Evolution of the Deep-Sea Gulper Eel Mitochondrial Genomes: Large-Scale Gene Rearrangements Originated Within the Eels

Molecular Biology and Evolution, Nov 2003

Recent studies have demonstrated that deviations from the typical vertebrate mitochondrial gene order are more frequent than initially thought. Such deviations, however, are minor, with inversions and/or translocations of a few genes being involved and tandem duplication of the gene regions followed by deletions of genes having been invoked as mechanisms originating in such novel gene order. During the course of molecular phylogenetic studies on the Elopomorpha (eels and their allies), we found that mitochondrial genomes (mitogenomes) from the two deep-sea gulper eels, Eurypharynx pelecanoides (Eurypharyngidae) and Saccopharynx lavenbergi (Saccopharyngidae), exhibit an identical gene order which greatly differs from that of any other vertebrates. Phylogenetic analysis using the mitogenomic data from 59 species of fish not only confirmed a single origin of such a gene order with confidence but also indicated that it had been derived from the typical vertebrate gene order. Detailed comparisons of the gulper eel gene order with that of typical vertebrates suggested that occurrence of a single step, large-scale duplication of gene region extending >12 kb, followed by deletions of genes in a common ancestor of the two species, most parsimoniously accounts for this unusual gene arrangement.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://mbe.oxfordjournals.org/content/20/11/1917.full.pdf

Evolution of the Deep-Sea Gulper Eel Mitochondrial Genomes: Large-Scale Gene Rearrangements Originated Within the Eels

Jun G. Inoue 0 Masaki Miya Katsumi Tsukamoto 0 Mutsumi Nishida 0 0 Ocean Research Institute, University of Tokyo , Tokyo , Japan; and Department of Zoology, Natural History Museum & Institute , Chiba , Japan Recent studies have demonstrated that deviations from the typical vertebrate mitochondrial gene order are more frequent than initially thought. Such deviations, however, are minor, with inversions and/or translocations of a few genes being involved and tandem duplication of the gene regions followed by deletions of genes having been invoked as mechanisms originating in such novel gene order. During the course of molecular phylogenetic studies on the Elopomorpha (eels and their allies), we found that mitochondrial genomes (mitogenomes) from the two deep-sea gulper eels, Eurypharynx pelecanoides (Eurypharyngidae) and Saccopharynx lavenbergi (Saccopharyngidae), exhibit an identical gene order which greatly differs from that of any other vertebrates. Phylogenetic analysis using the mitogenomic data from 59 species of fish not only confirmed a single origin of such a gene order with confidence but also indicated that it had been derived from the typical vertebrate gene order. Detailed comparisons of the gulper eel gene order with that of typical vertebrates suggested that occurrence of a single step, large-scale duplication of gene region extending .12 kb, followed by deletions of genes in a common ancestor of the two species, most parsimoniously accounts for this unusual gene arrangement. - Vertebrate mitochondrial gene order was initially considered conservative because the complete nucleotide sequences of the mitochondrial genomes (mitogenomes) from mammals (Anderson et al. 1981, 1982; Bibb et al. 1981) and the African clawed frog (Roe et al. 1985) showed a common gene order. As the complete mitochondrial DNA (mtDNA) sequences have been determined for a number of vertebrates (269 species as of 11 June 2003; National Center for Biotechnology Information [NCBI] Web site: http://www.ncbi.nlm.nih.gov/PMGifs/Genomes/ 7742.html), it has been recognized that deviations from the typical gene order are not so rare in vertebrates (fig. 1; Boore 1999). With the exception of placental mammals, examples include all major lineages of vertebrates such as lampreys (Lee and Kocher 1995), amphibians (Macey et al. 1997; Sumida et al. 2001), reptiles (Kumazawa and Nishida 1995; Quinn and Mindell 1996; Macey et al. 1997), birds (Desjardins and Morais 1990, 1991; Quinn and Wilson 1993; Mindell, Sorenson, and Dimcheff 1998), marsupials (Paabo et al. 1991; Janke et al. 1994), and fish (Miya and Nishida 1999; Inoue et al. 2001b; Miya, Kawaguchi, and Nishida 2001). Such gene rearrangements, however, are local (most notably within a cluster of tRNA genes), and no example has been found that involves genomic-scale rearrangements. Gulper eels are well-known deep-sea creatures, collectively including four families placed in the order Saccopharyngiformes, that occur at bathypelagic depths (.1,000 m) throughout the worlds oceans (Bertelsen, Nielsen, and Smith 1989). Their bizarre appearance (fig. 1), which results from extremely enlarged, deformed gapes at the tip of the eels snake-like head and body, has attracted much attention from many researchers (e.g., Bertelsen, Nielsen, and Smith 1989; Robins 1989). Such specialized anatomical features have led investigators to suppose that the phylogenetic positions of these animals are outside the bony fish (e.g., Tchernavin 1946), although they have generally been considered to be close relatives of the true eels (order Anguilliformes), because they have a leaf-like leptocephalous larval stage (Greenwood et al. 1966; Inoue and Miya 2001). During the course of molecular phylogenetic studies of the Elopomorpha (eels and their allies) using mitogenomic data, we found an unusual mitochondrial gene order for one of the gulper eels, Eurypharynx pelecanoides (fig. 1). This article describes novel gene organization in the mitogenomes of two species of gulper eels, E. pelecanoides and Saccopharynx lavenbergi. We employed a polymerase chain reaction (PCR)-based approach developed by Miya and Nishida (1999) for sequencing the complete mitogenomes of these fishes. To explore the origin of such unique mitogenomes, we subjected the mitogenomic data to phylogenetic analysis, and we discuss here the possible mechanisms generating such a gene arrangement in the two gulper eels. Mitochondrial DNA sequences were obtained from six individuals from two species representing two families in the order Saccopharyngiformes (Robins 1989). For the monotypic family Eurypharyngidae, one individual of E. pelecanoides was used as a specimen for determination of the complete mtDNA sequence, and the other four individuals, including two leptocephalous larvae, were used for determination of partial sequences of four major gene junctions (fig. 2) to confirm the gene order in the E. pelecanoides mitogenome. For the family Saccopharyngidae, a single individual of S. lavenbergi was used as a specimen for determination of the complete mtDNA sequence. Voucher specimens were deposited in the fish collections at the Natural History Museum & Institute, Chiba, Japan (CBM-ZF), and at the University of Washington (UW): E. pelecanoides (CBM-ZF 10311: Hyuga-Nada, FIG. 1.The typical gene order and its derivatives (represented linearly) in vertebrate mitochondrial (mt) genomes. Thick bars correspond to gene regions that presumably underwent tandem duplication and subsequent deletions of genes resulting in gene rearrangements. The gene map of the gulper-eel mitogenome is represented linearly, with five blocks of gene regions which retained the typical vertebrate gene order being colored. All protein-coding genes are encoded by the H strand except for the ND6 gene (underlined). Transfer RNA (tRNA) genes are designated by single-letter amino-acid codes; those encoded by the H and L strands are shown outside/above and inside/below the gene maps, respectively. 12S and 16S indicate 12S and 16S ribosomal RNA genes; ND1-6, and 4L, NADH dehydrogenase subunits 16, and 4L genes; COIIII, cytochrome c oxidase subunits IIII genes; ATPase 6 and 8, ATPase subunits 6 and 8 genes; cyt b, cytochrome b gene; CR, control region; L1, tRNALeu(UUR); L2, tRNALeu(CUN); S1, tRNASer(UCN); S2, tRNASer(AGY). All relevant information regarding the mitochondrial gene rearrangements is available from http://www.jgi.doe.gov/ programs/comparative/Mito_top_level.html curated by J. L. Boore. Long PCR strategies for the complete mtDNA sequences of E. pelecanoides and S. lavenbergi are described below. Two pairs of long-PCR primers (S-LA-16S-L H12293-Leu and L12321-Leu S-LA-16S-H for E. pelecanoides; and L250816S Sala-ND4-H and Sala-COI-L Sala-ND1-H for S. lavenbergi) amplified two segments that covered the entire mitogenome in each species. The thumbnail illustrations and photographs are reproduced through the courtesies of the Tokai University Press (Masuda et al. 1984), Andromeda Oxford (Banister 1987), and Marshall Editions (Whitfield 1998). southern Japan; CBM-ZF 10312, 10313, 10316, and 10317: equatorial central Pacific Ocean) and S. lavenbergi (UW 045633: Eastern Pacific). A portion of the epaxial musculature (ca. 0.25 g) was excised from fresh specimens of each species and immediately preserved in 99.5% ethanol. Total genomic DNA was extracted using the Qiagen QIAamp tissue kit following the manufacturers protocol. Polymerase Chain Reaction and Sequencing The entire mitogenomes of two gulper eels were amplified using a long-PCR technique (Cheng, Higuchi, and Stoneking 1994). Five fish-universal and three speciesspecific long-PCR primers (S-LA-16S-LH12293-Leu and L12321-Leu S-LA-16S-H for E. pelecanoides; L2508 16S Sala-ND4-H and Sala-CO1-L Sala-ND1-H for S. lavenbergi; fig. 1) were used to amplify the entire mitogenome of each eel in two reactions. The entire mitogenomes of the other four E. pelecanoides individuals were amplified with three fish-universal primers and one species-specific long-PCR primer (L250816S EupeND2-H and L12321-Leu S-LA-16S-H). Four speciesspecific primers (Eupe-ND2-H for E. pelecanoides; Sala-CO1-L, Sala-ND1-H, and Sala-ND4-H for S. lavenbergi) were alternatively designed with reference to the partial nucleotide sequences from the ND2 gene of E. pelecanoides and the COI, ND1, and ND4 genes of S. lavenbergi determined from each total DNA using fish-universal primers. The long-PCR products were diluted with TE buffer (1:20) for subsequent use as PCR templates, except for a region intervening between the two long-PCR primers (between S-LA-16S-H and S-LA-16S-L, and H12293-Leu and L12321-Leu for E. pelecanoides; fig. 1), in which total genomic DNA was used alternatively. A total of 82 fish-universal primers (Miya and Nishida 1999, 2000; Inoue et al. 2000, 2001a, 2001b, 2001c, 2001d; Ishiguro, Miya, and Nishida 2001; Kawaguchi, Miya, and Nishida 2001) were used in various combinations to amplify contiguous, overlapping segments of the entire mitogenome for each of the two species. Species-specific primers were designed in cases where no appropriate fishuniversal primers were available. A list of PCR primers used for a specific species is available from J.G.I. upon request. Long-PCR and subsequent PCR reactions were carried out as previously described (e.g., Miya and Nishida 1999; Inoue et al. 2003). Double-stranded PCR products, purified using a PreSequencing Kit (USB), were subsequently used for direct cycle sequencing using dye-labeled terminators (Applied Biosystems). The primers used were the same as those for PCR. All sequencing reactions were performed according to the manufacturers instructions. Labeled fragments were analyzed by means of a Model 377 DNA sequencer (Applied Biosystems). DNA sequences were analyzed using the computer software package program DNASIS version 3.2 (Hitachi Software Engineering). Locations of the 13 protein-coding genes were determined by comparisons of DNA or amino acid sequences of bony fish mitogenomes. The 22 tRNA genes were identified by their cloverleaf secondary structures (Kumazawa and Nishida 1993) and anticodon sequences. The two rRNA genes were identified by sequence similarity and secondary structure (Gutell, Gray, and Schnare 1993). Sequence data are available from DDBJ/ EMBL/GenBank with accession numbers AB046473 and AB046475AB046490 for E. pelecanoides, and AB047825 for S. lavenbergi. The data matrices from Inoue et al. (2001d) and Miya, Kawaguchi, and Nishida (2001) were combined and redundant taxa were eliminated. Mitogenomic sequences from the two gulper eels and two teleosts (Gymnothorax kidako, AP002976 [Inoue et al. 2003] and Salmo salar, U12143 [Hurst et al. 1999]) were added. Because unambiguous alignments of the two rRNA genes on the basis of secondary structure models were not feasible (Miya and Nishida 2000), they were not used in the analyses. The ND6 gene was not used in the phylogenetic analyses because of its heterogeneous base composition and consistently poor phylogenetic performance (e.g., Zardoya and Meyer 1996; Miya and Nishida 2000). Also, tRNA loops, tRNASer(AGY) (unstable inferred secondary structures in some species), and other ambiguously aligned regions, such as the 59 and 39 ends of several protein-coding genes, were excluded form the analyses. In addition, 3rd codon positions in the protein-coding genes that could positively mislead an analysis of higher-level relationships of actinopterygians (Miya and Nishida 2000) were excluded from the analyses, leaving 7,984 available nucleotide positions from the 12 protein-coding genes and 21 tRNA genes. Bayesian phylogenetic analyses were conducted using MrBayes 2.01 (Huelsenbeck and Ronquist 2001). The general time reversible model with some sites assumed to be invariable and with variable sites assumed to follow a discrete gamma distribution (GTR I ; Yang 1994) was selected as the best-fit model of nucleotide substitution (ModelTest version 3.06; Posada and Crandall 1998). We set the maximum likelihood parameters in MrBayes as follows: lset nst 6 (GTR), rates invgamma (I ), and basefreq estimate (estimated proportion of base types from the data). The Markov chain Monte Carlo process was set so that four chains (three heated and one cold) ran simultaneously. We conducted two independent runs for 1 million generations, with trees being sampled every 100 generations, each of which started from a random tree. Two independent analyses converged on similar loglikelihood scores and reached stationarity (lack of improvement in ML scores) at no later than 120,000 generations. Thus, the first 1,200 trees were discarded from each analysis, and the remaining 17,602 combined samples from two independent analyses were used to generate a 50% majority rule consensus tree, with the percentage of samples recovering any particular clade representing that clades posterior probability (Huelsenbeck and Ronquist 2001). The total lengths of the E. pelecanoides and the S. lavenbergi mitogenomes are 18,978 bp and 18,495 bp (except for a portion of the putative control region for S. lavenbergi), respectively (see Supplementary Material online). The genome content of these two gulper eels includes two rRNA, 22 tRNA, and 13 protein-coding genes, as found in other vertebrates (fig. 2). Also as in other vertebrates, most genes of these two species are encoded on the H-strand, except for the ND6 and eight tRNA genes. All features are similar in length to those in other bony fishes with the exception of the COI gene and the control region for E. pelecanoides and the COI and cyt b genes for S. lavenbergi (approximately 100, 800, 150, and 30 bp longer than those in other bony fishes, respectively). The E. pelecanoides mitogenome contains three noncoding regions longer than 200 bp (fig. 2; see also FIG. 2.Comparison of mitochondrial gene orders among typical vertebrates, E. pelecanoides, and S. lavenbergi. Five blocks of gene regions (A, tRNAGlutRNAGln; B, ATP8-ND3; C, tRNALeu(CUN)ND6; D, tRNAIletRNALys; and E, tRNAArgtRNASer(AGY) gene regions) retain the typical vertebrate gene order with the exception of several tRNA genes (arrowheads). The partial mtDNA sequences for four gene junctions (vertical bar: ND1ATP8, ND3-ND5, ND6-ND2, and ND4-cyt b regions), the gene order of which greatly differ from that of other typical vertebrates, were determined for four additional E. pelecanoides individuals (data available form DDBJ/EMBL/GenBank with accession numbers AB046475AB046490). Genes are depicted and labeled as in figure 1. Supplementary Materials online; noncoding region [nc1 and 2] and control region [CR]). The nc1, located between tRNASer(UCN) and tRNAAsp genes, and the nc2, located between the tRNAAsp and COII genes, contain two and four pseudogenes corresponding to degenerating copies of tRNAAsp gene, respectively. A noncoding region (1,818 bp) located between the cyt b and tRNAPhe genes appears to correspond to the control region. The S. lavenbergi mitogenome also contains three noncoding regions longer than 200 bp (noncoding region [nc] and control regions [CR1 and CR2]). The nc, located between the tRNALeu(CUN) and ND5 genes, contains at least two pseudogenes corresponding to degenerating copies of the tRNALeu(CUN) gene. CR1 (992 bp), located between two tRNA gene clusters (TP and IM regions), and CR2 (. 837 bp), located between cyt b and tRNAPhe genes, appear to correspond to the control region, and they have completely identical sequences over 800 bp, indicating that they are undergoing concerted evolution (see Kumazawa et al. 1998). With the exception of the two duplicated control regions (CR1 and CR2) in the S. lavenbergi mitogenome, the mitogenomes from the two deep-sea gulper eels exhibit an identical gene order (fig. 2). However, the mitochondrial gene order of the two gulper eels differs greatly from that of all other vertebrates known to date. When some tRNA genes were excluded from the comparisons (fig. 2), we identified five gene blocks (A, tRNAGlutRNAGln; B, ATP8-ND3; C, tRNALeu(CUN)ND6; D, tRNAIletRNALys; and E, tRNAArgtRNASer(AGY) gene regions), the gene order of which is identical to that of typical vertebrates. Phylogenetic Positions of Two Gulper Eels Figure 3 is a 50% majority rule consensus tree of the 17,602 combined samples from two independent Bayesian analyses of the 59 mitochondrial nucleotide sequences from the concatenated 12 protein-coding (no 3rd codon positions), plus 21 tRNA genes (stem regions only) using the GTR I model of sequence evolution (Yang 1994). With the exception of a few nodes within the Neoteleostei, most internal branches were supported by high Bayesian posterior probabilities ( 95%). It should be noted that the resultant tree explicitly demonstrates not only that the two gulper eels form a sister-group relationship with a high posterior probability (100%), but also that they are deeply nested within the Anguilliformes (true eels), strongly suggesting that they have been derived from an eel-like ancestor. Recently, more than 140 complete mtDNA sequences from the actinopterygian fishes have been determined (e.g., Miya and Nishida 1999, 2000; Inoue et al. 2000, 2001a, 2001b, 2001c, 2001d, 2003; Ishiguro, Miya, and Nishida 2001, 2003; Kawaguchi, Miya, and Nishida 2001; Miya, Kawaguchi, and Nishida 2001; Miya et al. 2003; Saitoh et al. 2003), and several examples of gene rearrangements have been reported (fig. 1). It should be noted that such gene rearrangement events involve few genes and that the gene arrangement of the gulper eel mitogenomes differs greatly from that of other vertebrates. To deduce the evolution of the gene arrangements in the two gulper eels, inference for the ancestral organization based on a reliable phylogenetic framework is required. Although the gulper eels have been included in the FIG. 3.The 50% majority rule consensus tree derived from Bayesian analysis of mitogenomic data from 57 teleosts and two outgroup species using MrBayes 2.01 (Huelsenbeck and Ronquist 2001) with GTR I model (Yang 1994) of sequence evolution. Numerals indicate Bayesian posterior probabilities (shown as percentages). Branch lengths were estimated by constrained ML analysis using PAUP*4.0b10 (Swofford 2002) with the GTR I model of sequence evolution. The scale indicates expected nucleotide substitutions per site. Thick branches indicate occurrences of the gene rearrangements. FIG. 4.Proposed mechanism of mitochondrial gene rearrangement in gulper eels in a model of tandem duplication of a gene region and subsequent gene deletions. (A) Typical vertebrate mitochondrial gene order. (B) Tandem duplication in the tRNAIlecontrol region (thick bar) and subsequent deletions of redundant genes, resulting in the observed gene order of gulper eels (C). Genes are depicted and labeled as in figure 1. Elopomorpha based solely on a distinct pelagic larval form, termed the leptocephalus, no one has corroborated their phylogenetic position using character matrices derived from morphological or molecular data (Inoue and Miya 2001). Bayesian analysis using the mitogenomic data from 59 species which fully represent teleostean fish diversity (fig. 3) not only corroborated that the novel gene order of the two gulper eels originated in a common ancestor of the two species, but also demonstrated that their origin was independent from those of the other novel gene orders. It also appears that the novel gene order of Conger myriaster, another member of the Elopomorpha, has an origin independent from that of gulper eels. It should be noted that Anguilla japonica, which has the typical vertebrate gene order, was confidently placed as a sister species of the two gulper eels. The most parsimonious reconstruction of the gene rearrangement events on the phylogenetic tree indicated that the aberrant gene order of the two gulper eels is derived from that of typical vertebrates. Two major mechanisms have been proposed to explain the gene rearrangements in vertebrate mitogenomes (Lee and Kocher 1995; Kumazawa et al. 1996). One is the tandem duplication of gene regions as a result of slipped strand mispairing, followed by the deletions of genes (Moritz and Brown 1986; Levinson and Gutman 1987). Although the dynamics of mitochondrial gene arrangements have not been explained in some invertebrates (e.g., Kurabayashi and Ueshima 2000; Machida et al. 2002; Tomita et al. 2002), vertebrate mitochondrial gene rearrangements may well be explained by such a mechanism (Desjardins and Morais 1990; Quinn and Wilson 1993; Kumazawa and Nishida 1995; Kumazawa et al. 1996; Mindell, Sorenson, and Dimcheff 1998; Miya and Nishida 1999; Inoue et al. 2001b), and its feasibility is also supported by the frequent polymorphic duplications of mtDNA sequences (e.g., Stanton et al. 1994; Gach and Brown 1997; Mindell, Sorenson, and Dimcheff 1998; Miya and Nishida 1999). The other suggested mechanisms invoke the illicit priming of replication by tRNAs and the resultant integration of tRNA genes around the control region (Cantatore et al. 1987; Jacobs et al. 1989). Although the second model is not excluded at this point, the first model is favored as the mechanism of gene rearrangement in the gulper eel mitogenomes. Assuming that a long DNA fragment in the tRNAIleCR region of the typical organization is duplicated (fig. 4), subsequent deletions of redundant genes would give rise to the rearranged gene organization in gulper eels. Such tandem duplication and subsequent deletions most parsimoniously resulted in the observed gene order and associated intergenic spacers in these two gulper eels. Multiple deletions of redundant genes seemed to be incomplete in the two gulper eels, owing to several stretches of noncoding sequences (fig. 2) occurring around the genes involved in the rearrangement (fig. 4). Although nc1 and nc2 in E. pelecanoides and nc in S. lavenbergi are composed of repetitive pseudogenes corresponding to degenerating copies of adjacent tRNA gene, CR1 and CR2 in S. lavenbergi mitogenome have identical sequences over 800 bp, and CR1 is located at the putative duplicated position while CR2 is located at the original position of the control region. This observation in the S. lavenbergi


This is a preview of a remote PDF: https://mbe.oxfordjournals.org/content/20/11/1917.full.pdf

Jun G. Inoue, Masaki Miya, Katsumi Tsukamoto, Mutsumi Nishida. Evolution of the Deep-Sea Gulper Eel Mitochondrial Genomes: Large-Scale Gene Rearrangements Originated Within the Eels, Molecular Biology and Evolution, 2003, 1917-1924, DOI: 10.1093/molbev/msg206