Semen extender triggers a mild physiological inflammatory response in the uterus without disrupting sperm-uterine immune crosstalk in vitro in cattle

Abstract

Artificial insemination (AI) in cattle involves introducing frozen-thawed sperm, a minimal amount of seminal plasma, and a significant volume of semen extender (SE) into the uterus. Previous studies have demonstrated that sperm interact with bovine endometrial epithelia via TLR 2/1, triggering a weak inflammatory response to clear the endometrium. This study investigated the impact of the major component of the insemination dose, egg yolk-based SE, on the uterine immune response in vitro. The results showed that SE did not affect sperm kinetic parameters or the entry of sperm into the uterine glands. SE alone significantly upregulated the mRNA expression of inflammatory cytokines (NFKB2, TNF, IL1B, CXCL8), TLR2/1, and the inflammasome NLRP3, while downregulating NOD1. Immunofluorescence analyses confirmed the upregulation of the strong inflammatory marker TNF alongside TLR2 and the downregulation of NOD1 in the uterine epithelium, similar to the effects observed with sperm. When combined with sperm, SE did not enhance the protein or mRNA expression of these markers, except for IL1B and CXCL8. In silico analyses revealed a strong affinity between triglycerides (the primary components of egg yolk) and TLR2/1, suggesting a potential role in stabilizing heterodimerization. These findings demonstrate that egg yolk-based SE, independent of sperm, triggers a mild physiological inflammatory response mediated by the TLR2/1 and NOD1 signaling pathways. The suppression of NOD1 by sperm and SE ensures a controlled and weak immune response in the uterus. Notably, despite the SE-induced inflammation, the sperm-uterine immune crosstalk was not disrupted, suggesting that SE does not negatively impact the physiological interactions between sperm and the uterus during AI.

Artificial insemination (AI) has revolutionized the breeding industry, particularly for cattle, by offering greater efficiency and genetic control than those achievable through natural mating [1]. During AI, semen is extensively diluted with a semen extender (SE) to maximize the number of females that can be inseminated from a single ejaculation. Semen extenders serve multiple purposes, including pH regulation [2], bacterial contamination control [3], mitigation of cryogenic damage [4, 5], reduction of oxidative stress [6], and the provision of essential energy sources [7]. Egg yolk-based extenders have been widely used to preserve the viability of bovine sperm during cryopreservation [8]. In addition to egg yolk, cryoprotectants such as glycerol, antibiotics, vitamins, and other additives are incorporated into SE formulations [7, 9, 10]. When extended semen is packaged in a straw, most of the content is SE, with only minimal amounts of seminal plasma (SP) and sperm [8]. Artificial insemination involves the direct deposition of semen into the uterus, with a large volume of SE, facilitating the direct interaction between sperm and the uterine microenvironment.

Upon deposition in the bovine uterus, sperm induce an acute, transient inflammatory response. This response is crucial for preparing the endometrium for decidualization, receptivity and embryo implantation [11, 12]. It is characterized by the upregulation of pro-inflammatory cytokines and rapid infiltration of leukocytes, predominantly polymorphonuclear neutrophils (PMNs), into the uterine lumen to remove excess sperm and pathogens [13]. Toll-like receptors (TLRs), some of the earliest described pattern recognition receptors (PRRs), play crucial roles in the innate immune system [14]. Notably, in cattle, sperm utilize the TLR2/1 signaling pathway to trigger a weak inflammatory response [15,16,17].

Nucleotide-binding oligomerization domain receptors (NOD-like receptors, NLRs) are pivotal intracellular PRRs involved in diverse biological processes beyond pathogen recognition [18]. Among NLR family members, NOD1 and NOD2 have been the most extensively studied [19]. NOD1 is expressed in various cell types, whereas NOD2 is restricted to hematopoietic cells [20]. Increased NOD1 expression in human decidual stromal cells is associated with recurrent pregnancy loss [21], whereas its downregulation in the ovine endometrium during early pregnancy suggests a potential role in maintaining a receptive endometrial environment [22]. Although the role of NOD1 in bovine pregnancy remains relatively unexplored, these findings underscore its importance in regulating immune responses and maintaining pregnancy.

Another well-studied NLR, NLRP3, is crucial for inflammasome assembly [23]. Inflammasomes are multi-protein complexes that, upon recognizing pathogen-associated molecular patterns or damage-associated molecular patterns (DAMPs), recruit and activate Caspase-1. Activated Caspase-1 cleaves pro-IL1B and pro-IL18, producing mature inflammatory cytokines [24]. Although NLRP3 activation is essential for various physiological processes, including embryo implantation in humans, its dysregulation is associated with recurrent pregnancy loss and other reproductive disorders. In bovine pregnancy, the specific role of NLRP3 is not well understood [25, 26]. However, given its importance in other species, it likely plays a significant role in immune regulation and pregnancy maintenance in cattle.

Previous studies have shown that washed bull sperm (sperm without SE) induce an acute and transient inflammatory response in the bovine uterus [16, 27,28,29]. However, in addition to the sperm, the insemination dose contains a substantial amount of SE. Both fresh and frozen SE from different origins trigger immune responses, particularly by inducing neutrophil extracellular trap (NET) formation in vitro in cattle [30]. This raises the question of whether SE can modulate sperm-triggered uterine inflammatory response during AI. We hypothesized that exposure of bovine endometrial tissues to insemination dose components, particularly egg yolk-based SE, may modulate sperm-triggered inflammatory mediators and alter the uterine immune microenvironment. Furthermore, we proposed that sperm and/or SE could regulate NOD signaling alongside TLR2/1 in the inflammatory cascade.

To test this hypothesis, we conducted in-depth investigations to elucidate the molecular mechanisms underlying the regulation of the uterine immune cascade by SE and sperm. Using in vitro and ex vivo approaches, we assessed the impact of SE and sperm in modulating the uterine immune milieu. Additionally, in silico analyses examined potential ligands within the SE (triglycerides [TGs], phospholipids, and cholesterol [31]) and their role in uterine inflammation. This knowledge will advance our understanding of breeding-induced endometrial immune modulation and ultimately contribute to the improvement of AI techniques in cattle.

Materials and Methods

Ethics statement

All animal experimental procedures were approved by the Committee on the Ethics of Animal Experiments of the Obihiro University of Agriculture and Veterinary Medicine (OUAVM) (Permit number 22-222).

Experimental model

We used in vitro, ex vivo, and in silico approaches to investigate the immunological responses of bovine endometrial epithelium to SE. To assess the impact of semen-induced inflammation on the bovine uterus at the cellular level, bovine endometrial epithelial cells (BEECs) were co-cultured with SE or sperm. Semen-induced protein expression changes in the bovine uterus were analyzed using immunofluorescence (IF) in a semen-uterine explant model. To investigate whether SE directly influences sperm behavior, we analyzed sperm motility using computer-assisted sperm analysis (CASA) and assessed sperm-uterine gland interactions in explants. In silico analysis was conducted to explore the immune effects of potential ligands within SE. The detailed experimental methodology is illustrated in Fig. 1.

Fig. 1.

Schematic illustration of the experimental methodology. (A) Separation of sperm and semen extender (SE). Frozen-thawed commercial semen straws were used to separate sperm from SE. After the first centrifugation of semen, a sperm pellet was obtained. The supernatant was then centrifuged twice to obtain the SE. (B) In vitro culture. A simplified in vitro co-culture model was used to investigate the inflammatory response of bovine endometrial epithelial cells (BEECs) challenged by semen components for 3 h. Co-cultured media were analyzed for sperm kinetics using computer-assisted sperm analysis. RT-qPCR analysis was conducted to investigate the mRNA expression of inflammatory cytokines, Toll-like receptors (TLRs), and NOD-like receptors (NLRs). (C) Ex vivo culture. To evaluate the sperm/SE-induced inflammatory response in the endometrium, bovine endometrial explants were co-cultured with semen components for 3 h. Immunostaining analyses were performed to evaluate the inflammatory marker TNF and receptor markers TLR2, NOD1, and NLRP3 proteins in the endometrium. (D) In silico analysis. Docking and molecular dynamics (MD) simulations were used to investigate the immune effects of potential ligands within the SE. The binding affinity of three major SE components—triglycerides, phospholipids, and cholesterol—were explored for their interaction with TLR2 dimers.

Isolation and in vitro culture of BEECs

Macroscopically healthy bovine uteri from post-pubertal, non-pregnant cattle, free from microbial inflammation, abnormal coloration, or pathological lesions, were collected at a slaughterhouse (Doto Plant Tockachi Factory, Obihiro, Japan). The whole reproductive tracts were transported to the laboratory under sterile conditions, and the uterine horn, ipsilateral to the active follicle, was used for cell culture. Epithelial cells were isolated and cultured following a previously described method [32], with modifications. The uterine lumen was washed twice with 25 ml of sterile Ca²⁺- and Mg²⁺-free Hanks’ balanced salt solution (HBSS) supplemented with 1% v/v penicillin-streptomycin (15140122, Gibco, Thermo Fisher Scientific, Waltham, MA, USA) and 1 mg/ml bovine serum albumin (BSA) (A7888, Sigma-Aldrich, St. Louis, MO, USA). Twenty-five milliliters of enzyme solution (sterile HBSS containing 500 μg/ml collagenase I [C2674, Sigma-Aldrich], 50 μg/ml deoxyribonuclease I [043-26773, FUJIFILM Wako pure chemical corporation, Osaka, Japan], and 1 mg/ml BSA) was then infused into the uterine lumen via a catheter. Epithelial cells were isolated by incubation at 38.5°C for 45 min. The cell suspension was filtered through a double-layer of 100-μm nylon mesh. The filtrate was washed twice by centrifugation (10 min at 180 × g) with Dulbecco’s modified Eagle medium (DMEM/F12, 12400024, Gibco, Thermo Fisher Scientific) supplemented with 1% v/v amphotericin B (15290018, Gibco, Thermo Fisher Scientific), 1% v/v penicillin-streptomycin, and 100 μg/ml BSA. The final pellet was resuspended in 5 ml of equilibrated culture media (DMEM/F12 supplemented with 1% v/v amphotericin B, 50 μg/ml gentamicin sulfate [G1264, Sigma-Aldrich], and 10% v/v heat-inactivated fetal bovine serum [FBS, S1400, Ireland-origin, Biowest, Nuaillé, France]). The cells were then seeded in 25 cm² flasks (156340, Nunc, Thermo Fisher Scientific) at 38.5°C in a humidified atmosphere with 5% CO₂. The culture medium was replaced every 48 h. Upon reaching 70–80% confluence, the cells were collected with trypsinizing (2.5 mg/ml trypsin; 7409, Sigma-Aldrich in 0.2 mg/ml EDTA; D05125 Sigma-Aldrich), transferred to 24-well plates (142475, Nunc, Thermo Fisher Scientific), and cultured up to around 90% confluence (first passage). β-estradiol (E8875, Sigma-Aldrich, 1 ng/ml) and progesterone (P8783, Sigma-Aldrich, 50 pg/ml) were added to the culture media (DMEM/F12, 1% v/v amphotericin B, 50 μg/ml gentamicin, and 5% v/v FBS) at preovulatory concentrations [33].

Sperm and SE preparation

Commercial semen straws were acquired from three Holstein bulls with known fertility belonging to the Genetics Hokkaido Association in Hokkaido, Japan. Three semen straws from different bulls were thawed each time in a water bath at 38.5°C for 30 sec, pooled, and centrifuged at 300 × g for 5 min to separate sperm pellet from the SE-containing supernatant. The supernatant was carefully collected and centrifuged at 500 × g for 5 min, twice, to remove any sperm present in the SE. The absence of sperm in SE was confirmed microscopically prior to each experiment. The obtained SE was used as the cryopreserved SE with sperm (SE ^+sperm). Additionally, freshly prepared SE diluents A (egg yolk, trisodium citrate, glycerin, antibiotics) and B (egg yolk, trisodium citrate, glycerin, antibiotics, glucose, fructose, and raffinose) were obtained from the Genetics Hokkaido Association and frozen in the laboratory at −30°C. The two diluents were mixed in a 1:1 ratio in the laboratory to prepare the commercial SE at the time of the experiment. This was used as the SE frozen without exposure to sperm (SE ^-sperm).

Cell culture challenge with SE and sperm

The sub-confluent BEEC monolayers (after the first passage) were washed twice with phosphate-buffered saline (PBS, P3813, Sigma-Aldrich) and equilibrated in a medium supplemented with 0.1% (v/v) FBS and 50 μg/ml gentamicin at 38.5°C. The BEECs were exposed either to culture medium alone or to medium containing 5%, 10%, or 20% (v/v) SE. In the successive model, BEECs were co-cultured with 5 × 10⁶/ml sperm with or without SE (5%, 10%, or 20% [v/v]). Additionally, BEECs were challenged in the same experimental setting with either control medium alone or sperm alone, 5%, 10%, or 20% (v/v) SE separated from the semen straw (SE ^+sperm); or 5%, 10%, or 20% (v/v) newly prepared frozen commercial SE (SE ^-sperm). Following the application of treatments, BEECs were incubated at 38.5°C in a humidified 5% CO₂ environment for 3 h, followed by washing the cells twice with PBS, lysing with TRIzol (15596018, Invitrogen, Thermo Fisher Scientific), and storing at −80°C until RNA extraction. These experiments were repeated five times using epithelial cells from five different uteri (n = 5).

Blockage of TLR2 pathway

For blockage of the TLR2 pathway in BEECs, sub-confluent BEECs monolayers in 24-well plates were incubated with TLR2/1 antagonist (CU-CPT22; 614305, Sigma-Aldrich) at 0.1 μM (36.24 ng/ml for 3 h, then washed before being exposed to the sperm or SE for an additional 3 h.

Assessment of sperm motility parameters using CASA

After a 3-h co-culture period, BEEC-co-culture media primed with sperm or sperm with SE (5%, 10%, or 20% [v/v]) were used to assess sperm motility and kinetic parameters. The CASA system (SMAS, DITECT, FHK, Tokyo, Japan) was used as previously described [34, 35]. The sperm motility and kinematic parameters included total motility (%), progressive motility (%), velocity along the straight line (VSL, μm/sec), velocity along the curvilinear/actual path (VCL, μm/sec), velocity along the average/smooth path (VAP, μm/sec), amplitude of lateral head (ALH, μm), beat cross frequency (BCF, Hz), straightness (STR, %), linearity (LIN, %), and hyperactivated motility (%).

PCR protocol

RNA extraction, cDNA synthesis, and RT-qPCR were performed as described in the previous protocol [36]. The extracted RNA was resuspended in RNA storage solution (AM7001, Invitrogen, Thermo Fisher Scientific) and quantified using a spectrophotometer (DS-11, DeNovix, Wilmington, DE, USA). A total of 1 μg of RNA was used for cDNA synthesis as previously described [35], and the synthesized cDNA was stored at −30°C. RT-qPCR of target genes —nuclear factor kappa B subunit 2 (NFKB2), tumor necrosis factor (TNF), interleukin 1 beta (IL1B), C-X-C motif chemokine ligand 8 (CXCL8), Toll-like receptor 1 (TLR1), Toll-like receptor 2 (TLR2), Toll-like receptor 4 (TLR4), Toll-like receptor 6 (TLR6), nucleotide-binding oligomerization domain containing 1 (NOD1), NLR family pyrin domain containing 3 (NLRP3), and actin beta (ACTB) (Supplementary Table 1)— was carried out using SYBR Green Supermix (1725271B02, Bio-Rad Laboratories, Hercules, CA, USA) on an iCycler iQ (Bio-Rad Laboratories) [37]. The calculated cycle threshold (Ct) values were normalized using ACTB as an internal housekeeping gene, and fold change between samples was quantified using the ΔΔCt method [38].

Challenge of endometrial explants with SE and sperm

Preovulatory endometrial explants were prepared according to a method described previously [27]. Endometrial explants were exposed to either modified Tyrode balanced salt solution (TALP) alone, or a medium containing 5 × 10⁶/ml sperm, 10% (v/v) SE, or 5 × 10⁶/ml sperm with 10% (v/v) SE for 3 h. Three independent experiments were performed using explants from three individual cows (n = 3). At the end of the experiment, the explants were rinsed with TALP and fixed in 4% paraformaldehyde solution (16120141, FUJIFILM Wako) for immunofluorescence analysis.

Assessment of sperm number in uterine glands

The sperm number within the uterine glands was assessed based on a previous report [29]. To visualize the behavior of sperm in the endometrial explants, videos of JC1 (502560032, AdipoGen, San Diego, CA, USA)-labeled sperm in the explants were recorded after 30 min. Recorded videos were used to count the sperm in the glands using the ImageJ software version (1.54 g). Counts were made for three glands of equal size in each experimental treatment.

Immunofluorescence analysis

The IF was performed based on a previous report, with minor modifications [35]. Briefly, the tissues were incubated overnight with primary antibodies against TNF, TLR2, NOD1, and NLRP3 (Supplementary Table 2) at 4°C in a humidified chamber. Afterward, the sections were incubated with Alexa Fluor-conjugated secondary antibodies (Supplementary Table 2) for 30 min. The sections were then washed, and coverslips were mounted using mounting medium containing DAPI (H-1200, Vector Laboratories, Newark, CA, USA). Finally, the sections were observed under a fluorescence microscope (BZ-X800, Keyence, Osaka, Japan) to visualize the distribution of TNF, TLR2, NOD1, and NLRP3 proteins in the endometrial tissue. The fluorescence intensity of the images was analyzed using ImageJ [37].

Statistical analyses

Data are presented as the mean ± standard error of mean (SEM) from 3–5 independent experiments. Statistical analyses were performed using the SAS software (SAS Institute Inc., Cary, NC, USA). One-way analysis of variance (ANOVA), followed by Bonferroni’s test, was used to compare the mean differences among the groups. The significance of the data was determined based on their respective p-values, where * P < 0.05, ** P < 0.01, *** P < 0.001, and ^### P < 0.001.

Computational methodology

Triglycerides, phospholipids (Dipalmitoylphosphatidylcholine, DPPC), and cholesterol are the major components of egg yolk [31], with potential involvement in inflammation through TLR receptors, particularly TLR2 and TLR4 [39]. Therefore, we investigated whether TG, phospholipids, and cholesterol could serve as potential ligands for TLR2 dimerization forms. To explore the affinity of TG, phospholipids, and cholesterol for TLR2 dimers (TLR2/1 and TLR2/6), we utilized bovine TLR2 structures modeled via homology modeling, along with TG, phospholipid, and cholesterol structures obtained from CHARMM-GUI. Pam3CSK4 (PAM3) and Pam2CSK4 (PAM2) served as positive controls because they are well-known ligands for TLR2/1 and TLR2/6 heterodimers, respectively. The effects of TG, phospholipids, and cholesterol as potential ligands for TLR2 were evaluated by comparing them with these positive controls [16, 40]. Docking simulations using AutoDock VINA assessed the interactions between TG, phospholipids, cholesterol, PAM3, and PAM2 with TLR2 dimers, and the structures were optimized through molecular dynamics (MD) simulations using GROMACS 2020 with CHARMM 27 force field parameters.

Further computational details, including docking, MD validation, glycoprotein preparation, MD parameters, and MM/PBSA binding free energy (BFE) calculations, can be found in the Supplementary Methods.

Results

Semen extender did not modulate sperm kinetic parameters

To evaluate whether SE had any effect on sperm, sperm motility and kinetic parameters were analyzed using CASA in BEECs co-culture media. The BEECs were primed for 3 h with either sperm alone or sperm in combination with SE. None of the SE concentrations tested (5%, 10%, and 20%) altered the sperm parameters, including the percentage of total, progressive, and hyperactively motile sperm, and the mean values of sperm motion parameters such as VSL, VCL, VAP, LIN, STR, BCF, and ALH compared with those in the control (Supplementary Fig. 1).

Semen extender did not modulate the sperm number within the uterine gland in bovine uterine explants

To evaluate whether SE modulates sperm interactions with the endometrial epithelia, sperm-endometrial interactions were analyzed using the sperm-endometrial ex vivo model. Fluorescence video microscopy of sperm labeled with JC-1-stain on sperm midpieces showed that the sperm glided over the surface epithelium and entered the uterine gland, as previously described [29]. We evaluated motile sperm within the glands after 30 min of incubation with either sperm alone or sperm with 10% SE, with explants. Sperm with 10% SE did not differ from the sperm-alone group in the number of sperm entering the uterine gland (Supplementary Fig. 2).

Semen extender and sperm upregulate the expression of inflammatory cytokines in BEECs

BEECs were exposed to either the control medium alone or 5%, 10%, or 20% (v/v) SE alone for 3 h to evaluate the impact of SE on BEECs. SE alone upregulated the inflammatory cytokines in BEECs. Briefly, exposure to 5% and 10% SE increased the epithelial cell expression of IL1B (P < 0.01) and CXCL8 (P < 0.001) (Fig. 2A). Moreover, exposure of epithelial cells to 20% SE significantly increased the expression of all inflammatory genes: NFKB2, TNF, (P < 0.05), IL1B (P < 0.01), and CXCL8 (P < 0.001) compared with those in the control (Fig. 2A).

Fig. 2.

The impact of semen extender (SE) and sperm on inflammatory gene expression in bovine endometrial epithelial cells (BEECs). (A) mRNA expression of the transcription factor NFKB2 and inflammatory cytokines TNF, IL1B, and CXCL8 in BEECs stimulated with SE (5%, 10%, and 20%). (B) mRNA expression of TLRs (TLR2, TLR1, TLR4, and TLR6) and NOD1 in BEECs stimulated with SE (5%, 10%, and 20%). (C) mRNA expression of transcription factor NFKB2 and inflammatory cytokines TNF, IL1B, and CXCL8 in BEECs stimulated with sperm and sperm plus SE (5%, 10%, and 20%). (D) mRNA expression of TLRs (TLR2, TLR1, TLR4, and TLR6) and NOD1 by the stimulation with sperm and sperm plus SE (5%, 10%, and 20%). (E) mRNA expression of transcription factor NFKB2 and inflammatory cytokines (TNF, IL1B, and CXCL8) in control BEECs and TLR2 antagonist-pretreated BEECs stimulated with 10% SE. Data are presented as the mean relative expression ± SEM from 5 independent experiments. Data were analyzed using the ANOVA with Bonferroni’s mean comparisons procedure. Asterisks indicate significant differences compared with the control alone. Hash marks indicate significant differences compared with sperm (* P < 0.05, ** P < 0.01, *** P < 0.001, ^### P < 0.001).

BEECs were stimulated with either sperm (5 × 10⁶/ml) alone or sperm with SE (5%, 10%, or 20% [v/v]) to evaluate the impact of SE on the sperm-triggered inflammatory response. Sperm alone increased the expression of TNF (P < 0.05), IL1B, and CXCL8 (P < 0.001) (Fig. 2C) in BEECs. All combined sperm and SE groups showed increased expression of all inflammatory genes (NFKB2, TNF (P < 0.05), IL1B, and CXCL8 (P < 0.001)), compared with those in the control. Furthermore, 20% SE upregulated the sperm-induced expression of IL1B, whereas 5%, 10%, and 20% SE exhibited similar effects on CXCL8 (P < 0.001) (Fig. 2C).

Semen extender and sperm both utilize TLR2/1 signaling in BEECs

Twenty percent of SE increased the expression of both TLR2 and TLR1, whereas 10% SE induced only TLR2 expression (P < 0.05) (Fig. 2B). None of the SE concentrations tested increased the expression of TLR4 or TLR6 (Fig. 2B). Sperm induced the expression of both TLR2 and TLR1 (P < 0.05), but did not affect TLR4 nor TLR6 expression (Fig. 2D). All combined sperm and SE groups showed increased expression of TLR2 (P < 0.05), whereas only 10% SE with sperm increased the TLR1 (P < 0.05) expression (Fig. 2D). None of the combined treatment groups altered expression of TLR4 or TLR6 (Fig. 2D).

TLR2/1 blockage prevented the SE-stimulated pro-inflammatory genes BEECs

To elucidate the functional role of TLR2/1 in SE-BEECs interaction, BEEC monolayers were incubated with a TLR2/1 antagonist for 3 h prior to SE exposure for 3 h. Our RT-qPCR results showed that the pre-incubation of BEECs with 0.1 μM of TLR2/1 antagonist prevented the SE-stimulated NFKB2, TNF, and CXCL8 expression. (Fig. 2E).

Semen extender and sperm downregulated NOD1 and upregulated NLRP3 gene expression

Gene expression of NOD1 and NLRP3 was evaluated following exposure of BEECs to control medium alone or 5%, 10%, or 20% (v/v) SE alone for 3 h. Similarly, experiments were conducted with sperm (5 × 10⁶/ml) alone or sperm with SE at 5%, 10%, or 20% (v/v). All treatment groups showed upregulation of NLRP3 (P < 0.01 and P < 0.001) compared with that in the control medium alone (Figs. 2A and 2C). The 10% and 20% SE treatments significantly upregulated the sperm-induced expression of NLRP3 (P < 0.001) (Fig. 2C). Notably, reduced expression of NOD1 (P < 0.001) was observed in both sperm and SE groups (Figs. 2B and 2D).

Immunofluorescence localization of TNF, TLR2, NOD1, and NLRP3 in bovine endometrial epithelium challenged by sperm or SE

Immunofluorescence analysis confirmed TNF, TLR2, NOD1, and NLRP3 protein localization in the endometrial tissues of the sperm, 10% SE, and sperm with 10% SE treatment groups. Owing to the natural variability in tissue morphology and the specific plane of sectioning, some differences in the structural appearance were observed among the micrographs. However, the essential structural elements, such as the luminal and glandular epithelium and stroma, were consistently present and recognizable across all images. The expression of the investigated proteins was concentrated in the endometrial luminal and glandular epithelia, with less immunoreactivity observed in the stromal tissues. When sperm, 10% SE, or a combination of sperm and 10% SE were incubated with uterine explants, all treatments significantly upregulated TNF (P < 0.05) and TLR2 (P < 0.01) expression in uterine tissues (Figs. 3A and 4A). Conversely, all treatment groups showed significantly reduced levels of NOD1 (P < 0.01) protein in uterine tissues (Fig. 4B). Although NLRP3 mRNA was upregulated in the invitro BEEC culture, NLRP3 protein was not modulated in the ex vivo model (Fig. 3B).

Fig. 3.

Immunofluorescence staining of TNF and NLRP3 inflammatory proteins in bovine endometrial explants exposed to sperm and semen extender. (A) The TNF protein expression (green) in bovine endometrial explants exposed to sperm and semen extender (SE) (10%). DAPI was set as the nuclear counterstain (blue), with a merged image showing DAPI in blue and TNF in green. Normal mouse IgG was used as the primary antibody in the IgG control. (B) The NLRP3 protein expression (red) in bovine endometrial explants exposed to sperm and SE (10%). DAPI was used as the nuclear counterstain (blue), with a merged image showing DAPI in blue and NLRP3 in red. Normal rabbit IgG was used as the primary antibody in the IgG control. Semiquantitative scoring of the corrected total cell fluorescence (CTCF) of TNF and NLRP3 expressed as arbitrary fluorescence units. CTCF values are presented as the mean relative expression ± SEM from 3 independent experiments. Data were analyzed using the ANOVA with Bonferroni’s mean comparisons procedure. Asterisks indicate significant differences (* P < 0.05). Scale bar = 50 μm.

Fig. 4.

Immunofluorescence staining of TLR2 and NOD1 receptors protein in bovine endometrial explants exposed to sperm and semen extender (SE). (A) TLR2 protein expression (red) in bovine endometrial explants exposed to sperm and 10% SE. DAPI was used as the nuclear counterstain (blue), with a merged image showing DAPI in blue and TLR2 in red. Normal rabbit IgG was used as the primary antibody in the IgG control. (B) NOD1 protein expression (green) in bovine endometrial explants exposed to sperm and 10% SE. DAPI was used as the nuclear counterstain (blue), with a merged image showing DAPI in blue and NOD1 in green. Normal mouse IgG was used as the primary antibody in the IgG control. Semiquantitative scoring of corrected total cell fluorescence (CTCF) of TNF and NLRP3 expressed as arbitrary fluorescence units. CTCF values are presented as the mean relative expression ± SEM from three independent experiments. Data were analyzed using the ANOVA with Bonferroni’s mean comparisons procedure. Asterisks indicate significant differences (** P < 0.01). Scale bar = 50 μm.

Semen extender separated from frozen semen straw and newly prepared frozen SE both similarly induce inflammatory gene expression in BEECs

BEECs were stimulated with control medium alone, sperm alone, 5%, 10%, or 20% (v/v) SE separated from semen straw (SE ^+sperm); or 5%, 10%, or 20% (v/v) newly prepared frozen commercial SE (SE ^-sperm). The SE preparations did not show differences in inflammatory gene expression between the SE preparations. Both SE preparations induced inflammatory gene expression compared with the control medium alone (Supplementary Fig. 3; P < 0.01, P < 0.001).

Triglycerides show strong affinity for stabilizing TLR2/1 heterodimerization

In this study, we evaluated the interactions of TGs, phospholipids, and cholesterol—the main components of egg yolk—with the TLR2/1 and TLR2/6 dimers using in silico methods. Docking and 300 ns MD simulations demonstrated that TG remained stably bound to the main binding sites of both TLR2 dimers (Fig. 5A). Binding free energy calculations using the MM/PBSA method revealed a significant affinity between TG and TLR2 dimers, suggesting the TG as a potential ligand for TLR2 signaling pathways (Supplementary Table 3). Additionally, center of mass analysis confirmed that TG effectively penetrated and interacted with the hydrophobic binding pockets of TLR2/1 and TLR2/6, supported by hydrophobic interactions with key residues (Fig. 5B and Supplementary Figs. 4 and 5). For the other ligands, phospholipids easily penetrated the main pocket of TLR2 in both dimer forms; however, they did not strongly interact with TLR1 or TLR6, suggesting a possible antagonistic effect of phospholipids (Supplementary Fig. 6). Additionally, cholesterol also penetrated the TLR2 pocket in both TLR2/1 and TLR2/6 dimers, indicating an antagonistic role (Supplementary Fig. 7).

Fig. 5.

Molecular binding of triglycerides (TG) with Toll-like receptors (TLR2/1 and TLR2/6). (A) The images demonstrate the positions of TG molecules and TLRs at the start (after docking) and the end (after 300 ns of molecular dynamics [MD] simulation). Docking predicts how a small molecule (like TG) might fit into a specific site on a protein. MD simulation models the movements and interactions of atoms over time, allowing us to observe how TG and the TLRs interact and change shape. Enlarged views show fatty acid chains of TG penetrating the main binding site of TLR2 and TLR1 and interacting with key residues to stabilize TLR2/1 heterodimer. In contrast, for the TLR2/6 complex, the fatty acids primarily interact with the main binding site of TLR2, but not with TLR6 without stabilizing the TLR2/6 heterodimer. TLR2, TLR1, and TLR6 are shown in green, magenta, and orange, respectively. TG is represented in stick form. The structures for TLR2/1-PAM3 and TLR2/6-PAM2 (positive control forms) are not shown here. (B) The graph shows the distance between the center of mass (COM) of TG, PAM3, and PAM2 and the primary TLR binding sites over time. The COM represents the average position of the mass of each molecule. Measuring its distance to the binding site of TLRs helps track how TG moves into the binding pocket and supports dimer stabilization by remaining bound to the receptors.

Discussion

Following AI, sperm, a key component of the insemination dose, induces a weak immune response in the bovine uterus via the TLR2/1 pathway [16, 29]. However, the potential immune impact of the SE, another major component of the insemination dose, has not been extensively explored. In the present study, we investigated the effect of an egg yolk-based SE on uterine immune responses in cattle. Our findings revealed that the SE itself could activate the TLR2/1 signaling pathway, leading to the release of inflammatory cytokines and a mild inflammatory response in the bovine uterus, independent of sperm. Interestingly, the cytokines released by both the sperm and extender appeared to suppress NOD1 signaling, resulting in a balanced, weak, and transient uterine inflammatory response that is crucial for endometrial preparation and implantation.

A typical ejaculate from a dairy bull contains approximately 5–10 billion sperm, which can be used to inseminate 300–1,000 cows when appropriately extended. Both egg yolk- and skimmed milk-based SEs have been employed for bovine semen cryopreservation; however, egg yolk-based extenders have been particularly prevalent because of their proven effectiveness [8]. When extended semen is packaged in 0.25 ml or 0.5 ml straws, only about 2% of the total volume consists of SP and 1% of the sperm pellet. The remaining 97% comprised the SE [41]. Thus, in addition to sperm, the SE could play a significant role in the uterine immune response after AI.

To investigate whether the SE directly affected sperm kinetics or sperm-uterine interactions, we evaluated its impact on these parameters. Our data revealed that the SE did not modulate sperm kinetics or the ability of sperm to enter the uterine glands, indicating that the extender did not directly influence sperm in modulating uterine immune responses.

Further analysis revealed that the purified endometrial epithelial cells exhibited increased expression of the inflammatory markers NFKB2, TNF, IL1B, and CXCL8 following exposure to the extender, similar to the response observed after sperm exposure. When the extender was combined with sperm, the combined group demonstrated an upregulation of the inflammatory response only in IL1B and CXCL8 compared to sperm alone. Similar to sperm, the extender induced mRNA expression of TLR2/1 and protein expression of TLR2 in endometrial explants. Furthermore, the TLR2/1 blockade in BEECs did not upregulate the expression of inflammatory cytokines. These findings suggest that the SE induces inflammation through the same TLR2/1 signaling pathway, independent of sperm.

Previous studies investigating SE-induced immune responses in cattle are limited. Recently, one group quantified and compared NETs formation following the in vitro incubation of bovine PMNs with SEs of different origins and conditions [30]. Their findings demonstrated that SE, both fresh and frozen, induced NETs formation in PMNs, with fresh SEs exhibiting greater capacity for NETs formation. NETs are critical components of the innate immune response, and their formation is one of the primary mechanisms by which neutrophils eliminate invading pathogens, including sperm. Although we did not directly investigate NET formation in our study, our findings on SE-induced cytokine expression in the bovine endometrium aligned with those of previous reports.

Analysis of the SE separated from frozen-thawed semen straws revealed no significant difference in uterine inflammatory gene expression compared with the newly prepared frozen extender. This suggests that sperm or any bioactive compounds derived from sperm during the cryopreservation do not interfere with the impact of the SE on the inflammatory response.

To identify the specific compounds within the egg yolk-based extender responsible for the observed effect, we initially focused on TGs (62%), phospholipids (33%), and cholesterol (5%), which are the major components of egg yolk-based extenders [31]. Computational analysis was employed to evaluate the potential interactions between these components and the TLR2 signaling pathway. Our results indicated that TGs could act as ligands for TLR2, particularly through TLR2/1 and TLR2/6 dimers. Previous studies have demonstrated that egg yolk TG can induce the production of inflammatory cytokines in the small intestine [42]. Furthermore, de-lipidation of egg yolk, which reduces TGs, significantly inhibits NFKB activation by downregulating TLR2 [43]. Additionally, increased TG levels are associated with a significant TLR2 upregulation, suggesting a direct association between TGs and TLR2 signaling [39]. These results suggest that TLR2 has a strong capacity to act as a TG receptor. Collectively, these findings reveal that TGs from egg yolk play a pivotal role in regulating the TLR2/1 pathway, contributing to the observed inflammatory response in the present study.

Although the role of TLR2/1 cascade in sperm-uterine immune crosstalk has been extensively studied, the involvement of other PRRs found in the bovine endometrium in sperm/semen-induced inflammation remains unclear. We focused on NLRs, the second-most-characterized PRR family in mammalian cells [44]. Among the NLRs, members of the NLRC subgroups, such as NOD1 and NOD2, have been implicated in initiating innate immune responses to intracellular pathogens and DAMPs, playing key roles in the immune system [45]. NOD1 is widely expressed in various cell types, including both hematopoietic and non-hematopoietic origins, whereas NOD2 is restricted to cells of hematopoietic origin and intestinal epithelia [20]. Therefore, we focused on the role of NOD1 in sperm/extender-triggered uterine inflammation. Our results revealed a marked decrease in NOD1 expression, both at the mRNA and protein levels, following the challenge of the endometrial epithelia with semen components.

Previous studies have demonstrated an association between NOD1 and reduced inflammation [22, 46]. Following parturition, cows experience reduced PMN function, which is associated with the suppression of inflammation. The suppression of NOD1 expression may contribute to reduced PMN function [47]. Additionally, the interaction between NOD1 and TLR2 was investigated in human enterocyte-like Caco-2/TC7 cells, which demonstrated a clear interaction [48]. TLR2/1 activation by Pam3csk4 significantly decreases NOD1 expression [48]. These findings collectively suggest that the interplay between the TLR2/1 and NOD1 signaling pathways plays a crucial role in regulating the inflammatory response induced by SE in the bovine endometrium.

The proposed working hypothesis for the SE- and sperm-induced inflammatory response in the endometrium, and the maintenance of immune homeostasis is illustrated in Fig. 6. Molecular patterns on sperm and in SE, possibly including TGs, are initially sensed by TLR2/1, triggering signal transduction through MYD88-dependent downstream signaling, leading to the release and translocation of NFKB and ultimately the production of pro-inflammatory cytokines, primarily TNF, pro-IL1B, and chemokines like CXCL8. Upon receiving the priming signal from TLR2/1, NLRP3 forms an inflammasome and activates pro-IL1B. CXCL8 is a potent neutrophil chemoattractant that plays a crucial role in recruiting immune cells to sites of inflammation in the endometrium [49, 50]. Although insemination-induced inflammation in the uterus is physiological, it must be maintained in a weak, transient state [13]. Increased expression of CXCL8, along with other pro-inflammatory cytokines, may contribute to the recruitment of neutrophils and other immune cells to the endometrium. Thus, the released inflammatory cytokines may suppress the NOD1 receptor through a negative feedback loop, thereby reducing NFKB to maintain the immune homeostasis and keeping the sperm- and SE-induced inflammation weak at optimal levels.

Fig. 6.

Schematic illustration of the working hypothesis for semen extender and sperm regulation and maintenance of immune homeostasis in the endometrium. The following numbers indicate the steps depicted in red numbers within purple circles in the working hypothesis. 1. Molecular patterns on sperm and in semen extender (possibly including triglycerides) are sensed by transmembrane protein TLR2/1, initiating the NFKB pathway. 2. Activated NFKB translocates to the nucleus, inducing pro-IL1B, TNF, and CXCL8 cytokine expression. 3. Simultaneously, NFKB-induced NLRP3 leads to IL1B processing. 4. Released inflammatory cytokines may suppress NOD1 in the cytoplasm, possibly through negative feedback, thereby reducing NFKB to maintain the uterine immune response at a physiological and weak level.

In conclusion, our findings demonstrated that egg yolk-based SE, independent of sperm, triggered a mild physiological inflammatory response in the bovine endometrium. This response was mediated by the activation of the TLR2/1 and NOD1 signaling pathways. The interplay between these pathways, in response to sperm and SEs, plays a crucial role in regulating and maintaining the uterine immune response at controlled, weak levels, which is essential for preparing the endometrium for implantation without causing tissue damage. Despite SE-induced inflammation, the immunological crosstalk between the sperm and uterus remained intact, suggesting that SE did not adversely affect the physiological interactions between sperm and the uterine environment during AI. Further investigations are essential to comprehensively understand the intricate relationship between SE, sperm, and uterine immunity, including the role of NOD1, in enhancing AI outcomes in cattle.

Conflict of interests

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as potential conflicts of interest.