Introduction Nutritional ketosis has been shown to significantly alter oxidative metabolism during exercise (1-3), restoring metabolic flexibility, and reducing the reliance on glycolysis to sustain high intensity exercise. Limited carbohydrate availability can result in muscle protein breakdown for gluconeogenesis in energetic stress, such as starvation or endurance exercise (4). Therefore, we tested whether nutritional ketosis could reduce muscle proteolysis in exercise, by determining changes in intramuscular concentration of branched-chain amino acids (BCAA) and their relationship to glycolysis. Methods After providing informed consent, and following an overnight fast, 10 male endurance-trained volunteers (VO2max 5.4±0.2 L/min; age 27.6±1.6 y) completed 3 trials of 1h fixed-intensity cycling at 75% WMax in a single-blind crossover design. Before each trial, participants drank taste-matched isocaloric drinks containing ≥96% of calories from carbohydrate (maltodextrin:fructose, 5:1; CHO), ketone ester (573 mg/kg BW; (R)-3-hydroxybutyl (R)-3-hydroxybutyrate; KE), or fat (FAT). Muscle biopsies were taken immediately pre- and post- exercise from vastus lateralis muscle. Muscle metabolites were extracted via the Folch method and analysed for intramuscular D-β-hydroxybutyrate (DβHB), pyruvate, leucine + isoleucine, valine, glucose and glycolytic intermediates using a triple quadrupolar mass spectrometer (Waters, UK). Serial blood samples were obtained via an IV catheter and analysed for DβHB and lactate. Athletes maintained a standard diet for 24h prior to the CHO and KE trials, and an isocaloric high-fat (~70%), low-carbohydrate (~5%) diet prior to the FAT trial. 3-way repeated measures ANOVA with post-Hoc Tukey corrections were used to determine statistical significance (considered as p<0.05). Results Intramuscular leucine + isoleucine was reduced by ~50% following exercise with KE supplementation, compared to CHO or FAT (p<0.05). No exercise differences in valine were seen. After 1h of exercise at 75% WMax, intramuscular leucine + isoleucine positively correlated with pyruvate (r=0.74, p<0.01) and lactate (r=0.61, p<0.05), while DβHB positively correlated with intramuscular glucose (r=0.55, p<0.05) and negatively correlated with pyruvate (r=0.46, p<0.05), glycolytic intermediates (r=0.52, p<0.05) and leucine/isoleucine (r=0.51, p<0.05). Conclusion In this work we have shown that nutritional ketosis reduces the rise in intramuscular BCAA levels during exercise, supporting previous evidence that ketosis tightly regulates glycolysis (and therefore pyruvate), ultimately reducing the requirement for muscle deamination(5). Such metabolic effects have a sound evolutionary basis, limiting the catabolism of carbohydrates and skeletal muscle protein for gluconeogenesis in starvation. These findings may also provide a key to reducing muscle catabolism, or enhancing recovery following sustained exercise.