Molecular Ecology (2014) 23, 5393–5397

NEWS AND VIEWS

OPINION

Mitochondrial genomes of domestic animals need scrutiny NI-NI SHI,*†1 LONG FAN,‡1 YONG-GANG Y A O , † § M I N - S H E N G P E N G * † and YA-PING ZHANG*† ¶ *State Key Laboratory of Genetic Resources and Evolution, and Yunnan Laboratory of Molecular Biology of Domestic Animals, Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming, Yunnan 650223, China; †Kunming College of Life Science, University of Chinese Academy of Sciences, Kunming, Yunnan 650204, China; ‡School of Life Sciences, The Chinese University of Hong Kong, Shatin, Hong Kong, SAR 999077, China; §Key Laboratory of Animal Models and Human Disease Mechanisms, Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming, Yunnan 650223, China; ¶Laboratory for Conservation and Utilization of Bio-Resources & Key Laboratory for Microbial Resources of the Ministry of Education, Yunnan University, Kunming, Yunnan 650091, China

More than 1000 complete or near-complete mitochondrial DNA (mtDNA) sequences have been deposited in GenBank for eight common domestic animals (cattle, dog, goat, horse, pig, sheep, yak and chicken) and their close wild ancestors or relatives, as well. Nevertheless, few efforts have been performed to evaluate the sequence data quality. Herein, we conducted a phylogenetic survey of these complete or near-complete mtDNA sequences based on mtDNA haplogroup trees for the eight animals. We show that errors due to artificial recombination, surplus of mutations and phantom mutations do exist in 14.5% (194/1342) of mtDNA sequences and all of them should be treated with wide caution. We propose some caveats for future mtDNA studies of domestic animals. Keywords: data quality, mtDNA, phylogeny

domestic

animal,

haplogroup,

Received 16 July 2014; revised 22 September 2014; accepted 03 October 2014 Mitochondrial DNA (mtDNA) is one of the most widely used markers in exploring genetic diversity and tracing evolutionary history for human, domestic and wild animals. Since 2006, we witnessed the change from partial mtDNA sequence (e.g. control region) to complete Correspondence: Min-Sheng Peng, Fax: +86-871-6519-5430; E-mail: [email protected] or Ya-Ping Zhang, Fax: +86 871-6519 5430; E-mail: [email protected] 1 These authors contributed equally to this work.

© 2014 John Wiley & Sons Ltd

mitochondrial genome in abundant studies of various domestic animals (Wang et al. 2014). Hitherto more than 1000 complete or near-complete mtDNA sequences have been deposited in GenBank for eight common domestic animals (cattle, dog, goat, horse, pig, sheep, yak and chicken) and their close wild ancestors or relatives (Table 1; Table S1, Supporting information). However, most researchers took those reported mtDNA genomes in their data mining and comparative analyses without any scrutiny for data quality. Unfortunately, it is known that some sequences have been shown to contain sequencing errors (Achilli et al. 2012; Miao et al. 2013) and nuclear mtDNA (NUMT) contaminations (Hassanin et al. 2010), similar to problems found in human mtDNA studies (Yao et al. 2008, 2009). Most released mtDNA genome sequences of domestic animals were generated through several PCRs and Sanger sequencing reactions (e.g. Pang et al. 2009; Yu et al. 2013). These practices are labour-intensive and prone to errors, as has been well demonstrated in human mtDNA data generated through similar experimental protocols (Bandelt et al. 2006). It triggers us to ask: do similar errors also occur in the mitochondrial genome data of domestic animals? Herein, we summarized three major kinds of errors (Table 2) and screened these errors in 1342 complete or near-complete mtDNA sequences from eight domestic animals and their close wild ancestors or relatives (Table 1; Table S1, Supporting information). We adopted the phylogenetic strategy developed in human mtDNA analyses (Yao et al. 2009) based on highresolution genealogy (i.e. mtDNA haplogroup tree). In brief, mtDNA haplogroup tree can be constructed with the parsimony-like methods such as networks (Bandelt et al. 1999). The haplogroups can be defined by following a specific diagnostic mutational motif. Specifically, the members that belong to a certain haplogroup can be characterized by a string of mutations that define the internal branch. This branch directs to the haplogroup (internal node) in the tree (van Oven & Kayser 2009). When comparing with the defined reference sequence (Bandelt et al. 2014), variants in each sequence can be scored. According to the haplogroup tree, the variants can be mapped on the internal branch as diagnostic mutations or on the terminal branch (tip). As a result, sequences with anomalous variants showing conflicts with their known phylogenetic status can be identified and shall be treated with caution. This phylogenetic method has been proved to be powerful and sensitive for human mtDNA data quality assessment (Bandelt et al. 2005, 2006; Salas et al. 2005a; Kong et al. 2008; Yao et al. 2009). Since 2007, the mtDNA haplogroup trees for pig (Wu et al. 2007), cattle (Achilli et al. 2008, 2009; Bonfiglio et al. 2010, 2012), horse (Achilli et al. 2012), chicken (Miao et al. 2013) and sheep (Lancioni et al. 2013) have been constructed. In

5394 N E W S A N D V I E W S : O P I N I O N Table 1 Summary of (near-)complete mtDNA genome sequences analysed in this work Animals

Related species

Released by GenBank*

Reference sequence

Haplogroup nomenclature

Cattle and Aurochs

Bos taurus B. indicus B. primigenius B. grunniens Canis lupus Equus caballus E. przewalskii Gallus gallus Sus scrofa Ovis aries Capra hircus

266 10 2 79 447 254 9 66 127 47 35

V00654

I, P, Q, R and T

GQ464259† EU789787† JN398377

A–D A–F A–R

AP003321 EF545567 AF010406 GU068049

A – I and X – Z A, D and E A–E A–C

Yak and wild yak Dog and grey wolf Horse and Przewalski’s horse Chicken and Red Junglefowl Pig and wild boar Sheep Goat

*Access time: 1 June 2014. † Reference sequences are defined in this work. Table 2 Three kinds of common errors occurring in mtDNA data analysed in this study Errors

Common phenotypes

Major causes

Artificial recombination

Missing diagnostic mutations; mis-added diagnostic mutations of different haplogroups Excessive unusual mutations, especially transversions, indels and heteroplasmic mutations Mutations are laboratory specific and occur repeatedly on different haplogroup backgrounds

Sample mix-up; contamination

Surplus of mutations Phantom mutations

Sequencing errors; NUMTs contamination Low quality of sequencing; technical pitfalls of NGS

Table 3 mtDNA genome sequences with potential errors in eight domestic animals

Animals

Released by GenBank

Cattle Chicken Dog Goat Horse Pig Sheep Yak

278 66 447 35 263 127 47 79

Artificial recombination 1 11 20 7 7

Surplus of mutations

Phantom mutations

Percentage of potential errors, %

1 1 5 4 2 19 2 1

7 + 34*

15.46 1.51 3.57 11.42 35.74 32.28 4.25 10.12

1 + 71* 15

*Detected in next-generation sequencing data (Table S6, Supporting information).

terms of the available trees, we followed the proposed caveats (Bandelt et al. 2005, 2006; Kong et al. 2008; Yao et al. 2008, 2009) to investigate the data quality. The previously defined reference sequences (Table 1) were adopted as the consistent standards to avoid confusion and misunderstanding (Salas et al. 2012; Bandelt et al. 2014). We discovered that potential errors occurred in cattle, horse, pig, sheep and chicken (Table 3; Table S2, Supporting information). For instance, in five pig mitochondrial genomes published recently (Yu et al. 2013), two sequences (GenBank Accession nos. KC250273 and KC469586) were problematic (Fig. 1a). The mis-added variants C9553T-C9605T in KC250273 (which belongs to haplogroup D1e) are the

diagnostic motif of haplogroup E1a. Similarly, motifs T2374C (i.e. @2374)-T2613C, A5672G-C5678T-T5708CT5753C, C9156T-C9157T-C9225T-T9234C and T14812CC14839T in KC469586 (which belongs to haplogroup D1a1a) characterize haplogroup E. The missed and misadded mutations most likely represent as ‘recombinants’ of separate segments from different samples of haplogroups E1a and E that are common in European pigs. The errors of artificial recombination are probably due to sample mix-up or contamination with European pig breed(s). We followed the same approach to check mitochondrial genome data from other domestic animals. We first reconstructed mtDNA haplogroup trees for dog, goat and © 2014 John Wiley & Sons Ltd

N E W S A N D V I E W S : O P I N I O N 5395 (a)

1254+A 2369d @2374 2613 5672 5678

4842 9553 9605 9917T

D1e

5708 5753 8017 9156 9157 9225

2065C 3594 14710

9234 10616 14812 14839 @15220 16111

D1a1a

300 599 8135 11014A

Reference EF545567

278 451 14198 15220

EU789672 @2227 @2232 EU789745 KM113774 @2837 3279 @3296 14895 12587 15892 16140

FJ237000

D1

E1a

240 240 240 11432 500 500 500 12064 2374 2374 2374 16301

9605 @9837 13478

D2

A6a

4471 8581 15337 15838

5948 11944 15025 15721 16292

European Pig 109 131 137C 142+nA 144 152 157 293 574 1168 2064 2071 2613 3102 3366 3451 3640 3873T 4030 4081 4420 4438 4489A 4737 4754 5168T 5207 5552 5678 5880 5963

A1a 180 4459 305 5628 451 5672 500 5708 1412 7339 2339 7669 4015 7837

8017 12649 9435 13049 10070 14213 10100 15685 10680 16111 11184 16258 11259

A1 3064 4939 5463A 5636 5753

Asian Pig 322 389 1175 1225 1304 1639 1990 2335 3088 3999 4342

4369 4711 4846 5369 5873 6092 7009 7108 7447 8498

A3b1 16172

A3b

E1

D 240 1315 5797 6164 6952

6296 6970 7513 9156 14639

A 8773 13581 9157 13997 9789 14399 9837 14738 10529 14839 10816 14998 10944 15151 11071 15277 11366 @15337 12109 16492

Root

E 6138 11830 6171 12241 6298 12298 6508 12370 6925T 12469 7022 12518 7321 12583 7486 12675 7750 12958 8371 12962 8413 13433A 8605 13605 8713 13838 8743 14209 8761 14313A 9058 14561C 9225 14812 9234 14947 9553 15109 9752 15127 9973 15362 10484 15625 10517 15661 10753 15961 11018 16294 11162 16297 11189 16358 11289 16456 11372 16552 11683 16564 11786 16608

8762 15533 16086

8318 11877 16005

A3 2227 7115 2837 7517 3296 7719 5427 8109 5721T 9136 5898 15381 6884G 15610

A6

767 3279 5031 5565 7361 8063 12119 12962 13566 13765 15007 15607 15645

A2’3 10357 14386G

A1’2’3 2232

A1’2’3’4’5 @15607

A 3470 11966 3599 12816 4518 13264 5010 13711 6471 15438 8426 15983 10407

yak (Tables S3–S5, Supporting information) based on the parsimony-like method; bibliographically, this method is usually applied in domestic animals (e.g. Wu et al. 2007; Achilli et al. 2008; Miao et al. 2013) as well as in human (van Oven & Kayser 2009) mtDNA studies. The hierarchical haplogroup nomenclature systems were updated from previous studies (Pang et al. 2009; Wang et al. 2010; Doro et al. 2014), referring to the rules in human mtDNA phylogeny (van Oven & Kayser 2009). Not unexpectedly, several potential errors are detected in the mtDNA sequences of these animals (Table 3; Table S2, Supporting information). For instance, a dog sequence (EU789672) from our previous study (Pang et al. 2009) was suspected to be corroded by artificial recombination. The diagnostic variants G2232A and T2227C-T2837C-G3296A, which define haplogroups A1’2’3 and A3, respectively, are missing (Fig. 1b). It was probably due to sample mix-up after replacing the © 2014 John Wiley & Sons Ltd

Fig. 1 The flawed mitochondrial genome sequences in pigs (a) and dogs (b). The nucleotide positions in the sequences from pigs and dogs were scored relative to the reference sequences EF545567 and EU789787, respectively. Transitions are shown on the branches, and transversions are further annotated by adding suffixes. Deletions and insertions are indicated by ‘d’ and ‘+’, respectively. Mutations towards a base identical-bystate to the reference sequences are indicated with the prefix @. Suspected mutations due to errors are in red, and their corresponding phylogenetic status in haplogroup trees are underlined.

(b)

KC469586

KC250273

Reference EU789787

Z 291 2840A 2981 3962 4119 7655 7927 8091 8510 11164 12387 12416A 13469 13588 13809 13951 15115 15170 15607

Root

fragment around nucleotide position 2200–3300 of EU789672 with another sample of nonhaplogroup A1’2’3 (most likely haplogroup A6, Fig. 1b). We re-amplified this mtDNA and sequenced the entire mitochondrial genome. Indeed, the suspected artificial recombination is confirmed (Fig. 1b). The corrected sequence has been updated in GenBank as KM113774. In addition to artificial recombination, we show that phantom mutation and poor sequencing quality do exist in mtDNA data of domestic animals (Table 3; Table S2, Supporting information). When using the flawed sequences, it is prone to getting erroneous claims. In some cases, it is expected that the surplus of mutations can ‘increase’ length of branch in the phylogeny and ‘accelerate’ molecular evolution (e.g. detection of positive selection and violation of molecular clock). Indeed, there are several cases reported in human mtDNA studies (e.g. Salas et al. 2005b; Liu et al.

5396 N E W S A N D V I E W S : O P I N I O N 2012). Moreover, phantom mutations basically are detected due to technological pitfalls in next-generation sequencing platforms (especially for homopolymers) (Table S6, Supporting information). Similarly, data qualities of some human mtDNA sequences generated via next-generation sequencing platforms were less than satisfactory (Bandelt & Salas 2012). Along with the progress of sequencing techniques, accumulation of mitochondrial genome sequences from domestic animals is accelerating. The common practice of generating and analysing mtDNA data should be carried out with the necessary caution. Analyses based on traditional phylogenetic software, at least in the related literatures (e.g. Pang et al. 2009; Yu et al. 2013), are inefficient to discern those errors. Thus, we suggest researchers to follow some caveats from human mtDNA studies during experiments and data analyses (Bandelt et al. 2005, 2006; Salas et al. 2005a; Kong et al. 2008; Yao et al. 2008, 2009). Phylogenetic analyses based on mtDNA haplogroup trees are recommended, although it may be difficult for a nonexpert or a novice in mtDNA analysis. Developing bioinformatic tools to make the related analyses convenient and efficient should be encouraged in the future. Furthermore, we suggest that these flawed sequences identified in this study (Table S2, Supporting information) should be never used in future data-mining analyses, unless a correct one was updated. We welcome the researchers of the original study to take our warnings into consideration and double check these sequences tagged with ‘flawed’, just as what we did for the dog sequence EU789672.

Acknowledgements We thank Adeniyi C. Adeola for technical assistance. This work was supported by the 973 programme (2013CB835200, 2013CB835204). The Youth Innovation Promotion Association, Chinese Academy of Sciences (CAS), provided support to M-S.P. M-S.P. is a member of Youth Innovation Promotion Association, CAS.

References Achilli A, Olivieri A, Pellecchia M et al. (2008) Mitochondrial genomes of extinct aurochs survive in domestic cattle. Current Biology, 18, R157–R158. Achilli A, Bonfiglio S, Olivieri A et al. (2009) The multifaceted origin of taurine cattle reflected by the mitochondrial genome. PLoS ONE, 4, e5753. Achilli A, Olivieri A, Soares P et al. (2012) Mitochondrial genomes from modern horses reveal the major haplogroups that underwent domestication. Proceedings of the National Academy of Sciences of the United States of America, 109, 2449–2454. Bandelt HJ, Salas A (2012) Current next generation sequencing technology may not meet forensic standards. Forensic Science International Genetics, 6, 143–145. Bandelt HJ, Forster P, R€ ohl A (1999) Median-joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution, 16, 37–48. Bandelt HJ, Kong QP, Parson W, Salas A (2005) More evidence for non-maternal inheritance of mitochondrial DNA? Journal of Medical Genetics, 42, 957–960.

Bandelt HJ, Kivisild T, Parik J et al. (2006) Lab-specific mutation processes. In: Human Mitochondrial DNA and the Evolution of Homo Sapiens (eds Bandelt H, Macaulay V, Richards M), pp. 117– 146. Springer, Berlin, Heidelberg. Bandelt HJ, Kloss-Brandst€ atter A, Richards MB, Yao YG, Logan I (2014) The case for the continuing use of the revised Cambridge Reference Sequence (rCRS) and the standardization of notation in human mitochondrial DNA studies. Journal of Human Genetics, 59, 66–77. Bonfiglio S, Achilli A, Olivieri A et al. (2010) The enigmatic origin of bovine mtDNA haplogroup R: sporadic interbreeding or an independent event of Bos primigenius domestication in Italy? PLoS ONE, 5, e15760. Bonfiglio S, Ginja C, De Gaetano A et al. (2012) Origin and spread of Bos taurus: new clues from mitochondrial genomes belonging to haplogroup T1. PLoS ONE, 7, e38601. Doro MG, Piras D, Leoni GG et al. (2014) Phylogeny and patterns of diversity of goat mtDNA haplogroup A revealed by resequencing complete mitogenomes. PLoS ONE, 9, e95969. Hassanin A, Bonillo C, Nguyen BX, Cruaud C (2010) Comparisons between mitochondrial genomes of domestic goat (Capra hircus) reveal the presence of numts and multiple sequencing errors. Mitochondrial DNA, 21, 68–76. Kong QP, Salas A, Sun C et al. (2008) Distilling artificial recombinants from large sets of complete mtDNA genomes. PLoS ONE, 3, e3016. Lancioni H, Di Lorenzo P, Ceccobelli S et al. (2013) Phylogenetic relationships of three Italian merino-derived sheep breeds evaluated through a complete mitogenome analysis. PLoS ONE, 8, e73712. Liu J, Wang LD, Sun YB et al. (2012) Deciphering the signature of selective constraints on cancerous mitochondrial genome. Molecular Biology and Evolution, 29, 1255–1261. Miao YW, Peng MS, Wu GS et al. (2013) Chicken domestication: an updated perspective based on mitochondrial genomes. Heredity, 110, 277–282. van Oven M, Kayser M (2009) Updated comprehensive phylogenetic tree of global human mitochondrial DNA variation. Human Mutation, 30, E386–E394. Pang JF, Kluetsch C, Zou XJ et al. (2009) mtDNA data indicate a single origin for dogs south of Yangtze River, less than 16,300 years ago, from numerous wolves. Molecular Biology and Evolution, 26, 2849–2864. Salas A, Carracedo A, Macaulay V, Richards M, Bandelt HJ (2005a) A practical guide to mitochondrial DNA error prevention in clinical, forensic, and population genetics. Biochemical and Biophysical Research Communications, 335, 891–899. Salas A, Yao YG, Macaulay V et al. (2005b) A critical reassessment of the role of mitochondria in tumorigenesis. PLoS Medicine, 2, e296. Salas A, Coble M, Desmyter S et al. (2012) A cautionary note on switching mitochondrial DNA reference sequences in forensic genetics. Forensic Science International Genetics, 6, e182–e184. Wang Z, Shen X, Liu B et al. (2010) Phylogeographical analyses of domestic and wild yaks based on mitochondrial DNA: new data and reappraisal. Journal of Biogeography, 37, 2332–2344. Wang G-D, Xie H-B, Peng M-S, Irwin D, Zhang Y-P (2014) Domestication genomics: evidence from animals. Annual Review of Animal Biosciences, 2, 65–84. Wu GS, Yao YG, Qu KX et al. (2007) Population phylogenomic analysis of mitochondrial DNA in wild boars and domestic pigs revealed multiple domestication events in East Asia. Genome Biology, 8, R245. Yao YG, Kong QP, Salas A, Bandelt HJ (2008) Pseudomitochondrial genome haunts disease studies. Journal of Medical Genetics, 45, 769–772. Yao YG, Salas A, Logan I, Bandelt HJ (2009) mtDNA data mining in GenBank needs surveying. American Journal of Human Genetics, 85, 929–933; author reply 933.

© 2014 John Wiley & Sons Ltd

N E W S A N D V I E W S : O P I N I O N 5397 Yu G, Xiang H, Wang J, Zhao X (2013) The phylogenetic status of typical Chinese native pigs: analyzed by Asian and European pig mitochondrial genome sequences. Journal of Animal Science Biotechnology, 4, 9.

Supporting information Additional supporting information may be found in the online version of this article. Table S1 Access numbers of (near-)complete mtDNA genome sequences.

M.-S.P. and Y.-P.Z. designed research, N.-N.S. conduct the experiment, N.-N.S., L.F. and M.-S.P. preformed the analysis, N.-N.S., Y.-G.Y., M.-S.P. and Y.-P.Z. wrote the manuscript.

Table S2 Potential errors identified in eight domestic animals. Table S3 mtDNA haplogroup tree of dog and grey wolf. Table S4 mtDNA haplogroup tree of goat.

doi: 10.1111/mec.12955

Table S5 mtDNA haplogroup tree of yak.

Data accessibility The mtDNA sequence alignments were deposited in Dryad (doi:10.5061/dryad.cc5kn).

© 2014 John Wiley & Sons Ltd

Table S6 Potential errors detected in next-generation sequencing data.