Integrating proteomics and metabolomics to evaluate impact of semen collection techniques on the quality and cryotolerance of goat semen Authors: Chunyan Li, Jiachong Liang, Larbi Allai, Bouabid Badaoui, Qingyong Shao, Yina Ouyang, Guoquan Wu, Guobo Quan, Chunrong Lv Published: 2024 ABSTRACT AbstractsResults of artificial insemination (AI) are affected by changes in sperm quality and the function throughout collection and preservation procedures. Proteome and metabolome alterations of sperm treated with the different procedures in goat, however, aren’t fully understood. To this end, we sought to investigate the impacts of rectal probe electrostimulation (EE) and artificial vagina (AV) semen collection methods on the quality and the cryotolerance of goat sperm, with additional focus on proteomic and metabolomic analyses. Semen samples were collected from Yunshang black goats and categorized into four groups: fresh sperm collected via AV (XAZ), fresh sperm collected via EE (XEZ), frozen sperm post-AV collection (DAZ) and frozen sperm post-EE collection (DEZ). Four comparisons (XAZ vs. XEZ, DAZ vs. XAZ, DEZ vs. XEZ, DAZ vs. DEZ) were performed, respectively. This study first evaluated sperm motility, acrosome integrity, plasma membrane integrity, mitochondrial activity, and reactive oxygen species (ROS) levels. The results indicated that there were no significant differences in fresh sperm quality parameters between the EE and AV methods. However, notable differences emerged post-cryopreservation. Specifically, the AV method proved more advantageous in preserving the motility, integrities of acrosome and plasma membrane, mitochondrial activity of frozen sperm compared to the EE method. Through the multi-omics approaches, a total of 210 differentially abundant proteins (DAPs) related to sperm characteristics and function were identified across the four comparations. Moreover, 32 differentially abundant metabolites (DAMs) were detected. Comprehensive bioinformatics analysis underscored significant molecular pathways in the co-enrichment of DAPs and DAMs, particularly focusing on the citrate cycle, ROS, oxidative phosphorylation, and glycine, serine, and threonine metabolism etc. We elucidated the differential impacts of AV and EE collection methods on the quality and cryotolerance of goat semen from omics perspectives, which offer a critical foundation for further exploration into optimizing semen collection and cryopreservation techniques in goat breeding program. Introductions The global decline in livestock genetic diversity has prompted numerous countries to initiate comprehensive gene bank conservation programs. These initiatives focus on the collection, processing, storage, and management of animal genetic resources, including germplasm cells, blood, tissues and DNA, which necessitate significant investments1. Central to these efforts is the establishment and refinement of sperm banks, where semen collection and cryopreservation are pivotal procedures. Nowadays, rectal probe electrostimulation (EE) and artificial vagina (AV) are two prevalent methods for livestock semen collection, each with its advantages and limitations. AV, simulating natural mating ejaculation, is usually preferred method. However, EE serves as a viable alternative in scenarios where males are untrained or estrous females are unavailable2. Concurrently, the advancement of semen cryopreservation technology, a practice has been ongoing over 70 years3, facilitates long-term semen preservation and significantly contributes to the advancement of artificial insemination (AI) in livestock daily production and breeding4. AI is an advanced assisted reproductive technique that makes a more efficient use of semen, meet the protection of livestock genetic resources and the increasing demand for livestock products5,6. Nonetheless, cryopreservation process can adversely affect semen quality, such as impairments of sperm motility, fertility due to physical–chemical and oxidative stress lesions7,8. Additionally, it’s also imperative to consider the adverse factors of semen freezing front-end, i.e. the choice of semen collection technology, which affecting the sperm’s response to the cryopreservation process. To the best of our knowledge, compared to the AV, the EE operation caused more stress and pain to male’s muscles because of electrical pulses, which affects animal welfare and semen quality like more volume but lesser concentration, altered pH and composition2,9,10, resulting in sperm’s different response to cryopreservation process. Hence, the impact of EE and AV on semen quality is a crucial area of study, particularly in the context of the beneficial effects of proteomics and metabolomics on semen cryobiology. Mature sperm is highly differentiated end-terminal cell that lacks transcriptional and translational activity, thus, molecular components of protein and metabolite are essential for maintaining sperm viability and functionality prior to oocyte fertilization11,12. Variations in ejaculate quality and post-thaw fertility attributable to the AV and EE collection methods have been documented, except for the unknown about the effects of AV and EE on molecular component changes of sperm2,13,14. Considering these findings, this study aimed to explore the differential impacts of AV and EE collection methods on the semen quality of Yunshang black goats, a newly recognized superior breed in China. The investigation will encompass both fresh and frozen sperm samples, employing an integrative multi-omics approach. Our goal is to identify key proteins and metabolites, along with their correlative biological pathways, to deepen our understanding on the molecular mechanisms of AV and EE collection methods differ on sperm quality. This insight is expected to inform the selection of optimal collection methods for the preservation of semen resources, thereby contributing to the broader goal of livestock genetic conservation. Materials and methods Ethics statement The ethical committee of Yunnan Animal Science and Veterinary Institute (Kunming city, Yunnan province, China) has approved all experiments including animal usage in this study (approval No. 202103001). All authors strictly obey Regulations on the Administration of Laboratory Animals (Order No.2 of the State Science and Technology Commission of People’s Republic of China, 1988) and Regulations on the Administration of Experimental Animals of Yunnan Province (the Standing Committee of Yunnan Provincial People’s Congress 2007.10). All methods in this study were carried out in accordance with relevant guidelines and regulations. All methods were reported in accordance with The ARRIVE guidelines 2.0. Experimental design and workflow To investigate different impacts of semen collection techniques (EE & AV) on the quality and cryotolerance of goat semen, the quality of fresh sperm and cryopreserved sperm collected by the two techniques was evaluated by varieties of analyses, respectively. Specific parameters analyzed included motility and motile indices, integrities of plasma membrane and acrosome, reactive oxygen species (ROS) levels and mitochondrial activity. At the same time, tandem mass tag (TMT)-based quantitative proteomic analysis combined with ultra-high performance liquid chromatography Q-Exactive Plus Orbitrap mass spectrometry (UHPLC-Q-Exactive Orbitrap MS/MS) analysis were performed to evaluate the potential differences of the two semen collection techniques on proteome and metabolome of fresh sperm and cryopreserved sperm, respectively (Fig. S1). Animal and semen preparation Ten healthy Yunshang black goats (approximately 2–3 years old and with live body weights of 82.87 ± 1.70 kg, and the scrotal peripheral volumes of 708.40 ± 32.74 mL) were randomly selected. All were housed at a breed protection farm from the Yi Xingheng Animal Science and Technology Co. Ltd (Kunming, Yunnan province, China, located at 102°37’ East and 24°57’ North, in 2022), and kept in the same condition of feeding and management. Fresh ejaculates (twice ejaculates per male in 10 min) were successively collected by rectal probe electrostimulation and artificial vagina, respectively. Then, the collected ejaculates of each male by the same collection method were pooled and placed in a water bath at 37.0 °C, and the pooled semen quality was immediately assessed using the computer-assisted sperm analysis system (CASA, Microptic, Barcelona, Spain). Only each pooled semen sample with a minimum volume of 1.0 mL, total motility greater than 75% and a concentration greater than 3 × 109 /mL could be used for the further analysis15,16. The pooled semen of each male was subsequently split to two aliquots. One aliquot was used as fresh sperm source (sperm source A) while the rest was used for cryopreservation (sperm source B). Hence the sperm source A consists of two types of sperm collected by rectal probe electrostimulation (XEZ group) and artificial vagina (XAZ group), the sperm source B consists of two types of sperm collected by rectal probe electrostimulation (DEZ group) and artificial vagina (DAZ group), respectively. Semen freezing medium The above semen aliquot (sperm source B) was gently diluted using the freezing medium at RT (25 °C) at a final sperm concentration of about 3 × 108 /mL17. The freezing medium was made up of 20% (v/v) AndroMed® (Minitube, Germany), 80% (v/v) double-distilled water, 5 mg/mL lactose (Sigma-Aldrich, USA), and 3 µg/mL vitamin B12 (Cisen Pharmaceutical, China)7. Freeze–thaw processing The extended semen samples were loaded into the labeled 0.25 mL straws (IMV technologies, L′Aigle, France) and equilibrated at 4 °C for 3 h, then placed in liquid nitrogen vapor for 8 min (-80 °C), immediately frozen in liquid nitrogen for a week18,19. For the thawing process, frozen specimens were incubated in a water bath at 37 °C for 30 s. All thawed semen were centrifuged at 10,000 × g for 3 min at 4 °C, and the pellets were washed thrice before being subjected to omics analysis18,20. Evaluation of sperm motility For all pooled semen samples (n = 20), the motilities were evaluated using the CASA which default parameters as previously described21. Plasma membrane and acrosome integrities Integrity of plasma membrane was assessed by the hypo-osmotic swelling test (HOST) with GENMED kit (GMS14017, GENMED Scientifics Inc. USA). Briefly, 10 µL of sperm sample was mixed with 100 µL of pre-heated hypo-osmotic solution and incubated at 37 °C for 30 min. Afterwards, the 10 µL mixture was smeared onto a pre-heated glass slide with a coverslip. 400 spermatozoa were counted under a phase-contrast microscope (Axio Vert A1, Germany). Spermatozoa with swelling of the head and tail, coiled or shortened of the tail were considered as intact. Integrity rate of spermatozoa was calculated7,22. Additionally, acrosome integrity was evaluated by fluorescein isothiocyanate labeled pea agglutinin (FITC-PSA) test, and the detailed method given in our previous report23. Mitochondrial activity and ROS levels Rhodamine 123 (R123) combined with PI were used to measure sperm mitochondrial activity. Firstly, 5 µL of post-thaw semen was added to 500 µL TALP solution, followed by 1.0 µL of Rh123 (0.5 mM). After 30 min of incubation at 37 °C in the dark, the pellet was collected after centrifugation at 500 × g for 3 min. 500 µL assay buffer and 5 µL of PI (300 mM) were used to suspend and incubate the pellet for 10 min. The mixture was evaluated by flow cytometer (BD Accuri™ C6, USA)22. For intracellular ROS, 5 μL of post-thaw semen was diluted with 500 μL TALP solution, followed by 0.5 μL DCFH-DA and 5 μL PI for 60 min at 25 °C in the dark; the mixture was evaluated by flow cytometer24. TMT-based proteomic investigation Protein extraction and digestion For the proteomic analysis, six male subjects were randomly selected from the ten goats tested previously mentioned. Thus, there were 24 bio-samples used in this part analysis based on the above experimental design. Firstly, all bio-samples underwent triple washing with isotonic phosphate-buffered saline (PBS), followed by centrifugation at 10,000 × g for 3 min at 4°C18. For the extraction of proteins, a 30 mg sperm pellet from each sample was subjected to sonication on ice using a high intensity ultrasonic processor, this process was conducted in a lysis buffer containing 8 M urea (Sigma, Germany), and 1% protease inhibitor cocktail (Calbiochem, Germany). Post-sonication, the supernatants were collected by centrifuging at 12,000 × g for 20 min at 4°C25. Protein concentrations in these samples were determined using the BCA protein assay kit (sparkJade, China), with the results detailed in Table S1. Subsequently, 15 μg of the extracted protein from each sample was resolved by 10% SDS-PAGE, and the protein profiles were visualized on gels (Fig. S2). For the digestion process, 300 µg of protein from each sample was reduced with DTT to a concentration of 100 mM at 100 °C for 5 min, then cooled and treated with 200 µL of UA buffer (8 M urea, 150 mM Tris–HCl, pH = 8.0). Post-centrifugation at 12,000 × g for 15 min, the supernatant was discarded, and the pellet was re-treated with 200 µL of UA buffer. Subsequent steps involved alkylation in the dark at room temperature (RT) for 30 min using 50 mM iodoacetamide (IAA, Sigma), followed by centrifugation at 12,000 × g for 10 min. The samples were then washed with 100 µL of UA buffer and centrifuged at 12,000 × g for 10 min. This step was followed by a buffer exchange with 100 µL of NH4HCO3 buffer and further centrifugation at 14,000 × g for 10 min. The digestion process involved an overnight incubation with trypsin at 1:50 trypsin-to-protein mass ratio, followed by a second digestion phase for 4 h at 1:100 trypsin-to-protein mass ratio26. TMT tag labeling and fractionation of peptides Peptides were initially desalted using a Strata X C18 Solid Phase Extraction (SPE) column (Phenomenex) and subsequently dried via vacuum centrifugation. Each peptide sample, quantifying 100 μg, was dissolved in 100 μL of triethylammonium bicarbonate (TEAB, 100 mM, pH8.5). These samples were then labeled using TMTpro 16-Plex reagents (ThermoFisher Scientific, USA)27,28. Briefly, the peptide solution was mixed with the TMTpro labeling regent (20 µL of reagent per 100 µL of peptides) and incubated for 1 h at RT. The regent was previously dissolved in 200 μL of anhydrous acetonitrile for 5 min at RT. Labeling of six pooled fractions from each group was performed as follows: PXEZ group labeled with 131N (PXEZ1, PXEZ4), 131C (PXEZ2, PXEZ5), and 132N (PXEZ3, PXEZ6) tags; PXAZ group labeled with 129C (PXAZ1, PXAZ4), 130N (PXAZ2, PXAZ5), and 130C (PXAZ3, PXAZ6) tags; PDEZ group labeled with 126 (PDEZ1, PDEZ4), 127N (PDEZ2, PDEZ5), and 127C (PDEZ3, PDEZ6) tags; PDAZ group labeled with 128N (PDAZ1, PDAZ4), 128C (PDAZ2, PDAZ5), and 129N (PDAZ3, PDAZ6) tags. Each batch of labeling included an internal standard reference sample, labeled with 132C. The reaction was quenched by adding 5 µL of 5% hydroxylamine to each sample and incubating for 15 min. The differentially labeled peptides were then mixed in equal proportions, desalted, and vacuum-dried29. The dried peptides were fractionated using a high pH reverse-phase column from Pierce™ high pH Reversed-Phase Peptide Fractionation kit (ThermoFisher), following the manufacturer’s protocol. The peptides were combined into 10 fractions, resulting in a total of 20 fractions. Finally, the tryptic peptides from each sample were dried by vacuum centrifugation. LC–MS/MS analysis The TMT labeled samples were subjected to fractionation using an Orbitrap Fusion Lumos mass spectrometer, interfaced with an Easy nLC 1200 system (Thermo Scientific). Each peptide fraction was loaded onto a trap column (100 µm × 20 mm, 5 µm, C18, Dr Maisch GmbH, Germany) using buffer A (composed of 0.1% formic acid and 2% acetonitrile in water). The separation was then carried out on a C18 analytical column (75 µm × 150 mm, 3 µm particle size, Dr Maisch GmbH) employing a linear gradient of buffer B (95% acetonitrile and 0.1% formic acid in water). The gradient profile was as follows: an increase from 2 to 8% over 2 min, 8% to 28% over 71 min, 28% to 40% over the next 79 min, then a rise to 100% over 81 min and holding at 100% for an additional 90 min. The entire process was conducted at a constant flow rate of 300 nL/min. Mass spectrometry (MS) analysis was performed using a Q Exactive HF-X mass spectrometer (Thermo Scientific) in the positive-ion mode. The data-dependent acquisition was set for a 90-min duration. A full MS scan was acquired over a range of 400 to 1800 m/z, at a resolution of 60,000 @m/z 200. The automatic gain control (AGC) target was set at 3e6 ions with a maximum injection time (max IT) of 50 ms. Afterwards, MS2 spectra was obtained in the collision-induced dissociation mode. The setting for MS2 included a normalized collision energy of 32%, a resolution of 45,000 @m/z 200, an AGC target of 1e5, a max IT of 50 ms, and an isolation window for the selected precursor of 1.6 m/z25,30,31. Database search and bioinformatics analysis LC–MS/MS raw data files were processed using the Proteome Discoverer software (version 2.4, Thermo Scientific). This process involved data interpretation and protein identification against a specific database32,33. The search parameters were specified as follows: the database used was “uniport-Capra hircus (goat)-9925 (77,732 sequences).fasta”, a target-reverse database derived from the Uniprot protein database (https://www.uniprot.org/taxonomy/9925), this database was employed to calculate the false discovery rate (FDR). TMTpro labels were used as isobaric tags. The mass tolerance for reporter ion MS2 was set to 0.005 Da. Trypsin/P was selected as the cleavage enzyme, following for a maximum of two missed enzymatic cleavage sites. The main search peptide tolerance for precursor ions was established as 10 ppm, while the mass error tolerance for fragment ions was 0.02 Da. Fixed modifications included Carbamidomethyl (C), TMTpro (K), TMTpro (peptide N-term), whereas variable modifications comprised oxidation (M) and acetylation (protein N-term). The FDR was adjusted to be less than 0.01. Both razor and unique peptides were utilized for protein quantification. Bioinformatics analyses were conducted using Perseus software, Microsoft excel, and R statistical computing software. Differentially abundant proteins (DAPs) were screened based on a fold change (FC) cutoff of > 1.20 or < 0.833 and P-value of < 0.05. Protein abundance data underwent hierarchical clustering. Sequence annotation leveraged information was extracted from UniProtKB/Swiss-Prot, gene ontology (GO), and the kyoto Encyclopedia of genes and genomes (KEGG)34. GO and KEGG enrichment analyses were executed using Fisher’s exact test, with FDR correction for multiple testing. GO terms were annotated using the UniProt-GOA database (http://www.ebi.ac.uk/GOA/) combined with Inter-ProScan tool (https://www.ebi.ac.uk/interpro/), covering three categories: biological process (BP), cellular component (CC), and molecular function (MF). The KASS service (https://www.genome.jp/tools/kaas/) and DAVID tool (https://david.ncifcrf.gov/) were used for KEGG pathway annotation35. Enriched GO and KEGG pathways were considered statistically significant at a P-value < 0.05. Additionally, the wolf-psort service (https://www.genscript.com/wolf-psort.html) was applied to predict subcellular localization. The NCBI’s Blast version 2.2.26 (http://blast.ncbi.nlm.nih.gov/Blast.cgi) facilitated the construction of protein–protein interaction networks. Interactions among DAPs were considered based on a confidence score of > 0.7. The R package “networkD3” (https://cran.r-project.org/web/packages/networkD3/) was utilized for the visualization of these networks. Parallel Reaction Monitoring (PRM) Quantification For the quantification of sperm samples using Parallel Reaction Monitoring (PRM), the preparation of samples (n = 6), total protein extraction, and peptide acquisition followed the protocols established in the TMT-based proteomic procedure. We used 2 µg of tryptic peptides from each sample for LC-PRM/MS analysis. Chromatographic separation of peptides was performed using an Easy-nLC 1200 UPLC system (Thermo Scientific) with a flow rate set at 300 nL/min. In brief, Peptide from each fraction was loaded onto a trap column (100 µm × 20 mm, 5 µm, C18, Dr Maisch GmbH) in buffer A (consisting of 0.1% formic acid and 2% acetonitrile in water). This was followed by separation on a C18 column (75 µm × 150 mm, 3 µm, Dr Maisch GmbH) employing a linear gradient of buffer B (95% acetonitrile and 0.1% formic acid in water). The gradient progressed from 2 to 5% for 5 min, then from 5 to 23% for 45 min, and subsequently from 23 to 40% for 50 min, before finally escalating climbing to 100% for 52 min and holding at 100% for 60 min. The separated peptides were then analyzed using a Q Exactive HF-X mass spectrometer (Thermo Scientific) in positive ion mode. The full scan range for MS was set to 300–1200 m/z at a resolution of 60,000 @m/z 200. The AGC target was 3e6 ions, and the max-IT was 50 ms. The parameters for MS2 were as follows: a resolution of 30,000 @m/z 200, an AGC target of 1e6 ions, the max IT of 100 ms, an activation type of HCD, an isolation window of 1.6 m/z, and a normalized collision energy set at 28. The PRM-MS/MS data were analyzed using the MaxQuant engine and processed through Skyline software (v.4.1)36. Untargeted metabolomic investigation This study conducted an untargeted metabolomic analysis using UHPLC-Q-Exactive Orbitrap MS/MS. Ten male subjects (n = 10) provided sperm samples for this analysis. For metabolite extraction, 50 mg of each sperm pellet was initially dissolved in 200 µL of precooled water, followed by the addition of 800 μL of a precooled methanol/acetonitrile solution (1/1, v/v). This mixture was then subjected to vortexing and ultrasonic disruption in an ice bath for 1 h using a cell ultrasonic crusher (Biosafer, USA). A 50 µL aliquot of the homogenate solution was utilized to determine protein concentration using a BCA kit, following the manufacturer’s instructions. The remaining mixture was further processed by adding 600 µL of acetonitrile/methanol (1:1) solvent, vortexed for 30 s, and incubated at -20 °C for 1 h to facilitate protein precipitation. This was followed by centrifugation at 16,000 × g for 20 min at 4°C37. The resulting 600 μL supernatant was then transferred to a new tube, dried in a vacuum concentrator, and reconstituted in 1 mL of extraction solvent (methanol/acetonitrile/water, 2/2/1). After additional vortexing and ultrasonic disruption for 10 min, the sample was centrifuged at 16,000 × g for 15 min at 4 °C. The clear supernatant was subsequently filtered through a 0.22 μm microfilters and transferred to a sample vial for LC–MS analysis7. Quality control (QC) samples were prepared by mixing equal aliquots of the supernatant from all sperm samples to normalize the data for metabolic profiling. Chromatographic separation of the above samples was performed on an UHPLC-ESI-Q-Orbitrap-MS system (Shimadzu Nexera X2 LC-30AD, Japan), utilizing an Acquity UPLC® HSS T3 column (2.1 mm × 100 mm, 1.8 μm). The mobile phases consisted of phase A (25 mM ammonium acetate and 25 mM ammonia hydroxide in water, pH9.75) and phase B (acetonitrile). The gradient profile of phase B started at 0% for 2 min, increased to 48% for 6 min, then climbing to 100% for 10 min and holding at 100% for 12 min, reduced back to 0% for 12 min, and held at 0% for additional 15 min. All performed at a constant flow rate of 0.3 mL/min, auto-sampler temperature at 4 ℃, and an injection volume of 6 μL. MS and MS2 analyses were conducted using a Q-Exactive Plus mass spectrometer (Thermo Scientific), with the following settings: sheath gas flow rate at 30 Arb, auxiliary gas flow rate at 5 Arb, capillary temperature at 320 ℃, full MS resolution at 70,000 @m/z 200, MS2 resolution at 17,500 @m/z 200, the max IT was 100 ms for MS and 50 ms for MS2. The isolation window for MS2 was 2 m/z and the normalized collision energy at 20/30/40 in NCE mode. The spray voltage was set at 3.8 kv for positive mode or -3.2 kv for negative mode, respectively38. The raw data were processed using MSDIAL software for peak alignment, retention time correction and peak area extraction. The metabolites were identified by accuracy MS matching (mass tolerance < 20 ppm) and MS2 matching (mass tolerance < 0.02 Da), which were matched with HMDB and MassBank databases, combined with self-built metabolite standard library. For extracted data, the ion peaks with missing value ≤ 50% in the group were retained for subsequent statistical analysis. The total peak areas of positive and negative ion data were normalized, respectively, the peaks of positive and negative ions were integrated and their pattern recognition was carried out with R software (v. 4.0.3). After preprocessing by unit variance scaling, the data were carried out by multivariate statistical analysis and KEGG enrichment analysis. For multivariate statistical analysis, R and R packages was used for all multivariate data analysis and modeling. Data were mean-centered using Pareto scaling. Models were built on principal component analysis (PCA), orthogonal partial least squares discrimination analysis (OPLS-DA). For KEGG enrichment analysis, the Fisher’s exact testing and FDR correction for multiple testing were performed, respectively. Enriched KEGG pathways were statistically significant at the P value < 0.05 level. Statistical analysis The data obtained in this study were presented and subjected to statistical analysis using Prism 8.0 software (GraphPad), with results displayed as means ± standard error of the mean (SEM). The evaluation of differences between groups was conducted through an analysis of variance (ANOVA) in IBM SPSS26.0 software (SPSS Inc., USA). Prior to applying ANOVA, the homogeneity of variance was verified. Subsequent to ANOVA, Duncan’s multiple range test was employed for post-hoc comparisons, contingent upon the assumption of equal variances among groups. Functional enrichment analyses were also conducted, and a corrected P-value of less than 0.05 was deemed statistically significant39. This statistical approach ensured a rigorous assessment of the data, allowing for meaningful and reliable interpretations of the study’s findings. Results Sperm motility and motile indices The motility and motile indices of both fresh and post-thaw goat spermatozoa, collected via AV and EE, were summarized in Table 1. The cryopreservation process significantly reduced the motility of goat spermatozoa (P < 0.01). Notably, post-thaw sperm collected by AV exhibited higher motility (43.96 ± 1.67%) compared to that collected by EE (38.69 ± 0.69%) (P < 0.05). Additionally, post-thaw sperm in both groups showed extremely significantly lower velocities [curvilinear (VCL, μm/s), rectilinear (VSL, μm/s) and mean (VAP, μm/s)] as compared to fresh sperm (P < 0.01). Thereinto, fresh sperm collected by EE showed significantly lower VAP (71.06 ± 3.30 μm/s) than that (85.49 ± 4.80 μm/s) collected by AV (P < 0.01). Also, EE significantly increased oscillation (WOB, %) of post-thaw sperm as compared to AV (P < 0.01). Detailed raw data on sperm motility and motile indices were described in Table S2.Table 1Effects of the two collection methods on the motility and motile indices of fresh sperm (XEZ & XAZ) and post-thaw sperm (DEZ & DAZ).GroupMotility (%)VCL (μm/s)VAP (μm/s)VSL (μm/s)STR (%)WOB (%)LIN (%)DEZ38.69 ± 0.69Ba47.86 ± 2.09Aa29.58 ± 2.07Aa23.05 ± 2.06Aa58.19 ± 2.83ab61.68 ± 1.42B37.54 ± 2.16abDAZ43.96 ± 1.67Bb53.71 ± 2.27Aa34.15 ± 1.77Aa27.44 ± 1.72Aa60.81 ± 1.16a54.19 ± 1.46Aa38.29 ± 1.61bXEZ88.34 ± 2.15Ac108.35 ± 2.72Bb71.06 ± 3.03B53.30 ± 3.06Bb59.25 ± 2.42ab50.25 ± 2.40Aa37.16 ± 3.19abXAZ90.85 ± 2.30Ac117.82 ± 5.38Bb85.49 ± 4.80C60.31 ± 3.29Bb53.24 ± 1.56b52.94 ± 1.83Aa30.95 ± 7.08aWithin the same column, values with different lowercase superscripts differ significantly (P < 0.05) and different uppercase superscripts differ significantly extremely (P < 0.01). Table 1Effects of the two collection methods on the motility and motile indices of fresh sperm (XEZ & XAZ) and post-thaw sperm (DEZ & DAZ).GroupMotility (%)VCL (μm/s)VAP (μm/s)VSL (μm/s)STR (%)WOB (%)LIN (%)DEZ38.69 ± 0.69Ba47.86 ± 2.09Aa29.58 ± 2.07Aa23.05 ± 2.06Aa58.19 ± 2.83ab61.68 ± 1.42B37.54 ± 2.16abDAZ43.96 ± 1.67Bb53.71 ± 2.27Aa34.15 ± 1.77Aa27.44 ± 1.72Aa60.81 ± 1.16a54.19 ± 1.46Aa38.29 ± 1.61bXEZ88.34 ± 2.15Ac108.35 ± 2.72Bb71.06 ± 3.03B53.30 ± 3.06Bb59.25 ± 2.42ab50.25 ± 2.40Aa37.16 ± 3.19abXAZ90.85 ± 2.30Ac117.82 ± 5.38Bb85.49 ± 4.80C60.31 ± 3.29Bb53.24 ± 1.56b52.94 ± 1.83Aa30.95 ± 7.08aWithin the same column, values with different lowercase superscripts differ significantly (P < 0.05) and different uppercase superscripts differ significantly extremely (P < 0.01). Effects of the two collection methods on the motility and motile indices of fresh sperm (XEZ & XAZ) and post-thaw sperm (DEZ & DAZ). Within the same column, values with different lowercase superscripts differ significantly (P < 0.05) and different uppercase superscripts differ significantly extremely (P < 0.01). Sperm acrosome and plasma membrane integrities This study examined the impacts of the two semen collection methods, AV and EE, on the acrosome integrity (ACRI, %) and plasma membrane integrity (PMI, %) of goat spermatozoa, as detailed in Table S2 and illustrated in Fig. 1A and 1B. The findings revealed that the freezing process significantly compromised both PMI and ACRI of goat spermatozoa (P < 0.01). Furthermore, analysis indicated that cryopreserved spermatozoa collected via AV method exhibited higher ACRI (P < 0.01) and PMI (P < 0.05) compared to those collected using the EE method.Fig. 1Levels of ACRI, PMI, ROS, MITA in all sperm samples from DEZ, DAZ, XEZ and XAZ groups. Lowercase letters differ indicate P < 0.05, uppercase letters differ indicate P < 0.01. Levels of ACRI, PMI, ROS, MITA in all sperm samples from DEZ, DAZ, XEZ and XAZ groups. Lowercase letters differ indicate P < 0.05, uppercase letters differ indicate P < 0.01. Mitochondrial activity and ROS levels The impact of two semen collection methods on reactive oxygen species (ROS) levels and mitochondrial activity in goat spermatozoa were examined. It was observed that the freezing process significantly increased intracellular ROS content (%) of goat spermatozoa (P < 0.01). AV and EE collection methods haven’t significant impact on the ROS content of fresh and post-thaw spermatozoa, respectively (Fig. 1C, P > 0.05). In addition, cryopreserved spermatozoa collected via the AV method exhibited higher mitochondrial activity (MITA, %) compared to those collected using the EE method (Fig. 1D, P < 0.05). Raw data about ROS content and MITA of spermatozoa were described in Table S2. Identification of proteomic information by TMT-based quantitative technology To further explore underlying molecular influences of two semen collection methods on goat spermatozoa, TMT-based quantitative proteomics experiments were performed. In total, 23,379 unique peptides were identified from 852,615 MS/MS spectrums, with 2,461 corresponding proteins identified and quantified (Table S3). Quality control results of the proteins were shown in Fig. S3. DAPs were identified based on fold changes among comparisons of the four groups, using the screening criteria intergroup ratio (FC > 1.2 or < 0.833, P < 0.05). After screening, the top 10 DAPs (upregulated & downregulated) across the four comparative groups (PDAZ vs. PDEZ, PDAZ vs. PXAZ, PDEZ vs. PXEZ, PXAZ vs. PXEZ) were given in Table S4. PRM Quantification To validate the accuracy of the TMT-based proteomic data, PRM technique was utilized. We selected twenty-three DAPs, each represented by at least two unique peptides, for cross-validation across the four comparative groups (PDAZ vs. PDEZ, PDAZ vs. PXAZ, PDEZ vs. PXEZ, PXAZ vs. PXEZ). The validation outcomes, as detailed in Table S5 and Fig. S4, demonstrated that the abundance patterns of these selected proteins were consistent with the trends observed in the TMT-based data. This consistency serves to reinforce the reliability of our proteomic findings. Classification and hierarchical clustering analysis Classification and hierarchical clustering of quantified proteins were conducted using the Mfuzz method, as illustrated in Fig. 2. The analysis identified various sperm protein clusters, each exhibiting distinct abundance patterns across different sample groups. Cluster 1 exhibited high protein abundance in the PXAZ group, and the lowest level in the PXEZ group, a gradual increase in abundance levels were observed in the PDAZ and PDEZ groups. Cluster 2 showed a very low abundance levels in the PXAZ and PDAZ, but significantly higher levels in PXEZ and PDEZ groups. Cluster 4 demonstrated low abundance level in the PXAZ, peaking in the PXEZ, and those followed by the lowest abundance levels in the PDAZ and PDEZ. Cluster 3 exhibited a pattern of gradually decreasing abundance from PXAZ to PDEZ. Whereas, Cluster 5 presented a pattern of gradually increasing abundance from PXAZ to PDEZ.Fig. 2Classification and hierarchical clustering of the quantified proteins. Proteins were classified into five clusters according to their abundance pattern in four groups by the Mfuzz method, with the minimum SD set as 0.2, the number of clustered set as 5 and P < 0.05. Individual protein was represented by a single row, and each group was represented by a single column in the heatmap (A). The top five enriched GO-BP and KEGG functional terms in each cluster were also exhibited (B). Classification and hierarchical clustering of the quantified proteins. Proteins were classified into five clusters according to their abundance pattern in four groups by the Mfuzz method, with the minimum SD set as 0.2, the number of clustered set as 5 and P < 0.05. Individual protein was represented by a single row, and each group was represented by a single column in the heatmap (A). The top five enriched GO-BP and KEGG functional terms in each cluster were also exhibited (B). According to the above abundance patterns, we focused the proteins’ information influenced by the two semen collection methods (in cluster 1, 2 and 4) and those affected by the freezing process (in cluster 3 and 5), as detailed in Table S6. Within these clusters, 12 proteins in PDAZ vs. PDEZ, 214 proteins in PDAZ vs. PXAZ, 380 proteins in PDEZ vs. PXEZ, and 128 proteins in PXAZ vs. PXEZ were identified as differentially abundant, respectively (Fig. 3A).Fig. 3Functional classification and pathway analysis of DAPs from four comparisons. Venn diagram displaying the numbers of DAPs (A). The DAPs were involved in various biological processes (B). The DAPs were involved in various biological pathways (C). PPI networks construction of the DAPs, different colors represented the Betweenness score values of significative proteins in the process of analysis using the cystoscope (v.3.9.1) (D). Functional classification and pathway analysis of DAPs from four comparisons. Venn diagram displaying the numbers of DAPs (A). The DAPs were involved in various biological processes (B). The DAPs were involved in various biological pathways (C). PPI networks construction of the DAPs, different colors represented the Betweenness score values of significative proteins in the process of analysis using the cystoscope (v.3.9.1) (D). Functional classification and pathway analysis of DAPs Bioinformatics techniques were employed to analyze the DAPs. The GO and KEGG functional annotations of these DAPs were determined using the DAVID database (https://david.ncifcrf.gov/home.jsp)40. Additionally, protein domain analysis was conducted using the SMART web tool (https://smart.embl.de)41. All results detailed in Table S7. GO analysis categorized the DAPs into three groups: biological process (BP), cellular component (CC) and molecular function (MF). The BP category showed significant involvement of DAPs in processes like ATP metabolic and biosynthetic process (16.10%, 62 proteins), mitochondrial respiratory chain complex assembly and mitochondrial transport (11.16%, 43 proteins), oxidative phosphorylation (9.09%, 35 proteins), ATP synthesis coupled electron transport (6.75%, 26 proteins), plasma membrane bounded cell projection assembly (4.94%, 19 proteins) (Fig. 3B). In the CC category, the significant DAPs were predominantly found in membrane bounded organelle (16.01%, 53 proteins), membrane (11.18%, 37 proteins), 9 + 2 motile cilium and motile cilium (9.97%, 33 proteins), sperm flagellum (3.93%,13 proteins), membrane protein complex (3.93%, 13 proteins), extracellular space (3.62%, 12 proteins), and respiratory chain complex I (3.32%,11 proteins) (Fig. S5A). With regard to MF, the significant DAPs were mostly associated with catalytic activity and acting on a protein (46.28%,143 proteins), oxidoreductase activity (11.97%, 37 proteins), oxidoreductase activity, acting on NAD(P)H (5.83%,18 proteins), and the NADH dehydrogenase activity (4.85%, 15 proteins) (Fig. S5B). KEGG enrichment analysis revealed that the significant DAPs were enriched in several pathways linked to sperm function, such as apoptosis (map04210, 4 proteins), glycolysis/gluconeogenesis (map00010, 10 proteins), oxidative phosphorylation (map00190, 36 proteins), fatty acid degradation (map00071, 2 proteins), pyruvate metabolism (map00620, 5 proteins), and the citrate cycle (map00020, 6 proteins) (Fig. 3C). Furthermore, a protein–protein interaction (PPI) network of all DAPs was constructed using the STRING database (v.11.5, https://cn.string-db.org/) and visualized using the Cystoscope (v.3.9.1). A P-value of less than 0.05 indicated significant correlations within the interaction networks, and key hub proteins were identified using the Betweenness (Fig. 3D). To sum up, this analysis identified 210 significant differential proteins associated with the characteristics and functions of goat sperm across four groups. These findings were derived from patterns of protein abundance combined with the functional enrichment analysis (Table S8). Metabolic findings of sperm samples using UHPLC-QE-MS In our study, a comprehensive metabolomic analysis was performed on sperm samples from the four experimental groups using an untargeted metabolomics approach with UHPLC-QE-MS. After quality control (QC), a total of 10,858 peaks were retained across both positive (POS) and negative (NEG) ion modes. We identified 495 metabolites in POS and 271 metabolites in NEG after mapping with MS2 Metabo name. These details were presented in Table S9. For QC assessment, principal component analysis (PCA) and orthogonal partial least squares discrimination analysis (OPLS-DA) were employed. The R2Y and Q2 values from permutation testing, as shown in Fig. S6, S7, were both close (or equal) to 1, indicating reliable model performance with no over-fitting in all OPLS-DA models. Identification and functional annotation of differential metabolites Differentially abundant metabolites (DAMs) between groups were identified using variable importance in projection (VIP) score greater than 1 in the OPLS-DA models and a student’s t-test (P-value < 0.05). We identified 535 DAMs across the four comparisons (Fig. 4A, Table S10), and the top 10 DAMs (upregulated & downregulated) across the four comparative groups (MDAZ vs. MDEZ, MDAZ vs. MXAZ, MDEZ vs. MXEZ, MXAZ vs. MXEZ) were given in Table S11. Utilizing the HMDB database (https://hmdb.ca/)42, these DAMs were predominantly categorized into lipids, organic acids and derivatives, benzenoids, organoheterocyclic compounds, phenylpropanoids and polyketides, nucleosides and analogues, alkaloid and derivatives (Fig. S8). KEGG pathway analysis revealed significant enrichment in various metabolic pathways, with specific pathways differing among the comparisons (Fig. 4B, 4C, and S9, Table S12). In total, we screened 32 DAMs and 17 significant biological pathways related to sperm function (Table S13).Fig. 4Functional annotation analysis of DAMs. Venn diagram displaying the numbers of DAMs (A). The DAMs were involved in various biological pathways in the MDAZ vs.MDEZ comparision (B), and the DAMs were involved in various biological pathways in the MXAZ vs.MXEZ comparision (C). Functional annotation analysis of DAMs. Venn diagram displaying the numbers of DAMs (A). The DAMs were involved in various biological pathways in the MDAZ vs.MDEZ comparision (B), and the DAMs were involved in various biological pathways in the MXAZ vs.MXEZ comparision (C). Correlation analysis between the significant DAPs and DAMs We conducted a correlation analysis to understand the relationship between the proteomic and metabolomic profiles in sperm. This was performed using R package software (v4.0.1) combined with Cytoscape (v3.8.2). A total of 64 proteins and 34 metabolites were to be co-enriched in 51 biological pathways across the four comparisons (Table S14). KEGG pathway enrichment results were presented in Fig. 5, S10. Notably, 10 significant molecular pathways involving the co-enrichment of 33 DAPs and 10 DAMs were identified (Table S15), including the citrate cycle (TCA cycle) (PDHB, DLST, DLD, IDH3B, citrate, thiamin pyrophosphate), chemical carcinogenesis-ROS (GSTO2, GSTM3, ATP5F1B, NDUFB11, LOC102183616, NDUFA10, UQCRQ, NDUFB10, ATP5F1C, COX6B2, NDUFB9, NDUFB3, NDUFA8, ATP5F1E, NDUFC2, NDUFB5, NDUFS4, LOC102187154, MT-CO1, GSTO2, NAD + , oxidized glutathione), oxidative phosphorylation (PPA2, LOC102168533, LOC108638252, ATP5F1B, NDUFB11, LOC102183616, NDUFA10, UQCRQ, NDUFB10, ATP5F1C, COX6B2, NDUFB9, NDUFB3, NDUFA8, ATP5F1E, NDUFC2, NDUFB5, NDUFS4, LOC106501717, LOC102187154, NDUFB6, MT-CO1, NAD + , phosphoric acid), glycine, serine and threonine metabolism (DLD, PGAM2, L-aspartate, choline, betaine, glycerate). These analyses provide valuable insights into the molecular mechanisms influencing the quality and freezing tolerance of goat sperm.Fig. 5Bubble of the significant KEGG pathways in metabolome or proteome between DAZ vs. DEZ and XAZ vs. XEZ comparisons, respectively. Bubble of the significant KEGG pathways in metabolome or proteome between DAZ vs. DEZ and XAZ vs. XEZ comparisons, respectively. Discussion Impact of semen collection methods on sperm functional parameters The process of freezing and thawing has been shown to compromise the quality of goat semen, notably affecting the motility, plasma membrane and acrosome structures, as well as the levels of ROS. These alterations significantly influence the sperm’s capacity to fertilize and subsequent embryo development18,43. In the present study, we evaluated the functional parameters of fresh and post-thaw goat sperm collected using rectal probe electrostimulation (EE) and artificial vagina (AV) methods, respectively. Our findings revealed that the EE and AV methods differed significantly only in the velocity average pathway (VAP) of fresh sperm. For post-thaw sperm, although ROS levels did not show a significant difference between the two methods, sperm motility, integrities of plasma membrane and acrosome were significantly better preserved by the AV method compared to EE. Sperm motility is a critical functional characteristic reflecting sperm quality, essential for the activation of the sperm flagellum and successful fertilization44. The acrosome and plasma membrane, crucial morphological indices of sperm, play a significant role in maintaining the integrity of membrane surface proteins. This integrity is vital for fertilization success, especially following stress-inducing procedures like freezing and storage45–47. Furthermore, an excess of ROS can cause oxidative damage to phospholipid-bound polyunsaturated fatty acids (PUFAs) and cholesterol in sperm, triggering oxidative stress and initiating apoptosis48–50. Based on these observations, while the EE method does not severely impair post-thaw sperm quality, the AV method appears to be more advantageous in preserving sperm function. Semen collection method influences the protein profile of goat sperm Sperm, as a highly specialized cell, is generally inactive in transcriptional and translational activities once it leaves the male reproductive tract. As a result, the composition, communication and regulation of protein and metabolite molecules play important roles in maintaining the pre-fertilization function of sperm, especially proteins population, involved in post-translational modification (PTM), to fulfill its’ specific function before fertilizing an oocyte51,52. Our previous studies have elucidated the effects of freezing and thawing process on proteins of ram sperm collected via AV method15,23. Here, we combined proteomic and metabolomic analyses for classification and identification of proteins and metabolites in sperm biological system, to provide a comprehensive explanation of the differential impacts of AV and EE collection methods on the quality and cryotolerance of goat semen. Firstly, we concentrated on the proteomic alterations in fresh and post-thaw goat sperm collected via AV and EE methods, respectively. Utilizing TMT-based sequencing and bioinformatics analysis, 33 differentially abundant proteins implicated in sperm function across the PXEZ, PXAZ, PDEZ, PDAZ groups were identified. These proteins are linked to various pathways, such as FoxO and AMPK signaling, ROS, citrate cycle (TCA cycle), OXPHOS, glutathione metabolism, glycine, serine and threonine metabolism. These pathways could elucidate the reduced motility observed in post-cryopreservation sperm and the differing impacts of collection methods on post-thaw sperm motility. For instance, certain proteins like NADH dehydrogenase [ubiquinone] 1 (NDUFB5, NDUFB6, NDUFB10, NDUFC2), ATP synthase subunit, cytochrome b-c1 complex subunit 8 (UQCRQ), and cytochrome c oxidase (COX6B2) were more abundant in post-thaw sperm. It has been reported that complex I (NADH dehydrogenase) can directly catalyse the electrons transfer from NADH to ubiquinone, belong to the first enzyme in respiratory chain, it’s accessory subunits such as NDUFB5, NDUFB6, NDUFB10, NDUFC2 may be involved in the regulation of enzymatic activity, stability of the complex or auxiliary functions during establishing a platform for ATP synthesis by complex V7,53–56. ATP synthases are macromolecular machines consisting of an ATP-hydrolysis-driven soluble F1 motor and a proton-translocation-driven membrane-embedded F0 motor, which positioned to oppose each other’s action on a shared rotor subcomplex57. Subunits of ATP synthase usually located on F1 or F0 ensure the structural coupling and promote both ATP synthesis or hydrolysis58,59. Proteins like UQCRQ, a component of ubiquinol-cytochrome c reductase binding that ensures the complex III function60,61. COX6B2 acts as a specific component of complex IV in sperm, is also noted for its involvement in mitochondrial OXPHOS62. Additionally, proteins specific to the PDEZ vs. PXEZ comparison, including NDUF family members (NDUFB3, NDUFB9, NDUFB11, NDUFA8), GSTO2, GSTM3, and cytochrome b-c1 complex subunit 7 (QCR7) were identified. These proteins are associated with OXPHOS and catalytic activities, vital for maintaining post-thaw sperm function. Concretely, as a functionally active enzyme, GSTO2 exerts roles in both glutathione reductase and dehydroascorbate reductase reactions63. In mammalian sperm, GSTO2 is mainly localized on the plasma membrane and may be involved in an oxidative or reductive reaction during sperm capacitation64. For GSTM3, has been reported as important antioxidant enzyme to protect boar sperm during freezing and thawing process65,66. QCR7 is one of subunits of the ubiquinol-cytochrome c oxidoreductase complex, it maybe interact with other proteins or enzymes and contribute to electron transfer process in mitochondria67,68. Conversely, downregulated proteins during sperm cryopreservation, such as cytochrome c oxidase subunit 1(MT-CO1), isocitrate dehydrogenase [NAD] subunit (IDH3B), glucose-6-phosphate isomerase (GPI), dihydrolipoyl dehydrogenase (DLD), were significantly enriched in pathways crucial for mitochondrial function, ROS production, and fertility69–72. Notably, the EE method specifically affected the abundance of nine proteins, mainly enriched in OXPHOS (ATP5F1B, ATP5F1C, NDUFS4, NDUFA10, PPA2), TCA cycle (PDHB, DLST), AMPK signaling pathway (RAB2A), and glutathione metabolism (LAP3), which are integral to acrosome formation and protection, ATP production, and sperm motility and fertility73–78. Furthermore, this study examined the effects of semen collection methods on sperm functional proteins. In fresh sperm, compared to EE, the AV method showed downregulation in proteins mainly involved in OXPHOS (NDUFS4, ATP5F1B, ATP5F1E) and AMPK signaling pathway (RAB2A). In contrast, upregulated proteins were primarily associated with OXPHOS (NDUFC2, NDUFA8, NDUFB3, NDUFB6, NDUFB10, QCR7, COX6B2) and glycolysis/gluconeogenesis (G6PC3). This indicates the critical role of mitochondrial OXPHOS in energy provision for sperm73,79–85. In cryopreserved sperm, two downregulated proteins (Ig-like domain-containing protein, uncharacterized protein with the accession is A0A452E2P1) identified warrant further investigation due to their involvement in key signaling pathways related to sperm survival and motility86,87. This analysis highlights the significant impact of semen collection methods on the protein profile of goat sperm, underscoring the importance of these methods in influencing sperm quality and functionality post-cryopreservation. Semen collection method changes the metabolite profile of goat sperm We explored the extensive changes in metabolites in both fresh and cryopreserved-thawed sperm, originating from two distinct semen collection methods. Following UHPLC-QE-MS sequencing and bioinformatic analysis, a total of 32 DAMs and their correlation with 17 significant biological pathways relevant to sperm function across the four comparative groups were determined. Through the integration of functional enrichment analysis between the proteome and metabolome, we highlighted 10 DAMs enriched in key pathways, including OXPHOS (NAD + and phosphoric acid), ROS (NAD + , oxidized glutathione), AMPK signaling pathway (quercetin), TCA cycle (citrate, thiamin pyrophosphate), glutathione metabolism (L-glutamic acid, oxidized glutathione), glycine, serine and threonine metabolism (choline, L-aspartate, betaine), as well as glycolysis/gluconeogenesis and pyruvate metabolism (thiamin pyrophosphate). Nicotinamide adenine dinucleotide (NAD +), both a substrate and crucial product in the OXPHOS processs88, is essential for glucose glycolysis, TCA cycle, energy balance, and DNA repair88–90. Environmental stress can impact the redox status of NAD + , leading to pathological changes in spermatogenesis tissues91, Interestingly, NAD + level is inversely correlated with human sperm quality92, and we observed a similar trend in goat sperm, where NAD + was more abundant in frozen than in fresh samples. Glutathione (GSH), synthesized in two ATP-dependent steps and recognized as an endogenous antioxidant due to its cysteine thiol moiety, is typically oxidized to glutathione disulfide (GSSG) by pro-oxidants like ROS and RNS93. The GSH/GSSG ratio is a key marker in cellular redox status94, and an increased ratio is associated with reduced cellular free radical generation95. In human semen, this ratio correlates positively with semen quality96. Our finding indicated that GSSG is significantly more abundant in frozen sperm, suggesting its potential as a stable negative biomarker for mammalian semen quality assessment. Quercetin, known for its antioxidant, anti-inflammatory, and antimicrobial properties97,98, has been increasingly applied to sperm cryopreservation. Its dose-dependent protective effects have been demonstrated in various species, including dog, Egyptian buffalo, and stallion sperm99–101. In this study, we found endogenous quercetin in goat sperm, significantly higher in post-thaw sperm collected by the EE, and enriched in the AMPK signaling pathway, potentially decreasing oxidative stress-induced apoptosis102–105. Moreover, differential metabolites such as thiamin pyrophosphate (TTP) and choline, which are mainly involved in the TCA cycle, glycolysis/gluconeogenesis, pyruvate metabolism, and glycine, serine and threonine metabolism, were downregulated in cryopreserved sperm compared to fresh samples collected by the AV. TTP, an active form of vitamin B1 and a cofactor in energy metabolism antioxidative pathways, may serve as a positive biomarkers for sperm quality, given the decreased ATP generation and antioxidant capacity of frozen sperm106–108. Choline, functioning similarly to vitamin B and acting as a precursor for various metabolites, is crucial for normal cellular functioning109,110. Our study suggests that choline’s abundance positively correlates with sperm quality. Additionally, betaine, a key downregulated metabolite in DEZ vs. XEZ, involved in glycine, serine and threonine metabolism, has been shown to improve sperm motility, reduce antioxidant damage, and maintain normal methylation levels in sperm freezing and artificial insemination contexts111–113. Lastly, L-aspartate, a protein building block and precursor for multiple metabolites such as organic acids and amino acids114,115, has been shown to deaminate to oxaloacetate by aspartate aminotransferase in cytoplasmic metabolic pathway, and oxaloacetate exerts a key role in the formation of malic acid needed for the TCA cycle in mitochondria116. Recently, Xu et al. reported that L-aspartate in seminal plasma enhances sperm freezability117, combined with our finding, where L-aspartate was more abundant in fresh sperm collected via AV method compared to the EE, suggesting its potential as a biomarker of sperm quality, deserving further investigation. Conclusions In summary, our study demonstrates that the AV collection method is superior in maintaining the functional indices of frozen spermatozoa, including the motility, integrities of acrosome and plasma membrane, as well as mitochondrial activity, when compared to the EE method. By integrating multi-omics techniques, we have for the first time generated proteomic and metabolomic profiles of fresh and frozen goat spermatozoa collected via both AV and EE methods. Among the four groups compared, a total of 210 DAPs related to sperm characteristics and functions were identified. Additionally, 32 DAMs were detected, which are involved in various metabolic functions. Our comprehensive bioinformatic analysis highlighted 10 significant molecular pathways co-enriched among 33 DAPs and 10 DAMs, such as the TCA cycle, ROS, OXPHOS, and glycine, serine, and threonine metabolism. These findings provide pivotal insights into the molecular mechanisms influencing the quality and freezing tolerance of goat spermatozoa collected via AV and EE methods. This understanding is instrumental in guiding the selection of the most appropriate semen collection method for the preservation of goat genetic resources. Supplementary Information Supplementary Information 1. Supplementary Information 2. Supplementary Information 3. Supplementary Information 4. Supplementary Information 5. Supplementary Information 6. Supplementary Information 7. Supplementary Information 8. Supplementary Information 9. Supplementary Information 10. Supplementary Information 11. Supplementary Information 12. Supplementary Information 1. Supplementary Information 2. Supplementary Information 3. Supplementary Information 4. Supplementary Information 5. Supplementary Information 6. Supplementary Information 7. Supplementary Information 8. Supplementary Information 9. Supplementary Information 10. Supplementary Information 11. Supplementary Information 12.