The Mitochondrial Genome of the Lizard Calotes versicolor and a Novel
Gene Inversion in South Asian Draconine Agamids
Sayed A. M. Amer* and Yoshinori Kumazawa*
*Graduate School of Science, Nagoya University, Nagoya, Japan; and Zoology Department, Faculty of Science,
Cairo University, Giza, Egypt
A complete mitochondrial DNA (mtDNA) sequence was determined for the lizard Calotes versicolor (Reptilia;
Agamidae). The 16,670-bp genome with notable shorter genes for some protein-coding and tRNA genes had the same
gene content as that found in other vertebrates. However, a novel gene arrangement was found in which the proline
tRNA (trnP) gene is located in the light strand instead of its typical heavy-strand position, providing the first known
example of gene inversion in vertebrate mtDNAs. A segment of mtDNA encompassing the trnP gene and its flanking
genes and the control region was amplified and sequenced for various agamid taxa to investigate timing and mechanism
of the gene inversion. The inverted trnP gene organization was shared by all South Asian draconine agamids examined
but by none of the other Asian and African agamids. Phylogenetic analyses including clock-free Bayesian analyses for
divergence time estimation suggested a single occurrence of the gene inversion on a lineage leading to the draconine
agamids during the Paleogene period. This gene inversion could not be explained by the tandem duplication/random loss
model for mitochondrial gene rearrangements. Our available sequence data did not provide evidence for remolding of the
trnP gene by an anticodon switch in a duplicated tRNA gene. Based on results of sequence comparisons and other
circumstantial evidence, we hypothesize that inversion of the trnP gene was originally mediated by a homologous DNA
recombination and that the de novo gene organization that does not disrupt expression of mitochondrial genes has been
maintained in draconine mtDNAs for such a long period of time.
Introduction
The typical vertebrate mitochondrial genome
(mtDNA) is approximately 16–18 kbp in size, containing
37 genes for 13 proteins, 2 rRNAs and 22 tRNAs, as well
as a major noncoding region or control region (CR) including sites for the initiation of transcription and replication
(Anderson et al. 1981). Complete or nearly complete
mtDNA sequences have been reported from several hundreds of vertebrate taxa dominated by placental mammals
and fishes (see, e.g., Jameson et al. 2003). Gene organization of animal mtDNAs occasionally changes, and such
gene rearrangements could provide some phylogenetic information. Gene rearrangements so far reported for vertebrate mtDNAs are most typically, but not exclusively,
local rearrangements among clustering tRNA and protein
genes, as well as tandem duplication of a continuous
mtDNA segment (see, e.g., Moritz et al. 1987; Stanton
et al. 1994; Macey et al. 1997; Mindell et al. 1998; Inoue
et al. 2001; Kumazawa and Endo 2004; Mueller and Boore
2005; Kurabayashi et al. 2006; San Mauro et al. 2006).
However, none of these cases accompany a gene inversion
in which a gene switched its encoded strand. On the other
hand, the gene inversion is common in invertebrate
mitochondrial genomes (Boore 1999), suggesting some unknown difference in evolutionary constraints in gene
rearrangements between vertebrate and invertebrate
mtDNAs.
Agamidae, one of lizard families, is distributed in tropical and subtropical regions of Eurasia, Africa, and Australasia (Pough et al. 2004, p. 122–125). Based on molecular
and morphological data, Macey et al. (2000) grouped the
Agamidae into 6 subfamilies: Uromastycinae (genus Uromastyx distributed in Africa–South Asia), Leiolepidinae
Key words: Calotes versicolor, Agamidae, mitochondrial genome,
gene inversion, homologous DNA recombination, gene rearrangement.
E-mail: [email protected].
Mol. Biol. Evol. 24(6):1330–1339. 2007
doi:10.1093/molbev/msm054
Advance Access publication March 22, 2007
Ó The Author 2007. Published by Oxford University Press on behalf of
the Society for Molecular Biology and Evolution. All rights reserved.
For permissions, please e-mail: [email protected]
(Leiolepis in Southeast Asia), Amphibolurinae (Australian–
New Guinean clade plus Southeast Asian Physignathus
cocincinus), Hydrosaurinae (Hydrosaurus in Indonesia
and Philippine), Agaminae (African–West Asian clade),
and Draconinae (South or Southeast Asian clade). Monophyly of each agamid subfamily has been corroborated
by some morphological (Moody 1987) and molecular
(e.g., Honda et al. 2000; Macey et al. 2000; Amer and
Kumazawa 2005a) studies.
Complete or nearly complete mtDNA sequences
have been reported from relatively fewer taxa of lizards
(Kumazawa and Nishida 1999; Janke et al. 2001;
Kumazawa 2004, 2007; Kumazawa and Endo 2004; Amer
and Kumazawa 2005b; Macey et al. 2005, 2006; Zhou
et al. 2006), among which 2 are from Agamidae: the amphibolurine Pogona vitticepes (Amer and Kumazawa 2005b)
and the agamine Xenagama taylori (Macey et al. 2006).
Because some novel gene rearrangements have been described in agamid mtDNAs (Macey et al. 1997, 2006;
Amer and Kumazawa 2005b), agamid mtDNAs may represent one of the most frequently rearranged mitochondrial
genomes among vertebrates. It thus would be intriguing to
explore more gene rearrangements in diverse agamid lineages for understanding molecular evolutionary processes
and mechanisms of gene rearrangements in animal
mtDNAs.
Here, we report the complete mtDNA sequence of
Calotes versicolor representing the 3rd agamid subfamily
Draconinae. This genome has a novel gene order regarding
the proline tRNA gene (trnP). Unlike mtDNAs from any
other vertebrates studied so far, the C. versicolor mtDNA
had a trnP gene encoded by the heavy strand. This finding
prompted us to examine the occurrence of a similar gene
arrangement in other South Asian agamids. By estimating
reliable phylogenetic relationships and divergence times,
we discuss the timing and possible mechanism of the gene
inversion, which should contribute to better understanding
of molecular evolutionary mechanisms operating on vertebrate mitochondrial genomes.
Gene Inversion in C. versicolor mtDNA 1331
FIG. 1.—(A) Gene organization of Calotes versicolor mtDNA shown linearly with columns which approximate sizes of individual genes, in which
the inversion of the trnP gene is highlighted. (B) Gene organization around the CR for Xenagama taylori (Macey et al. 2006) in which the trnP gene is
translocated from the 5#end to the 3#end of the CR. (C) Gene organization around the CR for Agama impalearis in which the trnF gene is translocated
from the 3#end to the 5#end of the CR. Note that the CR was not completely sequenced for A. impalearis (see table 1). Heavy-strand–encoded and lightstrand–encoded genes are shown above and below a column, respectively. Primers used for amplification and sequencing of the trnP-containing region
(rcytb-1L and r12S-1H), as well as those used for the PCR assay (rGlu-1L and rPro-1L) are shown for their location and orientation. Gene abbreviations
used are: rrnS and rrnL, 12S and 16S rRNAs; nad1–6, NADH dehydrogenase subunits 1–6; cox1–3, cytochrome oxidase subunits I–III; atp6and atp8,
ATPase subunits 6 and 8; cob, cytochrome b; and one-letter codes of amino acids, tRNA genes specifying them (L1 and L2 for leucine tRNA genes
specifying, respectively, UUR and CUN codons and S1 and S2 for serine tRNA genes specifying, respectively, AGY and UCN codons). CR, OH, and
OL stand for the control region, the heavy-strand replication origin, and the light-strand replication origin, respectively.
Material and Methods
Sequencing and Gene Identification
Total DNA was extracted from C. versicolor muscle tissues and used as a template to amplify and sequence parts of
the genes for cytochrome b (cob) and reduced form of nicotinamide adenine dinucleotide dehydrogenase subunit 2
(nad2) with conserved primers described by Kumazawa
and Endo (2004). These sequences were used for designing
taxon-specific primers (see supplementary table S1 [Supplementary Material online] for their sequences and locations)
for the long and accurate (LA)–polymerase chain reaction
(PCR) amplifications of spacing regions between them.
The amplified products (5 and 10 kb in length) were purified
by an agarose gel electrophoresis and used as templates for
nested PCRs (product length: 0.5–1 kb) that were conducted
with a set of conserved reptile-oriented primers (Kumazawa
and Endo 2004) and new taxon-specific primers (see supplementary table S1, Supplementary Material online). Amplified fragments by the nested PCR were directly sequenced
for both strands with the amplification primers or appropriate
internal primers, which were then assembled into a contiguous mtDNA sequence. For typical conditions of the LAPCR, the nested PCR, and the sequencing reaction, see Amer
and Kumazawa (2005b). The C. versicolor mtDNA sequence has been deposited in the DDBJ/EMBL/GenBank
databases with an accession number AB183287. Individual
genes and features in it (see supplementary table S2, Supplementary Material online) were identified in light of their
counterparts from other vertebrates as described by Amer
and Kumazawa (2005b).
Characterization of the trnP-Containing Region in Other
Agamids
An mtDNA fragment encompassing the trnP gene and
parts of the surrounding genes and the CR was successfully
amplified for various agamid taxa from South or Southeast
Asia (Acanthosaura capra, Acanthosaura crucigera, Calotes cf. calotes, Calotes cf. emma, Draco sp., Gonocephalus grandis, Hydrosaurus cf. weberi, Leiolepis cf. reevesi,
and Physignathus cocincinus) and West Asia–Africa (Agama impalearis, and Uromastyx benti) using primers rcytb1L and r12S-1H (Kumazawa and Endo 2004) (see supplementary table S3 [Supplementary Material online] for their
sequences and figure 1 for their locations and orientations).
The amplified products being approximately 2 kbp were
then sequenced with the above-mentioned primers and
newly designed taxon-specific internal primers (see supplementary table S3, Supplementary Material online). These
sequences have been deposited in the DDBJ/EMBL/GenBank databases with accession numbers shown in table 1.
A quick PCR survey was also conducted for the presence/absence of the trnP gene inversion in more agamid
taxa available in our laboratory. Primers rGlu-1L and
rPro-1L (see supplementary table S3, Supplementary Material online) were used for this assay. PCR products being
approximately 1.2 kbp would be obtained if the inversion
exists in the corresponding mtDNA (see fig. 1A).
Phylogenetic Analyses
Nucleotide sequences of an mtDNA segment between
nad1 and cytochrome oxidase subunit I (cox1) genes for
various agamid taxa were collected from the database
(see supplementary table S4 [Supplementary Material online] for accession numbers) and used for phylogenetic
analyses after unalignable and gap-containing sites were
deleted (1,490 bp in total). The aligned nucleotide sequences can be obtained from S.A.M.A. upon request. The phylogenetic analyses were done primarily by maximum
likelihood (ML) method with PAUP* 4.0b10 (Swofford
2003) by heuristic searches with the NNI branch swapping
1332 Amer and Kumazawa
Table 1
Agamid Taxa Used in This Study with Their Status for Proline tRNA Gene
Accession Number
Museum Number
trnP Status
Uromastycinae
Leiolepidinae
Amphibolurinae
Uromastyx benti
Leiolepis cf. reevesi
Pogona vitticepes
AB266887
AB266888
AB166795
NSMT-H4670
NUM-Az389
NUM-Az383
Normal
Normal
Normal
Hydrosaurinae
Agaminae
Physignathus cocincinus
Hydrosaurus cf. weberi
Agama impalearis
AB263945
AB263944
AB271238a and
AB271239a
DQ008215
AB266884
AB263946
AB263941
AB263942
AB183287
AB263943
AB263939
NUM-Az386
NSMT-H4602
—b
Normal
Normal
Translocated
This study
This study
Amer and Kumazawa
(2005b)
This study
This study
This study
CAS 225502
NSMT-H4596
NSMT-H4597
NSMT-H4599
NSMT-H4598
NUM-Az382
NSMT-H4600
NSMT-H4601
Translocated
Inverted
Inverted
Inverted
Inverted
Inverted
Inverted
Inverted
Macey et al. (2006)
This study
This study
This study
This study
This study
This study
This study
Subfamily
Draconinae
Taxon
Xenagama taylori
Acanthosaura capra
Acanthosaura crucigera
Calotes cf. calotes
Calotes cf. emma
Calotes versicolor
Draco sp.
Gonocephalus grandis
Reference
NOTE.—Accession numbers refer to the sequences encompassing trnP gene for these taxa. NUM, NSMT, and CAS stand for respectively, Nagoya University Museum,
National Science Museum, Tokyo, and California Academy of Sciences, San Francisco. Normal and inverted status for trnP stand for (cob–trnT–trnP–CR–trnF–rrnS) and
(cob–trnT–trnP–CR–trnF–rrnS), respectively. For 2 different translocated organizations involving the trnP gene, see figure 1.
a
Sequencing is incomplete in the middle of the major noncoding region.
b
Museum voucher numbers are not available because DNA was extracted from tail tissues cut from live individuals.
and 10 random taxon additions. The general reversible
model (GTR þ I þ G) and parameters optimized by Modeltest 3.0 (Posada and Crandall 1998) were used. Bootstrapping replicates were set to 500.
Assuming the topological relationships thus obtained,
divergence times at nodes were estimated using the multidistribute package (Thorne et al. 1998) after optimizing parameters for the model F84 and the gamma distribution for
8 categories by PAML (Yang 1997). We then used estbranches/multidivtime programs to estimate divergence
times by Bayesian Markov Chain Monte Carlo method
without assuming the molecular clock. Some key parameters used for this were the following: numsamps, 10,000;
sampfreq, 100; burnin, 100,000; rttm and rttmsd, 1.0
(see Kumazawa 2007); rtrate and rtratesd, 0.6; brownmean
and brownsd, 0.4; and minab, 1.0. Two iguanid taxa were
used as an outgroup. In the multidivtime analyses, we used
2 kinds of constraints for the nodal divergence times. Assuming that Laudakia microlepis of Zagros vicariantly diverged from northern Laudakia caucasia 9 MYA due to
mountain uplifts caused by the Indian and Arabian collision
(Macey et al. 1998), we constrained this divergence to be 8–
10 MYA. We also constrained the divergence between
Agamidae and Chamaeleonidae to be 123–76 MYA. This
second constraint is based on independent molecular time
estimates for this divergence using complete mtDNA sequences including other reptiles, mammals, birds, and amphibians (Kumazawa 2007).
Results and Discussion
Genomic Features of C. versicolor mtDNA
The complete mtDNA sequence of C. versicolor determined as described in Material and Methods was
16,670-bp long and had the usual set of 37 genes and a major noncoding region (fig. 1A). Genes for 2 rRNAs, 13 pro-
teins, and 20 out of 22 tRNAs were encoded in the same
relative position and orientation as found in other vertebrate
mtDNAs reported so far. One of the other tRNA genes
specifying glutamine was present at the 5# end of the isoleucine tRNA gene rather than at the 3# end of it in the typical vertebrate gene organization. This is a gene
arrangement that occurs widely in Agamidae and Chamaeleonidae (Macey et al. 1997, 2000).
The remaining tRNA gene specifying proline conserved its relative position in the mtDNA gene organization
but was encoded by the heavy strand, rather than by the
light strand as found for other vertebrates. Thus, a gene inversion for trnP likely occurred on an agamid lineage leading to C. versicolor by perfectly conserving organizations
for 36 other genes and the major noncoding region (fig. 1A).
To the best of our knowledge, this is the first example of
gene inversion in vertebrate mtDNAs after .600 complete
mtDNA sequences were reported from various groups of
vertebrates (Jameson et al. 2003).
The major noncoding region between the trnP and
trnF genes (1,504 bp in size) harbored 6 direct repeats
of a 66-bp unit and an extremely AT-rich region in the right
part close to trnF. Intensive search for possible cloverleaf
structures in the left part of the major noncoding region
failed to find any functional tRNA genes in it. This major
noncoding region containing motifs of the conserved sequence blocks I and III (see supplementary table S2, Supplementary Material online) may be regarded as the CR for
replication and transcription (Clayton 1992). The base composition of the C. versicolor mtDNA is skewed similarly as
in other vertebrate mtDNAs with more A-T base pairs than
G-C base pairs and much more A and C contents in the light
strand than in the heavy strand (Asakawa et al. 1991). The
overall base composition of the light-strand sequences is A,
33.8%; T, 26.0%; C, 27.0%; and G, 13.3%.
Gene Inversion in C. versicolor mtDNA 1333
3'
A
G
T
T
T
T
C
T
TCTA A C
AGAT G A
T
A
A A A CA
A
A A
C G A
T
T A
G C
G C
C T
T
T
T
A
T G G
5' C
A
A
A
A
G
A
T
A
A A TTTG
5' C
A
A
A
A
G
A
T
A
A
A
C
TTTG
A
A A A A A A C AT
AGGT C A
T
A
A G
T
T A
G C
G C
C T
T
T
T
A
T G G
Acanthosaura capra
5' C
A
A
A
A
G
A
T
A
A A TTTG
A
A A A CA
C A
C
T
G
G
C
T
T
T G
3'
A
G
T
T
T
T
C
T
TCCA C C
AGGT C A
T
A
G A
C
A
C
C
G
T
A
G
Calotes versicolor
A 3'
G
T
T
T
T
C
T
TCCAG G
3'
A
C G
A T
A T
A T
A T
G C
A T
C C
T
TTCGG
C
A
A A TTTG
AAGC C
C C
C
C
A A A CA
A
C A
T A A
T
T A
G C
G C
C G
C
T
T
A
T G G
5'
A
Gonocephalus grandis
5'
A
C
A
A
A
A
G
A
T
A
A A
TTTG
3'
A
G
T
T
T
T
C
T
A
TCCAA G
AGGT T C
T
C
A
A A A CA
T A
C G A
C
T A
G C
G C
C T
T
T
T
A
T G G
Calotes cf. calotes
Draco sp.
A 3'
G
T
T
T
T
C
T
TCCA C C
AGGT C G
C
A
A
A A A CA
C A
C G A
C
T A
G C
G C
C G
T
T
T
A
T G G
5'
A
C
A
A
A
A
G
A
T
A
A
A
TTTG
Calotes cf. emma
FIG. 2.—Cloverleaf secondary structures of the inverted trnP genes for several draconine agamids. All nucleotide sequences shown here are lightstrand sequences. Conserved nucleotides found for most of mitochondrial trnP genes of vertebrates (Kumazawa and Nishida 1993) are highlighted in
bold. The secondary structure of Acanthosaura crucigera trnP gene is not shown to avoid redundancy.
Size Reduction of Some Structural Genes
When protein genes were aligned to their counterparts
from other vertebrates, C. versicolor genes for cob, nad2,
and nad6 were somewhat (10–15 bp) shorter than those
from most other vertebrates (data not shown). In addition,
C. versicolor small rRNA gene (rrnS) was approximately
100-bp smaller than those from most other vertebrates,
which is due to considerable truncation in 4 distinct regions
of its secondary structure (data not shown). Alignment of
individual gene sequences suggested that the size shortening of these structural genes is a phenomenon not specific to
C. versicolor but applicable for wider taxonomic groups of
agamids (data not shown).
Twenty-two tRNA genes encoded in the C. versicolor
mtDNA, having identical anticodon triplet sequences to
those for other vertebrates, can be folded into the cloverleaf
secondary structures. However, truncation of the TwC arm
(T arm), previously described for snake mitochondrial
tRNAs (Kumazawa et al. 1996), was also notable for some
of C. versicolor mitochondrial tRNA genes. The C. versicolor trnT gene had only 3 bp in the T stem. The C. versicolor trnM, trnC, and trnP genes also had a truncated T
stem (e.g., 4 bp in trnP as contrasted to 5 bp for most mitochondrial tRNAs; see fig. 2) or an extremely shorter number of bases in the T loop (only 1 or 2 bases if the regular 5
bp in the T stem are allowed; data not shown). This tendency of shortening the T arm size is also notable for
the corresponding tRNA genes of other agamids (Macey
et al. 2000; data not shown). Taken together, it is likely that
the agamid mtDNAs have evolved under an evolutionary
pressure to shorten their gene sizes.
Timing of the trnP Gene Inversion
In order to examine phylogenetic distribution of the
inverted state of trnP, we amplified and sequenced an
mtDNA region enclosed by the cob and rrnS genes for diverse groups of agamids using procedures described in Material and Methods. Table 1 lists agamid taxa for which
sequence information of this region is available, and they
cover all 6 subfamilies of Agamidae defined by Macey et al.
(2000). The results indicated that all 7 draconine taxa sequenced have the inverted trnP, whereas all nondraconine
agamids do not (table 1). In addition, the PCR assay (see
1334 Amer and Kumazawa
IQM to QIM
Iguana iguana (0)
Brachylophus fasciatus
Chamaeleo africanus
Chamaeleo dilepis
Leiolepis belliana (0*)
Pogona vitticepes (0)
100
Caimanops amphiboluroides
100
CR duplication
100
100
100
52
trnP inversion
81
100
93
100
0.1
Chamaeleonidae
Leiolepidinae (S.E. Asia)
Amphibolurinae
(Australia-New Guinea and
S.E. Asia for P. cocincinus)
Ctenophorus ornatus
Physignathus cocincinus (0)
Hydrosaurinae (S.E. Asia)
Hydrosaurus sp. (0*)
Gonocephalus grandis (1)
98
85
Bronchocela cristatella (1')
Ceratophora aspera
Salea horsfieldii
Pseudocalotes brevipes
100
Draconinae
94
Japalura splendida (1')
(S. and S.E. Asia)
Acanthosaura capra (1)
Calotes versicolor (1)
Calotes calotes (1)
Otocryptis wiegmanni
Draco blanfordii (1*)
Laudakia caucasia
99
Laudakia erythrogastra
Laudakia microlepis
100
37 ±5
55 ±7
Iguanidae
100
Agaminae
(Africa-W. Asia)
Agama agama
Agama impalearis (2)
FIG. 3.—An ML tree constructed using 1,490 bp of the mtDNA segment between nad1 and cox1 genes. ML bootstrap values are shown on the
corresponding branches when they are more than 50%. The root is placed on an arbitrary position on a lineage leading to iguanids based on previous
phylogenetic studies (see, e.g., Pough et al. 2004, p. 121–129; Kumazawa 2007). Divergence times estimated as described in Material and Methods are
also shown for their mean and standard deviation (mean ± SD) at some key nodes. Numbers in parentheses indicate the status of the trnP gene: 0, taxa
with the normal position and orientation for the trnP gene; 1, taxa with the trnP gene inversion as confirmed directly by sequencing; 1#, taxa with the
trnP gene inversion as inferred by the PCR assay; and 2, taxa with the noninverted trnP gene but with some tRNA gene translocations around the CR.
Numbers with an asterisk mean that the trnP gene status was based on sequence data of other congeneric species.
Material and Methods) supported that 3 more draconines
(Bronchocela cristatella, Japalura splendida, and Pseudocalotes sp.) have the inverted trnP (data not shown). We
confirmed that the inverted trnP genes of all the 7 draconines can assume a functional secondary structure (fig.
2), arguing against the possibility that they are pseudogenes. These tRNA genes retain not only universally
conserved bases (e.g., T8, T33, and R37) but also those
conserved within trnP genes from other vertebrates
encoded by the opposite strand (see bold nucleotides in
fig. 2).
Among the nondraconine taxa examined, those from
Uromastycinae, Leiolepidinae, Amphibolurinae, and Hydrosaurinae showed the typical organization of genes in this
region (i.e., cob–trnT–trnP–CR–trnF–rrnS with underlined
genes encoded by the light strand), whereas 2 representative
species from Agaminae showed different gene arrangements. In Xenagama taylori (Macey et al. 2006), the trnP
gene appears to have been translocated to the 3# end of CR
in the vicinity of trnF (see fig. 1B). Agama impalearis
showed another different arrangement in which trnF was
translocated into the genes for trnT and trnP (fig. 1C).
These arrangements found in Agaminae do not involve inversion of genes from their typical coding orientation.
Figure 3 shows an ML tree for the agamid phylogeny
constructed using partial mtDNA sequences between nad1
and cox1 genes. Topological relationships supported by the
ML analyses with .50% bootstrap values (see fig. 3) were
also reconstructed by Bayesian analyses with appreciably
high (.90%) posterior probabilities (data not shown). This
phylogeny is consistent with trees with larger taxonomic
samplings constructed by Macey et al. (2000) using the same
gene region. The trees (fig. 3; Macey et al. 2000) supported
basal divergence of Leiolepidinae and Uromastycinae, followed by divergences of amphibolurines and hydrosaurines.
Macey et al. (1997, 2000) found that a tRNA gene rearrangement from trnI–trnQ–trnM to trnQ–trnI–trnM occurred in an ancestral lineage of Acrodonta (Agamidae þ
Chamaeleonidae). Amer and Kumazawa (2005b) found that
duplication and insertion of CR into the nad5–nad6 boundary occurred in the amphibolurine lineage after the divergence of Physignathus cocincinus. The present study
shows that the inverted state of trnP is shared by all examined draconine taxa, strongly suggesting a single occurrence
of the gene inversion in the draconine ancestral lineage (fig.
3). Divergence time estimation further suggested that this
inversion event took place approximately 37–55 MYA during the Paleogene period (fig. 3). Because genus Draco represents an earliest shoot-off among the draconines (Macey
et al. 2000; fig. 3), it seems difficult to further narrow down
the timing of the gene inversion by the current approach.
Within the Agaminae, we conducted the PCR assay
with some more taxa (e.g., Laudakia nupta, Laudakia stellio, and Pseudotrapelus sinaitus) and did not find evidence
for the trnP inversion (data not shown). This further corroborates our conclusion for the single occurrence of the trnP
gene inversion within Draconinae. In order to elucidate the
timing for occurrence of 2 different gene translocations
Gene Inversion in C. versicolor mtDNA 1335
FIG. 4.—Possible mechanisms (Aand B) for the trnP gene inversion
from the typical vertebrate organization to the organization of the
draconine agamids through intermediate states. (A) The tRNA gene
remolding from a tandemly duplicated trnT gene through an anticodon
switch (TGT to TGG). (B) Direct gene inversion by a homologous DNA
recombination with the aid of accidentally formed inverted repeat
sequences (hatched regions). Genes are shown in the abbreviated forms
(see caption of fig. 1).
within Agaminae (figs. 1B and C), relevant mtDNA regions
will be sequenced from more diverse agamine taxa and reported elsewhere.
Possible Mechanism of Gene Inversion
Although gene rearrangements have been found in
mtDNAs of various vertebrate groups (Macey et al.
1997; Mindell et al. 1998; Inoue et al. 2001; Kumazawa
and Endo 2004; Mueller and Boore 2005; Kurabayashi
et al. 2006; San Mauro et al. 2006; references therein), these
rearrangements did not accompany a switch of the sense
strand for any genes. Because polymorphic duplications
and/or deletions of an mtDNA segment are present in diverse groups of vertebrates (see, e.g., Moritz et al. 1987;
Holt et al. 1988; Stanton et al. 1994; Gach and Brown
1997; Kajander et al. 2000), molecular mechanisms for
these rearrangements have been ascribed to the tandem duplication/random loss model (Moritz et al. 1987; Boore
2000; San Mauro et al. 2006). However, this model does
not account for gene rearrangements involving a switch
of an encoded strand, such as the one found for trnP of draconine agamids.
Figure 4 depicts 2 possible candidates of molecular
mechanisms that might account for the trnP inversion in
draconines. The 1st model invokes the tRNA gene remolding proposed to account for some tRNA gene rearrangements in invertebrate mtDNAs (Cantatore et al. 1987;
Higgs et al. 2003; Rawlings et al. 2003; Lavrov and Lang
2005). There is a strong evidence for the gene remolding in
these cases based on sequence similarities of the corresponding tRNA genes. This mechanism can potentially mediate the strand switch of a tRNA gene if a template tRNA
gene for duplication was encoded by an opposite strand in
relation to a gene that would be replaced and deleted. The
template tRNA gene for the model of figure 4A is an adjacent trnT gene which is encoded by the heavy strand,
whereas a trnP gene, which will later be replaced and deleted, is encoded by the light strand.
We attempted to evaluate this model using available
tRNA gene sequences for draconine and nondraconine agamids (figs. 5 and 6). If the model of figure 4A were the case,
one would expect to observe a higher sequence similarity
between trnT and trnP genes in draconines than in nondraconines. Alternatively, if the other mechanisms such as the
homologous DNA recombination (fig. 4B) were the case,
one would not observe the significant increase of the
trnT–trnP similarity in draconines. Figure 6 shows that
the trnT–trnP similarities are not significantly higher
(P . 0.05 by Student’s t-test) in draconines than in nondraconines. Figure 6 also shows that the trnT–trnP similarities
in draconines are not significantly higher (P . 0.05 by Student’s t-test) than an accidental level of sequence similarity
between other neighboring tRNA genes (i.e., trnI vs. trnM,
trnA vs. trnN, and trnC vs. trnY) using the same taxon sampling. These results do not provide supportive evidence for
the tRNA remolding model of figure 4A. The same conclusion was reached when trnT and trnP sequences at the immediate common ancestor of draconines were inferred (fig.
5) and compared directly (fig. 6).
Although current taxonomic sampling includes more
closely related taxa in the draconine group than in the nondraconine one (fig. 3), this should not affect the abovementioned conclusion because these comparisons were
not made for orthologous genes. This conclusion also does
not appear to be misled by the differential base composition
between mtDNA strands (Asakawa et al. 1991). Indeed,
trnP genes encoded by the heavy strand in draconines possess a somewhat different base composition from that for
trnP genes encoded by the light strand in nondraconines
(see Supplementary fig. S1, Supplementary Material online), suggesting that the inverted trnP genes of extant draconines have already been adapted to the general trend of
the strand-specific base composition after 37–55 Myr of the
inversion event. However, in the tRNA remolding model of
figure 4A, this factor, if anything, would operate to increase
the trnT–trnP similarities for draconines as compared with
those for nondraconines, which was not actually detected
by the sequence comparisons (fig. 6). Thus, our conclusion
of no tRNA gene remolding is likely not an artifact by the
base composition bias.
In the tRNA gene remolding model (fig. 4A), anticodon sequences of trnT and trnP genes can be converted by
a single-base substitution (TGT to TGG) but a number of
other base substitutions, especially in the acceptor and T
stem regions, must be assumed to convert from trnT to trnP
in draconines (fig. 5). By contrast, trnP gene sequences of
draconine and nondraconine agamids, which are encoded
by a different strand from each other, are clearly more similar to each other than they are to trnT genes (figs. 5 and 6).
Because identity elements for each tRNA are usually confined to relatively small numbers of nucleotide positions
(Giegé et al. 1998), this pattern of sequence similarities cannot be explained by rapid sequence convergence after the
tRNA gene remolding. We also considered the possibility
1336 Amer and Kumazawa
FIG. 5.—Nucleotide sequences of the trnP and trnT genes in draconine and nondraconine agamids. The trnP gene sequences are shown for all
7 draconines sequenced in this study. However, trnP gene sequences for nondraconines and trnT gene sequences for draconines are shown for only
some selected taxa to avoid redundancy. Alignment was taken using the secondary structure model for mitochondrial tRNAs (Kumazawa and Nishida
1993). Dots denote the nucleotide identity to the first sequence. All nucleotide sequences correspond to the sense-strand sequences. Possible trnP and
trnT gene sequences at the common ancestor of the 7 draconine agamids as manually inferred by the parsimony criterion (Farris optimization) based on
the phylogenetic relationships in figure 3 and Macey et al. (2000) are also shown below. Note that the ancestral sequences were not reconstructed for
unalignable regions in the D and T loops, as well as 2 bp in the T stem.
that a template tRNA gene for the tRNA remolding was
supplied by a tRNA gene other than the neighboring trnT
gene (e.g., an antisense sequence of the trnP gene itself).
However, none of other tRNA gene and its antisense sequence was found to have a prominently high sequence
similarity with the trnP gene in Calotes versicolor (data
not shown).
An alternative model shown in figure 4B takes advantage of accidentally formed inverted repeat sequences (IRS)
to invert the trnP gene by homologous DNA recombination. The role of IRS in causing a gene inversion by the
homologous DNA recombination is well known (see,
e.g., Watson et al. 2004, p. 294–295). For example, circular
chloroplast genomes contain long IRS, with which intervening unique sequences between them are frequently inverted and the chloroplast genomes exist as equimolar
mixtures of 2 isomers differing from each other in the relative orientation of the intervening sequences (Stein et al.
1986; Hiratsuka et al. 1989).
Additional support for the model of figure 4B comes
from the following arguments.
1. The inversion of trnP gene in draconines took place by
conserving its relative position in the mtDNA gene
organization (fig. 1A), favoring the local DNA recombination (fig. 4B) rather than other mechanisms,
such as retrotransposition.
2. As described above, nucleotide sequences of the
inverted and noninverted trnP genes are quite similar
and likely homologous, supporting the direct gene
inversion (fig. 4B) rather than a homoplasious origin of
the trnP function (fig. 4A). However, inspection of the
mtDNA sequences of extant draconines did not show
a more direct evidence of the DNA recombination (e.g.,
a remnant of IRS) presumably because of substantial
sequence changes for 37–55 Myr (data not shown).
3. There is now growing evidence for the occurrence of
homologous DNA recombination in vertebrate mtDNAs
(Thyagarajan et al. 1996; Hoarau et al. 2002; Kraytsberg
et al. 2004; Tsaousis et al. 2005).
Rarity of Gene Inversion in Vertebrate mtDNAs
Why are gene inversions so rare in vertebrate mtDNAs
although they appear to be more frequent in invertebrate
mtDNAs (Boore 1999; references therein)? As far as invertebrate chordates are concerned, there exist multiple examples of gene inversions. Gene orders of cephalochordate
mtDNAs retain a certain level of similarity to vertebrate
counterparts but a lineage of cephalochordates did allow
an inversion of 4 neighboring genes (Nohara et al.
2005). Ascidian mtDNAs appear to have allowed numerous
gene inversions that led to a quite extreme gene order in
which all genes are now encoded by 1 of 2 mtDNA strands
(Yokobori et al. 1999).
Here, we discuss 2 main possibilities to account for the
rarity of gene inversion in vertebrate mtDNAs although we
do not necessarily rule out other potential causes. First,
inversional mutations may be less frequent due to a lower
recombinational activity in vertebrate mtDNAs. However,
to our knowledge, there is no study in which mitochondrial recombinational activities were compared between
Gene Inversion in C. versicolor mtDNA 1337
FIG. 6.—Pairwise sequence similarities between neighboring tRNA
genes. Open columns show the similarities between designated pairs of
tRNA genes averaged among 6 draconine agamids (Acanthosaura capra,
Calotes cf. calotes, Calotes cf. emma, C. versicolor, Draco sp., and
Gonocephalus grandis), whereas hatched columns denote those averaged
among 5 other agamids (Uromastyx benti, Leiolepis cf. reevesi, Pogona
vitticepes, Agama impalearis, and Xenagama taylori). Gray columns show
the similarities within individual tRNA genes averaged among all possible
pairs of draconine versus nondraconine comparisons: I (isoleucine),
M (methionine), A (alanine), N (asparagine), C (cysteine), Y (tyrosine),
T (threonine) and P (proline). In each column, a range for one standard
deviation (i.e., mean ±SD) is shown. A column with an asterisk refers to
the base matching percentage between the inferred ancestral trnP and
trnT genes of the draconine agamids (see fig. 5). Total numbers of sites
used for the comparisons were 49 sites for trnI, trnM, trnA, trnN, trnT, and
trnP and 38 sites for trnC and trnY, where all sites in the D arm were
excluded due to the unusual secondary structure of the trnC genes. All sites
in the D, extra, and T loops, as well as the last 2 base pairs in the T stem,
were also excluded for all tRNA genes for the difficulty of alignment.
vertebrates and invertebrates. Description of frequent recombinations in selectively neutral CR sequences of vertebrate mtDNAs (Hoarau et al. 2002; Kraytsberg et al. 2004)
may not support this possibility.
Second, inverted gene organizations may be more frequently deleterious and thus less frequently tolerated in vertebrate mtDNAs. The typical gene organization of
vertebrate mtDNAs may be at an optimum state in some
respects. In vertebrate systems, large polycistronic RNA
molecules transcribed from major promoters inside the
CR are processed into individual RNA components
(mRNAs, rRNAs, and tRNAs) by enzymatic activities that
target tRNA sequences between mRNA and rRNA sequences (Anderson et al. 1981; Ojala et al. 1981). On the other
hand, gene expressions from some invertebrate mtDNAs
may proceed in a somewhat different way using multiple
promoters (Elliott and Jacobs 1989). In addition, the tendency of clustering tRNA genes (see, e.g., Jacobs et al.
1989) does imply relaxation of their processing constraints
in some invertebrate mtDNAs.
As a positive evidence of such processing constraints
consistently operating in the evolution of vertebrate mitochondrial tRNA genes, a transposed methionine tRNA gene
in the parrot fish mtDNA left its pseudogene in its original
location as a processing signal for an adjacent protein gene
(Mabuchi et al. 2004). If tRNA genes with such constraints
were inverted, they would no longer play the original processing role. The trnP gene resides in a position where it
does not serve as a processing signal for any adjacent pro-
tein or rRNA gene (see fig. 1A). Only a limited number of
tRNA genes that are apparently free or relaxed from such
constraints (i.e., trnQ, trnA, trnN, trnC, 2 trnS isoacceptor
genes, and trnP, all of which are encoded by the light strand
in the typical gene organization) may be allowed to be inverted in vertebrates.
In addition to the processing constraints operating on
tRNA genes, the relative location and orientation of individual structural genes in vertebrate mtDNAs may be important to ensure controlled gene expressions in response
to, for example, exogenous regulatory factors such as temperature and adenosine triphosphate concentration, as well
as the requirement of synthesizing increased amounts of
rRNAs relative to mRNAs (reviewed in Attardi 1993).
The less frequent gene inversion may therefore reflect
the strength of these regulatory constraints in vertebrate
mtDNAs. Unfortunately, biochemical experiments that implied such regulatory constraints have been carried out
mostly using mammalian systems (Clayton 1992; Attardi
1993) and very little is known on invertebrate systems. Elucidation of molecular mechanisms for regulatory gene expressions in invertebrate mitochondria should therefore be
encouraged to examine this hypothesis in future.
Supplementary Material
Supplementary tables S1, S2, S3, and S4 and figure S1
are available at Molecular Biology and Evolution online
(http://www.mbe.oxfordjournals.org/).
Acknowledgments
We thank Mr K. Yagi (Remix Peponi Co., Nagoya,
Japan) for providing animal samples, Drs T. Nishikawa
(Nagoya University Museum) and S. Kawada (National
Science Museum, Tokyo) for help in depositing our specimens to museums, and Ms C. Aoki for experimental assistance. We also thank Dr T. P. Satoh (University of Tokyo)
for critically reading an earlier version of manuscript and
providing useful comments. This work was supported by
grants from the Daiko Foundation (the Visiting Fellowship
for Foreign Scholars number 10093) and from the Ministry
of Education, Culture, Sports, Science and Technology of
Japan (numbers 14902202 and 17570076).
Literature Cited
Amer SAM, Kumazawa Y. 2005a. Mitochondrial DNA sequences of the Afro-Arabian spiny-tailed lizards (genus Uromastyx; family Agamidae): phylogenetic analyses and evolution
of gene arrangements. Biol J Linn Soc. 85:247–260.
Amer SAM, Kumazawa Y. 2005b. Mitochondrial genome of
Pogona vitticepes (Reptilia; Agamidae): control region
duplication and the origin of Australasian agamids. Gene
346:249–256.
Anderson S, Bankier AT, Barrell BG, et al. (14 co-authors) 1981.
Sequence and organization of the human mitochondrial
genome. Nature. 290:457–465.
Asakawa S, Kumazawa Y, Araki T, Himeno H, Miura K,
Watanabe K. 1991. Strand-specific nucleotide composition
1338 Amer and Kumazawa
bias in echinoderm and vertebrate mitochondrial genomes.
J Mol Evol. 32:511–520.
Attardi G. 1993. The human mitochondrial genetic system. In:
DiMauro S, Wallace DC, editors. Mitochondrial DNA in
human pathology. New York: Raven Press. p. 9–25.
Boore JL. 1999. Animal mitochondrial genomes. Nucleic Acids
Res. 27:1767–1780.
Boore JL. 2000. The duplication/random loss model for gene
rearrangement exemplified by mitochondrial genomes of
deuterostome animals. In: Sankoff D, Nadeau JH, editors.
Comparative genomics. Dordrecht (the Netherlands): Kluwer
Academic Publishers. p. 133–147.
Cantatore P, Gadaleta MN, Roberti M, Saccone C, Wilson AC.
1987. Duplication and remoulding of tRNA genes during the
evolutionary rearrangement of mitochondrial genomes. Nature. 329:853–855.
Clayton DA. 1992. Transcription and replication of animal
mitochondrial DNAs. Int Rev Cytol 141:217–232.
Elliott DJ, Jacobs HT. 1989. Mutually exclusive synthetic
pathways for sea urchin mitochondrial rRNA and mRNA.
Mol Cell Biol. 9:1069–1082.
Gach MH, Brown WM. 1997. Characteristics and distribution of
large tandem duplications in brook stickleback (Culaea
inconstans) mitochondrial DNA. Genetics. 145:383–394.
Giegé R, Sissler M, Florentz C. 1998. Universal rules and
idiosyncratic features in tRNA identity. Nucleic Acids Res.
26:5017–5035.
Higgs PG, Jameson D, Jow H, Rattray H. 2003. The evolution of
tRNA-Leu genes in animal mitochondrial genomes. J Mol
Evol. 57:435–445.
Hiratsuka J, Shimada H, Whittier R, et al. (16 co-authors) 1989.
The complete sequence of the rice (Oryza sativa) chloroplast
genome: intermolecular recombination between distinct
tRNA genes accounts for a major plastid DNA inversion
during the evolution of the cereals. Mol Gen Genet. 217:
185–194.
Hoarau G, Holla S, Lescasse R, Stam WT, Olsen JL. 2002.
Heteroplasmy and evidence for recombination in the mitochondrial control region of the flatfish Platichthys flesus. Mol
Biol Evol. 19:2261–2264.
Holt IJ, Harding AE, Morgan-Hughes AJ. 1988. Deletions of
muscle mitochondrial DNA in patients with mitochondrial
myopathies. Nature. 331:717–719.
Honda M, Ota H, Kobayashi M, Nabhitabhata J, Yong H-S,
Sengoku S, Hikida T. 2000. Phylogenetic relationships of the
family Agamidae (Reptilia: Iguania) inferred from mitochondrial DNA sequences. Zool Sci. 17:527–537.
Inoue JG, Miya M, Tsukamoto K, Nishida M. 2001. Complete
mitochondrial DNA sequence of Conger myriaster (Teleostei:
Anguilliformes): novel gene order for vertebrate mitochondrial genomes and the phylogenetic implications for anguilliform families. J Mol Evol. 52:311–320.
Jacobs HT, Asakawa S, Araki T, Miura K, Smith MJ,
Watanabe K. 1989. Conserved tRNA gene cluster in starfish
mitochondrial DNA. Curr Genet. 15:193–206.
Jameson D, Gibson AP, Hudelot C, Higgs PG. 2003. OGRe:
a relational database for comparative analysis of mitochondrial genomes. Nucleic Acids Res. 31:202–206.
Janke A, Erpenbeck D, Nilsson M, Arnason U. 2001. The
mitochondrial genomes of the iguana (Iguana iguana) and the
caiman (Caiman crocodylus): implications for amniote
phylogeny. Proc R Soc Lond B Biol Sci. 268:623–631.
Kajander OA, Rovio AT, Majamaa K, Poulton J, Spelbrink JN,
Holt IJ, Karhunen PJ, Jacobs HT. 2000. Human mtDNA
sublimons resemble rearranged mitochondrial genomes found
in pathological states. Hum Mol Genet. 9:2821–2835.
Kraytsberg Y, Schwartz M, Brown TA, Ebralidse K, Kunz WS,
Clayton DA, Vissing J, Khrapko K. 2004. Recombination of
human mitochondrial DNA. Science. 304:981.
Kumazawa Y. 2004. Mitochondrial DNA sequences of five
squamates: phylogenetic affiliation of snakes. DNA Res.
11:137–144.
Kumazawa Y. 2007. Mitochondrial genomes from major lizard
families suggest their phylogenetic relationships and ancient
radiations. Gene. 388:19–26.
Kumazawa Y, Endo H. 2004. Mitochondrial genome of the
Komodo dragon: efficient sequencing method with reptileoriented primers and novel gene rearrangements. DNA Res.
11:115–125.
Kumazawa Y, Nishida M. 1993. Sequence evolution of
mitochondrial tRNA genes and deep-branch animal phylogenetics. J Mol Evol. 37:380–398.
Kumazawa Y, Nishida M. 1999. Complete mitochondrial DNA
sequences of the green turtle and blue-tailed mole skink:
statistical evidence for archosaurian affinity of turtles. Mol
Biol Evol. 16:784–792.
Kumazawa Y, Ota H, Nishida M, Ozawa T. 1996. Gene rearrangements in snake mitochondrial genomes: highly concerted
evolution of control-region-like sequences duplicated and
inserted into a tRNA gene cluster. Mol Biol Evol. 13:1242–1254.
Kurabayashi A, Usuki C, Mikami N, Fujii T, Yonekawa H,
Sumida M, Hasegawa M. 2006. Complete nucleotide
sequence of the mitochondrial genome of a Malagasy poison
frog Mantella madagascariensis: evolutionary implications
on mitochondrial genomes of higher anuran groups. Mol
Phylogenet Evol. 39:223–236.
Lavrov DV, Lang BF. 2005. Transfer RNA gene recruitment in
mitochondrial DNA. Trends Genet. 21:129–133.
Mabuchi K, Miya M, Satoh TP, Westneat MW, Nishida M. 2004.
Gene rearrangements and evolution of tRNA pseudogenes in
the mitochondrial genome of the parrotfish (Teleostei:
Perciformes: Scaridae). J Mol Evol. 59:287–297.
Macey JR, Fong JJ, Kuehl JV, Shafiei S, Ananjeva NB,
Papenfuss TJ, Boore JL. 2005. The complete mitochondrial
genome of a gecko and the phylogenetic position of the
Middle Eastern Teratoscincus keyserlingii. Mol Phylogenet
Evol. 36:188–193.
Macey JR, Larson A, Ananjeva NB, Fang Z, Papenfuss TJ. 1997.
Two novel gene orders and the role of light-strand replication
in rearrangement of the vertebrate mitochondrial genome. Mol
Biol Evol. 14:91–104.
Macey JR, Schulte JA 2nd, Ananjeva NB, Larson A, RastegarPouyani N, Shammakov SM, Papenfuss TJ. 1998. Phylogenetic relationships among agamid lizards of the Laudakia
caucasia species group: testing hypotheses of the biogeographic fragmentation and an area cladogram for the Iranian
Plateau. Mol Phylogenet Evol. 10:118–131.
Macey JR, Schulte JA 2nd, Fong JJ, Das I, Papenfuss TJ. 2006.
The complete mitochondrial genome of an agamid lizard from
the Afro-Asian subfamily agaminae and the phylogenetic
position of Bufoniceps and Xenagama. Mol Phylogenet Evol.
39:881–886.
Macey JR, Schulte JA 2nd, Larson A, Ananjeva NB, Wang Y,
Pethiyagoda R, Rastegar-Pouyani N, Papenfuss TJ. 2000.
Evaluating trans-tethys migration: an example using acrodont
lizard phylogenetics. Syst Biol. 49:233–256.
Mindell DP, Sorenson MD, Dimcheff DE. 1998. Multiple
independent origins of mitochondrial gene order in birds.
Proc Natl Acad Sci USA 95:10693–10697.
Moody SM. 1987. A preliminary cladistic study of the lizard
genus Uromastyx (Agamidae, sensu lato), with a checklist and
diagnostic key to the species. In: van Gelder JJ, Strijbosch H,
Bergers PJM, editors. Proceedings of the Fourth Ordinary
Gene Inversion in C. versicolor mtDNA 1339
General Meeting of the Society of European Herpetology,
Nijmegen, Holland. p. 285–288.
Moritz C, Dowling TE, Brown WM. 1987. Evolution of animal
mitochondrial DNA: relevance for population biology and
systematics. Annu Rev Ecol Syst. 18:269–292.
Mueller RL, Boore JL. 2005. Molecular mechanisms of extensive
mitochondrial gene rearrangement in plethodontid salamanders. Mol Biol Evol. 22:2104–2112.
Nohara M, Nishida M, Miya M, Nishikawa T. 2005. Evolution of
the mitochondrial genome in Cephalochordata as inferred
from complete nucleotide sequences from two Epigonichthys
species. J Mol Evol. 60:526–537.
Ojala D, Montoya J, Attardi G. 1981. tRNA punctuation model
of RNA processing in human mitochondria. Nature.
290:470–474.
Posada D, Crandall KA. 1998. Modeltest: testing the model of
DNA substitution. Bioinformatics. 14:817–818.
Pough FH, Andrews RM, Cadle JE, Crump ML, Savitzky AH,
Wells KD. 2004. Herpetology, third edition. London: Pearson
Education.
Rawlings TA, Collins TM, Bieler R. 2003. Changing identities:
tRNA duplication and remolding within animal mitochondrial
genomes. Proc Natl Acad Sci USA. 100:15700–15705.
San Mauro D, Gower DJ, Zardoya R, Wilkinson M. 2006. A
hotspot of gene order rearrangement by tandem duplication
and random loss in the vertebrate mitochondrial genome. Mol
Biol Evol. 23:227–234.
Stanton DJ, Daehler LL, Moritz CC, Brown WM. 1994.
Sequences with the potential to from stem-and-loop structures
are associated with coding-region duplications in animal
mitochondrial DNA. Genetics. 137:233–241.
Stein DB, Palmer JD, Thompson WF. 1986. Structural evolution
and flip-flop recombination of chloroplast DNA in the fern
genus Osmunda. Curr Genet. 10:835–841.
Swofford DL. 2003. PAUP*: phylogenetic analysis using
parsimony (* and other methods). Version 4. Sunderland
(MA): Sinauer Associates.
Thorne JL, Kishino H, Painter IS. 1998. Estimating the rate of
evolution of the rate of molecular evolution. Mol Biol Evol.
15:1647–1657.
Thyagarajan B, Padua RA, Campbell C. 1996. Mammalian
mitochondria possess homologous DNA recombination
activity. J Biol Chem. 271:27536–27543.
Tsaousis AD, Martin DP, Ladoukakis ED, Posada D, Zouros E.
2005. Widespread recombination in published animal mtDNA
sequences. Mol Biol Evol. 22:925–933.
Watson JD, Baker TA, Bell SP, Gann A, Levine M, Losick R.
2004. Molecular biology of the gene, fifth edition. San
Francisco (CA): Benjamin Cummings.
Yang Z. 1997. PAML: a program package for phylogenetic
analyses by maximum likelihood. CABIOS. 13:555–556.
Yokobori S, Ueda T, Feldmaier-Fuchs G, Pääbo S, Ueshima R,
Kondow A, Nishikawa K, Watanabe K. 1999. Complete DNA
sequence of the mitochondrial genome of the ascidian Halocynthia roretzi (Chordata, Urochordata). Genetics. 153:1851–1862.
Zhou K, Li H, Han D, Bauer AM, Feng J. 2006. The complete
mitochondrial genome of Gecko gecko (Reptilia: Gekkonidae)
and support for the monophyly of Sauria including Amphisbaenia. Mol Phylogenet Evol. 40:887–892.
Yoko Satta, Associate Editor
Accepted March 15, 2007