Article Text

This article has a correction. Please see:

PDF

No association of the mitochondrial DNA A12308G polymorphism with increased risk of stroke in patients with the A3243G mutation
  1. M Deschauer1,2,
  2. P F Chinnery1,
  3. A M Schaefer1,
  4. D M Turnbull1,
  5. R W Taylor1,
  6. S Zierz2,
  7. S Shanske3,
  8. S DiMauro3,
  9. K Majamaa4,
  10. E Wilichowski5,
  11. D R Thorburn6
  1. 1School of Neurology, Neurobiology and Psychiatry, The Medical School, University of Newcastle upon Tyne, UK
  2. 2Department of Neurology, Martin-Luther-University Halle-Wittenberg, Halle/Saale, Germany
  3. 3Department of Neurology, Columbia University College of Physicians and Surgeons, New York, USA
  4. 4Department of Neurology, University of Oulu, Finland
  5. 5Department of Paediatrics and Neuropaediatrics, Georg-August University Göttingen, Germany
  6. 6Murdoch Childrens Research Institute, Royal Children’s Hospital, Melbourne, Australia
  1. Correspondence to:
 Dr R W Taylor
 School of Neurology, Neurobiology and Psychiatry, The Medical School, Framlington Place, University of Newcastle upon Tyne, Newcastle upon Tyne NE2 4HH, UK; r.w.taylorncl.ac.uk

Statistics from Altmetric.com

There is a striking phenotypic variability among patients with the A3243G tRNALeu(UUR) gene mutation of mitochondrial DNA (mtDNA), the most common heteroplasmic mtDNA mutation. It is responsible for ∼80% of cases of MELAS (mitochondrial encephalomyopathy with lactic acidosis and stroke-like episodes), and is also associated with several other phenotypes including maternally inherited diabetes and deafness (MIDD) and chronic progressive external ophthalmoplegia (CPEO).1

Only 50% of patients carrying the A3243G mutation have stroke-like episodes1,2 and the reason for this clinical variability remains poorly understood. Although the percentage level of the A3243G mutation in clinically relevant tissues appears to be important, this relationship is far from clear.1 High percentage levels of the A3243G mutation in muscle are associated with stroke-like episodes, but approximately one in five patients harbouring >80% A3243G in muscle remain stroke free,1 suggesting that additional environmental and genetic factors may influence the phenotypic expression of this mutation.

One possibility is that background mtDNA sequence variation influences phenotype. There is a well-recognised association between the mtDNA genetic background (or haplogroup) and the risk of developing visual failure in another mtDNA disorder, Leber’s hereditary optic neuropathy,3 and a similar mechanism may influence the incidence of stroke-like episodes in patients harbouring the A3243G mutation. Intrafamilial clustering of clinical phenotypes in A3243G patients would indirectly support a role for the mtDNA background, though our own clinical experience suggests that there is significant clinical variability between families.

Pulkes et al have previously reported an increased risk of stroke associated with the presence of a homoplasmic, polymorphic (A12308G) variant in 48 patients with the A3243G mutation.2 The A12308G polymorphism, which is located in the second mitochondrial tRNA gene encoding leucine (tRNALeu(CUN)), occurs with a frequency of 21% in a population of European origin and defines the mtDNA super-haplogroup U/K together with two other polymorphisms (A11467G and G12372A). As haplogroup U has also been reported to be a risk factor for sporadic occipital stroke in patients with migraine,4 these observations could have profound implications for our understanding of mitochondrial genotype and its relationship to the clinical phenotype. Here we report on the investigation of the A12308G polymorphism in a larger group of well-characterised, unrelated A3243G index cases.

Methods

We carried out a large, multicentre study to investigate the A12308G polymorphism in a group of 107 unrelated family index cases harbouring the A3243G mutation. The patients (>95% Caucasian) were from England, Germany, USA, Australia, and Finland. All presented to a neurology clinic, where stroke-like episodes were diagnosed clinically by experienced neurologists based upon a characteristic clinical history and brain imaging with computed tomography (CT) or magnetic resonance imaging (MRI)5; in all cases, a molecular diagnosis of the A3243G mutation was made at a centre with expertise in the investigation of patients with mtDNA disorders.

To investigate the A12308G and G12372A polymorphisms, a 249 bp fragment spanning this mtDNA region was polymerase chain reaction (PCR)-amplified using a forward primer (5′ GATTGTGAATCTGACAACAGAGGCTT 3′; nt 12164–12189) and a reverse primer (5′ GGTTAACGAGGGTGGTAAGGATG 3′; nt 12412–12390). Amplified products were purified and sequenced using BigDye terminator cycle sequencing chemistries on an ABI 377 automated DNA sequence (Applied Biosystems, Warrington, UK).

Results

The A12308G polymorphism was present in 32 of the 107 patients, while 56 had a history of stroke-like episodes. Nine of the 56 patients with a history of a stroke and 23 of 51 patients without stroke were shown to harbour the A12308G polymorphism. Every patient with the A12308G polymorphism also harboured the G12372A variant, indicating that they belong to the same mtDNA super-haplogroup U/K.

As shown in fig 1, our study alone revealed an apparent negative association between stroke-like episodes and the A12308G polymorphism, an observation in direct contrast to the positive association found by Pulkes et al.2 Meta-analysis of all available data however, including the present study (n = 107) and the published study of Pulkes et al2 (n = 48), revealed that 16 of the 77 patients with a history of a stroke and 25 of 78 patients without stroke harboured the A12308G polymorphism. This did not show a statistically significant association between the A12308G polymorphism and stroke-like episodes (χ2 = 2.53, p = 0.112).

Figure 1

 Meta-analysis showing odds ratios (OR) and 95% confidence intervals (CI) for the original study by Pulkes et al,2 the data generated by this study, and the combined dataset including both studies. The squares represent the OR, with the size proportional to the study size. The horizontal lines show the CI for the OR. The diamond shows the OR for the combined dataset with CI that overlap 1, indicating a non-significant result.

Discussion

The aim of our study was to examine whether a previously described association between the A12308G polymorphism and an increased risk of stroke in patients with the A3243G mutation2 was reflected in a larger study group. In agreement with previous reports, 52% of our patients experienced stroke-like episodes1,2 and 30% harboured the A12308G polymorphism, confirming that our cohort of 107 A3243G index cases formed a representative sample. Despite studying a larger cohort of patients, we were not able to confirm the positive association between the A12308G polymorphism and an increased risk of stroke in patients with the A3243G mutation as reported previously.2 Meta-analysis of all the available data failed to prove any clear association between the A12308G polymorphism and stroke-like episodes.

The clinical diversity associated with the A3243G mutation clearly involves multiple factors. We have previously shown a correlation between clinical phenotype and mutation load in muscle.1 Age may well be a contributing factor, although there was a tendency for patients with stroke-like episodes in our group to be younger than those without. This argues against age as a risk factor for stroke-like episodes, as seen in common stroke.

Importantly our findings serve to highlight the difficulty of performing association studies on small numbers of patients. This is particularly difficult for mitochondrial genetic association studies because of the high variability of the mitochondrial genome. Understanding the phenotypic differences between patients with specific, pathogenic mtDNA mutations will ultimately involve studies of large cohorts of patients, unless we are able to gain clues from experimental studies that may highlight factors involved in the altered expression or segregation of mtDNA mutations.

Acknowledgments

The authors thank Geoff Taylor (University of Newcastle upon Tyne) for help with the sequencing. This study was supported by a fellowship from the European Neurological Society (MD), the Newcastle upon Tyne Hospitals NHS Trust (DMT and RWT) and the Wellcome Trust (PFC, DMT and RWT). DRT is supported by the Muscular Dystrophy Association and the Australian NHMRC.

References

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.

Linked Articles

  • Correction
    BMJ Publishing Group Ltd