Comparative Phylogenetic Study of Four Genes of Mitochondrial Genome in Tenpounder Fishes ( Order : Elopiformes )

The Elopiformes represent a group of seven known species of fishes found in marine and estuarine ecosystem and many of them are endemic to East Asia. To date published morphological and molecular phylogenetic hypothesis of Tenpounder fishes are part congruent and there are some areas of significant disagreement with respect to species relationships. The present study analyzed the sequence data from four genes (Cytb, CO1, 16S and 12S rRNA) of mitochondrial genome for the attempt to estimate the relationships among five species such as Elops saurus, E. affinis, E. smithi, E. machnata and E. hawaiensis and to assess the phylogenetic utility of these markers. The Kimura 2parameter (K2P) genetic distance, average nucleotide frequencies, nucleotide substitution patterns and phylogenetic trees were reconstructed using neighbour joining (NJ) method. The interspecies K2P genetic distance was 0.0158 and intraspecies distance was 0.0042 based on the barcoding gene, CO1 sequence data. Whereas the interspecies K2P genetic distance was 0.6140 and intraspecies genetic distance was 0.0020 based on the Cytb data. The four mitochondrial marker genes used in this study showed different type of cluster and we could not confirm the relationship between the five Elops species. This is due to the independent mutation rate of each mtDNA genes. However, this problem can be overcome by analysing in parallel other gene markers.


Introduction
The ladyfishes or tenpounders (genus Elops) are widely distributed in tropical-subtropical, marine and coastal waters (Mcbride et al., 2010).Six species of Elops are recognized worldwide (Eschmeyer and Fong, 2008), but the taxonomy of the group is poorly known and some authors recognize fewer species (Nelson, 2006).Taxonomic uncertainty of Elops is exemplified by the ladyfish, E. saurus, currently recognized as the only species of Elops in the western Atlantic Ocean (Mcbride et al., 2010).Smith (1989) also noted that E. saurus and E. smithi had largely allopatric distributions.Recent work has failed to support the phenotypic hypothesis for this two species (Mcbride et al., 2010).Mcbride et al. (2010) examined specimens of E. saurus and Elops sp. using common morphological, meristic characters and Cytb data that have been used to distinguish six species of Elops worldwide (E.saurus, E. affinis, E. lacerta, E. senegalensis, E. machnata and E. hawaiensis).Many taxonomic studies of Elops examined less than 20 specimens per species and for some species even single specimen also used.Hence, there may be even more species of Elops awaiting discovery (Mcbride et al., 2010).Phylogenetically, Obermiller and Pfeiler (2003) identify E. saurus and E. smithi based on 12S and 16S rRNA gene sequence.So far E. senegalensis and E. lacerta are not yet entered into the GenBank database and there is no phylo-genetic analysis work has been made using molecular tools.Based on the study of Mcbride et al. (2010) E. saurus and E. smithi showed ecological specialization in tropical and subtropical habitats may be a foundation for speciation.
Recently three independent research groups have published results from molecular phylogenetic studies of the elopomorphs using mitochondrial ribosomal DNA sequences (Filleul and Lavoue, 2001;Obermiller and Pfeiler, 2003;Wang et al., 2003).Wang et al. (2003) analyzed the elopomorph interrelationships based on the complete 12S rRNA gene sequences (1073 bp) from 42 teleosts including the Elopomorpha (34 spp.).Forey et al. (1996) employed a total-evidence approach in analyzing elopomorph relationships, combining partial nucleotide sequences of the mitochondrial 12S rRNA (345 bp), 16S rRNA (535 bp), and nuclear 18S rRNA (1870 bp) genes and morphological data (56 characters) from 13 species, thereafter subjecting the combined data to MP analysis, and suggested that the saccopharyngiforms are deeply nested within the anguilliforms.Obermiller and Pfeiler (2003) also analyzed mtDNA sequences (754 bp) in segments of the 12S and 16S rRNA genes from 45 species including 33 elopomorphs, nine osteoglossomorphs, and three clupeomorphs and did not support the monophyly of Elopomorpha.Inoue et al. (2004) investigated Elopomorpha monophyly and interrelationships at the ordinal level using complete mitochondrial genomic data from 33

Sequence analysis
The CO1 gene partial sequences of Elops machnata were unambiguously edited using BioEdit sequence editor, aligned using CLUSTAL-W and checked manually.Haplotype definitions have been submitted to the NCBI GenBank (Acc.No. KF006255, KF006256, KF006257).To give more support to the present data, CO1, Cytb, 16S rRNA and 12S rRNA sequences of other available species were retrieved from GenBank and the details are given in Tab. 1.Nucleotide diversity, genetic variation, nucleotide composition and pairwise evolutionary distance among haplotypes were determined by Kimura 2-Parameter method (Kimura, 1980) using the software program MEGA 3.1 (Kumar et al., 2004).The neighbour-joining (NJ) trees for CO1, Cytb, 16S rRNA and 12S rRNA were constructed and to verify the robustness of the internal nodes of these trees, bootstrap analysis was carried out using 1000 pseudoreplications.

Results and discussion
A total of 58 sequences were from five Elops species were included and analysed in this study.Simplicity and un-ambiguity were observed among the sequences and no introns, deletions or stop codons were observed any of the CO1 and Cytb sequences.The CO1 sequence analysis revealed that the average nucleotide frequency was A = 24.16± 0.78%, T = 26.74± 0.17%, G = 18.96 ± 0.60, C = 30.14± 1.72 (Fig. 1).The interspecies transition and transversion was 86.23 and 13.78 respectively (Fig. 2).Tajima's statistics for E. machnata and E. hawaiensis were not significantly (p<0.01)different among individuals.Kimura 2 Parameter (K2P) genetic distance in four Elops species is given in Tab. 2 (below diagonal).The K2P genetic distance was high (0.023) between E. hawaiensis and E. affinis.Very low K2P distance (0.003) was exhibited between E. hawaiensis and E. machnata.The Cytb sequence analysis among the four species showed that the average nucleotide frequency was A = 25.19 ± 0.65%, T = 26.54± 0.12%, G = 18.24 ± 0.53, C = 30.03± 0.90 (Fig. 1).The interspecies transition and transversion was 57.01 and 42.99 respectively and the R value was 1.28 (Fig. 2) based on the Cytb data.Tajima's statistics for Cytb sequence in E. smithi was not significantly (p<0.01)different among individuals (Tab.4).The K2P genetic distance based on the Cytb data among the Elops species is given in Tab. 3. The K2P genetic distance was high (1.231) between E. saurus and E. affinis and the distance was low (0.010) between E. hawaiensis and E. smithi.
The 16S rRNA gene sequence analysis showed the average nucleotide frequency was A = 31.63± 0.82%, T = 13.99 ± 0.42%, G = 23.39 ± 0.83, C = 30.99± 1.20 (Fig. 1).The interspecies transition and transversion was 52.96 and 37.04 respectively and the R value was 1.70 (Fig. 2).The K2P genetic distance based on 16S rRNA gene data of Elops species is given in Tab. 4. The high (0.048) K2P species represent the major teleostean and elopomorph lineages.Mitogenomic data strongly supported the order Elopiformes occupied the most basal position in the elopomorph phylogeny, with the Albuliformes and a clade comprising the Anguilliformes and the Saccopharyngiformes forming a sister group.
Mitochondrial DNA provides a potential tool for studying population and phylogenetic analysis and the different genes of mitochondrial genome are used for phylogeny analysis at different levels of taxa, family, species and individual's level.Hence, an attempt has been made to report the phylogenetic analysis based on the updated nucleotide sequence data from GenBank for the four regions of the mitochondrial genome (Cytb, CO1, 16S rRNA and 12S rRNA) to assess the pattern of species relationship and also to examine the rates and types of nucleotide substitutions among the Tenpounder fish species.

Sample collection
Forty specimens of Elops machnata were collected from Vellar estuary (Lat 11°29' N; Lon 79°46' E) southeast coast of India.Immediately after the collection, the specimens were kept in the icebox and the fishes were identified up to the species level using the FAO fish identification sheets (Thomson, 1984).The voucher specimens are maintained in Marine Biotechnology Laboratory, CAS in Marine Biology, Annamalai University.The fin-clips were preserved in 95% ethanol and stored at 4°C until used.

DNA isolation
The DNA was isolated by standard Proteinase-K/Phenol-Chloroform-ethanol method (Sambrook et al., 1989) and the concentration of isolated DNA was estimated using a UV spectrophotometer.The DNA was diluted in TAE buffer to a final concentration of 100 ng⁄ μL.

Gene amplification and sequencing
The CO1 gene was amplified in a 50 μL volume with 5 μL of 10X Taq polymerase buffer, 2 μL of MgCl2 (50 mM), 0.25 μL of each dNTP (0.05 mM), 0.5 μL of each primer (0.01 mM), 0.6 U of Taq polymerase and 5 μL of genomic DNA.The primers used for the amplification of the CO1 gene were FishF1-5' TCAACCAACCACAAAGACAT-TGGCAC 3' and FishR1-5' TAGACTTC TGGGTG-GCCAAAGAATCA 3' (Ward et al., 2005).The thermal regime consisted of an initial step of 2 min at 95°C followed by 35 cycles of 40 s at 94°C, 45 s at 54°C and 1 min at 72°C followed in turn by final extension of 10 min at 72°C.The PCR products were visualized on 1.5% agarose gels, and the most intense products were selected for sequencing.The cleaned up PCR product was sequenced by a commercial sequencing facility (Ramachandra Innovis, Chennai, India).genetic distance was observed between E. saurus and E. hawaiensis and low (0.024) among E. saurus and E. affinis.The 12S rRNA gene sequence analysis result showed the average nucleotide frequency was A = 28.81 ± 0.43%, T = 21.19 ± 0.31%, G = 26.44 ± 0.32, C = 23.56 ± 0.70 (Fig. 1).The transition and transversion among the species was 37.85 and 62.15 respectively and the R value was 0.68 (Fig. 2).The K2P genetic distance based on 12S rRNA gene sequence data of the fish species is given in Tab. 2 (above diagonal).The K2P genetic distance was high (0.035) between E. saurus and E. affinis and low (0.004) genetic distance was observed between E. hawaiensis and E. machnata.
The neighbour-joining method was actually employed in this study to get a solid phylogenetic information by the dendrogram.All the 58 sequences of the tenpounder fishes were subjected in the phylogenetic analysis.The neighbour joining trees by K2P model for the four mitochondrial genes were created to provide a graphical representation of the patterning of divergence of five Elops species.The neighbour joining phylogenetic tree based on CO1 gene sequences is given in Fig. 3.As per the NJ tree, two distinct clades as two sub-trees within the same genus were recognized with high bootstrap value.Among the two sub-trees, one has an independent assemblage of E. hawaiensis and E. machnata with 98% bootstrap value.Another clade representing the other two species such as Elops machnata JF493410 a ---3.
Elops Species identification and phylogenetic relationship based on traditional methods and molecular methods are mostly concordant (Ward et al., 2005).The efficiency of species identification by molecular methods is judged by E. saurus and E. affinis.Fig. 4 shows the neighbour joining phylogenetic tree based on Cytb gene sequences.In this tree, two distinct clusters are raised with high bootstrap value.Among the two clusters, one large has an assemblage of E. hawaiensis, E. saurus and E. smithi with 54% bootstrap value.Another clade representing the other one species, E. affinis.
The neighbour joining phylogenetic tree based on 16S rRNA gene sequences is given in Fig. 5.As per this tree, three distinct clades as sub-trees within the Elops genus were recognized with high bootstrap value.Among the three sub-trees, one has the cluster of E. saurus with 84% In such cases, a secondary independent molecular marker is required to solidify or confirm identification if applicable (Smith et al., 2007).In this study five Elops species were found genetically distinct from each other based on four mtDNA gene sequences which demonstrates simplicity and unambiguity.Morphologically very similar species like E. affinis and E. saurus form sister clade by all the four gene based NJ trees.Whereas, E. machnata and E. hawaiensis form an independent sister clade in NJ tree of CO1 gene.Because of the data deficient in GenBank we could not clearly resolve E. smithi from all other species.The observed genetic divergence from CO1 gene is sufficient to differentiate individuals of different Elops species.In this study the level of intra-species variation was low which may be due to low number of haplotype identified in the sample with limited numbers collected for this study.Similarly, Lakra et al. (2011) reported very low intra-specific genetic divergence for scombroid fishes and Ward et al. (2005) showed in many marine teleost species.Peris et al. (2009) also reported very low interspecies genetic distance for Indian the levels of intraspecific homogeneity and interspecific heterogenenity displayed by the intended method (Hallden et al., 1994;Lievens et al., 2001).Mitochondrial CO1 gene, as an attractive "species barcode", its high efficiency in species identification has been reported in Australia marine fishes (Ward et al., 2005), Canadian freshwater fishes (Hubert et al., 2008), ornamental fishes in the market of North America (Steinke et al., 2009b) and marine fishes of Japan (Zhang and Hanner, 2011).Due to the high efficiency in species identification, some ichthyologists advocate the inclusion of a DNA barcode in the formal description of species (Victor, 2007;Astarloa et al., 2008).Somehow, it deserves attention to recent speciation, introgressive hybridization, and taxonomic splitting, which  The genetic distance between Elops sp. and E. saurus, which occur sympatrically in Florida (Smith, 1989), was 0.021, similar to that found between the allopatric E. saurus and E. hawaiensis (0.024) (Obermiller et al., 2003) which inhabit different ocean basins.The present study revealed that the K2P genetic distance between the allopatric E. saurus and E. hawaiensis was 0.014 based on CO1 data and it was 0.013 based on Cyt b and 12S rRNA gene sequence data respectively.
It is unfeasible to build the phylogeny of Elopiforme fishes only based on mitochondrial DNA fragments alone.The disadvantage of 16S rDNA sequences is the lack of discrimination power among closely related species.However, this problem can be overcome by analysing in parallel other gene markers.Polyphyly/paraphyly in the NJ tree probably results from "bad taxonomy" when named species fail to identify the genetic limits of separate evolutionary entities, particularly for perplexing taxa involving cryptic species (Nice and Shapiro, 2001).All the four mitochondrial marker genes used in this study showed different type of cluster and we could not confirm the re-carangid fishes.For many marine fishes, there is a lack of phylogeographic structure among populations (Palumbi, 1994;Hellberg et al., 2002).Zhang (2011) reported that, individuals from long distance localities, some intraspecific genetic variations reduced to zero within families Carangidae, Sciaenidae, and Mullidae.However, some pairwise K2P distances exceeded 1.00% within the coastal species such as Acentrogobius caninus, Scomber japonicus, Terapon jarbua, Upeneus sulphureus, Elops hawaiensis, Gymnothorax pseudothyrsoideus, and Dendrophysa russelii.Based on the CO1 data, the present study also confirmed that the interspecies K2P genetic variation is more than 1.00%.It implied that biological mechanisms were responsible for the fluctuation of intraspecific genetic divergences in marine fishes.Ward et al. (2005) reported an overall higher GC content in fishes based on complete MtDNA genome ranging from 38.4-43.2% and in CO1 alone it was 42.2-47.1%,which reflects the 3 rd base variation.Peris et al. (2009) also reported considerable variation was exhibited in carangids in the 3rd base position.Ravitchandirane et al. (2012) showed the mean GC content was 36.8 -42.6% among the nine Nemipterus species.In this study it has been observed  lationship between the Elops species.This is due to the independent mutation rate of each mtDNA genes.Different nucleotide positions and genes within mtDNA are known to evolve at heterogeneous rates within a lineage (Brown et al., 1982;Gillespie, 1986;Moritz et al., 1987), and particular mtDNA genes (such as cytochrome oxidase) also show rate differences as high as fivefold across lineages (Brown and Simpson, 1982;Crozier et al., 1989).As per Zhang (2011), if we cannot set a threshold of the genetic variation in species delimitation, we find ourselves sunk in the dilemma facing new or cryptic species.On the one hand, the morphological taxonomy cannot give a definite identification.On the other hand, we cannot claim that it may be a new species based on molecular analysis without the species delimitation (Zhang, 2011).An assumed threshold is helpful to expedite discovery of new species and biodiversity, especially in dealing with little-studied biota, although a single, uniform threshold for species delimitation seems arbitrary because the rates of molecular evolution vary widely within and among lineages (Zuckerkandl and Pauling, 1965;Will and Rubinoff, 2004;De-Salle et al., 2005).

Conclusions
The study has successfully assessed the utility of the four genes (Cytb, CO1, 16S and 12S rRNA) of mitochondrial genome to estimate the relationships among five Elops species such as Elops saurus, E. affinis, E. smithi, E. machnata and E. hawaiensis.The four mitochondrial marker genes used in this study showed different type of cluster and we could not confirm the relationship between the five Elops species.This is due to the independent mutation rate of each mtDNA genes.However, this problem can be overcome by analysing in parallel by other gene markers also.Further studies involving all the Elopiformes in the world and also by increasing the sample size in future studies will clarify the issue.

Fig. 1 .
Fig. 1.Percentage of nucleotide composition in four mitochondrial genes of Elops species Fig. 2. The nucleotide substitution patterns of four mitochondrial genes based on the Kimura (1980) 2-parameter model (Ts-Transition; Tv-Transversion; R-estimated Transition/ Transversion bias)

Fig. 4 .
Fig. 4. Neighbour joining tree of Elops species based on Cytb gene sequences

Fig. 5 .
Fig. 5. Neighbour joining tree of Elops species based on 16S rRNA gene sequences Tab. 1. Elops species and their mitochondrial genes with accession number and reference Tab. 2. K2P genetic distance between Elops species based on CO1 gene sequences (below diagonal), based on 12S rRNA gene sequences (above diagonal) Tab. 3. K2P genetic distance between Elops species based on Cytb gene sequences Tajima's Neutrality Test for CO1 and Cytb genes in Elops species M = number of sequences, S = Number of segregating sites, ps = S/m, Θ = ps/a1, π = nucleotide diversity, and D = Tajima test statistic