- Email: [email protected]
A novel mitochondrial tRNAGlu (MTTE) gene mutation causing chronic progressive external ophthalmoplegia at low levels of heteroplasmy in muscle
Journal of the Neurological Sciences 298 (2010) 140–144
Contents lists available at ScienceDirect
Journal of the Neurological Sciences j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n s
Short communication
A novel mitochondrial tRNAGlu (MTTE) gene mutation causing chronic progressive external ophthalmoplegia at low levels of heteroplasmy in muscle Charlotte L. Alston a, James Lowe b, Douglass M. Turnbull a, Paul Maddison c, Robert W. Taylor a,⁎ a b c
Mitochondrial Research Group, Institute for Ageing and Health, The Medical School, Newcastle University, Framlington Place, Newcastle upon Tyne, NE2 4HH, UK School of Molecular Medical Sciences, University of Nottingham, Nottingham, UK Department of Clinical Neurology, Nottingham University Hospitals NHS Trust, Nottingham, UK
a r t i c l e
i n f o
Article history: Received 9 July 2010 Received in revised form 30 July 2010 Accepted 6 August 2010 Keywords: Mitochondrial DNA Chronic progressive external ophthalmoplegia tRNAGlu Single fibre studies Pathogenicity
a b s t r a c t Mitochondrial respiratory chain defects are associated with diverse clinical phenotypes in both adults and children, and may be caused by mutations in either nuclear or mitochondrial DNA (mtDNA). We report the molecular genetic investigations of a patient with chronic progressive external ophthalmoplegia (CPEO) and myopathy where muscle biopsies taken 11 years apart revealed a progressive increase in the proportion of cytochrome c oxidase (COX)-deficient fibres. Mitochondrial genetic analysis of the early biopsy had seemingly excluded both mtDNA rearrangements and mtDNA point mutations. Sequencing mtDNA from individual COX-deficient muscle fibres in the second biopsy, however, identified an unreported m.14723 T N C substitution within the mitochondrial tRNAGlu (MTTE) gene, which fulfilled all canonical criteria for pathogenicity. The m.14723 T N C mutation was absent from several tissues, including muscle, from maternal relatives suggesting a de novo event, whilst quantitative analysis of the first muscle biopsy confirmed a very low level of the mutation (7% mutated mtDNA), highlighting a potential problem whereby pathogenic mtDNA mutations may remain undetected using established screening methodologies. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Defects of the mitochondrial respiratory chain are associated with a diverse and ever-growing spectrum of clinical phenotypes in both adults and children, and may be caused by mutations in either the nuclear or mitochondrial genome (mtDNA). Where mitochondrial disease is due to mtDNA mutations, those occurring within mitochondrial (mt-) tRNA genes are the most common, with over 150 discrete pathogenic mutations identified across each of the 22 mt-tRNA genes [1]. They are associated with a number of distinctive clinical presentations ranging from the multi-organ, syndromic phenotypes of MERRF and MELAS, to isolated symptoms such as CPEO, diabetes or myopathy. In spite of the clinical and genetic heterogeneity associated with mt-tRNA mutations, determinants of pathogenicity are typically shared across the spectrum [2–4]. Pathogenic mutations are typically heteroplasmic, a term that describes the presence of both wildtype and mutant mtDNA molecules in a cell. They are present at high levels of heteroplasmy in post-mitotic tissues, such as muscle; are often lost from rapidly dividing cells, such as blood leucocytes [5] and are typically associated with a mosaic pattern of cytochrome c oxidase (COX) deficiency. Their presence at high levels of heteroplasmy in
⁎ Corresponding author. Tel.: + 44 191 2223685; fax: + 44 191 2824373. E-mail address: [email protected] (R.W. Taylor). 0022-510X/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2010.08.014
clinically-relevant tissues facilitates molecular diagnosis by direct sequencing of the whole mitochondrial genome. We present a patient with chronic progressive external ophthalmoplegia (CPEO) and COX-deficient muscle fibres associated with a novel m.14723 T N C mt-tRNAGlu mutation which was present at low levels in the first of two muscle biopsies, thereby evading detection by established sequencing methodologies. The m.14723 T N C mutation clearly segregated with a functional, biochemical defect as measured by COX activity in individual muscle fibres, and the absence of this mutation in muscle and several mitotic tissues from the patient's daughter indicates that it is not transmitted through the female germline. 2. Case report The index patient is a 63-year-old female who initially developed bilateral eyelid droop in early childhood, which was evident in photographs taken at the age of five years. She underwent corrective squint surgery at the age of seven years, and has had slight amblyopia in the left eye, having never suffered from significant double vision. She underwent further ophthalmological review at the age of 38 for worsening bilateral ptosis, subsequently requiring surgical correction on two separate occasions. By the age of 39, she had begun to develop neck and proximal upper limb ache and on neurological review at the age of 51 years, she had subtle neck and proximal upper limb weakness. Needle electromyography revealed myopathic features whilst CK results were normal. Histocytochemical analysis of a needle
C.L. Alston et al. / Journal of the Neurological Sciences 298 (2010) 140–144
quadriceps muscle biopsy taken in 1997 reported low levels (1–2%) of COX-deficient fibres (Fig. 1A). A repeat biopsy was performed 11 years later, in 2008, to aid genetic diagnosis. At present, neurological examination reveals significant ophthalmoplegia with exodeviation in primary gaze, without double vision. She has bilateral ptosis, facial weakness, and mild proximal limb weakness, with no other neurological abnormalities. Repeated ECG findings have been normal. Family history is somewhat unremarkable; her deceased parents were first cousins and her sister has multiple sclerosis, although no evidence of ophthalmoplegia. The patient's 39 year old, younger daughter is currently well, with no abnormal neurological symptoms or signs. Her older, 41 year old daughter has not been examined, but evidence of ophthalmoplegia has not been reported. 3. Materials and methods 3.1. Muscle histology and histochemistry Standard histological and histochemical analyses of quadriceps muscle biopsy were performed on fresh frozen sections (10 μm). Cytochrome c oxidase (COX) staining was carried out on the patient's initial muscle biopsy whilst investigations on the subsequent patient biopsy included the determination of sequential COX/succinate dehydrogenase (SDH) activities using standard methods [6]. 3.2. Molecular genetic studies Total DNA was extracted from several tissues including blood, buccal and urinary epithelium and skeletal muscle using standard
141
procedures. Blood, urine and buccal samples were obtained with consent from clinically-unaffected relatives (patient's sister and patient's daughter), whilst patient anxiety prompted her daughter to have a muscle biopsy. Mitochondrial DNA rearrangements were investigated in muscle DNA using established long-range PCR protocols [7], followed by direct sequencing of the entire mitochondrial genome, performed either in muscle homogenate DNA or DNA from individual COX-deficient muscle fibres isolated by lasermicrocapture, essentially as described elsewhere [8,9].
3.3. Assessment of m.14723 T N C mutation load by PCR-RFLP analysis To determine the level of heteroplasmy of the novel m.14723 T N C variant, we designed a PCR-restriction fragment length polymorphism (RFLP) assay as follows: a 211 bp PCR product spanning the mutation site was amplified using a forward mismatch primer 5′-AGAATAATAACACACCCGACGACAC-3′ (nt 14543–14567; mismatch nucleotide shown in bold ) and a reverse primer (nt 14753–14733), GenBank Accession number NC_012920 [10]; the mismatch nucleotide creates a Hpy99I restriction site, providing an internal control for digestion. The m.14723 T N C mutation creates a further Hpy99I restriction site, permitting the detection of mutant mtDNA amplicons which cut twice to generate bands of 161, 29 and 21 bp, from wild type mtDNA amplicons which cut only once yielding bands of 190 and 21 bp. Prior to the last cycle of PCR, 5 μCi [α-32P]dCTP (3000 Ci/mmol) was added. Labelled products were precipitated, digested overnight with 10U Hpy99I, separated through a 12% nondenaturing polyacrylamide gel and the radioactivity in each fragment was quantified using ImageQuant software (Amersham Biosciences/GE Healthcare).
Fig. 1. Muscle biopsy and mtDNA analysis. Histochemical analysis of patient muscle biopsy. COX staining highlights both COX-deficient fibres (unstained, marked with asterix) and COX-positive fibres (brown) in (A) the 1997 biopsy and (B) the 2008 biopsy; both biopsy sections have been counterstained with haematoxylin to visualise nuclei resulting in slight staining of adipose material. (C) Sequential COX-SDH analysis reveals numerous COX-deficient fibres on the 2008 biopsy section with mitochondrial aggregates corresponding to ragged-red fibres. (D) mtDNA sequencing chromatogram reveals the m.14723T N C substitution, present at very high levels in a COX-deficient fibre compared to a wild type control sequence (E). Sequencing of patient's 1997 muscle biopsy homogenate DNA sample (F) reveals very low levels of the m.14723 T N C mutation although higher levels were apparent in the 2008 muscle biopsy (G).
142
C.L. Alston et al. / Journal of the Neurological Sciences 298 (2010) 140–144
3.4. Assessment of m.14723 T N C mutation load by quantitative pyrosequencing The detection of low level m.14723 T N C heteroplasmy in the original muscle DNA homogenate prompted confirmation by an alternative method to quantitate mtDNA mutation load. Pyromark Assay Design Software v.2.0 (Qiagen) was used to design locus specific PCR and pyrosequencing primers, which amplified a 90 bp PCR product spanning the m.14723 nucleotide using a forward primer (nt 14688–14711) and a biotinylated reverse primer (nt 14777–14757). Pyrosequencing was achieved on the Pyromark Q24 platform according to the manufacturer's protocol, employing a mutationspecific pyrosequencing primer (nt 14704–14721). Pyromark Q24 software was used to quantify the m.14723 T N C heteroplasmy levels by directly comparing the relevant peak heights of both wild type and mutant mtDNA at this site [11].
4. Results 4.1. Histology and histochemistry COX histochemistry of the initial muscle biopsy taken in 1997 revealed 1–2% COX-deficient fibres (Fig. 1A), which first prompted the investigation of this patient. Analysis of the subsequent biopsy, taken 11 years later, demonstrated a significant increase in the proportion of COX-deficient fibres (7% of total biopsy) (Fig. 1B). Sequential COX-SDH analysis revealed that many of the COX-deficient fibres also exhibited subsarcolemmal mitochondrial aggregates which would correspond to ragged-red fibres (Fig. 1C), clearly demonstrating an underlying mitochondrial aetiology.
4.2. Molecular genetics investigations All initial mtDNA genetic testing was performed on muscle DNA which was referred to our laboratory and was extracted from the first (1997) skeletal muscle biopsy. Long-range PCR revealed only full length, wild type products with no evidence of any mtDNA rearrangement. Similarly, sequencing of the entire mitochondrial genome in this sample showed no pathogenic mtDNA mutations. Our patient was re-biopsied in 2008, and having excluded mtDNA deletions in this sample, we proceeded to sequence the entire mitochondrial genome from laser microdissected COX-deficient muscle fibres, revealing a novel m.14723 T N C substitution within the MTTE gene (Fig. 1D). Levels of m.14723 T N C heteroplasmy in COX-positive, biochemically normal, muscle fibres were significantly lower (data not shown). Reanalysis of old chromatograms suggested that the 1997 muscle homogenate sample exhibited very low levels of the m.14723 T N C mutation (Fig. 1F), with higher levels present in the 2008 biopsy (Fig. 1G).
Fig. 2. Characterisation of a novel m.14723 T N C MTTE substitution. (A) Last hot cycle PCRRFLP analysis confirms a low level of m.14723T N C mutation in the patient’s 1997 muscle biopsy (lane 3) with an increased mutation load in the 2008 muscle biopsy (lane 4). Results indicate an absence of m.14723 T N C in patient blood (lane 5). Control samples are also shown; uncut DNA sample (lane 1), control DNA sample (lane 2). (B) Pyrograms illustrating (i) low level of m.14723TN C heteroplasmy in the 1997 muscle biopsy; (ii) higher heteroplasmy levels in the 2008 patient biopsy and (iii) absence of m.14723 T N C in a muscle biopsy from the patient's clinically-unaffected daughter. (C) Graphical representation of single fibre PCR analysis clearly shows a marked segregation of the novel m.14723 T N C mutation with a biochemical defect in COX-deficient muscle fibres.
5. Discussion 4.3. Quantitative analysis of the m.14723 T N C mutation We quantified the level of m.14723 T N C heteroplasmy at 7% mutant load in the patient's initial biopsy and at 34% heteroplasmy in her later biopsy by last hot cycle PCR-RFLP analysis (Fig. 2A), which was confirmed by Pyrosequencing analysis (Fig. 2B). The m.14723 T N C substitution was absent in all other patient tissues tested including blood, urine and buccal epithelial cells, and was absent in all available tissues from the patient’s daughter – including skeletal muscle biopsy (Fig. 2B) – and from her sister. Single fibre analyses revealed higher levels of m.14723 T N C heteroplasmy in COXdeficient fibres (99.7 ± 0.5% (n = 10)) compared to COX-positive fibres (33.7 ± 27.8% (n = 8)), a statistically significant finding (P b 0.001, two-tailed Student's t test) (Fig. 2C).
We describe a patient who presented with CPEO and muscle biopsy changes suggestive of a mtDNA defect, in whom a novel m.14723 T N C mitochondrial tRNA mutation was identified. This novel substitution was present at an uncharacteristically low level of mtDNA heteroplasmy in an initial muscle biopsy, so low in fact that it was missed by the whole mtDNA genome analysis of muscle homogenate DNA, the established diagnostic screen for the detection of rare or novel mtDNA point mutations. This patient's suggestive clinical presentation of CPEO, ptosis and myopathy – combined with a number of COX-deficient muscle fibres on muscle biopsy – prompted further molecular investigation, and only following mtDNA genome sequencing of isolated, biochemically-deficient muscle fibres was the causative m.14723 T N C mutation identified.
C.L. Alston et al. / Journal of the Neurological Sciences 298 (2010) 140–144
Single fibre analysis of COX-deficient and COX-positive muscle fibres revealed a clear correlation between m.14723 T N C mutation load and the respiratory chain abnormality, corroborating the pathogenicity of this novel mtDNA substitution. Assessment of the m.14723 T N C mutation in the non-invasive tissues (e.g. buccal and urinary cells) from the patient and her maternal relatives confirmed that the mutation was confined to patient muscle, and is therefore likely to have arisen sporadically. Our patient has undergone two muscle biopsies, taken eleven years apart, which demonstrate an increase in the frequency of COXdeficient fibres over time and a concomitant increase in m.14723T N C mutation load between muscle homogenate DNA samples. Following a reanalysis of the original muscle homogenate sequencing chromatograms, the m.14723 T N C substitution was visible but at a very low level of heteroplasmy (7% mutation load) — well below the quoted sensitivity level of the sequencing assay [8,12] thus explaining our failure to identify it during initial sequencing studies (Fig. 1F). There is certainly a risk that at such low levels, novel heteroplasmic mtDNA substitutions may be merely dismissed as sequencing background artefact, or of being masked by this phenomenon completely. Interestingly, sequencing of the repeat (2008) muscle biopsy homogenate clearly identified the m.14723 T N C substitution (Fig. 1G), this time at a higher level of heteroplasmy (34% mutation load), correlating with the increased number of COX-deficient muscle fibres on muscle biopsy; pyrosequencing confirmed results for all samples assayed by PCR-RFLP. Similar cases are reported in the literature where mt-tRNA point mutation levels are shown to have increased over time, often correlating with a deteriorating clinical picture in terms both of histopathological changes on muscle biopsy and symptom severity, as characterised by our index case [13]. The novel m.14723 T N C substitution is not a recognised polymorphic variant [14,15], is situated in the highly conserved dihydrouridine arm of the glutamic acid tRNA (Fig. 3), and is predicted to disrupt the Watson and Crick base pairing. Interestingly, a further published pathogenic mtDNA mutation occurring at the adjacent nucleotide (m.14724G N A) also disrupts the Watson and Crick base pairing, and is situated within the same highly conserved dihydrouridine region. Despite the markedly different clinical presentation (index patient harbouring m.14724G N A had a neonatal presentation dominated by hypotonia, macrocephaly and cerebellar ataxia), the presence of another substitution mutation in the same region of the tRNA adds further weight to our hypothesis that the m.14723 T N C substitution forms the molecular basis of our patient's presentation. The canonical scoring system proposed by McFarland et al. [4], when applied to the m.14723 T N C substitution, deems it to be probably
143
pathogenic, with a score of 11 out of a possible 20 points. The absence of transmitochondrial cybrid or steady state mt-tRNAGlu level data, and the absence of respiratory chain enzyme results, precludes it from the “definitely pathogenic” category. mt-tRNA mutations present at low levels may never reach the threshold for categorisation under “definitely pathogenic” according to the criteria set out by McFarland et al. and perhaps this method for establishing pathogenicity may benefit from some flexibility in the case of dominant-acting or low level mt-tRNA mutations, whereby the weighting of single fibre data is greater than currently ascribed. This is the seventh discrete mutation to be described within the MTTE gene. As commonly observed in other mt-tRNA genes, MTTE mutations are associated with a wide range of clinical phenotypes including exercise intolerance [16], myopathy [17,18], diabetes [19,20], progressive encephalopathy [21,22] and encephalomyopathy [23]. A homoplasmic m.14674 T N C MTTE gene mutation has recently been described as the cause of infantile, reversible COX deficiency myopathy [24]. Despite the phenotypic heterogeneity associated with MTTE mutations, the m.14723 T N C mutation in our patient represents the first mutation in this mt-tRNA gene to be associated with a clinical presentation of CPEO. The presence of an mtDNA mutation that is confined to muscle poses a difficult situation in genetic counselling. The m.14723 T N C mutation does not represent a germline mutation due to the absence of m.14723 T N C in the patient's non-invasive tissues, yet mutationfree oocytes cannot be guaranteed. It is not standard practice for unaffected individuals to undergo muscle biopsy due to the invasive nature of the procedure and time/financial restraints, but the degree of anxiety caused by the uncertain genetic status to the patient's daughter was sufficient to deem it a viable option to fully exclude maternal transmission. In turn, we have been able to show that this novel mt-tRNA mutation has not been transmitted, although whether this is due to the de novo mutational event occurring after the mesodermal differentiation of myoblasts from germ cells, or if it is due to the non-transmissible nature of the mutation is unclear. The mechanisms determining which mt-tRNA mutations are transmitted from mother to child remain elusive [25]. Given the spectrum of devastating clinical phenotypes that can be attributed to mt-tRNA mutations, further investigation will certainly lead to better understanding of their pathogenic nature with an aim to improve counselling of our patients and in particular, may prove beneficial when predicting recurrence risks for future pregnancies. Our index case represents the third incidence of a pathogenic mt-tRNA mutation present at low levels of heteroplasmy in muscle homogenate. Previous reports have described a dominant-
Fig. 3. Conservation and molecular location of the novel mt-tRNA substitution. (A) The m.14723 T N C substitution is located within the highly conserved dihydrouridine arm of the mitochondrial glutamate tRNA molecule; the base pair between the affected m.14723 nucleotide and its Watson and Crick correlate at m.14731 is highly conserved across numerous species (B) indicating that it plays an important role in the tertiary structure and/or function of the tRNA molecule.
144
C.L. Alston et al. / Journal of the Neurological Sciences 298 (2010) 140–144
acting mt-tRNA mutation [1] and – as in our index case – a pathogenic mt-tRNA mutation present at low levels in muscle homogenate [26]. Taken together, these unusual clinical cases highlight the value of investigating isolated COX-deficient fibres, particularly when the usual avenues of investigation have previously been explored and excluded. Whole mtDNA genome sequencing of COX-deficient fibres represents a sensitive and effective method for elucidating the genetic basis of patients' mitochondrial disease; furthermore, the description and thorough investigation of resultant mtDNA variants are of utmost importance in both the counselling and clinical management of patients. Acknowledgements DMT and RWT acknowledge the continuing financial support of a Wellcome Trust Programme Grant (074454/Z/04/Z) and the UK National Commissioning Group which funds the “Rare Mitochondrial Disorders of Adults and Children” Diagnostic Service (http://www. mitochondrialncg.nhs.uk/index.html). References [1] Sacconi S, Salviati L, Nishigaki Y, Walker WF, Hernandez-Rosa E, Trevisson E, et al. A functionally dominant mitochondrial DNA mutation. Hum Mol Genet 2008;17: 1814–20. [2] DiMauro S, Schon EA. Mitochondrial DNA mutations in human disease. Am J Med Genet 2001;106:18–26. [3] Kondrashov FA. Prediction of pathogenic mutations in mitochondrially encoded human tRNAs. Hum Mol Genet 2005;14:2415–9. [4] McFarland R, Elson JL, Taylor RW, Howell N, Turnbull DM. Assigning pathogenicity to mitochondrial tRNA mutations: when “definitely maybe” is not good enough. Trends Genet 2004;20:591–6. [5] McDonnell MT, Schaefer AM, Blakely EL, Chinnery PF, McFarland R, Turnbull DM, et al. Noninvasive diagnosis of the 3243A N G mitochondrial DNA mutation using urinary epithelial cells. Eur J Hum Genet 2004;12:778–81. [6] Old SL, Johnson MA. Methods of microphotometric assay of succinate dehydrogenase and cytochrome c oxidase activities for use on human skeletal muscle. Histochem J 1989;21:545–55. [7] Blakely EL, He L, Taylor RW, Chinnery PF, Lightowlers RN, Schaefer AM, et al. Mitochondrial DNA deletion in “identical” twin brothers. J Med Genet 2004;41:e19. [8] Taylor RW, Taylor GA, Durham SE, Turnbull DM. The determination of complete human mitochondrial DNA sequences in single cells: implications for the study of somatic mitochondrial DNA point mutations. Nucleic Acids Res 2001;29:E74. [9] Taylor RW, Barron MJ, Borthwick GM, Gospel A, Chinnery PF, Samuels DC, et al. Mitochondrial DNA mutations in human colonic crypt stem cells. J Clin Invest 2003;112:1351–60.
[10] Andrews RM, Kubacka I, Chinnery PF, Lightowlers RN, Turnbull DM, Howell N. Reanalysis and revision of the Cambridge reference sequence for human mitochondrial DNA. Nat Genet 1999;23:147. [11] White HE, Durston VJ, Seller A, Fratter C, Harvey JF, Cross NC. Accurate detection and quantitation of heteroplasmic mitochondrial point mutations by pyrosequencing. Genet Test 2005;9:190–9. [12] Hancock DK, Tully LA, Levin BC. A standard reference material to determine the sensitivity of techniques for detecting low-frequency mutations, SNPs, and heteroplasmies in mitochondrial DNA. Genomics 2005;86:446–61. [13] Weber K, Wilson JN, Taylor L, Brierley E, Johnson MA, Turnbull DM, et al. A new mtDNA mutation showing accumulation with time and restriction to skeletal muscle. Am J Hum Genet 1997;60:373–80. [14] Ruiz-Pesini E, Lott MT, Procaccio V, Poole J, Brandon MC, Mishmar D, et al. An enhanced MITOMAP with a global mtDNA mutational phylogeny. Nucleic Acids Res 2007;35:D823–8. [15] Ingman M, Gyllensten U. mtDB: Human Mitochondrial Genome Database, a resource for population genetics and medical sciences. Nucleic Acids Res 2006;34:D749–51. [16] Mayr JA, Moslemi AR, Forster H, Kamper A, Idriceanu C, Muss W, et al. A novel sporadic mutation G14739A of the mitochondrial tRNAGlu in a girl with exercise intolerance. Neuromusc Disord 2006;16:874–7. [17] Bruno C, Sacco O, Santorelli FM, Assereto S, Tonoli E, Bado M, et al. Mitochondrial myopathy and respiratory failure associated with a new mutation in the mitochondrial transfer ribonucleic acid glutamic acid gene. J Child Neurol 2003;18:300–3. [18] Anitori R, Manning K, Quan F, Weleber RG, Buist NR, Shoubridge EA, et al. Contrasting phenotypes in three patients with novel mutations in mitochondrial tRNA genes. Mol Genet Metab 2005;84:176–88. [19] Hao H, Bonilla E, Manfredi G, DiMauro S, Moraes CT. Segregation patterns of a novel mutation in the mitochondrial tRNA glutamic acid gene associated with myopathy and diabetes mellitus. Am J Hum Genet 1995;56:1017–25. [20] Hanna MG, Nelson I, Sweeney MG, Cooper JM, Watkins PJ, Morgan-Hughes JA, et al. Congenital encephalomyopathy and adult-onset myopathy and diabetes mellitus: different phenotypic associations of a new heteroplasmic mtDNA tRNA glutamic acid mutation. Am J Hum Genet 1995;56:1026–33. [21] Uusimaa J, Finnila S, Remes AM, Rantala H, Vainionpaa L, Hassinen IE, et al. Molecular epidemiology of childhood mitochondrial encephalomyopathies in a Finnish population: sequence analysis of entire mtDNA of 17 children reveals heteroplasmic mutations in tRNAArg, tRNAGlu, and tRNALeu(UUR) genes. Pediatrics 2004;114:443–50. [22] Pereira C, Nogueira C, Barbot C, Tessa A, Soares C, Fattori F, et al. Identification of a new mtDNA mutation (14724G N A) associated with mitochondrial leukoencephalopathy. Biochem Biophys Res Commun 2007;354:937–41. [23] Pancrudo J, Shanske S, Bonilla E, Daras M, Akman HO, Krishna S, et al. Mitochondrial encephalomyopathy due to a novel mutation in the tRNAGlu of mitochondrial DNA. J Child Neurol 2007;22:858–62. [24] Horvath R, Kemp JP, Tuppen HA, Hudson G, Oldfors A, Marie SK, et al. Molecular basis of infantile reversible cytochrome c oxidase deficiency myopathy. Brain 2009;132:3165–74. [25] Elson JL, Swalwell H, Blakely EL, McFarland R, Taylor RW, Turnbull DM. Pathogenic mitochondrial tRNA mutations — which mutations are inherited and why? Hum Mutat 2009;30:E984–92. [26] Al-Dosary M, Whittaker RG, Haughton J, McFarland R, Goodship J, Turnbull DM, et al. Neuromuscular disease presentation with three genetic defects involving two genomes. Neuromuscul Disord 2009;19:841–4.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n s
Short communication
A novel mitochondrial tRNAGlu (MTTE) gene mutation causing chronic progressive external ophthalmoplegia at low levels of heteroplasmy in muscle Charlotte L. Alston a, James Lowe b, Douglass M. Turnbull a, Paul Maddison c, Robert W. Taylor a,⁎ a b c
Mitochondrial Research Group, Institute for Ageing and Health, The Medical School, Newcastle University, Framlington Place, Newcastle upon Tyne, NE2 4HH, UK School of Molecular Medical Sciences, University of Nottingham, Nottingham, UK Department of Clinical Neurology, Nottingham University Hospitals NHS Trust, Nottingham, UK
a r t i c l e
i n f o
Article history: Received 9 July 2010 Received in revised form 30 July 2010 Accepted 6 August 2010 Keywords: Mitochondrial DNA Chronic progressive external ophthalmoplegia tRNAGlu Single fibre studies Pathogenicity
a b s t r a c t Mitochondrial respiratory chain defects are associated with diverse clinical phenotypes in both adults and children, and may be caused by mutations in either nuclear or mitochondrial DNA (mtDNA). We report the molecular genetic investigations of a patient with chronic progressive external ophthalmoplegia (CPEO) and myopathy where muscle biopsies taken 11 years apart revealed a progressive increase in the proportion of cytochrome c oxidase (COX)-deficient fibres. Mitochondrial genetic analysis of the early biopsy had seemingly excluded both mtDNA rearrangements and mtDNA point mutations. Sequencing mtDNA from individual COX-deficient muscle fibres in the second biopsy, however, identified an unreported m.14723 T N C substitution within the mitochondrial tRNAGlu (MTTE) gene, which fulfilled all canonical criteria for pathogenicity. The m.14723 T N C mutation was absent from several tissues, including muscle, from maternal relatives suggesting a de novo event, whilst quantitative analysis of the first muscle biopsy confirmed a very low level of the mutation (7% mutated mtDNA), highlighting a potential problem whereby pathogenic mtDNA mutations may remain undetected using established screening methodologies. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Defects of the mitochondrial respiratory chain are associated with a diverse and ever-growing spectrum of clinical phenotypes in both adults and children, and may be caused by mutations in either the nuclear or mitochondrial genome (mtDNA). Where mitochondrial disease is due to mtDNA mutations, those occurring within mitochondrial (mt-) tRNA genes are the most common, with over 150 discrete pathogenic mutations identified across each of the 22 mt-tRNA genes [1]. They are associated with a number of distinctive clinical presentations ranging from the multi-organ, syndromic phenotypes of MERRF and MELAS, to isolated symptoms such as CPEO, diabetes or myopathy. In spite of the clinical and genetic heterogeneity associated with mt-tRNA mutations, determinants of pathogenicity are typically shared across the spectrum [2–4]. Pathogenic mutations are typically heteroplasmic, a term that describes the presence of both wildtype and mutant mtDNA molecules in a cell. They are present at high levels of heteroplasmy in post-mitotic tissues, such as muscle; are often lost from rapidly dividing cells, such as blood leucocytes [5] and are typically associated with a mosaic pattern of cytochrome c oxidase (COX) deficiency. Their presence at high levels of heteroplasmy in
⁎ Corresponding author. Tel.: + 44 191 2223685; fax: + 44 191 2824373. E-mail address: [email protected] (R.W. Taylor). 0022-510X/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2010.08.014
clinically-relevant tissues facilitates molecular diagnosis by direct sequencing of the whole mitochondrial genome. We present a patient with chronic progressive external ophthalmoplegia (CPEO) and COX-deficient muscle fibres associated with a novel m.14723 T N C mt-tRNAGlu mutation which was present at low levels in the first of two muscle biopsies, thereby evading detection by established sequencing methodologies. The m.14723 T N C mutation clearly segregated with a functional, biochemical defect as measured by COX activity in individual muscle fibres, and the absence of this mutation in muscle and several mitotic tissues from the patient's daughter indicates that it is not transmitted through the female germline. 2. Case report The index patient is a 63-year-old female who initially developed bilateral eyelid droop in early childhood, which was evident in photographs taken at the age of five years. She underwent corrective squint surgery at the age of seven years, and has had slight amblyopia in the left eye, having never suffered from significant double vision. She underwent further ophthalmological review at the age of 38 for worsening bilateral ptosis, subsequently requiring surgical correction on two separate occasions. By the age of 39, she had begun to develop neck and proximal upper limb ache and on neurological review at the age of 51 years, she had subtle neck and proximal upper limb weakness. Needle electromyography revealed myopathic features whilst CK results were normal. Histocytochemical analysis of a needle
C.L. Alston et al. / Journal of the Neurological Sciences 298 (2010) 140–144
quadriceps muscle biopsy taken in 1997 reported low levels (1–2%) of COX-deficient fibres (Fig. 1A). A repeat biopsy was performed 11 years later, in 2008, to aid genetic diagnosis. At present, neurological examination reveals significant ophthalmoplegia with exodeviation in primary gaze, without double vision. She has bilateral ptosis, facial weakness, and mild proximal limb weakness, with no other neurological abnormalities. Repeated ECG findings have been normal. Family history is somewhat unremarkable; her deceased parents were first cousins and her sister has multiple sclerosis, although no evidence of ophthalmoplegia. The patient's 39 year old, younger daughter is currently well, with no abnormal neurological symptoms or signs. Her older, 41 year old daughter has not been examined, but evidence of ophthalmoplegia has not been reported. 3. Materials and methods 3.1. Muscle histology and histochemistry Standard histological and histochemical analyses of quadriceps muscle biopsy were performed on fresh frozen sections (10 μm). Cytochrome c oxidase (COX) staining was carried out on the patient's initial muscle biopsy whilst investigations on the subsequent patient biopsy included the determination of sequential COX/succinate dehydrogenase (SDH) activities using standard methods [6]. 3.2. Molecular genetic studies Total DNA was extracted from several tissues including blood, buccal and urinary epithelium and skeletal muscle using standard
141
procedures. Blood, urine and buccal samples were obtained with consent from clinically-unaffected relatives (patient's sister and patient's daughter), whilst patient anxiety prompted her daughter to have a muscle biopsy. Mitochondrial DNA rearrangements were investigated in muscle DNA using established long-range PCR protocols [7], followed by direct sequencing of the entire mitochondrial genome, performed either in muscle homogenate DNA or DNA from individual COX-deficient muscle fibres isolated by lasermicrocapture, essentially as described elsewhere [8,9].
3.3. Assessment of m.14723 T N C mutation load by PCR-RFLP analysis To determine the level of heteroplasmy of the novel m.14723 T N C variant, we designed a PCR-restriction fragment length polymorphism (RFLP) assay as follows: a 211 bp PCR product spanning the mutation site was amplified using a forward mismatch primer 5′-AGAATAATAACACACCCGACGACAC-3′ (nt 14543–14567; mismatch nucleotide shown in bold ) and a reverse primer (nt 14753–14733), GenBank Accession number NC_012920 [10]; the mismatch nucleotide creates a Hpy99I restriction site, providing an internal control for digestion. The m.14723 T N C mutation creates a further Hpy99I restriction site, permitting the detection of mutant mtDNA amplicons which cut twice to generate bands of 161, 29 and 21 bp, from wild type mtDNA amplicons which cut only once yielding bands of 190 and 21 bp. Prior to the last cycle of PCR, 5 μCi [α-32P]dCTP (3000 Ci/mmol) was added. Labelled products were precipitated, digested overnight with 10U Hpy99I, separated through a 12% nondenaturing polyacrylamide gel and the radioactivity in each fragment was quantified using ImageQuant software (Amersham Biosciences/GE Healthcare).
Fig. 1. Muscle biopsy and mtDNA analysis. Histochemical analysis of patient muscle biopsy. COX staining highlights both COX-deficient fibres (unstained, marked with asterix) and COX-positive fibres (brown) in (A) the 1997 biopsy and (B) the 2008 biopsy; both biopsy sections have been counterstained with haematoxylin to visualise nuclei resulting in slight staining of adipose material. (C) Sequential COX-SDH analysis reveals numerous COX-deficient fibres on the 2008 biopsy section with mitochondrial aggregates corresponding to ragged-red fibres. (D) mtDNA sequencing chromatogram reveals the m.14723T N C substitution, present at very high levels in a COX-deficient fibre compared to a wild type control sequence (E). Sequencing of patient's 1997 muscle biopsy homogenate DNA sample (F) reveals very low levels of the m.14723 T N C mutation although higher levels were apparent in the 2008 muscle biopsy (G).
142
C.L. Alston et al. / Journal of the Neurological Sciences 298 (2010) 140–144
3.4. Assessment of m.14723 T N C mutation load by quantitative pyrosequencing The detection of low level m.14723 T N C heteroplasmy in the original muscle DNA homogenate prompted confirmation by an alternative method to quantitate mtDNA mutation load. Pyromark Assay Design Software v.2.0 (Qiagen) was used to design locus specific PCR and pyrosequencing primers, which amplified a 90 bp PCR product spanning the m.14723 nucleotide using a forward primer (nt 14688–14711) and a biotinylated reverse primer (nt 14777–14757). Pyrosequencing was achieved on the Pyromark Q24 platform according to the manufacturer's protocol, employing a mutationspecific pyrosequencing primer (nt 14704–14721). Pyromark Q24 software was used to quantify the m.14723 T N C heteroplasmy levels by directly comparing the relevant peak heights of both wild type and mutant mtDNA at this site [11].
4. Results 4.1. Histology and histochemistry COX histochemistry of the initial muscle biopsy taken in 1997 revealed 1–2% COX-deficient fibres (Fig. 1A), which first prompted the investigation of this patient. Analysis of the subsequent biopsy, taken 11 years later, demonstrated a significant increase in the proportion of COX-deficient fibres (7% of total biopsy) (Fig. 1B). Sequential COX-SDH analysis revealed that many of the COX-deficient fibres also exhibited subsarcolemmal mitochondrial aggregates which would correspond to ragged-red fibres (Fig. 1C), clearly demonstrating an underlying mitochondrial aetiology.
4.2. Molecular genetics investigations All initial mtDNA genetic testing was performed on muscle DNA which was referred to our laboratory and was extracted from the first (1997) skeletal muscle biopsy. Long-range PCR revealed only full length, wild type products with no evidence of any mtDNA rearrangement. Similarly, sequencing of the entire mitochondrial genome in this sample showed no pathogenic mtDNA mutations. Our patient was re-biopsied in 2008, and having excluded mtDNA deletions in this sample, we proceeded to sequence the entire mitochondrial genome from laser microdissected COX-deficient muscle fibres, revealing a novel m.14723 T N C substitution within the MTTE gene (Fig. 1D). Levels of m.14723 T N C heteroplasmy in COX-positive, biochemically normal, muscle fibres were significantly lower (data not shown). Reanalysis of old chromatograms suggested that the 1997 muscle homogenate sample exhibited very low levels of the m.14723 T N C mutation (Fig. 1F), with higher levels present in the 2008 biopsy (Fig. 1G).
Fig. 2. Characterisation of a novel m.14723 T N C MTTE substitution. (A) Last hot cycle PCRRFLP analysis confirms a low level of m.14723T N C mutation in the patient’s 1997 muscle biopsy (lane 3) with an increased mutation load in the 2008 muscle biopsy (lane 4). Results indicate an absence of m.14723 T N C in patient blood (lane 5). Control samples are also shown; uncut DNA sample (lane 1), control DNA sample (lane 2). (B) Pyrograms illustrating (i) low level of m.14723TN C heteroplasmy in the 1997 muscle biopsy; (ii) higher heteroplasmy levels in the 2008 patient biopsy and (iii) absence of m.14723 T N C in a muscle biopsy from the patient's clinically-unaffected daughter. (C) Graphical representation of single fibre PCR analysis clearly shows a marked segregation of the novel m.14723 T N C mutation with a biochemical defect in COX-deficient muscle fibres.
5. Discussion 4.3. Quantitative analysis of the m.14723 T N C mutation We quantified the level of m.14723 T N C heteroplasmy at 7% mutant load in the patient's initial biopsy and at 34% heteroplasmy in her later biopsy by last hot cycle PCR-RFLP analysis (Fig. 2A), which was confirmed by Pyrosequencing analysis (Fig. 2B). The m.14723 T N C substitution was absent in all other patient tissues tested including blood, urine and buccal epithelial cells, and was absent in all available tissues from the patient’s daughter – including skeletal muscle biopsy (Fig. 2B) – and from her sister. Single fibre analyses revealed higher levels of m.14723 T N C heteroplasmy in COXdeficient fibres (99.7 ± 0.5% (n = 10)) compared to COX-positive fibres (33.7 ± 27.8% (n = 8)), a statistically significant finding (P b 0.001, two-tailed Student's t test) (Fig. 2C).
We describe a patient who presented with CPEO and muscle biopsy changes suggestive of a mtDNA defect, in whom a novel m.14723 T N C mitochondrial tRNA mutation was identified. This novel substitution was present at an uncharacteristically low level of mtDNA heteroplasmy in an initial muscle biopsy, so low in fact that it was missed by the whole mtDNA genome analysis of muscle homogenate DNA, the established diagnostic screen for the detection of rare or novel mtDNA point mutations. This patient's suggestive clinical presentation of CPEO, ptosis and myopathy – combined with a number of COX-deficient muscle fibres on muscle biopsy – prompted further molecular investigation, and only following mtDNA genome sequencing of isolated, biochemically-deficient muscle fibres was the causative m.14723 T N C mutation identified.
C.L. Alston et al. / Journal of the Neurological Sciences 298 (2010) 140–144
Single fibre analysis of COX-deficient and COX-positive muscle fibres revealed a clear correlation between m.14723 T N C mutation load and the respiratory chain abnormality, corroborating the pathogenicity of this novel mtDNA substitution. Assessment of the m.14723 T N C mutation in the non-invasive tissues (e.g. buccal and urinary cells) from the patient and her maternal relatives confirmed that the mutation was confined to patient muscle, and is therefore likely to have arisen sporadically. Our patient has undergone two muscle biopsies, taken eleven years apart, which demonstrate an increase in the frequency of COXdeficient fibres over time and a concomitant increase in m.14723T N C mutation load between muscle homogenate DNA samples. Following a reanalysis of the original muscle homogenate sequencing chromatograms, the m.14723 T N C substitution was visible but at a very low level of heteroplasmy (7% mutation load) — well below the quoted sensitivity level of the sequencing assay [8,12] thus explaining our failure to identify it during initial sequencing studies (Fig. 1F). There is certainly a risk that at such low levels, novel heteroplasmic mtDNA substitutions may be merely dismissed as sequencing background artefact, or of being masked by this phenomenon completely. Interestingly, sequencing of the repeat (2008) muscle biopsy homogenate clearly identified the m.14723 T N C substitution (Fig. 1G), this time at a higher level of heteroplasmy (34% mutation load), correlating with the increased number of COX-deficient muscle fibres on muscle biopsy; pyrosequencing confirmed results for all samples assayed by PCR-RFLP. Similar cases are reported in the literature where mt-tRNA point mutation levels are shown to have increased over time, often correlating with a deteriorating clinical picture in terms both of histopathological changes on muscle biopsy and symptom severity, as characterised by our index case [13]. The novel m.14723 T N C substitution is not a recognised polymorphic variant [14,15], is situated in the highly conserved dihydrouridine arm of the glutamic acid tRNA (Fig. 3), and is predicted to disrupt the Watson and Crick base pairing. Interestingly, a further published pathogenic mtDNA mutation occurring at the adjacent nucleotide (m.14724G N A) also disrupts the Watson and Crick base pairing, and is situated within the same highly conserved dihydrouridine region. Despite the markedly different clinical presentation (index patient harbouring m.14724G N A had a neonatal presentation dominated by hypotonia, macrocephaly and cerebellar ataxia), the presence of another substitution mutation in the same region of the tRNA adds further weight to our hypothesis that the m.14723 T N C substitution forms the molecular basis of our patient's presentation. The canonical scoring system proposed by McFarland et al. [4], when applied to the m.14723 T N C substitution, deems it to be probably
143
pathogenic, with a score of 11 out of a possible 20 points. The absence of transmitochondrial cybrid or steady state mt-tRNAGlu level data, and the absence of respiratory chain enzyme results, precludes it from the “definitely pathogenic” category. mt-tRNA mutations present at low levels may never reach the threshold for categorisation under “definitely pathogenic” according to the criteria set out by McFarland et al. and perhaps this method for establishing pathogenicity may benefit from some flexibility in the case of dominant-acting or low level mt-tRNA mutations, whereby the weighting of single fibre data is greater than currently ascribed. This is the seventh discrete mutation to be described within the MTTE gene. As commonly observed in other mt-tRNA genes, MTTE mutations are associated with a wide range of clinical phenotypes including exercise intolerance [16], myopathy [17,18], diabetes [19,20], progressive encephalopathy [21,22] and encephalomyopathy [23]. A homoplasmic m.14674 T N C MTTE gene mutation has recently been described as the cause of infantile, reversible COX deficiency myopathy [24]. Despite the phenotypic heterogeneity associated with MTTE mutations, the m.14723 T N C mutation in our patient represents the first mutation in this mt-tRNA gene to be associated with a clinical presentation of CPEO. The presence of an mtDNA mutation that is confined to muscle poses a difficult situation in genetic counselling. The m.14723 T N C mutation does not represent a germline mutation due to the absence of m.14723 T N C in the patient's non-invasive tissues, yet mutationfree oocytes cannot be guaranteed. It is not standard practice for unaffected individuals to undergo muscle biopsy due to the invasive nature of the procedure and time/financial restraints, but the degree of anxiety caused by the uncertain genetic status to the patient's daughter was sufficient to deem it a viable option to fully exclude maternal transmission. In turn, we have been able to show that this novel mt-tRNA mutation has not been transmitted, although whether this is due to the de novo mutational event occurring after the mesodermal differentiation of myoblasts from germ cells, or if it is due to the non-transmissible nature of the mutation is unclear. The mechanisms determining which mt-tRNA mutations are transmitted from mother to child remain elusive [25]. Given the spectrum of devastating clinical phenotypes that can be attributed to mt-tRNA mutations, further investigation will certainly lead to better understanding of their pathogenic nature with an aim to improve counselling of our patients and in particular, may prove beneficial when predicting recurrence risks for future pregnancies. Our index case represents the third incidence of a pathogenic mt-tRNA mutation present at low levels of heteroplasmy in muscle homogenate. Previous reports have described a dominant-
Fig. 3. Conservation and molecular location of the novel mt-tRNA substitution. (A) The m.14723 T N C substitution is located within the highly conserved dihydrouridine arm of the mitochondrial glutamate tRNA molecule; the base pair between the affected m.14723 nucleotide and its Watson and Crick correlate at m.14731 is highly conserved across numerous species (B) indicating that it plays an important role in the tertiary structure and/or function of the tRNA molecule.
144
C.L. Alston et al. / Journal of the Neurological Sciences 298 (2010) 140–144
acting mt-tRNA mutation [1] and – as in our index case – a pathogenic mt-tRNA mutation present at low levels in muscle homogenate [26]. Taken together, these unusual clinical cases highlight the value of investigating isolated COX-deficient fibres, particularly when the usual avenues of investigation have previously been explored and excluded. Whole mtDNA genome sequencing of COX-deficient fibres represents a sensitive and effective method for elucidating the genetic basis of patients' mitochondrial disease; furthermore, the description and thorough investigation of resultant mtDNA variants are of utmost importance in both the counselling and clinical management of patients. Acknowledgements DMT and RWT acknowledge the continuing financial support of a Wellcome Trust Programme Grant (074454/Z/04/Z) and the UK National Commissioning Group which funds the “Rare Mitochondrial Disorders of Adults and Children” Diagnostic Service (http://www. mitochondrialncg.nhs.uk/index.html). References [1] Sacconi S, Salviati L, Nishigaki Y, Walker WF, Hernandez-Rosa E, Trevisson E, et al. A functionally dominant mitochondrial DNA mutation. Hum Mol Genet 2008;17: 1814–20. [2] DiMauro S, Schon EA. Mitochondrial DNA mutations in human disease. Am J Med Genet 2001;106:18–26. [3] Kondrashov FA. Prediction of pathogenic mutations in mitochondrially encoded human tRNAs. Hum Mol Genet 2005;14:2415–9. [4] McFarland R, Elson JL, Taylor RW, Howell N, Turnbull DM. Assigning pathogenicity to mitochondrial tRNA mutations: when “definitely maybe” is not good enough. Trends Genet 2004;20:591–6. [5] McDonnell MT, Schaefer AM, Blakely EL, Chinnery PF, McFarland R, Turnbull DM, et al. Noninvasive diagnosis of the 3243A N G mitochondrial DNA mutation using urinary epithelial cells. Eur J Hum Genet 2004;12:778–81. [6] Old SL, Johnson MA. Methods of microphotometric assay of succinate dehydrogenase and cytochrome c oxidase activities for use on human skeletal muscle. Histochem J 1989;21:545–55. [7] Blakely EL, He L, Taylor RW, Chinnery PF, Lightowlers RN, Schaefer AM, et al. Mitochondrial DNA deletion in “identical” twin brothers. J Med Genet 2004;41:e19. [8] Taylor RW, Taylor GA, Durham SE, Turnbull DM. The determination of complete human mitochondrial DNA sequences in single cells: implications for the study of somatic mitochondrial DNA point mutations. Nucleic Acids Res 2001;29:E74. [9] Taylor RW, Barron MJ, Borthwick GM, Gospel A, Chinnery PF, Samuels DC, et al. Mitochondrial DNA mutations in human colonic crypt stem cells. J Clin Invest 2003;112:1351–60.
[10] Andrews RM, Kubacka I, Chinnery PF, Lightowlers RN, Turnbull DM, Howell N. Reanalysis and revision of the Cambridge reference sequence for human mitochondrial DNA. Nat Genet 1999;23:147. [11] White HE, Durston VJ, Seller A, Fratter C, Harvey JF, Cross NC. Accurate detection and quantitation of heteroplasmic mitochondrial point mutations by pyrosequencing. Genet Test 2005;9:190–9. [12] Hancock DK, Tully LA, Levin BC. A standard reference material to determine the sensitivity of techniques for detecting low-frequency mutations, SNPs, and heteroplasmies in mitochondrial DNA. Genomics 2005;86:446–61. [13] Weber K, Wilson JN, Taylor L, Brierley E, Johnson MA, Turnbull DM, et al. A new mtDNA mutation showing accumulation with time and restriction to skeletal muscle. Am J Hum Genet 1997;60:373–80. [14] Ruiz-Pesini E, Lott MT, Procaccio V, Poole J, Brandon MC, Mishmar D, et al. An enhanced MITOMAP with a global mtDNA mutational phylogeny. Nucleic Acids Res 2007;35:D823–8. [15] Ingman M, Gyllensten U. mtDB: Human Mitochondrial Genome Database, a resource for population genetics and medical sciences. Nucleic Acids Res 2006;34:D749–51. [16] Mayr JA, Moslemi AR, Forster H, Kamper A, Idriceanu C, Muss W, et al. A novel sporadic mutation G14739A of the mitochondrial tRNAGlu in a girl with exercise intolerance. Neuromusc Disord 2006;16:874–7. [17] Bruno C, Sacco O, Santorelli FM, Assereto S, Tonoli E, Bado M, et al. Mitochondrial myopathy and respiratory failure associated with a new mutation in the mitochondrial transfer ribonucleic acid glutamic acid gene. J Child Neurol 2003;18:300–3. [18] Anitori R, Manning K, Quan F, Weleber RG, Buist NR, Shoubridge EA, et al. Contrasting phenotypes in three patients with novel mutations in mitochondrial tRNA genes. Mol Genet Metab 2005;84:176–88. [19] Hao H, Bonilla E, Manfredi G, DiMauro S, Moraes CT. Segregation patterns of a novel mutation in the mitochondrial tRNA glutamic acid gene associated with myopathy and diabetes mellitus. Am J Hum Genet 1995;56:1017–25. [20] Hanna MG, Nelson I, Sweeney MG, Cooper JM, Watkins PJ, Morgan-Hughes JA, et al. Congenital encephalomyopathy and adult-onset myopathy and diabetes mellitus: different phenotypic associations of a new heteroplasmic mtDNA tRNA glutamic acid mutation. Am J Hum Genet 1995;56:1026–33. [21] Uusimaa J, Finnila S, Remes AM, Rantala H, Vainionpaa L, Hassinen IE, et al. Molecular epidemiology of childhood mitochondrial encephalomyopathies in a Finnish population: sequence analysis of entire mtDNA of 17 children reveals heteroplasmic mutations in tRNAArg, tRNAGlu, and tRNALeu(UUR) genes. Pediatrics 2004;114:443–50. [22] Pereira C, Nogueira C, Barbot C, Tessa A, Soares C, Fattori F, et al. Identification of a new mtDNA mutation (14724G N A) associated with mitochondrial leukoencephalopathy. Biochem Biophys Res Commun 2007;354:937–41. [23] Pancrudo J, Shanske S, Bonilla E, Daras M, Akman HO, Krishna S, et al. Mitochondrial encephalomyopathy due to a novel mutation in the tRNAGlu of mitochondrial DNA. J Child Neurol 2007;22:858–62. [24] Horvath R, Kemp JP, Tuppen HA, Hudson G, Oldfors A, Marie SK, et al. Molecular basis of infantile reversible cytochrome c oxidase deficiency myopathy. Brain 2009;132:3165–74. [25] Elson JL, Swalwell H, Blakely EL, McFarland R, Taylor RW, Turnbull DM. Pathogenic mitochondrial tRNA mutations — which mutations are inherited and why? Hum Mutat 2009;30:E984–92. [26] Al-Dosary M, Whittaker RG, Haughton J, McFarland R, Goodship J, Turnbull DM, et al. Neuromuscular disease presentation with three genetic defects involving two genomes. Neuromuscul Disord 2009;19:841–4.