Mol Biol Rep (2012) 39:5767–5774 DOI 10.1007/s11033-011-1387-9

Molecular characterization and tissue expression profile of three novel ovine : ATP5O, NDUFA12 and UQCRH from muscle full-length cDNA library of black-boned sheep

R. S. Ye • H. B. Pan • G. F. Yin • Y. Huang • S. M. Zhao • S. Z. Gao

Received: 9 October 2010 / Accepted: 17 December 2011 / Published online: 3 January 2012 Ó Springer Science+Business Media B.V. 2012

Abstract Three novel ovine genes were obtained from no strand in UQCRH. One potential signal peptide structure muscle full-length cDNA library of black-boned sheep. in ATP5O protein were found. NDUFA12 and UQCRH Sequence analysis revealed that nucleotide sequences of have the extremely low possibility of signal peptides. these genes were not homologous to any of the known sheep Meanwhile, RasMol was used for visualizing the PDB files or goat genes, but these genes have high similarity to ATP generated by Swiss-Model in cartoon or three-dimensional synthase subunit O (ATP5O), NADH dehydrogenase (ubi- format. ATP5O and UQCRH protein were modeled by quinone) 1 alpha subcomplex, 12 (NDUFA12) and ubi- Swiss-Model. Tissue expression profile indicated that the quinol-cytochrome c reductase hinge protein (UQCRH) ovine ATP5O, NDUFA12 and UQCRH genes could be genes of other mammal animals (accession number: expressed in all detected tissues including muscles, heart, FJ546085, FJ546078 and FJ546083). The alignment anal- liver, spleen, lung, kidney and adipose tissues, but the ysis showed that the ovine ATP5O, NDUFA12 and expression abundance of these genes were various in the UQCRH genes and proteins have closer genetic relation- different tissues. Our experiment supplied the primary ships with the ATP5O, NDUFA12 and UQCRH genes and foundation for further researches on these three ovine genes. proteins from cattle. Conserved domain prediction showed that these three genes included OSCP, NDUFA12 super- Keywords Sheep ATP5O NDUFA12 UQCRH family and UCR-hinge superfamily domains respectively. Structure predication Tissue expression The deduced sequence of ATP5O, NDUFA12 and UQCRH protein had 213, 145 and 91 residues, with a molecular weight of approximately 23419.66, 17089.50 and Introduction 10657.75 Da and a theoretical isoelectric point of 9.90, 9.68 and 4.45. The secondary structure prediction revealed that Construction of a full-length cDNA library is an effective 60% helix structure in ATP5O, 60% coils in NDUFA12 and way to understand the functional expression of genes in muscle tissue, and in addition, novel genes for further R. S. Ye and H. B. Pan have contributed equally to this work and research can be found in the library. Muscle full-length could be regarded as the first authors. cDNA library of black-bone sheep was constructed in our previous study [1]. In the subsequent study, many new Electronic supplementary material The online version of this ovine genes were founded by conducting large-scale article (doi:10.1007/s11033-011-1387-9) contains supplementary material, which is available to authorized users. sequencing of colonies. A novel PSAM6 of black- bone sheep was reported in 2009 [1]. Meanwhile, three R. S. Ye H. B. Pan G. F. Yin Y. Huang other novel ovine genes-ATP synthase subunit O (ATP5O), & & S. M. Zhao ( ) S. Z. Gao ( ) NADH dehydrogenase (ubiquinone) 1 alpha subcomplex, Yunnan Key Laboratory of Animal Nutrition and Feed Science, Yunnan Agricultural University, Kunming 650201, China 12 (NDUFA12) and ubiquinol-cytochrome c reductase e-mail: [email protected] hinge protein (UQCRH) were isolated and characterized S. Z. Gao from the same muscle full-length cDNA library of black- e-mail: [email protected] boned sheep. 123 5768 Mol Biol Rep (2012) 39:5767–5774

Defects in mitochondrial biogenesis are the most fre- containing 10 units of MMLV reverse transcriptase (Pro- quent cause of mitochondrial disorders [2]. ATP5O, mega, Belgium), 1 mM dNTP mixture (Promega, Bel- NDUFA12 and UQCRH are nuclear encoded mitochondrial gium), 40 units of recombinant RNasin ribonuclease genes [3–5], which are involved in oxidative phosphoryla- inhibitor (Promega, Belgium) and 0.5 ll of oligo (dT) 18 tion (OXPHOS). OXPHOS is a process where electrons are (Promega, Belgium) in sterilized water and buffer supplied passed from NADH and FADH2 along a series of carrier by the manufacturer. After incubation at 42°C for 60 min, molecules [2, 6]. Protons are pumped across the inner the mixture was heat treated at 95°C for 5 min. cDNA mitochondrial membrane to produce a proton gradient and samples were kept in -20°C for Real-time PCR detection. in a final step ATP is produced from ADP and phosphate. This process occurs in the respiratory chain, which consists Sequence analysis of five multiprotein complexes [7]. ATP5O is a nuclear encoded subunit of complex V of the respiratory chain and The cDNA sequence prediction was conducted using Gen- influences the transmission of conformational changes and Scan software (http://genes.mit.edu/GENSCAN.html). The proton conductance [3]. NDUFA12 encodes an accessory protein prediction and analysis were performed using the subunit of complex I which is required either at a late step in Conserved Domain Architecture Retrieval Tool of BLAST the assembly of complex I, or in the stability of complex I at the National Center for Biotechnology Information [4]. The gene coding for mitochondrial hinge protein (NCBI) server(http://blast.ncbi.nlm.nih.gov/), the ClustalW UQCRH gene is a subunit of complex III [5]. software (http://www.ebi.ac.uk/Tools/clustalw2/index.html) In the present experiment, we will analyze the coding and DNAstar software. sequences of ovine ATP5O, NDUFA12 and UQCRH genes Isoelectric point and molecular weight (pI/Mw) of ovine basing on the conserve sequence information of human, three novel gene-coding proteins (ATP5O, NDUFA12 and mouse or cattle, compared the homology of ovine EST UQCRH) was analyzed in compute pI/Mw (http://web. sequence information with coding sequences of these three expasy.org/compute_pi/)[8]. Secondary structure of these genes of cattle or other mammals, subsequently perform proteins was predicted using PSIPRED (http://bioinf.cs.ucl. some necessary protein physical and chemical character- ac.uk/psipred/)[9]. DNAstar software was employed to istics as well as structure predication of deduced proteins, predict these proteins’ physiology characteristics which and finally detect the distribution of these include hydrophobicity, amphilicity, flexibility and genes in different tissues. The study will establish the antigenecity. primary foundation for understanding these three ovine Signal peptides prediction (neural networks signal pep- genes. tides prediction, signalP-NN and hidden Markov model signal peptides prediction, signalP-HMM) was accom- plished by SignalP 3.0 Server (http://www.cbs.dtu.dk/ Materials and methods services/SignalP/)[10]. Transmembrane helices prediction was conducted in TMpred analysis (http://www.ch.embnet. Sample preparation, total RNA extraction and reverse org/software/TMPRED_form.html)[11]. Swiss-Model (http:// transcript swissmodel.expasy.org/) was employed to homology mod- eling of three ovine proteins, and RasMol V2.7.5 was used to Twelve black-bone sheep (six females and six males) at the visualize the PDB files generated by Swiss-Model [12]. age of 12 weeks were slaughtered, respectively. The lon- gissimus dorsi muscles and other tissue samples including heart, liver, spleen, lung, kidney and adipose tissue of all Tissue expression profile animals were collected. Total RNA was extracted from all the samples using the Total RNA Extraction Kit (Invitro- Twelve samples of each tissue in black-boned sheep were gen, America) as per the manufacturer’s instructions. Total used. Real-time PCR was performed using the IQTM RNA concentration was then quantified by measuring the SYBR Green Supermix (Bio-Rad) following the manu- absorbance at 260 nm in a photometer (Eppendorf Bio- facturer’s instructions. The 18S rRNA (195 bp) gene was photometer). Ratios of absorption (260/280 nm) of all regarded as an internal control. The specific primers were: preparations were between 1.8 and 2.0. Aliquots of RNA 18S rRNA, 50-GCGGCTTTGGTGACTCTA-30 (forward samples were subjected to electrophoresis through a 1.4% primer) and 50-CTGCCTCCTTGGATGTG-30 (reverse agarose formaldehyde gel to verify their integrity. RNA primer); ATP5O gene, 50-GGTGATTTCCGCATTT-30 was stored at -80°C prior to use. (forward primer) and 50-TTCAGTCAGAGTGGCTTC-30 Reverse transcription was performed using the RNA (reverse primer); NDUFA12 gene, 50-GCAAATGATGT (2 lg) described above in a final volume of 25 ll GAGGGTT-30 (forward primer) and 50-TTACGAGCAG 123 Mol Biol Rep (2012) 39:5767–5774 5769

TAGGTGGTT-30 (reverse primer); UQCRH gene, 50-AG cattle (100%), human (89%) and mouse (90%). The ovine GACGAGCAAAGGATG-30 (forward primer) and 50-GC UQCRH protein has high homology with UQCRH protein CCAGGTGATGAAGG-30 (reverse primer). Conditions for from cattle (98%), human (94%) and mouse (93%).The cycling were 94°C for 4 min, followed by 40 cycles of ovine ATP5O, NDUFA12 and UQCRH have common denaturation at 94°C for 1 min; annealing at 50°C for conserved domains with highly homologous proteins from 1 min for ATP5O gene; 61°C for 1 min for NDUFA12 other mammals (Table 1; Fig. S2 in Supplementary gene; 60°C for 1 min for UQCRH gene; 55°C for 1 min for material). 18S rRNA gene; extension at 72°C for 1 min; and fluo- From the results obtained above, it can be concluded rescence reading at 82°C for 10 s. The results were that these genes belong to the sheep-ATP5O, NDUFA12 expressed as fold changes described previously by Zhao and UQCRH genes. Based on the results of the alignment et al. [13]. of ATP5O, NDUFA12 and UQCRH proteins in four ani- mals, phylogenetic trees were constructed using the DNAstar software (Fig. S3 in Supplementary material). Results Physical and chemical characteristics of deduced Analysis of sequence identity and phylogenetic tree proteins

The nucleotide sequence analysis using the BLAST soft- The deduced amino acid sequence of ATP5O, NDUFA12 ware at NCBI server (http://www.ncbi.nlm.nih.gov/ and UQCRH, had 213, 145 and 91 residues, with a Mw of BLAST) revealed that these genes were not homologous approximately 23419.66, 17089.50 and 10657.75 Da and a to any of the known sheep or goat genes and they were then theoretical pI of 9.90, 9.68 and 4.45, predicted by the deposited into the GenBank database (Accession number: Compute pl/Mw program (Table 1). The results obtained FJ546085, FJ546078 and FJ546083).The sequence predic- from the protein package in DNAstar software is shown in tion was carried out using the GenScan software and results Fig. 1. Algorithms used in this analysis were hydropho- showed that the 763-, 563-, and 539-bp cDNA sequences bicity plot, flexible regions, antigenic index and surface represent three single genes which encoded 213, 145 and accessibility. 91 amino acids, respectively. Further BLAST analysis of these CDS revealed that Prediction and analysis of protein structures sheep ATP5O gene has high homology with ATP5O genes and conserved domains from cattle (96%), human (86%), and mouse (79%). The sheep NDUFA12 gene has high homology with NDUFA12 The putative domain of the protein encoded by the ATP5O, gene from cattle (99%), human (91%), and mouse (86%). NDUFA12 and UQCRH genes showed that OSCP domain The sheep UQCRH gene has high homology with UQCRH of sheep ATP5O gene, NDUFA12 domain of sheep gene from cattle (96%), human (89%), and mouse (83%) NDUFA12 gene and UCR-hinge domain of sheep UQCRH (Table 1; Fig. S1 in Supplementary material). gene (Fig. S4 in Supplementary material). These putative proteins were also blasted using the PSIPRED can predict protein secondary structure with Conserved Domain Architecture Retrieval Tool of BLAST higher accuracy based on its position-specific iterated at the NCBI server (http://www.ncbi.nlm.nih.gov/BLAST) BLAST (PSI-BLAST) Algorithm. The prediction of ATP5O and their conserved domain were identified as ATP5O, protein revealed that helix structure account for about 60%, NDUFA12 and UQCRH protein. Further BLAST analysis especially position from 35 to 80, and position 85–135 and of these proteins showed that ovine ATP5O protein has four strand structures were found in the whole protein, the high homology with ATP5O protein from cattle (97%), remaining structures are coils that account for 25%. Con- human (86%) and mouse (81%). The ovine NDUFA12 versely, among NDUFA12 protein structure, account about protein has high homology with NDUFA12 protein from one quarter of the whole secondary structure composition,

Table 1 Homology of genes Accession Gene Nucleotide sequence (%) Amino acid sequence (%) pI and weight of and pI and weight of deduced number deduced protein protein Cattle Human Mouse Cattle Human Mouse

FJ546085 ATP5O 96 86 79 97 86 81 9.90/23419.66 FJ546078 NDUFA12 99 91 86 100 89 90 9.68/17089.50 FJ546083 UQCRH 96 89 83 98 94 93 4.45/10657.75

123 5770 Mol Biol Rep (2012) 39:5767–5774

Fig. 1 Secondary structure of three ovine novel gene-coding proteins. a ATP5O, b NDUFA12, c UQCRH

while coils account for 60%, three strands found. Interest- UQCRH, both signalP-NN and signalP-HMM prediction ingly, there is no strand in UQCRH secondary structure, revealed that there has an extremely low possibility of helix structures take major part, which may determined by signal peptides (NDUFA12 = 0.001, UQCRH = 0.000, length of UQCRH, all these results were shown in Fig. 2. not plotted in figure). There are two-method used for signal peptide prediction in SignalP 3.0 Server, which are signalP-NN and signalP- Swiss-Model homology modeling HMM. SignalP-NN showed that there was one potential signal peptide structure in ATP5O protein, and its cleavage One model of ovine ATP5O was selected (E value: 4.21e- site likely located between position 15 and 16 (MA- 54, QMEAN Z score: -2.57, QMEAN4: 0.55). Ovine ALAVSGLSQQVRCFSTS). SignalP-HMM prediction ATP5O protein was modeled by taking Bovine oligomycin indicated the similar results with high probability. Both sensitivity conferral protein N-terminal domain (ID: results were shown in Fig. 3. Regarding to NDUFA12 and 2bo5A) as template. ATP5O modeled residue range from

123 Mol Biol Rep (2012) 39:5767–5774 5771

Fig. 2 Hydrophobicity, amphilicity, flexibility and antigenecity of three ovine novel gene-coding proteins, a ATP5O, b NDUFA12, c UQCRH

24 to 143, which has 95.83% sequence identity with tem- tissues (Fig. 5). The ovine ATP5O gene was highly plate sequence. Similarly, UQCRH protein (Modeled res- expressed in muscle and liver tissues. The ratio of ATP5O idue range: 26 to 91) was modelled by the template of gene to 18S rRNA gene in muscle and liver tissues reached Chicken cytochrome bc1 complex [1bccH] (E value: 1.73 and 1.22 respectively. The ovine NDUFA12 gene was 2.86e-21, QMEAN Z score: -1.22, QMEAN 4:0.57), it is highly expressed in spleen and weakly expressed in adipose worth mentioning that modeling sequence has a 100% tissue. The relative expression abundances of NDUFA12 identity with template sequence. For NDUFA12, there are gene in these two tissues were 1.88 and 0.29 respectively. no templates in database available for modeling. Cartoon The relative expression level of ovine UQCRH gene was and molecular surface models of ATP5O and UQCRH, the lowest in lung, which was 0.46. quality plots of these two proteins were shown in Fig. 4.

Tissue expression profiles Discussion

Tissue expression profile indicated that the ovine ATP5O, ATP5O, NDUFA12 and UQCRH genes are nuclear enco- NDUFA12 and UQCRH genes were expressed in all ded mitochondrial genes [3–5] which are involved in detected tissue, but the expression abundance of ATP5O, OXPHOS [2, 6]. The OXPHOS system consists of five NDUFA12 and UQCRH genes varies in the different multisubunit complexes, complex I, II, III, IV and V [14].

123 5772 Mol Biol Rep (2012) 39:5767–5774

Fig. 3 ATP5O signal peptides analysis using signalP-NN and signalP-HMM prediction in SignalP 3.0 Server. a signalP- NN prediction, Cmax = 0.589 (cutoff: 0.32); Ymax = 0.336 (cutoff: 0.33); Smax = 0.600 (cutoff: 0.87). Its cleavage site most likely located between position 15 and 16. b signalP- HMM prediction also show high probability of signal peptide (0.899), the cleavage sites is between position 15 and 16 (0.854 probability)

ATP5O is a subunit of complex V [3], which is a subunit of Based on the important functions described above, mitochondrial ATP synthase present in the stalk region OXPHOS associated with energy metabolism, and the vast between the F0 and F1 segments, crucial for coupling the majority of cellular ATP is produced by mitochondrial energy to the synthesis of ATP [10]. NDUFA12, along with OXPHOS [17], so they are potentially related to ovine other NADH dehydrogenase subcomplex formed complex production. The study of these genes is of great importance I[4]. UQCRH, namely ubiquinol-cytochrome c reductase to increase ovine health, adaptability, and meat quality. hinge protein, is a subunit of complex III [5]. The product Our experiment established the primary foundation for of NDUFA12 and UQCRH genes serve as electrons further research on these three ovine genes. transporter of the respiratory chain. Electron-transfer pro- Three OXPHOS associated genes were isolated from the cess is coupled to the pumping of protons to the inter- muscle full-length cDNA library in our previous experiment membrane space, creating an electrochemical gradient [1]. This proved that cDNA library is a basic and efficient which is used by ATP synthase to produce ATP [15, 16]. In tool for finding new genes. In this study, we not only isolated conclusion, ATP5O, NDUFA12 and UQCRH genes play and cloned the complete coding sequence of ovine ATP5O, an important role in ATP production. NDUFA12 and UQCRH genes but also performed the Recent studies showed that differential expression of sequence analysis and physical and chemical characteristics these genes resulted in mitochondrial dysregulation, and and structure predication as well as tissue expression profile brought about a number of mitochondrial diseases [2]. analysis. The results of alignments showed that the homol- Chronic kidney disease is associated with overexpression ogy of the ovine ATP5O, NDUFA12 and UQCRH genes and of ATP5O, UQCRH and other OXPHOS associated genes protein more close to the genes and proteins from the cattle. [17]. ATP5O skeletal muscle mRNA expression can reg- Chen et al. [18] cloned a cDNA corresponding to human ulate in vivo glucose metabolism and influence insulin ATP5O. The predicted polypeptide is 213 amino acids long sensitivity [4]. 123 Mol Biol Rep (2012) 39:5767–5774 5773

Fig. 4 The predicted cartoon model, molecular surface model of ATP5O, UQCRH protein and its model quality plot. a Cartoon model. The left is ATP5O, the right is UQCRH; b Molecular surface model. The left is ATP5O, the right is UQCRH. c Estimated absolute quality of models. The left is ATP5O, the right is UQCRH. Regarding to local model quality, estimated per-residue inaccuracy visualized using a color gradient from white (more reliable regions) to black (potentially unreliable regions)

with more than 80% identity to the bovine and mouse format. ATP5O, UQCRH protein were Modeled by Swiss- homologs [18]. pI/Mw calculation is good for protein iso- Model, only high quality models with low E value, low lation and purification. Compute pI/Mw was used for ana- absolute value Z score, high QMEAN4 value were selected lyzing ATP5O, NDUFA12 and UQCRH protein according [19]. to their deduced amino acids sequence. The results indicated From its expression distribution, these genes were that these protein possessed different pI/Mw. DNA- detected in the muscles, heart, liver, spleen, lung, kidney STAR.Lasergene.v7.1 package is universal, powerful and and adipose tissues. This data indicated that these genes graphical software for sequence analysis. As a main protein may serve as the basic function in cell of different tissues. prediction according to amino acid sequences, protean was However, the expression abundance varied in the different employed to predict ATP5O, NDUFA12, and UQCRH tissues. Northern blot analysis of ATP5O detected proteins’ physiology characteristics. It is easy to get hydro- expression in all human tissues examined and at highest phobicity, amphilicity, flexibility and antigenecity position levels in muscle and heart [18]. To explain these tissue of amino acid residues of the corresponding protein, which is expression differences explicitly, further research based on beneficial for analyzing senior structure and immunology these primary results is needed. characteristics. In conclusion, we first isolated the ovine ATP5O, Swiss-Model is classical online modeling software, and NDUFA12 and UQCRH genes and performed necessary widely used for modeling unknown protein sequence by sequence and bio-informational analysis and tissue homology. RasMol was used for visualizing the PDB files expression profile detection. This established the primary generated by Swiss-Model in cartoon or three-dimensional foundation for further research on these three ovine genes.

123 5774 Mol Biol Rep (2012) 39:5767–5774

a muscle ATP50 mRNA expression and glucose uptake in young 2.5 twins. PLoS ONE 4(3):e4793 4. Ostergaard E, Rodenburg RJ, van den Brand M, Thomsen LL, 2 Duno M, Batbayli M, Wibrand F, Nijtmans L (2011) Respiratory chain complex I deficiency due to NDUFA12 mutations as a new 1.5 cause of . J Med Genet. doi:10.1136/jmg. 2011.088856 1 5. Modena P, Testi MA, Facchinetti F, Mezzanzanica D, Radice 0.5 MT, Pilotti S, Sozzi G (2003) UQCRH gene encoding mito- ATP5O gene chondrial Hinge protein is interrupted by a translocation in a soft- tissue sarcoma and epigenetically inactivated in some cancer cell Relative expression of 0 heart liver spleen lung kidney adipose muscle lines. Oncogene 22:4586–4593 Tissues 6. Garbian Y, Ovadia O, Dadon S, Mishmar D (2010) Gene b expression patterns of oxidative phosphorylation complex I sub- 2.5 units are organized in clusters. PLoS ONE 5(4):e9985 7. Devenish RJ, Prescott M, Boyle GM, Nagley P (2000) The oli- 2 gomycin axis of mitochondrial atp synthase: OSCP and the pro- ton channel. J Bioenerg Biomembr 32:507–515 1.5 8. Bjellqvist B, Hughes GJ, Pasquali CH, Paquet N, Ravier F, Sanchez JCH, Frutiger S, Hochstrasser DF (1993) The focusing 1 positions of polypeptides in immobilized pH gradients can be predicted from their amino acid sequences. Electrophoresis 0.5

NDUFA12 gene 14:1023–1031

Relative expression of 9. Bryson K, McGuffin LJ, Marsden RL, Ward JJ, Sodhi JS, Jones 0 heart liver spleen lung kidney adipose muscle DT (2005) Protein structure prediction servers at University Tissues College London. Nucleic Acids Res 33(Web Server issue):W36– c W38 1.2 10. Bendtsen JD, Nielsen H, von Heijne G, Brunak S (2004) Improved prediction of signal peptides: SignalP 3.0. J Mol Biol 1 340:783–795 0.8 11. Hofmann K, Stoffel W (1993) TMbase—a database of membrane spanning proteins segments. Biol Chem Hoppe-Seyler 374:166 0.6 12. Arnold K, Bordoli L, Kopp J, Schwede T (2006) The SWISS- 0.4 MODEL Workspace: a web-based environment for protein UQCRH gene 0.2 structure homology modelling. Bioinformatics 22:195–201 13. Zhao S, Ma H, Zou S, Chen W (2007) Effects of in ovo Relative expression of 0 heart liver spleen lung kidney adipose muscle administration of DHEA on lipid metabolism and hepatic lipo- Tissues genetic genes expression in broiler chickens during embryonic development. Lipids 42:749–775 Fig. 5 Tissue expression profile of three ovine genes. Relative 14. Berg JM, Tymoczko JL, Stryer L (2002) Biochemistry, 5th edn. expression level of genes was indicated by the ratio of genes to 18S WH Freeman. Co., New York rRNA gene. a ATP5O, b NDUFA12, c UQCRH 15. Carolyn MS, Eric AS (2000) Mitochondrial respiratory chain diseases and mutations in nuclear DNA: A Promising Start? Brain Pathol 10:442–450 16. Joshi S, Javed AA, Gibbs LC (1992) Oligomycin sensitivity- Acknowledgments This work was supported by National Key conferring protein (OSCP) of mitochondrial ATP synthase. The Foundation Research Development Project of China (973 Project, No. carboxyl-terminal region of OSCP is essential for the reconsti- 2007CB116201). tution of oligomycin-sensitive H?-ATPase. J Biol Chem 267:12860–12867 17. Carroll J, Fearnley IM, Skehel JM, Shannon RJ, Hirst J, Walker JE (2006) Bovine complex I is a complex of 45 different subunits. References J Biol Chem 281:32724–32727 18. Chen H, Morris MA, Rossier C, Blouin J-L, Antonarakis SE (1995) Cloning of the cDNA for the human ATP synthase OSCP 1. Hu H, Liu YG, Zhao SM, Deng WD, Gao SZ (2009) Molecular subunit (ATP50) by exon trapping and mapping to characterization and tissue expression of ovine PSAM6 gene 21q22.1–q22.2. Genomics 28:470–476 from muscle Full-Length cDNA library of Black-Boned Sheep. 19. Benkert P, Biasini M, Schwede T (2011) Toward the estimation Anim Biotechnol 20(4):238–241 of the absolute quality of individual protein structure models. 2. DiMauro S, Schon EA (2003) Mitochondrial respiratory-chain Bioinformatics 27(3):343–350 diseases. N Engl J Med 348:2656–2668 3. Ro¨nn T, Poulsen P, Tuomi T, Isomaa B, Groop L, Vaag A, Ling C (2009) Genetic variation in ATP5O is associated with skeletal

123