FEBS Letters 588 (2014) 450–454

journal homepage: www.FEBSLetters .org

Parallel evolution of IDH2 gene in cetaceans, primates and bats

0014-5793/$36.00 � 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.febslet.2013.12.005

⇑ Corresponding author at: State Key Laboratory of Emerging Infectious Diseases,Li Ka Shing Faculty of Medicine, University of Hong Kong, Hong Kong.

E-mail address: shenyy@mail.kiz.ac.cn (Y.-Y. Shen).

Wei-Ming Ai a, Shao-Bo Chen a, Xiao Chen a,e, Xue-Juan Shen b, Yong-Yi Shen b,c,d,⇑a Department of Marine Science, School of Life Science, Wenzhou Medical College, Wenzhou 325035, Chinab Joint Influenza Research Centre (SUMC/HKU), Shantou University Medical College, Shantou 515041, Chinac State Key Laboratory of Emerging Infectious Diseases, Li Ka Shing Faculty of Medicine, University of Hong Kong, Hong Kongd State Key Laboratory of Genetic Resources and Evolution, Kunming Institute of Zoology, the Chinese Academy of Sciences, Kunming 650223, Chinae Guangxi Key Lab for Mangrove Conservation and Utilization, Guangxi Mangrove Research Center, Beihai 536000, China

a r t i c l e i n f o

Article history:Received 14 August 2013Revised 3 December 2013Accepted 13 December 2013Available online 25 December 2013

Edited by Takashi Gojobori

Keywords:DolphinBatAdaptive evolutionEnergy metabolismParallel evolution

a b s t r a c t

Cetaceans and primates both have large brains that require large amounts of aerobic energy metab-olism. In bats, the cost of flight makes locomotion energetically demanding. These mammaliangroups may represent three independent evolutionary origins of an energy-demanding lifestyle inmammals. IDH2 encodes an enzyme in the tricarboxylic acid cycle in the mitochondrion, whichplays a key role in aerobic energy metabolism. In this study, we cloned and sequenced this genein two cetaceans, and 19 bat species, and compared the data with available primate sequences totest its evolution. We found significant signals of parallel evolution in this gene among these threegroups. Parallel evolution of this gene may reflect their parallel evolution towards a higher demandfor energy.� 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction

Among mammals, primates have the largest brains relative tobody size. Together with primates, cetaceans are among the bright-est mammals. The absolute brain mass of the Bottlenose Dolphin(Tursiops truncatus) is 1500–1700 g, which is slightly greater thanthat of humans (1300–1400 g) [1] and about four times that ofthe chimpanzee (400 g) [2]. Even after controlling for body mass,the brain-to-body mass ratios of dolphins are among the highestin non-human mammals. Neurons are the most energy-demandingcells [3], thus a large brain requires greater amounts of aerobic en-ergy metabolism, and adaptive evolution of energy metabolismgenes in large brain mammals has been documented [4,5].

Bats are the only mammals that developed the ability of sus-tained (flapping) flight. Flight is among the most energy demandingof all activities. Similar to birds, bats require energy metabolic ratesthat are 3–5 times greater than the maximum observed during exer-cise in similar-sized terrestrial mammals [6–8]. Hence, a significantmetabolic barrier must have been crossed to allow the evolution offlight [8].

Bats, cetaceans and primates have very different lifestyles, butthey independently developed higher energy demands during their

evolution. Their energy metabolism genes have undergone adap-tive evolution to fit their high energy demands [9–14]. Here weask whether some energy metabolism genes have evolved conver-gently to adapt to their common demand for higher levels ofenergy demand.

IDH2 encodes mitochondrial NADP(+)-dependent isocitratedehydrogenase, an enzyme that catalyzes the oxidative decarbox-ylation of isocitrate to 2-oxoglutarate. This enzymatic reaction isa part of the citric acid cycle – also known as the tricarboxylic acidcycle (TCA cycle). In all aerobic organisms, the TCA cycle generatesenergy through the oxidization of acetate derived from carbohy-drates, fats and proteins into carbon dioxide and water. Mutationsin TCA genes cause metabolic disorders. For example, the associa-tion of mutations in the IDHs (IDH1, IDH2) and metabolic disordershas been widely studied [15–18].

Since cetaceans, primates and bats need to elevate their maxi-mum energy metabolism ratio, and IDH2 has a vital role in aerobicrespiration, here we study the molecular evolution of IDH2 in thesemammalian groups.

2. Materials and methods

2.1. Source of data and primary treatments

We sequenced complete IDH2 gene from 19 bats (Aselliscus stol-iczkanus, Chaerephon plicata, Cynopterus sphinx, Eonycteris spelaea,

Fig. 1. IDH2 gene tree. A Bayesian tree for IDH2 was generated with MrBayes.Numbers at the nodes indicates posterior probabilities.

W.-M. Ai et al. / FEBS Letters 588 (2014) 450–454 451

Hipposideros armiger, Hipposideros larvatus, Hipposideros pomona, Iaio, Miniopterus fuliginosus, Myotis laniger, Myotis ricketti, Nyctalusplancyi, Pipistrellus abramus, Rhinolophus blythi, Rhinolophusmacrotis, Rhinolophus paradoxolophus, Rhinolophus rex, Rhinolophussinicus, Rousettus leschenaulti). In additions, complete IDH2sequences from the dolphin and other mammalian species werecollected from the Ensembl database.

Due to lack of fresh samples from cetaceans, we only sequencedparts of the IDH2 gene from two cetaceans (Neophocaena phocaeno-ides, Balaenoptera musculus). IDH2 genes from ten other mamma-lian species (Ailurus fulgens, Bos grunniens (three samples), Canislupus familiaris, Cervus albirostris, Equus kiang, Felis lynx, Felis silves-tris bieti, Ochotona curzoniae, Pseudois nayaur, Rhinopithecus roxell-ana) were also sequenced. The partial IDH2 sequences were usedto detect whether the parallel amino acid site were conserved.

For bats, RNA was isolated from fresh muscle tissues with theRNAiso™ Plus Kit (Takara, China). RT-PCR was performed usingthe PrimeScript™ RT-PCR Kit (Takara, China). For the other sam-ples, due to the lack of fresh tissue, we used frozen tissue to extractDNA by standard phenol/chloroform methods. PCR reactions wereperformed in a 50 ll volume containing 5 ll of 10 � reaction buf-fer, 0.2 mM dNTPs, 0.2 lM each primer, 1.5U Taq DNA polymerase(TaKaRa Biosystems) and 10 ng total DNA. PCR products were puri-fied on spin columns (Watson Biotechnologies Inc., Shanghai) andwere directly sequenced for both strands by using BigDye™ Termi-nator Cycle Sequence Kit (ABI Applied Biosystems 3730). Primerswere shown in Table S1. All the sequences were deported to Gen-Bank (accession numbers: HQ161111–HQ161159).

2.2. Phylogenetic analyses

Nucleotide sequences of IDH2 were aligned using CLUSTALX1.81 [19], with manual adjustment. Best fit models for nucleotidesubstitutions were determined by ModelTest [20] under the Akaikeinformation criterion. The computer algorithm PhyML [21] wasused to construct maximum-likelihood (ML) phylogenies of thenucleotide and amino-acid data under their best-fitting models.Bayesian inference (BI) phylogeny were constructed using MrBayes[22]. For the Bayesian analysis two separate runs were performedwith four Markov chains. Each run was conducted with 3 � 106

generations and sampled every 100 generations. A consensus treewas calculated after omitting the first 25% trees as burn-in.

2.3. Molecular evolutionary analyses

Tests for selection were carried out using the Codeml programimplemented in PAML [23,24]. Four models were used: (1) one-ratiomodel; (2) free-ratio model; (3) two-ratio model and (4) branch-sitemodel. LRT statistics were calculated between the following modelpairs: (1) the two-ratio model versus the one-ratio model were com-pared to test whether the x ratio is significantly different from thatof other mammals; (2) test 1 (branch-site model versus site modelM1a) and test 2 (branch-site model versus branch-site model withfixed x1 = 1) for branch-site model [25] were conducted. Significantdifferences were identified by calculating twice the difference in log-likelihood values (2DlnL) between the two models, which follow achi-squared (v2) distribution with degrees of freedom equalingthe difference in the number of parameters estimated by the pairof models.

2.4. Convergent evolution analyses

The sequences of the internal nodes were reconstructed usingBayesian methods [26]. Convergent and parallel amino acid substi-tutions along bats and dolphins were detected. The statistical

significance of these amino acid changes was tested with the meth-od developed by Zhang and Kumar [27].

3. Results and discussion

We built ML and BI trees based on the nucleotide sequences of19 bats, one cetacean, dog, cow, panda, horse, mouse, rat, human,macaque, gorilla and chimpanzee. The length of aligned nucleotidesequences for IDH2 was 1356 base pairs (bp). The gene trees forIDH2 (Fig. 1) were basically the same as the well-accepted speciestree [28–30] for all methods.

Bats, cetaceans and primates represent three groups of mam-mals that independently adapted to higher energy demands. Con-vergent and/or parallel sequence evolution may be responsible forthe parallel and convergent functional changes to fit for this adap-tation [31–37]. When a similar-appearing trait evolves indepen-dently in two or more lineages from different ancestral states,then it is defined as convergent evolution; while if similar traitsevolve independently in two lineages from the same ancestralstate, then it is parallelism [38]. Since IDH2 plays vital roles in en-ergy production, we tested whether convergent/parallel evolutioncould be found in this gene among these lineages. Branches b, dand p lead to the common ancestor (CA) bat, dolphin and primatesrespectively (Fig. 1). We reconstructed the sequences at the ances-tral nodes and mapped amino acid changes along these branches.We found branches b and d shared one amino acid change(V186A). Parallel evolution on these two branches differed statisti-cally from the random expectation (P < 0.05). Interestingly, wefound that the parallel amino acid change (V186A) was alsoevolved in parallel on the primate lineage (Fig. 2A).

IDH2 encodes a key enzyme in the tricarboxylic acid cycle [39]and was found to have a parallel amino acid replacement at position186 of valine (V) to alanine (A) on three mammalian lineages. Dol-phins, in addition to having large brains, adapted to the aquatic life-style through ancestors that had a long history of a semi-aquaticlifestyle, a lifestyle which incurs energy costs that are higher thanrunning [40]. Thus the energetics of swimming, in addition to brainsize, must have played a major role in the evolution of this group[41]. The dolphins, thus, have multiple factors that suggest that theyrequire elevated levels of energy production. Diving vertebrates are

human chimpanzee gorilla orangutan macaque marmoset mouse lemur tree shrew mouse rat kangaroo rat guinea pig squirrel rabbit pika cetaceans cow pig horse catdogbats

shrew hedgehog lesser hedgehog tenrec hyrax elephant opposum wallaby platypus turkey zebra finch anole lizard

Xenopus tropicalis fugu tetraodon stickleback medaka

mammalia

avianreptiliaamphibian

fish

A BV186A

V186A

V186A

Fig. 2. Parallel evolution of IDH2 in cetaceans, primates and bats. (A) The topology of IDH2 is marked with the V186A amino acid change. (B) Sequences of IDH2. Branchesmarked in red are the three branches that exhibit parallel evolution.

452 W.-M. Ai et al. / FEBS Letters 588 (2014) 450–454

exposed to progressive asphyxia and hypoxia during their dives.Cetaceans have developed a number of mechanisms that allow themto tolerate prolonged apnea periods and increase their diving capac-ity, including elevated hemoglobin, myoglobin concentrations, andhigh mitochondrial content [42]. Previous studies have suggestedthat genes that associated with energy metabolism have undergoneadaptive evolution in the dolphin to adapt for their higher energy de-mand [12,13]. The maximum metabolic rates in bats is high [6–8].Therefore, during the development of flight in bats, they would haveneeded to elevate their maximum metabolic rate. Larger brains

Fig. 3. Protein structure of IDH2. Three-dimensional representa

needs greater levels of aerobic metabolism [4,5] and larger brainsevolved on the dolphin and anthropoid ancestry of humans. It hasbeen found that an increased proportion of energy metabolismgenes have undergone adaptive evolution on the human lineage[10,43]. Therefore, our finding that IDH2 underwent parallel evolu-tion in the dolphins, bats and primates are not unexpected. Furtheranalyses of IDH2 gene sequences from diverse vertebrate species(data includes sequences from 31 mammals, two birds, one amphib-ian, one reptile and four fishes) showed that this amino acid site isconservative (Fig. 2B). This observation suggests that the parallel

tion of relevant amino acid residues in mammalian IDH2.

W.-M. Ai et al. / FEBS Letters 588 (2014) 450–454 453

changes at this amino acid site is not due to random mutation, andmay have functional importance. Parallel evolution of this genemay reflect their parallel evolution of increased demands for energy.

The site of the parallel amino acid change (186) is located in a bsheet of the IDH2 protein structure (Fig. 3, pdb code: 4ja8 [44]).Residues 140 and 172, where substitutions cause metabolic disor-ders [18], are also mapped in Fig. 3. The graphic representations ofthe 3D structures were created with the program VMD [45]. Thechange of the residue at position 186 should cause additional sterichindrance due to the increased size of the valine side chain relativeto alanine, and should alter the structure of the protein. Furtherfunctional experiments are needed to prove this suggestion.

In addition to the analysis of convergent evolution, we alsotested for adaptive selection acting on this gene. Selective pressurewas evaluated using the PAML package and the test results are pre-sented in Table S2. The one-ratio model obtained an average x (Ka/Ks ratio) of 0.0180. For specific branches, we alternatively set theCA bat (branch b) or dolphin (branch d in Fig. 1) as the foregroundbranch. In both conditions, although the ratios of x for the fore-ground branches were greater than background branches, theywere less than 1 (xb = 0.1756 while x0 = 0.0169; xd = 0.0497while x0 = 0.0172). For the branch-site models, both test 1 (ModelA versus Model 1a) and test 2 (Model A versus null model) wereused to control the false positive signals. However, we failed to de-tect any signals for positive selection. Other studies that detectedsignals for convergent/parallel evolution also failed to detect anysignal for positive selection [33,34,36,46].

As it is relatively easy to amplify and sequence mtDNA, it iswidely used to study the role of mitochondria in adaptation in ani-mals [47–49]. However, the respiratory chain has a dual (mitochon-drial and nuclear) genetic foundation. In this study, we cloned theIDH2 gene – a key nuclear gene in the citric acid cycle, and demon-strated that during the emergence of bats, cetaceans and primatesfrom their ancestors, it has undergone convergent evolution, likelyreflecting their shared demand for elevated energy.

Data release

GenBank: Accession numbers of sequences HQ161111–HQ161159.

Acknowledgments

This work was supported by Grants from the National NaturalScience Foundation of China (No. 31172080). We thank ProfessorDavid Irwin for editing the paper.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.febslet.2013.12.005.

References

[1] Harrison, G.A., Tanner, J.M., Pilbeam, D.R. and Baker, P.T. (1988) HumanBiology: An Introduction to Human Evolution, Variation, Growth, andAdaptability, Oxford University Press, Oxford.

[2] Tobias, P.V. (1971) Human skeletal remains from the Cave of Hearths,Makapansgat, northern Transvaal. Am. J. Phys. Anthropol. 34, 335–367.

[3] Attwell, D. and Laughlin, S.B. (2001) An energy budget for signaling in the greymatter of the brain. J. Cereb. Blood Flow Metab. 21, 1133–1145.

[4] Doan, J.W., Schmidt, T.R., Wildman, D.E., Uddin, M., Goldberg, A., Hüttemann,M., Goodman, M., Weiss, M.L. and Grossman, L.I. (2004) Coadaptive evolutionin cytochrome c oxidase: 9 of 13 subunits show accelerated rates ofnonsynonymous substitution in anthropoid primates. Mol. Phylogenet. Evol.33, 944–950.

[5] Schmidt, T.R., Wildman, D.E., Uddin, M., Opazo, J.C., Goodman, M. andGrossman, L.I. (2005) Rapid electrostatic evolution at the binding site for

cytochrome c on cytochrome c oxidase in anthropoid primates. Proc. Natl.Acad. Sci. USA 102, 6379–6384.

[6] Pasquis, P., Lacaisse, A. and Dejours, P. (1970) Maximal oxygen uptake in fourspecies of small mammals. Respir. Physiol. 9, 298–309.

[7] Thomas, S.P. and Suthers, R.A. (1972) The physiology and energetics of batflight. J. Exp. Biol. 57, 317–335.

[8] Maina, J.N. (2000) What it takes to fly: the structural and functionalrespiratory refinements in birds and bats. J. Exp. Biol. 203, 3045–3064.

[9] Goodman, M. and Sterner, K.N. (2010) Phylogenomic evidence of adaptiveevolution in the ancestry of humans. Proc. Natl. Acad. Sci. USA 107, 8918–8923.

[10] Goodman, M., Sterner, K.N., Islam, M., Uddin, M., Sherwood, C.C., Hof, P.R., Hou,Z.-C., Lipovich, L., Jia, H., Grossman, L.I. and Wildman, D.E. (2009)Phylogenomic analyses reveal convergent patterns of adaptive evolution inelephant and human ancestries. Proc. Natl. Acad. Sci. USA 106, 20824–20829.

[11] Goldberg, A., Wildman, D.E., Schmidt, T.R., Huttemann, M., Goodman, M.,Weiss, M.L. and Grossman, L.I. (2003) Adaptive evolution of cytochrome coxidase subunit VIII in anthropoid primates. Proc. Natl. Acad. Sci. USA 100,5873–5878.

[12] Shen, Y.-Y., Zhou, W.-P., Zhou, T.-C., Zeng, Y.-N., Li, G.-M., Irwin, D.M. andZhang, Y.-P. (2012) Genome-wide scan for bats and dolphin to detect theirgenetic basis for new locomotive styles. PLoS One 7, e46455.

[13] Sun, Y.-B., Zhou, W.-P., Liu, H.-Q., Irwin, D.M., Shen, Y.-Y. and Zhang, Y.-P.(2013) Genome-wide scans for candidate genes involved to the aquaticadaptation of dolphins. Genome Biol. Evol. 5, 130–139.

[14] Shen, Y.Y., Liang, L., Zhu, Z.H., Zhou, W.P., Irwin, D.M. and Zhang, Y.P. (2010)Adaptive evolution of energy metabolism genes and the origin of flight in bats.Proc. Natl. Acad. Sci. USA 107, 8666–8671.

[15] Reitman, Z.J. and Yan, H. (2010) Isocitrate dehydrogenase 1 and 2 mutations incancer: alterations at a crossroads of cellular metabolism. J. Natl. Cancer Inst.102, 932–941.

[16] Yan, H., Parsons, D.W., Jin, G., McLendon, R., Rasheed, B.A., Yuan, W., Kos, I.,Batinic-Haberle, I., Jones, S. and Riggins, G.J. (2009) IDH1 and IDH2 mutationsin gliomas. N. Engl. J. Med. 360, 765–773.

[17] Borodovsky, A., Seltzer, M.J. and Riggins, G.J. (2012) Altered cancer cellmetabolism in gliomas with mutant IDH1 or IDH2. Curr. Opin. Oncol. 24, 83–89.

[18] Yen, K.E., Bittinger, M.A., Su, S.M. and Fantin, V.R. (2010) Cancer-associatedIDH mutations: biomarker and therapeutic opportunities. Oncogene 29, 6409–6417.

[19] Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F. and Higgins, D.G.(1997) The CLUSTAL_X windows interface: flexible strategies for multiplesequence alignment aided by quality analysis tools. Nucleic Acids Res. 25,4876–4882.

[20] Posada, D. (1998) MODELTEST: testing the model of DNA substitution.Bioinformatics 14, 817–818.

[21] Guindon, S. and Gascuel, O. (2003) A simple, fast, and accurate algorithm toestimate large phylogenies by maximum likelihood. Syst. Biol. 52, 696–704.

[22] Huelsenbeck, J.P. and Ronquist, F. (2001) MRBAYES: Bayesian Inference ofPhylogenetic Trees, Oxford University Press, pp. 754–755, Oxford UniversityPress.

[23] Yang, Z. (1997) PAML: a program package for phylogenetic analysis bymaximum likelihood. Comput. Appl. Biosci. 13, 555–556.

[24] Yang, Z. (2007) PAML 4: phylogenetic analysis by maximum likelihood. Mol.Biol. Evol. 24, 1586–1591.

[25] Zhang, J., Nielsen, R. and Yang, Z. (2005) Evaluation of an improved branch-sitelikelihood method for detecting positive selection at the molecular level. Mol.Biol. Evol. 22, 2472–2479.

[26] Zhang, J. and Nei, M. (1997) Accuracies of ancestral amino acid sequencesinferred by the parsimony, likelihood, and distance methods. J. Mol. Evol. 44,S139–S146.

[27] Zhang, J. and Kumar, S. (1997) Detection of convergent and parallel evolutionat the amino acid sequence level. Mol. Biol. Evol. 14, 527–536.

[28] Teeling, E.C., Springer, M.S., Madsen, O., Bates, P., O’Brien, S.J. and Murphy, W.J.(2005) A molecular phylogeny for bats illuminates biogeography and the fossilrecord. Science 307, 580–584.

[29] Murphy, W.J., Eizirik, E., O’Brien, S.J., Madsen, O., Scally, M., Douady, C.J.,Teeling, E., Ryder, O.A., Stanhope, M.J., de Jong, W.W. and Springer, M.S. (2001)Resolution of the early placental mammal radiation using Bayesianphylogenetics. Science 294, 2348–2351.

[30] Murphy, W.J., Eizirik, E., Johnson, W.E., Zhang, Y.P., Ryder, O.A. and O’Brien, S.J.(2001) Molecular phylogenetics and the origins of placental mammals. Nature409, 614–618.

[31] Zhang, J. (2006) Parallel adaptive origins of digestive RNases in Asian andAfrican leaf monkeys. Nat. Genet. 38, 819–823.

[32] Shen, Y.-Y., Liang, L., Li, G.-S., Murphy, R.W. and Zhang, Y.-P. (2012) Parallelevolution of auditory genes for echolocation in bats and toothed whales. PLoSGenet. 8, e1002788.

[33] Shen, Y.Y., Liu, J., Irwin, D.M. and Zhang, Y.P. (2010) Parallel and convergentevolution of the dim-light vision gene RH1 in bats (order: Chiroptera). PLoSOne 5, 259–282.

[34] Shen, Y.Y., Lim, B.K., Liu, H.Q., Liu, J., Irwin, D.M. and Zhang, Y.P. (2012)Multiple episodes of convergence in genes of the dim light vision pathway inbats. PLoS One 7, e34564.

[35] Li, Y., Liu, Z., Shi, P. and Zhang, J. (2010) The hearing gene Prestin unitesecholocating bats and whales. Curr. Biol. 20, R55–R56.

454 W.-M. Ai et al. / FEBS Letters 588 (2014) 450–454

[36] Li, G., Wang, J., Rossiter, S.J., Jones, G., Cotton, J.A. and Zhang, S. (2008) Thehearing gene Prestin reunites echolocating bats. Proc. Natl. Acad. Sci. USA 105,13959–13964.

[37] Liu, Y., Cotton, J.A., Shen, B., Han, X., Rossiter, S.J. and Zhang, S. (2010)Convergent sequence evolution between echolocating bats and dolphins. Curr.Biol. 20, R53–R54.

[38] Zakon, H.H. (2000) Convergent evolution on the molecular level. Brain Behav.Evol. 59, 250–261.

[39] Bruns, G.A.P., Eisenman, R.E. and Gerald, P.S. (1976) Human mitochondrialNADP-dependent isocitrate dehydrogenase in man-mouse somatic cellhybrids. Cytogenet. Genome Res. 17, 200–211.

[40] Williams, T.M. (1999) The evolution of cost efficient swimming in marinemammals: limits to energetic optimization. Philos. Trans. R. Soc. Lond. B 354,193–201.

[41] Hui, C.A. (1987) Power and speed of swimming dolphins. J. Mamm. 68, 126–132.

[42] Zenteno-Savın, T., Vázquez-Medina, J.P., Cantu-Medellın, N., Ponganis, P.J.and Elsner, R. (2012) Oxidative Stress in Aquatic Ecosystems, Wiley-Blackwell.

[43] McGowen, M.R., Grossman, L.I. and Wildman, D.E. (2012) Dolphin genomeprovides evidence for adaptive evolution of nervous system genes and amolecular rate slowdown. Proc. R. Soc. B 279, 3643–3651.

[44] Wang, F., Travins, J., DeLaBarre, B., Penard-Lacronique, V., Schalm, S., Hansen,E., Straley, K., Kernytsky, A., Liu, W., Gliser, C., Yang, H., Gross, S., Artin, E.,Saada, V., Mylonas, E., Quivoron, C., Popovici-Muller, J., Saunders, J.O., Salituro,F.G., Yan, S., Murray, S., Wei, W., Gao, Y., Dang, L., Dorsch, M., Agresta, S.,Schenkein, D.P., Biller, S.A., Su, S.M., de Botton, S. and Yen, K.E. (2013) Targetedinhibition of mutant IDH2 in leukemia cells induces cellular differentiation.Science 340, 622–626.

[45] Humphrey, W., Dalke, A. and Schulten, K. (1996) VMD: visual moleculardynamics. J. Mol. Graph. 14, 33–38.

[46] Sugawara, T., Terai, Y., Imai, H., Turner, G.F., Koblmüller, S., Sturmbauer, C.,Shichida, Y. and Okada, N. (2005) Parallelism of amino acid changes at the RH1affecting spectral sensitivity among deep-water cichlids from LakesTanganyika and Malawi. Proc. Natl. Acad. Sci. USA 102, 5448–5453.

[47] da Fonseca, R.R., Johnson, W.E., O’Brien, S.J., Ramos, M.J. and Antunes, A. (2008)The adaptive evolution of the mammalian mitochondrial genome. BMCGenomics 9, 119–140.

[48] Shen, Y.Y., Shi, P., Sun, Y.B. and Zhang, Y.P. (2009) Relaxation of selectiveconstraint on avian mitochondrial DNA following the degeneration of flightability. Genome Res. 19, 1760–1765.

[49] Sun, Y.-B., Shen, Y.-Y., Irwin, D.M. and Zhang, Y.-P. (2011) Evaluating the rolesof energetic functional constraints on teleost mitochondrial-encoded proteinevolution. Mol. Biol. Evol. 28, 39–44.