The Complete Mitochondrial Genome of The

c Indian Academy of Sciences

RESEARCH ARTICLE

The complete mitochondrial genome of the yellow-browed bunting, Emberiza chrysophrys (Passeriformes: Emberizidae), and phylogenetic relationships within the genus Emberiza

QIONGQIONG REN1, JIAN YUAN1, LIQIAN REN2, LIQIN ZHANG1, LEI ZHANG1, LAN JIANG1, DONGSHENG CHEN3, XIANZHAO KAN1,3∗ and BAOWEI ZHANG2∗

1The Institute of Bioinformatics, College of Life Sciences, Anhui Normal University, Wuhu 241000, People’s Republic of China 2School of Life Sciences, Anhui University, Hefei 230039, People’s Republic of China 3The Provincial Key Laboratory of the Conservation and Exploitation Research of Biological Resources in Anhui, Wuhu 241000, People’s Republic of China

Abstract Mitochondrial genomes have proved to be powerful tools in resolving phylogenetic relationships. Emberiza chrysophrys (least concern species: IUCN 2013) is a passerine bird in the bunting family, Emberizidae. The complete mitochondrial genome of E. chrysophrys was sequenced. This circular mitochondrial genome was 16,803 bp in length, with an A+T content of 52.26%, containing 13 protein-coding genes (PCGs), two rRNAs, 22 tRNAs and a putative control region (CR). The CR of E. chrysophrys was divided into three conserved domains. Six conserved sequence boxes in the central conserved domain II were identiﬁed as F, E, D, C, b and B. An obvious positive AT-skew and negative GC-skew bias were found for all 28 genes encoded by the H strand, whereas it was the reverse in the remaining nine genes encoded by the L strand. Remarkable rate heterogeneity was present in the mitochondrial genome of E. chrysophrys. Notably, unusual slow rate of evolution in the mitochondrial CR of E. chrysophrys was detected, which is rarely seen in other birds. Phylogenetic analyses were carried out based on 13 PCGs that showed E. pusilla was the sister group of E. rustica, and the monophyly of Emberiza was established.

[Ren Q., Yuan J., Ren L., Zhang L., Zhang L., Jiang L., Chen D., Kan X. and Zhang B. 2014 The complete mitochondrial genome of the yellow-browed bunting, Emberiza chrysophrys (Passeriformes: Emberizidae), and phylogenetic relationships within the genus Emberiza. J. Genet. 93, 699–707]

Introduction Emberizini which comprised of Old World and New World Emberizini (Alström et al. 2008). The genus Emberiza, along The yellow-browed bunting (Emberiza chrysophrys), least with the other three monotypic genera Melophus, Miliaria concern species (IUCN 2013), is a member of Passeriformes. and Latoucheornis sensu Sibley and Ahlquist (1990), belongs Passeriformes is the largest order of Aves and consists of to the Old World Emberizini clade (Alström et al. 2008). The 123 families (Clements et al. 2013). Traditionally, the avian genus Miliaria and the remaining two genera Melophus and passerine family Fringillidae has the subfamilies Fringilli- Latoucheornis have been moved to Emberiza by Clements nae and Emberizinae, and the latter could be divided into et al. (2013) and Gill and Donsker (2013), respectively. ﬁve tribes: Cardinalini, Emberizini, Icterini, Parulini and The genus Emberiza currently comprises of 44 recognized Thraupini (Sibley and Ahlquist 1990). Recently, the tribe species, distributed throughout Europe, Asia and Africa (Gill Emberizini has been treated as a new family, Emberizidae and Donsker 2013). Based on cytochrome b (cytb)geneand (Clements et al. 2013; Gill and Donsker 2013). Based on the ornithine decarboxylase gene (ODC), Emberiza is divided previous phylogenetic analysis of combined mitochondrial into four main clades (A, B, C and D) (Alström et al. and nuclear data, the result did not support the monophyly of 2008). Clade A can be divided into three subclades forming a trichotomy in the phylogenetic tree (Alström et al. 2008). ∗ Due to polytomous topology, phylogenetic relationships For correspondence. E-mail: Xianzhao Kan, [email protected]; Baowei Zhang, [email protected]. among most species in subclade A1, such as E. pusilla and Qiongqiong Ren, Jian Yuan and Liqian Ren contributed equally to this work. E. rustica, remain unresolved. Keywords. Emberizidae; Passeriformes; mitochondrial genome; phylogenetics; rate heterogeneity; Emberiza chrysophrys.

Journal of Genetics, Vol. 93, No. 3, December 2014 699 Qiongqiong Ren et al.

In animals, the mitochondrial genome (mtDNA) is typi- other species of Passeriformes. Codon usage and nucleotide cally a small, double-stranded circular and compact molecule, composition statistics were computed using MEGA ver. 5.2 generally encoding 37 genes: 13 protein-coding genes, two (Tamura et al. 2011). Composition skew analysis was calcu- ribosomal RNA genes and 22 tRNA genes (Boore 1999; lated using the formulae AT skew = [A − T]/[A + T] and Burger et al. 2003). In addition, it contains a noncoding GC skew = [G − C]/[G + C] (Perna and Kocher 1995), region, known as control region (CR) (Wolstenholme 1992; respectively. The gene map of the E. chrysophrys mitochon- Boore 1999), which is the potential origin for mitochondrial drial genome was initially generated with OGDRAW (Lohse DNA replication and RNA transcription. et al. 2007) and modiﬁed manually. Compared to nuclear genes, mitochondrial DNA (mtDNA) has apparent advantages for phylogenetic studies: Rate heterogeneity a conservative gene content, higher mutation rates, lack of recombination and being maternally inherited (Rubtsov and For the six species of Emberiza, the rates (number of vari- Opaev 2012). Here we present the complete mitochondrial able sites, nucleotide diversity, ratio of nonsynonymous- genome of E. chrysophrys. Based on the new data, combined to-synonymous substitutions rates) and patterns (transition- with previous reports from other species, we tried to address: to-transversion (ts/tv) ratios) of evolution for each gene (i) the observed features of mitochondrial genome of yellow- was determined with DnaSP ver. 5.0 (Librado and Rozas browed bunting, (ii) comparative analysis of mitochondrial 2009), Datamonkey (Delport et al. 2010) and MEGA ver. genomes between yellow-browed bunting and its related 5.2 (Tamura et al. 2011). Rates heterogeneity across the species, and (iii) phylogenetic relationships within the genus mtDNAs of six Emberiza species were assessed by estimat- Emberiza based on the available complete mitochondrial ing the average pairwise sequence divergence per base (π) genomes. in 300 nt windows at every 10 nt. Sliding window analyses using DnaSP ver. 5.0 (Tamura et al. 2011) was carried out. Materials and methods

Sample collection and DNA extraction Phylogenetic analysis The E. chrysophrys specimen (code Kan-K0095) was col- Concatenated nucleotide sequences of 13 protein-coding ◦ ◦ lected from Wuhu (31 21 N, 118 22 E), Anhui, China. The genes (PCGs) in E. chrysophrys and five other known voucher specimen was deposited in the College of Life Sci- mitochondrial genomes of Emberiza were analysed to ences, Anhui Normal University, China. Total genomic DNA understand the phylogenetic relationships within Emberiza was extracted from the muscle tissue using the standard (table 2). Two species from Icteridae (Agelaius phoeniceus phenol–chloroform methods (Sambrook et al. 2001). and Amblyramphus holosericeus) were selected as outgroups. The nucleotide sequences of 13 PCGs were aligned using PCR amplification and sequencing MAFFT 7.058 (Katoh and Standley 2013). An alignment of The entire mt-genome of E. chrysophrys was amplified into 11,397 nucleotides (including 13 stop codons) was obtained. four long overlapping fragments by long and accurate (LA)- A substitution saturation analysis was performed for subsets polymerase chain reaction (LA-PCR) kit (Takara, Dalian, in the first, second and third codon positions using DAMBE China). The resulting amplified fragments were used as tem- (Xia 2013). The results showed that none of the substitu- plates for the nested PCR with 14 primer sets (table 1). Long tions from the three codon positions were saturated. Phy- and subsequently nested PCRs were carried out as designed logenetic reconstructions were conducted using maximum by Kan et al. (2010a). The PCR products were sequenced parsimony (MP), maximum likelihood (ML), and Bayesian directly on ABI-PRISM 3,730 sequencer using BigDye Ter- inference (BI) analyses. MP analysis was performed with minator ver. 3.1 Cycle sequencing kit (Applied Biosystems, PAUP* 4.0b10 (Swofford 2003), with indels being treated Foster City, USA), with their corresponding primers, and as missing character states. A heuristic search was imple- all fragments were sequenced with primer walking on both mented with 100 random addition sequence replicates, tree strands. bisection-reconnection (TBR) branch swapping, the MUL- TREES option and a maximum of 1000 trees saved from each replicate. To evaluate the relative robustness of clades Gene identification and genome analyses found in most parsimonious trees, bootstrap analyses were Sequences were assembled and checked using the performed (Felsenstein 1985). We employed 1000 replicates Sequencher 4.14 (Gene Codes Corporation, Ann Harbor, using the same heuristic search with only 100 maximum trees USA) and BioEdit 7.1.3 software (Hall 1999). The bound- saved per round. PhyML 3.0 (Guindon et al. 2010) was used aries of protein-coding genes were initially identified via for ML analyses of nucleotide dataset with 1000 bootstrap DOGMA (Wyman et al. 2004) using the default setting, and replicates. The best fitting model for the nucleotide dataset refined by aligning with the mitochondrial genomes from was performed with jModelTest2 (Darriba et al. 2012), and

700 Journal of Genetics, Vol. 93, No. 3, December 2014 Mitogenome of Emberiza chrysophrys

Table 1. The primers used in the study.

Primer pair no. Name Sequences (5–3) Size (bp) Sources

1 KLPASXF1 ACCCATTCATCATCATCGGC 20 This study KLPASMTR1 TTGTGGTTTTGGAGCTTGTACG 22 This study 2 KLPASMTF2 ATCTCCAACTCCCAAAGCT 19 This study ZPASMDF2(1) CCTACGACCTGAAAAGCC 18 This study ZPASMDF2(2) ACATCGCCTACACTTGCTC 19 This study KLPASMTR2 GGTATCTAATCCCAGTTTG 19 This study 3 KLPASMTF3 CCCACGGGTATTCAGCAGT 19 This study KLPASMTR3 ACTCTTTGTTGATGGCTGCT 20 This study 5 KLPASMTF4 GAGGTGAAAAGCCAATCGAGC 21 This study KLPASMTR4 GCTAGGGAGAGGATTTGAACC 21 This study KLPASMTF5 AGTCCTACGTGATCTGAGTT 20 This study KLPASMTR5 GGCCCGATAGCTTGTTTAG 19 This study 6 KLPASMTF6 GATAAAGTGAACATAGAGGT 20 This study KLPASMTR6 ATCGAAGCCCATCTGCCTA 19 This study 7 KLPASMTF7 GCCTTCAAAGCCTTAAACAA 20 This study KLPASMTR7 ATGGATAGGACGTAGTGGAA 20 This study 8 ZPASMTF8 GGAGGCACAATCAAATGG 18 This study KLPASMTR8 CTGGTACATAGCTTCTTAAT 20 This study 9 ZPASMTF9 CTACGGACAGTGCTCAGAA 19 This study ZPASMTR9 GAGGGCTTGGATTGCTTG 18 This study 10 KLPASMTF10 AGAACTAGGAGGACAATGAC 20 This study ZPASMTR10 CGATGGATGTTCAGGGAGT 19 This study 11 KLPASMTF11 CCTTCTACAATGCTAAAAAT 20 This study KLPASMTR11 CTTTCACTTGGATTTGCACC 20 This study 12 KLPASMTF12 AAAACCTTCTTACCTGCCGA 20 This study KLPASMTR12 CTTTTGAGTAGAATCCTGCT 20 This study 13 ZPASMTF13 CACAGCCTAAACGGAGAAC 19 This study KLPASMTR13 GAAGGCAGTTGCTATGAGG 19 This study 14 KLPASMTF14 CCCACACCATCAAACATCTC 20 This study KLPASMTR14 GGCTTACAAGACCAATG 17 This study

Table 2. Species examined in the study as classiﬁed by Clements et al. (2013).

Species Family Accession Reference

Emberiza rustica Emberizidae KC831775 Chen et al. (2014) Emberiza pusilla Emberizidae NC_021408 Pan et al. (2013a) Emberiza spodocephala Emberizidae NC_021445 Hu et al. (2013) Emberiza tristrami Emberizidae NC_015234 Kan et al. (2013) Emberiza aureola Emberizidae NC_022150 Pan et al. (2013b) Agelaius phoeniceus Icteridae NC_018801 Powell et al. (2013) Amblyramphus holosericeus Icteridae NC_018802 Powell et al. (2013)

the best model according to Akaike information criterion The pattern of codon usage in the E. chrysophrys mtDNA (AIC) was GTR + G. BI analysis of nucleotide dataset was was also investigated (figure 4). There were 3786 codons performed with MrBayes 3.2 (Ronquist et al. 2012)using for all the 13 PCGs after excluding the termination codons. the GTR model. Four Markov chains were run for 1,000,000 Among their, the codons that were most used Leu (17.56%), generations (sampling every 100 generations) allowing ade- Ala (8.32%),Thr (8.19%), Ser (7.64%), and Ile (7.58%) (figure 4). quate time for convergence. After discarding the first 2500 trees (25%) as burn-in, the remaining 7500 sampled trees were used to estimate the 50% majority rule consensus Results and discussion tree and Bayesian posterior probabilities (BPP). All Markov chain Monte Carlo (MCMC) runs were repeated twice to Genomic organization confirm consistent approximation of the posterior parameter The complete mtDNA sequence of E. chrysophrys was distributions. 16,803 bp in size (figure 1), and the sequence has been

Journal of Genetics, Vol. 93, No. 3, December 2014 701 Qiongqiong Ren et al.

Figure 1. Genomic map and complete mitochondrial genome organization in E. chrysophrys. Genes encode on the heavy or light stands which are shown outside or inside the circular gene map, respectively. The inner ring displays the GC content. 22 tRNA genes are designated by single-letter amino acid codes. The ﬁgure was initially generated with OGDRWA and modiﬁed manually. UNs, number of unassigned nucleotides after the gene; negative values represent overlapping nucleotides. ‘-’ in stop codon indicates termination codons completed via polyadenylation. deposited in GenBank (accession number: NC_015233). The rRNAs, 22 tRNAs and a putative control region (CR), with compact arrangement of E. chrysophrys mtDNA was simi- an identical gene order and length as observed in the above lar to typical avian mtDNA (Dove et al. 2008). The mito- four Emberiza species (Kan et al. 2010a;Huet al. 2013; chondrial genome contains 13 protein-coding genes, two Pan et al. 2013a;Chenet al. 2014). Based on the distribution

Figure 2. Schematic representation of CR organization in E. chrysophrys. ETAS, extended termination-associated sequence; F through B boxes, conserved sequence boxes in the central domain; CSB, conserved sequence block; CSB-like, a sequence similar to the CSB; LSP, light-strand transcription promoter; HSP, heavy-strand transcription promoter.

702 Journal of Genetics, Vol. 93, No. 3, December 2014 Mitogenome of Emberiza chrysophrys

Figure 3. Base composition of E. chrysophrys and other Emberiza mitochondrial genomes. A-T content, A-T skew and G-C skew are computed for each single gene and for many genomic regions following the legend below the chart.

of the conserved motifs in other avian CRs (Castro et al. 2010; Nucleotide composition and codon usage Kan et al. 2010b; Zhang et al. 2012;Kanet al. 2013), the CR of E. chrysophrys can be divided into three domains: The genome E. chrysophrys had an overall A + T content ETAS (extended termination-associated sequence) domain I of 52.26%, which was similar in value among the known (nt 1–432), central conserved domain II (nt 433–848) and Emberiza mitogenomes (range 52.26–52.92%) (figure 3). As CSB (conserved sequence block) domain III (nt 849–1218) in most vertebrates, guanine (G) was the rarest nucleotide, (figure 2). Further, six conserved sequence boxes (F, E, D, which was due to a strong bias against the use of gua- C, b and B) were found in the central conserved domain II nine at the third codon position (San Mauro et al. 2004; (figure 2). Castro et al. 2010). GC and AT skews were measures of

Journal of Genetics, Vol. 93, No. 3, December 2014 703 Qiongqiong Ren et al.

Table 3. Genomic characteristics of Embriza mtDNAs.

Genome Protein-coding gene lrRNA gene srRNA gene tRNA genes Control region Length Length AT% AT% AT% AT% Length Length Length Length Species (bp) AT% (bp) (all) (1st) (2nd) (3rd) (bp) AT% (bp) AT% (bp) AT% (bp) AT%

E. chrysophrys 16,803 55.3 11,397 51.0 48.1 58.6 46.4 1608 54.5 976 50.5 1545 57.9 1218 55.6 E. tristrami 16,789 52.6 11,394 51.3 48.0 58.5 47.4 1602 55.7 974 50.3 1542 57.7 1217 57.5 E. pusilla 16,790 52.4 11,400 51.1 47.8 58.6 47.0 1600 55.0 976 50.3 1543 57.7 1212 56.6 E. spodocephala 16,796 52.9 11,400 51.7 47.9 58.6 48.8 1601 55.5 977 50.3 1541 57.9 1215 56.5 E. rustica 16,798 52.8 11,400 51.5 48.1 58.6 47.8 1600 55.1 976 50.5 1545 58.3 1216 57.2 E. aureola 16800 53.0 11400 51.8 48.2 58.7 48.3 1602 55.2 975 50.3 1544 58.2 1218 57.6

AT is the content of nucleotide A + T, and AT% is the percentage of A + T content.

Figure 4. Codon distribution in six Emberiza species mitogenome. Numbers to the left refer to the total number of codons. CDsp T, codons per thousand codons. Codon families are given on the x-axis.

Table 4. Rates and patterns of evolution among mitochondrial genes and species of Emberizidae.

Gene Length (bp) Var. sites (%) π dN/dS ts/tv srRNA 979 55 (5.62) 0.02323 – 6.78 lrRNA 1609 111 (6.90) 0.03010 – 6.26 nad1 978 190 (19.43) 0.08303 0.01484 10.68 nad2 1041 225 (21.61) 0.09366 0.05572 10.06 cox1 1551 265 (17.09) 0.07136 0.01414 8.28 cox2 684 95 (13.89) 0.05750 0.01055 12.39 atp8 168 29 (17.26) 0.07103 0.09786 4.09 atp6 684 115 (16.81) 0.07057 0.01350 8.83 cox3 786 100 (12.72) 0.05403 0.00346 8.27 nad3 351 64 (18.23) 0.08205 0.04664 8.90 nad4L 297 54 (18.18) 0.07924 0.02180 11.26 nad4 1380 222 (15.00) 0.07024 0.01742 7.95 nad5 1818 284 (16.09) 0.07332 0.03744 6.48 cytb 1143 177 (15.49) 0.06311 0.01863 7.78 nad6 519 81 (15.61) 0.06667 0.02405 7.19 CR 1219 192 (15.75) 0.06772 – 2.05 tRNAs 1546 82 (5.30) 0.02330 – 11.93 Overall 16816 2389 (14.21) 0.06049 – 6.85

Var. sites, variable sites; π, nucleotide diversity; dN, nonsynonymous rate; dS, synonymous rates; ts/tv, transition-to-transversion ratio. nad, NADH dehydrogenase subunit; cox, cytochrome c oxidase subunit; cytb, cytochrome b; CR, control region.

704 Journal of Genetics, Vol. 93, No. 3, December 2014 Mitogenome of Emberiza chrysophrys

The pattern of codon usage in the E. chrysophrys mtDNA was also investigated (ﬁgure 4). There were 3786 codons for all the 13 PCGs after excluding the termination codons. Among them, codons that were most used are as follows: Leu (17.56%), Ala (8.32%),Thr (8.19%), Ser (7.64%) and Ile (7.58%) (ﬁgure 4).

Rates and patterns of mitochondrial gene evolution Comparison of the evolutionary rate of different genes is needed to understand the patterns of molecular evolution of the mtDNA. With no large insertions, deletions, or duplications in mitochondrial genomes, gene order and genome size are conserved among the six Emberiza taxa. Figure 5. Rate heterogeneity across four bunting mitochondrial Among the six mtDNA sequences, 2389 nucleotides sites genomes. A sliding window of 300 bp with step size 10 bp and are variable (table 4). In the protein coding region, the average pairwise sequence divergence (π) was recorded. Gene most variable region of the genome by percent variable sites abbreviations are as given in table 4. are the nad2, followed by nad1, nad3 and nad4L. tRNA genes are characterized by the lowest percentage variability, compositional asymmetry. The AT and GC skews for the followed by srRNA and lrRNA genes (table 4). Further, slid- mitochondrial genome of E. chrysophrys were 0.133 and ing window analysis of the complete nucleotide alignment −0.380, respectively. Interestingly, for all six Emberiza of six available Emberiza aves mtDNAs provide an indica- mtDNA, an obvious positive AT-skew and negative GC-skew tion of average pairwise sequence divergence (π) which dis- bias were found for all 28 genes encoded by the H strand, plays a pattern of heterogeneity across Emberiza genomes whereas the reverse were detected in the remaining nine (figure 5). Not unexpectedly, rates are reduced in the regions genes encoded by the L strand (figure 3). The AT composition associated with the srRNA and lrRNA genes, elevated in at the first codon position of E. chrysophrys mitogenome the nad1 and nad2 genes. In addition to the nad6 gene, PCGs was 48.1% (table 3). The values of the second and third rates are also noticeably higher in nad gene regions than in codon positions were 58.6 and 46.4%, respectively (table 3). cox or atp genes. The patterns of overall and synonymous

Figure 6. Inferred phylogenetic relationship among six Emberiza. Phylogenetic tree of the relationships among six Emberiza based on the nucleotide sequences of the 13 mitochondrial protein-coding genes. Age- laius phoeniceus and Amblyramphus holosericeus served as outgroups. Branch length and topology are taken from the Bayesian inference analyses. Numbers beside the nodes specify bootstrap percentages from ML (500 replicates) and MP (1000 replicates) and posterior probabilities from BI. *Represent low bootstrap. Scale bar represents substitutions/site.

Journal of Genetics, Vol. 93, No. 3, December 2014 705 Qiongqiong Ren et al. rate heterogeneity observed provide additional data about weekly supported in ML and MP but strongly supported in BI a phenomenon of rate heterogeneity observed at different analysis. Based on the phylogeny of cytochrome b and ODC hierarchical levels in passerine birds (Marshall et al. 2013). genes, Alström et al. (2008) found the clade (E. chrysophrys Interestingly, unusual slow rate of evolution was observed in + (E. tristrami + E. variabilis) which is the only part of the CR of E. chrysophrys mitochondrial genome. Rate was clade A1 that receives consistent high support. The analyses slightly higher in CR than in cox2, cox3, cytb and nad6 genes. on phylogenetic relationship between E. chrysophrys and The general rule for metazoan CRs has been that it appears E. tristrami strongly support Alström et al. (2008) result. to evolve faster than the rest of mitochondrial genome, espe- Further, a sister relationship between E. pusilla and E. rustica cially in humans and aves (Aquadro and Greenberg 1983; are recovered as a highly supported clade (bootstrap value Mindell 1997). However, several reports, including this 100% and 99% in ML and MP, respectively, posterior prob- study, have challenged the generality of this observation. A abilities =1.00 in BI). These two clades share a most recent slow rate of DNA substitution in the CR of mtDNA has been common ancestor with the E. aureola. Additional data would found in butterflies (Taylor et al. 1993), gulls (Crochet and be needed to resolve the phylogenetic relationship with the Desmarais 2000), fishes and mammals (Brown et al. 1986; genus Emberizidae. Tang et al. 2006). The transition-to-transversion ratio (ts/tv) in PCGs varies from atp8 (4.09) to cox2 (12.39). Transversions are typi- Acknowledgments cally rare in the slow-evolving regions (like the tRNAs), but the patterns among PCGs are variable and does not reflect The authors would like to thank Binghua Li, Xifeng Li and Lei Chen for assistance, suggestions and comments. This study differences in either synonymous or nonsynonymous rates was supported by the Key Laboratory of Biotic Environment among the genes. Unusually high ts/tv ratios were observed and Ecology Safety in Anhui Province, the Key Programme of for cox2, nad4L, nad1 and nad2 genes, which have inter- Natural Science Foundation of the Anhui Higher Education Institu- mediate rates in both these substitution categories; the low tions (no. KJ2013A127), and the Anhui Provincial Natural Science level of transversions in these genes were most likely due to Foundation (no. 1308085MC38). peculiarities in base composition. Typically, the highest dN/dS was observed for the atp8 gene and the lowest for the cox1 gene among Passeriformes References (Marshall et al. 2013); within Emberiza the highest dN/dS Alström P., Olsson U., Lei F., Wang H.-T., Gao W. and Sundberg P. was also observed for atp8 while the cox3 gene contained 2008 Phylogeny and classification of the Old World Emberizini more variable synonymous sites (table 4), which suggest (Aves, Passeriformes). Mol. Phylogenet. Evol. 47, 960–973. that functional constraints rather than positive selection were Aquadro C. F. and Greenberg B. D 1983 Human mitochondrial strong for this bar-coding gene. This was also the conclusion DNA variation and evolution: analysis of nucleotide sequences of Kerr (2011), which contradicts earlier speculation that the from seven individuals. Genetics 103, 287–312. Boore J. L. 1999 Animal mitochondrial genomes. Nucleic Acids high between-species variation and low within-species varia- Res. 27, 1767–1780. tion in cox1 were due to recurrent bouts of positive selection Brown G. G., Gadaleta G., Pepe G., Saccone C. and Sbisà E. 1986 (Kerr et al. 2007). Moreover, an average dN/dS for the nad Structural conservation and variation in the D-loop-containing genes was about five times high compared to the cox genes, region of vertebrate mitochondrial DNA. J. Mol. Biol. 192, which was similar to Fringilla (table 4)(Marshallet al. 503–511. Burger G., Gray M. W. and Franz Lang B. 2003 Mitochondrial 2013). Higher dN/dS ratios at different hierarchical levels for genomes: anything goes. Trends Genet. 19, 709–716. the nad genes relative to the cox genes were consistent with Castro J. A., Picornell A. and Ramon M. 2010 Mitochondrial relaxation of functional constraints on the former or height- DNA: a tool for populational genetics studies. Int. Microbiol. 1, ened restraints on the latter. Mishmar et al. (2006) provide 327–332. evidence that three nuclear DNA complex I genes have been Chen J, Pan T., Hu L., Ren L., Yu L. and Zhang B. 2014 Mitochon- drial genome of the Emberiza rustica (Emberizidae: Emberiza). subjected to adaptive selection during primate evolution and Mitochondrial DNA 25, 284–285. the mitochondrial nad subunits have coevolved. Clements J., Schulenberg T., Iliff M., Sullivan B., Wood C. and Roberson D. 2013 The eBird/Clements checklist of birds of the world: version 6.8. (http://www.birds.cornell.edu/ Phylogenetic analyses clementschecklist/downloadable-clements-checklist). Accessed 2 March 2014. The MP, ML and BI trees, which were based on the Crochet P.-A. and Desmarais E. 2000 Slow rate of evolution in the nucleotide dataset of 13 protein-coding genes, showed iden- mitochondrial control region of gulls (Aves: Laridae). Mol. Biol. tical topologies (figure 6). The monophyly of Emberiza Evol. 17, 1797–1806. was strongly supported (bootstrap value 100% in ML and Darriba D., Taboada G. L., Doallo R. and Posada D. 2012 jModel- MP, posterior probabilities = 1.00 in BI) (figure 6). E. Test 2: more models, new heuristics and parallel computing. Nat. Methods 9, 772. chrysophrys was found more close to E. tristram (bootstrap Delport W., Poon A. F., Frost S. D. and Pond S. L. K. 2010 value 99% in ML and MP, respectively, posterior probabil- Datamonkey 2010: a suite of phylogenetic analysis tools for ities = 1.00 in BI), and sister to the E. spodocephala with evolutionary biology. Bioinformatics 26, 2455–2457.

706 Journal of Genetics, Vol. 93, No. 3, December 2014 Mitogenome of Emberiza chrysophrys

Dove C. J., Rotzel N. C., Heacker M. and Weigt L. A. 2008 Using Mishmar D., Ruiz-Pesini E., Mondragon-Palomino M., Procaccio DNA barcodes to identify bird species involved in birdstrikes. V., Gaut B. and Wallace D. C. 2006 Adaptive selection of mito- J. Wildl. Manage. 72, 1231–1236. chondrial complex I subunits during primate radiation. Gene 378, Felsenstein J. 1985 Confidence limits on phylogenies: an approach 11–18. using the bootstrap. Evolution 39, 783–791. Pan T., Wang H., Zhu X., Zhou W., Hou Y., Yu L. et al. 2013a Gill F. and Donsker D. 2013 IOC World bird names (ver3.4). Avail- Mitochondrial genome of the Emberiza pusilla (Emberizidae: able at (http://www.worldbirdnames.org). Accessed 2 March Emberiza). Mitochondrial DNA 24, 382–384. 2014. Pan T., Ren L. Q., Zhu X. X., Yan L. H., Hu C. C., Chang Q. Guindon S., Dufayard J.-F., Lefort V., Anisimova M., Hordijk W. and Zhang B. W. 2013b Mitochondrial genome of the Emberiza and Gascuel O. 2010 New algorithms and methods to estimate aureola (Emberizidae: Emberiza). Mitochondrial DNA (in press). maximum-likelihood phylogenies: assessing the performance of Perna N. T. and Kocher T. D. 1995 Patterns of nucleotide com- PhyML 3.0. Syst. Biol. 59, 307–321. position at fourfold degenerate sites of animal mitochondrial Hall T. A. 1999 BioEdit: a user-friendly biological sequence genomes. J. Mol. Evol. 41, 353–358. alignment editor and analysis program for windows 95/98/NT,. Powell A. F., Barker F. K. and Lanyon S. M. 2013 Empirical evalua- Nucleic Acids Symp. Ser. 41, 95–98. tion of partitioning schemes for phylogenetic analyses of mitoge- Hu L., Pan T., Zhu X., Wang H., Hou Y., Yu L. et al. 2013 nomic data: an avian case study. Mol. Phylogenet. Evol. 66, Mitochondrial genome of the black-faced bunting (Emberiza 69–79. spodocephala). Mitochondrial DNA 25, 165–166. Ronquist F., Teslenko M., Van Der Mark P., Ayres D. L., Darling IUCN 2013 IUCN red list of threatened species. Version 2013.2. A., Höhna S. et al. 2012 MrBayes 3.2: efficient Bayesian phylo- Available at (http://www.iucnredlist.org). Accessed 5 March genetic inference and model choice across a large model space. 2014. Syst. Biol. 61, 539–542. Kan X. Z., Li X. F., Lei Z. P., Wang M., Chen L., Gao H. Rubtsov A. and Opaev A. 2012 Phylogeny reconstruction of the et al. 2010a Complete mitochondrial genome of Cabot’s yellowhammer (Emberiza citrinella) and pine bunting (Emberiza tragopan, Tragopan caboti (Galliformes: Phasianidae). Genet. leucocephala) based on song and morphological characters. Biol. Mol. Res. 9, 1204–1216. Bull. 39, 715–728. Kan X. Z., Li X. F., Zhang L. Q., Chen L., Qian C. J., Zhang X. Sambrook J., Russell D. W. and Russell D. W. 2001 Molecular W. et al. 2010b Characterization of the complete mitochondrial cloning: a laboratory manual (3-volume set). genome of the Rock pigeon, Columba livia (Columbiformes: San Mauro D., García-París M. and Zardoya R. 2004 Phylogenetic Columbidae). Genet. Mol. Res. 9, 1234–1249. relationships of discoglossid frogs (Amphibia: Anura: Discoglos- Kan X., Yuan J., Zhang L., Li X., Yu L., Chen L. et al. 2013 Com- sidae) based on complete mitochondrial genomes and nuclear plete mitochondrial genome of the Tristram’s Bunting, Ember- genes. Gene 343, 357–366. iza tristrami (Aves: Passeriformes): the first representative of Sibley C. G. and Ahlquist J. 1990 Phylogeny and classification of the family Emberizidae with six boxes in the central conserved birds. Yale University Press, New Haven, Connecticut. domain II of control region. Mitochondrial DNA 24, 648–650. Swofford D. L. 2003 PAUP* Phylogenetic analysis using parsi- Katoh K. and Standley D. M. 2013 MAFFT multiple sequence mony (* and other methods). Version 4.0 b10. alignment, software version 7: improvements in performance and Tamura K., Peterson D., Peterson N., Stecher G., Nei M. and Kumar usability. Mol. Biol. Evol. 30, 772–780. S. 2011 MEGA5: molecular evolutionary genetics analysis using Kerr K. C. 2011 Searching for evidence of selection in avian DNA maximum likelihood, evolutionary distance, and maximum par- barcodes. Mol. Ecol. Notes 11, 1045–1055. simony methods. Mol. Biol. Evol. 28, 2731–2739. Kerr K. C., Stoeckle M. Y., Dove C. J., Weigt L. A., Francis C. M. Tang Q., Liu H., Mayden R. and Xiong B. 2006 Comparison of and Hebert P. D. 2007 Comprehensive DNA barcode coverage of evolutionary rates in the mitochondrial DNA cytochrome b gene North American birds. Mol. Ecol. Notes 7, 535–543. and control region and their implications for phylogeny of the Librado P. and Rozas J. 2009 DnaSP v5: a software for compre- Cobitoidea (Teleostei: Cypriniformes). Mol. Phylogenet. Evol. hensive analysis of DNA polymorphism data. Bioinformatics 25, 39, 347–357. 1451–1452. Taylor M., Mckechnie S. W., Pierce N. and Kreitman M. 1993 Lohse M., Drechsel O. and Bock R. 2007 OrganellarGenome- The lepidopteran mitochondrial control region: structure and DRAW (OGDRAW): a tool for the easy generation of high- evolution. Mol. Biol. Evol. 10, 1259–1272. quality custom graphical maps of plastid and mitochondrial Wolstenholme D. R. 1992 Animal mitochondrial DNA: structure genomes. Curr. Genet. 52, 267–274. and evolution. Int. Rev. Cytol. 141, 173–216. Marshall H. D., Baker A. J. and Grant A. R. 2013 Com- Wyman S. K., Jansen R. K. and Boore J. L. 2004 Automatic anno- plete mitochondrial genomes from four subspecies of common tation of organellar genomes with DOGMA. Bioinformatics 20, chaffinch (Fringilla coelebs): new inferences about mito- 3252–3255. chondrial rate heterogeneity, neutral theory, and phyloge- Xia X. 2013 DAMBE5: a comprehensive software package for data netic relationships within the order Passeriformes. Gene 517, analysis in molecular biology and evolution. Mol. Biol. Evol. 30, 37–45. 1720–1728. Mindell D. P. 1997 Avian molecular evolution and systematics. In Zhang L., Wang L., Gowda V., Wang M., Li X. and Kan X. 2012 Mitochondrial control region sequences as tools for understand- The mitochondrial genome of the Cinnamon Bittern, Ixobrychus ing evolution (ed. A. J. Baker and H. Dawn Marshall), pp. 51–82. cinnamomeus (Pelecaniformes: Ardeidae): sequence, structure Academic Press, Toronto. and phylogenetic analysis. Mol. Biol. Rep. 39, 8315–8326.

Received 11 March 2014; accepted 30 May 2014 Unedited version published online: 25 June 2014 Final version published online:12 November 2014

Journal of Genetics, Vol. 93, No. 3, December 2014 707