Background: Mitochondrial disorders are generally not associated with a clear phenotype-genotype relationship, which complicates the understanding of the disease and genetic counseling.
Objective: To investigate the relationship between the muscle and blood mitochondrial DNA mutation load and phenotype.
Design: Survey.
Setting: The Neuromuscular Research Unit, Rigshospitalet, Copenhagen, Denmark.
Participants: Fifty-one persons with the 3243A>G point mutation of mitochondrial DNA, and 20 healthy control subjects.
Methods: We recorded the maximal oxygen uptake (Vo(2)max), maximal workload, resting and peak-exercise plasma lactate levels, muscle and blood mutation load, muscle morphology, and presence of diabetes mellitus and hearing impairment in all subjects.
Results: Muscle mutation load (mean +/- SE, 50% +/- 5%; range, 2%-95%) correlated with Vo(2)max and resting plasma lactate level (P<.001; R>/=0.64). All persons except 5 with a muscle mutation load above 50% had abnormal Vo(2)max and morphology on muscle biopsy findings. Persons with hearing impairment and diabetes mellitus had a muscle mutation load above 65%. The mutation load in blood (mean +/- SE, 18% +/- 3%; range, 0%-61%) did not correlate with Vo(2)max, resting plasma lactate levels, or presence of hearing impairment or diabetes mellitus.
Conclusions: This study demonstrates a close relationship between the muscle mutation load and phenotype in persons carrying the 3243A>G mutation. The lack of correlation between the mutation load in blood and symptoms from other tissues emphasizes the importance of assessing phenotype-genotype correlations in the same tissue in mitochondrial disease. The results indicate that the threshold of muscle mutation load at which oxidative impairment occurs can be as low as 50%, which is as much as 40% lower than that suggested by in vitro studies.