Introducing a novel human mtDNA mutation into the Paracoccus denitrificans COX I gene explains functional deficits in a patient

Neurogenetics (2006) 7: 51–57 DOI 10.1007/s10048-005-0015-z

ORIGINA L ARTI CLE

Simona Lucioli . Klaus Hoffmeier . Rosalba Carrozzo . Alessandra Tessa . Bernd Ludwig . Filippo M. Santorelli

Introducing a novel human mtDNA mutation into the Paracoccus denitrificans COX I gene explains functional deficits in a patient Received: 16 February 2005 / Accepted: 22 August 2005 / Published online: 12 November 2005 # Springer-Verlag 2005

Abstract We identified a novel mutation (S142F) in the human mtDNA CO I gene in a patient with a clinical phenotype resembling mitochondrial cardioencephalomyopathy. To substantiate pathogenicity, we modeled the identified mutation in the homologous gene in Paracoccus denitrificans and analyzed the biochemical consequences. We observed a deleterious effect on enzyme activity, with a lack of heme a3. Taking advantage of the extensive structural homology between the bacterial enzyme and the mammalian core complex, we conclude that the novel S142F mutation is disease-related. This approach can be used in other cases to support the pathogenicity of novel variants in the mitochondrial genome. Keywords Cytochrome c oxidase . MELAS . mtDNA mutation . Bacterial model . Paracoccus denitrificans

Introduction Cytochrome c oxidase (COX), the terminal component of the mitochondrial respiratory chain, catalyzes the exergonic reduction of molecular oxygen to water. Mammalian COX is a heteromeric complex consisting of three catalytic subunits (CO I to III), encoded by mitochondrial DNA (mtDNA), and ten “accessory” subunits, encoded by nuclear genes. These “accessory” subunits are possibly involved in the structural stabilization of the complex, and probably in the modulation of its catalytic activity, whereas CO I and CO II contain all the redox-active prosthetic groups. The function of CO III remains largely unknown. S. Lucioli . R. Carrozzo . A. Tessa . F. M. Santorelli (*) Molecular Medicine, IRCCS, “Bambino Gesù” Children’s Hospital, Rome, Italy e-mail: [email protected] Fax: +39-06-68592024 K. Hoffmeier . B. Ludwig Molecular Genetics, Institute of Biochemistry, J.W. Goethe Universität, Frankfurt, Germany

The redox centers involved in the electron transfer are two heme A moieties (a and a3) and two copper centers (CuA and CuB) [1–3]. Different clinical phenotypes associated with COX deficiency have been described, generally characterized by early onset of symptoms and fatal outcome. Most of the syndromes are inherited as autosomal recessive traits. To date, no mutation in any of the nucleus-encoded COX subunits has been identified. Instead, all the genetically defined cases so far reported are due to mutations in genes encoding factors involved in the biogenesis and assembly of the complex, including SURF1, COX10, SCO1, SCO2, and COX15 [4–8]. Mutations in the SURF1 gene are by far the most prevalent cause of infantile COX deficiency, whereas mutations in the other nuclear genes involved in COX biogenesis are considerably less common. Maternally inherited mutations in mtDNA-encoded COX subunits are rarer, often occur in adults, and are associated with multisystem diseases, including mitochondrial encephalomyopathy with lactic acidosis and stroke-like episodes (MELAS) [9–19]. In several cases of COX deficiency, however, the responsible gene remains unknown. New mutations in the mitochondrial genome are generally considered disease-related when they fulfill canonical criteria for pathogenicity, which include a high degree of conservation throughout evolution of the mutated residue, heteroplasmy, segregation of the mutation with the disease phenotype in the family, and absence in controls. Nonetheless, pathogenicity remains provisional until the functional significance of the mutation has been established or additional patients have been identified. The three subunits forming the catalytic core of COX are highly conserved among species and the determination of high-resolution structures of the enzyme from beef heart and from Paracoccus denitrificans has confirmed the high degree of structural similarity between the core subunits of the mammalian and bacterial enzymes [20, 21]. We employed a composite strategy to demonstrate the functional significance of a novel mutation (S142F) in CO I in a patient affected by a severe cardioencephalomyopathy resembling MELAS. As consensus criteria for pathogenic

52

changes were not readily applicable in our case, we created a bacterial model of the identified mtDNA alteration—by mutagenesis of the homologous residue in P. denitrificans —and analyzed its biochemical consequences to clarify the effect of mutations in the catalytic core of COX.

histidine tag at the carboxy terminus and mobilized into P. denitrificans AO1, a strain deleted in the genomic copies of both its ctaDI and ctaDII genes, by triparental mating (for experimental details, see [23, 24]). Transconjugant Paracoccus mutant strains were selected for their plasmid-coded streptomycin resistance, and representative clones further characterized.

Methods Case report

Purification of bacterial cytochrome c oxidase

This 30-year-old woman was referred to us with a 14-year history of exercise intolerance, easy fatigue, and muscle weakness. Because she had been adopted, her family history was unknown. Echocardiography, performed when the subject was 22, revealed hypertrophied left ventricle and interventricular septum. A brain MRI taken when the subject was 25 years old showed normal results, whereas resting blood lactate level was elevated (twice normal). At the age of 33, the patient had a stroke-like episode affecting the parieto-occipital cortex; her cardiologic conditions worsened, and the patient died 1 year later of a cardiac arrest. No autopsy was performed. Histochemical stain for oxidative phosphorylation and spectrophotometric measurement of respiratory chain enzyme activities in muscle, isolation of total DNA from muscle homogenate, and screening for the most common pathogenic mtDNA mutations were all performed using standard methods. The entire mtDNA from skeletal muscle was amplified in 13 overlapping fragments and submitted to direct sequencing using BigDye chemistry on an ABI 377 (Applera, Foster City, CA, USA).

Spectroscopic and enzymatic analysis of the mutant oxidases was performed on the purified enzyme complexes in order to identify their molecular properties unequivocally. Strains were grown aerobically at 32°C on a succinate medium [25] that included streptomycin sulfate. Cells were harvested during exponential growth, and membranes obtained by mechanical rupture at 4°C using a Manton– Gaulin homogenizer, operating at 400 bars for 10 min, followed by differential centrifugation. Cytoplasmic membranes were solubilized with n-dodecyl-β-D-maltoside. Cytochrome c oxidase was purified by affinity chromatography on Ni2+-nitrilotriacetic acid agarose (Qiagen, Germany) by means of a subunit I with a carboxy-terminal His6 tag. Correct composition and integrity of the assembled COX complexes was tested by Coomassie stain of SDS-polyacrylamide gel electrophoresis (SDS-PAGE), by Western blotting with monoclonal and polyclonal antibodies directed against subunits I, II, and III of COX, and by redox and ligand binding spectra. Reduced-oxidized difference spectra were recorded between 500 and 650 nm. The oxidized state was obtained by adding ferricyanide to the sample, and the reduced state by adding a few grains of solid sodium dithionite. Quantitation of cytochrome oxidase (heme aa3) was based on the heme A signal, using an extinction coefficient (Δɛred-ox) of 23.4 mM−1 cm−1 at 605–630 nm [26].

General methods Plasmid DNA isolation and restriction analysis, agarose DNA gel electrophoresis, and related techniques were performed essentially as previously reported [22]. Antibiotic concentrations in agar plates and in liquid media were as follows: rifampicin 60 μg/ml, ampicillin 50 μg/ml, streptomycin sulfate 25 μg/ml. Generation of mutants in P. denitrificans In order to create the subunit I site-directed mutants in the P. denitrificans aa3-type oxidase, plasmid pKH40, containing ctaDII (the gene expressing subunit I), was mutagenized using the Altered Sites in vitro mutagenesis protocol (Promega, Heidelberg, Germany) with minor modifications. Experimental details have been previously described ([23, 24]; see also references cited there). Sequences of the mutagenic primers are available on request. In order to create plasmids capable of replicating and expressing in P. denitrificans, a 1,042-bp XbaI–BsrGI fragment was restricted from each of the mutated pKH40 plasmids. Then, it was introduced into pKH163, a derivative of pRI2 containing a copy of ctaDII encoding a six-

COX activity Activity of cytochrome c oxidase was assayed spectrophotometrically at 25°C in 20 mM Tris–HCl (pH 7.5), 1 mM EDTA, and 0.2 g/l dodecyl maltoside. The reaction was measured as the rate of oxidation of reduced horse heart ferrocytochrome c as shown by a decrease in absorbance (A) using an Δɛred-ox of 21 mM−1 cm−1 at 550 nm. The reaction was initiated by the addition of 20 μM ferrocytochrome c. Ligand binding spectra with carbon monoxide Carbon monoxide (CO) difference spectra of oxidase were performed by sequential scanning of the same sample, bubbling over 5 min into 0.5 ml of dithionite-reduced cytochrome aa3 at 25°C. This protocol results in the complete saturation of the oxidase with CO. The amount of COligated heme a3 was estimated by subtracting the dithio-

53

Results

The heme A to protein stoichiometry of each of the purified oxidases was determined by using the pyridine hemochrome method to measure the amount of heme A and the modified Lowry method to determine the amount of protein [27]. Pyridine hemochromogen spectra were obtained from redox samples containing 20% (vol/vol) pyridine, using an Δɛ of 25 mM−1 cm−1 at 587–620 nm [28].

Histochemistry in a deltoid muscle biopsy sample from the patient revealed a low COX stain, whereas the succinate dehydrogenase (SDH) stain showed an increased signal in about 30% of fibers (Fig. 1). Mitochondrial respiratory chain enzyme activities in muscle showed a partial COX deficiency (residual activity relative to citrate synthase was 48% of control values). Direct sequencing of the entire mtDNA from skeletal muscle revealed a homoplasmic novel mutation (6328C>T), which resulted in a missense substitution of serine 142 with phenylalanine (S142F, Fig. 2). The mutation is located at the beginning of the fourth N-terminal transmembrane helix of CO I and it was virtually homoplasmic by a mispairing restriction fragment length polymorphism analysis (Fig. 2). Single muscle fiber studies were not performed due to paucity of residual tissue. Sequencing of the remaining mtDNA led to the identification of three additional variants (4562A>G, 8860G> A, and 13539A>G), which are listed in the public mitochondrial database (http://www.mitomap.org) as polymorphisms. They were therefore considered of unlikely pathogenic significance. We introduced a copy of the ctaDII gene carrying one of the three point mutations (S178F, S178Y, and S178A) into the P. denitrificans host strain AO1. The ctaDII gene is the bacterial homolog of the human mtDNA CO I gene. Bacterial COX variants S178A, S178Y, and S178F (a mutation homologous to that found in our patient) were expressed and studied in the isolated enzyme complex to investigate the pathogenic role of the patient’s mutation. The Coomassie blue-stained SDS-polyacrylamide gel electrophoresis profile of the purified oxidases were found in

Fig. 2 Electropherogram of the mtDNA flanking residue serine 142 in the human CO I gene in a control subject (a) and in the patient (b). The mutation is indicated by an arrow. The amino acid sequence is superimposed. (c) Use of a PCR-restriction fragment length polymorphism (PCR-RFLP) method to confirm the 6328C>T (Ser142Phe) mutation. In brief, a 231-bp fragment was amplified using primers MTF-6130 (5′–3′, GAGGCTTTGGCAACTGAC

TAG) and MTR-6360 (5′–3′, CTAGGTGTAAGGAGAAGATGGT TAGGTCTCCG) (mismatched nucleotide is underlined). Cleavage with the endonuclease MspI (Roche, Hamburg, Germany) generates 133-, 66-, and 32-bp fragments in a normal control (C). The presence of the 6328C>T mutation (P) abolishes a site of cleavage resulting in 133- and 98-bp fragments. M DNA molecular marker size

Fig. 1 Histochemistry in the patient’s biopsied muscle using succinate dehydrogenase (left, a) and cytochrome c oxidase stains (right, b)

nite-reduced spectrum from that of the reduced sample + CO, and then compared with the total amount of cytochrome c oxidase (estimated on the basis of the heme A signal). The amount of CO binding enzyme was estimated by using an Δɛ of 3.5 mM−1 cm−1 at 592–608 nm [26]. Heme A content

54

Fig. 3 Subunit composition of purified cytochrome oxidase variant forms. Equal amounts (10 μg protein) of oxidase samples were separated on 12% Laemmli SDS-PAGE and stained with Comassie Blue. Molecular weight marker was included to check product sizes. The location of subunits I, II, and III are indicated by arrows. Lane A contains S178A oxidase; lane B, S178F; lane C, S178Y; lane D, wild-type oxidase isolated from cells grown to mid-log phase and purified on Ni2+-nitriloacetic acid agarose. The presence of subunit IV (not shown) was confirmed in a different gel system

close agreement with those of the wild-type enzyme, although variants S178F and S178Y showed a dramatic decrease in the amount of subunit III (Fig. 3). An additional band of about 75 kDa was found in S178F mutants, which is likely an aggregation artifact of the SDS gel. As reported elsewhere, subunits II of purified bacterial cytochrome oxidase is resolved as a doublet with different relative intensities [29, 30]. Western blotting, using antibodies to CO I, CO II, and CO III, further confirmed the diminished levels of CO III in the S178F and S178Y mutants (Fig. 4). The enzymatic turnover number of the mutant enzyme S178A (1,039 s−1) was not significantly different from that of the wild-type enzyme (1,096 s−1). In contrast, the activity of the mutant enzymes S178F and S178Y was dramatically decreased (3.0 and 2.3 s−1, respectively).

Fig. 4 Western blot analysis of cytochrome oxidase samples (0.5 μg protein) from the wild-type (a), S178Y (b), S178F (c), and S178A (d) strains. Subunits were separated by 12% Tris-Glycine SDS-PAGE, blotted onto nitrocellulose membrane, and incubated with subunit-specific polyclonal antibodies to CO I and CO II and monoclonal antibodies to CO III

Fig. 5 Ligand binding difference spectra probing for carbon monoxide binding in the purified enzyme variants. Although the CO-reacting S178A enzyme shows a typical 592-nm peak virtually identical to that of the wild-type enzyme, the S178F and S178Y mutant enzyme spectra lack a detectable absorption in this spectral region

The modified proteins were analyzed for heme a and a3 content by visible spectroscopy, and their CO binding capacity was also evaluated. Figure 5 shows the CO difference spectra. A typical CO difference spectrum shows a positive peak at 592 nm (see wt trace in Fig. 5). Although the amount of CO-reacting S178A enzyme was virtually identical to that of the wild-type enzyme, the S178F and S178Y enzymes showed no detectable CO binding. The S178F mutant showed—on the basis of the Lowry value— a heme/protein ratio of 1.10, whereas in the S178A and in the wild-type enzyme, this ratio was 2.19 and 1.90, respectively. Heme determinations, both in native conditions and after pyridine treatment, were consistent with the presence of a single heme A.

Discussion A mutation in mtDNA in our patient was suspected on the basis of her clinical features and the histochemical and biochemical analyses of her muscle tissue. The discovery of a novel, virtually homoplasmic mtDNA mutation in muscle—the only tissue we were able to analyze—in association with a clinical phenotype that has already been associated with mtDNA alterations made it likely that the S142F change was the cause of the patient’s syndrome. This novel mutation affects a highly conserved residue; it was not identified in a large set of ethnically matched control chromosomes and is not listed in the mitochondrial polymorphism database (http://www.mitomap.org). It remains unclear why the S142F mutation has, at the detected mutant load of greater than 95%, only an approximately 50% effect on the maximal activity of COX in skeletal muscle. A similar observation was reported in association with the L196I mutation in the mtDNA CO I gene, proposing also for skeletal muscle tight in vivo control of mitochondrial oxidative phosphorylation by cytochrome c oxidase [31]. An appealing alternative possibility is that the S142L change has a hypomorphic, rather than null, effect,

55

determining a partial reduction, rather than abolition, of the COX catalytic activity, at least in skeletal muscle. As no functional studies could be performed in a cellular model, only indirect evidence is available for the pathogenicity of the S142F variant. We reasoned that the bacterial enzyme, given its relative simplicity, might constitute a suitable model for studying its effects. Previous studies of human mutations in bacterial COX show that the bacterial mutants behave like their human counterparts, and can thus be used to characterize in detail the effects of known disease-related mutations on respiratory function [32]. Ser178 is located at the beginning of the fourth transmembrane helix of COI; it was replaced with phenylalanine in the P. denitrificans gene to model the patient’s homologous mutant protein. We also replaced Ser178 with another bulky residue (tyrosine) and with a smaller residue (alanine) to test whether the side chain containing Ser178 is needed for the catalytic activity of the enzyme. To obtain data unbiased by fluctuating levels of expression of the mutant enzyme complexes, detergent-purified oxidases were analyzed for their integrity of assembly, and for their spectral and enzymatic properties. Introducing the human mutation into P. denitrificans produces a conspicuous enzyme defect. Both the S178F and S178Y mutations had a deleterious effect on enzyme activity, which dropped to a negligible value compared to that of the wild-type enzyme, whereas the S178A enzyme showed a level of activity similar to that of the wild-type enzyme. Ser178 is located far from the redox centers of cytochrome c oxidase, and it is unlikely that mutations of this residue directly affect the catalytic activity of the enzyme. Therefore, spectral analysis of ligand-bound oxidase using CO was carried out to assay for possible changes in the binuclear center where oxygen reduction takes place. Carbon monoxide, mimicking oxygen interaction at this site, binds to reduced heme a3 and induces an optical shift that is easily monitored in the visible region. Contrary to wild-type and S178A enzymes, the absence of affinity for CO observed in variants S178F and S178Y provided initial evidence that both mutant strains lack heme a3. The hypothesis that the heme a3 site was not occupied in the S178F enzyme was strengthened by comparison of the heme content of the isolated protein S178F measured by native and by pyridine hemochrome spectra, which, taken together, ruled out the possibility of any unusual coordination of the heme a moiety. Lack of heme a3 and, as a result, hampered enzymatic activity may arise from a structural defect in the binuclear site that occurs when a bulky side chain (Phe or Tyr) is introduced. Inspection of a space-filling model of the bacterial subunit I from both the bacterial and the mitochondrial oxidase (Fig. 6) reveals that Ser178 is within H-bonding distance of the backbone carbonyl oxygen of Ser148. Upon the replacement of Ser178 with a bulky aromatic side chain, a conformational distortion of this loop region by a tilt or a torsion of transmembrane helix III or IV, or both, might lead to a loss of heme a3, and probably affect the binding of COIII to the enzyme. This result is consistent with the structural data, which show the helix IV of COI close to the

Fig. 6 Schematic representation of human CO I structure (partial view on region of helices III–IV). The structure was modeled with SWISS MODEL [30] using human CO I amino acid sequence and Bos taurus CO I structure (PDB code 1v54)

helix I of COIII, both in P. denitrificans and in Bos taurus [20, 21]. The binding of su III is at least weakened by both the F and Y mutations, and one may speculate that this subunit may still be associated with the complex under native membrane conditions, or in a supercomplex background. Here, we apply more rigid conditions of purification, freeing the enzyme complex, by rigid detergent solubilization, of both its membrane phospholipid environment and of further protein partner complexes. This would also explain the observed differences in enzymatic activity between the patient and the purified bacterial enzyme complex (see also below). Altogether, the data we obtained led us to conclude that the decrease in activity and CO reactivity observed in the bacterial S178F mutant enzyme matched the direct observations of the patient. In view of the extensive structural subunit I homology between the two enzyme complexes, the same effect can be expected in the bacterial and human enzymes, where serine at position 142 is within H-bonding distance of the backbone carbonyl oxygen of Ser115 (Fig. 6), suggesting that the S142F mutation plays a pathogenic role through alteration of the proper folding of the enzyme. Summarizing, we report a new mtDNA variant associated with human pathology and, by means of a simpler bacterial system, demonstrate its functional significance. Under specific circumstances, this system allows the generation of information relating to pathogenicity, which is just as useful as that obtained through the more laborious generation of patient-derived transmitochondrial hybrids. As already reported by our group and others, this approach can be applied—with modifications—to virtually any alteration in the mitochondrial genome and it can also generate further information on structure and function of mito-

56

chondrially encoded components of the respiratory chain [32–36]. Acknowledgements We are indebted to Andrea Herrmann for excellent technical assistance and to Catherine J. Wrenn for kindly editing and reviewing the manuscript. The financial support of the Italian Ministry of Health (“Ricerca corrente 200202G000567” and “Ricerca finalizzata” grants to FMS) and Deutsche Forschungsgemeinschaft (SFB 472 grant to B.L.) is gratefully acknowledged.

References 1. Schultz BE, Chan SI (2001) Structures and proton-pumping strategies of mitochondrial respiratory enzymes. Annu Rev Biophys Biomol Struct 30:23–65 2. Richter OM, Ludwig B (2003) Cytochrome c oxidase—structure, function, and physiology of a redox-driven molecular machine. Rev Physiol Biochem Pharmacol 147:47–74 3. Brzezinski P, Larsson G (2003) Redox-driven proton-pumping by heme-copper oxidases. Biochim Biophys Acta 1605:1–13 4. Tiranti V, Hoertnagel K, Carrozzo R, Galimberti C, Munaro M, Granatiero M, Zelante L, Gasparini P, Marzella R, Rocchi M, Bayona-Bafaluy MP, Enriquez JA, Uziel G, Bertini E, DionisiVici C, Franco B, Meitinger T, Zeviani M (1998) Mutations of SURF-1 in Leigh disease associated with cytochrome c oxidase deficiency. Am J Hum Genet 63:1609–1621 5. Valnot I, von Kleist-Retzow JC, Barrientos A, Gorbatyuk M, Taanman JW, Mehaye B, Rustin P, Tzagoloff A, Munnich A, Rotig A (2000) A mutation in the human heme A farnesyltransferase gene (COX10) causes cytochrome c oxidase deficiency. Hum Mol Genet 9:1245–1249 6. Valnot I, Osmond S, Gigarel NS, Mehaye B, Amiel J, CormierDaire V, Munnich A, Bonnefont JP, Rustin P, Rotig A (2000) Mutations of the SCO1 gene in mitochondrial cytochrome c oxidase deficiency with neonatal-onset hepatic failure and encephalopathy. Am J Hum Genet 67:1104–1109 7. Papadopoulou LC, Sue CM, Davidson MM, Tanji K, Nishino I, Sadlock JE, Krishna S, Walker W, Selby J, Glerum DM, Coster RV, Lyon G, Scalais E, Lebel R, Kaplan P, Shanske S, De Vivo DC, Bonilla E, Hirano M, DiMauro S, Schon EA (1999) Fatal infantile cardioencephalomyopathy with COX deficiency and mutations in SCO2, a COX assembly gene. Nat Genet 23:333–337 8. Antonickà H, Mattman H, Carlson CG, Glerum D, Hoffbuhr KC, Leary SC, Kennaway NG, Shoubridge EA (2003) Mutations in COX15 produce a defect in the mitochondrial heme biosynthetic pathway, causing early-onset fatal hypertrophic cardiomyopathy. Am J Hum Genet 72:101–114 9. Comi GP, Bordoni A, Salani S, Franceschina L, Sciacco M, Prelle A, Fortunato F, Zeviani M, Napoli L, Bresolin N, Moggio M, Ausenda CD, Taanman JW, Scarlato G (1998) Cytochrome c oxidase subunit I microdeletion in a patient with motor neuron disease. Ann Neurol 43:110–116 10. Bruno C, Martinuzzi A, Tang Y, Andreu AL, Pallotti F, Bonilla E, Shanske S, Fu J, Sue CM, Angelini C, DiMauro S, Manfredi G (1999) A stop-codon mutation in the human mtDNA cytochrome c oxidase I gene disrupts the functional structure of complex IV. Am J Hum Genet 65:611–620 11. Karadimas CL, Greenstein P, Sue CM, Joseph JT, Tanji K, Haller RG, Taivassalo T, Davidson MM, Shanske S, Bonilla E, DiMauro S (2000) Recurrent myoglobinuria due to a nonsense mutation in the COX I gene of mitochondrial DNA. Neurology 55:644–649 12. Clark KM, Taylor RW, Johnson MA, Chinnery PF, Chrzanowska-Lightowlers ZM, Andrews RM, Nelson IP, Wood NW, Lamont PJ, Hanna MG, Lightowlers RN, Turnbull DM (1999) An mtDNA mutation in the initiation codon of the cytochrome C oxidase subunit II gene results in lower levels of the protein and a mitochondrial encephalomyopathy. Am J Hum Genet 64:1330–1339

13. Rahman S, Taanman JW, Cooper JM, Nelson I, Hargreaves I, Meunier B, Hanna MG, Garcia JJ, Capaldi RA, Lake BD, Leonard JV, Schapira AH (1999) A missense mutation of cytochrome oxidase subunit II causes defective assembly and myopathy. Am J Hum Genet 65:1030–1039 14. Campos Y, Garcia-Redondo A, Fernandez-Moreno MA, Martinez-Pardo M, Goda G, Rubio JC, Martin MA, del Hoyo P, Cabello A, Bornstein B, Garesse R, Arenas J (2001) Earlyonset multisystem mitochondrial disorder caused by a nonsense mutation in the mitochondrial DNA cytochrome c oxidase II gene. Ann Neurol 50:409–413 15. Keightley JA, Hoffbuhr KC, Burton MD, Salas VM, Johnston WS, Penn AM, Buist NR, Kennaway NG (1996) A microdeletion in cytochrome c oxidase (COX) subunit III associated with COX deficiency and recurrent myoglobinuria. Nat Genet 12: 410–416 16. Manfredi G, Schon EA, Moraes CT, Bonilla E, Berry GT, Sladky JT, DiMauro S (1995) A new mutation associated with MELAS is located in a mitochondrial DNA polypeptide coding gene. Neuromuscul Disord 5:391–398 17. Hanna MG, Nelson IP, Rahman S, Lane RJ, Land J, Heales S, Cooper MJ, Schapira AH, Morgan-Hughes JA, Wood NW (1998) Cytochrome c oxidase deficiency associated with the first stop-codon point mutation in human mtDNA. Am J Hum Genet 63:29–36 18. Tiranti V, Corona P, Greco M, Taanman JW, Carrara F, Lamantea E, Nijtmans L, Uziel G, Zeviani M (2000) A novel frameshift mutation of the mtDNA COIII gene leads to impaired assembly of cytochrome c oxidase in a patient affected by Leigh-like syndrome. Hum Mol Genet 9:2733–2742 19. Hoffbuhr KC, Davidson E, Filiano BA, Davidson M, Kennaway NG, King MP (2000) A pathogenic 15-base pair deletion in mitochondrial DNA-encoded cytochrome c oxidase subunit III results in the absence of functional cytochrome c oxidase. J Biol Chem 275:13994–14003 20. Iwata S, Ostermeier C, Ludwig B, Michel H (1995) Structure at 2.8 Å resolution of cytochrome c oxidase from Paracoccus denitrificans. Nature 376:660–669 21. Tsukihara T, Aoyama H, Yamashita E, Tomizaki T, Yamaguchi H, Shinzawa-Itoh K, Nakashima R, Yaono R, Yoshikawa S (1996) The whole structure of the 13-subunit oxidized cytochrome c oxidase at 2.8 Å. Science 272:1136–1144 22. Sambrook J, Russel DW (2004) Molecular cloning: a laboratory manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY 23. Pfitzner U, Odenwald A, Ostermann T, Weingard L, Ludwig B, Richter OMH (1998) Cytochrome c oxidase (heme aa3) from Paracoccus denitrificans: analysis of mutations in putative proton channels of subunit I. J Bioenerg Biomembr 30: 89–97 24. Pfitzner U, Hoffmeier K, Harrenga A, Kannt A, Michel H, Bamberg E, Richter OMH, Ludwig B (2000) Tracing the Dpathway in reconstituted site-directed mutants of cytochrome c oxidase from Paracoccus denitrificans. Biochemistry 39:6756– 6762 25. Ludwig B (1986) Cytochrome c oxidase from Paracoccus denitrificans. Methods Enzymol 126:153–159 26. Ludwig B, Schatz G (1980) A two-subunit cytochrome c oxidase (cytochrome aa3) from Paracoccus denitrificans. Proc Natl Acad Sci U S A 77:196–200 27. Markwell MA, Haas SM, Bieber LL, Tolbert NE (1978) A modification of the Lowry procedure to simplify protein determination in membrane and lipoprotein samples. Anal Biochem 87:206–210 28. Berry EA, Trumpower BL (1987) Simultaneous determination of hemes a, b, and c from pyridine hemochrome spectra. Anal Biochem 161:1–15 29. Zhen Y, Qian J, Follmann K, Hayward T, Nilsson T, Dahn M, Hilmi Y, Hamer AG, Hosler JP, Ferguson-Miller S (1998) Overexpression and purification of cytochrome c oxidase from Rhodobacter sphaeroides. Protein Expr Purif 13:326–336

57 30. Hosler JP, Fetter J, Tecklenburg MM, Espe M, Lerma C, Ferguson-Miller S (1992) Cytochrome aa3 of Rhodobacter sphaeroides as a model for mitochondrial cytochrome c oxidase. Purification, kinetics, proton pumping, and spectral analysis. J Biol Chem 5(267):24264–24272 31. Varlamov DA, Kudin AP, Vielhaber S, Schröder R, Sassen R, Becker A, Kunz D, Haug K, Rebstock J, Heils A, Elger CE, Kunz WS (2002) Metabolic consequences of a novel missense mutation of the mtDNA CO I gene. Hum Mol Genet 11:1797– 1805 32. Bratton M, Mills D, Castleden CK, Hosler J, Meunier B (2003) Disease-related mutations in cytochrome c oxidase studied in yeast and bacterial models. Eur J Biochem 270:1222–1230 33. Schwede T, Kopp J, Guex N, Peitsch MC (2003) SWISSMODEL: an automated protein homology-modeling server. Nucleic Acids Res 31:3381–3385

34. Lunardi J, Darrouzet E, Dupuis A, Issartel JP (2004) The nuoM arg368his mutation in NADH:ubiquinone oxidoreductase from Rhodobacter capsulatus: a model for the human nd4-11778 mtDNA mutation associated with Leber’s hereditary optic neuropathy. Biochim Biophys Acta 1407:114–124 35. Kerscher S, Grgic L, Garofano A, Brandt U (2004) Application of the yeast Yarrowia lipolytica as a model to analyse human pathogenic mutations in mitochondrial complex I (NADH: ubiquinone oxidoreductase). Biochim Biophys Acta 1659:197– 205 36. Carrozzo R, Murray J, Santorelli FM, Capaldi RA (2000) The T9176G mutation of human mtDNA gives a fully assembled but inactive ATP synthase when modeled in Escherichia coli. FEBS Lett 486:297–299