We examined the effect of carbohydrate (CHO) availability on whole body and skeletal muscle protein utilization at rest, during exercise, and during recovery in humans. Six men cycled at ∼75% peak O2 uptake (V̇o2peak) to exhaustion to reduce body CHO stores and then consumed either a high-CHO (H-CHO; 71 ± 3% CHO) or low-CHO (L-CHO; 11 ± 1% CHO) diet for 2 days before the trial in random order. After each dietary intervention, subjects received a primed constant infusion of [1-13C]leucine and l-[ring-2H5]phenylalanine for measurements of the whole body net protein balance and skeletal muscle protein turnover. Muscle, breath, and arterial and venous blood samples were obtained at rest, during 2 h of two-legged kicking exercise at ∼45% of kicking V̇o2peak, and during 1 h of recovery. Biopsy samples confirmed that the muscle glycogen concentration was lower in the L-CHO group versus the H-CHO group at rest, after exercise, and after recovery. The net leg protein balance was decreased in the L-CHO group compared with at rest and compared with the H-CHO condition, which was primarily due to an increase in protein degradation (area under the curve of the phenylalanine rate of appearance: 1,331 ± 162 μmol in the L-CHO group vs. 786 ± 51 μmol in the H-CHO group, P < 0.05) but also due to a decrease in protein synthesis late in exercise. There were no changes during exercise in the rate of appearance compared with rest in the H-CHO group. Whole body leucine oxidation increased above rest in the L-CHO group only and was higher than in the H-CHO group. The whole body net protein balance was reduced in the L-CHO group, largely due to a decrease in whole body protein synthesis. These data extend previous findings by others and demonstrate, using contemporary stable isotope methodology, that CHO availability influences the rates of skeletal muscle and whole body protein synthesis, degradation, and net balance during prolonged exercise in humans.