The Complete Nucleotide Sequence of the Mitochondrial DNA of the Dogfish, Scyliorhinus canicula

Corresponding author: Gabriel Gachelin, Unité de Biologie Moléculaire du Gène, Unité INSERM U277, Institut Pasteur, 25 rue du Dr. Roux, 75015 Paris, France., ggachel{at}pasteur.fr (E-mail).

Communicating editor: N. TAKAHATA

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have determined the complete nucleotide sequence of the mitochondrial DNA (mtDNA) of the dogfish, Scyliorhinus canicula. The 16,697-bp-long mtDNA possesses a gene organization identical to that of the Osteichthyes, but different from that of the sea lamprey Petromyzon marinus. The main features of the mtDNA of osteichthyans were thus established in the common ancestor to chondrichthyans and osteichthyans. The phylogenetic analysis confirms that the Chondrichthyes are the sister group of the Osteichthyes.

THE mitochondrial DNA (mtDNA) of most animals is a self-replicating, 15- to 17-kb-long, circular DNA molecule. Animal mtDNA codes for 13 mitochondrial proteins, 22 mitochondrial tRNAs, and two mitochondrial-specific ribosomal RNAs, the 12S and 16S rRNAs. It also contains DNA regions aimed at controlling its replication and transcription. From the evolutionary viewpoint, mtDNAs are "small genomes" that coevolve at their own rate with the organism in which they are lodged. Thus, mtDNA sequences are widely used to construct phylogenetic trees. The genomic organization of the mtDNA (i.e., the relative location of the genes and the structure of the control regions of the mitochondrial genome) has also evolved, and the comparison of the genetic maps of mtDNA has been used to describe some key points in the evolution of animals [e.g., the myxine/lamprey/dogfish relationships (DELARBRE et al. 1997 ) and the complex evolution of snakes (KUMAZAWA and NISHIDA 1995 )]. However, although a number of complete nucleotide sequences of mtDNA have been published, the relative lack of data hinders efficient statistical analysis. Concerning the evolution of fish, the sequence of mtDNA of the sea lamprey [Petromyzon marinus (LEE and KOCHER 1995 )] is known and so are several sequences of mtDNA of the Osteichthyes (TZENG et al. 1992 ; CHANG et al. 1994 ; ZARDOYA et al. 1995 ; NOACK et al. 1996 ; ZARDOYA and MEYER 1996 , ZARDOYA and MEYER 1997 ). No complete sequence of the mtDNA of any chondrichthyan has been published yet, however, although the mtDNA sequences of several members of this group are needed to examine their relationships with osteichthyans. In this article, we describe the complete nucleotide sequence of a chondrichthyan fish, the dogfish Scyliorhinus canicula. The sequence extending from the tRNA-Leu gene to the beginning of the COI gene of the S. canicula mtDNA has already been reported (DELARBRE et al. 1997 ). From the comparison of the mtDNA genomic map with those of other animals, as well as a study of the main features of the DNA sequence, the dogfish appears very similar to osteichthyans and significantly different from the sea lamprey.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Animals:
A specimen of S. canicula was caught offshore Roscoff (Brittany, France). The animal was anesthesized, killed, and dissected. The organs were immediately frozen in liquid nitrogen and stored at -78°.

Preparation of DNA:
Total DNA was prepared from dogfish muscles by proteinase K digestion according to conventional procedures (HOGAN et al. 1994 ).

Isolation and sequencing of mtDNA:
Overlapping fragments of mtDNA were obtained by PCR that was run on total genomic DNA using degenerate primers. The sequences of the primers used for PCR amplification are given in Table 1. Two strategies for obtaining PCR-amplified DNA were used.

For PCR using the 1–18 primers (i.e., from the 16S rRNA to ND4), the conditions were the following: 300 ng total DNA; the enzyme was the Pfu Exo+ polymerase (Stratagene, La Jolla, CA); and the thermal cycles were 3 min at 94°, followed by 50 cycles 1 min at 94°, 1 min at 48–55°, depending upon the primers, and 5 min at 72°. The last cycle was ended by incubating for 10 min at 72°. These PCR products were purified by electroelution, phosphorylated using the polynucleotide kinase (Pharmacia Fine Chemicals, Piscataway, NJ), and ligated at the EcoRV site of dephosphorylated KS Bluescript vector (Stratagene) using the Rapid DNA Ligation kit (Boehringer Mannheim, Indianapolis, IN). XL1Blue competent bacteria (Stratagene) were transformed with the ligation products. Several recombinant clones were selected for each PCR product, and plasmid DNA was recovered using the Clearcut Miniprep kit (Stratagene). The cloned PCR products were sequenced first using the M13 (-40) and KS reverse primers and subsequently using primers derived from the sequence determined (Table 1). Thus, primer walking was used throughout, using primers located on both strands. Three hundred base pairs were read using each primer. The entire sequence was determined on both strands. Finally, the 16S rRNA was sequenced twice on both strands starting from two independent clones. All overlapping sequences were found to be identical.

For PCR using the VL1 and VL2 primers, namely for the isolation of the ND4-16S rRNA segment of the mtDNA, the conditions were the following: the Gene Amp XL PCR kit from Perkin Elmer-Cetus (Norwalk, CT) was used with the rTth DNA polymerase XL and the Amplivax system. To determine the frequency of errors, three independent amplifications were carried out, and all three products were sequenced. The PCR conditions were as follows. Oligonucleotides (200 ng) were mixed with 100–200 ng DNA/tube. A complex PCR cycle with autoextension was used: 1 min at 93°, followed by 1 min at 93° and 12 min at 68° for 16 cycles; this was followed by 20 cycles, 1 min at 93° and 12 min at 68°, with an autoextension of 15 sec/cycle after the sixth cycle. The last cycle was followed by 15 sec at 72°. The final PCR product was not cloned, but was directly sequenced on both strands by primer walking (300 bp in each walk) using the primers indicated in Table 1. Three independent PCR products were sequenced.

Phylogenetic analyses:
All protein-coding sequences of the mtDNA of S. canicula, Branchiostoma lanceolatum (N. SPRUYT, C. DELARBRE, G. GACHELIN and V. LAUDET, unpublished results), Crossostoma lacustre (TZENG et al. 1992 ), Cyprinus carpio (CHANG et al. 1994 ), Gallus gallus (DESJARDINS and MORAIS 1990 ), Latimeria chalumnae (ZARDOYA and MEYER 1997 ), Oncorhynchus mykiss (ZARDOYA et al. 1995 ), and P. marinus (LEE and KOCHER 1995 ) were aligned manually. The alignment was based on the translation of the nucleotide sequences into amino acid sequences using the vertebrates' mitochondrial genetic code, except for the AGA codon in B. lanceolatum, which is read as a glycine (DELARBRE et al. 1997 ). The alignment was refined using the physical amino acid similarity criterion. All inserted gaps were triplets and were treated as both missing data and a fifth nucleotide (additional state). Overlapping positions exist in the ATPase8/ATPase6, ND4L/ND4, and ND5/ND6 genes. They were used either as duplicates (giving these positions double the weight of the others) or as singles. All protein-coding genes of each sample were combined to create the data set.

The maximum parsimony (MP) method was applied to the data set using an exhaustive search implemented in PAUP (version 3.1.1; SWOFFORD 1993 ) to find the most parsimonious tree. Different ponderation schemes were used: transitions and transversions were given equal weight, or transversions were given double the weight of transitions. The neighbor-joining (NJ) method (SAITOU and NEI 1987 ) was used to reconstruct a tree based on the Kimura two-parameter distance matrix (KIMURA 1980 ; transversions were given double the weight of transitions) implemented in PHYLIP (version 3.5c; FELSENSTEIN 1993 ). The robustness of the MP and NJ trees were tested using the bootstrap method, with 200 replications each (FELSENSTEIN 1985 ), implemented in PHYLIP and using the decay index, i.e., Bremer support (BREMER 1988 ).

A Dayhoff distance matrix was computed using the Pam matrix (DAYHOFF 1978) implemented in PHYLIP. The same amino acid alignment described above was used.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Genomic organization of the mtDNA of S. canicula:
The mtDNA of S. canicula (GenBank accession number Y16067) is 16,697 bp long. The 13 protein-coding genes, the 22 tRNA genes, and the two rRNA genes were identified through the zones of sequence similarity with the corresponding genes and encoded proteins of the carp, C. carpio (CHANG et al. 1994 ).

The gene order of the S. canicula mtDNA is depicted in Table 2. The complete nucleotide and deduced amino acid sequences are given in Figure 1.

View larger version (330K): In this window In a new window Download PPT slide   Figure 1. Complete nucleotide and encoded amino acid sequences of the mtDNA of S. canicula. As in Table 2, the nucleotides are numbered starting from the first nucleotide of the gene coding for the ND1 protein. The name of each gene is placed at the 5' of the coding sequence and above it. The limits of the genes are indicated by bars. The initiation codons and the tRNA anticodons are underlined. The stop codons are indicated by an asterisk. The putative regulatory sequences of the control are indicated above the corresponding nucleotide sequences, which are also underlined. The arrowheads indicate the orientation of the tRNAs.

Irrespective of minor differences in the size of several genes, the organization of S. canicula mtDNA was found to be identical to that of the bony fish, the complete mtDNA sequence of which is known [namely C. carpio (CHANG et al. 1994 ), Protopterus dolloi (ZARDOYA and MEYER 1996 ), Polypterus ornatipinnis (NOACK et al. 1996 ), L. chalumnae (ZARDOYA and MEYER 1997 ), Gadus morhua (JOHANSEN and BAKKE 1996 ), O. mykiss (ZARDOYA et al. 1995 ), and C. lacustre (TZENG et al. 1992 )]. The orientation of the genes also is the same as in the Osteichthyes: all coding sequences are located on the H strand, with the exception of the ND6 gene and the eight tRNA genes, which are located on the L strand (Table 2).

By contrast, three main features make the gene organization of S. canicula different from that of the agnathan P. marinus (LEE and KOCHER 1995 ):

1. The 1050-bp-long control region of S. canicula (about as long as in the carp, 928 bp; the lungfish, 1184 bp; and the trout, 1000 bp) is located between tRNA -Pro and -Phe genes, whereas the structurally different 675-bp-long control region of P. marinus (LEE and KOCHER 1995 ) is split into two regions, NC1 and NC2, respectively, by the insertion of two tRNA genes (Thr and Glu). It is located between the ND6 and CYTb genes.

2. As already described (DELARBRE et al. 1997 ), the origin of replication of the L strand of the mtDNA is localized between the tRNA-Cys and tRNA-Asn genes of S. canicula, but could not be identified in P. marinus.

3. The tRNA-Thr gene is located 3' to the CYTb gene in S. canicula, whereas in P. marinus, it is located in the middle of the control region, adjacent to the tRNA-Glu gene. The tRNA-Glu of S. canicula lies between the ND6 and the CYTb genes.

Salient features of the control region:
Several structural elements were evidenced in the 1050-bp-long control region; they are identified in Figure 1. These structural elements were two hairpin-like regions (hairpins 1 and 2), a 10-bp-long perfect palindrome (12893–12902), two 23-bp-long repeats (repeats 1 and 2), and three putative conserved sequence blocks [CSB I, II, and III, (WALBERG and CLAYTON 1981 )]. Several putative termination-associated sequences (TAS; DODA et al. 1981 ) were identified. Among them, the sequence ACATTAAACTAAT (13433–13446) is nearly identical to mouse consensus TAS and TAS-4 of the bichir, and the sequence ACATACTATG (12999–13008) is nearly identical to TAS-3 of the bichir. The AT-rich sequence of P. marinus [non-coding region 2 (LEE and KOCHER 1995 )] located between tRNA-Glu and CYTb was not identifiable in either S. canicula or the Osteichthyes. The presence of the origin of replication of the L strand in S. canicula had already been reported (DELARBRE et al. 1997 ). The presence/absence of these control elements makes the organization of the control region of S. canicula very similar to that of the Osteichthyes (JOHANSEN and BAKKE 1996 ).

Lack of distinctive traits of the S. canicula tRNAs:
The 22 tRNA genes were identified on the basis of their location and sequence. They were from 67 to 75 bp long. All of them, except for tRNA-Ser (AGY), were able to fold into a cloverleaf secondary structure. There is no feature that would make them different from the tRNAs of osteichthyans. As shown in Table 2, most of the tRNA genes are located on the H strand, with the exception of the tRNA-Glu, -Pro, -Gln, -Ala, -Asn, -Ser (TCN), -Cys, and -Tyr, which are on the L strand.

Distinctive features of the protein-coding sequences and of the encoded proteins:
The complete nucleotide sequence of the mtDNA of S. canicula is depicted in Figure 1, as well as the translation of the open reading frames using the vertebrate mitochondrial code and the limits of the genes.

The codon usage was determined (Table 3). We have compared the codon usage of S. canicula with that of P. marinus (LEE and KOCHER 1995 ) and C. carpio (CHANG et al. 1994 ). There are no major differences between the three codon usages. In all three animals, the TTG and GTG codons were predominantly used in ND6. Some codons preferentially used by S. canicula are also preferentially used by P. marinus: for example, the codon for Phe is predominantly TTT in P. marinus and S. canicula, whereas it is TTC for C. carpio. Conversely, among the codons coding for Ser, AGC is preferentially used in C. carpio and S. canicula, whereas AGT is predominant for Petromyzon. The codon TCG for Ser is very rarely used in S. canicula. Overall, the minor preferences shown by the codon usage of S. canicula are rather on the "Petromyzon side." The initiation codons were ATG for all genes, with the exception of GTG for the COI gene, as in most chordates studied so far. The stop codon for the ND6 gene was TAG. All other stop codons were TAA. The TAA stop codons of COII, ND4, and cytochrome b mRNAs should have been created by direct polyadenylation on a terminal T (OJALA et al. 1981 ). The same stop codons are used by the carp and rainbow trout. AGA is not used as a stop codon in S. canicula, as it is in the carp, trout, and cod. In some other fish, the AGA or AGG are used as stop codons, once in L. chalumnae and P. dolloi and twice in P. ornatipinnis. In P. marinus, the AGA is used five times as a stop codon.

Concerning the nature of the codons, we noted the comparatively rare usage of G at the third position in the fourfold degenerate codons (5.8%), as in most vertebrates, a low value that contrasts with the 12 and 24% observed in the hagfish and lancelet, respectively [based on the nucleotide sequences of ND1, ND2, ATPase 8, and ATPase 6 genes (DELARBRE and GACHELIN 1997 ; DELARBRE et al. 1997 )]. The A nucleotide is the most used nucleotide at the third position (39.7%). The figures for T and C were 27.8 and 26.7%, respectively.

The coding sequences of several proteins were found to partly overlap, as in most species studied. The sequence coding for ATPase 8 overlaps by 10 nucleotides with the coding for ATPase 6 as in the sea lamprey, lungfish, bichir, coelacanth, and chicken, whereas the overlap is seven nucleotides in the lancelet (DELARBRE and GACHELIN 1997 ) and carp. The sequences coding for the ND4L and ND4 proteins overlap by seven nucleotides as in all chordates studied so far. Finally, the ND5/ND6 coding genes overlap by four nucleotides as in the lungfish, carp, and sea urchin, whereas the overlap is as long as 16 nucleotides in the sea lamprey.

The overall amino acid usage has also been determined. It shows no preferential usage of any given amino acid, compared to the sea lamprey and several teleosts. This is in contrast with preferential usages already reported in the partial sequences of the hagfish and lancelet (DELARBRE et al. 1997 ).

Intragroup and intergroup similarities, as defined by the sequences of the cytochrome b protein:
No complete mtDNA sequence of other chondrichthyans is available so far. However, the sequence of the cytochrome b gene has been determined in numerous species, including other elasmobranchs. This allowed a more precise analysis of the relation of S. canicula to different groups of fish. Thus, the deduced amino acid sequence of the cytochrome b of S. canicula was aligned with that of several chondrichthyans, to check for the homogeneity of the Chondrichthyes group, then with that of the agnathan sea lamprey and a subset of bony fishes. The percentage of identity of the amino acid sequences was 83, 87, 87, and 83% with Heterodontus francisci, Sphyrna tiburo vespertina, Carcharinus plumbeus, and Carcharodon carcharias (MARTIN and PALUMBI 1993 ), respectively. It was 67% with P. marinus (LEE and KOCHER 1995 ) and 78, 77, 77, 75, 74, and 76% for C. carpio (CHANG et al. 1994 ), O. mykiss (ZARDOYA et al. 1995 ), C. lacustre (TZENG et al. 1992 ), P. dolloi (ZARDOYA and MEYER 1996 ), P. ornatipinnis (NOACK et al. 1996 ), and L. chalumnae (ZARDOYA and MEYER 1997 ), respectively. The percentage of sequence similarity among cytochrome b amino acid sequences between chondrichthyans was 87%. It was 92% between the carp and loach and 89% between the carp and rainbow trout. On the basis of the cytochrome b amino acid sequences, the mtDNA of the dogfish thus behaves as a "standard" chondrichthyan mtDNA, but once more appears to be much closer to the Osteichthyes than to the Agnatha.

Intergroup similarities of all mitochondrial proteins:
Considering that S. canicula is representative of the Chondrichthyes, we have aligned, with minimal insertions/deletions, the deduced amino acid sequences of all proteins of S. canicula with the corresponding proteins of several animals ranging from cephalochordates to birds (alignments are available upon request). The rate of evolution of the amino acid sequences differed for each protein. Considering the two extremes, the COI sequence was found to be extremely well conserved throughout evolution, whereas the ATPase 8 was the least conserved (data not shown). A distance Dayhoff matrix was computed (Table 4). It showed a low similarity of S. canicula mitochondrial proteins with those of a cephalochordate (B. lanceolatum) and an agnathan (P. marinus). The sequence similarity of the S. canicula mitochondrial proteins was higher with those of actinopterygians (C. carpio, C. lacustre, and O. mykiss) and an actinistian (L. chalumnae), than with those of a tetrapod (G. gallus).

Phylogenetic position of the dogfish:
The final data set was composed of eight taxa and 11,532 sites when overlapping positions are duplicated, with 7465 variable sites, including 4888 informative sites. The lancelet was used as an outgroup. Using the MP method without weighting (equal weight given to transitions and transversions) and gaps treated as missing data, a single most-parsimonious tree was obtained (Figure 2A, length = 18,171 steps, C.I. = 0.685, R.I. = 0.311). If gaps were treated as a fifth state, the same resulting tree was computed (length = 18,665 steps, C.I. = 0.690, R.I. = 0.311). When transversions were given double the weight of transitions, the same result was obtained, and any gaps were treated as missing data (length = 27,746 steps) or as a fifth nucleotide (length = 27,943 steps). Thus, the dogfish appeared as the sister group of the Osteichthyes (sarcopterygians and actinopterygians). The node was supported by a 100% bootstrap value and a 202 Bremer index. Using the NJ method, the NJ tree topology was found to be identical to the MP tree topology (Figure 2B). The position of the dogfish was also supported by a 100% bootstrap value. All internal branches appeared short in comparison to terminal branches. This holds particularly true for the chicken; in the same way, the minimal and maximal number of autapomorphies of the chicken were high compared to other taxa.

View larger version (16K): In this window In a new window Download PPT slide   Figure 2. Phylogenetic position of S. canicula. (A) The most parsimonious tree (PAUP 3.1.1, exhaustive search, unweighted parsimony, gaps = missing data) computed out of the 11,532 aligned positions. Tree length = 18,171 steps, C.I. = 0.685, R.I. = 0.311. The values above the branches indicate either synapomorphies or autapomorphies, according to the optimization of informative sites (table of linkage); numbers below the branches indicate the Bremer support (italics) and the bootstrap proportion (bold). (B) Phenogram constructed using the NJ method (Kimura's two-parameter distance, PHYLIP 3.5c). The numbers placed at the nodes indicate bootstrap proportion. The length of the branches is proportional to divergence between taxa (ranging from 0 to 1). A divergence of 0.1 corresponds to the length of the bar below the figure.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The primary conclusion reached from the study of the mtDNA of S. canicula is its high similarity with the mtDNA of the Osteichthyes: identical genomic organization, same loop of replication of the L strand, same codon usage, and high similarity of the amino acid sequences. Conversely, the genome organization and the DNA and amino acid sequences make the mtDNA of P. marinus very different from that of S. canicula. When studying the ND1-COI sequence, we had already suggested that some traits of the archetypal organization of the mtDNA of the Osteichthyes were present in the Chondrichthyes. The present data demonstrate that the overall organization and, presumably, the ways of the replication of the mtDNA of osteichthyans, were established in the common ancestor to chondrichthyans and osteichthyans.

The phylogenetic position of the dogfish we have found corresponds to the admitted consensus concerning the phylogeny of Recent chordates, with the chondrichthyans as the sister group of osteichthyans: a unique and identical tree was obtained using either MP or NJ. The result obtained using these eight taxa, which are relevant to the problem of the phylogenetic position of the dogfish, was reached using nearly all codons, in contrast to the need for a selection of a subset of codons pointed out by others (NAYLOR and BROWN 1998 ).

However, the search for identity at the amino acid sequence level reveals a much higher similarity of the dogfish with teleosts and actinistians. This result does not, however, conflict with the phylogenetic position of the dogfish. The overall identity of amino acid sequences does not necessarily indicate close relationships. The different rates of evolution of mtDNA can account for the observed similarities between the amino acid sequences of the dogfish and those of teleosts. Indeed, the substitution rate of mtDNA is slow in sharks and teleosts and faster in homeothermic species (THOMAS and BECKENBACH 1989 ; MARTIN et al. 1992 ; ADACHI et al. 1993 ).

The results of the phylogenetic analysis and of the amino acid sequences, however, must be considered in the light of the paleontological record. The current consensus about extant jawed vertebrate (gnathostome) interrelationships is chondrichthyans (including elasmobranchs and holocephalans) are the sister group of the osteichthyans (including actinopterygians and sarcopterygians). The monophyly of fossils and Recent chondrichthyans is supported by six morphological characters (MAISEY 1986 ), but only one of them (the presence of prismatic calcified cartilage) is unambiguous or nonhomoplastic. The monophyly of fossil and recent osteichthyans is supported by at least 18 unambiguous morphological characters (MAISEY 1986 ). The fact that the amino acid sequence of chondrichthyans (at any rate an elasmobranch) is closer to that of actinopterygians than to that of lampreys is no surprise, and it provides further support for the monophyly of the gnathostomes, which is already supported by at least 37 morphological characters (JANVIER 1996A , JANVIER 1996B ; MAISEY 1986 ). The minimum age provided by paleontological data for the divergence between lampreys and gnathostomes is ~470 million years, but the minimum age for the divergence of the gnathostome taxa is practically identical because the earliest known chondrichthyans, actinopterygians, and sarcopterygians are all late Silurian in age (~410 million years).

In the framework of the current gnathostome phylogeny, the similarity between the dogfish and teleost amino acid sequences results from either homoplasy or symplesiomorphy. In the latter case, the close similarity between the dogfish and teleost sequences reflects the general (ancestral) condition for Recent gnathostomes.

	FOOTNOTES

¹ These authors contributed equally to this work.

	ACKNOWLEDGMENTS

The work carried out in Paris has been supported in part by grants from the Institut National de la Santé et de la Recherche Médicale (INSERM). The work carried out in Lille has been supported by grants from the Institut Pasteur de Lille and the Centre National de la Recherche Scientifique (CNRS).

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ADACHI, J., Y. CAO, and M. HASEGAWA, 1993  Tempo and mode of mitochondrial DNA evolution in vertebrates at the amino acid sequence level: rapid evolution in warm-blooded vertebrates. J. Mol. Evol. 36:270-281[Medline].

BREMER, K., 1988  The limits of amino acid sequence data in angiosperm phylogenetic reconstruction. Evolution 42:795-803.

CHANG, Y.-C., F.-L. HUANG, and T.-B. LO, 1994  The complete nucleotide sequence and gene organization of carp (Cyprinus carpio) mitochondrial genome. J. Mol. Evol. 38:138-155[Medline].

DAYHOFF, M. O., R. M. SCHWARTZ and B. C. ORCUTT, 1978 Atlas of Protein Sequence and Structure, Vol. 5, pp. 342–352. National Biomedical Research Foundation, Washington, DC.

DELARBRE, C. and G. GACHELIN, 1997  A unique cDNA coding for subunits 8 and 6 of mitochondrial adenosine triphosphatase of the lancelet Branchiostoma lanceolatum, an ancestor of vertebrates. Biochem. Gen. 35:181-187[Medline].

DELARBRE, C., V. BARRIEL, S. TILLIER, P. JANVIER, and G. GACHELIN, 1997  The main features of the craniate mitochondrial DNA between the ND1 and the COI genes were established in the common ancestor to the lancelet. Mol. Biol. Evol. 14:807-813[Abstract].

DESJARDINS, P. and R. MORAIS, 1990  Sequence and gene organization of the chicken mitochondrial genome. A novel gene order in higher vertebrates. J. Mol. Biol. 212:599-634[Medline].

DODA, J. N., C. T. WRIGHT, and D. A. CLAYTON, 1981  Elongation of displacement loop strands in human and mouse mitochondrial DNA is arrested near specific template sequences. Proc. Nat. Acad. Sci. USA 78:6116-6120[Abstract/Free Full Text].

FELSENSTEIN, J., 1985  Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:783-791.

FELSENSTEIN, J., 1993 PHYLIP (Phylogeny Inference Package), version 3.5c. Computer program distributed by the author. Department of Genetics, University of Washington, Seattle.

HOGAN, B., R. BEDDINGTON, F. COSTANTINI and E. LACY, 1994 Isolating high molecular weight DNA from mouse tails, in Manipulating the Mouse Embryo: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

JANVIER, P., 1996a Early vertebrates, in Oxford Monographs in Geology and Geophysics, No. 33. Oxford University Press, Oxford.

JANVIER, P., 1996b  The dawn of vertebrates: characters versus common ascent in the rise of current vertebrate phylogenies. Paleontology 39:259-287.

JOHANSEN, S. and I. BAKKE, 1996  The complete mitochondrial DNA sequence of Atlantic cod (Gadus morhua): relevance to taxonomic studies among codfishes. Mol. Marine. Biol. Biotechnol. 5:203-214[Medline].

KIMURA, M., 1980  A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J. Mol. Evol. 16:111-120[Medline].

KUMAZAWA, Y. and M. NISHIDA, 1995  Variations in mitochondrial tRNA gene organization of reptiles as phylogenetic markers. Mol. Biol. Evol. 12:759-772[Abstract].

LEE, W.-J. and T. D. KOCHER, 1995  Complete sequence of a sea lamprey (Petromyzon marinus) mitochondrial genome: early establishment of the vertebrate genome organization. Genetics 139:873-887[Abstract].

MAISEY, J. G., 1986  Heads and tails: a chordate phylogeny. Cladistics 2:201-256.

MARTIN, A. P. and S. R. PALUMBI, 1993  Protein evolution in different cellular environments: cytochrome b in sharks and mammals. Mol. Biol. Evol. 10:873-891[Abstract].

MARTIN, A. P., G. J. P. NAYLOR, and S. R. PALUMBI, 1992  Rates of mitochondrial DNA evolution in sharks are slow compared with mammals. Nature 357:153-155[Medline].

NAYLOR, G. J. P. and W. M. BROWN, 1998  Amphioxus mitochondrial DNA, chordate phylogeny, and the limits of inference based on comparisons of sequences. Syst. Biol. 47:61-76[Medline].

NOACK, K., R. ZARDOYA, and A. MEYER, 1996  The complete mitochondrial DNA sequence of the Bichir (Polypterus ornatipinnis), a basal ray-finned fish: ancient establisment of the consensus vetebrate gene order. Genetics 144:1165-1180[Abstract].

OJALA, D., J. MONTOYA, and G. ATTARDI, 1981  tRNA punctuation model of RNA processing in human mitochondria. Nature 290:470-474[Medline].

SAITOU, N. and M. NEI, 1987  The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol. Evol. 4:406-425[Abstract].

SWOFFORD, D. L., 1993 PAUP: Phylogenetic Analysis Using Parsimony, version 3.1.1. Computer program distributed by the Illinois Natural History Survey, Champaign, IL.

THOMAS, W. K. and A. T. BECKENBACH, 1989  Variation in salmonid mitochondrial DNA: evolutionary constraints and mechanisms of substitution. J. Mol. Evol. 29:233-245[Medline].

TZENG, C.S., C.F. HUI, S.C. SHEN, and P.C. HUANG, 1992  The complete nucleotide sequence of the Crossostoma lacustre mitochondrial genome, conservation and variation among vertebrates. Nucleic Acids Res. 20:4853-4858[Abstract/Free Full Text].

WALBERG, M. W. and D. A. CLAYTON, 1981  Sequence and properties of the human KB cell and mouse L cell D-loop regions of mitochondrial DNA. Nucleic Acids Res. 9:5411-5421[Abstract/Free Full Text].

ZARDOYA, R. and A. MEYER, 1996  The complete nucleotide sequence of the mitochondrial geome of the lungfish (Protopterus dolloi) supports its phylogenetic position as a close relative of land vertebrates. Genetics 142:1249-1263[Abstract].

ZARDOYA, R. and A. MEYER, 1997  The complete DNA sequence of the mitochondrial genome of a "living fossil," the coelacanth (Latimeria chalumnae). Genetics 146:995-1010[Abstract].

ZARDOYA, R., A. GARRIDO-PERTIERRA, and J. M. BAUTISTA, 1995  The complete nucleotide sequence of the mitochondrial DNA genome of the rainbow trout, Oncorhynchus mykiss.. J. Mol. Evol. 41:942-951[Medline].

This article has been cited by other articles:

				A.-S. Rasmussen and U. Arnason Molecular studies suggest that cartilaginous fishes have a terminal position in the piscine tree PNAS, March 2, 1999; 96(5): 2177 - 2182. [Abstract] [Full Text] [PDF]