ABSTRACT: In addition to well-known mutations at nucleotide pair 8344 and8356 in mitochondrial DNA in patients with myoclonus epilepsy associatedwith ragged-red fibers (MERRF), we found a new G-to-A point mutation atnucleotide 8363 in two Japanese families. The probands had the typicalclinical characteristics of MERRF. Since the 8363 mutation was present in aheteroplasmic state, and seen in none of 92 patients with other mitochon-drial diseases or 50 normal individuals, this mutation is thought to be dis-ease-related and probably specific to MERRF. As seen in muscle biopsieswith the previous two mutations, focal cytochrome c oxidase (CCO) defi-ciency was the most characteristic finding. With single fiber analysis, theCCO-negative fibers contained a higher percentage of mutant DNA (88.4 ±6.6%) than CCO-positive fibers (65.1 ± 8.0%). These findings suggest thatmutations in tRNALys coding region are related to the MERRF phenotypeand are responsible for the reduced CCO activity. © 1997 John Wiley & Sons,Inc. Muscle Nerve, 20, 271–278, 1997.Key words: myoclonus epilepsy associated with ragged-red fibers; mito-chondrial DNA; tRNALys; point mutation; cytochrome c oxidase

MYOCLONUS EPILEPSY ASSOCIATEDWITH RAGGED-RED FIBERS: A G-TO-AMUTATION AT NUCLEOTIDE PAIR8363 IN MITOCHONDRIAL tRNALys INTWO FAMILIES

MATSUKO OZAWA, MD, 1,2,3 ICHIZO NISHINO, MD,2 SATOSHI HORAI, PhD, 4

IKUYA NONAKA, MD 1,2 and YU-ICHI GOTO, MD1,2 *

1Department of Laboratory Medicine, National Center Hospital for Mental, Nervousand Muscular Disorders, Tokyo, Japan2Department of Ultrastructural Research, National Institute of Neuroscience, NCNP,4-1-1 Ogawahigashi, Kodaira, Tokyo, Japan3Second Department of Pediatrics, Toho University School of Medicine, Tokyo,Japan4Department of Human Genetics, National Institute of Genetics, Mishima, Japan

Received 5 June 1996; accepted 31 August 1996

Recent genetic studies of mitochondrial DNA(mtDNA) have revealed the complexity of genotype–phenotype relationship in mitochondrial diseases.15

It is necessary to clarify this relationship in order tounderstand its pathomechanism and for developingnew therapeutic strategies for these diseases. Thecomplexity includes genetic and phenotypic hetero-geneities associated with mtDNA9 and the possiblenuclear DNA involvement. In this context, it is im-portant to describe how each genetic abnormality isrelated to phenotypic expression at organelle, cellu-lar, tissue, and individual levels.

Myoclonus epilepsy associated with ragged-red fi-

bers (MERRF) is one of the mitochondrial encepha-lomyopathies characterized by myoclonus epilepsy,cerebellar ataxia, dementia, and myopathy.7 An A-to-G transition mutation at nucleotide (nt) 8344(8344 mutation) in mitochondrial tRNALys has beenidentified in approximately 80% of the patients.25,30

In 1992, a second mutation, a T-to-C transition mu-tation at nt 8356 (8356 mutation), in the sametRNALys gene, was reported in patients with MERRFor MERRF/mitochondrial myopathy, encephalopa-thy, lactic acidosis, and strokelike episodes(MELAS).26,32

The 8344 mutation was shown to cause impair-ment of mitochondrial protein synthesis in both cul-tured myoblasts2,3 and r0 cell transformants fromMERRF patients.3,31 In studies of 8344 mutation-carrying cells, the reduced protein synthesis was

*Correspondence to: Dr. Yu-ichi Goto

CCC 0148-639X/97/030271-08© 1997 John Wiley & Sons, Inc.

8363 Mutation in MERRF MUSCLE & NERVE March 1997 271

caused by the premature termination of translationat the lysine codon, due to the deficiency of amino-acylation of tRNALys.6

We report a third mutation in the mitochondrialtRNALys in two Japanese families with MERRF, fur-ther supporting the contention that mutations in thetRNALys coding region are responsible for MERRFand cytochrome c oxidase (CCO) deficiency.

CASE REPORT

The pedigrees of two families are shown in Figure 1.In family A, patient II-1, an 8-year-old boy, developedmyoclonus epilepsy, generalized seizures, and cer-ebellar ataxia at 7 years. He also had mild muscleweakness with hypotonia. Lactate and pyruvate levelswere elevated in blood and cerebrospinal fluid(CSF). He had cardiac insufficiency with decrease inthe ejection fraction by echocardiography, which im-proved after treatment with Cardiochromet (Insti-tuto Sieroterapico Milanese Serafino Belfauti, Mi-lano, Italy). His mother, I-2, developed convulsionsat age 33 years.

In family B, patient II-2, a 15-year-old girl, hadmyoclonus epilepsy and cerebellar ataxia since theage of 9 years. Physical examination revealed marked

and generalized muscle weakness and atrophy, andmoderate mental deterioration. Her mother, patientI-2, a 52-year-old woman, had no myoclonus but dys-arthria which began in her 40s. Serum lactate andpyruvate levels were significantly elevated after aero-bic exercise, but were normal at rest.

MATERIALS AND METHODS

Pathological and Biochemical Analysis of Muscle Bi-opsies. Biopsies of the biceps brachii muscle werequickly frozen in isopentane alcohol cooled by liquidnitrogen for histochemical and biochemical studies.Serial frozen sections were stained with hematoxylinand eosin (H&E), modified Gomori trichrome(mGT), and succinate dehydrogenase (SDH), and abattery of histochemical methods,5 including CCOstain.24 On cross sections of each biopsy, the num-bers of ragged-red fibers (RRFs) on mGT stain andCCO-negative fibers on CCO stain among approxi-mately 3000 fibers of each specimen were deter-mined.

Mitochondria were isolated from either fresh orfrozen samples and respiratory chain enzyme activi-ties were determined biochemically.14

DNA Preparation. DNA samples were preparedfrom the muscle biopsies of 35 MELAS, 33 MERRF,and 24 chronic progressive external ophthalmople-gia (CPEO) patients by the method described previ-ously.8 Heparinized blood samples of some membersin both families were prepared by the method ofMiller et al.16 We also used 50 placental DNA from30 Japanese, 15 Europeans, and 5 Africans as normalcontrols.12

Sequence Analysis. We designed 11 sets of primersto amplify the DNA fragments which cover all the 22mitochondrial tRNA regions. Amplified fragmentsby a polymerase chain reaction (PCR) were purifiedby MicroSpint S-400 HR Columns (Pharmacia Bio-tech) and sequenced directly with PRISMt ReadyReaction Terminator Cycle Sequence Kit and amodel 377 automatic DNA sequencer (Applied Bio-systems, Perkin Elmer). Sequence data were com-pared with the sequence described by Anderson etal.1

PCR with a Mismatch Primer and Sna BI Digestion.For the detection of A-to-G transition mutation at

nt 8363, we used a modified forward primer (58-AGATTAAGAGAACCAACACCTCTTTTACGT-38: nt8333–8362) with mismatches (underlined) and a re-verse primer (58-GGGGGTAATTATGGTGGGCC-38:nt 8391–8410). The PCR amplification was per-

FIGURE 1. Top: pedigree of two families and the proportion ofmutant mtDNA in DNA extracted from muscle and blood. Bottom:Sna BI digestion of PCR-amplified mtDNA extracted from muscleand blood of patients and available family members. In the pres-ence of 8363 mutation, an 87-bp PCR DNA fragment is cut into57-bp and 30-bp fragments in heteroplasmic fashion. s, asymp-tomatic individuals: , mildly affected patient; d, severely af-fected patients; B, DNA extracted from blood; M, DNA extractedfrom muscle; N, DNA extracted from normal individual; n.d., notdone. *Percentage of mutant DNA.

272 8363 Mutation in MERRF MUSCLE & NERVE March 1997

formed for 30 cycles (20 s at 94°C, 20 s at 55°C, and1 min at 72°C) on a Thermal Cycler (PJ9600, Ap-plied Biosystems, Perkin Elmer) followed by Sna BIdigestion according to the manufacturer’s protocol(Behringer Manheim). If there was a transition mu-tation at nt 8363, the PCR-amplified products (87bp) and two cleaved fragments (57 bp and 30 bp)were detected on a 4% agarose gel (NuSievet 3:1Agarose, FMC Bioproducts) stained by ethidium bro-mide. Fragments originating from normal sequenceswere not digested by Sna BI.

Plasmid Construction. The PCR products encom-passing the entire tRNALys region were cloned withTA Cloningt kit (Invitrogen) according to themanufacturer’s protocol. Plasmids with the mutantsequence were detected by the methods describedabove. Both mutant and wild-type plasmid DNA werepurified by a plasmid DNA separation kit (PlasmidMini Kitt, Qiagen) and diluted to 10 µg/mL.

Dot Blot Hybridization. The first PCR amplification(10 s at 94°C, 10 s at 55°C, and 30 s at 72°C, for 30cycles) was performed with the forward primer (58-AGCCCACTGTAAAGCTAACT-38: nt 8291–8310)and the reverse primer (58-GGGGGTAATTATG-GTGGGCC-38: nt 8391–8410) to product 120-bpfragments. The PCR products were electrophoresedthrough 4% agarose gel (NuSievet 3:1 Agarose,FMC Bioproducts). For the exclusion of the nonspe-cific band, the target fragments were eluted and pu-rified using the spin column (Microcont and Micro-p u r et , A m i c o n ) . T h e n t h e s e c o n d P C Ramplification (10 s at 94°C, 10 s at 55°C, and 30 s at72°C, for 10 cycles) was performed. The concentra-tions of the PCR products were determined by aspectrophotometer (DU 640t, Beckman). Four hun-dred microliters containing one of the PCR productswas fixed on a nylon membrane (HyBondt-N+, Am-ersham) using Milliblot systemt (Millipore) and anUltraviolet Crosslinkert (Amersham), and then hy-bridized to the oligonucleotide probe (58 -TCTTTACAGTAAAATGCCCCA-38) labeled withECLt oligo 38-labeling system (Amersham) to detectmutants. The hybridized membrane was exposed onX-ray film and the signal intensity of the labeled frag-ment was counted by Image Mastert (PharmaciaBiotech). The standard curve was constructed by aseries of plasmid DNA of known mutant proportions(Fig. 2).

Single Muscle Fiber Analysis. Eight-micrometer-and 25-µm-thick frozen sections from patient I-2 infamily B were stained with CCO24 and SDH,5 respec-

tively. Single fibers were isolated from the SDH-stained section after careful examination of CCO ac-tivity in an adjacent serially cut section. Fibers wereisolated using a microcapillary tube under an in-verted microscope.18,29 Each isolated fiber was put in20 µL of water in a 600-µL -Eppendorf tube, frozenat −80°C, and then boiled for 10 min, and rapidlychilled on ice. The proportion of the mutant mtDNAwas determined by the dot blot hybridizationmethod described above.

RESULTS

Pathological and Biochemical Analysis of Muscle.Pathological and biochemical findings in muscles

of 3 patients with the 8363 mutation (II-1 family A,I-2 and II-2 in family B) are summarized in Table 1.

In mGT-stained sections, RRFs were detected in0.5% of the muscle fibers in patient I-2 in family B(Fig. 3), and some RRF-like fibers were found inpatient II-2 in family B. Patient II-1 in family A hadno RRFs but there were apparent myopathic changessuch as fiber size variability in type 1 fibers as well astype 2B fiber atrophy. With the SDH stain, stronglySDH-reactive blood vessels (SSVs)10 were seen onlyin patient II-1 in family A. CCO staining revealedthat there were scattered to clustered CCO-negative

FIGURE 2. Standard curve for quantitation of 8363 mutation.

8363 Mutation in MERRF MUSCLE & NERVE March 1997 273

fibers in all patients showing focal enzyme defi-ciency. Without exception all RRFs and SSVs wereCCO-negative.

Biochemically, all 3 patients had respiratorychain enzyme deficiency; 1 had complex I defi-ciency, another had complex IV deficiency, and thethird had combined deficiency of both complexes.

Detection of 8363 Mutation. Sequence analysis wasapplied to genomic DNA from the patient II-1 in afamily A, and we found a G-to-A transition at nt 8363in the 8-year-old boy with MERRF (II-1, family A),harboring neither the 8344 nor the 8356 mutation.This site is located at the end of the aminoacyl ac-ceptor stem (Fig. 4A), and the nucleotides are highly

Table 1. Histochemical findings and mitochondrial respiratory chain enzyme activities

PatinetAge at musclebiopsy (years)

Muscle pathologyRespiratory chain enzyme activity

(nmol/min/mg mitochondrial protein)

RRF SSVCCO-deficient

fibersNCCR

(27.3 ± 11.6)*SCCR

(76.6 ± 17.7)*CCO

(33.3 ± 16.1)*

II-1 in family A 7 — + 30% 7.0 253.7 29.4II-2 in family B 15 0.5%† − 80% 2.0 145.1 7.0I-2 in family B 48 0.5% − 75% 11.8 62.9 6.6

RRF, ragged-red fibers (fibers with the presence of granular accumulations in the subsarcolemmal area with the modified Gomori trichrome stain);SSV, strongly succinate dehydrogenase-reactive blood vessels; CCO, cytochrome c oxidase; NCCR, NADH cytochrome c reductase; SCCR, succinatecytochrome c reductase*Control values ± SD.†RRF-like fibers (fibers stained not as intensely as RRF, but darker than normal fibers).

FIGURE 3. Muscle pathology from patient I-2 in family B (A–C). There is moderate variation in fiber size with scattered ragged-red fibers(*), which have high enzyme activity (B). Not only ragged-red fibers, but numerous normal-appearing fibers have defective cytochromec oxidase activity (C). (D) Normal control; (A) modified Gomori-trichrome; (B) succinate dehydrogenase; (C–D) cytochrome c oxidase;(A–C) (serial sections) ×300; (D) ×350.

274 8363 Mutation in MERRF MUSCLE & NERVE March 1997

conserved through evolution from humans to seaurchins (Fig. 4B).13,27

By using a PCR-plus Sna BI digestion, we exam-ined DNA extracted from muscle and blood samplesfrom patients with MELAS (with or without the 3243or 3271 mutation in tRNALeu(UUR) and CPEO (withor without large-scale deletion) as well as MERRF(with or without the 8344 or 8356 mutation intRNALys) and the placental DNA from normal indi-viduals of the three racial groups (Table 2). Thesestudies revealed that only one additional family withMERRF (family B), with neither the 8344 nor the8356 mutation, had this mutation. To determinewhether the two mutations were genetically related,we performed sequence analysis of entire tRNA re-gions for patients I-2 and II-2 in family B. Unlike

family A, family B had a rare polymorphic basechange, a T-to-G transversion mutation at 10410, intRNAArg.11 This suggests that these two families arelikely to be genetically different.

Proportion of the Mutant mtDNA. The proportionsof mutant mtDNA in muscle and blood from 2 familymembers, calculated by dot blot hybridization,seemed to correlate with clinical severity such as ageat onset (Fig. 1 and Table 1).

Single Muscle Fiber Analysis. Single muscle fiberanalysis in 1 patient (I-2 in family B) revealed thatthe proportions of mutant mtDNA were 88.4 ± 6.6%in CCO-negative fibers and 65.1 ± 8.0% in CCO-positive fibers. The proportion of mutant mtDNA in

FIGURE 4. (A) Proposed secondary structure of tRNALys showing the mutation at nucleotide 8363 in the aminoacyl acceptor stem. (B)Comparison of the tRNALys gene sequence in different species. Boxes show the high degree of conservation of guanine at nucleotide8363 and of cytosine (C) at nucleotide 8295.

8363 Mutation in MERRF MUSCLE & NERVE March 1997 275

CCO-negative fibers was higher than that in CCO-positive fibers, and this difference was statistically sig-nificant (P < 0.001) (Fig. 5).

DISCUSSION

In MERRF, approximately 80% of patients have beenreported to have the 8344 mutation in mtDNA, butthree families, including a Japanese family, have the8356 mutation.21,26,32 Both mutated sites are locatedin the TCC stem of tRNALys (Fig. 4A). A Japanesefamily with both MERRF and MELAS symptoms hada different mutation at nt 7512 in tRNASer, but not intRNALys.19 Therefore, one may question whethermutations in tRNALys coding region are related spe-cifically to the MERRF phenotype.

We conclude, based on three lines of evidence,that the novel 8363 mutation is disease-specific. Thefirst is that the mutation was found in two unrelatedJapanese families, in which the probands had the

typical clinical characteristics of MERRF, includingmyoclonus epilepsy, cerebellar ataxia, and myopathywith maternal inheritance. The patients in these twofamilies are considered to be genetically divergent,because family B had a rare polymorphic change ofa T-to-G transversion mutation at 10410.

The second line of evidence is that the 8363 mu-tation was heteroplasmic. Heteroplasmy is an impor-tant clue to determine whether a given nucleotidesubstitution is disease-related. Several studies usingthe r0-cell system have confirmed that the activityand/or synthesis of respiratory chain enzymes aredecreased when the proportion of the mutantmtDNA exceeds a certain level, i.e., the thresholdlevel. The difference in proportions of the mutantmtDNA (heteroplasmy) from tissue to tissue hasbeen reported to be minimal in the 8344 mutation,28

but significant in the 3243 mutation.4 The propor-tions of the 8363 mutation in blood and skeletalmuscle were also not significantly different in ourMERRF patients. Different proportions among mu-tations in tRNALeu(UUR) and tRNALys might accountfor the different phenotypic expressions betweenMELAS and MERRF.

The third line of evidence is based on singlemuscle fiber analysis. In 1 patient (I-2, family B), theproportion of mutant mtDNA in CCO-negative fi-bers was higher (88.4 ± 6.6%) than that of CCO-positive fibers (65.1 ± 8.0%), confirming that in-creased mutant mtDNA was responsible forimpairment of mitochondrial function.

The morphological alterations in patients withthis mutation appeared to be mild; typical RRFs werefound in only 1 patient and comprised 0.5% of thetotal fibers and SSVs in 1 patient. The striking find-ing was, however, the presence of numerous fiberswith no CCO activity as confirmed by histochemicalanalysis. We have found that all RRFs are always CCO

Table 2. 8363 mutation in diseased and normal individuals.

Number examined 8363 mutation

MELAS (n = 35)With 3243 20 0With 3271 1 0Without 3243 or 3271 14 0

MERRF (n = 33)With 8344 13 0Without 8344 or 8356 20 2

CPEO (n = 24)With deletion 9 0With no deletion 15 0

Normal controls (n = 50)Japanese 30 0European 15 0African 5 0

MELAS, mitochondrial myopathy, encephalopathy, lactic acidosis, andstrokelike episodes; MERRF, myoclonus epilepsy associated withragged-red fibers; CPEO, chronic progressive externalophthalmoplegia.

FIGURE 5. Proportion of mutant DNA in single fibers from patient I-2 in family B. The proportions of mutant mtDNA are 88.4 ± 6.6% inCCO-negative fibers and 65.1 ± 8.0% in CCO-positive fibers. The proportion of mutant mtDNA in CCO-negative fibers is higher than thatin CCO-positive fibers, and this difference is statistically significant (P < 0.001).

276 8363 Mutation in MERRF MUSCLE & NERVE March 1997

deficient in muscles from MERRF patients.20 In con-trast, RRFs in MELAS patients with the tRNALeu(UUR)

mutation have variable CCO activity.9 In patientsharboring the 3243 mutation, the mutant mtDNAcomprised 95 ± 3% in CCO-negative RRFs, 90 ± 6%in CCO-positive RRFs, and 56 ± 21% in normal fi-bers, indicating that CCO deficiency is not the onlydeterminant for mitochondrial proliferation.17,22

These findings together with those in our presentstudy strongly suggest that decrease in CCO activityis a primary biochemical defect and a trigger for RRFformation in patients with mitochondrial tRNALys

mutations.This mutation was reported by Santrelli et al. in

two unrelated families with maternally inherited car-diomyopathy and hearing loss,23 which indicatesthere may be a phenotypic variability with this mu-tation. Cardiomyopathy is sometimes critical for theprognosis of MELAS and MERRF patients. Four of10 patients with the 8344 mutation, reported previ-ously by us, had cardiac involvement and 2 of themdeveloped progressive heart failure.2 One of our pa-tients with 8363 mutation (II-1 in family A) also hadmild cardiac symptoms, which should be carefullyobserved.

In conclusion, the 8363 mutation is related toMERRF and probably induces a tRNALys dysfunctionand CCO deficiency. Further studies using r0 celltransformants containing the mutant genome arenecessary to clarify functional similarities and differ-ences between the 8344 and 8363 mutations at thecellular level.

We wish to express our cordial thanks to Professor T. Aoki (TohoUniversity School of Medicine, Tokyo, Japan) for his helpful sug-gestions and advice on this work and Mrs. R. Oketa (NCNP) forher technical assistance. We also thank Drs. Y. Osawa, A. Watan-abe, and J. Kohyama (Tokyo Medical and Dental University, To-kyo, Japan) for referring and providing clinical information onpatients in family A, and Drs. A. Maruyama, H. Yamamoto, and M.Fujimoto (St. Marianna University School of Medicine, Kawasaki,Japan) in family B.

REFERENCES

1. Anderson S, Bankier AT, Barrell BG, de Bruijn MHL, CoulsonAR, Drouin J, Eperon IC, Nierlich DP, Roe BA, Sanger F,Schreier PH, Smith AJH, Staden R, Young IG: Sequence andorganization of the human mitochondrial genome. Nature1981;290:457–465.

2. Boulet L, Karpati G, Shoubridge EA: Distribution and thresh-old expression of the tRNALys mutation in skeletal muscle ofpatients with myoclonic epilepsy and ragged-red fibers(MERRF). Am J Hum Genet 1992;51:1187–1200.

3. Chomyn A, Meola G, Bresolin N, Lai ST, Scarlato G, AttardiG: In vitro genetic transfer of protein synthesis and respira-tion defects to mitochondrial DNA-less cells with myopathy-patient mitochondria. Mol Cell Biol 1991;11:2236–2244.

4. Ciafaloni E, Ricci E, Shanske S, Moraes CT, Silvestri G, Hirano

M, Simonetti S, Angelini C, Donati A, Garcia C, Martinuzzi A,Mosewich R, Servidei S, Zammarchi E, Bonilla E, DeVivo DC,Rowland LP, Schon EA, DiMauro S: MELAS: clinical features,biochemistry, and molecular genetics. Ann Neurol 1992;31:391–398.

5. Dubowitz V: Histological and histochemical stains and reac-tions, in Dubowitz V (ed): Muscle Biopsy. London, BailliereTindall, 1985, pp 19–40.

6. Enriquez JA, Chomyn A, Attardi G: MtDNA mutation inMERRF syndrome causes defective aminoacylation of tRNALys

and premature translation termination. Nat Genet 1995;10:47–55.

7. Fukuhara N, Tokiguchi S, Shirakawa K, Tsubaki T: Myoclonusepilepsy associated with ragged-red fibres (mitochondrial ab-normalities): disease entity or a syndrome? J Neurol Sci 1980;47:117–133.

8. Goto Y, Koga Y, Horai S, Nonaka I: Chronic progressive ex-ternal ophthalmoplegia: a correlative study of mitochondrialDNA deletions and their phenotypic expression in musclebiopsies. J Neurol Sci 1990;100:63–69.

9. Goto Y: Clinical features of MELAS and mitochondrial DNAmutations. Muscle Nerve 1995;3(suppl):107–112.

10. Hasegawa H, Matsuoka T, Goto Y, Nonaka I: Strongly succi-nate dehydrogenase-reactive blood vessels in muscles frompatients with mitochondrial myopathy, encephalopathy, lac-tic acidosis, and stroke-like episodes. Ann Neurol 1991;29:601–605.

11. Hattori Y, Goto Y, Sakuta R, Nonaka I, Mizuno Y, Horai S:Point mutations in mitochondrial tRNA genes: sequenceanalysis and chronic progressive external ophthalmoplegia(CPEO). J Neurol Sci 1994;125:50–55.

12. Horai S, Hayasaka K: Intraspecific nucleotide sequence dif-ferences in the major noncoding region of human mitochon-drial DNA. Am J Hum Genet 1990;46:828–842.

13. Howard TJ, David JE, Veerabhadracharya BM, Andrew F:Nucleotide sequence and gene organization of sea urchinmitochondrial DNA. J Mol Biol 1988;202:185–217.

14. Koga Y, Nonaka I, Sunohara N, Yamanaka R, Kumagai K:Variability in the activity of respiratory chain enzymes in mi-tochondrial disorders. Acta Neuropathol 1988;76:135–141.

15. Larsson N-G, Clayton DA: Molecular genetic aspects of hu-man mitochondrial myopathies. Ann Rev Genet 1995;29:151–178.

16. Miller SA, Dykes DD, Polesky HF: A simple salting out proce-dure for extracting DNA from human nucleated cells. NuclAcids Res 1988;16:1215.

17. Moraes CT, Ricci E, Bonilla E, DiMauro S, Schon EA: Themitochondrial tRNALeu(UUR) mutation in mitochondrial en-cephalomyopathy, lactic acidosis, and strokelike episodes(MELAS): genetic, biochemical, and morphological correla-tions in skeletal muscle. Am J Hum Genet 1992;50:934–949.

18. Moraes CT, Schon EA: Detection and analysis of mitochon-drial DNA and RNA in muscle by in situ hybridization andsingle-fiber PCR, in Attardi AM, Chomyn A (eds): Methods inEnzymology. Oxford, Academic Press, 1996. vol 264, pp522–540.

19. Nakamura M, Nakano S, Goto Y, Ozawa M, Nagahama Y,Fukuyama H, Akiguchi I, Kaji R, Kimura J: A novel pointmutation in the mitochondrial tRNASer(UCN) gene detected ina family with MERRF/MELAS overlap syndrome. Biochem Bio-phys Res Commun 1995;214:86–93.

20. Ozawa M, Goto Y, Sakuta R, Tanno Y, Tsuji S, Nonaka I: The8,344 mutation in mitochondrial DNA: a comparison betweenthe proportion of mutant DNA and clinicopathologic find-ings. Neuromusc Discord 1995;5:483–488.

21. Ozawa M, Sano M, Goto Y, Nonaka I: The 8356 mutation ofmitochondrial DNA in patients with mitochondrial encepha-lomyopathy. Jpn J Hum Genet 1995;40(suppl):128.

22. Petruzzella V, Moraes CT, Sano MC, Bonilla E, DiMauro S,Schon EA: Extremely high levels of mutant mtDNAs co-localize with cytochrome c oxidase-negative ragged-red fibers

8363 Mutation in MERRF MUSCLE & NERVE March 1997 277

in patients harboring a point mutation at nt 3243. Hum MolGenet 1994;3:449–454.

23. Santrelli FM, Mak SC, El-Schahawi M, Casali C, Shanske S,Baram TZ, Madrid RE, DiMauro S: Maternally inherited car-diomyopathy and hearing loss associated with a novel muta-tion in the mitochondrial tRNALys gene (G8363A). Am J HumGenet 1996;58:933–939.

24. Seligman AM, Karnovsky MJ, Wasserkrug HL, Hanker JSV:Nondroplet ultrastructural demonstration of cytochrome oxi-dase activity with a polymerizing osmiophilic reagent, diami-nobenzidine (DAB). J Cell Biol 1968;38:1–14.

25. Shoffner JM, Lott MT, Lezza AMS, Seibel P, Ballinger SW,Wallace DC: Myoclonic epilepsy and ragged-red fiber disease(MERRF) is associated with a mitochondrial DNA tRNALys

mutation. Cell 1990;61:931–937.26. Silvestri G, Moraes CT, Shanske S, Oh SJ, DiMauro S: A new

mtDNA mutation in the tRNALys gene associated with myo-clonic epilepsy and ragged-red fibers (MERRF). Am J HumGenet 1992;51:1213–1217.

27. Steinberg S, Misch A, Sprinzi M: Compilation of tRNA se-quences and sequence of tRNA genes. Nucl Acids Res 1993;21:3011–3015.

28. Tanno Y, Yoneda M, Tanaka K, Kondo R, Hozumi I,Wakabayashi K, Yamada M, Fukuhara N, Ikuta F, Tsuji S: Uni-form tissue distribution of tRNALys mutation in mitochondrialDNA in MERRF patients. Neurology 1993;43:1198–1200.

29. Tokunaga M, Mita S, Murakami T, Kumamoto T, Uchino M,Nonaka I, Ando M: Single fiber analysis of mitochondrialmyopathy, encephalopathy, lactic acidosis, and stroke-like epi-sodes (MELAS). Ann Neurol 1994;35:413–419.

30. Yoneda M, Tanno Y, Horai S, Ozawa T, Miyatake T, Tsuji S: Acommon mitochondrial DNA mutation in the t-RNALys of pa-tients with myoclonus epilepsy associated with ragged-red fi-bers. Biochem Int 1990;21:789–796.

31. Yoneda M, Miyatake T, Attardi G: Complementation of mu-tant and wild-type human mitochondrial DNAs coexistingsince the mutation event and lack of complementation ofDNAs introduced separately into a cell within distinct organ-elles. Mol Cell Biol 1994;14:2699–2712.

32. Zeviani M, Muntoni F, Savarese N, Serra G, Tiranti V, CarraraF, Mariotti C, DiDonato S: A MERRF/MELAS overlap syn-drome associated with a new point mutation in the mitochon-drial DNA tRNALys gene. Eur J Hum Genet 1993;1:80–87.

278 8363 Mutation in MERRF MUSCLE & NERVE March 1997