www.thelancet.com/neurology Vol 9 August 2010 829 Review Lancet Neurol 2010; 9: 829–40 Mitochondrial Research Group, Institute for Ageing and Health, Newcastle University, Newcastle upon Tyne, UK (R McFarland MRCPCH, Prof R W Taylor PhD, Prof D M Turnbull FRCP) Correspondence to: Prof Douglass M Turnbull Mitochondrial Research Group, Institute for Ageing and Health, Newcastle University, Newcastle upon Tyne NE2 4HH, UK d.m.turnbull@ncl.ac.uk A neurological perspective on mitochondrial disease Robert McFarland, Robert W Taylor, Douglass M Turnbull Disruption of the most fundamental cellular energy process, the mitochondrial respiratory chain, results in a diverse and variable group of multisystem disorders known collectively as mitochondrial disease. The frequent involvement of the brain, nerves, and muscles, often in the same patient, places neurologists at the forefront of the interesting and challenging process of diagnosing and caring for these patients. Mitochondrial diseases are among the most frequently inherited neurological disorders, and can be caused by mutations in mitochondrial or nuclear DNA. Substantial progress has been made over the past decade in understanding the genetic basis of these disorders, with important implications for the general neurologist in terms of the diagnosis, investigation, and multidisciplinary management of these patients. Introduction Over 20 years ago, Holt and colleagues 1 reported the first association between a defect in mitochondrial DNA and human disease, and this was quickly followed later the same year by a second report from Wallace and colleagues. 2 Since then, the number of disease-associated mitochondrial DNA mutations has expanded rapidly and mutations have been identified that cause classic mitochondrial syndromes such as mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes (MELAS), myoclonic epilepsy and ragged red fibres (MERRF), neuropathy, ataxia, and retinitis pigmentosa (NARP), Kearns-Sayre syndrome, and maternally inherited Leigh syndrome. 3 The importance of nuclear genetic mutations in causing mitochondrial dysfunction and human disease has become increasingly clear over the past decade. Indeed, the putative role of mitochondrial respiratory chain deficiency in the pathogenesis of a wide range of neurological disorders has been the subject of intense scientific scrutiny. 4–7 In this Review, we discuss mitochondrial respiratory chain disease in the context of clinical neurology, providing an overview of not only the neurological aspects but also the multisystem effects of mitochondrial disease that neurologists must consider in both children and adults with this disorder. The genetic aetiology of mitochondrial disease has a substantial effect on the investigations requested by treating physicians and the type of counselling that is provided to families. We have therefore discussed the underlying genetics in depth, along with diagnostic and management strategies. Mitochondria and mitochondrial genetics Mitochondria undertake many vital metabolic functions, probably the most important of which is oxidative phosphorylation, the principal method for generating ATP. 3 This process is dependent on five intramembrane complexes and two mobile electron carriers (coenzyme Q 10 and cytochrome c), which transport electrons between them. Supercomplexes (ie, respirasomes) are combinations of two or more respiratory chain complexes that can further enhance electron transfer. Although their role in the in-vivo action of the human mitochondrial respiratory chain remains contentious, evidence in favour of a multimeric organisation is accumulating. 8 An interesting legacy of the primeval origins of mitochondria 9 is the persistence of a 16·6 kb, double- stranded circle of DNA (mitochondrial DNA). This semi- autonomous genome encodes 13 structural subunit polypeptides and the machinery (22 transfer RNA molecules and 2 ribosomal RNA molecules) necessary for intramitochondrial protein synthesis. 10 Mitochondrial DNA is present in multiple copies, and in any single cell a small number of these genomes might contain mutations; however, the proportion of mutated DNA is usually so small that for practical purposes the tissue can be regarded as homoplasmic (genetically uniform). By contrast, for several mitochondrial DNA mutations, tissue variation in the level of heteroplasmy (the existence of two or more distinct mitochondrial genomes at high concentrations within the same tissue) has a direct effect on the resultant phenotype and even small decreases in the concentrations of wild-type mitochondrial DNA might be sufficient to cause disease. 11 However, the variable or single organ phenotypes that occur with homoplasmic mutations 12,13 and the apparent dominant nature of some mitochondrial transfer RNA (mitochondrial tRNA) mutations 14 suggest that other, as yet undefined, factors are also important in determining the phenotype. Although nuclear genetic mutations causing mitochondrial dysfunction have been associated with several so-called new clinical phenotypes, some nuclear gene mutations can result in clinical phenotypes that are similar to those in primary mitochondrial DNA disease and the distinction between the two is not always clinically obvious. 15 Prevalence of mitochondrial disease Recent estimates of prevalence suggest that mitochondrial disease is more common than previously thought. Both the mitochondrial tRNA mutations MTTL1, m.3243A>G and MTRNR1, m.1555A>G (aminoglycoside-induced deafness) have frequencies of up to 1 in 400 in the general population, 16–18 but many patients with these mutations remain asymptomatic. Clinical prevalence studies report that mitochondrial disease caused by mutations in mitochondrial DNA affects 9·2 in 100 000 adults aged less than 65 years, and up to 16·5 in 100 000 people aged less than 65 years who have a