Sex sorting of bull semen is now commonly used in dairy farming to produce more female calves, speed up genetic improvement, and make herd replacement easier to plan. However, the quality of sex-sorted semen is often different among bulls and herds. The pregnancy rates are usually lower than when utilizing regular semen (Thongkham et al., 2021); (Álvarez Gallardo et al., 2024). This difference makes it harder for farmers to make good decisions. It also reveals the need for biological markers to predict how well the semen will work, not just describe its traits.

Sperm function depends mainly on energy supply, membrane structure, and redox balance. There is acareful balance between making ATP and producing reactive oxygen, revealing a trade-off between energy use and stress. ATP comes from two sources: mitochondrial energy in the midpiece and glycolysis in the tail (Amaral, 2022)(Ricci et al., 2023)(Schmidt et al., 2024). These factors affect how the tail moves, which controls sperm movement. These energy limits also affect how ready the sperm is for capacitation, how long it can keep moving, and how well it survives during processing and storage.

Metabolomics helps us understand these biological processes as a whole system. Earlier studies drew a clear link between metabolite patterns and fertility in bulls, in both fresh and frozen semen. They found certain metabolites and pathways that can separate

high fertility bulls from low fertility (Talluri et al., 2022)(Giaretta et al., 2025). These outcomes reveal that studying small molecules can give useful biochemical markers to support regular sperm quality tests.

The environment strongly affects semen quality and molecular signals. Heat stress is known to lower semen quality in bulls, and its effects can last for weeks after being exposed to high temperatures. Changes in fertility by season also reveal that heat stress may have delayed effects on sperm production (Capela et al., 2022)(Khan et al., 2023). These environmental problems are very crucial in humid tropical areas like Indonesia. Constant heat, uneven energy in forage diets, and weak cold storage systems can disturb the balance between glycolytic and oxidative energy use. They can also change membrane lipids and affect the success of sex sorting (Silva et al., 2023). Even though global interest in this topic is increasing, there is still little research on how metabolite patterns differ among X and Y sperm and how these differences relate to semen quality and movement. This lack of knowledge is obvious in tropical dairy farms in Indonesia (Chang et al., 2025).

Even though more people worldwide are interested in this topic, there is still little research on the differences in metabolite patterns among X and Y sperm and how these differences relate to semen quality and movement. This lack of information is especially clear in tropical dairy farms in Indonesia (Chang et al., 2025). (Firdaus et al., 2024) found a pregnancy rate of about 63% utilizing albumin-sedimented sexed semen, which was greater than unsexed or PDGC-sexed semen (around 47%). (Nurcholis et al., 2024) also found pregnancy rates of 65% and 86%, depending on the season and semen type. Overall, these outcomes reveal that sex-sorted semen can give similar fertility outcomes to regular semen, but its performance in the field is still not stable in Indonesia.

This study looks for metabolite markers that clearly show the difference between X and Y sperm in sex-sorted Holstein Friesian bulls. It also studies how these metabolites relate to semen quality and sperm movement. Ultimately, this research aims to find useful markers that can be used in Indonesia’s humid tropical dairy farms.

In X-bearing spermatozoa Figure 8A, the three profiled metabolites—D-(−)-fructose, sn-glycerol-3-phosphocholine (sn-GPC), and L-(+)-lactic acid—showed generally positive correlations with key quality traits, including motility, viability, and membrane and acrosome integrity; meanwhile, sperm abnormalities were negatively correlated (r ≈−0.42 to −0.15). L-(+)-lactic acid showed the strongest correlations by membrane integrity (r ≈ 0.81) and track straightness (STR; r ≈ 0.86). Strong associations were also observed among kinematic parameters and the sugar/phosphocholine axis: wobble (WOB) correlated very strongly with D-(−)-fructose (r ≈ 0.98) and sn-glycerol-3-phosphocholine (sn-GPC; r ≈ 0.99), by additional positive correlations across linearity and velocity metrics (e.g., VSL, LIN, STR; r ≈ 0.78–0.86). In Y-bearing spermatozoaFigure 8B, acetyl-L-carnitine displayed a selective association, correlating most strongly by progressive motility parameters—VSL (r =0.82), LIN (r = 0.84), and WOB (r = 0.86)—and revealing moderate correlations by DAP (r = 0.42) and VAP/STR (r ≈ 0.26/0.24). Correlations by total movement, live sperm, and membrane or acrosome health were weak (|r| ≤ ~0.18), and DSL had a small negative correlation (r ≈ −0.12). Overall, these outcomes suggest that in X sperm, carbohydrate and phosphocholine are linked to stable and straight movement, and lactic acid helps keep the membrane healthy. In Y sperm, acetyl-L-carnitine is more linked to fast, straight movement than overall sperm health.

This study aimed to find small-molecule markers that can tell the difference between X and Y sperm and connect them to semen quality and movement in humid tropical conditions. This helps address the problem of unstable field outcomes by sex-sorted semen. The outcomes in Table 1 reveal that X sperm had better overall movement and acrosome health, while Y sperm had stronger membranes, fewer shape problems, greater speeds (VCL, VAP, VSL), longer movement paths (DCL, DAP, DSL), and better straightness (STR). These differences suggest that X and Y sperm may use different energy strategies, not just small variations of the same system.

The metabolomic data support this biological interpretation. Both unsupervised and supervised multivariate analyses clearly separated X- and Y-bearing sperm, by tight inside-of-group dispersion , and hierarchical clustering grouped samples primarily by chromosomal type Figure 4, indicating a coherent biochemical organization rather than random processing noise. Consistent univariate differences across features Figure 3 and Table 3, along with a greater number of retained HMDB-annotated metabolites and greater mean peak intensities in X-sperm Table 2, suggest that the phenotypic differences observed in Table 1 are underpinned by a definable molecular basis that is reproducible across specimens.

Inside this framework, X-bearing sperm appear to employ a carbohydrate-focused strategy. Enrichment of pathways related to glycolysis/gluconeogenesis, pyruvate metabolism, and other carbohydrate modules Figure 6 and Table 4 aligns with elevated levels of D-(−)-fructose, L-(+)-lactic acid, and sn-glycerol-3-phosphocholine (sn-GPC) among the top discriminatory metabolites Figure 7. These metabolites also have positive links with stable movement and membrane health Figure 8. Simply put, ATP made through glycolysis in the tail helps keep the flagellum beating. Lactate release through transporters helps control pH inside the cell and keep ion balance. These conditions help keep the sperm moving straight and protect the membrane after sorting and storage, matching the X sperm outcomes in Table 1(Amaral, 2022)(Lasorsa et al., 2023)(Rotimi et al., 2024). The link between sn-glycerol-3-phosphocholine (sn-GPC) and stable movement suggests phosphocholine may help keep the membrane flexible and support signals needed for capacitation. This fits with earlier studies on lipids related to fertility in bull semen (Longobardi et al., 2020)(Talluri et al., 2022). Among the pathways in Table 4, glycerophospholipid metabolism showed the biggest impact. This highlights the crucial role of membrane structures in controlling sperm movement and capacitation signals. This matches the sn-glycerol-3-phosphocholine (sn-GPC) patterns and their link to movement (Longobardi et al., 2020)(Talluri et al., 2022)(Correnti et al., 2023). For carbohydrate metabolism, the pathways of fructose, mannose, glycolysis, and pyruvate have a stronger effect on sperm function than other sugar pathways. This reveals that ATP made through glycolysis is very crucial to keep the sperm tail moving and to control lactate and pH levels. This matches the greater levels of D-(−)-fructose and L-(+)-lactate, which are linked to how the sperm moves (Amaral, 2022)(Lasorsa et al., 2023)(Rotimi et al., 2024). Overall, these patterns reveal that the types of membrane lipids and the glycolysis process are very crucial. They are key in connecting metabolite changes to how the sperm moves.

In contrast, Y-bearing sperm displayed a metabolic pattern linked to carnitine and oxidation. This fits with their faster and straighter movement. Acetyl-L-carnitine was a crucial metabolite Figure 7 and was strongly related to movement speed Figure 8. This happens because acetyl-L-carnitine helps bring acyl groups into the mitochondria, making energy production in the sperm midpiece more efficient. Changes in ether-lipid and glycerophospholipid metabolism Figure 6; Table 4 reveal that the membrane lipids are being reorganized in a coordinated way. This may help the mitochondria work better and improve flagellar signaling. This process helps explain why Y-sperm have greater STR and speed values, as revealed inTable 1. These outcomes also indicate that Y-sperm depend more on oxidative metabolism, while X-sperm rely more on glycolysis. This pattern has also been seen in studies of bull sperm and seminal plasma. In astheno-oligozoospermic bulls, disrupted fatty acid and ketone body metabolism has been linked to reduced motility. Similarly, (Correnti et al., 2023) reported that variations in acylcarnitines and phospholipids in seminal plasma were positively associated with sperm motility and morphology, reinforcing that metabolomic signatures are closely connected to fertility-related functional traits.

These mechanistic insights guide the selection of a minimal, translational biomarker panel. While ten features achieved perfect AUCs in internal cross-validation Table 5, several did so through quasi-separation by extreme log₂ fold-change values, a pattern that can lead to overestimating discriminative performance. Focusing on metabolites by well-defined, non-overlapping distributions—D-(−)-fructose, L-(+)-lactic acid, and sn-glycerol-3-phosphocholine (enriched in X-sperm) and acetyl-L-carnitine (enriched in Y-sperm)—recapitulates the class separation observed in Figures 1–2 while maintaining physiological relevance Figure 7. Because each metabolite can be quantified via targeted LC–MS or enzymatic assays, this panel offers a practical tool for laboratory quality control, enabling verification of fraction identity and detection of atypical batches prior to field deployment, directly linking molecular signatures to the divergent motility phenotypes summarized in Table 1. These outcomes can be directly used to improve extender formulas and help make better decisions on farms. The data in Figure 8, Figure 6, and Table 4 support making X-sperm media that give the right amount of glycolysis sources (like fructose) and choline/phosphocholine. This can help keep the sperm membrane strong and make their movement more stable, which matches the better motility and acrosome integrity seen in X-sperm Table 1. On the other hand, the Y-sperm metabolite pattern in Figure 8 reveals that adding carnitine may help keep their movement straight and fast. Antioxidants can also help reduce oxidative stress, which could lower the STR and speed seen in Table 1. In humid tropical dairy systems like those in Indonesia, farmers still face problems such as heat stress, changing energy in feed, and weak cold storage, as mentioned in the Introduction. Utilizing the four-metabolite pattern Figure 7 to help sort sperm batches and choose the right extender can be a helpful way to connect lab sperm separation with better pregnancy outcomes on farms (Mateus et al., 2023);(Schmidt et al., 2024).

The strength of this study comes from several pieces of clear evidence: multivariate separation Figures 1–2, steady univariate differences Figure 3 and Table 3, clear sample grouping Figure 4, logical pathway patterns Figure 6 and Table 4, and strong links to phenotypes Figure 8. All of these support a solid understanding of the different movement and quality patterns revealed in Table 1. Nonetheless, several crucial limitations should be considered when interpreting these outcomes. The sample size was modest, which may exaggerate apparent group separation even by cross-validation. Some AUC-perfect features in Table 5 reflect presence/absence patterns rather than stable quantitative changes. HMDB-based identifications in Table 3 may include isobaric or isomeric ambiguities, and many pathway hits rely on single-metabolite mappings, making enrichment outcomes hypothesis-generating rather than definitive. Potential confounding factors by sorting procedures, extender composition, and cryo-handling cannot be fully excluded, and the correlations revealed in Figure 8 do not establish causality or guarantee generalizability across bulls, seasons, or facilities.

Translating these outcomes into practical applications will require targeted quantification of the four-metabolite panel in independent cohorts, mechanistic manipulation of glycolytic substrates, choline donors, monocarboxylate transport, and carnitine in controlled media, and prospective evaluation of the metabolite score Figure 7 in relation to conception or pregnancy outcomes under varying thermal and management conditions. Such studies would link the biochemical patterns in Tables 1–5 and Figures 1–8 to on-farm fertility outcomes, transforming a laboratory-based signature into a robust decision-making tool.

Overall, the combined analysis of movement and metabolite data reveals that X- and Y-sperm have different movement patterns. These differences are caused by how they use energy and how their membranes are built. X-sperm mainly use carbohydrate and phosphocholine pathways, which help keep the membrane stable and control how the sperm moves. In contrast, Y-sperm rely on a carnitine-based oxidative pathway that supports faster and straighter movement. These metabolic processes help explain the differences in sperm quality and movement patterns in Table 1. This study combines past models of sperm energy use with sex-sorted sperm performance, giving a practical view for dairy farms in humid tropical areas. However, the research only used frozen semen from one cattle breed and did not test fertility in live animals, different environments, or various insemination conditions. This means the outcomes still need more testing to prove their reliability and usefulness in real situations.

This study reveals that X- and Y-sperm have clear and consistent differences in both traits and metabolites. X-sperm have greater overall movement and better acrosome integrity. Y-sperm have stronger membrane integrity, fewer shape problems, and move faster in a straight line. The metabolomic analysis revealed clear biomarker differences. X-sperm had greater levels of D-(−)-fructose, L-(+)-lactic acid, and sn-glycerol-3-phosphocholine. Y-sperm had greater levels of acetyl-L-carnitine. These outcomes confirm that certain metabolites are linked to the differences in quality and movement among X- and Y-sperm.