Mitochondrial genome of the moon jelly Aurelia aurita (Cnidaria, Scyphozoa): A linear DNA molecule encoding a putative DNA-dependent DNA polymerase.

Abstract

The 16,937-nuceotide sequence of the linear mitochondrial DNA (mt-DNA) molecule of the moon jelly Aurelia aurita (Cnidaria, Scyphozoa) - the first mtDNA sequence from the class Scypozoa and the first sequence of a linear mtDNA from Metazoa - has been determined. This sequence contains genes for 13 energy pathway proteins, small and large subunit rRNAs, and methionine and tryptophan tRNAs. In addition, two open reading frames of 324 and 969 base pairs in length have been found. The deduced amino-acid sequence of one of them, ORF969, displays extensive sequence similarity with the polymerase [but not the exonuclease] domain of family B DNA polymerases, and this ORF has been tentatively identified as dnab. This is the first report of dnab in animal mtDNA. The genes in A. aurita mtDNA are arranged in two clusters with opposite transcriptional polarities; transcription proceeding toward the ends of the molecule. The determined sequences at the ends of the molecule are nearly identical but inverted and lack any obvious potential secondary structures or telomere-like repeat elements. The acquisition of mitochondrial genomic data for the second class of Cnidaria allows us to reconstruct characteristic features of mitochondrial evolution in this animal phylum.

Full text

Page 1

Mitochondrial genome of the moon jelly Aurelia aurita
(Cnidaria, Scyphozoa): A linear DNA molecule
encoding a putative DNA-dependent
DNA polymerase
Zhiyong Shaoa, Shannon Grafb, Oleg Y. Chagac, Dennis V. Lavrova,d,⁎
aInterdepartmental Genetics Graduate Program, Iowa State University, Ames, Iowa 50011, United States
bDepartment of Microbiology and Immunology, Georgetown University Medical Center, Washington, DC 20007, United States
cDepartment of Cell and Molecular Biology, Feinberg School of Medicine, Northwestern University, Chicago, Illinois, United States
dDepartment of Ecology, Evolution and Organismal Biology, Ames, Iowa 50011, United States
Received 18 April 2006; received in revised form 20 June 2006; accepted 23 June 2006
Available online 18 July 2006
Received by M. Di Giulio
Abstract
The 16,937-nuceotide sequence ofthe linear mitochondrial DNA (mt-DNA)molecule of the moon jelly Aureliaaurita(Cnidaria, Scyphozoa)– the
firstmtDNAsequencefromthe classScypozoa and the firstsequence ofa linear mtDNAfrom Metazoa–has beendetermined.Thissequence contains
genes for 13 energy pathway proteins, small and large subunit rRNAs, and methionine and tryptophan tRNAs. In addition, two open reading frames of
324 and 969 base pairs in length have been found. The deduced amino-acid sequence of one of them, ORF969, displays extensive sequence similarity
with the polymerase [but not the exonuclease] domain of family B DNA polymerases, and this ORF has been tentatively identified as dnab. This is the
first report ofdnab in animal mtDNA. The genes inA. aurita mtDNA are arranged in two clusters with opposite transcriptional polarities; transcription
proceedingtowardtheendsofthemolecule.Thedeterminedsequencesattheendsofthemoleculearenearlyidenticalbutinvertedandlackanyobvious
potentialsecondarystructuresortelomere-likerepeatelements.TheacquisitionofmitochondrialgenomicdataforthesecondclassofCnidariaallowsus
to reconstruct characteristic features of mitochondrial evolution in this animal phylum.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Linear mtDNA; Mitochondrial evolution; dnaB; Gene order
1. Introduction
Although mitochondrial genomes are known to vary exten-
sively in size, structure, and gene content across diverse euka-
ryotic groups, those of multicellular animals (Metazoa) are
considered to be largely uniform (Lang et al., 1999). A typical
metazoan mtDNA is usually depicted as a small (∼16 kpb),
circular molecule that carries a conserved set of 37 compactly
arrayed genes coding for 13 proteins, two rRNAs and 22 tRNAs
(Boore, 1999). Other features commonly associated with animal
mtDNA include the lack of introns, multiple deviations in the
genetic code, unusual and/or highly reduced rRNA and tRNA
primary and secondary structures, and the presence of a single
large non-coding region containing the replication origins and
transcriptional promoters (Wolstenholme, 1992). Although the
idea of a “typical” animal mtDNA has its merits, a broader
sampling of diverse animal phyla revealed multiple exceptions
(Armstrong et al., 2000; Helfenbein et al., 2004). This is espe-
cially true for mtDNAs of nonbilaterian (“lower”) animals,
which lack many features associated with the “typical” animal
mtDNA and vary more in size and gene content (Beagley et al.,
1995; Ender and Schierwater, 2003; Lavrov et al., 2005).
Mitochondrialgenomes inthe phylumCnidariaare amongthe
most atypical among Metazoa. First, among the four traditionally
Gene 381 (2006) 92–101
www.elsevier.com/locate/gene
Abbreviations: SSU-rRNA and LSU-rRNA, small and large subunit
ribosomal RNAs; mtDNA, mitochondrial DNA; mt, mitochondrial.
⁎Correspondingauthor.253BesseyHall,DepartmentofEcology,Evolutionand
Organismal Biology, Iowa State University, Ames, Iowa 50011, United States.
E-mail address: dlavrov@iastate.edu (D.V. Lavrov).
0378-1119/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.gene.2006.06.021

Page 2

recognized cnidarian classes, Anthozoa, Cubozoa, Scyphozoa
and Hydrozoa, only the Anthozoa have a circular mitochondrial
genome,whiletheothershavelinearmtDNA(Bridgeetal.,1992,
1995). Second, out of the 22–25 tRNA genes present in mtDNA
of most other animals, only two,trnM andtrnW, have been found
in cnidarian mtDNA. Other tRNAs required for mitochondrial
protein synthesis are assumed to be encoded in the nuclear
genomeand imported into the mitochondria. Third, foreign DNA
is frequently found in cnidarian mtDNA, including a putative
mismatch repair protein (MutS) in the soft coral Sarcophyton
glaucum and the sea pen Renilla kolikeri (Pont-Kingdon et al.,
1995, 1998) and group I introns with or without the homing
endonucleases of the LAGLI-DADG type in the sea anemone
Metridium senile (Beagley et al., 1996) and several hard corals
(van Oppen et al., 2002; Fukami and Knowlton, 2005). Finally,
cnidarians use a minimally modified genetic code for mitochon-
drialtranslation,withTGA=tryptophanastheonlydeviation,and
encode well-conserved ribosomal and transfer RNAs.
Our knowledge of cnidarian mtDNA is mostly limited to the
class Anthozoa, for which several complete mitochondrial se-
quences are available: those from the soft coral S. glaucum
(Beaton et al., 1998), the sea anemone M. senile (Beagley et al.,
1998), and six hard corals Acropora tenuis (van Oppen et al.,
2002), Montipora cactus, Anacropora matthai (GenBank
Access number NC_006902; NC_006898), Montastraea annu-
laris, Montastraea franksi, and Montastraea faveolata (Fukami
and Knowlton, 2005). The other three classes are represented by
a few,mostly partial gene sequences (Pont-Kingdon et al., 2000;
Dawson and Jacobs, 2001; Schroth et al., 2002). It is unclear,
therefore, whether the features described for anthozoan mtDNA
are specific for this class or characteristic of the whole phylum
Cnidaria. To resolve this question we determined a nearly
complete sequence of the linear mitochondrial genome from the
scyphozoan Aurelia aurita, which we describe here.
Scyphozoans (jellyfish), are exclusively marine cnidarians
with a predominant medusoid stage in the lifecycle. They live in
all oceans, from polar to tropic. Some inhabit the deep sea, but
most live close to the coastal waters. There are approximately
200 species in this class, which are traditionally divided into
four orders (Brusca and Brusca, 2002), although one of these
orders (Stauromedusae) may represent an independent lineage
within the phylum (Collins, 2002; Marques and Collins, 2004;
Collins and Daly, 2005). The moon jelly Aurelia is one of the
most common and widely distributed species of jellyfish and a
popular research organism (Arai, 1996). Traditionally only two
species are recognized in the genus, A. aurita and A. limbata,
the latter is restricted to north polar oceans. However, presence
of several cryptic species within A. aurita is possible (Dawson
and Jacobs, 2001; Schroth et al., 2002).
2. Materials and methods
2.1. DNA extraction, PCR, and sequencing
Scyphistomae(scyphozoanpolyps)ofA.auritawerecollected
at the Marine Biological Station of the St. Petersburg University
(Chupa Inlet, Kandalaksha Bay of the White Sea; 66°19′N,
033°40′E) and cultured in the lab. After strobilation (an asexual
process of transverse fission), ephirae (young medusae) were
gathered and stored in 70% ethanol. Total DNA was extracted
from∼10ephiraeaccordingtotheprotocoldescribedinBerntson
et al. (1999). Primers designed to match generally conserved
regions of the animal mtDNA were used to amplify short frag-
ments from cox1 (Folmer et al., 1994), and nad4 (nad4F: 5′-
CCKAARGCYCAYGTKGARGCYCC-3′, nad4-R: 5′-GAR-
GAWCAKAWWCCRTGAGCAATYAT-3′). Specific primers
(Aurelia-cox1-F1: 5′-AACGTTGTAGTGACCGCTCATGC-3′,
Aurelia-cox1-R1: 5′CTTGGAAAGGCCATATCTGGAGC-3′,
Aurelia-nad4-F1: 5′-TCATGGACACTTTTTCCTGTGGC-3′,
Aurelia-nad4-R1: 5′-ATATATTACAGCAATTGACCCAAGC-
3′, Aurelia-rnl-F1: 5′- GTAACTCTGACCGTGATGAAG
TAGC-3′, Aurelia-rnl-R1: 5′-ATATCATAATTCAACATC
GAGGTGGC-3′) were designed based on these sequences and
previouslyavailable(Bridgeetal.,1995)partialrnldataandwere
usedwithTakara®LAPCRkitstoamplify∼15kbofthemtDNA
in two overlapping fragments, between nad4 and cox1 and bet-
ween cox1 and rnl. The peripheral regions of the molecule were
amplified using Step-Out PCR (Wesley and Wesley, 1997) and
reamplifiedbyconventionalPCRwithspecificprimersdesigned
atthe ends of the molecule. PCR reaction products were purified
by three serial passages through Ultrafree™ (30,000 NMWL)
columns (Millipore) and used as templates in dye-terminator,
cycle-sequencing reactions according to supplier's (Perkin
Elmer®) instructions. Both strands of each amplification product
were sequenced by primer walking, using an ABI 377 Automated
DNA Sequencer. Sequences were assembled using the STADEN
software suite (Staden, 1996). tRNA genes were identified by the
tRNAscan-SE program (Lowe and Eddy, 1997), other genes were
identified by similarity searches in local databases using the
FASTA program (Pearson, 1994), and in GenBank at NCBI using
the BLAST network service (Benson et al., 2003). The secondary
structures of rRNA genes were derived by analogy to other pub-
lished rRNA gene structures, and drawn using the RnaViz 2
program (De Rijk et al., 2003). A. aurita mtDNA sequence is
available via Genbank accession number DQ787873.
2.2. Phylogenetic analyses of protein data
Amino acid sequences of individual proteins were aligned
three times using ClustalW 1.82 (Thompson et al., 1994) with
differentcombinationsofopening/extensiongappenalties:10/0.2
(default), 12/4 and 5/1. For the last alignment, no increased gap
penalties near existing gaps, no reduced gap penalties in hydro-
philic stretches and no residue-specific penalties were applied.
ThethreealignmentswerecomparedusingSOAP(Löytynojaand
Milinkovitch, 2001), and the positions which were identical
among them were included in phylogenetic analyses. The final
concatenated data set was 2864 amino acids in length.
We performed a maximum likelihood(ML)searchfor the best
treeandestimatedbootstrapsupportvaluesasimplementedinthe
PHYML(v.2.4.4)program(GuindonandGascuel,2003) usinga
gamma+invariant model with 8 categories, estimated α-param-
eter,themtREVmatrixofamino-acidsubstitutions,andestimated
frequencies of amino acids. Bayesian inferences (MB) used
93Z. Shao et al. / Gene 381 (2006) 92–101

Page 3

MrBayes 3.1.1 (Ronquist and Huelsenbeck, 2003). We ran four
Markov Chain Monte Carlo (MCMC) chains for 1,100,000
generations, using the mtREV model of amino acid substitutions
with gamma+invariant distributed rates. Trees were sampled
every 1,000th cycle after the first 100,000 burn-in cycles.
3. Results and discussion
3.1. Genome structure, organization and nucleotide composition
The mitochondrial genome of A. aurita has been previously
characterized as a monomeric linear molecule about 18 kbp in
length (Bridge et al., 1992). We determined the sequence of a
16,937 bp segment of the genome that encodes 13 known
mitochondrialproteins,2tRNAs,largeandsmallsubunitrRNAs,
and two ORFs (Fig. 1). PCR amplification, cloning, and sequen-
cing of the rest of the mtDNAwere problematic. The sequence at
one end of the determined segment is repeated in an inverted
orientation at the other end. We view these repeated sequences as
partsoflonginvertedterminalrepeatsthatconstituteahallmarkof
linear plasmids and linear mitochondrial genomes, derived from
integration of such plasmids into circular mtDNA (Meinhardt et
al.,1990,1997;NosekandTomaska,2003).Linearmitochondrial
genomes are known in diverse phylogenetic groups (Vahrenholz
etal.,1993; Burger etal.,2000; Forget etal.,2002),but thisisthe
first description of such a genome within the Metazoa.
The genes in A. aurita mtDNA are arranged into two clusters,
one of which is in the opposite transcriptional orientation to the
other.Transcriptionproceedsinthedirectionstowardstheendsof
themolecule,with15genestranscribedinonedirectionwhiletwo
knowngenesand(tentatively)twoORFsintheother.Thechange
in the transcriptional polarity occurs between cox1 and cox2,
where a 93 bp non-coding region is located. Cox1 and cox2 are
located in a cluster of six genes, which are arranged nearly
identically in A. aurita and S. glaucum (Fig. 1). Two other
arrangements, +nad6+nad3+nad4L and +nad2+nad5, are also
conserved between these organisms. Furthermore the same three
arrangements of protein-coding genes are largely conserved in
three demosponge mtDNAs [except that there are tRNA genes
presentintheintergenicregions(Lavrovetal.,2005)]andsomay
be ancestral for all animals.
The A+T content of A. aurita mtDNA is 66.7%, slightly
higherthaninother cnidarianmtDNAsbutwellwithintherange
reported for other metazoans. There is a slight AT-skew between
the two strands (0.09), but almost no GC-skew (−0.03) for the
wholegenome.However,strongGC-skew(N0.3)isfoundinthe
sequencesencodingORFs andtheterminal endsof themolecule
(Table 1).
3.2. Protein-coding genes
Thirteen protein-coding genes found in most other animal
mtDNA (atp6, atp8, cob, cox1–3, nad1–6, nad4L) are also
present in A. aurita genome. These A. aurita genes, coding for
protein subunits involved in respiration and oxidative phosphor-
ylation, are similar in size to their homologues in other cnidarian
mtDNAsandsharewiththemonaverage54%(range21–81%)of
sequence identity (Table 2). All protein-coding genes are inferred
to have complete termination codons: either TAA (12 genes) or
TAG(cob).Twelveprotein-codinggenesstartwithATGinitiation
codons while one (nad3) is inferred to begin with GTG. We
inferredthatnad3overlapswiththeadjacentgenesby10(nad4L)
and 47 (nad6) nucleotides, respectively (Fig. 1). However, it is
possible that nad6 terminates with the CGTcodon (see below) in
which case the latter overlap would be reduced to 2 nt.
Inadditiontothe typical set ofprotein-codinggenes described
above, two ORFs of 324 and 969 nucleotides have been found
downstream of rnl, close to the end of the linear molecule. The
deduced amino-acid sequence of one of them, ORF969, displays
extensive sequence similarity with the family B DNA poly-
merases (Fig. 2) and this ORF has been tentatively identified as
dnab. Genes encoding a family B DNA polymerase have been
previously found in mtDNA from several other organisms, inclu-
ding the linear genome of the golden algae (chrysophyte) Oc-
hromonas danica (Coleman et al., 1991), and the circular
genomes of the red algae Porphyra purpurea (Burger et al.,
Fig. 1. Gene map of A. aurita mtDNA and gene order comparison with Sarcophyton glaucum mtDNA. Protein and ribosomal genes (large open boxes) are: atp6, 8 —
subunits 6, and 8 of the F0ATPase, cox1–3–cytochrome c oxidase subunits 1–3, cob — apocytochrome b (cob), nad1–6 and nad4L — NADH dehydrogenase
subunits 1–6 and 4L, rns and rnl — small and large subunit rRNAs. tRNA genes (hatched boxes) are identified by the one-letter code for their corresponding amino
acid. Orf324 and orf969 are two large open reading frames found in A. aurita mtDNA. Large intergenic regions are shown by shaded boxes; flanking sequences — by
two large arrows. Transcriptional direction for each gene is indicated by an arrow. Three conserved gene clusters between A. aurita and S. glaucum are underlined and
interconnected with arrows.
94Z. Shao et al. / Gene 381 (2006) 92–101

Page 4

1999), carrot (Robison and Wolyn, 2005), and basidiomycete
Agrocybe aegerita (Bois et al., 1999). Related genes are also
found in linear mitochondrial plasmids of protists, plants and
fungi(Robisonetal.,1991;RobisonandHorgen,1996;Rousvoal
et al., 1998; Robison and Wolyn, 2005). The evolutionary
proximity between mitochondrial and plasmid polB genes within
eachofthesegroupsoforganismsindicatesthatthemitochondrial
polB genes originated by the integration of linear plasmids into
mtDNA (Mouhamadou et al., 2004).
In comparison to its homologues in other organisms, the polB
present in A. aurita mtDNA is truncated at the 5′ end and is
inferred to code for a protein of 323 amino acids, approximately
200 aa smaller than any previously reported DNA polymerases.
Although this may suggest that the ORF represents a pseudogene
on its way to elimination as reported previously in some other
species [e.g., Weber et al. (1995)], the analysis of its sequence
(below) suggests a different interpretation.
Most DNA-dependent DNA polymerases possess two distinct
catalytic activities, a synthetic DNA polymerization activity and a
proofreading 3′–5′ exonuclease activity [reviewed in Joyce and
Steitz (1994)]. These two enzymatic activities have been mapped
in two structurally independent domains of the enzyme: the
exonuclease activityinthe N-terminaldomainandthe polymerase
activity in the C-terminal domain (Braithwaite and Ito, 1993). A
comparison of the amino acid sequence inferred from A. aurita
ORF969 with those of other DNA-dependent DNA polymerases
indicates that the exonuclease domain has been mostly lost in the
A. aurita protein, while the polymerase domain remains mostly
intact and contains at least four out of five evolutionary conserved
motifs (Dx2SLYP, Kx3NSxYG, Tx2A/GR, YxDTDS) character-
istic ofthisdomaininfamilyBDNApolymerases(Trunigeretal.,
1998). This clearly non-random pattern of the DNA loss and
conservationsuggestthatthepolymerasedomainismaintainedby
selection pressure. However, its function remains unclear because
the lack of the exonuclease domain is known to detrimentally
affect polymerization activity of the enzyme (Freemont et al.,
1986; Truniger et al., 1998). Interestingly, A. aurita hypothetical
POLB lacks a particular insertion between motifs POLIIa and
Table 1
Nucleotide composition data for different groups of genes, ORFs, and non-coding regions in Aurelia aurita mtDNA
Coding sequencesORFs rRNA-genestRNA-genes IntergenicFlanking repeatsa
(Total) (3rd positions)
%G
%A
%T
%C
%A+T
AT-skew
GC-skew
Total (bp)
17.0
28.9
37.6
16.6
66.5
−0.13
0.01
11.8
36.9
37.2
14.1
74.1
−0.01
−0.09
3977
10.5
37.4
30.0
22.0
67.4
0.11
−0.35
1293
16.8
38.0
28.8
16.4
66.8
0.14
0.01
2763
19.1
32.6
29.1
19.1
61.7
0.06
0.00
141
8.4
45.0
37.4
9.2
82.4
0.09
−0.04
131
21.1/10.3
37.6/30.8
30.2/38.5
11.0/20.5
67.9/69.2
0.11/−0.11
0.31/−0.33
417/35111,931
aThe nucleotide composition data is shown separately for the repeat units upstream of cob/downstream of orf969. The discrepancies in the nucleotide composition
data for these repeats reflect a) the opposite orientation of repeats and b) different lengths of sequences determined at the two ends of the molecule.
Table 2
Comparison of mitochondrial protein genes of the jellyfish Aurelia aurita (A.a.), the hard coral Acropora tenuis (A.t.), the sea anemone Metridium senile (M.s.) and the
soft coral Sarcophyton glaucum (S.g.)
Protein genesNumber of encoded amino acidsa
Percent amino acid identity Predicted initiation and
termination codons
A.a. A.t.M.s.S.g. A.a./A.t. A.a./M.s.A.a./S.g. M.s./S.g.In Aurelia auritab
atp6
atp8c
cob
cox1
cox2
cox3
nad1
nad2
nad3
nad4
nad4L
nad5
nad6
234
67
379
526
241
261
323
439
119
480
100
605
190
232
72
384
533
247
262
327
365
118
491
99
611
197
229
72
393
530
248
262
334
385
118
491
99
600
202
237
72
386
531
253
261
325
457
117
495
97
605
185
64
21
60
79
63
70
61
39
54
52
42
48
32
63
22
63
81
62
75
61
39
56
54
50
50
35
62
22
64
80
59
76
61
30
60
49
48
49
34
57
26
68
81
71
77
70
41
70
58
57
59
38
ATG (−1)
ATG (5)
ATG (2)
ATG (93)
ATG (93)
ATG (−1)
ATG (−1)
ATG (1)
GTG (−47)
ATG (0)
ATG (−10)
ATG (0)
ATG (14?)
TAA (−1)
TAA (−1)
TAG (END)
TAA (0)
TAA (5)
TAA (11)
TAA (0)
TAA (0)
TAA (−10)
TAA (2)
TAA (−1)
TAA (−2)
TAG (−47)
aData for A. tenuis are from van Oppen et al. (2002), for M. senile are from Beagley et al. (1998) and for S. glaucum are from Beaton et al. (1998).
bThe numbers of non-coding nucleotides upstream and downstream of a gene are shown in parenthesis after initiation and termination codons. The negative
numbers indicate that the genes are overlapping. END indicates that the flanking sequence is adjacent to the gene.
cSequence identities of atp8 are uncertain due to alignment ambiguities.
95Z. Shao et al. / Gene 381 (2006) 92–101

Page 5

POLIIb that is specific for protein-primed DNA polymerases and,
at least in phi29 DNA polymerase, also plays a role in strand
displacement and processivity (Rodriguez et al., 2005).
3.3. Codon usage and genetic code
Theanalysesofthecodonusageamongthe13energypathway
protein genes and, separately, in the dnab are shown in Table 3.
Thetableiscompiledbasedonaminimallymodifiedgeneticcode
(TGA=tryptophan as the only deviation) deduced for mitochon-
drial protein synthesis in A. aurita. However, the specificity of
CGN codons, tentatively identified as arginine, remains suspect
due to their rarity in the genome. Indeed, among the 13 energy
pathwayproteingenes,twocodonsCGCandCGGareneverused
andtwoothercodonsinthe CGNfamilyareusedonlyonceeach.
The scarcity of CGN codons does not reflect the absence of
arginine in mitochondrial proteins: two other codons specifying
arginine (AGA and AGG) are found 73 times. Neither can it be
Fig. 2. Alignment of the ORF969 encoded by the Aurelia aurita mtDNAwith DNA-dependent DNA polymerases from other organisms. The sequences are derived
from: aa-orf969 — Aurelia aurita mtDNA; pa-pAL2-1 — pAL2-1 plasmid of Podospora anserina (X60707); aa-mt — Agrocybe aegerita mtDNA (AF061244); ai-
AI2 — plasmid AI2 of Ascobolus immerses (P22374); ni-Kal — kalilo plasmid of Neurospora intermedia (X52106); od-mt — Ochromonas danica mtDNA
(NC002571); ppu-mt — Porphyra purpurea mtDNA (NC_002007); pp-mF — mF plasmid of Physarum polycephalum (D29637), bv-mt — Beta vulgaris mtDNA
(NC_002511); dc-mt — Daucus carota mtDNA (AY521591); prd1 — bacteriophage PRD1 (NC001421); phi29 — bacteriophage phi29 (X53370). Five evolutionary
conserved motifs characteristic of polymerase domain in family B DNA polymerases are those listed by Truniger et al. (1998). Numbers within each sequence indicate
the number of amino-acid residues excluded from the alignment.
Table 3
Codon usage among the 13 energy pathway protein genes and ORF969
A.a.ORFA.a.ORFA.a.ORF A.a. ORF
Phe TTT
TTC
TTA
TTG
CTT
CTC
CTA
CTG
ATT
ATC
ATA
ATG
GTT
GTC
GTA
GTG
220
78
271
95
85
14
103
28
162
42
252
123
132
27
116
16
10
9
10
4
8
5
8
2
12
8
14
5
6
0
3
0
Ser TCT
TCC
TCA
TCG
CCT
CCC
CCA
CCG
ACT
ACC
ACA
ACG
GCT
GCC
GCA
GCG
138
57
73
13
4
7
0
11
4
2
1
12
5
9
0
3
2
6
0
Tyr TAT
TAC
TAA
TAG
CAT
CAC
CAA
CAG
AAT
AAC
AAA
AAG
GAT
GAC
GAA
GAG
123
41
11
12
9
1
0
1
5
8
4
12
11
25
4
7
3
6
2
Cys TGT
TGC
TGA
TGG
CGT
CGC
CGA
CGG
AGT
AGC
AGA
AGG
GGT
GGC
GGA
GGG
26
9
62
25
1
0
1
0
58
26
55
18
62
31
115
61
1
0
0
0
1
2
0
1
2
1
9
3
1
1
5
3
LeuTERTrp
92
LeuPro74
24
46
5
79
43
70
6
117
64
63
His58
20
66
16
78
56
88
34
67
28
74
26
Arg
Gln
IleThrAsnSer
LysArg
Met
ValAla AspGly
Glu
4
96Z. Shao et al. / Gene 381 (2006) 92–101