Elsevier

Neuromuscular Disorders

Volume 14, Issue 2, February 2004, Pages 162-166
Neuromuscular Disorders

A novel sporadic mutation in cytochrome c oxidase subunit II as a cause of rhabdomyolysis

https://doi.org/10.1016/j.nmd.2003.10.011Get rights and content

Abstract

Disorders of the mitochondrial genome are an important cause of neurological disease, with patients presenting a variety of different phenotypes. Exercise induced muscle pain and myoglobinuria have been described with a number of metabolic defects, but because of the enormous variability of the mitochondrial genome identifying causative mitochondrial DNA mutations can be extremely difficult. Since mitochondrial tRNA genes were considered to be hot spots for mutation, sequencing was initially often confined to these genes. In a patient with symptoms and signs of exercise intolerance and myoglobinuria we originally ascribed pathogenicity to a mitochondrial-tRNAPhe mutation but here we show that the true pathogenic mutation was a novel mutation in the gene encoding subunit II of cytochrome c oxidase. We believe that this study demonstrates the importance of whole mitochondrial genome sequencing and of access to large sequence databases.

Introduction

Mitochondrial tRNA (mt-tRNA) genes appear to be particularly vulnerable to mutation and a large number of substitutions have now been described in almost all of the mt-tRNA genes [1]. Some mutations such as the 3243A>G (MELAS) [2] in mt-tRNALeu[UUR] and the 8334A>G (MERRF) [3] in mt-tRNALys have been described on many occasions and in different ethnic groups, indicating a relatively high incidence of disease. However, little is known about the frequency of other mt-tRNA mutations within the general population, most having been reported on not more than one or two occasions. This low reporting incidence is especially problematic when deciding whether a novel change is responsible for disease, particularly if the index family is small. Limited opportunity for correlation with clinical phenotype, restricted access to clinically relevant tissue and the spontaneous nature of many of these mutations confound attempts to define particular mutations as pathogenic.

In order to circumvent some of these difficulties, three cardinal criteria have been adopted for assigning pathogenicity to a suspected mitochondrial DNA (mtDNA) mutation. Firstly, the absence of the mutation from a large population of control genome sequences. This has been made easier by the sequencing and publication of a large number of genomes collected from a variety of sources, for example the University of Uppsala (http://www.genpat.uu.se/mtDB/index.html) and the Mitokor [4] databases, and has resolved problems such as reporting and ascertainment bias which have previously hampered local databases. However, the disease status of the individuals contributing to these large databases is an important consideration and often remains unknown. Secondly, the mutation must disrupt a functionally and phylogenetically conserved base pairing. While this is a useful consideration, the hallmark of the mitochondrial genome is divergence and clinically benign polymorphic variations do occur at phylogenetically conserved sites [5]. Finally, in a system where both wild type and mutated mtDNA co-exist (heteroplasmy), the proportion of mutated mtDNA must be directly correlated with the presence or absence of disease, and there should be a tissue-specific threshold for expression of a mitochondrial abnormality. Homoplasmic mutations (100% mutated mtDNA) are an obvious exception where other nuclear and tissue specific factors must be important in determining their pathogenicity [6], [7], [8].

Although not flawless, satisfying these criteria has generally been accepted as a reliable means of assigning pathogenicity to novel mtDNA mutations. In a previous report, at a time when we limited diagnostic sequencing to genomic ‘hotspots’, these criteria were utilised to conclude that a novel mt-tRNAPhe mutation was pathogenic [9]. Subsequent analysis of the remainder of this patient's mitochondrial genome has, however, revealed a previously unreported mutation (7989T>C) in the COII gene encoding subunit II of cytochrome c oxidase (COX). This COII mutation is more compatible with the clinical, biochemical and histochemical findings and has forced us to re-examine the process of assigning pathogenicity to the 606A>G mutation in mt-tRNAPhe.

Section snippets

Patient

Previously described in 1997, this patient first presented at 30 years old with acute onset of muscle pain and weakness 3 h after strenuous lifting [9]. Since then he has developed other symptoms including recurrent episodes of myoglobinuria, fatigue, weakness, bowel dysmotility and paroxysmal orthostatic tachycardia syndrome.

mtDNA sequencing

Total DNA was extracted from tissue samples using standard phenol/chloroform DNA extraction procedures. The entire sequence of the mitochondrial genome was determined

Re-evaluation of the molecular diagnosis

We revisited this patient's molecular diagnosis when we discovered the 606A>G mutation at almost homoplasmic levels in muscle of an unrelated patient with a completely different phenotype. mtDNA sequencing of the patient with myoglobinuria had previously been limited to the mt-tRNA genes and we therefore extended this sequencing to include the entire coding region of his mitochondrial genome.

Sequencing of the coding region of the mitochondrial genome

There are 13 changes from the revised Cambridge reference sequence [11] including 4 neutral

Discussion

Making a diagnosis of mitochondrial disease is rarely straightforward and many factors must be considered before concluding that a defect in mtDNA is responsible. Evidence is gathered from histochemical, biochemical and molecular genetic analyses of tissue (usually muscle), and a diagnosis is then made on the basis of the available information. Particularly for rare or singleton cases, the molecular diagnosis of a mitochondrial disease is often not absolute. The present case is a salutary

Acknowledgements

We thank the Wellcome Trust, the Muscular Dystrophy Campaign and the Newcastle upon Tyne Hospitals NHS Trust for their financial support.

References (18)

There are more references available in the full text version of this article.

Cited by (60)

  • Neuromuscular Manifestations in Mitochondrial Diseases in Children

    2016, Seminars in Pediatric Neurology
    Citation Excerpt :

    The importance of mutations in mtDNA COX genes in causing exercise intolerance and myoglobinuria was confirmed by 3 reports on 4 patients. Three of them were young adults with life-long exercise intolerance, exercise-related myalgia, and at least 1 episode of myoglobinuria; however, all 3 were sporadic cases harboring presumably somatic; heteroplasmic mutations confined to skeletal muscle; and affecting COX I (G6708A),20 COX II (T7989C),21 and COX III (T9789C).22 The fourth patient, who harbored a heteroplasmic mutation (G7970T) in COX II, had a severe encephalomyopathy with short stature, bilateral cataracts, sensorineural hearing loss, myopathy, ataxia, cardiac arrhythmia, and depression.22,24

  • Diagnosis of mitochondrial myopathies

    2013, Molecular Genetics and Metabolism
View all citing articles on Scopus
View full text