Are exercise-induced genes induced by exercise?

Numerous human in vivo studies on skeletal muscle gene expression have investigated the effects of given interventions. These have been founded on the assumption that presampling can be regarded as a representative control for postintervention sampling. However, many genes are responsive to the metabolic status, which varies during the day, so that observed differences in gene expression between the pre- and post-sample may therefore be a result of the daily variations rather than an intervention. Furthermore, the sampling itself can cause a local stress response, which may also influence the expression of some genes in later samples from the same localized area. To test this, we performed a short-term human endurance exercise study in which muscle biopsies were obtained from healthy untrained individuals (n = 14) before and in the hours after exercise to measure the expression of mRNA for previously reported exercise-related genes (e.g., PPARgamma coactivator-1alpha (PGC-1alpha), pyruvate dehydrogenase kinase 4 (PDK4), MyoD, p21, (heat shock protein 72 (HSP72), lipoprotein lipase (LPL), citrate synthase (CS), and glucose transporter 4 (GLUT4)). To test for changes unrelated to exercise, one half of the subjects did not exercise. As suspected, several presumed exercise-induced genes were induced even without the exercise. Our data demonstrate that presampling is not always a representative control for postintervention sampling, illustrating that use of presampling can cause erroneous interpretations of the underlying induction signals.