Proc. Nati. Acad. Sci. USA Vol. 88, pp. 10614-10618, December 1991 Cell Biology

Introduction of disease-related mitochondrial DNA deletions into HeLa cells lacking mitochondrial DNA results in mitochondrial dysfunction (mitochondrial encephalomyopathy/mtDNA deletion/intercelhlular mtDNA transfer/mitochondrial dysfunction)

JUN-ICHI HAYASHI-*, SHIGEO OHTAt, AIKO KIKUCHIt, MASAKAZU TAKEMITSUt, YU-ICHI GOTOt, AND IKUYA NONAKAt *Department of Biochemistry, Saitama Cancer Center Research Institute, Ina-machi, Saitama 362, Japan; tDepartment of Biochemistry, Jichi Medical School, Minamikawachi-machi, Tochigi 329-04, Japan; and tDivision of Ultrastructural Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry, Kodaira, Tokyo 187, Japan

Communicated by Gottfried Schatz, August 1, 1991

Mutant mitochondrial DNA with large-scale ABSTRACT deletions (A-mtDNA) has been frequently observed in patients with chronic progressive external ophthalmoplegia (CPEO), a subgroup of the mitochondrial encephalomyopathies. To excdude involvement of the nuclear genome in expiression of the mitochondrial dysf~1nction characteristic of CPEO, we introduced the mtDNA of a CPEO patient into' clonal mtDNA-less HeLa cells and isolated cybrid clones. Quaqtitation' of A-mtDNA in the cybrids revealed that A-mtDNA was selectively propagated with higher levels of A-mtDNA correlating with slower cellular growth rate. In these cybrid clones, translational complementation of the missing tRNAs occurred only when A-mtDNA was 60% resulted in progressive inhibition of overall mitochondrial translation as well as reduction of cytochrome c oxidase a#ctivit'tthroughout the organelle population. Because these'cybrids shared the same nuclear background as HeLa cells, these results suggest that large-scale deletion mutations of mtDNA alone are sufficient for the mitochondrial dysfunction characteristic of CFEO.

Molecular evidence that mitochondrial DNA (mtDNA) mutations may-beassociated with human diseases was first reported by Holt et al. (1), who observed heteroplasmy of wild-type mtDNA (wt-mtDNA) and mutant mtDNA with a large-scale deletion (A-mtDNA) in the skeletal muscles of patients with mitochondrial encephalomyopathies. These diseases are usually defined'by morphological and functional abnormalities of muscle mitochondria, and they have been divided into three distinct subgroups (2): (i) chronic progressive external ophthalmoplegia (CPEO) including KearnsSayre syndrome, (ii) myoclonus -,epilepsy associated with ragged red fibers (MERRF), and (iii) mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke-like episodes (MELAS). Heteroplasmy of wt- and A-mtDNA has been frequently observed in CPEO includipg Kearns-Sayre syndrome (3, 4) and in Pearson syndrome (4, 5). Moreover, histochemical studies have suggested that deletion mutations of mtDNA are related to focal cytochrome c oxidase (COX) deficiency in the skeletal' muscles of CPEO patients (6-8). Recently, heteroplasmy of wt-mtDNA and mutant mtDNA with a point mutation in the mitochondrial tRNALYS and tRNAUbUR genes was shown to be closely associated with MERRF (9) and MELAS (10-12), respectively. However, there is as yet no convincing-evidence to explain how mutant mtDNA induces mitochondrial dysfunction, or whether the

mutations of mtDNA alone are sufficient to cause mitochondrial dysfunction in these diseases. In the present study, by cytoplasmically transmitting CPEO-derived mtDNA to clonal mtDNA-less HeLa EB8 cells, we found that A-mtDNA had a clear propagational advantage over wt-mtDNA and that its consequent accumulation to over 60o of the'total mtDNA resulted in a progressive inhibition of overall mitochondrial translation as well as a reduction of COX activity. As -the mtDNA-recipient EB8 cells were clonal and thus shared the same nuclear background, we conclude that large-scale deletion mutations of mtDNA induced the mitochondrial dysfunction characteristic of the disease directly without the help of any defects in the nuclear genome.

MATERIALS AND METHODS Cell Culture. mtDNA-less HeLa EB8 cells were isolated by the procedure of King and Attardi (13) through long-term treatment of HeLa cells with ethidium bromide (14). Primary skin fibroblasts were obtained from a 14-year-old CPEO patient who had clinical characteristics of Kearns-Sayre syndrome and had 97% A-mtDNA in skeletal muscles. The deletion was 51% base pairs (bp) long with a breakpoint between nucleotides 8563 and 13,758 (Y.-i.G., I.N., and S. Horai, unpublished data). The fibroblasts were used for enucleation after the 23rd passage in culture. Cells were grown in either a glucose-rich medium, RPMI+pyruvate [RPMI1640 (Nissui Seiyaku, Tokyo) containing glucose (2 mg/ml), pyruvate (0.1 mg/ml), and 10% fetal bovine serum], or a medium without glucose, DM170 [DM170 (Kyokuto Kagaku, Tokyo) containing galactose (0.9'mg/ml), pyruvate (0.5 mg/ml), and 10o fetal bovine serum]. Iolation of Cybrids. Intercellular transmission of mtDNA was carried out (15) by fusion of enucleated fibroblasts with EB8 cells, which were resistant to 6-thioguanine. In brief, the patient-derived fibroblasts of the 23rd passage grown on round disks were enucleated by centrifugation'(23,000 x g at 34WC for 10 min) in the presence of cytochalasin B (10 ,ug/ml). By this treatment, >85% of the cells were enucleated. Then, the cytoplasts were mixed and fused with EB8 cells in the presence of 50o (wt/vol) polyethylene glycol 1500 (Boehringer Mannheim). After the fusion, the residual nonenucleated parental fibroblasts and hybrids (fibroblast-EB8) were completely eliminated in a selective medium with 6-thioguaAbbreviations: CPEO, chronic progressive external ophthalmoplegia; MERRF, myoclonus epilepsy associated with ragged red fibers; MELAS, mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke-like episodes; COX, cytochrome c oxidase; wt-mtDNA, wild-type mitochondrial DNA; A-mtDNA, mitochondrial DNA with large deletion(s).

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

10614

Cell

Proc. Natl. Acad. Sci. USA 88 (1991)

Biology: Hayashi et al.

nine. In this mtDNA transfer system, however, as no method for genetic selection to remove parental EB8 cells was available, the colonies that grew in the medium with 6-thioguanine were either cybrids or the parental EB8 cells. However, growth of cells with mtDNA (cybrids) is expected to be slightly better than that of mtDNA-less cells, so for preferential selection of cybrids we picked up 127 colonies growing relatively fast. Of 127 clones, 6 clones had mtDNA. mtDNA Analysis. The total DNA (2-5 jig) extracted from 2 x 105 cells was digested with the single-cut restriction enzyme Pvu II. The fragments separated by 0.8% agarose gel electrophoresis were then transferred to nitrocellulose membranes and hybridized with [a-32P]dCTP-labeled HeLa mtDNA. The membranes were washed and exposed to x-ray film (Fuji-RX, Fuji Film, Tokyo) for 1 hr at -80'C. To quantitate the A-mtDNA contents in the cybrid clones, the membranes were exposed to imaging plates (Fuji Film) for 5 min, and the radioactivity of each fragment was measured using a bioimaging analyzer, Fujix BAS 2000 (Fuji Film). Growth Curves. Cells were plated at 104 cells per dish (50 mm diameter) in RPMI+pyruvate medium and in DM170 medium, and the cells in individual dishes were counted at intervals. Analysis of Mitochondrial RNA. Total RNA (10 jug/lane) separated by electrophoresis in a 1.5% agarose/formaldehyde gel was transferred to nylon membranes and hybridized with five different antisense oligonucleotide probes labeled at the 5' end with [y-32P]ATP: 12S rRNA (nucleotide positions 1280-1261); ND1 (3740-3717); ATPase8 (8435-8414); ND4 (11,186-11,167); and ND5 (13,970-13,954). The radioactivCybrid Clones

Parents

A

Fb EB 16kbp

I-

1 2 3 4 5 6 t

40

"

10 Om 0

mtDNA

contents (%) 17 B

B

Fusion genes

0 53 0 0 34 47

-

A*77 amino acids

B*141 amino acids

RESULTS AND DISCUSSION Cytoplasmic Transfer of CPEO-Derived mtDNA to mtDNALess HeLa Cells. Cytoplasmic transfer of fibroblast mtDNA from a CPEO patient to mtDNA-less HeLa EB8 cells was achieved by fusing enucleated fibroblasts with the EB8 cells. Of the six cybrid clones containing mtDNA, three had only wt-mtDNA, while the other three also contained various proportions of A-mtDNA (Fig. 1). The restriction patterns of Hha I, which can distinguish HeLa mtDNA from other human mtDNA (20, 21), showed that both wt- and A-mtDNA w

a)

0 89

C

0

0

~~~~~~5196 astPArg

bp deletion tNSerLu PB tRNARrLeu ATPase8IR_ NAGIY tELNA~ tRNA ATPaCse t Ii ND3 ND4L ND4 tRN ISN

ATPase8/ND5 ATPase6/ND5 Fusion mRNA 1292 nucleotides Fusion proteins

ities of the transcripts were measured with the bioimaging analyzer. Analysis of Mitochondrial Translation Products. [35S]Methionine labeling of the mitochondrial translation products was carried out as described (16). In brief, cells (4 x 106) in a dish were incubated for 2 hr with [35S]methionine in the presence of emetine (0.2 mg/ml), and the mitochondrial fraction was obtained by homogenization in 0.25 M sucrose/1 mM EGTA/10 mM Hepes-NaOH, pH 7.4, followed by differential centrifugation. Proteins of the mitochondrial fraction (50 jig per lane) were separated by SDS/urea/ polyacrylamide gel electrophoresis. For quantitative estimation of [35S]methionine-labeled polypeptides, the dried gel was exposed to the imaging plate for 6 hr, and the radioactivities of total and individual polypeptides were measured with the bioimaging analyzer. COX activity was measured in the mitochondrial fraction as described (17). COX Electron Micrographs. Cultured cells were fixed in 2% glutaraldehyde/0.05 M phosphate buffer, pH 7.4, for 15 min and stained with COX (18). The cells were then postfixed in OS04 for 15 min and embedded in epoxy resin.

A 110/uun

a.

M

llkbp

A

1' 5'

; e:~a~e

i

10615

50-

5

z 0

_ ND5

ATPase6

ND6 4)

Go i8 12,1,j3,4,16 20 4! (weeks) Fusion Time after

rO %R

ATPaseB/ND5 frame shift ATPSe6D5

FIG. 1. (A) Southern blot analysis of Pvu II restriction patterns of mtDNA from parent cells and their cybrid clones. Lanes: Fb, fibroblasts isolated from the CPEO patient; EB, mtDNA-less HeLa EB8 cells; 1-6, cybrid clones 1-6, respectively, in week 4 after fusion. Total DNA (2-5 gg per lane) extracted from the cells was analyzed. Lane 1', cybrid clone 1 in week 13 after fusion; lane 5', cybrid clone 5 in week 10 after fusion. For quantitation of mtDNA in lanes 1' and 5', Pvu II digests of an equal amount of total DNA (5 jig per lane) were analyzed. (B) Transcription and translation maps around the region of the deletion breakpoint. In the maps, the lengths of the mtDNA genes are not proportional to their real lengths. The deletion was 5196 bp long with a breakpoint between nucleotides 8563 and 13,758. Nucleotide positions are based on the notation of Anderson et al. (19). This large-scale deletion mutation is supposed to make two fusion genes, ATPase8/ND5 and ATPase6/ND5. The transcripts should be 1292 nucleotides long, containing parts of the ATPase8, ATPase6, and ND5 genes and the antisense RNA sequence of ND6. The transcripts have translation initiation sites of the ATPase8/ND5 fusion gene and ATPase6/ND5 fusion gene at 8366 and 8527, respectively. In the ATPase6/ND5 fusion gene, translation was terminated at 13,792 due to a frameshift following the deletion breakpoint, making fusion protein A with 77 amino acids, whereas the ATPase6/ND5 fusion gene was translated entirely without a frameshift, making fusion protein B with 141 amino acids.

B .8E z z

:+glucose

:+pyruvat rC

+galactose + pyruvate

10 /

A

1

0 Os

2 4 6 8 10 Time (days)

2 4 6 8 10 Time (days)

FIG. 2. (A) Propagation profiles of A-mtDNA in cybrid clones 2, 5, and 6. *, Cybrid clones cultivated in glucose-rich medium, RPMI+pyruvate; o, clone 5 cultivated in medium without glucose, DM170; A, subclones of clone 5. (B) Effects of A-mtDNA contents and medium on cell growth. Cell growth was examined using three subclones of clone 5 with the lowest (72%, A, arrow A), intermediate (78%, A, arrow B), and highest (93%, A, arrow C) contents of A-mtDNA. HeLa cells (o), mtDNA-less HeLa EB8 cells (o), and cybrid clone 1 without A-mtDNA (e) were used as controls. Cells were grown in RPMI+pyruvate medium (+glucose, +pyruvate; Left) or DM170 medium (+galactose, +pyruvate; Right).

10616

Cell Biology: Hayashi et al.

Proc. Natl. Acad. Sci. USA 88 (1991) O loo.12SRNA

B

A Probes 1 2 3 4 5 EB He I 12S rRNA PI

75v .\;



"

-ATP8

-ND4L

-I -. PrtenB

X

.

I

0 25

-ATP65 Fusion

~Cytb

-

I

C

vf

80 80 100 40 Relative mtDNA Contents (%)

- 0- Fiv 75

Total Inorporation

i 50 F

.2

£ 25i

B

o

75 R

50 2. x

COX Activt .

.

25 8 .

.

0 80 100 40 Relative a mtDNA Contents (%)

FIG. 4. Analysis of mitochondrial translation and COX activity in cells with various A-mtDNA contents. (A) Protein synthesis in mitochondria. Lanes: 1, cybrids containing 0o A-mtDNA (clone 1); 2, 55% A-mtDNA (clone 5); 3, 61% A-mtDNA (clone 2); 4, 67% A-mtDNA (clone 2); 5, 78% A-mtDNA (clone 5); EB, mtDNA-less HeLa EB8 cells; He, HeLa cells. After [35S]methionine labeling of mitochondrial translation products, proteins of the mitochondrial fraction (50 jug per lane) were separated by SDS/urea/polyacrylamide gel electrophoresis. ND5, COI, ND4, Cytb, ND2, ND1, COII, COIII, ATP6, ND3, ATP8, and ND4L are polypeptides assigned to mtDNA genes. Positions of the predicted fusion proteins A (solid arrowhead) and B (open arrowhead) are indicated (see Fig. 1B). (B and C) Quantitative estimation of mtDNA translation products and COX activity. In the case of the fusion protein B, cells with 55% A-mtDNA were assumed to have 55% relative content of the fusion protein B.

Cell

Biology: Hayashi et al.

2A). To disclose this threshold, we isolated six subclones 19 weeks after clonal isolation of clone 5. The variation in A-mtDNA contents observed in the subclones was biased to higher contents of A-mtDNA (Fig. 2A) and the subclones with higher contents grew poorly (Fig. 2B). Such A-mtDNAbiased partitioning of mtDNA could be due to its propagational advantage. Thus, it appears that A-mtDNA can accumulate up to a threshold, beyond which a further increase in A-mtDNA is selected against for reasons most likely associated with cell growth. The threshold decreased to 50% in medium without glucose, which was lethal to cells with >72% A-mtDNA as well as to EB8 cells (Fig. 2). This implies that the thresholds in dividing tissues in vivo may also vary depending on environmental factors, such as the sorts of energy sources and their concentrations. The threshold in the in vivo fibroblasts would be below 34-53% (Fig. 2A), if one assumes that A-mtDNA also accumulated during the 23 passages of the primary culture before fusion. This would enable cells without A-mtDNA to develop even when random partitioning of mtDNA was biased to A-mtDNA. As cells without A-mtDNA grew faster (Fig. 2B), as suggested by Moraes et al. (22), dividing tissues with low thresholds of A-mtDNA would be able to eliminate A-mtDNA. In contrast, dividing tissues with high thresholds can accumulate A-mtDNA as skeletal muscle does in CPEO. Thus, in the human disease state, diversity in both of the cellular microenvironment and the cellular requirement of energy might permit different thresholds with subsequent diverse effects on cellular physiology. The propagation profiles of A-mtDNA in dividing cells (Fig. 2) explain well the "progressive nature" of the disease, and this could be a good model to account for the propagation of A-mtDNA in the Pearson syndrome, where 80-90%o A-mtDNA has been observed in bone marrow cells and lymphocytes (23). Effects of A-mtDNA Proportions on Transcription, Translation, and COX Activity. We analyzed the influence of mtDNA genotypes (the ratio of A-mtDNA to wt-mtDNA) on mitochondrial transcription and translation and also studied their effect on the phenotypic expressions characteristic of the disease by using cybrids with various contents of A-mtDNA. Northern blot analysis (Fig. 3) showed that as the contents of A-mtDNA increased, the contents of the transcripts of ND4 and ATPase6/8 genes missing in A-mtDNA decreased correspondingly. The transcripts of the putative "fusion genes" of length 1.3 kbp (see Fig. 1B) increased proportionally with the increase of A-mtDNA. All the mRNAs and rRNAs tested were transcribed almost proportionally to the contents of their corresponding genes (Fig. 3B). This is consistent with the previous observations that A-mtDNA was transcriptionally active (6, 7, 24). Next we analyzed the mitochondrial translation products by [35S]methionine labeling. The relative mobilities of 12 polypeptides in HeLa cells and cybrids containing only wt-mtDNA (Fig. 4A) were almost identical with those in HeLa cells reported by Chomyn et al. (25), and these 12 polypeptides could thus be assigned to mtDNA genes. One additional polypeptide, which corresponded well to the predicted fusion protein B (see Fig. 1B), was observed in cells with 55% and 61% A-mtDNA (Fig. 4A). Moreover, the significant increase in radioactivity of the ATPase8 band exclusively in cells with 55% A-mtDNA suggested that this increase was due to comigration of the predicted fusion protein A with the ATPase polypeptide. Thus, these polypeptides could be the fusion proteins, although their amino acid sequences were not determined. Quantitative estimation of mtDNA translation products and COX activity showed that both overall translation and COX activity decreased, particularly when >60% A-mtDNA

~ .mlp.

10617

Proc. Natl. Acad. Sci. USA 88 (1991)

A..

N~~~

A~~~~~~~~A

AII&

41* cybnd clone (w:th175%" 5 A-mtDNA) A-mtDNA)M(A)Land

mitochondiainsW-cybrid clone

5 were

COX-negative,

.

whe 9

FIG5. CO electro migah of 'yn clon 1wthu

tRNA ccursonlyblwteifecinpito missin inlcto 0 (Fig. 411 and C)'.. Belo th point' had accumulated

A-mtDNA whereasDti otaov hslvl a hc on transaio for available tRN~~~~~ for a.ocr competition mrprsoteybins ecloded exclthu4)G.Moeoer thepuetatvrfson B.And sivelyby (A-mDN were rdetectabe incelswith-mDA55% These floeatuwres idcat-e tat i-tnA 61%conri smia er though1 o eeven "RN cmpemnttin"apeas Th bytmio CXprosteinseNoded fbivloe tRN genes AmitDNAhadri los A-mtndral wnere temransaedwith the hepiofthee orrelysponingd the inflection pointteras Blo acmlaevel were 4 above hamdN elowe thebtA waft potens 6% mdecreased sigifcntlyensasila correspondedwl ing genei contentls(Fig.s11. Thisd leelns ohi coulMoexplain why Nuakaseetio alp2)riotefindeviodenc forlu A-mDN abov in cells with taslational complAeementation ndicprovidesevidenceug A-mtDNA) 7% ahseaue th6nlcio1on for the hyp othesisvpropose gbynShoubrodgeetsal.c(7)dtha 90% with hel-mftDNAThusonin -mtDNAineusce fibranslaegmnt it cnbe tasupposedthtfranslationalcomp ementaio ofwhen abeowethe inflection point, ofetran% -misigtRNA occurs only whic pobeointh abv thsilevlat ArmtDinA, wherease sitnislostlyi

imDay aocur competitiona for valaleetRAsi for translation

ithcne"tRNA

appar tobevrysmia complementation" ha rasatoa

to the "rRNA complementation" observed in cells with various proportions of chloramphenicol-resistant and chlor-

10618

Cell Biology: Hayashi et al.

amphenicol-sensitive mtDNA (26). However, such translational complementation and competition could not occur if wt- and A-mtDNA or chloramphenicol-sensitive and chloramphenicol-resistant mtDNA were segregated in different mitochondria. COX electron micrographs (Fig. 5) clearly showed that individual mitochondria in cells with 75% A-mtDNA had lost COX activity uniformly and were considerably swollen, with disorganized inner membranes. In addition, the COX activity was recovered in most of mitochondria in the cells with 60%o, mitochondrial translation may become limiting, and the translation phase may be shifted from complementation to competition, resulting in progressive inhibition of overall mitochondrial translation as well as reduction of COX activity. In this study, we excluded the influence of any nuclear gene on the mitochondrial dysfunction by introducing the mtDNA of a CPEO patient into clonal EB8 cells. Thus, the cybrids shared the same nuclear background and consequently the results showed that large-scale deletion mutations of mtDNA alone were sufficient for the observed respiration defects. A possibility cannot be excluded completely that the A-mtDNA possesses one or more additional mutations and that these also contribute to the observed changes. However, even if this is the case, it is concluded that accumulation of A-mtDNA induces the mitochondrial dysfunction characteristic of CPEO. Recently, some point mutations of mitochondrial tRNA genes have been shown to be closely associated with mitochondrial encephalomyopathies other than CPEO (9-12). Our mtDNA transmission system would also be applicable for studying the roles of these tRNA mutations. More recently, Chomyn et al. (27) have reported cotransfer of a MERRF-associated tRNALYS mutation and respiration defects into mtDNA-less human cells. We are grateful to Dr. S. Horai for communicating mtDNA sequence data prior to their publication. We thank Dr. Y. Kagawa and Dr. K. Nakachi for discussion and critical reading of the manuscript. This work was partly supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture, Japan. 1. Holt, I. J., Harding, A. E. & Morgan-Hughes, J. A. (1988) Nature (London) 331, 717-719. 2. DiMauro, S., Bonilla, E., Zeviani, M., Nakagawa, M. & DeVivo, D. C. (1985) Ann. Neurol. 17, 521-538. 3. Zeviani, M., Moraes, C. T., DiMauro, S., Nakase, H., Bonilla, E., Schon, E. A. & Rowland, L. P. (1988) Neurology 38, 1339-1346. 4. McShane, M. A., Hammans, S. R., Sweeney, M., Holt, I. J.,

Proc. Natl. Acad. Sci. USA 88 (1991)

5. 6. 7. 8. 9. 10.

11. 12. 13. 14.

15. 16. 17. 18. 19.

20. 21.

22.

23. 24. 25. 26. 27.

Beattie, T. J., Brett, E. M. & Harding, A. E. (1991) Am. J. Hum. Genet. 48, 39-42. Rotig, A., Colonna, M., Bonnefont, J. P., Blanche, S., Fischer, A., Saudubray, J. M. & Munnich, A. (1989) Lancet i, 902-903. Mita, S., Schmidt, B., Schon, E. A., DiMauro, S. & Bonilla, E. (1989) Proc. Natl. Acad. Sci. USA 86, 9509-9513. Shoubridge, E. A., Karpati, G. & Hastings, E. M. (1990) Cell 62, 43-49. Goto, Y., Koga, Y., Horai, S. & Nonaka, I. (1990) J. Neurol. Sci. 100, 63-69. Shoffner, J. M., Lott, M. T., Lezza, A. M. S., Seibel, P., Ballinger, S. W. & Wallace, D. C. (1990) Cell 61, 931-937. Goto, Y., Nonaka, I. & Horai, S. (1990) Nature (London) 348, 651-653. Kobayashi, Y., Momoi, M. Y., Tominaga, K., Momoi, T., Nihei, K., Yanagisawa, M., Kagawa, Y. & Ohta, S. (1990) Biochem. Biophys. Res. Commun. 173, 816-822. Kobayashi, Y., Momoi, M. Y., Tominaga, K., Shimoizumi, H., Nihei, K., Yanagisawa, M., Kagawa, Y. & Ohta, S. (1991) Am. J. Hum. Genet. 49, 590-599. King, M. P. & Attardi, G. (1989) Science 246, 500-503. Hayashi, J.-I., Yonekawa, H., Watanabe, S., Nonaka, I., Momoi, M. Y., Kagawa, Y. & Ohta, S. (1991) in Progress in Neuro-pathology, eds. Sato, T. & DiMauro, S. (Raven, New York), Vol. 7, pp. 93-102. Hayashi, J.-I., Yonekawa, H. & Tagashira, Y. (1989) Cancer Res. 49, 4715-4720. Mariottini, P., Chomyn, A., Riley, M., Cottrell, B., Doolittle, R. F. & Attardi, G. (1986) Proc. Nati. Acad. Sci. USA 83, 1563-1567. Orii, Y. & Okunuki, J. (1965) J. Biochem. (Tokyo) 58, 561-568. Seligman, A. M., Karnovsky, M. J., Wasserkrug, H. L. & Hanker, J. S. (1968) J. Cell Biol. 38, 1-14. Anderson, S., Bankier, A. T., Barrell, B. G., deBruijn, M. H. S., Coulson, A. R., Drouin, J., Eperon, I. C., Nierlich, D. P., Roe, B. A., Sanger, F., Schreier, P. H., Smith, A. J. H., Staden, R. & Young, I. G. (1981) Nature (London) 332, 548551. Hayashi, J.-I., Werbin, H. & Shay, J. W. (1986) Cancer Res. 46, 4001-4006. Hayashi, J.-I., Yonekawa, H., Murakami, J., Tagashira, Y., Pereira-Smith, 0. L. & Shay, J. W. (1987) Exp. Cell Res. 172, 218-227. Moraes, C. T., Schon, E. A., DiMauro, S. & Miranda, A. F. (1989) Biochem. Biophys. Res. Commun. 160, 765-771. Rotig, A., Cormier, V., Blanche, S., Bonnefont, J. P., Ledeist, F., Romero, N., Schmitz, J., Rustin, P., Fischer, A., Saudubray, J. M. & Munnich, A. (1990) J. Clin. Invest. 86,1601-1608. Nakase, H., Moraes, C. T., Rizzuto, R., Lombes, A., DiMauro, S. & Schon, E. A. (1990) Am. J. Hum. Genet. 46, 418-427. Chomyn, A., Mariottini, P., Cleeter, M. W. J., Ragan, C. I., Matsuno-Yagi, A., Hatefi, Y., Doolittle, R. F. & Attardi, G. (1985) Nature (London) 314, 592-597. Wallace, D. C. (1982) in Techniques in Somatic Cell Genetics, ed. Shay, J. W. (Plenum, New York), pp. 159-187. Chomyn, A., Meola, G., Bresolin, N., Lai, S. T., Scarlato, G. & Attardi, G. (1991) Mol. Cell. Biol. 11, 2236-2244.