A single bout of exercise is known to improve insulin sensitivity for at least 24-48h (Mikines et al. 1988). However, the molecular modulations underlying this effect are not well characterised. The aim of this study was to investigate the effect of prior exercise on basal and insulin mediated changes in the expression of metabolic genes involved in glucose transport (glucose transporter GLUT4, calpain-10) and oxidation (pyruvate dehydrogenase kinases (PDK)-2 and -4). Muscle PDK controls the activity of the pyruvate dehydrogenase compex (PDC), the enzyme which controls the rate-limiting step in glucose oxidation (Randle 1986), whereas calpain-10 has been associated with GLUT4 mediated glucose transport (Paul et al 2003). Eight healthy male individuals (age 24 ± 2years, BMI 24 ± 1, mean ± SEM) underwent a hyperinsulinaemic (80 mU/l) euglycaemic (4.5 mmol/l) clamp for 240min 24h after 90min of one-legged cycling at moderate intensity (~60% of maximal oxygen uptake) while keeping the other leg sedentary (control leg). During the subsequent 24h recovery period, subjects consumed a normal mixed diet (55% CHO, 30% fat, 15% protein). Muscle biopsy samples (vastus lateralis) were obtained from both legs after exercise (day 1), and before and after 4h of insulin infusion (day 2) for the determination of mRNA content of the genes under investigation (normalised to alpha actin) by quantitative real-time PCR using Taqman probes, and protein expression using Western blots. Exercise decreased serum insulin concentrations (7.08 ± 0.79 vs 3.55 ± 0.42 mU/l, P<0.01, 1-way ANOVA, relevant assumptions were verified) and increased plasma FFA levels (0.27 ± 0.05 vs 0.69 ± 0.11 mmol/l, P<0.001). Neither exercise nor insulin affected the mRNA levels of GLUT4 and the diabetes-linked gene calpain-10. However, at the end of exercise, skeletal muscle PDK4 mRNA content was 1.5-fold lower in the exercise leg than the control leg (1.39 ± 0.25 vs 2.14 ± 0.36, P<0.05, 2-way ANOVA). Surprisingly, on day 2, muscle PDK4 mRNA level decreased by 2.2-fold (0.97 ± 0.17 vs 2.14 ± 0.36, P<0.05) in the control leg and was no different from the exercise leg before and after the insulin clamp. On the other hand, prior exercise increased PDK2 mRNA content in the exercise leg only (0.93 ± 0.05 vs 0.74 ± 0.04, P<0.01). However, these exercise-induced changes in PDK2 and PDK4 transcripts did not lead to changes in their mitochondrial protein expression. These results show that a single bout of exercise is associated with significant changes in key genes regulating glucose oxidation but not transport. Furthermore, the exercise-induced downregulation of PDK4 gene expression in both exercised and non-exercised skeletal muscle raises the possibility that an ‘extramuscular’ factor associated with improvement in insulin sensitivity might be involved in the regulation of PDK4 gene expression in human skeletal muscle.