In skeletal muscle of humans, transcription of several metabolic genes is transiently induced during recovery from exercise when no food is consumed. To determine the potential influence of substrate availability on the transcriptional regulation of metabolic genes during recovery from exercise, 9 male subjects (aged 22-27) completed 75 minutes of cycling exercise at 75% V¿o₂max on 2 occasions, consuming either a high-carbohydrate (HC) or low-carbohydrate (LC) diet during the subsequent 24 hours of recovery. Nuclei were isolated and tissue frozen from vastus lateralis muscle biopsies obtained before exercise and 2, 5, 8, and 24 hours after exercise. Muscle glycogen was restored to near resting levels within 5 hours in the HC trial, but remained depressed through 24 hours in the LC trial. During the 2- to 8-hour recovery period, leg glucose uptake was 5- to 15-fold higher with HC ingestion, whereas arterial plasma free fatty acid levels were 3- to 7-fold higher with LC ingestion. Exercise increased (P < .05) transcription and/or mRNA content of the pyruvate dehydrogenase kinase 4, uncoupling protein 3, lipoprotein lipase, carnitine palmitoyltransferase I, hexokinase II, peroxisome proliferator activated receptor ¿ coactivator-1a, and peroxisome proliferator activated receptor a. Providing HC during recovery reversed the activation of pyruvate dehydrogenase kinase 4, uncoupling protein 3, lipoprotein lipase, and carnitine palmitoyltransferase I within 5 to 8 hours after exercise, whereas providing LC during recovery elicited a sustained/enhanced increase in activation of these genes through 8 to 24 hours of recovery. These findings provide evidence that factors associated with substrate availability and/or cellular metabolic recovery (eg, muscle glycogen restoration) influence the transcriptional regulation of metabolic genes in skeletal muscle of humans during recovery from exercise.