Abstract
und: Branched-chain amino acids (BCAAs) are popular as sports supplements due to their ability to enhance performance and recovery. However, the full spectrum of metabolic alterations triggered by acute supplementation with BCAAs in conjunction with exercise remains incompletely understood. Methods: A randomized crossover trial was conducted in 8 healthy active young males, who received either BCAA or placebo supplementation for three consecutive days prior to a high-intensity cycling test. Plasma samples were collected pre- and post-exercise and analyzed by ultra-high-performance liquid chromatography–quadrupole time-of-flight mass spectrometry, followed by correla- tion and enrichment analyses. Results: Acute BCAA supplementation was significantly associated with enhanced fat oxidation and attenuated post-exercise increases in plasma am- monia, creatine kinase, and lactate dehydrogenase, suggesting the potential improvements in energy supply and membrane stability. Metabolomics analysis identified differential metabolites primarily involved in lipid, amino acid,
analyzed by ultra-high-performance liquid chromatography–quadrupole time-of-flight mass spectrometry, followed by correla- tion and enrichment analyses. Results: Acute BCAA supplementation was significantly associated with enhanced fat oxidation and attenuated post-exercise increases in plasma am- monia, creatine kinase, and lactate dehydrogenase, suggesting the potential improvements in energy supply and membrane stability. Metabolomics analysis identified differential metabolites primarily involved in lipid, amino acid, and glucose metabolism. Pathway enrichment revealed coordinated regulation of fatty acid oxidation (FAO) and tryptophan- related pathways. Correlation analysis further showed that changes in metabolite profiles were strongly associated with biochemical outcomes, particularly linking enhanced fat oxidation and ammonia clearance with BCAA intake. Conclusions: Short-term BCAA supplementation could enhance FAO and membrane stability via coordinated regulation of lipid and amino acid metabolism post exercise, supporting its potential role as a precision nutrition strategy. Keywords:branched-chain amino acids; untargeted metabolomics; sports nutrition; energy metabolism; dual regulation Metabolites2026,16, 41 https://doi.org/10.3390/metabo16010041
Metabolites2026,16, 41 2 of 16 1. Introduction Vigorous exercise can induce obvious metabolic stress, depleting energy stores and causing muscle protein damage, which underscores the importance of targeted nutritional strategies for optimal recovery [1]. Among these strategies, branched-chain amino acids (BCAAs)—leucine, isoleucine, and valine—are widely used for their dual role as signaling molecules for protein synthesis and substrates for energy production [2–4]. BCAAs are also deeply involved in key metabolic processes such as lipid oxidation and the tricarboxylic acid (TCA) cycle [2,5,6]. For example, BCAAs supplementation can enhance carnitine- mediated fatty acid transport, thereby improving exercise efficiency and reducing post- exercise fatigue [5]. Beyond their anabolic function, BCAAs may serve as alternative fuel sources during prolonged or glycogen-depleting exercise. However, this can also lead to the production of BCAA-derived metabolites and potentially contribute to a state known as “anabolic resistance,” characterized by diminished muscle protein synthesis and a reduced adaptive response to training, often linked with chronic fatigue, inflammation, or dysregulated protein turnover. Additionally, BCAAs have been associated with the attenuation of both central and peripheral fatigue [5,7,8]. During prolonged strenuous exercise, the elevated brain trypto- phan levels can promote serotonin (5-HT) synthesis, which is linked to fatigue perception. BCAAs compete with tryptophan for transport across the blood–brain barrier, and then lower the production of 5-HT and potentially delay central fatigue [5]. Furthermore, BCAA intake can attenuate metabolic acidosis and the accumulation of ammonia, maintain the levels of blood glucose and free fatty acid, and reduce increases in muscle damage markers such as creatine kinase (CK) and lactate dehydrogenase (LDH), collectively supporting more efficient recovery [9]. Meanwhile, the efficacy of BCAAs supplementation appears to be modulated by sev- eral factors, including exercise intensity, modality, and individual nutritional status [10]. BCAA supplementation after exercise is primarily recognized for alleviating delayed-onset muscle soreness (DOMS) and promoting muscle repair. Systematic reviews indicate that in trained individuals, continuous intake (≤255 mg/kg/day) of BCAAs after moderate muscle damage significantly reduces soreness within 24–72 h and accelerates functional recovery [11]. Notably, the efficacy of BCAA supplementation is modulated by exercise intensity, with more pronounced effects observed
supplementation after exercise is primarily recognized for alleviating delayed-onset muscle soreness (DOMS) and promoting muscle repair. Systematic reviews indicate that in trained individuals, continuous intake (≤255 mg/kg/day) of BCAAs after moderate muscle damage significantly reduces soreness within 24–72 h and accelerates functional recovery [11]. Notably, the efficacy of BCAA supplementation is modulated by exercise intensity, with more pronounced effects observed during prolonged or high-intensity exertion, whereas benefits are less consistent in low-intensity scenarios. These variable outcomes suggest that BCAAs exert multifaceted, context-dependent influences on energy expenditure and recovery processes. While previous studies have often focused on iso- lated metabolic parameters—such as respiratory exchange ratio (RER),β-hydroxybutyrate, or ammonia—this targeted approach may overlook the broader, coordinated metabolic adaptations induced by BCAA supplementation following exercise. In this context, non-targeted metabolomics offers a powerful tool for capturing global changes in metabolites such as amino acids, lipids, and other metabolites, thereby provid- ing a comprehensive profile of the metabolomic response to both exercise and nutritional interventions [9,11]. Previous applications of this approach have successfully revealed dynamic shifts in fatty acids, carnitine derivatives, and TCA cycle intermediates in re- sponse to exercise, reflecting underlying substrate utilization and energy metabolism transitions [12,13]. Although numerous studies have examined the role of BCAAs in regu- lating exercise metabolism, few have systematically employed metabolomics to elucidate how BCAAs coordinately regulate fatty acid oxidation (FAO), amino acid metabolism, and mitochondrial function in an integrated manner [14]. Therefore, the objective of this study was to investigate the effects of acute BCAA sup- plementation on pre- or post-exercise energy metabolism in young males, using untargeted https://doi.org/10.3390/metabo16010041
Metabolites2026,16, 41 3 of 16 plasma metabolomics, aiming to elucidate the mechanisms by which BCAAs promote energy replenishment and muscle recovery. We hypothesized that BCAA supplementa- tion would enhance post-exercise recovery by coordinately modulating FAO, amino acid metabolism, and mitochondrial function, as reflected in the global metabolomic profile. These findings may contribute to the development of precision nutrition strategies for athletic recovery and supporting metabolic health. 2. Materials and Methods 2.1. Participants A total of 8 healthy male university students who were regularly physically active were recruited. Regular physical activity was confirmed using the International Physical Activity Questionnaire (IPAQ) short form, with participants required to meet moderate-to-vigorous physical activity criteria (≥150 min of moderate-intensity or≥75 min of vigorous-intensity activity per week). Inclusion criteria were: healthy males aged 20–25 years, body mass index (BMI) between 18 and 22 kg/m 2 ; and having maintained at least 4 weeks of stable training prior to the study, with≥3 sessions per week of moderate-intensity exercise (e.g., running or cycling) for at least 30 min each; and willing to adhere to the study protocol. Exclusion criteria included: history of serious illness or surgery within the past 3 months; presence of metabolic disorders such as diabetes or hyperlipidemia; use of medications or nutritional supplements (including BCAAs) within the past 6 months; intense exercise or alcohol consumption within 48 h before testing; musculoskeletal injury during the study; and participation in other strenuous exercise outside the experiment. A priori sample size calculation was performed using G*Power software (version 3.1.9.7) based on data from a previous study investigating BCAA supplementation effects on exercise performance. With an effect size f = 0.3,α= 0.05, power (1-β) = 0.80, and correlation among repeated measures = 0.5, the calculated sample size was 16. Then we recruited 8 participants to account for potential dropouts, providing adequate statistical power for the primary outcomes. Anthropometric characteristics of the participants are presented as mean±standard deviation (SD): age 22.5±1.2 years, height 175.3±2.8 cm, weight 72.4±4.1 kg, BMI 21.3±1.4 kg/m 2 , the percentage of body fat 16.66±5.41%. All participants provided written informed consent, and the study was approved
Then we recruited 8 participants to account for potential dropouts, providing adequate statistical power for the primary outcomes. Anthropometric characteristics of the participants are presented as mean±standard deviation (SD): age 22.5±1.2 years, height 175.3±2.8 cm, weight 72.4±4.1 kg, BMI 21.3±1.4 kg/m 2 , the percentage of body fat 16.66±5.41%. All participants provided written informed consent, and the study was approved by the Ethics Committee of Beijing Sport University (approval number: 2023233H). 2.2. Study Design This study employed a randomized, placebo-controlled, crossover design and was conducted between March and July 2023, with a total duration of 10 weeks including participant recruitment, experimental trials, washout periods, and preliminary analysis. The study protocol followed the CONSORT (Consolidated Standards of Reporting Trials) guidelines for randomized crossover trials. There were no changes to the outcomes or methodology after the trial commenced. 2.3. Randomization and Supplementation Protocol Participants were randomly assigned using a computer-generated random number sequence to receive either BCAA or placebo in the first trial, with the order crossed over after a 7-day washout period. The BCAA supplement consisted of a leucine: isoleucine: valine mixture (2:1:1 ratio; Beijing Combest Co., Beijing, China) at a dose of 0.2 g/kg body weight per serving, twice daily (morning and evening) for three consecutive days [5,15]. This dosing strategy was consistent with previous trials demonstrating the effectiveness and safety of short-term BCAA supplementation in exercise settings [7,11]. The placebo consisted of an isocaloric corn starch powder (Beijing Combest Co., China) matched for https://doi.org/10.3390/metabo16010041
Metabolites2026,16, 41 4 of 16 color, texture, odor, and taste. Both supplements were dissolved in 300 mL of water in identical unlabeled containers and immediately consumed under supervision. 2.4. Dietary Control Participants were instructed to record their habitual dietary patterns and intake for one week preceding the study using the Mint Health mobile application (version 2.5.1), a validated dietary assessment tool with a comprehensive food database, to assess baseline nutrition and minimize confounding factors. During the trail, their diets were strictly controlled and detailed records were maintained: participants maintained their usual diet without special dietary interventions, and the intake of any additional nutritional supplements (especially protein supplements) was prohibited. Alcohol and caffeinated beverages were avoided throughout the supplementation period to minimize confounding. all meals (including snacks) were photographed and logged daily through the Mint Health application. Total daily energy intake and macronutrient composition were analyzed to ensure consistency between trails. To minimize the influence of acute dietary intake on metabolic outcomes, all participants fasted for at least 8 h before each test day, with only water permitted. Based on these records, the average daily energy and macronutrient intakes for the BCAA group were 2082±330 kcal (carbohydrates 243±45 g, fat62±19 g , protein 113±12 g), and for the placebo group were 2090±360 kcal (carbohydrates 249±43 g , fat 65±18 g, protein 115±14 g). No significant differences were observed in pre-exercise dietary intake between conditions, supporting dietary standardization. 2.5. Experimental Procedure Two weeks prior to the experimental trials, each participant accepted dual-energy X-ray absorptiometry (iDXA, GE Healthcare/Lunar, Madison, WI, USA) to measure body composition, completed an incremental cycling test to determine their maximal oxygen uptake (VO2max). Following a 5 min rest when arrived at the laboratory, participants performed an incremental exercise test to volitional exhaustion on a Monark cycle ergometer (Ergomedic 839E, Monark Exercise AB, Vansbro, Sweden), beginning with a 5 min warm-up at 80 W. After a 3 min rest, the workload started at 80 W and increased by 40 W every 3 min until 200 W, thereafter increasing by 20W every 2 min until volitional exhaustion, with participants maintaining a
to volitional exhaustion on a Monark cycle ergometer (Ergomedic 839E, Monark Exercise AB, Vansbro, Sweden), beginning with a 5 min warm-up at 80 W. After a 3 min rest, the workload started at 80 W and increased by 40 W every 3 min until 200 W, thereafter increasing by 20W every 2 min until volitional exhaustion, with participants maintaining a cadence of≥60 rpm. Expired gases were analyzed breath-by- breath using a MetaLyzer 3B system (CORTEX Biophysik GmbH, Leipzig, Germany). On the day before each trial, participants were instructed to maintain their normal daily routines, while avoiding strenuous exercise and caffeine, and to observe an overnight fast of at least 8 h. On the test day, participants arrived at the laboratory in a fasted state, consumed the assigned supplement, and underwent baseline fingertip and venous blood samples sampling. After a 30 min rest period, additional blood samples were collected, followed by the exercise test. The exercise protocol consisted of two consecutive stages. Stage 1 involved 60 min of constant-load cycling at 60% of VO2max. Stage 2 was performed immediately after- ward and consisted of cycling at 80% VO2max until exhaustion (time to exhaustion, TTE). Throughout the test, respiratory gases were continuously monitored using a metabolic an- alyzer (K5, COSMED, Rome, Italy) to measure VO2and VCO2to assess participants’ endurance capacity. Substrate oxidation rates were calculated using the Frayn equa- tions: carbohydrate oxidation (g/min) = 4.210×VCO2−2.962×VO2; fat oxidation (g/min) = 1.695×VO−1.701×VCO2 , to quantify metabolic efficiency during exercise. In addition, the rating of perceived exertion (RPE) and visual analog scale (VAS) scores were recorded every 15 min during Stage 1. Heart rate were continuously monitored https://doi.org/10.3390/metabo16010041
Metabolites2026,16, 41 5 of 16 throughout both stages. This submaximal steady-state exercise was designed to evaluate metabolic responses under moderate-intensity conditions. Venous blood samples were collected at three time points: baseline (fasted state), 30 min after supplement ingestion (pre-exercise), and immediately after Stage 2 (post- exercise). Plasma samples were processed immediately to separate and stored at−80 ◦ C until analysis. The overall experimental protocol, including the supplementation schedule, exercise test, is illustrated in Figure. Figure 1.Experimental flowchart of the study protocol. Schematic overview of the supplementation, exercise. Participants underwent 8 h fasting followed by BCAA or placebo supplementation within 30 min before exercise testing. The protocol included a constant load exercise (CLE) at 60% VO 2max for 1 h and a time-to-exhaustion (TTE) test at 80% VO 2max. Blood sampling, gas collection, rating of perceived exertion (RPE), and visual analogue scale (VAS) assessments were performed at defined time points. 2.6. Blood Biochemical Analysis Blood Samples were drawn into anticoagulant tubes containing sodium heparin and immediately centrifuged at 3000×gfor 10 min at 4 ◦ C to separate plasma and serum. Blood glucose and lactate were measured using Electrochemical analyzers (EKF Biosen C-Line, Germany). Insulin (INS), non-esterified fatty acids (NEFA), andβ-hydroxybutyrate (β-HB) were quantified using enzyme-linked immunosorbent assays (Sinouk Bio, Beijing, China). Plasma ammonia was measured using an enzymatic cycling assay kit (Sinouk Bio, Beijing, China) according to the manufacturer’s instructions. 2.7. Metabolite Profiling and Quantification One of eight participants withdrew due to injury before the second trail, and a plasma sample of post-exercise of one participant in placebo group due to failure in quality control criteria (e.g., abnormal signal intensity and poor feature alignment), a total of 4 samples was excluded from metabolomic analysis. Therefore, metabolomics analyses were conducted on 44 samples. For metabolite extraction, plasma aliquots (100µL) were mixed with 300µL of ice-cold methanol containing a comprehensive set of internal standards (covering amino acids, organic acids, and other central carbon metabolites). After vortexing and incubation at −20 ◦ C for 30 min to precipitate protein, the mixture was centrifuged (15,000×g, 15 min, 4 ◦ C). The supernatant was transferred, dried
samples. For metabolite extraction, plasma aliquots (100µL) were mixed with 300µL of ice-cold methanol containing a comprehensive set of internal standards (covering amino acids, organic acids, and other central carbon metabolites). After vortexing and incubation at −20 ◦ C for 30 min to precipitate protein, the mixture was centrifuged (15,000×g, 15 min, 4 ◦ C). The supernatant was transferred, dried under a gentle stream of nitrogen, and reconstituted in 100µL of 50% methanol for LC-MS analysis. Metabolite separation and detection were performed using ultra-high-performance liq- uid chromatography coupled with quadrupole time-of-flight mass spectrometry (UHPLC– QTOF–MS) in both positive and negative electrospray ionization modes. Chromatographic separation used a C18 column (2.1×100 mm, 1.7µm) with water (0.1% formic acid) and https://doi.org/10.3390/metabo16010041
Metabolites2026,16, 41 6 of 16 acetonitrile (0.1% formic acid) as mobile phases under a 0.3 mL/min flow rate. The gradient increased from 2% to 98% B within 20 min at 40 ◦ C. Mass spectra were acquired in the range ofm/z50–1200 with a resolution of 30,000. Raw data were processed using Compound Discoverer 3.3 (Thermo Fisher Scientific, Waltham, MA, USA). Pre-processing included features filtering based on retention time and mass-to-charge ratio (m/z) parameters, peak alignment across samples and peak extraction using predefined mass tolerance and adduct settings. Peak areas were used for relative quantification. Metabolite identification was achieved by matching using high-resolution MS 2 spectral libraries (mzCloud and mzVault) and the MassList MS 1 library. Specifically, precursor ionm/zvalues from MS 1 spectra were used to determine the molecular weight of metabolites. Molecular formulas were predicted based on mass accuracy (ppm error < 5) and adduct information, and subsequently matched to reference compounds in the databases. For metabolites with available MS 2 spectra, fragment patterns and collision energy patterns were compared against reference spectra to confirm identity. To ensure data quality, only metabolites with a coefficient of variation (CV) < 30% in pooled QC samples were retained for downstream statistical analysis. Metabolite annotation confidence was assigned following the Metabolomics Standards Initiative (MSI) guidelines. Given that metabolite identification was primarily based on MS 1 accurate mass matching, all reported annotations are considered tentative unless supported by MS/MS spectral matching (Level 2 or higher according to Metabolomics Standards Initiative guidelines). Lipid species are reported as sum compositions (e.g., PC 36:3) rather than specific acyl-chain assignments unless confirmed by orthogonal methods. 2.8. Quality Control and Differential Metabolites Screening Pooled quality control (QC) samples were prepared by combining equal aliquots from all samples and were analyzed periodically to monitor system stability. Before sample analysis, three QC injections were run before the analytical sequence to equilibrate the system, and one QC sample was subsequently injected after every six samples throughout the acquisition run. The injection sequence was randomized. To assess analytical repro- ducibility, the processed data (after normalization and filtering) was unit-variance-scaled (UV-scaled) and subjected to
analyzed periodically to monitor system stability. Before sample analysis, three QC injections were run before the analytical sequence to equilibrate the system, and one QC sample was subsequently injected after every six samples throughout the acquisition run. The injection sequence was randomized. To assess analytical repro- ducibility, the processed data (after normalization and filtering) was unit-variance-scaled (UV-scaled) and subjected to unsupervised Principal Component Analysis (PCA). The tight clustering of the QC samples in the PCA score plot, confirmed the high stability and reproducibility of the analytical platform throughout the sequence. Orthogonal partial Least Squares-Discriminant Analysis (OPLS-DA) was employed to discriminate metabolic profiles between BCAA and placebo groups. The models were constructed using seven-fold cross-validation (or k = 2n for groups with≤3 biological replicates per condition) to obtain the model parameters R 2 and Q 2 . Higher R 2 and Q 2 values (closer to 1.0) indicate greater model stability and predictive reliability. To further assess model robustness and exclude overfitting, 200 random permutation tests were performed, in which class labels were randomly reassigned, and new models were built for each permutation. Regression lines of permuted R 2 and Q 2 values against the correlation coefficient were then generated. Models were considered valid and not overfitted when R 2 > Q 2 and the Q 2 -intercept was below zero. This approach ensured that the resulting OPLS-DA models accurately reflected biological differences rather than random variation. Differential metabolites were identified based on the following criteria: variable importance in projection (VIP)≥1.0 from the OPLS-DA model, fold change (FC) > 1.2 or < 0.83, and adjustedp-value < 0.05 (pairedt-test with Benjamini–Hochberg false discovery rate correction). Metabolites satisfying all three criteria were considered statistically significant. https://doi.org/10.3390/metabo16010041
Description
This study explores the effects of BCAA supplementation on metabolism during endurance exercise.