Gene xxx (2014) xxx–xxx

GENE-39552; No. of pages: 10; 4C:

Contents lists available at ScienceDirect

Gene

j ourna l homepage: www.e lsev ie r .com/ locate /gene

Mitogenomic circumscription of a novel percomorph fish clade mainlycomprising “Syngnathoidei” (Teleostei)

Ha Yeun Song a,⁎, Kohji Mabuchi a, Takashi P. Satoh b, Jon A. Moore c, Yusuke Yamanoue d,Masaki Miya e, Mutsumi Nishida a,f

a Atmosphere and Ocean Research Institute, The University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa, Chiba 277-8564, Japanb National Museum of Nature and Science, Collection Center, 4-1-1 Amakubo, Tsukuba, Ibaraki 305-0005, Japanc Florida Atlantic University, Wilkes Honors College, Jupiter, FL 33458, USA & Harbor Branch Oceanographic Institution, Fort Pierce, FL 34946, USAd Fisheries Laboratory, Graduate School of Agricultural and Life Sciences, The University of Tokyo, 2971-4 Bentenjima, Maisaka-cho, Nishi-ku, Hamamatsu 431-0214, Japane Natural History Museum and Institute, Chiba, 955-2 Aoba-cho, Chuo-ku, Chiba 260-8682, Japanf University of the Ryukyus, 1 Senbaru, Nishihara-cho, Okinawa 903-0213, Japan

Abbreviations: A, adenosine; BA, Bayesian; Bp, bprobability(ies); C, cytidine; G, guanosine; MCMC,mitogenome, mitochondrial genome; ML, maximum likenucleotide(s); nt, nucleotide(s); PP(s), posterior prothymidine.⁎ Corresponding author.

E-mail address: songhy@aori.u-tokyo.ac.jp (H.Y. Song)

http://dx.doi.org/10.1016/j.gene.2014.03.0400378-1119/© 2014 Elsevier B.V. All rights reserved.

Please cite this article as: Song, H.Y., et al., M(Teleostei), Gene (2014), http://dx.doi.org/1

a b s t r a c t

a r t i c l e i n f o

Article history:Received 25 July 2013Received in revised form 10 March 2014Accepted 20 March 2014Available online xxxx

Keywords:CallionymoideiDactylopteridaeMullidaePercomorphaSyngnathiformesSeahorses

Percomorpha, comprising about 60% of modern teleost fishes, has been described as the “(unresolved) bush atthe top” of the tree, with its intrarelationships still being ambiguous owing to huge diversity (N15,000 species).Recentmolecular phylogenetic studies based on extensive taxon and character sampling, however, have revealeda number of unexpected clades of Percomorpha, and one of which is composed of Syngnathoidei (seahorses,pipefishes, and their relatives) plus several groups distributed across three different orders. To circumscribethe clade more definitely, we sampled several candidate taxa with reference to the previous studies and newlydetermined whole mitochondrial genome (mitogenome) sequences for 16 percomorph species acrosssyngnathoids, dactylopterids, and their putatively closely-related fishes (Mullidae, Callionymoidei, Malacanthidae).Unambiguously aligned sequences (13,872 bp) from those 16 species plus 78 percomorphs and two outgroups(total 96 species) were subjected to partitioned Bayesian andmaximum likelihood analyses. The resulting trees re-vealed a highly supported clade comprising seven families in Syngnathoidei (Gasterosteiformes), Dactylopteridae(Scorpaeniformes), Mullidae in Percoidei and two families in Callionymoidei (Perciformes). We herein proposedto call this clade “Syngnathiformes” following the latest nuclear DNA studies with some revisions on the includedfamilies.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Modern teleost fishes are themost species rich and diversified groupof all the vertebrates (Nelson, 2006). They include 26,840 species ac-counting for 96% of all extant fishes, and dominate the world's rivers,lakes, and oceans (Nelson, 2006). The top of the teleostean phylogenieshas been described as the “(unresolved) bush at the top of the tree”(Nelson, 1989), and is occupied by Percomorpha (sensu Miya et al.,2003, 2005; Wiley and Johnson, 2010), which comprises some 15,322species (57% of all living teleosts) placed in 13 orders, 269 families,and 2537 genera (calculated from Nelson, 2006). Considering its enor-mous species diversity coveringmore than half of the living fish species,together with their ancient origin (180–200 million years ago, Azuma

ase pair(s); BP(s), bootstrapMarkov chain Monte Carlo;lihood; mt, mitochondria(l); n,babilitie(s); r, RY-coding; T,

.

itogenomic circumscription0.1016/j.gene.2014.03.040

et al., 2008; 100–120 million years ago, Near et al., 2012) and thewide-ranging variations not only in morphology but also in ecologyand physiology (see Helfman et al., 2009), it is no wonder that theintrarelationships of the Percomorpha remain highly ambiguous(“unresolved bush” as mentioned above) despite numerous compara-tive anatomical studies (for recent reviews, see Helfman et al., 2009;Nelson, 2006) and early molecular phylogenetic studies based onshorter sequences (mostly b1 kp) from limited taxonomic representa-tion (e. g., see Kocher and Stepien, 1997).

A series of recent studies based on whole mitochondrial genome(mitogenome) sequences (N15 kb) has, however, shed a new light onthese relationships. These studies revealed unexpected higher-level re-lationships that are often inconsistent with traditional systematics. Theexamples include polyphyly of the suborder Labroidei (Mabuchi et al.,2007), close relationships between the orders Tetraodontiformes andLophiiformes (and Caproidei) (Miya et al., 2003; Yamanoue et al.,2007), polyphyly of the order Gasterosteiformes (Kawahara et al.,2008), and phylogenetic affinity between the series “Atherinomorpha”and the percomorph fishes that spawn demersal eggs with filaments(Setiamarga et al., 2008). Interestingly these mitogenomic studies,

of a novel percomorph fish clade mainly comprising “Syngnathoidei”

2 H.Y. Song et al. / Gene xxx (2014) xxx–xxx

together with earlier studies by Miya et al. (2003, 2005), have consis-tently reproduced eight unnested major clades (Clades A to H) withinthe Percomorpha, and most of the eight clades have been reproducedalso in the recent nuclear gene studies using long (N7 kb) concatenatedsequences (Betancur-R et al., 2013; Near et al., 2013) (Table 1). As a firststep towards the resolution of percomorph phylogeny, revision and cir-cumscription of each of the eight clades with more comprehensivetaxon sampling should be very important.

In this study, we have focused on one of the eight clades, “Clade D,”whichwas tentatively named as such in Kawahara et al. (2008) [follow-ed by Yagishita et al. (2009) and this study]. Until the recent nucleargene analyses, molecular phylogenetic studies using partial mito-chondrial and shorter nuclear DNA (total b 5 kb) indicated that thesyngnathoids may have phylogenetic affinities with the familiesDactylopteridae (flying gurnards), Mullidae (goatfishes) and/or thesuborder Callionymoidei (dragonets) (Dettai and Lecointre, 2005; Liet al., 2009; Smith and Craig, 2007; Smith and Wheeler, 2006; see Fig.1c, d, e, h). Although the recent nuclear gene studies indicated mono-phyly of these fishes (Betancur-R et al., 2013; Near et al., 2012, 2013;see Fig. 1i, j, k), none of these and the previous studies have includedall of the families to be tested (see Table 2).

To circumscribe the percomorph clade mainly comprising“Syngnathoidei” more definitely, we newly determined whole mito-chondrial genome sequences for the five species of the Syngnathoidei,one species of Dactylopteridae, and their putatively closely-related fami-lies Mullidae (4 spp.), Callionymoidei (5 spp.) and Malacanthidae (1sp.).Unambiguously aligned sequences (13,872 bp) from those 16 speciesplus 78 percomorphs (including previously published nine syngnathoids,two dactylopterids, one callionymoid, and one malacanthid) and twooutgroups (total 96 species) were subjected to partitioned Bayesian(BA) andmaximum likelihood (ML) analyses. The resulting trees revealeda highly supported clade that solely comprises benthic or semi-benthicspecies placed in three different percomorph orders (Gasterosteiformes,Scorpaeniformes, and Perciformes).

2. Materials and methods

2.1. Taxonomic sampling

We used whole mitogenome sequences from 96 higher teleosts, ofwhich those from 75 species were identical to those used in the previ-ous phylogenetic study of “Gasterosteiformes” by Kawahara et al.(2008). In traditional systematics, the suborder Syngnathoidei is placedwithin the order Gasterosteiformes, together with the other suborderGasterosteoidei (sticklebacks). The study of Kawahara et al. (2008)investigated relationships of all the 11 families of Gasterosteiformesand found that the order was polyphyletic, comprising three line-ages [Gasterosteoidei (minus Indostomidae), Syngnathoidei, andIndostomidae] that have diverged basally within the percomorphs. Inthat study, the clade comprising Syngnathoidei and Dactylopteridae was

Table 1Eight major clades within the Percomorpha consistently reproduced in the previous and presen(2009)] with the corresponding clades reproduced in the latest two nuclear gene studies (Beta

Clades in: Representative member(series, orders, suborde

Mitogenomicstudies

Betancur-R et al.(2013)

Near et al. (2013)

A “Anabantomorphariae” V Synbranchiformes, ChanB “Ovarentaria” VII (“Ovarentaria”) Atherinomorpha, MugilC “Gobiomorpharia” IV (“Gobiiformes”) Gobioidei, Apogonidae,D “Syngnathiformes” III (“Syngnathiformes”a) Syngnathoidei, CallionyE “Carangimorphariae” VI Pleuronectiformes, CaraF “Scombriformes” III (“Scombriformes”) Bramidae, CentrolophidG “Perciformes” XIII (“Perciformes”) Cottoidei, ScorpaenoideH “new bush at the top” VIII–XII, XIV Lophiiformes, Tetraodon

a Not including the family Mullidae.

Please cite this article as: Song, H.Y., et al., Mitogenomic circumscription(Teleostei), Gene (2014), http://dx.doi.org/10.1016/j.gene.2014.03.040

recovered (and tentatively named as Clade D, see Fig. 1f), but their possi-ble closely related groups, such as the Mullidae and Callionymoidei werenot included.

To circumscribe the “Clade D”with more extensive taxon sampling,we added whole mitogenome sequences from the following 21 speciesto the data set used in Kawahara et al. (2008) (total=96 spp.):five spe-cies of Syngnathoidei (all fivewere newly determined here), six speciesof Callionymoidei (five newly determined), four species ofMullidae (4),one species of Dactylopteridae (1), two species of Malacanthidae (1),two species of Nomeidae, and one species of Arripidae. The familyMalacanthidae was included because a relatively recent comparativeanatomical study (Imamura, 2000) proposed its close relationship tothe family Dactylopteridae. The families Nomeidae and Arripidae wereincluded because of their possible affinity to Dactylopteridae as shownin the mitogenomic trees of Yagishita et al. (2009) (not shown inFig. 1). Among the 96 species included in the final data set, ninety-fourwere percomorphs (sensuMiya et al., 2003, 2005), and the remain-ing two [Polymixia japonica (Polymixiidae) and Beryx splendens(Berycidae)] were used as collective outgroups to root the percomorphtree with reference to Miya et al. (2003, 2005). Of the 94 percomorphspecies, 14 were syngnathoids, representing all the seven families inthe suborder. A list of the 96 species is provided in Table 3, along withDDBJ/EMBL/GenBank accession numbers and references.

2.2. DNA extraction, PCR, and sequencing

Total genomic DNA was extracted from frozen specimens usingAquaPure genomic DNA isolation kit (Qiagen), following themanufacturer's protocol. The mitogenome sequences of the 16 spe-cies were amplified in their entirety using a long PCR technique(Cheng et al., 1994; Miya and Nishida, 1999) with six fish-versatilelong PCR primers. Long PCR conditions followed Miya and Nishida(1999), with the following four combinations of the long PCRprimers: L2508-16S + H12293-Leu; L2508-16S + H15149-CYB;L2508-16S + H1065-12S; and L12321-Leu + S-LA-16S-H (Inoue et al.,2001, 2004; Kawaguchi et al., 2001). The long PCR products dilutedwith sterile water (1: 9 to 19) served as templates for subsequent shortPCR. We used fish-versatile PCR primers in various combinations to am-plify contiguous, overlapping segments of the entire mitogenome foreach of the 16 species (the locations and sequences of primers availablefromHY. S.). The short PCR conditions followedMiya andNishida (1999).

Double-stranded short PCR products were purified using an Exosap-IT enzyme reaction (GE Healthcare Bio-Sciences Corp., Piscataway, NT,USA). These were subsequently used for the direct cycle sequencingwith dye-labeled terminators (BigDye terminator ver. 3.1, AppliedBiosystems, Foster City, CA, USA) and using the same primers as thoseused in the short PCRs. All sequencing reactions were carried out ac-cording to the manufacturer's instructions. Labeled fragments were an-alyzed on an ABI PRISM 3130 xlGenetic Analyzer (Applied Biosystems).

t mitogenomic studies [A–H: names following Kawahara et al. (2008) and Yagishita et al.ncur-R et al., 2013; Near et al., 2013) and representative members.

srs, and families in Nelson, 2006)

nidae, Indostomidae, Osphronemidaeomorpha, Blennioidei, Gobiesocoidei, Cichlidae, Embiotocidae, PomacentridaeKurtidaemoidei, Dactylopteridae, Mullidaengidae, Centropomidae, Echeneidae, Sphyraenidae, Xiphiidaeae, Chiasmodontidae, Scombridae, Gempylidae, Nomeidae, Trichiuridaei, Platycephaloidei, Gasterosteidae, Notothenidae, Percidae, Serranidaetiformes, Acanthuroidei, Centrarchidae, Chaetodontidae, Labridae, Lutjanidae, Sparidae

of a novel percomorph fish clade mainly comprising “Syngnathoidei”

3H.Y. Song et al. / Gene xxx (2014) xxx–xxx

2.3. Sequence editing and alignment

The sequence editing and alignment were both conducted with thecomputer program ATGC ver. 4.0 (Genetyx Corporation, Tokyo, Japan).We carefully examined the concatenated sequences using DNASIS ver3.2 (Hitachi software Engineering Co. Ltd). For each individualprotein-coding gene, we aligned the sequences for the 96 species,with reference to the translated amino acid sequence using MAFFT v.6 (Katoh and Toh, 2008) and being corrected manually using MacCladever. 4.08 (Maddison and Maddison, 2000). All positions including stopcodon and ambiguously aligned regions were excluded from thesubsequent phylogenetic analyses. The 12S and 16S rRNA and 22 tRNAgene sequences were aligned using the software Proalign ver. 0.5(Löytynoja andMillinkovitch, 2003) with default setting of the parame-ters. Regions with posterior probabilities (PPs) of ≤50% were excludedfrom the subsequent phylogenetic analyses. All sequences from the L-strand-encoded eight tRNA genes were converted into complementarystrand sequences. The ND6 gene and the putative control region wereexcluded because of its heterogeneous base composition and poor phy-logenetic performance (Miya and Nishida, 2000). The complete data setincluded 13,872 positions, comprising 12 protein-coding genes (10,815positions), two rRNA coding genes (1706 positions) and 22 tRNA genes(1351 positions).

2.4. Phylogenetic analysis

Phylogenetic analyses were conducted on three data sets: thefirst matrix includes all aligned positions of gene-coding regions ofmitogenomic sequences (123nRTn; “n” denotes nucleotides); the sec-ond data set includes only transversion (R-Y) substitutions at the thirdcodon positions of protein-coding genes (12n3rRTn; subscript “r” de-notes RY-coding). This matrix was constructed by replacing the thirdcodon purines (A/G) with “R”, and the third codon pyrimidines (C/T)with “Y” (Phillips and Penny, 2003; Phillips et al., 2004); the thirddata sets include all the concatenated nucleotide sequences of themitogenomes except the third codon positions (12nRTn). The data ma-trices were then subjected to partitioned Bayesian (BA) and partitionedmaximum-likelihood (ML) analyses. Five (123nRTn and 12n3rRTn) andfour (12nRTn) partitions were set for each analysis.

Partitioned BA analyses were conducted with Mr. Bayes ver. 3.1.2(Ronquist and Huelsenbeck, 2003) for the above three data sets. TheGTR + I + Γ model [selected as the best model by ModelTest 3.06(Posada and Crandall, 1998) for each partition] was used for nucleotidesubstitution model (Yang, 1994). We assumed that all of the model pa-rameters were unlinked and that the rate multipliers were variableacross partitions and used the default settings for the priors on the pro-portion of invariable sites (0–1) and gamma shape parameters(0.1–50.0). A Dirichlet distribution was assumed for the rate matrixand base frequencies, and every tree topology was assumed equallyprobable. For the third codon positions in 12n3rRTn coding, we used ar-bitrarily “A” and “C” instead of “R” and “Y” respectively, and set a singlerate category for this partition (lset nst= 1) instead of six (lset nst= 6)to avoid unnecessary estimation of transitional changes during thecalculations.

The Markov chain Monte Carlo (MCMC) process was set so thatfour chains (three heated and one cold) ran simultaneously. We con-ducted two independent runs for each data set and continued for12,000,000 cycles, with one in every 1000 trees being sampled. “Sta-tionarity” (lack of improvement in the likelihood score) was checkedgraphically using Tracer ver. 1.5 (available from http://tree.bio. ed. ac.uk/software/tracer/) and all trees and parameters before stationaritywere discarded as “burn-in.” After confirming the agreement of the es-timated parameters between the two independent runs using Tracer,we pooled all post “burn-in” trees from the two runs. Posterior probabil-ities of phylogenies and their internal branches were estimated on thebasis of these pooled trees.

Please cite this article as: Song, H.Y., et al., Mitogenomic circumscription(Teleostei), Gene (2014), http://dx.doi.org/10.1016/j.gene.2014.03.040

Partitioned maximum-likelihood (ML) analyses were performedwith RAxML 7.2.6 (Stamatakis, 2006), a program implementing anovel, rapid-hill-climbing algorithm. For each data set, a rapid bootstrapanalysis (-f a) and search for the best-scoringML treewere conducted inone single program run, with the GTR+ I + Γmodel (Yang, 1994). Therapid bootstrap analyses were conducted with 1000 replications, withfour threads running in parallel. The program finally conducted ML op-timization for every 5th bootstrapped tree to search for the best-scoringML tree.

2.5. Testing alternative phylogenetic hypotheses

Alternative tree topologies were compared to our best-scoring MLtree (from RY-coding data set) using the likelihood-based AU test(Shimodaira, 2002).We first created the constraint topologies consider-ing monophyly of alternative hypotheses using MacClade and conduct-ed ML analyses using RAxML with those constraints. The resultingconstrained ML trees were used to compute the per-site log likelihoodsusing RAxML (-f g option) and outputs were subjected to AU tests usingCONSEL (Shimodaira and Hasegawa, 2001). A value of p b 0. 001 wasconsidered significantly different.

3. Results and discussion

3.1. Mitochondrial genome organization

The whole mitogenome sequences of the 16 species newly deter-mined here were registered in the DDBJ/EMBL/GenBank (Table 3). Thegenome contents and gene arrangements were identical to those oftypical vertebrates, except for Hoplolatilus cuniculus (Malacanthidae).Details of the gene contents found in the species will be discussedelsewhere.

3.2. Major clades within Percomorpha and sister group of Clade D

Partitioned BA and ML analyses based on the three data sets yieldeda total of 6 tree topologies. Although we were unable to determinewhich data set recovered the most likely phylogeny, we consideredthat the results from the 12n3rRTn data set (third codon positionswere RY-coded) represented the best estimate of phylogeny, becausesuch re-coding is expected to remove the likely noise from quickly sat-urated transitional changes in the third positions, and to avoid a lack ofsignal by retaining all available positions in the data set (for details, seediscussions in Saitoh et al., 2006). The resultant tree from the 12n3rRTndata set derived from the partitioned BA analysis is shown in Fig. 2, withstatistical support [posterior probabilities (PPs) from the BA analysisand bootstrap probabilities (BPs) from the ML analysis] indicated oneach internal branch.

Although topological differences were found between the resultsfrom the two methods (see arrowheads in Fig. 2) together with thoseamong three different data sets (data not shown), eight major clades[Clades A–H in Fig. 2; names of the clades followed Kawahara et al.(2008) and Yagishita et al. (2009) with some expansions] were consis-tently recovered with high statistical support (PPs = 100%, BPs ≥ 95%)in all of the six topologies. As noted in Kawahara et al. (2008), cladescorresponding to these eight major clades (excluding the newly addedspecies) have been repeatedly recovered in the previous mitogenomicstudies, despite differences in taxonomic coverage (Mabuchi et al.,2007; Miya et al., 2003; Setiamarga et al., 2008; Yamanoue et al.,2007). Except for the Clade H, they have been almost completely recov-ered also in the latest nuclear gene studies (see Table 1).

The taxonomic contents of each clade examined here were asfollows: Clade A (Synbranchiformes and Indostomidae); Clade B(Mugiliformes, Blenniidae, Gobiesocidae, and “Atherinomorpha”) =“Ovalentaria” in Wainwright et al. (2012); Clade C (Gobioidei);Clade D (mainly Syngnathoidei; for details, see below); Clade E

of a novel percomorph fish clade mainly comprising “Syngnathoidei”

4 H.Y. Song et al. / Gene xxx (2014) xxx–xxx

Please cite this article as: Song, H.Y., et al., Mitogenomic circumscription of a novel percomorph fish clade mainly comprising “Syngnathoidei”(Teleostei), Gene (2014), http://dx.doi.org/10.1016/j.gene.2014.03.040

Table 2Comparisons of previous molecular phylogenetic studies that sample syngnathoids and related taxa.

a) b) c) d) e) f) g) h) i) j) k) l)

Chenet al.(2003)

Smith andWheeler(2004)

Dettai andLecointre(2005)

Smith andWheeler(2006)

Smith andCraig(2007)

Kawaharaet al.(2008)

Dettai andLecointre(2008)

Li et al.(2009)

Nearet al.(2012)

Betancur-Ret al. (2013)

Nearet al.(2013)

Thisstudy

Molecular markers nc/mt nc/mt nc/mt nc/mt nc/mt wmt nc nc nc nc/mt nc wmtNucleotide lengths(bp)

2420 3425 3021 4721 4036 13,832 713 3405 7587 20,853 8577 13,872

Monophyly or not(except Creediidae)

No No No No No Yes No No Yes Yes Yes Yes

SyngnathoideiPegasidae – • – • – • – – – – • •

Solenostomidae – – – – – • – – – – – •

Syngnathidae – – • – – • • • • • • •

Centriscidae – – – • – • – • • • • •

Macroramphosidae • – • – – • • • • – • •

Fistulariidae • – • – – • – • • • • •

Aulostomidae • • • • • • • • – • • •

DactylopteroideiDactylopteridae • • • • • • • • – • • •

PercoideiMullidae • – • • • – • • – • • •

CallionymoideiCallionymidae • – • • – – • • • • • •

Draconettidae – – – • – – – – – – – •

a)–l) correspond to those molecular hypotheses shown in Fig. 1. Dots and dashes represent sampled and unsampled families, respectively, in those studies.

5H.Y. Song et al. / Gene xxx (2014) xxx–xxx

(Pleuronectiformes and Carangidae); Clade F [Scombridae, Stromateoidei(Nomeidae) and Arripidae] = “Pelagia” in Miya et al. (2013); Clade G[Scorpaeniformes (in part), Percidae, Trichodontidae, Zoarcoidei(Zoarcidae and Pholidae) and Gasterosteoidei (not includingIndostomidae)], and Clade H (Tetraodontiformes, Lophiiformes,Caproidae, Sparidae, Lutjanidae, Emmelichthyidae, and Malacanthidae).Correspondence between thesemitogenomic clades and the recently rec-ognized clades based on nuclear genes was shown in Table 1.

In most of the previous mitogenomic studies, Clade C (Gobioidei,mainly including benthic species) formed a sister-group relationshipwith CladeD (Dactylopteridae or Syngnathoidei+ Dactylopteridae), al-though statistical support was not high (BPs ≤ 64%; Kawahara et al.,2008; Yagishita et al., 2009). In the latest mitogenomic study of Miyaet al. (2013), however, “Pelagia” corresponding to our Clade F was thesister to the clade corresponding to our Clade D with not so high statis-tical supports (BPs≤ 75%). In the recent studies based on concatenatedmultiple nuclear gene sequences based on extensive taxon sampling[Near et al., 2012, 2013 (Fig. 1i, k); Wainwright et al., 2012 (notshown); Betancur-R et al., 2013 (Fig. 1j)], the clade including themembers of our Clade D was recovered as a sister-group of a clademainly comprising suborder Scombroidei corresponding to our CladeF (=“Pelagia”), which was supported by high statistical values (BPs N

90%). In the present mitogenomic analyses, a sister-group of Clade Dwas ambiguous, depending on both data sets and analyses. For example,Clade C (Gobioidei) was placed as a sister-group of the Clade D in threeof the six topologies (BA of 12n3rRTn and 123nRTn, and ML of 123nRTn:the result of the first analysis is shown in Fig. 2), while Clade F(Scombridae, Stromateoidei, Arripidae) was placed as a sister-group ofClade D in the remaining three analyses. Further analysis with morecomprehensive taxonomic and character sampling would be neededto confirm a sister-group of Clade D.

Fig. 1.Molecular phylogenetic hypotheses for the “CladeD” fishes from the previous (a–k) and pasterisks for Syngnathoidei, solid circles for Dactylopteridae, triangles for Mullidae, and solid s

Please cite this article as: Song, H.Y., et al., Mitogenomic circumscription(Teleostei), Gene (2014), http://dx.doi.org/10.1016/j.gene.2014.03.040

3.3. Monophyly of Clade D and alternative hypotheses

In the present analyses, all species from 7 families of Syngnathoideiwere confidently placed in Clade D, together with those of the familiesDactylopteridae, Mullidae, and two callionymoid families(Callionymidae and Draconettidae) (Fig. 2). Although the monophylyof most of these families among percomorphs has been presented inthe two recent nuclear studies (Betancur-R et al., 2013; Near et al.,2013; see Fig. 1j, k), that of all the families (Fig. 1l) has never been pre-sented before. The present result is the first solid support for theirmonophyly based on mitogenomic data.

The suborder Syngnathoidei, the main member of the Clade D, hastraditionally been placed in the order Gasterosteiformes, together withthe suborder Gasterosteoidei (sticklebacks, including Indostomidae). Asin the previous studies (e. g., Kawahara et al., 2008; see Fig. 1f), our resultsagain recovered the polyphyly of the order [Syngnathoidei in Clade D,Indostomidae in Clade A and Gasterosteoidei (minus Indostomidae) inClade G] within the Percomorpha. Monophylies of these three lineages(Syngnathoidei, Indostomidae and Gasterosteoidei) and any two ofthese three were confidently rejected by statistical comparisons (AUtests, p b 0. 001; Table 4).

The family Dactylopteridae has traditionally been placed in the orderScorpaeniformes (Nelson, 2006). Taxonomic treatment of this enigmaticfamily, however, has been controversial. Based on morphological analy-ses, it was once thought to be related to the syngnathoids (Pietsch,1978), but it was later placed in the Scorpaeniformes followingWashington et al. (1984) and Eschmeyer et al. (1990). Johnsonand Patterson (1993) rebutted these hypotheses, and established anew order Dactylopteriformes to include only Dactylopteridae.Based on 20 putative synapomorphies, Imamura (2000) proposed thatthe Dactylopteridae was nested within a percoid family Malacanthidae.

resent (l) studies. Classification followsNelson (2006). The CladeD families are denoted byquares for Callionymoidei. The family Creediidae is denoted by arrows.

of a novel percomorph fish clade mainly comprising “Syngnathoidei”

Table 3List of species examined with DDBJ/EMBL/Genbank accession numbers and references. Classification follows Nelson (2006).

Order Family Species Accession number References

Polymixiiformes Polymixiidae Polymixia japonica AB034826 Miya and Nishida (2000)Beryciformes Berycidae Beryx spelendens AP002939 Miya et al. (2001)Ophidiiformes Ophidiidae Bassozetus zenkevitchi AP004405 Miya et al. (2003)

Bythitidae Diplacanthopoma brachysoma AP004408 Miya et al. (2003)Lophiiformes Lophiidae Lophius americanus AP004414 Miya et al. (2003)

Lophius litulon AP004413 Miya et al. (2003)Chaunacidae Chaunax tosaensis AP004416 Miya et al. (2003)

Chaunax abei AP004415 Miya et al. (2003)Caulophrynidae Caulophryne pelagica AP004417 Miya et al. (2003)Melanocetidae Melanocetus murrayi AP004418 Miya et al. (2003)

Mugiliformes Mugilidae Crenimugil crenilabis AP002931 Miya et al. (2001)Mugil cephalus AP002930 Miya et al. (2001)

Atheriniformes Melanotaeniidae Melanotaenia lacustris AP004419 Miya et al. (2001)Atherinidae Hypoatherina tsurugae AP004420 Miya et al. (2003)

Beloniformes Adrianichthyidae Oryzias latipes AP004421 Miya et al. (2003)Scomberesocidae Cololabis saira AP002932 Miya et al. (2001)Exocoetidae Exocoetus volitans AP002933 Miya et al. (2001)

Cyprinodontiformes Aplocheilidae Kryptolebias marmoratus AF283503 Lee et al. (2001)Poeciliidae Gambusia affinis AP004422 Miya et al. (2003)

Gasterosteiformes Hypoptychidae Hypoptychus dybowskii AP004437 Miya et al. (2003)Aulorhynchidae Aulorhynchus flavidus AP009196 Kawahara et al. (2008)Gasterosteidae Gasterosteus aculeatus AP002944 Miya et al. (2001)Indostomidae Indostomus paradoxus AP004438 Miya et al. (2003)Pegasidae Eurypegasus draconis AP005983 Kawahara et al. (2008)

Pegasus volitans AP005984 Kawahara et al. (2008)Solenostomidae Solenostomus cyanopterus AB277725 Kawahara et al. (2008)

Solenostomus paradoxus AP012308 This studySyngnathidae Hippocampus kuda AP005985 Kawahara et al. (2008)

Microphis brachyurus AP005986 Kawahara et al. (2008)Doryrhamphus japonicus AP012309 This studyPhycodurus eques AP012313 This studySyngnathus schlegeli AP012318 This study

Aulostomidae Aulostomus chinensis AP009197 Kawahara et al. (2008)Fistulariidae Fistularia commersonii AP005987 Kawahara et al. (2008)

Fistularia petimba AP012312 This studyMacroramphosidae Macroramphosus scolopax AP005988 Kawahara et al. (2008)Centriscidae Aeoliscus strigatus AP009198 Kawahara et al. (2008)

Synbranchiformes Synbranchidae Monopterus albus AP002945 Miya et al. (2001)Mastacembelidae Mastacembelus favus AP002946 Miya et al. (2001)

Scorpaeniformes Dactylopteridae Dactyloptena tiltoni AP004440 Miya et al. (2003)Dactyloptena peterseni AP002947 Miya et al. (2001)Dactyloptena orientalis AP012311 This study

Scorpaenidae Helicolenus hilgendorfii AP002948 Miya et al. (2001)Sebastes schlegeli AY491978 Kim and Lee (2004)

Triglidae Satyrichthys amiscus AP004441 Miya et al. (2003)Cottidae Cottus reinii AP004442 Miya et al. (2003)Cyclopteridae Aptocyclus ventricosus AP004443 Miya et al. (2003)

Perciformes Percidae Etheostoma radiosum AY341348 Broughton and Reneau (2006)Carangidae Caranx melampygus AP004445 Miya et al. (2003)

Trachurus japonicus AP003091 Mabuchi et al. (2007)Carangoides armatus AP004444 Miya et al. (2003)Trachurus trachurus AB108498 Takashima et al. (2006)

Emmelichthyidae Emmelichthys struhsakeri AP004446 Miya et al. (2003)Lutjanidae Pterocaesio tile AP004447 Miya et al. (2003)Sparidae Pagrus major AP002949 Miya et al. (2001)

Pagrus auriga AB124801 UnpublishedMalacanthidae Hoplolatilus cuniculus AP012490 This study

Branchiostegus japonicus EU861052 Oh et al. (2010)Mullidae Mulloidichthys vanicolensis AP012310 This study

Parupeneus multifasciatus AP012314 This studyUpeneus japonicus AB355921 This studyUpeneus tragula AB355920 This study

Arripidae Arripis trutta AP006810 Yagishita et al. (2009)Zoarcidae Lycodes toyamensis AP004448 Miya et al. (2003)Pholidae Enedrias crassispina AP004449 Miya et al. (2003)Trichodontidae Arctoscopus japonicus AP003090 Mabuchi et al. (2007)Blenniidae Petroscirtes breviceps AP004450 Miya et al. (2003)

Salarias fasciatus AP004451 Miya et al. (2003)Callionymidae Synchiropus altivelis AP006027 This study

Neosynchiropus moyeri AP012315 This studySynchiropus splendidus AP012317 This studyCallionymus curvicornis AP012307 Song et al. (2012)Callionymus enneactis AP012316 This study

Draconettidae Draconetta xenica AP006028 This studyGobiesocidae Aspasma minima AP004453 Miya et al. (2003)

Arcos sp. AP004452 Miya et al. (2003)

6 H.Y. Song et al. / Gene xxx (2014) xxx–xxx

Please cite this article as: Song, H.Y., et al., Mitogenomic circumscription of a novel percomorph fish clade mainly comprising “Syngnathoidei”(Teleostei), Gene (2014), http://dx.doi.org/10.1016/j.gene.2014.03.040

Table 3 (continued)

Order Family Species Accession number References

Rhyacichthyidae Rhyacichthys aspro AP004454 Miya et al. (2003)Eleotridae Eleotris acanthopoma AP004455 Miya et al. (2003)Gobiidae Acanthogobius hasta AY486321 Kim et al. (2004)Scombridae Thunnus thynnus AY302574 Broughton and Reneau (2006)

Euthynnus alletteratus AB099716 UnpublishedAuxis rochei AB103467 UnpublishedAuxis thazard AB105447 UnpublishedKatsuwonus pelamis AB101290 UnpublishedThunnus alalunga AB101291 UnpublishedScomber scombrus AB120717 Takashima et al. (2006)

Nomeidae Psenes cyanophrys AP011067 Yagishita et al. (2009)Cubiceps pauciradiatus AP006038 Yagishita et al. (2009)

Caproidae Antigonia capros AP002943 Miya et al. (2001)Pleuronectiformes Paralichthyidae Paralichthys olivaceus AB028664 Saitoh et al. (2000)

Pleuronectidae Platichthys bicoloratus AP002951 Miya et al. (2001)Tetraodontiformes Balistidae Sufflamen fraenatus AP004456 Miya et al. (2003)

Monacanthidae Stephanolepis cirrhifer AP002952 Miya et al. (2001)Tetraodontidae Takifugu rubripes AP006045 Yamanoue et al. (2006)Molidae Mola mola AP006238 Yamanoue et al. (2004)

Masturus lanceolatus AP006239 Yamanoue et al. (2004)

7H.Y. Song et al. / Gene xxx (2014) xxx–xxx

In the present mitogenomic study, the Dactylopteridae was confi-dently placed in Clade D, while other members of Scorpaeniformesand Malacanthidae were placed in Clades G and H, respectively.Monophylies of those lineages (Dactylopteridae +Clade G andDactylopteridae + Malacanthidae) were statistically rejected byAU tests (Table 4). In the latest two nuclear studies (Betancur-Ret al., 2013; Near et al., 2013; see Fig. 1j, k), the Dactylopteridaeformed a robust monophyletic group with the other members ofour Clade D.

The phylogenetic position of theMullidae has not been examined sat-isfactorily (Kim, 2002). Some implied the close affinity with Lutjanidae(Regan, 1913), while others suggested Sparidae (Boulenger, 1904;Gosline, 1984). In the present analyses, Lutjanidae and Sparidae wereboth placed in Clade H. When monophyly was constrained on either ofthe two lineages (Mullidae + Lutjanidae and Mullidae + Sparidae), theresulting ML trees were confidently rejected by AU tests (Table 4). Inthe latest nuclear studies (Betancur-R et al., 2013; Near et al., 2013),Lutjanidae and Sparidae were both placed distantly from Mullidae,which formed a robust monophyletic group with the other members ofour Clade D.

The phylogenetic position of the Callionymoidei has also been enig-matic, but its close affinity with Gobiesocidae has been proposed by Gill(1996) and Springer and Johnson (2004) based on morphological anal-yses, also by Gosline (1970), Winterbottom (1993), and Johnson andPatterson (1993). Springer and Orrell (2004), based on morphologicalanalyses, proposed that the Callionymoidei (Draconettidae) was sisterto the clade of Gobiesocidae + Blennioidei. Based on mitogenomicsequences from 100 higher teleosts, Miya et al. (2003) demonstratedthat members of the Gobiesocidae were closely related to theBlennioidei within an expanded Atherinomorpha. The latter clade hasbeen reconfirmed by Setiamarga et al. (2008) and subsequently ex-panded and named “Ovalentaria” inWainwright et al. (2012). The latestnuclear gene studies (Betancur-R et al., 2013; Near et al., 2013) recov-ered the “Ovalentaria” (=our Clade B) being distantly placed from thespecies of Callionymoidei. Using our mitogenomic data, the monophylyof Gobiesocoidei (Clade B) plus Callionymoidei (Clade D) was confi-dently rejected by AU test (Table 4).

Greenfield et al. (2008) suggested close relationships between theCallionymoidei and Batrachoididae based on morphological analyses.Mitogenomic study by Miya et al. (2005), however, demonstrated thatthe batrachoidid species were nested within a clade comprising twospecies of the Synbranchiformes plus a species of Indostomus (corre-sponding to our Clade A), while two dactylopterids (the only represen-tatives from our Clade D) were placed far from the clade. In the recentnuclear gene studies (Betancur-R et al., 2013; Near et al., 2012, 2013;

Please cite this article as: Song, H.Y., et al., Mitogenomic circumscription(Teleostei), Gene (2014), http://dx.doi.org/10.1016/j.gene.2014.03.040

Fig. 1i, j, k), “Batracoidiformes” were placed at a basal position ofPercomorpha, far from the members of our Clade D.

In the nuclear gene tree of Betancur-R et al. (2013) (Fig. 1j), theclade comprising the members of our Clade D was named as“Syngnathiformes”, and it included the family Creediidae (indicatedby arrows in Fig. 1). This family was, however, distantly locatedfrom the members of our Clade D in the nuclear gene trees ofSmith and Wheeler (2006) and Near et al. (2013) (Fig. 1d, l). Thepresent mitogenomic study did not include the family. Apparentlya phylogenetic position of the Creediidae needs further studiesbased on multiple markers.

3.4. Families included in the Clade D

Seven syngnathoid families, Callionymidae, Mullidae, andDactylopteridae were respectively recovered as monophyleticgroups with 100% PPs and BPs (note that Macroramphosidae,Centriscidae, Aulostomidae, and Draconettidae were each repre-sented by a single species in this study). Interrelationships amongthese families, however, remained ambiguous with the exceptionof monophyly of Macroramphosidae plus Centriscidae (PPs =100%, BPs ≥ 97%; Fig. 2). Most of the inter-familial relationshipswere connected by extremely short internal branches and the resultingnodal support was mostly b50% across the data sets and analyses. Thesituation remains unchanged when two long-branches from theSolenostomidae and Aulostomidae were removed from the analy-ses. Among the recent nuclear studies (Betancur-R et al., 2013;Near et al., 2012, 2013; Fig. 1i, j, k), there were many disagreementsin the interrelationships among the family members of our Clade D.Thus more extensive taxon and character sampling would be neces-sary to obtain a more robust picture of their phylogenies.

All of the present Clade D families include benthic or semi-benthicspecies, with most of the members have more or less small tubularmouths (except Dactylopteridae and Mullidae) and armored bodieswith bony plates (except Mullidae and Callionymoidei). The fishes ofthe suborder Syngnathoidei (264 species placed in seven families, 62genera: Nelson, 2006), themainmembers of the clade, exhibit such dis-tinct appearancewith unorthodox locomotion and unique reproductivemode. For example, shrimpfishes (the family Centriscidae) swim in a ver-tical position with head-down, seahorses (the subfamily Hippocampinaein the family Syngnathidae) do sowith raise head, and trumpetfishes (thefamily Aulostomidae) often swim alongside larger fish with head tilteddown (Nelson, 2006). Seamoths (the family Pegasidae) “walk” over thebottom, as flying gurnards (the family Dactylopteridae in the orderScorpaeniformes) do. It is noteworthy that pipefishes and seahorses

of a novel percomorph fish clade mainly comprising “Syngnathoidei”

Fig. 2. Phylogenetic tree from partitioned Bayesian analysis based on 12n3rRTn data set. Topological incongruities between analyses (partitioned Bayesian and ML analyses) denoted byarrowheads. Numbers beside internal branches indicate Bayesian posterior probabilities (PPs) andML bootstrap probabilities (BPs) from 1000 replicates, respectively (shown as percent-ages). “–” indicates node not recovered inML analysis with N50% BPs. Thick branches supported by 100% PPs and BPs. The branch-lengths of the syngnathoid families of the Aulostomidaeand Solenostomidae are shortened to a half. The Clade D families are denoted by asterisks for Syngnathoidei, solid circles for Dactylopteridae, triangles for Mullidae, and solid squares forCallionymoidei, as in Fig. 1.

8 H.Y. Song et al. / Gene xxx (2014) xxx–xxx

Please cite this article as: Song, H.Y., et al., Mitogenomic circumscription of a novel percomorph fish clade mainly comprising “Syngnathoidei”(Teleostei), Gene (2014), http://dx.doi.org/10.1016/j.gene.2014.03.040

Table 4Statistical likelihood-based AU tests between unconstrained and constrained trees.

Alternative hypotheses tested Likelihood-based AU test Remarks

−In L Δ In L

Unconstrained 360,784.5671 (Best)Constrained

Syngnathoidei + Gasterosteoidei 361,422.1785 637.6114* Gasterosteiformes (sensu Nelson, 2006)Syngnathoidei + “Gasterosteoidei” 361,342.2973 557.7302* Gasterosteiformes minus IndostomidaeGasterosteoidei (including Indostomidae) 361,006.4981 221.9310* Gasterosteoidei (sensu Nelson, 2006)Syngnathoidei + Indostomidae 360,961.9104 177. 3433*Dactylopteridae + Clade G 360,936.1333 151.5663* Dactylopterid's affinity to other ScorpaeniformesDactylopteridae + Malacanthidae 360,982.7323 198.1653* Imamura (2000)Mullidae + Lutjanidae 361,103.4772 318.9101* Regan (1913)Mullidae + Sparidae 360,993.6338 209.0667* Boulenger (1904) and Gosline (1984)Callionymoidei + Gobiesocoidei 361,001.6728 217.1057* Gill (1996) and Springer and Johnson (2004)

Statistically significant differences p b 0. 001 are denoted by asterisks.

9H.Y. Song et al. / Gene xxx (2014) xxx–xxx

(the family Syngnathidae) are the only vertebrates in which the maleliterally becomes “pregnant” (Helfman et al., 2009). Although interrela-tionships among the families of the Clade D remained unclear, theirmonophyly, which was supported by the present mitogenomic and re-cent nuclear data with high statistical support, will provide a new basisfor the understanding of the evolution of these interesting fishes.

From a taxonomic standpoint, morphological synapomorphies forthis newly defined clade will be expected; however, the search of suchcharacters is not straightforward, because of the wide variety of theincluded taxa formerly distributed across four different percomorphsuborders. Apparently themorphological definition of the new clade re-quires careful studies bymorphologists as that found in a recent study ofa Lophiiformes/Tetraodontiformes sister group relationship (Chanetet al., 2013).

3.5. Name of the Clade D

We propose to call this interesting clade as “Syngnathiformes” fol-lowing the latest nuclear studies with some revisions on the includedfamilies. The name “Syngnathiformes” was used in Near et al. (2012,2013) and Betancur-R et al. (2013)with the familymembers somewhatdifferent from each other. The analyses of Near et al. (2012) did not in-clude the families Dactylopteridae and Mullidae. Near et al. (2013) in-cluded the two families in their analyses, but their “Syngnathiformes”did not include the Mullidae, although it formed a monophyleticgroup together with a clade including other members of our Clade Dand a clade corresponding to our Clade F (Fig. 1k). We prefer that“Syngnathiformes” include the Mullidae based on the results both ofthe present study and Betancur-R et al. (2013) (Fig. 1j, l). As mentionedabove, The “Syngnathiformes” of Betancur-R et al. (2013) include theCreediidae (Fig. 1j), but inclusion of this family needs further studies.

Conflict of interest

The authors declare no conflicts of interest.

Acknowledgments

Wewould like to thankKeiichiMatsuura andGento Shinohara at theNational Museum of Nature and Science (Japan) for providing somematerials.We also thank Ryouka Kawahara and Naoki Yagishita for pro-viding the sequences used in Kawahara et al. (2008) and Yagishita et al.(2009), and Jun G. Inoue for valuable discussion and suggestions. Thisstudy was supported by Grants-in-Aid from the Ministry of Education,Culture, Sports, Science and Technology (15380131, 17207007,19207007, 22580229 and 23370041). This is Harbor Branch Oceano-graphic Institution contribution no. 1917.

Please cite this article as: Song, H.Y., et al., Mitogenomic circumscription(Teleostei), Gene (2014), http://dx.doi.org/10.1016/j.gene.2014.03.040

References

Azuma, Y., Kumazawa, Y., Miya, M., Mabuchi, K., Nishida, M., 2008. Mitogenomic evalua-tion of the historical biogeography of cichlids toward reliable dating of teleostean di-vergences. BMC Evolutionary Biology 8, 215.

Betancur-R, R., Broughton, R.E., Wiley, E.O., Carpenter, K., López, J.A., Li, C., Holcroft, N.I.,Arcila, D., Sanciangco, M., Cureton II, J.C., Zhang, F., Buser, T., Campbell, M.A.,Ballesteros, J.A., Roa-Varon, A., Willis, S., Borden, W.C., Rowley, T., Reneau, P.C.,Hough, D.J., Lu, G., Grande, T., Arratia, G., Ortí, G., 2013. The tree of life and a new clas-sification of bony fishes. PLOS Currents 18, 1–45.

Boulenger, G.A., 1904. Fishes: systematic account of Teleostei. The Cambridge Natural His-tory, vol. VII. Macmillan, London 541–727 (Figs 325–440).

Broughton, R., Reneau, P., 2006. Spatial covariation of mutation and nonsynonymous sub-stitution rates in vertebrate mitochondrial genomes. Molecular Biology and Evolution23, 1516–1524.

Chanet, B., Guintard, C., Betti, E., Gallut, C., Dettai, A., Lecointre, G., 2013. Evidence for aclose phylogenetic relationship between the teleost orders Tetraodontiformes andLophiiformes based on an analysis of soft anatomy. Cybium 37, 179–198.

Chen, W.J., Bonillo, C., Lecointre, G., 2003. Repeatability of clades as a criterion of reliabil-ity: a case study for molecular phylogeny of Acanthomorpha (Teleostei) with largernumber of taxa. Molecular Phylogenetics and Evolution 26, 262–288.

Cheng, S., Higuchi, R., Stoneking,M., 1994. Completemitochondrial genome amplification.Nature Genetics 7, 350–351.

Dettai, A., Lecointre, G., 2005. Further support for the clades obtained by multiple molec-ular phylogenies in the acanthomorph bush. Comptes Rendus Biologies 328,674–689.

Dettai, A., Lecointre, G., 2008. New insights into the organization and evolution of verte-brate IRBP genes and utility of IRBP gene sequences for the phylogenetic study of theAcanthomorpha (Actinopterygii: Teleostei). Molecular Phylogenetics and Evolution48, 258–269.

Eschmeyer, W.N., Bailey, R.M., other collaborators, 1990. Catalog of the Genera of RecentFishes. California Academy of Sciences, San Francisco (697 pp.).

Gill, A.C., 1996. Comments on an intercalar path for the glossopharyngeal (cranial IX)nerve as a synapomorphy of the Paracanthopterygii and on the phylogenetic positionof the Gobiesocidae (Teleostei: Acanthomorpha). Copeia 1996, 1022–1029.

Gosline, W.A., 1970. A reinterpretation of the teleostean fish order Gobiesociformes. Pro-ceedings of the California Academy of Sciences 38, 363–382.

Gosline, W.A., 1984. Structure, function, and ecology of the goatfishes (family Mullidae).Pacific Science 38, 312–323.

Greenfield, D.W., Winterbottom, R., Collette, B.B., 2008. Review of the toadfish genera(Teleostei: Batrachoididae). Proceedings of the California Academy of Sciences 59,665–710.

Helfman, G.F., Collette, B.B., Facey, D.E., Bowen, B.W., 2009. The Diversity of Fishes: Biolo-gy, Evolution and Ecology, Second edition. Wiley-Blackwell, West Sussex, UnitedKingdom.

Imamura, H., 2000. An alternative hypothesis on the phylogenetic position of the familyDactylopteridae (Pisces: Teleostei), with a proposed new classification. IchthyologicalResearch 47, 203–222.

Inoue, J.G., Miya, M., Tsukamoto, K., Nishida, M., 2001. Complete mitochondrial DNA se-quence of the Japanese anchovy Engraulis japonicus. Fisheries Science 67, 828–835.

Inoue, J.G., Miya, M., Tsukamoto, K., Nishida, M., 2004. Mitogenomic evidence for themonophyly of elopomorph fishes (Teleostei) and the evolutionary origin of the lepto-cephalus larva. Molecular Phylogenetics and Evolution 30, 274–286.

Johnson, G.D., Patterson, C., 1993. Percomorph phylogeny: a survey of acanthomorphsand a new proposal. Bulletin of Marine Science 52, 554–626.

Katoh, K., Toh, H., 2008. Recent developments in the MAFFTmultiple sequence alignmentprogram. Briefings in Bioinformatics 9, 286–298.

Kawaguchi, A., Miya, M., Nishida, M., 2001. Complete mitochondrial DNA sequence ofAulopus japonicus (Teleostei: Aulopiformes), a basal Eurypterygii: longer DNA se-quences and higher-level relationships. Ichthyological Research 48, 213–223.

Kawahara, R., Miya, M., Mabuchi, K., Lavoué, S., Inoue, J.G., Satoh, T.P., Kawaguchi, A.,Nishida, M., 2008. Interrelationships of the 11 gasterosteiform families (sticklebacks,pipefishes, and their relatives): a new perspective based on whole mitogenomesequences from 75 higher teleosts. Molecular Phylogenetics and Evolution 46,224–236.

of a novel percomorph fish clade mainly comprising “Syngnathoidei”

10 H.Y. Song et al. / Gene xxx (2014) xxx–xxx

Kim, B.J., 2002. Comparative anatomy and phylogeny of the family Mullidae (Teleostei:Perciformes). Memoirs of the Graduate School of Fisheries Sciences, Hokkaido Uni-versity 49, 1–74.

Kim, I.C., Lee, J.S., 2004. The complete mitochondrial genome of the rockfish Sebastesschlegeli (Scorpaeniformes, Scorpaenidae). Molecules and Cells 17, 322–328.

Kim, I.C., Kweon, H.S., Kim, Y.J., Kim, C.B., Gye, M.C., Lee, W.O., Lee, Y.S., Lee, J.S., 2004. Thecomplete mitochondrial genome of the javeline goby Acanthogobius hasta(Perciformes, Gobiidae) and phylogenetic considerations. Gene 336, 147–153.

Kocher, T.D., Stepien, C.A., 1997. Molecular Systematics of Fishes. Academic Press, SanDiego, California.

Lee, J.S., Miya, M., Lee, Y.S., Kim, C.G., Park, E.H., Aoki, Y., Nishida, M., 2001. The completeDNA sequence of the mitochondrial genome of the self fertilizing fish Rivulusmarmoratus (Cyprinodontiformes, Rivulidae) and the first description of duplicationof a control region in fish. Gene 280, 1–7.

Li, B., Dettai, A., Cruaud, C., Couloux, A., Desoutter-Meniger, M., Lecointre, G., 2009.RNF213, a new nuclear marker for acanthomorph phylogeny. Molecular Phyloge-netics and Evolution 50, 345–363.

Löytynoja, A., Millinkovitch, M.C., 2003. A hidden Markov model for progressive multiplealignment. Bioinformatics 19, 1505–1513.

Mabuchi, K., Miya, M., Azuma, Y., Nishida, M., 2007. Independent evolution of the special-ized pharyngeal jaw apparatus in cichlid and labrid fishes. BMC Evolutionary Biology7, 10.

Maddison, W.P., Maddison, D.R., 2000. MacClade, Version 4. Sinauer Associates, Sunderland,Massachusetts.

Miya, M., Nishida, M., 1999. Organization of the mitochondrial genome of a deep-sea fish,Gonostoma gracile (Teleostei: Stomiiformes): first example of transfer RNA gene rear-rangements in bony fishes. Marine Biotechnology 1, 416–426.

Miya, M., Nishida, M., 2000. Use ofmitogenomic information in teleosteanmolecular phy-logenies: a tree based exploration under the maximum-parsimony optimality criteri-on. Molecular Phylogenetics and Evolution 17, 437–455.

Miya, M., Kawaguchi, A., Nishida, M., 2001. Mitogenetic exploration of higher teleosteanphylogenies: a case study for moderate-scale evolutionary genomics with 38 newlydetermined complete mitochondrial DNA sequences. Molecular Phylogenetics andEvolution 18, 1993–2009.

Miya, M., Takeshima, H., Endo, H., Ishiguro, N.B., Inoue, J.G., Mukai, T., Satoh, T.P.,Yamaguchi, M., Kawaguchi, A., Mabuchi, K., Shirai, S.M., Nishida, M., 2003. Major pat-terns of higher teleostean phylogenies: a new perspective based on 100 complete mi-tochondrial DNA sequences. Molecular Phylogenetics and Evolution 26, 21–138.

Miya, M., Satoh, T.P., Nishida, M., 2005. The phylogenetic position of toadfishes (orderBatrachoidiformes) in the higher ray-finned fish as inferred from partitioned Bayes-ian analysis of 102 whole mitochondrial genome sequences. Biological Journal ofthe Linnean Society 85, 289–306.

Miya, M., Friedman, M., Satoh, T.P., Takeshima, H., Sado, T., Iwasaki, W., Yamanoue, Y.,Nakatani, M., Mabuchi, K., Inoue, J.G., Poulsen, J.Y., Fukunaga, T., Sato, Y., Nishida, M., 2013. Evolutionary origin of the Scombridae (tunas and mackerels): members of aPaleogene adaptive radiation with 14 other pelagic fish families. PloS One 8 (9),e73535.

Near, T.J., Eytan, R.I., Dornburg, A., Kuhn, K.L., Moore, J.A., Davis, M.P., Wainwright, P.C.,Friedman, M., Smith, W.L., 2012. Resolution of ray-finned fish phylogeny and timingof diversification. Proceedings of the National Academy of Sciences of the UnitedStates of America 109, 13698–13703.

Near, T.J., Dornburg, A., Eytan, R.I., Keck, B.P., Smith,W.L., Kuhn, K.L., Moore, J.A., Price, S.A.,Burbrink, F.T., Friedman, M., Wainwright, P.C., 2013. Phylogeny and tempo of diversi-fication in the superradiation of spiny-rayed fishes. Proceedings of the National Acad-emy of Sciences of the United States of America 101, 12738–21743.

Nelson, G., 1989. Phylogeny of major fish groups. In: Fernholm, B., Bremer, K., Jörnvall, H.(Eds.), The Hierarchy of Life. Excerpta Medica, Amsterdam, pp. 325–336.

Nelson, J.S., 2006. Fishes of the World, fourth ed. Wiley, New York.Oh, D.J., Oh, B.S., Jung, M.M., Jung, Y.H., 2010. Complete mitochondrial genome of three

Branchiostegus (Perciformes, Malacanthidae) species: genome description and phylo-genetic considerations. Mitochondrial DNA 21, 151–159.

Phillips, M.J., Penny, D., 2003. The root of the mammalian tree inferred from whole mito-chondrial genomes. Molecular Phylogenetics and Evolution 28, 171–185.

Phillips, M.J., Delsuc, F., Penny, D., 2004. Genome-scale phylogeny and the detection ofsystematic biases. Molecular Phylogenetics and Evolution 21, 1455–1458.

Pietsch, T.W., 1978. Evolutionary relationships of the sea moths (Teleostei: Pegasidae)with a classification of gasterosteiform families. Copeia 1978, 517–529.

Posada, D., Crandall, K.A., 1998. MODELTEST: testing the model of DNA substitution.Bioinformatics 14, 817–818.

Regan, C.T., 1913. The classification of the percoid fishes. Annals and Magazine of NaturalHistory 12, 111–145.

Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes 3: Bayesian phylogenetic inference undermixed models. Bioinformatics 19, 1572–1574.

Please cite this article as: Song, H.Y., et al., Mitogenomic circumscription(Teleostei), Gene (2014), http://dx.doi.org/10.1016/j.gene.2014.03.040

Saitoh, K., Hayashizaki, K., Yokoyama, Y., Asahida, T., Toyohara, H., Yamashita, Y., 2000.Complete nucleotide sequence of Japanese flounder (Paralichthys olivaceus) mito-chondrial genome: structural properties and cue for resolving teleostean relation-ships. The Journal of Heredity 91, 271–278.

Saitoh, K., Sado, T., Mayden, R.L., Hanzawa, N., Nakamura, K., Nishida, M., Miya, M., 2006.Mitogenomic evolution and interrelationships of the Cypriniformes (Actinopterygii:Ostariophysi): the first evidence toward resolution of higher-level relationships ofthe world's largest freshwater fish clade based on 59 whole mitogenome sequences.Journal of Molecular Evolution 63, 826–841.

Setiamarga, D.H.E., Miya, M., Yamanoue, Y., Mabuchi, K., Satoh, T.P., Inoue, J.G., Nishida, M.,2008. Interrelationships of Atherinomorpha (medakas, flyingfishes, killifishes, silver-sides, and their relatives): the first evidence based on whole mitogenome sequences.Molecular Phylogenetics and Evolution 49, 598–605.

Shimodaira, H., 2002. An approximately unbiased test of phylogenetic tree selection. Sys-tematic Biology 51, 492–508.

Shimodaira, H., Hasegawa,M., 2001. CONSEL: for assessing the confidence of phylogenetictree selection. Bioinformatics 17, 1246–1247.

Smith, W.L., Craig, M.T., 2007. Casting the percomorph net widely: the importance ofbroad taxonomic sampling in the search for the placement of serranid and percoidfishes. Copeia 35–55.

Smith, W.L., Wheeler, W.C., 2004. Polyphyly of the mail-cheeked fishes (Teleostei:Scorpaeniformes): evidence from mitochondrial and nuclear sequence data. Molecu-lar Phylogenetics and Evolution 32, 627–646.

Smith, W.L., Wheeler, W.C., 2006. Venom evolution widespread in fishes: a phylogeneticroad map for the bioprospecting of piscine venoms. The Journal of Heredity 97,206–217.

Song, H.Y., Satoh, T.P., Mabuchi, K., 2012. Complete mitochondrial genome sequence ofthe dragonet Callionymus curvicornis (Perciformes: Callionymoidei: Callionymidae).Mitochondrial DNA 23, 290–292.

Springer, V.G., Johnson, G.D., 2004. Study of the dorsal gill-archmusculature of teleostomefishes, with special reference to the Actinopterygii. Bulletin of the Biological Society ofWashington 11 (1–235 and plates 205).

Springer, V.G., Orrell, T.M., 2004. Phylogenetic analysis of 147 families of Acanthomorphfishes based primarily on dorsal gill-archmuscles and skeleton. Bulletin of the Biolog-ical Society of Washington 11, 237–260.

Stamatakis, A., 2006. RAxML-VI-HPC: maximum likelihood-based phylogenetic analyseswith thousands of taxa and mixed models. Bioinformatics 22, 2688–2690.

Takashima, Y., Morita, T., Yamashita, M., 2006. Complete mitochondrial DNA sequence ofAtlantic horse mackerel Trachurus trachurus andmolecular identification of two com-mercially important species T. trachurus and T. japonicus using PCR-RFLP. FisheriesScience 72, 1054–1065.

Wainwright, P.C., Smith, W.L., Price, S.A., Tang, K.L., Sparks, J.S., Ferry, L.A., Kuhn, K.L.,Eytan, R.I., Near, T.J., 2012. The evolution of pharyngognathy: a phylogenetic andfunctional appraisal of the pharyngeal jaw key innovation in Labroidei and beyond.Systematic Biology 61, 1001–1027.

Washington, B.B., Eschmeyer, W.N., Howe, K.M., 1984. Scorpaeniformes: relationships. In:Moser, H.G., Richards, W.J., Cohen, D.M., Fahay, M.P., Kendall Jr., A.W., Richardson, S.L.(Eds.), Ontogeny and Systematics of Fishes. Spec. Publ. No. 1, Amer. Soc. Ichthyol.Herp. , pp. 438–447.

Wiley, E.O., Johnson, G.D., 2010. A teleost classification based onmonophyletic groups. In:Nelson, J.S., Schultze, H.-P., Wilson, M.V.H. (Eds.), Origin and Phylogenetic Interrela-tionships of Teleosts. Verlag Dr. Friedrich Pfeil, München, Germany, pp. 123–182.

Winterbottom, R., 1993. Search for the gobioid sister group (Actinopterygii: Percomorpha).Bulletin of Marine Science 52, 395–414.

Yagishita, N., Miya, M., Yamanoue, Y., Shirai, S.M., Nakayama, K., Suzuki, N., Satoh, T.P.,Mabuchi, K., Nishida, M., Nakabo, T., 2009. Mitogenomic evaluation of the unique fa-cial nerve pattern as a phylogenetic marker within the perciform fishes (Teleostei:Percomorpha). Molecular Phylogenetics and Evolution 53, 258–266.

Yamanoue, Y., Miya, M., Matsuura, K., Katoh, M., Sakai, H., Nishida, M., 2004. Mitochondri-al genomes and phylogeny of the ocean sunfishes (Tetraodontiformes: Molidae). Ich-thyological Research 51, 269–273.

Yamanoue, Y., Miya, M., Inoue, J.G., Matsuura, K., Nishida, M., 2006. The mitochondrial ge-nome of spotted green pufferfish Tetraodon nigroviridis (Teleostei: Tetraodontiformes)and divergence time estimation amongmodel organisms in fishes. Genes & Genetic Sys-tems 81, 29–39.

Yamanoue, Y., Miya, M., Matsuura, K., Yagishita, N., Mabuchi, K., Sakai, H., Katoh, M.,Nishida, M., 2007. Phylogenetic position of tetraodontiform fishes within the higherteleosts: Bayesian inferences based on 44 whole mitochondrial genome sequences.Molecular Phylogenetics and Evolution 45, 89–101.

Yang, Z., 1994. Maximum likelihood phylogenetic estimation from DNA sequences withvariable rates over sites: approximate methods. Journal of Molecular Evolution 39,306–314.

of a novel percomorph fish clade mainly comprising “Syngnathoidei”