Mol. Cells, Vol. 22, No. 3, pp.
314-322
The Complete Mitochondrial Genome of Pollicipes mitella
(Crustacea, Maxillopoda, Cirripedia): Non-Monophylies of
Maxillopoda and Crustacea
Jong Tae Lim and Ui Wook Hwang*
Department of Biology, Teachers College, Kyungpook National University, Daegu 702-701, Korea.
(Received August 23, 2006; Accepted October 10, 2006)
The whole mitochondrial genome (14,915 nt) of Pollici-
pes mitella (Crustacea, Maxillopoda, Cirripedia, Tho-
racica) was sequenced and characterized. It is the
shortest of the 31 completely sequenced crustacean mi-
tochondrial genomes, with the exception of a copepod
Tigriopus japonicus (14,628 nt). It consists of the usual
13 protein-coding genes, 22 tRNA genes, 2 rRNA genes,
and 1 relatively short non-coding region (294 nt). The
thoracican cirripeds apart from Megabalanus volcano
have the same arrangement of protein-coding genes as
Limulus polypemus, but there are frequent tRNA gene
translocations (at least 8). Some interesting transloca-
tion features that may be specific to the thoracican cir-
riped lineage are as follows: 1) trnK-trnQ lies between
the control region and trnI, 2) trnA-trnE lies between
trnN and trnS1, 3) trnP lies between ND4L and trnT,
and 4) trnY-trnC lies between trnS2 and ND1. In P.
mitella there are two trnL genes (L1 and L2) in the typi-
cal crustacean positions (ND1-L1-LrRNA and CO1-L2-
CO2). The present result is compared and discussed
with the other three cirriped mitochondrial genomes
from one pedunculate (Pollicipes polymerus) and two
sessiles (Tetraclita japonica and M. volcano) published so
far. Mitochondrial protein phylogenies reconstructed by
the BI and ML algorithms show that the thoracican
Cirripedia is monophyletic (BPP 100/BP 100) and asso-
ciated with Remipedia (BPP 98/BP 35). In addition, Oli-
gostraca, including Ostracoda, Branchiura, and Pentas-
tomida, is a monophyletic group (BPP 99/BP 68), and is
basal to all the other examined arthropods. Remipedia
+ Cirripedia appears as an independent lineage within
Arthropoda, apart from Thoracopoda (Malacostraca,
Branchiopda, and Cephalocarida). The Thoracopoda is
paraphyletic to Hexapoda. The present result suggests
* To whom correspondence should be addressed.
Tel: 82-53-950-5911; Fax: 82-53-950-6809
E-mail: uwhwang@knu.ac.kr
Molecules
and
Cells
©KSMCB 2006
that the monophylies of Crustacea and Maxillopoda
should be reconsidered.
Keywords: Cirripedia; Crustacean Phylogeny; Gene Re-
arrangement; Maxillopoda; Mitochondrial Genome; Pol-
licipes mitella; Thoracica.
Introduction
Pollicipes mitella (Crustacea, Maxillopoda, Cirripedia,
Thoracica) is a member of the barnacles, which are major
crustacean species. Barnacles live in brackish water, inter-
tidal to abyssal, on rocks, shells, and other hard substrates,
in aquaculture worldwide. Before the series of Darwin’s
studies (1851, 1854) on fossil and extant barnacle taxa
they were generally believed from their superficial fea-
tures to be mollusks. The order Thoracica is a representa-
tive order of cirriped maxillopods. Since its earliest fos-
sils date back to the Silurian (Wills, 1998), it is believed
to be one of the most slowly evolving species of maxillo-
pods. Therefore it has been considered a crucial taxon for
understanding the evolution of the whole crustacean as-
semblage (Lavrov et al., 2004; Mallat et al., 2004).
Mitochondrial genes have often been used for explor-
ing deep branching animal phylogenies (Hwang et al.,
2001) as well as for research in the field of population
genetics (Park et al., 2004). Recently, complete mito-
chondrial genome-based phylogenetic studies have given
rise to great interest in exploring deep branching patterns
of arthropods because mitochondrial genome-based phy-
logenies have often suggested interesting, sometimes
striking, topologies compared to the traditional arthropod
phylogenetic framework (Hassanin, 2006; Hwang et al.,
2001; Lavrov et al., 2004; Nardi et al., 2003). For in-
stance, mitochondrial protein phylogenies revealed the
existence of an unfamiliar subgroup of Chelicerata +
Myriapoda within arthropods (Hwang et al., 2001), and
 Jong Tae Lim & Ui Wook Hwang
also suggested that Collembola, an endognathan hexapod,
is the most primitive Pancrustacea apart from the other
hexapods (Nardi et al., 2003).
Up to the present 31 crustacean mitochondrial genomes,
including those of 3 cirripeds (1 pedunculate cirriped,
Pollicipes polymerus, and 2 sessile cirripeds, Tetraclita
japonica and Megabalanus volcano) have been com-
pletely sequenced. Nevertheless, the phylogenetic posi-
tion of the cirripeds in the Crustacea based on mitochon-
drial genome information, such as gene arrangements and
nucleotide and amino acid sequences of the 13 protein
coding genes, tRNA structures, A + T skewness, etc has
never been thoroughly considered.
Here we report the complete sequence of the mitochon-
drial genome of the pedunculate Pollicipes mitella and
characterize it in detail. The results are compared with
those for the other 3 cirripeds, P. polymerus, T. japonica
and M. volcano whose complete mitochondrial genomes
have been sequenced. In addition, we performed a phy-
logenetic analysis using the deduced amino acid sequences
of the 12 mitochondrial protein coding genes apart from
ATPase 8 by the maximum-likelihood method and Bayes-
ian inference. We discuss the phylogenetic position of the
cirripeds, and the phylogenetic relationships of the maxil-
lopod crustaceans within Arthropoda, in the light of the
resulting mitochondrial protein phylogeny and the com-
parison of cirriped mitochondrial genomes (especially gene
re-arrangement patterns).
Materials and Methods
Long PCR amplification, and sequencing of the P. mitella
mitochondrial genome Total cellular DNA was isolated from a
specimen of the barnacle P. mitella (Crustacea, Maxillopoda,
Cirripedia, Thoracica) collected in Guryong-Po, Pohang, Korea.
In order to design suitable PCR primers we first amplified a ca.
600-bp portion of the 1rRNA gene (the large subunit ribosomal
RNA gene) using universal primers (16SA: 5′- CGC CTG TTT
ATC AAA AAC AT-3′; 16SB: 5′-CCG GGT GAA CTC AGA
TCA-3′) and a ca. 800-bp portion of the NADH dehydrogenase
subunit 4 (ND4) gene using ND4 (+) 5′-GTR GGK TAY CAR
CCY GAR CG-3′ and ND51 (-) 5′-AHA KAA AVA GMA RVG
CCT TAA A-3′. The amplified PCR fragments were sequenced
and two pairs of long PCR primer sets were designed from the
partial sequences of the lrRNA and ND4. We amplified the
whole P. mitella mitochondrial genome using the Expand Long
Template PCR System (Roche Biochemicals) with two primer
sets: Myr (+) 5′-ATG CTA CYT TWA YRC AGT CAW DAT
ACY GC-3′, and PoNa 5′-CTT AGG AAG ATG GGG TTT ATG
AGG-3′ for PCR amplification of ca. 10.3 kb, and Lpo16b (+)
5′-AAA TCG ACA AGA AGG TTT GCG ACC TCG ATG TTG-
3′ and PoNb 5′-ACA TAG AAA GCC GGC AAC TTA ACC-3′
for amplifying the remaining part (ca. 3.8 kb). The first PCR
cycle included 2 min at 92°C, 30 s annealing at 65°C, and 13
315
min elongation at 68°C. During the last 20 cycles, elongation
times were increased to 20 s per cycle. The reaction was fin-
ished by a 20 min final elongation at 68°C.
The two purified long-PCR fragments (3.8 and 10.3 kb) were
cloned using pGEM-T Easy Vector (50 ng) (Promega Co., USA)
with T4 DNA ligase (Promega Co., USA). Both constructs were
transformed into Escherichia coli DH5-α. The correct recombi-
nants were selected by the blue/white colony selection method
using X-gal and IPTG. Plasmid DNAs were purified using an
AtmanBio Plasmid Miniprep Kit (Takara Co., Japan), sequenced
using a primer walking method with the ABI PRISM BigDye
terminator system and analyzed on an ABI3700 model auto-
matic sequencer (Genotech Co., Korea).
Sequence analysis and genome annotation The genome se-
quences obtained were analyzed using GeneJockey II v1.6 soft-
ware (Biosoft, Inc.), and P. mitella mitochondrial genome com-
ponents (genes) were identified by comparison with the mito-
chondrial genomes of other metazoan animals. The 13 protein
coding genes were confirmed using BLAST searches and the
Clustal X v1.8 multiple sequence alignment program (Thomp-
son et al., 1997). Transfer RNA genes were identified using the
program tRNAscan-SE 1.21 (Lowe and Eddy, 1997), and uniden-
tified tRNA genes were located by eye by comparison with inver-
tebrate mitochondrial anticodon sequences and the secondary
structures of tRNA transcripts. The complete mitochondrial ge-
nome sequence of P. mitella and its annotation data are deposited
under GenBank accession number, AY514042.
Amino acid sequence alignment of protein-coding genes, and
phylogenetic analyses Using the amino acid sequences deduced
from 12 protein-coding genes of the 34 taxa listed in Table 1,
multiple sequence alignments were prepared using Clustal X
with the default option (Thompson et al., 1997). ATPase 8 was
not included because it is too short and variable to use in a phy-
logenetic analysis. The 12 separate amino acid sequence align-
ments were refined by eye and concatenated to a single multi-
ple-sequence alignment with the Gblock program (Castresana,
2000). The alignment finally obtained consisted of the 34 taxa
and 2,401 sites. Gblock has often been used to extract conserved,
well-aligned regions, and to generate a single file made up of
concatenated conserved regions.
The refined and concatenated amino acid sequence alignment
was subjected to both maximum likelihood (ML) and Bayesian
Inference (BI) analyses (Huelsenbeck and Ronquist, 2001). For
the phylogenetic analyses, the best-fitting evolutionary model
was searched with ProtTest ver 1.2 (Abascal et al., 2005). The
GTR + I + Γ model (general time reversible model + among-
site-rate-variation modeled by a gamma distribution + a separate
parameter for invariable sites) was selected as the best-fitting
model among the 32 models tested for ML and BI. ML analysis
was performed using PHYML v2.4.4 (Guindon and Gascuel,
2003). It started from one of the Bio NJ trees, with proportions
of I (0.140) and Γ (0.770) estimated with ProtTest ver. 1.2. For
the ML tree, bootstrap proportions with 100 replicates were also
316 Cirriped Mitochondrial Genome and Crustacean Phylogeny

Table 1. Taxanomic classification, species name, and GenBank accession numbers of the complete mitochondrial genomes examined in
the present study.
Taxonomic classification Species name Accession No.
Phylum Arthropoda
Subphylum Hexapoda
Class Insecta
Order Diptera Drosophila yakuba X03240
Zygentoma Tricholepidion gertschi AY191994
Collembola Gomphiocephalus hodgsoni AY191995
Subphylum Crustacea
Class Malacostraca Harpiosquilla harpax NC006916
Squilla mantis NC006081
Macrobrachium rosenbergii NC006880
Penaeus monodon NC002184
Marsupenaeus japonicus NC007010
Panulirus japonicus NC004251
Cherax destructor NC011243
Pagurus longicarpus NC003058
Eriocheir sinensis NC006992
Pseudocarcinus gigas NC006891
Portunus trituberculatus NC005037
Callinectes sapidus NC006281.
Branchiopoda Artemia franciscana NC001620
Daphnia pulex NC000844
Triops cancriformis NC004465
Triops longicaudatus NC006079
Cephalocarida Hutchinsoniella macracantha AY456189
Cirripedia Megabalanus volcano NC006293
Tetraclita japonica NC008974
Pollicipes polymerus AY456188
Pollicipes mitella Present
Remipedia Speleonectes tulumensis AY456190
Branchiura Argulus americanus AY456187
Ostracoda Vargula hilgendorfii AB114300
Pentastomida Armillifer armillatus AY456186
Subphylum Myriapoda
Class Chilopoda Lithobius forficatus NC_002629
Diplopoda Narceus annularus AY055727
Subphylum Chelicerata
Class Xiphosura Limulus polyphemus AF216203
Phylum Annelida Lumbricus terrestris U24570
Phylum Mollusca Katarina tunicata U09810
Phylum Chordata
Subphylum Vertebrata Homo sapiens NC001807

obtained using SEQBOOT and CONSENSUS in Phylip 3.5c
(http://evolution.genetics.washington.edu/phylip/software.html).
BI tree was reconstructed using MrBayes v3.0b4 (Huelsenbeck
and Ronquist, 2001; Mau et al., 1999) with the following op-
tions: three independent Markov chains (1 hot chain and 2 cold
chains), 500,000 metropolis-coupled MCMC generations, tree
sampling every 100 generations, and a burn-in of 150,000 trees
produced at the initial stage.
Results and Discussion
The mitochondrial genome of the barnacle P. mitella
The length of the P. mitella mitochondrial genome is
14,915 bp. It is the shortest crustacean mitochondrial ge-
nome except for that of Tigriopus japonicus (14,628 nt).
The P. mitella mitochondrial genome contains the stan-
dard complement of 13 protein-coding genes, 22 tRNA
 Jong Tae Lim & Ui Wook Hwang
Fig. 1. Circular map of the Pollicipes mitella (Maxillopoda, Cir-
ripedia, Thoracica, Pedunculate) mitochondrial genome. The posi-
tions and orientations of the genes are shown. Positive integers at
gene junctions indicate the lengths of intergenic spacers, and
negative integers indicate numbers of overlapping nucleotides
between adjacent genes. The genes are abbreviated as follows: A6
and A8 (genes for subunits 6 and 8 of ATPase), CO1-CO3 (genes
for cytochrome C oxidase subunits 1−3), Cytb (gene for apocyto-
chrome coenzyme b), ND1-ND6 and ND4L (genes for NADH
dehydrogenase subunits 1−6 and 4L), lrRNA and srRNA (genes
for 16S and 12S subunit rRNA genes) and CR (control region;
unassigned region). One-letter amino acid abbreviations are used
to label the corresponding tRNA genes. The anticodons of the
pairs of trnL and trnS genes are indicated in parentheses to distin-
guish the members of each pair (L1, trnLCUN; L2, trnLUUR; S1,
trnSAGN; S2, trnSUUR).
genes, 2 rRNA genes, and a non-coding A + T rich region
(294 bp; a possible control region, CR) which is found in
most other metazoans (Fig. 1). Some of the genes overlap
and others have gaps between genes. In total there are 22
overlapping nucleotides between genes, as shown in Fig.
1. The 13 protein coding genes and 22 tRNA genes are very
similar in length to those of the other three cirripeds (Pol-
licipes polymerus, Tetraclita japonica and Megabalanus
volcano). All the gene orientations and gene junctions are
shown in Fig. 1.
The lengths of the lrRNA, srRNA and CR in the four cir-
riped species are given in Table 2. In P. mitella, CR (294
bp) and srRNA (759 bp) are relatively short, but lrRNA is
relatively long. The CR (230 bp) of Squilla mantis (Crusta-
cea, Hoplocarida), which has no apparent function (Cook,
2005), is the shortest known so far among the crustaceans.
Although P. mitella (294 bp) and T. japonica (263 bp) have
slightly larger CRs, they are still very short.
Gene order The order and orientation of all 13 protein-
coding genes in the P. mitella mitochondrial genome
(apart from Megabalanus) are identical to that of Limulus
polyphemus. It is known that Limulus has the ancestral
317
Table 2. Comparison of sequence lengths and G + C contents of
large and small subunit rRNA genes (lrRNA and srRNA) and
control regions from four thoracican cirriped species.
lrRNA srRNA Control region
Length G + C
Length GC Length GC
(bp) (%)
T. japonica 1313 28.3 801 35.2 263 27.7
M. volcano 1301 27.1 799 32.8 370 24.8
P. mitella 1324 31.6 759 32.0 294 28.5
P. polymerus 1307 27.8 773 35.3 332 30.4
gene arrangement of the chelicerates (even arthropods).
However, at least 8 of the tRNA gene positions/orienta-
tions in P. mitella differ from those of Limulus (Table 3 and
Fig. 2). Comparative analysis of the gene orders reveals
some interesting tRNA gene rearrangements which may be
phylogenetically important characteristics of the thoracican
cirriped lineage: 1) trnK-trnQ lies between the control re-
gion and trnI, 2) trnA-trnE lies between trnN and trnS1, 3)
trnP lies between ND4L and trnT, and 4) trnY-trnC lies
between trnS2 and ND1 (Figs. 1 and 2; Table 3). These
unique characteristics could serve as useful genetic markers
for rapidly identifying fossils or cyprid larvae of thoracican
cirripeds, which are not easy to identify.
According to the crustacean mitochondrial genome data
published to date, the positions of trnLCUN (L1) and
trnLUUR (L2) link crustaceans to hexapods (L1 positioned
between ND1 and 16S rRNA, and L2 between CO1 and
CO2). In cirripeds the positions of L1 and L2 are identical
to those in other malacostracan crustaceans (Fig. 1).
However, in other maxillopods such as Ostracoda, Cope-
poda, and Branchiura, their positions vary greatly (Table
3). Interestingly, L1-L2 between ND1 and 1rRNA shown
in Ostracoda is usually found in two other arthropod sub-
groups (Myriapoda and Chelicerata), arthropod relatives
(Onychophora and Tardigrada), and other metazoan ani-
mals (Boore, 1999; Boore et al., 1998).
Base composition and gene content The nucleotide com-
position of the 13 protein-coding genes in the P. mitella
mitochondrial genome is as follows: A, 33.9%; C, 22.1%;
G, 11.6%; T, 31.4%. The total G + C content is 34.7%,
which is similar to the 35.5% of Artemia franciscana.
This G + C content falls within the range described for
other arthropods. It is slightly lower than in two other
maxillopods (39.6% in the copepod T. japonicus and
38.4% in the ostacod Vargula hilgendorfii) and in a bran-
chiopod (37.7% in Daphnia pulex), but higher than in a
malacostracan (29.7% in Portunus trituberculatus) and a
collembolan (25.9% in Gomphiocephalus hodgsoni) (Ta-
ble 3). In crustaceans, comparison of the entire mito-
chondrial genomes reveals that Argulus americanus
(Branchiura) has the highest and V. hilgendorfii (Ostra-
318 Cirriped Mitochondrial Genome and Crustacean Phylogeny

Table 3. Comparison of base compositions and interesting tRNA gene rearrangements of mitochondrial genomes in 12 crustaceans, 1
insect, 1 myriapod, 1 chelicerate, and 1 pentastomid species.
G+C ND2 - CO1
Classification and species name Length (bp) ND1 – lrRNA CO1-CO2
contents (%) (Cytb/S2 – ND1)
Crustacea
Cephalocarida
Hutchinsoniella macracantha 16491 28.1 L1 - -
Branchiopoda
Artemia franciscana 15822 35.5 L1 L2 WIQCY
Triops cancriformis 15101 31.2 L1 L2 WCY
Malacostraca
Pagurus longicarpus 15630 36.6 L1 L2 - WQC
Penaeus monodon 15984 29.4 L1 L2 WCY
Remipedia
Speleonectes tulumensis 18372 32.5 - L2 WY
Cirripedia
Tetraclita japonica 15194 33.9 L1 L2 W (CY)*
Megabalanus volcano 15107 31.6 L1 L2 W (YKQC)*
Pollicipes polymerus 15634 32.9 L1 L2 W (YC)*
Pollicipes mitella 14915 34.7 L1 L2 W (YC)*
Ostracoda
Vargula hilgendorfii 15923 38.4 L1 L2 - WCY
Branchiura
Argulus americanus 15102 22.1 - L2 CQY
Pentastomida
Armillifer armillatus
16747 37.8 - L2 WCQY
Hexapoda
Drosophila yacuba 16019 21.4 L1 L2 WCY
Chelicerata
Limulus polyphemus 14985 32.4 L1 L2 - WCY
Myriapoda
Lithobius forficatus 15695 32.1 L1 L2 - WY
* Indicates Thoracica-specific (or possibly Cirripedia-specific) tRNA gene arrangements between the Cytb/S2 and ND1 genes.

Fig. 2. Comparison of the gene arrangements in cirriped mitochondrial genomes. Mitochondrial genome gene arrangements of (a) Pollici-
pes mitella and P. polymerus (Thoracica, Pedunculata) and (b) Tetraclita japonica and Megabalanus volcano (Thoracica, Sessilia) are
shown. Protein-encoding and rRNA genes are indicated by larger boxes and tRNA genes by smaller boxes. Each genome map is
standardized to commence from the CR (control regions; shaded boxes) in the orientation in which the genes are transcribed from left to
right except where underlined. Thick underlines indicate the opposite transcriptional polarity. Genes are abbreviated as in Fig. 1, and lrR
and srR indicate lrRNA and srRNA genes, respectively. Transfer RNA genes are designated by the one-letter code for the specified amino
acid, with numerals differentiating cases where there are two tRNAs for the same amino acid. The tandem array of two copies of trnC is
only found in P. polymerus.
 Jong Tae Lim & Ui Wook Hwang 319

Table 4. Organizations of the 13 protein coding genes in the Pol-

licipes mitella mitochondrial genome.
Start End
Gene Start End* Strand
codon codon
ND3 63 423 ATG T- H
ND5 808 2511 ATG TAA L
ND4 2576 3909 ATG T- L
ND4L 3902 4195 ATT TAA L
ND6 4326 4811 ATG TAA H
Cytb 4814 5953 ATG TAG H
ND1 6164 7081 ATA TAG L
CO1 10946 12490 ACA TAA H
CO2 12565 13248 ATG TAA H
A8 13312 13470 ATT TAA H
A6 13464 14129 ATG TAA H
CO3 14129 14915 ATG T- H
ND2 9884 10882 ATG TAA H
* Designated as the last base of the codon encoding the final amino
acid.

trnSAGN, which have TTT and TCT anticodons, respec-

Fig. 3. The putative secondary structures of the 22 tRNAs of the
Pollicipes mitella mitochondrial genome.
coda) the lowest G + C content (Table 3).
Translation initiation and termination signals Data on
the termini of the 13 protein-coding genes are summa-
rized in Table 4. Out of the 13 protein-coding genes in P.
mitella, 9 (ND2, ND3, ND4, ND5, ND6, Cytb, CO2, CO3
and ATP 6) start with an ATG initiation codon, 2 (ATP8
and ND4L) with ATT, 1 (CO1) with ACA (T), and 1
(ND1) with ATA (M). While 10 protein-coding genes are
terminated with TAA or TAG stop codons, the other 3
(CO3, ND3 and ND4) are inferred to have an incomplete
termination codon (T-). In the latter cases the transcripts
may be modified to generate a complete TAA termination
signal by polyadenylation after the polycistronic RNA is
cut, as demonstrated for other metazoan mitochondrial
genomes (Ojala et al., 1980).
tRNAs and codon usage Most of the tRNAs are obtained
by identifying intergenic regions capable of forming typi-
cal clover-leaf structures. In P. mitella there are 22 puta-
tive tRNA genes, as generally observed in metazoan mi-
tochondrial genomes. Their predicted secondary struc-
tures are shown in Fig. 3. Their anticodon sequences are
identical to those commonly found in other arthropod
mitochondrial genomes, with the exceptions of trnK and
tively, as alternatives to the more common CTT and GCT.
These anticodons match at the third wobble position. The
trnK and trnSAGN anticodon variants have been found in
Pagurus longicarpus (Hickerson and Cummingham, 2000),
Panulirus japonicus (Yamauchi et al., 2002), T. japonicus
(Machida et al., 2002) and Apis mellifera (Crozier and
Crozier, 1993), while the trnK anticodon variant alone has
been found in Penaeus monodon (Wilson et al., 2000) and
Heterodoxus macropus (Shao et al., 2001).
The pattern of codon usage of the protein coding genes
of the P. mitella mitochondrial genome is summarized in
Table 5. As in most metazoan mitochondrial genomes there
is a clear tendency to use A+T rich codons and an A or a T
in the third codon position, reflecting the high A + T con-
tent of the P. mitella mitochondrial genome. The most fre-
quently used codons are AUU (233 times) and UUU (222
times), whereas there are only 2 UAG and 1 AGG.
Crustacean phylogeny based on mitochondrial genomes
Mitochondrial protein phylogeny was reconstructed by the
ML and BI methods, which are considered relatively resis-
tant to long-branch attraction artefacts. The trees have an
identical topology, and show that P. mitella is allied with
the other species of the same genus, P. polymerus (BI
posterior probability in percent, BPP 99, and ML boot-
strapping value in percent, BP 40). The two other cir-
ripeds examined in this study, M. volcano and T. japonica,
are clustered together (BPP 58 and BP 78) and belong to
the family Balanidae (order Thoracica). The two Pollici-
pes species (Family Lepadomorpha) and M. volcano and
T. japonica (Family Balanidae) form a strong mono-
phyletic group (BPP 100 and BP 100), supporting the
320 Cirriped Mitochondrial Genome and Crustacean Phylogeny

Table 5. Codon usages of the 13 protein coding genes in the Pollicipes mitella mitochondrial genome.
AA* Codon No. AA Codon No. AA Codon No. AA Codon No.
Phe UUU 222 Ser UCU 119 Tyr UAU 87 Cys UGU 026
UUC 109 UCC 047 UAC 60 UGC 008
Leu UUA 171 UCA 056 Ter UAA 8 Trp UGA 078
UUG 114 UCG 014 UAG 2 UGG 028
CUU 109 Pro CCU 057 His CAU 37 Arg CGU 026
CUC 027 CCC 037 CAC 37 CGC 005
CUA 083 CCA 034 Gln CAA 43 CGA 019
CUG 036 CCG 010 CAG 14 CGG 007
Ile AUU 233 Thr ACU 076 Asn AAU 73 Ser AGU 031
AUC 076 ACC 036 AAC 59 AGC 012
Met AUA 140 ACA 048 Lys AAA 68 AGA 071
AUG 083 ACG 016 AAG 16 AGG 001
Val GUU 084 Ala GCU 090 Asp GAU 41 Gly GGU 067
GUC 023 GCC 051 GAC 38 GGC 026
GUA 111 GCA 044 Glu GAA 74 GGA 108
GUG 045 GCG 011 GAC 22 GGG 044
* AA refers to amino acid.
* No. indicates the number of usages of the corresponding codon.
* There are 3658 amino acid residues, and 3 incomplete termination codons (only T).

monophyly of the order Thoracica (Cirripedia).
Interestingly, the BI and ML trees show that the Cir-
ripedia are associated with the Remipedia (BPP 98 and BP
35). The Remipedia + Cirripedia lineage has a very weak
attraction to the Myriapoda + Chelicerata clade (BPP 65
and BP 31). Schram and Hof (1998) reported a number of
similarities between Remipedia and Maxillopoda, includ-
ing Cirripedia, as follows: loss of eyes in the adult stage,
a carapace-less short-cephalothorax of the head and first
trunk segment, loss of the maxillary endopod, a threeseg-
mented endopod of the trunk limbs, and a defined precoxa
of the maxillule. Such similar characters may point to a
close relationship between Remipedia and Cirripedia. In
addition to morphological data, some molecular phyloge-
netic approaches such as those of Spears and Abele
(1998), based on 18S rDNA data, and Lavrov et al. (2004),
based on mitochondrial amino acid sequence data, also
imply Remipedia + Cirripedia with the undisputed node
confidence values.
The members of the Oligostraca, including Ostracoda,
Pentastomida, and Branchiura, are grouped together (BPP
99 and BP 68), and the resulting group is found to be
basal to all the other arthropods. Oligostracans share re-
duced numbers of true body segments. Within the mono-
phyletic Oligostraca, Argulus americanus (Branchiura)
and Armillifer armillatus (Pentastomida) appear as sisters
with high node confidence values (BPP 100 and BP 100),
supporting the Ichyostraca. Then Vargula hilgendorfii
(Ostracoda) joins with the Ichyostraca clade.
Thus, the maxillopods examined here (Cirripedia, Os-
tracoda, and Branchiura) are divided into two different
lineages: Cirripedia as a sister taxon of Remipedia, and
Branchiura and Ostracoda within the Oligostraca. This
indicates that the Maxillopoda may not be a monophyletic
group.
Since the mitochondrial genome organization and
amino acid sequences of the copepod T. japonicus (Cope-
poda) are totally different from those of any other arthro-
pods, T. japonicus is excluded from the present analysis.
Most molecular phylogenetic studies exclude some ar-
thropod species such as maxillopods because of their
long-branch attractions or irregular accelerated evolution,
extreme protein amino acid composition bias, or peculiar
gene arrangement, etc (e.g. Giribet et al., 2001; Hassanin,
2006; Hwang et al., 2001; Peterson and Eernisse, 2001).
When we add T. japonicus to the analysis, Copepoda is
not clustered with Cirripedia. It even attracts not only
with ingroup taxa appearing as relatively long branches in
the arthropods but also with reference taxa such as Mol-
lusca or Annelida. Interestingly, regardless of inclu-
sion/exclusion of Copepoda, the resulting tree topologies
indicate identical phylogenetic relationships between the
ingroup taxa except for the production of an additional
lineage of Copepoda. But the overall node confidence
values are lower (data not shown).
The present results support the monophyly of Thoraco-
poda (Zrzavy et al., 1998), including Malacostraca, Bran-
chiopda, and Cephalocarida, (BPP 98 and BP 39), as well
as the idea that Thoracopoda is paraphyletic to Insecta (Fig.
4). Within the Thoracopoda, Chephalopoda is grouped with
 Jong Tae Lim & Ui Wook Hwang
Fig. 4. Phylogenetic position of the Cirripedia, and crustacean
phylogenetic relationships inferred from the amino acid sequences
of 12 mitochondrial protein-coding genes (excluding the ATPase 8
gene). The phylogenetic analyses were performed by the Bayesian
inference (BI) and maximum likelihood (ML) methods. The aa-
alignment length employed is 2401, which is obtained for the
mitochondrial genomes of the 34 taxa listed in Table 1. For both
the ML and BI analyses, GTR + I + Γ is used as an amino acid
substitution model. Log likelihood value of the BI tree shown
here is – 67531.031. The values above/below branches on the tree
are Bayesian posterior probabilities in percent (BPP: left numbers
of the slashes) and ML bootstrap proportions (BP: right numbers
of slashes). Bold indicates the thoracican cirriped species. Gray
boxes indicate the maxillopod species examined. The white box
shows that the phylogenetic position of Cherax dustructer differs
between the topologies of the BI and ML trees.
Collembola (BPP 98 and BP 59). The Collembola +
Chephalopoda lineage is found to be a sister of all the re-
maining thoracopods and hexapods. Malacostraca is more
closely related to true insects than Branchiopoda, although
the node confidence is relatively weak (BPP 62 and BP 41).
According to the present phylogenetic analysis, crusta-
ceans are divided into three independent evolutionary
lineages as follows: 1) Oligostraca group, including Os-
tracoda, Branchura, and Pentastomida, 2) Cirripedia and
Remipedia group, and 3) Thoracopoda, including Mala-
costraca, Branchiopoda, and Cephalocarida. This suggests
that the monophylies of Crustacea and Maxillopoda should
321
be seriously reconsidered, although we cannot unambigu-
ously exclude the possibility that the problems result from
artefacts of the phylogenetic analyses caused by long-
branch attractions.
To resolve the problematic crustacean phylogenies it
will be necessary to obtain more mitochondrial genome
sequence data from a variety of maxillopods (especially
missing maxillopods like Mystacocarida and Copepoda),
and to develop phylogenetic-analysis tools and appropri-
ate evolutionary models to overcome artefacts due to
long-branch attraction.
Acknowledgments This research was supported by grants from
the Korea Research Foundation (KRF-2002-015-CS0033) and
the Basic Research Program of the Korea Science and Engineer-
ing Foundation (No. R01-2004-000-10930-0) awarded to UWH.
References
Abascal, F., Zardoya, R., and Posada, D. (2005) ProtTest: selec-
tion of best-fit models of protein evolution. Bioinformatics 12,
12.
Boore, J. L. (1999) Animal mitochondrial genomes. Nucleic
Acids Res. 27, 1767−1780.
Boore, J. L., Lavrov, D. V., and Brown, W. M. (1998) Gene trans-
location links insects and crustaceans. Nature 392, 667−668.
Castresana, J. (2000) Selection of conserved blocks from multi-
ple alignments for their use in phylogenetic analysis. Mol.
Biol. Evol. 17, 540−552.
Cook, C. E. (2005) The complete mitochondrial genome of the
stomatopod crustacean Squilla mantis. BMC genomics 6, 105.
Crozier, R. H. and Crozier, Y. C. (1993) The mitochondrial ge-
nome of the honeybee Apis mellifera: complete sequence and
genome organization. Genetics 133, 97−117.
Darwin, C. R. (1851) Fossil Cirripedia. A monograph on the
fossil Lepadidae, or, pedunculated cirripedes of Great Britain.
in Palaeontographical Society. London.
Darwin, C. R. (1854) Fossil Cirripedia. A monograph of the
fossil Balanidae and Verrucidae of Great Britain. Palaeonto-
graphical Society. London.
Giribet, G., Edgecombe, D., and Wheeler, W. C. (2001) Arthro-
pod phylogeny based on eight molecular loci and morphology.
Nature 413, 157−161.
Guindon, S. and Gascuel, O. (2003) A simple, fast, and accurate
algorithm to estimate large phylogenies by maximum likeli-
hood. Syst. Biol. 52, 696−704.
Hassanin, A. (2006) Phylogeny of Arthropoda inferred from
mitochondrial sequences: Strategies for limiting the mislead-
ing effects of multiple changes in pattern and rates of substi-
tution. Mol. Phylogenet. Evol. 38, 100−116.
Hickerson, M. J. and Cummingham, C. W. (2000) Dramatic
mitochondrial gene rearrangements in the hermit crab Pagu-
rus longicarpus (Crustacea, Anomura). Mol. Biol. Evol. 17,
639−644.
Huelsenbeck, J. P. and Ronquist, F. (2001) MrBays: Bayesian
inference of phylogenetic trees. Bioinformatics 17, 754−755.
322 Cirriped Mitochondrial Genome and Crustacean Phylogeny
Hwang, U. W., Friedrich, M., Tautz, D., Park, C. J., and Kim, W.
(2001) Mitochondrial protein phylogeny joins myriapods
with chelicerates. Nature 413, 154−157.
Lavrov, D. V., Brown, W. M., and Boore, J. L. (2004) Phyloge-
netic position of the Pentastomida and (pan)crustacean rela-
tionships. Proc. R. Lond. B. 271, 537−544.
Lowe, T. M. and Eddy, S. R. (1997) tRNAscan-SE: a program
for improved detection of transfer RNA genes in genomic
sequence. Nucleic Acids Res. 25, 955−964.
Machida, R. J., Miya, M. U., and Nishida, S. (2002) Complete
mitochondrial DNA sequence of Tigriopus japonicus (Crus-
tacea: Copepoda). Mar. Biotechnol. 4, 406−417.
Mallatt, J. M., Garey, J. R., and Shultz, J. W. (2004) Ecdysozoan
phylogeny and Bayesian inference: first use of nearly com-
plete 28S and 18S rRNA gene sequences to classify the ar-
thropods and their kin. Mol. Phylogenet. Evol. 31, 178−191.
Mau, B., Newton, M. A., and Larget, B. (1999) Bayesian phy-
logenetic inference via Markov Chain Monte Carlo methods.
Biometrics 55, 1−12.
Nardi, F., Spinsanti, G., Boore, J. L., Carapelli, A., Dallai, R., et
al. (2003) Hexapod origins: monophyletic or paraphyletic?
Science 299, 1887−1889.
Ojala, D., Merkel, C., Gelfand, R., and Attardi, G. (1980) The
tRNA genes punctuate the reading of genetic information in
human mitochondrial DNA. Cell 22, 393−403.
Park, J.-K., Choe, B. L., and Eom, K. S. (2004) Two mitochon-
drial lineages in Korean freshwater corbicula (Corbiculidae:
Bivalvia). Mol. Cells 17, 410−414.
Peterson, K. J. and Eernisse, D. J. (2001) Animal phylogeny and
the ancestry of bilaterians: inferences from morphology and
18S rDNA gene sequences. Evol. Dev. 3, 170−205.
Schram, F. R. and Hof, C. H. J. (1998) In Arthropod Fossils and
Phylogeny, G. D. Edgecombe, G. D. (ed.), pp. 233−302,
Columbia Univ. Press, New York.
Shao, R., Campbell, N. J. H., and Barker, S. C. (2001) Numer-
ous gene arrangements in mitochondrial genome of the wal-
laby louse, Heterodoxus marcropus (Phthiraptera). Mol. Biol.
Evol. 18, 858−865.
Spears, T. and Abele, L. G. (1998) Crustacean phylogeny inferred
from 18S rDNA; in Arthropod relationships, Fortey, R. A. and
Thomas, R. H. (eds.), pp. 169−187, Chapman & Hall, London.
Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F.,
and Higgins, D. G. (1997) The CLUSTAL-X windows inter-
face: flexible strategies for multiple sequence alignment
aided by quality analysis tools. Nucleic Acid. Res. 25, 4876−
4882.
Wills, M. A. (1998) A phylogeny of Recent and fossil Crustacea
derived from morphological characters; in Arthropod relation-
ships, R. A. Fortey and R. H. Thomas (eds.), pp. 189−209,
Chapman & Hall, London.
Wilson, K., Cahill, V., Ballment, E., and Benzie, J. (2000) The
complete sequence of the mitochondrial genome of the crusta-
cean Penaeus monodon: are malacostracan crustaceans more
closely related to insects than to branchiopods? Mol. Miol.
Evol. 17, 863−874.
Yamauchi, M. M., Miya, M. U., and Nishida, M. (2002) Com-
plete mitochondrial DNA sequence of Japanese spiny lobster,
Panulinus japonicus (Crustacea: Decapoda). Gene 295, 89−
96.
Zrzavy, J., Hypsa, V., and Vlaskova, M. (1998) Arthropod phy-
logeny: taxonomic congruence, total evidence and condi-
tional combination approaches to morphological and molecu-
lar data sets; in Arthropod relationships, Fortey, R. A. and
Thomas, R. H. (eds.), pp. 97−10, Chapman & Hall, London.