Article Text

Download PDFPDF

Epidemiology of the mitochondrial DNA 8344A>G mutation for the myoclonus epilepsy and ragged red fibres (MERRF) syndrome
  1. A M Remes1,2,3,
  2. M Kärppä1,2,3,
  3. H Rusanen1,2,3,
  4. K Majamaa1,2,3,
  5. I E Hassinen2,
  6. J S Moilanen2,3,
  7. S Uimonen4,
  8. M Sorri4,
  9. P I Salmela5,
  10. S-L Karvonen6,
  11. S-L Karvonen7
  1. 1Department of Neurology, University of Oulu, Oulu, Finland
  2. 2Department of Medical Biochemistry and Molecular Biology, University of Oulu
  3. 3Biocentre, University of Oulu
  4. 4Department of Otorhinolaryngology, University of Oulu
  5. 5Department of Internal Medicine, University of Oulu
  6. 6Department of Dermatology, University of Oulu
  7. 7Department of Dermatology, University of Helsinki, Helsinki, Finland
  1. Correspondence to:
 Professor K Majamaa, Department of Neurology, University of Oulu, PO Box 5000, FIN-90014 Oulu, Finland;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

The myoclonus epilepsy and ragged red fibres (MERRF) syndrome is a maternally inherited progressive mitochondrial encephalomyopathy caused by a 8344A>G mutation in the MTTK gene that encodes mitochondrial tRNA for lysine. Its common clinical features include myoclonic and tonic-clonic seizures, ataxia, and myopathy, but other features have also been reported, including lipoma, diabetes mellitus, optic atrophy, peripheral neuropathy, hearing loss, and dementia.1

The population frequencies of pathogenic mutations in mitochondrial DNA (mtDNA) are not well known, but the Finnish healthcare organisation provides good opportunities to carry out studies on molecular epidemiology. We have previously determined the frequency of 3243A>G, the most common cause of the MELAS syndrome (mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes), to be 16/100 000 in the adult population of Northern Ostrobothnia.2 We report here on the identification of patient groups with common clinical features of the MERRF syndrome, in a comparable population and the resulting determination of the prevalence of the 8433A>G mtDNA mutation.


The prevalence area considered here is the province of Northern Ostrobothnia in northern Finland, with a total population of 353 895 on 31 December 1994 (prevalence date), including 245 201 persons ≥20 years of age. Adult patients with diagnoses that are commonly associated with the 8344A>G mutation1 were identified as being at risk with respect to mitochondrial disorders, and we therefore screened the population for patients ≥20 years of age who had disorders such as ataxia, diabetes mellitus, epilepsy, lipoma, myopathy, ophthalmoplegia, optic atrophy, peripheral neuropathy, and sensorineural hearing impairment (table 1). These were ascertained as described in detail previously.2 The research protocol was approved by the ethics committee of the Medical Faculty of the University of Oulu, Finland, and the Finnish Ministry of Social Affairs and Health.

Table 1

Criteria used in the screening of the patient groups

DNA from blood samples was purified using the QIAamp Blood Kit (QIAGEN) and a fragment encompassing the MTTK gene was amplified by PCR in the presence of 35S-dATP. The 8344A>G mutation was detected by restriction fragment analysis using BglI. After digestion, the samples were electrophoresed through a 6% acrylamide gel, which was dried and autoradiographed at −72°C overnight using Kodak XAR-5 film with an intensifying screen. Amplified DNA from a subject known to harbour the mutation was included in each restriction digestion and electrophoresis. The degree of mutant heteroplasmy in this sample was 59%.


We identified 818 patients with signs or symptoms that have been associated with MERRF (table 1), and samples obtained from 621 of these were examined for the 8344A>G mutation. None of the patients harboured the mutation (95% confidence intervals (CI) 0 to 3.67). The prevalence of 8344A>G in the adult population of Northern Ostrobothnia was thus calculated to be 0–1.5/100 000. The estimated frequency of 8344A>G in northern Finland is much lower than that of 3243A>G, but comparable to that found in two previous studies: 0.25/100 000 (95% CI 0.01 to 0.50) among adult patients in a single neurology centre in United Kingdom over a 10 year period,3 and 0 to 0.25/100 000 (95% CI) in a population based study among children in western Sweden.4 The 8344A>G mutation is not absent in Sweden or Finland, however, as the authors are aware of two families in southern Finland who possess it, and a few such families have been reported in Sweden.4

The frequency of 3243A>G has been found to be four times that of 8344A>G in the United Kingdom.3 Furthermore, gene analyses in a molecular diagnostic laboratory have revealed that the ratio of these two mutations among 2000 patients with features of mitochondrial disorders is 4,5 suggesting that the frequency ratio between the two is fairly constant. The 3243A>G MELAS mutation appears to be clearly more common than 8344A>G also among Finnish patients that was ascertained in a population based manner.

MtDNA mutations are a comparatively common cause of neurometabolic disorders in both adults and children, but they vary in prevalence. The most common mtDNA point mutations seem to be 11778G>A, 3243A>G and 3460G>A, while 8344A>G is infrequent. The 3243A>G mutation has arisen several times in a population2 and is not faced with any strong selection pressure,6 but the low frequency of 8344A>G suggests either that this gene is not a hot spot for mutational events, or that the mutation is rapidly eliminated in a population. Indeed, the two mutations lead to different biochemical consequences at the cellular level. The MERRF mutation impairs mitochondrial translation more severely than does the MELAS mutation.7 Evolutionarily, these two mutations may therefore be faced with different negative selection and may explain the differences in population frequencies.


The authors thank Ms Anja Heikkinen for her expert technical assistance. This study was supported by grants from the Medical Research Council of the Academy of Finland and the Sigrid Juselius Foundation.