Mitochondrial DNA (mtDNA) replication is thought to be an integral part of exercise-training-induced mitochondrial adaptations. Thus, mtDNA level is often used as an index of mitochondrial adaptations in training studies. We investigated the hypothesis that endurance exercise training-induced mitochondrial enzymatic changes are independent of genomic dosage by studying mtDNA content in skeletal muscle in response to six weeks of knee-extensor exercise training followed by four weeks of deconditioning in one leg, comparing results to the contralateral untrained leg, in 10 healthy, untrained male volunteers. Findings were compared to citrate synthase activity, mitochondrial complex activities, and content of mitochondrial membrane markers (porin and cardiolipin). One-legged knee-extensor exercise increased endurance performance by 120%, which was accompanied by increases in power output and peak oxygen uptake of 49% and 33%, respectively (p < 0.01). Citrate synthase and mitochondrial respiratory chain complex I⁻IV activities were increased by 51% and 46⁻61%, respectively, in the trained leg (p < 0.001). Despite a substantial training-induced increase in mitochondrial activity of TCA and ETC enzymes, there was no change in mtDNA and mitochondrial inner and outer membrane markers (i.e. cardiolipin and porin). Conversely, deconditioning reduced endurance capacity by 41%, muscle citrate synthase activity by 32%, and mitochondrial complex I⁻IV activities by 29⁻36% (p < 0.05), without any change in mtDNA and porin and cardiolipin content in the previously trained leg. The findings demonstrate that the adaptations in mitochondrial enzymatic activity after aerobic endurance exercise training and the opposite effects of deconditioning are independent of changes in the number of mitochondrial genomes, and likely relate to changes in the rate of transcription of mtDNA.