Survey of human mitochondrial diseases using new genomic/proteomic tools
© Plasterer et al., licensee BioMed Central Ltd 2001
Received: 8 February 2001
Accepted: 26 April 2001
Published: 1 June 2001
We have constructed Bayesian prior-based, amino-acid sequence profiles for the complete yeast mitochondrial proteome and used them to develop methods for identifying and characterizing the context of protein mutations that give rise to human mitochondrial diseases. (Bayesian priors are conditional probabilities that allow the estimation of the likelihood of an event - such as an amino-acid substitution - on the basis of prior occurrences of similar events.) Because these profiles can assemble sets of taxonomically very diverse homologs, they enable identification of the structurally and/or functionally most critical sites in the proteins on the basis of the degree of sequence conservation. These profiles can also find distant homologs with determined three-dimensional structures that aid in the interpretation of effects of missense mutations.
This survey reports such an analysis for 15 missense mutations, one insertion and three deletions involved in Leber's hereditary optic neuropathy, Leigh syndrome, mitochondrial neurogastrointestinal encephalomyopathy, Mohr-Tranebjaerg syndrome, iron-storage disorders related to Friedreich's ataxia, and hereditary spastic paraplegia. We present structural correlations for seven of the mutations.
Of the 19 mutations analyzed, 14 involved changes in very highly conserved parts of the affected proteins. Five out of seven structural correlations provided reasonable explanations for the malfunctions. As additional genetic and structural data become available, this methodology can be extended. It has the potential for assisting in identifying new disease-related genes. Furthermore, profiles with structural homologs can generate mechanistic hypotheses concerning the underlying biochemical processes - and why they break down as a result of the mutations.
As we move into the post-genome era, bioinformatics tools provide a powerful means of organizing and integrating information related to hereditary diseases. We present here a prototype study of this approach. Using a database of all identifiable proteins that contribute to the structure and/or biosynthesis of yeast (Saccharomyces cerevisiae) mitochondria, we have generated a library of protein profiles  that enables us to identify homologs in a wide range of genomes, including human. As a result, we have available sets of highly accurate amino-acid sequence alignments over the conserved regions that correspond to these profiles. By mapping known human disease-related mutations onto these alignments we can investigate the relationships to sequence conservation - and to sequence variations. In the cases where one or more of the identified homologs has a determined structure, these mutations can be interpreted in structural terms, allowing us to draw mechanistic inferences.
We have used profile analysis based on Bayesian priors, conditional probabilities that allow the estimation of the likelihood of an event on the basis of prior occurrences of similar events, to make this library of sequence profiles. The strength of prior-based profile analysis lies in its ability to capture homologous protein domains as related objects. Each domain is represented as a matrix of amino-acid probabilities (more precisely, log likelihood values) where each position in the domain has an estimated probability of being occupied by any particular amino acid. The similarities or degrees of residue conservation between homologous sequences reveal positions critical for protein structure and function. A profile-induced multiple alignment displays immutable positions, physicochemically conserved positions and highly variable positions. The examples presented in this paper show how such alignments provide insight into human mitochondrial pathologies.
Mitochondrial proteins are an attractive subset of the cellular proteome to analyze because they perform a number of correlated functions and interact extensively with one another. Their properties and behavior reflect an integrated system, and the effects of different mutational changes can be meaningfully cross-compared. The survey of this miniature proteome should reveal many of the features that characterize the larger complete proteomes of eukaryotic cell types. Mitochondrial pathologies also present uniquely complex evolutionary and genetic features as a result of their origin within organelles derived from ancient symbionts and the strictly maternal inheritance pattern followed by an important subset of the relevant genes.
Mitochondrial genome overview
The mitochondrial genome itself encodes only a handful of genes. The largest mitochondrial genome discovered to date - of the protist Reclinomonas americana - codes for 97 gene products (67 proteins and 30 structural RNAs) in a span of 69,034 base pairs . Human mitochondrial DNA is much smaller, encoding 37 gene products (13 proteins, 2 rRNAs and 22 tRNAs) over 16,569 base pairs . All its encoded proteins are directly involved in oxidative phosphorylation and include components of complex I (NADH dehydrogenase subunits ND1, ND2, ND3, ND4, ND4L, ND5, ND6), complex III (cytochrome b), complex IV (cytochrome oxidase subunits I, II and III) and complex V (mitochondrial ATPase subunits 6 and 8) . All other human genes encoding mitochondrial proteins (total estimated at around 1,100 ) are transcribed in the nucleus, translated in the cytoplasm and their products imported into mitochondria. The interplay between organellar and nuclear genomes accounts for a large part of the diversity in mitochondrial pathologies.
Each human cell contains multiple mitochondria, often dynamically interconnected in a complex reticular network [6,7], and each mitochondrion may contain ten or more mitochondrial DNA (mtDNA) molecules. Usually all copies of mtDNA are identical, a state known as homoplasmy. Occasionally, however, mtDNA mutations occur and there arise two (or more) populations of mtDNA, a state called heteroplasmy . In fact, heteroplasmic mtDNA ought to be relatively common, considering that human mtDNA mutates 10-20 times faster than nuclear DNA as a result of inadequate proofreading by mitochondrial DNA polymerases  and limited mtDNA repair capability. This expectation is to some extent borne out by the relative prominence of mitochondrial disorders arising from mutations in the mtDNA - although it is also important to note that such mutations, being comparatively easy to identify by sequencing, will naturally have been among the first to be characterized.
Schapira has devised a classification scheme for mitochondrial defects  based primarily on whether or not the affected gene product is directly involved in oxidative phosphorylation and whether it is encoded by mtDNA or nuclear DNA.
A primary defect in oxidative phosphorylation defines class I. Distinct mitochondrial pathologies may overlap more than one class, for example Leigh syndrome (see below). Subclass Ia includes diseases arising from mutations in mtDNA genes encoding subunits of proteins involved in oxidative phosphorylation, mitochondrial tRNA genes and mitochondrial rRNA genes. This subclass has three related categories, depending on the nature of the underlying mutations: (i) large-scale mtDNA deletions and duplications; (ii) point mutations and small rearrangements in protein-coding mtDNA regions; and (iii) small-scale mtDNA mutations in tRNA and rRNA genes . In order for a class Ia pathology to present itself, the mtDNA mutation must occur in a significant fraction of the heteroplasmic mtDNA population, as high as 60% for mtDNA deletions  (Ia-i) and even up to 95% for tRNA mutations (Ia-iii) . Class Ib mitochondrial diseases arise from mutations in any of the 70-odd nuclear genes encoding protein subunits involved in oxidative phosphorylation . These mutations can occur in exons or introns as well as promoters.
Class II mitochondrial disorders stem from secondary defects in oxidative phosphorylation. Class IIa genetic pathologies are characterized by mutations in nuclear mitochondrial protein genes whose products are imported into the mitochondria but are not subunits of the oxidative phosphorylation apparatus. There are four subtypes: (i) mtDNA abnormality due to mutations in nuclear genes affecting mtDNA transcription, translation or replication; (ii) direct mtDNA damage or defects in mtDNA repair; (iii) defective oxidative phosphorylation subunit import; and (iv) defective oxidative phosphorylation subunit assembly . Type IIb pathologies arise from endogenous and exogenous toxins.
Selected mutations associated with human mitochondrial disorders
Near active site
Near active site
Near active site
Near active site
Notin active site
T151del (1 nt)
Truncation - see text
A183del (10 nt)
Truncation - see text
ATT → ATG
Predicted tight turn
784del (2 nt)
2228ins (1 nt = A)
Primary mutational disorders of oxidative phosphorylation (class I)
Leber's hereditary optic neuropathy
Leber's hereditary optic neuropathy (LHON), the most common cause of male adolescent blindness, is attributed to mutations in mtDNA genes encoding subunits of complex I (NADH dehydrogenase (ND) - alternatively termed NADH CoQ1 reductase). The most frequently encountered mutation (50% of cases among Europeans, 95% among Asians) is G11778A in ND4 (Arg340 → His). Additionally, G3460A (Ala52 → Thr) in ND1, and both T14484C (Met64 → Val) and G14459A (Ala72 → Val) in ND6 rank among the other primary LHON-associated mutations [4,11].
Lower copy numbers of this mutation (~ 70%) are associated with neuropathy, ataxia and retinitis pigmentosa (NARP) whereas high copy numbers (>90%) give rise to Leigh syndrome . Mutating the conserved leucine to arginine at position 156 is a drastic change, disrupting a γ-branched aliphatic residue with a basic guanidino group, a major chemical and presumably structural substitution. In this case structural data are available (see Figure 4b) and clearly indicates the key location of this residue at the protein-protein interface of two of the subunits of the F0 component of complex V (ATP synthase).
Whereas only the human sequence uses arginine at this position, all other taxa use a large hydrophobic amino acid - either tyrosine or leucine; note that even an arginine side chain has a substantial hydrophobic segment bearing its terminal guanidino group. Substituting the small turn-forming glycine residue could have a significant impact on proper folding of this subunit. As shown in Figure 5b, Arg263 is located quite close to the active site of the enzyme, a region of the structure where even small perturbations could have large effects on activity.
Mutating the conserved arginine residue at position 554 to a tryptophan causes a loss of a polar residue and replacement with a much larger aromatic group. It is not apparent from the location of this residue in the structure (Figure 7b) why this change should have such an effect on function. It lies on the surface at a considerable distance from the active site and does not have any apparent role in subunit-subunit interactions. In fact, however, the high degree of sequence conservation over this entire region (Figure 7a) suggests that SDHA is fairly intolerant of mutations here. Proper interaction with the lipid bilayer may partly explain the phenomenon.
Secondary mutational disorders in oxidative phosphorylation (class II)
Mitochondrial neurogastrointestinal encephalomyopathy
Class II mitochondrial disorders are due to mutations in nuclear mitochondrial protein genes. Mitochondrial neurogastrointestinal encephalomyopathy (MNGIE) is characterized by ptosis, progressive external ophthalmoplegia, gastrointestinal dismotility, thin body habitus, peripheral neuropathy, myopathy, leukoencephalopathy and lactic acidosis and multiple mtDNA deletions or mtDNA depletion, or both. It is a class II Δ-mtDNA depletion disorder. It is believed that the pathogenic mechanism relates to aberrant thymidine metabolism giving rise to impaired mtDNA replication and/or mtDNA maintenance . Missense mutations in the enzyme thymidine phosphorylase (TP) (Gly145 → Arg, Gly153 → Ser, Lys222 → Ser, Glu289 → Ala and a few others) have been found in multiple MNGIE patients, and are believed to cause the pathology .
Deafness-dystonia, or Mohr-Tranebjaerg syndrome, is a mitochondrial protein import disorder. Mutations in DDP1 (deafness/dystonia peptide 1) lead to deafness-dystonia. In two cases seen to date, a 1 bp deletion in exon 1 (T151del) results in a frameshift, leading to incorporation of 25 new amino acids after Glu38, followed by a stop codon. In a second case, a 10 bp deletion in exon 2 causes a frameshift at Met48, leading to incorporation of 12 new residues and a stop codon . These proteins are homologous to the yeast mitochondrial import protein Tim8p (YJR135W-A) .
Friedreich's ataxia also falls into the category of a class II mtDNA damage/repair disorder. It is characterized by progressive gait and limb ataxia, lack of tendon reflexes, dysarthria and pyramidal weakness in inferior limbs . Lack of frataxin protein causes a build-up of iron in the matrix spaces of the mitochondria. As iron accumulates, it competes with catalase for hydrogen peroxide and forms hydroxyl radicals (Fe2+ + H2O2 → Fe3+ + OH- + OH·) - so-called reactive oxygen species that can attack mtDNA, membrane lipids, carbohydrates and proteins. This can damage the entire oxidative phosphorylation system, especially the highly vulnerable iron-sulfur centers. The shortage of frataxin does not result from formation of mutant proteins, but rather from expansion of a GAA repeat in the first intron of the frataxin gene. This expanded repeat is believed to form localized RNA secondary structures that interfere with transcription [31,32]. Yfh1p (YDL120W), the frataxin homolog in yeast, interacts with the yeast mitochondrial intermediate peptidase (Oct1, YKL134C) to lower matrix iron concentration, directly and by competition .
A pair of cation transporters, Mmt1 (YMR177W) and Mmt2 (YPL224C) in yeast, controls iron influx into the mitochondria. This other half of iron homeostasis could also easily have a role in Friedreich's ataxia and X-linked sideroblastic anemia.
Hereditary spastic paraplegia
The yeast m-AAA protease complex is itself regulated by two prohibitin proteins, Phb1 (YGR132C) and Phb2 (YGR231C) . Paraplegin homologs, analogous to the yeast paralogs Afg3p, Rca1p and Yme1p, must almost certainly exist, and human prohibitins may be expected as well. Point mutations in conserved regions of these proteins can be expected to cause missense mutations and translational truncations that will also lead to hereditary spastic paraplegia.
The process of understanding genetic disease typically proceeds through three stages: first, recognition of the disease state or syndrome including its hereditary character; second, discovery and mapping of the related mutation(s); and third, elucidation of the biochemical/biophysical mechanism leading to the disease phenotype. Sickle-cell anemia provides the classic example. In the case of mitochondrial diseases there are more than 100 described conditions (OMIM , MITOP , and WUSTL  websites) and a comparable (and growing!) number of mutations . As already described for LHON, Leigh syndrome and HSP, the causative mutations are not necessarily confined to a single protein-coding gene, or even to the nuclear versus the mitochondrial genome. Put differently, there is no one-to-one mapping between mutations (or even whole genes) and a particular defined disease entity. Thus, each disease spectrum may correspond to several sets of mutations, each set affecting one protein or protein subunit. Obviously, mutations in mt-tRNA, mt-rRNA and proteins that affect mitochondrial gene expression will also potentially have pleiotropic effects.
The results presented here address the third stage of understanding genetic disease, namely the characterization of the disease phenotype at the molecular level with a view to explaining its biochemical mechanism. In principle, the approach can be applied to any mutated protein-coding gene. When structural information is available, the exact mechanism of disease causation may be ascertainable at the molecular level - and testable by means of carefully targeted experiments. A further extension of this basic-knowledge approach lies in the direction of molecular evolution. As we will show elsewhere (T.P., T.F.S. and S.C.M., unpublished observations), the extended sets of homologs assembled using prior-based profiles provide a rich source of phylogenetic information. Not only should it be possible to sketch out the history of the development of different biochemical machineries, it should also be possible to establish which parts of the proteins are most critical to function on the basis of amino-acid sequence conservation correlated with the structural contexts.
Knowledge gained by the approach described here has obvious application to both the diagnosis and therapy of mitochondrial diseases. Once a mutation has been characterized, the exact nature of the associated illness is known with precision. This has a bearing upon choice of therapy and allows a much more exact prognosis than a characterization simply as, say, Leigh syndrome. In addition, the information from sequence alignments has predictive power: mutations not yet recorded in the literature might turn up at key positions in some of the proteins known to be affected in mitochondrial diseases. Individuals carrying such changes might be alerted to the possible harmful consequences and take whatever precautionary measures seem appropriate. (Note that substitutions in the human sequence that match amino-acid residues found at the corresponding locations in taxonomically distant sequences are less likely to prove harmful than completely novel substitutions.) This application of informatics data will certainly grow in importance as acquisition of additional single-nucleotide polymorphism (SNP) data becomes increasingly common and new paralogs and orthologs are added to the sequence database. Finally, depending on the nature of the affected protein and the frequency of occurrence of diseased individuals, it may be possible to use the data for attempts at rational drug design or gene therapy.
The work reported here represents a preliminary overview of the field. Eventually, an exhaustive survey should be completed, continuously updated and made publicly available. That will require precise characterization of many more mutations as well as expansion of the database of mitochondrial protein profiles beyond what we have constructed using yeast as a model. Such efforts are underway.
Materials and methods
Yeast mitochondrial protein database (YMPD)
We created a database of >430 yeast mitochondrial proteins using biochemical keyword searches for known mitchondrial processes from the Saccharomyces Genome Database , the Munich Information Center for Protein Sequences , the Yeast Proteome Database , and MitBASE , and by monitoring the current literature. Each member of this set of proteins is characterized by at least one prior-based profile  that represents the most conserved region(s).
As S. cerevisiae does not use a canonical mitochondrial complex I we have substituted the NADH dehydrogenase protein subunits from R. americana to initialize these profiles.
Developing prior-based profiles
Prior-based profiles , the principal tools used in this work, represent evolutionarily conserved, sequence-based, functional domains. Once created, each profile can be used to search all available databases in an attempt to locate all homologous sequence matches - which in ideal cases correspond to single protein domains. This expanded set of homologous sequences is then multiply-aligned using the profile as a template. The set of multiply-aligned sequences defines sequence-domain boundaries, allows putative functional assignments for previously unidentified proteins and highlights key conserved residues of members of a homologous set of proteins. As new sequences (and sequenced genomes) become available, our set of profiles can identify the additional homologs with high sensitivity and specificity .
The profile-defining set
A defining set with a wide taxonomic spread should be optimal for locating distant homologs from as many species as possible. Ideal profiles have a defining set composed of one protein each from a eukaryote, an archaeon, a Gram-positive bacterium and a Gram-negative bacterium, and one yeast mitochondrial protein. To create the defining set for each profile, we established a superset of potential family members using BLAST searches  against our Biomolecular Engineering Research Center (BMERC) all-genomes database and Swiss-Prot . Each identified yeast mitochondrial protein serves as a 'seed' (initializing protein) for a BLAST search. The cutoff scores are set at an expectation value of E= 10-6 using the standard SEG and XNU filters for low-complexity regions . From these BLAST hits we selected the initial defining set of maximum taxonomic spread. In some cases, BLAST fails to locate sequence homologs that have diverged greatly from the seed mitochondrial protein. We then used an iterative procedure to generate a profile using local dynamic programming, to maximize the information content. For such cases, the initial profile is run against the all-genomes database by exhaustive local dynamic programming. Then the next highest scoring member from a different kingdom is added to the defining set. We iterated this process until the ideal taxonomic spread was achieved or the information content of the profile dropped below ten amino-acid equivalencies .
Expanding the set of profile homologs
Once a profile is built we attempt to find all sequences sharing this sequence domain. This is done by heuristic and/or exhaustive approaches. In the heuristic approach the profile searches BMERC's all-genomes and Swiss-Prot by filtering the larger databases using BLAST with an expectation value cutoff of E = 10-2, followed by a short-in-long profile search. In the exhaustive approach, the profile searches the all-genomes database or the current version of Swiss-Prot using short-in-long dynamic programming.
Multiple alignment sets
Each profile induces a multiple alignment for any set of proteins that contain it. The profiles only align those subregions of a protein that they match. These regions incorporate the most conserved amino-acid positions among the sequences that make up the profile's defining set. We have chosen defining sets that contain the shortest informative protein sequence in order to maximize the chance of producing single-domain profile objects. A longer sequence in the defining set might make a slightly higher-scoring profile by the addition of a few carboxy- or amino-terminal residues, but only at the risk of missing slightly shorter members of the set when the profile is used to search the databases.
Homologous structures from the Protein Data Bank
We routinely search the RCSB Protein Data Bank , for potential structural homologs. Profile matches with Z-scores above 10 are retained and used to align the profile-matched region to a structural domain. These profile-induced structural alignments can be visualized in Rasmol , and images built using Molscript  and Rasters3D .
Additional data files
The following additional data files are available with this paper online: complete alignments for the profile-covered regions of the sequences shown in Figure 1 (additonal data file 1), Figure 2 (additonal data file 2), Figure 3 (additonal data file 3), Figure 4 (additonal data file 4), Figure 5 (additonal data file 5), Figure 6 (additonal data file 6), Figure 7 (additonal data file 7), Figure 8 (additonal data file 8), Figure 9 (additonal data file 9), Figure 10 (additonal data file 10) and Figure 11 (additonal data file 11).
We thank Greg McAllister for helpful discussions of transmembrane protein structures. This work was partially supported by US Department of Energy Grant DE-FG02-98ER62558 and NSF Grant DBI-9807993.
- Das S, Smith TF: Identifying nature's protein Lego set. In Advances in Protein Chemistry. Edited by Richards FM, Eisenberg DS,Kim PS. San Diego: Academic Press;. 2000, 159-183.Google Scholar
- Lang BF, Burger G, O'Kelly CJ, Cedergren R, Golding GB, Lemieux C, Sankoff D, Turmel M, Gray MW: An ancestral mitochondrial DNA resembling a eubacterial genome in miniature. Nature. 1997, 387: 493-497.PubMedView ArticleGoogle Scholar
- Taanman JW: The mitochondrial genome: structure, transcription, translation and replication. Biochim Biophys Acta. 1999, 1410: 103-123.PubMedView ArticleGoogle Scholar
- Leonard JV, Schapira AH: Mitochondrial respiratory chain disorders I: mitochondrial DNA defects. Lancet. 2000, 355: 299-304.PubMedView ArticleGoogle Scholar
- Mitochondrial disorders. [http://www.neuro.wustl.edu/neuromuscular/mitosyn.html]
- Rizzuto R, Pinton P, Carrington W, Fay FS, Fogarty KE, Lifshitz LM, Tuft RA, Pozzan T: Close contacts with the endoplasmic reticulum as determinants of mitochondrial Ca2+ responses. Science. 1998, 280: 1763-1766.PubMedView ArticleGoogle Scholar
- Skulachev VP: Mitochondrial filaments and clusters as intracellular power-transmitting cables. Trends Biochem Sci. 2001, 26: 23-29.PubMedView ArticleGoogle Scholar
- Schapira AH: Mitochondrial disorders. Biochim Biophys Acta. 1999, 1410: 99-102.PubMedView ArticleGoogle Scholar
- Hayashi J, Ohta S, Kikuchi A, Takemitsu M, Goto Y, Nonaka I: Introduction of disease-related mitochondrial DNA deletions into HeLa cells lacking mitochondrial DNA results in mitochondrial dysfunction. Proc Natl Acad Sci USA. 1991, 88: 10614-10618.PubMedPubMed CentralView ArticleGoogle Scholar
- Chomyn A, Martinuzzi A, Yoneda M, Daga A, Hurko O, Johns D, Lai ST, Nonaka I, Angelini C, Attardi G: MELAS mutation in mtDNA binding site for transcription termination factor causes defects in protein synthesis and in respiration but no change in levels of upstream and downstream mature transcripts. Proc Natl Acad Sci USA. 1992, 89: 4221-4225.PubMedPubMed CentralView ArticleGoogle Scholar
- Riordan-Eva P, Harding AE: Leber's hereditary optic neuropathy: the clinical relevance of different mitochondrial DNA mutations. J Med Genet. 1995, 32: 81-87.PubMedPubMed CentralView ArticleGoogle Scholar
- Brown MD, Voljavec AS, Lott MT, MacDonald I, Wallace DC: Leber's hereditary optic neuropathy: a model for mitochondrial neurodegenerative diseases. FASEB J. 1992, 6: 2791-2799.PubMedGoogle Scholar
- Zeviani M, Bertagnolio B, Uziel G: Neurological presentations of mitochondrial diseases. J Inherit Metab Dis. 1996, 19: 504-520.PubMedView ArticleGoogle Scholar
- Trounce I, Neill S, Wallace DC: Cytoplasmic transfer of the mtDNA nt 8993 T → G (ATP6) point mutation associated with Leigh syndrome into mtDNA-less cells demonstrates cosegregation with a decrease in state III respiration and ADP/O ratio. Proc Natl Acad Sci USA. 1994, 91: 8334-8338.PubMedPubMed CentralView ArticleGoogle Scholar
- Santorelli FM, Shanske S, Macaya A, DeVivo DC, DiMauro S: The mutation at nt 8993 of mitochondrial DNA is a common cause of Leigh's syndrome. Ann Neurol. 1993, 34: 827-834.PubMedView ArticleGoogle Scholar
- Marsac C, Benelli C, Desguerre I, Diry M, Fouque F, De Meirleir L, Ponsot G, Seneca S, Poggi F, Saudubray JM, et al: Biochemical and genetic studies of four patients with pyruvate dehydrogenase E1 alpha deficiency. Hum Genet. 1997, 99: 785-792. 10.1007/s004390050449.PubMedView ArticleGoogle Scholar
- Dahl H-HM: Getting to the nucleus of mitochondrial disorders: Identification of respiratory chain-enzyme genes causing Leigh syndrome. Am J Hum Genet. 1998, 63: 1594-1597.PubMedPubMed CentralView ArticleGoogle Scholar
- Tiranti V, Hoertnagel K, Carrozzo R, Galimberti C, Munaro M, Granatiero M, Zelante L, Gasparini P, Marzella R, Rocchi M, et al: Mutations of SURF-1 in Leigh disease associated with cytochrome c oxidase deficiency. Am J Hum Genet. 1998, 63: 1609-1621.PubMedPubMed CentralView ArticleGoogle Scholar
- Poyau A, Buchet K, Bouzidi MF, Zabot MT, Echenne B, Yao J, Shoubridge EA, Godinot C: Missense mutations in SURF1 associated with deficient cytochrome c oxidase assembly in Leigh syndrome patients. Hum Genet. 2000, 106: 194-205. 10.1007/s004390051028.PubMedView ArticleGoogle Scholar
- Bourgeron T, Rustin P, Chretien D, Birch-Machin M, Bourgeois M, Viegas-Pequignot E, Munnich A, Rotig A: Mutation of a nuclear succinate dehydrogenase gene results in mitochondrial respiratory chain deficiency. Nat Genet. 1995, 11: 144-149.PubMedView ArticleGoogle Scholar
- Parfait B, Chretien D, Rotig A, Marsac C, Munnich A, Rustin P: Compound heterozygous mutations in the flavoprotein gene of the respiratory chain complex II in a patient with Leigh syndrome. Hum Genet. 2000, 106: 236-243. 10.1007/s004390051033.PubMedView ArticleGoogle Scholar
- Nishino I, Spinazzola A, Hirano M: Thymidine phosphorylase gene mutations in MNGIE, a human mitochondrial disorder. Science. 1999, 283: 689-692. 10.1002/(SICI)1097-010X(19990601)283:7<689::AID-JEZ7>3.0.CO;2-T.PubMedView ArticleGoogle Scholar
- Miyadera K, Dohmae N, Takio K, Sumizawa T, Haraguchi M, Furukawa T, Yamada Y, Akiyama S: Structural characterization of thymidine phosphorylase from human placenta. Biochem Biophys Res Commun. 1995, 212: 1040-1045. 10.1006/bbrc.1995.2074.PubMedView ArticleGoogle Scholar
- Brown NS, Bicknell R: Thymidine phosphorylase, 2-deoxy-D-ribose and angiogenesis. Biochem J. 1998, 334: 1-8.PubMedPubMed CentralView ArticleGoogle Scholar
- Jin H, May M, Tranebjaerg L, Kendall E, Fontan G, Jackson J, Subramony SH, Arena F, Lubs H, Smith S, et al: A novel X-linked gene, DDP, shows mutations in families with deafness (DFN-1), dystonia, mental deficiency and blindness. Nat Genet. 1996, 14: 177-180.PubMedView ArticleGoogle Scholar
- Koehler CM, Leuenberger D, Merchant S, Renold A, Junne T, Schatz G: Human deafness dystonia syndrome is a mitochondrial disease. Proc Natl Acad Sci USA. 1999, 96: 2141-2146.PubMedPubMed CentralView ArticleGoogle Scholar
- Tranebjaerg L, Hamel BC, Gabreels FJ, Renier WO, Van Ghelue M: A de novo missense mutation in a critical domain of the X-linked DDP gene causes the typical deafness-dystonia-optic atrophy syndrome. Eur J Hum Genet. 2000, 8: 464-467.PubMedView ArticleGoogle Scholar
- Koehler CM, Merchant S, Schatz G: How membrane proteins travel across the mitochondrial intermembrane space. Trends Biochem Sci. 1999, 24: 428-432.PubMedView ArticleGoogle Scholar
- Leuenberger D, Bally NA, Schatz G, Koehler CM: Different import pathways through the mitochondrial intermembrane space for inner membrane proteins. EMBO J. 1999, 18: 4816-4822.PubMedPubMed CentralView ArticleGoogle Scholar
- Rotig A, de Lonlay P, Chretien D, Foury F, Koenig M, Sidi D, Munnich A, Rustin P: Aconitase and mitochondrial iron-sulphur protein deficiency in Friedreich ataxia. Nat Genet. 1997, 17: 215-217.PubMedView ArticleGoogle Scholar
- Delatycki MB, Williamson R, Forrest SM: Friedreich ataxia: an overview. J Med Genet. 2000, 37: 1-8.PubMedPubMed CentralView ArticleGoogle Scholar
- Sakamoto N, Chastain PD, Parniewski P, Ohshima K, Pandolfo M, Griffith JD, Wells RD: Sticky DNA: self-association properties of long GAA.TTC repeats in R.R.Y Triplex structures from Friedreich's ataxia. Mol Cell. 1999, 3: 465-475.PubMedView ArticleGoogle Scholar
- Branda SS, Yang ZY, Chew A, Isaya G: Mitochondrial intermediate peptidase and the yeast frataxin homolog together maintain mitochondrial iron homeostasis in Saccharomyces cerevisiae. Hum Mol Genet. 1999, 8: 1099-1110.PubMedView ArticleGoogle Scholar
- Allikmets R, Raskind WH, Hutchinson A, Schueck ND, Dean M, Koeller DM: Mutation of a putative mitochondrial iron transporter gene (ABC7) in X-linked sideroblastic anemia and ataxia (XLSA/A). Hum Mol Genet. 1999, 8: 743-749.PubMedView ArticleGoogle Scholar
- Stultz CM, White JV, Smith TF: Structural analysis based on state-space modeling. Protein Sci. 1993, 2: 305-314.PubMedPubMed CentralView ArticleGoogle Scholar
- White JV, Stultz CM, Smith TF: Protein classification by stochastic modeling and optimal filtering of amino acid sequences. Math Biosci. 1994, 119: 35-75.PubMedView ArticleGoogle Scholar
- Casari G, De Fusco M, Ciarmatori S, Zeviani M, Mora M, Fernandez P, De Michele G, Filla A, Cocozza S, Marconi R, et al: Spastic paraplegia and OXPHOS impairment caused by mutations in paraplegin, a nuclear-encoded mitochondrial metallopro-tease. Cell. 1998, 93: 973-983.PubMedView ArticleGoogle Scholar
- Steglich G, Neupert W, Langer T: Prohibitins regulate membrane protein degradation by the m-AAA protease in mitochondria. Mol Cell Biol. 1999, 19: 3435-3442.PubMedPubMed CentralView ArticleGoogle Scholar
- Online Mendelian Inheritance in Man. [http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=OMIM]
- Mitochondrial Project. [http://www.mips.biochem.mpg.de/proj/medgen/mitop/]
- Chervitz SA, Hester ET, Ball CA, Dolinski K, Dwight SS, Harris MA, Juvik G, Malekian A, Roberts S, Roe T, et al: Using the Saccharomyces Genome Database (SGD) for analysis of protein similarities and structure. Nucleic Acids Res. 1999, 27: 74-78.PubMedPubMed CentralView ArticleGoogle Scholar
- Mewes HW, Frishman D, Gruber C, Geier B, Haase D, Kaps A, Lemcke K, Mannhaupt G, Pfeiffer F, Schuller C, et al: MIPS: a database for genomes and protein sequences. Nucleic Acids Res. 2000, 28: 37-40.PubMedPubMed CentralView ArticleGoogle Scholar
- Hodges PE, McKee AH, Davis BP, Payne WE, Garrels JI: The Yeast Proteome Database (YPD): a model for the organization and presentation of genome-wide functional data. Nucleic Acids Res. 1999, 27: 69-73.PubMedPubMed CentralView ArticleGoogle Scholar
- de Pinto B, Malladi SB, Altamura N: MitBASE pilot: a database on nuclear genes involved in mitochondrial biogenesis and its regulation in Saccharomyces cerevisiae. Nucleic Acids Res. 1999, 27: 147-149.PubMedPubMed CentralView ArticleGoogle Scholar
- Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ: Basic local alignment search tool. J Mol Biol. 1990, 215: 403-410. 10.1006/jmbi.1990.9999.PubMedView ArticleGoogle Scholar
- Bairoch A, Apweiler R: The SWISS-PROT protein sequence data bank and its supplement TrEMBL in 1999. Nucleic Acids Res. 1999, 27: 49-54.PubMedPubMed CentralView ArticleGoogle Scholar
- Wootton JC, Federhen S: Analysis of compositionally biased regions in sequence databases. Methods Enzymol. 1996, 266: 554-571.PubMedView ArticleGoogle Scholar
- Smith RF, Smith TF: Automatic generation of primary sequence patterns from sets of related protein sequences. Proc Natl Acad Sci USA. 1990, 87: 118-122.PubMedPubMed CentralView ArticleGoogle Scholar
- Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, Shindyalov IN, Bourne PE: The Protein Data Bank. Nucleic Acids Res. 2000, 28: 235-242.PubMedPubMed CentralView ArticleGoogle Scholar
- Sayle RA, Milner-White EJ: RASMOL: biomolecular graphics for all. Trends Biochem Sci. 1995, 20: 374-PubMedView ArticleGoogle Scholar
- Kraulis PJ: MOLSCRIPT: a program to produce both detailed and schematic plots of protein structures. J Appl Crystallogr. 1991, 24: 946-950. 10.1107/S0021889891004399.View ArticleGoogle Scholar
- Merritt E, Bacon D: Raster3D: photorealistic molecular graphics. Methods Enzymol. 1997, 277: 505-524.PubMedView ArticleGoogle Scholar
- Amino acid classes. [http://bmerc-www.bu.edu/description/aaclasses.html]