variability and treatment influenced sperm quality parameters during prolonged storage.

The effect of curcumin on motility parameters after cold storage at 4 °C for 1 and 96 h is summarized in Table 1. After 1 h of refrigeration, significant differences among treatments were observed for several kinematic parameters. Curvilinear velocity (VCL) showed higher values in the control and C1 samples than in the C2 and C3 groups. Average path velocity (VAP) showed a similar pattern, with significantly higher values in the control and C1 samples compared with C2 and C3. In contrast, straightness (STR) values were significantly higher in the C2 and C3 groups compared with the control and C1 samples. The C3 group showed higher values for beat cross frequency (BCF) compared with the other concentrations.

In relation to other parameters, VSL (straight-line velocity), LIN (linearity), WOB (wobble), and ALH (lateral head displacement), no significant differences were observed among treatments at 1 h of refrigeration.

After 96 h of refrigerated storage, curcumin concentration influenced several motility parameters. Samples supplemented with 0.5 mM curcumin (C3) showed significantly lower values for VCL, VSL, and VAP compared with the control, C1, and C2 groups. No significant differences in velocity parameters were detected among the control, C1, and C2 samples. Control and C1 semen samples showed significantly higher values for ALH and BCF than C2 and C3. Conversely, wobble (WOB) values were significantly higher in the C3 group compared with the control.

Analysis of the statistical model indicated that, at 1 h of refrigeration, the stallion factor had a significant effect on most motility parameters, reflecting inter-individual variability. After 96 h of storage, curcumin concentration and the stallion × curcumin interaction significantly influenced most velocity-related parameters (Table 2).

Bacterial growth in semen samples during refrigerated storage was evaluated after 1 and 96 h using two methods: traditional culture on MacConkey agar and Petrifilm® Aerobic Count plates. Both methods were used to estimate cultivable aerobic bacterial growth under the different curcumin concentrations.

After 1 h of refrigeration, traditional bacterial culture showed significantly higher bacterial growth in the control samples compared with all curcumin-supplemented groups. No significant differences were observed among the different curcumin concentrations. In contrast, Petrifilm® analysis revealed similar bacterial counts in the control, C1, and C2 groups, whereas samples supplemented with 0.5 mM curcumin (C3) showed significantly lower bacterial growth compared with the other treatments (Figure 5).

After 96 h of refrigerated storage, both traditional culture and Petrifilm® analysis demonstrated significantly higher bacterial growth in the control samples compared with curcumin-supplemented extenders. Samples supplemented with 0.25 mM (C2) and 0.5 mM (C3) curcumin showed the lowest bacterial counts in Petrifilm® plates. In traditional culture, samples supplemented with 0.125 mM curcumin (C1) also showed significantly lower bacterial growth compared with the control (Figure 6).

Overall, curcumin supplementation was associated with reduced cultivable aerobic bacterial growth during refrigerated storage compared with the control extender without curcumin, with more pronounced differences observed after prolonged storage.

Refrigerated storage of equine spermatozoa is associated with a progressive decline in sperm quality, mainly due to cold shock, oxidative imbalance, and alterations of plasma and acrosomal membranes (4). In this study, curcumin was evaluated primarily to determine whether its antimicrobial properties could be exploited during storage without compromising basic sperm quality parameters. Under short-term refrigerated conditions, curcumin supplementation did not induce detectable changes in total motility, sperm viability, or acrosomal integrity, regardless of the concentration tested. This finding indicates that free curcumin does not exert acute toxic effects on equine epididymal spermatozoa during early cooling and that sperm function is initially preserved. The absence of early effects is consistent with the buffering capacity of equine semen extenders, which are designed to maintain sperm quality during the first phase of refrigerated storage (4, 5, 37).

In contrast, after prolonged storage, a clear dose and time dependent response became evident. Higher curcumin concentrations were associated with reduced sperm viability and acrosomal integrity, together with alterations in motility, whereas the lowest concentration tested did not differ from the control. These results suggest that curcumin may exert cumulative detrimental effects on spermatozoa when present at elevated concentrations for extended periods, rather than inducing immediate toxicity.

This concentration-dependent behavior is consistent with previous findings in the equine species. Rossi et al. (31) reported that low curcumin concentrations did not affect stallion sperm motility after short-term storage, whereas higher concentrations significantly reduced motility (44). Although storage conditions and experimental design differ, both studies support the existence of a threshold concentration above which curcumin becomes detrimental to equine spermatozoa, which may be related to a dose-dependent transition from antioxidant to pro-oxidant effects under specific storage conditions (45).

Curcumin is known to display a dual antioxidant and pro-oxidant behavior depending on concentration, exposure time, and cellular environment (30, 46). At higher concentrations, curcumin may interfere with membrane lipid organization, mitochondrial activity, or intracellular redox homeostasis, ultimately compromising sperm viability and membrane integrity. Importantly, none of the concentrations tested in the present study improved sperm quality parameters, indicating that, under refrigerated conditions, free curcumin should not be considered a sperm-protective additive but rather a compound with a limited safety window. Recent evidence further supports this interpretation. Nasiri-Foomani et al. (47) demonstrated that curcumin-loaded niosomal nanoparticles significantly improved motility, membrane functionality, and lipid peroxidation in cooled stallion semen, whereas free curcumin showed weaker or inconsistent effects.

These findings suggest that the antioxidant potential of curcumin may be constrained by its physico-chemical properties, such as limited aqueous solubility and stability, which could restrict its effective bioavailability under semen storage conditions. In this context, the lack of sperm-protective effects observed in the present study is consistent with the use of free curcumin at the concentrations tested.

Beyond total motility, computer-assisted sperm analysis revealed that curcumin supplementation influenced sperm movement characteristics in a concentration- and time-dependent manner (48, 49). Under short-term refrigeration, higher curcumin concentrations were associated with reduced velocity-related parameters and increased straightness, indicating slower but more rectilinear trajectories. While progressive movement is required for fertilization, reduced velocity may compromise sperm transport efficiency within the female reproductive tract (50).

After prolonged storage, these alterations became more pronounced. Exposure to the highest curcumin concentration resulted in marked reductions in velocity parameters together with decreased ALH and BCF, suggesting impaired flagellar beat amplitude and frequency. These changes are compatible with compromised axonemal function and altered energy metabolism, and they likely precede or accompany the observed decline in sperm viability and membrane integrity (51).

Together, these findings indicate that elevated curcumin concentrations do not simply reduce the proportion of motile spermatozoa but also alter the qualitative features of sperm movement, which may have functional implications for fertilization. This reinforces the importance of considering detailed kinematic parameters when evaluating the biological impact of extender additives.

Semen is not a sterile biological fluid, and bacterial contamination is a well-recognized factor negatively affecting sperm viability, fertility potential, and storage lifespan (12, 37). During refrigerated storage, bacterial metabolism may contribute to the production of reactive oxygen species and other toxic by-products that impair sperm membrane integrity, mitochondrial activity, and motility (52). In addition, the use of contaminated semen has been associated with an increased risk of post-breeding endometritis in mares, highlighting the clinical importance of controlling microbial proliferation in semen preservation protocols (13).

In the present study, supplementation of an antibiotic-free extender with curcumin consistently reduced cultivable bacterial growth during refrigerated storage. This inhibitory effect was observed using both MacConkey agar and Petrifilm® Aerobic Count plates, indicating a genuine antimicrobial action of curcumin under the experimental conditions employed. The reduction in bacterial growth became more evident after prolonged refrigeration, supporting a cumulative antimicrobial effect over storage time.

Comparable antimicrobial effects of curcumin and other natural bioactive compounds have been previously reported in semen preservation systems across species (19, 34, 52, 53). Although curcumin is generally less potent than conventional antibiotics, its capacity to limit bacterial proliferation without contributing to antimicrobial resistance underscores its potential as a complementary strategy in semen extenders. Importantly, antimicrobial activity is likely less dependent on intracellular uptake and bioavailability than antioxidant protection, which may explain why free curcumin effectively reduced bacterial growth despite failing to improve sperm quality parameters in the present study (34).

The microbiological assessment in this study was focused on cultivable aerobic and facultative anaerobic bacteria, as bacterial growth was evaluated using MacConkey agar—primarily selective for Gram-negative microorganisms—and Petrifilm® Aerobic Count plates for total aerobic bacterial counts. This methodological approach is supported by previous studies reporting that Gram-negative aerobic and facultative anaerobic bacteria, particularly members of the Enterobacteriaceae family, constitute the predominant microbial population in equine semen during storage (12, 37, 52). Strict anaerobic bacteria were not evaluated under the culture and incubation conditions employed and are generally considered less prevalent in routinely collected and refrigerated semen samples. Consequently, the antimicrobial effects observed in the present study should be interpreted as referring mainly to the bacterial populations most associated {Citation} with semen contamination and quality deterioration during refrigerated storage.

Epididymal spermatozoa provide a valuable experimental model for microbiological studies, as they are antibiotic-naïve and typically exhibit lower and more homogeneous bacterial contamination than ejaculated semen (36, 41, 42). This allows the evaluation of antimicrobial additives without confounding effects from seminal plasma or prior antibiotic exposure.

Although epididymal sperm initially lack seminal plasma proteins involved in sperm activation, washing and dilution in appropriate extenders can partially restore motility and functionality (53, 54). Importantly, several studies have reported acceptable pregnancy rates using cooled or cryopreserved epididymal sperm in horses, supporting its relevance for assisted reproduction and genetic conservation (42).

Clinically, epididymal sperm recovery represents the only option for preserving genetic material in cases of sudden death or severe trauma. While ejaculated semen differs in composition and microbial complexity, the epididymal model provides a relevant and practical framework for evaluating antimicrobial strategies under controlled conditions.