A single cell complementation class is common to several cases of cytochrome c oxidase-defective Leigh's syndrome
A single cell complementation class is common to several cases of cytochrome c oxidase-defective Leigh's syndromeMonica Munaro1, Valeria Tiranti1, Doriana Sandonà2, Eleonora Lamantea1, Graziella Uziel3, Roberto Bisson2 and Massimo Zeviani1,4,*
1Division of Biochemistry and Genetics and 3Division of Child Neurology, National Neurological Institute `C. Besta', via Celoria 11, 20133 Milan, Italy, 2Department of Experimental Biomedical Sciences, State University School of Medicine, via Trieste 75, 35121 Padua, Italy and 4Division of Molecular Medicine, Children's Hospital `Bambino Ges'-IRCCS, Piazza S. Onofrio 4, 00165 Rome, Italy
Received August 2, 1996;Revised and Accepted November 15, 1996
A generalized defect of complex IV (cytochrome c oxidase, COX) is frequently found in subacute necrotizing encephalomyelopathy (Leigh's syndrome), the most common mitochondrial disorder in infancy. We previously demonstrated the nuclear origin of the COX defect in one case, by fusing nuclear DNA-less cytoplasts derived from normal fibroblasts with mitochondrial DNA (mtDNA)-less transformant fibroblasts derived from a patient with COX-defective [COX(-)] Leigh's syndrome. The resulting cybrid line showed a specific and severe COX(-) phenotype. Conversely, in the present study, we demonstrate that a COX(+) phenotype could be restored in hybrids obtained by fusing COX(-) transformant fibroblasts of seven additional Leigh's syndrome patients with mtDNA-less, COX(-) tumor- derived rhoo cells. Both these results are explained by the presence of a mutation in a nuclear gene. In a second set of experiments, in order to demonstrate whether COX(-) Leigh's syndrome is due to a defect in the same gene, or in different genes, we tested several hybrids derived by fusing our original COX(-) cell line with each of the remaining seven cell lines. COX activity was evaluated in situ by histochemical techniques and in cell extracts by a spectrophotometric assay. No COX complementers were found among the resulting hybrid lines. This result demonstrates that all our cases were genetically homogeneous, and suggests that a major nuclear disease locus is associated with several, perhaps most, of the cases of infantile COX(-) Leigh's syndrome. This information should make it easier to identify the gene responsible.
In principle, faulty oxidative phosphorylation (OXPHOS) may be due to mutations in (i) nuclear or mitochondrial genes involved in the synthesis of individual respiratory subunits; (ii) nuclear genes involved in their import, assembly and function; and (iii) nuclear genes controlling the perpetuation, propagation and expression of the mitochondrial genome (1 ).
The most relevant contribution to the understanding of mitochondrial disorders has come from the discovery of an impressive, and ever expanding, number of disease-associated mutations affecting mitochondrial DNA (mtDNA) genes that either control the synthesis of, or code for, the 13 respiratory subunits specified by the mitochondrial genome (2 ). In addition, the existence of Mendelian disorders associated with re-arrangements in or decreased content of mtDNA (3 ) are proposed to be caused by mutated, trans-acting nuclear-coded factors, ultimately affecting the structural integrity or the copy number of mtDNA (4 ,5 ).
On the other hand, defects of nuclear genes involved in the synthesis and function of OXPHOS enzymes are believed to be responsible for familial cases compatible with autosomal recessive inheritance, and/or cases characterized by severe enzyme defects not associated with known mutations of mtDNA. However, with the exception of a single report of a defect of complex II due to a point mutation in a nuclear succinate dehydrogenase (SDH) gene (6 ), the attribution of these disorders to nuclear gene defects remains speculative, and the current classification is based on biochemical findings only.
Subacute necrotizing encephalomyelopathy, also known as Leigh's syndrome, is probably the most common disorder of the respiratory chain in infancy and childhood. Biochemically, a generalized defect of respiratory complex IV (cytochrome c oxidase, COX) is found in the majority of patients (7 ). However, deficiencies of pyruvate dehydrogenase complex (8 ) or respiratory complex I (NADH-coenzyme Q reductase, NADH-CoQ RD) (9 ,10 ), or the presence of mtDNA point mutations (11 -13 ) have also been reported.
We have demonstrated previously the nuclear origin of the defect in one COX-defective [COX(-)] Leigh's syndrome patient, by fusing nuclear DNA-less cytoplasts derived from normal fibroblasts with mtDNA-less transformant fibroblasts derived from the proband (14 ). The resulting cybrid line showed a specific and severe COX(-) phenotype, indicating that the nuclear complement contained the disease gene. We present here an alternative strategy, addressing the same question, based on the ability of a COX(-), mtDNA-less 143B-derived rhoo cell line to complement the COX(-) phenotype of transformant fibroblasts derived from seven additional Leigh's syndrome patients.
Moreover, to gain further insight into the genetic complexity of COX(-) Leigh's syndrome, we present here a strategy aimed at demonstrating whether individual cases of COX(-) Leigh's syndrome are due to a defect in the same gene, or in different genes. We created hybrids obtained by fusing the COX(-) cell lines of our original patient (14 ) with each of the other seven COX(-) cell lines. After selection by a double-antibiotic system, the presence of different cell complementation classes was evaluated in the heterokaryon hybrids by COX-specific histochemical and biochemical assays.
The puromycin-resistant (Puror) transformant fibroblasts from seven of our patients were polyethyleneglycol (PEG)-fused with the COX(-), neomycin-resistant (Neor) SFT.12-Neor transformant fibroblast clone derived from our eighth patient (see Materials and Methods). The SFT.12-Neor clone was used previously to demonstrate the nuclear origin of the COX defect in this patient (14 ).
Likewise, each of the the seven Puror transformant cell lines was PEG-fused with 143B-derived rhoo cells that were made Neor by transfecting plasmid pcDNA3 (143B.rhoo-Neor).
In both experiments, hybrids expressing both antibiotic resistances were selected against the parental cell lines by adding the neomycin analog G-418 and puromycin to the growth medium. A similar procedure was carried out to select the `control' hybrid Hy[C+fAR], generated by fusing Puror transformant cells from patient C with a Neor transformant fibroblast cell line derived from a normal subject.
After selection of cell hybrids, in order to verify the origin of the nuclear genomes of each hybrid line, we used a highly polymorphic repetitive sequence marker at locus D11S533 (14 ). An example of this analysis is shown in Figure 1 : the nuclear genotypes specific to the parental cells (in this case SFT.12-Neor, lane 1, versus BFT.13-Puror, lane 2, and SaFT.1-Puror, lane 3) were present in the corresponding hybrids Hy[S+B] lane 4, and Hy[S+Sa] lane 5.
The results of our biochemical assays are shown in Table 1 . All of the eight transformant fibroblast cell lines of our patients showed a marked reduction of COX activity, to ~5-10% of the controls' mean. No COX or complex I activity was found in the 143B.rhoo-Neor cells. As expected from the results of histochemistry, a defect of COX activity comparable with that of the parental cell lines was detected in all of the patient-derived hybrids (Table 0 ). By contrast, normal COX activities were obtained in the five rhoo-derived hybrids that were tested after 2 weeks of antibiotic selection. Normal COX activities were also obtained in fAR.12-Neor control fibroblast transformants, as well as in the original 143B.TK- cells. Complex I was found to be consistently normal in all cases, as was SDH in all of the cases that were tested for this activity.
Leigh's syndrome is the most common clinical presentation of COX deficiency in infancy (6 ). COX is composed of 13 protein subunits, three of which are encoded by the mitochondrial genome, while the remaining 10 are encoded by nuclear genes. The mitochondrially encoded subunits of COX (subunits I, II and III) provide the catalytic core of the enzyme, and form the transmembrane backbone essential for the proper assembly of the complex. By contrast, the functions of the nuclear DNA-encoded subunits have not been fully elucidated, but they may modulate the catalytic activity or the biosynthesis of COX, which is the central enzyme in the control of OXPHOS (15 ,16 ). Two subunits (subunits VIa and VIIa) are present as tissue-specific isoforms in humans, while no evidence of tissue-specific isoforms has been provided for the other eight subunits (17 ).
Several studies have tried in the past to characterize the structure, expression and kinetic properties of the defective enzyme in Leigh's syndrome (7 ,18 -20 ). However, the gene, or genes, responsible for this disease is/are still unknown. Given the double origin of the defective enzyme, and the extraordinary complexity of its genetic control (21 ,22 ), as indicated by several studies in yeast (23 -25 ) and other organisms (26 ,27 ), several possibilities can be hypothesized. These include mutations of either mtDNA-encoded (28 ) or nucleus-encoded COX subunits, or mutations in the gene repertoire controlling the expression, targeting and assembly of the enzyme.
On the basis of these considerations, and preliminary to the identification of the genetic basis of this important OXPHOS disorder, two fundamental questions still remain open. (i) Is the disease gene(s) part of the mitochondrial or of the nuclear gene complement? (ii) Is COX(-) Leigh's syndrome a genetically homogeneous or a genetically heterogeneous clinical entity?
In order to contribute to answering these questions, we have used cell culture systems based on the SV40 transformation of eight fibroblast cell lines, derived from as many Leigh's syndrome patients. These cell lines all expressed a clearcut deficiency of COX, as demonstrated by both histochemical and biochemical techniques.
In all of our eight cell lines, the gene responsible for the COX defect was nuclear, for the following reasons.
(i) The parental cell line that was common to all of the hybrids (SFT.12-Neor) belonged to a patient in whom the nuclear origin of the OXPHOS defect was proven in a previous work by using patient-derived rhoo cells (14 ). The latter provided the nuclear complement, while normal COX(+) cells provided the mitochondrial complement, for the creation of transmitochondrial cybrids. The cybrids resulted COX(-) by both histochemical and biochemical assays. These data were explained by the presence of a nuclear gene mutation capable of disrupting the OXPHOS competency of normal mitochondria.
(ii) Each of the remaining seven patient-derived parental cell lines were fused with a COX(-), mtDNA-less rhoo clone derived from 143B.TK- tumoral cells. The patients' cell lines were all made Puror, while the 143B.rhoo cells were made Neor, by transfection of suitable recombinant vectors expressing the corresponding resistance-conferring gene. The selection of the heterokaryon hybrids was obtained by prolonged treatment with both G-418 and puromycin: only hybrids derived from the fusion of a Neor parental cell with a Puror parental cell could survive this treatment, while both unfused parental cells and homokaryon hybrids were eliminated by exposure to the toxic effect of the one antibiotic for which they lacked resistance. After selection, heterokaryons were demonstrated by DNA genotyping of the highly polymorphic locus D11S533. Hybrids were then tested for the restoration of COX competency by COX-specific histochemistry in situ, and by a COX-specific biochemical assay on cell homogenates. Hybrids were consistently COX(+), indicating that (a) the 143B.rhoo nuclear genome was able to complement the OXPHOS defect and (b) the nuclear gene mutation behaved as a recessive trait in all cases. An identical result was obtained in the hybrid Hy[C+fAR], generated by fusing a COX(-), Puror cell line with a COX(+), Neor normal cell line, again indicating the recessive behavior of the COX defect. In an early report, Miranda and co-workers (29 ) showed that the COX(-) phenotype of cells derived from a Leigh's syndrome case could be corrected by fusing them with a HeLa cell variant. This result suggested the nuclear origin of the primary defect in their case. However, complementation of the COX defect by the residual HeLa mtDNA could not be firmly excluded, and the study was performed in a single patient. In our system, the only contribution to the correction of the COX(-) phenotype could derive from the nucleus of our rhoo cells, since these cells completely lack mtDNA and are themselves OXPHOS-incompetent.
In order to answer this question, we chose a strategy based on the identification of cell complementation classes. Using double-antibiotic selection, we created cell hybrids derived from one COX(-), Neor transformant fibroblast cell line, fused with each of seven different COX(-), Puror transformant cell lines.
In contrast to the results in patient-rhoo hybrids, histochemical and biochemical analyses showed that our patient-patient hybrids maintained the COX(-) phenotype of the parental cell lines, indicating that no complementation of the COX defect occurred in any case.
These data demonstrate that our eight Leigh's syndrome cases all belong to a single cell complementation class, suggesting genetic homogeneity of the trait in our series.
Alternative hypotheses to explain the maintenance of the COX defect in the presence of genetic heterogeneity are unlikely. For instance, the possibility that the COX defect cannot be complemented because of the presence of a dominant mutation is difficult to reconcile with the restoration of a COX(+) phenotype in the rhoo-derived hybrids, and with the maintenance of the COX(+) phenotype associated with the normal parental cell line fAR.12-Neor in the Hy[C+fAR] hybrid. Likewise, the possibility that the complementing gene(s) can be eliminated by partial chromosomal losses and genetic rearrangements in the patient-derived hybrids is difficult to reconcile with the uniformity of the COX(-) phenotype visualized cytochemically, and with the uniform COX(+) phenotype in all of the rhoo-derived hybrids as well as in the Hy[C+fAR] `control' hybrid.
Therefore, we propose that a single, identical gene is responsible for the COX(-) phenotype common to all of our patient-derived cell lines.
From a clinical point of view, six patients were diagnosed as having a typical COX(-) Leigh's syndrome, on the basis of common clinical, neuroradiological and morphological features. These included: early onset, rapidly progressive course, generalized hypotonia, pyramidal tract signs, cerebellar dysfunction, prominent abnormalities of eye and respiratory movements and symmetric necrotic lesions extending from the basal ganglia to the medulla oblongata. Interestingly, however, two patients (individuals B and Sa) were diagnosed as having `atypical' or `variant' Leigh's syndrome, according to the conclusions of two different clinical teams (30 ) (see Materials and Methods). Onset of the disease was slightly delayed in both, pyramidal tract signs were absent in one case and mild in the other and abnormal respiratory movements were absent. In one patient (individual B), ataxia and other cerebellar signs were absent, while myopathic symptoms were prominent in both.
Nevertheless, the results of our complementation experiments indicate that the variant cases and the typical cases were all genetically homogeneous. Clinical variations, overlap syndromes or different clinical entities in the presence of the same mutations are not uncommon in mitochondrial disorders (1 ). On the other hand, it is well known that different mutations in the same gene can account for the wide spectrum of clinical presentations and clinical severity frequently found in many hereditary disorders, especially those due to inborn errors of metabolism (1 ).
On the basis of these findings, we propose the existence of a major disease locus responsible for several, maybe for most, of the cases of infantile COX(-) Leigh's syndrome, although the existence of other disease loci associated with either Leigh's syndrome, or other less defined generalized defects of COX, is by no means excluded. For instance, it would be interesting to evaluate the existence of COX complementation between our patients' series and variants of Leigh's syndrome that recently have been identified in genetically isolated populations (31 ), and that show peculiar biochemical features of the COX defect (20 ).
The method described here, based on the screening of somatic cell hybrids for COX complementation, provides a powerful means of identifying a genetically homogeneous population of patients, in spite of relatively heterogeneous clinical presentations. We think that this is an obligatory step towards the identification of the responsible gene by means of linkage analysis in selected families, or other approaches based on positional cloning.
A summary of the clinical and laboratory findings of our eight patients is reported in Table 0 .
Six patients (individuals S, P, M, V, L and C) shared an apparently identical, rapidly progressive encephalopathy, characterized by the following features: early onset; generalized hypotonia with brisk tendon reflexes; truncal ataxia; oculomotor abnormalities including slow saccades, ophthalmoparesis or complex irregular eye movements; `central' abnormalities of ventilation, including episodes of apnoea and irregular hyperpnoea; rapidly progressive psychomotor regression leading to death from central ventilatory failure. In one case (patient P), prominent dystonia was present. Patient L was the younger sister of an individual who was affected by a virtually identical syndrome. Patient V was the identical twin sister of a similarly affected individual. In all patients, the CT scan or MRI revealed the presence of symmetric lesions scattered from the basal ganglia to the brainstem, including the cerebellum. In one case (individual S), necropsy examination showed the presence of necrotic lesions associated with glial and vascular proliferation, as typically described in subacute necrotizing encephalomyelopathy. We diagnosed these six patients as having `typical' Leigh's syndrome.
Two additional patients (individuals B and Sa) were included in the study. Patient Sa was the only product of distantly related parents (third-degree cousins), suggesting an autosomal recessive trait. Although the clinical features of these patients largely overlapped with those of the other patients (Table 0 2 ), they were referred to us as `atypical' Leigh's syndrome cases by our clinical team, on the basis of the following findings: later onset, predominantly myopathic signs, absence of `central' abnormalities of ventilation, no signs of peripheral neuropathy and mild or no abnormalities of eye movements. Moreover, the clinical and laboratory investigations on patient Sa were reported previously by other authors, who also considered this case as a variant of Leigh's syndrome, on the basis of clinical, morphological and neuroradiological features (30 ).
However, in all of the eight patients, lactic acid was above our normal range in blood and urine, and the muscle biopsy examination showed a severe decrease in the histochemical reaction to COX. Ragged-red fibers were consistently absent in all cases, while lipid accumulation was a distinct feature of the muscle biopsies of patients B and Sa. Biochemically, an isolated defect of COX (5-10% of the normal values) was detected in fibroblasts of all patients, in muscle homogenates of all tested patients (see Table 0 2 ) and in platelets of patient Sa (30 ).
Fibroblast cell lines were established from skin biopsies carried out on the patients, with the informed consent of the patients' parents. Fibroblast cell lines derived from seven of our eight patients (patients P, C, L, M, V, B and Sa) were co-transfected with two recombinant plasmids, namely pBABE and pMK. Plasmid pBABE (courtesy of Dr H. Land, Imperial Cancer Research Fund, UK) is a recombinant DNA vector in which a retroviral long terminal repeat and the SV40 promoter both regulate the expression of the gene conferring resistance to puromycin. Plasmid pMK is a recombinant plasmid expressing constitutively a thermosensitive variant of the SV40 T antigen, that is inactivated at 39oC or above. Transfection was followed by incubation at 33oC under continuous selection with puromycin. Puromycin-resistant (Puror) cell lines were named as follows: PFT.1-Puror (from patient P), CFT.3-Puror (from patient C), LFT.8-Puror (from patient L), MFT.05-Puror (from patient M), VFT.3-Puror (from patient V), BFT.13-Puror (from patient B) and SaFT.1-Puror (from patient Sa). A total of 2.5×106 cells for each cell line were harvested by trypsinization and resuspended in 230 µl of 20 mM HEPES, 137 mM NaCl, 5 mM KCl, 0.7 mM Na2HPO4, 6 mM dextrose, pH 7.0. Forty µg of plasmid in 20 µl of 1* TE containing 200 µg of herring sperm DNA were added to the cell suspension and electroporation was carried out at 250 V, 125 µF. A transformant fibroblast clone from our eighth patient (individual S), called SFT.12-Neor, was obtained previously by transfection with plasmid pRNS-1, as described (14 ). Transfection of pRNS-1 in human diploid cells results in a high frequency of stable, neomycin-resistant (Neor) transformants expressing the SV40 T antigen. Stable transformant clones were selected in Dulbecco's modified Eagle's medium (DMEM)/uridine (see below), 200 µg/ml of the neomycin analog drug G-418 (for fibroblasts transfected with pRNS-1), or 0.5 µg/ml of puromycin (for fibroblasts co-transfected with pMK + pBABE).
A similar procedure was adopted to stably transfect plasmid pcDNA3 (Invitrogen) into 143B.rhoo cells. Plasmid pcDNA3 is an eukaryotic vector constitutively expressing the neomycin resistance gene. Selection of Neor 143B.rhoo cells was carried out by adding G-418 to the culture medium (see above).
About 1×106 cells of the SFT.12-Neor fibroblast cell line were co-cultured with 1×106 cells of each of the Puror fibroblast cell lines in DMEM containing 4.5 g/l of glucose and 110 µg/ml of pyruvate, supplemented with 5% fetal bovine serum (FBS, Boehringer) and 50 µg/ml uridine (DMEM/uridine). After incubation for 1 day, fusion was carried out by treating the confluent cell monolayer with a 50% (w/v) solution of PEG in phosphate buffer, pH 7.4. Twenty four hours after fusion, cells were trypsinized and re-plated in a selective DMEM/uridine medium containing 200 µg/ml of G-418 and 0.5 µg/ml of puromycin. Hybrids were named according to the initials of the patients, as follows: Hy[S+P], Hy[S+C], Hy[S+L], Hy[S+V], Hy[S+M], Hy[S+Sa] and Hy[S+B]. In addition, the transformant cell line CFT.3-Puror was PEG-fused with a Neor, COX(+) transformant fibroblast clone, called fAR.1-Neor (14 ), to produce a `positive control' hybrid called Hy[C+fAR]. The histochemical and biochemical assays were carried out after at least 3 weeks of continuous selection, well after that cultures of each parental cell line, grown in parallel in the same selective medium, were eliminated completely. In some cases, however, COX-specific histochemistry was also performed at early stages, usually at 10-12 days after fusion. A similar procedure was used to obtain hybrids from a rhoo cell line derived from 143B.TK- osteosarcoma cells, fused with each of our Puror transformant cell lines. Hybrids were named according to the initials of the patients, as follows: Hy[C+rhoo], Hy[L+rhoo], Hy[M+rhoo], Hy[P+rhoo], Hy[V+rhoo], Hy[Sa+rhoo] and Hy[B+rhoo]. After fusion, parental cells and homokaryon hybrids were eliminated by double-antibiotic selection, as above.
Total DNA was extracted from the hybrids as well as from each type of cell line used in the fusions. Characterization of the nuclear DNA was carried out by using a highly polymorphic repetitive sequence at the D11S533 locus on chromosome 11q, as described (14 ).
COX activity was visualized cytochemically in cell cultures grown on coverslips, washed twice in phosphate-buffered saline (PBS) containing 1 mg/ml MgCl2 and 1 mg/ml CaCl2, and air-dried. Cells were pre-incubated for 15 min at room temperature in a buffer containing 50 mM Tris-HCl pH 7.6, 1.2 mM CoCl2, 10% sucrose. They were rinsed once in 10% sucrose, 0.1 M Na phosphate pH 7.6, and incubated for 6 h at 37oC in 0.1 M Na phosphate buffer, pH 7.6, containing 2 mg/ml of cytochrome c. Cells were rinsed once in 0.1 M Na phosphate buffer, pH 7.6 and 10% sucrose, once in PBS-Mg-CaCl2, and once in distilled water. Coverslips were mounted in glycerol-PBS and visualized under a light microscope (Zeiss).
Cell homogenates were prepared according to the digitonin-based method described by Robinson et al. (32 ) modified as in ref. 14 . The enzyme activities of COX and rotenone-sensitive NADH-CoQR (complex I) were measured twice in each assay (33 ). In several cases, we also measured succinate dehydrogenase (part of complex II) (33 ). Protein concentration was measured by the method of Lowry (34 ). Activities were expressed as nmol substrate/min/mg protein. We did not normalize the respiratory chain activities for citrate synthase, because citrate synthase is also expressed at high levels in the rhoo mitochondria. This could determine underestimation of the respiratory activities in the rhoo-derived hybrids.
We thank Professor H. L. Ozer and Professor H. Land for kindly providing us the pRNS-1 and pBABE constructs, respectively. We thank Dr M. Mora for useful technical advice, and Professor G. Attardi, Dr Patricio Fernandez-Silva and Dr E. Bertini for critical discussion. We are indebted to Ms B. Geehan for revising the manuscript. This work was partially supported by EU Human Capital and Mobility network grant to M.Z. on `Mitochondrial Biogenesis in Development and Disease', CNR Progetto Finalizzato Ingegneria Genetica and Telethon-Italy (grant no. 767 to M.Z. and grant n. 200 to R.B.).
1 Zeviani, M. and Taroni, F. (1994) Mitochondrial diseases. InHarding, A.E. (ed.), Genetics in Neurology, Bailliere Tyndall, London, pp. 315-334.
2 Wallace, D.C., Lott, M.T., Torroni, A., Brown, M.D. and Shoffner, J.M. (1993) Report of the committee on human mitochondrial DNA. In Cuticchia, A.J. and Pearson, P.L. (eds), Human Gene Mapping, pp. 813-845.
3 Zeviani M. (1992) J. Inherited Metab. Dis., 15, 456-471.MEDLINE Abstract
4 Suomalainen, A., Kaukonen, J., Amati, P., Timonen, R., Haltia, M., Weissenbach, J., Zeviani, M., Somer, H. and Peltonen, L. (1995) Nature Genet., 9, 146-151.MEDLINE Abstract&form=6&uid=95235559&Dopt=r">MEDLINE Abstract
5 Kaukonen, J.A., Amati, P., Suomalainen, A., Rstig, A., Piscaglia, M.-G., Salvi, F., Weissenbach, J., Fratta, G., Comi, G., Peltonen, L. and Zeviani, M. (1996) Am. J. Hum. Genet., 58, 763-769.MEDLINE Abstract&form=6&uid=96220229&Dopt=r">MEDLINE Abstract
6 Bourgeron, T., Rustin, P., Chretien, D., Birch-Machin, M., Bourgeois, M., Viegas-Pquinot, E., Munnich, A. and Rstig, A. (1995) Nature Genet., 11, 144-149.MEDLINE Abstract&form=6&uid=96024649&Dopt=r">MEDLINE Abstract
7 DiMauro, S., Servidei, S., Zeviani, M., DiRocco, M., DeVivo, D.C., DiDonato, S., Uziel, G., Berry, K., Hoganson, G., Johnsen, S.D. and Johnson, P.C. (1987) Ann. Neurol., 22, 498-506.MEDLINE Abstract&form=6&uid=88132592&Dopt=r">MEDLINE Abstract
8 Stansbie, D., Wallace, S.J. and Marsac, C. (1986) J. Inherited Metab. Dis., 9, 105-109.MEDLINE Abstract&form=6&uid=86309167&Dopt=r">MEDLINE Abstract
9 Fuji, T., Ito, M., Okuno, T., Mutoh, K., Nishikomori, R. and Mikawa, H. (1990) J. Pediatr., 116, 84-87.
10 Wijburg, F.A., Wanders, R.J., van Lie Peters, E.M., Vos, G.D., Loggers, H.G., Bolhuis, P.A., Herzberg, N.HH, Ruitenbeek, W., van Wilsem A., ten Houten, R. and Barth, P.G. (1991) J. Inherited Metab. Dis., 14, 297-300.MEDLINE Abstract&form=6&uid=92122750&Dopt=r">MEDLINE Abstract
11 Tatuch, Y., Christodoulou, J., Feigenbaum, A., Clarke, J.T.R., Wherret, J., Smith, C., Rudd, N., Petrova-Benedict, R. and Robinson, B.H. (1992) Am. J. Hum. Genet., 50, 852-858.MEDLINE Abstract&form=6&uid=92197632&Dopt=r">MEDLINE Abstract
12 Santorelli, F.M., Shanske, S., Macaya, A., DeVivo, D.C. and DiMauro, S. (1993) Ann. Neurol., 34, 827-834. MEDLINE Abstract&form=6&uid=94071416&Dopt=r">MEDLINE Abstract
13 de Vries, D.D., van Engelen, B.G., Gabreels, F.J., Ruitenbeek, W. and van Oost, B.A. (1993) Ann. Neurol., 34, 410-412.MEDLINE Abstract&form=6&uid=93371018&Dopt=r">MEDLINE Abstract
14 Tiranti, V., Munaro, M., Sandon, D., Lamantea, E., Rimoldi, M., DiDonato, S., Bisson, R. and Zeviani, M (1995) Hum. Mol. Genet., 4, 2017-2023.MEDLINE Abstract&form=6&uid=96154671&Dopt=r">MEDLINE Abstract
16 Kadenbach, B., Kuhn-Nentwig, L. and Buge, U. (1987) Curr. Top. Bioenerg., 15, 114-161.
17 DiMauro, S., Lombes, A., Nakase, H., Mita, S., Fabrizi, G.M., Tritschler, H.-J., Bonilla, E., Miranda, A.F., DeVivo, D.C. and Schon, E.A. (1990) Pediatr. Res., 28, 536-541.MEDLINE Abstract&form=6&uid=91074609&Dopt=r">MEDLINE Abstract
18 Glerum, M., Yanamura, W., Capaldi, R.A. and Robinson, B.H. (1988) FEBS Lett., 236, 100-104.
19 Merante, F., Petrova-Benedict, R., MacKay, N., Mitchell, G., Lambert, M., Morin, C., De-Braekeleer, M., Laframboise, R., Gagne, R. and Robinson, B.H. (1993) Am. J. Hum. Genet., 53, 481-487.
20 Lombes, A., Nakase, H, Tritschler, H.J., Kadenbach, B., Bonilla E., DeVivo, D.C., Schon, E.A. and DiMauro, S. (1991) Neurology , 41, 491-498.MEDLINE Abstract&form=6&uid=91187205&Dopt=r">MEDLINE Abstract
29 Miranda, A.F., Ishii, S., DiMauro, S. and Shay, J.W. (1989) Neurology, 39, 697-702.MEDLINE Abstract&form=6&uid=89219626&Dopt=r">MEDLINE Abstract
30 Angelini, C., Bresolin, N., Pegolo, G., Bet, L., Rinaldo, P., Trevisan, C. and Vergani, L. (1986) Neurology, 36, 1048-1052.MEDLINE Abstract&form=6&uid=86285460&Dopt=r">MEDLINE Abstract
31 Morin, C., Mitchell, G., Larochelle, J., Lambert, M., Ogier, H., Robinson, B.H. and DeBraekeleer, M. (1993) Am. J. Hum. Genet., 53, 488-496.MEDLINE Abstract
32 Robinson, B.H., Jewel, W., Goodyer, P. and Baudet, A. (1986) J. Clin. Invest., 77, 1422-1427.MEDLINE Abstract&form=6&uid=86196695&Dopt=r">MEDLINE Abstract
33 Darley-Usmar, V.M., Rickwood, D. and Wilson, M.T. (eds) (1987) Mitochondria-A Practical Approach. IRL Press, Oxford/Washington, DC.
34 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem., 193, 265-271.
This page is maintained by OUP admin. Last updated Fri Jan 31 11:45:11 GMT 1997. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996
