Article Text


The frequency of mtDNA 8994 polymorphism and detection of the NARP 8993 mutation
  1. R G F Gray,
  2. P A Davies,
  3. A Marshall,
  4. S K Heath
  1. Department of Clinical Chemistry, Diana, Princess of Wales Children's Hospital, Birmingham B4 6NH, UK
  1. Correspondence to:
 Dr R G F Gray, of Clinical Chemistry, Diana, Princess of Wales Children's Hospital, Birmingham B4 6NH, UK;

Statistics from

The highly polymorphic nature of the mitochondrial genome (mtDNA) has proved valuable to the population geneticist, but can cause serious problems in the identification of disease causing mutations.

A T→C or T→G transition at nt 8993 in human mtDNA is associated with an array of clinical phenotypes including Neuropathy, Ataxia and Retinitis Pigmentosa (NARP)1 and Leigh's syndrome.2 Conventionally, it is detected by polymerase chain reaction (PCR) amplification of the region containing the mutant sequence followed by digestion with restriction enzymes HapII or HpaII (recognition sequence c↓cgg) which recognise both sequence changes.1 The presence of either mutation results in the PCR product being cut into two fragments (343 bp and 206 bp), which can be separated and identified by agarose gel electrophoresis.

A polymorphic G→A transition in the adjacent base (nt 8994) abolishes the recognition site for the relevant restriction enzymes and, hence, in patients who also have the NARP 8993 mutation can result in a false negative diagnosis.

As a consequence, a method has been developed to detect the 8994 polymorphism using the restriction enzyme HaeIII applied to the same undigested PCR product used for detecting the 8993 mutation, which is in use by some laboratories who test for the 8993 mutation.3 Patients without the polymorphism have three HaeIII restriction sites in the amplified region resulting in four fragments (190 bp, 174 bp, 156 bp, 31 bp), while those with the polymorphism have one of these sites abolished giving rise to three fragments (190 bp, 187 bp, 174 bp). Under the gel electrophoresis conditions used (2% agarose with ethidium bromide staining), the 190 bp and 187 bp fragments run as a single band and the 31 bp fragment is usually not visible as it runs off the gel. Thus, in the presence of the 8994 polymorphism two fragments are seen, while in its absence three fragments are detected.

To exclude the NARP 8993 mutation, we routinely test all patients for both the 8994 and 8993 mutations. If the 8994 polymorphism is detected, the PCR product is sequenced to exclude the 8993 mutation.

The 8994 polymorphism has been reported in mtDNA haplogroup W, one of the rarer haplogroups specific to people of European origin.4 However, the polymorphism may also exist on its own in other ethnic groups. We have analysed muscle and blood samples for the 8993 mutation and the 8994 polymorphism in 547 patients from the West Midlands over a five year period. Analyses were performed using blood cell lysates and DNA extracted from muscle biopsies.

Six patients (1.1%) did not show the normal pattern of fragment sizes on HaeIII digestion for the 8994 polymorphism. On sequencing, four (0.7%) showed the 8994 polymorphism for which all were homoplasmic. One of these was of European origin, while three were from the Indian subcontinent or of Middle Eastern origin. The remaining two patients showed a different abnormal digestion pattern with HaeIII. On sequencing this was found to be the result of a G→A change at nucleotide 9025. This change has previously been reported as a polymorphism in the MITOMAP5 database. While this polymorphism resulted in an atypical restriction pattern, it would not have caused a misdiagnosis for the 8993 mutation.

In our study, the frequency of the 8994 polymorphism observed in the West Midlands population was very low (0.7%). Over a five year period we have diagnosed one family group with the NARP 8993 mutation in 547 samples from adults and children. Hence, the probability of a false negative diagnosis if a test for the 8994 polymorphism was not performed is 0.0013%. Despite the low frequency of the 8994 polymorphism in our population, it will remain necessary for us to use the restriction digestion method to detect this sequence change in order to exclude the 8993 mutation. This illustrates the problem of restriction digestion based methods for the detection of mutations in the highly polymorphic mitochondrial genome. A method using direct sequencing or a primer extension assay would obviate this problem and could be automated for high throughput diagnostic screening.


View Abstract

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.