Abstract Ethyl pyruvate (EP) has emerged as a promising compound with potential therapeutic benefits attributed to its anti-inflammatory and antioxidant properties. This study aimed to understand the effects of EP on plasma metabolites and immune cells in horses, utilizing advanced liquid chromatography-mass spectrometry (LC-MS)-based metabolomics, quantitative polymerase chain reaction (qPCR), and blood chemistry analyses. Our comprehensive analysis detected 2,366 ions, and 126 metabolites were accurately identified. Remarkably, EP administration induced significant changes in 28 metabolites at 1 h and 11 metabolites at 8 h, highlighting its time-dependent impact on metabolic pathways such as phenylalanine and arginine biosynthesis. Moreover, EP significantly lowered the expression of inflammatory markers interleukin (IL)-6 and heme oxygenase (HO)-1, indicating its potential as an anti-inflammatory agent. Blood chemistry analysis revealed notable reductions in glucose and triglyceride levels. These findings demonstrate that EP is a substance with potential effects on pathways associated with inflammation, oxidative stress, and metabolic processes. Supplementary Information The online version contains supplementary material available at 10.1038/s41598-024-75734-1. Keywords: Ethyl pyruvate, Metabolomics, Anti-inflammatory, Antioxidant, Biomarker Subject terms: Animal physiology, Biomarkers, Preclinical research, Molecular medicine __________________________________________________________________ Pyruvate is a natural end-product of glycolysis and serves as the initial substrate for the tricarboxylic acid (TCA) cycle^[30]1. It is not only a vital substance for energy production through the TCA cycle, but is also known to ameliorate various disease conditions through its anti-inflammatory and antioxidant actions^[31]1. However, the therapeutic effectiveness of pyruvate is limited due to its poor stability in aqueous solutions, resulting in the loss of function^[32]1,[33]2. To overcome this problem, a novel synthetic derivative of pyruvate, ethyl pyruvate (EP), has been formulated^[34]1,[35]2. It is more stable and safer than pyruvate, while retaining its beneficial effects^[36]1,[37]2. EP has been shown to exert various biological and pharmacological effects in both in vitro and in vivo experimental models^[38]1,[39]3. These effects are attributed to its influence on redox reactions and metabolic activities, as well as immune functions within the body^[40]1,[41]3. As a derivative of an endogenous substance, EP is also involved in glycolysis and cellular energy metabolism^[42]3,[43]4. Importantly, during these metabolic processes, reactive oxygen species (ROS) are inadvertently produced^[44]3,[45]5,[46]6. An imbalance between ROS and antioxidants can lead to functional and structural alterations in lipids, proteins, and deoxyribonucleic acid (DNA), causing cellular dysfunction and metabolic abnormalities^[47]3,[48]5,[49]6. EP demonstrates a potent ROS scavenging ability by regulating the expression of antioxidant genes such as nuclear factor erythroid 2-related factor 2 (Nrf2), antioxidant responsive element (ARE), heme oxygenase-1 (HO-1), and inducible nitric oxide synthase (iNOS)^[50]3,[51]7–[52]9. Alongside its protective effect on the intracellular redox processes, EP also exhibits anti-inflammatory action by modulating the signaling of immune cells such as high mobility group box (HMGB1), nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB), and mitogen-activated protein kinase (MAPK) and inhibiting the activation and proliferation of macrophages, dendritic cells, and T cells^[53]3,[54]10–[55]13. In equine studies, Jacobs et al. suggested that EP could be used as a treatment for endotoxemia in horses^[56]14. Schroeder et al. demonstrated a significant reduction in inflammatory gene expression after EP administration^[57]15. Additionally, Jonson et al. claimed that EP improves survival rates in cases of colonic volvulus by limiting ischemic injury and promoting intestinal mucosal repair^[58]16. These characteristics of EP have been shown to have benefits in the treatment of various diseases in experimental studies, and efforts to prove its safety and efficacy in the clinical setting are ongoing^[59]1,[60]3. To better understand the diverse effects of EP on the body, a thorough exploration of its mechanisms of action is crucial. Recently, untargeted metabolomics, an approach that does not limit itself to specific targets but instead investigates the entire metabolic profile of cells, has been utilized^[61]17,[62]18. This methodology facilitates a deeper understanding of the mechanism of action of a specific drug and captures unexpected metabolic changes, which can lead to the discovery of new biological targets and pathways. Such insights are essential for developing new therapeutic strategies that minimize drug side effects and enhance efficacy. Furthermore, untargeted metabolomics plays a vital role in discovering biomarkers crucial for the early diagnosis of diseases, monitoring disease progression, and evaluating treatment responses. In this study, we employed liquid chromatography-mass spectrometry (LC-MS)-based metabolomics, quantitative polymerase chain reaction (qPCR), and blood chemistry analyses to evaluate the impact of EP administration on the plasma metabolites and immune cells in horses. Our objective was to identify potential biomarkers and pathways associated with EP. This method not only provided new insights into the effects of EP but also served as a foundation for analyzing its beneficial impact across different species using the metabolomics approach. Materials and methods Materials EP (purity 98%) for administration and phenacetin (internal standard, ISTD) were purchased from Sigma-Aldrich (St. Louis, MO, USA). High-performance liquid chromatography (HPLC)-grade acetonitrile (ACN) and HPLC-grade deionized water (DW) were procured from J.T. Baker (Phillipsburg, NJ, USA). Formic acid (FA) was obtained from Junsei Chem (Chou-ku, Japan). Intravenous EP administration in horses Five horses participated in this study, including two warmblood mares, two Thoroughbred mares, and one Thoroughbred gelding, with an average age of 13 ± 6 years. Blood samples for the pre-administration (pre-admin) data were collected at one hour prior to EP administration. Each horse received 150 mg/kg of EP, administered as a continuous intravenous infusion in one liter of lactated Ringer’s solution over 60 min. Blood samples for LC-MS and laboratory analysis were drawn at one hour and eight hours post-administration (post-admin), and collected using three different types of tubes: serum separator tubes (SSTs) for biochemistry analysis, ethylene diamine tetra-acetic acid (EDTA) tubes for complete blood cell count (CBC) and qPCR, and heparin tubes for LC-MS analysis (BD Vacutainer^® blood collection tubes, Becton-Dickinson and Company, Franklin Lakes, NJ, USA). Anesthesia and euthanasia were not required for this experiment. All animal procedures were approved by the Korea Racing Authority (IACUC-2214, AEC-2213). We confirmed that all experiments in this study were performed in accordance with the relevant guidelines and regulations. All the procedure of the study is followed by the ARRIVE guidelines. LC-MS and qPCR analysis LC-MS and the qPCR analysis are described in detail in the supplementary materials. Data mining and peak annotation Excel files (with information including details such as the mass-to-charge ratio [m/z] values, retention times, and peak areas) for untargeted data analysis and the Mascot Generic Format (MGF) files for metabolite annotation were generated and exported from MZmine 3 (version 3.9.0). To interpret the peak, Global Natural Product Social Molecular Networking (GNPS, [63]https://gnps.ucsd.edu/ProteoSAFe/static/gnps-splash.jsp) and the SIRIUS software (version 5.8.3) were utilized, employing the GNPS Database and Human Metabolome Database (HMDB, [64]https://hmdb.ca/) library. The interpretation was conducted at level 2 annotation^[65]19. The area of the extracted data was initially corrected using the ISTD. The outliers were removed using a filter based on the quality control (QC) function and the dataset underwent preprocessing through total-ion-normalization, log-transformation, and Pareto-normalization using the MetaboAnalyst 6.0 platform ([66]https://www.metaboanalyst.ca)^[67]20. To validate the robustness of the analytical method, Pearson correlation analysis was performed on the QC samples using SPSS 27.0 (SPSS Inc., Chicago, IL, USA). The chromatograms and MS information in the raw data of LC-MS were explored using the Xcalibur software (Qual Browser, Thermo Fisher Scientific). The multivariate statistical analysis of plasma samples and metabolite enrichment analysis were performed using MetaboAnalyst 6.0. The lipid enrichment analysis was performed by The Lipid Pathway Enrichment Analysis platform (LIPEA, [68]https://hyperlipea.org). Statistical significance was determined using Student’s t-test, with a p-value of less than 0.05 considered significant. For data visualization, Excel 2016 (Microsoft, Seattle, WA, USA) and the GraphPad Prism software 6.0 (San Diego, CA, USA) were utilized. Results Untargeted metabolite profiling and statistical discrimination In a non-targeted metabolic approach conducted to study the changes in metabolites after EP treatment (each group, n = 5), a total of 2,366 ions were detected in the plasma samples. Of these, 126 metabolites were identified at Level 2 through the HMDB. Excluding one metabolite (PC (16:0/14:1(9Z)); mass error, -8.2 ppm), the precursor mass errors relative to the theoretical mass satisfied the criterion of being within 5 ppm (Table [69]S1). The abbreviations and full names of the chemicals investigated in this study are listed in Table [70]S2. The stability and quality of data from this non-targeted metabolic approach were ensured through statistical correlation of the QC samples (R^2 > 0.930) (Fig. [71]1). Fig. 1. [72]Fig. 1 [73]Open in a new tab Pearson correlation between QC samples. Among these, to impartially distinguish variables (n = 2,366) that significantly affected differentiation before and after EP administration over time, to identify significant markers between groups through multivariate analysis, the unsupervised method Principal Component Analysis (PCA) was attempted. However, the PCA results did not reveal a significant separation between the groups (Figure [74]S1). As a result, the supervised method, an Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA), was performed (Fig. [75]2). The results from the S-plot, which visualized information on the variable projected importance (VIP) scores (VIP > 1) and significant changes in the metabolites (p-value < 0.05), revealed significant metabolites in the 1-hour and 8-hour compared to the pre-admin. Among the identified metabolites, 24 exhibited increased levels in the 1-hour post-admin compared to the pre-admin, while 4 showed decreased levels (Fig. [76]2A; Table [77]1). Additionally, in the comparison between the pre-admin and the 8-hour post-admin, 6 metabolites increased while 5 decreased (Fig. [78]2B; Table [79]2). The OPLS-DA score plot utilized R²Y values (pre-admin vs. 1-hour post-admin, 0.878; pre-admin vs. 8-hour post-admin, 0.746) and was explained by the T-scores in the score plots (pre-admin vs. 1-hour post-admin, 13.7%; pre-admin vs. 8-hour post-admin, 14.2%). Furthermore, the plot demonstrated that the orthogonal T-scores (pre-admin vs. 1-hour post-admin, 27.9%; pre-admin vs. 8-hour post-admin, 29.6%) did not contribute to the variation affecting separation, thereby reducing the model’s noise, and enhancing the accuracy of differentiation. Fig. 2. [80]Fig. 2 [81]Open in a new tab Time-resolved Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA) score, loading S-plots before and after ethyl pyruvate (EP) administration at (A) 1 h, (B) 8 h. p[1] represents the loading vector of covariance in the first principal component, while p(corr)[1] represents the loading vector of correlation in the first principal component. Loading S-plot with annotated metabolites meeting criteria of |log[2](FC)| > 2, p-value < 0.05, and variable projected importance (VIP) > 1 (increased metabolites in red dot, decreased metabolites in blue dot). Detailed information has been summarized on the right side of the loading S-plot (* p-value < 0.05, ** p-value < 0.01). Table 1. Metabolites that exhibited significant changes 1 h after pre-admin (p-value < 0.05) (p-value < 0.05). No. Identification Log[2](FC) p-value 1 (14 S)-14,15-Dihydroxy-8(17),13(16)-labdadien-19-oic acid 2.2 * 9 Androsterone 2.7 * 13 CE(22:5(7Z,10Z,13Z,16Z,19Z)) 0.9 * 19 Dehydroabietic acid 2.8 * 20 Histidine -0.8 * 25 Phenylalanine 0.6 * 34 Homo-L-arginine -1.4 * 44 Linoleoyl ethanolamide 2.5 * 45 L-Tyrosine 0.4 * 46 LysoPA(18:1(9Z)/0:0) 0.7 * 47 LysoPC(0:0/16:0) 0.7 * 49 LysoPC(0:0/18:1(9Z)) 0.7 * 50 LysoPC(18:0/0:0) 0.3 * 53 LysoPC(20:4(8Z,11Z,14Z,17Z)/0:0) 0.8 * 56 LysoPC(P-18:1(9Z)/0:0) 0.6 * 58 LysoPS(18:0/0:0) 2.3 * 68 PC(16:0/14:1(9Z)) 1.8 * 81 PC(18:3(6Z,9Z,12Z)/16:0) 4.6 ** 82 PC(18:3(9Z,12Z,15Z)/19:0) -1.6 * 87 LysoPE(18:0/0:0) 1 ** 90 Phenylalanylleucine 2.1 ** 92 Platelet-activating factor 0.4 * 95 Retinoyl b-glucuronide 2.5 * 110 Valylleucine 3.4 * 112 Citrulline -1.1 * 114 LysoPC(16:0/0:0) 0.7 ** 115 Decanoylcarnitine 2 * 126 Docosapentaenoic acid (22n-3) 2.8 * [82]Open in a new tab Table 2. Metabolites that exhibited significant changes 8 h after pre-admin (p-value < 0.05) (p-value < 0.05). No. Identification Log[2](FC) p-value 1 (14 S)-14,15-Dihydroxy-8(17),13(16)-labdadien-19-oic acid 2.4 * 19 Dehydroabietic acid 2.6 * 34 Homo-L-arginine 0.7 * 54 LysoPC(20:5(5Z,8Z,11Z,14Z,17Z)/0:0) -0.6 * 85 PC(O-36:4) -0.6 * 87 LysoPE(18:0/0:0) -0.7 * 90 Phenylalanylleucine 3.2 ** 92 Platelet-activating factor -0.4 * 107 Threonylphenylalanine 2.1 ** 110 Valylleucine 7.9 * 121 alpha-Linolenic acid -1.8 * [83]Open in a new tab Pathway and metabolite set enrichment analysis We identified the metabolic pathways incorporating a total of 126 metabolites verified through the database (Fig. [84]3A,B). Among these, we distinguished the pathways containing the metabolites that exhibited significant temporal changes post-EP administration to analyze the time-dependent effects of EP (p-value < 0.05, n = 38) (Fig. [85]3C). The results indicated that metabolic pathways such as arginine biosynthesis, linoleic acid metabolism, tryptophan metabolism, and the biosynthesis of phenylalanine, tyrosine, and tryptophan were implicated (p-value < 0.05, pathway impact > 0.2). Additionally, the results from the lipid enrichment analysis performed in LIPEA included pathways involving linoleic acid metabolism, lipid metabolism, and arachidonic acid metabolism (transformed lipids > 5.5%). The identified metabolites and pathway-specific changes are schematically organized in Fig. [86]4A and B. Fig. 3. [87]Fig. 3 [88]Open in a new tab Analysis of 126 metabolites identified from the metabolomic study in plasma: (A) Results of the metabolic pathway analysis, and (B) Results of the enrichment analysis. (C) Enrichment analysis results of metabolites with significance from the time-course Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA) results in Fig. [89]2 (|log[2](FC)| > 2, p-value < 0.05, and variable projected importance [VIP] > 1). Fig. 4. [90]Fig. 4 [91]Open in a new tab (A) Schematic organization of metabolic pathways for plasma metabolites and immune cells after 1-hour ethyl pyruvate (EP) administration. We applied color coding to the text based on the Log[2] fold changes of metabolites, with significant changes (p-value < 0.05) highlighted in bold and marked with an asterisk (*). IL-6, HO-1, and glucose levels are shown based on qPCR and blood chemistry results (see Fig. 6). Substances shown in black text represent metabolites that were not detected by MS. (B) The relative intensity changes of metabolites involved in the significant metabolic pathways (phenylalanine biosynthesis, arginine biosynthesis, and steroid hormone biosynthesis) are described separately. Changes in plasma metabolites related to the lipid metabolic pathway are described in detail in Fig. [92]5A. Among the pathways identified through enrichment analysis, this study highlighted the changes in phenylalanine biosynthesis and arginine biosynthesis in horses, 1 h after EP administration (Figs. [93]3C and [94]4). Following EP administration, significant increases in phenylalanine and tyrosine in the phenylalanine biosynthesis pathway (p < 0.05) were observed. Conversely, the arginine biosynthesis pathway showed significant decreases in citrulline and homoarginine (p < 0.05) after EP treatment. Additionally, while the steroid hormone biosynthesis pathway showed lower significance in the enrichment analysis compared to other pathways, androsterone significantly increased one hour after EP administration. Although cortisol showed an increasing trend one hour after administration, it significantly decreased after eight hours (p < 0.05) (Figs. [95]4 and 6). Regarding lipid metabolism, changes in the metabolites were observed at 1-hour and 8-hour post-admin compared to the pre-admin group, after log[2] (fold change, FC) transformation, including 4 sphingolipids and 40 phospholipids. Furthermore, a cluster heatmap analysis between the 1-hour group, which showed the largest changes in lipid metabolites, and the pre-admin group revealed distinct differences. For phospholipids, significant increases were observed in Lysophosphatidylethanolamine (LysoPE) (n = 1), Lysophosphatidylcholines (LysoPC) (n = 6), Lysophosphatidic acid (LysoPA) (n = 1), and Phosphotidylcholine (PC) (n = 3) one hour after EP administration, with PC (18:3(9Z,12Z,15Z)/19:0) being the only one showing a significant decrease. After 8 h post-EP administration, significant decreases were observed in LysoPE (n = 1), LysoPC (n = 1), and PC (n = 1) compared to the pre-admin group. However, no significant changes were noted in the sphingolipids. The metabolic pathways related to the lipid changes in this study are schematically summarized in Fig. [96]5. Fig. 5. [97]Fig. 5 [98]Open in a new tab (A) Summary of lipid metabolic pathways. Metabolites in black text are annotated metabolites, and metabolites in gray text are not detected. (B) Cluster heat map of differential lipid metabolites. (C) Time-course log[2](FC) values of lipid metabolites relative to pre-admin intensity. Bold black squares indicate values with p < 0.05. Analysis of mRNA expression in peripheral blood mononuclear cells (PBMCs) To assess the impact of EP on immune cells, the changes in mRNA levels of six factors were measured in PBMCs: interleukin 6 (IL-6), human heat shock 70 kDa protein 6 (HSPA6), heme oxygenase 1 (HO-1), interleukin 10 (IL-10), superoxide dismutase type 1 (SOD-1), and Fos proto-oncogene, AP-1 transcription factor subunit (Fos) (Fig. [99]6A). These factors are related to immune regulation, stress response, antioxidation and cell protection, as well as cell integrity and function. The results showed that the mRNA expression of IL-6 and HO-1 significantly decreased one hour after EP administration, with ratios of 0.578 ± 0.225 and 0.0976 ± 0.180, respectively, against the pre-admin group (p < 0.05). The ratios were calculated by dividing the data from 1-hour post-admin data by the data from the pre-admin. Additionally, a statistically significant decrease was observed in HO-1 expression eight hours post-administration (0.362 ± 0.364, p < 0.05). No significant changes were observed in the remaining four factors. Fig. 6. [100]Fig. 6 [101]Open in a new tab (A) Quantitative PCR results for messenger ribonucleic acid (mRNA) expression related to cellular functions in peripheral blood mononuclear cells (PBMCs) following ethyl pyruvate (EP) administration. (B) Results of blood biochemical analyses after EP administration. Hematological and blood biochemical analysis The impact and safety of EP on horses were evaluated through a total of 18 blood parameters (Fig. [102]6B, Figure [103]S2). Among these, the glucose level was significantly decreased one hour after EP administration, from 90.8 ± 5.72 to 65.4 ± 13.89 mg/dL (p < 0.01), compared to the pre-admin group. Triglyceride (TG) levels substantially decreased one hour and eight hours after EP administration, from 38.4 ± 13.11 to 22.6 ± 3.05 and 23.8 ± 5.93 mg/dL, respectively (p < 0.05). The plasma cortisol level was notably decreased from 3.12 ± 1.25 to 1.12 ± 0.68 µg/dL eight hours after EP administration (p < 0.01). However, parameters related to lipid metabolism (total cholesterol, TC; lipase; total bilirubin; total protein), liver function (aspartate aminotransferase, AST; gamma-glutamyl transferase, GGT; alkaline phosphatase, ALP, alanine transaminase, ALT), kidney function (blood urea nitrogen, BUN; creatinine), immune response (white blood cells, WBC; albumin/globulin ratio), and electrolyte balance (calcium, sodium, potassium, chloride) showed no statistically significant differences between the groups. Discussion This study explored the changes in the metabolic pathways in horse plasma in response to EP administration, using LC-MS-based metabolomics, qPCR, and blood chemistry analysis. Non-targeted metabolite profiling identified 126 metabolites out of a total of 2,366 ions, with multivariate statistical techniques and pathway analysis highlighting significant changes in key metabolic pathways (phenylalanine biosynthesis, arginine biosynthesis, and lipid metabolism) following EP administration. Additionally, the study observed reductions in the expression of IL-6 and HO-1, as well as decreases in glucose, TG, and cortisol levels in immune cells induced post-EP administration. The effects of EP varied over time when monitored at 1 h and 8 h post-EP administration. More extensive impacts were noted in the 1-hour group, suggestive of the duration of the effects of EP. Previous studies indicate that the effects of EP are dose-dependent and tend to return to baseline levels 24 h post-administration, demonstrating EP’s transient bioavailability and efficacy^[104]21,[105]22. This study provides new insights into the duration of EP’s efficacy by monitoring its effects over short intervals post-administration and signifies the importance of employing various methodologies, including a metabolomics approach for the first time while studying the effects of EP. Phenylalanine and tyrosine metabolism plays a crucial role in maintaining normal biological functions by synthesizing compounds that are precursors to essential substances such as melanin, neurotransmitters, and hormones^[106]23,[107]24. These metabolites interact with pyruvate through various biochemical pathways during glycolysis and energy production^[108]25. Although the mechanisms underlying these pathway changes induced by ethyl pyruvate (EP) are not fully elucidated, one recently discovered mechanism involves the inhibition of pyruvate kinase (PK), which converts phosphoenolpyruvate (PEP) to pyruvate, leading to reduced ATP production^[109]26,[110]27. This inhibition by EP has shown to cause ATP depletion in vitro, yielding effective therapeutic results in diseases such as cancerous tumors and trypanosomiasis^[111]27. PEP is a crucial precursor in the biosynthesis of aromatic amino acids like tyrosine, phenylalanine, and tryptophan^[112]28. The changes in the metabolism of aromatic amino acids observed in this experiment suggest that the inhibition of PK by EP affects the conversion process of PEP. EP administration affected the arginine metabolic pathway. Previous studies have reported the anti-inflammatory and protective effects of EP against cellular damage, with one of the primary mechanisms being the inhibition of nitric oxide synthase (NOS) in the arginine biosynthesis pathway^[113]21,[114]29,[115]30. These effects are influenced by the regulation of NOS and the expression of amino acids, such as arginine and citrulline^[116]31–[117]33. In this study, a notable decrease in citrulline expression was observed. Recent studies in rodents have shown that citrulline independently regulates CD4 T cell proliferation and the function of FoxP3 + regulatory T cells^[118]34,[119]35, suggesting a potential link between EP’s immunomodulatory effects through citrulline. The immunomodulatory effects of EP are presumed to occur through various mechanisms, including the involvement of citrulline. The aforementioned phenylalanine and tyrosine are also associated with inflammation^[120]36,[121]37, and dehydroabietic acid, which significantly increased one hour and eight hours after EP administration, regulates inflammation by reducing NO production through the NF-κB and AP-1 pathways^[122]38,[123]39. While further research is needed to elucidate the detailed mechanisms and validate these potential effects, this study suggests a possible link between EP and inflammation, underscoring its potential as a therapeutic agent for inflammatory diseases. Based on this potential, an investigation of the mRNA expression in plasma PBMCs revealed that IL-6 expression, a factor involved in inflammatory regulation, antioxidation, and signaling, significantly decreased one hour after EP administration. Several in vitro studies have reported that EP reduces pro-inflammatory cytokines such as IL-1β, tumor necrosis factor-α (TNF-α), and IL-6 in PBMCs while increasing regulatory cytokines like IL-10 ^[124]40,[125]41. However, in this study involving horses, while IL-6 levels decreased, there were no significant changes in IL-10. The mRNA levels of HSPA6, related to cell protection and stress recovery, showed an increasing trend following EP administration. Notably, HO-1 significantly decreased 1 h post-administration and this effect persisted up to eight hours. Previous research indicates that the induction of HO-1 following EP administration, leading to suppression of HMGB1, is one of the key defense mechanisms during cellular stress and damage^[126]3,[127]30,[128]42. However, the findings of the current study did not align with these results. This suggests the necessity for extensive additional analysis across various animal species and organs and highlights the need to study the pattern of HO-1 expression in diverse environments and its effects on cell survival. Lipids are biochemical molecules essential for cell function and maintenance of life and are characterized by diverse structures and functions. These lipids are divided into several subclasses, including phospholipids, sphingolipids, and neutral lipids, and each of them performs crucial physiological roles such as cell membrane composition, energy storage, and signal transduction. Within the plasma, lipids are transported by lipoprotein complexes such as chylomicrons, very-low-density lipoproteins (VLDL), low-density lipoproteins (LDL), and high-density lipoproteins (HDL), which play a key role in the internal movement and distribution of lipids. A study by Hauser et al. demonstrated that lipid peroxidation induced by lipopolysaccharide significantly decreased following EP treatment, suggesting that EP possesses potential antioxidative capabilities that can regulate lipid metabolism and protect cells from oxidative stress^[129]43. However, a study by Robert et al., which involved consistent consumption of a 0.3% EP solution by rats over six weeks, observed no significant changes in the concentrations of TC, TG, and free fatty acids (FFAs), indicating that EP’s effects might be limited to phospholipids^[130]44. In this study, significant increases in phospholipid levels were observed in 11 out of the 40 phospholipids, one hour after the EP administration. Nevertheless, the observed decrease in phospholipid levels after 8 h suggests that the effects of EP on phospholipids may be transient. Despite this, following EP administration, the levels of certain phospholipids (lysoPE, lysoPC, lysoPA, PC) increased, whereas the levels of other phospholipids (PC) decreased, limited monitoring within the plasma, and the presence of various factors that can influence phospholipids suggest limitations to these possibilities. Additionally, the fact that changes in the phospholipids can regulate the activity of pyruvate oxidation enzymes in the energy metabolism process indicates a relationship between phospholipids and pyruvate metabolism^[131]45. Considering the potential of EP to enhance survival rates through its anti-inflammatory effects and long-term functional improvement, which may be related to the stabilization and functional enhancement of phospholipids in the cell membranes, protection, and regulation of the nervous system, further research on phospholipids appears necessary. Glucose levels significantly decreased one hour after EP administration. Previous research has indicated that EP improves secondary liver damage and renal complications caused by diabetes, enhances immunoregulatory components, and suppresses transcriptional regulators such as HMGB1, thereby preventing the progression of diabetes and lowering glucose levels^[132]46–[133]48. However, the direct effect of EP on the reduction of glucose levels remains uncertain, and the restoration of glucose levels in this study after eight hours suggests that the effect of the drug may be transient. The significant decrease in plasma TG levels observed after EP administration persisted beyond eight hours. Previous studies^[134]3,[135]49 have reported that EP ameliorates elevated TG levels due to alcoholic liver disease or carbon tetrachloride (CCL4)-induced liver damage in rodents, a finding that aligns with this study conducted on healthy horses, confirming that EP also reduces TG under healthy conditions. We utilized untargeted metabolite profiling through high-resolution mass spectrometry to measure in vivo changes induced by EP. This methodology was selected due to its flexibility in sample preparation and analysis compared to targeted approaches and can robustly capture a wide range of molecular interactions. This approach allowed us to screen for variations in metabolites induced by EP administration that were not revealed in previous targeted experimental models, providing new insights into metabolic pathways crucial for physiological maintenance and immunological functions. However, several limitations must be considered when interpreting these results. Although we identified altered metabolites and relevant pathways in horses using a non-targeted metabolomics approach with EP as a screening technique, targeted metabolomics is necessary to validate and better understand the regulated metabolome influenced by EP in future studies. Additionally, the effects of these identified metabolites on the inflammatory response should be determined by performing in vitro experiments, specifically examining cytokine expression in horse immune cell cultures treated with these metabolites. Conducting these experiments may provide a deeper understanding of the role of the altered metabolome by EP in inflammation and related diseases. Furthermore, a total of five horses were involved in this study, which may seem like a small number compared to experiments using other animal species. Conducting experiments with horses, which are large animals, requires substantial resources and incurs high costs, which limits the use of large numbers. Due to these physical and methodological constraints, understanding the various in vivo changes induced by specific substances necessitates considerable time, effort, and resources, making it technically challenging. This underscores the scarcity of experimental results obtained from studies involving horses. Also, this omics-based study is important because it identifies changes in previously unknown substances and facilitates faster and more efficient analyses in future experiments involving a larger number of animals across various species. Additionally, while this experiment focused on changes in blood metabolites and immune cells, it also established a foundation for comparing variations in other specimens such as urine and saliva. Therefore, to gain a deeper understanding of how specific substances induce changes across various biological samples, further studies using different biological specimens are necessary. Conclusion This study examined the effects of EP on plasma metabolites and immune cells in horses using LC-MS-based metabolomics, qPCR, and blood chemistry analyses. A total of 2,366 ions were detected, with 126 metabolites identified. EP administration resulted in changes to 28 metabolites at 1 h and 11 metabolites at 8 h, indicating its time-dependent effects. Significant changes were noted in phenylalanine biosynthesis, arginine biosynthesis, and lipid metabolism. EP also reduced IL-6 and HO-1 levels, suggesting its anti-inflammatory and antioxidant properties. Blood chemistry showed decreases in glucose and TG, highlighting EP’s potential therapeutic benefits. This study contributes to a deeper understanding of the physiological effects of EP by evaluating its impact on various metabolic pathways and immune responses over time. Given the transient efficacy and sustained changes in metabolites, EP shows significant potential as a strategic therapeutic agent for the treatment and prevention of diseases linked to specific pathways. Future research should consider these findings to explore the long-term effects of EP and its applicability across various animal models to further establish its therapeutic potential. Electronic supplementary material Below is the link to the electronic supplementary material. [136]Supplementary Material 1^ (510.5KB, docx) [137]Supplementary Material 2^ (505.4KB, csv) Acknowledgements